Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites [1st Edition] 9780081020029, 9780081020012

Sustainable and Nonconventional Construction Materials Using Inorganic Bonded Fiber Composites presents a concise overvi

1,953 84 38MB

English Pages 494 [473] Year 2017

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites [1st Edition]
 9780081020029, 9780081020012

Table of contents :
Content:
Front-matter,Copyright,List of contributors,Foreword,Summary,Introductory remarks—the nonconventional materials (NOCMAT) for sustainable infrastructure regeneration in 21st centuryEntitled to full textPart 1: Engineered vegetable and other natural fibers as reinforcing elements1 - Lignocellulosic residues in cement-bonded panels, Pages 3-16, Rafael F. Mendes, Alan P. Vilela, Camila L. Farrapo, Juliana F. Mendes, Gustavo H. Denzin Tonoli, Lourival M. Mendes
2 - The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus of cellulose nanofiber, Pages 17-26, Daman Panesar, Ramsey Leung, Mohini Sain, Suhara Panthapulakkal
3 - Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites, Pages 27-68, Sérgio Francisco Santos, Ronaldo Soares Teixeira, Holmer Savastano Junior
4 - Treatments for viable utilization of vegetable fibers in inorganic-based composites, Pages 69-123, Marie-Ange Arsène, Ketty Bilba, Cristel Onésippe
5 - New inorganic binders containing ashes from agricultural wastes, Pages 127-164, Jordi Payá, José Monzó, Maria Victoria Borrachero, Lourdes Soriano, Jorge L. Akasaki, Mauro M. Tashima
6 - New trends for nonconventional cement-based materials: Industrial and agricultural waste, Pages 165-183, Moisés Frías-Rojas, Maria Isabel Sánchez-de-Rojas-Gómez, César Medina-Martínez, Ernesto Villar-Cociña
7 - Alternative inorganic binders based on alkali-activated metallurgical slags, Pages 185-220, Maria Criado, Xinyuan Ke, John L. Provis, Susan A. Bernal
8 - The potential use of geopolymer for cleaning air, Pages 221-233, José Ramón Gasca-Tirado, Alejandro Manzano-Ramírez, José Luis Reyes-Araiza
9 - Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment, Pages 235-263, Philip Van den Heede, Nele De Belie
10 - Hatschek process as a way to valorize agricultural wastes: Effects on the process and product quality, Pages 267-290, Elena Fuente, Rocío Jarabo, Ángeles Blanco, Carlos Negro
11 - A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement, Pages 291-311, Luz M. Murillo, Silvio Delvasto, Marisol Gordillo
12 - Accelerated carbonation as a fast curing technology for concrete blocks, Pages 313-341, Caijun Shi, Zhenjun Tu, Ming-Zhi Guo, Dehui Wang
13 - Macro- and nanodimensional plant fiber reinforcements for cementitious composites, Pages 343-382, Shama Parveen, Sohel Rana, Raul Fangueiro
14 - Using vegetable fiber nonwovens cement composites as sustainable materials for applications on ventilated façade systems, Pages 385-397, Josep Claramunt, Mònica Ardanuy
15 - Potentialities of cement-based recycled materials reinforced with sisal fibers as a filler component of precast concrete slabs, Pages 399-428, Paulo R. Lopes Lima, Alex B. Roque, Cintia M. Ariani Fontes, José M. Feitosa Lima, Joaquim A.O. Barros
16 - Experimental investigations on cement-bonded rattan cane composites, Pages 429-444, Abel O. Olorunnisola
17 - Bi-component polyolefin fibers used for concrete and shotcrete applications, Pages 445-452, Stephen A.S. Akers, Josef Kaufmann, Eugen Schwitter
Index, Pages 453-466

Citation preview

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Related titles Handbook of Alkali-Activated Cements, Mortars and Concretes (ISBN: 978-1-78242-276-1) Eco-efficient Masonry Bricks and Blocks (ISBN: 978-1-78242-305-8) Recent Advances in Asphalt Materials (ISBN: 978-0-08-100269-8) Acoustic Emission and Related Non-destructive Evaluation Techniques in the Fracture Mechanics of Concrete (ISBN: 978-1-78242-327-0)

Woodhead Publishing Series in Civil and Structural Engineering

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites Edited by

Holmer Savastano Junior Juliano Fiorelli Sergio Francisco dos Santos

Woodhead Publishing is an imprint of Elsevier The Officers’ Mess Business Centre, Royston Road, Duxford, CB22 4QH, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, OX5 1GB, United Kingdom Copyright © 2017 Elsevier Ltd. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-08-102001-2 (print) ISBN: 978-0-08-102002-9 (online) For information on all Woodhead Publishing publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Matthew Deans Acquisition Editor: Gwen Jones Editorial Project Manager: Charlotte Rowley Production Project Manager: Poulouse Joseph Cover Designer: Greg Harris Typeset by MPS Limited, Chennai, India

List of contributors

Jorge L. Akasaki UNESP - Universidade Estadual Paulista, Sa˜o Paulo, Brazil Stephen A.S. Akers Akers Consulting GmbH, Mollis, Switzerland Mo`nica Ardanuy Universitat Polite`cnica de Catalunya, Terrassa, Spain Cintia M. Ariani Fontes State University of Feira de Santana, Feira de Santana, Bahia, Brazil Marie-Ange Arse`ne Universite´ des Antilles, Pointe-a`-Pitre, Guadeloupe Joaquim A.O. Barros University of Minho, Guimara˜es, Portugal Susan A. Bernal The University of Sheffield, Sheffield, United Kingdom Ketty Bilba Universite´ des Antilles, Pointe-a`-Pitre, Guadeloupe ´ ngeles Blanco Complutense University of Madrid, Madrid, Spain A Maria Victoria Borrachero Universitat Polite`cnica de Vale`ncia, Valencia, Spain Josep Claramunt Universitat Polite`cnica de Catalunya, Castelldefels, Spain Maria Criado The University of Sheffield, Sheffield, United Kingdom Nele De Belie Ghent University, Ghent, Belgium Maria Isabel Sa´nchez-de-Rojas-Go´mez Eduardo Construction Science (IETcc-CSIC), Madrid, Spain

Torroja

Institute

Silvio Delvasto Universidad del Valle, Cali, Colombia Gustavo H. Denzin Tonoli Federal University of Lavras, Lavras, Brazil Raul Fangueiro University of Minho, Guimara˜es, Portugal

for

xii

List of contributors

Camila L. Farrapo Federal University of Lavras, Lavras, Brazil Jose´ M. Feitosa Lima State University of Feira de Santana, Feira de Santana, Bahia, Brazil Elena Fuente Complutense University of Madrid, Madrid, Spain Jose´ Ramo´n Gasca-Tirado Universidad de Guanajuato, Celaya, Me´xico Khosrow Ghavami Pontifı´cia Universidade Cato´lica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil Marisol Gordillo Universidad Auto´noma de Occidente, Cali, Colombia Ming-Zhi Guo The Hong Kong Polytechnic University, Hong Kong, China Rocı´o Jarabo Complutense University of Madrid, Madrid, Spain Josef Kaufmann Empa, Swiss Federal Laboratories for Materials Science and Technology, Du¨bendorf, Switzerland Xinyuan Ke The University of Sheffield, Sheffield, United Kingdom Ramsey Leung University of Toronto, Toronto, ON, Canada Paulo R. Lopes Lima State University of Feira de Santana, Feira de Santana, Bahia, Brazil Alejandro Manzano-Ramı´rez Centro de Investigaciones y Estudios Avanzados del I.P.N., Quere´taro, Qro, Me´xico Ce´sar Medina-Martı´nez University of Extremadura, Ca´ceres, Spain Juliana F. Mendes Federal University of Lavras, Lavras, Brazil Lourival M. Mendes Federal University of Lavras, Lavras, Brazil Rafael F. Mendes Federal University of Lavras, Lavras, Brazil Jose´ Monzo´ Universitat Polite`cnica de Vale`ncia, Valencia, Spain Luz M. Murillo Universidad del Valle, Cali, Colombia Carlos Negro Complutense University of Madrid, Madrid, Spain

List of contributors

xiii

Abel O. Olorunnisola University of Ibadan, Ibadan, Nigeria Cristel One´sippe Universite´ des Antilles, Pointe-a`-Pitre, Guadeloupe Daman Panesar University of Toronto, Toronto, ON, Canada Suhara Panthapulakkal University of Toronto, Toronto, ON, Canada Shama Parveen University of Minho, Guimara˜es, Portugal Jordi Paya´ Universitat Polite`cnica de Vale`ncia, Valencia, Spain John L. Provis The University of Sheffield, Sheffield, United Kingdom Sohel Rana University of Minho, Guimara˜es, Portugal Jose´ Luis Reyes-Araiza Universidad Auto´noma de Quere´taro, DIPFI, Fac. de Ingenierı´a, Qro, Me´xico Moise´s Frı´as-Rojas Eduardo Torroja Institute for Construction Science (IETccCSIC), Madrid, Spain Alex B. Roque State University of Feira de Santana, Feira de Santana, Bahia, Brazil Mohini Sain University of Toronto, Toronto, ON, Canada Se´rgio Francisco Santos Sa˜o Paulo State University (UNESP), Guaratingueta´, SP, Brazil Holmer Savastano Junior University of Sa˜o Paulo (USP), Sa˜o Paulo, Brazil Eugen Schwitter Fibrotec ag, Mollis, Switzerland Caijun Shi Hunan University, Changsha, China Lourdes Soriano Universitat Polite`cnica de Vale`ncia, Valencia, Spain Mauro M. Tashima UNESP - Universidade Estadual Paulista, Sa˜o Paulo, Brazil Ronaldo Soares Teixeira University of Sa˜o Paulo (USP), Sa˜o Carlos, SP, Brazil

xiv

List of contributors

Zhenjun Tu Hunan University, Changsha, China; The Hong Kong Polytechnic University, Hong Kong, China Philip Van den Heede Ghent University, Ghent, Belgium; Strategic Initiative Materials (SIM vzw), project ISHECO within the program ‘SHE’, Ghent, Belgium Alan P. Vilela Federal University of Lavras, Lavras, Brazil Ernesto Villar-Cocin˜a Central University of Las Villas, Santa Clara, Cuba Dehui Wang Hunan University, Changsha, China

Foreword

The nonconventional materials (NOCMAT) conference has evolved rapidly since the first meeting that I attended in the beautiful coastal town of Joa˜o Pessoa, Paraı´ba State, Brazil, about 15 years ago. Inspired by the global need to develop sustainable buildings, the NOCMAT conference brings together multidisciplinary teams of researchers from the “North” and the “South” to present knowledge and ideas that are needed for the development of the next generation of ecofriendly building materials. This book presents advances in our scientific understanding of cementitious matrices, fibers, interfaces, and composites that are being developed for applications in sustainable buildings. The book is divided into four parts, with each part consisting of four or five chapters that provide cutting-edge new knowledge on nonconventional cementitious matrix composites that are produced from industrial or agricultural waste materials. In Part 1, the chapters explore the effects of processing on the mechanical properties and the durability of cement-based composites. This includes the effects of lignocellulosic residues, alkali treatment, and treatments that improve the properties of natural fibers that are used in inorganic matrix composites. The chapters provide the knowledge that is needed for the engineering of vegetable fibers. Part 2 of the book presents alternative binders that are produced from agricultural and industrial waste materials. These include: new inorganic binders that contain ash from agricultural waste; new trends in the development of inorganic binders; alternative inorganic binders that are based on alkali-activated slags; potential applications of air cleaning geopolymers; and a life cycle assessment of green concrete materials. These chapters provide some excellent examples of alternative binders. The processing and properties of cementitious matrix composite materials are presented in Part 3. This includes: the Hatcheck process as a way of adding value to waste materials; the mechanical and thermal properties of geopolymer composite materials; the use of carbonation in the fast curing of concrete blocks; and the effects of macro- and nanofibers as reinforcements in cementitious matrix composite materials. These chapters provide important new insights into the processing and properties (mechanical/thermal properties) of nonconventional cementitious matrix composite materials. Finally, Part 4 of the book explores the applications of natural fiber-reinforced composites in construction materials and components. These include: the applications of natural fiber-reinforced cement sheets in building envelopes; the potential uses of cement-based natural fiber-reinforced materials as fillers in one way slabs;

xvi

Foreword

and the reinforcement of concrete and shotcrete with bicomponent polyolefin fibers. All sections of the book are well edited and integrated into a coherent structure that spans the full range from fibers to binders/interfaces and matrix materials for emerging composite applications in sustainable buildings. The book will serve as a useful text for advanced graduate courses in civil, environmental, architectural, and materials engineering. I congratulate the editors, authors, and publishers on the publication of an important book that will serve as an excellent reference to those engaged in the development of ecomaterials for applications in sustainable buildings. Wole Soboyejo Worcester Polytechnic Institute, Worcester, MA, USA

Summary

There have been important advances in knowledge and technology regarding nonconventional composites in recent years. Considerable work has been done in enhancing the properties of sustainable composite material: lignocellulosic fiber of different types and scales have been applied, and surface treatments have been implemented to make the fibers more compatible with various inorganic matrices. It is expected that this volume will provide the reader with instructive and educational information for an accurate reflection on the sustainable applications of nonconventional composites, as well as a proper background for assessing future developments. Part 1 provides an introduction to engineered vegetable fibers as reinforcing elements, exploring their role in processing methods, and the mechanical and durability performance of the cement-based composites. Chapter 1 focuses on lignocellulosic residues in cement-bonded panels. Chapter 2 considers the effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus of cellulose nanofibers. Chapter 3 looks at the interfacial transition zone between lignocellulosic fibers and matrix in cement-based composites, while Chapter 4 discusses some treatments for the viable utilization of vegetable fibers in inorganicbased composites. Part 2 focuses on alternative inorganic binders based on agricultural and industrial wastes and some strategies to minimize global warming. Chapter 5 describes new inorganic binders containing ashes from agricultural wastes, while Chapter 6 considers new trends for nonconventional cement-based materials: industrial and agricultural waste. Chapter 7 describes alternative inorganic binders based on alkali-activated metallurgical slags. Chapter 8 brings the potential use of geopolymers for cleaning air and Chapter 9 discusses the sustainability assessment of potentially “green” concrete types using the life cycle assessment. Part 3 concentrates on the processing and characterization of nonconventional cementitious composites. Chapter 10 describes the Hatschek process as a way to valorize agricultural wastes and the effect on the process and product quality, while Chapter 11 discusses the mechanical and thermal properties of geopolymer composite materials. The strategy of using carbonation as fast curing for concrete blocks is evaluated in Chapter 12 and Chapter 13 reviews macro- and nanodimensional vegetable fibers as reinforcements for cementitious composites. Part 4 covers the application and performance of fibrous composites as construction materials and components. Chapter 14 considers natural fiber-reinforced cement sheets for building envelopes. Chapter 15 introduces the potentialities of cement-based materials reinforced with natural fibers as the filler component of

xviii

Summary

one-way slabs. Chapter 16 describes experimental investigations on cement-bonded rattan cane composites. Finally, Chapter 17 shows concrete and shotcrete using bicomponent polyolefin fibers. The editors would like to thank all the contributing authors for their outstanding work and for maintaining our ambitious schedule for this volume. They are also in debt to the external reviewers that kindly helped with the peer-review process of the chapters. Thanks are due to the Elsevier team, in particular Project Managers Charlotte Cockle, Charlotte Rowley and Poulouse Joseph who kept the editors in line throughout this process. Holmer Savastano Junior Juliano Fiorelli Sergio Francisco dos Santos

Introductory remarks—the nonconventional materials (NOCMAT) for sustainable infrastructure regeneration in 21st century Khosrow Ghavami1 and Holmer Savastano Junior2 1 Pontifı´cia Universidade Cato´lica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil, 2 University of Sa˜o Paulo (USP), Sa˜o Paulo, Brazil

Introduction Since the Industrial Revolution, human activity has put an increasing strain on the environment—consuming both renewable and nonrenewable resources at rates previously unseen in the natural ecology of planet Earth. The mechanization of processes, the creation of factories and the use of coal as a fuel source beginning towards the end of the 18th century improved the quality of life for people at the time by enabling a period of economic growth, but also created a dependency on the energy needed to fuel these new, mechanical technologies. In addition, new, synthetic materials began to be produced consuming large amounts of energy and resulting in the emission of pollutants into the environment. In the 200 years since the Industrial Revolution, about 337 billion metric tons of greenhouse gases have been added to the atmosphere, with an increasing percent attributed to industrial activities.1 The production of greenhouse gases at this rate is not sustainable and action must be taken to reduce the burden of human activities on the environment, particularly for the mitigation of climate change. Vegetable fibers, such as sisal, coconut fibers, hemp, wheat straw, bamboo, wood, and earth reinforced with wheat straw (in Persian called Kah-gel), were the main civil engineering materials. Many of the world’s most poverty stricken people live in remote locations, typically rich in natural resources. However, in many of these areas, the use of steel and reinforced concrete in construction has become a symbol of economic status. The local natural materials have been and can be used to create structures that can successfully meet the intended need, with a lower cost and environmental impact, considering precautions are taken to prevent deforestation. Using foreign construction materials raises the cost of construction in developing countries where a significant portion of the population lives in poverty.

xx

Introductory remarks—the nonconventional materials

Organizations like Brazilian Society of Non-Conventional Materials and Technologies (B.NOCMAT, in Portuguese Abmtenc) strive to achieve sustainable development and poverty alleviation by promoting the use of the natural resources.2 This introduction chapter presents a few historic and modern examples of using renewable natural materials, such as vegetable fibers, demonstrating the suitability of using these nonconventional materials in civil engineering applications, as well as their potential for use in modern construction. Also a short summary of the stateof-the-art and knowledge about NOCMAT is presented.

Natural materials in historic constructions Xerxes’ Pontoon Bridge The earliest written record of bridge construction described by Herodotus was a bridge built across the Euphrates River around 600 BC.3 He described a bridge built by Persian ruler, Xerxes, to cross the Hellespont (now the Dardanelles). This bridge consisted of two parallel pontoon bridges each made up of 314 and 360 boats which were tied to the riverbank and T anchored to the riverbed to support the weight of the Persian army, with two million men and horses. The Persian army needed to cross the river to meet the Greeks for the battle at Thermopylae in 480 BC. These bridges were built with natural locally available materials as presented in Fig. 1. Root Bridge of Cherrapunjee The region of Cherrapunjee, in India is one of the wettest places in the world, receiving up to 9.3 m of rainfall in one calendar month (July 1861). Besides this, Cherrapunjee is also home to bridges unlike any other in the world. These bridges are made from the roots of living trees as can be seen in Fig. 2.4 There have been no engineering analyses of the Cherrapunjee root bridges to date; however, making a few assumptions the maximum load capacity of this bridge can be estimated. The main assumptions used for this approximation were that the bridge acts as a cable submitted to a uniform applied load and that the

Figure 1 A depiction of Xerxes’ Pontoon Bridge, 480 BC.3

Introductory remarks—the nonconventional materials

xxi

Figure 2 Double-deck root bridge at Cherrapunjee.4

aerial roots of Ficus elastica have the same mechanical properties as the wood of Hevea brasiliensis, the more common species of rubber tree. Using the geometry of the bridge, the sag was determined and, along with the implementation of equilibrium equations, was subsequently used to solve the maximum load the bridge can support that would produce the maximum tensile stress of the bridge material.

Application of vegetable fibers The use of vegetable fiber reinforcement and of pozzolanic materials dates back more than two millennia.5 Asbestos fibers, for example, are known to have been used almost 4500 years ago to strengthen clay pots, whereas straw, reed, horse hair, and similar fibers are known to have been used in the early history of mankind in Mesopotamia to enhance fracture resistance and toughness of materials.6 Pozzolanic materials for construction were used more than 3000 years. The construction of Pasargadae in Iran (Persia) and Pantheon in Rome, for example, are standing monuments to the ingenuity and skills of the civil engineers who used a mixture of lime, volcanic ash, and crushed brick, stone with tufa, and pumice aggregates to create a structure that has withstood the ravages of time, of the elements, and of human vandalism. This proves that the combination of chemical and mineral admixtures with fiber reinforcement brought out a new composite material for civil construction that is tough, compactable, crack-resistant, and durable.7 It is therefore not surprising that since the 1970s there have been seen vast developments and advancements in the field of fiber reinforced composites (FRC).8 Four decades of research and development have not dimmed in any way the innovation created by the concept of fiber reinforcement of the cement and soil matrix, and its ability to transform a relatively brittle material to tolerate extensive damage, and develop high postcracking ductility and energy absorption capability. It is well known that the transfer of these well performed composites from the laboratory into practical application is not going to be easy or without problems. It

xxii

Introductory remarks—the nonconventional materials

is recognized that many of the in situ performance characteristics of these low cost energy saving composite materials when in full-scale construction, could be strongly influenced by size effects, by anisotropic behavior, fabrication processes, and aggressive environments. This is basically due to the fact that the laboratory experiments are in general, by necessity, carried out on small-scale samples. The importance of long periods of exposure of materials used in the structural elements cannot be ignored. The behavior of newly developed materials to aggressive environments is not yet fully established due to the lack of sufficient scientific data. The effects of scaling have strong influences on the serviceable behavior of materials and structures. The most significant size effects are a reduction in engineering properties, significant changes in dynamic response, and a larger range of failure modes due to the occurrence of more and larger flaws. They are generally ignored by the users. There are many premature failures in the history of civil engineering, where innovations in the laboratory results are interpolated to full scale construction without considering the original constraints of size, exposure regimes, time-dependent deformation, and deterioration processes in time.8 Worldwide urbanization created high population growth, which provoked immense and disproportionate consumption of the world’s energy and material resources after the Second World War. This consequently generated unrestrained waste and, widespread destruction of infrastructure through uncontrolled environmental pollution and resulting global warming, with insatiable demands on the construction industry such that sustainable growth of the cement and concrete industry has become a major challenge for the next millennium. However, we can only achieve sustainable growth if the materials we create and use, and the structures we design and build are cost-effective, give durable service performance over their specified design life, and above all, are such that their engineering capabilities are fully utilized and maximized in their service behavior. The serviceability of concrete caused by its deterioration in the environmental pollution and the climatic changes in modern time has become a major global problem. There is, now, a strong concern about the lack of durability of concrete. When the severity of exposure conditions is combined with poor quality materials and/or defective design and building practices, the process of deterioration is interactive, cumulative, interdependent, and very quick. This type of damage cannot be easily fixed, controlled, or stopped. The world we live in now is very different to the world that we inherited at the beginning of 20th century. The latter half of the 20th century saw unprecedented changes and innovations in materials technology, construction techniques, design processes, and analytical methods, as well as the economists creating and spending vast amounts of money in selling their products although they are highly damaging to our environment. One important factor that the capitalist societies have ignored is the dramatic changes being materialized globally through false propaganda advocating only profits for the owners including in the realm of the construction industries. There have been dramatic developments in the heights of buildings, bridge spans, offshore structures, and concrete technology, applying even where local durable materials exist. After creating in the laboratory fascinating NOCMAT materials which have been tested for all the engineering demands for their applications in practice

Introductory remarks—the nonconventional materials

xxiii

considering strength, stiffness, crack control, toughness and energy absorption capability, then one need to bear in mind three facts: (1) There is no guarantee that the application of the studied material in real structures would match that in the laboratory for the reasons discussed earlier. (2) There is not sufficient assurance for the satisfactory service life performance of the material, say for 40 years. (3) Such materials would not meet the demands of the infrastructure crisis into which the world and society have been immersed through the implementation of wild global capitalism. NOCMAT are imperative for the global needs of a sustainable infrastructure regeneration and rehabilitation, if we are to bring about a better world and just society to strengthen the Quality of Life of all its peoples. So now there is a challenge for the NOCMAT specialist. We need to create and produce materials superior to conventional concrete in terms of crack control, ductility, and energy absorption capability. The NOCMAT materials must have or can be expected to have a well-defined durable service life. They must be consistent with sustainable development. Additionally they are expected to meet an important ecological demand and to be cost-effective.

Reinforced adobes as energy saving construction materials Clay is the main material for adobe production. It is a natural, earthy, fine-grained material. Clay is composed essentially of silica, alumina and water. Adobe is a type of earthen building material including sandy clay, water, straw, and other organic materials. This natural building material is traditionally produced in handmade molded or compressed types and dried in the shadow of sunshine. The molded type is shaped via wooden forms in different dimensions. In general, adobes used in hot and dry climates are extremely durable. There are many known old buildings around the world with nearly 10,000 years of history. Therefore, adobe or earthen constructions are one of the oldest building materials in the human society. As an example, the Bam citadel located in Kerman-Iran is a fortified town completely built of adobes. This symbol of adobe architecture was built more than 2500 years ago and was inhabited up to the end of the 19th century by nearly 10,000 people. In the 20th century it was used as an army barracks until 1932.3,9 This town made of adobe reinforced with wheat straw is now designated as a world heritage site by UNESCO and included a fortress, more than 60 towers, caravanserais, a bazaar, and hundreds of houses. On December 26, 2003, the citadel was almost completely destroyed by an earthquake. Fig. 3 illustrates Bam citadel before earthquake. One of the most common types of reinforced adobe in the Bam citadel is the composition of sand, clay, straw, and water, known as Kah-gel in Persian language. In order to make Kah-gel, the ingredients are mixed very well with a combination of stirring and beating, to reach an homogenized appearance. The traditional way to achieve this combination is mixing with feet and tarps. The first step is throwing some crushed soil on a tarp; next sand is added and the two are then well mixed.

xxiv

Introductory remarks—the nonconventional materials

Figure 3 Bam citadel before earthquake of 2003.

The function of sand particles in the Kah-gel is similar to aggregates in concrete; each particle usually has rough angled sides preventing the high shrinkage of clay during the drying process. The straw acts as reinforcement, similar to the steel or polymer fibers in the cementitious matrix. The volume fractions of each component of the soil composites depend on their application. In a traditional method, before adding straw the weight fraction of sand is between 50% and 85%, and that for clay between 15% and 50%. The weight fraction of the straw varies according to the sand soil mixture, the more clay in the mix, the more straw is added. As a rule of thumb, the amount of straw will not be greater than 30%. One of the main objectives of using natural fibers as reinforcing elements with soil matrices is to prevent cracking of the soil resulting from shrinkage. Tensile shrinkage cracks in the soil are mainly due to rapid and nonuniform drying. Reinforcing fibers in the soil matrices prevent cracking by adhesion or bonding to the soil. The large application of earth as a construction material in the 21st century depends on the knowledge of the chemical, physical, and mechanical properties of the adobe. For example, X-ray diffraction techniques have allowed determining the structure of the clay minerals after the 1920s, while its chemical composition can be obtained using infrared spectra. In general it is agreed that the atomic lattices of most of the clay minerals consist of a unit in which a silicon atom is at the center of and equidistant from four oxygen or hydroxyls arranged to form a tetrahedron. Another structural unit consists of two sheets of closely packed oxygen atoms, hydroxyls have aluminum, iron, or magnesium atoms sandwiched between them. One of the most significant characteristics of adobe is its return to the natural cycle without any wastes. When adobe loses its performance, it would be returned to the environment as an agricultural soil. From the economical point of view, adobe is categorized as the cheapest building material, since clay as its primary raw material can be found in large quantities in different areas around the world; additionally the production process does not need high-energy consumption or

Introductory remarks—the nonconventional materials

xxv

complicated equipment. Moreover, using materials like wheat straw for reinforcing the soil is actually converting a minor agricultural product into a valuable building material. Adobe is one of the best energy saving materials used in building construction. The adobe is fabricated manually, dried in covered open space and voids are left in the adobe block. The voids cause a reduction in the heat transfer rate between the inside and outside of the building. As a comparison, the adobe thermal conductivity coefficient is from 0.5 to 0.7 W/(mK), whilst for concrete and burned brick it is 1.4 1.6 W/(mK) respectively. High permeability Air moisture, rain and snow are absorbed quickly in adobe, so adobe does not have a very good stability in humid regions. Because of water absorption, soil particles expand.10

Fibrous composite materials The cement-based composite reinforced with fibers are multiphase materials composed by two basic phases: the fibers and the inorganic matrix. Many composite materials currently used in civil construction are constituted of fragile matrices, especially pastes, mortars, and concrete produced with ordinary Portland cement. There is a growing application of mineral additions as pozzolanic materials that can partially substitute the conventional binder and, in some utilizations, it is also possible to work with clinker-free cement. This is the case for magnesium-based cement. This binder matrix produces mainly magnesium hydroxide [Mg(OH)2] and magnesium silicate hydrates (M S H) as the main compounds. Technical properties of this cement for fiber-cement production are equal to Portland cement, significantly reducing fiber degradation over time.11 The addition of fibers in the fragile matrix targets the improvement of mechanical properties, such as impact, tensile, and flexure strength. The postcracking behavior is envisaged based on strain-hardening performance or at least for the greater pseudo-plastic deformation of the composite material.12 Cementitious composites can be used in the fabrication of thin boards and the percentage of fibers is normally higher than 5% by volume.13 In this case, the fibers are mainly applied for enhancement of the mechanical strength and toughness. The use of cellulose pulp favors the better packing of the matrix, because of the morphological conditions and malleability of the fiber, making possible the industrial application of even higher amounts of fiber (around 10% by mass).14

Vegetable fiber-cement durability One concern regarding the use of vegetable fibers in conjunction with inorganic matrices is the embrittlement of the resulting composite in the long-term, due to the fiber decomposition promoted by the alkaline environment, in this particular case, the pore water present in the cement matrices. The movement of this alkaline pore water in and out of the vegetable fibers during the wetting and drying cycles will

xxvi

Introductory remarks—the nonconventional materials

accelerate their degradation. In the cement matrix, the cellulose and hemicellulose decomposition occurs mainly by the mechanisms of peeling and hydrolysis that degrade the extremities of the polymeric chains and provoke cuttings responsible for the reduction of the polymerization degree, respectively. Both mechanisms can be catalyzed by the temperature. The reduction of fiber-matrix adhesion (due to the shrinkage of the vegetable fiber in the drying stages) and the mineralization of the fibers (due to the precipitation of the hydration products in the inner voids of the fiber) also correspond to the embrittlement and decay of the composite as discussed by Mohr et al.15 To overcome or mitigate the durability concerns some strategies are adopted: G

G

G

G

Modification of the fiber surface for reducing the hydrophilic character and minimize water absorption, which can be achieved using silanes or methane cold plasma treatment.16 Thermal treatment for prior dimensional stabilization of the vegetable fiber using soak and dry cycles, which will promote the irreversible collapse of the fiber and consequent reduction of water absorption.17 Modification of the cement matrix by mineral additions (such as rice husk ash and bamboo leaf ash) that can reduce the availability of free calcium hydroxide and consequently the pH of the cementitious environment; it can also contribute to the porous refinement of the matrix.18,19 Accelerated curing in a modified atmosphere with an excess of CO2, which can stabilize the composite reducing open porosity, by fast carbonation of the cement matrix, contributing to the fiber preservation.20

Those solutions can be found currently in corrugated and flat sheets for external and internal application with the use of vegetable fiber in hybrid solutions with plastic fiber or as the sole reinforcing element of cement-based matrices. This book brings valuable contributions and examples of how the topic of inorganic bonded fiber composites can be explored to find new engineering solutions to meet the demands of society.

Concluding remarks There are many instances in the world where scientific investigations were interrupted or postponed, resulting in the failure of products that would greatly benefit humanity. That is why it is absolutely essential that methods be carried out correctly. In a world of growing populations and environmental stresses, solutions have to be found. The only way this is going to be accomplished is by research, whether it be rediscovering old technologies that our ancestors used, as is called now NOCMAT or creating completely new products. The same methodology is going to have to be utilized. When the first European explorers set out across the Atlantic or to the Far East they did not have the benefit of any synthetic fibers or materials. Then how did they manage to harness the winds using unwieldy sails? The answer was hemp; hemp was an extremely practical plant having far-reaching

Introductory remarks—the nonconventional materials

xxvii

applications in myriad different industries. Now, after some R&D applied into vegetable fibers, there is sufficient scientific information to use composites reinforced with natural fibers as a substitute for synthetic raw-material. Therefore the results of the research on NOCMAT could contribute to the need for cost-effective, durable, and ecofriendly construction materials for infrastructure without contributing much to the environmental pollution.

Acknowledgment The authors would like to thank the financial support granted by FAPERJ, FAPESP, CNPq, and CAPES. Thanks are also due to all students, colleagues, and technicians helping in the realization of the research programs realized during almost 40 years. Special thanks are due Prof. R. N. Swamy for years of working jointly on NOCMAT and to Mr. Arash Azadeh, Drs. Omar Pandoli, and Alexandr Zhemchuzhnikov among others. Finally thanks to Mrs. Ursula Ghavami who went through the English text.

References 1. Boden, T. A.; Marland, G.; Andres, R. J. Global, Regional, and National Fossil-Fuel CO2 Emissions. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, TN, 2009; Vol. 1. ,http://dx.doi. org/10.3334/CDIAC.. 2. Ghavami, K.; Azadeh, A. “Nonconventional Materials (NOCMAT) for Ecological and Sustainable Development,”. MRS Adv. 2016, 1 (53), 3553 3564. 3. Bennett, D.; Associates, D. B. The History and Aesthetic Development of Bridges. ICE, 2008. 4. Cherrapunjee [Online]. www.cherrapunjee.com of Cherrapunjee Holiday Resort. (accessed July 7, 2016). 5. Swamy, R. N. Fibre Reinforcement of Cement and Concrete. Mate´r. Constr. 1975, 8 (3), 235 254. 6. Swamy, R. N. The Magic of Synergy: Chemical and Mineral Admixtures for High Durability Concrete. In Proceedings of the International Conference on Role of Admix in High Performance Concrete, RILEM, 1999; pp 3 18. 7. Malhotra, V. M. A Global Review With Emphasis on Durability and Innovative Concrete. J. Am. Concr. Inst. 1988, 30, 120 130. 8. Naaman, A. E. High Performance Fiber Reinforced Cement Composites. In HighPerformance Materials: Science and Applications; Naaman, A. E., Ed.; World Scientific Publishing: Singapore, 2008; pp 91 153. 9. Sobral, H. S. Vegetable Plants and Their Fibres as Building Materials: Proceedings of the Second International RILEM Symposium; Routledge: Abingdon, UK, 2004. 10. Ghavami, K.; Toledo Filho, R. D.; Barbosa, N. P. Behaviour of Composite Soil Reinforced With Natural Fibres. Cem. Concr. Compos. 1999, 21 (1), 39 48. 11. Ma´rmol, G.; Savastano Junior, H. Study of the Degradation of Non-Conventional MgOSiO2 Cement Reinforced With Lignocellulosic Fibers. Cem. Concr. Compos. 2017, 80, 258 267.

xxviii

Introductory remarks—the nonconventional materials

12. Soltan, D. G.; das Neves, P.; Olvera, A.; Savastano Junior, H.; Li, V. C. Introducing a Curaua´ Fiber Reinforced Cement-Based Composite With Strain-Hardening Behavior. Ind. Crops Prod. 2017, 103, 1 12. 13. Bentur, A.; Mindess, S. Introduction. In Fibre Reinforced Cementitious Composites; Bentur, A,?Mindess, S., Eds.; 2nd ed. Taylor & Francis: New York, 2007; pp 1 10. 14. Savastano Junior, H.; Warden, P. G.; Coutts, R. S. P. Mechanically Pulped Sisal as Reinforcement in Cementitious Matrices. Cem. Concr. Compos. 2003, 25, 311 319. 15. Mohr, B. J.; Nanko, H.; Kurtis, K. E. Durability of Kraft Pulp Fiber-Cement Composites to Wet/Dry Cycling. Cem. Concr. Compos. 2005, 27, 435 448. 16. Barra, B. N.; Santos, S. F.; Bergo, P. V. A.; Alves, C.; Ghavami, K.; Savastano Junior, H. Residual Sisal Fibers Treated by Methane Cold Plasma Discharge for Potential Application in Cement Based Material. Ind. Crops Prod. 2015, 77, 691 702. 17. Ballesteros, J. M.; Rojas, M. F.; Santos, V.; Fiorelli, J. Potential of the Hornification Treatment on Eucalyptus and Pine Fibers for Fiber-Cement Applications. Cellulose 2017, 24, 2275 2286. 18. Pereira, C. L.; Savastano Junior, H.; Paya´, J.; Santos, S. F.; Borrachero, M. V.; Monzo´, J.; Soriano, L. Use of Highly Reactive Rice Husk Ash in the Production of Cement Matrix Reinforced With Green Coconut Fiber. Ind. Crops Prod. 2013, 49, 88 96. 19. Frı´as, Moise´s; Savastano, H., JR; Villar cocin˜a, Ernesto; Sa´nchez-de-Rojas-Go´mez, M. I.; Santos, S. e´rgio Francisco dos Characterization and properties of blended cement matrices containing activated bamboo leaf wastes. Cem. Concr. Compos. 2012, 34, 1019 1023. 20. Tonoli, G. H. D.; Pizzol, V. D.; Urrea, G.; Santos, S. F.; Mendes, L. M.; Santos, V.; John, V. M.; Frı´as, M.; Savastano Junior, H. Rationalizing the Impact of Aging on Fiber-Matrix Interface and Stability of Cement-Based Composites Submitted to Carbonation at Early Ages. J. Mater. Sci. 2016, 51, 7929 7943.

Lignocellulosic residues in cement-bonded panels

1

Rafael F. Mendes, Alan P. Vilela, Camila L. Farrapo, Juliana F. Mendes, Gustavo H. Denzin Tonoli and Lourival M. Mendes Federal University of Lavras, Lavras, Brazil

1.1

Introduction

Cement-bonded particleboards/panels are products manufactured from a mixture of Portland cement, chemical additives and particles generated from lignocellulosics.1 Generally, these products combine the good qualities of cement (relatively high resistance to water, fire, fungus, and termite infestation coupled with good sound insulation) with those of wood (high strength to weight ratio, nailability, and workability).2,3 Cement-bonded panels are used in building construction for their performance on fire resistance and thermal and acoustic insulation. Furthermore, the use of these panels allows saving of time, especially when applied in a modular way on the construction sites, resulting in the faster implementation of the construction.4,5 According to Latorraca and Iwakiri6 and Fan et al.,3 the conifer species, especially the genus Pinus, are the most used woods for the production of cement-based panels, since they present relatively good compatibility with cement, without affecting the cure of the panels. Nevertheless, most lignocellulosic species can be used as reinforcement in cement-bonded panels.7 The major requirement refers to the chemical composition of the lignocellulosic material, which may impair the hydration of the cement matrix.8 With the increasing production of cement-bonded panels due to growth in the construction sector, several species have being studied as alternatives of rapid growth, as well as the use of forest and agricultural residues.7,9,10 Large quantities of wood and agricultural residues are generated worldwide every year.11,12 Those residues include wood sawdust and bark from sawmill operations and lamination residues, sugar cane bagasse, rice husks, wheat straw, straw, corn cob and stalk, coconut pith, ground nut husk, among others.9,13 18 The use of the residues generated by the Brazilian agricultural and forestry industry into cement-bonded panels is an interesting solution for the wood demand problems of this sector; proper utilization of residues allows value addition to the lignocellulosic materials; improvement of the ductility, flexural, and tensile strengths, fracture toughness and crackinhibiting properties of the cement matrix; lightweight character; and improvement of the thermal insulation properties.19,20 The quality of the cement-bonded panels depends on several factors: the type of lignocellulosic material, density, particle size and volume, aspect ratio, mix design, mixing methods and processing and curing Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00001-2 © 2017 Elsevier Ltd. All rights reserved.

4

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

methods. In this context, this study aimed to evaluate the impact of different types of lignocellulosic feedstocks (Pinus and Eucalyptus residues, sugar cane bagasse and bamboo) on the performance of cement-bonded panels.

1.2

Material and methods

1.2.1 Raw materials Residues of lamination of Pinus oocarpa and Eucalyptus urophylla wood, sugar cane bagasse (Saccharum officinarum L.), bamboo (Bambusa vulgaris) and Pinus (P. oocarpa) shavings were used as raw materials. Trees of P. oocarpa (around 28-years-old), E. urophylla (around 20-years-old) and B. vulgaris (around 3-years-old) culms were obtained from experimental plantings in the Federal University of Lavras (UFLA), in Lavras/MG, Brazil. Sugar cane bagasse was collected in the Usina Monte Alegre, in Areado/MG, Brazil. Eucalyptus and Pinus logs were stored in a water tank at 65 C for 24 h. This process is required to promote the softening of the lignin, facilitating the lamination process. The lamination process was carried out to obtain veneers with 1.5-mm thickness. The residues generated during the veneers lamination were used to generate particles for production of cement-bonded panels. Bamboo culms, sugar cane bagasse and the residues of Pinus and Eucalyptus were processed in a hammer mill to generate “sliver” particles. After this step the particulate material was sieved, and the particles that passed through the 2-mm sieve and were retained in the 0.5-mm sieves, were selected. The resulting particles present dimensions of around 4 mm 3 1.5 mm 3 1.0 mm. The pinus logs were processed for removal of boards. The boards had their dimensions adjusted, which generated waste shavings with dimensions of around 25 mm 3 25 mm 3 2 mm. Ordinary Portland cement (CPV-ARI, NBR573321), water and calcium chloride (CaCl2) as a chemical additive were also used as raw materials for the panels. The chemical additive is used to accelerate the cement cure.

1.2.2 Physical and chemical characterization of the lignocellulosic feedstock The basic density of the different types of lignocellulosic materials were determined according to the procedures described in the NBR1194122 standard. Chemical constituents were determined using the sawdust obtained by the processing of the lignocellulosic materials in a Willey mill. The material that passed through 40-mesh sieve and was retained by the 60-mesh sieve, was conditioned at 22 6 2 C and RH 5 65 6 5% until stabilization and then used in the chemical analysis. The total content of extractives was determined following the NBR 1485323 standard; lignin content was evaluated according to the NBR 798924; ashes content according to NBR 1399925; while holocelluloses (H) were obtained by mass difference (H 5 100 2 extractives 2 lignin 2 ashes).

Lignocellulosic residues in cement-bonded panels

5

1.2.3 Experimental design and production of the cement-bonded panels The experimental design consisted of five treatments: (1) with sugar cane bagasse particles; (2) with bamboo particles; (3) residues of lamination of Eucalyptus; (4) residues of lamination of Pinus; and (5) Pinus shavings. For each treatment, three panels were produced with nominal apparent density (AD) of 1.30 g/cm3 and nominal dimensions of 480 mm 3 480 mm 3 15 mm. Panels were produced following the methodology suggested by Latorraca8 and Lopes26; and the variables used to calculate the mix-design were: nominal AD of 1.30 g/cm3 ; lignocellulosic: cement ratio of 1:2.5; water:cement ratio of 1:1.5; hydration rate of 0.25; and 4% of CaCl2. The components were mixed for 5 min and the resulting mass for obtaining each panel was distributed as a particle mat over aluminum sheets for pressing and clipping. Iron bars with thickness of 13 mm were placed on the aluminum plates for controlling the size and thickness of the panel. The panels were pressed at room temperature using 4 MPa. The pressure was applied until the pressed mat reached the thickness of the iron bars (13 mm), and subsequently clipped. The panels remained clamped for 24 h. Thereafter, the clamps were removed and the panels were stabilized in a controlled room at 22 6 2 C and relative humidity (RH) of 65 6 5% for a period of 28 days. Fig. 1.1 shows the production steps of cementbonded panels on the laboratory scale.

1.2.4 Determination of the panel properties and statistical analysis After curing, the cement-bonded panels were cut into smaller samples for the determination of their physical and mechanical properties. Water absorption (WA) and thickness swelling (TS) after 2 and 24 h, internal bonding and compressive strength (parallel to panel surface) were performed following the ASTM D103727 standard. Modulus of rupture (MOR) and modulus of elasticity (MOE) under bending were performed using the DIN 5236228 standard. Statistical analysis was conducted in a completely randomized design. For comparison between treatments, the analysis of variance and Scott-Knott average test were performed, both at 5% of significance level.

1.3

Results and discussion

1.3.1 Physical and chemical characterization of the lignocellulosic material Sugar cane bagasse presented the lower basic density (0.10 6 0.02 g/cm3 ), while bamboo particles presented basic density of 0.26 6 0.03 g/cm3 , and Eucalyptus and Pinus wood particles were 0.56 6 0.03 g/cm3 and 0.50 6 0.02 g/cm3 , respectively. Table 1.1 presents the chemical composition of the different types of lignocellulosic feedstocks evaluated.

6

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 1.1 Steps for production of the cement-bonded panels: (A) lignocellulosic residues; (B) particle generation hammer mill; (C) particle classification; (D) particle drying; (E) homogenization of raw materials in a mixer; (F) formation of the mattress; (G) pressing at room temperature and clipping; (H) finished cement-bonded panel.

Average and standard deviation values of the chemical constituents of the different lignocellulosic feedstocksa

Table 1.1

Lignocelullosic materials

Extractives

Lignin

Ash

Holocellulose

0.7 (0.1) b 1.3 (0.1) a 0.3 (0.0) c 0.3 (0.0) c

71.1 (0.7) a 65.4 (0.4) b 70.2 (1.6) a 59.3 (0.9) c

% Sugar cane bagasse Bambusa vulgaris Eucalyptus grandis Pinus oocarpa a

12.5 (1.0) a 9.2 (0.3) b 2.3 (0.3) c 10.8 (0.8) a

15.7 (0.9) b 24.1 (0.6) a 27.3 (1.4) a 29.6 (0.3) a

Averages followed by the same letter, in the same column, do not differ from each other by the Scott-knott test with a significance level of 5%.

Lignocellulosic residues in cement-bonded panels

7

Sugar cane bagasse presented a lower content of lignin than the other lignocellulosic materials evaluated, and a high average content of extractives and holocellulose. Bamboo presented the highest ash content. Eucalyptus presented the lowest content of extractives and ashes, and the highest content of lignin and holocellulose. Silva et al.29 evaluated the influence of age and position along the tree trunk in the chemical composition of Eucalyptus grandis obtained for 20-year-old trees: average values of 28.3% of lignin, 4.6% of extractive and 67.1% of holocellulose. Klock et al.30 stated that softwood presents a content of around 69 6 4% for holocellulose, 28 6 2% of lignin and 5 6 3% of extractives; while for hardwoods the values ranged from 70 6 5% holocellulose, 20 6 2% of lignin and 3 6 2% extractives. Marino et al.31 evaluated the chemical composition of 2- to 4-year-old Dendrocalamus giganteus and found a content of 8.4% of extractives, 19.4% of lignin, 0.7% of ash and 71.6% of holocellulose. Barros Filho32 obtained for sugarcane bagasse, derived from a sugar cane mill, the content of 9.4% of extractives, 16.3% of lignin, 0.8% of ashes (solid residues) and 73.5% of holocellulose. In general, the results obtained in the present study are consistent with those observed in other studies for the chemical analysis of different types of lignocellulosics.

1.3.2 Physical properties of the cement-bonded panels The average values obtained for the Apparent density (AD) of each treatment are given in Table 1.2. There was no significant difference between the average AD of the treatments. The average values are in the range of 1.24 1.30 g/cm3 . These panels presented approximate density values used on the industrial scale.33 Fig. 1.2 shows the average values obtained for WA after 2 h (WA2h) and 24 h (WA24h) of water immersion of the cement-bonded panels produced with different types of lignocellulosics. It is observed that the panels produced with residues from lamination of Eucalytus and Pinus wood presented the lowest values of WA2h, differing statistically from the panels produced with the other lignocellulosics. The panels manufactured with residues from lamination of Eucalyptus presented the lowest average values of WA24h, followed by the panels with residues from

Average and standard deviation values of apparent density of the cement-bonded panels with different lignocellulosic materialsa

Table 1.2

Lignocellulosic residues

Apparent density (g/cm3 )

Sugar cane bagasse Bamboo Eucalyptus (lamination) Pinus (lamination) Pinus (shavings)

1.30 (0.02) a 1.30 (0.03) a 1.28 (0.02) a 1.25 (0.02) a 1.24 (0.02) a

a

Average values followed by the same letter in the column do not differ from one another by the Scott-Knott test with a significance level of 5%.

8

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 1.2 Average and standard deviation values of water absorption after 2 h (WA2h) and after 24 h (WA24h) of the cement-bonded panels with lignocellulosic materials. Average values followed by the same letter (lower case for WA2h and upper case for WA24h) do not differ from each other by the Scott-Knott test with a significance level of 5%.

lamination of Pinus. The higher WA of the cement-bonded panels with sugar cane bagasse and bamboo is due to the increased surface area of those particles, since they present very low density and consequently there is a need to increase the number of particles in the panel with a predetermined density.13 Also, the lower content of lignin in the bamboo and sugar cane bagasse particles resulted in higher hydrophilicity than those particles with higher contents of lignin. Lignin has 1 2 free OH groups/C9, while the number of free OH groups in cellulose is 3 OH/C6.34 Based on this it is obvious that species with high lignin content have lower surface energy, negligible polar contribution, and decreased WA. The lower affinity of lignin-rich particles to water might be helpful to avoid fiber degradation, as alkaline pore water is the main agent for fiber degradation into the cementitious matrix.35 Additionally, according to Fan et al.,3 Tittelein et al.,36 and Tonoli et al.37 extractives prejudice the cement curing and the interaction between lignocellulosic particles and the cement matrix, leading to empty spaces between the matrix and the lignocellulosic particles/fibers. Fig. 1.3 shows the average TS after 2 h (TS2h) and 24 h (TS24h) of water immersion of the cement-bonded panels. Panels produced with residues of lamination of Eucalyptus and Pinus presented the lowest TS2h values, followed by the panels with sugar cane bagasse. Cement-bonded panels with residues from lamination of Eucalyptus presented the lowest TS24h values, because of the same reasons presented for the lower values of WA. Shaving particles promoted the highest WA and TS values, which is probably related to the pores generated between particles due to the poor packing into

Lignocellulosic residues in cement-bonded panels

9

Figure 1.3 Average and standard deviation values of thickness swelling after 2 h (TS2h) and after 24 h (TS24h) of the cement-bonded panels with the different lignocellulosic materials. Average values followed by the same letter (lower case for TS2h and upper case for TS24h) do not differ from each other by the Scott-Knott test with a significance level of 5%.

the matrix. In this case, the shaving particles were larger than the others, which probably led to large porosity in the cement-bonded panels. The same was observed by Olorunnisola,2 who described coconut husk particles impacting greatly on the WA properties. The Bison wood-cement board38 process suggests maximum values of 0.8% for TS2h and values between 1.2% and 1.8% for TS24h. In the other hand, Viroc33 establishes a maximum value of 1.5% for TS after 24 h of water immersion. Panels with residues of lamination of Pinus or Eucalyptus and sugar cane bagasse panels met the Bison criteria for TS. Only panels prepared with residues of lamination of Eucalytpus and Pinus met the specifications of Viroc.33 In general, the panels produced with lamination residues of Eucalyptus and Pinus wood showed the best values for the physical properties. This fact may be associated with the smaller dimensions of these particles, leading to better coating of the particles with the cement and thus better bonding of the particles with the matrix, proven in the internal bond test (Fig. 1.4), and associated with particle interaction with the cement matrix.

1.3.3 Mechanical properties of the panels Fig. 1.4 shows the average values obtained for modulus of the rupture (MOR) of the cement-bonded panels produced with different types of lignocellulosic material. The panels produced with residues from lamination of Eucalyptus and Pinus obtained the highest average values of the internal bond, presenting statistical

10

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 1.4 Average and standard deviation values of internal bonding for cement-bonded panels with the different lignocellulosic materials. Average values followed by the same letter do not differ from each other by the Scott-Knott test with a significance level of 5%.

equality and differentiating from the other treatments. The panels produced with sugar cane bagasse presented higher values of internal bonding than panels prepared with bamboo and Pinus shavings. Panels with Pinus shavings presented lower values of internal bonding, which is justified by the increase of voids depending on the geometry of the particles and their ability to mix with the cement particles. The fact that the panels with Eucalyptus presented higher internal bonding may be related to the lower extractives content (Table 1.2), which might affect the cement cure and prejudice the interaction of the lignocellulosic particles with the cement matrix.2,3,39,40 It is known that polysaccharides present in wood extractives are responsible for the inhibition of cement setting, due to a negative effect on the formation of the main hydrates (calcium silicate hydrate: C S H).36 Another factor that affects the internal bonding in the panels is the higher density of the particles, as observed in the preparation of the Eucalyptus panels, which required fewer particles when compared to the other feedstocks. The advantages of using a lower number of particles are improved packing and wettability of the particles by the cement.19,41 The Bison wood-cement board38 process and Viroc33 suggest a minimum value of 0.5 MPa for the internal bonding of cement-bonded panels. Only panels with residues of lamination of Eucalyptus and Pinus and the panels with sugar cane bagasse met those requirements. Figs. 1.5 and 1.6 present average MOR values and parallel compressive strength, respectively, of the cement-based panels produced from different types of lignocellulosic feedstock. The relationship observed for MOR and compressive strength was the same as discussed for the internal bonding, with the highest values for

Lignocellulosic residues in cement-bonded panels

11

Figure 1.5 Average and standard deviation values of modulus of rupture (MOR) of the cement-bonded panels with the different lignocellulosics. Average values followed by the same letter do not differ from each other by the Scott-Knott test with a significance level of 5%.

Figure 1.6 Average and standard deviation values of compressive strength of the cementbonded panels with the different lignocellulosics. Average values followed by the same letter do not differ from each other by the Scott-Knott test with a significance level of 5%.

panels with residues of lamination of Eucalyptus and Pinus wood. Panels with sugar cane bagasse presented intermediate results. This direct relationship between internal bonding, MOR, and compressive strength is related to the transmission of the stress that occurs between lignocellulosics and

12

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

cement particles,42 which was affected by factors such as the content of extractives, interaction between particle and cement matrix and particle dimensions.2,19,43 Fig. 1.7 shows the MOE values of the cement-bonded panels. Panels with residues from lamination of Eucalyptus and Pinus, and panels with particles of sugar cane bagasse present the highest MOE values. The panels with Pinus shavings obtained the lowest MOE values. Therefore, MOE values were also affected directly by the interaction between lignocellulosic particles and the cement matrix, presenting a close relationship with internal bonding, which was directly affected by the chemical composition and dimensions of the particles. As the lignocellulosic particle volume increases, the regions of stress concentration around adjacent particles become more diffuse, resulting in an increased resistance to the applied stress. However, the reduced cement content must remain high enough to afford a complete wetting of the particles. For example, at a lower cement wood ratio, complete wetting may not occur, thereby decreasing panel strength.44 This fact was observed when comparing wood particles and sugar cane bagasse. It is important to emphasize that the dimensions and geometry of the particles have strong effects on the composite properties.45 Therefore, some care must be taken when mixing wood strands with cement, in order to avoid the breakdown of the strands that would result in a decrease of the particle dimensions.46 Due to the elongated nature of strands, it is expected that there are some difficulties when mixing with cement, which could lead to incomplete coating/wetting of the particles with cement.47

Figure 1.7 Average and standard deviation values of modulus of elasticity (MOE) of the cement-bonded panels with the different lignocellulosics. Average values followed by the same letter in the column do not differ from one another by the Scott-Knott test with a significance level of 5%.

Lignocellulosic residues in cement-bonded panels

13

Additionally, the chemical composition of the lignocellulosics, such as hemicelluloses, starches, sugars, phenols, and hydroxylated carboxylic acids contents, can lead to hydration problems for the cement in the cement-bonded panels.48 The Bison wood-cement board38 process and Viroc33 determined a minimum value of 9 MPa for MOR. For MOE values, the Bison wood-cement board38 process suggests a minimum of 3000 MPa, while Viroc33 suggests a minimum of 4500 MPa. Therefore, only panels with residues from lamination of Eucalyptus and Pinus wood, and panels with sugar cane bagasse met the requirements stipulated for MOR and MOE.38 Any treatment met the requirements suggested by Viroc33 for MOE values.

1.4 ü ü ü ü ü

Conclusion

The panels produced with residues from lamination of Eucalyptus wood presented the best performance for all physical and mechanical properties. The panels with residues from lamination of Eucalyptus and Pinus wood met all the requirements of Bison wood-cement board38 process. The dimensions/geometry of the particles affected the physical and mechanical properties of the cement-bonded panels. Larger particles led to higher performance. The chemical composition of the lignocellulosics impact on the performance of the panels. The sugar cane bagasse presented greater potential for use in cement-bonded panels than bamboo particles.

Acknowledgments The authors would like to thank the Federal University of Lavras (UFLA), CNPq, FAPEMIG and CAPES, Brazil for supporting the experimental work and scholarships. Thanks also to the Graduation Program in Biomaterials Engineering (UFLA) and Rede Brasileira de Compósitos e Nanocompósitos Lignocelulósicos (RELIGAR).

References 1. Iwakiri, S. Paine´is de Madeira Reconstituı´da; FUPEF: Curitiba, 2005. 2. Olorunnisola, A. O. Effects of Husk Particle Size and Calcium Chloride on Strength and Sorption Properties of Coconut Husk Cement Composites. Ind. Crops Prod. 2009, 29, 593 598. Available from: http://dx.doi.org/10.1016/j.indcrop.2008.09.009. 3. Fan, M.; Ndikontar, M. K.; Zhou, X.; Ngamveng, J. N. Cement-Bonded Composites Made From Tropical Woods: Compatibility of Wood and Cement. Constr. Build. Mater. 2012, 36, 135 140. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2012.04.089. 4. Wolfe, R. W.; Gjinolli, A. Durability and Strength of Cement-Bonded Wood Particle Composites Made From Construction Waste. For. Prod. J. 1999, 49 (2), 24 31.

14

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

5. Matoski, A.; Hara, M. M.; Iwakiri, S.; Casali, J. M. Influence of Accelerating Admixtures in Wood-Cement Panels: Characteristics and Properties. Acta Sci. 2013, 35 (4), 655 660. Available from: http://dx.doi.org/10.4025/actascitechnol.v35i4.11261. 6. Latorraca, J. V. F.; Iwakiri, S. Effect of Particle Treatment of Eucalyptus dunni (Maid), Wood:Cement Ratio and Additives on the Physical and Mechanical Properties of Wood-Cement Boards. Cerne 2000, 6 (1), 68 76. 7. Ashori, A.; Tabarsa, T.; Amosi, F. Evaluation of Using Waste Timber Railway Sleepers in Wood-Cement Composite Materials. Constr. Build. Mater. 2012, 27 (1), 126 129. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2011.08.016. 8. Latorraca, J. V. F. Eucalyptus spp. na produc¸a˜o de paine´is de cimento-madeira. Thesis (D.Sc.), Universidade Federal do Parana´, 2000. 9. Mendes, R. F.; Mendes, L. M.; Abranches, R. A. S.; Santos, R. C. S.; Guimara˜es, J. B., Jr. Particleboards Produced with Sugar Cane Bagasse and Eucalyptus Wood. Sci. For. 2010, 38 (86), 285 295. Available from: http://dx.doi.org/10.1590/S1516-14392013005000004. 10. Farrapo, C. L.; Mendes, R. F.; Guimara˜es, J. B., Jr.; Mendes, L. M. Utilization of Pterocarpus violaceus Wood in the Particleboard Production. Sci. For. 2014, 42 (103), 329 335. 11. Almeida, R. R.; Del Menezzi, C. H. S.; Teixeira, D. E. Utilization of the Coconut Shell of Babac¸u (Orbignya sp.) to Produce Cement-Bonded Particleboard. Bioresour. Technol. 2002, 85, 159 163. doi:10.1016/S0960-8524(02)00082-2. 12. Karade, S. R. Cement-Bonded Composites From Lignocellulosic Wastes. Constr. Build. Mater. 2010, 24, 1323 1330. Available from: http://dx.doi.org/10.1016/ j.conbuildmat.2010.02.003. 13. Mendes, R. F.; Guimara˜es, J. B., Jr.; Santos, R. C.; Bufalino, L. The Adhesive Effect on the Properties of Particleboards Made From Sugar Cane Generated in the Distiller. Rev. Ciˆenc. Agra´r. 2009, 32, 209 218. 14. Barros Filho, R. M.; Mendes, L. M.; Novack, K. M.; Aprelini, L. O.; Botaro, V. R. Hybrid Chipboard Panels Based on Sugarcane Bagasse. Urea Formaldehyde and Melamine Formaldehyde Resin. Ind. Crops Prod. 2011, 33, 369 373. Available from: http://dx.doi.org/10.1016/j.indcrop.2010.11.007. 15. Mendes, R. F.; Guimara˜es, J. B., Jr.; Santos, R. C.; Ce´sar, A. A. S. Efeito da associac¸a˜o de bagac¸o de cana, tipo e teor de adesivo na produc¸a˜o de paine´is aglomerados com madeira de pinus. Ciˆenc. Florestal 2012, 22 (1), 187 196. Available from: http://dx.doi. org/10.5902/198050985088. 16. Scatolino, M. V.; Silva, D. W.; Mendes, R. F.; Mendes, L. M. Uso do sabugo de milho na produc¸a˜o de paine´is aglomerados. Ciˆenc. Agrotecnol. 2013, 37 (4), 330 337. Available from: http://dx.doi.org/10.1590/S1413-70542013000400006. 17. Guimaraes, B. M. R.; Mendes, L. M.; Tonoli, G. H. D.; Bufalino, L.; Mendes, R. F.; Guimara˜es, J. B., Jr. Chemical Treatment of Banana Tree Pseudostem Particles Aiming the Production of Particleboards. Ciˆenc. Agrotecnol. 2014, 38 (1), 43 49. Available from: http://dx.doi.org/10.1590/S1413-70542014000100005. 18. Mendes, R. F.; Mendes, L. M.; Oliveira, J. E.; Savastano, H., Jr.; Tonoli, G. H. D. Modification of Eucalyptus Pulp Fiber Using Silane Coupling Agents with Aliphatic Side Chains of Different Length. Polym. Eng. Sci. 2015, 55 (6), 1273 1280. Available from: http://dx.doi.org/10.1002/pen.24065. 19. Aggarwal, L. K. Bagasse-Reinforced Cement Composites. Constr. Build. Mater. 1995, 17 (2), 107 112. doi:10.1016/0958-9465(95)00008-Z. 20. Mendes, L. M.; Mendes, R. F.; Tonoli, G. H. D.; Bufalino, L.; Mori, F. A.; Guimara˜es, J. B., Jr. Lignocellulosic Composites Made from Agricultural and Forestry Wastes in Brazil. Key Eng. Mater. 2012, 517, 556 563. Available from: http://dx.doi.org/10.4028/ www.scientific.net/KEM.517.556.

Lignocellulosic residues in cement-bonded panels

15

21. Associac¸a˜o Brasileira de Normas Te´cnicas. NBR 5733. Cimento Portland de alta resistˆencia inicial, especificac¸a˜o, 1983; 5p. 22. Associac¸a˜o Brasileira de Normas Te´cnicas. NBR 11941. Madeira—Determinac¸a˜o da densidade ba´sica, 2003; 6p. 23. Associac¸a˜o Brasileira de Normas Te´cnicas. NBR 14853. Madeira—Determinac¸a˜o do material solu´vel em etanol-tolueno e em diclorometano e em acetona, 2010; 3p. 24. Associac¸a˜o Brasileira de Normas Te´cnicas. NBR 7989. Pasta celulo´sica e madeira— Determinac¸a˜o de lignina insolu´vel em a´cido, 2010; 6p. 25. Associac¸a˜o Brasileira de Normas Te´cnicas. NBR 13999. Papel, carta˜o, pastas celulo´sicas e madeira—Determinac¸a˜o do resı´duo (cinza) apo´s a incinerac¸a˜o a 525 C, 2003; 4p. 26. Lopes, Y. L. V. Utilizac¸a˜o da madeira e cascas de Eucalyptus grandis Hill ex Maiden na produc¸a˜o de paine´is cimento-madeira. Thesis (M.Sc.), Universidade Federal de Lavras, 2004. 27. American Society for Testing and Materials. ASTM D-1037. Standard Methods of Evaluating Properties of Wood-Base Fiber and Particles Materials, 2006; 30p. ¨ R Holzfaserplaten Spanplatten Sperrholz. DIN 52362. Testing of Wood 28. Normen FU Chipboards Bending Test, Determination of Bending Strength, 1982; pp 39 40. 29. Silva, J. C.; Matos, J. L. M.; Oliveira, J. T. S.; Evangelista, W. V. Influˆencia da idade e da posic¸a˜o ao longo do tronco na composic¸a˜o quı´mica da madeira de Eucalyptus grandis Hill ex. Maiden. Rev. A´rvore 2005, 29 (3), 455 460. Available from: http://dx.doi.org/ 10.1590/S0100-67622005000300013. 30. Klock, U.; Mun˜iz, G. I. B.; Hernandez, J. A.; Andrade, A. S. Quı´mica da Madeira; Universidade Federal do Parana´: Curitiba, 2005. 31. Marinho, N. P.; Nisgoski, S.; Klock, U.; Andrade, A. S.; Muniz, G. I. B. Ana´lise quı´mica do bambu-gigante (Dendrocalamus giganteus wall. ex munro) em diferentes idades. Ciˆenc. Florestal 2012, 22 (2), 417 422. Available from: http://dx.doi.org/ 10.5902/198050985749. 32. Barros Filho, R. M. Paine´is aglomerados a base de bagac¸o de cana-de-ac¸u´car e resinas ure´ia formaldeı´do e melamina formaldeı´do. Thesis (M.Sc.), Federal de Ouro Preto, 2009. 33. VIROC. Caracterı´sticas Viroc. ,http://www.viroc.pt. (accessed January, 2016). 34. Kajanto, I.; Niskanen, K. Dimensional Stability. In Paper Physics. Papermaking Science and Technology; Niskanen, K., Ed.; Finnish Paper Engineers’ Association and TAPPI: Helsinki, Finland, 1998; pp 223 259. 35. Tonoli, G. H. D.; Almeida, A. E. F. S.; Pereira-Da-Silva, M. A.; Bassa, A.; Oyakawa, D.; Savastano, H., Jr. Surface Properties of Eucalyptus Pulp Fibres as Reinforcement of Cement-Based Composites. Holzforschung 2010, 64, 595 601. Available from: http:// dx.doi.org/10.1515/HF.2010.073. 36. Tittelein, P.; Cloutier, A.; Bissonnette, B. Design of a Low-Density Wood-Cement Particleboard for Interior Wall Finish. Cem. Concr. Compos. 2012, 34 (2), 218 222. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2011.09.020. 37. Tonoli, G. H. D.; Mendes, R. F.; Siqueira, G.; Bras, J.; Belgacem, M. N.; Savastano, H., Jr. Isocyanate-Treated Cellulose Pulp and Its Effect on the Alkali Resistance and Performance of Fiber Cement Composites. Holzforschung 2013, 67 (8), 853 861. Available from: http://dx.doi.org/10.1515/hf-2012-0195. 38. Bison Wood-Cement Board. Bison-Report [S.l.], 1978; 10p. 39. Pizzol, V. D.; Mendes, L. M.; Savastano, H., Jr.; Frı´as, M.; Davila, F. J.; Cincotto, M. A., et al. Mineralogical and Microstructural Changes Promoted by Accelerated Carbonation and Ageing Cycles of Hybrid Fiber-Cement Composites. Constr. Build. Mater. 2014, 68, 750 756. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.06.055.

16

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

40. Iwakiri, S.; Silva, L. S.; Trianoski, R.; Bonduelle, G. M.; Rocha, V. Y. Avaliac¸a˜o do potencial de utilizac¸a˜o da madeira de Schizolobium amazonicum “Parica´” e Cecropia hololeuca “Embau´ba” para produc¸a˜o de paine´is aglomerados. Acta Amazonica 2010, 40 (2), 303 308. Available from: http://dx.doi.org/10.1590/S0104-77602012000200015. 41. Savastano, H., Jr.; Warden, P. G.; Coutts, R. S. P. Brazilian Waste Fibres as Reinforcement for Cement-Based Composites. Cem. Concr. Compos. 2000, 22 (5), 379 384. doi:10.1016/S0958-9465(00)00034-2. 42. Matoski, A.; Iwakiri, S. Avaliac¸a˜o das propriedades fı´sico-mecaˆnicas de paine´is de cimento-madeira utilizando farinha de madeira com granulometria controlada. Floresta 2007, 37 (2). Available from: http://dx.doi.org/10.5380/rf.v37i2.8646. 43. Monteiro, S. N.; Terrones, L. A. H.; Carvalho, E. A.; Almeida, J. R. M. Efeito da interface fibra/matriz sobre a resistˆencia de compo´sitos polime´ricos reforc¸ados com fibras de coco. Rev. Mate´r. 2006, 11 (4), 395 402. Available from: http://dx.doi.org/10.1590/ S1517-70762006000400005. 44. Moslemi, A. A.; Pfister, S. C. The Influence of Cement-Wood Ratio and Cement Type on Bending Strength and Dimensional Stability of Wood-Cement Composite Panels. Wood Fiber Sci. 1987, 19 (2), 165 175. 45. Sotannde, O. A.; Oluwadare, A. O.; Ogedoh, O.; Adeogun, P. F. Evaluation of CementBonded Particle Board Produced Fromafzelia africana Wood Residues. J. Eng. Sci. Technol. 2012, 7 (6), 732 743. 46. Papadopoulos, A. N.; Ntalos, G. A.; Kakaras, I. Mechanical and Physical Properties of Cement-Bonded OSB. Holz Roh Werkst. 2006, 64 (6), 517 518. Available from: http:// dx.doi.org/10.1007/s00107-005-0092-6. 47. Ma, L. F.; Yamauchi, H.; Pulido, O. R.; Sasaki, H.; Kawai, S. Production and Properties of Oriented Cement-Bonded Boards From Sugi (Cryptomeria japonica D. Don). ACIAR Proceedings; ACIAR: Canberra, ACT, 1998. 48. Miller, D. P.; Moslemi, A. A. Wood-Cement Composites: Effect of Model Compounds on Hydration Characteristics and Tensile Strength. Wood Fiber Sci. 1991, 23 (4), 472 482.

The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus of cellulose nanofiber

2

Daman Panesar, Ramsey Leung, Mohini Sain and Suhara Panthapulakkal University of Toronto, Toronto, ON, Canada

2.1

Introduction

The concept of using fibers as reinforcement in building materials is timeworn, such as with horsehair in mortar, or straw in mud and bricks. Since concrete displays relatively low tensile strength and ductility as a composite building material, research into steel and polymer fibers has shown how fiber reinforcement can be used to enhance ductility, tensile strength, toughness, fatigue strength, impact resistance, and energy absorption.1 4 However, with a pressing demand for building materials from renewable sources, the mechanical reinforcement of cement composites with vegetable fibers is of interest, whilst also considering their availability at a relatively lower cost in comparison to synthetic fibers.5 Nevertheless, load-induced failures in cement-based materials occur as a gradual multiscale process in which cracks initiate at the nanoscale level, where traditional macro- and microscale fiber reinforcements are not effective.6 Thus, theory suggests that nanofiber reinforcement would delay the formation of nanocracks, necessitating higher loads to initiate cracking and, therefore, improve the tensile strength of cement composites; this has been tested by recent research into carbon nanofibers.7 Yet, depending on the synthesis method used, the production of carbon-based nanofibers has been shown to be highly energy intensive,8,9 rousing a subsequent interest in cellulose nanofibers as an alternative reinforcement. Indeed, the use of cellulose nanofiber reinforcement in polymer composites has resulted in substantial improvements to mechanical properties, as reported in several recent papers.10 12 Wang and Sain reported that the tensile strength and the stiffness of a polyvinyl alcohol (PVA) composite reinforced with 5% soybean nanofiber increased by 58% and 170%, respectively, compared to pure PVA.11 Additionally, Seydibeyoglu and Oksman observed that the strength and the stiffness of a polyurethane (PU) matrix reinforced with 16.5% by weight of wood cellulose nanofibrils increased by approximately 500% and 3000%, respectively, compared to pure PU.10 More recently, Xu et al.12 showed that the flexural strength and the Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00002-4 © 2017 Elsevier Ltd. All rights reserved.

18

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

toughness of a polyethylene oxide (PEO) matrix reinforced with an additional 7% by weight of cellulose nanofiber improved by 92% and 732%, respectively.12 Thus, the use of nano-sized cellulose fibers in building materials, such as cement composites, has the potential to significantly enhance mechanical properties as a reinforcing agent with a very high specific surface area.5 Currently, research into the application of cellulose nanofibers to cement composite building materials is novel. Preliminary research has reported that optimum mechanical properties were observed in cement pastes reinforced with cellulose nanofibers at a fraction of 0.1% by mass, which increased the flexural strength and the energy absorption by approximately 106% and 184%, respectively, compared to reference cement pastes.5 However, higher nanofiber concentrations yielded diminishing mechanical properties of cement pastes, including a brittle failure mode, which were attributed to the difficulty of ensuring adequate dispersion of cellulose nanofibers within the composite matrix.5 This observed phenomenon of aggregation is alternatively called agglomeration or entanglement and is attributed to the high density of polarized hydroxyl groups at the surface of cellulose nanofibers that form additional weak hydrogen bonds between adjacent parts of the nanoparticles.13 However, the abundance of hydroxyl groups at the surface of cellulose nanofibers also makes it possible to attempt chemical modifications in order to introduce stable surface charges or tune surface energy characteristics, so as to obtain better dispersion and improve compatibility with nonpolar composite matrices.13 Yet, the main challenge with chemically modifying cellulose nanofibers lies in ensuring that the process only changes the surface structure of the nanofibers, while avoiding any polymorphic conversions that might deteriorate their reinforcing capability.13 This study examines the feasibility of utilizing alkali-treated cellulose nanofibers as reinforcement within cement composite building materials. A mild alkali treatment with sodium hydroxide (NaOH), also known as mercerization, was chosen for use in this study due to its longstanding establishment as a common, low cost, and simple process developed in 1844 by John Mercer for use in many textile industries. Three levels of sodium hydroxide concentrations were used in this study: 0%, 2.5%, and 4% NaOH concentration.

2.2

Experimental fiber preparation procedure and test methods

2.2.1 Alkali treatment Mechanical pulp, procured from Domtar Canada, was treated prior to mechanical defibrillation with an initial solution of 2.5%, and 4% NaOH at 60 C for 1 h. In addition, a control or untreated batch was prepared in a similar fashion. The solution was prepared by combining fine mechanical pulp fibers, NaOH pellets, and distilled water in a 1:20 fiber:solution ratio within a large beaker that was then

The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus

19

placed on top of a Corning PC-351 Hot Plate and subsequently mixed with a Canlab Caframo mechanical stirrer. After the 1 h treatment, the pulp was then washed thoroughly with distilled water through a 75 µm sieve until all the sodium hydroxide was eliminated and the pulp was alkali free, as determined by checking the pH periodically using pH paper. After disintegrating the neutralized pulp for 10 min, the pulp was then stored at 4 C in a cold room.

2.2.2 Mechanical defibrillation Nanocellulose fiber gel suspensions were prepared by mechanical defibrillation of both the control and alkali-treated batches of mechanical pulp. To ensure complete defibrillation, the batches of mechanical pulp were fed through a super fine disc grinder for 10 17 times. Thereafter, distilled water was added to the ground pulp to form a viscous translucent nanofiber gel suspension. To ensure comparable consistencies of gradation between the control and alkali-treated batches, sieve tests were performed.

2.2.3 Cellulose nanofiber film production A volume of nanofiber gel suspension was weighed for the purposes of further dilution in 200 mL of distilled water so as to produce a thin 70 µm film, 9.5 cm in diameter for a desired density of 0.7 g/cm3 after vacuum-filtration. To prepare the nanofiber gel suspension for filtration, the 200 mL diluted volume was first stirred for 1 h on a magnetic hotplate at a moderate speed so as to avoid the formation of bubbles. Next, a Bu¨chner vacuum funnel (attached to a Bu¨chner vacuum flask) was wet with distilled water, adjusted for horizontal alignment, and then lined with a filter paper disc, on top of which was placed a membrane disc. The 200 mL diluted volume of nanofiber gel suspension was then poured into the vacuum funnel and filtered down into the vacuum flask, after which the remaining nanofiber film was sandwiched with another membrane disc. Each nanofiber film sandwich was then removed from the vacuum funnel, sandwiched between stacks of drying hand-sheets, and pressed under 3.45 Bar (50 psi) for three intervals of 5 min each until dry. The damp hand-sheets in contact with the sandwich were exchanged for dry handsheets after each interval. After pressing, the membranes were removed and the remaining nanofiber films were then cut into multiple dumbbell shaped strips with a die according to ASTM D 638 (type V). After cutting, the nanofiber strips were dried at 60 C for 24 h, after which they were stored in a desiccator to preserve the strips from atmospheric moisture prior to mechanical testing.

2.2.4 Mechanical testing The tensile tests were performed with an Instron Model 3367 testing machine linked to a computer in tensile mode with a load cell of 1 kN in accordance with standard method ASTM D 638. Tensile tests were performed at a crosshead speed

20

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

of 2.5 mm/min. All the reported values are the average of at least five successful tests.

2.2.5 Transmission electron microscopy The degree of fibrillation of both the control and alkali-treated cellulose nanofibers was investigated using a Hitachi H-7000 120 kV Transmission Electron Microscope (TEM) equipped with a tungsten filament electron source and whose images were captured on an Advanced Microscopy Technology XR60 CCD camera at magnification factors of both 90003 and 400003.

2.3

Results and discussion

Table 2.1 presents the mean value 6 one standard deviation of tensile strength and Young’s modulus of the alkali-treated and untreated cellulose nanofiber strips. The coefficient of variation (COV) for the control specimens was marginally lower than that of the 4% alkali-treated specimens for both tensile strength and Young’s modulus, however all COVs were low in this comparison at less than 5%. It can be noted that the alkali-treated strips showed improved values of tensile strength and modulus at an increase of 20.2% and 24.0%, respectively, compared to the untreated strips. A subsequent investigation of the 2.5% alkalitreated specimens suggests a similar comparison against the untreated control specimens with similarly improved values of tensile strength and Young’s elastic modulus at an increase of 20.5% and a slight decrease of 0.8%, respectively. However, it should be noted that the 2.5% alkali-treated specimens had higher COVs than the 4% alkali-treated specimens, perhaps explained by irregularities in dispersion of nanofibers subsequent to the comparatively lower concentration of NaOH during treatment. Fig. 2.1 presents a direct tensile test curve, which also illustrates the 4% alkalitreated cellulose nanofiber strips as having a mean maximum applied load higher than that of the untreated cellulose nanofiber strips. Table 2.1

Properties of cellulose nanofiber strips

Sample

Tensile strength (MPa)

Young’s modulus (GPa)

Coefficient of variation of tensile strength (%)

Coefficient of variation of young’s modulus (%)

Control 2.5% Alkali-Treated 4% Alkali-Treated

149.20 6 4.15 179.87 6 21.88 179.40 6 7.20

8.94 6 0.30 8.87 6 1.13 11.09 6 0.50

2.78 12.16 4.01

3.36 12.74 4.51

The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus

21

Figure 2.1 Measured direct tensile test curve showing the direct tensile load versus specific displacement.

The mean maximum applied load of the 4% alkali-treated strips was 23.68 N compared to the mean maximum applied load of the control strips at 18.35 N. The mean maximum displacement for both sets of strips was 0.576 mm. Since we cannot directly measure the tensile strength of individual nanofibers, we measured the tensile strength of nanofiber strips cut from films as a comparative approximation. Yet, it should be noted that single nanocellulose fibers, of which the films are comprised, would have a much higher strength than the films themselves, with cellulose nanocrystals estimated to have an elastic modulus of approximately 150 GPa and a tensile strength of nearly 10 GPa.14 Therefore, the improvements observed with the alkali-treated strips were expected for a variety of reasons. During alkali treatment, it is likely that the NaOH reacted with hydroxyl groups on the surface of the cellulose nanofibers, disrupting their hydrogen bonds and increasing surface roughness.15 With this breakdown in cellular structure, a certain amount of lignin and hemicellulose was probably removed from the cellulose nanofibers, causing them to split into smaller filaments; a phenomenon termed fibrillation.15 This phenomenon of fibrillation is indicated by the TEM images shown in Fig. 2.2A D, which represent interwoven cellulose nanofibers that appear more dispersed when alkali-treated at 4% NaOH, as compared alongside nonalkali-treated fibers from the control batch at both 90003 and 400003 magnification factors. As a consequence of the dissolution of these lignin and hemicellulose fractions, the intrafibrillar regions would have become less dense and rigid, allowing for the rearrangement of the fibrils along with the direction of tensile deformation and strengthening their tensile characteristics.15,16 Furthermore, this fibrillation process would have increased the effective surface area available for contact with the matrix and possible modification through reaction sites.16,17 Comparatively, the phenomenon of fibrillation is less apparent in the TEM images shown in Fig. 2.3A B, which represent alkali-treated cellulose nanofibers at 2.5% NaOH that appear less dispersed than those that were alkali-treated at 4% NaOH, at both 90003 and 400003 magnification factors. These images provide support for an explanation of the findings that the 2.5% alkali-treated specimens had higher COVs than the 4% alkali-treated specimens, in suggesting that the lower

22

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 2.2 TEM images showing the effect of fiber treatment (A) control fibers at 90003 ; (B) 4% NaOH-treated fibers at 90003; (C) control fibers at 400003; and (D) 4% NaOHtreated fibers at 400003.

concentration of NaOH during treatment led to irregularities in the subsequent dispersion of the nanofibers. Certainly, similar mercerization processes have also been reported to increase the crystallinity index of sisal fibers through the removal of lignin and hemicellulose fractions, leading to a better packing of cellulose chains.15 Additionally, several authors have reported similar observations of fibrillation and a breakdown in cellular structure with sisal fibers,18,19 flax,20 jute,21 and coir.22 Furthermore, several authors have reported on the successful enhancement of mechanical properties when incorporating alkali-treated natural fibers in composite applications, due to better interfacial adhesion.23 26 In future investigations, it would therefore be expected that the inclusion of alkali-treated cellulose nanofibers would modify the functional properties of

The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus

23

Figure 2.3 TEM images of 2.5% NaOH-treated fibers (A) 90003 and (B) 400003.

Table 2.2

Fiber cement composite flexural strength

Fiber content

0 0.05N 0.1N 0.2N 0.4N

Control specimens (no fiber treatment) Mean (MPa)

Standard deviation (MPa)

4.63 8.24 7.81 7.33 7.53

0.55 0.84 0.71 0.45 0.39

Specimens containing treated fibers

Mean (MPa)

Standard deviation (MPa)

7.41 6.08 7.73 7.69

0.17 0.38 0.9 0.44

nanocellulose dispersed concrete materials. However, it is important to note that the continuing challenges associated with the degree of dispersion of nanocellulose in a concrete/cement matrix should be addressed by finding a pragmatic solution to narrow down the surface energy differences among the individual components. Preliminary flexural strength results of nanofiber-cement composites are shown in Table 2.2 and were obtained via methodology similar to that of prior investigations from this research group.5 Each mean represents the average of three or four specimens. In general, it is shown that the incorporation of 0.05 0.4 N nanofibers, statistically significant to a 95% confidence level, improved the 28-day flexural strength in tests of comparison against the reference cement paste specimens (without any fibers). Furthermore, although the standard deviation needs to be considered in interpretation of the data, it is shown that at low fiber content (0.05 and 0.1 N) the mean flexural strength is lower for the mercerized specimens (4% NaOH treatment),

24

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

but at relatively higher fiber contents (0.2 and 0.4 N), there is no statistically significant effect of the 4% NaOH mercerization treatment. Further examination and testing of fiber cement composites for fibers treated with varying percentages of NaOH will be helpful to better understand the effect of treated nanofibers on the mechanical performance of the cement composites. That is, despite similar values of tensile strength and Young’s elastic modulus between the 2.5% NaOHand 4% NaOH-treated fibers, the 2.5% NaOH-treated nanofibers appear less dispersed than those treated at 4% NaOH. It would be worthwhile to examine and verify how differences in nanofiber dispersion characteristics throughout a concrete/cement matrix may affect the flexural strength and other properties of the cement composite; comparing both the less dispersed 2.5% NaOH fibers against the more well dispersed 4% NaOH-treated fibers. Ultimately, it may be of interest to repeat these experiments with nanofibers treated with a higher concentration of NaOH.

2.4

Conclusions

The chemical surface treatment of cellulose nanofibers with sodium hydroxide was investigated in this work and compared to control untreated fibers. Alkali treatment was found to improve slightly the mechanical properties of the treated nanofiber strips, which could be attributed to the rearrangement of fibrils along the direction of tensile deformation, as a result of the dissolution of lignin and hemicelluloses fractions. An important outcome to note is that although the 2.5% NaOH treatment and 4% NaOH-treated fibers yielded similar tensile strength, the COV of the 2.5% NaOH was approximately three times larger. This could be attributed to irregularities in dispersion of nanofibers subsequent to the comparatively lower concentration of NaOH during treatment. While the surface hydroxyl groups of cellulose nanofibers have allowed for the possibility of the chemical modifications in this study and future expansions of it, a deterioration of the reinforcing capabilities of these nanofibers is a concern of utilizing these treatments, certainly at higher concentrations of chemicals. As such, alternative physical treatments using low temperature plasma or electrical corona discharge are of future interest to investigate as they may potentially modify the surface structures of cellulose nanofibers without altering their bulk morphology and reinforcing capabilities. An emerging topic of further research from this study is the investigation of the mechanical properties of cement composites upon incorporation of these alkalitreated cellulose nanofibers, in addition to quantification of nanofiber dispersion within a cement paste model. In addition to the aforementioned physical treatments, further studies on alternative chemical treatments and concentrations in relation to mechanical strength development, cement microstructure, and durability will be needed to encourage increased utilization of cellulose nanofibers in cement-based materials.

The effect of sodium hydroxide surface treatment on the tensile strength and elastic modulus

25

Acknowledgments The authors would like to thank and acknowledge the financial assistance of this research by Professor Sain’s Ontario Research Fund Research Excellence (ORF-RE) BioNib Project for materials and equipment support. We also acknowledge support from Professor Panesar’s NSERC Discovery Grant. Additionally, the authors would like to thank Alessandra de Souza Fonseca, Lynn He, and Viktoriya Pakharenko for their assistance with this investigation.

References 1. Balendran, R. V.; Zhou, F. P.; Nadeem, A.; Leung, A. Y. T. Influence of Steel Fibres on Strength and Ductility of Normal and Lightweight High Strength Concrete. Build. Environ. 2002, 37 (12), 1361 1367. Available from: http:\\dx.doi.org\10.1016/S03601323(01)00109-3. 2. Banthia, N.; Moncef, A.; Chokrii, K.; Sheng, J. Micro-fiber Reinforced Cement Composites Uniaxial Tensile Response. Can. J. Civil Eng. 1994, 21 (6), 999 1011. DOI. 10.1139/l94-105. 3. Khaloo, A. R.; Afsharii, M. Flexural Behaviour of Small Steel Fibre Reinforced Concrete Slabs. Cem. Concr. Compos. 2005, 27 (1), 141 149. Available from: http://dx. doi.org/10.1016/j.cemconcomp.2004.03.004. 4. Zhang, J.; Stang, H. Fatigue Performance in Flexure of Fiber Reinforced Concrete. ACI Mater. J. 1998, 95 (1), 58 67. 10.14359/351. 5. Onuaguluchi, O.; Panesar, D. K.; Sain, M. Properties of Nanofibre Reinforced Cement Composites. Constr. Build. Mater. 2014, 63, 119 124. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2014.04.072. 6. Metaxa, Z. S.; Konsta-Gdoutos, M. S.; Shah, S. P. Mechanical Properties and Nanostructure of Cement-Based Materials Reinforced with Carbon Nanofibers and Polyvinyl Alcohol (PVA) Microfibers. ACI Special Publication 2010, 270, 115 124. Available from: http://dx.doi.org/10.14359/51663743. 7. Metaxa, Z. S.; Konsta-Gdoutos, M. S.; Shah, S. P. Carbon Nanofiber-reinforced Cement-based Materials. Transp. Res. Rec. 2010, 2142, 114 118. Available from: http://dx.doi.org/10.3141/2142-17. 8. Khanna, V.; Bakshi, B. R. Carbon Nanofiber Polymer Composites: Evaluation of Life Cycle Energy Use. Environ. Sci. Technol. 2009, 43 (6), 2078 2084. Available from: http://dx.doi.org/10.1021/es802101x. 9. Kim, H. C.; Fthenakis, V. Life Cycle Energy and Climate Change Implications of Nanotechnologies. J. Ind. Ecol. 2013, 17 (4), 528 541. Available from: http://dx.doi. org/10.1111/j.1530-9290.2012.00538.x. 10. Seydibeyoglu, M. O.; Oksman, K. Novel Nanocomposites based on Polyurethane and Micro Fibrillated Cellulose. Compos. Sci. Technol. 2008, 68 (3-4), 908 914. Available from: http://dx.doi.org/10.1016/j.compscitech.2007.08.008. 11. Wang, B.; Sain, M. Isolation of Nanofibres from Soybean Source and their Reinforcing Capability on Synthetic Polymers. Compos. Sci. Technol. 2007, 67 (11-12), 2521 2527. Available from: http://dx.doi.org/10.1016/j.compscitech.2006.12.015. 12. Xu, X.; Liu, F.; Jiang, L.; Zhu, J., et al. Cellulose Nanocrystals vs. Cellulose Nanofibrils: A Comparative Study on their Microstructures and Effects as Polymer Reinforcing Agents. ACS Appl. Mater. Interfaces 2013, 5 (8), 2999 3009. Available from: http://dx.doi.org/10.1021/am302624t.

26

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

13. Islam, M. T.; Alam, M. M.; Zoccola, M. Review on Modification of Nanocellulose for Application in Composites. Int. J. Innov. Res. Sci. Eng. Technol. 2013, 2 (10), 5444 5451. ISSN: 2319-8753. 14. Lu, P.; Hsieh, Y. L. Preparation and Properties of Cellulose Nanocrystals: Rods, Spheres, and Network. Carbohydr. Polym. 2010, 82 (2), 329 336. Available from: http://dx.doi.org/10.1016/j.carbpol.2010.04.073. 15. Sreekumar, P. A.; Thomas, S. P.; Saiter, J. M.; Joseph, K., et al. Effect of Fiber Surface Modification on the Mechanical and Water Absorption Characteristics of Sisal/polyester Composites Fabricated by Resin Transfer Molding. Compos. Part A 2009, 40 (11), 1777 1784. Available from: http://dx.doi.org/10.1016/j.compositesa.2009.08.013. 16. Gassan, J.; Bledzki, A. K. Possibilities for Improving the Mechanical Properties of Jute/ epoxy Composites by Alkali Treatment of Fibres. Compos. Sci. Technol. 1999, 59 (9), 1303 1309. Available from: http:\\dx.doi.org\10.1016/S0266-3538(98)00169-9. 17. Cao, Y.; Shibata, S.; Fukumoto, I. Mechanical Properties of Biodegradable Composites Reinforced with Bagasse Fiber before and after Alkali Treatments. Compos. Part A 2006, 37 (3), 423 429. 10.1016. 18. Cyras, V. P.; Iannance, S.; Kenny, J. M.; Vazquez, A. Relationship between Processing and Properties of Biodegradable Composites based on PCL/starch Matrix and Sisal Fibers. Polym. Composites 2001, 22 (1), 104 110. Available from: http://dx.doi.org/ 10.1002/pc.10522. 19. Vazquez, A.; Dominguez, V.; Kenny, J. M. Bagasse Fiber-polypropylene based Composites. J. Thermoplast. Compos. Mater. 1999, 12 (6), 477 497. Available from: http://dx.doi.org/10.1177/089270579901200604. 20. Sharma, H. S. S.; Fraser, T. W.; Mccall, D.; Lyons, G. Fine Structure of Chemically Modified Flax Fiber. J. Text. Ins. 1995, 86 (4), 539 548. Available from: http://dx.doi. org/10.1080/00405009508659033. 21. Ray, D.; Sarkar, B. K. Characterization of Alkali-treated Jute Fibers for Physical and Mechanical Properties. J. Appl. Polym. Sci. 2001, 80 (7), 1013 1020. Available from: http://dx.doi.org/10.1002/app.1184. 22. Sreenivasan, S.; Iyer, P. B.; Krishna, I. K. R. Influence of Delignification and Alkali Treatment on the Fine Structure of Coir Fibers (CocosNucifera). J. Mater. Sci. 1996, 31 (3), 721 726. Available from: http://dx.doi.org/10.1007/BF00367891. 23. Alvarez, V. A.; Ruscekaite, R. A.; Vazquez, A. Mechanical Properties and Water Absorption Behavior of Composites Made from a Biodegradable Matrix and Alkalinetreated Sisal Fibers. J. Compos. Mater 2003, 37 (17), 1575 1588. Available from: http://dx.doi.org/10.1177/0021998303035180. 24. Edeerozey, A. M. M.; Akhil, H. M.; Azhar, A. B.; Ariffin, M. I. Z. Chemical Modification of Kenaf Fibres. Mater. Lett. 2007, 61 (10), 2023 2025. Available from: http://dx.doi.org/10.1016/j.matlet.2006.08.006. 25. Mwaikambo, L. Y.; Ansell, M. P. Chemical Modification of Hemp, Sisal, Jute and Kapok Fibers by Alkalization. J. Appl. Polym. Sci. 2002, 84 (12), 2222 2234. Available from: http://dx.doi.org/10.1002/app.10460. 26. Valadez-Gonzalez, A.; Cervantes-Uc, J. M.; Olayo, R.; Herrera-Franco, P. J. Effect of Fiber Surface Treatment on the Fiber-matrix Bond Strength of Natural Fiber Reinforced Composites. Compos. Part B Eng. 1999, 30 (3), 309 320. Available from: http:\\dx.doi. org\10.1016/S1359-8368(98)00054-7.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

3

Se´rgio Francisco Santos1, Ronaldo Soares Teixeira2 and Holmer Savastano Junior3 1 Sa˜o Paulo State University (UNESP), Guaratingueta´, SP, Brazil, 2University of Sa˜o Paulo (USP), Sa˜o Carlos, SP, Brazil, 3University of Sa˜o Paulo (USP), Sa˜o Paulo, Brazil

3.1

Introduction

The cementitious composites reinforced with lignocellulosic fibers are complex materials. Thus, cement-based composite design must be considered for the interrelation between interfacial transition zone, fibers, matrix, and processing method as depicted in the schematic illustration shown in Fig. 3.1. The bulk structure (matrix) of the composite among others aspects shown in Fig. 3.1 is fundamental for durability and mechanical performance of the fiber-cement. Therefore, some aspects must be considered, such as particles size distribution, water/binder ratio and rheological properties in the fresh state, morphology and dimensions of the fibers, production and curing processes. It is important to note that these aspects are interconnected. The basic particulate materials for these composites are ordinary Portland cement (OPC) and mineral addition (filler and/or pozzolanic mineral). In general, the mineral additions are added to reduce the overall cost of the cement-based composite product and to improve particle size distribution. Besides, pozzolanic minerals are silica-rich materials and influence the amount and kind of hydrates formed and thus the porosity and finally the durability of these composites. The use of mineral additions as supplementary materials leads to a significant reduction of the production of clinker, the principal component of Portland cement, which releases CO2 from the calcination of limestone and is also highly energy-intensive. Fibers are essential for the ductile performance of fiber-cement, a composite with a brittle matrix to prevent catastrophic failures and to promote cracking control. The use of short random lignocellulosic fibers in different scales (nano- and micro-lengths) as reinforcement in cement-based composite materials has been studied to partially replace the synthetic fibers, mainly polyvinyl alcohol (PVA) and polypropylene (PP) fibers. Several efforts are made to enable new air-cured composites reinforced with lignocellulosic and synthetic fibers or only lignocellulosic fibers. The mechanical Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00003-6 © 2017 Elsevier Ltd. All rights reserved.

28

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.1 The tetrahedral interrelation of the main factors of the cement-based composite design.

properties of the fiber-reinforced cement composites depend not only on the performance of the fiber, but also on the degree to which an applied load is transmitted to the fibers by the matrix phase. Therefore, important to the stress transference aspect is the magnitude of the interfacial bond between the fiber and matrix phases. The interrelations of these fibers and matrix of the cement composite also depend on how it is processed. The effects of different processing methods on the rheological behavior of the cement paste with fibers in the fresh state, applying different types of fabrication facilities, e.g., slurry dewatering followed by pressing (forming an initial mixture with approximately 20% solids) and extrusion (forcing a Bingham pseudoplastic mixture through a die), strongly interfere on the interfacial transition zone. Interactions of fibers with cement matrix depend on interfacial properties or micromechanics parameters, such as chemical and frictional bond. Interfacial properties are also governed by the matrix Young’s modulus, fiber content and fiber diameter, length (aspect ratio), and fiber Young’s modulus. Therefore, differently from a synthetic fiber it is difficult to control the stress transfer between lignocellulosic fibers and matrix because of the effect of water absorption on dimensional stability, complex surface chemistry, and variability in diameter and strength along the length. In spite of the challenges and drawbacks of lignocellulosic fibers, they possess important advantages over synthetic counterparts within the context of a sustainable economy and innovative construction and are widely available in most developing and industrialized countries.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

29

This review analyzes some basic characteristics of the interfacial transition zone between particles in Portland cement paste with mineral additions and between cement paste and short random lignocellulosic fibers. The influence of the water and particle size distribution in the matrix, curing process, such as accelerated carbonation in the early curing, and the age of the composite, on the characteristics of the transition zone, is also evaluated and correlated to mechanical properties of the composite.

3.2

Aspects of the bulk cementitious matrix

The state of particle dispersion significantly affects the processing of particulate systems, has a marked effect on the fiber-cement suspensions and on the porosity of the final product and, thus, on the mechanical strength. Specially, particle size distribution is a very important parameter for processing many cement products, as powder packing can be tailored by selecting raw materials with suitable sizes and fractions.18 For this reason, the design of bulk cementitious matrix of a fibercement in which various raw materials are properly combined and adjusted, with different particle size range, is very much recommended. Fiber-cement commonly comprises a broad particle size distribution containing a large fraction of micro- and submicro-sized particles in combination with aggregates and agglomerates that are clusters of aggregates (Fig. 3.2). The driving force for the formation of aggregates and agglomeration as three-dimensional structures is caused by collisions between particles suspended in a fluid and by the action of an existing attraction force known as van der Waals. That is a force caused by the interaction between the permanent or induced electric dipoles present inside the particles (Fig. 3.3). This process is known as flocculation. In general, the main approach used for stabilization of the particles flocculation or dispersion is by means of chemical additives based on (1) electrostatic, (2) steric, and (3) electrosteric mechanisms.9,10

Figure 3.2 Schematic illustration of the difference between primary particle size, aggregates, and agglomerates.

30

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.3 Schematic illustration of the electrical double layer developed around charged particles in cement pore solution.7

Figure 3.4 Schematic particle size distribution: (A) the frequency of primary particles, aggregates, and agglomeration; (B) modification of the particle size distribution of a raw material during periods of mixture T1, T2, and T3.

The real particle size distribution is very complex to obtain because a raw material is constituted by primary particles, aggregates, and agglomerates that change over time depending on the process and chemical additive applied (Fig. 3.4). Another factor that can change the particle packing condition is the particle morphology. Most of the studies related to packing models commonly assume the spherical shape due to the intrinsic complexity of representing and controlling geometries of irregular particles.11,12 Besides, there are no specific procedures clearly defined in the literature to quantify the influence of particle morphology. Although there is a particle shape effect on packing density that decreases with the presence of more irregular particles, this is the same if spaces between the cement particles are partially occupied with smaller fine filler (Fig. 3.5).

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

31

Figure 3.5 The volume occupied by a stable particle structure with coarse irregular particles compared to the volume occupied by a stable particle structure with fine irregular filler (white small particles).

Figure 3.6 Schematic illustrations about packing cement particles (without scale): (A) the wall effect with initial particle size; (B) packing size particles, before and after hydration process.

However, the use of a spherical shape permits to understand some phenomena such as wall effect. The interfacial zone between fine and coarse particles in fibercement has been considered as an area of weakness, both in terms of packing and strength. This assumption is mainly associated with the nonorganized arrangement of fine particles at the region close to a much larger particle (Fig. 3.6A). The presence of such packing flaws can result in fiber-cement that are, for example, more susceptible to chemical attack due to their higher local porosity. The porosity can be decreased with the hydration products of a cement paste (a suspension of cement particles in water). The cement matrix is constituted of portlandite, ettringite, and mainly calcium silicate hydrate (CSH), that is the binding nonstoichiometric and semicrystalline phase in all cementitious materials, which is one of the most complicated and intriguing material systems, but is not enough to guarantee complete filling of the pores (Fig. 3.6B).

32

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

For this reason, it is necessary for a discrete particle size distribution from nanometers to millimeters scales to design a microstructure of a fiber-cement as indicated in Fig. 3.7. The characteristic behavior of nanoparticles and their reactive mechanism, with complex materials like Portland cement, have to be studied in detail in order to utilize the full effect of nanomaterials for the improvement of the performance of the construction materials. Several theoretical and experimental researches were carried out in order to predict the global volume occupied by spherical particles. Two different methodologies were mainly used by many researchers: a discrete distribution, considering the particles individually; or a continuous one, by using a combination of particles with wide range of sizes. However, despite these various improvements to understand particles packings, only a few models are commonly applied for the design structure of cement-based composites: Furnas, Andreassen, and Alfred.14,15 Alfred’s model, developed by Dinger and Funk,16 is more representative and combines Andreassen’s approach (but limiting the smaller particle size) and a mathematical review of Furnas’ distribution method. However, it is also necessary to take into account how these particulate raw materials are arranged during their processing step to achieve a maximum packing, a better interface between particles, and, consequently, minimize quantity of pores in the cement composites. Thus, knowledge about rheological behavior related to the processing method of cement paste is recommended.

Figure 3.7 Specific surface area versus particle size of raw materials used in fiber-cement production. Based on Santos, S.F.; Tonoli, G.H.D.; Mejia, J.E.B.; Fiorelli, J.; Savastano Jr., H. Nonconventional cement-based composites reinforced with vegetable fibers: a review of strategies to improve durability. Mater. Constr. 2015, 65 (317), 120. doi: 10.3989/mc.2015.05514.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

33

The rheology of slurries consists in the study of the deformation or viscous fluids flow from the application of a tension or external pressure. Particularly, it is known that the distribution of particles has also a strong influence on the rheological properties of cement paste,6,17 which can generate two groups of forces with different natures that are prevalent on the particles: gravity and surface forces. In the case of micrometric or nanometric particles with high specific surface area, surface forces predominate. For millimeter particles, with small specific surface area, the gravitational forces are prevalent.14 Moreover, cement paste presents a thixotropic phenomenon (apparent viscosity decreases with duration of stress) that relates to rheological behavior.18 Therefore, the field of rheology of folders is essential for their efficient homogenization and process optimization, mainly in the extrusion process, as will be discussed with more detail below. Some examples of particle morphology of mineral additions, such as sugar cane bagasse ash, metakaolin, and crystalline silica, used as filler and pozzolanic materials, are shown in Fig. 3.8. In the case of a formulation, the bimodal mixing

Figure 3.8 Morphological characteristics of the particles observed by scanning electron microscopy: (A, B) sugar cane bagasse ash; (C, D) metakaolin; and (E, F) crystalline silica.

34

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

behavior of nonspherical particles, in relation to packing density, is the same as that of spherical particles, but probably promotes lower values of apparent densities of the composites.11 The internal porosity of the particles is another complex factor in obtaining high density packaging. The particles may be totally dense, with closed internal porosity, or with open porosity. Further to the morphological effect, particles with open pores also affect the processing, since they have a greater capacity for the absorption of liquid phases through the pores (Fig. 3.8B). The particle size distributions of the OPC and some mineral additions are shown in Fig. 3.9. It is observed that 50% of the particles are smaller than 12 μm, 15 μm,

Figure 3.9 Discrete and cumulative distribution (CPFT, Cumulative Percent Finer Than) of Portland cement particles (A, B) limestone; (C) sugar cane bagasse ash; (D) metakaolin; and (E) crystalline silica.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

35

Figure 3.10 Particle size distributions of the reference formulations (RD) and sugar cane bagasse ash (CD), metakaolin (MD), and crystalline silica (SD): (A) discrete; (B) accumulated (CPFT).

140 μm, 22 μm, and 5.4 μm for the Portland cement, limestone filler, sugar cane bagasse ash, metakaolin, and crystalline silica, respectively. Most of the particles (90%) are less than 32 μm, 61 μm, 314 μm, 55 μm, and 16 μm, respectively. In addition, it was noted that particle distributions of the Portland cement and limestone filler are monomodal and similar. It was observed that the particle size distributions of bagasse ashes and metakaolin have a slight bimodal distribution profile. Fig. 3.10 shows different curves of particle distributions obtained by Alfred approach. Reference formulation (RD) has 70 vol% and 27 vol% of Portland cement and limestone filler respectively. The remaining 3 vol% was filled by lignocellulosic pulp. Other formulations are constituted with 40 vol%, 27 vol%, and 30 vol% of Portland cement, limestone filler, and mineral addition as partial replacement of Portland cement (sugar cane bagasse ash—CD; metakaolin—MD; and crystalline silica—SD) respectively. The bimodal profile of the CD curve can be easily identified. The distribution shows that more than 5% of the particles have equivalent diameter above 100 μm. On the other hand, the particle distribution profile of the SD formulation presents a higher amount of fine particles in relation to other profiles. Using fine and superfine particles is interesting for most cement-based composites. It is based on the assumption that in a cementitious composite the density is restricted by intergranular voids that are filled in with water instead of solid particles. The bimodal profile is also interesting because it increases the probability to fill voids due to a wide range of sizes. Thus, eliminating these empty spaces by successively adding fine particles, such as colloidal and amorphous silica, induces lower water consumption and higher workability.

3.3

Relevant aspects related to the lignocellulosic fiber

The use of lignocellulosic fibers, such as wood (eucalyptus and pines) and nonwood (sisal, green coconut, yam, bamboo, fique, hemp, flax, jute, curaua´, and ramie)

36

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

fibers, as reinforcement in cement-based composites has been studied to partially replace the synthetic counterparts, especially glass and polymeric fibers in construction materials.13,19,20 However, in addition to the problem of durability there is the issue of the interface between the fibers with the cementitious matrix. The mechanical characteristics of cement-based composites reinforced with fibers depend not only on the properties of the fiber, but also on the degree to which an applied load is transmitted to the fibers by the matrix phase. Furthermore, changes in the mechanical properties over time can occur due to microstructural changes in the fibermatrix interface and bulk, as a consequence of the continued hydration process that occurs in the fiber surroundings, i.e., if the interfacial transition zone becomes denser due to reprecipitation of cement hydration products in the porous layers. However, at the same time, increasing porosity of the cement composite in the interfacial zone occurs, as there is a fluctuation of the local water/binder ratio. The higher water/binder ratio close to the fiber leads to a reduction of the fibermatrix contact surface, which may result in lower frictional bond stress.21 In this way, there is an additional challenge for the fiber-cement: the hydrophilic nature of the lignocellulosic fibers that depends on surface porosity, lumens, roughness, surface chemical groups, and surface energy in the fibermatrix interface (Fig. 3.11). In the early curing, lignocellulosic fiber shrinks after significant moisture loss to the cementitious matrix, i.e., the surrounding cement paste (matrix) becomes a rigid structure and the fiber is partially debonded from matrix during the curing process. This phenomenon promotes the weakening of the interfacial transition zone between fiber and matrix. It is critical to control the fibermatrix interaction with lignocellulosic fiber, because, differently from synthetic fibers, such as glass and polymeric ones (Fig. 3.12A), which are fabricated with an almost unchanged shape (Fig. 3.12A), lignocellulosic fibers are nonuniform not only in their cross-section, but also along the longitudinal axis, as shown in Fig. 3.12B. The diameter range distributions are illustrated in Fig. 3.13. This depends on whether the fiber is taken from stems, stalks, or leaves, the climate and soil conditions at the plant location, the age of the plant, and the preconditioning of the fibers. Lignocellulosic fibers have significant variations in chemical constitution, diameter, and surface roughness resulting in the significant scattering in fiber mechanical properties. For example, with a nonuniform cross-section and composition along the longitudinal axis, the ultimate tensile strength values of the sisal fibers varied between 129 and 378 MPa, as shown in Fig. 3.14. For this reason, a statistical distribution of diameters and ultimate strength values must be used to analyze mechanical behavior of lignocellulosic fibers instead of average values, commonly applied in the case of synthetic fibers, as well as the method applied to measure area of fiber cross-section should always be informed in articles and technical texts.22 These characteristics of the bond between the lignocellulosic macrofiber reinforcement in cement-based materials still remain relatively unexplored. Thus, problems between lignocellulosic fiber interface and matrix must be analyzed differently from composites reinforced with synthetic fibers, although similar physical and mechanical parameters can be used. The interface can be defined as the

Figure 3.11 Relevant characteristics of vegetable fibers. (A) Schematic illustration of a lignocellulosic fiber with some surface characteristics and fibermatrix interface; (B) details of a yam fiber surface showing pores and microfibrils; and (C) cross-section of a curaua´ fiber showing irregular lumens.

Figure 3.12 Scanning Electron Microscopy (SEM) micrographs: (A) Polypropylene fiber surface; and (B) Sisal (Agave sisalana) fiber surface.

Figure 3.13 Frequency distribution of the equivalent diameters of the cross-section of two lignocellulosic fibers obtained from stereoscopy measurement: (A) Sisal; and (B) Green coconut.

Figure 3.14 Ultimate tensile strength frequency distribution of the sisal fibers.22

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

39

three-dimensional boundary between the fiber and the matrix. This interaction can occur through three mechanisms: mechanical coupling or micromechanical interlocking of the two materials, physical coupling such as van der Waals or electrostatic interaction, and covalent bonding (by way of a coupling agent) between the fiber and the matrix (see Fig. 3.15 for some examples). These interactions create an interphase region which is a three-dimensional region near to the fiber with properties different from either the fiber or the matrix.23 Silva et al.24 studied the effect of fiber shape and morphology on interfacial bond and cracking behaviors of sisal fiber cement-based composites. An experimental investigation was performed to understand the pullout behavior of sisal fibers from the cement matrix. In addition, the effect of curing age and fiber embedment length on the fibermatrix interface was investigated. It was found that the sisal fiber morphology plays an important role in the bond strength. The possibility of applications for the lignocellulosic fibers has stimulated the interest to modify their surfaces for specific applications. In the majority of the cases, the fiber surface modification is an essential requirement to improve the

Figure 3.15 Schematic depictions of various interactions at the fibermatrix interface: (A) micromechanical interlocking; (B) permanent or induced dipole interactions; (C) chemical bonding; (D) chain entanglement/fibrillation; and (E) adhesion by embrittlement of the fiber.23

40

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

interfacial compatibility between fiber and matrix (improving adhesion) and thus yielding enhanced mechanical performance of the fibercement interface.13,22,2528 In addition, new approaches on lignocellulosic fibers are used to minimize their hydrophilic nature in cement matrix, such as chemical and physical treatments and a thermal treatment named the hornification process.29,30 Furthermore, mechanical treatments also are applied as fibrillation of the fibers’ surface. Fig. 3.16 shows what happens in the fiber cell wall with refining/fibrillation or nanofibrillation.31,32 The main effect of refinement (refiner or grinding) in cellulose fiber structure is the fibrillation of the fibers’ surface because of the mechanical treatment. Fig. 3.17 shows what happens in the fiber cell wall after refining. The micro/nanofibrils were more exposed in the refined pulp (Fig. 3.17B) than microfibrils in unrefined fiber (Fig. 3.17A). This is a consequence of the mechanical treatment promoted in the fiber surface during refinement. The refining energy is sufficient to partially rupture the bonds between microfibrils, leading to a more fibrillated surface. In fibercement composites, the higher the fibrillar surface of fibers, the higher their capacity to bond with cement matrix.13 These fibrillated and short fibers are responsible for the formation of a net inside the composite mixture with the consequent retention of cement matrix particles during the dewatering stage of the manufacturing process. Better fibermatrix interface adhesion and mechanical performance can be achieved by increasing the specific surface area of the fiber, by reducing the fiber diameter, and by promoting a rough surface with better mechanical anchorage in the matrix.13 Macrofibrils and microfibrils are constituted of cellulose nanocrystals (also reported in the literature as cellulose whiskers), nanofibrils, cellulose crystallites, or crystals that are considered as the crystalline domains of lignocellulosic fibers, and can be isolated by several chemical and mechanical methods (Fig. 3.18).

