Hydrometallurgy 179 (2018) 238–244

Contents lists available at ScienceDirect

Hydrometallurgy
journal homepage: www.elsevier.com/locate/hydromet

Effects of grain size gradation on the porosity of packed heap leach beds T
a a,⁎ b
Shuo Zhang , Wenying Liu , Giuseppe Granata
a
Department of Materials Engineering, University of British Columbia, 309-6350 Stores Road, Vancouver, BC V6T 1Z4, Canada
b
Faculty of Science and Engineering, Department of Resources and Environmental Engineering, Waseda University, Tokyo, Japan

A R T I C LE I N FO A B S T R A C T

Keywords: Fluid flow through packed ore/rock beds is among the critical processes that control the release of valuable
Fluid flow metals as well as substances potentially harmful to the environment. The properties of fluid flow in porous media
Heap/dump leaching are associated with the structure of the pores through which the fluid flows which, in turn, is influenced by grain
Waste rock drainage size gradation, grain shape and packing method. In this study, we investigated the effect of three types of grain
Pore size distribution
size gradation on porosity and pore size distribution using the bulk density and the computed tomography (CT)
Packed bed permeability
scanning methods. It was generally observed with the uniformly graded grains that the porosity decreased as the
mean grain size increased until a limit was reached. The porosities of the well graded grains were lower than
those of the uniformly graded grains in the coarse size range, but there was no difference in the porosities
between the two types of gradation in the fine size range. Furthermore, the influence of the packing method on
the well graded grains was more pronounced than on the uniformly graded grains, implying relative ease of
compaction of well graded grains. The proportion of fine grains in the gap graded grains influenced the por-
osities, which firstly decreased and then increased. The pore size distribution of the gap graded grains showed a
high degree of heterogeneity compared to those of the uniformly graded and the well graded grains. The findings
indicate that in ore/rock dumps fine particles account for the bulk of the porosity and possibly the bulk of any
pore water content, suggesting that fine particles are likely to contribute most of the leachable substances.

1. Introduction
The leaching of substances from packed ore/rock beds is of great
significance for valuable metal production via heap/dump leaching as
well as management of mine waste rock drainage. Heap/dump leaching
is a widely applied hydrometallurgical process for treating copper
oxides, secondary sulfides, and possibly low-grade primary sulfides
(Petersen, 2016; Yin et al., 2018). Mine waste rock drainage is among
the most significant environmental challenges facing the global mining
industry due to its dynamics and persistence (Johnson and Hallberg,
2005). Examples are selenium release from coal waste rock (Hendry
et al., 2015), arsenic contamination in diverse geological and climatic
settings (Williams, 2001), and acid mine drainage (Betrie et al., 2016).
On the other hand, mine waste and drainage represent vast opportu-
nities for valuable products recovery given the need for low-cost re-
sources, increasingly stringent environmental conditions, and improved
mineral processing and separation technologies (Lottermoser, 2011;
Nordstrom et al., 2017; Santos and Ladeira, 2011).
The complexity of the leaching process arises from the intertwined
fundamental processes occurring simultaneously in packed ore/rock
beds that are typically of highly heterogeneous nature in chemical and
physical properties (Amos et al., 2015; Zhang and Liu, 2017). Fluid
movement through porous reactive materials is a critical process that
provides reactants to and mobilizes products from leaching reactions
and microbial activities (Liu and Hashemzadeh, 2017; Velleux et al.,
2006; Villeneuve et al., 2017). Fluids can percolate slowly and uni-
formly, leading to a stable wetting front; or move rapidly and pre-
ferentially along paths of least resistance, bypassing a considerable
fraction of the porous media (Hendrickx and Flury, 2001). Channeled
and preferential flow in waste rock piles is common given regions of
vastly different porosity and permeability (Eriksson and Destouni,
1997). The ratio of uniform to preferential flow is crucial for de-
termining the amount of solutes leached from solid matrix and trans-
ported by moving fluids (Jarvis, 2007).
Grain size distribution is among the key physical properties that
control permeability and fluid flow (Stockwell et al., 2006). Various
models have been developed to describe the relationships between
permeability and the statistical parameters that describe the grain size
distribution of the porous media (Carrier, 2003; Chapuis, 2004; Masch
and Denny, 1966; Shepherd, 1989). These empirical models relate grain
size distribution to pore size distribution and porosity, and use them as
dimensionless input parameters (Arya and Paris, 1981; Yu and

