Mine Waste Management in China: Recent Development [1st ed. 2020] 978-981-32-9215-4, 978-981-32-9216-1

Table of contents :
Front Matter ....Pages i-ix
Traditional Treatment of Mine Waste (Di Wu)....Pages 1-9
Solutions for Surface Disposal of Mine Tailings (Di Wu)....Pages 11-19
Case Study of Surface Consolidated Tailings Stockpile (Di Wu)....Pages 21-35
Case Study of Surface Cemented Tailings Discharge (Di Wu)....Pages 37-47
Solutions for Underground Placement of Mine Tailings (Di Wu)....Pages 49-55
Properties of Cemented Tailings Backfill (Di Wu)....Pages 57-114
Case Study of Cemented Tailings Backfill (Di Wu)....Pages 115-122
Solutions for Underground Placement of Coal Mine Waste (Di Wu)....Pages 123-126
Properties of Cemented Coal Gangue-Fly Ash Backfill (Di Wu)....Pages 127-194
Case Study of Cemented Coal Gangue-Fly Ash Backfill (Di Wu)....Pages 195-204

Citation preview

Di Wu

Mine Waste Management in China: Recent Development

Mine Waste Management in China: Recent Development

Di Wu

Mine Waste Management in China: Recent Development

123

Di Wu School of Energy and Mining Engineering China University of Mining and Technology-Beijing Beijing, China

ISBN 978-981-32-9215-4 ISBN 978-981-32-9216-1 https://doi.org/10.1007/978-981-32-9216-1

(eBook)

© Springer Nature Singapore Pte Ltd. 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Acknowledgements

The author would like to acknowledge the support from: • China Scholarship Council (CSC); • Yue Qi Young Scholar Project, China University of Mining and Technology, Beijing; • National Key Technology R&D Program; • National Natural Science Foundation of China (Grants: No. 51674263; 51404271); • State Key Laboratory for Coal Resources & Safe Mining, China University of Mining & Technology (No. SKLCRSM16KFC04); • Beijing Key Laboratory for Precise Mining of Intergrown Energy and Resources; • University of Ottawa, The University of Western Australia, Curtin University; The University of New South Wales, China University of Mining & Technology, University of Science and Technology Beijing; • China Minmetals Corporation, Magang (Group) Holding Co., Ltd., China National Gold Group Corporation, Ansteel Group Corporation Limited, Shandong Gold Group, Shanxi Coking Coal Group Co., Ltd., Sinosteel Maanshan Institute of Mining Research Co., Ltd., China Academy of Building Research; • COMSOL Co., Ltd., Gold Fields Limited; Maptek Group; Magnetite Mines Limited. The author would also like to express sincere gratitude to his: • Supervisors (Sijing Cai, Mamadou Fall, Andy Fourie); • Colleagues (Baogui Yang, Yunbing Hou, Jiachen Wang, Jishan Liu, Xuejie Deng); • Students (Xiaolong Zhao, Donglin Fan, Tengfei Deng, Runkang Zhao, Yongan Zhang, Chaowu Xie, Wudi Zhang, Shuai Liu, Wentao Hou).

v

Contents

1

Traditional Treatment of Mine Waste . . . . . . . . . . . . . . . . . . 1.1 Traditional Disposal of Tailings . . . . . . . . . . . . . . . . . . . . 1.1.1 Tailings Dam Failures . . . . . . . . . . . . . . . . . . . . 1.1.2 Environmental Impacts of Tailings Dam Failures . 1.2 Traditional Disposal of Coal Mine Waste . . . . . . . . . . . . . 1.2.1 Hazardous Impacts on the Environment . . . . . . . . 1.2.2 Limited Utilization of Coal Mine Waste . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

1 1 3 3 5 5 7 8

2

Solutions for Surface Disposal of Mine 2.1 Dry Stacking of Tailings . . . . . . . 2.2 Paste Discharging of Tailings . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

11 11 14 18

3

Case Study of Surface Consolidated Tailings Stockpile . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Experiment Procedure . . . . . . . . . . . . . . . . 3.2.3 UCS and Wetting Tests . . . . . . . . . . . . . . . 3.2.4 Numerical Simulation . . . . . . . . . . . . . . . . . 3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . 3.3.1 Selection of Binder Type . . . . . . . . . . . . . . 3.3.2 Determination of Binder Content . . . . . . . . . 3.3.3 Effect of Binder on Dewaterability of Filter . 3.3.4 Industrial Application . . . . . . . . . . . . . . . . . 3.4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

21 21 23 23 24 25 25 25 25 25 27 31 34 35

Tailings . . . .......... .......... ..........

. . . .

. . . .

. . . .

vii

viii

Contents

....... .......

37 38

....... ....... .......

40 40 42

5

Solutions for Underground Placement of Mine Tailings . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49 54

6

Properties of Cemented Tailings Backfill . . . . . . . . . . . . . . 6.1 Rheological Properties of Fresh CPB . . . . . . . . . . . . . . 6.1.1 Mathematical Modeling . . . . . . . . . . . . . . . . . 6.1.2 Model Validation and Simulation . . . . . . . . . . 6.2 Flowability of Fresh CPB . . . . . . . . . . . . . . . . . . . . . . 6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB 6.3.1 Model Development . . . . . . . . . . . . . . . . . . . . 6.3.2 Model Validation . . . . . . . . . . . . . . . . . . . . . . 6.3.3 Model Application . . . . . . . . . . . . . . . . . . . . . 6.4 Thermo-Hydro-Mechanical Behavior of Hydrating CPB References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

Case Study of Cemented Tailings Backfill . . . . . . . . . . . . . . . . . . . . 115 7.1 Practice of CTB Technology in Baixiangshan Iron Mine . . . . . . 115 7.2 Optimization of CTB Technology in Xincheng Gold Mine . . . . 117

8

Solutions for Underground 8.1 Solid Backfill . . . . . . 8.2 Fluid Backfill . . . . . . References . . . . . . . . . . . . .

9

Properties of Cemented Coal Gangue-Fly Ash Backfill . . . . . 9.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Rheology and Flowability of Fresh CGFB . . . . . . . . . . . 9.2.1 Rheology of Fresh CGFB . . . . . . . . . . . . . . . . . 9.2.2 Flowability of Fresh CGFB . . . . . . . . . . . . . . . 9.3 Mechanical Performance of Hardened CGFB . . . . . . . . . 9.3.1 UCS and UPV of CGFB . . . . . . . . . . . . . . . . . 9.3.2 Mechanical Performance Based on AE . . . . . . . 9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB . . 9.4.1 Development of the THC Model . . . . . . . . . . . . 9.4.2 Validation of the Developed THC Model . . . . . 9.4.3 Application of the Developed Model . . . . . . . . . 9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB 9.5.1 Experimental Programs . . . . . . . . . . . . . . . . . . . 9.5.2 Results and Discussion . . . . . . . . . . . . . . . . . . .

4

Case Study of Surface Cemented Tailings Discharge . . . . 4.1 Selection of Tailings Disposal Method . . . . . . . . . . . . 4.2 Determination of Strength and Thickness of the Base Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Numerical Modeling . . . . . . . . . . . . . . . . . . . 4.2.2 Simulation Results and Discussions . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

57 57 59 62 69 71 73 81 89 107 112

Placement of Coal Mine Waste . . . . . . 123 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

127 128 128 128 132 138 139 145 150 151 153 156 161 162 166

Contents

ix

9.6

Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6.1 Mathematical Formulation of the THMC Model . 9.6.2 Verification of the Model and Simulation Results References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill 10.1 Materials and Tests . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Preparation of CGFB Materials . . . . . . . . . . . . . . . 10.2.1 Production of Coal Gangue . . . . . . . . . . . . 10.2.2 Provision of Fly Ash and Binder . . . . . . . . 10.2.3 Provision of Water and Additive . . . . . . . . 10.2.4 System for CGFB Preparation . . . . . . . . . . 10.2.5 Electrical Manipulative System . . . . . . . . . 10.2.6 Additional Appliance for Clean Production 10.3 Transportation of CGFB Materials . . . . . . . . . . . . . 10.4 Placement of CGFB Materials . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

173 173 181 192

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

195 196 196 196 199 199 200 201 201 202 203

Chapter 1

Traditional Treatment of Mine Waste

Abstract The traditional treatments of tailings, coal gangue, and fly ash are introduced, respectively. The disadvantages of these disposing methods are also presented, such as environmental contamination or even personnel casualties. Therefore, the conventional disposal of mine waste needs to be improved and new solutions desperately need to be developed. Keywords Tailings · Coal gangue · Fly ash · Impoundment · Dump · Pollution Mining industry certainly contributes to the prosperous development of modern society by providing valuable mineral resources that are widely used in different professions, such as manufacture of airplane, automobile and ship, electronic information, chemical engineering, construction, and so on. However, the mineral extraction also produces a huge amount of solid waste, such as tailings (from hard rock mines) and coal gangue (from coal mines), as well as fly ash (from coal combustion).

1.1 Traditional Disposal of Tailings The traditional treatment of tailings is to store them in impoundments behind dams. Figure 1.1 shows a tailings dam located in Hebei Province of China. Generally, tailings dams are built with readily available local materials, rather than the concrete used, for example, in water-retention dams (Kossoff et al. 2014). The tailings dam is commonly constructed by soil, waste rock, and the tailings themselves (Álvarez-Valero et al. 2009; Bruno 2007; Chakraborty and Choudhury 2009; DixonHardy and Engels 2007a; Younger and Wolkersdorfer 2004). However, the tailings dam is not a finalized full capacity structure; its height is raised with the increase of tailings storage quantity (Lottermoser 2007). The initial volumes of the tailings ponds can be expanded by elevating the embankments in three commonly used ways: upstream, vertically (center-line), and downstream (Martin and McRoberts 1999). Figure 1.2 schematically illustrates the three methods for raising the tailings dams. As explained by Fig. 1.2, upstream raising technique is accomplished by placing new construction materials within the existing impoundment, and center-line raising © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_1

1

2

1 Traditional Treatment of Mine Waste

Fig. 1.1 A tailings dam located in Hebei Province of China

Fig. 1.2 Schematic illustrations of upstream, downstream and center-line sequentially raised tailings dams (Kossoff et al. 2014)

method is achieved by piling up new construction materials directly on top of the existing embankment, while downstream raising is to place the new materials outside the impoundment (Kossoff et al. 2014). From Fig. 1.2, it can be obviously detected that the construction method of upstream raising is the most cost-effective since the fewest building materials are required for elevating the embankment (Soares et al. 2000). Hence, the upstream method has become the main method for mine tailings disposal in China (Wei et al. 2008). However, the tailings dams constructed by the upstream method are observably less stable than that built by the other two ones (Kwak et al. 2005). Besides, the tailings impoundments constructed by the upstream method can obtain the minimal volume expansion.

1.1 Traditional Disposal of Tailings

3

Fig. 1.3 Some of the destructions resulted from tailings dam failure on August 10, 2008

1.1.1 Tailings Dam Failures The failure of tailings impoundments is one of the main concerns of the mining industry (Dixon-Hardy and Engels 2007b). This is because the tailings dam failures can cause severe environmental pollution, economic loss or even human casualty. Rico et al. (2008) have stated that the tailings dams constructed by the upstream raised method is the most likely to fail, in comparison with that built by the other two methods. In China, failures of tailings dams have resulted in serious consequences. For example, on August 10, 2008, a failure occurred in iron tailings impoundment located in Shanxi province, northwest China, killing hundreds of people (Yin et al. 2011). Figure 1.3 illustrates some of the devastating spectacles induced by this accident. Additionally, Yin et al. (2011) have also listed several main failures of tailings dams and ensuing consequences of human death in Table 1.1.

1.1.2 Environmental Impacts of Tailings Dam Failures In addition to causing immediate casualties, the failures of tailings dams may result in serious environment pollution over a relatively long-term period that may also create deaths due to toxicity. The leakage of contaminants such as As and Pb into

4

1 Traditional Treatment of Mine Waste

Table 1.1 Main tailings dam failures reported in China and the consequences Name of the failed tailings dam

Building method

Consequences

Huogudu, Yunnan Tin Group Co.

Upstream

171 Killed

Niujiaolong, Shizhuyuan Nonferrous Metals Co.

Upstream

49 Killed

Longjiaoshan, Daye Iron Ore mine

Upstream

31 Killed

Dachang, Nandan Tin mine

Upstream

28 Killed

the expansive environment, almost certainly results in increased rates of pathology and, by extension, mortality (Kossoff et al. 2014). On April 25, 1998, the failure of Aznalcóllar dam at the Boliden Los Frailes Ag-Cu-Pb-Zn mine facility caused a wide spread of pollutants (e.g., As) in the largely agricultural Guadiamar basin and thus threatened local ecological environment (López-Pamo et al. 1999; Grimalt et al. 1999; Hudson-Edwards et al. 2003). For instance, as a direct result of the spill, all of the fish and shellfish present in the polluted watercourses were killed (Grimalt et al. 1999). On January 30, 2000, a tailings dam breakage occurred at Baia Mare, Romania (Macklin et al. 2003). At least 120 tonnes of cyanide and metallic elements entered the catchment with the released tailings, causing serious contamination and fish deaths (Lásló 2006). Cyanide is a kind of extremely toxic substance and can result in biological deaths, thus the cyanide waste is recommended to be treated and removed before being deposited in impoundments (Hoskin 2003). Except for the direct toxic hard to water and thus to the aquatic animals and plants, the aftermath of a tailings dam flood also exerts negative influence on soil and thus on vegetation and arable crops (Kossoff et al. 2014). The possible medium- to longer-term negative effects of a dam spill on the floodplain environment are illustrated by data (in Table 1.2) from the Pb–Zn mine in Chenzhou, China (Kossoff et al. 2014; Liu et al. 2005). On August 25, 1985, a tailings dam failure occured following heavy rains, inundating the Dong River valley. Strips of farmland 400 m wide along both river banks were covered with a 15 cm-thick layer of black sludge (Liu et al. 2005). Only some selected areas were remediated, but some other agricultural land was left without remediating (Liu et al. 2005). In August 2002, the measuring results from the unremediated soil indicated that eleTable 1.2 Floodplain soils and sediment concentrations following tailings dam spills reported in China Cd (mg/kg)

Cu (mg/kg)

Background soil, hill 10 km from mine

2.08

Remediated soil, 9.5 km from mine

2.7

72.18

Unremediated soil, 8 km from mine

7.57

135.83

All samples are taken in August 2002

25.95

Pb (mg/kg) 60.49 321.11 1088.3

Zn (mg/kg) 140.48 416.61 1000.71

1.1 Traditional Disposal of Tailings

5

Fig. 1.4 Crops polluted by released tailings spill

ment concentrations exceeded the Chinese soil maximum allowable concentration (MAC) standard (Liu et al. 2005). Figure 1.4 shows the arable crops that are immersed in the water contaminated by released tailings spill in west of Hunan Province, China. In summary, the seriously unacceptable consequences induced by tailings dam failures have been discussed above. Besides, regardless of the likelihood of failure, the existence of tailings dams occupies vast expanses of land, especially valuable farmland, let alone the huge capital costs of constructing and maintaining tailings dams. For instance, the annual operational cost for all the tailings dams in China is estimated to be higher than 750 million.

1.2 Traditional Disposal of Coal Mine Waste 1.2.1 Hazardous Impacts on the Environment In 2008, China had produced 2716 Mtons raw coal, accounting for approximately 40% of the total production and became the biggest producer of coal in the world (Liu and Liu 2010). Therefore, in China coal gangue generated from coal resources exploitation has contributed to the largest increase in the industrial solid waste produced annually. There are almost 4.5 billion tons of coal gangue stockpiled into more than 1700 waste dumps which occupied 150 km2 of land (Bian et al. 2009; Zhao et al. 2008). Besides, the annual generation of coal gangue is estimated to be more

6

1 Traditional Treatment of Mine Waste

than 315 million tons from underground coal mining operations (Liu and Liu 2010). The traditional management of coal gangues by dumping them in cone-shaped heaps may exert adverse influences in terms of environmental, social, and economic issues (Bell et al. 2000; Franks et al. 2011; Glauser et al. 2005; Szczepanska 1999). If coal gangue is exposed to the air and piled on land, landslides or even explosions sometimes may occur in the dumps, which may cause direct personnel casualties (Fan et al. 2014). In 2005, an explosion accident occurred in a coal gangue dump in China, killing 8 people and hurting 122 persons, as shown in Fig. 1.5 (Fan et al. 2014; Wang et al. 2008). In addition to immediate injuries, the spontaneous combustions of coal gangue dumps often occur, inducing indirect harm such as smoke poisoning. Detailed hazards result from environmentally unfriendly stockpiled coal gangue dumps are summarized as follows (Liu and Liu 2010): • Natural calamity, such as landslide and debris flow due to improperly piled gangue. • Poison releasing, natural weathering and rainwater drenching causing the poisonous into the soil and underground water. • Spontaneous combustion, the poisonous gas emitting into the atmosphere. • Acid rain formation near the gangue mountain. • Noxious substance polluting the groundwater. In comparison with the coal gangue that is generated directly from coal mining, fly ash, which is also a kind of solid wastes, is produced from the coal-related industry. Coal is combusted in power plants to generate electricity, and fly ash is exactly the

Fig. 1.5 An explosion accident of a coal gangue dump

1.2 Traditional Disposal of Coal Mine Waste

7

by-product from coal burning. The discharge of fly ash on land is also a hazard to the environment. The annual production of fly ash in China is more than 150 million tons (Liu and Liu 2010). Since the fly ash is very fine and light, it is very easy to fly with wind, and if not properly disposed, it can possibly contaminate the environment of the mining area by polluting the water, atmosphere, soil and occupying land (Liu and Liu 2010). The disadvantages of the traditional disposal of fly ash are summarized as follows (Liu and Liu 2010): • Potential toxicant may pollute the groundwater and soil. • Occasionally pile up occupies land and contaminates soil. • Atmosphere contamination is prone to occur in large area due to long distance spread of the fly ash with wind. • The air-suspended fly ash particles badly threaten the respiratory systems of people. The following example can further indicate the terrible consequences of inappropriate disposal of fly ash. On December 28, 2008, a massive release of fly ash occurred as a result of the collapse of a containing dam at Tennessee Valley Authority’s Kingston dam (Kossoff et al. 2014). The coal ash spilled into the River Emory, part of Tennessee’s catchment. The fly ash contained high levels of As (75 mg/kg) and Hg (0.15 mg/kg), as well as significant levels of radioactivity (Ruhl et al. 2009). In addition, the fine particle size of the fly ash made it easy to disperse in the air after its release, particularly from the ground surface of the floodplain (Kossoff et al. 2014).

1.2.2 Limited Utilization of Coal Mine Waste Chugh and Patwardhan (2004) have indicated that some coal gangue with high content of carbon can be generally mixed with other coal for mine-mouth power generation. Commonly, the coal gangue is mainly reused either as input ingredient to improve the performance of traditional constructing material (e.g. cement), or to prepare new building material, such as calcined products with high coal gangue, load-bearing and non-load-bearing hollow bricks, or as lightweight aggregate for the replacement of clay (Liu and Liu 2010). In addition, coal gangue has been accepted in many places as alternative aggregates in embankment, road, pavement, foundation and building construction, pyrites extraction, zeolites production, and so on (Liu and Liu 2010; Sun and Li 2008). Although some fly ash is recycled and thus used in making civil construction materials, there is still a limit to the application demand of fly ash in construction industry (Liu and Liu 2010). Kikuchi (1999) summarized three ways to treat fly ash in Finland, which are given as follows: i.

Alkali treatment can transform fly ash to zeolite;

8

1 Traditional Treatment of Mine Waste

ii. Potassium silicate fertilizer is produced from coal ash and has a higher receptivity in the soil than that of conventional fertilizers; iii. Emission of sulfur dioxide is controlled by flue gas desulfurization using fly ash. Although engineers and researchers have proposed numerous ways to reuse coal mine wastes, for reducing their surface discharge and stockpile. However, the achievements and benefits have still been limited. The current utilizations of coal mine wastes (such as recycling them as construction materials) are inadequate to treat such huge amounts of existing and ever-increasing coal mine wastes. Therefore, there is an urgent need to introduce and develop an effective solution to resuse, and thus significantly reduce the aboveground discharged quantities of coal mine wastes.

References Álvarez-Valero, A., Sáez, R., Pérez-López, R., Delgado, J., Nieto, J.M.: Evaluation of heavy metal bio-availability from Almagrera pyrite-rich tailings dam (Iberian Pyrite Belt, SW Spain) based on a sequential extraction procedure. J. Geochem. Explor. 102, 87–94 (2009) Bell, F.G., Stacey, T.R., Genske, D.D.: Mining subsidence and its effect on the environment: some differing examples. Environ. Geol. 40, 135–152 (2000) Bian, Z., Dong, J., Lei, S., Leng, H., Mu, S., Wang, H.: The impact of disposal and treatment of coal mining wastes on environment and farmland. Environ. Geol. 58, 625–634 (2009) Bruno, B.: Hydrogeotechnical properties of hard rock tailings from metal mines and emerging geo-environmental disposal approaches. Can. Geotech. J. 44(9), 1019–1052 (2007) Chakraborty, D., Choudhury, D.: Investigation of the behavior of tailings earthen dam under seismic conditions. Am. J. Eng. Appl. Sci. 2, 559–564 (2009) Chugh, Y.P., Patwardhan, A.: Mine-mouth power and process steam generation using fine coal waste fuel. Resour. Conserv. Recycl. 40(3), 225–243 (2004) Dixon-Hardy, D.W., Engels, J.E.: Methods for the disposal and storage of mine tailings. Land Contam. Reclam. 15, 301–317 (2007a) Dixon-Hardy, D.W., Engels, J.E.: Guidelines and recommendations for the safe operation of tailings management facilities. Environ. Eng. Sci. 24(5), 625–637 (2007b) Fan, G., Zhang, D., Wang, X.: Reduction and utilization of coal mine waste rock in China: a case study in Tiefa coalfield. Resour. Conserv. Recycl. 83, 24–33 (2014) Franks, D.M., Boger, D.V., Côte, C.M., Mulligan, D.R.: Sustainable development principles for the disposal of mining and mineral processing wastes. Resour. Policy 36, 114–122 (2011) Glauser, S., McAllister, M.L., Milioli, G.: The challenges of sustainability in mining regions: the coal mining region of Santa Catarina. Brazil. Nat. Resour. Forum 29, 1–11 (2005) Grimalt, J.O., Ferrer, M., Macpherson, E.: The mine tailing accident in Aznalcollar. Sci. Total Environ. 242, 3–11 (1999) Hoskin, W.A.N.D.A.: Abandoned mine sites: problems, issues and options. In: Puura, E., Marmo, L., D’Alessandro, M. (eds.) Workshop on Mine and Quarry Waste-The Burden from the Past. Workshop Organised in the Frame of the JRC Enlargement Project: Inventory, Regulations and Environmental Impact of Toxic Mining Wastes in Pre-accession Countries “PECOMINES”, 27th–28th May 2002, Ortno, Italy, pp. 53–55 (2003) Hudson-Edwards, K.A., Macklin, M.G., Jamieson, H.E., Brewer, P., Coulthard, T.J., Howard, A.J., Turner, J.: The impact of tailings dam spills and clean-up operations on sediment and water

References

9

quality in river systems: the Ríos AgrioGuadiamar, Aznalcóllar, Spain. Appl. Geochem. 18, 221–239 (2003) Kikuchi, R.: Application of coal ash to environmental improvement: transformation into zeolite, potassium fertilizer, and FGD absorbent. Resour. Conserv. Recycl. 27(4), 333–346 (1999) Kossoff, D., Dubbin, W.E., Alfredsson, M., Edwards, S.J., Macklin, M.G., Hudson-Edwards, K.A.: Mine tailings dams: Characteristics, failure, environmental impacts, and remediation. Appl. Geochem. 51, 229–245 (2014) Kwak, M., James, D.F., Klein, K.A.: Flow behaviour of tailings paste for surface disposal. Int. J. Miner. Process. 77, 139–153 (2005) Lásló, F.: Lessons learned from the cyanide and heavy metal accidental water pollution in the Tisa River basin in the year 2000. In: Dura, G., Simeonova, F. (eds.) Management of Intentional and Accidental Water Pollution, pp. 43–50 (2006) Liu, H., Liu, Z.: Recycling utilization patterns of coal mining waste in China. Resour. Conserv. Recycl. 54, 1331–1340 (2010) Liu, H., Probst, A., Liao, B.: Metal contamination of soils and crops affected by the Chenzhou lead/zinc mine spill (Hunan, China). Sci. Total Environ. 339, 153–166 (2005) López-Pamo, E., Barettino, D., Antón-Pacheco, C., Ortiz, G., Arránz, J.C., Gumiel, J.C., MartínezPledel, B., Aparicio, M., Montouto, O.: The extent of the Aznalcóllar pyritic sludge spill and its effects on soils. Sci. Total Environ. 242, 57–88 (1999) Lottermoser, B.: Mine Wastes: Characterization, Treatment and Environmental Impacts. Springer, Berlin, Heidelberg, New York (2007) Macklin, M.G., Brewer, P.A., Balteanu, D., Coulthard, T.J., Driga, B., Howard, A.J., Zaharia, S.: The long term fate and environmental significance of contaminant metals released by the January and March 2000 mining tailings dam failures in Maramures County, upper Tisa Basin, Romania. Appl. Geochem. 18, 241–257 (2003) Martin, T.E., McRoberts, E.C., 1999. Some considerations in the stability analysis of upstream tailings dams. In: Proceedings of Tailings & Mine Waste’99, pp. 287–302 Rico, M., Benito, G., Salgueiro, A.R., Díez -Herrero, A., Pereira, H.G.: Reported tailings dam failures: a review of the European incidents in the worldwide context. J. Hazard. Mater. 152, 846–852 (2008) Ruhl, L., Vengosh, A., Dwyer, G.S., Hsu-Kim, H., Deonarine, A., Bergin, M., Kravchenko, J.: Survey of the potential environmental and health impacts in the immediate aftermath of the coal ash spill in Kingston, Tennessee. Environ. Sci. Technol. 43, 6326–6333 (2009) Soares, L., Arnez, F.I., Hennies, W.T.: Major causes of accidents in tailing dam due to geological and geotechnical factors. In: Mine Planning and Equipment Selection—International Symposium, pp. 371–376 (2000) Sun, X., Li, X.: New technology of waste-filling replacement mining on strip coal pillar. J China Coal Soc. 33(3), 259–263 (2008) (in Chinese) Szczepanska, J.: Distribution and environmental impact of coal-mining wastes in Upper Silesia, Poland. Environ. Geol. 38(3), 249–258 (1999) Wang, Y., Sheng, Y., Gu, Q., Sun, Y., Wei, X., Zhang, Z.: Infrared thermography monitoring and early warning of spontaneous combustion of coal gangue pile. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Beijing (2008) Wei, Z., Yin, G., Li, G., Wang, J., Wan, L., Shen, L.: Reinforced terraced fields method for fine tailings disposal. Miner. Eng. 22, 1053–1059 (2008) Yin, G., Li, G., Wei, Z., Wan, L., Shui, G., Jing, X.: Stability analysis of a copper tailings dam via laboratory model tests: A Chinese case study. Miner. Eng. 24, 122–130 (2011) Younger, P.L., Wolkersdorfer, C.: Mining impacts on the fresh water environment: technical and managerial guidelines for catchment scale management. Miner. Water Environ. 23, 2–80 (2004) Zhao, Y., Zhang, J., Chou, C., Li, Y., Wang, Z., Ge, Y., Zheng, C.: Trace element emissions from spontaneous combustion of gob piles in coal mines, Shanxi, China. Int. J. Coal Geol. 73, 52–62 (2008)

Chapter 2

Solutions for Surface Disposal of Mine Tailings

Abstract The technologies of dry stacking and cemented tailings paste are developed to partly replace the traditional treatment, which stores tailings in impoundments. The technological processes for these two technologies are introduced, respectively. Some case applications of dry stacking and paste tailings disposal are also presented. These examples of applications successfully demonstrate that the dry stacking and cemented tailings paste technologies can dispose tailings more effectively with less impact on the environment, in comparison with the traditional treatment of tailings. Keywords Metal mine · Surface disposal · Dry stacking · Cemented tailings · Paste The disadvantages of traditionally discharging mine tailings in tailings impoundments have been discussed before. Instead of the conventional disposing methods of tailings, two main solutions for surface disposal of mine tailings have been proposed and thus used in some applications. These two solutions are the technologies of dry stacking (or dewatered stockpiling) and cemented tailings paste for surface deposition, respectively.

2.1 Dry Stacking of Tailings The dry stacking technology was initially introduced and thus employed in the Western Australian refineries of Alcoa World Alumina Australia (Cooling et al. 2002). Besides, Cooling (2007) further explained the technological process of the dry stacking utilized in Alcoa, as shown in Fig. 2.1. It can be noticed from Fig. 2.1 that, the dry stacking technology uses a superthickener to dewater the tailings for producing a thickened slurry, which is then pumped to layers over the storage areas for self dewatering under the combined effects of drainage and evaporative drying (Cooling et al. 2002; Cooling 2007).

© Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_2

11

12

2 Solutions for Surface Disposal of Mine Tailings

Fig. 2.1 Schematic diagram of the dry stacking process (Cooling 2007)

Although the initial costs of implementing the dry stacking operations at Alcoa’s three Western Australian refineries exceeded $150 million, some other benefits could still be achieved in the application of dry stacking (Cooling 2007): • A higher density deposit can be achieved, hence the overall volume of stored tailings can be significantly reduced. • The progressive stacking allows the deposit to be taken to a height, which would not be economic with conventional wet impoundments. • The higher density and increased deposit height means less land is used. • The exposure of less land area to residue. • The drained condition of the dry stack and the smaller footprint significantly reduce the risk of groundwater contamination. • Improved surface stability and drainage mean that completed areas can be reclaimed and revegetated quickly. • Safety hazards to people and wildlife are reduced. With advances in dewatering technologies, especially the development of large capacity vacuum and pressure filter technology, discharging the tailings in an unsaturated state, rather than slurry and/or paste, can thus be achieved (Davies 2011). Figure 2.2 illustrates another dry stacking process with the applications of both thickener and filter. Lara et al. (2013) reported that at least five operations of dry stacking of filtered tailings were successfully conducted in South America: La Coipa, Mantos Verde,

2.1 Dry Stacking of Tailings

13

Fig. 2.2 Schematic diagram of the dry stacking technology using filter (Lara et al. 2013)

and El Peñon projects, located in Chile, as well as the Cerro Lindo and Catalina Huanca projects, in Peru, and one of them is exemplarily shown in Fig. 2.3. Besides, Stone et al. (2016) indicated an operation of dry stacking of filtered tailings at San José silver mine in Oaxaca State, Mexico, as shown in Fig. 2.4. Moreover, Karamad and Seyf (2016) stated that the application of tailings filtration in the dry stacking operations in National Iranian Copper Industries Co. (NICICO) is due to the limited water resources. In comparison with conventional disposal of tailings in ponds, the dry stacking method allows the storage of tailings without any confining embankment. This can significantly eliminate the problem of tailings dam instability or even failure.

Fig. 2.3 The dry stacking operation of Cerro Lindo Project (Lara et al. 2013)

14

2 Solutions for Surface Disposal of Mine Tailings

Fig. 2.4 The dry stacking operation at San José silver mine of Mexico (Stone et al. 2016)

2.2 Paste Discharging of Tailings In addition to dry stacking, paste disposal of tailings is an alternative treatment, which is to transform dilute slurry of tailings into paste tailings (PT) or cemented paste tailings (CPT), and then discharge, as schematically demonstrated in Fig. 2.5. Cemented paste tailings (CPT) is a pumpable, flowable, non-Newtonian fluid generally prepared by mixing mine tailings, water, and binder. CPT is generally a nonsegregating mixture with a solid concentration of 70–85% and contains enough fines (at least 15% particles less than 20 microns) to prevent settlement and particle

Tailings slurry

Flocculant Binder

Paste thickener

Pump

Mixer

Pump

CPT

PT Fig. 2.5 Schematic diagram for explaining disposal of paste tailings and cemented paste tailings

2.2 Paste Discharging of Tailings

15

segregation when it is transported through a pipeline for disposal (Landriault 1995; Klein and Simon 2006). The CPT technology has been introduced as an innovative solution for surface tailings disposal (Bussière 2007; Cincilla et al. 1997; Deschamps et al. 2008; Shuttleworth et al. 2005; Theriault et al. 2003; Verburg 2002). The key advantages of the CPT disposal are summarized as follows (Benzaazoua et al. 2002, 2004; Cadden et al. 2003): (a) Improved tailings hydro-geotechnical properties; (b) Small amount of free water at the paste surface thereby reducing the dimensions of tailings-retaining structures in the case of underwater storage; (c) Homogeneity of paste induces less particles segregation; (d) Enhanced strength and durability of the resultant monolith; (e) Development of acid neutralization potential and pollutants stabilization capability of the matrix due to addition of alkaline binders. Lara and Maldonado (2016) have reported the practice of high density (or paste) tailings deposits at Animon Mine, which is located 150 km north-east of Lima-Peru in an area of moderate rainfall and moderate to high seismic activity. Since 2009, this mine uses a deep cone thickener (DCT) to thicken tailings (3720 tpd) and mine sludges (93 tpd) until 70% solids content is reached. Afterwards, the tailings are transported by pumping to the deposition area where they are placed in different zones in a cyclical manner to allow desiccation and consolidation, as demonstrated in Fig. 2.6. The deposited tailings have high specific gravity (Sg = 2.9–3.0) and silty clay particle size (80–90% of fines) (Lara and Maldonado 2016). Lara and Maldonado (2016) have also summarized the benefits of implementing paste tailings deposit instead of conventional tailings disposal (Lara and Maldonado 2016): • Maximize the tailings deposit capacity, to maximize the useful life of mine.

Fig. 2.6 Flow chart of the paste tailings disposal at Animon Mine (Lara and Maldonado 2016)

16

2 Solutions for Surface Disposal of Mine Tailings

Fig. 2.7 View of field paste tailings deposit at Animon Mine (Lara and Maldonado 2016)

• Minimize the tailings dam volume and delay the tailings dam raising, to reduce costs. • Minimize the water sent to the tailings deposit, to minimize the water treatment. • Maximize the recovery of process water, to minimize the use of fresh water. • Reduction of recirculating sludge from underground mine, to improve water quality (sediment control) and water recovery from the tailings deposit. Furthermore, Fig. 2.7 shows the view of field paste tailings deposit at Animon Mine. Cacciuttolo and Holgado (2016) have reported several successful cases of paste tailings disposal in Chile: Demo Plant Paste Tailings, Las Cenizas Paste Tailings, Delta Paste Tailings, and Alhué Paste Tailings. The Demo Plant Paste Tailings program was conducted by Collahuasi Mine, which was situated in northern Chile. The Demo Plant was located on the side of a tailings storage facility (TSF), below the tailings distribution tank and parallel to the tailings discharge drop boxes system. The demo plant had a 22 m in diameter, 18 m high DCT, which was the key equipment for producing paste tailings (Cacciuttolo and Holgado 2016). Figure 2.8 displays the overview of in situ paste tailings disposal at TSF of Collahuasi Mine. The Las Cenizas Paste Tailings operation employed a DCT with a height of 16 m and a diameter of 17 m to form paste tailings for surface discharge (Cacciuttolo and Holgado 2016). The field scene of paste tailings disposal at TSF of Las Cenizas Mine is shown in Fig. 2.9. The Delta Paste Tailings project was located in Ovalle valley in the northern region of Chile, approximately 330 km north of Santiago city. The tailings slurry from the process plant with 35% solids was transported by centrifugal pumps to a DCT with a diameter of 12 m and height of 8 m, and the underflow tailings of the DCT were then pumped to the TSF (Cacciuttolo and Holgado 2016). The Delta paste tailings thickening plant and field scene of paste tailings disposal can be investigated in Fig. 2.10. The Alhué Paste Tailings program was operated by Alhué Mine, which was located 180 km south of Santiago city in central Chile. Since 2016 this mine started to utilize a DCT (with a diameter of 17 m and height of 12 m) to make paste tailings and

2.2 Paste Discharging of Tailings

17

Fig. 2.8 Overview of in situ paste tailings disposal at TSF of Collahuasi Mine (Loan et al. 2011; Valdebenito 2012)

Fig. 2.9 DCT overview and paste tailings disposal at TSF of Las Cenizas Mine (Flores 2011; Valdebenito 2012)

discharge them in TSF (Cacciuttolo and Holgado 2016). Figure 2.11 illustrates the on-site paste tailings disposal and DCT used at Alhué Mine. In China, the cases of paste tailings disposal have also been reported. For instance, Wu et al. (2011) have introduced the practice of paste tailings discharge in tailings pond at Wushan Mine, which is located in Inner Mongolia and belongs to China National Gold Group Co., Ltd. Sun et al. (2018) have discussed the paste tailings disposal in collapse pits at Tong Keng Mine, which is located in Guangxi province and operated by Guangxi China Tin Group Co., Ltd. Moreover, Lu et al. (2018) has described the operation of paste tailings disposal in open pits at Shirengou Iron Mine, which is attached to HBIS Group Co., Ltd. and located in Heibei province, China.

