Shear strength characterisation of in-pit mud to ensure lowwall ...

1

Shear strength characterisation of in-pit mud to ensure lowwall stability

Timothy Alexander Vangsness

B. Eng

A thesis submitted for the degree of Doctor of Philosophy at

The University of Queensland in 2020

School of Civil Engineering

Geotechnical Engineering Centre

2

Abstract

Weakened basal spoil and in-pit ‘mud’ caused by heavy rainfall events and flooding have been

associated with open-cut strip mine lowwall failures, particularly in Australia’s Bowen Basin. Large-

scale lowwall failures cause considerable disruptions to mining associated with a loss of production,

damaged infrastructure, and the potential loss of life. The main objective of this research funded by

the Australian Coal Associated Research Program (ACARP) was to characterise this in-pit mud to

determine its parameters for use in design. An improved understanding of in-pit mud characteristics

will reduce the mud removal requirements, or remove them entirely, resulting in enhanced mine

safety and economics.

Thirteen samples of mud and six samples of spoil were obtained from three mines within the Bowen

Basin. The BMA Coal spoil shear strength framework was utilised for categorisation, with the

selection of materials ranging from competent to incompetent. Each sample was thoroughly

characterised with respect to its physical, chemical, mineralogical and geotechnical properties.

Physical, chemical and mineralogical testing involved measurement of the materials in situ moisture

content, specific gravity, pH, electrical conductivity, total suction, Emerson class, Atterberg limits,

X-ray diffraction, exchangeable cations, and cation exchange capacity. The particle size distributions

included dry sieving for material that did not agglomerate during drying at 60⁰, and wet sieving after

24 hours of soaking in a water bath without dispersant, followed by testing of the fine fraction without

dispersant.

Characterisation of the mud and spoil showed the majority of materials were dominated by Quartz,

Kaolinite, Illite-Smectite and Albite. High levels of sodium Smectite were associated with materials

having finer particle size distributions, higher liquid limits, and increased levels of degradation upon

exposure to water. Typically, the geotechnical competency of the mud was related to the competency

of the spoil it formed from.

Accelerated degradation testing results showed that the majority of degradation occurs within the first

24 hours. Wetting and drying cycles produced a faster rate of degradation than prolonged saturation.

From these results, the development of a modified slake durability testing methodology allowed for

rapid identification of highly degradable spoil prone to slaking and dispersion.

Geotechnical testing involved standard consolidation of the -2.36 mm fractions in a water bath, and

large slurry consolidometer consolidation of select mud samples at -19 mm, and direct shear testing

both as sampled, and after 24 hours of soaking in water. Spoil scalped to -19 mm was tested in a large

direct shear box measuring 300 x 300 x 200 mm. Mud samples were tested in a standard direct shear

box with dimensions of 60 x 60 x 30 mm, scalped to -6.7 mm.

3

Consolidation testing revealed the least settlement associated with the coarse-grained muds,

associated with stability in situ. Hydraulic conductivities calculated produced a range from 1.4x10-9

to 0.9x10-11 m/s, with mud formed from more competent spoil typically having higher conductivities,

representative of the lower range of the material due to the effects of scalping. The large slurry

consolidometer simulating truck and shovel loading conditions determined significant pore pressures

develop in very fine-grained materials; however, negligible pressures developed within the coarse-

grained mud, highlighting the potential for safe loading in situ if managed correctly.

Shear strength testing indicated the majority of mud materials had significantly higher friction angles

than the 18⁰ that is typically assumed, with results ranging between 25⁰ and 36⁰. In contrast to the

tested spoil, the mud had lower average apparent cohesion measurements. Two mud materials with

significant degrees of degradation had friction angles lower than 15⁰, but relatively high values of

apparent cohesion. On average, dry material had higher shear strength values than wet material. The

influence of wetting and drying cycles on shear strength showed that the majority of strength

reduction occurs within the first cycle.

Correlations between the material characterisation results and the geotechnical testing allowed for the

development of a multivariate regression model to be developed, using the fractions of sand and

gravel to predict the shear strength of in-pit mud with an r2 of 0.87. This model allows for quick

estimations of mud friction angles using standardised, cheap testing methods.

2D slope stability modelling using Slide 7.0 revealed that using the parameters obtained during the

laboratory testing, there is potential for spoiling onto in-pit mud; typically found with material derived

from Category 3, competent spoil. The results highlighted the potentially conservative design that

results from the use of remoulded strength assumptions adopted from the BMA Coal spoil shear

strength framework.

This research has extensively characterised a range of in-pit muds, identifying material that has the

potential to be safely spoiled upon in-pit. It has also allowed for the rapid identification of material

prone to degradation and provided a model for predictions of shear strength with minimal testing

requirements. Application of these results will improve handling techniques, mine safety, and

treatment of in-pit mud.

4

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written

by another person except where due reference has been made in the text. I have clearly stated the

contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance,

survey design, data analysis, significant technical procedures, professional editorial advice, financial

support and any other original research work used or reported in my thesis. The content of my thesis

is the result of work I have carried out since the commencement of my higher degree by research

candidature and does not include a substantial part of work that has been submitted to qualify for the

award of any other degree or diploma in any university or other Tertiary institution. I have clearly

stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,

subject to the policy and procedures of The University of Queensland, the thesis be made available

for research and study in accordance with the Copyright Act 1968 unless a period of embargo has

been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright

holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright

holder to reproduce material in this thesis and have sought permission from co-authors for any jointly

authored works included in the thesis.

5

Publications included in this thesis

Vangsness, T.A., Williams, D.J., Ahn, J., Miles, B. (2018). Accelerated Degradation and Modified

Slake Durability Testing of Clay Mineral-rich Coal Mine Spoil. Tailings and Mine Waste Conference

2018. – partially incorporated as Chapter 6.0

Contributor Statement of Contribution

Vangsness, T.A. Laboratory Testing (60%)

Wrote and edited paper (80%)

Williams, D.J. Wrote and edited paper (20%)

Ahn, J Laboratory Testing (20%)

Miles, B Laboratory Testing (20%)

6

Other publications during candidature

Williams, D.J., Vangsness, T.A., Bergin, L., Smith, A. (2016). Shear strength characterisation of in-

pit mud to ensure lowwall stability. Tailings and Mine Waste Conference 2016.


Williams, D.J. Sample collection (30%)


Vangsness, T.A. Sample collection (30%)

Laboratory Testing (100%)


Bergin, L. Sample collection (40%)

Edited Paper (5%)

Smith, A. Edited Paper (5%)

Vangsness, T.A., Williams, D.J., Islam, S., Smith, A., Bergin, L. (2018). Consolidation and Shear

Strength Testing and Stability Analysis of Coal Mine Spoil Degraded to Mud. Tailings and Mine

Waste Conference 2018.


Vangsness, T.A. Laboratory Testing (80%)


Williams, D.J. Wrote and edited paper (20%)

Islam, S Laboratory Testing (20%)

Smith, A Edited Paper (5%)

Bergin, L Edited Paper (5%)

Contributions by others to the thesis

Professor David Williams advised on all aspects of the thesis, including design, review of published

work, and review of the final thesis.

7

Statement of parts of the thesis submitted to qualify for the award of

another degree

No works submitted towards another degree have been included in this thesis.

Research Involving Human or Animal Subjects

No animal or human subjects were involved in this research.

8

Acknowledgements

• Professor David Williams for his guidance, mentorship and council throughout my

undergraduate and postgraduate years;

• Stuart Whitton for his sharing his passion and insight into the world of soil;

• My mother, father, brother and sister for their never ending support, compassion, wisdom

and love;

• Adrian Smith of Pells Sullivan Meynink (PSM) for his guidance in project methodology and

help with interpretation of data;

• Leigh Bergin of BHP Mitsubishi Alliance (BMA) and all personnel that provided access and

collection to the researched materials;

• Mark Raven of CSIRO Land & Water for considerable input and advice with respect to the

mineralogical and geochemical analysis and interpretation.

• ACARP for its funding and support;

• The University of Queensland for access to and the use of their facilities;

• A number of undergraduate students who assisted in laboratory testing, particularly Jiwoo

Ahn, Ben Miles and Nick Hutley; and

• All members of my Academic board of supervisors.

Dicebat Bernardus Carnotensis nos esse quasi nanos, gigantium humeris

insidentes, ut possimus plura eis et remotiora videre, non utique proprii visus

acumine, aut eminentia corporis, sed quia in altum subvenimur et extollimur

magnitudine gigantea

9

Financial support

This research was supported by funding from the Australian Coal Association Research Program

(ACARP).

Keywords

Shear strength, flooding, spoil, degradation, weathering, consolidation, permeability, slope stability,

numerical simulation.

10

Australian and New Zealand Standard Research Classifications

(ANZSRC)

ANZSRC code: 090501, Civil Geotechnical Engineering, 100%

Fields of Research (FoR) Classification

FoR code: 0905, Civil Engineering, 100%

11

TABLE OF CONTENTS

1 INTRODUCTION ................................................................................................................ 33

1.1 RESEARCH BACKGROUND.................................................................................... 33

1.2 RESEARCH OBJECTIVES ........................................................................................ 35

1.3 RESEARCH SCOPE AND METHODOLOGY.......................................................... 36

1.4 RESEARCH HYPOTHESIS ....................................................................................... 36

1.5 AUSTRALIAN COAL ASSOCIATION RESEARCH PROGRAM PROJECT

C25040 ................................................................................................................................... 37

1.6 THESIS STRUCTURE ................................................................................................ 37

2 LITERATURE REVIEW.................................................................................................... 38

2.1 BOWEN BASIN OPEN-PIT COAL MINING ........................................................... 39

2.1.1 Bowen Basin geology ............................................................................................... 39

2.1.2 Open-pit strip mining ............................................................................................... 41

2.1.3 Effects of segregation ............................................................................................... 49

2.2 SPOIL PILE STABILITY AND DESIGN .................................................................. 52

2.2.1 Identification of spoil pile failure mechanisms ........................................................ 52

2.2.2 Influence of water on spoil pile stability .................................................................. 56

2.2.3 Stability related to in-pit mud and weak basal material ........................................... 61

2.2.4 Drained or undrained failure .................................................................................... 67

2.2.5 Lowwall slope performance evaluation ................................................................... 69

2.2.6 Angle of repose and its influences ........................................................................... 70

2.2.7 Shear strength and its influences .............................................................................. 72

2.2.8 Degradation of spoil and flood materials ................................................................. 75

2.3 SHEAR STRENGTH CHARACTERISATION OF BOWEN BASIN SPOI, IN-PIT

MUD AND FLOOR LAYERS .............................................................................................. 86

2.3.1 Shear strength of spoil .............................................................................................. 86

2.3.2 Shear strength of in-pit mud and basal layers .......................................................... 92

2.4 CLASSIFICATION MODELS FOR SPOIL AND IN-PIT MUD .............................. 96

12

2.4.1 BMA Coal State-of-the-Art framework for spoil categorisation ............................. 97

2.5 LITERATURE REVIEW COMMENTARY AND CONCLUSIONS ...................... 103

2.5.1 Spoil category expansion and refinement .............................................................. 103

2.5.2 Verification by testing ............................................................................................ 103

2.5.3 Material degradation considerations ...................................................................... 104

2.5.4 Consideration of in-pit mud parameters ................................................................. 104

3 MINE SITE OBSERVATIONS ........................................................................................ 105

3.1 PRELIMINARY SITE VISITS ................................................................................. 105

4 PROJECT PLAN AND RESEARCH METHODOLOGY ............................................ 109

4.1 SAMPLING METHODOLOGY ............................................................................... 109

4.2 SAMPLE IDENTIFICATION SYSTEM AND SAMPLING SUMMARY ............. 109

4.3 MATERIAL SAMPLING – MINE SITE A (26 APRIL 2015) ................................. 110

4.3.1 Mine Site A sampling locations ............................................................................. 110

4.3.2 Mine Site A samples............................................................................................... 111

4.4 MATERIAL SAMPLING – MINE SITE B (2 JUNE 2015) ..................................... 119

4.4.1 Mine Site B sampling locations.............................................................................. 119

4.4.2 Mine Site B samples ............................................................................................... 122

4.5 MATERIAL SAMPLING – MINE SITE C (3 JUNE 2015) ..................................... 128

4.5.1 Mine Site C Sampling locations ............................................................................. 128

4.5.2 Mine Site C samples ............................................................................................... 130

4.6 MATERIAL SAMPLING – MINE SITE A AND MINE SITE B (30 NOVEMBER

2016) 134

4.6.1 Mine Site A and Mine Site B sampling locations .................................................. 134

4.6.2 Mine Site A and Mine Site B samples ................................................................... 134

4.7 PHYSICAL AND CHEMICAL CHARACTERISATION ....................................... 137

4.7.1 As-sampled moisture content ................................................................................. 137

4.7.2 Specific gravity....................................................................................................... 137

13

4.7.3 Total suction ........................................................................................................... 137

4.7.4 Atterberg limits....................................................................................................... 137

4.7.5 Emerson class number ............................................................................................ 138

4.7.6 Chemical characterisation ...................................................................................... 139

4.7.7 X-ray diffraction ..................................................................................................... 139

4.7.8 Cation exchange capacity and exchangeable cations ............................................. 140

4.8 PARTICLE SIZE DISTRIBUTION .......................................................................... 142

4.9 DEGRADATION TESTING OF SPOIL ................................................................... 144

4.9.1 Varied saturation durations .................................................................................... 144

4.9.2 Multiple wetting and drying cycles ........................................................................ 144

4.9.3 Spoil degradation testing program ......................................................................... 144

4.10 GEOTECHNICAL CHARACTERISATION ............................................................ 147

4.10.1 Small-scale consolidometer testing ........................................................................ 147

4.10.2 Large slurry consolidometer ................................................................................... 147

4.10.3 Small-scale and large-scale shear strength testing ................................................. 148

5 MATERIAL CHARACTERISATION TEST RESULTS ............................................. 151

5.1 PHYSICAL CHARACTERISATION ....................................................................... 151

5.1.1 As-sampled moisture state...................................................................................... 151

5.1.2 Total suction ........................................................................................................... 153

5.1.3 Specific gravity....................................................................................................... 155

5.1.4 Particle size distributions........................................................................................ 157

5.2 GEOTECHNICAL CHARACTERISATION ............................................................ 172

5.2.1 Atterberg limits....................................................................................................... 172

5.2.2 Emerson class number ............................................................................................ 175

5.3 CHEMICAL CHARACTERISATION...................................................................... 177

5.3.1 pH, electrical conductivity and total dissolved solids ............................................ 177

5.4 MINERALOGICAL AND GEOCHEMICAL CHARACTERISATION ................. 180

14

5.4.1 X-Ray diffraction, cation exchange capacity and exchangeable cations ............... 180

5.5 MATERIAL CHARACTERISATION TEST CONCLUSIONS .............................. 186

6 DEGRADATION TEST RESULTS ................................................................................. 189

6.1 WETTING AND DRYING CYCLES, AND PROLONGED SATURATION ........ 189

6.1.1 Results of degradation testing of C3S-20 ............................................................... 190

6.1.2 Results of degradation testing of C3S-13 ............................................................... 194

6.1.3 Conclusions of prolonged saturation and wetting and drying cycle degradation testing

198

6.2 ACCELERATED DEGRADATION AND MODIFIED SLAKE DURABILITY

TESTING OF SPOIL ........................................................................................................... 200

6.2.1 Testing methodology and sampling ....................................................................... 200

6.2.2 Physical characterisation of spoil ........................................................................... 203

6.2.3 Degradation testing results ..................................................................................... 204

6.2.4 Discussion of degradation results ........................................................................... 208

6.2.5 Accelerated degradation conclusions ..................................................................... 213

7 CONSOLIDATION TEST RESULTS ............................................................................. 214

7.1 STANDARD CONSOLIDOMETER RESULTS ...................................................... 214

7.1.1 Discussion and conclusions of consolidometer test results .................................... 230

7.2 LARGE SLURRY CONSOLIDOMETER TEST RESULTS ................................... 232

7.2.1 Test results for C3M-08 ......................................................................................... 236




7.2.5 Discussion and conclusions of slurry consolidometer test results ......................... 246

8 SHEAR STRENGTH TEST RESULTS .......................................................................... 250

8.1 SPOIL AND MUD DIRECT SHEAR TEST RESULTS .......................................... 250

8.1.1 Spoil material test results ....................................................................................... 250

8.1.2 Mud material test results ........................................................................................ 255

15

8.1.3 Conclusions of direct shear test results .................................................................. 260

8.2 INFLUENCE OF PIT FLOODING ON SPOIL AND MUD SHEAR STRENGTH 262

8.2.1 C3S-13 spoil and associated mud ........................................................................... 262

8.2.2 C3S-20 spoil and associated mud ........................................................................... 264

8.3 INFLUENCE OF DEGRADATION ON SPOIL SHEAR STRENGTH .................. 266

8.3.1 Conclusions of spoil degradation shear strength test results .................................. 276

9 IN-PIT MUD CATEGORISATION, SHEAR STRENGTH ESTIMATION, AND

LOWWALL STABILITY ................................................................................................. 277

9.1 CATEGORISATION AND SHEAR STRENGTH ESTIMATION OF IN-PIT MUD

277

9.1.1 Categorisation and shear strength estimation conclusions: .................................... 290

9.2 TESTING METHODOLOGY FOR CHARACTERISING IN-PIT MUD ............... 291

9.3 STABILITY MODELLING OF IN-PIT SPOIL AND MUD ................................... 292

9.3.1 Stability modelling conclusions: ............................................................................ 304

10 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH ......... 306

10.1 CONCLUSIONS AND SIGNIFICANT OUTCOMES ............................................. 306

10.1.1 Spoil and in-pit mud characterisation ..................................................................... 306

10.1.2 Degradation of spoil ............................................................................................... 307

10.1.3 Consolidation of spoil and mud.............................................................................. 308

10.1.4 Shear strength of spoil and mud ............................................................................. 309

10.1.5 Categorisation, shear strength estimation and modelling of in-pit mud ................ 309

10.2 RECOMMENDATIONS FOR FUTURE RESEARCH ............................................ 311

11 REFERENCES ................................................................................................................... 312

16

Tables

Table 2.1 Key dragline components for strip mining, (adapted from Prytherch 2012; Simmons 2009;

Williams 2015) ................................................................................................................................... 44

Table 2.2 Suggested Factor of Safety relationships for open-pit coal mining (adapted from Simmons

2009) .................................................................................................................................................. 70

Table 2.3 Properties related to strength and slake durability (adapted from McLemore et al. 2009) 80

Table 2.4 Results of statistical analysis of spoil density and triaxial shear strength data from Bowen

Basin spoil materials (adapted from Williams & Zou 1991) ............................................................. 86

Table 2.5 Shear strength parameters derived from direct shear testing (adapted from Naderian &

Williams 1996) ................................................................................................................................... 89

Table 2.6 Bowen Basin spoil shear strengths, (adapted from Simmons & McManus 2004) ............ 90

Table 2.7 Strengths for -2.36 mm spoil specimens tested dry in a 60 mm direct shear box, reproduced

(adapted from Hiung 2016) ................................................................................................................ 91

Table 2.8 Strengths for -2.36 mm spoil specimens tested wet in a 60 mm direct shear box, reproduced

(adapted from Hiung 2016) ................................................................................................................ 91

Table 2.9 Shear strength parameters of drained residual or remoulded Bowen Basin material ........ 94

Table 2.10 Shear strength parameters of undrained Bowen Basin basal material ............................. 95

Table 2.11 Typical shear strength parameter ranges for soil (adapted from Dorador et al. 2017) .... 96

Table 2.12 BMA design parameters for Category 1 to 4 spoil in unsaturated, saturated and remoulded

states, (adapted from Simmons & McManus 2004) .......................................................................... 99

Table 4.1 Spoil and Mud Identification Details ............................................................................... 109

Table 4.2 Mine Site A sampling details ........................................................................................... 111

Table 4.3 Mine Site B sampling details ........................................................................................... 122

Table 4.4 Mine Site C samples ........................................................................................................ 130

Table 4.5 Mine Site B and Mine Site A samples (2016) ................................................................. 135

Table 4.6 Summary of physical and chemical characterisation testing ........................................... 139

Table 4.7 Summary of mineralogical and geochemical characterisation testing ............................. 141

Table 4.8 Summary of Particle Size Distribution Testing ............................................................... 143

Table 4.9 Summary of degradation testing program ........................................................................ 146

17

Table 4.10 Summary of geotechnical testing ................................................................................... 150

Table 5.1 As-sampled gravimetric moisture content of all spoil and mud samples ........................ 152

Table 5.2 As-sampled moisture state of -2.36 mm scalped spoil samples....................................... 153

Table 5.3 Specific gravity of all spoil and mud samples ................................................................. 155

Table 5.4 D90, D50, D10, Cu and Cc for Category 3 spoil wet and dry sieving .................................. 158



Table 5.7 D90, D50, D10, Cu and Cc for Category 3 mud wet sieving ............................................... 163

Table 5.8 D90, D50, D10, Cu and Cc for Category 2 desiccated mud wet and dry sieving ................. 165

Table 5.9 D90, D50, D10, Cu and Cc for Category 1 mud wet sieving ............................................... 166

Table 5.10 Atterberg limits and plasticity index for all spoil and mud samples.............................. 172

Table 5.11 Emerson class test results for all spoil samples ............................................................. 175

Table 5.12 pH, electrical conductivity and total dissolved solids for all spoil and mud samples ... 177

Table 5.13 Mineralogical analysis via X-ray diffraction for all spoil and mud samples, excluding

amorphous materials ........................................................................................................................ 181

Table 5.14 Exchangeable cations and cation exchange capacity of all spoil and mud samples ...... 183

Table 6.1 Degradation of C3S-20 subjected to prolonged soaking ................................................. 191

Table 6.2 Degradation of C3S-20 subjected to wetting and drying cycles ...................................... 192

Table 6.3 Degradation of C3S-13 subjected to prolonged soaking ................................................. 194

Table 6.4 Degradation of C3S-13 subjected to wetting and drying cycles ...................................... 195

Table 6.5 Spoil physical characterisation ........................................................................................ 203

Table 6.6 D90, D50 and D10 before and after degradation testing of fresh spoil ............................... 205

Table 6.7 X-ray diffraction analysis for all spoil samples ............................................................... 209

Table 6.8 Exchangeable cations and cation exchange capacity of all spoil samples ....................... 209

Table 6.9 Key mineralogical and geochemical characteristics of spoil samples ............................. 212

Table 7.1 Initial and final conditions for all spoil specimens tested ................................................ 214

Table 7.2 Initial and final conditions for all mud specimens tested ................................................ 215

18

Table 7.3 Settlement for all spoil specimens tested ......................................................................... 217

Table 7.4 Settlement for all mud specimens tested .......................................................................... 218

Table 7.5 Final void ratio and compression index values for all spoil and mud specimens tested . 221

Table 7.6 Coefficient of Consolidation values for all spoil and mud specimens tested .................. 224

Table 7.7 Coefficient of volume change for all spoil and mud specimens tested............................ 225

Table 7.8 Hydraulic conductivity for all spoil and mud specimens tested ...................................... 228

Table 7.9 Slurry consolidometer test results for C3M-08 ................................................................ 237

Table 7.10 Slurry consolidometer test results for C3M-18 .............................................................. 239



Table 7.13 End of testing state for all slurry consolidometer specimens tested .............................. 248

Table 8.1 BMA shear strength parameters for different categories and mobilisation modes .......... 250

Table 8.2 Direct shear strength results for spoil tested dry and wet ................................................ 251

Table 8.3 Direct shear strength results for mud tested dry and wet ................................................. 256

Table 8.4 Direct shear strength test results of C3S-13 and associated muds .................................. 263

Table 8.5 Direct shear strength test results for C3S-20 and associated mud ................................... 264

Table 8.6 Direct shear strength test results for all spoil tested dry, wet and degraded .................... 269

Table 9.1 BMA shear strength parameters for categories and mobilisation modes ........................ 279

Table 9.2 Wet sieved particle size distributions for all mud samples .............................................. 280

Table 9.3 Direct shear strength test results for mud tested wet ....................................................... 282

Table 9.4 Mud gravel and sand-size fraction correlated to friction angle and shear strength ......... 285

Table 9.5 Multivariate regression statistics for prediction of friction angle using Gravel and Sand %

for -6.7 mm fraction of mud............................................................................................................. 286

Table 9.6 Comparison of laboratory tested friction angle and shear strength and predicted values using

multivariate analysis ........................................................................................................................ 287

Table 9.7 BMA shear strength parameters for categories and mobilisation modes ........................ 292

Table 9.8 Direct shear strength of spoil tested dry and wet ............................................................. 293

Table 9.9 Direct shear strength of mud tested dry and wet .............................................................. 293

19

Table 9.10 Slide 7.0 model geometry and assumptions ................................................................... 295

Table 9.11 Factor of safety for all spoil samples with different materials at toe of lowwall ........... 298

Table 9.12 Factor of safety for spoil with associated mud at the toe of lowwall ............................ 300

Table 9.13 Factor of safety for Category 3 spoil with laboratory tested muds at the toe of lowwall

.......................................................................................................................................................... 302

20

Figures

Figure 1.1 Rainfall anomaly for Queensland (adapted from http://www.bom.gov.au/) .................... 34

Figure 1.2 – BMA Spoil Strength Categories (adapted from Simmons and McManus, 2004) ......... 35

Figure 2.1 Map of basins within eastern Australia (left), and a conceptual model for the evolution of

the Bowen Basin (right) (adapted from Fielding et al. 1996) ............................................................ 40

Figure 2.2 Box cut, strip cuts and spoil piles (adapted from Humphrey 1984) ................................. 42

Figure 2.3 Blasted overburden in a strip mine (adapted from Prytherch 2012) ................................. 43

Figure 2.4 Typical strip coal mine highwall (left) and lowwall (right) in the Bowen Basin ............. 44

Figure 2.5 Dragline dimensional extents (adapted from Prytherch 2012) ......................................... 46

Figure 2.6 Dragline working bench instability (adapted from ‘Norwich Park Dragline Recovery’

2008) .................................................................................................................................................. 46

Figure 2.7 Dragline lowwall construction, (adapted from Duran 2013) ............................................ 47

Figure 2.8 A view of a truck and shovel operation in Hunter Valley Region in NSW (adapted from

Mitra & Onargan 2012)...................................................................................................................... 48

Figure 2.9 Schematic section of dragline spoil dump fabric and phreatic surface (adapted from

Simmons & McManus 2004) ............................................................................................................. 49

Figure 2.10 Schematic section of truck spoil dump fabric and phreatic surface (adapted from

Simmons & McManus 2004) ............................................................................................................. 50

Figure 2.11 Truck end dumping in the Bowen Basin ........................................................................ 50

Figure 2.12 Conceptual model of the particle-size distribution of a rock pile (adapted from McLemore

et al. 2009).......................................................................................................................................... 51

Figure 2.13 Schematic structure of a spoil pile formed by haul trucks end-dumping from a tip-head

(adapted from Williams 2015) ........................................................................................................... 51

Figure 2.14 Rainfall anomalies for Queensland’s Bowen Basin (adapted from

http://www.bom.gov.au) .................................................................................................................... 52

Figure 2.15 Geometric features of five surveyed spoil failures (adapted from Richards et al. 1981)53

Figure 2.16 Variables involved in two-wedge limit equilibrium analysis (adapted from Richards et al.

1981) .................................................................................................................................................. 54

Figure 2.17 Superficial failure mechanism (adapted from Simmons & McManus 2004) ................. 55

21

Figure 2.18 Two-wedge dump failure due to weak floor material (adapted from Poulsen et al. 2014)

............................................................................................................................................................ 56

Figure 2.19 Intermediate scale multi-wedge failure (adapted from Simmons & McManus 2004) ... 56

Figure 2.20 A sacrificial flooded pit within the Bowen Basin........................................................... 57

Figure 2.21 Conceptual hydrogeological model of a stockpile (adapted from Beale 2017) ............. 58

Figure 2.22 Hydrostratigraphy of segregated dumps – terraced construction (adapted from Smith et

al. 1995) ............................................................................................................................................. 59

Figure 2.23 Three-zone model for moisture conditions within spoil dumps (adapted from Simmon &

Fityus 2016) ....................................................................................................................................... 60

Figure 2.24 Spoiling into in-pit mud, adapted from (Prytherch 2012) .............................................. 62

Figure 2.25 Mud cleanout via the key bridge method (adapted from Prytherch 2012) ..................... 63

Figure 2.26 Potential geotechnical hazards for dragline in-pit bench operations (adapted from

Simmons 2009) .................................................................................................................................. 63

Figure 2.27 In-pit mud-dam storage observed in the Bowen Basin................................................... 64

Figure 2.28 Geotechnical instabilities related to weak basal material (adapted from Prytherch 2012)

............................................................................................................................................................ 66

Figure 2.29 Concept of mechanism by which undrained behaviour is invoked at base of spoil pile

(adapted from Duran 2013) ................................................................................................................ 68

Figure 2.30 Annual probability of failure versus consequence for various engineering structures

(adapted from Whitman 1984) ........................................................................................................... 69

Figure 2.31 Angle of repose of granular materials (adapted from Simons & Albertson 1960)......... 71

Figure 2.32 Schematic frequency distribution for natural hillslope angles in the United Kingdom

(adapted from Carson & Petley 1970) ............................................................................................... 72

Figure 2.33 Standard direct shear apparatus ...................................................................................... 73

Figure 2.34 Shear characteristics of dense and loose sands (adapted from Jackson 2015) ............... 74

Figure 2.35 Peak and residual shear strength and Coulomb envelopes (adapted from Jackson 2015)

............................................................................................................................................................ 74

Figure 2.36 Factors affecting weathering (adapted from Lan et al. 2003) ......................................... 76

Figure 2.37 Degradation of stable aggregate due to slaking and dispersion ...................................... 77

22

Figure 2.38 Weak cementation of spoil within the Bowen Basin ...................................................... 78

Figure 2.39 Slaking of clay-rich mine spoil from the Bowen Basin.................................................. 79

Figure 2.40 Erosion of dispersive spoil pile observed within the Bowen Basin ............................... 83

Figure 2.41 Scheme for determining class numbers of aggregates (adapted from Emerson 1967) .. 83

Figure 2.42 The influence of remoulding water content on dispersion (left), and influence of

exchangeable sodium on dispersion (right) (adapted from Seedsman & Emerson 1985) ................. 84

Figure 2.43 Effect of saturation on the shear strength of Bowen Basin spoil and a comparison with

other values determined either experimentally or by back-analysis of spoil pile instability (adapted

from Seedsman et al. 1988) ................................................................................................................ 87

Figure 2.44 Compilation of the strength of rockfill as measured in large triaxial tests (adapted from

Leps 1970), compared with direct shear values for coal mine spoil (adapted from Seedsman et al.

1988) .................................................................................................................................................. 88

Figure 2.45 Calculated shear strength of 100m deep spoil pile using average shear strength parameters

(adapted from Kho et al. 2013) .......................................................................................................... 90

Figure 2.46 Comparison of friction angle and cohesion for Boyd et al. (1978), Seedsman et al. (1988),

Simmons & McManus (2004) and Hiung (2016) .............................................................................. 92

Figure 2.47 Shear strength of bulk spoil (a) unsaturated – as sampled; (b) saturated (adapted from

Richards et al. 1981) .......................................................................................................................... 93

Figure 2.48 Spoil categories and attributes (adapted from Simmons & McManus 2004)................. 98

Figure 2.49 Spoil structure attribute to be used with Figure 2.48 (adapted from Simmons & McManus

2004) .................................................................................................................................................. 98

Figure 2.50 Degradation of dry Category 3 spoil (left) to mud (right) .............................................. 99

Figure 2.51 Conceptual strength modes for spoil (adapted from Simmons & McManus 2004) ..... 100

Figure 2.52 Conceptual strength modes for spoil modified to explain the linear shear strength

approximation adopted in framework (adapted from Bradfield et al. 2013) ................................... 100

Figure 2.53 Mohr Diagram showing framework linear fit with respect to actual strength envelope (not

to scale) (adapted from Bradfield et al. 2013) ................................................................................. 101

Figure 3.1 Mine Site C Ramp 23, spoil failure and draglining into mud producing slumping of spoil

and bow-waving of mud .................................................................................................................. 105

Figure 3.2 Mine Site C Ramp 23, Category 3 (left) and Category 4 and 2 (right) spoil ................. 106

23

Figure 3.3 Mine Site C Ramp 22, flooded pit (~40 m deep) ........................................................... 106

Figure 3.4 Mine Site C Ramp 6S, floor mud from Category 3 spoil ............................................... 107

Figure 3.5 Mine Site B Ramp 11S, closed by flooding since late 2010 .......................................... 107

Figure 3.6 Mine Site B Ramp 5S, flooded pit surrounded by largely Category 3 overburden and spoil

.......................................................................................................................................................... 108

Figure 3.7 Mine Site B Ramp 5S, in-pit mud at base of Category 3 overburden and spoil lowwall

.......................................................................................................................................................... 108

Figure 4.1 Mine Site A sampling locations...................................................................................... 110

Figure 4.2 Mine Site A Ramp 10N sampling location C3M-01 ...................................................... 111

Figure 4.3 Mine Site A Ramp 10N sample C3M-01 ....................................................................... 112


Figure 4.5 Mine Site A Ramp 10N Sample C1M-02 ....................................................................... 113





Figure 4. 10 Mine Site A Ramp 10N sampling location C3M-05 ................................................... 115

Figure 4.11 Mine Site A Ramp 10N sample C3M-05 ..................................................................... 116

Figure 4.12 Mine Site A Ramp 10N sampling Location C2M-06 ................................................... 116


Figure 4.14 Mine Site A Ramp 10N sampling Location C2M-07 ................................................... 117


Figure 4.16 Mine Site B Ramp 5S sampling ................................................................................... 119

Figure 4.17 Mine Site B Ramp 5S sampling location C3M-08 ....................................................... 119

Figure 4.18 Mine Site B Ramp 5S sampling location C3S-10 ........................................................ 120




24


Figure 4.23 Mine Site B Ramp 5S Sample C3M-08 surface texture (with 20-cent coins for scale)

.......................................................................................................................................................... 122

Figure 4.24 Mine Site B Ramp 5S sampling location C3S-10 surface texture ................................ 123

Figure 4.25 Mine Site B Ramp 5S sample C3S-10 sieving to -53 mm ........................................... 123

Figure 4.26 Mine Site B Ramp 5S sample C3M-12 surface texture................................................ 124

Figure 4.27 Mine Site B Ramp 5S sample C3S-13 fine-grained surface texture ............................ 124

Figure 4.28 Mine Site B Ramp 5S sample C3S-13 coarser-grained below surface ........................ 125

Figure 4.29 Mine Site B Ramp 5S sample C3S-13: (a) -53 mm, and (b) +53 mm ......................... 125

Figure 4.30 Mine Site B Ramp 5S sample C3S-13: +53 mm .......................................................... 126

Figure 4.31 Mine Site B Ramp 5S sample C2S-16 agglomerated surface texture .......................... 126

Figure 4.32 Mine Site B Ramp 5S sample C1S-17.......................................................................... 127

Figure 4.33 Mine Site B Ramp 1N sample C1S-17 surface texture ................................................ 127

Figure 4.34 Mine Site C sampling location C3M-18 ....................................................................... 128

Figure 4.35 Mine Site C sampling location C3S-20 ........................................................................ 128



Figure 4.38 Mine Site C Ramp 6S sample C3M-18 surface crusting .............................................. 130

Figure 4.39 Mine Site C Ramp 6S sample C3S-20 surface PSD (with 22.9 cm diameter plates for

scale) ................................................................................................................................................ 131

Figure 4.40 Mine Site C Ramp 5S sample C3S-20: (a) -53 mm, and (b) +53 mm ......................... 131

Figure 4.41 Mine Site C Ramp 5S sample C3S-20 weighing + & -53 mm fractions ...................... 132

Figure 4.42 Mine Site C Ramp 5S sample C3S-20 +53 mm ........................................................... 132

Figure 4.43 Mine Site C Ramp 22 sample C1M-23 ........................................................................ 133

Figure 4.44 Mine Site C Ramp 14 sample C2S-24 .......................................................................... 133


Figure 4.46 Mine Site A Ramp 50S sampling location C3M-32 ..................................................... 134

Figure 4.47 Mine Site B Ramp 5S sample C3M-30 ........................................................................ 135

25

Figure 4.48 Mine Site A Ramp 50S sample C3M-32 ...................................................................... 136

Figure 4.49 Helium pycnometer ...................................................................................................... 137

Figure 4.50 WP4 dewpoint potential meter ..................................................................................... 138

Figure 4.51 Atterberg limit test apparatus ....................................................................................... 138

Figure 4.52 Wet sieving apparatus (left), addition of suspension solution to stack (top right), filtering

of sieved sample for drying and weighing (bottom right) ............................................................... 142

Figure 4.53 Agitation of 1000cc solution (left) & hydrometer analysis of solution with the control

cylinder and temperature gauge (right) ............................................................................................ 143

Figure 4.54 Slake durability apparatus ............................................................................................ 145

Figure 4.55 Schematics of consolidometer testing in a water bath (tested “wet”) .......................... 147

Figure 4.56 Large slurry consolidometer apparatus schematic........................................................ 148

Figure 4.57 Schematic of direct shear box shear strength test ......................................................... 149

Figure 4.58 Large-scale direct shear machine (300 mm x 300 mm x 200 mm high) ...................... 149

Figure 5.1 As-sampled moisture content of all spoil and mud samples .......................................... 153

Figure 5.2 As-sampled gravimetric moisture content and total suction of all spoil samples .......... 154

Figure 5.3 Specific gravity of all spoil and mud samples ................................................................ 156

Figure 5.4 Overall particle size distribution curves of Category 3 spoil -53 mm fraction .............. 157



Figure 5.7 Overall particle size distribution curves of all dry spoil -53 mm fraction ...................... 161

Figure 5.8 D90 values for all dry and wet sieved spoil samples ....................................................... 162

Figure 5.9 Overall particle size distribution curves of all spoil wet and dry -53 mm fraction ........ 162

Figure 5.10 Overall particle size distribution curves of all Category 3 mud samples -53 mm fraction

.......................................................................................................................................................... 163

Figure 5.11 Overall particle size distribution curves of Category 2 dried mud -53 mm fraction .... 164

Figure 5.12 Overall particle size distribution curves of all Category 1 mud -53 mm fraction ........ 165

Figure 5.13 Overall particle size distribution curves of all mud samples -53 mm fraction ............. 166

Figure 5.14 D90 values for all wet sieved spoil and mud samples ................................................... 167

26

Figure 5.15 Overall particle size distribution curves of all Category 3 spoil and mud samples -53 mm

fraction ............................................................................................................................................. 168


fraction ............................................................................................................................................. 168


fraction ............................................................................................................................................. 170

Figure 5.18 Overall particle size distribution curves Mine Site B R5S spoil and mud -53 mm fraction

.......................................................................................................................................................... 170

Figure 5.19 Overall particle size distribution curves Mine Site C R6S spoil and mud -53 mm fraction

.......................................................................................................................................................... 171

Figure 5.20 Overall particle size distribution curves Mine Site A mud -53 mm fraction ............... 171

Figure 5.21 As-sampled gravimetric moisture content and Atterberg limits of all spoil and mud

samples ............................................................................................................................................. 173

Figure 5.22 Plasticity chart for all spoil and mud samples .............................................................. 174

Figure 5.23 Liquid and plastic limits for all spoil and mud samples ............................................... 174

Figure 5.24 C3S-10 and C3S-13 Emerson class test results ............................................................ 176



Figure 5.27 Variation of gravimetric moisture content and electrical conductivity for all spoil and

mud samples ..................................................................................................................................... 178

Figure 5.28 Variation of gravimetric moisture content and pH for all spoil and mud samples ....... 179

Figure 5.29 Variation of pH and electrical conductivity for all spoil and mud samples ................. 179

Figure 5.30 Mineralogical analysis via X-ray diffraction for all spoil and mud samples ................ 182

Figure 5.31 Cation exchange capacity and exchangeable cations for all spoil and mud samples ... 184

Figure 5.32 Calculated levels of Smectite and Illite for all spoil and mud samples ........................ 185

Figure 5.33 Smectite cations present in all spoil and mud samples ................................................. 185

Figure 6.1 C3S-20 prolonged saturation preparation ....................................................................... 190

Figure 6.2 Plastic-wrapped to avoid evaporation during submersion .............................................. 190

Figure 6.3 Degradation of C3S-20 when subjected to prolonged saturation ................................... 191

27

Figure 6.4 Degradation of C3S-20 when subjected to wetting and drying cycles ........................... 192

Figure 6.5 Electrical conductivity measurements of C3S-20 for soaking duration and wet/dry cycles

.......................................................................................................................................................... 193

Figure 6.6 Comparison of wetting and drying cycles versus saturation duration for C3S-20 ......... 194

Figure 6.7 Degradation of C3S-13 when subjected to prolonged soaking ...................................... 195

Figure 6.8 Degradation of spoil sample C3S-13 subjected to wetting and drying cycles ............... 196

Figure 6.9 Electrical conductivity measurements of C3S-13 for saturation duration and wet/dry cycles

.......................................................................................................................................................... 197

Figure 6.10 Comparison of wetting and drying cycles versus soaking duration for C3S-13 .......... 197

Figure 6.11 Comparison of C3S-13 and C3S-20 during prolonged soaking and wetting and drying

cycles ................................................................................................................................................ 199

Figure 6.12 Spoil material C3S-13 .................................................................................................. 200





Figure 6.17 As-sampled moisture content and Atterberg limits ...................................................... 203

Figure 6.18 Plasticity index versus liquid limit for all spoil ............................................................ 204

Figure 6.19 Particle size distribution of C3S-13 before and after degradation................................ 206

Figure 6.20 Particle size distribution of C3M-20 before and after degradation .............................. 206




Figure 6.24 Influence of degradation method on particle size reduction ......................................... 211

Figure 6.25 Modified slake durability degradation analysis per cycle ............................................ 211

Figure 7.1 Final dry density for all spoil and mud specimens tested ............................................... 216

Figure 7.2 Initial and final void ratio for all spoil and mud specimens tested ................................. 217

Figure 7.3 Settlement for -4.7 mm all loose-placed spoil specimens tested wet ............................. 218

28

Figure 7.4 Settlement for -4.7 mm all loose-placed mud specimens tested wet .............................. 219

Figure 7.5 Comparison of settlement for -4.7 mm spoil and mud specimens tested wet ................ 220

Figure 7.6 Initial dry density versus settlement at 1,000 kPa stress for all spoil and mud specimens

tested ................................................................................................................................................ 220

Figure 7.7 Applied stress versus void ratio for all spoil specimens tested ...................................... 222

Figure 7.8 Applied stress versus void ratio for all mud specimens tested ....................................... 222

Figure 7.9 Compression index values for all spoil and mud specimens tested ................................ 223

Figure 7.10 Typical calculation of the coefficient of consolidation for an oedometer specimen .... 225

Figure 7.11 Void ratio versus coefficient of volume change for -4.7 mm spoil tested wet ............. 226

Figure 7.12 Void ratio versus coefficient of volume change for -4.7 mm mud specimens tested wet

.......................................................................................................................................................... 227

Figure 7.13 Comparison of void ratio versus coefficient of volume change for all -4.7 mm spoil and

mud specimens tested wet ................................................................................................................ 227

Figure 7.14 Hydraulic conductivity values for all spoil and mud specimens tested ........................ 229

Figure 7.15 Hydraulic conductivity at 1,000 kPa applied stress for all spoil and mud specimens tested

.......................................................................................................................................................... 230

Figure 7.16 C3M-08 sampling ......................................................................................................... 234




Figure 7.20 Particle size distribution of samples tested in slurry consolidometer ........................... 236

Figure 7.21 Slurry consolidometer stress and pore water pressure plots for C3M-08 ..................... 237

Figure 7.22 Slurry consolidometer void ratio versus effective stress for C3M-08 .......................... 238

Figure 7.23 Slurry consolidometer hydraulic conductivity versus effective stress for C3M-08 ..... 238


Figure 7.25 Slurry consolidometer void ratio versus effective stress for C3M-18 .......................... 240



29

Figure 7.28 Slurry consolidometer void ratio versus effective stress for C1S-02 ........................... 243


Figure 7.30 Slurry consolidometer stress and pore water pressure data for C1S-23 ....................... 245

Figure 7.31 Slurry consolidometer void ratio versus effective stress plot for C1S-23 .................... 245

Figure 7.32 Slurry consolidometer hydraulic conductivity versus effective stress plot for C1S-23 246

Figure 7.33 Slurry consolidometer settlement versus time for all mud specimens tested ............... 247

Figure 8.1 Direct shear strength results for Category 3 spoil tested dry and wet ............................ 252



Figure 8.4 Apparent cohesion and friction angle for all spoil ......................................................... 254

Figure 8.5 Secant friction angle versus applied normal stress for Category 1, 2 and 3 spoil tested dry

and wet ............................................................................................................................................. 255

Figure 8.6 Direct shear strength results of Category 3 mud tested dry and wet .............................. 257



Figure 8.9 Friction angle and apparent cohesion for all mud specimens tested .............................. 259

Figure 8.10 Secant friction angle versus applied normal stress for Category 1, 2 and 3 mud samples

tested dry and wet ............................................................................................................................ 260

Figure 8.11 Comparison of spoil and mud apparent cohesion and friction angle ........................... 261

Figure 8.12 C3S-13, associated mud C3M-08 (lowwall) and C3M-30 (highwall) sampling locations

.......................................................................................................................................................... 262

Figure 8.13 Direct shear strength test results for C3S-13 spoil and associated C3M-08 mud ........ 263

Figure 8.14 C3S-20 and associated mud C3M-18 sampling locations ............................................ 264

Figure 8.15 Direct shear strength test results for C3S-20 spoil and associated C3M-18 mud ........ 265

Figure 8.16 C3S-13 as-sampled and after three wetting and drying cycles ..................................... 266




30


Figure 8.21 Particle size distribution for C3S-13 scalped to pass 6.7 mm and tested dry, soaked and

after wet/dry cycles .......................................................................................................................... 270

Figure 8.22 Direct shear strength results for C3S-13 scalped to pass 6.7 mm and tested dry, wet and

degraded ........................................................................................................................................... 270

Figure 8.23 Particle size distribution of C3S-20 scapled to pass 6.7 mm and tested dry, soaked and



degraded ........................................................................................................................................... 271

Figure 8.25 Particle size distribution of C3S-16 spoil scalped to pass 6.7 mm and tested dry, soaked

and after wet/dry cycles ................................................................................................................... 272


degraded ........................................................................................................................................... 273

Figure 8.27 Particle size distribution of C3S-24 scalped to pass 6.7 mm and tested dry, soaked and



degraded ........................................................................................................................................... 274

Figure 8.29 Particle size distribution of C3S-17 scalped to pass 6.7 mm and tested dry, soaked and



degraded ........................................................................................................................................... 275

Figure 9.1 Spoil Categories and Attributes (adapted from Simmons & McManus 2004)............... 278

Figure 9.2 Spoil Structure attribute to be used in association with Figure 9.1 (adapted from Simmons

& McManus 2004) ........................................................................................................................... 278

Figure 9.3 Coarse and fine fractions of degraded spoil ................................................................... 279

Figure 9.4 Overall particle size distribution curves of all mud samples .......................................... 281

Figure 9.5 Source material category compared with D90 ................................................................. 281

Figure 9.6 Apparent cohesion versus friction angle for all mud specimens .................................... 283

Figure 9.7 Secant friction angle versus applied normal stress for Category 1, 2 and 3 mud dry and wet

.......................................................................................................................................................... 284

31

Figure 9.8 A model for predicting the friction angle of in-pit mud ................................................. 288

Figure 9.9 Comparison of laboratory tested friction angle and predicted friction angle of mud

materials ........................................................................................................................................... 289

Figure 9.10 Comparison of laboratory tested shear strength and predicted shear strength assuming

zero cohesion for all mud samples ................................................................................................... 289

Figure 9.11 Two-wedge spoil pile failure mechanism (adapted from Philip et al. 1981) ............... 294

Figure 9.12 Dozer push lowwall geometry ...................................................................................... 295

Figure 9.13 Truck and shovel lowwall geometry ............................................................................ 296

Figure 9.14 Dragline lowwall geometry .......................................................................................... 296

Figure 9.15 Dragline lowwall with undercut toe geometry ............................................................. 296

Figure 9.16 Factor of safety for dragline lowwall with different base material at the toe ............... 299

Figure 9.17 Factor of Safety for Category 3 spoil with associated mud at the toe of a dragline lowwall

.......................................................................................................................................................... 300

Figure 9.18 Factor of safety for Category 3 spoil with different mud at the toe of lowwall ........... 303

Figure 9.19 Factor of safety comparison for Category 3 spoil against mud equivalent spoil with

different mud at toe of lowwall ........................................................................................................ 303

32

List of Abbreviations

ACARP – Australian Coal Association Research Program

ANZSRC – Australian and New Zealand Standard Research Classifications

BMA – BHP Mitsubishi Alliance

BOM – Bureau of Meteorology

CEC – Cation Exchange Capacity

CSIRO - Commonwealth Scientific and Industrial Research Organisation

D10 – Particle size 10% of material passes

EC – Electrical Conductivity

FOR – Fields of Research

FOS – Factor of Safety

LL – Liquid Limit

PI – Plasticity Index

PL – Plastic Limit

POF – Probability of Failure

PSD – Particle Size Distribution

PSM – Pells Sullivan Meynink

UDC – Utah Development Company

33

1 INTRODUCTION

Large flooding events can lead to the degradation of spoil and floor materials in open-cut mines

creating in-pit ‘mud’. Due to difficulty in sampling and associated costs, key geotechnical parameters

of in-pit mud have been derived from back-analysed failures with limited laboratory testing for

confirmation (Mallett et al 1983; Williams 2015). Best practice involves the removal of water and

mud prior to lowwall spoil dumping. This is a costly exercise and has associated risks which could

potentially be mitigated through better understandings of the geotechnical properties of the in-pit mud

and how it forms.