Figure 3.16 Schematic illustration: (A) unrefined; (B) refined; and (C) nanofibrillated fibers. Based on Afra, E.; Yousefi, H.; Hadilama, M.M.; Nishino, T. Comparative effect of mechanical beating and nanofibrillation of cellulose on paper properties made from bagasse and softwood pulps. Carbohydr. Polym. 2013, 97, 725730. doi: 10.1016/j.carbpol.2013.05.032.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

41

Figure 3.17 Typical scanning electron micrographs of: (A) unrefined (Canadian Standard Freeness, CSF 600 mL); and (B) refined (CSF 130 mL) Eucalyptus bleached fibers.13

Figure 3.18 (A) The molecular structure of cellulose, with glucose molecules alternately rotated 180 (solid line, covalent bonds; dashed line, hydrogen bonds); (B) cellulose microfibrils, with both crystalline and noncrystalline regions, aggregated into a macrofibril.33

42

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.19 Scanning Transmission Electron Microscopy (STEM) of unbleached bamboo nanofibrillated cellulose pulp after 10 cycles by mechanical nanofibrillation method.32

Fig. 3.19 shows nanofibrillated unbleached bamboo organosolv pulp prepared using the mechanical nanofibrillation method. In this method, the cellulose pulp was passed between a static grindstone and a rotating grindstone revolving at 1700 rpm. A commercial grinder, Supermasscolloider Mini, model MKCA 6-2, with two grinding stones of aluminum oxide (Al2O3), model MKGA 6-80#, produced by Masuko Sangyo Co., Ltd., was used.32

3.4

Fibermatrix interactions

The homogeneous distribution of fibers in cement paste in the fresh state depends on the workability or rheological properties that change during processing due to the aspect ratio and the hydrophilic nature of lignocellulosic fiber, microfibrils, and thixotropic behavior of the cement paste.34 Microfibrils have a higher specific surface area, and consequently, can promote a better mechanical anchorage in the cementitious matrix without the effect of the drying shrinkage of fibers. In a composite with a distribution of fibers, micro-, and nanofibrils the fracture process occurs at different scales in the microstructure promoting unique mechanical characteristics caused by the fiber-bridging and pullout action that occurs across fine multiple microcracks and cracks (Fig. 3.20). The bridging action contributes to the composite’s toughness by activating the fibermatrix interface where energy is dissipated through the debonding of the interface and fiber pullout.36

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

43

Figure 3.20 Schematic fracture process in a fiber-cement microstructure with different scales and mechanisms: (1) debonding; (2) bridging; (3) pullout; and (4) fractured fiber. Based on Coutts, R.S.P. Wood Fiber Reinforced Cement Composites. In Natural Fiber Reinforced Cement and Concrete; Swamy, R.N., Ed.; Blackie: Glasgow, 1988, pp. 208242.

The composite properties of the interface between the synthetic fibers and cement matrix have been examined in many studies available in the literature. For example, Yang et al.37 developed a strength-controlled model that is based on the interfacial bonding performance. The interfacial bonding is characterized by chemical and frictional bond strength. Single fiber pullout tests are carried out to simulate the interfacial behavior of the fiber-cement composite matrix. However, there is little research that investigates or presents models of interfacial transition zone between lignocellulosic fiber and cement matrix. As discussed above, there are several characteristics in the cementitious matrix associated to the particles size distribution, packaging of particles, rheological behavior of the cement paste in the fresh state to the hydration process, that by themselves increase the complexity of the microstructure of a composite. Introducing fibers and nanofibers makes the microstructure of a fiber-cement more complex. Fig. 3.21 illustrates packing of cement particles with cellulose microfibrills and lignocellulosic fiber, before and after the hydration process. Degradation process occurs when swelling/drying of the cellulosic fibers develop stress at interface regions leading to microcracking mechanism in the matrix around swollen/dried fibers and this promotes capillarity and fluids transportation via microcracks, i.e., the shrinkage of the fiber pulp generates microcracks between the fiber and the cement matrix, with consequent weakness in the interfacial transition zone. However, it is well known that, during the curing process, the hydration products form a shell around the unhydrated cement particle, slowing down the diffusion of water to its interior. This phenomenon limits the hydration rate and, as a result, the cores of the cement particles hydrate slowly. When nanofibrillated cellulose is present in the cement paste, it initially adheres to the cement particles and remains in the hydration product shell (i.e., the high density CaOSiO2H2O (CSH)). It could form a path to transport water from the external pore to the inner unhydrated cement core. This may facilitate a larger portion of cement reacting in comparison

44

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.21 Schematic illustration about packing cement particles (without scale) with cellulose microfibrils and lignocellulosic fiber, before and after the hydration process.

with the cement pastes without nanofibrillated cellulose. The mechanism of water molecules diffusing along the nanofibrillated pulp networks in the hydration products shell is referred here as short-circuit diffusion.34 Thus, there is in the cementitious microstructure reinforced with lignocellulosic pulp and nanofibrillated cellulose, a balance between both phenomena: formation of microcracks and formation of CSH in the interfacial transition zone.

3.5

Processing and curing methods

This section discusses the influence of the mineral additions, two techniques of processing: slurry dewatering followed by pressing and extrusion; and carbonation accelerated curing in the interfacial transition zone between lignocellulosic fibers and matrix. First, there is a comparison between different mineral additions to evaluation the effects on the cement composites using slurry dewatering followed by pressing as main process. Second, there is a discussion of the extrusion process and rheological behavior of the fresh cement paste with and without cellulose pulp as well as with PP or curaua´ fibers using the BenbowBridgwater model and ram extruder. Third, a comparison will be made between processes: slurry dewatering followed by pressure versus extrusion. Finally, there is a brief outline of some research about accelerated carbonation for curing and its influence on the interfacial transition zone.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

45

3.5.1 Slurry dewatering followed by pressing The mix design for the composites (Table 3.1) and particle size distributions shown in Fig. 3.10 was established based on commercial formulations used in the fibercement industry reinforced only with Eucalyptus pulp. The slurry dewatering followed by pressing technique was carried out, as an approximate reproduction of the Hatschek process applied in the industrial production of fiber-cement (Fig. 3.22). For the preparation of the fiber-cement pads, eucalyptus cellulosic pulp was previously dispersed in water by mechanical stirring at 3200 rpm for 5 min. Then, Portland cement, limestone filler, and other mineral additions were added, forming an initial mixture with approximately 20% solids, and the mixture was stirred at 1000 rpm for an additional 5 min. The slurry was transferred to an evacuable casting box, where vacuum (approximately 80 kPa gauge) was applied until a solid surface was formed. The nominal dimensions of the pad used here were approximately 200 mm 3 200 mm 3 15 mm. The pad of each formulation was pressed at 3.2 MPa for 5 min, and then sealed in a plastic bag

Mix design of the fiber-cement composites reinforced with Eucalyptus pulp

Table 3.1

Raw material

Unrefined unbleached Eucalyptus pulp (CSF 664 mL)a Ordinary Portland cement (OPC)—ASTM C150, Type I38 Limestone filler Sugar cane bagasse ash (CD) Metakaolin (MD) Crystalline Silica (SD)

Content (% by mass) RD

CD

MD

SD

3 70 27   

3 40 27 30

3 40 27

3 40 27

30 30

a

The Canadian Standard Freeness (CSF) test is a widely recognized standard measure of the drainage properties of pulp suspensions and relates well to the initial drainage rate of the wet pulp pad during the dewatering process.

Figure 3.22 Slurry dewatering followed by pressing molding technique. (1) A slurry with approximately 20% by mass of solid materials; (2) casting box placed under an initial vacuum for removal of excess water, forming a solid surface; (3) the moist pad was tamped flat; (4) the vacuum was reapplied; and (5) the pad was then transferred to an oiled steel plate, pressed at 3.2 MPa.

46

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

to keep it wet and held at room temperature for 2 days and subsequently immersed in water saturated with cement for 26 days, with the goal of completing most of the hydration reactions. Mechanical tests were performed in equilibrium with the temperature and air humidity of the laboratory about 20 days after curing for nonaged composites, using a servo-hydraulic mechanical testing machine MTS model 370.02 controlled by MultiPurpose TestWare System. Prismatic specimens were prepared using a diamond saw blade refrigerated by water, prior to grinding and final polishing of the specimen sides, using alcohol as lubricant, and having nominal dimensions of 12 mm x 16 mm x 80 mm. The three-point bending test configuration with span of 64 mm, eight specimens and specific type of specimen and parameters of test were used for modulus of rupture (MOR) and energy of fracture determination. The MOR was calculated using a cross-head speed of 5 mm/min. The fracture energy was obtained to evaluate the influence of toughening mechanisms promoted by difference of particle size distribution (Fig. 3.10) and lignocellulosic fibers, such as pullout and bridging, on mechanical performance of the composites. The fracture energy test was performed with the SENB-type specimen. The specimens with a centered flat notch with 30% of specimens height and notch tip in “V” with angle of about 30 were prepared using diamond disc of 0.5 mm thick. A cross-head speed of 10 μm/min was adopted and controlled by the actuator displacement to guarantee stable growth of the crack.20 The work done by the machine to completely propagate the crack along the specimen divided by two times the projected area of the fracture surface (cross-section of the specimen) was used to obtain the fracture energy, γ WoF. The integration of the forcedisplacement curve was made up to the point where the force decreased to 5% of its maximum value reached during the test. The water to binder ratio is shown in Fig. 3.23, which indicates the amount of water retained in the system. The MOR (Fig. 3.24) represents the combined effects of the stress distribution at the fibermatrix interfaces and the mechanical strength of the matrix. Fig. 3.25 shows the average values of fracture energy (γ WoF) of the matrices and indicate different fibermatrix interfaces. The pullout and bridging mechanisms consume a greater amount of energy, because of the shear stresses generated at the fibermatrix interface until the composite fractures completely. Therefore, in the matrix with mineral additions, the brittle behavior was minimized and there was a better balance between the fibermatrix adhesion strength.

3.5.2 Extrusion process Extrusion is a forming process consisting of forcing a viscous mixture through a die.39 The extruder allows for the manufacture of building elements with mechanical and physical characteristics similar or even better than those from other processes, such as slurry dewatering followed by pressing technique.3944 The extrusion of ceramic materials represents part of the overall production process of ceramic products. It comprises all essential process-specific steps such as feeding,

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

47

Figure 3.23 Water to binder ratio of cement matrix with eucalyptus pulp of mix design with OPC and limestones (RD), sugar cane bagasse ash (CD), metakaolin (MD) and crystalline silica (SD).

Figure 3.24 Average values and standard deviations of the modulus of rupture (MOR) of the reference composite and composites with mineral additions.

Figure 3.25 Average values and standard deviations of the fracture energy (γ WoF) of the reference matrices and with mineral additions.

48

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

compacting, flowing, and pressure build-up. This method displays versatility in the preparation of varied cross-sectional configuration building components, not limited to plates, as in many conventional processes. It can be applied to produce cementitious composite as tiles, plates, and tubes.4044 According to Zhou and Li,45 the extruders used in the ceramic industry are adaptable for the extrusion process of fiber-cement materials. Furthermore, the extruder enables the use of synthetic and natural fibers as reinforcement. The patent of Shah et al.43 demonstrates the potential of the extrusion process to produce more resistant fiber-cement composites. The patent publication in 200946 describes a new method and apparatus for extruding fiber-cement with cellulosic pulps. However, the main information on the variables of the extrusion process and raw materials used are protected. In addition, the improvement and optimization of these technologies still require consistent trials due to the number of variables involved and the aspects related to the durability of the final product. Therefore, the technical implementation and consolidation of the extrusion process of fiber-cement materials in the productive sector can enable the establishment of industry in sparsely populated regions due to the low cost of the production line. Furthermore, the extrusion process allows the flexibility to adapt this production line in low-capacity industry, in order to meet the small markets. The barrier of high investments for the implementation of fiber-cement industries would be overcome with the viability of the extrusion process, thus allowing the arising of new suppliers in the market, making it more competitive. Therefore, such a scenario would benefit consumers due to fierce competition, with better and more diversified products. Additionally, it increases the probability of the use of alternative raw materials, such as fibers and vegetable materials, and lignocellulosic residues, in order to minimize their costs.

3.5.2.1 Rheology of extruded mixtures Rheology is the study of the deformation and flow of matter and it deals with the flow of complex fluids, because it continuously deforms when there is an applied stress. The rheological behavior interferes strongly in the interfacial transition zone between fiber and matrix. There are two aspects, the yield stress (which is the minimum stress that has to be applied for material flow) promoted by shear stress (τ) and the viscosity (η).47 Viscosity is the measure of the internal friction of a fluid (Fig. 3.26). The application of rheological concepts in cement paste is of fundamental importance to allow easy molding and mitigation of defects, minimizing the occurrence of cracks, detachment, and stains that may result from inappropriate rheological behavior for the application.48 The molding processes applied differ from each other here with respect to the rate of molding, the molding forces as a function of the water content and the proportion and type of material, the geometry of the product, the necessary complexity of the machine, and the application of additives.49,50 Therefore, it is necessary to use additives to ensure the convenient rheological behavior that allows the hydration of cement, due to the necessity of using reduced amounts of water, avoiding the phase migration.39,49

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

49

Figure 3.26 (A) Representation of the viscous flow of a volume of material in response to an applied shear force. If there are two plates (area, A), separated by distance (separation height, y), viscosity can be defined as the ratio of the applied shear stress and the change in velocity dυ with distance dy in a direction perpendicular to and away from the plates. (B) The yield stress is the applied stress necessary to exceed a resistance in order to make a structured material flow.

Figure 3.27 Basic classification and representation of rheological behavior according to shear stress as a function of shear rate (A) and time (B) Based on Oliveira, I.R.; Studart, A.R.; Pileggi, R.G.; Pandolfelli, V.C. Dispersion and Packing of Particles—Basic Principles and Applications to Ceramic Processing, 1st edition; Fazendo Arte Editorial, 2000; 224 pp (in Portuguese).

Many constitutive models have been developed to represent the relationship, among which the Bingham model and HerschelBulkley models are the most commonly used for cement-based flow because of their simplicity.18,51 At the extrusion process, cementitious paste presents a Bingham pseudoplastic behavior (Fig. 3.27A). This rheological behavior of cementitious mixture is extremely relevant to allow its extrudability.52 In addition, cement paste has a thixotropic behavior (Fig. 3.27B).4 The definition of thixotropy can be described as the following: the decrease of viscosity with time

50

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

by application of shear and the recovery of viscosity when the material is at rest. The basic physical description for thixotropical behavior cannot fully indicate the nature of thixotropical behavior of cementitious materials since the hydration of cement will result in different phases and states of cementitious materials. The development of thixotropy of cement paste is due to the changes of structure from one phase to another phase.18,48 When Portland cement is mixed with water, each phase dissolves at least partially, leading to the establishment of a supersaturated solution with respect to different hydrates, which can precipitate.53 The thixotropy of cementitious materials can be considered as the agglomeration of particles when shearing is not applied to the cement particles. Once the external shearing is applied into the paste, the particles will be separated. It is noted that the reversible behavior of agglomeration, separation, and aggregation of cement particles contribute to thixotropic behavior of cement paste.18 Besides yield stress and viscosity, thixotropy, a property showing the time-dependent change in viscosity, is also observed and used for characterizing concrete rheological behavior.51 The rheology of the fiber-cement mixture for extrusion comprises liquid (continuous) or water and binder ratio, particulate (solid), and the reinforcing phase (fiber), which strongly depends on the morphology of the particles, particle size distribution, chemical characteristics and specific surface area of the raw materials, volume fraction and length of the fibers, as well as the degree of homogeneity of the mixture.50,52 Specially, the quantity of water influences the flow of the cement paste with fibers in the extrusion process because it wets the surfaces of the particles and reacts chemically with the cement. Besides, rheological modifiers are necessary to interfere in the rheological behavior of the cement paste.49,54 Many studies have demonstrated the benefits of adding fibers to cement-based materials. On the other hand, the fibers are presented mainly as an obstacle to the movement of the other two phases (water and particles) in the extrusion process. The workability is decreased due to the aspect ratio and the hydrophilic nature if the fiber is lignocellulosic. However, the slim shape of fibers causes them to rotate and to orientate during the flow, thus the fibers remain predominantly aligned in the direction of flow of the mixture in the extrusion process; it will facilitate the movement of particulates, according to Fig. 3.28.54 In fact, to evaluate the extrusion process, there are some techniques that can collaborate for the assessment of rheological behavior of the fresh cement paste with fibers, such as BenbowBridgewater model, capillary rheology,55,56 squeeze flow,57 and rheometers or viscometers.4,58 Fibers are often highly concentrated in cementitious suspensions. The interaction among fibers and of fibers with particles (fillers, mineral additions, cement) determines the kind of network they form. In flowing suspensions, three kinds of forces coexist to various degrees: forces of colloidal origins (interactions between particles resulting in repulsion or attraction); Brownian randomizing force, which ensures that particles (mainly smaller than 1 μm) are in constant movement; viscous forces acting on particles, which are proportional to the local velocity difference between the particle and the surrounding fluid. Fiberfiber interactions include hydrodynamic effects as well as mechanical contacts.51,59

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

51

Figure 3.28 Movement and orientation of the fibers in the extrusion process. Based on Srinivasan, S.; Deford, D.; Shah, P.S. The use of extrusion rheometry in the development of extrudate fiber-reinforced cement composites. Concr. Sci. Eng. 1999, 1 (11), 2636.

The fibers contribute to the formation of an internal network in fiber-cement in fresh state and increase the yield stress (Fig. 3.26B). A Bingham pseudoplastic behaves like a solid below the yield value; when the stress exceeds the yield value it starts to flow (Fig. 3.27A). The flow curve has been reported to fit several different mathematical forms, all of which indicate the existence of a yield stress.4,51 During the flow, mechanical contacts between fibers and fibers and also between fibers and particles (aggregates and agglomerates) increase the resistance to flow and the viscosity. However, during the mixture an external shearing is applied into the paste, the particles will be separated contributed to thixotropic behavior of the cement paste; at the same time fibers contribute to increase yield stress. Few studies have been evaluated in the literature on the effect of synthetic and lignocellulosic fibers, pulp, and nanofibrils on rheological characteristics of cementitious materials. However, it is known that the fiber volume (same fiber type), the aspect ratio, and the fiber diameter influence rheological characteristics. The ram extrusion mechanism is represented by the BenbowBridgwater model, which is based on the assumption of plastic deformation in die-entry and plug flow in die-land. As illustrated in Fig. 3.29A, the total extrusion pressure drop (Pe) through a circular die-entry can be calculated by the equation proposed by Benbow et al.60 According to model, the pressure, P, applied in the fresh cement paste to pass cylindrical dies is calculated by Eq. 3.1.55,60,61 P 5 2ðσ0 1 αV Þln

    D0 L 1 4ðτ 0 1 βVÞ D D

(3.1)

where D0 and D are the extruder barrel diameter and die diameter, respectively, and L is the die-land length. The barrel inner diameter D0 and its die inner diameter D. V is the velocity of the paste in the die-land which, due to slip at the wall of the die-land, is a constant independent of radius. σ0, α, τ 0, and β are parameters which describe the rheology of the paste. σ0, initial bulk yield stress, and α, velocity dependent factor for convergent flow, are associated with the movement of the

52

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.29 (A) Representation of BenbowBridgwater ram extrusion mechanism. (B) Details of ram extruder. Adapted from Benbow, J.J.; Jazayeri, S.H.; Bridgware, J. The flow of paste through dies of complicated geometry. Powder Technol. 1991, 65, 393401. (A Special Volume Devoted to the Second Symposium on Advances in Particulate Technology).

paste as it flows suffering extension as it passes from the barrel to the entry of the die-land. τ 0, initial wall shear stress, and β, velocity dependent factor for parallel flow, are associated with the resistance to flow in a duct of constant cross-section. The rheological behavior can be tested by preparing a variety of cement pastes and then studying the relationship between pressure drop P and velocity V during a ram extrusion experiment using a ram extruder (Fig. 3.29B). In the ram extruder the die-land length to diameter ratios (L/D) can be changed to 1, 4, and 8, for example, as well as the velocity (V) of cement paste in the die-land. Ram extruders can be employed as a form of rheometer to provide a means of characterizing the bulk and interfacial rheology of paste-like materials tailored for extrusion. The influence of cellulosic pulp in the rheological behavior of the cement paste by means of ram extruder was investigated. The mix design used in this study is presented in Table 3.2. The same water to binder ratio about 0.33 was used. Hydroxypropyl methylcellulose and carboxylate polyether were added as rheological modifiers. Therefore, for each mix design the load in ram extruder was recorded as a function of time and ram displacement. For each die-land length, ram velocities of 0.05 mm/s, 0.08 mm/s, 1.2 mm/s, and 1.8 mm/s were used. These ram velocities correspond to respectively the average extrusion velocity of 6.5 mm/s, 8.9 mm/s, 13.3 mm/s, and 20 mm/s in the die-land, according to Srinivasan et al.54

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

Table 3.2

53

Mix design used in the production of fresh cement-based

pastes Raw material

Portland cement [ASTM C150] Limestone Bleached Eucalyptus pulp

Content (% by mass) Matrix

Matrix 1 Cellulosic pulp

70.0 30.0 

70.0 27.0 3.0

Figure 3.30 Influence of the cellulosic pulp on extrusion pressure as function of the extrusion velocity.

In fact, it is known that cellulose and its derivatives (ether and ester) are widely used as viscosity modifying agents for the production of self-compacting concretes, dry set mortars, and pastes, among other applications.62,63 Fig. 3.30 indicates that the cellulosic pulp acts as rheological modifier decreasing the pressure to extrudate fresh cement paste. The influence of the synthetic and lignocellulosic fibers on rheological behavior of cement paste was also investigated. Portland cement, limestone, cellulosic pulp, curaua´, or PP fibers, both with a monodispersed distribution of length of 10 mm, and 1 wt% of hydroxypropyl methylcellulose by dry mass of particulate materials were mixed and homogenized at low speed (mixture distributive), in a mechanical intensive mixer (capacity of 10 L) for 5 min. After this stage, water and carboxylate polyether were added fractionally for 2 min. All raw materials were mixed at high speed for another 5 min for high shear mixing to break down the agglomerates generated during the wet mixing stage. The mix design used in this study is presented in Table 3.3. In the ram extruder it was used the average extrusion velocity of 5.9 mm/s, 8.9 mm/s, 13.3 mm/s, and 20 mm/s in the die-land.

54

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Mix design used in the production of fresh cement-based pastes without and with PP or curaua´ fibers by extrusion

Table 3.3

Raw material

Portland cement (ASTM C-150) Limestone Bleached Eucalyptus pulp Polypropylene fiber (PP) Curaua´ fiber Water/Binder ratio

Content (% by mass) Reference

PP (10 mm)

Curaua´ (10 mm)

69.95 27.08 2.98   0.33

68.55 26.54 2.92 2  0.33

67.79 26.54 2.89  3.08 0.36

Figure 3.31 Influence of the cellulosic pulp and sisal fibers on extrusion pressure as function of the extrusion velocity.

Fig. 3.31 shows that the pressure of the cement suspension with fibers is determined by the nature of interactions between the fibers and fibers and particles, which depends mainly on the fiber concentration, chemical and physical characteristics, and aspect ratio. In both systems the results show that at sufficiently velocity levels, the cement paste viscosity tends to reach a steady-state. However, there is a clear difference between the suspensions with PP or curaua´ fibers. The curaua´ fibers increase viscosity and yield stress of the suspension because of its high quantity of lumens (Fig. 3.11C) that facilitates the sequestration of the water utilized for lubricating the fine particles in the cement paste flow. Besides, the roughness and diameter of the curaua´ fibers increase the fiberfiber interaction (collisions) due to the presence of nonlongitudinally oriented fibers along the barrel wall that causes a larger network structure formed by adhesive forces, mechanical interaction, and interlocking during the flow. Consequently, it increases the work required to align

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

55

Figure 3.32 Scanning electron microscopy (SEM) with the backscattered electron image (BSEI) mode of the extruded cement pastes reinforced with (A) curaua´ or (B) polypropylene (PP) fiber.

the fibers at the entrance of the die-land and interferes in the interfacial transition zone between fiber and matrix of the hardened composite (Fig. 3.32). After ram extruder test, it is noted that the interfacial transition zone between curaua´ and cement matrix fiber has more porosity than with PP fiber (Fig. 3.32). Among many factors that interfere in the interfacial transition zone between fiber and matrix, such as characteristics with respect to the water/binder ratio of a fresh cement paste, the degree of chemical bond strength of the fibers, and the type of fiber, part of the porosity is promoted by the processing method.

3.5.2.2 Assessment of the processes: Slurry dewatering versus extrusion Processing effects in cementitious composites were evaluated by some researchers.56,64 However, the objective of this section is to compare the mechanical performance of matrices, with the same formulation and particle size distribution (Table 3.1 and Fig. 3.10). The cement-based composites were fabricated by slurry dewatering (RS) and by extrusion (RE) processes. The water soluble polymers, hydroxypropyl methylcellulose with 86,000 average molecular weight and 5.39 cP viscosity (at 2% concentration in water at 20 C), provided by Aditex and high range water reducer polyether carboxylic (commercially named ADVA 190 and provided by Grace) were used as rheological modifiers to promote pseudoplastic behavior of the composite. Each additive was applied in the proportion representing 1% of the total mass of the particulate raw materials, and was required to enable the extrusion process. The mixture was homogenized in a mechanical Amadio planetary mixer (capacity of 20 L) during 5 min at low speed (125 rpm), 5 min at medium speed (220 rpm), and finally 5 min at high speed (450 rpm). The mixture was transferred to a Gelenski MVIG05 laboratory extruder (Fig. 3.33). The linear speed of the extruder was approximately

56

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.33 (A) Front view of the extruder. (B) Side view illustrating the specimen outlet through the die. (C) Superior view of the extruded specimens on a metal plate.

4 mm/s and cross-section die width/height ratio was 3.3 (50 mm 3 15 mm). The mixture was recirculated into the extruder for 5 min before tailoring the samples. Samples with nominal dimensional of 15 mm 3 50 mm 3 200 mm were extruded and immediately transferred to the steel plates for hardening and initial curing. The water to cement ratio achieved in the slurry dewatering process was 0.34 and by extrusion process it was 0.32. The composites were characterized after 28 days of age. Fig. 3.34 shows the microstructure of the composites processed by slurry dewatering (RS) and by extrusion (RE). It was observed qualitatively in the micrographs that little difference was found between the both microstructures. In fact, Fig. 3.35 shows that there is no significant statistical difference between the average values of the MOR of the fiber-cement RS and RE. The results indicate that the MOR test was not very sensitive to determine the differences of the stress transfer from the matrix to the fibers. Unlike MOR test (cross-head speed of 5 mm/min), Figs. 3.36 and 3.37 indicate results obtained of the fibermatrix interactions during the quasistatic fracture process (cross-head speed of 10 μm/min) in the fiber-cement. These interactions are strongly dependent on the physical and chemical adhesion, shear stress resistance, and mechanical anchorage induced by deformations on the fiber surface or by the overall complex geometry of the cellulosic fiber. In addition, the fibermatrix interactions in the composites also depend on randomly oriented fibers, i.e., the effect of fiber orientation. As fracture energy test (γ WoF) is executed sufficiently slow, the average value represents the average of the set of all the processes of debonding and pullout of the fibers into composites during the fracture. These results suggest that extrusion process collaborates to organize better microstructure of the fiber-cement than slurry dewatering process.

Figure 3.34 Scanning electron microscopy (SEM) with the backscattered electron image (BSEI) mode of the composites processed (A) by slurry dewatering (RS) and (B) by extrusion (RE).

Figure 3.35 Average values and standard deviations of the modulus of rupture (MOR) of the composites processed by slurry dewatering (RS) and extrusion (RE).

58

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.36 Typical loaddisplacement curves obtained from the fracture energy tests of the composites processed by slurry dewatering (RS) and extrusion (RE).

Figure 3.37 Average values and standard deviations of the fracture energy (γ WoF) of the composites obtained by slurry dewatering (RS) and extrusion (RE).

3.5.3 Accelerated carbonation for curing The interaction between carbon dioxide and hydrated Portland cement at atmospheric pressure and ambient temperature conditions is a relatively well-known phenomenon, according to the diagram in Fig. 3.38. However, it becomes more complex in fiber-cement system because of lignocellulosic fibers. The development of a comprehensive model that describes the variation of composition of phases in a fiber-cement system submitted to accelerated carbonation

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

59

Figure 3.38 Diagram with principal mechanisms for accelerated carbonation in Portland cement.20

during hydration is considered a challenge for researchers working in this subject, since it must consider several parameters that influence the equilibrium of each phase and the interaction thereof, as shown in Fig. 3.39. It is known that nonuniform cross-sections of lignocellulosic fibers also promote a larger volume of capillary pores in the fiber/matrix interfacial zones that are considerably different from that of the bulk matrix, away from the interface. Accelerated carbonation at early age of the fiber-cement is an alternative route to partially mitigate the problem with interfacial transition zone between lignocellulosic fiber and cement matrix with filling of CaCO3 content and, consequently it decreases porosity, promotes higher density in the interface, guaranteeing a good fibermatrix adhesion.19,20,66 This approach corroborates for long-term durability issues of lignocellulosic fiber-cement and also in order to make fibercement composites more stable since their initial ages under different humidity conditions. Almeida et al.19 evaluated the effects of accelerated carbonation in the early stages of hydration of eucalyptus pulp reinforced cementitious composites produced by the slurry dewatering technique. Accelerated carbonation was applied after 2 days of controlled curing, which was investigated aiming to find a durable composite with lignocellulosic pulp as an exclusive reinforcement. The effect of carbonation curing on the mechanical, physical, and microstructural properties of composites at 28 days of age, after 200 and 400 accelerated ageing cycles and one

60

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.39 Diagram with the principal factors to promote accelerated carbonation in fibercement composites. Based on Ferna´ndez Bertos M.; S.J.R. Simons; Hills C.D.; Carey P.J. A review of accelerated carbonation technology in the treatment of cement-based materials and sequestration of CO2. J. Hazard. Mater. 2004, B112, 193205. doi: 10.1016/j. jhazmat.2004.04.019.

year of natural weathering was evaluated (Fig. 3.40). According to researchers, mechanical properties were better for the composites subjected to accelerated carbonation at early stages of hydration and these properties were maintained after accelerated and natural aging, indicating their improved durability. These results were associated to a lower porosity because of the densification in the interface between fibers and matrix by the higher precipitation of CaCO3. Cracks and pores are observed in the cement matrix surrounding both the cellulose and PVA fibers in noncarbonated composite (Fig. 3.41A).67 Pizzol et al.67 showed that the CaCO3 formed from the carbonation reaction is precipitated in the pore structure of the matrix, filling the voids, improving the contact between fibers and cement matrix and thus preventing penetration of the water into the pores (Fig. 3.41B). This result indicates that the carbonation rate of CSH and calcium sulfoaluminates occurred in a high extent than the other crystalline phases. The CSH gel forms and is progressively decalcified, being converted to CaCO3 and SH (silicate hydrates).20,67 Santos et al.20 investigated the effects of the supercritical carbonation, after precuring for three days, on extruded fiber-cement reinforced with bleached Eucalyptus pulp and residual sisal fibers. The supercritical carbonation curing in the initial age led to lower porosity in the composites and, consequently, to higher bulk density, which improved the microstructure by sealing the opened pores with calcium carbonate in the vicinities of the lignocellulosic fibers. The lower porosity of the carbonated composite is kept after 200 soak and dry accelerated aging cycles

Figure 3.40 Average values of modulus of rupture (MOR) versus apparent void volume.19

Figure 3.41 Scanning electron microscopy (SEM) with the backscattered electron image (BSEI) mode of the cut and polished surface of the fiber-cement composites: (A) noncarbonated and (B) carbonated after 10 h of accelerated carbonation. Arrow 1: cracks and pores around the fibers. Arrow 2: improved interface between fibers and matrix. Arrow 3: mineralized cellulose fiber. Based on Pizzol, V.D.; Mendes, L.M.; Frezzatti, L.; Savastano Jr., H.; Tonoli, G.H.D. Effect of accelerated carbonation on the microstructure and physical properties of hybrid fibercement composites. Miner. Eng. 2014, 59, 101106. doi: 10.1016/j.mineng.2013.11.007.

62

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 3.42 Discrete pore size distribution with curves of extruded fibercement carbonated and carbonated, submitted to supercritical CO2: (A) before; and (B) after exposure to 200 soaking and drying cycles (200c). Based on Santos, S.F.; Tonoli, G.H.D.; Mejia, J.E.B.; Fiorelli, J.; Savastano Jr., H. Nonconventional cement-based composites reinforced with vegetable fibers: Aa review of strategies to improve durability. Mater. Constr. 2015, 65 (317), 120. doi: 10.3989/mc.2015.05514.

(Fig. 3.42). Besides, after accelerated aging, the average values of fracture energy of the carbonated and noncarbonated composites decrease approximately 28 and 56%, respectively.

3.6

Concluding remarks

The use of fibers in engineered composites and new materials for multipurpose applications is very common. However, according to the works of various researchers, the presence of reinforcement, mainly lignocellulosic fibers, in cement paste creates a weak interfacial transition zone, i.e., the formation of porosity surrounding the fibers. The challenge is to control simultaneously numerous factors during the production of a cement-based composites that affect the interfacial transition zone such as the particles size distribution, water/binder ratio, rheological properties in the fresh state, shape and diameter distribution of the particles, aspect ratio and type of the fibers, curing process, and type of production process. These factors must be known and evaluated before the elaboration of a mix design and before choosing a process and type of curing. Based on discussions in the present review, the following conclusions can be made: G

Particle size distribution, as well as the zeta potential, is a very important parameter for processing many cement products, as powder packing can be tailored by selecting raw materials with suitable sizes and fractions. The interfacial zone between fine and coarse particles in fiber-cement has been considered as an area of weakness, both in terms of packing and strength. For this reason, it is necessary for a discrete particle size distribution from nanometers to millimeters scales using a diversity of mineral addition when designing a microstructure of a fiber-cement.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

G

G

G

63

The use of lignocellulosic fibers, such as wood (eucalyptus and pines) and nonwood (sisal, green coconut, yam, bamboo, fique, hemp, flax, jute, curaua´, and ramie) fibers, as reinforcement in cement-based composites has been studied as a partial replacement of the synthetic fibers. However, lignocellulosic fibers are nonuniform not only in their cross-section, but also along the longitudinal axis. This fact means that it is difficult to pack these fibers in the microstructure. Thus, a statistical distribution of diameters and ultimate strength values must be used to analyze mechanical behavior lignocellulosic fibers instead of average values, commonly applied in the case of synthetic fibers. It is also important to mention that the method applied to measure area of fiber cross-section must be informed in articles and technical texts. Besides, the refinement (refiner or grinding) lignocellulosic fibers can be a good strategy to improve the interfacial transition zone between fiber and matrix in the microstructure. It is known that the type of process has a strong influence on interfacial transition zone of the fiber-cement composite. The efficiency of two processes were assessed at laboratory scale: slurry dewatering and extrusion. These results suggest that the extrusion process collaborates to organize better microstructure of the fiber-cement than the slurry dewatering process. However, it depends on the rheological characteristics of the fresh cement paste. Accelerated carbonation at an early age of the fiber-cement is a developing technology and a strategy to partially mitigate the problem with the interfacial transition zone mainly between lignocellulosic fiber and cement matrix with filling of CaCO3 content. Consequently, it decreases porosity and promotes a higher density in the interface, guaranteeing a good fibermatrix adhesion and a better mechanical behavior.

Acknowledgments The authors acknowledge the Brazilian financial support from Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP, Grant nos. 2008/04769-9, 2009/10614-0, 2009/ 17293-5, 2010/16524-0, 2012/51467-3, and 2013/03823-8), Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq, Grant nos. 472133/2009-8, 305792/20091, 309796/2012-1, and 406429/2015) and Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior (Capes, Grant no. 3886/2014). The authors thank the Brazilian companies Fibria S.A. and Infibra Ltda. for their collaboration and continuous support to research.

References 1. Fennis, S. A. A. M.; Walraven, J. C. Using particle packing technology for sustainable concrete mixture design. Heron 2012, 57 (2), 73102. 2. Olhero, S. M.; Ferreira, J. M. F. Influence of particle size distribution on rheology and particle packing of silica-based suspensions. Powder Technol. 2004, 139, 6975. Available from: http://dx.doi.org/10.1016/j.powtec.2003.10.004. 3. Wong, V.; Chan, K. W.; Kwan, A. K. H. Applying Theories of Particle Packing and Rheology to Concrete for Sustainable Development. Organ. Technol. Manag. Constr.: Int. J. 2013, 5 (2), 841855. Available from: http://dx.doi.org/10.5592/otmcj.2013.2.3. 4. Pedrajas, C.; Rahhal, V.; Talero, R. Determination of characteristic rheological parameters in Portland cement pastes. Constr. Build. Mater. 2014, 51, 484491. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.10.004.

64

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

5. Quanji, Z. Thixotropic Behavior of Cement-Based Materials: Effect of Clay and Cement Types. M.Sc. Thesis; Iowa State University, 2010. 6. Vance, K.; Kumar, A.; Sant, G.; Neithalath, N. The rheological properties of ternary binders containing Portland cement, limestone, and metakaolin or fly ash. Cem. Concr. Res. 2013, 52, 196207. Available from: http://dx.doi.org/10.1016/j.cemconres.2013.07.007. 7. Talero, R.; Pedrajas, C.; Rahhal, V. Performance of Fresh Portland Cement Pastes— Determination of Some Specific Rheological Parameters. (Chapter 4) In Rheology— New Concepts, Applications and Methods; Durairaj, R., Ed.; 2013. pp 5779. Available from: http://dx.doi.org/10.5772/53761. 8. Knop, Y.; Peled, A.; Cohen, R. Influences of limestone particle size distributions and contentson blended cement properties. Constr. Build. Mater. 2014, 71, 2634. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.08.004. 9. Merkus, H. G. Particle Size Measurements: Fundamentals, Practice, Quality (Particle Technology Series), 1st edition; Springer: Dordrecht, 2009, 534 pp. 10. Reed, J. S. Principles of Ceramics Processing, 2nd edition; Wiley-Interscience: New York, 1995, 658 pp. 11. Zhao, S.; Zhang, N.; Zhou, X.; Zhang, L. Particle shape effects on fabric of granular random packing. Powder Technol. 2017, 310, 175186. Available from: http://dx.doi. org/10.1016/j.powtec.2016.12.094. 12. Jia, X.; Williams, R. A. A packing algorithm for particles of arbitrary shapes. Powder Technol. 2001, 120 (3), 175186. doi: 10.1016/S0032-5910(01)00268-6. 13. Santos, S. F.; Tonoli, G. H. D.; Mejia, J. E. B.; Fiorelli, J.; Savastano, H., Jr. Nonconventional cement-based composites reinforced with vegetable fibers: a review of strategies to improve durability. Mater. Constr. 2015, 65 (317), 120. Available from: http://dx.doi.org/10.3989/mc.2015.05514. 14. Oliveira, I. R.; Studart, A. R.; Pileggi, R. G.; Pandolfelli, V. C. Dispersion and Packing of Particles—Basic Principles and Applications to Ceramic Processing, 1st edition; Fazendo Arte Editorial, 2000, 224 pp (in Portuguese). 15. Liu, S.; Ha, Z. Prediction of random packing limit for multimodal particle mixtures. Powder Technol. 2002, 126, 283296. doi: 10.1016/S0032-5910(02)00075-X. 16. Dinger, D. R.; Funk, J. E. Particle packing, Part II: review of packing of polydisperse particle systems. Interceram 1992, 41 (2), 9597. 17. Bentz, D. P.; Ferraris, C. F.; Galler, M. A.; Hansen, A. S.; Guynn, J. M. Influence of particle size distributions on yield stress and viscosity of cementfly ash pastes. Cem. Concr. Res. 2012, 42, 404409. Available from: http://dx.doi.org/10.1016/j.cemconres. 2011.11.006. 18. Quanji, Z. Thixotropic Behavior of Cement-Based Materials: Effect of Clay and Cement Types. M.Sc. Thesis; Iowa State University, 2010. 19. Almeida, A. E. F. S.; Tonoli, G. H. D.; Santos, S. F.; Savastano, H., Jr. Improved durability of vegetable fiber reinforced cement composite subject to accelerated carbonation at early age. Cem. Concr. Compos. 2013, 42, 4958. Available from: http://dx.doi.org/ 10.1016/j.cemconcomp.2013.05.001. 20. Santos, S. F.; Schmidt, R.; Almeida, A. E. F. S.; Tonoli, G. H. D.; Savastano, H., Jr. Supercritical carbonation treatment on extruded fibercement reinforced with vegetable fibres. Cem. Concr. Compos. 2015, 56, 8494. Available from: http://dx.doi. org/10.1016/j.cemconcomp.2014.11.007. 21. Choi, W.-C.; Jang, S.-J.; Yun, H.-D. Interface bond characterization between fiber and cementitious matrix. Int. J. Polym. Sci. 2015, 2015, 112. Available from: http://dx.doi. org/10.1155/2015/616949.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

65

22. Barra, B. N.; Santos, S. F.; Bergo, P. V. A.; Alves, C., Jr.; Ghavami, K.; Savastano, H., Jr. Residual sisal fibers treated by methane cold plasma discharge for potential application in cement based material. Ind. Crops Prod. 2015, 77, 691702. Available from: http://dx.doi.org/10.1016/j.indcrop.2015.07.052. 23. Gillespie, J. W., Jr.; Devault, J. B.; Edie, D. D.; Gabara, V.; Haulik, T. J.; Kardos, J. L.; Schadler, L. S. High-Performance Structural Fibers for Advanced Polymer Matrix Composites. Committee on High-Performance Structural Fibers for Advanced Polymer Matrix Composites; National Research Council: Washington D.C., 2005, 70 pp, ISBN: 0-309-54943-4. 24. Silva, F. A.; Mobasher, B.; Soranakom, C.; Tolˆedo Filho, R. D. Effect of fiber shape and morphology on interfacial bond and cracking behaviors of sisal fiber cement based composites. Cem. Concr. Compos. 2011, 33, 814823. Available from: http://dx.doi. org/10.1016/j.cemconcomp.2011.05.003. 25. Tonoli, G. H. D.; Rodrigues Filho, U. P.; Savastano, H., Jr.; Bras, J.; Belgacem, M. N.; Rocco Lahr, F. A. Cellulose modified fibres in cement based composites. Compos. Part A: Appl. Sci. Manuf. 2009, 40 (12), 20462053. Available from: http://dx.doi.org/ 10.1016/j.compositesa.2009.09.016. 26. Ardanuy, M.; Claramunt, J.; Tolˆedo Filho, R. D. Cellulosic fiber reinforced cementbased composites: A review of recent research. Constr. Build. Mater. 2015, 79, 115128. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.01.035. 27. Bismarck, A.; Mohanty, A. K.; Aranberri-Askargorta, I.; Czapla, S.; Misra, M.; Hinrichsen, G.; Springer, J. Surface characterization of natural fibers; surface properties and the water up-take behavior of modified sisal and coir fibers. Green Chem. 2001, 3, 100107. Available from: http://dx.doi.org/10.1039/b100365h. 28. Monteiro, S. N.; Lopes, F. P. D.; Barbosa, A. P.; Bevitori, A. B.; Silva, I. L. A.; Costa, L. L. Natural lignocellulosic fibers as engineering materials—an overview. Metall. Mater. Trans. A 2011, 42A (10), 29632974. Available from: http://dx.doi.org/10.1007/ s11661-011-0789-6. 29. Claramunt, J.; Ardanuy, M.; Garcı´a-Hortal, J. A.; Tolˆedo Filho, R. D. The hornification of vegetable fibers to improve the durability of cement mortar composites. Cem. Concr. Compos. 2011, 33, 586595. Available from: http://dx.doi.org/10.1016/j.cemconcomp. 2011.03.003. 30. Mejia, J. E. B.; Santos, S. F.; Ma´rmol, G.; Savastano, H., Jr.; Fiorelli, J. Evaluation of cellulosic pulps treated by hornification as reinforcement of cementitious composites. Constr. Build. Mater. 2015, 100, 8390. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2015.09.044. 31. Afra, E.; Yousefi, H.; Hadilama, M. M.; Nishino, T. Comparative effect of mechanical beating and nanofibrillation of cellulose on paper properties made from bagasse and softwood pulps. Carbohydr. Polym. 2013, 97, 725730. Available from: http://dx.doi. org/10.1016/j.carbpol.2013.05.032. 32. Correia, V. C.; Santos, V.; Sain, M.; Santos, S. F.; Lea˜o, A. L.; Savastano, H., Jr. Grinding process for the production of nanofibrillated cellulose based on unbleached and bleached bamboo organosolv pulp. Cellulose 2016, 23 (5), 29712987. Available from: http://dx.doi.org/10.1007/s10570-016-0996-9. 33. Gibson, L. J. The hierarchical structure and mechanics of plant materials. J. R. Soc. Interface 2012, 9, 27492766. Available from: http://dx.doi.org/10.1098/rsif.2012. 0341.

66

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

34. Cao, Y.; Zavaterri, P.; Youngblood, J. Y.; Moon, R.; Weiss, J. The influence of cellulose nanocrystal additions on the performance of cement paste. Cem. Concr. Compos. 2015, 56, 7383. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2014.11.008. 35. Coutts, R. S. P. Wood Fibre Reinforced Cement Composites. In Natural Fibre Reinforced Cement and Concrete; Swamy, R. N., Ed.; Blackie: Glasgow, 1988; pp 208242. 36. Bentur, A.; Mindess, S. Fibre Reinforced Cementitious Composites, 2nd edition; Taylor & Francis: New York, 2007, 449 pp. 37. Yang, E.-H.; Li, V. C. Strain-hardening fiber cement optimization and component tailoring by means of a micromechanical model. Constr. Build. Mater. 2010, 24, 130139. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2007.05.014. 38. ASTM C 150. American Society for Testing and Materials—ASTM C150/C150M-11. Standard Specification for Portland Cement, 2011. 39. Shao, Y.; Marikunte, S.; Shah, S. P. Extruded fiber-reinforced composites. Concr. Int. 1995, 17 (4), 4852. 40. Qian, X.; Zhou, X.; Mu, B.; Li, Z. Fiber alignment and property direction dependency of FRC extrudate. Cem. Concr. Res. 2003, 33 (10), 15751581. doi: 10.1016/S00088846(03)00108-X. 41. Takashima, H.; Miyagai, K.; Hashida, T.; Li, V. C. A design approach for the mechanical properties of polypropylene discontinuous fiber reinforced cementitious composites by extrusion molding. Eng. Fract. Mech. 2003, 70 (78), 853870. doi: 10.1016/ S0013-7944(02)00154-6. 42. Teixeira, R. S.; Tonoli, G. H. D.; Santos, S. F.; Fiorelli, J.; Savastano, H., Jr.; Lahr, F. A. R. Extruded Cement Based Composites Reinforced with Sugar Cane Bagasse Fibres. Key Engineering Materials 2012, 517, 450457. Available from: http://dx.doi. org/10.4028/www.scientific.net/KEM.517.450. 43. Shah, Y.; Shao, S.; Marikunte, S. Method of Making Extruded Fiber Reinforced Cement Matrix Composites. Patent Application Publication. US 1999/6,528,151 B1, 1999. 44. Li, Z. J.; Mu, B.; Chui, S. N. C. Static and dynamic behavior extruded sheets with short fibers. J. Mater. Civil Eng. 2001, 13 (4), 248254. Available from: http://dx.doi.org/ 10.1061/(ASCE)0899-1561(2001)13:4(248). 45. Zhou, X.; Li, Z. Manufacturing cement-based materials and building products via extrusion: from laboratory to factory. Civil Eng. 2015, 168 (6), 1116. Available from: http://dx.doi.org/10.1680/cien.14.00065. 46. Chen, H. Burwood R.J., Maxwell I.A., Goringe N.S. Method and Apparatus for Extruding Cementitious Articles. Patent Application Publication. US 2009/0218720 A1, 2009. 47. Glatthor, A.; Schweizer, D. Rheological lab testing of building formulations. In Proceedings of the ConChem International Exhibition & Conference, Germany, Augsburg, November 1994; Verlag Ziolkowsky: Augsburg, 1994. 48. Cardoso, F. A.; Fujii, A. L.; Pileggi, R. G.; Chaouche, M. Parallel-plate rotational rheometry of cement paste: Influence of the squeeze velocity during gap positioning. Cem. Concr. Res. 2015, 75 (9), 6674. 49. Ho¨lzgen, M.; Quirmbach, P. Additive for Extrusion. In Extrusion in Ceramics; Ha¨ndle, Frank, Ed.; Springer: Berlin, Germany, 2007; pp 233244. 50. Gleißle, W.; Graczyk, J. Rheology and Extrudability of Ceramic Compounds. In Extrusion in Ceramics; Ha¨ndle, Frank, Ed.; Springer: Berlin, Germany, 2007; pp 175187. 51. Coussot, P. Introduction to the Rheology of Complex Fluids. In Understanding the Rheology of Concrete; Roussel, Nicolas, Ed.; Woodhead Publishing Limited: Cambridge, UK, 2012; pp 322.

Interfacial transition zone between lignocellulosic fiber and matrix in cement-based composites

67

52. Blackburn, S.; Lawson, T. A. Mullite—alumina composites by extrusion. J. Am. Ceram. Soc. 1992, 75 (4), 953957. Available from: http://dx.doi.org/10.1111/j.1151-2916.1992. tb04165.x. 53. Taylor, H. F. W. Cement Chemistry; Academic Press Limited: London, 1990, 475 pp. 54. Srinivasan, S.; Deford, D.; Shah, P. S. The use of extrusion rheometry in the development of extrudate fibre-reinforced cement composites. Concr. Sci. Eng. 1999, 1 (11), 2636. 55. Benbow, J. J.; Oxley, E. W.; Bridgwater, J. The extrusion mechanics of pastes—the influence of paste formulation on extrusion parameters. Chem. Eng. Sci. 1987, 42 (9), 21512162. doi: 10.1016/0009-2509(87)85036-4. 56. Kuder, K. G.; Shah, S. P. Processing of high-performance fiber-reinforced cement-based composites. Constr. Build. Mater. 2010, 24 (2), 181186. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2007.06.018. 57. Toutou, Z.; Roussel, N.; Lanos, C. The squeezing test: a tool to identify firm cement-based material’s rheological behaviour and evaluate their extrusion ability. Cem. Concr. Res. 2005, 35 (10), 18911899. Available from: http://dx.doi.org/10.1016/j.cemconres.2004.09.007. 58. Emoto, T.; Bier, T. A. Rheological behavior as influenced by plasticizers and hydration kinetics. Cem. Concr. Res. 2007, 37, 647654. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2007.01.009. 59. Gru¨newald, S. Fibre Reinforcement and the Rheology of Concrete. In Understanding the Rheology of Concrete; Roussel, Nicolas, Ed.; Woodhead Publishing Limited: Cambridge, UK, 2012; pp 229256. 60. Benbow, J. J.; Jazayeri, S. H.; Bridgware, J. The flow of paste through dies of complicated geometry. Powder Technol. 1991, 65, 393401. (A Special Volume Devoted to the Second Symposium on Advances in Particulate Technology). 61. Benbow, J. J.; Bridgwater, J. The cutting of paste extrudates. Chem. Eng. Sci. 1993, 48 (17), 30883091. doi: 10.1016/0009-2509(93)80175-P. 62. Brumaud, C.; Baumann, R.; Schmitz, M.; Radler, M.; Roussel, N. Cellulose ethers and yield stress of cement pastes. Cem. Concr. Res. 2014, 55, 1421. Available from: http:// dx.doi.org/10.1016/j.cemconres.2013.06.013. 63. Betioli, A. M.; Gleize, P. J. P.; Silva, D. A.; John, V. M.; Pileggi, R. G. Effect of HMEC on the consolidation of cement pastes: Isothermal calorimetry versus oscillatory rheometry. Cem. Concr. Res. 2009, 39, 440445. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2009.02.002. 64. Peled, A.; Shah, S. P. Processing effects in cementitious composites: extrusion and casting. J. Mater. Civil Eng. 2003, 15 (2), 192199. doi: 10.1061/(ASCE)0899-1561(2003) 15:2(192). 65. Ferna´ndez Bertos, M.; Simons, S. J. R.; Hills, C. D.; Carey, P. J. A review of accelerated carbonation technology in the treatment of cement-based materials and sequestration of CO2. J. Hazard. Mater. 2004, B112, 193205. Available from: http://dx.doi.org/ 10.1016/j.jhazmat.2004.04.019. ˇ 66. Savija, B.; Lukovi´c, M. Carbonation of cement paste: understanding, challenges, and opportunities. Constr. Build. Mater. 2016, 117, 285301. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2016.04.138. 67. Pizzol, V. D.; Mendes, L. M.; Frezzatti, L.; Savastano, H., Jr.; Tonoli, G. H. D. Effect of accelerated carbonation on the microstructure and physical properties of hybrid fibercement composites. Miner. Eng. 2014, 59, 101106. Available from: http://dx.doi.org/ 10.1016/j.mineng.2013.11.007.

68

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Further reading Yan, L.; Kasal, B.; Huang, L. A review of recent research on the use of cellulosic fibres, their fibre fabric reinforced cementitious, geo-polymer and polymer composites in civil engineering. Composites Part B 2016, 92, 94132. Available from: http://dx.doi.org/ 10.1016/j.compositesb.2016.02.002.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

4

Marie-Ange Arse`ne, Ketty Bilba and Cristel One´sippe Universite´ des Antilles, Pointe-a`-Pitre, Guadeloupe

4.1

Introduction

Vegetable biomass is composed of the whole dead or alive cells of plants, which major components are cellulose, hemicelluloses, and lignin.1 The structure and composition of biomass vary according to species but also to localization in the plant.24 Biomass is a network of fibrous cells, whose essential function is to give mechanical strength in order to face to environmental constraint. Cellulose is a linear carbohydrate macromolecule with a (C6H10O5)n formula. The polymerization degree allows distinguishing cellulose present in various types of biomass.2 This macromolecule crystalizes in fibrils divided in a matrix made of hemicelluloses and lignin, which are the main constituents of cellular walls. Hemicelluloses are complex nonaromatic molecules, with short chains based on C5 or C6 polysaccharides monomers. Hemicelluloses are amorphous.2,5 Their polymerization degrees are inferior to those of cellulose. It is the difference of solubility in diluted alkaline solutions that allows separation of hemicelluloses and cellulose.3 Amorphous lignin is a mixture of phenolic polymers with high molecular weight.2,5 The main precursors of the biosynthesis of lignin are coumarylic, conferylic, and synapylic alcohols.5 Each main constituent of vegetable fibers has specific enzymatical, thermal, and chemical strength, which contribute to the macroscopic behavior of biomass. Crystallized cellulose fibrils give mechanical properties to materials.6 Various studies have shown the possibility of introducing vegetable fibers in composites due to ecosystem, economy, capability of potential applications, improvement in durability strength and ductility, low energy consumption concerns.79 Once vegetable fibers are introduced in a cementitious matrix, the elaborated composite is less sustainable than cementitious matrix without fiber. The main limitations of using vegetable fibers are: G

G

High amount of water absorbed by fibers, which leads to an increase in the water/cement ratio, close to the fiber/matrix interface,10 and creates a porous zone (thickness 10100 μm11), which is called the interfacial zone. Variable fiber/matrix adhesion due to nature of linkages present in fiber constituents and at their surface. Adhesion is facilitated when polar covalent linkages are at the fiber surface12 and the surface of fibers is rough.13

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00004-8 © 2017 Elsevier Ltd. All rights reserved.

70

G

G

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Mineralization of vegetable fiber due to migration of calcium hydroxide in the fiber lumen.14,15 High vulnerability to chemical attacks, particularly in alkaline environment.16 According to Johnston17 within the major constituents of vegetable fibers, cellulose is the more resistant to basic solutions.

There are two ways to improve vegetable fiber durability in cementitious matrix: (1) modification of the matrix by using pozzolanic materials and (2) pretreatment of fibers before its introduction in the cementitious matrix. In this chapter, we will focus on pretreatment of fibers.

4.2

Pretreatment methods

The pretreatment of fibers improves vegetable fibers/cementitious matrix adhesion and durability of composites.18,19 Those pretreatments aim to modify the fiber surface by eliminating the components responsible for its degradation in cement. There are two types of pretreatment methods: G

G

Chemical pretreatment11,1824 Physical pretreatment18,2529

4.2.1 Chemical treatment As vegetable fibers are mainly composed of hydroxyl groups, they can support chemical modifications, which allow to clean surface fibers, modify the chemistry of the fiber surface, reduce humidity, and increase roughness in order to improve fiber/matrix interface30 by activating or introducing new groups leading to create linkages between fibers and matrix.19 Various pretreatments can be used: acidic, alkaline, coating, and drying/rewetting cycles. In the literature, the more commonly reported methods are alkaline processes well known in delignification3134 and acidic processes.3537 In recent years greener treatment processes have been proposed.7,9

4.2.1.1 Acidic pretreatment18,24 The effects of acidic pretreatment on vegetables fibers vary with the nature and the concentration of the acid and also with temperature. Acids easily hydrolyze the hemicelluloses. By hydrolysis, they form their corresponding monomer elements: D-glucose, D-mannose, D-xylose, L-arabinose, L-rhamnose, D-glucuronic acid, 4-methyl-D-glucuronic acid, and D-galacturonic acid. The hydrolysis of hemicellulose is a result of the rupture of acetyl bonds and the formation of a “hexose” water-soluble component. Increasing temperature also enhances the solubilization of hemicellulose. However, the increased yield of sugars can be masked by the instability of the resulting sugars. For example, fructose is not stable above 120 C. The increase in the concentration of the acid or hydrogen ions decreases the selectivity of polysaccharides hydrolysis. For environmental and

Treatments for viable utilization of vegetable fibers in inorganic-based composites

71

economic reasons, sulfuric acid is the most widely used acid reactant. The sugars are released efficiently by treatment with hot dilute acid. However, the use of chloridric or fluoridric acids is preferred to obtain polar molecules and to increase acidic fractions. Increasing the concentration of sulfuric acid can also produce the progressive removal of cellulose. However, it is difficult to remove all the cellulose. Indeed, acid treatments can produce weak or strong acid groups according to the nature of the acid. Hence, cellulose with sulfuric acid forms cellulose sulfates with a higher molecular weight than cellulose. The viscosity also increases after the acidic treatment of cellulose. Furthermore, the viscosity is higher after HCl than sulfuric acid treatment, although the particle sizes are the same after both treatments. Concentrated mineral acids reduce crystallinity and disrupt the association of lignin with cellulose. They also dissolve hemicellulose. The action of acid on cellulose can entail the attack of glucosidic bonds by hydronium ions. It also induces failure of the microfibrils and the separation of the constituents of the fiber. The lignin in hardwood species is partly dissolved by sulfuric acid during the acid hydrolysis. The acid treatment decreases the aliphatic hydroxyl content and thus decreases the polarity of the lignin molecules without derivatization, but enhances the antioxidant properties. In the case of olive stones, after 1 h of sulfuric acid (72%) treatment, hemicellulose is totally removed without significant alteration of the cellulose and lignin structure. Above 2 h of this sulfuric acid treatment, cellulose begins to dissolve, and lignin is modified.

4.2.1.2 Alkaline pretreatment18,24 All ester-linked molecules of the hemicellulose and other cell-wall components can be cleaved by alkali, increasing the hydrophilicity and hence the solubility of the material.38 It is also known that cellulose swells when treated with strong alkaline solutions. This swelling is due to the disruption of interfibrillar bonds, or breakage of adjacent bond, which bind cellulose molecules together. The native cellulose is transformed into a more reactive cellulose form, due to the easier penetration of the reagents. These treated cellulose fibers are called “mercerized,” after the inventor Mercer (1844). Due to their reticulation, lignin in situ is usually insoluble in all solvents. Nevertheless, some alkali labile linkages between lignin monomers or between lignin and polysaccharides may be broken by alkali treatment. Carboxylic or phenolic groups, ionized in alkaline solution, might also promote the solubilization of lignin, either by increasing the solubility of individual fragments or by inducing the swelling of the cell wall. It is generally thought that the main fragments of lignin in alkaline media arise from the cleavage of α-aryl ether bonds in polyphenolic units. The amount of lignin removed by alkali treatment varies with the nature of the fiber. Alkali treatment of lignocellulosic substances disrupts the cell wall and dissolves hemicellulose and lignin by hydrolyzing acetic acid esters and by swelling cellulose, decreasing the crystallinity of cellulose. The biodegradability of the cell wall increases, due to cleavage of the bonds between lignin and hemicellulose or lignin and cellulose.

72

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

In the Kraft process that is used in the paper industry, the alkaline reactant that is used is a mixture of sodium hydroxide, sodium sulfite, and sodium carbonate. Other alkaline solutions include sodium hydroxide, soda-anthraquinone, sulfite anthraquinone,33 alkaline sulfite anthraquinone, calcium hydroxide, potassium hydroxide, lithium hydroxide, ammonia, and hydrogen peroxide. It was shown on wheat straw treatment that calcium hydroxide and liquid ammonia at 20 C are less aggressive alkaline reagents than sodium hydroxide. After calcium hydroxide or liquid ammonia treatments for 6 h at 20 C, lignin was not extracted. However, in the case of sodium hydroxide pretreatment under the same conditions, 40 wt% of the lignin and 60 wt% of the hemicellulose were dissolved.18

4.2.1.3 Coating In the case of vegetable fiber-reinforced cement composites with Kraft or hardwood, silane pretreatment has been found to increase moisture-cycling resistance.39 Rocha Ferreira et al.9 indicate that polymer coating treatment reduces water absorption and is effective in the sealing process of natural fibers. Alkyltrialkoxysilanes R0 Si(OR)3 are used in many industrial application as coupling agent to improve adhesion between a polymeric matrix and inorganic materials. They have also been used in vegetable fiberpolymer composites27,40 to modify fiber surface. The mechanism of these coupling reactions is related to the effect of two reactive surface groups and the reactions are different according to the substrate in contact. On the one hand, alkoxy groups (OR) allow the silane to bind to surface OH groups after hydrolysis and, on the other hand, the alkyl groups R0 increase the compatibility with organic compounds and the hydrophobic character of the surface, thus leading to an increase in the strength of the interface in the polymer matrix. Usually, silane treatment is carried out with a dilute silane solution from 0.2% to 20 wt%.41,42 These conditions have the following advantages: G

G

G

An increase in silane solubility An improvement in the thickness of the surface film The development of a uniform cover on the surface

Water produces moderate hydrolysis of the silane and leads to silanol. This behavior allows the adhesion of silanes onto the OH groups of the fiber substrate. After solvent evaporation, the residual silanol groups can condense with the hydroxyl groups of the substrate, or can lead, by “self-condensation” reactions, to a polysiloxane network on the surface. In aqueous media, partially or totally hydrolyzed silanes are reactive molecules that change with time by self-condensation of silanol groups or condensation of silanol with alkoxy groups to form dimers or oligomers. They are commonly used as commercial water repellents. Silanes grafted onto the bagasse fiber surface form a network of polysiloxane molecules.23 They change the morphology of the fibers by causing swelling of the bagasse fibers and increase fiber dimensions and porosity. Silane treatment also decreases water adsorption by the fibers; this effect is more pronounced after a dialkydialkoxysilane (R2Si(ORv)2) treatment. Modifications are more extensive

Treatments for viable utilization of vegetable fibers in inorganic-based composites

73

with dialkydialkoxysilane treatment than with alkyltrialkoxysilane (RSi(OR0 )3) treatment.23 Silane coatings are good candidates for improving the fiber/matrix interface stability and enhance cement adhesion onto the fibers.23 The fibers are not pulled out after the fracture of composite. After pyrolysis, silane grafting seems to be less extensive. The combination of pyrolysis and silane treatments creates a lower porosity than the corresponding silane treatment alone.23 Dimensional changes are not influenced by additional pyrolysis. Composite setting is not modified by pyrolysis pretreatment. The only important modification is the hydrophilic character of the fibers; it strongly decreases with the cumulative contribution of both pyrolysis and silane treatments.

4.2.1.4 Drying/rewetting cycles The authors proposed successive drying and rewetting cycles as a way to improve the dimensional and thermal stability of cellulosic fibers against environmental changes in cement mortar composites and to reduce the water gradient in the fibermatrix interface.7 Drying and rewetting cycles cause principally shrinkage of the vegetable fibers due to the formation of hydrogen bonds in cellulose. This reversible effect called “hornification”7 occurs in the cell wall of the fibers, resulting in intensely bonded structures. The sequence of cycles is as follows: (1) drying in an oven with air circulation at 60 C during 7 h; (2) rewetting by soaking overnight; (3) disintegration of the wet pulp for 30,000 revolutions; and (4) filtration of the pulp suspension through a Buchner funnel (150 mesh, 88105 m). These authors have demonstrated “hornificated” fibers measuring viscosity and curl index (gradual and continuous curvature), observing morphology of fibers. They expect that the hornificated fibers will have higher stiffness and tensile strength as a consequence of the formation of a more closely packed crystalline. Rocha Ferreira et al.9 also propose rewetting and drying cycles to reduce fibers hydrophilicity and promote the stiffening of the polymeric structure present in lignocellulosic materials. The first step of the hornification cycle is a rewetting at 22 C during 3 h to reach the maximum water absorption capacity of fibers, followed by drying maintained at 80 C during 16 h and a cool down to 22 C in order to avoid thermal shock to the fibers. The cycle is repeated 10 times. They also observed thermal stability of hornificated fibers and indicated: G

G

an increase in cellulose crystallites, in hydrogen bridges, in hydrophobicity of lignin, in tensile strength and strain at failure, in adhesion bond by stiffening of polymeric structure of fiber-cells; and in parallel a decrease of water-absorption capacity.

4.2.2 Physical treatment Physical pretreatments of fibers essentially modify the surface properties of fibers and have few consequences on the chemical composition. They mainly influence linkages between fibers and matrix, i.e., the adhesion between them.

74

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Regardless of high-energy consumption, mechanical pulping presents a range of advantages over chemical pulps. As noted by Coutts and Ridikas43 effluent treatment and disposal are less troublesome, chemical requirements are lower, pulp price is cheaper (around half the price of Kraft pulp), and mills find economic viability at small scale.

4.2.2.1 Mechanical: pulping As exposed for softwood pulping,24 low-temperature thermo-mechanical processes with defibration temperatures in vicinity of 130 C are expected to provide wellfibrillated fibers (although with some fiber damage incidence). Higher temperatures (135170 C) are supposed to provide no fibrillation and also smooth ligninencased fibers, with poor bonding and beating properties and also with low conformability of the fibers. The Kraft or sulfate process is considered the predominant procedure for pulping all over the world. In the Kraft process that is used in the papermaking industry,27 the alkaline reactant is a mixture of sodium hydroxide, sodium sulfite, and sodium carbonate.27,44,45 Other alkaline solutions include sodium hydroxide,4447 soda-anthraquinone,44,45 sulfite anthraquinone,45 alkaline sulfite anthraquinone,45 calcium hydroxide,47 potassium hydroxide,47 lithium hydroxide,47 ammonia,47 and hydrogen peroxide.47 One of the possible treatments to enhance mechanical performance of cellulosic pulp is a refinement process, which is carried out in the presence of water, usually by passing the suspension of pulp fibers through a refiner disc comprised of a relatively narrow gap between the rotor and the stator.48,49 In the papermaking industry, refining or beating is an important stage in preparing the pulp aiming to improve the ability to form bonds among fibers. This is reached with the mechanical treatment of pulp due to the changing of the fiber structures and properties. The fiber treatment, a combination of flexing, shearing, rolling, and compressing, in the presence of water, has as the main function to plasticize it. This process, called internal fibrillation, starts with the removal of the remaining primary wall. The subsequent plasticization increases either wet flexibility or lateral conformability. It is generally assumed that the more highly plasticized fibers will compress more readily and, as a consequence, the area of interfiber and matrix/fiber contact will be greater.50 As proposed by Tonoli et al.51 in a study on cement-based composites reinforced with refined sisal pulp, the refinement causes intense fibrillation of the plant fibers, promoting their plasticization and mechanical anchorage in the cement-based matrix. However, excessive refinement also causes higher damage to the filaments, with expressive shortening and increase of fines. Pulp beating also plays an important role in composites subjected to accelerated aging tests. In this situation the predominant effect seems to be the improved densification of the composite. The better packing provided by the plasticized fibers and the continued hydration of the composites subjected to wet/dry cycling were combined to create an additional enhancement in the modulus of rupture, the limit of proportionality, and the modulus of elasticity. The water absorption also presented a significant reduction that

Treatments for viable utilization of vegetable fibers in inorganic-based composites

75

can be partially explained by the gradual carbonation of cement matrix. These results seem very promising in view of the potential applications of such nonconventional composites for cost-effective housing and infrastructure.

4.2.2.2 Thermal: pyrolysis Pyrolysis is a heat treatment used in the absence of oxygen in order to avoid oxidation and combustion. Its action on a surface leads to modification of the structure and degradation of botanical constituents of biomass according to temperature.28 The structural and chemical composition evolution of pyrolyzed sugar-cane bagasse fibers were studied using FTIR spectroscopy and elementary analysis, respectively, in the temperature range 200800 C.28 The principal features of this investigation were: 1. Structure modifications appear at 200 C and are intensified between 300 C and 400 C. These are characterized by (a) a decrease of intensities of the OH, CO, C5C lines and (b) the formation of alkyl bands. 2. Between 300 C and 400 C, modifications intensify and there is a preferential dehydration of the constituents of sugar-cane bagasse. As a matter of fact, medium temperatures (300320 C) favor dehydration of carbohydrates (hemicelluloses and cellulose), while higher ones (340380 C) facilitate dehydration of lignin. The continuation of the heat treatment leads to saturation of the aromatic rings and then rupture of the CC linkages formed in lignin, the release of water, CO and CO, molecules, and rearrangement of carbohydrates and lignin structures. High heating rates seem to favor the release of hydrogen (H2) and/or hydrocarbon molecules. 3. The thermal decomposition of bagasse progressively leads to the evolution of lowmolecular-weight compounds until all chemical linkages disappear and a char containing essentially carbon and inorganic oxides 800 C) is formed.

In this study, the authors thus showed a decrease of hemicellulose and extractible contents, no dimensions variation of fibers, and an increase of roughness at the surface of sugar-cane bagasse fibers, which could later increase the adhesion between fibers and matrix. They proposed that the best compromise between mass loss and sugars degradation is a pyrolysis at 200 C during 2 h. Heat-treated bagasse (200300 C) is less hygroscopic than untreated biomass; Bilba et al.23 also confirmed this observation in a study of silane treatment of bagasse fiber for reinforcement of cementitious composites. This is due to a different chemical composition and the disappearance of OH groups. This is an essential property for making composite materials with potentially improved thermal, mechanical, durability, and dimensional stability properties. In a study about arrow-root/cement composites29 using FTIR the authors showed that after a pyrolysis of arrow-root fibers between 200 C and 700 C: G

G

There is a structural evolution of the char pointed out by differences of intensities of peaks variable according to crop of arrow-root fibers for the same pyrolysis temperature. At 700 C, botanical constituents are not completely transformed in graphitic carbon.

Shafizadeh2 and Marcilla et al.52 show that the first process in the thermal decomposition of vegetable materials corresponds to hemicellulose decomposition.

76

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

This produces a low yield of chars. Several kinetic studies have concluded that this involves a first-order reaction, with activation energy between 100 and 112 kJ/mol, for the pyrolysis of hemicellulose from sugar-cane bagasse.53 This increases to 153 kJ/mol for the pyrolysis of hemicellulose from almonds.52 Shafizadeh2 studied the thermal decomposition of cellulose and found that, below 300 C, this reaction results in the formation of char, water, carbon monoxide, and carbon dioxide. However, between 300 C and 500 C, pyrolysis results in the formation of tars that consist largely of anhydrous sugars, oligosaccharides, and dehydration of pyran and furans. Above 500 C, flash pyrolysis produces volatiles or gaseous molecules of low molecular weight from secondary reactions of tars and the interaction between char, water, and CO2 at high temperature. The temperature of thermal decomposition also increases with the heating rate.54 The kinetic studies concluded that the pyrolysis of cellulose involves a first-order reaction with activation energy between 200 and 240 kJ/mol.52,54 Under isothermal conditions, between 280 C and 320 C, the thermal decomposition of cellulose is auto accelerated.55 Cellulose decomposition is the second process in the decomposition of lignocellulosic material, observed in almond chars by Marcilla et al.52 The pyrolysis of lignin occurs in inert atmospheres at high temperatures over a wide temperature domain.54,56 The thermal decomposition of lignin is also affected by the acid pretreatment used for cellulose extraction.57 This is responsible for the breakdown of the three-dimensional structure of lignin. A combination of chemical and/or thermal treatments can also be realized.35,57,58 Acid and alkaline treatments were proposed by Abou-Yousef et al.35 for sugar-cane delignification, hydrothermal et alkali pulp treatment by Garrote et al.58 acid and thermal treatment by Caballero et al.57

4.3

Influence of treatment on morphology and chemical composition of fiber

The influence of chemical treatments using acid, alkaline solution or drying/ rewetting cycles, and physical treatment are investigated here.

4.3.1 Influence on morphology and texture The effects of treatment of fibers can be seen using scanning electron microscopy (SEM) images to examine general aspects such as roughness, shape, and fiber dimension measurements. As can be shown by different study, the morphology and structures can be strongly modified by the fiber treatment.

4.3.1.1 Origin of fibers Fiber morphology is related to the botanical species and location of the fiber in the plant. The influence of the location in the plant has been highlighted by Arse`ne et al.18 for coconut fiber extracted from the fruit (coir) and the leaf sheath (coco sheath) (Fig. 4.1). Fig. 4.1 also shows the organization of the microfibers in the

Treatments for viable utilization of vegetable fibers in inorganic-based composites

Figure 4.1 SEM images of pyrolyzed coconut, bagasse, banana, and arrow-root fibers.18,29

77

78

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

different layers of the fibers for banana fiber extracted from the pseudo-stem and the leaf as well as for fiber extracted from banana pseudo-stem and for fiber extracted from arrow-roots. The dimensions of the cells on the base of an elliptic shape (D is major axis and d is minor axis) and the thickness of the wall (w) have been extracted from SEM observations and allow to evaluate some fibers characteristics as the cell surface density (cell mm22), the fraction of material and the interface as define by Arse`ne et al.18 From these observations, it is clear that the fiber morphology is related to the botanical species and location of the fiber in the plant. The SEM images of the cross-section of the fibers after fracture are presented in Fig. 4.1AF. The shapes, and consequently, the diameters of the fibers, are very different from one botanical species to another. These fibers appear as a bundle of unitary cells. At a macroscopic scale the dimensions of the fiber (length, width, thickness) depend on preparation method used to obtain the fiber (e.g., pulping, screening, cutting). These dimensions are important to evaluate the aspect ratio of the fiber, an important factor for the reinforcement capacity of the fiber. Table 4.124 summarizes the dimensions and aspect ratios observed for fiber obtained by crushing and sieving (0.41 mm) the raw material and the effect of the treatment on the aspect ratio.

4.3.1.2 Chemical treatment of fibers Acid and alkaline hydrolysis The pretreatment of vegetable fibers induces modifications of the surface, as shown in Fig. 4.2 for banana and bagasse fiber exposed to acid or alkaline hydrolysis.18 Chemical treatment by H2SO4 (5 wt%) and Ca(OH)2 (5 wt%) modifies the surface of the fibers. The strongest modification is observed for the alkaline treatment, which strongly damages the structure of the fiber: the hemicelluloses surrounding the cellulose skeleton are dissolved.

Coating Bagasse fibers 0.41 mm have been coated using silane solution (0.56 wt%). Different structures of silanes have been used, such as S1 a trialokoxysilane and S2 an alkyl-dialkoxysilane. The SEM images in Fig. 4.3 show the treated fiber surfaces. While SEM cannot directly identify silanes, their effect on bagasse-fiber morphology can be seen. Silane treatment of unpyrolyzed (Fig. 4.3A) or pyrolyzed bagasse fibers (Fig. 4.3B) changes the surface aspect of the fibers. After a 6 wt% silane S1 treatment, the surface of the fiber appears rougher with striations (Fig. 4.3C and D). After treatment with a 6 wt% silane S2 solution, the surface texture of the fibers looks “granular” (Fig. 4.3E and F). We suggest that either silane S2 partly covers the surface or causes, upon drying, cracking of the surface microfibrils. In addition we noted a densification of the fibers: the hollow structure of the pyrolyzed fibers (Fig. 4.3B) disappears after treatment with a 6% silane S1 solution (Fig. 4.3D).

Table 4.1

Macroscopic dimensions of some vegetable fibers24

Fiber

Arrow-Root (D1) Arrow Root (D2) Bagasse (B) Banana Trunk (BT) Banana Leaf (BL) Coconut Coir (CC1) Coconut Coir (CC2) Coconut Sheath (CT) nd, data nondetermined.

Aspect ratio L/w

Untreated fiber dimensions (st. dev.) Length (mm)

Width (µm)

Thickness (µm)

Untreated

Pyrolysis

Acid

Alkaline

Silane

5.77 [ 6 3.35] 5.45 [ 6 2.01] 3.69 6 2.15] 1.9 [ 6 0.64] 1.70 [ 6 0.91] 29.35 [ 6 8.17] 2.7 [ 6 2.46] 5.47 [ 6 3.08]

140.48 [ 6 86.14] 104.17 [ 6 46.98] 567.50 [ 6 329.40] 820.24 [ 6 264.49] 834.52 [ 6 245.94] 331.78 [ 6 198.72] 683.21 [ 6 279.42] 338.09 6 258.23]

83.03 [ 6 38.6] 64.58 [ 6 29.61] 161.25 [ 6 90.75] 150.29 [ 6 86 .94] 160.42 [ 6 55 .26] 273.57 [ 6 150.94] 215 [ 6 98.67] 177.68 [ 6 134.76]

4170

4182

nd

nd

nd

5285

4359

nd

nd

nd

6.523

623

613

415

712

2.312.6

623

819

1027

nd

211

216

nd

nd

nd

88107

84103

nd

nd

nd

413

310

nd

nd

nd

1631

2035

nd

nd

nd

80

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 4.2 SEM images of untreated and chemically treated bagasse and banana fibers.18

The results show an increase in the dimensions of the fibers after silane treatment.18 This trend is not significant considering the standard deviation of the results. This is observed for the three dimensions of the fibers. To conclude, a direct consequence of the increase in the external surface area that could be valuable, if the strength of the interface is preserved, is the

Treatments for viable utilization of vegetable fibers in inorganic-based composites

81

Figure 4.3 SEM images of bagasse fibers: (A) untreated, (B) pyrolyzed, (C) treated with S1, (D) pyrolyzed and treated with S1, (E) treated with S2, and (F) pyrolyzed and treated with S2.23

improvement in composite properties due to better contact between matrix and fiber.

Hornification (drying/rewetting cycles) Here, drying/rewetting cycles, also called hornification treatment, consist of a process inspired by Rocha Ferreira et al.9 The first step of the hornification cycle is a rewetting at 25 C for 2 h to reach the maximum water-absorption capacity of the fibers, followed by a drying at 80 C for 16 h and a cool down to 25 C in order to avoid thermal shock to the fibers. The cycle is repeated 10 times. This process has been performed on long fibers (l 5 4 cm) extracted from coir (CC) and bagasse (BAG). The dimensions of the fibers are given in Table 4.2 and the SEM images shown in Fig. 4.4. The dimensions of the raw fibers (BAG, CC) are always higher than the mean diameter of the hornificated fibers (BAGcms, CCcms).

82

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 4.2

Average diameters of the fibers used in the hornification

study59 Diameters (cm)

BAG Fine CC Average CC Thick CC

Raw

Pyrolyzed

Hornificated

0.076 6 0.023 0.054 6 0.005 0.07 6 0.01 0.1155 6 0.0323

0.0545 6 0.0139 0.0465 6 0.0074 0.0565 6 0.0138 0.078 6 0.017

0.067 6 0.016 0.052 6 0.007 0.065 6 0.014 0.097 6 0.029

BAG: bagasse fibers, CC: coir fibers

The long fiber from bagasse or coir is made of bundles of unitary fibers, as observed for short fibers (0.41 mm). These unitary fibers have different diameters.

4.3.1.3 Physical treatment of fibers Mechanical pulping Tonoli et al.51 studied the effect of refinement using a sisal Kraft pulp with a value of Canadian Standard Freeness of 680. The most affected parameters are fibrous materials content (increase of 233%), fine content (increase of 54%), length (decrease of 52%), and coarseness (decrease of 37%), while width is mostly constant. From the size distribution it was also noted that beating increases the portion of fibrous material with length of 01.5 mm by approximately 50% for the most refined pulp.

Thermal pyrolysis SEM images (Fig. 4.5) indicate that pyrolysis treatments (Fig. 4.5B, D, and F) increase the surface roughness from the raw fiber state (Fig. 4.5A, C, and E). To conclude, with the exception of the dry/rewetting cycles, treatments induce the following surface modifications: G

G

G

The surface roughness increases after treatment Severe damage of the internal structure appear after alkaline hydrolysis (0%100%) Coating increases fiber dimensions

4.3.2 Influence on chemical composition 4.3.2.1 Botanical composition As for morphological and textural properties, the chemical and elemental composition of the raw bagasse, coconut, and banana fibers vary with botanical species and with the location in the plant. For all of them, the most important constituents are cellulose (21%40%), lignin (15%47%), and hemicelluloses (12%27%). The major chemical elements (in wt%) are carbon (37%49%) and oxygen (38%46%).18,24 Acid treatment increases the cellulose content and strongly decreases the hemicellulose content. Such behavior is expected since attack by sulfuric acid results in

Treatments for viable utilization of vegetable fibers in inorganic-based composites

83

Figure 4.4 SEM images of long fibers from bagasse and coir, before and after hornification.59 BAG: raw bagasse fibers, CC: raw coir fibers, BAGcms: bagasse fibers after hornification, CCcms: coir fibers after hornification.

the dissolution of hemicellulose and cellulose, as stated above.57,59 The increase of cellulose content, compared to their levels in raw fiber, can be explained by a preferential dissolution of the other components, and by the reaction of cellulose with sulfuric acid. Indeed, the reaction of cellulose with sulfuric acid forms an ester C6H8(SO4H)2 that increases the mass fraction of cellulose. The action of acid on cellulose can also entail the attack of glucosidic bonds by hydronium ions. This induces the failure of the microfibrils and the separation of the fiber constituents.

84

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 4.5 SEM images of untreated and pyrolyzed fibers extracted from bagasse, banana, and arrow-root.18,29

Alkaline attack affects mostly lignin and extractives content.18 It also results in a decrease in the total fiber mass. The amount of lignin and extractives removed by alkali treatment varies with the nature of the fiber. The removed amount is greater for banana trunk than for bagasse fibers. The removal of lignin and hemicellulose, and the corresponding “enrichment” in cellulose, has been shown by Sun et al.60 for wheat straw fibers. Mosier et al.61 also reported similar findings. In the current study, the decrease of the amount of hemicellulose obtained for bagasse fiber is consistent with prior reports. After pyrolysis at 200 C, both the sugar-cane bagasse and banana-trunk fibers18 exhibited a decrease in the amount of extractive content. This was due to the elimination of volatile compounds such as methanol and acetic acid. The decrease is also attributed to the easier decomposition of hemicellulose. In the case of sugar-cane bagasse, the lignin and the cellulose content increase, while for the banana-trunk fiber, the lignin content decreases and the cellulose content increases. From results reported in the literature,2,29 the preferential transformation of hemicelluloses is expected before that of cellulose, which precedes the transformation of lignin. Lignin decomposition usually begins before cellulose transformation is completed. This occurs over a wide temperature range. Hence, an increase in both lignin and cellulose contents would be expected after pyrolysis at 200 C. It is interesting to

Treatments for viable utilization of vegetable fibers in inorganic-based composites

85

note that the unexpected decrease in lignin content was observed in fibers in which the total amount of component was less than 100%. The hydrophilic character of such fibers increases with pyrolysis. Due to OH groups, the fibers fix water by forming hydrogen bonds.2 After hornification, the analysis of the botanical content, considering the standard deviation of the results, indicates that for bagasse and coir fiber after pyrolysis or hornification treatment, the cellulose content is similar to the cellulose content of untreated fiber. The lignin content is higher for pyrolyzed fiber than for untreated fiber or after drying/rewetting cycles. These results are in agreement with those discussed previously by Arse`ne et al.18 who confirm that pyrolysis decreases hemicellulose content. While for coconut fiber (coir) the hemicellulose content of treated fibers is statistically lower than for untreated fiber, for lignin content, its highest value was observed after drying/ rewetting cycles (CCcms). As lignin content takes part in fiber rigidity and cellulose content in mechanical strength,62 according to these results, we could expect pyrolyzed bagasse fibers and fibers after drying/rewetting cycle to be the highest.

4.3.2.2 Surface groups pH value of surface groups (pHpzc) The impact of a lime solution on water absorption and wettability of sugar-cane bagasse fibers (0.41 mm) was investigated. The following processing conditions were applied to the fibers. They varied according to four parameters: G

G

G

G

Lime-solution concentration: 0.2 and 5 wt% Duration of impregnation of fibers into lime solution: 30 min and 2 h Temperature of treatment: 25 C and 100 C Volume of water in which the fibers are treated: 12.5 and 25 wt%

Thus a four-variable experimental design was established as shown by Table 4.3.

Four-variable experimental design for the study of the impact of a lime solution on water absorption and wettability of sugar-cane bagasse fibers63

Table 4.3

Type of fiber

Duration of treatment

Temperature of treatment

Volume of water

Lime-solution concentration

1 2 3 4 5 6 7 8

21 21 21 1 21 1 21 1

21 21 1 1 21 21 1 1

21 21 21 21 1 1 1 1

21 1 1 21 1 21 21 1

Duration of impregnation of fibers into lime solution: 21 corresponds to 30 min and 11 corresponds to 2 h; temperature of treatment: 21 corresponds to 25 C and 11 corresponds to 100 C; volume of water in which the fibers are treated: 21 corresponds to 12.5 wt% and 11 corresponds to 25 wt%; lime-solution concentration: 21 corresponds to 0.2 wt% and 11 corresponds to 5 wt%.

86

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 4.4

pHpzc of bagasse fibers after soaking 24 h in water63

Type of fiber

pHpzc

Effect of dry/rewetting processing parameters on pHpzc

1 2 3 4 5

7.51 6.87 8.43 6.71 7.55

Effect of concentration Effect of duration Effect of temperature Effect of volume of water Standard deviation of pHpzc (ΔpHpzc)

6 7 8

5.61 6.69a 7.32

0.46 0.46 0.20 0.29 0.19

Refer to fiber types described in Table 4.3. a Extrapolated value.

Treated fibers are numbered from 1 to 8 depending on their respective treatment: e.g., type 1 (by consequence treatment 1) corresponds to the lowest parameters: 30 min of treatment, 25 C, 12.5 wt% for the volume of water, and lime solution 0.2 wt%. The effect of alkaline treatment has been studied through the pH value of surface groups: pHpzc. The value of pHpzc (Table 4.4) allows the evaluation of the behavior of the fiber in its environment. When the fiber is in a solution in which the pH is higher than the pHpzc of the fiber, the fiber will have an acid behavior. The most significant parameters of this study are the concentration and the duration of the treatment. When increasing the concentration of lime solution, the pHpzc increases. When increasing the duration of the treatment, the pHpzc decreases. If the first observation is quite obvious, the second one (decrease of pHpzc) might be surprising, but it illustrates the acidification of the solution over time.

XPS analysis X-ray photoelectron spectroscopy analysis of unpyrolyzed bagasse fiber without and with coating by silane S1 and silane S2 exhibit O, C, and Si peaks.64,65 From the experimental Si electron energy bonds, according to various authors6370 the following observations can be deduced: G

G

G

Mainly peaks related to SiC, SiO2C2, and SiO2 bonds are reported on the surface; Peaks related to Si are apparently 1 nm underneath the surface after silane S1 or S2 treatment; and The signature of Si 2p apparent at 1nm below the surface, for unpyrolyzed and pyrolyzed fibers is present with low intensity, which corresponds to the inorganic elements of the fibers.71

Treatments for viable utilization of vegetable fibers in inorganic-based composites

87

Obviously, the presence of Si atoms is more important after silane treatment, showing that silanes are really grafted onto the surface of the fiber, and all samples present SiOxCy bonds due to linkages between the organic surface and the silane. The effect of pyrolysis pretreatment is not noticeable.

4.3.2.3 Infrared spectroscopy analysis—the effect of pyrolysis The effect of pyrolysis treatment on chemical structure has been investigated by Bilba et al.28,29 A detailed FTIR spectroscopy study has been performed for bagasse fibers to optimize the pyrolysis parameters. In case of bagasse fibers, 200 C has been identified as the optimized pyrolysis temperature. FTIR spectra of banana trunk and coconut sheath confirm the degradation of cellulose, hemicellulose and lignin occurs from 200 C, these structural modifications explain the morphological changes.29

4.3.2.4 Thermogravimetric analysis Effect of pyrolysis treatment on fibers Effect of pyrolysis treatment on fiber has been studied by Rodier72 for bagasse and Bilba et al. for arrow root fibers.27 Their main observations on mass loss curves are (1) water evaporation, (2) main decomposition of organic matter, and (3) final organic matter decomposition: G

G

G

The first zone between 29 C and 150 C presents a small variation of mass from 4.8% to 8% for untreated and pyrolyzed bagasse fibers respectively. It corresponds to the release of humidity from the fiber. As could be expected pyrolyzed fibers have got less water than untreated ones. This result is consistent with a decrease on IR spectra of bagasse fibers of OH the band. The second zone (150420 C) presents the highest mass loss with a decrease of 64.5% for the untreated fibers and 61.8% for pyrolyzed fibers. It corresponds to the decomposition of the lignocellulosic biomass made of hemicelluloses, cellulose and lignin. The mass loss is more important for untreated than for pyrolyzed ones. This suggests a higher amount of lignocellulosic compounds in the untreated fiber. The last zone (420900 C) corresponds to the decomposition of the fiber up to a constant mass.

For pyrolyzed fiber only cellulose decomposition takes place. The pyrolysis removes the hemicelluloses. To confirm the identification of the phenomenon observed by differential scanning calorimetry (DSC), thermal analysis has been performed on bagasse fiber and cellulose, hemicellulose, and lignin.73 The temperatures corresponding to the minimal heat flow are directly in relation with the distribution of botanical constituents in the fibers. Bilba et al.27 studied the effect of pyrolysis treatment on the amount of fiber on composite thermal degradation for arrow-root fibers. For the pyrolyzed arrow-root fiber a large band appears between 30 C and 240 C and two exothermal peaks, around 320 C and 460 C, are seen, identified as combustion of carbohydrates (320 C) and lignin (460 C).56,74

88

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Effect of pyrolysis treatment and amount of fibers on thermal degradation of composite Thermal degradation present two levels of interest: a practical one related to the durability of the material in the case of thermal solicitation and a fundamental one for the identification of phases, components, and phenomenon. The curves of heat flow versus temperature are realized using DSC in different zones, which correspond mainly to endothermal degradations.

4.3.2.5 Hydrosoluble components by NMR One important parameter in cement-bonded composite materials is setting. The presence of hydrosoluble components strongly affects setting.72 The identification of hydrosoluble compounds has been explored by nuclear magnetic resonance (NMR) on extracted alkaline solution from untreated or pyrolyzed fibers. The effect of alkaline treatments optimization has also been considered. The NMR results are compared to FTIR ones.

Study of hydrosoluble compounds extracted from untreated bagasse fibers The signals observed at 3.2 and 5.5 ppm in 1H spectrum (Fig. 4.6A) and 60 and 100 ppm for the 13C spectrum (Fig. 4.6B) suggest the presence of hemicellulose in the sample.75 The 13C spectrum presents six intense signals, which could indicate that hemicellulose is made of hexoses. These signals are at 98.6, 74.2, 72.2, 71.0, 70.4, and 66.3 ppm. Other less intense signals appear in the spectrum at 103.2, 103.1, and 53.7 ppm indicating the presence of other types of units in the extract. The correlation spectroscopy and heteronuclear single quantum correlation allows the identification of the base unit of the sample as α-D glucopyranose.

Study of hydrosoluble compounds extracted from treated fibers Hydrosoluble compounds from fibers 1, 2, 3, 7, and 8 (Table 4.3) have been studied by NMR. Comparison of 1H and 13C spectra with spectra of untreated fibers shows that all the samples are built from the same saccharide unit and that this unit is different from the repeat unit identified for untreated fibers. The signals observed in the 1H and 13C spectra of hemicellulosic extracts from treated bagasse fibers are characteristic of β-xylopyrannose units, substituted in 1!4.76 The β0 anomerie can be deduced from signals between 4.1 and 4.5 ppm. The structure of hemicellulosic fractions determined by NMR confirms the observations from infrared spectra. Indeed, the FTIR spectra of the fibers indicate on one part the presence of a xylopyranose structure among the hemicelluloses of the fibers, and on other part an alteration, more or less important, of the constituent when applying the treatments. The structure of xylopyranose type is the one funded in the hydrosoluble extracts of the treated fibers, as determined from 1H and 13C spectra. Comparison of the FTIR spectra of untreated fibers before and after soaking did not show difference of intensity of the characteristic band of the xylopyranose type. This observation is reinforced by the structure of glucopyranose type of the hemicelluloses extracted from the fibers.

6.0

5.5

5.0

4.5

4.0

(A)

3.5

3.0

2.5

2.0

ppm

(B)

Figure 4.6 RMN spectra of cristallized hydrosoluble compounds from untreated bagasse fibers.63 (A) RMN 1H spectra; (B) RMN 13C spectra of cellulose in D2O.

90

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

4.4

Influence of fiber treatment on setting of composites73

In cementitious materials, setting is a critical step, thus for different type of vegetable fibers the setting of composites has been examined. A detailed study was performed, which focused on the influence of fiber treatment on setting of composites. Vegetable fibers/cement composites of various compositions were prepared. Blended cement composites were prepared, mixing Portland cement with vegetable fibers and water. A thermos bottle was used as an adiabatic chamber. Time and temperature (60.1 C) were measured all along the setting. The setting temperature (the highest temperature of hydration) and the setting time (time to reach the highest temperature of hydration) were recorded. Plot of temperature versus setting time for the different paste composites (vegetable fibers/cement) prepared leads to similar profile curve (Fig. 4.7). Three zones can be observed in the graphs: G

An endothermic part during the first stage of hydration 40

Reference Cellulose Hemicellulose Lignin

Temperature (°C)

35

Raw bagasse

30

25

20

0

5

10

15 Duration (h)

20

25

30

Figure 4.7 Influence of the different botanical components of bagasse fiber of the setting of composite material.73 Reference: cement paste without fibers, cellulose: cement paste with 1 wt% of commercial cellulose powder (SigmaAldrich), hemicellulose: cement paste with 1 wt% of commercial hemicellulose powder (SigmaAldrich), lignin: cement paste with 1 wt% of commercial hemicellulose powder (SigmaAldrich), chemical treatment: cement paste with 1 wt% of chemically treated bagasse fibers.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

G

G

91

A second zone characterized by an exothermic peak A third zone of slow cooling during the final part of the setting

The setting of untreated vegetable fibers/cement composites is delayed compared with unreinforced cement paste (control).

4.4.1 Effect of chemical treatment of fibers The results of the chemical treatment (elimination of hemicellulose by acidic HCl hydrolysis) of bagasse fibers on the setting of composites are seen in Fig. 4.7, which shows that the consequences are a diminution of hydration temperature and an increase in the setting time. The behavior of the composite with chemically treated fibers is approximately the result of weighted additions of curves of cellulose/ cement and lignin/cement samples for the setting time.

40

Reference Cellulose Hemicellulose Lignin Chemical treatment

Temperature (°C)

35

30

25

20 0

5

10

15

20 25 Duration (h)

30

35

40

Figure 4.8 Influence of the thermal treatment of the bagasse fibers on the setting of the composites.73 Reference: cement paste without fibers; Raw bagasse: cement paste with 1 wt% of untreated bagasse fibers; Heat-treated bagasse 175 C: cement paste with 1 wt % of pyrolyzed bagasse fibers at 175 C; Heat-treated bagasse 200 C: cement paste with 1 wt% of pyrolyzed bagasse fibers at 200 C; Heat-treated bagasse 225 C: cement paste with 1 wt% of pyrolyzed bagasse fibers at 225 C; Heat-treated bagasse 250 C: cement paste with 1 wt% of pyrolyzed bagasse fibers at 250 C.

92

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

4.4.2 Effect of temperature of pyrolysis of fibers The influence of the thermal treatment of bagasse fibers before admixture with cement and water has also been studied. The results are presented in Fig. 4.8. The thermal treatment of bagasse improves the setting of composites, the setting being shortened and the maximum hydration temperature being obtained for composites with pyrolyzed fibers at 200 C and 250 C. As shown previously, pyrolysis of bagasse at these temperatures destroys the hydrosoluble and hemicellulose compounds unfavorable to the setting. The best compromise for the best setting, in the shortest time, is obtained for a mixture with a bagasse fiber heat treated at 200 C.

Figure 4.9 Setting of composites made of unpyrolyzed and pyrolyzed fibers coated with silane (S1 or S2).23 BS1: composite containing raw bagasse fibers coated with silane S1; PBS1: composite containing pyrolyzed bagasse fibers coated with silane S1; BS2: composite containing raw bagasse fibers coated with silane S2; PBS2: composite containing pyrolyzed bagasse fibers coated with silane S2.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

93

4.4.3 Effect of silane coating The effect of silane treatment on material setting was studied for composites reinforced by 2 wt% pyrolyzed or unpyrolyzed fibers treated with silane S1 or S2. The maximum temperature recorded during composites setting, here termed “setting temperature,” is shown in Fig. 4.9B. The corresponding setting time is seen in Fig. 4.9A. The treatment of unpyrolyzed fibers with 0.5 wt% solution of silane (S1 or S2) reduces setting time and increases the hydration temperature of composites. At this low concentration the water-repellent effect of silane enhances composite setting. Silane treatment avoids (1) competition between absorption of water and hydration of cement and (2) release of sugar by the fibers. After fibers treatment with 0.5 wt% of silane, the variation in temperature is the same. The highest settings are observed at lower silane concentration. The maximum setting temperature decreases to a minimum for 4 wt% silane for raw fibers and pyrolyzed fibers treated with silane S2. In the case of pyrolyzed fibers treated with silane S1, the minimum setting temperature is obtained after treatment with 2 wt% silane. Trends in the setting time are different for each silane. Treatment for S2 for both pyrolyzed and pyrolyzed leads to a constant setting time for all silane concentrations. The setting time increases with increasing content of silane S1. The effect is greater for pyrolyzed fibers. Composites with pyrolyzed fibers treated with silane S1 exhibit setting times 410 h longer than the corresponding composite with unpyrolyzed fibers. This observation confirms that this behavior is not related to the presence of free silane in the cement paste. To enhance the composite setting parameters, the optimal concentration of the silane solution for fiber treatment is less than 2 wt% to obtain the highest setting temperature and 0.5 wt% in order to have the shortest setting time.

4.5

Influence on physical-chemical properties of fibers and composites

4.5.1 Water absorption and wettability of fibers 4.5.1.1 Alkali treatment of sugar-cane bagasse fibers The impact of a lime solution (Table 4.3) on water absorption and wettability of sugar-cane bagasse fibers (0.41 mm) was investigated. The results showed that whatever the duration or temperature or lime-solution concentration, in the studied conditions, no influence on the water absorption of sugar-cane bagasse fibers was observed. Indeed, these alkali-treated fibers absorb around 11%13% of water relative to their dry mass. By using a K100 tensiometer (Kru¨ss, Germany), the contact angles between the treated sugar-cane bagasse fibers and water were determined. The contact-angle measurement shows the ability of a liquid to spread on a surface by wettability. The

94

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

method consists of measuring the angle of the tangent of the profile of a drop deposited on the substrate with the surface of the substrate. It allows measuring the surface energy of the liquid or of the solid. From the value of contact angle, it is thus possible to deduce the hydrophilic or hydrophobic characteristic of a surface. For all the raw and treated bagasse fibers, we obtained a contact angle θ  90 degrees in water. Consequently, it means that the surface of alkaline bagasse fibers is hydrophobic toward water. This study has to be completed by measurement of contact angle between those treated fibers and an alkaline solution whose pH would simulate that of a cementitious matrix.

4.5.1.2 Pyrolysis treatment of sugar-cane bagasse fibers

Mass of water absorbed (g of water/ g of fiber)

Rodier’s PhD thesis72 was about the study of the water absorption of raw and pyrolyzed (N2, 2 L/h, 2 h) sugar-cane bagasse fibers. They were short fibers ranging 0.41 mm in diameter. Fig. 4.10 shows the results. The curves of absorbed water masses by the raw and pyrolyzed sugar-cane bagasse fibers have similar patterns but differences reside in the absorbed water masses. The raw fibers absorb 2575 wt% of water in addition to the pyrolyzed fibers, whatever the duration. From 20 h, the curves reach a saturation plateau with an absorbed water mass of 4.2 g of water/g of fibers for the raw fibers and 2.7 g of water/g of fibers for the pyrolyzed fibers. Bilba and Ouensanga 28 showed that sugar-cane bagasse fibers heat-treated at 200 C are less hygroscopic than raw fibers. The hydrophilic nature of the fibers results from the formation of hydrogen bonds between the water molecules and the molecules exhibiting OH groups on the surface of the bagasse fibers.2 The decrease in the intensity of the band characteristic of the group -OH obtained by the 5

4

3

2

1 Raw bagasse fibers Pyrolyzed bagasse fibers

0 0

50

100

150

200

250

300

350

400

Time (h)

Figure 4.10 Water absorption of raw and pyrolyzed sugar-cane bagasse fibers.72

Treatments for viable utilization of vegetable fibers in inorganic-based composites

95

pyrolyzed fibers confirms the results obtained for the reduction of the mass of water absorbed by the pyrolyzed fibers (see Section 4.3.2.3).

4.5.1.3 Silane coating and pyrolysis treatment of sugar-cane bagasse fibers Bilba and Arse`ne23 compared the water absorption of, on the one hand unpyrolyzed sugar-cane bagasse fibers (noted B) treated by silane coating, and the other pyrolyzed sugar-cane bagasse fibers (noted PB) treated by silane coating. The silane used was alkyltrialkoxysilanes R0 Si(OR)3 (noted S1) and a dialkyl-dialkoxysilane (noted S2). The silane solutions concentrations were 0.5%, 4%, and 6% by volume. The water uptake by the different treated or untreated fibers was estimated under partial water pressure equal to 0.83 P0, where P0 is the saturation vapor pressure of pure water at 35 C. The main observations were the following: G

G

G

G

G

G

Fibers absorb from 4 to 10 wt% of their weight Unpyrolyzed fibers B are more hygroscopic than pyrolyzed fibers PB Water absorption is lower after silane treatment for both unpyrolyzed or pyrolyzed fibers Combination of pyrolysis and silane treatment has a synergistic effect, i.e., decreases the absorption of water by the fibers Silane S2 usually exhibits a greater repellent effect than silane S1 Using solutions of up to 6 wt% silane, water absorption decreases with increasing silane content of the treatment solutions

To conclude, the influence of treatments of sugar-cane bagasse fibers on water absorption and wettability of those fibers can be summarized as follows: G

G

G

Whatever the chosen parameters of alkali treatment (lime-solution concentration: 0.2 and 5 wt%, duration of impregnation of fibers into lime solution: 30 min and 2 h, temperature of treatment: 25 C and 100 C, volume of water in which the fibers are treated: 12.5 and 25 wt%), no influence was observed on water absorption or wettability of fibers. When using pyrolysis treatment (N2, 2 L/h, 2 h, 200 C), there is a decrease in the mass of water absorbed (63%) by the bagasse fibers, thus making them less hygroscopic. This point was confirmed by the work of Bilba and Arse`ne.23 Silane-coating treatment of fibers leads to less water absorption; the higher the silane solutions concentration, the lower the amount of water absorbed. Moreover, when silane coating is combined with pyrolysis, there is synergistic effect, i.e., a greater decrease of the water absorption by fibers.

4.5.2 Density behavior 4.5.2.1 Helium pycnometer density behavior of fibers Alkali treatment of sugar-cane bagasse fibers In the study where a four-variable experimental design was established (Table 4.3), it appears that this lime-solution treatment of fibers, whatever the parameters (duration, temperature or concentration, volume of water) in our conditions, does not

96

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Helium pycnometer density of sugar-cane bagasse fibers treated by lime solution and the influence of the different parameters and standard deviation63

Table 4.5

Type of fiber

Helium pycnometer density (g/cm3)

Untreated 1 2 3

1.66 1.67 1.82 1.72

4

1.71

5 6 7 8 Standard deviation

1.73 1.73 1.70 1.77 0.017

Concentration influence Duration of treatment influence Temperature influence Volume of water influence (water used for the treatment) Standard deviation

0.007 0.002 0.015 8.10217 0.017

Refer to fiber types described in Table 4.3.

Helium pycnometer density of different types of fibers and a comparison of treatments59

Table 4.6

Fibers Helium pycnometer density (g.cm23)

BAGb

BAGpyr

BAGcms

CCb

CCpyr

CCcms

0.70 [60.02]

1.13 [60.04]

1.26 [60.05]

1.27 [60.02]

1.28 [60.04]

1.32 [60.02]

BAGb, raw bagasse fibers; BAGpyr, pyrolyzed bagasse fibers; BAGcms, hornificated bagasse fibers; CCb, raw coir fibers; CCpyr, pyrolyzed coir fibers; CCcmc, hornificated coir fibers.

affect the helium pycnometer density of treated fibers as shown by Table 4.5. The one less affecting parameter is volume of water, while temperature leads to a small increase of the Helium pycnometer density of treated fibers.

Hornification (drying/rewetting cycles) of sugar-cane bagasse and coconut-coir fibers Hornification treatment consists of a process inspired by Rocha Ferreira et al.9 The first step of the hornification cycle is a rewetting at 25 C for 2 h to reach the maximum water-absorption capacity of the fibers, followed by drying at 80 C for 16 h

Treatments for viable utilization of vegetable fibers in inorganic-based composites

97

and a cool down to 25 C in order to avoid thermal shock to the fibers. The cycle is repeated 10 times. Hornification was used (see section Hornification (drying/rewetting cycles)) as treatment of bagasse fibers (BAG) and coconut-coir fibers (CC) in order to compare those treated fibers with raw (BAGb, CCb) and pyrolyzed fibers (BAGpyr, CCpyr). The hornificated fibers are called BAGcms and CCcms. All the fibers were 4 cm long. The diameters, obtained by optic microscope, are presented in Table 4.2. Three classes of diameters were observed for the coconut-coir fibers: fine, average, and thick diameters. The results concerning the real density are given in Table 4.6. For the same species, whatever the treatment, true density of fibers is of the same range. When one compares the real density according the nature of fibers, pyrolyzed bagasse fibers are lighter than pyrolyzed coir fibers. These behaviors may be related to the fact that pyrolyzed bagasse fibers exhibit higher amount of cellulose than pyrolyzed coir fibers (see Section 4.3.2.1). According to a material safety datasheet (FMC biopolymer) the bulk density of cellulose ranges from 0.2 to 0.5 g/cm3, i.e., cellulose is lighter than lignin, the density of which varies between 0.63 and 0.72 g/cm3.76

Silane coating and pyrolysis treatment of sugar-cane bagasse fibers Bilba and Arse`ne23 studied the pore-size distribution of untreated and pyrolyzed sugar-cane bagasse fibers, which is substantially the same for unpyrolyzed sugar-cane bagasse fibers (B) and pyrolyzed (PB). There are mainly macropores (.0.02 μm): 97.86% for unpyrolyzed and 97.84% for pyrolyzed sugar-cane bagasse fibers. The effect of pyrolysis is evident for the largest pores (50100 μm) obtained; they represent 15.59% from unpyrolyzed fibers and 17.26% from pyrolyzed fibers. Pyrolysis causes some loss of material, which is one of the main mechanisms involved in the creation of the largest pores. The corresponding real density measurement is 1.88 g/cm3 for unpyrolyzed fibers and 2.03 g/cm3 for pyrolyzed fibers. Regarding the botanical analysis, we noted that the increase in porosity (4.27%) corresponds to the release of water and extractives that represent 4.65% of the total mass of fibers. As extractive compounds that are present vary widely, this assertion cannot be based on numerical data alone but is also supported by thermal degradation studies of the fibers observed by DSC.23

Porosity and density of treated and untreated sugar-cane bagasse fibers23

Table

4.7

Total porosity (%) Density of porous matter (g/cm3) Helium pycnometer density (g/cm3)

B

PB

B6S

PB6S

69.00 0.59 1.88

74.00 0.65 2.03

91.26 0.30 3.48

85.69 0.49 3.41

B, untreated bagasse fibers; PB, pyrolyzed bagasse fibers; B6S, unpyrolyzed bagasse fibers treated with silane solution (6%); PB6S, pyrolyzed bagasse fibers treated with silane solution (6%).

98

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

The lighter compounds, mainly volatiles and extractives, are eliminated after the pyrolysis step. It is also possible to suggest the creation of pores that are closed or inaccessible to mercury intrusion. It is also interesting to note that porosity analysis confirms previous dimensional studies showing an increase in the dimensions of the fibers after silane treatment.29 For both unpyrolyzed and pyrolyzed fibers, treatment with 6 wt% silane enhances porosity as seen in Table 4.7. Pyrolyzed fibers that are initially more porous than unpyrolyzed fibers exhibit after silane treatment lower porosity than fibers only subjected to silane treatment. Raw fibers are more sensitive to the increase in porosity. These observations are in perfect agreement with the dimensional analysis. A hypothesis for the increase in porosity can therefore be proposed. When hydrated, the presence of water in the pore leads to swelling of the fibers. Upon drying, water is released and silanes act as a frame that maintains pores and cell walls in their hydrated configuration, with high pore volume, even with normal humidity. Silanes are known to react with hygroscopic surface OH groups.27 Chemical analyses indicate that the water content also decreases after pyrolysis. Pyrolysis reduces the hydrophilic properties of the fibers and consequently the number of OH groups on the surface. This decrease in OH groups could explain the lower sensitivity of pyrolyzed fibers treated with silanes. To conclude, the influence of treatments of fibers on the real density of those fibers can be summarized as follows: G

G

G

G

For the chosen parameters of alkali treatment, no influence was observed on macroscopic density of sugar-cane bagasse fibers. When applying pyrolysis treatment to sugar-cane bagasse fibers, there is an increase in the real density of fibers (160%) but the same treatment applied to coconut-coir fibers does not have any influence. Moreover, after pyrolysis, there is a decrease of porosity of sugar-cane bagasse fibers (7%), which is in agreement with dimensional variation. The hornification decreases the real density of bagasse fibers (55%) whereas it has no influence on coconut-coir fibers. Silane treatment increases real density (167% to 85%) and total porosity (115% to 132%) of bagasse fibers.

4.5.2.2 Helium pycnometer density behavior of cement paste composites elaborated with treated fibers A preliminary study at the laboratory COVACHIM-M2E aimed to determine the true density of ordinary Portland cement pastes reinforced with 5 wt% of coconutcoir fibers or 5 wt% of coconut-leaf sheaths, sieved from 0.4 to 1 mm of diameter. Those fibers were used raw (GFB, CCB) or pyrolyzed N2, 2 L/h, 2 h—(GFP, CCP) or alkali-treated attack by a 5 wt% alkaline solution of Ca(OH)2—(GFBA, CCBA). The mean value of the measured helium pycnometer densities of the samples is shown in Fig. 4.11. The addition of the fibers in cement paste reduces the helium pycnometer density, creating voids or pores within the material. This is the result of the increase in volume of the composite material and thus a loss in weight. The average values of density of the cement pastes reinforced with

Treatments for viable utilization of vegetable fibers in inorganic-based composites

99

Heliumpycnometerdensityofcementpastes (g/cm–3)

3

2

1

0 REF28

GFB28

CCB28 GFBA28 CCBA28 GFP28

CCP28

Figure 4.11 Helium pycnometer density of cement pastes reinforced with coconut-coir fibers or coconut-leaf sheaths at 28 days.77 REF28: control (cement paste) at 28 days; GFB28: cement paste/raw coconut-leaf sheaths fibers at 28 days; GFBA28: cement paste/alkali coconut-leaf sheaths fibers at 28 days; GFP28: cement paste/pyrolyzed coconut-leaf sheaths fibers at 28 days; CCB28: cement paste/raw coconut-coir fibers at 28 days; CCBA28: cement paste/alkali coconut-coir fibers at 28 days; CCP28, cement paste/pyrolyzed coconut-coir fibers at 28 days.

CC are lower than that of the cement pastes reinforced with leaf sheaths (GF). Treatments mainly affect composites reinforced with coconut-coir fibers.