⁎
Corresponding author.
E-mail address: wenying.liu@ubc.ca (W. Liu).

https://doi.org/10.1016/j.hydromet.2018.06.014
Received 14 February 2018; Received in revised form 19 June 2018; Accepted 23 June 2018
Available online 27 June 2018
0304-386X/ © 2018 Elsevier B.V. All rights reserved.
S. Zhang et al. Hydrometallurgy 179 (2018) 238–244

Standish, 1991). Our previous study has shown that dump leach pads
contained an extremely wide range of particle sizes and that spatial
segregation of fine and coarse particles existed (Zhang and Liu, 2017).
However, it was unclear how the segregation would affect porosity
behavior and pore structure. To assess the impact of particle size dis-
tribution on fluid flow properties, the current study attempted to
measure porosity directly in beds of varying particle sizes through the
use of computerized tomography and digital image analysis. The image
analysis would provide direct visual evidence on the change of porosity
behavior as the ratio of fine and coarse particles was varied. These
techniques have been applied to examine particle damage and copper
grain exposure by crushing (Kodali et al., 2011), and to aid the mod-
elling of particle scale leaching kinetics (Lin et al., 2016). It is not the
purpose of this paper to advance the image analysis techniques, but to
assess how fine particles affect porosity behavior and the implications
for substance leaching from ore/rock dumps.
2. Materials and methods
2.1. Sample grain preparation
Drain rock, i.e., barren gravel, was used in this study as samples to
investigate the effect of particle size distribution on porosity. The drain
rock with an average diameter of 0.75 in. was crushed and screened to
collect grains between two adjacent sieve apertures. Fourteen such
samples were collected, the mean particle size of which was calculated
as the average size of the two adjacent sieve openings. The calculated
mean particle sizes of the sample grains in descending order were 6.83,
4.83, 3.19, 2.19, 1.60, 1.10, 0.85, 0.60, 0.43, 0.30, 0.21, 0.15, 0.11, and
0.045 mm.
2.2. Sample grain gradation
Table 1
Three well graded samples tested, each as a mixture of four uniformly graded
grains with each size accounting for 25% by mass in the mixture.
Sample no. Sizes of the four uniformly-graded grains in the mixture, mm
1 6.83 4.83 3.19 2.19
2 1.60 0.85 0.60 0.30
3 0.21 0.15 0.11 0.045
parameters: the mean grain size d50; the coefficient of uniformity, de-
fined as Cu = d60/d10; and the coefficient of curvature, defined as
Cc = d30 2/d60 ∙d10 . The terms d10, d30, d50 and d60 are the screen sizes
being passed by 10%, 30%, 50% and 60% by mass of the particles. The
higher the value of Cu the larger the range of the grain sizes in the
sample. For the grains to be classified as well graded, the following two
criteria must be met: Cu ≥ 6 and 1 < Cc < 3. If such criteria were not
met, the grains were classified as poorly graded, either uniformly
graded or gap graded.
The 14 samples collected between the two adjacent sieves had a
narrow size variation and their Cu values were calculated to be ap-
proximately 1. Therefore, they were used to represent the uniformly
graded grains. These uniformly graded grains were mixed in different
size combinations and at various proportions to prepare the well graded
and the gap graded samples. Three well graded samples were prepared,
each as a mixture of four uniformly graded grains with each size ac-
counting for 25% by mass in the mixture (Table 1). Table 2 shows the
coefficient of uniformity and the coefficient of curvature of the three
well graded samples, which met the two criteria: Cu ≥ 6 and
1 < Cc < 3. Six gap graded samples were prepared, each as a mixture
of two uniformly graded grains, one representing fine grains and the
other representing coarse grains (Table 3). For each size combination,
the proportion of fine grains was varied as 0%, 25%, 50%, 75% and
100% by mass.

