Abstract

The understanding of the engineering behaviour of unsaturated soil is totally dependent on the water retention characteristic curve (WRCC). In this paper, a comprehensive study of the WRCCs of pond ash along with the ash's geotechnical behaviour has been made. The WRCC has been drawn experimentally using a Fredlund device based upon the pressure plate technique for both wetting and drying cycles. Further, an investigation was carried out to study WRCC hysteresis of pond ash. There exists a considerable hysteresis in drying and wetting curves of pond ash sample. The different WRCC models were used to fit the experimental WRCC data. The effect of compaction on WRCC was also studied. The air entry value in the case of a loose sample is low and the sample gets nearly desaturated at low soil suction as compared to a dense sample. Also, the wetting WRCC is predicted using the Feng and Fredlund model as it is difficult and time consuming to measure the whole hysteresis. The predicted results are compared with the measured wetting WRCC. Since the direct measurement of unsaturated hydraulic conductivity is difficult to obtain in engineering practices, the unsaturated hydraulic conductivity function is predicted using the measured WRCC as the input parameter using SEEP/W software.

INTRODUCTION

In a thermal power plant, a considerable amount of coal ash is produced as a by-product from the combustion of pulverized coal. The products formed during combustion of coal are categorized as fly ash, bottom ash and vapours. The fly ash being lighter in weight leaves the boiler and is collected in the electrostatic precipitation hoppers either by using a vacuum system or a pressure system. The bottom ash is heavier and is generally disposed of as a wet disposal in which ash is mixed with water in an approximately 1:10 ratio. The ash slurry so formed is transported through ash dykes to an ash pond for safe disposal. This disposed ash in an ash pond is termed as pond ash. In view of large scale generation of pond ash, it is prudent to explore as many avenues as possible for ways of utilizing the ash. Pond ash has found various applications in the field of geotechnical and geo-environmental engineering. Use of coal ash conserves the scarce natural resources on one hand and saves the precious land which is otherwise occupied by the ash pond and many environmental issues related to the deposited ash. To encourage the usefulness of pond ash, the geotechnical and unsaturated behaviour of pond ash is required to be ascertained. Currently, pond ash has been successfully used as a filling material in embankments of roads and railways, in low lying areas and in dams, and in the majority of these practices the unsaturated behaviour of pond ash prevails.

There are various materials present in the geotechnical field whose characteristics are not in line with the theory of saturated soil mechanics. The existence of different fluid phases causes the occurrence of unsaturated behaviour which is inconsistent with the concepts and principles of classical engineering practice. The direct practice of measuring unsaturated soil in the field involves measurement of soil suction in unsaturated phase, which is relatively time consuming and complex to perform as compared to the test for saturated soil. Other methods for the study of soil in unsaturated phase have been suggested wherein the water retention characteristic curve (WRCC) and permeability coefficient are used (Fredlund & Rahardjo 1993). Soil material near the ground surface where the level of the water table is beneath the ground surface has negative pore-water pressures, resulting in possible reduction in degree of saturation. Study of WRCC can assist in the understanding of distribution of pore water pressure, which is important in describing the possible mechanism for the behaviour of soil in unsaturated phase. WRCCs have various applications in the geotechnical field related to natural slopes that are subjected to environmental change, in contaminant retention ponds, in slope stability of deep vertical or near-vertical excavations, bearing strength for shallow foundations, and road and railway structures.

