Wet-dry cycle of expansive soil stabilized with fly ash and waste foundry sand on bearing capacity and swelling potential

The behavior of expansive soil, which is sensitive to water, makes it susceptible to moisture content changes, resulting in the soil becoming soft during the rainy season and soil shrinkage during the dry season. This research aims to study the effects of the wet-dry cycle on the bearing capacity and swelling potential of stabilized expansive clay soil using two types of industrial waste, namely Fly Ash (FA) and Waste Foundry Sand (WFS), as soil improvement and industrial waste management efforts. The wet-dry cycle is assumed to represent the changes in the environment due to seasonal variations. The wetting cycle is conducted by immersing the soil samples in water, while the drying cycle is achieved by allowing them to air dry at room temperature. The FA content is set at 9% and the WFS content at 15% by weight of dry soil. The California Bearing Ratio (CBR) test is performed after the completion of the wet-dry cycle, while the swelling test is conducted during the wet-dry cycle. Each sample undergoes one, two, and three cycles of wetting and drying. The results showed that the CBR value increased with the addition of FA and WFS. The CBR value of natural soil is 8.89% while the CBR value of soil+9%FA+15%WFS is 16.02%. During the wetting phase, the CBR value decreases with increasing moisture content. After undergoing the third cycle, the CBR value and development potential tend to remain constant.


Introduction
Weather changes cause the soil to undergo wet and dry conditions repeatedly.Wet and dry conditions in expansive soil result in damage to infrastructure due to soil expansion when wet and soil shrinkage when dry.One of the efforts to improve expansive soil is by adding additive materials such as Fly ash [1][2][3] or industrial waste such as Waste Foundry sand (WFS) to enhance bearing capacity and reduce the potential for soil expansion [4][5][6].Industrial waste, i.e., fly ash and waste foundry sand, have compounds that can still be utilized for the improvement of expansive soil and can serve as alternatives to additive materials other than cement and lime [7][8].The durability of additive materials is determined by the mineralogical and microscopic content of the stabilized soil [9].Several previous studies have shown that wet-dry cycles lead to changes in bearing capacity and potential for expansion in expansive soil [10][11].The longevity of the improved material's resistance to environmental changes is also a crucial factor to consider.
This research aims to analyze the use of FA and WFS as additive materials in improving expansive soils.The focus of this study is to examine the bearing capacity and expansiveness of stabilized soils using industrial waste materials FA and WFS, considering the effects of wetting and drying cycles.

Methods
This research was conducted in several stages, including material preparation, sample preparation, testing of physical properties, testing of mechanical properties such as compaction tests, California Bearing Ratio (CBR) tests, and testing of soil swelling potential.The testing of physical properties and mechanical properties was based on ASTM standards.CBR test based on ASTM D-1833.Chemical analysis of oxide elements (XRF) was also performed to analyze the chemical components of the soil.

Test specimen preparation
The disturbed soil was obtained from the industrial area of East Karawang, Ciampel District, Karawang, West Java, Indonesia, at a depth of 0.5m-3m from the natural soil surface.The FA content was set at 9% and the WFS content was set at 15% by dry weight of the soil.The Fly Ash used was sourced from the Paiton Jawa Timur Power Plant in Indonesia, while the WFS used was sourced from a metal casting factory in the Surya Cipta industrial area, Karawang, West Java, Indonesia.The soil was mixed with FA and WFS in a dry state and then moistened with the optimum moisture content (OMC).

Testing of natural soil physical properties
Testing of the natural soil physical properties was conducted to determine the soil type and characteristics in a laboratory.The physical properties test was conducted on dry soil that passed ASTM sieve number 40.Furthermore, specific gravity test was conducted according to ASTM D-162 standard, Liquid limit, plastic limit, and plasticity index according to ASTM D-4318 standard and Shrinkage limit according to ASTM D-427 standard.

Compaction, CBR, swelling, and wet-dry cycle testing
Proctor modified compaction test was conducted on the natural soil and the mixture of expansive soil (+ 9% FA + 15% WFS) to determine the Optimum Moisture Content (OMC) and Maximum Dry Density (MDD) values.The wet-dry (W/D) cycle was performed to simulate environmental changes [11].The soil mixture samples containing 9% FA and 15% WFS at the OMC were compacted using a modified Proctor compaction mold.The compacted samples were then placed in tightly sealed containers and cured for 4 days.This phase is referred to as the 0th cycle.After the curing process, the samples were immersed in water for 4 days.The height difference of the samples was measured to determine the volume change and swelling potential.Each sample underwent wet-dry cycles for 1, 2, and 3 periods.One period consisted of immersion for 4 days followed by air drying for 1 day.The CBR test was conducted after the wet-dry cycles were completed, whereas the swelling test was performed during the wet-dry cycles.Chemical analysis of the soil (X-Ray Fluorescence/XRF test) was conducted at the initial condition (0th cycle) and after 3 wet-dry cycles.