4.5.2.3 Apparent density78 and water-accessible porosity78 of mortar composites elaborated with treated fibers Pyrolysis treatment of sugar-cane bagasse fibers According to Rodier’s PhD thesis,72 a study was carried out on composite materials obtained by mixing raw or pyrolyzed sugar-cane bagasse fibers (0.41 mm of thickness, 26 wt% of the dry constituents of matrix) with either a modified ternary matrix or commercial cement. The matrix called “ternary” was a cementitious mixture of Portland cement without mineral addition (CEM I 52.5 N, 80 wt%), natural pozzolan (NP, 15 wt%), and bagasse ashes (BAs 5 wt%). Hence, in this ternary matrix, 20 wt% of the ordinary Portland cement was replaced by NP and BAs. Another commercial Portland cement, called CLA, was used as a reference in this study. It is a CEM II 32.5 N containing 83 wt% of clinker and approximately 17 wt% of NP. In this way, the ternary matrix and CLA present about the same amount of mineral replacement and the same NP.79 Composites were made by mixing the matrix (ternary or CLA) to an amount of raw or pyrolyzed sugar-cane bagasse fibers ranging from 2 to 6 wt%. The method used to name the sheets of composites was as follows: the name of the matrix

Figure 4.12 Apparent density of composites elaborated with (A) cement CLA and (B) ternary matrix at 28 and 90 days (curing chamber).72

Treatments for viable utilization of vegetable fibers in inorganic-based composites

101

(ternary or CLA) is followed by xB or xP (B for raw, P 5 pyrolyzed and x 5 2, 4, 6 wt%). For aging, the composite pastes were placed in a curing chamber (23 C, 50% relative humidity) as the standard curing conditions or immersed in water at 23 C and 40 C as an accelerated aging process. The samples were placed in the standard curing conditions for 28 and 90 days and in water for 180 and 360 days to simulate the behavior of the materials in tropical weather conditions after a few years of rainy season. The apparent densities of the composites produced from the ternary matrix and commercial cement CLA at 28 and 90 days in a curing chamber are shown in Fig. 4.12. At 28 days, taking into account the standard deviations, the pyrolysis of the bagasse fibers has no influence on the apparent densities of the composites produced from the ternary matrix. At the same date, the same behavior was observed for composites produced with commercial cement, CLA, except for a content of 6 wt% of fibers, for which the treatment of bagasse fibers decreased by 4.57% the apparent density of the composite. At 90 days, for content of 4 and 6 wt% of fibers, the pyrolysis of the fibers decreases by 2.81% and 7.45%, respectively, the apparent densities of the composites elaborated with the ternary matrix. The same behavior is observed for composites made from commercial cement. Decreases of 2.89% and 2.27%, respectively, were noted. The apparent densities of ternary matrix composites are generally higher than those of commercial cement. The phenomenon could be explained by the differences in the composition of the cements and the fact that the helium pycnometer density of the ternary matrix is higher than that of CLA (1.57 6 0.01 g/cm3). The water-accessible porosity of the composites elaborated with the ternary matrix and the commercial cement CLA at 28 and 90 days, after curing chamber, are presented in Fig. 4.13. At 28 days, taking into account the standard deviations, the pyrolysis of the bagasse fibers did not have a significant effect on the water-accessible porosity of the composites for amounts of 2 wt% of fibers. For a content of 4 wt% of fibers, the pyrolysis of the fibers slightly increases (6.1%) the water-accessible porosity of the composite elaborated with commercial cement CLA. However, for the composite prepared from the ternary matrix, the pyrolysis of the bagasse fibers has no significant influence taking into account the standard deviations. For a content of 6 wt% of fibers, the pyrolysis of the fibers reduces the wateraccessible porosity of the composite produced from the ternary matrix by 10.48%. On the other hand, the decrease is 20.41% for the composite elaborated with commercial cement CLA. At 90 days, the pyrolysis of the bagasse fibers did not have a significant effect on the water-accessible porosity of the composites produced with the ternary matrix or commercial cement CLA, except for the composite containing 4 wt% of fibers (CLA), for which a decrease of 9.23% of the water-accessible porosity is observed.

Figure 4.13 Water-accessible porosity of composites elaborated with (A) cement CLA and (B) ternary matrix at 28 and 90 days (curing chamber).72

Figure 4.14 Apparent densities of the composites produced from the ternary matrix (POU15BAG5) with (A) raw bagasse fibers and (B) pyrolyzed bagasse fibers and (POU15BAM5) with (C) raw bagasse fibers and (D) pyrolyzed bagasse fibers at 180 and 360 days cured in various environments.72

104 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

The apparent densities of the composites produced from the ternary matrices (POU15BAG5, POU15BAM5) at 180 and 360 days cured in various environments are shown in Fig. 4.14. In this part of study, Rodier72 worked with two kinds of cementitious matrices: partial substitution by (1) volcanic natural pozzolan (NP) or bagasse ashes (BAs) (POU15BAG5) and (2) NP or 5 wt% of bamboo-trunk ash (POU15BAM5). At 180 days the pyrolysis of the bagasse fibers leads to a decrease in the apparent densities of the composites produced with bagasse ash and aged in an aqueous bath at 23 C. At this time, it is necessary to increase the temperature of the bath to 40 C in order to observe the same behavior for the composites made with bambootrunk ash. This could be due to degradation of the fibers in the matrix, which is more consistent at 40 C than at 23 C in the aqueous baths. At 360 days pyrolysis had no significant influence on the apparent densities of composites made from bagasse ash and bamboo-trunk ash, with the exception of composites containing 4 and 6 wt% of fibrils and aged in an aqueous bath at 23 C and in climatic chamber. The water-accessible porosity of the composites produced from the ternary matrices (POU15BAG5, POU15BAM5) at 180 and 360 days cured in various environments are shown in Fig. 4.15. At 180 and 360 days for the available data, taking into account the standard deviations, the heat treatment of bagasse fibers has no significant effect on wateraccessible porosity for composites aged in a climatic chamber, whatever the type of ash used. For composites aged in an aqueous bath at 40 C, the heat treatment of the bagasse fibers reduces the water-accessible porosity of composites made with bamboo-trunk ash at 180 days. At that date, the same behavior was observed for 2 wt% of fiber content for composites made with bagasse ashes. At 360 days, taking into account the standard deviations, the pyrolysis of bagasse fibers has no significant influence on the water-accessible porosities for composites, with the exception of the composite prepared with bagasse ash containing 4 wt% of fibers. At 180 days for composites aged in aqueous bath at 23 C, taking into account the standard deviations, the heat treatment of bagasse fibers does not have a significant influence on the water-accessible porosity of composites elaborated with bamboo-trunk ash, except those containing 4 wt% of fibers. However, for composites made from bagasse ash, the heat treatment of bagasse reduces the wateraccessible porosity of composites up to 4 wt% of fibers. At 360 days, the pyrolysis of bagasse fibers decreases the water-accessible porosity for composites made with bamboo-trunk ash up to a content of 4 wt% of fibers. For composites made with bagasse ash, this content does not exceed 2 wt% of fibers.

Alkali treatment and pyrolysis treatment of sugar-cane bagasse fibers One´sippe et al.80 studied the real density of CBAGP and CBAGB. These acronyms are for composites made of cement matrix reinforced with different amounts of vegetable fibers (varying from 0 to 3 wt%). The fibers where sugar-cane bagasse fibers ranging between 0.4 and 1 mm. Two fiber treatments were performed:

Climatic chamber 180d

Water 23°C 180d

Water 40°C 180d

Climatic chamber 360d

Water 23°C 360d

Water 40°C 360d

(B) POU15BAG5 Pyrolyzed fibers

90

90

80

80

Water accessible porosity (%)

Water accessible porosity (%)

(A) POU15BAG5 Raw fibers

70 60 50 40 30 20 10

Water 23°C 180d

Water 40°C 180d

Climatic chamber 360d

Water 23°C 360d

Water 40°C 360d

70 60 50 40 30 20 10

0

0 2

(D) POU15BAM5 Pyrolyzed fibers

4 Fiber content (%)

6

2

(C) POU15BAM5 Raw fibers Climatic chamber 180d

Water 23°C 180d

Water 40°C 180d

Climatic chamber 360d

Water 23°C 360d

Water 40°C 360d

80 70 60 50 40 30 20

4 Fiber content (%)

6

Climatic chamber 180d

Water 23°C 180d

Water 40°C 180d

Climatic chamber 360d

Water 23°C 360d

Water 40°C 360d

90 Water accessible porosity (%)

90

Water accessible porosity (%)

Climatic chamber 180d

80 70 60 50 40 30 20 10

10

0

0 2

4 Fiber content (%)

6

2

4 Fiber content (%)

6

Figure 4.15 Water-accessible porosity of the composites produced from the ternary matrix (POU15BAG5) with (A) raw bagasse fibers and (B) pyrolyzed bagasse fibers and (POU15BAM5) with (C) raw bagasse fibers and (D) pyrolyzed bagasse fibers at 180 and 360 days cured in various environments.72

106 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 4.8 Density and water-accessible porosity of cement composites reinforced with bagasse fibers80 Method

He Hg

Sample

Control

CBAGP

CBAGB

wt% fiber

0

1.5

3

1.5

3

Helium pycnometer density (g.cm23) Mercury-intrusion pycnometry density (g.cm23) Mercury-intrusion porosity (%)

2.6648 [60.0108] 2.49923

2.8277 [60.0223] 1.6155

2.5526 [60.0091] 1.4561

2.8316 [60.0045] 1.5744

2.5681 [60.0186] 1.6024

64.98

43.85

44.23

48.73

47.89

CBAGP, composites with pyrolyzed bagasse fibers; CBAGB, composites with alkaline bagasse fibers.

G

G

Pyrolysis under controlled atmosphere (N2, 2 L/h, 2 h, 200 C) Chemical treatment: attack by a 5 wt% alkaline solution of Ca(OH)2

Matrix was made of ordinary Portland cement, sand, water, CaCO3, bentonite, silica fume, acrylic styrene polymer, and cellulose pulp. Table 4.8 presents the helium pycnometer density results of this study. As reported in literature81 the authors observed that the higher the fiber content (up to 3 wt%) the lighter the composite (i.e., the lower the density). No effect of fiber treatments on the macroscopic (method He) density of composites was noticed. The mercury intrusion porosity of the matrix (control specimen) decreased by the addition of fibers. In the range of this study, whatever the fiber content or treatment, the porosity values are of the same order. The influence of treatment of fibers on density of cement composites reinforced by these fibers can be summarized as follows: G

G

G

Alkali treatment or pyrolysis treatment of sugar-cane bagasse fibers have no effect on real density or the porosity of composites. However, the addition of fibers to the matrix decreases the real density of composites (58% to 72% according to treatment and amount of fibers). For curing in climatic chamber (23 C, 50% RH): At 90 days, the pyrolysis of the bagasse fibers leads to a decrease (from 2% to 7%) in the density of the composites made from ternary or CLA matrices reinforced by the treated fibers. The pyrolysis of the bagasse fibers leads to the reduction (from 5% to 20%) of the wateraccessible porosity of the composites containing 6 wt% of fibers prepared from the CLA matrix at 28 days.

For curing in aqueous bath (23 C and 40 C): G

G

The apparent density of the composites increases with the fiber content up to 32%, at 180 days. At 360 days, the environment and the pyrolysis of the fibers have no significant influence on the apparent density of composites, regardless of the type of ash. The pyrolysis has no effect on water-accessible porosity of composites.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

107

Thermal conductivity of treated sugar-cane bagasse fibers according to modeling80

Table 4.9

Pyrolyzed bagasse fibers wt% of fibers Volume fraction of fibers Helium pycnometer density (g/cm3) Thermal conductivity (W/m/K)

4.6

1.5 0.1909 1.6155

3 0.2818 1.4561 0.1781

Alkaline bagasse fibers 1.5 0.2115 1.5744

3 0.2096 1.6024 0.1092

Influence on thermal and mechanical properties

4.6.1 Thermal conductivity 4.6.1.1 Thermal conductivity of alkali-treated and pyrolyzed fibers Due to the process required (hot-wire method) for the determination of thermal conductivity, it is very difficult to measure the thermal conductivity of vegetable fibers because this method requires two identical planar samples. However, One´sippe et al.80 proposed for the first time modeling based on the MaxwellEucken equation allowing estimation of the thermal conductivity kf of the treated bagasse fibers used in their study. The main results, with a good correlation (R2 5 0.99), are given in Table 4.9. According to these modeling results, it appears that thermally treated sugar-cane bagasse fibers exhibit higher thermal conductivity than alkali-treated ones. Bagasse fibers belong to the Poacea family.82 According to Ramanaiah et al.83 thermal conductivity of broom grass (which belongs to the Poaceae family) is evaluated around 0.1303 W/m/K. In fact, there is no bibliographic study estimating the thermal conductivity of the raw bagasse fibers, so it assumed to be similar to broom grass. In this logic, it could be said that the pyrolysis increases the thermal conductivity of the fibers while the alkaline treatment decreases it. Within the framework of the problem of thermal insulation in buildings in tropical environments, the thermal treatment of fibers appears to be the best.

4.6.1.2 Thermal conductivity of cement paste elaborated with treated fibers In this section, sugar-cane and coconut-leaf sheaths fibers were treated by pyrolysis or alkaline solutions. The mean values of the measured thermal conductivities are shown in Fig. 4.16. Ordinary Portland cement paste reinforced with coconut fibers has relatively low thermal conductivity ranging from 0.50 to 0.60 W/m/K. This corresponds to a

108 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 4.16 Thermal conductivity of cement paste reinforced with coconut-coir fibers or coconut-leaf sheaths at 28 days.77 REF28: control (cement paste) at 28 days; GFB28: cement paste/raw coconut-leaf sheaths fibers at 28 days; GFBA28: cement paste/alkali coconut-leaf sheaths fibers at 28 days; GFP28: cement paste/pyrolyzed coconut-leaf sheaths fibers at 28 days; CCB28: cement paste/raw coconut-coir fibers at 28 days; CCBA28: cement paste/alkali coconut-coir fibers at 28 days; CCP28: cement paste/pyrolyzed coconut-coir fibers at 28 days.

Figure 4.17 Thermal conductivity of composites reinforced with treated sugar-cane bagasse fibers (365 day old specimen).80 CBAGP: composites reinforced with pyrolyzed bagasse fibers; CBAGB: composites reinforced with alkaline bagasse fibers.

significant decrease (40%) in the thermal conductivity of cement paste, which is in agreement with Khedari et al.80 It can be observed that the pyrolyzed fibers materials have a lower thermal conductivity than other reinforced cement pastes and that the most low thermal conductivity values were obtained for the coconut-coir fibers/cement pastes.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

109

4.6.1.3 Thermal conductivity of mortar composites elaborated with treated fibers Alkali treatment and pyrolysis treatment of sugar-cane bagasse fibers Thermal conductivity has been found (hot-wire method) for composites CBAGP and CBAGB,80as shown in Fig. 4.17. As usual in the literature78,81,84 adding vegetables fibers decreases the thermal conductivity of matrix because the thermal conductivity of porous medium is inversely proportional to the voids in the specimen and the voids occur from packing of fibers (especially short fibers). According to One´sippe et al.80 the effect of fiber treatment on thermal conductivity of composites is stronger for pyrolyzed bagasse fibers than for alkaline bagasse fibers. The authors explain that fact by the chemical composition: alkaline bagasse fibers exhibit higher amounts of cellulose and extractives than for pyrolyzed bagasse fibers.

Pyrolysis treatment of sugar-cane bagasse fibers and aging of composites In this section, an analysis of the work of Rodier’s PhD thesis72 is proposed. This work was about the comparison of (1) ternary matrix reinforced by raw or pyrolyzed sugar-cane bagasse fibers and (2) cement (CLA) reinforced by the same fibers. For aging, the composite pastes were placed in a curing chamber (23 C, 50% RH) as the standard curing conditions or immersed in water at 23 C and 40 C as an accelerated aging process. The samples were placed in the standard curing conditions for 28 and 90 days and in water for 180 and 360 days to simulate the behavior of the materials in tropical weather conditions after a few rainy seasons. The main results were as follows79: G

G

G

In general terms, adding fibers to matrix decreases, as described in literature, the thermal conductivity of composites. The effect of fiber treatment on thermal conductivity of mortar composites is stronger for pyrolyzed bagasse fibers than for alkaline bagasse fibers, i.e., mortar composites made with pyrolyzed bagasse fibers allow a lower diffusion of heat than those made with alkaline bagasse fibers. Ordinary Portland cement paste reinforced by thermally treated coconut-coir fibers have lower thermal conductivity than composites reinforced by coconut-leaf sheaths fibers.

For curing in climatic chamber (23 C, 50% RH): G

The incorporation of raw and pyrolyzed bagasse fibers in the ternary matrix and in the CLA leads to a decrease (45%) of the thermal conductivity of the composites.

For curing in aqueous bath (23 C and 40 C): G

At 180 and 360 days the pyrolysis of the bagasse fibers and the type of ash used in the cementitious matrix do not have a significant influence on the thermal conductivity of the composites.

110 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Comparison of pyrolysis and hornification on specific heat (50 C) of sugar-cane bagasse fibers and coconut-coir fibers59

Table 4.10

Fibers Specific heat (J.g21.K21)

Bb

Bpyr

Bcms

Cb

Cpyr

Ccms

1.75 [60.60]

1.52 [60.12]

1.69 [60.12]

1.83 [60.01]

1.73 [60.02]

1.83 [60.01]

Bb, raw bagasse fibers; Bpyr, pyrolyzed bagasse fibers; Bcms, hornificated bagasse fibers; Cb, raw coir fibers; Cpyr, pyrolyzed coir fibers; Ccms, hornificated coir fibers.

4.6.2 Specific heat capacity (Cp) 4.6.2.1 Specific heat of treated fibers: pyrolysis treatment and hornification of sugar-cane bagasse fibers and coconutcoir fibers According to the hornification study (see section Hornification (drying/rewetting cycles)), when bagasse fibers are hornificated, they show higher values of specific heat than pyrolyzed or raw fibers (Table 4.10). Knowing that the specific heat represents the amount of energy to be brought by heat exchange to increase by one kelvin the temperature of one gram of material, in terms of thermal insulation, it is more advantageous to use the pyrolyzed bagasse fibers. The hornification does not seem to have an effect on the specific heat of the thicker coir fibers. Moreover, taking into account the standard deviations, in the context of thermal insulation, the most suitable fibers are coir fibers.

4.6.2.2 Specific heat (Cp) of cement paste elaborated with treated fibers In this section sugar-cane bagasse fibers and coconut-leaf sheaths fibers were pyrolyzed and alkali treated. The addition of coconut fibers to cementitious paste leads to a significant decrease of 65% of composite materials, Cp. A less significant reduction of Cp ( 8%) can be observed for the cement paste reinforced with alkali-treated coconut-coir fibers in comparison with cement paste reinforced with raw fibers. Taking into account the standard deviations, the higher specific heat is obtained for the cement paste reinforced with thermally treated coconut-coir fibers, i.e., this formulation is the best suitable for thermal insulation problem.

4.6.2.3 Specific heat (Cp) of mortar composites elaborated with treated fibers Alkali treatment and pyrolysis treatment of sugar-cane bagasse fibers in mortar matrix Whatever the treatment of fibers (Fig. 4.18), adding fibers to the cementitious matrix leads to a decrease of specific heat value. This decrease is more pronounced for higher fiber content.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

111

Figure 4.18 Specific heat (30 C) of composites reinforced with treated sugar-cane bagasse fibers (365 days old specimen).80 CBAGP: composites reinforced with pyrolyzed bagasse fibers; CBAGB: composites reinforced with alkaline bagasse fibers.

Considering the standard deviations, no effect of fiber treatment could be observed on the specific heat of composites CBAGB with 3 wt% of fiber content according to One´sippe et al.80 But with 1.5 wt% fiber content, the pyrolysis decreases the specific heat more than the basic treatment.

Pyrolysis treatment of sugar-cane bagasse fibers in ternary matrix In their study of ternary cementitious matrix, Arse`ne et al.79 measured the specific heat. The influence of treatments of fibers on specific heat of composites reinforced with these fibers can be summarized as follows, in the case of bagasse fibers: G

G

G

In general terms, adding fibers to cementitious matrices decreases (up to 28%) the specific heat of composites. For low fiber content (1.5 wt%), the pyrolysis decreases (17%) the specific heat more than the basic treatment. The type of fibers (bagasse or coconut-leaf sheaths fibers) has an impact on the specific heat: indeed, coconut fibers reduce the specific heat more than bagasse fibers.

For curing in climatic chamber (23 C, 50% RH): G

G

No effect of the pyrolysis on specific heat of the composites was observed. For curing in aqueous bath (23 C and 40 C): No effect of the pyrolysis on specific heat of the composites was observed.

112 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 4.19 Tensile strength of treated and untreated fibers.18

Figure 4.20 Stress at break of (A) sugar-cane bagasse fibers and (B) thick coconut-coir fibers.59 Bb: raw bagasse fibers; Bpyr: pyrolyzed bagasse fibers; Bcms: hornificated bagasse fibers; Cb: raw coir fibers; Cpyr: pyrolyzed coir fibers; Ccmc: hornificated coir fibers.

4.6.3 Tensile strength, Young modulus, stress, and energy at break 4.6.3.1 Fibers mechanical properties Pyrolysis treatment of sugar-cane bagasse, banana, and coconut fibers18 According to Arse`ne et al.18 the strongest pyrolyzed fibers are the bagasse fibers (426 6 335 MPa) and banana-leaf fibers (22 6 7 MPa) are the weakest. Note that this partly due to the nature of the plant and the location in the plant from which the fibers were extracted. Furthermore, a wide variability of strength was observed between different fibers of the same species, as illustrated by the large standard deviations. The elongation to failure was also shown to be similar (  7%9.2%) for most of the fibers. However, the coconut-sheath fibers exhibited a higher elongation to failure, presumably as a result of the bidimensional structure of these fiber networks in the plant.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

113

In the case of the chemically treated fibers, an analysis of the statistical variations in the fibers strengths was realized. The authors found that the strength variations of pyrolyzed fibers were well characterized by the lognormal distribution. The mean values associated with this distribution are presented in Fig. 4.19. These clearly show that the fiber strengths increase by a factor between 3 and 5 after pyrolysis treatment. Alkali attack has no significant effect on fiber strength, while acid attack degrades fiber strength for banana and bagasse fibers.

Hornification and pyrolysis treatment of sugar-cane bagasse fibers and coconut-coir fibers It is necessary to remember that the hornification treatment (cms) (see section Hornification (drying/rewetting cycles)) was applied to untreated sugar-cane bagasse fibers (Bb) and untreated coconut-coir (Cb) fibers, which was compared to the pyrolysis treatment (Bpyr and Cpyr). The stresses at break of the fibers are shown in Fig. 4.20A and B. The results for the coir fibers are presented for the thick class diameter (because they are the most significant). The standard deviations show a great variability in the breaking strengths of the fibers studied, as already shown by others.9,18 For bagasse fibers, the treatment has no influence on the tensile strengths considering the standard deviations. The treatments would not affect the strength of the bagasse fibers, which is confirmed by the botanical analysis showing similar cellulose content (see Table 4.11). On the other hand, for coir fibers, the treatments led to an increase in the tensile strength. In fact, the tensile stresses of the treated fibers are higher compared to the raw fibers; the treatments increase the tensile stress. These results are in line with those of other authors9,18 who found an increase in stress at the break for the treated bagasse and banana-trunk fibers. The Young’s modulus (E) or the longitudinal modulus of elasticity makes it possible to evaluate the stiffness of the fibers, as shown in Fig. 4.21A and B. The Young’s modulus increases after pyrolysis for the bagasse fibers but remains substantially equal to the drying/wetting cycle. These results are in agreement with those of the botanical analysis (Table 4.5) showing that the lignin, which, as can be recalled, confers its rigidity on the fibers and has a higher content following pyrolysis. Concerning thicker coir fibers the treatments increase the Young’s modulus. Indeed, these results are in agreement with the botanical analysis.18 The energy at the break accounts for the ductility of the fibers, i.e., their ability to withstand the propagation of cracks; this energy at break was calculated from the area under the curve strength 5 f(stress). The Mean and standard deviations of break energy measurements per fiber type were calculated and are listed in Table 4.12. Taking into account the standard deviations, the treatments do not influence the breaking energy of the treated fibers. The bagasse fibers would be more resistant to the propagation of cracks than the coir fibers.

Table 4.11

Chemical composition of untreated and hornificated fibers59 Fibers

Chemical Composition in principal compounds

Plant

Name

Treatment

Lignina

Extractivesa

Cellulosea

Hemicellulosea

Sugar-cane Bagasse

B BP

 Pyrolysis

Bcms CC

Drying/ rewetting cycles 

CCP

Pyrolysis

CCcms

Drying/ rewetting cycles

21.96 35.10 [ 6 1.94] 15.58 [ 6 3.47] 22.17 [ 6 9.00] 36.49 [6.47] 42.22 [ 6 1.20]

3.92 2.33 [ 6 0.63] 1.43 [ 6 0.19] 2.04 [ 6 0.12] 3.68 [ 6 0.16] 1.11 [ 6 0.10]

48.68 45.84 [ 6 4.84] 43.39 [ 6 3.56] 33.26 [ 6 6.87] 38.80 [ 6 3.39] 35.18 [2.25]

25.46 14.36 [ 6 3.2] 27.97 [ 6 6.33] 20.07 [ 6 2.95] 14.51 [ 6 0.69] 13.42 [1.18]

Coconut Coir

a

Weight % of principal compounds in dry fibers (hemicellulose, cellulose, lignin, extractives). Weight % of dry matter.

b

Totalb of principal compounds 100 97.63 88.37 77.47 93.48 91.93

Treatments for viable utilization of vegetable fibers in inorganic-based composites

115

Figure 4.21 Young’s modulus of (A) untreated and treated sugar-cane bagasse fibers and (B) untreated and treated coconut-coir fibers.59 Bb: raw bagasse fibers, Bpyr: pyrolyzed bagasse fibers, Bcms: hornificated bagasse fibers, Cb: raw coir fibers, Cpyr: pyrolyzed coir fibers, Ccmc: hornificated coir fibers.

Energy at break of different fibers and comparison of the type of fibers and treatments59

Table 4.12

Fibers

Bb

Bpyr

Bcms

Cb thick

Cpyr thick

Ccms thick

Energy at break (1022 J)

44.45 [633.64]

48.45 [635.40]

59.61 [628.66]

17.87 [619.88]

50.01 [6130.69]

69.02 [648.88]

Bb, raw bagasse fibers; Bpyr, pyrolyzed bagasse fibers; Bcms, hornificated bagasse fibers; Cb thick, thick raw coir fibers; Cpyr thick, thick pyrolyzed coir fibers; Ccmc thick, thick hornificated coir fibers.

10

Sugar cane Bagasse

Banana trunk

Cement paste (Reference)

9 8

Strength (MPa)

7 6 5 4 3 2 1 0 Pyrolysis

Acid treat.

Basic treat

Raw fiber

No fiber

NFRC = Natural Fiber Reinforced Composites Figure 4.22 Three point-bending strength at 28 days of NFRC with treated and untreated sugar-cane bagasse and banana-trunk fibers. The samples were stored in air at 24 6 1 C and relative humidity 5 85 6 2%.18 NFRC: Natural fiber-reinforced composites.

116 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

4.6.3.2 Composite mechanical properties Pyrolysis treatment of sugar-cane bagasse, banana, and coconut fibers Arse`ne et al.18 also evaluated the mechanical behavior of paste composites elaborated with these treated fibers. The matrix was made of ordinary Portland cement. This mechanical study allows the selection of the treatment, the type of fibers, and the amount of fibers best suited for good mechanical properties of composites. The effects of fiber treatment on composite strength are shown in Fig. 4.22. With the exception of fibers exposed to acid attack, the reinforcement with bagasse fibers improves the strength of cement paste while banana fibers, even those pyrolyzed, do not significantly increase the strength of composites. In the case of sugar-cane bagasse composites, the addition of raw fibers doubles the strength compared to the unreinforced cement paste. For sugar-cane bagasse, the acid treatment degrades the composites strength. The pyrolysis treatment conducts to an increase of mean value of the composite strength, but the difference is not statistically significant to conclude an enhancement of the strengthening effect of the fibers after pyrolysis. The effect of alkali treatment is weaker but with the same trend as the pyrolysis treatment. Strength slightly increases with increasing reinforcement volume fraction in the case of sugar-cane bagasse fibers exposed to alkali attack. The best properties were obtained after pyrolysis treatment of sugar-cane bagasse. For banana fibers, the effects of pyrolysis or chemical treatment are very limited. Also, such treatments do not increase the strength of the composites. The effects of chemical treatment are not directly correlated with the strength of the treated fibers. The treated banana-trunk fibers do not significantly increase the composite strength. In the case of banana-trunk or sugar-cane bagasse, the composites are strengthened with bagasse fibers. The optimal amount of sugar-cane bagasse fibers is 2 wt%. In the case of composite reinforced with 15 wt% of pyrolyzed bananatrunk fibers, there is no significant difference between the average bending strength of the composites and that of plain cement paste. However, the strength generally decreases with increasing fiber content. Maximum strength was obtained for composites reinforced with 1 wt% of banana-trunk fibers.

Alkali treatment and pyrolysis treatment of sugar-cane bagasse fibers and coconut-leaf sheath fibers When using coconut-coir fibers and coconut-leaf sheath fibers, the effects of thermal and chemical treatments on bending strength of ordinary Portland cement paste composites are also increased (125% for pyrolyzed coconut-leaf sheath and 118.5% for pyrolyzed coconut-coir fibers). It has been demonstrated28 that the fibers treated by pyrolysis or basic hydrolysis have higher cellulose content and a considerable decrease in extractives than raw fibers. This means that treatment of the fibers leads to a dissolution of the

Treatments for viable utilization of vegetable fibers in inorganic-based composites

117

disorganized particles or molecules, conferring surface roughness to the fibers, which improves adhesion between fibers and cementitious matrix.

Pyrolysis treatment of sugar-cane bagasse fibers: effect of curing conditions According to Arse`ne et al.79, after 28 and 90 days of curing the decrease of bending strength is always higher for CLA composites than for ternary composites, whatever the age and amount of added raw fibers; any trend is noted for composites made of pyrolyzed fibers. This shows that either chemical composition of the matrix, porosity, or fiber/matrix adhesion influences the flexural strength of materials. The lowest decrease is noted in ternary composites containing raw fibers showing that raw fibers are less degraded in the presence of the ternary matrix, probably because less calcium hydroxide, which conducts later on to mineralization of fibers and reduces their mechanical properties, is produced in the presence of ternary matrix. The authors suggest this related behavior is due to composite porosity. Whatever the age, taking into account standard deviations, flexural strengths of composites cured in water, either at 23 C or 40 C, are similar; they are superior to the ones of materials cured in curing chamber. Flexural strengths slightly decrease with duration of curing. Whatever the curing environment, pyrolysis of fibers does not have any effect on flexural strength. The highest strengths are noted for composites cured in water, probably because the presence of external water favors formation of CSH (hydrated calcium silicates) responsible for mechanical properties of materials.71 The influence of fiber treatment on the mechanical properties of fibers can be summarized as follows: G

G

G

G

G

G

G

G

The tensile strengths of coconut, banana, and bagasse fibers vary over a wide range. The fiber strengths are greater in the pyrolyzed condition. The strengths of the pyrolyzed fibers can be ranked as follows: bagasse, banana-trunk, coco-sheath, coconut-coir, and bananaleaf fibers. Pyrolysis treatment increases tensile strength of the fibers, at least by a factor 3, in the case of bagasse. The tensile strength increases slightly for banana-trunk fibers and decreases slightly for sugar-cane bagasse fibers after alkaline attack, but decreases with acid attack. The hornification treatment would not affect the strength of the bagasse fibers; on the contrary pyrolysis and hornification treatments increases tensile strengths of coconut-coir fibers (thick diameter). Concerning the influence of fiber treatments on the mechanical properties of composites: The three-point bending strengths of the composites reinforced with pyrolyzed banana fibers do not present significant differences with plain cement paste. The bending stress of composites reinforced with pyrolyzed coconut-leaf sheath fibers is higher than with other kind of fibers or treatment. For different curing (climatic chamber 23 C, 50% RH, or aqueous bath 23 C and 40 C): At 90 days, the fibers degrade the bending strength of the composites, no matter the treatment of the fibers and no matter the binder (curing chamber). Curing in water favors high flexural strength which is interesting for outdoor use.

118 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

4.7 G

G

G

G

Conclusions

In this chapter, various treatments of fibers were explored, such as chemical and physical processes of treatment with combined steps, e.g., soaking 1 alkaline treatment or pyrolysis 1 coating. Green processes like hormification were also discussed. All the treatments induce morphological and chemical composition modifications but also affect structural linkages. These changes affect the physical-mechanical properties of the fibers but also their behavior with a cementitious matrix. Thus the treatment has to be adapted to the final use of the fibers and the composites. Treatments such as pyrolysis conduct to interesting properties of fibers but in global point of view, durability and price of fibers can be negatively impacted. A balance between the technical and environmental issues of material has to be found to choose the right process. The hornification process (drying/rewetting cycles) does not affect strongly the morphology of fibers.

References 1. Luttge, U.; Kluge, M.; Bauer, G. Botanique - Traite´ fondamental, Edition Lavoisier Tec & Doc, 1992. 2. Shafizadeh, F. Introduction to Pyrolysis of Biomass. J. Anal. Appl. Pyrol. 1982, v 3 (4), 283305. 3. Marquez Montesino, F.; Arse`ne, M.-A.;. Orea Igraza, U.; Carballo Abreu, L.; Ouensanga, A. Influence of Tree Trunk Part on Charcoal Elaboration: Case of Eucalyptus Wood. In: Proceedings of Carbon conference, July, 2003, Oviedo, Spain. 4. Wang, N.; Low, M. J. D. Spectroscopic Studies of Carbon. XV The Pyrolysis of Lignin. Mater. Chem. Physics 1990, 26, 6780. 5. Angenault, J. La chimie  Dictionnaire encyclope´dique, Edition Dunod, 1991. 6. Bledzki, K.; Gassan, J. Composites Reinforced With Cellulose Based Fibres. Prog. Polymer Sci. 1999, 24 (2), 221274. Available from: http://dx.doi.org/10.1016/s00796700(98)00018-5. 7. Claramunt, J.; Ardanuy, M.; Garcia-Hortal, J. A. Effect of Drying and Rewetting Cycles on the Structure and Physicochemical Characteristic of Softwood Fibers for Reinforcement of Cementitious Composites. Carbohydrate Polymers 2010, 79, 200205. Available from: http://dx.doi.org/10.1016/j.carbpol.2009.07.057. 8. Motaung, T. E.; Danandjiwala, R. D. Effect of Alkali and Acid Treatment on Thermal Degradation Kinetics of Sugar Cane Bagasse. Ind. Crops Products 2015, 74, 472477. Available from: http://dx.doi.org/10.1016/j.indcrop.2015.05.062. 9. Rocha Ferreira, S.; De Andrade Silva, F.; Lopes Lima, P. R.; Dias Toledo Filho, R. Effect of Hornification on the Structure, Tensile Behavior and Fiber Matrix Bond of Sisal, Jute and Caraua Fiber Cement Based Composite Systems. Constr. Building Mater. 2017, 139, 551561. Available from: http://dx.doi.org/10.2016/j.conbuildmat.2016.10.004. 10. Savastano Junior, H.; Agopyan, V. Transition Zone Studies of Vegetable Fiber-Cement Paste Composites. Cement Concrete Composites 1999, 31 (4), 232243. Available from: https://doi.org/10.1016/s0958-9465(98)00038-9.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

119

11. Arse`ne, M.-A.; Savastano Junior, H.; Allameh, S.M.; Ghavami, K.; Soboyejo, W. Cementitious Composites Reinforced With Vegetable Fibers. In: Proceedings of the First Interamerican Conference on Non-Conventional Materials and Technologies in the Eco-Construction and Infrastructure IAC-NOCMAT, 2003, Joao-Pessoa Brazil. Proceedings, ISBN: 85-98073-02-4; 2003. 12. Cerny, R.; Rovnanikova, P. Transport Processes in Concrete; Spon Press: London, 2002. 13. Mohanty, A. K.; Misra, M.; Drzal, L. T. Surface Modifications of Natural Fibers and Performance of the Resulting Biocomposites: An Overview. Composite Interface 2001, 8 (5), 313343. Available from: http://dx;doi.org/10.1163/156855401753255422. 14. Almeida, A.; Toledo Filho, R.; Melo Filho, J. Cement Composites Reinforced by Short Curaua Fibers. Mate´ria (Rio de Janeiro) 2010, 15 (2). Available from: http://dx.doi.org/ 10.1590/S1517-70762010000200010. 15. Toledo Filho, R.; Fairbairn, E. M. R.; Deandrade Silva, F.; Melo Filho, J. A. Durability of Compression Molded Sisal Fiber Reinforced Mortar Laminates. Constr. Building Mater. 2009, 23 (6), 24092420. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2008.10.012. 16. Saran, S. Reinforced Soil and Its Engineering Applications; I.K. International publishing house Pvt Limited: New Delhi, 2010. 17. Johnston, C.D. Fibers Reinforced Cements and Concretes. Advances in Concrete Technology Volume 3 Cordona and Breach Science Publishes, 2001. 18. Arse`ne, M.-A.; Bilba, K.; Okwo, A. S.; Soboyejo, A. B. O.; Soboyejo, W. O. Chemically and Thermally Treated Vegetable Fibers for Reinforcement of CementBased Composites. Mater. Manuf. Processes 2007, 22 (2), 214227. Available from: http://dx.doi.org/10.1080/10426910601062065. 19. Li, X.; Tabil, L. G.; Panigraphi, S. Chemical Treatments of Natural Fiber for Use in Natural Fiber-Reinforced Composites: A Review. J. Polymers Environ. 2007, 15, 2533. Available from: http://dx.doi.org/10.1007/s10924-006-0042-3. 20. Rong, M. Z.; Zhang, M. Q.; Liu, Y.; Yang, G. C.; Zeng, H. M. The Effect of Fiber Treatment on the Mechanical Properties of Unidirectional Sisal-Reinforced Epoxy Composites. Compos. Sci. Technol. 2001, 61 (10), 14371447. Available from: http:// dx.doi.org/10.1016/S0266-3538(01)00046-X. 21. Mwaikambo, L. Y.; Ansell, M. P. The Effect of Chemical Treatment on the Properties of Hemp, Sisal, Jute and Kapok Fibers for Composites Reinforcement. Die angewandte makromolekulare chemie 1999, 272 (1), 108116. doi:10.1002/(SICI)1522-9505 (19991201)272:1. 22. John, M. J.; Anandjiwala, R. D. Recent Developments in Chemical Modification and Characterization of Natural Fiber-Reinforced Composites. Polymer Composites 2008, 29 (2), 187207. Available from: http://dx.doi.org/10.1002/pc.20461. 23. Bilba, K.; Arse`ne, M.-A. Silane Treatment of Fiber for Reinforcement of Cementitious Composites. Composites Part A 2008, 39 (9), 14881495. Available from: http://dx.doi. org/10.1016/j.compositesa.2008.05.013. 24. Arse`ne, M.-A.; Bilba, K.; Savastano Junior, H.; Ghavami, K. Treatment of Non-Wood Plant Fibers Used as Reinforcement in Composite Materials. Mater. Res. 2013, 16 (4), 903923 ISSN 1516-1439. 25. Mukhopadhyay, S.; Fangueiro, R. Physical Modification of Natural Fibers and Thermoplastic Films for Composites: A Review. J. Thermoplastic Composite Mater. 2009, 22 (2), 135162. Available from: http://dx.doi.org/10.1177/0892705708091860.

120 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

26. Shindha, E.; Panigrahi, S. Effect of Plasma Treatment on Structure, Wettability of Jute Fiber and Flexural Strength of Its Composite. J. Composite Mater. 2009, 43 (17), 17911802. doi/abs/10.1177/0021998309338078. 27. Bilba, K.; Arse`ne, M.-A.; Ouensanga, A. Study of Banana and Coconut Fibers; Botanical Composition, Thermal Degradation and Textural Observations. Bioresource Technol. 2007, 98 (1), 5868. Available from: http://dx.doi.org/10.1016/j. biortech.2005.11.030. 28. Bilba, K.; Ouensanga, A. Fourier Transform Infrared Spectroscopic Study of Thermal Degradation of Sugarcane Bagasse. J. Anal. Appl. Pyrol. 1996, 38, 6173. Available from: https://doi.org/10-1016/2370(96)00952-7. 29. Bilba, K.; Arse`ne, M.-A. Etude de mate´riaux composites fibres de dictame/ciment. Revue des composites et des mate´riaux avance´s 2007, 17 (3), 327350. 30. Arrakhiz, F. Z.; Malha, M.; Bouhfid, R.; Benmoussa, K.; Qaiss, A. Tensile, Flexural and Torsional Properties of Chemically Treated Alfa, Coir and Bagasse Reinforced Polypropylene. Composites: Part B 2013, 47, 3541. Available from: http://dx.doi.org/ 10.1016/j.compositesb.2012.10.046. 31. Xiao, B.; Sun, X. F.; Sun, R. Chemical, Structural, and Thermal Characterizations of Alkali-Soluble Lignins and Hemicelluloses, and Cellulose From Maize Stem, Rye Straw, and Rice Straw. Polymer Degradat. Stability 2001, 74, 307319. Available from: http://dx.doi.org/10.1016/S0141-3910(01)00163-X. 32. Sun, J.; Sun, X. F.; Zhao, H.; Sun, R. C. Isolation and Characterization of Cellulose From Sugarcane Bagasse. Polymer Degradat. Stability 2004, 84, 331339. Available from: http://dx.doi.org/10.1016/j.polymdegradstab.2004.02.008. 33. Khristova, P.; Kodsachia, O.; Patt, R.; Khider, T.; Karrar, I. Alkaline Pulping With Additives of Kenaf From Sudan. Ind. Crops Products 2002, 15, 229235. Available from: http://dx.doi.org/10.1016/S0926-6690(01)00118-2. 34. Aziz, S. H.; Anselle, M. P.; Clarke, S. J.; Panteny, S. R. Modified Polyester Resins for Natural Fiber Composites. Composites Sci. Technol. 2005, 65 (3-4), 525535. Available from: http://dx.doi.org/10.1016/j.compscitech.2004.08.005. 35. Abou-Yousef, H.; El-Sakhawy, M.; Kamel, S. Multistage Pulping by Using Alkali/ Cairo’s Acid Treatment. Ind. Crops Products 2004, 21 (3), 337341. Available from: http://dx.doi.org/10.1016/j.indcrop.2004.05.001. 36. Das, P.; Ganesh, A.; Wangikar, P. Influence of Pretreatment for Deashing of Sugarcane Bagasse Pyrolysis Products. Biomass Bioenergy 2004, 27 (5), 445457. Available from: http://dx.doi.org/10.1016/j.biombioe.2004.04.002. 37. Pouteau, C.; Cathala, B.; Dole, P.; Kurek, B.; Monties, B. Structural Modification of Kraft Lignin After Acid Treatment: Characterization of the Apolar Extracts and Influence on the Antioxidant Properties in Polypropylene. Ind. Crops Products 2005, 21, 101108. Available from: http://dx.doi.org/10.1016/j.indcrop.2004.01.003. 38. Li, Y.; Mai, Y.-W.; Ye, L. Sisal Fiber and Its Composites: A Review of Recent Developments. Compos. Sci. Technol. 2000, 60, 20372055. Available from: http://dx. doi.org/10.1016/S0266-3538(00)00101-9. 39. Blankenhorn, P. R.; Blankenhorn, B. D.; Silsbee, M. R.; Dicola, M. Effects of Fiber Surface Treatments on Mechanical Properties of Wood Fiber-Cement Composites. Cement Concrete Res. 2001, 31, 10491055. Available from: http://dx.doi.org/10.1016/ S0008-8846(01)00528-2. 40. Pothan, L. A.; Thomas, S. Polarity Parameters and Dynamic Mechanical Behavior of Chemically Modified Banana Fiber Reinforced Polyester Composites. Compos. Sci.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

41.

42.

43. 44.

45.

46.

47.

48. 49. 50. 51.

52.

53.

54.

55.

56.

121

Technol. 2003, 63, 12311240. Available from: http://dx.doi.org/10.1016/S0266-3538 (03)00092-7. Pehanich J.L., Blankenhorn P.R., Silsbee M.R. Wood Fiber Surface Treatment Level Effects on Selected Mechanical Properties of Wood Fibers Composites. Cement and concrete research, v. 34, Issue 1, p. 59-65, 2004. ISSN: 0008-8846. Coutinho, F. B.; Costa, T. H. S.; Carvalho, D. L.; Gorelova, M. M.; De Santa Maria, J. C. Thermal Behaviour of Modified Wood Fibers. Polymer Testing 1998, 17, 299310. Available from: http://dx.doi.org/10.1016/S0142-9418(97)00059-7. Coutts, R. S. P.; Ridikas, V. Refined Wood Fiber-Cement Products. Appita 1982, 35 (5), 395400. Agopyan, V.; Savastano Junior, H.; John, V.M.; Cincotto, M.A. Developments on Vegetable Fiber-Cement Based Materials in Sao Paulo, Brazil: An Overview. Cement Concrete Composites, 27, 527536, 2005. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2004.09.004 Mathur, V. K. Composite Materials From Local Resources. Constr. Building Mater. 2006, 20 (7), 470477. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2005.01.031. Tan, T.; Rahbar, N.; Allameh, S. M.; Kwofie, S.; Dissmore, D.; Ghavami, K. Mechanical Properties of Functionally Graded Hierarchical Bamboo Structures. Acta biomaterialia 2011, 7 (10), 37963803. Available from: http://dx.doi.org/10.1016/j. actbio.2011.06.008. Barra, B.; Bergo, P.; Alves, C.; Savastano Junior, H.; Ghavami, K. Effects of Methane Cold Plasma in Sisal Fibers. Key Eng. Mater. 2012, 517, 458468. Available from: http://dx.doi.org/10.4028/www.scientific.net/KEM.517.458. Britt, K. W. Chapters 1-3 2nd ed.; Handbook of Pulp Technology, 5. Van Nostrand Reinhold: New York, 1970. Clark, J. A. Chapters 6-8 12-14 Pulp Technology and Treatment for Paper; Miller Freeman: San Francisco, 1987. Mackenzie, A.W. A Guide to Pulp Evaluation. CSIRO Division of Forest Products Australia, 1994. Tonoli, G. H. D.; Joaquim, A. P.; Arse`ne, M. A.; Bilba, K.; Savastano Junior, H. Performance and Durability of Cement Based Composites Reinforced With Refined Sisal Pulp. Mater. Manufact. Processes 2007, 22, 149156. Available from: http://dx. doi.org/10.1080/10426910601062065. Marcilla, A.; Conesa, J. A.; Asensio, M.; Garcia-Garcia, S. M. Thermal Treatment and Foaming of Chars Obtained From Almond Shells: Kinetic Study. Fuel 2000, 79, 829836. Available from: http://dx.doi.org/10.1016/S0016-2361(99)00206-9. Paris, O.; Zollfrank, C.; Zickler, G. A. Decomposition and Carbonisation of Wood Biopolymers-A Microstructural Study of Softwood Pyrolysis. Carbon 2005, 43, 5366. Available from: http://dx.doi.org/10.1016/j.carbon.2004.08.034. Varhegi, G.; Antal Junior, M. J.; Jakab, E.; Szabo, P. Kinetic Modeling of Biomass Pyrolysis. J. Anal. Appl. Pyrolysis 1997, 42, 7387. Available from: https://doi.org/ 10.1016/S0165-2370(96)00971-0. Capart, R.; Khezami, L.; Burnham, A. K. Assessment of Various Kinetic Models for the Pyrolysis of Microgranular Cellulose. Thermochimica acta 2004, 417, 7989. Available from: http://dx.doi.org/10.1016/j.tca.2004.01.029. Tsujiyama, T.; Miyamori, A. Assignment of DSC Thermograms of Wood and Its Components. Thermochimica acta 2000, 351, 177181. Available from: http://dx.doi. org/10.1016/S0040-6031(00)000429-9.

122 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

57. Caballero, J. A.; Marcilla, A.; Conesa, J. A. Thermogravimetric Analysis of Olive Stones With Sulphuric Acid Treatment. J. Anal. Appl. Pyrolysis 1997, 44, 7588. Available from: http://dx.doi.org/10.1016/S0165-2370(97)00068-5. 58. Garrote, G.; Eugenio, M. E.; Diaz, M. J.; Ariza, J.; Lopez, F. Hydrothermal and Pulp Processing of Eucalyptus. Bioresource Technol 2003, 88, 6168. Available from: http:// dx.doi.org/10.1016/S0960-8524(02)00256-0. 59. Nuissier, M. Mise en oeuvre de fibres ve´ge´tales longues: de l’obtention a` la caracte´risation. Thesis (M.Sc.) Universite´ des Antilles, 2016. (In French) 60. Sun, R.; Lawther, J. M.; Banks, W. B. Influence of Alkaline Pre-Treatements on the Cell Wall Components of Wheat Straw. Ind. Crops Products 1995, 4 (2), 127145. Available from: https://doi.org/10.1016/0926-6690(95)00025-8. 61. Mosier, N.; Wyman, C.; Dale, B.; Elander, R. Features of Promising Technologies for Pretreatment of Lignocellulosic Biomass. Bioresource Technol. 2005, 96 (6), 673686. Available from: http://dx.doi.org/10.1016/j.biortech.2004.06.025. 62. Kalia, S.; Thahur, K.; Kiechel, M.-A.; Schauer, C.-L. Surface Modification of Plant Fibers Using Environment Friendly Methods for Their Application in Polymer Composites, Textile Industry and Antimicrobial Activities: A Review. J. Environ. Chem. Eng. 2013, 1, 97112. Available from: http://dx.doi.org/10.1016/j.jece.2013.04.009. 63. Nasso, I. Utilisation de fibres ve´ge´tales comme renfort de mate´riaux cimentaires. Report of postdoctoral activities, Universite´ des Antilles et de la Guyane, 2010. (In French). 64. Bilba, K. Elaboration et caracte´risation de mate´riaux composites. HDR thesis, Universite´ des Antilles et de la Guyane, 2006. (In French). 65. Valadez-Gonzalez, A.; Cervantes-Uc, J. M.; Olayo, R.; Herrera-Franco, P. J. Chemical Modification of Henequen Fibres With an Organosilane Coupling Agent. Composites Part B: Engineering 1999, 30, 321331. Available from: http://dx.doi.org/10.1016/ S1359-8368(98)00055-9. 66. Shimoda, K.; Park, J. S.; Hinoli, T.; Kohyama, A. Influence of SiC Nano-Sized Analyzed by X-ray Photoelectron Spectroscopy on Basic Powder Characteristics. Appl. Surface Sci. 2007, 253, 94509456. Available from: http://dx.doi.org/10.1016/j.apsusc.2007.06.023. 67. Tserki, V.; Zafeiropoulos, N. E.; Simon, F.; Panayiotou, C. A Study of the Effect of Acetylation and Propionylation Surface Treatments on Natural Fibres. Compos. Part A: Appl. Sci. Manufact. 2005, 36, 11101118. Available from: http://dx.doi.org/10.1016/j. compositesa.2005.01.004. 68. Park, B. D.; Wi, S. G.; Lee, K. H.; Singh, A. P.; Yoon, T. H.; Kim, Y. S. X-ray Photoelectron Spectroscopy of Rice Husk Surface Modified With Maleated Polypropylene and Silane. Biomass Bioenergy 2004, 27, 353363. Available from: http://dx.doi.org/10.1016/j.biombioe.2004.03.006. 69. Balat, M.; Berjoan, R.; Pichelin, G.; Rochman, D. High-Temperature Oxidation of Sintered Silicon Carbide Under Pure CO2 at Low Pressure: Active-Passive Transition. Appl. Surface Sci. 1998, 133, 115123. Available from: http://dx.doi.org/10.1016/ S0169-4332(98)00193-7. 70. Streemany, M.; Ghosh, T. B.; Pai, B. C.; Chakraborty, M. XPS Studies on the Oxidation Behavior of SiC Particles. Mater. Res. Bullet. 1998, 33 (2), 189198. Available from: http://dx.doi.org/10.1016/S0025-5408(97)00222-5. 71. Bilba, K.; Arse`ne, M.-A.; Ouensanga, A. Influence of Chemical Treatment of Vegetable Fibres on Insulating Behavior of Vegetable Fibres/Cement Composites. In: Proceedings of Brazilian Conference on Non-Conventional Materials and Technologies: Affordable Housing and Infrastructure, Pirassununga, 2004.

Treatments for viable utilization of vegetable fibers in inorganic-based composites

123

72. Rodier, L. Mate´riaux de construction en zone tropicale humide  potentialities de sousproduits ou de mate´riaux naturels locaux en substitution ou en addition a` la matrice cimentaire. PhD thesis, Universite´ des Antilles et de la Guyane, 2014. (In French). 73. Bilba, K.; Arse`ne, M.-A.; Ouensanga, A. Sugarcane Bagasse Fibre Reinforced Cement Composites. Part I. Influence of Botanical Components of Bagasse on the Setting of Bagasse/Cement Composite. Cement Concrete Composites 2003, 25, 9196. Available from: http://dx.doi.org.10.1016/S0958-9465(02)00003-3. 74. Ouensanga, A.; Picard, C. Thermal Degradation of Sugar Cane Bagasse. Thermochimica acta 1988, 125, 8997. Available from: https://doi.org/10.1016/0040-6031(88)87213-7. 75. Justiz-Smith, N. G.; Virgo, G. J.; Buchanan, V. E. Potential of Jamaican Banana, Coconut Coir and Bagasse Fibres as Composite Materials. Mater. Character. 2008, 59, 12731278. Available from: http://dx.doi.org/10.1016/j.matchar.2007.10.011. 76. Tomani, P.; Axegard, P.; Berglin, N.; Nordgren, D.; Berghel, T. LignoBoost kraft lignin  a new renewable fuel and a valuable fuel additive. International Bioenergy & Bioproducts Conference, Atlanta GA USA, March 1416, 2011. 77. TINGUE Eliphe`te. Etude de l’influence de traitements chimiques de fibres ve´ge´tales sur les proprie´te´s de l’interface fibres/matrice de mate´riaux Composites Fibres Ve´ge´tales/Ciment (CFVC). Thesis (M.Sc.) Universite´ des Antilles et de la Guyane, 2009. (In French). 78. Norme Francaise-European Norm. NF EN18-459. Be´ton-Essais pour be´ton durci-Essai de porosite´ et de masse volumique, Normalisation franc¸aise des ciments. 2010. (In French). 79. Arse`ne, M.-A.; Bilba, K.; Onesippe, C.; Rodier, L. Thermal and Flexural Properties of Bagasse/Cement Composites. Green Mater 2016, 1, 112. Available from: http://dx.doi. org/10.1680/jgrmat.15.00012. 80. Onesippe, C.; Passe-Coutrin, N.; Toro, F.; Delvasto, S.; Bilba, K.; Arse`ne, M.-A. Sugar Cane Bagasse Fibers Reinforced Cement Composites: Thermal Considerations. Composites: Part A 2010, 41, 549556. Available from: http://dx.doi.org/10.1016/j. compositesa.2010.01.002. 81. Khedari, J.; Suttisonk, B.; Pratinthong, N.; Hirunlabh, J. New Lightweight Composite Construction Material With Low Thermal Conductivity. Cement Concrete Composites 2001, 23, 6570. Available from: http://dx.doi.org/10.1016/S0958-9465(00)00072-X. 82. Verardi, A.; Blasi, A.; De Bari, I.; Calabro, V. Steam Pretreatment of Saccharum Officinarum L. Bagasse by Adding of Impregnating Agents for Advanced Bioethanol Production. Ecotoxicol. Environ. Safety 2016, 134, 293300. Available from: http://dx. doi.org/10.1016/j.ecoenv.2015.07.034. 83. Ramanaiah, K.; Ratna Prasad, A. V.; Hema Chandra Reddy, K. Thermal and Mechanical Properties of Waste Grass Broom Fiber-Reinforced Polyester Composites. Mater. Design 2012, 40, 103108. Available from: http://dx.doi.org/10.1016/j. matdes.2012.03.034. 84. Xu, Y.; Chung, D. D. L. Effect of Sand Addition on the Specific Heat and Thermal Conductivity of Cement. Cement Concrete Res. 2000, 30, 11751178 PII: S0008-8846 (99)00206-9.

Further reading 1.

Martin, H. J.; Schulz, K. H.; Bumgardner, J. D.; Walters, K. B. An X.P.S. study on the Attachement of Triethoxysilybutyraldehyde to Two Titanium Surfaces as a Way to Bond Chitosan. Appl. Surface Sci. 2008, 254 (15), 45994605. Available from: http:// dx.doi.org/10.1016/j.apsusc.2008.01.066.

New inorganic binders containing ashes from agricultural wastes

5

Jordi Paya´1, Jose´ Monzo´1, Maria Victoria Borrachero1, Lourdes Soriano1, Jorge L. Akasaki2 and Mauro M. Tashima2 1 Universitat Polite`cnica de Vale`ncia, Valencia, Spain, 2UNESP - Universidade Estadual Paulista, Sa˜o Paulo, Brazil

5.1

Introduction: sustainability, new binders, use of wastes, and circular economy

In recent decades, there has been a growing interest in sustainability. In this context, developments in the design of methods for reducing negative human impact are being highlighted. In particular, concerning the building and construction sectors, many efforts are being carried out to develop more sustainable materials and procedures. In these sectors, many manufactured products are produced using huge amounts of raw materials and energy. As can be seen in Fig. 5.1, in the area of materials and procedures, concrete is the most important component of building activities in terms of sustainability. There are several reasons for this: (1) the volume of concrete produced annually is more than 14,000 million cubic metres; (2) the component of concrete with the highest environmental impact is Portland cement (OPC); (3) usually the binder used is OPC, whose manufacture requires high energy consumption; and (4) the fabrication of OPC (more than 3600 million metric tons per year) contributes 5%7% of annual anthropogenic global CO2 emissions.1,2 Specifically, OPC is the key to the sustainability of concrete. The production of 1 ton of OPC clinker releases 0.81 ton of CO2, taking into account the chemical release (limestone decomposition) and energy release (fossil fuel combustion). Many developments have been achieved regarding the reduction of clinker content in commercial Portland cements by blending with supplementary cementing materials (SCMs) (BFS, fly ash (FA), and silica fume (SF), among others).3 In 1995, the percentage of blended Portland cements was 44%, whereas this percentage increased to 55% in 2000, and 80% in 2009.4 The increase of this percentage in the use of blended cements will probably be limited by technological issues, being that in some cases the percentage of blending affects significantly the mechanical and durability properties. Many reports related to the using of wastes in the production of Portland cement and concrete have been published.5 An immense range of wastes is amenable for use in concrete; however, their exact

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00006-1 © 2017 Elsevier Ltd. All rights reserved.

128 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 5.1 Materials in building and construction sectors: the importance of different materials in terms of sustainability.

composition can vary depending on the specific source: hazardous elements or compounds, and special pretreatments, may limit their use. In the last decade, a radical change in the future of binders for construction has been proposed by the scientific community: the admission that other cements are possible is a logical consequence of the simple approach to scientific activity.6 Thus, calcium sulfoaluminate (CSA) cements have been proposed as an alternative to Portland cements because the base phase is ye’elimite, whose clinkerization requires lower temperatures (12001300 C) and less limestone as raw material than Portland cement clinker. A reduction of 50% of the CO2 emissions compared with Portland clinker can be achieved. However, for this new cement it is not easy to use wastes for blending. More possibilities for using wastes are found for alkaline cements.79 In this case, a waste or a mixture of wastes (precursor) is activated by means of a high-alkaline solution (usually, alkali hydroxide/silicate solution). Depending on the nature of the precursor and the concentration of the activating solution, binding materials with different properties are achieved. An interesting possibility for the future is to combine OPC and alkali-activated cement, termed “hybrid cements.” In this case, large amounts of SCMs can be used in combination with very low amounts of OPC (10%30% of the mixture). The circular economy plays an important role in the concrete industry and the applications of the material. It is well known that in developed countries, more than 30% of the total volume of residues is related to construction and demolition wastes. An important part of these wastes is composed of concrete and its constitutive elements (hydrated products from binder and different-sized aggregates). A part of the aggregates from demolished concrete can be recycled to prepare new concrete; however, other important parts present difficulties for reuse, or the alternative reuse is low-grade. This is the main reason for the huge consumption of

New inorganic binders containing ashes from agricultural wastes

129

Figure 5.2 The circular economy: contributions from different activities to concrete production.

raw materials by the concrete industry, and especially the Portland cement-based concrete industry. The key to enhancing the circularity of the economy is to open the door to the using of alternative SCMs generated in other activities (industry, energy, agroindustry, urban/domestic, etc.), as can be seen in Fig. 5.2. Additionally, the design of new binders with low or nil OPC content will favor the circular economy in the concrete sector. Among the wastes from different activities, those from agroindustry are increasingly relevant for several reasons. Firstly, many agrowastes are composed of biomass, which presents the potential for valuation in the energy sector; the ashes obtained after combustion may contain relevant chemical compounds and mineral phases, which make appropriate their valuation in cementing materials for construction and the building industry. Secondly, the crop areas and the crop types are varied, and agrowastes can be found primarily in developing countries: thus, some agrowastes and their derived products can be used in the same place where they were generated, reducing transportation costs and facilitating their availability for companies and for rural and disadvantaged communities. Thirdly, many ashes derived from agrowaste (sometimes referred to as “agro-ashes”) contain important chemical elements from the point of view of cementing materials: silicon, aluminum, and sodium/potassium, chemically combined as reactive compounds (vitreous phases and soluble phases). The addition of these ashes to the cementing matrix could modify the microstructure and the alkalinity of the binding

130 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

system, and thus the corresponding cementing mixtures would be of interest for the development of natural fiber composites (e.g., fibrocement).10,11

5.1.1 Alkali activation (AA): precursors and alkaline activating reagents  basic principles This type of cement is known by various names: alkaline cements, alkali-activated materials (AAMs), and geopolymers, among others. Their production is carried out by mixing a solid precursor (aluminasilicate-based mineral material) and a highly concentrated alkaline solution. Usually, the precursor is prepared from by-products: BFS, other slags (from copper, nickel, or electric arc furnace slags), FAs from thermoelectric power plants, ceramic waste, or construction and demolition wastes.9 The liquid activator is prepared from chemical reagents: sodium or potassium hydroxide and waterglass (sodium silicate solution or other appropriate solution, potassium silicate); in some cases, alkali carbonates or calcium salts (calcium hydroxide) have also been tested for preparing activating solutions. A distinction in this type of cement must be made in order to differentiate the activation conditions and the cementing products formed. Usually, when a calcium-rich mineral admixture is selected as precursor (e.g., BFS or C-class FA), the term AAM is employed. On the other hand, for low-calcium precursors (very rich in silica and alumina compounds), the term “geopolymer” (GP) is preferred (considered by many authors a subset of AAM). In Fig. 5.3, the different compositions of these cements are compared with Portland cement (OPC), calcium aluminate cement (CAC), and CSA-based cements. It may be noticed that the calcium content of AAM (see alkali-activated slag (AAS)/WG and AAS/N in Fig. 5.3) is lower than

Figure 5.3 Comparative percentage ratios of compositions of Portland cement (OPC), calcium sulfoaluminate cement plus gypsum (CSA+G), calcium aluminate cement (CAC), alkali-activated materials (AAM: blast furnace slag based materials, with waterglass AAS/ WG, with sodium hydroxide AAS/N) and geopolymers (GP: fly ash based materials AA/FA and metakaolin based materials AA/MK). (Axis key for mass percentage ratios: C=CaO; ˆ S=SiO2; A=Al2O3; N=Na2O; S=SO 3).

New inorganic binders containing ashes from agricultural wastes

131

that of OPC, CAC, and CSA, and GP presents the lowest calcium content (see AA/FA and AA/MK in Fig. 5.3). Additionally, GP cements require higher alkali concentrations M+ (added to the activating solution), and usually contain more Al than AAM. Alkali-activated blast furnace slag cement (AAS) is the most widely studied in the AAM group. Usually, the precursor is activated by means of a 24 molar solution of sodium hydroxide, although higher concentrations (reaching 8 molar in some cases) and the presence of soluble silicate (from waterglass) have also been applied. The mixtures develop strength when cured at room temperature. In these AAM systems, a calcium aluminosilicate hydrate (C-A-S-H gel) is produced as the main binding phase, being very similar in structure to the C-S-H (chain structure) gel obtained from hydration of Portland cement.8 Activation of low-calcium content precursors (e.g., low-calcium fly and metakaolin (MK)) is carried out by means of 812 molar concentrations of alkali and the presence of soluble silicate (usually the SiO2/Na2O molar ratio in the activating solution ranges from 0.5 to 2).7 The cementing process usually requires curing at high temperatures (6090 C range). In these GP systems, sodium aluminosilicate hydrate gel (N-A-S-H) is produced, which presents a three-dimensional disordered structure (pseudo-zeolitic network).7

5.1.2 Agricultural wastes in inorganic binders Traditionally, those SCMs to be blended with a Portland cement clinker have an industrial source: BFS, FAs, or SF. Consequently, most of the commercially blended cements include one or more of these SCM derived from industrial by-products. Additional SCM are also used: filler limestone, thermally activated clays or schists. Some of them present pozzolanic activity, that is, the ability to react with hydrated lime to form insoluble hydrates with cementing properties. This reactivity is based on the fact that these SCM (also called pozzolans or pozzolanic materials) contain reactive silica (SiO2, S); in some cases, reactive alumina (Al2O3, A) also takes part in this reaction. The corresponding reaction products are calcium silicate hydrates (CSHs) (C-S-H) or calcium aluminosilicate hydrates (C-A-S-H) in gel or semicrystalline form: CaðOHÞ2 þSiO2 þH2 O ! C-S-H

(5.1)

CaðOHÞ2 þSiO2 þAl2 O3 þH2 O ! C-A-S-H

(5.2)

Fortunately, wastes from agriculture can be transformed into appropriate SCM for blending Portland cement.12 Plants absorb silicon from soil as silicic acid and this accumulates in the cellular tissues as hydrated silica or opal (SiO2  nH2O). This silica compound is located in specific cells or parts of the cells (e.g., phytolites).13 The silica content is between 0.1% and 10% dry weight of higher plants. In general, monocots accumulate more silica than dicots, although there may be differences even at the genus level.

132 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 5.4 Ternary diagram for classification of ashes from agrowastes according to their chemical compositions. Shaded zone includes appropriate biomass ashes for Portland cement blends. Adapted from Vassilev, S.V.; Baxter, D.; Andersen, L.K.; Vassileva, C.G. An Overview of the Chemical Composition of Biomass. Fuel, 2010, 89, 913933. doi: 10.1016/j.fuel.2009.10.022.

The removal of organic matter from the agricultural waste yields an ash composed of several inorganic compounds.14 Vassilev et al. proposed a classification of biomass ashes according to their chemical composition (Fig. 5.4). The most interesting ashes from the point of view of use as SCM are those that correspond to the “S” zone (high acid zone), because of the high percentage of silica. Rice husk ash (RHA) belongs to this type of ash. Interest is also growing in medium-acid ashes located in the S-MA and K-MA zones. In recent years, many ashes derived from biomass have been studied for use in preparing Portland-based blended cements12: sugarcane bagasse ash (SCBA),15 palm oil fuel ash (POFA),16 corn cob ash,17 sugarcane straw ash (SCSA),18 bamboo leaf ash,19 banana leaf ash,20 olive biomass ash,21 and cocoa almond bark ash,22 among others. In all these cases, the following three fundamental aspects of possible developments must be addressed23: chemical aspects, especially related to the silica content, chlorides, alkali, and loss on ignition; physical aspects, fundamentally linked to the granulometry and accessible porosity of the particles; and, finally, mineralogical aspects, focused on the crystallinity or amorphicity of the silica phases and other compounds.

New inorganic binders containing ashes from agricultural wastes

5.2

133

Alkali-activated binders containing ashes from agricultural wastes: case studies

5.2.1 Rice husk ash As is well known, RHA is a siliceous waste material that can present excellent properties when used as a pozzolanic material in blended Portland cements. In this section, the use of RHA in the production of alkali-activated binders will be summarized. Due to the high amount of silicon dioxide in its composition, RHA can be used to modify the SiO2/Al2O3 ratio of precursors or to generate sodium silicate solutions. Bouzo´n et al.24 reported the development of an alkali-activated binder based on a fluid catalytic cracking (FCC) catalyst activated by an alkaline solution prepared by refluxing aqueous mixtures of ground or unprocessed RHA and NaOH. According to the authors, the effectiveness of the process depends on the reflux time. Activation of FCC by a mixture of RHA/NaOH produced mortars with compressive strength in the range of 3141 MPa. This value was similar to the compressive strength of a control mortar prepared using a commercial waterglass. Bernal et al.25 performed a study to produce AAS binders using silicates based on alkaline activators derived from chemical reactions between NaOH and SF or RHA, demonstrating the feasibility of this process. Binders produced using this procedure developed higher compressive strength (.100 MPa after 28 days of curing) than binders produced using commercial sodium silicate solution. When exposed to temperatures of about 600 C, AAS presented a significant reduction in compressive strength, while only the RHA-based system retained measurable strength at 800 C. The use of RHA as a precursor is more common than its use for preparing the activating solution. The binary mixture FARHA has been studied in various articles. Detphan and Chindaprasirt26 studied the influence of various factors in the preparation of geopolymers with RHA. They tested the influence of the temperature of burning of the rice husk (580940 C), the ratio Na2SiO3/NaOH, the ratio FA/RHA, the effect of curing (delay time before heat curing, duration of heat curing, heat curing temperature) and the fineness of the RHA. The authors concluded that the optimum burning temperature was 690 C and an increase in the fineness of the RHA improved the reactivity. The mixtures with a ratio of Na2SiO3/ NaOH equal to 4.0 obtained the best compressive strengths, and all the mortars in which FA was replaced by RHA had slightly decreased compressive strengths compared with the mortar containing only FA. Songpiriyakij et al.27 studied the same system, and in this case the authors obtained mortars with high compressive strengths when they replaced FA with rice husk bark ash RHBA (RHBA was obtained from a biomass electric power plant burning rice husks and bark at a ratio of 70:30). The mixture with the ratio SiO2/ Al2O3 equal to 15.91 (40 FA:60 RHBA) had a 73 MPa compressive strength after 90 days of curing. The authors concluded that the compressive strength of the mixtures increased with less Na2O.

134 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Hwang and Huynh28 studied the effect of the alkali activator and RHA content in the type of mixture. They cured the mortars at 35 C and at 50% relative humidity. The optimum concentration of NaOH was 10 M and the poorest was 8 M. The authors proposed an RHA content of 35% to obtain the best compressive strength in FA/RHA mixtures. Riahi et al.29 studied the performance of a mixture of 80 FA:20 RHA using the Taguchi method. The authors studied an L9 Taguchi array (three factors at three levels), where the factors were curing temperature (25, 70 and 90 C), oven curing time (2, 4 and 8 h), and sodium hydroxide concentration (5, 8 and 12 M). They concluded that the optimum conditions were a curing temperature of 90 C and concentration of 5 or 8 M (5 M with 2 days of curing and 8 M with 7 days of curing). The oven curing time had no significant effect on compressive strength. In an analogous way, Riahi and Nazari30 published another paper employing the Taguchi method. In this case the authors made mortars with mixtures 80 FA:20 RHA and tested four factors at three levels: percentage of nano-silica (1, 2, and 3 wt%), oven curing temperature (25, 70, and 90 C), oven curing time (2, 4, and 8 h), and NaOH concentration (5, 8, and 12 M). The authors reached the same conclusions as the previously cited paper, i.e., optimal conditions of 8 M and 90 C and established the addition of 3 wt% SiO2 or Al2O3 nanoparticles as the optimum percentage. Karim et al.31 proposed a ternary system using RHAground granulated blast furnace slag (GGBS)FA and studied the effect of using NaOH, KOH, or Ca(OH)2 as activator reagents. The authors made five types of mixtures, varying the percentage of GGBS, RHA, and FA. The activator solutions were prepared using different percentages of the total weight of the binder (2.5, 5.0, and 7.5%) and with different molar concentrations (1.0, 2.5, 5.0, and 7.5 M). The setting time of the mixture was influenced by the quantity of RHA and the type of activator: the samples with higher RHA had greater setting time. The use of NaOH as activator contributed to shorter setting times. At the same time, the use of NaOH contributed to reduced mortar flow compared with the effect of the other two activators.RHAGGBSFA mixture with 304030 proportion yielded the best mechanical behavior. For this system the quantity and type of activator was a critical factor in the development of compressive strength: the use of 2.5 wt% of KOH or Ca(OH)2 was insufficient to obtain a good mechanical behavior and the mortars achieved a low strength (less than 25 MPa after 90 days of curing); when NaOH or KOH were added at a weight of 5%, the activation was successful (40 MPa at 90 days); when KOH was used, a dose of 7.5% was necessary to obtain the highest strength (more than 50 MPa at 28 days). In terms of molarity, the concentration of 2.5 mol  L21 was sufficient to achieve a good compressive strength for NaOH-activated mortars: with this type of mixture, mortars with compressive strengths higher than 42 MPa after 28 days were achieved. Finally, Nazari studied the performance of FARHA geopolymers using artificial neural networks and fuzzy logic prediction.32,33 These studies incorporated the use of palm oil clinker (POC) as a fine and coarse aggregate. In these papers, the authors concluded that the use of POC particles reduced the compressive strength in relation to reference mortar, but resulted in good resistance to water absorption.

New inorganic binders containing ashes from agricultural wastes

135

Table 5.1 Physical properties of metakaolin-based geopolymers using commercial potassium silicate (G-KS) and rice husk ash (G-RHA)34 Property 3

Bulk density (Kg/cm ) Open porosity (%) Thermal expansion coefficient (1/K) Specific heat (J/kg  K) Thermal diffusivity (mm2/s) Thermal conductivity (W/m  K)

G-KS

G-RHA

1415 33.38 104 252.76 0.22 0.244

1338 35.54 104 515.30 0.16 0.170

Villaquira´n-Caicedo et al.34 prepared novel binary and ternary geopolymers from MK and boiler slag/MK, using alternative silica sources including RHA and KOH to create the alkaline activator in situ. The resulting thermal properties (specific heat, diffusivity, and thermal conductivity) appeared to be associated with the cumulative pore volume, water content, and microstructure of the geopolymer, due to their effect on the heat flow in the material. The geopolymer with RHA showed the lowest thermal conductivity and the lowest thermal diffusivity. The physical properties of the MK-based geopolymer with RHA (G-RHA) compared with the geopolymer (G-KS), produced using traditional commercial potassium silicate (K2SiO3/KOH), are summarized in Table 5.1. Other types of binders have been prepared by blending RHA with different mineral admixtures. Rattanasak et al.35 developed a process to prepare aluminosilicate composites (ASC) with high volumes of RHA. RHA as the source of silica and Al (OH)3 as the source of aluminum were used, and for stabilizing the mixture, boric acid (1 wt%) was introduced. Solutions of 10 M NaOH and Na2SiO3 were used to activate the geopolymerization. Mass ratios of RHA/Al(OH)3 between 70/30 and 99/1 were studied, and the pastes were cured at different temperatures. After heat curing at 115 C, the pastes containing 2.520 wt% Al(OH)3 showed no sign of disintegration in boiling water or swelling. The compressive strength at 90 days was 20 MPa for 10 wt% of Al(OH)3. This mortar also showed good resistance to 3 vol% H2SO4 solution, with only a slight loss of strength after 90 days immersion, but in 5 wt% MgSO4 solution after the same immersion time, a larger strength reduction took place. Other authors36 have used kaolin as a precursor in the formation of geopolymer, in order to reduce heat energy consumption. The reagents Na2SiO3 and 10 M NaOH combined at a ratio Na2SiO3/NaOH equal to 3/2 were used as activator. Under these conditions, the geopolymeric mixture required high temperature (80 C for 24 h) to be solidified. However, the addition of bagasse ash (BA) or RHA allowed solidification by curing at room temperature. The geopolymers produced with kaolin mixed with either BA or RHA exhibited very much higher strengths after 91 days than those from a single ash (100% BA and 100% RHA). The compressive strength

136 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

achieved (21.3 MPa at 91 days) using 80% kaolin with 20% RHA geopolymer was superior to that achieved with BA (16.80 MPa at 91 days). Moreover, He et al.37 synthesized a geopolymer from two industrial wastes, red mud (RM) and RHA. The amorphous silica present in the RHA after dissolution with NaOH was used as activator. They studied a variety of parameters, including curing time, RHA/RM ratio, RHA particle size, and alkalinity. The final products were also characterized by X-ray diffraction technique, scanning electron microscopy (SEM), compressive strength test, and chemical analysis. The mechanical properties of the RHA/RM geopolymers were highly complex. A compressive strength of 20.46 MPa was reached with a mass ratio RHA/RM=0.5. Microstructural and compositional analysis showed that the final products were composed of an amorphous geopolymer binder with other crystalline phases as fillers. The authors concluded that uncertainties in the composition, microstructure, and extent of RHA dissolution, as well as other side reactions, may cause difficulty in the practical applications of this kind of geopolymer. Finally, some researchers have used RHA to design geopolymeric lightweight matrices. Thus, Pimraksa et al.38 synthesized lightweight geopolymeric materials from highly porous siliceous materials, such as diatomaceous earth (DE) and RHA, with high SiO2/Al2O3 ratios of 13.033.5 and Na2O/Al2O3 ratios of 0.663.0. The RHA was used to adjust the composition of the silica phase. Fig. 5.5 shows the

Figure 5.5 Compressive strength development and bulk density values for geopolymers made with diatomaceous earth calcined at 800 C and rice husk ash (RHA). Adapted from Pimraksa, K.; Chindaprasirt, P.; Rungchet, A.; Sagoe-Crentsil, K.; Sato, T. Lightweight Geopolymer Made of Highly Porous Siliceous Materials with Various Na2O/ Al2O3 and SiO2/Al2O3 Ratios. Mater. Sci. Eng. A, 2011, 528, 66166623. doi: 10.1016/ j.msea.2011.04.044.

New inorganic binders containing ashes from agricultural wastes

137

compressive strength and bulk density of geopolymer pastes made by mixtures of DE (calcined at 800 C) and RHA. The strength increases to 24 Kg/cm2 due to the Si-O-Al bonds formed in the geopolymeric structures. The authors detected an optimum in the quantity of RHA (40 wt%). The density values were low (0.921.01 g/cm3), which is highly desirable.

5.2.2 Sugarcane bagasse ash SCBA is an agroindustrial waste material composed mainly of silicon dioxide, which has been extensively used in Portland cement blended mortars and concretes.39 However, studies related to the use of SCBA as precursor in the production of alkali-activated binders are less numerous. Tippayasam et al.36 reported the use of 100% SCBA as precursor for the production of alkali-activated binders, obtaining low compressive strength (5.68 MPa after 91 curing days). However, when binary systems such as kaolin/SCBA or MK/SCBA were used, higher compressive strengths were yielded as compared with 100% SCBA (16.80 MPa after 91 curing days)36. Castaldelli et al.40 reported a study using an SCBA obtained from the sugarcane industry with a high proportion of organic matter (about 25%). Alkali-activated pastes and mortars using alkaline solutions of 5 mol  kg21 of Na+ and a SiO2/Na2O molar ratio of 1.46, different BFS/SCBA proportions, and two curing conditions (room temperature at 20 C and thermal bath at 65 C) were assessed. Mortars cured at room temperature for 90270 days yielded notably higher compressive strength (5565 MPa) and lower total porosity (7.5%10.0%) than mortars cured for 3 days at 65 C (4254 MPa and 9.5%12.5%, respectively). Amin et al.41 assessed the possibility of using SCBA and calcined clay in the production of alkali-activated systems. The authors calcined the clay for 10 hours at different temperatures: 600, 700, 800, 900, and 1000 C. The mixtures were prepared using different SiO2/ Al2O3 ratios, activated with 18 mol  L21 sodium hydroxide solution and cured for 7 days at 60 C. The obtained results showed that clay calcined in the range 600900 C due to the formation of metakaolinite in this calcination temperature range reduced in the setting time and increased in the compressive strength (over 16 MPa) of the alkali-activated pastes. However, at 1000 C, an increase in the setting time and a decrease in the mechanical strength were observed. The authors also concluded that SiO2/Al2O3 ratio of 2.7 and NaOH concentration of 3 M were the optimum conditions for achieving the highest compressive strength (about 20 MPa). Pereira et al.42 performed a study of the mechanical and durability properties of alkali-activated mortar based on SCBA and BFS. In this case, the SCBA was obtained by means of an autocombustion process of bagasse, resulting in an SCBA with low loss on ignition (about 4.4%). Three different solutions, sodium hydroxide (8 mol  L21), sodium silicate (8 mol  L21 of Na+ and SiO2/Na2O molar ratio of 0.5), and potassium hydroxide (8 mol  L21), and 4% of SCBA replacing BFS were assessed. The authors reported that although the SCBA presented a large amount of quartz (soil contamination), the SCBA was a reactive material in a high-alkaline

138 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

medium. Replacement of 25% BFS with SCBA generated mortars that yielded similar compressive strength to the control (BFS only). Additionally, the alkaliactivated mortars exhibited better durability than plain Portland cement (OPC) mortars, especially in ammonium chloride, acetic acid, and sodium sulfate environments. In other work, Castaldelli et al.43 assessed the use of FA/SCBA in the production of alkali-activated binders using potassium silicate as the alkaline activating solution. In Fig. 5.6 the FTIR spectra of the raw materials (FA and SCBA) and alkali-activated pastes (SiO2/K2O, molar ratio of 0.75) with different FA/SCBA proportions, after 3 and 7 days of curing at 65 C, are depicted. The bands related to Si-O were identified at 1070 and 459 cm21 for FA and at 1056, 780, and 490 cm21 for SCBA. For the alkali-activated pastes a new band centrred at 1006 cm21 was attributed to the product formed by the alkaline activating reaction, since this band did not appear in the FA and SCBA spectra. The compressive strengths of mortars are influenced by the SiO2/K2O molar ratio and the FA/SCBA ratio, with values in the range 23.436.4 MPa after 7 days of curing at 65 C and 28.040.3 MPa after 270 days of curing at 20 C.13 The use of SCSA in the production of alkali-activated binders has also been reported in the literature. The calcination of sugarcane leaves into SCSA tends to

Figure 5.6 FTIR spectra of alkali-activated pastes for different proportions of FA/SCBA after 3 and 7 days of curing at 65 C. Source: Castaldelli, V.N.; Moraes, J.C.B.; Akasaki, J.L.; Melges, J.L.P.; Monzo´, J.; Borrachero, M.V., et al. Study of the Binary System Fly Ash/Sugarcane Bagasse Ash (FA/SCBA) in SiO2/ K2O Alkali-activated Binders. Fuel, 2016, 174, 307316. doi: 10.1016/j.fuel.2016.02.020.

New inorganic binders containing ashes from agricultural wastes

139

form a disordered material with irregular shapes and some original cellular structures (stomata and phytoliths).13 Moraes et al.44 assessed the compressive strength of binary systems BFS/SCSA (100/0, 75/25, and 50/50) for mortars activated with 8 mol  kg21 of Na+ and cured at room temperature for 7 and 28 days. Comparing the compressive strength of mortars produced using BFS only, increments of 57% and 148% for mortars with 25% of SCSA were observed after 7 and 28 curing days, respectively.

5.2.3 Palm oil fuel ash The palm oil industry is present in countries like Malaysia, Indonesia, and Thailand. Malaysia is the largest producer of palm oil and palm products in the world. POFA is the ash generated from burning palm oil plant residues (husk and shell). The production of POFA grows every year, and it is usually disposed of in landfills. Malaysia alone produces annually around 3 million tons of POFA, and this production rate is likely to increase in the coming years. There is another less known waste produced during combustion: palm oil boiler ash (POBA) or bottom ash. POBA contains coarser particles than POFA and a slightly different composition in the percentage of oxides. The chemical composition of POFA shows that it is rich in silicon dioxide ($40%), with low percentages of Al2O3, and a relatively high content of CaO (around 11%), most likely from soil and fertilizer.45,46 The morphology of POFA has been observed by SEM: POFA particles are porous with irregular shapes, as can be seen in Fig. 5.7. A few studies have used POFA or POBA alone as unique geopolymeric precursors. Salih et al.46 proposed a method to activate POFA: NaOH solution was mixed with Na2SiO3 to produce six different ratios (0.5, 1.0, 1.5, 2.0, 2.5, and 3.0). The specimens were cured at 60 C for 2 h, and then kept in laboratory conditions. The best compressive strength was obtained by the mortar with the ratio Na2SiO3/ NaOH equal to 2.5: this mortar presented a resistance of 32 MPa at 28 days. The authors also concluded that in POFA geopolymer, C-S-H gel was formed rather

Figure 5.7 SEM images of palm oil ashes (POFA).

140 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

than N-A-S-H gel. Zarina et al.47 investigated the effect of the solid/liquid (S/L) and Na2SiO3/NaOH ratios in a geopolymer with only POBA as precursor. The maximum compressive strength (11.5 MPa) was obtained using POBA with S/L and Na2SiO3/NaOH ratios of 1.5 and 2.5, respectively. The compressive strength obtained in this study was low, since the mortar was cured at a high temperature (80 C); at room temperature 0.4 MPa was reached, and 3.2 MPa for 50 C. Many studies have investigated the use of POFA combined with other mineral admixtures, such as MK, FA), and GGBS. Mijarsh et al.48 used a treated POFA (POFA heated at 500 C to remove the unburned carbon, named TPOFA) combined with SF, calcium hydroxide (CH), and aluminum hydroxide (AH). In the first article, the authors applied the Taguchi method for assessing parameters affecting the development of compressive strength. A total of 25 experiments were performed to study six factors (effect of CH, SF, and AH; effect of sodium hydroxide concentration; effect of Na2SiO3/NaOH ratio; effect of solid/alkaline activator ratio). The optimum parameters were 20% Ca(OH)2, 5% SF, 10% Al(OH)3, Na2SiO3/NaOH ratio 5 2.5, and solid/activator ratio 5 2.13. The mortar with these parameters had a compressive strength of 47.27 MPa at 7 days of curing. Mijarsh et al. published two articles in 2015. In the first article49 they studied mixtures with different proportions of source materials (pure TPOFA, TPOFACH 80:20; TPOFACH 90:10; TPOFASF 95:5; TPOFACHAHSF 65:20:10:5). A combination of waterglass and NaOH was chosen as the alkaline activator, and the mortars were cured for 1 h at 27 C, then cured at 75 C for 24 h, and finally cured at 27 C. At 28 days of curing the mixture containing pure TPOFA as precursor had compressive strength equal to 28.81 MPa while the mortar containing TPOFACHAHSF had 52.48 MPa. It was observed that the type of reaction product was directly dependent on the composition of the mixture, and that with a longer curing time the gel type C-A-S-H was predominant. In the other article, Mijarsh et al.,50 investigated the effect of the concentration of Na2SiO3, the curing time, and the delay time (time during which the specimens were left at room temperature until they were placed in the oven for curing at high temperature). The authors performed this study using the optimum system described in the previous article. They concluded that the amount of reaction products depended on the concentration of the alkaline activator, and that there was a correlation between these products and the mechanical strength. The mortars with the highest concentration of Na2SiO3 (c. 113 Kg/m3 of mortar) had the best developed strength at 120 days. Another group of geopolymers composes the system POFAMK. The group of Hawa et al. have published four articles on this system. In two articles they prepared mortars with 0%, 5%, 10%, and 15% substitution of MK by POFA.51,52 The ratio Na2SiO3/NaOH was 2.5 and the mixing was performed at high temperature. The mortars were cured in an oven at 80 C for 1, 2, or 4 h, and after this, were cured at room temperature until the testing date. They concluded that the optimum substitution was 5% and optimum curing time was 4 h (70 MPa at 28 days). The same group published two articles on the use of POFA, MK and PWA (Parawood ash). In the first article, they studied the properties of pavement

New inorganic binders containing ashes from agricultural wastes

141

repair material based on geopolymeric materials. Geopolymer mortars were prepared with MK and different proportions, between 5% and 15%, of substitution of MK by POFA or PWA. The curing conditions and the process of mixture were the same as in the aforementioned study. They concluded that POFA was superior to PWA in strength characteristics. The bond strength to OPC mortar was better in the mortars with PWA and POFA than the mortar with only MK.53 In the second article, they studied the same mixtures as in the previous articles under more aggressive conditions. They concluded that the mortars with more than 30% of PWA or 15%20% POFA displayed greater expansion because of their greater porosity and CaO content.54 Finally, Ismail et al.55 published a paper in which MK was replaced with POFA at proportions between 0% and 80%. They studied the compressive strength after 3 and 7 days. The replacement of MK by POFA improved the workability and compressive strength with 40% and 60% of replacement (nearly 70 MPa at 7 days). Kabir et al.56 studied a ternary system, in which the mixtures had different proportions of MK, POFA, and GGBS and were activated with Na2SiO3/NaOH (ratio 2.5). The mixture with the highest compressive strength was the mortar with 20% MK, 35% GGBS, and 45% POFA (48 MPa after 28 days). They concluded that the mixtures with percentages of POFA higher than 45% had less strength because the quantity of Al was insufficient. For all mortars the compressive strength was better with higher molarities of NaOH. There are various articles focusing on the use of POFA and slags. Salih et al.45 prepared a binder with high strength development at room temperature. They had substituted POFA by GGBS at 5% (10%50%), and a Na2SiO3/NaOH mixture was chosen as an alkaline activator (ratio 2.5). A larger proportion of GGBS in the mixture produced a decrease in the setting time: if the replacement was more than 20% the reduction in the setting time was especially great. Fig. 5.8 shows the compressive strength (Cs) of alkali-activated POFA paste at 28 days. As can be seen, the paste with 50% of GGBS has strength more than double that of the paste with 100% POFA. The inclusion of GGBS produces a higher degree of polymerization and a higher reaction rate, and in this system a gel of C-A-S-H type was the main reaction product. Karim et al.57 fabricated a geopolymer using slag, POFA, and RHA; these materials were activated only with NaOH (used at 2.5%, 5.0%, and 7.5% by weight of precursor). The authors fixed the content of RHA, while the slag and POFA were distributed at 3:2 ratio. They made mixtures with materials before and after grinding to study the influence of the grinding process. The setting time of the mixtures was influenced by the quantity of Al2O3 and SiO2: the mixtures with more SiO2 had longer setting times, while the mixtures with more Al2O3 had shorter setting times. The mortar with 42% slag, 28% POFA, 30% RHA, and 5% NaOH was considered as the optimum mixture for this type of geopolymer. Yusuf et al. described, in various articles in the period 201415, the use of POFA and slag. The POFA in all the articles was calcined at 500 C and milled to obtain an ultrafine material.5860 In 2014 Yusuf et al. published three articles using POFA and GBFS. In the first article,58 they studied the optimum parameters in

142 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites 90 80 70

Rc (MPa)

60 50 40 30 20 10 0

POFA

10% GGBS

20% GGBS

30% GGBS

40% GGBS

50% GGBS

Figure 5.8 Compressive strength at 28 days for geopolymeric pastes with palm oil fuel ash (POFA) and ground granulated blast furnace slag (GGBS). Adapted from Salih, M.A.; Farzadnia, N.; Ali, A.A.A.; Demirboga, R. Development of High Strength Alkali Activated Binder Using Palm Oil Fuel Ash and GGBS at Ambient Temperature. Constr. Build. Mater., 2015, 93, 289300. doi: 10.1016/j.conbuildmat.2015.05.119.

mixtures of POFA/GBFS, and concluded that the percentage of GBFS should be maintained at 20% and that thermal curing at 65 C for 24 h produced mortars and concretes with good performance. In the second article,59 the effect of the H2O/NaOH ratio on the strength of mortars with POFA and GBFS was studied. They concluded that there is a recommended range of the H2O/NaOH ratio (18.920.5) to achieve a concrete with higher compressive strength (higher than 50 MPa) and good workability. A high H2O/NaOH ratio improves the ionic mobility and produces less amorphous products, but under these conditions the compressive strength is lower (3442 MPa). Finally, the same authors60 studied the effect of curing methods and concentration of NaOH on the strength of POFA/GBFS geopolymer. They concluded that the best curing method was oven curing at 65 C, but that water curing may be a complementary method. The best ratio of Na2SiO3/ NaOH was 2.5, and when using concentrations of NaOH more than 10 M, negative effects on the compressive strength were observed. Additionally, the group of Yusuf et al. presented two papers61,62 regarding the durability of mortars. They studied the performance of geopolymer in sulfate and sulfuric acid environments. The study in sulfate environment used mortars made with 100% POFA and 80% POFA:20% GBFS, and the mortars were exposed to 5%wt Na2SO4 and 5%wt MgSO4 for 6 months. The authors concluded that the mortars with GBFS had better sulfate resistance and that the compressive strength loss was higher for mortars subjected to the Na2SO4 environment. In mortars submerged in MgSO4, surface deposits of calcium sulfate salts were observed.61 The performance of this type of mortar exposed to a sulfuric acid environment was also examined,62 and the authors concluded that the optimum range of slag substitution was 20%40% and that the effect of the added water content

New inorganic binders containing ashes from agricultural wastes

143

(5 to 15%wt respect to solid precursor) was an important point in the degree of sulfuric resistance. The presence of too much water increased the permeability and enhanced the mobility of the ions. The binary systems with POFA containing fly ashes (POFAFA) were studied. The first study using FA was published in 2011 by Ariffin et al. 63 The authors studied geopolymer concrete with various FA:POFA ratios (0:100, 30:70, 50:50, and 70:30), and analyzed the influence of NaOH molarity (8 and 14 M) and the ratio Na2SiO3/NaOH (2.5 and 1.0). They concluded that the compressive strength was significantly influenced by the FA content: the mortar with the highest strength was 70 FA:30 POFA (25 MPa at 28 days, cured at room temperature). The best ratio Na2SiO3/NaOH was 2.5 and the optimum NaOH molarity was 14 M. The geopolymer concrete possessed higher compressive strength at higher curing temperature. Ariffin et al.64 published a paper in which the sulfuric acid resistance using the best mixture from the previous article (70 FA:30 POFA) was assessed. The mortars were exposed to 2% solution of sulfuric acid for 18 months; a mortar made with OPC was prepared as a control sample. The control mortar OPC presented a mass loss around 20% and a compressive strength loss of 68% after 18 months. The geopolymer mortar had a mass loss around 8% and a compressive strength loss of 35% in the same period. The N-A-S-H gel (geopolymer FAPOFA) appeared to have better resistance to sulfate attack than the C-S-H gel (OPC mortars). Bhutta et al.65 studied the performance of concrete with a proportion of 70% FA and 30% POFA exposed to a 5%wt sodium sulfate solution for up to 18 months, removing the solution every month. The principal form of damage to OPC concrete was the formation of gypsum and ettringite. The geopolymeric mortars have better chemical resistance to sulfate attack, with a mass loss of only around 4%, and had a more stable cross-linked aluminosilicate structure. Liu et al.66 published two articles regarding the fabrication of a lightweight geopolymer concrete. The materials employed in the articles were the same as above: they used FA and POFA as binders, and included the use of oil palm shell (OPS) as a lightweight aggregate. The authors made a concrete with 20% POFA and 80% FA, and used Na2SiO3/NaOH as an alkaline activator (ratio of 2.5). They concluded that to obtain the requirements of the splitting tensile strength then the content of POFA may be lower than 20%. Liu et al.67 published another study investigating the thermal conductivity of some lightweight geopolymers based on 20%wt POFA and 80%wt FA. It was demonstrated that this material has lower thermal conductivity (0.470.54 W/mK) than conventional materials (blocks 0.60 W/mK and bricks 0.90 W/mK). This material can be categorized as a structural and insulating concrete. Ranjbar et al.68 employed mixtures of POFA:FA with different percentages of substitution of FA by POFA (between 0 and 100%). The molarity of the NaOH and the ratio of Na2SiO3 to NaOH were 16 M and 2.5, respectively. The mortars were cured for 24 h at 65 C. Mortars with only FA reached 97% of their ultimate compressive strength after 7 days (higher than 35 MPa). The mortar with only POFA increased slightly in strength from 7 days (11 MPa) up to 112 days (26 MPa). This group also executed a study in which the previously cited geopolymers were

144 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

exposed to high temperatures up to 1000 C.69 They concluded that all the geopolymer mixtures gained strength when exposed to temperatures up to 300 C; the mixtures with high content of FA achieved the highest strength at 300 C, while the mortars with high content of POFA increased in strength until 500 C. The stability at high temperatures (8001000 C) was better in the mortars with high FA content. Bashar et al.70 studied the effect of varying the molarity of the alkali activator and the effect of the fine aggregate type on the FAPOFA system. The authors kept constant the binder:fine aggregate, POFA:FA, water/binder, and alkaline activator/binder ratios (1:1.5, 1:1, 0.2, and 0.4, respectively), and varied the molarity of NaOH and the proportion of natural, manufactured, and quarry dust sand. They concluded that the mixtures with high molarity of NaOH have higher compressive strengths and that the use of quarry dust or manufactured aggregate did not affect the compressive strength development with respect to the mortar with natural sand. The same researchers published an additional paper71 in which POFAFAGGBS systems with manufactured sand were analyzed. They concluded that the GGBSPOFA system have better performance than the FAPOFA system, but recommended the use of percentages below 40% of POFA. The best compressive strength was obtained by the mortar with 70 GGBS:30 POFA (66 MPa at 28 days). Finally, Hussin et al.72 studied the performance of POFAFA geopolymer at elevated temperatures. The authors heated the mortar up to 800 C and compared the evolution of the geopolymer mortar (30 POFA:70 FA) versus OPC mortar. They observed that OPC mortars failed at 400 C while the geopolymeric mortar lost strength at 600 C. The mass loss of the OPC mortar at 800 C was nearly 45% while the mass loss of the geopolymer was just under 15%. The phases detected by XRD in the geopolymer were hydroxysodalite and analcime in the range of 200400 C, nepheline at 600 C, and albite at 800 C.

5.2.4 Wood ash Biomass FAs produced by burning wood have been studied recently in terms of geopolymer design. A few articles have proposed the use of these ashes.73,74 Rajmma et al.73 studied the alkali activation of biomass FA (produced by burning waste from eucalyptus forests) and MK. Different mixtures were prepared, varying the molarity of NaOH, the ratio of Na2SiO3/NaOH, and the percentage of MK. The mortar of only biomass FA gave a compressive strength of 18 MPa at 10 days while the mortar incorporating 40% of MK yielded around 38 MPa. Ban et al.74 proposed the fabrication of a low-alkalinity geopolymer with the use of FA and wood ash (WA). The authors concluded that the WA with the presence of arcanite (K2SO4) and calcium hydroxide (Ca(OH)2) could be considered as an alkali activator. They used a pressurized forming technique to fabricate mortar blocks with compressive strengths above 18 MPa.

New inorganic binders containing ashes from agricultural wastes

5.3

145

Hydrated lime-based binders containing ashes from agricultural wastes: case studies

The growing demand for new construction is causing problems for the manufacture of building materials, due to the availability of natural resources. Moreover, large energy consumption, large air and water pollution, and a substantial environmental impact are created in the production of conventional construction materials such as OPC. In order to improve the sustainability of the fabrication of construction materials, it is necessary to adopt cost-effective, environmentally appropriate technologies and upgrade traditional techniques with available local materials. We must work in a sustainable environment, where we can meet our needs without seriously compromising those of future generations.75 In that sustainable environment, the main issues are reductions in the use of natural and nonrenewable raw materials, and the atmospheric emission of harmful gases, especially those involving a contribution to the greenhouse effect (greenhouse gases, GHGs).76 Many agricultural wastes have pozzolanic properties, which can be a source of supplementary material to improve sustainability in Portland cement mortars and concrete. The using of agrowaste in sustainable construction materials provides a solution that offers a reduction in natural resource as well as energy consumption.77 The utilization of agricultural waste could provide the breakthrough needed to make the industry more environmentally friendly and sustainable. Agricultural wastes, such as RHA, wheat straw ash, SCBA, and hazel nutshell ash, are pozzolanic materials with interesting properties.78 Lime was known in the sixth millennium BC as a building material for mortars and coatings. It has been found to have been used in ancient Egypt, the Assyrian Empire, Classical Greece, and the Roman Empire. Moreover, outside the Mediterranean area, it was used by the Mayas, Incas, and Aztecs in America and by the earliest Chinese dynasties and also the first Indian dynasties.79 Air lime mortars have advantageous properties because they have the particularity of being permeable to water vapor (but not to rain), which allows the walls to transpire, providing hygroscopic properties to regulate the humidity of the environment.80 Hydraulic mortars have been used since the Greek and Roman times. In these, lime is mixed with pozzolans (volcanic ash or ground ceramic) and the mixture is hardened under water. The work of Cook et al.,81 Mehta,82 and other researchers has shown that a Portland cement substitute can be obtained by mixing RHA and hydrated lime for use in construction.83 In construction, about 50% of Portland cement (OPC) is used for secondary applications such as masonry and plastering; however, the performance of OPC in terms of strength is not always needed for these purposes. The strength of the binder needed for these applications is in the range of 415 MPa. RHAlime binder can replace Portland cement for such applications. The strength of RHA-based binder depends on the reactivity of the ash and the limeRHA ratio.84 In the following sections of this chapter the state of the art of using different types of agricultural waste ashes in mixtures with lime is summarized.

146 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

5.3.1 Rice husk ash Walker and Pavia85 studied the influence of mineral admixture properties (particle size, specific surface area, chemical and mineral composition, amorphousness, and water demand) that affect the reactivity and strength of limeadmixture pastes. Materials with pozzolanic behavior, such as pulverized fuel ash, calcined clay, microsilica, RHA, red brick dust, tile and yellow brick dust, were assessed. The researchers concluded that amorphousness determined pozzolan reactivity and also strength to a much greater extent than any other pozzolan property, and also established that the specific surface area of the pozzolan governed the water demand of the paste. Souza et al.86 studied the hydration of pastes containing calcium hydroxide and either RHA or SCBA in various initial CaO/SiO2 (C/S) molar ratios. The authors evaluated how the origin of the silica in the system and the initial C/S ratio affected the final products. It was concluded that the nature of the C-S-H from the pozzolanic reaction between RHA and Ca(OH)2 depended on the C/S ratio of the mixture. For low C/S ratio mixtures, the main product was C-S-H (I), while for higher C/S ratio pastes, there were both C-S-H (I) and C-S-H (II), or only C-S-H (II). The RHA mixtures with C/S ratios between 0.6 and 1.0 had all Ca(OH)2 consumed after 50 days of curing at 40 C. In RHA pastes, the C-S-H appeared as fibrillar structures, deposited on the surface of the ash. Finally, the morphology of the C-S-H formed in SCBACa(OH)2 was different from that formed by RHA. The hardening of mortars composed of cement, RHA and lime with different compositions was studied by Cizer et al.87,88 Thermal analysis, mechanical strength testing, and SEM were used. Three different mortar compositions were studied: cementRHA mortars, cementRHAlime mortars, and RHAlime mortars. After Portland cement (OPC) was partially replaced by RHA and lime (L), in the OPC-RHA-L 107020 and 105040 proportions, hardening occurred as a result of combined effect of the cement hydration, pozzolanic reaction, and carbonation reaction. Although cement hydration contributed to the early strength development of the mortars, carbonation was much more pronounced at later stages due to the porosity of the mortars. Flexural strength reduction was recorded after 28 days of curing for cementRHAlime mortars containing 10 wt% of cement. Pavia et al.89 studied the effect of RHA on the properties of hydrated lime (calcium limeCL90s) mortars with a view to improving their properties and making them more sustainable. The variation in mortar properties was related to the activity of the RHA (specific surface area, fineness, reactivity, composition, and amorphousness). The results suggested that the RHA was probably produced at low temperature (400500 C), and highly reactive ash with high specific surface area contained crystalline silica (cristobalite) and unburnt cellulosic material. Pozzolanic hydrates were clearly present after 24 h, progressively increasing in size and amount (at 3 and 7 days) and linking to each other to form continuous networks after 14 days. The authors concluded that the replacement of lime by RHA improved mortar workability, lowering the water/binder ratio and the amount of water required to reach a specific consistency. RHA also enhanced the bulk

New inorganic binders containing ashes from agricultural wastes

147

density and lowered the difference between real and bulk densities, thus resulting in lower porosity. The replacement of lime by RHA enhanced the strength, and accelerated the setting and early strength gain. Finally, with increasing RHA content, lime mortars became progressively stiffer and more elastic, yet remained plastic and thus underwent significant strain before failure. A discussion on the comparison of RHA samples from different field ovens (annular enclosure, enclosure made of bricks and pit burning) was made by Nair et al.90 The authors also recommended a feasible method to produce RHA pozzolan in uncontrolled conditions in rural environments as an alternative to cement for building applications requiring lower strengths. Different types of ashes were produced and the long-term strength of RHA pozzolan with lime or cement was investigated to suggest a sustainable, affordable option in rural building applications, especially for rural housing in Kerala, a southern state of India. LimeRHA and cementRHA mortars in 1:3 proportions (by weight) were prepared for different proportions of limeRHA and cementRHA combinations. The comparison of the properties of ashes from different field ovens confirmed the superiority of annular enclosure over the other two field ovens. These results also showed that RHA samples with lower values of loss on ignition and higher specific surface area resulted in the production of relatively higher-strength pozzolan-based mortars. For rural building applications, sophisticated ovens and techniques are not feasible or affordable to attain the above criteria. Billong et al.91 studied the effects of MK on the density and grindability (capacity for being ground) of powdered mixtures, and the chemical and mineralogical composition, setting time, microstructure, water absorption, and mechanical properties of hydrated RHAlime binders. This research was motivated by the fact that the alumina content in MK can contribute to the formation of new phases during the hydration of hydrated lime-based binders. From the results obtained, it was concluded that MK increased the density of the mixtures (from 2.59 g/cm3 for 100%MK to 2.29 g/cm3 for 100%RHA) and decreased their grindability (specific surface area: 8.12 m2/g for 100%MK to 13.38 m2/g for 100%RHA). The presence of MK decreased the SiO2 content of the binders and increased their Al2O3 and Fe2O3 contents. CSH gel and hydrated gehlenite (C2ASH8) were the main phases formed during the pozzolanic reaction in the presence of MK. No reduction in flexural and compressive strengths was observed after 28 days for binders containing MK. The mixture of 25% MK and 75% RHA, which was recommended, presented greater flexural and compressive strengths than mixtures of binders with RHA or MK, as the only pozzolan. Water absorption for all the tested mortars was less than 20%. Mendez et al.92 presented preliminary results on the influence of the RHA/lime ratio on the workability and strength of mortars. The results showed that mortar workability improved with increasing RHA/lime ratio. The compressive strength (Cs) of mortars with different RHA/lime ratios was studied, in the same way, for 28 days curing time at 20 C, and an increase in Cs with increasing RHA/lime ratio was observed. However, for 90 and 180 days curing times, the maximum Cs was obtained for an RHA/lime ratio equal to 2. The lowest and highest Cs values

148 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

obtained were 6 and 18 MPa, respectively, when 20 C curing temperature was used. Above 65 C of curing temperature, a similar tendency of Cs with respect to the RHA/lime ratio was observed. A preliminary study of binders for use in mortar tiles revealed that at least low quantities of Portland cement must be included in the binder composition in order to obtain short-term strengths that facilitate the tile demolding process. In general, it can be concluded that RHAlime mixtures show good results. They could be a technologically viable low-cost alternative to basic mortars, to be used mainly in developing countries or rural areas, in rice-producing countries, and in regions with thermal processing of rice husk to derive energy and provide RHA, and limited to situations that do not require high mechanical strength at short curing times. The use of RHA as studied in this report would reduce the production of clinker, which contributes to the greenhouse effect. This would also promote environmentally friendly waste management, because the RHA used is an agricultural waste material and has lower economic and environmental costs than OPC. Stroeven et al.93 used RHA in the economic production of low- to high-strength hydraulic binders. Some vegetable wastes contain relatively large amounts of silica; one of them is rice husk. This is very interesting because rice is a major crop in many developing countries, and it is available in large quantities. RHA can be produced without using expensive fuels, such as in the case of Portland cement, where during incineration heat can be transformed into mechanical energy for grinding RHA. The RHA used in the above mentioned investigations was from the Arusha region in Northern Tanzania. RHA was used in combination with Portland cement (OPC) (RHAOPC), with OPC and lime (RHAlimeOPC), and with lime (RHAlime). The compressive strength for low-strength mortars for rural applications over a period of 1 year ranged from 1.55 to 13.6 MPa depending on the composition of the ternary mixture RHAlimeOPC. Koteswara et al.94,95 analyzed the influence of mixtures of RHA and lime on the stabilization of expansive soil and marine soil. Problems with lack of dimensional stability of soils have been reported all over the world. In the rainy season expansive soils swell and in summer they shrink. Because of this alternating swelling and shrinkage, lightly loaded civil engineering structures like residential buildings, pavements, and canal linings are severely damaged. The stabilization of expansive soils with various additives has been highly successful. In this study, RHA, lime, and gypsum were added to an expansive soil, which resulted in considerable improvement in the strength characteristics of the soil. The following conclusions were drawn: the liquid limit of the expansive soil was decreased by 22% with the addition of 20% RHA+5% lime, and the free swelling index of the expansive soil was reduced by 88%. The unconfined compressive strength (UCS) of the expansive soil was increased by 548% with the addition of 20% RHA + 5% lime + 3% gypsum after 28 days curing. The CBR (California Bearing Ratio) value of the expansive soil was increased by 1350% with the same addition after 14 days curing. RHA can potentially stabilize expansive soil either alone or mixed with lime and gypsum. The utilization of industrial wastes like RHA, lime, and gypsum is an alternative method to reduce the construction cost of roads, particularly in the rural areas of developing countries.

New inorganic binders containing ashes from agricultural wastes

149

Koteswara et al.95 also studied the effect of RHA and lime on the strength properties of marine soil. In this study the properties of marine soil with RHA were determined and the performance of marine soil when stabilized with RHA was evaluated (CBR increased 282% by addition of RHA to the tested soil, and increased 449% with additional 9% of lime). It was concluded that the stabilized marine soil was suitable to use as a subgrade material for pavement construction and also for various foundations of buildings. Shrivastava et al.96 evaluated the feasibility of using RHA with lime to stabilize black cotton soil. This kind of soil exhibits high swelling and shrinking when exposed to changes in moisture content and hence has been found to be highly problematic for engineering considerations. This behavior is attributed to the presence of montmorillonite. To counter the high degree of expansiveness, innovative and nontraditional research on waste utilization is currently gaining importance. Soil improvement using waste material such as slag, RHA, SFs etc. in geotechnical engineering has been practiced from an environmental point of view. A series of laboratory experiments was conducted on 5%-lime mixed black cotton soil blended with RHA at 5%, 10%, 15% and 20% by weight of dry soil. The experimental results showed a significant increase in CBR and UCS. The CBR values increased by 287% and the UCS improved by 30% The differential free swelling of the black cotton soil was reduced by 87% with an increase in the RHA content from 0% to 20%. From this investigation it can be concluded that RHA has the potential to improve the black cotton soil to have less shrinkage and less expansion. Behak and Peres97 studied the reactivity of RHA, and of mixtures with sandy soil and lime, as a function of the kind of rice husk and the temperature of burning. X-ray diffraction analyses and loss-on-ignition tests were carried out on RHA. X-ray diffraction analyses, UCS tests, and splitting tensile strength tests were conducted on mixtures of sandy soil with different RHA and lime contents. The results showed that the optimal reactivity of the RHA was reached for a controlled temperature range of 650800 C, providing a significant increase in the compressive strength and stiffness of the mixtures. The results indicated that the stabilization of sandy soils with RHA and lime could provide an alternative material for subbase and base layers of low-volume traffic pavements in regions where highperformance materials are not available, such as rice production regions. The use of such material would provide significant improvement to the road networks, with socioeconomic consequences. It would also contribute to the preservation of the environment by employing a residue and reducing the exploitation of deposits of nonrenewable resources such as soils and rocks. Two methods based on electrical conductivity measurements to evaluate the pozzolanic reactivity of RHA have been proposed. The first one is a very simple method based on electrical conductivity and pH measurements, proposed by Tashima et al.98 In this method, calcium hydroxide: pozzolan water suspensions were monitored. In the experiments, calcium hydroxide was suspended in deionized water to yield a lime-saturated suspension. Three testing temperatures were selected (40, 50, and 60 C). The addition of siliceous pozzolan (RHA or densified silica

150 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 5.9 Loss in conductivity (Lc) (%), for CH:RHA suspensions (2:8 to 4.5:5.5) at 40 C. Adapted from Tashima, M.M.; Soriano, L.; Monzo´, J.; Borrachero, M.V.; Akasaki, J.L.; Paya´, J. New Method to Assess the Pozzolanic Reactivity of Mineral Admixtures by Means of pH and Electrical Conductivity Measurements in Lime:Pozzolan Suspension. Materiales de Construccio´n, 2014, 64, 316. doi: 10.3989/mc.2014.00914

fume, DSF) to the saturated lime suspension could result in unsaturation of the system, depending on the testing time, testing temperature, and reactivity of the pozzolan. The unsaturation degree was calculated by measuring the loss in electrical conductivity Lc (Fig. 5.9). When unsaturation was reached, the loss of electrical conductivity was higher than 30% and the variation of the pH was higher than 0.15 units. A classification of pozzolanic materials into three reactivity levels (high, medium and low) was proposed (Fig. 5.10). Amorphous RHA was classified as a high-reactivity material, and crystalline RHA as a low-reactivity material. The second method was proposed by Viet-Thien-An et al.,99 in which the pozzolanic reactivity of RHA and SF was assessed by measuring the concentration of silicon, calcium ions, and pozzolanic reaction product in saturated portlandite solution at 40 C, in addition to electrical conductivity measurements. The differences in surface structure and alkali content, and possible differences in the nature of dissolved Si, between RHA and SF are important parameters influencing their Ca2+ ion adsorption and dissolutionprecipitation behavior, and hence the electrical conductivity of portlanditepozzolan suspensions. SF possesses a higher pozzolanic reactivity than RHA. The higher the specific surface area of RHA, the higher the pozzolanic reactivity.

