Swelling Characteristics of Compacted Claystone-Bentonite Mixtures

High-density claystone mixtures have been suggested as liners in radioactive waste repositories. However, this material can also be used as a liner in landfills. This study focuses on swelling characteristics of compacted claystone-bentonite at a low density for landfill applications. Claystone was taken from the Banjarbakula landfill project in Banjarbaru. The bentonite used is a commercially available bentonite from Indonesia. Claystone was mixed with 5, 10, 15, and 20% of bentonite on a dry weight basis. The mixture was statically compacted with moisture contents of 10, 15, and 20% to achieve a dry density of 16 kN/m3. A swell-load test was carried out using conventional oedometer equipment to obtain the swelling potential and swelling pressure of the samples. The results show that the swelling potential and swelling pressure increased with an increase in bentonite contents. At a bentonite content of more than 10%, both increased significantly. This research also revealed a linear relationship between swelling potential and swelling pressure.


Introduction
The claystone-bentonite mixture is a material recommended for use as a liner for radioactive waste [1][2][3][4]. Aside from its economic and environmental benefits, the use of claystone has the advantage of containing minerals and chemical compositions that are compatible with the host rock [1]. However, the minerals in claystone vary by region. Other materials, such as bentonite, are required to achieve the permeability requirement as a liner.
Bentonite is required because it contains the mineral montmorillonite, which absorbs more water and continues to expand. When using bentonite as a sealing material, the swelling property is important. The swelling potential and swelling pressure are two important aspects of this swelling characteristic [1,5,6], [7]. The swelling potential, also known as the percent swelling, is the ratio of deformation to the initial sample thickness. Meanwhile, swelling pressure is the pressure required to keep the void ratio constant as the sample absorbs water [1]. By this definition, there are three methods for determining swelling pressure: the constant volume test, the swell under load test, and the swell-load test [1]. According to some researchers, different methods produce different values of swelling pressure due to differences in loading and wetting conditions [1,8]. Furthermore, swelling pressure is highly dependent on the initial dry density of the soil specimen. The swelling pressure increases as the initial dry density increases. The swelling pressure, on the other hand, is unaffected by the initial water content. Meanwhile, both the initial dry density and the moisture content of the compacted soil influence the swelling potential [5][6][7]. A mixture of claystone and bentonite is typically recommended as a radioactive waste liner. As a result, materials must have a high dry density. Wang et al. [1] investigated the swelling properties of a claystone-bentonite mixture with a dry density of 17-19 kN/m 3 . Zhang [2], on the other hand, used a sample with a dry density of 18.8-19.5 kN/m 3 . For landfills, the required density is not too high as long as the permeability requirements for Indonesia are less than 1×10 -8 m/s [9,10]. This research focuses on the swelling characteristics of compacted claystone-bentonite with a dry density at maximum Proctor compaction (i.e., 16 kN/m 3 ). To obtain the swelling potential and swelling pressure of the samples, a series of swell load tests were performed. The effects of bentonite content (5-20%) and sample moisture content (10-20%) on swelling characteristics were studied.

Materials
Claystone and bentonite are the two materials used in this study. The claystone used derives from the Banjarbakula landfill project in Banjarbaru. It is estimated that 8000 m 3 of claystone was dumped during the project's construction. The claystone has a specific gravity of 2.60 and a water content of 2.75% on average in the field. The soil contains 5.5% sand, 43.9% silt, and 51.6% clay and has a liquid limit of 50.76%, a plastic limit of 20.95%, a plasticity index of 29.8%, and a shrinkage limit of 9.74%. Ca 2+ , with a 4.3 meq/g exchangeable cation, is the dominant exchangeable cation in the claystone. Meanwhile, the bentonite used is commercial bentonite from Indonesia. The specific gravity of the material is 2.71, with a hygroscopic water content of 14.1%. Bentonite is composed of 1.4% fine sand, 8.3% silt, and 90.3% clay. The sample has a liquid limit of 351.71%, a plastic limit of 44.68%, a shrinkage limit of 41.89%, and a plasticity index of 307.03%. The main exchangeable cation of Ca 2+ in bentonite has a capacity of 18.70 meq/g.

Techniques and Procedures 2.2.1. Sample preparation
The claystone was subjected to a Proctor compaction test, which yielded an optimum moisture content of 15% and a maximum dry density of 16 kN/m 3 . This density was used in this study, along with an additional moisture content at dry of optimum (i.e., 10%) and wet of optimum (i.e., 20%). Crushed claystone was sieved with a No. 40 mesh. Bentonite was mixed at a certain weight to achieve dry weight contents of 5, 10, 15, and 20% in the mixture. After evenly mixing the claystone and bentonite, water was gradually added with a sprayer to attain a moisture content of 10, 15, and 20%. The sample was statically compacted in a mold with a diameter of 6.2 cm and a height of 2 cm to achieve a dry density of 16 kN/m 3 . Claystone with the same density and moisture content was tested for comparison. Claystone and bentonite are referred to as "CS" and "B" in this paper. Meanwhile, moisture content is denoted by the letter "w." Table 1 summarises the conditions and sample codes used. 20% 100CSw20 95CS5Bw20 90CS10Bw20 85CS15Bw20 80CS20Bw50