The method for grading the sample grains, i.e., the classification of
the particle size distribution of the sample grains, was in accordance
with the Unified Soil Classification System. The sample grains were
classified as well graded and poorly graded. The latter was further di-
vided into uniformly graded and gap graded. A well graded sample
contained grains of a wide range of sizes, with each size well re-
presented in the sample. A gap graded sample had an excess or defi-
ciency of certain particle sizes, also called the bimodal size distribution.
The uniformly graded grains referred to the case where the grains had
similar sizes with a narrow size variation. Fig. 1 shows a schematic of
the three types of grain gradation used in this study.
The gradation of the sample grains was characterized by three
2.3. Porosity measurements
2.3.1. Bulk density method
The sample grains were packed into a custom designed test column
of 3.75 in. in diameter and 5 in. in height using two packing methods:
random loose packing and random close packing. For the random loose
packing, grains were directly packed to the target volume without
tapping or shaking the test column, while the random close packing
involved tapping and shaking the test column until no further decrease
in the volume was observed. The porosity of the packed grains was
calculated from its bulk density and the density of the solid grains,
which was measured to be 2.64 g/cm3 by the water displacement
method. Each test was repeated five times to obtain the average por-
osity and the standard deviation as the error bar.

2.3.2. Computed tomography (CT) scanning method

The computed tomography (CT) scanning method allows for non-
invasive 3D imaging of specimens up to 100 mm in diameter and
160 mm in length, with resolution of 5–200 μm. In this study, a micro-
CT specimen scanner (Scanco Medical μCT100) was used to acquire
images of packed grains. Table 4 shows all the grain sizes tested and the

Table 2
Coefficient of uniformity (Cu) and coefficient of curvature (Cc) of the well
graded samples.
Sample no. d10, mm d30, mm d50, mm d60, mm Cu, d60/ Cc, d302/
d10 (d60d10)

1 0.02 0.06 0.11 0.12 6 1.6

2 0.14 0.43 0.60 0.86 6 1.5
Fig. 1. A schematic showing the particle size distribution curves for the three
3 0.67 2.02 3.19 4.05 6 1.5
types of grain gradation: uniformly graded, well graded and gap graded.