1.1 Research Background

Lowwall geotechnical instability due to the formation of in-pit mud has been a concern within the

industry since the commencement of large-scale open strip mining in the Bowen Basin during the wet

period of the early 1970s, which was met with a spate of lowwall failures (Gonano 1980). A rise in

failure rates was observed again in the late 1990s, and yet again in the period from 2008 to 2013

(Williams 2015); all coinciding with particularly wet climatic periods, as shown in Figure 1.1,

reproduced from the Bureau of Meteorology.

Investigations into these and subsequent lowwall failures led to a number of recommendations for

spoiling, including:

• removal water from the pit where degradable floor materials are present, prior to the dumping

of lowwall spoil;

• removal any mud that has formed and water from the pit, prior to the dumping of lowwall

spoil; and

• avoiding placement of spoil such as problematic Tertiary or weathered materials at the base

of the spoil pile, as such materials are prone to degradation on wetting, which can lead to

failure of the subsequently placed lowwall.

The removal of mud and water from the pit prior to spoiling is costly and is not always possible where

access is restricted, or there are concerns for safety. There is a need to better characterise those spoil

and floor materials that have a low potential to weaken on exposure to water, as their recognition may

reduce or eliminate the mud removal requirements.

34

Figure 1.1 Rainfall anomaly for Queensland (adapted from http://www.bom.gov.au/)

For the characterisation of in-pit mud, use will be made of the State-of-the-Art BMA Coal Spoil

Categories, published by Simmons & McManus (2004), as shown in Figure 1.2. This categorisation

methodology has been adopted throughout the Bowen Basin Coalfields, and to some extent in the

Hunter Valley Coalfields.

The focus of the research will be sampling fresh and wet spoil and floor materials that have the highest

potential to maintain strength on breaking down within a flooded pit, which will likely comprise

Category 3 and 4 spoil, however, fresh and water-softened Category 1 and 2 spoil will also be

sampled. Investigations will involve testing of the physical, chemical, mineralogical, geochemical

and geotechnical properties of these materials, with focus placed on how they form, how to handle

them appropriately when they do.

35

Figure 1.2 – BMA Spoil Strength Categories (adapted from Simmons and McManus, 2004)

1.2 Research Objectives

Past research has been conducted investigating the shear strength properties of spoil from the Bowen

Basin both fresh and degraded, however, data is limited with respect to in-pit mud. The majority of

shear strength parameters used in design have been derived from back-analysed failures. Due to the

potential benefits of improved characterisation of this material, and the importance of improving mine

safety, the objectives of this research are:

• to geologically identify and sample selected fresh and degraded spoil and floor materials to

assess their potential for water-softening;

• to characterise physically, chemically and mineralogically in the laboratory the representative

materials sampled, including testing for slake durability;

• to simulate degradation of spoil in a laboratory environment with aims of identifying

relationships between the material characteristics and their resultant parameters;

• to test spoil and floor materials under consolidation using standard oedometer’s and a large

slurry consolidometer to determine parameters relevant to settlement and consolidation under

loading, and to understand the pore water pressures within the materials;

• to carry out laboratory shear strength testing on fresh, moistened and water-softened

specimens of the spoil and floor materials sampled;

• to relate the laboratory shear strength of the spoil and floor materials tested to their physical,

chemical and mineralogical characteristics; and

36

• to develop testing protocols for the identification of degradable spoil and floor materials,

allowing for the development and improvement of design guidelines for enhancing the

geotechnical stability of lowwalls for both durable and water-softened spoil.

1.3 Research Scope And Methodology

To investigate the properties of in-pit mud, and the influences they have on stability, the methodology

is as follows:

• Representative mud and spoil sampling, with an emphasis on material likely to maintain

structure and strength during wetting;

• Laboratory physical and chemical characterisation testing;

• Laboratory mineralogical and geochemical testing;

• Laboratory geotechnical investigations including degradation, consolidation and shear

strength testing;

• Framework development to estimate in-pit mud parameters using relationships found between

the material characteristics and the geotechnical properties; and

• Stability modelling to determine the impact the tested spoil and mud properties have on

stability, and to provide insight into improvements of the current assumptions used by

industry.

1.4 Research Hypothesis

• Upon wetting, coal mine spoil and pit floors will exhibit varied degrees of degradation related

to the state of weathering, degree of blasting, time of exposure, and the geological and

mineralogical properties, resulting in a loss of shear strength;

• Wetting and drying cycles will increase the rate of degradation of coal mine spoil, affecting

materials with higher fractions of swelling clays more severely;

• The shear strength of in-pit mud will be related to the material source, allowing for the

identification of spoil that will not break down rapidly, and will be suitable for spoiling upon

if handled correctly; and

• Improved knowledge of the geotechnical properties of in-pit mud will allow for increases in

modelling accuracy and the development of in situ handling methodologies.

37

1.5 Australian Coal Association Research Program Project C25040

This research has been funded by the Australian Coal Association Research Program (ACARP)

Project C25040. It builds upon the related 4-year ACARP Project C19022 on the Implications of

Settlements of High Coal Mine Spoil on Stored Volume and Stability undertaken by Professor David

Williams of The University of Queensland (UQ) from 2010 to 2014, and the 3-year ACARP Project

C20019 on the Stability of very High Spoil Piles, undertaken by Dr John Simmons of Sherwood

Geotechnical and Research Services and Professor Stephen Fityus of the University of Newcastle.

Both of these projects involved the laboratory shear strength testing of a range of spoil materials,

including those that degrade on contact with water.

1.6 Thesis Structure

This thesis consists of multiple chapters investigating specific aspects of the characterisation and

testing of in-pit mud, and sampled spoil materials. The topics of each chapter are as follows:

• Section 1: A project introduction detailing scope and objectives;

• Section 2: A literature review including commentary;

• Section 3: Mine site observations of in-pit mud at numerous sites;

• Section 4: Material sampling, project plan and testing methodologies;

• Section 5: Physical, chemical, mineralogical and geochemical characterisation of spoil and

mud;

• Section 6: Degradation testing of spoil;

• Section 7: Standard consolidation and large slurry consolidation of spoil and mud;

• Section 8: Direct shear testing of spoil and mud;

• Section 9: Prediction of mud shear strength and implications of results on lowwall stability;

• Section 10: Conclusions and future research; and

• Section 11: References.

38

2 LITERATURE REVIEW

Weak foundations and floor materials below dumped spoil have been the cause for many significant

coal mine lowwall failures in the Bowen Basin (Nguyen & Welsh 1981; Richards et al. 1981;

Seedsman et al. 1988). There is a concern that mud produced by the degradation of in-pit spoil and

floor materials could cause instabilities in future lowwalls if spoil is placed on top. Characterisation

of degraded material is most often through the interpretation of back-analysed failures (Gonano 1980;

Seedsman et al. 1988). The locations of these weak materials often present safety issues with respect

to access. Furthermore, due to the rate of mining, economics and standard operating procedures,

degraded in-pit mud is rarely tested with strength parameters assumed for design. Another reason for

the limited information available is the reluctance of the mining community to make information

about lowwall failures public. There is currently work underway to improve knowledge of these

problematic materials and to better understand all the issues involved (Williams 2015; Simmon &

Fityus 2016).

Section 2.1 discusses the open-pit coal mining operations of the Bowen Basin in Australia. This

includes the geological aspects of coal formation, the methodology behind open-pit strip mining, and

the impacts typical handling processes have on the physical structure of a lowwall.

Section 2.2 investigates spoil pile stability and design within the Bowen Basin, reviewing the causes

of past failures and the derived parameters related to their occurrence. The review includes an

investigation into failure mechanisms, the occurrence and impact of water within a spoil pile, the

influential parameters of the spoil, and the degradation mechanisms that are associated with the

formation of in-pit mud.

Section 2.3 reviews the literature around shear strength characterisation of Bowen Basin spoil, in-pit

mud and basal layers within lowwalls. Results of past testing have been compiled in relation to the

strength of in-pit mud, highlighting areas that require further investigation.

Section 2.4 discusses the classification and State-of-the-Art categorisation of spoil and in-pit mud,

looking at standardised practice, and systems utilised by mines within the Bowen Basin for visual

categorisation.

Section 2.5 summarises the findings of the literature review, highlighting areas of research that will

improve the stability of lowwalls through improved classification and categorisation of in-pit mud.

39

2.1 Bowen Basin Open-pit Coal Mining

Commercial coal mining has taken place in the Bowen Basin since the 1890s, however, large-scale

mining has commenced since the mid to late 1970s (Richards et al. 1981). Due to the nature of the

coal formation, the size of the waste dumps created are also increasing. Spoil dump heights have

already reached above 350 m, with plans to go higher being implemented (Simmons 2009). The

stability of these dumps is dependent on a number of factors, including geometry, construction

methods, and the parameters of the spoil, which is largely influenced by its geological history. This

section focuses on the formation of coal measures and their relevance to lowwall stability.

2.1.1 Bowen Basin geology

The geology of Australia has been discussed in depth by Fielding et al. (1996) and Mallett et al (1983),

who both detail the large amount of valuable minerals within the Bowen Basin, particularly fossil

fuels. It contains the largest black coal reserves in Australia and is one of the world’s largest deposits.

Its general layout has been detailed in Figure 2.1. The Bowen Basin is up to 10 km thick, composed

of Permian and Triassic aged material (Fielding et al. 1996), situated in a sub-humid, sub-tropical

climate. The exposed formation is 550 to 600 km long, and 250 km wide. A comprehensive study on

the formation and sedimentological features of the Bowen Basin was conducted by Mallett et al

(1983). Of consideration for mining operations were the identification of weak lacustrine beds and

the presence of finely interbedded sequences vulnerable to damage during mine blasting. Steeply

dipping inter-seam units were also recognised as potential causes of instability.

2.1.1.1 Coal formation

The coal found in the Bowen Basin is created through the burial of peat mires in alluvial or estuarine

depositional systems, considered low energy environments. As these mires are buried beneath soil,

they become compressed under increasing temperature and pressure (Christoulas et al. 1987). The

biological and geological processes that take place result in the formation of peat, which then turns

into coal (Taylor et al. 2009). A conceptual model of the evolution of the Bowen Basin is shown in

Figure 2.1.

For the formation of commercially usable coal to occur, the range of acceptable temperatures and

pressures is quite small. This is important as it means the parameters of coal measures overburden

are related more to geological attributes than the geographical ones (Simmons & McManus, 2004).

2.1.1.2 Geotechnical considerations of coal measures

Coal measures define the carboniferous system in which coal forms. In a low energy environment,

the coal measures typically consist of fine- and coarse-grained sedimentary rocks, including but not

limited to claystone, mudstone, siltstone, shale and sandstone (Rayner & Hemingway 1974). These

40

sedimentary rocks are usually considered to have low strength, and to be anisotropic. Upon the

excavation of the coal seam, the spoil produced from the coal measures overburden will likely have

low strength, with variable susceptibility to degradation on exposure to water and oxygen based on

the anisotropic sedimentation. The environments in which coal seams usually form also contain clay

which can form as bands within and around the seam. Not all clay bands lithify, and due to overburden

stress and compaction, these bands of clay are often found in a sheared state (Simmons 2009).

Figure 2.1 Map of basins within eastern Australia (left), and a conceptual model for the

evolution of the Bowen Basin (right) (adapted from Fielding et al. 1996)

Over geological time, coal measures are exposed to weathering, erosion and oxidation. The extent of

these can vary and will be pronounced along structural defects such as faults (McLemore et al. 2009).

This creates requirements for site-specific considerations of the geological formations and the impacts

they can have on the structural stability of the excavation and the dumped spoil. Formations such as

bedding planes that have been exposed to weathering can act as potential slipping planes (Simmons

2009).

41

2.1.2 Open-pit strip mining

Open-pit strip mining is a common technique for accessing coal close to the surface (Atwood 1975).

To collect the coal in an open-pit mine, the overburden must be removed from above the coal seam.

This removal is associated with processes such as drilling, blasting, loading and hauling (Hartman &

Mutmansky 2002). These processes will impact the parameters of the spoil based on the type of

explosives used, the magnitude of the explosions and the method of spoil transportation.

Open-cut strip mining starts with an initial box cut into the overburden extending along the strike of

the coal seam. This box cut is then extended in the direction of the coal seam dip, with the wall above

the coal referred to as the highwall (Atwood 1975). Overburden can be removed with a number of

machines including but not limited to draglines, trucks and shovels, excavators, dragline hoppers, and

varied sizes of dozers depending on the suitability of the mine site and its resources (Mitra & Onargan

2012). Once excavated, the highwall is either moved behind the progressing machinery and placed

into a spoil pile referred to as the lowwall, or it is dumped externally. An example of this and the

associated terminology is provided in Figure 2.2, showing the fundamentals of dragline excavation

detailed in Section 2.1.2.3.

Depending on the depth of the seam and the rate of mining, the strip mining method can result in

millions of cubic meters of spoil production. The spoil that is produced will consist of a large range

of rocks and soils based on the geographical and geological location of the mine and the mining

process used. Spoil is considered to be highly heterogeneous, and the degree of heterogeneity will

increase with time in terms of composition and particle size (Fityus et al. 2008). This will be

associated with increasing complexity in modelling of the lowwall, and determination of appropriate

material parameters.

42

Figure 2.2 Box cut, strip cuts and spoil piles (adapted from Humphrey 1984)

2.1.2.1 Influence of blasting

To access the coal seam, the coal measures overburden must be removed. In an open-cut strip mine,

this is most commonly conducted through drilling and blasting. This involves the placement of

explosives into the overburden, which is then ignited in a controlled manner, fracturing the rock. The

blast itself must be carefully engineered to avoid dilution and ore loss, accounting for factors such as

the amount of overburden, economics, emissions, design requirements, rock density and the rocks

uniaxial compressive strength (UCS) (McLemore et al. 2009). An example of overburden post-

blasting has been provided in Figure 2.3. Due to the anisotropic nature of coal measures, accurate

blasting can be difficult, with efficacy largely influenced by geological features and the skill of the

engineers.

The magnitude of the blast will have a significant influence on the degree of fracturing, and hence,

the particle size of the spoil produced. Excessive blasting force will result in an overproduction of

fines. If too little force is used, the overburden may not be sufficiently disintegrated, resulting in

hazardous scenarios that may require re-drilling and re-blasting, as well as increased difficulty in

handleability for objects such as large boulders. This is an area of extensive research, modelling and

optimisation (Singh et al. 2016; Abbaspour et al. 2018).

A method referred to as throw blasting is often used to assist in the process of moving the overburden

from the highwall to the lowwall. This method involves designing the explosion to throw the material

43

away from the highwall, resulting in reduced overburden removal requirements. This, however, can

result in rapid loading of any material currently on the floor of the pit (Singh et al. 2016). The

implications of rapid loading are discussed in Section 2.2 with respect to drained and undrained

behaviour and require careful considerations when the pit contains open water or in-pit mud.

Figure 2.3 Blasted overburden in a strip mine (adapted from Prytherch 2012)

2.1.2.2 Lowwall formation

An example of spoil being stored in-pit as a lowwall is provided in Figure 2.4. For large scale mining

efforts, in-pit dumping is most commonly conducted with a dragline positioned on a bench, or with

the use of trucks that transport the spoil to the desired location. Recent advancements in mining

processes now show sites using an integration of both methods (Westcott 2004; Mitra & Onargan

2012).

Each method for spoil transportation and placement has influences on the mining logistics,

construction speeds and the structural integrity of the produced spoil piles. Lowwalls constructed

using a dragline are dumped in a top-down manner, in contrast to a truck and shovel operation which

will be built from the bottom up. The stability implications for each method have been discussed in

Section 2.1.2.3 and 2.1.2.4 respectively.

HIGHWALL

LOWWALL BLASTED

OVERBURDEN

44

Figure 2.4 Typical strip coal mine highwall (left) and lowwall (right) in the Bowen Basin

2.1.2.3 Dragline operations

Draglines are large pieces of machinery used to move significant amounts of bulk solids. In the

context of a strip mine, it allows for the movement of overburden from the highwall to the lowwall

to expose the coal seam below.

The fundamental limitations for a dragline are the seam floor dipping angle, the reach, and the dig

depth. For large coal deposits with shallow slopes and suitable dig depths, draglines are the most

feasible option (Mitra & Onargan 2012). Table 2.1 details the key dimensional aspects of a dragline

relevant to strip mining and lowwall construction. The dig depth and dump height measurements are

shown in Figure 2.5.

Table 2.1 Key dragline components for strip mining, (adapted from Prytherch 2012; Simmons

2009; Williams 2015)

Machine dig depth Dependent on the length of the hoist and the drag ropes, and the length of the

boom. Medium-sized dragline dig depths are 48-54m but can range from 40-60m.

Machine dig radius A function of boom length (75-110 m) and boom angle (30-40⁰).

Machine dump height A function of boom length, boom angle and hoist limits, with typical dump heights

between 45 and 90 m.

Dump radius Conventional draglines can only dump at the operating radius of the machine

Machine working grade Maximum working grade is ±3%

HIGHWALL

LOWWALL

45

The draglines pick spoil up from in front of the machine, and place it behind, creating what is referred

to as the rill batter, which can be from 45 m to 90 m above the operating bench. The typical rill batter

has an approximate angle of repose of 37⁰, with results ranging from 35⁰ to 45o (Williams 2015)

dependent on the parameters of the spoil.

Draglines move down the highwall, excavating the overburden in strips. The strip width is typically

50 to 90 m and a few kilometres long (Westcott 2004). A cross-sectional view of the formation of a

lowwall using a dragline is represented in Figure 2.7, showing blasting of the overburden, movement

of the spoil towards the lowwall via the use of a bridge, and lastly, exposure of the coal seam. This

method can be used for single seam geology or multiple seams with adjustment to digging depths and

procedures. Pre-stripping is a technique that can also be used to account for limits in the draglines dig

depth, in which other machinery is used to create the first cut, reducing the relative dig depth

requirements. The spoil profile created by a dragline is limited in height by the size of the dragline

used. Dragline spoil profiles are usually limited to 80 m and 120 m total thickness (Simmons &

McManus 2004).

Progression of the lowwall requires surcharging of the working platform by other machinery to create

a stable foundation for the dragline. As described in Table 2.1 the machine working grade must also

be within ±3%. The stability will depend on the geometry of the mining pit, the working bench, the

spoil parameters within the pad, and the foundation conditions. Instabilities of these temporary

constructions can result in loss of production, infrastructure and life. An example of a failure

involving a dragline on a working bench with inadequate stability is shown in Figure 2.6. The stability

concerns around in-pit benches for draglines are discussed in Section 2.2 with respect to degraded,

in-pit mud at the base.

46

Figure 2.5 Dragline dimensional extents (adapted from Prytherch 2012)

Figure 2.6 Dragline working bench instability (adapted from ‘Norwich Park Dragline

Recovery’ 2008)

47

Figure 2.7 Dragline lowwall construction, (adapted from Duran 2013)

48

2.1.2.4 Truck dumping

With respect to mine design, truck dumping provides more flexibility and design control than a

dragline, allowing for accountability of varying overburden depths and thicknesses, and smaller

deposits (Westcott 2004), albeit at a higher price. The process consists of a shovel at the highwall

which loads the trucks with overburden spoil. The trucks then drive the spoil across to the lowwall,

placing it in ascending levels by driving it up to and along the running surface, and dumping it off the

edge, or by some variant of paddock dumping. A truck and shovel operation is shown in Figure 2.8.

Figure 2.8 A view of a truck and shovel operation in Hunter Valley Region in NSW (adapted

from Mitra & Onargan 2012)

Due to the constant movement of machinery creating “running surfaces”, the spoil undergoes greater

levels of compaction than with loosely placed dragline spoil dumps. This will have implications on

the formation of perched water tables, further discussed in Section 2.2.2. The truck and shovel method

has less separation of large and fine particles as they have less distance to fall in contrast to dragline

dumping (Williams 2015). Dump lifts range from 5 to 30 m, with the allowable height dictated by

safety and slope stability.

2.1.2.5 Dragline and truck and shovel integration

A study was conducted on a theoretical mine simulating a standard operation within Australian strip

mines to determine the differences in operating costs for a dragline and a truck and shovel operation.

The outcomes of this study show that it is largely dependent on the mine, the geometry, and the

required pace of progression (Westcott 2004), with results showing an integrated method is often the

most economic choice due to an increase in flexibility. The integrated use of equipment will impact

the construction of a lowwall, increasing the modelling complexity.

49

2.1.3 Effects of segregation

The degree and effects of segregation will differ depending on the type of dumping used and the

height of fall. Spoil particles tend to segregate by particle size during the placement via end dumping

from a dragline bucket or truck. This scenario is explained by Bagnold’s grain dispersive pressure,

and through particle kinematics (Middleton, 1970).

The initial momentum of the larger particles as they are dumped will cause them to ravel further down

the slope. For both the dragline and the truck and shovel methods of dumping, this results in a layer

of coarser spoil at the slope base, covered with layers that will alternate between fine and coarse

particle size distributions (Simmons & McManus 2004). This separation of coarse and fine particles

is illustrated in Figure 2.9 and Figure 2.10 for draglines and truck and shovel operations respectively.

For the truck and shovel method, Figure 2.10 also depicts the trafficked surface the trucks travel on

that can result in perched water tables forming within the dump. There can also be fines hang-up at

the crest due to shear strength characteristics and matric suction (Williams 2015). An image of a truck

end-dumping has been provided in Figure 2.11.

The spoils particle size distribution is related to it’s friction angle, and hence, it's shear strength, as

discussed in Section 2.2.6. This implies that with larger spoil ravelling to the bottom of the dumped

spoil, zones of higher shear strength can form. This effect will also be observed within the spoil pile

between the alternated layers of coarse and fine spoil. The variation in particle size will also be related

to the hydraulic conductivity of that section. This means that within the lowwall itself, there will be

alternating zones of high and low hydraulic conductivity which may or may not be interconnected.

This is discussed further in Section 2.2.2.

Figure 2.9 Schematic section of dragline spoil dump fabric and phreatic surface (adapted

from Simmons & McManus 2004)

50

Figure 2.10 Schematic section of truck spoil dump fabric and phreatic surface (adapted from

Simmons & McManus 2004)

Figure 2.11 Truck end dumping in the Bowen Basin

The segregation that occurs during dumping has been detailed by McLemore et al. (2009) who

highlighted five layers observed within spoil piles constructed via end dumping, detailed in Figure

2.12. The heterogeneity of the spoil pile will increase as it advances due to the variant nature of coal

measures and mining processes. A conceptual drawing of this increase is shown in Figure 2.13.

51

Figure 2.12 Conceptual model of the particle-size distribution of a rock pile (adapted from

McLemore et al. 2009)

Figure 2.13 Schematic structure of a spoil pile formed by haul trucks end-dumping from a

tip-head (adapted from Williams 2015)

52

2.2 Spoil Pile Stability and Design

This section will discuss the failure mechanisms of lowwalls and dumped spoil. The review will

include typical geometry, the physical parameters of spoil, water and the resultant pore water

pressures that develop within the spoil pile, and the conditions of the foundation, all of which are

considered critical factors of stability (Gómez et al. 2002).

2.2.1 Identification of spoil pile failure mechanisms

Large scale commercial open-cut strip mines in the Bowen Basin began during the early 1970’s, when

wet, muddy pit floors were the norm due to periods of heavy rainfall as depicted in Figure 2.14,

reproduced from the Bureau of Meteorology.

Figure 2.14 Rainfall anomalies for Queensland’s Bowen Basin (adapted from

http://www.bom.gov.au)

This period was marked by a number of lowwall slope failures (Gonano 1980; Seedsman et al. 1988).

Most of these failures occurred within approximately two weeks of construction of the spoil pile. The

wet, muddy pit floors were identified as the major cause of the failures, however, the dip of the pit

floor was also considered to have a major influence on spoil pile stability (Williams 2015). In 1974,

the Commonwealth Scientific and Industrial Research Organisation (CSIRO) and Utah Development

Company (UDC) put together a collaborative program to investigate these failures. Their aims as

summarised by Richards et al. (1981) were:

53

• to resolve the mechanisms operative in the stress-deformation behaviour of the highwalls

and spoil piles;

• to develop and define methods of predicting possible unstable areas; and

• to recommend possible control and remedial measures, recognising that total prevention of

all failures in the mine may not be the most economical or desirable solution.

Gonano (1980) published research on failures from Goonyella mine site recorded and analysed over

seven years. The research utilised physical mapping of the failures and identification of the shear

deformation locations. This data identified two common instability mechanisms. The first was

recognised as shallow, circular failures due to slope undercutting. The second mechanism identified

was a mode of large-scale failure related to a weak basal plane at the base of the lowwall that dipped

in the same direction as the coal seam.

Richards et al. (1981) published results of the laboratory and field testing used to investigate these

failures. Alongside the accurate surface surveying of Gonano (1980), subsurface instrumentation used

by Fuller & Cox (1987) was also used comprised of inclinometer strings, shear strips, and piezometers

installed during the failure to measure groundwater pressures. The geometric features of the failures

analysed are recorded in Figure 2.15.

Figure 2.15 Geometric features of five surveyed spoil failures (adapted from Richards et al.

1981)

This body of research resulted in the conceptualisation of the large slope failure mechanism observed

within the Bowen Basin region, specified by an active and a passive wedge, with movement laterally

along a weak basal layer. A limit equilibrium model was developed that utilises cohesive and

54

frictional material parameters acting along the defined failure planes of each wedge, displayed in

Figure 2.16.

Figure 2.16 Variables involved in two-wedge limit equilibrium analysis (adapted from

Richards et al. 1981)

This conceptual model was further developed by Nguyen et al. (1984), with the creation of a

numerical model that allowed for the calculation of a Factor of Safety for a scenario in which the

main body of the spoil has different parameters than the moisture-softened material at its base, noting

the influences of the berm width and the coal wedge length. Stability charts were developed based on

the model, and a number of lowwall toe conditions were discussed including a coal wedge, an

excavated and rehandled floor, buttress spoil, and a spoil buckwall.

The result of the findings at the Goonyella mine sparked an investigation into all of BHP’s open-cut

coal mines (Simmons & Yarkosky 2017). These studies involved back-analyses of failures and

correlations to laboratory testing of the associated spoil and floor materials. The test results with

respect to shear strength are discussed further in Section 2.3.

Simmons & McManus (2004) published updates to the understood failure mechanisms observed in

Australian coal strip mines, describing the most common forms of instability witnessed in spoil piles

as superficial mechanisms, deep-seated multi-wedge mechanisms, and multi-wedge rill mechanisms

acting on a weak base.

2.2.1.1 Superficial failure mechanisms

Superficial mechanisms have been described as failures that do not affect the overall stability of the

spoil pile. They are on the surface, acting as shallow arcs or undercut slumps. Figure 2.17 shows this

type of failure and where it occurs. The cause of propagation is most often due to batter undercutting

at angles steeper than the rill angle of repose, or due to perched water tables within the spoil.

55

Figure 2.17 Superficial failure mechanism (adapted from Simmons & McManus 2004)

A conference paper on the influence vegetation has on the stability of slopes (Schor & Gray 1995)

discusses the important role water plays in both shallow mass wasting and surficial erosion processes.

If water rises within the spoil pile, it can cause seepage out of the open face which will often result

in instabilities such as skin slump.

2.2.1.2 Multi-Wedge failure mechanisms

The results of in situ investigations and laboratory testing since the 1970s identified the multi-wedge

failure mechanism. The impacts of these failures can be severe. A large scale two-wedge dump failure

similar to those that are common in the Bowen Basin was investigated by Poulsen et al. (2014), with

before and after formations shown in Figure 2.18. An extensive study using numerical methods

including limiting equilibrium and finite difference models were used for the analysis. The study

indicated that the failure occurred through the mobilisation and resultant residual strength parameters

of the foundational material.

56

Figure 2.18 Two-wedge dump failure due to weak floor material (adapted from Poulsen et al.

2014)

Large scale failures are extremely dangerous if not monitored and understood and accounted for. The

example shown by Poulsen et al. (2014) that resulted in 14 deaths highlights this, further reinforcing

the importance of a reliable and accurate spoil characterisation and handling.

Simmons & McManus (2004) discuss another form of multi-wedge failure, depicted in Figure 2.19.

This type of failure is described as intermediate in scale and takes place in high rilled slopes. They

involve three components listed as toe bulging, mass sliding along a rill surface, and an upper steep

sliding surface or vertical tension crack. This form of failure is again a result of a weak basal layer.

Figure 2.19 Intermediate scale multi-wedge failure (adapted from Simmons & McManus

2004)

2.2.2 Influence of water on spoil pile stability

Water tables within a spoil pile can reduce the Factor of Safety by up to a factor of two (Williams

2015). Water influences the stability of spoil piles, both physically and chemically. Chemically,

exposure of clay-mineral rich spoil to water can result in degradation, causing breakdown of the spoil,

57

resulting in a loss of shear strength. This degradation occurs over time depending on the mineralogy

and geological history of the spoil. This occurrence is further discussed in depth in Section 2.2.8.

Physically, water flow into and out of the spoil pile, as well as the storage of it within the spoil pile,

will influence its stability.

2.2.2.1 Water inflow and outflow

With initial placement, the individual spoil particles may be saturated, but the spoil mass as a whole

is considered dry as there is no free water within the void spaces. Spoil piles increase in water content

over time due to the inflow and entrapment of water (Williams & Rohde 2008). Water inflow into a

lowwall can come from multiple sources. These include the infiltration of incident rainfall and rainfall

runoff from the catchment above the spoil pile, ingress from the adjacent pit, and groundwater

inflows. Water can also be introduced into the pit intentionally, such as after a flooding event, in

which a sacrificial pit will be used to store the water while another is being mined. An image of a

sacrificial flooded pit within the Bowen Basin has been provided in Figure 2.20.

Figure 2.20 A sacrificial flooded pit within the Bowen Basin

A conceptual hydrogeological model illustrated in Figure 2.21 shows a simplified overview of water

inflow and outflow of a spoil pile. Water entering the system includes rainfall, surface water runoff

from surrounding areas, alluvial flow, and natural water contained within the spoil itself. Water

leaving the system includes evaporation and drainage out of the toe, from a perched water table out

of the face of the spoil pile, or through seepage into the ground below.

Beale (2017) emphasises that a model must consider the geology of the material, the changes the

materials will experience over time, their natural moisture content, the external inflow of water,

redistribution of water within the stockpile, and water leaving the stockpile. Even with a simplified

model, the scenario is complex and difficult to model.

58

Figure 2.21 Conceptual hydrogeological model of a stockpile (adapted from Beale 2017)

2.2.2.2 Unsaturated spoil

As water enters the spoil pile, it will begin to flow downward through unsaturated, or partially

saturated zones. As the spoil wets-up, the majority of the water will initially go into storage within

the spoil. The spoil will continue to wet-up, remaining unsaturated, until its hydraulic conductivity

and the connectivity of water in the pores are sufficient to allow flow and drain down. Coarse-grained,

durable spoil will only need approximately 25% of its void space to saturate before breakthrough

occurs, while well-graded, weathered spoil, will need to wet-up to about 60% saturation to allow

drain down. While drain-down occurs, full saturation will not be reached, implying seepage can be

observed in unsaturated conditions (Williams 2006).

2.2.2.3 Saturated spoil and perched water tables

Perched water tables occur when full saturation is reached within the spoil pile, defined by a phreatic

surface. This phreatic surface will produce a static head and potentially develop pore pressures under

loading. Saturated zones will develop in areas of low permeability and trafficked surfaces, as well as

in contours produced by the topography of the site. A diagram showing multiple perched water tables

for a terraced construction is shown in Figure 2.22, proposed by Smith et al. (1995).

59

Figure 2.22 Hydrostratigraphy of segregated dumps – terraced construction (adapted from

Smith et al. 1995)

Saturated zones can occur anywhere within the spoil pile. Saturated spoil and resultant water tables

are often observed at the base of a spoil pile, however, to what extent this water table rises within the

spoil pile is a complicated matter that is not fully understood, dependent on the water balance of the

site and the spoil parameters (Okagbue 1986; Simmons & McManus 2004; McLemore et al. 2009;

Kho et al. 2013).

2.2.2.4 A three-zoned model for moisture within a spoil dump

Simmon & Fityus (2016) conducted a study on the moisture and water flow within a spoil pile, with

their results producing a simplified three-zone model that can be used for developing an

understanding of the dynamics of water within a spoil pile, displayed in Figure 2.23.

The model describes an upper, middle and lower zone. The upper zone is defined as material exposed

to rainfall and evapotranspiration. The water balance and resultant moisture content of this material

will be dictated by the environmental conditions of the mine site. The lower zone makes contact with

the foundation, and is expected to be an area of seepage, with a higher moisture content and a potential

water table. The middle zone is defined by areas of high and low permeability depending on the

construction method used that can contain perched water tables. For modelling, Simmon & Fityus

(2016) recommend unsaturated parameters for the middle zone, and saturated parameters for the

lower zone, as defined by the strength mobilisation modes recommended by Simmons & McManus

(2004), further discussed in Section 2.4.1.2.

60

Figure 2.23 Three-zone model for moisture conditions within spoil dumps (adapted from

Simmon & Fityus 2016)

2.2.2.5 Influence of water on stability

Direct observations from some Bowen Basin coal mines and a mine in Indonesia of relative conditions

show that a zone of saturated spoil builds up to form a phreatic surface to a thickness not exceeding

5 m above the base of the pile, and tapering to the downslope toe. This has become a general rule of

thumb for lowwall design (Simmons & McManus 2004), but analysis should still be conducted on a

case by case basis.

During the construction of a spoil pile, increased loading will have a direct effect on the spoil itself,

the foundations of the pile, and the basal material. Increased pressure on the spoil will cause a

reduction in void space, resulting in decreased hydraulic conductivity. Any zones of saturation have

the potential to develop pore pressures, which will reduce the effective normal load, leading to lower

shear strengths within the lowwall (McLemore et al. 2009). Increased loading will have the same

effects of reduced hydraulic conductivity on the foundation and any basal materials. As weak basal

materials are identified as a key cause of instability, it is important to understand loading rates and

pore pressure development, and to account for them within the lowwall design (Beale 2017).

Due to the complexity associated with modelling water flow within and around a lowwall, Simmons

(2009) states “confidence in the reliability of finite element groundwater modelling can only be

justified from back analysis and from calibration to monitoring data.” Due to the fast progression of

61

spoil piles, the difficulty of installation of test equipment and associated costs, moisture contents and

resultant pore pressures within spoil piles are not commonly measured. Further testing in this area

could improve our understanding of the hydrogeological model, and the stability of spoil piles in

general.

2.2.3 Stability related to in-pit mud and weak basal material

The floor material that will be the future lowwall foundation underlies the coal seam. As the mine

progresses, this foundation will be covered with spoil. Inadequate foundation preparation is the

primary cause of dump failures, with economic reasons being the deciding factor for their treatment

(Simmons & McManus 2004). It is therefore critical that a technical understanding of the foundation

parameters is known and accounted for in a design. To handle poor quality foundation material,

techniques such as blasting, the formation of spoil-backfilled keys, cross-ripping and terracing can be

used (Simmons & Yarkosky 2017).

Large scale lowwall failures in the Bowen Basin have most commonly occurred due to weakened

basal material (Seedsman et al. 1988). In-pit mud is, therefore, a serious consideration with respect

to design and management, as it will be located at the base of the placed spoil. If the mud acts as a

slipping plane, instabilities can occur in the lowwall, and within temporary structural features such

as the bridge utilised by a dragline to move the spoil from the highwall to the lowwall.

An example of the difficulty in handleability of in situ mud is depicted in Figure 2.24. If the highwall

is blasted onto the in-pit mud, a procedural issue arises in which material must be re-handled to access

the mud. This is associated with increased costs, increased fines production, and management of

potential slope instabilities during the process.

62

Figure 2.24 Spoiling into in-pit mud, adapted from (Prytherch 2012)

63

When mud is encountered in situ, Simmons & Yarkosky (2017) discuss a handling procedure known

as a key-bridge method to remove the weak zone (also referred to as a mud-pass), as shown in Figure

2.25. Modelling of this is complicated in two dimensions, and the complexity in three-dimensional

analysis arises with the number of unknown parameters and assumptions required to produce the

model. The potential slip paths are highlighted in Figure 2.26.

Figure 2.25 Mud cleanout via the key bridge method (adapted from Prytherch 2012)

Figure 2.26 Potential geotechnical hazards for dragline in-pit bench operations (adapted from

Simmons 2009)

Simmons (2009) explains that hazards arise if low strength surfaces have not been disrupted during

the blasting process, including shears, clay bands within the overburden, the coal seam, and the

immediate floor. Of the risks that exist, there are three methods of handling them:

64

• Remove the weak material;

• Analysis to determine acceptable likelihoods of instability; and

• Positioning the dragline so that the failure paths of the instabilities do not pass through

where the dragline is positioned.

A mud pass can involve the creation of a mud-dam within the dump profile. Simmons & Yarkosky

(2017) comment that this process requires significant skill from the operators and can produce further

hazards with respect to reshaping of the lowwall dump profile that must be recognised and addressed.

An image of a mud-dam has been provided in Figure 2.27, where mud has been isolated from the

active highwall, allowing mining to commence.

Figure 2.27 In-pit mud-dam storage observed in the Bowen Basin

Instabilities associated with weakened basal material and in-pit mud have been displayed in Figure

2.28 to provide context for scale and occurrence.

65

66

Figure 2.28 Geotechnical instabilities related to weak basal material (adapted from Prytherch

2012)

67

2.2.4 Drained or undrained failure

For coarse-grained spoil, large voids indicate drained conditions will exist, allowing water to flow

freely through the voids, with the stresses being held by the spoil framework. This contrasts with

undrained conditions, where the water in the voids cannot flow out fast enough, resulting in an

increase in pore pressures with no change in volume due to the incompressible nature of water. For

complex scenarios such as spoil piles, the internal spoil may be drained or undrained and can be

within these idealised conditions, acting in a partially drained manner (Williams 2015).

While typically drained conditions are observed, Duran (2013) investigated the behaviour of two

spoil piles in the Bowen Basin that appeared to exhibit undrained behaviour. Seven possible scenarios

were analysed involving weak basal material at the bottom of a spoil pile. One such scenario is

depicted in Figure 2.29, which shows the progression of a strip mine that was inundated with water.

The water within the pit was removed, and the next strip loading was placed against the previous

spoil pile. Upon completion of the strip, a large double wedge failure occurred, with shearing along

the previously saturated basal zone.

The failures were examined using the Morgenstern-Price method of slides. The models were

generated and calibrated with measurements recorded in situ to determine the best parameters to

represent the failure that occurred. It was found that for two scenarios, excellent matches existed

between the critical failure paths observed and undrained strength parameters. Duran (2013)

concludes that for undrained behaviour to be considered, the following aspects are required:

• Materials highly prone to slaking; and

• Previous spoil that has undergone flooding, with undrained behaviour applicable to areas

where additional spoil loading occurs in the current strip.

For investigations into in-pit mud, it is therefore critical to determine whether the material is likely

to exhibit drained or undrained conditions, as this will influence how the material should be tested in

the laboratory. Ideal conditions for maximum shear strength will be under drained conditions. It is,

therefore, most likely that in-pit muds with coarse-grained particle size distributions will have the

highest likelihood of ensuring stability.

68

Figure 2.29 Concept of mechanism by which undrained behaviour is invoked at base of spoil

pile (adapted from Duran 2013)

69

2.2.5 Lowwall slope performance evaluation

It is important to evaluate slope performance in the context of strip mining. Depending on the purpose

of the constructed slope and the duration it is expected to remain standing, variations in the required

Factor of Safety (FOS) can be justified. Figure 2.30 shows the annual probability of failure in contrast

to the failure consequence for various engineering structures, with mine pit slopes considered a low

cost, high probability event. While this gives context to potential acceptable failure rates, from the

mining perspective, the acceptable rate is a function of the consequences of a failure occurring,

including but not limited to the loss of life, damaged infrastructure and the loss of production rates,

as well as large scale considerations such as a social license to mine (Williams 2015). It is, therefore,

a site-wide or companywide consideration that must be made about what is acceptable.

Figure 2.30 Annual probability of failure versus consequence for various engineering

structures (adapted from Whitman 1984)

A range of typically accepted Factors of Safety have been detailed in Table 2.2. The FOS has been

related to the Probability of Failure (POF), including indications on what movement movements

could be expected and appropriate responses. Depending on the risks and consequences considered,

a range of FOS’s is justifiable.

70

The FOS is calculated as per Equation (1), in simplified terms:

𝐹𝑎𝑐𝑡𝑜𝑟 𝑜𝑓 𝑆𝑎𝑓𝑒𝑡𝑦 (𝐹𝑂𝑆) =𝑆ℎ𝑒𝑎𝑟 𝑆𝑡𝑟𝑒𝑛𝑔𝑡ℎ

𝑆ℎ𝑒𝑎𝑟 𝑆𝑡𝑟𝑒𝑠𝑠 (1)

For general slope design, a FOS of 1.2 is considered a good general target (Simmons 2009). There

may be cases where values as low as 1.05 are acceptable (Williams 2015).

Table 2.2 Suggested Factor of Safety relationships for open-pit coal mining (adapted from

Simmons 2009)

FOS POF INDICATIONS OF MOVEMENT ACTION/RESPONSE

1.5 to 1.2 < 0.001 Minor, stationary cracking acceptable

1.2 to 1.1 0.001 to

0.02

Surface cracks opening, some

ravelling, observable bulging, minor

scarp formation

Check sensitivity to assumptions,

review risk assessment

1.1 to 1.0 0.2 to 0.8 Significant observable movement,

loosening and isolated rolling of

rocks, scarps, mechanism forming

Seek stabilisation options, introduce

access restrictions, upgrade monitoring

<1.0 >0.8 Continuing movement, mechanism

well defined

Manage consequences, implement

stabilisation options

Table 2.2 provides both the FOS and the POS. The use of POF provides a more realistic depiction of

stability, however, generation of probabilities requires variations in expected material parameters to

be known. This is difficult due to the anisotropic nature of coal measures that results in the complex

heterogeneity of lowwalls, consisting of material prone to degradation. Standard mining practice

utilises FOS’s within design, and until data collection improves, it will likely remain the accepted

approach (Simmons 2009).

2.2.6 Angle of repose and its influences

The angle of repose is defined as the steepest angle achievable of a dumped material that does not

result in slumping or collapse (Beakawi Al-Hashemi & Baghabra Al-Amoudi 2018). In the case of

lowwall design, the material is placed by end dumping with trucks or with a dragline. Both methods

rely on gravity to carry material down the slope. Typical coarse-grained material handled in this way

will settle at its angle of repose (Williams 2015). The following factors have been noted to influence

the angle of repose of granular materials (Rowe 1962):

• particle size, shape and surface roughness (increasing with increases in these parameters);

• the specific gravity of the particles (increasing with increasing specific gravity);

• height of fall (decreasing with increasing height of fall);

71

• amount of water present (increasing with the addition of water to a maximum, and

decreasing with further saturation);

• the curvature of the slope (concave curvature resulting in slopes about 3o steeper than

convex curvature, with planar slope angles in between);

• base conditions; and

• whether the slope is natural or artificial.

Testing of Simons & Albertson (1960) investigated the angle of repose in relation to the particle size

distribution, with results shown in Figure 2.31. As the irregularity and size of the particles increase,

so does the angle of repose. Typical mine spoil is classified between sub-angular or subrounded.

The degradation of particles over time can also influence the angle of repose. Degradation results in

an increase in fines as the material breaks down. An increase in fines will be accompanied by an

increase in the density of the material. As the slope flattens, it will also become more stable (Hustrulid

et al. 2001).

Carson & Petley (1970) undertook a study on natural slopes and their angle of repose. Figure 2.32

shows a schematic of the frequency of slopes of different angles. The three thresholds shown are the

angle of repose (dry), half-angle of repose (saturated) and slope wash. This highlights how much a

material’s angle of repose can change due to the influence of degradation and saturation, subsequent

erosion and instabilities over time.

Figure 2.31 Angle of repose of granular materials (adapted from Simons & Albertson 1960)

72

Figure 2.32 Schematic frequency distribution for natural hillslope angles in the United

Kingdom (adapted from Carson & Petley 1970)

2.2.7 Shear strength and its influences

The shear strength of a soil is described as the shear stress that can be resisted prior to deformation.

There are many models for the calculation of shear strength in soil mechanics, of which the most

commonly adopted for lowwall design with saturated conditions is determined as per Equation (2)

(Terzaghi 1936).

= 𝑐′ + (𝑛

− uw) 𝑡𝑎𝑛 ′ (2)

where:

= shear strength

c’ = apparent cohesion

n = normal stress on the failure plane

uw = pore-water pressure

(n – uw) = effective normal stress on failure plane

’ = effective friction angle

When plotted on a graph of shear strength in relation to normal stress, Mohr-Coulomb criterion can

be applied to determine the values of apparent cohesion and the effective friction angle, as defined

by the intercept of the y-axis and the strength envelope that intersects multiple Mohr circles

respectively. These relationships have been plotted in Figure 2.35.

The friction angle is a complex function of the (Hustrulid et al. 2001):

• particle size distribution (reducing with decreasing particle size);

73

• particle shape and surface roughness (increasing with increasing angularity and surface

roughness);

• strength and specific gravity of individual particles;

• state of compaction (increasing with increasing compaction);

• applied stress level (decreasing with increasing stress, resulting in a curved strength

envelope passing through the origin);

• drained or undrained failure conditions; and

• degree of saturation.

The cohesion of a material is determined by the strength of the material under zero normal stress.

This strength results from the interlocking of dense particles, the electrostatic bonds between clay and

silt-sized particles, and matric suction defining the capillary forces between particles (McLemore et

al. 2009).

In order to determine values for apparent cohesion and effective friction angles, conventional devices

used include the simple direct shear, the standard direct shear, and triaxial testing (Babalola 2016).

For the application of laboratory data to in situ conditions, the apparatus must reflect the conditions

expected.

For the shearing of basal material in the multi-wedge failure mechanism, the standard direct shear

produces a representative shearing plane. A cross-section of the device is provided in Figure 2.33. A

representative specimen of the material in situ is placed inside the shear box, with a normal force

applied to it from above. The upper and lower box halves move apart horizontally, resulting in

shearing along a defined surface. During the test, the horizontal displacement is measured, as well as

the applied shear force.

Figure 2.33 Standard direct shear apparatus

74

As the sample shears, the recorded shear strength will be a function of the testing conditions, and the

parameters of the material. Figure 2.34 shows the change in shear stress with increases in shear strain

(horizontal displacement).

For dense, normally consolidated material, the shear stress will increase to a peak strength, after

which the material will lose strength and return to its residual shear strength. For loosely placed

material, no peak strength may be observed, with a continual increase in shear stress trending towards

its residual strength. This change in stress is associated with an increase in volume for dense material

and a decrease for loose material. With increasing strain, the overall void ratio for both tends towards

the critical voids ratio, defined by the state in which a material does not contract or dilate when

subjected to shear (Jackson 2015).

Figure 2.34 Shear characteristics of dense and loose sands (adapted from Jackson 2015)

With multiple shear tests conducted at differing normal stresses, Coulomb envelopes can be plotted,

with examples provided for peak and residual conditions in Figure 2.35. When analysing a material,

it is therefore critical to determine what stresses are relevant to the conditions in situ.

Figure 2.35 Peak and residual shear strength and Coulomb envelopes (adapted from Jackson

2015)

Spoil piles have been tested all around the world, with results summarised by McLemore et al. (2009),

with results bearing showing friction angle values of 38⁰ to 45⁰, and ranges from 21⁰ to 55⁰. Values

75

for cohesion range from 0 to 239 kPa depending on the materials and the testing conditions. The

testing of material within the Bowen Basin has been further discussed in Section 2.3.

2.2.8 Degradation of spoil and flood materials

Coal measures consist of material which has been deposited and formed over geological time, with

influences of pressure, erosion and weathering. Weathering is used to describe the physical, chemical

and biological changes that occur throughout the coal measures, typically within the top 50-100 m

(Neuendorf et al. 2005). Influences of weathering include colour changes, oxidisation, particle size

changes, changes in cementation, and changes in mineralogy.

