Minerals Engineering 22 (2009) 840848

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

Suitability of bentonite-paste tailings mixtures as engineering barrier material

for mine waste containment facilities
M. Fall *, J.C. Clestin, F.S. Han
Department of Civil Engineering, University of Ottawa, 161 Colonel By, Ottawa (Ontario), Canada K1N 6N5

a r t i c l e i n f o a b s t r a c t

Article history: This study investigates the feasibility of using bentonite-paste tailings (BPT) as a barrier (liner, cover)
Received 26 November 2008 material for mine waste containment facilities. Improvements of the hydraulic properties are realized
Accepted 28 February 2009 by using compaction to densify the paste tailings in the rst stage and mixing an additive-like natural
Available online 3 April 2009
bentonite to paste tailings to further reduce the voids that control hydraulic conductivity in the second
stage. A signicant decrease in hydraulic conductivity is observed with these transformations. Values that
Keywords: are as low as 1  10 9 and 4  10 9 cm/s are obtained in 8% and 4% bentonite and BPT, respectively.
Paste backll
Based on these satisfactory values, additional investigations are conducted to evaluate the freezethaw
Tailings
Acid mine drainage
and wetdrying performance of BPT up to 5 and 6 cycles, respectively. The results show that negligible
Recycling to acceptable changes in hydraulic conductivity occur. None of the changes reach one order of magnitude.
Barrier As a nal step, a cost analysis is undertaken to evaluate the economical benets that can be obtained from
such new material. When compared to conventional compacted claybentonite barrier or sandbenton-
ite with 12% bentonite concentration, it is found that the 4% BPT is less expensive by 66%. The results
place this recycled BPT material as a promising candidate for barrier design while reducing the amount
of waste to be managed and the cost of surface tailings management.
2009 Elsevier Ltd. All rights reserved.

1. Introduction mentioned tailings disposal facilities can result in serious

The mining industry continues to strongly contribute to the
economy of many countries around the world (e.g. Australia, Can-
ada, and South Africa). However, it produces a huge amount of so-
lid waste, such as tailings. The tailings result from the
hydrometallurgical processing during the mineral extraction/liber-
ation process (Benzaazoua et al., 2004). In most cases, tailings are
reputed to be hazardous for the environment, health and economy.
This is because most of these tailings are rich in pyrite and pyrrho-
tite (Elberling et al., 1994). These sulphides, when they come in
contact with water and oxygen, impose high risks to the environ-
ment due to the potential of formation of acid mine drainage
(AMD) (Fall and Merkel, 2001; Ritcey, 2005; Amde et al.,
2007a,b) which can damage the ecosystems, and animal and hu-
man lives. Hence, the management and disposal of tailings is a ma-
jor technical challenge for the mining industry as well as an
environmental concern in many countries.
The tailings have been traditionally stored in tailings disposal
facilities (e.g. dams, ponds and other types of surface impound-
ments) located at the mine surface in a variety of different
methods. However, the degradation and/or failure of the afore-
* Corresponding author. Tel.: +1 613 562 5800x6138.
E-mail address: mfall@eng.uottawa.ca (M. Fall).
environmental and geotechnical problems with severe social and
economical ramications. These risks and consequences associated
with the conventional tailing impoundments, the immense opera-
tion and maintenance cost of these impoundments as well as pub-
lic perception and stricter regulations regarding the disposal of
such waste have made the mining industry and engineering com-
munity consider alternate usage of the tailings, and develop new
strategies of tailings management which should be environmen-
tally sound and cost-effective. Therefore, nding alternative ap-
proaches for tailings management and usage of tailings as
construction materials have become a great challenge for the min-
ing and civil engineering community.
In this scenario, the mining industry has implemented cemen-
ted paste tailings (CPT) technology as a cost-effective means of tail-
ings management and construction materials in mining operations.
CPT is a relatively new engineering material containing 7085% of
tailings (proportion of particles with size lower than 20 lm should
be at least 20%), 27% binder and 1530% water (Fall et al., 2008,
2004; Yilmaz et al., 2003; Landriault, 1995). Since its rst usage
in the Bad Grund Mine (Germany) (Yilmaz et al., 2004; Fall et al.,
2008) in the late 1970s, CPT has gained tremendous popularity
among mining operators. CPT offers the advantage of increasing
ore extraction capability by substituting pillars with CPT struc-
tures. CPT technology can allow the storage of up to 60% of the tail-
ings produced underground, thereby leading to a drastic reduction

