Behaviour of Expansive Soil

There are various methods of determining the behaviour of the expansive soil. They are based on the results of the test that are obtained under fully lateral restraint conditions of the odometer. It is a challenging endeavour to study the swelling behaviour of the expansive soils. This is different from the non-expansive and the low-expansive soils. The reason for this difficulty is that the process of the flow of the transient is accompanied by the change of the volume of the soils. Due to this, all the soil changes as the change of the soil suction, the void ration and hydraulic conductivity, are all influenced. The fundamental properties of clays are therefore the low permeability and the very high entry value of the soil suction to reach the unsaturated state. The high expansive soils have a high tendency to retain moisture content during the process of desiccation. It is for this reason that its permeability function becomes extremely small with the decrease of the void ratio and the saturation zone. This happens before reaching the entry value of the air. Thus in order to realize the saturated zone of the highly expansive soils, we need a much greater value of suction.

Considering the above common observations into the focal point of the experimental work of this research, the research will lead to:

Ø  The significance of selecting the appropriate materials magnitudes in order to create the extremely expansive soil that moreover simulates the field condition or can be used in the design of the geotechnical engineering.

Ø  The substance of the selected materials proportions in providing a sensible testing duration together with demonstrating a noteworthy volumetric change.

Ø  The importance of executing suitable testing performance to conquer adequate results and analyses.

The chapter dominantly dwells on the selection of the materials, determination of the geotechnical properties of engineering and the description of the preparation methods. Beyond here, there is an address of the apparatus that are used in the process. The results of the research will in any way describe the influence of the mineralogy and physic-chemical or the swelling behaviour of the selected soil.

Research methods

Most of the procedures that have been drawn in this chapter are based on carrying out experiments in order to determine the results to the problems that have been posted. The main research method is the carrying out of experiments.

4.2 Materials Selection and Soil Property Tests

Most of the samples used for experimental research as revealed in chapter 3 were artificially prepared. It is more preferable to testing the mechanical behaviour than natural expansive soils for the reason of eliminating the possible variability between samples such as heterogeneity, and inconsistency of chemical or mineral composition (Sharma, 1998).

In that consideration, two materials have been selected in order to create artificial highly-expansive soil namely; Bonny silt and Black hills bentonite. The Bonny silt is a material obtained from a sedimentary formation located at Bonny Dam, in eastern Colorado State (Bicalho, 1999). It has the classification properties of LL = 25%, PL = 21%, Casagrande classification CL-ML and Specific Gravity (Gs) = 2.63. The grain size distribution of Bonny silt is shown in Figure 4.1. As the Bonny silt has particles of sand size, Number 40 sieve (0.425 mm opening) was used to retain those particles and the remaining of bonny silt material is mixed with Black Hills Bentonite.

The black hills bentonite is a material acquired from commercial vendors (Mile hi Ceramics Inc.). This meticulous sodium bentonite has principal smectite mineral, which is montmorillonite. The sodium bentonite is acknowledged as clay of high swelling, viscous, and thixotropic as a unique chattel. Those properties give bentonite outstanding uses for various geotechnical engineering applications that are explained in detail by Murray (2007).

The assortment of soil sizes (all samples are prepared from slurry) is based on introductory shrinkage tests as elucidated later in this chapter. It is established that the slowest shrinkage test is observed when using a mixture of 20% bentonite to 80 % Bonny silt ratio, which is not applicable since the shrinkage tests are solely performed as a baseline for establishing swelling tests of various initial moisture contents. Conversely, a mixture of 60 % bentonite to 40 % Bonny silt ratio shows the quickest desiccation process but causes cracks that are widening with test duration.  Therefore, the 40 % bentonite to 60 % Bonny silt is selected in this research in order to sustain between the observable degree of evasiveness and the suitability of swelling test duration.

Two tests have been performed in this study to determine the selected soil properties:

Ø  A specific gravity (Gs) test (e.g. see ASTM D854 - 06 e1), in which 2.53 is determined as the Gs of the selected soil.

