The Fabrication, Testing and Application of Fibre Cement Boards [1 ed.] 9781527512320, 9781527505766

Citation preview

The Fabrication, Testing and Application of Fibre Cement Boards

The Fabrication, Testing and Application of Fibre Cement Boards Edited by

Zbigniew Ranachowski and Krzysztof Schabowicz

The Fabrication, Testing and Application of Fibre Cement Boards Edited by Zbigniew Ranachowski and Krzysztof Schabowicz This book first published 2018 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2018 by Zbigniew Ranachowski, Krzysztof Schabowicz and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-5275-0576-6 ISBN (13): 978-1-5275-0576-6

CONTENTS

Introduction ................................................................................................ 1 Chapter One ................................................................................................. 7 Fabrication of Fibre Cement Boards By Krzysztof Schabowicz and Tomasz GorzelaĔczyk Introduction ........................................................................................... 7 Primary raw materials ......................................................................... 11 Properties of PVA fibres ..................................................................... 11 Raw materials preparation and mixing zone ........................................ 14 Detailed operation of Hatschek machine and fibre film formation ...... 23 Details of other FCB fabrication process stages .................................. 28 Chapter Two .............................................................................................. 41 Fibre Cement Boards Testing By Zbigniew Ranachowski Mechanical properties of FCB ............................................................ 41 European Standard EN 12467 .............................................................. 42 Measuring board dimensions, tolerances, straightness and squareness of edges ................................................................. 44 Measuring apparent density ................................................................. 45 Measuring mechanical resistance......................................................... 46 Testing water impermeability .............................................................. 56 Testing water vapour permeability by the determination of the water vapour resistance value µ ........................................... 56 Testing moisture content during FCB production process ................... 58 Durability against warm water ............................................................. 65 Durability against soak–dry cycles ...................................................... 66 Moisture movement test ....................................................................... 66 Durability against freeze–thaw cycles ................................................. 68 Durability against heat–rain ................................................................. 69 Reaction to fire..................................................................................... 69 Information on parameters determined based on standard tests and indicated on the product label .................................................. 71 FCB testing using ultrasound ............................................................... 71 Application of Lamb waves for detecting structural defects ................ 78

vi

Contents

By Dariusz Jarząbek Application of advanced optical microscopy ....................................... 89 Examples of application of scanning electron microscopy (SEM) ...... 96 X-ray microtomography (micro-CT) application in visualization of the fibre distribution and crack detection ................................... 98 Chapter Three .......................................................................................... 107 Applications of Fibre Cement Boards By Krzysztof Schabowicz and Tomasz GorzelaĔczyk Introduction........................................................................................ 107 Example applications of fibre cement boards .................................... 107 Conclusions ....................................................................................... 121 References ............................................................................................... 123

INTRODUCTION

Fibre cement board (FCB) is a versatile green building material, strong and durable. It serves as a substitute for natural wood and wood based products such as plywood or oriented strand boards (OSB). The properties of FCB as a construction material make it preferable for use as a ventilated façade cladding for both newly built and renovated buildings, interior wall coverings, balcony balustrade panels, base course and chimney cladding, and enclosure soffit lining. FCB can be applied to unfinished, painted, or simply impregnated surfaces [1-3]. Fibre cement components have been used in construction for over 100 years, mainly as roofing covers in the form of corrugated plates or nonpressurized tubes. They were invented by a Czech engineer, Ludwik Hatschek (1856-1914) [4]. In 1900, he developed and patented a technology for manufacturing light, tough, durable, and non-flammable asbestos cement sheeting that he called “Eternit.” Fibre cement composite–compared to the components of concrete– shows improved toughness, ductility, flexural capacity, and crack resistance. A major advantage of the fibre reinforcement is the behaviour of the composite after cracking has started, as the fibres bridge the matrix crack (matrix-bulk material, in which the fibres are dispersed). Then, the stress is transferred to the high-resistance system of fibres, capable of withstanding the load. Post-cracking toughness allows the use of FCBs in construction. Moreover, the fibres can reduce the unwanted free plastic shrinkage within the structure, decrease the thermal conductivity, and improve the acoustic insulation and fire resistance. For many years asbestos has been regarded as the most suitable material for reinforcing FCBs. White asbestos, a silicate mineral / Mg3(Si2O5)(OH)4 / naturally forms long and durable fibrous crystals with each visible fibre composed of millions of thin microscopic fibrils. It is found in many geological deposits throughout the world. The fibres reinforce FCB components only when added in a specific quantity (approx. 10% wt.) and are uniformly dispersed throughout the cementitious matrix. A highly complex procedure is required to achieve this goal under efficient industrial conditions. Ludwig Hatschek solved the problem by inventing a machine with a rotating sieve and a vat containing a diluted fibre slurry, Portland cement, and mineral components. A thin

2

Introducction

film of an F FCB is formed on a movin ng felt partiallly wrapped arround the sieve, similaar to the proceedure used in paper sheet m making. Startiing in the early 1920s, the mass prroduction of asbestos a FCB B expanded th hroughout the world. C Corrugated rooofing covers can c still be fouund in many buildings, b mostly in ruural areas (see Fig. 1). It is wiidely known that Ludwig g Hatschek ddied from pu ulmonary disease. Thee death of ann Englishwom man, Nellie K Kershaw, in 19 924 from pulmonary asbestosis waas the first caase to be desscribed in thee medical literature [5]]. In the early 1970s, a glob bal effort to leegislate for thee removal of asbestos ffrom industriaal products waas initiated.

Fig. 1. Corruugated roofing made of asbestos FCB, now expected to bee removed due to health risks.

Fibre cem ment was a major m user of assbestos, and aas such new reeinforcing fibres were bbeing sought as a alternativess to asbestos iin this class off building material. Jam mes Hardie and a Coy Pty Ltd. L started m manufacturing g asbestos cement prodducts in Austrralia in 1917. The company ny extended prroduction to New Zeaaland, Indoneesia, Malaysia, and the U US. In the mid-1940s, m during the ppost-war years, when theree was a shortaage of asbesto os, James menced a research on repllacing asbestoos with cellulo Hardie comm ose fibres [6]. The fiibres studied included baagasse, grounndwood, wheat straw, cement bagss, and brown paper. p Howev ver, the work w was discontinu ued when asbestos beccame more acccessible. Reg gardless, the iidea of using cellulose fibres was sstill considereed by the com mpany. Beginnning in the 1960s, the company m manufactured a product called “Hardiflexx” containing 8% (wt.)

The F Fabrication, Teesting and Application of Fibree Cement Board ds

3

asbestos andd 8% (wt.) wood w fibres. In those dayys, no technology was available to maintain com mpatibility beetween the ceellulose fibress and the matrix to create a prooduct of suffficient qualiity. The con nsiderable shortcominggs of the orgaanic fibres aree due to their insufficient long-term l durability inn an alkaline environment e compared to thhe cementitiou us matrix. Water absorrption of the matrix m could cause an alkaali attack on the t fibres resulting inn volume variation, v fib bre fractures , pull-out resistance r degradation,, and mineraliisation of the fibres. f An effecctive method to lower thee alkalinity oof the matrix was the addition of ppozzolanic coomponents in the form of fiine-ground qu uartz sand containing aamorphous siliica. Pozzolaniic silicate in thhe presence of alkaline calcium hyddroxide produuces a hydrated calcium siliicate, a substaance with lower alkalinnity [7]: 3Ca++ + 2H2SiO4 + 2OH 2 - + 2H2O ĺ Ca3[H2Si2 O7](OH)2 · 3H H2O In the laate 1980s, assbestos fibres were successsfully replaceed in the following prroducts: flat sheet, s corrugaated roofing, and moulded products by the leadinng manufactuurers, including g James Harddie [6] and Cem mbrit [2]; however, oother manufaccturers, mosttly in Easterrn Europe and a Asia, continued thhe use of asbesstos until the first f decade off the 2000s. A mass pproduction off green FCB was w made posssible after the optimal source of thhe reinforcingg fibres was determined. N Nowadays, a kraft lap pulp derivedd from hardw wood, e.g., Pin nus radiata oor Red Cedar, is used. Cellulose iss the main component c off the kraft puulp (>50% wt.). w This polymeric suubstance has the ability to form long annd hard to breeak linear chains madee out of alternnating pairs of dehydrated glucose / C6H10O5 / as shown in Figg. 2.

Fig. 2. Single ppolymer molecuule–a basic comp ponent of the linnear chain in celllulose.

4

Introducction

Multiple hydroxyl grroups in the glucose from m one chain form the hydrogen bonds with oxxygen atoms in adjacent chains, thus forming microfibrils with high tennsile strength. The microfibrrils have a diameter of approx. 100 nm and, in turrn, form thickeer (30-40 μm) fibres. The microfibrils can be positiioned at a diffeerent angle in relation r to the longitudinal axis a of the fibre. At sm mall angles, ann increase in fibre f stiffness compared to the fibres made out off microfibrils positioned p at laarger (40-50º) angles can be observed [8]. The elassticity/stiffnesss of the fibress can thus be controlled by choosing the source off cellulose or by b the mechan nical processinng of the kraftt pulp (the most commoon industrial soource of cellulose fibres). Coutts inn [9] has reporrted that theree is a cellulosee–cement com mpatibility in the sense that the matrrix hardening is not affecteed by the pressence of a certain amoount of celluloose. Optimal interfacial boond between the fibre and the matrrix can be estaablished. Excessive bond sttrength would d result in material em mbrittlement, while w a weak bond would reduce the composite c strength. Sinnce cement is an alkaline material m it conttains metal (C Ca, Si, Al, Fe) hydroxxyl groups. Cellulose C fibrres contain hydroxyl gro oups and carboxylic groups. Theese groups form f covalennt bonds capable of producing eefficient fibree–matrix bond ds. Thereforee, the cellulosse matrix composite can show both high flexural strength and fracture tough hness that should be baalanced by thee optimal amo ount of compoonents. The meethods of innvestigation into i the meechanical (an nd other) properties oof a variety of FCB products will be diiscussed in Chapter C 2. Many manuufacturers alsso add 3-5% % synthetic ffibres, e.g., polyvinyl p alcohol (PV VA, PVOH) fibbres to FCB. The techhnology is useed for severall reasons [10]]. The added fibres f are non-toxic aand feature high h tensile strength andd flexibility. PVA is hydrophilic and is capablle of forming strong bondss between the hydroxyl group and tthe cement matrix m in the presence of w water. The material m is cheap and w widely availabble. It is used in FCB in thhe form of fibrres 30-50 μm in diameeter. Fig. 3 shoows the PVA polymer moleecule.

Fig. 3. Formaation of polyvinnyl alcohol (PVA A).

The F Fabrication, Teesting and Application of Fibree Cement Board ds

5

Introducing the two aforemention ned types off fibres allow ws better control of thheir length disstribution with hin the matrixx, while the mechanical strength of tthe compositee is roughly proportional too the fibre asp pect ratio, i.e., to the diameter/lenggth coefficien nt. It also eliiminates the common problem of thin fibre “curling” togeth her when disppensed into the t slurry [11], and at the same timee allows the fiibres to be envveloped by ceement and filler particlles. Fig. 4 shhows two po olarized light micrographs of FCB containing 66% (wt.) kraft cellulose fibrres and 2% (w wt.) PVA fibres. Figures 4, 2-31, 2-32, 2-33, 2-34, 2-35, 2-36, 2-37, and 3-14 are reproduced in the centrefo fold in colour for f improved readability.

Fig. 4. Micrrographs of the fibre cemen nt board contaiining 6% (wt.) of kraft cellulose fibrres and 2% (wt.) PVA fibres. Top micrograpph shows PVA fibres and mall strand of cellulose c fibres. bottom microograph shows sm

CHAPTER ONE FABRICATION OF FIBRE CEMENT BOARDS KRZYSZTOF SCHABOWICZ AND TOMASZ GORZELAēCZYK FACULTY OF CIVIL ENGINEERING, WROCàAW UNIVERSITY OF TECHNOLOGY

Introduction A modern process of FCB fabrication consists in laying thin paper-like films on top of each other until a desired sheet thickness is obtained [13, 14]. The process distributes the reinforcing fibres in planar uniformity, taking the best advantage of the reinforcing fibres to increase the in-plane strength of the sheet. Thus, the strength of sheets made using this process is approximately 50% greater than the sheets formed to full thickness in a single action using a filter press or the extruding processes. In a process detailed below, a thin film formed from a diluted fibre slurry is usually 0.25-0.4 mm thick and each FCB comprises a stack of these films. Thus, the final sheet consists of approx. 20-30 or more thin films. A large number of layers partially suppresses the imperfection of the considered method: the films formed on a sieve are not uniform in composition but due to sedimentation they have a fibre-rich side and a fibre-poor side. Additionally, 2.5-3 mm long fibres bridge the sieve holes, slightly blocking the feed of other particles and forming a fibre-rich layer. The portions formed last can be relatively fibre-poor. On this account, an advanced fibre orientation and distribution devices have been developed and introduced into the actual chain of FCB fabrication. Fig. 1-1 shows a flowchart of a sample fibre cement board production process. The particular sub-processes named in the flowchart are detailed below.

Chapter One

8

WATER AND WASTE RECOVERY SYSTEM

Batching mix components

Mixing until homogenous pulp is obtained

PREPARATION ZONE

Preparing raw materials

Forming boards in special forming machine (Hatschek machine or flow-on machine)

Wet cutting and stacking

Pressing to enhance properties

Transferring

Taking boards off stacks and carrying

Natural maturing in climatic chambers and in air for 7-14 days (maximally 28 days) Finishing – dustless cutting with water jet

Fig. 1-1. Flowchart of sample fibre cement board production process.

Figure 1-2 shows a sample flowchart of the technological process of fabrication of cellulose fibre cement boards, including seven zones corresponding to the specific production stages [12]. Fig. 1-3 shows a detailed diagram of zones no. 1 and no. 2 in order to outline their complexity. The moisture content of the board varies at each stage and is determined using methods described in the next chapter.

Fabrrication of Fibree Cement Boardds

9

Fig. 1-2. Floowchart of tecchnological pro ocess of fabriccation of cellu ulose fibre cement boardds.

The fabrrication of cellulose FCB starts s in the ppreparation zo one (zone 1), where ceellulose fibress are mixed with w water in a mixer (pulpeer) until a uniform disppersion is prooduced. Bulk components c (ccement with additives) a in specific pproportions arre added to th he batched waater and mixeed until a homogenouss plastic comppound is obtained. There arre two main categories c of cellulose fibre cement: low-temperaature cured annd high-tempeerature or autoclave ccured. Low (air) ( cured fo ormulations uusually contaain larger amounts of Portland cem ment combineed with fine-gground fillerss such as clays, silica fume, groundd limestone, or o fly ash. Auutoclaved form mulations contain less Portland cem ment and more pozzolanic coomponents an nd fillers. The mixxture describbed above iss transferred to a board forming machine. Thhe Hatschek or flow-on machine m is uused (zone 2) to form boards with a fixed thickkness of 4-14 mm. m The nextt (optional) prroduction stage involvves the pressinng of stacked d boards (zonne 3). A presssing force suitable for the type off board fabriccated (exterioor or interior cladding board) is appplied. Warm m (due to cem ment hydrationn heat) cellullose fibre cement boarrds directly affter pressing or forming arre transferred to a precuring tunneel (zone 4) where w they rem main for abouut 14 hours. Next, N the boards can bbe taken off thhe stack and placed p on a palllet. This musst be done as quickly aas possible sinnce the boardss are still quitee warm and damp, d and so they shouuld not be alloowed to cool and a dry too muuch or too quiickly.

10

Chapter One

Fig. 1-3. Diagram of zones no. 1 and no. 2.

The boards placed on the pallets are left to mature and are cured in steady thermal-moisture conditions, e.g., in special airtight tarpaulin tents (zone 5) for about 14 days. After that time, the boards obtain the proper bending strength, and some moisture is removed naturally. After the maturation period, the boards are transferred to a final drying oven (zone 6) where they are subjected to three-stage drying at 180qC, 160qC, and

Fabrication of Fibre Cement Boards

11

120qC, respectively at stage 1, 2, and 3. Then, the boards are cooled naturally for about 20-30 minutes depending on their thickness. This is a critical stage of the fabrication process. Boards with excessive moisture content are not fit for further treatment, such as impregnation or painting. At the last fabrication stage, the boards are trimmed, and, if necessary, their surface is ground at the edges (zone 7).

Primary raw materials Table 1-1 shows the typical primary raw materials used in the fabrication of naturally maturing fibre cement boards. Table 1-1. Typical primary raw materials used in fabrication of fibre cement boards. type of raw material

Standard ratio

approximate composition ~ 60% ~ 8% ~ 2% ~ 30% ~ 100%

Cement Cellulose (dry) PVA Kaolin or lime Total Additives & admixtures Hyperplasticizer ~ 0.1 l/t *) Didecyldimethylammonium chloride ~ 0.1 l/t *) (DDAC) or bromide (DDAB) Perlite ~ 1 kg/t *) Mica ~ 1 kg/t *) Microsphere ~ 1 kg/t *) Antifoaming agent ~ 0.26 l/t*) *) l/t or kg/t = litres or kilograms per ton of finished product

Properties of PVA fibres PVA (polyvinyl alcohol) fibres are a major component in the fabrication of naturally maturing cement fibre boards. The basic specifications of PVA fibres, based on the data for Kuralon (manufactured by Kuraray America Inc.) [40], are shown below. Table 1-2 shows the basic specifications of PVA fibres compared to other commercially available fibres.