0892-6875/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mineng.2009.02.011
 M. Fall et al. / Minerals Engineering 22 (2009) 840848
of surface land disturbance (Fall et al., 2008). Moreover, the maxi-
mum underground disposal of potential acid generating tailings
represents a signicant environmental benet (Verburg, 2001).
The reactivity of the sulphide minerals in the CPT system (i.e. its
ability to develop AMD) is negligible due to the high water satura-
tion of the CPT, binder hydration products acting as a physical bar-
rier to oxygen and porosity reduction caused by the cement
hydration process (Fall et al., 2004).
However, although CPT can allow the safe storage of a huge
amount of mine tailings produced in underground openings, a
large quantity of the tailings (at least 40%) will still remain and
thus, have to be stored and managed in the aforementioned con-
ventional tailings impoundments and/or emerging tailings disposal
facilities with all associated geo-environmental, economical and
geotechnical safety concerns. Over the last few years, new and
modied surface tailings disposal methods have been introduced
to reduce the geotechnical and environmental hazards associated
with conventional tailings impoundments. These new methods in-
clude densied tailings (thickened, surface paste, and ltered tail-
ings which are also called dry stack tailings), co-disposal of tailings
and waste-rock. In summary, these densied tailings have been
signicantly dewatered to a point where they form a dense mate-
rial with a solid content, usually between 50% and 70% (for thick-
ened tailings), 70% and 80% (for paste tailings) and over 85% (for
ltered tailings). The densied tailings are transported to the dis-
posal area through pipelines by pumping (for thickened and paste
tailings), conveyor or truck (for ltered tailings). The main advan-
tages of the technology of densied tailings include, but are not
limited to, reducing seepage from stored tailings and minimizing
the requirement for retaining dams due the self-supporting ability
of the densied tailings (reduction of capital cost). However, it
should be mentioned that the densied tailings do not fully elim-
inate the AMD and seepage problem. Furthermore, the relatively
low degree of saturation of the ltered tailings can be very favour-
able for AMD development. Finally, mine co-disposal involves the
mixing and disposal of tailings together with coarse mine waste.
This further improves drainage properties and increases the rate
of consolidation, resulting in a greater nal strength and geotech-
nical stability (Struthers, 1999). In addition, the co-mixing of
waste-rock and tailings reduces the air and water permeability of
the co-disposal mass, thereby resulting in the improvement of
resistance to AMD development.
The above summarizes the emerging methods of surface tail-
ings management. Although they have improved geotechnical per-
formance and reduced environmental risks (e.g. AMD, seepage) of
tailings disposal facilities, they are still based on the permanent
surface storage of tailings and the containment concept, and are
also relatively unproven. These new methods (densied tailings)
also require larger surfaces of land compared to the conventional
tailings impoundments. Hence, if space is a constraint, this will
immediately exclude densied tailings. Moreover, none of them
can fully eliminate the AMD and seepage problems. Lastly, the
mining companies are still reluctant to switch from conventional
tailings impoundments to these new tailings management ap-
proaches although wet deposition is more problematic.
Considering the facts mentioned above, it can be concluded that
the environmental risks related to conventional and modern tail-
ings surface disposal facilities will still remain a pressing issue
and concern in the coming years or decades. Therefore, the devel-
opment or improvement of new technology to eliminate or limit
AMD and seepage associated with tailings disposal facilities as well
as the constructive recycling of tailings in mining and/or civil engi-
neering construction is needed. One of the most cost-effective
techniques used in surface mine waste management to eliminate
or limit AMD development and seepage (i.e. control acid and/or
pollutants migration) into underlying aquifers is a barrier (liner,
841
cover), which can also be a cost-effective means of recycling tail-
ings as explained below.
The mining industry relies on a wide range of liner types to
minimize the migration of efuent from tailings impoundments
to surface and/or ground waters. The major types of liners used
to control seepage from tailings disposal facilities are geosynthetic
clay liners, which are very expensive, and natural deposited and
compacted liners made of local or imported clays. One of the most
desirable properties in liner design is low hydraulic conductivity
that prevents migration of hazardous uids to underlying areas
where wastes are deposited. In the case of compacted clay, the
Environmental Protection Agency (EPA) requires a hydraulic con-
ductivity less than 10 7 cm/s (Daniel, 1993) of the compacted clay.
Previous studies performed on paste tailings (PT) materials re-
vealed that its saturated hydraulic conductivity is in the range of
10 510 6 cm/s, i.e. close to the minimum saturated hydraulic
conductivity required for liner design according to the EPA. This
means as soil permeability is governed by the proportion and nat-
ure of the ner fraction (Shaee, 2008), PT mixed with other ne or
swelling materials, such as bentonite, can be potentially used as
liner for tailings disposal facilities, in regions where preferred
materials, such as natural clay, are not available for the design. This
will result in reduction in the cost of the liner design as well as a
good opportunity to recycle tailings into mining constructions that
can promote environment sustainability. However, the recycling of
PT materials has not yet been proposed as liner materials. This may
be due to the fact that PT materials can contain (but not always) a
high quantity of sulphide minerals that can generate AMD. How-
ever, the use of environmental desulphurization (e.g. Leppinen
et al., 1997) can allow the separation of the sulphide (acid generat-
ing) and non-sulphide minerals, thereby producing non-acid gen-
erating PT materials. The latter can be used as liner materials
while the acid generating tailings will be used in the CPT prepara-
tion. The technical and economical feasibility of environmental
desulphurization have been proven by several studies (e.g.
McLaughlin and Stuparyk, 1994; Yalcin et al., 2004). Furthermore,
non-reactive natural tailings (i.e. tailings with very low sulphide
content and do not generate AMD) are largely available in many
mining areas, such as eastern Canadian mining sites. Thus, there
is a need to study the suitability of the mixture of non-acid gener-
ating PT and bentonite as liner for tailings disposal facilities.
However, in practice, the liner alone is not enough to control
AMD and its impact on the groundwater, since in the long term,
inltration water can enter into the tailings impoundment and be-
come acid. When the inltration water cannot escape via seepage,
it will accumulate until the available storage is lled and then dis-
charge over the lip of the liner (Robertson et al., 1998). Thus, an
effective means to control water inltration as well as oxygen sup-
ply to the tailings impoundment is the construction of a cover. One
