Landfills and Waste Containment

Performance of Compacted Silty Loess as Landfill Liner Material

G.M. Aiassa
Universidad Tecnológica Nacional, Facultad Regional Córdoba, Argentina
(gaiassa@scdt.frc.utn.edu.ar)
P.A. Arrúa
Universidad Tecnológica Nacional, Facultad Regional Córdoba, Argentina
(parrua@cbasicas.frc.utn.edu.ar)

ABSTRACT This work investigated the feasibility of using silty loess from the centre of Argentina as
landfill liner material. The study were conducted both at laboratory and field scales. The materials tested
were natural silt and silt-bentonite mixtures. For the laboratory tests infiltrations cell were used, and samples
with different method, dry unit weight and water content were prepared by direct compaction inside the
mold. For field study a small test pad was constructed and sealed double ring infiltrometers were used. For
specials compaction conditions, the hydraulic conductivities were on the order of 10-9 m/s and achieve the
required value by the international regulations.

INTRODUCTION In this study, laboratory and field test were

Compacted soils are widely used for liner systems
to isolate waste materials from the environment.
The effectiveness of the system is controlled by
the permeability and liner thickness. Generally,
the regulations establish a maximum allowable
permeability and a minimum liner thickness. For
EPA (1993) the minimum thickness needed to
minimize the number of cracks or imperfections is
two feet (> 0,60 m) and the permeability must be
lower than 1×10-9 m/s.
Utilization of local excavated soils from different
engineering projects may be an economically
convenient alternative (Wijeyesekera et al., 2001;
Met et al., 2004; Bozbey and Guler, 2005). Use of
these materials in pure or stabilized form as
compacted soil liners, enables many small towns,
whose economy does not allow the use of
geosynthetic materials, to solve the problem of
waste disposal through environmentally sound and
affordable alternative in cost.
conducted to explore if the silty loess from the
centre of Argentina can be used as landfill liner
material. Test were conducted on both pure a
stabilized with sodium bentonite. The laboratory
tests include falling head permeability carried out
on samples prepared under different compaction
conditions. Field tests were conducted by sealed
double ring infiltrometers (SDRI) on a test pad
constructed in the University campus. The effect
of different compaction condition and aggregate
amount of sodium bentonite were studied.
LITERATURE REVIEW
The centre of Argentina, in particular the Province
of Córdoba, is mostly covered by soil deposited by
wind. These soils, called loess, have special
characteristics related to their geological origin.
They consist mainly of silt, and small amounts of
fine sand and clay (Teruggi, 1957; Zárate, 2003).
Loessic undisturbed soils, have a low density
macroporous structure, and may undergo large
deformations by increasing, isolated or combined,
external loads and water content. Because of this
346 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

group are in the unstable soils, and are classified The aggregate used is a natural sodium bentonite
as collapsible soils (Rocca et al., 2006). To use supplied by the company MINARMCO SA. The
these soils in the construction of embankments material properties are presented in Table 1.
improvement by compaction is needed. However,
only compaction may not be enough to meet the TABLE 1 Loess soil and bentonite properties
permeability requirements or specifications ------------------------------------------------------------
established by technical studies, so it is necessary Property Soil Bentonite
to use some type of stabilizer added to improve ------------------------------------------------------------
Liquid limit (%) 24,8 395
performance.
Plastic Index (%) 4,4 344
The permeability of compacted fine soils mainly
depends on the structure formed during the Particles <0,075mm (%) 85 100
manufacturing process. The most influential
variables on the structure of compacted soils are Particles <0,002mm (%) 7 80
water content, energy applied and compaction ------------------------------------------------------------
method. In general, permeability decreases when
the water content is close to optimum or slightly The addition of bentonite changed soil properties
wet of optimum, increase energy and use the in the way it is presented in Figure 1. When the
knead method (Mitchell et al., 1965; Benson et al., bentonite content increased, the liquid limit and
1994; Wang and Benson, 1995). plasticity index increased too, in a linear
relationship.
The soil composition can significantly affect the
50
permeability. Soils with greater amounts of active Liquid limit
clay minerals and fine particles generally have 40 Plasticity index
lower permeability, because they increase the
thickness of the diffuse double layer. In general, 30
(%)

permeability decreases with increases in liquid 20

limit and plasticity index, because these indicators
are directly related to the mineralogy and clay 10
content of soil (Benson et al., 1994; Mitchell and
0
Soga, 2005).
0 2 4 6 8 10 12 14 16
Bentonite content (%)
The use of sodium bentonite as an aggregate to Fig.1 Index properties of loess-bentonite mix.
reduce the permeability of compacted soils has
been studied by several authors (Gleason et al., Regarding the compaction conditions, the increase
1997; Kaya and Durukan, 2004; Roberts and in bentonite content decreased the maximum dry
Shimaoka, 2008). Bentonite is a clay mineral unit weight and slightly increased the optimum
consisting primarily of montmorillonite. This compaction water content.
mineral has strong attraction of water and
experiment swelling, which results in a reduced EXPERIMENTAL PROCEDURE
hydraulic conductivity. Falling head permeability tests were carried out in
laboratory. Rigil-wall cell was employed and
MATERIALS samples were tested in the Standard Proctor
The soil used is a silty-clay loess. The samples compaction mold (ASTM, D-5856).
were taken in the area south of Córdoba city, at
depths of 1,0 to 5,0 meters. It is a low plasticity In field tests sealed double ring infiltrometer was
soil and is designated as CL-ML in the Unified employed (ASTM, D-5093). An experimental
System of Soil Classification (USCS) (Table 1).
 Aiassa and Arrúa 347

compacted liner was constructed. The used was and increase the bentonite content. The addition of
pure loess without bentonite added. The bentonite reduces its effect when the compactness
dimension was 1,50 by 1,50 meters in plan and of the soil increases.
0,40 meters thickness.
1,0E-06 17
RESULTS AND ANALYSIS
Figure 2 shows the results obtained for pure loess

Dry unit weight, Ȗd (kN/m3)

BC: 5% 16
soil samples compacted under Standard Proctor 1,0E-07

Permeability, k (m/s)
energy and different water content. In this case
bentonite was not added (BC=0%). Near the 15
optimum water content the minimum permeability
1,0E-08
was obtained.
14
1,0E-05 18
1,0E-09
13
Dry unit weight, Ȗd (kN/m3)

Permeability test
1,0E-06 17
Compaction curve
Permeability, k (m/s)

BC: 0%
1,0E-10 12
1,0E-07 16 10 12 14 16 18 20 22 24
Molding water content, wc (%)
Fig.3 Influence of structure on the permeability of
1,0E-08 15 compacted loess-bentonite mix under Standard Proctor
energy (Bentonie added 5%).

1,0E-09 14 1,0E-05
Permeability test 0%BC
Permeability, k (m/s)

Compaction curve 5%BC

1,0E-06
1,0E-10 13 7%BC
Maximum
6 8 10 12 14 16 18 20 22 1,0E-07 compacted
Molding water content, wc (%) loess
1,0E-08
Fig.2 Influence of structure on the permeability of
compacted loess under Standard Proctor energy (no 1,0E-09
Natural
loess
bentonite added).
1,0E-10
In the samples of bentonite added, lower 0,35 0,40 0,45 0,50 0,55
permeabilities were achieved (Figure 3). Particular Porosity, n
important reduction can be observed on the dry Fig.4 Permeability of samples prepared with different
side of optimum. Both, with or without bentonite, compactness and sodium bentonite added (BC). Static
a permeability under 1×10-9 m/s can be achieved compaction method.
under particular compaction conditions.
The field results are presented in Figure 5. Two
Figure 6 shows results on samples prepared by tests were realized and same results were obtained.
static compaction at different levels of Field results are consistent with laboratory.
compactness and added of bentonite. Samples
were prepared between limits defined by natural CONCLUSIONS
loessic soil porosity, and that for the compacted The main conclusions of this work are:
soil to optimum water content and Standard x The increased in content of sodium bentonite
Proctor energy. There are significant reductions in caused significant reductions in permeability,
the permeability when the soil porosity decrease
348 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

independent of the dry unit weight and
compaction water content.
x The addition of sodium bentonite in loess
soils, improve the hydraulic performance of
civil works such as landfill liner.
x The compactness of the soil was one of the
main variable affect the permeability.
800
Infiltration, V (cm3)
hydraulic containment applications. Journal of
Geotechnical and Geoenvironmental
Engineering, ASCE, 123(5), 438-445.
Kaya, A. and Durukan, S. (2004). Utilization of
bentonite-embedded zeolite as clay liner.
Applied Clay Science, 25, 83-91.
Met, I.; Akgün, H. and Tükmenoglu, A.G. (2004).
Environmental geological and geotechnical
investigations related to the potential use of

Ring 01
600 Ring 02 Ankara clay as a compacted landfill liner
material, Turkey. Environmental Geology,
400 47(2), 225-236.
Ȗd = 16,4 kN/m3 Mitchell, J.K. and Soga, K. (2005). Fundamentals
w = 16,2%
200 of soil behavior. Third Edition. John Wiley &
k = 2×10-9 m/s
Sons.
0
Mitchell, J.K.; Hooper, D.R. and Campanella, R.G.
0 100 200 300 400 500 600 700 (1965). Permeability of compacted clay.
Time, t (hours)
Journal of the Soil Mechanics and Foundations
Fig.5 Double ring infiltration on compacted soil liner.
Division, ASCE, 91(4), 41-65.
ACKNOWLEDGMENT Roberts, A.A. and Shimaoka, T. (2008). Analytical
This work was supported by the National study on the suitability of using bentonite
Technological University and the Ministry of coated gravel as a landfill liner material. Waste
Science and Technology of the Province of Management, 28, 2635-2644.
Cordoba. In particular the authors thank the Rocca, R.J.; Redolfi, E.R. and Terzariol, R.E.
support of UTN, Facultad Regional Córdoba, Civil (2006). Características geotécnicas de los loess
Engineering Department. de Argentina. Revista Internacional de
Desastres Naturales, Accidentes e
REFERENCES Infraestructura Civil, 6(2), 149-166.
ASTM, (2002). Soil and Rock. Annual book of Teruggi, M.E. (1957). The nature and origin of
ASTM standards. Argentinian loess. Journal of Sedimentary
Benson, C.H.; Zhai, H. and Wang, X. (1994). Petrology, 27, 322-332.
Estimating hydraulic conductivity of compacted Wang, X. and Benson, C.H. (1995). Infiltration
clay liners. Journal of Geotechnical and saturated hydraulic conductivity of
Engineering, ASCE, 120(2), 366-387. compacted clay. Journal of Geotechnical
Bozbey, I. and Guler, E. (2005). Laboratory and Engineering. ASCE, 121(10), 713-722.
field testing for utilization of an excavated soil Wijeyesekera, D.C.; O´Connor, K. and Salmon,
as landfill liner material. Waste Management, D.E. (2001). Design and performance of a
26, 1277-1286. compacted clay barrier through a landfill.
EPA (1993). Solid waste disposal facility criteria. Engineering Geology, 60, 295-305.
U.S. Environmental Protection Agency, Zárate, M.A. (2003). Loess of southern South
Technical Manual. EPA/530-R-93-017. America. Quaternary Science Reviews, 22,
Gleason, M.H.; Daniel, D.E. and Eykholt, G.R. 1987-2006.
(1997). Calcium and sodium bentonite for
 Pond Ash As A Landfill Liner – A Review

Aditi Varma
UG Student,Delhi College of Engineering
(adizmail2006@gmail.com)
Mayank Saxena
UG Student,Delhi College of Engineering
(saxena.mayank.2007@gmail.com)
Ankit Garg
UG Student,Delhi College of Engineering
(ankit.civil.dce@gmail.com)
Neelesh Khanna
UG Student,Delhi College of Engineering
(neilkhanna123@gmail.com)
Pranjal Sharma
UG Student,Delhi College of Engineering
(pranjal22688@gmail.com)
Prof.Raju Sarkar
Deptt.of Civil Engg. Delhi College of Engineering
(sarkar.raju@yahoo.com)

ABSTRACT: With increasing use of thermal power as an energy source, coal ash utilization has become the
need of the hour. It has long been regarded as a waste material having potential environmental implications.
However with identification of its various geo-technical characteristics it can be fruitfully utilized. Due to
presence of certain radioactive and toxic elements, pond ash is likely to contaminate soil and ground water
system, which impresses upon the need of a landfill liner. In this paper the use of pond ash as a potential
landfill liner has been discussed. The effect of Carbonic Aluminate Salt (CAS) as a hardening agent to
eliminate seepage of various hazardous elements into ground water has also been highlighted.

Keywords : Pond ash; landfill liner; Carbonic Aluminate Salt (CAS); Pozzolana;Stabilisation of pond ash

INTRODUCTION

Evidently the main by-product of thermal power
stations is coal ash which further gives rise to fly ash.
It is basically the fly ash from which pond ash is
obtained. As mentioned earlier, pond ash had hitherto
been regarded as a waste material. However in the
ongoing research works, the possibility of using
locally available fine grained soil to improve the
compaction and permeability properties of pond ash
is being considered. An attempt can be made to use
pond ash as a landfill liner when small quantities of
fine grained soil and CAS(Carbonic Aluminate Salt)
are added to it.Being used as a landfill liner,pond ash
can be fruitfully utilized in preventing various
hazardous elements from seeping into ground water.
Before seeking ways of utilization and stabilization
of these by-products mainly pond ash, following
definitions need to be mentioned.
350 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE1 Chemical properties cementitious properties but when present in finely

divided form, along with moisture, reacts chemically
COMPOUNDS %(by mass) to add certain properties of concrete. Fly ash, Micro-
Silica, Rice-Husk, GGBS (Ground Granulated Blast
SiO2 61.85
furnace Slag), Metakaoline are a few examples of
Al2O3 30.48
pozzolanic materials.
Fe2O3 3.23
SiO2+ Al2O3+ Fe2O3 95.56 Coal Ash: This term is encompasses all terms listed
TiO2 2.19 above
K2 O 0.72
P2O5 0.90 Boiler slag is a molten ash collected at the base of
CaO 0.64 slag tap and cyclone boilers that is quenched with
LOI 1.27 water and shatters into black, angular particles having
a smooth glassy appearance
Few Definitions:
Bottom Ash: This refers to the ash collected from PROPERTIES OF POND ASH
the bottom of a furnace, after being ground in a One of the basic principles of pond ash stabilization
clinker grinder unit. is evaluation of properties of pond ash.

Coarse Pond Ash or Pond Ash at Inflow

TABLE: Physical Properties
Point: This refers to the coarser fraction of pond ash TABLE 2 Physical Properties
usually counteracted at the inflow point in an ash
pond where the slurry from PARAMETERS VALUES
the pipeline is received. Specific gravity 2.00
Bulk density, g/cm3 0.95
Porosity, % 52.50
Fine Pond Ash or Pond Ash at Outflow Point: This Coeff.ofpermeablty,cm/s 5.5472×10– 4
refers to the finer fraction of pond ash usually
counteracted near the outflow point (decant pond Particle size analysis results
zone) in an ash pond where clear water is decanted. D10, µm 8.74
D50, µm 28.10
Chimney Ash: This term is synonymous to fly ash. D90, µm 74.26
Specific surface area, m2/cc 0.293
Boiler Ash: This term is synonymous to bottom ash.
Chemical Properties

The results of chemical analysis show that pond ash
comprises of oxides of silicon (SiO2), aluminum
(Al2O3), iron (Fe2O3) and very little amount of
calcium (CaO) and some minor quantities of other
metal oxides like titanium and potassium etc. the sum
of oxides of silicon (SiO2), aluminum (Al2O3) and
iron (Fe2O3) is greater than 70%. Generally increase
in lime content causes slight change in liquid limit
and considerable increase in Plasticity index.
Pozzolana is a natural or artificial material
containing silica in reactive form. In other words,
pozzolana is a siliceous or siliceous & aluminous
material which otherwise possess little or no
Such pozzolanic materials, in combination with OPC
or PPC/PSC can sustain extreme aggressive
water/environmental exposure conditions.
As is evident, pond ash is mainly composed of Al2O3,
SiO2 and Fe2O3 and very little CaO. It is Non-
pozzolanic due to very low fresh lime content.
However an addition of 7% slaked lime of the total
solid (pond ash and lime mixture)yields a very good
strength. Unlike pond ash it is well known that the
presence of free lime content in fly ash has a major
effect on the self-hardening potential and pozzolonic
reactivity.

POND ASH STABILISATION
This can be achieved by addition of suitable
admixtures or stabilizers in order to improve the
stability or bearing power of the ash. Better the
degree of mixing higher is the strength.
There are some additives to improve properties
• Lime
• Sodium hydroxide
• Sodium Carbonate
• Calcium Chloride
Need for Pond Ash Stabilization
• Limited Financial Resources to provide an
alternate source for landfill.
• Effective utilization of pond ash and other
suitable stabilizing agents.
• Encouraging the use of Industrial Wastages
in building low cost of landfill.
UTILISATION OF POND ASH
With increase in number of coal based thermal power
stations, there has been enormous generation of fly
ash as well as pond ash. Ash disposal involves design
and installation of ash ponds, which not just cover
quite a large area at each plant site, but also create
aesthetic as well as hygienic environmental
implications. This requirement of immediate ash
disposal has initiated several scientific as well as
industrial researches where in techniques of making
pond ash a useful by-product are being developed.
The problem of disposing pond ash can be reduced
by utilizing it in large quantities in various
engineering works. Till date pond ash has been
effectively used in construction sites, brick making
etc.As per the data collected from Panipat Thermal
Power Plant, the following can be stated:
• For a 110MW electricity production, one
production unit emits 18-20 tonnes of ash per hour.
The ash is very hazardous as discussed above, and
contains high percentage of unburnt carbon. The
details are as follows:
TABLE 3 % of unburnt Carbon
Unit No. % of unburnt carbon in
Pondash Bottom ash
1 6 10
Varma et al. 351
3 3.6 9.75
4 2.4 8.2
5 1.5 5.3
6 1.2 4.29
7 1.04 4.6
• Combustion of coal generates enormou
quantities of ash that are disposed-off i
vast areas of land adjoining the powe
station. Geochemical and radiometri
analysis of the pond ash depicts th
quantity of toxic elements that are likely
to contaminate the soil and the
groundwater system. Trace element
analysis depicts that toxic elements (Pb,
Cu, Ni, Fe, As) are sufficiently enriched
in pond ash, and preferably in the lighter
size fractions. Chemical analysis of the
water samples near the ash ponds indicate
that there has been high concentration of
trace elements (Al, Li,Ni, Fe, As, Zn, B,
Ag, Sb, Co, Si, Mo, Ba, Rb, Se, Pb V, Cr,
Cu, Cd, Mn, Sr), whose distribution,to
large extent, is directed by the ash
deposited in the adjoining areas.The
enrichment of some elements (Al, Fe, As
and Mn) above WHO guidelines for
drinking water denotes significant
contamination of the groundwater from
the toxic elements leached from the ash
pond.
• Groundwater is prone to metal
contamination due to waste disposal
and leachate percolation. Coal-fired
power stations have remained under
scrutiny for quite a while,as potential
sources of mercury and other heavy
metal pollutants. When huge
quantities of ash accumulate for long
time periods in the disposal site,
hazardous substances tend to
percolate through the soil layers and
eventually reach the groundwater.
352 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE 4 W.H.O guidelines for drinking water Chandrapur (M.P) 58.2 89.2 301.2
Farakka( West Bengal) 84.1 98.8 297.1
ELEMENT PERMISSIBLE Raichur (Karnataka) 83.1 102.5 334.1
LIMIT (in ppm)
Al 0.2 Talchir(Orrisa) 79.2 96.3 291.6
As 0.01 Bokaro (Bihar) 102.9 154.7 233.0
Ba 0.7
Cr 0.05 Ramagundam(Andhra 92.5 120.3 255
Pradesh)
Cu 2 Neyvelli (T.N.) 64 126.9 370
Fe 0.3
Mn 0.1 Amarkantak(M.P.) 18.5 22.2 37
Ni 0.02 Bandel(West Bengal) 18.5 18.5 37
Indraprastha(Delhi) 22.2 63 255.0
Durgapur(West Bengal) 18.5 29.6 66.7
The EPA (Environment Protection Agency) estimates Korba(M.P) 22.2 44.3 137.1
that about 300 ponds for coal ash exist nationwide.
The power industry estimates that the ponds contain Nasik(Maharashtra) 44.5 37 96
tens of thousands of pounds of toxic heavy metals.

Over the years, the US government has found Without federal guidelines, regulations of the ash
increasing evidence that coal ash ponds and landfills ponds vary by state. Most lack liners and have no
taint the environment and pose risks to humans and monitors to ensure that ash and its contents don't
wildlife. In 2000, when the EPA first floated the idea seep into underground aquifers. Industries have
of a national standard, the agency came across 11 previously experimented with clay being used as a
cases of water pollution linked to ash ponds or landfill liner. Several layers of clay were laid in and
landfills. In 2007, that list grew to 24 cases in 13 around the landfills to obstruct any kind of leakage in
states with another 43 cases where coal ash was the the underlying ground water.
likely cause of pollution. However, the attempt proved to be a failure, as
several gaps existed in this kind of liner. No matter
The leaks and spills are blamed for abnormalities in how many layers of clay were laid, it did not serve
tadpoles. The heads and fins of certain fish species the purpose completely. Clay being a porous material
were deformed after exposure to the chemicals. In could not, at all, cater to the need of a potential
2006, the EPA concluded that disposal of coal waste landfill liner.
in ponds elevates cancer risk when metals leach into
Effect of addition of soil on pond ash-soil mixture
drinking water sources.
TABLE6 Effect of addition of soil on pond ash-soil
TABLE 5: Radioactivity in pond ash from different mixture
thermal power stations in India MIX SP. OMC PERMEABLTY
DESIGNA- GRV (cm/s)
Thermal power station Activity TION TY
concentration(Bq/kg) 100%pond 2.14 27.5% 0.00012
228 40
Ac K ash + 0%soil
226
Ra 90%pond ash 2.193 25.4% 0.000623
Allahabad 78.4 89.1 362.7 + 10%soil

Angul(Orrisa) 78.5 86.5 278.1 80%pond ash 2.246 24.8% 0.0000467

Badarpur(Delhi) 75.5 88.1 286.4 + 20%soil

Bakreshwar (West 76.3 87.5 288.1

Bengal)
 Varma et al. 353

Effect of addition of hardening agent on pond ash- REFERENCES:

soil mixture

TABLE7 Effect of addition of Hardening Agent on 1. .Appolonia D J D. (1980).Soil-Bentonite

Pond ash-soil Mixture Slurry Trench Cut-offs.. Journal of
Geotechnical Engineering, ASCE, vol 106,
MIX SP. OMC PERMEABLTY no GT4, pp 399-417.
DESIGNATN GRVTY (cm/s)
2. Dutta, Juneja & Vats D K.
90%pond ash + 2.188 23.6% 0 .0000315 (1997),.Permeability Study of Amended
5%soil + 5%
SoilLiners for Landfills.. Indian
CAS
Geotechnical Journal, vol 27, no 2, pp 162-
80%pond ash + 2.238 22.6% 0.000006 174
15%soil +
5%CAS 3. Kaniraj S K and Havangi V J. (1999),
Geotechnical Characteristics of Flyash-soil
Mixture.. Geotechnical Engineering Journal,
vol 30, no 2, pp 129-147.
The hardening agent used in this study is carbonic
aluminate salt (CAS) with its various constituents as
4. Kaniraj S R and Havangi V J.(
cement (80%), Al2(SO4 )3 (9%), Na2CO3 (6%) and
1999).Geotechnical Characteristics of
CaO (5%)
Flyash-soil Mixture.. Geotechnical
Engineering Journal, vol 30, no 2, pp 129-
With the increase in soil content at 5% CAS addition,
147.
the corresponding permeability decreases;
and that permeability of the order of 10-7 is obtained
which makes it suitable for use as liner in land fills. 5. Kumar Virender Dr. et al. (February 2002.)
Thus CAS has remarkable effect on the permeability Journal of The Institution of Engineers
of pond ash-soil matrix. (India), vol 82, pt CV, no 4,.

CONCLUSION 6. Mandal and Sengupta.Radionuclide and

Evidently, it is possible to reduce the permeability of
pond ash with the addition of locally available fine
size soil by compacting the mixed material at their
optimum content to their maximum dry density,
though in doing so, larger quantity of soil is required.
For its suitability as a liner to landfills, this
combination requires the addition of a hardening
agent (CAS), which further reduces the permeability
and the quantity of soil to be used in the pond ash-
soil matrix. However, from these studies it is difficult
to infer whether permeability of these materials
would not increase due to prolonged accumulation of
leachate generated from the landfills in which these
are to be used as liners.
trace element contamination around
Kolaghat Thermal Power Station, West
Bengal –Environmental implications
Department of Geology and Geophysics,
Indian Institute of Technology, Kharagpur
721302, India WHO table
7. Mishra D P Dr. Consolidation
Characteristics of Stowed Pond Ash and
Pond Ash-lime Mixture
8. Sawa Kohei, Masashi Kamon, Seishi
Tomokisa and Nagahide Naito. (1998),
.Waste Oil Hardening Treatment bt
Industrial Waste Material.. International
Conference on Fly Ash Disposal and
Utilization, CBIP, New Delhi, pp 361-367
 Laboratory Investigation of Moisture Content Redistribution on the Base
of Double Composite Liner Systems

Farzad Meysami Azad

School of Civil Engineering, University of Sydney, NSW 2006, Australia
David W Airey
School of Civil Engineering, University of Sydney, NSW 2006, Australia
R Kerry Rowe
GeoEngineering Centre at Queen’s-RMC, Queen’s University, Kingston, Ontario, Canada
Abbas El-Zein
School of Civil Engineering, University of Sydney, NSW 2006, Australia

ABSTRACT Geomembranes (GMs) and geosynthetic clay liners (GCLs) are used as part of double
composite liner systems (DCLSs) for hazardous waste and municipal solid waste landfills. This paper
presents the results of a study of the spatial and temporal redistribution of moisture at the base of DCLSs
under isothermal conditions. This investigation gives a better understanding of the time taken for the
moisture content to reach equilibrium. The effect of GCL’s carrier on final GCL water content is also
investigated. The study of water distribution will be extended with more tests incorporating temperature
gradients and overburden pressure.

INTRODUCTION CCL or a GCL serve to minimize that leakage. In

Modern landfill barrier systems include a composite
liner intended to reduce the outward transport of
contaminants to a negligible level. The composite
liner is comprised of a high density polyethylene
(HDPE) geomembrane (GM) and either a
geosynthetic clay liner (GCL) or a compacted clay
liner (CCL) and for a well designed and constructed
liner, the composite action of the HDPE GM and clay
minimizes the leakage of contaminant to
groundwater (Rowe 2005). Given the long
contaminating lifespan of modern landfills, the long-
term performance of lining systems is extremely
important (Rowe et al. 200).
The use of geosynthetic materials in composite liners
has increased because of their relatively low cost
(Malusis & Shakelford 2002) and generally good
performance for the period of time that they have
been monitored (Rowe 2005, Mitchell et al. 2007).
Although HDPE GMs represent the primary barrier
to leachate flow, leakage through perforated wrinkles
and other defects in the HDPE GM can occur and a
particular, composite liners involving a GCL have
been shown to be highly effective in preventing
groundwater contamination when the GCL is
adequately hydrated (Rowe 2005). However for the
composite action to be most effective, it is important
that the clay liner below the HDPE GM be intact and
not fractured for the contaminating lifespan of the
landfill (Rowe 2005, Southen 2005).
Thermal gradients induced by temperature generated
by the waste (Collins 13) can cause moisture
movement at the base of the landfill with a
consequently decrease the water content in clay
liners. This complex mechanism has caused
desiccation cracking in clay liners reported at various
sites and in laboratory experiments (Bowders et al.
17, Sangam & Rowe 2002, Rowe 2005, Southen &
Rowe 200). In particular, Southen & Rowe (200,
2005) conducted small and large scale laboratory
tests to examine the effect of thermal gradients on
GCLs used in single composite liners.
 Azad et al. 355

The objective of this paper is to investigate, the
spatial and temporal redistribution of moisture in a
GCL used in a DCLSs under isothermal condition.
The time taken for the moisture content to reach
equilibrium and the effect of GCL’s carrier are also
investigated. This study is conducted as a precursor
of a future examination of the effect of temperature
gradient on the moisture in a GCL with time.
TESTING PROGRAM
particle size distribution of the soil is given in Figure
2 with 12% passing the 0.075 mm sieve. A standard
proctor compaction test performed on the soil
(ASTM D 68) and gave a maximum dry density of
1.7g/cm 3 and an optimum water content of 11.2%
Figure 3. A summary of the main properties of the
other materials studied is given in table 2.
100.00
90.00
80.00

Percent Finer
Apparatus 70.00

Five polyvinylchloride (PVC) cells were constructed 60.00

with an internal diameter of 160mm, 325mm high 50.00

and a wall thickness of mm (Figure 1). The bottom 40.00

of the cell was filled with 250mm of compacted 30.00

20.00
silty-sand soil representative of subsoil and a double
10.00
composite liner system was constructed over this
0.00
subsoil.
0.001 0.01 0.1 1 10 100
Grain Size (mm)
Fig.2 Grain size distribution of the subsoil
1.735

1.73
Dry density (g/cm3)

1.725

1.72

1.715

1.71

1.705
0 2 4 6 8 10 12 14 16

Fig.1 Schematic of test cell Moisture content (%)

Fig.3 Compaction curve for subsoil

Starting at top of the subsoil, the DCLSs consisted of
GCL overlain by a 1.5mm thick secondary HDPE
GM, a 5.3mm geonet, and geotextile (forming the
leak detection/ secondary leachate collection
system), a primary GCL, and a 1.5mm thick primary
HDPE GM. The cell was sealed using a 10mm PVC
plate and adhesive-sealant on the top and bottom
boundaries.
Material properties
Tests were conducted using a silty-sand subsoil with
nominal properties as indicated in table 1. The
Test description
The soil samples were mixed with water to bring
its water content to 11% and then stored in plastic
bags to allow the moisture in the soil to reach
equilibrium. The soil was then tamped into the test
cell in four layers. Water content samples were
taken from each layer. The initial moisture
contents profile of the soil in five cells prior to
GCL installation are given in Figure .
356 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE 1 Subsoil properties

Optimum Maximum Grain size distribution
water content dry density Silt content Sand content Gravel content
11.2% 1.74(g/cm2) 12% 82% 6%

TABLE 2 Material properties based on manufacturers published data

Reference
Geosynthetic Clay Liners (GCLs) www.geofabrics.com.au
Nominal total mass/unit area (g/m2) 4390
Bentonite mass/unit area (g/m2) 4000
Nonwoven cover geotextile (g/m2) 270
Woven carrier geotextile (g/m2) 110
Geotextile (GT) www.globalsynthetics.com.au
Mass (g/m2) 280
Geomembranes (HDPE GMs) www.gseworld.com
Thickness (mm) 1.50
Geonet www.globalsynthetics.com.au
Thickness (mm) 5.3

25 The test cells were opened every five days and the
GCL was removed for moisture content
20
T1-GT1 determination and then returned. Since there was
T2-GT1 no source of water for primary system this
Depth (cm)

15 T3-GT1
T4-GT1 represent a severe scenario where moisture can
10
T5-GT2
evaporate from the GCL into the airspace and then
be lost when the system is opened (the leak
5
detection system is “breathing” with moisture loss
0
from the layer with exchange of very humid air in
8.5 9 9.5 10 10.5 11 11.5 12 12.5 the layer with less humid air from the
Water content (%) atmosphere). The cells were excavated at the end
Fig.4 Initial moisture content profile for soil of each test to assess the temporal variation of
samples water content within the subsoil as soil suction
equilibrated.
The GCL samples were hydrated by gradually
applying water until the initial water content RESULTS
(75%) was reached. The secondary GCL was then Based on theoretical considerations, isothermal
placed over the foundation soil, followed by conditions don’t lead to desiccation (Southen &
1.5mm HDPE GM. A 5.3 mm layer of Geonet was Rowe 2005) but the GCL gravimetric water
also placed over the secondary HDPE GM to content just prior to the placement of waste is
simulate the secondary leachate collection system significant.
followed by GT, primary GCL and primary HDPE
GM. A steel block was placed over these layers to The first four tests were nominally identical except
produce an overburden stress of 1.7kPa, to ensure for the test duration, as indicated in table 3. In
intimate contact between the layers. these tests the woven carrier GT was in contact
with the GT and subsoil for the primary and
 Azad et al. 357

secondary GCL respectively. In the fifth test, the
GCL was reversed (so that the nonwoven
geotextile was in contact with the GT and subsoil)
to examine the effect of GCL’s carrier on moisture
uptake (T5-GT2). The choice of initial water
content of GCLs was based on what would be
expected from uptake of water into a GCL laid on
moist subsoil (Southen & Rowe 2005).
the secondary GCL water content increased by
50%. This confirmed that observed by Southern
and Rowe (2005) and Rayhani et al. (2008). The
GCLs water contents are presented in Figure 8.
200
T4-GT1 (Primary GCL)

Gravimetric water content (%)

180
T4-GT1 (Secondary GCL)
160
Water contents 140
GCLs are highly effective in preventing 120

groundwater contamination when they are 100

adequately hydrated (Rowe 2005). Variation of 80

60
GCLs moisture content was plotted with time in
40
Figure 7. After a period of 170 days,
20
approximately 32% decrease in primary GCL 0
water content was observed due evaporation of 0 50 100 150 200
water into the air space until there was an Time (days)

equilibrium between the relative humidity of the Fig.7 Gravimetric water contents for T-GT1
air and the suctions in the GCL. There was then
loss of this moisture in the air when the samples 170
160
were weighed followed by more moisture transfer 150
Initial
Final (Primary GCL)
from the GCL when the cell was resealed. In one 140
130 Final (Secondary GCL)
sense this is an artifact of the test and the moisture
Water Content (%)

120
110
loss would likely have been much less if there had 100
been less opening of the cell to monitor moisture 90
80
content. However this could occur in a real 70
60
landfill situation if there was significant airflow 50
40
through the leak detection system (e.g. to remove 30
gases or to cool the liner). 20
10
0
T1-GT1 T2-GT1 T3-GT1 T4-GT1
Due to uptake of water from the subsoil, there was
an approximately 61% increase in secondary GCL Fig.8 GCL gravimetric water contents
water content Figure 7. Within the first 75 days,

TABLE 3 Test conditions

Test T1-GT1 T2-GT1 T3-GT1 T-GT1 T5-GT2
Test duration (days) 20 40 60 170 20
Initial subsoil water content (%) 10.9 10.7 10.8 10.48 10.9
Final subsoil water content (%) 9.6-11.2 8.3-11.4 8.1-11.8 7.85-10.74 9.3-11.1
Initial primary GCL water content (%) 75 75 75 75 75
Final primary GCL water content (%) 66.1 59.6 55.9 42.4 63.4
Initial secondary GCL water content (%) 75 75 75 75 75
Final secondary GCL water content (%) 99.0 110.8 122.3 136 118.8
Applied stress (kPa) 1.7 1.7 1.7 1.7 1.7
358 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Effect of carrier on final GCL water content The primary GCL in T5-GT2 had lower moisture
The gravimetric water content distribution content than that in T1-GT1. Gravimetric moisture
measured during excavation of two tests where the content versus time is plotted in Figure 11. GCL’s
GCL had the woven down (T1-GT1) and the carrier appears to have some effect on rate of
nonwoven down (T5-GT2) respectively is shown moisture redistribution in both GCLs.
in Figure . After 20 days, the gravimetric water 125

Gravimetric water content (%)

content for T1-GT1 was found to range from .6%
below the GCL to 11.2% at the lower boundary. 100

For T5-GT2, the water content of the subsoil was a

75
little lower ranging between .3% below the GCL
and 11.1% at the lower boundary, respectively. 50
Primary GCL (T1-GT1 woven down)
Thus final subsoil water content below the GCL Primary GCL (T5-GT2 nonwoven down
was lower than the initial value, primarily due to 25 Secondary GCL (T1-GT1 woven down)
uptake of water by the secondary GCL Figure 10. Secondary GCL (T5-GT2 nonwoven down)
0
Over a period of 20 days, the secondary GCL 0 5 10 15 20 25
water content increased from 75 to % when the Time (days)
woven GT was in contact with the subsoil (T1-
Fig.11 The effect of GCL carrier on rate of moisture
GT1) which is smaller than the increase from 75 to redistribution
11% when the nonwoven was in contact with the
soil (T5-GT2).
CONCLUSION
30 The spatial and temporal redistribution of moisture
25 at the base of DCLSs under isothermal condition
was studied.
Height (cm)

20
A reduction in primary GCL water content was
15
Initial T1-GT1 observed due to evaporation of water into the air in
10 Final T1-GT1 the leak detection system and subsequent loss of
Initial T5-GT2 this moisture from the leak detection layer when
5 Final T5-GT2
high humidity air was replaced by low humidity
0
air. This reduction was marginally greater when
8 9 10 11 12
Water content (% )
the nonwoven GT was in contact with the subsoil.
Fig.9 Subsoil gravimetric water contents assessed at Observations of the secondary GCL showed
the termination of T1-GT1 and T5-GT2
uptake of water from the subsoil but even though
there was an increase to about 136%, the GCL and
160
initial soil had not reached equilibrium after 170 days
140 Final (primary GCL)
Final (secondary GCL) despite an initial water content of about 11% in the
120
water contents (%)

soil and 75% in the GCL. Thus at low stress, it

100
may take a significant amount of time for the GCL
80
to fully hydrate. The tests indicate that, other
60
things being equal, moisture uptake was likely to
40
be faster with the nonwoven geotextile component
20
of the GCL in contact with the subsoil than with a
0
T1-GT1 T5-GT2
woven carrier resting on the subsoil.
Fig.10 GCLs gravimetric water contents

It is important to emphasize that these tests were
simply to establish a sense of initial conditions
before a thermal gradient is established. In order to
fully investigate the behavior of GCL in a basal
liner system more testing with a temperature
gradient and overburden pressure are needed.
REFERENCES
Bowders, J. J., Daniel, D. E., Wellington, J. and
Houssidas, V. (1997) Managing desiccation
cracking in compacted clay liners beneath
geomembranes, Proc. Geosynthetics ’97, Long
Beach, 527-540.
Collins, H. J. (1993) Impact of the temperature
inside the landfill on the behaviour of barrier
systems, Proc. 4th International Landfill
Symposium, S. Margherita di Pula, Cagliari,
Italy, 1: 17-32.
Malusis, M.A., Shackelford, C.D. (2002).
Coupling effects during steady-state solute
diffusion through a semipermeable clay
membrane. Environmental Science & Western Ontario, London, Ontario, Canada.
Technology, ACS 36 (6), 1312–1319.
Mitchell, J.K., Alvarez-Cohen, L., Atekwana, E.,
Burns, S.E., Gilbert, R.B., Kavazanjian, E.,
O’Riordan, W.H., Rowe, R.K., Shackelford,
C.D., Sharma, H.D. and Yesiller, N. (2007).
Assessment of the performance of engineered
waste containment barriers, Report of the
Committee to Assess the Performance of
Engineered Barriers to National Research
Council of the U.S. National Academies, ISBN-
13: 978-0-309-10809-6 (121p.).
Azad et al. 359
Rayhani M.H.T., Rowe R.K., Brachman R.W.I.,
Siemens G. and Take W.A. (2008), Closed-
system investigation of GCL hydration from
subsoil, GeoEdmonton 2008, 61st Canadian
Geotechnical Conference & 9th Joint CGS/IAH-
CNC Groundwater Conference.
Rowe, R.K., Quigley, R.M., Brachman, R.W.I.,
Booker, J.R. (200) Barrier Systems for Waste
Disposal Facilities, E & FN Spon, Taylor &
Francis Books Ltd, London, 579p.
Rowe, R.K. (2005). Long-Term Performance of
Contaminant Barrier Systems, 45th Rankine
Lecture, Geotechnique, 55 (9), 631-678.
Sangam, H.P. and Rowe, R.K. (2002) Effects of
exposure conditions on the depletion of
antioxidants from HDPE geomembranes,
Canadian Geotechnical Journal, 39 (6):1221-
1230.
Southen, J. M. (2005), Thermally Driven Moisture
Movement Within and Beneath Geosynthetic
Clay Liners, PhD thesis, The University of
Southen, J. M. and Rowe, R. K. (2004).
Investigation of the behavior of geosynthetic
clay liners subjected to thermal gradients in
basal liner applications, Journal of ASTM
International, 1(2) ID JAI11470.
Southen J. M. and Rowe R.K. (2005), Laboratory
investigation of geosynthetic clay liner
desiccation in a composite liner subjected to
thermal gradients, Journal of Geotechnical and
Geoenvironmental Engineering, 131(7), 925-
35.
 Hydraulic Conductivity of a Tropical Soil
(Non-Lateritic) and Bentonite Mixtures for Base Linear
Purposes
Thiago Luiz Coelho MORANDINI
Núcleo de Geotecnia, Universidade Federal de Ouro Preto, Brazil (thiagomorandini@yahoo.com.br)
Adilson do Lago LEITE
Departamento de Engenharia Civil, Universidade Federal de Ouro Preto, Brazil (alleite@em.ufop.br)

ABSTRACT: Tropical soils cover vast regions of the globe and have great potential to be use as a compacted
clay liner (CCL) in waste disposal facilities. When required, bentonite clay can be added to these soils so as
to improve their hydraulic properties (expected reduction in the permeability). This paper focuses the
characterization and the determination of the hydraulic conductivity (tests in flexible wall permeameter) of
mixtures of a non lateritic tropical soil and bentonite in proportions of 0, 3, 6, 9 and 12%. The results show a
significant decrease in the hydraulic conductivity of the samples with the addition of bentonite, as well as
increasing the confining pressures. KEYWORDS: tropical soils, bentonite clay, hydraulic conductivity,
compacter clay liner

INTRODUCTION
The concept of a sustainable environment is
prominent nowadays. A sustainable environment
in solid waste disposal facilities means a correct
management and storage of waste, as well as
monitoring of the leachate and gas generated in
the site. Liner systems are essential components
of these facilities, playing two different roles: 1)
covers, which restrict the infiltration from
precipitation and 2) base liners, that reduce or
mitigate leachate flow towards soil and
groundwater.
Tropical soils, in turn, have great potential to be
used as compacted clay liners (CCL), due their
wide distribution over the globe and some
favourable geotechnical properties when
compacted, such as high unit weight and shear
strength. However, sometimes CCL’s composed
of tropical soils do not meet hydraulic
conductivity criteria for waste containment
barriers (e.g. 10-7cms-1) because of their peculiar
mineralogy, usually composed of quartz, caulinite
and aluminium-iron oxides/hydroxides.
Additionally, their low activity decreases their
contaminant sorption capacity.
A possible alternative to reduce the hydraulic
conductivity of compacted tropical soils for waste
containment purposes is the addition of bentonite.
Some papers have showed the improvement
provided by the addition of bentonite to sandy
soils, including indication of bentonite quantities
to be added. Daniel (1987) and Keeney et al.
(1992) demonstrate that the best bentonite content
would be up to 12% (dry basis).
Other authors investigate different properties of
the bentonite addition to lateritic soils, including
Anderson & Hee (1995), Osinubi & Nwaiwu
(2002), Farnezi & Leite (2007) and Batista &
Leite (2009). In spite of their great contribution to
the understanding of the behaviour of these
mixtures, none of them deals exclusively with
tropical soils of non-lateritic behaviour.
Nogami & Villibor (1995), through MCT
classification, identified two main classes from
the tropical soils. Lateritic soils, formed in well-
drained, generally correspond to more superficial
layers between 2 m and 10 m of thickness. They
are characterized by fine-grain aggregation, quite
resistant to water action, high permeability and
 Morandini and Leite 361

high mechanical strength. And non-lateritic soils
(saprolitic soils), usually lie in the underlying
layers to lateritic soils. Its thickness is varied
from a few centimeters to tens meters,
characterized by remnants of the matrix rock, as
schistosity, layers, spots and voids. In contrast to
the lateritic soils, which can be transported or
residual soils, non-lateritic soils are genuinely
residual soils, characterized by geotechnical
properties (mechanical and hydraulic) extremely
varied (unpredictable by simple identification).
This paper focuses on the determination of index
properties and the hydraulic conductivity of
mixtures of different proportions of bentonite and
a non-lateritic tropical soil sample. Laboratory
tests were conducted to evaluate the hydraulic
conductivity of mixtures of 0%, 3%, 6%, 9% and
12% of bentonite (dry weight basis), using a
flexible wall test apparatus (triaxial cell). A
constant head of 50 kPa were applied for
confining pressures of 20 kPa, 40 kPa and 80 kPa.
MATERIALS AND METHODS
Sample Properties
sample came from Boa Vista city, Paraiba State,
north-eastern Brazil.
The tropical soil sample, referred here as SN
sample, came from an outcrop by the road that
connects the cities of Mariana to Ponte Nova,
Minas Gerais State, south-eastern Brazil. In place,
it was a typical dark red and thick residual soil
profile, originated from gneiss rocks.
In the laboratory, both samples were prepared
according to the Brazilian Standard ABNT-NBR
6457, that recommends harrowing,
homogenisation, sieving and stockpiling in
appropriate containers. The proportions of soil
and bentonite of the mixtures were based on the
literature reviews and previous work of one of the
authors (e.g. Day & Daniel, 1985; Anderson &
Hee, 1995; Shackelford, 1994 and 2000; Farnezi
& Leite, 2007 and Batista & Leite, 2009), as
follows (dry weight):
- SN: 100% natural soil;
- SN03: 97% natural soil and 3% of bentonite;
- SN06: 94% natural soil and 6% of bentonite;
- SN09: 91% natural soil and 9% of bentonite;
- SN12: 88% natural soil, 12% of bentonite.

A commercial sodium bentonite Brasgel sample The tests (references) used to determine sample
was used to compose the mixtures. According to properties are depicted in Table 1.
the fabricant (Bentonit União Nordeste S/A), this

TABLE 1. Methods for determination of the sample properties.

Test
MCT classification
Grain size analysis
Liquid limit determination
Plastic limit determination
Activity determination
Specific gravity of solids
determination
Compaction
Blue metilene adsorption test
Properties
Lateritic character
Grains size
Liquid limit (ωL)
Plastic limit (ωP)
Activity index (A)
Specific gravity of solids (Gs)
Optimum moisture content (wot)
and Maximum dry unit weight
(ρdmax)
Cation Exchange Capacity (CEC)
and Specific Surface (SS)
Reference
Nogami & Villibor (1995)
ABNT-NBR7181 (1984)
ABNT-NBR6459 (1984)
ABNT-NBR7180 (1984)
Skempton (1953), cited in Mitchell
(1993)
ABNT-NBR6508 (1984)
ABNT-NBR7182 (1986)
Pejon (1992)
362 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Hydraulic Conductivity Tests
This property was determined using a flexible
wall (triaxial cell) method, as shown in Figure 1.
As any triaxial cell, the apparatus applies a
pressure at the base of the sample (p1) and another
one at the top (p2), as well as a confining pressure
Fig. 1. Schematic of the flexible wall permeameter apparatus.
The procedure used in the tests was based on the
recommendations of Head (1998), as follows:
(1) Initially the soil samples were compacted
(Proctor Normal) at moisture contents (distilled
water) of 2% wet of optimum, to be after trimmed
to dimensions of 5 cm in diameter and 10 cm
long.
(2) In saturation, a maximum backpressure of 300
kPa was applied to saturate the soil samples. The
B valve was maintained closed during this stage.
Small pressure increments were induced at p1 and
σ3 (σ3 > p1) until the equalization in p2.
system (σ3). An electronic volume meter and a
transducer for pore pressure measurement (u)
were also connected to p1.
The following condition was applied to samples
to avoid soil failure induced by the increments of
pore pressure: σ3 > p1 > p 2.
(3) By closing the valve A' and applying σ3, it was
possible to estimate the B parameter. The sample
was considered saturated when B values were
greater than 94%.
(4) After saturation, the soil samples were
consolidated under the respective confining stress
to ensure that the measured water volumes of
percolation were a consequence of flow, not of
consolidation (reduction in the void ratio). The
effective confining pressures were 20, 40 and 80
kPa. They were based on possible loads on
landfills.
 Morandini and Leite 363

hydraulic head (h) equivalent to 509,7 cm of

(5) The sample height L (cm) after consolidation water.
could be estimated using equation (1), where ∆V
(cm³) is the variation of the sample volume and L0 (8) The flow rate, Q (cm³/s), was estimated by
and V0 are the initial length and initial volume of applying equation (6) to the linear portion of the
the sample, respectively. diagram that relates the flow volume, Vf (cm3),
with time, t (s).
 1 ∆V 
L = L0 1 − 
 (1) ∆V f
 3 V0  Q= (6)
∆t
The cross section area of the sample after
The hydraulic gradient i (cm/cm) is given by
consolidation, A (cm2), was estimated using
equation (7).
equation (2). Equation (3) was used to calculate
the void ratio during the tests.
h
i= (7)
V − ∆V  2 ∆V  L
A= 0 = A0 1 − 
 (2)
L  3 V0  Finally, the hydraulic conductivity K (cm/s) is
estimated by the Darcy Law (equation 8).
 ∆V 
e = e0 − (1 + e0 ) 
 (3) Q
 V0  K= (8)
Ai
(6) The pore pressure was estimated prior to
percolation so as to establish the same effective RESULTS AND DISCUSSIONS
confining pressure of the consolidation phase.
Head (1998) states that the application of two Sample Properties
different pressures at the base and top of the
sample (respectively, p1 and p2) induces a non Sample properties are presented in Table 2. It can
linear pore pressure over the sample length. An be seen that the addition of bentonite has not
average pore pressure (u) can then be obtained increased the clay content of the tropical soil
using equation (4). sample (SN), 42%, as expected. On the other
hand, the influence of bentonite addition was very
u=
1
(2 p1 + p 2 ) (4) significant on the Atterberg Limits and physic-
3 chemical parameters, demonstrated by the
substantial increase in the plastic index (PI),
Equation (5) was used to calculate the effective cation exchange capacity (CEC) and specific
confining pressure inside the sample (σ'3). surface (SS).

σ 3 = σ 3' + u = σ 3' +
1
(2 p1 + p 2 ) (5)
Figure 2 illustrates the Atterberg Limits increase
as a reflex of the bentonite addition. It worth to
3
point out the quasi linear behaviour of PI (R2 =
(7) The hydraulic head inside the sample is 0,9827). As no variation of the plastic limit ( ωP)
imposed by the difference of the base and the top was noticed, the increase in PI is due exclusively
pressure (p1 - p2). These pressures were by the increment in the liquid limit ( ωL).
maintained constant and equal to p1 of 300 kPa
and p2 of 250 kPa, which gives a constant
364 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE 2. Sample properties.

PROPERTY / SAMPLE SN SN03 SN06 SN09 SN12 BK
Bentonite (%) 0 3 6 9 12 100
Grain-size analysis 1
Clay (%) 42 42 42 42 42 91
Silt (%) 6 3 6 6 6 9
Fine Sand (%) 6 13 10 8 10 0
Medium Sand (%) 36 30 32 32 32 0
Coarse Sand (%) 8 10 8 10 8 0
Gravel (%) 2 2 2 2 2 0
Atterberg Limits
Liquid Limit ωL (%) 51,9 77,7 96,6 117,0 156,0 682,5
Plastic Limit ωP (%) 29,6 27,6 28,7 26,8 27,8 90,6
Plasticity Index PI (%) 22,3 50,1 67,9 90,2 128,2 591,9
Activity (Skempton,1952) 0,53 1,19 1,62 2,15 3,05 6,50
Specific Gavity
Gs (g/cm³) 2,840 2,828 2,817 2,803 2,791 2,452
Compaction
Maximum dry unit weight ρdmax (g/cm³) 1,680 1,655 1,620 1,602 1,588 NR 2
Optimum moisture content wot (%) 21,5 22,5 23,3 24 24,5 NR 2
Physico-chemical
Cation Exchange Capacity CEC (cmol/100g) 6,9 13,5 18,6 25,3 36,0 99,4
Specific Surface SS (m2/g) 53,8 105,3 145,6 197,3 281,3 775,9
1
Grain-size scale of the Massachusetts Institute of Technology.
2
No results.