18

2 Solutions for Surface Disposal of Mine Tailings

Fig. 2.10 DCT and field disposing scene of the Delta Paste Tailings project (Flores 2011; Valdebenito 2012)

Fig. 2.11 The field paste tailings disposal and DCT used at Alhué Mine (SEA 2015)

References Benzaazoua, M., Belem, T., Bussière, B.: Chemical factors that influence the performance of mine sulphidic paste backfill. Cem. Concr. Res. 32(7), 1133–1144 (2002) Benzaazoua, M., Marion, P., Picquet, I., Bussière, B.: The use of paste fill as solidification and stabilization process for the control of acid mine drainage. Miner. Eng. 17(2), 233–243 (2004) Bussière, B.: Colloquium 2004: hydro-geotechnical properties of hard rock tailings from metal mines and emerging geo-environmental disposal approaches. Can. Geotech. J. 44, 1019–1054 (2007) Cacciuttolo, C., Holgado, A.: Management of paste tailings in Chile: a review of practical experience and environmental acceptance. In: Paste 2016—19th International Seminar on Paste and Thickened Tailings. Chile (2016) Cadden, A., Newman, P., Fordham, M.: New developments in surface paste disposal of mine wastes. In: Proceedings of Disposal Mineral Industry, June 2003, London (2003) Cincilla, W.A., Landriault, D.A., Verburg, R.B.M.: Application of paste technology to surface disposal of mineral wastes. In: Proceedings of the Fourth International Conference on Tailings and Mine Waste’97, Fort Collins, Colorado, 13–16 January, Balkema, Rotterdam, pp. 343–356 (1997) Cooling, D.J., Hay, P.S., Guifoyle, L.: Carbonation of bauxite residue. In: Proceedings of the 6th International Alumina Quality Workshop, Brisbane, pp. 185–190 (2002) Cooling, D.J.: Improving the sustainability of residue management practices—Alcoa World Alumina. In: Fourie, A.B., Jewell, R.J. (eds.) Paste 2007—Proceedings of the Tenth International

References

19

Seminar on Paste and Thickened Tailings, pp. 3–15. Australian Centre for Geomechanics, Perth (2007) Davies, M.: Filtered dry stacked tailings-the fundamentals. In: Proceedings Tailings and Mine Waste 2011, Vancouver, Canada. Keevil, BC (2011) Deschamps, T., Benzaazoua, M., Bussière, B., Aubertin, M., Belem, T.: Micro structural and geochemical evolution of paste tailings in surface disposal conditions. Miner. Eng. 21(4), 341–353 (2008) Flores, I.: Disposición de relaves en pasta Faena Cabildo. In: 3rd Paste Tailings Seminar, RELPAS, Santiago, Chile (2011) Karamad, E., Seyf, H.: Water and tailings management costs in NICICO’s copper concentrator development projects. In: Paste 2016–19th International Seminar on Paste and Thickened Tailings. Chile (2016) Klein, K., Simon, D.: Effect of specimen composition on the strength development in cemented paste backfill. Canad. Geotech. J. 43, 310–324 (2006) Landriault, D.: Paste backfill mix design for Canadian underground hard rock mining. In: Proceedings of the 12th Annual CIM Mine Operation Conference, Timmins, Ontario, February 1995, pp. 1–10. Canadian Institute of Mining and Metallurgy (CIM), Montreal, Que (1995) Lara, J., Maldonado, R.: Design and operational experience of the Animon high density tailings deposit, Peru. In: Paste 2016–19th International Seminar on Paste and Thickened Tailings. Chile (2016) Lara, J., Pornillos, E., Muñoz, H.: Geotechnical-geochemical and operational considerations for the application of dry stacking tailings deposits-state-of-the-art. Paste 2013, pp. 251–262. Belo Horizonte, Brazil. Jewell, R.J., Fourie, A.B., Cadwell, J., Pimenta, J. (eds.) Australian Centre for Geomechanics, Perth (2013) Loan, C., Villanueva, M.L., Saldía, N.: Collahuasi paste thickener-pilot and full scale results. In: Proceedings of the 14th International Seminar on Paste and Thickened Tailings, April 2011, Perth, Australia Lu, H., Qi, C., Chen, Q., Gan, D., Xue, Z., Hu, Y.: A new procedure for recycling waste tailings as cemented paste backfill to underground stopes and open pits. J. Clean. Prod. 188, 601–612 (2018) SEA (Servicio de Evaluación Ambiental) EIA: Depósito de Relaves en Pasta Minera Florida (2015). Viewed December 2015 at: http://seia.sea.gob.cl/documentos/documento.php?idDocumento= 7889337 Shuttleworth, J.A., Thomson, B.J., Wates, J.A.: Surface paste disposal at Bulyanhulu e practical lessons learned. In: Proceedings of the 8th International Seminar on Paste and Thickened Tailings—Paste 2005, Santiago, Chili, pp. 207–218 (2005) Stone, D., Pearson, K., Pacora, J. Paste backfill at the San José Silver Mine, Mexico. In: Paste 2016–19th International Seminar on Paste and Thickened Tailings. Chile (2016) Sun, W., Wang, H., Hou, K.: Control of waste rock-tailings paste backfill for active mining subsidence areas. J. Clean. Prod. 171, 567–579 (2018) Theriault, J.A., Frostiak, J., Welch, D.: Surface disposal of paste tailings at the Bulyanhulu gold mine, Tanzania. In: Proceedings of Sudbury 2003 Mining and the Environment, May 25–28, Sudbury, Ontario (2003) Valdebenito, R.: Tranques de relaves espesados y en pasta en Chile. In: FLSmidth Tailings Disposal Seminar, May 2012, Santiago, Chile (2012) Verburg, R.B.M.: Paste technology for disposal of acid-generating tailings. Mining Environ. Manage. 14–18 (2002) Wu, A., Yang, S., Wang, H., Jiao, H., Xiao, Y.: Status and trend of paste disposing technology of ultrafine tailings. Min. Technol. 11(3), 4–8 (2011)

Chapter 3

Case Study of Surface Consolidated Tailings Stockpile

Abstract Waste tailings have increasingly been used as materials to backfill mined-out areas. However, the tailings produced cannot be fully utilized for void backfilling, so other storage locations are needed, such as worked out open pits and subsidence areas. A CTS (consolidated tailings stockpile) technology for disposing tailings in mining-induced subsidence areas is presented. The corresponding experiments and calculations required before this technology can be implemented have been discussed. UCS tests were conducted to determine the preferred type of binder. Numerical simulations were developed to select the optimum binder dosage. The outcomes of filtration experiment indicated that the addition of binder can improve the processing capability of the filter. Finally, based on these obtained outcomes, the technical process of the CTS technology in an engineering application was introduced in detail, and the social and environmental benefits of using this technology in Xishimen Iron Mine were also discussed. Keywords Tailings · Consolidation · Backfill · Subsidence areas · Filtering

3.1 Introduction The mining industry provides people valuable mineral products, but in the meantime, it also creates enormous volumes of mine waste. As the biggest developing country in the world, China, during the past few decades, has achieved rapid development, which is undoubtedly supported by the utilization of the extracted mineral products. However, the disposal of these mine waste has become a serious environment problem, arousing wide and intensive social concerns. By the end of 2012, there are more than 12,000 tailings impoundments in China (Yin et al. 2011), storing about 12 billion tons of tailings. The tailings from mineral processing plants contain beneficiation reagents and most of them are poisonous. Therefore, the traditional disposing method of storing tailings slurries in the tailings impoundments without any treatment may cause serious environment contamination. For example, in 1998, the spill of the Aznalcóllar mine tailings induced by dam failure resulted in severe pollution to the environment (López-Pamo et al. 1999; Gallart et al.

© Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_3

21

22

3 Case Study of Surface Consolidated Tailings Stockpile

1999). Moreover, the failure of tailings dam can also lead to economic loss or even casualty accidents. In addition to the above-discussed disadvantages of conventional disposal of tailings, it has become more and more difficult for mining enterprises to find places to newly build impoundments to store tailings, especially in China. For these reasons, novel solutions for disposal of tailings are urgently needed to replace the traditional treatment. During the past decades, dry stacking has been developed and used for surface disposal of tailings. The benefit of this technique is reducing some of the risks associated with conventional practice, most signally obviating the necessity of dams, increasing water recycling and facilitating rapid reclamation (Mizani et al. 2013). However, the hidden risk and environmental impact of using this method still remain. For instance, rainfalls may cause the massive collapse of the stacked “dry” tailings or even debris flows (Sun et al. 2018). For another example, wind may blow up the “dry” tailings, causing dust or even sandstorms. Consequently, there is an urgent need to improve the conventional dry stacking technology. Enlightened by the cemented paste tailings technology, the technique of consolidated tailings stockpile (or cemented dry stacking) is created and utilized (Hou et al. 2011). This approach is mixing some binder into the filtered tailings and then stockpiling them on the surface. The cementation of the mixed binder can effectively consolidate the tailings together, preventing the “dry” tailings from sliming in water, and also flying with the wind. Xishimen Iron Mine, which is located at Hebei province of China, has used pillarless sublevel caving method to extract iron ore for decades. Therefore, large collapse pits are formed on surface, as shown in Fig. 3.1. The storage capacity of the existing tailings impoundment is not enough, and the government does not approve this mine to build a new reservoir. For this reason, this mine attempts to take advantage of the collapse pits for disposing tailings. This solution can not only relieve the limited capacity of current tailings storage, but also contribute to a balanced treatment of both waste tailings and subsidence areas. In this chapter, the practice of consolidated tailings stockpile in the subsidence areas of Xishimen Iron Mine is employed as a case study to illustrate the utilization of this technology. The organization of this chapter is set as follows. First, the materials and experimental procedure are introduced in Sect. 3.2. Second, the obtained results and discussions are presented in Sect. 3.3, including the outcomes of laboratory tests and numerical simulations. Finally, the conclusions are indicated and summarized in Sect. 3.4.

3.2 Materials and Methods

23

Fig. 3.1 Subsidence areas of Xishimen Iron Mine

3.2 Materials and Methods 3.2.1 Materials The consolidated tailings stockpile (CTS) mixtures are comprised of tailings, cement, and water. The tailings used are from Xishimen Iron Mine. Two types of binders are selected for the current study, and they are respectively ordinary Portland cement 425# and slag Portland cement 325#. The water used is tap water. Tables 3.1 and 3.2 exhibit the physical characteristics and main chemical compositions of the tailings Table 3.1 Physical characteristics of the tailings used

Tailings

Density (kg/m3 )

D10 (µm)

D30 (µm)

D50 (µm)

D60 (µm)

Cu

Cc

2860

54.1

117.2

186.8

224.0

4.14

1.13

Table 3.2 Main chemical compositions of the tailings used Chemical compositions (mass%) Tailings

SiO2

Al2 O3

Fe2 O3

CaO

MgO

K2 O

Na2 O

MnO

29.31

6.42

4.02

26.08

14.66

1.55

1.68

0.82

24

3 Case Study of Surface Consolidated Tailings Stockpile

Table 3.33 Main chemical compositions of the binder used Chemical compositions (mass%) SiO2

Al2 O3

Fe2 O3

CaO

MgO

SO3

Ordinary cement

19.31

4.93

3.12

63.15

3.26

2.32

Slag cement

23.48

6.26

2.39

57.40

3.31

2.02

Fig. 3.2 Particle size distribution of the tailings used

used. Table 3.3 lists the main chemical compositions of the ordinary cement and slag cement. Figure 3.2 illustrates the particle size distribution of the tailings.

3.2.2 Experiment Procedure The experimental process for preparing the CTS mixtures is described as follows: dry tailings and water are mixed to make slurries with mass concentration of 40%. After that, the prepared tailings slurries are put into a mixer, and meanwhile, cement is also added into the mixer. When the tailings slurries and cement are mixed homogeneously, they are delivered to a filter. After filtering, the produced filter cakes, which possess a mass concentration of around 80%, are then placed into curing cubes with dimension of 7 cm × 7 cm × 7 cm in length × width × height to form cubic CTS samples, and then the CTS specimens are cured in HSBY-60B standard curing chamber at the temperature of 20 ± 1 °C and for various periods of 3, 7 and 28 days. When the scheduled curing age is completed, the CTS specimens are subjected to uniaxial compressive strength (UCS) tests.

3.2 Materials and Methods

25

3.2.3 UCS and Wetting Tests The UCS tests on the CTS specimens are then conducted using a computer-controlled mechanical press, which has a normal loading capacity of 50 kN. During the test, the displacement rate of the compressive loading is set at 0.5 mm/min. The CTS specimens prepared with different binder contents are subjected to the wetting tests (or soaking experiments). The CTS specimens tested are immersed in water for 90 min, and then they are observed that if they have been defeated and dispersed or not.

3.2.4 Numerical Simulation Numerical simulating studies are conducted to determine the content of binder selected for preparing the CTS mixtures.

3.3 Results and Discussion 3.3.1 Selection of Binder Type Since the mechanical performance of CTS mixtures can significantly ensure the safety of subsequent reclamation on them. An important parameter used in practice to judge the mechanical performance of CTS is its uniaxial compressive strength (UCS), and the measurement of UCS is also convenient. Therefore, in this section, several laboratory experiments are carried out to measure the UCS values of CTS samples. As indicated before, two types of binder (ordinary cement and slag cement) are selected to prepare the CTS specimens, and the present experimental study aims to find the better binder. The UCS performances of these CTS samples are shown in Fig. 3.3. From this figure, it can be clearly noticed that the UCS values of the CTSs prepared by slag cement are higher than that by ordinary cement. As a result, slag cement is selected as the binder for preparing CTS mixtures, which are delivered into the target subsidence areas. As expected, Fig. 3.3 also illustrates that the UCS of CTS increases with curing age, and a higher cement ration is associated with a higher UCS value.

3.3.2 Determination of Binder Content The content of the binder used is significantly related to the stability of CTS mixtures and also the cost. Therefore, it is essential to determine the binder content. In this

26

3 Case Study of Surface Consolidated Tailings Stockpile

Fig. 3.3 Effects of cement-to-tailings ratio and curing age on UCS of CTS

section, the CTS structures cured for the same age (3 days) but with 5 different kinds of binder contents (2.0, 2.5, 3.0, 3.5 and 4.0%) are chosen to conduct a simulation study, for selecting a binder content that is both mechanically and economically applicable. The binder used is the slag cement shown in Table 3.3. Figure 3.4 shows the profile map of the collapse pit of Xishimen Iron Mine which will be filled with

Fig. 3.4 Profile map of the collapse pit with mesh generation

3.3 Results and Discussion

27

Table 3.4 Mechanical parameters of the CTS structures with various binder contents Binder content (%)

Bulk modulus (MPa)

Shear modulus (MPa)

Cohesion (kPa)

Density (kg/m3 )

Internal friction angle (°)

2.0

15.97

8.68

83

1927

24

2.5

17.39

9.45

104

1976

26

3.0

19.66

10.68

137

2000

27

3.5

20.17

10.96

159

2028

28

4.0

21.33

11.59

212

2043

30

Table 3.5 Mechanical parameters of the collapse pit Bulk modulus (GPa)

Shear modulus (GPa)

Cohesion (MPa)

Density (kg/m3 )

Internal friction angle (°)

37.2

22.3

55

2700

45

the CTS structures. Tables 3.4 and 3.5 list the mechanical parameters of the CTS structures and collapse pit used for numerical simulation. Figure 3.5 demonstrates the plastic zone distribution in the filled CTS structures that are prepared by various dosage of binder. From these figures, it can be noticed that when the binder content selected is 2.0%, a plastic zone is clearly developed within the filled CTS structure from the surface to the bottom. When increasing the binder content to 2.5%, although the plastic zone becomes obviously smaller, it still develops through the whole filled CTS structure. However, if the binder content used is equal or more than 3.0%, no penetrative plastic zone can be found in the filled CTS structure. Thus, the optimum binder content should be 3.0%, which can not only ensure the stability of the filled CTS but also more cost-effective in comparison with the binder content of 3.5% and 4.0%. The results of wetting tests on the CTS specimens (with dimension of 7 × 7 × 7 cm3 ) prepared with various binder (slag cement) contents can be observed in Fig. 3.6. From Fig. 3.6 it can be clearly found that, after 90 min’s soaking in tap water, the CTS specimens prepared with more than 1.8% content of slag cement can endure the water softening. The obtained results strongly support the availability of the above simulation study, which selects 3.0% as the binder content for preparing the CTS mixtures.

3.3.3 Effect of Binder on Dewaterability of Filter The cemented tailings slurries should be filtered to form CTS mixtures before being placed in the subsidence areas. Therefore, the processing capability of filter is a key factor significantly affecting the efficiency of CTS production and transportation. In

28

3 Case Study of Surface Consolidated Tailings Stockpile

(a) Binder content: 2.0%

(c) Binder content: 3.0%

(b) Binder content: 2.5%

(d) Binder content: 3.5%

(e) Binder content: 4.0% Fig. 3.5 Plastic zone distribution in the filled CTS structures using different binder content

this section, the impact of binder on the processing capability of filter is investigated, by experimentally comparing the processing capability of the filter on tailings slurries with that on cemented tailings slurries. The slurries samples are added with different contents of binder (0 and 3%) and dewatered with different periods (21 and 25 s). For each sample, 3 measurements are carried out, and the average data are collected. The binder used in this experiment is the slag cement shown in Table 3.3, and the filter used is a ceramic filter (as shown in Fig. 3.7), which mainly includes a ceramic filter plate, a vacuum pump, a vacuum flask, a bullet valve, a vacuum rubber hose, and a piezometer. The filter plate is coated with a layer of membrane, which has a pore diameter of 1.8 µm and a thickness of 0.45–0.55 µm.

3.3 Results and Discussion

Fig. 3.6 The CTS specimens before (a) and after (b) the wetting tests

Fig. 3.7 The ceramic filter used in the present study

29

30

3 Case Study of Surface Consolidated Tailings Stockpile

The measured results of this experiment are listed in Table 3.6. Figure 3.8 demonstrates the effect of binder on the processing capability of the filter with different dewatering time. It can be discovered from this figure and Table 3.6 that, the dewaterability of the ceramic filter on cemented tailings slurries is notably higher than that on tailings slurries in spite of the dewatering time. In other words, the addition of binder on tailings slurries increases the processing capability of the ceramic filter on them. The reason may be ascribed to the fact that the binder reacts with water to form hydration products. The generated products are cementitious and thus can bond solid particles together, which is notably beneficial to the process of solid-liquid separation (or dehydration) functioned by the filter. Table 3.6 Measured data of the filter dewatering test Slurry type

Concentration Binder of the content (%) slurries (%)

Dewatering time (s)

Concentration Processing of the filtered capability cakes (%) (kg/m2 h)

Tailings slurries

40

21

79.9

418.7

25

80.4

389.6

Cemented tailings slurries

40

21

80.5

622.4

25

80.7

643.2

0 3

Fig. 3.8 Effect of binder on the filter processing capability

3.3 Results and Discussion

31

3.3.4 Industrial Application Based on the above discussions, the binder used for preparing CTS mixtures is chosen as slag cement and the corresponding dosage is 3%. The technological progress for the preparation and placement of CTS mixtures in industrial application is schematically displayed in Fig. 3.9. It is seen in this figure that, the tailings slurries from the mineral processing plant are transported into a thickener. After concentrating, the thickened tailings slurries are pumped into a mixer, and meanwhile, cement is also added into the mixer. When the tailings slurries and cement are blended homogeneously, they are transferred to a filter. After filtering, the produced filter cakes (i.e., CTS mixtures) are delivered onto a belt conveyor, for being conveyed into the collapse pit (or subsidence areas). The main processes of the CTS technology include concentration of tailings slurries, addition of binder, filtration, and transportation of CTS. It should be pointed out that, the mass concentrations of the thickened tailings slurries and filter cakes are about 40 and 80%, respectively. The reason why the addition of binder is after thickening but before filtering has been explained in Sect. 3.3.3. In addtion, the filtered cakes are not able to provide Fig. 3.9 Flowsheet for preparation and placement of CTS mixtures

Tailings slurries feed

Binder silo

Thickener

Mixer Underflow pump

Transfer pump

Filter Water recycling pond

Belt conveyor

Collapse pit

32

3 Case Study of Surface Consolidated Tailings Stockpile

enough reacted water for the progress of binder hydration, hence the addition of binder should not be after the filtering (i.e., the binder should not be added into the filtered cakes). Figure 3.10 shows the thickener (type: HRC18) used in the field and Fig. 3.11 shows the filter (GPT60-120) used. Tables 3.7 and 3.8 illustrate the main technical specifications of the thickener and filter, respectively. Figure 3.12 displays the overview of disposing CTS mixtures in the field. The utilization of CTS technology in Xishimen Iron Mine can achieve very significant social and environmental benefit, as illustrated as follows: (i)

The discharge of CTS in subsidence areas effectively avoids the land (especially the cultivated land) occupation and also saves the storage of tailings impoundment. (ii) This technology can surely reduce the new construction of tailings impoundments, and thus relieve the relevant risk, such as failure of tailings dam and water pollution. (iii) The application of CTS technology can notably contribute to the land reclamation and ecological environment reconstruction in the subsidence areas.

Fig. 3.10 View of the thickener used in the field

3.3 Results and Discussion

33

Fig. 3.11 View of the filter used in the field Table 3.7 Main technical specifications of the thickener

Table 3.8 Main technical specifications of the filter

Parameter

Value

Unit

Maximum tailings throughput

30

t/h

Diameter

18

m

Height

10.45

m

Volume of distribution

1200

m3

Parameter

Value

Unit

Filtering area

64

m2

Diameter of the filter disc

2.6

m

Number of the filter disc

8

–

Dimension (length × width × height)

5565 × 3290 × 3190

mm

Weight

15.9

T

Power of spindle motor

5.5

kw

Speed of mainshaft

0.1–1

rpm

34

3 Case Study of Surface Consolidated Tailings Stockpile

Fig. 3.12 View of CTS mixtures disposal in the field

3.4 Concluding Remarks Traditional treatment of tailings in tailings ponds has caused serious problems, such as dam breakage accident and environment contamination, and these issues have drawn continuous and close attention by the government and people in China. In response, waste tailings are widely and intensively used as backfill materials to fill underground stopes. However, not all the waste tailings produced can be placed underground. For this reason, some open pits (of closed-open pit mines) and collapse pits (produced by caving mining method) can be used for the placement of tailings. This chapter presents a new management of tailings in mining-induced subsidence areas by using CTS technology. Based on the obtained results in this study, the following conclusions can be drawn. (1) Slag cement is more suitable than ordinary Portland cement in the preparation of CTS mixtures, due to the fact that the slag cement can improve the CTS performance both mechanically and economically. (2) An optimum dosage of binder (3.0%) in making CTS is selected from a range of choices. This selection can ensure the stability of the filled CTS structures in the pit at the minimum cost. (3) The addition of binder is a key step for the application of CTS technology. The test results indicate that adding the binder after thickening but before filtering

3.4 Concluding Remarks

35

can improve the capability and efficiency of the filter. These findings provide a crucial contribution to the design and implementation of the whole procedure of the CTS technology. (4) The introduction and practice of CTS technology in Xishimen Iron Mine is a good attempt in replacing the traditional management of tailings in ponds. This solution can also provide inspirations for the reclamation and sustainable development of similar mines. This study can provide useful information for the design, preparation, and utilization of CTS mixtures. Although the CTS technology is introduced to relieve the damage of tailings ponds, the hazard of the CTS itself should be considered and investigated. For instance, the hazardous and noxious substances contained in the CTS may pollute the environment through wind and rainfall. Therefore, a future study should focus on the assessment of the environmental performance of CTS mixtures.

References Gallart, F., Benito, G., Vide, M.J.P., Benito, A., Prió, J.M., Regüés, D.: Fluvial geomorphology and hydrology in the dispersal and fate of pyrite mud particles released by the Aznalcóllar mine tailings spill. Sci. Total Environ. 242(1–3), 13–26 (1999) Hou, Y., Tang, J., Wei, S.: Research on tailings’ cementation and discharging technology. Metal Mine 6, 59–62 (2011) (in Chinese) López-Pamo, E., Barettino, D., Antón-Pacheco, C., Ortiz, G., Arránz, J.C., Gumiel, J.C., MartínezPledel, B., Aparicio, M., Montouto, O.: The extent of the Aznalcóllar pyritic sludge spill and its effects on soils. Sci. Total Environ. 242, 57–88 (1999) Mizani, S., He, X., Simms, P.: Application of lubrication theory to modeling stack geometry of high density mine tailings. J. Non-Newton. Fluid 198, 59–70 (2013) Sun, W., Wang, H., Hou, K.: Control of waste rock-tailings paste backfill for active mining subsidence areas. J. Clean. Prod. 171, 567–579 (2018) Yin, G., Li, G., Wei, Z., Wan, L., Shui, G., Jing, X.: Stability analysis of a copper tailings dam via laboratory model tests: A Chinese case study. Miner. Eng. 24, 122–130 (2011)

Chapter 4

Case Study of Surface Cemented Tailings Discharge

Abstract After over a hundred years’ mining operations, a large open pit is formed in Gushan Open pit Mine. This open pit is planned to store waste tailings. Since mining operations will still be proceeded under the open pit. Hence, a combination of cemented tailings discharge (CTD) aboveground and backfill underground should be used, which is called B-CTD system. The technical process for the operation of the B-CTD system is introduced. A base structure should be formed at bottom of the open pit, ensuring the safety of both underground mining and surface tailings disposal. A numerical study has been conducted to determine the thickness and strength of the base structure. Keywords Open pit · Cemented tailings discharge · Backfill · Base structure · Transition For open pit mining mines, open pits are often formed when the orebodies are mined. If the tailings can be consolidated and discharged into open pits without threatening the underground mining activities, the following benefits can be possibly achieved: (1) Expansion of existing tailings impoundment or construction of new tailings ponds can be avoided; (2) Occupation of cultivated land can be reduced; (3) Open pits can be harnessed and the environment of mine area can be improved. Gushan Open pit Mine locates at Dangtu County, Anhui Province, China, and belongs to Gushan Mining Company, which is operated by Magang (Group) Holding Co., Ltd. Gushan Open pit Mine, with over a hundred years of iron ore mining history, will be closed at the end of 2019, and the mining operation will transform from open pit mining to underground mining. The being closed pit has a depth of about 170 m and a volume of about 27.5 million m3 , as shown in Fig. 4.1. The mine plans to use the open pit as a storage of tailings. Before placement of tailings, the bottom of the open pit should be reinforced with anti-seepage treatment, in order to ensure that the discharge of tailings in the open pit would not exert detrimental influence on the underground mining operation. Therefore, a base structure is planned to build at the open pit bottom.

© Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_4

37

38

4 Case Study of Surface Cemented Tailings Discharge

Fig. 4.1 View of the Gushan open pit

4.1 Selection of Tailings Disposal Method Two schemes are proposed for the tailings disposal in Gushan Open pit Mine, being cemented dry stacking (CDS) and cemented tailings (or paste) discharging (CTD). For the CDS technology, a new system for preparation and placement of the consolidated tailings stockpile (CTS, as discussed in Chap. 3) needs to be constructed. According to some relevant content presented in Chap. 3, Table 4.1 illustrates the technological parameters designed for the CDS system. As described before, mining operations will be continued under the closed open pit, and a backfill mining method will be employed. For this reason, a backfill system needs to be built, for preparing, transporting and placing backfill materials. The backfill system can be used, if a new delivery system is added, for also preparing cemented tailings and discharging them into the surface open pit. The system of Table 4.1 Main technological parameters for the CDS system

Technological parameter

Value

Processing capacity of tailings

300 t/d

Solids content of tailings feeded

18–20%

Flow of tailings feeded

30–36 m3 /h

Content (or percentage) of binder used

2–3%

Amount of binder used

0.5–0.8 t/h

Processing capacity of filter

70 kg/m2 h

Solids content of tailings filtered (or filter cakes)

75–80%

4.1 Selection of Tailings Disposal Method

39

backfill combined with the surface disposal can be called B-CTD system. On the basis of the design for the backfill system being built, Table 4.2 shows the technological parameters for the B-CTD system. It should be noted that the tailings processed (1009.33 t/d) and feeded (102 ~ 122 m3 /h) in Table 4.2 include the tailings backfilled underground and discharged aboveground. Figure 4.2 schematically illustrates the technical process for the B-CTD system operation. Based on the relevant data in Tables 4.1 and 4.2, the budget estimates for establishing the CDS and B-CTD systems are both calculated, with the results shown in Tables 4.3 and 4.4, respectively. It has been stated before that the tailings processed and feeded in Table 4.2 include that backfilled underground (accounting for about 70%) and discharged aboveground (30%), so it is regarded that the cost for surface CTD accounts for 30% of the budget estimate of the whole B-CTD system. Consequently, the cost for surface CTD is calculated to be 3,620,700 CNY, which is lower than the cost of the CDS system Table 4.2 Main technological parameters for the B-CTD system

Technological parameter

Value

Processing capacity of tailings

1009.33 t/d

Solids content of tailings feeded

18–20%

Flow of tailings feeded

102–122 m3 /h

Cement-tailings ratio

1:4–1:6

Amount of binder used

56 t/d

Backfilling capacity of the system

80–100 m3 /h

Solids content of fresh cemented tailings materials

70%

Tailings feed

Thickener

Binder

Open pit Mixer Underflow pump

Transfer Pump

Stope

Fig. 4.2 Technical process for the B-CTD system operation

40 Table 4.3 Budget estimate for establishing the CDS system

4 Case Study of Surface Cemented Tailings Discharge Item Pipeline

200,000

Device for binder feed

1,200,000

Filter

1,200,000

Backwater pump and auxiliary pipeline

500,000

Belt convey configuration

200,000

Inspection control device

500,000

Civil construction and auxiliary facility

400,000

Cost for designing the system

200,000

Contingency Total cost

Table 4.4 Budget estimate for establishing the B-CTD system

Cost (CNY)

Item Civil construction

500,000 4,000,000

Cost (CNY) 933,000

Equipment acquisition

9,600,000

Equipment installation

960,000

Equipment operation Total cost

576,000 12,069,000

(4,000,000 CNY). It is indicated that the CTD technology is better than the CDS technology economically. In addition, the discharge of cemented tailings through a pipeline is more flexible than the placement of consolidated tailings via a belt. The base structure of the open pit can also be formed by the cemented tailings materials with high cement-tailings ratio when the CTD technology is used. Finally, the CTD technology is chosen as the scheme for tailings disposal.

4.2 Determination of Strength and Thickness of the Base Structure 4.2.1 Numerical Modeling Figure 4.3 shows the 3D DTM model of the Gushan open pit developed by 3DMine. The 3D DTM model of the Gushan open pit is implemented into the software COMSOL Multiphysics to form a geometric model (Fig. 4.4a), with mesh generation as shown in Fig. 4.4b. Table 4.5 lists the mechanical parameters of the geotechnical materials used for model computation. During the simulation, gravity stress is considered, without

4.2 Determination of Strength and Thickness of the Base Structure

41

Fig. 4.3 3D DTM model of the Gushan open pit

Fig. 4.4 Geometric model of the Gushan open pit with mesh generation Table 4.5 Mechanical parameters of the geotechnical materials (CTM: cemented tailings materials) Parameter Density (kg

Orebody m−3 )

Elastic modulus (GPa)

Rock

Soil

CTM

4120

2700

2100

1920

14

18

3

5.44

Poisson’s ratio

0.27

0.25

0.22

0.23

Internal friction angle (°)

42

29

22

26

0.35

0.27

0.07

0.16

0.051

4.7 × 10−4

Cohesive force (MPa) Permeability coefficient (m

d−1 )

0.12

7.8 ×

10−7

42

4 Case Study of Surface Cemented Tailings Discharge

Fig. 4.5 Initial stress distribution in the open pit

considering the tectonic stress, which is insignificant since the open pit locates at shallow ground. The geometric model with input parameters is imported into COMSOL for simulation. The initial model calculation results are shown in Fig. 4.5. It can be seen that the overall stress distributed in the open pit gradually increases from top to bottom because of the gravity effect.

4.2.2 Simulation Results and Discussions The thickness and strength of the base structure at the bottom of the open pit are associated with the cost for constructing the base structure and also the safety of mining operation underneath. Based on an overall consideration of cost and safety, the thickness values (3, 4, 5, 6, 7, 8, 9 m) and strength values (0.2–2.0 MPa) are selected for simulation. A typical section that is crossing both the open pit and orebody is selected for visually displaying the simulation results, as shown in Fig. 4.6. The thickness values of 3 m are taken as an example to show the simulation results. Figure 4.7 graphically demonstrates the plastic zone distribution within the underground stope and the surface discharged cemented tailings materials when the open pit is fully filled. The thickness of the base structure is 3 m, and its strength varies from 0.2 to 2.0 MPa. Figure 4.8 displays the vertical displacement of the model when the open pit is fully filled and the base structure is 3 m thick.

4.2 Determination of Strength and Thickness of the Base Structure

43

Fig. 4.6 Typical section crossing the open pit and orebody

Figure 4.9 reveals the stress distribution in the underground stope and the surface CTMs when the open pit is fully filled. The base structure is also 3 m thick and its strength also changes from 0.2 to 2.0 MPa. According to analysis of the all the model simulation results, a both economically and securely suitable strength value for the base structure is obtained corresponding to each kind of thickness (3, 4, 5, 6, 7, 8, 9 m). Table 4.6 shows the suitable strength value for the base structure at each thickness.