Chapter One

12

Main advantages of Kuralon fibres: ƒ ƒ ƒ ƒ

non-toxic, long-term presence on the market, established manufacturing process, high quality.

Table 1-2. Specifications of PVA fibres compared to other commercially available fibres [40]. type of fibre Parameter tenacity (cN/dtex) Young’s modulus alkali resistance

Kuralon

PET

Nylon

PAN

PP

Aramid

Carbon

11-14

6-8

5

2-4

6-8

22

13-23

250-300

80145

40-80

3070

30110

400700

11902370

w

X

X

X

w

X

w

X

X

X

X

X

•

w

•

w

X

w

w

adhesion to w cement weather w resistance w = good, • = normal, X = bad

Figure 1-4 shows micrographs of the cross-sections of fibre cement boards reinforced with different fibres.

Fabrrication of Fibree Cement Boardds

13

a)

b)

c)

Fig. 1-4. Miccrographs of cross-sections c of o fibre cementt boards reinfo orced with different fibrees [40]: a) Kuraalon fibres; b) polypropylene p fi fibres; c) ARG fibres. f

Chapterr One

14

Figure 1-5 shows the relationship between fibree tenacity and d bending strength of ffibre cement boards b reinforcced with diffeerent fibres [40 0].

Fig. 1-5. Relaationship betweeen fibre tenacitty and bending strength of fibrre cement boards [40].

R Raw materiaal preparattion and miixing zone The preparaation zone of a fibre cementt boards fabriccation usually y includes the followinng: ƒ ƒ ƒ ƒ ƒ ƒ ƒ ƒ

2 cem ment and addditive (e.g., lim mestone powdder) silos, eacch with a capaccity of about 100-200 1 m3, a watter tank with a capacity of 20-30 2 m3, 4 daiily stock silos with a capacity of 15-20 m 3, a bellt conveyor for fo transferring cellulose annd waste pap per to the pulpeer, a pullper with a cappacity of 12-16 m3, ( buffer pulpp tank) with a capacity an inntermediate ceellulose tank (a of 200-40 m3, up too 4 refiners (deepending on th heir type), 2 cellulose tanks (refined pulp storage s tanks)) with a capacity of 40-

Fabrrication of Fibree Cement Boardds

ƒ ƒ ƒ ƒ

15

60 m3, a mixxer (for mixinng all the comp ponents) with a capacity off 5-7 m3, an inntermediate filller tank (a bu uffer tank for the mixed liq quid pulp) with a capacity of 10-15 m3, a dennsity calibratinng tank with a capacity of 22-4 m3, chem mical agents taank (a flocculaant batching ddevice).

The prodduction of booards starts at a the two siloos located ou utside the production building (Fiig. 1-6), from which thhe raw mateerials are transported bby a belt convveyor (situated under the siilo (Fig. 1-7)) and then by a pneumaatic conveyinng system to th he daily stockk silo in the prroduction building (Figg. 1-8).

Fig. 1-6. Raw w material silos,, with a capacity y of 200 m3 eacch, outside the shop s floor.

16

Chapterr One

Fig. 1-7. Presssure vessel andd belt conveyor under silo.

Fig 1-8. Sampple daily stock silos, with a cap pacity of 15 m3 each.

The firstt production stage involvees preparing tthe cellulose, which is the source ffor the fibres that reinforcee the compositte fibre cemen nt boards in the celluulose preparaation zone. The T cellulose (and waste paper, if required) is fed in weigheed-out portion ns (bales) into the pulper viia the belt conveyor (F Fig. 1-9) with integrated i scaales.

Fabrrication of Fibree Cement Boardds

17

Fig. 1-9. Sam mple belt conveeyor with scales (weighing ceellulose and waaste paper) feeding weighhed-out portionns into the pulpeer.

Next, thee main water is i supplied fro om the tank (F Fig. 1-10) to the t pulper in order to m macerate and pre-defibre th he cellulose. T The cellulose is pulped in the pulpeer (Fig. 1-11). Water is used u throughoout the entiree pulping process.

18

Chapterr One

Fig. 1-10. Sam mple water tankk with a capacitty of 25 m3.

The celluulose content in the pulp iss 4% (wt.). A biocide is battched into the pulper tto protect thee cellulose fiibres against biodegradatio on. Also, depending oon the type of fabricated boards, a hydropphobizing ageent can be added at thee same time.

Fabrrication of Fibree Cement Boardds

19

Fig. 1-11. Sam mple pulper witth a capacity off 14 m3.

Since, ass mentioned earlier, e some cellulose c can be reclaimed from the waste paperr, the cellulosse fibres shou uld undergo a special impregnation process. Thhis can be donne in accordaance with, e.gg., US Patentt No. US 7.244.388 B B2 “Method off producing sttable cellulosee fibres with improved i biological sttability and prroducts made of them,” pateent date 17 Ju uly 2007. The celllulose fibress obtained using u this m method featurre higher biological rresistance and stability. Even E though the previou usly used methods of treating celluulose with biiologically toxxic compound ds would result in a hhigher resistannce of the celllulose to dec omposition, they t were not complettely satisfactory, since the fibres had to be cleaned before use and the enerrgy demand of o the cleanin ng process waas very high, while w the loss of fibree length was quite q significaant. Studies shhow that the treatment

20

Chapterr One

of cellulose fibres with didecyldimeth d hylammonium m chloride (D DDAC) or bromide (DDAB), carried out as per the method ddisclosed in th he patent yields the bbest results. Thanks to th hese substancces and som me copper content, the product is chharacterized by b a very goood biological stability, but not at thhe expense off a significant increase in ennergy demand d through the cleaningg process orr fibre length h loss. The ttreated fibress provide excellent reiinforcement of o the fibre ceement board pproducts and guarantee g high resistannce to degradaation. The preppared pulp is transferred t fro om the pulperr via a system m of pipes and pumps tto the buffer pulp p tank (Fig g. 1-12). Next,, the pulp is trransferred from the tannk to the refinners, where it is i refined. Dep epending on th he type of refiner, up too four refinerss (Fig.1-13) caan be used in the fabrication n plant.

Fig. 1-12. Sam mple buffer pullp tank with a capacity of 30 m 3.

Fabrrication of Fibree Cement Boardds

21

Fig. 1-13. Vieew of four refinners.

The refinned pulp is conveyed to one of the twoo storage tank ks, with a capacity of, e.g., 50 m3 eaach. Dependin ng on the proccess requiremeents, pulp with fibress reclaimed from waste paper, iimpregnated with a hydrophobizzing agent, caan be stored in one tank, while pulp with w nonimpregnatedd cellulose fibbres can be stored s in the other tank. Fig. F 1-14 shows sampple pulp storagge tanks. The pulpp is transferrred from the storage tankks to a mixeer with a capacity of 44.6 m3, wheree it is mixed with w other raw materials, i.e.., cement, additives (e.g., limestonee powder), an nd chemical aagents, thus giving g the liquid celluulose-cement mix the required rheoological (con nsistency) properties. A All of the com mponents (exccept for the ppulp) are batched from the daily stoock silo and auuxiliary plantss. The liquuid mix is trannsferred to an n 11 m3 bufferr tank (Fig. 1-16). The liquid mix is transferred from f the tank via a system of pipes to th he second zone, wheree the boards are a formed in the Hatschek machine, as described d below.

22

Chapterr One

Fig. 1-14. Reffined pulp storaage tanks with capacity c of 50 m 3.

Fig. 1-15. Mixer with a capaacity of 4.6 m3.

Fabrrication of Fibree Cement Boardds

23

Fig. 1-16. Liqquid mix bufferr tank with a cap pacity of 11 m3 .

D Detailed op peration off Hatschek machine and d fibre film m formation n Fig. 1-17 shhows a diagrram of a Hattschek machiine. The main n section includes onee (or more) vat(s) v with a cylindrical c sieeve rotating in contact with a diluteed water-baseed fibre slurry y capable of fo forming a filteering film and mineral materials, inccluding Portlaand cement. T The sieve is drriven by a continuous ffelt wrapped around the to op of a rotatinng cylinder by y a couch roller. The filtering film moves on th he felt to a foorming roller. The felt continuouslyy travels betw ween a drive ro oller and a taill roller.

24

Chapterr One

Fig. 1-17. Haatschek machinee diagram.

The sheeets are formedd in the Hatsch hek machine in accordancee with the following prrocedure [13]. When the cllean sieve is ppulled under the t slurry in the vat, w water from thhe slurry runs through the sieve to depo osit a soft porous film of fibres andd cement on the surface oof the sieve. The T sieve moves the ffilm onto the felt. f The exceess water is reemoved when the sieve is pressed too the felt. Thhe felt transferrs the film onnto the formin ng roller. When the reequired numbber of films is wrapped aroound the forming roller to form a shheet of a requiired thickness, the stack of films is remo oved from the roller annd laid out fllat to form a board. The ddraining of su uccessive films in conntact with eacch other undeer pressure is sufficient to bind the films togethher and form a continuous so olid material. Typical sieve mesh siizes are approx. 0.4 mm, w whereas the no on-fibrous particles aree significantlyy smaller (approx. 50 μm inn diameter). Therefore, T these particlles can freelyy pass through h the sieve ouut of the form med layer, whereas the fibres are rettained on the sieve. Thus, eentrapment off the nonfibrous mateerials within thhe film on thee surface of thhe sieve depen nds on the structure of the fibre layerrs formed. The effiiciency with which the filter f layer reetains the no on-fibrous material deppends on thee fineness off the cellulosse kraft pulp. This is achieved byy a special proccessing, i.e., by b crushing thhe fibres to red duce their diameter witthout affectinng their length h. High qualityy kraft pulp caan form a network of fine fibres caapable of trap pping at least the larger no on-fibrous particles. Thhis forms an initial i filter laayer on the suurface of the sieve s and

Fabrication of Fibre Cement Boards

25

allows the formation of the film. Once formed, the filter layer begins to trap the non-fibrous particles, which in turn fill up the spaces between the fibres. Thus, the pore size of the film is rapidly reduced and the finest non-fibrous particles are retained in the film. However, due to the removal of excess water, some pores may remain in the FCB material, constituting approx. 5-15 % of the volume porosity, as clearly shown in in Fig. 4. Therefore, the film structure is fibre-rich on the sieve side and relatively fibre-poor on the opposite side. This structure is similar in the other layers, forming a structure often referred to as a “cake” in terms of filtration theory. The continuous fibre network prevents the film from being deformed due to the stresses exerted in the fabrication process. Using a standard formula [16], the amount of liquid filtered in a unit of time per area of filter medium can be correlated to the pressure drop in the Hatschek machine, the filter cake (film) thickness, and the properties of the slurry being filtered. ଵ ஺

ȉ

ௗ௏ ௗ௧

ൌ

ο௉

(1)

௥ఓ௟

where: l = cake thickness μ = slurry viscosity ǻP = pressure on the film on the sieve immersed in the vat A = surface area of the cake ହȉሺଵି௘ሻమ ȉௌ మ

r= is a specific resistance of the cake ௘య e = material constant V = volume of the liquid filtrate t = time The ratio v of the volume of the cake to the volume of the filtrate can be expressed as follows: ௃௏ఘ

‫ ݒ‬ൌ  ሺଵି௃ሻሺଵି௘ሻఘ

ೞ ି௃௘ఘ



where: J = solid content of the feed slurry ȡ = density of water ȡs = density of solids in the slurry (other variables described above)

(2)

Chapterr One

26

Considerring that eqss. (1-2) relatee to the condditions presen nt in the Hatschek m machine withh a sieve raadius R and a felt speeed s and approximatiing v by solid content c, Co ooke in [13] ppresented the following f relationship for the cakee thickness l produced p in tthe Hatschek machine from the sluurry accordingg to the parameeters specifiedd above: ݈ ൌ ܴ݇ට

௖ ௦

(3)

m consttants of the sllurry feed and d the film where k includes all material propertiees. Equationn (3) can be used u to predictt how the form mation of a fillm on the Hatschek m machine is affeected by chang ges in the opeerating conditions. The following caan be concluded, provided other o conditioons remain uncchanged: -

largee machines willl produce thiccker films, solid content c willl increase film m thickness, increease in machinne speed will reduce r film thhickness.

In [13], Cooke describbes the experrimental resultts of 52 meassurements of film thickkness carried out for slurries with differrent solid con ntent (1.85.1%) and aat different felt speed using g the same maachine. An an nalysis of the data calcculated accordding to (3) hass revealed thatt the relation between b l and was estimated wiith a correlatio on coefficientt R = 0.5648; however, the scatter oof the results was w significantt. Uniform m film and shheet thicknesss, as well ass board levellness, are difficult to oobtain using Hatschek H machines. Anotheer relevant ob bservation is that the ooverflow of thhe machine iss apparently uuniform on av verage, as indicated byy fluctuations in the depth of o the overflow w at different points of the vat. Thoose fluctuationns are usually small, and soo the flow ratee does not vary much aacross the widdth of the sievee. It is appparent that thee flow within the vat of thee Hatschek machine m is laminar, andd so any changges in the soliid content in tthe feed at thee tub inlet will take plaace during itss flow around d the tub to thhe overflow. Thus, T any persistent loow spot or higgh spot in the sheet thickneess is due to th he low or high local soolid content inn the feed, respectively. Gen enerally, the hiigh or the low spot is aassociated witth a disturbancce in the levell of the feed to the tub. Local variattions in the feeed level or so olid content w will result in ch hanges in film thickneess.

Fabrication of Fibre Cement Boards

27

Any sheets consistently thicker on one side only are the result of a gradient in the solid content of the feed from one side of the machine to the other. The gradients can usually be observed, when the machine is fed from one side only. Usually, the thicker side is closer to the feed since the water tends to separate from the solids that are concentrated closer to the tub. It is therefore critically important that the feed is uniform along the entire width of each tub and is maintained as such throughout the entire process. The most common solution to this problem is to use the “fishtail” shaped distribution pipes with a narrow included angle. This ensures that there is no segregation of the solids in the feed, and a uniform distribution of solids is maintained across the sieve. It is not uncommon for tubs to be fitted with two or more fishtail feeders to feed the fractions of the width of the tub. It is also clear that where the film thickness of the sheet varies, it can be brought under control and compensated for by changing the machine speed in accordance with the relationship in eq. (3). The productivity of the machine could also be significantly improved by simultaneously increasing its speed and the solid content of the feed. However, there are two factors that limit the solid content that can be used in the tub. The first one is that increased viscosity of the feed increases the resistance of the sieve rotating in the tub. The other is that increased solid content increases the occurrence of curls, preventing the formation of smooth films. The increase in the viscosity of the feed is due to the increase in the solid content in the slurry. This slows the rate of film formation as shown in eq. (1) and increases the friction of the sieve. When the increase in resistance is too high, the sieve may slip across the surrounding felt. This may result in uneven film formation, since a thicker film will develop on the section of the sieve immersed when the sieve stops. When the sieve starts again, the next section of the film will be thinner, and the difference will be difficult to compensate for. Flocculation (curling) of the feed at higher solid content is of equal importance. It results from the “crowding” of the particles together, in particular, entangled fibres. The fibres, in turn, trap other particles in their interstices, forming irregular lumps. Thus, smooth uniform films cannot be formed, and the appearance, as well as other properties of the sheets, are compromised. Thus, there are limitations to the solid content of the feed slurries before formation, otherwise operational problems may occur within the Hatschek machines. Furthermore, eqs. (1-3) are not valid under extreme conditions, since the assumed uniformity of the slurry no longer applies.

Chapterr One

28

Dettails of otheer FCB fabrication prrocess stagees This subsecction presentss other devicees used in thhe fabrication n of fibre cement boarrds. It should be b noted that some of the ddevices are opttional.

Weet board trim mming system m The wett board freshhly formed in n the Hatscheek machine should s be trimmed to the required size. s Edge trim mmers (Fig. 1-18) are used for this purpose.

Fig. 1-18. Edgge trimmers.

W Waste recoveery system Wet board trimming waste is geneerated in the pproduction pro ocess. As previously m mentioned, boards b are triimmed after leaving the Hatschek machine, annd the waste reecovery system m is installed there. Fig. 1-19 shows the waste rrecovery devvice. The trim mmings are transferred by b a belt conveyor too a sluice, whhere a measurring system m monitors the density d of the waste diiluted with waater. Then, th he waste is traansferred to a recovery device from m which the sluurry is returneed by a pumpping system to o a buffer tank and reuused in the fabbrication of fib bre cement booards.

Fabrrication of Fibree Cement Boardds

29

Fig. 1-19. Vieew of waste reccovery system.

Pressingg and heat treatment of bboards The nexxt production stage involv ves further trreatment of the fresh boards. Trim mmed boards are a placed on a moving steeel plate (Fig. 1-20) and transported tto an electroppneumatic device (a stacker)) that stacks th he boards together witth steel platess (dividers). The T dividers ar are oiled to prrevent the boards from m sticking. Figg. 1-21 shows a sample sttacker. The boards are transferred tto a roller traansfer system (Fig. 1-22) tthat, dependin ng on the intended usee of the boardds, transfers th he stacks of fibbre cement bo oards to a press (Fig. 11-23) or directtly to a prelim minary heat treeatment tunneel (Fig. 124).

30

Chapterr One

Fig. 1-20. Steeel plates used for f stacking fibrre cement boardds.

Fig. 1-21. Staacker.

Fabrrication of Fibree Cement Boardds

31

Fig. 1-22. Roller transfer sysstem.