TABLE 3. Hydraulic conductivity, void ratio and total

porosity results.
PROPERTY /
SN SN03 SN06 SN09 SN12
SAMPLE
Bentonite (%) 0 3 6 9 12
Hydraulic conductivity, K x 10-7 (cm/s)
20 17,4 1,85 0,389 0,169 0,113
σ3 40 11,5 1,24 0,213 0,086 0,043
(kPa)
80 8,07 0,705 0,110 0,031 0,012
Void Ratio, e
Fig. 2. Atterberg limits variation with bentonite 20 0,686 0,738 0,743 0,756 0,754
σ3 40 0,681 0,735 0,738 0,749 0,748
addition. (kPa)
80 0,673 0,728 0,731 0,742 0,743
Hydraulic Conductivity Porosity, n (%)
20 40,7 42,5 42,6 43,1 43,0
Table 3 presents the results of K determination for σ3 40 40,5 42,4 42,5 42,8 42,8
(kPa)
all the confining pressures, including also the 80 40,2 42,1 42,2 42,6 42,6
void ratio (e) and total porosity (n). Figure 3, in
turn, depicts the K variation for the different It is clear from Figure 3 the influence of the
bentonite contents. bentonite clay over the hydraulic conductivity of

the samples, decreasing K from the order of 10-5
to 10-6 to the order of 10-8 to 10-9 cm/s.
Fig. 3. Hydraulic conductivity versus bentonite content.
As K = 10-7 cm/s seems to fulfil the requirements
for most regulations and landfill projects in the
world, Figure 3 demonstrates that this value was
achieved around 3,5% of bentonite content for all
the confining pressures. Additionally, Figure 3
shows that the decrease in K is non linear over the
proportions of bentonite used in the tests.
When the confining pressure is plotted against the
hydraulic conductivity (Figure 4), some
observations can be highlighted:
• An almost linear relation was observed for all
samples;
• The influence of the confining pressure is
more expressive on the samples with higher
proportions of bentonite: 6, 9 and 12% (e.g.
SN06, SN09 and SN12 samples). This is due to
the higher plasticity of these samples.
Fig. 4. Hydraulic conductivity of the samples under
different confining pressures.
Morandini and Leite 365
Figure 5 presents the hydraulic conductivity
variation with the porosity after consolidation. It
shows that the porosity increases with addition of
bentonite, obtaining, as expected, the inverse to
K. This tendency highlights the proposal of
Chapuis (1990), where the correlations between
hydraulic conductivity and porosity do not show
expected results.
Fig. 5. Hydraulic conductivity versus sample void ratio
under different confining pressures.
As shown in Figure 5, the process of
consolidation in hydraulic conductivity
determination decreases the porosity of the
samples. On the other hand, the addition of
bentonite increases this porosity, resulting in
unknown change in the structure of samples.
Therefore, a more detailed study is necessary to
establish a model of hydraulic conductivity and
void ratio of soils.
CONCLUSIONS
Some important conclusions can be related to this
study, as follows:
(1) The bentonite considerably increased the
plasticity, specific surface and cation exchange
capacity of the tropical soil sample. The increase
in the last two parameters represents an
improvement in terms of enhancing the sorption
capacity for contaminant retention;
366 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
(2) The hydraulic conductivity of the tropical soil
experimented significant reduction with the
increase of bentonite under all the confining
pressures used in the tests. To achieve K values
below 10-7 cm/s, a content of 3,5 or 4% of
bentonite is recommended;
(3) The higher plasticity of the SN06, SN09 and
SN12 increased the dependency of the hydraulic
conductivity on the confining pressure, which has
led to a greater reduction in the permeability of
the samples;
(4) The increase in the bentonite content has led
to a small increase in the void ratio and porosity
of the samples. Therefore, this trend is not so
conclusive.
Some suggestions can be offered to future
research on this matter:
- Other tropical soil samples should be evaluated;
- Assessment of the influence of the lateritic
degree on the hydraulic conductivity;
- Study the change of void ratio of tropical soils
with the addition of bentonite;
- Evaluation of shear strength with the addition of
bentonite in tropical soils.
ACKNOWLEDMENTS
The authors would like to thank the Coordenação
de Aperfeiçoamento de Pessoal de Nível Superior
(CAPES) for the grant provided to the first author.
Special acknowledments is also given to the
Bentonit União Nordeste S/A, represented by the
engineer Pedro Paulo Furtado Gouveia, who
kindly donated the bentonite samples used in this
research.
BIBLIOGRAPHIC REFERENCES
Acar, Y. B. & Olivieri, I. (1989). Pore fluid effects on
the fabric and hydraulic conductivity of laboratory
compacted clay. Transportation Research Record,
1219, 144-159.
Anderson, S. A. & Hee, B. H. (1995). Hydraulic
conductivity of compacted lateritic soil with
bentonite admixture. Environmental & Engineering
Geoscience, 1 (3), 299-312.
ABNT (1984). Soil Grains Passing the Sieve 4,8 mm –
Determination of Solids Density. NBR 6508, Rio
de Janeiro, 8 p. (In Portuguese).
ABNT (1984). Soil - Determination of the Liquid
Limit. NBR 6459, Rio de Janeiro, 6 p. (In
Portuguese).
ABNT (1984). Soil - Determination of Plastic Limit.
NBR 7180, Rio de Janeiro, 6 p. (In Portuguese).
ABNT (1984). Soil - Grain-size Distribution Analyses.
NBR 7181, Rio de Janeiro, 13 p. (In Portuguese).
ABNT (1986) Soil - Compaction Test. NBR 7182, Rio
de Janeiro, 10 p. (In Portuguese).
ABNT (1986) Soil Samples - Preparation for
Characterization and Compaction tests. NBR 6457,
Rio de Janeiro, 9 p. (In Portuguese).
Batista, P. (2006). Geotechnical assessment of
mixtures of soil lateritic with cement and bentonite
used in vertical blinds. Dissertation. Universidade
Federal de Ouro Preto. Ouro Preto. (In Portuguese).
Benson, C. H.; Zhai, H. & Wang, X. (1994).
Estimating hydraulic conductivity of compacted
clay liners. Journal of Geotechnical Engineering,
120 (2), 366-385.
Chapuis, R. P. (1990). Sand-bentonite liners:
predicting permeability from laboratory tests.
Canadian Geotechnical Journal, 27, 47-57.
Daniel, D. E. (1987). Hydraulic conductivity tests for
clay liners. Geotechnical and Geohydrological
Aspects of Wast Managemen, 15-34.
Day, S. R. & Daniel, D. E. (1985). Hydraulic
conductivity of two prototype clay liners. Journal
of Geotechnical Engineering, 111 (8), 255-270.
Farnezi, M. K. (2005). Alternative technical
improvement of tropical soils for use in sealing
systems baseline, through the application of
bentonite. Dissertation. Universidade Federal de
Ouro Preto. Ouro Preto. (In Portuguese).
Farnezi, M.K. & Leite, A.L. (2007) Lateritic soil and
bentonite mixtures assessment for liner usage
purpose. Soils and Rocks, 30 (2), 102-112.

Head, K. (1986). Manual of soil laboratory testing.
V.3. Pentech Press, London, 712p.
Keeney, T.; van Veen, M.; Swallow, M.; Singalia, M.
(1992). Hydraulic conductivity of compacted
bentonite-sand mixtures. Canadian Geotechnical
Journal, 29 (3), 364-374.
Mitchell, J.K. (1993). Fundamentals of Soil Behavior.
2 ed. John Wiley & Sons, New York, 437 pp.
Nogami, J. S. & Villibor, D. F. (1995). Flooring low
cost lateritic soils. São Paulo, Villibor, 213 p. (In
Portuguese).
Osinubi, K. J. & Nwaiwu, C. M. O. (2002).
Compacted lateritic soils as hydraulic barriers in
waste containment systems. In: International
Congress On Environmental Geotechnics, 4TH,
Rio de Janeiro, Balkema, Rotterdam, p.225-230.
Morandini and Leite 367
Pejon, O.J. (1992). Engineering Geological Mapping
of the Piracicaba, SP (1:100000 scale): A Study of
methods, Characterization and Presentation of the
Attributes. PhD Thesis, Engineering School of Sao
Carlos, University of São Paulo, pp. 59-71 (In
Portuguese).
Shackelford, C. D. (1994). Waste-soil interactions that
alter hydraulic conductivity. Hydraulic
Conductivity and Waste Contaminant Transport in
Soil. ASTM STP 1142, p. 111-166. Philadelphia.
Shackelford, C. D. & Jefferis, S. A. (2000).
Geoenvironmental engineering for in situ
remediation. In: GEOENG, Melbourne, 2000,
preprint.
Collapse Potential of a Lateritic Clay Liner by Contact with the Liquid
Phase of Red Mud
R M Q L Braga
Ideal Faculty, Belem, Brazil (risetemaria@yahoo.com.br)
M E G Boscov
University of São Paulo, Sao Paulo, Brazil (meboscov@usp.br)
M L Costa
Federal University of Para, Belem, Brazil (mlc@ufpa.br)

ABSTRACT This paper describes an investigation on the compatibility of a lateritic compacted soil to be used in the
construction of bottom liners and dikes in red mud disposal basins with the red mud liquid phase, i.e. maintenance of
geotechnical properties after contact with concentrated NaOH solutions. Compatibility was investigated by means of
simple and double collapse consolidation tests with water and sodium hydroxide solutions at the concentrations of
0.12%,. 1.23% and 5.00%, which is the range observed in disposed red mud. Applied pressures varied from 10 kPa to
320 kPa. Soil specimens compacted at the optimum water content of standard energy presented a significant increase
in compressibility when infiltrated by NaOH solutions.

INTRODUCTION
In Brazil, millions of tons of red mud, as wastes
generated by alumina production are called,
have to be disposed of each year. Red mud
results from the digestion of bauxite by sodium
hydroxide solutions according to Bayer process.
Predominant minerals are hematite, anathasium,
goethite (bauxite components) and sodalite
(formed during the transformation of bauxite
into alumina). The liquid phase, consisting of a
concentrated sodium hydroxide solution,
presents pH value of circa 12.
Red mud is deposited in basins lined with
geomembranes to protect the underlying aquifer
from the liquid phase of the mud. Disposal
basins located in flat lands are conformed by
encircling dikes also lined with geomembranes
to avoid contact with the mud and thus insure
structural safety, since caustic solutions may
alter soil structure and consequently the
mechanical and hydraulic properties of the soil.
Geomembranes, however, may present holes
even after excellent installation (Qian et al.
2002). Estimate of the behavior of soil barriers
exposed to caustic solutions should be
mandatory for environmental impact assessment
of red mud disposal sites.
This paper describes an on-going investigation
on the compatibility (maintenance of
geotechnical properties) of a lateritic soil to be
used in bottom liner and dike construction with
the liquid phase of red mud. Compatibility was
here assessed by the identification and
measurement of collapse of compacted soil
when in contact with caustic solutions.
COLLAPSE
Collapsible soils undergo a remarkable void
ratio reduction by wetting, with or without the
action of a surcharge; as the fluid seeps through
the soil it softens or destroys the bonds between
soil particles (Barden and Sides 1971).
 Braga, Boscov and Costa 369

Conditions for the occurrence of soil collapse Chemical analysis by X-ray fluorescence
are: an open and generally meta-stable structure, indicate the predominance of SiO2 (71-85%)
an applied effective stress sufficiently high to and Al2O3 (5-16%), and the presence of F2O3
develop potential instability, a high suction to (1.5–2.0%) and TiO2 (1.0-1.5%) concretions.
promote temporary gain of rigidity and strength,
and that removal of this suction causes collapse. Organic carbon content (0.2% to 0.4%) is low
In some soils cementation of contacts between but compatible with most Brazilian soils,
particles adds to the effect of suction, and especially for samples collected after the
collapse is also related to dissolution of natural removal of a 0.5m-thick superficial organic
cements. Infiltration of contaminant solutions layer. The cation exchange capacity of 3.27
may also cause collapse due to alterations in the cmolc/kg may be explained by the
distribution of superficial electric charges. mineralogical composition of the soil, whose
predominant clay mineral is kaolinite (CEC of 3
Collapsible soils may be identified by direct to 15 cmolc/kg, Mitchell 1976). Cation
methods, i.e. that measure collapse magnitude exchange capacity is also within the range for
and quantify the collapse potential, and indirect Brazilian soils, 79% of which present CEC
methods, which recognize collapse potential by lower than 10 cmolc/kg (Fadigas et al. 2002).
means of other parameters e.g. consistency
limits and chemical analysis. TABLE 1 Geotechnical characterization of the soil
Property Result
Single and double collapse oedometer tests are Grain-size distributiona (%)
the most employed direct methods to identify Medium sand 5
and measure soil collapse. A single collapse Fine sand 42
Silt 8
oedometer test is performed according to the Clay 45
procedure of a consolidation test (ASTM Liquid limit, wLb (%) 33
D2435) until a given vertical stress is reached, Plasticity limit, wPb (%) 21
when the soil is saturated and consequent Plasticity index, PIb (%) 12
displacements are measured. Double collapse Skempton activity, Ac 0.27
oedometer tests are simultaneously executed Optimum water contentd (%) 17.5
with two soil specimens, one at the natural Maximum dry unit weightd (kN/m3) 17.5
water content and the other saturated from the Hydraulic conductivitye (m/s)
1x10–9
beginning of the test (Jennings and Knight (optimum water content, Proctor energy)
Organic carbon contentf (%) 0.2-0.4
1957, 1975). pH (Distilled water) g 4.6-4.8
pH (KCl) g 4.0-4.1
MATERIALS AND METHODS Cation exchange capacityh (cmolc/kg) 3.27
Point of zero chargei 5.4
Soil Geotechnical classificationj ML
a
The investigated soil is a lateritic clay from Para ASTM 5422
b
State, in the Amazonian region of Brazil. ASTM D4318
c
Samples were collected at the depth of 1.5 m in A = PI/C (C = clay content)
d
the area where basins for red mud disposal will ASTM D698
e
Constant-head flexible-wall permeability test
be built. The geotechnical characterization of f
Walkley-Black Method (Walkley and Black 1934)
the soil determined by standard methods is g
ASTM D49721
presented in Table 1. h
Ion exchange resin (Rodella and Alcarde 1994)
i
Equation by Keng and Uehara (1975)
j
The soil is constituted primarily of quartz and Unified System of Soil Classification
kaolinite as determined by X-ray difractometry.
370 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Collapse Tests
Single and double collapse oedometer
consolidation tests were executed with distilled
water and NaOH solutions at concentrations of
0.12%, 1.23% and 5.00%, which is the range
observed in disposed red mud. Soil specimens
were compacted at the optimum water content
of standard Proctor effort, according to
compaction specifications for bottom liners.
Applied pressures varied from 10 kPa to 320
kPa. Special consolidation cells of stainless steel
were built for the tests with NaOH solutions.
Collapse potential under a given effective
vertical stress is defined as the height decrease
due to wetting relative to the initial height of the
soil specimen, and may be evaluated by
equations (1) or (2). Equation (2), originally
developed for double tests, takes into account
that different soil specimens may present
different initial void ratios and isolates volume
decrease due to wetting from the effect of
different compressibility between soil
specimens.
∆H ∆e
I= = (1)
Hi 1 + ei
∆H ε v ( wet ) − ε v ( nat )
I= = (2)
Hi 1 − ε v ( nat )
where I = collapse potential, Hi = height of the
soil specimen before wetting, ∆H = height
decrease due to wetting, ei = pore volume of the
soil specimen before wetting, ∆e = pore volume
decrease due to wetting, εv(nat) = volumetric
strain at the natural water content due to the
application of a given effective vertical stress,
εv(wet) = volumetric strain due to the application
of a given effective vertical stress and wetting
for single tests; and volumetric strain due to the
application of a given effective vertical stress
for a soil specimen wetted from the beginning
of the test for double tests.
For civil engineering purposes, collapse may be
classified as light, moderate and high for I(%) =
2, 6 and 10, respectively (Lutenegger and Saber
1988).
RESULTS AND DISCUSSION
Results of single and double collapse
oedometric tests are presented in Fig. 1 to 5.
Fig. 1 presents volumetric strain as a function of
applied vertical effective stress for double
collapse tests with 5 specimens: a specimen at
the compacted water content, a specimen wetted
from the beginning of the test with water, and
three specimens wetted from the beginning of
the test with different NaOH solutions.
Vertical stress (kPa)
10 100 1000
0
Volumetric strain (%)
5
10
15
20
25
30
Natural water content Water
NaOH 0.12% NaOH 1.23%
NaOH 5%
Fig. 1 Results of double collapse oedometric tests
with water and NaOH solutions at concentrations of
0.12%, 1.23% and 5.00% as wetting fluids
(concentration of 5.00% is equivalent to a solution of
50 g/L of NaOH in distilled water).
The compacted soil, as expected, does not
collapse when saturated with water. However,
the contact with NaOH solutions increases soil
compressibility significantly.
Potential collapse values for each liquid and
applied vertical stress are shown in Table 2. It
can be noticed that the potential collapse is
negative when water is the wetting fluid, which
can be explained by a slight expansion or to a
 Braga, Boscov and Costa 371

difference in the initial void ratios of the
specimens. For NaOH solutions the potential
collapse increases, reaching between 0.5 and 6.7
for the concentration of 0.12%, 5.6 to 12,0 for
the concentration of 1.23%, and 5.7 to 14.6 for
5.00%. These values evince a structural effect of
the permeation of caustic solution through the
soil altering its compressibility.
σ = applied vertical effective stress
εv(nat) = volumetric strain at the natural water content
due to an applied effective vertical stress
εv(wet) = volumetric strain due to an applied effective
vertical stress for a soil specimen wetted from the
beginning of the test
I = potential collapse
ei = initial pore volume of the soil specimen
∆e = pore volume decrease due to wetting

TABLE 2 Collapse potential evaluated for single and Figures 2 to 5 present the results of single
double collapse odeometer tests. collapse consolidation tests for water and NaOH
Double collapse oedometer tests
solutions. Associated potential collapse values
Fluid σ εv(nat) εv(wet) I
are also presented in Table 2.
(kPa) (%) (%) (%)
Water 40 2.25 2.20 -0.1
Vertical stress (kPa)
80 3.50 3.44 -0.1
1 10 100 1000
160 5.18 5.15 0.0 0
320 7.35 7.24 -0.1
2
NaOH 40 2.25 2.73 0.5 Vometric strain (%)
0.12% 80 3.50 4.84 1.4 4
160 5.18 8.30 3.3 6
320 7.35 13.53 6.7
8
NaOH 40 2.25 7.65 5.6
1.23% 80 3.50 11.36 8.1 10
160 5.18 14.74 10.1 12
320 7.35 18.45 12.0 14
NaOH 40 2.25 7.49 5.7 Water, 40 kPa Water, 80 kPa
5% 80 3.50 11.25 8.0 Water, 160 kPa Water, 320 kPa
160 5.18 16.65 12.1 Fig. 2 Results of single collapse oedometer tests with
320 7.35 26.65 14.6 water as wetting fluid for vertical stresses of 40, 80,
Single collapse oedometer tests 160 and 320 kPa.
σ εv(nat) εv(wet) I
(kPa) (%) (%) (%) Vertical stress (kPa)
Water 40 1.15 1.31 0.2 1 10 100 1000
80 3.32 3.38 0.1 0
160 5.56 5.60 0.1 2
Vometric strain (%)

320 9.80 9.84 0.0

4
NaOH 40 2.70 2.70 0.0
0.12% 80 4.41 4.68 0.3 6
160 5.59 7.61 2.2 8
320 7.61 9.00 1.5 10
NaOH 40 3.41 2.85 -0.6
1.23% 80 12
9.50 12.26 3.1
160 7.05 10.09 3.3 14
320 8.60 13.21 5.0 NaOH 0.12%, 40 kPa NaOH 0.12%, 80 kPa
NaOH 0.12%, 160 kPa NaOH 0.12%, 320 kPa
NaOH 40 3.50 3.65 0.2
5.00% 80 5.05 8.24 3.4 Fig. 3 Results of single collapse oedometer tests with
160 4.98 10.75 6.1 NaOH solution at concentration of 0.12% as wetting
320 6.44 10.09 3.9 fluid for vertical stresses of 40, 80, 160 and 320 kPa.
372 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Vertical stress (kPa) When the specimen was flooded with NaOH
1 10 100 1000 solution at the concentration of 0.12%, a volume
0
decrease corresponding to light risk of collapse
2 was noticed at the vertical stresses of 160 kPa
Vometric strain (%)

4 and 320 kPa. When the soil was in contact with

6 NaOH solutions with concentrations of 1.23%
and 5.00%, risk ranged from light to moderate
8
under vertical stresses equal or higher than 80
10
kPa. However, potential collapse measured in
12 the single tests was much lower than that
14 obtained by double tests at the same applied
NaOH 1.23%, 40 kPa NaOH 1.23%, 80 kPa stresses.
NaOH 1.23%, 160 kPa NaOH 1.23%, 320 kPa

Fig. 4 Results of single collapse oedometer tests with The controversy about the best method to
NaOH solution at concentration of 1.23% as wetting measure collapse potential should be analised in
fluid for vertical stresses of 40, 80, 160 and 320 kPa. the light of the clarity of evidence, i.e.
measurable collapse (which in this research was
Vertical stress (kPa)
1 10 100 1000
best observed by means of double collapse
0
oedometer tests than by single tests), as well as
in the procedure that best reproduces future
2
events in the field. For red mud disposal basins,
Vometric strain (%)

4 bottom liners and dikes may be or not in contact

6 with the interstitial solution of the red mud
8 during the elevation of the waste mass, that
10
depending on the moment when geomembrane
faults may permit infiltration of the solution into
12
the soil.
14
NaOH 5.00%, 40 kPa NaOH 5.00%, 80 kPa
NaOH 5.00%, 160 kPa NaOH 5.00%, 320 kPa
CONCLUSIONS
Fig. 5 Results of single collapse oedometer tests with High collapse deformations result from the
NaOH solution at concentration of 5.00% as wetting
contact of a compacted soil with concentrated
fluid for vertical stresses of 40, 80, 160 and 320 kPa.
NaOH solutions. Compacted soils at optimum
water content, generally stable regarding water
Results show that collapse does not occur by
saturation, may undergo a significant increase in
water flooding at any of the applied vertical
compressibility by infiltration of caustic
stresses. In fact, the soil structure at the
solutions. An increase in compressibility may
optimum water content is not meta-stable, as it
lead to a decrease in the safety of dams and
corresponds to the minimum possible porosity
liners of red mud disposal areas in case of
for that compaction effort. Initial saturation
geomembrane defects or failure. This aspect
degree of the tested specimens, varying from 69
must be emphasized to environmental protection
e 83%, is not usually associated to high suctions
agencies, designers, constructors and mining
either. After flooding with water, the degree of
companies. Present results must be completed
saturation reached 78 a 100%; the respective
with shear strength and permeability tests, as
decrease in suction is not likely to provoke
well as scanning microscopy and viscosity tests
collapse in the soil.
to assess microstructure dispersion.

REFERENCES
Barden L. and Sides G. (1971). Engineering
behavior and structure of compacted clay.
Journal of the Soil Mechanics and
Foundations Division, 97 (12), 1708-1709.
Fadigas, F.S., Amaral-Sobrinho, N.M.B.,
Mazur, N., Santos, L.H.C. and Freixo, A.A.
(2002). Concentrações naturais de metais
pesados em algumas classes de solos
brasileiros. Bragantia, 61(2), 151-159.
Jennings J.E. and Knight K. (1957). The
additional settlement of foundations due to a
collapse of the structure of sandy subsoils on
wetting. International Conference on Soil
Mechanics and Foundation Engineering.
Londres, (1), 316 – 319.
Jennings J.E. and Knight K. (1975). A guide to
construction or on with materials exhibiting
additional settlement due to “collapse of
grain structure”. Regional Conference for
Africa on Soil Mechanics and Foundation
Engineering. Durban, 99, 105 p.
Lutenegrer A.G. and Saber R.T. (1988).
Determination of collapse potential of soils.
Geotechnical Testing Journal, 11(3), 173-
178.
Braga, Boscov and Costa 373
Keng J.C.W. and Uehara G. (1974). Chemistry,
mineralogy and taxonomy of Oxisols and
Ultisols. Soil Crop Science Society of
Florida, Belle Glade, 33:119-126.
Mitchell J.K. (1976). Fundamentals of Soil
Behavior, John Wiley & Sons.
Qian X., Koerner R.M. and Gray D.H. (2002).
Geotechnical aspects of landfill design and
construction. Prentice Hall, Upper Saddle
River.
Rodella A.A. and Alcarde, J.C. (1994).
Avaliação de materiais orgânicos
empregados como fertilizantes. Scientia
Agricola, 51( 3), 556 - 562.
Walkley, A. and Black, I. A. (1934). An
examination of Degtjareff method for
determining soil organic matter and a
proposed modification of the chromic acid
titration method. Soil Sci. 37:29-37.
 Effect of pH on Sorption Characteristics of Bentonite
K. M. Nithya
Research Scholar, Department of Civil Engineering, IIT Madras, Chennai (kmnithya@gmail.com)
D. N. Arnepalli
Assistant Professor, Department of Civil Engineering, IIT Madras, Chennai (arnepalli@iitm.ac.in)
S. R. Gandhi
Professor, Department of Civil Engineering, IIT Madras, Chennai (srgandhi@iitm.ac.in)

ABSTRACT
Heavy metal retention capacity of soil depends on its sorption characteristics and buffer capacity. This is an
important consideration while selecting an efficient barrier material for containing industrial or municipal
solid waste, which may contain large amount of toxic and hazardous heavy metals. With this in view the
present study attempted to investigate the effect of pH on the sorption characteristics of bentonite which is
chief mineral in a clayey soil used as liner. To achieve this, commercially available sodium bentonite powder
has been used and characterized for its physico-chemico-mineralogical characteristics along with sorption
characteristics. In order to obtain sorption characteristics the batch sorption experiments were conducted over
a wide range of pH and liquid to solid ratio, L/S. Based on the results, it has been concluded that sorption
characteristics of the bentonite highly depends on pH and L/S.

INTRODUCTION different mechanisms such as (1) cation

The concentration and the mobility of heavy
metals in soils and sediments have been
widely studied by the researchers quite
elaborately in the past. Although many heavy
metal in trace level concentrations act as
essential elements for the normal
development of the biological cycles,
however these heavy metals do become toxic
at higher concentrations. Most of the heavy
metals get accumulated in excess of preferred
level, in certain geoenvironment either in the
natural phenomena and/or due to human
activities, such as aggressive agricultural
practices, transport, unplanned industrial
activities and indiscriminate waste disposal
practices (Hani, 1990). In order to minimize
the severity of the geoenvironmental
degradation, clay minerals have been,
popularly, used as barrier material at various
waste disposal and containment facilities
(such as engineered landfills), due to their
high specific surface area, low cost and
abundance availability (Bailey, 1999). In
general, these liner materials constitute of
smectite mineral of the montmorillonite
group, which has Na+ or Ca+ end members
(Kaya and Durukan, 2004). Montmorillonite
mineral can sorb heavy metals through two
exchange in the inter layers resulting from the
interactions between ions and permanent
negative charge and (2) formation of inner-
sphere complexes through Si-O- and Al-O-
groups at clay particle edges. Both the
mechanisms are pH dependent because,
during acid conditions most silanol and
aluminol groups are protonated; therefore, in
particular for the latter, acidification may lead
to an increase in mobility of metals bound to
soil (Batchelor, 1998). Further, retention of
heavy metal in geomaterials is very much
dependent on pH, since they are likely to
precipitate around neutral pH by forming
compounds such as hydroxides, sulfates, and
chlorates species. Hence the ability of the
geomaterial to accumulate heavy metal ions
and retard their mobility is quite depends on
the susceptibility of the soil to changes its pH
(buffer capacity). This is an important
consideration while evaluating the soil
mineral as barrier material to attenuate
contaminants consisting of toxic heavy
metals. For this reason, it is very much
essential to understand the influence of pH on
the sorption capacity of montmorillonite.
With this in view, Cu2+ was chosen as model
contaminant in this study, as copper is one of
 Nithya, Arnepalli and Gandhi 375

the principal heavy metals responsible for presented in Table1. The consistency limits
causing degradation of the geoenvironment (Atterberg limits) such as liquid limit, LL,
and associated health effects (NRC, 2000) and plastic limit, PL, and shrinkage limit, SL and
it is most commonly present in both differential free swell index, FSI were
municipal and industrial leachate. Sorption of determined as per the guidelines presented in
Cu2+ ions on bentonite depends on both ASTM D4318 and ASTM D427, respectively.
charge characteristics of the sorbent and The results obtained are presented in Table 1.
properties of metal ions (Adhikari et al.,
2003). Further factors such as metal TABLE 1 Physico-chemico-mineralogical and
concentration, pH, ionic strength, type and geotechnical characteristics of the geomaterial
concentration of competing ions, the liquid: Property Value
solid ratio and temperature also influence the Physical characteristics
sorption phenomena (Chen et al., 2002). Specific gravity, G 2.61
Gravel (%) 0
In view of the above, an attempt is made in Sand (%) 1.2
this present work is to study the sorption Silt (%) 33.6
Clay (%) 65.2
mechanism of Cu2+ on bentonite clay and the
LL (%) 206
effect of pH, and the amount of sorbent on the
PL (%) 43.2
sorption capacity of the bentonite. SL (%) 9.5
USCS Classification* CH
MATERIALS AND METHODS FSI (%) 475
Sample of commercially available sodium Geotechnical characteristics
bentonite (denoted as BT) was used in the Ȗdmax (g/cc) 1.38
present study. Copper, Cu2+, used as a model OMC (%) 30.3
contaminant and is prepared using the CuSO4 k×10-11 m/s 5.29
5H2O and deionized water (DI). It can be Chemical characteristics
noted that the background concentration of pH 10.3
Cu2+ in the geomaterial and in the pore CEC (meq./100g) 91
solution is negligible. To avoid the SSA (m2/g) 85.6
precipitation of Cu2+, solution pH equals to 2, Carbonates (%) 9
3, 4 and 5 was chosen for the present study Organic (%) 11.9
and pH was adjusted to the desired value Major oxides (%) Na2O=0.24;MgO=0.55;
using 1N NaOH and 0.1M HNO3 solutions. Al2O3=14.5;SiO2=42.9;
P2O5=1.2; K2O=1.8;
The concentration of the Cu2+ is determined
CaO=2.9; TiO2 =2.5;
with the help of atomic absorption Fe2O3=32.5; SO3=0.17;
spectrophotometer (AAS). Details of the Cl= 0.31.
experimental investigations that were carried Mineralogical characteristics
out on these samples are presented in the Minerals Present Illite, montmorillonite,
following. Hematite, feldspar,
Muscovite
Physicao-chemical, mineralogical and *Unified Soil Classification System (ASTM D2487)
geotechnical characteristics
The specific gravity, G, of the geomaterial The compaction characteristics of the
was obtained using a pycnometer (ASTM geomaterial such as maximum dry density,
D854). The particle size distribution Ȗdmax, and optimum moisture content, OMC,
characteristics of the samples were obtained were obtained as per the guidelines presented
by following the guidelines presented in in ASTM D698 and results are presented in
ASTM D422 and obtained results are Table 1. Further, coefficient of permeability
376 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
(k) was obtained with the help of one
dimensional consolidation test setup by
following the guidelines presented in ASTM
D2435 and results are reported in Table 1.
As depicted in Table 1, chemical
characteristics such as cation exchange
capacity, CEC, (as per IS 2720), carbonates
(Hesse et al., 1972), organic matter (ASTM
D2974) and pH (ASTM D4972) of the
samples were also obtained. The specific
surface area, SSA, of these samples was
obtained by employing the nitrogen gas
and the distilled water) were performed to
determine the sorption capacity of the bottle
and the initial concentration of the heavy
metal in the sorbent (Yong and
Phadungchewit, 1993), respectively. These
corrections were incorporated while
computing the corrected initial concentration
of the solution Ci and Ce (i.e., the
concentration of heavy metal present in the
solution after equilibration time, in mg/l).
Later, the amount of heavy metal sorbed on
the sorbent (geomaterial) (in mg/kg), Cs, can
be computed using the following relationship.
adsorption technique (BET) using surface
area analyzer and the obtained results are
presented in Table 1. In addition to this, the
mineralogical composition of the material was
also determined with the help of X-ray
Diffraction Spectrometer, XRD, (make
Phillips 2400), using a graphite
monochromator and Cu-.Į UDGLDWLRQ 7KH
chemical composition of the material in the
form of major oxides was determined using
an X-ray Fluorescence setup, XRF, (Phillips
1410, Holland) and the results obtained are
presented in Table 1.
Batch sorption experiments
Methodology
In the batch sorption experiments, a known
amount of soil is allowed to interact with
known initial concentration of heavy metal
solution under controlled humidity,
temperature and liquid to solid ratio, L/S. The
experiments were terminated when the
concentration of Cu2+ ion present in the solute
reaches practically constant value i.e., the
equilibrium between ions sorbed by the
geomaterial and of those in the solution. The
difference between the initial solute
concentration Ci and the concentration of
equilibrium solution Ce is proportional to the
amount of ions sorbed by the geomaterial.
The blank (sampling bottle filled with a
certain concentration of heavy metal without
the geomaterial) and control experiments
(sampling bottle filled with the geomaterials
C s Ci  Ce  u ( L / S ) (1)
Equilibration Time Determination test
As the sorption is a physico-chemical and
time dependant phenomena, hence it is
essential to estimate the optimal geomaterial-
contaminant interaction time at which the
geomaterial exhibits maximum sorption
capacity (characteristics). For this purpose the
recommendations made by Roy et al., (1992)
and ASTM D4646 were adopted, equilibrium
is considered to be achieved when there is no
more than a 5% variation in the concentration
of the solution over 24-h duration. In order to
obtain the equilibration sorption time, ts,
experiments were conducted with different
L/S and geomaterial is allowed to interact
with heavy metal solution of initial
concentration, Ci, equals to 500 mg/l. The L/S
is defined as the ratio of mass of solution to
mass of solids (geomaterial) used in the tests.
The ASTM D4646 and Roy et al. (1992) have
suggested that the minimum L/S should not
be less than 4 to avoid the difficulties
associated with separation, mixing and
analysis after the soil-contaminant interaction.
With this in view, the present study
considered five L/S values such as 10, 20, 50,
100, and 200 to establish sorption
characteristics. The mixture of metal solution
and solid particles (geomaterial) in
polypropylene bottle was gently shaken by
mounting them on a mechanical shaker for
different interaction times (5 min. to 240 h).
Later, sample bottles were removed from the

mechanical shaker and their contents were
centrifuged at 1000 rpm for 30 min., which
helps in separating solid particles from the
solution. The clear solution was decanted
from these bottles and was filtered using a 45
µm filter paper. The filtrate was analyzed for
their metal concentration using AAS. For the
sake of brevity, the solution concentration,
Csolution, corresponding to L/S of 20 and 200
for different interaction times were plotted as
depicted in Fig. 1.
RESULTS AND DISCUSSION
As depicted in Fig 1, the concentration of
Cu2+ in the solution (i.e. for a certain L/S),
decreased gradually from its initial
concentration Ci (of 500 ppm) to significantly
low values of (5 ppm and 350
ppmrespectively) over a period of time for a
given geomaterial. Further the solution
concentration of attains practically a constant
value corresponding to interaction time of 96
h and beyond. To avoid repetition, the results
corresponding to other L/S are not presented
herein. It is noted that the interaction time is
varying from 72h to 96h for other heavy
metals with selected geomaterials. This
indicates that 96 h can be considered as
equilibrium interaction time while performing
batch sorption experiments to obtain sorption
characteristics of the geomaterials.
As sorption characteristics of the various
geomaterials for given heavy metals ions is
often pH-dependant (Stumm, 1992; Farrah &
Pickeringf, 1977). Hence attempts were made
to demonstrate the influence of pH on
sorption of Cu2+ on the bentonite (sorbent)
was studied over a wide range of pH values of
2-5 and for a L/S values of 10, 20, 50, 100,
200 and the obtained results are presented in
Fig.2. It can be noted from the figure that, the
sorption of Cu2+ ions was maximal
corresponding to pH of 5, and incidentally
this finding is coinciding with those reported
in the literature (Bellir et al., 2005). By
decreasing the acidity of the solution, the
removal rate (%) as well as the amount sorbed
Nithya, Arnepalli and Gandhi 377
per unit mass (Cs), showed a positive
variation.
As it can be seen from Fig 2, the sorption
capacity of the bentonite increased as pH of
the solution was increased from 2 to 5.
Further from the figure we can observe that
lowest sorption rates were obtained at pH 2,
this may be attributed to the fact that, the low
pH solute consists of more number of H+ ions
and in turn compete with Cu2+ to get sorbed
on the limited amount of available negatively
charged surface of the bentonite clay. Further,
the concentration of H+ ions diminishes
corresponding to high pH values as solution
may consists copper in its hydroxide form
CuOH+. This situation limits the competition
between H+ and copper ions (Bhattacharya
et.al, 1984).
600
L/S
500
20
200
400
Csolution(ppm)
300
200
100
0
96 h
0 50 100 150 200 250 300
t (h)
Fig 1. Equilibration time curves for copper
solution with bentonite
(For colour figure, refer to CD)
In addition to this, the increase of solution pH
may favour the precipitation of copper ions
and then limits the availability copper ions to
get sorbed on the sorbent. Hence precipitation
phenomenon may be responsible for
eliminating the copper ions from the solution,
as the majority of the copper hydroxide may
get precipitated corresponding to pH of 5.6.
The basic mechanism that govern the sorption
characteristics of bentonite at pH ranging
between 2 and 5 are sorption and ion
exchange. At these pH levels exchangeable
cations, present at the exchangeable sites, i.e.,
378 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Na+, K+, Ca2+ and Mg2+ , are exchanged for reported by Vengris for sorption of nickel,
Cu2+ cations in the aqueous solutions. copper and zinc on modified clay sorbent.

12.0k 6.0k 120

L/S=10 L/S=20

9.0k 100
4.0k
Cs (mg/kg)

Removal rate (%)

6.0k 80
pH pH
2 2.0k 2
3.0k 60
3 3
4 4
5 5 40 L/S
0.0 0.0 10
0 20 40 60 80 100 0 10 20 30 40 50
20
16.0k 40k
20 50
L/S=50 100
L/S=100 200
0
12.0k 30k 0 1 2 3 4 5 6 7 8
Cs (mg/kg)

pH
8.0k 20k
pH pH
Fig 3. Effect of pH on Cu2+ sorption on
2 2
4.0k 3 10k 3
bentonite at different L/S values
4 4
5 5 (For colour figure, refer to CD)
0.0 0
0 40 80 120 160 0 100 200 300 400
Ce (mg/l) Ce (mg/l) CONCLUSION
50k
L/S=200
The sorption characteristics of bentonite were
40k determined for Cu2+ over wide range of pH
values and amount of sorbent present in the
Cs (mg/kg)

30k
20k
10k
0
0 100
Ce (mg/l)
Fig 2. Sorption of copper ions on bentonite at
different pH values
(For colour figure, refer to CD)
Figure 3 shows the relationship between %
removal rate & pH for all L/S values
considered for the present study. The removal
rate (%) is increased rapidly as L/S values
decreases. Availability of increasing the
number of exchangeable sites ensured the
enhanced uptake of Cu2+. However, the
amount of sorbed per unit mass of the sorbent
at its equilibrium decreased with a decrease in
the L/S values. Although the number of
sorption sites per unit mass of a sorbent
should remain constant, independent of the
total sorbent mass, increasing the amount of
sorbent in a fixed volume reduces the number
of available sites as the effective surface area
is likely to decrease. Similar results have been
solute i.e., liquid to solid ratio, L/S. The
pH
2
results obtained from the batch sorption
3
4
experiments indicate that, the retention
5 characteristics of the geomaterial is sensitive
200 300 400 500
to the pH of the solute and L/S ratio. A unit
change in the soil solution pH results
significant change in its retention capacity and
hence the sorption mechanism of heavy
metals in the soils. The decrease of solution
acidity increases amount of sorbed ion on the
sorbent. Further increase of L/S value results
decrease of removal rate, however sorbed
Cu2+ on unit mass (Cs) of the sorbent
increased at equilibrium
REFERENCES
Adhikari T. and Singh M. V. (2003) Sorption
characteristics of lead and cadmium in some soils
of India, Geoderma, 114, 81-92.
ASTM D422 (1994) Standard test method for
particle size analysis of soils, Annual Book of
ASTM Standards, 04.08, ASTM, West
Conshohocken, PA, USA, 10-16.
ASTM D427 (1994) Standard test method for
shrinkage factors of soils by Mercury method,

Annual Book of ASTM Standards, 04.08, ASTM,
West Conshohocken, PA, USA, 21-24.
ASTM D2487 (1994) Standard classification of
soils for engineering purposes (Unified Soil
Classification System), Annual Book of ASTM
Standards, 04.08, ASTM, West Conshohocken,
PA, USA, 206-215.
ASTM D4318 (1994) Standard test method for
liquid limit, plastic limit and plasticity index of
soils, Annual Book of ASTM Standards, 04.08,
ASTM, West Conshohocken, PA, USA, 551-561.
ASTM D 4972 (2001) Standard Test method pH of
soils, Annual Book of ASTM Standards, 04.08,
ASTM, West Conshohocken, PA, USA.
ASTM D2435 (2004) Standard Test Methods for
One-Dimensional Consolidation Properties of Soils
Using Incremental Loading, Annual Book of ASTM
Standards, 04.08, ASTM, West Conshohocken,
PA, USA.
ASTM D 4646 (2004) Standard test method for 24-
h batch-type measurement of contaminant sorption
by soils and sediments, Annual Book of ASTM
Standards, 04.11, ASTM, West Conshohocken,
PA, USA, 44-48.
ASTM D854 (2006) Standard test methods for
specific gravity of soil solids by water pycnometer
method, Annual Book of ASTM Standards, 04.08,
ASTM, West Conshohocken, PA, USA.
ASTM D 2974 (2007) Standard Test Methods for
moisture, Ash, and Organic Matter of Peat and
Other Organic Soils, Annual Book of ASTM
Standards, 04.08, ASTM, West Conshohocken,
PA, USA.
ASTM D698 (2007) Standard test method for
laboratory compaction characteristics of soils using
light effort, Annual Book of ASTM Standards,
04.08, ASTM, West Conshohocken, PA, USA.
Bailey S.E. Olin T.J. Bricka R. and Adrian D.D.
(1999) A review of potentially low-cost sorbents
for heavy metals, Water Res, 33(11), 2469–2479.
Batchelor B. (1998) Leach models for
contaminants immobilized by pH-dependent
mechanisms, Environmental Science and
Technology, 32(11), 1721–1726.
Bellir K. Bencheikh-Lehocine M. Meniai H. A.
and Gherbi N. (2005) Study of the retention of
heavy metals by natural material used as liners in
landfills, Desalination, 185, 111–119.
Bhattacharya A. K. and Venkobacher C. (1984)
Removal of cadmium (II) by low cost adsorbent,
Nithya, Arnepalli and Gandhi 379
Journal of Environmental Engineering, 110 (1),
110-122.
Chen J. and Du G. C. (2002) The production and
application of environmentally friendly materials,
Beijing, Chemical Industry Press.
Daniel D. E. (1993) Geotechnical practice for
waste disposal, Chapman &Hall, London, UK.
Farrah H. and Pickeringf W.F. (1977) The effect
of pH on the retention of Cu, Pb, Zn and Cd by
clay-humic acid mixtures, Water Air Soil Pollution,
8, 189-179.
Hani H. (1990) The analysis of inorganic and
organic pollutants in soil with special regard to
their bioavailability, International Journal of
Environmental Analytical Chemistry, 39(2):197–
208.
Hesse P.R. (1972) A textbook of soil chemical
analysis, Chemical Publishing, New York.
Kaya A. and Durukan S. (2004) Utilization of
bentonite-embedded zeolite as clay liner, Applied
Clay Science, 25, 83-91.
NRC (2000) Copper in Drinking Water, National
Research Council, National Academy Press,
Washington, D.C.
Roy W.R. Krapac I.G. Chou S.F.J. and Griffin
R.A. (1992) Batch- type procedures for estimating
soil adsorption of chemicals, U.S.EPA, technical
resource document, EPA/530-SW-87-006-F,
Washington D.C.
Stumm W. (1992) Chemistry of the solid–water
interface, John Wiley &Sons, New York, 21–34,
229–232.
USEPA (United States Environmental Protection
Agency. (1987) Requirements for Hazardous
Waste Landfill Design, Construction, and closure,
EPA /625/4-89/022, USEPA Cincinnati.
Vengris T. Binkiene R. and Sveikauskite A. (2001)
Nickel, copper and zinc removal from waste water
by a modified clay sorbent, Applied Cay Science,
18,183-190.
Yong R.N. and Phadungchewit Y. (1993) pH
influence on selectivity and retention of heavy
metals in some clay soils, Canadian Geotechnical
Journal, 30, 821-833.
 Modern Landfill Pollution Barriers with the Use of Geosynthetics
J K Pries
BBG Bauberatung Geokunststoffe GmbH & Co. KG, Espelkamp, Germany
(jpries@bbgeo.com)
R Clitus
NAUE GmbH & Co. KG, Chennai, India (crobinson@naue.com)

ABSTRACT Since decades, geosynthetics like geomembranes, geotextiles, geogrids and geosynthetic clay
liners are increasingly used in sealing systems for environmental protection (landfill liners etc.). The paper
will give examples of state-of-the-art sealing systems for environmental protection using geosynthetic
materials, focussing on the advantages of composite liner systems as the most reliable pollution barriers. In
landfill engineering, basal linings and caps are designed in two different manners to maximise containment of
potential pollutants. Basal lining systems include geosynthetic sealing and drainage components to contain
the fluids and contaminants which leach from the waste as well as divert these fluids to a collection point for
treatment. Geosynthetic landfill caps utilise sealing and drainage layers to prevent precipitation from
penetrating into the landfill body, minimising the generation of leachate in closed facilities and minimizing
also the release of landfill gas into the atmosphere. The possibility is given that a complete geosynthetic
solution protects the environment from potential contamination of pollutants.

INTRODUCTION less space, save resources, reduce installation

Engineered landfill sealing systems are an
important tool in the containment strategy for
waste disposal. The principal goal of an
engineered landfill sealing system is to avoid the
contamination of the groundwater and the subsoil
and to prevent or control infiltration of
precipitation, thereby controlling or preventing
leachate production and to control the
development and emission of landfill gas, thus
preventing its emission into the atmosphere. To
achieve these goals a properly designed system
with appropriately designed and selected
components is required.
Geosynthetic landfill caps utilise sealing and
drainage layers to prevent precipitation from
penetrating into the landfill body, minimizing the
generation of leachate in closed facilities. The
possibility is given that a complete geosynthetic
solution protects the environment from potential
contamination of pollutants. The European
communities council directive 1999/31/EC as well
as the US Environmental Protection Agency
(EPA) Federal Regulations have defined mainly
mineral sealing and drainage layers, excepting the
geomembrane. The defined mineral drainage and
sealing layers can be replaced by geosynthetic
components, if the equivalency can be proven.
Compared to conventional construction methods,
they are also typically more economical, require
costs and demonstrate technical advantages.
GEOSYNTHETIC FUNCTIONS IN
LANDFILLS
Sealing
Acting as liquid and gas barriers, geomembranes
have become a fundamental component in landfill
engineering, due to the greater need for
groundwater protection. High density
polyethylene (HDPE) geomembranes, specifically
those with a certificate from governmental side,
and a thickness of more than 1.5 mm, are mostly
used.
In regions where local soils are not suitable for the
construction of hydraulic barriers, geosynthetic
clay liners (GCLs) can be used to protect the
hydrogeological resources by providing an
effective barrier against potential risks for the
environment. Typical compacted clay liners
(CCLs) are difficult as well as expensive to build.
A geosynthetic clay liner can replace or augment
the typical CCL in most designs to minimize the
thickness of the sealing system as well as the costs
for installation. To be able to judge the sealing
properties of a GCL and a CCL, a comparison of
the permeability, under consideration of the
permittivity (k/d-value), can be worked out. A
comparison of the permittivity considers the
hydraulic conductivity (kf-value [m/s]) as well as
the effective thickness of the sealing system.

Additionally a comparison of the flow rate under
consideration of the effective hydraulic head can
be used.
Protection
Geomembranes, structures, coated materials as
well as related construction elements must often
be protected from potential mechanical damage.
Without suitable protection, damage may occur by
sharp-edged objects such as stones, from the
unevenness of the subsoil or by the cover material.
Mechanically bonded needle-punched nonwovens
and composite materials manufactured from
polypropylene (PP) are normally used for
protection layers. Specific to nonwoven
geotextiles, the protection function is directly
related to the thickness and mass per unit area, the
heavier and thicker nonwoven provides better
protection.
Drainage
Drainage materials are required for the surface
collection of precipitation, the subsurface
collection and diverting of groundwater, as well as
the general collection of fluids and their diverting
into a drainage system. Drainage systems are
typically designed with individual material layer
or in combination with other components to create
pre-formed composite drainage elements.
Composite drainage elements consist of at least
one filter layer and one drainage layer. The filter
layer is required for the flowing and diverting of
fluids to a collection point, without the build-up
pressure. Single and multiple component
geosynthetic drainage systems made with high
density polyethylene (PE) as well as
polypropylene (PP), often replace the
conventional thick mineral drainage layer as liquid
or gas collection system. The long-term
transmissivity [m²/s] gives comparable values
about the discharge capacity.
Separation
As a separation layer, geotextiles are used to
prevent adjacent soil layers or fill material from
intermixing. Synthetic nonwovens that exhibit an
elongation capacity are the materials of choice in
most applications. The selection of a suitable
Pries and Clitus 381
product depends upon the grain size of the fill
material and the operational loads to be expected.
Filtration
In filtration applications nonwoven geotextiles are
applied to retain soil particles while allowing the
vertical passage of liquids through the filter media.
There are two aspects of filtration that should be
considered when designing. The mechanical filter
efficiency (does the fabric have sufficient soil
retention capacity?) and the hydraulic filter
efficiency (does water divert without a hydraulic
pressure build-up?). As with mineral filter layers
the geotextile thickness directly benefits the long-
term mechanical and hydraulic efficiency of the
filter.
REGULATIONS FOR SEALING SYSTEMS
Landfill sealing systems
The US Environmental Protection Agency (EPA),
Federal Regulations and the European Council
Directive on the Landfill of Waste describe
prescriptive cross sections for hazardous and non-
hazardous waste landfill base and capping sealing
systems. These systems are designed to prevent
the escape of containments from land filled waste
into underlying soils (base sealing system) and the
intrusion of precipitation into the waste or the
release of landfill gas from the waste (capping
sealing systems). Typical cross sections of base
sealing systems are shown in Figure 1 and
Figure 2.
Waste
Leachate collection
(50 cm)
Geomembrane
Geological barrier
(1 m) < 1x10-9 m/s
Fig. 1 EU directive for base seals (non-hazardous)
(For colour figure, refer to CD)
382 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Waste Geosynthetic base and capping sealing systems

Filter soil
There are significant economic as well as
performance benefits derived from geosynthetic
Leachate collection layer alternatives compared to conventional, thick
(30cm) material layers. Natural resources as well as
> 10-4 m/s
airspace volumes are preserved; construction
HDPE geomembrane volumes, traffic and transportation costs are
1.5 mm (60 mils)
reduced, construction timeframes are reduced and
performance can remain at the same level or
Compacted clay liner higher.
(60 cm) d 1 x10-9 m/s
In figure 5 a geosynthetic alternative capping
sealing system according to the EU-directive for a
non-hazardous landfill is shown.
Subgrade Top soil (• 1 m)
Fig. 2 US-EPA directive, for base seals (non- Top soil (• 1m) Geosynthetic drainage layer
hazardous)

Topsoil (> 1 m) Geosynthetic clay liner

Drainage layer (> 50 cm)

Geosynthetic gas drainage layer
Waste
Waste
Impermeable mineral layer Levelling layer
Fig. 5 EU directive equivalent capping solution
(non-hazardous)(For colour figure, refer to CD)
Gas control
Waste

In recent years serious consequences in

Fig. 3 EU directive for landfill caps (non-hazardous)
consideration of the mineral sealing systems are
HDPE geomembrane
known, if settlements of the waste body appear.
1.5 mm (60 mils) The settlements result into deformations of the
CCL which can lead to a permeable CCL. The
most serious problem for a top liner is the
desiccation (less critical for a geological barrier in
base sealing systems). The results and the end of
the quotation of the status-workshop "Desiccation
Compacted clay liner behaviour of mineral sealing layers in landfill
(45 cm)
< 1x10-7 m/s
surface sealing systems" (2002) of the German
Geotechnical Society is given in the following:
"However, it has to be pointed out that in spite of
Gas vent the realisation of the causes and the existing
Waste empirical experiences no definite
Fig. 4 US-EPA directive for landfill caps (non- recommendations can be given at the moment for
hazardous) (For colour figure, refer to CD) the long term functional design of landfill capping

sealing systems with compacted clay liners
(CCLs)." The replacement of a CCL with a GCL
can minimize the effects of desiccation. Once
covered, a geosynthetic clay liner quickly hydrates
with existing soil moisture to form a low
permeable barrier.
In many cases where a CCL is used as sealing
component, they are subjected to installation
conditions that can cause desiccation cracking and
create preferential flow paths. In these cases the
geomembrane is the essential part for
environmental protection. In cases when the CCL
has very high water content, it can cause stability
problems when installed on steep slopes.
Latest product developments of GCLs have
additionally a polymeric coating added to the GCL
which is placed on the top of GCL side during
installation. These polymeric coatings act as a root
barrier as well as a desiccation barrier and allow
the moisture to stay in the bentonite in water-free
periods. Therefore desiccation of the bentonite
does not occur and typical clay cracking of the
bentonite is prevented."
Change of the water content after
installation
Desiccation: Sealing problems!
Re-hydration: Stability problems!
Fig. 6 Desiccated compacted clay liner (CCL)
(For colour figure, refer to CD)
For sealing applications multidirectional shear
strength transmitting needle-punched GCLs can
Pries and Clitus 383
replace difficult to build compacted clay liners in
non-hazardous landfills as well as hazardous
landfill capping sealing systems. Due to their
excellent multi-axial strain behaviour, needle
punched GCLs retain their sealing capability
under high elongation, differential settlement
conditions of landfill capping systems. Steep slope
applications up to 2.5(H):1(V) are possible with
needle-punched nonwovens as carrier and cover
components of these GCLs.
Desiccated sample Re-hydrated sample
"Self-healing effect" of a GCL
Fig. 7 Radiogram of desiccated and re-hydrated
GCLs "GCL's self healing effect"
When used in combination with an HDPE
geomembrane a geosynthetic clay liner can
achieve the superior hydraulic performance
obtained with exceptional intimate contact, and
can minimize lateral flow of leachate – in the
unlikely event of a puncture. The combination of a
non-polar sealing element (geomembrane) with a
polar element (mineral sealing element) is a very
effective system with regard to the retention
capacity of pollutants. Such systems are defined as
"composite" sealing systems (see figure 8).
To achieve best pollution retention results in a
composite sealing system with a mineral sealing
layer, a minimum thickness of 2.5 mm of the
HDPE geomembrane is defined by law in
Germany.
The transport of pollutants through the pore
channels takes place with the water flow or is
powered by differences between concentrations
(diffusion). Inorganic, water-soluble organic and
water insoluble organic pollutants are contained in
the water and transported through soil layers with
the gravitation or respectively with the hydraulic
gradient or by diffusion.
384 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

The mass transport of pollutants in the pore Berlin, Germany, will keep the diffusion rate at a
channels of the mineral sealing layer is possible. In low level.
geomembranes, which do not have a pore volume 160
the mass transport in the pore channels does not 1
1 trichlorthylene
2 chloroform

Residual permeation (permeation rates) [g/(m² · d)]

exist. Here the mass transport takes place based on 140
3 tetrachloroethylene
concentration differences by diffusion. Due to the 4 toluene
120 5 carbon tetrachloride
different possibilities of the mass transport in the 6 xylene
various layers, different pollutants can be 7 chlorobenzene
100 thickness =
transported. 2,5mm
Considering a geomembrane in a composite liner 2
80
system, which has no holes, the inorganic 3
4
pollutants, such as cations and anions, are water 60
insoluble and will be retained by the geomembrane. 5
6
Water insoluble organic pollutants enter into the 40
7
geomembrane and pass the geomembrane because
20
of diffusion processes. The water insoluble organic
pollutants will be retained in the contact area 0
0 1,0 2,0 3,0 4,0 5,0 6,0 7,0
between the geomembrane and the mineral sealing Thickness of geomembranes [mm]
layer. Water soluble organic pollutants diffuse
Fig. 9 Diffusion of concentrated hydrocarbons and
through the geomembrane and can be transported chlorinated hydrocarbons through HDPE geomembrane
by the water flow in the pore channels of the (For colour figure, refer to CD)
mineral sealing layer. Therefore the water soluble
organic pollutants can reach the ground water (see The physical requirements and chemical resistance
Figure 8). of HDPE geomembranes for landfill caps are
inorganic
inorganic pollutan
pollutants comparable to those needed in landfill basal
water-soluble organicorg
water-soluble pollutants
linings. HDPE geomembranes have been
water-insoluble organico
water-insoluble
pollutants
pollutants successfully tested for their resistance to many
k=0 geomembrane chemical environments and are considered
suitable for landfill sealing systems in the bases
and for covers.
i=0 mineral sealing (CCL* / GCL**)
Structured or textured HDPE geomembranes are
k>0 used to increase shear force transmission between
the adjacent components under more challenging
conditions (e. g. steep slopes) to ensure long-term
groundwater stability of the sealing system against sliding.