0.2 MPa

0.4 MPa

0.6 MPa

0.8 MPa

Fig. 4.7 Plastic zone distribution within the model when the open pit (with a 3 m thick base structure) is fully filled

44

4 Case Study of Surface Cemented Tailings Discharge

1.0 MPa

1.2 MPa

1.4 MPa

1.6 MPa

1.8 MPa

2.0 MPa

Fig. 4.7 (continued)

4.2 Determination of Strength and Thickness of the Base Structure

0.2 MPa

0.4 MPa

0.6 MPa

0.8 MPa

1.0 MPa

1.2 MPa

1.4 MPa

1.6 MPa

Fig. 4.8 Vertical displacement of the model with a 3 m thick base structure

45

46

4 Case Study of Surface Cemented Tailings Discharge

1.8 MPa

2.0 MPa

Fig. 4.8 (continued)

0.2 MPa

0.4 MPa

0.6 MPa

0.8 MPa

Fig. 4.9 Stress distribution in the model with a 3 m thick base structure

4.2 Determination of Strength and Thickness of the Base Structure

1.0 MPa

1.2 MPa

1.4 MPa

1.6 MPa

1.8 MPa

2.0 MPa

47

Fig. 4.9 (continued) Table 4.6 Suitable strength value for the base structure at each thickness Thickness of the base structure (m) Strength (MPa)

3

4

5

6

7

8

9

>2.0

1.6–1.8

1.6

1.4

1.2–1.4

1.0

0.8

Chapter 5

Solutions for Underground Placement of Mine Tailings

Abstract The technology of cemented paste backfill (CPB) has been increasingly and widely used for mine tailings management and ground control. CPB, which is a mixture of tailings, binder, water, and additives if necessary, is transported underground through pipeline by gravity or gravity/pump or pump/gravity. Some case applications have been introduced to present the representative technological processes for preparation of CPB, which is the key to CPB technology utilization. Keywords Metal mine · Tailings · Cemented paste backfill · Gravity · Pump · Thickening In recent years, the technology of cemented tailings backfill (CTB) or cemented paste backfill (CPB) has been introduced and utilized as an effective application for disposing waste mine tailings (Chen et al. 2018; Fall and Benzaazoua 2005; Ghirian and Fall 2014; Helinski et al. 2010; Kesimal et al. 2003; Qi et al. 2018; Wu et al. 2014, 2016; Yilmaz et al. 2009). CPB is an engineered mixture prepared by blending tailings, binder, and water, as well as additives (such as slag, fly ash, retarder, water reducer, suspending agent, and so on) if necessary. Generally, the tailings from the mineral processing operations are thickened, as shown in Fig. 5.1. The CPB prepared is flowable and usually placed into underground mined-out stopes through pipeline by gravity or gravity/pump or pump/gravity, as shown in Fig. 5.2. Tariq and Yanful (2013) have summarized the benefits associated with the underground disposal of CPB mixtures. They state that CPB (underground disposal) technology has a number of advantages over other mine tailings management strategies including: (a) deposition into mine voids created by mining operation, thus providing an enhanced level of local and regional stability to the ore body in addition to providing applicable and economical disposal of mining-associated waste (Rankine and Sivakugan 2007); (b) increase in the available ore reserves by acting as secondary ground support pillars favoring mine stability (Benzaazoua et al. 2004a); © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_5

49

50

5 Solutions for Underground Placement of Mine Tailings

Tailings feed

Thickener

Binder

Mixer Transfer Pump

Underflow pump Stope

Fig. 5.1 The key step of thickening tailings to prepare CPB

Fig. 5.2 Basic configurations for underground placement of CPB (Thomas et al. 1979; Belem and Benzaazoua 2008)

5 Solutions for Underground Placement of Mine Tailings

51

Fig. 5.3 Schematic diagram of a working platform formed by CPB (Hassani and Bois 1992; Belem and Benzaazoua 2008)

(c) about 60% decrease in the amount of sulphidic waste that has to be disposed on the surface, thus reducing pollution and negative environmental impacts (Benzaazoua et al. 2004b); (d) absence of free water in the paste fill system thereby avoiding drainage requirements during curing period and resulting in faster stope cycle time. In underground mining applications, paste backfill can serve as a construction material to create a floor (such as Fig. 5.3) to mine on top of, a wall to mine next to, and a roof or head cover to mine under (Landriault et al. 1997). Due to the advantages of CPB technology in both disposing mine tailings and mined-out areas, it has been widely and intensively utilized all over the world. Figure 5.4 demonstrates the technological process for CPB preparation at the Louvicourt mine, Canada (Cayouette 2003). Mill tailings are first fed into a thickener to increase solids concentration from 35 wt% to approximately 55–60 wt%. Flocculent is added to aid filtration. The thickened tailings are then pumped from the thickener to a high-capacity holding tank. From the surge tank, the thickened tailings are gravity-fed to disc filters operating alone or in parallel to produce filter cake with a solids concentration of approximately 70–82 wt%. The filter cake is then discharged onto a belt (or reversible) conveyor and fed to a screw feeder for weighing. Finally, filter cake batches are mixed in a spiral (or screw) mixer with binder and water added for about 45 s to produce CPB (Belem and Benzaazoua 2008; Cayouette 2003). Figure 5.5 shows a photo of the CPB preparing plant at Sibanye Gold, Cooke Operations (van der Spuy et al. 2016). Figure 5.6 illustrates the technological process for CPB preparation at Tonglushan copper mine, China (Cai and Wang 2012). Figures 5.7 and 5.8 reveal the technological processes for CTB preparation at Qixiashan lead-zinc mine and Zhangmatun mine, China, respectively (Cai and Wang 2012).

52

5 Solutions for Underground Placement of Mine Tailings

Fig. 5.4 Technological process for preparation of CPB at the Louvicourt mine, Canada (Cayouette 2003)

Fig. 5.5 CPB preparation plant at Sibanye Gold, Cooke Operations (van der Spuy et al. 2016)

5 Solutions for Underground Placement of Mine Tailings

53

1-Total tailings bin; 2-Belt filter press; 3-Cement bin; 4-Cement truck; 5-Slag bin; 6-belt conveyor; 7-Screw conveyor; 8-Paddle mixer; 9-Double-screw mixer; 10-Double-piston pump

Fig. 5.6 Technological process for CPB preparation at Tonglushan copper mine, China (Cai and Wang 2012)

1-Mill total tailings; 2-tailings pool; 3-Pump; 4-Mixer of flocculant & tailings; 5,6-Tailings bin; 7-Flocculant adding; 8-Cement truck; 9-Cement bin; 10-Dust cleaner; 11-Double-screw conveyor; 12-Screw weigher; 13-1st mixer; 14-2nd mixer; 15-Pulp pump; 16-High pressure water pump

Fig. 5.7 Technological process for CTB preparation at Qixiashan lead-zinc mine, China (Cai and Wang 2012)

54

5 Solutions for Underground Placement of Mine Tailings

1-Tailings pool; 2-Clamshell crane; 3-Vibrating feeder; 4-Belt conveyor; 5-Weigher; 6-Dust cleaner; 7-Double-axial mixer; 8-High speed mixer; 9-Cement bin; 10-Water tower; 11-Screw weigher; 12-Screw conveyor; 13-Concrete pump; 14-Filling pipe; 15-Controlling room

Fig. 5.8 Technological process for CTB preparation at Zhangmatun mine, China (Cai and Wang 2012)

References Belem, T., Benzaazoua, M.: Design and application of underground mine paste backfill technology. Geotech. Geol. Eng. 26, 147–174 (2008) Benzaazoua, M., Fall, M., Belem, T.: A contribution to understanding the hardening process of cemented paste. Miner. Eng. 17(2), 141–152 (2004a) Benzaazoua, M., Marion, P., Picquet, I., Bussière, B.: The use of paste fill as solidification and stabilization process for the control of acid mine drainage. Miner. Eng. 17(2), 233–243 (2004b) Cai, S., Wang, H.: Modern backfill theory and technology. Metallurgical Industry Press (2012) (in Chinese) Cayouette, J.: Optimization of the paste backfill plant at Louvicourt mine. CIM Bull. 96(1075), 51–57 (2003) Chen, Q., Zhang, Q., Qi, C., Fourie, A., Xiao, C.: Recycling phosphogypsum and construction demolition waste for cemented paste backfill and its environmental impact. J. Clean. Prod. 186, 418–429 (2018) Fall, M., Benzaazoua, M.: Modeling the effect of sulphate on strength development of paste backfill and binder mixture optimization. Cem. Concr. Res. 35(2), 301–314 (2005) Ghirian, A., Fall, M.: Coupled thermo-hydro-mechanical-chemical behavior of cemented paste backfill in column experiments: part II: mechanical, chemical and microstructural processes and characteristics. Eng. Geol. 170, 11–23 (2014) Hassani, F., Bois, D.: Economic and technical feasibility for backfill design in Quebec underground mines. Final report 1/2, Canada-Quebec Mineral Development Agreement, Research & Development in Quebec Mines. Contract no. EADM 1989–1992, File no. 71226002 (1992) Helinski, M., Fahey, M., Fourie, A.: Behavior of cemented paste backfill in two mine stopes: measurements and modeling. J. Geotech. Geoenviron. Eng. 137(2), 171–182 (2010) Kesimal, A., Ercikdi, B., Yilmaz, E.: The effect of desliming by sedimentation on paste backfill performance. Miner. Eng. 16(10), 1009–1011 (2003) Landriault, D., Verburg, R., Cincilla, W., Welch, D.: Paste technology for underground backfill and surface tailings disposal applications. Short course notes. In: CIM technical workshop. Vancouver, British Columbia, Canada, 27 Apr 1997

References

55

Qi, C., Fourie, A., Chen, Q., Zhang, Q.: A strength prediction model using artificial intelligence for recycling waste tailings as cemented paste backfill. J. Clean. Prod. 183, 566–578 (2018) Rankine, R.M., Sivakugan, N.: Geotechnical properties of cemented paste backfill from Cannington Mine, Australia. Geotech. Geol. Eng. 25(4), 383–393 (2007) Tariq, A., Yanful, E.K.: A review of binders used in cemented paste tailings for underground and surface disposal practices. J. Environ. Manage. 131, 138–149 (2013) Thomas, E.G., Nantel, J.H., Notely, K.R.: Fill technology in underground metalliferous mines, p. 293. Canada, International Academic Services Limited (1979) van der Spuy, B., Snyman, J., Steinmann, J.: Design and commissioning of the Cooke 2 shaft backfill reticulation system. In: Paste 2016–19th international seminar on paste and thickened tailings. Chile (2016) Wu, D., Fall, M., Cai, S.: Numerical modelling of thermally and hydraulically coupled processes in hydrating cemented tailings backfill columns. Int. J. Min. Reclam. Env. 28(3), 173–199 (2014) Wu, D., Zhang, Y., Liu, Y.: Mechanical performance and ultrasonic properties of cemented gangue backfill with admixture of fly ash. Ultrasonics 64, 89–96 (2016) Yilmaz, E., Benzaazoua, M., Belem, T., Bussière, B.: Effect of curing under pressure on compressive strength development of cemented paste backfill. Miner. Eng. 22(9), 772–785 (2009)

Chapter 6

Properties of Cemented Tailings Backfill

Abstract A mathematical model is developed to predict and assess the evolution of the rheological properties of CPB under the coupled effects of temperature and progress of binder hydration. The prediction ability of the model is verified, and the validation test results show that there is good agreement between the predicted and experimentally measured rheological properties of CPB. The flowability of fresh CPB through a loop pipe has also been discussed. Furthermore, two mathematical models are developed to describe the thermo-hydraulic coupled behavior and thermo-hydromechanical behavior of hydrating CPB, respectively. Data from field and laboratory studies are employed to validate the developed two models. Some applications of the validated models are also presented. Keywords Cemented paste backfill · Hydration · Rheology · Tailings · Coupled processes · THM · Model · Validation

6.1 Rheological Properties of Fresh CPB After the preparation of fresh CPB by mixing tailings, binders, and water in a backfilling plant which is usually located at the surface of the mine, the CPB forms a kind of slurry with relatively high density. The next key step is to place these fresh CPB mixtures into underground open stopes for the purpose of disposal. Hence, one of the key properties of fresh CPB is its transportability, which is related to its fluidity or flowability. The fresh CPB must show acceptable flowability to enable efficient pumping/delivery from the CPB plant to underground stopes. Therefore, the flowability or the transportability of fresh CPB is crucial for efficient and cost-effective mine backfill operations. Fresh CPB with poor flowability not only affects efficiency in pumping/delivery to stopes but can also result in pipe clogging which is associated with financial ramifications for the mine. For instance, if the fresh CPB is not flowable enough and pipe clogging occurs, the transportation system should be temporarily discontinued and the pipeline network disconnected (if necessary) to clear the pipes, which otherwise results in unfavorable consequences in which production progress is delayed and the operating costs of the mining companies are increased.

© Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_6

57

58

6 Properties of Cemented Tailings Backfill

Therefore, fluent transportability of fresh CPB should be ensured in order to avoid the occurrence of pipe clogging. It is well-known that characterizing the rheological properties and behavior of fresh CPB is the key step toward ensuring the good flowability of CPB. This stems from the fact that the transportation, pumping, and placement of CPB depend on its rheological properties or behavior. In addition to external factors such as temperature and loading pressure, the rheological behavior of transported fresh CPB also depends on internal elements (e.g., density and concentration of the CPB mixtures, characteristics of the CPB mixture constituents, pH). Therefore, there is the need to better understand the rheological behavior of fresh CPB under various thermal loading conditions and their changes with time. The main sources of heat that fresh CPB can be subjected to during its transport from the backfill plant to the underground excavation are schematized in Fig. 6.1. As shown in this figure, regardless which system is in use, that is, the gravity, pump/gravity or gravity/pump system, there is always friction between the inner sidewall of the pipe and the fresh CPB when it is transported through the pipeline. In this process, the kinetic energy of the fresh CPB converts to heat energy, that is to say, frictional heating occurs and contributes to the temperature development or heat generation within the fresh CPB (Fig. 6.1—➀). Furthermore, depending on the depth of the underground opening stope, the transportation distance can be sometimes extremely long. Therefore, the cement hydration process may start when the fresh CPB is transported through the pipeline during a relatively lengthy amount of time, which can release non-negligible amounts of heat and thus increase the temperature of the CPB during its transport (Fig. 6.1—➁). The exposed rock mass is a primary heat load source in any deep level mining operation. These hot rock temperatures will obviously increase the temperature of the transported CPB because of the thermal interactions between the rock and the CPB contained in the pipe (Fig. 6.1—➂). In

Fig. 6.1 Underground CPB transportation systems and thermal affecting factors (Wu et al. 2013)

6.1 Rheological Properties of Fresh CPB

59

addition, the geographical location of the mine is also a factor that can affect the temperature of open pit mines and mines situated at relatively shallow depths, and thereby influence the temperature of the transported CPB. The temperatures of these mines are strongly influenced by the climate of the region (Fig. 6.1—➃). These various temperatures, to which the CPB can be subjected during its transport, infer that one of the most challenging engineering tasks to ensure efficient and cost-effective transport of CPB is the understanding and prediction of the effects of temperature on the rheological behavior of CPB with time. Variations in yield stress and viscosity are directly associated with whether fresh CPB can be transported or how much load should be applied or energy (e.g. pumping) should be consumed to ensure flowability. Therefore, it is crucial to understand and model the rheology of fresh CPB under the coupled influence of temperature and time. This will contribute to a better optimization of CPB flow characteristics which could result in the improvement of CPB transport parameters, and thereby reduce the risk of pipe clogging.

6.1.1 Mathematical Modeling The modeling procedure is based on the mechanism of computational fluid dynamics (CFD), which can be applied in simulating fluid flow as well as the accompanied heat transfer process. The heat transfer in fluids and the general mathematical model can be presented in the following form (COMSOL 2014): ρc Cc

  ∂ Tc + ρc Cc u c · ∇T = ∇ · k T ∇T + Q ∂t

(6.1)

where ρ c , C c and k T are the density, specific heat capacity and thermal conductivity of the fresh CPB, respectively; uc is the velocity field (i.e., the volumetric flow rate), and Q is the heat source term. The degree of cement hydration is introduced to describe the fraction of reacted cement amount, and can be defined as follows (De Schutter 1999; Schindler and Folliard 2005): α(t) =

H (t) HT

(6.2)

where α(t) is the degree of cement hydration at the time t, H(t) is the accumulated heat released by cement hydration until time t, and H T is the total heat when all the cement reacts ultimately. In order to reveal the effect of time t on the degree of cement hydration, the following expression can be obtained (Schindler 2004; Schindler and Folliard 2005):

60

6 Properties of Cemented Tailings Backfill

    τT β α(t) = αf · exp − t

(6.3)

where τ T is the time parameter of cement hydration when the temperature of fresh CPB is T; β is the shape parameter of cement hydration; and α f is the ultimate degree of cement hydration, which can be calculated in the following form (Kjellsen et al. 1991): αf =

1.031 · w/c 0.194 + w/c

(6.4)

where w/c is water-to-cement ratio. Based on the calculated results of Eq. (6.4), if the ultimate degree of cement hydration is 1, w/c is 6.258. Therefore, when the w/c value is higher than 6.258, it is assumed that the αf value is 1, while if the w/c value is equal to or lower than 6.258, the equation is valid and thus the αf value can be calculated. Cement hydration process releases significant amounts of heat that can bring about the temperature development in fresh CPB, which can further affect its rheology and thus its flowability. The following expression is employed to illustrate this thermal progress, during which cement hydration, heat, temperature, and elapsed time are linked together (Schindler and Folliard 2005): Q ch = HCEM ·

  τ β  β · τ 1 1 E Tc Tc · − · α(t) · · t τTr · t R 273 + Tr 273 + Tc

(6.5)

where Qch is the heat generating rate of cement hydration, H CEM is the heat generated by cement hydration, T c is the temperature of fresh CPB, T r is the reference temperature, τ Tc and τ Tr are respectively the time parameters of cement hydration at temperature T c and T r , E is the apparent activation energy, and R is the universal gas constant. The following equation is to reveal the effect of the heat produced by cement hydration on the temperature evolution of fresh CPB versus time (Schindler 2004): ρc Cc

dTc = Q ch dt

(6.6)

The apparent activation energy E varies with the change of the CPB temperature, which is constant with the value of 33,500 J/mol when is T c equal to or higher than 20 °C and on the contrary when T c is lower than 20 °C (Kim et al. 2001): E(Tc ) = 33500 + 1470 · (20 − Tc )

(6.7)

Cement hydration includes all the chemical reactions of the compounds that comprise cement. Therefore, the term H CEM is the total heat that is produced by all the compounds (Schindler 2004):

6.1 Rheological Properties of Fresh CPB

61

HCEM = 500 pC3 S + 260 pC2 S + 866 pC3 A + 420 pC4 AF + 624 pSO3 + 1186 pFreeCaO + 850 pMgO

(6.8)

where pi is the weight ratio of i-th compound in terms of total cement content. In addition to the heat generated by cement hydration, the heat transfer process between the CPB and its surroundings can also contribute to its temperature evolution. For this reason, Fourier’s law is presented here to demonstrate this heat transfer process induced by temperature gradient between the fresh CPB and its surroundings (Nasir and Fall 2009): Q T = −k T · ∇T

(6.9)

where QT is the transferred heat flux, k T is the thermal conductivity of the fresh CPB, and ▽T is the temperature gradient. The thermal conductivity of the fresh CPB, which is not affected by the evolution of cement hydration and temperature, is considered to be constant, with the evidence experimentally proved by Celestin and Fall (2009). According to the conservation of momentum, the following expression can be applied to fresh CPB (Wu et al. 2013): ρc

du = ρc u c = ∇σ + ρc g dt

(6.10)

where σ is the stress tensor, and g is the gravitational acceleration. The fresh CPB flow performs non-Newtonian behaviors, and its fluid rheology can be described by Bingham model, which is a basic nonlinear rheological model and can be represented by the following equation (Cooke 2001): τs = τy + μB γ

(6.11)

where τ s , τ y , and μB are respectively the shear stress, yield stress and Bingham plastic viscosity of the fresh CPB flow, and γ is the shearing rate. The influence of temperature on the viscosity of fresh CPB can be expressed by the following equation (Wu et al. 2013): μc = μr · exp(E/RTc )

(6.12)

where μc and μr are the viscosities of fresh CPB at T c and T r , respectively. The following equation can be used to reveal the effect of cement hydration process on the viscosity of fresh CPB (Papo and Caufin 1991): μc = μc0 + (1000 − μc0 ) · (t/tv )n

(6.13)

where μc0 is the initial viscosity of the fresh CPB; n is a parameter indicating the kinetics of cement hydration progress; and t v is the moment at which the fresh CPB

62

6 Properties of Cemented Tailings Backfill

possesses a relatively high characteristic viscosity, which can be further calculated by the equation of water-to-cement ratio as follows (Papo and Caufin 1991): Y

tv = tv0 + X · w/c − (w/c)0

(6.14)

where t v0 is the initial value of t v ; (w/c)0 is the initial value of w/c; and X and Y are the experimentally determined parameters. Not only will the viscosity of the CPB change, but also its yield stress due to the coupled effect of temperature and binder hydration (time). To predict this coupled effect on the shear yield stress of CPB, the following equation can be used (Petit et al. 2006): τy (t) = τy0 (T ) + θ · exp(ε/T ) · t

(6.15)

where τ y0 (T ) is the initial yield stress of fresh CTB slurry at a given temperature T; θ and ε are experimental constants depending on the mixing proportion of fresh CPB. By combing the above equations, the coupled model can be achieved. The model will be implemented into COMSOL Multiphysics to predict the coupled influence of cement hydration and temperature variation on the evolution of the CPB rheological properties.

6.1.2 Model Validation and Simulation To validate the results of the developed model, various sets of laboratory experimental results were used: (i)

a set of available experimental data from rheometer tests performed on various types of CPB samples cured at various times; (ii) another set of experimental data obtained from vane shear tests performed on CPB samples cured at different times and temperatures; (iii) a set of data that resulted from slump tests conducted on CPB of different ages and cured at different temperatures. (1) Validation against rheometer test results Silicate cement 325# was selected for preparing the fresh CPB samples. The particle size distribution of the tailings used is shown in Fig. 6.2. The laboratory rheological test was conducted by utilizing Brookfield R/S plus Rheometer, and Table 6.1 illustrates the physical characteristics of the three sets of CPB slurry samples. As an example, Fig. 6.3 shows the rheometer test results of 1# CPB sample. Table 6.2 summarizes the rheometer testing results of all the fresh CPB slurry samples. The obtained test results of 1# CPB sample were compared with the outcomes predicted by the developed numerical model. Figure 6.4 displays the comparison of

6.1 Rheological Properties of Fresh CPB

63

Fig. 6.2 Particle size distribution of the tailings used

Table 6.1 The general physical characteristics of the fresh CPB samples

Samples Tailings density

1# (t/m3 )

2#

3#

2.63

2.63

2.63

Solid concentration by mass (%)

66

68

70

Initial temperature (°C)

21.5

21.5

21.5

Cement-to-tailings ratio

1/8

1/8

1/8

Water to cement ratio

4.6

4.2

3.9

Slump (cm)

20

18

16

the predicted viscosity evolution of the fresh CPB with the experimental data, and Fig. 6.5 shows the contrast between the predicted and tested evolution of the shear stress of the fresh CPB. Figures 6.4 and 6.5 show well agreement between the predicted and experimental results. It can be concluded that the validity and applicability of the developed numerical model is testified. (2) Validation against vane shear test results In order to reveal the coupled effect of temperature and binder hydration (i.e., curing time) on the rheological properties (yield stress) of fresh CPB flow, laboratory vane shear tests were conducted on CPB samples cured at various temperatures (3, 20 and 35 °C) and curing times (0.25, 1, and 2 h) by Fall and Ghirian (2012). The mix components of the CPB are given in Table 6.3. The experimental results were compared with the simulation outcomes, as shown in Figs. 6.6 and 6.7. It can be noticed that the experimental and simulated results agree well with each other. The yield stress of fresh CPB increases with increases in temperature and advance in curing time.

64

6 Properties of Cemented Tailings Backfill

(a) Evolution of the fresh CPB’s viscosity versus time

(b) Bingham regression analysis of the fresh CPB’s shear stress versus shear rate Fig. 6.3 Rheometer test results of 1# CPB sample Table 6.2 Summary of the rheometer testing results

Samples

Solids content (%)

Bingham plastic viscosity (Pa s)

1#

66

0.79

Yield stress (Pa) 393.20

2#

68

1.45

759.61

3#

70

1.83

1363.15

6.1 Rheological Properties of Fresh CPB

Fig. 6.4 Comparison of the predicted and tested results of the viscosity

Fig. 6.5 Comparison of the predicted and tested results of the shear stress

65

66 Table 6.3 Main parameter and property inputs used for the model

6 Properties of Cemented Tailings Backfill Parameter

Value

Binder content (%)

4.5

w/c

7.6

Tailings particle size

D60 = 31.50 μm; D50 = 22.50 μm; D10 = 1.9 μm

Backfill thermal conductivity (W/(m k))

Varies

Initial temperature

Variable (3, 20, 35 °C)

Initial viscosity (Pa s)

Varies

Initial yield stress (Pa)

Varies

R (gas constant) [J/(mol K)]

8.314

Fig. 6.6 Yield stress versus temperature of fresh CPB flow for various curing times

Fig. 6.7 Yield stress versus curing time of fresh CPB flow for various temperatures

6.1 Rheological Properties of Fresh CPB

67

(3) Validation against slump test results One of the most commonly utilized field tests to judge the ease of transport and placement of fresh CPB is the slump test. The data from experimental investigations conducted by Fall and Ghirian (2012) that studied the effects of temperature and binder hydration on the slump and spreading of CPB are used here for validation purposes. The characteristics of the CPB mixtures, on which the slump tests were performed are given in Table 6.3. Figure 6.8 graphically shows the contrast between the experimental and simulated results of slump versus spreading of fresh CPB flow at different temperatures. From this figure, it can be seen that these two sets of outcomes agree well with each other, and both the simulation and laboratory results reveal that the slump and spreading of fresh CPB flow decrease with increases in temperature. This is because higher temperatures speed up the generation of cement hydration products which make the CPB more viscous. Figure 6.9 demonstrates a comparison of the experiment and simulation of slump versus temperature of fresh CPB flow at different curing times. It is found that the simulated outcomes agree well with the experimental ones, and both of the results reveal that the slump of fresh CPB decreases with increases in temperature and curing time. The good agreement between the obtained numerical results and experimental values over a wide range of temperatures and curing times (Figs. 6.8 and 6.9) confirms the validity of the developed numerical model and allows us to use the numerical model to simulate the slump cone test on aging CPB under various thermal loading conditions. Figure 6.10 demonstrates an example of the simulation results of fresh CPB flow shape development versus time after removing the cone. In this simulation process, the CPB temperature is 20 °C, and the time for the flow to spread is recorded

Fig. 6.8 Slump versus spreading of fresh CPB flow for various temperatures

68

6 Properties of Cemented Tailings Backfill

Fig. 6.9 Slump versus temperature of fresh CPB flow for various curing times

Fig. 6.10 Simulation results of slump test for fresh CPB

6.1 Rheological Properties of Fresh CPB

69

Fig. 6.11 Effect of temperature on the slump flow of fresh CPB

until 3 s with a time interval of 1 s. Furthermore, on the condition of keeping the same elapsed time (1 s), a numerical simulation was conducted to reveal the effect of different temperatures on the fresh CPB flow, and the results are graphically illustrated in Fig. 6.11. This figure reveals a difference in fluidity between the CPB samples cured at different temperatures. It is evident from these results that the fluidity of the flow of CPB at lower temperatures is significantly higher than that with higher temperatures.

6.2 Flowability of Fresh CPB In this section, the flowability of fresh CPB through a loop pipe will be described and discussed. The pipe flow model developed by Wu et al. (2005) is used for the numerical study. The simulation properties of the CPB slurry samples used are shown in Table 6.4. Figure 6.12 shows the geometric model of the loop pipe used, which is built in the backfill laboratory of China University of Mining and Technology, Beijing. The entire length of the loop pipe is 42 m, with an inner diameter of 80 mm. The capability of the stirring chest (the box in Fig. 6.12) is 0.3 m3 . Two pressure tapings (the two

70

6 Properties of Cemented Tailings Backfill

Table 6.4 Main simulation properties of the CPB slurry samples used Solids content (%)

Cement-totailings ratio (c/t)

Density (g cm−3 )

74

1:6

1.927

318.754

0.74334

1:8

1.930

266.043

0.72819

75

76

77

78

Rheological properties Bingham plastic viscosity (Pa s)

Yield stress (Pa)

1:10

1.929

230.663

0.66801

1:6

1.951

406.793

0.89757

1:8

1.949

388.725

0.88683

1:10

1.951

309.345

0.73887

1:6

1.976

674.824

1.29244

1:8

1.974

601.208

1.26577

1:10

1.976

597.381

1.18191

1:6

2.002

893.757

1.27768

1:8

2.000

854.219

1.24675

1:10

2.002

756.071

1.23131

1:6

2.028

1426.723

1.75711

1:8

2.026

1397.652

1.57218

1:10

2.028

1371.921

1.13533

Fig. 6.12 Geometric model of the loop pipe used

6.2 Flowability of Fresh CPB

71

(a) c/t=1:6

(b) c/t=1:8

(c) c/t=1:10 Fig. 6.13 Pressure distribution of the fresh CPB with 74% solids content through the loop pipe

blue points in Fig. 6.12) are fixed in the pipe, with a distance (between them) of 10 m. The inlet flow is 40 m3 /h. The fresh CPB samples with solid contents of 74 and 78% are taken as two examples. Figures 6.13 and 6.14 illustrate the pressure distributions of the fresh CPBs with 74 and 78% solids contents through the loop pipe, respectively. The pressure drop is applied to evaluate the flowability of fresh CPB through pipeline. The pressure drops between the two tapings in the loop pipe for all of the CPB samples are summarized in Table 6.5, according to the model simulation results. It can be seen from this table that, the pressure drop value increases with the increases of both c/t and solids content.

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB The underground disposal of tailings in mine voids supported by the CPB technology requires the filled CPB showing acceptable environmental properties, such as hydraulic conductivity. Consequently, susceptibility to AMD (acid mine drainage) and ability to release contaminants into the mine areas and/or groundwater are relevant aspects that need to be addressed as part of the environmental design criteria for CPB (Kesimal et al. 2003; Yilmaz et al. 2004).

72

6 Properties of Cemented Tailings Backfill

(a) c/t=1:6

(b) c/t=1:8

(c) c/t=1:10 Fig. 6.14 Pressure distribution of the fresh CPB with 78% solids content through the loop pipe Table 6.5 Pressure drops between the two tapings in the loop pipe for all of the CPB samples

Solids content (%)

c/t

Pressure drop (kPa/m)

74

1:10

18.26

1:8

20.65

75

76

77

78

1:6

23.47

1:10

22.97

1:8

28.21

1:6

29.44

1:10

41.86

1:8

55.19

1:6

57.68

1:10

50.23

1:8

55.19

1:6

57.68

1:10

78.54

1:8

85.05

1:6

88.57

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

73

It is well-known that thermal factors can exert a significant effect on the permeability or hydraulic conductivity of a CPB. The thermal factors mainly include the temperature exchange between the surrounding environment and CPB, as well as the heat generated by hydration of binder using to prepare CPB. Hence, the understanding and modeling of thermal processes (heat generation and evolution, thermal properties), development of suction and hydraulic conductivity (hydraulic processes) as well as the interactions between these thermal and hydraulic processes within hydrating CPB are crucial for the design of cost-effective and environmentally friendly CPB structures (Wu et al. 2014).