Pressingg (at min. 9000 tonnes) aims to remove eexcess water from the boards and tto compact thee material, and d thus improvve its parameteers. Then, thhe fibre cemennt boards are transferred too the prelimin nary heat treatment tuunnel (Fig. 1--24). Heat treeatment in thee tunnel at 80qC lasts approximateely 8-10 houurs. During th he process, th the board acq quires its preliminary mechanical sttrength, in parrticular its bennding strength h.

Fig. 1-23. Preess.

32

Chapterr One

Fig. 1-24. Preeliminary heat treatment t tunnel.

After heeat treatment in the tunnell, the stacks are transferreed by the rollers to a destacker (F Fig. 1-25), wh here the dividders are remo oved and transferred tto a cleaner and a then to an n oiler, after w which the prelliminarily matured fibrre cement boards are stack ked (but this time without dividers) on wooden pallets. The pallets with the t fibre cem ment boards arre placed inside the sppecial tents (F Fig. 1-26), wh here the boardds are left to mature m for about two w weeks.

Fig. 1-25. Destacker.

Fabrication of Fibre Cement Boards

33

Fig. 1-26. Tents for maturing boards.

Water recovery system A considerable amount of recoverable water is used in the production process. The water is recovered by the components of the return water system. The main source of recovered water is the first production stage. Water is filtered off the cellulose-cement mix in the Hatschek machine by a system of suction pumps. Then, water finds its way to a pair of return water tanks (Fig. 1-27) via a system of channels in the floor and under the machine. The tank design allows water to be fed into its middle section, and, before mixing with the remaining water, it reaches a depth of about 5 m. Cement suspension is preliminarily removed from the water (clean water is at the top, while dirty water is at the bottom of the tank).

34

Chapter One

Fig. 1-27. System of return water tanks (left – tank 1, right – tank 2).

An overflow is installed between return water tank 1 and return water tank 2 (Fig. 1-27), whereas the preliminarily cleaned water overflows from the top of tank 1 to tank 2. Sedimentation (the settling of cement, cellulose, and other component particles) takes place in both tanks. Since the settling particles can block the tanks’ outlets, a special chemical agent (flocculant) is added to the process on one side using the device shown in Fig. 1-28. The flocculant binds the particles into more liquid coagulants. On the other side, the valve in the tank bottom is automatically opened for a short time at frequent intervals and the water is discharged to a special well to remove the excess sediment accumulated at the bottom of the tank. Water is fed from return water tank 1 to a density calibrating tank, where it is used to dilute the mix to a required concentration of 10-15%. It makes it possible to use most of the sediment and to significantly reduce the amount of waste products. Water also flows to the well during discharge (lasting a few seconds), a process that is repeated every few minutes in order to remove the excess sediment from the bottom of the tank. Return water tank 2 supplies water to the water tank (which, in turn, supplies water to the pulper) and to the Hatschek machine.

Fabrication of Fibre Cement Boards

35

Fig. 1-28. Flocculant batching device.

Finishing the boards Partial drying of the boards Partial drying of the fibre cement boards in the drying tunnel shown in Fig. 1-29 is an optional fabrication process stage. The aim is to reduce the relative moisture content of the boards to a few percent before varnishing. The process is divided into three stages: 1) heating up to 120qC; 2) drying at approx. 180qC; and 3) cooling. The boards are transferred in the drying tunnel by a roller conveyor. A system of fans diffusing air in the drying tunnel chambers keeps the ambient conditions constant. The whole process of moving the boards through the drying tunnel is automated and controlled by a system of sensors. If necessary, the transfer speed is increased or decreased to ensure optimal and safe drying conditions. Thereby, the boards are protected against damage or inadequate heating (and thus incorrect drying conditions). The finished boards leave the kiln on rollers and are placed on a pallet by, e.g., a stacker.

36

Chapterr One

Fig. 1-29. Tunnnel for drying boards.

ng, drilling, an nd painting Edging, grinding, cuttin o fibre cemeent boards iss cutting, The nexxt stage in thhe finishing of edging, andd milling. Inn the producction processs, particularly y during pressing, the edges of thhe boards can n be damagedd. In order forr the end product to satisfy the set s requiremen nts, it shouldd undergo th he edging process, whiich involves retrimming, r grrinding, and m milling. Dependiing on the orrder and the intended usee of the fibree cement boards, theiir surface cann also be polished. The poolished board ds can be varnished. Fibre cem ment boards prepared p in th his way can bee optionally trransferred to the panell saw shown in Fig. 1-30 to be cut innto smaller sizes. This process is laargely mechannized and aided by a compuuter system.

Fabrrication of Fibree Cement Boardds

37

Fig. 1-30. Pannel saw.

Dependiing on the orrder, the fibree cement boarrds can be cu ut to any shape using a water-jet cuutting machine shown in Fiig. 1-31. The computer controlled jeet of pressurized (to 3800 bars) water w with an abrasive agent can cut the bboard to any shape. s

Fig. 1-31. Waater-jet cutting machine. m

38

Chapter One

The final optional stage in the fabrication process of fibre cement boards is painting, which requires a board varnishing line. The sections of a varnishing line are shown in Figs 1-32 and 1-33.

Fig. 1-32. Section of the varnishing line: feed roller system followed by a system that applies individual coats of varnish.

Fig. 1-33. Section of the varnishing line: multiple-hearth furnace.

Fabrication of Fibre Cement Boards

39

The varnishing line is divided into two sections. The first section includes a workstation, to which the roller system delivers a pallet with the boards and all sides of the boards are manually spray varnished. Next, the pallet with the boards is transferred to a drying room. Varnish is applied to the top and the bottom of the fibre cement board in the second section of the line. Individual boards are transferred via a conveyor belt to the cleaning brushes and then to the tunnel, where the boards are heated. Then, the boards move over the rollers while they apply a prime coat (hardened by UV lamps) to the bottom. Then, two prime coats are applied to the top surface of the board. The board is transferred to another section, where it is dried with hot air before it is moved to a mechanized shop. The dried boards enter a varnishing chamber where, e.g., two moving double nozzles apply a required varnish coating. In the final step, the varnish is dried in a drying furnace, where the varnished boards are left for approx. 45 minutes, until the cross-linking is completed and the varnish cures.

CHAPTER TWO FIBRE CEMENT BOARDS TESTING ZBIGNIEW RANACHOWSKI INSTITUTE OF FUNDAMENTAL TECHNOLOGICAL RESEARCH, POLISH ACADEMY OF SCIENCES

Mechanical properties of FCB The mechanical properties of an FCB depend on its composition and the orientation of the fibres within the sheet. Fibre orientation depends on the operating conditions of the fabricating machine during sheet formation. Thus, the relative strengths and strains until failure generally change in the orientation parallel and perpendicular to the main sheet formation on the felt. Before the cracking of the matrix, the stress within the sheet in each direction is transferred by both the matrix and the fibres. The relative amount of stress transferred by each depends on the relative elastic moduli of the matrix and the fibres, the orientation of the fibres and their ratio [14]. Nowadays, FCB contains an experimentally determined concentration of fibres, in such a way that, after cracking, the matrix does not carry any load and, instead, the entire tensile or flexual load is carried by the fibre network. The fibres are usually mostly organized in a pattern, and in every other film they tend to run parallel to the sheet surface and lie at the same but numerically opposite angles to the machine direction. Thus, the ultimate directional flexural or tensile strength may be predicted in relation to the mechanical parameters of the fibres and the matrix by using the following formulas [15]: ௟

ߪ஼௎ெ஽ ൌ  ‫ݒ‬௙ ߪி௎ ሺͳ െ  ሻܿ‫ߠݏ݋‬ ௟೎

݈௖ ʹ݈ െ  ݈௖ ൌ  ‫ݒ‬௙ ߬ ሺ ሻܿ‫ߠݏ݋‬ ݀ ݈

(4)

Chapter Two

42 ௟

ߪ஼௎௑஽ ൌ  ‫ݒ‬௙ ߪி௎ ሺͳ െ  ሻ‫ߠ݊݅ݏ‬ ௟೎

(5)

݈௖ ʹ݈ െ  ݈௖ ൌ  ‫ݒ‬௙ ߬ ሺ ሻ‫ߠ݊݅ݏ‬ ݀ ݈

where: ıCU = composite strength ıFU = fibre strength vf = fibre volume fraction IJ = fibre/matrix bond strength l = fibre length lc = critical fibre length MD = machine direction XD = cross direction Ĭ = average angle between the fibre and the machine direction A critical fibre length is a combination of fibre and matrix material parameters. It is defined as a minimum fibre length l, at which the center of the fibre reaches the ultimate strength sf when the matrix achieves the maximum shear strength tm , so that l/d should exceed sf / 2tm. In practice, the largest fibre lengths available in a processed kraft pulp are 2.5-3 mm long. The effect of strength anisotropy in FCB products due to their fabrication method is resolved according to European Standard EN 12467, a procedure for determining board bending strength in which the strength is evaluated in both orthogonal directions. A detailed study of the gradual decrease of FCB mechanical parameters, FCB durability, and potential environmental hazards can be found in [15].

European Standard EN 12467 Basic FCB testing procedures are detailed in European Standard EN 12467, which was approved by the 33 member countries of European Committee for Standardization (CEN), i.e., Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic, Denmark, Estonia, Finland, F. Y. R. of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, and Great Britain. The standard was prepared by CEN Technical Committee No. 128 “Roof covering products for discontinuous laying and products for wall cladding.” The latest version of the standard was approved on 4 August 2012 [17]. Generally speaking, the standard

Fibre Cement Boards Testing

43

specifies the technical requirements for and establishes methods of inspection and testing as well as acceptance conditions for the use of FCB for both external and internal wall and ceiling finishes. The standard describes the composition of FCB as including the following ingredients: -

fibre of any non-asbestos material, cement complying with EN 197-1 or with technical specifications relevant for the country of use and/or calcium silicate, process aids, fillers, aggregates, and pigments.

Table 2-1 shows the sample content of FCB composition for both external and internal applications. Table 2-1

raw material

cement fillers cellulose fibres PVA fibres (more expensive than cellulose kraft) hydrophobic agent

% of content by weight external cladding board

internal cladding board

50-70 27-32 6-12

44-64 33-43 6 10

0.25-2

-

0.3-0.9

-

FCB products are divided into four categories (A, B, C, and D) according to their weather resistance. Category A includes boards intended for applications where they may be subjected to heat, high moisture, and severe frost. Category B includes boards intended for similar conditions but protected from or not subjected to severe weather conditions. Category C includes boards for internal applications subjected to heat and moisture but not frost, and Category D includes boards for rigid underlayer applications. FCB products are also divided into five classes (1, 2, 3, 4, and 5) according to the results of the bending strength tests. The standard assessment of FCB properties takes into consideration the following requirements: -

board dimensions, shape, and tolerances, apparent density,

Chapter Two

44

-

mechanical resistance, water impermeability, water vapour permeability by the determination of the water vapour resistance value μ, moisture content measured during FCB production process, durability against warm water, durability against soak–dry and freeze–thaw cycling, durability against heat–rain, reaction to fire.

The most important information for the product characteristics evaluated by the tests discussed above will be presented in the form of a product label in accordance with the EN Standard. Apart from the standard testing procedures, some new and not yet established methods are presented at the end of this chapter.

Measuring board dimensions, tolerances, straightness and squareness of edges Fibre cement boards are usually large-sized components. They are subject to shrinkage when dried, and their dimensions should be properly evaluated to ensure proper mounting. Therefore, Standard EN 12467 defines a precise location and the number of repetitions of each dimension measurement for each board size. In addition, the Standard specifies the following tolerances for nominal dimensions. Two levels of fabrication qualities are specified for the length and width tolerances. Within the first level, the tolerances vary from + 3 mm for small boards (maximum dimension < 600 mm) to + 5 mm for the largest components (maximum dimension > 1600 mm). Within the second level, the tolerances vary from + 4 mm for small boards (maximum dimension < 600 mm) to + 8 mm for the largest components (maximum dimension > 1600 mm). Another critical board parameter is its thickness. The nominal thickness of the board must be specified by the manufacturer. For nontextured boards, the nominal thickness is taken as an average value. However, for textured boards it is taken as a maximum value. The procedure for determing which thickness measurement is used depends on whether the board is textured or not. In each case, the thickness is checked as a first step by carrying out of three measurements within an accuracy of 0.1 mm and then calculating the arithmetic mean, as well as the difference between the extreme values. These derived values must meet the relevant standard tolerances. The tolerances vary from + 0.6 mm for thin boards to

Fibre Cement Boards Testing

45

-2 to +3 mm for thick and textured ones. For non-textured boards, the procedure should be repeated two to three times, depending on their size. For large (> 1600 mm) textured boards, eight measurements around the entire surface of the board are required. A more accurate procedure is required for measuring the average thickness of textured boards, and can be used to calculate the modulus of rupture (MOR) and apparent density. In this case, it is recommended to immerse the sample in water. This can be expressed with the following formula: ݁ ൌ 

ଵ଴଴଴௏

(6)

௟௪

where: e = specimen thickness in millimetres, V = fluid volume displacement in cm3, l = board length in millimetres, w = board width in millimetres. The straightness of the edges is measured with a ruler or a portable square. The measurement involves stretching a metal string or wire between the edges of the board and measuring the greatest distance between the edge and the string. The straightness of the ruler must be at least 0.3 mm/m and the right angle must be accurate to at least 0.1%. The tolerances on the straightness of the edges are defined as a percentage of the length of the edge of a relevant dimension and must conform to the following values: 0.1% for Level I or 0.3% for Level II. The squareness of the edges is measured with a portable square by placing two adjacent corners of the board in turn between the arms of the square, keeping one side pressed against the full length of one arm and the other in contact with another one at least at one point. In this position, the greatest distance of the board edge from the arm of the square is measured. The tolerances on the squareness of the board are defined as a maximum distance measured, as defined above per unit of the board length and must be in accordance with the following values: 2 mm/m for Level I or 4 mm/m for Level II.

Measuring apparent density The apparent density d of a porous material is based on the external dimensions of its sample and the sample weight. The Standard states that

46

Chapter Two

in the case of testing FCB the manufacturer must specify in his literature the minimum value of d for each category and each class of sheet. The apparent density determined by the completed measurements must be not less than this value. Typical apparent densities are as follows: dry cellulose fibres – 400 kg/m3; PVA fibres – 1300 kg/m3; concrete – 2400 kg/m3; light FCB of class D – 1000 kg/m3; tough FCB of class A – 1600 ~ 1800 kg/m3 Apparent density is determined as follows. First, the sample volume V is measured by immersion in water or another method within an equivalent accuracy (at least 1%). In the case of immersion in water, the specimen must be saturated in water beforehand. Then, the mass m of the specimen is measured after drying it in a ventilated oven maintained at 100 ºC to 105 ºC for 24 hours. The mass should be measured within an accuracy of 0.1%. The apparent density is then given by the following formula: d = m / V.

Measuring mechanical resistance The mechanical behaviour of brittle cement composite materials is determined by the flexural test. The main parameters related to FCB measured during the test are the modulus of rupture (MOR) and the modulus of elasticity (MOE). MOR is the maximum stress reached by the material under mechanical load. MOE is the tangent of the slope angle of the stress versus the deflection curve during initial elastic deformation. The equations for calculating those parameters depend on the bending configuration, i.e., three – or four – point bending. They also depend on the distances between the supports (also referred to as the length of the support span), their relationship with the position of the force(s), and the specimen dimensions. The existing guidelines laid down in EN 12467 on how to perform testing of MOR and MOE are presented below. Another parameter that allows one to measure the strength of the fibre system within the FCB is the energy absorbed during the flexural test, defined as the area under the force vs. the displacement curve from the beginning until the limit of the test is reached. The guidelines on how to perform the tests have been developed by the International Union of Laboratories and Experts in Construction Materials, Systems and Structures (RILEM). Regarding the energy absorbed during the flexural test measurement, RILEM recommends whichever occurs first to be used as a limit: the value of the ordinate corresponding to 40% MOR or the deflection corresponding to 10% span. All RILEM guidelines for testing and use of construction materials are available in [18]. The Standard TFR1 of RILEM (“Test for the determination of MOR and LOP of thin fibre

F Fibre Cement Bo oards Testing

47

reinforced sections”) estaablishes the fo our-point bendding configura ation as a preferred test configuratiion. LOP stan nds for the paarameter referrred to as “the limit off proportionaliity,” i.e., the limit of the elaastic region off loading. This connfiguration allows the partt of the speciimen between n the two loads to be subjected to be b a pure ben nding constantt effort and th he point’s application tto stay away from the breaaking point off the piece, in n order to ensure that the correct reesults are obtained. By conntrast, in a th hree-point bending connfiguration, thhe only pointt subjected too pure bendin ng is the point just unnder the load application point. p Moreov er, the load presses p on the surface of the specim men near or in n the breakingg cross-sectio on, which for soft mateerials could affect the flexu ural test resultts. However, ISO I 8336 and EN 124467 also recom mmend the three-point bennding configurration for testing fibre cement speciimens. Fig. 2-1 shows the foour-point bend ding configurration of the specimen loading baseed on a Lloydd EZ 50 machine. The machhine is operating at the Institute of Fundamental Technological Research, Warsaw, Poland [19]. The system is fitted withh a LVDT sen nsor recordingg the deflection of the bent specim men with a resoolution of 1 μm m.

Fig. 2-1. Loaading machine in a four-point bending confiiguration for testing fibre cement specim mens.