There are various methods already described for measurement of WRCC in the laboratory (Fredlund & Rahardjo 1993). Agus & Schanz (2005) compared the WRCC obtained through various methods and all these are based on the principles of relative humidity. The instruments used for measuring the WRCC have limitations in that the WRCC can be measured up to a specific range of matric suction. In the present study, the Fredlund SWC-150 device was used, which is a pressure plate apparatus that can measure WRCC up to matric suction of 1,500 kPa. However, the soil with low permeability becomes dry at considerably high matric suction and the value of matric suction to be applied to make the sample completely dry is 106 kPa. Therefore it becomes necessary to predict WRCC precisely for high range of matric suction until the sample gets nearly desaturated. There are a variety of models and estimation techniques developed from research studies for the measurement of WRCC (Sreedeep & Singh 2011). It was found that estimation techniques can be used for seepage modelling of soil in unsaturated phase wherein the water storage function is estimated from WRCC and the permeability function is calculated from saturated hydraulic conductivity. The results obtained for different models and estimation techniques generally vary as they have different sets of assumptions and input parameter. Therefore a comparative study is required for the prediction of WRCC. Moreover, the WRCC of a soil is generally hysteretic; i.e. behaviour of soil in all climate types is not the same. It is necessary to understand the behaviour of material in all the phases, i.e. drying and wetting phases, and the measurement of both drying and wetting WRCCs is extremely time consuming and therefore quite costly (Topp & Miller 1966; Dane & Wierenga 1975; Zeng et al. 2012). Therefore, it is essential to develop a means of predicting the wetting phase of WRCC from the drying curve, and vice versa (Pham et al. 2003). However, the soil will show different dry densities or water contents in the field; the influence of these geotechnical properties could play an important role in affecting the soil behaviour in unsaturated phase (Likos et al. 2014). The effect of dry density and initial water content on the WRCC has been studied by researchers on different types of soil. The initial water content has a significant influence on the movement of water through the material in unsaturated phase. The stress, structure and aggregation also significantly influence the WRCC of soil whereas the effect of initial void ratio is minimal (Vanapalli et al. 1999).

Recently, the pond ash had been used in bulk as a subgrade of pavements, embankments, filling of low lying areas, ash dykes, and landfill covers. In all these applications, the pond ash is compacted at maximum dry unit weight (MDD) and then remains in the unsaturated condition for most of the service life period. Therefore, to understand the effect of various parameters affecting WRCC and consequently unsaturated hydraulic conductivity of pond ash, a comprehensive study is carried out to understand the behaviour of soil in unsaturated phase.

EXPERIMENTAL INVESTIGATION

Material

The pond ash used was collected from an ash pond situated on the bank of River Satluj in Ropar (India). A representative ash pond sample was collected from an outflow point. at a distance of approximately 450 m from the ash dyke.

Geotechnical characterization

The specific gravity of the sample is determined as per the procedures mentioned in IS: 2720-Part 3 (1980) (equivalent to ASTM D854-14 (2014)). The specific gravity test is based upon the determination of the dry weight of a pond ash sample and the weight of the same sample with water added in a container of known volume. The specific gravity of pond ash comes out to be 2.13. The particle size distribution of pond ash is found as per the method outlined in IS: 2720-Part 4 (1985) (equivalent to ASTM D422 (2002)). The grain size distribution of the portion of ash particles which were retained on a 0.075 mm sieve was determined through sieve analysis. The remaining material which is finer than 0.075 mm was collected and a hydrometer test was conducted. The ash sample is in the range of silt size containing 90% silt sized particles. The pond ash sample used in the study is non-plastic and is classified as inorganic silt with low plasticity as per IS: 1498 (2002) (equivalent to ASTM D2487-11 (2011)). To measure the MDD and optimum moisture content (OMC) of pond ash, the standard Proctor test was conducted as described in IS: 2720-Part 7 (1980) (equivalent to ASTM D698 (2012)). The MDD and OMC of the pond ash sample is 10.67 kN/m3 and 35.5% respectively.

The permeability coefficient of pond ash was determined as per the method prescribed in IS: 2720-Part 17 (1986) (equivalent to ASTM D2434-68 (2006)). The sample of ash was prepared at dense and loose state, i.e. at 95% MDD and 70% MDD respectively, to determine the coefficient of permeability of samples (Jakka et al. 2010). The values of coefficient of permeability for ash sample are shown in Table 1. It was found that the permeability coefficient lies in the range of sand and silt. The value of permeability of pond ash is greater than 10−9 m/s and hence could not be used as barrier material independently. However, with addition of bentonite the value of permeability could be reduced in order to be used as barrier material (Sobti & Singh 2017). The loss on ignition (LOI) test was performed as per IS: 1917-Part 1 (1991). In this study, the pond ash was first oven dried and the temperature of the muffle furnace was increased incrementally until the furnace temperature reached 1,000 °C. The LOI value of the pond ash is 6.25%. This value indicates the presence of low unburned carbon which shows that the spontaneous heating of pond ash is minimal. Therefore, the pond ash, which is waste material, could be used in bulk as a filling material in construction of embankments without any concern of overheating.