6.3.1 Model Development The general mathematical model for heat transfer in porous media (COMSOL 2014) is presented in the following equation: (ρC)eq

  ∂T + ρF CF u F · ∇T = ∇ · keq ∇T + Q ∂t

(6.16)

where ρ F is the fluid density; C F is the fluid heat capacity at constant pressure; (ρC)eq is the equivalent volumetric heat capacity of the solid-fluid system at constant pressure; k eq is the equivalent thermal conductivity of the solid-fluid system; uF is velocity field of the fluid; and Q is the heat source. For CPB materials, the degree of binder hydration can be mathematically expressed in the following form (De Schutter 1999; Schindler and Folliard 2005): α(t) =

H (t) HT

(6.17)

where α(t) is the degree of cement hydration at the time t, H(t) is the accumulated heat released by cement hydration until time t, and H T is the total heat when all of the binders react ultimately. The exponential function expressed as follows can characterize the development of the degree of binder hydration (Schindler 2004; Schindler and Folliard 2005):  τ β α(te ) = αu · exp − te

(6.18)

where α(t e ) is the degree of binder hydration at an equivalent age t e , τ is the hydration time parameter at the reference temperature, β is the hydration shape parameter, and α u is the ultimate degree of binder hydration. The equivalent age can be written in the form (Schindler 2004):

74

6 Properties of Cemented Tailings Backfill

te =

τ ·t τT

(6.19)

where τ T is the hydration time parameter at the temperature of materials, and t is the chronological age. By combining Eqs. (6.18) and (6.19):     τT β α(t) = αu · exp − t

(6.20)

α u can be calculated in the following form (Kjellsen et al. 1991): αu =

1.031 · w/c 0.194 + w/c

(6.21)

where w/c is the water-binder ratio. In CPB systems, full hydration of the binder, i.e., an ultimate degree of binder hydration of 1, can be expected, since high w/c ratios (5–15) are commonly applied to prepare CPB. Thus, in Eq. (6.21), when w/c < 6.258, the equation is valid; if w/c ≥ 6.258, it can be assumed that α u = 1. The following equation can be used to determine the temperature development in CPB curing under adiabatic conditions (Jonasson et al. 1995): dH dT QH = = dt ρp · Cp dt



1 ρp · Cp

(6.22)

where T is the temperature of CPB (°C), ρ p is the density of CPB (kg/m3 ), C p is the specific heat capacity of CPB (J/kg/°C), QH is the rate of heat generation (W/m3 ), and H is the heat of binder hydration (J/m3 ). By using the equivalent age maturity method, the following formula can be used to describe the coupled effect of temperature and time on the rate of heat generation by binder hydration (Schindler and Folliard 2005):   τ β  β · τ E 1 1 T T · α(t) · · · − Q H (t) = HT · t τ ·t R 273 + Tr 273 + Tc

(6.23)

where T r is the reference temperature, T c is the temperature of the materials, E is the apparent activation energy, and R is the universal gas constant [8.314 J/(mol K)]. The apparent activation energy that characterizes the binder hydration process can be proposed in the following equation (D’Aloia and Chanvillard 2002; Kim et al. 2001): E(Tc ) = 33, 500 + 1470 · (20 − Tc ) Tc < 20 ◦ C

(6.24)

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

75

E(T c ) is constant with the value of 33,500 J/mol when is T c equal to or higher than 20 °C. When the mineral admixtures slag and fly ash (FA) are used, the total heat H T can be calculated by (Maekawa et al. 1999; Maekawa and Ishida 2002): HT = (HCEM · pCEM + 209 · pFA + 461 · pSLAG ) · Cc

(6.25)

where pCEM is the cement mass ratio of the total cement content, pFA is the FA mass ratio, and pSLAG is the slag mass ratio; C c is the cementitious material (cement, FA and slag) content (namely, binder content) (g/m3 ), and H CEM is the total heat of cement hydration (J/g), which can be calculated by Eq. (6.8). The fluid mass flow rate for cement-based materials (like CPB), can be calculated by the following equation (Mainguy et al. 2001; Poyet et al. 2011; Richards 1931): m F = ρF · u F = −ρF K

kr grad(P) μF

(6.26)

where mF is the fluid mass flow rate, K is the intrinsic permeability of the porous media, μF is the dynamic viscosity of the fluid, k r is the relative permeability to the fluid, and P is the fluid pressure. It should be pointed out that the fluid consists of water and air, and thus, Darcy’s law is used for water and air (Bourgeois et al. 2002): m W = ρW · u W = −ρW K m A = ρA · u A = −ρA K

krW grad(PW ) μW

(6.27)

krA grad(PA ) μA

(6.28)

where ρ W and ρ A are the densities of water and air, μW and μA are the dynamic viscosities of water and air, k rW and k rA are the relative permeabilities to water and air, and PW and PA are the water pressure and air pressure, respectively. Since grad(P) is the fluid pressure gradient caused by capillary absorption (or suction), the following equation can be obtained (Fall and Samb 2007; Krus et al. 1997; Talor et al. 1999): grad(PW ) = grad(PA ) = grad(P) = ∇ pc

(6.29)

where pc is the capillary pressure. The relative permeabilities to water and air are respectively represented in the forms (Luckner et al. 1989; Mualem 1976; van Genuchten 1980): krW (Seff ) =



   X 2 1/ X Seff 1 − 1 − Seff

(6.30)

76

6 Properties of Cemented Tailings Backfill

krA (Seff ) =



 2X 1/ X 1 − Seff 1 − Seff

(6.31)

where X is the material parameter to be evaluated on the basis of experimental results, S eff is the effective saturation degree, defined as follows (van Genuchten 1980): Seff =

θ − θr θs − θr

(6.32)

where θ is the volumetric water content, and θ s and θ r are the saturated and residual water content, respectively. On the basis of validated experimental results, Abdul-Hussian and Fall (2011) obtained the following exponential function for predicting the evolution of residual water content with the degree of binder hydration: θr = je−hα

(6.33)

where α is the degree of binder hydration, and j and h are the adimensional parameters of the cement-based materials to be determined. During the hydration process, only the capillary water is consumed by selfdesiccation (or hydration) and contributes to further hydration. In the present study, it should be pointed out that capillary water loss due to phase transition (e.g., liquid water to water vapor) is ignored. Furthermore, Abdul-Hussian and Fall (2011) provided the following equations: θs − θr θ = θr +

X 1 + (i pc )Y

(6.34)

i = ae−bα

(6.35)

X = f α −1

(6.36)

where, a, b and f are the adimensional parameters that depend on the cemented porous media, α is the binder hydration degree (given in percentage) and Y =

1 1− X

(6.37)

The integration of Eqs (6.32) and (6.34) yields:  Seff = which further leads to:

1 1 + (i pc )Y

X (6.38)

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

pc =

 Y1 1  −1/ X Seff − 1 i

77

(6.39)

The volumetric water content is defined as: θ = φF · Seff

(6.40)

where φ F is the porosity. The evolution of the CPB porosity versus degree of binder hydration is revealed by the following equation (Abdul-Hussian and Fall 2011): φ(α) = Gα + φ0

(6.41)

where φ(α) is the porosity of the CPB when it reaches the degree of hydration α, G is an empirical material parameter to be experimentally determined, and φ 0 is the initial (or early age) porosity of the CPB or the porosity of the CPB when it starts to harden. With the combination of the Eqs. (6.33), (6.34), (6.38), (6.40) and (6.41), the effective saturation degree can be expressed in the following form: Seff =

je−hα Gα + φ0 + je−hα − θs

(6.42)

The dynamic viscosity of the fluid (water and air) is strongly temperaturedependent. The following approximate formula is used to calculate the dynamic viscosity of water (Thomas and Sansom 1995): μW = 0.6612(T − 229)−1.562

(6.43)

In a similar way, a simple temperature related function of the dynamic viscosity of air is presented (Alnajim 2004): μA = 3.85 × 10−8 T

(6.44)

The time-dependent evolution of the intrinsic permeability of the CPB can be described by Fall et al. (2009): K = K T · A · (UCSt /UCSmax )B

(6.45)

where K T is the permeability of the tailings being used, UCSt is the UCS of the CPB for a given time t, UCSmax is the maximum UCS of the CPB, and A and B are adimensional fitting parameters to be determined for each CPB mix. In order to approximate the degree of binder hydration, Abdul-Hussian and Fall (2011) proposed the following relation between UCSt and UCSmax :

78

6 Properties of Cemented Tailings Backfill

α(t) =

UCSt UCSmax

(6.46)

where α(t) is the degree of binder hydration at curing time t. Then, the intrinsic permeability can be rewritten as: K = K T · A · [α(t)] B

(6.47)

For the sake of simplicity, air is considered to be an ideal gas (Baggio et al. 1997; Bary and Sellier 2004): ρA =

MA pA RA T

(6.48)

with M A as the molar mass of air, RA the universal gas constant of air, and pA the absolute pressure of air. The temperature-dependent evolution of water density is expressed as follows (Tong et al. 2010):   pW − 101, 325 −1 ρW = ρT 1 − KW

(6.49)

where K W (= 2.15 × 109 Pa) is the bulk modulus of water, pW is the water pressure (Pa), and ρT = 1000.066219 + 0.0209229T − 0.00602137T 2 + 0.0000163T 3

(6.50)

which is the density at atmospheric pressure, with T as the temperature (°C). Equation (6.16) can further be expressed in the following form as CPB is a type of porous medium and cementitious material: (ρC)eq

  ∂T + ρF CF u F · ∇T + ∇ · −keq ∇T = Q H ∂t

(6.51)

where QH is the heat from binder hydration. It should be stated that heat production by binder hydration is an accumulated process with the development of the height of the backfill, which may take a long time in cases of large backfill volume (Nasir and Fall 2009). Therefore, the backfilling rate will be introduced to describe the dynamic process of heat generation: Q H = Q H (t) · v · t · As

(6.52)

where QH (t) is the heat generation rate (W/m3 ), As is the cross-section area of the stope (m2 ), and v is the backfilling rate (m/s), defined by:

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

v=

H t

79

(6.53)

where H is the height of the backfill (m). The third term of Eq. (6.51) reveals the heat conduction between the surrounding rock and CPB, which is derived from Fourier’s law of heat conduction (Nasir and Fall 2009): q = −keq ∇T

(6.54)

where q is the heat flux vector; and k eq is related to the thermal conductivities of the solid (k p ) and the fluid (k F ) by: keq = φp kp + φF kF

(6.55)

The equivalent volumetric heat capacity of the CPB is calculated as: (ρC)eq = φp ρp Cp + φF ρF CF

(6.56)

where ρ p is the solid density, C p is the solid heat capacity at constant pressure, and φ p denotes the solid’s volume fraction, which is related to the volume fraction of the fluid φ F (or equivalent to the porosity) by: φ p + φF = 1

(6.57)

It should be noted that because of the low permeabilities of the rock and CPB, and the assumption that there is no significant air flow in the stope voids, heat transfer through convection (or advection) has been disregarded in the above equation (Mainguy et al. 2001; Nasir and Fall 2009; Poyet et al. 2011). As mentioned above, as the fluid includes water and air, Eqs. (6.55)–(6.57) should then be rewritten as follows, respectively: keq = φp kp + φW kW + φA kA

(6.58)

(ρC)eq = φp ρp Cp + φW ρW CW + φA ρA CA

(6.59)

φ p + φW + φ A = 1

(6.60)

where φ W and φ A are, respectively, the volume fractions of water and air, k W and k A are respectively the thermal conductivities of water and air, and C W and C A are the heat capacities of water and air at constant pressure. φ W and φ A can be respectively calculated from the following equations: φW = φF Seff

(6.61)

80

6 Properties of Cemented Tailings Backfill

φA = φF (1 − Seff )

(6.62)

The initial (or early age) porosity of the CPB can be calculated by (Krus et al. 1997):  ρ0 · 100% = 1 − φ0 ρt

(6.63)

where ρ 0 is the initial bulk density of the CPB, and ρ t is the true density of the CPB. It should be pointed out that the bulk density of the CPB is not constant, which changes as hydration progresses. Therefore, with the help of Eqs. (6.41) and (6.63), there is: ρp = ρt · (1 − φF ) = ρt · [1 − (Gα + φ0 )]

(6.64)

For porous media, there is a function for calculating the equivalent thermal conductivity of the solid-fluid system in the form of (Somerton et al. 1973; Wang and Beckermann 1993):    (6.65) keq = kpv + Seff kpl − kpv where k pv is the thermal conductivity of the porous media when the saturation degree is zero while k pl corresponds to the case where the porous media is fully saturated with water. By making use of all corresponding equations above, the final thermo-hydrochemical coupled model can be presented in the following form:   MA 2 pA C A [1 − (Gα + φ0 )] ρt Cp + (Gα + φ0 )Seff ρW CW + (Gα + φ0 )(1 − Seff ) RA T ⎧    X 2 √ ⎪ 1/ X ⎪ ⎪ 1 − 1 − S S eff eff ∂T ⎨ − ρW CW ⎪ ∂t 0.6612(T − 229)−1.562 ⎪ ⎪ ⎩ ⎫    X 2 √ ⎪ 1/ X ⎪ ⎪ Seff 1 − 1 − Seff ⎬ MA + · p C A A −8 2 ⎪ 3.85 × 10 RA T ⎪ ⎪ ⎭

   K T A(α) B · ∇ pc · ∇T + ∇ · −keq ∇T   τ β  βτ E 1 1 T T v · t · As α(t) − = HT t τt R 273 + Tr 273 + Tc

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

81

The mathematical model couples the thermal, hydraulic and chemical (THC) processes. The THC model is implemented into COMSOL, in which The Darcy’s Law Interface under the Porous Media and Subsurface Flow branch and The Heat Transfer in Porous Media Interface are coupled together to simulate thermo-hydraulic coupled processes in hydrating CPB.

6.3.2 Model Validation 6.3.2.1

Validation of the Thermal Prediction Ability

For test and verification of the validity of the developed model, the model simulating results were validated against the results obtained by field and laboratory investigations. Table 6.6 shows the main input parameters utilized to validate the proposed THC model. It should be pointed out that the binder type for the field study and Laboratory Study 1 is only Portland cement type I (PCI), while Laboratory Study 2 utilized PCI/FA (50/50) (i.e., the mass ratio of PCI to FA is 50–50%) and PCI/SLAG (50/50) (i.e., the mass ratio of PCI to slag is 50–50%). The physical and chemical properties of PCI are shown in Table 6.7 (Abdul-Hussian and Fall 2011). Table 6.6 Main parameters used for simulation to validate the THC model Parameters

Field study

Lab. Study 1

Lab. Study 2

Initial CPB temperature T 0 (°C)

26.0

24.5

22.0

Initial rock temperature T 1 (°C)

32.2

24.5

22.0

Initial air temperature T 2 (°C)

Varies

–

22.0

Equivalent thermal conductivity [W/(m K)]

Varies

Varies

Varies

Backfill specific heat [J/(kg K)]

950

1250

1000

Rock thermal conductivity [W/(m K)]

2.67

–

–

Rock specific heat [J/(kg K)]

1000

1000

1000

Cement activation energy (J/mol)

Varies

Varies

Varies

Binder content (kg/m3 )

239

146

200

R (gas constant) [J/(mol K)]

8.314

8.314

8.314

Table 6.7 Characteristics of Portland cement type I Blaine (m2 /g)

Chemical composition (mass%) PCI

MgO

CaO

SiO2

Al2 O3

Fe2 O3

SO3

2.65

62.82

18.03

4.53

2.70

3.82

1.30

82

6 Properties of Cemented Tailings Backfill

(1) Validation against field results Williams et al. (2001) conducted field research on cemented backfill by observing stress and strain changes in the backfill in order to evaluate if temperature compensation was required for the instrumented strain gauges. Temperature evolution with time caused by the cement hydration heat was investigated. For purposes of comparison with the field results and to validate the simulated model, temperature evolution was calculated with the developed model, and the temperature development at the midpoint of the CPB was recorded. The geometry of the simulated model is revealed in Fig. 6.15 and some of the necessary input parameters can be found in Table 6.6. Figure 6.16 shows a comparison of the temperature history between the present study and the field data. As shown in Fig. 6.16, there is a good agreement between Fig. 6.15 Geometry of the simulated model (against field study)

Fig. 6.16 Comparison between the results of simulation and field study

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

83

the temperature predicted by the developed THC model and the field study results. Although the time to reach the temperature peak is different, the maximum temperature is almost the same. It is deduced that the differences in temperature results from the complicated and varied boundary conditions in the field. (2) Validation against Lab. Study 1 A laboratory CPB column study carried out by De Souza and Hewitt (2005) was used to validate the proposed model in the current study. Figure 6.17 demonstrates the geometry of the simulated model in COMSOL, with some necessary input parameters which are also shown in Table 6.6. For the purpose of validation, the temperature development in the middle of the CPB was investigated and Fig. 6.18 presents a comparison between the results obtained by simulation and the laboratory study. It can be seen from Fig. 6.18 that there is a good agreement between the THC model simulations and the laboratory results. Nevertheless, the differences in the temperature evolution may be attributed to the varying boundary conditions in the laboratory. (3) Validation against Lab. Study 2 The results from another laboratory CPB small column study conducted at University of Ottawa were used to validate the developed THC model. The CPB column (as shown in Fig. 6.19) is 50 cm high and the height of the CPB is 40 cm, with the thickness of the top freeboard at 5 cm and the bottom cap at 5 cm. A temperature sensor was located exactly right in the middle of the CPB to investigate the temperature evolution versus time within the CPB structure. Figure 6.20 illustrates a comparison between the results obtained from simulation and the laboratory investigation. It can be observed that the temperature development predicted by the simulated THC model agrees well with that investigated in the laboratory. Fig. 6.17 Geometry of the simulated model (against Lab. Study 1)

84

6 Properties of Cemented Tailings Backfill

Fig. 6.18 Comparison between the results of model simulation and Lab. Study 1 Fig. 6.19 Geometry of the simulated model (Lab. Study 2)

Fig. 6.20 Comparison between the results of model simulation and Lab. Study 2

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

6.3.2.2

85

Validation of the Hydraulic Prediction Ability

Table 6.8 shows the main input parameters of the model used for simulation. (1) Validation against Experimental Study Abdul-Hussian and Fall (2012) conducted an experimental study on the thermohydro-mechanical behavior of CTB columns in drained and undrained (UD) conditions. The columns were filled up to a height of 150 cm in three filling sequences at 50 cm for each sequence. There was a rest period of 24 h between each sequence. The suction and temperature developments as well as the water drainage were monitored during the curing process, while the saturated hydraulic conductivity and thermal conductivity of the CPB at various depths were tested after the completion of curing. The PCI was used in the column experiments. To test the performance of the developed model, the suction developments as well as the saturated hydraulic conductivity at the middle point of each layer (i.e., at depths of 25, 75 and 125 cm) were simulated. Moreover, the water drained was computed and its evolution with time was simulated. Figure 6.21 compares the saturated hydraulic conductivity of a CPB between the model simulations and the laboratory tests. It is seen that regardless of the drainage conditions, there are no significant differences between the hydraulic conductivity values (after 28 days of curing) obtained by the numerical simulations and experimental investigations. The reason for the slightly lower value of saturated hydraulic conductivity in the bottom layer of the undrained column can be attributed to the impact of cement hydration heat on the refinement of the pore structure of the cementitious material. Figure 6.22 reveals a validation of the current modeling study on the suction development within the columns under drained and undrained conditions against Table 6.8 Main parameters used in the THC model for validation purpose Parameters

Experimental study

Field test

Initial CPB temperature T 0 (°C)

20

23

Curing temperature (°C)

20

23

Initial air temperature (°C)

20

23

Equivalent thermal conductivity [W/(m * K)]

Variable

Variable

Apparent activation energy (J/mol)

Variable

Variable

2100

2180

Initial bulk density

(kg/m3 )

Dry density (kg/m3 )

1600

1652

Binder content (wt%)

4.5

4.5

Solid mass concentration (wt%)

73

75.8

Water-to-binder ratio

8.2

7.4

Initial effective saturation degree (%)

100

100

Initial porosity

0.5

0.53

86

6 Properties of Cemented Tailings Backfill

Fig. 6.21 Comparison between simulations and experiments on saturated hydraulic conductivity after 28 days of curing

(a) Drained condition

(b) Undrained condition

Fig. 6.22 Comparison between simulations and experiments on matrix suction (Wu et al. 2014)

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

87

the experimental investigations. As shown in Fig. 6.22, there is a good agreement between the suction predicted by the developed THC model and the experimental data; it is also interesting to see that the THC model can catch the effect of the filling sequences on the evolution of the suction within the CPB column very well. It can be observed that both the simulation and laboratory results reveal that the matrix suction values of the three layers under drained and undrained conditions increase from 0 kPa to values in the range of 60–70 kPa. The reason for these observations can be explained as follows: when the hydration of the CPB progresses, the binder hydration degree increases, which results in a decrease of the saturation degree and thus an increase in the matrix suction. It is well-known that self-desiccation occurs within a CPB during binder hydration, which consumes water and thereby reduces the effective saturation degree. That is why with the elapse of curing time, the suction increases for the three layers on both drained and undrained conditions until the saturation degree reaches its ultimate value. Regardless of the drainage conditions, the sharp decrease of suction in the bottom layer on the first and second days is due to the fact that the addition of the middle and the top layers provide fresh water to the bottom layer. A similar explanation can be applied to the sharp decrease of suction in the middle layer on the second day. It can be observed that after 24 h, but before pouring the middle layer, the matrix suction value of the drained column is higher than that of the undrained column. In addition, after about 5 days, the suction in the bottom layer reaches a constant value, and this ultimate value of the drained column is obviously higher than that of the UD column. These are due to the fact that, in contrast with the undrained column, the bottom layer of the drained column loses more water, resulting in a lower degree of saturation and thus a higher value in the matrix suction. It can also be found that under the undrained condition, the ultimate suction value of the middle layer is obviously higher than that of the bottom layer. This can be attributed to the fact that a certain amount of water in the middle layer flows to the bottom layer, which decreases the saturation degree of the middle layer and meanwhile, increases the saturation degree of the bottom layer. However, for the drained condition, the bottom layer can lose more water than the middle layer, which offsets the effect of the water from the middle layer on the difference in saturation degree in the two layers. That is why the ultimate suction value of the middle layer is almost the same as that of the bottom layer in the drained column. According to the comparisons of these figures, it can be seen that both the THC model and the experimental outcomes indicate that the top layers of the drained and undrained columns achieve the highest suction value among the three layers. This is because the top of the columns is uncovered, and in addition to self-desiccation, evaporation can cause a significant decrease of the saturation degree of the top layer and thereby result in an increase of its matrix suction. Figure 6.23 illustrates a comparison between the experimentally obtained evolution of the percentage of water drained and that which resulted from THC simulation. As shown in this figure, the percentage of water drainage predicted by the developed THC model agrees well with that measured in the laboratory. It can be observed that after about 50 min, the water drainage content will not increase. This is due to the fact

88

6 Properties of Cemented Tailings Backfill

Fig. 6.23 Comparison between simulations and experiments on drained water content

that with the accumulation of binder hydration products, the pore structures within the CPB will be refined until 50 min have passed, and the pores are fine enough so that no more water will drain from them. (2) Validation against Field Test A field test was carried out in underground mine environments to reveal the influence of self-weight consolidation settlement of paste backfill on the evolution of its physical and mechanical characteristics, under the undrained (UD), half-drained (HD) and fully drained (FD) conditions (Belem et al. 2006). The three columns were filled up to a height of 300 cm in two filling sequences of 150 cm over 24 h, with a section area of 31.5 × 30.5 cm2 . The water drainage of the paste backfill with time was investigated. PCI/SLAG (20/80, i.e., the mass ratio of PCI to slag is 20–80%) was selected as the binder to conduct the simulation. Figure 6.24 presents a comparison of the proportion of water drained from the HD and FD columns between the simulations and the field investigations.

Fig. 6.24 Comparison between simulations and field tests on drained water content

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

89

The results from the comparison show that the amount of drained water calculated by the developed THC model agrees well with that acquired by the filed measurements. As expected, the amount of water drained from the FD column is twice that from the HD column. It can also be observed that regardless of the drainage condition, no water drainage occurs after about 72 h. This is due to the fact that with the development of binder hydration, the generated hydration products can refine the pore structures of the CPB until the pores are fine enough to prevent any more water drainage beyond 72 h.

6.3.3 Model Application 6.3.3.1

Application of the Thermal Prediction Ability

Since the ability of the THC model to predict the temperature development in CPB has been validated, it could be used to simulate the thermal response in CPB under the effect of various factors in practice. The simulated model is illustrated in Fig. 6.25, where T 0 is the initial CPB temperature, T 1 is the initial rock temperature, T 2 is the initial air temperature, W is the width of the backfilling stope, and H is the height of the stope. Table 6.9 shows the main input parameters utilized for the application of the proposed THC model, and the main simulation results will be shown in the following sections. It should be noted that three types of binders were selected to conduct the study: PCI/FA (50/50), PCI/SLAG (50/50) and PCI (100%) (i.e., the mass ratio of PCI is 100% in the binder). (1) Effect of stope geometry (1.1) Stope size As CPB is placed into a backfilled stope, which could have different sizes, it is, therefore, important to simulate the effect of stope size on the temperature development of CPB. According to field conditions, a stope can be different and ranges from small to large. Four stope sizes (H × W ), including 5 m × 2.5 m, 10 m × 5 m, 20 m × Fig. 6.25 Geometry of the simulated model for application to practical situations

90 Table 6.9 Main parameters applied in the THC model simulations

6 Properties of Cemented Tailings Backfill Parameters

Values

Initial CPB temperature T 0 (°C)

20.0 (varies during parametric study)

Initial rock temperature T 1 (°C)

20.0 (varies during parametric study)

Initial air temperature T 2 (°C)

20.0

Equivalent thermal conductivity [W/(m K)]

Varies

Backfill specific heat [J/(kg K)]

950

Rock thermal conductivity [W/(m K)]

2.67

Rock specific heat [J/(kg K)]

1000

Cement activation energy (J/mol)

Varies

Binder content (kg/m3 )

200 (varies during parametric study)

R (gas constant) [J/(mol·K)]

8.314

Stope size (H × W ) (m × m)

20 × 10 (varies during parametric study)

Backfilling rate (m/d)

2.5 (varies during parametric study)

10 m and 60 m × 30 m, are selected to conduct simulation in this section, with the results shown in Fig. 6.26. As presented in this figure, a larger stope size brings about a greater temperature rise, but requires a lower temperature increase rate to reach the maximum temperature within the CPB structure. As expected, with an increase in the stope size, a larger amount of binder will hydrate, thereby leading to more heat production. Meanwhile, under the same backfilling rate, a larger stope size will require more time to fill the stope, and this is why a larger stope size leads to a lower temperature increase rate to reach the temperature peak within the CPB. (1.2) Stope shape In practice, a stope shape varies from site to site, depending on the in situ situations, such as the operating conditions and deposit location of the ore. As an important shape feature, the H/W ratio (H: stope height; W: stope width) is an identification parameter to describe the stope shape. Shape ratios of 4, 1 and 0.25 are selected to conduct the simulation of the thermal response of CPB structures to stope shape. Figure 6.27 presents the effect of stope shape on the temperature development within the CPB structures. From this figure, it is apparent that a greater temperature increase is associated with a smaller H/W ratio.

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB Fig. 6.26 Effect of stope size on the heat development within CPB

(a) PCI/FA(50/50)

(b) PCI/SLAG (50/50)

(c) PCI (100%)

91

92

6 Properties of Cemented Tailings Backfill

(a) PCI/FA(50/50)

(b) PCI/SLAG (50/50)

(c) PCI (100%)

Fig. 6.27 Effect of stope shape on the heat development within CPB

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

93

Fig. 6.28 Effect of mineral admixtures on the heat development within CPB

(2) Effect of mineral admixtures In order to demonstrate the effect of mineral admixtures, three types of binders were selected to conduct the following study: PCI/FA (50/50), PCI/SLAG (50/50), and PCI (100%). The temperature development at the point that is 2 m above the bottom of the backfill stope was observed. Figure 6.28 reveals the result of temperature evolution versus time within the CPB structures affected by the supplement of mineral admixtures. By replacing 50% of the PCI with FA or slag, the maximal temperature decreases because the heat released by the hydration of PC is much more than that of FA or slag. In addition, Fig. 6.29 shows the temperature distribution in the CPB and the surrounding rock, from which the effect of mineral admixtures on the temperature rise and distribution in the CPB is clear. Although the addition of mineral admixtures into PCI will decrease the heat generation within the CPB, the mixed binder can still bring about a significant temperature rise. Therefore, in practical backfilling operations, it could be favorable to proportionally replace PCI with FA or slag in the binder, which can reduce the preparation cost of the CPB. (3) Effect of initial rock temperatures In order to determine the effect on the thermal response of the CPB structure to different thermal conditions (e.g., deep mine), simulation is conducted at various initial rock temperatures (0, 20, 40 °C). Figure 6.30 represents the temperature development of CPBs and their surrounding media after 7 days of curing, while Fig. 6.31 demonstrates the effect of initial rock temperatures on temperature distribution after 30 days of curing. According to these figures, higher initial rock temperatures lead to greater temperature rises and longer times to keep the higher temperatures within the CPB structures. The reason is that a high curing temperature is beneficial to the temperature rise within CPB by accelerating binder hydration, which is consistent with the results obtained by Fall et al. (2010).

94

6 Properties of Cemented Tailings Backfill

Fig. 6.29 Temperature distribution in CPB and surrounding rock

(4) Effect of backfilling rate A small stope of 10 m × 5 m and a large stope of 60 m × 30 m were considered, and backfilling rates of 1, 2.5, 5 and 10 m/d were chosen to separately conduct simulation. Figure 6.32 displays the effect of the backfilling rate on heat development with a small CPB structure at a point located 2 m above the bottom of the stope. It can be observed that a higher backfilling rate generates higher temperatures and helps to maintain the higher temperatures for a longer amount of time. This is because a fast backfilling rate accelerates the heat generation rate and at the same time, reduces heat dissipation to the surroundings. Figure 6.33 presents the effect of the backfilling rate on heat development within a large CPB structure at a point located 2 m above the bottom of the stope. It can be determined that a higher backfilling rate is associated with a greater temperature rise.

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

95

Fig. 6.30 Temperature distribution in CPB and the surrounding rock after 7 days subjected to different initial rock temperatures

As a result, when conditions permit (e.g., pumping capacity) in backfilling practices, it is favorable to use a higher backfilling rate to achieve higher temperatures for the binder hydration within the CPB.

96

6 Properties of Cemented Tailings Backfill

Fig. 6.31 Temperature distribution in CPB and the surrounding rock after 30 days subjected to different initial rock temperatures

(5) Effect of backfilling strategy In backfilling practices in the field, the selection of backfilling strategy (with or without an initial plug) is important. For example, in order to keep the barricades from bearing excess strength, some stopes are filled in two stages, with the initial plug cured for a day or more (Thompson et al. 2009). The simulation for a stope size (H × W ) of 40 m × 20 m with a binder content of 200 kg/m3 , is carried out and a backfilling rate of 5 m/d is applied. Three backfilling strategies were studied and analyzed: first, no plug was used; second, an initial plug of 5 m in height was filled at 5 m/d, followed by a rest period of 1 day, and then the backfilling of the remaining 35 m was completed at a rate of 5 m/d; and finally, an initial plug of 5 m in height was filled at 5 m/d, and then the remaining 35 m

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

(a) PCI/FA(50/50)

(b) PCI/SLAG (50/50)

(c) PCI (100%)

Fig. 6.32 Effect of backfilling rate on the heat development within CPB for a small stope

97

98

6 Properties of Cemented Tailings Backfill

(a) PCI/FA(50/50)

(b) PCI/SLAG (50/50)

(c) PCI (100%)

Fig. 6.33 Effect of backfilling rate on the heat development within CPB for a large stope

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

99

Fig. 6.34 Effect of backfilling strategy on temperature distribution within different CPB after 7 days of filling

was filled at a rate of 5 m/d after a rest period of 3 days. Figure 6.34 demonstrates the effect of the backfilling strategy on temperature distribution within a hydrating CPB. As expected, it can be observed that regardless of binder type, the strategy with an initial plug is more beneficial to the temperature development within CPB than the one without a plug, and longer curing times for the plug will result in greater temperature rise. This is because a filling strategy with an initial plug that has a rest period can provide more time for binder hydration in the plug than that without a plug, and thus more heat is produced, which also favors temperature development within the remaining parts of the CPB structure.

6.3.3.2

Application of the Hydraulic Prediction Ability

After verification, the proposed model will be applied to carry out several simulation studies, in order to reveal the effects of different factors (shape of CTB, binder

100

6 Properties of Cemented Tailings Backfill

content, initial CTB temperature, curing temperature) on water seepage flow rate in CTB. The shape of CTB, which influences the quality or performance of CTB, is dependent on the mould where CTB is placed and cured. Thus, two main CTB shapes (cube and cylinder, as shown in Fig. 6.35) are used to conduct the simulation, and the symmetric plane for each shape (i.e., the two green faces in Fig. 6.35) are selected as the geometry for simulation. The radius of the cylinder is 0.11 m, the side length of the cube is 0.2 m, and the heights for the cube and the cylinder are both 0.2 m. According to the dimensions of these two shapes, they are the same in volume (π × 0.112 × 0.2 = 0.23). Table 6.10 lists the main parameters used for model simulation. It should be stated that, in Table 6.10, binder content represents the binder consumption per unit volume of CTB, and the binder used is Portland Cement type I.

Effect of CTB Shape Figure 6.36 demonstrates the comparison of the temperature (unit: K) distribution on each symmetric plane of the two CTBs after 1 day’s curing. It can be noticed from Fig. 6.36 that the temperature at the center of CTB is higher than that at anywhere within CTB. This is due to the fact of the thermal diffusion along the temperature gradient. From Fig. 6.36 it can also be seen that, after curing 1 day, the peak temperatures of the cubic and cylindrical CTB are respectively 295.61 and 295.79 K. Namely, the peak temperature of the cylindrical CTB is slightly higher than that of the cubic one. This is because the lateral area (2 × π × 0.11 × 0.2 ≈ 0.14 m2 ) of the cylindrical

(a) Cube Fig. 6.35 Shape of the CTB (dimension in m)

(b) Cylinder

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

101

Table 6.10 Main parameters used for model simulation Parameters

Sections Section 6.3.3.2.1

Section 6.3.3.2.2

Section 6.3.3.2.3

Section 6.3.3.2.4

CTB shape

Cube, cylinder

Cube

Cube

Cube

Cement content (kg m−3 )

150

100, 150, 200

150

150

Initial CTB temperature (°C)

25

25

5, 25, 35

25

Curing temperature (°C)

20

20

20

1, 20, 40

Reference temperature (°C)

20

20

20

20

CTB is smaller than that of the cubic one (4 × 0.2 × 0.2 =0.16 m2 ). Therefore, the heat dissipating through the lateral face of the cylindrical CTB to its surroundings is less than that of the cubic. This is why the peak temperature at the center of the cube is lower than that of the cylinder. The velocity (unit: m/s) fields of the water seepage flow on the symmetric planes of the two CTB shapes cured for 1 day are demonstrated by Fig. 6.37. It can be observed that the highest velocity (7.7646 × 10−4 m/s) within the cylindrical CTB is faster than that (7.0018 × 10−4 m/s) within the cubic CTB. This is attributed to the fact that, as discussed before, the temperature of the cylinder is slightly higher than that of the cube, speeding up the velocity of the water seepage flow.

Effect of Cement Content Figure 6.38 reveals the effect of various cement content on the seepage flow rate of CTB. From this figure, it can be found that the water seepage flow rate within CTB decreases with the increase of cement content. The reason for this outcome is that larger amount of cement implies more cement hydration products are generated. Therefore, the pore connectivity within CTB becomes worse, which makes the flowing of the seepage water more difficult. From Fig. 6.38 it can also be seen that, with the curing time, the seepage flow rate increases until it reaches a maximum value, and then it decreases until its value becomes zero. This is due to the fact that, when the cement hydration proceeds with curing time, it releases considerable amount of heat to raise the temperature within CTB, which greatly increase the water seepage flow rate to its peak value. Afterwards, the water seepage flow rate decreases dramatically due to the precipitation of hydration products in the pores of CTB. When the pore structures become too dense for seepage water to flow, the value of seepage flow rate reaches zero.

102

6 Properties of Cemented Tailings Backfill

(a) The symmetric plane of the cubic CTB

(b) The symmetric plane of the cylindrica lCTB Fig. 6.36 Temperature distribution on the symmetric plane of the CTB

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

(a) The symmetric plane of the cubic CTB

(b) The symmetric plane of the cylindrical CTB Fig. 6.37 Velocity field of the water seepage flow on the symmetric plane of the CTB

103

104

6 Properties of Cemented Tailings Backfill

Fig. 6.38 Effect of cement content on the seepage flow rate within CTB versus curing time

Figure 6.39 shows the effect of cement content on the seepage flow velocity (unit: m/s) in CTB with the curing time of 1 day. It can be noticed that from this figure, increasing the cement content can reduce the water seepage flow velocity in CTB. This is because increasing the cement consumption for preparing CTB leads to the increase of cement hydration products, and these products accumulate and precipitate to refine the pores within CTB and in turn break the interconnectivity among these pores.

Effect of Initial CTB Temperature The effect of different initial CTB temperatures (5, 25, and 35 °C) on the evolution of water seepage flow rate within CTB versus curing age is graphically illustrated by Fig. 6.40. It can be noticed that increasing the initial temperature of CTB can reduce its seepage flow rate. This is because increasing the initial temperature of CTB is beneficial to the generating of hydration products, which is obstructive to the seepage water flow.

Effect of Curing Temperature Three kinds of curing temperatures (1, 20, and 40 °C) are selected to develop this simulation. The influence of curing temperature on the evolution of water seepage flow rate versus curing time is revealed by Fig. 6.41. It can be noticed that the seepage flow rate in CTB decreases with the increase of curing temperature. This is attributed to the fact that raising the curing temperature can accelerate the process of

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

105

(a) 100 kg/m3

(b) 150 kg/m3 Fig. 6.39 Effect of cement content on the seepage flow velocity field on the symmetric plane of the CTB structure

106

6 Properties of Cemented Tailings Backfill

(c) 200 kg/m3 Fig. 6.39 (continued)

Fig. 6.40 Effect of initial CTB temperature on the seepage flow rate within CTB versus time

6.3 Themo-Hydraulic Coupled Behavior of Hydrating CPB

107

Fig. 6.41 Effect of curing temperature on the seepage flow rate within CTB versus curing time

cement hydration and thus increase the hydration products. As discussed before, the increased number of cement hydration products is unfavorable to the flow of seepage water, because of the refinement of the pore structures within CTB.

6.4 Thermo-Hydro-Mechanical Behavior of Hydrating CPB Once prepared and placed underground, the CPB structure is simultaneously subjected to mechanical (M, e.g., geomechanical conditions of the mine, filling rate and strategy, backfill self-weight), hydraulic (H, e.g., water drainage, suction, pore water pressure development), chemical (C, e.g., cement hydration), and thermal (T) loads from early to advanced ages (Fall and Ghirian 2014). Hence, the performance of the CPB structure is controlled by coupled multiphysics processes, including thermal (T), hydraulic (H), mechanical (M) and chemical (C) (THMC) processes. Understanding and modeling of the coupled THMC processes that occur in CPB are crucial for reliably assessing and predicting the performance of CPB structures (Cui and Fall 2015). Figure 6.42 depicts the main THMC processes that can affect the CPB behavior. Recently, Cui and Fall (2015) have developed a coupled multiphysics model to analyze and predict the THMC behavior of CPB. The modeling process and the coupled model can be found in their study (Cui and Fall 2015). Since the thermo-hydraulic coupled behavior of hydrating CPB has been discussed in the above section. The mechanical behavior of hydrating CPB will be presented in the current section.