of the approaches used in mining operations is the soil cover. The
details about the various types of soil covers and their characteris-
tics are given in Nicholson et al. (1989); Yanful (1993); Yanful et al.
(2006), etc. Covering tailings with a soil layer that has a low
hydraulic conductivity limits accessibility of water to the tailings
(Yanful et al., 2006). In addition, when the soil cover is placed close
to saturation and maintained at this high water content, oxygen
accessibility is also restricted (Yanful et al., 2006). Thus, a ne-
grained soil cover with minimum hydraulic conductivity, but max-
imum degree of saturation can drastically reduce the accessibility
of water and oxygen to the tailings (Yanful et al., 2006). Hence,
non-reactive PT materials blended with small amount of bentonite
can be an adequate soil cover. However, our knowledge about the
hydraulic properties of bentonite-paste tailings (BPT) mixture is
limited.
In consideration of the facts that are mentioned above, a re-
search program has been conducted at the University of Ottawa
842
to study the suitability of the mixtures of non-acid generating PT
and bentonite as the liner and cover for waste containment facili-
ties. If PT can be recycled in engineering barrier constructions, then
signicant amount of tailings can be reutilized in the mining waste
management which helps to reduce environmental problems as
well as costs associated with surface tailings management. A part
of the obtained results, i.e. those related to one of the key param-
eters (saturated hydraulic conductivity) used to judge the suitabil-
ity of BPT mixtures as a liner or cover will be presented in this
paper. The main objectives of this paper are:
To present the results of the experimental evaluation of the
saturated hydraulic conductivity of compacted PT;
To present the results of the experimental evaluation of the
saturated hydraulic conductivity of compacted BPT mixtures;
To present the results of the effect of environmental stress
(wetdrying and freezethaw cycles) on the saturated hydrau-
lic conductivity of compacted BPT mixtures; and
To discuss the economical benets obtained from BPT
barriers.
2. Experimental program
2.1. Materials used
2.1.1. Tailings
In this study, articial tailings prepared industrially from
ground silica and available commercially are used. They have sim-
ilar physical properties as natural tailings from the milling process.
The articial tailings have an advantage where they are exempt of
chemicals that can interfere with the test results. The principal
mineral component of this material is quartz whose percentage
is listed as 99%. The weight proportions of particles with a diame-
ter lower than 20 lm is around 40% (i.e. higher than the minimum
proportion required to develop paste tailings). For comparative
tasks, the grain size distributions of tailings from nine eastern
Canadian mines are presented in Fig. 1 with articial tailings.
When observing this graph, it can be noticed that the tailings con-
tain no clay particles. It can also be observed that the silica, with
about 45 wt% ne particles (<20 lm), can be classied as medium
tailings. Moreover, the tailings are tested for index properties (li-
quid and plastic limits) following ASTM D 4318. However, the tail-
ings are unworkable in order to determine any of these parameters
at the laboratory. Aubertin et al. (1996) reported from extrapola-
tion, a plastic limit of 17.5%, but no liquid limit. From the experi-
100
90
Volume percent (cumulative)
80
70
60
50
40
9 Canadian mines
30
artificial tailings
20
Bentonite
10
0 of optimum lies around 20% moisture content and a maximum
0.0001 0.001 0.01 0.1 1 10
Grain size (m )
Fig. 1. Grain size distribution of the articial tailings and bentonite used compared
to the tailings from nine Canadian mines from east Ontario.
M. Fall et al. / Minerals Engineering 22 (2009) 840848
mental work, the used tailings can be classied as sandy silts of
low plasticity, ML in the Unied Soil Classication System. Specic
gravity tests were conducted following the ASTM D854-02. The va-
lue determined is 2.66 for the tailings. It should be mentioned that
besides the ground silica, natural tailings were also used to prepare
some PT samples for comparison with respect to the compaction
behaviour of tailings. The natural tailings have a grain size curve
close to that of the ground silica. The natural tailings originate from
hard magmatic rocks and are characterised by the dominance of
sulphide minerals (pyrite), quartz and paragonite.
2.1.2. Bentonite
Bentonite was obtained from a commercial distributor. The so-
dium mineral type of sodium bentonite was chosen as it presents
good swelling potential, absorbing almost ve times its weight
and occupying a volume up to 15 times its bulk dry volume at full
saturation (Kashir and Yanful, 2001). The swelling property confers
the ability to plug voids in a mix and therefore, improve hydraulic
conductivity (Kashir and Yanful, 2001). The grain size distribution
of the bentonite determined by the hydrometer test is shown in
Fig. 1. A percentage of 92% clay can be noticed in the bentonite.
The specic gravity value determined is 2.58 for the bentonite. In
order to verify the minerals present in the bentonite, X-ray diffrac-
tion of the powder material was performed using Scintag XDS 2000
XRD. The results of the diffractogram (Fig. 2) show that the domi-
nant clay mineral present is smectite (sodium montmorillonite).
Other important minerals present, but in less intensity, are kaolo-
nite, clinoptilolite, quartz and calcite (Fig. 2). In contrast to the tail-
ings used, the Atterberg limits of the bentonite were well dened.
A liquid limit of 657 and plastic limit 52 were determined, con-
ducting to a plasticity index of 605. Therefore, bentonite can be
classied in the CH group (inorganic, fat clays, highly plastic clay).
2.1.3. Mixing water
Deionised water was used to ensure that no chemical parameter
was involved during this series of tests.
2.2. Preparation of the specimens
For the plain tailings and BPT mix, the anhydrous material was
rst homogenized in the dry state. Once mixing was completed,
the amount of deionised water was gradually added until there
was a uniform hydration of blend. After compaction in a proctor
mould, a cutting cylinder of 5 cm in diameter was used for sam-
pling undisturbed cores for hydraulic conductivity testing. Thereaf-
ter, specimens were removed from the cylindrical sampling tube
using a mechanical press and stored in a polyethylene packaging
for 24 h prior to conducting the hydraulic conductivity test. Hy-
drated samples were trimmed and installed in a triaxial cell where
the hydraulic conductivity measurement took place.
2.3. Testing methods and procedures
2.3.1. Compaction tests
Compaction was performed following ASTM D 698-00a for
Standard Proctor. The rational of compaction was to increase the
dry density and thereby to reduce the void in the material. To come
up with the optimum moisture content (OMC), initially, several
specimens of raw PT were prepared and tested at water contents
from the range of 4% to 24% in gradual increments of 2% of water.
The compaction curve drawn from these tests showed that the wet
100 1000 density of 1580 kg/m3. is determined. For proper characterization
of compaction behaviour, tests were conducted at least on two
points on the wet side, two points on the dry side and at the opti-
mal water content. The water content from the optimum was con-