For the context of this research, degradation will be used to discuss the physical and chemical changes

of overburden once exposed to the atmosphere through mining-related activities such as blasting,

scraping and hauling. Common physical influences of degradation are described by Fookes et al.

(1971) and Birkeland (1999) as freeze/thaw, thermal expansion and contraction related to temperature

fluctuations, crushing, abrasion, varying pressure and biologically related causes including plant roots

and living organisms. Chemical degradation involves oxidation, acid-based reactions, cementation,

solubility controlled precipitation and soil digenesis (Fookes et al. 1971; Birkeland 1999; Lan et al.

2003).

With increasing physical and chemical degradation is a decreased resemblance of the parent material

as existent in the coal measures overburden. Degradation post-exposure can influence every aspect

of the spoil, including colour, bulk texture, mineralogy, mineral texture, particle texture, cementation

and water chemistry(McLemore et al. 2009). Figure 2.36 adapted from Lan et al. (2003) summarises

the processes involved in degradation (referred to as weathering for the context of their research).

The physical and chemical components of degradation are complimentary in that physical breakdown

exposes more surface area, allowing for more chemical reactions to take place.

These changes caused by degradation will impact the geotechnical parameters of the spoil, altering

the shear strength, compressibility, permeability and overall structural stability of lowwalls and spoil

piles in general. Weathering occurs over hundreds of years whereas degradation can occur within

hours. The anisotropic nature of coal measures is reflected in the structural integrity of the spoil

produced during mining, resulting in significant variations of behaviour upon exposure.

For the formation of in-pit mud and quantification of its impacts on stability, key considerations are

the spoil, and floor materials slake durability, cementation ability and dispersion parameters

(Seedsman & Emerson 1985). Coal measures and the resultant spoil are typically weakly-cemented

and sensitive to water. Furthermore, the pore water within the spoil may have high levels of salinity.

This can exacerbate the effects of fresh rainwater on the degradation of spoil due to osmotic pressures.

76

A simplified schematic for the way a spoil aggregate can degrade is provided in Figure 2.37, showing

that it can slake, disperse, or do both. Examples of the visual impact of slaking and dispersion are

also provided.

Figure 2.36 Factors affecting weathering (adapted from Lan et al. 2003)

77

Figure 2.37 Degradation of stable aggregate due to slaking and dispersion

2.2.8.1 Cementation

Cementation is caused by particles being bound by clay, carbonates or hydroxides, as well as

amorphous material (Neuendorf et al. 2005). The degree of cementation will be dependent on

environmental considerations, including water availability and its chemistry.

Cementation can be both supportive and detrimental to the stability of a spoil pile, related to the

strength and degree of cementation, and where it occurs. Sections of a spoil pile can be influenced by

cementation, creating zones of lower permeability. This can result in preferential flow and the

possibility of perched water tables (Graupner et al. 2007). Cementation has been observed within

spoil piles throughout the world and can be a contributing factor to the integrity of a spoil pile

(Nguyen & Welsh 1981; Chigira & Oyama 2000; Karem 2005; Stockwell et al. 2006; Marescotti et

al. 2008). An example of weak cementation observed at a mine within the Bowen Basin is depicted

in Figure 2.38.

78

Figure 2.38 Weak cementation of spoil within the Bowen Basin

2.2.8.2 Slake durability

Clay-mineral rich soil and rocks are highly sensitive to changes in their water content. Upon exposure,

their structural integrity can deteriorate rapidly, having significant impacts on spoil pile stability. The

deterioration is recognisable as fracturing, flaking and dissolution (Erguler & Ulusay 2009).

Slaking is most often observed in spoil that contains high quantities of swelling clays, with the

degradation related to how the spoil’s moisture content changes. The main mechanisms of slaking on

immersion in water are the pressures associated with the compression of entrapped air related to

matric suction, and osmotic forces caused by differentials in ionic concentrations, seen most

commonly in swelling clay minerals (Seedsman 1986). Once the strength of the bonding is exceeded,

cracks will begin to propagate, allowing the release of entrapped air from the voids. An example of

clay-mineral rich rock slaking observed in the Bowen Basin is provided in Figure 2.39, with a clear

distinction made between the exposed and unexposed sections, typical of susceptible surface layer

spoil.

The resistance of a material to slaking is referred to as the materials slake durability. The degree of

swelling and slaking observed will be dependent on the mineralogy, chemical composition, strength

of the interparticle bonds, pore water chemistry and moisture conditions (Seedsman 1986).

As a rock slakes, changes are observed in the microstructure and mineralogy. Slaking is related to

reductions in shear strength, decreased permeability and void spaces, and increased settlement. A

summary of the generalised changes caused by slaking compiled by (McLemore et al. 2009) is

provided in Table 2.3 after research results compiled by Koncagül & Santi (1999) and Cheema et al.

(2004).

79

Figure 2.39 Slaking of clay-rich mine spoil from the Bowen Basin

Multiple methods exist for testing the slake durability of a rock. Common methods for assessment

include the slake index test (Deo 1972), the slake durability test (Franklin & Chandra 1972), the jar

slake test (Wood & Deo 1975), and the free swell test (Olivier 1979). All methods involve wetting

pieces of rock and determining the degree to which they slake. Attempts have also been made to test

the slake durability mimicking in situ conditions (Selig et al. 1983). Standardised methodologies have

been created such as the rock swelling and slake durability tests (AS 4133.3.4 2005; ASTM D3744 /

D3744M-18 2018), and numerous advanced analysis methods have and are being developed to

address limitations and provide extensions (Dhakal et al. 2002; Erguler & Ulusay 2009; Cano &

Tomás 2016; Kikumoto et al. 2016).

80

Table 2.3 Properties related to strength and slake durability (adapted from McLemore et al.

2009)

FACTORS STRENGTH TEST SLAKE DURABILITY TEST

Microstructure

Angularity Increase Decrease

Grain size

Coarse

Fine

Decrease

Increase

Increase

Decrease

Degree of alignment Decrease Decrease

Packing Density (dense) Increase Increase

Sutured/straight grain to grain

contact Increase Increase

Porosity

High

Low

Decrease

Increase

Depends on permeability

Depends on permeability

Degree of bonding

Well

Weakly

Increase

Decrease

Increase

Decrease

Mineralogy

Grains Depend on type of minerals Depend on type of minerals

Cementing and bonding material

Quartz

Clay Minerals

Increase

Decrease

Increase

Decrease

Various

Permeability

High

Low

Decrease

Increase

Increase

Decrease

Diagenesis and Metamorphosis Increase Increase

Water content

High

Low

Decrease

Increase

Increase

Decrease

Soft soluble minerals Decrease Decrease

Microfractures Decrease Decrease

Inclusions Decrease No effect

During the lowwall failures that occurred in the Bowen Basin in the 1970s, testing involved

investigating the failure mechanisms and reductions in shear strength caused by in-pit flooding and

heavy rain periods. Seedsman (1986) researched the behaviour of Bowen Basin clay shales in water,

collected from Goonyella mine, South Blackwater and Moura. Variance in slaking was observed

81

between the different mine sites highlighting the sensitivity differences that can be observed within

the same geographical region and geological formation. Four stages of breakdown were recognised,

being (a) swelling with minor fracturing, (b) disintegration of the sample into a coarse-grained pile

with some clay dispersion, (c) disintegration into a fine-textured pile with the dispersion of clay, and

(d) complete breakdown with formation of a layer of dispersed material. A key observation was the

importance of multiple wetting and drying cycles before any deterioration in the clay shales was

detected. A note was also made that slaking occurs more rapidly on unconfined surfaces, resulting in

the potential for one side of a rock degrading, as shown previously in Figure 2.39. These two

observations have critical implications for the impacts of rainfall, flooding, and location of the spoil

within the lowwall.

Further investigations into the slaking behaviour of spoil at the base of spoil piles in the Bowen Basin

were undertaken to determine the influence they would have on stability by Mostofa (2015) and

Vosolo (2017). Mostofa (2015) examined a high-quality spoil sample (Category 3 with respect to

BMA spoil category framework detailed in Section 2.4.1) in saturated conditions under a loads of

300, 600 and 900 kPa to determine the degree of slaking that would take place over 2, 90 and 180

days. Results of the testing showed that degradation increased with time and pressure. A decrease in

shear strength was observed for the degraded spoil in contrast to the spoil as sampled. Vosolo (2017)

also examined high-quality spoil (Category 3); however, a focus was placed on the influence of

wetting and drying cycles while under loading. Results of the testing indicated increases in slaking

with increasing pressure and the number of wetting and drying cycles.

Simmon & Fityus (2016) also investigated the degradation of mine spoils collected from the Bowen

Basin. Their work highlighted the complex nature of degradation, showing that spoils with similar

textures do not always react the same way when exposed to changes in water and humidity, with the

latter deemed less significant. Results show that some spoil will break down into smaller rock

fragments, while others will completely slake into mining soils. The slaking and cementation were

related to clay content and degree of cementation, highlighting the importance of the spoil mineralogy

over the rock fabric for identification of material prone to slaking. Furthermore, the results show

evidence that spoil within the lowwall, away from the surface, is less likely to degrade severely. Spoil

on the surface exposed to multiple cycles of wetting and drying and thermal changes will degrade at

a much faster rate.

The literature surrounding the slake durability of coal measures within the Bowen Basin focuses on

the spoil on the surface and within the lowwalls, and the degree of slaking that is expected. These

results highlight the potential for a clay-mineral rich spoil to degrade rapidly upon exposure to water.

This is important in consideration of what material is placed at the base of the spoil pile, and when

82

considering the impacts of flooding within a pit, with recognised Category 3 materials showing a high

degree of slake durability. It is therefore likely that spoil and floor materials exist that will retain

enough strength to be spoiled upon, even after saturation. This also means there are clay-mineral rich

materials that will degrade rapidly and should be not placed at the base of any construction.

2.2.8.3 Dispersion

Clay minerals are particles composed of minute crystalline sheets made of hydrous aluminum

silicates. Different structures of these sheets are observed in different clay types, with the most

common structures associated with the Kaolin, Illite, Smectite and Chlorite groups (Eberl et al. 1984).

Due to the physical structure of these clays and the cations present, largely different behaviours can

be observed in the presence of water which are of critical importance to the susceptibility of mine

spoil to degradation.

Of interest is Smectite which consists of two tetrahedral sheets and one octahedral sheet between

them. Between the interface of these sheets, water has the potential to be adsorbed as a function of

the cations present. The surface area of a clay particle is negatively charged. This causes positively

charged ions to be electrostatically attracted to the clay particles. The four most common cations

found are calcium, magnesium, potassium and sodium (Robertson et al. 1999). The strongest bonds

between clay particles are formed due to calcium (Ca2+) and magnesium (Mg2+) ions. Potassium (K+)

and sodium (Na+) ions both form weak bonds, resulting in greater susceptibility to swelling and

dispersion (also referred to as deflocculation) of the clay particles.

Dispersive clays will cause reductions in permeability, restricting water flow, increased swelling

pressures and subsequent slaking. As a result, they will be associated with spoil that experiences large

reductions in shear strength when exposed to water. Dispersive clay-mineral rich spoil will also erode

more readily, with an example of such erosion in Figure 2.40. Due to the nature of dispersive clays,

prolonged and intense rainfall can have significant detrimental impacts on the geotechnical stability

of a clay-mineral rich spoil pile (Chowdhury & Nguyen 1987). Recognition of dispersive spoil is

therefore highly beneficial, allowing for careful consideration of the material that is placed at the base

of a lowwall.

83

Figure 2.40 Erosion of dispersive spoil pile observed within the Bowen Basin

In order to determine the dispersive qualities of a material, a framework for class determination was

developed by Emerson (1967) that has been widely adopted. This Emerson class framework

categorises a material based on if it slakes, if it displays any dispersion, and to what degree it

disperses, with a flowchart displayed in Figure 2.41.

Figure 2.41 Scheme for determining class numbers of aggregates (adapted from Emerson

1967)

84

Due to the failures observed in the Bowen Basin in the 1970’s, Seedsman & Emerson (1985) looked

into the roles of clay-mineral rich spoil and the impacts they have on stability. The site focused on

was Goonyella mine, with the spoil noted to consist of Permian coal measures composed of

interbedded sandstones, siltstone and claystones. Seedsman & Emerson (1985) note that for clays to

disperse, the exchangeable sites of the clay must be occupied by sodium ions, however, there is also

the potential for material to disperse if reworked at water contents higher than their water content for

dispersion.

The analysis undertaken by Seedsman & Emerson (1985) involved six materials from slightly

weathered sandstone to mudstones, composed of Montmorillonite (a subclass of Smectite) and Illite.

Samples were classified as per the Emerson class testing detailed in Figure 2.41 and underwent a

comprehensive geochemical analysis.

Testing involved comparisons of the material as an intact chip to a pulverised condition with particle

size distributions passing 2 mm and 0.5 mm. The first intact chip showed dispersion of 3% in the

<20-micron fraction. Reduction of the aggregate particle size resulted in a 17% increase in measured

dispersed fines. Further reworking with water increasing dispersion to 95%. It was also determined

that swelling and dispersion could occur from osmotic pressures between the clay particles depending

on the strength of the attractive electrostatic forces, with the change in forces produced by variations

in the sodium cations identified as strong enough to disintegrate weak rock completely. For a sample

manipulated to contain more calcium ions, the required remoulding water content was significantly

increased, indicating the presence of calcium ions greatly reduce the dispersive potential of the clay.

The graphed results related to these findings are provided in Figure 2.42.

Figure 2.42 The influence of remoulding water content on dispersion (left), and influence of

exchangeable sodium on dispersion (right) (adapted from Seedsman & Emerson 1985)

85

From the investigations into coal measures overburden degradation, three key considerations were

drawn that will influence the degree and rate:

• Increased reworking will increase degradation;

• Increased water present during reworking will increase degradation; and

• Increased exchangeable sodium cation dominance will increase degradation.

These results imply that the more a material is reworked, the higher the chances are that it will degrade

on exposure to water. It is therefore critical that clay-mineral rich spoil with high exchangeable

sodium dominance is kept away from water, and any reworking is minimised. Upon saturation,

substantial degrees of degradation can be expected. These findings are in line with recent research

conducted by Simmon & Fityus (2016), with their degradation testing showing that the degree of

degradation was more dependent on the mineralogy of the material than the characteristic fragment

size or rock fabric.

2.2.8.4 Leaching

Leaching describes the process in which salts are removed from a material via hydraulic gradients or

through diffusion. It can cause significant changes in both the physical and the chemical properties

of clays, related to changes in pore water chemistry, removal of cementing agents, and resultant base

exchanges in the clay minerals. These can all be related to compressibility and decreases the shear

strength.

In the context of strip mining, this is likely to occur due to water movements within the spoil pile,

rainfall on the spoil, and in-pit flooding. Leaching lowers the materials liquid limit, plasticity index,

consolidation pressure, increases its compressibility, decreases its undisturbed and remoulded shear

strength, and increases sensitivity which is defined as the ratio of undisturbed shear strength to

remoulded shear strength (Brand & Brenner 1981). In the presence of swelling, dispersive clays,

leaching can occur rapidly. This highlights the importance of identifying spoil containing sensitive

swelling clays so it can be managed accordingly.

86

2.3 Shear Strength Characterisation of Bowen Basin Spoi, In-pit mud and Floor Layers

This section investigates the shear strength testing results of spoil, in-pit mud and floor layers of

material within the Bowen Basin open-cut strip mines obtained from laboratory testing, and through

back-analyses of failures.

2.3.1 Shear strength of spoil

With the geotechnical instabilities that started to occur in the Bowen Basin during the 1970s,

investigations and measurements of the shear strength of the coal measures spoil were undertaken.

These have included triaxial testing (Boyd et al. 1978; Dunbavan & Welsh 1982) analysed in

Williams & Zou (1991), direct shear testing (Naderian & Williams 1996; Kho et al. 2013; Bradfield

et al. 2013), and numerous results from back-analysed failures (Gonano 1980; Richards et al. 1981).

The majority of characterisation has been conducted on lowwall spoil and floor materials, with the

shear strength parameters of the in-pit mud mostly determined through back-analysis.

Table 2.4 displays the results of a statistical analysis completed by Williams & Zou (1991), in which

data from Boyd et al. (1978) was analysed and categorised as samples from the mass of the spoil, or

from the base, both of which were scalped to -19 mm. Results indicated the materials were likely

between drained and undrained conditions, with the spoil mass recognised generally as frictional, and

the spoil base as cohesive.

Table 2.4 Results of statistical analysis of spoil density and triaxial shear strength data from

Bowen Basin spoil materials (adapted from Williams & Zou 1991)

SPOIL PARAMETER SIZE OF DATA SET MEAN VALUE STANDARD DEVIATION

Total unit weight (kN/m3) 61 18.2 2.2

Cohesion of mass spoil (kPa) 62 73.1 77.7

Friction angle mass spoil (⁰) 62 28.1 10.8

Cohesion of base spoil (kPa) 87 121.8 83.1

Friction angle of base spoil (⁰) 87 8.0 4.0

Seedsman et al. (1988) conducted direct shear testing of Bowen Basin spoil in both dry (in situ) and

saturated conditions. These results were analysed in terms of friction and cohesion in Figure 2.43,

showing the differentiation between cemented, poorly lithified and weathered spoil. The highest shear

strengths were associated with the cemented spoil, followed by the poorly lithified spoil, and lastly

the weathered rock. Large ranges of cohesion were observed for all three categories of material.

87

Figure 2.43 Effect of saturation on the shear strength of Bowen Basin spoil and a comparison

with other values determined either experimentally or by back-analysis of spoil pile instability

(adapted from Seedsman et al. 1988)

The data from Seedsman et al. (1988) was also adjusted for comparison with the Leps (1970) data for

low and high-quality rockfill by adjusting the cohesion to zero, with results showing cemented spoil

was equivalent to low density, poorly graded rockfill. This has been plotted in Figure 2.44 adapted

from Seedsman et al. (1988).

88

Figure 2.44 Compilation of the strength of rockfill as measured in large triaxial tests (adapted

from Leps 1970), compared with direct shear values for coal mine spoil (adapted from

Seedsman et al. 1988)

Richards et al. (1981) and Peter et al. (1996) of CSIRO undertook testing of spoil, characterising the

materials physically, chemically and geotechnically, including shear strength testing, moisture

retention characteristics, consolidation, direct shear testing and triaxial testing. Sampling took place

at 163 locations, with testing conducted on seven of the most clayey and gravelly materials created

by combining samples due to quantity requirements. The tested spoil showed friction angles above

38⁰ for all compacted specimens. Upon saturation, the gravelly samples showed no reduction in

friction angle, however, the clayey materials reduced to between 12⁰ and 24⁰. One observation was

that the shear strength envelopes developed in the analysis of the results showed non-linear behaviour

above 400 kPa; a result which has been observed multiple times by other research studies (Morris

89

1990; Simmons 1995), implying the potential for zero cohesion and a reducing friction angle with

increasing normal stress. While this will influence the Factor of Safety of superficial failures, it is less

of a concern for deep-seated failures (Williams 2015).

Naderian & Williams (1996) conducted direct shear testing in a 100 mm device on coal mine spoil

sampled from the Jeebropilly mine (Moreton Basin, South East Queensland). Their investigation looked

at claystone and sandstone tested at the in situ moisture content, and after saturation. Results of the testing

detailed in Table 2.5 show that saturation reduced the cohesion and the friction angle of both

materials. The most significant reduction was observed in the claystone, with a friction angle

reduction of 20⁰, from 38.6⁰ down to 18.6⁰, in comparison to the sandstone which had a reduction of

6.7⁰. This quantifies the susceptibility of clay-mineral rich material to degrade upon wetting, which

is of concern if placed at the base of a spoil pile or lowwall.

Table 2.5 Shear strength parameters derived from direct shear testing (adapted from

Naderian & Williams 1996)

TEST

CONDITIONS

CLAYSTONE SANDSTONE

Cohesion (kPa) Friction Angle (⁰) Cohesion (kPa) Friction Angle (⁰)

As sampled 32.0 32.0 36.5 40.5

Saturated 25.4 25.4 8.9 33.8

Simmons (1995) published and later updated in Simmons & McManus (2004) a shear strength

framework for the categorisation of Bowen Basin spoil produced from coal measures. These

categories were produced from the dataset of Boyd et al. (1978), with its applicability confirmed for

the Permian, Triassic and Jurassic coal basins of Eastern Australia (Simmons & McManus 2004).

Four categories were assigned to different qualities of mine spoil. Each category was assigned values

of unit weight, cohesion and friction angle for unsaturated, saturated and remoulded conditions,

detailed in Table 2.6. The categorisation of spoil using this framework is further discussed in Section

2.4.1. The analysis shows that for the proposed framework, with increasing spoil category, an increase

in both friction angle and cohesion is observed, in comparison to results of a decreasing cohesion

with increasing friction angle for Boyd et al. (1978) and Seedsman et al. (1988).

Kho et al. (2013) attempted to contrast the datasets from Boyd et al. (1978), Simmons & McManus

(2004) and Seedsman et al. (1988) in terms of shear strength as per Equation (2), with results plotted

in Figure 2.45. Dry spoil is highlighted brown, and basal or saturated materials is highlighted blue.

Results are comparable between the two sets of parameters of Boyd et al. (1978) and the proposed

framework of Simmons & McManus (2004), likely due to the category framework being partially

based off of the testing results of Boyd et al. (1978). The results obtained by Seedsman et al. (1988)

90

were typically higher, comparable to the Category 4 spoil parameters proposed by Simmons &

McManus (2004).

Table 2.6 Bowen Basin spoil shear strengths, (adapted from Simmons & McManus 2004)

SPOIL

CATEGORY

UNSATURATED SATURATED REMOULDED

Unit

Weight

(kN/m2)

Cohesion

(kPa)

Friction

Angle

(deg)

Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Cohesion =

0 kPa, Friction

Angle (deg)

1 18 1 20 1 25 2.5 20 1 0 18 3 18 1.5

2 18 1 30 15 28 3 20 1 15 7.5 23 2.5 18 1.5

3 18 1 50 15 30 2 20 1 20 10 25 2.5 18 1.5

4 18 1 50 15 35 2.5 20 1 0 30 1.5 28 2

Figure 2.45 Calculated shear strength of 100m deep spoil pile using average shear strength

parameters (adapted from Kho et al. 2013)

Further testing of mine spoil was included in the research conducted by Hiung (2016), looking at

samples obtained from four mine sites in Queensland and New South Whales, composed of fresh,

well cemented, weakly cemented and weathered clay-mineral rich rock. The direct shearing test

results have been summarised in Table 2.7 and Table 2.8, respectively. Hiung (2016) contrasted these

results with the ones obtained by Boyd et al. (1978), Seedsman et al. (1988) and Simmons &

McManus (2004), with results plotted in Figure 2.46 in terms of cohesion and friction angle. Results

of the comparison show similar ranges of friction angles for all data sets. Variations are observed for

0

100

200

300

400

500

600

700

800

AV

ER

AG

E S

HE

AR

ST

RE

NG

TH

FO

R 1

00

m

HIG

H S

PO

IL P

ILE

(k

Pa)

91

the cohesion with Hiung (2016) showing similar values to Simmons & McManus (2004), but lower

results than Boyd et al. (1978) and Seedsman et al. (1988).

Table 2.7 Strengths for -2.36 mm spoil specimens tested dry in a 60 mm direct shear box,

reproduced (adapted from Hiung 2016)

MINE SPOIL DESCRIPTION COHESION (kPa) FRICTION ANGLE (⁰)

Jeebropilly Clay 4.4 12.5

Weathered rock 28.6 27.2

Mt Owen Mudstone-siltstone 2.5 34.7

Mudstone 29.3 32.9

Mt Arthur 3-month old Sandstone 5.9 35.3

2-year old degraded spoil 7.0 32.5

Hunter Valley

Operations

1-day old Siltstone 23.6 33.0

Rehandled 20-year old spoil 31.0 30.4

Averages 16.5 29.8

Ranges 3 to 31 ±14 13 to 35 ±11

Table 2.8 Strengths for -2.36 mm spoil specimens tested wet in a 60 mm direct shear box,

reproduced (adapted from Hiung 2016)

MINE SPOIL DESCRIPTION COHESION (kPa) FRICTION ANGLE (⁰)

Jeebropilly Clay 0.0 24.1

Weathered rock 6.5 19.5

Mt Owen Mudstone-siltstone 6.0 30.2

Mudstone 16.1 33.1

Mt Arthur 3-month old Sandstone 11.9 34.1

2-year old degraded spoil 8.2 26.2

Hunter Valley

Operations

1-day old Siltstone 0.1 31.6

Rehandled 20-year old spoil 3.0 30.1

Averages 6.5 28.6

Ranges 0 to 16 ±8 20 to 34 ±11

92

Figure 2.46 Comparison of friction angle and cohesion for Boyd et al. (1978), Seedsman et al.

(1988), Simmons & McManus (2004) and Hiung (2016)

2.3.2 Shear strength of in-pit mud and basal layers

The shear strength of in-pit mud and basal layers has been discussed since the 1970s. Most of the

analysis and understanding has been developed through back-analysis of past failures due to the

difficulty of sampling the material and the associated costs. Back-analysis requires assumptions to be

made on what the conditions of water are within the spoil pile, the strength of the spoil above the

basal layer, and the conditions in which failure occurred e.g. drained or undrained, saturated or

unsaturated (Nguyen 1985; Seedsman et al. 1988; Ulusay et al. 1995; Duran 2013).

Richards et al. (1981) undertook physical sampling of both the spoil and floor in their analysis of a

failure at Goonyella mine in the Bowen Basin. Samples were collected through borehole drilling and

core barrelling through the spoil and into the floor. Their results provide insight into the shear strength

of the spoil using both triaxial and direct shear testing to determine peak and residual strengths. The

shear strength of the spoil increased with depth, excluding a very low strength area measured just

above the floor explained by the presence of groundwater observed in relevant boreholes. The floor

beneath this weak layer was found to have significantly higher strengths.

Results of the direct shearing of the spoil show peak shear strengths and sharp reductions to residual

shear strength, as shown in Figure 2.47. For the unsaturated, as sampled spoil, the residual shear

strength was observed as a large reduction in friction angle, but a negligible change in cohesion. For

93

the saturated sample, an increase in cohesion was observed along with a substantial reduction in

friction angle, from 32⁰ to 3⁰.

Richards et al. (1981) conclude that remoulding was observed during initial stages of construction,

and therefore consider the residual values appropriate for the basal plane in the stability calculations

of newly completed spoil piles.

Figure 2.47 Shear strength of bulk spoil (a) unsaturated – as sampled; (b) saturated (adapted

from Richards et al. 1981)

This spoil based focus of testing is reflected in the work of Seedsman et al. (1988), Simmons (2009)

Bradfield et al. (2013) and Hiung (2016). Modelling of these failures are typically conducted under

the assumption of a weak basal layer, and that layer is most often assigned residual parameters

(Gonano 1980; Seedsman et al. 1988; Duran 2013). A summary of residual and remoulded shear

strengths for drained materials from the Bowen Basin found in the literature have been detailed in

Table 2.9. Most results show or assume a cohesion of 0 kPa, with the range of results up to 125 kPa

depending on the moisture conditions. For the friction angle, results ranged between 3⁰ and 30⁰

depending on the type of material and the testing conditions.

94

Table 2.9 Shear strength parameters of drained residual or remoulded Bowen Basin material

MATERIAL DESCRIPTION METHOD

RESIDUAL /

REMOULDED SHEAR

STRENGTH (DRAINED) REFERENCE

Cohesion

(kPa)

Friction

Angle (⁰)

Light grey mudstone Direct shear 20 6

(Dunbavan & Welsh 1982) Light grey sandy mudstone Direct shear 0 13

Top side of clay band Direct shear 20 8

Weak basal layer in spoil dump Direct shear 0 14 (Dunbaven & Driver 1988)

Clayey spoil from weathered soft rock Triaxial 0-75 7-17 (Mallett et al 1983)

Blocky spoil resistant to slaking Triaxial 0-50 20-30

Medium order dispersive swelling

clay spoil dry Direct shear 125 15

(Richards et al. 1981). Medium order dispersive swelling clay

spoil wet Direct shear 15 15-30

Clayey basal floor Triaxial 50 3

Poorly lithified mudstone Direct shear 45 24 (Seedsman et al. 1988)

Sheared floor Direct shear 0 11 (Seedsman et al. 1995)

Fine grained clay-rich high plasticity

spoil Fit to data 0 18

(Simmons & McManus

2004)

Fine grained low plasticity spoil Fit to data 0 18

Spoil with large clasts and low

plasticity Fit to data 0 18

Large blocky spoil with minor fines

and slaking Fit to data 0 28

Sub-floor Back

analysed 0 10-18 (Duran 2013)

95

The possibility of undrained failures within the Bowen Basin has been discussed as results of back-

analysis of past failures through the identification of failures that are best described by undrained

parameters. (Seedsman et al. 1988; Duran 2013). The shear strength parameters of undrained failure

derived from the available literature within the Bowen Basin are detailed in Table 2.10.

Table 2.10 Shear strength parameters of undrained Bowen Basin basal material

SPOIL DESCRIPTION METHOD SHEAR STRENGTH

(UNDRAINED) Su/v REFERENCE

Grey mudstone Triaxial “low” due to lack of samples (Dunbavan & Welsh

1982)

Poorly lithified mudstone Direct shear 0.29-0.37, average 0.33

(Seedsman et al. 1988) Montmorillonitic weathered material Direct shear 0.23

Kaolinite material Direct shear 0.47 (rapid draining, potentially

drained)

Weak basal layer in spoil dump Back

analysis

0.08-0.35, typically between

0.18 and 0.22 (Duran 2013)

For a slope that has failed, it is expected that back-analysis will find remoulded conditions for a basal

layer as the best fit due to the failure mechanism. This does not, however, account for the slopes that

do not fail while having degraded material at their base. There is, therefore, scope for investigations

into the shear strength variance of in-pit mud and degraded spoil for the identification of material that

can potentially be spoiled onto without promoting geotechnical instability.

96

2.4 Classification models For Spoil and In-pit Mud

Numerous standards exist around the world for the classification of soil and rocks. These

classification models can be divided into two systems; those which classify based on particle size and

plasticity, and those that classify based on bulk behaviour (Dorador et al. 2017). An example of the

former is the Unified Soil Classification System (USCS) ASTM D2487-11 (2011), and the latter the

British Standards Institution Standard (Dumbleton 1981). These classifications can be used to predict

the behaviour of materials, with classification based off parameters measured in the field, or in a

laboratory.

From these classifications, efforts have been made globally to determine appropriate shear strength

parameters for common materials, including but not limited to work undertaken by Koloski et al.

(1989), Carter & Bentley (1991), Dorador et al. (2017) and Das & Sobhan (2017). A summarised

table of results produced by Dorador et al. (2017) is provided in Table 2.11.

Table 2.11 Typical shear strength parameter ranges for soil (adapted from Dorador et al.

2017)

MATERIAL

UNIFIED SOIL

CLASSIFICATION

SYSTEM (USCS) SYMBOL

COHESION

(kPa)

EFFECTIVE

FRICTION

ANGLE (⁰)

Gravels, gravel with sand, alluvial deposits

(high energy), well-graded sand, angular

grains

GW, GP, GM, SW 0 30-45

Outwash (galacial), volcanic soil (lahar) GW, GP, GM, GW, SP, SM 0-50 25-40

Alluvial (low energy), uniform sand, roud

grains SW, SM, SP, ML 0-25 15-30

Glaciolacustrine SP, SM, ML 0-140 15-35

Lacustrine soil (inorganic) SP, SM, ML 0-10 5-20

Silty sand SM 0 30-34

Till, silty clays, sand-silt mix SM, ML 0-200 34-45

Clayey sands, sand-clay mix, volcanic soil

(tephra) SC, SM, ML 0-50 20-35

Silt (non-plastic clayey silts ML 0-30 30-35

Sandy clay, silty clay, clays (low plasticity) CL, CL-ML 0-20 18-34

Clays (high plasticity), clayey silts CH, MH 0 19-28

Silt loam, clay loam, silty clay loam ML, OL, CL, MH, OH, CH 0-20 18-32

Lacustrine soil (organic) OL, PT 0-10 0-10

97

Using these classification systems, parameters can be derived relative to the material type. While it

is ideal to classify a material accurately, it is not always possible due to budget and time constraints.

This is typical of the strip mining process in which large amounts of coal measures overburden is

moved at a rapid rate. In-depth analysis of all the material types present would not be feasible or

realistic. It is, therefore, a requirement that a quick and straightforward method of classification is

used to allow operators and engineers to get an estimation of spoil parameters quickly.

One such model has been developed by Simmons & McManus (2004) and has gained widespread

adoption throughout the mines of the Bowen Basin and across Australia, with the categorisation

framework also being looked at for use in Indonesia.

2.4.1 BMA Coal State-of-the-Art framework for spoil categorisation

BMA’s spoil category framework was developed in 2004, based on two decades of extensive in-

house research and collaboration with CSIRO on dragline-scale dumps. Shear strength results were

based on laboratory tests with empirical adjustments using back analysis of several large spoil dump

failures (Bradfield et al. 2013).

Concerning the design of spoil piles, in practice, there are many unknowns. Because of this and the

variant nature of spoil, failures do occur. The use of a framework such as the one implemented within

BMA allows for a structured system that can be built upon, with potential for continual improvement.

The current BMA framework used for categorising and associating shear strength parameters to

unsaturated, saturated and remoulded spoil overburden is based off the particle size distribution, the

liquid limit, the structure, the age and the cohesiveness. Each of these parameters is given weighted

percentages which are used to determine the category of the material. The specified category can then

be used to assign the material with estimated values of unit weight, friction angle and cohesion.

2.4.1.1 Framework structure

The bulk spoil is described using the terms of “framework” and “matrix”. The framework is the larger

particles and is considered to act as the main pathway for stresses. The matrix is the finer material

within this framework, filling the void spaces. The consistency of the material is determined when in

a moist state and considered either cohesive or cohesionless. The liquid limit is determined through

site investigation standards or via field investigation methodology. The age is related to the degree of

weathering that has taken place. The greater the weathering, the weaker the material is considered to

be. Of the five parameters, the highest weighting is for the liquid limit at 29%, followed by the

consistency and structure at 22.6%.

The weighted system is shown in Figure 2.48. Visual identification aides used in determining the

material category have been shown in Figure 2.49. Photos of Category 3 spoil and associated mud

98

(considered remoulded as per Table 2.12) have been provided in Figure 2.50, obtained during a site

visit to a mine within the Bowen Basin in during 2016.

Figure 2.48 Spoil categories and attributes (adapted from Simmons & McManus 2004)

Figure 2.49 Spoil structure attribute to be used with Figure 2.48 (adapted from Simmons &

McManus 2004)

99

Figure 2.50 Degradation of dry Category 3 spoil (left) to mud (right)

The associated Mohr-Coulomb shear strength spoil parameters for each category have been given in

Table 2.12. The values for cohesion and the friction angle were calculated from a straight-line fit

based off data for normal stresses ranging from 60 to 90 m. These parameters were derived from

laboratory triaxial and direct shear tests and were verified through back-analysis using observed

mechanisms of failure and the Sarma method, calculated with GALENA, a slope stability modelling

program (Simmons & McManus 2004).

Table 2.12 BMA design parameters for Category 1 to 4 spoil in unsaturated, saturated and

remoulded states, (adapted from Simmons & McManus 2004)

SPOIL

CATEGORY


Unit

Weight

(kN/m2)

Cohesion

(kPa)

Friction

Angle

(deg)

Unit

Weight

(kN/m3)

Unit

Weight

(kN/m2)

Cohesion

(kPa)

Friction Angle

(deg)

1 18 1 20 1 25 2.5 20 1 0 18 3 18 1.5

2 18 1 30 15 28 3 20 1 15 7.5 23 2.5 18 1.5

3 18 1 50 15 30 2 20 1 20 10 25 2.5 18 1.5

4 18 1 50 15 35 2.5 20 1 0 30 1.5 28 2

2.4.1.2 Framework strength mobilisation modes

For each spoil category, the parameters given are based on the state of the strength mobilisation mode

of the spoil – either unsaturated, saturated or remoulded.

The unsaturated spoil condition is defined as the state the material is in after initial dumping. In this

state, the material is free draining with no water held within the void spaces of the spoil framework

(Simmons & McManus 2004). The near-saturated spoil state occurs when water is introduced into

the void spaces to a saturation level that does not drain freely. The presence of water can result in

degradation as observed during the shear strength testing related to slaking and softening of the spoil,

and as such, is assigned lower strength parameters than that of unsaturated spoil. The remoulded

conditions describe the material once it has undergone shearing or has been significantly disturbed.

This creates a preferential plane of weakness, resulting in large losses of shear strength and is

irreversible.

A conceptual model of these three remoulded states is provided in Figure 2.51. The unsaturated and

saturated modes are obtained from results of triaxial testing. The remoulded mode is obtained through

direct shear testing. Figure 2.52 shows visually the expected decrease in shear strength from the

highest (unsaturated) to the lowest (remoulded).

100

It is important to note that these curves would theoretically flatten out with increasing stress, as

investigated by Bradfield et al. (2013), Simmon & Fityus (2016) and Bradfield (2017). For the current

framework in use, a linear fit was used to obtain values for the friction angle and the cohesion. This

straight-line fit is graphed in Figure 2.52 with indications of its intended range of applicability.

Figure 2.51 Conceptual strength modes for spoil (adapted from Simmons & McManus 2004)

Figure 2.52 Conceptual strength modes for spoil modified to explain the linear shear strength

approximation adopted in framework (adapted from Bradfield et al. 2013)

101

Figure 2.53 highlights the difference between an estimated framework linear fit and an actual strength

envelope proposed by Simmons (2009). The linear envelope was fitted to provide reasonably accurate

values for a dump 30 to 120 m high. For a dump 0 to 30 m high, the actual strength envelope is

potentially overestimated by the linear framework fit. The lower excavated slope of a dragline spoil

dump is noted commonly in practice to sit at an angle steeper than the angle of repose. This effect is

short-term and is attributed to the matric suction effect of the fine-grained material.

Simmons & McManus (2004) suggest that, in addition to the effects of matric suction, the non-linear

shear strength envelope helps explain this problem. At heights greater than 120 m, the framework

quickly overestimates the shear strength and the friction angle. At a dump height of 400 m, the shear

strength parameters inferred by the framework would be higher than would be realistic, leading to

potential instability if they were in design.

Figure 2.53 Mohr Diagram showing framework linear fit with respect to actual strength

envelope (not to scale) (adapted from Bradfield et al. 2013)

Conclusions from Bradfield (2017) investigating the impacts of high pressures on shear strength

concluded that the framework is still applicable with the following limitations:

• Category 2 or Category 3 spoil containing clasts with UCS > 5 MPa;

• The fine fraction of the spoil has a liquid limit < 35%;

• The clasts have high slake durability, low swell potentials and low dispersion potentials; and

• Category 1 strengths are only considered for dump heights up to 120 m.

102

2.4.1.3 Application of framework in practice

Simmons & McManus (2004) describe a test that was undertaken, in which 105 bulk samples were

collected. Two mine site personnel were required to categories the material visually. The results

showed different answers for 45 of the samples. The worksheet was then used to reattempt

categorisation of 13 bulk samples that were difficult to test previously. Again, using two personnel

with different levels of experience, there was 100% agreement on the categories of the samples.

103

2.5 Literature Review Commentary and Conclusions

The current BMA shear strength framework has been applied and used for hundreds of back-analysis

situations (Simmons & McManus 2004). It is a popular tool throughout the Australian coal industry

partially due to the movement of personnel between sites and companies. In its current state, it

provides adequate starting points for the categorisation of spoil. Application and adaption to

investigate the formation and potential categorisation of spoil that degrades into in-pit mud is of

consideration, with the following limitations and potential extensions identified.

2.5.1 Spoil category expansion and refinement

The current framework allows for the choice of four categories. The most common categories of

material found on site are Category 2 and 3, with Category 1 and 4 being relatively rare. Furthermore,

there is often variability within a spoil pile. This variability along with the high occurrence of

Category 2 and 3 material can make it challenging to identify which is which, and how to model the

spoil pile as a whole (Simmons & McManus 2004). As the classification is typically conducted

visually, it is crucial that different operators can obtain the same result consistently.

To account for this issue, one solution is to expand the selection of categories. This would involve

refining the parameters used to identify the material so that classifications can be more accurately

made. The introduction of more categories would also provide greater accuracy when investigating

failures using back analysis. The reason that this was not originally done was due to the data used to

create the framework. Review of the triaxial test data could not identify an alternative sorting that

was considered more reliable (Simmons & McManus 2004). More data collection, in collaboration

with past results, will allow for greater refinement and potential expansion of the current framework

categories.

2.5.2 Verification by testing

The authors of the original framework Simmons & McManus (2004) found that additional testing

was required for some clay-mineral rich lithologies. They state that the test results were not too

different from what the framework had suggested, however, acknowledged that this will not always

be the case.

While the framework has proven reliable for many typical cases, more data is always beneficial for

the less common materials and lithologies. Through more sampling and testing, refinement will be

possible. This is important for Category 1 material in particular, as it is not as common as the other

categories of materials and is often the most troublesome.

104

2.5.3 Material degradation considerations

The current framework was not designed to account for spoil susceptible to significant slaking. The

material will typically appear blocky initially, however once dumped, weathering can quickly induce

slaking. Williams (2015); Hiung (2016), Simmon & Fityus (2016), Mostofa (2015), Vosolo (2017)

and Bradfield (2017) have all investigated spoil from the Bowen Basin and the influence of saturation

on the particle size and shear strength, with results yet to be used in updating the framework.

Simmons & McManus (2004) discuss an observation of this occurrence in the field. The material at

the bottom of the spoil pile that was originally blocky underwent slaking, dispersion and inter-particle

void collapse. It is therefore important to have a means of identifying material susceptible to slaking

and to adapt the framework to account for this degraded state as witnessed with in-pit mud formations.

2.5.4 Consideration of in-pit mud parameters

Since the development of the BMA Coal shear strength framework, little consideration has been

found in the literature for the application of the framework to in-pit mud, and the parameters of in-pit

mud. This is due to a lack of testing, standard operating procedures requiring the removal of in-pit

mud, and remoulded strength parameters being assumed and assigned to in-pit mud for stability

modelling purposes.

The remoulded parameters used in the BMA Coal framework were obtained through the studies of

back-analysis of past failures such as in Seedsman & Emerson (1985). There is, therefore, room for

an investigation into the shear strength parameters of in-pit mud, and the potential variations that can

occur. The assumption of in-pit mud being assigned weak strength parameters is likely a combination

of conservative design to avoid failures, and the back-analysed strength parameters being low as only

failures are typically investigated. For scenarios in which the mud cannot be removed, and spoil is

placed on top, cases where no failure have occurred should also be investigated and added to the

literature. There is little data on this as the practice of spoiling onto mud in situ is frowned upon as

per standard operational procedures (Prytherch 2012).

Investigations into the classification of the physical, chemical and geotechnical properties relating to

in-pit mud and its formation will be highly beneficial towards ensuring lowwall stability post-flooding

events, and after periods of heavy rainfall. This is particularly beneficial for pits in which mud

removal is not an option. Testing of the causes and rates of degradation of spoil is also of interest,

with previous work (Emerson 1967; Seedsman & Emerson 1985; Seedsman 1986) highlighting key

identifiers such as the available exchangeable sodium cations and their relation to slake durability.

Through the correlation of in-pit mud parameters and the geotechnical parameters, significant

improvements in the modelling and handling of the mud can be made.

105

3 MINE SITE OBSERVATIONS

3.1 Preliminary Site Visits

Preliminary site visits took place on Tuesday 5 May 2015, investigating Mine Site C and Mine Site

B. At Mine Site C, Ramp 23, 22, 8, and 6S were visited. At Mine Site B Mine, inspections included

Ramp 9S, 11S, 7S, 5S, 4S and N1. Selected photographs from the Mine Site C and Mine Site B visits

are shown in Figure 3.1 to Figure 3.7.

Figure 3.1 Mine Site C Ramp 23, spoil failure and draglining into mud producing slumping of

spoil and bow-waving of mud

Figure 3.1 depicts a mud cleanout in progress. A dragline is using selective placement of spoil to push

in-pit mud towards the mining lowwall. This process allows for the highwall to be safely accessed. It

does however cause a cell of mud to be created which must be managed correctly to ensure lowwall

106

stability in both the short and long term. Spoil can be placed upon the mud to promote consolidaton.

If this spoil is placed too rapidly, as observed in Figure 3.1, bow-waving can occur identified by

tension cracks forming in the spoil. This can result in a significant loss in strength and must be

carefully managed and accounted for in design.

Figure 3.2 Mine Site C Ramp 23, Category 3 (left) and Category 4 and 2 (right) spoil

Figure 3.2 highlights the variance in spoil that can be found in situ in close proximity. An example

of Category 3 spoil is shown on the left consisting of grey boulders and gravel. On the right, an

example of Category 4 spoil is shown next to some Category 2 spoil. A clear differentiation is shown

between particle size and colour. It was noted that Category 4 spoil was less commonly found than

Category 2 and 3 spoil.

Figure 3.3 Mine Site C Ramp 22, flooded pit (~40 m deep)

Figure 3.3 shows a sacrificial mining pit. This pit was flooded purposefully so that an adjacent pit

could be mined. This water is of interest as it will potentially result in the degradation of all the

material it saturates. The rate of rise in the water table can also have impacts on the stability of the

lowwall.

107

Figure 3.4 Mine Site C Ramp 6S, floor mud from Category 3 spoil

Figure 3.4 shows flooding that reaches the highwall and the associated mud formed from the floor

material. In order for the highwall to be mined, this water and degraded mud must be managed

appropriately as in the future it will become the base of the lowwall.

Figure 3.5 Mine Site B Ramp 11S, closed by flooding since late 2010

Figure 3.5 shows a mining ramp that has been closed due to flooding for multiple years. Closure of a

pit impacts the productivity and profitability of a mine and highlights the importance of management

of any in-pit water and mud.

108

Figure 3.6 Mine Site B Ramp 5S, flooded pit surrounded by largely Category 3 overburden

and spoil

Figure 3.6 shows a flooded pit and the quality of the spoil surrounding it. This pit was surrounded by

Category 3 spoil, often attributed with relatively high shear strength. It was observed that the material

appeared to be relatively resilient to current and past water tables in comparison to Category 1 or 2

spoil, maintaining both particle size and structural stability. This observance indicates material which

could have the potential for withstanding in-pit flooding.

Figure 3.7 Mine Site B Ramp 5S, in-pit mud at base of Category 3 overburden and spoil

lowwall

Figure 3.7 shows the material at the current water level. A mixture of spoil and in-pit mud was found.

The mud was observed to be composed of clay, silt and sand, with sand being the dominant fraction.

109

4 PROJECT PLAN AND RESEARCH METHODOLOGY

This section details the sampling locations for all collected materials, and the testing methodology

utilised for the characterisation of the samples.

4.1 Sampling Methodology

Due to the location of the material and accessibility, only manual handling was used for collection.

The containers were therefore limited to 20 L, with spoil scalped at -53 mm. Scalping for the collected

mud samples was not required. All samples were freighted to the UQ Geotechnical Laboratories in

the Geotechnical Engineering Centre. All 20 L containers were sealed to maintain moisture during

transit.

4.2 Sample Identification System AND SAMPLING SUMMARY

Each material was classified based on its collection location and its BMA spoil category. Samples

taken directly from the lowwall were classified as spoil. Samples collected below the water level or

at a location previously underwater were classified as in-pit “mud”, with reference made to the

assigned source spoil category. For each material collected details of the mine, ramp, source BMA

spoil category, sampling date, an ID No. to differentiate various materials collected from the same

location, and a unique ID Code indicating the source spoil category, material type and sample number,

are included in Table 4.1. For referencing simplicity, the ID Code is used hereafter to identify the

materials.

Table 4.1 Spoil and Mud Identification Details

MINE RAMP SOURCE

SPOIL

SPOIL

/MUD

SAMPLING

DATE ID NO. ID CODE

Mine Site A Ramp 10 North

CAT 3 SPOIL 26/04/2015 #1 C3S-01

CAT 1 SPOIL 26/04/2015 #2 C1S-02

CAT 3 SPOIL 26/04/2015 #3 C3S-03

CAT 3 SPOIL 26/04/2015 #4 C3S-04

CAT 2 MUD 26/04/2015 #5 C3M-05

CAT 2 MUD 26/04/2015 #6 C3S-06

CAT 2 MUD 26/04/2015 #7 C3M-07

Ramp 50S CAT 1 MUD 30/11/2016 #32 C1M-32

Mine Site B Ramp 5 South

CAT 3 MUD 2/06/2015 #8 C3M-08

CAT 3 SPOIL 2/06/2015 #10 C3S-10

CAT 3 MUD 2/06/2015 #12 C3M-12

CAT 3 SPOIL 2/06/2015 #13 C3S-13

110

CAT 2 SPOIL 2/06/2015 #16 C3S-16

CAT 3 SPOIL 30/11/2016 #30 C3S-30

Ramp 1 North CAT 1 SPOIL 2/06/2015 #17 C3S-17

Mine Site C

Mine

Ramp 6 South CAT 3 MUD 3/06/2015 #18 C3M-18

CAT 3 SPOIL 3/06/2015 #20 C3S-20

Ramp 22 CAT 1 MUD 3/06/2015 #23 C1M-23

Ramp 14 CAT 2 SPOIL 3/06/2015 #24 C2S-24

4.3 Material Sampling – Mine Site A (26 April 2015)

Leigh Bergin of BMA collected samples from Mine Site A Mine on 26 April 2015. Details and photos

of the collected materials are included in Sections 4.3.1 and 4.3.2.