* CCL = Compacted Clay Liner Geomembranes must be protected from

** GCL = Geosynthetic Clay Liner mechanical damage by cover materials or any
Fig. 8 Pollutant retention of composite liner systems underlying source. If using a mineral drainage
(For colour figure, refer to CD) layer on top of the geomembrane, mechanically
The diffusion of concentrated hydrocarbons and bonded PP nonwovens made of crimped staple
chlorinated hydrocarbons through HDPE fibres are often installed rather than a sand
geomembranes reduces with increasing thickness protection layer. To avoid significant
of the used membrane (see Figure 9). The geomembrane deformation during installation and
minimum requirement for geomembrane to ensure long-term protection, needle-punched
thicknesses of 2.5 mm according to German law, nonwoven geotextiles with a minimum mass per
which are also approved by BAM (Federal unit area of 800 g/m² are recommended.
Institute for Materials Research and Testing),

When used in combination with an HDPE
geomembrane, geosynthetic drainage composites
not only fulfil the filter and drainage function, but
also act as a protection layer for the geomembrane.
The drainage geocomposite as a protection and
drainage system also prevents deformation in the
HDPE geomembrane (see Table 1).
TABLE 1: Mass per unit area of geotextiles or
drainage composites as protection layer in consideration
of the height of the waste and the cover soil
Height of Mass per unit
Mass per unit
top soil area drainage
area nonwoven
layer composite
Up to 1.0 m > 800 g/m² > 1.000 g/m²
Up to 1.5 m >1.200 g/m² > 1.000 g/m²
In landfill caps, geosynthetic drainage systems can
effectively transmit and direct the percolated
rainwater to a collection or drainage point. Proper
drainage minimises water accumulation on the
sealing system which increases liner efficiency
and improves the stability of the capping sealing
system.
State-of-practice geosynthetic drainage systems
consist of three individual layers made of the
following components:
¾ filter nonwoven – protects the drainage core
from clogging,
¾ drainage layer – transmits the water in the
filter level,
¾ filter/protection nonwoven – acts as filter or
separation layer to the mineral component
The composite bonding of the three layers ensures
the uniform shear force transmission within the
drainage system. Due to the uniform bonding,
some geosynthetic drainage composites are also
suitable for the use in steep slopes.
The stable structure of the core layer of some
drainage geocomposites provides the long-term
drainage performance of these products. Drainage
geocomposites can also be used as the gas venting
and protection layer beneath the liner capping
system. Temporary covers, which are installed
during landfill operations, can minimize
Pries and Clitus 385
precipitation infiltration and reduce gas emissions,
when applied prior to the final or temporary seal.
A 1.5 mm thick HDPE geomembrane is ideal for
this application.
CONCLUSIONS
The repair or remediation of a capping sealing
system of a landfill is technically difficult and
expensive. It is for this reason that high quality
products, which can ensure long-term
performance, need to be selected during the design
phase of a project. Complete geosynthetic
solutions can be employed to address the site-
specific requirements of a project. Products
certified by government regulators are available
for single-layered and composite liner systems.
A desiccated CCL will never be a homogeneous
impermeable sealing component. Also a quality
installation of geosynthetics is necessary to avoid
any failure in the systems. A wide range of
geosynthetic products and styles allows efficient,
project specific solutions, even for the most
challenging applications.
Considering the economic and ecologic
advantages of the geosynthetic solution, the
surface sealing system with components like
geomembranes, geosynthetic clay liners and
geosynthetic drainage systems should be the
preferred solution.
REFERENCES
European communities' council directive
1999/31/EC
BAM (August 1992); Report No.: UBA-FB
102 03 412: Permeability behaviour of
composite liner systems in landfills and
existing waste deposits as to materials which
are hazardous to water“, Berlin
Rowe, R. K. (1998) Geosynthetics and the
minimization of contaminant migration
through barrier systems beneath soild waste.
Keynote Lecture, Proceedings of the Sixth
International Conference on Geosynthetics,
Atlanta, March, Vol.1, pp. 27-103, Industrial
Fabrics Association International, St. Paul.
MN.
US Environmental Protection Agency (EPA)
 Shedi Soil as Liner for Impoundments

C Ramakrishnegowda
Nagarjuna College of Engineering and Technology, Devanahalli, Bangalore, India
(rkgowda_32@yahoo.com)
P V Sivapullaiah
Indian Institute of Science, Bangalore, India (siva@civil.iisc.ernet.in)

ABSTRACT A natural soil available near and around Mangalore, called Shedi soil containing mixed
minerals of kaolinite and smectite, is assessed whether it can exploited for t liner applications for waste
impoundments. The hydraulic conductivity has been arrived at by conventional oedometer testing with water
and with a variety of fluids such as sodium hydroxide, phosphoric acid, and sodium carbonate solutions. At
any void ratio the conductivity generally increases with contaminants. The increase in the hydraulic
conductivity depends on both the type and concentration of contaminant. The concentration of contaminant at
which maximum hydraulic conductivity that occurs due to soil contaminant interaction has been identified.

INTRODUCTION exchangeable cations; ii) adsorbed anions, and

Liners are frequently constructed with natural
materials serving as the primary barrier to
liquid/leachate movement (Brandl, 1992; Rowe et
al., 1995). Clays across sections to be installed at
waste disposal sites. Soil liners are preferred
because of their low cost, large leachate
attenuation capacity and resistance to damage and
puncture. Clay is the most important component
of soil liners because the clay fraction of the soil
ensures low hydraulic conductivity. The hydraulic
conductivity of soils is the most important
parameter involved in the assessment of
contaminant migration in the sub surface (Mitchell
& Jaber 1990, Mitchell & Madsen, 1987). The
hydraulic conductivity is influenced several
factors. It is generally known that permeability of
remoulded soils depends on the moulding water
content, compaction method and compactive
effort. Further the properties of permeants, such as
its concentration and duration of exposure also
play very important role. The variations in the soil
parameters are considered in terms of i) associated
variations in permeating fluid (Lakshmikantha &
Sivapullaiah, 2005; Sivapullaiah,2009), which are
reflected through changes due to (i) flocculation of
the clay, (2) dissolution of the clay minerals
(aluminosilicates) and (3) dissolution of other
minerals (e.g., CaCO3) in the clay soil. Significant
increase in hydraulic conductivity may result from
shrinkage of the soil matrix in the presence of
concentrated organic solvents, and acid base
dissolution of the soil constituents (Yang
&Barbour, 1992). The paper presents the
variations in the hydraulic conductivity of shedi
soil with three different type contaminants, viz.,
acid, salt alkali solutions.
MATERIALS AND METHODS
Shedi Soil
The naturally occurring shedi soil has been
procured from an industrial area of Surathkal, near
Mangalore, Karnataka State of India. The soil was
collected by open excavation from a depth of two
meter from natural ground level. In the coastal
 Ramakrishnegowda and Sivapullaiah 387

lateritic belt, where the top lateritic soil is Table 1. Physical Properties of Shedi Soil
followed by weak lithomargic clay of whitish
(shedi) colour. Shedi soil was collected after
removing the top lateritic soil. The soil was hand Property Value
sorted to remove stones and vegetative matter if Particle size distribution
any. The soil was further dried and pulverized Sand size, % 34
and sieved on 4.75mm sieve to eliminate gravel Silt size, % 38
fractions if any. Finally the soil was sieved Clay size, % 28
through 425-micron sieve and used. X-ray Consistency limits
diffraction studies of Shedi soil indicated peaks Liquid limit, % 60.6
7.14, 3.57, 2.34, 2.5 A0 pertaining to 1:1 (1 silica Plastic limit, % 37.3
Shrinkage limit, % 27.0
sheet and 1 alumina sheet) Kaolinite mineral and
Plasticity Index, % 23.3
peaks 14.4, 1.54 A0 that are characteristic peaks
of saponite, a smectite group mineral. Specific gravity 2.49
Geotechnical properties of the soil are presented in Compaction characteristics
Table 1 Optimum moisture content, % 27
Maximum dry density, kN/m3 14.23
Consolidation test and Hydraulic Conductivity Consolidation characteristics
Coefficient of consolidation cv, 0.403
One-dimensional consolidation tests were cm2/min
conducted in modified Oedometer to measure Compression index, Cc 0.104
permeability using different concentration of
contaminant fluids as per ASTM D2435-04 Coefficient of permeability (k), m/s 2.71x10-9
method. To reduce the side friction between the Strength parameters:
rings and specimen, the inner surface of the rings Cohesion, ccu (kPa) 36
were lubricated with silicon grease. Shedi soil and Angle of internal friction, (Icu) 1.54q
soil contaminated with varying concentrations of
contaminants fluid were statically compacted to
the Proctor’s maximum dry density in a 60 mm RESULTS AND DISCUSSIONS
diameter and 20 mm high consolidation ring to a
thickness of 14 mm. The ring was then mounted in Hydraulic Conductivity of Shedi Soil
a consolidation cell and positioned in the loading
frame. At a nominal pressure of 5 kPa, the sample It is well recognised that consolidation testing
was inundated with water. A load increment ratio method offers best means of measuring the
of one was adopted and the load increment changes in the hydraulic conductivity of soils with
duration was kept as 24 hours or until primary different chemicals. Also this method can be used
consolidation was complete. At the end of to obtain hydraulic conductivity at different
consolidation under the applied pressure 10, 20, pressure and at different void ratios. Since the
50, 100, 200, 400 and 800 kPa falling head samples used for measuring the hydraulic
permeability test is carried out. The falling head conductivity were already compacted, the
test is continued for 24 hours observing the variations in the void ratios and consequently the
variations of the water level with time. The hydraulic conductivity were restricted. It is
consolidation test is then resumed. Constant dial observed that the hydraulic conductivity increases
gauge readings have been noticed well with in 12 from 0.662 x 10-9 m/s to 2.96 x 10-9m/s as the void
to 18 hours even at varying concentrations. An ratio is increased from 0.24 to 0.47. Similarly the
observation period of 24 hours is used for the hydraulic conductivity decreases as the effective
permeability measurement. pressure is increased. The effect of variations in
the hydraulic conductivity of shedi soil due to
presence of various contaminants has been studied.
388 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

more dispersed structure reducing the hydraulic

Effect of Sodium Hydroxide conductivity.
The hydraulic conductivity of the soil varies with
applied pressure and with void ratio. However,
void ratio achieved in the soil in any contaminant
depends on the applied pressure.

Effect of Applied Pressure

As expected the hydraulic conductivity of alkali

contaminated soils decreases with increase in
vertical effective pressure. This is more due to
reduction in the void ratio with increase in
pressure. It is generally seen that with any applied
pressure the hydraulic conductivity increases with
increase in the concentration of alkali solution. Figure 1 Hydraulic Conductivity versus Void Ratio
However, the rate of increase with increase in Plots for Sodium Hydroxide Treated Soil
pressure is not the same for all samples. There is
steep increase in the hydraulic conductivity at
higher concentration beyond 6%. This might be Table 2 Hydraulic Conductivity x 10-9, m/s with
due to attack of mineral structure by strong alkali. contaminants
It is well known that alkali can attack silica of soil
structure. Thus the increase in hydraulic Sodium Sodium
Phosphoric
Hydroxide Carbonate
conductivity might be more due to attack of Contaminant Acid (E-
(E-09, (E-09,
mineral structure by alkali and consequent change (%) 09, m/s)
m/s) m/s)
in the particle arrangement. It is thus necessary to at e = 0.2
at e = 0.4 at e = 0.3
compare the effect of contaminant on the 0 2.30 0.95 1.65
hydraulic conductivity at comparable void ratios.
1.0 2.90
Effect of Void Ratio 1.5 1.05
2.0 0.93
Figure 1 shows the variation of hydraulic
2.5 4.10
conductivity of soil with different alkali
contaminated soil with void ratio. Void ratio for 3.0 1.28
the alkali contaminated soil has been taken from 3.5 1.5
void ratio pressure relationships. It is clearly seen 4.0 2.11
that the hydraulic conductivity increases steeply 4.5 2.11
with increase in void ratio with any concentration
5.0 2.2
of alkali. The steepness increases with increase in
the concentration of alkali. Thus to assertion the 6.0 2.00 1.45
effect of alkali on the increase in hydraulic 7.5 4.32 5.49
conductivity they are compared at any given void 8.0 4.83 2.73
ratio and summarized in Table 2. As seen from 10.0 1.75 1.87
Table 2 at the same void ratio, the hydraulic
10.5 2.98
conductivity of shedi soil with sodium hydroxide
up to 6% is lower than the soil itself. This 12.0 3.30 1.50
decrease may be due to fabric changes that occur
with alkali solution. At high pH the soil assumes

However the hydraulic conductivity is higher with
8% slightly but start decreasing with further
increases in the amount of sodium hydroxide. It is
known that soil alkali interaction can lead to
formation of zeolites. At a low concentration the
formation is negligible. But with high
concentration of sodium hydroxide more and more
formation occurs which would block the pores
reducing the hydraulic conductivity. These
compounds which are generally known to be
swelling type may not increase the compressibility
but can reduce the hydraulic conductivity. This
clearly demonstrates that higher concentration of
sodium hydroxide changes the nature of particle
arrangement and/or produces some new minerals.
Effect of Phosphoric Acid and Sodium Carbonate
The effect of phosphoric acid and sodium
carbonate on the hydraulic conductivity of shedi
soil is similar to that of sodium hydroxide. The
hydraulic conductivity of soil at different void
ratio is compared with that of sodium hydroxide at
different concentrations. But the maximum
hydraulic conductivity that occurs is different with
different contaminants. The maximum hydraulic
conductivity of shedi soil with different
contaminants are shown in Figure 2
Figure 2 Maximum Hydraulic Conductivity Values with
different contaminants
Ramakrishnegowda and Sivapullaiah 389
CONCLUSIONS
The type and amount of contaminants affect the
hydraulic conductivity of shedi soil differently.
These variations can some time be explained by
the changes that occur in the volume change
behaviour of soil in the presence of contaminants,
whereas on other counts, the hydraulic
conductivity is affected by formation and
dissolution of minerals or reaction products. The
hydraulic conductivity with any contaminant can
not increase beyond 10-8 m/s.
REFERENCES
Brandl H. (1992) Mineral liners for hazardous
waste containment. Geotechnique,, 42, 57-65.
Lakshmikantha H. and Sivapullaiah P.V. (2005)
Relative performance of lime stabilized
amended clay liners in different pore fluids.
Geotechnical and Geological Engineering, 24
(5), 1425-48.
Mitchell J.R. and Jaber M. (1990) Factors
Controlling the long-term Properties of Clay
Liners. Geotechnical Special Publication No. 25,
ASCE, 884-105.
Mitchell J.K. and Madsen F.T. (1987) Chemical
Effects on Clay Hydraulic Conductivity.
Geotechnical Practice for Waste Disposal,
ASCE, Geotechnical Special Publication No. 13,
87-116.
Sivapullaiah P. V. (2009) Clay Liner Systems for
waste disposal facilities. Chapter 6 in “Solid
Waste Management and Engineered Landfills”,
Ed. G V. Rao and R. S. Sasidhar, Sai Master
Geoenvironmental Services Pvt. Ltd. (Sages). P
109-115. Hyderabad, 2009.
Rowe K.R., Quigley R.M. and Booker R. B. (1995)
Clayey Barrier Systems for Waste Disposal
Facilities. E & FN Spon, London.
Yang N. and Barbour S.L. (1992) The Impact of
Soil Structure and Confining Stress on the
Hydraulic Conductivity of Clays in Brine
Environments. Canadian Geotechnical Journal,
29, 730-739.
 Effect of Lime Treatment on Pore Size and Permeability
of MSW Liners
Z. Metelková
Charles University, Faculty of Science, Institute of Hydrogeology, Engineering Geology and
Applied Geophysics, Prague, Czech Republic (metelkov@natur.cuni.cz)
J. Boháþ
Charles University, Faculty of Science, Institute of Hydrogeology, Engineering Geology and
Applied Geophysics, Prague, Czech Republic (bohac@natur.cuni.cz)
I. SedláĜová
Institute of Chemical Technology, Prague, Czech Republic (sedlaroi@vscht.cz)

ABSTRACT The effect of adding CaO to facilitate compaction in producing a mineral liner is
presented. Adding CaO modified positively the density-moisture relation of the soil used typically
in the liners, it however increased its hydraulic conductivity. Mercury intrusion porosimetry and
nitrogen adsorption are used to characterise the pore space in the treated soil and to estimate the
gradual development of new minerals in the soil. The hydraulic conductivity seemed to be
influenced by dissolution-precipitation processes rather than a grow of new minerals.

INTRODUCTION Petry, 1992) hypothesised that lime-stabilised

In Czechia the geology barrier of waste
deposits must typically be improved by a clay
liner (0.5 to 1.0 metres thick) of the hydraulic
conductivity k  1×10-9ms-1. For many landfills
loess (brick-earth; silty to sandy clay, CL) has
been used in building the clay liners, primarily
due to its relatively low plasticity, good
workability and low cost. However when the
in-situ water content is too high the contractors
tend to use lime treatment to facilitate
compaction, which on the other hand influences
the pore distribution and hydraulic properties.
The findings about the influence of lime on
permeability are not conclusive. Fossberg
(1965) reported that admixture of 10% by dry
mass of hydrated lime Ca(OH)2 resulted in a
remarkable decrease of saturated hydraulic
conductivity k of a high plasticity clay by
almost one order of magnitude after 7 days of
curing, and by more than 2 orders of magnitude
after 28 days of curing. The difference was
explained by reduction in pore size due to
stabilisation and by formation of gels and
clogging up the pores.
Townsend & Klym (1966, in McCallister &
soils would increase the pore volume due to
flocculation. Their tests showed an increase in
hydraulic conductivity after adding lime to
heavy clays.
An increase in k by up to two orders of
magnitudes after adding lime to four types of
soils ranging from clay to silt was reported by
Brandl (1981). He concluded that the more
cohesive and more reactive the untreated soil
was, the more its hydraulic conductivity k
increased due to immediate flocculation. The
maximum increase in k for inactive silt was at
the admixture of 1% of slaked lime. However
with time the fines were “...embedded within a
gelatinous ... mass,...(and)... hardening products
of the binder grow(ed) into the voids of the soil
aggregates changing the void structure”. Due to
the decrease of pores and forming a film of
bound water k decreased with increasing curing
time.
El Rawi & Awad (1981) carried out falling
head permeability tests on compacted
specimens. At the optimum water content of the
tested silty-sandy clay 3% of hydrated lime
increased k by about one order of magnitude,

and there was practically no other change in k
adding 5% or 7% of lime. The authors
explained the increase in k by flocculation and
by an increase of the particle size and thus the
pore size due to formation of gel around the
original particles.
Osinubi (1998) tested lateritic soil. With
increasing lime content hydraulic conductivity
increased until 4% of hydrated lime and
decreased with the higher content of lime (4%
to 8%). Curing had a similar effect: k increased
with curing period 5 and 15 days, while after 30
days of curing there was a drop in k. The
decrease with lime content was explained by
exceeding the “lime fixation point”, i.e. by
“excess” lime being available for pozzolanic
reactions. The decrease with curing time,
reported also for low lime content of 2%
(below the “lime fixation point” was said to be
due to gel formation. McCallister & Petry
(1992) showed that hydraulic conductivity of
three tested high plasticity clays was
substantially increased by adding 1% to 3%
lime, and that with a further increase of lime
(up to 10%) k dropped but remained higher
than with the untreated soils. The maximum
increase by about two orders of magnitute was
achieved at the “lime modification optimum”
LMO. Until LMO is reached, the lime addition
results in flocculation and increase of hydraulic
conductivity k. The pozzolanic reactions do not
occur on a large scale due to relative low pH in
the soil. Above LMO pozzolanic reactions take
place and k decreases.
Locat et. al (1996) reported maximum
flocculation of silty clay at 3% of lime, where
there was the maximum of hydraulic conducti-
vity k. A further addition of lime decreased
k and at 10% of lime k was about one order
of magnitude lower than that of untreated soil
In the case of a saprolitic soil de Brito Gãlvao
et al. (2004) measured an increase of k at 2% of
added Ca(OH)2 followed by a decrease in k
with lime content between 2 and 8%. With
lateritic soil k decreased monotonously on
adding lime. The difference in the response to
lime treatment was explained by aggregation of
Metelková, Bohá č and Sedlářová 391
the saprolitic soil on adding 2% of lime, while
filling the fine pores with further lime
admixture. Lateritic soil was supposed to
exhibit much less aggregation as lime was
found to reduce permeability.
The amount of CaO used typically to facilitate
compaction is 1% or 2%. It is likely that “lime
modification optimum” or “lime fixation point”
is not achieved at this percentage, and therefore
hydraulic conductivity of the treated clayey soil
would increase, without the positive effect of
development of secondary minerals.
Three categories of soil pores are recognized:
micropores with the width below 0.002 µm,
mesopores from 0.05 µm to 0.002 µm and
macropores exceeding 0.05 µm. The
distribution of the macropores can be
determined by mercury intrusion porosimetry
(MIP) and the distribution of the mesopores by
MIP or by nitrogen adsorption. Micropores
could be determined by nitrogen or carbon
dioxide adsorption. Russo et. al (2007)
investigated the influence of curing time and
moisture on the pore distribution in the
compacted mixture of alluvial silty soil with
3% of CaO. Porosity increased due to adding
CaO and decreased due to the curing time.
Addition of CaO produced flocculation of the
clay minerals and increased porosity.
Bin et al. (2007) studied the influence of 8 and
10% of lime on the structure of aggregates in
expansive soil. Nitrogen adsorption showed
that addition of CaO lead to an increase in the
amount of mesopores and macropores. The
increase was related to a reduction in the
amount of the micropores.
Le Runigo et al. (2008) investigated
permeability and pore distribution in a lime
treated silt. They observed a decrease of the
amount of pores between 0.1 – 1 ȝm (3% of
CaO) with the increasing curing time due to
formation of carbonates and cementitious
compounds. A smaller amount of cementitious
compounds was determined also at 1% CaO,
without significant modificatin of the pore
structure. A decrease of available lime during
curing time was shown.
392 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
FIELD TRIAL COMPACTION
During an extension of a waste landfill in the
central Bohemia, a local clayey silt (Horky
Loess, liquid limit 35%, plasticity limit 18%)
was used in the stages of the landfill
construction. The grading curve of the soil is
summarised in Fig.1. Main constituents of the
soil are 41.9% of quartz, 30.1% of illite, 16.8 %
of calcite, 5.9% of montmorillonite and 2.6 %
of kaolinite. In 2006 the soil was too wet (to
meet the criteria of stiffness and dry density)
and the contractor decided to use lime treatment
(active CaO 91.34%, CO2 3.08%) to facilitate
compaction.
Fig.1 Grading Curve of Horky Loess
During the trial compaction the loess was used
first without any additives, then 1% of CaO was
added. The trial fields were compacted in two
drifts. Each of them were 0.3 m thick. The
compaction effort was the same for all drifts.
Hydraulic conductivity and degree of
compaction (ȡd/ȡdmaxPS; PS means Proctor
Standard; ȡd means dry density) were
determined using undisturbed samples. Degree
of compaction exceeded 97% for all samples.
However the values of hydraulic conductivity
of some samples lay above 1×10-9 m/s not
satisfying the criterion stipulated by the project.
It was decided that the Horky loess could not be
used with lime in the mineral liner and the
geosynthetic clay liner was used instead.
LABORATORY STUDY
A laboratory study was carried out to explain
the influence of lime treatment on the
compactibility and hydraulic conductivity of the
particular soil. The same lime and the soil from
the same pit (Horky Loess) were used as in the
field trial compaction. The grains larger than 2
mm were removed for the laboratory testing.
The water content–density relationships (PS)
were determined at CaO contents of 0%, 1%,
2%, 4% and 8%. The specimens with 1% of
CaO after 1, 7, 14 and 28 days of curing were
studied. The macropore distribution was
determined by mercury porosimetry, the
mesopore distribution was determined by
mercury porosimetry and nitrogen adsorption
and hydraulic conductivity was measured in the
triaxial cells and flexible wall permeameters.
Preparation of the specimens and test
procedures
For determining compactibility the required
amount of water was added to 600g of dry soil
and allowed to homogenise for 24 hours. Then
the required amount of CaO (1 - 8%) with
respect to the dry density was carefully mixed
with the soil by a blender. After 2 hours of
curing a sample of the diameter of 100 mm and
height of 38 mm was compacted by the Proctor
Standard compaction effort relevant for the first
layer only.
For permeability testing the specimens (height
35 mm, diameter 38 mm) were cut from a
sample with 1% of CaO (ȡd =1569 kg/m3, water
content w = 23.1%). The specimens were left to
cure for 1, 7 or 14 days under constant moisture
and temperature. The sample was compacted by
PS effort relevant for the first layer (the final
height of cca 38 mm).
After curing constant head permeability tests
were carried out in the triaxial cell. The back
pressure of 300 kPa was applied and the
hydraulic gradients were of about 30. The tests
were carried in an air-conditioned laboratory
(22ºC) and the measured hydraulic
conductivities were corrected with respect to
10ºC.
For determination of the macropore and
mesopore distribution using MIP four samples
with different CaO contents (1, 2, 4, 8%) were
compacted by the PS effort relevant to the first
layer at the water content close to optimum
(achieved ȡd were 1618, 1527, 1501, and 1478

kg/m3). For the sample with 1% CaO the
mesopore distribution was also determined by
nitrogen adsorption (BJH method). For the
analyses, small specimens were taken from the
centre of the samples and the pore distribution
was determined after drying in a laboratory
freeze dryer.
Test results
Density – moisture relations
The compaction curves of the samples with 0%,
1%, 2%, 4% and 8% of CaO are shown in Fig.2.
The maximum dry density decreases with
increasing lime content. The drop of ȡd between
0% and 1% of CaO is greater than 20 to 30
kg/m3, which are the typical values reported in
practice.
Fig.2 Dry Density – Water Content Relation
(Laboratory Study)
Mercury intrusion porosimetry (MIP)
Results of MIP of a sample with 1% of CaO are
shown in Fig.3. The soil was mixed with lime
by hand and compacted in the standard way in
three layers (PS). Two specimens (a,b) were
taken from each top, middle and bottom thirds
of the sample after 1 day of curing. Both top
third (TT) specimens exhibited a higher
cumulative volume of pores with the radius of
2-62 ȝm than the middle (MT) and bottom (BT)
thirds. The two specimens from the BT differ
slightly in the pores of the radius of 0.8-2 ȝm,
while the pair from the MT exhibits a similar
cumulative volume. Figure 3 showed that the
hand mixing was not suitable for preparing the
sample, and a blender was used for further
samples instead.
Metelková, Bohá č and Sedlářová 393
Fig.3 Cumulative Pore Volume - Pore Size Radius
of Two Treated Specimens from TT, MT and BT,
Respectively
The development of the pore space with time
(in 1, 7, 14 and 28 days of curing) for different
CaO contents (1, 2, 4 and 8% of CaO) is shown
in Fig. 4. In the legend the porosities computed
from MIP are included.
The porosity increases with the amount of CaO.
For the specimens with 1 and 2% CaO no trend
in the change of porosity with the curing time
was found. However the porosity decreases
consistently with the curing time at 4 and 8% of
CaO. The changes in porosity are rather small
and close to the accuracy of the method.
Nevertheless the trend seems to indicate the
development of new minerals in the pores.
The most frequent radius (rpeak) of the pores
increases with the amount of CaO. The number
of pores with rpeak increases with the amount of
CaO. For 1% of CaO the number of the most
frequent pore size seems to increase with the
curing time. In the specimens with 4 and 8% of
CaO it is exactly the opposite. 2% of CaO
could be considered the lime modification
optimum (LMO). Above LMO the pozzolanic
reactions and the newly created minerals result
in the rpeak reduction and in the decrease of
porosity with time.
At and below LMO the lime admixture was not
sufficient for a pronounced formation of new
minerals. Some smaller amount of new
minerals might be formed after 14 days of
curing. However the measurements taken on
different specimens could not produce a clearer
data.
394 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

BJH method
The pore volume of the sample with 1% of CaO
was also determined by nitrogen adsorption
using BJH method. The method is used for
computing the cumulative volumes from the
desorption isotherm (Barret et al., 1951). The
specimens were taken from the same part of the
sample as for MIP and they were dried in the
laboratory freeze dryer within 1 day. Figure 5
shows the cumulative volume curves of
a)
mesopores after different curing time.

b) Fig.5 BJH Cumulative Pore Volume Curves of the

Sample with 1% of CaO after 1, 7, 14 and 28 Curing
Days

The change of the cumulative volume curves

with the curing time shows a decrease of the
amount of the pores r > 0.04 ȝm. The shape of
the curve changes after 28 days. The number of
pores between 0.05 – 0.008 ȝm increases more
slowly than in the other specimens, which is
attributed to the growth of new minerals.
c) The average pore radius computed from the
BJH desorption cumulative pore volume at the
pore radius interval 0.002- 0.03 ȝm decreases
from 5.53 to 3.12 nm with the increasing curing
time.

d)
Fig.4 The Pore Size Distribution of the Specimens
with a) 1%, b) 2%, c) 4% and d) 8% of CaO after 1,
7, 14 and 28 Curing Days
Permeability test
The specimens with 1% of CaO were placed in
the triaxial cell after 1, 7 and 14 curing days. In
Fig.6 there is the dependence of the hydraulic
conductivity on the leaching time. The
hydraulic conductivity decreases with the
increasing leaching time and the initial rate of
the decrease is higher after 7 curing days than
after 14 days. However the change in hydraulic
conductivity is not believed related to the

development of new minerals, which are
supposed to grow only in the pores with r <0.04
ȝm. Water mainly circulates in the bigger pores.
The decrease of k is hypothesised to be due to
the dissolution and precipitation. The new
minerals grow in the specimen cured for 14
days and the amount of available lime
decreases. The dissolution – precipitation
processes are not intensive.
Fig.6 Hydraulic Conductivity – Leaching Time –
Laboratory Prepared Samples with 1% of CaO
CONCLUSION
The admixture of lime has decreased the dry
density and increased the optimum water
content of the tested clayey soil. It confirms the
positive influence of lime admixture on
compactibility known in the practice.
Porosity increases with the amount of lime due
to stronger flocculation. The number of the
most frequent pore radius rpeak increases with
the amount of lime. Above 2% of CaO (LMO)
the number of rpeak and rpeak decreases with
curing time due to newly created minerals. At
and below LMO the new minerals were not
demonstrated by the mercury intrusion
porosimetry.
Nitrogen adsorption showed a reduction of the
average pore radius with increasing curing time
at the interval 0.002- 0.03 ȝm.
The ongoing permeability testing indicates that
the hydraulic conductivity decreases with the
leaching time and increases with the curing
time.
ACKNOWLEDGEMENT
The financial support by the research grants
GAUK 62008 and MSM0021620855 is
gratefully acknowledged.
Metelková, Bohá č and Sedlářová 395
REFERENCE
Barret E.P., Joyner L.G. and Halenda P.P.
(1951) The determination of pore volume
and area distributions in porous substances.
I. Computations from nitrogen isotherms, J.
Am. Chem. Soc., 73, 373-380.
Bin S., Zhibin L., Yi C. and Xiaoping Z. (2007)
Micropore structure of aggregates in treated
soils, J. Mater. Civil. Eng., 19(1), 99-104.
Brandl H. (1981) Alteration of soil parameters
by stabilization with lime, Proc. 10th
ICSMFE, 3, 587 – 594.
El Rawi N.M. and Awad A.A. (1981)
Permeability of lime stabilized soils, J.
Transp. Eng. ASCE, 107(1), 25-35.
Fossberg P.E. (1965) Some fundamental
engineering properties of a lime-stabilized
clay, Proc. 6th ICSMFE, 1, 221-225.
de Brito Galvão T.C., Elsharief A. and Simões
G.F. (2004) Effects of lime on permeability
and compressibility of two tropical residual
soils, J. Environ. Eng. ASCE, 130(8), 881-
885.
Le Runigo B., Cuisinier O., Cui Y.J., Ferber V.
and Deneele D. (2008) Impact of initial state
on the fabric and permeability of a lime-
treated silt under long-term leaching, Can.
Geotech. J., 46, 1243-1257.
Locat J., Tremblay H. and Leroueil S. (1996)
Mechanical and hydraulic behaviour of a
soft inorganic clay treated with lime, Can.
Geotech. J., 33, 654-669.
McCallister L.D. and Petry T.M. (1992) Leach
tests on lime treated clays, Geotech. Test. J.,
15(2), 106 -114.
Osinubi K.J. (1998) Permeability of lime-
treated lateritic soil, J. Transp. Eng. ASCE,
124(5), 465-46.
Russo G., Dal Vecchio S. and Mascalo G.
(2007) Microstructure of a lime stabilised
compacted silt, Experim. Unsat. Soil Mech.,
Springer Proc. In Phys., 112, 49 – 56.
 Effect of Salt Solutions Found in Landfill Leachate on some Properties
of Soil-Bentonite Mixtures as Liners
N. Shariatmadari
Department of Civil Engineering, IUST, Tehran, Iran (shariatmadari@iust.ac.ir)
M. Salami
Department of Civil Engineering, IUST, Tehran, Iran (mzsalami@gmail.com)

ABSTRACT Landfill liners play an important role in keeping the leachate from migrating and polluting
groundwater. Leachate generated from landfills includes some salts. In this study, effect of NaCl and CaCl2
solutions on some properties of mixtures of a local clay specimen and bentonite has been investigated.
Laboratory tests including 1D consolidation, compaction tests, liquid limit and free swell tests were carried out.
Results indicated that increasing the salt concentration led to a decrease in liquid limit, free swell and an
increase in hydraulic conductivity of the soil mixtures. Also, divalent cations increased the hydraulic
conductivity to a greater extent.

INTRODUCTION during hydration (Grim,1968, Mitchell, 1993).

The growth of population increased the production
of waste to a great extent. Therefore waste
disposal has become a huge environmental issue.
A good landfill liner material is one with low
hydraulic conductivity under various
environmental conditions. Its role is to prevent
leachate generated from the waste from infiltrating
to the environment thus minimizing the risk of
groundwater contamination. Clays are suitable
material for landfill liners with their high
adsorption capacities and low hydraulic
conductivity. Encapsulation systems vary, like the
environments which they are built in, and the
components of encapsulation systems are as
diverse and complex as the waste content of the
landfill (Bergaya et al., 2006)
In industrial and domestic waste landfills, natural
bentonite are considered to be used (Ebina et al.,
2004). A mixture of locally available soil and
bentonite may be employed as landfill liners.
Bentonite consists of different minerals; The
primary mineral in bentonite is montmorillonite. It
is a 2:1 layer consisting of an octahedral sheet
located between two silica sheets. The large
specific surface of montmorillonite, also its large
net negative charge result in adsorption of a large
number of hydrated cations as well as adsorption
of water molecules, and interlayer separation
Due to the change of leachate content generated
from waste, liners are exposed to various chemical
and biological processes. Thus liners should be
able to resist the passage of a wide variety of pore
fluids. As a results, distilled or tap water is not a
complete representative of field conditions.
Consistency limits state some very basic
mechanical data about the soil (Arasan &
Yetimoglu, 2008), therefore their evaluation can
be helpful.
The aim of this paper is to observe the effect of
salt solutions found in landfill leachate in various
concentrations on the performance of soil-
bentonite mixtures as landfill liners. In this study,
data obtained from experiments including
compaction tests, liquid limits, free swell tests and
hydraulic conductivity using 1-D consolidation
tests on soil-bentonite mixtures are presented.
BACKGROUND
Numerous studies have been made on the effect of
different fluids on the properties of soils. Some of
the more recent researches are stated here.
Gleason et al. (1997) showed that water
adsorption behavior in two different kinds of
bentonite, namely Na-bentonite and Ca-bentonite,
when permeated with CaCl2 in high concentration,
 Shariatmadari and Salami 397

did not differ very much due to the high The grain size distribution of Varamin Clays
concentration of the electrolyte. Stern & shows that it has 93% clay-silt content and 7%
Shackleford (1998) showed that permeability of sand content. The results of XRF analysis of
sand-bentonite mixtures increased when using Varamin Clay are shown in Table 2.
0.5M CaCl2 solutions by up to two orders of
magnitude. Another investigation conducted by The pore fluids used were DI water and two salt
Alawaji (1999), showed the effect of NaNO3 and solutions namely NaCl and CaCl2 in various
Ca(NO3)2 in four concentration of 0.1, 0.5, 1 and concentrations.
4 N on mixtures of sand and 20% bentonite. The
results indicated that these salts decreased TABLE 2 XRF Analysis of Varamin Clay
Atterberg limits of the mixtures to a great extent. Component Percent
Shackelford et al. (2000), Jo et al. (2001) and SiO2 1.54
Shan & Lai (2002) stated that an increase in free Fe2O3 6.90
swell of bentonite causes the hydraulic Fe3O4 1.40
conductivity to decrease. Singh & Prasad (2007) Al2O3 15.2
investigated the effect of aluminum hydroxide and MgO 3.20
acetic acid-which are commonly found in landfill CaO 1.15
leachate-in controlled conditions in the laboratory, Na2O 1.03
K2O 2.80
on bentonite. Results showed that hydraulic
SO3 0.17
conductivity decreased by 17% and 12% Cl- 0.03
respectively. Yilmaz et al. (2008) stated that by CaCO3 2.80
increasing salt concentrations, the hydraulic LOIa 11.10
conductivity of CH clays increases, and that of a
Loss of Ignition
CL clays decreases. In this study the effect of
NaCl and CaCl2 solutions on some properties of EXPERIMENTAL PROCEDURE
soil-bentonite mixtures is investigated.
Free Swell Test
MATERIALS The free swell of bentonite was determined as per
The materials used in this study were clay from ASTM D5890. The bentonite was ground to 100%
the city of Varamin located near north of Iran passing through a #200 US standard mesh.
(referred as Varamin Clay hereafter), and Although some modifications were made
commercially available bentonite supplied by Iran according to Roberts & Shimaoka (2008): In order
Barit Falat Company. Some of the properties of to lessen the effect of friction between the
these soils are given in Table 1. swelling clay and walls of the cylinder, a few
changes in proportions were made. First samples
TABLE 1 Properties of Used Soils were dried at 1050C; then approximately 40 g of
Varamin dried clay were added to 1800 ml of salt solution
Properties Bentonite
Clay in a 2000 ml graduated cylinder. The bentonite
Specific Gravity 2.67 2.53 was added in increments of 2 g over a period of
approximately 10 minutes; the samples were
Liquid Limit (%) 34.2 199.4 settled for 3 hours until all 40 g of the bentonite
Plasticity Index (%) 13.7 157.9 was added. After adding the final increment of the
OMC(%), 14, sample, the cylinder walls were washed with the
NDa salt solution and brought to the 2000 ml volume
MDD(gr/cm3) 1.86
a
Not Determined mark. Samples were set undisturbed for 2 months
in which after the changes in swelling and pH of
the samples were recorded.
398 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Compaction Test
To investigate the compaction characteristics, the
standard Proctor compaction test was done as per
ASTM D698 on Varamin Clay with no bentonite
using DI water, and mixtures of bentonite and soil
using DI water, NaCl and CaCl2 in concentrations
of 1 N.
Liquid Limit Test
The liquid limit was done according to British
standard BS1377:Part2:1990. The samples were
prepared in two proportions of 100:10 and 100:20
(100 is the total weight and 10 and 20 are the
weight of the bentonite). The pore fluids used
were DI water and Sodium Chloride and Calcium
Chloride solutions with various concentrations
namely 0.01, 0.1, 1 and 2 N. The soil mixtures
were exposed to the salt solutions then sealed in
plastic bags to remain for at least 24 hours. Then
after, liquid limit tests were carried out on the
samples.
Consolidation Test
Consolidation test were carried out following
procedures described in ASTM D2435 in order to
determine the hydraulic conductivity and
compressibility of the samples. One-dimensional
consolidation test apparatus -in the absence of
triaxial permeameter- can be used to assess
hydraulic conductivity. The consolidation test on
the selected soils was conducted in the standard
consolidometers using rings 50 mm in diameter
and 20 mm thick. Pore fluids were DI water and
CaCl2 and NaCl solutions in two concentrations of
0.1N and 1N. The soil samples were hand
remoulded in the consolidation ring at initial
moisture content pertaining to their liquid limit,
taking caution not to trap any air in the specimens.
Both sides of the sample were covered with filter
papers and porous stones, placed in the
consolidation cell and in the loading frame. The
porous stones were used in damp condition to
prevent absorption of water from the sample. The
inside of the consolidation ring was lubricated
with a thin layer of silicon grease to decrease
friction between the ring and the soil specimen.
The liquid surrounding the consolidation ring had
the same composition as the saturating fluid. In
order to minimize evaporation the consolidation
cell was enclosed within a plastic bag and was
allowed to equilibrate for at least 24 hours before
executing the test.
The samples were initially loaded with a stress of
25 kPa, increasing by an increment ratio of 1 to a
maximum stress of 400 kPa. Changes in the
deformation gauge were recorded manually.
The hydraulic conductivity of the soil specimens
was calculated by the following equation:
k cv mv J w
Where cv is the coefficient of consolidation, mv is
the coefficient of volume change. Also Ȗw is the
unit weight of the pore fluid.
RESULTS AND DISCUSSION
Free Swell Test
Figure 1 shows the results of free swell tests
carried out on bentonite for NaCl and CaCl2 in
two concentrations of 0.1N and 1 N. It can be seen
that the most swelling occurred for bentonite in DI
water. In comparison between NaCl and CaCl2 at
the same salt concentration, more swelling
occurred for NaCl, which is particularly
significant at 0.1N. Increasing the salt
concentration, led to a decrease in swelling
volume. Increasing the cations in the bulk solution
results in the gradient of free energy thus making
water leave the interlayer region (Mishra et al.,
2005). The pH of the specimens were in the
neutral range throughout the experiment, and
therefore did not affect the swelling volume.
600
DI water
500
swell volume (ml/40 gr)
400
NaCl, 0.1N
300
200
NaCl, 1N
CaCl2, 0.1N
CaCl2, 1N
100
0
Salt solutions
Fig.1 Free Swelling Response of Bentonite to
Different Salt Solutions
 Shariatmadari and Salami 399

2000 2000
DI water- 0 % Bentonite NaCl-10% Bentonite
1950 DI water - 10 % Bentonite
(a) 1950 NaCl-20% Bentonite (b)
DI water -20% Bentonite 1900

Dry Density (kg/m 3)

1900 CaCl2-10% Bentonite
Dry Density (kg/m 3)

CaCl2- 20% Bentonite

1850 1850
1800 1800
1750 1750
1700 1700
1650 1650
1600 1600
1550 1550
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Moisutre Content (%) Moisutre Content (%)

Fig.2 Results of Proctor tests using (a) DI water (b) Salt Solutions

Compaction Test the cation valence decreases the inter-particle

The compaction curves of the samples can be seen repulsion which results in particles moving more
in Fig.2. The curves show that by increasing the freely in lower water contents, therefore reducing
bentonite content in the mixtures, the optimum the liquid limit of the mixtures (Warkentin, 1961).
water content increases while the maximum dry
unit weight decreases. This behavior is predictable Consolidation Tests
due to the activity of bentonite. The absorbed Figures 4 and 5 show the variation of hydraulic
water surrounding the clay particles have volume conductivity with respect to void ratio for salt
that leads to an increase in the water content and solutions in two concentrations namely 0.1N and 1
decrease in the dry unit weight (Kaya et al.,2006). N. As expected, with decreasing void ratio,
hydraulic conductivity of the mixtures decreases.
Increasing the salt concentration of the pore fluid
(comparison between Fig.2(a) and Fig.2(b)) led to
a decrease in the optimum water content and an
increase in maximum dry weight. By increasing
the salt concentration of the pore fluid, the diffuse
double layer thickness decreases. Therefore using
the same amount of compaction energy, the
particles move closer to each other and the dry
density increases (Singh & Prasad, 2007). The
difference between the affect of NaCl and CaCl2
solutions is negligible.

Liquid Limit Test

Results of the liquid limit tests are shown in Fig.3.
From the graphs it is shown that, increasing the Fig.3 Liquid Limit of Soil Mixtures
bentonite content, increases the liquid limit of the
soil mixtures. Also when salt solutions are used as
pore fluids, the liquid limit decreases. The type of The reason is the decreased void space available
salt has a greater effect on the LL in the 100:20 for flow of the pore fluid and possibly the re-
mixtures, and for the 100:10 mixtures the liquid orientation of the particles perpendicular to the
limit with both NaCl and CaCl2 are similar. direction of the flow (Quigley & Thompson,
Furthermore with increasing salt concentration in 1966).
both 100:10 and 100:20 mixtures, the LL
decreases. Increasing the salt concentration and
400 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

1.00E-08 1.00E-08
Hydraulic Conductivity (m/s)

Hydraulic Conductivity (m/s)

CaCl2-0.1N-10% Bentonite NaCl-0.1N-10% Bentonite
CaCl2-0.1N-20% Bentonite NaCl-0.1N-20% Bentonite
CaCl2- 1N -10% Bentonite NaCl- 1N -10% Bentonite
1.00E-09 CaCl2-1N-20% Bentonite 1.00E-09 NaCl-1N-20% Bentonite
DI water- 10% Bentonite DI water- 10% Bentonite
DI water-20% Bentonite DI water-20% Bentonite
1.00E-10 1.00E-10

1.00E-11 1.00E-11

1.00E-12 1.00E-12
0.00 0.30 0.60 0.90 1.20 0.00 0.30 0.60 0.90 1.20
Void Ratio Void Ratio

Fig.4 Effect of NaCl solutions on hydraulic Fig.5 Effect of CaCl2 solutions on hydraulic
conductivity of different soil mixtures conductivity of different soil mixtures

Increasing the salt solution from 0 to 0.1 to 1 N SUMMERY AND CONCLUSIONS

led to an increase in hydraulic conductivity which
is seen in 100:20 mixtures; although in 100:10
mixtures this increase is not significant. In
comparison between CaCl2 and NaCl, it is seen
that hydraulic conductivity is increased to a
greater extent for CaCl2, probably because of less
swelling and higher flocculation of bentonite in
the mixtures, which is more evident in 100:20
mixtures (Mishra et al., 2005). These trends are
consistent with diffuse double layer theory. Figure
6 shows the relationship between compression
index, Cc, and salt concentration.
0.3
0.25
Compression Index (Cc)
NaCl-20% Bentonite
CaCl2-20% Bentonite
A number of tests have been conducted on
mixtures of a local clay specimen and bentonite.
From these experiments it was seen that by
increasing the salt concentration and cation
valance, the swelling volume of bentonite
decreases.
The liquid limit of the mixtures decreases by
adding salt solutions which is more evident in
100:20 mixtures.
Also, it is shown that hydraulic conductivity of the
soil mixtures increases when salt solutions are
used as pore fluids, and the cation valance
attributes to a greater increase of this factor.
These trends are consistent with diffuse double
layer theory.

NaCl-10% Bentonite

0.2 CaCl2-10% Bentonite

REFERENCES
0.15

0.1
0.05
0 0.2
Concentration (N)
Fig.6 Relationship Between Salt Concentrations and
CC
It is evident that Cc is affected by the type of
cation and salt concentration. It is seen that by
increasing the salt concentration, the Cc value
decreases. Also, soil mixtures with a higher
bentonite content (20% bentonite) has a greater Cc
value.
Alawaji H.A. (1999) Swell and Compressibility
Characteristics of Sand-Bentonite Mixtures
0.4 0.6 0.8 1
Inundated with Liquids, Applied Clay Sci., 15,
411-430.
Arasan S. and Yatimoglu T. (2008) Effect of
Inorganic Salt Solutions on the Consistency
Limits of Two Clays, Turkish Jl. of Eng. & Env.
Sci., 32, 107-115.
Bergaya F., Theng B.K.G. and Lagaly G. (2006)
Handbook of Clay Sci., Developments in Clay
Sci., 1, Elsevier Ltd.
Ebina T., Minja R.J.A., Nagase T., Onodera Y. and
Chatterjee A. (2004) Correlation of Hydraulic

Conductivity of Clay-Sand Compacted
Specimens with Clay Properties, Applied Clay
Sci., 26, 3-12.
Gleason M., Daniel D.E. and Eydholt G.R. (1997)
Calcium and Sodium Bentonite for Hydraulic
Containment Applications, Jl. of Geotech. &
Geoenv. Eng. ASCE, 123:5, 438-445.
Grim R.E. (1968) Clay Mineralogy, 2nd ed.
McGraw-Hill, New York.
Jo H.Y, Katsumi T., Benson C.H. and Edil T.B.
(2001) Hydraulic Conductivity and Swelling of
Nonprehydrated GCLs Permeated with Single-
Species Salt Solutions, Jl. of Geotech. &
Geoenv. Eng. ASCE, 127:7, 557-567.
Kaya A., Durakan S., Oren A.H. and Yukselen Y.
(2006) Determining the Engineering Properties
of Bentonite-Zeolite Mixtures, Turkish
Chamber of Civil Engineers, Digest, 17, 1075-
1088.
Mishra A.K., Ohtsubo M., Li L. and Higashi T.
(2005) Effect of Salt Concentrations on the
Permeability and Compressibility of Soil-
Bentonite Mixtures, Jl. of Fac. of Agr., 50:2,
837-849.
Mitchell J.K. (1993) Fundamentals of Soil
Behaviour, 2nd ed. Wiley, New York.
Quigley R.M. and Thompson C.D. (1966) The
Fabric of Anisotropically Consolidated
Sensitive Marine Clay, Can. Geotech. Jl., 3:2,
61-73.
Shariatmadari and Salami 401
Roberts A.A. and Shimaoka T. (2008) Analytical
Study on the Suitability of Using Bentonite
Coated Gravel as a Landfill Liner Material,
Waste Mng., 28, 2635-2644.
Shackelford C.D., Benson C.H., Katsumi T., Edil,
T.B. and Lin L. (2000) Evaluation of the
Hydraulic Conductivity of GCLs Permeated
with Non-standard Liquids, Geotex. &
Geomembr., 18:2-4, 133-161.
Shan H-Y, and Lai Y.J. (2002) Effect of Hydrating
Liquid on the Hydraulic Properties of
Geosynthetic Clay Liners, Geotext. &
Geomembr., 20:1, 19-38.
Singh S. and Prasad A. (2007) Effects of
Chemicals on Compacted Clay Liner, The
Electronic Jl. of Geotech. Eng. (EJGE), 12,
Bundle D.
Stern R.T. and Shackelford C.D. (1998)
Permeation of Sand-Processed Clay Mixtures
with Calcium Chloride Solution, Jl. of Geotech.
& Geoenv. Eng. ASCE, 124:3, 231-241.
Warkentin B.P. (1961) Interpretation of the Upper
Plastic Limit of Clays, Nature, 190, 287-288.
Yilmaz G., Yetimoglu T. and Arasan S. (2008)
Hydraulic Conductivity of Compacted Clay
Liners Permeated with Inorganic Salt Solutions,
Waste Mng. & Research, 26:5 ,464-473.
 Effect of Glassgrid Geocell Inclusion on Flexural Behavior of Cover
Soil for Low Level Radioactive Waste Disposal Facility
A. Rawat
Indian Institute of Technology Bombay, Mumbai, India (abhishek_rawat786@yahoo.co.in)
R. R. Rakesh
Bhaba Atomic Research Centre, Mumbai, India (rakeshraviranjan@yahoo.co.in)
J. N. Mandal
Indian Institute of Technology Bombay, Mumbai, India (cejnm@civil.iitb.ac.in)

ABSTRACT Behavior of cover system under the flexural loading is one of the deciding factors for
satisfactory performance of any landfill project. A beam testing program is carried out to evaluate the
effect of inclusion of glassgrid geocell on the flexural strength of cover soil. The size of beam was selected
as 100×100×300 mm. Beams were casted with soil, collected from one of the operating near surface
radioactive waste disposal facility site in India. Two different opening size geocell 50u50u100 mm and
50u60u100 mm were used as reinforcement. The three point loading system for flexural loading system is
used for the test with the strain rate of 1mm/min. The results were obtained in form of load deflection
curve. Crack length and location was recorded in each test. Test results show the improvement in the
flexural rigidity of unreinforced soil beams due to the inclusion of glassgrid as reinforcement. Beam tests
conditions were modeled with the Finite Element Analysis software PLAXIS V8 version. The values of
maximum central deflection of beam for different conditions of test obtained in PLAXIS V8 version were
compared with the experimental results and matched reasonably.

INTRODUCTION protect the ingress of rainwater from the top.