The TFR R1 test develooped by RILE EM recommennds a support span and specimen thhickness ratioo equal to or higher thann 20. In a four-point f bending connfiguration (seee Fig. 2.2), th he following equations are valid for

Chapter Two

48

MOR and MOE expressed in megapascals (MPa), according to this Standard:

Fig. 2-2. Four-point bending configuration of specimen loading.

‫ ܴܱܯ‬ൌ  ‫ ܧܱܯ‬ൌ 

ி௟ ௕௛మ ଶଷி௟ య ଵ଴଼௙௕௛య

(7) (8)

where: F = loading force (N), l = length of the support span (mm), f = maximum deflection (mm), b = specimen width (mm), h = specimen thickness (mm),

Fig. 2-3. Three-point bending configuration of specimen loading.

When applying the same parameters in a three-point bending configuration (see Fig. 2.3) in accordance with EN 12467, the following equations are valid for a single measurement of MOR and MOE: ‫ ܴܱܯ‬ൌ  ‫ ܧܱܯ‬ൌ 

ଷி௟ ଶ௕௛మ ሺிమ ିிభ ሻ௟ య ସሺ௙మ ି௙భ ሻ௕௛య

(9) (10)

F Fibre Cement Bo oards Testing

49

where: F, l, b, h = identical paarameters as in eqs. (7) andd (8), F2 and F1 are loads takken from two points on the linear section n of the plot, beloow the limit of o proportionality, f2 and f1 are deflectionns correspondiing to the loadds selected, in millimettres. The loadds F2 and F1 selected s for MOE M calculatiion were intro oduced to compensate for an irreggular course of the loadding force veersus the specimen deeflection curvee during the bending test. F 2 and F1 shou uld ensure that the deteermined MOE E corresponds to the slope oof the F(f) curve within the initial liinear range. FCB F boards sh hould also bee tested with the t upper surface (expposed after insstallation) faciing the appliedd load. Fig. 2-4 shows an exaample of the load-deflectio l on curve plotted by the authors usinng the three--point bendin B specimen with the ng of an FCB following diimensions: l = 150 mm, b = 40 mm, and h = 8 mm.

Fig. 2-4. Loaad-deflection cuurve with markeed sections usedd to determine MOR and MOE parameeters.

The authhors have perfformed the beending tests oon three sets of o Class 2 FCB specim mens under thee conditions deescribed abovve. The remain ning three sets of speciimens of the same s types were stored in aan electric ov ven at 230 ºC for two hours. The heat h treatmen nt resulted in the decompo osition of cellulose fibbre bulk reinfoorcement due to pyrolysis. Table 2-2 con ntains the description oof the tested specimens s of three t panel typpes.

Chapter Two

50 Table 2-2 set number / properties

1

2

3

set designatioon

A

B

C

type of boardd

FCB for exterior use, 6% of fibres of good quality q and av. length of 2.5 mm

board thicknness [mm]

FC CB for exterior usse, 6% of recycled fibres an nd av. length off 1 mm

FCB for inteerior use, 4% of fibress of nonkraft, raw ceellulose and av. length of 2.5 mm

10

10

10

average bending strength [MPa]

12..30

8.50

3.00 0

density [kg/m m3]

1700

1400

1200 0

view of boarrds

Fig. 2-5 to 2-7 show w the results of the mechaanical tests, where w the curve on thhe left shows the behaviou ur of the speccimens in as--delivered state and thee curve on thee right shows the t behaviourr of the specim mens with the reinforceement destroyyed. Photograp phs of the cro ss-section of the t tested boards beforre and after buurning are sho own below thhe curves. Bassed on the analysis of the curves, the board sp pecimens in as-delivered state are capable of w withstanding a load close to FMAX for a period neccessary to destroy the reinforcemennt system. The T specimenns that underrwent the cellulose fibbre decompossition demonstrated the bbrittle characteer of the rupture, i.e.,, the break off the material cross-section appeared imm mediately after reaching the critical FMAX level. Moreover, the higher values v of deflection an and thus the higher h work of flexural testt Wf were recorded for

F Fibre Cement Bo oards Testing

51

the material with longer and a non-recycled fibres.

Fig. 2-5. Loadd-deflection cuurve at constant deflection rate for type A fibrre concrete specimens wiith 6% fibre content. Tests carried c out on m materials in ass-delivered state (left) annd after destructtion of the reinforcement (righht). View of thee fractured specimen (below).

Fig. 2-6. Loadd-deflection cuurve at constant deflection rate for type B fibrre concrete specimens w with 4% raw cellulose fibre. Tests carried out on materiials in asdelivered statte (left) and afteer destruction of o the reinforcem ment (right). View V of the fractured speccimen (below).

Chapter Two

52

Fig. 2-7. Loadd-deflection cuurve at constant deflection rate for type C fibrre concrete specimens wiith 4% non-kraaft, raw cellulosse fibres. Testss carried out on n materials in as-delivereed state (left) annd after destrucction of the reinnforcement (rig ght). View of the fractureed specimen (below).

After thee tests, the energy e absorbed during thee flexural tesst Wf was determined as the work made m over thee deflection ccurve, i.e., equ ual to the integrated prroduct of applied stress and d strain per unnit cross-sectiion of the tested boardd. A calculatted Wf valuee is referred to as toughn ness. The fracture wass initiated at initial force F0 = 2N and progressed until u final decrease to 40% maxim mum load F0.4MAX , accordding to the following f 0 formula:

W

f

1 S

F 0 . 4 MAX

䌿

F dda

F0

(11)

where: S = specimen cross-seection [m2], a = specimen deflectioon under the loading roller [m]. Faults duue to incorrecct fabrication of type C booard, leading to a nonuniform disttribution of fibres, fi resulted d in a consideerable declinee of work throughout tthe flexural teest Wf. A sligh ht increase inn FMAX can be observed in Fig. 2.5--2.7 (right cuurve), probably due to phhase transition ns in the

Fibre Cement Boards Testing

53

matrices during heating. Table 2-3 shows the comparison of all determined Wf values. The experimental results presented above enabled the authors to formulate the following conclusions. The rhree-point flexural test is an effective method of analysis of the board resistance to applied mechanical load. Both microstructural phases of the board, i.e., the brittle matrix and the fibre reinforcement, carry the load during bending or tension. The test results have revealed that the destruction of reinforcement resulted in a 4050% loss of toughness in respect to the boards with uniformly distributed reinforcement made up of fibres of good quality and the required length (6 mm). For boards reinforced with a recycled cellulose with a reduced fibre length, featuring a significantly lower toughness, approximately 70% loss of toughness was observed following the destruction of the reinforcing skeleton. The tests carried out on a low-quality material with non-uniform fibre distribution showed that the pyrolysis of the remaining fibres caused the highest decrease in toughness. Table 2-3 specimen code flexural toughness Wf of type A specimens, [J/m2] specimen code flexural toughness Wf of type B specimens, [J/m2] specimen code flexural toughness Wf of type C specimens, [J/m2]

A31

A32

A33

A34

A35

A36

0.98

1.34

0.83

0.49

0.40

0.49

B31

B32

B33

B34

B35

B36

0.45

0.50

0.71

0.13

0.14

0.23

C31

C32

C33

C34

C35

C36

0.50

0.41

0.40

0.04

0.03

0.04

Since MOR and, to some extent, MOE are the key mechanical parameters of FCBs, a detailed procedure for determining both parameters is available in EN 12467. The ratio of the machine support span l to the

Chapter Two

54

specimen dimensions and other parameters must conform to the following: -

l to specimen thickness ratio is greater than or equal to 15, l to deflection at rupture ratio is greater than or equal to 20, specimen length is greater than or equal to l plus 40 mm, specimen width is greater than or equal to five times the nominal specimen thickness, support diameter is between 3 and 25 mm, preferred specimen dimensions are 250 x 250 mm, preferred span is 200 mm.

After cutting the specimens by following a procedure to obtain a representative set of samples, the specimens in all four quality categories should be conditioned in accordance with the procedure detailed in Table 2-4. Table 2-4 test type

conditioning procedure

Category A/B product test

24 h immersion in water for board thickness < 20 mm 48 h immersion in water for board thickness > 20 mm

Category C/D product test

7 to 14 days in ambient laboratory conditions

New and category A/B/C product test

Prior to the bending test: 7 to 14 days in ambient laboratory conditions followed by 24 h immersion in water for board thickness < 20 mm or 48 h immersion in water for board thickness > 20 mm. The specimens must be tested immediately upon removal from the water.

New and category D product test

7 to 14 days in ambient laboratory conditions

Fibre Cement Boards Testing

55

Both the modulus of rupture and the modulus of elasticity of FCB products according to the Standard mentioned above must be estimated using an arithmetic mean of at least ten values, i.e., two x five measurements in two perpendicular directions. There are generally two types of conditioning procedures. Immersion in water is required for higher categories (A and B) boards. In this case, the next mechanical test is performed in wet conditions. Boards in the lower categories (C and D) are conditioned in ambient laboratory conditions. In the latter case, a mechanical test is performed in ambient laboratory conditions. The test must be performed at a constant deflection rate with accuracy and repeatability error less than or equal to 3%. Due to the usual variation of thickness that occurs in FCB products, the board thickness should be determined at two points along the loading line. After the board has been tested in one direction, the procedure recommends reassembling the pieces and performing the measurement in the perpendicular direction. Generally, the MOR parameter is not recommended as a measure of mechanical toughness for textured boards. In that case, the MOE should be applied. The minimum moduli of rupture (MOR) required to comply with EN 12467 are specified in Table 2-5. Table 2-5 minimum modulus of rupture at wet conditions [MPa]

minimum modulus of rupture at ambient laboratory conditions [MPa]

classes

category A and B

classes

category C and D

1

4

1

4

2

7

2

7

3

13

3

10

4

18

4

16

5

24

5

22

The specimen loading duration must be between 10 and 30 seconds. Table 2-6 shows typical MOR values as determined by the authors for different FCB products and FCB applications.

Chapter Two

56 Table 2-6 product no.

application

bending strength class

density [kg/m3]

bending strength [MPa]

1

exterior

2

1600

26.5

2

exterior

2

1600

30.0

3

interior

3

1700

12.8

4

interior

3

1200

14.0

5

interior

3

1000

9.5

6

interior

3

1300

10.2

7

interior

3

1800

9.3

Testing water impermeability Testing water impermeability involves a visual inspection of three specimens per test. The specimen dimensions must correspond to the actual size of small boards. For large boards, the dimensions should be at least 600 mm x 500 mm, except for narrow boards, where the dimensions should be 600 mm x the maximum available width. The specimens should be kept at ambient laboratory conditions for at least seven days before testing. The test is carried out by applying a special frame sealed at the top of the specimen face. The face is filled with water to a height of 20 mm above the board face. The test lasts 24 hours at laboratory conditions. The test results are deemed positive if moisture can be observed on the under face of the board, but no water drops have formed by the end of the test period.

Testing water vapour permeability by determining water vapour resistance μ The water vapour transmission rate of building materials is a key parameter for assessing the moisture protection of the final construction. The method should be used to test the drying processes for the boards used in underlayer applications (Category D FCBs). Water vapour resistance μ can be determined in a specialized laboratory in accordance with EN ISO 12572 [20]. Water vapour resistance is related to vapour diffusion across the analysed material, and the results are compared to the process carried out in stagnant air at normal pressure.

Fibre Cement Boards Testing

57

That value can be determined under two different conditions. To induce the diffusion of vapour, the tested specimen can be placed between the following two air volumes at relative humidity (RH): between 0% and 65% RH at room temperature (dry conditions) or between 65% and 100% RH at room temperature (wet conditions). A circular specimen is placed in a vessel with an internal diameter of 100 or 200 mm as its top closure and sealed up to the vessel’s rim. To maintain a constant relative humidity of 65% inside the vessel, a saturated potassium nitrate solution is transferred to the vessel. Alternatively, to maintain a relative humidity of 100%, distilled water is used instead of the salt. The vessel is placed in a climate chamber that provides both a stable temperature and a stable humidity. Water vapour is diffused throughout the material specimen due to the partial vapour pressure gradient between the air layers adjacent to the material surfaces. After the diffusion flux reaches a steady state (approx. 1 h), a constant change in weight of the vessel per unit of time can be recorded using a precision scale. The diffusion flux G under these conditions can be calculated as follows: G = ǻm / ǻt

(12)

where: ǻm = absolute vessel mass [kg] increase (in dry conditions) or decrease (in wet conditions) as determined by measurements at t1 and t2 ǻt = t2 – t1 In the next step, a diffusion conductance factor W is calculated as follows: ܹ ൌ 

ீ

(13)

஺ο௣ೇ

where: A = diameter of specimen [m2], ǻ pV = absolute differential pressure between the pressure in the vessel p1 and the pressure in the climate chamber p2, p1 and p2 = pressure [Pa] calculated based on temperature Ĭ [ºC] and humidity ij [%] in accordance with the thermodynamic conditions as follows: భళǤమలవഇ

‫ ݌‬ൌ ͸ͳͲǤͷ߮݁ మయళǤయశഇ

(14)

Chapter Two

58

The water vapour diffusion coefficient į is calculated as follows: į=Wd

(15)

where: d = specimen thickness [m] Finally, the required water vapour diffusion resistance factor μ is calculated as follows: ߤ ൌ ߜ௔ Ȁߜ

(16)

where: įa =

଴Ǥ଴଼ଷ௣బ ோೇ ்௣

ሺ

் ଶ଻ଷ

ሻଵǤ଼ଵ ሾ

௞௚ ௠௦௉௔

ሿ is the ambient air-water vapour

diffusion coefficient, p0 = 1013.25 [hPa], Rv = 462 [Nm/kgK], T = thermodynamic temperature [K], p = air pressure [hPa]. The measurements in dry conditions can be compromised with significant errors compared to the measurements in wet conditions, since it is not possible to maintain the air volume inside the vessel at exactly 0% RH. Besides, determining the water vapour diffusion resistance factor in dry conditions results in values approx. five times higher than those determined in a different thermodynamic state, i.e., in wet conditions.

Testing moisture content during FCB fabrication process Cellulose fibre cement board moisture content is a key parameter for providing information about the quality of the board and its treatment during the fabrication process. Thus, it is essential to measure and control the board moisture content at each fabrication step. The authors tested the moisture content of the exterior cladding and the interior cladding boards during the final fabrication step, i.e., downstream of the final drying oven [12]. The tests were carried out using a standard drying-weighing method [21, 22] and a non-destructive dielectric method (using a Trotec T650 portable meter). This handheld device was capable of measuring the moisture content distribution to a depth of 4 cm from the surface in

F Fibre Cement Bo oards Testing

59

building maaterials, walls,, ceilings, floors, etc. Non--destructive tests were carried out at the top surface of sttacked 1200 u 3050 mm m boards. Measuring ppoints locatedd at least 100 mm m from the board edges and a about 200-255 mm m from one annother, as sho own in Fig. 2--8, were selected. Each board was measured at 72 points. Approximatel A ly 300 cellulose fibre cement boarrds were testedd in total.

Fig. 2-8. Loccation of moistuure content meeasuring points on tested cellu ulose fibre cement boardd [12].

Since thhe reading deppends on the bulk density of the tested material, the authors developed a custom calibration curve, shown in Fig g. 2-9, to determine thhe moisture coontent of the tested t boards. Small specim mens were cut out in places where the t dielectric measurementts had been caarried out and those sspecimens weere tested usin ng the dryingg-weighing method m in accordance with the following proced dure. The speecimens were weighed hieve a stable gravity consttant mass. and subsequuently dried att 105qC to ach Then, a perccentage mass moisture conttent was calcuulated accordiing to the following reelation:

Chapter Two

60

wm

mw  m s >% @ ms 

(17)

where: mW = masss of the speciimen with thee actual moistuure content [g], ms = masss of the speciimen dried at 105qC [g]. Table 2-7 shows the reesults of the specimen moissture content test t for a single batchh of cellulose fibre f cement boards. b

Fig. 2-9. Callibration curve used to determ mine the moistture content off cellulose fibre cement bboards [12].

Fibre Cement Boards Testing

61

Table 2-7

board symbol PZ 1/07.08.2013/

PW 47/16.09.2013/

measuring point

moisture content in % exterior cladding board

interior cladding board

1

7.23

-

2

7.14

-

3

6.90

-

4

6.54

-

5

6.13

-

6

5.57

-

7

5.19

-

9 …

4.88 …

-

72

3.37

-

1

-

8.50

2

-

8.39

3

-

8.12

4 . .

-

7.69 . .

72

-

3.97

Figures 2-10 and 2-11 show mean moisture content distribution maps for the exterior and interior cladding fibre cement boards which were tested.

62

Chapter Two

Fig. 2-10. M Mean moisturre content disstribution mapp for exterior cladding fibre cement bboard [12].

Fig. 2-11. M Mean moisturre content disstribution mapp for interior cladding fibre cement bboard [12].

Table 2--8 shows the experimentally e y determined mass moisturre content (wm) of the ccellulose fibree cement board ds at six fabriccation stages.