4.3.1 Mine Site A sampling locations

The sampling locations of the materials collected at Mine Site A are shown in Figure 4.1.

Figure 4.1 Mine Site A sampling locations

111

4.3.2 Mine Site A samples

The samples collected at Mine Site A are listed in Table 4.2 and shown in Figure 4.2 to Figure 4.15.

Table 4.2 Mine Site A sampling details

SAMPLE ID SAMPLE DESCRIPTION NUMBER OF 20 L

BUCKETS (-53 mm)

WET MASS

(kg)

Ramp 10N

C3M-01 Fresh Category 3 wet mud in pit floor 2 60

C1M-02 Old Category 1 mud 2 60

C3M-03 Fresh Category 3 wet mud in floor,

granular with minor fines 2 60


granular 1 30


majority fines 1 30

C2M-06 Dried Category 2 mud 2 60

C2M-07 Dried Category 2 mud 2 60

Figure 4.2 Mine Site A Ramp 10N sampling location C3M-01

112

Figure 4.3 Mine Site A Ramp 10N sample C3M-01


113

Figure 4.5 Mine Site A Ramp 10N Sample C1M-02


114



115


Figure 4. 10 Mine Site A Ramp 10N sampling location C3M-05

116


Figure 4.12 Mine Site A Ramp 10N sampling Location C2M-06

117


Figure 4.14 Mine Site A Ramp 10N sampling Location C2M-07

118


119

4.4 Material Sampling – Mine Site B (2 June 2015)

The main location of the sampling was Ramp 5S, with minor sampling in Ramp 1N.

4.4.1 Mine Site B sampling locations

The sampling locations at Mine Site B are shown in Figure 4.16 to Figure 4.22.

Figure 4.16 Mine Site B Ramp 5S sampling

Figure 4.17 Mine Site B Ramp 5S sampling location C3M-08

Approximate high water level

120

Figure 4.18 Mine Site B Ramp 5S sampling location C3S-10


121




122

4.4.2 Mine Site B samples

The samples collected at Mine Site B are described in Table 4.3, and shown in Figure 4.23 to

Figure 4.33.

Table 4.3 Mine Site B sampling details

SAMPLE ID SAMPLE DESCRIPTION

NUMBER OF

20 L BUCKETS

(+53 mm)

WET MASS (kg)

-53 mm +53 mm

Ramp 5S

C3M-08 10-year old Category 3 mud just above

current water level (previously flooded) 6 180 -

C3S-10 10-year old Category 3 spoil above

C3M-08 mud 6 168 19 (10.2%)

C3M-12 10-year old Category 3 wet mud from

edge of flooded pit 3 90 -


historical water level 18 547.5 40 (6.8%)


historical water level 3 90 -

Ramp 1N

C1S-17 Fresh Category 1 (Tertiary) spoil 3 90 -

Figure 4.23 Mine Site B Ramp 5S Sample C3M-08 surface texture (with 20-cent coins for

scale)

123

Figure 4.24 Mine Site B Ramp 5S sampling location C3S-10 surface texture

Figure 4.25 Mine Site B Ramp 5S sample C3S-10 sieving to -53 mm

124

Figure 4.26 Mine Site B Ramp 5S sample C3M-12 surface texture

Figure 4.27 Mine Site B Ramp 5S sample C3S-13 fine-grained surface texture

125

Figure 4.28 Mine Site B Ramp 5S sample C3S-13 coarser-grained below surface

(a) (b)

Figure 4.29 Mine Site B Ramp 5S sample C3S-13: (a) -53 mm, and (b) +53 mm

126

Figure 4.30 Mine Site B Ramp 5S sample C3S-13: +53 mm

Figure 4.31 Mine Site B Ramp 5S sample C2S-16 agglomerated surface texture

127

Figure 4.32 Mine Site B Ramp 5S sample C1S-17

Figure 4.33 Mine Site B Ramp 1N sample C1S-17 surface texture

128

4.5 Material Sampling – Mine Site C (3 June 2015)

At Mine Site C, the main location of the sampling was Ramp 6S, with minor sampling in Ramps 22

and 14. Access to Ramp 23, where a spoil pile failure had previously occurred, had not been

established.

4.5.1 Mine Site C Sampling locations

The sampling locations at Mine Site C are shown in Figure 4.34 to Figure 4.37.

Figure 4.34 Mine Site C sampling location C3M-18

Figure 4.35 Mine Site C sampling location C3S-20

129



130

4.5.2 Mine Site C samples

The samples collected at Mine Site C are described in Table 4.4, and shown in Figure 4.38 to

Figure 4.44.

Table 4.4 Mine Site C samples


NUMBER OF

20 L BUCKETS

(+53 mm)

WET MASS (kg)

-53 mm +53 mm

Ramp 6S

C3M-18 Fresh (6-month old) Category 3 mud

from floor 6 180 -

C3S-20 Fresh (6-month old) Category 3 spoil

from lower bench slope 18 562.5

44

(7.3%)

Ramp 22:

C1M-23 6.5-year old Category 1 wet mud from

edge of flooded pit 3 90 -

Ramp 14:

C2S-24 Fresh Category 2 spoil 3 87 -

Figure 4.38 Mine Site C Ramp 6S sample C3M-18 surface crusting

131

Figure 4.39 Mine Site C Ramp 6S sample C3S-20 surface PSD (with 22.9 cm diameter plates

for scale)

(a) (b)

Figure 4.40 Mine Site C Ramp 5S sample C3S-20: (a) -53 mm, and (b) +53 mm

132

Figure 4.41 Mine Site C Ramp 5S sample C3S-20 weighing + & -53 mm fractions

Figure 4.42 Mine Site C Ramp 5S sample C3S-20 +53 mm

133

Figure 4.43 Mine Site C Ramp 22 sample C1M-23

Figure 4.44 Mine Site C Ramp 14 sample C2S-24

134

4.6 Material Sampling – Mine Site A and Mine Site B (30 November 2016)

A second sampling visit was undertaken with intentions of collecting a large amount of one Category

3 and one Category 1 mud for testing. The Category 3 mud was collected from Mine Site B’s Ramp

5S at the highwall. The Category 1 mud was collected from Mine Site A’s Ramp 50S at the highwall.

4.6.1 Mine Site A and Mine Site B sampling locations

The sampling locations of both mines are shown in Figure 4.45 and Figure 4.46.


Figure 4.46 Mine Site A Ramp 50S sampling location C3M-32

4.6.2 Mine Site A and Mine Site B samples

The samples collected at Mine Site C are described in Table 4.5, and shown in Figure 4.47 to

Figure 4.48.

135

Table 4.5 Mine Site B and Mine Site A samples (2016)


NUMBER OF

20 L BUCKETS

(+53 mm)

WET MASS (kg)

-53 mm +53 mm

Ramp 5S

C3M-30 10-year old Category 3 wet mud from

bottom of highwall

14 (4 over

19 mm) 590 -

Ramp 50S

C3M-32 6.5-year old Category 1 wet mud from

edge of flooded pit

9 (none over

53 mm) 243 -

Figure 4.47 Mine Site B Ramp 5S sample C3M-30

136

Figure 4.48 Mine Site A Ramp 50S sample C3M-32

137

4.7 Physical and Chemical Characterisation

4.7.1 As-sampled moisture content

The as-sampled gravimetric moisture content of all samples collected was determined by oven drying

according to AS 1289.2.1.1 (2005) as the mass of water/mass of solids, expressed as a %. Due to the

potential presence of carbonaceous material in the samples, they were dried in a 60oC to avoid

combustion.

4.7.2 Specific gravity

The specific gravity of all samples collected was determined from the average of 10 readings on dried

samples crushed to -1 mm. Testing was carried out in accordance with AS 1289.3.5.2 (2002) using a

vacuum pycnometer with helium, as shown in Figure 4.49.

Figure 4.49 Helium pycnometer

4.7.3 Total suction

The total suction was determined from the average of three readings on the -2.36 mm fraction of all

spoil samples collected. Testing was carried out in accordance with AS 1289.2.2.1 (1998) using a

WP4 Dewpoint Potential Meter, as shown in Figure 4.50.

4.7.4 Atterberg limits

The Atterberg limits describe the plasticity of a material. The liquid limit (LL; the gravimetric

moisture content at which the material starts to flows) of all samples collected was determined in

accordance with AS 1289.3.1.1 (2009), using the Casagrande method. The plastic limit (PL; the

138

gravimetric moisture content at the material starts to act as a solid) of all samples collected was

determined in accordance with AS 1289.3.2.1 (2009). The Casagrande LL and PL apparatus are as

shown in Figure 4.51. The plasticity index (IP) was calculated as the difference between the LL and

the PL, as per AS 1289.3.3.1 (2009).

Figure 4.50 WP4 dewpoint potential meter

Figure 4.51 Atterberg limit test apparatus

4.7.5 Emerson class number

To identify materials prone to slaking, the Emerson class number of all samples collected was

determined in accordance with AS 1289.3.8.1 (2017).

139

4.7.6 Chemical characterisation

The pH, electrical conductivity and total dissolved solids of all samples collected were determined

from the average of three readings in accordance with AS 1289.4.3.1 (1997) and AS 1289.4.4.1

(2017), using a hand-held pH-EC meter in paste samples.

Table 4.6 Summary of physical and chemical characterisation testing

CO

DE

CA

TE

GO

RY

SO

UR

CE

GR

AV

IME

TR

IC

MO

IST

UR

E

CO

NT

EN

T

TO

TA

L S

UC

TIO

N

AT

TE

RB

ER

G

LIM

ITS

SP

EC

IFIC

GR

AV

ITY

EL

EC

TR

ICA

L

CO

ND

UC

TIV

ITY

pH

EM

ER

SO

N C

LA

SS

C3S-01 CAT 3 MUD √ √ √ √ √

C1S-02 CAT 1 MUD √ √ √ √ √

C3S-03 CAT 3 MUD √ √ √ √ √

C3S-04 CAT 3 MUD √ √ √ √ √

C3M-05 CAT 2 MUD √ √ √ √ √

C3S-06 CAT 2 MUD √ √ √ √ √

C3M-07 CAT 2 MUD √ √ √ √ √

C1M-32 CAT 1 MUD √ √ √ √ √

C3M-08 CAT 3 MUD √ √ √ √ √

C3S-10 CAT 3 SPOIL √ √ √ √ √ √ √

C3M-12 CAT 3 MUD √ √ √ √ √

C3S-13 CAT 3 SPOIL √ √ √ √ √ √ √

C3S-16 CAT 2 SPOIL √ √ √ √ √ √ √

C3S-30 CAT 3 MUD √ √ √ √ √

C3S-17 CAT 1 SPOIL √ √ √ √ √ √ √

C3M-18 CAT 3 MUD √ √ √ √ √

C3S-20 CAT 3 SPOIL √ √ √ √ √ √ √

C1M-23 CAT 1 MUD √ √ √ √ √

C2S-24 CAT 2 SPOIL √ √ √ √ √ √ √

4.7.7 X-ray diffraction

For each of the spoil and mud samples collected, 400 grams were sent for XRD Analysis by Mark

Raven at CSIRO Land and Water Flagship Urrbrae SA. As per information provided by Mark Raven,

the samples were subjected to the methodology detailed below:

• A 1.5 g sub-sample was ground for 10 minutes in a McCrone micronizing mill under

ethanol. The resulting slurry was oven-dried at 60°C then thoroughly mixed in an agate

140

mortar and pestle before being lightly-pressed into a stainless-steel sample holder for X-ray

diffraction analysis.

• The XRD patterns of the as-received samples showed swelling clay minerals present and

hence were Ca saturated twice using 1M CaCl2 followed by water, and then ethanol wash

before oven drying at 60°C (samples were centrifuged at 6000 rpm after each step). The

oven-dried samples were thoroughly mixed with an agate mortar and pestle before being

lightly back pressed into stainless steel sample holders to achieve random orientation of the

mineral particles for XRD analysis.

• XRD patterns were recorded with a PANalytical X'Pert Pro Multi-purpose Diffractometer

using Fe filtered Co Ka radiation, auto divergence slit, 2° anti-scatter slit and fast

X'Celerator Si strip detector. The diffraction patterns were recorded in steps of 0.016° 2

theta with a 0.4 s counting time per step and logged to data files for analysis.

• Quantitative analysis was performed on the XRD data using the commercial package

SIROQUANT from Sietronics Pty Ltd. The results are presented in two forms, normalised

to 100%, and hence do not include estimates of unidentified or amorphous materials and

after analysis using an internal standard to determine amorphous and unidentifiable content.

4.7.8 Cation exchange capacity and exchangeable cations

The samples sent to CSIRO Land and Water Flagship at Urrbrae SA for XRD analysis were also

subjected to chemical analysis at the same facility. The chemical analysis performed identified

exchangeable cation content, Cation exchange capacity (CEC) and soluble salt content.

Ammonium (NH4+) CEC and exchangeable cations were determined by CSIRO Land and Water

Analytical Services Unit using method 15D2 given in the “Australian Laboratory Handbook of Soil

and Water Chemical Method” (Rayment & Higginson 1992).

141

Table 4.7 Summary of mineralogical and geochemical characterisation testing

CODE CATEGORY SOURCE X-RAY

DIFFRACTION

CATION

EXCHANGE

CAPACITY

EXCHANGEABLE

CATIONS

C3S-01 CAT 3 MUD √ √ √

C1S-02 CAT 1 MUD √ √ √

C3S-03 CAT 3 MUD √ √ √

C3S-04 CAT 3 MUD √ √ √

C3M-05 CAT 2 MUD √ √ √

C3S-06 CAT 2 MUD √ √ √

C3M-07 CAT 2 MUD √ √ √

C1M-32 CAT 1 MUD √ √ √

C3M-08 CAT 3 MUD √ √ √

C3S-10 CAT 3 SPOIL √ √ √

C3M-12 CAT 3 MUD Similar to C3M-08



C3S-30 CAT 3 MUD √ √ √


C3M-18 CAT 3 MUD √ √ √


C1M-23 CAT 1 MUD √ √ √


142

4.8 Particle Size Distribution

The Australian Standards used for determining the particle size distribution (sieving and hydrometer)

the methodology were AS 1289.3.6.1 (2009) and AS 1289.3.6.3 (2009). Australian Standards were

used to conduct a particle size distribution test on each of the representative samples obtained.

Depending on the sample, typical ranges of tested material were 1 to 3 kg per test. A wet (washing)

sieving procedure was used to allow for the breakdown of agglomerates not accounted for with dry

sieving, or for materials that underwent significant agglomeration upon drying. The test equipment

for wet sieving is shown in Figure 4.52. For the analysis of the fines, the hydrometer setup is shown

in Figure 4.53.

Unless stated otherwise, dispersant was not used as per the standards to simulate the true in situ

particle size distribution of the material without causing further degradation. During wet sieving, the

material passing the 75 µm sieve was collected and used for hydrometer testing. The fractions

retained on each sieve were dried and weighed.

Figure 4.52 Wet sieving apparatus (left), addition of suspension solution to stack (top right),

filtering of sieved sample for drying and weighing (bottom right)

143

Figure 4.53 Agitation of 1000cc solution (left) & hydrometer analysis of solution with the

control cylinder and temperature gauge (right)

Table 4.8 Summary of Particle Size Distribution Testing

CODE SOURCE

CATEGORY TYPE DRY SIEVING

WET SIEVING

(W/O

DISPERSANT)

HYDROMETER

(W/O

DISPERSANT)

C3S-01 CAT 3 MUD √ √




C3M-05 CAT 2 MUD √ √

C3S-06 CAT 2 MUD √ √ √

C3M-07 CAT 2 MUD √ √ √













144

4.9 Degradation Testing of Spoil

Two testing programs were conducted to investigate the degradation of fresh spoil. To understand

how spoil materials would behave under different conditions, the influence of prolonged saturation

and wetting and drying cycles were investigated. Once conditions were identified that would produce

appropriate degrees of in situ degradation, the methodologies were applied to all fresh spoils

collected, with methodologies discussed in Section 4.9.3.

4.9.1 Varied saturation durations

Prolonged saturation was used to determine how different durations of flooding would influence a

spoil sample, and at what rate the spoil degrades. To test the spoil, representative samples of 1.5 kg

were prepared for each material out of the -19 mm fraction. Each sample was exposed to different

durations of saturation, after which its particle size distribution was determined via wet sieving and

hydrometer analysis. During saturation, measurements were made of the electrical conductivity for

each sample, as well as the total dissolved salts.

4.9.2 Multiple wetting and drying cycles

Multiple wetting and drying cycles were used on representative samples of 1.5 kg of spoil out of the

-19 mm fraction. The aim of this testing was to determine how rainfall or flooding could degrade

spoil materials and at what rate. For each sample investigated, the particle size distribution was

analysed after three wetting and drying cycles had been completed. After each wetting cycle, the

electrical conductivity and total dissolved solids were measured.

4.9.3 Spoil degradation testing program

To simulate the degradation of the spoil materials on wetting-up, three methods of degradation with

tap water were used. Dry sieving was used to determine the particle size distribution of the as-sampled

spoil. This is the state it would be in on visual classification using the BMA framework. Submersion

with tap water over 24 hours was used to break down any loosely conglomerated particles. Wetting

and drying cycles were used to accelerate the rate of degradation and promote slaking of the material.

The modified slake durability test was used to reduce the time required for the material to degrade,

while still undergoing wetting and drying cycles in a reliable and repeatable manner.

4.9.3.1 24 Hour Saturation

A 1.5 kg representative sample of each spoil material was collected and submerged in tap water for

24 hours. To promote saturation, each sample was gently agitated to remove air pockets and wet-up

dry zones. Once submerged, each sample was covered with a lid to reduce evaporation and stored at

145

room temperature. After 24 hours, the particle size distributions were determined by wet sieving and

hydrometer analysis.

4.9.3.2 Wetting and drying cycles

To simulate in situ conditions during a wet season or with cycling flooding, wetting and drying cycles

were applied to the spoil samples. For each material, a representative sample of 1.5 kg was prepared,

and placed into a stainless-steel tray. For each degradation cycle, tap water simulating rain and mine

runoff was used to submerge the material for 96 hours, which was covered by a lid to reduce

evaporation, and was stored at room temperature. Each spoil sample was then placed in a 60°C oven

to dry, prior to the next wetting and drying cycle. For each sample, three wetting cycles and two

drying cycles were conducted. At the end of the third wetting-up, the particle size distribution was

determined by wet sieving and hydrometer analysis.

4.9.3.3 Modified slake durability testing

For each spoil material, a representative 1 kg dry mass sample was prepared, limited by the size of

the slake durability cells. Each sample was left within a rotating slake durability cell until all the

particles finer than 2 mm had passed through. The sample and cell were then weighed. The method

detailed in AS 4133.3.4 (2005) was applied, with the following adjustments:

• three wetting cycles instead of two;

• 60°C oven instead of 110°C to avoid combustion of any carbonaceous material; and

• particles finer than 2 mm were left in tap water for 24 hours.

After the third wetting and drying cycle, the particle size distribution was determined by dry sieving

for particles coarser than 2.36 mm. For particles finer than 2.36 mm, the particle size distribution was

determined by wet sieving and hydrometer analysis. An image of the slake durability apparatus is


Figure 4.54 Slake durability apparatus

146

Table 4.9 Summary of degradation testing program

CODE CATEGORY SOURCE

PR

OL

ON

GE

D

SA

TU

RA

TIO

N

MU

LT

IPL

E W

ET

TIN

G

AN

D D

RY

ING

CY

CL

ES

DEGRADATION TESTING

Sa

tura

tio

n

(24

Ho

urs

)

3X

Wet

tin

g a

nd

Dry

ing

Mo

dif

ied

Sla

ke

Du

rab

ilit

y T

esti

ng

C3S-13 CAT 3 SPOIL √ √ √ √ √



C3S-20 CAT 3 SPOIL √ √ √ √ √


147

4.10 Geotechnical Characterisation

4.10.1 Small-scale consolidometer testing

Compression testing provides a means of assessing the settlement of coal mine spoil and mud in wet

conditions. In Stage 1, the compression behaviour of the spoil was assessed in a conventional

laboratory consolidometer measuring 76 mm in diameter by about 20 mm deep, and capable of

applying stresses of up to 1,000 kPa (see Figure 4.55). The testing was carried out broadly in

accordance with AS 1289.6.6.1 (1998), on -4.6 mm scalped samples. Scalping was carried out by dry

sieving of air-dried samples. The degree of scalping is made necessary by the depth of the specimens

in the oedometer being limited to a maximum of about 20 mm, in turn limiting the maximum particle

size to about 1/10th (or at most 1/5th) of this dimension.

Specimens of each material tested were prepared loose in the consolidometer and were subjected to

on average 24-hour loading stages in increments of 25 kPa, 50 kPa, 100 kPa, 200 kPa, 500 kPa, and

1,000 kPa, with the maximum loading increment being limited by the capacity of the testing machine.

Figure 4.55 Schematics of consolidometer testing in a water bath (tested “wet”)

4.10.2 Large slurry consolidometer

A purpose-built large slurry consolidometer was manufactured by Wille Geotechnik for the

University of Queensland’s Geotechnical Engineering Centre. A schematic of the device is shown in

Figure 4.56. The consolidation cell was designed to allow for both top and bottom drainage, providing

the ability to simulate numerous in situ conditions.

The consolidation cell has an internal diameter of 150 mm and a height of 410 mm. Current load cells

have a capacity of 10 kN, allowing for up to 566 kPa of applied stress to be measured at the top and

bottom of the cell. This load cell placement also allows the skin friction of the cell to measured and

accounted for within the results. Seven pore water transducers allow for pore water pressure

measurement at different heights within the cell.

148

The large slurry consolidometer was used to measure the effect of continuous loading on mud samples

in a slurry state. A continuous loading rate of 0.1 kPa/min was used up to a maximum applied stress

of 500 kPa. The final applied stress of 500 kPa was held until all pore water pressures had dissipated

from the tested specimen.

Figure 4.56 Large slurry consolidometer apparatus schematic

4.10.3 Small-scale and large-scale shear strength testing

Shear strength testing was largely carried out using a standard 60 mm by 60 mm (by approximately

30 mm high specimen) direct shear box, broadly in accordance with AS 1289.6.2.2 (1998). The

samples were scalped to pass a 6.7 mm sieve to ensure a specimen height to maximum particle size

ratio of at least five. A 300 mm by 300 mm by nominally 200 mm high large direct shear box was

used for the coarse-grained Category 3 spoil tested dry as-sampled (scalped to -53 mm on-site). Due

to sample requirements, only a select number could be tested in the large direct shear box.

All specimens were placed loose, to represent loose dumping and a normally-consolidated specimen.

The spoil specimens tested “dry” were placed at their as-sampled gravimetric moisture content, which

was less than a few percent. The specimens to be tested wet were soaked for 24 hours in deionised

water to simulate flooding, prior to the normal stress being applied.

In both the small and large direct shear boxes, three single-stage tests were carried out on each

material type, at nominal normal stresses of 200 kPa (or 250 kPa), 500 kPa and 1,000 kPa, simulating

spoil depths of approximately 10 m (or 12.5 m), 25 m and 50 m. Each specimen was left to settle and

consolidate under the applied normal stress before shearing at a relatively slow rate of 0.1 mm/min

149

for the small shear box, and 1 mm/min for the large shear box, with intentions of inducing drained

behaviour. The specimens were sheared to 10% of the shear box length. If the material reached a peak

shear strength prior to 10% shear displacement, it was recorded. If not, the shear strength at 10% was

recorded. The normal and shear stresses at failure were corrected for the reduction in area due to shear

displacement.

Figure 4.57 Schematic of direct shear box shear strength test

Figure 4.58 Large-scale direct shear machine (300 mm x 300 mm x 200 mm high)

150

Table 4.10 Summary of geotechnical testing

CODE CATEGORY SOURCE

CONSOLIDATION DIRECT SHEAR

Sm

all

-Sca

le O

edo

met

er

La

rge

Slu

rry

Co

nso

lid

om

eter

In S

itu

Mo

istu

re C

on

ten

t

24

Ho

ur

Sa

tura

tio

n

3x

Wet

tin

g a

nd

Dry

ing

Cy

cles

C3S-01 CAT 3 MUD √ 60 X 60

C1S-02 CAT 1 MUD √ √ 60 X 60

C3S-03 CAT 3 MUD √ 60 X 60

C3S-04 CAT 3 MUD √ 60 X 60

C3M-05 CAT 2 MUD √ 60 X 60

C3S-06 CAT 2 MUD √ 60 X 60

C3M-07 CAT 2 MUD √ 60 X 60

C1M-32 CAT 1 MUD √ 60 X 60

C3M-08 CAT 3 MUD √ √ 60 X 60

C3S-10 CAT 3 SPOIL √ 60 X 60 300 X 300 60 X 60

C3M-12 CAT 3 MUD √ 60 X 60

C3S-13 CAT 3 SPOIL √ 300 X 300 60 X 60

C3S-13 +

C3M-08 CAT 3

SPOIL+

MUD 1.5:1 √ 300 X 300

C3S-16 CAT 2 SPOIL √ 60 X 60 60 X 60

C3S-30 CAT 3 MUD √ 60 X 60 60 X 60

C3S-17 CAT 1 SPOIL √ 60 X 60 60 X 60

C3M-18 CAT 3 MUD √ √ 60 X 60

C3S-20 CAT 3 SPOIL √ 300 X 300 60 X 60

C3S-20 +

C3M-18 CAT 3

SPOIL+

MUD 1.5:1 √ 300 X 300

C1M-23 CAT 1 MUD √ √ 60 X 60

C2S-24 CAT 2 SPOIL √ 60 X 60 60 X 60 60 X 60

151

5 MATERIAL CHARACTERISATION TEST RESULTS

A large testing program was conducted following the methodology detailed in Section 4.7, with aims

of characterising numerous spoil and mud materials collected from three mines within the Bowen

Basin. This chapter focuses on the physical, chemical, mineralogical and geochemical test results. All

spoil materials are discussed with respect to their assigned BMA spoil category. The mud materials

were labelled with respect to the category of their spoil source from which they formed.

5.1 Physical Characterisation

5.1.1 As-sampled moisture state

For each of the spoil and mud materials collected, the as-sampled gravimetric moisture content (mass

of water/mass of solids, expressed as a percentage) and total moisture content (mass of water/total

mass, expressed as a percentage) are given in Table 5.1. The moisture content was determined on the

-19 mm fraction to ensure accuracy and repeatability. Typically, the lowest gravimetric moisture

contents were obtained for the spoil materials collected above any previous flooding levels (1.1 to

3.3%). C3S-10, a spoil material that was previously below past flooding levels had a moisture content

of 9.5%, highlighting the spoils ability to retain moisture near the surface of the spoil pile post-

flooding. Moderate moisture contents were recorded for dried mud materials that had previously been

exposed to flooding (C2M-06 and C2M-07).

The as-sampled gravimetric moisture contents of the Category 3 muds ranged from 13.3 to 94.5%.

That of the Category 2 muds ranged from 4.5 to 12.9%, and that of the Category 1 muds ranged from

9.5 to 90.9%. The values given in Table 5.1 are also plotted in Figure 5.1.

152

Table 5.1 As-sampled gravimetric moisture content of all spoil and mud samples

CODE SOURCE

CATEGORY TYPE

GRAVIMETRIC

MOISTURE

CONTENT (%)

TOTAL MOISTURE

CONTENT (%)

C3S-10 CAT 3 SPOIL 9.5 8.7

C3S-13 CAT 3 SPOIL 1.1 1.1

C3S-20 CAT 3 SPOIL 2.1 2.1

C3M-01 CAT 3 MUD 20.1 16.7

C3M-03 CAT 3 MUD 24.0 19.4

C3M-04 CAT 3 MUD 13.3 11.7

C3M-05 CAT 3 MUD 20.5 17.0

C3M-08 CAT 3 MUD 15.1 13.1

C3M-12 CAT 3 MUD 25.0 20.0

C3M-30 CAT 3 MUD 16.0 13.8

C3M-18 CAT 3 MUD 94.5 48.6

C2S-16 CAT 2 SPOIL 3.3 3.2

C2S-24 CAT 2 SPOIL 1.9 1.8

C2M-06 CAT 2 MUD 12.9 11.4

C2M-07 CAT 2 MUD 4.5 4.3

C1S-17 CAT 1 SPOIL 2.0 1.9

C1M-02 CAT 1 MUD 46.1 31.5

C1M-23 CAT 1 MUD 90.9 47.6

C1M-32 CAT 1 MUD 9.5 8.6

153

Figure 5.1 As-sampled moisture content of all spoil and mud samples

5.1.2 Total suction

Table 5.1 shows the total suction for each of the collected spoil samples. The range of all values was

from 50.4 to 68.5 MPa, with one outlier at 0.7 MPa for C3S-10, which was collected below a previous

flooding level.

Table 5.2 As-sampled moisture state of -2.36 mm scalped spoil samples

CODE SOURCE

CATEGORY TYPE

GRAVIMETRIC

MOISTURE

CONTENT (%)

TOTAL

SUCTION

(MPa)

C3S-10 CAT 3 SPOIL 9.5 0.7

C3S-13 CAT 3 SPOIL 1.1 62.1

C3S-20 CAT 3 SPOIL 2.1 44.6

C2S-16 CAT 2 SPOIL 3.3 51.9

C2S-24 CAT 2 SPOIL 1.9 50.4

C1S-17 CAT 1 SPOIL 2.0 68.5

Figure 5.2 shows the gravimetric moisture content plotted against the total suction for each spoil

specimen. There is no clear relationship between material category, moisture content and total suction

for the materials collected.

154

Figure 5.2 As-sampled gravimetric moisture content and total suction of all spoil samples

155

5.1.3 Specific gravity

For each spoil and mud material, the specific gravity was calculated. No clear relationship is apparent

between the specific gravity, the source BMA spoil category. Carbonaceous material was identified

in several materials, associated with a reduction in specific gravity. The results given in Table 5.3 are

also plotted in Figure 5.3.

Table 5.3 Specific gravity of all spoil and mud samples

CODE SOURCE

CATEGORY TYPE SPECIFIC GRAVITY

C3S-10 CAT 3 SPOIL 2.43



C3M-01 CAT 3 MUD 2.56

C3M-03 CAT 3 MUD 2.49

C3M-04 CAT 3 MUD 2.56

C3M-05 CAT 3 MUD 2.57

C3M-08 CAT 3 MUD 2.23

C3M-12 CAT 3 MUD 2.27

C3M-30 CAT 3 MUD 2.56

C3M-18 CAT 3 MUD 2.40



C2M-06 CAT 2 MUD 2.71

C2M-07 CAT 2 MUD 2.59


C1M-02 CAT 1 MUD 2.55

C1M-23 CAT 1 MUD 2.60

C1M-32 CAT 1 MUD 2.23

Pure Water 1.00

Pure Coal 1.30

Nominal mineral matter 2.65

156

Figure 5.3 Specific gravity of all spoil and mud samples

157

5.1.4 Particle size distributions

All collected samples were analysed to determine their particle size distribution. For the spoil, both

wet and dry sieving were used to characterise the materials. The purpose of dry sieving was to

determine the in situ particle size distribution. Wet sieving was undertaken to breakdown any weakly

agglomerated fines that would not be identified during standardised dry sieving, and to promote

slaking in highly degradable material.

The particle size distributions of all the materials collected were analysed with respect to their BMA

spoil category, with results plotted in Figure 5.4 to Figure 5.20. For all materials, the D90, D50, D10,

Cu and Cc values were calculated and are given in Table 5.4 to Table 5.9. D90 values for all spoils

tested wet and dry, and all muds tested wet are plotted in Figure 5.8 and Figure 5.14, respectively.

Figure 5.4 Overall particle size distribution curves of Category 3 spoil -53 mm fraction

158

Table 5.4 D90, D50, D10, Cu and Cc for Category 3 spoil wet and dry sieving

CODE SOURCE

CATEGORY TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C3S-10 CAT 3 SPOIL DRY 36.51 12.04 7.616 2.351 0.272 44.32 1.689

C3S-10 CAT 3 SPOIL WET 24.64 5.136 2.861 0.365 0.056 91.97 0.465

C3S-13 CAT 3 SPOIL DRY 32.25 8.829 6.436 2.957 0.738 11.96 1.341

C3S-13 CAT 3 SPOIL WET 17.39 4.640 3.077 1.323 0.148 31.40 2.552

C3S-20 CAT 3 SPOIL DRY 34.87 17.66 14.22 7.949 2.142 8.246 1.671

C3S-20 CAT 3 SPOIL WET 22.83 5.355 2.835 0.232 0.016 338.9 0.638

Figure 5.4 details the particle size distribution of all Category 3 spoil after dry and wet sieving. C3S-

10 is a 10-year-old spoil sampled below a previous flooding level, and approximately 5m above the

current water level. Contrasting dry sieving to wet sieving after 24 hours of saturation shows up to a

magnitude of size reduction, displayed by increases in sand and silt content. C3S-13 was collected

from the same spoil pile above any previous flooding levels. The results show a similar particle size

distribution to C3S-10 when tested dry. Upon wetting, the degradation experienced was lower than

that of C3S-10, with approximately half a magnitude of size reduction. Due to the age of these two

materials, it is likely that some degradation has already occurred to some degree due to the

environmental conditions.

C3S-20 is relatively fresh spoil that was collected six months after the blasting of the highwall. The

particle size distribution when tested via dry sieving was the coarsest of all the Category 3 spoils.

Upon wetting, it experienced the largest reduction in particle size, with significant increases in sand

and silt content. The presence of clay was also observed. Due to the relatively young age of this spoil,

it is interesting to observe the significant amount of degradation after only 24 hours of exposure to

water. This is highlighted with respect to the D90, D50, and D10 values showing a reduction of 35%,

80% and 99%, respectively.

Figure 5.5 shows the wet and dry particle size distributions obtained for the two Category 2 spoil

materials collected. C2S-16 is a 5-year old spoil that was collected above any historical water levels.

C2S-24 was freshly exposed spoil. The coarsest dry particle size distribution was obtained with C2S-

16. Upon exposure to 24 hours of wetting and wet sieving, degradation of two and three orders of

magnitude in particle size was observed. These results show that the spoil was very weakly cemented.

Less degradation is observed in C2S-24, with a size distribution reduction of approximately one

magnitude when exposed to soaking and wet sieving.

159

From the two Category 2 spoils tested, a high degree of variability is observed. This is likely due to

numerous reasons, including the sampling locations, age of the spoil, and aspects such as the

mineralogical composition and chemical parameters.



CODE SOURCE

CATEGORY TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C2S-16 CAT 2 SPOIL DRY 32.11 11.31 8.006 3.617 0.504 22.43 2.294

C2S-16 CAT 2 SPOIL WET 1.467 0.162 0.044 0.006 0.001 124.1 0.180

C2S-24 CAT 2 SPOIL DRY 23.10 4.335 1.961 0.430 0.123 35.28 0.347

C2S-24 CAT 2 SPOIL WET 2.583 0.302 0.185 0.050 0.009 34.19 0.944

Figure 5.6 displays the results of the particle size distribution analysis for wet and dry sieving for

C1S-17, a Category 1 spoil. C1S-17 was sampled fresh from a Tertiary spoil pile. Dry sieving

categorised the material as a Gravelly SAND, with very little content measured above 10 mm.

Soaking and wet sieving caused one to two orders of magnitude of particle size degradation. The

results of the wet sieving show a Silty SAND. These results show that the material was likely weakly

agglomerated. No clay-sized particles were observed.

160



CODE SOURCE

CATEGORY TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C2S-16 CAT 2 SPOIL DRY 15.12 4.027 1.746 0.337 0.118 15.04 0.918

C2S-16 CAT 2 SPOIL WET 0.915 0.233 0.174 0.058 0.015 15.04 0.918

The particle size distribution curves obtained on dry sieving of air-dried Category 1, 2 and 3 spoils

are presented in Figure 5.7, which shows that the dry-sieved Category 3 spoil is generally more

coarse-grained than the dry-sieved Category 2 spoil, which is, in turn, more coarse-grained than the

dry-sieved Category 1 spoil, as would be expected.

With respect to the spread of particle size data, those for dry-sieved Category 3 spoil had a slightly

smaller spread than those for dry-sieved Category 2 spoil. The D50 (size through which 50% of the

particles pass) values range from 6.5 to 14.2 mm for dry-sieved Category 3 spoil, from 1.9 to 8 mm

for dry-sieved Category 2 spoil, and 1.8 mm for the dry-sieved Category 1 spoil tested.

161

Figure 5.7 Overall particle size distribution curves of all dry spoil -53 mm fraction

Figure 5.8 shows the D90 for all spoil wet and dry sieving. With respect to dry sieving, a clear

differentiation is seen between the categories. For the spoil after soaking and wet sieving, the

Category 3 materials remain the coarsest. For the Category 2 and 1 spoil, there is less visible

differentiation.

Figure 5.9 highlights the differences in particle size distribution for dry sieving and wet sieving for

all categories of spoil. The results of the dry sieving show a smaller spread than that of wet sieving.

Furthermore, the distinction between categories is more pronounced for dry sieving. After 24 hours

of soaking followed by wet sieving, there is little differentiation between the Category 1 and 2 spoils

below the D90 value. For wet sieving, the coarsest particle size distribution is observed for the

Category 3 materials, with a similar spread as seen in the dry sieving. All categories of materials

tested by dry sieving were Sandy GRAVEL, degrading on wet sieving to combinations of clay, silt,

sand and gravel.

162

Figure 5.8 D90 values for all dry and wet sieved spoil samples

Figure 5.9 Overall particle size distribution curves of all spoil wet and dry -53 mm fraction

Figure 5.10 shows the particle size distributions obtained via soaking and wet sieving of all the

sampled Category 3 muds, with values of D90, D50, D10, Cu and Cc recorded in Table 5.7. Eight

samples were collected in total from three mine sites. The typical spread of results is shaded, with up

163

to two orders of magnitude variation. C3M-18 is identified as a clear outlier. Values for D90 values

ranged from 9.7 to 41 mm. Most materials were Silty, Sandy GRAVEL or Silty, Gravelly SAND.

C3M-18 was composed almost entirely of silt-sized particles. C3M-01, C3M-03, C3M-04 and C3M-

05 were all fresh muds sampled within the same pit, showing the variation that can occur within small

distances. C3M-08 and C3M-12 were mud samples that were ten years old. C3M-18 was mud

collected from degraded spoil/floor material associated with C3S-20, a 6-month old spoil.

Figure 5.10 Overall particle size distribution curves of all Category 3 mud samples -53 mm

fraction

Table 5.7 D90, D50, D10, Cu and Cc for Category 3 mud wet sieving

CODE

SOURCE

CATEGOR

Y

TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C3M-01 CAT 3 MUD WET 41.26 3.422 0.776 0.066 0.010 360.1 0.135

C3M-03 CAT 3 MUD WET 21.58 2.475 0.793 0.061 0.011 216.9 0.133

C3M-04 CAT 3 MUD WET 30.59 7.459 4.254 0.308 0.011 675.1 1.148

C3M-05 CAT 3 MUD WET 26.64 8.633 6.596 3.538 0.058 149.53 25.11

C3M-08 CAT 3 MUD WET 13.81 1.903 0.985 0.239 0.049 38.76 0.613

C3M-12 CAT 3 MUD WET 18.56 1.868 0.969 0.206 0.021 89.50 1.091

C3M-30 CAT 3 MUD WET 9.734 2.961 1.752 0.247 0.013 224.3 1.557

C3M-18 CAT 3 MUD WET 0.063 0.016 0.012 0.008 0.007 2.28 0.557

164

Figure 5.11 shows the particle size distribution of two Category 2 mud materials. Both materials had

previously been exposed to flooding. The water was removed, and both materials underwent drying

in situ. The results show that for both mud specimens, there is a significant degree of degradation

when exposed to water. C2M-06, when sampled dry, had a particle size distribution that would be

described as a Sandy GRAVEL. Upon soaking and with wet sieving, degradation of up to three orders

of magnitude is observed, showing a Clayey Sandy SILT. This shows the material is weakly cemented

with very few coarse particles present. This is reinforced by a reduction in the D90 of over 98%. This

is important to note as purely visual observation can lead to incorrect particle size distribution

assumptions.

C2M-07 shows less degradation, with a particle size reduction of one to two orders of magnitude,

going from a Sandy GRAVEL to a Silty Gravely SAND. The D90 for C2M-07 went from 40 to

9.3 mm, which is considerable, however relatively small in comparison to C2M-06. While both

materials have similar dry particle size distributions, exposure to water had a drastically different

influence on each. This shows the difficulty that could be expected in trying to predict the degradation

of each material visually.

Figure 5.11 Overall particle size distribution curves of Category 2 dried mud -53 mm fraction

165

Table 5.8 D90, D50, D10, Cu and Cc for Category 2 desiccated mud wet and dry sieving

CODE

SOURCE

CATEGOR

Y

TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C2M-06 CAT 2 MUD DRY 29.54 10.46 7.364 3.284 0.602 17.36 1.713

C2M-06 CAT 2 MUD WET 0.573 0.015 0.011 0.005 0.002 8.94 0.995

C2M-07 CAT 2 MUD DRY 40.98 6.337 4.182 1.301 0.280 22.67 0.956

C2M-07 CAT 2 MUD WET 9.328 0.105 0.271 0.035 0.007 15.12 1.701

Figure 5.12 shows the wet sieved particle size distributions of all sampled Category 1 muds. C1M-

02 is an old Category 1 mud of unknown age. The material was submerged in water for a prolonged

duration. C1M-23 is 6.5-year-old mud produced from a Tertiary spoil. Samples were collected from

the edge of a flooded pit. C1M-32 is a fresh mud collected from a pit floor near the highwall. The

results show all three materials have an extremely fine particle size distribution composed mostly of

silt, with some sand and gravel-sized particles present. All distributions have a spread of less than

one magnitude.

Figure 5.12 Overall particle size distribution curves of all Category 1 mud -53 mm fraction

166

Table 5.9 D90, D50, D10, Cu and Cc for Category 1 mud wet sieving

CODE

SOURCE

CATEGOR

Y

TYPE

DRY OR

WET

SIEVING

D90

(mm)

D60

(mm)

D50

(mm)

D30

(mm)

D10

(mm) Cu Cc

C1M-02 CAT 1 MUD WET 0.271 0.058 0.043 0.013 0.009 6.233 0.319

C1M-23 CAT 1 MUD WET 0.073 0.017 0.008 0.004 0.004 4.806 0.307

C1M-32 CAT 1 MUD WET 0.493 0.018 0.013 0.009 0.008 2.154 0.566

Figure 5.13 shows the distributions of all tested mud materials, categorised based on their source’s

BMA spoil. The smallest spread of results is observed with the Category 1 muds, all showing

extremely fine particle sizes comprised mostly of silt. The Category 3 mud shows a spread of results

from one to two orders of magnitude. The largest range was observed within the Category 2 mud

samples tested. The coarsest distributions were related to the Category 3 mud, followed by Category

2 and 1 mud. From Figure 5.13, it can be observed that the differentiation of Category 2 and 1 material

could be difficult. The results also show differing degrees of cementation observed, indicating the

large potential variation of materials that can be found in situ. The one Category 3 outlier shown as

almost entirely silt-sized particles indicates that spoil the mud originates from does not necessarily

predict the particle size distribution that can be expected of the material once it is exposed to

saturation and resultant degradation.

Figure 5.13 Overall particle size distribution curves of all mud samples -53 mm fraction

167

Figure 5.14 shows the D90 values of all the Category 1, 2 and 3 muds and spoils after wet-sieving. On

average, the Category 3 D90 values are significantly higher than both Category 2 and Category 1

materials. Within the Category 3 materials, there is no clear differentiation between the spoil and the

mud samples. For the Category 2 spoil and mud materials, again there is again no clear differentiation.

C2M-07 shows a significantly higher D90 than both spoil samples, equivalent to the worst Category

3 mud sample. The D90 values for the Category 1 spoil and mud materials are all extremely low in

comparison to the Category 2 and 3 materials.

Figure 5.15 shows the dry and wet particle size distribution curves for all Category 3 spoil and mud

materials. Wetting resulted in approximately an order of magnitude reduction in particle sizes. Most

of the collected muds had higher sand and silt-sized fractions than the wet spoil.

Figure 5.16 shows the dry and wet particle size distribution curves for Category 2 spoil and mud

samples. Wetting of the Category 2 spoil samples resulted in approximately three orders of magnitude

reduction in particle sizes, highlighting a high potential for breakdown in water. The tested Category

2 muds showed a large difference in particle size for the two materials, indicating potential difficulties

that could arise when categorising this material in situ. This could also result in inaccurate estimates

of the materials geotechnical parameters, including consolidation, permeability and shear strength,

all of which are related to lowwall stability.

Figure 5.14 D90 values for all wet sieved spoil and mud samples

168

Figure 5.15 Overall particle size distribution curves of all Category 3 spoil and mud samples -

53 mm fraction


53 mm fraction

169

Figure 5.17 shows the dry and wet particle size distribution curves for all Category 1 spoil and mud

materials. Wetting of the Category 1 spoil specimens showed about an order of magnitude reduction

in particle sizes from those obtained on dry sieving and wetting of the mud samples showed about a

further order of magnitude reduction.

Figure 5.18 shows the variation of particle size distributions of all Category 3 spoil and mud collected

within one mining pit. Spoil was collected from the lowwall above the previous flooding height (C3S-

13) and below the previous flooding height (C3S-10). Mud was collected at the water’s surface from

both the lowwall (C3M-08 and C3M-12) and the highwall (C3M-30). The results show degradation

from dry to wet sieving for the spoil below the previous flooding level was significantly higher than

the degradation observed from spoil which had never been exposed to prolonged saturation. The mud

collected at both the lowwall and highwall had very similar particle size distributions. The particle

size distribution of the spoil collected below the previous flooding height had a very similar

distribution to the sampled muds when wet sieved.

Figure 5.19 shows the particle size distribution of 6-month-old spoil (C3M-20) collected from the

lowwall above any past water levels, and fresh mud (C3M-18) produced by in-pit flooding. The

results show significant degradation of the spoil when exposed to water, with two to three orders of

magnitude particle size reduction after 24 hours of soaking. The mud collected on the pit floor has a

particle size distribution that is two to three orders of magnitude lower than the wet-sieved spoil.

Figure 5.19 depicts material that is highly susceptible to degradation. If a design were to be based off

the results of the fresh, dry spoil, it is possible that the spoil shear strength would be overestimated if

the spoil was to undergo any reasonable degree of saturation from heavy rainfall or flooding.

Figure 5.20 shows the large degree of variability that can occur within a single mine pit. Within this

one pit, there were mud samples identified and collected, ranging from Category 1 to Category 3.

Two Category 2 muds (C2M-06 and C2M-07) were within the pit in a dried, re-agglomerated state.

Both the dry and wet sieving results of these two samples were included, showing the potential

degradation that reasonably competent material can experience when exposed to water. Of the

materials collected, the coarsest was the Category 3 muds, all of which were collected on the floor of

the pit and base of the lowwall.

170


53 mm fraction

Figure 5.18 Overall particle size distribution curves Mine Site B R5S spoil and mud -53 mm

fraction

171

Figure 5.19 Overall particle size distribution curves Mine Site C R6S spoil and mud -53 mm

fraction

Figure 5.20 Overall particle size distribution curves Mine Site A mud -53 mm fraction

172

5.2 Geotechnical Characterisation

5.2.1 Atterberg limits

Each material was tested to determine its plastic and liquid limits. From these results, the plasticity

index was calculated and plotted on a plasticity chart. The resulting classification for each material

and the associated liquid and plastic limits are given in Table 5.10, with the range of each category

provided. The in situ moisture content and Atterberg limits for each material are plotted in

Figure 5.21. The plasticity chart is plotted in Figure 5.22.