108

6 Properties of Cemented Tailings Backfill

Fig. 6.42 Schematic of interactions among multiphysics processes within CPB (envir.: environment) (Cui and Fall 2015)

Fig. 6.43 Geometry and mesh of the simulated model and monitoring point: a first stage of backfilling; b Second stage of backfilling; and c third stage of backfilling (Cui and Fall 2015)

6.4 Thermo-Hydro-Mechanical Behavior of Hydrating CPB

109

Fig. 6.44 Predicted vertical settlement versus experimental data of vertical settlement (Cui and Fall 2015)

In order to experimentally study the coupled THMC behavior of CPB, experiments on high columns with CPB were carried out in previous research (Ghirian and Fall 2013, 2014). In the experiment, staged backfilling strategy was adopted and completed in 3 layers that were 50 cm each in height for 3 consecutive days. Each backfilling stage was completed in a very short time (within 5 min). One day of curing time (or delay) between each lift was adopted. The high column tests were performed under insulated-undrained conditions. During the high column test, extensive laboratory testing and measurements were carried out, including the monitoring of the variation of the pore water pressure, temperature distribution, vertical settlement within the CPB and the UCS (Cui and Fall 2015). The model characteristics, mesh and monitoring point are shown in Fig. 6.43. The predictive ability of the coupled model is verified against the experimental data of vertical settlement measured on the surface of the top layer in the CPB. A comparison between the simulated and measured values is presented in Fig. 6.44. From this figure, it can be noticed that the modeling results are in very good agreement with the measured data (Cui and Fall 2015). Apart from the measurement of vertical displacement of the high CPB column, a series of UCS tests were also performed on CPB specimens sampled from the column after 7, 28, 90, and 150 days of curing. The obtained UCS data are used to further validate the mechanical component of the developed model (Cui and Fall 2015). Typical validation results are shown in Fig. 6.45. As presented in this figure, the good agreement between predicted outcomes and experimental data of stress–strain curve indicates that the mechanical component of the developed model is capable to

110

6 Properties of Cemented Tailings Backfill

Fig. 6.45 Comparison of numerical results and measured data of stress–strain relation (Cui and Fall 2015)

predict relatively well the time-dependent stress–strain response of the CPBs. In other words, the model can reproduce well the strain hardening-softening of hydrating CPB (Cui and Fall 2015). For validating the capability of the developed model to describe the deformation under triaxial compression stress conditions, triaxial compression tests on CPBs of different ages and corresponding simulation are performed. For the preparation of CPB samples, the ground silica tailing, ordinary PCI and tap water are employed. 4.5% PCI and a water-cement ratio (w/c) of 7.6 is used as mix proportions. For measurement of axial displacement, a linear variable displacement transformer (LVDT) was utilized. Under 200 kPa confining pressure, a minimum of 3 samples were tested for each curing time (Cui and Fall 2015). The geometry with mesh is presented in Fig. 6.46. As indicated in Fig. 6.47, a good agreement between predicted results and experimental data is obtained in terms of axial strain and volumetric strain. Both simulation results and measured data demonstrate that the volumetric strain first decreases with deviator stress (i.e., volumetric contraction), then, the volumetric expansion appears.

6.4 Thermo-Hydro-Mechanical Behavior of Hydrating CPB

111

Fig. 6.46 Mesh and geometry of triaxial compression test (Cui and Fall 2015)

This is due to the fact that, as the deviatoric stress increases, the elastic volumetric strain develops within CPB samples (Cui and Fall 2015).

112

6 Properties of Cemented Tailings Backfill

Fig. 6.47 Comparison of simulated results and experimental data of triaxial compression tests (Cui and Fall 2015)

References Abdul-Hussian, N., Fall, M.: Unsaturated hydraulic properties of cemented tailings backfill that contains sodium silicate. Eng. Geol. 123(4), 288–301 (2011) Abdul-Hussian, N., Fall, M.: Thermo-hydro-mechanical behavior of sodium silicate-cemented paste tailings in column experiments. Tunn. Undergr. Space Technol. 29, 85–93 (2012) Alnajim, A.: Modélisation et simulation du comportement du béton sous hautes températures par une approche thermo-hygro-mécanique couplée. Application à des situations accidentelles, Ph.D. thesis, Université Marne La Vallée (2004) Baggio, P., Bonacina, C., Schrefler, B.A.: Some consideration on modelling heat and mass transfer in porous media. Transp. Porous Media 28(3), 233–251 (1997) Bary, B., Sellier, A.: Coupled moisture-carbon dioxide-calcium transfer model for carbonation of concrete. Cem. Concr. Res. 34(10), 1859–1872 (2004) Belem, T., El Aatar, O., Bussière, B., Benzaazoua, M., Fall, M., Yilmaz, E.: Characterization of self-weight consolidated paste backfill. In: Jewell, R., Lawson, S., Newman, P. (eds.) Proceedings of the 9th International Seminar on Paste and Thickened Tailings, Limerick, Ireland, pp. 333–345. ACG Bourgeois, F., Burlion, N., Shao, J.F.: Modelling of elastoplastic damage in concrete due to desiccation shrinkage. Int. J. Numer. Anal. Methods Geomech. 26(8), 759–774 (2002) Celestin, J., Fall, M.: Thermal conductivity of cemented paste backfill material and factors affecting it. Int. J. Min. Reclam. Environ. 23(4), 274–290 (2009) COMSOL: COMSOL Multiphysics 5.0. http://www.comsol.com (2014)

References

113

Cooke, R.: Design procedure for hydraulic backfill distribution systems. J. S. Afr. Inst. Min. Metall. 101:97–102 (2001) Cui, L., Fall, M.: A coupled thermo-hydro-mechanical-chemical model for underground cemented tailings backfill. Tunn. Undergr. Space Tech. 50(50), 396–414 (2015) De Schutter, G.: Hydration and temperature development of concrete made with blast-furnace slag cement. Cem. Concr. Res. 29, 143–149 (1999) De Souza, E., Hewitt, A.: The contribution of cemented backfill to heat loads. In: 8th International Mine Ventilation Congress. Brisbane, pp. 87–93 (2005) D’Aloia, L., Chanvillard, G.: Determining the “apparent” activation energy of concrete: Ea—numerical simulations of the heat of hydration of cement. Cem. Concr. Res. 32(8), 1277–1289 (2002) Fall, M., Adrien, D., Celestin, J.C., Pokharel, M., Toure, M.: Saturated hydraulic conductivity of cemented paste backfill. Miner. Eng. 22(15), 1307–1317 (2009) Fall, M., Celestin, J.C., Pokharel, M., Touré, M.: A contribution to understanding the effects of curing temperature on the mechanical properties of mine cemented tailings backfill. Eng. Geol. 114, 397–413 (2010) Fall, M., Ghirian, A.: Laboratory vane shear tests and slump tests on cemented paste backfill. Report-Ottawa (2012) Fall, M., Ghirian, A.: Coupled thermo-hydro-mechanical-chemical evolution of cemented paste backfill and implications for backfill design-experimental results. In: 11th International Symposium on Mining with Backfill, MineFill 2014. Perth, Australia, pp. 183–196 (2014) Fall, M., Samb, S.S.: Pore structure of cemented tailings materials under natural or accidental thermal loads. Mater. Charact. 59(5), 598–605 (2007) Ghirian, A., Fall, M.: Coupled thermo-hydro-mechanical-chemical behavior of cemented paste backfill in column experiments. Part I: physical, hydraulic and thermal processes and characteristics. Eng. Geol. 164, 195–207 (2013) Ghirian, A., Fall, M.: Coupled thermo-hydro-mechanical-chemical behavior of cemented paste backfill in column experiments: part II: mechanical, chemical and microstructural processes and characteristics. Eng. Geol. 170, 11–23 (2014) Jonasson, J.E., Groth, P., Hedlund, H.: Modeling of temperature and moisture field in concrete to study early age movements as a basis for stress analysis. In: Proceedings of the International RILEM Symposium on Thermal Cracking in Concrete at Early Ages, pp. 42–52. London (1995) Kesimal, A., Ercikdi, B., Yilmaz, E.: The effect of desliming by sedimentation on paste backfill performance. Miner. Eng. 16(10), 1009–1011 (2003) Kim, J.K., Han, S.H., Lee, K.M.: Estimation of compressive strength by a new apparent activation energy function. Cem. Concr. Res. 31(2), 217–225 (2001) Kjellsen, K.O., Detwiler, R.J., Gjørv, O.E.: Development of microstructures in plain cement pastes hydrated at different temperatures. Cem. Concr. Res. 21(1), 179–189 (1991) Krus, M., Hansen, K.K., Kiinzel, H.M.: Porosity and liquid absorption of cement paste. Mater. Struct. 30(7), 394–398 (1997) Luckner, L., Genuchten, M.T.V., Nielsen, D.R.: A consistent set of parametric models for the twophase flow of immiscible fluids in the subsurface. Water Resour. Res. 25(10), 2187–2193 (1989) Maekawa, K., Chaube, R., Kishi, T.: Modeling of Concrete Performance: Hydration. Microstructure Formation and Mass Transport, E&FN SPON, London (1999) Maekawa, K., Ishida, T.: Modeling of structural performances under coupled environmental and weather actions. Mater. Struct. 35(10), 591–602 (2002) Mainguy, M., Coussy, O., Baroghel-Bouny, V.: Role of air pressure in drying of weakly permeable materials. J. Eng. Mech. 127(6), 582–592 (2001) Mualem, Y.: A new model for predicting the hydraulic conductivity of unsaturated porous media. Water Resour. Res. 12, 513–522 (1976) Nasir, O., Fall, M.: Modeling the heat development in hydrating CPB structures. Comput. Geotech. 36(7), 1207–1218 (2009) Papo, A., Caufin, B.: A study of the hydration process of cement pastes by means of oscillatory rheological techniques. Cem. Concr. Res. 21, 1111–1117 (1991)

114

6 Properties of Cemented Tailings Backfill

Petit, J.Y., Khayat, K.H., Wirquin, E.: Coupled effect of time and temperature on variations of yield value of highly flowable mortar. Cem. Concr. Res. 36(5), 832–841 (2006) Poyet, S., Charlesa, S., Honoré, N., L’hostisa, V.: Assessment of the unsaturated water transport properties of an old concrete: Determination of the pore-interaction factor. Cem. Concr. Res. 41(10), 1015–1023 (2011) Richards, L.A.: Capillary conduction of liquids through porous mediums. Physics 1(5), 318–333 (1931) Schindler, A.K.: Effect of temperature on the hydration of cementitious materials. ACI Mater. J. 101(1), 72–81 (2004) Schindler, A.K., Folliard, K.J.: Heat of hydration models for cementitious materials. ACI Mater. J. 102(1), 24–33 (2005) Somerton, W.H., Keese, J.A., Chu, S.L.: Thermal behavior of unconsolidated oil sands. In: Proceedings of 48th Annual Fall Meeting of the Society of Petroleum Engineers, September 1973, Paper SPE-4506 Las Vegas, USA (1973) Talor, S.C., Hoff, W.D., Wilson, M.A., Green, K.M.: Anomalous water transport properties of Portland and blended cement-based materials. J. Mater. Sci. Lett. 18(23), 1925–1927 (1999) Thomas, H.R., Sansom, M.R.: Fully coupled analysis of heat, moisture and air transfer in unsaturated soil. J. Eng. Mech. 121(3), 392–405 (1995) Thompson, B.D., Grabinsky, M.W., Bawden, W.F., Counter, D.B.: In-situ measurements of cemented paste backfill in long-hole stopes. In: Proceedings of the 3rd Canada-US Rock Mechanics Symposium and 20th Canadian Rock Mechanics Symposium (RockEng09), Toronto (2009) Tong, F., Jing, L., Zimmerman, R.W.: A fully coupled thermo-hydro-mechanical model for simulating multiphase flow, deformation and heat transfer in buffer material and rock masses. Int. J. Rock Mech. Min. Sci. 47(2), 205–217 (2010) van Genuchten, M.T.: A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44(5), 892–898 (1980) Wang, C.Y., Beckermann, C.: A two-phase mixture model of liquid-gas flow and heat transfer in capillary porous media-I. Formulation. Int. J. Heat Mass Transf. 36(11), 2747–2758 (1993) Williams, T.J., Denton, D.K., Larson, M.K., Rains, R.L., Seymour, J.B., Tesarik, D.R.: Geomechanics of reinforced cemented backfill in an underhand stope at the Lucky Friday Mine, Mullan, Idaho. US Department of Health and Human Services (2001) Wu, D., Fall, M., Cai, S.: Coupling temperature, cement hydration and rheological behaviour of fresh cemented paste backfill. Miner. Eng. 42, 76–87 (2013) Wu, D., Fall, M., Cai, S.: Numerical modelling of thermally and hydraulically coupled processes in hydrating cemented tailings backfill columns. Int. J. Min. Reclam. Env. 28(3), 173–199 (2014) Wu, D., Yang, B., Liu, Y.: Pressure drop in loop pipe flow of fresh cemented coal gangue-fly ash slurry: experiment and simulation. Adv. Powder Technol. 26(3), 920–927 (2015) Yilmaz, E., Kesimal, A., Ercidi, B.: Strength development of paste backfill simples at Long term using different binders. In: Proceedings of 8th Symposium MineFill04, China, pp. 281–285 (2004)

Chapter 7

Case Study of Cemented Tailings Backfill

Abstract Experimental test, theoretical calculation, and numerical simulation are all conducted to determine both the optimum solids content and cement-to-tailings ratio of the CTB materials of Baixiangshan Iron Mine. Besides, another case study reports the improvement of CTB technology used in Xincheng Gold Mine. The optimization work solved the problem of material segregation and also increased the solids content of the backfill materials. Keywords Cemented tailings backfill · Bleeding rate · Flocculating agent · Drag reduction agent · Deslimed tailings

7.1 Practice of CTB Technology in Baixiangshan Iron Mine Baixiangshan Iron Mine, located in Dangtu County, Anhui Province, China, belongs to Gushan Mining Company, which is operated by Magang (Group) Holding Co., Ltd. A working sketch of Baixiangshan Iron Mine is shown in Fig. 7.1. Baixiangshan Iron Mine has an iron ore reserve of more than 130 million tons, and most are magnetic iron ore. The commissioning of this mine was since 2013, and the output of raw iron ore per year for this mine is more than 2 million tons. Due to limited land for construction of tailings impoundments, this mine applies cemented tailings backfill (CTB) technology for tailings management. The CTB solution can significantly reduce the threat to the environment posed by mining operations. This mine employs two main mining methods: upward drift stoping with backfill and sublevel open stoping with delayed backfill. A gravity-fed backfill reticulation system is used to prepare, transport, and place the CTB materials. Figure 7.2 demonstrates the technological process for preparing CTB at Baixiangshan Iron Mine. The tailings used are from the dressing plant of Baixiangshan Iron Mine, and the binder used is ordinary Portland cement 425#. The water used is tap water. Figure 7.3 shows the particle size distribution of the tailings. Tables 7.1 presents the chemical compositions of the tailings used. The CTB mixtures with dimension of 7 × 7 × 7 cm3 , various ratios of cementto-tailings (c/t: 1:4, 1:6, 1:8, and 1:10) and solids contents (58, 60, 62, and 64%) © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_7

115

116

7 Case Study of Cemented Tailings Backfill

Fig. 7.1 Working sketch of Baixiangshan Iron Mine

Thickener

Storehouse Binder

Mixer

Borehole Fig. 7.2 Technological process for preparing CTB at Baixiangshan Iron Mine

7.1 Practice of CTB Technology in Baixiangshan Iron Mine

117

Fig. 7.3 Particle size distribution of the tailings used

Table 7.1 Chemical compositions of the tailings used Components SiO2 35.81% Components Na2 O 0.515% Components ZnO 0.0176%

MgO

CaO

Fe2 O3

Al2 O3

P2 O5

K2 O

18.30%

10.08%

9.82%

8.64%

2.64%

1.92%

S

TiO2

MnO

Cl

V2 O5

SrO

0.479%

0.389%

0.130%

0.126%

0.0408%

0.0296%

CeO2

Co3 O4

CuO

Rb2 O

NiO

ZrO2

0.0116%

0.0094%

0.0091%

0.0052%

0.0050%

0.0034%

and cured for different periods (3, 7 and 28d) are subjected to UCS (Fig. 7.4), slump (Fig. 7.5), bleeding rate (Fig. 7.6) and rheology tests, respectively. Tables 7.2 and 7.3 summarize the UCS and slump test results, respectively. Table 7.4 lists the bleeding rate test results. In addition to experimental study, theoretical calculation and numerical simulation are also conducted. On the basis of the test, calculation, and simulation results, the optimum solids content and cement-to-tailings ratio for preparing the CTB materials are obtained, associated with the rheological parameters, as shown in Table 7.5.

7.2 Optimization of CTB Technology in Xincheng Gold Mine Xincheng Gold Mine is located at Laizhou bay, Jiaodong peninsula, Sahndong province, China, and belongs to the Shandong Gold Group, Ltd. The mine started to be set up in 1975, and put into production in 1980. After three times of expansion,

118 Fig. 7.4 UCS test

Fig. 7.5 Slump test

Fig. 7.6 Bleeding rate tests

7 Case Study of Cemented Tailings Backfill

7.2 Optimization of CTB Technology in Xincheng Gold Mine

119

Table 7.2 UCS test results of the CTB mixtures c/t 1:4

1:6

1:8

1:10

Density (g/cm3 )

Solids content (%)

UCS (MPa) 3d

7d

28d

58

1.57

0.292

0.652

2.701

60

1.63

0.510

1.094

3.104

62

1.68

0.612

1.252

3.402

64

1.76

0.740

1.564

3.784

58

1.60

0.152

0.340

1.510

60

1.58

0.172

0.364

1.727

62

1.62

0.232

0.508

2.125

64

1.70

0.272

0.640

2.571

58

1.58

0.156

0.240

1.066

60

1.59

0.160

0.280

1.273

62

1.69

0.180

0.332

1.664

64

1.77

0.220

0.396

2.041

58

1.62

0.120

0.188

0.835

60

1.60

0.132

0.224

1.047

62

1.69

0.152

0.273

1.436

64

1.78

0.193

0.334

1.842

Table 7.3 Slump (cm) test results of the CTB mixtures 58%

60%

62%

64%

1:4

28.1

27.5

26.0

23.5

1:6

28.0

27.6

26.0

22.3

1:8

28.0

27.4

25.8

21.8

1:10

28.1

27.3

25.8

21.3

62%

64%

Table 7.4 Bleeding rate test results of the CTB mixtures 58%

60%

1:4

11.98%

9.26%

7.36%

4.68%

1:6

12.47%

9.65%

7.72%

4.86%

1:8

12.86%

9.94%

8.07%

5.29%

1:10

13.11%

10.06%

8.13%

5.97%

Table 7.5 Optimum data for preparing the CTB mixtures

Solids content

c/t

Viscosity (Pa s)

Yield stress (Pa)

62%

1:4

0.38

178.99

120

7 Case Study of Cemented Tailings Backfill

Fig. 7.7 Current technological process for preparing CDTB at Xincheng Gold Mine

Fig. 7.8 Creep test results of CDTBs

production capacity of the mine from the beginning 500 t/d expanded to recently 6000 t/d. This mine employs two main mining methods: mechanized panel cut-and-fill mining with backfill and mechanized panel uphand heading mining with backfill. Currently, Xincheng Gold Mine utilized cemented deslimed tailings backfill (CDTB) materials for management of mine waste and also treatment of mined-out areas. The CDTB materials were transported through pipeline to underground mined stopes by gravity. Figure 7.7 shows the technological process for preparing the CDTB materials at Xincheng Gold Mine. Some problems of the current CDTB craft used in Xincheng Gold Mine were as follows:

7.2 Optimization of CTB Technology in Xincheng Gold Mine

121

Table 7.6 Rheology test results of fresh CDTBs Solids content (%)

Viscosity of fresh CDTB (Pa s) Cement-to-deslimed tailings 1:4

1:6

1:8

1:10

1:20

65

1.000

1.000

1.000

6.137

32.438

66

1.000

1.000

1.000

38.671

71.060

67

0.965

1.000

1.000

11.527

9.801

68

1.000

1.278

1.000

18.587

81.908

69

1.000

1.436

1.000

165.263

3.253

70

1.000

1.205

1.191

1.257

2.366

71

1.569

1.698

1.923

3.751

4.437

72

2.476

1.890

3.804

7.189

7.662

73

3.000

2.804

3.900

10.084

13.260

74

4.892

4.688

8.010

16.449

23.407

75

10.535

9.379

13.961

25.751

26.632

76

18.233

22.438

10.612

50.914

34.448

77

48.092

60.663

47.120

55.259

41.279

78

114.715

146.978

93.187

65.829

59.357

79

285.009

172.361

145.080

121.040

55.070

80

385.389

327.307

319.047

260.183

187.931

(1) The distance for conveying the CDTB was very long, and the configuration of the backfill reticulation system was very complicated, and pipe choke happened frequently during filling operation. (2) Mass density of fill pulp was only about 68%, which led to the loss of cement and fine tailings particles during draining when the CDTB materials were filled into a stope, and resulting in strength reduction of placed CDTB structures as well as contamination of underground drifts and working faces. (3) For the uphand heading stoping method, the problem of low pulp density became more serious, for instance, the placed CDTB cannot contact and thus to support the roof of mined-out heading. (4) Recovery of deslimed tailings was around 65%, i.e., nearly 35% of very fine tailings needed to be stored in tailings pond, which posed a threat to stability of the tailings dam. According to rheology, compressibility and creep tests in the laboratory, as well as in situ investigation, the CDTB technology currently utilized in Xincheng Gold Mine was optimized. Table 7.6 presents the rheology testing results of fresh CDTB materials. Table 7.7 shows the compressibility test results of hardened CDTB structures. Figure 7.8 demonstrates the creep test results of hardened CDTB structures. The field monitoring data are presented in Fig. 7.9.

122 Table 7.7 Compressibility test results of CDTBs

7 Case Study of Cemented Tailings Backfill Loading Total deformation value (mm) stress (MPa)

Strain (%)

0.5

0.432

0.27

1

1.261

0.79

2

1.898

1.19

4

2.431

1.52

6

3.619

2.26

9

4.348

2.72

15

4.960

3.10

22

6.089

3.81

Fig. 7.9 Field monitoring data

The optimization study indicated that flow behavior of fresh CDTB was improved by admixing a flocculating agent (Minefill309) and a drag reduction agent (Minefill501). Besides, the segregation of fresh CDTB had also been effectively controlled, and the solids content of fresh CDTB had been increased to 75–78%. The delivery of CDTB had become more stable and pipe wearing had also been reduced.

Chapter 8

Solutions for Underground Placement of Coal Mine Waste

Abstract An environmentally friendly solution for underground discharge of coal mine wastes is to place these wastes into underground mined-out areas. Generally, the wastes are made into solid backfill or fluid backfill, being delivered underground through belts or pipelines. The technological processes for both the solid backfill mining technology and fluid backfill mining method are introduced. Keywords Coal mine · Solid waste · Gob · Solid backfill · Fluid backfill In China, a recently developed solution for underground discharge of coal mine wastes is making the wastes into backfill materials and thereafter placing them in underground voids (or mined-out areas). According to the states of the backfill materials, two major methods are commonly used, and they are respectively solid backfill and fluid backfill.

8.1 Solid Backfill The solid backfill mining technology, developed by the China University of Mining and Technology with independent intellectual property rights, has been successfully applied in more than ten mining areas in China (Huang et al. 2011; Zhang et al. 2009, 2015). In the solid backfill mining operation, coal gangue, fly ash, and some other solids are used as filling materials and delivered to the gob of the underground working face after being treated (Zhang et al. 2015). The key step of the solid backfill mining technology is implementing the procedure of solid waste transport and backfill into the coal mining operation. Figure 8.1 schematically illustrates the technological process of solid backfill mining technology. In addition to the disposal of coal mine wastes, this technology can also control ground subsidence and improve recovery percent of coal resources. From Fig. 8.1, it can be clearly seen that through the feeding, transportation, and backfill systems, the aboveground deposited coal mining wastes are placed underground during the solid backfill mining operation. Figure 8.2 further graphically reveals the underground delivery and backfill of the coal mine wastes. As shown © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_8

123

124

8 Solutions for Underground Placement of Coal Mine Waste

Fig. 8.1 Schematic diagram for the solid backfill mining operation (Zhang et al. 2015)

Fig. 8.2 Underground placement of the coal mine waste (Zhang et al. 2016)

8.1 Solid Backfill

125

in this figure, the mine wastes are delivered and filled into the gob behind the coal mining face through the backfilling scraped conveyor and compacted by the tamping arm of the hydraulic support.

8.2 Fluid Backfill The fluid backfill operation is to prepare the coal mine wastes into fluid materials and then backfill them underground through pipeline. Commonly, the coal mine wastes are mixed with water and binder to form cemented backfill materials (CBM). Afterwards, the CBM is placed into the underground gob through pipeline transportation. Figure 8.3 schematically illustrates the placement of CBM through pipeline transportation into the underground mined-out areas. As shown in this figure, the pump is used as an option to provide the power for transporting the CBM. If the vertical distance from the mixer to the underground void is able to provide enough head to overcome the pressure drop (or resistance loss) for transporting the CBM through pipeline, the CBM can be delivered by gravity without the pump. The transportation by gravity flow is also associated with significant economic performance. Otherwise, when the vertical distance or the gravity is not enough, the optional pump should be equipped to supplement the inadequate head for the transportation of CBM. Option Pump

Mixer

Ground surface

Main roof

Caved zone

Immediate roof

CBM

Coal seam

Pipeline

Hydraulic support

Fig. 8.3 Underground placement of CBM through pipeline transportation

126

8 Solutions for Underground Placement of Coal Mine Waste

The fluid backfill operation has been experimentally used in some coal mines in China, such as Taiping Coal Mine, Suncun Coal Mine, Tonger Coal Mine and Xiaotun Coal Mine (Xing et al. 2013). Currently, for fluid backfill in China, there are two main operations: paste backfill and hydraulic backfill with highly concentrated slurry at a usual solid concentration of 75–80% by weight. In contrast with solid backfill and paste backfill, the hydraulic backfill with highly concentrated slurry is more favorable than them, respectively in terms of ground control and cost saving (Li and Yang 2011). In terms of underground disposal of coal mine waste, a common solution is to mix coal gangue, fly ash, binder, water (and additive if necessary) to prepare highly concentrated material that is called cemented coal gangue-fly ash backfill (CGFB). In the following Chap. 9, the properties of CGFB mixtures will be discussed, and a case study of fluid backfill practice applying the CGFB solution in Xinyang Coal Mine will be presented in Chap. 10.

References Huang, Y., Zhang, J., Zhang, Q., Nie, S.: Backfilling technology of substituting waste and fly ash for coal underground. Environ. Eng. Manag. J. 10, 769–775 (2011) Li, Y., Yang, B.: Stowing mining technology development and classification of Chinese modern coal mine. Coal Min. Technol. 16(5), 1–4 (2011) (in Chinese) Xing, J., Yang, B., Li, Y., Li, Z., Yang, P., Li, W., Ding, P.: Discussion on the development direction of coal mine filling mining technology. Coal Mine Saf. 44(12), 189–191 (2013) (in Chinese) Zhang, J., Miao, X., Guo, G.: Development status of backfilling technology using raw waste in coal mining. J. Min. Saf. Eng. 26(4), 395–401 (2009) Zhang, J., Zhang, Q., Sun, Q., Gao, R., Germain, D., Abro, S.: Surface subsidence control theory and application to backfill coal mining technology. Environ. Earth Sci. 74, 1439–1448 (2015) Zhang, J., Li, B., Zhou, N., Zhang, Q.: Application of solid backfilling to reduce hard-roof caving and longwall coal face burst potential. Int. J. Rock Mech. Min. Sci. 88, 197–205 (2016)

Chapter 9

Properties of Cemented Coal Gangue-Fly Ash Backfill

Abstract Cemented coal gangue-fly ash backfill (CGFB), which is a mixture of coal gangue, fly ash, cement, and water, is being extensively used for subsidence control and waste management. A pipe flow model is developed for predicting the flow behavior of fresh CGFB slurry in the pipe loop. This model is validated by the contrast of simulation outcomes and loop testing results. Afterwards, the uniaxial compressive strength (UCS) and ultrasonic pulse velocity (UPV) of hardened CGFB are investigated. The UCS and UPV values of CGFB increase with increasing fly ash dosage and solid content. One of the most significant evaluation criteria for CGFB is its mechanical performance, which is subjected to coupled thermo-hydro-chemical (THC) effects. A THC coupled model is developed to investigate the coupled THC behavior of CGFB and its evolution versus time. A favorable agreement between the modeling results and experimental data can validate the capability of the developed model to simulate the coupled THC responses in CGFB. Both column and block experiments are conducted to investigate the thermal, hydraulic and mechanical performance of CGFB. The results indicate that the hydraulic and mechanical behaviors of the CGFBs are significantly affected by thermal factors, contributing to a better understanding of the thermo-hydro-mechanical (THM) behavior CGFB. Furthermore, the CGFB performance is subjected to the thermal (T), hydraulic (H), mechanical (M) and chemical (C) processes and THMC coupling effects. A THMC coupled model is developed to describe and analyze the coupled THMC processes that occur in CGFB. The simulation results of the developed model are compared with the experimentally tested data from three case studies, validating the capability of the THMC model. The modeling results can also contribute to a better design of stable, durable and environment-friendly CGFB structures. Keywords Cemented coal gangue-fly ash backfill · Loop test · Pressure drop · Uniaxial compressive strength · Ultrasonic pulse velocity · THMC · Coupled model As described before, coal gangue, fly ash, binder, and water (and additive if necessary) are commonly blended to prepare highly concentrated material that is called cemented coal gangue-fly ash backfill (CGFB). In the current chapter, the fluid characteristics of freshly formed CGFB slurry and the solid properties of hardened CGFB © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_9

127

128

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Table 9.1 Chemical composition of the CG and FA used Chemical compositions (wt%) SiO2

Al2 O3

Fe2 O3

CaO

K2 O

TiO2

S

CG

41.86

23.43

5.09

23.74

0.82

1.36

3.70

FA

56.89

31.89

5.38

1.84

1.39

1.95

0.66

structure will be presented and discussed. It is stated that coal gangue and fly ash are abbreviated as CG and FA, respectively, in the following sections.

9.1 Materials The CG and FA materials used in this study are, respectively, obtained from a coal mine in the northwest of China and a power plant near the mine. The binder used is ordinary Portland cement 425# and the water used is tap water. The chemical properties of the CG and FA samples are shown in Table 9.1 and Fig. 9.1 illustrates the X-ray diffraction (XRD) testing results for the CG and FA. According to the data from Table 9.1 and Fig. 9.1, contents of Al2 O3 and SiO2 contained in the CG are, respectively, 23.43 and 41.68 wt%, indicating that the CG samples used can be classified as argillaceous rocks (with Al2 O3 of 15–30 wt% and SiO2 40–60 wt%), which are suitable to produce CGFB mixtures. In addition, SiO2 content in the FA is high (56.89 wt%), which provides activity for the FA to participate in hydration. Table 9.2 gives the particle size compositions of the CG and FA. Furthermore, Figs. 9.2 and 9.3 demonstrate the particle size distributions of the CG and FA used. It is pointed out that the maximum particle size of the CG should be smaller than 15 mm.

9.2 Rheology and Flowability of Fresh CGFB 9.2.1 Rheology of Fresh CGFB A rheometer test is carried out to investigate the rheology of fresh CGFB slurry. Table 9.3 shows the mix proportions and densities of the fresh CGFB slurry samples (1#–4#) used for the rheometer tests. The slurry samples 1# and 3# are taken as examples, and Fig. 9.4 graphically displays the rheometer testing results of these two slurry samples. Figs. 9.4a, b, respectively, reveal the shear stress evolutions of 1# and 3# CGFB slurry samples versus shear rate. As illustrated in Fig. 9.4, the shear stress of the CGFB correlates well with the shear rate in a linear relationship. The correlation coefficient (R) for 1# slurry sample is 0.97, while that for 3# is 0.98. According to

9.2 Rheology and Flowability of Fresh CGFB

129

Fig. 9.1 XRD profiles of the CG (a) and FA (b) samples Table 9.2 Particle size composition of the CG and FA used D10 (μm)

D30 (μm)

D50 (μm)

D60 (μm)

D90 (μm)

CG

204.722

856.589

1952.857

2736.752

6634.312

FA

10.779

43.826

82.471

112.358

400.008

130

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.2 Particle size distribution of the CG Fig. 9.3 Grain size distribution of the fly ash

Table 9.3 Mix proportions and densities of the fresh CGFB slurry samples Concentration (wt%)

CG content (wt%)

FA content (wt%)

Cement content (wt%)

Density (kg m−3 )

1#

78.0

46.0

22

10

1980

2#

78.5

46.5

22

10

2035

3#

79.0

47.0

22

10

2060

4#

79.5

47.5

22

10

2098

9.2 Rheology and Flowability of Fresh CGFB

131

Fig. 9.4 Rheometer testing results of 1# (a) and 3# (b) CGFB slurry samples

Fig. 9.4, it can also be concluded that the CGFB slurry behaves like Bingham plastic fluid (Chandel et al. 2009; Wu et al. 2013): τ = μB γ + τy

(9.1)

where τ and τ y are, respectively, the shear stress and yield stress of the fluid (the CGFB slurry in the present study), μB is the Bingham plastic viscosity, γ is the shear rate. Based on the Bingham plastic model expressed by Eq. (9.1) and the rheometer testing results as shown in Fig. 9.4, the Bingham plastic viscosity and yield stress for 1# CGFB slurry sample are respectively 5.9581 Pa s and 164.76 Pa, and those for 3# sample are separately 8.2779 Pa s and 165.69 Pa. Moreover, based on the rheometer tests and Bingham linear-regression analyses, Table 9.4 summarizes the rheological parameters (viscosity and yield stress) for all the four CGFB slurry samples.

132

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Table 9.4 Rheological parameters of the CGF slurry samples Concentration (wt%)

Bingham plastic viscosity (Pa s)

Yield stress (Pa)

Correlation coefficient

1#

78.0

5.9581

164.76

0.97

2#

78.5

7.1637

165.12

0.97

3#

79.0

8.2779

165.69

0.98

4#

79.5

9.3268

167.53

0.97

9.2.2 Flowability of Fresh CGFB 9.2.2.1

Loop Test

A test loop system is employed to investigate the pipe flow behavior of CGF slurry and measure the pressure drop data of different CGFB slurry samples. A schematic layout of the test loop system is presented in Fig. 9.5. The pipeline is totally 35 m long, and the pipe inner diameter is 120 mm. The slurry is supplied from the slurry tank (3 in Fig. 9.5), which has a capability of 1 m3 . A mixer (2 in Fig. 9.5) is equipped in the slurry tank to keep the slurry at homogeneous state. The pipe loop and slurry tank are both fabricated with stainless steel. The slurry is circulated through the testing system by a centrifugal pump (1 in Fig. 9.5), which can offer 80 m3 slurry per unit hour and 12 MPa pumping pressure at the maximum power. The slurry volumetric flow rate is measured by using an electromagnetic flow meter (11 in Fig. 9.5), and the concentration profile of the slurry is measured by a sampling probe (12 in Fig. 9.5). As shown in Fig. 9.5, the differential pressure between the pressure tapings 5 and 6 are investigated as the partial pressure drop, while the one between the pressure

Fig. 9.5 Schematic diagram of the test loop system

9.2 Rheology and Flowability of Fresh CGFB Table 9.5 Loop testing results of the fresh CGFB slurry samples

133

Concentration (wt%)

Partial pressure drop (kPa)

Total pressure drop (kPa)

1#

78.0

2.43

19.97

2#

78.5

3.36

26.24

3#

79.0

4.82

33.36

4#

79.5

5.67

37.58

tapings 4 and 7 is as the total pressure drop (including the partial pressure drop). The pipe length between the pressure tapings 4 and 7 is 6.8 m. The partial pressure drop and total pressure drop data are respectively collected by two differential pressure transducers (8 and 9 in Fig. 9.5). Before collecting the pressure data, sufficient time is allowed to ensure that the slurry flows at a uniform and continuous state in the pipe loop. The slurry flow status can be determined both by visually observing from an observation chamber (10 in Fig. 9.5) and by examining the readings of flowing rate and pressure drop trend. All the readings from the differential pressure transducers and electromagnetic flow meter are recorded in a PC-controlled data acquisition system. During loop testing of the slurry, valve 14 is turned off and valve 13 is turned on. Generally, the loop testing process lasts for half an hour, while experimental data are recorded. After completing the loop test, valve 13 is turned off and valve 14 is turned on, and the waste slurry is drained out from the discharge outlet (15 in Fig. 9.5). The partial pressure drop (differential pressure between tapings 5 and 6 in Fig. 9.5) and total pressure drop (differential pressure between tapings 4 and 7) data of these samples are experimentally obtained, as tabulated in Table 9.5.