Ø  A liquid limit test (LL) (e.g. see ASTM D4318 - 05), which is one of Atterberg limits usually performed to determine a moisture content at the interchange state of clay soil from liquid to plastic. The LL of the selected soil is 126.4 % which gives an indication of considerably large volume changes as moisture content of the soil changes.

A laboratory mixer is used to thoroughly mix two quantities of the soil by which 300 % of water content per each quantity is added to form slurry.  The total mass of the sample is approximately 18.7 Kg. This amount is sufficient to produce a big sample of approximately 10 mm in thickness and 610 mm in diameter so that, multiple specimens can be obtained simultaneously for various testing as explained in the next sections.

4.3 Double Drainage Consolidation Test

The one-dimensional consolidation test is one of the most essential geotechnical tests; it is usually performed by using saturated fine-grain soil. Therefore, the sample produced in this test can be considered as remoulded sample that has uniformly pore size distribution (e. g., see Fityus and Buzzi, 2009). The role of this test is to produce big samples that have as small thickness as possible to maintain the subsequent tests' durations convenient for the experimental program period in this research. However, since the time-rate of consolidation is governed by sample height, drainage path and hydraulic conductivity of the soil, it is decided to perform double drainage consolidation test.

4.3.1 Experimental Apparatus

The apparatus , which are utilized for sample preparation consist of an aluminium container that has inner diameter of approximately 610 mm and inner height of 537 mm, a top aluminium platen that has four holes for allowing the excess water in sample pores to expel from the top surface, two porous stone disks, two fine filter papers, a top aluminium ring, a bottom aluminium plate, a rubber O-ring to seal between the container and the bottom plate, and connection rods that connect the top ring with the bottom plate.  A schematic assemblage of the container is shown in Figure 4.2.

The Material Testing System (MTS) is the machine used for consolidation test. Its model is 661.23A-02 with the load cell capacity of 500 KN/ 110 Kips. It can be functioned by two modes namely; displacement-control mode and load-control mode.

4.3.2 Experiment Procedure

The double drainage consolidation test has been performed in two stages:

1. After preparing the slurry as mention earlier, it is poured in the assembled container and left uncovered for one day to allow for self consolidating and solidifying the upper portion. The previous strategy is found effective to support the weight of the filter paper and the top porous stone when are placed. Moreover, splashing the slurry from the sides of the top porous stone is prevented.  The time-rate of consolidation under the weight of the porous stone is one day. Afterward, the top platen is placed to provide an additional loading step that has duration of another one day. Then, weights of .... Kg is distributed on top of the top platen to further consolidate the slurry for one day. Next, the weights are quickly replaced by approximately of ... Kg of wooden pieces to cover the entire space of the container as shown in Figure 4.3.

2. In the second stage, The MTS machine is used to continue the consolidation process under displacement-control mode. An additional steel plate is placed on the middle of the wooden pieces to keep the loading cell always in the center position during the test. Therefore, the load distribution on the slurry is more accurately measured by the fixed load cell under a constant displacement rate of 1.2x10-5.0 m/min that is controlled by the base of the MTS machine. The gradual increase of the load is recorded onto the personal computer. Figure 4.4 shows the installation of the container in the MTS machine. The consolidation process last for about 9 to 10 days to reach the designated thickness of the sample and the pre-consolidation pressure is determined to be approximately......

As soon as the consolidation test is stopped, the strain control mode is switched to load-control mode to hold the pre-consolidation pressure as it is. Then, the excess water on top of the sample is extracted completely to assure that it will not be absorbed back into the sample pores when the sample is unloaded in the MTS machine. The container is then, removed and all the weights on the sample are extracted. The container is disassembled and the sample is cut into several pies and preserved in sealed plastic bags. It is found difficult to prevent the samples from losing their moisture in the plastic bags for a long time, therefore,  they are preserved in a plastic container and stored in the humidity room in which the relative humidity is fluctuating between ... to ..... Moreover, additional wetted paper towels are set at the bottom of the closed plastic container to provide a relative humidity of approximately 97.0 % as illustrated in Figure 4.5.