Table 1

Basic properties of pond ash

Test Pond ash 
Specific gravity 2.13 
Sand (%) 7.40 
Silt (%) 90.40 
Clay (%) 2.20 
D10 (mm) 0.016 
D60 (mm) 0.030 
Coefficient of uniformity 1.87 
Coefficient of curvature 0.92 
Maximum dry unit weight (kN/m310.67 
Optimum moisture content (%) 35.50 
Coefficient of permeability (m/s) Dense 9.1 × 10−7 
 Loose 2.5 × 10−6 
LOI value (%) 6.25 
Test Pond ash 
Specific gravity 2.13 
Sand (%) 7.40 
Silt (%) 90.40 
Clay (%) 2.20 
D10 (mm) 0.016 
D60 (mm) 0.030 
Coefficient of uniformity 1.87 
Coefficient of curvature 0.92 
Maximum dry unit weight (kN/m310.67 
Optimum moisture content (%) 35.50 
Coefficient of permeability (m/s) Dense 9.1 × 10−7 
 Loose 2.5 × 10−6 
LOI value (%) 6.25 

Water retention characteristic curve

The study of geotechnical characteristics showed that the index properties and permeability coefficient of pond ash lies between sand and silt. Thus, the pond ash could be used in various geotechnical applications as filling material and in most of these applications the material is compacted to OMC where the unsaturated behaviour of material prevails. In unsaturated soil, the behaviour of soil is dependent upon the negative pore pressure. Thus it is necessary to understand the unsaturated behaviour of pond ash, for which the WRCC needs to be studied. In the present study, Fredlund SWCC (soil water characteristic curve) device, which is a pressure plate device, was used to determine the WRCC experimentally. Figure 1 shows the actual instrument used in this study. The preparation of sample was done by compacting the sample at the desired density into a mould. The mould was then partially submerged in a container filled with water to achieve possible saturation in such a way that the height of water remains 2 mm below the top of the mould so that the sample gets saturated through capillary action. After the sample became saturated, it was placed on a ceramic disk of 1,500 kPa; i.e air entry value (AEV) of the ceramic disk is 1,500 kPa. The ceramic disk acts like a semi-permeable medium as it permits the passage of water through it but does not allow the air to pass, up to the AEV of the ceramic disk. The desired soil suction was applied to the sample. The soil suction was increased incrementally from 0 to 10, 20, 50, 100, 200 and 400 kPa. At each value of soil suction applied, the water coming out from the sample was calculated using volumetric tubes attached to the device. After the equilibrium was attained, the pressure was increased and we again noted the reading. During this process it was necessary to flush out the air entrapped in the water using a flushing device. This eliminated the error due to presence of air, and the accurate reading of water level corresponding to a particular pressure was noted down. After each increment of soil suction, the water content of pond ash sample decreased and this process continued until the sample attained the value of residual water content (Singh & Singh 2018).

Figure 1

Fredlund SWC-150 device.

Figure 1

Fredlund SWC-150 device.

RESULTS AND DISCUSSIONS

Drying water retention characteristic curve

The drying stage represents the removal of water from the sample while increasing the matric suction, which was determined using a Fredlund SWCC device. The experimental data of matric suction (Ψ) and volumetric water content (θw) for pond ash in the case of drying is shown in Figure 2. It has been observed from the test data that the sample gets nearly desaturated at application of pressure of 400 kPa. The AEV and residual suction (Ψr) were determined by graphical method (Singh & Singh 2018) and the AEV was found to be 30 kPa.

Figure 2

Drying water retention characteristic curve.

Figure 2

Drying water retention characteristic curve.

Modelling of experimental data

The experimental data obtained was best fit using the equation proposed by Fredlund and Xing (FX) (Fredlund & Xing 1994), Van Ganuchten (VG) (Van Ganuchten 1980) and estimation method. The FX method is a closed-form type solution which develops the volumetric water content function based on the soil fit parameters (a, n, m). The governing equation of the FX is as follows: 
formula
(1)
where
  • θ = the volumetric water content

  • θs = the saturated volumetric water content

  • a,n,m = soil fit parameters, where a is in kPa

  • Ψ = soil suction (kPa).