239
S. Zhang et al.
Table 3
Six gap graded samples, with the proportion of fine grains varying as 0%, 25%,
50%, 75% and 100% in each sample.
Sample no. Coarse grain size, mm Fine grain size, mm
1 6.83 0.05
2 4.83 0.11
3 3.19 0.15
4 2.19 0.21
5 1.60 0.30
6 0.85 0.60
corresponding sample holder size and resolution. For each sample, a
sample holder of an appropriate size was selected to minimize wall
effects. For all tests, the grains were packed in the sample holder using
the random loose packing method. Due to the expensive and time-
consuming nature of the CT scanning method, all tests were carried out
without replication.
Images of the packed grains were taken and imported as a stack to
the ImageJ software for image processing. The images were first cali-
brated using the sample holder diameter as the reference size. By using
the built-in method for global thresholding that assigned each pixel a
value below or above the specified threshold value, the images were
segmented into binary images consisting of the ore particles as back-
ground and the pore space as foreground. Finally a plugin, called 3D
objects counter, was used to measure the volume of individual pores.
The generated data was then exported to a Microsoft Excel spreadsheet
to calculate the total pore volume as the summation of all individual
pore volumes. The corresponding porosity was calculated as the total
pore volume divided by the total volume of the packed column.
3. Results and discussion
3.1. Effect of particle size distribution on porosity for the uniformly graded
grains
Fig. 2A compares the porosities obtained by the bulk density
method under the random loose and random close packing conditions.
It was generally observed that the porosity increased as the mean grain
size decreased until a limit was reached. For the random loose packing
the porosity increased from about 38% to 52% as the mean grain size
decreased from 3.19 mm to 45 μm. This was different from the case of
the random close packing, where the porosity was observed to increase
from 38% to 52% in a rather narrow size range from 100 μm to 45 μm.
The trends in Fig. 2 identified critical particle sizes above which the
porosity no longer responded to the variation in the mean grain size.
The critical particle size where such transition occurred was 3.19 mm
for the random loose packing, much larger than the 100 μm for the
random close packing. The observed increase in the porosity with
Table 4
Grain sizes tested and the corresponding sample holder size and resolution for the CT scanning method.
Sample No. Sample type Particle size
mm
1 Uniformly graded 0.088
2 0.11
3 0.21
4 0.60
5 2.19
6 6.83
7 Well graded 0.045 + 0.11 + 0.15 + 0.21
8 0.30 + 0.60 + 0.85 + 1.60
9 2.19 + 3.19 + 4.83+ 6.83
10 Gap graded 0.21 + 2.19
Note: 25% by mass of each size component in the well graded samples; the fraction of fine grains varied as 0, 25%, 50% and 75% in the gap graded samples.
240
Hydrometallurgy 179 (2018) 238–244
decreasing grain size was ascribed to the decrease in the grain spheri-
city with decreasing grain size (Rogers and Head, 1961). The random
packings of non-spherical grains is reported to result in porosities larger
than those for spheres (Ouchiyama and Tanaka, 1984).
Fig. 2B shows the porosities obtained by the CT scanning method in
comparison to those obtained by the bulk density method, both by the
random loose packing. The result obtained by the CT scanning method
showed that the porosity increased from 33% to 46% as the mean
particle size decreased from 2.19 mm to 88 μm. This trend was in
general agreement with that obtained by the bulk density method.
However, the critical particle size at which porosity ceased to respond
to grain size was not observed with the CT scanning method, because
the maximum grain size tested with this method was 2.19 mm, smaller
than the transition grain size of 3.19 mm identified with the bulk
density method. Note that the porosity of the 6.83 mm sample was not
included, because the sample holder size used (the largest size avail-
able) was not sufficiently large to provide a reliable porosity value.
Another critical grain size was identified, above which the porosity
no longer responded to the packing method. Specifically, when the
mean grain size was larger than 3.19 mm, the two packing methods
generated porosities of no significant difference, indicating that the
packing method did not influence the porosity in the coarser particle
range. When the mean grain size was smaller than 3.19 mm, the mea-
sured porosities for the close random packing were lower than those for
the loose random packing. The shaking of the column during the close
random packing allowed for more grains to be added to the column,
leading to an increase in the bulk density and accordingly, a decrease in
the porosity of the packed column. These results suggested the im-
portance of the heap/dump construction method in influencing porosity
values (Lahmira et al., 2016), which could change fluid flow properties
and reaction kinetics (Kohfahl et al., 2011).
Fig. 3 presents the CT images for all uniformly graded samples. By
image thresholding, these images were segmented into binary images,
from which the porosities were obtained. Note that the larger but fewer
voids in the larger particle sizes yielded porosities lower than the finer
particles. The large but fewer voids of the large particles would confer
greater permeability than the small but abundant voids of the small
particles. Fig. 4 demonstrates an example of a binary image after the
image processing procedure.
3.2. Effect of particle size distribution on porosity for the well graded grains
Fig. 5A shows the porosity values derived by the bulk density
method with the random loose and random close packing for the well
graded grains. In both cases, the porosities decreased as the mean grain
size increased from 0.13 mm to 0.84 mm, at which a porosity limit was
reached. Further increase in the mean grain size from 0.83 mm to
4.26 mm did not lead to any change in the porosity. The results ob-
tained by the CT scanning method were generally consistent with those
Sample holder diameter (mm) Resolution (μm)
14 7.4
14 7.4
14 7.4
14 7.4
19 10
48 24.6
14 7.4
19 10
48 24.6
19 10
S. Zhang et al.
Fig. 2. Porosities of the packed uniformly graded grains at varying grain sizes: (A) derived by the bulk density method; (B) derived by the CT scanning method in
comparison with the bulk density measurements under random loose packing condition.
by the bulk density method (Fig. 5B).
A comparison of the two types of grain gradation showed that the
well graded grains behaved differently from the uniformly graded
grains (Fig. 6). For the finer grains < 0.83 mm in size, the well graded
grains yielded porosities similar to that of the uniformly graded grains.
This indicated that porosities were independent of grain size gradation
in the fine size range. In contrast, in the coarser grain size range, the
porosities of the well graded grains were lower, implying that it was
easier to compact the well graded grains than the uniformly graded
grains.
Furthermore, the influence of the packing method on the well
graded grains was more pronounced than for the uniformly graded
grains. The porosities for the well graded grains responded to the
packing method for the entire size range tested, as opposed to the case
of the uniformly graded grains, for which the porosity was independent
of the mean grain size when it was larger than 3.19 mm (Fig. 2A). These
results provide experimental evidence that not only grain size but also
grain size distribution are critical factors in determining porosities of
packed ore/rock beds. The coefficient of uniformity and the coefficient
of curvature could be effective parameters that represent the grain size
Fig. 3. CT images of the uniformly graded samples (mm) by random loose packing: (A) 0.045; (B) 0.11; (C) 0.21; (D) 0.60; (E) 2.19; (F) 6.83; sample holder sizes and
image resolutions shown in Table 4.
241
Hydrometallurgy 179 (2018) 238–244
distribution for predicting porosity values.
3.3. Effect of particle size distribution on porosity for the gap graded grains
Gap graded samples were prepared as mixtures of two uniformly
graded grains with a large size variation, one representing fine grains
and the other representing coarse grains. Fig. 7A shows that with in-
creasing fraction of fine particles, the porosities firstly decreased and
then increased. A similar trend was reported by Wickland et al. (2006)
in the investigation of co-disposal of tailings and waste rock. To verify
the result, one sample, the mixture of 0.21 mm and 2.19 mm grains, was
selected for porosity measurement by the CT scanning method. Fig. 7B
shows that the trend of the curve was similar to that derived by the bulk
density method.
A possible explanation for the initial decrease was that the coarse
particles were dominant and the fine particles filled the void space
formed between the coarse particles. As the fine grain fraction in-
creased further, the fine grains played an increasingly dominant role
and the coarse particles were embedded in a fine particle matrix, re-
sulting in an increase in the porosity. The CT images shown in Fig. 8
S. Zhang et al. Hydrometallurgy 179 (2018) 238–244