9.2.2.2

Development of the Pipe Flow Model

During the loop test, the fresh CGFB slurry is assumed as a single-phase fluid flow in the loop pipe. The software package COMSOL Multiphysics is applied here to conduct the numerical simulation. COMSOL Multiphysics provides various interfaces (such as Acoustics, Fluid Flow, Heat Transfer, Structural Mechanics and so on) for science and engineering applications (COMSOL 2014). For this study, the Fluid Flow interface is employed to solve the problem of pipe loop flow. Furthermore, the Fluid Flow interface also includes several branches, such as Single-Phase Flow, Thin-Film Flow, Multiphase Flow, Porous Media and Subsurface Flow and so on, as shown in Fig. 9.6. As discussed above, the CGFB slurry in the pipe loop is assumed to be a single-phase flow. Therefore, the Single-Phase Flow branch of the Fluid Flow interface in COMSOL Multiphysics is chosen to simulate the flow of CGFB slurry. In addition, the Single-Phase Flow branch also contains several sub-branches, such as Laminar Flow, Turbulent Flow, Creeping Flow, Pipe Flow and so on (as illustrated in Fig. 9.6). Since the CGFB slurry flows in a pipe loop, thus in the present study, the

134

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Interface

Branch

Sub-branch Laminar Flow

Acoustics

Single-Phase Flow Thin-Film Flow

COMSOL Multiphysics

Fluid Flow Heat Transfer

Turbulent Flow Creeping Flow Pipe Flow

Multiphase Flow Porous Media and Subsurface Flow

Structure Mechanics Fig. 9.6 Partial simulation and the currently used modules included in COMSOL Multiphysics

sub-branch Pipe Flow is employed to provide the simulation platform for analyzing the pipe flow behavior of the CGFB slurry. The governing equations for the pipe flow model, which is used to describe the flowing behavior of the CGFB slurry in the pipe loop, are presented below (COMSOL 2014): 0 = −∇ p − f D

ρ u|u| + F 2dh

∇ · (Aρu) = 0

(9.2) (9.3)

where p is the pressure, f D is the Darcy friction factor, ρ is the CGFB slurry density, d h is the hydraulic pipe diameter, u is the fluid velocity in the tangential direction of the pipe curve segment, F is the volume force (such as gravity), A is the cross-section area of the pipe. The term f D in Eq. (9.2) can be expressed as follows (COMSOL 2014): 1/b  b b + f D,turb f D = f D,lam

b f D,lam

 1.143   10.67 + 0.1414 RHeBe He 64 = +  1.16 ReB ReB 1 + 0.0149 RHeBe ReB b f D,turb = 4 × 10a0 (ReB )−0.193

40,000 ReB    a0 = −1.47 · 1 − 0.146 · exp −2.9 × 10−5 He b = 1.7 +

(9.4)

(9.5)

(9.6) (9.7) (9.8)

9.2 Rheology and Flowability of Fresh CGFB

135

where ReB is the Reynolds number, which can be written in the following form: ReB =

ρdh v ρdi v = μB μB

(9.9)

where, d h is the hydraulic pipe diameter, which equals to the inner diameter of the loop pipe (d i ); v is the flow velocity of the CGFB slurry, which can be further calculated by Eq. (9.10); μB is Bingham plastic viscosity of the CGFB slurry. v=

π

Q  di 2

(9.10)

2

where Q is the volumetric flow rate. The item H e in Eq. (9.5) can be presented by (COMSOL 2014): He =

ρdh2 τB μ2B

(9.11)

where τ B is the yield stress of the CGFB slurry. Based on the rheometer tests, the yield stresses and Bingham plastic viscosities of the CGFB slurry samples can be obtained. With the help of other necessary data (such as volumetric flow rate, slurry density, pipe inner diameter, and so on), the Darcy friction factor (f D ), which is an input parameter for the pipe flow model, can be worked out according to the above corresponding equations. Furthermore, after importing all the input parameters (the input parameters will be demonstrated in the following section) into the pipe flow model in COMSOL Multiphysics, the model can compute by itself and provide the results of pressure distributions in the pipe loop visually. According to these pressure data, the pressure drop between any two locations in the pipe loop can be obtained.

9.2.2.3

Validation of the Model

According to the test loop system demonstrated in Fig. 9.5, the generalized geometric model of the pipe loop can be established, as shown in Fig. 9.7. The geometric model is imported into COMSOL Multiphysics to conduct simulating work. During the numerical simulation, the pressures at the four faces (faces A, B, C, and D in Fig. 9.7) are investigated. It should be pointed out that, the locations of these four faces in the pipe loop exactly correspond to where the four pressure tapings (4, 5, 6, and 7 in Fig. 9.5) locate. During the simulation, it is assumed that the viscosity (i.e., the experimentally obtained Bingham plastic viscosity) and density of the CGFB slurry do not change with time or temperature. The main input parameters for simulation are listed in Table 9.6.

136

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.7 Geometric model of the test loop

Table 9.6 Main parameters used for simulating the loop flow of the CGFB slurry samples Parameters

Values for the CGF slurry samples 1#

Density

(kg/m3 )

2#

3#

4#

1980

2035

2060

2098

Yield stress (Pa)

164.76

165.12

165.69

167.53

Bingham plastic viscosity (Pa s)

5.9581

7.1637

8.2779

9.3268

Inner diameter of the pipe (m)

0.12

0.12

0.12

0.12

Initial atmospheric pressure (Pa)

101,325

101,325

101,325

101,325

Volumetric flow rate (m3 /h)

80

80

80

80

Pumping pressure (MPa)

12

12

12

12

The pump mass flow rate (kg/s)

44

44

44

44

1# CGFB slurry is taken as an instance, its pressure distribution in the pipe loop can be obtained by the pipe flow model simulation, as shown in Fig. 9.8. According to the simulation results, the pressures at the faces A, B, C, and D (or the locations of pressure tapings 4, 5, 6, and 7) in the pipe loop can be obtained. Therefore, the partial pressure drops and total pressure drops of these CGFB slurry samples in the pipe loop can be calculated. The simulated results are compared with the experimentally measured ones, as demonstrated in Fig. 9.9. It can be found from Fig. 9.9 that the simulated results agree well with the tested ones. This indicates that the model (pipe flow model) simulation results are valid, and it is suitable for this model to simulate the pipe loop flow of CGFB slurry. Figure 9.9 also illustrates that both the total pressure drop and partial pressure drop increase with the increase in the solid concentration of the CGFB slurry. This is because increasing the solid concentration results in the increase of the slurry Bingham plastic viscosity (as demonstrated in Table 9.4). As a result, it becomes more difficult for the CGFB slurry with a higher value of solid concentration to flow in the pipe loop. That is to say, more pressure is needed to transport the slurry (with a higher value of concentration as well as viscosity) through the pipeline. This explains the reason why increasing

9.2 Rheology and Flowability of Fresh CGFB

Fig. 9.8 Pressure (unit: Pa) distribution of 1# CGFB slurry sample in the pipe loop

Fig. 9.9 Comparison between the loop test results and the model simulation ones

137

138

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

the slurry concentration leads to raising the pressure drop of the slurry during its flow in the pipe loop.

9.3 Mechanical Performance of Hardened CGFB As discussed previously, except for the function of disposing coal mine waste, the CGFB technology can also contribute to the ground control. This requires the CGFB structure to perform satisfactory mechanical performance. Figure 9.10 schematically demonstrates the role of CGFB in controlling roof settlement and ground subsidence in a coal mine. Once placed underground, the fresh CGFB behind the hydraulic support is required to harden quickly (this is because the hydraulic support needs to advance rapidly to keep high production efficiency), for letting the hardened CGFB structure to play its role of supporting the soft overlying strata. For this reason, mechanical performance is one of the most important parameters of a hardened CGFB structure. Triaxial compressive strength (TCS) and uniaxial compressive strength (UCS) are the two main criteria indicating the mechanical performance of a structure. In comparison with TCS test, UCS test is more convenient and cost saving, and can quickly determine the mechanical performance of cement-based structures (e.g., Hughes and Bahramian 1965; Fall and Pokharel 2010). In addition, UCS test can be incorporated into routine quality control programs at the mine (Fall and Samb 2009). For the current study, UCS test is used for evaluating the mechanical performance of CGFB. As a type of nondestructive testing (NDT) measurement, ultrasonic technique is also used to estimate the mechanical performance of cement-based materials, by measuring the ultrasonic pulse velocity (UPV) through them (e.g., Chotard et al. 2001; Demirboga et al. 2004; Trtnik et al. 2009). Hence in the present study, the mechanical performance of CGFB is also assessed by UPV test. As another NDT measurement, acoustic emission (AE) monitoring has been widely and intensively used in the field of civil engineering for structural health monitoring (e.g., Shiotani et al. 2001; Shiotani 2006; Carpinteri et al. 2011; Prem and Murthy 2016). During the past few years, AE monitoring technique has also Ground surface Main roof Immediate roof Coal seam Hydraulic support

Fig. 9.10 Effect of CGFB on controlling ground surface subsidence

Caved zone CGFB

9.3 Mechanical Performance of Hardened CGFB

139

Table 9.7 Mix proportion of the CGFB samples subjected to UCS and UPV tests Samples

CG (wt%)

FA (wt%)

Binder (wt%)

Solid content (wt%) (%)

1

46.5

20

10

76.50

2

47.5

20

10

77.50

3

48.5

20

10

78.50

4

47.5

19

10

76.50

2

47.5

20

10

77.50

5

47.5

21

10

78.50

been employed to assess the mechanical performance of cement-based materials (e.g., Abadelrahman et al. 2014; Benavent-Climent et al. 2012; Gong et al. 2014; Liu et al. 2013; Sagasta et al. 2016). The current study will also evaluate the mechanical performance of CGFB with the AE technique.

9.3.1 UCS and UPV of CGFB 9.3.1.1

Preparation of the CGFB Samples

Based on the mix proportion presented in Table 9.7, a total number of 60 CGFB samples (there are five kinds of mixes, and each mix is in triplicate and cured for four kinds of periods) with admixture of different fly ash contents are prepared. The required amount of coal gangue, fly ash, cement and water are mixed and homogenized in a mixer (as shown in Fig. 9.11a) until obtaining the desired CGFB mixtures. Afterwards, the produced CGFB mixtures are poured into curing cubes of 7 × 7 × 7 cm in length × width × height to form cubic samples, which are then cured in HSBY-60B standard curing chamber (Fig. 9.11b) at temperature of 20 ± 1 °C and for periods of 3, 7, 14 and 28 days.

9.3.1.2

Testing Methods

After the specific curing periods (3, 7, 14 and 28 days), the CGFB samples are subjected to the UPV tests. By taking advantage of the ultrasonic pulse method, the UPV testing (Fig. 9.11c) measures longitudinal P-wave velocities through the test media. During the UPV testing, the longitudinal P-wave velocity (V p ) through the CGFB is measured and recorded. The longitudinal P-wave velocity is calculated by the following equation: Vp = d/t

(9.12)

where d is the distance between the transmitter and receiver, t is the travel time.

140

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.11 Preparation and testing of the CGFB samples: a mixing; b cured CGB samples; c UPV tests; d UCS tests

After the UPV testing, the UCS tests are carried out on the same CGFB specimens (Fig. 9.11d) by using a computer-controlled mechanical press, which offers a normal loading capacity of 50 kN. During the compressive test, the displacement rate of the press is set at 0.5 mm/min.

9.3.1.3

Results of the Tests

The average values of UCS and UPV for each kind of CGFB mixture are taken as the representative ones to proceed with the following discussions. (1) Effect of FA content The UCS development of the CGFB samples with different dosages of fly ash versus curing periods is illustrated in Fig. 9.12. From this figure, it can be found that, on the condition of using the same solids content, UCS values of the CGFB samples increase with the increase of fly ash dosage. From Fig. 9.12 it can also be noticed that the UCS of CGFB increases with the elapsed curing time. This is ascribed to the fact that as the curing time elapses, more and more hydration products (C–S–H and ettringite) are generated. These products

9.3 Mechanical Performance of Hardened CGFB

141

Fig. 9.12 Effect of fly ash dosage on UCS values of CGB samples versus curing time

will fill the pore space within CGFB and improve bonding between coal gangue particles, leading to an increase in the strength of CGFB. The effect of fly ash dosage on the UPV values of the CGFB is also investigated, with the obtained results graphically displayed in Fig. 9.13. With the increase in the fly ash dosage for making CGFB, UPV values increase irrespective of the solid content or the curing time. This is because raising the dosage of fly ash results in increased number of hydration products for filling the pore structures (UPV in air is lower than that in any mineral skeleton such as rock and cementitious materials). (2) Effect of solid content Figure 9.14 demonstrates the effect of solid content on the evolution of UCS development versus curing time at the same fly ash content of 20 wt%. It is evident that the UCS values of CGFB samples increase with the rise of solid content. Besides, Fig. 9.14 also illustrates that the UCS of CGFB increases with the elapsing of curing age regardless of solid content. The effect of solids content on the UPV in CGFB samples is graphically presented in Fig. 9.15. It can be seen from this figure that, decreasing the solid content leads to the reduction of UPV values. (3) Correlation between UCS and UPV There are five groups of CGFB samples at four kinds of curing periods (3, 7, 14, and 28 days), so a total number of 40 values (20 for UCS and 20 for UPV) can be employed to find out the relationship between UCS and UPV. It is concluded that

142

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.13 Effect of fly ash dosage on UPV values of CGB samples versus curing time

Fig. 9.14 Effect of solids content on UCS values of CGFB samples versus curing time

9.3 Mechanical Performance of Hardened CGFB

143

Fig. 9.15 Effect of solids content on UPV values of CGB samples versus curing time

the UCS and UPV of the CGFB samples prepared at different solid content are in an exponential relationship with the highest the correlation coefficient (R), as exhibited in Table 9.8 and Fig. 9.16. It is observed from Fig. 9.17 that, the correlation between the UCS and UPV can be described by the following equation: UCS = 0.0159e0.0035·UPV

(9.13)

With the help of Eq. (9.13), the UCS of CGFB can possibly be obtained by conducting UPV test. As demonstrated in Table 9.8, although the above Eq. (9.13) has the highest correlation coefficient (0.959), it does not necessarily indicate the goodness-of-fit of this equation. Therefore, t- and F-tests are conducted to further validate the above Eq. (9.13). The t-test is used to determine the significance of the correlation coefficient (R), while the F-test is carried out to determine the significance of the regression, respectively by comparing the computed t and F values with the tabulated ones applying Table 9.8 Results of the curve fitting

Types of curve fitting

Correlation coefficient (R)

Linear

0.883

Logarithmic

0.868

Exponential

0.959

Power

0.953

144

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.16 Correlation of UCS and UPV for CGFB samples at different solid content

Fig. 9.17 Comparison of the calculated UCS values against the tested ones

null hypothesis. The results for the t- and F-tests are illustrated in Table 9.9. As shown in this table, the computed t value and F value are both greater than the tabulated ones, indicating that Eq. (9.13) is valid. Table 9.9 Results of t- and F-tests for Eq. (9.13) Equation (9.13)

Tabulated t value

Computed t value

Tabulated F value

Computed F value

±2.10

14.35

4.41

205.85

9.3 Mechanical Performance of Hardened CGFB

Infrared thermography

145

Compressive loading Rock mechanics testing apparatus

PCI-2 AE Monitor System CGFB Sample

AE sensor

Rubber band

Fig. 9.18 Schematic diagram of the UCS test coupled with AE monitoring and thermal infrared observation

Figure 9.18 shows the comparison between the calculated values of UCS and the measured ones. It can be found that the calculating and testing results agree well with each other, indicating that the exponential model is available to estimate the UCS values of CGFB with the measured data of UPV.

9.3.2 Mechanical Performance Based on AE 9.3.2.1

Preparation of the CGFB Specimens

Based on the mix proportion shown in Table 9.10, a total number of 72 CGFB samples are prepared: the CGFBs are cured at 4 kinds of temperatures (20, 50, 75, and 90 °C) for 3 kinds of periods (1d, 3d, and 7d) and tested by three kinds of experimental programs (measurement of elasticity modulus, UCS tests coupled with AE monitoring and thermal infrared observation), with each testing program conducted in triplicate. The required amount of CG, FA, binder, and water are blended and homogenized in a mortar mixer until obtaining the desired CGFB mixtures. Afterwards, the CGFBs are poured into curing moulds (10 cm × 10 cm × 10 cm Table 9.10 Mix proportion of the CGFB specimens

Solids content (wt%)

CG content (wt%)

FA content (wt%)

Binder content (wt%)

80.0

50.0

20.0

10.0

146

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

in length × width × height) to form cubic specimens. These CGFB specimens are cured in a standard curing chamber (type: HS-225) at different curing temperatures and for various curing time.

9.3.2.2

Testing Programs

Since the CGFB specimens are cured at 4 kinds of temperatures (20, 50, 75 and 90 °C) for 3 kinds of periods (1, 3 and 7 days), there are a total number of 12 curing processes. When each curing process of the CGFB specimens are completed with the required age and temperature, two sets of CGFB specimens are taken out from the curing chamber. One set of the CGFB specimens are subjected to the measurement of elasticity modulus in triple, and the average value is used. The other set of the CGFB samples are subjected to the UCS tests also in triple. The rock mechanics testing apparatus (type: TAW-2000), which has the maximum load of 2000 kN and can display the stress–strain relation in real time, is used for the UCS tests. During the process of UCS test, the CGFB sample is subjected to AE monitoring and thermal infrared observation simultaneously. PCI-2 AE Monitor System is used to record and display the characteristic parameters of the AE signals. An infrared thermography is used to record the average infrared radiation temperature (AIRT) evolution of the CGFB specimen during the UCS test. The camera used is Fluke Ti400 with the temperature range between −20 and 1200 °C, resolution of 320 × 240 pixel, and sensitivity of 0.05 °C. Figure 9.18 schematically illustrates the UCS test on a cubic CGFB sample coupled with AE monitoring and thermal infrared observation in the current study. Before the test, two thin films of Vaseline are coated on the end surfaces of the two AE sensors, respectively, in order to ensure favorable contact of the sensors with the CGFB sample. A rubber band is used to fix the two AE sensors and made them contact with the CGFB sample all the time during the test. During the procedure of testing and monitoring, the compressive loading rate is kept at 0.2 mm/min, the monitoring sampling rate of AE signals is set as 1 MSPS (i.e., million samples per second), and the gain for preamplifier is 40 dB. In order to minimize the interference of noise, the threshold value is set at 55 dB.

9.3.2.3

Results and Discussions

(1) UCS and elastic modulus Figure 9.19 illustrates the effect of curing temperature (20, 50, 75 and 90 °C are selected) on the UCS development in CGFB. As expected, the UCS of CGFB increases with the curing time. Besides, it can also be found that a higher curing temperature leads to a higher UCS in the CGFB structure. Especially in the very early age (0–3 days), the UCS of CGFB significantly increases with the curing temperature. This is ascribed to the fact that during this curing period, the effect of a

9.3 Mechanical Performance of Hardened CGFB

147

Fig. 9.19 Effect of curing temperature on the evolution of UCS of CGFB versus curing time

higher curing temperature is highly remarkable to accelerate the binder hydration progress, which generates a large amount of hydration products to increase the UCS of CGFB. From the 3rd to 7th day, the binder hydration rate decreases, thus the development of UCS in the CGFB also slows down. Figure 9.19 also shows that when the curing temperature increases from 20 to 50 °C and then to 75 °C, the UCS of CGFB increases notably. When the curing temperature increases from 75 to 90 °C, the UCS values of the CGFB specimens cured for 1d and 3d increase slowly, but the UCS of CGFB cured for 7d decreases. This is due to the fact that, when the curing temperature increases in an appropriate range, it can exert a positive influence on the UCS development by promoting the binder hydration course. However, if the curing temperature is too high, it would exert a negative effect on the UCS development in CGFB, by generating excessive thermal stress on the CGFB and thus destroying its structure. From 0 to 3rd day, the binder hydration progress is strong and very sensitive to the curing temperature. Raising the curing temperature during this period can effectively speed up the binder hydration rate and thus contribute to the UCS development. Thereafter (3–7d) the binder hydration process slows down and the effect of curing temperature on it weakens. The positive contribution of increasing the curing temperature to the UCS development is not able to conquer the UCS reduction induced by the negative effect of too high curing temperature. It can be expected that there must exist a certain threshold for the curing temperature, and if the curing temperature exceeds the threshold, the UCS of CGFB would decrease. Figure 9.20 shows the influence of curing temperature on the elastic modulus of the CGFB specimens. As expected, the elastic modulus of CGFB develops with the curing time. Similar results (with Fig. 9.19) can be seen in Fig. 9.20 that a higher

148

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.20 Effect of curing temperature on elastic modulus of CGFB versus curing time

curing temperature is associated with a higher value of elastic modulus of the CGFB specimen, except that a too high curing temperature (90 °C) exerts a negative effect on the elastic modulus. (2) Stress evolution and cumulative count of AE Taking the CGFB specimens cured for 7 days as examples, Fig. 9.21 displays the effect of curing temperature on the stress evolution of the CGFB structures versus time and the cumulative count of AE during the UCS tests. All these four figures (a, b, c and d) illustrate almost the same results of regularity: the cumulative counts of AE achieve the maximum number when the stress reaches its peak value. The whole stress evolutionary process versus time in the CGFB under UCS testing can be divided into 5 stages (Fig. 9.22): (1) initial compaction stage (o–a); (2) linear elastic stage (a–b); (3) elastoplastic stage (b–c); (4) plastic stage (c–d); (5) failure stage (d–e). Due to the uniaxial compression induced closure of the pores and micro cracks in the CGFB specimen, almost no AE events occur during the initial stage of loading. In the linear elastic stage, since few cracks are generated inside the CGFB structure, very few AE signals can be recorded. When it comes to the elastoplastic stage, the volumetric strain of the CGFB structure changes from compression to expansion before failure and the AE counts begin to increase with the increase of stress. Afterwards, with the continuous increase of the stress and when it reaches to the maximum, the continuous generation of new fracture and extension of old cracks result in the failure of the CGFB structure, and meanwhile, the cumulative counts of AE also reach the peak at the end of the plastic stage. And then during the failure stage, the CGFB sample still has bearing capacity to a certain extent, hence a

9.3 Mechanical Performance of Hardened CGFB

a) curing temperature: 20 °C

c) curing temperature: 75 °C

149

b) curing temperature: 50 °C

d) curing temperature: 90 °C

Fig. 9.21 Effect of curing temperature on stress evolution and cumulative count of AE in CGFB cured for 7d

Fig. 9.22 Development of stress in the CGFB structure versus time during UCS testing

150

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

definite number of AE events can be recorded. After that, the cumulative counts of AE gradually decrease to zero. It can also be noticed from Fig. 9.22 that, the higher the peak stress (i.e., UCS) of the CGFB is, the more the maximum cumulative counts of AE are. This is because more energy is released during the uniaxial compressive deformation and damage of a CGFB structure with a higher value of stress, consequently, more AE events occur during this process. The mutation of AE cumulative counts is a sign of the failure of CGFB samples. The results shown in Fig. 9.22 indicate that the CGFB structures have different AE features at different loading stages. Therefore, the evolution of AE signals can be used to predict if the CGFB structures are stable under compressive load and when they are damaged by destabilization.

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB Due to the composition of CGFB, the binder used will chemically react with water. This chemical reaction is called binder hydration, generating hydration products, consuming water and releasing heat. The generation of hydration products results in the hardening and strength gain of the CGFB, and the consumption of water leads to the decrease of pore water pressure or the suction development within the CGFB, while the generation of heat contributes to the temperature increment of the CGFB. As discussed before, in addition to the preparation of CGFBs for disposing mine waste, another significant application is to backfill mined-out areas with them for the purpose of ground subsidence control. This requires the CGFB structures to possess satisfactory mechanical performance, which is dramatically affected by the chemical (e.g., binder hydration), thermal (e.g., self-heating, heat exchange with surroundings) and hydraulic (e.g., suction development) factors. The interaction of these three factors is indicated as follows: • Chemical (C)–Thermal (T) interaction. Binder hydration releases heat to increase the CGFB temperature, while the temperatures of CGFB and its surroundings affect the rate of binder hydration. • Chemical (C)–Hydraulic (H) interaction. Binder hydration process consumes water to contribute to the suction development in the CGFB. However, it should be pointed out that the suction development exerts insignificant influence on the binder hydration process. • Thermal (T)–Hydraulic (H) interaction. Temperature variation affects the suction evolution of the CGFB, but the suction development exerts insignificant impact on the thermal factor. The CGFB, which is a type of cement-based geo-materials, is subjected to the coupled thermal (T), hydraulic (H) and chemical (C) factors. These factors influence each other, together exerting THC coupled effects on the mechanical properties of CGFB. Hence, the objective of this section is to develop a numerical model to

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB

151

analyze and predict the THC behavior of CGFB. The numerical simulation results of the model are compared with the data from an experimental study.

9.4.1 Development of the THC Model (1) Thermal equations CGFB mixture is a type of self-heating composites, due to the fact that the binder used to prepare the CGFB reacts with the mixed water to release heat. Within the CGFB, heat convection due to fluid flow and heat radiation are insignificant, hence these two ways of thermal transfer are not considered in the present study. In consideration of self-heating and thermal conduction between the CGFB and its surroundings, the following equation can be obtained (COMSOL 2014): (ρC)eq

  ∂T − ∇ · keq ∇T = Q H ∂t

(9.14)

where, (ρC)eq is the equivalent volumetric heat capacity of the CGFB (a solidfluid composite) at constant pressure; T is the temperature; t is the time; k eq is the equivalent thermal conductivity of the CGFB; and QH is the term representing heat generation, which can be expressed as follows: Q H = C B · qh

(9.15)

where, C B is content of the binder used for preparing the CGFB (kg/m3 ); qh is the heat generated by binder hydration per unit time and mass (W/kg), which can be presented as below (De Schutter and Taerwe 1995):  

 EA 1 1 − qh = qm · c · sin(π · α)a · exp(−b · α) · exp R Tr Tc

(9.16)

where, qm is the maximum heat generation rate at the temperature of 20 °C; α is the degree of binder hydration; a, b, and c are experimentally determined constants; T r is the reference temperature (i.e., 20 °C); T c is the temperature of the CGFB; and E A is the apparent activation energy, which is related with T c (Jin Sang and Kwang 2001): if T c ≥ 20 °C, E A = 33,500 J/mol; while if T c ≤ 20 °C, E A = 33,500 + 1470 × (293 − T c ). (2) Hydraulic equations The hydraulic properties of a CGFB structure mainly include hydraulic conductivity, suction, and (positive) pore water pressure. In the present study, we only focus on the suction development, which is induced by the binder hydration (i.e., the binder chemically reacts with water) within the CGFB. The suction of CGFB can be calculated

152

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

by using the following formula derived from the Kelvin equation (Abdul-Hussian and Fall 2011): ψ =−

RT ln h C vwo wv

(9.17)

where ψ is the total suction (kPa) of the CGFB, R is the universal (molar) gas constant [8.31,432 J/(mol K)], vwo is the specific volume of water [(1/ρ w ) (m3 /kg)], wv is the molecular mass of water vapor (18.016 g/mol), and hC is the relative humidity of the CGFB. CGFB is a cement-based material as similar to concrete, therefore the expression developed to calculate the relative humidity of concrete is applied to CGFB (Zhang and Yang 2011): h C = 100 + X · exp(Y · S)

(9.18)

where X and Y are fitting parameters; and S is the CGFB saturation, which can be expressed as follows (Hansen 1986): S=

Vcw + Vgw Vcw + Vgw + Vcs

(9.19)

Vcw = ϕ0 − 1.32(1 − ϕ0 )α

(9.20)

Vgw = 0.6(1 − ϕ0 )α

(9.21)

Vcs = 0.2(1 − ϕ0 )α

(9.22)

where V cw is the capillary water content (m3 /m3 ), V gw is the gel water content (m3 /m3 ), V cs is the chemical shrinkage (m3 /m3 ), ϕ 0 is the initial porosity of the CGFB or the porosity of the CGFB when it begins to harden from the paste phase, α is the binder hydration degree. (3) Equations for chemical reactions The main chemical reaction in CGFB is binder hydration, and the rate of the binder hydration process can be described by the binder hydration degree, which is expressed in the following forms (Kjellsen and Detwiler 1993; Poole et al. 2007; Schindler and Folliard 2005):   τ β α = αu · exp − te αu =

1.031 · r + 0.5 pFA + 0.3 pS 0.194 + r

(9.23) (9.24)

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB

t te =

−

e

EA R



1 Tc

− T1r

153



dt

(9.25)

0

where β is the shape parameter for the binder hydration, τ is the time parameter of the hydration at the reference temperature, α u is the ultimate degree of the binder hydration, t e is the equivalent age, r is the water-to-binder ratio, and pFA and pS are the proportions of cement replaced by fly ash and slag, respectively. As demonstrated above, the thermal equations (T) and hydraulic equations (H) are correlated by temperature, the thermal equations (T) and chemical equations (C) are interrelated by the degree of binder hydration, and the hydraulic equations (H) and chemical equations (C) are linked with each other by the binder hydration degree. Furthermore, the degree of binder hydration is an expression of temperature, therefore the three groups of equations can be coupled with each other by temperature. The coupled THC model is incorporated the numerical program COMSOL Multiphysics to conduct numerical simulation.

9.4.2 Validation of the Developed THC Model (1) Laboratory experiment A laboratory study investigating the temperature and suction evolutions of the CGFB versus time is applied to validate the corresponding results predicted by the developed THC model. The mix proportion of the materials for preparing the CGFB mixture is presented in Table 9.11. The CG used is from a coal mine in Northwest of China, and the FA used is from a power plant near the coal mine. These materials are mixed uniformly until obtaining a homogeneous paste, and then the plastic CGFB paste is poured into a cubic container, which is impermeable and has the dimension of 0.3 × 0.3 × 0.6 m3 in length × width × height. The container possesses enough stiffness to restrict the deformation of the contained CGFB. A tensiometer (with the measurement accuracy of 0.1 kPa) and a temperature transmitter (measurement accuracy is 0.01 °C) are fixed to investigate the variations of suction and temperature within the CGFB, respectively. Figure 9.23 shows the schematic diagram of the designed experimental apparatus. As shown in Fig. 9.23, the container is uncovered, leaving the space (0.5 m thick) for the CGFB to contact with air. This is to imitate the practical mine backfill situations in the abovementioned coal mine that the surface of the backfilled CGFB is exposed to the environment, while the sides and the bottom are surrounded by Table 9.11 Composition of the CGFB

Coal gangue (wt%)

Fly ash (wt%)

Binder (wt%)

Water (wt%)

49

20

10

21

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

0.15 m

Data collector

0.25 m

0.15 m

0.05 m

154

Temperature transmitter

0.15 m

Tensiometer CGFB 0.15 m

Container 0.3 m

Fig. 9.23 Schematic diagram of the experimental apparatus

the adjacent steel plates. The data observed by the temperature transmitter and tensiometer are collected and displayed by the data collector. A thermometer is used to monitor the temperature of the surrounding environment in the laboratory. The room temperature is monitored to be nearly constant (±1 °C). (2) Model validation The following procedure illustrated in Fig. 9.24 is designed to validate the developed THC model. Figure 9.25 demonstrates the geometry of the simulated model. The (external) boundary condition of this model is set as thermal insulation. The developed THC model is used to predict the CGFB temperature development versus curing time, and the prediction results are compared with the corresponding data obtained by experiment investigation. The main input parameters for validating the developed THC model can be found in Table 9.12. The comparison is graphically illustrated in Fig. 9.26. It is observed from this figure that, the temperature development at point PT within the CGFB structure versus time simulated by the developed THC model agrees satisfactorily with that obtained by the experiment investigation, except for some temperature differences probably due to the varied boundary conditions during the test. Figure 9.26 also demonstrates the temperature distribution within the CGFB structure when the temperature at point PT reaches its peak value. The results revealed by Fig. 9.26 can evidently manifest

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB Fig. 9.24 Flow chart for validating the developed model

PT PS

155

Input parameters for the initial and Geometry AND boundary conditions (as listed in Table 2), Initial time ti=0 CGFB curing process, Ending time te=7d

Simulating the temperature evolution at point PT versus time

Simulating the temperature evolution at point PS versus time Calculating the suction evolution at point PS versus time with the help of Eq.s (4)-(12)

Contrasting with the experimental investigating results (as shown in Figures 4 and 5)

that the proposed model possess the capability to simulate the temperature evolution of the CGFB. After validating the ability of the developed model for simulating CGFB temperature development, this model is then applied to predict the suction evolution of the CGFB. The comparison between the model prediction and the experimental observation is shown in Fig. 9.27. From this figure, it can be observed that the suction evolution versus curing time at point PS within the CGFB predicted by the THC model is in favorable agreement with the experimentally observed results. This indicates that the proposed THC model is also capable of predicting CGFB suction development.