Table 5.10 Atterberg limits and plasticity index for all spoil and mud samples

CODE SOURCE

CATEGORY TYPE

PLASTIC

LIMIT

(%)

LIQUID

LIMIT

(%)

PLASTICITY

INDEX (%)

PLASTICITY

CHART

CLASSIFICATION

C3S-10 CAT 3 SPOIL 20.0 28.1 8.1 CL/OL

C3S-13 CAT 3 SPOIL 19.4 26.6 7.2 CL/OL

C3S-20 CAT 3 SPOIL 19.4 40.3 20.9 CL/OL

C3M-01 CAT 3 MUD 17.4 30.2 12.7 CL/OL

C3M-03 CAT 3 MUD 18.6 31.5 12.9 CL/OL

C3M-04 CAT 3 MUD 16.9 27.8 10.9 CL/OL

C3M-05 CAT 3 MUD 18.6 30.7 12.1 CL/OL

C3M-08 CAT 3 MUD 18.0 26.9 8.9 CL/OL

C3M-12 CAT 3 MUD 16.5 26.2 9.6 CL/OL

C3M-30 CAT 3 MUD 18.5 29.4 10.9 CL/OL

C3M-18 CAT 3 MUD 26.5 61.8 35.3 CH/OH

Category 3 Ranges 16.5-26.5 26.2-61.8 7.2-35.3 CL-CH

Category 3 Averages 19.1 32.7 13.6 CL

C2S-16 CAT 2 SPOIL 12.1 21.5 9.4 CL/OL

C2S-24 CAT 2 SPOIL 18.7 39.4 20.7 CL/OL

C2M-06 CAT 2 MUD 25.1 44.6 19.5 CL/OL

C2M-07 CAT 2 MUD 20.1 33.3 13.2 CL/OL

Category 2 Ranges 12.1-25.1 21.5-44.6 9.4-20.7 CL


C1S-17 CAT 1 SPOIL 14.4 21.9 7.5 CL/OL

C1M-02 CAT 1 MUD 19.6 28.1 8.5 CL/OL

C1M-23 CAT 1 MUD 27.1 61.9 34.7 CH/OH

C1M-32 CAT 1 MUD 15.9 41.7 25.7 CL/OL

Category 1 Ranges 14.4-27.1 21.9-61.9 7.5-34.7 CL-CH


173

Figure 5.21 shows the as-sampled moisture states of the materials compared with their Atterberg

limits. The results show that the as-sampled moisture state of all spoil materials was well below their

plastic limit. A larger variation was observed for the tested muds, ranging from samples with as-

sampled moisture contents below their plastic limit to well above their liquid limit.

Figure 5.21 As-sampled gravimetric moisture content and Atterberg limits of all spoil and

mud samples

The plasticity index and liquid limit for all spoil and mud materials are plotted in Figure 5.22. The

results show that all materials are classified as clays of low plasticity (CL), apart from C3M-18 and

C1M-23, which classify as clays of high plasticity (CH). There is no clear relationship between the

spoil classification and their Atterberg limits. This is noteworthy since 29% of the weighting of the

BMA spoil classification framework is based on the liquid limit.

Figure 5.23 shows the plastic and liquid limits plotted for all samples tested. No relationship can be

seen between these parameters and the spoil classifications. Furthermore, there is no clear

differentiation between the plastic and liquid limits of the spoil and mud samples. There is, however,

a strong relationship, with an r2 value of 0.6971, between the plastic and liquid limits of the tested

materials.

174

Figure 5.22 Plasticity chart for all spoil and mud samples

Figure 5.23 Liquid and plastic limits for all spoil and mud samples

175

5.2.2 Emerson class number

The Emerson class test is used to determine if the tested soil will deflocculate in water, with results

dividing the soil into one of seven classes. Emerson class testing took place on all spoil materials

collected. The results given in Table 5.11 show that for C3S-10 and C3S-13, slaking took place with

no dispersion. Further testing revealed the presence of calcite determining that both materials were

considered Class 4, indicating a moderate potential for erosion. All remaining spoil materials were

classified as Class 2, with slaking and some dispersion identified, indicating a high to very high

potential for erosion. Images taken of the specimens during testing are shown in Figure 5.24,

Figure 5.25 and Figure 5.26.

Table 5.11 Emerson class test results for all spoil samples

CODE SOURCE

CATEGORY TYPE

EMERSON CLASS

NUMBER INTERPRETATION

C3S-10 CAT 3 SPOIL 4 Slaking with no dispersion. Calcite

present C3S-13 CAT 3 SPOIL 4

C3S-20 CAT 3 SPOIL 2

Slaking with some dispersion C2S-16 CAT 2 SPOIL 2



176

Figure 5.24 C3S-10 and C3S-13 Emerson class test results



C3S-10

C1S-16 C2S-16

C3S-13

C3S-20 C2S-24

177

5.3 Chemical characterisation

5.3.1 pH, electrical conductivity and total dissolved solids

The pH, electrical conductivity and total dissolved solids for all spoil and mud materials are given in

Table 5.12. The measured pH values range from 7.75 to 9.27, with no clear relationship to spoil

category. The electrical conductivity ranges from 246 to 8920 µS/cm, with mud typically having

higher values than spoil. The gravimetric moisture content of each materials is plotted against its

electrical conductivity in Figure 5.27.

Table 5.12 pH, electrical conductivity and total dissolved solids for all spoil and mud samples

CODE SOURCE

CATEGORY TYPE PH

EC

(µS/CM) TDS (PPM)

C3S-10 CAT 3 SPOIL 8.80 295 214

C3S-13 CAT 3 SPOIL 8.96 246 174

C3S-20 CAT 3 SPOIL 9.27 1186 841

C3M-01 CAT 3 MUD 8.60 1417 1000

C3M-03 CAT 3 MUD 8.62 1490 1060

C3M-04 CAT 3 MUD 8.61 1399 991

C3M-05 CAT 3 MUD 8.51 1912 1360

C3M-08 CAT 3 MUD 8.74 1257 894

C3M-12 CAT 3 MUD 8.85 1219 849

C3M-30 CAT 3 MUD 8.98 2900 2050

C3M-18 CAT 3 MUD 8.56 2450 1730

C2S-16 CAT 2 SPOIL 8.21 1893 1350

C2S-24 CAT 2 SPOIL 8.82 2520 1780

C2M-06 CAT 2 MUD 8.98 1926 1360

C2M-07 CAT 2 MUD 9.04 977 694

C1S-17 CAT 1 SPOIL 8.43 4240 3000

C1M-02 CAT 1 MUD 8.47 3390 2400

C1M-23 CAT 1 MUD 8.72 1346 957

C1M-32 CAT 1 MUD 7.75 8920 6260

Figure 5.27 shows the gravimetric moisture content in relation to electrical conductivity for each

material. For the spoil specimens, a trend of increasing electrical conductivity with decreasing spoil

category is observed. On average, the moisture content of the mud materials collected on site was

higher than the spoil samples. One Category 1 specimen (C1M-32) is shown to have a significantly

higher electrical conductivity than all other materials tested.

178

Figure 5.28 shows the relationship between gravimetric moisture content and pH for all spoil and

mud materials. Most materials had pH values of between 8 and 9. For the Category 3 materials, a

weak trend is observed with the pH values for mud typically being lower than the tested spoils. For

the Category 1 and 2 materials, there is no clear relationship between gravimetric moisture content,

spoil category and pH. C1M-32 was found to have the lowest pH of 7.75.

Figure 5.27 Variation of gravimetric moisture content and electrical conductivity for all spoil

and mud samples

Figure 5.29 highlights the relationship between pH and electrical conductivity for all tested spoil and

mud materials. A linear trendline is plotted to the relationship showing an r2 value of 0.544, indicating

a moderately strong trend between a decreasing pH and an increased electrical conductivity.

179

Figure 5.28 Variation of gravimetric moisture content and pH for all spoil and mud samples

Figure 5.29 Variation of pH and electrical conductivity for all spoil and mud samples

180

5.4 Mineralogical and Geochemical Characterisation

5.4.1 X-Ray diffraction, cation exchange capacity and exchangeable cations

X-ray diffraction was used to obtain an understanding of the clay types and crystalline minerals

present in each of the spoil and mud materials. Representative samples of each material were analysed

by Mark Raven at CSIRO Land and Water Flagship Urrbrae SA. The results of the analysis are given

in Table 5.13 and are plotted in Figure 5.30, excluding the amorphous and unidentified materials. The

identified clays and minerals are normalised to 100%.

Table 5.13 shows that all materials were dominated by Quartz, Kaolinite, Illite-Smectite and Albite.

There are several other minerals observed in lower quantities in each material tested. Comparison of

the categories show that on average, Category 3 materials had less Quartz than Category 2 materials,

which had less than the Category 1 materials. The highest levels of Illite-Smectite were found in the

Category 3 materials, followed by the Category 1 materials. All material categories had similar levels

of Kaolinite and Albite. From these results, there is no clear relationship between a spoil or mud

source BMA spoil category and its mineralogical composition.

181

Table 5.13 Mineralogical analysis via X-ray diffraction for all spoil and mud samples,

excluding amorphous materials

CODE

QU

AR

TZ

KA

OL

INIT

E

CA

LC

ITE

AR

AG

ON

ITE

DO

LO

MIT

E

SID

ER

ITE

ILL

ITE

-

SM

EC

TIT

E

MIC

A/I

LL

ITE

CH

LO

RIT

E

AN

AT

AS

E

PY

RIT

E

AL

BIT

E

OR

TH

OC

LA

SE

HA

LIT

E

GE

OT

HIT

E

C3S-10 35 15 0 0 2 3 33 4 0 1 <1 7 0 0 0

C3S-13 35 6 0 0 1 2 40 5 0 <1 <1 11 0 0 0

C3S-20 28 11 0 0 4 1 44 1 1 1 0 9 0 <1 0

C3M-01 35 12 1 0 1 4 38 5 0 1 0 3 0 0 0

C3M-03 41 7 0 0 1 2 39 6 0 <1 <1 4 0 0 0

C3M-04 34 8 0 0 2 3 42 6 0 <1 <1 5 0 0 0

C3M-05 37 6 <1 0 3 6 36 6 0 <1 <1 5 0 0 0

C3M-08 38 14 <1 0 2 4 30 4 0 1 0 7 0 0 <1

C3M-12 36 8 0 0 2 1 39 3 0 1 <1 10 0 0 0

C3M-30 24 15 0 0 0 0 48 3 2 1 0 6 0 <1 0

C3M-18 35 12 1 0 1 4 38 5 0 1 0 3 0 0 0

Category

3

Averages

34.3 10.2 0.12 0 1.8 2.6 38.9 4.3 0.3 1 0 6.7 0 0 0

C2S-16 52 13 0 0 2 0 23 1 0 1 0 4 0 0 4

C2S-24 29 16 <1 0 9 4 28 <1 0 <1 0 13 0 0 0

C2M-06 32 14 3 0 2 <1 37 3 0 1 0 3 0 0 5

C2M-07 34 11 1 0 3 3 39 2 0 <1 0 6 0 0 0

Category

2

Averages

36.8 13.5 1.3 0 4 2.3 31.8 2 0 1 0 6.5 0 0 2.3

C1S-17 60 6 2 0 0 0 21 2 0 <1 0 6 2 <1 0

C1M-02 39 11 1 3 1 1 32 6 0 1 0 4 1 <1 0

C1M-23 24 19 0 0 <1 <1 46 <1 3 1 0 3 0 <1 3

C1M-32 30 17 0 0 0 1 38 3 0 1 <1 10 0 0 0

Category

1

Averages

38.3 13.3 0.8 0.8 0.3 0.7 34.3 3.7 0.8 1.0 0.0 5.8 0.8 0.0 0.8

182

Figure 5.30 Mineralogical analysis via X-ray diffraction for all spoil and mud samples

The cation exchange capacity (CEC) is a measure of the soil or clays ability to hold and exchange

cations. It is calculated by adjusting the charge or valance of each cation present for its atomic weight

to obtain the equivalent atomic weight in grams per positive charge, converting it to mille

equivalents/100 g (by multiplying by 10), and summing for all the cations present. This can be said

to be normalising the cation suite present for valance and atomic weight, thus providing a measure of

the number of potentially free negatively charged sites per 100 g of the soil or clay under

consideration. Since Smectites have the largest number of these sites available, (Illite-Smectite can

also have exchange sites, depending on age and structure), it can also be a measure of the Smectite

quantity present and the propensity to expend and hold water depending on the valance of the cation

present.

The CEC and exchangeable cation concentrations are given in Table 5.14, with calculated averages

for each spoil category. The total exchangeable cations and the CEC are plotted in Figure 5.31. The

CEC values of the spoil and mud tested ranged from 10 to 21 cmol(+)/kg. The tested Category 3 and

2 materials ranged from 10 to 15 cmol(+)/kg, with one Category 3 outlier of 19 cmol(+)/kg. The

Category 1 materials showed a larger degree of variation, from 10 to 21 cmol(+)/kg. The same

relationships are observed for the total exchangeable cations, with the tested Category 1 spoil and

mud materials showing the highest magnitude of exchangeable cations, with one outlier from a

183

Category 3 mud (C3M-18). The samples were most often dominated by magnesium cations, followed

by sodium and calcium. The least common cation across all materials was potassium.

Table 5.14 Exchangeable cations and cation exchange capacity of all spoil and mud samples

CODE SOURCE

CATEGORY TYPE

EXCHANGEABLE CATIONS AT pH 8.5

(cmol(+)/kg)

CATION

EXCHANGE

CAPACITY

(NH4+)

(cmol(+)/kg)

Ca2+ Mg2+ Na+ K+ Total

Cations

C3S-10 CAT 3 SPOIL 1.6 6.3 3.1 0.6 11 10

C3S-13 CAT 3 SPOIL 3.6 7.1 0.7 1.2 13 13

C3S-20 CAT 3 SPOIL 4.0 4.7 4.9 0.6 14 15

C3M-01 CAT 3 MUD 2.9 5.2 2.8 0.7 12 12

C3M-03 CAT 3 MUD 3.2 5.6 3.1 0.8 13 12

C3M-04 CAT 3 MUD 2.6 5.5 2.8 0.7 12 11

C3M-05 CAT 3 MUD 3.1 5.9 2.9 0.7 12 12

C3M-08 CAT 3 MUD 1.9 4.6 2.7 0.5 10 11

C3M-12 CAT 3 MUD Not tested due to similarity to C3M-08

C3M-30 CAT 3 MUD 0.9 6 6.2 0.8 14 12

C3M-18 CAT 3 MUD 4.3 7.4 7.2 0.6 20 19

Category 3 Averages 2.8 5.8 3.6 0.7 13.1 12.7

C2S-16 CAT 2 SPOIL 1.2 6.6 4.6 0.2 13 11

C2S-24 CAT 2 SPOIL 2.9 5.7 4.6 0.7 14 13

C2M-06 CAT 2 MUD 2.1 5.3 3.5 0.5 11 11

C2M-07 CAT 2 MUD 3.0 7.4 2.6 0.7 14 13

Category 2 Average 2.3 6.3 3.8 0.5 13.0 12.0

C1S-17 CAT 1 SPOIL 3.2 4.9 6.9 0.4 15 14

C1M-02 CAT 1 MUD 3.0 7.7 6.5 0.8 18 14

C1M-23 CAT 1 MUD 5.3 10.0 7.1 0.5 23 21

C1M-32 CAT 1 MUD 5.3 8.8 7.3 0.4 22 18

Category 1 Average 4.2 7.9 7.0 0.5 19.5 16.8

For each material, the CEC is dominated by the Kaolinite and Illite-Smectite components. From the

XRD analysis, it is not possible to determine to what degree the Smectite has turned into Illite. It is,

however, important to understand this relationship, as Smectite can be highly reactive to water, and

can cause a large degree of shrinkage and stiffening on drying and swelling and slaking on wetting.

To estimate the percentages of Illite and Smectite within each material, it was assumed that the CEC

of Illite, Smectite and Kaolinite are 20 cmol+/kg, 100cmol+/kg and 10 cmol+/kg, respectively. Using

184

this relationship, the measured CEC and the known percentage of Illite-Smectite, an estimate of the

Illite percentage was calculated using:

𝐼𝑙𝑙𝑖𝑡𝑒% =(𝑇𝑜𝑡𝑎𝑙 𝐶𝐸𝐶) − (𝐼𝑆% ∗ 𝑆𝑚𝑒𝑐𝑡𝑖𝑡𝑒 𝐶𝐸𝐶) − (𝐾𝑎𝑜𝑙𝑖𝑛𝑖𝑡𝑒% ∗ 𝐾𝑎𝑜𝑙𝑖𝑛𝑖𝑡𝑒 𝐶𝐸𝐶)

(𝐼𝑆% ∗ 𝐼𝑙𝑙𝑖𝑡𝑒 𝐶𝐸𝐶) − (𝐼𝑆% ∗ 𝑆𝑚𝑒𝑐𝑡𝑖𝑡𝑒 𝐶𝐸𝐶) (3)

From Equation (3), the Smectite content can be determined by subtracting the Illite percentage from

the total Illite-Smectite percentage. The results of this analysis are shown in Figure 5.32.

Figure 5.31 Cation exchange capacity and exchangeable cations for all spoil and mud samples

Figure 5.32 shows the estimated levels of Smectite and Illite in all the tested materials. Within each

spoil category, a range of results was observed. Higher levels of Smectite are recorded in the Category

2 and Category 1 materials, however C3S-20 and C3S-18 also show relatively high levels, both of

which were collected from the same mine pit. High levels of Smectite are seen in C3M-18, C1S-17,

C1M-23 and C1M-32. The lowest tested levels were measured in C3S-10, C3S-13, C3M-04, C2M-

06.

Figure 5.33 shows the exchangeable cations associated with the estimated levels of Smectite in each

of the materials. The highest levels of sodium Smectite were found in the Category 1 materials.

Relatively high levels were also recognised in C3S-20 and C3M-18. The lowest levels of sodium

Smectite measured were in C3S-13, C3S-10 and C3M-04.

185

Figure 5.32 Calculated levels of Smectite and Illite for all spoil and mud samples

Figure 5.33 Smectite cations present in all spoil and mud samples

186

5.5 Material Characterisation Test Conclusions

An extensive material characterisation testing program of the collected spoil and mud materials was

conducted, investigating numerous physical, chemical, mineralogical and geochemical parameters.

The analysis of these results was related to the source BMA spoil category to determine its

applicability, limitations, and potential improvements for categorisation of in situ spoil and mud.

Each sampled material was tested for its as-sampled gravimetric moisture content. The results show

that spoil above any previous flooding levels had very low moisture contents, ranging from 1.1 to

3.3%, associated with no recent rainfall. Spoil collected from below a previous flooding level just

below the surface had a higher moisture content; however, it was still below the materials plastic

limit. These results highlight spoil’s ability to retain moisture below the surface, even when visually

the conditions appear dry. This indicates that once a spoil pile is inundated with water, it may not

readily drain. This is significant for the modelling of lowwalls with respect to consideration of

appropriate strength mobilisation modes, and where to apply saturated or unsaturated material

parameters.

Large variability in as-sampled moisture content was observed for the mud samples, ranging from

below their plastic limits to well above their liquid limits. For both the spoil and the mud samples,

there was no clear relationship between the spoil category and as-sampled moisture content, with

results largely influenced by the sampling locations and weather conditions.

The specific gravity of each spoil sample was determined, showing a range from 2.2 to 2.7, with the

majority between 2.4 and 2.6. Low specific gravities were associated with materials that were

observed to have trace amounts of carbonaceous material. There was no clear relationship between

specific gravity and the spoil category.

Each of the sampled spoils had their total suction calculated. For spoil collected above any previous

flooding levels, values ranged between 44 MPa and 68 MPa. For C3M-10, a spoil collected below a

previous flooding level, an average total suction of 0.7 MPa was recorded. There was no relationship

between the spoil category and the total suction, with results reflective of the materials history post-

excavation.

For all spoil materials sampled, and dried mud samples, dry sieving was conducted to determine the

materials in situ particle size distribution scalped at -53 mm. A clear distinction between the spoil

categories was identified, with the representative Category 3 spoil specimens being the coarsest,

followed by the Category 2 spoil, and lastly the Category 1. These results are clearly reflected in the

D90 results graphed in Figure 5.8 for the spoil, and Figure 5.14 for the mud. Wet sieving of the

187

sampled spoil resulted in moderate to significant degradation, with the largest degree of reduction

observed in the Category 2 and 1 spoil, with both having very similar distributions post sieving.

All mud material particle size distributions were analysed via wet sieving, due to agglomeration

observed on drying. The smallest spread of results was from the Category 1 muds, showing the finest

distributions composed mostly of silt, with trace amounts of sand and gravel. Of the two Category 2

muds tested, a large variation was present of almost three orders of magnitude, highlighting the

potential variability that can be encountered in situ. The coarsest size distribution was found in the

Category 3 muds, with a range of up to two orders of magnitude, comprising a mixture of silt, sand

and gravel-sized particles. One Category 3 mud outlier was identified as C3M-18, having one of the

finest particle size distributions out of all the mud materials tested, composed almost entirely of silt.

The variation of particle size distributions from three mine sites was examined. Of key interest are

two pits composed of Category 3 materials that were shown to have vastly different spoil and mud

distributions, as depicted in Figure 5.18 and Figure 5.19 for C3S-13 and C3S-20 respectively. This

indicates the spoil category does not necessarily predict the degradation behaviour that can be

expected with flooding. This is not currently accounted for in the current BMA spoil category

framework.

The results of the particle size distribution testing show that in a dry state, the BMA spoil

classifications for the tested spoil was in agreeance with the determined size distributions. After wet

sieving, however, differentiation between the Category 2 and 1 spoil and mud materials was not

possible. Furthermore, large variations are evident in all material categories not predicted by the

materials original BMA spoil classification.

The pH, EC and TDS of each spoil and mud material was calculated. The pH values ranged from 7.8

to 9.3. The electrical conductivity results ranged from 246 to 8,920 µS/cm, with mud typically having

higher values than spoil. For the spoil materials, a trend of increasing electrical conductivity with

decreasing spoil category is observed. A comparison of pH against electrical conductivity showed a

moderately strong inverse relationship, with an r2 value of 0.544 shown in Figure 5.29.

The Atterberg limits and plasticity index of each material was determined. Plastic limits ranged from

12.1 to 27.1%. Liquid limits ranged from 21.5 to 61.9%. A reasonably strong relationship between

plastic and liquid limits was identified with an r2 value of 0.6971. All materials were classified as

clays of low plasticity, except for two mud samples classified as clays of high plasticity.

With respect to assigning a spoil category, 29% of the weighting is based on the material’s liquid

limit, with ranges of <20% for Category 4, 20 to 35% for Category 3, 35 to 50% for Category 2, and

>50% for Category 1 spoil. For this data set, aside from C3M-18 and C3M-23, all materials had liquid

188

limits within the range from 20 to 45%, with no clear relationship between the tested liquid limit and

the material’s assigned spoil category. There was also no clear distinction between the results of the

spoil and mud materials.

Each of the spoil materials sampled underwent Emerson class testing. For C3S-10 and C3S-13, both

were classified as Class 4 materials, indicating a moderate potential for erosion. The remaining spoil

specimens were classified as Class 2 materials due to the identification of slaking and dispersion,

indicating a high to very high potential for erosion. These results are significant, as it both identifies

spoil that has the potential to breakdown on exposure to water, and spoil that appears to be resistant.

To identify the mineralogical composition of all spoil and mud materials, as well as the geochemical

parameters of each, X-ray diffraction, cation exchange capacity, and exchangeable cations were

measured. The results show all tested specimens were dominated by Quartz, Kaolinite, Illite-Smectite

and Albite. Calculations were conducted to estimate the level of Smectite in each material, with

specimens of C3M-18, C1S-17, C1M-23 and C1M-32 returning the highest percentages. The lowest

levels of Smectite were present in C3S-10, C3S-13, C3M-04 and C2M-06. There was no clear

distinction between mineralogical composition and each spoil category. The cation exchange capacity

results were similar for the Category 3 and 2 materials (10 to 15 cmol(+)/kg, and higher in the

Category 1 materials (14 to 21 cmol(+)/kg). There was one Category 3 mud outlier (C3M-18) with a

value of 19 cmol(+)/kg.

With respect to exchangeable cations, the majority of specimens tested were dominated by

magnesium cations, followed by sodium and calcium. The least common cation was potassium. The

available cations were analysed with respect to the calculated Smectite percentage, with a focus on

the sodium cations due to their propensity to swell when hydrated. It was found that the highest levels

of sodium Smectite were found in the Category 1 materials, and the Category 3 outliers C3M-18 and

its associated spoil, C3M-20.

189

6 DEGRADATION TEST RESULTS

To understand how clay mineral-rich spoil degrades to mud in situ, several degradation tests were

conducted. Section 6.1 details the results of testing two Category 3 spoils under different durations

of saturation, and different numbers of wetting and drying cycles. The results from these two sections

were used to determine at what rate these spoil specimens degrade, and to what extent, as well as the

implications this has for clay mineral-rich spoil in situ.

The spoil samples tested were C3S-13, a 10-year old Category 3 spoil, and C3S-20, a 6-month old

Category 3 spoil. The conclusions from the extended saturation and wetting and drying cycles were

then used to conduct a study involving multiple degradation techniques on all spoil materials that had

not been influenced by in-pit flooding prior to sampling. These test results are discussed in

Section 6.2, with comparisons made to each of the spoil’s physical and chemical parameters.

Through identification of material characteristics related to degradation, improvements can be made

with respect to their categorisation, and their handling. The use of a modified slaked durability test to

identify quickly spoil prone to degradation has also been discussed. The spoil materials examined

were C3S-13, C3S-16, C3S-17, C3S-20 and C3S-24.

6.1 Wetting and Drying Cycles, and Prolonged Saturation

The influence of wetting and drying cycles, as well as prolonged saturation, were investigated on two

Category 3 spoils. C3S-13 and C3S-20 were chosen, as a large difference was observed in the mud

that was formed during flooding in situ, and through the results of wet sieving after 24 hours, as

discussed in Section 1425.1.4. For both materials, several representative 1.5 kg specimens scalped to

-19 mm were prepared with known initial particle size distributions obtained via dry sieving. Each of

the representative specimens was then exposed to a set duration of submersion in tap water or a known

number of wetting and drying cycles.

Figure 6.1 shows the laboratory setup, with the spoil material placed within trays and submerged in

tap water. To ensure full submersion, each specimen was filled with 52% of the spoil’s dry mass.

Each specimen was covered with clear plastic to avoid evaporation during periods of soaking, as


190

Figure 6.1 C3S-20 prolonged saturation preparation

Figure 6.2 Plastic-wrapped to avoid evaporation during submersion

6.1.1 Results of degradation testing of C3S-20

The C3S-20 representative specimens were subjected to seven increasing durations of soaking and

up to seven wetting and drying cycles. After degradation, each specimen was wet sieved without

dispersant for particles coarser than 75 m and analysed via hydrometer for particles finer than

75 m. The results of the prolonged saturation testing and the wetting and drying cycles for C3S-20

are plotted in Figure 6.3 and Figure 6.4, respectively, with details of D90, D50 and D10 recorded in

Table 6.1and Table 6.2.

191

Table 6.1 Degradation of C3S-20 subjected to prolonged soaking

CODE

SOAKING

DURATION

(days)

D90

(mm)

D50

(mm)

D10

(mm) Cu Cc

ELECTRICAL

CONDUCTIVITY

(µS/cm)

C3S-20 Dry sieve 14.7 8.61 0.388 27.1 3.995 1,190

C3S-20 1 11.3 1.27 0.011 295 0.463 2,050

C3S-20 2 11.6 1.29 0.007 417 0.405 2,620

C3S-20 4 11.3 1.06 0.008 348 0.309 2,870

C3S-20 8 11.7 1.09 0.006 461 0.325 3,060

C3S-20 16 11.1 0.531 0.005 408 0.226 3,620

C3S-20 33 8.64 0.396 0.005 301 0.068 3,960

C3S-20 64 9.99 0.271 0.007 242 0.085 4,370

Figure 6.3 Degradation of C3S-20 when subjected to prolonged saturation

Figure 6.3 shows the degree of degradation increases with the soaking duration. The greatest

reduction is observed within the first 24 hours, as weakly agglomerated sand and silt-sized particles

disperse. The largest increase is observed in the silt and sand-sized fractions. After the initial 24 hours

of soaking, the least amount of degradation is observed in the gravel-sized fraction. The Cu and Cc

values in Table 6.1classify all the tests as poorly graded.

The changes in C3S-20’s particle size distribution during numerous wetting and drying cycles are

plotted in Figure 6.4. The results show that one wetting and drying cycle produces more degradation

192

than just 24 hours of soaking, with a reduction in particle sizes across all fractions. The result of

multiple wetting and drying cycles produces up to a magnitude of reduction in the materials particle

size distribution. As with the 24 hours of soaking, negligible increases in clay content are observed.

The results of this testing show that increasing cycles of wetting and drying result in increasing

degrees of degradation in this material.

Table 6.2 Degradation of C3S-20 subjected to wetting and drying cycles

CODE

NUMBER OF

WETTING

AND DRYING

CYCLES

D90

(mm)

D50

(mm)

D10

(mm) Cu Cc

ELECTRICAL

CONDUCTIVITY

(µS/cm)

C3S-20 Dry sieve 14.7 8.61 0.388 27.1 4.00 1,190

C3S-20 24-hour soak 11.3 1.27 0.011 295 0.463 2,050

C3S-20 1 8.80 0.37 0.003 592 0.163 1,910

C3S-20 2 7.86 0.298 0.003 269 0.373 3,310

C3S-20 3 8.01 0.676 0.006 268 0.227 4,010

C3S-20 4 6.62 0.355 0.005 213 0.071 4,180

C3S-20 5 6.26 0.216 0.005 122 0.139 4,350

C3S-20 6 6.41 0.264 0.005 129 0.041 4,990

C3S-20 7 5.17 0.102 0.006 39.3 0.226 5,330

Figure 6.4 Degradation of C3S-20 when subjected to wetting and drying cycles

193

Figure 6.5 shows the electrical conductivity measurements taken during the prolonged saturation, and

the wetting and drying cycles. The analysis shows that two wetting and drying cycles are

approximately equivalent to 10 days of saturation. As with the reduction in particle size, the greatest

increase in electrical conductivity occurs within the first 24 hours of soaking. For the various

prolonged durations of soaking, the increases in conductivity begin to slow down after 30 days. For

the wetting and drying cycles, most increases occur within the first three cycles; however, after seven

cycles, large increases are still observed in association with the reduction in particle size plotted in

Figure 6.4.

Figure 6.6 illustrations a comparison between the two methods. The results show clearly that wetting

and drying cycles cause a faster reduction in particle size for the material. It is also important to

highlight the initial reduction in particle size during the first 24 hours of soaking. One wetting and

drying cycle is approximately equivalent to 16 days of saturation with respect to the resulting particle

sizes. The associated in-pit mud has also been plotted, comprised almost entirely of silt.

Figure 6.5 Electrical conductivity measurements of C3S-20 for soaking duration and wet/dry

cycles

194

Figure 6.6 Comparison of wetting and drying cycles versus saturation duration for C3S-20

6.1.2 Results of degradation testing of C3S-13

Representative specimens of C3S-13 were subjected to up to three wetting and drying cycles, and

three different durations of saturation to allow for contrast with C3S-20. The results of the prolonged

saturation testing and the wetting and drying cycles for C3S-13 are plotted in Figure 6.7 and

Figure 6.8, respectively, with the D90, D50, D10, Cc, Cu and the electrical conductivity values given in

Table 6.3 and Table 6.4 for the two cases.

Table 6.3 Degradation of C3S-13 subjected to prolonged soaking

CODE

SOAKING

DURATION

(days)

D90

(mm)

D50

(mm)

D10

(mm) Cu Cc

ELECTRICAL

CONDUCTIVITY

(µS/cm)

C3S-13 Dry Sieve 15.0 4.88 0.535 11.8 1.41 250

C3S-13 1 10.5 2.41 0.129 29.7 2.68 530

C3S-13 16 10.1 3.26 0.024 185 11.6 670

C3S-13 32 9.91 2.67 0.044 93.0 9.39 770

C3S-13 64 9.27 2.87 0.045 95.4 7.70 860

Figure 6.7 shows the degradation results of prolonged soaking on C3S-13. The results show the

largest reduction occurring during the first 24 hours of soaking. After the first 24 hours, negligible

further degradation occurred in the sand and gravel-sized fractions. An increase is observed in the

silt-sized fraction, with approximately 5% more silt-sized particles after 64 days.

195

Figure 6.7 Degradation of C3S-13 when subjected to prolonged soaking

Table 6.4 Degradation of C3S-13 subjected to wetting and drying cycles

CODE

NUMBER OF

WETTING

AND DRYING

CYCLES

D90

(mm)

D50

(mm)

D10

(mm) Cu Cc

ELECTRICAL

CONDUCTIVITY

(µS/cm)

C1S-13 Dry Sieve 15.0 4.88 0.535 11.8 1.41 250

C1S-13 24-hour soak 10.5 2.41 0.129 29.7 2.68 530

C1S-13 3 9.33 2.24 0.030 117 5.21 810

C1S-13 5 8.86 2.38 0.015 240 6.39 1,125

C1S-13 7 7.34 1.93 0.028 102 3.92 2,570

The influence of wetting and drying cycles on C3S-13 is shown in Figure 6.8. The particle size after

24 hours of soaking has also been included. The largest degree of degradation occurs with the first

wetting and drying cycle, which results in larger sand and silt-sized fractions than after 24 hours of

soaking. After exposure to up to seven wetting and drying cycles, negligible degradation is observed,

highlighting the resistance of this material to slaking.

The electrical conductivity readings for both the prolonged soaking and the wetting and drying cycles

are plotted in Figure 6.9. For prolonged saturation, the largest increase in conductivity is observed

after the first 24 hours and reducing at a rapid rate with increased time intervals. The wetting and

196

drying cycles, however, show a significant increase after the first cycle, equivalent to 46 days of

soaking. With each cycle, relatively large increases in conductivity occur.

Figure 6.10 shows the progression of C3S-13 from fresh spoil to in situ, in-pit mud, contrasting the

difference between prolonged saturation and wetting and drying cycles. The wetting and drying

cycles show higher degrees of degradation than prolonged soaking, with one wetting and drying cycle

causing a larger reduction than 64 days of soaking in all size fractions. In comparison to the associated

in-pit mud (C3M-08 and C3M-12), the results show that prolonged soaking resulted in a similar

degree of silt-sized fractions. Wetting and drying cycles produced a larger number of fines than found

in the in-pit mud. This could be due to the in-pit mud being fully submerged up to the sampling date,

or due to fines being washed away from the sampling location. It is also possible that the energy input

of the 60oC oven is greater than what is experienced in situ.

Figure 6.8 Degradation of spoil sample C3S-13 subjected to wetting and drying cycles

197

Figure 6.9 Electrical conductivity measurements of C3S-13 for saturation duration and

wet/dry cycles

Figure 6.10 Comparison of wetting and drying cycles versus soaking duration for C3S-13

198

6.1.3 Conclusions of prolonged saturation and wetting and drying cycle degradation testing

Prolonged saturation and wetting and drying cycles were applied to investigate how clay mineral-rich

spoil is likely to behave when exposed to flooding, and rainfall events. The results have shown that

wetting and drying cycles are more effective at breaking down weakly cemented material promoting

the slaking of larger-sized fractions within the spoil. The same effects are observed to a lesser degree

with prolonged saturation. For both materials tested, the largest reduction was observed within

24 hours of soaking. This is due to weakly cemented fines dispersing and breaking apart.

For C3S-20, prolonged durations of soaking resulted in increased reductions in particle size. This was

not observed in the testing of C3S-13, with negligible changes occurring after the first 24 hours of

soaking. With respect to wetting and drying cycles, C3S-20 continued to degrade with each cycle.

For C3S-13, most of the degradation occurred after the first cycle, with increasing numbers of cycles

resulting in slightly higher amounts of sand and silt, with little to no change in the gravel-sized

fraction.

For each cycle or duration of soaking, the electrical conductivity was measured. Increases in

conductivity occurred quickly at the start of prolonged soaking but slowed down significantly after

the first 24 hours for both materials. With each wetting and drying cycle, a large increase was

observed in the measured conductivity for both materials.

From these results, it is observed that C3S-20 is far more prone to degradation via slaking. Fresh spoil

that has recently been exposed is more susceptible to degradation due to a change in stresses and

introduction to surface weathering. As it is ten years old, it is likely C3S-13 has already undergone

some degree of degradation in situ prior to sampling and was originally a coarser material. The

contrast between the degradation testing of the two spoil materials is highlighted in Figure 6.11,

showing the range of degradation results, the original spoil, and the in situ mud particle size

distributions. The wetting and drying cycles and prolonged saturation are shown with brown and blue

shading, respectively.

To simulate potential in situ degradation, wetting and drying cycles are more effective than prolonged

soaking for promoting slaking of the spoil, and the time requirement of the test is greatly reduced,

allowing for spoil prone to slaking to be identified at a rapid rate.

199

Figure 6.11 Comparison of C3S-13 and C3S-20 during prolonged soaking and wetting and

drying cycles

200

6.2 Accelerated Degradation and Modified Slake Durability Testing of Spoil

Three methods were utilised to promote the degradation of the sampled mine spoil within a laboratory

setting to predict potential degradation in situ. The purpose of this testing was to identify if the spoil’s

BMA category could reflect the degree of degradation it underwent.

6.2.1 Testing methodology and sampling

Dry sieving was used to determine the in situ particle size of each of the spoil materials collected.

Representative samples were then exposed to three degradation methods. The first method involved

submerging the samples in tap water for 24 hours, followed by wet sieving. Wetting and drying cycles

were used to promote slaking of the materials, and to break up any weakly cemented fines,

representative of cycling flooding or large rainfall events. Lastly, a modified slake durability test was

used as a comparison with three wetting and drying cycles to determine if it was a suitable method

for rapidly and cheaply identifying material prone to slaking.

Photos of the spoil source are included in Figure 6.12 to Figure 6.16, with each photo containing

either 20-cent coins (2.9 cm) or disposable plates (15 cm). Two Category 3 materials, two Category

2 materials, and one Category 1 material was selected and analysed to highlight variability within and

between the categories.

Figure 6.12 Spoil material C3S-13

201



202



203

6.2.2 Physical characterisation of spoil

The Atterberg limits and specific gravities of these spoil samples are summarised in Table 6.5 and

Figure 6.17. The as-sampled gravimetric moisture content ranged from 1.1 to 3.3%, with no

significant variation in moisture content between the spoil categories. The liquid limits ranged from

21.5 to 40.3%. The plastic limits varied from 12.1 to 19.4%, with the highest values being for the

clay mineral-rich Category 3 spoil materials. The plasticity index values ranged from 7.25 to 20.9%,

showing no strong relationship with spoil category. To further interpret the Atterberg limit data, a

plasticity chart is shown in Figure 6.18, which classifies all spoils as clays of low plasticity, with

C3S-20 and C1S-17 having a higher plasticity index and higher plasticity. The specific gravities

covered a narrow range from 2.57 to 2.68.

Table 6.5 Spoil physical characterisation

CODE SOURCE

CATEGORY TYPE

PLASTIC

LIMIT (%)

LIQUID

LIMIT (%)

PLASTICITY

INDEX (%)

SPECIFIC

GRAVITY

C3S-13 CAT 3 SPOIL 19.4 26.6 7.25 2.61

C3S-20 CAT 3 SPOIL 19.4 40.3 20.9 2.57

C2S-16 CAT 2 SPOIL 12.1 21.5 9.4 2.67

C2S-24 CAT 2 SPOIL 18.7 39.4 20.7 2.68

C1S-17 CAT 1 SPOIL 14.4 21.9 7.5 2.62

Figure 6.17 As-sampled moisture content and Atterberg limits

204

Figure 6.18 Plasticity index versus liquid limit for all spoil

6.2.3 Degradation testing results

The particle size distribution curves for each spoil tested under the four different conditions are

plotted in Figure 6.19 to Figure 6.23. The sizes through which 90%, 50% and 10% of the particles

pass (D90, D50 and D10) are summarised as seen in Table 6.6.

As seen in Figure 6.19 to Figure 6.23 and Table 6.6, for a given test condition the particle size

distributions were coarsest for the Category 3 spoil and generally finest for the Category 1 spoil, with

the Category 2 spoils generally intermediate. An exception was the dry sieving of Category 2 spoil

C2S-24, which was slightly finer-grained than the dry sieving of Category 1 spoil C1S-17. This is

due to the higher Illite-Smectite content of C2S-24 compared with that of spoil C1S-17. The two

Category 3 spoils had similar dry particle size distributions, but spoil C3S-20 degraded far more than

C3S-13. The two Category 2 spoils had distinctly different dry particle size distributions, but similar

degraded particle size distributions.

The least degradation, of less than half an order of magnitude in particle size, was seen for Category

3 spoil C3S-13. The highest degree of degradation, of about two orders of magnitude in particle size,

was seen for Category 2 spoil C2S-16. Category 1 spoil C3S-20 showed about 1.5 orders of

magnitude of degradation, and both Category 2 spoil C2S-24 and Category 1 spoil C1S-17 showed

about one order of magnitude of degradation. Generally, for all spoil samples, degradation was least

205

after 24-hour soaking; most after modified slake durability testing, and intermediate after three

wetting and drying cycles. However, these differences were less than the difference between the dry

and degraded particle size distributions.

Table 6.6 D90, D50 and D10 before and after degradation testing of fresh spoil

CODE SOURCE

CATEGORY TYPE D90 (mm) D50 (mm) D10 (mm)

Dry Sieving

C3S-13 CAT 3 SPOIL 15.0 4.88 0.535

C3S-20 CAT 3 SPOIL 16.4 7.08 0.394

C3S-16 CAT 2 SPOIL 12.8 4.18 0.234

C3S-24 CAT 2 SPOIL 7.42 0.561 0.114

C3S-17 CAT 1 SPOIL 8. 017 0.617 0.105

24-hour Soaking

C3S-13 CAT 3 SPOIL 10.5 2.41 0.129

C3S-20 CAT 3 SPOIL 10.8 0.832 0.005

C3S-16 CAT 2 SPOIL 2.0 0.176 0.009

C3S-24 CAT 2 SPOIL 17.4 4.64 3.077

C3S-17 CAT 1 SPOIL 0.750 0.170 0.015

3 Wetting and Drying Cycles

C3S-13 CAT 3 SPOIL 9.48 2.20 0.024

C3S-20 CAT 3 SPOIL 13.8 0.374 0.003

C3S-16 CAT 2 SPOIL 4.58 0.082 0.001

C3S-24 CAT 2 SPOIL 1.57 0.051 0.003

C3S-17 CAT 1 SPOIL 0.433 0.061 0.007

Modified Slake Durability Testing

C3S-13 CAT 3 SPOIL 9.46 1.72 0.010

C3S-20 CAT 3 SPOIL 9.37 0.378 0.003

C3S-16 CAT 2 SPOIL 1.84 0.054 0.003

C3S-24 CAT 2 SPOIL 1.17 0.152 0.003

C3S-17 CAT 1 SPOIL 0.577 0.095 0.002

206

Figure 6.19 Particle size distribution of C3S-13 before and after degradation

Figure 6.20 Particle size distribution of C3M-20 before and after degradation

207



208


6.2.4 Discussion of degradation results

The results of the mineralogical characterisation testing using XRD analysis are summarised in

Table 6.7, and the results of the exchangeable cation and cation exchange capacity (CEC) testing are

summarised in Table 6.8

The spoil samples are dominated by Quartz (28 to 60%), Kaolinite (6 to 16%), Illite-Smectite (21 to

44%) and Albite (4 to 13%). The highest concentration of Illite-Smectite is present in the Category 3

spoils. The lowest levels of Illite-Smectite are seen in C2S-16 and C1S-17, both of which are

dominated by quartz (52% and 60%, respectively). Trace amounts of numerous other clays and

minerals are present in the other materials, ranging from 0 to 9%.

There is no clear distinction between exchangeable cations, CEC or mineralogical composition with

respect to the BMA spoil categories associated with each material. The Category 3 spoils contain the

highest quantities of Illite-Smectite. The Category 1 spoil contains the highest Quartz content and a

relatively low Illite-Smectite content.

209

Table 6.7 X-ray diffraction analysis for all spoil samples

CODE

QU

AR

TZ

KA

OL

INIT

E

CA

LC

ITE

AR

AG

ON

ITE

DO

LO

MIT

E

SID

ER

ITE

ILL

ITE

-

SM

EC

TIT

E

MIC

A/I

LL

ITE

CH

LO

RIT

E

AN

AT

AS

E

PY

RIT

E

AL

BIT

E

OR

TH

OC

LA

SE

HA

LIT

E

GE

OT

HIT

E

C3S-13 35 6 0 0 1 2 40 5 0 <1 <1 11 0 0 0

C3S-20 28 11 0 0 4 1 44 1 1 1 0 9 0 <1 0

C2S-16 52 13 0 0 2 0 23 1 0 1 0 4 0 0 4

C2S-24 29 16 <1 0 9 4 28 <1 0 <1 0 13 0 0 0

C1S-17 60 6 2 0 0 0 21 2 0 <1 0 6 2 <1 0

Table 6.8 Exchangeable cations and cation exchange capacity of all spoil samples

CODE SOURCE

CATEGORY TYPE

EXCHANGEABLE CATIONS AT pH 8.5

(cmol(+)/kg)

CATION

EXCHANGE

CAPACITY

(NH4+) Ca2+ Mg2+ Na+ K+

Total

Cations

C3S-13 CAT 3 SPOIL 3.6 7.1 0.7 1.2 13 13

C3S-20 CAT 3 SPOIL 4.0 4.7 4.9 0.6 14 15

C2S-16 CAT 2 SPOIL 1.2 6.6 4.6 0.2 13 11

C2S-24 CAT 2 SPOIL 2.9 5.7 4.6 0.7 14 13

C1S-17 CAT 1 SPOIL 3.2 4.9 6.9 0.4 15 14

The most common cations present are magnesium and sodium. C3S-13 has noticeably lower levels

of exchangeable sodium. All materials had low levels of potassium cations. C1S-17 had the highest

sodium cation content. The CEC for all the materials range from 11 to 15 cmol+/kg, with no clear

differentiation between the material categories.

To explain the large variations in the degradation results obtained for the five spoil materials tested,

Figure 6.24 shows the cumulative area under the particle size distribution curves divided by the total

area of the curves for each specimen and each degradation method. The larger the area ratio, the finer

the overall particle size distribution. For all materials, the coarsest particle size distribution was

obtained via dry sieving. Submersion in water for 24 hours resulted in a substantial reduction in the

proportion of coarse-grained particles, with wetting and drying cycles causing further degradation of

coarse-grained particles. The modified slake durability testing had similar results to the three wetting

and drying cycles, and in some cases, slightly greater degradation. This indicates the presence of

material prone to slaking, and varying degrees of cementation within the agglomerated particles.

210

There was no significant difference between the as-sampled moisture contents or specific gravities of

the spoil samples tested. The spoil samples were all collected on the same day, under dry conditions,

from two mines within close proximity, so the consistently low as-sampled moisture content would

be expected. The similar specific gravities are consistent with normal mineral matter containing a low

carbonaceous content. However, there were significant differences between the Atterberg limits of

the samples, with two samples plotting higher on the Plasticity chart.

Figure 6.25 provides further insight into the behaviour of these spoil materials as they degrade. The

plot shows the percentage of mass retained within the drum during the modified slake durability test

for each wetting and drying cycle. For the two Category 3 spoils, the greatest amount of material is

retained in the drum initially as predicted by the materials dry sieving particle size distribution. With

each cycle, a steadily reducing amount of material can be seen to be retained. C3S-20 shows a

reduction of material at a similar rate to C3S-13. It does, however, lose a larger amount after the first

cycle, indicating weakly cemented material breaking apart, or rapid slaking.

C2S-16 shows a significant decrease in retained material after the first cycle, similar to its reduction

in particle size after 24 hours of soaking. C2S-24 shows most of the specimen passes through the

2 mm drum before the initial wetting and drying cycle. A large reduction is shown to occur after the

first cycle, with a steady decrease observed for the following cycles. C1S-17 shows similar results to

C2S-24, with most of the specimen passing through the 2 mm drum before the first cycle. After the

first cycle, all the material was seen to pass through, apart from one large piece of gravel, which

remained and did not slake during the second and third cycles.

211

Figure 6.24 Influence of degradation method on particle size reduction

Figure 6.25 Modified slake durability degradation analysis per cycle

212

The key XRD and CEC results are given in Table 6.9, showing the dominant components for all spoil

samples to be Quartz, Kaolinite and Illite-Smectite. For each material, the CEC will be dictated by

the Kaolinite and Illite-Smectite components. From the XRD analysis, it is not possible to determine

to what degree the Smectite has turned into Illite. It is, however, important to understand this

relationship, since Smectite can be highly reactive to water, and can cause a large degree of shrinkage

and stiffening on drying and swelling and slaking on wetting. For this study, it was assumed that the

CEC of Illite, Smectite and Kaolinite are 20 cmol+/kg, 100 cmol+/kg and ten cmol+/kg, respectively.

Using this relationship, the measured CEC and the known percentage of Illite-Smectite, the Illite

percentage was estimated from Equation (1).