300
250
Montmorillonite
200
(Na-Smectite)
Intensity, CPS
150
100
Clinoptilolite
50
0
5 10
sidered to ensure that all tests took place in a range susceptible of
leading to maximum dry density (i.e. minimum void ratio) for all
other tests. Thus, the indication of maximum density from the
raw PT was used as a reference for all tests with modied PT pre-
pared with bentonite. Each test was repeated at least twice and the
average value was considered as the water and dry densities of the
sample were tested.
2.3.2. Hydraulic conductivity tests
The exible wall permeameter principle was used following
ASTM D 5084-00 to perform hydraulic conductivity tests. The efu-
ent head was kept constant at 10 kPa, while inuent head was cal-
culated from desired gradient to be applied as a function of the
sample thickness. Saturation was achieved by bridging the inuent
and efuent lines and applying backpressure. Saturation was con-
sidered complete by verifying inuent intake against efuent
water volume supply until they became equal.
During the preliminary hydraulic conductivity tests conducted
at room temperature, samples of compacted plain PT were pre-
pared at various water contents with a thickness of 7.5 cm. The re-
corded values were in the order of 10 610 5 cm/s under a
hydraulic gradient of 30 as recommended by ASTM D 5084. Inves-
tigations carried out by Aubertin et al. (1996) and Qiu and Sego
(2001) reported hydraulic conductivity of compacted and consoli-
dated tailings to be almost in the same range of 10 710 5 cm/s.
When it was time to test the BPT samples, it was observed during
the recording operation that even after periods of 36 h, these sam-
ples remained unsaturated and no ow took place. This led to an
inference that the value of the hydraulic conductivity has signi-
cantly decreased. For this reason, in the rst period, the thickness
of the samples was reduced to an average of 3.5 cm. As this mod-
ication did not appear to stimulate ow, an additional measure
consisting in gradually increasing the hydraulic gradient up to a -
nal value of 600 was adopted to provoke ow. This is the value at
which an acceptable ow takes place in a reasonable period of
time. It was rst veried that the compacted material was suf-
ciently consistent (Shaee, 2008) in accepting this gradient with-
out perturbation of the results. Also, this gradient can be
considered acceptable as literatures report values up to 1800 (Ka-
shir and Yanful, 2001) used during permeability tests in low
hydraulic conductivity materials. The effective stress maintained
in the cell was an average of 130 kPa, which is a value that can
M. Fall et al. / Minerals Engineering 22 (2009) 840848 843
Clinoptilolite Clinoptilolite
Kaolinite
Quartz Calcite
Quartz
Kaolinite
Calcite
15 20 25 30 35 40 45 50 55 60
Degrees 2-theta - Cu K Radiation
Fig. 2. X-ray diffractogram for the bentonite.
be produced by overlying waste (Kashir and Yanful, 2001) or found
in the majority of tailings management facilities (Qiu and Sego,
2001). The constant hydraulic gradient of 600 was applied during
tests using controlled air pressure from the control panel. Addi-
tional care was also taken to collect specimens as much as possible
at the middle of the compaction mould so they can all be consid-
ered under the same compactive condition. Finally, the hydraulic
conductivity was calculated using the constant head principle.
Hydraulic conductivity was conducted over a range of water con-
tent plus or minus 4% OMC at different water contents. Each
hydraulic conductivity test was repeated at least two times and
the average value was considered as the hydraulic conductivity
of the sample tested.
2.3.3. Freezethaw cycles
Surface tailings disposal facilities in permafrost-affected areas
or cold regions can be subjected to a series of freezethaw cycles
that may affect the hydraulic conductivity of the lining and cover
materials. This can reduce the function of these materials as
hydraulic barriers. It is therefore, critical to determine the impact
of freezethaw cycles on the hydraulic conductivity of the pro-
posed engineering barriers. To evaluate the performance of the
proposed barrier materials to freezethaw conditions, compacted
BPT samples were subjected to cyclic freezethaw. Samples were
rst prepared at a water content that corresponded to the min-
imum hydraulic conductivity which was determined during pre-
vious compaction accompanied with hydraulic conductivity tests
at 20 C. Hydraulic conductivity tests were carried out over sam-
ples submitted to 1 and up to 5 cycles of freezethaw. ASTM D
6035-96 was followed during freezethaw stress application by
ascertaining that all ice lenses melted and moisture in samples
became at equilibrium. It was arranged so that all of the samples
were submitted to the freezethaw inuence at their moulded
water content. This necessitated the preparation of an equivalent
number of samples and cycles that tested by sample type as
each sample was discarded after testing because it was saturated
during the process. The thermal loading was applied tri-dimen-
sionally under hermetic conditions to prevent modication in
the initial water content during the process of temperature
change by enclosing specimens in sealed plastic bags. Each test
(freeze/thaw and hydraulic conductivity test) was repeated
twice.
844
2.3.4. Wetting and drying cycles
Since tailings impoundments can be located in climatic regions
that are characterised by alternate wet and dry seasons, the spec-
imens were subjected to a series of drying and wetting cycles to as-
sess the effect of such on the hydraulic conductivity of BPT. The
following procedure was adopted. Samples were rst prepared at
a water content corresponding to the minimum hydraulic conduc-
tivity which was determined during previous compaction accom-
panied with hydraulic conductivity tests at 20 C. After
measuring the initial hydraulic conductivity, the samples were re-
moved from the permeameter and dried in an oven at 45 C for
24 h. The 45 C drying temperature was selected to approximate
hot summer temperatures and minimize the possible impact of
high temperature on the bentonite clay mineral. After completion
of the rst drying, the specimens were cooled to room temperature
prior to submerging them in distilled water and their hydraulic
conductivities were determined again. This completed the rst cy-
cle of testing and measurement. Wetting and drying cycles were
repeated a number of times. The compacted BPT specimens were
subjected to zero (as-compacted state), 1, 2, 3, 4 and 6 cycles of
wetting and drying, respectively. Hydraulic conductivity determi-
nations were made after each cycle. Each test (wet-drying and
hydraulic conductivity test) was repeated twice for all specimens
to ensure the repeatability of the results.
3. Results and discussion
3.1. Compaction test results
Fig. 3 presents the compaction curve for plain tailings using an
effort of 592.7 kJ/m3 as prescribed by ASTM D 698-00a. Similarly,
the results of compaction tests conducted on samples made of nat-
ural tailings are presented for the objective of comparison only.
The maximum dry density of 1580 kg/m3 was obtained for a water
content of around 19%. This point corresponds to the one where
minimum voids are achieved. This point eventually provides the
range for best hydraulic conductivity of the material itself and
mixes to be enhanced or improved with additives. The information
gathered from this curve is used as guidance for the approximate
location of the optimum water content for subsequent mixes of
PT with bentonite as additives.
As the optimal water content was determined to lie in the 19%
range, further tests were conducted on BPT mixes containing ben-
tonite in proportions of 2%, 4% and 8% by weight to determine the
effect of the bentonite on the dry density. Proctor tests were con-
ducted on the dry side up to 16% and also on the wet side, and
up to 24% of the optimum water content initially determined in
plain PT. The results of the standard proctor curves obtained by
varying the water content with the compaction tests are plotted
1640
Dry density (kg/m3)
1540
1440
1340
1240
1140
0 0.05 0.1 0.15
Water content
Fig. 3. Compaction curve of plain paste tailings.
M. Fall et al. / Minerals Engineering 22 (2009) 840848
and presented in Fig. 4. In this graph, it appears that the material
with lower maximal dry density (1580 kg/m3) corresponds to the
plain PT. This can result from the fact that even the coefcient of
curvature of 1.36 lies in the range of 13, and the coefcient of uni-
formity of 2.86 is lower than 6, allowing the tailings to be consid-
ered as well graded. It so happens that specimens containing
bentonite develop better packing density. In such conditions, the
voids left after compaction in the plain tailings are more signicant
than the voids left in samples mixed with bentonite. When observ-
ing the samples mixed with 2% of bentonite, one will notice that
the compaction quality has improved, giving rise to a maximum
dry density of 1610 kg/m3. This result of a better packing density
is realized with the presence of bentonite which has ner grains
in comparison to the tailings. The ne grains have the ability to ll
parts of the voids that could be left during the compaction of the
mixture. The best improvement is realized with the addition of
4% of bentonite which gives a dry density of 1640 kg/m3. Up to this
proportion, however, further increasing the bentonite content does
not lead to a proportional increase of the dry density. Instead, one
can notice a drop in the dry density value of the sample made with
8% bentonite. The value fell to about 1610 kg/m3, a value which is
almost similar to the mix with 2% bentonite. Furthermore, in the
sample with 8% bentonite, the peak corresponding to the dry of
optimum appears to be less dened (at). The curve presents a at
summit which is often noticed in compacted clays. This behaviour
and shape in the curve is merely due to the fact that after the pres-
ence of bentonite have contributed to densify the tailings by sup-
plying appropriate nes to ll the voids and build the best
packing density, the relatively high proportion of nes supplied
by the bentonite has a detrimental effect on the mix. The mix tends
to behaves like a clay where the tailings grains are prevented in
having good contact each other, leading them to slide in the plastic
bentonite paste. Otherwise, the bentonite lm plays the role of a
lubricant for the tailings. This property of bentonite is sometimes
used in well drillings (Falode et al., 2008).
3.2. Hydraulic conductivity test results
In the rst stage, permeability tests are conducted along a com-
paction process in a range of 10% to 24% water content (Fig. 5)
compared to the range of 2% to 24% water content for the compac-
tion results of plain PT. Tests for hydraulic conductivity at a water
content less than 9% were impractical to be conducted in the triax-
ial cell, because sampling and manipulating the compacted cohe-
sionless core from the cylindrical cutter failed. Similarly, tests at
a water content higher than 24% were not possible due to the
paste-like consistency of the plain tailings. A continuous narrow
pattern of hydraulic conductivity values was difcult to draw
immediately. However, a certain decreasing trend was recognized
in the scatter of permeabilities ranging from 1.3  10 5
1640
1540
Dry density (kg/m3)
1440
1340 0% Bent
2% Bent
artificial tailings 1240 4% Bent
natural tailings 8% Bent
1140
0.2 0.25 0.3 0 0.05 0.1 0.15 0.2 0.25 0.3
Water content
Fig. 4. Effect of bentonite proportion on the compaction curves of BPT.