605 mm

310 mm

550 mm

Fig. 9.25 Geometry of the simulated model

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

50 mm

156

Air (TE) CGFB (T0) Container (TB) 300 mm

9.4.3 Application of the Developed Model After the validation of the developed THC model, it is utilized to simulate some issues in terms of engineering applications. (1) Effect of binder hydration degree on CGFB temperature development (T–C coupling) Water-to-binder ratio (w/b) is one of the most important design specifications for CGFB, since it can affect the degree of binder hydration according to Eq. (9.24). The influence of the chemical field (C, the binder hydration degree) on the thermal filed (T, the temperature development) within the CGFB is discussed. Based on Eq. (9.24), by changing the contents of coal gangue and binder, and keeping the contents of fly ash and water unchanged, three kinds of water-to-binder ratios and thus binder hydration degrees for the CGFB are obtained, as shown in Table 9.13. The input parameters listed in Table 9.12 and the varied α u (0.88, 0.91 and 0.93) are implemented into the developed model (with its geometry shown in Fig. 9.25) for numerical simulation. Figure 9.28 reveals the effect of binder hydration degree on the temperature development within the CGFB. As expected, the temperature of the CGFB increases with the increase in the ultimate binder hydration degree. This is because a higher ultimate degree of binder hydration indicates that more binder can finally hydrate, in turn generating more heat to motivate the temperature development in the CGFB. Regardless of the ultimate degree of binder hydration, after curing for almost 1 day,

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB Table 9.12 Input parameters for validation of the developed model

157

Parameters

Values

Initial CGFB temperature T 0 (°C)

12.0

Initial temperature of the container T B (°C)

12.0

Environment temperature T E (°C)

12.0

CGFB thermal conductivity [W/(m*K)]

2.85

CGFB specific heat [J/(kg*K)]

1200

Thermal conductivity of the container [W/(m*K)]

19

Specific heat of the container [J/(kg*K)]

1464

E A (apparent activation energy) (J/mol)

Varies

CGB density

(kg/m3 )

2000

Water density (kg/m3 )

1000

Binder content (kg/m3 )

600

α u (ultimate binder hydration degree)

0.91

β (hydration shape parameter)

30

τ (hydration time parameter) (h)

0.6

qmax,293 (W/kg)

2.19

a

0.667

b

3.0

c

2.6

X

−8.63e9

Y

−23.796

Some of the data in Table 9.12 are based on the studies (De Schutter and Taerwe 1995; Zhang and Yang 2011; Poole et al. 2007; Schindler and Folliard 2005; Bouguerra et al. 1997; Khan 2002; Yoon et al. 2014; Ukrainczyk and Matusinovi´c 2010; Parrott 1988)

the temperature (at point PT ) within the CGFB increases to the peak, due to generated heat from binder hydration. Afterwards, the binder hydration process slows down, and thus the heat generation and accumulation in the CGFB is unable to conquer the heat diffusion from inside the CGFB to its surroundings (due to the temperature difference induced heat transfer). This leads to the CGFB temperature decline. In practice, it may be believed that increasing the binder percentage in the CGFB can always lead to temperature increment and thus the strength increase, regardless of the w/b ratio of the CGFB. However, the data in Table 9.13 and Fig. 9.28 together illustrate that the w/b ratio plays a critical role in the temperature development within the CGFB. The results indicate that a CGFB with a higher w/b ratio due to the lower binder content show a higher temperature. This is due to the fact that the decrease of binder percentage is associated with the increase of water-to-binder ratio for a given water content, which is a key factor determining the binder hydration process and thus the temperature rise. This finding suggests that, without enough water, simply

158

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.26 Comparison between THC model simulation and experimental investigation results of temperature

Fig. 9.27 Comparison between THC model prediction and experiment observation results of suction

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB

159

Table 9.13 Mix proportion of the CGFB used for THC model application CG (wt%)

FA (wt%)

Cement (wt%)

Water (wt%)

w/b

αu

44

20

15

21

0.6

0.88

49

20

10

21

0.7

0.91

52.75

20

6.25

21

0.8

0.93

Fig. 9.28 Effect of binder hydration degree on the temperature development in CGFB

raising the binder percentage would not necessarily be beneficial for the CGFB with respect to temperature induced strength increase. (2) Effect of binder hydration degree on CGFB suction development (H–C coupling) Three kinds of ultimate binder hydration degrees (0.88, 0.91, and 0.93 as tabulated in Table 9.13) are selected to conduct the current analysis, which aims to reveal the effect of the chemical field (C, the final degree of binder hydration) on the hydraulic filed (H, the suction development) within the CGFB. The impact of binder hydration degree on the suction evolution of the CGFB versus curing time is shown in Fig. 9.29. From Table 9.13 and Fig. 9.29 it is observed that at the same water content, the suction in the CGFB increases with increasing the ultimate degree of binder hydration. The reason has been pointed out above, a higher ultimate binder hydration degree implies the binder reacts more sufficiently, consuming more water to increase the suction. Figure 9.29 also reveals that the suction (at point PS ) of the CGFB increases rapidly at the early age (0–1d), and then the suction increment slows

160

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.29 Effect of binder hydration degree on the CGFB suction evolution with time

down until after about 3 days, the suction no longer increases but remains unchanged. The reason is that the CGFB suction development with curing time is in accordance with the binder hydration progress. During the earlier period, the binder hydration is violent, consuming a large amount of water to make the suction develop at a fast rate. Thereafter, the binder hydration slows down, hence less water is consumed, making the slow development of suction. Eventually, after curing for almost 3 days, the binder hydration process basically finishes, that is to say, no more water is consumed to contribute to the suction development. (3) Effect of temperature on CGFB suction development (T–H coupling) 1. Effect of initial CGFB temperature on its suction development Three types of initial CGFB temperatures (1, 20 and 40 °C) are chosen in this simulation. Figure 9.30 reveals the influence of the initial CGFB temperature on the suction evolution of point PS in the CGFB versus curing time. As expected, the initial CGFB temperature indeed exerts a significant effect on its suction development, and the CGFB suction increases with the increase of the initial temperature. It is generally acknowledged that temperature is a vital factor affecting the rate of chemical reaction. A higher temperature situation is associated with a higher chemical reaction rate. Hence, increasing the initial CGFB temperature can significantly make the binder hydrate more rapidly, and thereby consuming more water to promote suction development.

9.4 Thermo-Hydro-Chemical Coupled Behavior of CGFB

161

Fig. 9.30 Effect of initial CGFB temperature on the suction development in CGFB

2. Effect of environmental temperature on CGFB suction development In addition to the temperature of the CGFB itself, the temperature of the environment around the CGFB also affects its suction development. Therefore, this simulation aims to reveal the effect of the environment temperature (5, 20 and 35 °C are selected) on the CGFB suction evolution with curing time. As illustrated in Fig. 9.31, increasing the environment temperature contributes to the CGFB suction increment. The reason has been stated above, a high temperature is beneficial to the binder hydration, and thus the self-desiccation, associating with the consumption of water and hence the development of suction.

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB Since the CGFB is designed and utilized to alleviate the pollution resulted from surface discharge of solid wastes, it should be ensured that the backfilled CGFB materials exert the lowest negative impact on the underground circumstances. This requires the CGFB possessing favorable environmental performance, which is significantly related to the thermal, mechanical and hydraulic (e.g., water drainage) behaviors of CGFB, as well as the interaction and coupling of these behaviors.

162

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.31 Effect of environmental temperature on the suction development in CGFB

9.5.1 Experimental Programs 9.5.1.1

Test Device Developed for CGFB Column Experiments

An experimental device (as shown in Fig. 9.32) is designed and fabricated to investigate the behavior of CGFB during its curing under pressure. A transparent acrylic column (Fig. 9.33) that is 15.0 cm in internal diameter, 16.0 cm in external diameter and 23.0 cm in height is used to place the freshly prepared CGFB sample, and the acrylic column is inserted into a steel column (7 in Fig. 9.32) that is 16.0 cm in internal diameter, 17.0 cm in external diameter and 23.0 cm in height. An internal circular slot, which is 17.0 cm in diameter and 0.5 cm in depth, is cut in a steel portable plate (13 in Fig. 9.32) for the placement of the steel column. The portable plate, which is embedded in a steel base plate (14 in Fig. 9.32), is designed and made for the convenience of casting the CGFB sample (9 in Fig. 9.32). Before the CGFB casting, the portable plate is pulled out from the base plate, and after completing the casting, the portable steel plate along with the CGFB sample and the two (steel and acrylic) columns are inserted into the base plate. Prior to pouring the CGFB into the acrylic column, a discoid permeable stone (12 in Fig. 9.32), which is 15.0 cm in diameter and 0.5 cm in thickness, is placed at the acrylic column bottom (above the circular slot of the portable steel plate). Thereafter, the CGFB sample is filled into the acrylic column, until the CGFB height reaches to 20.0 cm, and the top freeboard of the acrylic column (with the thickness of 2.5 cm) is left uncovered. This is in accordance with real underground mining conditions.

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB

1- Compression bar 5- Balance plate 9- CGFB sample

2- Support rod

3- Top plate

6- Axial loading piston

10- Lateral pressure sensor

13- Portable plate

14- Base plate

163

4- Spoke-type pressure sensor

7- Steel column

8- Acrylic column

11- Temperature sensor

15- Aqueduct

16- Bottle

12- Permeable stone 17- Steel leg

Fig. 9.32 Schematic diagram of the developed experimental device

A circular through-hole, which is 2.2 cm in diameter and used for water drainage (from inside the CGFB sample), is opened at the center of the portable plate, and beneath this through-hole, another circular through-hole, which has the same inner diameter (2.2 cm) with the one in the portable plate, is opened at the center of the base plate. An aqueduct (15 in Fig. 9.32) is used to link the through-holes with a bottle (16 in Fig. 9.32), which is used to contain the drained water from the CGFB sample. Until no more water comes out from the CGFB sample, the already gathered water in the bottle is weighed by an electronic scale. Two through-holes are respectively opened at one side of the steel and acrylic columns, for fixing a small tip lateral pressure sensor (10 in Fig. 9.32, measurement range of the lateral pressure sensor is from 0 to 2 MPa) prior to casting the plastic CGFB specimen, and another two through-holes are at the opposite side of the two columns, for attaching a small tip temperature sensor (11 in Fig. 9.32, measuring range is from −40 to 180 °C). The lateral pressure sensor is applied to investigate the lateral pressure evolution of the CGFB sample, while the temperature sensor is employed to monitor the temperature development in the CGFB. An axial loading

164

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.33 Transparent acrylic column

piston (6 in Fig. 9.32), which is linked with a compression bar (1 in Fig. 9.32, the compression bar is connected with an air cylinder, and the air cylinder is further connected with an air compressor, which can provide 0.7 MPa air pressure in maximum), is vertically pressed down onto the top surface of the columnar CGFB sample by using high air pressure. Between the compression bar and the loading piston, there is a spoke-type pressure sensor (4 in Fig. 9.32) and a steel balance plate (5 in Fig. 9.32). The pressure sensor is implemented to tell the data of the air pressure on the balance plate supplied by the air compressor, and the vertical load on the columnar CGFB sample can be calculated by using the following equation: PV =

G bp + G lp + PA SC

(9.26)

where PV is the vertical load on the CGFB sample, Gbp and Glp are the weights of the balance plate and the loading piston, respectively, S C is the cross-sectional area of the columnar CGFB sample, and PA is the air pressure value provided by the air compressor. The data observed and captured by the lateral pressure sensor, temperature sensor and spoke-type pressure sensor are all collected by a PC-controlled recording instrument, which also has a screen for displaying curves of the lateral pressure, temperature and vertical pressure in real time. A thermohygrometer is applied to monitor the changes in room temperature and RH during curing the CGFB sample.

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB

165

Table 9.14 Mix proportions of the CGFB samples (1#–3#) Solids content (wt%)

CG content (wt%)

FA content (wt%)

1#

80.0

50.0

24.0

2#

80.0

50.0

22.0

8.0

3#

80.0

50.0

20.0

10.0

9.5.1.2

Binder content (wt%) 6.0

Preparation of the CGFB Samples

Coal gangue, fly ash, cement, and water are mixed in a cement mortar mixer to prepare the CGFB mixtures. The mixing time is set as 3 min for each barrel of mixes, and two barrels of mixes are needed for preparing one columnar CGFB sample that is 15.0 cm in diameter and 20.0 cm in height. Two mixers work simultaneously to make the mixes and after 3 min, the two barrels of mixes are poured into the column that is used to place the CGFB sample. It takes no more than 5 min to complete the whole process of one CGFB sample preparation and placement. This experimental study tests three kinds of CGFB samples, and the test period for each sample is 30 days. The three types of CGFB mixtures are prepared according to the mix proportions listed in Table 9.14.

9.5.1.3

Experimental Testing Programs

(1) Water drainage test The bottle (16 in Fig. 9.32) is applied to collect the drained water from inside the tested CGFB samples, and an electronic scale is utilized to measure the weight of the drained water. The water drainage (%) can be calculated as: Water drainage =

Drained water × 100% Total water used to prepare the CGFB

(9.27)

(2) Lateral pressure investigation The lateral pressure sensor (10 in Fig. 9.32) is used to measure the lateral pressure of the CGFB, and the measurement lasts for 30 days. After curing for one day, a vertical mechanical stress is loaded on the tested CGFB by the compression bar (1 in Fig. 9.32). That is to say, from the second day on, the lateral pressure sensor investigates the lateral pressure of the CGFB under vertical pressure. (3) Temperature measurement The development of temperature in the CGFB samples versus curing time (30 days) is monitored by the temperature sensor (11 in Fig. 9.32). Similarly, a vertical load is acted on the tested CGFB after 1 day of curing age.

166

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

(4) UCS test A computer-controlled mechanical press is utilized to conduct UCS tests on the CGFB samples (in undrained condition) which are 10 × 10 × 10 cm3 in dimension and prepared according to Table 9.14. The CGFB samples are cured for 1, 3 and 7 days and at curing temperatures of 20 and 50 °C. A total number of 54 CGFB samples are prepared, and for each kind of CGFB (e.g., 1# sample cured for 1 day at the curing temperature of 20 °C), three samples are tested, and the average data are collected.

9.5.2 Results and Discussion 9.5.2.1

Water Drainage

The water drainage of the CGFB sample in drained condition is investigated. The water starts to discharge from the CGFB immediately after its casting, and about three hours later, no more water flows out from the CGFB sample. Figure 9.34 demonstrates the water drainage results of the CGFB samples (1#, 2# and 3# with varied mix proportions listed in Table 9.14). The water drainage (percentage by weight, i.e., wt%) data for the CGFB samples 1#, 2# and 3# are 8.25, 8, and 7.77%, respectively. This is attributed to the fact that the three CGFB samples have different mix proportions, especially cement-to-fly ash (PC/FA) ratio. It is detected that a higher PC/FA ratio leads to a lower percentage of water drainage. This is because a higher PC/FA ratio is associated with more cement hydration reactions, which not only consume more evaporable water but also generate more hydration products to refine the pore structure within the CGFB. The refinement of the pore structure

Fig. 9.34 Effect of PC/FA ratio on water drainage of the CGFB

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB

167

Fig. 9.35 Effect of curing temperature on water drainage of the CGFB

effectively hinders the water seepage from inside the CGFB. This is also associated with favorable environmental performance of the CGFB. Taking the CGFB sample 3# as an example, Fig. 9.35 reveals the variation in the water drainage of the CGFB specimens cured at different temperatures (20 and 50 °C) but the same humidity (95%). This figure states that the curing temperature significantly affects the water drainage of the CGFB. A higher curing temperature is associated with a lower content of drained water. This is due to the reality that, a higher curing temperature can dramatically accelerate the binder hydration progress. Thereby, the amount of hydration products (e.g., C–S–H, Ca(OH)2 ) increases with the curing temperature. This is in favor of refining the pore structures within the CGFB, and thus retaining more water in the CGFB (namely, less water is drained out).

9.5.2.2

Evolution of Temperature

Figure 9.36 illustrates the evolutions of the temperatures of the three CGFB samples with curing time (30 days) in drained condition. For all the three samples, their temperatures rapidly reach the highest value around 1 day of curing time after casting. This is due to the fact that the heat of binder hydration contributes to the temperature development. From this time point (1 day) up to 3 days, the heat of hydration gradually dissipates and thereby, the temperatures of the three CGFB samples start to decrease, because of the heat exchange between the CGFBs and their surrounding environment. From Fig. 9.36 it is observed that the highest temperature develops in the CGFB sample 3#, while the sample 1# has the lowest temperature. This is due to the fact that the sample 3# has the highest PC/FA ratio, and a higher PC/FA ratio signifies that more cement hydrates and thus generates more heat to contribute to a higher temperature development. From the 4th day to 30 days, the temperatures of the three CGFB

168

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.36 Temperature evolutions of the CGFB samples versus curing age

samples almost have no changes. This is ascribed to the thermal equilibrium between the CGFBs and the environment. However, it is noticed that the CGFB sample 3# still possesses the highest temperature among the three samples. As discussed before, a higher PC/FA ratio is associated with more cement hydrations, which generate more hydration products to reduce the porosity of the CGFB. This is in favor of the storage of hydration heat that develops within the CGFB. It should be pointed out that, after 1 day of curing time, a sudden (and later on a constant) vertical pressure is loaded onto the CGFB. It can be investigated that this mechanical load has no significant influence on the development of temperature within the CGFB.

9.5.2.3

Evolution of Lateral Pressure

Lateral pressure developments of the three different CGFB samples versus curing time in drained condition are illustrated in Fig. 9.37. After curing for 1 day, a vertical stress that is 800 kPa is pressed onto the CGFB, and this mechanical load is maintained for 2 h. This is to simulate a sudden roof falling on the backfill in field. Afterwards, the vertical load is changed to a constant value of 80 kPa and kept on until 30 days. From Fig. 9.37, it can be seen that the lateral pressures for all the three CGFB samples increase with the curing age, and there is a sudden increase in the lateral pressure when loading the vertical stress after 1 day of curing. This is because after curing for 1 day, the CGFB mixtures are not completely hardened but still plastically flowable. The equation below developed for concrete can be employed to explain the lateral

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB

169

Fig. 9.37 Lateral pressure evolutions of the CGFB samples versus curing age

pressure of fresh CGFB (Schjödt 1955; Levitsky 1973): PC = λC γ H

(9.28)

where λC is the relationship between vertical and horizontal pressure, γ is the concrete specific weight, and H is the concrete height. Since the CGFB is also a kind of cement-based materials like concrete, the increase in the lateral pressure of a plastic CGFB mixture induced by a sudden vertical load can be explained by the following equation:   P = λ γ  H  + PV

(9.29)

where P is the lateral pressure of the plastic CGFB, λ is the parameter describing the relation between vertical and lateral pressure, γ  is the CGFB specific weight, H  is the CGFB height, and PV is the vertical load. From Fig. 9.37, it can be found that different CP/FA ratios have no significant influence on evolution of the lateral pressure in the CGFB samples in the first day. Nevertheless, when the constant vertical stress (80 kPa) is loaded, the lateral pressures of the CGFBs increase with the curing time and a higher PC/FA ratio leads to a higher lateral pressure development in the CGFB. After 30 days of curing time, the monitored lateral pressure values for the CGFB samples 1#, 2# and 3# are 0.31, 0.61 and 0.99 MPa, respectively. The increase in the lateral pressures of the CGFBs with curing age can be explained by creeping effect. The CGFB samples are placed in

170

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

the steel column which is nonflexible and rigid. After curing for 1 day, the already hardened CGFB mixture is assumed to be an elastic–plastic medium. When loading the constant vertical stress onto the CGFB, it will deform laterally, but the steel column restricts its lateral deformation except for the opening where the lateral pressure sensor is installed. The CGFB specimen can deform through the opening and the sensing film of the lateral pressure sensor can perceive its deformation. Although the vertical load stays constant, the deformation of the CGFB through the opening gradually increases with the elapse of curing period. This is exactly due to the creeping effect. Sellier et al. (2016) have developed a concrete creep model that considers the effects of curing time, nonlinearity, multi-axiality, hydration and temperature. Since the CGFB in the current study is also a cementitious material like concrete, this model is used here to describe the creeping behavior of the CGFB under vertical load. The increase in the lateral pressure of the CGFB with curing time can be explained by the following equations (Benboudjema et al. 2005; Ladaoui et al. 2011, 2013; Vidal et al. 2014): P(t) = ϕ(t) · E 0

(9.30)

  t ϕ(t) = kref C C C ln 1 + β  C H = αw α

(9.32)

C T = Cw Cp

(9.33)

T



H

Ew Cw = exp − R 

Ep 1 Cp = exp − − R T

M

 1 1 − T Tref  1 if T > Tthr Tthr

(9.31)



Cp = 1 if T ≤ Tthr

(9.34) (9.35) (9.36)

where, P(t) is the lateral pressure (investigated by the lateral pressure sensor) at time t, ϕ(t) is the creep function, E 0 is the modulus of deformation, k ref is the reference creep coefficient, C T is the influence of temperature on creep potential, C H is the influence of humidity, C M is the nonlinear effect of mechanical load, β is a parameter characterizing the initial material state, α w is the volume water content, α is the total volume of porosity, C w and C p are two functions which consider the effect of temperature on viscosity and creep potential respectively, E w is the activation energy of water viscosity (that is about 17,000 J/mol), T ref is the reference temperature, T thr is the threshold temperature (that is about 45 °C), T is the material temperature, R is the gas constant (which is 8.31 J/mol/K), and E p is about 25,000 J/mol. According to the above equations, the lateral pressure of the CGFB indeed increases with the elapse of the curing time. In addition, it can also be noticed in

9.5 Thermo–Hydro–Mechanical Coupled Behavior of CGFB

171

Fig. 9.36 that, a higher PC/FA ratio results in a higher lateral pressure development. This can also be qualitatively explained by the above Eqs. (9.30)–(9.36). As discussed before, increasing the PC/FA ratio causes the increase of cement hydration reactions. This consequence is not only beneficial to the generation of more hydration products (to refine the CGFB pore structures of and thus to reduce its porosity), but also the release of more hydration heat (to increase the CGFB temperature). In this study, the CGFB is confined laterally, and only a hole is left for it to expand. On the condition of loading the same vertical stress (80 kPa), it is easily figured out that, the CGFB sample that has a lower porosity will yield more lateral expansion and less vertical compression than the one which has a higher porosity (as shown in Fig. 9.38). Meanwhile, with the combination of thermal expansion, the increase of temperature (T ) and the decrease of porosity (α) together lead to the increase in the lateral pressure of the CGFB, on the basis of the above discussions and expressions (5)–(11). This is the reason that the CGFB specimen 3# has the highest lateral pressure development, since it possesses the highest PC/FA ratio among the three samples.

Fig. 9.38 Schematic diagram for explaining the effect of porosity on the CGFB lateral expansion

172

9.5.2.4

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Evolution of Strength

The effect of curing temperature on the strength development of the CGFB sample 3# (which is taken as an example) in undrained condition is presented in Fig. 9.39. From this figure, it is seen that the curing temperature plays a significant role in the CGFB strength development with curing time. As expected, the UCS of the CGFB develops with time, and a higher curing temperature brings about a higher strength development. This is because a higher curing temperature can accelerate the binder hydration process, which generates significant amount of hydration products (e.g., C–S–H gel). The C–S–H gel is regarded as the main binding composition in hardened cement-based materials (Gani 1997), such as CGFB mixes. These hydration products can create bonds between aggregates of the CGFB and thus facilitate strength. This explains why the UCS of the CGFB increases with the curing temperature. In addition, another factor should also be considered as a contributor to the curing temperature rise induced strength development of CGFB at early ages. This factor is that, as the curing time goes on, the amount of hydration products gradually increases and precipitates in the pores of the CGFB, and the pore structures within the CGFB are much more refined by larger amount of hydration products at a higher curing temperature. This can be graphically verified by the results of SEM tests performed on the CGFB samples cured for 7 days, at 20 °C (Fig. 9.40a) and 50 °C (Fig. 9.40b), respectively. From Fig. 9.40 it can be apparently observed that, the CGFB which is cured at a higher temperature (50 °C) has a denser microstructure with a finer distribution of the pores than the CGFB cured at the temperature of 20 °C.

Fig. 9.39 Effect of curing temperature on the UCS development in the CGFB

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

(a) Cured at 20 °C

173

(b) Cured at 50 °C

Fig. 9.40 Effect of curing temperature on the microstructure within the CGFB

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB It has been described before that, the CGFB technology is introduced for underground disposal of coal mine waste. Once prepared and placed underground, the CGFB should possess satisfactory stability, durability, and environment-friendly properties, and these properties are strongly affected by the thermal (T), hydraulic (H), mechanical (M) and chemical (C) processes as well as the THMC couplings. Understanding and modeling of the coupled THMC processes that occur in CGFB are significant for evaluating the performance of CGFB and thereby designing stable, durable and environment-friendly CGFB structures. The objective of this study is to develop a numerical model to analyze and assess the THMC behavior of CGFB. The numerical simulation results predicted by the THMC model are compared with the data of 3 case studies.

9.6.1 Mathematical Formulation of the THMC Model 9.6.1.1

Thermal Equations

The temperature of CGFB is mainly affected by two factors, which are self-heating and heat exchange between the CGFB and its surroundings. The former factor results from the exothermic reaction of binder hydration, and the latter includes thermal convection, thermal conduction, and heat radiation. It should be pointed out that the heat radiation is insignificant and thus not considered in the present study. (1) Binder hydration Binder hydration is an exothermal reaction generating heat within CGFB. A previous study has provided a formula for describing the hydration heat generating process (De Schutter and Taerwe 1996):

174

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

 

 E ap 1 1 · qh = qmax · a1 · sin(π · α)a2 · exp(−a3 · α) · exp − R Tr Tc

(9.37)

where, qh is the binder hydration heat produced per unit time by weight; qmax is the maximum heat production rate at the temperature of 20 °C; a1 , a2 , and a3 are constants determined by experiments; α is the degree of binder hydration; T r is the reference temperature; T c is the CGFB temperature; R is the universal gas constant; and E ap is the apparent activation energy, which is dependent on the temperature of CGFB and can be calculated by the following expressions (Jin Sang and Kwang 2001): when T c < 20 °C, E = 33,500 + 1470 (20 − T c ); otherwise when T c ≥ 20 °C, E = 33,500 (J/mol). (2) Thermal convection A CGFB skeleton is a kind of porous medium, which is formed by solid grains as well as liquid water and dry air in pores. Thermal convection within CGFB occurs when heat transfer through both liquid water and dry air flow. Since the convective heat interaction between the solid grains and dry air is insignificant. Therefore, the heat transfer process via dry air is ignored in this study. The convective heat through liquid water flow (Qw ) can be expressed by the following equation: Q w = ρw Cw u w · ∇T

(9.38)

where ρ w is the density of liquid water, C w is the specific heat capacity of liquid water at constant pressure, uw is the velocity field of liquid water, and ∇T is the gradient of temperature. (3) Thermal conduction Fourier’s law is employed to characterize the heat conduction process between CGFB and its surroundings (Nasir and Fall 2009): q = −keq · ∇T

(9.39)

where, q denotes the conductive heat flux vector, and k eq refers to the equivalent thermal conductivity of CGFB. Somerton et al. have provided an equation for calculating k eq as follows (Somerton et al. 1973): keq = kd +

√ ωe · (ks − kd )

(9.40)

where k d is the thermal conductivity of the porous medium (like CGFB in this study) in a completely dry condition, k s is the thermal conductivity of the porous medium when it is fully saturated with water, and ωe denotes the effective saturation degree of the porous medium. Therefore, the following equation is employed here to calculate k s of CGFB (Côté and Konrad 2005; Ghirian and Fall 2013):

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

ks = kg1−φ · kwφ

175

(9.41)

where, k g and k w denote the mean thermal conductivity values of the coal gangue grains and water, respectively, and φ is the porosity. The condition that all the pores of a porous medium are completely dry can refer to the situation that these pores are fully saturated with air. As a consequence, the above Eq. (9.41) can be derived to calculate k d (Cui and Fall 2015): kd = kg1−φ · kaφ

(9.42)

where k a is the value for the thermal conductivity of air. (4) Integrated expression In consideration of the heat conduction and convection in CGFB, the heat conduction between the CGFB and its surroundings, as well as the heat generation from binder hydration, the following expression can be derived: (ρC)eq

∂T + Q w + ∇q = Q H ∂t

(9.43)

where (ρC)eq is the equivalent volumetric heat capacity of the porous CGFB skeleton at constant pressure, and QH refers to the heat generated by binder hydration and can be obtained by using the following equation: Q H = Wb · q h

(9.44)

where W b is content of the binder used for preparing the CGFB.

9.6.1.2

Hydraulic Equations

(1) Fluid flow As discussed before, the heat transfer through dry air in the CGFB is ignored. Therefore, only the liquid water flow process is considered in this study. With the help of extended Darcy’s law, the following equation can be used to analyze the liquid water flow within the CGFB (Mainguy et al. 2001; Poyet et al. 2011): u w = −ki

krw ∇ Pw μw

(9.45)

where k i is the intrinsic permeability of CGFB, k rw is the relative permeability of CGFB with respect to liquid water, μw is the dynamic viscosity of water, and Pw is the pore water pressure. The permeability and hydraulic conductivity (i.e., coefficient of permeability) of CGFB to liquid water have the following relationship:

176

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

ki =

K μw γw

(9.46)

where, K is the hydraulic conductivity of CGFB, and γ w is the volume weight of water. The following equation is employed to calculate the hydraulic conductivity of CGFB (Ghirian and Fall 2013):   K = K A · exp Aα B

(9.47)

where K A is the hydraulic conductivity of the aggregate (coal gangue) being used, α is the degree of binder hydration, and A and B are the experimentally determined fitting parameters. According to the study of Ghirian and Fall (2013), the values of A and B are −8.173 and 4.035, respectively. The relative permeability k rw in Eq. (9.45) is a function of effective saturation degree (ωe ), as expressed in the following form (Mualem 1976; van Genuchten 1980; Luckner et al. 1989): krw =

√

 m 2  ωe · 1 − 1 − ωe1/m

(9.48)

where, m is the material parameter, and ωe can be further written in the following form: ωe =

θv − θr θs − θr

(9.49)

where, θ v is the volumetric water content, and θ s and θ r are the saturated and residual water contents, respectively. The residual water content (θ r ) of cemented backfill materials (such as CGFB) can be calculated by an empirical expression as follows (Abdul-Hussain and Fall 2011): θr = f 1 · exp(− f 2 · α)

(9.50)

where, f 1 and f 2 are the fitting constants, and the values used for them are 1.31 and 7.54, respectively, based on Abdul-Hussain and Fall’s data (Abdul-Hussain and Fall 2011). (2) Matric suction On the basis of the van Genuchten model, θ v can be calculated by the following equation (van Genuchten 1980; Abdul-Hussain and Fall 2011): θs − θr θv = θr +  m 1 1 + (l · ψ) 1−m

(9.51)

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

177

where ψ is the matric suction; and l and m are the fitting parameters, which can be calculated by (Abdul-Hussain and Fall 2011; Ghirian and Fall 2013): l = f 3 · exp( f 4 · α)

(9.52)

m = f 5 · α f6 + f 7

(9.53)

where, f 3 , f 4 , f 5 , f 6, and f 7 are the fitting constants: f 3 = 0.21 kPa−1 , f 4 = 6.92, f 5 = 0.0415, f 6 = 4.23 and f 7 = 0.41. The integration of Eqs. (9.49) and (9.51) yields: 1 ωe =  m 1 1 + (l · ψ) 1−m

(9.54)

The matric suction (ψ) can be characterized by the following equation (AbdulHussain and Fall 2011): ψ =−

RT ln Hr Vw Mw

(9.55)

where V w is the specific volume of water (i.e., 1/ρ w ), M w is the molecular mass of water, and H r is the relative humidity. Zhang and Yang (2011) have provided an empirical expression to describe the relative humidity (H r ) of cemented-based materials (e.g., CGFB): Hr = 100 + f 8 · exp( f 9 · ω)

(9.56)

where, f 8 and f 9 are the fitting parameters (the values used for them in this study are −8.63e9 and −23.8, respectively); and ω is the saturation degree, which can be defined by the following equation (Hansen 1986): ω=

Jcw + Jgw Jcw + Jgw + Jcs

(9.57)

where, J cw, J gw , and J cs are the capillary water content, gel water content and chemical shrinkage, respectively, and all of them can be defined by the function of the degree of binder hydration, as expressed as follows separately: Jcw = φ0 − 1.32(1 − φ0 )α

(9.58)

Jgw = 0.6(1 − φ0 )α

(9.59)

Jcs = 0.2(1 − φ0 )α

(9.60)

178

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

where φ 0 is the initial porosity of the porous medium. With the combination of Eqs. (9.45)–(9.60), the velocity field of liquid water (uw ) and matric suction can both be expressed with the function of binder hydration degree (α).

9.6.1.3

Mechanical Equations

For a hydrating CGFB structure, the binder hydration progress generates heat and thus causes thermal stress within the CGFB, and in addition, the binder reacts with water to induce chemical shrinkage. Furthermore, when the CGFB structure is placed into an underground stope for ground support, it suffers stress from surrounding rocks, and this stress results in elastic–plastic deformation in the CGFB. Therefore, based on the above discussions, the total strain of the CGFB structure should include elastic, plastic and thermal strains as well as chemical shrinkage: ε = εel + εpl + εth + εch

(9.61)

where ε is the total strain, εel , εpl and εth are the elastic strain, plastic strain, and thermal strain, respectively, and εch is the chemical shrinkage. For the simplicity of modeling, the CGFB structure is considered as isotropic, thus the constitutive relationship between the effective stress and elastic strain of the CGFB is expressed in the following form:   σeff = Dεel = D ε − εpl − εth − εch

(9.62)

where σ eff is the effective stress; and D is the elasticity matrix, which can be written in terms of Lamé parameters λ and μ as follows: ⎡

λ + 2μ λ λ 0 0 ⎢ λ λ + 2μ λ 0 0 ⎢ ⎢ λ λ λ + 2μ 0 0 ⎢ D=⎢ ⎢ 0 0 0 μ0 ⎢ ⎣ 0 0 0 0μ 0 0 0 0 0

⎤ 0 0⎥ ⎥ ⎥ 0⎥ ⎥ 0⎥ ⎥ 0⎦ μ

(9.63)

λ and μ are defined by the following equations, respectively: λ=

Eυ (1 + υ)(1 − 2υ)

(9.64)

E 2(1 + υ)

(9.65)

μ=

where E is the elastic modulus, and υ is the Poisson’s ratio.