Low and Intermediate Level radioactive Waste (L
& ILW) are generated at various stages of nuclear
fuel cycle, which includes the mining and milling
of uranium ore, fuel fabrication, reactor operation
and spent fuel reprocessing. Besides these sources,
radioactive wastes are produced as a result of the
ever-increasing use of radioisotopes in medicine,
industry and agriculture. After proper treatment
and conditioning of these wastes the
solid/solidified waste is disposed in the various
disposal modules viz., Earth Trench (ET)/Stone
Lined Trench (SLT), Reinforced Concrete Trench
(RCT) and Tile Holes (TH) of Near Surface
Disposal Facility (NSDF) (Raj et al. 2006). After
operation of these disposal modules it is planned
to have a closure of these disposal modules to
Multilayer cap may be a feasible option for this.
As radioactive waste disposal and its management
is concerned for a very long duration (few
hundreds of year) a program has been initiated to
study the design of multilayer cap and its behavior
for a sufficient long time period. Some of the
initial work done in this field have been compiled
and published by International Atomic Energy
Agency (IAEA) in IAEA TECDOC-1260.
Considerable research has also been carried out to
assess the flexural and tensile strength of soil, clay
liners reinforced with different types of
geosynthetics by Mandal 1988, Indraratna et al.
1996, Kaniraj and Gayathri 2006, Wang 2007.
However, the applications of these researches are
limited to municipal or other conventional waste

management. In the present study, experimental
results on flexural strength of soil beam have been
compared with the mathematical modeling
performed by the PLAXIS V8.
LABORARTARY INVESTIGATION
Material and Methods
The soil was collected from one of the Near
Surface Radioactive Waste Disposal Facility
(NSDF) operated in India. These soil samples
were collected from three different locations from
a depth of 1 meter from ground level and mixed
thoroughly to make a representative sample. These
samples were analyzed in the laboratory as per IS
specifications and the properties of the soil sample
is presented in Table 1.
TABLE 1 Geotechnical Properties of Soil
Property Value
Natural Water Content (%) 8.70
Specific Gravity 2.70
Atterburg Limits
(a) Liquid Limit (%) 48.82
(b) Plastic Limit (%) 38.53
(c) Shrinkage Limit (%) 24.60
(d) Plasticity Index (%) 10.29
Optimum moisture content (%) 20.50
Maximum dry density of soil (g/cc) 1.68
Permeability of soil (cm/sec) 3.79×10-5
Shear strength parameters of soil
Undrained cohesion (kPa) 14.50
Angle of internal friction (degree) 20
To study the flexural strength, beams were casted
from the collected soil. The size of beam were
taken as length = 300 mm, width = 100 mm and
depth = 100 mm to maintain the length to width
ratio as 3:1. All the beams were casted at
Optimum Moisture Content (OMC).
The flexural strength was determined on two
different types of beams, namely (i) unreinforced
beam, and (ii) beams reinforced with geocell.
Rawat, Rakesh and Mandal 403
Three point loading arrangement was used while
performing the tests. The experiments were
performed at constant strain rate of 1 mm/min.
The tests were carried out on universal tensile
strength machine with certain fabrications.
BEAM TESTING (UNREINFORCED CASE)
24 hours oven dried soil samples, sieved through
4.75 mm sieve were used to cast the beam at
OMC. Standard proctor compaction method was
used for the compaction. Total number of blows
was calculated on the basis of energy imparted by
the standard proctor test (594 kJ/m3). Beams were
compacted in three layers, so the number of blows
for the mould size of 300u100u100 mm and
volume of 30, 00,000 mm3 is obtained as 78 per
layer. Total 9 beams were tested in which three
beams were tested for unreinforced case. During
the test, three point loading system was used for
the flexure loading with the strain rate of
1mm/min. The complete arrangement of loading
system with the Universal Tensile Strength
Machine is shown in Fig. 1.
Load
cell Loading arm
Computer Monitor
Soil Beam Display Board
C.P.U.
Loading Pad
Fig. 1 Loading System Arrangement in Beam Test
(For colour figure, refer to CD)
The pictures during the test of unreinforced soil
beam are shown in Fig. 2. Load deflection
response of soil beam unreinforced case is
depicted in Fig. 3. The parameters related with the
flexural loading as moment of resistance, radius of
curvature and modulus of flexure related with the
unreinforced beams are presented in Table 2 and
3.
404 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

(a) Soil Beam Unreinforced Case (Before the Test) (b) After Test
(For colour figure, refer to CD) Fig. 4 Beam Test with Geocell having Opening Size
Loading arm 50x50x100 mm (For colour figure, refer to CD)

Results of the soil beam test reinforced with glass

Crack
grid geocell having opening size as 50x50x100
mm is shown in Table 2 and 3.

TABLE 2 Result of the Soil Beam Test Unreinforced

(b) After Test (Insight Shows the Crack Pattern) and Reinforced Case
Fig. 2 Beam Test of Cover Soil (Unreinforced Case)
(For colour figure, refer to CD) Test condition Load at Max. central
which beam deflection
fails (N) (mm)
Unreinforced case 150.67 0.30
50x50x100 mm 351.70 11.20
50x60x100 mm 300.20 10.10

TABLE 3 Parameters for Soil Beam Test Unreinforced

and Reinforced Case

Test condition Resisting Radius Modulus of

Fig. 3 Load-Deflection Curve for Unreinforced Soil moment of flexure
Beam
(kN-m) curvature

Reinforced with Geocell Unreinforced 0.0113 37.50 5.08 u 104

Geocell was self-made by glass grid having 50x50x100 mm 0.0263 1.01 3.11 u 103
different opening size as 50×50×100 mm and
50×60×100 mm. In this way the effect of opening 50x60x100 mm 0.0225 1.04 1.90u 103
size on confinement effect can also be observed.
Fig. 4 depicts the testing of soil beam reinforced The comparison of load-deflection curve for
with geocell having opening size 50×50×100 mm. geocell having different opening size is presented
in Fig. 5. The results indicate that the geocell
having opening size 50 u 50 u 100 mm provide a
better reinforcement compared to 50×60×100 mm.

(a) Before Test (For colour figure, refer to CD)

 Rawat, Rakesh and Mandal 405

400

350

300
Load in Newton

250

200

150

100 5 x 5 x 10 cm
(a) Mesh
50 5 x 6 x 10 cm

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Deflection in mm

Fig. 5 Load-deflection Curve for Soil Reinforced with

Geocell having Different Opening Size
(For colour figure, refer to CD)

FINITE ELEMENT ANALYSIS OF BEAM

TEST WITH PLAXIS V8
The FEM computer code PLAXIS V8 was used to
simulate the beam test experiment. The results of
the modeling study were compared in form of
maximum central deflection with the experimental
study. Table 4 shows the material properties.
(b) Deformed mesh (For colour figure, refer to CD)
Fig. 6 Generation of mesh in Finite Element
Analysis (FEM) and deformed mesh for soil beam
without reinforcement.

Table 4 Material properties (local soil) The vertical displacement of soil due to flexure
loading is shown in Fig. 7. The extreme vertical
Properties of Soil Value
displacement is obtained as 0.258 mm.
Unsaturated unit weight 1.653E-05
(N/mm3)
Permeability of soil (mm/day) 32.8
Cohesion (kPa) 14.5
Angle of internal friction 20
2
Young’s modulus (N/mm ) (Eref) 8.2
Poisson’s ratio 0.32 (a) Vertical displacement (0.258 mm)
(For colour figure, refer to CD)

Soil Beam without Reinforcement Fig. 7 Extreme vertical displacement for soil beam
Soil beam without reinforcement is modeled in without reinforcement
software Plaxis V8. Fig. 6 shows mesh and deform
mesh shape during analysis. Fig. 8 shows the geometric model of soil beam
reinforced with glass grid geocell having opening
size 50x50x100 mm.
406 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Distribution of vertical displacement for glass grid

geocell with opening size 50×50×100 mm
reinforced beam is shown in Fig. 10.

Fig. 8 Geometric Model of Soil Beam Reinforced with

Glass Grid Geocell with Opening Size 50×50×100 mm
(For colour figure, refer to CD)
The geogrid properties used in the modeling is
presented in Table 5. The values of applied load in
simulating the experiment are presented in Table
6.
TABLE 5 Geogrid Properties Modeled as Geocell
Fig. 10 Vertical Displacement (extreme Uy 9.86 mm)
for Glass Grid Geocell Reinforced Beam having
Opening Size 50×50×100 mm.
(For colour figure, refer to CD)
TABLE 7 Comparison in Between Experimental and
Finite Element Study

Properties Glass grid geocell Extreme Extreme

Material type Elastic Test vertical vertical Percentage
Tensile strength (kN/m) 60.97 description displacement displacement error
Interaction coefficient 0.45 in experiment in FEM (%)
study (mm) analysis (mm)

TABLE 6 Values of Applied Load 50x50x100 11.2 9.86 (+) 11.9

mm
Specification of Geocell Applied load
50x60x100 10.1 10.81 (-) 7.0
Opening size 50×50×100 mm 351.7 (N)
mm
Opening size 50×60×100 mm 300.2 (N)

The deformed mesh for soil beam reinforced with
geocell having opening size as 50×50×100 mm is
presented in Fig. 9.
Fig. 9 Deform Mesh Shape for Soil Beam Reinforced
with Glass Grid Geocell having Opening Size
50×50×100 mm. (For colour figure, refer to CD)
RESULTS AND DISCUSSION
Soil beam reinforced with glass grid geocell with
different opening size shows elastic behavior
(during the initial stage of loading up to deflection
0.4 to 0.8 mm). Geocell as a reinforcement enables
soil beam to sustain more load. The load
deflection curve also improves. The higher value
of central deflection (up to 25 % of the depth of
beam) is observed. The use of geocell as
reinforcement/confinement of the soil provides a
significant increase in flexural strength, and
enables the beams to sustain much larger
deformations. The reinforced soil beams
demonstrate a definite ductile-plastic (post-peak)
behavior following the pre-peak elasto-plastic
curve, whereas the unreinforced beams are

generally characterized by a post-peak brittle
failure mode. FEM analysis with PLAXIS V8
shows well agreement with experimental test
results.
REFERENCES:
Indraratna, B. and Lasek, G. (1996). “Laboratory
evaluation of the load-deflection behavior of
clay beams reinforced with galvanized wire
netting”, Geotextiles and Geomembranes, 14,
555-573.
International Atomic Energy Agency Document.
(1995). “Principles of radioactive waste
management”. Safety series, No. 111-F, 3-9.
International Atomic Energy Agency. November
1999.”Technical considerations in the design of
near surface disposal facilities for radioactive
waste”, IAEA-TECDOC-1256, 36-50.
Rawat, Rakesh and Mandal 407
Kaniraj, S.R. and Gayathri, V. (2006). “Behavior
of fibre-reinforced cement stabilized fly ashes”,
Journal of Testing and Evaluation, 34(4), 1-8.
Mandal, J. N. (1988). “Reinforced soil beam
subjected to bending”, The Indian Geotechnical
Journal, 18 (3), 279-288.
Raj, K., Prasad, K.K., Bansal, N.K. (2006).
“Radioactive waste management practices in
India”, Nuclear Engineering and Design. 236(7-
8), 914-930.
Wang, J. J., Zhu, J. G. and Zhang, H. (2007).
“Experimental study on fracture toughness and
tensile strength of clay”, Engineering Geology,
94, 65-75.
 Design Rationale for Construction and Monitoring of Unsaturated Soil Covers
at the Rocky Mountain Arsenal

L.O. Williams
US Environmental Protection Agency, Denver, USA (williams.laura@epa.gov)

J.G. Zornberg
The University of Texas at Austin, Austin, USA (zornberg@mail.utexas.edu)

S.F. Dwyer
Dwyer Engineering, Albuquerque, USA (dwyerengineering @yahoo.com)

D.L. Hoyt
Pacific Western Technologies, Denver, USA (dhoyt@pwt.com)

G.A. Hargreaves
US Environmental Protection Agency, Denver, USA (hargreaves.greg@epa.gov)

ABSTRACT: Unsaturated soil covers were designed for contaminated areas of the Rocky Mountain Ar-
senal, a high-profile hazardous waste facility located near Denver, Colorado, USA. The soil cover system
has three functions: control infiltration, control erosion, and prevent biota intrusion. These three functions
are achieved by means of unsaturated soil, capillary barrier material, and crushed concrete. This paper
provides an overview of the rationale used to design each of these components, the challenges presented
by the interdependent criteria, and the post-construction monitoring program.

INTRODUCTION to that of a US Resource Conservation and Recov-

The use of unsaturated soil covers to isolate ha-
zardous waste from the surrounding environment
is considered an alternative approach for waste
containment within the US regulatory framework.
The Rocky Mountain Arsenal (RMA) site, located
18.5 km northeast of Denver, Colorado, USA, is a
hazardous waste facility regulated under the US
Environmental Protection Agency’s (EPA) Super-
fund program. A key element of the cleanup re-
medy was to interrupt the exposure pathways by
(a) placing the most contaminated soil and struc-
ture demolition debris in two regulated landfills
constructed on-site; and (b) consolidating less-
contaminated soil below unsaturated soil covers in
six highly contaminated areas considered too risky
for excavation. These unsaturated soil covers,
which span over 183.3 hectares (ha) of RMA,
were required to have an ‘equivalent’ performance
ery Act (RCRA) Subtitle C cover with a goal of
zero percolation; hence, their designation at RMA
as “RCRA-Equivalent Covers.”
The RMA covers were required to control storm
water infiltration, as well as prevent biointrusion
and control erosion. An overview of the rationale
used to design each of these components, the chal-
lenges presented by interdependent criteria, and
post-construction monitoring are presented herein.
In 1942, the US Army established the RMA site to
manufacture chemical warfare agents and incen-
diary munitions for use in World War II. Private
companies, including the Shell Oil Company,
manufactured pesticides at the site from 1952 to
1982. Disposal practices resulted in contamination
of the soil, structures, surface water, and ground-

Fig. 1 RMA RCRA-Equivalent Covers (For colour figure, refer to CD)
water to levels that posed unacceptable health
risks to humans and the environment.
RCRA-EQUIVALENT COVER DESIGN
The first RCRA-Equivalent cover system designed
and constructed at RMA was over the Shell Dis-
posal Trenches area (Shell Cover) and is approx-
imately 8.5 ha. The five remaining RCRA-
Equivalent covers (Fig. 1) are the Basin F Cover,
which extends approximately 41.8 ha, and the In-
tegrated Cover System (ICS), which encompasses
133 ha and consists of four adjacent consolidation
projects (Basin A, Complex (Army) Trenches,
Lime Basins, and South Plants Central Processing
Area). A cross section of the RMA covers is
shown in Fig. 2a and 2b.
The RCRA-Equivalent cover design used both
evapotranspiration (ET) and capillary barrier con-
cepts to control infiltration. Two different mate-
rials were used for the capillary barrier due to cost
considerations and difficulties experienced during
construction of the Shell Cover. As shown in Fig-
ures 2a and 2b, the RCRA-Equivalent cover sys-
tems include the following components, from bot-
tom to top:
• Biointrusion Component, designed to prevent
biota from accessing underlying contaminated
Williams et al. 409
soil and constructed of concrete cobbles (at least
0.41 m thick) overlain by a chokestone layer,
which provides a uniform surface for placement
of the subsequent capillary barrier material.
• Capillary Barrier Component, consists of a
nonwoven geotextile for the cover constructed
first (Shell Cover). The design for the subse-
quent covers (ICS and Basin F) was modified to
use a 0.03 to 0.08 m-thick layer of well-graded,
washed pea gravel.
• Unsaturated Soil Component, a 1.22 m-thick
layer of soil with specific geotechnical and
agronomic characteristics.
• Vegetation Component, uses native grasses
compatible with a short-grass prairie habitat.
Site-specific objectives and conditions included
the use of on-site borrow soil and compatibility of
the final cover slopes and vegetation with the des-
ignated future use of the site as a National Wild-
life Refuge. The large areas that required covers
placed additional constraints on the resulting de-
sign and construction efforts.
DESIGN RATIONALE
Each of the cover components discussed above
provides one or more functions. The biota control
function of the covers is achieved by the biointru-
410 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

sion and vegetation components. The infiltration gradation (e.g. aggressive water attack, acidic
control function is achieved by the integrated re- aqueous solutions, sulfates) (RVO 1997).
sponse of the unsaturated soil, capillary barrier,
and vegetation components. Finally, the erosion To deter invasion by prairie dogs, the seed mix for
control function is achieved by the unsaturated the cover vegetation included tall grass species.
soil and vegetation components, along with the Based on a study (RVO 1997), a gradation with at
grading and drainage control features of the cover least 33% of the cobble diameters ranging from
system. The characteristics of each of the cover 0.15 to 0.31 m would prevent a badger from push-
functions are described below. ing a cobble to the surface while, at the same time,
having voids that are too small to provide access
Biota Control Function of small rodents like prairie dogs or pocket go-
The primary design criteria for the biointrusion phers. A thickness of 0.41 to 0.46 m for the
layer were established for the predominant bur- (BBM) layer was selected and the BBM extended
rowing animal species present at RMA: badgers 15 m beyond the perimeter of the contaminated
and prairie dogs. Because the covers must isolate soils. The BBM was placed below the 1.22 m-
the waste left in place in perpetuity, it was critical thick unsaturated soil layer to address freeze-thaw
that the biota barrier material (BBM) be a highly and chemical degradation concerns (RVO 1997).
durable material that was resistant to animals,
freeze-thaw action, chemical breakdown from the While natural materials, such as crushed granite,
overlying cover soils, and moisture-induced de- were a likely option for use as BBM, an opportu-
(a) nity arose to recycle high-strength concrete from
Diverse mixture of the adjacent and recently decommissioned Denver
native plants Vegetation
Component
Stapleton International Airport. Acceptability cri-
teria for this recycled concrete option required a
Soil with Organic
Amendments (0.3 m)
Unsaturated Soil minimum compressive strength of 13.79 MPa and
Component
(1.22 m) a unit weight of more than 20.4 kN/m3 (RVO
AZ Soil
Capillary Barrier 1997). Laboratory testing of the Stapleton run-
Component
Nonwoven geotextile
or pea gravel Biointrusion ways and aprons indicated that the concrete was
Component
Chokestone (0 to 80 mm) acceptable due to its high density, durability,
Crushed Concrete
(0.41 to 0.46 m) hardness and evidence of limited aggregate segre-
Gradefill gation or surface deterioration (RVO 1997). The
Waste
volume of the Stapleton concrete was sufficient to
construct the 183.3 ha of covers at RMA. In addi-
(b) tion to meeting the design criteria, recycling the
Diverse mixture of airport concrete was cost-effective, eliminated
native plants high truck traffic through the adjacent communi-
Soil with Organic ties for constructing this cover layer, and pro-
Amendments moted EPA’s mission to protect human health and
AZ Soil the environment through the reduction, reuse, or
Nonwoven Geotextile recycling of materials.
Chokestone
Crushed Concrete Infiltration Control Function
Unsaturated Soil
Gradefill Unsaturated soil covers are an alternative ap-
proach that has been deployed at a number of sites
Fig. 2 Cross-section of the RMA Covers: (a) Sche- worldwide (Zornberg et al. 2003, Dwyer et al.
matic View; (b) Exposed cut in Shell Cover. 2006). ET and moisture storage significantly in-
(For colour figure, refer to CD) fluence the performance of unsaturated soil cover

systems. The innovation of this approach is that
basal percolation control is partly achieved
through storage of moisture that infiltrates during
precipitation events until it is released back to the
atmosphere through ET.
Based on site conditions and studies available at
the time (e.g. Melchior 1997), a quantitative per-
colation criterion (QPC) of 1.3 mm/year was se-
lected for the RMA alternative cover design (RVO
1998). A field demonstration was conducted to
prove the RMA covers would perform equivalent
to a RCRA Subtitle C cap. This equivalence dem-
onstration directly compared the measured field
percolation from four test plots to the QPC.
Four test plots consisting of unsaturated soil lay-
ers, approximately 9.14 m by 15.24 m, were de-
signed and constructed using on-site soils, but
without a biointrusion or capillary barrier compo-
nent. Data for each test plot was monitored be-
tween 1998 and 2003 for basal percolation, preci-
pitation, moisture content and overland runoff
(Kiel et al 2002). Basal percolation was collected
in pan lysimeters, which involved a geocomposite
drainage layer underlain by geomembrane (RVO
1998). Evaluation of the lysimeter data indicated
that all the test plots satisfied the QPC. However,
subsequent scrutiny of the moisture content data
indicated that the design criterion had been
achieved because a capillary barrier had developed
within the constructed test plots at the interface
between the soil layer and the underlying geo-
composite drainage layer.
The requirement to duplicate in full-scale con-
struction the successful infiltration control
achieved in the test plots led to additional perfor-
mance criteria on the cover design process. In ad-
dition to adopting a 1.22 m-thick soil layer and in-
corporating an underlying geotextile to create a
capillary barrier as in the test plots, the cover de-
sign required quantification of the soil properties,
soil placement conditions and agronomic characte-
ristics. An important criterion for selection of on-
site soils to be used for cover construction was
that their texture be within a defined Acceptable
Zone (AZ).
Williams et al. 411
The AZ for soil texture was based on the field
demonstration, hydraulic property testing and per-
colation modeling of the successful test plot soils.
The AZ was defined using the U.S. Department of
Agriculture textural triangle and identified accept-
able ranges of silt, sand, and clay content. In addi-
tion, the RCRA-Equivalent Covers were to be
compacted within a range from 75% to 85% of the
Standard Proctor maximum dry density with the
intent of promoting vegetation (TTECI 2005).
Capillary Barrier
Capillary barriers develop when placing a fine-
grained soil over a coarse-grained soil or a geotex-
tile. To further the assess the capillary barrier de-
velopment in the test plots, column tests were
conducted to assess four proposed capillary barrier
interfaces using a fine-grained soil layer placed
over: (a) a geocomposite drainage layer similar to
that used in the equivalence demonstration; (b)
geotextile with chokestone beneath it; (c) chokes-
tone; and (d) gravel. Each of the proposed inter-
faces was equally successful in developing a capil-
lary barrier (TTFWI 2005).
The Shell Cover includes a nonwoven geotextile
as a capillary barrier material underlying the fine-
grained unsaturated soil layer. Use of a geotextile
layer also acts as a filter, minimizing the migration
of soil particles into the underlying chokestone.
The selected geotextile was bright orange to serve
as a deterrent to human intruders from excavating
into the underlying contaminated soils (Fig. 4)
Vegetation
A diverse mixture of native plants was identified
in the design to maximize water removal, be com-
patible with the surrounding Wildlife Refuge and
remain resilient given unpredictable changes in the
environment resulting from pathogen and pest
outbreaks, disturbances (overgrazing, fire, etc.)
and climatic fluctuations. The design of the seed
mix for the cover vegetation involved the follow-
ing considerations:
• Erosion control
412 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
• A deep root system to increase the ET contribu-
tion
• Cool and warm season species to promote ET
for long periods, increasing water storage capac-
ity of the cover soils
• Wildlife deterrence
• Exclusion of forbs to allow the potential use of
herbicides for weed control.
The borrow identified for cover construction was
subsurface, on-site soil; thus, the design required
micronutrients to be added to the top 0.30 m and
the clay and calcium carbonate content was li-
mited to 40% and 15%, respectively. As previous-
ly mentioned, the 1.22 m-thick soil layer was
placed at a relatively loose density ranging from
75% to 85% of the Standard Proctor maximum dry
density to enhance vegetation growth.
Erosion Control Function
The design objectives included minimizing ero-
sion by wind and water as well as maximizing ru-
noff and minimizing ponding. Calculations for
long-term erosion due to storm water and wind
predicted a soil loss below 10 mm in 100 years.
As a result, an additional 0.15 m of soil was added
to the minimum cover thickness estimated to be
needed to meet the percolation criteria (1.07 m)
for a total cover thickness of 1.22 m.
Vegetation
Once established, plants are expected to dissipate
wind energy and reduce eolic erosion. The shal-
low root system of seeded native plants enhances
the soil surface resistance to water and wind ero-
sion. In addition, plants intercept some of the rain
before it impacts the ground surface, thereby re-
ducing the potential for water erosion.
Cover Configuration
The slope selected for the cover design was 3%. In
addition, overland flow lengths were limited to
102 m to minimize rill and gully formation. There-
fore, to minimize the amount of gradefill needed
to achieve the overall 3% slopes, a “broken back”
design was adopted that consisted of long, low
slope drainage channels that cut through the large
Fine-grained
AZ Soil
Nonwoven Geotextile
Fig. 4 Soil placement over the orange nonwoven
geotextile used as capillary barrier in the Shell
Cover (For colour figure, refer to CD)
cover areas. This minimized the overall cover
height and created multiple drainages to direct
storm water flow. The final design for all the
RMA RCRA-Equivalent covers includes approx-
imately 34 km of drainage channels ranging from
31 to 460 m in length, at grades ranging from
0.3% to 1%.
A full RCRA Subtitle C cover was designed where
concentrated flow would occur in the low drainage
slopes. The Subtitle C cover system includes a
geosynthetic clay liner, a 1.5 mm-thick geomem-
brane, a geocomposite drainage layer, and gravel
with a slotted drainage pipe down the flowline. In
addition, the channel surface is lined with concrete
to reduce variability in the final drainage surface
and promote storm water flow off the cover.
POST-CONSTRUCTION MONITORING
Construction of the Shell Cover was completed in
2007. The five other RCRA-Equivalent Covers are
expected to be completed in 2010. Until the vege-
tation is fully established, the performance of the
RCRA-Equivalent Covers is rigorously monitored
to assess functionality. Generally, monitoring of
the covers includes visual observations of the cov-
er for damage, inspection of the vegetation, and
percolation monitoring using lysimeters. In addi-
tion, the Shell Cover is instrumented with water
content reflectometers to measure moisture con-
tent within the cover soil. Visual inspections and
percolation monitoring is conducted monthly, qua-

litative vegetation inspections are conducted se-
miannually, and quantitative vegetation inspec-
tions are performed annually. Annual Cover Re-
ports that document inspection findings,
percolation monitoring data, vegetation assess-
ment data, and maintenance activities, are ex-
pected to be issued in November of each year.
Long-term monitoring of the biointrusion and ca-
pillary barrier components is not conducted be-
cause they are located beneath the 1.22 m-thick
soil layer. However, any breach in these compo-
nents can be observed during monthly inspections
conducted for burrowing animals through the soil
component of the cover. Immediate removal of
prairie dogs is expected to keep badgers away
from the covers, as prairie dogs are one of their
chief prey. Settlement monuments were installed
to monitor for soil loss and/or settlement of the en-
tire soil cover. The monuments consist of a pipe
and base plate that sit on top of the BBM layer and
extends to the cover soil surface. These settlement
monuments are monitored as part of the cover in-
spections to identify needed repairs such as rills,
gullies, excessive sheet erosion, settlement, indica-
tions of ponding, and overall integrity of the cover
drainages.
CONCLUSIONS
The design of an alternative cover for containment
of highly contaminated waste has inherent con-
straints that result from the biota, infiltration, and
erosion control functions that must be met by the
cover system. In particular, design and construc-
tion of these components required integration of
multiple site-specific criteria to achieve a compat-
ible, interactive system design. As monitoring data
become available, the functionality of the cover
systems will be regularly evaluated and, ultimate-
ly, a determination of an “operational and func-
tional” cover will be made.
ACKNOWLEDGMENTS
The authors are thankful to the US Army and Te-
tra Tech EC, Inc. for their partnership during the
remediation at RMA and to Richard McPeek for
photographs taken in support of EPA field over-
sight of the cover construction.
Williams et al. 413
REFERENCES
Dwyer, S.F., Rager, R., and Hopkins, J. (2006).
Cover System Design Guidance and Require-
ments Document. Los Alamos National La-
boratory report LA-UR-06-4715.
Kiel, R.E., Chadwick, D.G., Lowrey, J., Mackey,
C.V., Greer, L.M. (2002) “Design of evapo-
transpirative (ET) covers at the Rocky Moun-
tain Arsenal.” Proceedings: SWANA 6th An-
nual Landfill Symposium.
Melchior, S. (1997). “In-situ studies of the per-
formance of landfill caps (Compacted clay lin-
ers, geomembranes, geosynthetic clay liner,
capillary barriers).” Land Contamination and
Reclamation. 5(3), 209-216.
Remediation Venture Office (RVO). (1997) Ba-
sin A Consolidation and Remediation Project,
Final Design Package (Issued for Construc-
tion). Section 3.3, Human/Biota Barrier De-
sign Analysis.
RVO. (1998) Final RCRA-Equivalent Cover
Demonstration Project, Comparative Analysis
and Field Demonstration Design Scope of
Work.
Tetra Tech EC, Inc. (TTECI) (2005). RCRA-
Equivalent Cover Post-Demonstration Geo-
technical Evaluation, Final, Summary Report
for Acceptance Zone Development and Densi-
ty Requirements for RCRA-Equivalent Cover
Soils.
Tetra Tech FW, Inc. (TTFWI). 2005. Final Ca-
pillary Break Test Report.
Zornberg, J.G., LaFountain, L., and Caldwell, J.A.
(2003). “Analysis and Design of Evapotranspir-
ative Cover for Hazardous Waste Landfill.”
Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, Vol. 129, No. 5, pp. 427-
438.
 Performance Evaluation of various Clay-based Landfill Covers
subjected to Differential Settlements: Centrifuge Study
S. Rajesh
Indian Institute of Technology Bombay, Mumbai, India (rajesh.s@iitb.ac.in )
B.V.S. Viswanadham
Indian Institute of Technology Bombay, Mumbai, India (viswam@civil.iitb.ac.in )

ABSTRACT The objective of this paper is to study the deformation behavior of barriers made out of clay
and sand-bentonite blend subjected to continuous differential settlements through centrifuge model studies.
Motor-based differential settlement simulator was used for inducing differential settlements at 40 gravities.
Digital image analysis was adopted for estimating soil strains. Performance of soil barriers was monitored at
various distortion levels. An attempt has been made to evaluate limiting distortion level for both barriers.
The analysis and interpretation of centrifuge test results indicate that a significant delay in the crack initiation
was observed for a sand-bentonite barrier when compared to compacted soil barrier.

INTRODUCTION Keeping in view of the potential of compacted soil

Imperviousness of clay based soil barrier provided
in the cover system of municipal solid waste
landfills and low-level radioactive waste disposal
sites is essential to mitigate the environmental
hazard. Compacted soil barrier is generally made-
up of naturally available fine-grained soil.
However, when impervious natural soils are not
readily or economically available then compacted
mixtures of bentonite and sand or compacted
mixture of bentonite and in-situ soil can be used.
From literature, it was observed that a failure of
soil barrier due to excessive settlements of waste,
in particular, differential settlements is one of the
reasons for cracking failure of impervious barriers
of cover systems. The problem due to differential
settlements is more pronounced in cover systems
when compared to bottom-lining systems because
the former is subjected to low overburden pressure
and constructed on an unstable foundation. Soil
barriers tend to crack, if the induced tensile strains
due to differential settlements exceed the
permissible tensile strain of the barrier material.
barrier and bentonite-enriched soils and problem of
differential settlements, in the present study, an
attempt has been made to evaluate the deformation
behavior of soil barriers at the onset of differential
settlements using centrifuge modeling technique.
Application of centrifuge modeling technique to
the present study is relevant because the loss of
integrity of soil barriers is highly influenced by the
presence of prototype stress conditions. The
centrifuge tests reported herein were performed at
40 gravities. The 4.5 m radius large beam
centrifuge at IIT Bombay having a capacity is
250g-ton with a maximum payload of 2.5t at 100g
was used. The detailed specifications are discussed
by Viswanadham & Rajesh (2009).
EXPERIMENTAL METHODOLOGY
A mixture of commercially available kaolin clay
and locally available poorly graded sand in the
proportion of 80:20 (by dry weight) was used to
represent a compacted soil barrier (CSB). It has a
liquid limit of 38%, plasticity index of 16%,

maximum dry unit weight and optimum moisture
content (standard Proctor compaction) are 15.9
kN/m3 and 22%. A blend of 20% commercially
available bentonite clay and 80% poorly graded
sand by dry weight was used to represent sand-
bentonite barrier (SBB). Liquid limit, plasticity
index, maximum dry unit weight and optimum
moisture content (standard Proctor compaction) of
model SBB was found to be 69%, 46%, 17 kN/m3
and 15.5% respectively. The coefficient of
permeability of CSB and SBB moist-compacted at
their respective maximum dry unit weight and
optimum moisture content are 0.4x10-9m/s and
3.2x10-10m/s respectively. Both model soil barrier
materials were chosen in such a way that they
represent properties of barrier materials used in
landfills (Benson et al. 1999).
Since, majority of soil barriers are compacted at
wet side of optimum, it was decided to prepare soil
barriers (both CSB and SBB) at moisture content
equal to 5% wet of optimum and corresponding
dry unit weight. As the thickness of the soil barrier
commonly adopted in landfill covers range from
0.6m to 1.2m, a 1.2m thick soil barrier was
modeled. Two centrifuge tests at 40g were
conducted one for each soil barrier type to study
the influence of soil barrier type on the
deformation behavior of a respective soil barrier
subjected to continuous differential settlements.
The maximum central settlement of 25mm (1m in
prototype dimensions) was induced at a settlement
rate of 1 mm/min (in model dimensions).
Test procedure
Differential settlements were simulated in a
centrifuge using Motor based Differential
Settlement Simulator (MDSS). Figure 1 shows a
cross-section of the test set-up used in the present
study. A 30mm thick model soil barrier along with
sacrificial layers i.e. a layer of fine sand (30mm
thick) and coarse sand (30mm thick) was prepared
at normal gravity as shown in Fig. 1. A 30mm
thick soil layer corresponds to 1.2m thick at 40
gravities. Upon completion of the soil barrier layer,
plastic markers were inserted along the cross-
section of a soil barrier at a spacing of 20 mm
centre to centre from the centre of the soil barrier
to measure deformation profiles during various
Rajesh and Viswanadham 415
stages of the test. A measured quantity of water
was placed above the soil barrier in order to assess
the infiltration of water through the soil barrier
while inducing differential settlements. In order to
avoid leakage of water between the sides of the
container and the soil barrier, water tight seal made
up of a thick bentonite paste was applied all along
the sides of the soil barrier, as shown in Fig.1.
Pore Pressure Transducers (PPTs) were placed
above the surface of the soil barrier to monitor the
water level and to establish water breakthrough. A
digital camera was placed on the front side of the
model; Charge Coupled Device (CCD) video
camera was mounted on top of the model. Desired
settlement rate of 1mm/min (in model dimensions)
was achieved by regulating the speed of motor of
MDSS system. At various stages of central
settlements, photographs of front elevation of the
model were captured through image acquiring
software and were later used for image analysis to
compute deformation profiles.
Fig. 1 Cross-section view of model test package
(All dimensions are in mm)
[Modified after Rajesh & Viswanadham, 2009]
ANALYSIS AND INTERPRETATION
Images captured with the help of an image
capturing software at various settlement stages
from 0 m to 1 m in the intervals of 0.2 m were
digitized by using a module map edit of GRAM++
package (GRAM++ 2004). As shown in Fig. 2, a
grid of permanent markers was placed on the
internal side of front Perspex sheet at pre-
416 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

determined distance and were used to scale and
control the images captured. The error involved in
digitizing permanent markers i.e., standardizing
the image can be obtained from the GRAM++
software. Once the error is within the tolerable
limit, digitization of the discrete markers was
carried-out. The coordinates of discrete markers
were obtained with reference to the coordinates of
permanent markers. When the horizontal distance
from centre of the soil barrier x is zero, the value
of settlement is defined as a central settlement a
(Figs. 1 and 2). Settlement ratio, a/amax is defined
as the ratio of central settlement a at any stage of
deformation to the maximum central settlement
amax (for both the tests, amax is equal to 25mm, in
model dimensions). Distortion level, a/l is the ratio
of central settlement a at any particular stage of
deformation to a horizontal distance (l), where l is
found to be 200mm (in model dimensions). This is
defined as a horizontal distance from the centre of
the soil barrier to a point along the surface of the
barrier, where induced settlements cease to be
diminished.
could be obtained for various stages of central
settlements. Moreover, higher magnitude of
settlements occurred near the central portion of the
soil barrier and negligible settlements occurred
beyond x = 8m. Strains along the top surface of the
soil barrier were computed using combined
bending and elongation method (Lee & Shen,
1969). Outer fiber strain along the top surface of
the barrier İof is the algebraic summation of
elongation strain and bending strain. The
methodology adopted in computation of strain was
described by Viswanadham & Rajesh (2009).
Figure 4 shows the variation of outer fiber strain
for a 1.2m thick sand-bentonite barrier. Sign
convention adopted for tensile strain is positive. It
can be observed that as the central settlement
increases, both the compressive as well as tensile
strains increase. Moreover, maximum outer fiber
strains (tensile) (İof)max were observed at the zone
of maximum curvature for all central settlement
stages.
Horizontal distance from x (m)
Deformation profiles (m)

0 2 4 6 8 10 12 14
0.0

x 0.2

0.4 SBB
a = 0.2m
220 mm 0.6 a = 0.4m
Rmin a = 0.6m
0.8 a = 0.8m
a = 1.0m
a 1.0
l
Fig. 3. Variation of measured deformation profiles with
350 mm horizontal distance for a 1.2m thick SBB.
Permanent markers Discrete markers Horizontal distance from x (m)
0 2 4 6 8 10 12 14
6
Outer fiber strain (%)

Fig. 2. Front elevation of a deformed SBB (+) Tension

4 (-) Compression
(For colour figure, refer to CD)
2
The measured co-ordinates of a row of discrete
0
markers are approximated with an exponential
equation of the normal distribution to get the -2 SBB
a = 0.2m; a = 0.4m
deformation at various stages along the -4 a = 0.6m; a = 0.8m
longitudinal axis of the soil barrier. Figure 3 shows a = 1.0m
the deformation profile of the sand-bentonite -6
barrier under various stages of central settlement.
Fig. 4. Variation of outer fiber strain with horizontal
It can be noticed that smooth deformation profiles
distance for a 1.2m thick SBB.
 Rajesh and Viswanadham 417

When a strain value increases beyond the centrifuge test results of both CSB and SBB is
permissible value of the soil barrier material, then given in Table.1.
there may be an occurrence and propagation of the
5
crack, which can hamper the functionality of the
CSB
cover system. Once after ascertaining central 4 SBB

(Hof)max (%)
settlement at which initiation of cracking occurred
along the surface, value of strain at crack initiation 3
İc (which is defined herein as the value of an outer 2
fiber strain at the zone of maximum curvature at
the onset of cracking of the soil barrier) can be 1
determined. Figure 5 presents the variation of 0
maximum outer fiber strain with a/l and a/amax for 1 10 100 1000
Rmin (m)
both compacted soil and sand bentonite barriers.
With an increase in a/amax and a/l, a significant
Fig. 6. Variation of maximum outer fiber strain with
increase in the value of (İof)max was observed for radius at the zone of maximum curvature.
both the soil barriers. It can also be noticed that a
delay in initiation of the crack was observed for Table 1. Summary of centrifuge test results.
SBB when compared to CSB. This implies that Test Type (İof)max *dc *wc (a/l)lim
SBB has sustained relatively higher distortion level
RCL4 CSB 3.39 % 30 4.29 0.042
at the onset of cracking. Moreover, for various RCL14 SBB 3.98 % 12 2.25 0.062
values of a/amax and a/l, magnitude of (İof)max of
CSB – Compacted soil barrier; SBB- Sand bentonite
SBB was found be higher than CSB.
barrier; (İof)max - Maximum outer fiber strain; dc –
a/l depth of cracks in mm; wc – average width of cracks in
0.000 0.025 0.050 0.075 0.100 0.125 mm (*at the end of centrifuge test given in model
4 dimensions); (a/l)lim – limiting distortion level.

3
(Hof)max (%)

The performance of a soil barrier as an effective

2
1 Hc for CSB
0
0.0 0.2 0.4 0.6
a / amax
Fig. 5. Variation of maximum outer fiber strain with
settlement ratio and distortion level.
Figure 6 shows the variation of (İof)max with the
radius at the zone of maximum curvature Rmin for
various stages of central settlements. It can be
observed that both soil barriers are found to have
identical pattern of variation. However, magnitude
of radius of the SBB was found to have marginally
lower values for all values of central settlements
when compared to CSB. The summary of
hydraulic barrier can be best illustrated through
CSB infiltration of water (as a permeant placed above
SBB the soil barrier) through the soil barrier. This
infiltration of water can be directly observed by the
Hc for SBB
reduction in the volume of water measured using
0.8 1.0 PPT's, during various stages of central settlements.
The volume change ratio, VCR is the ratio of
volume of water at a particular settlement stage (V)
to that of initial volume of water (Vo) placed on the
surface of the soil barriers. The Infiltration ratio
(IFR) can be determined as IFR 1  VCR . The
value of IFR ranges from zero to one. IFR = 0
indicates negligible infiltration and complete
infiltration of water through the soil barrier for IFR
= 1.
Figure 7 shows the variation of IFR with a/l and
a/amax. For both the soil barriers, it can be observed
that there exists a gentle variation in IFR values up
to a certain a/amax and a/l and thereafter a steep
variation was observed. Steep variation of IFR
418 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
indicates water breakthrough of the soil barrier.
The distortion level corresponding to the water
breakthrough is termed as limiting distortion level,
(a/l)lim. When cracks extend sufficient width and
depth, water has a tendency to flow through the
soil barrier. The limiting distortion level has been
determined using back tangent method (Fig. 7).
The (a/l)lim value for CSB and SBB was found to
be 0.042 and 0.062 respectively. The higher value
of limiting distortion level for SBB indicates that
SBB has an ability to withstand relatively higher
distortion levels when compared to a compacted
soil barrier. However, it can be noticed that both
the soil barriers lost their integrity in terms of
water breakthrough at higher distortion levels.
a/l
0.000 0.025 0.050 0.075 0.100
1.0
CSB
SBB
0.8
0.6
IFR
0.4 (a/l)lim = 0.042
(a/l)lim = 0.062
0.2
0.0
0.0 0.2 0.4 0.6 0.8
a / a max
Fig. 7 Variation of IFR with a/amax and a/l
During post-test investigations on deformed soil
barriers, intactness of water sealing arrangement
was verified thoroughly; hence, water lost while
inducing continuous differential settlement was
primarily due to the infiltration of water through
cracks. When strain level due to deformation
exceeds the permissible limit of the soil barrier,
hydraulic barrier has a tendency to crack. By
comparing Figs. 5 and 7, it can be inferred that
water breakthrough has not occurred immediately
after initiation of crack but it has occurred after
cracks have extended to sufficient depth and width.
In other words, distortion level at the initiation of
crack for CSB and SBB was found to be 0.018 and
0.031 respectively; where as, the limiting
distortion level was found to be almost twice the
value of distortion level at the initiation of crack
for both the soil barriers. The strain value
corresponding to limiting distortion level for CSB
and SBB was found to be 0.83% and 1.52%.
The status of both the soil barriers after subjecting
to a distortion level of 0.125 (at the end of
centrifuge test) is shown in Figure 8. It can be
observed that CSB tends to experience multiple
cracks with few cracks extending up to full-depth
of the soil barrier. However, in the case of a sand-
bentonite soil barrier, a single wide crack
extending up to the entire breadth of the container
but extends only up to a partial depth, at the zone
of maximum curvature was observed. The average
0.125 width of cracks of CSB and SBB were found to be
4.29mm and 2.25mm (in model dimensions). The
beneficial effect of self healing nature of the
bentonite and higher plasticity index can be
observed starting from delay in initiation of cracks
to sustaining higher distortion level. However, an
adequate care should be taken in retaining the
moisture content of SBB. Figure 8 also shows the
top view of both the soil barriers after about 24
hours testing in a centrifuge. It can be noticed that
a CSB with lower plasticity index has not shown
1.0
any significant variation except in the reduction of
moisture content; whereas, SBB has shown
significant desiccation cracks apart from reduction
in the moisture content. Moreover, in the present
study the moist-compacted SBB was tested
without giving adequate time for hydration of
bentonite occupied within sand grains. Further
studies in this direction are warranted to study the
influence of type and content of the bentonite and
self-healing nature of bentonite on the
deformation behaviour of SBB.
CONCLUSIONS
Based on the analysis and interpretation of
centrifuge test results, the following conclusions
are drawn:
 Rajesh and Viswanadham 419

30mm

i) Cross-section view ii) Top view at the zone of iii) Top view at the zone of
maximum curvature maximum curvature (After 24
a) Compacted soil barrier hours)

30mm

i) Cross-section view ii) Top view at the zone of iii) Top view at the zone of
maximum curvature maximum curvature (After 24
b) Sand bentonite barrier hours)
Fig. 8. Status of soil barriers after inducing a distortion level of 0.125 (For colour figure, refer to CD)
REFERENCES
1) A 1.2 m thick compacted soil barrier moist- Benson C.H., Daniel D.E. and Boutwell G.P.
compacted at wet side of optimum was found (1999). Field performance of compacted clay
to experience multiple cracking pattern. The liners, Journal of Geotechnical and
value of İc and maximum İof was found to be Geoenvironmental Engineering, ASCE, 125(5),
0.32% and 3.39% respectively. The limiting 390-403.
distortion level was found to be 0.042. GRAM++ (2004). http://www.csre.iitb.ac.in/gram++
2) A 1.2m thick moist-compacted SBB at wet Lee K.L. and Shen C.K. (1969). Horizontal
side of optimum was found to experience movements related to subsidence, Journal of
single wide crack at the zone of maximum Soil Mechanics and Foundation division,
curvature. The value of İc and maximum İof ASCE, 94 (6), 139-166.
was found to be 0.75% and 3.98 respectively. Rajesh S. and Viswanadham B.V.S. (2009).
The limiting distortion level was found to be Evaluation of geogrid as a reinforcement layer
0.062, which is 48 % more than SBB. in clay based engineered barriers, Applied Clay
It can be concluded from the present study that the
Science, 46(2), 153-165.
performance of sand bentonite barrier was found
to superior than the compacted soil barrier in Viswanadham B.V.S. and Rajesh S. (2009).
sustaining relatively higher distortion levels. Centrifuge model test on clay based engineered
barriers subjected to differential settlement,
Applied Clay Science, 42 (3-4), 460-472.
 Effect of Channel Flow on Water Balance in a Layered Inclined Soil
Cover over Mine Waste Rock
Q. Song
University of Western Ontario, Ontario, Canada (qsong3@uwo.ca)
E.K. Yanful
University of Western Ontario, Ontario, Canada (eyanful@eng.uwo.ca)

ABSTRACT Two soil covers consisting of silty clay and sand and sloped at 20% were installed over
reactive waste rock in the laboratory. One cover had a 10-cm wide sand-filled channel in the barrier layer;
the other cover was a normal one without channel. A finite element commercial software, VADOSE/W, was
used to model the measured water balance in the two covers. The measured results indicated that channel
flow had a significant effect on interflow and percolation, and marginal effect on actual evaporation and
change in soil water storage. The modeled water balance components agreed well with measured values.

INTRODUCTION barrier forms when a fine-grained material is

The management of acid rock drainage (ARD) is a
major environmental problem facing the mining
industry. Measures to control ARD are divided
into two categories (Johnson & Hallberg 2005):
the treatment of drainage (or migration control)
and prevention of ARD generation (or source
control). The source control method uses
techniques to create conditions that eliminate or
reduce the oxidation of sulphide-bearing minerals,
hence minimizing ARD generation.
A method of source control is the use of an
engineered soil cover to prevent water and oxygen
ingress into the underlying waste. Soil covers may
be classified into resistive covers or water balance
covers according to the design mechanism and
cover function. Resistive covers employ materials
having very low hydraulic conductivity to resist
water and oxygen entering the barrier layer
(Benson et al. 1994). Water balance covers either
have high water storage capacity to hold water in
the cover layers during rainfall periods (such as
monolithic barriers) or use capillarity principle to
minimize water and oxygen ingress into the
wastes (such as capillary barriers). A capillary
placed over a coarse-grained material. Hence, a
high saturation in the barrier layer (the fine-
grained layer) is maintained due to the contrast in
the unsaturated hydraulic properties of the two
materials, which can minimize the oxygen ingress
into the underlying wastes (e.g. Yanful 1993,
Khire et al. 2000, Ogorzalek et al. 2008).
Preferential flow in unsaturated soils is not an
unusual phenomenon (Hagedorn & Bundt 2002).
There are three kinds of preferential flow
mechanisms: macropore flow (Beven & Germann
1982), fingered flow (Selker et al. 1992), and
funnel flow (Kung 1990, Walter et al. 2000). Out
of the three mechanisms, macropore flow through
the barrier layer is a major concern because it can
have an adverse impact on soil cover performance
(e.g. Kelln et al. 2007). Macropores in soils can
result from a number of factors including
freeze/thaw and wet/dry cycling, plant root
penetration and animal intrusion (e.g. Beven &
Germann 1982).
Computer modeling provides a cost-effective
approach for the design and evaluation of a soil

cover. For this reason, a number of computer
programs, such as Soilcover (USG 1997),
HYDRUS (Simunek et al. 2006), and
VADOSE/W (Geo-slope 2004), have been used in
cover analysis. VADOSE/W couples water flow
with heat transport to simulate unsaturated water
flow in porous media. It has successfully been
used by researchers to model water flow in soil
covers (e.g. Adu-Wusu et al. 2007).
In the present study, two soil covers were tested.
One cover had a 10-cm wide sand filled channel
(Test 1), which represented the aggregate of
fissures, cracks, and other macropores that control
preferential flow in the field; the other cover was
an intact cover without channel pathway (Test 2).
The macroporosity of the channel with respect to
the tested barrier layer (silty clay) was 3.4%,
which is similar to those of other fracture
networks reported by other researchers (e.g.
Germann & Beven 1981, Perret et al. 1999). The
commercial computer program VADOSE/W was
employed to predict measured water balance of the
two tested soil covers. The objectives of the study
were (1) to investigate the effects of channel flow
on water balance in the soil covers; (2) to simulate
the measured water balance using VADOSE/W.
MATERIALS AND METHODS
Materials
Construction sand (fine-medium sand) and silty
clay, obtained from suburban London, Ontario,
Canada, were used as cover materials. The waste
rock used in the experiments was obtained from
Mattabi mine site near Ignace, Ontario, Canada.
Before its use, the waste rock was crushed, sieved
(through No. ½ inch sieve) and flushed with tap
water to remove debris and coatings adhering to
the waste rock particles. The sand, silty clay and
waste rock had saturated hydraulic conductivity of
2.8x10-4, 9.5x10-11, and 3.1x10-4 m/s, respectively.
Song and Yanful (2010a, b) have presented details
of the materials used in the experiments.
Description of experimental unit
The experimental unit was comprised of a box
(made of 1/2” thick PVC sheets) and a rainfall
Song and Yanful 421
simulation system (including a rainfall distributor,
two reservoir tanks, and a pump). The box used in
the experiments measured 120 cm x 120 cm x 25
cm (width x height x thickness) in inside
dimensions with three outlets (for collection of
percolation) at the bottom and two outlets (for
collection of runoff and interflow) at left and one
inlet (for by-pass flushing purpose) at the right
side. The whole box was reinforced with steel
angle bars following the installation of the soil
cover. The box sat on a steel frame 40 cm high,
and the rainfall distributor was placed 30 cm
above the box. A table fan was used to circulate
ambient air around the box.
Cover installation, instrumentation and
measurements
The sections of soil covers for the two tests and
associated instrumentation are presented in Figs.
1a and 1b, respectively. Apart from the sand filled
channel in test 1, the two tests had similar soil
installation and instrumentation. Instrumentation
included soil temperature probes, soil water
pressure sensors, volumetric water content rods,
and oxygen concentration sampling ports. Soil
water pressure and temperature were measured
and stored hourly in a datalogger. The volumetric
water contents of the cover soils and waste rock
were measured once a day. Gaseous oxygen
concentrations in the soil covers were sampled
every three days and measured using an oxygen
analyzer. Other measurements included water
balance components (runoff, interflow, percolation,
precipitation) and ambient environmental
conditions (air temperature and relative humidity).
Collected percolation was sampled and analyzed
for pH, electrical conductivity, sulphate
concentration, alkalinity, acidity, and dissolved
metals. Test 1 was run from 8 February to 8 July
2008, while Test 2 was monitored from 3 October
2008 to 3 March 2009.
Estimation of actual evaporation
As there was no runoff, actual evaporation from
the cover system was calculated according to the
following water balance equation (Eq. 1):
Vprec - Vintf - Vperc - Vevap= ± Vstor (1)
422 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

lope lope
20% s W10 20% s W10'
P10 P10'
T1 W11 T1' W11'
Sand T2
P11 Sand T2'
P11'
P1 W1 O7 P1' W1' O7'
P7 W7 P7' W7'
W2 O8 W2' O8'
P2 W12 P2' W12'
O1 O1' P12'
T3 T3'

Sand Channel
P12
O2 O9 O2'
W3 Silty Clay Barrier W13 W3' Silty Clay Barrier P13' W13'
P3 P13 P3' O9'
P8 W8 W8'
P8'
O6 P14 W14 O6'
W4 W4' W14'
P4 O3 P4' O3' P14'
T4 O10 T4' O10'

W5 O11 W5' O11'

P5 P5'
O4 P15 W15 O4' W15'
T5 P15'
T5'
P9 W9 W9'
O5 O5' P9'
W6 Waste Rock W6' Waste Rock
P6 P6'

P1-P15 Suction sensors O1-O11 Oxygen sampling ports P1'-P15' Suction sensors O1'-O11' Oxygen sampling ports
W1-W15 TDR rods T1-T5 Temperature sensors W1'-W15' TDR rods T1'-T5' Temperature sensors
Unit: cm Unit: cm

Fig. 1a Section of soil cover with channel flow Fig. 1b Section of soil cover without channel flow
pathway (Test 1) and instrumentation pathway (Test 2) and instrumentation

where Vprec is the volume of precipitation (m3);
Vintf is the interflow (m3); Vperc is the percolation
(m3); Vevap is the actual evaporation (m3); and Vstor
is the change in water storage (m3). A positive
change in water storage meant that the total water
content increased in the system, while a negative
value meant the total water content decreased.
Computer Modeling
Modeling method A step-by-step modeling
method was used to simulate the measured water
balance components for the two tests. Specifically,
the modeling processes (usually employing
unsaturated hydraulic conductivity functions) were
based on the drying and wetting stages of the soil
covers. The soil covers were first subjected to 18
days of drying, followed by 6 days of wetting and
then 61 days of drying. Following that, the soil
covers were subjected to another wet–dry cycle (7
days of wetting and 60 days of drying). The drying
and wetting processes were only simulated in the
sand layer during the modeling because: (1) the
measured water content and matric suction in the
sand layer showed significant difference in the
precipitation and evaporation stages; (2) the water
content and suction in the silty clay and waste
rock were relatively stable during the precipitation
and evaporation periods.
Input data for modeling For water flow modeling
in VADOSE/W, the input data included soil
properties (soil water characteristic curve,
unsaturated hydraulic conductivity, thermal data),
and climatic conditions. The soil water
characteristic curves (SWCCs) of the materials
were measured and fitted with the equation by
Fredlund & Xing (1994). Then the unsaturated
hydraulic conductivity functions of the soils were
obtained by fitting the saturated hydraulic
conductivities of the soils and the corresponding
SWCCs with the function embedded in
VADOSE/W. Input thermal data included mass
specific heat capacity and thermal conductivity
values of the soils. The climate data input in
VADOSE/W were daily temperature (min & max),
relative humidity (min & max), wind speed,
potential evaporation, and precipitation and its
duration.
Boundary and initial conditions The measured
indoor climate data were employed as the upper
boundary condition; other sides of the box were

assigned no-flux conditions for water flow except
at the outlets, where seepage points or no-flux
points were specified dependent on the modeled
process (drying or wetting). Similarly, apart from
the upper climate boundary, the other three sides
were set to a no-flux boundary for heat transport.
For initial conditions, -0.8 m, -0.3 m, and -0.1 m
were specified at nodes in the sand layer, silty clay
layer, and waste rock, respectively, for both tests.
These values were similar to those measured in the
soils at the beginning of the tests. The initial
temperature was specified as 22°C for both tests.
Detailed SWCCs, unsaturated hydraulic
conductivity functions and input thermal data were
presented by Song and Yanful (2010b).
RESULTS AND DISCUSSION
Measured and modeled water balance in Test 1
and Test 2 are presented in Figures 2a and 2b,
respectively. In general, water balance
components predicted from the computer
modeling had good agreement with the measured
water balance components for both tests. The
maximum discrepancy between the measured and
modeled water balance components occurred for
actual evaporation. For example, the modeled
actual evaporation (27.7%) was larger by 4.1%
than that measured (23.6%) in Test 1, and the
modeled actual evaporation (32.4%) in Test 2 was
greater by 4.5% than that measured (27.9%). This
phenomenon could be attributed to the different
methods used to calculate actual evaporation in the
computer modeling and in the measurement
method. In VADOSE/W, actual evaporation was
calculated directly based on the measured climatic
data and model calculated suction at the surface of
the soil cover, while the measured actual
evaporation was obtained from the water balance
equation (Eq. 1). The discrepancies between the
measured change in soil water storage, actual
evaporation and the corresponding modeled values
in the second wet-dry cycle were relatively larger
than those in the first wet-dry cycle as shown in
Figs. 2a and 2b, which implied that the hydraulic
properties (SWCC and unsaturated hydraulic
conductivity) of the sand needed to be modified
slightly for the second wet-dry cycle to get better
fit between measured and modeled results.
Song and Yanful 423
Results from both tests indicate that soil water
storage increased during the rainfall events, and
larger rainfall intensity resulted in larger change in
soil water storage. For instance, the average
intensities for the first and second rainfall events
in Test 1 were 27.34 mm/d and 21.16 mm/d
respectively (data not shown); correspondingly,
the maximum changes in soil water storage during
the first and second rainfall events in Test 1 were
8.5% and 8.1%, respectively. In Test 2, the
average precipitation intensities for the first and
second rainfall events were 27.24 mm/d and 21.08
mm/d; as a result, the measured calculated
maximum change in soil water storage was 9.1%
in the first rainfall events and 8.8% in the second
rainfall events. Higher rainfall intensity also
produced larger interflow. For example, the
measured interflow in the first rainfall events in
Test 2 was 38.9%, which was larger than that in
the second rainfall events (33.9%). Similar results
were found for Test 1.
The measured water balance components in Test 1
and Test 2 showed that the final interflow
increased from being 35.3% of the total rainfall in
Test 1 to 72.8% in Test 2; while percolation
decreased from 43.0% in Test 1 to zero in Test 2
during the test period. Actual evaporation only
slightly increased from 23.6% in Test 1 to 27.9%
in Test 2. These tested results implied that
increased percolation in the soil cover with
channel flow (Test 1) was mainly produced by the
decrease in interflow when compared to the
measured percolation and interflow in the soil
cover without channel flow (that is, intact soil
cover or Test 2). Final cumulative soil water
storage change increased from being -1.9% of the
total precipitation (a negative value meant a
decline in the total soil water content in the cover
system compared to the initial total soil water
content) in Test 1 to -0.6% in Test 2, which
implied that there was less soil water loss in Test 2
than in Test 1 during the test periods and the intact
soil cover would provide better protection to the
underlying waste rock. It is possible that the
effects of the channel or preferential flow on the
water balance components also depend on other
factors, such as location of the preferred pathway,
424 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

geometry of the flow pathway, geometry of sloped water balance components (and the performance)
cover, and characteristics of soils. The climatic of the inclined soil cover would be complex, and
environment would also impact the effects of the further research is necessary for getting more
channel flow. From this point of view, the effect knowledge of the effects to improve soil cover
of the preferential flow in barrier layers on the design practice.
Prec-measured Prec-applied Interf-measured Interf-modeled
Perc-measured Perc-modeled Evap-measured Evap-modeled
Stor-measured Stor-modeled
100
Cumulative percentage (%)