5.3.2 Sugarcane bagasse/straw ash There are some studies related to the use of SCBA and SCSA as mineral admixtures for both lime stabilized soil100 and soil block production.101,102 Nevertheless,

New inorganic binders containing ashes from agricultural wastes

151

Figure 5.10 Template for classification of pozzolan reactivity: High reactivity zone (1), medium (2) and low (3). Adapted from Tashima, M.M.; Soriano, L.; Monzo´, J.; Borrachero, M.V.; Akasaki, J.L.; Paya´, J. New Method to Assess the Pozzolanic Reactivity of Mineral Admixtures by Means of pH and Electrical Conductivity Measurements in Lime:Pozzolan Suspension. Materiales de Construccio´n, 2014, 64, 316. doi: 10.3989/mc.2014.00914

most studies are related to the use of SCBA/SCSA as pozzolanic material.18,103 In all cases, the calcination processes of SCBA/SCSA are crucial. In 1998, Martirena et al.104 reported the use of wastes from the sugar industry (SCBA and SCSA) as pozzolan in lime/pozzolan binders. In that study, SCBA was extracted directly from the boilers of a sugar factory while SCSA was sampled from heaps of open-air burnt straw. Based on the obtained results, the authors concluded that SCBA can be classified as a pozzolanic material, but is influenced by the calcination temperature and by the incomplete combustion that takes place in the boilers. These factors affect the reactivity of SCBA due to the crystallization of silica present in the ash and the presence of impurities, mainly unburned material. In the same way, SCSA was classified as a very reactive pozzolanic material that fulfils the main requirements for pozzolans, probably due to the low temperatures reached during the calcination process providing an amorphous structure for the silica present in the ash. In other studies performed by Martirena et al.,105,106 the authors proposed the construction of a rudimentary, low-tech incinerator (Fig. 5.11) in order to produce ashes with enhanced properties. Some considerations were taken into account in the design of the incinerator: the temperature in the furnace cannot reach 700 C, the cooling of the burning chamber was performed through a natural air draught to evacuate exhaust gases, and the speed of the airflow must be slow enough to avoid a significant loss of ash through the chimney. As conclusions, the authors detected three main problems in the low-tech incinerator that reduced the reactivity of the obtained ash: long residence time, cooling rate of the ash, and low output of ash. The influence of calcining temperature on the reactivity of SCBA/SCSA by means of kinetic parameters was reported in several papers.107110 At temperatures of about 800 C, the SCBA/SCSA presented a very high pozzolanic reactivity even for short curing times (, 7 days of curing). Ashes obtained at 1000 C presented a decreased activation rate for the first days of reaction but, for long curing times, their activity was similar to SCBA obtained at 800 C.

152 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 5.11 Low-tech incinerator to produce reactive pozzolans. Source: Martirena, F.; Middendorf, B.; Day, R.L.; Gehrke, M.; Roque, P.; Martı´nez, L., et al. Rudimentary, Low Tech Incinerators as a Means to Produce Reactive Pozzolan Out of Sugar Cane Straw. Cement Concrete Res., 2006, 36, 10561061. doi: 10.1016/j.cemconres.2006.03.016.

Cordeiro et al.103 reported a study of the pozzolanic activity and filler effect of SCBA, collected from a sugar and alcohol factory, in Portland cement and lime mortars, using different grinding procedures for SCBA. According to the authors, SCBA showed appropriate physicochemical properties, and its pozzolanic reactivity depended on the particle size distribution and fineness. In 2015, Moraes et al.18 performed a specific study related to the pozzolanic activity of SCSA obtained by an autocombustion process. The pozzolanic reactivity of SCSA was assessed by means of thermogravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), and SEM for both OPC and calcium hydroxide pastes and the compressive strength test for OPC mortars. TGA of calcium hydroxide: SCSA pastes (3:7 mass ratio, cured at 20 C) showed that all the available calcium hydroxide was completely fixed by SCSA after only 3 days of curing, and FTIR analysis confirmed these results. SEM images reported an amorphous structure formed by the pozzolanic reaction, in which some C-S-H gels appeared as needle-like products (Fig. 5.12). Marduwar et al.101 assessed the possibility of using SCBA in the production of bricks. The bricks prepared using SCBA, quarry dust, and lime achieved all the requirements of class 3.5 common burnt clay bricks. Bricks containing 50% SCBA, 30% quarry dust, and 20% lime exhibited a water absorption of 19.7% (less than 20%), compressive strength of about 6.59 MPa (higher than 3.5 MPa), and lower thermal conductivity (κ) when compared with commercial clay brick. The improvement of the durability and mechanical properties of compacted soil blocks using SCBA and lime was reported by Alave´z-Ramı´rez et al.102 The

New inorganic binders containing ashes from agricultural wastes

153

Figure 5.12 SEM images of SCSA/calcium hydroxide pastes (3:7 mass ratio, cured at 20 C): (A) and (B) pastes after 28 days of curing. Source: Moraes, J.C.B.; Akasaki, J.L.; Melges, J.L.P.; Monzo´, J.; Borrachero, M.V.; Soriano, L., et al. Assessment of Sugar Cane Straw Ash (SCSA) as Pozzolanic Material in Blended Portland Cement: Microstructural Characterization of Pastes and Mechanical Strength of Mortars. Constr. Build. Mater., 2015, 94, 670677. doi: 10.1016/j.conbuildmat.2015.07.108.

compacted soil blocks containing 10% SCBA and 10% lime presented a significant improvement in their mechanical and durability properties. Moreover, the authors concluded that the combination of SCBA and lime as a replacement for OPC in the stabilization of soil blocks was an environmentally friendly alternative due to the reduction in energy consumption and CO2 emissions.

5.3.3 Other biomass ashes Nakanishi et al.111 studied the behavior of two elephant grass ashes as pozzolans, elephant grass Cameroon (EGC) and Napier (EGN), calcined under controlled conditions (700 C). They reported preliminary scientific aspects (chemical and mineralogical characterization, pozzolanic properties, identification and evolution of hydrated phases, and morphology) of the use of these new industrial wastes in ash/Ca (OH)2 pastes. Both ashes showed high pozzolanic activity after 7 days of curing, with at least 85% of fixed lime in 1:1 by weight pastes. The thermogravimetric and micro-Raman analysis of elephant grass ashes/Ca(OH)2 pastes confirmed that the main hydrated phases during the pozzolanic reaction consisted of CSH gels. The results obtained support the viability of recycling elephant grass ashes in order to obtain future SCMs as an alternative to traditionally used pozzolans, leading to environmental and socioeconomic benefits. Cobreros et al.112 characterized barley straw ash (BSA), and other natural and artificial pozzolans from Mexico, such as wheat straw ash, volcanic ash, fired clay brick, and FA. The pozzolanic properties of BSA were compared with those of the pozzolans mentioned above. To determine the pozzolanic properties, the water-soluble fraction, fineness, lime-pozzolan compressive strength development, and lime consumption (using an electrical conductimetric method) were also evaluated. Additionally, the evolution of crystalline phases during the pozzolanic reaction was determined using X-ray diffraction and the change of micromorphology during

154 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

the pozzolanic reaction was evaluated by SEM. It was concluded that BSA was a pozzolanic material with good reactivity. However, it exhibited less pozzolanic activity in comparison with the dust of fired clay brick and FA. A comprehensive investigation of the mechanical strength of co-fired biomass, pure biomass, and coal FAs with hydrated lime mortar has been conducted under a carbonation-free environment with moist conditions, in a matrix of mixing ratios, curing temperatures, and testing days, by Wang.113 He concluded that wood FA from pure wood combustion, due to the much larger particle size and high unburned carbon content, had much poorer mechanical performance than coal and cofired biomass ash samples. The results obtained suggest that cofired biomass ashes obtained from cofiring biomass fractions (generally ,25% biomass by mass) have similar or better mechanical performances than coal ash samples, resulting mainly from pozzolanic reactions. The same researcher, Wang,114 studied the quantitative kinetics of pozzolanic reactions of class C and F and cofired biomass FAs with Ca(OH)2 (CH) under carbonation-free conditions. The gram mass of chemically combined CH per gram of FA increased with the CH mixing ratio, curing temperature, and curing time. The chemically combined mass ratio was approximated at 30/70 (CH/FA), although it varied slightly from ash to ash. The quantitative kinetics showed that class C and F, cofired biomass and pure biomass FAs shared the same diffusion mechanism. Finally, it was concluded that cofired biomass FA deserved to be re-evaluated for use in concrete, rather than being excluded from it by ASTM C 618 for its “noncoal” origin. Carrasco-Hurtado et al.115 have studied the effect of adding bottom ash from the process of biomass plant combustion to obtain calcium silicate samples, by partial replacement of the sources of lime and Portland cement with biomass bottom ash (BBA) in increments from 10% to 90% of the dry weight of the mixtures, for use as masonry units with low thermal conductivity. The experimental programme included a wide range of methods for the characterization of this by-product, including physical, chemical, and mineralogical properties as well as various testing methods of the developed material, such as mechanical strength, porosity, microstructure, freezethaw, and thermal conductivity. The optimal values are those containing a 1:1 ratio of SiO2/CaO, with compressive strength ranging from 25.21 MPa (BBA/CH) to 61.11 MPa (BBA/OPC) and thermal conductivity from 0.564 W/m  K (BBA/CO) to 0.773 W/m  K (BBA/CH) (where BBA is bottom ash, CO is calcium oxide, CH is calcium hydroxide and OPC is Portland cement). The initial results make it possible in principle to obtain calcium silicate samples with low thermal conductivity according to British Standard EN 7712:2011.

5.4

Final remarks

Countries are now looking for alternative construction materials to minimize greenhouse gas emissions. Most countries have adopted a new sustainable development agenda and global agreement on climate change, which materializes in the 17 Sustainable Development Goals and 169 targets. The global agreement on climate

New inorganic binders containing ashes from agricultural wastes

155

change is specified in Goal 13: “Take urgent action to combat climate change and its impacts.” From the above described case studies, one can conclude that many possibilities for using biomass waste ashes are available in terms of the preparation of alternative binders to Portland cement, to reduce climate change: 1. The preparation of alkali-activated cements from mixtures of precursor and high-alkalinity solutions allows the inclusion of biomass ashes. Firstly, as a part of the precursor, because biomass ashes usually have low percentages of reactive alumina (Al2O3), the main reagent together with silica (SiO2) to form cementing gels: the sole use of biomass ashes does not allow for good performance. Secondly, in a new approach to prepare high-sustainability binders, as a part of the alkaline solution: exploitation of the silica content in the biomass ashes as a chemical component in the sodium/potassium silicate-based activating solutions. Undoubtedly, this research approach will have interesting advances in the future. 2. The use of biomass waste ashes in the preparation of lime-based binders also has interest because it opens the possibility of the preparation of simple binders for application in purposes in which strength is not the main criterion.

Most biomass wastes are located in developing countries. If sound technological solutions are provided, recycling these wastes could become a low-cost and lowcarbon alternative solution ensuring a supply of cheap, easy-to-use, and durable construction materials for developing countries in the future. It is important to remember that “climate change is a global challenge that does not respect national borders: emissions anywhere affect people everywhere. It is a matter that requires solutions that need to be coordinated at the international level and it needs international cooperation to help developing countries move towards a low-carbon economy” (United Nations, UN http://www.un.org/sustainabledevelopment/climatechange-2/. Accesed 2017-03-13). The wide variety of applications for different biomass ashes in the development of new binders offers the possibility to apply them in the preparation of new composites in which the reinforcing phase is related to natural fibers or particles derived from biomass (Fig. 5.13). These materials will

Figure 5.13 The preparation of new biomass-based high sustainability composites.

156 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

be characterized by the valorization of alternative wastes in the preparation of new biomass-based high-sustainability composites.

References 1. Gartner, E. M.; Macphee, D. E. A Phisico-chemical Basis for Novel Cementitious Binders. Cement Concrete Res. 2011, 41, 736749. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2011.03.006. 2. Schneider, M. Process Technology for Efficient and Sustainable Cement Production. Cement Concrete Res. 2015, 78, 1423. Available from: http://dx.doi.org/10.1016/j. cemconres.2015.05.014. 3. Meyer, C. The Greening of the Concrete Industry. Cement Concrete Compos. 2009, 31, 601605. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2008.12.010. 4. Scheneider, M.; Romer, M.; Tschudin, M.; Bolio, H. Sustainable Cement Production— Present and Future. Cement Concrete Res. 2011, 41, 642650. Available from: http:// dx.doi.org/10.1016/j.cemconres.2011.03.019. 5. Phair., J. W. Green Chemistry for Sustainable Cement Production and Use. Green Chem. 2006, 8, 763780. Available from: http://dx.doi.org/10.1039/B603997A. 6. Shi, C.; Ferna´ndez-Jime´nez, A.; Palomo, A. New Cements for the 21st Century: The Pursuit of an Alternative to Portland Cement. Cement Concrete Res. 2011, 41, 750763. Available from: http://dx.doi.org/10.1016/j.cemconres.2011.03.016. 7. Provis, J. L.; Van Deventer, J. S. J. Geopolymers. Structures, Processing, Properties and Industrial Applications. A volume in Woodhead Publishing Series in Civil and Structural Engineering; Woodhead Publishing: Cambridge, UK, 2009. 8. Provis, J. L.; Van Deventer, J. S. J. Alkali Activated Materials: State-of-the-Art, Rilem TC 224-AAM; Springer: Dordrecht, 2014. 9. Pacheco-Torgal, F.; Labrincha, J.; Leonelli, C.; Palomo, A.; Chindaprasit, P. Handbook of Alkali-Activated Cements, Mortars and Concretes. Woodhead Publishing Series in Civil and Structural Engineering (number 54); Woodhead Publishing: Cambridge, UK, 2015. 10. Ma´rmol, G.; Santos, S. F.; Savastano, H., Jr; Borrachero, M. V.; Monzo´, J.; Paya´, J. Mechanical and Physical Performance of Low Alkalinity Cementitious Composites Reinforced with Recycled Cellulosic Pulp from Cement Kraft Bags. Ind. Crops Prod. 2013, 49, 422427. Available from: http://dx.doi.org/10.1016/j.indcrop.2013.04.051. 11. Pereira, C. L.; Savastano, H., Jr; Paya´, J.; Santos, S. F.; Borrachero, M. V.; Monzo´, J., et al. Use of Highly Reactie Rice Husk Ash in the Production of Cement Matrix Reinforced with Green Coconut Fiber. Ind. Crops Prod 2013, 49, 8896. Available from: http://dx.doi.org/10.1016/j.indcrop.2013.04.038. 12. Aprianti, E.; Shafigh, P.; Bahri, S.; Farahani, J. N. Supplementary Cementitious Materials Origin from Agricultural Wastes  A Review. Constr. Build. Mater 2015, 74, 176187. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.10.010. 13. Rosello´, J.; Soriano, L.; Santamarina, M. P.; Akasaki, J. L.; Melges, J. L. P.; Paya´, J. Microscopy Characterization of Silica-Rich Agrowastes to be used in Cement Binders: Bamboo and Sugarcane Leaves. Microsc. Microanal 2015, 21, 13141326. Available from: http://dx.doi.org/10.1017/S1431927615015019. 14. Vassilev, S. V.; Baxter, D.; Andersen, L. K.; Vassileva, C. G. An Overview of the Chemical Composition of Biomass. Fuel 2010, 89, 913933. Available from: http://dx. doi.org/10.1016/j.fuel.2009.10.022.

New inorganic binders containing ashes from agricultural wastes

157

15. Fairbairn, E. M. R.; Americano, B. B.; Cordeiro, G. C.; Paula, T. P.; Toledo Filho, R. D.; Silvoso, M. M. Cement Replacement by Sugar Cane Bagasse Ash: CO2 Emissions Reduction and Potential for Carbon Credits. J. Environ. Manage. 2010, 91, 18641871. Available from: http://dx.doi.org/10.1016/j.jenvman.2010.04.008. 16. Safiuddin, M.; Abdus Salam, M.; Jumaat, M. Z. Utilization of Palm Oil Fuel Ash in Concrete: A Review. J. Civil Eng. Manage. 2011, 17 (2), 234247. Available from: http://dx.doi.org/10.3846/13923730.2011.574450. 17. Adesanya, D. A.; Raheem, A. A. Development of Corn Cob Ash Blended Cement. Constr. Build. Mater. 2009, 23 (1), 347352. Available from: http://dx.doi.org/10.1016/ j.conbuildmat.2007.11.013. 18. Moraes, J. C. B.; Akasaki, J. L.; Melges, J. L. P.; Monzo´, J.; Borrachero, M. V.; Soriano, L., et al. Assessment of Sugar Cane Straw Ash (SCSA) as Pozzolanic Material in Blended Portland Cement: Microstructural Characterization of Pastes and Mechanical Strength of Mortars. Constr. Build. Mater. 2015, 94, 670677. Available from: http:// dx.doi.org/10.1016/j.conbuildmat.2015.07.108. 19. Villar-Cocin˜a, E.; Morales, E. V.; Santos, S. F.; Savastano, H., Jr.; Frı´as, M. Pozzolanic Behavior of Bamboo Leaf Ash: Characterization and Determination of the Kinetic Parameters. Cement Concrete Compos. 2011, 33, 6873. Available from: http://dx.doi. org/10.1016/j.cemconcomp.2010.09.003. 20. Kanning, R. C.; Portella, K. F.; Braganc¸a, M. O. G. P.; Bonato, M. M.; antos, J. C. M. Banana Leaves Ashes as Pozzolan for Concrete and Mortar of Portland Cement. Constr. Build. Mater. 2014, 54, 460465. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2013.12.030. 21. Cruz-Yusta, M.; Ma´rmol, I.; Morales, J.; Sa´nchez, L. Use of Olive Biomass Fly Ash in the Preparation of Environmentally Friendly Mortars. Environ. Sci. Technol. 2011, 45, 69916996. Available from: http://dx.doi.org/10.1021/es200968a. 22. Silva, R. B.; Fontes, C. M. A.; Lima, P. R. L.; Gomes, O. F. M.; Lima, L. G. L. M.; Moura, R. C. A., et al. Biomass Ash from Cocoa Agroindustry: Characterization and Use as a Cement Substitute. Ambiente Construı´do 2015, 15, 321334. Available from: http://dx.doi.org/10.1590/s1678-86212015000400053. 23. Paya´, J.; Monzo´, J.; Borrachero, M.V. Outstanding aspects on the use of rice husk ash and similar agrowastes in the preparation of binders, in I Pro-Africa Conference: NonConventional Building Materials Based On Agroindustrial Wastes, Pirassununga  October SP, Brazil, 2010; pp 1819. 24. Bouzo´n, N.; Paya´, J.; Borrachero, M. V.; Soriano, L.; Tashima, M. M.; Monzo´, J. Refluxed Rice Husk Ash/NaOH Suspension for Preparing Alkali Activated Binders. Mater. Lett. 2014, 115, 7274. Available from: http://dx.doi.org/10.1016/j, matlet.2013.10.001. 25. Bernal, S. A.; Rodrı´guez, E. D.; Mejı´a De Gutie´rrez, R.; Provis, J. L. Performance at High Temperature of Alkali-activated Slag Pastes Produced with Silica Fume and Rice Husk Ash Based Activators. Materiales de Construccio´n 2015, 65, 110. Available from: http://dx.doi.org/10.3989/m.c.2015.03114. 26. Detphan, S.; Chindaprasirt, P. Preparation of Fly Ash and Rice Husk Ash Geopolymer. Int. J. Miner. Metall. Mater. 2009, 16, 720726. Available from: http://dx.doi.org/10.1016/S1674-4799(10)60019-2. 27. Songpiriyakij, S.; Kubprasit, T.; Jaturapitakkul, C.; Chindaprasirt, P. Compressive Strength and Degree of Reaction of Biomass-and Fly Ash-based Geopolymer. Constr. Build. Mater. 2010, 24, 236240. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2009.09.002.

158 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

28. Hwang, C. L.; Huynh, T. P. Effect of Alkali-activator and Rice Husk Ash Content on Strength Development of Fly Ash and Residual Rice Husk Ash-based Geopolymers. Constr. Build. Mater. 2015, 101, 19. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2015.10.025. 29. Riahi, S.; Nazari, A.; Zaarei, D.; Khalaj, G.; Bohlooli, H.; Kaykha, M. M. Compressive Strength of Ash-based Geopolymers at Early Ages Designed by Taguchi Method. Mater. Des. 2015, 37, 443449. Available from: http://dx.doi.org/10.1016/j.matdes.2012.01.030. 30. Riahi, S.; Nazari, A. The Effects of Nanoparticles on Early Age Compressive Strength of Ash-based Geopolymers. Ceramics Int. 2012, 38, 44674476. Available from: http:// dx.doi.org/10.1016/j.ceramint.2012.02.021. 31. Karim, M. R.; Zain, M. F. M.; Jamil, M.; Lai, F. C. Development of a Zero-cement Binder using Slag, Fly Ash, and Rice Husk Ash with Chemical Activator. Adv. Mater. Sci. Eng., 2015, 2015, 114. Available from: http://dx.doi.org/10.1155/2015/247065. 32. Nazari, A. Fuzzy Logic for Prediction Water Absorption of Lightweight Geopolymers Produced from Waste Materials. Ceramics Int. 2012, 38, 47294736. Available from: http://dx.doi.org/10.1016/j.ceramint.2012.02.058. 33. Nazari, A. Artificial Neural Networks Application to Predict the Compressive Damage of Lightweight Geopolymer. Neural Comput. Appl. 2013, 23, 507508. Available from: http://dx.doi.org/10.1007/s00521-012-0945-y. 34. Villaquira´n-Caicedo, M. A.; Mejı´a De Gutie´rrez, R.; Sulekar, S.; Davis, C.; Nino, J. C. Thermal Properties of Novel Binary Geopolymers based on Metakaolin and Alternative Silica Sources. Appl. Clay Sci. 2015, 118, 276282. Available from: http://dx.doi.org/ 10.1016/j.clay.2015.10.005. 35. Rattanasak, U.; Chindaprasirt, P.; Suwanvitaya, P. Development of High Volume Rice Husk Ash Alumina Silicate Composites. Int. J. Miner. Metall. Mater. 2010, 17, 654660. Available from: http://dx.doi.org/10.1007/s.12613-010-0370-0. 36. Tippayasam, C.; Keawpapasson, P.; Thavorniti, P.; Panyathanmaporn, T.; Leonelli, C.; Chaysuwand, D. Effect of Thai Kaolin on Properties of Agricultural Ash Blended Geopolymers. Constr. Build. Mater. 2014, 53, 455459. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2013.11.079. 37. He, J.; Jie, Y.; Zhang, J.; Yu, Y.; Zhang, G. Synthesis and Characterization of Red Mud and Rice Husk Ash-based Geopolymer Composites. Cement Concrete Compos. 2013, 37, 108118. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2012.11.010. 38. Pimraksa, K.; Chindaprasirt, P.; Rungchet, A.; Sagoe-Crentsil, K.; Sato, T. Lightweight Geopolymer Made of Highly Porous Siliceous Materials with Various Na2O/Al2O3 and SiO2/Al2O3 Ratios. Mater. Sci. Eng. A 2011, 528, 66166623. Available from: http:// dx.doi.org/10.1016/j.msea.2011.04.044. 39. Cordeiro, G. C.; Toledo Filho, R. D.; Fairbairn, E. M. R. Effect of Calcination Temperature on the Pozzolanic Activity of Sugar Cane Bagasse Ash.. Constr. Build. Mater. 2009, 23 (10), 33013303. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2009.02.013. 40. Castaldelli, V. N.; Akasaki, J. L.; Melges, J. L. P.; Tashima, M. M.; Soriano, L.; Borrachero, M. V., et al. Use of Slag/Sugar Cane Bagasse Ash (SCBA) Blends in the Production of Alkali-activated Materials. Materials 2013, 6, 31083127. Available from: http://dx.doi.org/10.3390/ma6083108. 41. Amin, N.; Faisal, M.; Muhammad, K.; Amin, W. Geopolymerization with Bagasse Bottom Ash and China Clay, Effect of Calcination Temperature and Silica to Alumina Ratio. RSC Adv. 2015, 5, 6781467819. Available from: http://dx.doi.org/10.1039/ c5ra04525h.

New inorganic binders containing ashes from agricultural wastes

159

42. Pereira, A.; Akasaki, J. L.; Melges, J. L. P.; Tashima, M. M.; Soriano, L.; Borrachero, M. V., et al. Mechanical and Durability Properties of Alkali-activated Mortar based on Sugarcane Bagasse Ash and Blast Furnace Slag. Ceramics Int. 2015, 41, 1301213024. Available from: http://dx.doi.org/10.1016/j.ceramint.2015.07.001. 43. Castaldelli, V. N.; Moraes, J. C. B.; Akasaki, J. L.; Melges, J. L. P.; Monzo´, J.; Borrachero, M. V., et al. Study of the Binary System Fly Ash/Sugarcane Bagasse Ash (FA/SCBA) in SiO2/K2O Alkali-activated Binders. Fuel 2016, 174, 307316. Available from: http://dx.doi.org/10.1016/j.fuel.2016.02.020. 44. Moraes, J. C. B.; Queiroz, D. C. A.; Akasaki, J. L.; Melges, J. L. P.; Borrachero, M. V.; Paya´, J., et al. Possibilities of Reusing Sugar Cane Straw Ash in the Production of Alternative Binders. Key Eng. Mater. 2015, 668, 304311. Available from: http://dx. doi.org/10.4028/www.scientific.net/KEM.668.304. 45. Salih, M. A.; Farzadnia, N.; Ali, A. A. A.; Demirboga, R. Development of High Strength Alkali Activated Binder Using Palm Oil Fuel Ash and GGBS at Ambient Temperature. Constr. Build. Mater. 2015, 93, 289300. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2015.05.119. 46. Salih, M. A.; Ali, A. A. A.; Farzadnia, N. Characterization of Mechanical and Microstructural Properties of Palm Oil Ash Geopolymer Cement Paste. Constr. Build. Mater. 2014, 65, 592603. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2014.05.031. 47. Zarina, Y.; Mustafa Al Bakri, A. M.; Karamudin, H.; Khairul Nizar, I.; Rafiza, A. R.; Andrei, V. S. Effect of Solids to Liquids, Na2SiO3 to NaOH and Curing Temperature on the Palm Oil Boiler Ash (Si + Ca) Geopolymerisation System. Materials 2015, 8, 22272242. Available from: http://dx.doi.org/10.3390/ma8052227. 48. Mijarsh, M. J. A.; Megat Johari, M. A.; Ahmad, Z. A. Synthesis of Geopolymer from Large Amount of Treated Palm Oil Fuel Ash: Application of the Taguchi Method in Investigating the Main Parameters Affecting Compressive Strength. Constr. Build. Mater. 2014, 52, 473481. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.11.039. 49. Mijarsh, M. J. A.; Megat Johari, M. A.; Ahmad, Z. A. Compressive Strength of Treated Palm Oil Fuel Ash Based Geopolymer Mortar Containing Calcium Hydroxide, Aluminium Hydroxide and Silica Fume as Mineral Additives. Cement Concrete Compos. 2015, 60, 6581. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2015.02.007. 50. Mijarsh, M. J. A.; Megat Johari, M. A.; Ahmad, Z. A. Effect of Delay Time and Na2SiO3 Concentration on Compressive Strength Development of Geopolymer Mortar Synthesized from TPOFA. Constr. Build. Mater. 2015, 86, 6474. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.03.078. 51. Hawa, A.; Tonnayopas, D.; Prachasaree, W. Performance Evaluation and Microstructure Characterization of Metakaolin Based Geopolymer Containing Oil Palm Ash. Scientific World J. 2013, 2013, 19. Available from: http://dx.doi.org/10.1155/2013/857586. 52. Hawa, A.; Tonnayopas, D.; Prachasaree, W.; Taneerananon, P. Investigating the Effects of Oil Palm Ash in Metakaolin Based Geopolymer. Ceramics-Silika´ty 2013, 57 (4), 319327. 53. Hawa, A.; Tonnayopas, D.; Prachasaree, W.; Taneerananon, P. Development and Performance Evaluation of Very High Early Strength Geopolymer for Rapid Road Repair. Adv. Mater. Sci. Eng., 2013, 2013, 19. Available from: http://dx.doi.org/ 10.1155/2013/764180. 54. Hawa, A.; Tonnayopas, D.; Prachasaree, W. Performance Evaluation of Metakaolin based Geopolymer Containing Parawood Ash and Oil Palm Ash Blends. Mater. Sci. 2014, 20 (3), 339343. Available from: http://dx.doi.org/10.5755/j01.ms.20.3.4543.

160 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

55. Ismail, M.; Yusuf, T. O.; Noruzman, A. H.; Hassan, I. O. Early Strength Characteristics of Palm Oil Fuel Ash and Metakaolin Blended Geopolymer Mortar. Adv. Mater. Res. 2013, 690-693, 10451048. Available from: http://dx.doi.org/10.4028/www.scientific. net/AMR.690-693.1045. 56. Kabir, S. M. A.; Alengaram, U. J.; Jumaat, M. Z.; Sharmin, A.; Islam, A. Influence of Molarity and Chemical Composition on the Development of Compressive Strength in POFA based Geopolymer Mortar. Adv. Mater. Sci. Eng., 2015, 2015, 115. Available from: http://dx.doi.org/10.1155/2015/647071. 57. Karim, M. R.; Zain, M. F. M.; Jamil, M.; Lai, F. C. Fabrication of a Non-cement Binder Using Slag, Palm Oil Fuel Ash and Rice Husk Ash with Sodium Hydroxide. Constr. Build. Mater. 2013, 49, 894902. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2013.08.077. 58. Yusuf, M. O.; Johari, M. A. M.; Ahmad, Z. A.; Maleshuddin, M. Evaluation of Alkaline Activated Ground Blast Furnace Slag-ultrafine Palm Oil Fuel Ash based Concrete. Mater. Design 2014, 55, 387393. Available from: http://dx.doi.org/10.1016/j. matdes.2013.09.047. 59. Yusuf, M. O.; Johari, M. A. M.; Ahmad, Z. A.; Maleshuddin, M. Effects of H2O/Na2O Molar Ratio on the Strength of Alkaline Activated Ground Blast Furnace Slag-ultrafine Palm Oil Fuel Ash based Concrete. Mater. Design 2014, 56, 158164. Available from: http://dx.doi.org/10.1016/j.matdes.2013.09.078. 60. Yusuf, M. O.; Johari, M. A. M.; Ahmad, Z. A.; Maleshuddin, M. Influence of Curing Methods and Concentration of NaOH on Strength of the Synthesized Alkaline Activated Ground Slag-ultrafine Palm Oil Fuel Ash Mortar/Concrete. Constr. Build. Mater. 2014, 66, 541548. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.05.037. 61. Yusuf, M. O. Performance of Slag Blended Alkaline Activated Palm Oil Fuel Ash Mortar in Sulfate Environments. Constr. Build. Mater. 2015, 98, 417424. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.05.012. 62. Yusuf, M. O.; Johari, M. A. M.; Ahmad, Z. A.; Maleshuddin, M. Evaluation of SlagBlended Alkaline-activated Palm Oil Fuel Ash Mortar Exposed to the Sulphuric Acid Environment. J. Mater. Civil Eng. 2015, 27 (12), 04015058. Available from: http://dx. doi.org/10.1061/(ASCE)MT.1943-5533.0001315. 63. Ariffin, M. A. A.; Hussin, M. W.; Bhutta, A. R. Mix Design and Compressive Strength of Geopolymer Concrete Containing Blended Ash from Agro-Industrial Wastes. Adv. Mater. Res. 2011, 339, 452457. Available from: http://dx.doi.org/10.4028/www.scientific.net/AMR.339.452. 64. Ariffin, M. A. M.; Bhutta, M. A. R.; Hussin, M. W.; Mohd Tahir, M.; Aziah, N. Sulfuric Acid Resistance of Blended Ash Geopolymer Concrete. Constr. Build. Mater. 2013, 43, 8086. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.01.018. 65. Bhutta, M. A. R.; Hussin, W. M.; Azreen, M.; Mohd Thair, M. Sulphate Resistance of Geopolymer Concrete Prepared from Blended Waste Fuel Ash. J. Mater. Civil Eng. 2014, 26 (11), 04014080. Available from: http://dx.doi.org/10.1061/(ASCE)MT.19435533.0001030. 66. Liu, M. Y. J.; Chua, C. P.; Alengaram, U. J.; Jumaat, M. Z. Utilization of Palm Oil Fuel Ash as Binder in Lightweight Oil Palm Shell Geopolymer Concrete. Adv. Mater. Sci. Eng. 2015, 2014, 16. Available from: http://dx.doi.org/10.1155/2014/610274. 67. Liu, M. Y. J.; Alengaram, U. J.; Jumaat, M. Z.; Mo, K. H. Evaluation of Thermal Conductivity, Mechanical and Transport Properties of Lightweight Aggregate Foamed Geopolymer Concrete. Energy Build. 2014, 7, 238245. Available from: http://dx.doi. org/10.1016/j.enbuild.2013.12.029.

New inorganic binders containing ashes from agricultural wastes

161

68. Ranjbar, N.; Mehrali, M.; Behnia, A.; Alengaram, U. J. Compressive Strength and Microstructural Analysis of Fly Ash/Palm Oil Fuel Ash based Geopolymer Mortar. Mater. Design 2014, 59, 532539. Available from: http://dx.doi.org/10.1016/j.matdes.2014.03.037. 69. Ranjbar, N.; Mehrali, M.; Alengaram, U. J.; Metselaar, H. S. C.; Jumaat, M. Z. Compressive Strength and Microstructural Analysis of Fly Ash/Palm Oil Fuel Ash based Geopolymer Mortar Under Elevated Temperatures. Constr. Build. Mater. 2014, 65, 114121. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.04.064. 70. Bashar, I. I.; Alengaram, U. J.; Jumaat, M. Z.; Islam, A. The Effect of Variation of Molarity of Alkali Activator and Fine Aggregate Content on the Compressive Strength of the Fly Ash. Adv. Mater. Sci. Eng. 2014, 2014, 113. Available from: http://dx.doi. org/10.1155/2014/245473. 71. Islam, A.; Alengaram, U. J.; Jumaat, M. Z.; Bashar, I. I. The Development of Compressive Strength of Ground Granulated Blast Furnace Slag-Palm Oil Fuel Ash based Geopolymer Mortar. Mater. Design 2014, 56, 833841. Available from: http://dx. doi.org/10.1016/j.matdes.2013.11.080. 72. Hussin, M. W.; Bhutta, M. A. R.; Azreen, M.; Ramadhansyah, P. J.; Mirza, J. Performance of Blended Ash Geopolymer Concrete at Elevated Temperatures. Mater. Struct. 2015, 48, 709720. Available from: http://dx.doi.org/10.1617/s11527-014-0251-5. 73. Rajamma, R.; Labrincha, J. A.; Ferreira, V. Alkali activation of biomass fly ashmetakaolin blends. Fuel 2012, 98, 265271. Available from: http://dx.doi.org/10.1016/ j.fuel.2012.04.006. 74. Ban, C. C.; Ken, P. W.; Ramli, M. The Hybridizations of Coal Fly Ash and Wood Ash for the Fabrication of Low Alkalinity Geopolymer Load Bearing Block Cured at Ambient Temperature. Constr. Build. Mater. 2015, 88, 4155. Available from: http:// dx.doi.org/10.1016/j.conbuildmat.2015.04.020. 75. Brundlandt, G. H. Our Common Future, World Commission on Environment and Development; Oxford University Press: New York, 1987. 76. Paya´, J. La transmutacio´n sostenible de los residuos para nuevas materias primas en el a´mbito del concreto. Dyna 2012, 3847. 77. Madurwar, M. V.; Ralegaonkar, R. V.; Mandavgane, S. A. Application of Agro-waste for Sustainable Construction Materials: A Review. Constr. Build. Mater. 2013, 872878. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2012.09.011. 78. Parianti, E.; Shafigh, P.; Bahri, S.; Farahani, J. N. Supplementary Cementitious Materials Origin from Agricultural Wastes - A Review. Constr. Build. Mater. 2015, 176187. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.10.010. 79. Alejandre Sanchez, F. J. In La cal: Investigacio´n, patrimonio y restauracio´n; Sevilla, Universidad, Ed.; Secretariado de publicaciones, Sevilla: Spain, 2014 (ISBN: 978-8447215-07-2). 80. Matteini, M.; Moles, A. La quı´mica en la restauracio´n Ed. Nerea, Donostia, Spain, 2001. (ISBN:978-84-89569-54-6) 81. Cook, D. J.; Pama, R. P.; Paul, B. K. Rice Husk Ash-lime Cement Mixes for Use in Masonry Units. Build. Environ. 1977, 12, 18. Available from: http://dx.doi.org/ 10.1016/0360-1323(77)90031-2. 82. Metha, P.K. The Chemistry and Technology of Cements Made from Rice Husk Ash. In: Proceedings of a Joint Workshop Held In Peshawar, Pakistan; Regional center for Technology Transfer, 1979. 83. Boateng, A. A.; Skeete, D. A. Incineration of Rice Hull for Use as a Cementitious Material: The Guyana Experience. Cement Concrete Res. 1992, 20, 795802. Available from: http://dx.doi.org/10.1016/0008-8846(90)90013-N.

162 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

84. Billong, N.; Melo, U. C.; Kamseu, E.; Kinuthia, J. M.; Njopwouo, D. Improving Hydraulic Properties of Limerice Husk Ash (RHA) Binders with Metakaolin (MK). Constr. Build. Mater. 2011, 2, 1572161. Available from: http://dx.doi.org/10.1016/ j.conbuildmat.2010.11.013. 85. Walker, R.; Pavia, S. Physical Properties and Reactivity of Pozzolans and Their Influence on the Properties of Lime-pozzolans Pastes. Mater. Struct. 2011, 44, 11391150. Available from: http://dx.doi.org/10.1617/s11527-010-9689-2. 86. Souza, L. M. S.; Fairbairn, E. M. R.; Toledo Filho, R. D.; Cordeiro, G. C. Influence of Initial CaO/SiO2 Ratio on the Hydration of Rice Husk Ash-Ca(OH)2 and Sugar Cane Bagasse Ash-Ca(OH)2 Pastes. Quimica Nova 2014, 37, 16001605. Available from: http://dx.doi.org/10.5935/0100-4042.20140258. 87. Cizer, O.; Van Balen, K.; VAN Gemert, D.; Elsen, J. Carbonation and Hydration of Mortars with Calcium Hydroxide and Calcium Silicate Binders. In Sustainable Construction Materials and Technologies; Chun, Claisse, Naik, Ganjian, Eds.; Taylor & Francis: London, 2007; pp 611621. 88. Cizer, O.; Van Balen, K.; Van Gemert, D.; Elsen, J. Hardening of Mortars Made from Cement, Rice Husk Ash and Lime. Proc. Inst. Civ. Eng. Constr. Mater. 2009, 162, 1927. Available from: http://dx.doi.org/10.1680/coma.2009.162.1.19. 89. Pavia, S.; Walker, R.; Veale, P.; Wood, A. Impact of the Properties and Reactivity of Rice Husk Ash on Lime Mortar Properties. J. Mater. Civ. Eng. 2014, 26, 04014066. doi:10.1061/(ASCE)MT.1943-5533.0000967. 90. Nair, D. G.; Jagadish, K. S.; Fraaij, A. Reactive Pozzolanas from Rice Husk Ash: An Alternative to Cement for Rural Housing. Cement Concrete Res. 2006, 36, 10621071. Available from: http://dx.doi.org/10.1016/j.cemconres.2006.03.012. 91. Billong, N.; Melo, U. C.; Kamseu, E.; Kinuthia, J. M.; Njopwouo, D. Improving Hydraulic Properties of Lime-rice Husk Ash (RHA) Binders with Metakaolin (MK).. Constr. Build. Mater. 2011, 25, 21572161. Available from: http://dx.doi.org/10.1016/j. conbuildmat.20110.11.013. 92. Mendez, R.; Borrachero, M. V.; Paya´, J.; Monzo´, J. Mechanical Strength of Lime-rice Husk Ash Mortars: A Preliminary Study. Key Eng. Mater. 2012, 517, 495499. Available from: http://dx.doi.org/10.4028/www.scientific.net/KEM.517.495. 93. Stroeven, P.; Bui, D. D.; Sabuni, E. Ash of Vegetable Waste used for Economic Production of Low to High Strength Hydraulic Binders. Fuel 1999, 78, 153159. doi:10.1016/S0016-2361(98)00143-4. 94. Koteswara, R. D.; Pranav, P. R. T.; Anusha, M. Stabilization of Expansive Soil with Rice Husk Ash, Lime and Gypsum: An Experimental Study. Int. J. Eng. Sci. Technol. 2011, 3, 80768085. 95. Koteswara, R. D.; Rameswara, G. V. V.; Pranav, P. R. T. A Laboratory Study on the Effect of Rice Husk Ash & Lime on the Properties of Marine Clay. Int. J. Eng. Innov. Technol. 2012, 2 (1), 345353. 96. Shrivastava, D.; Singha, I. A. K.; Yadav, R. K. Effect of Lime and Rice Husk Ash on Engineering Properties of Black Cotton Soil. Int. J. Eng. Sci. Technol. 2014, 3 (2), 292296 http://www.ijerst.com/ijerstadmin/upload/IJEETC_537b21ef09451.pdf. 97. Behak, L.; Nun˜ez, W. P. Effect of Burning Temperature on Alkaline Reactivity of Rice Husk Ash with Lime. Road Mater. Pavement Des 2013, 14 (3), 570585. Available from: http://dx.doi.org/10.1080/14680629.2013.779305. 98. Tashima, M. M.; Soriano, L.; Monzo´, J.; Borrachero, M. V.; Akasaki, J. L.; Paya´, J. New Method to Assess the Pozzolanic Reactivity of Mineral Admixtures by Means of pH and Electrical Conductivity Measurements in Lime:Pozzolan Suspension. Materiales de Construccio´n. 2014, 64 (316). Available from: http://dx.doi.org/10.3989/mc.2014.00914.

New inorganic binders containing ashes from agricultural wastes

163

99. Viet-Thien-An, V.; Robler, C.; Danh-Dai, B.; Horst-Michel, L. Pozzolanic Reactivity of Mesoporous Amorphous Rice Husk Ash in Porlandite Solution. Constr. Build. Mater. 2014, 59, 111119. Available from: http://dx.doi.org/10.1016/j.conbuildmat. 2014.02.046. 100. Osinubi, K. J.; Ijimdiya, T. S.; Nmadu, I. Lime Stabilization of Black Cotton Soil Using Bagasse Ash as Admixture. Adv. Mater. Res. 2009, 62-64, 310. Available from: http://dx.doi.org/10.4028/www.scientific.net/AMR.62-64.3. 101. Madurwar, M. V.; Mandavgane, S. A.; Ralegaonkar, R. V. Development and Feasibility Analysis of Bagasse Ash Bricks. J. Energy Eng. 2015, 141 (3), 19. Available from: http://dx.doi.org/10.1061/(ASCE)EY.1943-7897.0000200. 102. Alave´z-Ramı´rez, R.; Montes-Garcı´a, P.; Mirtı´nez-Reyes, J.; Altamirano-Jua´rez, D. C.; Gochi-Ponce, Y. The Use of Sugarcane Bagasse Ash and Lime to Improve the Durability and Mechanical Properties of Compacted Soil Blocks. Constr. Build. Mater. 2012, 34, 296305. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2012.02.072. 103. Cordeiro, G. C.; Toledo Filho, R. D.; Tavares, L. M.; Fairbairn, E. M. R. Pozzolanic Activity and Filler Effect of Sugar Cane Bagasse Ash in Portland Cement and Lime Mortars. Cement Concrete Compos. 2008, 30, 410418. Available from: http://dx.doi. org/10.1016/j.cemconcomp.2008.01.001. 104. Martirena-Herna´ndez, J. F.; Middendorf, B.; Gehrke, M.; Budelmann, H. Use of Wastes of Sugar Industry as Pozzolan in Lime-pozzolana Binders: Study of the Reaction. Cement Concrete Res. 1998, 28, 15251536. Available from: http://dx.doi.org/10.1016/ S0008-8846(98)00130-6. 105. Martirena-Herna´ndez, J. F.; Betancourt-Rodrı´guez, S.; Middendorf, B.; Rubio, A.; Martı´nez-Ferna´ndez, L.; Machado-Lo´pez, I., et al. Pozzolanic Properties of Residues of Sugar Industries (first part). Materiales de Construccio´n 2000, 50, 7178. Available from: http://dx.doi.org/10.3989/mc.2000.v50.i260.392. 106. Martirena, F.; Middendorf, B.; Day, R. L.; Gehrke, M.; Roque, P.; Martı´nez, L., et al. Rudimentary, Low Tech Incinerators as a Means to Produce Reactive Pozzolan Out of Sugar Cane Straw. Cement Concrete Res. 2006, 36, 10561061. Available from: http://dx.doi.org/10.1016/j.cemconres.2006.03.016. 107. Frı´as, M.; Villar-Cocin˜a, E.; Valencia-Morales, E. Characterisation of Sugar Cane Straw as Pozzolanic Material for Construction: Calcining Temperature and Kinetic Parameters. Waste Manage. 2006, 27, 533538. Available from: http://dx.doi.org/ 10.1016/j.wasman.2006.02.017. 108. Frı´as, M.; Villar-Cocin˜a, E. Influence of Calcining Temperature on the Activation of Sugar-Cane Bagasse: Kinetic Parameters. Adv. Cement Rese. 2007, 19, 109115. Available from: http://dx.doi.org/10.1680/adcr.2007.19.3.109. 109. Villar-Cocin˜a, E.; Valencia-Morales, E.; Gonza´lez-Rodrı´guez, R.; Herna´ndez-Ruı´z, E. Kinetics of the Pozzolanic Reaction between Lime and Sugar Cane Straw Ash by Electrical Conductivity Measurement: A Kinetic-Diffusive Model. Cement Concrete Res. 2003, 33, 517524. Available from: http://dx.doi.org/10.1016/S0008-8846(02)00998-5. 110. Villar-Cocin˜a, E.; Frı´as, M.; Valencia-Morales, E. Sugar Cane Wastes as Pozzolanic Materials: Application of Mathematical Model. ACI Mater. J. 2008, 105 (3), 258264. Available from: http://dx.doi.org/10.14359/19822. 111. Nakanishi, E. Y.; Frı´as, M.; Martı´nez-Ramı´rez, S.; Santos, S. F.; Rodrigues, M. S.; Rodrı´guez, O., et al. Characterization and Properties of Elephant Grass Ashes as Supplementary Cementing Material in Pozzolan/Ca(OH)2 Pastes. Constr. Build. Mater. 2014, 73, 391398. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.09.078. 112. Cobreros, C.; Reyes-Araiza, J. L.; Manzano-Ramirez, A.; Nava, R.; Rodriguez, M.; Mondrago´n-Figueroa, M., et al. Barley Straw Ash: Pozzolanic Activity and Comparison

164 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

with Other Natural and Artificial Pozzolans from Me´xico. Bioresources 2015, 10, 37573774. Available from: http://dx.doi.org/10.15376/biores.10.2.3757-3774. 113. Wang, S. Compressive Strengths of Mortar Cubes from Hydrated Lime with Cofired Biomass Fly Ashes. Constr. Build. Mater. 2014, 50, 414420. Available from: http:// dx.doi.org/10.1016/j.conbuildmat.2013.09.045. 114. Wang, S. Quantitative Kinetics of Pozzolanic Reactions in Coal/Cofired Biomass Fly Ashes and Calcium Hydroxide (CH) Mortars. Constr. Build. Mater. 2014, 51, 364371. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.10.057. 115. Carrasco-Hurtado, B.; Corpas-Iglesias, F. A.; Cruz-Perez, N.; Terrados-Cepeda, J.; Perez-Villarejo, L. Addition of Bottom Ash from Biomass in Calcium Silicate Masonry Units for Use as Construction Material with Thermal Insulating Properties. Constr. Build. Mater. 2014, 52, 155165. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2013.11.018.

New trends for nonconventional cement-based materials: industrial and agricultural waste

6

Moise´s Frı´as-Rojas1, Maria Isabel Sa´nchez-de-Rojas-Go´mez1, Ce´sar Medina-Martı´nez2 and Ernesto Villar-Cocin˜a3 1 Eduardo Torroja Institute for Construction Science (IETcc-CSIC), Madrid, Spain, 2 University of Extremadura, Ca´ceres, Spain, 3Central University of Las Villas, Santa Clara, Cuba

6.1

Introduction

One of the problems facing 21st century society in the wake of worldwide population growth is the generation of vast amounts of different kinds of industrial waste. Stockpiling or incinerating this waste has highly adverse effects on the environment and public health and poses technical issues as well. In contrast, its reuse as raw materials for certain industries can carry significant added value in addition to protecting natural spaces, conserving nonrenewable natural mineral resources, and reducing the emission of climate change-inducing greenhouse gases. Cement manufacture for construction consumes large amounts of raw materials (limestone and clay) and accounts for 5.0% 7.0% of all the CO2 released into the atmosphere. The cement industry also reuses large amounts of industrial waste in different stages of production, however, primarily as active additions to Portland clinker. Commercial cement manufacturing standards in place the world over since the second half of the 20th century envisage the use of active additions such as silica fume, fly ash, blast furnace slag, metakaolin, and calcined natural pozzolans to improve the performance of blended cements. Nonetheless, the use of these additions may be significantly conditioned by availability shortfalls, low production, environmental standards, and shipping costs. For that reason, the scientific community has turned its gaze in recent years to new types of recyclable industrial waste and by-products to generate more eco-efficient pozzolans than those conventionally used. Europe’s 20201 strategy envisages the future implementation of models based on highly efficient material and energy resources, particularly for the manufacture of cement binders, the materials most abundantly used in construction. For several years, the Material Recycling research team working out of Spain’s Eduardo Torroja Institute (IETcc, a National Research Council [CSIC] body) has been analyzing many types of waste that vary in origin and in their organic or inorganic nature. This chapter describes both lines of work. The focus in inorganic Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00007-3 © 2017 Elsevier Ltd. All rights reserved.

166 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

waste is on spent fluid catalytic cracking (FCC) catalysts, fired clay (CC), construction and demolition waste (C&DW) refuse, and coal mining waste (CW); and in organic waste, on agroindustrial refuse, which must be thermally treated to acquire pozzolanic properties. The origin of the starting waste and the activation conditions have a significant impact on the properties of the ash obtained. To substantiate that assertion, this chapter compares the performance of sugar cane bagasse activated under optimal conditions in the laboratory and on an industrial scale for use as biomass fuel to obtain cleaner and cheaper energy than afforded by fossil fuels such as coke and coal.2,3 This wide variety of industrial waste, for its pozzolanic properties can have a direct application in improving the performances of the natural fiber-reinforced composites, allowing a decrease in the alkalinity of the cement matrix, a greater adherence of theinterfacial transition zone (ITZ) cement/fiber, and a high durability of the fibers by reducing its mineralization in highly alkaline media.4

6.2

Waste origin and characterization

6.2.1 Inorganic waste Whilst the generation of organic and inorganic industrial waste and by-products poses problems for industrialized societies, technological progress is rendering industrial processes ever more efficient, sustainable, and eco-friendly. The use of waste in different stages of production is standard practice in construction, for instance. The active additions approved for cement manufacture, according to European standard EN 197-1,5 include fly ash and silica fume. Nonetheless, as these by now conventional pozzolans will be in short supply in the future, alternative materials must be sought. That has opened up new lines of research. This section addresses research on four of the many materials that could initially be used for cement manufacture: (1) spent FCC catalysts; (2) clay waste (CC and C&DW); and (3) CW. 1. Cracking has become a key procedure in refineries to enhance gasoline production at the expense of heavier and less valuable products, such as kerosene and fuel oil. In FCC a fine, powdery catalyst is carried upward by a hot stream of oil vapor to a reactor where the reaction takes place. The burnt powder is sedimented out as waste and the vapors are fractionated to separate gasoline from other products. The properties of this waste make it usable for cement manufacture.6 10 A spent fluid catalytic cracking (FCC) furnished by REPSOL was used in the study described hereunder (Fig. 6.1A). 2. Fired clay waste (CC) was also analyzed. European standard EN 197-14 on the manufacture of ordinary cement defines natural calcined pozzolan (CEM II/A and B-Q) as “materials of volcanic origin, clays, shales or sedimentary rocks activated by thermal treatment.” As a rule, clay minerals are highly pozzolanic when heated at 600 1000 C. The loss of chemically combined water during calcination destroys the crystalline network of the clay constituents, rendering their components amorphous or vitreous. Such thermodynamic instability is one of the major factors governing pozzolanic behavior.11 Kaolinite

New trends for nonconventional cement-based materials: industrial and agricultural waste

167

Figure 6.1 (A): (FCC) catalyst; (B): fired clay waste (discards (CC) and C&DW); (C): coal mining waste (CW). or montmorillonite-like clay minerals, or a combination of the two, acquire pozzolanicity under controlled heating at temperatures of 540 980 C. Illite-like clays and shales with large proportions of vermiculite, chlorite, and mica must be fired at higher temperatures to be activated.12 In Spain, in the wake of the economic crisis, the manufacture of fired clay materials for construction has declined drastically. Nonetheless, the percentage of market rejects (CC) continues to be high, while clay-based materials account for 54% of the 40 million tonnes of C&DW generated in the country yearly. While the polluting power of this inert waste is low, it poses a severe environmental problem because its stockpiling mars the landscape (Fig. 6.1B). The authors have observed the presence of pozzolanicity in fired clay discards,13 15 noting that hydrated products similar to those obtained with other pozzolanic materials are formed in the respective pozzolanic reactions. Toledo et al.16 studied the reuse of fired clay industry waste from plants in Brazil with satisfactory results using cement replacement ratios of 10% and 20%. Velosa et al.17 reported that expanded clay waste was apt for use as a pozzolanic addition in concrete with a hydraulic lime binder. Further to research on C&DW18,19 proving that this waste is active, its authors were awarded a Spanish patent for blended cement manufacture.20 3. The third type of inorganic waste analyzed was CW from an open-pit mine furnished by Hullera Vasco-Leonesa, Sociedad Limitada, located in the Spanish province of Leon. This waste was thermally activated in a laboratory muffle furnace at 600 C for 2 h to obtain a product (Fig. 6.1C) subsequently analyzed in a number of studies.21 27

Sample fineness plays a significant role in pozzolanic activity, particularly at the earliest stages. The spent FCC catalyst had a Blaine specific surface of 5900 cm2/g, determined as recommended in European standard EN 196-6.28 After crushing, the fired clay materials were ground in a ball grinder to specific surfaces of 3500 and 4700 cm2/g for CC and C&DW, respectively. The burnt CW was crushed and ground to a particle size of under 90 μm. The chemical composition of the materials determined with X-ray fluorescence (XRF) is given in Table 6.1. The FCC catalyst had a high silica and alumina content, with small percentages of iron, alkalis, and magnesium. No calcium or sulfur was detected, and its loss on ignition (LOI) was attributed to the residual coal. The chemical composition of the brick and roof tile discards was similar to the composition of other pozzolanic materials, i.e., it was highly acidic with a predominance of silica, alumina, and iron oxide and a low SO3 content.

168 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Chemical composition of inorganic waste (determined with XRF)

Table 6.1

Materials

CaO

SiO2

Al2O3

MgO

Fe2O3

SO3

Na2O

K2O

LOI

0.79 2.70 0.27

0.0 0.58 0.76 0.17

0.17 3.10 1.99 3.09

4.10 3.44 12.07 3.19

(wt%) FCC CC C&DW CW

12.41 20.66 4.20

42.12 52.88 43.04 56.63

51.27 16.80 11.54 25.29

0.99 3.64 2.51 0.77

0.37 5.29 4.34 4.64

Figure 6.2 XRD spectra for inorganic waste.

The chemical composition of the C&DW was as expected for waste management plant material with over 20% fired clay. The majority components were SiO2, Al2O3, and Fe2O3, along with up to 20.66% CaO and 2.70% SO3. The low sulfate content ruled out the presence of significant amounts of plaster. High LOI was attributable to the calcite in the sample, identified by X-ray diffraction pattern. The CW sample also had high SiO2 and Al2O3 contents but less than 5% CaO. X-ray diffraction (XRD) analysis showed that the catalyst (Fig. 6.2) had low crystallinity, with diffraction lines attributed to a faujasite family-like hydrogen aluminosilicate (HAS) pertaining to the tectosilicate class of zeolites. All the fired clay refuse studied exhibited a similar mineralogical composition, with quartz (Q), illite (I), calcite (C), dolomite (D), hematite (H), orthoclase (O), and anorthite (An) as the main crystalline components. No evidence of gypsum was found (Fig. 6.2). The coal tailings contained primarily quartz (Q), micas (M), and feldspars (F). Scanning electron microscope (SEM) studies showed that the catalyst consisted of spherical particles (Fig. 6.3) smaller than 30 μm, although other irregularly shaped particles, perhaps fragments of broken spheres, were also present. Further to microanalysis, the spheres comprised primarily Si and Al, with an Si/Al ratio, at 0.80, consistent with the XRF findings. The particles in the crushed and ground fired clay material (CC) were irregular and differed in size. The chemical composition of this waste included aluminum,

New trends for nonconventional cement-based materials: industrial and agricultural waste

169

Figure 6.3 SEM micrographs showing the morphology of: (A) FCC; (B) CC; (C) C&DW; and (D) CW.

potassium, and a small percentage of iron, denoting the presence of feldspars (Table 6.1). Quartz-high particles in the samples were also observed. SEM identified spongy metakaolin (MK) clusters and more compact quartz clusters in the thermally treated coal tailings. Mg, K, Ca, Fe, and S deposits were observed in varying proportions on the surface of both types of clusters.

6.2.2 Organic waste Agroindustrial waste has become the object of research around the world in the last 10 15 years in light of its industrial applicability, which reduces the adverse effects of stockpiling or uncontrolled incineration.29 32 The literature describes a wide spectrum of inert agroindustrial waste that is directly usable in the cement industry as supplementary cementitious materials, given its (postheating) high reactive silica content.33 41 All these authors reported that the high pozzolanic activity in agroindustrial ash depends directly on the nature and origin of the waste as well as on the activation conditions. The research discussed hereunder focused on four types of agroindustrial wastes, differing in nature and activating conditions: (1) Brazilian bagasse activated thermally on a laboratory scale (LBA); (2) industrial ash from a sugar processing plant at Pirassununga, Sa˜o Paulo, Brazil, that uses 100% of a variable mix of bagasse and sugar cane leaves (IBA) as biomass fuel, with ratios of up to 60/40 depending on availability; (3) Brazilian bamboo leaf ash, activated on a laboratory scale (BA); and (4) industrial ash from a Spanish plant that uses a 3:1 (wt%) mix of straw and forestry waste (SFA). The laboratory scale ash was heated to 600 C for

170 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

2 h in an electrical furnace, while the industrial IBA was heated to around 800 C in a boiler for a few minutes and the SFA to 1200 C for 15 20 min. As Fig. 6.4 shows, after milling and sieving to a particle size of under 90 μm, the various types of ash exhibited a range of colours: red in LBA, ochre in IBA, white in BA, and greyish black in SFA. The chemical composition of the ash as determined with XRF is given in Table 6.2. The ash from sugar cane waste contained primarily silica and alumina, which together accounted for 80% of the total, followed by iron and potassium oxides. Bamboo leaf ash had 77% silica and 8.50% calcium oxide. Lastly, SFA ash had the lowest silica content (49.6%) and 12.3% calcium and 13.3% potassium oxides. Other authors found different values than shown in Table 6.2, particularly in connection with IBA: both higher and lower Al2O3 contents have been reported for industrial bagasse (up to 20.7%),42 and as low as 5.7%,43 while Chuslip et al.43 also observed an Fe2O3 content of 3.54%. The SFA ash contained 0.22 wt% chloride,

Figure 6.4 Ashes analyzed.

Chemical composition of ashes from organic waste determined with XRF

Table 6.2

Oxides

LBA

IBA

BA

SFA

76.93 0.62 0.35 0.19 2.25 8.54 0.11 1.37 4.44 0.04 1.27 3.83 1952

49.64 3.13 2.13 0.33 3.38 12.35 1.87 0.73 13.24 0.20 4.26 8.47 2296

(wt%) SiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O SO3 K2O TiO2 P2O5 LOI Cl (ppm)

69.40 11.26 5.41 0.21 1.28 2.51 0.09 1.83 3.45 1.38 1.61 1.56 1054

60.14 12.53 10.35 0.20 2.10 3.11 0.16 0.11 6.06 2.73 1.47 1.03 151

New trends for nonconventional cement-based materials: industrial and agricultural waste

171

Figure 6.5 XRD patterns for four types of starting ash.

a finding to be borne in mind when manufacturing future blended cements. The replacement ratio of this ash could be conditioned by the # 0.1% chloride ceiling for commercial cements laid down in European standard EN 197-1 2011.5 According to the mineralogical analysis, amorphous material prevailed in all the ashes studied, with some crystalline compounds whose composition varied depending on origin and activation conditions. The amorphous is a broad peak between 2θ angles 15 and 25 degrees on the XRD patterns for the agroindustrial ash reproduced in Fig. 6.5. A review of the diffractograms revealed that the ash obtained in the laboratory (BA and LBA) contained traces of calcite (Ca) and quartz (Q), the main crystalline components, whereas quartz and hematite (He: Fe2O3) were identified in the industrial IBA ash, and quartz, gehlenite (Ge: Ca2MgSi2O7) and sylvite (NaCl) in SFA ash. The presence of these crystalline compounds in industrial ash was attributed to contamination during collection and storage prior to its use as biomass.

6.3

Pozzolanic behavior and modeling in pozzolan/Ca(OH)2 systems

One of the essential factors in the valorization of industrial waste as an active addition in Portland cement is its reactivity with the calcium hydroxide in the medium. To that end, for decades the IETcc-CSIC Material Recycling team has been using an accelerated chemical method13 based on the standardized procedure for determining cement pozzolanicity.44 In the IETcc method, 1 g of material is suspended in 75 mL of lime-saturated solution (instead of the 20 g of blended cement in 100 mL of water specified in the standard). With this approach, the amount of lime fixed over time can be monitored and the kinetic parameters that govern the pozzolanic reaction can be quantified with a diffusive-kinetics model.45,46

172 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

6.3.1 Pozzolanicity of inorganic waste The pozzolanicity findings (Fig. 6.6) showed that reactivity was highest in the FCC catalyst and the burnt coal tailings and lowest in the fired clay waste. Fineness factor is another consideration for, as noted, it plays a significant role at early ages. Due to its lower specific surface, CC reacted more slowly than the other materials. After 7 days, however, the activity observed for fired clay discards was comparable to the findings for FCC. C&DW activity, in turn, grew with its fired clay content, as reported in earlier studies. This refuse behaved like the 100% clay-based CC taken from material discarded by firing plants.

6.3.2 Pozolanicity of organic waste The amount of lime fixed by 1, 3, 7, 28, and 90 day specimens of each type of ash analyzed is shown in Fig. 6.7. In all cases, the ash exhibited high pozzolanicity, for the 90 day samples consistently fixed over 90% of the available lime, values

Figure 6.6 Inorganic waste pozzolanicity versus reaction time.

Figure 6.7 Organic waste pozzolanicity versus reaction time.

New trends for nonconventional cement-based materials: industrial and agricultural waste

173

similar to those observed for highly reactive standard pozzolans such as silica fume and metakaolin.33,47 Such good pozzolanic performance was attributed to the amorphism, low density, and large specific surface of agroindustrial ash. Reaction rate, i.e., the rate at which the lime in the medium is fixed, is instrumental to material pozzolanicity and has a direct impact on the ultimate applicability of the respective blended cement. After a few days of reaction, the ashes studied were ranked by reaction rate as follows: BA . LBA . IBA . SFA. In other words, the ash generated on a laboratory scale had a higher early-age reaction rate than the industrial (boiler) ash. The lower reaction rate in the latter was directly related to the presence of quartz and iron oxides from soils where the crops were planted. Frı´as et al.48 reported that different types of cogeneration ash from a Brazilian sugar plant exhibited significant quartz mineral contamination, with particle sizes of over 1 mm. The concomitant dilution effect had a very adverse impact on ash pozzolanicity. Given the high amorphous content of agroindustrial ash, the primary pozzolanic reaction product is a C-S-H gel. Its CaO/SiO2 ratio of around 1 is much lower than observed in gels generated by cement particles, where values are normally around 2 3.49,50

6.3.3 Modeling the pozzolanic reaction in the pozzolan/Ca(OH)2 system The diffusive kinetic model described by Villar et al.45 was used to calculate the kinetic parameters (Eq. 6.1):     2:65259 Exp 2 3tτ 21 1 Exp τt τ1 29:4732 Exp 2 τt τ1 Ct 5 1 1 Ccorrection D K (6.1) 

where: D is the effective diffusion coefficient K is the reaction rate constant τ is a time constant (the time interval during which the pozzolan radius declines to 37% of its initial radius, rs) Ct is the absolute loss of CH concentration in the pozzolan/lime system over time Ccorrection is a parameter that corrects for the remaining CH concentration not consumed in the reaction (not all the CH is consumed in some systems).

Lime concentration versus time was fitted successively to the kinetic control model (second and third terms), the diffusive control model (first and third terms), and a mixed (kinetic-diffusive) control model (all terms). Of the three, the kinetic control model was found to provide the best fit to the experimental data, based on an exhaustive analysis of the major statistical parameters, including the: correlation coefficient (r), coefficient of multiple determination (R2), standard error (SE), 95% confidence intervals, residual sum of squares (RSS), residual scatter, residual

174 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Reaction rate constant, parameter τ, and statistical parameters for the pozzolans

Table 6.3

WASTE

τ (h)

Reaction rate constant K (h21)

Ccorr

Free energy activation (kJ/mol)

Corr. coeffic. (r)

Coeff. multiple determin. (R2)

C&DW FCC CC CW LBA IBA BA SFA

97.5 6 2.4 55.5 6 1.1 117.4 6 0.9 37.5 6 3.5 23.6 6 0.6 133.9 6 12 7.9 6 0.6 144.3 6 8.6

(6.19 6 0.82).1024 (3.14 6 0.65).1023 (6.76 6 0.54).1024 (6.05 6 0.8).1023 (3.73 6 0.08).1022 (1.07 6 0.10).1023 (1.25 6 0.09).1021 (6.2 6 0.86).1024

3,78 6 0.05 1.95 6 0.18 2.46 6 0.16 3.11 6 0.43 0.21 6 0.06 0.54 6 0.08 1.36 6 0.07 1.58 6 0.09

117.3 6 0.5 113.1 6 0.8 117.1 6 0.2 111.4 6 0.7 106.7 6 0.8 115.9 6 0.5 103.5 6 0.1 117.3 6 0.4

0.9602 0.9308 0.9786 0.9816 0.9886 0.9808 0.9980 0.9912

0.9569 0.9268 0.9713 0.9816 0.9870 0.9765 0.9969 0.9891

probability and analysis of variance (which rigorously assesses the process of fitting the model to the experimental data). Further to the findings in Table 6.3, the reaction rate constant was one order of magnitude higher in inorganic waste FCC and CW (1023) than in fired clay waste C&DW and CC (1024). Of the agroindustrial waste, BA, at 1021  h21, exhibited the highest reactivity, followed by LBA (1022  h21) and IBA (1023  h21). Reactivity was lowest for SFA, with an order of magnitude of 1024  h21. Note the higher pozzolanic reactivity in bamboo leaf and bagasse ash generated under laboratory-controlled conditions than in any of the other types of waste analyzed.48,51,52

6.4

Rheological behavior of blended cement pastes

The blended and control cements exhibited differential rheological behavior, attributed primarily to the nature, chemical composition, and pozzolan content of the two materials. Pursuant to the European legislation in effect,5 rheology is assessed in terms of the water content required for normal consistency (water demand), initial setting time, and volume increases. Binary cement pastes were prepared with each type of ash studied at cement replacement ratios of 10% and 20% and variable water/binder ratios depending on the water required in each case.

6.4.1 Inorganic waste The relative water demand (NCW) studies conducted showed that the cements containing FCC required from 6% to 12% more water for 10% and 20% replacement ratios (Fig. 6.8). This was attributed to the extreme fineness of the catalyst,6 which prompted a higher water demand for a given consistency. CC waste also raised the water demand by amounts that depended on the replacement ratio: at 10% demand rose by 3.6% and at 20%, by 9.6%. Water demand increased with

New trends for nonconventional cement-based materials: industrial and agricultural waste

175

Figure 6.8 Water demand in blended cements (percentage relative to the control).

Figure 6.9 Evolution of initial setting time for blended cements compared to the reference cement.

C&DW as well, although to a lesser extent than with CC. These findings are as expected, the porosity of fired clay is greater than other waste materials, a higher percentage of the former would raise water demand more steeply. This same trend is found in the CW waste, which increase the water demand between 3% and 6.6% compared to the reference paste. Initial setting time was retarded slightly by FCC at both replacement ratios (Fig. 6.9). In contrast, fired clay materials (CC and C&DW) shortened the initial setting time (60 min), a finding attributed to the calcium carbonate present in these materials, as reported by other authors.53,54 Calcined CW brought the initial setting time forward by 40 min, likewise at 10% and 20% replacement. The volume stability findings showed no shrinkage or expansion in any of the industrial waste materials, all of which were standard EN 197-15-compliant in this respect.

176 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 6.10 Water demand in blended cements (percentage relative to the control).

6.4.2 Organic waste The relative difference in water demand between the control (CEM I 42.5R) and the sample pastes is shown in Fig. 6.10. Three of the four ash types analyzed, i.e., activated bamboo leaf ash, laboratory scale sugar cane ash, and ash industrially processed from bagasse and leaf of sugar cane, induced a substantial rise in water demand (positive values), which was greater at the higher replacement ratio. The 20% BA blended cement paste required up to 46% more water than the reference paste, followed by IBA and LBA ash. These findings were in line with the results for other agroindustrial waste, such as rice husk ash and paper mill sludge.55,56 These findings for the water demand in the agroindustrial ash analyzed were related to a number of physical characteristics of the calcined ash: greater volume, lower density, extreme fineness, organic matter content, and very porous particles, all resulting from thermal activation.57,58 The findings for ash SFA differed from the results for the other three. Water demand was 7% 7.5% lower than in the reference cement, irrespective of the percentage of ash added to the blend. In other words, this ash (straw and forestry waste) acted as a fluidizing addition in blended cements. As its use is unprecedented, a tentative explanation for its behavior may be suggested: when heated at high temperatures, a chemical compound possibly present in the waste (resin in the forestry waste) may have yielded a denser, more compact, less porous ash than observed in the other materials that may have acted as a water repellent. Fig. 6.11 shows the initial setting times for all the blended cements prepared with 10% and 20% ash. The addition of agroindustrial ash was observed to consistently retard setting, more intensely when the ash content was higher. Setting was impacted least in ash BA, with a delay of only 30 min at 20% replacement, followed by LBA and IBA, where delays of 60 70 min were recorded. Lastly, setting began 70 130 min later in SFA than in the reference paste. These delays in initial setting time were related to the presence of minority elements in the agroindustrial ash, such as Zn, Pb, Cr, Sr, and Ni, most of which have a retarding

New trends for nonconventional cement-based materials: industrial and agricultural waste

177

Figure 6.11 Increase in initial setting times for blended cements compared to the reference material.

effect.59 The concentration of these elements in the agroindustrial ash was under 100 ppm for all except strontium (Sr), for which values of up to 600 ppm were found in IBA. This should not be regarded as a drawback, however, for mortars and concretes shipped from the factory must reach the worksite in conditions of workability that ensure satisfactory on-site placement and construction of the unit of work at issue. Consequently, where medium distances are involved, the retarding effect of these elements is beneficial. The expansion identified in all the blended cements was under 1 mm, a value well within the 10 mm ceiling specified in the existing legislation.

6.5

Mechanical strength of blended cement mortars

Mortar specimens containing the 10% and 20% blended cements were prepared and cured as laid down in the existing legislation to determine compressive strength.5,60

6.5.1 Inorganic waste Fig. 6.12 shows the compressive strength values for the 2, 7, 28, and 90 day mortars at 0%, 10%, and 20% replacement ratios. The 2-day compressive strength for the pozzolan blended mortars was observed to be lower than for the reference ordinary Portland cement mortar (OPC); except for the 10% CW and 10% CC mortars. Whilst that pattern remained practically the same in the 7-day specimens, after 28 days significant changes were recorded: strength was higher in the mortars containing 10% and 20% FCC than in the control, by up to 35% in the latter. Ninety day strength was the same or greater than in the reference ordinary Portland cement mortar (OPC) in all but not for mortars containing C&DW. It is known, however, that C&DW can be optimized at the recycle plant by raising its fired clay content.

178 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 6.12 Mortar compressive strength versus reaction time.

Percentage increment in compressive strength of experimental mortars relative to the reference mortar by hydration time

Table 6.4

Time

Content

LBA

7 days

10% 20% 10% 20% 10% 20%

1 22.5 1 9.5 1 25.0 1 12.6 1 27.2 1 13.2

28 days 90 days

IBA 2 10.6 2 10.0 2 5.9

BA

SFA

2 0.7 2 6.7 1 2.5 2 2.5 1 0.1 2 3.3

2 6.5 2 18.8 2 0.6 2 14.7 2 0.2 2 10.7

6.5.2 Organic waste The compressive strength values for standardized blended cement mortar specimens are given in Table 6.4 for both 10% and 20% replacement ratios for all but IBA, for which 20% replacement strength is shown. The specimens were cured in water for 7, 28, or 90 days. As expected, compressive strength varied depending on the nature, origin, and content of the ash addition. In much the same manner as found for the other properties, mechanical strength was higher in the ash obtained on a laboratory scale under controlled conditions than in the ash generated industrially. The cement mortars with 10% LBA exhibited 22% 27% higher strength, and with 20% replacement, 10% 13% higher compressive strength, than the reference. In the mortars containing BA, strength was comparable to the values for the reference material, with only minor positive or negative increments. Lastly, the industrial ash (IBA and SFA) exhibited lower compressive strength than the control, although as in the preceding case, the differences narrowed with curing time and were smaller at the lower ash content. A 6% decline was

New trends for nonconventional cement-based materials: industrial and agricultural waste

179

observed in compressive strength for IBA at 20% replacement, whereas for SFA strength dropped by 10% relative to the reference at that same ratio. The lower strength performance of blended mortars containing industrial ash was attributed primarily to the higher activation temperatures than used in the laboratory. More intense heat would induce more crystalline (due to recrystallization of amorphous silica) and lower specific surface ash. Moreover, the dilution effect induced by the presence of clay- and lime-based minerals in this ash would also lower its pozzolanicity.48,57,61 The high content of unburnt carbon in SFA (8.47%) would reduce strength even further in the respective mortar.43

6.6

Conclusions

Social and economic development and gradual population growth worldwide, with the concomitant increase in demand for food and goods, are generating ever larger amounts of organic and inorganic industrial waste and by-products. In most cases, this waste is neither managed nor controlled by the competent authorities, but rather stockpiled in rubbish tips or incinerated irregularly, adversely affecting the environment, society and public health. This trend is incompatible with sustainable development. The ongoing aim to use waste as raw materials across a wide range of industries has therefore been included in the main European strategy for a circular economy, based on “closing the life cycle” of food, goods, services, waste, materials, water, and energy. Given its technical and energy aspects and environmental imperatives, cement manufacturing is one of the industries most keen on applying this new industrial waste as supplementary cementitious materials that can serve as alternatives to conventional pozzolans. This chapter addresses the research conducted on various types of current waste that can be used as active additions in the preparation of future eco-efficient and innovative binary cements. The inorganic waste studied (FCC, fired clay materials, coal mining) has been found to yield active cement additions. The feasibility of its use on an industrial scale will be limited by its availability and management. C&DW performance as a pozzolan improves substantially when its fired clay material content is raised. Agroindustrial waste ash is an inexhaustible source of excellent active additions in blended cements, which conform to the existing legislation on the manufacture of commercial cements. Further to scientific and technical findings, a collection, shipping, and storage protocol is needed for the agroindustrial waste used as biomass fuel to reduce the contamination of the resulting ash by minerals present in the soil, thereby enhancing its cementitious properties.

Acknowledgements This research was funded by the Spanish Ministry of the Economy and Competitiveness under coordinated projects MAT2012-37005-CO3-01, BIA2015-65558-C3-1-2-3R (MINECO/FEDER),

180 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

BIA 2010-21194-C03-01, BIA 2013-48876-C3-1-R and BIA 2013-48876-C3-2-R, as well as by the Spanish National Research Council (CSIC) under projects i-LINK 0675 (CSIC-FAPESP) and I-COOPA 20089 (CSIC). The authors wish to thank the University of Extremadura-CSIC (SOSMAT) and Autonomous University of Madrid-CSIC (Geomateriales) associated units, the Grupo Cementero Lafarge (Villaluenga de la Sagra, Toledo, Spain) and the Brazilian companies UTE Baldin and Abengoa (Pirassununga, Brazil) for the support provided. The assistance received under the framework partnering agreements between CSIC/IETcc and the University of Sa˜o Paulo (Department of Biosystems Engineering), Brazil and CSIC/IETcc and the National Technical University (Regional Faculty at Santa Fe), Argentina, is likewise gratefully acknowledged. Also, to the CNPq (Process n. 313782/2013-0, Project PVE).

References 1. EUROPA 2020. Una estrategia para un crecimiento inteligente, sostenible e integrador, 2010. http://eur-lex.europa.eu/legal-ontent. 2. Werther, J.; Saenger, M.; Hartge, E. U.; Ogada, T.; Siagi, Z. Combustion of Agricultural Residues. Prog. Energy Combust. Sci. 2001, 26, 1 27. 3. Rahman, A.; Rasul, M. G.; Khan, M. M. K.; Sharma, S. Recent Development on the Uses of Alternative Fuels in Cement Manufacturing Process. Fuel 2015, 145, 84 99. 4. Tonoli, G. H. D.; Pizzol, V. D.; Urea, G.; Santos, S. F.; Mendes, L. M.; John, V. M.; Frı´as, M.; Savastano, H. Rationalizing the Impact of Aging on Fiber Matrix Interfase and Stability of Cement based Composites Submitted to Carbonation at Early Ages. J. Mater. Sci. 2016, 51 (17), 7929 7943. Available from: http://dx.doi.org/10.1007/ s10853-016-0060-z. 5. EUROPEAN STANDARD. UNE EN 197-1. Cement - Part 1: Composition, specifications and conformity criteria for common cements, 2011. 6. Garcı´a de Lomas, M.; Sa´nchez de Rojas, M.I.; Frı´as, M.; Mu´jika, R. Comportamiento cientı´fico-te´cnico de los cementos Portland elaborados con catalizador FCC. Aplicacio´n de la normativa vigente. Monografı´a 412. Instituto de Ciencias de la Construccio´n Eduardo Torroja del Consejo Superior de Investigaciones Cientı´ficas (CSIC), 2006. 7. Garcı´a de Lomas, M.; Sa´nchez de Rojas, M. I.; Frı´as, M. Pozzolanic Reaction of a Spent Fluid Catalytic Cracking Catalyst in FCC-cement Mortars. J. Ther. Anal. Calorim. 2007, 90 (2), 443 447. 8. Paya´, J.; Monzo´, J. M.; Borrachero, M. V. Physical, Chemical and Mechanical Properties of Fluid Catalytic Cracking Catalyst Residue (FC3R) Blended Cements. Cement Concrete Res. 2001, 31 (1), 57 61. 9. Borrachero, M. V.; Monzo´, J. M.; Paya´, J.; Vunsa, C.; Vela´zquez, S.; Soriarno, L. Spent Fluid Catalytic Cracking Catalyst for Improving Early Strength of Portland Cement. ACI Mater. J. 2014, 111 (1), 59 66. 10. Sa´nchez De Rojas, M. I.; Garcı´a De Lomas, M.; Asensio, E.; Frı´as, M.; Medina, C. Cementos con FCC: cumplimiento de requerimientos normativos. Cemento y Hormigo´n 2016, (972), 24 30. 11. Hea, C. H.; Osbzckb, B.; Makovicky, E. Pozzolanic Reactions of Six Principal Clay Minerals: Activation, Reactivity Assessments and Technological Effects. Cement Concrete Res. 1995, 25 (8), 1691 1702.

New trends for nonconventional cement-based materials: industrial and agricultural waste

181

12. Calleja, J. Las Puzolanas. Monografı´a 281. Instituto Eduardo Torroja de la Construccio´n y Cemento. Patronato de Investigacio´n Cientı´fica y Te´cnica Juan de la Cierva del Consejo Superior de Investigaciones Cientı´ficas (CSIC), 1969. 13. Sa´nchez de Rojas, M. I.; Frı´as, M.; Rivera, J.; Escorihuela, M. J.; Marı´n, F. Research about the Pozzolanic Activity of Waste Materials from Calcined Clay. Materiales de Construccio´n 2001, 51 (261), 45 52. 14. Sa´nchez de Rojas, M. I.; Marı´n, F.; Rivera, J.; Frı´as, M. Morphology and Properties in Blended Cements with Ceramic Waste Materials Recycled as Pozzolanic Addition. J. Am. Ceram. Soc. 2006, 89 (12), 3701 3705. 15. Sa´nchez de Rojas, M. I.; Frı´as, M.; Rodrı´guez Largo, O.; Rivera, J. Durability of Blended Cement Pastes Containing Ceramic Waste as a Pozzolanic Addition. J. Am. Ceram. Soc. 2014, v. 97 (5), 1543 1551. 16. Toledo Filho, R. D.; Goncalves, J. P.; Americano, B. B.; Fairbairm, E. M. R. Potential for Use of Crushed Waste Calcined-Clay Brick as a Supplementary Cementitious Material in Braz. Cement Concrete Res. 2007, v. 37 (9), 1357 1365. 17. Velosa, A. L.; Cachim, P. B. Hydraulic-Lime Based Concrete: Strength Development Using a Pozzolanic Addition and Different Curing Conditions. Constr. Build. Mater. 2009, 23 (5), 2107 2111. 18. Asensio, E.; Medina, C.; Sa´nchez de Rojas, M.I.; Frı´as, M. Blended cements based on C&DW. Influence in the pozzolanicity. Proceedings of the International Conference on Construction Materials and Structures (ICCM 2014). Johannesburg, South Africa. Noviembre. pp. 370 377, 2014. ISBN: 978-1-61499-7. 19. Medina, C.; Zhu, W.; Howind, T.; Frı´as, M.; Sa´nchez de Rojas, M. I. Effect of the Constituents (Asphalt, Clay Materials, Floating Particles and Fines) of Construction and Demolition Waste on the Properties of Recycled Concretes. Constr. Build. Mater. 2015, 79 (1), 22 33. 20. Sa´nchez de Rojas, M.I.; Frı´as, M.; Asensio, E.; Medina, C. Ceramic Waste Useful for Cement Manufacture, Obtention Proceeding and Cements that are Included. Patent No publication ES2512065. 2015. 21. Frı´as, M.; Vigil, R.; Sa´nchez de Rojas, M. I.; Medina, C.; Andre´s, J. Scientific Aspects of Kaolinite based Coal Mining Wastes in Pozzolan/Ca(OH)2 System. J. Am. Ceramic Soc. 2012, 95 (1), 386 391. 22. Frı´as, M.; Vigil, R.; Sa´nchez de Rojas, M. I.; Andre´s, J.; Medina, C. Effect of the Activated Coal Mining Wastes on the Properties of Blended Cement Matrixes. Cement Concrete Compos. 2012, 34 (5), 678 683. 23. Vigil, R.; Frı´as, M.; Garcı´a, R.; Martı´nez, S.; Ferna´ndez, L. Mineralogical Transformation of Coal-Mining Waste into Cementing Materials for Blended Cements: Activation Temperature and Pozzolanic Properties. J. Coal Geol. 2014, 132, 123 130. 24. Garcı´a, R.; Vigil, R.; Frı´as, M.; Rodrı´guez, O.; Martı´nez, S.; Ferna´ndez, L.; Sonsoles, I.; Villar, E. Mineralogical Study of Calcined Coal Wastes in Pozzolan/Ca(OH)2 System. Appl. Clay Sci. 2015, 108, 45 54. 25. Frı´as, M.; Rodrı´guez, O.; Vigil de la Villa, R.; Garcı´a, R.; Martı´nez, S.; Ferna´ndez, L.; Vegas, I. The Influence of Activated Coal Mining Wastes on the Mineralogy of Blended Cement Pastes. J. Am. Ceram. Soc. 2016, 99 (1), 300 307. 26. Vegas, I.; Cano, M.; Arribas, I.; Rodrı´guez, O.; Frı´as, M. Physical-Mechanical Behaviour of Binary Cements Blended with Thermally Activated Coal Mining Waste. Constr. Build. Mater. 2015, 99, 169 174.

182 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

27. Garcı´a, R.; Vigil, R.; Frı´as, M. From coal-mining waste to construction material: a study of its mineral phases. Environ. Earth Sci. 2016, 75 (6), 1 8. Available from: http://dx. doi.org/10.1007/s12665-016-5494-8. 28. European Standard. UNE En196-6. Method of testing cement. Part 6: Determination of the fineness, 2010. 29. Saval, S. Aprovechamiento de Residuos Agroindustriales: Pasado, Presente y Futuro. BioTecnologı´a 2012, 16 (2), 14 46. 30. Pizzol, V. D.; Mendes, L. M.; Savastano, H.; Frı´as, M.; Davila, F. J.; Cincotto, M. A.; John, V. M.; Tonoli, G. H. D. Mineralogical and Microstructural Changes Promoted by Accelerated Carbonation and Ageing Cycles of Hybrid Fiber Cement Composites. Constr. Build. Mater. 2014, 68, 750 756. 31. Siddique, R. Waste Materials and by Products in Concrete; Springer: Berlin, 2008. 32. Ferra´ndiz-Mas, V.; Bond, T.; Garcı´a-Alcocel, E.; Cheeseman, C. R. Lightweight Containing Expanded Polystyrene and Paper Sludge Ash. Constr. Build. Mater. 2014, 61, 285 292. 33. Frı´as, M.; Rodrı´guez, O.; Sa´nchez de Rojas, M. I. Paper Sludge, An Environmentally Sound Alternative Source of MK based Cementitious Materials. A Review. Constr. Build. Mater. 2015, 74, 37 48. ´ lvarez, A.; Lavı´n, A. G.; Bueno, J. L. Study of Biomass 34. Garcı´a, R.; Pizarro, C.; A Combustio´n Wastes. Fuel 2015, 148, 152 159. 35. Chiou, I. J.; Chen, C. H. Reuse of Incinerated Ash from Industrial Sludge Derived Fuel. Constr. Build. Mater. 2013, 49, 233 239. 36. Loh, Y. R.; Sujan, D.; Das, C. A. Sugar Cane Bagasse - The Future Composite Material: A Literature Review. Resour. Conserv. Recy. 2013, 75, 14 22. 37. Nakanishi, E. Y.; Frı´as, M.; Martı´nez, S.; Santos, S. F.; Rodrigues, M. S.; Rodrı´guez, O.; Savastano, H. Characterization and Properties of Elephant Grass Ashes as Supplementary Cementing Material in Pozzolan/Ca(OH)2 System. Constr. Build. Mater. 2014, 73, 391 398. 38. Nakanishi, E. Y.; Frı´as, M.; Martı´nez, S.; Santos, S. F.; Rodrigues, M. S.; Vigil, R.; Rodrı´guez, O.; Savastano, H. Investigating the Possible Usage of Elephant Grass Ash to Manufacture the Eco-friendly Binary Cements. J. Clean. Prod. 2016, 116, 236 243. 39. Garcı´a, R.; Vigil, R.; Gon˜i, S.; Frı´as, M. Fly Ash and Paper Sludge on the Evolution of Ternary Blended Cement: Mineralogy and Hydrated Phases. J. Mater. Civil Eng. 2014, 27 (9). 04014249-1. 40. Ettu, L. O.; Arimanwa, J. L.; Nwachukwu, K. C.; Awodiji, C. T. G.; Amanze, P. C. Strength of Ternary Blended Cement Concrete Containing Corn Cob Ash and Pawpaw Leaf Ash. Int. J. Eng. Sci. 2013, 2 (5), 84 89. 41. Aprianti, E.; Shafigh, P.; Bahri, S.; Farahani, J. N. Supplementary Cementitious Materials Origin from Agricultural Waste- A Review. Constr. Build. Mater. 2015, 74, 176 187. 42. Paya´, J.; Monzo, J.; Borrachero, M. V.; Dı´az-Pinzo´n, L.; Ordo´n˜ez, L. M. Sugar Cane Bagasse Ash: Studies on its Properties for Reusing in Cocnrete Production. J. Chem. Technol. Biotechnol. 2002, 77, 321 325. 43. Chusilp, N.; Jaturapitakkul, Ch; Kiattikomol, K. Effect of LOI of Ground Bagasse Ash on the Compressive Strength and Sulfate Resistance of Mortars. Constr. Build. Mater. 2009, 23, 3523 3531. 44. European Standard. UNE EN 196-5. Methods of Testing Cement - Part 5: Pozzolanicity Test for Pozzolanic Cement. 2011. p. 15.

New trends for nonconventional cement-based materials: industrial and agricultural waste

183

45. Villar, E.; Frı´as, M.; Valencia, E.; Sa´nchez de Rojas, M. I. An Evaluation of Different Kinetic Models for Determining the Kinetic Coefficients in Sugar Cane Straw-Clay Ash/ lime System. Adv. Cement Res. 2006, 18 (1), 17 26. 46. Frı´as, M.; Villar, E.; Valencia, E. Characterization of Sugar Cane Straw Waste as Pozzolanic Material for Construction: Calcining Temperature and Kinetic Parameters. Waste Manage. 2007, 27 (4), 533 538. 47. Sa´nchez de Rojas, M. I.; Rivera, J.; Frı´as, M. Influence of the Microsilica State on Pozzolanic Reaction Time. Cement Concrete Res. 1999, 29 (6), 945 949. 48. Frı´as, M.; Villar, E.; Savastano, H. Brazilian Sugar Cane Bagasse Ashes from the Cogeneration Industry as Active Pozzolans for Cement Manufacture. Cement Concrete Compos. 2011, v. 33 (4), 490 496. 49. Singh, N. B.; Das, S. S.; Singh, N. P.; Dwivedi, V. N. Hydration of Bamboo Leaf Blended Portland Cement. Indian J. Eng. Mater. Sci. 2007, 14, 69 76. 50. Frı´as, M.; Sa´nchez de Rojas, M.I. Uso integral de residuos en la fabricacio´n de nuevos cementos: Parte 2. In Sostenibilidad en construccio´n a trave´s de los materiales, Frı´as, M.; Sa´nchez de Rojas, M.I. Eds., Madrid, Espan˜a, 2014; pp 55 60. 51. Villar, E.; Frı´as, M.; Herna´ndez, J.; Savastano, H. Pozozlanic Behaviour of a Bagasse Ash from the Boiler of a Cuban Sugar Factory. Adv. Cement Res. 2013, 25 (3), 136 142. 52. Frı´as, M.; Savastano, H.; Villar, E.; Sa´nchez de Rojas, M. I.; Santos, S. Characterization and Properties of Blended Cement Matrices Containing Activated Bamboo Leaf Wastes. Cement Concrete Compos. 2012, 34 (9), 1019 1023. 53. Kakali, G.; Tsivilis, S.; Aggeli, E.; Bati, M. Hydration Products of C3A, C3S and Portland Cement in the Presence of CaCO3. Cement Concrete Res. 2000, 30 (7), 1073 1077. 54. Frı´as, M.; Rodrı´guez, O.; Vegas, I.; Vigil, R. Properties of Clacined Clay Waste and Its Influence on Blended Cement Bahviour. J. Am. Ceram. Soc. 2008, 91 (4), 1226 1230. 55. Frı´as, M.; Sa´nchez de Rojas, M.I. Artificial Pozzolans in Eco-Efficeint Concrete. In Eco-Efficient Concrete, Pacheco, F.; Jalali, S.; Labrincha, J; John, V.M. Eds., Cambridge, UK, 2013; pp 105 122. 56. Frı´as, M.; Vegas, I.; Vigil, R.; Garcı´a, R. Recycling of Waste Paper Sludge in Cements: Characterization and Behavior of New Eco-efficient Matrices. In Integrated Waste Management- II, Kumer, S. Ed., Rijeka, Croatia, 2011, pp 301 318. 57. Bie, R. S.; Song, X. F.; Liu, Q. Q.; Ji, X. Y.; Chen, P. Studies on Effects of Burning Conditions and Rice Husk Ash Blending Amount on the Mechanical Behaviour Of Cement. Cement Concrete Compos. 2015, 55, 162 168. 58. Soares, M. M. N. S.; Poggiali, F. S. J.; Bezerra, A. C. S.; Figueiredo, R. B.; Aguilar, M. T. P.; Cetlin, P. R. The Effect of Calcination Conditions on the Physical and Chemical Characteristics of Sugar Cane Bagasse Ash. Revista Escola de Minas, Ouro Preto 2014, 67 (1), 33 39. 59. Stephan, D.; Maleki, H.; Knofel, D.; Eber, B.; Hardtl, R. Influence of Cr, Ni and Zn on the Properties of Puere Clinker Phases. Part I. C3S. Cement Concrete Compos. 1999, 29, 545 552. 60. European Standard. UNE - EN 196-1. Method of Testing Cement. Part 1: Determination of Strength. 2005. 61. Bahurudeen, A.; Santhanam, M. Influence of Different Processing Methods on the Pozzolanic Performance of Sugar Cane Bagasse Ash. Cement Concrete Compos. 2015, v. 56, 32 45.

Alternative inorganic binders based on alkali-activated metallurgical slags

7

Maria Criado, Xinyuan Ke, John L. Provis and Susan A. Bernal The University of Sheffield, Sheffield, United Kingdom

7.1

Introduction

There is a growing global interest in maximizing the reuse and recycling of waste and industrial by-products, to reduce the ecological footprint linked to disposal and treatment of wastes. One rapidly growing strategy for future-proofing the sustainability of the construction industry is the utilization of wastes or industrial by-products as partial or total replacements for Portland cement in concretes. Alkali-activation technology has been successfully utilized for the valorization and/or management of waste and industrial by-products, producing alternative inorganic binders1 with performance similar to that of Portland cement, but with a fraction of the CO2 emissions.2 Numerous different types of raw materials have been studied for production of alkali-activated cements. However, significant efforts have been focused on the utilization of ground granulated blast furnace slag (GBFS), derived from the iron-making industry, as it is well known that alkali-activated cements derived from this by-product can develop a microstructure comparable (in many aspects) to that of a blended Portland cement,3 high mechanical strength,4 and durability over a long service life, as demonstrated by the assessment after 3060 years in service of some of the standing concrete buildings made with these cements.5,6 Despite the advantageous properties developed by alkali-activated cements based on GBFS, this by-product is already highly exploited by the construction industry to produce slag cements,7 and therefore there is a burgeoning interest in determining if metallurgical slags other than GBFS can also be utilized to produce high quality alkali-activated inorganic cements. In this chapter, we present an overview of some of the metallurgical slags that have been used as raw materials for the production of alkali-activated cements, including a brief description of the process which generates the slag, the quantities generated, the distinctive features of the alkali-activated inorganic cements produced, and the pitfalls and potential opportunities for utilization that can be identified.

7.2

Ferrous slags

Ferrous slags comprise a group of nonmetallic oxide materials produced at high temperature as melts during the manufacturing of iron and steel, with chemistry Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00008-5 © 2017 Elsevier Ltd. All rights reserved.

186 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

designed to remove impurities from iron as it is refined, and separated from the metal in the melt flow due to differences in density. The chemical composition and physical properties of ferrous slags depend on the process from which they are derived, the nature and amounts of raw materials used during the metallurgical process, and the cooling process adopted to treat the slag. There are two main types of ferrous slags: blast furnace slag (BFS) derived from the iron-making process, and steel slag produced during the conversion of hot metal (in a basic oxygen furnace), sponge iron, or steel scrap (in an electric arc furnace (EAF)) into crude steel.8 The process of production of these slags is illustrated in Fig. 7.1. Stainless steel production, and the production of alloys such as ferronickel and ferrosilicon, also lead to the generation of slags of distinct characteristics and chemistry. Blast furnace slags will not be covered in this chapter, as detailed compilations of information related to alkali-activated inorganic cements based on this byproduct have recently been reported.2,3,9

7.2.1 Steel slags Steelmaking slag, or steel slag, is a by-product from the production of steel during the conversion of hot metal to crude steel in a basic oxygen furnace, or during the melting of scrap in an EAF. The slag is generated as a melt and is a complex solution of silicates and oxides that solidify upon cooling.11 Depending on the specific

Figure 7.1 Scheme of the production of blast furnace and steel slags. Adapted from Horii, K.; Tsutsumi, N.; Kitano, Y.; Kato, T. Processing and Reusing Technologies for Steelmaking Slag. Nippon Steel Technol. Report No 14, 2013, 123129.10

Alternative inorganic binders based on alkali-activated metallurgical slags

187

steel production process, three different types of steel slags can be produced: basic oxygen furnace slag (BOFS), EAF acid slag, and ladle furnace (LF) basic slag, also called refining slag (Fig. 7.1). The chemical composition of steel slags is highly variable, depending on the process route used and the amount of recycled materials added into the steelmaking process. This variability is identified from plant to plant, and from batch to batch even within the same plant,12,13 meaning that the utilization of this by-product is challenging and necessitating detailed quality control. A total of 21.4 Mt of steel slags were produced in Europe in 2012,14 whose distribution and routes of utilization are shown in Fig. 7.2. Steel slags have been used by the construction industry for several decades, particularly in China, where the use of these by-products for cement and concrete production is already standardized (GBT 20491-2006, GB 13590-2006).15 Other

Figure 7.2 Overview of the different types of steel slag produced in Europe in 2012 (top) and their applications (bottom). Adapted from EUROSLAG Statistics 2012. http://www.euroslag.com/PRODUCTS/ STATISTICS/2012/ (accessed Apr 11, 2016).

188 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

specialized Chinese standards are also available for steel slag utilization in road construction, brickwork, and as a supplementary cementitious material (JC/T 10872008, JC/T 1090-2008, and JC/T 1082-2008).16 In spite of this, of the 90 Mt of steel slag produced in China in 2010, just 20% were utilized for industrial purposes. Currently, China has more than 300 Mt of legacy steel slags landfilled, which represents a severe environmental problem.17 Therefore, there is an existing need for repurposing and valorizing large amounts of steel slag. The main mineral phases identified in steel slags include olivine, merwinite, C3S, β-C2S, C4AF, C2F, the “RO” phase (CaO-FeO-MnO-MgO solid solution), and free CaO.1820 Careful attention needs to be paid to the dimensional stability of cements containing steel slag, as high contents of free CaO and iron-rich phases can lead to volumetric changes.16,18 The content of free CaO must be limited below 3 wt.%, and the content of metallic Fe must be restricted to a maximum of 2 wt.%, when steel slag is used as a supplementary cementitious material in Portland cement based systems.16 The relative proportions of the mineral phases change depending on the chemical composition of the initial iron, and the steel produced. Generally steel slags have 1015 wt.% SiO2, 15 wt.% Al2O3, 4560 wt.% CaO, 720 wt. % Fe2O3, 313 wt.% MgO, and 14 wt.% P2O5.20,21 The presence of tricalcium silicate (C3S), dicalcium silicate (β-C2S), tetracalcium aluminoferrite (C4AF), and calcium ferrite (C2F) gives steel slag cementitious properties, which increase with its basicity (which can be calculated as the CaO/SiO2 ratio, or variants of this ratio which include, e.g., MgO and Al2O3 contents), although free CaO content also increases with the basicity of steel slag. The C3S and β-C2S contents in steel slag are much lower than those typically present in Portland cement, and its reactivity is limited by its thermal history. Thus, steel slag has been regarded as weakly hydraulic compared to Portland cement clinker,20,22 although it can exhibit very good cementitious properties under the action of chemical activators.19,20,2325 BOFS is produced during the primary stage of steel production, and is a major source of steel slag aggregate for roads construction. EAF is generated during the production of more specialist steels from the refining of scrap iron.26 BOFS and EAF slag are now used in asphaltic mixes and road-base layers.27 LF slag, or refining slag, is a basic slag produced in the final stages of steelmaking, when the steel is desulfurized in the transport ladle, during what is generally known as the secondary metallurgy process. Ladle slag generation is approximately one third of the total amount of slag usually produced in an EAF, with an estimated production of 4 million tonnes p.a. in Europe in 2007.21 Compared to BFS, ladle slag is richer in calcium oxide and poorer in silica. The average chemical composition of ladle slags is: SiO2, 1826 wt.%; Al2O3, 413 wt.%; CaO, 5056 wt.%; Fe2O3, 13 wt.%; MgO, 37 wt.%; TiO2, 0.30.9 wt.%; Na2O, 0.030.06 wt.%; K2O, 0.020.01 wt.%; SO3, 24 wt.%.12,23 The use of a calcium-rich raw material, such as ladle slag, as a precursor for alkali-activated cement production has great potential, as the alkali-activated products may achieve desirable setting and hardening characteristics without high temperature curing. Nikoli´c et al.28 assessed the dissolution of steel slags in KOH and NaOH solutions, identifying a correlation between the kinetics of dissolution of Si and Al

Alternative inorganic binders based on alkali-activated metallurgical slags

189

species and the chemical composition of the main binding phase, a calcium aluminum silicate hydrate (C-A-S-H) type gel, forming upon activation of these materials. Preferential leaching of Si from the steel slags was identified when using NaOH as activator, while the use of KOH solution promoted Al dissolution. Wang et al.29 identified the formation of similar reaction products when activating a steel slag with NaOH solutions with different concentrations, although increased alkalinity accelerated the reaction of the hydraulic phases present in the slag. Slight differences in the phase assemblages of the activated steel slag cements were observed at extended curing times, along with a limited mechanical strength (9 MPa after 360 days of room-temperature curing). This demonstrated that it is feasible to produce solid monoliths using steel slag as the sole silicate precursor, but optimization of the formulation of these cements is required for production of higher strength materials. Studies have also been carried out utilizing steel slags as a partial replacement in blends with other precursors, to produce alkali-activated cements. Alkali-activation of blends of steel slag and metakaolin produced hardened monoliths with good mechanical strengths (44 MPa at 28 days).20 This was associated with the formation of C-A-S-H, along with a sodium aluminosilicate hydrate (N-A-S-H) type gel, as the main reaction products. The addition of steel slag accelerated the reaction, and significantly improved the compressive strength of the metakaolin-based geopolymers, as a consequence of its latent hydraulic cementitious character. The addition of steel slag also improved the abrasion resistance of metakaolin geopolymers through the formation of a denser microstructure.20,25 Shi et al.18 reported that 20 wt.% replacement of BFS by steel slag in silicateactivated cements yielded higher mechanical strengths than those obtained when solely using BFS, although a higher content of steel slag replacement beyond 20 wt.% led to reductions in mechanical strength. Alkali-activated BFS/steel slag blended cements showed a dense microstructure and, if it is added in adequate proportions, the steel slag can contribute to mitigation of shrinkage. In a different study, sieved ladle slags with maximum particle sizes of 149, 74, and 44 μm were activated with sodium silicate.23 Upon mixing with water, all the slag fractions produced hardened monoliths, with compressive strengths between 15 and 20 MPa after 28 days of curing, with higher strengths generally observed when using finer steel slag fractions. The addition of an alkali activator promoted the development of much higher mechanical strengths, associated with the more extensive reaction of the hydraulic C2S polymorphs present in the slag. Natali Murri et al.24 studied the thermal properties of blended alkali-activated materials including ladle slag and metakaolin (MK), or ladle slag and fly ash (FA). Samples were cured for 7 days then exposed to temperatures of 1000 C, and the compressive strengths determined before and after heat treatment are shown in Fig. 7.3. The ladle slag/FA-based alkali-activated materials exhibited superior strength gain and better thermal stability than the ladle slag/metakaolin-based alkali-activated materials, as a consequence of the differences in the type of reaction products formed, and the reduced water content at which the ladle slag/FA samples were formulated and mixed.

190 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.3 Effect of the bulk Si/Al ratio in the binders on the compressive strength of alkaliactivated ladle slag, blended with metakaolin or fly ash, before and after heat treatment at 1000 C. Two ladle slag/fly ash mixes were formulated, with low calcium content (CaO: 5.36.8 wt.%) and high calcium content (CaO: 17.520.6 wt.%). Data from Natali Murri, A.; Rickard, W.D.A.; Bignozzi, M.C.; van Riessen, A., High Temperature Behaviour of Ambient Cured Alkali-Activated Materials based on Ladle Slag. Cement Concrete Res., 2013, 43, 5161. doi: 10.1016/j.cemconres.2012.09.011.

7.2.2 Ferronickel slag The production of ferronickel from lateritic ores follows a pyrometallurgical route including three important stages: (1) prereduction in rotary kilns; (2) reductive smelting in EAFs; and (3) ferronickel enrichment-refining in converters.30 During reductive smelting, two separate phases are formed into the EAF: the ferronickel alloy, and the slag that contains oxides from the gangue material along with the remaining iron. The slag produced in the EAF represents about 8090% of the mass of the feedstock material. As a result, a vast quantity of slag is generated in ferronickel production, which is a significant concern for ferronickel plants worldwide. For every tonne of ferronickel alloy produced, 4 tonnes of slag are generated,31 and the annual ferronickel slag production in Greece alone was about 1.7 Mt in 2007. Ferronickel slag has been used in a diversity of applications by the construction industry including as replacement of limestone aggregates for the construction of roads,32 and the production of cement and concrete,32 as an additive in Portland cement or a substitute for natural aggregates,33,34 however its utilization is limited

Alternative inorganic binders based on alkali-activated metallurgical slags

191

Figure 7.4 Phase diagram of the FeO-SiO2-MgO system. The gray-circled region corresponds to nominal slag compositions derived from several ferronickel smelting operations, neglecting any potential influence of Al2O3 content on the liquidus temperatures. Adapted from Diaz, C.; Landolt, C.; Vahed, A.; Warner, A.; Taylor, J., A Review of Nickel Pyrometallurgical Operations. JOM, 1988, 40, 2833. doi: 10.1007/s10853-006-0529-2.

at present. This slag can be described within the FeO-SiO2-Al2O3-MgO system (e.g., the simplified view shown in Fig. 7.4), with a composition particularly rich in iron oxides (3840 wt.%) and Al2O3 (811 wt.%). It also contains substantial amounts of MgO (26 wt.%), CaO (34 wt.%), and trivalent chromium oxides (23 wt.%), as well as traces of nickel (0.09%).35,36 The high silica and alumina contents make this slag suitable for the production of alkali-activated cements. The feasibility of producing cements by alkali-activation of granulated ferronickel slag was demonstrated by Golubnitchy and Koksharev,37 who produced heat-cured concretes with compressive strengths of up to 60 MPa using a sodium silicate activator. There also exists a Ukrainian specification describing the use of ferronickel slags in alkali-activated cements.38 The production of alkali-activated cements using ferronickel slag was evaluated in more detail by Komnitsas et al.,35 who produced monoliths reaching compressive strengths of 15 MPa after 28 days of curing, which retained mechanical strength when exposed to distilled or sea water, or when subjected to freezethaw cycles (Fig. 7.5). Conversely, the exposure of these cements to acidic solutions led to reductions in the compressive strength (Fig. 7.5), which was attributed to the breakage of the Si-O-Al bonds present in the main binder phase, and the consequent dealumination and liberation of silicates. Studies of alkali-activated ferronickel slag blended with supplementary sources of reactive Si and/or Al have also been carried out. The addition of kaolinite or metakaolin in these cements induced a monotonic reduction of the compressive strength as the percentage of clays added increased.39 The partial replacement of ferronickel slag by waste glass promoted the formation of a highly amorphous microstructure, and the inclusion of nitrate or sulfate additives into these inorganic binders reduced the compressive strength as the amount of additives increased.40 Maragkos et al.30 identified an increase in the compressive strength of alkaliactivated ferronickel slags as the solid/liquid ratio was reduced, although the workability

192 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.5 Change in compressive strength of ferronickel based activated cements, as a function of exposure to different aggressive environments, after initial curing at 80 C for 48 h then ambient temperature for 28 days. Freezethaw cycling was carried out weekly from 215 to +60 C. Data from Komnitsas, K.; Zaharaki, D.; Perdikatsis, V., Geopolymerisation of Low Calcium Ferronickel Slags. J. Mater. Sci., 2007, 42, 30733082. doi: 10.1007/s10853-006-0529-2.

of the paste was compromised at low liquids content. Increased concentrations of NaOH and soluble silicates promoted higher strength, associated with more extensive dissolution of the slag, and the formation of more crosslinked reaction products. The nature of the alkali activator cation also influenced the mechanical strength development of these cements: when utilizing potassium based activators more workable pastes are produced, and higher compressive strengths are obtained compared to sodium hydroxide-activated ferronickel slag cements, independent of the curing temperature.39 Golubnitchy and Koksharev37 and Sakkas et al.41 also reported on the thermal performance of these cements, identifying that they are effective heat flux barriers and can offer very good protection to structural concrete in case of fire, with compressive strength and thermal conductivity comparable to or exceeding the performance of existing fire protection materials.

7.2.3 Titaniferous slag Titaniferous slag is generated during the extraction of iron from ilmenite or titaniferous ores by carbothermic reduction smelting, which can take place in a direct current plasma or EAF.42 Titaniferous slag has a chemical composition in the approximate range: SiO2, 2328 wt.%; Al2O3, 1517 wt.%; CaO, 2532 wt.%; Fe2O3, 0.55 wt.%; MgO, 79 wt.%; MnO, 0.13 wt.%; TiO2, 1524 wt.%; and SO3, 0.20.4 wt.%. Its quality can be determined via calculation of a quality modulus K=(CaO+MgO+Al2O3)/(SiO2+MnO+TiO2), whose values range between 1.06 and 1.25,43 with a higher modulus generally indicating a more reactive slag. The

Alternative inorganic binders based on alkali-activated metallurgical slags

193

cooling conditions and the content of TiO2 in these slags strongly influence reactivity; slags containing more than 25 wt.% TiO2 present limited reactivity, especially when water cooled,43 as this favors crystallization of the slag and therefore restricts its applicability as a construction material. There are very few reports of the utilization of titaniferous slags to produce alkali-activated cements. Wang et al.44 produced alkali-activated titaniferous slag bricks using a combined NaOH/Ca(OH)2 activator at concentrations of 9 to 11 wt. %, along with minor additions of anhydrite. The resulting bricks achieved compressive strengths of up to 3040 MPa after 7 days of curing. Cheng et al.43 demonstrated the feasibility of utilizing titaniferous slag as the sole binder precursor for alkali-activated concretes, yielding compressive strengths of up to 95 MPa when applying autoclave curing. A strong relationship between the compressive strength and the slag cooling/quenching conditions was identified,43 and the type of activator and curing conditions of activated titaniferous slag specimens were also important. The partial substitution of titaniferous slag by BFS significantly increased the compressive strength developed of these alkali-activated binders.

7.2.4 Stainless steel slag During the production of stainless steel, particularly from scrap metal, several slags are generated in the different stages of metal processing; these include argon oxygen decarburation (AOD) slag produced in the AOD converter, and EAF slag. The final phase of the metallurgical operation by which stainless steel is produced takes place in a continuous casting ladle, where a molten layer of slag (ladle slag) is maintained on top of the liquid metal, serving to prevent its oxidation and minimize heat losses, as well as ensuring the desired alloy composition.45 These stainless steel slags tend to be more highly crystalline than BFSs due to their higher basicity (CaO/SiO2 ratio). They contain high amounts of CaO and their chemical composition consists of: SiO2, 2531 wt.%; Al2O3, 14 wt.%; CaO, 4759 wt.%; Fe2O3, 0.42 wt.%; MgO, 719 wt.%; and Cr2O3, 0.41.7 wt.%. The majority of the CaO is present in the form of dicalcium silicate (C2S), Fig. 7.6C. During cooling of the slag from its molten state, this phase undergoes polymorphic transformations from the high temperature hydraulic α and β forms to the nonhydraulic γ form, leading to an increase of volume of about 12%.46 This phenomenon results in pulverization of the slag into fine dust, which brings advantages for use as a binder component as it can reduce the energy and cost required in the slag milling process. The alkali-activation of relatively crystalline stainless steel slags, with high contents of CaO and SiO2, can under suitable conditions of alkali concentration and temperature yield a material with desirable binding properties. For example, Salman et al.47 activated the 245 μm fraction of continuous casting stainless steel slag with different solutions of sodium and potassium hydroxide at 80 C. The extent of reaction as measured by heat flow increased as the molarity of sodium hydroxide increased up to a concentration of 10 M, while using potassium hydroxide as activator, the cumulative heat flow increased up to 7.5 M activator concentration and then decreased beyond this value. The samples activated with sodium hydroxide showed higher total heat output and higher strengths than those activated

194 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.6 Compressive strengths of alkali-activated continuous casting stainless steel slag mortars after steam curing for 16 hours with (A) Na silicate activator and (B) K silicate activator, and (C) mineral composition of the slag used to produce the mortars, from Rietveld refinement. ¨ .; Pontikes, Y.; Vandewalle, L.; Blanpain, B.; van Balen, K., Data from Salman, M.; Cizer, O Effect of Curing Temperatures on the Alkali Activation of Crystalline Continuous Casting Stainless Steel Slag. Constr. Build. Mater., 2014, 71, 308316. doi: 10.1016/j. conbuildmat.2014.08.067.

with potassium hydroxide47; addition of silicates to the alkaline solutions can improve the mechanical properties achieved,45,48,49 while elevated temperature (steam) curing is also beneficial, Fig. 7.6.49 The porosity of the continuous casting stainless steel slag samples activated with K-containing activators was lower than that of the samples with sodium-containing activators. C-S-H type reaction products were produced with both activators at all

Alternative inorganic binders based on alkali-activated metallurgical slags

195

Figure 7.7 Schematic diagram of the general process for nonferrous slag generation. Adapted from User Guidelines for Waste and Byproduct Materials in Pavement Construction. U.S. Federal Highway Administration. http://www.fhwa.dot.gov/publications/ research/infrastructure/structures/97148/nfs1.cfm (accessed Mar 09, 2016).51

temperatures, while the formation of brucite via periclase hydration was also favored at higher steam curing temperatures. Salman et al.48 also observed the effect of the alkali type and the ratio of hydroxide to silicate on the hydration and mechanical strength development of the binders based on stainless steel refining slag. Consistent with the results for continuous casting stainless steel slags, the cumulative heat release associated with the formation of hydration products, and the compressive strengths achieved, were higher when higher amounts of silicates were present in the activating solutions.