Fibre Cement Boards Testing

63

Table 2-8 board state

fabrication stage (zone)

test method

determined moisture content wm [%]

liquid fibre cement mixture

1

-

100

freshly formed board

2

drying-weighing

35-50

board after pressing

3

drying-weighing

25-35

board after passing through pre-curing tunnel

4

drying-weighing

20-25

board after 14 days of maturing in tents

5

board after passing through final drying 6 oven

dielectric / drying-weighing

dielectric / drying-weighing

12-16 (exterior-cladding board) 13-18 (interior-cladding board) 3-7.5 (exterior-cladding board) 4-8.5 (interior-cladding board)

A comparison of the results presented in Table 2-8 shows that the moisture content of the board changes significantly in the course of the fabrication process. With regard to the quality of the finished board, the moisture content in the board is critical after the final stage. As is clear from the mean moisture content distribution maps for the boards tested by the authors, the distribution is uneven along the board length. This applies to the exterior cladding board, in which the final moisture content ranges from 3% to almost8 %, and to the interior cladding board, in which the moisture content is similar, ranging from4 % to almost9 %. The scatter of the moisture content results seems to be high; however, it is typical for the values measured in the boards downstream of the final drying oven. Most probably, it was due to the oven design and the location of heating elements.

64

Chapter Two

A non-uuniform distribbution of mo oisture contennt affects not only the strength paarameters of the board, but b also its further treattment, in particular im mpregnation, varnishing, or covering with other decorative d layers. The boards are characterized by poor surrface adhesio on, which shortens their service life under variablle weather connditions. Based onn the test ressults presented d here, an efffective board moisture content conntrol system was suggesteed and improovements to the oven structure weere introducedd. Thus, the heat distributioon during boaard drying and consequuently the mooisture conten nt distributionn in both thee exterior cladding boaard and the innterior claddin ng board becam me more unifform. Fig. 2-12 shows the final dryinng configuratiion in the FCB B plant.

Fig. 2-12. Finnal drying oven used in fabricaation of cellulosse fibre cement boards.

As part of the study, the mean mo oisture contentt distribution maps for the finishedd (downstream m of the finall drying ovenn) exterior and d interior cladding booards were plootted. Moisturre content meeasurements were w also

Fibre Cement Boards Testing

65

carried out for the boards at the earlier stages of the fabrication process to determine the changes in board moisture content throughout the entire production process, whereby a suitable board quality control system could be established.

Durability against warm water Ten pairs of specimens must be prepared for the test in the same way as for testing the MOR parameter in accordance with EN 12467. Each specimen pair must be cut adjacent to the fabricating machine direction from the same board and should be marked with the same number for further comparison of the test results. The first set of ten specimens is subjected to the standard bending strength test after conditioning as described above. The second set of specimens is immersed in hot water (60 + 2) qC for 56 + 2 days. The second set is conditioned after that period in accordance with the conditions listed in Table 2-4 and subjected to the bending strength test. The test results are processed to determine ten individual moduli of rupture ratio ‫ܴܯ‬௜ using the following formula: ‫ܴܯ‬௜ ൌ 

ெைோ೑೔ ெைோ೑೎೔

(18)

where: MOR fi = MOR of the i-th specimen after the warm water test, MOR fci = MOR of the i-th reference specimen. Standard deviation s and average R are determined for the population of MRi. Based on the calculations, a parameter determining resistance of the analysed specimens to warm water test RL can be calculated according to the following statistical estimation: RL = R – 0.58 s

(19)

RL is the lower estimation of the uncertainty range of the average MOR ratio after the warm water test and must be at least 0.75.

66

Chapter Two

Durability against soak–dry cycles Mechanical strength loss after the soak–dry cycles is determined using the same procedure as described in the previous paragraph on ten pairs of specimens. In this case, a procedure described below replaces immersing a set of ten specimens in hot water. The soak-dry procedure consists of a required number of immersion cycles in water at ambient temperature for 18 h and then the same number of drying cycles in a ventilated oven at (60 + 5) qC and relative humidity below 20 % for 6 h. The 20% humidity level must be reached at least 3 h prior to the end of the 6-hour drying cycle. The required number of cycles is 50 cycles for category A boards and 25 cycles for other board categories. If necessary for technical reasons, an interval up to 72 h between cycles is allowed, during which the specimens must remain immersed. After the required number of cycles, the specimens are stored at ambient laboratory conditions for seven days. The next step involves conditioning of the set of specimens per the requirements listed in Table 24 and the bending strength test. The results of the mechanical strength test are processed using standard methods. Thus, a parameter determining the resistance of the analysed specimens to soak–dry cycles RL is calculated. RL is the lower estimation of the uncertainty range of the average MOR ratio after the test and must be at least 0.75.

Moisture movement test Both the moisture movement test and the freeze–thaw resistance test require a specialized conditioning chamber. The chamber must be fitted with a ventilation system and must be capable of maintaining both the positive and the negative constant temperature within + 2° accuracy and the relative humidity (RH) within + 2% accuracy at a full load at 30% RH or with + 5% accuracy at 90% RH. A sample configuration which meets the requirements specified above is a type C 20/350 chamber manufactured by Clima Temperatur Systeme operating at 3.2 kW (see Fig. 2-13). The chamber allows operation at -20 to +180°C and at 10 to 98% relative humidity and conforms to DIN40050.

F Fibre Cement Bo oards Testing

67

Fig. 2-13. T Type C 20/3550 conditionin ng chamber m manufactured by Clima Temperatur S Systeme.

The moisture movemeent test consissts in measurinng the specim men length at low relaative humidityy and the len ngth of the same specim mens after conditioningg and a periood of storagee at increasedd humidity. The T tests should be carried out on two specimeens 300 mm loong and 75 mm m wide. One specim men is cut in parallel p to the longer dimennsion of the board b and the other is ccut at a right angle a to the lo onger dimensioon of the boarrd. The speccimens are prre-conditioned d at 30% relaative humidity y at 23°C until the weight loss or gaain during a 24 2 h period dooes not exceed d 0.1% of their weightt. After condiitioning, the length l and weeight of the specimens are recorded. Since minnor changes in the physiical propertiees of the specimens aare to be expected, the dim mensions are measured wiithin 0.02 mm accuraccy and the weiight is measurred within 0.11% accuracy relative r to the weight. After determining thee physical paraameters of thee specimens as a per the procedure ddescribed abovve, the specim mens are condditioned at 90 0% RH at 23°C. The sspecimens aree left at these conditions unntil they reach h a steady state, i.e., unntil the weighht loss or gain n in a 24 h pperiod does no ot exceed 0.1% of theiir weight.

Chapter Two

68

A relative change in specimen length as measured before and after storage at increased humidity, expressed as a percentage of the initial length, is referred to as the linear moisture movement Lm and is recorded after the test along with the test date and identification of the batch from which the specimen board was taken.

Durability against freeze–thaw cycles Ten pairs of specimens are prepared in the same way as for the warm water resistance test. The first set of ten specimens is subjected to the standard bending strength test after conditioning as specified in Table 2-4. The second set of specimens is immersed in water at ambient temperature for 48 h and is subjected to freeze–thaw cycles. The set is conditioned in accordance with the conditions listed in Table 2-4 and subjected to the bending strength test. The following number of cycles applies to each tested FCB category: 100 cycles for category A, and 25 cycles for categories B, C, and D. The freeze–thaw cycle procedure is carried out as follows. The specimens are placed in the conditioning chamber and frozen until a temperature of (–20 + 4) °C is reached in 1 to 2 h and held at this temperature for another hour. The specimens are immersed in a slowly heated water bath. The bath should reach a temperature of (20 + 4) °C within 1 to 2 h and the temperature is maintained for another hour. The specimens should be separated during cycling to allow free circulation of air and water around them. Each freeze/thaw cycle must last from 4 to 6 hours; however, the interval between two cycles, during which the specimens are immersed in water at 20°C, should not exceed 72 h. The results are processed to determine ten individual moduli of rupture ratio ‫ܴܯ‬௜ using the formula (18) presented in the previous paragraph. Standard deviation s and average R are determined for the test results. Based on the calculations, a parameter determining the specimen resistance to freeze–thaw cycles RLFT can be calculated according to the following statistical estimation: RLFT = R – 0.58 s

(20)

RLFT is the lower estimation of the uncertainty range of the average MOR ratio after the freeze–thaw test and must be at least 0.75.

Fibre Cement Boards Testing

69

Durability against heat–rain Category A and B boards should be tested using this method at 50 and 25 cycles, respectively. The specimens are fastened in a vertical position to the framing system according to the manufacturer’s instructions. The actual frame dimensions should provide a minimum surface area of 3.5 m2 and a maximum surface area of 12 m2. Two specimens should be used if the surface area of the board is greater than 1.8 m2. Otherwise, a sufficient number of boards should be fastened to cover a surface area of at least 3.5 m2. The frame with the boards fixed to it is subjected to a series of operations, i.e., a single cycle per Table 2-9. Table 2-9 Operation

Duration

water spray

2 h 50 min + 5 min

pause

10 min + 1 min

radiant heat

2 h 50 min + 5 min

second pause

10 min + 1 min

total cycle duration

6 h + 12 min

After the test, the board surface is visually inspected. Any observed cracks, delamination, warping, bowing, etc. must not affect the product performance.

Reaction to fire FCB boards made of Portland cement and fillers, including clays, silica fume, ground limestone, or fly ash, and up to 12% (wt.) organic fibres, generally feature good resistance to fire; however, due care should be taken when testing this property due to related risks to life and health. Boards containing 1% or less of organic substances (either by mass or volume) are deemed to satisfy the requirements set forth for performance Class A1 per EN 13501-1 regarding reaction to fire and, in accordance with the Commission Decision No. 96/603/EC, are excluded from the list. The boards not listed in the Commission Decision must be tested and classified in accordance with EN 13501-1 [23]. The boards are classified as Class A2 “non-combustible products.” Tests are required to determine the resulting smoke emission and production of flaming droplets/particles during combustion. The tested boards are installed on a test assembly in a

Chapter Two

70

manner corresponding to their intended use in accordance with the manufacturer’s specifications. Five examples of suitable test assemblies can be found in paragraph 7.5.2.2 of EN 12467. The test assembly is a corner set-up made out of two timber frame supporting structures, 1.5 m high each, to which the FCB boards are fastened. One of the frames forms a long wing 1 m long, while the other forms a short wing half of that length. The test assembly is installed inside an apparatus capable of performing a single burning item test according to EN 13823. The tested boards are exposed to an open fire created by a propane gas burner with a thermal output of 30 kW placed at the bottom of the corner formed by the test assembly. The test specimen and the burner are located under a hood in an enclosure. A thermally insulated smoke exhaust pipe is installed in the hood. A set of sensors is installed inside the exhaust pipe, including a hot air flow sensor, a set of air flow temperature sensors, and an optical smoke density sensor. The single burning item test is carried out in accordance with the following procedure. An initial constant air flow of 0.6 m3 is maintained inside the testing enclosure. After 5 minutes of preheating the test assembly with a low power burner, the assembly is heated at full power for 1560 seconds. Within that period six parameters are recorded and evaluated to classify the product specimen in certain specific categories. The parameters required for Class 2 reaction to fire are listed in Table 210. Table 2-10 measured parameter

requirement in respect to Class A2

fire growth rate [W/s]

< 120

lateral wing flame spread

the flame must not reach the outer edge of the wing

total heat release [MJ]

< 7.5 within 600 seconds from recording

smoke production rate [m2/s2]

< 30 plus total smoke production of 50 m2 per 600 s indicates level s1 < 180 plus total smoke production of 200 m2 per 600 s indicates level s2 higher quantities indicate level s3

Fibre Cement Boards Testing formation of flaming droplets / particles detected visually per 600 seconds

71

no droplets detected indicates level d0 detected droplets, but faded within 10 s each indicates level d1 detected droplets, active for more than 10 s indicates level d2

Products not tested in accordance with EN 13501-1 which contain more than 1% organic matter are classified as Class F.

Information on product parameters determined based on standard tests and indicated on the product label Essential information on product parameters are included on the packaging: – – – – – –

identification of the manufacturer, European Standard for fibre cement boards, i.e., EN 12467, product size and/or name, category and class according to EN 12467, date of manufacture, “NT” mark for non-toxic products, – class of reaction to fire.

FCB testing using ultrasound Ultrasound testing (UT) methods can be used to quickly determine bulk FCB density or to detect local delamination due to fabrication defects. Other authors have reported the use of low-frequency ultrasound (50-200 kHz) to identify small defects (a few centimetres in size) in concrete structures on-site [24, 25]. However, low frequency range ultrasound cannot be used to detect smaller defects in relatively thin fibre cement boards. The experiments performed by the authors revealed that the required sensitivity of the ultrasound methods used to determine the structural properties of FCB is achieved when the ultrasound wavelength is similar to the size of the defect. The wavelength Ȝ is in the following relation to the frequency f of the emitting source and propagation velocity c of the traveling ultrasonic longitudinal waveform: Ȝ=c/f

(21)

72

Chapter Two

Thus, foor the propaggation velocitty of 1000-26600 m/s reco orded for FCBs, the auuthors recomm mend 1 MHz freequency to achhieve the propaagation of wavelengths in the rangee of 1-2.5 mm m. To test thhe applicabilitty of the described m method, the autthors carried out o a series off tests using a UTC110 ultrasonic teester by Eurosonic. The tesster was conffigured to opeerate with ultrasonic trransducers at a frequency beetween 1 and 25 MHz. Fig. 2-144 shows the teest configuratiion including Videoscan traansmitting and receiviing transduceers, emitting a 19-mm uultrasonic beam at a frequency oof 1 MHz. Thhe operating frequency f wass selected to minimize the attenuatiion of the longitudinal elasttic waves proppagating in th he fibrous material tested. A custom m designed holder h with arrticulated join nts and a spring was ddeveloped to ensure correcct coupling beetween the traansducers and the surfface of the tessted board. Fiig. 2-15 show ws the detailed d view of the holder.

Fig. 2-14. Ultrasonic test configuration c for f FCB includding the ultraso onic tester connected to a laptop, a referrence specimen n, and the transdducers holder.

F Fibre Cement Bo oards Testing

73

Fig. 2-15. D Detailed view of o the custom designed holdder for correctt coupling between the uultrasonic transdducers and both h sides of the booard.

Ultrasonnic waveform velocity v1 caan be determinned using the described d configuratioon and two different d metthods. The fi first is a tran nsmission method usinng two transduucers, the firsst emitting andd the second receiving the waveforrm. This methhod relies on the t measurem ment of time difference d between tim me t0, when thee transmitting g transducer iss triggered, an nd time t2, when the w waveform reacches the secon nd transducerr. Thus, the waveform w velocity cann be expressedd as follows: v1 = d / (t2 – t0)

(22)

where: d = boardd thickness The metthod describedd above is afffected by a s ignificant erro or t1 – t0, because afteer triggering t0 a dead time required to ggenerate the waveform w can be obseerved. The measurements m performed byy the authors revealed that the aveerage t2 – t0, i.e., a measu ured waveforrm time of fliight, was approx. 7 μss, whereas ann unbiased wa aveform time oof flight was approx. a 6 μs. The trannsmission metthod of measu uring the wavveform velocitty is used for the low w-density booards, where high disper ersion of pro opagating waveforms ooccurs. For the majority of FCB boardss characterizeed by a low w/medium dispersion oof ultrasound, a more accuraate second echho method can n be used

74

Chapter Two

to measure the ultrasonic waveform velocity v2. T This method uses two transducers placed on botth sides of the tested board. The propagatting wave travels across the bulk material, m is refflected from tthe edge of th he board, and then retturns and travvels along thee same path oonce more. Therefore, T two echoes are recorded by b the receiviing transducerr. In this case, the time difference bbetween the arrrival of the fiirst and the seecond echo t2 – t1 is not affected by any unwantedd delays. The time differennce between th he arrival of the first aand the seconnd echo indicaates double crrossing of thee material thickness, heence: v2 = 2d / (t2 – t1)

(23)

Fig. 2-166 shows a schhematic diagraam of both meethods for dettermining the ultrasonnic waveform velocity, and the actual m measurement reesults are shown in Fiig. 2-17 and 2-18. 2 Fig. 2-17 and 2-18 shhow a short waveform w immediatelyy after triggeriing the transdu ucer. The effeect is due to th he energy of high enerrgy pulse leakk from the tran nsmitting transsducer to the receiving transducer. The initial waveforms w do not affect the determin nation of velocity, sinnce the signal is processed at a preset gatedd time intervalls.

Fig 2-16. Diaagram of the reeflection and trransmission meethods for meaasuring the ultrasonic waaveform velocityy.

The ultrrasonic materiial tester auto omatically me measures the waveform w time of flighht and may be used in both methods. m The measurementt involves preparing thhree gates foor measuring the waveform m arrival tim me with a preset ampllitude and tim me range usin ng an intercoonnection betw ween the tester and thhe laptop. Affter several tests on typicall boards, it was w found that the arrivval time t1 appplying the seccond echo moode must be within w the

Fibre Cement Boards Testing

75

range of 4-7 μs and the arrival time t2 must be within the range of 12-18 μs. Additionally, the arrival time for the third gate in the transmission mode must be set to 7-12 μs. The gates must register the occurrence of the ultrasonic signal when its amplitude exceeds a preset level (10% of fullscale range) in reference to 40 dB of the tester receiver amplifier. After setting the parameters per the procedure above, the tester operates the device in automatic mode to quickly determine the waveform velocity. To correct the coupling between the rough board surface and the transducers and to avoid moisture diffusion into the board, both silicone grease and a thin Scotch tape fixing was applied. The authors performed a series of measurements to evaluate a range of uncertainty of the measured arrival time. A 25-mm thick steel calibration block conforming to EN 12223 and ISO 2400 requirements was used. The manufacturer of the calibration block declares that the longitudinal wave velocity measured inside the block is exactly 5920 m/s with the uncertainty below + 0.5%. The authors obtained the following results after ten measurements: -

average value of t1 = 5.519 us, with standard deviation 0.58% of the average, average value of t2 = 13.824 us, with standard deviation 0.3% of the average.