Fig. 4. (A) An example of a CT image of the 2.19 mm sample; (B) corresponding 8-bit binary image after image processing.

Fig. 5. (A) Porosity of well graded grains derived by the bulk density method under the random loose and close packing; (B) comparison of the porosities by the bulk
density and the CT scanning methods, both by the loose random packing method.

Fig. 6. Comparison of porosities between the uniformly graded grains and the well graded grains under random loose packing condition: (A) by the bulk density
method; (B) by the CT scanning method.

illustrate the filling of the void space as the fraction of fine grains was
increased. Another observation was that the magnitude of the porosity
change was much more marked for the samples with a larger size
variation. As the difference in the size of the two components became
smaller, the effect of fine particle addition on the porosity became less
significant. This was supported by the most pronounced effect observed
for the combination of 6.83 mm and 0.045 mm and the least significant
effect for the combination of 0.85 mm and 0.60 mm. This result was
consistent with the previous conclusion that it is more difficult to
compact particles of a more uniform size distribution.
3.4. Qualitative analysis of pore size distribution
The complex nature of pore structure cannot be fully characterized
by a single porosity value. Porosity does not provide any information on
pore size distribution, pore connectivity and tortuosity, all of which
play key roles in influencing permeability and flow properties (Xiong
et al., 2016). In this study the pore size distributions of the three types
of gradation were only qualitatively examined by comparing their CT
images (Fig. 9). More quantitative information on the effect of particle
size distribution on pore structure can be extracted from the images,
which will be the focus of future work. The heterogeneity in the pore
size distribution was not apparent in the case of the uniformly graded
particles and the well graded particles. In contrast, the pore size

242
S. Zhang et al. Hydrometallurgy 179 (2018) 238–244

Fig. 7. (A) Porosities with increasing fraction of fine grains for the six gap graded samples under the loose packing condition: (A) by the bulk density method (B) by
the CT scanning method in comparison with those derived by the bulk density method for one gap graded sample.

Fig. 8. CT images of one gap graded sample, the mixture of 2.19 mm and 0.21 mm, with a varying fraction of 0.21 mm grains by mass: (A) 25%; (B) 50%; (C) 75%.

Fig. 9. Binary images after image processing to show the pore size distribution at two different vertical locations: (A) and (B) for uniformly graded grains with a mean
particle size of 2.19 mm; (C) and (D) for one well graded sample, a mixture of 1.60, 0.85, 0.60 and 0.30 mm, 25% each by mass; (E) and (F) for gap graded grains with
25% 0.21 mm and 75% 2.19 mm.

distribution for the gap graded grains presented a high degree of het- 3.5. Practical relevance
erogeneity as well as the segregation of fine and coarse grains; pores of
large sizes were observed in the coarse grain region with some being The experimental results obtained using grain size ranging from
filled by fine grains. 6.83 mm to 45 μm show that the porosity decreased as the grain size