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

179

It should be pointed out that the elastic modulus and Poisson’s ratio for a CGFB are not constant, and they are dependent on the CGFB properties affected by the proceeding of binder hydration. The binder hydration generates hydration products for contributing to the hardening of the CGFB, and thus the variation in the elastic modulus and Poisson’s ratio of the CGFB. The following equation is used to describe the relation between the elastic modulus and the degree of binder hydration (De Schutter and Taerwe 1996): E = Eu



α − α0 αu − α0

X (9.66)

where E u is the elastic modulus when the degree of binder hydration is equal to 1; α u is the ultimate binder hydration degree; α 0 refers to a certain value of hydration degree, below which no strength develops in the CGFB; X is a material constant that depends on the mix components of the CGFB. The following expression is employed to calculate the ultimate binder hydration degree (α u ) of CGFB (Kjellsen et al. 1991): αu =

1.031 · w/c + 0.5 · pFA + 0.3 · pSlag 0.194 + w/c

(9.67)

where w/c refers to the water-to-cement ratio; and pFA and pSlag are the proportions of fly ash and blast furnace slag in the binder, respectively. It should be stated that if the calculated value of α u exceed 1 according to Eq. (9.67), the value used for α u is deemed as 1. The evolution of the Poisson’s ratio with the binder hydration degree can be predicted by the following equation (Bittnar 2006; Boumiz et al. 1996; Galaa et al. 2011; Sayers and Grenfell 1993):   υ = 0.5 · exp(Y1 α) + Y2 · α Y3 · exp Y4 α Y5

(9.68)

where, Y 1 , Y 2 , Y 3 , Y 4, and Y 5 are the fitting parameters. Based on the study conducted by Cui and Fall (2016), Y 1 = −0.2, Y 2 = −15,000, Y 3 = 7, Y 4 = −11, and Y 5 = 0.7. The effective stress (σ e ) for a CGFB structure (a porous medium) can be written in the following form (Nuth and Laloui 2008): σeff = σ − φ Pw δi j

(9.69)

where σ is the total stress vector, δ ij is the Kronecker’s delta: δ ii = 1; δ i=j = 0. The strain increment due to thermal factors can be defined by the following equation: dεth = βth · dT · I

(9.70)

180

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

where β th is the coefficient for linear thermal expansion, and I is the second-order volumetric unity tensor. β th can be calculated by the following expression (Choktaweekarn and Tangtermsirikul 2009; Cui and Fall 2015; Kamali et al. 2004; Neville and Brooks 1987):   2vcg · (1/Cb − 1) · βp − vcg    βth = βp − [(w/c) · vw + vc ] · 1 + N1 · (1 − φ) N2 /E cg + 2vcg · (1/Cb − 1) (9.71) where C b is the binder content, vcg , vw , and vc are the specific volumes of coal gangue, water and cement, respectively, N 1 and N 2 are the fitting parameters (the values used for them are 46.03 GPa and 3.16, separately, according to the study of Cui and Fall (2015)), E cg is the elastic modulus of coal gangue used, and β p and φ can be expressed in the following forms, respectively:  D1 βuc vc · (1 − α) + D2 βhp vc + Rccw/hc (vccw + vaw ) · α βp = (w/c) · vw + vccw + (1/Cb − 1) · vcg φ=

(w/c) · vw − (vccw + vaw )Rccw/hc · α (w/c) · vw + vccw + (1/Cb − 1) · vcg

(9.72) (9.73)

where D1 and D2 are the fitting constants (the values used for them in this study are 0.28 and 1.23, respectively, according to the data reported by Cui and Fall (2015)); β uc and β hp are respectively the thermal expansion coefficients of unhydrated cement and hydration products; vccw and vaw are the specific volumes of chemically combined water and absorbed water, respectively, and according to Brouwers’s study (Brouwers 2004): vccw = 0.72 (cm3 /g), vaw = 0.90 (cm3 /g); and Rccw/hc , which represents the mass ratio of chemically combined water and hydrated cement, can be calculated by the following equation (Powers and Brownyard, 1947): Rccw/hc = 0.187 pC3 S + 0.158 pC2 S + 0.665 pC3 A + 0.213 pC4 AF

(9.74)

where pi is the weight proportion of the clinker composition. For the convenience of numerical computation, the thermal expansions of unhydrated cement and hydration products are neglected, that is to say, β uc = 0 and β hp = 0. Therefore, Eq. (9.71) can be rewritten in the following form: 2vcg2 · (1/Cb − 1)    (9.75) βth = [(w/c) · vw + vc ] · 1 + N1 · (1 − φ) N2 /E cg + 2vcg · (1/Cb − 1) The strain increment due to chemical factors can be defined as follows (Gawin et al. 2008): dεch = βch · dα · I

(9.76)

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

181

where, β ch is the chemical shrinkage coefficient, which can be calculated by the following equation based on several previous studies (Brouwers 2004; Cui and Fall 2015; Powers and Brownyard 1947): βch =

9.6.1.4

(2vw − vccw − vaw ) · Rccw/hc (w/c) · vw + vccw + (1/Cb − 1) · vcg

(9.77)

Equations for Characterizing the Chemical Process

The main chemical reaction in a CGFB is the process of binder hydration, and the rate of this process can be characterized by the degree of binder hydration (α), which is defined by the following expression (Kjellsen and Detwiler 1993; Poole et al. 2007; Schindler and Folliard 2005):   τ  α = αu · exp − te

(9.78)

where α u is the ultimate binder hydration degree, which can be calculated by Eq. (9.31);  is the shape parameter of the binder hydration; τ is the time parameter of the binder hydration at the reference temperature; t e is the equivalent age, which is defined as: t te =

 

E ap 1 1 dt · exp − − R Tc Tr

(9.79)

0

The above mathematical modeling process indicates that the thermal equations (T), hydraulic equations (H), mechanical equations (M) and equations of chemical process (C) can be mutually coupled with each other by the evolution of the degree of binder hydration with time.

9.6.2 Verification of the Model and Simulation Results The developed THMC coupled model is implemented into the software COMSOL Multiphysics to conduct numerical analysis and calculation, and the simulated results are compared with the outcomes obtained by three experimental cases, for verifying the validity and applicability of the model. Table 9.15 provides the input parameters, boundary conditions and initial values used for the numerical computations of each experiment study. The CGFB mixtures

182

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Table 9.15 Input parameters, boundary conditions and initial values used for model verification Case test 1

Case test 2

Case test 3

w/c ratio

2.1

2.1

2.1

Dimension (length × width × height)

0.3 m × 0.3 m × 0.55 m

0.15 m × 0.15 m × 0.2 m

0.1 m × 0.1 m × 0.1 m

Initial CGFB temperature

12 °C

12 °C

N/A

Environment temperature

12 °C

12 °C

N/A

CGFB thermal conductivity

2.85 W/(m K)

2.85 W/(m K)

N/A

CGFB specific heat

1200 J/kg K

1200 J/kg K

N/A

Thermal conductivity of the container

19 W/(m K)

19 W/(m K)

N/A

E ap (apparent activation energy)

Varies

Varies

Varies

CGB density

2000 kg/m3

2000 kg/m3

2000 kg/m3

α0

0

0

0

αu

0.91

0.91

0.91



30

30

30

τ

0.6 h

0.6 h

0.6 h

qmax

2.19 W/kg

2.19 W/kg

N/A

a1

2.6

2.6

2.6

a2

0.667

0.667

0.667 3.0

a3

3.0

3.0

X

2.2

2.2

2.2

Eu

1900 MPa

1900 MPa

1900 MPa

Top surface

Free

Load

Displacement

Lateral sides

Roller

Roller

Free

Mechanical module

Bottom side

Fixed

Fixed

Fixed

Volume force

Gravity

Gravity

Gravity

Top surface

Mass flux

Mass flux

N/A

Lateral sides

Insulated

Insulated

Insulated

Bottom side

Insulated

Open

N/A

Volume force

Gravity

Gravity

Gravity

Initial value

Hydraulic head = 0

Hydraulic head = 0

N/A

Hydraulic module

Thermal module Top side

12 °C

12 °C

N/A

Lateral sides

12 °C

12 °C

N/A (continued)

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

183

Table 9.15 (continued) Case test 1

Case test 2

Case test 3

Bottom side

12 °C

12 °C

N/A

Initial temperature

12 °C

12 °C

N/A

Some of the data in Table 9.2 are based on the referenced studies (Cui and Fall 2015; De Schutter and Taerwe 1996; Poole et al. 2007; Schindler and Folliard 2005; Zhang and Yang 2011)

Table 9.16 Mix proportion of the CGFB used

Solids content (wt%)

Coal gangue (wt%)

Fly ash (wt%)

Cement (wt%)

79.0

49.0

20.0

10.0

used in the three cases are prepared with the same mix proportion, as shown in Table 9.16. (1) Model validation against experiment 1 An experimental study that investigates the evolutions of temperature and pore water pressure in CGFB versus time is conducted to validate the developed THMC model. As shown in Fig. 9.41a, the CGFB mixtures are placed into a cubic container, which has adequate stiffness to restrict the lateral deformation of the CGFB. Two transducers are buried within the CGFB at the positions of P1 and P2, in order to monitor the variations of temperature and pore water pressure, respectively. The geometry of the simulated model is built in COMSOL Multiphysics, as shown in Fig. 9.41b. Figure 9.42 reveals a comparison between the numerically predicted and experimentally measured data of the temperature (at the point P1) evolution versus time (0–7d) in the CGFB. It can be noticed that the simulated evolution of temperature (b)

15cm

5cm

(a)

P1 25cm

Ambient CGFB

15cm

P2

Fig. 9.41 Geometry of the simulated model in Experiment 1: a schematic diagram of the test; b geometric model developed in COMSOL

184

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.42 Comparison of temperature (at P1) evolution versus time between model simulation and experiment investigation

agrees well with the experimental results, except for some misfits that are due to the varied room temperature during the test. The comparison results prove that the model prediction is valid, and the model is capable to simulate temperature evolution of CGFB. At the very early age (0–1d), a large amount of heat is generated by binder hydration, contributing to a rapid temperature rise within the CGFB. Afterwards, since the binder hydration process slows down and the heat exchanges between the CGFB and its surroundings, the CGFB temperature decreases until it remains steady (close to the ambient temperature). Figure 9.43 shows the simulated spatial distribution of temperature in the CGFB for different curing ages (1, 3 and 7 days). A contrast between the model prediction and experimental observation of the evolution of pore water pressure at the point P2 with time (0–7d) in the CGFB is demonstrated in Fig. 9.44. It is observed from this figure that, there is a favorable agreement between the simulated results and measured data. Both the predicted and monitored outcomes indicate that the pore water pressure rapidly decreases in the early age (0–1d). This is because, during this period, the binder hydration process is intense, consuming plenty of pore water and thus contributing to the strong development of pore water pressure. After about 3 days, the pore water pressure at the point P2 in the CGFB maintains almost steady at the value of −175 kPa. In order to understand the spatial distribution of pore water pressure within the whole CGFB structure, Fig. 9.45 shows the numerically simulated spatial distribution of pore water pressure in the CGFB at different periods (0, 1, 3 and 5 days). In the beginning, the distributed pore water pressure in the CGFB is positive, but with the

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

185

Fig. 9.43 Simulated evolution of temperature distribution in the CGFB at different curing ages: a 1 day; b 3 days; c 7 days

consumption of pore water due to binder hydration, the negative pore water pressure develops from the top domain of the CGFB. From the view of the total distribution of the pore water pressure in the CGFB, its value increases with depth from the top of the CGFB to the bottom. This is ascribed to the fact of water migration induced by gravity. (2) Model validation against experiment 2 As schematically presented in Fig. 9.46a, the CGFB structure is cured in a rigid cylinder container (with the thickness of 1 cm) under a constant vertical pressure of 0.1 MPa. A circular drainage hole with the diameter of 2.2 cm is cut at the center of the container bottom for water drainage, and a permeable membrane is covered above the drainage hole. The monitoring point P3 is utilized for temperature investigation, and P4 is for lateral pressure observation. Figure 9.46b presents the geometric model for the test that is established in COMSOL. A contrast between the simulated results and monitored data of the temperature (at the point P3 of the CGFB) evolution versus time is presented in Fig. 9.47. From

186

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.44 Comparison of pore water pressure (at P2) evolution versus time between model simulation and experiment investigation

this figure, it is found that the predicted temperature evolution basically matches the measured data, except for some misfits which are probably ascribed to the varied boundary conditions during the experiment. This further verifies the capability of the developed model for temperature prediction. As expected, the CGFB temperature of experiment 2 is faster to reach the peak value than that of experiment 1, but the peak temperature of the CGFB in experiment 2 is lower than that in experiment 1. This is due to the difference between the sample dimensions. Figure 9.48 demonstrates the evolution of lateral stress at P4. It is noticed from this figure that the predicted results basically agree with the experimentally measured data. Both the simulation and experiment results show that the lateral stress gradually increases with time, until after about 3 days, the lateral stress values become steady. This is due to the fact that, before the time point (3d), the CGFB structure is plastic and a vertical load on it can contribute to the increase of the lateral stress. However, after 3 days, the CGFB becomes hardened due to the process of binder hydration. The constant vertical load can no longer contribute to the increase in the lateral stress, expect for loading a higher vertical pressure. As an example, Fig. 9.49 shows the water velocity field in the CGFB at the initial time (0 h) simulated by the developed model. From this figure, it is noted that, under the effect of gravity and pore pressure, water flows to the drainage hole. Additionally, near the drainage hole, where the arrows are intensively distributed, the velocity of water flow is high. According to the predicted velocity field, the evolution of the average water flow velocity through the drainage hole versus time can be obtained, as shown in Fig. 9.50. From this figure, it can be found that the average water flow velocity through the drainage hole gradually decreases with time, until 4 h later,

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

187

Fig. 9.45 Simulated evolution of pore water pressure distribution in the CGFB for different ages: a 0 day; b 1 day; c 3 days; d 5 days

(a)

(b)

P3

CGFB

P4

20cm

3cm

Uniform load

2.2cm 15cm

Fig. 9.46 Geometry of the simulated model in Experiment 2: a schematic diagram of the test; b geometric model developed in COMSOL

188

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.47 Comparison of temperature (at P3) evolution versus time between model simulation and experiment investigation

Fig. 9.48 Comparison between model simulation and experiment investigation results of lateral stress (at P4) evolution with time

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB

189

Fig. 9.49 Simulated velocity field distribution in the CGFB (0 h)

Fig. 9.50 Predicted evolution of average water flow velocity through the drainage hole versus time

no more water drains out. This is because the binder hydration process not only consumes pore water, but also generates hydration products to refine the pores in the CGFB and thus block the water flowing channel. The following equation is used to calculate the drainage water flow: t Qd =

va Sd dt

(9.80)

0

where Qd is the quantity of water flow through the drainage hole at time t, va is the average water flow velocity through the drainage hole, S d is the cross-sectional area of the drainage hole: S d = π · 1.12 (cm2 ). According to the above Eq. (9.80), the drainage quantity of water flow can be predicted. Figure 9.51 reveals a comparison between the predicted and experimentally observed data of the drainage quantity of water flow. It is indicated from this figure

190

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Fig. 9.51 Comparison between model simulation and experiment measurement of drainage flow

that the numerical simulation is consistent with the experimental measurement. This also indicates that the developed model is provided with the capability for water drainage prediction. (3) Model validation against experiment 3 In order to understand the effect of thermal process on the mechanical performance of CGFB, experiment 3 is conducted. In this test, 12 groups of CGFB specimens are prepared and they are cured for 1, 3 and 7 days at the curing temperatures of 20, 50, 75 and 90 °C, respectively. When finishing the curing process, uniaxial compression tests are conducted on these CGFB specimens to obtain the stress–strain relationships. Each group includes 3 samples with the dimension of 10 × 10 × 10 cm3 , and the average value tested for each group is figured out and used. Figure 9.52 shows the stress evolution with strain in the CGFB samples cured at various temperatures (20, 50, 75, and 90 °C) and for different ages (1, 3, and 7 days). It is found that the model prediction results coincide well with the tested values, except for some misfits that may be attributed to the varied boundary conditions during the experiment. As expected, the CGFB peak stress increases with time and temperature. Especially at the early age (0–3d), increasing the curing temperature leads to a remarkable increase in the peak stress value of the CGFB. The comparison results further verify the validity of the developed model, as well as its ability for predicting the mechanical behavior of CGFB.

9.6 Thermo–Hydro–Mechanical–Chemical Coupled Behavior of CGFB Fig. 9.52 Comparison of simulated results and experimental data of uniaxial compressive tests of the CGFB samples cured at different temperatures and ages: a 1 day; b 3 days; c 7 days

191

192

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

References Abadelrahman, M., Elbatanouny, M.K., Ziehl, P.H.: Acoustic emission based damage assessment method for prestressed concrete structures: modified index of damage. Eng. Struct. 60, 258–264 (2014) Abdul-Hussian, N., Fall, M.: Unsaturated hydraulic properties of cemented tailings backfill that contains sodium silicate. Eng. Geol. 123(4), 288–301 (2011) Benavent-Climent, A., Gallego, A., Vico, J.M.: An acoustic emission energy index for damage evaluation of reinforced concrete slabs under seismic loads. Struct. Health Monit. Int. J. 11, 69–81 (2012) Benboudjema, F., Meftah, F., Torrenti, J.: Interaction between drying, shrinkage, creep and cracking phenomena in concrete. Eng. Struct. 27(2), 239–250 (2005) Bittnar, Z.: Microstructure-based micromechanical prediction of elastic properties in hydrating cement paste. Cem. Concr. Res. 36(9), 1708–1718 (2006) Bouguerra, A., Laurent, J., Goual, M., Queneudec, M.: The measurement of the thermal conductivity of solid aggregates using the transient plane source technique. J. Phys. D Appl. Phys. 30(30), 2900–2904 (1997) Boumiz, A., Vernet, C., Tenoudji, F.C.: Mechanical properties of cement pastes and mortars at early ages: evolution with time and degree of hydration. Adv. Cem. Based Mater. 3(3), 94–106 (1996) Brouwers, H.J.H.: The work of powers and Brownyard revisited: part 1. Cem. Concr. Res. 34, 1697–1716 (2004) Carpinteri, A., Lacidogna, G., Niccolini, G.: Damage analysis of reinforced concrete building by the acoustic emission technique. Struct. Control Health Monit. 18, 660–673 (2011) Chandel, S., Singh, S.N., Seshadri, V.: Deposition characteristics of coal ash slurries at higher concentrations. Adv. Powder Technol. 20, 383–389 (2009) Choktaweekarn, P., Tangtermsirikul, S.: A model for predicting coefficient of thermal expansion of cementitious paste. Sci. Asia. 35, 57–63 (2009) Chotard, T., Gimet-Breart, N., Smith, A., Fargeot, D., Bonnet, J.P., Gault, C.: Application of ultrasonic testing to describe the hydration of calcium aluminate cement at the early age. Cem. Concr. Res. 31(3), 405–412 (2001) COMSOL: COMSOL Multiphysics 5.0 (2014). http://www.comsol.com Côté, J., Konrad, J.M.: A generalized thermal conductivity model for soils and construction materials. Can. Geotech. J. 42, 443–458 (2005) Cui, L., Fall, M.: A coupled thermo-hydro-mechanical-chemical model for underground cemented tailings backfill. Tunn. Undergr. Sp. Tech. 50(50), 396–414 (2015) Cui, L., Fall, M.: An evolutive elasto-plastic model for cemented paste backfill. Comput. Geotech. 71, 19–29 (2016) De Schutter, G., Taerwe, L.: General hydration model for Portland cement and blast furnace slag cement. Cem. Concr. Res. 25(95), 593–604 (1995) De Schutter, G., Taerwe, L.: Degree of hydration-based description of mechanical properties of early age concrete. Mater. Struct. 29(6), 335–344 (1996) Demirboga, R., Turkmen, I., Karakoc, M.B.: Relationship between ultrasonic velocity and compressive strength for high-volume mineral-admixtured concrete. Cem. Concr. Res. 34(12), 2329–2336 (2004) Fall, M., Pokharel, M.: Coupled effects of sulphate and temperature on the strength development of cemented tailings backfills: Portland cement-paste backfill. Cem. Concr. Compos. 32, 819–828 (2010) Fall, M., Samb, S.S.: Effect of high temperature on strength and microstructural properties of cemented paste backfill. Fire Safety J. 44(4), 642–651 (2009) Galaa, A.M., Thompson, B.D., Grabinsky, M.W., Bawden, W.F.: Characterizing stiffness development in hydrating mine backfill using ultrasonic wave measurements. Can. Geotech. J. 48(8), 1174–1187 (2011) Gani, M.S.J.: Cement and Concrete, 1st edn. Chapman and Hall, London (1997)

References

193

Gawin, D., Pesavento, F., Schrefler, B.A.: Modeling of cementitious materials exposed to isothermal calcium leaching, considering process kinetics and advective water flow. Part 1: theoretical model. Int. J. Solids Struct. 45, 6221–6240 (2008) Ghirian, A., Fall, M.: Coupled thermo-hydro-mechanical-chemical behavior of cemented paste backfill in column experiments. Part I: physical, hydraulic and thermal processes and characteristics. Eng. Geol. 164, 195–207 (2013) Gong, C., Li, C., Zhao, K.: Experimental study on b-value characteristics of acoustic emission of cemented filling body under loading and unloading test. J. Min. Safety Eng. 31(5), 788–794 (2014) (in Chinese) Hansen, T.C.: Physical structure of hardened cement paste: a classical approach. Mater. Struct. 19(114), 423–436 (1986) Hughes, B.P., Bahramian, B.: Cube tests and the uniaxial compressive strength of concrete. Mag. Concr. Res. 17(53), 177–182 (1965) Jin Sang, H.H., Kwang, M.L.: Estimation of compressive strength by a new apparent activation energy function. Cem. Concr. Res. 31(2), 217–225 (2001) Kamali, S., Moranville, M., Garboczi, E., Prené, S., Gérard, B.: Hydrate dissolution influence on the Young’s modulus of cement pastes. In: Proceedings Fracture Mechanics of Concrete Structures (FraMCoS-V), pp. 631–638. Routledge, Vail, Colorado (2004) Khan, M.I.: Factors affecting the thermal properties of concrete and applicability of its prediction models. Build. Environ. 37(6), 607–614 (2002) Kjellsen, K.O., Detwiler, R.J.: Later-age strength prediction by a modified maturity model. ACI Mater. J. 90(3), 220–227 (1993) Kjellsen, K.O., Detwiler, R.J., Gjørv, O.E.: Development of microstructures in plain cement pastes hydrated at different temperatures. Cem. Concr. Res. 21(1), 179–189 (1991) Ladaoui, W., Vidal, T., Sellier, A., Bourbon, X.: Effect of a temperature change from 20 to 50 °C on the basic creep of HPC and HPFRC. Mater. Struct. 44(9), 1629–1639 (2011) Ladaoui, W., Vidal, T., Sellier, A., Bourbon, X.: Analysis of interactions between damage and basic creep of HPC and HPFRC heated between 20 and 80 °C. Mater. Struct. 46(1–2), 13–23 (2013) Levitsky, M.: Analytical determination of pressure on formwork. Proc. ASCE 99, 551–564 (1973) Liu, X., Liu, K., Zhao, K., Hu, J.: Strain rate effects and acoustic emission characteristics of tailing cement backfill under uniaxial compression. Electron. J. Geotech. Eng. 18, 3919–3932 (2013) Luckner, L., Genuchten, M.T.V., Nielsen, D.R.: A consistent set of parametric models for the twophase flow of immiscible fluids in the subsurface. Water Resour. Res. 25(10), 2187–2193 (1989) Mainguy, M., Coussy, O., Baroghel-Bouny, V.: Role of air pressure in drying of weakly permeable materials. J. Eng. Mech. 127(6), 582–592 (2001) Mualem, Y.: A new model for predicting the hydraulic conductivity of unsaturated porous media. Water Resour. Res. 12, 513–522 (1976) Nasir, O., Fall, M.: Modeling the heat development in hydrating CPB structures. Comput. Geotech. 36(7), 1207–1218 (2009) Neville, A.M., Brooks, J.J.: Concrete Technology. Longman Scientific & Technical Harlow, New York (1987) Nuth, M., Laloui, L.: Effective stress concept in unsaturated soils: clarification and validation of a unified framework. Int. J. Numer. Anal. Met. Geomech. 32, 771–801 (2008) Parrott, L.J.: Moisture profiles in drying concrete. Adv. Cem. Res. 1(3), 164–170 (1988) Poole, J.L., Riding, K.A., Folliard, K.J., Juenger, M.C.G., Schindler, A.K.: Methods for calculating activation energy for Portland cement. ACI Mater. J. 104(1), 303–311 (2007) Powers, T.C., Brownyard, T.L.: Studies of the physical properties of hardened Portland cement paste. ACI J. Proc., 249–336 (1947) Poyet, S., Charlesa, S., Honoré, N., L’hostisa, V.: Assessment of the unsaturated water transport properties of an old concrete: determination of the pore-interaction factor. Cem. Concr. Res. 41(10), 1015–1023 (2011) Prem, P.R., Murthy, A.R.: Acoustic emission and flexural behaviour of RC beams strengthened with UHPC overlay. Constr. Build. Mater. 123, 481–492 (2016)

194

9 Properties of Cemented Coal Gangue-Fly Ash Backfill

Sagasta, F.A., Benavent-Climent, A., Roldán, A., Gallego, A.: Correlation of plastic strain energy and acoustic emission energy in reinforced concrete structures. Appl. Sci. 6, 84 (2016) Sayers, C., Grenfell, R.: Ultrasonic propagation through hydrating cements. Ultrasonics 31(3), 147–153 (1993) Schindler, A.K., Folliard, K.J.: Heat of hydration models for cementitious materials. ACI Mater. J. 102(1), 24–33 (2005) Schjödt, R.: Calculation of pressure of concrete on forms. Proc. ASCE 81, 1–16 (1955) Sellier, A., Multon, S., Buffo-Lacarrière, L., Vidal, T., Bourbon, X., Camps, G.: Concrete creep modelling for structural applications: non-linearity, multi-axiality, hydration, temperature and drying effects. Cem. Concr. Res. 79(1), 301–315 (2016) Shiotani, T.: Evaluation of repair effect for deteriorated concrete piers of intake dam using AE activity. Adv. Mater. Res. 13, 175–180 (2006) Shiotani, T., Ohtsu, M., Ikeda, K.: Detection and evaluation of AE waves due to rock deformation. Constr. Build. Mater. 15, 235–246 (2001) Somerton, W.H., Keese, J.A., Chu, S.L.: Thermal behavior of unconsolidated oil sands. In: Proceedings of 48th Annual Fall Meeting of the Society of Petroleum Engineers, Sept 1973, paper SPE-4506 Las Vegas, USA (1973) Trtnik, G., Kavcic, F., Turk, G.: Prediction of concrete strength using ultrasonic pulse velocity and artificial neural networks. Ultrasonics 49(1), 53–60 (2009) Ukrainczyk, N., Matusinovi´c, T.: Thermal properties of hydrating calcium aluminate cement pastes. Cem. Concr. Res. 40(1), 128–136 (2010) van Genuchten, M.T.: A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44(5), 892–898 (1980) Vidal, T., Sellier, A., Ladaoui, W., Bourbon, X.: Effect of temperature on the basic creep of highperformance concretes heated between 20 and 80 °C. J. Mater. Civ. Eng. 27(7), B4014002 (2014) Wu, D., Fall, M., Cai, S.: Coupling temperature, cement hydration and rheological behaviour of fresh cemented paste backfill. Miner. Eng. 42, 76–87 (2013) Yoon, S., Macphee, D.E., Imbabi, M.S.: Estimation of the thermal properties of hardened cement paste on the basis of guarded heat flow meter measurements. Thermochim. Acta 588(7), 1–10 (2014) Zhang, D., Yang, W.: Theoretical study and tests on relative humidity change of concrete under self-desiccation effect. J. Yangtze River Sci. Res. Instit. 28(9), 44–47 (2011) (in Chinese)

Chapter 10

Case Study of Cemented Coal Gangue-Fly Ash Backfill

Abstract Coal mine waste is used as backfill materials and mixed with hydraulic binder and additive to fill underground mined-out areas. Xinyang Coal Mine is selected as the case mine to present its significant practice in not only minimizing the surface disposal of mine solid wastes but also providing solution for ground control. The technology of cemented coal gangue-fly ash backfill (CGFB, mixture of binder, coal gangue, fly ash, water and additive if necessary) is introduced and utilized in this mine to dispose both waste and gob. Based on rheology, loop, and UCS (uniaxial compressive strength) tests, the optimum mix proportion and solid concentration for preparing the CGFB mixtures are achieved. The technological process of using the CGFB technology, which includes preparation, transportation, and placement of the CGFB materials, are also described and discussed. Keywords Underground coal mine · Coal gangue · Fly ash · Backfill · Waste management In this chapter, the use of CGFB technology in Xinyang Coal Mine is selected as a case study to describe and discuss the practice of coal mine solid waste management. As to Xinyang Coal Mine, the coal mine wastes that have been and are being produced cause serious social and environmental problems which urgently need to be solved. In addition, great quantities of coal resources depot under the villages for protecting the ground structures such as buildings. As a result, the utilization of CGFB technology in Xinyang Coal Mine can achieve some benefits as follows: • The usage of coal gangue and fly ash can recycle coal mine wastes, notably reducing the environmental contamination resulted from the surface discharge of them and thus improving the ecological environment of the mining area. • The CGFB technology can provide a solution for extracting the coal resources under the buildings without relocation of the villages above, contributing to substantial saving of cost. • The utilization of CGFB technology can set up a demonstration for similar coal mines to reduce surface disposal of mine solid waste and also safely extract the trapped coal resources to the utmost extent. • The CGFB technology can also control ground subsidence, easing the contradiction between local residents and the mine. © Springer Nature Singapore Pte Ltd. 2020 D. Wu, Mine Waste Management in China: Recent Development, https://doi.org/10.1007/978-981-32-9216-1_10

195

196

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill

10.1 Materials and Tests The CGFB mixtures of Xinyang Coal Mine are prepared by blending CG, FA, binder, additive, and water. The binder used is ordinary Portland cement 425# bought from the market, and the water is tap water. The additive used includes water reducer, suspending agent and early strength agent. The particle size distributions and chemical properties of the CG and FA have been demonstrated in Chap. 9. Required amounts of CG, FA, binder, additive, and water are mixed to form homogeneous CGFB mixtures. After that, the plastic CGFB slurries are subjected to rheology and loop tests. Besides, the testing CGFB slurries are poured into standard molds and cured in curing chamber to form hardened CGFBs, which are then taken out at different curing age and thereby subjected to UCS tests. The rheology, loop and UCS tests on the CGFB mixtures have also been discussed in Chap. 9, as well as the corresponding experimental results.

10.2 Preparation of CGFB Materials The CGFB materials are prepared by mixing CG, FA, binder, additive, and water, thus the preparation of CGFB materials includes provision of CG, FA, binder, additive and water, respectively, as well as mixing of these compositions.

10.2.1 Production of Coal Gangue CG is the major component in the CGFB mixtures. Hence, a large enough place should be provided for storing the stock dump of raw CGs that come from coal waste piles. The storage place for the stock dump of raw CGs in Xinyang Coal Mine covers an area of 1600 m2 , as shown in Fig. 10.1. Figure 10.2 graphically displays the concise process for producing made-up CGs that are used for preparing the CGFB materials. The raw CGs from the stock dump are loaded into a storage by shovel cars. It should be stated that the raw CG is hard and has large particle size, and if the raw CGs with these properties are employed to prepare CGFB mixtures, the pipeline that transports the CGFBs is very easy to be abraded. Therefore, the raw CGs should be crushed by crushed until the maximum particle size (MPS) of them meets the proper requirement. If the MPS of the crushed CGs is smaller than 15 mm, they can become qualified products and thus be delivered to the storage for made-up CGs. Only these made-up CGs can be applied to prepare CGFBs. The crusher used in the field is shown in Fig. 10.3 and the vibrosieve used is displayed in Fig. 10.4.

10.2 Preparation of CGFB Materials

197

Fig. 10.1 The storage place for the stock dump of raw CGs Storage of raw CG

Crusher

Conveyor

Vibrosieve

MPS > 15 mm

Conveyor

Elevator

Storage of made-up CG

MPS < 15 mm Conveyor

Fig. 10.2 Schematic flow diagram for the production of made-up CG

198

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill

Fig. 10.3 The crusher used in the field

Fig. 10.4 The vibrosieve used in the field

10.2 Preparation of CGFB Materials

199

Fig. 10.5 The silos for storing FA and binder

10.2.2 Provision of Fly Ash and Binder The FAs used for preparing the CGFB mixtures are from a power plant near Xinyang Coal Mine, and the binder used is ordinary Portland cement 425# that is provided by the nearby market. The FA and binder are separately stored in the corresponding silos, as shown in Fig. 10.5.

10.2.3 Provision of Water and Additive The process of water provision is schematically illustrated in Fig. 10.6. The water used for preparing the CGFB mixtures is pumped through pipeline from the impounding reservoir and thereby stored in the water storage tank. The process of additive provision is schematically demonstrated in Fig. 10.7. Since the additive used includes water reducer, suspending agent and early strength agent, they are blended with water in the mixing pool and then pumped through pipeline into the additive storage tank. Impounding reservoir

Pump

Fig. 10.6 Schematic flow chart for the water provision

Pipeline

Water storage tank

200

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill Mixing pool

Pump

Additive storage tank

Pipeline

Fig. 10.7 Schematic flow chart for the additive provision

10.2.4 System for CGFB Preparation When the CG, FA, binder, additive, and water are ready for preparing CGFB mixtures, the next key step is to blend these component materials. Figure 10.8 illustrates the whole CGFB preparation system, which includes two constituent parts: one is for operation and the other for back up. It can be seen from this figure that, the produced additive is blended with water in a vertical mixer, and then they are delivered into a horizontal mixer. The stored FA and binder as well as the made-up CG are also conveyed into the horizontal mixer, where the five constituent materials are blended together until the qualified homogeneous CGFB slurries are formed. After that, the prepared CGFB slurries are transferred to a storage and thus ready to be pumped into the gob through pipelines. It should be indicated that the quantities of the five ingredients composing the CGFB mixtures are measured and controlled by precise metering equipment installed in the system. Besides, the horizontal mixer (in comparison with the vertical one) can ensure full contact of the admixture ingredients, and thus provide well mixing consequence. The horizontal mixer is also equipped with intense and active stirring capability. At work Water

Additive

FA

CG Binder

Conveyor

Vertical mixer

Horizontal mixer Coal seam Storage Gob

Pump

Fig. 10.8 Schematic diagram of the CGFB preparation system

Back up

10.2 Preparation of CGFB Materials

201

10.2.5 Electrical Manipulative System The whole procedure for the preparation and transportation of CGFB is controlled by an electrical manipulative system (EMS). The EMS is mainly constituted by a video monitoring system (VMS) and an industrial computer-controlled system (ICCS). The entire CGFB preparation and transportation process can be visually displayed by the VMS. In addition, when all the corresponding parameters are input and installed, the weighing, loading, feeding, mixing, splitting, and transporting of the materials can all be automatically controlled by the ICCS.

10.2.6 Additional Appliance for Clean Production During the progress of CGFB preparation, dust control is a significant issue needs to be concerned. In the stages of crushing, conveying and feeding CG, as well as feeding and conveying of FA and binder, dust removal installations (as illustrated in Fig. 10.9) are implemented for dust extraction.

Fig. 10.9 The dust removal installation

202

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill

10.3 Transportation of CGFB Materials After being prepared, the CGFB materials are pumped through a pipeline system into underground mined-out areas. The entire pipeline contains different transporting pipes that are arranged at various locations. Figure 10.10 graphically exhibits the flow chart of the pipes layout for backfilling the CGFB mixtures and Fig. 10.11 schematically shows the pipeline transportation of the CGFBs. The overall length of the entire filling pipeline is 3500 m, including the pipes above ground (40 m), pipes through borehole (260 m), pipes along roadways (3000 m) and pipes behind working face (200 m). Fig. 10.10 Flow chart for the layout of the CGFB filling pipes

Pump

Pipes above ground

Pipes along roadways

Pipes through borehole

Pipes behind working face

Branching pipes

Output of CGFBs

Fig. 10.11 Schematic diagram for the pipeline transportation of the CGFB materials

10.4 Placement of CGFB Materials

203

Fig. 10.12 Schematic diagram for the discharge of CGFB materials behind working face

10.4 Placement of CGFB Materials Figure 10.12 schematically displays discharge of the CGFB materials through 6 branching pipes behind the mining face. Generally, the inner surface of the entire pipeline should be lubricated by water flush before transporting the CGFB mixtures, and cleaned by water wash after completing the transportation. At these stages, all the six valves fixed along the pipeline behind working face (the blue pipe in Fig. 10.12) are kept open, but all the six valves installed respectively in the six branching pipes (the green ones in Fig. 10.12) should be closed, and the drain pipe (the red one in Fig. 10.12) is used for water drainage. After the pipeline lubrication, and when the CGFB materials are carried to the mining face through the pipeline along roadways (the yellow one in Fig. 10.12), the opening and closing of the 12 valves are cooperatively controlled, until the gobs behind the mining face are uniformly filled with the CGFB materials. It should be pointed out that, these freshly discharged CGFB materials are plastic, and it takes at least 4 h for them to harden and thus stand on their own. Therefore, a type of hydraulic support is specially designed and manufactured for placement of the CGFB mixtures behind the working face, as schematically illustrated in Fig. 10.13. The hydraulic support provides a solid chamber, which is behind the hydraulic props and like a curing box, affording protective environment (necessary time and space) for the hardening progress of the CGFB mixtures. When the hardened and self-supporting CGFB structures are formed, the chamber can move forward with the advance of the mining face, and thereafter the CGFB structures expose and thus provide significant roof support. The movement of the chamber is powered by hydraulic pressure, and the working resistance of the hydraulic support is required to be higher than 4500 kN.

204

10 Case Study of Cemented Coal Gangue-Fly Ash Backfill

Fig. 10.13 Schematic diagram of the special hydraulic support for placing CGFB materials