The fitting parameters a, n and m were used to predict the WRCC of samples using Geostudio software SEEP/W version 2012. Table 2 shows the value of fitting parameters. The results obtained experimentally from the Fredlund SWC-150 device are up to certain limit. As the matric suction reaches beyond that limit, the values of volumetric water content can be numerically predicted through different models using SEEP/W. It is observed from Figure 3 that the FX equation fitted well with the measured data for pond ash sample.

Table 2

Parameters for drying WRCC for dense sample

WRCC model Description Pond ash 
Fredlund and Xing Saturated volumetric water content 0.478 
Residual volumetric water content 0.136 
Residual matric suction 120 
Air entry value (kPa) 30 
Volumetric water content at inflection point 0.368 
a (kPa) 50 
n 1.632 
m 0.9616 
s 2.374 × 10−3 
WRCC model Description Pond ash 
Fredlund and Xing Saturated volumetric water content 0.478 
Residual volumetric water content 0.136 
Residual matric suction 120 
Air entry value (kPa) 30 
Volumetric water content at inflection point 0.368 
a (kPa) 50 
n 1.632 
m 0.9616 
s 2.374 × 10−3 
Figure 3

Measured and predicted drying WRCC using different methods (GSD: grain size distribution).

Figure 3

Measured and predicted drying WRCC using different methods (GSD: grain size distribution).

Van Genuchten proposed a closed-form type solution, whose governing equation is as follows: 
formula
(2)
where
  • θ = the volumetric water content,

  • θs = the saturated volumetric water content,

  • θr = the saturated volumetric water content,

  • Ψ = soil suction (kPa)

  • a, n, m = curve fitting parameters where a is in kPa

The soil fit parameters a and n were used to best fit the VG equation. It is observed from Figure 3 that the VG equation fitted well with the measured data for soil suction up to 100 kPa. After 100 kPa the WRCC derived from the VG function shifts upward as compared to the measured WRCC. The WRCC can also be estimated using SEEP/W. From the geotechnical characterization it is observed that the pond ash sample lies in the range of silt. Therefore, a silt sample function was selected to predict the WRCC. In the estimation method the WRCC of the pond ash sample was also predicted using grain size parameters (D10, D60), consistency limits and saturated water content in SEEP/W. The measured and predicted WRCCs are shown in Figure 3.

It is observed from Figure 3 that the FX method closely relates with the experimental result. The WRCC predicted from VG method closely fits up to 100 kPa; after that there is variation in VG results as compared to experimental results. The estimation methods using grain size and sample function shows wide variation in WRCC as compared to the experimental curve. It is significant to understand the variation between experimental and analytical models. There are certain limits of experimental equipments up to which the soil suction can be induced in the sample. The soil of low permeability requires high matric suction to reach residual water content. Hence, it is required to understand the relation of experimental and analytical models. Therefore, the calculated a, n and m parameters can be used to produce the WRCC up to 106 by fitting these parameters in FX, VG and estimation methods. In the present study it is concluded that the FX method closely fits with the experimental results; therefore this method is best fit to predict the WRCC of pond ash material.

Hysteresis of water retention characteristics curve

Hysteresis in the WRCC reveals the behaviour of soil in unsaturated phase in all climate types. It was found by researchers that the total hysteresis can be quantified by calculating the total area between the drying and wetting WRCCs (Yang et al. 2004). A significant hysteresis is observed between drying and wetting of pond ash as shown in Figure 4. Thus it is necessary to study the WRCC of pond ash for all climate types. The occurrence of this hysteresis may be because, during wetting, larger pores control the movement of water, whereas in draining processes the smaller pores control the flow.

Figure 4

Hysteresis of WRCC of pond ash.

Figure 4

Hysteresis of WRCC of pond ash.

Influence of compaction on WRCC

To study the effect of compaction on WRCC, the pond ash samples were prepared at 95% MDD and 70% MDD using different compaction efforts. It is observed from Figure 5 that initial water content of the loose sample is more than that of the dense sample due to presence of more voids. As the suction in the sample increases, the dissipation of water in the loose sample is rapid as the water is not tightly enclosed in the pores in the case of the loose sample. The soil parameters obtained through WRCC are shown in Table 3. The AEV of the loose sample and dense sample is 14 kPa and 30 kPa respectively. As pointed out by studies, the water content initially present would affect the pore-size distribution. As the matric suction reaches 400 kPa both samples get nearly desaturated. The residual water content in the case of the loose sample is 0.1, which is less than that of the dense sample, i.e 0.14.