Rainfall events 1

50
Rainfall events 2

-50

-100
0 20 40 60 80 100 120 140 160
Tim e (day)
Fig. 2a Measured and modeled water balance components in Test 1 (soil cover with channel). Prec - precipitation,
Interf - interflow, Perc - percolation, Evap - actual evaporation, Stor - soil water storage change
Prec-measured Prec- applied Interf-measured Interf-modeled
Perc-measured Perc-modeled Evap-measured Evap-modeled
Stor-measured Stor-modeled
100
Cumulative percentage (%)

Rainfall events 1
50
Rainfall events 2

-50

-100
0 20 40 60 80 100 120 140 160
Tim e (day)
Fig. 2b Measured and modeled water balance components in Test 2 (soil cover without channel). Prec - precipitation,
Interf - interflow, Perc - percolation, Evap - actual evaporation, Stor - soil water storage change
(For colour figure, refer to CD)
CONCLUSIONS significantly in the soil cover with channel flow
Channel flow occurring in the barrier layer of a (Test 1), and increased percolation was mainly
soil cover can considerably influence water transferred from the decreased interflow because
balance in the soil cover. Compared to water the sum of the measured interflow and percolation
balance components in the intact soil cover (Test in both the intact soil cover and the soil cover with
2), interflow decreases and percolation increases channel flow was close to each other. Channel

flow occurring in Test 1 had only small effects on
actual evaporation and soil water storage change
relative to those measured in Test 2. The modeled
water balance components agreed well with those
measured for both Test 1 and Test 2 when using
the step-by-step modeling method, in which the
wetting and drying processes in the surface sand
layer had different unsaturated hydraulic
conductivity functions.
REFERENCES
Adu-Wusu C. Yanful E.K. Lanteigne L. and
O'Kane M. (2007) Prediction of the water
balance of two soil cover systems, Geotech.
Geol. Eng., 25, 215-237.
Benson C.H. Bosscher P.J. Lane D.T. and Pliska
R.J. (1994) Monitoring system for hydrologic
evaluation of landfill covers, Geotech. Test. J.,
17, 138-149.
Beven K. and Germann P. (1982) Macropores and
water flow in soils, Water Resour. Res., 18,
1311-1325.
Fredlund D.G. and Xing A. (1994) Equations for
the soil-water characteristic curve. Can.
Geotech. J., 31, 521-532
Geo-slope (2004) VADOSE/W 2004 user’s
manual. GEO-SLOPE International Limited,
Calgary, Alberta, Canada.
German P. and Beven K. (1981) Water flow in
soil macropores 1. An experimental approach,
J. Soil Sci., 32, 1-13.
Hagedorn F. and Bundt M. (2002) The age of
preferential flow paths, Geoderma, 108, 119-
132.
Johnson D.B. and Hallberg K.B. (2005) Acid mine
drainage remediation options: a review, Sci.
Total Environ., 338, 3-14.
Kelln C. Barbour L. and Qualizza C. (2007)
Preferential flow in a reclamation cover:
hydrological and geochemical response, J.
Geotech. Geoenviron. Eng., 133, 1277-1289.
Khire M.V. Benson C.H. and Bosscher P.J. (2000)
Capillary barriers: Design variables and water
balance, J. Geotech. Geoenviron. Eng., 126,
695-708.
Song and Yanful 425
Kung K-J.S. (1990) Preferential flow in a sandy
vadose zone: 2. mechanism and implications,
Geoderma, 46, 59-71.
Ogorzalek A.S. Bohnhoff G.L. Shackelford C.D.
Benson C.H. and Apiwantragoon P. (2008)
Comparison of field data and water-balance
predictions for a capillary barrier cover, J.
Geotech. Geoenviron. Eng., 134, 470-486.
Perret J. Prasher S.O. Kantzas A. and Langford C.
(1999) Three-dimensional quantification of
macropore networks in undisturbed soil cores,
Soil Sci. Soc. Am. J., 63, 1530–1543.
Selker J.S. Steenhuis T.S. and Parlange J.-Y.
(1992) Wetting front instability in
homogeneous sandy soils under continuous
infiltration, Soil Sci. Soc. Am. J., 56, 1346-
1350.
Simunek J. van Genuchten M.Th. And Sejna M.
(2006) The HYDRUS software package for
simulating the two- and three-dimensional
movement of water, heat, and multiple solutes
in variably-saturated media. Technical manual
Version 1.0. PC Progress, Prague, Czech
Republic.
Song Q. and Yanful E.K. (2010a) Effect of
channelling on water balance, oxygen
diffusion and oxidation rate in mine waste rock
with an inclined multilayer soil cover. J.
Contam. Hydrol., 114, 43-63.
Song Q. and Yanful E.K. (2010b) Laboratory and
numerical modeling of water balance in a
layered sloped soil cover with channel flow
pathway over mine waste rock. Environ. Earth
Sci., DOI: 10.1007/s12665-010-0488-4.
USG (1997) The design of soilcover systems.
SoilCover Version 4.0 User Manual.
Unsaturated Soils Group, Dept. of Civil Eng.,
University of Saskatchewan, Saskatoon,
Saskatchewan, Canada.
Walter M.T. Kim J.-S. Steenhuis T.S. Parlange J.-
Y. Heilig A. Braddock R.D. Selker J.S. and
Boll J. (2000) Funnelled flow mechanisms in a
sloping layered soil: laboratory investigation,
Water Resour. Res., 36, 841-849.
Yanful E.K. (1993) Oxygen diffusion through soil
covers of sulphidic mine tailings, J. Geotech.
Eng., 119, 1207-1228.
 Study On Gas Transport And Water Retention In Landfill Cover Soil:
A Sri Lankan Case Study
R.H.K.. Ranasinghe
Graduate Student, Department of Civil Engineering University of Moratuwa, Moratuwa, Sri Lanka ,
(kaushalyaranasinghe@gmail.com)
U.P. Nawagamuwa
Senior Lecturer, Department of Civil Engineering, University of Moratuwa,Moratuwa, Sri Lanka,
(udeni@civil.mrt.ac.lk)
P. N. Wickramarachchi
Graduate Student, Faculty of Civil and Environmental Engineering, Saitama University, Saitama, Japan
(praneeth1977@yahoo.co.uk)
K.Kawamoto
Associate Professor, Faculty of Civil and Environmental Engineering, Saitama University,, Japan,
(kawamoto@mail.saitama-u.ac.jp)
T.Komatsu
Professor, Faculty of Civil and Environmental Engineering, Saitama University, Saitama, Japan,
(komatsu@mail.saitama-u.ac.jp)

ABSTRACT In Sri Lanka, engineered landfills are not usually available and the common procedure is to
cover the waste with a good cover soil without providing gas venting facilities. Hazardous gaseous and liquid
landfill emissions are major problems of landfills. Hence cover soil of the landfill plays a major role in
emission of landfill gases, because once the solid waste is covered with soil these gases are released to the
atmosphere with high pressure, through this cover soil. Therefore, studying the cover soil parameters are of
paramount importance in evaluating its future gas diffusion. The Soil gas diffusion coefficient (Dp) and Air
permeability (ka) govern the transport and emission gases in the unsaturated zone. In this study, soil gas
diffusivity (Dp/Do, Do is gas diffusion coefficient in free air) and ka was measured for the soil of Maharagama
landifill final cover soil at different soil water matric potentials.

INTORDUCTION solid waste is ultimately converted to gaseous and

Waste is an unwanted material arising from
human and animal activities, which is in solid
form. Generation of municipal solid waste (MSW)
increases with rapid urbanization due to
accelerated socio economic development.
Inadequate waste management and disposal
practices combined with the tropical climatic
influence results in increasing environmental
problems (Visvanathan et al, 2004).
The generated solid waste from various places
finally ended up in waste disposal sites. Once
municipal solid waste is placed in such a disposal
sites, a complex sequence of biologically,
chemically and physically mediated events occur
related to hazardous gaseous and liquid landfill
emissions. A Significant fraction of the
biodegradable portion of the municipal
liquids as end products during the predominately
anaerobic stabilization of solid waste organic
fractions.
Landfill gas (LFG) is typically 40 to 60 %
methane, with the balance being mostly carbon
dioxide. Various trace gases such as hydrogen
sulfide, water vapor, ammonia, and a variety of
volatile organic compounds (VOCs) are also
found in land fill gases. (Cooper et al, 1992). It is
very important to study the landfill gases because
most of these gases are toxic, capable of polluting
atmosphere and give bad odor, health problems,
pollute the water bodies and the main problem is
green house effect. Methane is one of the most
effective green house gases and many studies has
investigated that landfills are one of the major

source, of emitting green house gases to the
atmosphere.
However, in Sri Lanka most of the waste disposal
sites are uncontrolled waste disposal sites (without
gas venting facilities and no landfill liner system
to prevent contamination of groundwater from the
leachate). Therefore, the environmental pollution
from Sri Lankan landfills is very significant. Since
there is no gas venting and collecting facilities
provided, the gases generated inside the landfills
are released to the atmosphere through the final
cover soil. Hence the final cover soil of the landfill
is playing a major role in emitting the landfill
gases, because once the solid waste is covered
with a cover soil, these gases are releasing to the
atmosphere with high pressure, through this final
cover soil.
This paper discusses the effects of compaction and
soil moisture content on gas transport and water
retention in landfill cover soil. In Sri Lanka, there
are several locations where waste has been
covered with a cover soil subsequently and these
locations have been used for different purposes.
Maharagama waste disposal site is such a location
where a mini cricket ground has been constructed
for the local residents. There had been a waste
layer of 3-4 m height and for the construction,
waste layer was first compacted. Then 3-4 layers
of soil (Laterite) were compacted and finally a 1.5
m thick cover had been placed and compacted
above the waste layer. In this study Dp/D0 and ka
were calculated based on the measurements from
landfill cover soil at Maharagama waste disposal
site.
LITERATURE REVIEW
Accurate description of soil-gas transport
parameters in unsaturated soil profiles is needed
when investigating the emission of greenhouse
gases to the atmosphere as discussed by Osozawa
& Hesegawa, (1995). The transport of gases in soil
is basically controlled by diffusion and convection
in the air-filled pore space where gases can only
move freely through the soil. Dp governs the gas
transport due to concentration gradient as
Ranasinghe et al. 427
proposed by Petersen et al. (1996). If the soil gas
transport is taken place as a result of pressure
difference, that gas movement is called as
convection flow. To model the convective gas
transport, the parameter, ka is used. Soil-gas
transport processes in a natural soil profile are
largely governed by variations in soil-water
content (water retention), soil texture, and organic
matter content (Moldrup et al. 2004). Knowing the
variations of Dp and ka at a field scale is vital in
simulating the fate, transport and emissions of
green house gases and volatile organic compounds
in soil and vadose zone (Resurreccion et al, 2007).
The ka value provides useful information about
soil structure, and is used for characterizing soil
pore geometry. The ka value is related to Dp, and
these two transport parameters have been linked
together to express air-filled pore connectivity and
equivalent pore diameter, (Ball, 1981 and Moldrup
et al., 2001). Dp/Do and air filled porosity (H) were
used to describe the continuity and tortuosity of
micro pores in the soil by Ball (1981). Soil water
matric potential (< or pF=-log (<)) provides a
measure of soil moisture status and is directly
linked to the soil water content (T) and
consequently to H through the soil water
characteristic curve and soil water characteristic
models (Ex Campbell, 1974).
MATERIALS AND METHODS
Basic soil tests such as particle size distribution
test, soil pH, specific gravity test, electrical
conductivity and standard Proctor compaction test
were done for the soil samples collected from the
Maharagama landfill cover soil in order to identify
the soil type and the other physical and chemical
properties of the soil.
Soil samples collected from the site were
compacted (according to ASTM standards) at
different moisture contents. From the Proctor
compaction mould, two 100 cm3 core cutter
samples were collected at each moisture levels
These collected samples were tested for ka and Dp
(normal compacted samples). When the
428 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

measurements were done samples were saturated RESULTS AND DISCUSSION

and then drained in to different pF conditions. At
each moisture content ka and Dp measurements
were taken (pF controlled samples).In order to
find the effect of compaction and the moisture
content of the sample on gas transport parameters,
the samples were compacted in to different
compaction levels and different pF conditions
were. The ka and Dp were measured in all the
samples.

The ka was measured by the modified field air

permeameter proposed by Iversen et al. (2001).
For the calculation of the ka, Darcy’s law was used.
The DP was measured by using the diffusion
chamber developed by Currie (1960a) and
Fig.1. Variation of ka with H and comparison of
recommended by Rolston & Moldrup (2002). existing power law models with the observed data.
Oxygen was used as tracer gas and measured as a
(For colour figure, refer to CD)
function of time in the diffusion chamber.
In controlling pF of the samples, two apparatus
were used. For small pF values (usually < 2.0)
sand box apparatus was used and for higher values,
pressure plate extractors were used.

Methane concentration of the Maharagama landfill

was measured using a laser methane detector, and
using these data, methane concentration contours
was produced.
Power-law type models for ka can be expressed in
general form (Hamamoto et al. 2009; Moldrup et
al.1998) as,
K
ka § H ·
¨¨ ¸¸ (1)
k a ,100 © H 100 ¹

ka,100 and H100 are soil air permeability and soil air
content at 100 cm soil water matric potential
respectively. Where ƾ represents the combined
Fig.2. Variation of Dp/D0 with H
effects of tortuosity and connectivity of air-filled
pores. Kawamoto et al. (2006) found that ƾ is a
function of the slope of soil-water characteristic
curve in log-log coordinate system.

TABLE 1. Composition and physical properties of
Maharagama waste disposal site final cover soil
Particle size fraction (%)
Gravel Sand Silt Clay
(4.75- (0.075-
( > 4.75 (<0.005
0.075 0.005
mm) mm)
mm) mm)
10 40 35 15
Figure 1 shows the ka variation with the H and
comparison of existing power law models with the
observed data. These models are of the same
power law type model with different K
An increase in bulk density due to soil compaction
consequently decreases the continuity of air-filled
pores primarily in the inter-aggregate pore space
region. In this research, the ka and Dp/Do was
measured as a function of H.
3
For the low soil air contents (H < 0.1 cm /cm ) all 3
three power law models are fitted well. However,
with high soil air contents, power law models are
deviating. Modrup (1998) model over estimates
the experimental values while other two models
underestimate the experimental values. Further,
the Mualem (1976) model shows the minimum
deviation. When b=14.5 curve is fitted with the
observed values.
Figure 2 shows the Dp/Do variation with H.
According to the figure 2 the Dp/Do is
monotonically increase with the H. There is no
very significant difference between the pF
controlled samples and the normal compacted
samples. It implies that Dp/Do is not depends on
Soil structure properties.
Even though the methane concentration of
Maharagama fill site was measured, the values
were very much closer to the atmospheric
Ranasinghe et al. 429
Particle Bulk Total
Soil density density porosity EC
pH
texture
ȡs (g/cm3) ȡb (g/cm3) Ɏ mS/m
Silty
Sand 2.77 § 1.90 § 0.35 5.4 32
methane concentration except at two points. These
two high concentration points (hotspots) are very
closer to the lamp posts which were driven up to
the waste layer. Further lower compaction of soil
was observed surrounding the posts. It can be
assumed that these points are functioning as
ventilation pipes since the remaining ground is
well compacted.
CONCLUSION
ka and Dp/D0 were measured as a function of H in
different T. In normal compacted samples, ka
initially increased with H and when H is
correspondent to field water content (|0.1
cm3/cm3), its variation was not so significant.
Finally, ka was increased significantly with drier
condition. These all scenario may suggest a
change in the structure formation with T. In the
case of pF controlled samples, ka increased
exponentially. Dp/Do increased with H and power
law model fitted well. Finally, it can be conclude
that the Maharagama waste disposal site final
cover soil is a very less gas exchangeable material
but a good capping soil.
REFERENCES
Ball, B. C. (1981) Pore characteristics of soils
from two cultivation experiments as shown by
gas diffusivities and air permeabilities and air-
filled porosities. J. Soil Sci. 32:483-498.
Campbell, G. S.(1974) A simple method for
determining unsaturated conductivity from
moisture retention data. Soil Sci. 117:311-314.
430 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Cooper C.D., Reinhart D.R., Seligman D., Keely
D., and Rash F., (1992),eport # 92-2,Civil and
Environmental Engineering Department,
University of Central Florida ,pp 11-14
Currie, J.A. (1960a). Gaseous diffusion in porous
media. Part 1.- A non-steady state method. British
J. Appl. Phys. 11: 314-317.
Hamamoto S, Moldrup P, Kawamoto K, Komatsu
T(2009) Effects of particle size and soil
compaction on gas transport parameters in
variably-saturated, sandy soils.Vadose Zone
Journal (in press).
Iversen B. V., Schjønning P., Poulsen T. G., and
Moldrup P. (2001). In-situ, on-site and laboratory
measurements of soil air permeability: Boundary
conditions and measurement scale. Soil Sci.
166:97-106.
Kawamoto K, Moldrup P, Schjonning P, Iversen
VB, Komatsu T, Rolston DE (2006) Gas transport
parameters in the vadose zone: Development and
tests of power-law models for air permeability.
Vadose Zone Journal 5, 1205-1215.
Moldrup P, Poulsen TG, Schjonning P, Olesen T,
Yamaguchi T (1998) Gas permeability in
undisturbed soils: measurements an predictive
models. Soil Science 163, 180-189.
Molderup, P.,Olsen T., Yoshikawa S., Komatsu
T., and Rolston, (2004). Three porosity model for
predicting the gas diffusion coefficient in
undisturbed soil. Soil Sci. Aoc. Am J. 68: 750-
759.
Moldrup, P.,Olsen T. , Komatsu t., Schjønning, P.
and Rolson, D. E., (2001). Tortousity Diffusivity
and permeability in the soil liquid and gaseous
phases. Soil Sci. Soc. Am. J 65:613-623
Mualem, Y. (1976). "A new model for predicting
the hydraulic conductivity of unsaturated porous
media". Water Resources Research 12: 513–522.
Osozawa, S., and Hasegawa.S. (1995). Diel and
seasonal changes in carbon dioxide concentration
and flux in an Andisol. Soil Sci. 160: 117-124.
Petersen, L. W., El-Farhan Y. H., Moldrup P., D.
Rolston E., and Yamaguchi T.(1996). Transient
diffusion, adsorption, and emission of volatile
organic vapors in soils with fluctuating low water
contents. J. Environ. Qual. 25:1054–1063.
Resurreccion A.C., Kawamoto K., Komatsu T.,
Molderup P., Ozaki N., and Rolston D. E., (2007)
Gas transport parameters along field transects of a
volcanic ash soil. Soil Sci.vol. 172: 3-15.
Rolston D.E. and Moldrup P. (2002). Chapter 4.3.
Gas Diffusivity,In: J.H. Dane and G. C. Topp
(eds.).
Methods of Soil Analysis, Part 4, SSSA Book
Ser.5, ASA ans SSSA, Madison, WI, pp 1113-
1139
Visvanathan C., Tubtimthai O. Kuruparan
P.(2004) Influence of landfill top cover design on
Methane Oxidation: pilot scale lysimeter
experiments under tropical conditions: APLAS
Kitakyushu 2004, Third Asian-Pacific landfill
symposium, pp 387-394.
 Characterization and Decay of Organic Matter Content of MSW in a
Geotechnical Perspective
C. C. Gomes
COBA, S.A (crgomes@iol.pt)
M. L. C. Lopes
Faculty of Engineering of Porto University ( lcosta@fe.up.pt)

ABSTRACT This paper presents the results of total and degradable organic matter content of MSW
determined by calcination at high temperatures and by chemical oxidation with hydrogen peroxide. The
results, analyzed according to the average age of the samples of MSW, allow the establishment of laws of
organic matter decay. From the proposed models and the velocity of the degradation processes is also
analysed the evolution of the constant of hydrolysis, kh’, taken as a parameter that controls the rate of these
processes.

INTRODUTION than that of easy or medium biodegradability and

MSW’s properties (physical, chemical and
mechanical) change over time mainly due to
biodegradation processes. In a geotechnical
perspective, the amount of organic matter of the
waste and its degree of decomposition strongly
affects the mechanical behaviour of the deposited
waste, namely in terms of the magnitude and rate
of settlements.
With time, most of MSW are converted into inert
components, so, one of the direct links between
the processes of biodegradation and the
composition of the waste is a decrease in the
levels of degradable organic matter with ageing
of waste.
Different methodologies can be used to
characterize the organic matter content of the
MSW, however each one has different destruction
capacities of the organic matter and so its
applicability and results comparation should be
taken carefully.
Considering the geotechnical classifications of
MSW, whose main objective is to group materials
with similar geotechnical behaviour, it seems
essential to differentiate the total organic matter
also analyse that in terms of the age of the waste.
SITE AND WORKS DESCRIPTION
The results presented in this paper come from a
larger research carried out in the landfill of Sto
Tirso (North of Portugal) between 2001 and 2007.
The main target of this research was the
evaluation of the waste biophysical, chemical and
mechanical properties and their evolution with
waste ageing, as well as, the characterization of
the deformational landfill’s behaviour during
time.
This sanitary landfill was filled between 1995 and
January 2005 with some direct waste (10%) and
mostly with pre-treated waste (90%, from
composting and triage), corresponding to a total
income of around 700 000 tonnes. The lower
levels of the landfill also included waste from an
old dump, located in the place where the landfill
was built (untreated waste).
The landfill have 1:1.5 (v:h) to 1:2 (v:h) lateral
slopes, with 5 m to 6 m height, limited by 3 m to
5 m width bench. The majority of the waste was
deposited in successive layers (0.5 m to 1.0 m
height), through small width rectangular areas
432 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
transversal to the landfill, till reaching one bench
level. After the deposition of each small area the
deposition go on in a similar adjacent area and
only the top of the waste (at each bench level)
was covered with soils. So, it could be considered
that the waste between two consecutive benches
has the some average age for each point of the
landfill (designated in the paper by layer).
Among others, the research included different
laboratory tests (physical and mechanical
characterization of waste and chemical
characterization of leachate) and landfill
monitoring (with earth and pore pressure cells,
temperature cells, open tube piezometers,
inclinometers tubes with magnetic settlement
plates or spiders and topographic control marks).
All the works were concentrated in three different
areas (A, B and C, designated in the paper by
station) each one corresponding to different ages
of waste (A youngest; B intermediate; and C
oldest waste). Details of this study are presented
in Gomes (2008).
The monitoring data, namely from earth pressure
cells and from topographic surveys, and the
landfill owner’s data allow the reconstitution of
the landfill constructive history and the evolution
of parameters such as waste and leachate height,
total and layers settlements, total and effective
stresses, temperatures, volumetric unit weight,
among others. In that way, it was possible to
estimate, with an acceptable error, the ages of
each deposited layer and, consequently, the
average age of the collected samples.
ORGANIC MATTER CONTENT
General Comments
In terms of geotechnical characterization, the
majority of the reported values of organic matter
content are determined through visual
identification of organic materials, by high
temperatures combustion (also designated by
calcination or ignition loss) or, rarely, by
chemical oxidation. Also, these results are rarely
analysed in terms of the age of the waste.
The nutritional value of the waste as well as the
determination of the main classes of organic
components (lipids, proteins, etc) or the
molecular and structural information of the
organic components, can also be used to follow
the evolution of the organic matter with waste
maturation, but, in generally, these are used only
in detailed analyses done with others proposes
than that of the waste geotechnical
characterization.
The characterization of organic matter through
visual identification is only possible in fresh
waste, while for older waste one of the other
methodologies should be used (calcination or
chemical oxidation), as proposed by Brauer et al.
(2005).
The incineration at high temperatures determines
the total organic matter content (also designated
by volatile solids (VS), total volatile solids (STV)
or content of "ashes"). That’s why the
consideration of the organic matter content
determined by calcination as an indicator of
biodegradation of organic fraction of waste is, by
rule, inadequate as it contains materials of very
different degradability, from readily to very slow
degradability. To characterize this seems
preferable to use a chemical oxidation method,
which is more related with the degradable
fraction of organic content (Brauer et al., 2005).
The chemical oxidation can be performed with
different strong oxidizing agents, in which the
most relevant are the potassium dichromate
(K2Cr2O7), the sodium hypochlorite (NaOCl), the
potassium permanganate (KMnO4) and the
hydrogen peroxide (H2O2). The first 3 methods
are usually carried out in samples of leachate or
by leaching the samples of MSW (eluate), while
the attack by hydrogen peroxide (H2O2) can be
made in samples of the liquid or solid fraction of
the MSW. Although these methods are proposed
with the same purpose, the oxidizing potential
developed by each one is different, giving them
different capacities for organic matter destruction.
Moreover, the practical application of methods
 Gomes and Lopes 433

that use samples of the liquid fraction (leached or and then the organic matter content. Figure 2
eluate) shows some difficulties, especially when shows the variation of moisture content (by
the determination is made by titration with visual drying at 60° C) with the average age of waste
indicator. (corresponding to the difference between the date
of sampling and date of half-time of deposition of
Laboratory Tests Results the layer in question).
160
Based on the results of particle size distribution

Moisture content (Wd, %)

140 Station A
and type and form classification of 27 waste 120 Station B
samples collected from boreholes in the 3 stations 100
of the landfill, was noticed that the ageing of the 80 Station C
waste results in a decrease of particle size and 60
quantity of fabrics and plastics and in an increase 40 y = -0.026 t + 117.768
2
20 R = 0.512
of the amount of mineral components and non-
0
identifiable materials (fine fraction). In
0 500 1000 1500 2000 2500 3000 3500 4000
consequence, the quantity of materials with Average age (days, t)
sheet/strip and wire/tube forms also decreases, Fig.2 Variation of moisture content with age of waste.
while the amount of material with granular form (For colour figure, refer to CD)
increases and flattened remains relatively
constant.
In this research two different methods were used
According to the solid waste classification to characterize the organic matter content:
proposed by Grisolia et al. (1995) or Landva & calcination at 550º C ±50º C (OMT) and chemical
Clark (1990), the amount of inert or inorganic oxidation of the solid fraction with hydrogen
components increase and the amount of highly peroxide (OMOXID). In the first case, the tests were
deformable and biodegradable (readily and conducted in accordance with the ASTM D2974
slowly) elements decrease with ageing (Fig.1). standard (developed for organic soils) and in the
second one following the methodology proposed
80
by the Portuguese Road Department (EP) in S.9-
MSW geotechnical classification (%)

Grisolia et al. (1995) Landva & Clark (1990)

70
53 standard.
60 Initial
Station A
50
Station B The chemical oxidation by potassium dichromate
40 Station C was also tried but some difficulties were found in
30 the identification of the change of colour (as in
20 general the leachate is a very dark colour liquid,
10 the identification of the change in colour of the
0
indicator requires a large dilution of the previous
Putrescible +
deformable

fine fraction
putrescible

sample).
Biodegrad.+
fine fraction

Inorganic
Inert /
stable

Higly

Non-

Figure 3 shows the relationship of total (OMT)

and oxidable (OMOXID) organic matter content
Fig.1 MSW geotechnical classification according to
with moisture content (Wd). In spite the range of
Grisolia et al. (1995) and Landva & Clark (1990).
(For colour figure, refer to CD) values have similar shape, the magnitude of the
values of total organic matter content are much
Smaller samples were taken from the previous higher (on average, about 6 times higher for
ones to characterize moisture and organic matter samples from Station A, 8 times for Station B and
content. At first, moisture content was analysed 9 times for Station C). The range of values of
434 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

total organic matter is similar to that proposed by 90 18

Landva & Clark (1990), although the authors do 80 Total OM 16

Degradable org. matter (%)

Degradable OM

Total organic matter (%)

not report the method used. 70 14
OMT = -7.482Ln (t) + 101.839
60 2
12
R = 0.490
100 18 50 10
OMT =21.3Ln (Wd) - 28.4

Degradable org. matter (%)

90 16
Total organic matter (%)

40 8
80 OMOXID=3.8Ln (Wd) - 5.0
14 30 6
70 12
60 20 4
10 OMOXID = -3.090Ln (t) + 29.025
50 10 2 2
8 R = 0.904
40 0 0
30 6
Landva & Clark (1990) 4 0 500 1000 1500 2000 2500 3000 3500 4000
20 Total OM Average age (days, t)
10 OMOXID=11.4Ln (Wd) - 47.7 Degradable OM 2
OMT =63.3Ln (Wd) - 264.9 Proposed correlation Fig.4 Variation of organic matter contents with age of
0 0
0 50 100 150 200 waste. (For colour figure, refer to CD)
Moisture content, Wd (%)
Fig.3 Variation of organic matter contents with According to studies of Straub & Lynch (1982, in
moisture content. (For colour figure, refer to CD) Ibarra, 1994) the speed of the decomposition
processes of waste is dependent on the
The influence on the moisture content of hydrolyzing rate and can be expressed as:
parameters, such as, the average precipitation of
the month preceding the collection of samples, d ( BDM )
k h ' BDM 
(1)
the deep of sampling and the relative amount of dt
some components of waste, namely those with Where BDM is the solid biodegradable matter, kh’
high capacity to absorb water, were also studied. is the hydrolysis constant modified by water
The results enhance that the moisture content content and t is the elapsed time.
seem to be more influenced by the composition
(namely the relative amount of textiles and According to these authors the value of kh’
paper/cards, although these last ones were not depends on the characteristics of the deposited
tested due to their small quantity in the samples) material, the temperature, the pH, and basically
and stage of decomposition of waste (expressed on the amount of water present in the landfill
by the content of organic matter) than by depth or during the degradation of organic matter.
by precipitation.
The values of the hydrolysis constant of MSW
The variation of both organic matter content with from Santo Tirso Landfill were determined using
ageing shows a logarithmic decreasing equations 2 or 3, resulting from the integration of
relationship, expressed by the equations of Fig 4. equation 1 in order of time (between t=0 and t).
These decreasing is more clear in the case of The correspondent values are presented in Fig 5
oxidable organic matter (OMOXID; R2 • 0.9), and Fig 6 shows the values of kh’ defined by the
which shows the sensitiveness of this parameter correlations of Fig 5.
concerning the evolution of biodegradable
organic matter of waste. MOOXID t  (2)
ln  k h '.t
MOOXID t 0 

or

MOOXID t  MOOXID t 0 .e  kh '.t (3)

 Gomes and Lopes 435

Average age (days, t) These values of kh’ fall within the range provided
0 500 1000 1500 2000 2500 3000 3500 4000 by Williams et al. (1987, in Ibarra, 1994) and
0.0 Park & Lee (1997) through tests carried out in
ln(MOOXID(t) / MOOXID(t0))

lysimeters, where kh’ ranged from 0.1 to 0.0008

-0.5 y = -0.0004.t - 0.8533 = -kh'.t days-1 and between 0.006 and 0.0002 days-1,
2
R = 0.6857 respectively. Also Ibarra (1994) and Gonzalez
-1.0
(1995) obtained similar values (from 0.003 to
-1.5 0.0003 days-1 and from 0.001 to 0.0004 days-1,
respectively) by adjusting the monitoring data of
-2.0 different sections of the sanitary landfill of
-2.5 Meruelo (Spain).

CONCLUSION
Average age (days, t) The characterization of the degradable organic
0 500 1000 1500 2000 2500 3000 3500 4000 matter content by the attack by hydrogen
0.0 peroxide or other similar method (chemical
MOOXID(t) / MOOXID(t0 )

0.2 oxidation) appears to be an important tool in the

0.4
assessment of the biodegradability of a landfill.
-0.0004 t -kh' t
0.6 y = 0.426 e =e
Although in the reviewed literature the
2
R = 0.686
0.8 application of this type of test in MSW is
1.0 practically non-existent, it is a simple
methodology, easily applicable in any phase of
1.2
the landfill and the results look promising.
Fig.5 Decay of degradable organic matter with age of The applicability of the results obtained from
waste. these tests in the prediction of MSW landfill
settlement was tested during the research carried
0.0035 out in the landfill of Santo Tirso. For such was
0.0030
used the Meruelo's settlement prediction model
(Ibarra, 1994; Gonzalez, 1995), with some
k h ' (days )

0.0025
-1

y = -0.0005 Ln (t) + 0.0048

2
adjustments (Gomes, 2008).
0.0020 R = 0.9580
0.0015 In this simulation the initial values of degradable
0.0010 organic matter, BDM(t1), and kh'(t1) (where t1 is
0.0005 the average age of the landfill at the beginning of
0.0000 the prediction) were determined from the
0 500 1000 1500 2000 2500 3000 3500 4000 correlations presented in Fig 4 to Fig 6 and the
Average age (days, t) results compared with the obtained through the
Fig.6 Variation of kh’ with age of waste. monitoring of various sections of the landfill and
considering different stages of the life of the
According to these data, kh’ decreases landfill (both during and after the deposition).
logarithmically with ageing of waste, ranging The results obtained through the adaptation of
from 0.0048 days-1, to t=1 day, until relatively this model were similar to those recorded by the
stable values of around 0.0006-0.0007 days-1 to monitoring and the deviations, for any of the
MSW older than about 2400 days (§ 6.5 years). analyzed situations, were lower than those
determined by the application of others settlement
436 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
prediction models (11 settlement prediction
models were considered).
Thus, it appears interesting to verify the
applicability of this type of research in other case
studies. A more consistent development of laws
of decay of organic matter, relying on simple
approaches, such as the presented, and verified by
more detailed studies (as lysimeter tests or other
types of bioreactors), could permit the
establishment of models of variation of the
organic matter content according to the
environmental conditions of landfills and age of
deposited waste.
The use of decay laws of the degradable organic
matter (similar to the presented) and parameters
that influence these processes (such as the
hydrolysis constant), if properly confirmed, could
be a key tool in predicting the biological
settlements for any period of life of a landfill
considered with interest.
REFERENCES
ASTM D2974-07a (2007) Standard Test Methods
for Moisture, Ash, and Organic Matter of Peat
and Other Organic Soils. American Society for
Testing and Materials, ASTM, 4 p.
Brauer C., Achour F., Bayard R., and Gourdon,
R. (2005). Characterization of organic matter
during maturation of municipal solid waste in
order to identify chemical parameters indicating
stabilization. Proceedings of the International
Workshop “Hydro-Physico-Mechanics of
Landfills”, Grenoble, France, 21-22 March
2005, 4 p.
Gomes C. G. (2008). Study of landfill behaviour.
Municipal solid waste physical, biochemical
and mechanical characterization. PhD Thesis,
Faculty of Engineering, Porto University,
Portugal (in Portuguese), 536 p.
Gonzalez J. H. P. (1995). Geotechnical behaviour
of controlled landfills of municipal solid waste.
PhD Thesis, School of Inginieros de Caminos,
Canales y Puertos, Cantabria University, Spain
(in Spanish), 587 p.
Grisolia M., Napoleoni Q. and Tancredi G.
(1995). Contribution to a technical classification
of MSW. Proceedings of the 5th International
Landfill Symposium, Cagliari, Sardinia, 2-6
October 1995, Vol.2, 703 – 710.
Ibarra A. A. (1994). Settlement model of
controlled landfills of municipal solid waste.
Master Thesis, School of Inginieros de
Caminos, Canales y Puertos, Cantabria
University, Spain (in Spanish), 138 p.
JAE S.9-53 (1953). Standard of JAE for
determining the percentage of organic matter in
soils. Junta Autónoma de Estradas (JAE),
Portugal, (in Portuguese)4 p.
Landva A. O. and Clark J. I. (1990). Geotechnics
of waste fill. Geotechnics of Waste Fills -
Theory and Pratice, Landva/Knowles editors,
ASTM – STP 1070, Philadelphia, 86 – 103.
Park H.I. and Lee S.S. (1997). Long-term
settlement behaviour of landfills with refuse
decomposition. Journal of Solid Waste
Technology and Management, Vol.24, Nº 4,
159-165.
Straub W. and Lynch D. (1982). Models of
landfill leaching: organic strength. Journal of
the Environmental Engineering Division, ASCE
108, 251-268.
Williams N., Pohland F., McGowan K., and
Sounders M. (1987). Simulation of leachate
generation from municipal solid waste.
Environmental Protection Agency, Hazardous
Waste Engineering Research Laboratory,
EPA/600/2-87/059, Cincinnati, United States.
 Characterization of the Shear Strength of Municipal Solid Waste for
Landfill Design
Sukhmander Singh
Nicholson Family Professor of Civil Engineering, Santa Clara University, Santa Clara, California
Mark Wooster
Senior Research Assistant, Civil Engineering Department, Santa Clara University, Santa Clara, California

ABSTRACT The paper presents, an analysis of available published data on the shear strength of Municipal
Solid Waste. The cohesive behaviour of municipal solid waste is critically examined. Data is plotted and
sorted by the method by which it is obtained. Scatter in the data is observed and discussed. A waste with
only frictional component is not a good predictor of shear behaviour. Case histories and lab data exhibited
large amounts of scatter. Neglect of highly comparessible nature of MSW in developing Mohr Coulomb
strength envelopes has been pointed out. The use of a plot of shear strength versus depth is recommended.

INTRODUCTION et. al., 2004). More recently, the effect of confining

Every day, the United States alone generates millions
of tons of municipal solid waste (MSW) that must be
processed. Landfilling is the least expensive method
of waste disposal and landfills are being built to
unprecedented heights. Landfill failures can expose
the public to a wide variety of health hazards and can
also create ecological and environmental disasters.
These and other reasons made the determination of
design parameters for landfills a growing field of
interest. Geotechnical parameters are important for the
design of each of the subsystems. This paper
examines the shear strength of MSW. This parameter
is important for both seismic and static slope stability.
Although much data has been published in past
decade, variability in the published data seems to
match with the variability in the landfill content from
place to place and time to time. Many question like the
applicability of Moahr Coulomb concept; and the
effect of age and compressibility on the evaluation of
strength parameter of MSW still remain unanswered.
Accordingly, no clear cut way for characterising the
shear strength of MSW has emerged. This paper
attempts to progress in this direction.
ALTERNATIVES FOR CHARACTERIZATION
OF MUNICIPAL SOLID WASTE
The shear strength of MSW is dependant upon
several variables. The type of materials in the waste
stream, the age of the waste, the degree of
compaction, the type and amount of daily cover all
can have an effect on shear strength (Thusyanthan
stress and the influence of the presence of fibrous
materials on the shear strain and shear strength of
MSW have been noted (Zekkos, et. al., 2008). The
ways in which shear strength parameters are
obtained also affect the values. Published shear
strength data has been obtained from direct lab
testing of samples, in-situ testing of landfills, back-
analyzing from failures, and lab testing on
simulated MSW. As a result of all these variables,
researchers have differed in their opinions about
how to best characterize MSW.
Many researchers have characterized shear
strength properties by proposing Mohr-Coulomb
strength envelopes derived from their
experimental data (Kavazanzian, 1999, Zekkos,
2005). Others reported results in terms of the same
parameters after obtaining the data from back-
calculations of failed landfills (Kavazanzian 2001,
Eid, et. al., 2000). Other researchers believed the
variability they observed in their samples was too
great for Mohr-Coulomb parameters to be usefully
developed for each sample tested (Siegel, et. al.,
1990). For example, Siegel, et. al, (1990) tested
samples from the OII landfill, perhaps the most
studied landfill, and made the observations that
“given the refuse variability deriving Mohr-
Coulomb angles of internal friction and cohesion
intercept for individual samples was deemed
inappropriate’. Singh and Murphy (1990)
suggested that Mohr Coulomb’s theory might not
be applicable to characterizing the shear strength
of MSW. The reason they gave is one that other
438 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
researchers have discussed, but none of them have
stated that it makes the Mohr-Coulomb parameters
irrelevant.
The cause of this controversy is that the soils that
form the foundation for the landfill fail at a much
lower strain than the MSW. In fact, the shear strength
of the MSW is not fully mobilized until strains of
around 30%. But, soil fails at much lower strains, and
so strain compatibility is a problem. However, data
for MSW at lower strains is still useful for design
purposes. Some researchers published stress-strain
curves for samples of MSW in shear; and the most
common form of data found in the literature is in the
form of Mohr Coulomb cohesion intercepts and
friction angles.
Cohesive Properties of Municipal Solid Waste
The strength of MSW in shear is similar to a soil in
many ways. It can be thought of for conceptual
purposes (and many argue for other purposes as well)
as a fibrous soil (Zekkos,2008). Direct shear tests by
Landva and others have shown that the shear strength
of MSW depends on the nature of the test (Qian et al.,
2002) .
MSW exhibits a form of behaviour in shear, which is
referred to as cohesion, but is seen by others as only
apparent cohesion. Some call this apparent cohesion-
‘adhesion’ (Qian et. al., 2002). This behaviour comes
about mainly because MSW is a hodgepodge of
different materials and different shapes. The interlock
of these odd-sized ‘grains’ causes the MSW to exhibit
behaviour similar to what is called cohesion in clay
soils. In clay, cohesion is the result of water and
associated electrical charges, although the critical
state concept considers cohesion even in clays
resulting from interlocking (Schofield, 2005). The
cohesion in the MSW is the result of mechanical
interlock. As MSW degrades with age the effect of
the interlock would decrease and the cohesion should
drop.. Accordingly, cohesion in MSW cannot be
treated as a constant parameter. Though, the root
cause of the behaviour differs from that in soils, the
observed behaviour is quite similar. Cohesion is
observed in MSW although it can be correctly termed
apparent cohesion. Cohesion has been observed in
data on actual waste than in model wastes. The
cohesive behaviour of MSW is also reinforced in
practice by the abundance of incidents in which
vertical or nearly vertical cuts remain stable for years
without any signs of failure (Qian et. al., 2002).
Data from Model Waste Experiments
Testing of artificial waste and model waste provided
data points with no cohesion (Thusyanthan et al,
2004).
Data from Case Histories
As was mentioned earlier, another method of
collecting shear strength data is back-calculations
based on case histories. Data reported in literature
from various case histories was plotted. In almost all
of the cases, the shear strength data was obtained
after a landfill failure. The most common method was
to assume a factor of safety equal to unity and back
calculate the shear strength of the MSW involved
using standard geotechnical analyses. In this study,
only case studies using a factor of safety equal to one
for failure analysis were reported. Other case studies
estimate the factor of safety and then back-calculate
shear strength parameters. However, the shear
strength values are very sensitive to changes in the
factor of safety, and the factor of safety can’t be
known with certainty unless failure occurs. The most
common details presented by the researchers, and the
techniques used to calculate the shear parameters are
given in Qian, et al., 2002 and Zekkos (2005) among
others.
It is of interest that every case history for which shear
strength was found yielded data with cohesion. The
only point on the y-axis is one for which the friction
angle was not reported. (The friction angle in this
case is not zero; it was plotted as zero because no
other information was available in this case.) It
should also be noted that the lowest cohesion is
reported by Seed and Boulanger (1992) in the case
histories, plotted above, and is the lower bound for
five case studies reported by Howland and Landva
(1992). All the case history data led to the appearance
that MSW does indeed have cohesive properties.

However, no clear trendline emerged from the data,
although a general decrease in cohesion as the friction
angle increases was observed.
3.3 Lab and In-situ Testing on Actual Municipal
Solid Waste
Far more data points were available for lab and in-situ
testing of MSW than for any other category of data.
When lab methodology is carefully controlled, this
category of data can also be quite accurate. This was
the data that most researchers in the past have
graphed to find their lower bounds and ranges of
acceptable parameters.
In figure 1, below, it was seen that different studies
have yielded widely different results. This was due to
the fact that the samples tested were disturbed to
different degrees, in the case of lab tests. The inherent
heterogeneity of MSW can also be responsible for the
scatter. There was the additional complication that
some samples were tested in triaxial machines while
others were tested in direct shear. There could also be
a disparity between tests done in the field and those
done under controlled conditions. All of the data that
specified depth, however, was for depths of less than
40 meters from the landfill surface.
Figure 1: Shear strenth parameters from Lab and In-
situ testing MSW
(For colour figure, refer to CD)
Singh and Wooster 439
From figure 2, large scatter in data is seen. It has
been recognized that factors such as age,
composition, test procedures can affect the value of
shear strength parameters of MSW. More recently
the dependency of Mohr Coulomb parameters on the
confining pressure has also been shown (Zekkos,
2005). An additional factor which has not been paid
much attention to is the highly compressible nature of
MSW. As can be seen in Figure 4 below, at high
normal stresses, the large compressibility of MSW,
would invalidate the attempt to use a single Mohr
Coulumb envelope for tests on samples at widely
varying confining stresses. Accordingly, the more
appropriate way is to develop a plot of shear strength
versus depth of MSW, especially for high landfills to
account for the effects of compressibility, unit weight
and high confining pressures.
Figure 2: Compression curves for the samples taken
from BH5 (Chen et al, 2008)
CONCLUSIONS
The shear strength of MSW is of increased
importance in landfill design due to the weak shear
interface between the geosynthetics in modern double
layer liners. MSW has a cohesive as well as a
frictional behaviour in shear loading. Published shear
strength data from models, case histories, back
calculations, and from lab or in-situ tests on actual
MSW was critically examined. It has been shown that
440 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
model or artificial waste is not a very good predictor
of shear behaviour. Case histories and lab data
exhibited large amounts of scatter, but lower bounds
and reasonable design parameters have been
suggested by the data. It is suggested that due to the
large compressibility of MSW at-high normal
stresses, a single Mohr Coulomb shear envelope for a
landfill may not be applicable. The use of strength
versus depth plot is more appropriate for
characterizing shear strength of MSW, especially for
high landfills.
REFERNCES
Chen, Y., Zhan, T.L.T., and ling, Wei-an, (2008),
Mechanical Properties of Municipal Solid Waste
from Suzhou Landfill in China, “Proceedings of
Geo-Congress 2008, ASCE New Orleans March
9-12, 2008.
Eid, H., et al, (2000). Municipal Solid Waste Slope
Failure, I: Waste and Foundation Soil Properties.
Journal of Geotechnical and Geoenvironmental
Engineering, 126-5, 2000.
Fasset, J., et al. Geotechnical Properties of Municipal
Solid Wastes and Their Use in Landfill Design.
Proc. WasteTech; 94 Charleston, SC, National
Solid Waste Management Association,
Washington D.C.
Howland, J.D. and Landva, A.O. (1992). “Stability
Analysis of a Muncipal Solid Waste Landfill.”
Proc. ASCE Specialty Conference on Stability
and Performance of Slopes and Embankments. II,
Berkeley, Cali., June 28- July 1.
Kavazanjian, et al. (2001). Municipal Solid Waste
slope Failure, I; Waste and Foundation Soul
Properties. Journal of Geotechnical and
Geoenvironmental Engineering, 127-9.
Kavazanjian, E, Jr., Matasovic, N. and Bachus, R.C.
(1999). Large-diameter static and cyclic
laboratory testing of Municipal Solid Waste.
Proceedings, Sardinia 1999, 7th International
Waste Management and Landfill Symposium, 4-8
October.
Landva, Knowles (editors) (1994). Geotechnics of
Waste Fills Theory and Practice. ASTM, Ann
Arbor, Michigan.
P. Pinto. Environmental Geotechnics. Proc. Third
International Congress on Environmental
Geotechnics. Lisbow, Portugal. 7-11 September.
Offprint. A.A. Balkema, Rotterdam, Brookfield.
Quian, X., et al, (2002). Geotechnical Aspects of
Landfill Design and Construction. Prentice Hall,
Upper Saddle River, New Jersey 07458.
Schofield, A; (2005) Disturbed Soil Properties and
Geotechnical Design: Thomas Telford, Publi-
shers, London, U.K.
Seed, R., R. Boulanger. Stability of and Performance
of Slopes and Embankments-II: A 25-Year
Perspective. Geotechnical and Special
Publication No. 31.
Singh, S., B.J. Murphy. Evaluation of the Stability of
Sanitary Landfills. Geotechnics of Waste Fills-
Theory and Practice, ATM STOP 1070, pp. 240-
258.
Thusyanthan et. al., (2004). Centrifuge Modeling of
Solid Waste Landfill Systems-Part I:
Development of a Model Municipal Solid
Waste. Geotechnical Testing Journal, 29-3.
GTJ1229.
Zekkos D; Bray, J.D., Stokoe K., Kavazanjian, E.,
Rathje, E., Athanasopoulos, G.A., Riemen, M.,
Matasovic, N., Lee, J.J., and Seos, B. (2008)
“Recent Findings on the Static and Dynamic
Properties of Municipal Solid Waste”
proceedings of Geo-Congress 2008; ASCE, New
Orleans. March 9-12, 2008.
Zekkos (Zeccos), D.P. (2005), “Evaluation of static
and dynamic properties of Municipal Solid-
Waste”, dissertation submitted in partial
fulfillment of the requirements for the degree of
Doctor of Philosophy, Department of Civil and
Environmental Engineering, University of
California, Berkeley, Fall. Also available at
http://waste.geoengineer.org
 Study on the Compressibility and Compression Mechanism for Municipal Solid
Waste in China
Z Y. Zhang
School of Civil Engineering and Architecture, Zhejiang Sci-Tech University, Hangzhou,
China (zhangzhenyinga@163.com)
D Z. Wu
School of Civil Engineering and Architecture, Zhejiang Sci-Tech University, Hangzhou,
China (wdz@hz.cn)

ABSTRACT By using a large scale compression testing instrument, the compressibility and compression
mechanism for municipal solid waste have been studied. A series of laboratory tests have been performed for
these test samples under different pressures. The results show that the settlement is affected by the initial
void ratio, the content of organic substance and the pressure applying on the waste. The settlement consists
of two parts, one part is due to the exert pressure, and the other part is due to the degradation of the organic
substance and the creep deformation. The compression index is 0.95 to 3.12.
KEYWORDS Municipal solid waste, stress compression, biodegradation, compression index

Since the municipal solid waste is a kind of highly

INTRODUCTION
Municipal solid waste (MSW) is a mixture of
wastes that are primarily of residential and
commercial origin (Neil Dixon, 2005). Some of
the components are stable while others degrade as
a result of biological and chemical processes
(Sandro L, 2002). It should be noted that the
composition of MSW varies from region to region
and country to country, and it has a close
relationship with the economic development level
(Wang & Nie, 2000). The content of organic
matter is the main ingredient in developed
countries and the content of inorganic substance is
the main ingredient in the developing countries. It
is a very concerning problem for people how to
deal with the waste (Zhang, 2005). Currently,
there are three methods to dealing with MSW:
composting, incineration and landfill, and landfill
is one of the most economic and feasible means of
disposing MSW (Hoe I. Ling et al, 1998).
Settlement estimation is a topic of concern in
MSW landfill management. Landfill capacity will
be increased if most settlement occurs during the
stage of filling. However, a large post closure
settlement may lead to surface pond, developing
of cracks, failure of the cover system(including
tearing of the geomembrane), and the gas
collection and the drainage pipes may be damaged.
asymmetric, extremely complex, multi-ingredient
and highly compressible material, it is very
difficult to study the compressibility and
compression mechanism of MSW. In order to
study this problem, the materials of MSW are
divided into three parts: the material that is easy to
be biodegraded, reinforced material that is
difficult to be biodegraded and incompressible
solid material. Basing on the study of properties of
MSW in Hangzhou, China (Zhang, 2010), the
laboratory test has been performed by using a new
developed large scale compression test instrument.
The testing results are analyzed, and the
compressibility of MSW is studied.
EXPERIMENTAL PROCEDURES
Testing Instrument
A new developed compression test instrument for
MSW is shown in Fig.1. The new compression test
instrument is composed of four parts: exerting load
system with hydraulic pressure, measuring system
with sensor devices for vertical displacement,
draining system for the water out of MSW and the
computer data processing system. The cross
sectional area of the testing cylinder is 0.3027
square meter. The diameter of the testing cylinder
is 0.621meter, and the height of the testing cylinder
is 0.300meter.
442 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Experimental Procedures
The materials of MSW are divided into three parts:
the material that is easy to be biodegraded,
reinforced material that is difficult to be
biodegraded and the incompressible solid material.
To investigate the influence of different ingredients
on the compressibility, three different proportions
of the three parts are selected, which are 50%, 15%,
35%; 65%, 10%, 25%; and 80%, 5%, 15%,
respectively. Before testing, the materials of MSW
are mixed, stirred uniform, and stewed for 24 hours.
The initial void ratio of the test specimen is 4.5.
Fig.1 New developed compression test instrument for The levels of the pressure applying on the test
MSW (For colour figure, refer to CD) specimen are 25, 50, 100, 200, 400 and 800kPa,
respectively. The testing time lasted for 24h for
Composition of the Test Material every level of the pressure, and the total testing
The materials of MSW are divided into three parts: time lasted for 144h. The procedure of the
the material that is easy to be biodegraded, compression test is described as follows: a) firstly,
reinforced material that is difficult to be applied load 25kPa on the test specimen, holding
biodegraded and incompressible solid material. the pressure by the hydraulic pressure system,
The material that is easy to be biodegraded recorded the settlement with the computer data
includes food, vegetable, fruit, paper, garden processing system. The compression test is
wastes, trees and wood; the reinforced material that performed and obeys the test criteria in China
is difficult to be biodegraded includes textile, (national regulation, 1999). b) 24h later, applied
rubber, plastics and animal bones; the the next load level 50kPa on the test specimen, and
incompressible solid material includes stone, sand, holding the pressure for 24h, recorded the
soil, brick, glass, ceramics and metal. When the settlement with the computer data processing
proportions of these three parts are 50%, 15% and system. c) Applied succeeding load levels 100, 200,
35%, the detail contents of the waste material are 400 and 800kPa on the test specimen in the same
listed in Table 1. method. After finishing the test on first test
specimen, the second and the third test specimen
TABLE 1 Composition of the Waste Material Used in are tested in the same method till the total data are
the Indoor Test (% by Weight) collected.
-------------------------------------------------------------------
Items Paper Vegetable Fruit Garden waste
TESTING RESULTS AND ANALYSIS
-------------------------------------------------------------------
Percentages 2 36 9 3
------------------------------------------------------------------- For the test specimen with the proportions of the
Items Plastic Rubber Bone Metal Brick three parts are 50%, 15% and 35%, the test results
------------------------------------------------------------------- are shown in Fig.2 and Table 2. It is easy to find
Percentages 3 3 4.5 1.8 2.2 that the compression strain increases with time,
------------------------------------------------------------------- and the increasing velocity of the compression
Items Stone Soil Ceramics Glass textile strain is very slow under the same pressure. When
------------------------------------------------------------------- the load is applied on the testing specimen, the
Percentages 2.2 25.2 1.8 1.8 4.5
compression strain is suddenly increases, and then
Total Percentage 100% it increase slowly. The water generated from the
Moisture content of original sample is 89% MSW is slowly drained out of the testing cylinder,
------------------------------------------------------------------- and then the compression strain is slowly increased.
 Zhang and Wu 443

drawn into a line. The slope of the back end line

is1.27, therefore, the compression index is 1.27,
which shows that the compressibility of MSW is
very big.