4.4 Shrinkage Test

The primary objective of this test is to produce specimens at various initial moisture contents and use them as a baseline for swelling tests. However, the constitutive relationship between the void ratio and moist content during the desiccation process is monitored as well.

Shrinkage is a response of soil particles movement caused by initial moisture content reduction or by introducing pore air into a soil mass over time. The soil particles movements are conceptually due to capillary tension development in water-air interface, which has a tendency to pull adjacent soil particles together until the shrinkage limit is reached. At shrinkage limit, the process of moisture content reduction continues without any further significant reduction in overall soil volume. In the field, this behaviour is caused by elevated temperatures (evaporation) or by lowering the groundwater table level.

In the case of unsaturated clay soil, both pore sizes and particle sizes are smaller such that the capillary tensions are higher and the relative particle movements are easier than in other types of soils. This phenomenon is similar to a reduction in overall saturated soil volume when subjected to an increase of effective overburden pressure, except that in cases of unsaturated expansive soils, the reduction in volume results from additional internal pressures caused by both, capillarity and osmotic effects.

On the basis of the theoretical relationship between RH and soil suction as shown in Figure 4.6, it is obvious that at the region of very high RH and very low soil suction, the slightly change in RH greatly influences the soil suction variation. For example, at RH =  . . . . %, the corresponding soil suction is. While the RH of ... % is equivalent to soil suction of ......  Therefore, the RH may have a significant influence to soil suction variation if it is considered as one of the state variables (effective stress, water content, temperature, external pressure, hydraulic gradient, and others) that is governing the soil system (pore size distribution, hydraulic conductivity, and other soil properties).

The quickest shrinkage process is occurred at room temperature where the RH is considerably low as compared to the humidity room. Since the samples have very high liquid limit value, the desiccation-induced cracking due to a rapid moisture content reduction if the samples are allowed to dry at room temperature, is highly possible. In order to avoid this problem and have a good control over the shrinkage process, all the specimens required for this test are placed in small plastic containers and the shrinkage process takes place in different environmental conditions as explained in the following sections.

4.4.1 Experimental Apparatus

The apparatus of this test are; an aluminium plumber that has a sharp edge is used as a cutting tool whose inner diameter is ..., small plastic containers with lids, a scale, a digital camera, and two closed desiccator containers where wetted paper towels are placed to increase the RH instead of the desiccating agent. In addition,  The humidity room is used for controlling the desiccation process especially in cases where specific moisture content of desiccated specimens are required as initial values used for swelling tests.

4.4.2 Experiment Procedure 

Firstly, the cutting tool is used to extract specimens from the samples prepared as discussed earlier. The specimens are then placed in the small plastic container and quickly closed to preserve the initial moisture contents.

Secondly, the digital camera is installed in such a way that the distance between the stationary digital camera and the specimen is fixed when the images of the specimens are taken periodically during the test as illustrated in Figure 4.7. In this manner, the error tolerance of total volume measurements will be minimized. The pixel ruler is utilized to measure the specimen dimensions in pixels and converted to metric units after collecting the images onto a personal computer.

Consequently, the initial values of total volume, void ratio, and moisture content of each specimen are determined. It should be noted that some issues are found to be difficult to handle such as the discrepancy of obtaining the initial moisture contents and the initial void ratios.  The rebounding nature of the soil due to unloading process after the end of the consolidation test as discussed in section 4.3 caused the expansive sample to relax and absorb some of the remaining water from the bottom filter paper and bottom porous stone. In addition, the difficulties of acquiring identical specimens, cause the inconsistency of obtaining the initial values of moisture content and void ratio.

By accepting the above matters, the shrinkage tests are then, initiated according to the following scheme:

Ø  Shrinkage process #1: Some specimens are allowed to desiccate at room temperature whose the RH is about 30.0 %. This approach provides the shrinkage limit by which the corresponding equilibrium moisture content of the specimens has the lowest value in the testing program.