7.3

Nonferrous slags

Nonferrous slags are wastes derived from pyrometallurgical extraction of nonferrous metals from natural ores, or from reprocessing of recycled metals. The main nonferrous slags produced worldwide are derived from the extraction of lead, zinc, nickel, copper, manganese, and phosphorus.50 Fig. 7.7 illustrates the general process leading to nonferrous slag generation. In most cases, these slags are classified as toxic wastes as they can contain high concentrations of heavy metals, which limits their utilization. Therefore, cementation of these slags can be seen as a method of solidification/stabilization to minimize the environmental impacts associated with landfill disposal, rather than specifically as a means of valorization.

7.3.1 Lead slag Lead smelter slag is produced through the process of extraction of lead (Pb) from ores, and from reprocessing of recycled Pb mainly from electronic devices and car batteries.52 Global primary production of 10.4 Mt of lead was reported in 2012, with an additional 5.8 Mt produced from secondary sources.53 It is estimated that for each tonne of metallic lead produced, roughly 100350 kg of slag is

196 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.8 General flow diagram for processing lead ores. Adapted from Piatak, N.M.; Parsons, M.B.; Seal, R.R., Characteristics and Environmental Aspects of Slag: A Review. Appl. Geochem., 2015, 57, 236266. doi: 10.1016/j. apgeochem.2014.04.009.

generated.52 Fig. 7.8 illustrates the general production process of lead slag, and Table 7.1 shows its usual range of chemical composition.54 In general these slags tend to present a highly variable composition depending on the source, very low CaO content compared with steel slags and BFS, and are rich in SiO2 and Fe2O3 (Table 7.1). Lead slags can also contain high concentrations of the heavy metals arsenic (As) and lead (Pb), and therefore some are classified as hazardous wastes. Alkali-activation of lead smelter slags has been carried out as a means for the solidification/stabilization of these materials, particularly targeting the immobilization of the lead and arsenic. Ogundiran et al.55 produced potassium silicateactivated binders based on FA/BFS blends, including a spent aluminate solution and lead slag in different proportions. The replacement of FA by lead slag up to 10 wt.% did not significantly change the mechanical strength of the monoliths produced (B100 MPa after 28 days curing). A significant reduction in the leachability of Pb was achieved by solidifying the lead slag within the blended activated cement (Fig. 7.9), although an increase in the leachability of arsenic was observed when the lead slag was alkali-activated, particularly for cements produced using the spent aluminate solution (Fig. 7.9). Onisei et al.56 produced alkali-activated cements based solely on lead slag, although these displayed significant cracking upon high temperature curing (70 C) (Fig. 7.10), and developed limited mechanical strength. The main binding phase formed upon activation was a silicon-rich gel as the content of alumina in the slag was low, also containing notable quantities of Pb, Fe and Zn. The chemical composition of the binding matrix formed in these cements is shown in the pseudo-ternary diagram in Fig. 7.10. Due to the lack of aluminum in the lead slag, the addition of FA as a secondary aluminum source was shown to be effective; an optimum lead slag/FA ratio of 70/30 was reported, yielding a compressive strength of 47 MPa after sealed curing, increased density and reduced water absorption.56 These cements

Alternative inorganic binders based on alkali-activated metallurgical slags

197

Chemical composition determined by X-ray fluorescence (values reported in wt.%) and titration (values reported in mg/kg) of lead smelter slags from several studies

Table 7.1

wt.%

Min

Max

mg/kg

Min

Max

Al2O3 CaO FeO total K2O MgO MnO Na2O SiO2 TiO2

1.74 0.45 3.16 0.23 0.37 0.09 0.02 17.6 0.01

11.1 23.1 59.6 2.6 5.4 9.0 1.4 54.6 5

As Ba Cd Co Cr Cu Ni Pb Zn

87 169 0.3 6.1 19 88.5 5.6 5000 701

2900 190000 700 185 700 7550 240 319190 120000

Data from Piatak, N.M.; Parsons, M.B.; Seal, R.R., Characteristics and Environmental Aspects of Slag: A Review. Appl. Geochem., 2015, 57, 236266. doi: 10.1016/j.apgeochem.2014.04.009.

immobilized Pb most likely via incorporation in the main aluminosilicate binding phase, as the formation of crystalline phases such as the Pb3SiO5 or Na6PbO5 which were tentatively identified during immobilization of Pb in aluminosilicate geopolymers57,58 was not identified. However, as was the case for the lead slag alone (Fig. 7.9), the leachability of arsenic increased upon alkali-activation of the lead slag/FA blend. Therefore, caution needs to be paid to leaching of toxic elements from the inorganic cements that can be produced via alkali-activation of lead slags, to assure that any released fractions are below the limits of classification of hazardous wastes.

7.3.2 Copper slag Copper slag, or fayalite slag, is a by-product obtained during the matte smelting, converting and refining of copper (Fig. 7.11).54,59 It has been estimated that for every tonne of copper produced, about 2.2 tonnes of slag is generated,60 as a consequence of the relatively low grades of copper concentrates now available. The global production of refined copper over the past eight years is shown in Fig. 7.12. Two types of slag can be produced: an air-cooled copper slag with light black color and a glassy appearance, and granulated copper slag which is porous with a vesicular structure.26 The specific gravity of copper slag varies as a function of the iron content, between 3.2 and 3.8 kg/m3, and its water absorption is typically very low (,0.2 wt.%),54,59 although this depends on the porous structure and also the extent of grinding used to reduce the slag to a fine powder. Copper slag is essentially a ferrous silicate whose range of chemical compositions is shown in Table 7.2, where the high variability in chemical composition of these slags is evident. Some of these slags can contain high concentrations of heavy metals, particularly arsenic, cadmium, and lead, leading to their classification as hazardous wastes. The concentrations of arsenic in copper slags depend on the furnace use during smelting,62 being higher during the conversion step of the copper ore.

Figure 7.9 Leached concentration of lead (left) and arsenic (right) from alkali-activated lead slag fly ash-blast furnace slag binders in distilled water. Binders were activated by spent aluminate solution (cement A) and potassium silicate (cement B). For comparison, the EU Landfill Directive concentration limits for nonhazardous waste are 10 ppm for Pb, and 2 ppm for As. Data from Ogundiran, M.B.; Nugteren, H.W.; Witkamp, G.J., Immobilisation of Lead Smelting Slag within Spent Aluminate—Fly Ash based Geopolymers. J. Hazard. Mater., 2013, 248249, 2936. doi: 10.1016/j.jhazmat.2012.12.040.

Alternative inorganic binders based on alkali-activated metallurgical slags

199

Figure 7.10 Micrograph (left) of an alkali-activated Pb slag based cement, and the chemical composition of the binding phase (right) determined by EDX analysis. Data from Onisei, S.; Pontikes, Y.; van Gerven, T.; Angelopoulos, G.N.; Velea, T.; Predica, V.; Moldovan, P., Synthesis of Inorganic Polymers Using Fly Ash and Primary Lead Slag. J. Hazard. Mater., 2012, 205206, 101110. doi: 10.1016/j.jhazmat.2011.12.039.

Figure 7.11 General flow diagram for processing copper ores (including secondary copper sources) and slag production. Adapted from Piatak, N.M.; Parsons, M.B.; Seal, R.R., Characteristics and Environmental Aspects of Slag: A Review. Appl. Geochem., 2015, 57, 236266. doi: 10.1016/j. apgeochem.2014.04.009.

Currently copper slags are mainly recycled for recovery of metals,50 although slags with reduced copper contents (less than 0.8 wt.% Cu) are either discarded as waste or sold as products with properties similar to those of natural basalt (crystalline) or obsidian (amorphous).60 Processed air-cooled and granulated copper slags have been used as fine and coarse aggregates for production of concretes,59,63 and as a supplementary cementitious material for partial replacement of Portland cement.64 Limestone is often added to copper slag/Portland blended cement as a complementary CaO source, as the copper slag itself is low in calcium.65,66 Deja et al.67 carried out one of the first studies using copper slag as precursor to produce sodium hydroxide-activated cements, and obtained monoliths with compressive strengths up to 45 MPa after 4 h of curing at 80 C. The immersion of these specimens either in water or a chloride/sulfate rich solution led to a slight increase in the mechanical strength after one year of exposure, demonstrating the long-term stability of these inorganic binders. Onisei et al.68 assessed the effects of the type of

200 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.12 Global production of refined copper. Data from Oracle Mining Corp. Why We Belive in Copper’s Long-term Value. http://www. oracleminingcorp.com/copper/ (accessed Apr 20, 2016).61

Chemical composition of copper slags determined by X-ray fluorescence (values reported in wt.%) and titration (values reported in mg/kg) from several studies Table 7.2

wt.%

Min

Max

mg/kg

Min

Max

Al2O3 CaO FeO total K2O MgO MnO Na2O SiO2 TiO2

0.01 0.15 0.67 0.01 0.09 0.03 0.01 9.82 0.1

18.9 21.9 62.0 4.83 6.45 6.55 4.31 70.7 11.8

As Ba Cd Co Cr Cu Ni Pb Zn

0.8 28 0.43 15 13 1400 2 6.2 4.4

75865 29000 14000 24104 7510 353580 935 183800 280000

Data from Piatak, N.M.; Parsons, M.B.; Seal, R.R., Characteristics and Environmental Aspects of Slag: A Review. Appl. Geochem., 2015, 57, 236266. doi: 10.1016/j.apgeochem.2014.04.009.

activator on the early reaction of fayalite slags, and on the microstructural evolution of the resulting activated cements. When using NaOH at high concentrations a very fast reaction of the slag was identified. Conversely, when using Na2O  rSiO2 solutions as activators, more moderate rates of reaction were obtained. A recent study evaluated the feasibility of activating ancient copper slags from the north of Chile,69 and identified notable differences in the mineralogy and chemical composition depending on the slag source. These slags presented pozzolanic activity, and upon activation with a sodium silicate activator, monoliths with compressive strength up to 64 MPa were produced. When specimens were cured at room

Unconfined compressive strength (MPa)

Alternative inorganic binders based on alkali-activated metallurgical slags

100

201

0 wt.% Copper smelter slag 50 wt.% Copper smelter slag 100 wt.% Copper smelter slag

80 60 40 20 0

60

70

80 90 Temperature (°C)

100

Figure 7.13 7-day compressive strength as a function of the curing temperature of sodium hydroxide-activated copper smelter slag/copper mine tailings binders, with different ratios of copper slag to mine tailings as marked. Data from Ahmari, S.; Parameswaran, K.; Zhang, L., Alkali Activation of Copper Mine Tailings and Low-calcium Flash-furnace Copper Smelter Slag. J. Mater. Civ. Eng., 2014, 27, 04014193-1-04014193-11. http://dx.doi.org/10.1061/(ASCE)MT-1943-5533.0001159.

temperature, this strength was achieved after 90 days of curing, although increasing the curing temperature to 65 C led to a similar strength being reached after 7 days. This elucidates that it might be possible to utilize existing stockpiles of copper slags for producing alkali-activated cements, and to contribute to remedying some of the existing environmental issues associated with the disposal of these slags. Ahmari et al.70 assessed the effect of partially replacing copper slag by copper mine tailings, on the microstructure and mechanical performance of alkali-activated cements produced with these precursors. The replacement of copper slag by copper mine tailings reduced the compressive strength of the alkali-activated cements (Fig. 7.13), depending on the curing temperature. Specimens cured at 60 C and 90 C, containing 50 wt.% copper slag showed comparable strengths to that obtained by samples with 100 wt.% copper slag. In copper slag-free activated binders an optimal curing temperature of 75 C was identified, while in specimens containing 50100 wt.% copper slag a higher curing temperature (90 C) yielded the best compressive strengths. These studies demonstrated that high compressive strength cements can be developed via alkali-activation of copper slags; however, the leachability of heavy metals from these inorganic cements remains unknown, which might limit their applicability as building materials.

7.3.3 Nickel and copper-nickel slag Nickel production from high-magnesium nickel oxide ores follows a pyrometallurgical process including prereduction, smelting in a blast furnace, and nickel

202 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

enrichment refining.71 During the smelting stage, besides the nickel alloy, nickel slag is also generated. The nickel slag is recycled back into the process until its nickel concentration is low, and is then disposed of in stockpiles.72 In China alone, the 2005 generation of nickel slag was 800,000 tonnes, and just 8% of this was utilized for construction purposes,72,73 with the remainder stockpiled or landfilled.72 In Greece 1.5 Mt of nickel slag were generated in 2000.74 Nickel slags are rich in SiO2 and Fe oxides, with traces of Al2O3 and CaO, sulfur, MgO and other oxides, depending on the ore source and process of Ni extraction.75 The percentages of the main oxides of nickel slag vary in the following ranges: SiO2, 3137 wt.%; Al2O3, 48 wt.%; CaO, 13 wt.%; Fe2O3, 3358 wt.%; MgO, 212 wt.%; Na2O, 0.041.2 wt.%; S, B0.7 wt.%; Cr, 0.10.4 wt.%; Ni, 0.030.2 wt.%.76 Nickel slags are vitreous with minor crystalline components enriched in iron, and have been identified as potential supplementary cementitious materials for production of blended Portland cements,74 in addition to use as aggregates for pavements.77 A few studies have been published utilizing nickel slag for producing alkaliactivated cements. Bin et al.78 produced alkali-activated nickel slag cements with good mechanical performance, although the contents of CaO and Al2O3 needed to be adjusted to increase reactivity in binder formation. Blocks were produced with these cements using nickel slag fractions with different finenesses,79 so that the fine fraction was used as the main precursor for binder formation, and the coarse fraction was used as a substitute for sand. These blocks developed 28.9 MPa after curing. Yang et al.80 added high-magnesia nickel slag to FA-based alkali-activated cements, identifying as main binding product a sodiummagnesium aluminosilicate type gel, although it remains unclear whether this is a single-phase product or an intimately intermixed combination of multiple disordered phases. The Si/Al ratio of the gel phase increased with high-magnesium nickel slag content, as this slag was acting as an additional silicon source. Addition of 20 wt.% nickel slag to the predominantly FA-based binder promoted the highest compressive strength, independent of the modulus of solution of the sodium silicate, and reduced the linear

Figure 7.14 Scanning electron micrographs of fracture sections of alkali-activated fly ashes with (A) 0 wt.% and (B) 20 wt.% addition of high magnesiumnickel slag. F, fly ash particles; G, gel phase formed; H, high-magnesiumnickel slag particles. Adapted from Yang, T.; Yao, X.; Zhang, Z., Geopolymer Prepared with High-magnesium Nickel Slag: Characterization of Properties and Microstructure. Constr. Build. Mater., 2014, 59, 188194. doi: 10.1016/j.conbuildmat.2014.01.038.

Alternative inorganic binders based on alkali-activated metallurgical slags

203

drying shrinkage of the specimens assessed. This was associated with the development of a dense microstructure in the blended cements (Fig. 7.14). The type of activator used also plays an important role in the effectiveness of nickel slag-based activated binders. Xu81 studied the use of blended activators (mainly mixes of Na2SiO3, Na2CO3 or Na2SO4), and found that for nickel slag with high MgO content (e.g., 15 wt.%), an increased content of Na2CO3 enhanced the compressive strength. Increasing the content of Na2SO4 in the blended activator also seemed to favor the development of good mechanical strengths. Depending on the ore from which nickel is obtained, some of the slags produced can also be enriched in other metals such as copper, which may also be extracted as a valuable commodity. Kalinkin et al.76 produced alkali-activated binders based on CuNi slags from such a process, and found that grinding of the slags increased their reactivity, particularly when a CO2 atmosphere was imposed during grinding. This increased reactivity was a consequence of both the reduced crystallinity of the slag, and the superficial modification of the particles. Enhanced release of Si and Al species from CO2-ground CuNi slags in alkaline media was identified, compared to slags ground in air.82 CO2-ground CuNi slags activated with sodium silicate reached compressive strengths of up to 50 MPa after 1 day of room temperature curing.83 Intergrinding of the slag with alkali-carbonates was also studied, and the resulting powders were mixed with water to produce binders.84 The reaction products formed in the hydrated CuNi slag/carbonate powders were comparable to those identified when alkali-activating CO2-ground CuNi slags, demonstrating the effectiveness of the intergrinding process for producing highly reactive precursors for “just add water” alkali-activated cements.

7.3.4 Zinc slag Zinc slag production follows a comparable process to that described above for copper and nickel slags. About 13% of the global production of zinc is carried out using an imperial smelting furnace process, and consequently the smelting furnace slag generation per year is limited (B1 Mt).85 The 2003 generation of imperial smelting furnace slag in some European countries was: UK 80,000 t, Italy 60,000 t, France 100,000 t, and Germany 100,000 t.85 Typical annual imperial smelting furnace slag generation in India is about 50,000 tonnes.86 The zinc slag composition varies according to the particular process used in refining. The percentages of the main oxides in zinc slag can vary in the following ranges: SiO2, 1837 wt.%; Al2O3, 817 wt.%; CaO, 1718 wt.%; Fe2O3, 1135 wt.%; MgO, 14 wt.%; Na2O, 0.40.7 wt.%; MnO, 0.81.3 wt.%; ZnO, 910 wt.%; SO3, 1.11.4 wt.%.86,87 Zinc slags have been used as supplementary cementitious materials for Portland cement replacement, after fine grinding (to 40005000 cm2/g Blaine fineness),88 and as a fine aggregate for concrete manufacture, particularly granulated leadzinc slag which has a high density (about 3860 kg/m3).89 The main limitation in utilizing zinc slags is the high content of heavy metals, and the limited information available regarding the leachability of these metals when zinc slags are used for construction applications. Alex et al.86 studied the influence of the grinding atmosphere (e.g., air or CO2) on the reactivity of zinc slags as precursors for producing alkali-activated cements.

204 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Differences in the kinetics of reaction of the activated zinc slags were identified, depending on the grinding atmosphere, so that longer induction periods were identified for the slag milled in air compared to that of slag milled in CO2. Despite the similarities between the zinc slags used, the CO2 grinding induced a higher reactivity. At all times of curing tested, alkali-activated cement based on CO2-milled slag developed a greater strength than was obtained when using an air-milled slag in the same binder mix design. No significant differences in the microstructure of the alkali-activated zinc slag cements were identified, although the cement produced with air-ground zinc slag seems to be more porous. Zinc slags have also been used in producing hybrid cements comprising 12 wt.% Portland clinker and 88 wt.% BFS plus zinc slag,87 which presented comparable performance to that of conventional Portland cements when exposed to 0.1 N HCl, sodium sulfate or seawater.

7.3.5 Manganese and silicomanganese slags Manganese slag is a by-product of ferromanganese alloy manufacturing. Two kinds of slag can be generated in the production of ferromanganese and ferromanganesesilicon alloys: air-cooled slag also known as lumpy slag, and water-cooled slag also known as granulated slag.90 Approximately 900 kg of manganese slag is generated per tonne of ferromanganese alloy produced,90 and the global production of ferromanganese alloys is shown in Fig. 7.15. The ground manganese slag has an angular shape as it results from the breakage of larger, brittle particles, and contains the mineral phases α0 -C2S, C3MS2,

Figure 7.15 Global production of ferromanganese alloy, excluding US production of ferromanganese alloys using electric arc processes as data are not publicly available. Data sourced from Feromanganese and Silicomanganese: World Production, by Country. Index mundi. http://www.indexmundi.com/en/commodities/minerals/manganese/ manganese_t9.html (accessed Apr 22, 2016).91

Alternative inorganic binders based on alkali-activated metallurgical slags

205

CaO  MnO  2SiO2 and C2AS, with the α0 -C2S in particular providing it with hydraulic reactivity.92 The fresh manganese slag is light yellow or light green, with a moisture content between 16%38% remaining from the granulation process. The packing density and true density of the slag are 700 kg/m3 and 2870 kg/m3, respectively. The manganese slag has a chemical composition of: SiO2, 1737 wt.%; Al2O3, 2127 wt.%; CaO, 1735 wt.%; Fe2O3, 0.11.2 wt.%; MnO, 521 wt.%; MgO, 79 wt.%; TiO2, 0.10.5 wt.%; Na2O, 0.21.6 wt.%; K2O, 0.11.5 wt.%; P2O5, 0.10.3 wt.%; SO3, 2.45 wt.%; and Cl, 0.12.8 wt.%.93 Two quality coefficients which are used to describe the reactivity of manganese slag, referred to as F8 and F9,94 are described in Eqs. 7.1 and 7.2: F8 ¼

C þ 0:5M þ A þ CaS S þ MnO

(7.1)

F9 ¼

C þ 0:5M þ A S þ ðMnOÞ2

(7.2)

where: C 5 CaO, M 5 MgO, A 5 Al2O3, S 5 SiO2, CaS 5 oldhamite.

Manganese slags have been assessed as potential supplementary cementitious materials for partial replacement of Portland cement, although its effectiveness is affected by the content of MnO in the slag.90,92 Pe´ra et al.93 characterized five manganese-rich slags from the ferroalloys industry, identifying that increased contents of MnO inhibited early hydration of Portland cement, and reducing the slag particle size of these slags did not notably enhance the hydraulic reactivity. Several studies have been carried out to improve the reactivity of manganese slags, mainly by mechanical grinding.95 Song et al.96 studied a wide range of alkali-activated manganese slag binders using sodium silicate and K2CO3 as activators. A high alkali dose was required to achieve a setting time that would be suitable for on-site implementation (Table 7.3), whereas the addition of potassium carbonate delayed the initial and final setting time of the paste produced. Ma et al.97 produced activated manganese slag cements, with minor additions of metakaolin to control the total MnO content (9.4 wt.% originally). A mixed sodiumpotassium silicate solution was used as activator, and the average compressive strength after 28 days is shown in Fig. 7.16. The partial replacement of manganese slag by metakaolin did not improve the compressive strength monotonically, as not only the MnO content, but also the total CaO content in the system influences the mechanical performance of these binders. Strength evolution of manganese slag-metakaolin binders was similar to that of a BFS-metakaolin binder at

206 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Setting time of alkali-activated manganese slag paste, using different types and doses of alkali activators, measured using the Vicat penetration methodology

Table 7.3

Na2O  1.6SiO2 (wt.%)

10 20 25 25 25

K2CO3 (wt.%)

0 0 0 1 2

Setting time Initial

Final

.30 h 1h 36 min 48 min 65 min

 2030 h 3.5 h 3.9 h 4.5 h

Data from Song, X.; Han, J.; Hao, Z.; Yang, M.; Sheng, G., Study on Alkali-Activated Manganese Slag Cementitious Materials. Mater. Rev. (Chongqing), 7, #22, 2009.

3 days, but the further strength gains beyond this age were not as high. Song et al.98 assessed alkali-activated cements based on blends of manganese slag with BFS (up to 30 wt.%) and 10 wt.% Portland cement, identifying an increase in the compressive strength with BFS addition. Silicomanganese (SiMn) slag is a by-product of the production of silicomanganese alloys manufactured by carbothermic reduction of manganesecontaining raw materials (a blend of minerals/ores with slags from other processes), in a submerged EAF at 16001650 C. Minor amounts of MgO-containing minerals (e.g., dolomite or olivine) may also be added.99 The amount of slag generated per tonne of SiMn metal produced is determined by the ore/slag ratio in the raw materials blend; increasing the content of FeMn slag at the expense of Mn-rich ore will lead to a larger slag/metal ratio in the SiMn process.99 It has been reported100 that for every tonne of SiMn alloy 1.21.4 tonnes of slag are produced, and so it is estimated that around 10 million tonnes of SiMn slag are produced every year. SiMn slag is typically amorphous in structure and with a glassy appearance (Fig. 7.17), and it can be described within the MnO-SiO2-CaO-Al2O3-MgO system. SiMn slag is characterized by a relatively high content of manganese oxide (about 10 wt.%) and it is usually rich in SiO2 (3843 wt.%), CaO (2530 wt.%), and Al2O3 (1215 wt.%), with lower levels of MgO (25 wt.%).101104 For BFS it has been reported93 that a high Mn content reduces hydraulic activity, affecting the development of early age binder properties (although this effect is less notable when using ultrafine slags), and therefore it has been assumed that the high content of Mn in SiMn slag might reduce its reactivity as a pozzolan. Several studies have been carried out assessing SiMn slag as a partial replacement for Portland cement,101 and the use of SiMn slag as recycled aggregate for concrete and pavement production.105 There is a good consensus that this slag has the potential to be used as a supplementary cementitious material with moderate reactivity at early age. The cements produced presented a phase assemblage generally comparable to that identified in BFS cement.

Alternative inorganic binders based on alkali-activated metallurgical slags

207

Figure 7.16 Compressive strength development of alkali-activated manganese slags (Mn_S) with 1030 wt.% replacement by metakaolin (MK) as marked. The compressive strength of blast furnace slag (BFS) blended with 30% metakaolin is shown for comparison. Adapted from Ma, S.; Chen, P. Study on Geopolymer Prepared by Manganese Slag. Concrete (Shenyang), 2008, 10, 8083.

The lower reactivity of SiMn slag at early ages of curing can be overcome by using an alkaline solution to accelerate the kinetics of dissolution of the slag and the consequent structural development of binding phases. Zhang et al.106 produced inorganic binders with 80 wt.% SiMn slag powder, 10 wt.% lime, and 10 wt.% anhydrite, with setting times comparable to those of Portland cement, and a compressive strength of 51 MPa after 28 days of curing, with the formation of a C-S-H type gel as the main reaction product. Kumar et al.107 evaluated the influence of mechanical activation of SiMn slag on its reactivity, and identified that SiMn slags

208 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 7.17 Photograph of (A) a SiMn slag glass and (B) pulverized SiMn slag.

Properties of sodium hydroxide (6 M)-activated SiMn slag binders

Table 7.4

Type of mill used for mechanical activation

Ball Attrition Vibration

Setting time (min)

Compressive strength (MPa)

Initial

Final

3 days

28 days

390 67 57

472 105 90

6 26 42

24 66 101

Data from Kumar, S.; Garcı´a-Trin˜anes, P.; Teixeira-Pinto, A.; Bao, M., Development of Alkali Activated Cement from Mechanically Activated Silico-manganese (SiMn) Slag. Cement Concrete Compos., 2013, 40, 713. doi: 10.1016/j.cemconcomp.2013.03.026.

with high specific surfaces are suitable precursors for production of alkali-activated cement. The strength development of these binders was attributed to the formation of a C-S-H type gel and a hydrotalcite type phase as main reaction products, similar to those typically identified in alkali-activated binders produced from BFS,108 along with an AFm-structured aluminate hydrate.107 A summary of the properties of the SiMn slag-based inorganic binders produced in that study is reported in Table 7.4, where it can be seen that using a vibration mill significantly enhances the reactivity of the SiMn slag, compared with a ball or attrition mill. Blended systems comprising BFS and SiMn slag have been also studied,109 as a means to enhance reactivity of the SiMn slag without mechanical activation. Provis et al.109 identified that the inclusion of up to 20 wt.% of SiMn slag did not modify the phase assemblage of BFS activated slag cements; however, minor changes in the structure of the hydrotalcite type reaction products were identified. This was

Alternative inorganic binders based on alkali-activated metallurgical slags

209

assigned to the potential substitution of Mn in the Mg and/or Al sites in these compounds,110 which suggested that Mn may be participating in the activation reaction of these blended systems. These studies are evidence that the utilization of SiMn slag is technically feasible for producing inorganic binders via alkali-activation; however its broader applicability might be limited by the moderate and often localized availability of SiMn slag around the world. Although the corrosion of steel rebars embedded in concretes containing SiMn slag has not yet been studied in detail, it is expected that the addition of this slag will alter the pore solution composition of the binders, and thus also the interfaces between steel reinforcing and the cement matrix.111 The Mn in this type of slag, which is significantly redox-active,112 can modify the redox potential of the system and thus possibly influence the corrosion process of steel reinforcement.

7.3.6 Phosphorus slag Phosphorus slag is a by-product of production of yellow phosphorus via the electric furnace method, where silica and carbon are added to remove impurities during the slagging process. About 7.5 tonnes of slags are generated during the production of one tonne of phosphorus.62 Phosphorus slag is mainly composed of calcium oxide (CaO) and silicon oxide (SiO2), with traces of minor components including 2.55 wt.% Al2O3, 0.22.5 wt.% Fe2O3, 0.53 wt.% MgO, 15 wt.% P2O5, and 02.5 wt.% F. This slag has a very low content of aluminum compared with BFS, which reduces its reactivity in pozzolanic-type processes. The chemical composition of phosphorus slags is strongly dependent on the nature of the phosphate ores used, with CaO/SiO2 ratios ranging from 0.8 to 1.2.19,113 Two main types of phosphorus slag can be produced: air-cooled and granulated phosphorus slags. The air-cooled phosphorus slag does not have cementitious properties and can be crushed and used as an aggregate. Conversely, the glass content of granulated phosphorus slag can be up to 98% as a consequence of the high viscosity of the molten slag. Shi et al.75 reported that the hydraulic activity of phosphorus slags can be determined considering their chemical composition, through a quality coefficient “K,” Eq. 7.3: K¼

CaO þ MgO þ Al2 O3 SiO2 þ P2 O5

(7.3)

High K values were associated with a greater amenability to alkaline activation. It has been reported114 that the partial replacement of silica by aluminum-rich materials during the phosphorus production promotes higher amounts of Al2O3 in the resulting slag, so that its reactivity can match that of GBFS. The presence of P2O5 in phosphorus slag has a negative effect on the early hydration of Portland cements.115 However, this is not the only factor limiting the utilization of these slags. Some phosphorus slags are more radioactive than the minimum safety levels, and consequently their handling and disposal must be consistent with that of a low

210 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

level radioactive waste. However, in China the radiation levels of the phosphorus slags are considered low, and these materials are standardized (Chinese standards JC/T 1088-2008 and JC/T 740-2006) as supplementary cementitious materials for production of blended phosphorus slag-Portland cements.19 Shi116 carried out one of the earlier studies on sodium silicate-activated phosphorus slag cements. An increased modulus of solution and activator dose reduced the setting time, independent of the content of soluble phosphorus in the slag. When using optimal activation conditions, production of mortars with up to 120 MPa compressive strength after 28 days of curing was achieved.117 The specific surface of phosphorus slags has a minor effect on its reactivity in alkali-activation, as long as the values are greater than 400 cm2/g.19 The type of activator influences the early age strength development of alkali-activated phosphorus slags; the use of NaOH as activator promotes higher early strengths, while silicate-based activators promote higher strength development at advanced times of curing. Therefore, sodium silicate solutions with a modulus between 1.2 and 1.5 are preferred for development of activated cements with improved mechanical properties.118120 The main crystalline reaction product identified in alkali-activated phosphorus slag cements is the zeolite analcime, and conversely to observations in alkali-activated BFSs121 the formation of AFm type phases is not observed, due to the low Al2O3 content of phosphorus slags.118 Some studies have demonstrated that it is possible to produce hybrid cements with high compressive strength by blending phosphorus Portland cements with minor fractions of sodium sulfate,122 calcined gypsum, or calcium sulfoaluminate, among other additives.123,124 Fig. 7.18 shows the influence of different additions on the compressive and flexural strengths of low-content (,40 wt.%) phosphorus slag cements. Compared with Portland cement based specimens, the hardened pastes containing phosphorus slag presented a reduced porosity and a smaller fraction of large pores at advanced ages of curing. Phosphorus slag-containing hybrid cements presented an improved stability compared with Portland cement specimens, when exposed to sulfate solutions. The addition of phosphorus slag also reduced the tendency toward deleterious alkaliaggregate reaction.123 Fang et al.125 also developed sodium silicate-activated phosphorus slag cements blended with FA. The replacement of phosphorus slag by FA decreased the compressive strength but increased the flexural strength slightly. The chemical and freezethaw resistance of phosphorus slag/FA cements were higher than those obtained for Portland cement, although at the cost of increased drying shrinkage. Conversely, the addition of up to 40 wt.% phosphorus slags to alkali-activated metakaolin binders led to an increased mechanical strength as a consequence of the simultaneous formation of C-A-S-H and N-A-S-H type gels, as the phosphorus slag supplied calcium in these cements.126

Alternative inorganic binders based on alkali-activated metallurgical slags

211

Figure 7.18 (A) Compressive and (B) flexural strength of alkali-activated hybrid cements containing phosphorus slag as part of the binder. C, Portland clinker; PS, phosphorus slag; G, gypsum; CG, calcined gypsum; NS, sodium sulfate; CAS, calcined alumstone.

212 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

7.4

General remarks

In general, there are three key characteristics needed for a metallurgical slag to be seen as an attractive precursor to alkali-activated cements: 1. Absence or low concentration of hazardous (radioactive or leachable toxic) species; 2. Sufficient and consistent chemical reactivity towards formation of desirable (strong, chemically and dimensionally stable) binder phases; 3. Availability in a sufficient quantity and for a long enough timeframe to enable commercially viable exploitation.

For valorization of metallurgical slags as a construction material, the control of toxicity of these industrial by-products as raw materials should always be borne in mind as a top priority. Chemical activation of metallurgical slags by alkalis may in some cases increase the leachability of toxic species including heavy metals. Also, depending on the country (or region) and purpose for which the products would be used, differences in restrictions and regulations should also be taken into account. Low hydraulicity and slow initial reaction could be the main technical challenges that most of the metallurgical slags are facing for use in binder production. However, these characteristics could be improved by mechanical activation using vibratory, ball, or attrition milling to increase the specific surface; by incorporation of other materials such as FA or metakaolin as a secondary aluminum source, or limestone as a complementary CaO source; by blending slag with minor fractions of sodium sulfate, calcined gypsum, or calcium sulfoaluminate, among other additions, to produce hybrid cements; and by controlling the grinding atmosphere of the slags to increase their reactivity. It may also be of interest to simultaneously use the fine fraction of a ground slag as the main precursor for binder formation and the coarse fraction as a mineral filler or partial substitute for sand. Among the slags discussed in this chapter, the most likely to meet this full set of criteria in general would appear to be those derived from ferrous metallurgy. However, in the specific locations where large-scale nonferrous processes are in operation, the other slags mentioned do have notable potential for utilization in alkali-activated binder production, subject to suitable characterization to ensure that the products generated are nonhazardous. These products can therefore form a useful component of the future “toolkit” of available alkali-activated binder systems, and may also offer routes to valorization of slags which are not so well suited to use in blends with Portland cement.

Acknowledgments The research leading to these results has received funding from the European Research Council under the European Union’s Seventh Framework Programme (FP/2007-2013) / ERC Grant Agreement #335928, and from the UK Engineering and Physical Sciences Research Council (EPSRC) under grant EP/M003272/1.

Alternative inorganic binders based on alkali-activated metallurgical slags

213

References 1. Bernal, S. A.; Rodrı´guez, E. D.; Kirchheim, A. P.; Provis, J. L. Management and Valorisation of Wastes through Use in Producing Alkali-Activated Cement Materials. J. Chem. Technol. Biotechnol. 2016, 91 (9), 23652388. Available from: http://dx.doi.org/ 10.1002/jctb.4927. 2. Provis, J. L.; Bernal, S. A. Geopolymers and Related Alkali-Activated Materials. Annu. Rev. Mater. Res. 2014, 44, 299327. Available from: http://dx.doi.org/10.1146/annurevmatsci-070813-113515. 3. Bernal, S. A.; Provis, J. L.; Ferna´ndez-Jime´nez, A.; Krivenko, P. V.; Kavalerova, E.; Palacios, M.; Shi, C. Binder Chemistry  High-Calcium Alkali-Activated Materials. In Alkali Activated Materials, State of the Art Report of RILEM TC 224-AAM; Provis, J. L., van Deventer, J. S. J., Eds.; RILEM/Springer: Dordrecht, 2014; pp 5991. Available from: http://dx.doi.org/10.1007/978-94-007-7672-2. 4. Bernal, S. A.; Mejı´a de Gutie´rrez, R.; Pedraza, A. L.; Provis, J. L.; Rodriguez, E. D.; Delvasto, S. Effect of Binder Content on the Performance of Alkali-Activated Slag Concretes. Cement Concrete Res. 2011, 41 (1), 18. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2010.08.017. 5. Buchwald, A.; Vanooteghem, M.; Gruyaert, E.; Hilbig, H.; De Belie, N. Purdocement: Application of Alkali-Activated Slag Cement in Belgium in the 1950s. Mater. Struct. 2015, 48 (12), 501511. Available from: http://dx.doi.org/10.1617/s11527-013-0200-8. 6. Provis, J. L.; Brice, D. G.; Buchwald, A.; Duxson, P.; Kavalerova, E.; Krivenko, P. V.; Shi, C.; van Deventer, J. S.; Wiercx, J. H. Demonstration Projects and Applications in Building and Civil Infrastructure. In Alkali Activated Materials, State of the Art Report of RILEM TC 224-AAM; Provis, J. L., van Deventer, J. S. J., Eds.; RILEM/Springer: Dordrecht, 2014; pp 309338. Available from: http://dx.doi.org/10.1007/978-94-0077672-2_11. 7. Lothenbach, B.; Scrivener, K.; Hooton, R. D. Supplementary Cementitious Materials. Cement Concrete Res. 2011, 41 (12), 12441256. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2010.12.001. 8. Shi, C. Steel Slag  Its Production, Processing, Characteristics and and Cementitious Properties. J. Mater. Civ. Eng. 2004, 16 (3), 230236. Available from: http://dx.doi.org/ 10.1002/chin.200522249. 9. Provis, J. L.; Palomo, A.; Shi, C. Advances in Understanding Alkali-Activated Materials. Cement Concrete Res. 2015, 78 (A), 110125. Available from: http://dx.doi. org/10.1016/j.cemconres.2015.04.013. 10. Horii, K.; Tsutsumi, N.; Kitano, Y.; Kato, T. Processing and Reusing Technologies for Steelmaking Slag. Nippon Steel Techn. Rep. 2013, 104, 123129. 11. van Oss, H. G. Slag-Iron and Steel. Department of the Interior and US Geological Survey Minerals Yearbook; USGS: Reston, VA, 2003. 12. Setie´n, J.; Herna´ndez, D.; Gonza´lez, J. J. Characterization of Ladle Furnace Basic Slag for Use as a Construction Material. Constr. Build. Mater. 2009, 23 (5), 17881794. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2008.10.003. 13. Jonczy, I. Variability of Chemical Composition of Metallurgical Slags After Steel Production. RMZ Materials and Geoenvironment; Materiali in Geookolje 2013, 60 (4), 263270. 14. EUROSLAG. Statistics 2012. http://www.euroslag.com/PRODUCTS/STATISTICS/ 2012/ (accessed Apr 11, 2016).

214 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

15. Standardization Administration of the People’s Republic of China. GB/T 20491-2006, Steel slag powder used for cement and concrete. 2006. p 9. 16. Li, G., Slag Valorisation in China: An Overview, In: 1st Slag Valorisation Symposium, 2009, Leuven, Belgium. pp 165176, 2009. 17. Yi, H.; Xu, G.; Cheng, H.; Wang, J.; Wan, Y.; Chen, H. An Overview of Utilization of Steel Slag. Procedia Environ. Sci. 2012, 16, 791801. Available from: http://dx.doi.org/ 10.1016/j.proenv.2012.10.108. 18. Shi, C.; Wu, X.; Tang, M. Research on Alkali-Activated Cementitious Systems in China: A Review. Adv. Cement Res. 1993, 5 (17), 17. Available from: http://dx.doi. org/10.1680/adcr.1993.5.17.1. 19. Shi, C.; Qian, J. High Performance Cementing Materials from Industrial Slags - A Review. Resour. Conser. Recy. 2000, 29 (3), 195207. Available from: http://dx.doi. org/10.1016/S0921-3449(99)00060-9. 20. Hu, S.; Wang, H.; Zhang, G.; Ding, Q. Bonding and Abrasion Resistance of Geopolymeric Repair Material Made with Steel Slag. Cement Concrete Compos. 2008, 30 (3), 239244. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2007. 04.004. 21. Kourounis, S.; Tsivilis, S.; Tsakiridis, P. E.; Papadimitriou, G. D.; Tsibouki, Z. Properties and Hydration of Blended Cements with Steelmaking Slag. Cement Concrete Res. 2007, 37 (6), 815822. Available from: http://dx.doi.org/10.1016/j. cemconres.2007.03.008. 22. Tang, M. An Investigation on Mineral Composition of Steel Slag for Cement Production. Research Report; Nanjing Institute of Chemical Technology: Nanjing, 1973. 23. Shi, C. Characteristics and Cementitious Properties of Ladle Slag Fines from Steel Production. Cement Concrete Res. 2002, 32 (3), 459462. Available from: http://dx.doi. org/10.1016/S0008-8846(01)00707-4. 24. Natali Murri, A.; Rickard, W. D. A.; Bignozzi, M. C.; van Riessen, A. High Temperature Behaviour of Ambient Cured Alkali-Activated Materials based on Ladle Slag. Cement Concrete Res. 2013, 43, 5161. Available from: http://dx.doi.org/10.1016/ j.cemconres.2012.09.011. 25. Bignozzi, M. C.; Manzi, S.; Lancellotti, I.; Kamseu, E.; Barbieri, L.; Leonelli, C. Mixdesign and Characterization of Alkali Activated Materials based on Metakaolin and Ladle Slag. Appl. Clay Sci. 2013, 73, 7885. Available from: http://dx.doi.org/10.1016/ j.clay.2012.09.015. 26. Yusof, M.A.W. Investigating the Potential for Incorporating Tin Slag in Road Pavements. Thesis (M.Sc.) University of Nottingham, 2005. 27. Motz, H.; Geiseler, J. Products of Steel Slags an Opportunity to Save Natural Resources. Waste Manage. 2001, 21 (3), 285293. Available from: http://dx.doi.org/10.1016/ S0956-053X(00)00102-1. 28. Nikoli´c, I.; Drinˇci´c, A.; Djurovi´c, D.; Karanovi´c, L.; Radmilovi´c, V. V.; Radmilovi´c, V. R. Kinetics of Electric Arc Furnace Slag Leaching in Alkaline Solutions. Constr. Build. Mater. 2016, 108, 19. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2016.01.038. 29. Wang, Q.; Yang, J. W.; Yan, P. Y. Influence of Initial Alkalinity on the Hydration of Steel Slag. Sci. China Technol. Sci. 2012, 55 (12), 33783387. Available from: http:// dx.doi.org/10.1007/s11431-012-4830-9. 30. Maragkos, I.; Giannopoulou, I. P.; Panias, D. Synthesis of Ferronickel Slag-based Geopolymers. Miner. Eng. 2009, 22 (2), 196203. Available from: http://dx.doi.org/ 10.1016/j.mineng.2008.07.003.

Alternative inorganic binders based on alkali-activated metallurgical slags

215

31. Balomenos, E.; Panias, D., Iron Recovery and Production of High Added Value Products from the Metallurgical by-products of Primary Aluminum and Ferro-nickel Industries, 3rd International Slag Valorisation Symposium, Leuven, Belgium. pp 161172, 2013. 32. Kirillidi, Y.; Frogoudakis, E., Electric Arc Furnace Slag Utilization. Proceedings of the 9th International Conference on Environmental Science and Technology, Rhodes, Greece, 2005. 33. Fidancevska, E.; Vassilev, V.; Milosevski, M.; Parvanov, S.; Milosevski, D.; Aljihmani, L. Composites based on Industrial Wastes III. Production of Composites of Fe-Ni Slag and Waste Glass. J. Univ. Chem. Technol. Metall. 2007, 42 (3), 285290. 34. Vassilev, V.; Fidancevska, E.; Milosevski, M.; Parvanov, S.; Milosevski, D.; HristovaVasileva, T. Composites based on Industrial Wastes IV. Production of Porous Composites from Fe-Ni Slag and Waste Glass. J. Univ. Chem. Technol. Metall. 2007, 42 (4), 369376. 35. Komnitsas, K.; Zaharaki, D.; Perdikatsis, V. Geopolymerisation of Low Calcium Ferronickel Slags. J. Mater. Sci. 2007, 42 (9), 30733082. Available from: http://dx.doi. org/10.1007/s10853-006-0529-2. 36. Diaz, C.; Landolt, C.; Vahed, A.; Warner, A.; Taylor, J. A Review of Nickel Pyrometallurgical Operations. JOM 1988, 40 (9), 2833. Available from: http://dx.doi. org/10.1007/s10853-006-0529-2. 37. Golubnitchy, A. V.; Koksharev, V. V. Heat Resistant Slag Alkaline Binder based on Granulated Slag of Ferronickel Production. In Proceedings of the First International Conference on Alkaline Cements and Concretes; Krivenko, P. V., Ed.; VIPOL Stock Company: Kiev, Ukraine, 1994; pp 371374. 38. The Ukrainian Cooperative State Enterprise of Agroindustrial Construction. TU 55910.20-001-90. Technical Specifications “A slag alkaline binder from ferronickel granulated slag”. Zhitomir, 1990. 39. Komnitsas, K.; Zaharaki, D.; Perdikatsis, V. Effect of Synthesis Parameters on the Compressive Strength of Low-calcium Ferronickel Slag Inorganic Polymers. J. Hazard. Mater. 2009, 161 (23), 760768. Available from: http://dx.doi.org/10.1016/j. jhazmat.2008.04.055. 40. Zaharaki, D.; Komnitsas, K.; Perdikatsis, V. Use of Analytical Techniques for Identification of Inorganic Polymer Gel Composition. J. Mater. Sci. 2010, 45 (10), 27152724. Available from: http://dx.doi.org/10.1007/s10853-010-4257-2. 41. Sakkas, K.; Nomikos, P.; Sofianos, A.; Panias, D. Utilisation of FeNi-Slag for the Production of Inorganic Polymeric Materials for Construction or for Passive Fire Protection. Waste Biomass Valorization 2013, 5 (3), 403410. Available from: http://dx. doi.org/10.1007/s12649-013-9278-z. 42. Sahu, K. K.; Alex, T. C.; Mishra, D.; Agrawal, A. An Overview on the Production of Pigment Grade Titania from Titania-rich slag. Waste Manage. Res. 2006, 24 (1), 7479. Available from: http://dx.doi.org/10.1177/0734242X06061016. 43. Chen, J.-X.; Chen, H.-B.; Xiao, P.; Zhang, L.-F. A Study on Complex Alkali-Slag Environmental Concrete. Proceedings of the International Workshop on Sustainable Development and Concrete Technology; Center for Transportation Research and Education: Beijing, China, May 2021, 2004. 44. Wang, J. Study on the Use of High Titaniferous Blast-Furnace Slag for Preparing Wall Bricks. Applied Chemistry; Shenyang University of Chemical Technology: China, 2004. ¨ .; Pontikes, Y.; Snellings, R.; Dijkman, J.; Sels, B.; Vandewalle, 45. Salman, M.; Cizer, O L.; Blanpain, B.; van Balen, K. Alkali Activation of AOD Stainless Steel Slag Under

216 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

46.

47.

48.

49.

50. 51.

52.

53. 54.

55.

56.

57.

58.

59.

Steam Curing Conditions. J. Am. Ceram. Soc. 2015, 98 (10), 30623074. Available from: http://dx.doi.org/10.1111/jace.13776. Kim, Y. J.; Nettleship, I.; Kriven, W. M. Phase Transformations in Dicalcium Silicate: II, TEM Studies of Crystallography, Microstructure, and Mechanisms. J. Am. Ceram. Soc. 1992, 75 (9), 24072419. Available from: http://dx.doi.org/10.1111/j.11512916.1992.tb05593.x. ¨ .; Pontikes, Y.; Snellings, R.; Vandewalle, L.; Blanpain, B.; Balen, Salman, M.; Cizer, O K. V. Investigating the Binding Potential of Continuous Casting Stainless Steel Slag by Alkali Activation. Adv. Cement Res. 2014, 26 (5), 256270. Available from: http://dx. doi.org/10.1680/adcr.13.00018. ¨ .; Pontikes, Y.; Snellings, R.; Vandewalle, L.; Blanpain, B.; Balen, Salman, M.; Cizer, O K. V. Cementitious Binders from Activated Stainless Steel Refining Slag and the Effect of Alkali Solutions. J. Hazard. Mater. 2015, 286, 211219. Available from: http://dx. doi.org/10.1016/j.jhazmat.2014.12.046. ¨ .; Pontikes, Y.; Vandewalle, L.; Blanpain, B.; van Balen, K. Effect Salman, M.; Cizer, O of Curing Temperatures on the Alkali Activation of Crystalline Continuous Casting Stainless Steel Slag. Constr. Build. Mater. 2014, 71, 308316. Available from: http:// dx.doi.org/10.1016/j.conbuildmat.2014.08.067. Shen, H.; Forssberg, E. An Overview of Recovery of Metals from Slags. Waste Manage. 2003, 23 (10), 933949. Available from: http://dx.doi.org/10.1016/S0956-053X(02)00164-2. User Guidelines for Waste and Byproduct Materials in Pavement Construction. U.S. Federal Highway Administration. http://www.fhwa.dot.gov/publications/research/infrastructure/structures/97148/nfs1.cfm (accessed Mar 09, 2016). Kreusch, M.; Ponte, M.; Ponte, H.; Kaminari, N.; Marino, C.; Mymrin, V. Technological Improvements in Automotive Battery Recycling. Resour. Conserv. Recy. 2007, 52 (2), 368380. Available from: http://dx.doi.org/10.1016/j.resconrec.2007.05.004. International Lead Association. Lead Production & Statistics. http://www.ila-lead.org/ lead-facts/lead-production--statistics (accessed Apr 14, 2016). Piatak, N. M.; Parsons, M. B.; Seal, R. R. Characteristics and Environmental Aspects of Slag: A Review. Appl. Geochem. 2015, 57, 236266. Available from: http://dx.doi.org/ 10.1016/j.apgeochem.2014.04.009. Ogundiran, M. B.; Nugteren, H. W.; Witkamp, G. J. Immobilisation of Lead Smelting Slag within Spent Aluminate—Fly Ash based Geopolymers. J. Hazard. Mater. 2013, 248249, 2936. Available from: http://dx.doi.org/10.1016/j. jhazmat.2012.12.040. Onisei, S.; Pontikes, Y.; van Gerven, T.; Angelopoulos, G. N.; Velea, T.; Predica, V.; Moldovan, P. Synthesis of Inorganic Polymers Using Fly Ash and Primary Lead Slag. J. Hazard. Mater. 2012, 205206, 101110. Available from: http://dx.doi.org/10.1016/j. jhazmat.2011.12.039. Palomo, A.; Palacios, M. Alkali-activated Cementitious Materials: Alternative Matrices for the Immobilisation of Hazardous Wastes: Part II. Stabilisation of Chromium and Lead. Cement Concrete Res. 2003, 33 (2), 289295. Available from: http://dx.doi.org/ 10.1016/S0008-8846(02)00964-X. Perera, D. S.; Aly, Z.; Vance, E. R.; Mizumo, M. Immobilization of Pb in a Geopolymer Matrix. J. Am. Ceram. Soc. 2005, 88 (9), 25862588. Available from: http://dx.doi.org/ 10.1111/j.1551-2916.2005.00438.x. Shi, C.; Meyer, C.; Behnood, A. Utilization of Copper Slag in Cement and Concrete. Resour. Conserv. Recy. 2008, 52 (10), 11151120. Available from: http://dx.doi.org/ 10.1016/j.resconrec.2008.06.008.

Alternative inorganic binders based on alkali-activated metallurgical slags

217

60. Gorai, B.; Jana, R. K.; Premchand. Characteristics and Utilisation of Copper Slag—A Review. Resour. Conserv. Recy. 2003, 39 (4), 299313. Available from: http://dx.doi. org/10.1016/S0921-3449(02)00171-4. 61. Copper. Why We Belive in Copper’s Long-term Value. Oracle Mining Corp. http:// www.oracleminingcorp.com/copper/ (accessed Apr 20, 2016). 62. Reuter, M.; Xiao, Y.; Boin, U. Recycling and Environmental Issues of Metallurgical Slags and Salt Fluxes. VII International Conference on Molten Slags, Fluxes and Salts; The South African Institute of Mining and Metallurgy: South Africa, 2004. 63. Khanzadi, M.; Behnood, A. Mechanical Properties of High-strength Concrete Incorporating Copper Slag as Coarse Aggregate. Constr. Build. Mater. 2009, 23 (6), 21832188. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2008.12.005. 64. Moura, W. A.; Gonc¸alves, J. P.; Lima, M. B. L. Copper Slag Waste as a Supplementary Cementing Material to Concrete. J. Mater. Sci. 2007, 42 (7), 22262230. Available from: http://dx.doi.org/10.1007/s10853-006-0997-4. 65. Yao, Y.; Wang, J.; Wang, M.; Wen, S.; Gao, X. A Method for Utilizing Copper Slag to Prepare Nuclear Cement. China Patent: CN104129933 B, 2015. 66. Pan, W.; Wang, T., Activated Copper Slag Composition and Gelling Properties. South China University of Technology (Natural Science Edition), 1, 1993, Guangzhou, China. 67. Deja, J.; Małolepszy, J., Long-term Resistance of Alkali-activated Slag Mortars to Chloride Solution. 3rd international conference on durability of concrete. Nice, France, 1994. 68. Onisei, S.; Lesage, K.; Blanpain, B.; Pontikes, Y. Early Age Microstructural Transformations of an Inorganic Polymer Made of Fayalite Slag. J. Am. Ceram. Soc. 2015, 98 (7), 22692277. Available from: http://dx.doi.org/10.1111/jace.13548. 69. Nazer, A.; Paya´, J.; Borrachero, M. V.; Monzo´, J. Use of Ancient Copper Slags in Portland Cement and Alkali Activated Cement Matrices. J. Environ. Manage. 2016, 167, 115123. Available from: http://dx.doi.org/10.1016/j.jenvman.2015.11.024. 70. Ahmari, S.; Parameswaran, K.; Zhang, L. Alkali Activation of Copper Mine Tailings and Low-calcium Flash-furnace Copper Smelter Slag. J. Mater. Civ. Eng. 2014, 27 (6). 04014193-1-04014193-11, http://dx.doi.org/10.1061/(ASCE)MT-1943-5533.0001159. 71. Wang, H.; Deng, H.-B.; Lu, X.-F. Important Laterite Nickel Ore Resources in the World: Present Situation and Exploitation.. Gansu Metall. 2009, 1, 6. 72. Sheng, G.; Zhai, J. Making Metallurgical Slag from Nickel Industry a Resource.. Metal Mine 2005, 10 (10), 6871. 73. Ma, M.; Ni, W.; Wang, Y. Crystallization Kinetics and Process of the Glass-Ceramic Produced by Nickel Slag. J. Univ. Sci. Technol. Beijing 2007, 29 (2), 168172. 74. Papadakis, V. G.; Tsimas, S. Greek Supplementary Cementing Materials and their Incorporation in Concrete. Cement Concrete Compos. 2005, 27 (2), 223230. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2004.02.011. 75. Shi, C.; Krivenko, P. V.; Roy, D. M. Alkali-Activated Cements and Concretes; CRC Press: Abingdon, 2006. 76. Kalinkin, A. M.; Kumar, S.; Gurevich, B. I.; Alex, T. C.; Kalinkina, E. V.; Tyukavkina, V. V.; Kalinnikov, V. T.; Kumar, R. Geopolymerization Behavior of CuNi Slag Mechanically Activated in Air and in CO2 Atmosphere. Int. J. Miner. Process. 2012, 112113, 101106. Available from: http://dx.doi.org/10.1016/j.minpro.2012.05.001. 77. Emery, J. J. Slag Utilization in Pavement Construction. ASTM Special Technical Publication 774; ASTM: Philadelphia, 1982. 78. Bin, X.; Yuan, X., Research of Alkali-Activated Nickel Slag Cement, Proceedings of the Second International Conference on Alkaline Cements and Concretes. Kiev, Ukraine. pp 531536, 1999.

218 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

79. Bin, X.; Yuan, X., Research of Building Block Made from Nickel Slag, Proceedings of the Second International Conference on Alkaline Cements and Concretes. Kiev, Ukraine. pp 712716, 1999. 80. Yang, T.; Yao, X.; Zhang, Z. Geopolymer Prepared with High-magnesium Nickel Slag: Characterization of Properties and Microstructure. Constr. Build. Mater. 2014, 59, 188194. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.01.038. 81. Xu, B., Alkali-excited Nickel Slag Cement and Its Production Process and Use in China. China Patent CN 1197773 A, 1998. 82. Kalinkina, E. V.; Gurevich, B. I.; Kalinkin, A. M.; Mazukhina, S. I. Interaction of Magnesia-ferriferous Slag with Sodium Hydroxide Solutions: Experimental and Physicochemical Modeling. Russ. J. Appl. Chem. 2015, 88 (7), 11271133. Available from: http://dx.doi.org/10.1134/S1070427215070046. 83. Kalinkin, A. M.; Kumar, S.; Gurevich, B. I.; Kalinkina, E. V.; Tyukavkina, V. V. Synthesis of Geopolymer Materials based on Slags of Nonferrous Metallurgy with the Use of Mechanoactivation. Glass Phys. Chem. 2014, 40 (1), 2630. Available from: http://dx.doi.org/10.1134/s1087659614010088. 84. Kalinkina, E. V.; Gurevich, B. I.; Kalinkin, A. M.; Mazukhina, S. I.; Tykavkina, V. V.; Zalkind, O. A. Binding Properties of Ferromagnesian Slags After Mechanical Activation with Alkaline-earth Carbonates. Inorg. Mater. 2014, 50 (11), 11791184. Available from: http://dx.doi.org/10.1134/s0020168514110065. 85. Morrison, C.; Hooper, R.; Lardner, K. The Use of Ferro-silicate Slag from ISF Zinc Production as a Sand Replacement in Concrete. Cement Concrete Res. 2003, 33 (12), 20852089. Available from: http://dx.doi.org/10.1016/S0008-8846(03)00234-5. 86. Alex, T. C.; Kalinkin, A. M.; Nath, S. K.; Gurevich, B. I.; Kalinkina, E. V.; Tyukavkina, V. V.; Kumar, S. Utilization of Zinc Slag through Geopolymerization: Influence of Milling Atmosphere. Int. J. Miner. Process. 2013, 123, 102107. Available from: http:// dx.doi.org/10.1016/j.minpro.2013.06.001. 87. Fernandez-Jimenez, A.; Flores, E.; Maltseva, O.; Garcia-Lodeiro, I.; Palomo, A. Hybrid Alkaline Cements. Part III. Durability and Industrial Applications. Rom. J. Mater. 2013, 43 (2), 195200. 88. Atzeni, C.; Massidda, L.; Sanna, U. Use of Granulated Slag from Lead and Zinc Processing in Concrete Technology. Cement Concrete Res. 1996, 26 (9), 13811388. Available from: http://dx.doi.org/10.1016/0008-8846(96)00121-4. 89. Alwaeli, M. Application of Granulated LeadZinc Slag in Concrete as an Opportunity to Save Natural Resources. Radiat. Phys. Chem. 2013, 83, 5460. Available from: http://dx.doi.org/10.1016/j.radphyschem.2012.09.024. 90. Rai, A.; Prabakar, J.; Raju, C. B.; Morchalle, R. K. Metallurgical Slag as a Component in Blended Cement. Constr. Build. Mater. 2002, 16 (8), 489494. Available from: http://dx.doi.org/10.1016/S0950-0618(02)00046-6. 91. Feromanganese and Silicomanganese: World Production, by Country. Index mundi. http://www.indexmundi.com/en/commodities/minerals/manganese/manganese_t9.html (accessed Apr 22, 2016). 92. Liu, R.-J.; Ding, Q.-J.; Chen, P.; Yang, G.-Y. Durability of Concrete Made with Manganese Slag as Supplementary Cementitious Materials. J. Shanghai Jiaotong Univ. 2012, 17 (3), 345349. Available from: http://dx.doi.org/10.1007/s12204-012-1284-y. 93. Pe´ra, J.; Ambroise, J.; Chabannet, M. Properties of Blast-Furnace Slags Containing High Amounts of Manganese. Cement Concrete Res. 1999, 29 (2), 171177. Available from: http://dx.doi.org/10.1016/S0008-8846(98)00096-9.

Alternative inorganic binders based on alkali-activated metallurgical slags

219

94. Smolczyk, H., Slag Structure and Identification of Slags. Proceedings of the 7th International Congress on the Chemistry of Cement, Paris, France, 1980. 95. Gao, Z.; Song, X.; Han, J. Study on Physical Properties and Mechanical-Chemistry of Mechanically Ground Manganese Slag. Mater. Rev. (Chongqing) 2010, 8 (#25). 96. Song, X.; Han, J.; Hao, Z.; Yang, M.; Sheng, G. Study on Alkali-Activated Manganese Slag Cementitious Materials. Mater. Rev. (Chongqing) 2009, 7 (#22). 97. Ma, S.; Chen, P. Study on Geopolymer Prepared by Manganese Slag. Concrete (Shenyang) 2008, 10, 8083. 98. Song, X.; Yang, M.; Han, J.; Hao, Z. Research on the Mechanical Property and Hydration Process of Alkali-activated Manganese Slag and Slag Cementitious Materials. China Concrete Cement Prod. 2010, 3, 912. 99. Olsen, S. E.; Tangstad, M. Silicomanganese Production  Process Understanding. Tenth International Ferroalloys Congress, INFACON X ‘Transformation through technology’; Document Transformation Technologies: Cape Town, South Africa, 2004. 100. Market Survey on Manganese Ore. Indian Bureau of Mines. http://ibm.nic.in/index. php?c=pages&m=index&id=504 (accessed Mar 21, 2016). 101. Frı´as, M.; Sanchez de Rojas, M. I.; Rodrı´guez, C. The Influence of SiMn Slag on Chemical Resistance of Blended Cement Pastes. Constr. Build. Mater. 2009, 23 (3), 14721475. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2008.06.012. 102. Allahverdi, A.; Ahmadnezhad, S. Mechanical Activation of Silicomanganese Slag and Its Influence on the Properties of Portland Slag Cement. Powder Technol. 2014, 251, 4151. Available from: http://dx.doi.org/10.1016/j.powtec.2013.10.023. 103. Ringdalen, E.; Gaal, S.; Tangstad, M.; Ostrovski, O. Ore Melting and Reduction in Silicomanganese Production. Metall. Mater. Trans. B 2010, 41 (6), 12201229. Available from: http://dx.doi.org/10.1007/s11663-010-9350-z. 104. Silicomanganese. Grupo FerroAtlantica. http://www.ferroatlantica.es/index.php/en/productos-fav/ferrosiliciofv (accessed Mar 21, 2016). ˇ ˇ Silikomanganska troska 105. Jadrijevi´c, A.; Bermanec, V.; Orˇsuli´c, D.; Zigoveˇ cki Gobac, Z. kao agregat za asfalt i beton. Graðevinar 2011, 63 (5), 441447. 106. Zhang, X.-F.; Ni, W.; Wu, J.-Y.; Zhu, L.-P. Hydration Mechanism of a Cementitious Material Prepared with Si-Mn Slag. Int. J. Miner. Metall. Mater. 2011, 18 (2), 234239. Available from: http://dx.doi.org/10.1007/s12613-011-0428-7. 107. Kumar, S.; Garcı´a-Trin˜anes, P.; Teixeira-Pinto, A.; Bao, M. Development of Alkali activated Cement from Mechanically Activated Silico-manganese (SiMn) Slag. Cement Concrete Compos. 2013, 40, 713. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2013.03.026. 108. Bernal, S. A.; Provis, J. L.; Walkley, B.; San Nicolas, R.; Gehman, J. D.; Brice, D. G.; Kilcullen, A.; Duxson, P.; van Deventer, J. S. J. Gel Nanostructure in Alkali-activated Binders based on Slag and Fly Ash, and Effects of Accelerated Carbonation. Cement Concrete Res. 2013, 53, 127144. Available from: http://dx.doi.org/10.1016/j. cemconres.2013.06.007. 109. Provis, J.L.; Hanafi, R.; Garcia-Trin˜anes, P., Valorisation of Wastes by AlkaliActivation-Progress, Opportunities and Pitfalls, 4th International Slag Valorisation Symposium, Leuven, Belgium, 2014. 110. Mills, S.; Christy, A.; Ge´nin, J.-M.; Kameda, T.; Colombo, F. Nomenclature of the Hydrotalcite Supergroup: Natural Layered Double Hydroxides. Mineral. Mag. 2012, 76 (5), 12891336. Available from: http://dx.doi.org/10.1180/minmag.2012.076.5.10.

220 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

111. Lee, W. E.; Zhang, S. Direct and Indirect Slag Corrosion of Oxide and Oxide-c Refractories. VII International Conference on Molten Slags, Fluxes and Salts; The South African Institute of Mining and Metallurgy: Cape Town, South Africa, 2004. 112. Bru¨ckner, A.; Lu¨ck, R.; Wieker, W.; Winkler, A.; Andreae, C.; Mehner, H. Investigation of Redox Reactions Proceeding During the Hardening Process of Sulfide Containing Cement. Cement Concrete Res. 1992, 22 (6), 11611169. Available from: http://dx.doi.org/10.1016/0008-8846(92)90045-W. 113. Chen, X.; Zeng, L.; Fang, K. Anti-crack Performance of Phosphorus Slag Concrete. Wuhan Univ. J. Nat. Sci. 2009, 14 (1), 8086. Available from: http://dx.doi.org/ 10.1007/s11859-009-0117-9. 114. Wu, X. A Study on Mineral Phases in Phosphorus Slag. J. Yinnan Build. Mater. 1984, 4, 514. 115. Shi, C.; He, F.; Jime´nez, A. F.; Krivenko, V. P.; Palomo, A. Classification and Characteristics of Alkali-activated Cements. J. Chinese Ceram. Soc. 2012, 40 (1), 6975. http://www.cnki.net/kcms/detail/11.2310.TQ.20111229.1907.011.html. 116. Shi, C. Study on Alkali Activated Phosphorous Slag Cement. J. Nanjing Inst. Chem. Technol. 1988, 10 (2), 110116. 117. Shi, C. Influence of Temperature on Hydration of Alkali Activated Phosphorous Slag. J. Nanjing Inst. Chem. Technol. 1989, 11 (1), 9499. 118. Chen, L.; Zhu, C.; Sheng, G. Mechanical Properties and Microstructures of Alkaliactivated Phosphorous Slag Cement. J. Chin. Ceram. Soc. 2006, 34 (5), 604609. 119. Chen, Y.; Fang, Y.; Jia, L.; Lu, F. Hardened Paste of Alkali-activated Phosphorous Slag Cement and Structure of Paste-Aggregate Interface. J. Hohai Univ. (Nat. Sci.) 2010, 38 (1), 7275. 120. Fang, Y.; Wang, C.; Zhu, D.; Li, Z. Performance of Alkali-activated Phosphor Slag-fly Ash Cementitious Material. J. Hohai Univ. (Nat. Sci.) 2007, 35 (5), 549552. 121. Ke, X.; Bernal, S. A.; Provis, J. L. Controlling the Kinetics of Reaction of Sodium Carbonate-activated Slag Cements Using Calcined Layered Double Hydroxides. Cement Concrete Res. 2015, 81, 2437. Available from: http://dx.doi.org/10.1016/j. cemconres.2015.11.012. 122. Shi, C.; Tang, X.; Li, Y. Studies on the Activation of Phosphorus Slag. Proceedings of the Third International Conference on the Use of Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, supplementary volume. American Concrete Institute SP-114: Norway, 1989. 123. Li, D.; Shen, J.; Mao, L.; Wu, X. The influence of Admixtures on the Properties of Phosphorous Slag Cement. Cement Concrete Res. 2000, 30 (7), 11691173. Available from: http://dx.doi.org/10.1016/S0008-8846(00)00291-X. 124. Li, D.; Shen, J.; Chen, L.; Wu, X. The Influence of Fast-setting/Early-strength Agent on High Phosphorous Slag Content Cement. Cement Concrete Res. 2001, 31 (1), 1924. Available from: http://dx.doi.org/10.1016/S0008-8846(00)00442-7. 125. Fang, Y.; Mao, Z.; Wang, C.; Zhu, Q. Performance of Alkali-activated Phosphor Slagfly Ash Cement and the Microstructure of Its Hardened Paste. J. Chin. Ceram. Soc. 2007, 35 (4), 451455. 126. Soleimani, M.; Naghizadeh, R.; Mirhabibi, A.; Golestanifard, F. The Influence of Phosphorus Slag Addition on Microstructure and Mechanical Properties of Metakaolinbased Geopolymer Pastes. Ceramics-Silika´ty 2013, 57 (1), 3338.

The potential use of geopolymer for cleaning air

8

Jose´ Ramo´n Gasca-Tirado1, Alejandro Manzano-Ramı´rez2 and Jose´ Luis Reyes-Araiza3 1 Universidad de Guanajuato, Celaya, Me´xico, 2Centro de Investigaciones y Estudios Avanzados del I.P.N., Quere´taro, Qro, Me´xico, 3Universidad Auto´noma de Quere´taro, DIPFI, Fac. de Ingenierı´a, Qro, Me´xico

8.1

Introduction

Geopolymers have been used in different applications, most of them are for construction and cementitious material. Nevertheless, some researchers have found new applications for these materials. Some of them are in fire protection,1 immobilizations of waste and toxic materials,2 and radioactive waste encapsulation.3 Recently, geopolymers have been used as support material for optical applications: color holder, color pH indicator, photoluminiscent material, and for volatile organic degradation by photocatalysis for instance.4,5 This research has been focused on incorporating titanium dioxide (TiO2) in the geopolymers with the aim of degrading volatile organic compounds present in the atmosphere. In this way, it is visualized that geopolymer will be part of future buildings by playing an important role in environment remediation.

8.2

What make geopolymers good for this kind of applications?

Geopolymers are environmentally friendly materials synthesized at low temperatures. In comparison with the production of cements, the amount of energy required for their synthesis and the volume of CO2 exhausted is quite low. Their synthesis is based on the reaction between aluminosilicates and polysilicates in an alkaline environment6 and their structure is balanced by cations as Al1, K1, Mg21, and others which can be ion exchanged. Because of their chemical composition, they are considered analogues of zeolites but with an amorphous structure.7,8 This similarity is an important issue that suggests they can be employed for similar aims. Studies have been undertaken where zeolites have been used for the reduction of NO,9 CO2,10 and methyl blue degradation, for example.11 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00009-7 © 2017 Elsevier Ltd. All rights reserved.

222 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

The aim of this study therefore is to find ways to modify geopolymers in such a way that they can degrade organic compounds by using TiO2 in the form of anatase. This anatase is a well-known semiconductor with photocatalytic activity.12,13

8.3

Optical characterization of geopolymers

Two sets of geopolymers were synthesized at three different temperatures (40 C, 60 C, and 90 C) to determine the amount of light transmitted, absorbed, and remitted at different wavelengths.14 The starting materials employed in all samples were metakaolin (MetaMax from BASF Corporation), sodium hydroxide, and sodium silicate, from SIDESA Corporation Mexico. Samples were prepared by mixing metakaolin, sodium hydroxide, sodium silicate, and distilled water to follow fixed molar oxide ratios: SiO2/Al2O3 5 3.3, Na2O/SiO2 5 0.25, Na2O/Al2O3 5 0.48, and H2O/Na2O 5 13.73. The first set of samples (40 C, 60 C, and 90 C), each 5 mm thick, was obtained by pouring the mixture in petri dishes and cut with a diamond disc. The second set of samples, at similar temperatures, was synthesized by pouring the mixture between two microscope slides (separated by 600 µm polyvinyl chloride (PVC) slides). One microscope slide was coated with 50 µm high absorbing black paint and the other with a water-based releasing agent to finally have a 550 µm thick geopolymer (Fig. 8.1). All samples were analyzed by measuring the scatter reflectance with a Perkin Elmer Lambda 900 UV/Vis/NIR spectrophotometer. The scatter reflectance of thick (5 mm) and thin (550 µm) samples was measured at different wavelengths (400800 nm) and the fraction of light transmitted, was calculated through Dahm & Dahm’s equation (Eq. 8.1).15 For Dahm and Dahm,15 a sample is regarded to have a fixed number of representative layers and this mathematical model (“Absorption/Remission function”)

Figure 8.1 Geopolymer (thickness: 550 µm). A 600 µm PVC is observed on the left of the figure. The geopolymer is in the middle over a 50 µm high absorbing black paint. Right, a section of the microscope slide.

The potential use of geopolymer for cleaning air

223

Figure 8.2 Representation of light absorbed, transmitted, and reflected (remitted according to Dahm & Dahm’s equation) in a geopolymer sample.15

relates the fraction of light absorbed (A), remitted (R), and transmitted (T) by each layer (Fig. 8.2). This function is determined by measuring the scatter reflectance (remission according to Dahm and Dahm15) of optically thick and thin samples. When samples are highly thick, the fraction of light transmitted is practically insignificant, and Eq. (8.1) is modified to Eq. (8.2) as a consequence. In these conditions, the fraction of light remitted is expressed as RN. The linear absorption (K) and remission coefficient (B), which are properties of the sample, are determined by measuring R in optically thin samples but on a black background. AðR; TÞ 5 ð½12R2 2 T 2 Þ=R 5 2K=B

(8.1)

AðRN ; 0Þ 5 ð12RN Þ2 =RN

(8.2)

By solving these equations, the apparent absorption, remission, and transmission fraction of light through geopolymers synthesized at 40 C, 60 C, and 90 C were calculated. A representation of the calculated light transmitted through different thicknesses is shown in Figs. 8.3AC respectively. Fig. 8.3D shows a comparison of transmission fraction of light for former 550 µm thick geopolymers at different temperatures. It was clearly observed that light with a wavelength below 350 nm is well transmitted through samples, having a maximum record for the geopolymer at 90 C. The next study employed to analyze the spectrum of samples incorporated with TiO2 was applying fluorescence. The samples were irradiated with light of a welldefined wavelength (below 350 nm) and the emitted light was then measured. In this way, a fluorescent or photoactive material should absorb some energy and emit light of lower energy. Fig. 8.4 shows the set of equipment used in this study. All devices employed were provided by Acton Corporation.

224 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 8.3 Graphics (A), (B) and (C) representing the calculated transmission fraction of light for geopolymers synthesized at 40 C, 60 C, and 90 C, respectively. Graphic (D) is a comparison of light transmitted through samples with thickness of 550 µm (40 C, 60 C, and 90 C).

Figure 8.4 Equipment used to analyze geopolymers by fluorescence. Light passes through monochromator A, and then is driven to the sample at B. At C, another monochromator drives light emitted by the sample at B to the photomultiplier D (Spectra Pro-2500i).

8.4

Incorporation of TiO2 in the geopolymer

TiO2 as anatase is a semiconductor that is excited with light of 380 nm12,13 and once incorporated into the geopolymer by ion exchange,5 can be easily excited and fluoresce with a lower energy light (Fig. 8.5). According to the semiconductor theory, the energy difference between the absorbed and emitted light by the TiO2 is the energy used to degrade adsorbed compounds on the geopolymer surface. This process is also known as photooxidation.16

The potential use of geopolymer for cleaning air

225

Figure 8.5 Fluorescence spectrum of ion exchanged geopolymer excited with light at 200, 230, 260, 290, and 315 nm. Table 8.1 Specific surface and pore width of geopolymers before and after ion exchange (TiO2-Geopolymer) Sample

Specific surface (m2/g)

Pore width (nm)

Geopolymer TiO2-Geopolymer

27 196

30.5 2.2

Fig. 8.5 depicts the fluorescence spectra of the first TiO2-geopolymer. A 90 C synthesized geopolymer was employed; which according to the previous results, allows light of 380 nm to be transmitted more easily. The process followed to incorporate TiO2 was ion exchange using a solution of ammonium titanyl oxalate monohydrate (NH4)2 TiO2 (C2O4)2-H20 (Sigma-Aldrich Mexico). The 550 µm thickness samples of 90 C-geopolymers were rinsed several times before and after ion exchange to remove any traces of competing ions.15 It was observed that porosity and pore size distribution was modified after ion exchange. These parameters were measured by nitrogen adsorption on a NOVA 2000e Quantachrome instruments and by following the Barret-Joyner-Halenda method17 (Table 8.1). These values were important because an increase in the specific surface may result in higher adsorption of volatile organic compounds. Fig. 8.6 shows the TiO2 molecules observed by a transmission electron microscope (JEOL, model JEM-1010). The incorporation of these molecules, which grow inside the geopolymer, modified the specific surface and pore width (Table 8.1). This porosity allows volatile organic compounds to access deeper active TiO2, and could reduce the active surface if they are blocked. The phenomenon happens when the volatile organic compound reaches an active site and is irradiated with light of equal or higher energy18 to oxidize the organic compound—it is known as photolysis.

226 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 8.6 TEM micrograph of the ion exchanged geopolymer where small clusters of TiO2 are indicated by arrows.

8.5

Design of a batch reactor to assess the degradation of 2-butanone

The first reactor built to assess the capacity of TiO2-geopolymer to degrade volatile organic compounds, was a 12 L cylindrical shape batch reactor (Fig. 8.7). Inside, a ultraviolet light of long wave (UVA) lamp whose relative spectra power is shown in Fig. 8.8, excited a single face TiO2-Geopolymer sample. This reactor was provided with two valves and a septum which allowed, a mixture of 80% N220% O2 gas, to wash off remaining gases and impurities. Once this cleaning was performed several times, the valves were closed and 2-butanone (Sigma Aldrich) was syringed through the septum. It is important to mention that the lamp employed in this study is known as a UVA lamp. This lamp does not emit high energy ultraviolet light which may degrade 2-butanone by itself.

8.6

Degradation of 2-butanone

Different 2-butanone concentrations ranging from 0.5 to 5 g m23 were syringed in the reactor through a pierceable septum placed on the head of the reactor. The lamp

The potential use of geopolymer for cleaning air

227

Figure 8.7 Batch reactor employed to assess degradation of 2-butanone.

Figure 8.8 Relative spectra power of Exo Terra 15 W UVA lamp employed in the reactor. This lamp does not emit high energy ultraviolet light which may degrade 2-butanone by itself.

was turned on once the reactor temperature reached 25 C and the 2-butanone concentration was steady over time. To determine the gas composition and concentration inside the reactor, gas samples (1 mL) were extracted frequently from the reactor using a 5 mL syringe (Supelco, Bellefonte, PA), and injected into a gas chromatograph (Agilent Technologies Inc., model 7890A, Santa Clara, CA) equipped with a DB-5 ms capillary column (30 m, 0.25 mm ID, 0.25 µm film thickness, Agilent Technologies). The gas chromatograph injection port temperature was 200 C, split 8:1, and the oven temperature was programmed from 40 C initially, then 20 C min21 up to 200 C where it was held for 4 min. High purity helium was used as carrier gas, at 1 mL min21. Chromatographic peaks were identified using a 5975C VL mass

228 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

spectrometer (Agilent Technologies). Ion source and quadrupole temperatures were 230 C and 150 C respectively. The transfer line was heated at 280 C. Electron impact data was collected from 33300 m/z at a scan rate of 5.27 samples per second, with an ionization voltage of 70 eV. Finally, compound identification was made by comparing mass spectral data from the Wiley 8th/NIST 2008 kit database (Agilent Technologies). System software control and data management/analysis were performed by means of Enhanced Chemstation rev. E.02.00 (Agilent Technologies).

8.7

Langmuir-Hinshelwood model

2-Butanone degradation may be considered as a decomposition reaction (A!B 1 C) where its rate of reaction is given by r 5 2 d[A]/dt 5 d[B]/dt 5 d[C]/dt. (B and C are the products and A is the reactant). This reaction is often classified as a heterogeneous catalytic reaction where mass transfer occurs between a gas (2Butanone) and a solid (TiO2-Geopolymer).19 To make this happens; gas concentration should increase at the interface between the solid and gas. This phenomenon, known as adsorption, has been considered before any reaction assumption. Several kinetic models have been developed so far and most of them can completely account for all possible factors including adsorption (e.g., light intensity, reactants concentration, oxygen concentration, water vapor content, and temperature).19 For heterogeneous catalysis, the adsorption is a main factor that has been considered before any chemical reaction arises.19 This chemical adsorption happens on the surface of the catalyst and it is in an equilibrium process between the bulk fluid and the species adsorbed on the catalyst. To formulate this kind of rate equation, the Langmuir-Hinshelwood model (Eq. 8.3) can be used to determine the disappearance rate (r) of reactant (2-butanone).19 r5

kKA CA 1 1 KA CA 1 KB CB 1 KC CC 1 ?

(8.3)

where r is the reaction rate (gm23 min21), CA (gm23) the concentration of reactant (2-butanone), and CB, CC and so on the concentration of products; k is the reaction rate constant (gm23 min21) and KA, KB and KC the adsorption equilibrium constant (m3 g21) for each substance present in the reaction. At the beginning of the reaction only 2-butanone is present (CA0) and products are not already formed; therefore, their terms in Eq. (8.3) vanish and a new expression arises. r0 5

kKA CA0 1 1 KA CA0

(8.4)

where, r0 is the initial reaction rate (gm23 min21), CA0 (gm23) the initial 2-butanone concentration, and KA is the adsorption equilibrium constant for 2-butanone (m3 g21).

The potential use of geopolymer for cleaning air

229

Figure 8.9 Retention time of 2-butanone and its mass spectra.

As a good fit of the experimental data to model is observed (Eq. 8.4), the chemical reaction is the limiting step in the Langmuir-Hinshelwood reaction mechanism. Otherwise, the adsorption is the rate-limiting step; described by Eq. 8.5, where kads (min21) is the adsorption constant of 2-butanone on the geopolymer. r0 5 kads CA0

(8.5)

In this research, the initial reaction rate (r0) was determined by deriving the kinetic slopes equation at the initial reaction time (r0 5 dCA0/dt at t 5 0). Finally, the reaction rate constant k, and the adsorption equilibrium constant KA were calculated by least square analysis of r0 versus CA0.

8.8

Results of the chromatograph

The first chromatograph obtained for pure 2-butanone was in order to determine its retention time (Fig. 8.9). This is the time taken for 2-butanone to travel through the chromatographic column to the detector and it is measured from the time at which the sample is injected to the moment at which the display shows a maximum peak height for that compound. In the same figure it is shown the mass-to-charge ratio (m/z) plot representing 2-butanone. This is a pattern representing the distribution of ions by mass in the sample that were detected in this particular case after 2.8 min of the sample injection. Samples of gas contained in the reactor were taken from the septum and two peaks appeared at 1.6 and 2.1 min on the chromatograph. According to the massto-charge ratio (m/z), these peaks were ascribed to acetaldehyde and acetone

230 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 8.10 Reaction products observed during the photolysis of 2-butanone. Acetaldehyde and acetone at 1.6 and 2.1 min respectively.

Initial reaction rate for different 2-butanone initial concentrations Table 8.2

Run

2-Butanone initial concentration, CA0 (gm23)

Initial reaction rate, r0 (gm23 min21)

1 2 3 4 5 6 7 8

0.5687 0.575 1.11 1.22 2.13 3.24 4.1 5.108

0.000173 0.00016 0.00017 0.000179 0.0001814 0.0001827 0.000199 0.0002

respectively (Fig. 8.10). These reaction products were also found by Vincent et al. for the degradation of 2-butanone by TiO2 (P25 Degussa) on a fiberglass support.20 Table 8.2 shows the reaction rate computed for each initial concentration of 2butanone injected in the reactor. The relative quantities of each reaction product and 2-butanone were determined by a direct proportion of areas under the corresponding peaks in the chromatograph and the reaction rate was determined by deriving the concentration at the early stages of the reaction (r0 5 dCA0/dt at t 5 0). Fig. 8.11 shows a representative behavior of change of 2-butanone concentration through the photolysis in the TiO2-geopolymer. The reaction rate was determined only at the early stage of the reaction (r0 5 dCA0/dt at t 5 0). Pairs of data (CA0 and r0) were analyzed, resulting in a linear correlation (Fig. 8.12). As a good fit to Eq. (8.5) was observed, it was deduced that the

The potential use of geopolymer for cleaning air

231

Figure 8.11 A representative behavior of 2-butanone concentration (0.5687 gm23) versus time.

Figure 8.12 Plot of the initial reaction rate (r0) versus 2-butanone initial concentration (C0).

adsorption phenomena (the rate at which 2-butanone reaches the TiO2-geopolymer) was the limiting step. This is an important issue because 2-Butanone concentration should increase at the interface between the solid and gas in order for degradation to succeed. In other words, a good concentration of 2-butanone should be present in order for the TiO2-geopolymer to do its work.

8.9

Conclusions

Geopolymers are environmentally friendly materials that have been used for construction and cementitious material mainly. Their applications have increased as they have been used for optical purposes.

232 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

In this research, it has been found how geopolymers can transmit, absorb, and remit light at different wavelengths in accordance with their temperature synthesis and sample thickness. In addition, they can ion exchange their cations in such a way that TiO2 can grow inside their pores. In this study it was found geopolymers synthesized at 90 C allowed light of 380 nm to reach the TiO2 clusters more easily than at other temperatures. It was found that samples of 550 µm thick transmitted over 50% of light (380 nm wavelength) to ensure the degradation of 2-butanone on the surface of the geopolymer. The ion exchange process modified the geopolymer specific surface, pore width, and morphology as a consequence of the occupancy of pores by TiO2. These TiO2 clusters were easily reached by light emitted by the low intensity UVA lamp (15 W) in the photoreactor. The gas samples taken from the photoreactor gave evidence of the 2-butanone photolysis. The peaks ascribed to acetaldehyde and acetone were detected by chromatography and they were the same products observed in other studies for the degradation of 2-butanone. The results mentioned above showed that geopolymer can be an excellent material for remediating the environment or can be used as a photocatalyst support in other chemical reactions. Their low temperature synthesis and their chemical composition make them sustainable materials for potential applications in future buildings.

Acknowledgments The authors want to thank CONACYT for the financial support, to Lourdes Palma Tirado (UNAM) and Miriam Rodriguez Olvera (CICATA) for their technical assistance.

References 1. Giancaspro, J.; Balaguru, P.; Lyon, R. Fire Protection of Flammable Materials Utilizing Geopolymers. 49th International SAMPE symposium and exhibition, 2004, International SAMPE symposium and exhibition. Long Beach, CA, 2004, pp 11251138. 2. Zhang, J. G.; Provis, J. L.; Feng, D. W.; Van Deventer, J. S. J. Geopolymers for Immobilization of Cr61, Cd21, and Pb21. J. Hazard. Mater. 2008, 157 (23), 587598. Available from: http://dx.doi.org/10.1016/j.jhazmat.2008.01.053. 3. Pereira, C. F.; Luna, Y.; Querol, X.; Antenucci, D.; Vale, J. Waste Stabilization/ Solidification of an Electric Arc Furnace Dust Using Fly Ash-based Geopolymers. Fuel J. 2009, 88 (7), 11851193. Available from: http://dx.doi.org/10.1016/j.fuel.2008.01.021. 4. Mackenzie, K. J. D.; O’leary, B. Inorganic Polymers (Geopolymers) Containing Acidbase Indicators as Possible Color-change Humidity Indicators. Mater. Lett. 2009, 63 (2), 230232. Available from: http://dx.doi.org/10.1016/j.matlet.2008.09.053.

The potential use of geopolymer for cleaning air

233

5. Gasca-Tirado, J. R.; Manzano-Ramı´rez, A.; Villasen˜or-Mora, C.; Mun˜iz-Villareal, ´ valos, J. C.; Amigo´-Borra´s, V.; Nava-Mendoza, R. M. S.; Zaldivar-Cadena, A.; Rubio-A Incorporation of Photoactive TiO2 in an Aluminosilicate Inorganic Polymer by Ion Exchange. Micropor. Mesopor. Mater. 2012, 153, 282287. Available from: http://dx. doi.org/10.1016/j.micromeso.2011.11.026. 6. Davidovits, J. Geopolymers  Inorganic Polymeric New Materials. J. Ther. Anal. 1991, 37 (8), 16331656. Available from: http://dx.doi.org/10.1007/BF01912193. 7. Xu, H.; Van Deventer, J. S. J. The Geopolymerisation of Alumino-silicate Minerals. Int. J. Miner. Process. 2000, 59 (3), 247266. Available from: http://dx.doi.org/10.1016/s03017516(99)00074-5. 8. Fernandez-Jimenez, A.; Palomo, A. Composition and Microstructure of Alkali Activated Fly Ash Binder: Effect of the Activator. Cement Concrete Res. 2005, 35 (10), 19841992. Available from: http://dx.doi.org/10.1016/j.cemconres.2005.03.003. 9. Lange, J.-P.; Klier, K. U.v.-vis-n.i.r. Studies of Fe(II)-A Zeolite. Zeolites 1994, 14 (6), 462468. Available from: http://dx.doi.org/10.1016/0144-2449(94)90173-2. 10. Anpo, M.; Yamashita, H.; Ichihashi, Y.; Fuji, Y.; Honda, M. Photocatalytic Reduction of CO2 with H2O on Titanium Oxides Anchored within Micropores of Zeolites: Effects of the Structure of the Active Sites and the Addition of Pt. J. Phys. Chem. B 1997, 101 (14), 26322636. Available from: http://dx.doi.org/10.1021/jp962696h. 11. Arbuj, S. S.; Hawaldar, R. R.; Mulik, U. P.; Wani, B. N.; Amalnerkar, D. P.; Waghmode, S. B. Preparation, Characterization and Photocatalytic Activity of TiO2 Towards Methylene Blue Degradation. Mater. Sci. Eng. B 2010, 168 (13), 9094. Available from: http://dx.doi.org/10.1016/j.mseb.2009.11.010. 12. Luttrell, T.; Halpegamage, S.; Tao, J.; Kramer, A.; Sutter, E.; Batzill, M. Why is Anatase a better Photocatalyst than Rutile? -Model Studies on Epitaxial TiO2 Films. Sci. Rep. 2014, 4 (4043), 18. Available from: http://dx.doi.org/10.1038/srep04043. 13. Fukahori, S.; Ichiura, H.; Kitaoka, H.; Tanaka, H. Capturing of Bisphenol A Photodecomposition Intermediates by Composite TiO2-zeolite Sheets. Appl. Catal. B 2003, 46 (3), 453462. Available from: http://dx.doi.org/10.1016/s0926-3373(03)00270-4. 14. Mun˜iz-Villareal, M. S.; Manzano-Ramı´rez, A.; Sampieri-Bulbarela, S.; Gasca-Tirado, ´ valos, J. C.; Pe´rez-Bueno, J. J.; Apatiga, L. M.; J. R.; Reyes-Araiza, J. L.; Rubio-A Zaldivar-Cadena, A.; Amigo´-Borra´s, V. The Effect of Temperature on the Geopolymerization Process of a Metakaolin-based Geopolymer. Mater. Lett. 2011, 65 (6), 995998. Available from: http://dx.doi.org/10.1016/j.matlet.2010.12.049. 15. Dahm, D. J.; Dahm, K. D. Representative Layer Theory for Diffuse Reflectance. Appl. Spectrosc. 1999, 53 (6), 647654. Available from: http://dx.doi.org/10.1366/ 0003702991947298. 16. Ollis, D. F.; Pelizzetti, E.; Serpone, N. Photocatalysis: Fundamentals and Applications. In Photocatalysis: Fundamentals and Application; Serpone, N., Pelizzetti, E., Eds.; Wiley-Interscience: New York, 1989; pp 603637. 17. Roque-Malherbe, R. M. A. Adsorption and Diffusion in Nanoporous Materials; CRC Press: New York, 2007. 18. Kish, H. What is Photocatalysis. In Photocatalysis: Fundamentals and Application; Serpone, N., Pelizzetti, E., Eds.; Wiley-Interscience: New York, 1989; pp 18. 19. Hill, C. G. An Introduction to Chemical Engineering Kinetics and Reactor Design; John Wiley & Sons: Hoboken, NJ, 1977. 20. Vincent, G.; Queffeulou, A.; Marquaire, P. M.; Zahraa, O. Remediation of Olfactory pollution by Photocatalytic Degradation Process: Study of Methyl Ethyl Ketone (MEK). J. Photochem. Photobiol. A 2007, 191 (1), 4250. Available from: http://dx.doi.org/ 10.1016/j.jphotochem.2007.04.002.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

9

Philip Van den Heede1,2 and Nele De Belie1 1 Ghent University, Ghent, Belgium, 2Strategic Initiative Materials (SIM vzw), project ISHECO within the program ‘SHE’, Ghent, Belgium

9.1

Introduction

Nowadays, numerous introduction sections of scientific articles and books published on potentially “green” concrete with high cement replacement levels include often nothing more than a rather vague general justification for the research conducted. Usually, their authors state that because at least 50% of the ordinary Portland cement (OPC) is replaced with by-products from other industries, cementrelated greenhouse gas emissions can be reduced substantially. However, this acclaimed environmental benefit is almost never quantified in an objective way. Or even worse, it is roughly estimated that the reduction in environmental burden in terms of percentage should more or less resemble the cement replacement level. The lacking of an objectively quantified environmental score for these low-cement concrete types is in a way rather surprising because the required assessment tool to achieve this goal is certainly available and quite generally accepted under the name life cycle assessment (LCA). True, LCA was originally not developed for the construction industry. Moreover, at the start it was far from a well-developed methodology. The study of environmental impact of consumer products began in the late 1960s and early 1970s when environmental issues like resource and energy efficiency, pollution control, and solid waste became issues of broad public concern.1,2 Initially, those studies were not much more than simple comparisons between products to see which one was better. Soon it was recognized that for many of these products the largest portion of the environmental impact was not caused by its use, but by its production, transportation, or disposal.1 One of the first published examples of system analysis quantifying the resource requirements, emission loadings and waste flows of a production chain from “cradle-to-grave,” dates from 1974. The study was conducted by the Midwest Research Institute for the US Environmental Protection Agency.3 In fact, it was a follow-up of a study performed by the same institute for the Coca Cola Company in 1969 already, to enable an environmental comparison between different beverage containers. In the subsequent years, extensive efforts were done to harmonize and standardize LCA. It resulted in the publication of the ISO standards on LCA4,5 and the development of some important impact methods (e.g., Center of Environmental Science (CML)6, Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00010-3 © 2017 Elsevier Ltd. All rights reserved.

236 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Eco-Indicator 997 etc.). As such, the whole methodology had gained substantial importance by the first decade of the 21st century. It is also in this period that the LCA approach drew the attention of the construction sector.8 Moreover, environmental policy all over the world started to become more and more life cycle based, especially with respect to carbon footprinting. These specific circumstances convinced the Concrete & Environment Research Group of the Magnel Laboratory for Concrete Research of Ghent University to dedicate more research efforts to quantifying the environmental impacts related to cement and concrete production using LCA. It resulted in a thorough understanding of the actual sustainability of concrete with high volumes of the cement replaced with fly ash (FA), an industrial by-product of coal-fired electricity production.913 The functional unit (FU) choice and allocation of impacts approach related to industrial by-products were identified as very important issues when performing an LCA study on concrete. They have been addressed thoroughly in the first two case studies included in this book chapter (Section 9.4 and 9.5). Studying the high-volume FA concrete, also helped to gradually fine-tune the LCA methodology for other novel concrete types. As such, a third case study could be included in which the environmental impacts related to the incorporation of natural flax fibers in concrete are assessed (Section 9.6). Prior to the actual case studies, the necessary background on LCA of concrete, the studied concrete compositions, and the natural flax fiber are provided (Sections 9.2 and 9.3).

9.2

Background on life cycle assessment of concrete

9.2.1 General methodology LCA is defined as “the compilation and evaluation of the inputs, outputs and potential environmental impacts of a product system throughout its life cycle”.4 In other words, LCA is a tool for the analysis of the environmental burden of products at all stages in their life cycle. According to this definition, the impact of a product is normally studied from “the cradle to the grave.” Traditionally, a LCA study should consist of four major steps: (1) the definition of goal and scope; (2) the inventory analysis; (3) the impact analysis; and (4) the interpretation. Within the first step (1), the goal of the study should be defined very clearly. When aiming at a sustainability assessment of concrete with an alternative binder system, the goal is usually a proper quantification of the avoided cement-related CO2 emissions. However, nowadays it is actually recommended to do more than a mere carbon footprinting. One should have a clear idea on how the concrete will perform over a broad range of environmental impact categories. Once the main goal has been decided on, the scope should be set as such that it remains achievable. This means that the system boundaries and more importantly the FU should be in full agreement with this goal. The system boundaries may vary. A cradle-togate or cradle-to-grave approach could be considered. Transport of raw materials for concrete manufacturing or the initial construction of the infrastructure for producing those raw materials and the concrete could be accounted for or not.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

237

Depending on these choices, the quantified environmental impacts can be significantly different in value. For the three case studies presented in this chapter, transport of each constituent to the concrete plant was not taken into consideration since its impact is always very case specific. On the other hand, the infrastructure needed for concrete manufacturing was accounted for. The issue of the FU has been discussed more in detail in Section 9.2.2. The second step (2) on inventory analysis comprises the collection of data regarding the use of energy and raw materials, the emission to air, water, and soil, as well as the production of waste related to each concrete constituent and the operation of the concrete plant. The necessary information can be obtained directly from the industries involved using detailed questionnaires or from publicly available annual environmental reports (ERs) and environmental product declarations (EPDs). Obviously, data from questionnaires will result in a more reliable life cycle inventory (LCI) because ERs and EPDs will always hold a certain risk of misinterpretation and double counting. However, first hand data are not always provided by the companies because of confidentiality issues. As a consequence, the larger part of the LCIs is based on data from ERs, EPDs, and LCA related journals. Therefore, it is understandable that ISO 140445 requires detailed documentation referencing for all public sources used. A sensitivity analysis is also very useful. For instance, Josa et al.14 made an extensive comparative analysis of the available life cycle inventories of cement in the EU to obtain a general trend in CO2, NOx, SOx, and dust emissions. LCA databases (e.g., Ecoinvent15) are seen as another important data source. In the present study most of the required data were collected from Ecoinvent. Only superplasticizer inventory data were obtained from an environmental declaration published by the EFCA.16 One of the biggest challenges at this stage lies in the proper allocation of impacts to the alternative binders for “green” concrete. Usually those materials are by-products (FA, silica fume (SF), blast-furnace slag (BFS)) of other industries (coal-fired electricity production, silicon metal production, steel production). As a consequence, the overall impact of the industry should be divided in a fair manner between main products and by-products. This delicate issue has been dealt with in Section 9.2.3. The main aim of step (3), the impact analysis, is to connect each LCI result to the corresponding environmental impacts. Usually, this approach results into a classification of impact categories, each with a category indicator. Two main schools of methods can be distinguished.17 The first school comprises classical life cycle impact assessment (LCIA). Its category indicator is located right in between the LCI results and the category end points (where the environmental effect or damage occurs). The corresponding methods restrict quantitative modeling to relatively early stages in the cause-effect chain to limit uncertainties and group LCI results related to a certain environmental problem, into midpoint categories. Therefore, these methods (e.g., CML) are considered to be problem oriented. For example, a material’s impact on climate change can be expressed in kilograms CO2 equivalents. Obviously, this is merely a quantification of an emission that contributes to the problem of climate change and not a quantification of the actual environmental damage.

238 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

The second school focuses much more on the actual effect. So-called damage oriented impact methods (e.g., Eco-indicator 99) try to model the cause-effect chain up to the endpoint, or the actual environmental damage, sometimes with high uncertainties. With respect to climate change, the damage on human health is quantified in terms of disability adjusted life years (DALYs). This unit counts as a measure for the Years Lived Disabled (YLD) and the Years of Life Lost (YLL) due to this damage. According to Benetto et al.18 the problem related approach provides reliable results, although it is sometimes difficult to compare them with each other. On the other hand, damage oriented impact analysis allows for a much easier interpretation of the LCA output, but is considered to be not so reliable. One of the key parts of step (4), the interpretation phase, is the identification of the significant issues based on the results of the LCI and LCIA phases. It eventually results in an environmental score for each of the considered environmental impact categories.

9.2.2 Importance of the functional unit choice Of great importance is the definition of the FU. This unit is seen as the reference unit of the product system for which the environmental impact will be calculated.4,8 Literature review shows that the scale of this FU for LCA can vary significantly from the material (1) on to the structure level (2): 1. When comparing the environmental impact of different concrete mix designs, a smallscale FU on the material level can be appropriate. An efficiency indicator similar to the the binder intensity index proposed by Damineli et al.19 is seen as a first example. It measures the total amount of binder per m3 of concrete to deliver 1 MPa of strength. As such, it is in compliance with Damtoft et al.20 who see compressive strength as one of the key options to increase the efficiency of cement use. Still, a drawback of the binder intensity index is that it does not relate to a unit of service life. Another possible FU choice is the amount of concrete needed in a simple structural element (column, beam, slab,. . .) with a given mechanical load and a predefined service life in a given environment. This way, additional concrete manufacturing due to replacement or repair over time, is included in the LCA. The same goes for differences in strength. The use of a high strength concrete implicates that structure dimensions and thus the overall concrete amount needed can often be reduced considerably. The resulting environmental benefit will be visible in the LCA output. Habert and Roussel21 studied this for horizontal and vertical steel-reinforced concrete elements, yet without considering the service life aspect. 2. When LCA is used to evaluate the environmental impact of a specific structure, the FU usually corresponds with the structure itself. For a LCA study on pavements, Sayagh et al.22 used a 1 km pavement with a lane width of 3.5 m as FU. The service life was set at 30 years with a traffic load of 9.4 million trucks per lane within this period. Both durability and mechanical load aspects are included in the study that way. Park et al.23 calculated environmental impacts for 1 km of a four lane highway with a predefined service life of 20 years and repair once every 7 years. However, construction activities and maintenance are not always included. For instance, Chowdhury et al.24 chose a road section with a thickness, width and length of 600 mm, 2.5 m and 1000 m, respectively, as FU.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

239

The focus of the study was simply on the material level with no specification of a service life. Besides road structures, entire buildings can be the object of an environmental evaluation. Xing et al.25 studied the difference in impact between steel and concrete office buildings with a use life of 50 years. Both the construction and use phase were taken into account. One m2 of building area was adopted as FU. For the assessment of wood and steel frame housing construction, Gerilla et al.26 expressed the FU in kilogram emission per year and per m2 . It was assumed that the detached houses had a 150 m2 floor space and a design life of 35 years. The whole life cycle of the houses was considered. Specifying a 1 m2 area is not always correct.27 When comparing the construction related impact of in situ cast floors with precast floors, the smallest precast hollow core slabs on the market performed much better than a normally dimensioned in situ cast floor regarding structural strength. As a consequence, the spans achievable with precast concrete slabs are higher and this reduces the number of columns and spread footings. Therefore, the FU comprised the whole building, although the main goal of the study was only an evaluation of the floor type used.

In conclusion, regardless of the scale of the FU for a LCA study comparing the environmental impact of traditional and “green” concrete, the unit should be able to deal with strength and durability/service life differences between concrete types.

9.2.3 Importance of allocation While collecting the inventory data, attention needs to be paid to allocation. Problems occur whenever a system produces more than one product. Somehow, the environmental impacts have to be divided over the different end products. Although this allocation of impacts is preferably avoided, this is often impossible. When this is the case, the inputs and outputs of the system should be partitioned between its different products or functions in a way that reflects the underlying relationships between them, e.g., allocation by mass or by economic value.5 Allocation is of particular importance when industrial by-products like BFS, FA, or SF are in play. Whenever these materials are used as a cement replacing material, attention needs to be paid to their allocated environmental impact. When no impact is attributed to them, they are considered as wastes of the primary industrial processes. The environmental load is at the expense of its producer, i.e., the steel, electricity, or Si-metal producer. However, BFS, FA, and SF are no longer considered as merely waste, but as useful by-products. All three of them meet the necessary requirements imposed by the European Union directive 2008/98/EC28 to qualify for the by-product status: (1) further use of the substance is certain; (2) the substance or object is produced as an integral part of a production process; (3) the substance or object can be used directly without any further processing other than normal industrial practice; and (4) further use is lawful. Moreover, sometimes the use of these by-products is even highly recommended to obtain a sufficient performance. For instance, the use of slag in concrete has been proven beneficial to increase the concrete’s resistance to acid attack.29 Also, the use of FA in concrete tends to improve its durability performance in marine environments.9 As a consequence, the question arises as to whether it would not be more appropriate to

240 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

allocate a part of the environmental load to the concrete producer. Sayagh et al.22 studied two extreme allocation procedures for BFS. Firstly, the BFS is considered as a waste with no steel plant environmental loads allocated to it. Secondly, the BFS is seen as a by-product with 20% of the steel plant environmental flows allocated to the BFS. This percentage is in correspondence with the BFS/steel mass ratio. The environmental indicator results were found to be highly sensitive to the adopted allocation hypothesis. A 20% allocation by mass resulted in a contribution increase to the greenhouse effect of roughly 60%. Acidification and eutrophication potentials were found to be around 25% higher. Ecotoxicity potentials on the other hand almost doubled. Chen et al.30 and Chen31 evaluated the influence of three allocation procedures on the environmental impacts of BFS, FA, and SF: no allocation; allocation by mass; and allocation by economic value. The latter two approaches resulted in the calculation of a mass allocation coefficient Cm and an economic allocation coefficient Ce (Table 9.1) using Eqs. (9.1) and (9.2), respectively. In these formulae, m and h represent the mass and price of main and by-products. Cm 5

mby2product mmain product 1 mby2product

(9.1)

Ce 5

ðh  mÞby2product ðh  mÞmain product 1 ðh  mÞby2product

(9.2)

Chen et al.30 collected the required steel, BFS, electricity and FA data for these calculations from Althaus,32 Sokka et al.,33 Dones et al.,34 Dahlstro¨m and Ekins,35 Metal Bulletin,36 Ecocem,37 and EDF.38 The required Si-metal and SF data come from Chen.31 The resulting allocation coefficients shown in Table 9.1 were used to determine the amount of raw material use, energy use, emissions, and waste production attributable to the by-products BFS, FA, and SF.

Allocation percentages by mass and economic value for FA, BFS, and SF as calculated by Chen et al.30 and Chen31 Table 9.1

Product

Mass produced

Market price

Mass allocation (%)

Economic allocation (%)

Steel BFS Electricity FA Si-metal Silica fume

1 kg 0.24 kg 1 kWha 0.052 kg 1 kg 0.15 kg

400 h/t 40 h/t 0.1 h/kWh 20 h/t 1200 h/t 400 h/t

80.6 19.4 87.6 12.4 87.0 13.0

97.7 2.3 99.0 1.0 95.2 4.8

a

Equivalent to 0.367 kg of hard coal used to produce electricity.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

241

Each of these two allocation principles has its advantages and disadvantages. Mass allocation imposes enormous environmental impacts to the industrial by-products which may discourage the concrete industry to continue applying them as cement replacement.30 When comparing the mass allocated emission values for BFS, FA, and SF with the corresponding cement-related emissions in Van den Heede,13 it is for instance clear that considerably more CO2 (FA: 2.27 kg/kg; cement: 6 0.86 kg/kg), SOx (FA: 10.61 g/kg; cement: 6 0.53 g/kg), and NOx (FA: 4.67 g/kg; cement: 3.65 g/ kg) are emitted for 1 kg of FA. In fact, most of the mass allocated environmental impacts attributed to FA are high. This is mainly caused by the fact that very little FA (0.052 kg) is produced per kWh of electricity. As a consequence, the mass allocation coefficient of 12.4% needs to be applied to the impacts of about 19.2 kWh of electricity to obtain the impact of 1 kg of FA. Since a lot more BFS is produced per kg of steel (0.24 kg), the mass allocated impact of 1 kg BFS is much lower than the mass allocated impact of 1 kg FA. The same holds true for SF, with 0.15 kg of the material produced per kg Si-metal. There is one exception though. Because high amounts of CO2 emissions are inherent to the Si-metal production process, the CO2 emissions assigned to SF through mass allocation remain huge (4.51 kg/kg). When adopting the economic allocation principle, the impacts imposed onto 1 kg of BFS, FA and SF are much lower. Yet, for the latter by-product the CO2 emissions remain higher than for cement production (SF: 1.66 kg/kg; cement 0.86 kg/kg). Economic allocation is sensitive to price instability which can make the LCA outcome fluctuate considerably. When attributing an environmental load to BFS, FA or SF through mass allocation, the corresponding allocation coefficient will remain more or less constant over a long period of time.30

9.3

Studied concrete compositions

9.3.1 Concrete with high volumes of fly ash (and silica fume) In total, four concrete mixtures were studied for the evaluation of the environmental benefit of concrete with high volumes of FA (and SF) (Table 9.2). Mixture T(0.45) a is an OPC concrete composition with a minimum cement content and a maximum water-to-cement (W/C) ratio conforming to NBN B1500139 and NBN EN 20640 for exposure class XS2 which corresponds with environments where concrete is permanently submerged in sea water. Apart from the OPC mixture, a FA containing concrete composition (F15) conforming to the k-value concept of NBN B15 00139 was made. By using the k-value concept, the maximum fly ash-to-binder (F/B) ratio for a minimum total binder content equals only 15% which indicates that the environmental benefit of this partial cement replacement will be rather limited. To increase this benefit, a High-Volume Fly Ash (HVFA) mixture F50 was developed. Note that this HVFA mixture is characterized by a higher total binder content (cement 1 FA) and a lower water-tobinder ratio (W/B) compared to the reference T(0.45). This was mainly done to ensure a strength class based on the 28-day characteristic compressive strength that

242 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Mixture proportions, strength class, and estimated service life of mixtures T(0.45)a, F15, F50, and F40SF10

Table 9.2

Sand 0/4 (kg/m3 ) Aggregate 2/8 (kg/m3 ) Aggregate 8/16 (kg/m3 ) CEM I 52.5 N (kg/m3 ) Fly ash (kg/m3 ) Silica fume (kg/m3 ) Water (kg/m3 ) SP (ml/kg B) W/B F/B SF/B Strength class Service life

T(0.45)a

F15

F50

F40SF10

715 515 671 340 0 0 153 3.0 0.45 0 0 C45/55 32

696 502 654 317.6 56 0 153 3.0 0.41 15 0 C45/55 .100

645 465 606 225 225 0 157.5 4.0 0.35 50 0 C30/37 .100

791 687 454 170 136 34 119 12.0 0.35 40 10 C50/60 .100

is relatively high for this concrete type (C30/37). However, in comparison with the OPC reference T(0.45) (C45/55), the experimental strength class of the HVFA concrete is still lower. The same statement holds true when mixture F15 (C45/55) would be seen as the proper reference concrete. To improve the early age strength performance of concrete with a high FA content, a concrete composition with a ternary binder system was considered as well. The binder system of this mixture (F40SF10) consisted of 50% OPC, 40% FA, and 10% SF. By maintaining the same total binder content as the OPC reference T(0.45) (340 kg/m3 ) in combination with a W/B ratio of 0.35, a strength class of no less than C50/60 could be achieved. Thus, in terms of strength FA 1 SF mixture F40SF10 seems more advantageous than HVFA mixture F50.

9.3.2 Concrete with high volumes of blast-furnace slag Two concrete mixes with high amounts of BFS were investigated as well. There, the slag-to-binder ratios (S/B) amounted to 50% (S50) and 70% (S70), respectively. The same percentages were used in previous research by Gruyaert et al.4143 The total binder content and water-to-binder ratio of those mixtures amounted to 350 kg/m3 and 0.45, respectively. An additional OPC concrete reference composition T(0.45)b with the same values for those two characterizing factors was made for a comparative performance evaluation. The mixture proportions of all three mixtures are given in Table 9.3. Clearly, the strength performance of the BFS containing concrete compositions is considerably less in comparison with the reference concrete. In contrast with the studied FA concrete compositions the total binder content and W/B were not altered, nor was there SF introduced to improve the 28-day strength performance. The

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

243

Mixture proportions, strength class, and estimated service life of mixtures T(0.45)b, S50, and S70

Table 9.3

Sand 0/4 (kg/m3 ) Aggregate 2/8 (kg/m3 ) Aggregate 8/16 (kg/m3 ) CEM I 52.5 N (kg/m3 ) Blast-furnace slag (kg/m3 ) Water (kg/m3 ) SP (ml/kg B) W/B S/B Strength class Service life

T(0.45)b

S50

S70

781 619 480 350 0 157.5 1.4 0.45 0 C45/55 27

781 619 480 175 175 157.5 2.9 0.45 50 C30/37 .100

781 619 480 105 245 157.5 2.9 0.45 70 C25/30 .100

important consequences of this inequality in strength on the environmental score of the high-volume BFS concrete will be demonstrated in case study 1 (Section 9.4).