Fig.2 Relationship between compression strain ε and

time T for MSW

Table.2 shows that the total settlement is 160.3mm,

and is about 75% of the height of the test sample.
The total settlement is composed of three parts:
stress compression, consolidation compression and
creep compression. The consolidation compression
could be divided into two parts, one is due to the
discharging of the water from the void of the
municipal solid waste and the other is due to the
discharging of the water and the air which
generated from the degradation of organic
substance of municipal solid waste. The creep
compression is a settlement due to the holding-on
and rubbing each other in the materials of the
waste under the same stress. The amount of the
stress compression and the consolidation
compression are about 80% to 85% of the total
settlement, and the creep compression is about
15% to 20% of the total settlement.
Fig.3 Relationship between void ratio e and logarithm
pressure P for the first kind MSW
Fig.4 is for the test sample which the proportions
of the three parts are 65%, 10% and 25%. From
this figure, it can be found that the curve can be
drawn into a polygonal line. The slope of the back
end line is2.34, therefore, the compression index is
2.34, which shows that the compressibility of
MSW is also very big. At the last level of load, the
compression strain is very big, and the void ratio is
very small, and the minimal void ratio is 0.25.

TABLE 2 Settlement of Every Load Level

-------------------------------------------------------------------
Load (kPa) 25 50 100 200 400 800
-------------------------------------------------------------------
Settlement (mm) 51.7 73.2 113.2 135.7 149.7 160.3
-------------------------------------------------------------------

The relationship between void ratio e and

logarithm pressure P for three different proportions
samples are shown in Figs.3-5. It can be found that
it is a typical compression curve (e-log P curve).
Fig.3 is for the test sample which the proportions
of the three parts are 50%, 15% and 35%,
respectively. From this figure, it can be found that
the curve of front end and the back end could be
Fig.4 Relationship between void ratio e and logarithm
pressure P for the second kind MSW
Fig.5 is for the test sample which the proportions
of the three parts are 80%, 5% and 15%. From this
figure, it can be found that the curve can be drawn
into three lines. The slope of the middle line is 3.12,
and the slope of the back line is 0.95, which shows
444 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
that the compressibility of MSW is firstly
increased and then decreased. At the last level of
load, the compression strain is very big, and the
void ratio is very small, and the minimal void ratio
is closed to zero.
From figs.2-5, it can be found that the compression
strain is very big, and it is over 70%. The testing
process can be divided into three stages: initial
stage, drainage stage and creep deformation stage.
At the initial stage, the load is exerted suddenly,
and the settlement is increased rapidly. At the
drainage stage, the water generated from the MSW
is slowly drained out of the testing cylinder, and
then the settlement occurred. The settlement
increases with the weight of the water. At the creep
deformation stage, the settlement is slowly
increased under the same pressure on the testing
specimen, but it is not very big, not very small
either.
Fig.5 Relationship between void ratio e and logarithm
pressure P for the third kind MSW
CONCLUSIONS
In this paper, a series of laboratory tests have been
performed by using a new developed large scale
compression test instrument. The compressibility
and compression mechanism of municipal solid
waste have been studied. The laboratory test has
been performed for different proportions of
ingredients and different pressures. The testing
results show that the compression strain increases
with time, and the increasing velocity of the
compression strain is very large after the load is
applied, and decreases later. When the load is
applied, the compression strain is suddenly
increased, then increased slowly. The total
compression strain can be exceeded 70%. The total
settlement is composed of two parts: one is due to
stress compression, and the second is due to
biodegradation compression and creep
compression. In the biodegradation process, the
compression could be divided into two parts, one
part is due to the discharging of the water from the
void of the MSW and the other part is due to the
discharging of the water and the air which
generated from the degradation of organic
substance in MSW. The creep compression is a
settlement due to the holding-on and rubbing each
other in the materials under the same stress. The
compression index is 0.95 to 3.12, which shows
that the compressibility of MSW is very big.
ACKNOWLEDGMENT
This work is funded by National Natural Science
Foundation of China under Contract No.50778164
and No.50878197.
REFERENCES
Dixon N. and Jones D.R.V. (2005) Engineering
Properties of Municipal Solid Waste, Geotextiles
& geomembranes, 23 (2005), 205-233.
Hoe I.L; Dov L; Yoshiyuki M. and Toshinori K.
(1998) Estimation of Municipal Solid Waste
Landfill Settlement, Journal of Geotechnical and
Geoenvironmental Engineering, 124(1), 21-28.
Hongtao W. and Yongfeng N. (2001) Municipal
solid waste characteristics and management in
China, Journal of the Air and Waste
Management Association, 51(2), 250–263.
Sandro L.M; Miriam F.C. and Orencio M.V. (2002)
Constitutive Model for Municipal Solid Waste,
Journal of Geotechnical and Geoenvironmental
Engineering, 128(11), 940-951.
Standard for P.R. China.(1999) Standard for soil
test method (GB/T50123-1999), Plan press in
Peking, China.
Zhang Z.Y. (2005) Study on Compressibility and
Settlement of Landfill for Municipal Solid Waste,
Ph. D. Thesis. Zhejiang University, China.
Zhang Z.Y. and Wu C.f. (2010) The Results of
Direct Shear Test for Municipal Solid Waste in
Hangzhou, China, Journal of Zhejiang Sci-Tech
University, unpublished.
Geotechnical Characterization and Stability Study of Coal Mine Dumps
Aránzazu Sánchez de Ocaña Cerrada
Laboratorio de Geotecnia – CEDEX – Ministerio de Fomento, España (aranzazu.sanchez@cedex.es)
Herminia Cano Linares
Laboratorio de Geotecnia – CEDEX – Ministerio de Fomento, España (herminia.cano@cedex.es)
José Estaire Gepp
Laboratorio de Geotecnia – CEDEX – Ministerio de Fomento, España (Hjose.estaire@cedex.esH)
María Santana Ruiz de Arbulo
Laboratorio de Geotecnia – CEDEX – Ministerio de Fomento, España (maria.s.ruiz@cedex.es)

ABSTRACT The stability of two coal mine dumps, located in a village in the north of Spain, is analyzed in
this paper. The waste characterization was carried out by means of “in-situ” measurements and laboratory
testing. It is important to underline that the shear tests were performed in a 1x1 m box that allows the use of
blocks up to 20 cm. Stability calculations were carried out over ten different profiles of the two waste dumps.
In the most unfavourable and representative slopes of each waste dump, sensibility analysis were performed.

INTRODUCTION
El Bierzo, in the northwest of Spain is a region
where there used to be an important coal mining
activity. As a consequence of such industry there
are a lot of coal waste dumps spread in many of West waste dump

the valleys of the region. The Spanish

East waste dump
Government, through CIUDEN foundation, has a
program to restore those coal waste dumps.
As part of that program, CIUDEN entrusted
CIEMAT to make the “Supervision of the edafo-
landscape restoration of the waste dumps situated
in the village of Tremor de Arriba, León”. In the Fig 1: Topographic map showing waste dump situation
frame of those works, CIEMAT asked Laboratorio (For colour figure, refer to CD)
de Geotecnia- CEDEX to analyse the geotechnical
stability conditions of the two waste dumps, as a
previous step to the study of the edafo-landscape
measurements to be carried out in future.

SITE DESCRIPTION
The two waste dumps, identified as East and West,
occupy the lower part of the left and right slopes
of river Tremor valley. They are situated in the
middle of the village, as it can be seen in Figure 1.
Figure 2 shows a photograph of East waste dump,
where it can be seen that there are vegetation and Fig 2: General view of East waste dump
trees in the slopes of the waste dumps. (For colour figure, refer to CD)
The East waste dump has a surface of 3.82 Ha and Based on topographic information, slope
it is about 20 year old, while the West one has a inclination maps were made selecting for that
surface of 2.93 Ha and it is 50 year old. purpose significant slope zones associated with
the potential stability of the waste dumps, (Figures
446 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

3 and 4). Three zones were distinguished,

according to its inclination and the results
obtained in laboratory tests regarding average
friction angles: a) zones with inclination less than
35º (in green); b) zones with inclination greater
than 40º (in red) and c) zones with inclination
between 35 and 40º (in orange).

Fig 5: Greatest inclination zone in West waste dump

Fig 3: Slope inclination map (East waste dump)

(For colour figure, refer to CD)
Fig 6: Greatest inclination zone in East waste dump

Fig 7: Smallest inclination zone in West waste dump

Fig4: Slope inclination map (West waste dump)
(For colour figure, refer to CD)

Figures 5 and 6 show the profiles with greatest

inclination zones (profiles 2 and 6) while the
profiles with smallest inclination zones (profiles 1
and 9) can be seen in Figures 7 and 8. It can be
seen that the maximum height of East and West
waste dumps is, respectively, 55 and 75 m. The
maximum thickness of the coalmine wastes that
were dumped on the valley was estimated in 10 m
based on in situ observations. Fig 8: Smallest inclination zone in East waste dump
 Cerrada et al. 447

GEOTECHNICAL CHARACTERIZATION Grain size distribution

OF WASTE DUMP MATERIAL
Figure 10 shows the representative grain size
Material description distribution curves of the groups of materials
differenced in the waste dumps.
The material found in the waste dumps was 100

Passing(%)
mainly formed by a coarse soil with sands and 90 Shales Coal

Bituminous sandstones Bituminous shales

clay traces. The material was classified in four 80

70
different groups: bituminous shales, shales, 60
bituminous sandstones and coal. As it can be seen 50

in Figure 9, the grains have an angled roundness 40

and their shape is elongated, if they are shales or 30

sandstones, and cubic if they have a greater 20

10
proportion of coal rests. 0
1000,00 100,00 10,00 1,00 0,10 Diameter (mm) 0,01

Fig 10: Grain distribution size curves

(For colour figure, refer to CD)

The grain distribution size curves were similar for

the four groups in spite of the previous description
made by visual observations. The materials had an
average content of cobbles of 7%, 80% of gravel,
10% of sands and 3% of fine particles (<0.08 mm).

Atterberg Limits

The results of Atterberg Limits performed with the

fine fraction of the samples have been represented
in Figure 11.
60
Indice de Plasticidad (%)

50

CH
40

30 CL

Fig 9: Aspect of materials. Up: Bituminous shales and Coal

sandstones. Down: Coal. 20
Bituminous sandstones
Bituminous sandstones
MH
Shales

In situ density 10 Bituminous

h l
Bituminous
CL-ML
ML h l

0
The in situ density was determined by the 0 10 20 30 40 50 60 70 80 90 100

excavation of several holes, with different volume Límite Líquido (%)

Fig 11: Casagrande chart with the results of mine waste

ranging from 18 to 64 litres what implied the samples (For colour figure, refer to CD)
extraction of between 27.8 and 81 kg of material.
The results of the density determinations were The results indicate that bituminous sandstones
between 13 and 15 kN/m3. are classified as CL/ML, while the rest of
materials correspond to ML/OL group.
448 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Specific gravity The results of the five tests performed are

The nine specific gravity values obtained were in
the range between 1.94 and 2.79. The values
inferior to 2.2-2.3 correspond to materials
classified as coal while values around 2.6-2.7 to
represented in Figure 13, in terms of vertical stress
vs. shear stress. The samples in the 1 m3 box were
prepared with a density between 1.5 and 1.7
kN/m3, according to the density measured in situ.
250

bituminous sandstones and shales.

Shear stress (kPa)

200

150 WkPa)=11+Vtag 39º

Direct Shear tests
100

The direct shear test were performed in 30x30 cm 50

WkPa)=Vtag 40º
and 1x1 m boxes taking into account the grain size
0
distribution of the waste material. These boxes let 0 50 100 150 200 250 300

to use big size blocks that are more representative
of the material found in the waste dumps. To
compare the results, direct shear test in 6x6 cm
box were also performed, using less than 2 mm
size material. The samples were consolidated and
broken without filling the shear box with water.
It is interesting to remark that, to perform the
shear tests in the 1x1 m box, the material (around
6 m3 in total) was collected in situ by a backhoe at
depths between 0.5 and 1 m.
The large scale direct shear tests were performed
in a device equipped with a box of 1x1 m (Estaire,
2006). The maximum vertical load that can be
Vertical stress (kPa)
Bituminous shales (1x1) Bituminous shales (1x1)
Coal (30x30) Bituminous shales (30x30)
Bituminous shales (6x6)
Fig 13: Vertical stress vs. Shear stress obtained in the
direct shear tests. (For colour figure, refer to CD)
The strength parameters (cohesion and friction
angle) were obtained by fitting the test results with
lines, according to Mohr-Coulomb failure
criterion. The pairs of values of cohesion and
friction angle obtained with the individual
interpretation of test results are shown in Figure 14.
Friction angle (º)

40

imposed to the sample is 1000 kN. The maximum 39

horizontal displacement of the test box is 25 cm. 38

The horizontal load, with a maximum of 1000 kN, 37

can be imposed at a constant speed, ranging from 36

0.5 to 45 mm/min. The maximum size of particles 35

Bituminous shales (6x6) Bituminous shales (30x30)
used in the tests is 20 cm. Figure 12 shows a 34 Coal (30x30) Bituminous shales (1x1)

photograph of the equipment used in the tests. 33

0 5 10 15 20 25 30
Bituminous shales (1x1) Cohesion (kPa)

Fig 14: Values of cohesion and friction angle obtained

from the interpretation of direct share test results
(For colour figure, refer to CD)

Another step is to make a global interpretation of

the test results. This can be done in two different
ways also using Mohr-Coulomb failure criterion:
assuming the material only has a pure frictional
behaviour, as it corresponds to a granular material,
or assuming the material develops an apparent
cohesion coming from the mathematical fitting

Fig 12: 1x1m direct shear test device

(For colour figure, refer to CD)
 Cerrada et al. 449

Table 1 shows the representative strength TABLE 2. Parameters used in slope stability
parameters obtained with such interpretations that Type of Cohesion Friction Natural
can be seen in Figure 13. landslides (kPa) angle Density
(º) (kN/m3)
TABLE 1: Failure strength parametres Superficial 0 40 13
Interpretation 1 Interpretation 2 Deep 20 35 13
(W = c + Vv tg.I) (W = Vv tg I)
Reg. c Reg. An example of the graphical outputs obtained in
c (kPa) I (º) I (º)
Coef (kPa) Coef
the calculations is showed in Figure 15.
11,3 38 0,9835 0 40 0,9711
Note: Regression coefficients are expressed with four decimal 1.510

figures to be able to value the differences between them. F.S.=1,108

F.S.=1,510

The results shown in Figures 13 and 14 indicate 1.108

that a value of 38º can be considered as the

representative value of the friction angle, given
that the results measured in all the direct shear
boxes used in the tests, are equivalent.
Moreover, that value is consistent with the
observations made in situ regarding the stability of Figure 15: Graphical outputs of stability calculations.
the slope and that is why it has been considered Left: superficial slide. Right: deep slide.
valid for the slope inclination maps. (For colour figure, refer to CD)

SLOPE STABILITY ANALYSIS Figure 16 shows the factor of safety calculated in

all the sections of both waste dumps.
2
The slope stability analysis was performed with WEST Deep Slide
1,8 Superficial Slide
GEOSLOPE program, using Morgenstern – Price
Factor of Safety

method, as it satisfies all equilibrium conditions 1,6

with reasonable assumptions. 1,4
As it was said previously, the coalmine wastes 1,2
were dumped on the valley natural slopes, with
1
maximum thickness of 10 m. The resulting slope
0,8
inclinations were mainly between 35º and 45º. In
the calculations the piezometric level was situated 0,6
P1O P2O P3O P7O P8O
in the surface of the previous natural bedrock. Section
Two different kinds of slides were analysed: 2,2 Deep Slide
EAST Superficial Slide
superficial and deep slides. The first ones could 2
only remove the first meters of waste dumps, so it 1,8
Factor of Safety

was supposed that the material did not have any 1,6
cohesion. On the contrary, the deep slides were 1,4
supposed to be able to affect the whole thickness 1,2
of material. In this case, a cohesion value of 20 1
kPa was assigned and the friction angle was 0,8
decreased to 35º. 0,6
The stability analysis was made for ten different P4E P5E P6E P9E P10E
Section
sections of the two waste dumps. Table 2 shows
the most important parameters used in the stability Fig 16: Safety factors obtained for the most significant
calculations performed in static conditions. profiles of both waste dumps.
(For colour figure, refer to CD)
450 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE 3.- Results of reliability calculation

The slope stability analysis makes it possible to Cohesion Friction Factor of
Section
remark the following most important aspects: (kPa) angle (º) safety
- The factor of safety calculated for the deep 20 35 1.752
slides ranged from 1.98 to 1.51 that indicates 15 35 1.636
P7O 25 35 1.868
the stability of those slopes. These values are
20 33 1.658
consistent with observations made in situ.
20 37 1.850
- The factor of safety for superficial slides Standard deviation 0.15
ranged from 0.98 to 1.28. It indicates the 20 35 1.510
possibility to find landslides affecting to the 15 35 1.416
upper part of the coalmine waste deposits, P10E 25 35 1.604
especially in the zones where the inclination is 20 33 1.428
higher. 20 37 1.596
Standard deviation 0.13
In addition to the usual slope analysis, a reliability
calculation was done. This calculation provides an Considering the results of the table, the factor of
evaluation of the uncertainties linked to the safety can be accurately quantified for the
evaluation of geotechnical parameters values. In different sections as:
normal calculations, a concrete value of Section P7O => F.S. = 1.75 ± 0.15
geotechnical parameter is used but, in fact, a range Section P10E => F.S. = 1.50 ± 0.13
of values are valid for the same geotechnical
characterization of the material. These data confirm that the possible deep slide
surfaces are stable enough for the edafo-landscape
The aim of the reliability calculations in this paper measurements to be carried out without stability
is to provide a method to evaluate the range of problems.
variation of the factor of safety considering that
the values of cohesion and friction angle are into a REFERENCES
range. The Taylor series technique to estimate the Duncan et al. “Factors of safety and reliability in
standard deviation and the coefficient of variation geotechnical engineering”. Journal of
for the factor of safety was done as indicated in Geotechnical and Geoenvironmnental
Duncan (2000). As indicated in the paper, the Engineering”, 126 (4) pp 307-316. (2000)
cohesion value has a variation coefficient between
13 and 40% and between 2 and 13% for friction Estaire J. & Olalla, C. “Analysis of the strength of
angle. rockfills based on direct shear tests made in 1
m3 shear box.”. 22nd ICOLD, Barcelona. Q.86-
Taking into account these data, slope stability R.36, pp. 529-540. (2006).
analysis was made with a cohesion value ranging
between 15 and 25 kPa, and with values for
friction angle between 33º and 37º for the deepest ACKNOWLEDGEMENTS
landslides. The reliability analysis was made in
the representative sections of each slope (P7O & The authors of the paper wish to acknowledge
P10E). CIUDEN foundation and especially Avelino
García and Rocío Millán (CIEMAT) for their
Table 3 shows the factor of safety calculated contribution to the work.
changing the strength parameters.
 Constitutive Model for MSW Considering Creep and Biodegradation
Effects
G. L. Sivakumar Babu
Department of Civil Engineering, Indian Institute of Science, Bangalore 560012, India,
e-mail: gls@civil.iisc.ernet.in

Krishna R. Reddy
Department of Civil and Materials Engineering, University of Illinois at Chicago, 842 West Taylor Street,
Chicago, Illinois 60607, U.S.A, e-mail: kreddy@uic.edu

Sandeep K. Chouksey.
Department of Civil Engineering, Indian Institute of Science, Bangalore 560012, India,
e-mail: sandeepkc@civil.iisc.ernet.in

ABSTRACT: In this paper, a generalized constitutive model for MSW, based on the framework of
critical state concepts is proposed to incorporate the effects of mechanical creep and time dependent
biodegradation to calculate total compression under loading with time. To illustrate the general
applicability of the model, detailed parametric studies considering variations of different parameters are
conducted in term of variations of the settlement with time as affected by parameters. The influences of
strength and stiffness of MSW, compressibility parameters, and biodegradability parameters in
settlement-time response of MSW are highlighted. The model is useful for assessing the deformation and
stability of landfills and any post-closure development structures located on landfills.

Reddy and his coworkers (Reddy et al. 2009a,

INTRODUCTION
Prediction of municipal solid waste (MSW)
landfill settlement is required to assess the
integrity of cover systems and appurtenant
systems (gas and leachate collection pipes),
estimate the landfill airspace, and design the end-
use facilities (e.g., golf course,
industrial/commercial building). MSW settlement
is mainly attributed to: (1) physical and
mechanical processes that include the reorientation
of particles, movement of the fine materials into
larger voids, and collapse of void spaces; (2)
chemical processes that include corrosion,
combustion and oxidation; (3) dissolution
processes that consist of dissolving soluble
substances by percolating liquids and then forming
leachate; and (4) biological decomposition of
organics with time depending on humidity and the
amount of organics present in the waste.
2009b, 2009c) presented considerable data on the
geotechnical characteristics of landfills at different
stages of degradation. Babu et al. (2009a)
proposed a generalized MSW landfill settlement
model which accounts for the stress-strain
characteristics through a constitutive model based
on critical state concepts. The advantage of this
model is that it is based on the stress-strain
(constitutive) response of MSW and it is a general
model which can be applied to determine spatial
variations in settlement depending on the landfill
conditions (thickness, age, etc.). The model
accounts for mechanical compression and time-
dependent mechanical creep and biodegradation.
The model is validated with reference to the test
data of (a) fresh MSW obtained from working
phase of a landfill, (b) landfilled waste retrieved
from a landfill after a year of degradation, and (c)
synthetic MSW with controlled composition. The
model captures the stress-strain and pore water
pressure response of these three types of MSW
adequately. Babu et al. (2010) illustrated the
452 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

applicability of the model for a typical MSW effects. The elastic volumetric strain d H ve can be
landfill. The predicted settlement results were written as
compared with the predicted settlement results
dee N dpc
obtained using fourteen different reported models. d H ve  (2)
1 e 1  e pc
Two of these reported models were developed by
Marques and his collaborators (Marques, 2001;
and Marques et al., 2003) which also account for And, increment in plastic volumetric strain can be
mechanical compression, mechanical creep and written as
biodegradation similar to the constitutive model.
§ O N · ª dpc 2K dK º
d H vp ¨ ¸ « pc  2 » (3)
FOR PARAMETRIC STUDY © 1 e ¹ «¬ M  K 2 »¼

The settlement predictive models involve many

parameters related to the strength, compressibility
and biodegradation which vary widely. Literature
review indicates that the influence of the
parameters related to (i) shear strength ( I ), (ii)
compressibility (Cc), (iii) total biodegradable
strain ( EDG ), and (iv) biodegradation rate constant
(d) on the settlement response is significant.
The above formulations for increments in
volumetric strain due to elastic and plastic are well
established in critical state soil mechanics
literature.
The mechanical creep is a time dependent
phenomenon in exponential function given by
HC 
It may be stated that while the qualitative
influence of these parameters on settlement of
MSW is known, it is essential to obtain the values
in quantitative terms in terms of time-settlement
behavior for different parameters and this is
possible by conducting parametric studies using
suitable models or the approaches that consider all
the above factors. The following sections provide
an overview of the constitutive model of Babu et
al. (2009a) and provides results of parametric
study, including comparisons with two reported
models- Marques (2001) and Marques (2003).
CONSTITUTIVE MODEL (Babu et al. 2009a)
Considering elastic and plastic behavior as well as
mechanical creep and biological decomposition,
the total volumetric strain of the MSW under
loading is expressed as:
b'pc 1  e  ct c  (4)
where b is the coefficient of mechanical creep;
'pc is the change in mean effective stress, c is the
rate constant for mechanical creep; and t’ is the
time since application of the stress increment. The
time dependent biodegradation is given by
Hb EDG (1  edtcc ) (5)
where EDG is the total amount of strain that can
occur due to biological decomposition; d is the
rate constant for biological decomposition; and t "
is the time since placement of the waste in the
landfill. From Eq. (4), increment in volumetric
strain due to creep is written as:
d H vc cb'pcect cdt c (6)

p From Eq. (5), increment in volumetric strain due

dHv dH e  dH  dH c  dH b (1) biodegradation effect is written as:
v v v v

where d H e , d H p , d H c and d H b are the increments d H vb EDG e dt ccdt cc (7)

v v v v
of volumetric strain due elastic, plastic, time
dependent mechanical creep and biodegradation
 Babu, Reddy and Chouksey 453

In the present case t c time since application of the over 30 years (10,950 days, which is typically the
stress increment and t " time since placement of landfill closure time specified) and draw
the waste in the landfill are considered equal to‘ t ’. inferences with regard to time-settlement response
Using Eqs. (2), (3), (6), and (7) and substituting in of MSW.
Eq. (1), total increment in strain is given by
Influence of shear strength parameters
' ' For the calculation purpose, it is assumed that only
N dp § O  N · ª dp 2K d K º
dH v ''
¨ ¸« ''  » friction angle is variable and other parameters
1 e p © 1  e ¹¬ p M 2 K 2
¼ remain constant and values in the range of 10° to
 cb ' V e  ct dt  E DG e 40° are used. The model parameters used for the
(8) calculations and ultimate settlement at the end of
30 years are given in Table 1.

PARAMETRIC STUDY TABLE 1. Influence of friction angle ( I ) on MSW

settlement
A simple example of MSW landfill of 30 m
(I ) (Cc) ( EDG ) (d)(day-1) (m)
height has been considered which is assumed to be
filled in ten layers each of 3 m deep as shown in 10° 0.20 0.159 0.00114 0.71
Fig. 1. 20° 0.20 0.159 0.00114 0.64
30° 0.20 0.159 0.00114 0.58
40° 0.20 0.159 0.00114 0.53

Fig. 2 shows the time-settlement response for

MSW for friction angles. It can be noted that the
settlement at the end of 30 years (10,950 days)
corresponding to 10° is 0.71 m, whereas for MSW
represented by higher friction angle 40°, the
settlement is 0.53 m. For a particular time period,
the settlement is less for MSW with higher friction
angle. This shows that the shear strength in terms
of friction angle of the MSW influences the
ultimate settlement of the landfill considerably.

Time (days)
1 10 100 1000 10000
Fig. 1 MSW landfill scenario for estimation of 0
I 
settlement versus time.
I 
0.2 I 
At the top of landfill, a final cover system has
S ettlem en t (m )

I 
been assumed to be constructed which consists of 0.4
composite liner (compacted clay and
geomembrane) overlain by a sand drainage layer 0.6
and then a vegetative cover soil layer. The
objective of this study is to examine the results of
0.8
variation of landfill settlement with time for a
typical layer (for example, P6 layer) considering
1
the variations of different ranges of parameters
454 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Fig. 2 Time vs. settlement response of MSW for Influence of Biodegradation

different values of friction angle
Influence of Compression Index (Cc)
Using different values of compression index from
0.10 to 0.40, time-settlement responses is
predicted keeping the other parameters constant.
Comparison of values of maximum settlement
from all the three models and the model
parameters used are given in Table 2. It can be
noted that the higher values of compression index
are associated with larger settlements.
Time (days)
1 10 100
0 parameters are presented in Table 3.
To assess the settlement behavior of MSW with
respect to biodegradation effect, the three models
are used with different values of total
biodegradable strain, EDG (0.000159 - 0.159) and
the biodegradation rate is kept constant. Marques
et al. (2003) reported the values of EDG varying
from 0.214 to 0.131 with an average value of
0.159. In this study, this value as well as the
values lowered by 10, 100 and less have been used
to discern the effect. Comparison of time-
settlement response for different values of total
biodegradable strain ( EDG ) is shown in Figure 4..
The maximum settlements corresponding to
1000 10000 different values of EDG and additional constant

0.5
TABLE 3. Comparison of maximum values of
S ettlem ent (m)

1
settlement (in m) from different models keeping
constant parameters ( I =20°, Cc =0.20, d = 0.00114
1.5
day-1)
Cc = 0.10 Marques
Marques Proposed
Cc = 0.20 (EDG) et al.
2 (2001) model
Cc = 0.30 (2003)
Cc = 0.40
0.000158 0.872 0.993 1.038
2.5
0.00158 0.876 0.9374 1.058
Fig. 3 Time vs. settlement response of MSW for 0.0158 0.919 0.981 1.120
different values of compression index ( Cc ). 0.158 1.348 1.417 1.408

Time (days)
The general observation from all three models is
1 10 100 1000 10000
that with increasing values of compression index,
0
the predicted value of ultimate settlement is more. Edg = 0.000159
0.2 Edg = 0.00159
TABLE 2. Comparison of maximum values of 0.4
Edg = 0.0159
settlement (in m) from different models keeping Edg = 0.159
Settlement (m)

constant parameters ( I =20°, EDG =0.159, d = 0.00114 0.6

day-1) 0.8
Marques Marques et Proposed
(Cc) 1
(2001) al. (2003) model
0.10 1.017 1.075 1.030 1.2
0.20 13.47 1.417 1.334 1.4
0.30 1.676 1.759 1.821
1.6
0.40 2.005 2.101 2.164
Fig. 4 Time vs. settlement response of MSW for
different values of total biodegradable strain EDG ).
 Babu, Reddy and Chouksey 455

CONCLUDING REMARKS
From the plotted results, it is observed that the
model predictions of final settlements are In this paper, results of parametric study on the
influenced by the values of total biodegradable influence of strength, compressibility,
strain. biodegradation parameters on the settlement time
response of MSW are presented. As expected,
Influence of Rate of Biodegradation (d) shear strength parameters, the total biodegradation
Comparison of the settlement response for content and rate of biodegradation influence the
different values of d is shown in Fig. 5. It shows settlement response which can be predicted by the
that higher values of biodegradation rate constant model proposed. This aspect can be
cause higher settlement and hence enhancement of advantageously used in the design of bioreactor
biodegradation rates using leachate recirculation landfills to accelerate settlement and also improve
helps in accelerated settlements. The maximum strength response.
settlements observed corresponding to different
values of d are presented and constant parameters REFERENCES
used are presented in Table 4.
Babu Sivakumar, G. L., Reddy, K.R., and
Chouskey, S.K. (2009a). Constitutive Model
TABLE 4. Comparison of maximum values of
settlement (in m) from different models keeping for Municipal Solid Waste Incorporating
constant parameters ( I =20°, Cc =0.20, EDG= 0.159) Mechanical Creep and Biodegradation-Induced
Compression. Waste Management Journal, (30)
Marques Constit
Marques 11- 22.
(d ) (2001)
et al. -utive
(2003) model Babu Sivakumar, G. L., Reddy, K.R., and
0.0000114 0.923 0.974 0.891 Chouskey, S.K. and Kulkarni, H.S. (2010)
0.000114 1.196 1.219 1.112 Prediction of Long-term Municipal Solid
0.00114 1.348 1.417 1.328 Waste Landfill Settlement Using Constitutive
0.0114 1.349 1.418 1.344 Model. Practice Periodical of Hazardous,
Toxic, and Radioactive Waste Management,
ASCE, 14(3), 139-150.
Time (days)
1 10 100 1000 10000 Marques, A. C. M., Filz, G. M., and Vilar, O. M.
0 (2003). “Composite compressibility model for
d = 0.0000114 municipal solid waste.” J. Geotech.
0.2
d = 0.000114 Geoenviron. Eng., 129(4), 372–378.
0.4 d = 0.00114
Settlement (m)

d = 0.0114 Marques, A.C.M. (2001). “Compaction and

0.6 Compressibility of Municipal Solid Waste.”
0.8 PhD Thesis, Sao Paulo University, Sao Carlos,
1
Brazil.

1.2
Reddy, K.R., Hettiarachchi, H., Gangathulasi. J.,
Bogner, J.E., Lagier, T., 2009c. Geotechnical
1.4 properties of synthetic municipal solid waste.
1.6 International Journal of Geotechnical
Fig. 5 Time vs. settlement response of MSW for Engineering, 3(3), 429-438.
different values of biodegradation (d). Reddy, K.R., Hettiarachchi, H., Parakalla, N.,
Gangathulasi. J., Bogner, J.E., Lagier, T.,
2009b. Geotechnical properties of landfilled
456 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
p p p
municipal solid waste under short-term solid waste at Orchard Hills Landfill, USA.
leachate recirculation operations. Waste Waste Management, 29(2), 952-959.
Management & Research, 27(6), 578-587.
Wood, D.M., 1990. Soil Behaviour and Critical
Reddy, K.R., Hettiarachchi, H., Parakalla, N.S., State Soil Mechanics. Cambridge, University
Gangathulasi. J., Bogner, J.E., 2009a. Press, UK.
Geotechnical properties of fresh municipal
 Hydraulic Conductivity of Municipal Solid Waste
as a Function of Placement Conditions

N. Yesiller
Global Waste Research Institute, California Polytechnic State University, San Luis Obispo, California,
USA (nyesille@calpoly.edu)
W. W. Wong
Hultgren-Tillis Engineers, Concord, California, USA (wwwong@gmail.com)
J. L. Hanson
California Polytechnic State University, San Luis Obispo, California, USA (jahanson@calpoly.edu)

ABSTRACT This investigation was conducted to evaluate the effect of placement conditions on hydraulic
conductivity (k) of municipal solid wastes. A laboratory test program was conducted using a large-scale
rigid-wall dual-ring permeameter to determine k as a function of compaction conditions. The results
indicated that k is highly dependent on waste placement conditions, ranging from approximately 1 x 10-2
cm/s (dry of optimum) to 8 x 10-5 cm/s (optimum and wet of optimum). The decrease in k for wet samples
was attributed to particle swelling, as well as softening and subsequent breakdown of clusters of waste
components in analogy to hydraulic conductivity response of compacted clayey soils.

INTRODUCTION
The quantity of municipal solid waste (MSW)
disposed of in landfills continue to increase in the
U.S., even with increased efforts towards waste
diversion and recycling (USEPA 2008).
Engineering properties of wastes and placement
conditions affect the operation, short- and long-
term performance, and overall stability of landfills.
Hydraulic conductivity of MSW affects the
geotechnical response (e.g., dissipation of pore
water pressures, compressibility), biochemical
performance (e.g., migration of fine particles,
microbes, and nutrients affecting rate of physical
breakdown or decomposition of the waste
components), and hydraulic performance (e.g.,
leachate flow regimes) of landfills. High
variability was reported for k of MSW based on
experimental field and laboratory investigations
(e.g., Oweis et al. 1990, Powrie & Beaven 1999,
Reddy et al. 2008). Results span 7 orders of
magnitude (from 1 x 10-8 to 2 x 10-1 cm/s), as
summarized in Jain et al. (2006) and Reddy et al.
(2009). Hydraulic conductivity has been reported
to be a function of void ratio or dry unit weight
(Landva and Clark 1990, Durmusoglu et al. 2006,
Reddy et al. 2009). Similarly, several investigators
have provided correlations between k and vertical
effective stress (i.e., burial depth) [Blieker et al.
1995, Powrie and Beaven 1999]. Increased dry
unit weights, decreased void ratios, and increased
effective stresses all result in decreased k due to
decreased void space available for flow paths. In
addition, MSW changes physically and chemically
with time, which influences hydraulic properties.
Multiple factors affect the hydraulic conductivity
of MSW with time, including material softening,
particle swelling, and particle migration (Olivier &
Gourc 2007, Chen & Chynoweth 1995). Effects of
compaction conditions on k and variation of k as a
function of placement conditions have not been
identified.
This study was conducted to investigate the
variation of hydraulic conductivity of MSW as a
function of placement conditions. Hydraulic
conductivity of variably compacted wastes was
458 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

determined using a large-scale permeameter to TABLE 1 Waste components and preparation

investigate the relationship between the as-placed Maximum
Weight
structure of wastes (due to compaction) and Material Material Nominal
Fraction
hydraulic performance. These relationships have Category Used Particle Size
(%)
(mm)
implications for operation of landfills including
Shredded
leachate recirculation systems and bioreactor Paper
paper
3 x 32 19
operations. Paper Cardboard 50 x 50 6
Broken
EXPERIMENTAL TEST PROGRAM Glass 50 6
glass
Steel Nails 32 5
Test Materials Al
Aluminum 20 2
Laboratory hydraulic conductivity tests were shavings
conducted on samples prepared using MSW Other Al
20 0.3
metals shavings
constituent compositions provided by USEPA
Plastic
(2008). A total of 15 materials were used to Plastics
chips
10 13
represent the components of MSW. The materials Plastic
used for constructing the test samples are Plastics 50 x 50 3
bags
described in Table 1. Rubber/ Shredded
20 3
leather tires
Multiple materials were used for the broad Rubber/ Leather
50 x 50 1
categories of paper, plastics, and rubber and leather coupons
leather based on the data provided in Miller & Textile
Textiles 50 x 50 6
coupons
Clesceri (2002). The resulting mixture was
Wood
representative and reproducible and allowed for Wood
chips
30 7
investigation of hydraulic conductivity of MSW Other Concrete
without the influence of compositional variability 50 2
materials fragments
between test samples. Food Dog food 13 18
Grass
Yard 50 7
Test Samples clippings
An extensive laboratory test program was used to Other Soil 2 2
determine the compaction characteristics of the
MSW mixture described in the previous section.
4X Modified Effort
Compaction tests were conducted using modified
and 4X modified effort tests to represent high 6.5
compactive efforts observed in field compaction of
MSW. Details of the compaction analyses are 6.0
Dry Unit Weight (kN/m )
3

presented in Hanson et al. (2010). The Jdmax was

5.5
determined to be 5.9 kN/m3 and the wopt to be 56%
for the 4X modified test (Fig. 1). Hydraulic 5.0
conductivity tests were conducted on a total of 5
MSW samples prepared at predetermined moisture 4.5
content (w) – dry unit weight (Jd) conditions based
4.0
on the results of the 4X modified effort tests
(Table 2). The samples were prepared at w – Jd 3.5
pairs representing dry of optimum, optimum, and 0 20 40 60 80 100 120 140
wet of optimum conditions. Moisture Content (%)
Fig. 1. Compaction curve for MSW
 Yesiller, Wong and Hanson 459

TABLE 2 Hydraulic conductivity test samples progressed to assure consistent compaction with
Moisture Moist Unit Dry Unit depth. A geocomposite was placed on top of the
Content Weight (kN/m3) Weight (kN/m3) sample prior to placement of the drum lid. Once
11% 4.7 4.3 the permeameter had been sealed, the outlet from
30% 7.0 5.4
the Mariotte bottle was attached and air was
56%a 9.2 5.9
85% 10.2 5.5
purged from the water supply line.
110% 10.3 4.9
a
optimum moisture content

Hydraulic Conductivity Tests

Procedures used for testing soils were adopted for
the laboratory investigation. Tests were conducted
in a custom-built rigid-wall dual-ring permeameter.
A large-scale device was required to provide
appropriate scale for the relatively large particles
(as compared to soil) in the waste samples. The
permeameter was built using a 30 L steel drum
with a removable lid, and included a stainless steel
separation ring (50 mm height and 280 mm
diameter) and one inlet and two outlet ports. The
inlet port consisted of a 25-mm-diameter bulkhead
and the two outlet ports included a central drainage
port and a perimeter drainage port consisting of
25- and 19-mm-diameter bulkheads, respectively.
The base of the permeameter was lined with a
drainage and filtration system consisting of (from
top to bottom) a geocomposite (3 ply edge drain);
a 40-mm-thick layer of uniformly graded, angular
Fig. 2. Schematic diagram and photograph of
gravel; and another geocomposite reinforced with
permeameter (For colour figure, refer to CD)
steel mesh. A schematic diagram and photograph
of the permeameter are presented in Fig. 2.
RESULTS AND DISCUSSSION
The hydraulic conductivity of the MSW mixture
A large scale Mariotte bottle was used to provide
generally decreased with time for all of the tests
constant head conditions for the permeability tests.
(Wong 2009). The effect was more pronounced
The Mariotte bottle consisted of a 200-mm-outer
for optimum and wet of optimum samples due to
diameter 1.8-m-long acrylic tube with a wall
the longer test durations. The decrease in k was
thickness of 10 mm. The bubble tube was a 1.08-
attributed to time-dependent swelling of the waste
m-long acrylic tube with 25 mm outer diameter
components, potential migration of fines
and 3 mm nominal wall thickness.
throughout the waste sample causing blocked
A compression load frame was used to compact
pathways, and potential initiation of biological
wastes in the permeameter using static load
activity causing clogging.
application at the predetermined compaction
conditions (Table 2). Approximately 15 to 20 lifts
Significant variation in hydraulic conductivity of
were used to prepare each test sample in the
MSW was observed as a function of compaction
permeameter. The samples were prepared to a
conditions. Hydraulic conductivity decreased by
thickness of 356 mm. The lifts ranged in thickness
more than two orders of magnitude as moisture
from 25 mm to 75 mm and decreased as filling
content increased (Fig. 3). The greatest decrease
460 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

occurred between the dry of optimum and higher for the sample prepared at dry of optimum
optimum moisture contents with relatively low (w = 30%) than the sample prepared at wet of
variation in k at moisture contents above optimum optimum (w = 85%).
moisture conditions. The hydraulic conductivity
decreased from the maximum value for the driest
sample (1.28 x 10-2 cm/s) to the minimum value at Hydraulic Conductivity
optimum moisture content (7.99 x 10-5 cm/s). -1
10
The maximum hydraulic conductivity was

Hydraulic Conductivity (cm/s)

obtained for the minimum dry unit weight and the -2
10
minimum hydraulic conductivity was obtained for
the maximum dry unit weight (Fig. 4). The
general trend of decreasing hydraulic conductivity 10
-3

with increasing dry unit weight is consistent with

data provided in literature (e.g., Reddy et al. 2009).
However, the results from this test program 10
-4

indicate the presence of mechanisms that are

affected by compaction conditions extending
-5
beyond strict dry unit weight – hydraulic 10
conductivity correlations. At moisture contents 0 20 40 60 80 100 120
wet of optimum, only a slight increase in k was Moisture Content (%)
observed as compared to optimum moisture Fig. 3. Hydraulic conductivity versus compaction
moisture content
content conditions even though these samples were
prepared at significantly lower dry unit weights
than optimum. Dry of Optimum
Optimum
Wet of Optimum
The variation of k as a function of compaction -1
10
moisture content is described in analogy to the
hydraulic conductivity response of clayey soils
Hydraulic Conductivity (cm/s)

(e.g., Mitchell et al. 1965, Benson & Daniel 1990, -2

10
Daniel and Benson 1990). Even though the values
of hydraulic conductivity measured for MSW were
generally in the range of values expected for -3
10
granular soils, the variations in k between the dry
and wet of optimum moisture content resembled
-4
the response of clayey soils. The observed 10
behavior was attributed to numerous components
of the waste mixture behaving similar to clay
-5
clusters or clods. Preferential flow paths between 10
discrete clods that are present in dry conditions did 4 4.5 5 5.5 6
3
not exist at higher moisture contents (i.e., wet of Dry Unit Weight (kN/m )
optimum conditions). The softening of the clusters Fig. 4. Hydraulic conductivity versus dry unit weight
or clods at higher moisture content allowed for
breakdown of the waste components (i.e., clods) The high variability of hydraulic conductivity of
and resulted in reduced void spaces between clods MSW as a function of compaction conditions at
reducing flow of water (Benson & Daniel 1990, time of waste placement can result in zones of
Daniel & Benson 1990). At similar dry unit highly variable hydraulic behavior in a landfill
weights (i.e., void ratios), k was significantly system. The authors have observed highly varying

seasonal moisture and associated variable dry unit
weight conditions on a field scale (Hanson et al.
2010). Seasonal placement of wastes can result in
a stratified waste mass with alternating layers of
high and low hydraulic conductivity. Such
variability may affect the transport and extraction
of fluids in the landfill system and should be
considered in landfill design and operation.
CONCLUSIONS
Hydraulic conductivity of municipal solid waste
was determined as a function of placement
conditions. A large-scale dual-ring rigid-wall
permeameter was used in the testing program.
Tests were conducted on samples (356 mm height
and 356 mm diameter) prepared over a wide range
of predetermined moisture content – dry unit
weight conditions based on the results of prior 4X
modified effort compaction tests. In a given test, k
typically decreased with time. The decrease in k
was more pronounced for the higher moisture
content samples and was attributed to time
dependent swelling of the waste components,
potential migration of fines in a sample blocking
pathways, and potential biological activity causing
clogging, especially for relatively long duration
tests.
Significant variation in hydraulic conductivity of
MSW was observed as a function of compaction
conditions. Hydraulic conductivity decreased by
more than two orders of magnitude as moisture
content increased from the dry of optimum to the
wet of optimum. The greatest decrease occurred
between the dry of optimum and optimum
moisture contents with relatively low variation in k
at moisture contents above the optimum moisture
content. Hydraulic conductivity decreased from
1.28 x 10-2 cm/s for dry conditions (11% moisture
content) to 7.99 x 10-5 cm/s for optimum moisture
content (wopt = 56%). At similar dry unit weights
(i.e., void ratios), k was significantly higher for the
sample prepared at dry of optimum (30% moisture
content) than the sample prepared at wet of
optimum (85% moisture content). Variation of
moisture content and associated variations in
structure of waste samples had more significant
effects on hydraulic conductivity than direct dry
Yesiller, Wong and Hanson 461
unit weight – hydraulic conductivity correlations.
The decrease in k for wet of optimum compared to
dry of optimum conditions (even at similar dry unit
weights) was attributed to particle swelling and
breakdown of material structure (i.e., clods in
analogy to hydraulic conductivity response of
compacted clayey soils).
REFERENCES
Benson, C.H., and Daniel, D.E. (1990). Influence
of Clods on Hydraulic Conductivity of
Compacted Clay. Journal of Geotechnical
Engineering, 116(8): 1231-1248.
Bleiker, D.E., Farquhar, G., and McBean, E.
(1995). Landfill Settlement and the Impact on
Site Capacity and Refuse Hydraulic
Conductivity. Waste Mngt. & Res., 13, 533-554.
Chen, T., and Chynoweth, D. (1995). Hydraulic
Conductivity of Compacted Municipal Solid
Waste. Bioresources Technology, 51, 205-212.
Daniel, D.E. and Benson, C.H. (1990). Water
Content-Density Criteria for Compacted Soil
Liners, Journal of Geotechnical Engineering,
116(12): 1811-1830.
Durmusoglu, E., Sanchez, I.M., and Corapcioglu,
M.Y. (2006). Permeability and Compression
Characteristics of Municipal Solid Waste
Samples. Environmental Geology, 50, 773-786.
Hanson, J.L., Yesiller, N., VonStockhausen, S.A.,
and Wong, W.W. (2010). Compaction
Characteristics of Municipal Solid Waste.
Journal of Geotechnical and Geoenvironmental
Engineering, 136(8): 1095-1102.
Jain, P., Powell, J., Townsend, T., and Reinhart, D.
(2006). Estimating the Hydraulic Conductivity
of Landfilled Municipal Solid Waste Using the
Borehole Permeameter Test. Journal of
Environmental Engineering, 132(6): 645-652.
Landva, A.O., and Clark, J.I. (1990). Geotechnics
of Waste Fill. Geotechnics of Waste Fills -
Theory and Practice, STP 1070 , A. Landva
and E.G. Knowles, Eds., ASTM, 86-106.
462 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Miller, P. and Clesceri, N. (2002). Waste Sites as
Biological Reactors: Characterization and
Modeling, CRC Press, Boca Raton, FL.
Mitchell, J.K, Hooper, D.R., and Campanella, R.G.
(1965). Permeability of Compacted Clay.
Journal of the Soil Mechanics and Foundations
Division, Proc. of ASCE, 91(SM4): 41-64.
Olivier, F. and Gourc, J.-P. (2007). Hydro-
Mechanical Behavior of Municipal Solid Waste
Subject to Leachate Recirculation in a Large-
Scale Compression Reactor Cell. Waste
Management, 27(1): 44-58.
Oweis, I.S., Smith, D.A., Ellwood, R.B., and
Greene, D.S., (1990). Hydraulic Characteristics
of Municipal Refuse. Journal of Geotechnical
Engineering, 116(4): 539-553.
Powrie, W., and Beaven, R. (1999). Hydraulic
Properties of Household Waste and
Implications for Liquid Flow in Landfills.
Proceedings of the Institution of Civil
Engineers, Geotechnical Engineering, 235-247.
Reddy, K.R., Gangathulasi, J., Hettiarachchi, and
Bogner, J. (2008). Geotechnical Properties of
Municipal Solid Waste Subjected to Leachate
Recirculation. Geotechnics of Waste
Management and Remediation, GSP 177,
ASCE, 144-151.
Reddy, K.R., Hettiarachchi, H., Parakalla, N.,
Gangathulasi, J., Bogner, J., and Lagier, T.
(2009). Hydraulic Conductivity of MSW in
Landfills. Journal of Environmental
Engineering, 135(8): 677-683.
USEPA. (2008). Municipal Solid Waste
Generation, Recycling, and Disposal in the
United States: Facts and Figures for 2006.
http://www.epa.gov/osw/nonhaz/municipal/
msw99.htm. Last accessed: 3 June 2008.
Wong, W.W. (2009). Investigation of the
Geotechnical Properties of Municipal Solid
Waste as a Function of Placement Conditions,
M.S. Thesis, California Polytechnic State
University, San Luis Obispo, California.
 Geotechnical Aspects of Nuclear Waste Management

Padmanabhan G, Sundaramurthy C, Sivathanu Pillai.C, Kumar.P.V

Indira Gandhi Center for Atomic Research, India (padman@igcar.gov.in)

ABSTRACT Radioactive waste generated from all stages of Nuclear Fuel Cycle, industrial applications,
medical use and research needs to be safely managed for the protection of human and environment. In
contrast to the normal solid waste management and disposal, disposal of nuclear waste needs special attention
to reduce undue exposure to public and environment. Sites for locating Nuclear Waste disposal require
extensive geotechnical and hydro geological studies. This paper highlights geotechnical aspects of Nuclear
Waste management, site selection for locating waste disposal facilities and safety analysis to assess the
suitability of foundation material to minimize the radioactive release to public and environment.

INTRODUCTION facilities with the surrounding soil is also an

In contrast to normal solid waste disposal, the
philosophy of nuclear waste management is built
on waste minimization, its volume reduction and
effective confinement. Waste management
principle of Department of Atomic Energy, India is
based on delay and decay of radio active elements,
dilute and disperse, confinement through volume
reduction and safe disposal. For a safe disposal,
Nuclear Waste management and disposal facilities
are co-located with Nuclear Power Plants to avoid
radiation exposure during transportation of nuclear
waste. Facilities for storage and disposal of
various categories of nuclear waste are Interim
storage facilities, Near Surface Disposal Facilities,
and deep geological repositories. Near surface
disposal facilities are co located with associated
Nuclear Power Plants and fuel cycle facilities.
Sites for locating these facilities required to be
qualified in terms of its geotechnical and hydro
geological characters. Possible geohazard and
stability of sub surface material are to be assessed
for ensuring the safety of structures founded on the
sub surface. Interaction of this waste storage
important aspect to be considered in the design of
these facilities. This paper addresses the safety of
nuclear waste management facilities in relation
with geotechnical, geological and hydro geological
characters of a near surface disposal site.
STORAGE AND DISPOSAL FACILITIES
Storage of radioactive waste is required for
isolation, protection of environment and decay of
short lived radio nuclides. Storage involves
isolating treated nuclear waste by placing them in
engineered barriers, containing alpha bearing
waste and vitrified high level waste in interim
storage facilities pending deep geological disposal.
Disposal is the final step in the radio active waste
management.
SITING OF NUCLEAR WASTE STORAGE AND
DISPOSAL FACILITIES
Being an important step in waste disposal system,
systematic approach is required for siting of these
facilities from conceptual stage to site
confirmation stage. Siting principles has to be
outlined during the conceptual stage and candidate
464 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
sites were selected during the site survey stage.
The candidate sites have to be evaluated in detail
during the site characterization stage and the
characteristics evaluated are to be confirmed
during site confirmation stage. The various
important parameters to be considered during a
siting process of nuclear waste disposal facilities
are geology, hydrogeology, and geochemistry are
described in this section.
Geology
The objective of studying geological setting of the
site is to under stand the stratigraphy, lithology,
and mineralogical characters of the site during the
site characterization stage. Seismic refraction
survey and exploratory geotechnical investigations
will be carried out the evaluate sub surface
characteristics during this stage. These data’s are
required to assess the suitability of the site. A
typical profile obtained from seismic refraction
survey is shown in Fig 1.
Fig.1 -Typical soil profile from seismic refraction
survey (For colour figure, refer to CD)
The compression wave velocity obtained from the
seismic refraction survey and thickness of each
sub surface layers determined during the
investigation are useful to assess the suitability of
the site for locating these facilities. Geophysical
electrical sounding has to be carried out at
different locations and geological formations are to
be obtained by interpreting the vertical electrical
sounding resistivity data. A typical fence diagram
obtained for a disposal facility using electrical
sounding and bore hole data is shown in Fig 2.
Fig.2 –Sub surface model of a disposal Site
(For colour figure, refer to CD)
In the site confirmation stage extensive geological
and geotechnical investigations are to be carried
out to fully characterize the candidate site and
results obtained from preliminary investigations
will be confirmed at this stage.
Hydrogeology
Hydrogeological studies include the evaluation of
aquifer thickness, ground water flow and its
direction. Sites for locating nuclear disposal
facilities should have low ground water flow and
long flow path.
Aquifer parameters are to be determined by
pumping test during both pre and post monsoon.
The draw down data and recovery data are plotted
and the characteristics of aquifer were determined.
The thickness of the aquifer, hydraulic
conductivity and transmittivity of the aquifer will
be determined to assess the suitability of the
disposal site.
Fluctuation of water level in the aquifer in
response to recharge and discharge has to be
studied to evaluate the aquifer characteristics.
Spatial variation in ground water level is used to
indicate the ground water flow. The ground water
flow shall not be towards the water bodies to avoid
the transport of radio nuclides. Flow direction of
ground water observed for a particular month in a
disposal facility is shown in Fig 3.