Table 3

Values of soil fit parameters for drying WRCC of dense and loose sample of pond ash

Pond ash Dry density (kN/m3Saturated volumetric water content Air entry value (kPa) Residual volumetric water content a (kPa) n m s 
Dense 10.13 0.48 30 0.14 50 1.63 0.96 2.37 × 10−3 
Loose 7.47 0.51 14 0.10 23 1.25 0.80 3.66 × 10−3 
Pond ash Dry density (kN/m3Saturated volumetric water content Air entry value (kPa) Residual volumetric water content a (kPa) n m s 
Dense 10.13 0.48 30 0.14 50 1.63 0.96 2.37 × 10−3 
Loose 7.47 0.51 14 0.10 23 1.25 0.80 3.66 × 10−3 
Figure 5

Drying and Wetting WRCC for pond ash sample at dense and loose state.

Figure 5

Drying and Wetting WRCC for pond ash sample at dense and loose state.

It is also observed from Figure 5 that the area under hysteresis in the case of the loose sample is less as compared to the dense sample. This means the water content of sample remains the same at a particular suction for all climate types. Thus it can be concluded that there is significant effect of compaction effort on the WRCC of pond ash.

Prediction of main wetting curve

The reliable measurement of WRCC is essential for describing unsaturated flow. Although various laboratory equipment is now available to measure the WRCC, they are all time consuming for measurement of WRCC. Moreover, the WRCC of pond ash is hysteretic and measurement of the entire wetting curve is difficult and also it consumes more time to complete. In the present study the laboratory measurement of one drying and wetting WRCC using the Fredlund SWC device takes 15 days to complete. The greater the number of samples, the more complex the measurement process. Therefore, it is necessary to predict the hysteresis curve by measuring only one WRCC. The wetting WRCC is predicted using the equation presented by Feng and Fredlund as: 
formula
(3)
where
  • wu = water content at zero suction

  • c = water content at high suction

  • Ψ = soil suction (kPa)

  • b, d = soil fit parameters where b is in kPa.

The results revealed that by using only two points the entire boundary wetting curve could be predicted. The two points on the boundary wetting curve were measured and used to calculate the curve fitting parameters dw and bw. The calculated values of dw and bw were used in Equation (3) to predict the water content at different soil suction.

The two curve fitting parameters, i.e. dw and bw, to be used in prediction of the main wetting curve were calculated as follows: 
formula
(4)
 
formula
(5)
where
  • w1w = measured water content on wetting curve corresponding to first point Ψ1

  • w2w = measured water content on wetting curve corresponding to second point Ψ2

  • Ψ1,Ψ2 = soil suction representing the first and second point to be taken for prediction of wetting curve (kPa)

  • Ψ1w = soil suction representing the first predicted point on wetting curve (kPa)

  • wu = water content at zero suction

  • c = water content at high suction.

The comparison of predicted result and measured result for dense and loose samples is shown in Figure 6. It is observed from the figure that at low suction the predicted wetting curve is closely related to the measured wetting curve. At higher suction, the predicted wetting curve is slightly overestimated from the measured values in some cases. The performance of predicted values is quantitatively evaluated using the R2 statistical indicator, i.e. coefficient of determination. The R2 obtained for pond ash in dense and loose state is 0.98 and 0.91 respectively as shown in Figure 7.

Figure 6

Predicted and measured wetting WRCC for pond ash sample at dense and loose state.

Figure 6

Predicted and measured wetting WRCC for pond ash sample at dense and loose state.

Figure 7

R2 for predicted values for pond ash sample at dense and loose state.

Figure 7

R2 for predicted values for pond ash sample at dense and loose state.

Unsaturated hydraulic conductivity

Unsaturated hydraulic conductivity is an important soil property which affects the rate at which water percolates through the soil. The experimental measurement of hydraulic conductivity in unsaturated phase is a complex phenomenon as it is time consuming and expensive. Therefore, the alternative is to predict the unsaturated hydraulic conductivity using estimation methods. There are various estimation techniques which are acceptable to geotechnical engineering that can be used for determination of the unsaturated hydraulic conductivity function (Thieu et al. 2000). One of the approaches for the implementation of theories of unsaturated soils in practice is the determination of the WRCC, used along with saturated hydraulic conductivity. This approach is an indirect way for a determination of unsaturated hydraulic conductivity.