Swelling potential and swelling pressure tests
The swell under load technique was used for the swelling test. In this test, the oedometer was used to obtain both the swelling potential or swelling strain and swelling pressure in the same specimen. In the first step, the sample was subjected to a pressure of 6.9 kPa and inundated with water until equilibrium was reached, as indicated by no change in the dial reading during testing [11]. The load was then gradually added to the load increment ratio (LIR) in the same way as in the consolidation test (i.e., LIR = 2) until the void ratio equalled its initial value. Based on a swell-load test, the pressure at which the void ratio is equal corresponds to the swelling pressure [12]. Figure 1 depicts the swelling development over time for 100% claystone samples and claystonebentonite mixtures with bentonite contents ranging from 5 to 20%. Figure 1(a) represents the swelling development of a sample with a moisture content of 10%, while Figures 1(b) and 1(c) exhibit the swelling development of samples with moisture contents of 15% and 20%, respectively. As shown in the figure, the swelling strain increases over time and remains constant at a certain point. The swelling potential of the samples is determined by the swelling in which no noticeable difference is visible in the graph. The swelling strain was found to be constant over less than 100 minutes for 100% claystone samples (i.e., 100CSw10 and 100CSw15). Meanwhile, the swelling potential of other samples could be determined after more than 100 minutes, and samples with bentonite contents of 15% and 20% could be determined after more than 2000 minutes. As the percentage of bentonite in the mixture increased, the time to reach equilibrium rose.  Table 2 summarises the swelling potential of the samples, which is plotted as a function of bentonite content in Figure 2. In general, as shown in Figure 2, the swelling potential increased as the bentonite content increased. If the increase in swelling per percentage of bentonite is calculated as the difference in swelling divided by the addition of bentonite content, then for samples containing up to 10% bentonite, a 0.4-0.7% increase in swelling was obtained per one percent of bentonite. Meanwhile, for samples containing more than 10% bentonite (i.e., 15-20% bentonite), the swelling addition was 1.1-1.4% per one percent bentonite. As a result, samples with a bentonite contents greater than 10% showed a greater increase in swelling. Similarly, the difference in swelling potential data for each water content is divided by the percentage of water content to investigate the effects of water content. The increase or decrease in swelling potential per one percent water content was found to be 0.1-0.3%. It can be concluded that the initial water content of claystone-bentonite mixtures had no significant effect on their swelling potential.

Swelling Pressure
The swell load test result of a claystone-bentonite mixture sample utilised in this investigation is shown in Figure 3. When the sample was inundated with water, the void ratio increased from 0.40 to 0.48, as shown in Figure 3 (a) for the 90CS10Bw15 sample. This data was used as the amount of the swelling potential. The void ratio decreased with increasing pressure, and the initial void ratio was intersected at 52.32 kPa. In Figure 3 (b), the sample void ratio of 80CS20Bw20 grew during the wetting process from 0.39 to 0.61, then declined with rising pressure until it intersected with the sample's initial void ratio at 183.82kPa. The swelling pressures for 90CS10Bw15 and 80CS20Bw20 were 52.32kPa and 183.82kPa, respectively.  The swell load test results are summarised in Table 3 and plotted as a function of bentonite content in Figure 4. Figure 4 shows that the higher the bentonite content, the greater the swelling pressure of the sample. The average increase in swelling pressure per one percent addition of bentonite was 1.4, 5.87, 15.15, and 13.43% for samples containing 5, 10, 15, and 20% bentonite, respectively. As a result, it can be concluded that the highest increase in swelling pressure occurred when the bentonite content exceeded 10%.  Figure 4 also shows that specimens with a moisture content of 15% had the highest swelling pressure. This is attributed to the fact that the water content represents the optimum moisture content of the claystone sample used in this study. The swelling pressure results were influenced by the dominant claystone in the mixture. Despite having the same initial dry density, the swelling pressure obtained by the swell-load test method varied with moisture content. This is due to the fact that the microstructure of samples compacted at various water contents results in a different macro-micro pore distribution [13]. Furthermore, because the void ratio of the samples changed during the test, the stress paths traversed by the samples were also different. According to Sridharan et al [12], the swelling pressure tested using the swell-load method is timeconsuming. Meanwhile, the swelling potential can be obtained faster. The relationship between the swelling potential and swelling pressure of the claystone-bentonite mixture samples obtained in this study is depicted in Figure 5. Regardless of the sample's bentonite or water content, the relationship between the two can be determined by a straight-line equation with R 2 = 0.9512. (Equation 1).
where Ps and Sp denote swelling pressure in kPa and swelling potential, respectively. Moreover, the linear swelling pressure and swelling potential relationship was accomplished using the Ps (kPa) = 10.8 sp equation, whereas Ps and Sp were swelling pressures in kPa and swelling potential.  Figure 5. Swelling pressure and swelling potential correlation.

Conclusions
The swelling characteristics of a claystone-bentonite mixture, including its swelling potential and swelling pressure, have been described and analysed. The swelling potential of the sample increased as the bentonite content increased. The greatest increase occurred in samples containing more than 10% bentonite, with 1.1-1.4% per one percent bentonite. The sample's swelling pressure rose as the bentonite content rose. The greatest increase was also observed in samples with a bentonite content greater than 10%, with an average increase of 13.3-15.3% per one percent increase in bentonite in the mixture. The water content had no significant effect on the swelling potential or swelling pressure of the compacted claystone-bentonite investigated in this study.