Based on the results and the technical specifications of the tester and the calibration block, the authors evaluated the uncertainty of determination of wave velocity using the described method as approx. + 1% after at least ten readings for specimens with smooth and parallel surfaces. The uncertainty for measurements of typical FCB boards with rough surfaces is approx. + 2%. Fig. 2-17 shows the results of the Class A FCB tests (apparent density = 1834 kg/m3, thickness ~ 9 mm). Since the second echo method was used, gates 1 and 2 report valid data (see the report at the top of the screen in the figure below). The final segment of the oscillations due to the first reflection occurs in the third gate region. For the data quoted (t1 = 5.66 us, t2 = 15.09 us), the corresponding ultrasonic wave velocity is 1909 m/s.

76

Chapter Two

Fig. 2-17. W Waveforms recorrded by the ultrasonic materiial tester during Class A board testing..

Fig. 2-188 shows the reesults of the Class C C FCB tests (apparen nt density = 1140 kg/m m3, thickness ~ 8.2 mm). The transmisssion method was used and only a single reflecttion could bee determined due to the siignificant signal attenuuation. The thhird gate reporrted valid dataa, since the pro opagation speed was rrelatively low w. The corresp ponding ultrassonic wave veelocity in this case woould be 1136 m/s. m No distinnct correlationn between thee bending strenngth and the ultrasonic u wave velociity was determ mined by the authors. a Howeever, some dep pendence (cross correelation coefficcient R = 0.7)) between thee apparent den nsity and the wave veelocity was obbserved. Fig. 2-19 2 shows thhe dependencee between these two pparameters as determined for f the sevenn board types listed in Table 2-6. T The effect sugggests the abillity to test thee board density y and the occurrence of local delaamination and d fibre curls in the board d volume indirectly ussing the ultrassonic wave vellocity measureement techniq que.

F Fibre Cement Bo oards Testing

77

Fig. 2-18. W Waveforms recorrded by the ultrasonic materiial tester durin ng Class C board testing characterized by b low apparentt density.

Fig. 2-19. Reelation betweenn the apparent density of the board and the ultrasonic wave velocityy determined byy the authors.

78

Chapter Two

The effeect of a local >10% decrease in the avverage ultrasonic wave velocity migght be appliedd to detect strructural defectts in the tested boards. Fig. 2-20 shows exam mples of th he readings obtained using the instrumentattion developeed by the authors. a A m more advanced testing technique fo for detecting structural deffects will be presented in the next paragraph.

Fig. 2-20. Deetecting local board defects by y recording the ultrasonic wav ve velocity decrease >10% of the averagge value. Locall wave velocityy decrease is maarked with arrows and ciircles.

Application of Lamb L wavess for detectting structu ural defeccts The applicaation of longiitudinal ultrassonic waves for testing FCBs was described inn the previous paragraph. Th hese waves caan be applied when the dimensions of the tested items are sig gnificantly grreater than the applied wavelength.. The theory of the propaagation of low w-frequency and a large elastic wavees in plates waas developed by Lamb [26]]. The Lamb waves w are implementedd into the buulk of the tessted plate usiing wedge traansducers placed diagoonally to the item’s i surfacee. The team leed by Prof. Kaaczmarek of the Instittute of Mechaanics and App plied Computeer Science, Bydgoszcz University, P Poland and Assoc. A Prof. Scchabowicz of W Wrocáaw Univ versity of Science andd Technologyy have design ned a prototyype for an in nnovative scanner usinng the Lamb waves appro oach to test ccellulose fibre cement boards [27, 228].

F Fibre Cement Bo oards Testing

79

Fig. 2-21. Diiagram of the test t set-up for testing cellulo se fibre cemen nt boards: (top) diagram m; (bottom) actuual test set-up [2 28]. .

The authhors researcheed the applicaability of the apparatus for testing Fig. 2-21 sh fibre cemennt boards in the fabrication process. F hows the scanner for ddetecting structural defects. The appparatus is insttalled directly y on the FCB B fabrication line. The propagating Lamb wavee is introduceed via an airr gap into th he tested material witthout any conttact between the t transmitterr (T) at one en nd of the board, and thhe receiver (R R) at the other end of the booard. The Lam mb waves are generateed indirectly using u the long gitudinal wavees from air inttroduced into the boaard by the traansmitter (T) in the directiion of its surfface at a critical angle D; the anglee depends on the t propertiess of the tested material and, in this case, it is 15qq. The receiveer (which recoords the wavee leaking on to the boarrd surface at the t same from the maaterial) is inclined in relatio angle as thee transmitter. The transduccers were locaated approx. 100 mm

80

Chapter Two

above the board moving at a rate of approx. 20 mm/s. The apparatus was installed on an experimental FCB fabrication line. The tests were carried out using Ultran Group NCG ultrasonic air transducers with a basic frequency of 100 kHz, a transducer positioning system, a TiePie Engineering Handyscope HS3, a signal generator with a two-channel oscilloscope with a signal sampling frequency of up to 10 MHz, a signal amplifier, and a PC with dedicated control, data acquisition, and analysis software. The transmitter (T) on one end of the board was activated with a chirp signal with linearly modulated frequency. In this case, the frequency range of 80-120 kHz was applied to introduce a longitudinal wave with the required energy into the tested material and to generate the Lamb wave in the material. After the Lamb wave passed through the 1200 mm wide board, it was recorded by the receiver as a leaky Rayleigh wave (LRW). The signal was correlated with a reference signal, i.e., the signal sent to the transmitter from the signal amplifier, and the waveform with an amplitude of 0.1-2 V was obtained for further analysis. Overall, 160 signals at a 20-mm interval were recorded for each board. The board moved in the direction indicated by the black arrow in Fig. 2-21. The time signals for the Lamb wave propagating in the tested board were recorded during the tests. The amplitude of the signals at the 160 measuring points was the main parameter used in the analysis of the quality of the boards. The peak-to-peak (P2P) amplitude versus signal recording position was plotted. Since the large variations in the amplitude of the signals were due to the air fluctuations caused by the movement of the board, the data were averaged using the moving average (5 sampling points). Fig. 2-22 shows the sample measurement results for one of the tested boards with three defects detected. The signals recorded for each measuring point are shown at the top as a so-called brightness modulation presentation mode. The specific measuring points (along the entire length of the board) are marked on the horizontal axis, while the vertical axis represents the successive numbers of the set of signal samples, travelling parallel to the board width. The areas of potential defects are marked with black arrows.

Fibre Cement Boards Testing

81

Fig. 2-22. Results of the measurements obtained for the board: (top) scan in brightness modulation mode; (bottom) average signal amplitude P2P as a function of measurement position. The areas of potential defects are marked with black arrows [28].

Fig. 2-22 shows the diagram of averaged P2P signal amplitudes versus measurement position, also parallel to the board width (bottom). A significant decrease in the signal amplitude can be observed in the defective areas. The experimental results presented above prove that the defects in the board material would manifest themselves as sudden changes in the amplitude of the Lamb wave. Slow changes in the wave amplitude as a function of measurement location were observed in the course of the measurements. The changes were due to the differences in

Chapter Two

82

moisture content between the central section and the edges of the board. Therefore, the moisture content of the boards as a function of measurement location was determined independently using the dryingweighing method and compared with the results of the ultrasonic tests. The authors carried out the tests on a large population of boards to identify the material defects that might have arisen during the fabrication process. A relationship between the Lamb wave amplitude and the signal recording position was determined and analysed. Fig. 2-23 shows the diagrams of Lamb wave amplitudes recorded when scanning a set of 28 cellulose fibre cement boards selected from a single production batch. However, the nature of the observed amplitude variations within the boards has not been studied in detail. Most probably, it is due to the local material inhomogeneities resulting from the complexity of the process. An i–th parameter to determine the defect size Zi was calculated for each board according to the following statistical estimation: Ai  P V ,

Zi

(24)

where:

Ai

= mean amplitude of the Lamb wave in the board, P = mean amplitude in the reference board (without defects), V = standard deviation of the amplitude in the reference board. For boards of acceptable quality which meet the relevant requirements,

Z dr2

). Otherwise, the the Z value must be in the range from -2 to 2 ( i entire board is treated as defective, and local decreases in wave amplitudes are identified by checking if the following condition is present, namely, if the amplitude decrease at a given board position is higher than 3dB compared to the mean amplitude A i , the material defects at the position for which the decrease was identified. The sample results of Z values for a single batch of boards are shown in Fig 2-24, where the Z value for the board no. 11 is lower than -2, and so the board is treated as defective. Fig. 2-25 shows the measurement results for another set of 14 boards (bottom). A material defect – in this case a delamination, shown at the top (Fig. 2-25) – was identified in one of the boards using the non-contact ultrasonic method as part of the automatic inspection procedure. Fig. 2-24 shows the average Lamb wave amplitudes and standard deviation of the

F Fibre Cement Bo oards Testing

83

arithmetic avverage as a fuunction of the position of thhe set of 14 bo oards plus the amplitudde distributionn for the boarrd in which thhe defect was detected. A sharp local decrease inn the wave am mplitude alongg the board leength at a distance of aapprox. 1550 mm from its edge e was obseerved, contribu uting to a significant ddeviation from m the average for f the entire sset of boards.

Fig. 2-23. Diaagrams of the Lamb L wave amp plitudes recordded when scanning the set of 28 cellulosse fibre cement boards from a single productioon batch [28].

Fig. 2-24. Reesults for determ mining the parameter definingg the defect sizze Zi for a single batch oof boards. Boardd no. 11 is treatted as defectivee [28].

84

Chapter Two

The top picture in Fig. 2-25 shows an area of increased signal attenuation probably due to delamination at this location. Another type of imperfection analysed in the FCBs by the authors were cracks, which usually occur near the board edges, mainly as a result of transporting the boards on the rollers in the fabrication process. Fig. 2-26 shows the sample measurement results for several boards, including one with a crack as indicated by the Lamb wave amplitude graph highlighted in black. Compared to the amplitude graphs recorded for the other boards, the increase in signal occurring in a defective component involves a shift in the horizontal position by approx. 400 mm. A picture of the defective area of the board (Fig. 2-27) validates the results obtained using the non-contact ultrasonic method. Other tests carried out by the authors showed that the moisture content has a significant effect on the testing of the boards with the Lamb waves. In order to determine this effect, the moisture content tests were carried out using a non-destructive dielectric method described above. A Trotec T650 non-invasive dielectric capacitance moisture meter was used.

F Fibre Cement Bo oards Testing

85

Fig. 2-25. (T Top) the defective area determ mined based onn the largest decrease d in signal; (bottom) the measureement results fo or a set of 14 bboards taken fro om a batch different from m that shown in Fig. 2-24 [28].

86

Chapter Two

Fig. 2-26. Meeasurement resuults for a set off boards, one off which includeed a defect (crack) as inddicated by the Lamb L wave amp plitude graph hiighlighted in blaack [28].

Fig. 2-27. A ppicture of the defective d area of o the board testted using the Lamb L wave method, incluuding a crack coorresponding to o the position hiighlighted in Fiig. 2-26.

Overall, 48 measuringg points weree arranged at 2250 mm centtres along the tested booards. The deestructive moiisture contentt tests were caarried out on random specimens to verify the reesults obtaineed by the non n-invasive method. Squaree (250 u 250 mm) specimeens were cut out from dielectric m the boards, aand the moistuure content in n their volumee was measureed using a standard meethod by deterrmining the lo oss in specim men weight aftter drying on a tray over a source off heat, in accorrdance with thhe procedure described d in the previoous paragraphh. Figure 2--28 shows the results for a set of boards inn which the Laamb wave amplitude/nooise ratio was used u to detect the t amplitude ((lines at the top p).

F Fibre Cement Bo oards Testing

87

Fig. 2-28. Lam mb wave P2P amplitudes a verssus position meaasured in standard boards and in boardss with increasedd moisture conteent [28].

The grouup of lines at the t bottom corrresponds to tthe boards in which w the Lamb wavee amplitude was w too low to detect thhe defects du ue to the increased m moisture contennt.

Fig. 2-29. (Top) relationshipp between the moisture conteent; (centre) am mplitude of me of flight of the t Lamb wavee. Values measu ured in the the Lamb wavve; (bottom) tim FCB along itss length in the boards b stored in n unsuitable connditions [28].

88

Chapter Two

Some boards tested by the authors have revealed a similar and highly non-uniform spatial moisture content distribution (Fig. 2-29). The test results shown in Fig. 2-29 indicate that, in some cases, the Lamb wave amplitude could not be measured at a level required to detect the defects. The amplitude of the Lamb wave on the edge of the board was higher than at its centre or at its end, mainly due to the different moisture content, verified using the capacitive method. The differences were most probably due to the storage of the manufactured boards in unsuitable conditions. In accordance with the results determined by the authors, the Lamb wave velocity increases with the decrease in the moisture content. The behaviour is atypical for cement-based materials (e.g., concrete). Research with mostly contradictory results are available [29, 30]. However, more detailed studies of the influence of initial saturation and water content on the bulk and surface wave velocity and attenuation coefficient have documented a nonlinear behaviour of both the wave parameters on the water content. These studies have proven that in a low water content range (0-8%) a decrease in velocity can be observed (and a corresponding increase in attenuation), whereas above this low water content range, the inverse is true.

Fibre Cement Boards Testing

89

Applications of advanced optical microscopy

DARIUSZ JARZĄBEK INSTITUTE OF FUNDAMENTAL TECHNOLOGICAL RESEARCH, POLISH ACADEMY OF SCIENCES Due to its complexity, the microstructure of fibre concretes is difficult to visualize using conventional methods. On the one hand, due to the often complex 3D topography, it is usually impossible to focus on the entire surface of interest using a conventional optical microscope. Also, the fibres may protrude over the surface of the tested specimen up to a few hundred micrometres, making optical imaging even more challenging. On the other hand, scanning electron microscopy (SEM) techniques generally provide a higher depth of field; however, it requires the non-conductive specimens to be coated. Unfortunately, the thickness of the coating can obscure important details. Also, SEM techniques are limited to specimens that are small enough to fit inside the vacuum chamber and that can handle moderate vacuum pressure. They are also large and expensive; they must be housed in an area free of any electrical, magnetic, or vibrational interference; and they require handling by a knowledgeable and experienced staff. A promising alternative for the conventional methods that overcomes most of these disadvantages is extended focal imaging (EFI) optical microscopy [33]. This method enables the imaging of large areas by manual displacement of the specimen stage or by fully automated software aided scanning. Fig. 2-30 shows a DSX 510 type Olympus EFI camera. Fig. 2-31a and 2-31b show two conventional optical microscope images of the same broken FCB surface generated at two different focus planes. The surface of the matrix is visible; however, the protruded fibres are blurred in Fig. 2-31a. On the other hand, in Fig. 2-31b only the fibres are in focus. THe EFI technique was used to create the image shown in Fig. 231c. In this case, both the fibres and the matrix are clearly visible. The EFI procedure involves taking several images while the point of focus is moved down and the vertical (z) position is precisely measured. Using these images, the areas where the specimen is in focus are combined into one image with the entire specimen in focus. Using the data about the z position, a 3D image of the tested specimen can be created. This approach eliminates issues with surfaces that are not perfectly parallel to the focal plane, surfaces with complex 3D topography or surfaces with protruding fibres.

90

Chapter Two

Fig. 2-30. General view of DSX 510 type Olympus EFI camera.

The optical components of a modern microscope are very complex. To provide a well functioning microscope, the entire optical path must be precisely and carefully set up and controlled. The simplest method for illuminating the optical microscopy is referred to as the bright field. In this mode, the light path comes from the light source, passes through the objective lens, and is reflected off the surface of the specimen. Then, it returns through the objective lens and reaches the camera. Unfortunately, bright field microscopy usually has low contrast, which sometimes makes it difficult to evaluate the microstructure of fibre concrete. However, there are many other techniques for changing the path of the light in optical microscopes.

F Fibre Cement Bo oards Testing

91

Fig. 2-31. Coomparison betw ween conventional optical miccroscope imagees and EFI images: (a) thhe surface of the matrix is visib ble but the prottruding fibres are blurred; (b) only the ffibres are in foccus; (c) EFI image of the sampple in polarized d light; (d) EFI image off the sample in dark d field.

Modificaations improvve the contrast of the sam mple image. The T most common tecchniques for improving i thee image contrrast include dark d field, differential interference contrast, c and polarization. Two techniq ques, i.e., dark field annd polarizatioon, are of partiicular importaance in regard ds to fibre concretes.

92

Chapter Two

The darkk field is an illumination technique inn which the light l path passes throuugh the outer hollow ring of the objectiive lens and then t falls onto the speecimen at a hiigh angle of incidence. It rreflects off thee surface, passes throuugh the interioor of the objecctive lens, andd ultimately reeaches the camera. Figg. 2-31d show ws the imagee of the surfface of fibre concrete created in tthe dark fieldd. The dark field fi is highlyy effective in showing small voids on the tested surface. How wever, the mosst promising method m of optical imagging for the fibbre concretes is polarizationn.

Fig. 2-32. Comparison between two EFII imaging techhniques, bright field and polarization: (a) bright field;; (b) polarizatio on.