In this study, the unsaturated hydraulic conductivity was determined using SEEP/W 2012 software by inserting the value of the saturated coefficient of permeability and WRCC obtained using the FX model (Fredlund & Xing 1994) and VG model (Van Ganuchten 1980). The WRCC obtained for both the models differs at high suction range and hence there exists a variation in predicted unsaturated hydraulic conductivity for the FX and VG models. Figure 8 shows the variation of unsaturated hydraulic conductivity with matric suction for pond ash. It is observed from Figure 8 that the sample is nearly saturated near to zero suction. The saturated hydraulic conductivity of the sample at this stage is 9.1 × 10−7 m/sec. As matric suction in the sample is increased, the air starts entering the pores and the hydraulic conductivity goes on decreasing. It is significant to note from the figure that as the matric suction reaches the AEV, i.e 30 kPa, air starts entering the pore of samples progressively, resulting in the decrease in hydraulic conductivity. Further, as the sample gets nearly desaturated the pores get totally filled with air which resists the movement of water through the pores and hydraulic conductivity becomes negligible. Also, it is found that there is significant variation in hydraulic conductivity predicted from the FX and VG models for the matric suction greater than 104kPa.

Figure 8

Variation of unsaturated hydraulic conductivity with matric suction.

Figure 8

Variation of unsaturated hydraulic conductivity with matric suction.

CONCLUSIONS

In this study the detailed investigation of geotechnical characterics and unsaturated behaviour of pond ash was carried out. The following conclusions can be drawn based on the study.

  • The geotechnical characterization of pond ash sample reveals that particles of pond ash are mostly in the silt range. The LOI value of the ash sample is low and thus bulk use of pond ash as filling material can be done with no risk of self-heating.

  • WRCC is an important parameter to study the behaviour of soil in unsaturated condition. Most of the experimental methods for determining WRCC are limited to a suction pressure up to 1,500 kPa. Beyond that, the WRCC has to be extrapolated using different models. A comparative study of prediction of WRCC through various models and the estimation technique was conducted. The measured WRCC matches satisfactorily with predicted WRCC using the FX equation as compared to VG equation and estimation methods.

  • There is considerable hysteresis in the drying and wetting curves of pond ash sample, which indicates pronounced difference in behaviour of pond ash in wetting and drying cycles during the service life period. A pond ash embankment experiences change in water content as the climatic condition changes, which signifies that during wetting there is less capillary rise of water in the pores of pond ash; thus water content at high suction is low in wetting as compared to drying.

  • In the field, the density achieved in the compaction where coal ash is being used in embankments, subgrade etc. is 95–97% of MDD. The density of the coal ash deposited hydraulically in a pond ash bed is in loose condition and its density generally varies from from 60 to 70% of MDD. Thus, it is significant to understand the effect of compaction on the WRCC. A change in compaction shows a considerable variation in behaviour of WRCC. The AEV of the loose sample is less than that of the dense sample. The area under hysteresis of WRCC is also less in the case of loose sample.

  • The experimental measurement of the wetting curve of pond ash is time consuming and takes 8 to 10 days to complete. It is important to predict the wetting curve and thus prediction of the wetting curve was done using Feng and Fredlund model. The predicted results of the wetting WRCC was compared with the measured wetting WRCC to investigate the reliability of the model in precise prediction of the wetting WRCC. The predicted wetting WRCC shows a similar behaviour as that of the measured WRCC.

  • The measurement of unsaturated hydraulic conductivity using experimentation is complex. The alternative is to predict the unsaturated hydraulic conductivity using estimation methods. It was concluded from the estimation method that there is significant variation in hydraulic conductivity predicted from the FX and VG models using SEEP/W for a higher matric suction, i.e. greater than 104kPa.