2.E-05
Hydraulic conductivity (cm/s)
2.E-05
1.E-05
1.E-05
1.E-05
8.E-06
6.E-06
4.E-06
2.E-06
0.E+00
0 0.05 0.1 0.15 0.2
Water content
Fig. 5. Hydraulic conductivity scatter pattern of plain paste tailing compacted in
the range of 1024% water content
4.1  10 6 cm/s in the testing interval of water contents that were
considered. As reported by tests that were conducted on clays and
sands (Abichou et al., 2004), a minimum hydraulic conductivity is
expected at a certain point past the maximum dry density due to
minimum voids in this interval.
Based on hydraulic conductivity values determined from sam-
ples of raw compacted PT at various water contents, further
improvement was investigated by mixing bentonite in proportions
of 2%, 4% and 8% to tailings by total weight. Referring to results pre-
sented in Fig. 6, it can be noticed that the hydraulic conductivity
decreases in all samples as the water content is increased. For illus-
tration purposes, an example of superposition is presented for the
BPT mix containing 4% of bentonite in Fig. 7.
In this gure, it can be noticed that the slope of the decrease in
hydraulic conductivity is steeper at the dry of optimum up to an
early point at the wet optimum where it tends to remain constant.
It is also important to recall that no further hydraulic conductivity
testing was conducted up to the point where the consistency of the
material was no longer suitable for sampling or handling for
hydraulic measurements. In the global view of hydraulic conduc-
tivity behaviour versus water content and bentonite proportion,
the main observation from Fig. 6 is that hydraulic conductivity
evolution is inversely proportional to the bentonite content. Con-
trary to the compaction curves where the maximum dry density
decreases after optimal bentonite content, in the case of hydraulic
conductivity, the decreasing trend is continuous with the increase
in bentonite content. This is illustrated by the values where sam-
ples with higher bentonite content (8%) show a hydraulic conduc-
tivity of 6.27  10 10 cm/s at wet of optimum (Fig. 6). This low
value is followed by samples with 4% bentonite which show a va-
lue of 9.58  10 9 cm/s and nally, the highest values are recorded
in samples with 2% bentonite where the value is 4.35  10 8 cm/s.
2.E-07
Hydraulic conductivity (cm/s)
2.E-07
1.E-07
8.E-08
4.E-08
0.E+00
0.1 0.15 0.2
Water content
Fig. 6. Effect of different bentonite proportions and water content on hydraulic
conductivity of BPT.
M. Fall et al. / Minerals Engineering 22 (2009) 840848 845
6.E-08
1640
Hydraulic conductivity (cm/s)
5.E-08
Dry density (kg/m3)
1590
4.E-08
3.E-08
1540
2.E-08
1490
Compaction curve
1.E-08
Hydraulic conductivity
0.25 0.3
1440 0.E+00
0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 0.26
Water content
Fig. 7. Variation of hydraulic conductivity within the compaction curve of BPT (4%
Bentonite).
The explanation for this behaviour may entail two reasons. First,
ow in porous media relies on void spaces and interconnectivity
for uid movement. When proceeding with compaction, the higher
dry density is an indication of minimum voids achieved during the
process. This point would theoretically correspond to the mini-
mum hydraulic conductivity under minimum void ratio. On the
other hand, the fact that higher bentonite content more efciently
occupies the reduced pores left during compaction and contributes
to their further blockage, it thereby hinders the movement of u-
ids. Furthermore, the presence of bentonite involves the effect of
cation exchange capacity which is a process by which water mole-
cules are attracted by its double layer diffuse, signicantly retard-
ing the ow (Dixon et al. 1999). Finally, for the sample being
restrained in the triaxial cell, swelling pressure developed by ben-
tonite plays a signicant role in obstructing pores when hydrated
by the permeant. It is important to recall that the hydraulic con-
ductivities were conducted at the nearest same effective stress
and hydraulic gradient for all of the specimens. This provided a
uniform reference and prevented any ambiguity of interpretation
with respect to variable deformation intensities which could arise
in different loading conditions for each specimen. Hence, this
would result in shrinking or dilating the voids in different propor-
tions as reported by Shaee (2008).
3.3. Effect of freezethaw cycles on the hydraulic conductivity of BPT
Typical results of the effects of cyclic freezethaw on the
hydraulic conductivity of BPT are presented in Figs. 8a and 8b. In
Fig. 8a, the measured stabilized permeabilities for BPT with differ-
ent bentonite contents are plotted versus the number of freeze
thaw cycles. Fig. 8b shows the normalized hydraulic conductivity
of BPT with different bentonite contents for different freezethaw
cycles. The normalized hydraulic conductivity is expressed as the
ratio between the permeability obtained in the thawed BPT after
2% Bent N cycles (kN) and the permeability initially in the unfrozen BPT
4% Bent
(k0). From Fig. 8a, it can be observed that for 8% bentonite, the per-
meability values uctuate between 4.8  10 10 and 1.2  10 9 cm/
8% Bent
s which represents a 153% increase of the hydraulic conductivity,
or an increase factor of 2.5 (Fig. 8). In another way, the increase
is less than one order of magnitude. For the 4% bentonite mixes
presented in Fig. 8, minor uctuations in hydraulic conductivity oc-
cur in the range of 2.8  10 91.3  10 8 cm/s within the ve
freezethaw cycles which represents an increase of 355% or an in-
crease factor of 4.6 (Fig. 8b). This increase is also less than one or-
0.25 0.3
der of magnitude. The values for the 2% bentonite appear to
decrease slightly from 4.69  10 82.14  10 8 cm/s with the
number of cycles (Fig. 8a). It represents a decrease factor of only
0.46. In terms of percentage, it represents a decrease of 54% in
846 M. Fall et al. / Minerals Engineering 22 (2009) 840848