Table 6.9 Key mineralogical and geochemical characteristics of spoil samples

CO

DE

SO

UR

CE

CA

TE

GO

RY

TY

PE

QU

AR

TZ

(%

)

KA

OL

INIT

E (

%)

ILL

ITE

-

SM

EC

TIT

E (

%)

CE

C (

cmo

l+/k

g)

CA

LC

UL

AT

ED

ILL

ITE

(%

)

CA

LC

UL

AT

ED

SM

EC

TIT

E (

%)

CA

LC

UL

AT

ED

SO

DIU

M

SM

EC

TIT

E (

%)

C3S-13 CAT 3 SPOIL 35 6 40 13 35 5.5 0.3

C3S-20 CAT 3 SPOIL 28 11 44 15 38 6.4 2.2

C3S-16 CAT 2 SPOIL 52 13 23 11 17 6.4 2.3

C3S-24 CAT 2 SPOIL 29 16 28 13 21 7.3 2.4

C3S-17 CAT 1 SPOIL 60 6 21 14 9.5 12 5.2

From the estimated percentage of Smectite, the measured sodium cations were used to estimate the

overall percentage of sodium Smectite within each sample. Sodium Smectite is the most reactive of

all clay minerals due to Smectite being the most moisture-reactive clay mineral, and the mono-valent

sodium cation being weakly bonded.

Both Category 3 spoil materials have high levels of Illite-Smectite, with C3S-13 having an estimated

0.3% sodium Smectite and C3S-20 having an estimated 2.2% sodium Smectite. This is likely a

significant contributing factor to C3S-20 having a liquid limit 13.7% higher, and the higher degree of

degradation observed in comparison to that of C3S-13. The two Category 2 spoil materials have

similar and significant estimated percentages of sodium Smectite, explaining their substantial degrees

of degradation. C1S-17 had the highest estimated percentage of sodium Smectite at 5.2%, explaining

its highest degradation among the spoil materials tested.

The as-sampled moisture content and the specific gravities of the spoils tested were similar and would

not have influenced the degradation behaviour on wetting-up. The degradation test results indicated

213

high variability in the degree of degradation on wetting-up, which could not be attributed solely to

the assigned spoil category.

6.2.5 Accelerated degradation conclusions

While degradation on wetting-up tends to be more pronounced the lower the spoil Category, other

factors also contribute. These include, in order of decreasing importance, the clay mineralogy of the

spoil material, particularly the presence and percentage of sodium Smectite, the exchangeable cations

and CEC, and the liquid limit.

The higher the sodium Smectite percentage, the greater the observed degradation on wetting-up.

Similarly, the greater the percentage of the monovalent cation sodium, and the larger the CEC, the

greater the observed degradation is on wetting-up. Sodium Smectite is associated with swelling and

dispersion, which is reflected in the liquid limits, with material more susceptible to degradation also

having higher liquid limits.

The identification of the presence of sodium Smectite in spoil and its proportion is, therefore, a critical

component to identifying its potential to degrade on wetting-up, indicative of the quality of in-pit

mud that spoil may turn into.

The results of this testing also highlight the potential for the modified slake durability test to be used

to identify spoil prone to degradation, and hence whether the resulting mud needs to be removed prior

to subsequent spoiling or whether it can be safely left in place. Correct identification and handling of

material prone to slaking will result in large cost savings. The testing method is simple and repeatable,

allowing for any technician to conduct the testing with adequate training.

214

7 CONSOLIDATION TEST RESULTS

The results of the consolidation testing are discussed in Sections 7.1 and 7.2. Section 7.1 details the

oedometer results for all spoil and mud materials sampled in a 76 mm diameter by 20 mm deep 1 MPa

device as detailed in Section 4.10.1. Section 7.2 discusses the results of the large slurry

consolidometer testing conducted on C3M-08, C3M-18, C1M-02 and C1M-23, with the methodology

detailed in Section 4.10.2.

7.1 Standard Consolidometer Results

The results of standard oedometer testing of spoil and mud materials are detailed in this section. Due

to the large number of tests, only summaries of key results are provided and discussed. For the

oedometer testing, all test materials were sieved to -4.7 mm and placed loose at the as-sampled

gravimetric moisture content. Staged stress increments of 28 kPa, 50 kPa, 100 kPa, 200 kPa, 500 kPa

and 1,000 kPa were applied.

Table 7.1 and Table 7.2 give the initial and final conditions for the oedometer testing of all spoil and

mud specimens, respectively. Average values for each category and material source are included.

Table 7.1 Initial and final conditions for all spoil specimens tested

CO

DE

INIT

IAL

GR

AV

IME

TR

IC

MO

IST

UR

E

CO

NT

EN

T (

%)

FIN

AL

GR

AV

IME

TR

IC

MO

IST

UR

E

CO

NT

EN

T (

%)

INIT

IAL

DR

Y

DE

NS

ITY

(t/

m3)

FIN

AL

DR

Y

DE

NS

ITY

(t/

m3)

INIT

IAL

VO

ID

RA

TIO

FIN

AL

VO

ID

RA

TIO

C3S-10 9.5 9.7 1.31 1.97 0.86 0.24

C3S-13 1.1 11.4 1.35 2.01 0.94 0.30

C3S-20 2.9 12.6 1.47 1.94 0.76 0.32

Averages 4.5 11.2 1.37 1.97 0.85 0.29

C2S-16 3.3 14.4 1.48 1.93 0.80 0.38

C2S-24 2.4 14.4 1.44 1.94 0.86 0.39

Averages 2.8 14.4 1.46 1.93 0.83 0.39

C1S-17 2.7 13.3 1.37 1.94 0.91 0.35

Most of the spoil materials had a gravimetric dry based moisture content below 3.3%, with C3S-10

as the outlier due to being below a previous flooding level. When loosely placed, the initial dry

densities ranged from 1.3 to 1.5 t/m3, with initial void ratios from 0.76 and 0.94. Post-consolidation,

dry densities ranged from 1.93 to 2.01 t/m3, with the Category 3 spoil having slightly higher dry

densities than the Category 2 or Category 1 spoils. A large variation was observed in the final void

215

ratios, with the lowest values associated with the Category 3 spoil, followed by Category 2 and 1

spoils. The Category 2 and 1 spoils also had slightly higher final gravimetric moisture contents.

Table 7.2 Initial and final conditions for all mud specimens tested

CO

DE

INIT

IAL

GR

AV

IME

TR

IC

MO

IST

UR

E

CO

NT

EN

T (

%)

FIN

AL

GR

AV

IME

TR

IC

MO

IST

UR

E

CO

NT

EN

T (

%)

INIT

IAL

DR

Y

DE

NS

ITY

(t/

m3)

FIN

AL

DR

Y

DE

NS

ITY

(t/

m3)

INIT

IAL

VO

ID

RA

TIO

FIN

AL

VO

ID

RA

TIO

C3M-01 34.0 12.5 1.41 1.94 0.82 0.32

C3M-03 24.0 15.8 1.46 1.79 0.70 0.39

C3M-04 13.3 13.8 1.58 1.89 0.63 0.36

C3M-05 13.3 13.5 1.55 1.91 0.67 0.35

C3M-08 22.1 14.1 1.49 1.70 0.50 0.31

C3M-12 20.2 15.7 1.53 1.65 0.46 0.35

C3M-30 17.7 11.6 1.67 1.97 0.53 0.30

C3M-18 94.5 17.9 0.74 1.68 2.25 0.43

Averages 29.9 14.4 1.43 1.82 0.82 0.35

C2M-06 12.9 15.7 1.28 1.90 1.11 0.43

C2M-07 4.5 13. 5 1.45 1.92 0.79 0.35

Averages 8.7 14.6 1.37 1.91 0.95 0.39

C1M-02 46.1 15.8 1.24 1.82 1.07 0.40

C1M-23 113.8 26.3 0.65 1.54 3.00 0.68

C1M-32 5.6 18.5 1.09 1.58 1.05 0.41

Averages 55.2 20.2 0.99 1.65 1.70 0.50

The initial conditions of the tested mud materials had a large range due to the variation found in situ.

C3M-18 and C3M-23 both had very high initial and final moisture contents. For these specimens, a

very low initial dry density was recorded, with values of 0.74 t/m3 and 0.65 t/m3, respectively. On

average, the Category 3 mud had higher initial densities than the Category 2 mud, which in turn were

higher than the Category 1 mud. The measured final densities and moisture contents for the Category

2 and Category 3 muds were similar, however the range within the Category 3 materials was larger,

with values from 1.54 to 1.97 t/m3. The Category 1 mud on average had lower final densities and

higher moisture contents. With respect to final void ratios, the lowest value was obtained with the

Category 3 mud, followed by Category 2, and lastly Category 1.

216

Figure 7.1 shows the final dry densities for all spoil and mud specimens. The results show that on

average, the highest final dry densities are achieved with the spoil specimens, however similar values

are achieved with some of the Category 3 and Category 2 muds. Figure 7.2 shows the void ratios for

all spoil and mud specimens before loading, and at 1,000 kPa. A linear trendline was fitted to the data

with an r2 value of 0.69, indicating a strong relationship between the two variables.

Figure 7.1 Final dry density for all spoil and mud specimens tested

217

Figure 7.2 Initial and final void ratio for all spoil and mud specimens tested

The oedometer results are given in Table 7.3 and are plotted in Figure 7.3 for the spoil samples tested,

and in Table 7.4 and Figure 7.4 for the mud samples tested. The results are provided in terms of

cumulative settlement as a percentage observed at the end of each loading stage, with averages for

each category, and material source included.

Table 7.3 Settlement for all spoil specimens tested

CODE SETTLEMENT % UNDER APPLIED STRESS

28 kPa 50 kPa 100 kPa 200 kPa 500 kPa 1,000 kPa

C3S-10 18.1 21.2 24.2 27.1 30.7 33.4

C3S-13 8.9 13.0 17.7 22.6 28.5 33.0

C3S-20 7.9 10.2 13.7 17.3 21.7 24.6

Averages 11.6 14.8 18.6 22.3 27.0 30.3

C2S-16 7.0 11.2 15.1 17.8 20.6 23.1

C2S-24 7.6 10.4 12.8 16.7 22.2 25.4

Averages 7.3 10.8 14.0 17.3 21.4 24.2

C1S-17 11.0 14.4 18.7 22.4 26.5 29.4

There is no clear distinction between spoil category and settlement percentage. Of the spoil specimens

tested, a small range of 10% is observed between the results, with the largest initial settlement related

to C3S-10.

218

Figure 7.3 Settlement for -4.7 mm all loose-placed spoil specimens tested wet

Table 7.4 Settlement for all mud specimens tested

CODE SETTLEMENT % UNDER APPLIED STRESS


C3M-01 12.2 15.3 18.0 20.8 24.5 27.2

C3M-03 5.3 7.2 9.7 12.1 15.5 18.1

C3M-04 3.8 5.8 8.1 10.7 14.0 16.8

C3M-05 4.7 6.9 9.7 12.6 16.3 19.1

C3M-08 4.3 5.3 6.6 8.1 10.3 12.3

C3M-12 1.1 1.7 2.6 3.8 5.6 7.4

C3M-30 6.7 7.5 8.9 10.6 13.0 15.1

C3M-18 31.9 37.5 42.8 47.5 52.8 56.0

Averages 8.7 10.9 13.3 15.8 19.0 21.5

C2M-06 13.4 16.4 19.1 22.6 28.8 32.6

C2M-07 8.1 11.0 14.5 17.7 21.7 24.5

Averages 10.7 13.7 16.8 20.2 25.3 28.5

C1M-02 15.4 18.2 21.4 24.6 28.8 32.0

C1M-23 35.4 39.9 47.3 51.6 56.2 57.9

C1M-32 12.8 15.1 19.0 22.9 27.7 31.1

Averages 21.2 24.4 29.2 33.0 37.6 40.3

219

Figure 7.4 shows the % settlement results for all mud samples tested. C3M-18 and C3M-23 are both

observed to be significant outliers, showing a significant amount of settlement because of their high

initial gravimetric moisture contents (94.5% and 113.8%, respectively). A trend can be observed,

with the Category 3 mud samples typically showing the least settlement during consolidation, and at

the final applied stress of 1,000 kPa. Higher degrees of settlement are seen for the Category 2 and

Category 1 mud samples.

The settlement of the spoil and mud specimens is compared in Figure 7.5. A larger spread of results

is observed for the mud samples tested, with several of the Category 3 mud samples tested showing

less settlement than all of the spoil samples tested. Figure 7.6 compares the % settlement to the initial

dry density. The results show a strong relationship with an r2 value of 0.87.

Figure 7.4 Settlement for -4.7 mm all loose-placed mud specimens tested wet

220

Figure 7.5 Comparison of settlement for -4.7 mm spoil and mud specimens tested wet

Figure 7.6 Initial dry density versus settlement at 1,000 kPa stress for all spoil and mud

specimens tested

221

For each spoil and mud specimen, the final void ratios calculated at the end of each loading step are

shown in Table 7.5, as well as the calculated compression index. The void ratio versus the applied

stress is plotted in Figure 7.7 and Figure 7.8 for the spoil and mud, respectively. The compression

index for all materials is plotted in Figure 7.9. The compression index was calculated using:

𝐶𝑐 = (𝑒1 − 𝑒0)/(log10(2′ − ′

1)) (4)

where:

Cc = compression index;

e1/e2 = the void ratio calculated at a linear section; and

σ’2/σ’1 = the effective stress at a linear section.

Table 7.5 Final void ratio and compression index values for all spoil and mud specimens

tested

CODE VOID RATIO AT APPLIED STRESSES

COMPRESSION

INDEX

28 kPa 50 kPa 100 kPa 200 kPa 500 kPa 1,000 kPa Cc r2

C3S-10 0.52 0.46 0.41 0.35 0.29 0.24 0.078 0.998

C3S-13 0.76 0.68 0.59 0.50 0.38 0.30 0.130 1.000

C3S-20 0.62 0.58 0.51 0.45 0.37 0.32 0.084 0.999

Averages 0.63 0.57 0.50 0.43 0.35 0.29 0.097 0.999

C3M-01 0.59 0.54 0.49 0.44 0.37 0.32 0.075 0.998

C3M-03 0.61 0.58 0.54 0.49 0.44 0.39 0.061 1.000

C3M-04 0.57 0.53 0.50 0.45 0.40 0.35 0.059 0.999

C3M-05 0.59 0.55 0.50 0.46 0.39 0.35 0.067 1.000

C3M-08 0.43 0.42 0.40 0.38 0.34 0.31 0.033 0.994

C3M-12 0.44 0.43 0.42 0.40 0.38 0.35 0.025 0.978

C3M-30 0.43 0.41 0.39 0.37 0.33 0.30 0.036 0.988

C3M-18 1.21 1.03 0.86 0.71 0.54 0.43 0.216 0.988

Averages 0.61 0.56 0.51 0.46 0.40 0.35 0.072 0.993

C2S-16 0.67 0.60 0.53 0.48 0.43 0.38 0.078 0.972

C2S-24 0.72 0.67 0.62 0.55 0.45 0.39 0.094 0.994

Averages 0.70 0.63 0.57 0.51 0.44 0.39 0.086 0.983

C2M-06 0.83 0.77 0.71 0.64 0.50 0.43 0.113 0.993

C2M-07 0.64 0.59 0.53 0.47 0.40 0.35 0.082 0.998

Averages 0.74 0.68 0.62 0.55 0.45 0.39 0.098 0.996

C1S-17 0.70 0.64 0.55 0.48 0.40 0.35 0.099 0.994

C1M-02 0.70 0.64 0.55 0.48 0.40 0.35 0.099 0.994

C1M-23 0.75 0.69 0.62 0.56 0.47 0.40 0.095 1.000

222

C1M-32 1.59 1.40 1.11 0.94 0.75 0.68 0.258 0.960

Averages 0.79 0.74 0.66 0.58 0.48 0.41 0.107 0.999

Figure 7.7 Applied stress versus void ratio for all spoil specimens tested

Figure 7.8 Applied stress versus void ratio for all mud specimens tested

223

Figure 7.9 Compression index values for all spoil and mud specimens tested

From the results of the settlement data obtained during testing, the coefficient of consolidation cv was

calculated for each specimen via graphical construction using the log time method AS 1289.6.6.1

(1998). An example of this calculation is shown in Figure 7.10, plotting the settlement against time.

Analysis of the data allows for the identification of primary and secondary consolidation, from which

the coefficient of consolidation cv is calculated according to:

cv =

0.026H2

t50 (5)

where

cv = coefficient of consolidation;

H = average thickness of specimen for the load increment, in millimetres; and

t50 = time for 50% primary consolidation, in minutes.

224

The coefficient of consolidation values of all materials is given for each loading increment in

Table 7.6. From Figure 7.7 and Figure 7.8 the coefficients of volume change were calculated using:

𝑚v =

∆𝑒

∆’∗

1

1 + 𝑒 (6)

where:

mv = coefficient of volume change (m2/kN);

∆e = change in the void ratio of the specimen during before and after loading;

∆σ’ = increase in the pressure, in kilopascals, above the present overburden pressure; and

e = void ratio of the laboratory specimen at the end of the loading stage.

Table 7.6 Coefficient of Consolidation values for all spoil and mud specimens tested

CODE COEFFICIENT OF CONSOLIDATION, cv, AT APPLIED STRESSES


C3S-10 8.04x10-03 4.71x10-03 4.76x10-03 8.06x10-03 4.03x10-03 1.01x10-02

C3S-13 5.36x10-02 4.99x10-02 3.93x10-02 1.92x10-02 8.98x10-03 3.36x10-03

C3S-20 2.67x10-03 1.04x10-02 6.74x10-03 4.48x10-03 4.07x10-03 3.93x10-03

Averages 2.14x10-02 2.17x10-02 1.69x10-02 1.06x10-02 5.69x10-03 5.80x10-03

C3M-01 5.93x10-04 1.48x10-03 4.41x10-03 7.38x10-03 5.49x10-03 3.34x10-03

C3M-03 3.91x10-03 4.08x10-03 5.18x10-03 6.31x10-03 7.51x10-03 6.42x10-03

C3M-04 6.45x10-03 5.00x10-03 8.12x10-03 7.00x10-03 9.00x10-03 8.38x10-03

C3M-05 2.32x10-03 2.74x10-03 3.57x10-03 4.74x10-03 5.74x10-03 5.32x10-03

C3M-08 1.12x10-02 1.59x10-02 1.33x10-02 1.13x10-02 8.66x10-03 1.18x10-02

C3M-12 2.60x10-02 7.85x10-03 1.12x10-02 2.46x10-02 1.59x10-02 2.30x10-02

C3M-30 3.45x10-02 3.18x10-02 1.87x10-02 1.80x10-02 2.87x10-02 2.72x10-02

C3M-18 4.76x10-04 1.93x10-04 3.04x10-04 3.83x10-04 6.34x10-04 6.19x10-04

Averages 1.07x10-02 8.63x10-03 8.09x10-03 9.97x10-03 1.02x10-02 1.08x10-02

C2S-16 5.39x10-04 9.93x10-03 1.16x10-02 2.51x10-02 1.76x10-02 3.14x10-03

C2S-24 7.48x10-04 2.68x10-03 3.37x10-03 3.65x10-03 4.19x10-03 2.77x10-03

Averages 6.43x10-04 6.30x10-03 7.51x10-03 1.44x10-02 1.09x10-02 2.95x10-03

C2M-06 9.15x10-03 1.27x10-02 4.74x10-03 1.64x10-02 7.25x10-03 5.05x10-03

C2M-07 8.71x10-03 9.46x10-03 6.60x10-03 6.10x10-03 8.38x10-03 6.83x10-03

Averages 8.93x10-03 1.11x10-02 5.67x10-03 1.13x10-02 7.82x10-03 5.94x10-03

C1S-17 6.97x10-03 3.09x10-03 3.07x10-03 4.93x10-03 4.82x10-03 2.77x10-03

C1M-02 1.07x10-03 2.01x10-03 3.21x10-03 4.02x10-03 3.44x10-03 2.19x10-03

C1M-23 5.55x10-04 3.49x10-04 5.18x10-04 3.77x10-04 4.76x10-04 1.72x10-03

C1M-32 4.68x10-03 4.40x10-03 4.90x10-03 3.93x10-03 4.26x10-03 3.33x10-03

Averages 2.10x10-03 2.25x10-03 2.88x10-03 2.78x10-03 2.73x10-03 2.41x10-03

225

Figure 7.10 Typical calculation of the coefficient of consolidation for an oedometer specimen

The resulting values of mv for each of the spoil and mud samples are given in Table 7.7 and are plotted

in Figure 7.11 and Figure 7.12. A comparison of all the spoil and mud is shown in Figure 7.13.

Table 7.7 Coefficient of volume change for all spoil and mud specimens tested

CODE COEFFICIENT OF VOLUME CHANGE, mv, AT APPLIED STRESSES


C3S-10 8.04x10-03 4.71x10-03 4.76x10-03 8.06x10-03 4.03x10-03 1.01x10-02

C3S-13 5.36x10-02 4.99x10-02 3.93x10-02 1.92x10-02 8.98x10-03 3.36x10-03

C3S-20 2.67x10-03 1.04x10-02 6.74x10-03 4.48x10-03 4.07x10-03 3.93x10-03

Averages 2.14x10-02 2.17x10-02 1.69x10-02 1.06x10-02 5.69x10-03 5.80x10-03

C3M-01 5.93x10-04 1.48x10-03 4.41x10-03 7.38x10-03 5.49x10-03 3.34x10-03

C3M-03 3.91x10-03 4.08x10-03 5.18x10-03 6.31x10-03 7.51x10-03 6.42x10-03

C3M-04 6.45x10-03 5.00x10-03 8.12x10-03 7.00x10-03 9.00x10-03 8.38x10-03

C3M-05 2.32x10-03 2.74x10-03 3.57x10-03 4.74x10-03 5.74x10-03 5.32x10-03

C3M-08 1.12x10-02 1.59x10-02 1.33x10-02 1.13x10-02 8.66x10-03 1.18x10-02

C3M-12 2.60x10-02 7.85x10-03 1.12x10-02 2.46x10-02 1.59x10-02 2.30x10-02

C3M-30 3.45x10-02 3.18x10-02 1.87x10-02 1.80x10-02 2.87x10-02 2.72x10-02

C3M-18 4.76x10-04 1.93x10-04 3.04x10-04 3.83x10-04 6.34x10-04 6.19x10-04

Averages 1.07x10-02 8.63x10-03 8.10x10-03 9.96x10-03 1.02x10-02 1.08x10-02

C2S-16 5.39x10-04 9.93x10-03 1.16x10-02 2.51x10-02 1.76x10-02 3.14x10-03

C2S-24 7.48x10-04 2.68x10-03 3.37x10-03 3.65x10-03 4.19x10-03 2.77x10-03

C2M-06 9.15x10-03 1.27x10-02 4.74x10-03 1.64x10-02 7.25x10-03 5.05x10-03

226

Averages 3.48x10-03 8.44x10-03 6.57x10-03 1.51x10-02 9.68x10-03 3.65x10-03

C2M-07 8.71x10-03 9.46x10-03 6.60x10-03 6.10x10-03 8.38x10-03 6.83x10-03

C1S-17 6.97x10-03 3.09x10-03 3.07x10-03 4.93x10-03 4.82x10-03 2.77x10-03

C1M-02 1.07x10-03 2.01x10-03 3.21x10-03 4.02x10-03 3.44x10-03 2.19x10-03

C1M-23 5.55x10-04 3.49x10-04 5.18x10-04 3.77x10-04 4.76x10-04 1.72x10-03

C1M-32 4.68x10-03 4.40x10-03 4.90x10-03 3.93x10-03 4.26x10-03 3.33x10-03

Averages 2.10x10-03 2.25x10-03 2.88x10-03 2.78x10-03 2.73x10-03 2.41x10-03

Figure 7.11 Void ratio versus coefficient of volume change for -4.7 mm spoil tested wet

227

Figure 7.12 Void ratio versus coefficient of volume change for -4.7 mm mud specimens tested

wet

Figure 7.13 Comparison of void ratio versus coefficient of volume change for all -4.7 mm spoil

and mud specimens tested wet

228

Using the coefficient of consolidation and the coefficient of volume change, the saturated hydraulic

conductivity can be calculated from:

𝑘v = cv. mv. (7)

where:

kv = saturated hydraulic conductivity;

cv = coefficient of consolidation;

mv = coefficient of volume change; and

w = unit weight of water = 9.81 kN/m3.

The saturated hydraulic conductivity was calculated for all spoil and mud specimens, given in

Table 7.8, and are plotted in Figure 7.14, which show the calculated hydraulic conductivities for each

material at each stage of loading. Figure 7.15 shows only the hydraulic conductivities calculated at

1,000 kPa.

Table 7.8 Hydraulic conductivity for all spoil and mud specimens tested

CODE HYDRAULIC CONDUCTIVITY UNDER APPLIED STRESS (m/s)


C3S-10 6.31x10-08 7.94x10-09 3.80x10-09 3.10x10-09 6.82x10-10 8.05x10-10

C3S-13 1.85x10-07 1.05x10-07 4.40x10-08 1.19x10-08 2.41x10-09 4.43x10-10

C3S-20 8.14x10-09 1.16x10-08 5.45x10-09 1.89x10-09 7.47x10-10 2.99x10-10

Averages 8.54x10-08 4.14x10-08 1.78x10-08 5.62x10-09 1.28x10-09 5.16x10-10

C3M-01 2.92x10-09 2.39x10-09 2.86x10-09 2.54x10-09 8.92x10-10 2.46x10-10

C3M-03 7.74x10-09 3.75x10-09 2.75x10-09 1.73x10-09 9.84x10-10 3.90x10-10

C3M-04 8.98x10-09 4.62x10-09 3.99x10-09 2.00x10-09 1.16x10-09 5.43x10-10

C3M-05 4.07x10-09 2.84x10-09 2.20x10-09 1.51x10-09 8.31x10-10 3.58x10-10

C3M-08 1.76x10-08 7.84x10-09 3.58x10-09 1.85x10-09 6.67x10-10 5.28x10-10

C3M-12 1.04x10-08 2.19x10-09 2.00x10-09 2.90x10-09 9.91x10-10 8.90x10-10

C3M-30 8.75x10-08 1.27x10-08 5.42x10-09 3.44x10-09 2.55x10-09 1.36x10-09

C3M-18 7.91x10-09 7.60x10-10 5.46x10-10 3.39x10-10 2.31x10-10 9.01x10-11

Averages 1.84x10-08 4.63x10-09 2.92x10-09 2.04x10-09 1.04x10-09 5.51x10-10

C2S-16 1.43x10-09 2.11x10-08 1.04x10-08 8.09x10-09 2.06x10-09 1.95x10-10

C2S-24 2.19x10-09 3.61x10-09 1.83x10-09 1.70x10-09 9.59x10-10 2.31x10-10

Averages 1.81x10-09 1.24x10-08 6.13x10-09 4.89x10-09 1.51x10-09 2.13x10-10

C2M-06 5.01x10-08 2.01x10-08 3.15x10-09 7.17x10-09 2.07x10-09 5.50x10-10

C2M-07 2.71x10-08 1.36x10-08 5.29x10-09 2.36x10-09 1.39x10-09 4.97x10-10

Averages 3.86x10-08 1.69x10-08 4.22x10-09 4.76x10-09 1.73x10-09 5.24x10-10

C1S-17 3.05x10-08 5.38x10-09 3.21x10-09 2.31x10-09 8.89x10-10 2.16x10-10

229

Averages 3.05x10-08 5.38x10-09 3.21x10-09 2.31x10-09 8.89x10-10 2.16x10-10

C1M-02 6.93x10-09 2.99x10-09 2.56x10-09 1.66x10-09 6.63x10-10 2.04x10-10

C1M-23 1.08x10-08 1.16x10-09 1.42x10-09 3.33x10-10 1.64x10-10 1.34x10-10

C1M-32 2.43x10-08 5.29x10-09 4.65x10-09 1.93x10-09 9.25x10-10 3.21x10-10

Averages 1.40x10-08 3.15x10-09 2.88x10-09 1.31x10-09 5.84x10-10 2.20x10-10

Figure 7.14 Hydraulic conductivity values for all spoil and mud specimens tested

230

Figure 7.15 Hydraulic conductivity at 1,000 kPa applied stress for all spoil and mud

specimens tested

From the materials tested, a large range of results for hydraulic conductivity were obtained. The

conductivities of all materials ranged from 1.4x10-09 to 0.9x10-11 m/s, with values for both C3M-18

and C3M-23 being particularly low. At 1,000 kPa, the Category 3 spoil had higher hydraulic

conductivities than the Category 2 or 1 spoil materials. The Category 3 and 2 muds had similar values,

with slightly lower averages for the Category 1 mud.

It is important to note that these calculations are based off samples sieved to -2.36 mm. Based on the

material characterisation in Section 5.1.4, the coarser particles with the Category 3 spoil and mud

materials would have a large influence on the hydraulic conductivity of the material in situ.

7.1.1 Discussion and conclusions of consolidometer test results

All spoil and mud materials collected in situ were tested using a standard consolidometer. Each

specimen was soaked for 24 hours before consolidation. Initial dry densities of all spoil specimens

ranged from 1.3 to 1.5 t/m3 prior to consolidation, and from 1.9 to 2.0 t/m3 under an applied stress of

1,000 kPa. The lowest final void ratios were obtained by the Category 3 spoil. The same relationship

is seen reflected in the final moisture contents measured.

For the mud materials tested, the initial dry density mostly ranged from 1.3 to 1.6 t/m3, with the lowest

values for the Category 1 mud samples. C3M-18 and C1M-23 were both significant outliers with very

231

high initial moisture contents and high void ratios. Relative to the tested spoil materials, a large range

of final dry densities were recorded for the tested muds. For the Category 3 mud, values varied from

1.65 to 1.97 t/m3. Both Category 2 muds had initial dry densities above 1.9 t/m3. C1M-02 had the

highest value for a Category 1 mud at 1.82 t/m3, with the other two specimens below 1.6 t/m3. There

is no clear relationship between spoil category and the final dry densities of the tested muds. With

respect to the void ratio, on average, the lowest void ratios were obtained by the Category 3 mud,

followed by Category 2, and lastly Category 1. A strong relationship between the void ratio before

loading and at 1,000 kPa was identified, with an r2 value of 0.69.

As all materials were placed loosely, the recorded settlement is largely related to the initial conditions

of the testing. The range of settlement for all spoil specimens was within 10%, with average

settlements at 1,000 kPa ranging from 20 to 35%. A larger range was observed for the mud specimens,

ranging from 8 to 35%, with two significant outliers having settlements of 55% and 60% (C3M-18

and C1M-23, respectively). Typically, less settlement was observed for the Category 3 mud. As in

situ settlement can cause instabilities within a lowwall, these results highlight the likelihood of

Category 3 undergoing less settlement on loading. A very strong relationship between settlement

percentage and initial dry density was identified, with an r2 value of 0.866.

The compression index was calculated for all spoil and mud materials. For the spoil specimens, the

compression index ranged from 0.13 to 0.78. A larger range was observed for the mud samples tested,

with values from 0.025 to 0.13. C3M-18 and C1M-23 were both significant outliers with values of

0.22 and 0.26, respectively.

Using the coefficient of compressibility and the coefficient of volume change, the hydraulic

conductivity of all specimens was back-calculated, showing a spread of values ranging from 1.4x10-

09 to 0.9x10-11 m/s, with most values between 2x10-10 m/s and 6x10-10 m/s. On average, higher

hydraulic conductivities were associated with the Category 3 materials. As these calculations are

based on samples scalped to pass 2.36 mm, higher conductivities would be expected in situ for the

coarser-grained actual materials, particularly for Category 3 spoil and mud. For the low hydraulic

conductivity muds, these results indicate very long durations of time would be required for

consolidation to occur in situ, which is not realistic considering the rapid pace of strip mining.

232

7.2 Large Slurry Consolidometer Test Results

Conventional consolidation equipment such as the standard consolidometer or Rowe cell are unable

to adequately handle slurry-like material. The standard consolidometer uses a thin specimen with a

height of 20 mm, loaded vertically. To load these specimens, a loading cap must be placed on top.

For a slurry, there is inadequate strength in the specimen to maintain the weight of the loading cap.

The Rowe cell can test slurry-like soils during consolidation; however, it has several limitations.

During consolidation, it is important to maintain a level surface. This can be difficult using a Rowe

cell as deformation of the specimen’s centre can cause bowing in the porous disk. It is also limited in

height, making it difficult to test slurries with high moisture contents (Umehara & Zen 1980).

Due to these limitations and for the recording of the desired measurements, a purpose-built large

slurry consolidometer was used to analyse a select number of muds during consolidation. A schematic

of the large slurry consolidometer was provided in Section 4.10.2, with design details and

methodology discussed. For this testing, drainage was only allowed from the top of the cell to

simulate upward drainage of the mud sitting on an impermeable pit floor. A loading rate of

0.1 kPa/min was applied, up to a maximum stress of 500 kPa. For each test, measurements of the

applied stress, stress measured at the base, pore water pressure readings along the column and at the

base, and settlement were taken.

While the applied stress and the stress at the base were recorded, the distribution of stress down the

length of the specimen is unknown, although the measured pore water pressures give some indication.

The average stress can be estimated using Equations (8) and (9):

σavg =σt − σb

2 (8)

σavg = σb +

1

3∗ (σt − σb) (9)

where:

σavg = the average stress within the specimen;

σt = the applied stress at the top of the specimen; and

σb = the measured stress at the base of the specimen.

Equations (8) and (9) assume linear and parabolic stress distributions within the specimen,

respectively. As drainage is only allowed from the top of the specimen, it is therefore assumed that

the closest fit is obtained using Equation (9). To determine the average effective stress, the following

equation was used:

233

σ′avg = σ𝑎𝑣𝑔 − 𝛽. 𝑢𝑏 (10)

where:

σ’avg = the average effective stress within the specimen;

β = average pore water pressure/pore water pressure at the base of the specimen; and

ub = measured pore water pressure at the base of the specimen.

For the calculations of average effective stress, several authors have suggested suitable values of β.

For the purposes of this research, values suggested by Janbu et al. (1980 and Leroueil et al. (1985)

were used to calculate the following conditions:

β = 0.67 if ub/σavg < 0.40;

β = 0.63+0.1×ub/σavg if 0.40 < ub/σavg <0.70;

β = 0.5409+0.2273×ub/σavg if 0.70 < ub/σavg <0.9.2; and

β = 0.75 if ub/σavg < 0.92.

Wall friction is usually ignored when the height to diameter ratio is greater than 0.4. The dimensions

of the large slurry consolidometer result in a height to diameter ratio of 2. It is important to consider

wall friction in the analysis of the test results. The loss of applied stress due to friction between the

piston and specimen and the wall was calculated using:

𝜇 = σt − σb (11)

μ = friction along the wall

σt = the applied stress at the top of the specimen

σb = the measured stress at the base of the specimen

The hydraulic conductivity of each specimen was calculated. Due to the constant rate of loading,

analysis of the hydraulic conductivity cannot be calculated using the standard method applied in

standard consolidation testing. Instead, the method proposed by Davison & Atkinson (1990) was

used:

𝑘 = (

𝑤

∗ 𝐻

2 ∗ 𝑢ℎ ) . (

∆ℎ

∆𝑡) (12)

where:

k = the hydraulic conductivity of the specimen;

γw = the unit weight of water;

234

H = the current height of the specimen;

uh = the excess pore water pressure at the base of the specimen;

∆h = the change of height in the specimen during the selected change in time; and

∆t = the change of time.

Each specimen was placed into the large slurry consolidometer with it's in situ moisture content. To

ensure full saturation and a level surface for measurement, sample supernatant was added to ensure

that the top surface of the specimens was level. This additional water was accounted for within the

analysis. For each mud tested, details of the state of the specimen initially, at the end of the loading

stage and at the end of the test post pore water pressure dissipation, are provided.

Four different mud specimens were analysed in the large slurry consolidometer. These were C3M-

08, C3M-18, C1M-02 and C1M-23. Photos of each mud sample in situ are provided in Figure 7.16

to Figure 7.19. The particle size distributions of the four materials are plotted in Figure 7.20. Each

specimen was scalped to -19 mm match the dimensions of the testing equipment.

To determine the potential variability between materials, both Category 1 and 3 muds were

investigated. C3M-08 was selected due to its high sand and gravel composition, to investigate how a

coarse-grained mud would perform under consolidation. C3M-18 was another Category 3 mud, with

an extremely fine particle size distribution composed almost entirely of silt. Lastly, the two Category

1 materials C1M-02 and C1M-23 were examined as a representation of problematic material that

could be encountered in situ.

Figure 7.16 C3M-08 sampling

235




236

Figure 7.20 Particle size distribution of samples tested in slurry consolidometer

7.2.1 Test results for C3M-08

Details of the state of the C3M-08 specimen before, at the end of loading, and at the end of the test

are given in Table 7.9. The stresses and pore water pressures and settlement measured during the

testing are plotted in Figure 7.21. The average void ratio in comparison to the average effective stress

and calculated hydraulic conductivity are plotted in Figure 7.22 and Figure 7.23, respectively.

The specimen was initially fully submerged. Due to the self-weight settlement, approximately 10 mm

of supernatant was visually observed. This is also seen in Figure 7.21, with a very rapid settlement of

11 mm before increases in total and effective stress were observed. The settlement of this specimen

was observed to occur rapidly. Once the maximum applied load was achieved, negligible settlement

was recorded afterwards. Significant wall friction was calculated from the test results, with the wall

friction stopping close to 40% of the applied stress from reaching the base.

Throughout the entire test with a loading rate of 0.1 kPa/min, no significant pore water pressures were

produced, implying this material could be loaded at a faster rate, and that the calculated hydraulic

conductivity values given in Table 7.9 and plotted in Figure 7.25 underestimate the actual hydraulic

conductivity of this material. At the end of loading, the hydraulic conductivity was calculated as

3.09x10-09 m/s.

237

Table 7.9 Slurry consolidometer test results for C3M-08

C3M-08 INITIAL END OF LOADING FINAL CONDITIONS

Height (mm) 135.0 111.4 111.4

Settlement (%) 0 17.5 17.5

Applied Stress (kPa) 0 500.0 500.0

Base Stress (kPa) 0 303.7 301.7

Wall Friction (kPa) 0 196.3 198.3

Base Pore Water Pressure (kPa) 1.1 1.7 1.9

Average Effective Stress (kPa) 0 367.9 366.5

Gravimetric Moisture Content (%) 26.7 14.2 14.2

Dry Density (t/m3) 1.396 1.692 1.693

Void Ratio 0.6 0.32 0.32

Hydraulic Conductivity (m/s) 8.82x10-07 3.09x10-09 8.43x10-10

Figure 7.21 Slurry consolidometer stress and pore water pressure plots for C3M-08

238

Figure 7.22 Slurry consolidometer void ratio versus effective stress for C3M-08

Figure 7.23 Slurry consolidometer hydraulic conductivity versus effective stress for C3M-08

239


Details of the state of the C3M-18 specimen before, at the end of loading and at the end of the test




The in situ moisture content of the C3M-18 mud was higher than its liquid limit of 61.8%. Due to

this, the specimen had a relatively low dry density and a high void ratio. During loading, high pore

water pressures were measured. At the base of the sample, a decrease in pore water pressure was only

observed after three days of loading. Once the maximum applied stress had been reached, a steady

decline in pore water pressure was observed, with full dissipation after eight days.

The decrease in pore water pressure is associated with an increase in average effective stress,

indicating a transition from a slurry to a soil-like state. During this transition, the distance between

particles reduces, resulting in greater contact between particles. Due to the top only drainage of this

test, this transition will begin to take place at the top of the specimen while preferential pathways

form. An increase in average effective stress can be seen after the peak of each pore water pressure

measurement, starting at 80 mm above the base, followed by 40 mm above the base, and finally at

the base. As this transition occurs, an increase in wall friction also takes place, with a final value of

72 kPa. The hydraulic conductivity was calculated at the end of loading to be 4.65x10-11 m/s.



Height (mm) 113.0 58.3 54.1

Settlement (%) 0 48.4 52.1







Dry Density (t/m3) 0.705 1.367 1.472

Void Ratio 2.40 0.76 0.63


240


Figure 7.25 Slurry consolidometer void ratio versus effective stress for C3M-18

241







The pore water pressures recorded at 80 mm above the base, 40 mm above the base, and at the base

of the specimen rise with the increasing applied stress until between days 1 and 2, after which the

pressures begin to dissipate. Once the maximum applied stress is reached and becomes steady, a sharp

reduction in pore water pressure was observed at 40 mm above the base, and at the base of the

specimen. With reductions in pore water pressure, the average effective stress is seen to increase, as

does the wall friction. Most of the settlement was observed to occur during loading. Once the

maximum applied stress had been reached, only a further 1.1% settlement occurred over the

remaining seven days at that applied stress as the final pore water pressures had dissipated. The

hydraulic conductivity was calculated at the end of loading at 8.23x10-11 m/s.

242



Height (mm) 118.0 77.0 75.8

Settlement (%) 0 34.7 35.8







Dry Density (t/m3) 1.100 1.686 1.713

Void Ratio 1.32 0.51 0.49



243

Figure 7.28 Slurry consolidometer void ratio versus effective stress for C1S-02


244




test are plotted in Figure 7.30. The average void ratio in comparison to the average effective stress


C1M-23 was sampled in situ with a very wet moisture content, far above its liquid limit of 61.9%.

During loading, large pore water pressures were recorded at 40 mm above the base and at the base of

the specimen. These pore water pressures started to dissipate after 2.5 to 3 days, at a slow rate. After

achieving the maximum applied stress and under constant loading, another four days were required

for all the pore water pressures to dissipate. With a reduction in pore water pressures, increases in

average effective stress and wall friction are measured. The hydraulic conductivity was calculated at

the end of loading to be 6.71x10-11 m/s.



Height (mm) 103.0 51.7 47.2

Settlement (%) 0 49.8 54.2







Dry Density (t/m3) 694 1383 1516

Void Ratio 2.75 0.88 0.72


245

Figure 7.30 Slurry consolidometer stress and pore water pressure data for C1S-23

Figure 7.31 Slurry consolidometer void ratio versus effective stress plot for C1S-23

246

Figure 7.32 Slurry consolidometer hydraulic conductivity versus effective stress plot for C1S-

23

7.2.5 Discussion and conclusions of slurry consolidometer test results

The large slurry consolidometer is a device manufactured specifically for the consolidation of

material from a slurry-like state to a soil-like state. As it is a non-standard device, the analysis of the

results differed from conventional consolidation data. Due to this, literature was relied upon to

analyse the data accounting for the design differences, as discussed at the start of this section.

Four mud materials were investigated, namely C3M-08, C3M-18, C1M-02 and C1M-23. C3M-08

was chosen due to its sandy particle size distribution, to contrast against the other materials that were

identified to be highly problematic in situ. These materials have particle size distributions dominated

by the silt-sized fraction, with some sand-sized particles present.

Each material was consolidated under a constant loading pressure of 0.1 kPa/min until a maximum

applied stress of 500 kPa was achieved, which was held constant until all pore water pressures had

dissipated.

The settlement of all specimens is shown in Figure 7.33. C3M-08 showed self-weight settlement upon

placement into the consolidometer, with 11 mm of free water on the surface observed prior to loading.

During loading, no measurable pore water pressures developed in the specimen, indicating drained

conditions.

247

For C3M-18, C3M-02 and C3M-23, significant pore water pressures developed during the loading

phase with dissipation only starting to occur between 1.5 days and three days, implying undrained

loading. Prior to dissipation, low average effective stresses were calculated. After the pore water

pressures began to dissipate, an increase in average effective stress was observed and continued until

full pore water pressure dissipation was reached. This took 6 to 8 days. Due to top only drainage, the

pore water pressures dissipated from the top down, and finally at the base, via preferential pathways

which could potentially be along the side of the consolidometer.

Most of the settlement in all specimens occurred during the loading phase, with sharp reductions in

settlement observed once the pore water pressures begun to dissipate. For C3M-08, once the

maximum applied stress was achieved, negligible settlement occurred.

Significant friction on the side of the device occurred in C3M-08, due to the relatively coarse particle

size distribution of the material. For all other specimens, the friction loss remained below 100 kPa

(20% of the applied stress). Table 7.13 details the final state of each specimen at the end of testing.

The hydraulic conductivity values are calculated at the end of the constant loading. After constant

applied stress of 500 kPa was reached, a range of results was obtained as seen in each specimen’s

respective void ratio versus hydraulic conductivity plot.

Figure 7.33 Slurry consolidometer settlement versus time for all mud specimens tested

248

Table 7.13 End of testing state for all slurry consolidometer specimens tested

Parameter C3M-08 C3M-18 C1M-02 C1M-23

Height (mm) 111.4 54.1 75.8 47.2

Settlement (%) 17.5 52.1 35.8 54.2

Applied Stress (kPa) 500.0 500.0 500.0 500.0

Base Stress (kPa) 301.7 428.0 394.5 454.6

Wall Friction (kPa) 198.3 72.0 105.5 45.4

Base Pore Water Pressure (kPa) 1.9 7.5 3.2 9.8

Average Effective Stress (kPa) 366.5 446.9 427.5 463.1

Gravimetric Moisture Content (%) 14.2 26.3 19.2 27.5

Dry Density (t/m3) 1,693 1,472 1,713 1,516

Void Ratio 0.32 0.63 0.49 0.72

Hydraulic Conductivity (End of Loading) (m/s) 3.09x10-09 4.65x10-11 8.23x10-11 6.71x10-11

A comparison of results shows that C3M-08 had the lowest final gravimetric moisture content and

calculated void ratio, one of the highest dry densities, and a hydraulic conductivity two orders of

magnitude higher than that of the other materials tested. As no pore water pressures were generated

during testing, the specimen tested could have been loaded at a faster rate, resulting in quicker

settlement and hence higher calculated hydraulic conductivities. These results show that C3M-08 has

potential for in situ rehabilitation in terms of consolidation and permeability, potentially negating the

requirement to remove the mud prior to spoiling due to concerns of undrained behaviour on loading.

C3M-18, C1M-02 and C1M-23 all started with moisture contents well above their respective liquid

limits. Post consolidation at an applied stress of 500 kPa, final moisture contents closer to each

measured plastic limit were achieved. The final void ratios ranged from 0.49 to 0.72, with dry

densities from 1.470 to 1,710 t/m3. The hydraulic conductivities calculated at the end of loading

ranged from 4.65x10-11 to 8.23x10-11 m/s. These values are extremely low. In addition to the slow

dissipation of pore water pressures, the results of this testing show the difficulty that could arise in

situ upon loading, with high pore water pressures, low average effective stress, and low hydraulic

conductivities resulting in slow dissipation of pressure and reduced shear strength.

Results of the large slurry consolidometer testing show a range of results, highlighting material that

shows potential for in-pit rehabilitation with respect to consolidation and permeability, and materials

that are likely difficult to rehabilitate. Furthermore, the results show that C3M-18, while classified as

a Category 3 material, performs in a similar way to the Category 1 specimens tested. These results

highlight the importance of accurate classification and potential re-classification over time. It also

shows the impact mineralogy can have on degradation and the resultant material parameters.

249

The results of the large slurry consolidometer testing used a constant loading rate of 0.1 kPa/min up

to 500 kPa to show the nature of the selected materials during consolidation. It is, however, important

to note that the conditions of the test setup should mimic the scenario expected in practice. The

method of loading of spoil onto the mud must be replicated. These tests were conducted at a rate of

144 kPa/day, equivalent to approximately 8 m of fill with a unit weight of 18 kN/m3. Draglines can

produce spoil piles of the order of 60 to 100m high and with lineal rates of advance along the strip of

10 m/day (Duran 2013).

These results, therefore, show that for C3M-18, C1M-02 and C1M-23, this rate of rise would produce

excessive pore pressures and likely result in instabilities. If these muds were encountered, careful

management methods would be required if the mud were to remain in situ such as truck and shovel

systems. For C3M-08, however, the results show the potential for truck and shovel loading without

the build-up of pore pressures, promoting the potential for this material to be left in-pit, potentially

saving millions of dollars in removal costs and associated issues.

250

8 SHEAR STRENGTH TEST RESULTS

The BMA shear strength parameters for spoil categories and mobilisation modes are included in

Table 8.1. For each of the spoil categories, back-calculated and laboratory-based values of unit

weight, friction angle and apparent cohesion were selected for unsaturated, saturated and remoulded

conditions. While a reduction in shear strength parameters is shown in Table 8.1 from the unsaturated

to saturated conditions and from the saturated to remoulded conditions, there is a lack of data on the

shear strength of in-pit mud, detailed in Section 2.3 of the literature review. Due to this, a potentially

conservative remoulded shear strength is most often assumed for design when using the framework.

This section discusses the direct shear strength test results of all spoil and mud specimens tested.

Depending on the amount of material available and its particle size distribution, a 300 mm or 60 mm

direct shear box was used. For the 300 mm direct shear box, samples were scalped to pass 19 mm.

For the 60 mm direct shear box, samples were scalped to pass 6.7 mm. For both the spoil and mud

samples, tests were conducted on loose specimens at their as-sampled moisture content (“dry”), and

after 24 hours of soaking in tap water (“wet”). The specimens were left to consolidate under the

applied normal stress prior to shearing. The results of direct shear tests on dry and wet spoil and mud

are detailed in Section 8.1. Straight lines were fitted to the test data to give an apparent cohesion (c’,

or zero) and a friction angle (ɸ’).