9.3.3 Possible measures for active crack width control In Section 9.3.1, it has been pointed out that the durability and service life performance of the concrete is very important to define an adequate FU for LCA. In that perspective, one should be very aware of the fact that concrete is a material that has a low tensile strength. Therefore, it is very susceptible to cracking. The cracking can be induced by drying shrinkage, thermal contraction, restraints, differential settlement, and applied loads.44 These cracks logically serve as preferential pathways for aggressive substances. In steel-reinforced concrete applications, faster penetration of Cla, CO2, H2O, and O2 through cracks will cause faster onset of corrosion of the embedded reinforcing steel and hence compromise the structural safety and service life. Being aware of this, the maximum crack widths allowed are being limited in the model codes for structural design.45 For steel-reinforced and prestressed concrete structures those limiting values amount to 0.3 and 0.2 mm, respectively. Still, Van den Heede et al.12 found that accelerated chloride ingress in the crack region is observed, even if the crack width does not exceed 0.1 mm. This means that preferably one needs to limit the crack width even more to a crack width of around 0.05 mm that would normally be able to heal autogenously.46 The easiest way to achieve that involves adding more reinforcing steel. However, this is a rather costly and not necessarily environmentally friendly strategy.12 Another way to minimize the crack widths consists of incorporating microfibers.45 Fiberreinforced concrete already exists since the 1960s47,48 and the possibilities of fiberreinforced cement-based composites have been investigated ever since.49,50 According to Snoeck and De Belie44 a general distinction can be made between natural fibers, glass fibers, metal fibers, and synthetic fibers. In this case study, the

244 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

environmental benefits/impacts related to the use of only one type of natural fibers, i.e., flax fibers, were investigated. Flax is seen as a potentially interesting option for fiber reinforcement. It is much cheaper than the more commonly used synthetic fibers and it has excellent strength properties.51 The only problem is that the materials tend to degrade in alkaline environments like concrete. As a solution to this problem one could chemically treat these fibers to eliminate the alkali-sensitivity of the material. Snoeck et al.52 showed that mercerization with 2 m% concentration sodium hydroxide (NaOH) resulted in optimal multiple cracking of 160 3 40 3 10 mm3 mortar prisms with technical flax fibers (dosage: 2% v/v (30 kg/m3 ), diameter: 103 6 47 µm, length: 20 mm, density: 1500 kg/m3 ). After displacement-controlled four-point bending tests, around 7.5 cracks on average could be observed per sample. Their average crack width amounted to only 28 µm. This is below the 30 µm limit mentioned by Snoeck et al.52 that would allow for complete autogenous healing. Moreover, based on a 6-month immersion of treated slightly-retted hemp fibers in (alkaline) cement filtrate, the same authors concluded that the chemical treatment with NaOH is effective and similar behavior would be expected for natural fibers in general (including flax fibers).

9.4

Case study 1: strength and service life related functional units

9.4.1 Calculation setup LCA calculations were performed for all concrete compositions mentioned in Tables 9.2 and 9.3. Three different units were considered to demonstrate the importance of a correct FU choice, i.e., (1) 1 m3 of concrete; (2) the required concrete volume per unit of strength and service life; and (3) a steel-reinforced concrete slab with predefined mechanical load and design service life. The first one, representing just a volume of concrete, does obviously not account for any of the functionalities of the construction material. The second unit, which can easily be obtained by dividing 1 m3 of concrete by the minimum characteristic cube strength as indicated by the second number in the strength class name and the estimated service life of the concrete, is much more useful in that perspective. Probabilistic service life assessment, cf. fib Bulletin 3453, was used to estimate the time to chloride-induced steel depassivation, an event that is often assumed to correspond with the end of service life in engineering codes although visual damage is not yet being observed at that point. Since this book chapter focuses on the LCA calculations, the details regarding the service life assessment were not included here. The service life prediction outcomes of the seven concrete compositions mentioned in Tables 9.2 and 9.3 were already published in Van den Heede13 and Maes.54 They have been summarized in the same tables. With the strength and service life known, the required concrete volume per unit of strength and service life was calculated and reported in Table 9.4. Note that although the estimated service life amounted to considerably more than 100 years in Van den

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

245

Required concrete volume per 1 N/mm2 of strength and 1 year of service life (FU (2)) for each composition exposed to chloride-induced corrosion

Table 9.4

Composition

T(0.45)a F15 F50 F40SF10 T(0.45)b S50 S70

Strength

Strength

Service life

FU

class ()

(N/mm2 )

(years)

(m3 /(N/mm2  years))

C45/55 C45/55 C30/37 C50/60 C45/55 C30/37 C25/30

55 55 37 60 55 37 30

32 .100 .100 .100 27 .100 .100

5.68 3 1024 1.82 3 1024 2.70 3 1024 1.67 3 1024 6.73 3 1023 2.70 3 1024 3.33 3 1024

Steel reinforced slab (FU (3)) specifications (variable load: 5 kN/m2 , service life: 100 years) for each composition exposed to chloride-induced corrosion Table 9.5

Composition

Strength Class ()

Slab cross-section (mm2 )

# steel rebars ()

# repairs ()

FU (m3 )

T(0.45)a F15 F50 F40SF10 T(0.45)b S50 S70

C45/55 C45/55 C30/37 C50/60 C45/55 C30/37 C25/30

1000 3 160 1000 3 160 1000 3 185 1000 3 155 1000 3 160 1000 3 185 1000 3 195

6 6 5 6 6 5 5

3 0 0 0 3 0 0

1.54 0.79 0.92 0.77 1.54 0.92 0.97

Heede13 and Maes,54 a maximum timespan of only 100 years was assumed nevertheless in the calculation of the FU. Because the benefits of having a higher compressive strength class cannot always be fully utilized to reduce the overall dimensions of a structure, option (2) can in some cases overestimate the environmental benefit. Therefore, it can be recommended to use a realistic steel-reinforced element with given properties as a FU. Option (3) counts as a good example of such a unit. In this study, the slab mentioned, has a span and width of 5 m and 1 m, a variable load of 5 kN/m2 and a design service life of 100 years. All structural design calculations were done in accordance with Eurocode 2.45 Table 9.5 gives an overview of the necessary slab cross-section, the required concrete volume and the amount of reinforcing steel (rebar diameter: 16 mm) to manufacture slabs with each of the studied concrete compositions and the number of repairs needed to maintain them for 100 years.

246 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 9.6 Life cycle inventory data from Ecoinvent used for each concrete mixture Constituent

LCI from Ecoinvent

Sand 0/4 Aggregate 2/8 Aggregate 8/16 Cement Water

Sand, at mine/CH U Gravel, round, at mine/CH U Gravel, round, at mine/CH U Portland cement, strength class Z 52.5, at plant/CH U Tap water, at user/CH U

The inventory data were mainly collected from the renowned Ecoinvent database.15 Their proper short descriptions as mentioned in the database are mentioned in Table 9.6. Most of the data in Ecoinvent are specifically applicable for Switzerland or represent a European average and may not be completely valid for Belgium. Therefore, the available Ecoinvent inventory data for each concrete constituent were carefully analyzed to see whether the involved production processes deviated much from what is common practice in Belgium. Where possible, the necessary changes were made to make them more representative for the Belgian situation. Superplasticizer inventory data were obtained from an environmental declaration published by EFCA.16 Transport of each constituent to the concrete plant was not incorporated in the LCA since its environmental impact is always very case specific. With respect to the industrial by-products FA, SF, and BFS, no impact of the main industrial processes (i.e., coal-fired electricity, silicon metal and steel production) were allocated to them in any way for this first case study. Only the required limited basic treatment after their initial production to make them suitable for use in concrete were taken into consideration. Chen31 provided a basic dataset for drying and storage of FA. The same authors also published datasets that cover the basic treatment of SF and BFS. In the case of the former, it comprises just the local storage of the material in silos, while in the case of the latter, it includes granulation, drying, grinding, and storage. These datasets were compiled from the treatment data published by Construction Technology Laboratories, Inc.55 Based on this information, similar life cycle inventories for FA, SF, and BFS treatment have been developed in SimaPro using data from Ecoinvent for use in our case studies. The impacts related with the production process at a concrete plant were included by the partial assignment of the following LCI from Ecoinvent: “Concrete, normal at plant/CH U.” It comprises the whole process of producing 1 m3 of readymixed concrete, including all internal processes (transport, wastewater treatment, etc.) and infrastructure, and this for a traditional concrete composition. By removing the original concrete constituents and their transport from this inventory, a new LCI is obtained that simply represents the concrete production in general, without any link with a predefined concrete composition. This new LCI was assigned to each of the concrete compositions under investigation.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

247

Characterization factors and indicator units of the CML-IA impact method

Table 9.7

Impact category

Characterization factor

Indicator unit

Abiotic depletion

Abiotic depletion potential, fossil fuels (ADP) Global warming potential (GWP) Ozone depletion potential (ODP) Human toxicity potential (HTP) Freshwater aquatic ecotoxicity potential (FAETP) Marine aquatic ecotoxicity potential (MAETP) Terrestrial ecotoxicity potential (TETP) Photochemical ozone creation potential (POCP) Acidification potential (AP) Eutrophication potential (EP)

MJ (fossil fuels)

Climate change Ozone layer depletion Human toxicity Freshwater aquatic ecotoxicity Marine aquatic ecotoxicity Terrestrial ecotoxicity Photochemical oxidation Acidification Eutrophication

kg (CO2 eq) kg (CFC-11 eq) kg (1,4-DB eq) kg (1,4-DB eq) kg (1,4-DB eq) kg (1,4-DB eq) kg (C2H4 eq) kg (SO2 eq) kg (PO4 eq)

Given the problems with damage oriented impact methods (see Section 9.2.1), it was decided to use the well-known problem-oriented CML-IA impact method for the three case studies discussed in this book chapter. It is an update of the CML 2 baseline 2000 method which was released by the CML of Leiden University in the Netherlands. The list of best available practice impact categories drawn up by the SETAC (Society of Environmental Toxicology and Chemistry) Working Group on LCIA served as a basic list for the problem oriented impact method CML.56 For the LCAs conducted in this book chapter, we will mainly look at the baseline impact categories. For each category, a category indicator can be calculated based on the applicable characterization model and the characterization factors derived from the underlying model. The applicable characterization factors and indicator units per impact category have been summarized in Table 9.7. All subsequent LCA calculations were performed in SimaPro 8 equipped with the Ecoinvent 3 database.

9.4.2 Evaluation of the environmental impact Figs. 9.19.3 include the 10 baseline indicators of the CML-IA impact method for each of the three FUs that were considered. When looking at the impacts related to 1 m3 of concrete (Fig. 9.1), there seems to be a significant environmental benefit of all the concrete compositions with a binder system consisting of high volumes of FA, FA 1 SF, or BFS in comparison with their corresponding references. With OPC composition T(0.45)a as reference, the environmental benefit of HVFA mixture F50 ranges between 22% and 3 %, depending on the considered impact category. Where mixture F15, which meets the requirements of the k-value concept, is seen as reference this benefit slightly decreases to 18%28%. The FA 1 SF

248 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

0.0

ADP (× 10 1 MJ) 1.0 2.0 3.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70 0.0

ODP (× 10 -5 kg CFC-11 eq) 0.5 1.0 1.5

T(0.45)a F50 F40SF10 T(0.45)b S50 S70

FAETP (× 10 1 kg 1,4-DB eq) 0.0 0.5 1.0 1.5 T(0.45)a F50 F40SF10 T(0.45)b S50 S70 0.0

TETP (× 1.0

kg 1,4-DB eq) 2.0 3.0

F50 F40SF10 T(0.45)b S50 S70 10 -1 kg

F15 F50 F40SF10 T(0.45)b S50 S70

AP (× 2.0 4.0

F50 F40SF10 T(0.45)b S50 S70

6.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70 0.0 T(0.45)a

MAETP (× 10 5 kg 1,4-DB eq) 1.0 2.0 3.0 4.0 5.0 Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70

POCP (× 10 -2 kg C2H4 eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70

SO2 eq) 8.0 10.0 12.0 Cement FA SF BFS Sand Gravel Water SP Mixing

5.0

HTP (× 10 1 kg 1,4-DB eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

GWP (× 10 2 kg CO2 eq) 1.0 2.0 3.0 4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

10 -1

T(0.45)a

0.0

T(0.45)a

2.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

0.0

2.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

T(0.45)a

4.0

0.0 T(0.45)a F15 F50 F40SF10 T(0.45)b S50 S70

EP (× 10 -1 kg PO4 eq) 1.0 2.0 3.0

4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

Figure 9.1 Influence of the applied binder system on the 10 CML-IA baseline impact categories related to 1 m3 concrete without allocated impacts assigned to FA, SF, and BFS.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

0.0

2.0

ADP (× 10 -3 MJ) 4.0 6.0 8.0 10.0 12.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

F50 F40SF10 T(0.45)b S50 S70 0.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70

F50 F40SF10 T(0.45)b S50 S70 0.0 T(0.45)a

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70

F50 F40SF10 T(0.45)b S50 S70 0.0 T(0.45)a

T(0.45)a

F50 F40SF10 T(0.45)b S50 S70

F50 F40SF10 T(0.45)b S50 S70

0.0

F15 F50 F40SF10 T(0.45)b S50 S70

AP (× 10 -4 kg SO2 eq) 1.0 2.0 3.0 4.0 5.0

0.0

10.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

T(0.45)a

kg 1,4-DB eq) TETP (× 2.0 4.0 6.0 8.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F50 F40SF10 T(0.45)b S50 S70

0.0 T(0.45)a F15

F50 F40SF10 T(0.45)b S50 S70

4.0

MAETP (× 10 1 kg 1,4-DB eq) 1.0 2.0 3.0 4.0

POCP (× 10 -5 kg C2H4 eq) 0.5 1.0 1.5

2.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

6.0

HTP (× 10 -2 kg 1,4-DB eq) 1.0 2.0 3.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

10 -5

0.0

4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

FAETP (× 10 -3 kg 1,4-DB eq) 2.0 4.0 6.0 8.0 10.0

GWP (× 10 -1 kg CO2 eq) 1.0 2.0 3.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

ODP (× 10 -9 kg CFC-11 eq) 2.0 4.0 6.0 8.0 10.0 12.0

T(0.45)a

0.0

0.0

249

EP (× 10 -1 kg PO4 eq) 2.0 4.0 6.0 8.0 10.0 12.0 Cement FA SF BFS Sand Gravel Water SP Mixing

Figure 9.2 Influence of the applied binder system on the 10 CML-IA baseline impact categories related to the required concrete volume per unit of strength and service life without allocated impacts assigned to FA, SF, and BFS.

250 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

0.0

1.0

ADP (× 10 1 MJ) 2.0 3.0 4.0

T(0.45)a

5.0

Concrete Steel Repair

F15

F50

0.0 T(0.45)a F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70 0.0

ODP (× 10 -5 kg CFC-11 eq) 1.0 2.0 3.0 4.0

T(0.45)a

Concrete Steel Repair

F15 F50

HTP (× 10 1 kg 1,4-DB eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 T(0.45)a F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

S70 FAETP (× 10 1 kg 1,4-DB eq) 0.0 1.0 2.0 3.0

T(0.45)a

4.0

Concrete Steel Repair

F15 F50

0.0

F50

T(0.45)b

T(0.45)b

S50

S50 S70 0.0

TETP (× 10 -1 kg 1,4-DB eq) 1.0 2.0 3.0 4.0

T(0.45)a

Concrete Steel Repair

F15 F50

0.0

POCP (× 10 -2 kg C2H4 eq) 2.0 4.0 6.0 8.0 10.0 12.0

T(0.45)a

Concrete Steel Repair

F15 F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70

S70 0.0

F50

Concrete Steel Repair

F15 F40SF10

S70

MAETP (× 10 5 kg 1,4-DB eq) 1.0 2.0 3.0 4.0 5.0 6.0

T(0.45)a

F40SF10

F15

Concrete Steel Repair

F15

F40SF10

T(0.45)a

Concrete Steel Repair

F15

F40SF10

S70

GWP (× 10 2 kg CO2 eq) 2.0 4.0 6.0 8.0 10.0

AP (× 0.5

10 -1 kg 1.0

SO2 eq) 1.5

2.0

Concrete Steel Repair

0.0 T(0.45)a F15 F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

S70

1.0

EP (× 10 -1 kg PO4 eq) 2.0 3.0 4.0 5.0

6.0

Concrete Steel Repair

Figure 9.3 Influence of the applied binder system on the 10 CML-IA baseline impact categories related to a steel-reinforced concrete slab with a predefined mechanical load and service life without allocated impacts assigned to FA, SF, and BFS.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

251

composition turns out to be even more beneficial. This is mainly due to the fact that it has a lower total binder content than HVFA composition F50. In comparison with references T(0.45)a and F15, the impacts related to mixture F40SF10 are 31%48% and 27%45% lower. Also for the concrete containing high volumes of BFS (mixtures S50 and S70), the impacts were much lower (30%48% and 42% 67%, respectively) than the ones of OPC reference T(0.45)b. It should be noted though that these reductions in environmental impact per m3 of concrete can only be recorded when apart from the basic treatment of FA, SF, and blast-furnace, no allocated impacts of their corresponding main industries (coal-fired electricity, silicon metal and steel production) were assigned to them. From Fig. 9.1, it can be seen very clearly that the impacts related to basic treatment of the by-products are negligible to those of cement production. The environmental scores change completely when the required concrete volume per unit of strength and service life, a unit far more functional than 1 m3 of concrete, is subjected to LCA calculations (Fig. 9.2). Whenever the OPC concrete (T(0.45)a or T(0.45)b) is taken as reference now, the HVFA, FA 1 SF, and BFS concrete are a lot more sustainable. As such, composition F50 has impact indicators that are around 6368% lower. For composition F40SF10, the impacts are reduced with no less than 80%85%. Opposed to reference T(0.45)b, the reduction in impact for mixtures S50 and S70 amounts to around 72%79% and 71% 84%, respectively. The poor service life performance of the OPC references (T(0.45)a: 32 years, T(0.45)b: 27 years) in marine environments catalogued as exposure class XS2, is the main explanation for this. The higher strength class of the FA 1 SF composition also helped in that perspective. HVFA concrete F50 and BFS concrete mixtures S50 and S70 do not have the strength advantage, their promising environmental scores mainly result from the fact that they should guarantee a service life of at least 100 years when used in steel reinforced concrete structures. With composition F15 as reference, there seems to be only an environmental benefit (impacts 33%49%) for the FA 1 SF concrete. The high total binder content (450 kg/m3 ) without a substantial increase in strength class (C30/37) for HVFA concrete F50, results in impact indicators that exceed the ones of the F15 reference by 7%22%. With a uniformly load steel-reinforced concrete slab with a predefined uniform mechanical load and service life as FU, the influence of the concrete strength class on the resulting structural height and required number of steel reinforcements for the slab is also accounted for (Fig. 9.3). It indicates an environmental impact reduction compared to OPC concrete T(0.45)a of 34%54 % and 30%64% for concrete compositions F50 and FA 1 SF, respectively. With the k-value conforming composition F15 as reference, those percentage ranges only amount to 10%17% and 9%34%, respectively. The BFS containing concrete mixtures S50 and S70 are characterized by CML baseline impact indicators that are 37%62% and 39%71% lower than the one recorded for the corresponding reference concrete T(0.45)b. It should be noted that the environmental benefits obtained for the steelreinforced slab as FU are clearly not as pronounced as the ones belonging to the

252 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

previously considered FU, i.e., the required concrete volume per unit of strength and service life. The main reason for this is that the presence of a limited number of steel rebars (5 or 6) can already contribute in a major way to the overall score for the slab for some impact categories, like for example abiotic depletion, human toxicity, freshwater aquatic ecotoxicity, marine aquatic ecotoxicity, photochemical oxidation, and eutrophication. Although the concrete volume needed to initially manufacture the slab exceeds the required steel volume by far, it is the latter material that tends to dominate the overall value for those impact indicators. It demonstrates that reducing the structural height of concrete elements by using a concrete with a high strength class and consequently incorporating more reinforcing steel may not be the best option from an environmental point of view. Clearly, only the consideration of a realistic steel-reinforced concrete element as FU can reveal this kind of important environmental effects inherent to combined material use (concrete 1 steel). The easy-to-use required concrete volume per unit of strength and service life as FU evidently cannot. Therefore, it was decided to mainly use the steel-reinforced concrete slab as object of study further on.

9.5

Case study 2: allocated impacts for alternative binders

9.5.1 Calculation setup The calculation setup for the second case study was very similar in terms of the definition of goal and scope, the inventory analysis, and the impact method selection. The only major differences were the additional assignment of an economically allocated impact of coal-fired electricity, silicon metal and steel production to the industrial by-products FA, SF, and BFS next to their basic treatment and the focus on the steel reinforced concrete slab as FU. For the allocation of impacts, the economic allocation coefficients mentioned in Table 9.1 were applied to the following Ecoinvent life cycle inventories: electricity; hard coal, at power plant/BE U; MG-silicon, at plant/NO U; and Pig iron, at plant/GLO U. With respect to the FU, 1 m3 of concrete will also still be considered in the case study because it shows more clearly how the environmental impacts of the by-products increase when adopting an economic allocation approach.

9.5.2 Environmental impact assessment Although the economic allocation coefficients for the by-products are actually quite low (Table 9.1: FA: 1.0%, SF: 4.8%, BFS: 2.3%), they increase the impacts associated with each CML-IA baseline category quite substantially (Fig. 9.4). When looking at Abiotic depletion potential (ADP), Human toxicity potential (HTP), Freshwater aquatic ecotoxicity potential (FAETP), Marine aquatic ecotoxicity

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

0.0

ADP (× 10 1 MJ) 1.0 2.0 3.0

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b

S50 S70

0.0

ODP (× 10 -5 kg CFC-11 eq) 0.5 1.0 1.5

T(0.45)a F50 F40SF10 T(0.45)b

S50 S70

FAETP (× 10 1 kg 1,4-DB eq) 0.0 0.5 1.0 1.5 T(0.45)a F15 F50 F40SF10 T(0.45)b

S50 S70 0.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b

S50 S70

0.0

F15 F50 F40SF10 T(0.45)b S50 S70

2.0

AP (× 4.0

10 -1 kg 6.0

T(0.45)a F50 F40SF10 T(0.45)b

S50 S70

5.0

HTP (× 10 1 kg 1,4-DB eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70 0.0

MAETP (× 10 5 kg 1,4-DB eq) 1.0 2.0 3.0 4.0 5.0

T(0.45)a

Cement FA SF F50 BFS F40SF10 Sand Gravel T(0.45)b Water S50 SP S70 Mixing -2 POCP (× 10 kg C 2H 4 eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 F15

T(0.45)a

Cement FA SF BFS Sand Gravel Water SP Mixing

F15 F50 F40SF10 T(0.45)b S50 S70

SO2 eq) 8.0 10.0 12.0 Cement FA SF BFS Sand Gravel Water SP Mixing

GWP (× 10 2 kg CO2 eq) 1.0 2.0 3.0 4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

2.0

Cement FA SF BFS Sand Gravel Water SP Mixing -1 TETP (× 10 kg 1,4-DB eq) 1.0 2.0 3.0 4.0

T(0.45)a

0.0

2.0

Cement FA SF BFS Sand Gravel Water SP Mixing

F15

T(0.45)a

4.0

253

0.0 T(0.45)a F15 F50 F40SF10 T(0.45)b

S50 S70

EP (× 10 -1 kg PO4 eq) 1.0 2.0 3.0

4.0

Cement FA SF BFS Sand Gravel Water SP Mixing

Figure 9.4 Influence of the applied binder system on the 10 CML-IA baseline impact categories related to 1 m3 concrete with economically allocated impacts assigned to FA, SF, and BFS.

254 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

potential (MAETP), Photochemical ozone creation potential (POCP), Acidification potential (AP), and Eutrophication potential (EP) per m3 of concrete there is no environmental benefit whatsoever anymore for the HVFA composition F50, no matter whether composition T(0.45)a or F15 is seen as reference. Only the Global warming potential (GWP) (14%17%), Ozone depletion potential (ODP) (17%20%), and Terrestrial ecotoxicity potential (TETP) (7%9%) values are still somewhat lower. The same statement holds more or less true for composition F40SF10. For that mixture, the use of 10% SF additionally results in nonbeneficial ODP and TETP values, which leaves the mixture only with a reduced GWP (18%20%). With an economically allocated impact of steel production assigned to the BFS, the impact indicators of mixtures S50 and S70 also increase. Nevertheless, environmental benefits continue to exist for five out of ten CML baseline impact indicators: GWP (S50: 39%, S70: 55%), ODP (S50: 34%, S70: 48%), HTP (S50: a8%, S70: 11%), TETP (S50: 25%, S70: 36%) and AP (S50: 22%, S70: 32%). Based on the mere assessment of 1 m3 of concrete with economically allocated impacts attributed to the FA, SF, and BFS, it seems no longer advantageous to use high volumes of supplementary cementitious materials in concrete. However, one should remain aware of the fact that for instance the good service life performance of HVFA, FA 1 SF, and BFS concrete in comparison with OPC concrete may still render it worthwhile to use the industrial by-products on a large scale in marine concrete. To evaluate this in an objective way, one must look at the LCA output corresponding with the steel-reinforced concrete slab as FU (Fig. 9.5). For the HVFA concrete, significant reductions in environmental impact are still present in terms of ADP (28%), GWP (46%), ODP (47%), HTP (10%), TETP (38%), POCP (20%), and AP (21%). With respect to the FA 1 SF composition, this only holds true for ADP (20%), GWP (52%), and ODP (42%). With the k-value conforming F15 composition as reference, environmental benefits are barely being recorded anymore for F50 and F40SF10. In contrast with the slabs made of HVFA or FA 1 SF concrete, the slabs made of BFS compositions S50 and S70 still have reduced environmental impacts for all 10 CML-IA baseline impact categories. The percentages of reduction in relation to OPC reference T(0.45)b range between 14%58% for S50 and 6%65% for S70. As such, the BFS-based concrete mixtures seem the most promising ones from an overall environmental point of view. In general, it should be noted that research on concrete with high amounts of supplementary materials is quite often being justified by stating that those materials have the potential of reducing greenhouse gas emissions significantly. Fig. 9.5 indeed mentions important reductions in GWP for compositions F50, F40SF10, S50, and S70, even with allocated impacts assigned to the by-products. Nevertheless, one should always keep in mind the overall environmental performance of a material when quantifying its sustainability. Apparently, when using FA (and SF) in large amounts some other environmental issues surface, especially issues related to toxicity and ecotoxicity, acidification, and eutrophication, which certainly deserve further attention.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

0.0 T(0.45)a

1.0

ADP (× 10 1 MJ) 2.0 3.0 4.0

5.0

Concrete Steel Repair

F15

0.0

F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70

S70 0.0 T(0.45)a

ODP (× 10 -5 kg CFC-11 eq) 1.0 2.0 3.0 4.0 Concrete Steel Repair

F15 F50

HTP (× 10 1 kg 1,4-DB eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

T(0.45)a F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70

S70 0.0 T(0.45)a

FAETP (× 10 1 kg 1,4-DB eq) 1.0 2.0 3.0

4.0

Concrete Steel Repair

F15 F50

0.0

Concrete Steel Repair

F15 F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

MAETP (× 10 5 kg 1,4-DB eq) 1.0 2.0 3.0 4.0 5.0 6.0

T(0.45)a

F40SF10

S70 0.0

T(0.45)a

TETP (× 10 -1 kg 1,4-DB eq) 1.0 2.0 3.0

F50

0.0

4.0

Concrete Steel Repair

F15

POCP (× 10 -2 kg C2H4 eq) 2.0 4.0 6.0 8.0 10.0 12.0

T(0.45)a

Concrete Steel Repair

F15 F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70

S70 0.0

F50

Concrete Steel Repair

F15

F40SF10

F15

Concrete Steel Repair

F15

F50

T(0.45)a

GWP (× 10 2 kg CO2 eq) 2.0 4.0 6.0 8.0 10.0

T(0.45)a

F40SF10

255

AP (kg SO2 eq) 0.5 1.0 1.5

2.0

Concrete Steel Repair

0.0 T(0.45)a F15 F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

S70

1.0

EP (× 10 -1 kg PO4 eq) 2.0 3.0 4.0 5.0

6.0

Concrete Steel Repair

Figure 9.5 Influence of the applied binder system on the 10 CML-IA baseline impact categories related to a steel-reinforced concrete slab with a predefined mechanical load and service life with economically allocated impacts assigned to FA, SF, and BFS.

256 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Another remark should be made about the fact that assigning allocated impacts to supplementary cementitious materials is still considered very controversial. True, FA, SF, and BFS are indeed in agreement with all the requirements to catalog them as by-product instead of waste. Thus, an allocation of impacts would be mandatory in theory. However, this is for the moment still far from common practice in the cement and concrete industry. EPDs of blended cements for example, also containing FA, BFS, etc., are usually not based on the principle of allocation.57,58 Economic allocation is said to impose a fair impact to the by-products. Yet, this is not necessarily true because their availability and resulting prices vary from country to country. In countries like China and India that rely highly on coal-fired electricity production, FA is abundantly available and almost free, while in countries like Belgium FA has to be imported from its neighboring countries which automatically implies higher costs. As a consequence, economic allocation coefficients can vary over a broad range. The LCA output shown in Fig. 9.5, may be more or less representative for the Belgian situation. In other parts of the world, where there is more abundant by-product availability, the LCA output shown in Fig. 9.3 might be more representative. A uniformly applicable eco-profile for concrete with supplementary cementitious materials can simply not be determined.

9.6

Case study 3: natural fibers as reinforcement for service life extension

9.6.1 Calculation setup In this third case study it was evaluated for each of the studied concrete compositions how much the incorporation of technical flax fibers will add in environmental impact to the 10 CML-IA baseline impact categories. Although the dosage of 2% v/v (30 kg/m3 ) proposed by Snoeck et al.52 to have optimal multiple cracking conditions (see Section 9.3.3) is normally valid for a specific mortar composition of which the binder system consisted of cement and FA, the very same dosage was applied in concrete compositions T(0.45)a, F15, F50, F40SF10, T(0.45)b, S50, and S70. 1 m3 of concrete was adopted as FU because the service life aspects for flax fiber-reinforced concrete are for the moment not yet well understood. One could theoretically assume that with crack widths reduced to less than 30 µm the concrete would behave as if uncracked due to its complete autogenous healing capacity for fine cracks. The possibly improved durability and service life performance because of this have not been experimentally assessed though. Moreover, it is also still difficult to get a thorough understanding on how larger cracks of up to 0.3 mm in width shorten the service life. Most engineering model codes—also the ones used to estimate the time to chloride-induced steel depassivation for the studied concrete compositions (Section 9.4.1)—do not account for the presence of those cracks. Some first preliminary estimations indicated that the service life of mortar/concrete structures containing cracks 100300 µm in width would be

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

257

around 81% lower in comparison with sound concrete.59 However, the modeling approach for cracked concrete still needs to be improved further. In other words, a comparison in service life performance between (cracked) concrete with and without fibers is not possible. The rudimentary LCA calculation for flax fiber concrete in this section therefore just focuses on the additional impacts induced by the flax presence in the concrete without accounting for the reduced maintenance related impacts with time. To model technical flax fiber for this rudimentary LCA calculation, a LCI was compiled based on detailed information published by Le Duigou et al.60 According to this data source, the production of flax fiber consists of growing dry green flax stems. Then, a subsequent retting, scutching, and hackling process is required to obtain the final product. As such, everything was modeled in SimaPro 8 with data from Ecoinvent 3. The same was done for the chemical treatment with NaOH of the fibers. A dilution of NaOH, 50% in H2O, to 2% m/m NaOH was modeled by adding the required amount of water (used Ecoinvent data: Sodium hydroxide, 50% in H2O, production mix, at plant/RER U; water, dionized, at plant/CH U). Then, it was linked to the technical flax inventory in a way that 1.75 kg of 2% m/m aqueous NaOH solution is sufficient to adequately treat 1 kg of flax fiber, which is in agreement with in-house conducted experiments on the technical flax. In contrast with Le Duigou et al.60 no transport of materials between sites was included in this study. Note that with respect to the flax fiber, the potential reduction in greenhouse gas emissions due to photosynthesis during plant growth was not taken into account since this approach is not generally accepted in literature according to Le Duigou et al.60 In accordance with the same authors the preferred mass allocation approach was adopted to deal with the multiproduct output of flax cultivation (flax fibers, tows, shives, seeds, etc.). Just like for the previous two case studies an overall eco-profile was compiled in accordance with the CML-IA baseline impact method. The focus here was to evaluate how the impacts related to the use of a limited content of chemically treated flax fibers (30 kg per m3 of concrete) were significantly higher or lower than the ones related to the overall concrete volume.

9.6.2 Environmental impact assessment The incorporation of 30 kg of chemically treated technical flax fiber per m3 of concrete does not result in significant extra impacts for most of the CML-IA baseline impact categories (Fig. 9.6). The ADP, GWP, ODP, POCP, and AP values attributable to it usually comprise less than 15% of the overall impact indicator values. On the other hand, this is not the case when looking at freshwater aquatic ecotoxicity, marine aquatic ecotoxicity, human toxicity, eutrophication, and terrestrial ecotoxicity. Especially terrestrial ecotoxicity seems to be a problem. The treated fibers seem to be responsible for no less than 68% of the impact. The use of fertilizers (especially triple superphosphate) to cultivate the flax is mainly responsible for this high impact. Also the eutrophication and other (eco)toxicity scores are affected by the use of these chemical

258 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

0.0

ADP (×101 MJ) 1.0 2.0 3.0

T(0.45)a

Concrete Flax fiber Flax treatment

F15 F50

GWP (× 10 2 kg CO2 eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

4.0 T(0.45)a F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

Concrete Flax fiber Flax treatment

F15

S70 HTP (× 101 kg 1,4-DB eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

-5

0.0

ODP (× 10 kg CFC-11 eq) 0.5 1.0 1.5

T(0.45)a

2.0

Concrete Flax fiber Flax treatment

F15 F50

F15 F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

Concrete Flax fiber Flax treatment

T(0.45)a

S70 1

0.0 T(0.45)a

FAETP (×10 kg 1,4-DB eq) 0.5 1.0 1.5

2.0

Concrete Flax fiber Flax treatment

F15 F50

0.0 T(0.45)a F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70 0.0

T(0.45)a

TETP (× 10 -1 kg 1,4-DB eq) 1.0 2.0 3.0 4.0 5.0

POCP (× 10 -2 kg C2H 4 eq) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

6.0

Concrete Flax fiber Flax treatment

F15

T(0.45)a

F50

F50 F40SF10

T(0.45)b

T(0.45)b

S50

S50 S70

S70 0.0

F15 F50

Concrete Flax fiber Flax treatment

F15

F40SF10

T(0.45)a

5.0

Concrete Flax fiber Flax treatment

F15

F40SF10

S70

MAETP (×105 kg 1,4-DB eq) 1.0 2.0 3.0 4.0

AP (× 10-1 kg SO2 eq) 2.0 4.0 6.0 8.0 10.0 12.0 Concrete Flax fiber Flax treatment

0.0 T(0.45)a F15

EP (× 10-1 kg PO4 eq) 1.0 2.0 3.0

4.0

Concrete Flax fiber Flax treatment

F50

F40SF10

F40SF10

T(0.45)b

T(0.45)b

S50

S50

S70

S70

Figure 9.6 Influence of chemically treated technical flax fiber incorporation on the 10 CML-IA baseline impact categories related to 1 m3 of concrete with economically allocated impacts assigned to FA, SF, and BFS.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

259

products. This finding emphasizes once more that the usual focus on the carbon footprint when investigating the sustainability of potential new concrete types can be dangerous as it may overlook other environmental issues. When comparing the impacts related to the production of technical flax fiber and the impacts related to their treatment, one can conclude that the latter are usually rather negligible. Overall, it can be concluded that except for the problematic terrestrial ecotoxicity issues, one should not expect dramatic environmental consequences from adding flax fiber to concrete when simply studying concrete volumes. Most probably, if one could prove that this technique could indeed result in a prolonged concrete service life induced by effective crack width control, there would be substantial environmental benefits. The fact that the inclusion of service life in the FU definition has a very important effect on the LCA calculation outcome almost automatically implies this. Therefore, it remains most relevant that the durability and service life of flax fiber-reinforced concrete in comparison with ordinary cracked concrete should be investigated further in the future. The service life should then become part of an integrated LCA.

9.7 G

G

G

Conclusions

A proper FU choice is of great importance when assessing the environmental benefits and impacts related to potentially green concrete types with high cement replacement levels. The required concrete volume per unit of strength and service life or a steel-reinforced concrete slab with a predefined mechanical load and service life count as good examples. Without an allocation of impacts assigned to the industrial by-products FA, SF, and BFS, all studied concrete compositions with high volumes of FA (and SF) or BFS are more sustainable in terms of all 10 CML-IA baseline impact indicators than the reference concrete for exposure class XS2. With the slab as the FU the benefits are somewhat lower because the impacts related to the limited amount of reinforcing steel used tend to dominate the overall environmental scores. Applying economic allocation coefficients to the industrial by-products (FA: 1%, SF: 4.8%, and BFS: 2.3%) increases the impacts of the concrete with high cement replacement levels substantially, especially with respect to the impact categories dealing with (eco)toxicity, acidification, and eutrophication. As a consequence, there is no environmental benefit anymore for all 10 CML-IA baseline impact categories in comparison with traditional steel-reinforced concrete slabs. Note that this conclusion is not unambiguously valid everywhere. It largely depends on the local availability and the resulting prices of the by-products. The incorporation of 30 kg of chemically treated natural flax fibers per m3 of concrete to limit any occurring cracks in width does not add dramatic extra impacts to the composite material except in terms of terrestrial ecotoxicity. The latter phenomenon is induced by the extensive use of fertilizers to grow the flax. Further research is still needed though in order to see by how many years the use of crack width controlling flax fiber in concrete can extend concrete service life before definitive conclusions regarding the sustainability of the composite material can be drawn.

260 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Acknowledgments This research under the program SHE (Engineered Self-Healing materials), project ISHECO (Impact of Self-Healing Engineered materials on steel Corrosion of reinforced concrete) was funded by SIM (Strategic Initiative Materials in Flanders) and VLAIO (Flanders Innovation & Entrepreneurship). The financial support from the foundations for this study is gratefully appreciated.

References 1. Guine´e, J. B.; Heijungs, R.; Huppes, G.; Zamagni, A.; Masoni, P.; Ekvall, T., et al. Life Cycle Assessment: Past, Present, and Future. Environ. Sci. Technol. 2011, 45 (1), 9096. Available from: http://dx.doi.org/10.1021/es101316v. 2. Assies, J.A. Introduction Paper to Setac-Europe Workshop on Environmental Life Cycle Analysis of Products. Proceedings of a Setac-Europe workshop on Environmental Life Cycle Assessment of Products, December 23, Leiden, the Netherlands. 1991. 3. Hunt, R. G.; Franklin, W. E.; Welch, R. O.; Cross, J. A.; Woodal, A. E. Resource and Environmental Profile Analysis of Nine Beverage Container Alternatives; U.S. Environmental Protection Agency: Washington, DC, 1974. 4. International Organization for Standardization (ISO). ISO 14040. Environmental ManagementLife Cycle AssessmentPrinciples and Framework. 2006. p 28. 5. International Organization for Standardization (ISO). ISO 14044. Environmental ManagementLife Cycle AssessmentRequirements and Guidelines. 2006. p 54. 6. Heijungs, R.; Guine´e, J. B.; Huppes, G.; Lankreijer, R. M.; Udo de Haes, H. A.; Wegener Sleeswijk, A., et al. Environmental Life Cycle Assessment of Products. Guide & Backgrounds  October 1992; Centre for Environmental Science, Leiden University: Leiden, 1992. 7. Goedkoop, M.; Spriensma, R. The Eco-indicator 99A Damage Oriented Method for Life Cycle Impact Assessment; PRe´ Consultants bv: Amersfoort, 1999. 8. Desmyter, J.; Martin, Y. The Environmental Impact of Building Materials and Buildings (in Dutch); WTCB: Limelette, 2001. 9. Van den Heede, P.; Maes, M.; Gruyaert, E.; De Belie, N. Full Probabilistic Service Life Prediction and Life Cycle Assessment of Concrete with Fly Ash and Blast-furnace Slag in a Submerged Marine Environment: A Parameter Study. Int. J. Environ. Sustainable Dev. 2012, 11 (1), 3249. Available from: http://dx.doi.org/10.1504/ IJESD.2012.049141. 10. Van den Heede, P.; De Belie, N. Environmental Impact and Life Cycle Assessment (LCA) of Traditional and ‘green’ Concretes: Literature Review and Theoretical Calculations. Cement Concrete Compos. 2012, 34 (4), 431442. Available from: http:// dx.doi.org/10.1016/j.cemconcomp 2012.01.004. 11. Van den Heede, P.; De Belie, N. A Service Life Based Global Warming Potential for High-volume Fly Ash Concrete Exposed to Carbonation. Constr. Build. Mater. 2014, 55 (3), 183193. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.01.033. 12. Van den Heede, P.; Maes, M.; De Belie, N. Influence of Active Crack Width Control on the Chloride Penetration Resistance and Global Warming Potential of Concrete Slabs

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

13. 14.

15. 16. 17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

261

Made with Fly Ash 1 Silica Fume Concrete. Constr. Build. Mater. 2014, 67 (Part A), 7480. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.10.032. Van den Heede, P. Durability and Sustainability of Concrete with High Volumes of Fly Ash (Ph.D. thesis); Ghent University: Ghent, Belgium, 2014. Josa, A.; Aguado, A.; Heino, A.; Byars, E.; Cardim, A. Comparative Analysis of Available Life Cycle Inventories of Cement in the EU. Cement Concrete Res. 2004, 34 (8), 13131320. Available from: http://dx.doi.org/10.1016/j.cemconres.2003.12.020. Frischknecht, R.; Jungbluth, N. (Eds.). Overview and Methodology, Final Report Ecoinvent v2.0, No. 1. St-Gallen: Swiss Centre for Life Cycle Inventories, 2007. EFCA. EFCA Environmental Declaration Superplasticizing Admixures (2006), EFCA doc 325 ETG. http://www.efca.info/publications.html (accessed September 29, 2016). Jolliet, O.; Margni, M.; Charles, R.; Humbert, S.; Payet, J.; Rebitzer, G., et al. IMPACT 2002 1 : a new life cycle assessment methodology. International Journal of Life Cycle Assessment 2003, 10 (6), 324330. Available from: http://dx.doi.org/10.1007/ BF02978505. Benetto, E.; Rousseaux, P.; Blondin, J. Life cycle assessment of coal by-products based electric power plants. Fuel 2004, 83 (78), 957970. Available from: http://dx.doi.org/ 10.1016/S0016-2361(03)00258-8. Damineli, B. L.; Kemeid, F. M.; Aguiar, P. S.; John, V. M. Measuring the ecoefficiency of cement use. Cement Concrete Compos. 2010, 32 (8), 555562. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2010.07.009. Damtoft, J. S.; Lukasik, J.; Herfort, D.; Sorrentino, D.; Gartner, E. M. Sustainable development and climate change initiatives. Cement Concrete Res. 2008, 38 (2), 115127. Available from: http://dx.doi.org/10.1016/j.cemconres.2007.09.008. Habert, G.; Roussel, N. Study of Two Concrete Mix Design Strategies to Reach Carbon Mitigation Objectives. Cement Concrete Compos. 2009, 31 (6), 397402. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2009.04.001. Sayagh, S.; Ventura, A.; Hoang, T.; Franc¸ois, D.; Jullien, A. Sensitivity of the LCA Allocation Procedure for BFS Recycled into Pavement Structures. Resour. Conserv. Recy. 2010, 54 (6), 348358. Available from: http://dx.doi.org/10.1016/j. resconrec.2009.08.011. Park, K.; Hwang, Y.; Seo, S.; Seo, H. Quantitative Assessment of Environmental Impacts of Life Cycle of Highways. J. Constr. Eng. Manage. 2003, 129 (1), 2531. Available from: http://dx.doi.org/10.1061/(ASCE)0733-9364(2003)129. Chowdhury, R.; Apul, D.; Fry, T. A Life Cycle Based Environmental Impacts Assessment of Construction Materials Used in Road Construction. Resour. Conserv. Recy. 2010, 54 (4), 250255. Available from: http://dx.doi.org/10.1016/j. resconrec.2009.08.007. Xing, S.; Xu, Z.; Jun, G. Inventory Analysis of LCA on Steel- and Concreteconstruction Office Buildings. Energy Build. 2008, 40 (7), 11881193. Available from: http://dx.doi.org/10.1016/j.enbuild.2007.10.016. Gerilla, G. P.; Teknomo, K.; Hokao, K. An Environmental Assessment of Wood and Steel Reinforced Concrete Housing Construction. Build. Environ. 2007, 42 (7), 27782784. Available from: http://dx.doi.org/10.1016/j.buildenv.2006.07.021. Lo´pez-Mesa, B.; Pitarch, A.; Toma´s, A.; Gallego, T. Comparison of Environmental Impacts of Building Structures with in situ Cast Floors and With Precast Concrete Floors. Build. Environ. 2009, 44 (4), 699712. Available from: http://dx.doi.org/ 10.1016/j.buildenv.2008.05.017.

262 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

28. European Union. Directive 2008/98/EC of the European Parliament and the Council of 19 November 2008 on Waste and Repealing Certain Directives. Off. J. Eur. Union 2008, L312, 320. 29. Gruyaert, E.; De Belie, N., Van den Heede, P. Acid Resistance of Concrete Containing Blast-furnace Slag: Influence of the Pore Structure and Hydration Process. International RILEM TC211-PAEfinal Conference on Concrete in Aggressive Aqueous Environments:Performance, Testing and Modeling, June 35, Toulouse, France, 2009, pp 389396. 30. Chen, C.; Habert, G.; Bouzidi, Y.; Jullien, A.; Ventura, A. LCA Allocation Procedure Used as an Incitative Method for Waste Recycling: An Application to Mineral Additions in Concrete. Resour. Conserv. Recy. 2010, 54 (12), 12311240. Available from: http:// dx.doi.org/10.1016/j.resconrec.2010.04.001. 31. Chen, C. A Study of Traditional and Alternative Structural Concretes by Means of the Life Cycle Assessment Method (in French) (Ph.D. thesis); University of Technology of Troyes: Troyes, France, 2009. 32. Althaus, H. J. Life Cycle Inventories of Metals, Final Report Ecoinvent 2000, No. 10; Swiss Centre for Life Cycle Inventories: Du¨bendorf, 2003. 33. Sokka, L.; Koskela, S.; Seppa¨la¨, J. Life Cycle Inventory Analysis of Hard Coal Based Electricity Generation, The Finnish Environment 797; Finnish Environment Institute: Helsinki, 2005. 34. Dones, R.; Bauer, C.; Bollinger, R.; Burger, B.; Faist Emmenegger, M.; Frischknecht, R., et al. Life Cycle Inventories of Energy Systems: Results for Current Systems in Switzerland and Other UCTE Countries, Ecoinvent Report no. 5; Swiss Centre for Life Cycle Inventories: St-Gallen, 2007. 35. Dahlstro¨m, K.; Ekins, P. Combining Economic and Environmental Dimensions: Value Chain Analysis of UK Iron and Steel Flows. Ecol. Econ. 2006, 58 (3), 507519. Available from: http://dx.doi.org/10.1016/j.ecolecon.2005.07.024. 36. Metal Bulletin. Ferrous and Non-ferrous Market Prices 20042008. http://www.metalbulletin.com (accessed August 18, 2011). 37. Ecocem. Producer of Ground Granulated Blast-furnace Slag (GGBS), 2008. 38. EDF. Leading the Energy Change, Changer l’e´nergie Ensemble, 2008 Sustainable Development Report; EDF: Paris, 2008. 39. NBN. NBN B15-001. ConcreteSpecification, Performance, Production and Conformity-National Supplement to NBN EN 2061:2001 (in Dutch). 2012. p 29. 40. CEN. NBN EN 206. ConcreteSpecification, Performance, Production and Conformity. 2014. p 95. 41. Gruyaert, E.; Van den Heede, P.; De Belie, N. Chloride Ingress for Concrete Containing Blast-furnace Slag, Related to Microstructural Parameters. Proceedings of the 2nd International RILEM Workshop on Concrete Durability and Service Life Planning (ConcreteLife ’09), September 79, Haifa, Israel, 2009, pp 440448. 42. Gruyaert, E.; Van den Heede, P.; De Belie, N. A Comparative Study of the Durability of Ordinary Portland Cement Concrete and Concrete Containing (High) Percentages of Blast-furnace Slag. Proceedings of the International RILEM Conference on Material Science, September 68, Aachen, Germany, 2010, pp 241251. 43. Gruyaert, E.; Van den Heede, P.; Maes, M.; De Belie, N. Investigation of the Influence of Blast-furnace Slag on the Resistance of Concrete Against Organic Acid or Sulphate Attack by Means of Accelerated Degradation Tests. Cement Concrete Res. 2012, 42 (1), 173185. Available from: http://dx.doi.org/10.1016/j.cemconres.2011.09.009.

Sustainability assessment of potentially ‘green’ concrete types using life cycle assessment

263

44. Snoeck, D.; De Belie, N. From Straw in Bricks to Modern Use of Microfibers in Cementitious Composites for Improved Autogenous HealingA Review. Constr. Build. Mater. 2015, 95, 774787. Available from: http://dx.doi.org/10.1016/j.conbuildmat. 2015.07.018. 45. NBN. NBN EN 1992-1-1 ANB. Eurocode 2: Design of Concrete Structures - Part 1-1: General Rules and Rules for Buildings (in Dutch). 2010. p 33. 46. Jaroenratanapirom, D.; Sahamitmongkol, R. Self-crack Closing Ability of Mortar with Different Additives. J. Miner. Met.Mater. Soc. 2011, 21 (1), 917. 47. Romualdi, J. P.; Batson, G. Mechanics of Crack Arrest in Concrete. J. Eng. Mech. Div. 1963, 89, 147168. 48. Romualdi, J. P.; Mandel, J. A. Tensile Strength of Concrete Affected by Uniformly Distributed and Closely Spaced Short Lengths of Wire Reinforcement. ACI J. 1964, 61, 657670. Available from: http://dx.doi.org/10.14359/7801. 49. Zollo, R. F. Fiber-reinforced Concrete: An Overview After 30 Years of Development. Cement Concrete Compos. 1997, 19 (2), 107122. Available from: http://dx.doi.org/ 10.1016/S0958-9465(96)00046-7. 50. Brandt, A. M. Fiber-reinforced Cement-based (FRC) Composites After Over 40 Years of Development in Building and Civil Engineering. Compos. Struct. 2008, 86 (13), 39. Available from: http://dx.doi.org/10.1016/j.compstruct.2008.03.006. 51. Aziz, M. A.; Paramasivam, P.; Lee, S. L. Prospects of Natural Fibres Reinforced Concretes in Construction. Int. J. Cement Compos. Lightweight Concr. 1981, 3 (2), 123132. Available from: http://dx.doi.org/10.1016/0262-5075(81)90006-3. 52. Snoeck, D.; Smetryns, P. A.; De Belie, N. Improved Multiple Cracking and Autogenous Healing in Cementitious Materials by Means of Chemically-treated Natural Fibres. Biosys. Eng. 2015, 139, 8799. Available from: http://dx.doi.org/10.1016/j. biosystemseng.2015.08.007. 53. fib. Bulletin 34. Model Code for Service Life Design. 2006, p 116. 54. Maes, M. Combined Effects of Chlorides and Sulphates on Cracked and Self-healing Concrete in Marine Environments (Ph.D. thesis); Ghent University: Ghent, Belgium, 2015. 55. Construction Technology Laboratories, Inc. Life Cycle Inventory of Slag Cement Inventory Process: Project CTL, No 312012; Construction Technology Laboratories, Inc.: IL, 2003. 56. Guine´e, J. B., Ed. Handbook on Life Cycle Assessment, Operational Guide to the ISO Standards; Kluwer Academic Publishers: Dordrecht, 2002. 57. Febelcem. Environmental Product Declaration for Belgian CementCembureau EPD Format, Cement CEM II B/M 32.5; Febelcem v.z.w: Brussels, 2012. 58. Febelcem. Environmental Product Declaration for Belgian CementCembureau EPD Format. Cement CEM III/A 42.5 N LA; Febelcem v.z.w: Brussels, 2012. 59. Maes, M.; De Belie, N. Service Life Estimation of Cracked and Healed Concrete in Marine Environment. Proceedings of Concrete Solutions, 6th International Conference on Concrete Repair, June 2022, Thessaloniki, Greece, 2016, pp 341347. 60. Le Duigou, A.; Davies, P.; Baley, C. Environmental Impact Analysis of the Production of Flax Fibres to be Used as Composite Material Reinforcement. J. Biobased Mater. Bio. 2011, 5 (1), 153165. Available from: http://dx.doi.org/10.1166/jbmb.2011.1116.

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality

10

Elena Fuente, Rocı´o Jarabo, A´ngeles Blanco and Carlos Negro Complutense University of Madrid, Madrid, Spain

10.1

Introduction

The construction sector is among the leading industries when it comes to energy consumption, with a high CO2 footprint; it is responsible for the generation of waste and pollution. This is one of the drivers for the research and implementation of sustainable building materials. In recent years, there has been a growing interest in the use of natural fibers in many kinds of composite materials for construction.1 This is the case of the fibercement industry, especially after the use of asbestos as reinforcing fibers was forbidden by the EU.2 Since then, considerable efforts have been made to develop natural fiber-reinforced cement composites for affordable infrastructures and new technologies for the production of fiber-cement using cellulose and synthetic materials as reinforcement fibers have been developed.3,4 Fiber-cement produced in flat or corrugated shape is manufactured by the Hatschek process from a mixture of Portland cement, fibers, water, and some additives in a minor proportion. Its mechanical and physical properties must accomplish the requirements for being used as a material for roofing, internal/external walls, and facades. By minor modifications of the traditional Hatschek process, it is possible to produce useful and safe fiber-cement containing cellulose fibers instead asbestos. In fact it is the most common fiber-cement product nowadays. The major arguments supporting the use of lignocellulosic fibers is the low material density, which allows the achievement of high levels of specific properties, and the low cost and high availability.5 Although unbleached Kraft pine pulps is the most commonly used source of cellulose fibers in the manufacture of asbestos-free fiber-cement composites,4,6,7 agricultural waste fibers, as bagasse from sugar cane, bamboo, and coconut fibers,8 have been studied as cheaper alternatives due to their biodegradability, low specific mass, low cost, and availability.9 Moreover, their high permeability and lack of resistance to cracking, limited the long-term durability of the composites.10 Therefore, there are still some drawbacks to solve, such as, for example, the interaction of cellulose fibers with the cement matrix, which affects the manufacture Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00011-5 © 2017 Elsevier Ltd. All rights reserved.

268 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

process and product quality. The relevance of these drawbacks increases when the origin of cellulose fibers is as waste and even more so when that waste has been treated with a nonconventional pulping process, such as the organosolv pulping process. Detailed knowledge of the fiber morphology and of the interactions with the minerals and Portland cement is necessary to evaluate their potential application and their effect on the product quality.11 12 These interactions affect the flocculation of the fiber-cement mixture, which is key for the manufacture of fiber-cement by the Hatschek process, because it affects the formation of the sheet, the retention of solids, and the drainage of the suspension. Consequently, the use of agricultural wastes modifies the process productivity and the product quality, which is affected not only by the properties of the reinforcement fibers and of the matrix, but also by the flocculation process in the fiber-cement suspension.

10.1.1 Use of hemp wastes and corn stalks in fiber-cement manufacture Hemp or Cannabis sativa is one of the most useful industrial plants because most of its parts are valuable in industry and it has been used for more than 25,000 products around the world.13 Hemp fibers have a very high mechanical strength, low density, and high durability14; furthermore, they can be between 0.9 m and 5 m long. Because of that, they are used in the production of fiber-cement, reinforced concrete, thermal insulating boards, special paper production, thermoplastic reinforced composites, and other industries.15 19 Seeds are used in oil extraction and biodiesel production.20 However, even this plant has a poorly appreciated part, which is the hemp core or the leftover bast fiber, the inner bark, also called hemp shives or hemp hurds, and which typically ends up as landfill.15,16 In 2013 more than 56,000 tons of this waste was produced in the world.21 However, the hemp core could be very interesting in new products manufacturing. Hemp shives (core of the stem) contain 48% w/w cellulose, 21% 25% w/w hemicellulose, and 17% 19% w/w lignin.22 It has been observed that after dissolving the lignin and hemicelluloses, graphene-like carbon nanosheets remain with a high strength structure and remarkable electrochemical properties that can be used for a variety of diverse applications including energy storage.23 The main advantages of hemp core fibers in fiber-cement in comparison with pine fibers are: G

G

G

G

The lower price as it is a waste. The higher durability, because of the higher chemical resistance of hemp core fibers to alkaline degradation. The high permeability and lack of resistance to crack limited the long-term durability of the fiber-cement composites containing pine fibers.10 Competitive mechanical properties: Tensile strength of 0.75 GPa, Young’s Modulus of 50 GPa. These values for pine are below 0.8 GPa and 25 GPa, respectively.24 Li et al.25 observed that hemp fibers improved the mechanical properties of fiber-cement because the better properties of the hemp fibers compared to the pine fibers. Good thermal insulation and antiseptic properties.19

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 269

G

The use of hemp shives in fiber-cement could increase the feasibility of second generation biorefinery based on this agricultural waste. Valuable products could be obtained from the lignin and hemicelluloses separated in the organosolv process while the hemp shives organosolv pulp would be used as reinforcement in fiber-cement.20

Nowadays, hemp core is commercially used in composite materials for building as “hemp-lime or hempcrete,”26 “hemp-clay or hempcrete,” and even hempboard,17 which are very efficient for insulation in buildings.27 However, all of them have limited mechanical properties, such as the compressive strength of these composites is too low for installation of bearing structures, therefore, they are usually used in combination with a wooden load-bearing frame.28 Cement matrix has a higher compressive strength and a cement-based composite with hemp shives could accomplish the building requirements. Some mixtures of noncooked hemp shives with cement were patented in the 1990s.29 However, it is not feasible to use pieces of hemp shives in the Hatschek process because of the wide range of size. Furthermore, it has been observed that the shives contain water soluble compounds in the basic environment which inhibit cement hydration.30 32 The main compounds that inhibit cement hydration are sugars and hemicelluloses can release sugar under certain conditions.32 35 The shives have to be treated by mineralization to avoid the release of these components. Balciunas et al.35 optimized the compatibility of hemp shives and cement matrix by mineralizing hemp shives aggregates with a mineral complex (Al2(SO4)3 and Ca(OH)2). Another possibility is to use an alternative matrix of MgO-cement; Stevulova et al.36 and Cigasova et al.37 studied the possibility of preparing lightweight composites based on MgO-cement and hemp shives as filler, without mineralization. These researchers also studied the effect of chemical and physical modifications of hemp shives on the physical properties (density, thermal conductivity, and water absorbability) and compressive strength of composite.38,39 Furthermore, the use of shives without pulping reduces the interaction among shives and cement matrix, which limits the tensile and compressive strengths of the composite. Stevulova et al.36 observed that the increase in the size of hemp shives decreased the compressive strength of the composite. In addition the use of hemp shives instead of free fibers could notably affect the Hatschek process. Taking this into account, it seems logical to remove lignin before using hemp shives in order to separate the cellulose fibers increasing the homogeneity of the composite material and decreasing the amount of disturbance substances, such as, for example, extractives, hemicelluloses, and sugars, in the fibers. However, the use of pulp produced from hemp core fibers as reinforcing material to produce fiber-cement has not been studied until this decade and only limited research has been published.40,41

10.1.2 Use of corn stalks in fiber-cement manufacture There are many studies of using cereal agriculture wastes, such as rice husk, rice straw.42 44 However, corn is the most produced cereal worldwide, surpassing wheat and rice.21 Furthermore, the production of this cereal increases almost every year

270 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 10.1 Evolution of production in millions of tonnes of corn in the world.21

and it has doubled in the last 20 years (Fig. 10.1).21 The large amounts of wastes generated in the production of this crop, justify the study of corn stalk material in different applications.45,46 Corn stalks consist of a pithy core with an outer layer of long fibers. Currently, corn stalks are chopped and used for forage, left on the field, or baled for animal bedding.47 Research shows that corn stalks can be used in many applications, including human consumption and as a source of industrial raw material for the production of oil, alcohol, and starch.46,47 It can also be used to make reasonably good particleboard and fiberboard.45,48 Ajayi49 and Babatunde50 used the corn stalk as reinforcement material in fibercement boards and characterized the resulting composite. They did not use corn stalk fibers, but ground them into particles smaller than 2 mm using a hammer mill. These researchers produced fiber-cement boards of 6 mm with calcium chloride as additive and obtained acceptable durability, modulus of rupture (MOR) up to 4.5 MPa, and modulus of elasticity (MOE) of 3700 MPa. These studies showed the possibility of producing a fiber-cement board with acceptable mechanical properties by using corn stalks particles. Therefore, it could be of great interest to use pulps from corn stalk instead of particles. Jarabo et al.40 studied the use of pulps from hemp core and corn stalks cooked with NaOH and anthraquinone as a source of fibers in fiber-cement, and examined how the cooking conditions affected the behavior of the fibers in the composite. They concluded that it could be feasible to replace partially the commercial pine pulp commonly used in the Hatschek process by those pulps providing soft conditions (140 C, 30 min, 10% NaOH). However, there is a remarkable lack of knowledge on the use of wastes from corn stalk in the reinforced fiber-cement as there are only a few studies on the use of pulps from corn stalks as reinforcing fibers,40,51 despite the interesting composition of corn stalk, i.e., the percentage of cellulose of up to 60%, and lignin content as low as 6%, which is lower than those for pine trees. With the purpose of contributing to building up this knowledge, this chapter aims to study the effects of the use of cooked corn stalk and hemp core, obtained by means of the organosolv process, on the Hatschek process and the fiber-cement properties, taking into account the morphology of the fibers and their interactions

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 271

with the cement suspensions, as well as the use of these pulp fibers as reinforcement in cement-based composites.

10.1.3 Why use organosolv process to obtain the pulp for fiber-cement production? Nowadays there is an increasing interest in obtaining energy and different material products from lignocellulosic wastes and nonfood biomass feedstocks in an integrated way, to approach the ideal of zero wastes. This is the main idea of biorefineries.52 The various chemical compounds forming biomass can be employed in a variety of ways. Cellulose and hemicellulose may be separated out and fermented into fuel grade alcohol and acetone. The lignin has huge potential as a useful source of aromatic chemicals for numerous industrial applications.53,54 However, the number of biorefineries that use lignocellulosic biomass as feedstock that are currently in operation or under construction is still quite low; only three are commercial and the rest, six plants, are pilot plants or demonstration plants.55 Lignin, which makes up a significant fraction of agricultural waste, is in the most of the cases burnt to obtain energy. Only two of these plants, which are pilot plants, one in France and other in Sweden, benefit from lignin as a product or as raw material for synthesis of aromatic products. Therefore, a large percentage of the lignocellulosic material is converted into energy instead of other valuable products, although value added applications for lignin would substantially improve the economics of lignocellulosic biorefineries.53 55 The main problem of biorefineries is that most separation techniques employed by industry today are not optimal for providing industrially useful lignin because only the highly pure lignin is useful for synthesis. Traditional chemical pulping processes chemically alter or degrade the lignin to the point where it is no longer acceptable for use in many of its potential applications. And the steam explosion fractionation, which is the most common treatment of lignocellulosic biomass in commercial biorefineries, is limited to making the cellulose fraction suitable for further processing without recovering any purified lignin fraction.53,54,56 The organosolv process consists of treating biomass with an organic solvent or a combination of them, for example, ethanol, methanol, acetone, in water at moderate temperatures, i.e., from 40 to 80 C.57,58 The organic solvent extracts the lignin by solving it without chemical modification,58 60 while the hemicellulose is depolymerized through acid-catalyzed hydrolysis. Because of that, the organosolv process allows to fractionate the lignocellulosic biomass into its individual major fractions in contrast to other pretreatment technologies and to Kraft pulping. Furthermore, the degradation of cellulose caused by this process is much lower than that caused by traditional pulping processes (such as for example Kraft), and cellulose keeps its good mechanical properties.57 Therefore, it is expected that the production of organosolv pulp from agricultural wastes increases and it is worth studying the use of the organosolv pulps in fibercement production.

272 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

10.2

Materials and methods

10.2.1 Pulps from agricultural wastes Two different organosolv pulp fibers from agricultural residues, Corn stalk (Zea mays L.) and Hemp core (Cannabis sativa L.), have been evaluated as alternative sources of cellulose as reinforcement fibers in fiber-cement, manufactured by the Hatschek process. Organosolv pulping was performed with a solution of ethanolamine in water in a 25-L stainless steel rotator digester with a heat exchanger system and pressure control. The pulping conditions were maintained constant as follows: liquor to solid ratio: 6:1 and digester pressure 5 7 bars. Selected experimental conditions were: solvent concentration, temperature, and pulping time. The cooked pulps were defibered in a hydrapulper and screened through a Somerville vibratory flat screen with 0.15 mm slot size; the screened pulp was washed, pressed, crumbled and stored at 4 C. Table 10.1 summarizes the cooking conditions and chemical composition to obtain four types of pulps: two low yield pulps (one for each waste) by means of hard cooking conditions and two high yield pulps by means of soft cooking conditions. Commercial refined unbleached pine Kraft pulps (35  SR and Kappa 25.4 mL) were used as reference since this pulp is commonly used to provide cellulose fibers in the manufacture of fiber-cement by the Hatschek process. Pulps were characterized by means of a scanning electron microscope (SEM), JEOL, model JM-6400, and a Morfi, V7.9.13.E (Techpap, France). Fig. 10.2 and Table 10.2 show the SEM images and morphological characterizations of the pulps, respectively. The degree of corn stalks pulps refining was measured using a Canadian Standard Freeness (CSF), according to ISO 5267/2.61

Table 10.1 Cooking conditions and chemical composition of hemp core and corn stalk using the organosolv chemical process Corn stalk

Ethanolamine concentration (%) T ( C) Cooking time (min) Yield (%) Kappa index (mL/g) Uncooked mass (%) Intrinsic viscosity (mL/g) Extractives (%)

Hemp core

C1

C2

H1

H2

60 185 90 52 4.4 3.3 753 0.45

40 155 30 71 16.8 5.2 873 0.55

60 180 90 54 25.3 6.9 256 0.49

40 155 30 75 113.4 12.4 376 0.61

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 273

Figure 10.2 SEM images of the organosolv pulps.

Table 10.2

Morphological characterization of pulps

Pulps

C1

C2

H1

H2

Pine

20.5 342 29.4 0.14 17.8 10.4 1.66

14.3 367 30.8 0.18 10.9 8.14 1.64

45.2 345 24.0 0.06 13.9 6.52 1.68

28.5 352 27.3 0.10 11.7 6.08 1.48

10.8 456 25.5 0.19 19.1 7.3 1.62

0.29 0.59

0.33 0.75

0.20 0.59

0.57 1.07

30081 687

30574 696

8947

15474

Fibers Number of fibers (106/g) Average length (µm) Average width (µm) Coarseness (mg/m) Kinked fibers (%) Average Curl (%) Microfibrils (%)

Vessels Number of vessels per g (106/g) Length average (mm)

Fines Number of fines CSFa (mL) a

CSF, Canadian Standard Freeness.

49192 192

274 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

SEM images show that the organosolv cooking successfully separated the fibers, but there are some residues or “detrimental material” (uncooked particles) in the pulps; some of them are counted as “vessels” by the Morfi analyzer, although a vessel is actually a conduct or a piece of a conduct which carried the sap of the tree. The particles with low aspect ratio and width larger than the diameter of fibers are considered as vessels by the Morfi analyzer. Because of that, it is possible to compare the amount of uncooked particles of the pulps. These residues are larger and more abundant in the pulp H2 than in the others (Table 10.2). Therefore, the use of this pulp is more close to the use of hemp core pieces without pulping. It is noticeable that the length of the fibers and fines obtained from the organosolv pulps are much shorter than that of the pine fibers, which is related to the origin of these fibers. This could improve their dispersion in the matrix. The intensity of cooking increases the number of fibers, as a result of the higher dissolution of lignin and better separation among fibers. However, it decreases slightly the length and width, and notably the coarseness; that could indicate some degradation of the fiber or some external defibration resulting in some smaller or thinner fibers, which agrees with the higher percentage of microfibrils. The decrease in intrinsic viscosity indicates that some degradation of cellulose could be occurred. However, that decrease could be due to the lower amount of uncooked material. The uncooked material makes difficult the flow of the pulp through the capillary viscometer. C1 and H1, presented higher values of kinked fibers and average curl than those for the other pulps. This is due to the more aggressive cooking conditions used to obtain pulps C1 and H1, which increase the fibers degradation during the process.62 Pine pulp is refined to increase the microfibrils, which increase the interaction among fibers and between them and the matrix. Pulp refining produces an external fibrillation of the cells, which allows an increase in flexibility and the formation of bridges with other fibers. Refining also increases the percentage of fines in the pulp as many of the fibers are cut. This explains the higher percentage of fines in the pine pulp. The presence of fines improves solids retention during the formation of fiber-cement11 but it can reduce the resistance as it means there are a lower number of long fibers, which could affect mechanical properties. The percentage of microfibrills in organosolv pulps were similar to that in the pine fibers, therefore, they could be used without refining. Furthermore, refining is a high energy-consuming treatment, which increases the cost of the fibers. However, the CSF of the organosolv pulps is much higher than that for the pine pulp, because of the lower percentage of fines.

10.2.2 Fiber-cement composition The basic composition of the fiber-cement suspensions is summarized in Table 10.3. Five fiber-cement suspensions were prepared with different pulps: composites MC1 and MC2 were prepared by using the pulp C1 and C2 (Table 10.2), respectively, and composites MH1 and MH2 with pulps H1 and H2, respectively.

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 275

Composition of the suspensions to prepare the fiber-cement

Table 10.3

Raw materials

Dry weight (%)

Cellulose pulp ASTM II Portland cement Microsilica

5 91 4

Finally, the pine Kraft pulp was used to prepare the fiber-cement composite called MR, used as reference. The air curing process was used to prepare the fiber-cements and no synthetic fibers (for example, polyvinyl alcohol [PVA]) were used in order to enhance the effect of the cellulose fibers in the properties of the composites. Only one fibercement mixture was prepared with PVA fibers, in order to illustrate how the presence of PVA fibers hides the effect of cellulose fibers. It was obtained by using 3.2% of cellulose plus 1.8% of PVA fibers instead of the 5% of cellulose.

10.2.3 Flocculation of fiber-cement suspension The Hatschek process consists of the formation of thin paper-like films that are placed one on the other until the desired sheet thickness is reached. However, the different density, chemical composition, and hygroscopic character of cellulose fibers makes the interaction between these fibers and Portland cement much more complex30,63 and, therefore, a suitable flocculation is needed when using pulp fibers to improve the retention of minerals and fine fiber fragments during the formation of films and to reduce the recirculating load of fine particles in the water system.64 The flocculant used to study the behavior of fiber-cement suspensions and to prepare the fiber-cement specimens was an anionic polyacrylamide (APAM) with a molecular weight of 7.4 3 106 g/mol and a charge density of 13.4%, commonly used in the industrial Hatschek process.65 Flocculant was dissolved in distilled water to prepare solutions of APAM with a concentration of 1.5 g/L. A FBRM (focused beam reflectance measurement) M500L probe supplied by Lasentec and manufactured by Mettler Toledo, USA, was used to monitor the flocculation process and to determine the floc size, stability, and reflocculation ability. The FBRM offers the possibility of in situ and in real time particles characterization in a wide solid concentration interval. A laser beam is generated by a diode and focused on a focal point in the plane next to the external surface of the probe window, which is inside the suspension. The focal point describes a circular path at a constant speed of 2000 m/s due to the rotation of the focusing lens. When a particle intercepts the focal point path, the reflected light reaches the detector through the probe and the optical fiber. The detector receives light pulses. A chord length of the intercepting particle is determined as the product of the time duration of the light pulse by the linear speed of the focal point movement. Thousands of chord length measurements are collected per second, producing a histogram in which the

276 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

80

MR MH1 MH2 MC1 MC2

800 rpm

400 rpm

Evolution of flocs

40

30

20

100 ppm APAM

10 800

900

1000

Reflocculation

50

400 rpm

Deflocculation

60

Flocculation

Mean chord length (µm)

70

1100

1200

1300

1400

1500

Time (s) Figure 10.3 Evolution of the mean chord size during flocculation, defloculation, and reflocculation processes at different stirring speeds.41,51

number of the observed counts is sorted in several chord length bins over the range from 0.2 to 2000 µm (set at 5 s in this case). Each distribution is a function of the size and shape of particles in suspension. The evolution of the chord size distribution reflects the aggregation and dispersion of particles. Total number of counts measured per second, counts in specific size regions (population), mean chord length, and other statistical parameters can be easily calculated from the data.66,67 In a typical trial, the probe was immersed into 400 mL of fiber-cement suspension consisting of diluted water saturated with Ca(OH)2 from Portland cement, microsilica, and pulp fiber, stirred at 800 rpm. After 10 min, stirring speed intensity was reduced to 400 rpm. 100 ppm of APAM was added 5 min later, to induce flocculation, and the evolution of the flocs was studied at 400 rpm during 4 min. Then, the stirring speed intensity was increased to 800 rpm to break down the formed flocs (deflocculation) during 2 min and, finally, the stirring speed intensity was reduced again to 400 rpm to induce the reflocculation of the system (Fig. 10.3).40

10.2.4 Retention and drainage of fiber-cement suspension A VDT (vacuum drainage tester) were performed to study the retention and drainage, previously described by Negro et al.65 It has two jars separated by a barrier: the upper jar keeps the fiber-cement suspensions stirred up to the addition of the flocculant dosage. The second jar contains a mesh in the bottom to carry out the

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 277

dewatering of the suspension and it is connected to a vacuum pump and to a probe where filtrate is stored and weighted in real time and a final volume of filtrate is measured. In a typical trial, 400 mL of fiber-cement suspension, prepared with water saturated in Ca(OH)2, was stirred at 600 rpm for 6 min in the upper jar. Then, stirring intensity was decreased to 300 rpm and 100 ppm of APAM was added 5 min later. After 15 s of contact time between flocculant and mixture, the stirring was stopped, the barrier was removed, and the suspension was drained to the second jar in which an 18 mesh sieve was placed. The suspension was drained under vacuum (0.2 atm) through the filter and a computerized balance recorded the mass of drained water over time. The drainage curve was analyzed. Retention and cake humidity were gravimetrically determined from the formed cake.65,68

10.2.5 Preparation of fiber-cement specimens Fiber-cement specimens were prepared in the laboratory by means of a suspension vacuum dewatering technique followed by the pressing technique, described in detail by Savastano Jr. et al.9 The matrix materials and fibers were added and dispersed in Ca(OH)2 saturated water with a solids concentration of 20%. After stirring for 5 min, the suspension was rapidly transferred to an evacuable 125 3 125 mm2 casting box and an initial vacuum (60 kPa gage) drawn until the bulk of the excess water was removed and a solid surface formed. The moist pad was tamped flat and vacuum reapplied for 2 min. The pad was then transferred to an oiled steel plate and a wire mesh placed on top. Then, the pads were pressed under 3.2 MPa for 5 min. The cement-based composites were molded in plates measuring 200 3 200 mm and around 6 mm thick. Three pads were prepared for each formulation. After two days; the pads were removed from the sealed bags and were placed in water. Twenty-six days later, the pads were removed from the water and four flexural test specimens of 200 3 50 mm2 were wet diamond sawn from each pad. Eight pads were prepared to provide sufficient specimens for the determination of mechanical properties and four pads were prepared for the determination of their physical properties.

10.2.6 Mechanical and physical properties of fiber-cement Mechanical characterization was based on the RILEM recommendations 49.69 Bending tests were performed in the universal testing machine Emic DL-30,000 equipped with 1 kN load cell. A four point bending configuration was employed in the determination of the load (F) vs deflection (δ) curves. MOR, proportional limit (LOP), MOE, and specific energy (SE) of the specimens were calculated from the curves as specified by equations Eqs. (10.1) (10.4). A major span (L) of 100 mm and a constant deflection rate of 0.5 mm/min were used in the bending tests.9 Test data was digitally recorded and reduced using automatic data collection and processing facilities. Eight specimens were tested for each composite formulation.

278 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

MOR 5

FM L bd 2

(10.1)

LOP 5

FLOP L bd2

(10.2)

276FM L3 1296bδE d3

(10.3)

Energy absorbed bd

(10.4)

MOE 5 SE 5

where: FM is the maximum load; FLOP is the highest load that the specimen can sustain without any deviation from linear proportionality in the load-deflection curve; b and d are the specimen width and the specimen thickness, respectively. The SE, Eq. (10.4), is defined as energy absorbed during the flexural test divided by the specimen cross-sectional area. The energy absorbed is the area under the load-deflection curve at the point corresponding to a reduction in load carrying capacity to 50% of the maximum reached.70 Furthermore, load-deflection curves were converted to stress (σ) versus strain (γ) curves by using the equations Eqs. (10.5) and (10.6). γ5

δ L

(10.5)

σ5

FL bd2

(10.6)

The dimensions of the specimen cross section (b and d) were determined for at least three different points by means of a caliper. Thickness varied from side to side of the specimen by less than 5%. The mechanical properties of the fiber-cement specimens were measured 28 days after the pads were prepared. Water absorption, bulk density, and apparent porosity values were obtained at 28 days following the procedures specified in ASTM C 948 81.71 Four specimens were used to determine these properties.

10.3

Results

10.3.1 Effect of the use of organosolv pulps on the Hatschek process The flocculation process is important in the Hatschek process when fiber-cement is produced. This is due to its influence on mineral fines retention, and on the

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 279

dewatering rate of the thin fiber-cement sheets in the sieve rotating in contact with the suspension in the vat, which affects machine productivity and, as a consequence, the product formation and properties. Therefore, to determine the effect of using organosolv pulps on the Hatschek process, the flocculation process and the floc properties of the fiber-cement suspensions prepared with the different pulps have been studied and compared to those prepared with the commercial Kraft pine pulp. Then, the solids retention and drainage curves were determined. As shown in Fig. 10.3, the addition of APAM to the suspension caused a fast increase in the mean chord length due to the aggregation of particles to form larger flocs. A maximum value of this is reached at 10 or 15 s and then it starts decreasing at a different rate depending on the stability and cohesiveness of the flocs in such hydrodynamic conditions (evolution of flocs). When the stirring was increased to 800 rpm part of the remaining flocs were broken, decreasing the mean chord size (deflocculation). After this, the reflocculation ability of the system is shown by the slight increase in the mean chord size when the stirring speed was decreased to 400 rpm. The largest maximum mean chord size immediately after the addition of APAM was reached by the MR suspension. It can be observed that the use of corn organosolv pulps did not affect drastically the flocculation process, with the exception of a lower mean chord size increment during the flocculation stage. However, the use of hemp core pulps affected notably the flocculation process and the floc properties. The evolution of the mean chord size of the suspensions with C1, C2, H1, and H2 pulps indicates that the formed flocs were more stable than those formed from MR. The use of H2 increased notably the size of the flocs during the evolution stage, which means that the retention and drainage processes can be improved. Moreover, after the deflocculation and reflocculation stages, the mean chord size in MH2 suspension remained larger than the values obtained for MR. The use of H2 decreases the flocs size, but increases their stability as proved by the less steep slope of the curve of mean chord size versus time during the floc evolution stage and the larger final mean chord size compared to MR. The results indicate that the flocculation process and the floc properties are affected by the source of fibers and by the cooking conditions, especially in the case of using hemp core, and that, in this case, the use of soft cooking conditions favors the formation of larger and more stable flocs when APAM was added to the fiber-cement suspension. Large flocs could affect the formation of the fiber-cement sheets, but it could also favor solids retention and water drainage. Drainage takes place in two steps (Fig. 10.4): first, the suspension is filtrated and a cake is formed with a fast water removal, which corresponds to the first part of the drainage curves (linear part with a high slope); secondly, the cake is compressed and thickened and the water removal rate decreases towards zero. During the first stage, only water free to move among the flocs is removed, while part of the water inside the flocs is removed during the compression stage. The loss of most solids with the filtrate takes place during the first stage while the second stage determines the final humidity and the formation properties of the cake. Drainage time can be obtained as the time required to reducing the drainage curve slope to zero.

280 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Filtrate

Total filtrate drained

Step 1

Step 2

Time

Drainage time

Figure 10.4 Scheme of the two steps in the drainage of suspensions. 500

Filtrate (g)

400

300

MR MH1 MH2 MC1 MC2

200

100

0 0

200

400

600

800

1000

1200

1400

1600

1800

Time (s)

Figure 10.5 Drainage curves of the fiber-cement suspensions.41,51

Fig. 10.5 shows that the drainage of the mixture MH2, containing H2 pulp was the fastest and that the suspension containing H1 pulp, MH1, was the slowest, as expected from the flocculation studies. No strong variations in drainage rate were observed for MC1, containing corn stalk pulp; in this case the drainage rate was similar to that for MR and only the final recovered weight of filtrate was higher. However, MC2 drained almost as fast as MH2. The initial drainage rate is notably affected by the intensity of cooking conditions of the pulp and it was the slowest for MH1, which contains low yield pulp. There is not an appreciable compression stage in the drainage curve of MH2 and MC2. In the case of MH2, the large floc size allows the water to drain very fast and the absence of an appreciable

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 281

Table 10.4 Solids retention and humidity of cake after draining Fiber source

Retention (%)

Humidity (%)

H1 H2 C1 C2 Pine

65 6 2 19 6 6 35 6 2 17 6 2 57 6 2

51 6 1 65 6 1 58 6 3 54 6 1 58 6 1

compression stage indicates that the flocs were not easily deformed. The final weight of recovered filtrate is higher when organosolv pulps are used, especially when they are obtained by soft cooking conditions. It is worth noting that the final weight of filtrate, in these cases was around 400 g. Considering that only 400 mL of suspension was used, large amounts of minerals have passed through the wire. Surprisingly, the retention of solids is very low when the soft cooked pulps were used, despite the large floc size, and the highest solids retention was obtained with MH1, although the presence of this pulp decreased notably the floc size (Table 10.4). These values, together with the drainage curves, demonstrate that many of the mineral solids are lost with the drained water during the first drainage stage in the cases of MH2 and MC2 with only some of the mineral solids remaining in the cake and most of the fibers, whose high hygroscopicity explains the high humidity of the cakes formed in these two cases. The humidity of the cake decreases when the retention increases, and this is due to the higher mineral content of the cake (water absorbed by the mineral particles is much less than water absorbed by the fibers). The small mean chord length reached after flocculation of MH1, Fig. 10.3, indicates that there are a high percentage of small flocs and small particles in MH1. Therefore, they block the spaces among the flocs and decrease the drainage rate during the first step, which contributes to the increase in the retention of solids, but causes an increase in the drainage time. In the case of MH2 and MR, the spaces among the flocs are larger than those for MH1, and water can drain very fast and carry with it part of the nonflocculated minerals. This explains the high initial drainage rate observed and the lower retention compared with MH1. The flocs formed from MR are much weaker than those formed from MH2 and they are easily deformed, and, therefore, they are broken, deformed, and compressed by the vacuum forces and the shearing forces during the drainage, and the spaces among the flocs are blocked, which retains many of the mineral solids, in the case of MR, although decreases the drainage rate (compression stage). Furthermore, the interaction of Kraft pulp, which has a number of fines quite a lot higher than the organosolv pulps, with the flocculants and minerals was higher than that for the organosolv pulps, increasing retention of solids.

282 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Lignin covers the cellulosic surface and interferes in the interaction among fibers, flocculant, and mineral particles, because lignin, which is hydrophobic compared to cellulose, acts as a cross-linking agent within the walls of the cellules and avoids swelling and external fibrillation of the fiber, decreasing the specific surface able to interact with flocculant and mineral particles.72 When soft cooking conditions were used, especially in the case of H2, there was a high amount of lignin remaining, as shown by the high kappa index for H2 (Table 10.1). Therefore, mineral particles were not attached to the fibers in the flocs and they were drained with the water. Furthermore, the presence of large detrimental particles, detected as “vessels” in MH2 is related to the low solids retention and to the absence of a compression stage in the drainage curve of MH2, as the interaction between the detrimental particles and minerals is low. In summary, the effect of using organosolv pulps from hemp core and corn stalks on the Hatschek process depends strongly on the origin of the pulp and on the cooking conditions. Soft cooking conditions strongly decrease solids retention in the Hatcheck process. The use of low yield hemp core organosolv pulps increases solids retention in comparison with the Kraft pine pulp, however, the drainage will be slower, affecting the productivity of the machine. These effects could be solved if only a portion of pine fibers were replaced by the organosolv pulps.

10.3.2 How do the organosolv pulps affect to the Hatschek product properties? The quality of the organosolv pulp fibers is lower than the reference pulp. They are shorter and contain detrimental materials, which can affect the product. It could be especially relevant in the case of H2, because the size of the detrimental materials was the highest (Fig. 10.2). Table 10.5 shows that despite the differences in solids retention and drainage during the Hatschek process (slurry vacuum dewatering technique) the physical properties of the fiber-cement specimens with organosolv pulps were similar to those with Kraft pine pulp. The exception was fiber-cement MC2, the values of the water adsorption and the apparent porosity of the specimens with the organosolv Table 10.5

Physical properties of fiber-cement specimens

Fibercement

Fiber source

Bulk density (g/cm3)

Apparent porosity (%w)

Water absorption (%w)

MH1 MH2 MC1 MC2 MR

H1 H2 C1 C2 Kraft Pine pulp

1.725 1.700 1.677 1.689 1.713

33.7 32.8 32.6 34.4 34.3

19.5 19.4 19.5 20.4 20.1

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 283

Strain for MRPVA (mm/mm) 0

2x10–2

4x10–2

6x10–2

8x10–2

14 MR MH1 MH2 MC1 MC2 MRPVA

12

Stress (MPa)

10

8

6

4

2

0 0

5x10–3

10x10–3

15x10–3

20x10–3

Strain (mm/mm)

Figure 10.6 Stress strain curves of the fiber-cement composites.

pulps were slightly lower than those for the specimen used for reference. However, the value of the bulk density of the specimens MH1, MH2, MC1, and MC2 was very close to the reference composite. It is not possible to affirm the same statement on the mechanical properties. Fig. 10.6 shows the strain stress curves obtained for the specimens prepared with the five mixtures. The curves are the mean curve obtained from all the repetitions. It is worth noting that the stress strain curve of the specimens prepared with MH1 is very similar to that of MR, but the maximum strain is higher. It must be taken into account that the specimens have been prepared without PVA fibers. The effect of PVA on mechanical properties is so high that no significant differences could be observed when PVA fibers were added. In Fig. 10.6 the stress versus strain curve of a specimen with the Kraft pine pulp and PVA fibers (specimen called MRPVA) has been included to show this fact. Therefore, instead of focusing the study on the absolute values of the mechanical properties, it was focused on the differences between the properties of MR and those of MH1, MH2, MC1, and MC2, which indicate the effect of using organosolv pulps on the fibercement composites’ mechanical properties. It is worth noting that the maximum elongation before the failure of the specimen of fiber-cement MH1 is higher than that for MR, but the maximum load was lower for all the specimens with organosolv pulps. High elongation and load and

284 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 10.6 Reduction or increment in percentage (%) according to effect of replacing pine fibers by organosolv pulps on the mechanical properties of fiber-cement Fiber-cement

MOR

MOE

LOP

SE

MH1 MH2 MC1 MC2 Values for MR

19.8 49 19.1 15.8 8.1 MPa

6.6 1 12.4 2.4 1 6.4 15.7 GPa

31.4 61.1 24.1 21.3 7.45 MPa

25 75 56 58 0.48 kJ/m2

the large area under the stress versus strain curves can be associated with a good array of the fibers and solid particles into the composite.73 The other specimens have a lower ability to elongate before failure than MH1 and MR, as shown by the short strain hardening regions. Therefore, the specimens prepared from fiber-cement MH1 may be the ones with the best mechanical properties among the fiber-cement probes prepared with the studied organosolv fibers. Table 10.6 shows the effect of using organosolv pulps on the mechanical properties. The specimens of the fiber-cement MR have some superior mechanical properties. The effect of using organosolv pulps in fiber-cement suspensions and specimens is closely related to the fibers’ morphology and to the flocculation process, which depend on the cooking conditions. The addition of APAM to the fiber-cement suspensions prepared with the pulps obtained with hard cooking conditions (MH1 and MC1) induced the formation of small flocs, which slowed down the drainage process and enhanced the compression stage of the drainage (Fig. 10.5). This increased the retention of solids (Table 10.4), overcoming that for the MR suspension, partly by means of their interaction with the fiber and partly by blocking of the small voids among flocs. This is related to the better performance in mechanical properties of the fiber-cement specimens with MH1 and MC1 compared to the reference fiber-cement (MR); especially in the case of MH1. However, the use of H1 instead of Kraft pine pulp reduces slightly the values of MOR, MOE, LOP, and SE. These effects are much lower than fiber-cement MH2, which is the one with the most negative impact on the mechanical properties. The cooking conditions have a notable impact on mechanical properties. These results can be explained by the effect of cooking conditions on pulp morphology. The presence of large particles of detrimental material, detected as “vessels” by the Morfi analyzer in MH2 is related to the poor mechanical properties. The interaction of those “vessels” with the cement matrix is very poor and because of that, the MOR is quite low, because of the lower adhesion in the interfaces between detrimental material particles and the matrix. The hard cooking process at hard conditions reduces notably the number and size of detrimental material particles to values even lower than those for the C1 and C2 pulps. This leads to a more homogeneous distribution of the fibers in the suspension, which will improve the

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 285

interaction between them and the matrix causing this fiber-cement specimen to be better formed. Furthermore, the increase in the number of microfibrils and the slight decrease in coarseness cause in increase in the surface area that is available for chemical bonding between the fibers and the cement matrix. In addition, at the level of the microfibrils, the longitudinal tension in the fibers could involve not only tensional stresses, but also torsion effects as well. This torsion stress, which initially impedes the tension, could subside or relax as a result of the repeated stress.41 It is also probable that rearrangements and reorientations of the cellulose microfibrils and/or changes in the crystallinity fraction occur in the fibers.17 The use of corn stalk pulps reduced slightly the MOR, however, it affected notably the total SE absorbed by the fiber-cement, making it more rigid and fragile. No strong differences were observed. The number of “vessels” measured in both MC1 and MC2 pulps was similar as the effect of cooking conditions on the detrimental nonfibrillated material was not high in this case. This confirms the influence of those vessels on the interaction between fibers and cement. In summary, the use of organosolv pulps from hemp core and corn stalks is not detrimental to the physical properties of the fiber-cement product manufactured by the Hatschek process. However, in the absence of other kind of fibers, such as Kraft pine pulp or PVA fibers, it can deteriorate the mechanical properties of the product, but they are better than those obtained for the composites with corn stalk pieces by Ajayi,49 with the exception of the mechanical properties of MH2. Nevertheless, the effect on mechanical properties depends strongly on the origin of the pulp and on the cooking conditions and it is low when hemp core pulps of low yield are used. It is possible to suggest that this effect could be negligible when only a portion of Kraft pine pulp is replaced by the organosolv fibers or when the organosolv fibers are combined with PVA fibers.

10.4 G

G

G

G

G

Concluding remarks

The Hatschek process with air curing technology can contribute to valorizing the hemp core and corn stalks, when they are cooked by an organosolv process, without a significant impact on the process and product quality, providing that only a portion of commercial cellulosic fibers is replaced by the organosolv pulp. The use of low yield pulps, obtained with hard cooking conditions, is recommended instead of the high yield pulps, especially in the case of the valorization of hemp core fibers. The effect of cooking conditions is less important when the waste being valorized is corn stalks. The use of hemp core or corn stalks organosolv pulps does not affect the physical properties of the product. The effect on mechanical properties will be negligible if a portion of commercial cellulosic fibers is replaced by low yield hemp core organosolv pulp or by corn stalk organosolv pulp, however the use of high yield organosolv hemp core pulp can decrease the SE absorption and the MOR of the product. The presence of large particles of detrimental material in the pulp (1 mm or larger) can reduce the potential of that pulp to be valorized in the Hatschek process.

286 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Acknowledgments The authors thank the Ministry of Science and Innovation of Spain for funding the project: CTM2007 66793 C03 03 and the scholarship of Rocı´o Jarabo to accomplish her PhD Thesis. This research is part of a collaborative work between the Universidad Complutense de Madrid and the University of Sa˜o Paulo; the authors would also like to acknowledge Prof. Holmer Savastano Junior and all assistance offered by Prof. Gustavo H.D. Tonoli (Federal University of Lavras) and Mr. Zaqueu Dias de Freitas at the Laboratory of Construction and Ambience of the University of Sa˜o Paulo, Brazil. The authors would also like to acknowledge the contribution of the LEPAMAP group of the University of Girona in supplying organosolv pulps.

References 1. Prusty, J. K.; Patro, S. K. Properties of Fresh and Hardened Concrete Using Agro-waste as Partial Replacement of Coarse Aggregate-A Review. Constr. Build. Mater. 2015, 82, 101 113. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.02.063. 2. Commission of The European Communities. Commission Ditective 1999/77/CE, J. Eur. Commun. L 217, 18 20, 1999. 3. Hannant, D. J. Fibre Reinforcement in the Cement and Concrete Industry. Mater. Sci. Technol. 1995, 11 (9), 853 861. Available from: http://dx.doi.org/10.1179/mst.1995.11.9.853. 4. Coutts, R. S. P. A Review of Australian Research into Natural Fibre Cement Composites. Cement Concrete Compos. 2005, 27 (5), 518 526. Available from: http:// dx.doi.org/10.1016/j.cemconcomp.2004.09.003. 5. Pickering, K. L.; Aruan Efendy, M. G.; Le, T. M. A Review of Recent Developments in Natural Fibre Composites and Their Mechanical Performance.. Compos. Part A 2016, 83, 98 112. Available from: http://dx.doi.org/10.1016/j.compositesa.2015.08.038. 6. Bartl, A.; Mihalyi, B.; Marini, I. Applications of Renewable Fibrous Materials. Chem. Biochem. Eng. Q. 2004, 18 (1), 21 28. 7. Ismail, M. A. Compressive and Tensile Strength of Natural Fibre-reinforced Cement base Composites. Al-Rafidain Eng. J. 2007, 15 (2), 42 51. 8. Lertwattanaruk, P.; Suntijitto, A. Properties of Natural Fiber Cement Materials Containing Coconut Coir and Oil Palm Fibers for Residential Building Applications. Constr. Build. Mater. 2015, 94, 664 669. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2015.07.154. 9. Savastano, J. H.; Warden, P. G.; Coutts, R. S. P. Brazilian Waste Fibres as Reinforcement for Cement-based Composites. Cement Concrete Compos. 2000, 22 (5), 379 384. Available from: http://dx.doi.org/10.1016/S0958-9465(00)00034-2. 10. Savastano, J. H.; Santos, S. F.; Radonjic, M.; Soboyejo, W. O. Fracture and Fatigue of Natural Fiber-reinforced Cementitious Composites. Cement Concrete Compos. 2009, 31 (4), 232 243. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2009.02.006. 11. Tonoli, G. H. D.; Fuente, E.; Monte, C.; Savastano, H., Jr.; Rocco Lahr, F. A.; Blanco, A. Effect of Fibre Morphology on Flocculation of Fibre cement Suspensions. Cement Concrete Res. 2009, 39 (11), 1017 1022. Available from: http://dx.doi.org/10.1016/j. cemconres.2009.07.010. 12. Jarabo, R.; Monte, M. C.; Blanco, A.; Negro, C.; Tijero, J. Characterisation of Agricultural Residues Used as a Source of Fibres for Fibre-cement Production. Ind.

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 287

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

25.

26.

Crops Prod. 2012, 36 (1), 14 21. Available from: http://dx.doi.org/10.1016/j. indcrop.2011.07.029. Salentijn, E. M. J.; Zhang, Q.; Amaducci, S.; Yang, M.; Trindadea, L. M. New Developments in Fiber Hemp (Cannabis sativa L.) Breeding. Ind. Crops Prod. 2015, 68, 32 41. Available from: http://dx.doi.org/10.1016/j.indcrop.2014.08.011. Van Der Werf, H. M. G. Life Cycle Analysis of Field Production of Fibre Hemp, the Effect of Production Practices on Environmental Impacts. Euphytica 2004, 140 (1 2), 13 23. Troe¨dec, M.; Rachini, A.; Peyratout, C.; Rossignol, S.; Max, E.; Kaftan, O., et al. Influence of Chemical Treatments on Adhesion Properties of Hemp Fibres. J. Colloid Interface Sci. 2011, 356 (1), 303 310. Available from: http://dx.doi.org/10.1016/j. jcis.2010.12.066. Sedan, D.; Pagnoux, C.; Smith, A.; Chotard, T. Mechanical Properties of Hemp Fibre Reinforced Cement: Influence of the Fibre/matrix Interaction. J. Eur. Ceram. Soc. 2008, 28 (1), 183 192. Available from: http://dx.doi.org/10.1016/j.jeurceramsoc.2007.05.019. Placet, V. Characterization of the Thermo-mechanical Behaviour of Hemp Fibres Intended for the Manufacturing of High Performance Composites. Compos. Part A: Appl. Sci. Manufact. 2009, 40 (8), 1111 1118. Available from: http://dx.doi.org/ 10.1016/j.compositesa.2009.04.031. Dalmay, P.; Smith, A.; Chotard, T.; Sahay-Turner, P.; Gloaguen, V.; Krausz, P. Properties of Cellulosic Fibre Reinforced Plaster: Influence of Hemp or Flax Fibres on the Properties of Set Gypsum. J. Mater. Sci. 2010, 45 (3), 793 803. Available from: http://dx.doi.org/10.1007/s10853-009-4002-x. Elfordy, S.; Lucas, F.; Tancret, F.; Scudeller, Y.; Goudet, L. Mechanical and Thermal Properties of Lime and Hemp Concrete (“hempcrete”) Manufactured by a Projection Process. Constr. Build. Mater. 2008, 22 (10), 2117 2123. Available from: http://dx.doi. org/10.1016/j.conbuildmat.2007.07.016. Li, S. Y.; Stuart, J. D.; Li, Y.; Parnas, R. S. The Feasibility of Converting Cannabissativa L. oil into Biodiesel. Bioresour. Technol. 2010, 101 (21), 8457 8460. Available from: http://dx.doi.org/10.1016/j.biortech.2010.05.064. Faostat. Food and Agriculture Organization of The United Nations Statistics Division Database. http://faostat3.fao.org/browse/Q/QC/E (accessed Feb 01, 2016). Mostefai, N.; Hamzaoui, R.; Guessasma, S.; Aw, A.; Nouri, H. Microstructure and Mechanical Performance of Modified Hemp Fibre and Shiv Mortars: Discovering the Optimal Formulation. Mater. Des. 2015, 84, 359 371. Available from: http://dx.doi.org/ 10.1016/j.matdes.2015.06.102. Wang, H.; Xu, Z.; Kohandehghan, A.; Li, Z.; Cui, K.; Tan, X., et al. Interconnected Carbon Nanosheets Derived from Hemp for Ultrafast Supercapacitors with High Energy. ACS Nano 2013, 7 (6), 5131 5141. Available from: http://dx.doi.org/10.1021/ nn400731g. Mott, L.; Groom, L.; Shaler, S. Mechanical Properties of Individual Southern Pine Fibers. Part II. Comparison of Earlywood and Latewood Fibers with Respect to Tree Height and Juvenility. Wood Fiber Sci. 2007, 34 (2), 221 237. Li, Z.; Wang, L.; Wang, X. Compressive and Flexural Properties of Hemp Fiber Reinforced Concrete. Fiber Polym. 2004, 5 (3), 187 197. Available from: http://dx.doi. org/10.1007/BF02902998. Bruijna, P. B.; Jeppssona, K. H.; Sandinb, K.; Nilsson, C. Mechanical Properties of Lime-hemp Concrete Containing Shives and Fibres. Biosys. Eng. 2009, 103 (4), 474 479. Available from: http://dx.doi.org/10.1016/j.biosystemseng.2009.02.005.

288 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

27. Chabannes, M.; Garcia-Diaz, E.; Clerc, L.; Benezet, J. C. Studying the Hardening and Mechanical Performances of Rice Husk and Hemp-based Building Materials Cured Under Natural and Accelerated Carbonation. Constr. Build. Mater. 2015, 94, 105 115. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.06.032. 28. Gross, C. H.; Walker, P. Racking Performance of Timber Studwork and Hemp-lime Walling. Constr. Build. Mater. 2014, 66, 429 435. Available from: http://dx.doi.org/ 10.1016/j.conbuildmat.2014.05.054. 29. Rasetti, C. Process for the Treatment of a Hemp by-product, Aggregates Obtained by this Process and Concrete Incorporating these Aggregates. EP0384815, 1990. 30. Miller, D. P.; Moslemi, A. A. Wood cement Composites: Effect of Model Compounds on Hydration Characteristics and Tensile Strength. Wood Fibre Sci. 1991, 23 (4), 472 482. 31. Semple, K. E.; Evans, P. D. Wood cement Composites—Suitability of Western Australian Malle Eucalypt, Blue Gum and Melaleucas; RIRDC/Land and Water Australia/FWPRDC/MDBC: Kingston, 2004. 32. Diquelou, Y.; Gourlay, E.; Arnaud, L.; Kurek, B. Impact of Hemp Shive on Cement Setting and Hardening: Influence of the Extracted Components from the Aggregates and Study of the Interfaces with the Inorganic Matrix. Cement Concrete Compos. 2015, 55, 112 121. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2014.09.004. 33. Thomas, N. L.; Birchall, J. D. The Retarding Action of Sugars on Cement Hydration. Cement Concrete Res. 1983, 13 (6), 830 842. Available from: http://dx.doi.org/ 10.1016/0008-8846(83)%2090084-4. 34. Bilba, K.; Arsene, M.-A.; Ouensanga, A. Sugar Cane Bagasse Fibre Reinforced Cement Composites. Part I. Influence of the Botanical Components of Bagasse on the Setting of Bagasse/cement Composite. Cement Concrete Compos. 2003, 25 (1), 1 96. Available from: http://dx.doi.org/10.1016/s0958-9465(02)3-3. 35. Balciunas, G.; Pundiene, I.; Lekunaite-Lukosiune, L.; Vejelis, S.; Korjakins, A. Impact of Hemp Shives Aggregate Mineralization on Physical-mechanical Properties and Structure of Composite with Cementitious Binding Material. Ind. Crops Prod. 2015, 77, 724 734. Available from: http://dx.doi.org/10.1016/j.indcrop.2015.09.011. 36. Stevulova, N.; Kidalova, L.; Cigasova, J.; Junak, J.; Sicakova, A.; Terpakova, E. Lightweight Composites Containing Hemp Hurds. Procedia Eng. 2013, 65, 69 74. Available from: http://dx.doi.org/10.1016/j.proeng.2013.09.013. 37. Cigasova, J.; Stevulova, N.; Schwarzova, I.; Sicakova, A.; Junak, J.; Sahmenko, G. Application of Hemp Hurds in the Preparation of Biocomposites. IOP Conf. Ser. Mater. Sci. Eng. 2015, 96, 12023. Available from: http://dx.doi.org/10.1088/1757-899X/96/1/012023. 38. Cigasova, J.; Stevulova, N.; Terpakova, E.; Sicakova, A.; Junak, J.; Kidalova´, L. Chemically Treated Hemp Shives as a Suitable Organic Filler for Lightweight Composites Preparing. Procedia Eng. 2012, 42, 948 954. Available from: http://dx.doi. org/10.1016/j.proeng.2012.07.488. 39. Stevulova, N.; Cigasova, J.; Estokova, A.; Terpakova, E.; Geffert, A.; Kacik, F., et al. Properties Characterization of Chemically Modified Hemp Hurds. Materials 2014, 7 (12), 8131 8150. Available from: http://dx.doi.org/10.3390/ma7128131. 40. Jarabo, R.; Fuente, E.; Blanco, A.; Savastano, H.; Negro C. Use of Agricultural Wastes as a Source of Fibers in the Manufacture of Asbestos-free Fiber-cement. 12th International Inorganic-Bonded Fiber Composites Conference (IIBCC 2010), September 2010, Aalborg, Denmark. Proceedings. . . Sorensen, E., 2010, pp 98 104. 41. Jarabo, R.; Fuente, E.; Monte, M. C.; Savastano, H. J.; Mutje, P.; Negro, C. Use of Cellulose Fibers from Hemp Core in Fiber-cement Production. Effect on Flocculation, Retention, Drainage and Product Properties. Ind. Crops Prod. 2012, 39, 89 96. Available from: http://dx.doi.org/10.1016/j.indcrop.2012.02.017.

Hatschek process as a way to valorize agricultural wastes: effects on the process and product quality 289

42. Allam, M.; Garas, G. Recycled Chopped Rice Straw-cement Bricks: An Analytical and Economical Study. In WIT Transactions on Ecology and the Environment; Popov, V., Itoh, H., Mander, U., Brebbia, C. A., Eds.; , 140; WIT Press: Ashurst, 2010. 43. Balboul Shawia, N.; Ali Jabber, M.; Fadhil Mamour, A. Mechanical and Physical Properties of Natural Fiber Cement Board for Building Partitions. Phys. Sci. Res. Int. 2014, 2 (3), 49 53. 44. Ghofrani, M.; Mokaram, K. N.; Ashori, A.; Torkaman, J. Fiber-cement Composite Using Rice Stalk Fiber and Rice Husk Ash: Mechanical and Physical Properties. J. Compos. Mater. 2014, 49 (26), 3317 3322. Available from: http://dx.doi.org/10.1177/ 0021998314561813. 45. Wu, J.; Zhang, X.; Wan, J.; Ma, F.; Tang, Y.; Zhang, X. Production of Fiberboard Using Corn Stalk Pretreated with White-rot Fungus Trametes hirsute by hot Pressing Without Adhesive. Bioresour. Technol. 2011, 102 (24), 11258 11261. Available from: http://dx. doi.org/10.1016/j.biortech.2011.09.097. 46. Yang, X. S.; Zhang, S. J.; Zuo, Z.; Men, X.; Tian, S. Ethanol Production from the Enzymatic Hydrolysis of Non-detoxified Steam-exploded Corn Stalk. Bioresour. Technol. 2011, 102 (17), 7840 7844. Available from: http://dx.doi.org/10.1016/j. biortech.2011.05.048. 47. Muoneke, C. O.; Ogwuche, M. A. O.; Kalu, B. A. Effect of Maize Planting Density on the Performance of Maize/soybean Intercropping System in a Guinea Savannah Agroecosystem. Afr. J. Agric. Res. 2007, 2 (12), 667 677. 48. Chow, P. Dry Formed Composite Board from Select Agricultural Residues. World Consultation on Wood Based Panels; Food and Agriculture Organization of the United Nations: New Delhi, India, 1974. 49. Ajayi, B. Properties of Maize-stalk-based Cement-bonded Composites. Forest Prod. J. 2006, 56 (6), 51 55. 50. Babatunde, A. Durability Characteristics of Cement-bonded Particleboards Manufactured from Maize Stalk Residue. J. For. Res. 2011, 22 (1), 111 115. Available from: http://dx.doi.org/10.1007/s11676-011-0135-2. 51. Jarabo, R.; Monte, M. C.; Fuente, E.; Santos, S. F.; Negro, C. Corn Stalk from Agricultural Residue used as Reinforcement Fiber in Fiber-cement Production. Ind. Crops Prod. 2013, 43, 832 839. Available from: http://dx.doi.org/10.1016/j.indcrop.2012.08.034. 52. Forster-Carneiro, T.; Berni, M. D.; Dorileo, I. L.; Rostagno, M. A. Biorefinery Study of Availability of Agriculture Residues and Wastes for Integrated Biorefineries in Brazil. Resour. Conserv. Recy. 2013, 77, 78 88. Available from: http://dx.doi.org/10.1016/j. resconrec.2013.05.007. 53. Arthur, J.; Ragauskas, A. J.; Beckham, G. T.; Biddy, M. J.; Chandra, R.; Chen, F., et al. Lignin Valorization: Improving Lignin Processing in the Biorefinery. Science 2014, 344, 6185. Available from: http://dx.doi.org/10.1126/science.1246843. 54. Ouyang, X.; Wang, W.; Yuan, Q.; Li, S.; Zhangb, Q.; Zhaoc, P. Improvement of Lignin Yield and Purity from Corncob in the Presence of Steam Explosion and Liquid Hot Pressured Alcohol. RSC Adv. 2015, 5 (76), 61650 61656. Available from: http://dx.doi. org/10.1039/C5RA12452b. 55. Wild, P. J. De; Huijgen, W. J. J.; Linden, R.; Der, Van; Uil, H. Den; Snelders, J., et al. Organosolv Fractionation of Lignocellulosic Biomass for an Integrated Biorefinery. NPT Procestechnologie 2015, 1, 10 11. 56. Li, J.; Gellerstedt, G.; Toven, K. Steam Explosion Lignins; their Extraction, Structure and Potential as Feedstock for Biodiesel and Chemicals. Bioresour. Technol. 2009, 100, 2556 2561. Available from: http://dx.doi.org/10.1016/j.biortech.2008.12.004.

290 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

57. Muurinen, E. Organosolv Pulping. A Review and Distillation Study Related to Peroxyacid Pulping. Academic Dissertation, University of Oulu, Oulu, Finland, 2000. 58. De La Torre, M. J.; Moral, A.; Herna´ndez, M. D.; Cabeza, E.; Tijero, A. Organosolv Lignin for Biofuel. Ind. Crops Prod. 2013, 45, 58 63. Available from: http://dx.doi.org/ 10.1016/j.indcrop.2012.12.002. 59. Wildschut, J.; Smit, A. T.; Reith, J. H.; Huijgen, W. J. Ethanol-based Organosolv Fractionation of Wheat Straw for the Production of Lignin and Enzymatically Digestible Cellulose. Bioresour. Technol. 2013, 135, 58 66. Available from: http://dx.doi.org/ 10.1016/j.biortech.2012.10.050. 60. Huijgen, W.; Telysheva, G.; Arshanitsa, A.; Gosselink, R.; De Wild, P. Characteristics of Wheat Straw Lignins from Ethanol-based Organosolv Treatment. Ind. Crops Prod. 2014, 59, 85 95 Available from: http://dx.doi.org/10.1016/j.indcrop.2014.05.003. 61. International Organization For Standardization. ISO 5267-2. Pulps Determination of Drainability Part 2: “Canadian Standard” Freeness Method. 2001. p 14. 62. Fardim, P.; Dura´n, N. Modification of Fibre Surfaces During Pulping and Refining as Analysed by SEM, XPS and ToF-SIMS. Colloids Surf., A 2003, 223 (1), 263 276. Available from: http://dx.doi.org/10.1016/S0927-7757(03)00149-3. 63. Hatchmi, M. H.; Moslemi, A. A. Correlation between Wood Cement Compatibility and Wood Extractives. Forest Prod. J., 39 1990, (6), 55 58. 64. Blanco, A.; Fuente, E.; Alonso, A.; Negro, C. Optimal Use of Flocculants on the Manufacture of Fibre Cement Materials by the Hatschek Process. Constr. Build. Mater. 2010, 24, 158 164. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2007.06.017. 65. Negro, C.; Blanco, A.; San Pı´o, I.; Tijero, J. Methodology for Flocculant Selection in Fiber Cement Manufacture. Cement Concrete Compos. 2006, 28 (1), 90 96. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2005.07.003. 66. Negro, C.; Blanco, A.; San Pı´o, I.; Tijero, J. In-line Flocculation Monitoring in a Hatschek Machine for Fiber cement Manufacture. Compos. Part A 2007, 38 (1), 26 33. Available from: http://dx.doi.org/10.1016/j.compositesa.2006.01.027. 67. Blanco, A.; Fuente, E.; Negro, C.; Tijero, J. Flocculation Monitoring: Focused Beam Reflectance Measurement as a Measurement Tool. Can. J. Chem. Eng. 2002, 80 (4), 734 740. Available from: http://dx.doi.org/10.1002/cjce.5450800403. 68. Fuente, E.; Jarabo, R.; Moral, A.; Blanco, A.; Izquierdo, L.; Negro, C. Effect of Sepiolite on Retention and Drainage of Suspensions of Fiber reinforced Cement. Constr. Build. Mater. 2010, 24 (11), 2117 2123. Available from: http://dx.doi.org/ 10.1016/j.conbuildmat.2010.04.048. 69. RILEM Technical Committee 49 TFR. Testing Methods for Fibre Reinforced Cement based Composites. Mate´riaux et Constructions 1984, 17 (102), 441 456. 70. Tonoli, G. H. D.; Santos, S. F.; Joaquim, A. P.; Savastano, J. H. Effect of Accelerated Carbonation on Cementitious Roofing Tiles Reinforced with Lignocellulosic Fibre. Constr. Build. Mater. 2010, 24 (2), 193 201. Available from: http://dx.doi.org/10.1016/ j.conbuildmat.2007.11.018. 71. American Society For Testing and Materials. ASTM C948 81, Standard Test Method for Dry and Wet Bulk Density, Water Absorption, and Apparent Porosity of Thin Sections of Glass-fiber Reinforced Concrete. West Conshohocken, USA, 2000. 72. Hubbe, M. A.; Rojas, O. J. Colloidal Stability and Aggregation of Lignocellulosic Materials in Aqueous Suspension: A Review. Bioresources 2008, 3 (4), 1419 1491. 73. Bentur, A.; Mindess, S. Fibre Reinforced Cementitious Composites. Modern Concrete Technology Series, 2nd ed; CRC Press, Taylor & Francis Group: New York, 2006.