8.E-08 ing will take place with respect to the vertical water ow
a
Hydraulic conductivity (cm/s)

2% Bent
7.E-08 (Viklander, 1998).
4% Bent
6.E-08 8% Bent
In the case of the 2% bentonite BPT, the appearance of a slight
decrease of the permeability with freezethaw cycles can be attrib-
5.E-08
uted to the fact that it has a lower initial dry density, i.e. higher ini-
4.E-08
tial void ratio than the 8% and 4% bentonite BPT (Fig. 3). Due to its
3.E-08
less dense structure, a larger volume decrease during thawing will
2.E-08 occur in the 2% bentonite BPT than in the 8% and 4% bentonite BPT.
1.E-08 This volume decrease, due to the consolidation of the BPT matrix,
1.E-10 will increase the density. As well, larger tailings particles will come
0 1 2 3 4 5 6 closer, thereby leading to the decrease of the permeability. Similar
Cycles behaviour was observed on loose and dense ne-grained soils sub-
1.E+01 jected to freezethaw cycles (Viklander and Knutsson, 1998).
b
Hydraulic conductivity (k N /k0)

9.E+00 2% Bent In this present study, the changes of hydraulic conductivity of

8.E+00
7.E+00
6.E+00
5.E+00
4.E+00
3.E+00
2.E+00
1.E+00
1.E-10
0 1
Fig. 8. Evolution of the (a) hydraulic conductivity and (b) normalized hydraulic
conductivity (kN/k0) of BPT with freezethaw cycles.
the hydraulic conductivity (Fig. 9). The increase in permeability ob-
served for 8% and 4% bentonite BPT and after three freezethaw cy-
cles may be attributed to the formation of a network of micro-
cracks (not visible to the naked eye) and larger pores during freez-
ing. Indeed, when the BPT is exposed to freezing and thawing, the
pore water is turned into ice, and ice lenses can form. Ice forces will
act on the BPT skeleton due to a 9% volume increase during freez-
ing. This freezing pressure will push apart some tailings and clay
particles. This will result in formation of larger pores and/or mi-
cro-cracking of the BPT skeleton, and an increase of the pore con-
nectivity. However, from Fig. 8, it can be noted that after three
freezethaw cycles, the permeability of the 8% and 4% bentonite
decreases until reaching its initial value (Fig. 8b). This observed de-
crease can be attributed to the self-healing ability of the BPT. Due
to their low content in clay (compared to natural expansive soil),
the BPT are less cohesive than a natural expansive soil. When the
ice melts in the BPT, the cracks developed during freezing may
close after thawing, owing to the relatively low cohesion in the
BPT. This is because in less cohesive soil, grains may relocate dur-
ing thawing, which block cracks and cavities. Thus, some self-heal-
2.5E-08
4% Bent BPT observed with freezethaw cycles are relatively low (less than
8% Bent
one order of magnitude in hydraulic conductivity) compared to the
more substantial changes reported on materials such as compacted
clay (Kim and Daniel, 1992; Othman and Benson, 1993). Aside from
the mechanisms explained above, the freezing rate used in this
study contributes to a small change of the hydraulic conductivity
during the freezingthawing mechanisms. In the present study,
the heat removal rate brings moisture in the sample to the solid
2 3 4 5 6
state in a very fast period of time. Pre-test results from the evalu-
Cycles ation of the freezing rate of the samples have shown that it re-
quires only 2 h for the temperature to drop from 0 C to 19 C.
In this small interval of phase change, the formation of ice lenses
and advancement of the front (Hewitt and Daniel, 1997) are so fast
that no migration of water takes place, which can form large ice
lenses and affect the fabric of the material. Based on the denition
used by Othman and Benson (1993a) the freezing rate applied to
the samples in the present study would be considered as being
1.04  10 4 m/s while Othman and Benson (1993a) considered a
rate of 10 6 m/s as being fast. Therefore, freezing will take place
globally as a whole in the sample. This way, all ice lenses will de-
velop simultaneously, preventing a suction effect and cancelling
uid movement to large individual ice lenses which would exercise
stresses by their higher volume inside the material. The stresses
would generally be translated into cracks which would have an ef-
fect of damaging the material fabric (Hohmann-Porebska, 2003;
Othman and Benson, 1993). This assumption is supported by the
fact that all samples are inspected for visual cracking on their sur-
faces or for other possible damage. No visual signs are detected,
only an apparent lm of ice covered some parts of the samples.
One explanation is that each individual ice lens had built up on
its original pore position and stayed immobile as gradient termi-
nated early. Any expansion or shrinkage would have taken place
locally without signicantly extending the pore network connex-
ion. This inhibition of ow may become more pronounced by the
low hydraulic conductivity of the material, where more time
would have been required for the uid to advance within the cold
4.0 front. For the cycle completion, during the thawing, the fabric
Relative hyd. conductivity (kN/k0)

3.5
would only restore to the original fabric of samples without notice-
Hydraulic conductivity (cm/s)

2.0E-08 able damage.