Table 8.1 BMA shear strength parameters for different categories and mobilisation modes

SPOIL

CATEGORY


Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Cohesion =

0 kPa, Friction

Angle (deg)

1 18 1 20 1 25 2.5 20 1 0 18 3 18 1.5

2 18 1 30 15 28 3 20 1 15 7.5 23 2.5 18 1.5

3 18 1 50 15 30 2 20 1 20 10 25 2.5 18 1.5

4 18 1 50 15 35 2.5 20 1 0 30 1.5 28 2

8.1 Spoil and Mud Direct Shear Test Results

8.1.1 Spoil material test results

C3S-13 and C3S-20 were tested both dry and wet in the 300 mm direct shear box. C3S-10 was tested

only dry in the 300 mm direct shear box and tested wet in the 60 mm direct shear box due to a lack

of sample. C2S-16, C2S-24 and C1S-17 were also tested in the 60 mm direct shear box. The results

of the direct shear tests are given in Table 8.2 and are plotted in Figure 8.1 to Figure 8.3. Also given

in Table 8.2 are the corresponding BMA framework shear strength parameters. A summary of the

251

test results in terms of apparent cohesion and friction angle is included in Figure 8.4. The results are

plotted in terms of secant friction angle (assuming zero cohesion) versus applied stress in Figure 8.5.

Table 8.2 Direct shear strength results for spoil tested dry and wet

CODE CATEGORY SOURCE DRY WET

c’ (kPa) ’ (deg) c’ (kPa) ’ (deg)

C3S-10 CAT 3 SPOIL 97.3 30.4 23.1 34.5

C3S-13 CAT 3 SPOIL 63.8 31.9 27.9 30.5

C3S-20 CAT 3 SPOIL 76.3 38.1 0.0 29.6

Category 3 Averages 79.1 33.5 17.0 31.5

C3S-16 CAT 2 SPOIL 150.7 30.0 109.2 22.9

C3S-24 CAT 2 SPOIL 138.9 21.8 15.9 27.6

Category 2 Averages 144.8 25.9 62.5 25.3

C3S-17 CAT 1 SPOIL 118.8 30.9 13.9 26.4

BMA Shear Strength Design Parameters

BMA CAT 3 SPOIL 50 30 20 25



BMA CAT 1-3 SPOIL Remoulded 0 18

For all spoil samples tested, a moderate to a significant decrease in apparent cohesion between dry

and wet tests was observed, likely due to the loss of matric suctions within the samples on wetting.

Overall, the spoil samples tested gave shear strength parameters higher than the corresponding BMA

framework values, indicating that the use of the BMA shear strength parameters would result in

conservative designs.

The Category 3 spoil samples tested as-sampled gave dry apparent cohesion values ranging from 63

to 97 kPa, and dry friction angles ranging from 30 to 38o, somewhat higher than the corresponding

BMA framework values of 50 kPa and 30o, respectively. After 24 hours of soaking, the Category 3

spoil samples gave wet apparent cohesion values ranging from 0 to 28 kPa, somewhat lower than the

corresponding BMA framework value of 30 kPa, and friction angles ranging from 29 to 34o,

significantly higher than the corresponding BMA framework value of 25o.

The Category 2 spoil samples tested as-sampled gave dry apparent cohesion values very much higher

than the corresponding BMA framework value of 30 kPa, and dry friction angles somewhat lower

than the corresponding BMA framework value of 28o. After 24 hours of soaking, the Category 2 spoil

samples gave wet apparent cohesion values very much higher than the corresponding BMA

framework value of 15 kPa, and friction angles at or above the corresponding BMA framework value

of 23o.

252

The single Category 1 spoil sample tested as-sampled gave a dry apparent cohesion value very much

higher than the corresponding BMA framework value of 20 kPa, and a friction angle somewhat higher

than the corresponding BMA framework value of 25o. After 24 hours of soaking, the Category 1 spoil

sample gave wet apparent cohesion value very much higher than the corresponding BMA framework

value of 0 kPa, and a friction angle well above the corresponding BMA framework value of 18o.

Figure 8.1 Direct shear strength results for Category 3 spoil tested dry and wet

253



254

Figure 8.4 Apparent cohesion and friction angle for all spoil

Figure 8.4 illustrates the spread of results of the spoil shear strength testing both wet and dry in terms

of apparent cohesion and friction angle. Arrows are used to highlight the changes in shear strength of

the materials from the as-sampled moisture content (dry), to that after 24 hours of soaking (wet). For

all tests, a loss of apparent cohesion was experienced on soaking. For all but two materials, a reduction

in friction angle was also observed.

Figure 8.5 illustrates the variability between the tested spoils with respect to the calculated secant

friction angle against the applied stress; a technique utilised by Leps (1970). The tests gave similar

average strength parameters for Category 3, 2 and 1 spoil samples at their as-sampled moisture

content.

After soaking, the Category 3 and 2 spoil samples gave similar strength parameters under both

250 kPa and 500 kPa of applied normal stress. Overall, the Category 1 spoil sample gave the lowest

secant friction angles. A comparison between the calculated secant friction angles with those

calculated from the BMA framework suggested shear strength parameters shows that for all spoil

categories, tested under both dry and wet conditions, the test results were on average higher. All spoil

samples tested dry gave secant friction angles above the poor quality rockfill line reported by Leps

(1970), except for the Category 2 spoil at 1,000 kPa.

255

Figure 8.5 Secant friction angle versus applied normal stress for Category 1, 2 and 3 spoil

tested dry and wet

8.1.2 Mud material test results

The shear strength of all sampled mud materials was examined in a 60 mm direct shear box, with the

results of the testing given in Table 8.3. C3M-18 and C1M-23 both had in situ moisture contents

above their liquid limits, and extremely fine particle size distributions. Due to this, the material had

to be consolidated in a large slurry consolidometer before being tested in the direct shear apparatus.

Post-consolidation, the two mud materials were tested in a water bath in accordance with the wet

testing methodology. For all other materials, testing was conducted both as-sampled (dry) and after

24 hours of soaking (wet). C3M-05 was not tested dry due to a lack of material.

256

Table 8.3 Direct shear strength results for mud tested dry and wet



C3M-01 CAT 3 MUD 11.4 28.5 41.3 26.7

C3M-03 CAT 3 MUD 14.6 32.2 25.6 31.1

C3M-04 CAT 3 MUD 35.2 32.8 41.1 32.4

C3M-05 CAT 3 MUD - - 12.3 30.0

C3M-08 CAT 3 MUD 20.9 35.9 49.2 35.6

C3M-12 CAT 3 MUD 22.3 35.6 26.0 36.0

C3M-18 CAT 3 MUD NA NA 82.0 13.2

C3M-30 CAT 3 MUD 13.8 36.9 38.6 34.8

Category 3 averages 19.7 33.7 39.5 30.0

C3S-06 CAT 2 MUD 26.9 28.7 31.2 25.5

C3S-07 CAT 2 MUD 0.0 34.9 0.1 29.1


C3S-02 CAT 1 MUD 0 27.0 0 26.0

C3S-23 CAT 1 MUD NA NA 57.6 15.6

C3S-32 CAT 1 MUD 4.6 31.9 23.1 26.2







On average, the Category 3 mud gave generally higher apparent cohesion values and higher friction

angles during both dry and wet testing than the Category 2 mud, which in turn gave higher values

than the Category 1 mud tested dry. However, when tested wet, the Category 1 mud gave higher shear

strength parameters than the Category 2 mud.

The Category 3 mud samples tested at the as-sampled moisture content gave apparent cohesion values

ranging from 11 to 35 kPa, and friction angles ranging from 28 to 37o. After 24 hours of soaking, the

Category 3 mud samples gave a larger range of apparent cohesion values from 12 to 82 kPa and a

larger range of friction angles from 13 to 35o. C3M-18 mud was a clear outlier, showing a very high

apparent cohesion, and an extremely low friction angle on testing after pre-consolidation.

257

The two Category 2 mud samples tested showed little change in apparent cohesion on soaking, but a

reduction in friction angle to slightly lower than those for the Category 3 mud. The three Category 1

mud samples tested (C3M-23 after pre-consolidation) showed reasonably similar shear strengths

overall to Category 2 mud.

The mud samples tested generally gave shear strengths higher than the BMA framework saturated

shear strength parameters for the corresponding spoil category, which are significantly higher than

the remoulded shear strength parameters of each spoil category. This indicates that if allowed to

consolidate, the mud samples would have shear strengths much higher than is typically adopted for

design. However, C3M-18 and C1M-23 were outliers, with shear strengths below the BMA

framework saturated shear strengths for Category 3 spoil, up to 500 kPa applied stress. Their apparent

cohesion values were high while their friction angles were low, in contrast to the BMA framework

values for a remoulded spoil of zero apparent cohesion and 18o friction angle.

Figure 8.6 Direct shear strength results of Category 3 mud tested dry and wet

258



259

Figure 8.9 shows the apparent cohesion and friction angle for all tested muds, both wet and dry.

Grouping can be observed for most specimens tested, with Category 3 on average having a higher

apparent cohesion and friction angle than the Category 2 muds, which in turn showed higher apparent

cohesion values than the Category 1 mud. Excluding the outliers C3M-18 and C1M-23, all materials

had friction angles tested at over 25o, in comparison to the BMA framework remoulded shear strength

of 18o.

These results have also been plotted as secant friction angle versus applied normal stress in

Figure 8.10, with the BMA remoulded values included. Plotting the data in this manner again shows

the differentiation between categories, with secant friction angles of Category 3 mud typically being

higher than Category 2, which were in turn higher than Category 1. For the Category 3 and 1 mud, a

negligible difference is observed between dry and wet testing, with most materials being sampled

with moisture contents at or above their plastic limit. A larger difference is seen in the Category 2

muds as they were sampled dry with weak agglomeration, and hence, underwent more degradation

during the soaking period. C3M-18 and C1M-23 were excluded from the averages, being recognised

as clear outliers.

Figure 8.9 Friction angle and apparent cohesion for all mud specimens tested

260

Figure 8.10 Secant friction angle versus applied normal stress for Category 1, 2 and 3 mud

samples tested dry and wet

8.1.3 Conclusions of direct shear test results

All sampled spoil and mud materials underwent direct shear testing with loose placement, both at the

as sampled moisture content (dry), and after 24 hours of soaking in water (wet). Two samples with

moisture contents above their liquid limit had to be consolidated prior to testing.

The shear strength testing of the spoil showed that on average, dry testing resulted in the highest

recorded shear strengths. Soaking for 24 hours resulted in the decrease of both the apparent cohesion

and the effective friction angle. For all spoil materials tested, the final shear strengths determined

were above the values assumed by the BMA category shear strength framework. For dry spoil, the

strengths of all categories were equivalent to Leps (1970) poor quality rock fill.

Typically, the wet spoil strength was equivalent to the unsaturated BMA framework assumptions for

each respective spoil category. For the wet tests, on average the Category 3 spoil had higher strengths

than Category 2, which had higher strengths than Category 1. This was most noticeable with applied

stresses above 500 kPa, with a larger spread of results observed under low stress associated with loose

placement.

For the mud materials tested, on average, the Category 3 mud had a higher shear strength than the

Category 2 mud, which was higher than the Category 1 mud. This was observed for both wet and dry

261

testing. For the Category 3 and 1 mud, negligible reduction in shear strength was observed after

24 hours of soaking. Larger reductions were found in the Category 2 mud related to weakly

agglomerated material rapidly breaking down during soaking. All mud materials had similar friction

angles to those found in the tested spoil. On average, the apparent cohesion was lower. This

relationship is illustrated in Figure 8.11.

In relation to the BMA shear strength framework, the measured shear strengths were well above the

BMA remoulded assumptions of apparent cohesion and friction angle. For most materials, the mud

tested closer to the saturated assumptions of the framework for the respective materials spoil source

category. C3M-18 and C1M-23 were highlighted as significant outliers, with strengths well below

the other materials due to low friction angles. They were however very similar in shear strength to

the assumed remoulded conditions due to high apparent cohesion values. This contrasts with the

framework assumption of friction only.

Figure 8.11 Comparison of spoil and mud apparent cohesion and friction angle

262

8.2 Influence of Pit Flooding on Spoil and Mud Shear Strength

Section 8.1 discussed the shear strength test results of all spoil and mud specimens sampled from the

Bowen Basin. Within this section, two case studies were selected displaying the variation that can

occur within a mine pit, and the implications this has on lowwall design. The two case studies are

analysed with respect to sampling location, material collected and the impact of flooding. These test

results are then discussed with respect to the associated material strengths assigned to them with the

current BMA framework.

8.2.1 C3S-13 spoil and associated mud

C3S-13 was 10-year-old spoil collected from a flooded mining pit, above any previous flooding

levels. C3M-08 was collected from the current flooding level at the toe of the lowwall. C3M-30 was

collected from the highwall of the same pit, at the current flooding level. The sample locations are

highlighted in Figure 8.12. The results of the direct shear testing are given in Table 8.4 and are plotted

in Figure 8.13.

Figure 8.12 C3S-13, associated mud C3M-08 (lowwall) and C3M-30 (highwall) sampling

locations

The shear strength of C3S-13 tested dry provided a friction angle of 31.9o, and an apparent cohesion

of 63.8 kPa. Saturation resulted in a loss of apparent cohesion and a slight reduction in friction angle

of 1.4o. C3M-08 and C3S-30 tested at their in situ moisture content resulted in lower apparent

263

cohesion values, but notably higher friction angles by over 4o. After 24 hours of soaking, both mud

samples showed slight reductions in friction angle, but increases in apparent cohesion.

C3S-13 in its dry state as sampled in situ had a strength approximately equal to the assumed values

of the BMA shear strength framework. C3M-08 and C3M-30, however, showed much higher shear

strengths than predicted, with friction angles almost 10o higher than the saturated Category 3 spoil,

and apparent cohesion values approximately double. In relation to remoulded spoil, significantly

higher strengths were determined with friction angles close to twice that of the predicted value.

Table 8.4 Direct shear strength test results of C3S-13 and associated muds



C3S-13 CAT 3 SPOIL 63.8 31.9 27.9 30.5

C3M-08 CAT 3 MUD 20.9 35.9 49.2 35.6

C3M-30 CAT 3 MUD 13.8 36.9 38.6 34.8




Figure 8.13 Direct shear strength test results for C3S-13 spoil and associated C3M-08 mud

264

If a lowwall is designed with in-pit mud assumed to have no cohesion and a friction angle of 18o, a

scenario like this would require the removal of the mud prior to spoiling, which is a very costly and

time-consuming exercise. In this scenario, characterisation and testing of the in-pit mud highlights

the potential for the mud to remain if handled correctly.

8.2.2 C3S-20 spoil and associated mud

C3S-20 was a 6-month-old spoil collected from above any past flooding levels. C3M-18 was mud

collected from the same pit, at the base of the highwall. C3M-18 was spoil/floor material exposed to

wetting and saturation. The sample locations are highlighted in Figure 8.14. The results of the direct

shear testing are given in Table 8.5 and are plotted in Figure 8.15.

Figure 8.14 C3S-20 and associated mud C3M-18 sampling locations

Table 8.5 Direct shear strength test results for C3S-20 and associated mud


c’ (kPa) ’ (°) c’ (kPa) ’ (°)

C3S-20 CAT 3 SPOIL 76.3 38.1 0.0 29.6

C3M-18 CAT 3 MUD NA NA 82.0 13.2




265

C3S-20 at its as sampled in situ moisture content produced an apparent cohesion of 76.3 kPa, and a

friction angle of 38.1o, both of which are significantly higher than the assumed strengths provided by

the BMA framework. Upon wetting, however, significant decreases in shear strength are observed

with no calculated apparent cohesion, and a friction angle of 29.6o; an 8.5o decrease. In its wet

condition, the shear strength of the spoil is in-between the assumed strengths of unsaturated and

saturated spoil. C3M-18, the associated mud material, had an extremely high in situ moisture content

and was not able to be tested without prior consolidation. The consolidated material when sheared

had a high apparent cohesion of 82 kPa, and a low friction angle of 13.2o. Under low stresses, C3M-

18 gave shear strength values higher than those calculated for the BMA framework remoulded

strength values, but similar results at 1,000 kPa. A low friction angle would result in the remoulded

conditions being more conservative than the mud at higher stresses.

Figure 8.15 Direct shear strength test results for C3S-20 spoil and associated C3M-18 mud

For the scenario of C3S-20 at the lowwall and C3M-18 on the pit floor, the values for shear strength

that would be assumed using the BMA framework are relatively accurate, with the assumption that

the in-pit mud would be assigned a remoulded shear strength. It would, therefore, be important to

manage the mud correctly in relation to the results of a stability analysis, potentially requiring a mud

cleanout.

266

8.3 Influence of Degradation on Spoil Shear Strength

Section 6 discussed in detail the degradation of five spoil materials collected from the Bowen Basin

that had not been exposed to past flooding events. Each material underwent a series of particle size

distribution tests, including dry sieving, wet sieving after 24 hours of soaking in a water bath, wet

sieving after three wetting and drying cycles, and lastly wet sieving after a modified slake durability

test. The results highlighted a significant variation in the materials, with correlations made to the

spoils physical, chemical, mineralogical and geochemical parameters.

For each of these spoil materials, the specimens that had undergone wetting and drying cycles also

had their shear strength tested in a direct shear box. All materials were tested in a 60 mm direct shear

box. Each specimen was soaked for 24 hours prior to testing and submerged in a water bath during

shearing.

Photos of each spoil specimen in situ, and after the wetting and drying cycles are provided in

Figure 8.16 to Figure 8.20.

The results show that visually C3S-13 did not agglomerate during the wetting and drying cycles. For

all other specimens, a high degree of agglomeration and crusting on the surface is observed,

irrespective of the materials assigned spoil category. The largest shrinkage cracks are observed in

C1S-17, with moderate cracking in C3S-20, C2S-16 and C2S-24.

Figure 8.16 C3S-13 as-sampled and after three wetting and drying cycles

267




268


The results of the shear strength testing are given in Table 8.6, with values of the material tested as-

sampled in situ (dry), after 24 hours of soaking (wet), and after three wetting and drying cycles

(degraded).

In all cases, the dry shear strength of the tested spoil specimens was underestimated by the BMA

shear strength framework with respect to apparent cohesion, with the largest variations in the

Category 2 and 1 spoil. With respect to friction angle, all specimens had higher values except for

C3S-24, which was significantly lower.

After 24 hours of soaking, similar results are obtained, with the shear strength of the spoil typically

underestimated. Lastly, the degraded shear strengths of the spoil samples show significantly higher

results than the remoulded assumption of the BMA shear strength framework, and reasonably higher

strengths that the saturated assumptions.

On average, most of the change in shear strength occurs during the initial 24-hour soaking period.

Further wetting and drying cycles did not result in a loss of shear strength, and in some cases caused

an increase in apparent cohesion.

269

Table 8.6 Direct shear strength test results for all spoil tested dry, wet and degraded

CODE CATEGORY SOURCE DRY WET DEGRADED

c’ (kPa) ’ (°) c’ (kPa) ’ (°) c’ (kPa) ’ (°)

C3S-13 CAT 3 SPOIL 63.8 31.9 27.9 30.5 48.2 32.7

C3S-20 CAT 3 SPOIL 76.3 38.1 0.0 29.6 28.8 26.9

C3S-16 CAT 2 SPOIL 150.7 30.0 109.2 22.9 35.5 26.7

C3S-24 CAT 2 SPOIL 138.9 21.8 15.9 27.6 19.5 27.1

C3S-17 CAT 1 SPOIL 118.8 30.9 13.9 26.4 27.8 24.3


BMA CAT 3 SPOIL 50 30 20 25 0 18

BMA CAT 2 SPOIL 30 28 15 23 0 18

BMA CAT 1 SPOIL 20 25 0 18 0 18

Figure 8.21 shows the particle size distributions for C3S-13 spoil samples after scalping to pass

6.7 mm and tested dry, after soaking for 24 hours, and after three wetting and drying cycles.

Degradation on soaking and wetting and drying cycles is mainly observed in the sand and silt-sized

fractions.

Figure 8.22 shows shear strength versus applied normal stress plots for all three samples. Soaking

resulted in a reduction in apparent cohesion. Wetting and drying cycles resulted in less reduction in

apparent cohesion, and a slightly higher friction angle. These results should be compared with the dry

and wet testing in the large direct shear box of C3S-13 scalped to pass 19 mm, to assess scale effects.


6.7 mm and tested dry, after soaking for 24 hours, and after three wetting and drying cycles. Soaking

caused a large reduction in particle size, with large increases observed in the sand and silt-sized

fractions.

Wetting and drying cycles resulted in some increase in the sand-sized fraction and greater increases

in the silt-sized fraction. The influence of these reductions is observed in the shear strength of the

spoil plotted in Figure 8.24, which shows a significant reduction in shear strength when tested wet.

There is little further reduction in the overall shear strength on wetting and drying cycles, with a

higher apparent cohesion compensated by a lower friction angle. Again, these results should be

compared with the dry and wet testing in the large direct shear box of C3S-20 scalped to pass 19 mm,

to assess scale effects.

270

Figure 8.21 Particle size distribution for C3S-13 scalped to pass 6.7 mm and tested dry,

soaked and after wet/dry cycles

Figure 8.22 Direct shear strength results for C3S-13 scalped to pass 6.7 mm and tested dry,

wet and degraded

271

Figure 8.23 Particle size distribution of C3S-20 scapled to pass 6.7 mm and tested dry, soaked

and after wet/dry cycles


wet and degraded

272



caused a large reduction in particle size, with negligible breakdown on wetting and drying cycles.

Figure 8.26 shows a significant reduction in shear strength when tested wet, affecting both the

apparent cohesion and the friction angle. Wetting and drying cycles caused a reduction in apparent

cohesion, but an increase in friction angle, with overall shear strength largely unchanged above

500 kPa applied stress.



caused a significant reduction in gravel-sized particles, and an increase in sand and silt-sized particles.

Wetting and drying cycles caused further increases in fine sand and silt-sized fractions, with little

variation in the coarse sand and gravel-sized fractions.

Figure 8.28 shows a large reduction in apparent cohesion when tested wet, but an increase in friction

angle. Wetting and drying cycles resulted in a slight increase in apparent cohesion, without changing

the friction angle.

Figure 8.25 Particle size distribution of C3S-16 spoil scalped to pass 6.7 mm and tested dry,

soaked and after wet/dry cycles

273


wet and degraded

Figure 8.27 Particle size distribution of C3S-24 scalped to pass 6.7 mm and tested dry, soaked


274


wet and degraded



caused a significant reduction in gravel-sized particles, and an increase in sand and silt-sized particles.

Wetting and drying cycles caused further increases in fine sand and silt-sized fractions, with little

variation in the coarse sand and gravel-sized fractions.

Figure 8.30 shows a reduction in apparent cohesion when tested wet, but an increase in friction angle.

Wetting and drying cycles resulted in a slight increase in apparent cohesion and a slight decrease in

friction angle.

275

Figure 8.29 Particle size distribution of C3S-17 scalped to pass 6.7 mm and tested dry, soaked



wet and degraded

276

8.3.1 Conclusions of spoil degradation shear strength test results

The direct shear strength testing of a range of spoil samples scalped to pass 6.7 mm and tested dry,

after soaking, and after wetting and drying cycles, the largest reduction in particle size and shear

strength occurred following soaking for all samples apart from C3S-13, which showed negligible

shear strength change. Wetting and drying cycles caused some particle size degradation in most cases,

mainly in the fine sand and silt-sized fractions, and little change in overall shear strength. For all spoil

samples apart from C2S-16, an increase in apparent cohesion was matched by a decrease in friction

angle, likely related to an increase in the fine-grained fractions.

These results highlight the importance of keeping clay-mineral rich spoil dry, since most degradation

occurs on first wetting. If saturated, large reductions in shear strength can occur in clay mineral-rich

spoil irrespective of the original spoil classification, more closely related to the mineralogical and

geochemical parameters of the spoil, as discussed in Section 5.4. Due to this, best practice would be

improved by improved identification of material likely to degrade, and management to keep that

material away from the base of future spoil advances.

277

9 IN-PIT MUD CATEGORISATION, SHEAR STRENGTH

ESTIMATION, AND LOWWALL STABILITY

Thirteen samples of in-pit mud were collected from three mines within the Bowen Basin with aims

of determining appropriate material characteristics and shear strength data. These material parameters

are discussed extensively in previous chapters. Mud formed from a variety of different source

materials was sampled to identify the range that could be expected of mud in situ. With the ability to

categorise mud in situ, or with simple and standardised laboratory tests, improvements can be made

in designs of lowwalls to ensure their stability and to account for the mud, or to determine if the mud

has to be removed prior to spoiling.

Section 9.1 discusses the difficulties and limitations of providing shear strength parameter estimates

of in-pit mud based on assigned categories. An alternative model for estimating the shear strength

was provided, allowing for estimates to be made from particle size distribution test results. Section 9.2

details a set of standardised tests for the characterisation of in-pit mud, with intentions of providing

a methodology for analysis that will determine relevant physical, chemical, mineralogical and

geochemical material characteristics related to predicting the materials geotechnical parameters.

Section 9.3 investigates the stability of lowwalls with mud at the toe for four scenarios. Values from

the laboratory tests on spoil and mud are compared with the current BMA framework strength

assumptions, with a discussion on the variability.

9.1 Categorisation and Shear Strength Estimation of In-Pit Mud

For the categorisation of spoil, the current framework developed by Simmons and McManus (2004),

as shown in Figure 9.1 and visualised in Figure 9.2 was applied. The spoil’s category is determined

based on the predominant particle size, consistency, structure, liquid limit and age, with weightings

applied to each based on engineering judgement and extensive use within the industry.

278

Figure 9.1 Spoil Categories and Attributes (adapted from Simmons & McManus 2004)

Figure 9.2 Spoil Structure attribute to be used in association with Figure 9.1 (adapted from

Simmons & McManus 2004)

Figure 9.2 provides a visual aid in identifying the spoil category in terms of framework and matrix

structure. This is not possible with many mud samples as in their wet state, it can be hard to identify

large particles covered by finer material. Figure 9.3 shows highly degradable spoil after 24 hours of

soaking. On the right side of the image, identification of coarse fractions is difficult in contrast to the

left side, which was gently washed with water. Due to this difficulty, as well as material access and

testing costs, the source material is often used to categorise the mud. Without testing, assumptions

must be made that account for all possibilities, resulting in conservative values having to be used in

the design. For each spoil category, an apparent cohesion and friction angle were assigned for use in

design, as given in Table 9.1. Currently, there is no similar framework available for the estimation of

in-pit mud shear strength parameters.

279

Figure 9.3 Coarse and fine fractions of degraded spoil

The shear strength in terms of apparent cohesion and friction angle is dependent on a number of key

factors as discussed by Hustrulid et al. (2001). The apparent cohesion is related to the electrostatic

bonds between clay and silt-sized particles and capillary forces formed between particles resulting in

suction. The friction angle is determined by the particle size distribution, particle shape, surface

roughness, strength and the specific gravity of individual particles, the state of packing, and the

applied stress. For the mud materials collected Table 9.2 details the source category, D90, D50, D10,

Cu and Cc after wet sieving. The particle size distributions are plotted in Figure 9.4.

Table 9.1 BMA shear strength parameters for categories and mobilisation modes

SPOIL

CATEGORY


Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Cohesion =

0 kPa, Friction

Angle (deg)

1 18 1 20 1 25 2.5 20 1 0 18 3 18 1.5

2 18 1 30 15 28 3 20 1 15 7.5 23 2.5 18 1.5

3 18 1 50 15 30 2 20 1 20 10 25 2.5 18 1.5

4 18 1 50 15 35 2.5 20 1 0 30 1.5 28 2

280

Table 9.2 Wet sieved particle size distributions for all mud samples

CODE SOURCE

CATEGORY D90 D50 D10 Cu Cc

C3M-01 CAT 3 41.26 0.776 0.010 0.066 0.010

C3M-03 CAT 3 21.58 0.793 0.011 0.061 0.011

C3M-04 CAT 3 30.59 4.254 0.011 0.308 0.011

C3M-05 CAT 3 26.64 6.596 0.058 3.538 0.058

C3M-08 CAT 3 13.81 0.985 0.049 0.239 0.049

C3M-12 CAT 3 18.56 0.969 0.021 0.206 0.021

C3M-30 CAT 3 9.734 1.752 0.013 0.247 0.013

C3M-18 CAT 3 0.063 0.012 0.007 0.008 0.007

C2M-06 CAT 2 0.573 0.011 0.002 0.005 0.002

C2M-07 CAT 2 9.328 0.271 0.007 0.035 0.007

C1M-02 CAT 1 0.271 0.058 0.043 0.013 0.009

C1M-23 CAT 1 0.073 0.017 0.008 0.004 0.004

C1M-32 CAT 1 0.493 0.018 0.013 0.009 0.008

Figure 9.4 shows that on average, the Category 3 mud materials are coarser than the Category 2 and

1 mud, with significantly higher gravel and sand content. A large spread of results is observed for the

Category 2 mud, with C3M-06 showing a large percentage of silt-sized particles. For the Category 1

mud, very fine distributions are observed, comprising mostly of silt and sand-sized fractions. C3M-

18 has an extremely fine distribution, most like the Category 1 sourced muds. This is further

illustrated in Figure 9.5, showing the D90 values of all materials in relationship to their sources spoil

category. A power regression fit of the data results in a moderately strong r2 value of 0.472. A similar

fit of 0.467 was also found for an exponential relationship. Similar results were found for the D50 and

Cu values.

281

Figure 9.4 Overall particle size distribution curves of all mud samples

Figure 9.5 Source material category compared with D90

282

The measured shear strength parameters of the mud materials are given in Table 9.3 and are plotted

in Figure 9.6 as apparent cohesion versus friction angle, and in Figure 9.7 as secant friction angle

versus applied normal stress. The results in Figure 9.6 are for the wet test conditions (after 24 hours

of soaking in a water bath), while Figure 9.7 shows results for dry (as-sampled) and wet test

conditions.

Table 9.3 Direct shear strength test results for mud tested wet

As illustrated in Figure 9.6, two outliers (C3M-18 and C1M-23) show high apparent cohesion values

and low friction angles. For the rest of the samples, grouping is observed with the Category 3 sourced

mud showing higher friction angles than the Category 2 and 1 mud, ranging from 26 to 36o. For all

categories, a range of results is observed for apparent cohesion, from 0 to 30 kPa for Category 1 and

2, and from 12 to 50 kPa for Category 3. For a statistically significant analysis, further sampling and

testing of Category 1 and 2 muds is required.

CODE SOURCE

CATEGORY

WET SHEAR STRENGTH SECANT FRICTION ANGLE

c’ (kPa) ɸ’ (°) 250 kPa 500 kPa 1,000 kPa

C3M-01 CAT 3 41.3 26.7 34.9 29.8 28.4

C3M-03 CAT 3 25.6 31.1 35.3 33.2 32.0

C3M-04 CAT 3 41.1 32.4 38.9 35.6 33.8

C3M-05 CAT 3 12.3 30.0 33.2 29.8 30.6

C3M-08 CAT 3 49.2 35.6 41.2 40.2 37.0

C3M-12 CAT 3 26.0 36.0 40.3 37.7 36.9

C3M-30 CAT 3 38.6 34.8 37.9 38.9 35.8

C3M-18 CAT 3 82.0 13.2 20.4 19.2 17.0

CATEGORY 3 AVERAGE 39.5 30.0 35.3 33.0 31.4

CATEGORY 3 RANGE 12.3-82 13.2-36 20.4-41.2 19.2-40.2 17-37

CATEGORY 3 MINIMUM 12.3 13.2 20.4 19.2 17.0

C2M-06 CAT 2 31.2 25.5 31.4 28.3 26.8

C2M-07 CAT 2 0.1 29.1 25.9 31.0 28.7

CATEGORY 2 AVERAGE 15.6 27.3 28.7 29.6 27.8

CATEGORY 2 RANGE 0.1-31.2 25.5-29.1 25.9-31.4 28.3-31 26.8-28.7


C1M-02 CAT 1 31.3 24.8 31.4 26.8 26.2

C1M-23 CAT 1 57.6 15.6 20.7 19.6 18.1

C1M-32 CAT 1 23.1 26.2 30.0 28.1 27.2

CATEGORY 1 AVERAGE 37.3 22.2 27.4 24.8 23. 8

CATEGORY 1 RANGE 23.1-57.6 15.6-26.2 20.7-31.4 19.6-28.1 18.1-27.2


283

If a framework is to be developed like the current framework used for spoil, conservative values will

have to be chosen. The Category 3 material highlights this clearly with a range of friction angles from

26o to 36o and one significant outlier at 13.2o. The use of the mean, median, lowest value or a certain

percentile could result in shear strengths being chosen that are unconservative or overly conservative.

This is also relevant for the Category 1 and 2 muds. The test results show that aside from C3M-18

and C1M-23, friction angles for all other mud samples were approximately 25o or higher. For

comparison, a friction angle of 25o is 38% higher than the current remoulded strength assumption.

Figure 9.6 Apparent cohesion versus friction angle for all mud specimens

284

Figure 9.7 Secant friction angle versus applied normal stress for Category 1, 2 and 3 mud dry

and wet

As an alternative to using a category system for assigning shear strength parameters to in-pit mud, a

model was developed used to estimate the friction angle of the mud based on the percentage of sand

and gravel-sized particles present. The values used for developing the model are given in Table 9.4.

All samples were scalped to -6.7 mm before testing. This adjustment was made for the material

fractions to provide accurate percentages.

285

Table 9.4 Mud gravel and sand-size fraction correlated to friction angle and shear strength

A multivariate regression analysis was conducted using the gravel and sand fractions of each

material to predict the friction angle of the material. The model regression statistics, and predicted

values, are given in Table 9.5 and Table 9.6 respectively. The calculated model is described by:

𝐹𝑟𝑖𝑐𝑡𝑖𝑜𝑛 𝐴𝑛𝑔𝑙𝑒 (°) = 16.079 + 0.1674(𝐺𝑟𝑎𝑣𝑒𝑙%) + 0.2993(𝑆𝑎𝑛𝑑%) (13)

where:

Gravel% = Percentage gravel-sized, 2 to 6.7 mm fraction; and

Sand% = Percentage sand-sized, 0.06 to 2 mm fraction.

From the predicted model, an r2 value of 0.87 was obtained, indicating a very strong relationship

between the gravel-sized percentage, the sand-sized percentage, and the friction angles of the

materials. The p-values of the intercept, gravel-sized percentage and sand-sized percentage were all

below 0.05, confirming the statistical significance of each of the variables in the model.

CODE SOURCE

CATEGORY

FRACTION (%) –6.7 mm FRICTION

ANGLE (°)

SHEAR

STRENGTH AT

1,000 kPa

Gravel

(2 to 6.7 mm)

Sand

(0.06 to 2 mm)

C3M-01 CAT 3 14.2 35. 2 26.7 601.9

C3M-03 CAT 3 17.1 38.7 31.1 694.4

C3M-04 CAT 3 24.0 36.5 32.4 741.4

C3M-05 CAT 3 56.0 19.1 30.0 657.4

C3M-08 CAT 3 19.9 55.8 35.6 817.9

C3M-12 CAT 3 17.5 57.5 36.0 817.6

C3M-30 CAT 3 29.9 45.1 34.8 795.0

C3M-18 CAT 3 0.2 2.8 13.2 327.9

C2M-06 CAT 2 11.8 6.4 25.5 553.9

C2M-07 CAT 2 14.2 37.2 29.1 607.5

C1M-02 CAT 1 1.2 26.2 24.8 545.9

C1M-23 CAT 1 0.2 8.9 15.6 362.5

C1M-32 CAT 1 2.9 20.9 26.2 571.0

286

Table 9.5 Multivariate regression statistics for prediction of friction angle using Gravel and

Sand % for -6.7 mm fraction of mud

REGRESSION STASTICS

Multiple R 0.933

R Square 0.870

Adjusted R Square 0.845

Standard Error 2.778

Observations 13 ANOVA

df SS MS F Significance F

Regression 2 518.695 259.347 33.606 3.64x10-05

Residual 10 77.174 7.717

Total 12 595.869

COEFFICIENTS STD

ERROR t STAT P-VALUE

LOWER

95%

UPPER

95%

Intercept 16.079 1.621 9.919 1.71x10-06 12.467 19.691

Gravel (%) 0.167 0.055 3.033 1.26x10-02 0.044 0.290

Sand (%) 0.299 0.047 6.326 8.61x10-05 0.194 0.405

287

Table 9.6 Comparison of laboratory tested friction angle and shear strength and predicted

values using multivariate analysis

In Table 9.6 an estimate of the shear strength was calculated using the predicted friction angle and

contrasted to the measured value of shear strength at 1,000 kPa of applied load, adjusted for the

surface area within the direct shear box. For this calculation, an assumption of zero cohesion was

made. For the materials containing high clay and silt-sized fractions, not accounting for the cohesion

is likely to be conservative. However, due to the risks associated with geotechnical instability of

lowwalls, it is safer to assume conservative shear strengths for highly degraded spoil.

Using the percentages of gravel and sand-sized particles, the model is shown graphically in Table 9.8,

in which the predicted values for each material are plotted, with data labels indicating the percentage

variation from the laboratory test results. Diagonal lines have been added indicating friction angles

which can be read off once the size percentages of a chosen material are plotted. The use of a model

such as this allows for quick estimates of the materials friction angle to be made from a simple particle

size distribution test with greater accuracy than would be provided from a generalised category

framework.

CODE

FRICTION ANGLE (°) VARIABILITY SHEAR STRENGTH AT

1,000 kPa VARIATION

Tested Predicted (°) (%) Measured

Cohesion =

0 kPa,

Calculated

τ (kPa) (%)

C3M-01 26.7 29.0 2.3 8.5 601.9 615.9 15.3 2

C3M-03 31.1 30.5 -0.6 -1.9 694.4 654.5 -41.2 -6

C3M-04 32.4 31.0 -1.4 -4.2 741.4 665.4 -80.0 -10

C3M-05 30.0 31.2 1.2 4.1 657.4 672.9 20.4 2

C3M-08 35.6 36.1 0.5 1.5 817.9 792.3 -51.7 -3

C3M-12 36.0 36.2 0.2 0.6 817.6 797.8 -34.6 -2

C3M-30 34.8 34.6 -0.2 -0.6 795.0 759.3 -50.6 -5

C3M-18 13.2 16.9 3.8 28.8 327.9 326.0 -15.6 0

C2M-06 25.5 20.0 -5.6 -21.8 553.9 398.8 -163.4 -28

C2M-07 29.1 29.6 0.5 1.8 607.5 630.6 13.4 4

C1M-02 24.8 24.1 -0.6 -2.6 545.9 496.7 -47.1 -9

C1M-23 15.6 18.8 3.2 20.5 362.5 376.4 9.2 4

C1M-32 26.2 22.8 -3.4 -13.0 571.0 467.1 -103.6 -18

288

Figure 9.8 A model for predicting the friction angle of in-pit mud

For most of the tested materials, friction angle predictions vary between -13 and +8.5%. Three

materials that contained high levels of silt and clay-sized fractions had predicted values with error

above 20%. A comparison of the laboratory tested friction angles and the model predictions are shown

graphically in Figure 9.9. With an assumption of zero cohesion, the predicted shear strengths of the

tested mud materials have a range between -11% and +3%, with two outliers of -18% (C1M-32) and

-29% (C2M-06). C2M-06 shows a highly conservative prediction, with a large underestimation

related to the material having a very low sand content. The shear strength comparison of laboratory

and predicted values are plotted in Figure 9.10.

289

Figure 9.9 Comparison of laboratory tested friction angle and predicted friction angle of mud

materials

Figure 9.10 Comparison of laboratory tested shear strength and predicted shear strength

assuming zero cohesion for all mud samples

290

9.1.1 Categorisation and shear strength estimation conclusions:

This chapter has discussed the development of a model used to predict the friction angle of in-pit mud

based on the gravel and sand fractions of the material. The predicted friction angles show larger

variability for materials that contain large amounts of silt and clay-sized fractions. Overall, most

predictions varied between -13% and +8.5%. The statistical significance of the model was found to

be very high with an r2 value of 0.87, indicating a strong relationship between the parameters and the

friction angle. Each variable used in the multivariate regression was significant.

Calculations of the shear strength with an assumption of zero cohesion were compared with

laboratory-determined values. The largest variability above the laboratory-determined value was 3%.

For most materials, conservative results are calculated. For samples containing high amounts of silt

and clay, an increase is observed in the conservative nature of the predictions. For design, this is

desirable; however, improved accuracy is always beneficial.

To quickly estimate the shear strength of an in-pit mud material, this proposed model will provide

more accuracy than would be obtained based on an assumption of remoulded parameters from the

BMA spoil category framework.

By basing the predictions on material parameters that can be determined via wet sieving, estimations

can be made quickly and cheaply.

The key limitations and potential extensions of this model are:

• The sample size used for the multivariate regression analysis was 13. Future testing of in-pit

mud will allow for a larger sample size, which will be useful in improving the accuracy of

the model.

• As the model increases with inaccuracy in estimation of friction angle for materials with

high silt and clay content, its use for identification of material with more than 30% sand and

40% gravel is recommended.

291

9.2 Testing Methodology for Characterising In-Pit Mud

Accurate characterisation of in-pit mud is important for understanding its geotechnical behaviour

under different conditions and the influence it will have on lowwall stability. It will also allow for the

creation of a database with relevant material parameters, critical in further understanding indicators

for degradable material, to what degree degradation is likely to occur, and how to manage the mud

once it has formed. From the results of the research and analysis conducted, key material parameters

were identified relevant to the formation of the mud. To identify these parameters, the following tests

are recommended for accurate characterisation of in-pit mud prior to geotechnical testing.

Physical analysis:

• Moisture content (AS 1289.4.3.1 1997)

• Specific gravity (AS 1289.3.5.2 2002)

• Particle size distribution:

• Wet sieving (>0.075 mm fraction) (AS 1289.3.6.1 2009)

• Hydrometer (<0.075 mm fraction):

o with dispersant (AS 1289.3.6.3 2009)

o without dispersant (non-standardised)

• Atterberg limits:

o Liquid limit (AS 1289.4.3.1 1997)

o Plastic limit (AS 1289.3.2.1 2009)

o Plasticity index (AS 1289.3.5.2 2002)

Mineralogical and geochemical analysis:

• X-ray diffraction; and

• Exchangeable cations and cation exchange capacity (NH4+ exchange)

292

9.3 Stability Modelling of In-Pit Spoil and Mud

Within the Bowen Basin, several mining styles are used to gain access to coal seams below

overburden. These typically involve the use of trucks and shovels, dozers, and draglines of various

sizes to help move the overburden from the highwall to the lowwall post blasting. The presence of

degraded spoil and floor material referred to as mud between the highwall and lowwall can cause

issues with respect to logistics, material handling, and stability.

Due to the high rate of mining, costs of testing, and access difficulty for sampling, when present, any

degraded in-pit material is rarely sampled and tested to determine its physical characteristics and

geotechnical parameters. With respect to the spoil, the BMA spoil category framework that was

developed by Simmons and McManus (2004) allows for quick categorisation of spoil based mostly

on visual observations. For these categories, shear strength parameters have also been assigned based

on laboratory shear strength testing and the back-analysis of past failures. For the spoil tested,

envelopes were fitted to the data to create the categories and associated strengths listed in Table 9.7.

Table 9.7 BMA shear strength parameters for categories and mobilisation modes

SPOIL

CATEGORY


Unit

Weight

(kN/m2)

Cohesion

(kPa)

Friction

Angle

(deg)

Unit

Weight

(kN/m3)

Cohesion

(kPa)

Friction

Angle

(deg)

Cohesion =

0 kPa, Friction

Angle (deg)

1 18 1 20 1 25 2.5 20 1 0 18 3 18 1.5

2 18 1 30 15 28 3 20 1 15 7.5 23 2.5 18 1.5

3 18 1 50 15 30 2 20 1 20 10 25 2.5 18 1.5

4 18 1 50 15 35 2.5 20 1 0 30 1.5 28 2

The results of the direct shear testing for all spoil and mud materials collected are given in Table 9.8

and Table 9.9, split into dry (as-sampled) and wet (after 24 hours of soaking) results. For each

category, the average values of apparent cohesion and friction angle have also been calculated and

provided.

With these results, the implications of the variation in materials and the effect of weakened material

at the toe of a lowwall were investigated with respect to the lowwall stability. This is to replicate the

process that occurs in situ if the mud is to be covered by the next strip of overburden. Through

identification of the difference in the laboratory tested materials, and the estimated shear strengths

provided by the BMA framework, further progression can be made in the characterisation of spoil

and mud material, as well as methods for improved handling and understanding of weathered, in-pit

mud.

293

Table 9.8 Direct shear strength of spoil tested dry and wet


c’ (kPa) ’ (°) c’ (kPa) ’ (°)

C3S-10 CAT 3 SPOIL 97.3 30.4 23.1 34.5

C3S-13 CAT 3 SPOIL 63.8 31.9 27.9 30.5

C3S-20 CAT 3 SPOIL 76.3 38.1 0.0 29.6

CATEGORY 3 AVERAGE 79.1 33.5 17.0 31.5

C3S-16 CAT 2 SPOIL 150.7 30.0 109.2 22.9

C3S-24 CAT 2 SPOIL 138.9 21.8 15.9 27.6


C3S-17 CAT 1 SPOIL 118.8 30.9 13.9 26.4


Table 9.9 Direct shear strength of mud tested dry and wet


c’ (kPa) ’ (°) c’ (kPa) ’ (°)

C3M-01 CAT 3 MUD 11.4 28.5 41.3 26.7

C3M-03 CAT 3 MUD 14.6 32.2 25.6 31.1

C3M-04 CAT 3 MUD 35.2 32.8 41.1 32.4

C3M-05 CAT 3 MUD - - 12.3 30.0

C3M-08 CAT 3 MUD 20.9 35.9 49.2 35.6

C3M-12 CAT 3 MUD 22.3 35.6 26.0 36.0

C3M-18 CAT 3 MUD NA NA 82.0 13.2

C3M-30 CAT 3 MUD 13.8 36.9 38.6 34.8


C3S-06 CAT 2 MUD 26.9 28.7 31.2 25.5

C3S-07 CAT 2 MUD 0.0 34.9 0.1 29.1


C3S-02 CAT 1 MUD 0 27.0 0 26.0

C3S-23 CAT 1 MUD NA NA 57.6 15.6

C3S-32 CAT 1 MUD 4.6 31.9 23.1 26.2


The literature review discusses the potential failure mechanisms, with weakened floor material most

often resulting in multi-wedge failures involving an active and passive wedge. An example of this is

illustrated in Figure 9.11. To simulate this type of failure, all models were analysed using Sarma with

non-vertical slices with an assumption of non-circular failure. The back scarp angle was given a range

from 55 to 65o as is typically observed in failures in the Bowen Basin strip mines.

294

To assess the effects of different materials within a lowwall, two-dimensional slope stability

calculations were carried out using the commercial software Slide 7.0 (Rocscience Inc., 2013).

Models were created for four scenarios, to contrast the difference designs have on the stability of the

lowwall. Key features of these designs are provided in Table 9.10. The geometries of each lowwall

are shown in Figure 9.12 to Figure 9.15.

For the lowwall designed using a dozer, a maximum height of 88 m was chosen. Higher lowwall

designs were used for the other three methods based on standard construction practice and limitations

of the machinery. The highest lowwalls are created using a dragline, with the stability analysis

investigating a standard design, and an undercut toe scenario. To simplify comparisons, all models

have the same floor dip angle, water tables, failure assumptions and in-pit mud area dimensions.

Figure 9.11 Two-wedge spoil pile failure mechanism (adapted from Philip et al. 1981)

295

Table 9.10 Slide 7.0 model geometry and assumptions

PARAMETER DOZER

PUSH

TRUCK

AND

SHOVEL

DRAGLINE

DRAGLINE

WITH

UNDERCUT

TOE

Maximum height of spoil at the crest (m) 88 117 142 142

Dip of floor (deg) 7

The angle of repose of spoil (deg) 37

Length of material at the toe (m) 60

Height of material at the toe (m) 4

Height of water table (m) 0 to 5

Backscarp angle range (deg) 55-65

Unit weight of unsaturated material (kN/m3) 18. 0

Unit weight of saturated material (kN/m3) 20. 0

Method of analysis Sarma with non-vertical slices

Figure 9.12 Dozer push lowwall geometry

296

Figure 9.13 Truck and shovel lowwall geometry

Figure 9.14 Dragline lowwall geometry

Figure 9.15 Dragline lowwall with undercut toe geometry

Figure 9.15 involves the same geometry as Figure 9.14, but with an undercut toe. The undercutting

of the toe is conducted in practice when the coal seam is covered by spoil. For the purposes of this

investigation and in line with standard industry practice, the maximum height of undercutting is 10 m,

with an angle of 60o.