In this teechnique, the light path is the same as iin the bright field, but the contrastt is improveed by the po olarizers. A nnatural light wave is characterised by the light waves with w random vibration directions. d Polarizationn filters only let through liight waves viibrating parallel to the direction off transmissionn. The contrasst may be inccreased, if th he sample between thee polarizers chhanges the vib bration directtion of light. Fig. 2-32 shows the comparison of o the resultss of bright ffield (Fig. 2-32a) and polarization (Fig. 2-32bb) imaging. The differencce is clearly y visible. Polarizationn imaging proovides much better contrasst between th he matrix and the fibrres. Also, thee images in Fig. F 2-31a-c w were created using u the polarization illumination technique. Polarization allso works for different colours of tthe matrix. Fig. 2-33 show ws the image s of coloured d FCB in polarized ligght. The fibres are as clearlly visible as inn the case of the white matrix.

F Fibre Cement Bo oards Testing

93

wo images at diffferent magnificcation of a colooured fibre cemeent board. Fig. 2-33. Tw

FCB boaards can be reinforced with both cellulosse and PVA fibres. The polarization illumination technique allo ows both kindds of fibres to be easily identified. F Fig. 2-34 show ws both kinds of fibres and tthe differencee between them can bee identified inn in coloured d micrographss. The cellulo ose fibres have a charracteristic yelllow or brown n colour. In ccontrast, the dark d field technique dooes not allow one to disting guish betweenn the differentt kinds of fibres.

Fig. 2-34. D Distinguishing between two different kinnds of fibres using the polarization illumination tecchnique.

94

Chapter Two

Implemeentation of thee motorized z stage with a position meaasurement and control enables the capturing c of accurate a 3D im mages. Moderrn optical microscopess are fitted wiith z stages with w position reesolution belo ow 1 μm. 3D imagingg techniques can c be of partticular importtance when evaluating e the shape of a single fibbre of compossite materials.. In this case,, the best results are aachieved withh the dark field illuminationn. Fig. 2-35 shows s the comparison between the polarization (Fig. ( 2-35a) aand the dark field f (Fig. 2-35b) imagges of a singlle PVA fibre. The dark fieeld images sh how more detailed surffaces of the analysed a fibree, and the textture of the su urface can be visualizeed within an accuracy com mparable to thhat obtained using u the SEM technnique, while the details of o the texturre can sometimes be concealed unnder the metaal coating in th he polarizationn technique.

Fig. 2-35. Coomparison betw ween the images of a PVA fibree in polarized liight and in dark field: (a)) polarization; (b) ( dark field.

Furtherm more, 3D imagges of a singlee PVA fibre aare shown in Fig. F 2-36. 3D techniquue allows onee to precisely determine thhe shape a fibre and to measure its diameter.

A fibre: (a) the entire e fibre; (b) m magnified fibre surface. s Fig. 2-36. 3D images of a PVA

F Fibre Cement Bo oards Testing

95

Advanceed dedicated microscope m so oftware allowss the stitching g together of the microoscope imagess. The techniq que allows onee to capture th he images of large sam mples that do not n fit into thee microscope’ s field of view w. Fig. 237a shows tthe example of o a stitched image. The ssoftware inclu uded with the most addvanced microoscopes can also a stitch toggether 3D im mages and measure thee geometry of an entire sam mple (Fig. 2-377b). A disadv vantage of this techniquue is that it can c take severral hours to geenerate the im mage of a large samplee; however, thhe measurement can be autoomated.

Fig. 2-37. Stitching imaage section of o 5-mm lonng sample: (aa) planar; (b) 3D stitching.

To recappitulate, due too the recent developments d in optical miccroscopy, it has becom me a useful toool for the anallysis of FCB m microstructuree and will be fully utillized in the neear future. Ex xtended focal imaging allow ws one to capture images with a prractically unlim mited depth oof view. In that aspect, the EFI teechnique exceeeds the cap pabilities of the SEM teechnique. Furthermoree, optical imagging does nott require any sample coatin ng, which is a huge addvantage overr electron miccroscopy. It allso allows thee imaging of very larrge samples by stitching g together inndividual imaages and generating 33D images. Allso, the variou us illuminationn techniques allow a one to improve the contrast of the images and to disstinguish betw ween the matrix and tthe different kinds k of fibres used in the annalysed concrrete.

96

Chapter Two

Examples applications of scanning electron microscopy (SEM) The SEM technique has been extensively used to study the microstructure of fibre reinforced composites and to analyse fibre distribution and evaluate fibre-matrix bonding. The advantage of the SEM technique compared to light microscopy is its ability to create micrographs at high magnification (> 1000) with the acceptable resolution necessary for visualizing the interaction of small cementitious compounds with the surface and the lumen of single fibres approx. 30 μm in diameter [7, 31]. However, unlike light microscopy, the natural colours of the specimen are not reproduced and some vital information is lost. Usually, before analysis, the specimens are conditioned for several hours to maintain a specified level of moisture content and fractured to expose the fibres protruding from the broken surface. More extensive studies on fibre and matrix interactions using the SEM technique can be found in [7, 9, 32]. Coutts and Kightly [9] reported the following research. A microstructure of cementitious composite containing 2% (wt.) kraft made of Pinus Radiata, after preconditioning in an oven at 105°C for 24 hours, was analysed at 250X magnification. The broken fibre endings and the matrix material attached to the fibres were the proof that the dry composite constitutes a strong fibre-to-matrix bond. The other conditioning mode, i.e., 24 hours of immersion in water, resulted in a weak fibre-to-matrix bond. The micrographs for the latter conditioning mode showed large, pulled-out, and twisted fibres, mostly devoid of the adhering matrix. Coutts and Kightly also performed the test after conditioning of the same composite at 50% relative humidity and 22°C, i.e., intermediate conditions in respect to the conditions specified above. The examples of both fibre fracture and fibre pull-out were observed in the micrographs. The research described above is the evidence that the mechanical properties of the fibrous composite can be determined by analysing its microstructure at a specific magnification. The authors prepared fractured specimens to generate SEM micrographs using a Quanta FEG-250 Scanning Electron Microscope installed in the Wrocáaw Technological University. The specimens were made of Class A fibre cement composite, containing cement, fillers, 6% (wt.) cellulose fibres, and 1% (wt.) PVA fibres. The samples were preconditioned at 50% relative humidity and 22°C to allow for different modes of decomposition of the composite microstructure. Fig. 2-38, 2-39, and 2-40 show the micrographs at 250X magnification. Different types of fibre-to-matrix bond mechanisms can be observed.

F Fibre Cement Bo oards Testing

97

Fig. 2-38. Exaample of pulledd-out PVA fibree from the matrrix of Class A FCB. F

Fig. 2-39. Exxample of partiaally pulled-out cellulose c fibres from the matriix of Class A FCB.

98

Chapter Two

Fig. 2-40. Exxample of partiaally pulled-out cellulose c fibress with the matriix material attached to thheir surfaces: thhe specimen is made of the saame material ass shown in Figs. 2-38 andd 2-39.

X-rayy microtom mography (m micro-CT) application n in visualizzation of fiibre distrib bution and ccrack detecction Micro-CT hhas been usedd to visualize the microstruucture of cem mentitious materials for the last seveeral decades [34–36]. The eequipment forr material testing usingg X-ray tomoographic techn niques is curreently manufacctured by several com mpanies and iss capable of performing p tessts on small sp pecimens only few miillimetres in size s or large items i up to a few meters in i length. The devicess include a micro m focal sou urce of X-rayy radiation, a movable table for thee specimen, and a a flat pan nel with a raddiation detecto or with a resolution off 2000 x 20000 pixels. Fig. 2-441 shows a schematic diagram d of a standard laaboratory measuring ssystem. The analysed a micrrostructure cann be visualizeed on the cross-sectionns (tomogram ms) of the tested specim men using greyscale convention rrelated directlly to the amount of local rradiation abso orption by the materiall. The greysccale covers ap pprox. 200 grrey levels an nd can be ordered froom white (maximum ( absorption) a to black (m minimum absorption). Unhydrated cement particcles and aggreegate grains are a highly absorbent m materials. Thhe hydration products thaat cover mosst of the cementitiouss matrix shoow slightly lower l absorpption. The fillers and

F Fibre Cement Bo oards Testing

99

hydrated callcinates featurre an even low wer absorptionn, while the fibres f and high porositty areas are at a the end of the scale. Thhe image reso olution of microtomoggrams usually varies from 1 to 10 μm perr voxel (volum metric 3D pixel). Thee advantage of the micro o-CT techniqque is the ability a to reconstruct 3D images of the analy ysed objects aand to determine the volumetric fraction of the material occupied bby the bulk k matrix, aggregates, voids, cracks, fibres, etc. The T preparatioon of specimeens is not laborious, w while the scannning procedurre is performeed at room tem mperature and under attmospheric prressure.

Fig. 2-41. Opperating principlle of the X-ray microtomograpph.

The authhors analysedd three specim mens made oof three differrent FCB compositionns [37]. Acccording to th he requiremennts of the micro-CT m scanning prrocedure, i.e., 100X magnification, cylinndrical cores 7 mm in diameter andd 7 mm in height were cut from the boar ards by drilling g. The 10 mm thick sppecimens weree extracted fro om the differeent fibre cemeent panels supplied byy the manufaccturer. Prior to t analysis, th the panels weere tested using standaard proceduress to evaluate their t performaance. Materiall “A” was characterizeed by low moisture absorbaability nw (8-110%), and thee bending strength of panels was 11-13 1 MPa. Based B on the data provideed by the manufactureer, the concenntration of fibrres by weight in that materiial is 6 % for good quuality 2.5 mm m long cellulo ose fibres. M Material “B” was w made using the saame technoloogy as materiial “A,” but it contained 6% (wt.) recycled celllulose fibres 1 mm long, and the bendding strength of panels was 8.5-10 MPa. Materrial “C” was characterizedd by a lowerr bending

Chapter Two

100

strength of 33-4 MPa and contained 4% % (wt.) good qquality cellulo ose fibres as a 2.5 mm m long raw maaterial. Due to the fabricatioon process, disstribution of the celluulose fibres was w uneven in n the last maaterial. The sp pacing of fibres was non-uniform, and several fibre agglom merates (“curlls”) were present in thhe bulk materrial in type C panels, identtified by obseerving the fractured areeas of the matterial. Fig. 2-4 42 shows the ttested specimeens.

Fig. 2-42. Speecimens A, B, and a C tested by y the authors usiing micro-CT teechnique.

The maaterial was tested t using a Nanotom m 30 microto omograph manufactureed by General Electric at the Institute of M Materials and d Machine Mechanics in Bratislavaa. Fig. 2-43 shows s the deevice. The to omograph allows one to scan objects up to 30 3 x 30 x 3 0 cm. The following f parameters w were set: lampp voltage - 115 kV; lamp cuurrent - 95 miicroamps; shot exposittion time - 7550 ms. The eq quipment wass capable of producing p the followinng data sets in order to define the microstructurre of the specimen: -

-

digitaal specimen im mages represeenting cross-seections of testted object in trransverse or lengthwise direction d to tthe mean axiis of the cylinndrical specim men, threee-dimensional projections of the speecimens show wing the defecctive areas of the t tested specimen.

Fibre Cement Boards Testing

101

The spatial resolution of the reconstructed microstructure was 4.3 μm3 per voxel. The result of the micro-CT scanning was a set of tomograms (specimen cross-sections), recorded for every 5 μm of the specimen height. The set of tomograms consisted of 1200 cross-sections 1200 x 1200 bytes each. Fig. 2-44 shows the sample specimen cross-sections illustrating the material microstructure with clearly visible differences in fibre shape. The black coloured fibres in the image of specimen B are shorter and thinner than those shown in the image of specimen A. A large fibre agglomerate can also be observed in the third image. The recorded data allows one to present a greater number of crosssections within the single graph by using a special projection. However, the total number of bytes related to the consecutive voxels in such a dataset would be 1.5 x 109 bytes, and would require additional lengthy processing. Therefore, a specific subset was taken from the entire dataset, preferably cube-shaped for further processing. The resulting subset is referred to as a ROI (Region of Interest) and might include cracked areas in the proximity of the specimen surface, usually damaged by a drilling tool. The study included 3 ROIs 9303 (804,000,000) voxels each, representing 4 x 4 x 4 mm cubes and virtually extracted from the analysed specimens.

102

Chapter Two

Fig. 2-43. Nanotom 30 type miccrotomograph: (top) generral view; (bottom) X-rray source withh tungsten diap phragm and cyylindrical FCB specimen placed on the rotating turntabble.

Fig. 2-44. Crooss-sections of the analysed sp pecimens: (left)) A; (centre) B;; (right) C. The differencces in the fibre shapes s and fibree distribution arre visible.

Fibre Cement Boards Testing

103

Figure 2-45 shows the microstructure of the 4 x 4 x 4 mm cubes (ROIs), virtually extracted from the analysed specimens. The projection shows the voxels on three virtual cube surfaces. Therefore, more information on the investigated structures are shown graphically; however, this is still a small fraction of the collected data, i.e., three outer walls of the cubes.

Fig. 2-45. Representation of the microstructure of the specimens A, B, and C in the form of 4 x 4 x 4 mm cubes virtually extracted from each of the analysed specimens: (left) A; (centre) B; (right) C).

A different method for presenting the information included in the ROIs described above is to perform the brightness distribution of all voxels constituting the ROI ensembles. It was found that the magnitudes of brightness in different regions were included in the following ranges: the area of voids and fibres: 0-50 [arbitrary units, a.u.]; fillers: 50-140 [a.u.]; dense phases (unhydrated cement and fine aggregate grains: 140-170 [a.u.]. The greyscale brightness distribution of all voxels belonging to the investigated ROIs are shown in Figure 2-46. It is worth noting that the presence of dense phases can be observed as the peaks on the right side of the distributions mentioned above.

104

Chapter Two

Fig. 2-46. G Greyscale brighhtness distribu ution of all vvoxels belongin ng to the investigated R ROIs shown in Fig. 2-44 and 2-45. 2

In some circumstancees, the digitall data represeenting a locall level of radiation absorption of sccanned volume allows one tto automaticaally detect large volum me defects (ccracks). The procedure invvolves compaaring the brightness oof adjacent voxels to find the low-dennsity areas wiith a size exceeding tthe fibre dim mensions. Th he process iis referred to o as the determinatioon of voxel connectivity c an nd is further described in [38, 39]. The proceduure aims to siimulate the diffusion of gaases and liquiids in the interconnectted network im mmersed in bu ulk material. T The authors developed d special softtware to exaamine the vo oxel’s intercoonnection in datasets obtained thrrough micro-C CT scanning. At the beginnning, a certain n number of “walkerss” was distribbuted random mly across thee processed ROI. R The walkers occuupied one voxxel of space an nd could be uunderstood as a marked point in the dataset repressenting the analysed volum me of the specim men. The walkers woould migrate into the adjacent voxels ffollowing thee data on voxel brighhtness (i.e., material m density). The waalkers would jump in random dirrections, provvided that th he adjacent vvoxel belong ged to a permitted loow-absorptionn phase; otheerwise, the juumps were discarded. d After refreshhing the posiition of all waalkers, one eppoch of their action is completed bby the algoritthm. The num mber of epochhs is measureed by the

Fibre Cement Boards Testing

105

dimensionless integer of time t. The primary output of the random walk procedure is the walker’s mean-square displacement as a function of time (xi, yi, zi are the coordinates of a current position and n is a number of operating walkers):

 r(t)2 !

1 n

n

䌥 [(x ( t) i 1

i

(xi (0))2  ( yi ( t) ( yi (0))2  (zi ( t) (zi (0))2 ]

(25)

The key transport property referred to as the diffusion tortuosity IJ of a porous medium is related to the time-derivative of and can be expressed as follows:

W

A as t 䊻 䌲 d  r (t ) 2 ! / d t

(26)

where: A = a constant that depends on the implemented image lattice parameters and by some authors is taken as 1. Fig. 2-47 shows the image of the FCB specimen including a large crack. Graphical visualization of the voxel movements in the Class A specimen and in the cracked specimen are compared in Fig. 2-48.

Fig. 2-47. Front view of the analysed specimen 7 mm in diameter and 7 mm in height with a large crack.

106

Chapter Two

Fig. 2-48. Virtual paths taken by the “walkers” through fibres and voids in (left) specimen A, (right) cracked specimen showed in Fig. 2-47.

The mean-square displacements after 500,000 epochs in specimens A and B, as well as in the cracked specimen, were 279, 4, and 7160, respectively. The determined values of diffusion tortuosity IJ were 52, 350, and 6.5, respectively. It is worth noting that the parameters of the cracked specimen significantly differ from the non-defective specimens. The results of the data processing obtained by specimen analysis using micro-CT methods allowed the authors to conclude that the method presented here can be used as a tool for testing the microstructure of FCBs. The main advantage of the method is its ability to present the microstructure of the entire specimen volume in a relatively fast and nondestructive mode. However, the resolution of detected microstructure details depends on the size of the specimen under analysis.

CHAPTER THREE APPLICATIONS OF FIBRE CEMENT BOARDS KRZYSZTOF SCHABOWICZ AND TOMASZ GORZELAēCZYK FACULTY OF CIVIL ENGINEERING, WROCàAW UNIVERSITY OF TECHNOLOGY

Introduction The ventilated systems market amounts to close to 1 M (million) m2, about 19% of which are fibre cement boards. However, the market share of this type of boards continues to increase at a steady rate. The situation is clearly different on the framed partition systems market. The annual sales of gypsum boards amount to approximately 100 M m2. The share of fibre cement boards in the framed partition systems market amounts to approx. 2.5-3%, i.e., 4 M m2. Since the market share of fibre cement boards in both the exterior cladding panels market and the interior panels market is relatively low, the main competition comes from the gypsum board manufacturers. The manufacturers of drywall partitioning panels must pay more careful attention to the requirements of their customers who, when choosing a building material, consider not only low operating costs, but also the functionality and aesthetics of the interior. The competing manufacturers of fibre cement boards use different names for their products, and, depending on the company, the fibre cement board may be referred to as a cement board, a cellulose-cement board, a cement-fibre board, or a fibre cement board.