3.0

1.5E-08 2.5 3.4. Effect of wetdrying cycles on the hydraulic conductivity of BPT

2.0
The obtained results show that the hydraulic conductivity of
1.0E-08
1.5 BPT samples compacted wet of optimum increases within one to
1.0 two drying cycles. However, this increase is less than one order
5.0E-09 of magnitude. For illustration purposes, an example of the typical
hydr. Conductivity 0.5
Relative hydr. Conductivity results from the impact of wetdrying cycles on the hydraulic con-
0.0E+00 0.0 ductivity and relative (normalized) hydraulic (kN/k0) of BPT con-
0 1 2 4 6 taining 4% bentonite compacted wet optimum is presented in
Number of cycles
Fig. 9. From this gure, it can be observed that after two wetdry-
Fig. 9. Evolution of hydraulic conductivity and normalized hydraulic conductivity ing cycles, the permeability values increase from 5.8  10 9 (initial
(kN/k0) of BPT with wettingdrying cycles. value of hydraulic conductivity) to 2.0  10 9 cm/s which repre-
 M. Fall et al. / Minerals Engineering 22 (2009) 840848
sents only an increase factor of 3.5 (Fig. 9). In other words, the in-
crease is less than one order of magnitude. This increase in hydrau-
lic conductivity can be attributed to the desiccation cracks formed
in the BPT. These cracks are induced by the evaporation of the
water and consequent shrinkage. These cracks introduce a coarser
pore structure and create pathways for transport of uids, result-
ing in an increase of the hydraulic conductivity of the BPT. Similar
observations were made on clays subjected to wetdrying cycles
(e.g. Omidi et al., 1996; Rayhani et al., 2007). This observed low in-
crease of hydraulic conductivity is in good agreement with the re-
sults reported by Eigenbrod (2003), who demonstrated that Na
bentonitesand mixes do not experience permeability changes
subsequent to drying and wetting cycles. This is because the swell-
ing pressure of the Nabentonite exceeds the shearing resistance of
the paste and therefore, is able to completely ll the void space be-
tween the sand particles (Eigenbrod, 2003).
However from Fig. 9, it can be noted that the hydraulic conduc-
tivity decreases after two wetdrying cycles. It is believed that this
decrease of hydraulic conductivity is associated with self-healing
processes that affect the various soil types by different degrees
(Eigenbrod, 2003). Since the hydraulic conductivity of all samples
are conducted at the same level of effective stress, the clogging
of desiccation cracks by particles eroded from the fracture surfaces
during permeation and swelling of the bentonite particles near the
cracks should be considered as the main mechanisms responsible
for the healing of the desiccation cracks.
3.5. Cost effectiveness of BPT engineering barrier
For the proposed barrier material to be of interest, it should be
capable of adding value in comparison to existing alternatives
available in the construction practices aside from recycling capac-
ity. Based on the EPA minimum requirement for hydraulic conduc-
tivity and relatively good behaviour under freezethaw and wet
drying conditions, a cost analysis was conducted on the BPT mixes.
It is obvious that the additive materials mainly control the prices in
such mixtures. The investigation was conducted taking into ac-
count, variable additive proportion. A balance can be established
between saving costs and an acceptable level of permeability. To
appreciate such, costs were compared to reference barriers made
of sand and clay mixed with 12% bentonite. At a unit kilogram price
of $3.96 for bentonite, for any determined mass M of barrier, the
cost for 12% bentonite is estimated at $0.47 while the costs for
8%, 4% and 2% of bentonite are estimated at $0.32, $0.16, and
$0.08, respectively. Still using the 12% bentonite cost as a reference,
the benet realized can be estimated at 33%, 66% and 83% when
using BPT barriers containing 8%, 4% and 2% of bentonite, respec-
90%
80%
70%
60%
Benefit
50%
40%
30%
20%
10%
0%
12% 8% 4% 2%
Bentonite
Fig. 10. Relative benet realization when using BPT as compared to a conventional
12% bentonitesand mixture.
847
tively. The benet is presented graphically for ease of comparison
in Fig. 10.
4. Conclusions
This study has produced valuable data and information. The
compacted BPT mixtures exhibit low hydraulic conductivities. Val-
ues that are as low as 6.27  10 10 cm/s and 4.109 cm/s are ob-
tained in 8% and 4% bentonite, respectively. The study has also
shown that the proposed BPT barrier resists environmental stres-
ses (freezethaw and wetdrying cycles) relatively well. The
freezethaw cycles only have a light adverse effect on the hydrau-
lic conductivity of BPT. None of the samples developed changes in
the order of magnitude. Furthermore, the increase of hydraulic
conductivity caused by the wetdrying cycles is less than one order
of magnitude. It is found that after a specic number of freeze
thaw or wetdrying cycles, the hydraulic conductivity of the BPT
tends to decrease. The good response of the composite material
to environmental stresses can be mainly attributed to possible
resealing of micro-cracks with self-healing properties entailed by
the high swelling additives. It can be assumed that in the eld with
increasing stress levels, less freezing or desiccation induced cracks
will be formed thereby increasing the resistance of BPT to the
aforementioned environmental stresses. This is because the com-
pressive stress from the overburdening will help close pre-existing
cracks, and prevent the development of freezing or desiccation
cracks.
Cost assessment of the material in reference to 12% bentonite
(usually used in sand mixtures) shows that economical gains as
high as 83%, 67% and 33% can be achieved when using 2%, 4%
and 8% of bentonite, respectively. However, one has to evaluate
the ratio hydraulic conductivity versus economical returns.
From the promising results obtained, it can be concluded that
the BPT mixture has the potential to be used for liner or cover
materials for mine waste containment facilities. Furthermore, its
use can be extended to municipal waste containment facilities
(landlls). However, further research is needed to provide an
understanding of the mechanical behaviour, water retention ability
and durability of BPT. Moreover, the impact of leachates or acid
water on the hydraulic conductivity of BPT should be investigated.
Research in this area has not been done for the project in this pa-
per, but currently being implemented in a new project.
Acknowledgments