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

11

Luz M. Murillo1, Silvio Delvasto1 and Marisol Gordillo2 1 Universidad del Valle, Cali, Colombia, 2Universidad Auto´noma de Occidente, Cali, Colombia

11.1

Introduction

The alkaline activation of aluminosilicates is a research line that has achieved remarkable results by comparing their properties with those of materials made of Portland cement.1,2 Alkaline-activated cements, such as those made from ceramic tile wastes (CTW), are characterized as having a high content of alkalis and a low content of calcium.3 For this reason, the development of the mechanical strength and durability is attributed to the reaction product or N-A-S-H gel, an alkaline aluminosicate that presents a three-dimensional structure,2,4 quite different to C-S-H gel, calcium-silicate-hydrate, obtained from the OPC hydration.5 However, this gel can incorporate a small aluminum content, forming a C-(A)-S-H gel. Nowadays, the cement industry works together with the scientific community in order to minimize the negative environmental impact by using materials such as pozzolan (fly ash, silica fume, construction wastes, ceramic tiles among others).6,7 However, they are focusing on geopolymers due to their high strength, durability, and reduction of environmental impact.2,8,9 The term “geopolymer” was given to inorganic synthetic polymers of aluminosilicates derived from a chemical reaction known as geopolymerization where silica and aluminum are bound tetrahedronically by exchanging oxygen atoms, forming the basic unit, a sialate monomer (O-Si-O-Al-O). It carries a negative charge excess due to the replacement of Si41 for Al31. The charge balance in the polysiate structure is achieved by alkaline metallic cations (K1 or Na1).10 Theoretically, any aluminosilicate source can be used for preparing geopolymers, including aluminosicate minerals.11 Several authors have used different sort of wastes as an alternative material source, for instance: CTW which makes part of this research,3,6,12 fly ash,4,13 blast furnace slag,14,15 among others. The alkali activation of Portland cement blends containing high proportions of aluminosilicate materials renders hybrid binders. Reig et al.12 studied the influence

Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00013-9 © 2017 Elsevier Ltd. All rights reserved.

292 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

of the alkaline activator concentration, and the addition of calcium hydroxide on the compressive strength behavior, and the microstructure of stoneware waste samples, using NaOH and a sodium silicate reaction as activators. The results show that the activator concentration, and the calcium sodium molar ratio strongly influences the behavior in the fresh state, and in the hardening process of mortars made from ceramic wastes. A lineal evolution of the compressive strength by adding sodium was observed when SiO2 concentration was kept constant, achieving compressive strengths up to 36 MPa after 7 curing days at 65 C with a relative humidity of 95%. Fadaly et al.16 carried out the alkaline activation of ceramic tiles using highly-pure sodium hydroxide, and sodium silicate. The ceramic wastes were activated with a series of alkaline solutions with about the same quantity of silica, but with a different quantity of sodium oxide and water, in 10% 40% proportions. The obtained samples were cured at room temperature for 90 days, obtaining compressive strengths up to 55 MPa for the mixture with a SiO2/Al2O3 of 3.42, Na2O/SiO2 of 0.29, and H2O/Na2O of 1.47 ratios. Sun et al.17 used ceramic wastes (from tiles and blocks mainly) that were activated using alkaline hydroxides and/or sodium/potassium silicate solutions to synthesize the studied geopolymeric material. The synthesized geopolymer pastes were characterized by using mechanical tests, TG-DSC, TGA, DTA, scanning electron microscopy (SEM), X-Ray Diffraction (XRD), as well as the FT-IR analysis. The thermal behavior of the synthesized geopolymers was determined in terms of compressive strength evolution due to the exposure to 100, 200, 400, 600, 800 and 1000  C. The synthesized geopolymer pastes showed a maximum compressive strength of 71.1 MPa after 28 days; and the antithermal properties were favorable, showing a higher compressive strength (75.6 MPa) after a thermal treatment at 1000  C. According to the obtained results, the ceramic waste could be an appropriate material to form thermoset geopolymers. Puertas et al.6 studied the alkaline activation of six fired ceramic wastes produced in different plants of ceramic tile factories. For this purpose, three samples were selected from red-clay fired tiles, two other samples came from white-clay fired tiles, and finally, a fired waste sample was selected from a white-and-red waste mixture because some manufacturers do not separate wastes. The alkaline activation was achieved by using NaOH, obtaining a low alkaline activation after 8 curing days. The mechanical compressive strengths values did not overpass 14 MPa; for this reason, it can be stated that the amorphous state of wastes is not a determinant factor in their alkaline activation. The activator nature (NaOH dissolution) does not seem to be a factor that influences the process; the highest strengths are reached when the activator concentration is 6 M. The characterization study carried out on activated pastes has demonstrated that feldspatic phases are more susceptible to dissolve and react with alkaline dissolutions. In this research, binder pastes of hybrid systems were developed based on CTW together with Portland cement, which acts as the calcium provider. Also studied was the influence of the percentage of cementitious material, SiO2/Al2O3 and Na2O/SiO2 molar ratio to optimize its compressive strength, using the Response Surface Methodology (RSM).

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

11.2

293

Materials

Construction and demolition wastes from walls and floors have been applied to replace part of Portland cement in order to reduce the environmental impact generated not only from the cement production process, but also because of its wrong or poor disposal. These ceramic tiles debris were sampled from a demolished part of a building.

11.2.1 Characterization of wall and floor construction and demolition ceramic wastes Wall-and-floor construction and demolition ceramic wastes, and Portland cement were characterized by means of chemical, physical, mineralogical, microstructural tests, and using the thermogravimetric analysis (TGA).

11.2.1.1 Chemical characterization The chemical characterization of the used raw material is shown in Table 11.1. The loss on ignition (LOI) was determined by burning the samples at 1000 C. The CTW have a content of 64.51% silica content (SiO2), and 17.37% alumina (Al2O3), becoming a suitable material for the geopolymerization process. This material also has a content of 4.81% iron oxide (Fe2O3), hence its red color. Because this material is a waste mixture made of red and white clays, the rest of its chemical compound quantities are less than 2.1%. According to the obtained results, this material presents SiO2/Al2O3 molar ratio of 6.3, which is the starting point to carry out the activation process. The ordinary Portland cement (OPC) has a calcium oxide content higher than 50.0%, this means it is a great calcium source during the sample processing. This material is added in order to avoid the thermal curing. The OPC also has 20.73% of silica, and 5.63% of iron oxide. Other chemical compounds represent less than 5%.

Chemical composition of precursor CTW and OPC (% by weight) Table 11.1

Oxide

Ceramic tile wastes (CTW)%Wt

Ordinary Portland cement (OPC) (%wt)

CaO SiO2 Fe2O3 Na2O Al2O3 K2 O SO3 MgO LOI

2.08 64.51 4.81 17.37 0.85 2.09 0.05 0.94 3.96

52.69 20.73 5.63 4.54 0.15 0.41 3.14 2.24 9.81

294 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

11.2.1.2 Physical characterization Determining particle-size distribution The particle-size distribution is one of the factors that affects activated-material behavior; for this reason, it was determined in the ceramics tile wastes by means of the laser granulometric analysis. The average particle size for the CTW was 29.42 μm. This size was obtained after crushing the debris by hand; then, the sample was placed into a crusher, obtaining approximately 4-mm size particles. After that, a comminution was made in a laboratory ball mill for 3 h. This particle size is similar to the ones used in research carried out for the alkaline activation process of CTW.12,17,18 Fig. 11.1 shows the particle size distribution curves for CTW and OPC. The granulometric distribution for CTW is wide, indicating variability in the particles size: 10% of them present a size below 1.69 μm, 50% have a size below 18.83 μm, a medium-size particle of 23 μm is observed, with 90% below 72.87 μm. In the case of the Portland cement, a medium-size particle of 26 μm was obtained: 10% of the particles have a size below 2.40 μm, 50% a size below 16.98 μm, and 90% below 49.58 μm.

Characterization performed by using scanning electron microscopy (SEM) As it can be observed in Fig. 11.2, the CTW particles obtained after the grinding process present an irregular shape due to the fact that this material comes from wall-and-floor ceramic tiles that are formed by a very dense and strong structure.19 This is corroborated with the comminution process. This material has a wide size dispersion, confirming the result obtained in the granulometric analysis curve. 6

CTW OPC

5

Volume (%)

4

3 2 1 (B) 0 0.1

(A) 1

10

100

1000

Particle size (μm)

Figure 11.1 Granulometric distribution of the raw material: (A) Ceramic Tile Wastes (CTW); (B) Portland cement (OPC).

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

295

Figure 11.2 Low (A) and high (B) magnification SEM micrograph of Ceramic Tile Waste particles. Q

Q

CTW

An

Ab

Ab

M H

An H

M

M Q Q Q M

Q Q

H

A-C

OPC

A A-L A

Q

G

A-L

G BC

B B

10

B

A

A Q A

L A

20

A

A AQ

G

30

C L

A L

40

A B

50

L

60

2θ (degrees)

Figure 11.3 CTW and OPC diffractograms. mullite (M), quartz (Q), Hematite (H), Albite (Ab), anorthoclase (An), alite (A), larnite (L), gypsum (G), calcite (C) and brownmillerite (B).

11.2.1.3 Mineralogical and structural characterization X-Ray Diffraction (XRD) study CTW and Portland cement X-ray diffractograms are shown in Fig. 11.3. In the case of CTW, a material with a mainly crystalline structure is observed; quartz and

296 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

CTW 3 45

2

1

6 12 11 13 14 15 9 16 17 10 7

OPC

8 c d a

f g

e

p n q

h

b

o j

r

s

k l m

i

4000

3500

3000

2500

2000

Wavenumber

1500

1000

500

(cm–1)

Figure 11.4 Spectra of ceramic tile wastes (CTW) and Portland cement (OPC).

mullite are its most abundant phases. In addition, other phases such as albite, anorthoclase, and hematite are present. Phases such as Hatrurite, quartz, Larnite, gypsum, calcite, and brownmillerite (ferrous phase) can be observed in the cement diffractogram.

Study by using fourier transform infrared spectroscopy (FTIR) The infrared spectrum of the ceramic tiles wastes and the Portland cement can be observed in Fig. 11.4. Table 11.2 shows the assignment of vibration bands in this figure. The presence of quartz, in CTW, causes a series of bands located at 1084, 796 778 (double band), 693, 522, and 463 cm21 in the IR spectrum.20 The presence of aluminum octahedron in mullite is responsible for a series of bands located between 1180 and 1130 cm21, in the range from 560 to 550 cm2121,22; and at 630 cm21, it is possible to find a characteristic band of Al-O-Si.23 The band corresponds to the internal vibrations of primary structural units; in other words, SiO4 and AlO4 tetrahedrons, and Si-O-Si and Si-O-Al bridges are found in the wave number 470 cm21.24 The band, also intensive, associated with the vibrations of deformations of T-O bonds (internal vibrations of the tetrahedrons) is located at 466 cm21.25 In addition, the bands associated with the vibrations of O-H asymmetrical tensions can also be observed in the spectrum in a range from 3200 to 3700 cm21, as well as vibrations of H-O-H deformation of water in the region of 2300 2500 cm21, between 1600 and 1650 cm21, and around 1800 cm21 and 2000 cm21, which is in agreement with the research carried out by Mozgawa et al.26 Xu et al.27, and Komnitsas et al.18 In this material, it is possible to observe

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

Table 11.2

297

FTIR results for ceramic tile wastes and Portland

cement Ceramic tile wastes (CTW)

Ordinary Portland cement (OPC)

Bands

Wave number (cm21)

Characteristic bands

Bands

Wave number (cm21)

Characteristic bands

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

3596 2400 2004 1891 1801 1620 1164 1083 796 778 693 631 563 515 470 467 463

δ O-H-O (H2O) δ O-H-O (H2O) δ O-H-O (H2O) δ O-H-O (H2O) δ O-H-O (H2O) δ O-H-O (H2O) v T-O (mullite) δ Si-O (quartz) δ Si-O (quartz) δ Si-O (quartz) δ Si-O (quartz) v Si-O-Si y Al-O-Si v T-O (mullite) δ Si-O (quartz) Si-O-Si (TO4 Td) δ O-Si-O (TO4 Td) δ O-Si-O (TO4 Td)

a b c d e f g h i j k l m n o p q r s

3547 3406 2984 2876 2514 1799 1684 1621 1428 1114 1092 924 875 715 694 669 602 523 467

v as H2O gypsum v as H2O CaCO3 CaCO3 CaCO3 CaCO3 δ H2O sulfates δ H2O sulfates CO3 v as de SO4 CaSO4 v as Si-O (C3S) δ de CO3 δ de CO3 δ Si-O (quartz) δ (O-S-O) (SO4) δ (O-S-O) (SO4) δ O-Si-O (C3S) δ O-Si-O (TO4 Td)

the low intensity of bands associated to water; this is because CTW are materials that have undergone a sintering process, and are mainly porcelain stoneware, construction elements that have a low water absorption (below 5%). In the Portland cement spectra, bands between 3100 3700 cm21 can be observed, and are associated with the presence of H2O and OH because the cement can have some residual humidity. The water absorbed by gypsum is associated with the band located at 3550 cm21.28 Two bands, one at 1623 cm21, and a smaller one at 1684 cm21, come up in the spectrum of the anhydride cement due to the presence of water absorbed by sulfates. The bands located at 1796, 2513, 2875, 2983, and 1350 1550 cm21 are caused by the presence of calcium carbonate in the cement,28 in the same way that the presence of carbonates is associated with bands located at 1428 cm21, 714 cm21, and 878 cm21. Furthermore, two main bands, intensive and sharp (926 and 878 cm21) are associated with stretching/tension vibrations Si-O cm21 of the Alite, and the 523 cm21 band is associated with deformation vibrations O-Si-O of the main silicates in cement.29 The characteristic bands of sulfates are generally found in the range of 1100 1200 cm21. It is difficult to interpret this area of the spectrum because sulfates generate many peaks, thus generating

298 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

overlaps. The band located at 1092 cm21 corresponds to sulfate of anhydrous calcium (CaSO4), while the bands located at 602 and 660 cm21 are associated with semihydrated calcium sulfate (CaSO4  1/2H2O).30

11.2.1.4 Thermogravimetric analysis Figs. 11.5 and 11.6 show the results of the TGA, and the differential thermal analysis (DTA) carried out on the ceramic tiles wastes and on the Portland cement, respectively. For the CTW, only a change at 573 C was observed; this band was generated during the quartz inversion from α to β. This unique transformation is due to the fact that this material has already been exposed to a firing process (B1200 C). Some transformations are generated in this process, such as the kaolin dehydroxylation, quartz transformation from α to β, and the metakaolin decomposition in a spinel-type structure, amorphous silica, and mullite phase, where finally a low water absorption (below 0.5%) material is obtained.31 The TGA result shows a 0.45% total water loss attributed to water removal by humidity, this is consistent with the characteristics of a sintered material including porcelain stoneware which is the main component in CTW. The results of TGA and DTA analyses carried out on OPC first showed a mass loss at 100 C; this is related to the removal of absorbed water.32 The second mass loss is likely due to the dehydration of hydrated calcium sulfate present in the OPC (CaSO4  2H2O); this takes place between 0 C and 95 C.33 The third mass loss at 700 C is related to the dissociation of carbonates due to the removal of CO2.34

100.1

0.1 °C/mg

0.0

99.9 Weight (%)

%

–0.1

99.8

–0.2 99.7

–0.3

99.6

–0.4

99.5

0

200

400

600

800

1000

Temperature (°C)

Figure 11.5 Curves of TGA and DTA for the ceramic tile wastes.

1200

Temperature difference (°C/mg)

100.0

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

299

11.2.2 Characterization of fine aggregate To elaborate mortars and structural elements, a fine aggregate, natural river sand, was used; it has a fineness modulus of 2.76 (medium sand), a specific density equal to 2.56 g/cm3, and a 1.8 absorption. Fig. 11.7 shows its granulometric distribution. 100

0.05

0.00

Weight (%)

–0.05

%

90 –0.10 85 –0.15

80 0

200

400

600

800

1000

Temperature (°C)

Figure 11.6 Curves of TGA and DTA for the OPC. 100

Cumulative weight (%)

80

60

40

20

0 0.1

1 Grain size (mm)

Figure 11.7 Distribution of the fine aggregate particle size.

–0.2 1200

Temperature difference (°C/mg)

°C/mg

95

300 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

11.2.3 Paste sample preparation In order to produce hybrid geopolymers, Portland cement was added to the CTW. Hybrid systems formed by ceramic tiles wastes with a CTW/(CTW 1 OPC) ratios, between 0.85 and 0.95%, were elaborated using a sodium hydroxide alkaline solution (99% pure) dissolved in water, and a sodium silicate solution (SiO2 5 32.24%, Na2O 5 11.18% and H2O 5 55.85%). These materials were mixed using a 0.26 liquid/solid ratio for 4 min in a mixer. Afterwards, the mixture was placed in 2 cm cubic molds; then, these molds were vibrated for 1 min. Next, the paste samples were wrapped with a polyethylene film for 24 h at room temperature (B25 C). After that, the samples were removed from the molds, and wrapped again with polyethylene film, keeping them at room temperature until testing. In order to obtain an optimal sample, a compressive strength test was carried out using the RSM. This statistical technique is used to design an experiment to provide reasonable values of the variable response; in this case, it is the compressive strength at 28 days (CS 28D), and in this way, determines the mathematical model that better adjusts to the obtained data. Table 11.3 shows the factors and levels to design experiments; levels are established based on tests carried out previously. The treatments to be carried out according to the factors and levels are shown in Table 11.4. The optimal sample was analyzed using a Scanning Electron Microscopy SEM at the ages of 7, 28, and 90 curing days.

11.2.4 Mortars preparation The geopolymeric mortars were prepared with an optimal mixture of a hybrid CTW geopolymer paste. First, the ground CTW were mixed with the alkaline solution for 2 min; afterwards, the sand was added and mixed for 3 min, with a CTW/sand ratio equal to 1, and a 0.28 liquid/solid ratio. The mixture was placed in 5 cm thick molds in two layers, tamping down each layer; and then, the samples were wrapped with a plastic film, following the same procedure used with the alkali-activated pastes. Table 11.3

Factors and levels to design experiment modulus

Factors

Levels

SiO2/Al2O3 (S/A) Molar ratio

(2) low level [values 5 7.0] (1) high level [values 5 8.0] (2) lower level [values 5 0.08] (1) high level [values 5 0.18] (2) low level [values 5 0.85]

Na2O/SiO2 (N/S) Molar ratio Ceramic tile waste /(Ceramic tile waste 1 Portland cement) CTW/(CTW 1 OPC)

(1) high level [values 5 0.95]

Mean values S/A N/S CTW/(CTW 1 OPC)

7.50 0.13 0.9

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

11.3

301

Results and discussion

11.3.1 Response surface analysis From the hypothesis test, which examines if the errors follow a normal distribution, it was possible to prove that the quadratic model fits, in an acceptable way, with it at 76.7%. This can be observed in Fig. 11.8. The line forms an estimation of an Table 11.4

Mixtures for the alkaline activation process

Standard order

Run order

Point type

S/A

N/S

CTW/(CTW 1 OPC)

1 5 9 15 2 8 4 3 13 6 7 14 11 10 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

2 2 2 0 2 2 2 2 0 2 2 0 2 2 2

7 7 7.5 7.5 8 8 8 7 7.5 8 7 7.5 7.5 7.5 7.5

0.08 0.13 0.08 0.13 0.08 0.13 0.18 0.18 0.13 0.13 0.13 0.13 0.08 0.18 0.18

0.90 0.95 0.95 0.90 0.90 0.85 0.90 0.90 0.90 0.85 0.85 0.90 0.85 0.95 0.85

Figure 11.8 Normality of the residuals of the quadratic model applied to the compressive strength data at 28 curing days of hybrid mixtures based on alkaline-activated ceramic tile wastes.

302 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

accumulative distribution in the function of the population data; therefore, the data follows a normal distribution. The Levene test was used to determine the variance homogeneity, obtaining 25.6%, that complied with the variance homogeneity. For this reason, the application of this model is suitable for predicting the compressive strength of systems based on wall-and-floor CTW with calcium content that were alkaline-activated. The analysis of the compressive strength results using the RSM was carried out in specimens cured for 28 days. Based on the analysis of these results, using the least square method for the quadratic model, a model for the compressive strength variable response (MPa) was obtained with an adjustment of 95.25% (R2) of the model to the data. The expression that represents the quadratic model is shown in Eq. 3.3.1, where the significant coefficients are presented, not only because of linearity, but also those related to the interaction and quadratic effects (Table 11.5). 

 CTW N S N N CS28 5 451:3 2 134:7 2 1132 2 43:4 2 4074  CTW 1 OPC S A S S N S 1 253:1  S A (11.1) In order to assess the statistical significance of the regression model (Table 11.5), the variance analysis (ANOVA) was undertaken. It provides the variability of the compressive strength (Y), independently for each effect. Therefore,

Analysis of the variance for the proposed quadratic model for predicting the compressive strength of hybrid mixtures based on alkaline-activated ceramic tiles

Table 11.5

Source

Degrees of freedom

Sum of squares

Mean squares

F-value

p-value

Model Linear N/S S/A CTW/(CTW 1 OPC) Square N/S N/S 2-Way Interaction N/S S/A Error Lack of fit Pure error Total

5 3 1 1 1 1 1 1 1 8 6 2 13

2837.02 2303.63 300.96 1723.14 182.57 344.13 344.13 160.15 160.15 141.34 130.31 11.03 2978.36

567.40 767.88 300.48 1723.14 182.57 344.13 1344.13 160.15 160.15 17.67 21.72 5.51

32.12 43.46 17.03 97.53 10.33 19.48 19.48 9.06 9.06

0.000 0.000 0.003 0.000 0.012 0.002 0.002 0.017 0.017

3.94

0.216

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

303

the statistical significance was proved for each effect by comparing the sum of square of the regression (SCR 5 2837.02), with respect to the sum of square errors (SCE 5 141.34), and it was observed that the lineal effect contributes the most (SCL 5 2303.63) to explain the model variation. In relation to the hypothesis about the lack of adjustment for the proposed model, it was found that the model can predict the compressive strength (CS 28) to a significance level of 0.216. Thus, the optimization of the compressive strength was carried out.

11.4

Compressive strength optimization

An optimization was performed for the variable compressive strength response to 28 days of curing based on the proposed model. First, the conditions for which the optimization would be performed were established, that is, for a minimum value of 2.15 MPa, and for an objective value of 54.78 MPa of compressive strength, where the bigger the optimization is, the better it is (maximization). Fig. 11.9 shows the first proposed optimizations for the predictability function. The optimization plots show the effect of each factor (columns) on the responses (CS) and desirability or convenience (D) (rows). The vertical red lines on the graph represent the current factor settings. The numbers displayed at the top of a column show the current factor level settings (in red). The horizontal blue lines and numbers represent the compressive strength for the current factor level. It can be observed that for a N/S molar ratio of 0.08, a S/A molar ratio of 7, and a CTW/ (CTW 1 OPC) ratio equal to 0.85, a compressive strength of 58.14 MPa was obtained with a predictability or convenience (D) equal to 1. Fig. 11.10 shows the optimization suggested by the researcher, which is based on the predictability function, where for a N/S molar ratio of 0.08, a S/A molar relation of 7, and a CTW/(CTW 1 OPC) ratio of 0.89 or 0.11 of OPC, a compressive strength of 52.73 MPa was obtained with a predictability or convenience (D) equal to 96.10%. For this research, the CTW/(CTW 1 OPC) ratio should be 0.89 at maximum because a ratio higher than this value (Fig. 11.10) alters the predictability and the

Figure 11.9 Optimization prediction of the compressive strength of mixtures with a CTW/ (CTW 1 OPC) ratio based on alkaline-activated ceramic tiles.

304 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

compressive strength of the pastes in a negative way. This is why the cementitious material is important for both eliminating the thermal curing, and developing better mechanical compressive strengths during the alkaline activation process. The contour plots (Fig. 11.11), used to predict the compressive strength for a certain ratio, show that the compressive strength is higher than 50 MPa when the molar solution N/S is between 0.08 and 0.12, the S/A molar ratio is between 7.0 and 7.36, and the CTW/(CTW 1 OPC) ratio is between 0.85 and 0.91, validating what was stated in the previous paragraph.

Figure 11.10 Researcher’s prediction of the compressive strength of hybrid mixtures with a CTW/(CTW 1 OPC) ratio based on alkaline-activated ceramic tiles.

Figure 11.11 Contour plots for (A) N/S molar ratio versus CTW/(CTW 1 OPC) ratio (B) S/A molar ratio versus a CTW/(CTW 1 OPC) ratio, and (C) S/A versus N/S molar ratios.

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

305

According to the obtained results, it can be stated that for high S/A and N/S ratios, the compressive strength decreases due to the fact that with the increases in the S/A ratio, the reaction speed is slower. This fact makes the gel formation more difficult, which is essential for developing the material mechanical strength.1,35 In the formulations with high Na2O/SiO2 (N/S) ratios, it is expected that an alkalinity increase causes a higher degree of dissolution of the employed precursors, and that the presence of cations maintains the electric neutrality of the geopolymeric system. However, a high content of alkalis causes the polymerized nets to be less stable because the quantity of required cations for the charge balance is independent of the alumina content in the system. For this reason, a cation excess could cause its leaching, and its reaction with the atmospheric CO2, forming carbonates.3,36,37 The increase in the soluble silicates content improves the mechanical performance of the material according to the research results carried out by Criado et al.38; it is for that reason that there is not enough silica at low S/A ratios for the activation process. In the case of high S/A ratios caused by the increase of sodium silicate, an increase in the mixture viscosity is generated, accelerating the geopolymerization process, and reducing the setting time.39 These results are coherent with those obtained in this work, where with high S/A ratios, reductions of workability were observed due to viscosity increases, as well as a decrease in the mechanical strength. That is why an optimal quantity of this material should be added during its processing. The presence of calcium in the mixtures of alkali-activated CTW allows curing of the samples at room temperature and the development of mechanical strength. This is possibly due to the presence of cementitious gels, including C-A-S-H gel (that has sodium in its composition), and (N,C)-A-S-H gels (N-A-S-H gels with a high calcium content). These gels are responsible of the mechanical resistance development of the samples.29 The compressive strength of the samples was higher when the OPC content increased in the system (CTW/(CTW 1 OPC)) suggesting further reaction of the OPC.

11.4.1 Geopolymeric mortars Based on the results obtained in the optimization process of SiO2/Al2O3, Na2O/SiO2 molar ratios, and the CTW/(CTW 1 OPC) ratio, hybrid mortars were made using a proportioning of CTW/sand equal to 1, a SiO2/Al2O3 molar ratio equal to 7, a Na2O/ SiO2 molar ratio equal to 0.08, and a CTW/(CTW 1 OPC) ratio equal to 0.85. Fig. 11.12 shows the results of mechanical compressive strengths for the ages of 7, 28, and 90 days of curing; as time goes on, the compressive strength increased until it reached 30.06 MPa. This result is coherent with those obtained in hybrid paste mixtures, and the RSM design.

11.5

Scanning electron microscopy analysis

Fig. 11.13 shows the microstructure of alkaline-activated pastes containing calcium, S/A 5 7, N/S 5 0.08, and CTW/(CTW 1 OPC) 5 0.85 at the ages of 7, 28, and 90 days of curing (Figs. 11.13A, B, and C respectively). It can be observed that alkaline-activated CTW generates a denser structure as the curing ages increase

Compressive strength (MPa)

306 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0

7

28

90

Days of curing

Figure 11.12 Compressive strength of geopolymeric mortars.

Figure 11.13 Scanning electron microscopy of (A) alkali-activated CTW with OPC paste at the age of 7 days, the yellow circles highlight the unreacted particles; (B) alkali-activated CTW paste at the age of 28 days; (C) alkali-activated CTW paste at the age of 90 days.

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

307

EDS result for alkali-activated CTW with OPC (hybrid) at 90 days of curing

Table 11.6

Spectrum 1 2 3 4

C

O

Na

Al

Si

K

Ca

Fe

Total

7.46 14.06

44.49 47.91 46.25 47.82

7.43 6.98 7.99 6.33

4.53 7.16 3.36 6.20

25.74 23.59 22.55 19.71

2.01 1.85 1.24 0.59

12.99 8.45 9.47 5.28

2.80 4.07 1.69

100.00 100.00 100.00 100.00

Figure 11.14 Scanning electron microscopy of alkali-activated CTW with OPC at the age of 90 days. Four different EDS spectra were obtained from the spots marked in the image as 1, 2, 3, and 4.

(Figs. 11.13B and C). This is also coherent with the increase of the compressive strength obtained for geopolymeric mortars. Fig. 11.13A also shows some unreacted particles, indicating a relatively low stage of geopolymerization (particles inside the yellow circle), and some fissures that could have been generated by the shrinkage process or during the sample preparation because stresses are generated. There are no zeolitic compounds found in the micrographs. According to the results obtained by Energy Dispersive Spectroscopy (EDS) (Table 11.6) for the hybrid material sample at the age of 90 days of curing (Fig. 11.14), the selected areas report silica, oxygen, aluminum, calcium, and sodium as the main elements (spectrum 1, 2, 3, and 4). According to what was stated by Garcia Lodeiro et al.29 the reaction product in the activation process of

308 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

CTW with OPC can be a calcium sodium aluminosilicate gel, (N,C)-A-S-H gel, that might be a N-A-S-H gel that had taken up a certain amount of calcium. The spectrum3,4 identified the presence of carbon likely to be sodium carbonate— Na2CO3 monoclinic (Natrite)—due to atmospheric carbonation.40 In the reaction process of the alkali-activated hybrid systems, as well as the C-SH gel generated by the cement hydration, it precipitates a (N,C)-A-S-H gel (calcium sodium aluminosilicate) that has a three-dimensional structure. During the first stages of the reaction, the gel seems to be a N-A-S-H type that has incorporated small quantities of calcium in its composition; this element increases its content over time, causing a decrease in sodium.29

11.6

Conclusions

The SiO2/Al2O3 and Na2O/SiO2 molar ratios have a great effect on the development of the wall-and-floor CTW activation process, due to the charge balance of the Si and Al tetrahedron, influencing the obtained compressive strength properties. The RSM is a statistical tool used to predict the compressive strength of hybrid systems of alkali-activated CTW, thus obtaining the optimal molar ratios. Based on this statistical tool to obtain the maximum compressive strength in hybrid systems of alkaliactivated CTW, mixtures with a SiO2/Al2O3 molar ratio equal to 7.0, a Na2O/SiO2 ratio equal to 0.08, and CTW/(CTW 1 OPC) ratio of 0.85 are the optimal mixtures. The use of OPC as a calcium source in the geopolymeric systems significantly improves the compressive strength; this happens due to the reaction products generated during the process, including the formation of (N,C)-A-S-H gel which enhances the sample densification due to the microstructure modification. Moreover, using a calcium source avoids the use of thermo-curing of the samples to accelerate the activation process.

Acknowledgments The research presented here has been possible thanks to the project: “Investigacio´n de un material cementicio ecoeficiente para elementos de construccio´n de bajo costo,” contrato FP44842 3992015, financially supported by COLCIENCIAS “EL PATRIMONIO AUTONOMO FONDO NACIONAL DE FINANCIAMIENTO PARA LA CIENCIA, LA ´ N, FRANCISCO JOSE´ DE CALDAS” and to the TECNOLOGIA Y LA INNOVACIO Administrative Department of Science and Technologies (COLCIENCIAS) for the support provided for the development of this research.

References 1. Duxson, P.; Provis, J. L.; Lukey, G. C.; Van Deverter, J. S. J. The Role of Inorganic Polymer Technology in the Development of “Green Concrete”. Cement Concrete Res. 2007, 37, 1590 1597. Available from: http://dx.doi.org/10.1016/j.cemconres.2007.08.018.

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

309

2. Ferna´ndez-Jime´nez, A. Y.; Palomo, A. Composition and Microstructure of Alkali Activated Fly Ash Binder: Effect of the Activator. Cement Concrete Res. 2005, 35, 1984 1992. Available from: http://dx.doi.org/10.1016/j.cemconres.2005.03.003. 3. Reig, L.; Tashima, M. M.; Borrachero, M. V.; Monzo´, J.; Cheeseman, C. R.; Paya´, J. Properties and Microstructure of Alkali-activated Red Clay Brick Waste. Constr. Build. Mater. 2013, 43, 98 106. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2013.01.031. 4. Palomo, A.; Grutzeck, M. W.; Blanco, M. T. Alkali-Activated Fly Ashes: A Cement for Future. Cement Concrete Res. 1999, 29 (8), 1323 1329. Available from: http://dx.doi. org/10.1016/S0008-8846(98)00243-9. 5. Richardson, I. G. The Calcium Silicate Hydrates. Cement Concrete Res. 2008, 38, 137 158. Available from: http://dx.doi.org/10.1016/j.cemconres.2007.11.005. 6. Puertas, F.; Barba, A.; Gazulla, M. F.; Go´mez, M. P.; Palacios, M.; Martı´nez, S. Ceramic Wastes for Possible Use as a Raw Material in the Manufacture of Portland Cement Clinker: Characterization and Alkaline Activation. Mater. Constr. 2006, 56, 73 84 (In Spanish). Available from: http://dx.doi.org/10.3989/mc.2006.v56.i281.94. 7. Senhadji, Y.; Escadeillas, G.; Mouli, M.; Khelafi, H.; Benosman. Influence of Natural Pozzolan, Silica Fume and Limestone Fine on Strength, Acid Resistance and Microstructure of Mortar. Powder Technol. 2014, 254, 314 323. Available from: http:// dx.doi.org/10.1016/j.powtec.2014.01.046. 8. Smith, B. G. Durability of Silica Fume Concrete Exposed to Chloride in Hot Climates. J. Mater. Civ. Eng. 2001, 13 (1), 41 48. Available from: http://dx.doi.org/10.1061/ (ASCE)0899-1561. 9. Bakharev, T. Durability of Geopolymer Materials in Sodium and Magnesium Sulfate Solutions. Cement Concrete Res. 2005, 35 (6), 1233 1246. Available from: http://dx. doi.org/10.1016/j.cemconres.2004.09.002. 10. Davidovits, J. Soft Mineralurgy and Geopolymers. Proceedings Of Geopolymer 88, 1st European Conference On Soft Mineralurgy, 1. The Geopolymer Institute: Compiegne, France, June, 1988, 19 23. 11. Xu, H.; Van, J. S. J. Factors Affecting the Geopolymerization of Alkali-Feldspars. Miner. Metall. Process. 2002, 4, 209 214. 12. Reig, L.; Soriano, L.; Borrachero, M. V.; Monzo, J.; Paya, J. Influence of the Activator Concentration and Calcium Hydroxide Addition on the Properties of Alkali-activated Porcelain Stoneware. Constr. Build. Mater. 2014, 63, 214 222. Available from: http:// dx.doi.org/10.1016/j.conbuildmat.2014.04.023. 13. Fernandez, L. E.; Jimenez, J. R.; Ayuso, J.; Fernandez, J. M.; Brito, J. Maximum Feasible Use of Recycled Sand from Construction and Demolition Waste for Eco-mortar Production E Part-I: Ceramic Masonry Waste. J. Clean. Prod. 2014, 87 (1), 692 706. Available from: http://dx.doi.org/10.1016/j.jclepro.2014.10.084. 14. Cheng, T. W.; Chiu, J. P. Fire-resistant Geopolymer Produced by Granulated Blast Furnace Slag. Miner. Eng. 2003, 16, 205 210. Available from: http://dx.doi.org/ 10.1016/S0892-6875(03)00008-6. 15. Krivenko P.V., Skurchinskaya J.V. Fly Ash Containing Geocements. Proceedings of the International Conference on the Utilization of Fly Ash and Other Coal Combustion byProducts 91, September, 1991, Shanghai Research Institute of Building Science. Shanghai, China, pp 64-1 64-7. 16. Fadaly, E.; Mostafa, M.; Saraya, M. E. I.; Nassar, F. A.; Sokkary, T.; Didamony, H. Ecofriendly Cement from Ceramic Waste Geopolymarization. Impact J. 2014, 2 (5), 195 210.

310 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

17. Sun, Z.; Cui, H.; An, H.; Tao, D.; Xu, Y.; Zhai, J., et al. Synthesis and Thermal Behavior of Geopolymer-type Material from Waste Ceramic. Constr. Build. Mater. 2013, 49, 281 287. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.08.063. 18. Komnitsas, K.; Zaharaki, D.; Vlachou, A.; Bartzas, G.; Galetakis, M. Effect of Synthesis Parameters on the Quality of Construction and Demolition Wastes (CDW) Geopolymers. Adv. Powder Technol. 2015, 26 (2), 368 376. Available from: http://dx. doi.org/10.1016/j.apt.2014.11.012. 19. Restrepo, B. O. J. Ceramic Tile and Stoneware: A World in Constant Evolution; Centro Editorial Facultad de Minas: Colombia, 2011, In Spanish. 20. Gadsden, J. A. Infrared Spectra of Minerals and Related Inorganic Compounds; Butterworths: London UK, 1975. 21. Jin, X. H.; Gao, L.; Guo, J. K. J. The Structural Change of Diphasic Mullite Gel Studied by XDR and IR Spectrum Analysis. J. Eur. Ceram. Soc. 2002, 22 (8), 1307 1311. Available from: http://dx.doi.org/10.1016/S0955-2219(01)00447-2. 22. Temuujin, J.; Okada, K.; Mackenzie, K. J. D. Effect of Mechanochemical Treatment on the Crystallization Behaviour of Diphasic Mullite Gel. Ceram. Int. 1999, 25 (1), 85 90. Available from: http://dx.doi.org/10.1016/S0272-8842(98)00005-4. 23. Lee, W. K.; Van Deventer, J. S. J. Use of Infrared Spectroscopy to Study Geopolymerization of Heterogeneous Amorphous Aluminosilicates. Langmuir 2003, 19 (12), 8726 8734. Available from: http://dx.doi.org/10.1021/la026127e. 24. Flaningen, E. M.; Khatami, H.; Szymanski, H. A. Molecular Sieve Zeolites. Adv. Chem. 1971, 101, 201 229. Available from: http://dx.doi.org/10.1021/ba-1971-0101.ch016. 25. Ferna´ndez-Jime´nez, A.; Puertas, F. Effect of Activator Mix on the Hydration and Strength Behaviour of Alkali-activated Slag Cements. Adv. Chem. 2003, 15 (3), 129 136. Available from: http://dx.doi.org/10.1680/adcr.15.3.129.36623. 26. Mozgawa, W.; Bajda, T. Application of Vibrational Spectra in the Studies of Cation Sorption on Zeolites. J. Mol. Struct. 2006, 792, 170 175. Available from: http://dx.doi. org/10.1016/j.molstruc.2005.12.057. 27. Xu, H.; Van Deventer, J. S. J. The Geopolymerisation of Multiple Minerals. Int. J. Miner. Process. 2000, 59 (3), 247 266. Available from: http://dx.doi.org/10.1016/ S0892-6875(02)00255-8. 28. Pique, T.; Va´squez, A. Use of Fourier Transform Infrared Spectroscopy (FTIR). Concreto Cemento, Investigacio´n Desarrollo 2012, 3 (2), 62 71, In Spanish. 29. Garcia-Lodeiro.; Fernandez, A.; Palomo, A.; Macphee, D. E. Effect of Calcium Additions on N-A-S-H Cementitious Gels. J. Am. Ceram. Soc. 2010, 93 (7), 1934 1940. Available from: http://dx.doi.org/10.1111/j.1551-2916.2010.03668.x. 30. Va´zquez, T. Practical Applications of Infrared Absorption Spectroscopy in the Study Raw, Clinker and Anhydrous Portland Cement. Mater. Constr. 1979, 29 (175), 23 34 (In Spanish). Available from: http://dx.doi.org/10.3989/mc.1980.v30.i177.1065. 31. Colombian Technical Standards. NTC 919. Ceramic Tile. Definitions, Classification, Characteristics and Marking. 2000. (In Spanish). 32. Souza Santos P. Science and Technology for the Clays. Sa˜o Paulo, EDUSP, 1975. (In Portuguese). 33. Liew, A. G.; Idris, A.; Wong, C. H. K.; Samad, A. A.; Noor, M. J.; Baki, A. M. Incorporation of Sewage Sludge in Clay Brick and Its Characterization. Waste Manage. Res. 2004, 22 (4), 226 233. Available from: http://dx.doi.org/10.1177/ 0734242X04044989.

A study of a hybrid binder based on alkali-activated ceramic tile wastes and portland cement

311

34. Andrade, P. M.; Neto, H. S. N.; Monteiro, S. N.; Vieira, C. M. F. Effect of Phonolite Addition on Sintering Kaolinitic Clay. Ceraˆmica 2005, 51, 362 371 (In Portuguese). Available from: http://dx.doi.org/10.1590/S0366-69132005000400010. 35. Lee, W. K. W.; Van Deventer, J. S. J. Effects of Anions on the Formation of Aluminosilicate Gel in Geopolymers. J. Ind. Eng. Chem. 2002, 41, 4550 4558. Available from: http://dx.doi.org/10.1021/ie0109410. 36. Phair, J. W.; Smith, J. D.; Van Deventer, J. S. J. Characteristics of Aluminosilicate Hydrogels Related to Commercial “Geopolymers”. Mater. Lett. 2003, 57, 4356 4367. Available from: http://dx.doi.org/10.1016/S0167-577X(03)00325-2. 37. Stevenson, M.; Sagoe Crentsil, K. Relationship between Composition, Structure and Strength of Inorganic Polymers: Part 1 Metakaolin-derived Inorganic Polymers. J. Mater. Sci. 2005, 40 (8), 2023 2036. Available from: http://dx.doi.org/10.1007/s10853005-2794-x. 38. Criado, M.; Ferna´ndez-Jime´nez, A.; Palomo, A. Alkali Activation of Fly Ash: Effect of the SiO2/Na2O Ratio. Part I: FTIR Study. Micropor. Mesopor. Mater. 2007, 106 (1), 180 191. Available from: http://dx.doi.org/10.1016/j.micromeso.2007.02.055. 39. Pacheco-Torgal, F.; Castro-Gomes, J.; Jalali, S. Alkali-activated Binders: A Review. Part 1. Historical Background, Terminology, Reaction Mechanisms and Hydration Products. Constr. Build. Mater. 2008, 22, 1305 1314. Available from: http://dx.doi.org/ 10.1016/j.conbuildmat.2007.10.015. 40. Provis, J. L.; Harrex, R. M.; Bernal, A. S.; Duxson, P.; Van Deventer, J. S. J. Dilatometry of Geopolymers as a Means of Selecting Desirable Fly Ash Sources. J. Non-Cryst. Solids 2012, 358, 1930 1937. Available from: http://dx.doi.org/10.1016/j. jnoncrysol.2012.06.001.

Accelerated carbonation as a fast curing technology for concrete blocks

12

Caijun Shi1, Zhenjun Tu1,2, Ming-Zhi Guo2 and Dehui Wang1 1 Hunan University, Changsha, China, 2The Hong Kong Polytechnic University, Hong Kong, China

12.1

Introduction

It is well known that carbon dioxide is the main greenhouse gas, which had risen from preindustrial level of 280 ppm to about 403 ppm in 2015.1 This is accompanied by a gradual rise in mean global temperature as more solar radiation is trapped by the greenhouse effect.2 In 2015, the United Nations Framework Convention on Climate Change was signed by over 200 countries, agreeing in principle to limit the global average temperature increase to 2 C above preindustrial levels in this century, and that parties should take urgent actions to meet this long-term goal.3 By 2020, China has committed to decrease its CO2 emissions per unit of gross domestic product (GDP) by 40%45% from 2005 levels and to increase forest stock volume by 1.3 billion cubic meters in 2020.4 Therefore, there is an urgent need to develop carbon capture, utilization, and sequestration techniques to reduce the CO2 emissions into the atmosphere. As one of the major sources of greenhouse emissions, the cement industry contributes to about 7% of the total global CO2 emissions.5,6 Fortunately, accelerated carbonation as a fast curing technology for eco-efficient concrete blocks may hold the key to capturing and storing CO2 emissions from the cement industry for the production of value-added concrete products.7,8 If accelerated carbonation is applied immediately after casting, CO2 can chemically react with the silicate phases, mainly dicalcium siliciate (C2S) and tricalcium silicate (C3S), to form thermodynamically stable calcium carbonates in the presence of water or water vapor.913 If accelerated carbonation is performed after several hours of hydration, both calcium silicates and hydration products (e.g., calcium hydroxide [CH], calcium silicate hydrate [C-S-H], and ettringite) can react with CO2.1417 However, in either case, the hydration products of CH, C-S-H, and ettringite (AFt) are not abundant (even if accelerated carbonation is not applied immediately, cement is hydrated only for a short period of time), and their carbonation can be ignored, thus CO2 is expected to largely react with the silicate phases, rather than the hydration products.15 There is no possibility of carbon dioxide release after carbonation,10 while achieving a rapid increase in the strength of the carbonated concrete products due to the gradual decrease in porosity and a concomitant increase in the CO2 curing degree.4,18,19 The CO2 curing degree α, which is Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00015-2 © 2017 Elsevier Ltd. All rights reserved.

314 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

defined as the ratio of CO2 consumption during curing to the theoretical maximum consumed amount of CO2, can be calculated as follows20: α5

M2 2 M1 1 Mw 3 100% XCO2 max 3 MC

(12.1)

where: M1 5 mass of the specimens before accelerated carbonation [g]; M2 5 mass of the specimens after CO2 accelerated carbonation [g]; Mw 5 mass of moisture evaporated from the specimen after accelerated carbonation [g]; MC 5 mass of the cement in the specimen [g]; XCO2 max 5 the theoretical maximum CO2 consumption of the cement used. The theoretical maximum CO2 consumption of the cement used, which is based on the oxide contents (X) of the materials, can be calculated by using the following Eq. (12.2)21:  XCO2 max 5 0:785 XCaO 2 0:700XSO3 1 1:091XMgO 1 1:42XNa2 O 1 0:935XK2 O (12.2) where: XCaO 5 the mass percentage of CaO in cement; XSO3 5 the mass percentage of SO3 in cement; XMgO 5 the mass percentage of MgO in cement; XNa2 O 5 the mass percentage of Na2O in cement; XK2 O 5 the mass percentage of K2O in cement. According to the above Eq. (12.2), regardless of cement type, the theoretical maximum CO2 consumption increases with the CaO content, accordingly a much higher theoretical calculated value of the degree of CO2 curing may be obtained.22 Theoretically, one tonne of cement could consume half a tonne of CO2.23 In practice, the most widely used accelerated curing for concrete blocks is steam curing. However, it is an energy-intensive process and contributes a significant portion to the production costs. For instance, the energy consumption was 2300 kJ for each normal weight standard concrete block and 2500 kJ for each standard lightweight concrete block.4 The autoclave curing also has exceedingly high energy demands. The energy consumption is about 1.2 times higher than that needed by steam curing.24 During these curing processes, a temperature gradient exists in the concrete, and this nonuniform heating may cause cracking inside the concrete products.4 As an alternative to steam curing and/or autoclave curing, accelerated carbonation curing has fueled keen interest around the world since it was first investigated in the early 1970s.12,25,26 It has advantages not only in the capture and storage of carbon dioxide in response to climate change by storing CO2 as a solid calcium carbonate phase in carbonated concrete,4,9,20 but also in decreasing the duration

Accelerated carbonation as a fast curing technology for concrete blocks

315

time of curing at early ages and increasing the productivity and mechanical properties of concrete blocks with a very low energy consumption.4,9,20,22 It was found that the energy consumption of accelerated carbonation of concrete was nearly 500 kJ for each standard carbonated concrete block, which is about one-fifth of that of steam curing of concrete.27 Unlike weathering carbonation of hardened concrete in the atmosphere, the early compressive strength of concrete increased rapidly after the accelerated carbonation curing process.4 After being exposed for 5 min to 100% CO2 concentration under 0.4 MPa pressure, the compressive strength of carbonated C3S (3CaO  SiO2) and β-C2S (2CaO  SiO2) compacted pastes with a water-to-cement ratio of 0.08 both exceeded 20.0 MPa.28 After being exposed for 81 min to 100% CO2 at 0.1 MPa, the compressive strength of the C3S and β-C2S compacted mortars with a water-to-cement ratio of 0.125 was above 45 MPa, which was about three times higher than that of the hydrated reference.12 These researches also revealed that a water-to-cement ratio of around 0.13 was optimal for the CO2 consumption and the strength development of the compacted cement paste and mortar. However, at such a low water-to-cement ratio, it is difficult to prepare concrete, and there is not enough water available for calcium silicates dissolution and the subsequent formation of calcium carbonates.4,9,22 Fortunately, more recent research showed that concrete with a water-to-cement ratio higher than 0.35 can also achieve a substantial CO2 curing degree and satisfactory compressive strength by means of preconditioning the sample in a dry environment (T 5 22 6 3 C, RH 5 55 6 10%) for a few hours before being subjected to the accelerated carbonation process.4,9,22 After preconditioning in the dry environment, a water-to-cement ratio in the range between 0.16 and 0.20 was reported to be optimal for the CO2 curing degree and compressive strength of concrete.29 It was noteworthy that the rate of moisture evaporation during the dry preconditioning period has to be controlled so as to avoid cracking.4 The strength of carbonated concrete will be increased continuously with time during the further curing in a moist environment (T 5 22 6 3 C, RH . 95%) due to the hydration of unreacted cement particles in the concrete.22 Compared with those of steam curing, the carbonated concrete demonstrated better dimensional stability and a similar weathering performance.4 This chapter reviews the research progress on accelerated carbonation as a fast curing technique for concrete blocks. The impact of the combined use of preconditioning, subsequent water curing, and the selected accelerated carbonation on the efficiency of carbonation and the development of compressive strength of concrete is discussed. Moreover, the effects of a wide variety of factors, including raw materials, preconditioning conditions, CO2 concentration and pressure, and curing duration time, on the curing degree, mechanical properties, and dimensional stability of carbonated concrete blocks are examined. In addition, the kinetics and mechanisms of the accelerated carbonation, as well as the microstructure of carbonated concrete are also discussed. This review is expected to serve as a catalyst for a bettering understanding of the current landscape of accelerated carbonation as a fast curing technique, and tries to shed new light on the future research needs worth attention.

316 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

12.2

Development of accelerated carbonation technique

For a long time, CO2 in the atmosphere (0.03%0.06% v/v) has been used to cure lime-based materials to achieve the required stiffness and strength.4 It is well known that CO2 in the atmosphere also can penetrate slowly into hardened cement mortar and concrete and react with the alkaline hydration products to form calcium carbonates and other products, which is regarded as weather carbonation of concrete.4 Since the concentration of calcium hydroxide is usually higher than that of the other hydration products, the reaction is mainly between calcium hydroxide (CH) and CO2.30,31 The chemical reaction of CO2 with CH led to the formation of calcium carbonate on the surface of CH crystals and then its rate of reaction decreased with time because CO2 diffusion through this already carbonated layer is the rate-controlled step for further carbonation of CH.3235 As the available calcium hydroxide is gradually depleted via the reaction, another hydration product, C-S-H, becomes susceptible to carbonation.36 Carbonation of C-S-H resulted in the removal of calcium ions from the gel and the subsequent formation of calcium carbonate and decalcified C-S-H and/or amorphous silica gel.3742 Under the condition of weather carbonation, reactions between calcium silicates (C3S and C2S) and CO2 can be ignored because of their weak competition for CO2 in the presence of CH. It was found that the amount of CH consumed by carbonation is roughly 20 times and 50 times as great as that of C2S and C3S, respectively.36 It has been generally conceived that weather carbonation might result in a reduction in alkalinity and the cracking of the concrete cover, and thus impair the durability of reinforced concrete.4346 In contrast, in the accelerated carbonation of concrete at an early age, a large amount of tricalcium silicate and dicalcium silicate, rather than the hydration products, can chemically react with CO2 in the presence of water or water vapor to form stable calcium carbonates and to achieve a desirable strength within a few hours in a CO2 rich environment.12,25,47 For example, Castellote et al.48 observed that the amount of C3S and C2S decreased due to these constituents with CO2 in the presence of water or water vapor. This is because the formation of hydration products was rare within a short time of hydration prior to carbonation. Similar results were reported by Kashef-Haghighi et al.15 in which only approximately 14% of C3S and 25% of C3A were transformed to CH, C-S-H and AFt after 4 h of hydration. This accelerated carbonation curing technique is beneficial to many aspects of the cured concrete blocks, such as rapid compressive strength development,4,20,22 considerable sequestration of CO2,25,49 and improved dimensional stability.9 Accelerated carbonation curing of concrete dates back to the 1960s when it was introduced to reduce the shrinkage of concrete.50 The carbonation curing with flue gas (10%), pure CO2 (99.5%), or a combination of the two, was applied as a secondary curing regime after steam curing at early ages. In the 1970s, CO2 was used to accelerate the hydration reaction of cement in the presence of water or water vapor.25,26,28,47,51,52 In the late 1990s, this fast carbonation curing was employed to accelerate the setting in the cement-bonded particleboard production to yield dimensionally stable products for initial handling and to improve mechanical

Accelerated carbonation as a fast curing technology for concrete blocks

317

properties.5355 This process involved exposing cementitious materials to supercritical carbon dioxide (abbreviated as scCO2), which is a supercritical fluid phase of CO2 with the temperature and pressure exceeding 31.1 C and 7.38 MPa, respectively.56 Recent research has predominately focused on fast carbonation of concrete blocks using 100% CO2 at low pressure due to the global warming concerns.4,9,57 Shao et al.58 suggested that carbon dioxide recovered from cement kiln flue gas also can be used for concrete carbonation curing and thus the associated CO2 emissions might be reduced accordingly. It was also reported that the use of CO2 to cure recycled concrete aggregates was another attractive way of storing CO2 and improving their mechanical properties and durability.2,5,5962 More importantly, mortar samples prepared by the carbonated recycled concrete aggregates exhibited reduced drying shrinkage, water absorption and chloride migration coefficient compared with mortars made of uncarbonated recycled concrete aggregate.5,15,18,60,61,63

12.3

Factors influencing fast carbonation of concrete

12.3.1 Raw materials It is well known that Portland cement mainly consists of tricalcium silicate (3CaO  SiO2, denoted by C3S), dicalcium siliciate (2CaO  SiO2, denoted by C2S), tricalcium aluminate (3CaO  Al2O3, denoted by C3A), and tetracalcium aluminoferrite (4CaO  Al2O3  Fe2O3, denoted by C4AF). Previous studies showed that carbonation of C3S and C2S contributed to the CO2 consumption and rapid strength gain of cement-based materials.52,64 However, the difference in chemical compositions and physical properties of cement clinker minerals leads to different chemical kinetics and reactions with carbon dioxide. Berger et al.28 compared the early carbonation of the compacted C3S, β-C2S, and C3A pastes with a low water/solids ratio (0.050.175). It was found that the strength of the C3S and β-C2S pastes evolved rapidly, but the C3A paste gained a negligible strength. Moreover, the β-C2S paste had a higher initial strength compared to the C3S counterpart. Young et al.12 investigated the effect of CO2 curing time on the strength development of C3S and β-C2S mortars with a water-to-solid ratio of 0.125, and found that the rate of strength development of carbonated C3S mortars was higher than that of β-C2S mortars initially. However, after 81 min of accelerated carbonation, both C3S and β-C2S mortars achieved compressive strengths of approximately 48.0 MPa. Liu et al.65 investigated the accelerated carbonation of the main cement minerals (C3S, C2S, C3A, and C4AF) with a water-solid ratio of 0.126 under certain conditions. The results showed that the ability to absorb carbon dioxide is in the descending order of C3S . C2S . C3A . C4AF. For instance, 1 g of C3S and C2S could respectively absorb 0.168 g and 0.125 g of CO2 gas. On the other hand, carbonation of C3A and C4AF can be ignored due to their low content and the rare formation of ettringite and monosulphate (AFm).38,66 Bukowski et al.47 found that γ-C2S also could react with CO2 although it is not prone to hydration and, consequently,

318 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

it does not contribute to any strength development of ordinary concrete. In general, the reactions between CO2 and γ-C2S can be described as follows67: γ  2CaO  SiO2 1 2CO2 1 μH2 O ! 2CaCO3 1 SiO2  μH2 O

(12.3)

In practice, some supplementary cementitious materials, such as granulated blast furnace slag, steel slag, and coal fly ash, could also react with CO2 due to the presence of calcium oxide in these materials.16,23,6875 Increasing the specific area of these materials increases the reactivity of mineral phases and the subsequent CO2 curing degree.25 Furthermore, the surface of limestone powder can act as the thermodynamically favorable area for the precipitation of calcium carbonate and thus increase the CO2 curing degree.58,7680

12.3.2 Water-to-cement ratio The water-to-cement ratio is a very important factor that influences the accelerated carbonation of concrete blocks. High water content fills the pore system of concrete and impedes CO2 diffusion while low water content hinders the dissolution of CO2 and cementitious materials. Even so, this technology was successfully applied to fresh C3S and β-C2S pastes with a water-to-cement ratio varying from 0.05 to 0.175 and static compact forming.28 It was found that after being exposed for 5 min to 100% CO2 at 0.3 MPa immediately after compacting, the compressive strength of pastes was over 20 MPa. The compressive strength increased with an increasing water-to-binder ratio up to 0.13, after which a gradual decrease in strength was observed. Similar results were reported by Young et al.12 who studied fast carbonation of the C3S and β-C2S compacted mortars and found that a water-to-cement ratio of 0.125 was optimum for the strength development of those carbonated mortars. A much higher water-to-cement ratio may be inappropriate for accelerated carbonation due to the blocking of pores by water, which impaired the diffusion of CO2. However, such a low water-to-cement ratio made it difficult to prepare concrete, and not enough water was available for calcium silicates dissolution and the formation of calcium carbonates.4,9,22 Generally, concrete needs a water-to-cement ratio of 0.350.6 to facilitate the mixing and to maintain its workability.4 However, in the case of CO2 curing, excess water would block the microchannels and hinder the diffusion of CO2 into the concrete. As a result, this decreases the degree of CO2 curing and compressive strength of concrete.22 Recently, this problem was successfully addressed by the employment of preconditioning of the samples in a dry environment according to Shi et al.4,9,22 It was found that fast carbonation of concrete after preconditioning can significantly improve the degree of CO2 curing and increase the compressive strength of concrete.4,9,22 The use of preconditioning will be discussed in detail in the following section.

12.3.3 Sand-to-cement ratio and compaction pressure The effect of sand-to-cement ratio on the accelerated carbonation of concrete was rarely studied due to its insignificant role in the improvement of the CO2 curing

Accelerated carbonation as a fast curing technology for concrete blocks

319

degree, which was caused by the so-called dilution effect. In general, a decrease in strength was approximately linear with an increase in the sand-to-cement ratio.26 For the concrete prepared by compaction pressure, when higher compaction pressure was applied, samples always displayed a higher strength due to a decrease in porosity. However, for the concrete subjected to accelerated carbonation, a decrease of porosity may unfavorably hinder the diffusion of CO2 into the specimens, and subsequently decrease the CO2 curing degree, which in turn leads to the decrease of the compressive strength.20 It was revealed that compaction pressure and CO2 curing degree are inversely correlated. For example, Chang et al.81 investigated the influence of compaction pressure on the carbonation of calcium hydroxide prepared with compaction pressure at a range of 28 MPa and a constant water-to-solid ratio of 0.10. It was found that 2 MPa compaction pressure would have a higher increase in compressive strength than 8 MPa. To an extent, the reduction in porosity caused by higher compaction pressure might unfavorably cause a reduction in the diffusion rate of CO2 into concrete, and consequently result in a decrease of strength, at least to the extent to offset the gains in compressive strength.82 So it is necessary to obtain an optimal compaction pressure in the real manufacture of concrete blocks so as to allow the positive effect of the porosity reduction on strength to gain an upper hand over its negative effect on the CO2 curing degree. In other words, a proper sand-to-cement ratio accompanied with an appropriate casting method is necessary.9,22

12.3.4 Preconditioning before accelerated carbonation As mentioned previously, unlike steam curing of conventional concrete with a typical water-to-cement ratio of 0.350.6, concrete samples subjected to accelerated carbonation normally feature a very low water-to-cement ratio that ranges from 0.05 to 0.28.12 Further increase of the water-to-cement ratio generally leads to a decrease of the CO2 curing degree and compressive strength of carbonated concrete. This is explained by the fact that the water occupying a much higher portion of the overall pores in concrete would block the microchannels and hinder the penetration of CO2 into the concrete, and accordingly decrease the CO2 uptake and compressive strength of concrete.12,26 Shi et al.22 stated that, once the concrete blocks mixture was designed with a high water-to-cement ratio above 0.28, preconditioning is necessary to evaporate some water from the pore system so as to smooth the carbonation process, especially for the concrete with a water-to-cement ratio greater than 0.43. Selection and optimization of preconditioning curing conditions (e.g., relative humidity and temperature) is a critical step for the production of carbonated concrete with high CO2 consumption and excellent properties.4 In 2008, Shi et al.22 proposed an appropriate preconditioning curing process, in which the concrete was stored in a dry environment (T 5 22 6 3 C, RH 5 55 6 10%) before accelerated carbonation. Within 4 h of preconditioning, the CO2 curing degree of concrete sharply increased with the preconditioning time. However, under the moist environment (T 5 22 6 3 C, RH . 95%), the CO2 curing degree of the concrete decreased with the increase of the preconditioning time (Fig. 12.1).83 It indicated that the moisture loss rather than the duration time of preconditioning was the main driving

320 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

30

CO2 curing degree (%)

25

20

Preconditioned in the dry environment

15

Preconditioned in the moist environment

10 5

0

5

10

15

20

25

Preconditioning time (h)

Figure 12.1 Effect of preconditioning time on CO2 curing degree of concrete.83

Compressive strength (MPa)

15.0

2 h of CO2 curing after preconditioning period Immediately after preconditioning

10.0

5.0

0.0 2

4

8

12

18 Steam curing

Figure 12.2 Effect of preconditioning times (h) on compressive strength of concrete cured with CO2 at 0.14 MPa for 2 h, after dry preconditioning.4

force for CO2 consumption. In terms of compressive strength, after 218 h of preconditioning in the dry environment, all the concrete displayed almost the same strength after 2 h of accelerated carbonation, indicating that the total curing time could be reduced to 4 h (see Fig. 12.2). Compared with the concrete cured in steam for 24 h, the precured concrete, which was carbonated in a 100% CO2 chamber under 0.14 MPa for 2 h, demonstrated a similar strength development. Nonetheless the rate of moisture evaporation of concrete has to be controlled so as to avoid cracking.4

Accelerated carbonation as a fast curing technology for concrete blocks

321

12

Compressive strength (MPa)

10

8

6

4

5% 20%

2

50% 99.5%

0 0

4

8

12

16

20

24

Curing time (h)

Figure 12.3 Effect of the CO2 concentrations on compressive strength of the concrete specimens.20

12.3.5 Effect of concentration and pressure of CO2 During the accelerated carbonation process, the most important factors governing the diffusion of CO2 in concrete and the kinetics of carbonation reactions are the concentration and pressure of CO2.15 In fact, the two parameters have long been found to significantly affect the CO2 curing degree and compressive strength of concrete. In general, either increasing the CO2 concentration from 5% to 99.5% or increasing the CO2 pressure up to 0.2 MPa could accelerate carbonation and augment the strength development rate of concrete (Figs. 12.3 and 12.4).20 The reason is that a higher CO2 concentration encourages more rapid CO2 penetration, increases the CO2 gas solubility, and thus promotes the dissolving of Ca21 from the cement phases into the pore solution in concrete and the subsequent formation of carbonated products.84 However, if a large number of free water plugs the capillary pores, an increase in the CO2 concentration might not be sufficient to increase the CO2 curing degree of concrete.85 In other words, it is not always advantageous to increase the CO2 concentration to obtain a high CO2 curing degree.82,86 Similarly, carbonation with higher CO2 pressure led to deeper penetration of CO2 and a faster strength development.87,88 However, it should be noted that when the CO2 pressure was above 0.6 MPa, the pressure exerted an insignificant effect on both the CO2 curing degree and compressive strength.20,84 In some cases, increasing the CO2 pressure from 0.2 MPa to a higher value also had almost no additional effect.12 In a recent study, Xuan et al.57 have shown that the compressive strength of concrete blocks carbonated under a gas pressure of 0.01 MPa was even comparable with concrete subjected to carbonation under a gas pressure of 0.5 MPa, if the CO2 curing

322 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

CO2 curing degree (%)

40

30

20

5% 10

20% 50% 99.5%

0

0

4

8

12 16 Curing time (h)

20

24

Figure 12.4 Effect of the CO2 concentrations on CO2 curing degree of concrete specimens.20

time was extended to 24 h. From an economic perspective, it is highly recommended that a low CO2 gas pressure is employed to cure concrete blocks. Generally, the resulting microstructure of concrete carbonated under a high CO2 concentration and pressure is different from those subjected to weather carbonation. For hydrated cement pastes, increasing CO2 concentration might result in the increase of the polymerization of the C-S-H gel. When 3% CO2 was used, the microscopic changes of concrete were similar to those undergoing weather carbonation in an atmospheric environment. However, when carbonated at 10% and 100% CO2, the samples experienced a complete disappearance of the C-S-H gel and the concomitant formation of a polymerized Ca-modified silica gel.48 Furthermore, the percentage of calcite in the carbonated samples increased as the concentration of CO2 increased.48 The increase of the CO2 concentration led to the formation of more calcium carbonates, making the microstructure denser.82 Although the above studies demonstrated similar results, different interpretations still hold. Bukowski et al.47 considered that under high CO2 pressure, the main carbonated products with particle size as large as 10 μm were formed, which provided a less homogeneous and/or a weaker bond than smaller crystals (,1 μm) produced in the compacted specimens carbonated at 0.1 MPa of CO2 pressure. Shi et al.20 pointed out that there was a limit to extract the calcium ions, which could be combined with carbonic acid ions at such a high pressure, while Phung et al.89 thought that a combination of diffusion and advection should be taken into account if a high pressure was chosen. The longer the carbonation time, the more important the contribution of advection in the transportation of CO2 gas.90 Furthermore, when more carbonate particles deposited directly on the cement surface, further reactions were impeded. And this

Accelerated carbonation as a fast curing technology for concrete blocks

323

is particularly true in the case of higher CO2 pressure due to faster accumulation of carbonate particles on the unreacted cement particles.15 It was noteworthy that under supercritical conditions, however, instead of the CO2 concentration, CO2 pressure was the main driving force for CO2 diffusion in the concrete matrix.88 In this condition, elevating the CO2 pressure could significantly increase the strength of Portland cement composite.91

12.3.6 Further curing after accelerated carbonation Since accelerated carbonation of concrete has been shown to contribute to calcium carbonate precipitation and the strength development at an early stage, further curing can be used to strengthen the mechanical properties and durability of carbonated concrete due to further hydration of uncarbonated cement.9,13,92 The strength development of CO2-cured concrete continuously and gradually increased with time at a relatively lower rate due to the significant decrease of the available water after carbonation.12,13,22,93,94 The influence of further curing environments was studied by Shi and Wu.22 The authors found that the strength of the carbonated concrete blocks increased continuously with curing time in both a dry environment and a moist environment (Dry environment: T 5 22 6 3 C, RH 5 55 6 10% and/or moist environment: at T 5 22 6 3 C, RH . 95%). The compressive strength development of the concrete blocks after accelerated carbonation was similar to that of steam curing9 (See Fig. 12.5). It was suggested that carbonation may not hinder the further hydration of calcium silicates and thus allow the carbonated concrete to gain more strength. Unfortunately, there was an inversely proportional relationship between the CO2 curing degree already attained and the strength gain rate.19 This phenomenon was similar 20 SC –Steam curing; MC –Moist curing; CC –CO2 curing

Compressive strength (MPa)

18 16 14 12 10 8 6

SC

SC+28dMC

CC

CC+28d MC CC+90d MC

Curing method

Figure 12.5 Compressive strength development of the concrete blocks after steam and CO2 curing.9

324 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

to the one observed by He et al.13 When further water curing was applied, carbonated concrete with a residual water-to-cement ratio of 0.18 demonstrated the highest CO2 curing degree but a slightly lower strength gain rate than those with either a higher or lower residual water-cement ratio. It was also found that the subsequent water curing of the CO2-cured concrete could form C-S-H gel and ettringite crystals. In addition, the calcium carbonate formed as a result of carbonation was partially consumed during the subsequent water curing due to the reaction of the produced calcium carbonate with C3A, forming hydrated tetracalcium monocarboaluminate (3CaO  Al2O3  CaCO3  11H2O).13,94 Similar results were obtained during normal hydration of cement containing limestone powders.9599 The chemical reaction between calcium carbonate and C3A and the formation of hydrated tetracalcium monocarboaluminate are described as follows100102: 3CaO  Al2 O3 1 CaCO3 1 11H2 O ! 3CaOAl2 O3  CaCO3  11H2 O

12.4

(12.4)

Mechanism, carbonated products, and microstructure

12.4.1 Mechanism The mechanism of fast carbonation curing of concrete was first studied by researchers at the University of Illinois in the 1970s, with the intention of using carbon dioxide to accelerate strength development.25,28,49 In recent years, fast carbonation of concrete has been attracting more and more attention as it can potentially be a substitute for the traditional steam curing with the advantage of reducing energy consumption, capturing CO2 from industrial sources, and at the same time improving the performance of concrete. It is different from that of the natural carbonation of hardened concrete in an atmospheric environment due to the minor formation of hydration products before carbonation. Considering that carbonation was usually applied immediately after casting or within several hours of preconditioning, the contents of hydration products (e.g., CH, C-S-H, and ettringite) are not abundant, accordingly their carbonation can be ignored.38 In other words, during this fast carbonation curing process, CO2 is usually chemically reacted with the main silicate phases (C3S and C2S) in the presence of water to form stable calcium carbonates and silica gel without producing calcium hydroxide,12,36,48,103 as described in Eqs. (12.4) and (12.5) 3CaO  SiO2 1 3CO2 1 μH2 O ! SiO2  μH2 O ðSilica gelÞ 1 3CaCO3

(12.5)

2CaO  SiO2 1 2CO2 1 μH2 O ! SiO2  μH2 O ðSilica gelÞ 1 2CaCO3

(12.6)

The above reactions are strongly exothermic, with calculated heats of carbonation of C3S and β-C2S of 347 kJ/mol and 184 kJ/mol, respectively.25 In a study by

Accelerated carbonation as a fast curing technology for concrete blocks

325

Monkman and Shao,23 X-ray diffraction (XRD) analysis showed that significant consumption of C3S and C2S has occurred due to carbonation. Full carbonation, however, is not reached during the accelerated carbonation process because a dense calcium carbonate layer was formed on the surface of the cement particles and thus impeded their further carbonation.2,19,60,104 Shao et al.105 concluded that after 2 h of carbonation, Portland cement could gain more than 14% CO2 by mass of the dry cement in concrete. As shown in a recent study, after 18 h of preconditioning in an environmental chamber at a relative humidity of 50% and a temperature of 25 C and followed by accelerated carbonation in a sealed chamber filled with carbon dioxide gas at a pressure of 0.1 MPa for 4 days, the samples only reached 70% of the theoretical full carbonation degree.106 It may take long-term curing for full carbonation, which occurred when the phenolphthalein pink color disappeared completely or the pH value decreased to 8.3.44,107109 Although the formation of silica gel is still not clear due to its amorphous structure, it is generally accepted that the carbonation of the main silicates phases in the Portland system leads mainly to the formation of a CaO-SiO2-H2O-like gel initially,12,28 which is then transformed into amorphous silica gel.12 In contrast, the natural carbonation of concrete in atmospheric conditions occurs mainly between CO2 and the cement hydration products, such as C-S-H, CH, and hydrated calcium aluminum sulfates (AFt and AFm).110 The main chemical reaction involved can be described as follows48,52,111: C-S-H 1 CO2 ! CaCO3 1 SiO2  nH2 O

(12.7)

CaðOHÞ2 1 CO2 ! CaCO3 1 H2 O

(12.8)

3CaO  Al2 O3  3CaSO4  32H2 O ðsÞ 1 3CO2 ðaqÞ ! 3CaCO3 1 3CaSO4  2H2 O 1 2AlðOHÞ3 ðsÞ 1 9H2 O

(12.9)

Calcium hydroxide seems to be the first phase that is attacked during carbonation.112 However, some studies64,113 considered that CO2 simultaneously reacted with C-S-H and calcium hydroxide. Some researchers40,114 pointed out that the carbonation rate of CH may be initially more rapid than that of C-S-H due to the lower calculated CO2 partial pressure needed for carbonation of calcium hydroxide than that of C-S-H, but with the formation of calcium carbonate, the carbonation rate of CH decreased correspondingly. A recent study by Shi et al.98 predicted that carbonation of CH was followed by carbonation of C-S-H using thermodynamic modeling based on the observation that some high-Ca C-S-H phase was still present when the full carbonation of CH was achieved. In contrast, progressive carbonation of C-S-H led to a decrease of its Ca/Si ratio, making it more vulnerable to decomposition. And the rate of C-S-H decomposition increased with a decreasing Ca/Si ratio.37 It is widely accepted that the decomposition of C-S-H caused by carbonation involves two steps.37 Firstly, calcium is gradually removed from the interlayer and the defect sites in the silicate chains until the Ca/Si ratio decreases to 0.67. Further carbonation might result in the consumption of Ca21 ions in the principal layers.37

326 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Carbonation of C-S-H also resulted in the precipitation of three different polymorphs of crystalline calcium carbonates (calcite, aragonite, and vaterite) and some amorphous calcium carbonate (ACC), apart from the formation of decalcified C-S-H gel.115 The precipitation of metastable carbonates (aragonite and vaterite) seems to be related to the presence of highly decalcified C-S-H.31,116,117 In addition, the quantity of calcium carbonate formed by carbonation of C-S-H depends strongly on the initial Ca/Si ratio of the C-S-H gel. The higher the Ca/Si ratio, the higher the amount of produced calcium carbonate.41,42,113,118,119 Other calciumbearing phases, such as AFt and AFm, seem to be completely carbonated after carbonation.120 The carbonation of ettringite led to the formation of vaterite crystals.121 However, considering that there is proportionally less C3A and C4AF content in Portland cement and thus the content of formed AFt and AFm in the hydrated cement is small, their contribution to the overall carbonation is rather small, so accordingly it may be ignored.38,66

12.4.2 Carbonated products The carbonated products of concrete after the fast carbonation are normally identified as calcium carbonate, decalcified C-S-H, and/or silica gel.9,22 However, generation of these carbonated products depended on various factors, such as raw materials, water-to-cement ratio, preconditioning, CO2 concentration, pressure, time, and further curing (as discussed previously). It is well known that the polymorphism of calcium carbonate has three anhydrous crystalline structures, namely calcite, aragonite, and vaterite (in the order of decreasing solubility and increasing stability).122 In addition, calcium carbonate can also precipitate in ACC, hexahydrate calcium carbonate (HCC), and monohydrate calcium carbonate (MCC).123 Goto et al.124 investigated the carbonated products of C3S powder cured with 5%CO2-bearing N2 gas under saturated humidity at room temperature. After 48 h of carbonation, the carbonated products of the C3S powder were found to be mainly calcite with the presence of some vaterite and aragonite. However, for hydrated C3S, the carbonates formed after carbonation were mainly vaterite and aragonite due to carbonation of calcium silicate hydrate rather than the C3S phase.38,42,48,64 Increasing the hydration degree of the C3S might increase the mass ratio of vaterite and aragonite.64 In other words, vaterite was observed as a primary carbonation product at Ca/Si $ 0.67, whereas aragonite was favorably formed at Ca/ Si # 0.67.42 All the unstable carbonates (aragonite and vaterite) produced during the carbonation process might be transformed into more stable polymorphs after a further moist curing process,17 and calcite might be the ultimate product because it is the most stable polymorph of CaCO3.125 Shtepenko et al.126 investigated the carbonated products of accelerated-carbonated Portland cement using the complementary analytical techniques of X-ray diffraction analysis (XRD), Scanning electron microscope (SEM), thermogravimetry-differential thermal analysis (TG-DTA), and magic angle spinning nuclear magnetic resonance spectroscopy (MAS-NMR) and found that both calcite and aragonite were formed without the presence of vaterite. It is interesting to note that aragonite is preferably formed

Accelerated carbonation as a fast curing technology for concrete blocks

327

under high relative humidity and without added water, whereas, calcite is predominantly formed with added water. This helps to explain why the mainly carbonated products are normally identified as calcite rather than other forms for the carbonated concrete as mentioned by Shi et al.29 From a thermodynamic point of view, the produced carbonates could be described as two types. One type was a poorly defined crystalline calcium carbonate, which was decomposed at a very low temperature (below 600  C). The other was a well-defined crystalline calcium carbonate structure, which was usually decarbonated at temperatures above 600 C.124 In contradiction, Rostami et al.11 and Villain et al.127 suggested that the decomposition of poorly defined crystalline calcium carbonate occurred at a temperature between 540 C and 720 C, while the decarbonation of well-defined crystalline calcium carbonate structure occurred at a temperature between 720 C and 950 C. The appearance of the decalcified C-S-H and/or silica gel was amorphous, making it difficult to be distinguished from other phases. Usually, during the accelerated carbonation process, the initial carbonated products are a C-S-H-like gel, which is similar to that found in normal hydration.12 If carbonation is applied with carbon dioxide combined with water vapor, the C-S-H gel formed was minimal.25 Goto et al.124 however, questioned the formation of a silica-gel-like reaction product when they investigated the carbonation products of calcium silicates (e.g., C3S, β-C2S, and γ-C2S) under saturated humidity at room temperature and suggested that an amorphous calcium silicate hydrocarbonate binding phase was formed during the carbonation reaction because the ratios of H2O and CO2 for the carbonated products were similar to those for the basic hydrated calcium carbonate (2CaCO3  Ca(OH)2  1.5H2O) and amorphous basic hydrated calcium carbonate (Ca3(OH)6x(CO3)33x  3yH2O). Rostami et al.11 also suggested that early accelerated carbonation of cement paste after 18 h of preconditioning in a controlled air chamber led to the formation of an amorphous calcium silicate hydrocarbonate binding phase. El-Hassan et al.128 investigated reaction products in carbonation-cured lightweight concrete and found that the ratios of poorly hydrated C-S-H to amorphous CaCO3 were almost constant for the air-cured carbonated concrete, indicating that the amorphous carbonates and poorly hydrated C-S-H were one component. The formation of silica gel was also reported by some previous studies, which might result from the initial carbonated products of decalcified C-S-H, in which its CaO/SiO2 ratios progressively decreased with increasing the duration time of carbonation. It was similar to the carbonation of C-S-H phase in normal hydration concrete, which takes place in two steps. Firstly, calcium is gradually removed from the interlayer and the defect sites in the silicate chains until Ca/Si 5 0.67. After that calcium from the principal layer is consumed, forming an amorphous silica gel.37,48,98,113 It seems that the samples with Ca/Si 5 0.67 were most resistant to decomposition into amorphous silica gel.42,119 Accelerated carbonation of C-S-H in high CO2 concentrations might shorten the curing duration time required to completely decompose the CS-H gel.129 Another typical hydration product, CH, was usually not observed after accelerated carbonation of concrete. Although there is evidence in the literature113 that CO2 simultaneously reacts with C-S-H and CH, carbonation of CH may be initially more rapid than that of the C-S-H gel. This might be due to the formation of a layer of CaCO3 microcrystals at the surface of CH.31

328 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

12.4.3 Microstructure Accelerated carbonation is performed immediately after casting or within several hours of hydration. Therefore, the changes in microstructure induced by carbonation are mainly originating from the consumption of cement (mainly C3S and C2S), the precipitation of calcium carbonates, and the progressive decalcification of the initially formed C-S-H gel. During the accelerated carbonation process, both decalcified C-S-H gel and silica gel may be produced. The three different polymorphs of calcium carbonates—calcite, aragonite, and vaterite—which have different unit cell volumes, might be produced. Application of accelerated carbonation could improve the microstructure of both hydrated and fresh concrete mixtures. For hydrated concrete, the hydrated products (e.g., CH, C-S-H, and ettringite) were consumed to produce many cubic and crystalline-shaped particles,130 and consequently the microstructure became denser and smoother.18 SEM observations, as shown in Fig. 12.6, indicated that the structure of carbonated paste had a denser texture compared to the air-cured paste, where calcium silicate hydrates were present in the vicinity of calcium silicates after 20 h of curing (Fig. 12.7).11 The Energy Dispersive Spectroscopy (EDS) spot analysis suggested that monolithic amorphous products containing relatively high carbon contents were formed after carbonation.

Figure 12.6 Carbonated OPC paste at 20 h: (A) SEM photomicrograph; (B) EDS analysis.11

Figure 12.7 Air-cured OPC paste at 20 h: (A) SEM photomicrograph; (B) EDS analysis.11

Accelerated carbonation as a fast curing technology for concrete blocks

329

Figure 12.8 Carbonated OPC paste at 28 days: (A) SEM photomicrograph; (B) EDS analysis.11

As compared with those cured in air or water, carbonated concrete demonstrated lower porosity. In general, carbonation of C3S, C2S, CH, C-S-H, AFt, and AFm led to the solid volume increases of 92.5%, 108.7%, 11.5%, 23.1%, 44.9%, and 31.8%, respectively.5,31,52,111,131,132 Use of further water for curing increased the formation of hydration products, which were confirmed by the EDS analysis, as shown in Fig. 12.8, and thus further decreased the porosity. It was reported that the porosity of the carbonated concrete blocks decreased by 4.4%12.2% after 90 days of further hydration.13 Considering that the formation of hydration products could also lead to a reduction in porosity, the durability of carbonated concrete can be expected to be further improved.13,133 It should be pointed out that further hydration up to 28 days did not increase the CH content.13,92 The lack of CH could be partly due to the complete carbonation of the calcium silicate phases.92 Another cause of the disappearance of CH may be its reaction with the formation of silica gel during carbonation.13,134

12.5

Dimensional stability

Dimensional stability is one of the most important weathering properties of concrete, especially for accelerated carbonation-cured concrete blocks. The widely observed improvement in dimensional stability was attributed to modifications in the micropore structure and substantial reductions in porosity. Compared with steam-cured concrete blocks, the CO2-cured blocks demonstrated lower drying shrinkage, water absorption, and higher durability.9 A higher carbonation rate can favorably decrease the porosity of the carbonated concrete by sealing the open pores, and thus lead to continuous decrease of water absorption.135 Shi et al.9 investigated the weathering performance of carbonated concrete and compared it with those of steam curing with maximum curing temperature 5 55.8 C and found that fast carbonation can decrease the drying shrinkage (Fig. 12.9), and thus improve the dimensional stability of concrete. This finding was consistent with that of others.27,56,133,136,137 It was found that the CO2-cured cement mortar had excellent dimensional stability due to

330 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

0.10 Steam cured

0.09

CO2 cured

Drying shrinkage (%)

0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00

0

4

8

12

16

20

24

28

Drying time/Days

Figure 12.9 Drying shrinkage of concrete blocks after 180 days of winter weathering exposure.9

significant productions of CaCO3, which is a well crystallized product and exhibits a better dimensional stability than the main hydration products, C-S-H, in steam-cured concrete blocks.27 Carbonation-induced modifications in micropore structures of the matrix might be also responsible for the increase in dimensional stability,137 especially for the scCO2 treated concrete.138 In previous studies,139,140 it was found that the addition of limestone powders as a cement replacement would also improve the dimensional stability of the normal hydrated concrete by reducing the porosity, permeability, and sorptivity. Moisture reduction inside the CO2-cured specimens and the correspondingly slowed hydration reaction might also decrease the drying shrinkage and thus improve the dimensional stability.

12.6

Conclusions

Fast carbonation curing, an alternative to steam curing and/or autoclave curing, is a renewed and sustainable concrete curing technique for recycling and reutilization of CO2 gas emissions in response to climate change. The knowledge of fast carbonation of concrete is still developing and evolving. However, findings from the previous studies could be summarized as follows: 1. Accelerated carbonation as a fast curing technique for concrete blocks refers to a process whereby carbon dioxide mainly chemically reacts with the main silicate phases (C3S and C2S) to form stable calcium carbonates without producing calcium hydroxide either immediately after casting or within several hours of preconditioning, although minimal hydration products were formed before early accelerated carbonation.

Accelerated carbonation as a fast curing technology for concrete blocks

331

2. The ability of the four typical clinker phases in Portland cement to absorb carbon dioxide is in the descending order of C3S . C2S . C3A . C4AF. 3. A water-to-cement ratio of 0.350.6 can be used to facilitate the manufacture and subsequent accelerated carbonation of the concrete blocks by employment of the preconditioning in a dry environment (T 5 22 6 3 C, RH 5 55 6 10%) before carbonation so as to evaporate excess water from the pore system and improve the diffusion of CO2 in concrete. 4. Increasing CO2 concentration and pressure will accelerate carbonation of concrete and increase its strength development rate. However, if the CO2 pressure was greater than a certain value, i.e., 0.5 MPa, its influence on carbonation of concrete became rather ineffective. 5. Further water curing plays a critical role in influencing the mechanical properties and durability of carbonated concrete. The strength of carbonated concrete increased continuously with time during further water curing due to the hydration of unreacted cement particles. 6. The carbonated products of concrete after fast carbonation are normally identified as calcium carbonate, decalcified C-S-H gel, and/or silica gel. The morphology and thermodynamic stability of calcium carbonate was dependent on the available phases for carbonation and the curing parameters, and the finally produced calcium carbonate may be mainly calcite due to its super stability. The production of decalcified C-S-H gel or silica gel still remains a subject of debate, but the progressive decalcification of C-S-H gel and its ultimate transformation to amorphous silica gel were confirmed. 7. Due to the precipitation of calcium carbonate in the pores, the CO2-cured concrete blocks demonstrated a reduction in porosity, drying shrinkage, and water absorption, while displayed higher compressive strength and improved dimensional stability compared with their counterparts subjected to steam curing.

Despite the perceived advantages and benefits of using accelerated carbonation as an alternative to steam curing for concrete blocks, there are still some limitations worth special attention: 1. Apart from Portland cement, some common supplementary cementitious materials, such as granulated blast furnace slag, steel slag, and coal fly ash, can also react with CO2 due to the inclusion of calcium oxide in these materials. These materials (usually used as a partial cement replacement) merit keen attention so as to produce environmentallyfriendly concrete with low cement contents. 2. To confirm the suitability of using CO2 for concrete curing, durability tests should be performed following standard procedures in the future research. 3. Further water curing has been found to be beneficial for the strength development of carbonated concrete, but its effect on the durability properties needs to be examined in order to facilitate the production and wider application of carbonated concrete.

Acknowledgment The authors would like to acknowledge the financial support from Specialized Research Fund for the Doctoral Program of Higher Education (SRFDP) (20130161110019) and the Research Institute for Sustainable Urban Development of the Hong Kong Polytechnic University.

332 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

References 1. Ghacham, A. B.; Cecchi, E.; Pasquier, L. C.; Blais, J. F.; Mercier, G. CO2 Sequestration Using Waste Concrete and Anorthosite Tailings by Direct Mineral Carbonation in Gassolid-liquid and Gas-solid Routes. J. Environ. Manage. 2015, 163, 7077. Available from: http://dx.doi.org/10.1016/j.jenvman.2015.08.005. 2. Zhan, B. J.; Poon, C. S.; Shi, C. J. Materials Characteristics Affecting CO2 Curing of Concrete Blocks Containing Recycled Aggregates. Cement Concrete Compos. 2016, 67, 5059. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2015.12.003. 3. Olajire, A.; Bass, A. A Review of Mineral Carbonation Technology in Sequestration of CO2. J. Pet. Sci. Eng. 2013, 109, 364392. Available from: http://dx.doi.org/10.1016/j. petrol.2013.03.013. 4. Shi, C. J.; He, F. Q.; Wu, Y. Z. Effect of Pre-conditioning on CO2 Curing of Lightweight Concrete Blocks Mixtures. Constr. Build. Mater. 2012, 26 (1), 257267. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2011.06.020. 5. Zhang, J. K.; Shi, C. J.; Li, Y. K.; Pan, X. Y.; Poon, C. S.; Xie, Z. B. Performance Enhancement of Recycled Concrete Aggregates through Carbonation. J. Mater. Civ. Eng. 2015, 27 (11), 04015029. Available from: doi:10.1061/(Asce)Mt.1943-5533.0001296. 6. Shi, C. J.; A Fernandez, J.; Palomo, A. New Cements for the 21st Century: The Pursuit of an Alternative to Portland Cement. Cement Concrete Res. 2011, 41 (7), 750763. Available from: http://dx.doi.org/10.1016/j.cemconres.2011.03.016. 7. Kashef-Haghighi, S.; Ghoshal, S. Physico-Chemical Processes Limiting CO2 Uptake in Concrete during Accelerated Carbonation Curing. Ind. Eng. Chem. Res. 2013, 52 (16), 55295537. Available from: http://dx.doi.org/10.1021/ie303275e. 8. Hossain, M. U.; Poon, C. S.; Lo, I. M. C.; Cheng, J. C. P. Evaluation of Environmental Friendliness of Concrete Paving Eco-blocks Using LCA Approach. Int. J. Life Cycle Assess. 2016, 21 (1), 7084. Available from: http://dx.doi.org/10.1007/ s11367-015-0988-2. 9. Shi, C. J.; Wang, D. H.; He, F. Q.; Liu, M. Weathering Properties of CO2-cured Concrete Blocks. Resour. Conserv. Recy. 2012, 65, 1117. Available from: http://dx. doi.org/10.1016/j.resconrec.2012.04.005. 10. Perez-Moreno, S. M.; Gazquez, M. J.; Bolivar, J. P. CO2 Sequestration by Indirect Carbonation of Artificial Gypsum Generated in the Manufacture of Titanium Dioxide Pigments. Chem. Eng. J. 2015, 262, 737746. Available from: http://dx.doi.org/ 10.1016/j.cej.2014.10.023. 11. Rostami, V.; Shao, Y. X.; Boyd, A. J.; He, Z. Microstructure of Cement Paste Subject to Early Carbonation Curing. Cement Concrete Res. 2012, 42 (1), 186193. Available from: http://dx.doi.org/10.1016/j.cemconres.2011.09.010. 12. Young, J. F.; Berger, R. L.; Breese, J. Accelerated Curing of Compacted Calcium Silicate Mortars on Exposure to CO2. J. Am. Ceram. Soc. 1974, 57 (9), 394397. Available from: http://dx.doi.org/10.1111/j.1151-2916.1974.tb11420.x. 13. He, P. P.; Shi, C. J.; Tu, Z. J.; Poon, C. S.; Zhang, J. K. Effect of Further Water Curing on Compressive Strength and Microstructure of CO2-cured Concrete. Cement Concrete Compos. 2016, 72, 8088. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2016.05.026. 14. Zhang, D.; Shao, Y. X. Early Age Carbonation Curing for Precast Reinforced Concretes. Constr. Build. Mater. 2016, 113, 134143. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2016.03.048.

Accelerated carbonation as a fast curing technology for concrete blocks

333

15. Kashef-Haghighi, S.; Shao, Y. X.; Ghoshal, S. Mathematical Modeling of CO2 Uptake by Concrete During Accelerated Carbonation Curing. Cement Concrete Res. 2015, 67, 110. Available from: http://dx.doi.org/10.1016/j.cemconres.2014.07.020. 16. Zhang, D.; Cai, X. H.; Shao, Y. X. Carbonation Curing of Precast Fly Ash Concrete. J. Mater. Civ. Eng. 2016, 04016127. doi:10.1061/(ASCE)MT.1943-5533.0001649. 17. El-Hassan, H.; Shao, Y. X. Early Carbonation Curing of Concrete Masonry Units with Portland Limestone Cement. Cement Concrete Compos. 2015, 62, 168177. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2015.07.004. 18. Xuan, D. X.; Zhan, B. J.; Poon, C. S.; Zheng, W. Carbon Dioxide Sequestration of Concrete Slurry Waste and Its Valorisation in Construction Products. Constr. Build. Mater. 2016, 113, 664672. Available from: http://dx.doi.org/10.1016/j.conbuildmat. 2016.03.109. 19. Zhan, B. J.; Xuan, D. X.; Poon, C. S.; Shi, C. J. Effect of Curing Parameters on CO2 Curing of Concrete Blocks Containing Recycled Aggregates. Cement Concrete Compos. 2016, 71, 122130. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2016.05.002. 20. Shi, C. J.; Liu, M.; He, P. P.; Ou, Z. H. Factors Affecting Kinetics of CO2 Curing of Concrete. J. Sustain. Cement-Based Mater. 2012, 1 (12), 2433. Available from: http://dx.doi.org/10.1080/21650373.2012.727321. 21. Steinour, H. H. Some Effects of Carbon Dioxide on Mortars and Concrete-discussion. J. Am. Concrete Inst. 1959, 30 (2), 905907. 22. Shi, C. J.; Wu, Y. Z. Studies on Some Factors Affecting CO2 Curing of Lightweight Concrete Products. Resour. Conserv. Recy. 2008, 52 (8-9), 10871092. Available from: http://dx.doi.org/10.1016/j.resconrec.2008.05.002. 23. Monkman, S.; Shao, Y. X. Assessing the Carbonation Behavior of Cementitious Materials. J. Mater. Civ. Eng. 2006, 18 (6), 768776. Available from: http://dx.doi.org/ 10.1061/(Asce)0899-1561(2006)18:6(768). 24. Kawai, K. J.; Sugiyama, T.; Kobayashi, K.; Sano, S. Inventory Data and Case Studies for Environmental Performance Evaluation of Concrete Structure Construction. J. Adv. Concrete Technol. 2005, 3 (3), 435456. Available from: http://dx.doi.org/10.3151/jact.3.435. 25. Goodbrake, C. J.; Young, J. F.; Berger, R. L. Reaction of Beta-Dicalcium Silicate and Tricalcium Silicate with Carbon-Dioxide and Water-Vapor. J. Am. Ceram. Soc. 1979, 62 (3-4), 168171. Available from: http://dx.doi.org/10.1111/j.1151-2916.1979.tb19046.x. 26. Klemm, W. A.; Berger, R. L. Accelerated Curing of Cementitious Systems by Carbon Dioxide: Part I. Portland Cement. Cement Concrete Res. 1972, 2 (5), 567576. Available from: http://dx.doi.org/10.1016/0008-8846(72)90111-1. 27. Shi, C. J.; Wu, Y. Z. CO2 Curing of Concrete Blocks-research Toward Greener Production. Concrete Int. 2009, 31 (02), 3943. 28. Berger, R. L.; Klemm, W. A. Accelerated Curing of Cementitious Systems by Carbon Dioxide. Cement Concrete Res. 1972, 2 (6), 647652. Available from: http://dx.doi.org/ 10.1016/0008-8846(72)90002-6. 29. Shi, C. J.; He, P. P.; Tu, Z. J.; Cao, Z. Effects of Pre-conditioning on Process and Microstrucutre of Carbon Dioxide Cured Concrete. J. Chin. Ceram. Soc. 2014, 42 (8), 9961004. Available from: http://dx.doi.org/10.7521/j.issn.0454-5648.2014.08.07. 30. Saetta, A. V.; Schrefler, B. A.; Vitaliani, R. V. The Carbonation of Concrete and the Mechanism of Moisture, Heat and Carbon-Dioxide Flow-through Porous Materials. Cement Concrete Res. 1993, 23 (4), 761772. Available from: http://dx.doi.org/ 10.1016/0008-8846(93)90030-D. 31. Thiery, M.; Villain, G.; Dangla, P.; Platret, G. Investigation of the Carbonation Front Shape on Cementitious Materials: Effects of the Chemical Kinetics. Cement

334 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

Concrete Res. 2007, 37 (7), 10471058. Available from: http://dx.doi.org/10.1016/j. cemconres.2007.04.002. Garcia-Gonzalez, C.; Hidalgo, A.; Andrade, C.; Alonso, M. C.; Fraile, J.; Lopez-Periago, A. M., et al. Modification of Composition and Microstructure of Portland Cement Pastes as a Result of Natural and Supercritical Carbonation Procedures. Ind. Eng. Chem. Res. 2006, 45 (14), 49854992. Available from: http://dx.doi.org/10.1021/ie0603363. ¨ .; Balen, K. V.; Elsen, J.; Gemert, D. V. Real-time Investigation of Reaction Cizer, O Rate and Mineral Phase Modifications of Lime Carbonation. Constr. Build. Mater. 2012, 35, 741751. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2012.04.036. Galan, I.; Glasser, F. P.; Baza, D.; Andrade, C. Assessment of the Protective Effect of Carbonation on Portlandite Crystals. Cement Concrete Res. 2015, 74, 6877. Available from: http://dx.doi.org/10.1016/j.cemconres.2015.04.001. Richardson, I. G.; Groves, G. W.; Brough, A. R.; Dobson, C. M. The Carbonation of OPC and OPC/silica Fume Hardened Cement Pastes in Air Under Conditions of Fixed Humidity. Adv. Cement Res. 1993, 5 (18), 8186. Available from: http://dx.doi.org/ 10.1680/adcr.1993.5.18.81. Peter, M. A.; Muntean, A.; Meier, S. A.; Bohm, M. Competition of Several Carbonation Reactions in Concrete: A Parametric Study. Cement Concrete Res. 2008, 38 (12), 13851393. Available from: http://dx.doi.org/10.1016/j.cemconres.2008.09.003. Sevelsted, T. F.; Skibsted, J. Carbonation of C-S-H and C-A-S-H Samples Studied by 13 C, 27Al and 29Si MAS NMR Spectroscopy. Cement Concrete Res. 2015, 71, 5665. Available from: http://dx.doi.org/10.1016/j.cemconres.2015.01.019. Morandeau, A.; Thiery, M.; Dangla, P. Investigation of the Carbonation Mechanism of CH and C-S-H in Terms of Kinetics, Microstructure Changes and Moisture Properties. Cement Concrete Res. 2014, 56, 153170. Available from: http://dx.doi.org/10.1016/j. cemconres.2013.11.015. Galan, I.; Andrade, C.; Castellote, M. Thermogravimetrical Analysis for Monitoring Carbonation of Cementitious Materials Uptake of CO2 and Deepening in C-S-H Knowledge. J. Ther. Anal. Calorim. 2012, 110 (1), 309319. Available from: http://dx. doi.org/10.1007/s10973-012-2466-4. Borges, P. H. R.; Costa, J. O.; Milestone, N. B.; Lynsdale, C. J.; Streatfield, R. E. Carbonation of CH and C-S-H in Composite Cement Pastes Containing High Amounts of BFS. Cement Concrete Res. 2010, 40 (2), 284292. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2009.10.020. Black, L.; Garbev, K.; Gee, I. Surface Carbonation of Synthetic C-S-H Samples: A Comparison between Fresh and Aged C-S-H Using X-ray Photoelectron Spectroscopy. Cement Concrete Res. 2008, 38 (6), 745750. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2008.02.003. Black, L.; Breen, C.; Yarwood, J.; Garbev, K.; Stemmermann, P.; Gasharova, B. Structural Features of C-S-H(I) and Its Carbonation in Air  A Raman Spectroscopic Study. Part II: Carbonated Phases. J. Am. Ceram. Soc. 2007, 90 (3), 908917. Available from: http://dx.doi.org/10.1111/j.1551-2916.2006.01429.x. Bertos, M. F.; Simons, S. J.; Hills, C. D.; Carey, P. J. A Review of Accelerated Carbonation Technology in the Treatment of Cement-based Materials and Sequestration of CO2. J. Hazard. Mater. 2004, 112 (3), 193205. Available from: http://dx.doi.org/ 10.1016/j.jhazmat.2004.04.019. Papadakis, V. G.; Fardis, M. N.; Vayenas, C. G. Effect of Composition, Environmental Factors and Cement-lime Mortar Coating on Concrete Carbonation. Mater. Struct. 1992, 25 (5), 293304. Available from: http://dx.doi.org/10.1007/BF02472670.

Accelerated carbonation as a fast curing technology for concrete blocks

335

ˇ 45. Savija, B.; Lukovi´c, M.; Pacheco, J.; Schlangen, E. Cracking of the Concrete Cover Due to Reinforcement Corrosion: A Two-dimensional Lattice Model Study. Constr. Build. Mater. 2013, 44, 626638. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.03.063. 46. Johannesson, B.; Utgenannt, P. Microstructural Changes Caused by Carbonation of Cement Mortar. Cement Concrete Res. 2001, 31 (6), 925931. Available from: http:// dx.doi.org/10.1016/S0008-8846(01)00498-7. 47. Bukowski, J. M.; Berger, R. L. Reactivity and Strength Development of CO2 Activated Non-hydraulic Calcium Silicates. Cement Concrete Res. 1979, 9 (1), 5768. Available from: http://dx.doi.org/10.1016/0008-8846(79)90095-4. 48. Castellote, M.; Fernandez, L.; Andrade, C.; Alonso, C. Chemical Changes and Phase Analysis of OPC Pastes Carbonated at Different CO2 Concentrations. Mater. Struct. 2009, 42 (4), 515525. Available from: http://dx.doi.org/10.1617/s11527-008-9399-1. 49. Goodbrake, C. J.; Young, J. F.; Berger, R. L. Reaction of Hydraulic Calcium Silicates with Carbon-Dioxide and Water. J. Am. Ceram. Soc. 1979, 62 (910), 488491. Available from: http://dx.doi.org/10.1111/j.1151-2916.1979.tb19112.x. 50. Toennies, H. T. Artificial Carbonation of Concrete Masonry Units. J. Am. Concrete Inst. 1960, 31 (8), 737755. 51. Berger, R. L.; Young, J. F.; Leung, K. Acceleration of Hydration of Calcium Silicates by Carbon-dioxide Treatment. Nat. Phys. Sci. 1972, 240 (97), 1618. Available from: http://dx.doi.org/10.1038/physci240016a0. ˇ 52. Sauman, Z. Carbonization of Porous Concrete and Its Main Binding Components. Cement Concrete Res. 1971, 1 (6), 645662. Available from: http://dx.doi.org/10.1016/ 0008-8846(71)90019-6. 53. Hermawan, D.; Hata, T.; Umemura, K.; Kawai, S.; Nagadomi, W.; Kuroki, Y. Rapid Production of High-strength Cement-bonded Particleboard Using Gaseous or Supercritical Carbon Dioxide. J. Wood Sci. 2001, 47 (4), 294300. Available from: http://dx.doi.org/10.1007/Bf00766716. 54. Hermawan, D.; Hata, T.; Umemura, K.; Kawai, S.; Kaneko, S.; Kuroki, Y. New Technology for Manufacturing High-strength Cement-bonded Particleboard Using Supercritical Carbon Dioxide. J. Wood Sci. 2000, 46 (1), 8588. Available from: http:// dx.doi.org/10.1007/Bf00779559. 55. Simatupang, M. H.; Habighorst, C.; Lange, H.; Neubauer, A. Investigation on the Influence of the Addition of Carbon Dioxide on the Production and Properties of Rapidly Set Wood-cement Composites. Cement Concrete Compos. 1995, 17 (3), 187197. Available from: http://dx.doi.org/10.1016/0958-9465(95)00009-2. 56. Farahi, E.; Purnell, P.; Short, N. R. Supercritical Carbonation of Calcareous Composites: Influence of Curing. Cement Concrete Compos. 2013, 43, 4853. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2013.06.008. 57. Xuan, D. X.; Zhan, B. J.; Poon, C. S. Development of a New Generation of Eco-friendly Concrete Blocks by Accelerated Mineral Carbonation. J. Clean. Prod. 2016, 133, 12351241. Available from: http://dx.doi.org/10.1016/j.jclepro.2016.06.062. 58. Shao, Y. X.; Rostami, V. Z.; He Boyd, A. J. Accelerated Carbonation of Portland Limestone Cement. J. Mater. Civ. Eng. 2013, 26 (1), 117124. Available from: http:// dx.doi.org/10.1061/(ASCE)MT.1943-5533.0000773. 59. Kou, S. C.; Zhan, B. J.; Poon, C. S. Use of a CO2 Curing Step to Improve the Properties of Concrete Prepared with Recycled Aggregates. Cement Concrete Compos. 2014, 45, 2228. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2013.09.008. 60. Zhan, B. J.; Poon, C. S.; Shi, C. J. CO2 Curing for Improving the Properties of Concrete Blocks Containing Recycled Aggregates. Cement Concrete Compos. 2013, 42, 18. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.09.008.

336 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

61. Zhang, J. K.; Shi, C. J.; Li, Y. K.; Pan, X. Y.; Poon, C. S.; Xie, Z. B. Influence of Carbonated Recycled Concrete Aggregate on Properties of Cement Mortar. Constr. Build. Mater. 2015, 98, 17. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.08.087. 62. Zhan, B. J.; Poon, C. S.; Liu, Q.; Kou, S. C.; Shi, C. J. Experimental Study on CO2 Curing for Enhancement of Recycled Aggregate Properties. Constr. Build. Mater. 2014, 67, 37. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.09.008. 63. Xuan, D. X.; Zhan, B. J.; Poon, C. S.; Zheng, W. Innovative Reuse of Concrete Slurry Waste from Ready-mixed Concrete Plants in Construction Products. J. Hazard. Mater. 2016, 312, 6572. Available from: http://dx.doi.org/10.1016/j.jhazmat.2016.03.036. 64. Slegers, P. A.; Rouxhet, P. G. Carbonation of the Hydration Products of Tricalcium Silicate. Cement Concrete Res. 1976, 6 (3), 381388. Available from: http://dx.doi.org/ 10.1016/0008-8846(76)90101-0. 65. Liu, M.; Wu, H. Z.; Chang, J.; Ding, L. Carbonation of Cement Minerals and Clinker. J. Univ. Jinan (Sci. Technol.) 2010, 2, 003 1671-3559(2010)02-0115-04. 66. Shen, Q. Z.; Pan, G. H.; Bao, B. F. Influence of CSH Carbonation on the Porosity of Cement Paste. Mag. Concrete Res. 2015, 111. Available from: http://dx.doi.org/ 10.1680/jmacr.15.00286. 67. Higuchi, T.; Morioka, M.; Yoshioka, I.; Yokozeki, K. Development of a New Ecological Concrete with CO2 Emissions Below Zero. Constr. Build. Mater. 2014, 67, 338343. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.01.029. 68. Huntzinger, D. N.; Gierke, J. S.; Kawatra, S. K.; Eisele, T. C.; Sutter, L. L. Carbon Dioxide Sequestration in Cement Kiln Dust Through Mineral Carbonation. Environ. Sci. Technol. 2009, 43 (6), 19861992. Available from: http://dx.doi.org/10.1021/es802910z. 69. Eloneva, S.; Teir, S.; Salminen, J.; Fogelholm, C.; Zevenhoven, R. Fixation of CO2 by Carbonating Calcium Derived from Blast Furnace Slag. Energy 2008, 33 (9), 14611467. Available from: http://dx.doi.org/10.1016/j.energy.2008.05.003. 70. Eloneva, S.; Puheloinen, E.; Kanerva, J.; Ekroos, A.; Zevenhoven, R.; Fogelholm, C.-J. Co-utilisation of CO2 and Steelmaking Slags for Production of Pure CaCO3  Legislative Issuees. J. Clean. Prod. 2010, 18 (18), 18331839. Available from: http:// dx.doi.org/10.1016/j.jclepro.2010.07.026. 71. Morandeau, A.; Thi ry, M.; Dangla, P. Impact of Accelerated Carbonation on OPC Cement Paste Blended with Fly Ash. Cement Concrete Res. 2015, 67, 226236. Available from: http://dx.doi.org/10.1016/j.cemconres.2014.10.003. 72. Ati¸s, C. D. Accelerated Carbonation and Testing of Concrete Made With Fly Ash. Constr. Build. Mater. 2003, 17 (3), 147152. Available from: http://dx.doi.org/10.1016/ S0950-0618(02)00116-2. 73. Dong, B. Q.; Qiu, Q. W.; Gu, Z. T.; Xiang, J. Q.; Huang, C. J.; Fang, Y., et al. Characterization of Carbonation Behavior of Fly Ash Blended Cement Materials by the Electrochemical Impedance Spectroscopy Method. Cement Concrete Compos. 2015, 65, 118127. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2015.10.006. 74. Ko, M. S.; Chen, Y. L.; Jiang, J. H. Accelerated Carbonation of Basic Oxygen Furnace Slag and the Effects on Its Mechanical Properties. Constr. Build. Mater. 2015, 98, 286293. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.08.051. 75. Ghouleh, Z.; Guthrie, R. I. L.; Shao, Y. X. High-strength KOBM Steel Slag Binder Activated by Carbonation. Constr. Build. Mater. 2015, 99, 175183. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.09.028. 76. Tu, Z. J.; Guo, M. Z.; Poon, C. S.; Shi, C. J. Effects of Limestone Powder on CaCO3 Precipitation in CO2 Cured Cement Pastes. Cement Concrete Compos. 2016, 72, 916. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2016.05.019.

Accelerated carbonation as a fast curing technology for concrete blocks

337

77. Stockmann, G. J.; Wolff-Boenisch, D.; Bovet, N.; Gislason, S. R.; Oelkers, E. H. The Role of Silicate Surfaces on Calcite Precipitation Kinetics. Geochim. Cosmochim. Acta 2014, 135, 231250. Available from: http://dx.doi.org/10.1016/j.gca.2014.03.015. 78. Fernandez-Martinez, A.; Hu, Y.; Lee, B.; Jun, Y.; Waychunas, G. A. In situ Determination of Interfacial Energies between Heterogeneously Nucleated CaCO3 and Quartz Substrates: Thermodynamics of CO2 Mineral Trapping. Environ. Sci. Technol. 2012, 47 (1), 102109. Available from: http://dx.doi.org/10.1021/es3014826. 79. Lioliou, M. G.; Paraskeva, C. A.; Koutsoukos, P. G.; Payatakes, A. C. Heterogeneous Nucleation and Growth of Calcium Carbonate on Calcite and Quartz. J. Colloid Interface Sci. 2007, 308 (2), 421428. Available from: http://dx.doi.org/10.1016/j.jcis.2006.12.045. 80. Lin, Y. P.; Singer, P. C. Effects of Seed Material and Solution Composition on Calcite Precipitation. Geochim. Cosmochim. Acta 2005, 69 (18), 44954504. Available from: http://dx.doi.org/10.1016/j.gca.2005.06.002. 81. Chang, J.; Li, Y.; Cao, M. L.; Fang, Y. F. Influence of Magnesium Hydroxide Content and Fineness on the Carbonation of Calcium Hydroxide. Constr. Build. Mater. 2014, 55, 8288. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.12.099. 82. Cui, H. Z.; Tang, W. C.; Liu, W.; Dong, Z. J.; Xing, F. Experimental Study on Effects of CO2 Concentrations on Concrete Carbonation and Diffusion Mechanisms. Constr. Build. Mater. 2015, 93, 522527. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2015.06.007. 83. Zou, Q. Y.; Shi, C. J.; Zheng, K. R.; He, F. Q. Effect of Pre-conditioning on CO2 Curing of Block Concretes. J. Chin. Ceram. Soc. 2008, 11 (1), 116120. Available from: http://dx.doi.org/1007-9629(2008)01-0116-05. 84. Han, D. R.; Namkung, H.; Lee, H. M.; Huh, D. G.; Kim, H. T. CO2 Sequestration by Aqueous Mineral Carbonation of Limestone in a Supercritical Reactor. J. Ind. Eng. Chem. 2015, 21, 792796. Available from: http://dx.doi.org/10.1016/j.jiec.2014.04.014. 85. Houst, Y. F.; Wittmann, F. H. Depth Profiles of Carbonates Formed During Natural Carbonation. Cement Concrete Res. 2002, 32 (12), 19231930. Available from: http:// dx.doi.org/10.1016/S0008-8846(02)00908-0. 86. Galan, I.; Andrade, C.; Mora, P.; Sanjuan, M. A. Sequestration of CO2 by Concrete Carbonation. Environ. Sci. Technol. 2010, 44 (8), 31813186. Available from: http://dx. doi.org/10.1021/es903581d. 87. Mo, L.; Zhang, F.; Deng, M. Effects of Carbonation Treatment on the Properties of Hydrated Fly ash-MgO-Portland Cement Blends. Constr. Build. Mater. 2015, 96, 147154. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.07.193. 88. Zha, X. X.; Yu, M.; Ye, J.; Feng, G. L. Numerical Modeling of Supercritical Carbonation Process in Cement-based Materials. Cement Concrete Res. 2015, 72, 1020. Available from: http://dx.doi.org/10.1016/j.cemconres.2015.02.017. 89. Phung, Q. T.; Maes, N.; Jacques, D.; Bruneel, E.; Driessche, I. V.; Ye, G., et al. Effect of Limestone Fillers on Microstructure and Permeability Due to Carbonation of Cement Pastes Under Controlled CO2 Pressure Conditions. Constr. Build. Mater. 2015, 82, 376390. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2015.02.093. 90. Phung, Q. T.; Maes, N.; Jacques, D.; De Schutter, G.; Ye, G.; Perko, J. Modelling the Carbonation of Cement Pastes Under a CO2 Pressure Gradient Considering Both Diffusive and Convective Transport. Constr. Build. Mater. 2016, 114, 333351. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2016.03.191. 91. Garcia-Gonzalez, C. A.; Grouh, N.; Hidalgo, A.; Fraile, J.; Lopez-Periago, A. M.; Andrade, C., et al. New Insights on the Use of Supercritical Carbon Dioxide for the Accelerated Carbonation of Cement Pastes. J Supercrit. Fluids 2008, 43 (3), 500509. Available from: http://dx.doi.org/10.1016/j.supflu.2007.07.018.