REFERENCES

REFERENCES
Agus
S.
&
Schanz
T.
2005
Comparison of four methods for measuring total suction
.
Vadose Zone Journal
4
(
4
),
1087
1095
.
ASTM D2434-68
2006
Standard Test Method for Permeability of Granular Soils (Constant Head)
.
ASTM International
,
West Conshohocken, PA
,
USA
.
ASTM D2487-11
2011
Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System)
.
ASTM International
,
West Conshohocken, PA
,
USA
.
ASTM D422
2002
Standard Test Method for Particle Size Analysis of Soils
.
ASTM International
,
West Conshohocken, PA
,
USA
.
ASTM D698
2012
Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Standard Effort
.
ASTM International
,
West Conshohocken, PA
,
USA
.
ASTM D854-14
2014
Standard Test Methods for Specific Gravity of Soil Solids by Water Pycnometer
.
ASTM International
,
West Conshohocken, PA
,
USA
.
Fredlund
D. G.
&
Rahardjo
H.
1993
Soil Mechanics for Unsaturated Soils
.
John Wiley and Sons, Inc.
,
New York, USA
.
Fredlund
D. G.
&
Xing
A.
1994
Equations for the soil-water characteristic curve
.
Canadian Geotechnical Journal
31
(
4
),
521
532
.
GEO-SLOPE International Ltd
2012
Seepage Modeling with SEEP/W 2012 Version, An Engineering Methodology
.
IS: 1498
2002
Classification and Identification of Soils for General Engineering Purposes
.
Bureau of Indian Standards
,
New Delhi, India
.
IS: 1917-Part 1
1991
Chemical Analysis of Quartzite and High Silica Sand – Determination of Loss on Ignition
.
Bureau of Indian Standards
,
New Delhi, India
.
IS: 2720-Part 3
1980
Methods of Test for Soils – Determination of Specific Gravity
.
Bureau of Indian Standards
,
New Delhi, India
.
IS: 2720-Part 4
1985
Methods of Test for Soils – Grain Size Analysis
.
Bureau of Indian Standards
,
New Delhi, India
.
IS: 2720-Part 7
1980
Methods of Test for Soils – Determination of Water Content – Dry Density Relation Using Light Compaction
.
Bureau of Indian Standards
,
New Delhi, India
.
IS: 2720-Part 17
1986
Laboratory Determination of Permeability
.
Bureau of Indian Standards
,
New Delhi, India
.
Jakka
R. S.
,
Ramana
G. V.
&
Datta
M.
2010
Shear strength characteristics of loose and compacted pond ash
.
Geotechnical and Geological Engineering
28
,
763
778
.
Likos
W. J.
,
Lu
N.
&
Godt
J. W.
2014
Hysteresis and uncertainty in soil water-retention curve parameters
.
Journal of Geotechnical and Geoenvironmental Engineering
140
(
4
),
040130501
11
.
Singh
J.
&
Singh
S. K.
2018
Geotechnical characterization and WRCC for spatially varied pond ash within an ash pond
.
Indian Geotechnical Journal
1
11
.
Sreedeep
S.
&
Singh
D. N.
2011
A critical review of the methodologies employed for soil suction measurement
.
International Journal of Geomechancis
11
(
2
),
99
104
.
Thieu
N. T. M.
,
Fredlund
D. G.
&
Vu
H. Q.
2000
General partial differential equation solvers for saturated-unsaturated seepage
. In:
Proceedings of Asian Conference on Unsaturated Soils
,
UNSAT ASIA
,
Singapore
, pp.
201
206
.
Topp
G. C.
&
Miller
E. E.
1966
Hysteretic moisture characteristics and hydraulic conductivities for glass-bead media
.
Soil Science Society of America Journal
30
(
2
),
156
162
.
Van Ganuchten
M. T.
1980
A closed-form equation for predicting the hydraulic conductivity of unsaturated soils
.
Soil Science Society of America Journal
44
,
892
898
.
Yang
H.
,
Rahardjo
H.
,
Leong
E. C.
&
Fredlund
D. G.
2004
Factors affecting drying and wetting soil water characterstic curves of sandy soils
.
Canadian Geotechnical Journal
41
(
5
),
908
920
.
Zeng
Z. T.
,
Lu
H. B.
&
Zhao
Y. L.
2012
Wetting-drying effect of expansive soils and its influence on slope stability
.
Applied Mechanics and Materials
170
,
889
893
.