243
S. Zhang et al.
increased until a limit was reached, after which the porosity ceased to
respond to grain size. The particle size used constitutes the fine or very
fine end of the particle size spectrum in ore/rock dumps. Nevertheless,
it is this very range of sizes that likely contributes most of the leachable
substances. Indeed, experiments on copper leaching from sulfide-con-
taining ore/rock have shown that metal extraction from the fines is
higher than from the coarser size fractions (Bennett et al., 2012; Kodali
et al., 2011).
The trends observed in this study suggest that regions of fine grains
will account for the bulk of the porosity, and therefore the bulk of any
pore water content. If so, then any dissolution reactions would be
concentrated in those regions of fine grains. Of course, the dissolution
reaction is only one component of the leaching process: the second
component is the elution of the dissolved substance from the rock bed.
That elution component is a transport process dependent on perme-
ability, which will also be affected by the grain size and porosity. The
experimental results also show that for gap graded grains, typically
observed in ore/rock dumps, the relative fraction of fines to coarse
particles determined the porosity. This in turn dictates the geochemical
and hydrogeological properties of waste rock (Amos et al., 2015).
4. Conclusion
The effect of grain size gradation on porosity was studied using the
bulk density and the CT scanning methods. The grain size gradation
studied consisted of uniformly graded, well graded and gap graded
grains in the range of 0.045 to 6.83 mm. For the uniformly graded
grains, it was generally observed that the porosity decreased as the
grain size increased until a limit was reached, after which the porosity
was independent of grain size; furthermore, the porosity of the uni-
formly graded grains was independent of packing method when the
particle size was larger than 3.19 mm. The porosities of the well graded
grains were lower than those of the uniformly graded grains in the
coarse size range, but in the fine size range similar porosities were
obtained with these two types of grain size gradation. Furthermore, the
influence of the packing method on the well graded grains was more
pronounced than on the uniformly graded grains, implying relative ease
of compaction of the well graded grains. For the gap graded grains, as
the fraction of fine particles increased, the porosities firstly decreased
and then increased. The pore size distribution for the gap graded grains
showed a high degree of heterogeneity compared with the uniformly
graded and the well graded grains. The range of grain sizes used in this
study constitutes the fine or very fine end of the particle size spectrum
in ore/rock dumps. Nevertheless, it is this very range of sizes that likely
contributes most of the porosity and pore water and therefore leachable
substances.
References
Amos, R.T., Blowes, D.W., Bailey, B.L., Sego, D.C., Smith, L., Ritchie, A.I.M., 2015. Waste-
rock hydrogeology and geochemistry. Appl. Geochem. 57, 140–156.
Arya, L.M., Paris, J.F., 1981. A physicoempirical model to predict the soil moisture
characteristic from particle-size distribution and bulk density Data1. Soil Sci. Soc.
Am. J. 45, 1023–1030.
Bennett, C.R., McBride, D., Cross, M., Gebhardt, J.E., 2012. A comprehensive model for
copper sulphide heap leaching: part 1 basic formulation and validation through
column test simulation. Hydrometallurgy 127–128, 150–161.
Betrie, G.D., Sadiq, R., Nichol, C., Morin, K.A., Tesfamariam, S., 2016. Environmental risk
assessment of acid rock drainage under uncertainty: the probability bounds and
PHREEQC approach. J. Hazard. Mater. 301, 187–196.
244
Hydrometallurgy 179 (2018) 238–244
Carrier, W.D., 2003. Goodbye, Hazen; Hello, Kozeny-Carman. J. Geotech. Geoenviron.
129, 1054–1056.
Chapuis, R.P., 2004. Predicting the saturated hydraulic conductivity of sand and gravel
using effective diameter and void ratio. Can. Geotech. J. 41, 787–795.
Eriksson, N., Destouni, G., 1997. Combined effects of dissolution kinetics, secondary
mineral precipitation, and preferential flow on copper leaching from mining waste
rock. Water Resour. Res. 33, 471–483.
Hendrickx, J.M.H., Flury, M., 2001. Uniform and Preferential Flow Mechanisms in the
Vadose Zone, Conceptual Models of Flow and Transport in the Fractured Vadose