For each of these scenarios, several models were run, investigating the influence of a variety of

material combinations. Comparison has also been made to the results obtained using the BMA shear

strength framework. All sampled spoil materials were investigated under unsaturated (dry) and

saturated (wet) conditions, and with weakened mud at the toe of the lowwall. For the weakened mud,

two scenarios were investigated, using the remoulded shear strength assumption of the BMA

297

framework, and the average measured shear strength of laboratory tested muds of the same category,

as given in Table 9.1.

For all models analysed, the relationships between materials and the final factors of safety were

similar, with variable magnitudes dependent on design geometry. Figure 9.16 shows the final stability

analysis results of a lowwall constructed using the dragline technique illustrated in Figure 9.14.

The results show that for all spoils tested in each category, the laboratory tested materials had higher

factors of safety than the assumed shear strength parameters provided by the BMA framework. For

all materials tested, lower factors of safety resulted from saturated material at the toe of the lowwall.

It was also found that the laboratory mud average shear strength produced significantly higher factors

of safety, with the largest variation observed in the Category 3 materials.

All scenarios except for spoil material C2S-16 had a FOS below 1.3 for the BMA remoulded base

conditions. During a design phase, it would, therefore, be likely these designs would be considered

unsafe, and the degraded spoil/mud would have to be removed prior to dumping. When using the

results of the laboratory testing, however, for all materials tested in the laboratory, factors of safety

over 1.3 are achieved with wet and dry spoil at the base of the lowwall, as well as the tested average

laboratory muds. Using the framework values only, the Category 2 and 1 spoil designs have factors

of safety below 1.3 for unsaturated and saturated spoil at the base of the lowwall. In contrast to the

tested values, this could potentially lead to conservative designs.

298

Table 9.11 Factor of safety for all spoil samples with different materials at toe of lowwall

MATERIAL CATEGORY

FACTOR OF SAFETY

Unsaturated /

Dry

Saturated /

Wet BMA Remoulded

Laboratory Mud

Average

Truck and Shovel Lowwall

BMA-C3 CAT 3 734 1.601 1.160 1.722

C3S-13 CAT 3 1.923 1.864 1.253 1.876

C3S-20 CAT 3 2.408 1.866 1.382 2.178

BMA-C2 CAT 2 1.511 1.353 1.034 1.486

C2S-16 CAT 2 2.232 2.141 1.662 2.068

C2S-24 CAT 2 1.707 1.623 1.517 1.620

BMA-C1 CAT 1 1.209 0.948 0.946 1.282

C1S-17 CAT 1 2.088 1.882 1.463 1.818

Dozer Push Lowwall

BMA-C3 CAT 3 1.708 1.386 1.043 1.65

C3S-13 CAT 3 1.992 1.712 1.148 1.783

C3S-20 CAT 3 2.455 1.657 1.294 1.959

BMA-C2 CAT 2 1.356 1.138 0.896 1.254

C2S-16 CAT 2 2.584 2.394 1.626 2.037

C2S-24 CAT 2 2.011 1.874 1.481 1.868

BMA-C1 CAT 1 1.105 0.801 0.798 1.1

C1S-17 CAT 1 2.374 1.741 1.397 1.710

Dragline Lowwall

BMA-C3 CAT 3 1.471 1.301 1.001 1.461

C3S-13 CAT 3 1.633 1.571 1.102 1.575

C3S-20 CAT 3 2.044 1.577 1.236 1.769

BMA-C2 CAT 2 1.280 1.090 0.872 1.196

C2S-16 CAT 2 1.886 1.793 1.454 1.725

C2S-24 CAT 2 1.431 1.379 1.272 1.376

BMA-C1 CAT 1 1.056 0.786 0.783 1.039

C1S-17 CAT 1 1.779 1.571 1.291 1.527

Dragline Lowwall Undercut at Toe

BMA-C3 CAT 3 1.468 1.298 1.008 1.457

C3S-13 CAT 3 1.626 1.595 1.099 1.569

C3S-20 CAT 3 2.037 1.586 1.233 1.813

BMA-C2 CAT 2 1.228 1.080 0.868 1.152

C2S-16 CAT 2 1.898 1.802 1.455 1.743

C2S-24 CAT 2 1.433 1.399 1.278 1.387

BMA-C1 CAT 1 1.006 0.778 0.776 0.993

C1S-17 CAT 1 1.794 1.594 1.286 1.517

299

Figure 9.16 Factor of safety for dragline lowwall with different base material at the toe

During sampling, two spoil materials were collected that had associated in-pit mud. The shear

strengths of these materials and sampling locations are discussed in detail in Section 8.2. Using these

values as the material parameters in each of the models, the differences in factor of safety are

compared with the design using values from the BMA shear strength framework. The results of the

modelling are given in Table 9.12 and are plotted in Figure 9.17 for the dragline scenario.

The results in Figure 9.17 show in all scenarios. The tested Category 3 spoil’s had higher factors of

safety than the assumed BMA shear strength framework. With a remoulded base, a factor of safety

below 1.3 was calculated for all scenarios. For C3S-13 spoil and C3M-08 as the mud, significantly

higher factors of safety were calculated than assumed with the framework. This was also observed

for C3S-20 and C3M-18.

300

Table 9.12 Factor of safety for spoil with associated mud at the toe of lowwall

MATERIAL CATEGORY

FACTOR OF SAFETY

Unsaturated /

Dry

Saturated /

Wet BMA Remoulded Associated Mud

Truck and Shovel Lowwall

BMA-C3 CAT 3 1.734 1.601 1.160 1.160

C3S-13 CAT 3 1.923 1.864 1.253 1.946

C3S-20 CAT 3 2.408 1.866 1.382 1.840

Dozer Push Lowwall

BMA-C3 CAT 3 1.734 1.601 1.160 1.160

C3S-13 CAT 3 1.923 1.864 1.253 2.000

C3S-20 CAT 3 2.408 1.866 1.382 1.748

Dragline Lowwall

BMA-C3 CAT 3 1.734 1.601 1.160 1.160

C3S-13 CAT 3 1.923 1.864 1.253 1.675

C3S-20 CAT 3 2.408 1.866 1.382 1.451

Dragline Lowwall Undercut at Toe

BMA-C3 CAT 3 1.734 1.601 1.160 1.160

C3S-13 CAT 3 1.923 1.864 1.253 1.707

C3S-20 CAT 3 2.408 1.866 1.382 1.429

Figure 9.17 Factor of Safety for Category 3 spoil with associated mud at the toe of a dragline

lowwall

301

Table 9.13 details the stability analysis results of the laboratory tested mud, with assumed spoil

parameters obtained from the BMA shear strength framework, to highlight the variation within the

mud materials source categories, and between categories. Figure 9.18 illustrates these results for all

tested materials and scenarios. For most materials tested, the safety factors were the highest for truck

and shovel and the lowest for the dragline scenarios.

For all Category 3 mud except for C3M-18, the factors of safety were all above 1.3. C3M-18 was

below 1.3 for scenarios involving the dragline, but above 1.3 for the dozer push and truck and shovel

designs. For the two Category 2 muds investigated, both had factors of safety above 1.3 for all

designs. Varied results were obtained with the Category 1 muds. For the dragline designs, C1M-02

and C1M-23 were both below 1.3. On average, the Category 3 mud’s had higher factors of safety

than the Category 2 mud, which were higher than the Category 1 mud.

The lowest results were obtained using the BMA remoulded shear strength values, with unacceptable

safety factors being calculated in all cases. The results of this modelling and the shear strength testing

conducted in the laboratory show that assumptions made for in-pit mud have the potential to be highly

conservative.

302

Table 9.13 Factor of safety for Category 3 spoil with laboratory tested muds at the toe of

lowwall

SPOIL MUD

FACTOR OF SAFETY

Truck and

Shovel Dozer Push Dragline

Dragline

Undercut

BMA-C3 C3M-01 1.695 1.574 1.416 1.412

BMA-C3 C3M-03 1.714 1.59 1.458 1.491

BMA-C3 C3M-04 1.749 1.727 1.499 1.536

BMA-C3 C3M-05 1.685 1.471 1.394 1.392

BMA-C3 C3M-08 1.79 1.87 1.573 1.587

BMA-C3 C3M-12 1.765 1.737 1.536 1.568

BMA-C3 C3M-30 1.769 1.781 1.532 1.565

BMA-C3 C3M-18 1.584 1.485 1.077 1.185

BMA-C3 Remoulded 1.16 1.043 1.001 1.008

BMA-C3 C2M-06 1.685 1.475 1.366 1.369

BMA-C2 C2M-06 1.492 1.31 1.24 1.185

BMA-C3 C2M-07 1.59 1.36 1.308 1.303

BMA-C2 C2M-07 1.449 1.2 1.158 1.126

BMA-C2 Remoulded 1.034 0.896 0.872 0.868

BMA-C3 C1M-02 1.47 1.271 1.217 1.217

BMA-C1 C1M-02 1.234 1.009 0.965 0.982

BMA-C3 C1M-23 1.499 1.388 1.077 1.172

BMA-C1 C1M-23 1.272 1.147 0.988 0.897

BMA-C3 C1M-32 1.662 1.439 1.349 1.345

BMA-C1 C1M-32 1.306 1.165 1.11 1.061

BMA-C1 Remoulded 0.946 0.798 0.783 0.776

Figure 9.19 compares the influence the spoil strength above the mud has on the stability of the

lowwall for the Category 2 and 1 material. While large variations can occur within a pit, it is likely

the mud formed during a flooding event will be related to the spoil within the lowwall if properly

categorised. The use of a Category 3 spoil provided a higher factor of safety for the lowwall designs.

When using the muds associated spoil within the model, none of the Category 2 or 1 muds had factors

of safety above 1.3. This further highlights the importance of correct classification of spoil in situ.

303

Figure 9.18 Factor of safety for Category 3 spoil with different mud at the toe of lowwall

Figure 9.19 Factor of safety comparison for Category 3 spoil against mud equivalent spoil

with different mud at toe of lowwall

304

9.3.1 Stability modelling conclusions:

The implications of spoiling onto in-pit mud were investigated with respect to four different lowwall

designs, produced using a truck and shovel, a dozer push, and a dragline with both standard and

undercut toes. The stability of each design was calculated using the assumed shear strength

parameters of spoil and mud obtained from the BMA shear strength framework and the associated

strength mobilisation mode as per recommended by Simmons (2009), with unsaturated conditions

assumed for spoil above the water table, and saturated conditions for spoil below. This was contrasted

with the shear strengths determined through laboratory testing.

Results of the stability analysis with respect to spoil in unsaturated, saturated and remoulded

conditions show that for all tested spoils, higher factors of safety were achieved using the laboratory

tested strength values highlighted in Figure 9.16. For all spoil materials tested, dry and wet conditions

produced factors of safety above 1.3. The assumption of a remoulded base at the toe of the lowwall

resulted in safety factors below 1.3 for all scenarios except for C2S-16.

Figure 9.16 shows that for calculations using the average shear strength of the laboratory tested muds

associated with the spoil assigned its BMA spoil category strength parameters, factors of safety for

all materials tested were above 1.3. This highlights the potential for conservative lowwall designs

based on assuming the BMA framework suggested remoulded shear strength parameters. It was also

observed that for Category 2 and 1 spoil using the BMA shear strength design values, all scenarios

produced factors of safety below 1.3.

Further analysis was conducted for two spoil materials with associated in-pit mud. The factors of

safety obtained using the laboratory tested spoil and mud shear strength parameters were significantly

higher than that of the remoulded assumptions, depicted in Figure 9.17.

A set of models were computed using BMA’s Category 3 spoil shear strength assumptions, and the

laboratory tested mud materials at the base of the toe of each lowwall design, with results plotted in

Figure 9.18. Except for one outlier, all Category 3 muds produced safety factors above 1.3 in all

scenarios. Lower safety factors were calculated for the Category 2 and 1 mud materials.

For the Category 2 and 1 material, models were computed using BMA spoil material strength values

associated with the material category of the laboratory tested mud. These results were contrasted to

the Factor of Safety generated if the spoil was incorrectly categorised as a Category 3 material. A

reduction in the safety factor was calculated, resulting in all the Category 2 and 1 scenarios having a

safety factor below 1.3. This highlights the importance of correct spoil and mud in situ categorisation.

305

The results of this modelling highlight the possibility of safety spoiling onto in situ mud. If the mud

has enough strength and can consolidate without the build-up of excess pore water pressures, the

factor of safety is not drastically influenced by its presence. This scenario is most like for in-pit mud

formed from Category 3 spoil, that exhibits limited slaking and dispersion. Assuming remoulded

conditions for in situ mud results in moderate to highly conservative designs, and for the tested

Category 3 mud, could result in unnecessary material handling and expenses.

306

10 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE

RESEARCH

The fundamental reason for this research was to investigate the parameters of in-pit mud and how

they influence the stability of lowwalls within an open-cut coal strip mining scenario. The objectives

were to identify spoil prone to degradation, determine parameters that predict its degradation, and

identify the geotechnical properties of the degraded mud both simulated and sampled. In addition to

this, the objectives included modelling of lowwalls to determine if mud could be spoiled upon in situ

if managed correctly.

10.1 Conclusions and Significant Outcomes

This research thesis has involved an extensive amount of laboratory testing, fully characterising 19

samples consisting of spoil and mud. Valuable contributions have been made to the understanding of

how mud is formed in situ, how it behaves once formed, and the impact it has on the stability of coal

strip mine lowwalls, and spoil piles in general. The degradation of spoil and its resultant strength have

also been investigated, with methods developed to identify degradable material in a laboratory setting

quickly. Lastly, a model has been developed allowing for quick predictions of the effective friction

angle of in-pit mud, and for an assumption of zero apparent cohesion, the shear strength of the

material in drained conditions using particle size distribution results.

10.1.1 Spoil and in-pit mud characterisation

Thirteen mud samples and six spoil samples were collected and analysed with respect to their

physical, chemical, mineralogical and geochemical parameters. The results of the testing were used

to identify relationships between the assigned BMA spoil categories of the materials and the results

determined in the laboratory.

The gravimetric moisture content of the sampled spoil varied from 1.1 to 3.3%, with one spoil

material collected below a previous flooding level having a moisture content of 9.5%, indicating the

ability for spoil to retain moisture at the surface post-flooding. Of the muds sampled, moisture

contents were recorded from below the plastic limit to above the liquid limit of the materials. The

specific gravities of all spoil and mud materials ranged from 2.2 to 2.7 t/m3, with lower values

associated with increases in carbonaceous material. The Atterberg limits were determined, with

plastic and liquid limit ranges of 12.1 to 27.1%, and 21.5 to 61.9%, respectively. All materials were

classified as clays of low plasticity excluding two outliers identified as clays of high plasticity. The

results indicate potential refinement of the current liquid limit ranges within the BMA spoil category

framework is required.

307

The measured pH for the spoil and mud ranged from 7.75 to 9.27, and the electrical conductivities

ranged from 246 to 8,920 µS/cm. There was no apparent relationship between the source materials

category and the measured results. All spoil materials had Emerson classes of 2 or 4, indicating

moderate and high to a very high potential for erosion, respectively. The implications of this show

that there is spoil that can degrade rapidly on exposure to water, irrespective of initial conditions such

as particle size.

All materials were dominated by quartz, Illite-Smectite, Albite and Kaolinite, with no clear

relationship between composition and source material category. The measured cation exchange

capacities were from 10 and 21 cmol+/kg. The most common exchangeable cations were magnesium,

sodium and calcium. An approximation of the presence of sodium Smectite was calculated, with

results showing the highest levels associated with Category 1 mud, and one Category 3 mud outlier.

With respect to particle size distribution, the Category 3 spoil was on average coarser than the

Category 2 spoil, which was, in turn, coarser than the Category 1 spoil. The same relationship was

observed with the mud formed from categorised source materials; however, less differentiation was

observed between the Category 1 and 2 source muds. Varying amounts of degradation were observed

during wet sieving within each category, indicating that categorisation of the material in a dry state

could lead to incorrect assumptions of the competence of the material.

10.1.2 Degradation of spoil

Five of the spoil samples had not been exposed to flooding prior to sampling. Each of these materials

was tested under multiple conditions to determine the degree of degradation experienced and to relate

the degradation to the materials physical, chemical and mineralogical parameters, with intentions of

being able to identify highly degradable material before it is wet up.

Two spoil samples underwent a series of prolonged saturation and wetting and drying cycle tests.

Wetting and drying cycles resulted in higher amounts of particle breakdown, indicating the clay-

mineral rich spoil’s potential for slaking, swelling and dispersion in scenarios of cycling flooding or

heavy rainfall. For both spoil materials, the largest degree of breakdown was observed within the first

24 hours of soaking. This has significant implications for ensuring degradable materials are kept away

from sources of water if possible. Increasing numbers of wetting and drying cycles were found to

cause large increases in the measured electrical conductivities of both materials.

Based on the degradation results, all spoil samples were then subjected to dry sieving, 24 hours of

soaking, three wetting and drying cycles, and a modified slake durability test. It was found that in

decreasing order of importance, the factors related to the degree of degradation were clay mineralogy

of the spoil, the exchangeable cations, the cation exchange capacity and the liquid limit. The results

308

of the modified slake durability test were found to be repeatable and showed potential for the method

to be used to quickly identify spoil prone to degradation in water.

10.1.3 Consolidation of spoil and mud

All spoil and mud samples underwent consolidation testing with loose placement to simulate in situ

conditions. Final dry densities under an applied stress of 1,000 kPa ranged from 1.65 to 1.97 t/m3 for

the tested spoil, with the highest densities obtained from the Category 3 specimens. Final dry densities

of the mud ranged from 1.54 to 1.97 t/m3, with no apparent relation to the source materials category.

On average, the lowest void ratios were found with the Category 3 mud, followed by the Category 2

and 1 mud.

For the spoil specimens tested, settlement ranged from 20 to 35%. For the mud specimens tested,

settlement generally ranged from 8 to 35%, with two significant outliers of 55% and 60%. Typically,

less settlement was observed for the Category 3 materials.

The compression index ranged from 0.13 to 0.78 for the spoil specimens tested, and generally much

flatter from 0.025 to 0.113 for the mud specimens tested. Two significant mud outliers had values of

0.216 and 0.258. Back-calculated hydraulic conductivities ranged from 1.4x10-09 to 0.9x10-11 m/s

under an applied stress of 1,000 kPa. The Category 3 materials tested had on average higher hydraulic

conductivities than the Category 2 and 1 materials; however, considerable variability was observed.

In the absence of scalping, it is likely that typically, coarser particles in the Category 3 materials

would increase the hydraulic conductivity, suggesting that the calculated values are lower limits.

The large slurry consolidometer was used to analyse four mud samples. For mud with a fine-grained

particle size distribution, pore water pressures developed rapidly, with initial dissipation observed to

start after between 1.5 days and three days under a constant loading rate of 0.1 kPa/min. Full pore

water pressure dissipation took 6 to 8 days. One Category 3 mud, with a relatively coarse particle size

distribution of silt, gravel and sand, showed no pore water pressures. Hydraulic conductivity values

back-calculated from the dissipation of the pore water pressures reduced to 3.1x10-09 m/s towards the

end of loading, with higher conductivities likely with faster loading rates. The other three mud

specimens tested, having finer particle size distributions, gave much lower final hydraulic

conductivity values ranging from 4.7x10-11 to 8.2x10-11 m/s.

Results of the large slurry consolidometer show that coarse-grained muds show the most potential for

being spoiled onto in situ without developing detrimental pore water pressures, or without requiring

a long waiting period for pore pressures to dissipate. For the fine-grained muds tested, careful loading

would be required to allow for pore water pressure dissipation, using methods such as a dozer. If

309

loading is to occur too rapidly, the pore water pressures produced will drastically reduce the shear

strength of the material.

10.1.4 Shear strength of spoil and mud

All spoil and mud samples were tested both dry (as-sampled) and wet (after soaking for 24 hours in

a water bath prior to shearing). On average, the dry testing produced the largest shear strengths.

Soaking resulted in a decrease in both apparent cohesion and friction angle in most cases. Typically,

the shear strength of the spoil tested dry was higher than the shear strengths suggested by the BMA

Framework. The wet tested spoil was closer to the unsaturated BMA Framework shear strength,

indicating that for almost all cases, the BMA Framework was conservative to highly conservative.

On average, the Category 3 mud had higher shear strength than the Category 2 mud, which was higher

than the Category 1 mud. Two outliers were identified with extremely low friction angles of 12 to

15o, but high apparent cohesion values of up to 100 kPa. Most muds had friction angles above 25o,

with some up to 36o, with the highest values associated with Category 3 mud relatively high in sand

and gravel content. The majority of the tested materials had friction angles higher than the typical

assumption of 18o that is used in the industry, due to remoulded parameters being used for wet

conditions in the absence of laboratory testing results.

All spoil samples underwent wetting and drying cycles to cause degradation and to determine the

influence that would have on shear strength. The results show that the largest decrease in strength

occurs during the first 24 hours of wetting. After three wetting and drying cycles, the largest change

was observed in the silt and clay-sized fractions. The sand and gravel in most cases remained similar,

as did the tested shear strength. This indicates that once a spoil pile has been saturated, there will be

an irreversible loss of strength due to degradation.

10.1.5 Categorisation, shear strength estimation and modelling of in-pit mud

The use of the BMA Framework categories for in-pit mud and degraded material can lead to

conservative shear strengths used in design. Instead, a model was developed and proposed for the

estimation of in-pit mud friction angles. The model had an r2 value of 0.87, indicating a very strong

relationship using the gravel and sand-sized fractions to predict the friction angle of the mud.

Variability between the tested and predicted values for most in-pit mud materials ranged from -13%

to +8.5%, with increasing variability related to samples with higher percentages of silt and clay-sized

particles. Calculations using the predicted friction angles and an assumption of zero cohesion showed

that in comparison to the tested shear strength at 1,000 kPa stress, predictions were within 10%, with

the largest overestimation being 3% higher. Two materials underestimated the shear strength of the

material by up to 30%, both containing high silt and clay-sized fractions. With respect to design for

310

materials containing high silt and clay-sized fractions, conservative results are desirable as instability

can be highly dangerous and costly to remediate. Further sampling and testing would result in the

refinement of the model to improve shear strength estimates for fine-grained materials.

For the characterisation of in-pit mud and spoil, the recommended test methodology is provided for

materials to be characterised physically, chemically, mineralogically and geochemically. By

conducting these tests, significant parameters related to the geotechnical characteristics of the

material can be identified.

Four lowwall designs were analysed using the slope stability software Slide 7.0. The results of the

testing show that the laboratory tested shear strength parameters resulted in higher Factors of Safety

than the BMA framework suggested shear strengths. For all spoil materials tested, use of the BMA

framework remoulded shear strength parameters resulted in factors of safety below 1.3.

With the laboratory tested mud shear strengths applied at the base of the toe of the lowwall, factors

of safety greater than 1.3 were obtained for almost all materials tested. The results of the modelling

show potential for spoiling into in-pit mud if the shear strengths are given time to develop through

the dissipation of pore water pressures if they develop. Spoiling into in-pit mud is most possible for

mud formed from Category 3 spoil. For mud derived from spoil containing high silt and clay-sized

fractions, slow loading and extended periods between progressive strips would be required to allow

for adequate time for the pore water pressures to dissipate, likely requiring dozer or truck and shovel

operations rather than dragline or highwall blasting.

311

10.2 Recommendations for Future Research

Due to the continuous nature of strip mining and the potential risks associated with a weakened layer

at the base of a spoil pile, research into in-pit mud is critical in avoiding potential instabilities and

unnecessary costs. From the results of this project, and with future characterisation and modelling of

in-pit mud, there is potential for controlled spoiling onto in-pit mud. From this research project, the

following recommendations are made for future work.

Early identification of degradable material:

• Comparison of in-pit mud characterisation results to past borehole logs could allow for early

identification of poor-quality material, allowing it to be accounted for within design.

• Implementation of the modified slake durability test for analysing spoil, highlighting

material prone to degradation and comparing it to associated in-pit mud.

In-pit mud characterisation:

• Further categorisation of in-pit mud. The sample size for this project was 13, with a focus on

Category 3 materials. Future testing expanding the current set of results will allow for

refinement of material parameters and management techniques.

• Large slurry consolidation testing simulating spoiling into mud under different loading rates,

determining the impact of different spoil handling techniques on pore water pressures within

the mud, and the influence these have on stability.

Refinement of the proposed model for estimating in-pit mud friction angle:

• Refinement of the proposed model for predicting the friction angle of in-pit mud, with a

focus on improved accuracy for materials with high silt and clay-sized fractions.

Trial spoiling into mud and associated modelling:

• Trial spoiling into mud in situ is recommended, using dragline, truck and shove and dozing

methods to advance the spoil. There is potential for spoiling into in-pit mud if conducted at a

sufficiently slow rate, allowing excess pore water pressures to dissipate. A successful trial

could result in substantial cost savings in not having to remove in-pit mud.

• In association with a trial, consolidation and pore water pressure measurement would be

required during loading, and modelling of the results would be highly beneficial in

improving understanding of the response of the mud during loading.

312

11 REFERENCES

Abbaspour, H, Drebenstedt, C, Badroddin, M & Maghaminik, A 2018, ‘Optimized design of drilling

and blasting operations in open pit mines under technical and economic uncertainties by system

dynamic modelling’, International Journal of Mining Science and Technology, vol. 28, no. 6, pp.

839–848.

AS 1289.2.1.1 2005, ‘Methods of testing soils for engineering purposes - Soil moisture content tests

- Determination of the moisture content of a soil - Oven drying method (standard method)’, in

Standards Australia, Sydney: Standards Australia Limited.

AS 1289.2.2.1 1998, ‘Methods of testing soils for engineering purposes - Soil moisture content tests

- Determination of the total suction of a soil - Standard method’, in Standards Australia, Sydney:

Standards Australia Limited.

AS 1289.3.1.1 2009, ‘Methods of testing soils for engineering purposes - Soil classification tests -

Determination of the liquid limit of a soil - Four point Casagrande method’, in Standards Australia,

Sydney: Standards Australia Limited.


Determination of the plastic limit of a soil - Standard method’, in Standards Australia, Sydney:



Calculation of the plasticity index of a soil’, in Standards Australia, Sydney: Standards Australia

Limited.

AS 1289.3.5.2 2002, ‘Methods of testing soils for engineering purposes Soil classification tests -

Determination of the soil particle density of combined soil fractions - Vacuum pycnometer method’,

in Standards Australia, Sydney: Standards Australia Limited.


Determination of the particle size distribution of a soil - Standard method of analysis by sieving’, in

Standards Australia, Sydney: Standards Australia Limited.


Determination of the particle size distribution of a soil - Standard method of fine analysis using a

hydrometer’, in Standards Australia, Sydney: Standards Australia Limited.


Dispersion - Determination of Emerson class number of a soil’, in Standards Australia, Sydney:


313

AS 1289.4.3.1 1997, ‘Methods of testing soils for engineering purposes - Soil chemical tests -

Determination of the pH value of a soil - Electrometric method’, in Standards Australia, Sydney:


AS 1289.4.4.1 2017, ‘Methods of testing soils for engineering purposes - Soil chemical tests -

Determination of the electrical resistivity of a soil - Method for fine granular materials’, in Standards

Australia, Sydney: Standards Australia Limited.

AS 1289.6.2.2 1998, ‘Methods of testing soils for engineering purposes - Soil strength and

consolidation tests - Determination of shear strength of a soil - Direct shear test using a shear box’,

in Standards Australia, Sydney: Standards Australia Limited.

AS 1289.6.6.1 1998, ‘Methods of testing soils for engineering purposes - Soil strength and

consolidation tests - Determination of the one-dimensional consolidation properties of a soil -

Standard method’, in Standards Australia, Sydney: Standards Australia Limited.

AS 4133.3.4 2005, ‘Methods of testing rocks for engineering purposes - Rock swelling and slake

durability tests - Determination of the slake durability index of rock samples’, in Standards Australia,

Sydney: Standards Australia Limited.

ASTM D2487-11 2011, Standard Practice for Classification of Soils for Engineering Purposes

(Unified Soil Classification System), ASTM International, West Conshohocken, PA, accessed from

<www.astm.org>.

ASTM D3744 / D3744M-18 2018, Standard Test Method for Aggregate Durability Index, ASTM

International, West Conshohocken, PA, accessed from <www.astm.org>.

Atwood, G 1975, ‘The Strip-mining of Western Coal’, Scientific American, vol. 233, no. 6, pp. 23–

29.

Babalola, Z 2016, Direct shear and direct simple shear tests: a comparative study of the strength

parameters and their dependence on moisture and fines contents, University of Cape Town.

Beakawi Al-Hashemi, HM & Baghabra Al-Amoudi, OS 2018, ‘A review on the angle of repose of

granular materials’, Powder Technology, vol. 330, pp. 397–417.

Beale, G 2017, ‘Surface water and groundwater characterisation’, in Guidelines for Mine Waste

Dump and Stockpile Design, CSIRO Publishing, Collingwood, pp. 99–118.

Birkeland, P 1999, Soils and Geomorphology 3 edition., Oxford University Press, New York.

Boyd, GL, Komdeur, W & Richards, BG 1978, ‘Open strip pitwall instability at Goonyella mine -

causes and effects’, in Proceedings of The Australasian Institute of Mining and Metallurgy

314

Conference, The Australasian Institute of Mining and Metallurgy, Melbourne, Australia, pp. 139–

157.

Bradfield, L 2017, Reliable Shear Strength Estimation for Very-High Spoil Dumps, The University

of Newcastle.

Bradfield, L, Simmons, J & Fityus, S 2013, ‘Issues related to stability design of very high spoil

dumps’, in 13th Coal Operators’ Conference, The Australasian Institute of Mining and Metallurgy

and Mine Managers Association of Australia, University of Wollongong, pp. 376–386.

Brand, EW & Brenner, RP 1981, Soft clay engineering, Elsevier Science.

Cano, M & Tomás, R 2016, ‘Proposal of a New Parameter for the Weathering Characterization of

Carbonate Flysch-Like Rock Masses: The Potential Degradation Index (PDI)’, Rock Mechanics and

Rock Engineering, vol. 49, no. 7, pp. 2623–2640.

Carson, MA & Petley, DJ 1970, ‘The Existence of Threshold Hillslopes in the Denudation of the

Landscape’, Transactions of the Institute of British Geographers, no. 49, pp. 71–95.

Carter, M & Bentley, SP 1991, Correlations of soil properties, London Pentech.

Cheema, T, Al-Harthy, A, Al Aghbari, M & Al-Aufi, M 2004, ‘Influence of Grain Size and

Mineralogy on the Strength of Weak Rocks’, in Geotechnical Practice Publication, pp. 239–249.

Chigira, M & Oyama, T 2000, ‘Mechanism and effect of chemical weathering of sedimentary rocks’,

Developments in Geotechnical Engineering, vol. 55, pp. 267–278.

Chowdhury, RN & Nguyen, VU 1987, ‘Risk updating for rainfall-triggered spoil failures’,

Proceedings 2nd International Conference on Reclamation, Treatment and Utilisation of Coal

Mining Wastes, pp. 95–109.

Christoulas, S, Kalteziotis, N, Tsiambaos, G & Sabatakakis, N 1987, ‘Engineering Geology of Soft

Clays, Examples from Greece’, in Embankments on soft clays, Athens : Public Works Research

Center.

Das, B & Sobhan, K 2017, Principles of Geotechnical Engineering 9th edn, Cengage Learning US.

Davison, LR & Atkinson, JH 1990, ‘Continuous loading oedometer testing of soils’, Quarterly

Journal of Engineering Geology and Hydrogeology, vol. 23, no. 4, pp. 347–355.

Deo, P 1972, ‘Shales as Embankment Materials’, JTRP Technical Reports.

Dhakal, G, Yoneda, T, Kato, M & Kaneko, K 2002, ‘Slake durability and mineralogical properties of

some pyroclastic and sedimentary rocks’, Engineering Geology, vol. 65, no. 1, pp. 31–45.

315

Dorador, L, Cunning, J, Junqueira, F & Hawley, M 2017, ‘Material Characterisation’, in Guidelines

for Mine Waste Dump and Stockpile Design, CSIRO Publishing, Collingwood, pp. 79–97.

Dumbleton, MJ 1981, The British Soil Classification System for Engineering Purposes: Its

Development and Relation to Other Comparable Systems, Transport and Road Research Laboratory.

Dunbavan, M & Welsh, TC 1982, ‘Spoil Pile Instability, South Blackwater Mine Queensland’,

Commonwealth Scientific and Industrial Research Organisation, Institute of Energy and Energy

Resources, Division of Applied Geomechanics, vol. Report No. 35.

Dunbaven, M & Driver, IJ 1988, ‘Lowwall Stabilisation at Riverside Coal Mine’, Fifth Australia-

New Zealand Conference on Geomechanics: Prediction Versus Performance; Preprints of Papers,

p. 368.

Duran, A 2013, ‘Undrained behaviour in spoil piles’, in P Dight (ed), Proceedings of the 2013

International Symposium on Slope Stability in Open Pit Mining and Civil Engineering, Australian

Centre for Geomechanics, Perth, pp. 853–865.

Eberl, DD, Farmer, VC & Barrer, RM 1984, ‘Clay Mineral Formation and Transformation in Rocks

and Soils [and Discussion]’, Philosophical Transactions of the Royal Society of London. Series A,

Mathematical and Physical Sciences, vol. 311, no. 1517, pp. 241–257.

Emerson, WW 1967, ‘A classification of soil aggregates based on their coherence in water’, Soil

Research, vol. 5, no. 1, pp. 47–57.

Erguler, ZA & Ulusay, R 2009, ‘Assessment of physical disintegration characteristics of clay-bearing

rocks: Disintegration index test and a new durability classification chart’, Engineering Geology, vol.

105, no. 1, pp. 11–19.

Fielding, CR, Kassan, J, Draper, JJ & Geological Society of Australia. Australasian Sedimentologists

Group 1996, Geology of the Bowen and Surat Basins, Eastern Queensland / Christopher R. Fielding,

Jochen Kassan, John J. Draper., Geological Society of Australia, Springwood, N.S.W.

Fityus, S, Hancock, G & Wells, T 2008, ‘Geotechnical characteristics of coal mine spoil’, Australian

Geomechanics, vol. 43, no. 2, pp. 13–22.

Fookes, PG, Dearman, WR & Franklin, JA 1971, ‘Some engineering aspects of rock weathering with

field examples from Dartmoor and elsewhere’, Quarterly Journal of Engineering Geology and

Hydrogeology, vol. 4, no. 3, pp. 139–185.

Franklin, JA & Chandra, R 1972, ‘The slake-durability test’, International Journal of Rock Mechanics

and Mining Sciences & Geomechanics Abstracts, vol. 9, no. 3, pp. 325–328.

316

Fuller, PG & Cox, RHT 1987, Highwall and spoil pile failure processes identified by sub-surface

instramentation at Goonyella Mine, Commonwealth Scientific and Industrial Research Organisation

(CSIRO), Division of Applied Geomechanics.

Gómez, P, Díaz, M & Lorig, L 2002, ‘Stability analysis of waste dumps at Chuquicamata Mine,

Chile’, , vol. 20, pp. 61–68.

Gonano, LP 1980, An integrated report on slope failure mechanisms at Goonyella, November 1976,

Commonwealth Scientific and Industrial Research Organization, Institute of Earth Resources,

Division of Applied Geomechanics, Mt. Waverley, Vic.

Graupner, T, Kassahun, A, Rammlmair, D, A. Meima, J, Kock, D, Furche, M, Fiege, A, Schippers,

A & Melcher, F 2007, ‘Formation of sequences of cemented layers and hardpans within sulfide-

bearing mine tailings (mine district Freiberg, Germany)’, Applied Geochemistry, vol. 22, pp. 2486–

2508.

Hartman, HL & Mutmansky, JM 2002, Introductory Mining Engineering, John Wiley & Sons.

Hiung, AKK 2016, Bulking and Subsequent Self-Weight and Saturation Settlements, and

Geotechnical Stability of High Coal Mine Spoil Piles, The University of Queensland, School of Civil

Engineering.

Humphrey, J. D 1984, ‘The Fundamentals of the Dragline’, in Marion Division-Dresser Ind. Inc:

Marion, p. 29.

Hustrulid, WA, McCarter, MK & A, VZ, Dirk J (eds) 2001, Slope Stability in Surface Mining, Society

for Mining, Metallurgy, and Exploration, Littleton, Colo.

Jackson, R 2015, ‘Manual of Soil Laboratory TestingBook Review’, Environmental and Engineering

Geoscience, vol. 21, no. 3, pp. 247–248.

Janbu, N, Tokheim, O & Senneset, K 1980, ‘Consolidation tests with continuous loading’, Norvegian

Geotech Institute.

Karem, WA 2005, Development of a Predictive Model to Evaluate Mine Spoil Fills for Industrial

Development, University of Kentucky.

Kho, AK, Williams, DJ, Kaneko, N & Smith, NJW 2013, Shear strength parameters for assessing

geotechnical slope stability of open pit coal mine spoil based on laboratory tests, Australian Centre

for Geomechanics.

Kikumoto, M, Putra, AD & Fukuda, T 2016, ‘Slaking and deformation behaviour’, Géotechnique,

vol. 66, no. 9, pp. 771–785.

317

Koloski, JW, Schwarz, SD & Tubbs, DW 1989, ‘Geotechnical Properties of Geologic Materials’, in

Engineering Geology in Washington, Washington Division of Geology and Earth Resources, pp. 19–

26.

Koncagül, EC & Santi, PM 1999, ‘Predicting the unconfined compressive strength of the Breathitt

shale using slake durability, Shore hardness and rock structural properties’, International Journal of

Rock Mechanics and Mining Sciences, vol. 36, no. 2, pp. 139–153.

Lan, HX, Hu, RL, Yue, ZQ, Lee, CF & Wang, SJ 2003, ‘Engineering and geological characteristics

of granite weathering profiles in South China’, Journal of Asian Earth Sciences, vol. 21, no. 4, pp.

353–364.

Leps, TM 1970, ‘Review of Shearing Strength of Rockfill’, Journal of the Soil Mechanics and

Foundations Division, vol. 96, no. 4, pp. 1159–1170.

Leroueil, S, Kabbaj, M, Tavenas, F & Bouchard, R 1985, ‘Stress–strain–strain rate relation for the

compressibility of sensitive natural clays’, Géotechnique, vol. 35, no. 2, pp. 159–180.

Mallett et al, CW 1983, Stability of Spoil Piles and Highwalls in Deep Surface Mines,.

Marescotti, P, Carbone, C, De Capitani, L, Grieco, G, Lucchetti, G & Servida, D 2008, ‘Mineralogical

and geochemical characterisation of open-air tailing and waste-rock dumps from the Libiola Fe-Cu

sulphide mine (Eastern Liguria, Italy)’, Environmental Geology, vol. 53, pp. 1613–1626.

McLemore, VT, Fakhimi, A, van Zyl, D, Ayakwah, GF, Anim, K, Boakye, K, Ennin, F, Felli, P,

Fredlund, D, Gutierrez, LAF, Tachie-Menson, S & Viterbo, VC 2009, ‘Literature Review of Other

Rock Piles: Characterization, Weathering and Stability’, New Mexico Bureau of Geology and Mineral

Resources, p. 101.

Mitra, R & Onargan, T 2012, ‘Surface Coal Mining Methods in Australia’, in Mining Methods,

InTech.

Morris, PH 1990, The engineering properties and behaviour of coal tailings, The University of

Queensland, School of Civil Engineering, Australia.

Mostofa, MG 2015, Base material characterisation of spoil piles at BMA coal mines, Queensland

University of Technology.

Naderian, AR & Williams, DJ 1996, ‘Simulation of Open-cut Coal Mine Back-fill Behaviour’,

National Symposium on the Use of Recycled Materials in Engineering Construction: 1996;

Programme & Proceedings, p. 17.

Neuendorf, KKE, Mehl, JP & Jackson, JA 2005, Glossary of geology 5th edn, American Geological

Institute, Alexandria, VA.

318

Nguyen, VU 1985, ‘Discussion: Back calculations of slope failures by the secant method’,

Géotechnique, vol. 35, no. 3, pp. 386–388.

Nguyen, VU, Nemcik, JA & Chowdhury, RN 1984, ‘Some practical aspects of spoil pile stability by

the two-wedge model’, Mining Science and Technology, vol. 2, no. 1, pp. 57–68.

Nguyen, VU & Welsh, TC 1981, ‘Characteristics of Spoil Pile Failures In a Strip Coal Mine In

Tertiary And Permian Sediments’, in ISRM International Symposium, International Society for Rock

Mechanics and Rock Engineering.

‘Norwich Park Dragline Recovery’ 2008, accessed July 27, 2019, from

<http://www.miningmayhem.com/2009/09/norwich-park-dragline-recovery.html>.

Okagbue, CO 1986, ‘An investigation of landslide problems in spoil piles in a strip coal mining area,

West Virginia (U.S.A.)’, Engineering Geology, vol. 22, no. 4, pp. 317–333.

Olivier, HJ 1979, ‘Some aspects of the influence of mineralogy and moisture redistribution on the

weathering behaviour of mudrocks’, Proceedings of the 4th international conference on rock

mechanics, vol. 3, pp. 467–474.

Peter, P, Nefiodovas, A, Hignett, C, Wright, W, Wei, Y & Sheng, Y 1996, BHP Spoil Hydrology

Project Final Report, Geotechnical Testing of Spoil, Commonwealth Scientific and Industrial

Research Organisation (CSIRO).

Philip, MG, Boyd, GL & Eckersley, J 1981, ‘Stabilisation of large scale spoil piles in deep open strip

coal mines’, in Proceedings of Symposium on Strip Mining – 45 m and Beyond, The Australasian

Institute of Mining and Metallurgy, Melbourne, Australia., Central Queensland, pp. 119–129.

Poulsen, B, Khanal, M, Rao, AM, Adhikary, D & Balusu, R 2014, ‘Mine Overburden Dump Failure:

A Case Study’, Geotechnical and Geological Engineering, vol. 32, no. 2, pp. 297–309.

Prytherch, B 2012, ‘Goonyella Riverside Mine Dragline Operations’, in BHP Billiton Mitsubishi

Alliance Internal Documents.

Rayment, GE & Higginson, FR 1992, Australian Laboratory Handbook of Soil and Water Chemical

Method, Reed International Books Australia P/L, Trading as Inkata Press, Port Melbourne.

Rayner, DH & Hemingway, JE 1974, The Geology and Mineral Resources of Yorkshire, Yorkshire

Geological Society.

Richards, BG, Coulthard, MA & Toh, CT 1981, ‘Analysis of slope stability at Goonyella Mine’,

Canadian Geotechnical Journal, vol. 18, no. 2, pp. 179–194.

319

Robertson, GP, Sollins, P, Ellis, BG & Lajtha, K 1999, ‘Exchangeable ions, pH and cation exchange

capacity’, in Sollins, P. (Eds.), Standard Soil Methods in Long-term Ecological Research, Oxford

University Press, pp. 106–114.

Rowe, PW 1962, ‘The stress-dilatancy relation for static equilibrium of an assembly of particles in

contact’, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences,

vol. 269, no. 1339, pp. 500–527.

Schor, HJ & Gray, DH 1995, ‘Landform Grading and Slope Evolution’, Journal of Geotechnical

Engineering, vol. 121, no. 10, pp. 729–734.

Seedsman, R 1986, ‘The behaviour of clay shales in water’, Canadian Geotechnical Journal, vol. 23,

no. 1, pp. 18–22.

Seedsman, RW, Cameron, M & Peou, S 1995, ‘Geotechnical Investigations and Pit Wall Designs in

Faulted Ground at Eastern Creek South’, in Bowen Basin Symposium 1995, The Bowen Basin

Geologists Group and the Geological Society of Australia Incorporated Coal Geology Group, pp.

191–192.

Seedsman, RW & Emerson, WW 1985, ‘The role of clay-rich rocks in spoil pile failures at Goonyella

Mine, Queensland, Australia’, International Journal of Rock Mechanics and Mining Sciences &

Geomechanics Abstracts, vol. 22, no. 2, pp. 113–118.

Seedsman, RW, Richards, BG & Williams, DJ 1988, ‘The Possibility of Undrained Failure in Bowen

Basin Spoil Piles’, Fifth Australia-New Zealand Conference on Geomechanics: Prediction Versus

Performance; Preprints of Papers, p. 404.

Selig, E, Dusseault, M, Cimolini, P, Soderberg, H & Scafe, D 1983, ‘Rapid Index Tests for

Transitional Materials’, Geotechnical Testing Journal - GEOTECH TESTING J, vol. 6.

Simmon, J & Fityus, S 2016, The Stability of Very High Spoil Piles, Australian Coal Industries

Research Program.

Simmons, J 1995, ‘Slope stability for open pit coal mining’, in Australian Geomechanics, pp. 45–57.

Simmons, J 2009, Geotechnical Support for Open Pit Coal Mining, Australasian Institute of Mining

and Metallurgy, Carlton, Vic.

Simmons, J & McManus, D 2004, ‘Shear Strength Framework for Design of Dumped Spoil Slopes

for Open Pit Coal Mines’, in Advances in geotechnical engineering: The Skempton conference,

Conference Proceedings, Thomas Telford Publishing, pp. 981–991, accessed October 12, 2016, from

<http://www.icevirtuallibrary.com.ezproxy.library.uq.edu.au/doi/abs/10.1680/aigev2.32644.0026>.

320

Simmons, J & Yarkosky, R 2017, ‘Dragline spoils’, in M Hawley & J Cunning (eds), Guidelines for

Mine Waste Dump and Stockpile Design, CSIRO Publishing, Collingwood, pp. 233–256.

Simons, DB & Albertson, ML 1960, ‘Uniform Water Conveyance Channels in Alluvial Materials’,

Journal of the Hydraulics Division, vol. 86, no. 5, pp. 33–71.

Singh, PK, Roy, MP, Paswan, RK, Sarim, Md, Kumar, S & Ranjan Jha, R 2016, ‘Rock fragmentation

control in opencast blasting’, Journal of Rock Mechanics and Geotechnical Engineering, vol. 8, no.

2, pp. 225–237.

Smith, LA, López, DL, Beckie, R, Morin, K, Dawson, R & Price, W 1995, Hydrogeology of Waste

Rock Dumps, British Columbia Ministry of Energy, Mines and Petroleum Resources and CANMET.

Stockwell, J, Smith, L, Jambor, JL & Beckie, R 2006, ‘The relationship between fluid flow and

mineral weathering in heterogeneous unsaturated porous media: A physical and geochemical

characterization of a waste-rock pile’, Applied Geochemistry, vol. 21, no. 8, pp. 1347–1361.

Taylor, EL, Taylor, TN & Krings, M 2009, Paleobotany: The Biology and Evolution of Fossil Plants,

Academic Press.

Terzaghi, K 1936, ‘The shear resistance of saturated soils’, in The shear resistance of saturated soils,

International Conference Soil Mechanics Foundation Engineering, pp. 54–56.

Ulusay, R, Arikan, F, Yoleri, MF & Çaǧlan, D 1995, ‘Engineering geological characterization of coal

mine waste material and an evaluation in the context of back-analysis of spoil pile instabilities in a

strip mine, SW Turkey’, Engineering Geology, vol. 40, no. 1, pp. 77–101.

Umehara, Y & Zen, K 1980, ‘Constant Rate of Strain Consolidation for Very Soft Clayey Soils’,

Soils and Foundations, vol. 20, no. 2, pp. 79–95.

Vosolo, D 2017, Investigations on Geotechnical Engineering Properties of Coal Mine Spoil

Subjected to Slaking, Queensland University of Technology, School of Civil Engineering.

Westcott, P 2004, ‘Dragline or Truck/Shovel? Some Technical and Business Considerations’, in

University of New South Whales.

Whitman, RV 1984, ‘Evaluating Calculated Risk in Geotechnical Engineering’, Journal of

Geotechnical Engineering, vol. 110, no. 2, pp. 143–188.

Williams, D & Rohde, T 2008, ‘Rainfall infiltration into and seepage from rock dumps - A review’,

in A Fourie (ed), Proceedings of the First International Seminar on the Management of Rock Dumps,

Stockpiles and Heap Leach Pads, Australian Center for Geomechanics, Perth, Western Australia, pp.

79–89.

321

Williams, DJ 2006, ‘The case for revolutionary change to mine waste disposal rehabilitation’, in

Second International Seminar on Strategic versus Tactical Approahces to Mining, Australian Centre

for Geomechanics, Perth, Western Australia, pp. 1–10.

Williams, DJ 2015, Bulking and Subsequent Self-Weight and Saturation Settlements, and

Geotechnical Stability, of Deep Coal Mine Spoil Piles, Australian Coal Industry’s Research Program.,

Brisbane, Australia.

Williams, DJ & Zou, JZ 1991, ‘Stochastic finite element analysis of coal mine spoil pile stability’, in

Proceedings of 7th Conference of International Association for Computer Methods and Advances in

Geomechanics, Cairns, pp. 1411–1416.

Wood, L & Deo, P 1975, ‘A suggested system for classifying shale materials for embankments’,

International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, vol. 12,

no. 8, pp. 39–55.