Ø  Shrinkage process #2: Some specimens are permitted to dry in humidity room where the flocculation of the RH is between 91.0 % to 93.0 %. The test is stopped once the specimens have lost approximately 2.0 gm of water and the moisture distribution in the specimen is equilibrated.

Ø  Shrinkage process #3: Some specimens are permitted to dry in humidity room where the fluctuation of the RH is between 91.0 % to 93.0 %. The shrinkage test is stopped when the shrinkage limit is reached. Out of those specimens, some are further dried at room temperature until the shrinkage limit is attained.

Ø  Shrinkage process #4: Some specimens are allowed to dry in the desiccator's container that has the RH of 99.0%. The shrinkage test is stopped once the shrinkage limit is reached.

The above approaches will suggest additional investigations of the desiccation process and shrinkage deformation under the influence of the RH changes that will be discussed in chapter 5.  In the end, duplicate specimens are produced at similar moisture contents to be carried out for the free swell tests as explained in the next section.

4.5 Free Swelling Test

The free swelling test is widely known as the test used to provide a general designation of the swelling potential of expansive soils. It is traditionally performed by submerging the expansive soils with water in a gradual cylinder. The more improved technique by adopting the conventional oedometer device (as discussed in section ....), permitting the simplest method of measuring the axial swelling deformation with time. However, the free swell terminology is normally used to describe the bulk swelling deformation interpreted from very little to unconfined/unpressurized boundaries of the tested swelling soil, which can either be performed in one-dimensional perspective such as the latter approaches or in three-dimensional perspective such as the one performed in this study. Since the deformation phenomenon of the soil always of a three dimensional nature, the objective of this test is to periodically monitor the extreme case of free swelling behaviour as the specimen is permitted to swell in all directions when subjected to specified water infiltration at different rates. It is decided to select different infiltration rates for the following reasons:

The cumulative increase of moisture content of bulge soil can result both in a reduction of capillary tensions and an increased volume expansion. This is alike to the effect that occurs when the effectual overtax pressure on drenched swelling soil is unloaded.

4.5.1 Experimental Apparatus

The specimens used for shrinkage tests are utilized herein for free swelling tests. A plastic syringe is used to add water drops on the specimens at specific rates. The changes in bulk volume during water uptake are measured by using the same digital camera setup as discussed in section 4.4.2. To ensure the least moisture content loss of the specimens, the humidity room is used as storage.

4.5.2 Experiment Procedure

This is a simple test and yet, very essential to quantitatively indulging the factors which have chief influences on the swelling deformability of the particular costly soil used in this program. The selection of infiltration rates are former to performing the free swelling tests which are found to be satisfactory in between observing the progression of swelling deformation and the test duration.

The free swelling tests are therefore, performed as follows:

I. Approach #1: Some specimens are tested directly at their initial states by imposing different infiltration rates of 0.2 gm/d, 0.5 gm/d, and 1.0 gm/d. It is suggested that the amount of water per day should be added gradually because of the very low permeability of the specimens relative to the specified infiltration rates mentioned above. Particularly in this test, the soil deformation is a response to the initial and current soil moisture equilibrium which in turn, effects the current bulk volume measurements. Therefore, allowing some time for the specimens to be somewhat geometrically isotropic during the swelling process is essential for measuring the void ratios.

II. Approach #2: Some of the specimens that are dried at room temperature and humidity room, rewetted by increasing their moisture content hygroscopically in the glass container until the equilibrium state is reached. Out of those specimens, some of them are further rewetted by applying similar infiltration rates as in approach #1.

III. Approach #3: Some of the specimens that are dried at room temperature, rewetted by increasing their moisture content hygroscopically in the humidity room until the equilibrium state are reached. Out of those specimens, some of them are further rewetted by applying similar infiltration rates as in approach #1.

IV. Approach #4: Some of the specimens that are dried in glass container, rewetted by applying similar infiltration rates as in approach #1.

For all the above approaches, images are taken every day and before adding the next amount of water so that the bulk volume changes can be periodically determined.