Fig.3 –Ground water flow direction
Geochemistry
Geochemistry of the disposal site is required to be
evaluated to assess the suitability of groundwater
to avoid the transportation of radionuclide.
Geochemical condition should promote conditions
suitable for absorption and precipitation of
radionuclide and inhibit the formation of easily
transportable chemical compounds of
radionuclides. Corrosivity of the groundwater
towards engineered barrier is also to be studied to
evaluate the safety of the disposal system.
Hydrogen ion concentration, Redox potential,
Electrical conductance, determination of presence
of major ions like Potassium, Magnesium,
Calcium, Chloride are typically evaluated in a
disposal facility site.
The determination of Chloride and sulphate in sub
soil and ground water are required to determine
the type and quantity of cement used for
construction of sub surface structures. Figure 4
shows regional variation in concentration of
sulphate in ground water at a near surface
Disposal site. The chemical analysis on water and
soil samples showed that site falls under the
category of Class III as per Indian Standard IS
456- 2000 and hence it is recommended to use
Portland Pozzolana Cement with a minimum
cement content of 350 kg/m3.
Padmanabhan et al. 465
Fig.4 –Variation of Sulphate in mg/l
EVALUATION OF ENGINEEING PROPERTIES
OF A NEAR SURFACE DISPOSAL FACILITY
To assess the possible geo hazard and stability of
subsurface material detailed geotechnical
investigations are to be carried out at the candidate
site. From the detailed geotechnical investigations
carried out engineering properties of the sub soil is
determined to assess the interaction of waste
storage facilities with the surrounding soil.
A typical near surface disposal facility consists of
Stone lined earthen trenches, RCC trenches, tile
holes and Alpha Solid waste storage facilities.
Typically these structures are founded on
competent strata. The wastes stored in these
disposal facilities are retrievable. These structures
are designed for site specific earthquake levels so
that in the event of any unforeseen even there is
no breach of radio nuclides to the environment.
For a Near Surface Disposal facility located at a
particular site along East Cost of Peninsular India
geotechnical investigations are carried out as per
the requirement of USNRC 1.132 – Site
investigation for Nuclear Power Plants. Various
field and laboratory tests are carried out to
evaluate the engineering properties of the sub
surface material. The foundation strata of
weathered rock were characterized from the
results obtained from the field and laboratory
466 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

investigations. The Rock Mass Rating obtained for earthquake of Level 1 as per IAEA TECDOC
the top layer of weathered rock is indicated in the 1347 and FEMA 450.
Table 1.
Explanatory guide to FEMA 450 suggests
Table 1 : Rock Mass rating of weathered rock liquefaction potential can be evaluated from
Parameter Range of Rating empirical procedures based on the observed cases
values of liquefaction and measurements made in the field
Strength of 50-100 7 with conventional exploration methods. This
intact rock in
approach is based on investigation of whether or
Mpa
RQD 25-50% 8
not insitu soil deposit has actually developed
liquefaction during the past earthquakes. With an
Spacing of 60-200mm 8
intensity of shaking estimated by appropriate
discontinuities
Condition of Slightly rough 20 procedures, values of Cyclic Stress Ratio (CSR)
discontinuities surface believed to have occurred in insitu deposits during
Ground water Damp 10 an earthquake can be estimated and compared with
the penetration resistance of a sandy soil at any
RMR value 53 depth of an insitu deposit. Since it is known
whether or not liquefaction induced ground
damage occurred, it is possible to establish a
The allowable bearing pressure is estimated using threshold relation between the CSR and N-value of
different methods specified as per IS 12070- the SPT test and CSR and observed shear wave
Design and construction of Shallow foundations velocity. Liquefaction potential is evaluated based
on rocks. The least value from this is adopted as on both SPT N values using Seed & Idriss method
safe bearing pressure of shallow foundations and Shear wave velocity as per the guidelines of
resting on rock. Table 2 shows a typical range of FEMA 450 for FRFCF site.
allowable bearing capacity of weathered
Charnokite formation in a near surface disposal The results shows that factor of safety against
facility. liquefaction is more than 1.1 for all the methods in
all the bore holes and this suggests that the site is
Table 2: Allowable bearing pressure of Charnokite Rock safe against liquefaction. Figure 5 and 6 represents
formation the factor of safety against depth obtained from
Safe bearing Capacity SBC (Mpa) various methods of analysis as per FEMA 450.
UCC Point RQD RMR Pressure Liquefaction resistance Using SPT N values
load meter test (FEMA 450)

2.56- 10.06- 1-8 1.78-2.67 1.17-2.51 Factor of Safety

4.31 10.51 0 1 2 3 4 5 6
0

From this it is concluded that the safe bearing 1

capacity of the disposal structures resting on 2

weathered rock formation is of the order of 1Mpa. 3 BH29

Depth

BH 70
4 BH 122
Liquefaction and subsidence is major parameters 5

to be evaluated for the structures resting on top 6

soil surface. Detailed liquefaction analysis has to 7

be carried out for sandy strata where earthen and
Fig.5 –Factor of Safety against liquefaction Using SPT
RCC trenches are supported. Detailed liquefaction
N values for various Depth as per FEMA 450
analysis is carried out at a facility for Site specific
(For colour figure, refer to CD)
 Padmanabhan et al. 467

Liquefaction Resistance Using Shear Wave velocity characterization of the site for locating this facility
is of similar nature of that for near surface
Factor of Safety
disposal facility.
0 2 4 6 8 10
0
GEOLOGICAL DISPOSAL OF NUCLEAR
2
WASTE
4 Disposal of high level nuclear waste in deep
BH29
geological repository involves placement of
Depth

6 BH 70
BH122 treated waste at depths of kilometres from the
8 ground surface in appropriate rock formations like
10 granite, charnockite and basalt. A typical diagram
of deep geological repository is shown in Fig 8.
12

Fig.6 –Factor of Safety against liquefaction Using Shear

wave velocity values for various Depth as per FEMA
450 (For colour figure, refer to CD)

A comparison of results shows that Factor of

Safety obtained from SPT N value gives
conservative results when compared to results
obtained from other methods. Figure 7 shows the
factor of safety obtained from different methods
suggested by IAEA guidelines and FEMA 450.
Factor of Safety for bore holes using different
methods
IAEA TECDOC 1347
Method1 -BH29
Factor of Safety IAEA TECDOC 1347
0 2 4 6 8 10 Method 2-BH 29
FEMA 450 Using SPT
0 N values-BH29
IAEA TECDOC 1347
Method 1-BH70
2 IAEA TECDOC 1347
Method 2-BH70
FEMA 450 Using SPT
Fig.8 –Schematic diagram of deep geological repository
4 N values -BH70
IAEA TECDOC 1347
(For colour figure, refer to CD)
Method 1-BH 122
Depth

IAEA TECDOC 1347

6 Method 2-BH 122
FEMA 450 Using SPT
The Indian programme on geological repository is
8
N Values -BH 122
FEMA 450 Using
in initial stage and involves extensive geo
Shear Wave -BH29
FEMA 450 Using scientific investigations. The site characterization
Shear Wave -BH 70
10 FEMA 450 Using for deep geological repository is a challenge
Shear Wave -BH122
12
involving multi disciplinary engineering. Issues
relating to site investigation involve
Fig.7 –Factor of Safety against liquefaction for various
geomorphology, stratigrpahy, structural geology,
Depth (For colour figure, refer to CD)
tectonics, seismicity, rock mechanics, hydrology
FACILITY FOR INTREIM STORAGE OF and geo chemistry, waste-host rock interactions
WASTE and coupled thermal, geochemical, hydraulic,
Treated and conditioned High level waste is to be mechanical and geochemical interactions.
stored in facilities to reduce the decay heat
produced by these waste. Generally these facilities The characterization of a site for locating deep
are co located along with Waste Treatment plants. geological repository includes seismic cross hole
These structures are to be designed for site tests and hydraulic techniques to evaluate the
specific earthquake level 1 and the permeability characteristics of parent rock
formation. Modelling of ground water movement
468 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
is also required to evaluate the hydraulic
properties of the parent rock. Model has to be
conceptualized with the help of detailed
geological and geotechnical data and water level
fluctuation observed in the study area. The
construction of repository also imposes challenges
to the geotechnical engineering in terms of
construction of tunnels and thermal behaviour of
parent rock in the presence of nuclear waste.
CONCLUSION
The nuclear waste generated from different phases
of fuel cycle are treated, conditioned and stored in
safe places. Conditioned wastes are stored in Near
surface disposal facilities and high level wastes
are stored in Interim storage facilities pending
disposal in deep geological repository. Various
aspects of site selection as discussed earlier are
essential in identifying a potential site for locating
these facilities. Geotechnical characterization of
candidate site, hydro geological investigations are
also to be carried out in detail to assess the
suitability of the site for locating these facilities so
that there is no possible release of radio activity to
the environment. In Indian Scenario the nuclear
waste disposal sites are co located with Nuclear
Power Plants and other facilities and utmost care
is taken while selecting the site to minimize the
release of radioactivity to the environment.
REFERENCES
Arya S Design of structures and foundations for
vibrating machines
Bowles,J.E, Foundation analysis and design.
IAEA Safety Series No 50-SG-S8. Safety aspects
of foundations of Nuclear power plants.
IAEA TECDOC-1347 – Consideration of external
events in the design of nuclear facilities other
than nuclear power plants with emphasis on
earthquakes.
IS 456-2000- Code of practice for plain and
reinforced concrete.
IS 12070- Design and construction of Shallow
foundations on rocks
Ronald D.Andrus et.al. “Guide for Shear wave
based liquefaction potential evaluation”,
Earthquake Spectra, Volume 20, No2, pp.285-
308, May 2004.
USNRC guidelines 1.132 – Site investigation for
foundations of Nuclear Power Plants.
ACKNOWLEDGEMENT
The authors wish to express their
acknowledgement to Department of Geology of
Anna University, Chennai for carrying out the
required hydrogeological investigations.
Design Concept in the Revised Technical Manual for Seawalls of Waste
Disposal Facilities in Japan
Y. Watabe
Port and Airport Research Institute, Yokosuka, Japan (watabe@ipc.pari.go.jp)
K. Yamada
Institute of Technology, Penta-Ocean Construction Co., Ltd., Nasushiobara, Japan
T. Oki
JFE Steel Corporation, Tokyo, Japan

ABSTRACT In the coastal waste disposal facilities, most of the reclaimed wastes are submerged, and the
facilities are very flat as if they were containers without outlet. In this paper, the performance of the
hydraulic barrier seawall for the coastal waste disposal facilities is described from the different two points
of view: one is the hydraulic barrier function, which has been stated as the minimum requirement in the
structural code, and the other is the fail-safe system, which has been stated in the new technical manual,
with the functions of backup, proofing/monitoring, restoration, etc.

INTRODUCTION
In Japan, waste disposal facilities for municipal
solid wastes and/or industrial wastes have a
requirement to fulfill the ministerial ordinance
revised in 1998 (Prime Minister’s Office and
Ministry of Health and Welfare, 1998). The
ministerial ordinance is called as “Structural Code
1998” hereafter. Today, Structural Code 1998 is
promulgated from the Ministry of the Vertical hydraulic
Rubble mound
Environment. Structural Code 1998 has first and backfill rocks
barrier (Sheet piles) Hydraulic barrier
quantitatively described the performance of layer (clay deposit)

hydraulic barrier, which had been vague
instructions. In the descriptions, the double or
composite liner structure is required as the bottom
liners, indicating that the fail-safe concept has
been introduced.
In the inland waste disposal facilities, the leachate
from the bottom of the facility is collected by
gravitational flow through the drainage system,
and purified at the treatment plant. In contrast, in
the coastal waste disposal facilities, most of the
reclaimed wastes are submerged because of high
inside water level, and the facilities are very flat
as if they were containers without outlet (Fig.1).
Fig.1. Coastal waste disposal facilities.
The Structural Code 1998 has been established
through discussions mainly on the inland waste
disposal facilities; however, it is directly applied
to the coastal waste disposal facilities. Because
the coastal waste disposal facilities are generally
constructed as a part of port facilities, they also
have a requirement to fulfill the technical standard
for port facilities (The Japan Port and Harbour
Association, 1999; 2007).
The seawalls of the waste disposal facilities are
generally subjected by external forces, which
470 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
always vary with time; e.g. tide, ocean waves,
high tide water, tsunami, ocean stream, etc. The
seawalls require the performances as not only the
shore protection but also the hydraulic barrier. In
the technical standard for port facilities, the
hydraulic barrier seawall are required not to leak
out the wastes and contaminated liquid under the
external forces.
In conventional type shore protections, the
stability is generally ensured by installing a
permeable structure in order to decrease the
differential water pressure; e.g. a caisson is placed
on a permeable rubble mound. The performance
of the hydraulic barrier is the opposing role.
There are some difficulties on the interpretation
between two standards for the different purposes,
i.e. Structural Code 1998 and the technical
standard for port facilities. Under the
circumstances, a technical manual has been
published under the editorship of the Ports and
Harbours Bureau of Ministry of Transport
(Waterfront Vitalization and Environment
Research Center, 2000). The ministry was
renamed as the Ministry of Land, Infrastructure,
Transport, and Tourism as a part of the
government reorganization. The manual is called
as “Technical Manual 2000” hereafter.
The performance-based reliability design method
is expected to be widely adopted in design codes
for various facilities. In response to this global
trend, the technical standard for port facilities
(The Japan Port and Harbour Association, 1999)
has also been revised to introduce the
performance based design (The Japan Port and
Harbour Association, 2007). In addition, the
evolution of the hydraulic barrier technology has
been significant after the publication of the
Technical Manual 2000. Consequently, the
technical manual has been revised in 2008. The
revised manual is called as “the Technical Manual
2008” hereafter.
STRUCTURAL CODE 1998
According to Structural Code 1998, the hydraulic
barrier layer for the waste disposal facilities is
defined as a layer with a hydraulic conductivity
smaller than 1×10–7 m/s and a thickness of greater
than 5 m, or equivalent.
The hydraulic liners are described as follows: a) a
combination of a clay liner with a thickness
greater than 0.5 m and hydraulic conductivity
smaller than 1×10–8 m/s or equivalent, and
impermeable geomembrane liner; b) a
combination of an asphalt concrete layer with a
thickness greater than 0.05 m and hydraulic
conductivity smaller than 1×10–9 m/s or
equivalent, and an impermeable geomembrane
liner; and c) double layered impermeable
geomembrane liners with an intermediate buffer
layer such as non-woven fabric or others. The
performance of the impermeable layer, such as a
clay or asphalt liner, must be equivalent on the
leachate flux to that of the hydraulic barrier layer.
The intermediate layer must be installed in order
to prevent the simultaneous damage on the double
geomembranes. All of these descriptions aim to
ensure double layered impermeable liners. These
are some kinds of backup system based on the
fail-safe/fault-tolerance concept.
Meanwhile, the vertical hydraulic barriers are
required as follows: a) chemical grouting with a
Lugeon unit smaller than unity; b) diaphragm wall
with a thickness greater than 0.5 m and hydraulic
conductivity smaller than 1×10–8 m/s; c)
impermeable sheet pile wall. The performance of
the diaphragm wall is equivalent on the leachate
flux to that of the hydraulic barrier layer.
In the coastal waste disposal facilities,
impermeable geomembrane liners are available on
the back side surface of rubble mound seawall or
backfill; however, vertical hydraulic barrier walls
consisted of steel piles or caissons are used in
most cases. From the standpoints of workability,
cost, and capacity, vertical hydraulic barrier
seawalls consisted of steel sheet piles or steel pipe
piles are preferably used in the coastal waste
disposal facilities. The impermeable vertical
seawalls are considered one of the most useful

methods for the hydraulic barrier of waste
disposal facilities, because the steel seawalls can
be embedded deeply into the hydraulic barrier
layer. In Structural Code 1998, “impermeable
sheet pile wall” is required; however, there is no
quantitative description on the performance.
TECHNICAL MANUAL 2000
The Technical Manual 2000 has indicated that the
equivalent performance of hydraulic barrier layer
means equivalent on travel time t as expressed
with the following equation, not equivalent on
flux:
L2
t (1)
k ˜ 'h
where k is the hydraulic conductivity, L is the
thickness of the hydraulic barrier layer, and 'h is
the differential hydraulic head. According to this,
the hydraulic barrier layer described in the
Structural Code 1998 (k ” 1×10–7 m/s and L • 5
m) can be converted to k ” 1×10–8 m/s and L • 1.6
m. If the performance of the layer does not satisfy
this requirement, an additional artificial bottom
liner must be installed.
TECHNICAL MANUAL 2008
In both Structural Code 1998 and Technical
Manual 2000, the performance of the vertical
hydraulic barrier has not been clearly quantified.
Consequently, in Technical Manual 2008, the
performance of the vertical hydraulic barrier has
been quantitatively described. The vertical
hydraulic barriers are required to be with a
thickness greater than 0.5 m and hydraulic
conductivity smaller than 1×10–8 m/s. The
required performance of the vertical hydraulic
barrier is equivalent to that of the diaphragm wall
whose performance has been quantitatively
described in Structural Code 1998. However, “the
hydraulic conductivity” has to be replaced by an
equivalent parameter, because the hydraulic
conductivity cannot be directly defined for the
vertical hydraulic barrier consisted of such as the
steel piles.
Watabe, Yamada and Oki 471
In practice, the performance of the vertical
hydraulic barrier has been quantified by the
equivalent hydraulic conductivity ke, which is
defined as the following equation:
q L
ke ˜ (2)
B 'h
where q is the leachate flux per unit depth of a
joint, B is the interval distance of the joint, 'h is
the differential water head, and L is the virtual
thickness of the hydraulic barrier work (L = 0.5 m
is generally used for the coastal waste disposal
facilities). When the equivalent hydraulic
conductivity ke is smaller than 1×10–8 m/s for the
sheet pile wall, its performance is considered to
satisfy the description in Technical Manual 2008.
This is a unique concept for the coastal waste
disposal facilities, and an additional requirement
on Structural Code 1998. However, it is very
important on the application of Structural Code
1998 to the coastal waste disposal facilities,
because the ordinance has been established for
mainly the inland waste disposal facilities.
The fail-safe system has been introduced for both
the performance restoration and damage
minimization in case of defects or damages. In
addition, Technical Manual 2000 was published
after Structural Code 1998, most of the advanced
technologies had not yet been developed.
However, because various new technologies have
been developed up to today, Technical Manual
2008 has introduced these new technologies.
The following sections provide a detailed
explanation about the enhancement of the
fail-safe system and introduction of new
technology.
FAIL-SAFE SYSTEM
The enhancement of the fail-safe system is
stressed in Technical Manual 2008. Because the
fundamental performance of the seawall of the
waste disposal facilities is the hydraulic barrier,
the fail-safe system is required to be added in case
of damages on the hydraulic barrier. The fail-safe
472 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
system includes the backup system directly added
to the hydraulic barrier; the leachate control by
lowering the inside water level; the proofing/
monitoring of the impermeability; the restoration
function, etc.
The functions of the hydraulic barrier seawall are
listed below:
1) Hydraulic barrier function
The hydraulic barrier is the most fundamental
function to prevent the pollution of water
resources and/or ground waters.
2) Backup function
The backup function is a preparation in case of
damages on the main hydraulic barrier; e.g.
additional low-permeable material liner in
order to reduce the leachate flux; additional
parallel hydraulic barrier in order to reduce the
leaking risk; etc. The double installed hydraulic
barriers ensure the performance of the function
in case that one of the two barriers was
damaged. The double layered impermeable
liners described in Structural Code 1998 are
classified into this function.
3) Hydraulic gradient decreasing function
The leachate flux can be decreased by lowering
the inside water level, because water flow is
controlled by the magnitude and direction of
the hydraulic gradient.
4) Restoration function
The repairability including the self-restoration
mechanism is a key function in case of
damages.
5) Proofing and monitoring functions
Proofing function for the hydraulic barrier
performance is required when the installation
of the hydraulic barrier is completed. If it was
realized to eliminate any incipient defects, the
performance of the hydraulic barrier would be
ensured. In addition, monitoring function is
also useful in order to determine the need for
additional countermeasures in order to reduce
the leaking risk; such as restoration of the
hydraulic barrier, inside water lowering, etc.
Despite these advantages, both the proofing
and monitoring functions have not been
conducted in practice, because these have been
difficult to be done by the previous
technologies. After Technical Manual 2000,
new technologies have been developed to
realize these functions, and they have been
introduced in Technical Manual 2008.
6) Damage prevention function
Damage prevention function is useful against
not only either the asperity of the foundation or
the angular particles in the waste, but also large
deformation caused by either subsoil settlement
or lateral soil movement. In the coastal waste
disposal facilities, the countermeasures against
either wave force or uplift force are also
included in the damage prevention function.
The functions from 2) to 5) are classified into the
fail-safe systems.
In Technical Manual 2008, the vertical hydraulic
barriers with fail safe system are classified into
four types corresponding to both the joint
mechanisms and material combinations (Table 1).
In all the four types, the main hydraulic barrier
function is realized by joint treatment with
swellable grout for the steel sheet piles or mortal
grout for steel pipe piles. A different combination
of material and member is used in each type. In
type 1, the hydraulic barrier is realized by the
combination of the completely different two
impermeable systems, i.e. one is the steel sheet
piles or steel pipe piles and the other is
low-permeable layer. In type 2, the joint is
reinforced with grout-filled additional arc member.
In type 3, the treated joints of steel pipe piles and
H-shape steel piles are doubled. This type can
make a combination of the two different
impermeable systems, if the inside space between
the double joints is filled with an low-permeable
material. In type 4, the steel sheet pile or steel
pipe pile walls are installed double.
 Watabe, Yamada and Oki 473

Table 1. Fail-safe systems described in Technical is ideally profitable, because the clay is not only
Manual 2008. deformable but also non-deteriorating. A mixture
6 [RG (WPFCO GPVCN $ CEMWRU[UVGO 
$  'ZCO RNGU HQTVJGHCKNUCHGU[UVGO
of dredged clay and bentonite has been proposed
U[UVGO 
( as a hydraulic barrier material (Yamada et al.,
 6TGCGCVGF
GFLQKPV
PVU $ CEMHK
C NNY KVJ
VJ
HKNNGFY KVJ
VJ KO RGTO
RGT GCDNGQT
Waste Sea 2003). The variation of hydraulic conductivity
O QTVCNITQWV
I NQY RGTO
RGT GCDNG B B
QT
QT O CVGTKCN
F
F
due to sand content has been examined (Watabe et
UY GNNCDN
ITQWV
IT
CD G al., 2010). A field proof test for the vertical
 # FFK
F VKQP
QPCNLQKPVO GO DG
HKNNGFY
GF KVJ
VJ
DGT
hydraulic barrier of the mixture was executed at
KO RGT
RGTO GCDNG

Q NQY RGT

QT RGTO GCDNG
G
F Kure Port. Inside of “H” shaped steel pile wall
F
B
O CVGTKCN (0.5 m in thickness) was filled by the mixture of
 # FFK
F VKQP
QPCNLQKPVY KVJ
VJC
HKNNGFY KVJ
URCEGHK VJ
dredged clay and bentonite as shown in Fig.3. In
KO RGTO
RGT GCDNG

QTNQY RGT

Q RGTO GCDNG
G
F
B
practice, the joints are waterproofed with
F
O CVGTKCN expansive agent; however, in the proof test, the
 & QWDN
WDNGKO RGT
DCTTKGT
DC
RGTO GCDNG
GTY CNNU
joints were hydraulically conductive because the
F aim of the test is to evaluate the performance of
F B
the mixture. The equivalent hydraulic
conductivity ke smaller than 1×10–9 m/s was
realized.
RECENT DEVELOPMENTS OF VERTICAL
HYDRAULIC BARRIER TECHNOLOGY
In consideration of both the workability and
economical efficiency, the steel sheet pile is used
for a shallower depth and the steel pipe pile is
used for a deeper depth. Because the steel pile
itself is completely impermeable, the leakage path
is limited to a part of the joint. Therefore, (a) (b)
hydraulic barrier performance of the joint had to Fig.2. Improvements for the joint: (a) Sheet pile joint
be improved, and very useful methods have been with a pocket for waterproof grouting; (b) P – T joint
proposed (Watabe et al., 2007). using waterproof rubber plate filled up with mortar.

The performance as a hydraulic barrier of a joint

filled with expansive waterproof agent and P
(pipe) – T joint of steel pipe pile filled with
mortar have been evaluated in the laboratory
(Kinoshita et al., 2003). The equivalent hydraulic
conductivity ke of these two joint types showed
smaller than 1×10–8 m/s. Some improvements for
the joint are attempted; e.g. a joint with a pocket Fig.3. “H” shaped steel pile wall filled by the mixture
for waterproof grouting/rubber (Fig.2a), and a P – of dredged clay and bentonite
T joint with waterproof rubber plate (Fig.2b).
Field proof tests to evaluate the performance of Various arrangements of steel sheet piles or steel
joints of sheet piles were carried out in Kure Port, pipe piles are proposed as shown in Fig.4. Fig.4a
Hiroshima, Japan (Watabe et al., 2007). The shows an example of double-installed hydraulic
equivalent hydraulic conductivity ke smaller than barrier walls similar to the fail-safe concept
1×10–9 m/s was realized in the field. described about the bottom-liner in Structural
Code 1998. Fig.4b shows examples of
Utilization of natural clay for the hydraulic barrier single-installed hydraulic barrier with double
474 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

joints. The small inside space between the two REFERENCES

joints can be used for proofing and/or monitoring. Kinoshita, M., Oka, Y., Kita, H. and Yoshida, M.
Even if there is a defect or damage in the 2003. Cutoff wall using steel (pile) piles for
hydraulic barrier seawall, the joints can be filled landfill in coastal area, Tsuchi-to-Kiso, Vol.51,
by a low-permeable flexible material for the No.8, 34-36. (in Japanese)
restoration. Prime Minister’s Office, and Ministry of Health
㧔ㆤ᳓Ꮏᢿ㕙଀㧕 㧔ㆤ᳓Ꮏᐔ㕙଀㧕
Secondary
Primary
and Welfare (1998): Ministerial ordinance
Secondary Primary
barrier barrier determining engineering standards pertaining
barrier
╙ㆤ᳓Ꮏ ╙ㆤ᳓Ꮏ
barrier
to final disposal sites for municipal solid
Backfill
 wastes and/or industrial wastes. (in Japanese)
⼔ጯ
Dike 㓒ო
The Japan Port and Harbour Association (1999):
ᑄ᫈‛
Waste
ᶏᐩ㕙
Seabed Technical standards and commentaries for port
and harbour facilities. (in Japanese)
The Japan Port and Harbour Association (2007):
㓒ო
ㆤ᳓ጀ
Hydraulic barrier layer Technical standards and commentaries for port
(a) An example of double-installed steel pile wall: steel and harbour facilities. (in Japanese)
sheet pile wall and steel pipe pile wall Watabe, Y., Kinoshita, M., Yamada, K. and Oki,
: Impermeable
㧔ㆤ᳓Ꮏᢿ㕙଀㧕 㧔ㆤ᳓Ꮏᐔ㕙଀㧕 ㆤ᳓᧚
material T. (2007): Field performance and advanced
Backfill Primary
ㆤ᳓Ꮏ
barrier
Example
㧨଀㧝㧪 1 Example 2 Example 3 Example 4
㧨଀㧞㧪 㧨଀㧟㧪㧨଀㧠㧪 application of impermeable steel seawalls for
 confined disposal facilities in coastal areas,
⼔ጯ
Dike
JSCE Journal of Geotechnical and
ᑄ᫈‛
Waste

Geoenvironmental Engineering, 63(3),

ᶏᐩ㕙
Seabed
662–676. (in Japanese)
Watabe, Y., Yamada, K. and Suitor, K. (2010):
ㆤ᳓ጀ
ㆤ᳓
Hydraulic barrier layer Hydraulic conductivity and compressibility of
(b) Examples of single-installed hydraulic barrier with mixtures of Nagoya clay with sand or bentonite,
double joints. The inter-joint-space can be filled with a Géotechnique, 60. (accepted)
low-permeable flexible material Waterfront Vitalization and Environment
Research Center (2000): Technical manual for
Fig.4. Various arrangements of steel sheet piles or steel
pipe piles design, construction and management for
coastal waste disposal site. (in Japanese)
SUMMARY Waterfront Vitalization and Environment
In the present paper, the performance of the Research Center (2008): Technical manual for
hydraulic barrier seawall for the coastal waste design, construction and management for
disposal facilities was described from the coastal waste disposal site (revised version).
different two points of view: one is the hydraulic (in Japanese)
barrier function, which has been stated as the Yamada, K., Ueno, K., Hada, A., Tsuchida, T.,
minimum requirement in Structural Code 1998, Watabe, Y. and Imai, G. 2003, Development of
and the other is the fail-safe system, which has clayey water interception material in a coastal
been stated in Technical Manual 2008, with the disposal site, Soft Ground Engineering in
functions of backup, proofing/monitoring, Coastal Areas, A.A.Balkema, (Proc. the
restoration, etc. in consideration of combinations Nakase Memorial Symposium), 193-200.
of various hydraulic barrier mechanisms.
 Geotechnical Monitoring of Sanitary Landfills
J.H. Palma G.
Grupo de Geotecnia, Pontificia Universidad Católica de Valparaíso, Chile (jpalma@ucv.cl)
R. Espinace A.
Grupo de Geotecnia, Pontificia Universidad Católica de Valparaíso, Chile (respinac@ucv.cl)
P. Valenzuela T.
Grupo de Geotecnia, Pontificia Universidad Católica de Valparaíso, Chile (pamela.valenzuela@ucv.cl)

ABSTRACT Previous studies done by these authors have shown the stability of landfills may be
assessed through the equilibrium analysis using the traditional geotechnical methods used on soil.
Furthermore, it is plausible to determine and evaluate the resistance parameters during the global
stability analysis through the interaction of conventional geotechnical tests and models. This allows
the monitoring of the stability conditions of a landfill using limit equilibrium methods, analyzing the
resistance parameters and mechanical conditions through traditional geotechnical methods, which
may be used during the operation and abandonment stages of the facility.

INTRODUCTION for the landfill deformations.

The facilities intended to treat the solid wastes
generated in the cities of many countries must Considering the landfill projects must comply
currently fulfill significant technical with high environmental and sanitary security
requirements in order to protect the environment levels, it is necessary to create representative
and the population. Therefore, the antecedents to subsequently evaluate the
environmental geotechnic must actively geotechnical conditions during the operation and
participate in the selection, design, construction, abandonment periods.
operation, closing and rehabilitation stages of the
emplacement. In Chile, as in the developed In this paper, a low-cost methodology to assess
countries, the present trend involves the the geotechnical aspects and compressibility,
construction of intercommunal landfills. In carrying capacity and slope stability of landfills
many cases, landfills become quite significant, is suggested, which has been developed from
encompassing engineering projects and costs study and research experiences carried out during
aimed to avoid impacts on health and the the last 20 years by the authors.
environment.

The proper construction and safety must be GEOTECHNICAL MONITORING OF

ensured during the operational stage of a landfill;
thus, different geotechnical aspects must be
integrated, such as soil of foundation and slope
stability, material characteristics for the residue
covering, determination of the landfill resistance
and evolution of the resistant parameters in time,
compressibility analysis and stabilization periods
LANDFILLS
During the compressibility assessment, settling
data from Chilean landfills is used, as well as the
residue and landfill characteristics, which allow
the application of models to predict settlings,
such as the classical proposal by Sowers and the
Meruelo, Model Arias (1994) that included and
476 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
relates the fundamental parameters like time,
organic material content of residues, and
development humidity and rhythm of decay
processes. Figure 1 shows 10-year settling
estimations for different ages and thickness of
the landfill, which are set out for a Chilean
landfill. These estimations must be verified and
adjusted according to the data obtained from the
settling assessment.
Fig. 1 Settling estimation at a Chilean landfill, 10
years. (For colour figure, refer to CD)
Source: Aguilera, Galeas, 2005.
The load plate tests done on a Chilean landfill for
residues show a relatively lineal performance
until the maximum step and a clear absence of
failure, i.e. there is proportion between the
applied effort and the resulting deformation. The
covering-residue system shows a tight-
deformational curve exhibiting an elastic
response first stage until the covering gadding is
produced. Then, the lineal settling
corresponding to the deformation of the residue
bed is established, which entirely absorbs the
applied load increase without the additional help
of the covering (Figure 2).
Dynamic penetration tests done in Chile on large
landfills have shown homogeneity curves
between the resistance to penetration and depth,
which progressively and lineally increase with
depth and time (Figure 3).
Fig. 2 Results of the load plate tests (plate
diameter 45 cm) a) solid waste; b) covering.
Source: Aguilera, Galeas, 2005.
(For colour figure, refer to CD)
The geotechnical information recorded on the
compressibility, load capacity and slope stability
allows assess the general stability of a landfill.
The permanent geotechnical monitoring of the
slope compressibility and stability helps to
continuously evaluate the stability conditions of
a landfill during the operational and subsequent
closing stages through a security factor
established from the analysis of he resistance
parameters and the mechanical condition.
CONCLUSIONS
The permanent assessment of the geotechnical
aspects of a landfill during the operational and
abandonment stages allows the evaluation of the
stability conditions through the analysis of the
resistance parameters and the mechanical con-
ditions using traditional geotechnical methods.

Fig. 3 Dynamic penetration tests in a landfill in
Chile (For colour figure, refer to CD)
Source: Aguilera, Galeas, 2005.
The generation of information on the geometry
and deformability of a landfill requires the
quality of the deformation measurements to be
reliable on the real performance of the landfill,
trying to avoid precision changes in the
measurement instruments, operational
interventions affecting the position of the
measurement spots and the loss of measurement
continuity, among others.
The solid residue mass may experience an
increase of stress as a consequence of the effect
of the typical weight increase of the materials,
either by height or excessive declivity of slopes
or filling saturation. A decrease on the cutting
resistance of the residues during the
decomposition processes may also occur, causing
a redistribution of the shear effort, which may in
turn become excessive for the intrinsic resistance
of the material and cause landslides during the
Palama G, Espinace A and Valenzuela T 477
exploitation phase or after it has been closed
down. Therefore, the execution of a slope
stability analysis in a landfill results crucial in
order to constantly evaluate a series of factors
that may act individually or in group on the
instability of the filling body.
During the research done by the Geotechnical
Group of the PUCV, which have been carried out
in the most important landfills in Chile, a series
of operational measures have been suggested to
guarantee its stability, considering that according
the national and international experiences one of
the main causes for the low stability are the
leachings created in the fillings, which are
inadequately drained thus causing leaching
accumulation that along to the biogas pressure
increase the interstitial pressure and a gradual
reduction of the mechanical parameters
(cohesion and friction angles). These in turn
cause a reduction of the resistance forces on the
landslide surfaces and increase the weight of the
sliding wedge of the rubbish.
The main recommendations suggested by these
authors to assess the stability of landfills are now
presented, using a low-cost methodology
applicable to countries like Chile:
 During the design stage, the geometry of the
landfill must be determined certifying the
proper Security Factors. During the
operational and closing stages, a periodic
assessment of the geometry and horizontal
movements and settlings of the filling must
be carried out, using inclinometers and
topography, in order to contrast the real
geometry and the original design. Thus, it is
necessary to have a secure area at the landfill
for the location of the movement control
spots.
 During the design stage, the equipotential
levels of the landfill must be determined in
order to certify the Security Factors range
among those previously suggested. During
the operational and closing stages, a
478 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
continuous assessment of these levels must
be executed to determine they provide the
adequate security.
 A constant evaluation of the efficiency of the
biogas and leaching drainage systems must
be done, avoiding their accummulation in the
filling. The biogas and leaching pressure
generate losses in the cutting resistance of the
filling mass.
 An evaluation of the information related to
the allocated residues, international
experiences, back analysis and/or execution
of on site tests (load tests, dynamic
penetration, test pit, etc.) must be done to
choose the resistant parameters to be used
during the analysis and homogeneity, which
allow the evaluation of the geotechnical
conditions of the landfill.
 It is recommended to establish the stability
using limit equilibrium methods, such as
Bishop modified, Janbu or Spencer.
Besides the minimum recommendations
suggested by these authors, which should apply
to all landfills, the Technical Committee TC 5 of
ISSM GE, which included a sub-committee (SC
9) responsible for the assessment of the landfill
performance in view of extreme load conditions
(earthquakes, for instance), has suggested the
evaluation of other failure mechanisms that may
be present in these fillings, such as: distorsion of
the cutting resistance of the foundation residues
or soils; failure of the covering soil; bursting of
the internal slope of the dump; failure of the gas
collection system; geomembrane failure; landfill
landslide bursting caused by failures in the
foundation soil; sub-soil tectonic movements or
liquification of the foundation soil.
ACKNOWLEDGEMENTS
The authors would like to thank the Comisión
Nacional de Investigación Científica y
Tecnológica de Chile (CONICYT) and the
Dirección de Investigación de la Pontificia
Universidad Católica de Valparaíso, for their
support towards the research that gave rise to this
paper.
REFERENCES
Aguilera, K, Galeas, E. (2005) Estudio de la
estabilidad mecánica de los rellenos sanitarios
chilenos. Caso relleno sanitario Loma Los
Colorado. Tesis para optar al título de
Ingeniero Constructor de la Pontificia
Universidad Católica de Valparaíso.
Arias, A. (1994) Modelo de Asentamiento de
Vertederos Controlados de Residuos Sólidos
Urbanos"; Tesina de Magister en Ingeniería
Sanitaria y Ambiental D.C.T.A. y M.A.,
Universidad de Cantabria, España.
Palma, J.H., Espinace, R., Valenzuela, P. (2006)
Análisis de la estabilidad de rellenos sanitarios.
XXX Congreso de la Asociación
Interamericana de Ingeniería Sanitaria y
Ambiental, Punta del Este, Uruguay.
Shafer, A. L., Hargrove, J.Q and Harris, J.M.,
(2003) Stability Analysis for Bioreactor
Landfill Operations. Private communication.
 Swelling Characteristics of Bentonites Under high-alkali Conditions
for Radioactive Waste Disposal
Ko Sugiura
Ibaraki University, Ibaraki, Japan (08nm813f@hcs.ibaraki.ac.jp)
Hideo Komine
Ibaraki University, Ibaraki, Japan (hkomine@mx.ibaraki.ac.jp)
Kazuya Yasuhara
Ibaraki University, Ibaraki, Japan (yasuhara@mx.ibaraki.ac.jp)
Satoshi Murakami
Ibaraki University, Ibaraki, Japan (murakami@mx.ibaraki.ac.jp)

ABSTRACT Radioactive wastes are classified according to their ultimate disposal at high-level or
low-level radioactive waste disposal facilities. Bentonite materials, which have characteristics of
low-permeability and which are self-sealing when swollen by water absorption, are usually used as a
buffer material at disposal facilities. It is expected that engineered barriers might eventually confront
high-alkali conditions from mine timbers, which use concrete materials. For that reason, swelling
characteristics of bentonite might degrade after absorption of high-pH solutions. Therefore, this study
investigated the influence of a high-pH solution on swelling characteristics of commercial bentonite of
three types. Experimental results show the influence of high-pH solution levels from the viewpoint of
initial dry density and montmorillonite contents of the bentonite.

INTRODUCTION 12.5–13.5 (M. Codina et al. 2008)) issues from contact

Radioactive wastes, categorized as high-level
radioactive wastes or low-level radioactive wastes, are
created during electric power generation processes. To
keep them isolated from human life, these wastes can
be discarded at high-level radioactive waste disposal
facilities or at low-level disposal facilities. Figure 1
presents a summary of engineered barriers used at such
facilities. The engineered barrier is expected to be
produced with concrete and bentonite materials; each
material is studied for design of an
engineered barrier. The bentonite material is #EEGUU
VWPPGN
between concrete and groundwater. Swelling
characteristics of bentonite might degrade after
absorption of such a high-pH solution (Jo et al. 2001).
Therefore, to use bentonite for an engineered barrier, it
is important to research swelling characteristics of
bentonite in high-pH conditions. This study
investigated the influence of a high-pH solution on the
swelling characteristics of commercial bentonites of
three types. Specifically, experimental tests for swelling
$CEMHKNN OCVGTKCN
(KNNKPI OCVGTKCN
$WHHGT OCVGTKCN

mainly of low-permeability and is $CEMHKNN

OCVGTKCN
self-sealing after swelling from absorption of
$WHHGT
water (Fig. 2). Bentonite materials of many OCVGTKCN

/KPG VKODGTKPI
kinds have particular properties. After 1XGTRCEM *KIJNGXGN 9CUVG EQPVCKPGT
TCFKQCEVKXG YCUVG .QYFKHHWUKQP OCVGTKCN
construction of engineered barriers, it is
Fig. 1 Engineered barriers (left, high-level radioactive waste disposal;
forecast that a high-pH solution (pH right low-level waste disposal). (For colour figure, refer to CD)
480 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

TABLE 1 Fundamental properties of bentonite materials used this study

Bentonite GX (Kunigel-GX) Bentonite A (Kunigel-V1) Bentonite E (MX-80)
Type Sodium bentonite Sodium bentonite Sodium bentonite
Particle density: Mg/m㧟 2.65 2.79 2.88
Liquid limit: % 355.1 458.1 437.3
Plastic limit: % 22.8 23.7 38.0
Plasticity index 332.3 434.4 399.3
Montmorillonite content: % 41 57 73
Cation exchange capacity (meq/g) 0.854 1.166 1.348
+
Exchange capacity Na (meq/g) 0.521 0.631 0.646
Exchange capacity Ca2+ (meq/g) 0.314 0.464 0.522
Exchange capacity K+ (meq/g) 0.005 0.030 0.038
+
Exchange capacity Mg (meq/g) 0.015 0.041 0.142

pressure and swelling deformation with bentonite GX,
bentonite A, and bentonite E were conducted using a
high-pH solution. These samples of bentonite have
different fundamental properties. The experimental
results showed the influence of the high-pH solution
level from the viewpoint of initial dry density and
montmorillonite content of the bentonites. This paper
has shown swelling characteristics of bentonite GX,
bentonite A, and bentonite E under high-alkali
conditions: experimental tests for swelling pressure
tests and swelling deformation tests were done using
high-pH solutions.
SAMPLES AND SOLUTIONS
Bentonite materials
This study used bentonites of three types. Bentonite
GX is called Kunigel-GX (produced at the Tsukinuno
Mine in Yamagata prefecture, Japan). Bentonite A is
called Kunigel-V1 (Tsukinuno Mine). Bentonite E is
called MX-80 (produced in Wyoming, USA). TABLE
1 presents typical profiles of the bentonites. These
bentonites have characteristics of each material. For
example, bentonite GX contains 41% montmorillonite,
bentonite A contains 57% montmorillonite, and
bentonite E contains 80% montmorillonite.
High-pH solutions
The solution that issues from concrete materials has
high pH: 12.5-13.5. This study supplied distilled water
and aqueous sodium hydroxide of 0.1M NaOH (pH13)
and 1.0M NaOH (pH14) to simulate high-pH
conditions and to explore swelling characteristics of
bentonites under high-alkali conditions. Therefore, this
study did not examine calcium ions exuded from
concrete.

EXPERIMENTAL APPARATUS
Manufacture method of specimen
For this study, bentonite samples were made of
compacted specimens that were 10 mm height with 60
mm diameter. Bentonite samples were compacted with
Fig. 2 Example of bentonite swelling.

static compaction using an oil pressure jack; the
objective initial dry density was 1.50-1.85 Mg/m3.
Experimental apparatuses
Experimental tests were conducted for swelling
pressure tests and swelling deformation tests. Figure 3
portrays the swelling pressure test apparatus and
swelling deformation test. The swelling pressure test
measures pressure that occurs when increased sample
volume from absorption of solution is restricted. In the
experiment, the vertical deformation volume of a
bentonite specimen was measured using a displacement
gauge. The initial dry density was recalculated by the
specimen volume after experiments. Swelling
deformation tests measure the vertical deformation
volume from absorption of solution when the vertical
pressure (=19.6 kPa) is constant. For experiments,
8GTVKECN RTGUUWTGM2C
&KURNCEGOGPV
ICWIG
4KPI
+&OO
.QCF EGNN
*OO
&KURNCEGOGPV
ICWIG
4KPI
+&OO
*OO
5CORNG
&OO
*OO
2QTQWU UVQPG
2QTQWU UVQPG
Fig. 3 Summary of experimental apparatuses (left, swelling
pressure test apparatus; right, swelling deformation test
apparatus). (For colour figure, refer to CD)
silicon grease was added to the inner ring to reduce
friction between bentonite specimens and rings. Both
experiments were conducted during seven days.
EXPERIMENTAL RESULTS
Swelling characteristics of bentonites under
high-alkali conditions
This section described the results of swelling pressure
tests and swelling deformation tests under high-alkali
conditions. Experimental results of swelling pressure
Sugiura et al. 481
tests were arranged with maximum swelling pressure
during the experimental term and initial dry density.
Furthermore, results of swelling deformation tests
showed the swelling strain as defined by eq. (1)
(Komine and Ogata 1994).
'S (1)
Hs u 100
H0
Therein, İs is the swelling strain (%), 'S is the
swelling deformation (mm), and H0 is the initial
specimen height (mm). The maximum swelling strain
is calculated from the approximate curve with the
relation between the swelling strain and time. It shows
the maximum deformation volume during the
experimental test (Komine and Ogata 1994).
Figure 4 shows the relationship between the
maximum swelling pressure and the initial dry density
of each bentonite material. The figures portray the
influence of maximum swelling pressure in each
concentration of high-pH solution. Only the maximum
swelling pressure of bentonite GX decreased in 0.1M
NaOH (pH13). However, the maximum swelling
pressure of bentonite GX and bentonite A widely
5CORNG
decreased because of the effects of 1.0M NaOH (pH14)
‫&ޓ‬OO
*OO
(Figs. 4(a) and 4(b)). The characteristics of swelling
pressure of bentonite depend on the hydrogen ion
concentration of the supplied solution. The maximum
swelling pressure of bentonite E showed little
difference from those of the other samples (Fig. 4(c)).
Figure 5 shows the relationship between maximum
swelling strain and initial dry density of each bentonite
material. The figures portray the influence of maximum
swelling strain in each concentration of high-pH
solution. All bentonite decreased after it was exposed to
0.1M NaOH (pH13) and 1.0M NaOH (pH 14). The
trends show that bentonite swelling characteristics
depend also on the hydrogen ion concentration of the
supplied solution. The characteristics of swelling
deformation were more susceptible than the
characteristics of swelling pressure. In this case,
482 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

3000 5000 8000

㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉 㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉 㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉
Maximum swelling pressure Psmax(kPa)

Maximum swelling pressure Psmax(kPa)

Maximum swelling pressure Psmax(kPa)

NaOH : 0.10M, pH13 㪥㪸㪦㪟㩷㪑㩷0.10M, pH13 NaOH : 0.10M, pH13
2500 7000
NaOH : 1.0M, pH14 4000 NaOH : 1.0M, pH14 NaOH : 1.0M, pH14
㫐㩷㪔㩷㪇㪅㪇㪋㪎㪋㪈㪐㩷㪁㩷㪼㪵㩿㪌㪅㪐㪏㪍㪍㫏㪀 㫐㩷㪔㩷㪇㪅㪉㪊㪐㪏㪏㩷㪁㩷㪼㪵㩿㪌㪅㪊㪋㪉㪌㫏㪀 6000
2000 㫐㩷㪔㩷㪇㪅㪇㪇㪐㪍㪎㪌㪍㩷㪁㩷㪼㪵㩿㪍㪅㪏㪋㫏㪀
㫐㩷㪔㩷㪇㪅㪐㪊㪈㪇㪊㩷㪁㩷㪼㪵㩿㪋㪅㪐㪇㪏㪉㫏㪀
㫐㩷㪔㩷㪇㪅㪊㪌㪉㪇㪋㩷㪁㩷㪼㪵㩿㪌㪅㪈㪊㪉㪊㫏㪀 㫐㩷㪔㩷㪎㪅㪊㪐㪈㪍㩷㪁㩷㪼㪵㩿㪊㪅㪎㪐㪐㪊㫏㪀
㫐㩷㪔㩷㪇㪅㪇㪇㪈㪍㪇㪍㪐㩷㪁㩷㪼㪵㩿㪎㪅㪌㪊㪋㪍㫏㪀 3000 㫐㩷㪔㩷㪇㪅㪇㪇㪎㪐㪌㪐㪏㩷㪁㩷㪼㪵㩿㪎㪅㪇㪋㪍㫏㪀 㫐㩷㪔㩷㪇㪅㪋㪏㪇㪎㪊㩷㪁㩷㪼㪵㩿㪌㪅㪉㪎㪐㪐㫏㪀
5000
1500
2000 4000
1000
3000
1000
500 㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪘 2000
㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪞㪯 㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪜
0 0 1000
1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.85 1.55 1.6 1.65 1.7 1.75 1.8 1.85 1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.85
Initial dry density Ud0(Mg/m3) Initial dry density Ud0(Mg/m3) Initial dry density Ud0(Mg/m3)

(a) Bentonite GX (Kunigel-GX) (b) Bentonite A (Kunigel-V1) (c) Bentonite E (MX-80)

Fig. 4 Relationship maximum swelling pressure and initial dry density.

300 500 300

㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉 㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉 㪛㫀㫊㫋㫀㫃㫃㪼㪻㩷㫎㪸㫋㪼㫉
NaOH : 0.10M, pH13
Maximum swelling strainHsmax(%)

Maximum swelling strain Hsmax(%)

NaOH : 0.10M, pH13 NaOH : 0.10M, pH13

Maximum swelling strain Hsmax(%)

250 NaOH : 1.0M, pH14 250
NaOH : 1.0M, pH14 400 NaOH : 1.0M, pH14
㫐㩷㪔㩷㪄㪎㪌㪅㪍㪌㪉㩷㪂㩷㪉㪈㪋㪅㪊㪏㫏 㫐㩷㪔㩷㪄㪋㪋㪅㪊㪊㪍㩷㪂㩷㪈㪊㪎㫏
㫐㩷㪔㩷㪄㪎㪇㪅㪈㪎㪌㩷㪂㩷㪈㪉㪊㪅㪈㪊㫏 㫐㩷㪔㩷㪄㪈㪎㪇㪅㪉㪋㩷㪂㩷㪉㪉㪉㪅㪉㪍㫏 㫐㩷㪔㩷㪄㪍㪌㪅㪏㪎㪋㩷㪂㩷㪈㪈㪏㪅㪌㪌㫏
200 㫐㩷㪔㩷㪄㪌㪐㪅㪍㪎㪋㩷㪂㩷㪏㪊㪅㪋㪊㪍㫏 㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪞㪯 200 㫐㩷㪔㩷㪄㪍㪉㪅㪍㪈㪋㩷㪂㩷㪍㪎㪅㪌㪐㪈㫏
㫐㩷㪔㩷㪄㪊㪏㪅㪎㪈㪊㩷㪂㩷㪋㪎㪅㪍㪇㪐㫏
㫐㩷㪔㩷㪄㪌㪍㪅㪊㪍㩷㪂㩷㪋㪍㪅㪌㪎㪏㫏 㫍㪼㫉㫋㫀㪺㪸㫃㩷㫇㫉㪼㫊㫊㫌㫉㪼㩷㪑㩷㪈㪐㪅㪍㫂㪧㪸 300
150 150
200
100 100 㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪜
㫊㪸㫄㫇㫃㪼㩷㪑㩷㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪘 㫍㪼㫉㫋㫀㪺㪸㫃㩷㫇㫉㪼㫊㫊㫌㫉㪼㩷㪑㩷㪈㪐㪅㪍㫂㪧㪸
100 㫍㪼㫉㫋㫀㪺㪸㫃㩷㫇㫉㪼㫊㫊㫌㫉㪼㩷㪑㩷㪈㪐㪅㪍㫂㪧㪸
50 50

0 0 0
1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.85
Initial dry densityUdo(Mg/m3) Initial dry density Udo(Mg/m3) Initial dry density Udo(Mg/m3)

(a) Bentonite GX (Kunigel-GX) (b) Bentonite A (Kunigel-V1) (c) Bentonite E (MX-80)

Fig. 5 Relationship maximum swelling strain and initial dry density. (For colour figure, refer to CD)
bentonite specimens absorbed more solution with
swelling pressure because characteristics of swelling
deformation increased volume with absorbed solution
in low vertical pressure during specimen deformation.
Discussion of the reduction of swelling characteristic
of bentonites in high-pH solution
To investigate influences of characteristic of swelling
pressure in the high-pH solution, the “reduction ratio of
maximum swelling pressure” was calculated using eq.
(2). The Rsw-pre(%) reduction ratio was calculated as
Ps max  dw  Ps max  NaOH
Rsw pre
Ps max  dw
where Psmax-dw is the maximum swelling pressure in
distilled water (kPa) and Psmax-NaOH is the maximum
swelling pressure in the sodium hydroxide solution
(kPa).
Figure 6 shows the relationship between the reduction
ratio of maximum swelling pressure and initial dry
density in high-pH solutions. The figure shows that the
highest reduction ratio of bentonite GX was about
40-60%, bentonite A was about 20-55%, and bentonite
E was the lowest––under 10% in the 1.0M NaOH
(pH14). Results show that bentonite E is the most
insusceptible of the three type bentonites.
Montmorillonite contents of bentonite materials are the
following: bentonite GX, 41%; bentonite A, 57%; and
bentonite E, 80%. The latter might be the least
(2)
u 100 susceptible bentonite material by virtue of its higher
montmorillonite content by a high-pH solution.
Additionally, the reduction ratio of maximum swelling
pressure decreases concomitantly with increasing initial
dry density. Results suggest that bentonite, with more
 Sugiura et al. 483

Reduction of maximum swelling pressure Rsw-pre (%)

montmorillonite might be the least susceptible 100
Bentonite GX : NaOH, pH13
bentonite specimen. Bentonite GX : NaOH, pH14

To investigate influences of characteristics of swelling 80 Bentonite A : NaOH, pH14

Bentonite E : NaOH, pH14

deformation in the high-pH solution, the “reduction 㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪞㪯

㪤㫆㫅㫋㫄㫆㫉㫀㫃㫃㫆㫅㫀㫋㪼㩷㪺㫆㫅㫋㪼㫅㫋㩷㪋㪈㩼
60
ratio of maximum swelling strain” was calculated using
eq. (3). The reduction ratio Rsw-def(%) can be calculated 40
㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪘
using the following equation. 㪤㫆㫅㫋㫄㫆㫉㫀㫃㫃㫆㫅㫀㫋㪼㩷㪺㫆㫅㫋㪼㫅㫋㩷㪌㪎㩼

20
H s max  dw  H s max  NaOH (3) 㪙㪼㫅㫋㫆㫅㫀㫋㪼㩷㪜
R sw def u 100 㪤㫆㫅㫋㫄㫆㫉㫀㫃㫃㫆㫅㫀㫋㪼㩷㪺㫆㫅㫋㪼㫅㫋㩷㪏㪇㩼

H s max  dw 0
1.5 1.55 1.6 1.65 1.7 1.75 1.8 1.85 1.9
3
Initial dry density U (Mg/m )
In that equation, Hsmax-dw is the maximum swelling d0

Fig. 6 Reduction ratio of maximum swelling pressure

strain in distilled water (%) and Hsmax-NaOH is the
in high-pH solution.
maximum swelling strain in the sodium hydroxide (For colour figure, refer to CD)
solution (%).