Sample applications of fibre cement boards Sample applications of the fibre cement boards in both newly designed and renovated buildings, mostly located in Poland, are shown below.

108

Chapter Three

Fig. 3-1 and 3-2 show w the ventilateed façade, maade out of fibrre cement boards, of thhe Dialogue Centre C “Upheaavals” buildingg. The buildin ng houses an exhibitioon pavilion of the Natio onal Museum m in Szczecin n, where exhibitions and educatioonal events devoted to tthe recent history h of Szczecin annd the West Pomerania P reg gion are held.. The building g is sunk into the grouund of Solidaarity Square where w the Monnument to thee Victims of Decembeer 1970 standss. The Dialog gue Centre “U Upheavals” waas opened to the publicc at the end of o 2015 and th he beginning of 2016. Thee building was awardeed the prestigious Europeaan Prize for U Urban Public Space in 2016. The exterior claddding panels were attachhed using mechanical fixings.

Fig. 3-1. Sidee façade of the Dialogue D Centre “Upheavals” in Szczecin.

Appliications of Fibrre Cement Boarrds

109

Fig. 3-2. Froont façade and the entrance to the Dialoguee Centre “Uph heavals” in Szczecin.

w the façade of the “Bibliiotech” locateed on the Fig. 3-3 and 3-4 show Wrocáaw Uniiversity of Sciience and Tecchnology in Wrocáaw. W campus of W The buildinng houses thee Community y Library of Exact and Technical T Sciences. T The modern “Bibliotech”” building taakes the forrm of a monumentall gate leading from Grun nwaldzki Squaare to the caampus of Wrocáaw U University of Science and d Technologyy. The build ding was officially oppened on 14 November N 2014 as part off the celebration of the 70th anniverrsary of Wroccáaw Universitty of Science aand Technolo ogy. The façaade panels were attached using adhesivve fixings. Th he panels had been subbjected to autooclaving in th he fabrication pprocess.

110

Chapter Three

Fig. 3-3. View w of “Bibliotechh” façade.

Fig. 3-4. View w of “Bibliotechh” façade corneer.

Appliications of Fibrre Cement Boarrds

111

Fig. 3-5 and 3-6 show w the façades of the C-13 building, a fiive-storey building thaat houses the Integrated I Stu udent Centre oof Wrocáaw University U of Science aand Technologgy. The buildiing was comm missioned in 2007. 2 Due to its uniquue appearancee, it is referreed to as the ““cheese block k” or the “holed blocck.” The sourrce of inspirattion for its ddesigners was a punch tape, a old fo form of data sttorage. The façaade panels were w attached with mechannical fixings,, and the irregularly sspaced holes in the fibre cement boardds were madee using a water-jet cuttting technoloogy.

Fig. 3-5. View w of the C-13 building b façadee of Wrocáaw U University of Sccience and Technology aas seen from WyybrzeĪe WyspiaĔskiego Streett.

112

Chapter Three

Fig. 3-6. View w of the C-13 building b façadee of Wrocáaw U University of Sccience and Technology aas seen from Proofessors Avenu ue.

Fibre ceement boards can also bee used as a ccladding in renovated r buildings. T The followingg example is the C-6 buiilding that ho ouses the Chemistry F Faculty of Wrocáaw W University of Scieence and Tecchnology. Fig. 3-7 andd 3-8 show thhe views of the renovated bbuilding’s façaades. The façade pannels made of fibre cem ment boards were attacheed using mechanical fixings.

Fig. 3-7. View w of the corner of the C-6 building adjacent too the C-13 building.

Appliications of Fibrre Cement Boarrds

113

Fig. 3-8. View w of the C-6 buuilding façade.

on of the fibrre cement boards as a Another example of the applicatio e builddings of the Medical façade claddding are thhe recently erected University oof Warsaw. The facilities presented p beloow were consttructed at the Banach Campus at ĩw wirko and Wig gura Street inn Warsaw. Fig gs 3-9 and 3-10 show the façades of a modern building of the Warsaw Medical University Paediatric Hospital. H Its constructionn was comp pleted in September 2015. White fibre cemen nt boards andd black metaal framed windows (thhe colour scheeme consisten nt with the adj djacent buildin ngs) were used for thee façades. In order to “waarm up” and soften the raather cold technologicaal appearancee of the cam mpus, light gre reen was used d for the window linttels and the reectangular form ms were brokken by accentu uating the building’s entrances with w overhang gs. As a rresult, an in nteresting architecturall effect was produced p and the hospital’ s image was “warmed up.” The faççade panels were attached using u mechaniical fixings.

114

Chapter Three

Fig. 3-9. Façaade of the Warssaw Medical Un niversity Paediaatric Hospital.

Fig. 3-10. Bacck wing of the Warsaw Mediccal University P Paediatric Hospital.

Figure 3-11 shows the Warsaw w Medical U University Preclinical P Research Ceentre on the Ochota O Campu us. The buildinng was openeed in May 2013.

Appliications of Fibrre Cement Boarrds

115

Fig. 3-11. Faççade of the Warrsaw Medical University U Precllinical Research h Centre.

Fig. 3-122 shows the façade of a building b consstructed as paart of the extension off the Electronnics and Inforrmation Technnologies Facu ulty at the Warsaw Uniiversity of Technology. Thee building waas opened in September 2015. The fa façade panels made m of fibree cement boardds were attach hed using mechanical fixings.

Fig. 3-12. F Façade of the new building g constructed ffor the Electrronics and Information T Technologies Faaculty, Warsaw w University of Technology.

116

Chapter Three

Another example of using u fibre cem ment boards aas a façade clladding is shown in Fiigs 3-13, 3-14, and 3-15. Th he figures shoow views of th he façade of the Sciennce and Art Centre, C which is part of the StrzemiĔski Academy A of Arts in àóódĨ. Its constrruction was co ompleted in A April 2013.

Fig. 3-13. Façade of the Sciience and Art Centre, C StrzemiiĔski Academy of Arts in àódĨ.

Fig. 3-14. Façade of the Sciience and Art Centre, C StrzemiiĔski Academy of Arts in àódĨ.

Appliications of Fibrre Cement Boarrds

117

Fig. 3-15. Faççade of the Science and Art Centre C (view froom car park), SttrzemiĔski Academy of A Arts in àódĨ.

Ventilateed façades made m of fibre cement boarrds are also known k in many Europpean and othher countries, including Sccandinavian countries, c Russia, andd North Ameerica, but theey significanttly differ in the raw materials annd maturing teechniques used d in the fabric ation process.. Fig. 3-16 and 3-17 show s the view ws of the froont façade off a public building on Chistoprudnyyy Boulevard in Moscow. T The façade pan nels were attached usinng mechanicaal fixings.

Fig. 3-16. Fronnt façade of a puublic building on n Chistoprudnyyy Boulevard in Moscow. M

118

Chapter Three

Fig. 3-17. Side view of thee front façade of a public buuilding on Chisstoprudnyy Boulevard in Moscow.

Fig. 3-188 and 3-19 shoow the façadees of a residenntial complex located l at Zoologichesskiy Perelok in i Moscow. In n both cases tthe façade pan nels were attached usinng mechanicaal fixings.

Applications of Fibre Cement Boards

119

Fig. 3-18. Side view of the façade of a residential building, Zoologicheskiy Perelok in Moscow.

Fig. 3-19. Side view of the façade of a residential building, Zoologicheskiy Perelok in Moscow.

Chapter Three

120

To conclude, it is worth noting the advantages and disadvantages of fibre cement boards and the associated benefits and risks. The relevant analysis based on market research is shown in Table 3-1. Table 3-1. Advantages and disadvantages of fibre cement boards and associated benefits and risks.

Advantages x Unique and well automated technology x Good product with wide range of applications x Good strength parameters x Good fire-resistance parameters for the products conforming to the requirements of the relevant category

Benefits x Focus on one product will eliminate market cannibalism x Highlighting wide range of board applications x Positioning board as modern and eco-friendly technology, demanding but worth using

Disadvantages x New product, and so low product awareness x Risks related to implementation x Weak sales structures

Risks x No awareness of product and its applications x Association with asbestos board x Large number of different applications and characteristics of fibre cement boards makes communication difficult x Competition from gypsum boards and existing board producers x Difficult to change habits of contractors who associate fibre cement boards with greater labour intensity

Applications of Fibre Cement Boards

121

Conclusions According to the development strategy, in the next several years the focus will be put on developing an effective distribution system and efforts will be aimed at promoting new brands of fibre cement boards. Along with those measures, the manufacturers intend to focus on product development by introducing new families of decorative products including the following: – woodgrain finish boards, – stone finish boards, – boards with innovative irregular shapes or relief patterns. Decorative boards can be used for interior and exterior applications as a finishing and decorative element of the room. The product development will also cover roof applications, including roofing tiles, chimney cladding, etc., as well as the following: – – – –

soundproof partition wall systems, fire division systems with EI 30 and EI 120 min fire rating, ventilated insulation systems for single-family houses, interior design board fitting systems.

The introduction of these systems will enable sales of complementary products and increase the turnover and mark-ups for other necessary components, including the following: – – – –

fixings, adhesives and sealants, ventilated systems supports, insulation, etc.

C1 Micrograph of PVA fibres enforcing the microstructure of a fibre cement board, (Fig. 4, upper part).

C2 Micrograph of small strand of cellulose fibres in a fibre cement board, (Fig. 4, lower part).

C3 Comparison between the conventional optical microscope images and the EFI technique. (a) the surface of the matrix is visible but the protruding fibres are blurred; (b) only the fibres are in focus; (c) EFI image of the sample in polarized light; (d) EFI image of the sample in dark field, (Fig. 2-31).

C4 Comparison between two EFI imaging techniques. (a) bright field; (b) polarization, (Fig. 2-32).

C5 Two images at different magnification of a coloured fibre cement board, (Fig. 233).

C6 Distinguishing between two different kinds of fibres using the polarization illumination technique, (Fig. 2-34).

C7 Comparison between the images of a PVA fibre: (a) in polarized light; (b) in dark field, (Fig. 2-35).

C8 3D images of PVA fibre: (a) the entire fibre; (b) magnified fibre surface, (Fig. 2-36).

C9 Stitching image section of 5 mm long sample. (a) planar; (b) 3D stitching, (Fig. 2-37).

C10 Façade of Science and Art Centre, StrzemiĔski Academy of Arts in àódĨ (Poland), (Fig. 3-14).

REFERENCES

[1] Source: www.euronit.de/. [2] Source: www.cembrit.com/. [3] EN 12467 – “Fibre-cement flat sheets. Product specification and test methods” (2013). [4] Source: de.wikipedia.org/wiki/Ludwig_Hatschek. See also Rudolf Lehr, Landeschronik Oberösterreich, Wien and München: Verlag Christian Brandstätter, 2008. [5] W. E. Cooke, “Fibrosis of the Lungs due to the Inhalation of Asbestos Dust,” British Medical Journal 2 (1924), 147. [6] R. S. P. Coutts, “A Review of Australian Research into Natural Fibre Cement Composites,” Cement & Concrete Composites 27 (2005), 518-26. [7] M. Ardanuy et al., “Cellulosic Fiber Reinforced Cement-Based Composites: A Review of Recent Research,” Construction and Building Materials 79 (2015), 115-28. [8] J. Färber et al., “Cellulose Microfibril Angles in a Spruce Branch and Mechanical Implications,” J. Mater. Sci. 36 (2001), 5087-92. [9] R. S. P. Coutts and P. Kightly, “Bonding in Wood Fibre-Cement Composites,” J. Mater. Sci. 19 (1984), 3355-59. [10] S. Sasmal and G. Avinash, “Investigations on Mechanical Performance of Cementitious Composites Micro-Engineered with Poly Vinyl Alcohol Fibers,” Construction and Building Materials 128 (2016), 136-47. [11] A. J. Michell and G. Freischmidt, “Effect of Fibre Curl on the Properties of Wood Pulp Fibre Cement and Silica Sheets,” J. Mater. Sci. 25 (1990), 5225-30. [12] T. GorzelaĔczyk and K. Schabowicz, “Nondestructive Testing of Moisture in Cellulose Fibre Cement Boards,” Proc. of 3rd Int. Conference “Durability and Effectiveness of Repairing of Buildings,” PoznaĔ, 2014. [13] T. Cooke, “Formation of Films on Hatschek Machines.” Bonded Wood and Fibre Composites Conference, Idaho, 2002. Source: www.fibrecementconsulting.com/publications/publications.htm.

124

References

[14] V. C. Li, H. Stang, and H. Krenchel, “Micromechanics of Crack Bridging in Fibre-Reinforced Concrete,” Materials and Structures 26 (1993), 486-94. [15] A. M. Cooke, “Durability of Autoclaved Cellulose Fiber Cement Composites.” Conference on Inorganic Bonded Wood and Fbre Composites, Idaho, 2000, ibidem. [16] J. M. Coulson and J. F. R. Richardson, Chemical Engineering, 2nd ed., Oxford and New York: Pergamon Press, 1968, Vol. 2, Chapter 2. [17] European Standard EN 12467, “Fibre-cement flat sheets – Product specification and test methods” (2012), European Committee for Standardization, Brussels Av. Marnix 17. [18] RILEM Recommendations for the Testing and Use of Constructions Materials, ed. by RILEM, 1994. Source: www.rilem.net. [19] Z. Ranachowski et al., “Application of Acoustic Emission Method to Determine Critical Stress in Fibre Reinforced Mortar Beams,” Archives of Acoustics 37 (2012), 1-8. [20] Source: www.iso.org/standard/64988.html. [21] J. Hola and K. Schabowicz, “State-of-the-Art Non-Destructive Methods for Diagnostic Testing of Building Structures – Anticipated Development Trends,” Archives of Civil and Mechanical Engineering 10 (2010), 5-18. [22] V. M. Malhort and J. N. Carin, Handbook on Non-Destructive Testing of Concrete, CRC Press, 2004. [23] Source: www.apawood-europe.org/official-guidelines/europeanstandards/en-13501-1/. [24] B. Piwakowski et al., “Detection of the Damage in Concrete Cover Using Ultrasound,” In Nondestructive Testing of Materials and Structures, pp. 173-84, Warsaw, Poland: Institute of Fundamental Technological Research, 2002. [25] D. M. McCann and M. C. Forde, “Review of NDT Methods in the Assessment of Concrete and Masonry Structures,” NDT & E International 34 (2001), 71-84. [26] H. Lamb, The Dynamical Theory of Sound, London: Arnold, 1931. [27] M. Kaczmarek et al., “Noncontact Ultrasonic Nondestructive Techniques: State of the Art and Their Use in Civil Engineering,” Journal of Infrastructure Systems 23 (2017), 45-56. [28] R. Drelich et al., “Automated Control of Cellulose Fibre Cement Boards with a Non-Contact Ultrasound Scanner,” Automation in Construction 57 (2015), 55-63.

The Fabrication, Testing and Application of Fibre Cement Boards

125

[29] S. Popovics, “Effects of Uneven Moisture Distribution on the Strength of and Wave Velocity in Concrete,” Ultrasonics 43 (2005), 429-34. [30] V. Garnier et al., “Acoustic Techniques for Concrete Evaluation: Improvements, Comparisons and Consistency,” Construction and Building Materials 43 (2013), 598-613. [31] P. Goodhew et al., Electron Microscopy and Analysis, 3 ed., London: Taylor and Francis, 2001. [32] E. M. Bezerra et al., “Some Properties of Fiber-Cement Composites with Selected Fibers,” Proc. of NOCMAT 2004. Source: www.usp.br/constrambi/producao_arquivos/some _properties_pdf. [33] Detail and Depth with Panoramic and Extended Focal Imaging, Olympus Europa SE & Co. KG. Source: https://unicam.hu/files/ olympus-cellsens/cellSens_Attekintes.pdf. [34] S. Lu et al., “X-Ray Microtomographic Studies of Pore Structure and Permeability in Portland Cement Concrete,” Mater. Struct. 39 (2006), 611-20. [35] E. J. Garboczi, “Three-Dimensional Mathematical Analysis of Particle Shape Using X-Ray Tomography and Spherical Harmonics: Application to Aggregates Used in Concrete,” Cem. Concr. Res. 32 (2002), 1621-38. [36] M. Lanzón et al., “X-Ray Microtomography (ȝ-CT) to Evaluate Microstructure of Mortars Containing Low Density Additions,” Cem. Concr. Compos. 34 (2012), 993-1000. [37] K. Schabowicz et al., “Application of X-Ray Microtomography to Quality Assessment of Fibre Cement Boards,” Construction and Building Materials 110 (2016), 182-88. [38] Y. Nakashima and S. Kamiya, “Mathematica Programs for the Analysis of Three-Dimensional Pore Connectivity and Anisotropic Tortuosity of Porous Rocks Using X-ray Computed Tomography Image Data,” J. Nucl. Sci. Technol. 44 (2012), 1233-47. [39] Z. Ranachowski et al., “Application of X-Ray Microtomography and Optical Microscopy to Determine the Microstructure of Concrete Penetrated by Carbon Dioxide,” Archives of Metallurgy and Materials 59 (2014), 1451-57. [40] For more information, visit the following website: www.kuraray.us.com/products/fibers/kuralon-and-kuralon-k-ii/.