338 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

92. Rostami, V.; Shao, Y. X.; Boyd, A. J. Durability of Concrete Pipes Subjected to Combined Steam and Carbonation Curing. Constr. Build. Mater. 2011, 25 (8), 33453355. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2011.03.025. 93. Junior, A. N.; Filho, R. D. T.; Fairbairn, E. D. M. R.; Dweck, J. The Effects of the Early Carbonation Curing on the Mechanical and Porosity Properties of High Initial Strength Portland Cement Pastes. Constr. Build. Mater. 2015, 77, 448454. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.12.072. 94. Morshed, A. Z.; Shao, Y. X. Influence of Moisture Content on CO2 Uptake in Lightweight Concrete Subject to Early Carbonation. J. Sustain. CementBased Mater. 2013, 2 (2), 144160. Available from: http://dx.doi.org/10.1080/ 21650373.2013.797373. 95. Bonavetti, V.; Donza, H.; Menendez, G.; Cabrera, O.; Irassar, E. F. Limestone Filler Cement in Low w/c Concrete: A Rational Use of Energy. Cement Concrete Res. 2003, 33 (6), 865871. Available from: http://dx.doi.org/10.1016/S0008-8846(02)01087-6. 96. Mounanga, P.; Khokhar, M. I. A.; El Hachem, R.; Loukili, A. Improvement of the Early-age Reactivity of Fly Ash and Blast Furnace Slag Cementitious Systems Using Limestone Filler. Mater. Struct. 2011, 44 (2), 437453. Available from: http://dx.doi. org/10.1617/s11527-010-9637-1. 97. Sevelsted, T. F.; Herfort, D.; Skibsted, J. 13C Chemical Shift Anisotropies for Carbonate Ions in Cement Minerals and the Use of 13C, 27Al and 29Si MAS NMR in Studies of Portland Cement Including Limestone Additions. Cement Concrete Res. 2013, 52, 100111. Available from: http://dx.doi.org/10.1016/j. cemconres.2013.05.010. 98. Shi, Z.; Lothenbach, B.; Geiker, M. R.; Kaufmann, J.; Leemann, A.; Ferreiro, S., et al. Experimental Studies and Thermodynamic Modeling of the Carbonation of Portland Cement, Metakaolin and Limestone Mortars. Cement Concrete Res. 2016, 88, 6072. Available from: http://dx.doi.org/10.1016/j.cemconres.2016.06.006. 99. Tydlit, T. V.; Matas, T.; Cerny, R. Effect of w/c and Temperature on the Earlystage Hydration Heat Development in Portland-limestone Cement. Constr. Build. Mater. 2014, 50, 140147. Available from: http://dx.doi.org/10.1016/j. conbuildmat.2013.09.020. 100. Kakali, G.; Tsivilis, S.; Aggeli, E.; Bati, M. Hydration Products of C3A, C3S and Portland Cement in the Presence of CaCO3. Cement Concrete Res. 2000, 30 (7), 10731077. Available from: http://dx.doi.org/10.1016/S0008-8846(00)00292-1. 101. Lothenbach, B.; Saout, G. L.; Gallucci, E.; Scrivener, K. Influence of Limestone on the Hydration of Portland Cements. Cement Concrete Res. 2008, 38 (6), 848860. Available from: http://dx.doi.org/10.1016/j.cemconres.2008.01.002. 102. Zhang, Y. J.; Zhang, X. Research on Effect of Limestone and Gypsum on C3A, C3S and PC Clinker System. Constr. Build. Mater. 2008, 22 (8), 16341642. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2007.06.013. 103. Han, J. D.; Pan, G. H.; Sun, W.; Wang, C. H.; Cui, D. Application of Nanoindentation to Investigate Chemomechanical Properties Change of Cement Paste in the Carbonation Reaction. Sci. China Technol. Sci. 2012, 55 (3), 616622. Available from: http://dx.doi.org/10.1007/s11431-011-4571-1. 104. Shi, C. J.; Zou, Q. Y.; He, F. Q. Study on CO2 Curing Kinetics of Concrete. J. Chin. Ceram. Soc. 2010, 7, 11791184. Available from: http://dx.doi.org/10.14062/j. issn.0454-5648.2010.07.031.

Accelerated carbonation as a fast curing technology for concrete blocks

339

105. Shao, Y. X.; Monkman, S.; Tran, S. CO2 Uptake Capacity of Concrete Primary Ingredients. J. Chin. Ceram. Soc. 2010, 9, 008. Available from: http://dx.doi.org/ 10.14062/j.issn.0454-5648.2010.09.018. 106. El-Hassan, H.; Shao, Y. X. Carbon Storage Through Concrete Block Carbonation Curing. J. Clean Energy Technol. 2014, 2 (3), 287291. Available from: http://dx.doi. org/10.7763/JOCET.2014.V2.141. 107. Mohammed, M. K.; Dawson, A. R. I.; Thom, N. H. Carbonation of Filler Typed Selfcompacting Concrete and Its Impact on the Microstructure by Utilization of 100% CO2 Accelerating Techniques. Constr. Build. Mater. 2014, 50, 508516. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2013.09.052. 108. Mo, L. W.; Zhang, F.; Deng, M.; Panesar, D. K. Effectiveness of Using CO2 Pressure to Enhance the Carbonation of Portland Cement-fly ash-MgO Mortars. Cement Concrete Compos. 2016, 70, 7885. Available from: http://dx.doi.org/10.1016/j. cemconcomp.2016.03.013. 109. Thiery, M. L.; Dangla, P.; Belin, P.; Habert, G.; Roussel, N. Carbonation Kinetics of a Bed of Recycled Concrete Aggregates: A Laboratory Study on Model Materials. Cement Concrete Res. 2013, 46, 5065. Available from: http://dx.doi.org/10.1016/j. cemconres.2013.01.005. ˇ 110. Savija, B.; Lukovi´c, M. Carbonation of Cement Paste: Understanding, Challenges; Opportunities. Constr. Build. Mater. 2016, 117, 285301. Available from: http://dx. doi.org/10.1016/j.conbuildmat.2016.04.138. 111. Hidalgo, A.; Domingo, C.; Garcia, C.; Petit, S.; Andrade, C.; Alonso, C. Microstructural Changes Induced in Portland Cement-based Materials Due to Natural and Supercritical Carbonation. J. Mater. Sci. 2008, 43 (9), 31013111. Available from: http://dx.doi.org/10.1007/s10853-008-2521-5. 112. Fang, Y. F.; Chang, J. Microstructure Changes of Waste Hydrated Cement Paste Induced by Accelerated Carbonation. Constr. Build. Mater. 2015, 76, 360365. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.12.017. 113. Groves, G. W.; Brough, A.; Richardson, I. G.; Dobson, C. M. Progressive Changes in the Structure of Hardened C3S Cement Pastes Due to Carbonation. J. Am. Ceram. Soc. 1991, 74 (11), 28912896. Available from: http://dx.doi.org/10.1111/j.11512916.1991.tb06859.x. 114. Auroy, M.; Poyet, S.; Bescop, P. L.; Torrenti, J. M.; Charpentier, T.; Moskura, M. Bourbon, X. Impact of Carbonation on Unsaturated Water Transport Properties of Cement-based Materials. Cement Concrete Res. 2015, 74, 4458. Available from: http://dx.doi.org/10.1016/j.cemconres.2015.04.002. 115. Chang, J.; Fang, Y. F. Quantitative Analysis of Accelerated Carbonation Products of the Synthetic Calcium Silicate Hydrate (CSH) by QXRD and TG/MS. J. Ther. Anal. Calorim. 2015, 119 (1), 5762. Available from: http://dx.doi.org/10.1007/ s10973-014-4093-8. 116. Morandeau, A. E.; White, C. E. In situ X-ray Pair Distribution Function Analysis of Accelerated Carbonation of a Synthetic Calciumsilicatehydrate Gel. J. Mater. Chem. A 2015, 3 (16), 85978605. Available from: http://dx.doi.org/10.1039/C5TA00348B. 117. Goni, S.; Gazta Aga, M. T.; Guerrero, A. Role of Cement Type on Carbonation Attack. J. Mater. Res. 2002, 17 (07), 18341842. Available from: http://dx.doi.org/10.1557/ JMR.2002.0271. 118. Groves, G. W.; Rodway, D. I.; Richardson, I. G. The Carbonation of Hardened Cement Pastes. Adv. Cement Res. 1990, 3 (11), 117125. Available from: http://dx.doi.org/ 10.1680/adcr.1990.3.11.117.

340 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

119. Garbev, K.; Stemmermann, P.; Black, L. Structural Features of CSH (I) and Its Carbonation in Air—A Raman Spectroscopic Study. Part I: Fresh Phases. J. Am. Ceram. Soc. 2007, 90 (6), 900907. Available from: http://dx.doi.org/10.1111/j.15512916.2006.01428.x. 120. Hyvert, N.; Sellier, A.; Duprat, F.; Rougeau, P.; Francisco, P. Dependency of CSH Carbonation Rate on CO2 Pressure to Explain Transition from Accelerated Tests to Natural Carbonation. Cement Concrete Res. 2010, 40 (11), 15821589. Available from: http://dx.doi.org/10.1016/j.cemconres.2010.06.010. 121. Fernandez-Carrasco, L.; Torrens-Martin, D.; Martinez-Ramirez, S. Carbonation of Ternary Building Cementing Materials. Cement Concrete Compos. 2012, 34 (10), 11801186. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2012.06.016. 122. Sawada, K. The Mechanism of Crystallization and Transformation of Calcium Carbonates. Pure Appl. Chem. 1997, 69 (5), 921928. Available from: http://dx.doi. org/10.1351/pac199769050921. 123. Sanna, A.; Uibu, M.; Caramanna, G.; Kuusik, R.; Maroto-Valer, M. M. A Review of Mineral Carbonation Technologies to Sequester CO2. Chem. Soc. Rev. 2014, 43 (23), 80498080. Available from: http://dx.doi.org/10.1039/c4cs00035h. 124. Goto, S.; Suenaga, K.; Kado, T.; Fukuhara, M. Calcium Silicate Carbonation Products. J. Am. Ceram. Soc. 1995, 78 (11), 28672872. Available from: http://dx.doi.org/ 10.1111/j.1151-2916.1995.tb09057.x. 125. Song, K.; Kim, W.; Bang, J. H.; Park, S.; Jeon, C. W. Polymorphs of Pure Calcium Carbonate Prepared by the Mineral Carbonation of Flue Gas Desulfurization Gypsum. Mater. Des. 2015, 83, 308313. Available from: http://dx.doi.org/10.1016/j. matdes.2015.06.051. 126. Shtepenko, O.; Hills, C.; Brough, A.; Thomas, M. The Effect of Carbon Dioxide on β-dicalcium Silicate and Portland Cement. Chem. Eng. J. 2006, 118 (1), 107118. Available from: http://dx.doi.org/10.1016/j.cej.2006.02.005. 127. Villain, G.; Thiery, M.; Platret, G. Measurement Methods of Carbonation Profiles in Concrete: Thermogravimetry, Chemical Analysis and Gammadensimetry. Cement Concrete Res. 2007, 37, 11821192. Available from: http://dx.doi.org/10.1007/s11367015-0988-2. 128. El-Hassan, H.; Shao, Y. X.; Ghouleh, Z. Effect of Initial Curing on Carbonation of Lightweight Concrete Masonry Units. ACI Mater. J. 2013, 110 (4), 441450.. Available from: http://www.concrete.org/P....aspx?m 5 details&i 5 51685791. 129. Galan, I.; Andrade, C.; Castellote, M. Natural and Accelerated CO2 Binding Kinetics in Cement Paste at Different Relative Humidities. Cement Concrete Res. 2013, 49, 2128. Available from: http://dx.doi.org/10.1016/j.cemconres.2013.03.009. 130. Ramezanianpour, A. A.; Ghahari, S. A.; Esmaeili, M. Effect of Combined Carbonation and Chloride Ion Ingress by an Accelerated Test Method on Microscopic and Mechanical Properties of Concrete. Constr. Build. Mater. 2014, 58, 138146. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2014.01.102. 131. Chen, J. J.; Thomas, J. J.; Jennings, H. M. Decalcification Shrinkage Of Cement Paste. Cement Concrete Res. 2006, 36 (5), 801809. Available from: http://dx.doi.org/ 10.1016/j.cemconres.2005.11.003. 132. De Juan, M. S.; Gutierrez, P. A. Study on the Influence of Attached Mortar Content on the Properties of Recycled Concrete Aggregate. Constr. Build. Mater. 2009, 23 (2), 872877. Available from: http://dx.doi.org/10.1016/j.conbuildmat.2008.04.012.

Accelerated carbonation as a fast curing technology for concrete blocks

341

133. Rostami, V.; Shao, Y. X.; Boyd, A. J. Carbonation Curing versus Steam Curing for Precast Concrete Production. J. Mater. Civ. Eng. 2012, 24, 12211229. Available from: http://dx.doi.org/10.1061/(ASCE)MT.1943-5533.0000462. 134. Hanehara, S.; Tomosawa, F.; Kobayakawa, M.; Hwang, K. Effects of Water/powder Ratio, Mixing Ratio of Fly Ash; Curing Temperature on Pozzolanic Reaction of Fly Ash in Cement Paste. Cement Concrete Res. 2001, 31 (1), 3139. Available from: http://dx.doi.org/10.1016/S0008-8846(00)00441-5. 135. Santos, S. F.; Schmidt, R.; Almeida, A. E. F. S.; Tonoli, G. H. D.; Savastano, H., Jr. Supercritical Carbonation Treatment on Extruded Fibrecement Reinforced with Vegetable Fibres. Cement Concrete Compos. 2015, 56, 8494. Available from: http:// dx.doi.org/10.1016/j.cemconcomp.2014.11.007. 136. Farahi, E.; Purnell, P.; Short, N. R. Supercritical Carbonation of Calcareous Composites: Influence of Mix Design. Cement Concrete Compos. 2013, 43, 1219. Available from: http://dx.doi.org/10.1016/j.cemconcomp.2013.06.004. 137. Seneviratne, A. M. G.; Short, N. R.; Purnell, P.; Page, C. L. Preliminary Investigations of the Dimensional Stability of Super-critically Carbonated Glass Fibre Reinforced Cement. Cement Concrete Res. 2002, 32 (10), 16391644. Available from: http://dx. doi.org/10.1016/S0008-8846(02)00837-2. 138. Domingo, C.; Loste, E.; Gomez-Morales, J.; Garcia-Carmona, J.; Fraile, J. Calcite Precipitation by a High-pressure CO2 Carbonation Route. J. Supercrit. Fluids 2006, 36 (3), 202215. Available from: http://dx.doi.org/10.1016/j.supflu.2005.06.006. 139. Kwan, A. K. H.; Mckinley, M.; Chen, J. J. Adding Limestone Fines as Cement Paste Replacement to Reduce Shrinkage of Concrete. Mag. Concrete Res. 2013, 65 (15), 942950. Available from: http://dx.doi.org/10.1680/macr.13.00028. 140. Marzouki, A.; Lecomte, A.; Beddey, A.; Diliberto, C.; Ben Ouezdou, M. The Effects of Grinding on the Properties of Portland-Limestone Cement. Constr. Build. Mater. 2013, 48, 11451155. Available from: http://dx.doi.org/10.1016/j.conbuildmat. 2013.07.053.

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

13

Shama Parveen, Sohel Rana and Raul Fangueiro University of Minho, Guimara˜es, Portugal

13.1

Introduction

The civil construction industry is majorly dominated by cement as the binding material. Cementitious materials or composites have good compressive properties but they lack tensile strength. Steel reinforcing bars and synthetic fibers, such as glass, carbon, poly vinyl alcohol, or aramid fibers, are used to overcome the disadvantages of cementitious composites and also to achieve long postcrack behavior. The use of steel reinforcement makes cementitious materials more susceptible to environmental attacks causing corrosion. Steel or synthetic fiber reinforcement both have environmental impacts caused by its production process, which includes high energy consuming steps and leads to the release of harmful chemicals and gases to the atmosphere. Since 1970, the civil construction sector has been focusing on the sustainability of structures. Plant fibers are the most promising solution towards sustainability due to their eco-friendly nature along with good reinforcing performance. Therefore, over the last three decades researchers and scientists have focused increasingly on plant-based plant fiber reinforcements. Plant fibers are used in cementitious composites to increase the postcracking ductility, toughness and fracture energy.1 Plant fibers have many advantageous properties such as good tensile and flexural modulus, low density, low coefficient of thermal expansion, and so on. One of the major problems of plant fibers is the degradation and deterioration of their properties with time. Therefore, for successful application of plant fibers in cementitious matrix, the influence of different degradation parameters should also be thoroughly studied to understand and improve the durability of plant fibers in highly alkaline cementitious composites.1 Different chemical or physical modifications could also be performed on the surface of plant fibers to improve their durability.1 Plant fiber derivatives such as microcrystalline cellulose (MCC), microfibrillar cellulose (MFC), nanocrystalline cellulose (NCC), etc. and bacterial cellulose (BC) are now being considered as good options for the reinforcement of cementitious matrix. Their extraordinary mechanical, thermal, and optical properties can make them a potential substitute for steel and synthetic fibers.2 Owing to their higher crystallinity, MCC, NCC, and BC are considered to be much more stable towards degradation than plant fibers because of their highly crystalline structures and, Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00020-6 © 2017 Elsevier Ltd. All rights reserved.

344 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

therefore, their utilization within cementitious composites can provide superior durability as compared to ordinary plant fiber-reinforced cementitious composites. The application of nanocellulose in polymer composites has been included in this chapter just to provide the readers with sufficient background information and the techniques that can be used in cement-based composites.

13.2

Properties of portland cement

Concrete generally consists of Ordinary Portland Cement (OPC), which is known as the principal hydraulic binding agent), coarse aggregates, and fillers such as sand, admixtures, and water.3,4 The dry portion of Portland cement is composed of 63% calcium oxide, 20% silica, 6% alumina, 3% iron (III) oxide, and small amount of other materials including some impurities. These materials react with water with an exothermic reaction forming a mineral glue (known as “C-S-H” gel), calcium hydroxide, ettringite, monosulfate, unhydrated particles, and air voids. The molecular structure of C-S-H gel was not fully understood until recently. Researchers at the Massachusetts Institute of Technology (MIT, USA)5 recently proposed a structure, and according to that, cement hydrate consists of a long tetrahedral silica chain and calcium oxide in long range distances, where water causes an intralayer distortion in otherwise regular geometry (Fig. 13.1). The distortion in the structure due to the addition of water makes the cement hydrate robust.

Figure 13.1 The molecular model of C-S-H: the blue and white spheres are the oxygen and hydrogen atoms of water molecules, respectively; the green and gray spheres are inter- and intralayer calcium ions, respectively; the yellow and red sticks are silicon and oxygen atoms in silica tetrahedral. Source: Pellenq et al.5

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

345

Figure 13.2 Growth in plant fiber composites market. From: Lucintel, http://www.lucintel.com/lucintelbrief/potentialofnaturalfibercomposites-final. pdf

13.3

Properties of plant fibers and nanocellulose

In recent times, tremendous interest has been paid to various plant fibers (such as sisal, jute, flax, hemp, coir, etc.) both by the scientific community and industrial sectors for various applications including civil construction, automobiles, sports, aerospace, and geotechnical engineering. The global plant fiber composites market reached 1.6 billion euros in 2010, with a compound annual growth rate of 15% over the last five years.6 By 2016, plant fiber composite market is expected to reach 2.8 billion euros with a growth rate of 10% (Fig. 13.2).

13.3.1 Properties of plant fibers Plant fibers are low cost, light weight, nonhazardous, eco-friendly, and renewable materials possessing high specific mechanical properties and require lower energy during their growth and applications.7 Due to their lower carbon footprint and environmental benefits, plant fiber-based products are considered to possess a lower carbon footprint and higher sustainability.813 Table 13.1 lists the physical and mechanical properties of various plant fibers and compares them with the commonly used synthetic fibers.7,14 Among these fibers, cotton is known to be the most popular fiber for apparel sectors because of their comfort properties. Other fibers like sisal, jute, flax, coir, etc. have been mainly used for various technical applications. Flax fibers show the best mechanical properties among the various plant fibers. In comparison with the synthetic fibers, plant fibers present much lower mechanical properties, as can be seen from Table 13.1. However, owing to their much lower density as compared to synthetic fibers, they present very good specific mechanical properties (Fig. 13.3) and therefore, are of tremendous interest in applications demanding light weight.15 Therefore, these materials have huge potential to

346 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Table 13.1

Properties of selected plant and synthetic fibers

Type of fiber

Density (g/cm3)

Tensile strength (MPa)

Elastic modulus (GPa)

Elongation (%)

Jute Flax Hemp Ramie Sisal Coir Cotton E-glass S-glass Aramid Carbon

1.31.45 1.5 1.48 1.51 1.45 1.15 1.51.6 2.5 2.5 1.4 1.7

393773 3451100 514 400938 468640 131175 287800 20003500 4570 30003150 4000

1326.5 27.6 24.8 61.4128 9.422 46 5.512.6 70 86 6367 230240

78 2.73.2 1.6 1.23.8 37 1540 78 2.5 2.8 3.33.7 1.41.8

Source: Fangueiro, R.; Rana, S. (Eds.), Natural Fibres: Advances in Science and Technology Towards Industrial Applications. Springer, ISBN: 978-94-017-7515-1.

Figure 13.3 Comparison of specific mechanical properties of natural fibers with synthetic fibers. Source: Rana and Fangueiro.15

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

347

reduce the consumption of nonrenewable, nonenvironmentally friendly, and energyconsuming materials, such as concrete, metals, or synthetic fibers, in the above applications.

13.3.2 Properties of micro- and nanocellulose Cellulose is an abundant biopolymer, and with the progress in nanotechnology the nano form of cellulose, i.e., nanocellulose has gained tremendous attention. The terminology MFC was first coined in the early 1980s when ITT Rayonier issued patents and publications based on a totally new nanocellulose composition.16,17 In later years MFC was modified by acid hydrolysis to obtain NCC. Nanocellulose has now become a good alternative for other nanomaterials in various applications due to its remarkable mechanical properties, transparency, ability to form chiral nematic structures, and above all, owing to its lower health risk, environmental friendliness, and biodegradability.18,19 Researchers are working with nanocellulose in diverse fields. It can act as a reinforcing agent for various matrices because of excellent mechanical properties, as well as due to the presence of free hydroxyl groups which can be modified according to the needs. Nanocellulose is also being explored in the biomedical field for drug delivery, enzyme immobilization, tissue culture, etc. Because of its transparency and barrier properties it can be utilized in packaging and as transparent flexible films.18,19 Nanocellulose can be obtained by mainly two different approaches: a top-down approach and a bottom-up approach, as presented in Fig. 13.4. Nano cellulose

Top down approach

Bottom up approach

Acetobacter xylinum + saccharides (iguchi)

Mechanical defibrillation of cellulose fibers

Selective hydrolysis of cellulose fibers

Micro fibrillar cellulose (MFC)

Colloidal cellulose

Chemical modification / enzymatic modification + mechanical treatment

Micro crystalline cellulose (MFC)

Continuous gel of bacterial cellulose

Nano fibrillar cellulose (NFC)

Nano crystalline cellulose (NCC)

Bacterial cellulose (BC)

Figure 13.4 Different approaches for production of micro- and nanocellulose.

348 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Comparison of properties of nanocellulose with other high strength materials

Table 13.2

Material

Density (g/cm3)

CTE (1026/K) axial

Tensile strength (GPa) axial

Elastic modulus (GPa) Axial

Crystalline cellulose Kevlar-49 Fiber Clay nanoplatelets Carbon nanotubes Boron nanowhiskers

Transverse

1.6

0.1

7.5

120220 1157

1.4 

2 

3.5 

124130 2.5 170 





1163

270950 0.830



6

28

250360 

Source: Moon, R.J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Cellulose Nanomaterials Review: Structure, Properties and Nanocomposites. Chem. Soc. Rev. 2011, 40, 39413994.

The top-down approach involves enzymatic or chemical/physical processes to isolate nanocellulose from wood and forest/agriculture residues. In the bottom-up approach nanocellulose is obtained from glucose by bacteria. The isolated cellulosic materials with one dimension in the nanometer range are referred to as nanocellulose. Nanocellulose can be categorized as nanowhiskers or NCC and nanofibrillar cellulose (NFC). When plant products are subjected to strong acid conditions combined with sonication, they produce nanowhiskers or NCC. Nanowhiskers are rod-like structures resulting from the hydrolysis of noncrystalline domains. The dimension of the nanowhiskers depends on the source of cellulose; their length ranges between 100300 nm.1826 On the other hand, when plant products are subjected to high mechanical shearing without undergoing the hydrolysis steps, it results in NFC. The lateral dimension of NFC lies in the range of 1030 nm. They are generally present in bundles, in which the individual fibril’s lateral dimension is 5 nm.1826 The production method of nanocellulose is presented in Fig. 13.4 and the properties of nanocellulose have been compared with other materials in Table 13.2.

13.4

Processing of plant fiber-reinforced cementitious composites

Due to the corrosion problem of steel, there is an increasing need for alternative reinforcing materials for cementitious composites which can replace steel rebars. Following the use of various synthetic fibers, concrete has been also reinforced with various plant fibers, such as bamboo, coconut, sisal, flax, etc., for developing cost-effective and sustainable building constructions. However, prior to the use of

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

Table 13.3

349

Selected chemical surface treatment methods of plant

fibers27 Type of treatments

Advantages

Alkali treatment

Removal of lignin and hemicellulose, increases fiber tenacity, increase in surface roughness of fibers Reduction of moisture absorption, improvement of chemical resistance and durability Introduction of desired chemical groups to enhance compatibility with different matrices Improvement of interface, decrease in moisture absorption, increase in fiber tenacity Decrease in plant fiber’s hydrophilic nature Significant improvement of fiber/matrix interface through covalent bonding Significant improvement of fiber/matrix interface through covalent bonding Decrease in plant fiber’s hydrophilic nature

Acetylation Etherification Peroxide treatment Benzoylation Acrylation Silane treatment Permanganate treatment Graft copolymerization

Introduction of desired groups to enhance compatibility with different matrices and improvement of fiber/matrix interface

plant fibers in cementitious materials, they require various surface treatments, which have been discussed in Section 13.4.1.

13.4.1 Surface treatment of plant fibers More commonly, plant fibers are treated with various chemicals such as alkali, water repellents, silane, peroxides, permanganates, etc. to reduce their moisture absorption and to improve their compatibility with various matrices.27 Table 13.3 lists some commonly practiced chemical surface treatment methods of plant fibers and associated advantages.

13.4.2 Plasma surface treatment and grafting process Although quite efficient, the chemical surface treatment methods are not environmentally favorable due to production of waste chemicals and effluents. Recently, plasma surface modification has come out as a clean and dry surface modification technique of various polymeric fibers including plant fibers.2830 Plasma treatment can alter the surface characteristics in the nano scale without changing the bulk properties of the fibers.28 Industrial scale atmospheric plasma treatment machines have been developed for surface treatment of plant fibers in bulk at high processing speeds. Besides improving surface functionalities, wettability, as well as better plant fiber/matrix interfacial

350 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

properties, plasma treatment could also improve the mechanical properties of plant fibers.28,30 This opened the doorway of utilizing some newly explored plant fibers such as Quiscal in technical applications through improvement of their surface and mechanical properties by plasma treatment. Fig. 13.5 shows the atmospheric plasma treatment process and improvement of mechanical properties of Quiscal fibers through treatment with atmospheric plasma.30 Although plasma treatment can provide different types of functionalized surface depending on the type of gas used in the plasma reactor (air, oxygen, nitrogen, etc), this technique has limitations in terms of variety of surface modifications and stability of surface functional groups. Therefore, grafting of various polymers has also been carried out at the plasma functionalized surfaces in order to produce various stable functional groups as per the applications. This process, known as the plasmainduced grafting process, has been used recently by researchers to modify some plant fibers used in the apparel sectors, such as cotton, wool, silk, etc., to introduce flame retardancy properties.31,32 However, this process contains a chemical reaction step (grafting) and therefore, is associated with environmental pollution. The grafting step can be eliminated using the plasma polymerization process in which surface activation and High voltage Ceramics electrodes Banana fibers Power source Continuous fabric transmission device

Working gas air Rubber roll

(A)

20

35.0

Untreated

Fat, wax and impurities Moisture regain

30.0 90 kJ m–2

25.0

180 kJ m–2

10

20.0 %

Force (N)

15

15.0 5

10.0 5.0

0 0

0.5

1

1.5 2 Elongation (mm)

(B)

2.5

3

0.0 Untreated

90 kJ m–2

180 kJ m–2

(C)

Figure 13.5 Schematic of continuous atmospheric plasma treatment process for plant fibers (A), improvement in fiber tenacity (B), and removal of fiber impurities at different plasma treatment doses (C). Source: Relves et al.30

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

351

polymerization occur simultaneously. Plasma polymerization has been extensively used in anticorrosive surfaces and scratch resistance, chemical barrier, and water repellent coatings, since this is a clean and green (no solvent) process, reliable, reproducible, and suitable for wide varieties of monomers, different surfaces, and sample geometries.33,34 Recently, plasma polymerization of hexamethyldisiloxane and tetramethyldisiloxane on to polyester and high performance fibers have been reported to impart superhydrophobicity and heat resistance properties.33,34 Therefore, plasma polymerization process can also be utilized for plant fibers to create varieties of surface chemistries and topographies.

13.4.3 Dispersion of micro- and nanocellulose This section will help the readers to have a general idea about the nanocellulose dispersion techniques, and to know the application potential of these techniques for cementitious composites as well. These are general dispersion techniques used for mainly aqueous system, various solvents and polymer matrices. The dispersion of nanocellulose within cementitious composites has not been much studied. On the other hand, Section 13.4.4 contributes some information about nanocellulose dispersion within a cementitious matrix. NCC (nanowhiskers) have excellent mechanical and physicochemical properties.19 NCC can be used in various sectors due to its advantageous properties (such as high surface area, interesting mechanical and optical properties), renewability and abundance. The properties of nanomaterials can be fully explored when they are well dispersed within the matrix. Incorporation of NCC in aqueous medium, solvents, or polymeric resin is carried out using physical or chemical techniques, as shown in Fig. 13.6.19 In some cases, both physical and chemical techniques are used to obtain better dispersion in the matrix. Typically, NCC are produced as aqueous suspensions. The dispersion of NCC in hydrophobic resins, therefore, requires evaporation of water. Drying of NCC can be achieved through a freeze drying or spray drying process.3541 However, NCC tend to agglomerate during the drying process. The formation of agglomerates can be reduced by optimizing the rate, time period, and temperature of drying, and also by surface modification of NCC.3541

Figure 13.6 Classification of NCC dispersion techniques.

352 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

13.4.3.1 Chemical modification of micro and nanocellulose Chemical modification of NCC is performed to make desirable changes on the surface to widen its application in various matrices and solvents. Surface modification of NCC increases the compatibility with hydrophobic matrices and nonpolar solvents by decreasing the surface energy. NCC contain three hydroxyl groups in each pyranose ring. The hydroxyl group present on the 6th position is the primary hydroxyl group, which is more susceptible to any type of chemical modification.42 The chemical modification techniques can be classified as noncovalent and covalent techniques. Noncovalent modifications are mostly carried out using surfactant or polymer coating. Surfactants/polymer coatings are adsorbed on the surface of NCC without affecting the chemical morphology and therefore, securing the integrity and strength of NCC. Covalent modifications include acetylation, esterification, cationization, silylation, fluorescence labeling, polymer grafting, etc.

13.4.3.2 Noncovalent modification of micro and nanocellulose Noncovalent modification of MCC was first reported by Heux et al.43 MCC was dispersed in aqueous suspension with the help of an anionic surfactant (acid phosphate ester of alkyl phenol ethoxylate) using MCC-surfactant ratio of 4:1. The resultant surfactant coated MCC was freeze dried in the form of pallets. The surfactant coated MCC pallets were easily dispersed in nonpolar solvents using ultrasonication energy for a small duration.43 A similar procedure was also followed by Ljungberg et al.44 and Fortunati et al.45 for the dispersion of NCC.44,45 Researchers also used cationic surfactants to form a stable dispersion in organic solvents. Kaboorani et al.46 and Salajkova et al.47 used quaternary ammonium surfactant, hexadecyltrimethylammonium (HDTMA) bromide, in aqueous medium to obtain surfactant coated NCC. Surfactant coated NCC suspensions were then centrifuged to eliminate excess surfactant from the NCC surface and then freeze dried to obtain dried NCC powder.46,47 Cationic surfactant coated NCC can be easily dispersed in low polar solvents like tetrahydrofuran (THF). Another quaternary ammonium surfactant, cetyltrimethylammonium bromide (CTAB) has also been utilized due to its good adsorption onto NCC surface. According to Beaupre´ et al.,48 almost 60% of surface hydroxyl groups were covered by CTAB when used at 57.5 wt% with respect to NCC. CTAB coated NCC has been used for drug deliveries.48 Nanometal synthesis can also be done on the NCC surface using CTAB.49 The density and particle size of metal nanomaterials synthesized on NCC surface were controlled by CTAB concentration, pH, and the reduction time.4850 The use of cationic alkyl ammonium surfactants, didecyldimethylammonium bromide (DMAB) and CTAB, have also been reported to prepare NCC Pickering emulsions.51 The use of nonionic surfactant is also common to disperse NCC in hydrophobic polymer matrices. Kim et al.52 used sorbitan monostearate to improve the dispersion of NCC in THF. Sorbitan monoesterate was found to improve the stability of NCC dispersion within hydrophobic polystyrene matrix.52 Recently, 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-oxidized NCC whiskers were dispersed using Pluronic surfactants (Pluronic L61 and L121) for fabrication of epoxy nanocomposites.53 The use of

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

353

Pluronic led to better NCC dispersion and NCC/epoxy interfacial interactions, resulting in improved mechanical and thermal properties of the nanocomposites.

13.4.3.3 Covalent functionalization of micro- and nanocellulose Acetylation and esterification Acetylation and esterification of NCC have been carried out using a number of methods. In one of these techniques, acetylation was performed by treating fibrous as well as homogeneous NCC with acetic anhydride and acetic acid.54 In case of fibrous NCC, acetylation only occurred in the cellulose chains present on the surface of NCC, which surrounded the unreacted NCC core. On the other hand, uniform acetylation was obtained in case of homogeneous NCC caused by the progress of acetylation reaction into the core, owing to dissolution of surface acetylated cellulose. In another method, an NCC suspension was mixed with an aqueous emulsion of alkyenyl succinic anhydride and the mixture was subjected to freeze drying and heating to perform acetylation of NCC.55 Reaction with vinyl acetate in the presence of potassium carbonate catalyst has also been used to perform surface acetylation of NCC whiskers.56 An increase in the reaction time, however, led to complete destruction of the crystalline structure of NCC whiskers. A combined method of NCC synthesis and functionalization has also been developed recently.57 A mixture of acetic acid, hornificated cotton linters (HCL), and organic acids was used for the single-step synthesis and functionalization of NCC through the Fischer esterification process, as presented in Fig. 13.7. Gas phase esterification of NCC through evaporation in a large excess of palmitoyl chloride has also been

Figure 13.7 Single-step synthesis and functionalization of NCC through the Fischer esterification process. Source: Braun and Dorgan.57

354 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

reported.58 Refluxing of hydrolyzed NCC in organic acid chloride is another approach for the esterification of NCC.59 This method did not affect the crystalline core of NCC and resulted in grafting of NCC with organic fatty acids of different aliphatic chain lengths.

Cationization Recently, a one-step process of cationization of NCC surface has been developed. In this process, epoxypropyltrimethylammonium chloride (EPTMAC) was grafted to the NCC surface.60 The grafting occurred as a result of nucleophilic addition reaction of alkali-activated hydroxyl groups of NCC to the epoxy group of EPTMAC. The surface charge of NCC changed from negative to positive due to this grafting process and, as a result of positive surface charges, a stable aqueous suspension was obtained. The use of mild alkaline conditions in this process did not affect the original morphology or crystal structure of NCC.

Functionalization with fluorescein isothiocyanate For fluorescence bioassay and bioimaging applications, which are based on tracking of localization of the fluorophores, NCC has been covalently functionalized with fluorescein-50 -isothiocyanate (FITC).61 For this covalent functionalization, a threestep reaction route has been used, as shown in Fig. 13.8. In the first step, NCC surface was functionalized with epoxy functional groups through reaction with epichlorohydrin. In the second step, primary amino groups were introduced by opening the epoxy rings through reaction with ammonium hydroxide.

Figure 13.8 Scheme for functionalization of NCC with FITC. Source: Doug and Roman.61

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

355

In the final step, covalent bonding of FITC with NCC was achieved through reaction of primary amino groups of NCC with isothiocyanate groups present in FITC. The functionalization of NCC with FITC was confirmed through UV-Vis spectroscopy from the absorption peaks of FITC in the wavelength range of 450500 nm.

Functionalization with silanes Partial silylation of NCC whiskers has been carried out using a series of alkyldimethylchlorosilanes containing alkyl groups ranging from isopropyl to n-butyl, n-octyl and n-dodrecyl.62 When the degree of substitution (DS) was between 0.6 and 1, the silylated NCC whiskers were dispersed easily in medium polarity solvents like acetone and THF. No change in morphology or crystal structure was observed when DS was maintained below 0.6. However, when DS was increased above 1, the structural integrity was disrupted. According to the model developed by the researchers, silylated NCC at low DS maintained its structural integrity and was hydrophilic (as shown in Fig. 13.9). When the DS was moderate, the surface of NCC was hydrophobic and it could be dispersed in THF (Fig. 13.9). And lastly, when the DS was high, the surface chains were solubilized and silylation progressed into the NCC core, resulting in disruption of the crystal structure of NCC

Figure 13.9 Model explaining the silane functionalization of NCC at: (A) low DS showing onset of surface functionalization; (B) moderate DS showing surface functionalization; and (C) high DS showing disruption of NCC core. Source: Gousse´ et al.62

356 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

(Fig. 13.9). In addition to the above process, trimethyl silylation of NCC surface, derived from BC, has also been reported.25

TEMPO-mediated oxidation and functionalization TEMPO-mediated oxidation has been used to functionalize NCC with carboxylic groups.6365 This functionalization is carried out using TEMPO reagent in a NaBr and NaOCl environment by specifically oxidizing the primary hydroxymethyl groups.66 Only 50% of the surface hydroxymethyl groups are oxidized in TEMPOmediated oxidation, keeping the secondary hydroxyl groups intact. TEMPO-mediated oxidation is shown schematically in Fig. 13.10. This functionalization resulted in better aqueous dispersion of NCC due to electrostatic repulsion between the carboxylic groups and the resulting suspension showed a liquid-crystal-like behavior. The degree of oxidation of NCC in TEMPO-mediated oxidation process could be controlled by varying the molar ratio of NaOCl over the anhydroglucose unit of hydrolyzed cellulose. Higher NaOCl molar ratio resulted in higher oxidation degree and carboxyl content; however, excessive oxidation led to the degradation of amorphous region of NCC affecting the structural integrity.67 Once NCC is functionalized with carboxyl groups, it can be further grafted with various polymers to obtain different functionalities. Preparation of a “brush polymer” through grafting of poly (ethylene glycol) on to carboxyl functionalized NCC (obtained from TEMPO oxidation) has been reported.68 In a recent study, NCC, after TEMPOoxidation and amine grafting with tuned charge density, was used to control the morphology and stability of silver nanoparticles in aqueous suspensions.69

Functionalization through polymer grafting Functionalization of NCC surface through polymer grafting has been frequently reported. In one approach, polymer chains were grafted to the NCC surface (known

Figure 13.10 Schematic of TEMPO-mediated oxidation of NCC. Source: Habibi et al.66

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

357

as “grafting to” approach), whereas in the second approach, polymer chains were grown on the NCC surface (known as “grafting-from” approach) through a graft polymerization process. In the first approach, polycaprolactone (PCL) of different molecular weights was grafted to NCC through the isocyanate-mediated coupling reaction.21 At higher grafting density, crystallization of PCL on NCC surface was observed. Grafting of polyurethane onto NCC surface using the same approach has also been reported.70 Carboxylated NCC, obtained through TEMPO oxidation, could be grafted with polymers through the peptide coupling reaction. Through this approach, grafting of PEG-NH2 on to TEMPO-oxidized NCC surface was carried out through EDC/NHS [1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide/N-hydroxysuccinimide] carbodiimide chemistry at room temperature.68 The same approach has also been utilized for grafting of DNA to the NCC surface.71 Grafting of maleated polypropylene onto NCC surface has also been reported.44 NCC with thermos reversible aggregation behavior, which can be used for designing stimuliresponsive bio-based materials, was prepared through grafting of thermo-responsive polymers onto NCC surface via a peptidic coupling reaction.72 In the graft polymerization approach (i.e., “grafting-from”), polymer chains have been grown on the NCC surface through the atom transfer radical polymerization (ATRP) process. In this process, first the hydroxyl groups of NCC surface were esterified with 2-bromoisobuturyl bromide (BIBB). In the second step, the selected monomers were polymerized. The grafting process on NCC surface could be controlled very precisely using this surface initiated ATRP technique.73 The monomers which have been graft polymerized on NCC surface include styrene and N, N-dimethylaminoethyl methacrylate.7476 Azobenzene polymers were also grafted to NCC surface to produce a novel amphotropic hairy rod-like system exhibiting thermotropic and lyotropic liquid crystalline properties.77 Another approach which has been tried for graft polymerization on NCC surface was through ring-opening polymerization. PCL was grafted to NCC through this approach using stannous octoate (Sn(Oct)2) as the grafting and polymerizing agent.21 The use of microwave irradiation was also used to improve the grafting efficiency onto the NCC surface.78,79 Attempts were also made to use a novel in situ solvent exchange method for grafting of long-chain isocyanate groups onto NCC whiskers.80

13.4.4 Dispersion of micro- and nanocellulose within cementitious matrix Due to the lack of research studies in the field of nanocellulose-reinforced cementitious composites, very little information is available on the dispersion of nanocellulose within cementitious matrix. The hydrophilic nature and water retention capability of NCC and MCC influence the yield stress of cement paste and hydration kinetics of cementitious composites. A suspension of MCC in water was prepared by Hoyos et al.81 to study the amount of water absorbed by MCC. The aqueous suspension of MCC was prepared by mixing 0.5 g of MCC in 3 mL of water. The suspension was kept for three days at 25 C and centrifuged at 3000 rpm

358 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

for 25 min to segregate the two phases. The saturated MCC samples were then quantified for the absorbed water. It was observed that MCC absorbed 230% of water with respect to its mass and 100% with respect to its volume. A similar procedure of MCC saturation was also adopted to prepare MCC-reinforced cementitious composites. Saturated MCC suspensions were prepared by adding MCC in an adequate amount of water and storing the suspension for two days. Then, the saturated MCC suspension was mixed with cement in a Hobart planetary mixer and additional water was added during the mixing process.81 Cao et al.82 used dispersed NCC (5.38 wt.%. cellulose nano crystals (CNCs) in water) obtained from the sulfuric acid hydrolysis of cellulose fibers. 0.81% surface of NCC was grafted by sulfate groups, which ensured homogenous dispersion of NCC in aqueous suspension. NCC-reinforced cement composites were prepared by mixing diluted NCC suspensions (0.13.8 wt.% w.r.t. cement) and water with cement with the help of a vacuum mixer. The vacuum mixer was set at the speed of 400 rpm for 180 s, and there was a pause after 90 s for scrapping the mixture from the bowl. The vacuum mixer was used to minimize the air entrapment during the mixing process and it also maintained the consistency of cement pastes.82

13.5

Properties of plant fiber-reinforced cementitious composites

13.5.1 Influence of plant fibers on the properties of cementitious composites 13.5.1.1 Influence of plant fibers on flow behavior of cement It has been observed that the incorporation of plant fibers within cement mixtures reduces its workability, depending on the fiber volume fraction and aspect ratio.83,84 The decrease in cement flow behavior caused by addition of plant fibers is attributed to their hydrophilic nature and absorption of water from the cement mixture. Therefore, in order to obtain a cement mixture with sufficient workability, either researchers needed to treat the surface to reduce their hydrophilicity or they presaturated or increased the water/cement ratio used, taking into account the water absorption of the plant fibers.85

13.5.1.2 Influence of plant fibers on setting time of cement Plant fibers have shown a negative effect on the hydration behavior of Portland cement.8690 The reasons ascribed for this effect are (1) production of soluble sugars resulting from hydrolysis of lignin and partial solubilization of hemicellulose. Calcium compounds produced within the cementitious matrix due to dissolution of sugars retards the hydration process; (2) pectins present in plant fibers act as the inhibitor for the growth of calcium silicate hydrate (CSH); (3) carbohydrates and hemicelluloses present in wood and plant fibers decrease the rate of hydration of cement. The negative effect of plant fibers on cement hydration can be reduced through the addition of

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

359

pozzolan, treatment of plant fibers to remove lignin, increased curing temperature, addition of chemical accelerators and supplementary materials, etc.85

13.5.1.3 Influence of plant fibers on plastic shrinkage The addition of plant fibers was found to be beneficial to reduce plastic shrinkage and associated crack development in cement mortar.9195 The evaporation of water from the exposed surface of fresh mortar mixes results in volume contraction, which is known as the plastic shrinkage. Plant fibers could reduce considerably the maximum width and area of cracks, formed due to plastic shrinkage, owing to abridgement of cracks by fibers, reduced rate of settlement of particles, and decreased bleeding induced by fibers.95

13.5.1.4 Influence of plant fibers on drying shrinkage Drying shrinkage is a very important property of cementitious composites influencing their durability. Drying shrinkage results from the loss of capillary water from the hardened cement mixture, leading to contraction and crack formation within concrete. According to the previous studies, the addition of plant fibers, such as sisal, to cement mortar increases its drying shrinkage.93,96 This could be attributed to the high moisture absorption of plant fibers and also the increased porosity of mortar because of the addition of plant fibers. However, drying shrinkage of mortar strongly depends on the type and quantity of plant fiber, their surface characteristics and moisture absorption behavior. For example, Fig. 13.11 shows the influence of sisal and coconut fibers on the drying shrinkage of cement.

Figure 13.11 Drying shrinkage of cement containing sisal and coconut fibers (W, water cure; DC, damp cloth-cured; PDC, pressure 1 damp cloth-cured; M1, mortar mix; M1S225, mortar mix with 2% sisal; M1S325, mortar mix with 3% sisal; M1C225, mortar mix with 2% coconut fiber; M1C325, mortar mix with 3% coconut fiber). Source: Toledo Filhto et al.93

360 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

It can be observed that the addition of 2% and 3% of sisal fiber to the mortar mix led to an increase in drying shrinkage by 10% and 27%, respectively, in the case of water cured samples. Also, the drying shrinkage of composites containing 3% of sisal fibers was 8.2% higher than those reinforced with coconut fibers.93

13.5.1.5 Influence of plant fibers on mechanical properties Reinforcement of cementitious matrix with plant fibers can significantly influence its mechanical properties.97106 Among the various properties, the impact resistance of cement mortar could be significantly improved through reinforcement of plant fibers, particularly coir fibers due to their higher elongation at break and higher toughness as compared to other plant fibers.85 Impact resistance of cement mortar was improved by 18 times as compared to the unreinforced mortar specimens using coir fibers.85 Table 13.4 lists the flexural strength and toughness of mortar reinforced with different plant fibers. It can be observed that the flexural toughness of mortar increased with the fiber volume fraction of plant fibers, while the optimum fiber volume fraction to achieve high flexural strength was 0.080.1. Among the plant fibers, due to its high aspect ratio and consequently high specific surface area, abaca fibers led to a strong improvement of cement flexural properties.85

13.5.2 Effect of micro- and nanocellulose on the properties of cementitious matrix 13.5.2.1 Influence on flow behavior of cement paste The mini slump test is performed for the analysis of flow behavior of cement to check the workability, according to Standard ASTM: C-143. In the freshly prepared cement paste, small particles interact via colloidal forces, such as Van-der Waals, electrostatic repulsion, steric hindrance, and hydrogen bonding, and some bigger particles interact via direct contact like friction or collisions. The yield stress (τ 0) is the stress necessary to break those interactions and separate the particles.81 MCC (3 wt.% w.r.t. cement)-reinforced cement paste showed an increase of τ 0 by 2.6 times over plain cement paste. This may result in an increase of energy costs in construction. However, for certain construction applications, a higher τ 0 is necessary; for example, in rigid pavements where the fresh paste should retain its shape. Therefore, for these applications MCC reinforcement will be ideal.81 Cao et al.82 studied using a nanorheometer the influence of NCC [or CNC] vol.% on yield stress, as shown in Fig. 13.12. According to them, at lower NCC concentration (0.020.04 vol.% w.r.t cement), the yield stress decreased as compared to plain cement paste with the increase in NCC concentration. But on further increase in the NCC concentration, the yield stress again started to increase and at 0.3 vol.% of NCC (w.r.t. cement) the yield stress reached the value similar to that of plain cement. With a further increase of NCC (1.5 vol.% w.r.t. cement) the yield stress increased significantly as compared to plain cement paste. This contradictory behavior between 0.020.04 vol.% of NCC could be explained based on various

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

361

Mechanical properties of cementitious composites reinforced with plant fiber

Table 13.4

Type matrix

Fiber type

Fiber content (%w/w)

Fiber aspect ratio

Flexural strength (MPa)

Flexural toughness (kJ  m22)

Paste

 Refined softwood sisal kraft pulp

0 4 8 12 4 8 12 4 8 12 4 8 12 4 2 4 6 8 10 12 14 2 4 6 8 10

 53

11.8 6 3.7 19.2 6 1.9 23.5 6 0.8 25.0 6 2.1 16.5 6 0.6 21.5 6 1.6 20.3 6 1.4 15.5 6 1.3 19.5 6 1.4 20.1 6 2.5 15.6 6 0.8 21.4 6 0.9 22.2 6 1.3 14.4 6 1.0 10.9 6 1.5 12.1 6 1.3 16.2 6 1.0 17.4 6 0.9 18.6 6 1.2 19.2 6 1.5 21.8 6 1.7 17.5 6 2.0 21.8 6 2.1 26.3 6 1.6 27.3 6 3.2 24.7 6 3.9

0.04 6 0.01 0.64 6 0.09 1.32 6 0.11 1.93 6 0.42 0.39 6 0.06 0.92 6 0.13 1.41 6 0.20 0.21 6 0.03 0.53 6 0.08 1.01 6 0.15 0.29 6 0.04 0.82 6 0.11 1.50 6 0.18 0.58 6 0.17 0.07 6 0.01 0.15 6 0.02 0.23 6 0.02 0.32 6 0.03 0.45 6 0.07 0.54 6 0.05 0.70 6 0.06 0.47 6 0.10 0.93 6 0.254 1.76 6 0.48 2.08 6 0.33 2.19 6 0.78

9.2 6 0.7 9.9 6 0.8 11.3 6 0.8 12.7 6 1.2 15.9 6 1.2 16.7 6 1.0 15.0 6 1.7 10.3 6 1.6

0.25 6 0.02 0.45 6 0.03 0.62 6 0.07 0.84 6 0.08 1.64 6 0.17 2.05 6 0.29 2.49 6 0.47 2.47 6 0.46 3.07 6 0.58

Unrefined waste sisal kraft pulp Unrefined banana kraft pulp Unrefined eucalaptus kraft pulp Sisal strand refined bamboo kraft pulp

Refined abaca kraft pulp

Mortar

Unrefined sisal kraft pulp

0.5 1 1.5 2 4 6 8 10 12

122

127

61

89

400

Source: Onuaguluchi, O.; Banthia, N. Plant-based Natural Fiber Reinforced Cement Composites: A Review, J. Cem. Concr. Compos. 2016, 68, 96108.

362 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 13.12 Influence of NCC content (% by volume) on the yield stress of mortar paste. Source: Cao et al.82

interactive forces present between NCC particles during mixing with cement. At lower concentration, NCC particles behaved like cement admixtures (e.g., polycarboxylate) and promoted the degree of cement hydration by dispersing cement particles through steric stabilization mechanism. This resulted in lower yield stress as compared to plain cement paste. But, with the increase in NCC concentration, NCC particles started to agglomerate and the force required to break these agglomerates was very high, resulting in high yield stresses.82

13.5.2.2 Influence on hydration and mechanical properties of cementitious composites It was observed that incorporation of 3 wt.% of MCC decreased the mechanical properties, such as flexural and compressive strength, of cementitious composite in a normal curing period of 28 days.81 During accelerated curing conditions, which resulted in a higher degree of cement hydration, the above mechanical properties of MCC-reinforced cement just reached the values of plain cement mortar. As MCC could retain water due to its hydrophilic nature, at higher temperature (during accelerated curing) it released water leading to a greater formation of hydration products.81 According to Cao et al.82 NCC behaved as a water reducing agent at lower concentration (0.2 vol.%) and helped to disperse cement particles. The degree of hydration increased with the addition of an adequate amount of NCC. It was observed that NCC% could only be increased up to 0.5% as further increases resulted in the segregation of particles. The increase in flexural strength was 20%

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

363

Figure 13.13 Improvement of mortar’s flexural strength due to CNC addition (WRA, Water reducing agent; CNC, Cellulose nanocrystals; B3B, Ball on three ball flexural testing method). Source: Cao et al.82

and 30% in 3 and 7 days, respectively, with the addition of 0.2% of NCC (Fig. 13.13). The improvement of flexural strength due to NCC addition was attributed to the improved hydration of cement owing to the steric stabilization of cement particles and short circuit moisture diffusion.82 The short circuit diffusion was the penetration of water from the hydrated part of CSH (more dense part) to the unhydrated part with the help of NCC particles leading to better hydration,82 as shown in Fig. 13.14.

13.5.2.3 Effect on microstructure of cementitious composites According to C. G. Hoyos et al.81, MCC cement composites possess strong interface between MCC and hydration products of cement. The available hydroxyl groups of MCC can form hydrogen bonding with the hydration products of cement. MCC remains saturated with water and therefore, the CSH phase (cement hydration product) growing near MCC can utilize the water bound with MCC. Moreover, the size distribution of MCC is similar to CSH crystals making MCC a suitable reinforcement for cementitious matrix. The microstructure of MCCreinforced cement is presented in Fig.13.15.

364 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 13.14 Schematic diagrams explaining the short circuit diffusion mechanism: (A) plain cement mortar and (B) mortar reinforced with CNC. Source: Cao et al.82

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

365

Figure 13.15 FE-SEM image of cement-based materials with 3 wt.% of MCC at different magnifications. Source: Hoyos et al.81

(A)

(B) 0.35

30

Fiber 1

25 20 20 14

15

10 10

6 3

5

Fiber 2

Fiber 3

Fiber 4

0.3

21 Diameter in mm

Frequency %

25

1

0.25 0.2 0.15 0.1

0 0,08 0,11 0,14 0,17 0,2 0,23 - 0,26 - 0,29 0,1099... 0,1399... 0,1699... 0,1999... 0,2299... 0,2599... 0,2899... 0,3199...

0.05 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96

Diameter groups (mm)

Readings along the length of fiber

Figure 13.16 Variability of diameter of banana fibers: (A) frequency diameter distribution and (B) diameter variation along the fiber length. Source: Mukhopadhyay et al.107

13.6

Challenges with plant fiber-reinforced cementitious composites

13.6.1 Variability in plant fiber properties One of the inherent drawbacks of plant fibers is the variability of properties.85,107109 The variability of their physical and mechanical properties mainly originates from the variation in their chemical structure and composition, such as cellulose content, degree of polymerization, orientation of molecular chains, crystallinity, etc.85,107109 These parameters are highly dependent on the growth conditions of the plant and also on the fiber extraction methods. Therefore, fibers extracted from different parts of the plants or grown in different locations and weather conditions present huge variability in their length, cross-sectional area, and mechanical properties.85,107109 An example of variability of plant fiber diameter is presented in Fig. 13.16. The distribution of diameter of 100 fibers, presented in Fig. 13.16(A), shows high variability among the different fibers. Even, within the

366 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

same fiber, the diameter varies significantly along the length of fiber, as shown in Fig. 13.16(B). Besides chemical structure and composition, the mechanical properties of plant fibers were found to be highly dependent on the testing parameters, such as strain rate. A higher speed of testing and higher strain rate resulted in brittle failure mechanism and lower tenacity as compared to a lower strain rate which led to ductile failure of fibers, resulting in higher tenacity.107 High variability of plant fiber properties is a problem when developing products based on plant fibers. The prediction of properties and product designing become difficult. For structural applications, a higher safety margin is required when plant fibers are used as the reinforcement of cementitious composite-based structures.

13.6.2 Hydrophilicity of plant fibers Plant fibers are hydrophilic owing to the presence of functional groups such as hydroxyl in their structure.110 Therefore, plant fibers absorb a considerable amount of moisture from the surrounding environment. The moisture absorption capacity of plant fibers depends mainly on their chemical composition and crystallinity. Table 13.5 lists the moisture absorption of some selected plant fibers.110 The high moisture absorption of plant fibers leads to a number of problems when used for the reinforcement of cementitious materials such as85,110: (1) plant fibers swell due to absorption of moisture and shrink when moisture is removed due to dry atmosphere and elevated temperatures. Then, when plant fibers are used to reinforce cementitious matrix, their frequent swellingshrinking phenomena leads to formation of cracks. This leads to reduced mechanical performance and durability of cementitious composites. (2) High absorption of alkaline solution present within the cement mixture leads to degradation of plant fibers with time. These result in the deterioration of properties of plant fibers as well as plant fiberreinforced cementitious composites. (3) If plant fibers are not saturated, during mixing with cementitious materials they absorb considerable amount of water and reduce the water required for cement hydration. This leads to a reduced degree of cement hydration and, consequently, poor mechanical performance of cementitious Table 13.5

Moisture absorption of plant fibers

Fibers

Moisture absorption (%)

Sisal Coconut Bamboo Hemp Caesar wood Banana Piassava palm Date palm

110 93.8 145 85105 182 407 34108 6084

Source: Pacheco-Torgal and Jalali.110

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

367

composites. (4) Moisture absorption of plant fibers leads to breakage of hydrogen bonds between the fiber and cementitious matrix and, therefore, weakens the fiber/ matrix interface and, consequently, it deteriorates mechanical strength. Therefore, a number of fiber surface treatment methods have been tried to reduce the moisture absorption of plant fibers for applying them in cementitious composites.

13.6.3 Fiber/matrix interface Poor interface between the plant fibers and different matrices present one of the major problems with plant fiber composites. Poor interface results in inferior load transfer between fiber and matrix, resulting in lower mechanical performance. Before discussing interface between plant fiber/cement composites, a brief discussion has been provided about the plant fiber/polymer composite’s interface, just to provide a general idea about the interface in plant fiber composites. Due to the hydrophilic nature of the fibers, the interface formed between plant fibers and hydrophobic polymer matrices is very weak.27 On the other hand, the interface of plant fibers with hydrophilic matrices can be better as a result of interfacial hydrogen bonds. However, as discussed earlier, due to high moisture absorption of plant fibers, a breakage of interfacial hydrogen bonds occurs in the interfacial region, weakening the fiber/matrix bonding. The interface between different types of reinforcement and brittle cementitious matrices in different scales is also very weak, which results in pullout when subjected to loading. Fig. 13.17 shows the complete pullout of a bamboo culm from

Figure 13.17 Pullout test to measure bonding strength between bamboo culm and cementitious matrix (A) and imprint of bamboo fibers on cementitious matrix after pullout test (B) Source: Khare.111

368 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

the cement composites, representing a very weak interface. The bonding strength of this type of bamboo rebars, obtained from pullout testing, is listed in Table 13.6 and compared with steel rebars. It can be observed that the bonding strength of bamboo rebars is even lower than the smooth steel rebars and much lower than the rough steel rebars used for construction applications. Therefore, it is highly essential to improve fiber/matrix bonding in case of plant fiber composites and various approaches have been tried for this purpose, as discussed in Section 13.6.4.

13.6.4 Durability of plant fibers and reinforced structures Plant fibers are prone to degradation because of their high moisture absorption. Therefore, the long-term stability of plant fibers and plant fiber-reinforced cementitious composites is questionable. Additionally, the cement matrix presents an alkaline environment, which accelerates the degradation of plant fibers due to the dissolution of lignin and hemicellulose in alkaline solution from porous water. Fig. 13.18 shows the typical load-deflection curves in 3 point

Bonding strength of bamboo and steel rebars with cementitious matrix

Table 13.6

Rebar type

Bonding strength (MPa)

Bamboo Bamboo with epoxy Smooth steel Rough steel

0.81 0.32 1.33 6.87

Source: Pacheco-Torgal and Jalali.110

1200

0 cycles

Load (kN)

1000 800

5 cycles

600

10 cycles

400 25 cycles 200 0 0

0.25

0.50

0.75

1.00

1.25

Deflection (mm)

Figure 13.18 Load-deflection curves of kraft fiber-reinforced cement composites subjected to different wet/dry cycles. Source: Mohr et al.112

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

369

Figure 13.19 Fracture surface of kraft pulp fiber cement composites showing progression of fiber degradation: (A) without wet/dry cycles; (B) after 5 cycles; and (C) after 25 wet/dry cycles. Source: Mohr et al.112

bending mode of kraft pulp (KP) fiber (obtained from Slash pine softwood)-reinforced cement composites, subjected to wet/dry cycles.112 The degradation of fibers due to wet/dry cycles led to 43.5%52.0% loss of first crack strength, 50.8%72.4% loss of peak strength, and 97.5%98.8% loss of postcracking toughness of cementitious composites. The Scanning electron microscope (SEM) micrographs (Fig. 13.19) of fracture surface shows significant fiber pullout in the case of samples without wet/dry cycles, whereas significant fiber rupture occurred after 25 wet/dry cycles resulting from the brittleness of fibers after the degradation cycles. Different studies which revealed the inferior durability of plant fiberreinforced concrete are presented in Table 13.7 and various approaches to improve the durability of plant fiber-reinforced concrete are listed in Table 13.8. One example of improved durability of pretreated plant fiber-cement composites is shown in Fig. 13.20. It can be observed that kraft softwood pulp (i.e., pulp obtained through kraft pulping of pine) and cotton linter-reinforced cementitious composites could provide better resistance to accelerated aging conditions after the hornification treatment.130

370 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Deterioration of properties of plant fiber and plant fiber-reinforced cementitious composites owing to degradation

Table 13.7

Type of fibers

Degrading conditions

Deterioration of properties

Coir, sisal, jute, and hibiscus fibers

Water, saturated lime, and sodium hydroxide (NaOH) solutions

Sisal and coconut fibers

Alkaline solution

Kraft pulp fiberreinforced cement paste specimens Sisal and eucalyptus fiber-reinforced roofing titles

25 wet/dry cycles

Lignin, hemicellulose, and cellulose content decreased significantly.113 Cement mortar containing the degraded fibers showed reduced mechanical strengths Treated fibers completely lost their flexibility. Mortar containing treated fibers showed significant decrease in toughness114 Significant loss of mechanical properties112

Weathering conditions

Drastic reduction in toughness of cement composites115

Different approaches to reduce deterioration of performance of plant fiber-reinforced cementitious composites owing to degradation

Table 13.8

Approach

Strategies

Improvement of properties

Use of supplementary cementitious materials to substitute cement and reduce alkalinity

Undensified silica fume Binary and ternary blends of slag, metakaolin, and silica fume Calcium hydroxide free cement matrix

Reduced degradation of plant fiber cement-based composites116 Reducing degradation of pulp fiber cementitious composites subjected to wet/dry cycles117

Metakaolin and calcined waste crushed clay brick Low alkaline ground granulated blast furnace slag cement

Reduced loss of toughness and long-term embrittlement of sisal fiber-reinforced cement118 4 times increase in ultimate bending strength and 42 times increase in toughness of sisal fiber-reinforced cement composites119 Reinforcing coir fibers appeared undamaged after 12 years120

(Continued)

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

Table 13.8

371

(Continued)

Approach

Strategies

Improvement of properties

Pretreatment of plant fibers

Acetylation and silane treatment

Reduced moisture absorption of fibers leading to significant improvement in fiber/cement bond behavior121 Improvement of fiber strength, fiber/ cement bonding and toughness of plant fiber-reinforced cement composites122 Removal of residual lignin and extractives resulting in deterioration of tensile strength of fibers, increased fiber softness resulting in better fiber/ cement bond behavior, reduced fiber pull-out length and ductility of composites, increased peak mechanical strength of composites, reduced durability of composites resulting from increased mineralization of bleached fibers123126 Increases the fiber fineness, softness and fiber/cement interaction, modulus of rupture, limit of proportionality, and modulus of elasticity of cement composites127,128 Improves fiber/matrix interface, dimensional stability of plant fibers, and durability of plant fiber cement composites129131

Alkali treatment

Bleaching

Beating

Hornification, i.e., alternate drying and rewetting of fibers to irreversibly reduce water retention Pyrolysis at 200 C

Specialized composite processing

Accelerated carbonation curing

Dehydrates chemical components of plant fibers, improves surface roughness and enhances fiber/cement interfacial bonding132 Reduces alkalinity of cement mixtures, reduced pore volume due to precipitation of carbonate products, improves early strength gain, increases mechanical strength, increases durability by improving resistance to sulfate attack, water absorption, and chloride ion penetration, provides environmental benefits133137

372 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 13.20 Compressive strength of cementitious composites reinforced with (A) untreated kraft pulp (KP) and hornificated kraft pulp (HKP) and (B) untreated cotton linters (CL) and hornificated cotton linters (HCL); thin lines represent the aged (A) composites. Source: Claramunt et al.131

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

373

Figure 13.21 Stable microcrystalline cellulose suspensions prepared using Pluronic F 127: (A) aqueous suspensions; and (B) optical micrograph of 0.5% MCC suspension. Source: Parveen et al.139

13.6.5 Dispersion of micro- and nanocellulose Dispersion of nanomaterials is one of the primary problems in successful development of nanocomposites.138 Nanomaterials form agglomerates owing to strong attractive forces acting between them and it is a challenging task to disperse them homogeneously in an aqueous medium or in different matrices. Therefore, the first and foremost step towards developing nano/microcellulose-reinforced cementitious composites is to prepare highly homogeneous and stable aqueous suspensions, which can be subsequently added to the cement mixture. Direct mixing of micro/nanocellulose powder with cement mixture may not ensure homogeneous dispersion, as is also observed in case of carbon nanotubes.2 For practical applications in the civil construction industry, the micro/nanocellulose suspensions AWE should have high storage stability so that they can be stored for a long time period before mixing with cementitious materials. As discussed in Section 13.4.3, micro/nanocellulose can be dispersed in aqueous medium by various mechanical and chemical methods. Their dispersion stability can be improved through chemical functionalization or using surfactants. For example, Fig. 13.21 shows the stable suspensions of MCC prepared using Pluronic F 127 surfactant (BASF).139

13.7

Applications

Plant fibers have huge potential for applications in the construction industry. Several research studies conducted to date have proved that plant fibers could be low cost and light weight reinforcing materials for cementitious composites. The future of micro/nanocellulose-based cementitious composites is also promising for structural applications. Apart from the reinforcement of cementitious composites, plant fibers can have other applications in the building industry

374 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

such as (1) plant fibers can be used as thermal insulation materials in building construction to conserve energy.140 Plant fiber composites (from cotton and sunflower stalks) insulation materials exhibited suitable mechanical and thermal performances, which satisfactorily met the requirements of the Turkish TS 805 EN 601 insulation material application standard.141 The use of nanocellulosebased aerogel materials to improve thermal and sound insulation in buildings can be highly effective. (2) Micro/nanocellulose can be used in building construction to significantly reduce the permeability of cement-based materials and, consequently, to improve their durability.142 (3) Wet sprayed plant fibers can also be used as a surface curing agent for repairing concrete infrastructures.143

13.8

Conclusions and future trend

Plant fibers show high potential as reinforcing materials of cementitious composites. Addition of plant fibers showed significant enhancement in the impact resistance, toughness, and strength of cementitious composites. Additionally, they are advantageous to reduce the plastic shrinkage and associated crack formation of cementitious matrices. The use of nano- and micro-dimensional plant fibers such as nano- and microcellulose showed still higher improvements in the performance of cementitious composites. The problems associated with the plant fibers mainly come from their high moisture absorption. This led to their negative effect on cement hydration, drying shrinkage, interface with cement, and durability of cementitious composites. Therefore, it is highly essential to modify the surface of plant fibers to reduce their hydrophilicity. Among the various surface treatment methods, physical techniques such as plasma and corona treatments are much more favorable over the chemical techniques in terms of environmental pollution and sustainability. At present, machines for continuous plasma treatment of fibers and textiles on the industrial scale are available and can be utilized for developing plant fiber-based cementitious composites. Although micro- and nanocelluloses show huge potential as a reinforcement of cement, their dispersion still presents a big concern. Stable and well dispersed aqueous micro/nanocellulose suspensions could be obtained through their chemical functionalization, which introduces an additional step in the process. Therefore, more research and developments are required for successful application of micro- and nanocelluloses within cementitious composites. Despite tremendous research work on plant fiber-reinforced cementitious composites, up until now their practical applications are very few. One of the main problems in applying plant fibers within cementitious composites is the lack of design standards on fiber-reinforced concrete. Therefore, considerable effort should be directed towards developing suitable design codes and guidelines to increase the application of plant fibers in the civil construction industry.

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

375

References 1. Ardanuy, M.; Claramunt, J.; Filho, R. D. T. Cellulosic Fiber Reinforced Cement-based Composites: A Review of Recent Research. Constr. Build. Mater. 2015, 79, 115128. 2. Peng, B. L.; Dhar, N.; Liu, H. L.; Tam, K. C. Chemistry and Applications of Nanocrystalline Cellulose and Its Derivatives: A Nanotechnology Perspective. Can. J. Chem. Eng. 2011, 89 (5), 11911206. 3. Parveen, S.; Rana, S.; Fangueiro, R. A Review on Nanomaterial Dispersion, Microstructure and Mechanical Properties of Carbon Nanotube and Nanofiber Based Cement Composites. J.Nanomater. 2013, 2013, 119, Article ID 710175. 4. Parveen, S.; Rana, S.; Fangueiro, R.; Paiva, M. C. Microstructure and Mechanical Properties of Carbon Nanotube Reinforced Cementitious Composites Developed Using a Novel Dispersion Technique. Cem. Concr. Res. 2015, 73, 215227. 5. Pellenq, R. J.; Kushima, A.; Shahsavari, R.; Van Vliet, K. J.; Buehler, M. J.; Yip, S.; Ulm, F. J. A realistic molecular model of cement hydrates. Proc. Natl. Acad. Sci. 2009, 106 (38), 1610216107. 6. LucintelGlobal Management Consulting & Market Research Firm.-Opportunities in Natural Fiber Composites. http://www.lucintel.com/lucintelbrief/potentialofnaturalfibercomposites-final.pdf (accessed Nov 29, 2016). 7. Ahmad, F.; Choi, H. S.; Park, M. K. A Review: Natural Fiber Composites Selection in View of Mechanical, Light Weight, and Economic Properties. Macromol. Mater. Eng. 2015, 300 (1), 1024. 8. Rana, S.; Fangueiro, R., Eds. Fibrous and Textile Materials for Composite Applications; Springer: Singapore, 2016, ISBN 978-981-10-0234-2. 9. Rana, S.; Pichandi, S.; Parveen, S.; Fangueiro, R. Regenerated Cellulose Fibres and their Implications on Sustainability. In Roadmap to Sustainable Textiles and Clothing; Dr. Muthu, S., Ed.; Springer: Singapore, 2014; pp 239276. 10. Rana, S.; Pichandi, S.; Parveen, S.; Fangueiro, R. Biosynthetic Fibres: Production, Processing and Properties and Their Sustainability Parameters. In Roadmap to Sustainable Textiles and Clothing; Dr. Muthu, S., Ed.; Springer: Singapore, 2014; pp 109138. 11. Rana, S.; Pichandi, S.; Parveen, S.; Fangueiro, R. Natural Plant Fibres: production, Processing and Properties and Their Sustainability Parameters. In Roadmap to Sustainable Textiles and Clothing; Dr. Muthu, S., Ed.; Springer: Singapore, 2014; pp 135. 12. Rana, S.; Pichandi, S.; Karunamoorthy, S.; Bhattacharyya, A.; Parveen, S.; Fangueiro, R. Carbon Footprint of Textile and Clothing Products. In Handbook of Sustainable Apparel Production; Dr. Muthu, S., Ed.; CRC Press: New York, 2014; pp 141165. 13. Karunamoorthy, S.; Rana, S.; Begam, R.; Parveen, S.; Fangueiro, R. Life Cycle Assessment Studies Pertaining to Textiles and Clothing Sector. In Handbook of Sustainable Apparel Production; Dr. Muthu, S., Ed.; CRC Press: New York, 2014; pp 181205. 14. R. Fangueiro; S. Rana (Eds.). Natural Fibres: Advances in Science and Technology Towards Industrial Applications. 2016, Springer, ISBN 978-94-017-7515-1. 15. Rana, S.; Fangueiro, R., Eds. Advanced Composite Materials for Aerospace Engineering: Processing, Properties and Applications; Woodhead Publishing Ltd.: Cambridge, 2016, ISBN: 9780081009390. 16. A. F. Turbak; F. W. Snyder; K. R. Sandberg In Microfibrillated Cellulose, a New Cellulose Product: Properties, Uses, and Commercial Potential. Proceedings of the

376 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

17.

18.

19.

20.

21. 22.

23.

24.

25.

26. 27. 28.

29. 30.

31.

32.

33.

Ninth Cellulose Conference, pp 815827, Applied Polymer Symposia, Wiley: New York, 1983. F.W. Herrick; R. L. Casebier; J. K. Hamilton; K. R. Sandberg. Microfibrillated Cellulose: Morphology and Accessibility. In J. Appl. Polym. Sci. Appl. Polym. Symp.; (United States) 1983 (Vol. 37, No. CONF-8205234-Vol. 2). ITT Rayonier Inc., Shelton, WA. A. F. Turbak; F. W. Snyder; K. R. Sandberg. Inventors; International Telephone, Telegraph Corporation, Assignee. Microfibrillated Cellulose. United States patent US 4,374,702. Feb 22, 1983. Eichhorn, S. J.; Dufresne, A.; Aranguren, M.; Marcovich, N. E.; Capadona, J. R., et al. Review: Current International Research Into Cellulose Nanofibres and Nanocomposites. J. Mater. Sci. 2010, 45 (1), 133. Kvien, I.; Tanem, B. S.; Oksman, K. Characterization of Cellulose Whiskers and Their Nanocomposites by Atomic Force and Electron Microscopy. Biomacromolecules 2005, 6 (6), 31603165. Angles, M. N.; Dufresne, A. Plasticized Starch/tunicin Whiskers Nanocomposites. 1. Structural Analysis. Macromolecules 2000, 33 (22), 83448353. Fleming, K.; Gray, D.; Prasannan, S.; Matthews, S. Cellulose Crystallites: A New and Robust Liquid Crystalline Medium for the Measurement of Residual Dipolar Couplings. J. Am. Chem. Soc. 2000, 122 (21), 52245225. Habibi, Y.; Goffin, A. L.; Schiltz, N.; Duquesne, E.; Dubois, P.; Dufresne, A. Bionanocomposites Based on Poly (ε-caprolactone)-grafted Cellulose Nanocrystals by Ring-opening Polymerization. J. Mater. Chem. 2008, 18 (41), 50025010. Siqueira, G.; Bras, J.; Dufresne, A. Cellulose Whiskers versus Microfibrils: Influence of the Nature of the Nanoparticle and Its Surface Functionalization on the Thermal and Mechanical Properties of Nanocomposites. Biomacromolecules 2008, 10 (2), 425432. Gousse´, C.; Chanzy, H.; Excoffier, G.; Soubeyrand, L.; Fleury, E. Stable Suspensions of Partially Silylated Cellulose Whiskers Dispersed in Organic Solvents. Polymer 2002, 43 (9), 26452651. Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood., J. Cellulose Nanomaterials Review: Structure, Properties and Nanocomposites. Chem. Soc. Rev. 2011, 40 (7), 39413994. Yanjun, X.; Callum, A. S.; Hill, Z.; Xiao, H.; Militz, C.; Mai, S. Coupling Agents Used for Natural Fiber/polymer Composites: A Review. Compos. Part A 2010, 41, 806819. Mukhopadhyay, S.; Fangueiro, R. Physical Modification of Natural Fibers and Thermoplastic Films for Composites A Review. J. Thermoplast. Compos. Mater. 2009, 22 (2), 135162. Morent, R.; De Geyter, N.; Verschuren, J.; De Clerck, K.; Kiekens, P.; Leys, C. Nonthermal Plasma Treatment of Textiles. Surf. Coat. Technol. 2008, 202, 34273449. Relves, C.; Rana, S.; Fangueiro, R.; Castro, G. Characterization of Physical, Mechanical and Chemical Properties of Quiscal Fibres: The Influence of Atmospheric DBD Plasma Treatment. Plasma Chem. Plasma Process. 2015, 35 (5), 863878. Tsafack, M. J.; LevaloisGru¨tzmacher, J. Towards Multifunctional Surfaces Using the Plasmainduced Graftpolymerization (PIGP) Process: Flame and Waterproof Cotton Textiles. Surf. Coat. Technol. 2007, 201, 57895795. Kanchit Kamlangkla, S. K.; Hodak, J. LevaloisGru¨tzmacher, Multifunctional Silk Fabrics by Means of the Plasma Induced Graft Polymerization (PIGP) Process. Surf. Coat. Technol. 2011, 205, 37553762. Jimenez, M.; Bellayer, S.; Duquesne, S.; Bourbigot, S. Improvement of Heat Resistance of High Performance Fibers Using a Cold Plasma Polymerization Process. Surf. Coat. Technol. 2010, 205, 745758.

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

377

34. Ji, Y. Y.; Hong, Y. C.; Lee, S. H.; Kim, S. D.; Kim, S. S. Formation of Superhydrophobic and Waterrepellency Surface with Hexamethyldisiloxane (HMDSO) Coating on Polyethyleneteraphtalate Fiber by Atmosperic Pressure Plasma Polymerization. Surf. Coat. Technol. 2008, 202, 56635667. 35. Noishiki, Y.; Nishiyama, Y.; Wada, M.; Kuga, S.; Magoshi, J. Mechanical Properties of Silk FibroinMicrocrystalline Cellulose Composite Films. J. Appl. Polym. Sci. 2002, 86 (13), 34253429. 36. Svagan, A. J.; Hedenqvist, M. S.; Berglund, L. Reduced Water Vapour Sorption in Cellulose Nanocomposites with Starch Matrix. Compos. Sci. Technol. 2009, 69 (3), 500506. 37. Viguie´, J.; Molina-Boisseau, S.; Dufresne, A. Processing and Characterization of Waxy Maize Starch Films Plasticized by Sorbitol and Reinforced with Starch Nanocrystals. Macromolecular Bio. 2007, 7 (11), 12061216. 38. Kvien, I.; Sugiyama, J.; Votrubec, M.; Oksman, K. Characterization of Starch Based Nanocomposites. J. Mater. Sci. 2007, 42 (19), 81638171. 39. Angellier, H.; Molina-Boisseau, S.; Dole, P.; Dufresne, A. Thermoplastic Starch-waxy Maize Starch Nanocrystals Nanocomposites. Biomacromolecules 2006, 7 (2), 531539. 40. Orts, W. J.; Shey, J.; Imam, S. H.; Glenn, G. M.; Guttman, M. E.; Revol, J. F. Application of Cellulose Microfibrils in Polymer Nanocomposites. J. Polym. Environ. 2005, 13 (4), 301306. 41. Voronova, M. I.; Zakharov, A. G.; Kuznetsov, O. Y.; Surov, O. V. The Effect of Drying Technique of Nanocellulose Dispersions on Properties of Dried Materials. Mater. Lett. 2012, 68, 164167. 42. Roy, D.; Semsarilar, M.; Guthrie, J. T.; Perrier, S. Cellulose Modification by Polymer Grafting: A Review. Chem. Soc. Rev. 2009, 38 (7), 20462064. 43. Heux, L.; Chauve, G.; Bonini, C. Nonflocculating and Chiral-nematic Self-ordering of Cellulose Microcrystals Suspensions in Nonpolar Solvents. Langmuir 2000, 16 (21), 82108212. 44. Ljungberg, N.; Bonini, C.; Bortolussi, F.; Boisson, C.; Heux, L.; Cavaille´, J. Y. New Nanocomposite Materials Reinforced with Cellulose Whiskers in Atactic Polypropylene: Effect of Surface and Dispersion Characteristics. Biomacromolecules 2005, 6 (5), 27322739. 45. Fortunati, E.; Armentano, I.; Zhou, Q.; Iannoni, A.; Saino, E.; Visai, L.; Berglund, L. A.; , et al.Kenny, J. M. Multifunctional Bionanocomposite Films of Poly (lactic acid), Cellulose Nanocrystals and Silver Nanoparticles. Carbohydr. Polym. 2012, 87 (2), 15961605. 46. Kaboorani, A.; Riedl, B. Surface Modification of Cellulose Nanocrystals (CNC) by A Cationic Surfactant. Ind. Crops Prod. 2015, 65, 4555. 47. Salajkova´, M.; Berglund, L. A.; Zhou, Q. Hydrophobic Cellulose Nanocrystals Modified with Quaternary Ammonium Salts. J. Mater. Chem. 2012, 22 (37), 1979819805. 48. J.J. Beaupre´. Chemically Enhanced Water Removal In Papermaking, A Dissertation Submitted in Fulfillment of the Requirements for the Degree in the Doctor of Philosophy (in Chemical Engineering), University of Maine, Orono, ME (May 2012). 49. Padalkar, S.; Capadona, J. R.; Rowan, S. J.; Weder, C.; Won, Y. H.; Stanciu, L. A.; Moon, R. J. Natural Biopolymers: Novel Templates for the Synthesis of Nanostructures. Langmuir 2010, 26 (11), 84978502. 50. E. Fairman. Avoiding Aggregation During Drying and Rehydration of Nanocellulose. Honors Thesis. University of Maine, May 2014. 51. Hu, Z.; Ballinger, S.; Pelton, R.; Cranston, E. D. Surfactant-enhanced Cellulose Nanocrystal Pickering Emulsions. J. Colloid Interface Sci. 2015, 439, 139148.

378 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

52. Kim, J.; Montero, G.; Habibi, Y.; Hinestroza, J. P.; Genzer, J.; Argyropoulos, D. S.; Rojas, O. J. Dispersion of Cellulose Crystallites by Nonionic Surfactants in a Hydrophobic Polymer Matrix. Polym. Eng. Sci. 2009, 49 (10), 20542061. 53. Emami, Z.; Meng, Q.; Pircheraghi, G.; Manas-Zloczower, I. Use of Surfactants in Cellulose Nanowhisker/epoxy Nanocomposites: Effect on Filler Dispersion and System Properties. Cellulose 2015, 22 (5), 31613176. 54. Sassi, J. F.; Chanzy, H. Ultrastructural Aspects of the Acetylation of Cellulose. Cellulose 1995, 2 (2), 111127. 55. Yuan, H.; Nishiyama, Y.; Wada, M.; Kuga, S. Surface Acylation of Cellulose Whiskers by Drying Aqueous Emulsion. Biomacromolecules 2006, 7 (3), 696700. ¨ zmen, N.; Henry, N.; Harper, D.; Dadmun, M.; Se`be, G., et al. 56. C ¸ etin, N. S.; Tingaut, P.; O Acetylation of Cellulose Nanowhiskers with Vinyl Acetate Under Moderate Conditions. Macromol. Biosci 2009, 9 (10), 9971003. 57. Braun, B.; Dorgan, J. R. Single-step Method for the Isolation and Surface Functionalization of Cellulosic Nanowhiskers. Biomacromolecules 2008, 10 (2), 334341. 58. Berlioz, S.; Molina-Boisseau, S.; Nishiyama, Y.; Heux, L. Gas-phase Surface Esterification of Cellulose Microfibrils and Whiskers. Biomacromolecules 2009, 10 (8), 21442151. 59. de Menezes, A. J.; Siqueira, G.; Curvelo, A. A.; Dufresne, A. Extrusion and Characterization of Functionalized Cellulose Whiskers Reinforced Polyethylene Nanocomposites. Polymer 2009, 50 (19), 45524563. 60. Hasani, M.; Cranston, E. D.; Westman, G.; Gray, D. G. Cationic Surface Functionalization of Cellulose Nanocrystals. Soft Matter. 2008, 4 (11), 22382244. 61. Dong, S.; Roman, M. Fluorescently Labeled Cellulose Nanocrystals for Bioimaging Applications. J. Am. Chem. Soc. 2007, 129 (45), 1381013811. 62. Gousse´, C.; Chanzy, H.; Excoffier, G.; Soubeyrand, L.; Fleury, E. Stable Suspensions of Partially Silylated Cellulose Whiskers Dispersed in Organic Solvents. Polymer 2002, 43 (9), 26452651. 63. Li, B.; Xu, W.; Kronlund, D.; Ma¨a¨tta¨nen, A.; Liu, J.; Sma˚tt, J. H.; Peltonen, J.; Willfo¨r, S.; Mu, X.; Xu, C. Cellulose Nanocrystals Prepared via Formic Acid Hydrolysis Followed by TEMPO-mediated Oxidation. Carbohydr. Polym. 2015, 133, 605612. 64. Saito, T.; Isogai, A. TEMPO-mediated Oxidation of Native Cellulose. The Effect of Oxidation Conditions on Chemical and Crystal Structures of the Water-insoluble Fractions. Biomacromolecules 2004, 5 (5), 19831989. 65. Saito, T.; Hirota, M.; Tamura, N.; Isogai, A. Oxidation of Bleached Wood Pulp by TEMPO/NaClO/NaClO2 System: Effect of the Oxidation Conditions on Carboxylate Content and Degree of Polymerization. J. Wood Sci. 2010, 56 (3), 227232. 66. Habibi, Y.; Chanzy, H.; Vignon, M. R. TEMPO-mediated Surface Oxidation of Cellulose Whiskers. Cellulose 2006, 13 (6), 679687. 67. Montanari, S.; Roumani, M.; Heux, L.; Vignon, M. R. Topochemistry of Carboxylated Cellulose Nanocrystals Resulting from TEMPO-mediated Oxidation. Macromolecules 2005, 38 (5), 16651671. 68. Araki, J.; Wada, M.; Kuga, S.; Okano, T. Flow Properties of Microcrystalline Cellulose Suspension Prepared by Acid Treatment of Native Cellulose. Colloids Surf. A Physicochem. Eng. Asp. 1998, 142 (1), 7582. 69. Hoeng, F.; Denneulin, A.; Neuman, C.; Bras, J. Charge Density Modification of Carboxylated Cellulose Nanocrystals for Stable Silver Nanoparticles Suspension Preparation. J. Nanopart. Res. 2015, 17 (6), 14.

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

379

70. Cao, X.; Habibi, Y.; Lucia, L. A. One-pot Polymerization, Surface Grafting, and Processing of Waterborne Polyurethane-cellulose Nanocrystal Nanocomposites. J. Mater. Chem. 2009, 19 (38), 71377145. 71. Mangalam, A. P.; Simonsen, J.; Benight, A. S. Cellulose/DNA Hybrid Nanomaterials. Biomacromolecules 2009, 10 (3), 497504. 72. Azzam, F.; Heux, L.; Putaux, J. L.; Jean, B. Preparation by Grafting Onto, Characterization, and Properties of Thermally Responsive Polymer-decorated Cellulose Nanocrystals. Biomacromolecules 2010, 11 (12), 36523659. 73. Wang, J. S.; Matyjaszewski, K. Controlled/“living” Radical Polymerization. Atom Transfer Radical Polymerization in the Presence of Transition-metal Complexes. J. Am. Chem. Soc 1995, 117 (20), 56145615. 74. Morandi, G.; Heath, L.; Thielemans, W. Cellulose Nanocrystals Grafted with Polystyrene Chains through Surface-initiated Atom Transfer Radical Polymerization (SIATRP). Langmuir 2009, 25 (14), 82808286. 75. Yi, J.; Xu, Q.; Zhang, X.; Zhang, H. Chiral-nematic Self-ordering of Rodlike Cellulose Nanocrystals Grafted with Poly (styrene) in both Thermotropic and Lyotropic States. Polymer 2008, 49 (20), 44064412. 76. Yi, J.; Xu, Q.; Zhang, X.; Zhang, H. Temperature-induced Chiral Nematic Phase Changes of Suspensions of Poly (N, N-dimethylaminoethyl methacrylate)-grafted Cellulose Nanocrystals. Cellulose 2009, 16 (6), 989997. 77. Xu, Q.; Yi, J.; Zhang, X.; Zhang, H. A Novel Amphotropic Polymer based on Cellulose Nanocrystals Grafted with Azo Polymers. Eur. Polym. J. 2008, 44 (9), 28302837. 78. Chen, G.; Dufresne, A.; Huang, J.; Chang, P. R. A Novel Thermoformable Bionanocomposite based on Cellulose Nanocrystal-graft-poly (ε-caprolactone). Macromol. Mater. Eng. 2009, 294 (1), 5967. 79. Lin, N.; Chen, G.; Huang, J.; Dufresne, A.; Chang, P. R. Effects of Polymer-grafted Natural Nanocrystals on the Structure and Mechanical Properties of Poly (Lactic Acid): A Case of Cellulose Whisker-graft-polycaprolactone. J. Appl. Polym. Sci. 2009, 113 (5), 34173425. 80. Siqueira, G.; Bras, J.; Dufresne, A. New Process of Chemical Grafting of Cellulose Nanoparticles with a Long Chain Isocyanate. Langmuir 2009, 26 (1), 402411. 81. Hoyos, C. G.; Cristia, E.; Va´zquez, A. Effect of Cellulose Microcrystalline Particles on Properties of Cement Based Composites. Mater. Des. 2013, 51, 810818. 82. Cao, Y.; Zavaterri, P.; Youngblood, J.; Moon, R.; Weiss, J. The Influence of Cellulose Nanocrystal Additions on the Performance of Cement Paste. Cem. Concr. Compos 2015, 56, 7383. 83. Mansur, M. A.; Aziz, M. A. A Study on Jute Fibre Reinforced Cement Composites. Int. J. Cem. Compos. Lightweight Concr. 1982, 4 (2), 7582. 84. Savastano, H.; Agopyan, V.; Nolasco, A. M.; Pimentel, L. Plant Fibre Reinforced Cement Components for Roofing. Constr. Build. Mater. 1999, 13, 433438. 85. Onuaguluchi, O.; Banthia, N. Plant-based Natural Fiber Reinforced Cement Composites: A Review. J. Cem. Concr. Compos. 2016, 68, 96108. 86. Bilba, K.; Arsene, M.-A.; Quensanga, A. Sugar Cane Bagasse Fibre Reinforced Cement Composites. Part I. Influence of the Botanical Components of Bagasse on the Setting of Bagasse/cement Composite. Cem. Concr. Compos. 2003, 25, 9196. 87. Sudin, R.; Swamy, N. Bamboo and Wood Fibre Cement Composites for Sustainable Infrastructure Regeneration. J. Mater. Sci. 2006, 41, 69176924. 88. Sedan, D.; Pagnoux, C.; Smith, A.; Chotard, T. Mechanical Properties of Hemp Fibre Reinforced Cement: Influence of the Fibre/matrix Interaction. J. Eur. Ceram. 2008, 28, 183192.

380 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

89. Fan, M.; Ndikontar, M. K.; Zhou, X.; Ngamveng, J. N. Cement-bonded Composites Made from Tropical Woods: Compatibility of Wood and Cement. Constr. Build. Mater. 2012, 36, 135140. 90. Vaickelionis, G.; Vaickelioniene, R. Cemeny Hydration in the Presence of Wood Extractives and Pozzolan Mineral Admixtures. Ceram. Silik 2006, 50 (2), 115122. 91. Sanjuan, M. A.; Filho, R. D. Toledo Effectiveness of Crack Control at Early Age on the Corrosion of Steel Bars in Low Modulus Sisal and Coconut Fibre-reinforced Mortars. Cem. Concr. Res. 1998, 28 (4), 555565. 92. Toledo Filho, R. D.; Sanjuan, M. A. Effect of Low Modulus Sisal and Polypropylene Fibre on the Free and Restrained Shrinkage of Mortars at Early Age. Cem. Concr. Res. 1999, 29 (10), 15971604. 93. Toledo Filho, R. D.; Ghavami, K.; Sanjuan, M. A.; England, G. L. Free, Restrained and Drying Shrinkage of Cement Mortar Composites Reinforced with Vegetable Fibres. Cem. Concr. Compos. 2005, 27, 537546. 94. Boghossian, E.; Wegner, L. D. Use of Flax Fibres to Reduce Plastic Shrinkage Cracking in Concrete. Cem. Concr. Compos. 2008, 30, 929937. 95. Soleimani, T.; Merati, A.K.; Latifi, M.; Ramezanianpor, A.K. Inhibition of Cracks on the Surface of Cement Mortar Using Estabragh Fibers, Adv. Mater. Sci. Eng. 2013, 5. Article ID 656109, doi:10.1155/2013/656109. 96. Silva, F. A.; Toledo Filho, R. D.; Melo Filho, J. A.; Fairbairn, E. M. R. Physical and Mechanical Properties of Durable Sisal Fiber-cement Composites. Constr. Build. Mater. 2010, 24, 777785. 97. Claramunt, J.; Lucı´a, J. F.; Ventura, H.; Ardanuy, M. Natural Fiber Nonwoven Reinforced Cement Composites as Sustainable Materials for Building Envelopes. Constr. Build. Mater. 2016, 115, 230239. 98. Yan, L.; Chouw, N.; Huang, L.; Kasal, B. Effect of Alkali Treatment on Microstructure and Mechanical Properties of Coir Fibres, Coir Fibre Reinforcedpolymer Composites and Reinforced-cementitious Composites. Constr. Build. Mater. 2016, 112, 168182. 99. Dai, D.; Fan, M. Wood Fibres as Reinforcements in Natural Fibre Composites: Structure, Properties, Processing and Applications. Nat. Fibre Compos. Mater. Process. Propert 2014, 365. 100. Ma´rmol, G.; Santos, S. F.; Savastano, H.; Borrachero, M. V.; Monzo´, J.; Paya´, J. Mechanical and Physical Performance of Low Alkalinity Cementitious Composites Reinforced with Recycled Cellulosic Fibres Pulp from Cement Kraft Bags. Ind. Crops Prod. 2013, 49, 422427. 101. Xie, X.; Zhou, Z.; Jiang, M.; Xu, X.; Wang, Z.; Hui, D. Cellulosic Fibers from Rice Straw and Bamboo Used as Reinforcement of Cement-based Composites for Remarkably Improving Mechanical Properties. Compos. Part B Eng. 2015, 78, 153161. 102. Jiang, D.; Cui, S.; Xu, F.; Tuo, T. Impact of Leaf Fibre Modification Methods on Compatibility between Leaf Fibres and Cement-based Materials. Constr. Build. Mater. 2015, 94, 502512. 103. Hwang, C.-L.; Tran, V.-A.; Hong, J.-W.; Hsieh, Y.-C. Effects of Short Coconut Fiber on the Mechanical Properties, Plastic Cracking Behavior, and Impact Resistance of Cementitious Composites. Constr. Build. Mater. 2016, 127, 984992. 104. Silva, A.; Barzin Mobasher, F.; Filho, R. D. T. Fatigue Behavior of Sisal Fiber Reinforced Cement Composites. Mater. Sci. Eng. A 2010, 527 (C1), 55075513. 105. Silva, A.; Barzin Mobasher, F.; Soranakom, C.; Filho, R. D. T. Effect of Fiber Shape and Morphology on Interfacial Bond and Cracking Behaviors of Sisal Fiber Cement Based Composites. Cem. Concr. Compos. 2011, 33 (8), 814823.

Macro- and nanodimensional plant fiber reinforcements for cementitious composites

381

106. Jarabo, R.; Monte, M. A. C.; Blanco, A.; Negro, C.; Tijero, J. Characterisation of Agricultural Residues Used as a Source of Fibres for Fibre-cement Production. Ind. Crops Prod. 2012, 36 (1), 1421. ¨ . Banana FibersVariability 107. Mukhopadhyay, S.; Fangueiro, R.; Arpac, Y.; Sentu ¸ ¨ rk, U and Fracture Behaviour. J. Eng. Fibers Fabr. 2008, 3 (2), 3945. 108. Ce´lino, A.; Fre´our, S.; Jacquemin, F.; Casari, P. The Hygroscopic Behavior of Plant Fibers: A Review. Front. Chem. 2013, 1. 109. Thomason, J. L.; Carruthers, J.; Kelly, J.; Johnson, G. Fibre Cross-section Determination and Variability in Sisal and Flax and Its Effects on Fibre Performance Characterisation. Compos. Sci. Technol. 2011, 71 (7), 10081015. 110. Pacheco-Torgal, F.; Jalali, S. Cementitious Building Materials Reinforced with Vegetable Fibres: A Review. Constr. Build. Mater. 2011, 25, 575581. 111. Khare, L. Performance Evaluation of Bamboo Reinforced Concrete Beams. MS Thesis, Civil & Environmental Engineering; University of Texas: Arlington, TX, 2007. 112. Mohr, B. J.; Nanko, H.; Kurtis, K. E. Durability of Kraft Pulp Fiber-cement Composites to Wet/dry Cycling. Cem. Concr. Compos. 2005, 27, 435448. 113. Ramakrishna, G.; Sundararajan, T. Studies on the Durability of Fibres and the Effect of Corroded Fibres on the Strength of Mortar. Cem. Concr. Compos. 2005, 27, 575582. 114. Toledo Filho, R. D.; Scrivener, K.; England, G. L.; Ghavami, K. Durability of Alkali Sensitive Sisal and Coconut Fibres in Cement Mortar Composites. Cem. Concr. Compos. 2000, 22, 127143. 115. Roma, L. C., Jr.; Martello, L. S.; Savastano, H., Jr. Evaluation of Mechanical, Physical and Thermal Performance of Cement-based Tiles Reinforced with Vegetable Fibers. Constr. Build. Mater. 2008, 22, 668674. 116. Toledo Filho, R. D.; Ghavami, K.; England, G. L.; Scrivener, K. Development of Vegetable Fibre Mortar Composites of Improved Durability. Cem. Concr. Compos. 2003, 25, 185196. 117. Mohr, B.; Biernacki, J.; Kurtis, K. Supplementary Cementitious Materials for Mitigating Degradation of Kraft Pulp Fiber Cement-composites. Cem. Concr. Res. 2007, 37, 15311543. 118. Toledo Filho, R. D.; Silva, F. A. S.; Fairbairn, E. M. R.; Melo Filho, J. A. Durability of Compression Molded Sisal Fiber Reinforced Mortar Laminates. Constr. Build. Mater. 2009, 23, 24092420. 119. Silva, F. A.; Toledo Filho, R. D.; Melo Filho, J. A.; Fairbairn, E. M. R. Physical and Mechanical Properties of Durable Sisal Fiberecement Composites. Constr. Build. Mater. 2010, 24, 777785. 120. John, V. M.; Cincotto, M. A.; Sjostrom, C.; Agopyan, V.; Oliveira, C. T. A. Durability of Blast Furnace Slag-based Cement Mortar Reinforced with Coir Fibres. Cem. Concr. Compos. 2005, 27, 565574. 121. Bledzki, A. K.; Gassan, J. Composites Reinforced with Cellulose Based Fibres. Prog. Polym. Sci. 1999, 24, 221274. 122. Li, Z.; Wang, L.; Wang, X. Flexural Characteristics of Coir Fiber Reinforced Cementitious Composites. Fibers Polym. 2006, 7 (3), 286294. 123. Shukla, S. R.; Pai, R. S.; Shendarkar, A. D. Adsorption of Ni (II), Zn (II) and Fe (II) on Modified Coir Fibres. Sep. Purif. Technol. 2006, 47, 141147. 124. Tonoli, G. H. D.; Belgacem, M. N.; Bras, J.; Pereira-da-Silva, M. A.; Lahr, F. A. R.; Savastano, H., Jr. Impact of Bleaching Pine Fibre on the Fibre/cement Interface. J. Mater. Sci. 2012, 47, 41674177. 125. Coutts, R. S. P. A Review of Australian Research into Natural Fibre Cement Composites. Cem. Concr. Compos. 2005, 27, 518526.

382 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

126. Mohr, B. J.; Biernacki, J. J.; Kurtis, K. E. Microstructural and Chemical Effects of Wet/dry Cycling on Pulp Fiber-cement Composites. Cem. Concr. Res. 2006, 36 (7), 12401251. 127. Savastano, H., Jr.; Warden, P. G.; Coutts, R. S. P. Mechanically Pulped Sisal as Reinforcement in Cementitious Matrices. Cem. Concr. Compos. 2003, 25, 311319. 128. Tonoli, G.; Joaquim, A.; Arsene, M.; Bilba, K.; Savastano, H. Performance and Durability of Cement based Composites Reinforced with Refined Sisal Pulp. Mater. Manuf. Process 2007, 22, 149156. 129. Ferreira, S. R.; Lima, P. R.; Silva, F. A.; Toledo Filho, R. D. Effect of Sisal Fiber Hornification on the Fiber-matrix Bonding Characteristics and Bending Behavior of Cement Based Composites. Key Eng. Mater. 2014, 600, 421432. 130. Claramunt, J.; Ardanuy, M.; Garcı´a-Hortal, J. A. Effect of Drying and Rewetting Cycles on the Structure and Physicochemical Characteristics of Softwood Fibres for Reinforcement of Cementitious Composites. Carbohydr. Polym. 2010, 79, 200205. 131. Claramunt, J.; Ardanuy, M.; Garcı´a-Hortal, J. A.; Toledo Filho, R. D. The Hornification of Vegetable Fibers to Improve the Durability of Cement Mortar Composites. Cem. Concr. Compos. 2011, 33, 586595. 132. Arsene, M.; Okwo, A.; Bilba, K.; Soboyejo, A.; Soboyejo, W. Chemically and Thermally Treated Vegetable Fibers for Reinforcement of Cement-based Composites. Mater. Manuf. Process 2007, 22 (2), 214227. 133. Rostami, V.; Shao, Y.; Boyd, A. J. Durability of Concrete Pipes Subjected to Combined Steam and Carbonation Curing. Constr. Build. Mater. 2011, 25, 33453355. 134. Rostami, V.; Shao, Y.; Boyd, A. J. Durability of Concrete Pipes Subjected to Combined Steam and Carbonation Curing. Cem. Concr. Res. 2012, 42 (1), 186193. 135. Tonoli, G. H. D.; Santos, S. F.; Joaquim, A. P.; Savastano, H., Jr. Effect of Accelerated Carbonation on Cementitious Roofing Tiles Reinforced with Lignocellulosic Fibre. Constr. Build. Mater. 2010, 24, 193201. 136. Soroushian, P.; Wonb, J.; Hassan, M. Durability Characteristics of CO2-cured Cellulose Fiber Reinforced Cement Composites. Constr. Build. Mater. 2012, 34, 4453. 137. Almeida, A. E. F. S.; Tonoli, G. H. D.; Santos, S. F.; Savastano, H., Jr. Improved Durability of Vegetable Fiber Reinforced Cement Composite Subject to Accelerated Carbonation at Early Age. Cem. Concr. Compos. 2013, 42, 4958. 138. Sohel Rana, R.; Alagirusamy.; Joshi, Mangala A Review on Carbon Epoxy Nanocomposites. J. Reinf. Plast. Compos. 2009, 28 (4), 461487. 139. S. Parveen; S. Rana; R. Fangueiro; M.C. Paiva. 2016. A Novel Approach of Developing Micro Crystalline Cellulose Reinforced Cementitious Composites with Enhanced Microstructure and Mechanical Performance. Cem. Concr. Compos. 78, 2017, 146161. 140. Binici, H.; Aksogan, O.; Bodur, M. N.; Akca, E.; Kapur, S. Thermal Isolation and Mechanical Properties of Fibre Reinforced Mud Bricks as Wall Materials. Constr. Build. Mater. 2007, 21, 901906. 141. Fan, M.; Fu, F. Advanced High Strength Natural Fibre Composites in Construction; Woodhead Publishing: Cambridge, 2016513. 142. Fortunati, E.; Peltzer, M.; Armentano, I.; Torre, L.; Jimenez, A.; Kenny, J. M. Effects of Modified Cellulose Nanocrystals on the Barrier and Migration Properties of PLA Nano-biocomposites. Carbohydr. Polym. 2012, 90, 948956. 143. Shehata, M.; Navarra, M.; Klement, T.; Lachemi, M.; Schell, H. Use of Wet Cellulose to cure Shotcrete Repairs on Bridge Soffits. Part 1: Field Trials and Observations. Can. J. Civ. Eng. 2006, 33, 807814.

Using vegetable fiber nonwovens cement composites as sustainable materials for applications on ventilated fac¸ade systems

14

Josep Claramunt1 and Mo`nica Ardanuy 2 1 Universitat Polite`cnica de Catalunya, Castelldefels, Spain, 2Universitat Polite`cnica de Catalunya, Terrassa, Spain

14.1

Introduction

It is widely known that the building sector is one of the main areas responsible for energy consumption and CO2 emissions worldwide. Specifically, in the case of Europe, around 40% of the European Union’s total final energy consumption and 36% of its total CO2 emissions are due to the building sector.1 To improve the energy efficiency of the building sector requires action in both the manufacture of the building materials—that should have higher performance and be more sustainable and eco-friendly—and the constructive solutions—that should allow to reduce the energy consumption during buildings’ lifetime.2 During the lifetime of a building, the energy consumed is derived mainly from air conditioning in hotter or colder periods. The design of new construction solutions combined with adequate energy consumption management can lead to a significant reduction of the energy expenditure for lighting and air conditioning. More concretely, the construction solution known as the “ventilated fac¸ade” system has experienced a significant increase in use in Europe, especially the Mediterranean countries, due to both its good performance and its ease of construction.3 Ventilated fac¸ades are multilayered building envelopes consisting of an outer layer mechanically connected to an inner layer and a ventilated air gap that usually contains thermal insulation in contact with the inner layer. The material used for this outer layer has to fulfill certain requirements such as strength, flexibility, ductility, lightness, permeability, thermal and acoustic insulation, and durability, among others. Currently, the materials most commonly used for these envelopes are ceramics, natural stones, wood-resin and aluminum-resin composites, and, increasingly, fiber cements. Each of these materials has advantages with regard to some requirements but not for all of them. For example, ceramics and natural stones have excessive weight and stiffness, which limits their size and necessitates a complex supporting structure. Furthermore, partial breaking, which can lead to objects falling Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites. DOI: http://dx.doi.org/10.1016/B978-0-08-102001-2.00016-4 © 2017 Elsevier Ltd. All rights reserved.

386 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

on public roads, can be dangerous. Wood and aluminum composites are more flexible and lightweight but are less durable, less hard, and are much more expensive and less sustainable compared to conventional building materials. For these reasons, it is of practical importance to develop new materials for envelopes with the maximum aforementioned characteristics, primarily strength, ductility, flexibility, and durability, using environmentally friendly and low-cost raw materials and processes.4 The use of vegetable fibers to reinforce brittle matrices, such as cement, constitutes an interesting possibility that offers many advantages with respect to the utilization of other fibers or reinforcements. On the one hand, due to their mechanical properties, vegetable fibers can improve the ductility, flexibility, and crack resistance of the resulting material. On the other hand, and in the case of precasting materials such as fiber-cements, the use of vegetable fibers has emerged during the last decades as an interesting option to replace asbestos, allowing the development of materials with good performance at relatively low cost.5 8 Moreover, vegetable fibers are nonhazardous, renewable, and biodegradable, allowing the development of more sustainable construction materials. Many works describing the use of cellulose-based fibers as reinforcement for cement-based composites have been published.6,9 16 Nonetheless, in most of these papers the fibers are used in pulp or staple forms, limiting the improvement of the flexural strength and ductility of the composites due to the short length of the fibers and the maximum quantity that it is possible to mix with the cement matrix (around 4 8 wt.%). In order to overcome these problems, Toledo Filho et al.17 used sisal strands in the form of unidirectional fabrics to reinforce cement-based composites, achieving significant improvements in the tensile strength and toughness.18,19 Despite the interesting results obtained with these strands, it is difficult to make possible the industrial production of these composites using an automatized laminating process. In this sense, the use of textile preforms like nonwovens could bring benefits such as fabric handability and easier applicability in an automatized process. The problem with producing these composites lies in the difficulty of the infiltration of the cement through the textile structure. For this reason, it is important to design nonwoven structures and optimized procedures to allow a good infiltration of the cement through the reinforcement. Recently, we have developed a successful procedure to produce flax fiber nonwovens reinforced cement composites with high ductility.20 In this chapter, we present the procedure, methods of characterization, and mechanical performance of these materials.

14.2

Processing

14.2.1 Preparation of the reinforcement To prepare the nonwoven fabrics, the natural fibers with lengths around 6 cm need to be first opened and carded to form a thin web. Then, these thin webs are laid by the cross-laying method to form batts. Finally, these batts should be consolidated

Vegetable fiber nonwovens cement composites

387

Figure 14.1 Image of flax fibers in staple form (left) and the resulting nonwoven fabric (right).

by needle-punching to form nonwoven mats. Fig. 14.1 shows an image of the fibers and the nonwoven obtained. In order to obtain optimized nonwoven structures that can allow a good infiltration of the cement, the nonwovens can be designed with different entanglements, thicknesses, and weights. A detailed description of the machine parameters that can be used to prepare these nonwovens is described in Ventura et al.21 One of the main drawbacks of the use of vegetable fibers to reinforce cement matrix is the loss of mechanical properties after accelerated aging. It is well known that this lack of durability is mainly caused, on the one hand, by the calcium hydroxide (Portlandite) of the matrix, which degrades the fibers, and, on the other, by changes in environmental moisture, which induce dimensional changes in the vegetable fibers and hence a loss of physical contact with the matrix.11,22,23 To minimize the dimensional changes of the fibers, chemical or physical modifications can be carried out. One successful treatment is to subject the fibers to wetting cycles in water followed by drying cycles, causing shrinkage of the fibers and a reduction in water-absorption desorption. This simple and ecofriendly treatment, which uses water as the reagent, has been used successfully to obtain more durable cement composites reinforced with cellulose pulps14,24 or sisal strands.25 Moreover, these wet dry treatments can be carried out under different conditions to minimize the water absorbance of the fibers and hence to improve the durability of the composites.

14.2.2 Preparation of the matrix The preparation of the matrix must take into account various conditions. Thus, apart from minimizing the calcium hydroxide content to avoid degradation of the fibers, it is necessary to have a suitable fluidization and fineness of the particles to allow a good infiltration of the cement through the nonwoven. For example, as can be seen in Fig. 14.2 (left), the particles of sand are too big to allow a good infiltration of the cement through the nonwoven fabric. Nonetheless, for a matrix with finer sand it is possible to prepare composites with very good impregnation of the nonwoven (Fig. 14.2, right)

388 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

Figure 14.2 BSE SEM images of cement composites reinforced with flax nonwovens. 1- Preparation of the nonwovens with desired size

2- Soaking in cement matrix

4- Pressing 3- Molding and dewatering with vacuum

Figure 14.3 Scheme of the preparation of the composites.

On the other hand, to prepare composites with good infiltration, the initial water/ cement ratio has to be high, around 0.75, since the vegetable fibers absorb a lot of water.

14.2.3 Composite processing The laboratory-scale procedure to prepare cement composites with nonwoven preforms follows the scheme presented in Fig. 14.3. The first step to prepare the composites is to cut the nonwoven preforms. Then the nonwovens must be immersed in the cement paste, wetting the fabric with the help of a roller. After this, the wetted nonwovens must be placed cross-oriented on a mold to allow an isotropic reinforcement in the x and y directions. Moreover, one protective outer layer without fibers of around 1 mm thickness must also be placed on every face of the samples. This outer layer can contain a high percentage of sand and dyes or other active products, such as photocatalytic titanium oxide, to improve the self14-cleaning or give other properties to the sheets. Once all the layers are in the mold, a vacuum pump must be connected while a roller compactor

Vegetable fiber nonwovens cement composites

389

is passed to remove the air that may remain between layers and to ensure a good contact between all the nonwoven layers and the cement (Fig. 14.4). Then the composite must be compacted by a press, applying a pressure of around 3 MPa (Fig. 14.5). It is important that the mold apply homogeneous pressure to ensure that a uniform plate thickness is obtained. The pressure must be maintained until there is no further water exudation from the plate (before introducing the cement paste, it is important to place a synthetic paper filter on the bottom to allow retention of the solid paste during drainage). Once demolded, the specimens must be cured for 28 days at 20 C 6 1 C in a humidity chamber (,95% of relative humidity). This procedure is very easy to implement for high volume products at industrial scale, since the nonwovens can be supplied in a roll form and immersed in a continuous process, following similar steps to the Hatschek process.

Figure 14.4 Image of the mold connected to the vacuum pump with the nonwoven layers impregnated with the cement paste.

Figure 14.5 Image of the press for the compaction of the laminate.

390 Sustainable and Nonconventional Construction Materials using Inorganic Bonded Fiber Composites

14.3

Mechanical performance and durability

Usually, the fac¸ade sheets are subjected to bending stresses caused by the wind. This is why most of the tests performed on these plates described in the literature are bending tests made under three or four bending points. Nevertheless, tensile tests allow a better characterization of the adhesion between the matrix and the reinforcement. In both tests, it is possible to determine four parameters that define the behavior of the material: the limit of proportionality stress (LOP), the modulus of rupture (MOR), the modulus of elasticity (MOE), and the energy absorption or toughness. Fig. 14.6 shows a typical curve obtained for cellulose reinforced cementitious composites tested under bending loading. As shown, at the beginning of the curve there is a linear portion with a steep slope. Then, there is an abrupt change of the slope of the curve, being lower until the fracture of the material. The slope of the initial linear portion depends on the nature of the matrix and on the fiber content. After the cracking of the matrix, the strength is transferred to the fibers, showing in the composite as a strain-hardening or strain-softening behavior until the fracture of the sample. The numerical parameters obtained from this stress strain curve allow us to summarize the mechanical behavior of the composite. The LOP value is the maximum va