Reduction of maximum swelling strain Rsw-def (%)

160
Bentonite GX : NaOH, pH13
Figure 7 shows the relationship between the reduction 140
Bentonite GX : NaOH, pH14
Bentonite A : NaOH, pH13
ratio of maximum swelling strain and the initial dry 120
Bentonite A : NaOH, pH14
Bentonite E : NaOH, pH13
Bentonite E : NaOH, pH14
density in high-pH solution. In all samples, in the case 100 㫊㫆㫃㫌㫋㫀㫆㫅㩷㪑㩷㪈㪅㪇㪤㩷㪥㪸㪦㪟㪃㩷㫇㪟㪈㪋
of 0.1M NaOH (pH13), the reduction ratio of 80
maximum swelling strain is 30-40%, in the case of 60
1.0M NaOH (pH14), it is from 70-90%. Unlike the 40
swelling pressure test, the difference of the reduction 20 㫊㫆㫃㫌㫋㫀㫆㫅㩷㪑㩷㪇㪅㪈㪤㩷㪥㪸㪦㪟㪃㩷㫇㪟㪈㪊
ratio did not occur in relationship to the difference of
0
the samples. Accordingly, the reduction ratio of 1.5 1.55 1.6 1.65 1.7 1.75 1.8
3
Initial dry density U (Mg/m )
maximum swelling strain of each material is apparently d0

independent of the montmorillonite content. Fig. 7 Reduction ratio of maximum swelling strain
Characteristics of swelling deformation of bentonite of in high-pH solution.
(For colour figure, refer to CD)
various kinds in artificial seawater have been reported
(Komine et al. 2009). Results show that the reduction bentonite specimen (Fig. 8). Specifically, this definition
ratio of maximum swelling strain in artificial seawater means the density of the only montomorillonite in the
decreased as montmorillonite contents increased with specimen. The effective montmorillonite density was
vertical pressure of 1000 kPa. Therefore, effects of calculated as
characteristic of swelling deformation by the high-pH Cm Ue (4)
solution might depend on the montmorillonite content Uem
§ U ·
of bentonite material in the high vertical pressure.
However, for low vertical pressure (19.6 kPa), the
solution effect does not depend on montmorillonite
contents. The reason is the decreased effective
montmorillonite density of the specimens. The effective
montmorillonite density is defined density that mineral
of excluding montmorillonite is excluded in the
¨¨100  100  Cm  e ¸¸
© U nm ¹
where Uem is the effective montmorillonite density
(Mg/m3), Cm is the montmorillonite content (%), Ue is
the effective clay density (Mg/m3)––which means the
dry density in this study––and Unm is the particle density
of component minerals excluding montmorillonite in
484 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

bentonite, which is 2.81(Mg/ m3). O

UGO  GO
Figure 9 presents differences of effective 8GO
montmorillonite density after swelling deformation of
8QKF /QPVOQTKNNQPKVG
each bentonite specimen. Consequently, the effective
#HVGT CDUQTDKPI
8GO
montmorillonite density showed from 0.39 Mg/m3 to 8
OGO /QPVOQTKNNQPKVG GO
0.55 Mg/m3 in the case of 0.1M NaOH (pH 13), and
from 0.75 Mg/m3 to 0.88 Mg/m3 in the case of 1.0M /KPGTCN QH GZENWFKPI OQPVOQTKNNQPKVG
NaOH (pH 14). The results suggest that increased
㨯OGO KU OCUU QH OQPVOQTKNNQPKVG  EQPUVCPV
volume of bentonite specimens settles at the maximum 㨯8GO KU XQNWOG QH OQPVOQTKNNQPKVG CPF XQKF
YCVGT CPF CKT
swelling strain on the effective montmorillonite density KP DGPVQPKVG URGEKOGP

of the constant value in each concentration solution. Fig. 8 Effective montmorillonite density.
Therefore, it might be that the reduction ratio of the 1.6
Bentonite GX : NaOH, pH13
maximum swelling strain of each material did not 1.4
Bentonite GX : NaOH, pH14

after swelling deformationU㫄(Mg/m3)

Bentonite A : NaOH, pH13

Effective montmorillonite density

Bentonite A : NaOH, pH14
depend on the montmorillonite content (Fig. 7). 1.2 Bentonite E : NaOH, pH13
Bentonite E : NaOH, pH14
1 㫊㫆㫃㫌㫋㫀㫆㫅㩷㪑㩷㪈㪅㪇㪤㩷㪥㪸㪦㪟㪃㩷㫇㪟㪈㪋

CONCLUSIONS 0.8
This paper reported characteristics of swelling 0.6
pressure and deformation of three types of bentonite in 0.4
aqueous sodium hydroxide of pH13 and pH14. The 0.2
㫊㫆㫃㫌㫋㫀㫆㫅㩷㪑㩷㪇㪅㪈㪤㩷㪥㪸㪦㪟㪃㩷㫇㪟㪈㪊

following conclusions were drawn from results of these 0

experimental studies.
(1) The influence of high-pH solutions on the
characteristic of swelling pressure decreased
concomitantly with increasing initial dry density
and higher montmorillonite contents in the
bentonite samples. Therefore, samples are
apparently less susceptible to the high-alkali
solutions according to their montmorillonite
contents.
(2) The influence of the high-pH solution on the
swelling deformation characteristics did not depend
on the montmorillonite content in the bentonite
samples and initial dry density. Increasing volume
of the bentonite specimen settles at the maximum
swelling strain on the effective montmorillonite
density of the constant value in solutions of each
concentration.
(3) The influence of swelling characteristics of
bentonite specimens differs in terms of the confined
stress of swelling deformation of the bentonite
1.5 1.55 1.6 1.65 1.7 1.75 1.8
3
Initial dry density U (Mg/m )
d0
Fig. 9 Effective montmorillonite density after
swelling deformation.
(For colour figure, refer to CD)
specimen in high-pH solutions.
REFERENCES
M. Codina et al. (2008) : Design and characterization of
low-heat and low-alkalinity cements, Cement and
Concrete Research 38 (2008) 437–448.
Jo, H.Y., Katsumi, T., Benson, C.H., and Edil, T.B. (2001) :
Hydraulic conductivity and swelling of non-prehydrated
GCLs permeated with single specis salt solutions, Journal
of Geotechnical and Geoenvironmental Engineering,
ASCE, Vol.127, No.7, pp.557-567.
Komine, H. and Ogata, N., (1994) : Experimental study on
swelling characteristics of compacted bentonite, Canadian
Geotechnical Journal, 31 : 478-490.
Komine, H., Yasuhara, K., and Murakami, S. (2009) :
Swelling characteristics of bentonites in artificial seawater.
Canadian Geotechnical Journal, 46 : 177-189.
 Geotechnical Aspects of Municipal Solid Waste Landfill Site:
A Case Study of Surat City
Ms. Nehal Gaurav Dalal
Research Scholar, Applied Mechanics Department, S V National Institute of Technology, Surat – 395
007 Gujarat, India, e-mail: nbn81@yahoo.co.in
Dr Chandresh H Solanki
Assistant Professor, Applied Mechanics Department, S V National Institute of Technology, Surat –
395 007 Gujarat, India, e-mail: chandresh1968@yahoo.co.in
Dr Atul K Desai
Assistant Professor, Applied Mechanics Department, S V National Institute of Technology, Surat –
395 007 Gujarat, India, e-mail: akd@amd.svnit.ac.in

ABSTRACT:

The increased pressure on the environment due to urbanisation causes environmental impacts within geo-related
areas. Considering the vision of a sustainable development, the challenge of the geotechnical community is to
supply more environmental friendly construction techniques, reduce the use of natural resources and promote the
use of less harmful products. Land cost is increasing day by day in urban areas so it became a challenging effort
to put out a strong engineering solution for the sustainable structural stability at the costly places in the urban
areas of covered land filling. Solid waste disposal in landfills is the most economical form of disposal of waste
particularly in the developing country as compared to incineration. Landfill Engineering Design Problems
involve various aspects like determination of geotechnical properties of waste, land fill settlement, slope
stability, field performance of landfill cover and liner systems, seismic behaviour of the disposal site during earth
quake etc. Before buildings or other improvements can be constructed on a landfill site, geotechnical properties
of the landfill waste should be analyzed. This paper presents the geotechnical aspects of Municipal Solid Waste
landfill site of Surat City.
Key Words: Municipal Solid Waste, landfill settlement, seismic behaviour, geotechnical properties.

INTRODUCTION Geotechnical analyses performed for landfill

Now a day’s densely populated region of the
urban centres are under increasing
developmental pressures to use old landfill
sites. As the old and closed landfills are having
the limited end use in terms of recreational uses
like gardens and golf courses, as they are the
most popular and appear to be the best option
for maintaining environmental protection and
ensuring public safety. But it is now the
demand of time to gain some return from the
old landfills like infrastructure, commercial and
low income residential development.
To do the development of completed sanitary
landfills as building sites requires greater
attention for the problem like settlement and
combustible gases resulting from consolidation
and decomposition of waste materials.
design include static and seismic slope stability,
final cover sliding, foundation bearing capacity
and foundation settlement. Foundation settlements
are calculated to evaluate their effects on the
integrity and operation of the landfill system.
Other less conventional analyses performed include
waste compressibility, tension in geoeynthetic materials,
capacity of anchor trenches, geotextile clogging,
geotextile permittivity, geocomposite transmissibity,
pipe crushing and geocomposite collapse.
GEOTECHNICAL PROPERTIES OF MSW
It is important to choose suitably the
engineering properties indexes of the MSW
486 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
during geotechnical engineering analysis. The
local economical development level, customs,
climatic conditions, geological conditions and
filling time can influence the MSW’s complex
physicochemical ingredients and engineering
properties, which are not determined easily.
Determination of geotechnical properties of
MSW is difficult because:
x The inconsistent composition of
landfill material results in widely
variable properties;
x Obtaining samples of sufficient size to
be representative of field condition is
difficult;
x The erratic nature of the waste
particles makes sampling and testing
difficult; there are no generally
accepted sampling and testing
procedures for waste materials; and
x The properties change with time.
The landfill’s various and different ingredients
are the major reasons for its complex
engineering properties.
In general, the organic content gradually
decreases with depth because of thorough
biochemical degradation of the deep organic
matter, but the inorganic matter decomposed
slowly, so its relative content increased with the
reduction of organic content and increased soil
pressure. Also the disposal sites occasionally
are subjected to fire or abnormal heating during
day resulting into disintegration of organic
wastes. Over years it becomes insignificant.
CASE STUDY
Surat Municipal Corporation (SMC) is having a
waste dump fill sits, which was started in 1991
and the dumping activity is closed since 2002.
The area of open dump site is approximate 23
hectors. The amount of waste placed in the
dump site is approximately 35, 45040 MT.
The random core cutter samples were taken
from the site at the depth of 1m from surface.
(Worst condition for foundation at 1 m is
adopted for analysis). The following
observations are made from the samples.
TABLE 1 Geotechnical Parameters of SMC
Waste Dump Site
Parameter Range
Moisture content 12 to 18% in hot season
Dry density 1.33-1.53 gms/cc
Atterberg’s Limit
LL% 69-36
PL% 25-30 (30% of samples
PI% have LL with not possible
to roll in PL test.) 15-20
Free Swell Index Moderate 36 + 3
From table 1 it is shown that the moisture
content is in the range of 12 to 18 %, which is
lower than the values observed in the literature
as the test was conducted in the hot season. The
water content ranged from 20% to 60%.
(Xiang-rong et al., 2003). As depth increases
water content water content gradually
decreased. The water content closely related to
the local climate and the filling mode (Mitchell
et al., 1992). It is high in rainy and humid areas.
It is important for reducing water content to
classify the MSW before filling, to use covering
clay whose coefficient of permeability is less
than 10-7 cm/s, and to keep drainage
unobstructed.
The moisture content in landfills is highly
dependent on several interrelated factors
including: initial composition of the waste;
local climatic conditions; landfill operating
procedure (e.g., use and type of daily cover);
effectiveness of any leachate collection and
removal system; amount of moisture generated
by biological processes within the landfill; and
amount of moisture removed with landfill
gases.
The unit weight of MSW varies within a broad
range because, by its very nature, its
components are highly variable, and placement
procedures as well as environmental conditions
can greatly affect its unit weight. Each landfill
operator handles incoming waste differently
and achieves different levels of compaction.
Common difficulties in assessing MSW unit
weight include:
 Dalal, Solanki and Desai 487

x Separation of the contribution of daily soil The value of ¨Nc/m for the depth between 9.6
cover; to 7.2 m is 20.41 which show that subsoil may
x Assessing the changes in unit weight with be non cohesive silts, silty sand, sand gravel.
time and depth; the majority of reported (C = 0)
values reflect waste near or at the surface;
and Bearing Capacity Calculation:
x Obtaining data on moisture of the waste. Bearing Capacity calculation for foundation at
The typical range was 8 kN/m3 to 14 kN/m3 1 m & 2m for Ø = 0. (from DCPT graph)
(Fungaroli, 1979; Oweis et al., 1986; Landva et (Desai, 2005).
al., 1990; Qian et al., 1994; Timothy et al.,
2000). The value of unit weight for the present 1. For 1 m depth
study is reported between the ranges of 13-15 Nc = 5, Cu =8*Nc = 40 kPa
kN/m3 for the SMC waste dump site.
Safe Bearing Capacity =
Unit weight is also important for seismic
evaluations. The MSW unit weight affects the
= 120 kPa
calculated value of the small strain shear
2. For 2 m depth
modulus (the product of the mass density of the
Nc =10, Cu =8*Nc =80 kPa
material and the square of the shear wave
velocity) and thus influences the natural period Safe Bearing Capacity =
and seismic response of the landfill.
Kavazanjian (2001) notes that the impact of = 240 kPa
unit weight on seismic response “may be
beneficial or detrimental, depending on the
natural period of the landfill and the
predominant period of the earthquake.”
Dynamic Cone Penetration Test (DCPT) (IS
Code DCPT 4968 (Part I) 1976 reaffirmed
1997) was also conducted on the site. From the
graph in fig.2, the profile of the dump site is
summarized below. (Desai, 2005)

1. For depth up to 3.6 m

¨Nc/m = (12-0)/ (3.6-0) =3.33
The value of ¨Nc/m for first 3.6 m is
between 0-8 which shows that the soil
behaves as saturated clay clods in soft matrix
presenting behaviour Ø=0

2. For depth up to 7.2 m

¨Nc/m = (41-12)/ (7.2-3.6) =8.05
The value of ¨Nc/m for the depth between
Fig.1 DCPT Conducted at SMC Waste
3.6 to 7.2 m is 8.05 which shows that the soil
Dump Site
is normally consolidated clay in stiff (For colour figure, refer to CD)
consistency, (increasing sand content C- Ø)

3. For depth up to 9.6 m

¨Nc/m = (90-41)/ (9.6-7.2) =20.41
488 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Fig.2 Result of DCPT at various Depths
CONCLUSION:
Dynamic Cone Penetration Test (DCPT) is the
method which is quickly used to assess
variations of field material with respect to grain
size, density and moisture. It is possible to
conduct the test where normal drilling and
sampling by conventional method is not
possible.
Interpretation of test shows fairly good
representation of soil behaviour for feasibility
study.
The estimated bearing capacity obtained by
DCPT interpretation gives net Safe Bearing
Capacity (SBC) of 120 kPa at a depth of 1 m
and it will be almost double at the depth of 2m.
The effect of rain and environment is analysed
and it is desirable to do the ground
improvement in terms of providing stiffeners
(geofabric or wire mess metress). Further study
for field behaviour is being continued.
REFERENCES
1. Desai, M. D. (2005). “Ground property
characterization from insitu testing”, Indian
Geotechnical Society, Surat Chepter.
2. Fungaroli, A. A., Steiner, R. L. (1979).
“Investigation of sanitary landfill behaviour”,
Vol 1, Final Report, EPA- 600/2-79/053a.
Cincinnati Ohio, US Env. Protection Agency.
3. Kavazanjian, E., Jr. (2001). “Mechanical
properties of municipal solid waste.” Proc.,
Sardinia 2001, 8th Int. Landfill Symp.
4. Landva, A., Clark, J.I. (1990). “Geotechnics
of Waste Fills-Theory and Practice”, ASTM
STP, vol. 1070, pp. 86–106.
5. Mitchell, R.A., Mitchell, J.K. (1992).
"Stability Evaluation of Waste Landfill,"
Stability and Performance of Slope and
Embankments - II, ASCE Geotechnical
Special Publication No. 31, CA, p.1152-
1187.
6. Oweis, I., Khera, R.P. (1986). "Criteria for
Geotechnical Construction of Sanitary
Landfills", International Symposium on
Environmental Geotechnology, 1986.
7. Qian, X. D. (1994). “Analysis of Allowable
Reintroduction Rate for Landfill Leachate
Recirculation”, Lansing Michigan USA:
Michigan department of environmental
quality.
8. Timothy, D. S., Hisham, T. E., Douglas, W.
E., Paul, E. S. (2000). “Municipal solid waste
slope failure II: Stability analysis”, J.
Geotech. Emgrg., ASCE, 126(5), pp. 408-
419.
9. Xiang-rong ZHU, Jian-min Jin, Peng-fei
Fang. (2003). “Geotechnical behavior of the
MSW in Tianziling landfill, Journal of
Zhejiang University.” SCIENCE, May-June,
V. 4, No. 3, pp. 324-330.
 Study of Hydraulic Conditions in Simulated Landfill Bioreactors
Sapna Sethi
Dr.B.R.Ambedkar National Institute of Technology, Jalandhar, India (sethisapna@yahoo.co.in)
N C Kothiyal
Dr.B.R.Ambedkar National Institute of Technology, Jalandhar, India (kothiyalnc@nitj.ac.in)
A K Nema
Indian Institute of Technology, New Delhi, India (ak nema@civil.iitd.ac.in)

ABSTRACT Uniform moisture addition/leachate recirculation is the main requirement for bioreactor
landfills. So the hydraulic conditions in the landfills must be investigated to follow the path of moisture in the
landfills. Present study includes the study of hydraulic conditions in simulated landfill bioreactors using
sodium chloride as tracer. Plexiglass reactors of dimensions 60cm height and 17cm diameter were used as
simulated landfill bioreactors. Reactors were packed with solid waste collected from the dump site at IIT,
Delhi. Reactors were flushed with water and pulse of tracer is applied. Inflow of water was re-established
after the tracer application. Leachate samples were collected at regular time intervals. Study validates the
preferential flow paths in solid waste through the interpretation of breakthrough curves (BTCs).

1.0. INTRODUCTION becomes mandatory in such cases (Kruempelbeck,

In developing countries, larger portion of the
organically rich MSW goes to open dumps; only a
small portion is processed by composting. Open
dumping has already made plenty of damage to
environment in the form of water and air pollution.
To overcome the ill effects of open dumping at
future generations, waste disposal need to be made
environmentally safe and sustainable by adoption
of new scientific methods. Gradually, developing
countries like china and India are shifting from
open dumping to engineered or sanitary landfills
equipped with bottom liners, daily covers, leachate
and gas collection systems. Till now landfills were
operated on dry tomb technology, the principles
described in Subtitle D of the Resource
Conservation and Recovery Act (Federal Register,
1991) where impermeable soil covers are used to
avoid percolation of water and any other liquid to
reduce the quantity of leachate and gas. This dry
tomb technology decreases the rate of anaerobic
decomposition of the biodegradable fraction of
MSW. Due to insufficient moisture and its
distribution in conventional landfills, stabilization
of waste requires several decades to occur. So the
post closure care of landfills for several years
1999).
Presently, the concept of bioreactor landfills has
emerged only in European countries. It employs
leachate recirculation and/or water addition in the
landfill to enhance the rate of biological processes
and stabilize the organic fraction of the waste
within 8 to 10 years rather than several decades as
in conventional landfills. Solid waste degrades
anaerobically under optimum moisture conditions.
According to Solid Waste Association of North
America (SWANA), a bioreactor landfill is “a
controlled landfill or landfill cell where liquid and
gas conditions are actively managed in order to
accelerate or enhance biostabilization of the waste.
The bioreactor landfill significantly increases the
extent of organic waste decomposition, conversion
rates, and process effectiveness over what would
otherwise occur within the landfill.” Faster
stabilization of waste, increased settlement (15-
30% gain in landfill space), higher gas generation
rates are some important applications of bioreactor
landfills (Reinhart (1996); Warith (2002)).
Improved opportunities for leachate treatment and
storage; reduction of post-closure activities; and
490 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
abatement of greenhouse gases are some important
applications of leachate recirculation.
As compared to the simple sanitary landfills, the
bioreactor landfills need modification in terms of
design and operation. Uniform moisture
addition/leachate recirculation is the main
requirement for bioreactor landfills. MSW is
highly heterogeneous and in presence of non
uniform water flow, the waste matrix might be left
with dry spots with delayed stabilization.
Knowledge of the space and time variability of the
water storage in landfills is lacking to a great
extent. The spatial variability of water content is
large, ranging from saturated to dry conditions.
Dry spots regions in waste deposits have been
reported by several researchers (Harris 1979).So
the hydraulic conditions in the landfills must be
considered to follow the path of moisture in the
landfills. Generally the diverse nature of MSW
creates restricted channels and voids through
which moisture movement takes place. These
preferred paths influence the degradation of waste
in the bioreactor landfill, making it unmixed, water
deficient reactor (Bogner & Spokes 1993). Several
lab scale and field studies (Ehrig (1983); Blight et
al.,(1992); Bendz et al.,(1997);Rosqvist et
al.,(1997); Rosqvist et al.,(1999); Rosqvist et
al.,(2005); Uguccioni and Zeiss, (1997)) using
tracer tests have reported the existence of favoured
flow paths in landfills. Channelized flow through
the waste results in only a limited part of the total
waste mass taking part in water movement. So the
leachate quality reflects mainly processes within
the preferential flow paths and the close
surroundings only. Pollution potential of slow
moving paths and dry spots remains unpredicted.
So long term prediction of leachate quality and
quantity becomes imprecise (Benz & Flyhammar
1999).
Tracer tests are often used to trace the path of
flowing water. Ideally, tracer chemicals must
behave conservatively, with the only transport
processes influencing a conservative tracer being
advection, dispersion and diffusion. The most
common tracers used are fluorescent dyes and
halides such as chlorides, bromides and iodide. In
the developing countries, sodium chloride is
readily available and therefore may be suited for
tracer tests. Tracer test study has been conducted
for determination of flow with in landfills using a
non reactive chemical whose mobility is
exclusively determined by the flow of water. The
objective of this study is to investigate the
hydraulic conditions in the simulated landfill
bioreactors containing two types of solid wastes
using sodium chloride as a tracer before operating
them as leachate recirculation simulators. Tracer
travel time distributions were derived from
breakthrough curves (BTCs). Experimental
breakthrough curves indicate the occurrence of
favoured flow paths with in the solid waste. For a
pulse or spike input of dissolved tracer through the
packed bed, the breakthrough curve will usually be
approximately Gaussian.
2.0. MATERIALS AND METHODS
2.1. The solid Waste Material
Fresh solid waste samples were collected from the
dump site of IIT, Delhi. Sample selection was done
by coning and quartering method. Samples were
segregated manually on site with the help of
workers present there. Segregation of samples
were done into various physical components- Fruit
and Vegetable waste, yard waste, paper, plastics,
rags, glass, rubber, leather, metals, inerts etc. After
separation all components were weighed
separately and mixed uniformly. Moisture content
was analyzed immediately.
Two different types of solid waste samples were
used for the present study. The first sample (MSW)
consisted of shredded waste (size less than 4cm)
containing composting material, paper, plastic,
rags etc. as shown in table 1. The second sample
was pure organic waste (POW) consisting of fruit,
vegetable and yard waste only. Pure organic waste
was also shredded to the size less than 4cm (table
2). During the preparation of samples, size
reduction of materials was achieved using knives
and scissors so that moisture content of waste was
not affected. Before filling the reactors, waste
samples were stored at 4°C. Initial moisture
content of MSW and POW was 53% and 70%
respectively.

TABLE 1 Composition of MSW
_______________________________________
Waste fractions Percentage
_______________________________________
Fruit and Vegetable 60
Plastic 8 were flushed
Paper 8 water to decrease the background concentration of
Rubber 2 NaCl in the waste and also to create steady state
Rags 4
Yard waste 7
Inerts 11
_______________________________________
TABLE 2 Composition of POW
_______________________________________
Waste fractions Percentage
_______________________________________
Fruit and Vegetable 96
Yard waste 4 conductivity of leachate matched with the
_______________________________________
2.2. Landfill Simulators Set-up
Experiments were conducted using two reactors (A
& B) of plexiglass with diameter of 17cm and
height 60cm. Each reactor contained a metallic lid
having over head container of one litre capacity
with a stopper to regulate the flow inside the
container. A water distribution system was
installed below the lid of reactor at the top to add
liquid to the waste in the form of spray. Leachate
was collected from the bottom of the reactor in
plexiglass container. A 5cm thick layer of 2.5cm
gravel was placed at the bottom of each reactor for
leachate drainage. Leachate was collected from the
bottom of the reactor. Two nylon screens of 1mm
diameter were placed above and below the
drainage layer to retain MSW and stop particles
from leaching out. Reactor A was filled with
4.4Kg of MSW with wet weight compaction
density of 622Kg/m3 in a 40cm layer. Reactor B
was filled with 6Kg of pure organic waste in a
40cm layer with wet weight compaction density of
850 kg/m3. Headspace of 15cm at the top was left
in each reactor.
Sethi, Kothiyal and Nema 491
2.3. Tracer Experiments
Tracer experiments were conducted to investigate
hydraulic conditions in both the reactors. 11g of
NaCl is dissolved in 150 ml of water and applied
as tracer. Before applying tracer pulse, reactors
with approximately 24 litres of
flow conditions in the reactors. Electrical
conductivity of leachate was measured according
to standard method (APHA et al., 2005). A
relationship between conductivity and
concentration of NaCl was also determined
through another experiment discussed later. Pulse
of NaCl was applied at the rate of 2.5ml per
second. After the pulse application, water was
sprinkled at the same rate till electrical
conductivity that was before the application of
tracer. 6 litres of water was sprinkled for 1800
seconds and 2820 seconds in reactor A and B
respectively and leachate samples were collected at
the regular intervals. Breakthrough curves
(concentration of tracer verses time) were plotted
for these tracer experiments.
3.0. RESULTS AND DISCUSSION
3.1. Conductivity and Concentration of NaCl
An experiment was conducted to determine
relationship between conductivity of leachate and
concentration of NaCl. Conductivity of one litre of
distilled was measured before and after adding 2g
of NaCl. Conductivity was increased to
3.30mS/cm after addition of 2g of NaCl. 1000mg
of NaCl contributes 1.65mS/cm to the conductivity.
So the following equation can be used to convert
conductivity into concentration of NaCl.
NaCl (mg/L) =Conductivity (mS/cm)/1.65x10-3
3.2. Tracer Study Results
In both the reactors, water was supposed to be
flushed for 1800 seconds. But in case of pure
organic waste, compaction density was more so
water flow rate was slower as compared to MSW.
So leachate sample approached the background
492 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India

Fig. 1 Tracer Concentration at Reactor A Fig. 3 Experimental Breakthrough Curve for reactor A
(For colour figure, refer to CD) in terms of normalized mass flow rate, S/M0 where S is
the mass of tracer recovered at time t and M0 is the total
input mass of tracer. (For colour figure, refer to CD)

Fig. 2 Tracer Concentration at Reactor B Fig.4 Experimental Breakthrough Curve for reactor B in
(For colour figure, refer to CD) terms of normalized mass flow rate, S/M0 where S is the
mass of tracer recovered at time t and M0 is the total
input mass of tracer. (For colour figure, refer to CD)

concentration in 2820 seconds in case of pure
organic waste. In reactor A, the background
conductivity was 0.945mS/cm and water was
flushed after tracer application till conductivity
approached near to 0.945mS/cm. In reactor B, the
background conductivity was 1.34mS/cm and
water was flushed after tracer application till
conductivity approached near to 1.34ms/cm. 25-
30 leachate samples were collected for each
experiment. Conductivities of leachate samples
were measured and converted into concentration of
NaCl by the above mentioned method. The
concentration of tracer in leachate verses time
curves were plotted for reactor A and reactor B
given in figure 1and 2 respectively. Maximum
concentration of tracer in reactor A and in reactor
B was observed at 7 minutes and 12 minutes
respectively. Mass of tracer recovered in leachate
was 78% and 75% for A and B reactors
respectively.
The shapes of Experimental BTCs plotted for
reactor A and Reactor B in figure 3 and 4 were
positively skewed with tails indicating preferential
tracer transport. The early peaks at 420 seconds
and 720 seconds for reactor A and reactor B
respectively pinpointing rapid tracer flow in
preferential flow paths and tails indicate the slow
mass flow in not favoured , sluggish paths through
the solid waste. Compaction density and nature of
the waste also affected the flow of moisture
through the waste. The maximum concentration of
tracer in leachate reached earlier in reactor A
containing MSW with less compaction density as
compared to reactor B containing pure organic
waste with more compaction density.
3.0. CONCLUSIONS
Present study confirms the occurrence of
channeled or preferential flow in solid waste
landfills. Techniques enhancing waste
stabilization like water/leachate recirculation
through landfills are successful only in presence of
even moisture distribution in landfill. Due to
heterogeneous nature of waste, preferential or
channeled flow in the landfills will ultimately
delay the waste stabilization process in less
preferred paths in the waste. So the techniques
Sethi, Kothiyal and Nema 493
enhancing even moisture distribution in landfills
need to be improved further. Shredding, recycling,
segregation before placing a landfill may be useful
to promote uniform moisture distribution and
reducing differential water flow through the waste.
Nature of the waste and compaction density also
affects the moisture movement in the landfills.
4.0. REFERENCES
APHA, AWWA, WEF (2005) Standard Methods
for Water and Wastewater Examination, 21st ed.
American Public Health Association, Washington,
DC.
Benz, D. & Flyhammar, P. (1999) Channel flow
and its effects on long-term leaching of heavy
metals in MSW landfills. In:Proc.Sardinia
99,Seventh International Waste Management and
Landfill Symposium, Sardinia,Italy,October
1999,vol 2, pp. 43-50.
Bendz, D., Singh, V .P., & Åkesson, M. (1997)
Accumulation of water and generation of leachate
in a young landfill. Journal of Hydrology, 203,
1–10.
Blight, G.E., Ball, J.M., & Blight, J. J. (1992)
Moisture and suction in sanitary landfills in
semiarid areas. Journal of Environmental
Engineering, 118, 865–877.
Bogner, J., Spokas, K. (1993) Landfill CH4: rates,
fates, and role in global carbon cycle,
Chemosphere 26 (1–4), 369–386.
Ehrig, H-J. (1983) Quality and quantity of sanitary
landfill leachate, Waste Management & Research,
1, 53–68.
Federal Register (1991) United States Code of
Federal Regulations 40 CFR Part 258, Subtitle D
of the Resource Conservation and Recovery Act
(RCRA), Criteria for Municipal Solid Waste
Landfills (56 FR 50978).
Fellner, J., Doberi, G., Allgaier, G., Brunner, Paul
H. (2009) Comparing field investigations with
laboratory models to predict landfill leachate
emissions, Waste Management, 29, 1844-1855.
494 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Haris, M.R.R., 1979. A study of behavior of refuse
as a landfill material.PhD. Thesis, Dept. Civ. Eng.,
Portsmouth Polytechnic, 144pp.
Kruempelbeck, I., Ehrig, H. J. (1999). Long Term
Behaviour of Municipal Solid Waste Landfills in
Germany, Proceedings Sardinia 99, the Seventh
International Waste Management and Landfill
Symposium, Cagliari, Italy.
Reinhart, D., Al-Yousfi, A B. (1996) The Impact
of Leachate Recirculation on Municipal Solid
Waste Landfill Operating Characteristics. Waste
Management & Research, 14, 337–346.
Rosqvist, H., Bendz, D., Öman, C., & Meijer, J.-E.
(1997) Water flow in apilot-scale landfill. Proc.
Sardinia 97, Sixth International Landfill
Symposium, Sardinia, Italy, October 1997, vol. 2,
85–96.
Rosqvist, H. & Bendz, D. (1999) An experimental
evaluation of the solute transport volume in
biodegraded municipal solid waste, HESS, 3, 429–
438.
Rosqvist, H., Dollar, A., Fourie, B. (2005)
Preferential flow in municipal solid waste and
implications for long-term leachate quality:
valuation of laboratory-scale experiments .Waste
Management & Resources, 23: 1–14
Uguccioni, M., & Zeiss, C. (1997) Comparison of
two approaches to modeling moisture movement
through municipal solid waste. Journal of
Environmental Systems, 25, 4–63.
Warith, Mostafa (2002) Bioreactor landfills:
experimental and field results, Waste Management,
22(1) 7-17.
 Development of a Field-Validated Landfill CH4 Emissions Inventory
Methodology for California

Jean E. Bogner1
1
University of Illinois at Chicago, Chicago, Illinois, USA (jbogner@uic.edu)

Kurt Spokas
2
U.S. Dept. of Agriculture(USDA)/Agricultural Research Service, St. Paul, Minnesota
USA, (Kurt.Spokas@ARS.USDA.GOV)

Jeffrey Chanton3
3
Florida State University, Tallahassee, Florida, USA (jchanton@fsu.edu)

ABSTRACT We have developed a new field-validated landfill methane (CH4) emissions inventory
methodology for California. CALMIM (CAlifornia Landfill Methane Inventory Model) is an IPCC
(Intergovernment Panel on Climate Change) Tier III site-specific model based on expansion of previously-
published field-validated modeling and measurement approaches, integration with USDA globally-validated
climatic and soil microclimate models, extensive supporting laboratory studies, and field validation over 2
annual cycles at the coastal Marina Landfill (Monterey, CA) and the Scholl Canyon Landfill in the Los
Angeles area. Unlike previous inventory methods based on a theoretical multi-component first order kinetic
model for CH4 generation, CALMIM relies upon a more realistic emphasis on landfill CH4 emissions
inclusive of methanotrophic CH4 oxidation for daily, intermediate, and final cover materials over a typical
annual cycle. The model is a freely-available JAVA tool based on 1-D diffusion inclusive of oxidation
which calculates typical annual emissions for 10-min. time steps and 2.5-cm. depth increments for any
layered cover soil up to about 250 cm. thick. [Version 4.2 of CALMIM is available at
http://www.lmem.us/CALMIM/CALMIM42.exe]. This paper summarizes model development,
supporting laboratory and field studies, and comparisons of model results to field data.

1.1 BACKGROUND National GHG inventory emission estimates for

According to recent global inventories as
summarized in the Intergovernmental Panel on
Climate Change (IPCC) 4th Assessment Report
(Bogner et al., 2007; Monni et al., 2006; US EPA,
2006, Rogner et al., 2007), landfill CH4 emissions
are responsible for only about 1-2% of total global
anthropogenic greenhouse gas (GHG) emissions.
However, in countries with a history of landfilling,
landfill CH4 emissions can be one of the larger
sources of atmospheric CH4; for example, in the
U.S., landfill emissions are the second largest
anthropogenic source of atmospheric CH4 after
ruminant animals (US EPA, 2010). CH4 has a
relatively high 100-year global warming potential
of 25; because it also has a relatively-short
atmospheric lifetime of about 12 years, reductions
in emissions of individual CH4 sources can begin
to affect atmospheric concentrations in about a
decade (IPCC, 2007).
CH4, which are based on current IPCC Tier I and
Tier II national GHG inventory methodologies,
have high uncertainties (IPCC, 2006). These
uncertainties result from methodologies based on a
multicomponent first order kinetic CH4 generation
model relying on: (1) an assumption of
homogeneous internal landfill processes; (2)
uncertain annual national waste data (especially in
developing countries), and (3) questionable
assumptions regarding the “net” CH4 emissions to
the atmosphere including (a) a current allowance
for a maximum of 10% oxidation by indigenous
aerobic methane-oxidizing microorganisms
(methanotrophs) in cover materials, and (b) lack
of inclusion of the site-specific physical properties
of daily, intermediate, and final cover materials to
limit emissions over a typical annual cycle.
The purpose of this project was to develop an
improved, science-based field-validated landfill
496 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
CH4 emissions inventory model for California
(CALMIM, CAlifornia Landfill Methane
Inventory Model). We relied upon well-researched
and accepted theoretical relationships, previous
field and laboratory studies (Spokas et al., 2003,
2006; Bogner et al., 1997, 1999, 2010; Scheutz et
al., 2003, 2009; Liptay et al., 1998; Chanton and
Liptay, 2000), existing globally-validated U.S.
Dept. of Agriculture models for climate and soil
microclimate (Spokas and Forcella, 2006, 2009),
and extensive supporting laboratory studies on
CH4 oxidation completed specifically for this
project to quantify the limits and dynamics of CH4
oxidation in California landfill cover soils (Spokas
and Bogner, 2010). The model was field-validated
over two years at two California landfill sites
(Bogner et al., 2010, submitted), including the
northern coastal Marina Landfill (Monterey
County, CA) and the southern Scholl Canyon
Landfill (Los Angeles County, CA). Field
validation consisted of seasonal field campaigns
(wet/dry) to quantify CH4 emissions from daily,
intermediate, and final cover materials using
randomized static chamber techniques (>800
chamber measurements), as well as supporting
measurements for meteorology, soil temperature,
soil moisture, and CH4 oxidation using stable
carbon isotopic methods (Chanton and Liptay,
2000). Additional limited field validation at the
Lancaster, Kirby Canyon, and Tri-Cities Landfills
was made possible through the cooperation of
Waste Management, Inc., including field
measurements of CH4 emissions using both static
chambers and an above-ground method employing
a TDL (tunable diode laser) instrument for vertical
and horizontal radial plume mapping (Green et al.,
2009).
MODEL STRUCTURE AND METHODS
CALMIM is an IPCC Tier III model (IPCC, 2006)
with four (4) major integrated components (Fig. 1):
(1) Inventory Template;
(2) Meteorological Model;
(3) Soil Microclimate Model;
(4) 1-D Emissions/Oxidation Model.
With regard to (1), a data template inclusive of the
major California cover types, ADCs (Alternative
Daily Covers), and a customized “cover designer”
was developed. Because the major controls on
landfill methane emissions are the thickness and
properties of the cover materials and the
implementation of active landfill gas extraction,
the model requires site-specific input data.
Required inputs include the area of each type of
cover material (daily, intermediate, and final) and
whether active landfill gas extraction is currently
in place for each type of cover material (as % of
area). A customized “cover designer” also permits
any layered sequence of cover materials, including
geomembrane composite cover designs. An
objective of the project was to facilitate data input
by users and minimize the site-specific information
requirements to the most pertinent data needed for
this inventory. Site locations can be queried by
name and are linked to the California SWIS (Solid
Waste Information System) database. With regard
to (2) and (3), the meteorological and soil
microclimate models rely on modified versions of
the following USDA models: Global TempSIM,
Global RainSIM, Solarcalc, and STM2 (Spokas
and Forcella, 2006, 2009). In particular, the soil
temperature functions for STM2 (Soil Temperature
and Moisture2) were modified to accommodate the
landfill heat source. For air temperature, the
latitude and longitude of the site are used to
extrapolate the daily climatic conditions from an
interpolated database based upon 30-yr (1961 –
1990) air temperature records. These records were
previously compiled and interpolated by Legates
and Willmott (1990a and 1990b) with further
improvements by Willmott and Matsuura (1995).
Diurnal temperature ranges were interpolated from
monthly averages from data collected 1961-1990
and cross-validated by New et al. (1999). Average
precipitation was also extrapolated from the
average number of days in a month with rainfall as
compiled and interpolated by Legates and
Willmott (1990a and 1990b) with further
improvements by Willmott and Matsuura (1995).
Average monthly precipitation was taken from the
interpolated data collected 1961-1990 and cross-
validated by New et al. (1999). With regard to (4),
the emissions model is based on 1-D bidiffusional
flux of CH4 and O2 through cover materials where
the driving force is the gas concentration gradient

through the gas-filled pore space (based on
Campbell, 1985; Moldrup et al., 1999, 2000, 2004).
CALMIM California Landfill Inventory Model
...web-based, site-specific JAVA tool with 4 parts:
(1) Site Location, Cover Materials, and Landfill Gas R
(Interactive template)
(2) Environmental Simulation/Meteorology
(Global TempSIM
, Global RainSIM /SOLARCALC)
Field
Validation
and
Supporting (3) Microclimate Model:
Laboratory (Soil temperature and moisture: STM 2)
Studies
(4) Methaneoxidation/Emission Model
(1-Ddiffusion)
ȱ
Fig.
ȱ 1 CALMIM components.
(For colour figure, refer to CD)
Seasonal methane oxidation is modeled through
the use of scaled results relative to maximum rates
over the full range of expected field temperature
and moisture conditions based on extensive
laboratory batch incubations of California landfill
cover soils, as well as published literature. The
supporting laboratory studies consisted of batch
incubations of daily, intermediate, and final cover
materials from Marina and Scholl Canyon.
Methane oxidation rates were determined using a
time-series based on the decrease in headspace
CH4 as related to CH4 exposure history, variable
soil moisture, and soil temperature, including
diurnal temperature fluctuations (>2000
incubations). Further details regarding incubation
methods and results are given in Spokas and
Bogner (2010).
Field validation included extensive measurement
of CH4, CO2, and N2O emissions on daily,
intermediate, and final cover materials at the
Marina and Scholl Canyon sites in dedicated field
campaigns during both March and August of 2007
and 2008. The wet (March) and dry (August) field
campaigns consisted of more than 800 fluxes and
relied on static chamber techniques (Rolston,
Bonger, Spokas and Chanton 497
1986), which is the method of choice for
quantifying the variability of emissions across a
particular cover soil (including both CH4 emissions
and uptake of atmospheric CH4). Meteorological
variables and soil moisture/temperature profiles
were continuously monitored during 2007-2008.
The field measurements at Marina and Scholl
Canyon were supplemented by static chamber and
TDL data at Lancaster, Kirby Canyon, and Tri-
Cities through data-sharing from an independent
field-monitoring project by Waste Management,
Inc. which is focusing on field measurement of
landfill CH4 emissions. The fractional CH4
oxidation at all 5 sites was quantified through the
use of stable carbon isotopic methods developed
by Chanton and colleagues (Liptay et al., 1998;
Chanton and Liptay, 2000; Chanton et al., 2008)
which rely on the temperature-dependent
preference of methanotrophs for the isotope of
smaller mass (12C).
RESULTS AND DISCUSSION
Replotting of data from 7 well-monitored landfill
cells with engineered gas recovery (Spokas et al.,
2006) indicates that field measurements of
emissions cannot be correlated to modeled
emissions using a first order kinetic model for CH4
generation, which is the current IPCC Tier I
default model. Indeed, using the Tier I model for
our field validation sites indicated that the current
measured CH4 recovery greatly exceeded the
modeled generation. Using CALMIM, a
comparison of field-measured emissions to
modeled “net” CH4 emissions inclusive of
oxidation generally replicated field measurements,
in the worst case, to within the same order of
magnitude as field measurements. This is a
significant improvement in view of global
literature spanning 6-7 orders of magnitude
variability for field measurements of landfill CH4
emissions. Importantly, CALMIM also includes
the effect of engineered gas extraction through an
adjustment in the gas concentration gradient
through cover materials—and more than >90% of
the waste in place in California landfills has
engineered gas recovery. In most cases, CALMIM
replicated field results to better than an order of
magnitude over a typical annual cycle and was also
498 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
able to replicate observed seasonal variability.
Differences with field data were related to an
extreme drought year with respect to soil moisture
data and areas of high relief where the nearest
weather station data were not representative of
field conditions at the site. CALMIM includes an
“advanced” option to input additional site-specific
data, as available, to better match field conditions.
We suggest that an important field parameter for
field measurement is the CH4 concentration
gradient through daily, intermediate, and final
cover materials, as a direct measure of the driving
force for emissions. Because soil moisture- and
temperature-related transport properties in
CALMIM are largely based on the agricultural
literature, emission calculations are conservative
for landfill soils, which are more highly compacted
than agricultural soils. Another conservative
aspect of CALMIM is that boundary conditions do
not allow uptake of atmospheric CH4 (negative
emissions) while field data indicated that for 4 of
the 6 covers investigated, there were significant
seasonal percentages of negative fluxes. Caveats
with respect to the use of CALMIM include:
(a) As an inventory model for a typical annual
cycle for specific cover materials and gas recovery
status, CALMIM is not intended to replace
regulatory guidelines and models which address
shorter-term landfill operational practices (e.g.,
time periods less than 1 year);
(b) CALMIM is also not intended to replace
existing methods or models used for monetization
of carbon credits resulting from landfill gas
recovery projects.
ACKNOWLEDGEMENTS
Financial support for this project was provided by
the California Energy Commission (CEC) Public
Interest Energy Research (PIER) Program
[Contract # 500--05-039], G. Franco, Program
Manager. We gratefully acknowledge the support
of the Los Angeles County Sanitation District, the
Monterey Regional Waste Management District,
the former California Integrated Waste
Management Board (now Calrecycles), the Air
Resources Board (ARB), Waste Management, Inc.,
and the following individuals: Martin duSaire,
Nancy Barbour, Dean Peterson, Paul Roots, Tim
Badger, Chad Rollofson, Tia Phan, Lindsay
Watson, Lianne Endo, Kia Young, Mai Song Yang,
and David Hamrum.
REFERENCES
Bogner, J., Meadows, M., and Czepiel, P. (1997).
“Fluxes of Methane between Landfills and the
Atmosphere: Natural and Engineered
Controls.” Soil Use and Management 13:268-
277.
Bogner, J., Spokas, K., and Burton, E. (1999),
“Temporal Variations in Greenhouse Gas
Emissions at a Midlatitude Landfill, J.
Environ. Qual. 28: 278-288.
Bogner, J., Ahmed, M.A., Diaz, C., Gao, Q., Faaij,
A., Hashimoto, S., Mareckova, K., Pipatti, R.,
and Zhang, T. (2007) “Waste Management.”
In Climate Change 2007: Mitigation,
Contribution of Working Group III to the
Fourth Assessment Report of the
Intergovernmental Panel on Climate Change;
Metz, B.; Davidson, O.R.; Bosch, P.R.; Dave,
R.; Meyer, L.A.,Eds.; Cambridge University
Press, Cambridge, United Kingdom and New
York, NY USA.
http://www.ipcc.ch/publications_and_data/ar4
/wg3/en/ch10.html
Bogner, J., Spokas, K., and Chanton, J. (2010),
“Field Measurement of CH4, CO2, and N2O
Emissions from Daily, Intermediate, and Final
Cover Materials at Two California Landfills.”
submitted to Environ. Sci. & Tech.
Campbell, G.S. (1985) Soil Physics with Basic:
Tranport Models for Soil-Plant Systems,
Elsevier Science Publishers, Amsterdam.
Chanton, J., and Liptay, K. (2000) “Seasonal
Variation in Methane Oxidation in a Landfill
Cover Soil as Determined by an in situ Stable
Isotope Technique,” Global Biogeochemical
Cycles 14: 51-60.
Chanton, J.P., Powelson, D.K., Abichou, T., and
Hater, G., (2008) Improved field methods to
quantify methane oxidation in landfill cover
materials using stable carbon isotopes. Environ.
Sci. and Technol. 42: 665-670.
Green, R., Hater, G., Goldsmith, C.D, Chanton, J.,
Swan, N., and Abichou. T. (2009) “Estimates
of Methane Emissions from Three California

Landfills using Two Measurement
Approaches” Proceedings First International
Greenhouse Gas Measurement Symposium,
March 22-25, 2009, San Francisco, CA, Air &
Waste Management Association, Pittsburgh,
PA, USA.
IPCC (2006) 2006 IPCC Guidelines for National
Greenhouse Gas Inventories. IPCC/IGES.
Hayama, Japan. http://www.ipcc-
nggip.iges.or.jp/public/2006gl/vol5.html
IPCC (2007) Climate Change 2007 - The Physical
Science Basis: Contribution of Working Group
I to the Fourth Assessment Report of the IPCC
[Solomon, S., D. Qin, M. Manning, Z. Chen, M.
Marquis, K.B. Avery, M. Tignor and H.L.
Miller (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York,
NY, USA.
http://www.ipcc.ch/publications_and_data/ar4/
wg1/en/contents.html
Legates, D. R. and Willmott, C.J. (1990a) “Mean
Seasonal and Spatial Variability Global Surface
Air Temperature. “Theoretical and Applied
Climatology, 41:11-21.
Legates, D. R. and Willmott, C.J. (1990b) “Mean
Seasonal and Spatial Variability in Gauge-
Corrected, Global Precipitation. “International
Journal of Climatology, 10, 111-127.
Liptay, K., Chanton, J., Czepiel, P., and Mosher,
B. (1998) “Use of Stable Isotopes to Determine
Methane Oxidation in Landfill Cover Soils,”
Journal of Geophysical Research 103: 8243-
8250.
Moldrup, P., Olesen, T., Gamst, J., Schjonning, P.,
Yamaguchi, T. and Rolston, D.E. (2000)
“Predicting the Gas Diffusion Coefficient in
Repacked Soil: Water-induced Linear
Reduction Model.” Soil Sci. Soc. Am. J.
64:1588–1594.
New, M., Hulme, M., and Jones, P.D. (1999).
“Representing Twentieth Century Space-time
Climate Variability. Part 1: Development of a
1961-90 Mean Monthly Terrestrial
Climatology.” Journal of Climate, 12: 829-856.
Moldrup, P., Olesen, T., Yamaguchi, T.,
Schjønning, P., and Rolston, D.E. (1999) .
“Modeling Diffusion and Reaction in Soils: IV.
The Buckingham-Burdine-Campbell Equation
Bonger, Spokas and Chanton 499
for Gas Diffusivity in Undisturbed Soil. Soil
Sci. 164:542–551.
Moldrup, P., Olesen, T., Yoshikawa, S., Komatsu,
T., and Rolston, D.E. (2004) “Three-porosity
Model for Predicting the Gas Diffusion
Coefficient in Undisturbed Soil.” Soil Sci. Soc.
Am. J. 68:750–759.
Monni, S., R. Pipatti, A. Lehtilä, I Savolainen, and
S. Syri (2006) Global Climate Change
Mitigation Scenarios for Solid Waste
Management. VTT Publication 603. VTT
Technical Research Centre of Finland.
Rogner, H-H., Zhou, D., Bradley, R., Crabbé, P.,
Edenhofer, O., Hare, B., Kuijpers, L., and
Yamaguchi, M. (2007) “Introduction.” In
Climate Change 2007: Mitigation,
Contribution of Working Group III to the
Fourth Assessment Report of the
Intergovernmental Panel on Climate Change;
Metz, B.; Davidson, O.R.; Bosch, P.R.; Dave,
R.; Meyer, L.A., Eds.; Cambridge University
Press, Cambridge, United Kingdom and New
York, NY USA.
http://www.ipcc.ch/publications_and_data/ar4
/wg3/en/ch1.html
Scheutz, C, Bogner, J., Chanton, J., Blake, D.,
Morcet, M., and Kjeldsen, P. (2003)
“Comparative Oxidation and Net Emissions of
Methane and Selected Non-methane Organic
Compounds in Landfill Cover Soils, Environ.
Sci. Technol. 37: 5150-5158.
Scheutz, C., Kjeldsen, P., Bogner, J., deVisscher,
A., Gebert, J., Hilger, H., Huber-Humer, M.,
and Spokas, K. (2009), “Microbial Methane
Oxidation Processes and Technologies for
Mitigation of Landfill Gas Emissions.” Waste
Management and Research 27: 409-455.
Spokas, K., Graff, C., Morcet M., and Aran, C.
(2003) “Implications of the Spatial Variability
of Landfill Emission Rates on Geospatial
Analyses.” Waste Management, 23: 599-607.
Spokas, K., Bogner, J., Chanton, J., Morcet, M.,
Aran, C., Graff, C., Moreau-le-Golvan, Y.,
Bureau, N. and Hebe, I. (2006) “Methane Mass
Balance at Three Landfill Sites: What is the
Efficiency of Capture by Gas Collection
Systems?” Waste Management 26:516-525.
500 6th International Congress on Environmental Geotechnics, 2010, New Delhi, India
Spokas, K.A., and Forcella, F. (2006) “Estimating
Hourly Incoming Solar Radiation from Limited
Meteorological Data.” Weed Science 54:182-
189.
Spokas, K.A., and Forcella, F. (2009) “Software
Tools for Weed Seed Germination Modeling.”
Weed Science. 57:216-227.
US EPA [U.S. Environmental Protection Agency]
(2010) Inventory of U.S. Greenhouse Gas
Emissions and Sinks 1990-2008. USEPA #430-
R-10-00.
http://epa.gov/climatechange/emissions/usinve
ntoryreport.html
Willmott, C. J. and Matsuura, K. (1995) “Smart
Interpolation of Annually Averaged Air
Temperature in the United States.” Journal of
Applied Meteorology 34, 2577-2586