Zone. National Academy Press, Washington, D.C.
Hendry, M.J., Biswas, A., Essilfie-Dughan, J., Chen, N., Day, S.J., Barbour, S.L., 2015.
Reservoirs of selenium in coal waste rock: Elk Valley, British Columbia, Canada.
Environ. Sci. Technol. 49, 8228–8236.
Jarvis, N.J., 2007. A review of non-equilibrium water flow and solute transport in soil
macropores: principles, controlling factors and consequences for water quality. Eur.
J. Soil Sci. 58, 523–546.
Johnson, D.B., Hallberg, K.B., 2005. Acid mine drainage remediation options: a review.
Sci. Total Environ. 338, 3–14.
Kodali, P., Dhawan, N., Depci, T., Lin, C.L., Miller, J.D., 2011. Particle damage and ex-
posure analysis in HPGR crushing of selected copper ores for column leaching. Miner.
Eng. 24, 1478–1487.
Kohfahl, C., Graupner, T., Fetzer, C., Holzbecher, E., Pekdeger, A., 2011. The impact of
hardpans and cemented layers on oxygen diffusivity in mining waste heaps: diffusion
experiments and modelling studies. Sci. Total Environ. 409, 3197–3205.
Lahmira, B., Lefebvre, R., Aubertin, M., Bussière, B., 2016. Effect of heterogeneity and
anisotropy related to the construction method on transfer processes in waste rock
piles. J. Contam. Hydrol. 184, 35–49.
Lin, Q., Barker, D.J., Dobson, K.J., Lee, P.D., Neethling, S.J., 2016. Modelling particle
scale leach kinetics based on X-ray computed micro-tomography images.
Hydrometallurgy 162, 25–36.
Liu, W., Hashemzadeh, M., 2017. Solution flow behavior in response to key operating
parameters in heap leaching. Hydrometallurgy 169, 183–191.
Lottermoser, B.G., 2011. Recycling, reuse and rehabilitation of mine wastes. Elements 7,
405.
Masch, F.D., Denny, K.J., 1966. Grain size distribution and its effect on the permeability
of unconsolidated sands. Water Resour. Res. 2, 665–677.
Nordstrom, D.K., Bowell, R.J., Campbell, K.M., Alpers, C.N., 2017. Challenges in re-
covering resources from acid mine drainage. In: Wolkersdorfer, C., Sillanpää, M.,
Häkkinen, A. (Eds.), Mine Water and Circular Economy. Proceedings of the 2017
International Mine Water Association Conference, International Mine Water
Association, 25–30 June. Lappeenranta, Finland, pp. 1138–1147.
Ouchiyama, N., Tanaka, T., 1984. Porosity estimation for random packings of spherical
particles. Indust. Eng. Chem. Fundament. 23, 490–493.
Petersen, J., 2016. Heap leaching as a key technology for recovery of values from low-
grade ores – a brief overview. Hydrometallurgy 165 (Part 1), 206–212.
Rogers, J.J.W., Head, W.B., 1961. Relationships between porosity, median size, and
sorting coefficients of synthetic sands. J. Sediment. Res. 31, 467–470.
Santos, E.A., Ladeira, A.C.Q., 2011. Recovery of uranium from mine waste by leaching
with carbonate-based reagents. Environ. Sci. Technol. 45, 3591–3597.
Shepherd, R.G., 1989. Correlations of permeability and grain size. Ground Water 27,
633–638.
Stockwell, J., Smith, L., Jambor, J.L., 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. Appl. Geochem. 21,
1347–1361.
Velleux, M.L., Julien, P.Y., Rojas-Sanchez, R., Clements, W.H., England, J.F., 2006.
Simulation of metals transport and toxicity at a mine-impacted watershed: California
gulch, Colorado. Environ. Sci. Technol. 40, 6996–7004.
Villeneuve, S.A., Barbour, S.L., Hendry, M.J., Carey, S.K., 2017. Estimates of water and
solute release from a coal waste rock dump in the Elk Valley, British Columbia,
Canada. Sci. Total Environ. 601, 543–555.
Wickland, B.E., Wilson, G.W., Wijewickreme, D., Klein, B., 2006. Design and evaluation of
mixtures of mine waste rock and tailings. Can. Geotech. J. 43, 928–945.
Williams, M., 2001. Arsenic in mine waters: an international study. Environ. Geol. 40,
267–278.
Xiong, Q., Baychev, T.G., Jivkov, A.P., 2016. Review of pore network modelling of porous
media: experimental characterisations, network constructions and applications to
reactive transport. J. Contam. Hydrol. 192, 101–117.
Yin, S., Wang, L., Kabwe, E., Chen, X., Yan, R., An, K., Zhang, L., Wu, A., 2018. Copper
bioleaching in China: review and Prospect. Fortschr. Mineral. 8, 32.
Yu, A.B., Standish, N., 1991. Estimation of the porosity of particle mixtures by a linear-
mixture packing model. Ind. Eng. Chem. Res. 30, 1372–1385.
Zhang, S., Liu, W., 2017. Application of aerial image analysis for assessing particle size
segregation in dump leaching. Hydrometallurgy 171, 99–105.