The writers would like to acknowledge the National Sciences
and Engineering Research Council of Canada (NSERC), University
of Ottawa, and National Research Council (NRC) of Canada.
References
Abichou, T., Edil, T.B., Benson, C.H., Tawq, K., 2004. Hydraulic conductivity of
foundry sands and their use as hydraulic barriers. Geotechnical Special
Publication 127, 186200.
Amde, A.M., Dinku, A., Davies, O., 2007a. Abatement of acid mine drainage using
industrial waste products. Advances in Construction Materials. Springer. pp.
653665.
Amde, A.M., Dinku, A., Davies, O., 2007b. Abatement of acid mine drainage using
industrial waste products. Advances in Construction Materials. Springer, Berlin.
13p.
Aubertin, M., Bussiere, B., Chapuis, R.P., 1996. Hydraulic conductivity of
homogenized tailings from hard rock mines. Canadian Geotechnical Journal
33 (3), 470482.
Benzaazoua, M., Fall, M., Belem, T., 2004. A contribution to understanding the
hardening process of cemented pastell. Minerals Engineering 17 (2), 141152.
Daniel, D.E., 1993. Geotechnical Practice for Waste Disposal. Chapman and Hall,
London (683pp).
Dixon, D.A., Graham, J., Gray, M.N., 1999. Hydraulic conductivity of clays in conned
tests under low hydraulic gradients. Canadian Geotechnical Journal 36 (5), 815
825.
848 M. Fall et al. / Minerals Engineering 22 (2009) 840848
Eigenbrod, K.D., 2003. Self-healing in fractured ne grained soils. Canadian
Geotechnical Journal 40 (2), 435449.
Elberling, B., Nicholson, R.V., Scharer, J.M., 1994. A combined kinetic and diffusion
model for pyrite oxidation in tailings: a change in controls with time. Journal of
Hydrology 157, 4760.
Fall, M., Merkel, B. (Eds.), 2001. Kontamination aus der Nutzung von Ressourcen
Probleme und Lsungen. In: Proceedings of the 1st conference of Geo-
Environment of the Graduatenkolleg (272): Geowissenschaftliche und
Geotechnische Umweltforschung TU Freiberg Germany, Wissen. Mitt.
Institut fr Geologie, Heft 18. 20 Beitrge, 200 p.
Fall, M., Benzaazoua, M., Ouellet, S., 2004. Effect of tailings properties on paste
backll performance. In: Proceedings of the 8th International Symposia on
Mining with Backll, Beijing, China, pp. 193202.
Fall, M., Benzaazoua, M., Saa, E., 2008. Mix proportioning of underground cemented
paste backll. International Journal of Tunnelling and Underground
Construction 23 (1), 8090.
Falode, O.A., Ehinola, O.A., Nebeife, P.C., 2008. Evaluation of local bentonitic clay as
oil well drilling uids in Nigeria. Applied Clay Science 39 (12), 1927.
Hewitt, R.D., Daniel, D.E., 1997. Hydraulic conductivity of geosynthetic clay liners
after freezethaw. Journal of Geotechnical and Geoenvironmental Engineering
123 (4), 305313.
Hohmann-Porebska, M., 2003. Microfabric effects in frozen clays in relation to
geotechnical parameters. Applied Clay Science 21 (12), 7787.
Kashir, M., Yanful, E.K., 2001. Hydraulic conductivity of bentonite permeated with
acid mine drainage. Canadian Geotechnical Journal 38 (5), 10341048.
Kim, W., Daniel, D.E., 1992. Effects of freezing on hydraulic conductivity of
compacted clay. Journal of Geotechnical Engineering 118 (7), 10831097.
Landriault, D. 1995. Paste backll mix design for Canadian underground hard rock
mining. In: Proceedings of the 97th Annual General Meeting of the CIM. Rock
Mechanics and Strata Control Session. Nova Scotia, Halifax, pp. 229238.
Leppinen, J.O., Salonsaari, P., Palosaari, V., 1997. Flotation in acid mine drainage
control: beneciation of concentrate. Canadian Metallurgical Quaterly 36 (4),
225230.
McLaughlin, J., Stuparyk, R. 1994. Evaluation of low sulphur rock tailings production
at INCOs Clarabelle mill. In: Yalcin, (Ed.), Proceedings of the Innovations in
Mineral processing Conference, Sudbury, pp. 129146.
Nicholson, R.V., Gillham, R.W., Cherry, J.A., Reardon, E.J., 1989. Reduction of acid
generation in mine tailings through the use of moisture-retaining cover layers
as oxygen barriers. Canadian Geotechnical Journal 26 (1), 18.
Omidi, G.H., Thomas, J.C., Brown, K.W., 1996. Effect of desiccation cracking on the
hydraulic conductivity of a compacted clay liner. Water, Air, and Soil Pollution
89 (12), 91103.
Othman, M.A., Benson, C.H., 1939a. Effect of freeze-thaw on the hydraulic
conductivity of three compacted clays from Wisconsin. Transportation
Research Record 1369, 118125.
Othman, M.A., Benson, C.H., 1993. Effect of freezethaw on the hydraulic
conductivity and morphology of compacted clay. Canadian Geotechnical
Journal 30 (2), 236246.
Qiu, Y.J., Sego, D.C., 2001. Laboratory properties of mine tailings. Canadian
Geotechnical Journal 38 (1), 183190.
Rayhani, M.H.T., Yanful, E.K., Fakher, A., 2007. Desiccation-induced cracking and its
effect on the hydraulic conductivity of clayey soils from Iran. Canadian
Geotechnical Journal 44 (3), 276283.
Ritcey, M.G., 2005. Tailing management in gold plants. Hydrometallurgy 78, 320
(1-2 SPEC. ISS).
Robertson, A.M., Robertson, S., Kirsten. 1998. Alternative acid mine drainage
abatement measures, Report, 33p.
Shaee, A., 2008. Permeability of compacted granuleclay mixtures. Engineering
Geology 97 (34), 199208.
Struthers, S. (1999). Constructive recycling of mine wastes. In: Proceedings of
Fourth International Conference on Tailings and Mines Waste. Colorado, USA,
pp. 1725.
Verburg, R.B.M. (2001). Use of paste technology for tailings disposal: Potential
environmental benets and requirements for geochemical characterization.
IMWA Symposium, pp. 113.
Viklander, P., Knutsson, S., 1997. Permeability changes in a ne-grained till due to
freezing and thawing. In: Proceedings 8th International Symposium on Ground
Freezing. A.A. Balkema, Rotterdam, pp. 193202.
Yalcin, T., Papadakis, M., Hmidi, N., Hilsher, B. 2004. Desulphurization of Placer
Domes Muselwhite mine gold cyanidation tailings. CIM bulletin Technical
Paper, pp. 17.
Yanful, E.K., 1993. Oxygen diffusion through soil covers on sulphidic mine tailings.
Journal of Geotechnical Engineering ASCE 119 (8), 12071228.
Yanful, E.K., Mousavi, S.M., De Souza, L.P., 2006. A numerical study of soil cover
performance. Journal of Environmental Management 81 (1), 7292.
Yilmaz, E., Kesimal, A., Deveci, H., Ercikdi, B. 2003. The factors affecting the
performance of paste backll; physical chemical and mineralogical
characterization. First Engineering Sciences Congress for Young Researcher
(MBGAK03), Istanbul.
Yilmaz, E., Kesima, A., Ercidi, B. 2004. Strength development of paste backll
simples at Long term using different binders. In: Proceedings of 8th Symposium
MineFill04, China, pp. 281285.