Results and Discussions
The results of the physical properties test of the natural soil are presented in Table 1.Based on the Unified Soil Classification System (USCS) and American Association of State Highway and Transportation Officials (AASHTO) classification, the soil is classified as high plasticity clay.The results of the mechanical properties test of the natural soil are presented in Table 2.
The swelling potential of the natural soil is 9.82%, indicating a high degree of expansion [12][13].The presence of montmorillonite mineral in the expansive soil causes the clay mineral to easily expand.The mineral structure of montmorillonite consists of one layer of gibbsite sandwiched between layers of silica.The interlayer bonding between the silica layers is very weak due to van der Walls forces, and there is a deficiency of negative ions in the octahedral sites, allowing water and ions to easily enter and break the bonds.This is what causes the clay mineral to easily expand [11].Figure 1 shows that the addition of 9% FA and 15% WFS to expansive soil increases the CBR value.The CBR value of natural soil is 8.89%.The CBR value of soil + 9%FA + 15% WFS is 16.02%.The self cementing nature of fly ash is able to harden the soil when mixed with water.The addition of WFS serves as a filler in expansive soils capable of improving soil density thus increasing its bearing strength [8].This is consistent with previous studies [7,8,14] which state that Fly Ash and WFS can improve the CBR value.The swelling test aims to determine the percentage of soil swelling during soaking in water.The swelling value is calculated from the change in the initial height of the sample expressed in percent.The sample used in the swelling test is the same as the soaked CBR (Figure 1).In this test, observations are made during the soaking period.The swelling value of natural soil is 9.82% while the swelling value of soil+9%FA+15%WFS is 3.22% (Figure 2).The addition of FA and WFS additives to expansive soils can reduce the swelling value.Fly ash helps to bind the soil particles, forming a denser structure that is less susceptible to volume changes due to changes in moisture content.This binding of particles prevents cracks and fissure formation in the soil, which can significantly reduce swelling values [1].The CBR results for each cycle are presented in Table 3 and Figure 3.There was a decrease in the CBR value after the completion of the first cycle from dry to wet, from a CBR value of 8.89% to 0.94%.Furthermore, in the second cycle, the CBR value increased slightly to 1.30%.In the third cycle, the CBR value became 1.31%.Similarly, the CBR value of soil+9%FA+15%WFS has a similar pattern with the natural soil.In the first cycle from dry to wet, the CBR value went from 16.02% to 6.46%.In the second cycle the CBR value increased slightly to 6.53%, and at the end of the third cycle the CBR value became 6.6%.Drying for 1 day is insufficient to evaporate the absorbed water from the soaking process, resulting in a high moisture content and causing the CBR values to remain relatively constant during each.The large decrease in strength in the first cycle was due to the increase in moisture content of the expansive soil during immersion.Expansive soils contain the mineral Montmorillonite.This montmorillonite mineral has a very small size and a very large water attraction because the bond between crystals with van der wall forces between silica sheets is weak and lacks negative ions in the octahedral so that water and ions easily enter to break the bond.The binding of water in the first cycle of immersion is already very large, therefore in the next cycle the addition of the water binding process becomes insignificant [11].Soil stabilized with 9% FA and 15% WFS exhibits similar phenomena to natural soil.Due to the wettingdrying process, the soil strength decreases, indicating that the bond between the soil and FA/WFS becomes weaker due to the influence of water, resulting in a decrease in CBR value.However, the addition of FA and WFS as additive materials to expansive soil ensures that the CBR value still meets the requirements for road subgrade (>6%), even though the water content remains high due to repeated wetting-drying cycles.The swelling potential value in the soil with +9%FA+15%WFS mixture remains constant during the application of 3 cycles (Figure 4).The shrink-swell behavior does not reoccur even with the wetting-drying cycles [11].The swelling value of natural soil in the first cycle was 9.82%.After the second cycle, the swelling value increased to 9.98% and then slightly decreased in the third cycle to 9.91%.Similarly, the swelling value of soil+9%FA+15%WFS.The first cycle swelling value of soil+9%FA+15%WFS was 3.22%, in the second cycle it slightly increased to 3.25%, then decreased again at the end of the third cycle which was 3.23%.The swelling values in the three cycles did not show significant changes.Figure 4 shows that the swelling potential of the natural soil remains relatively constant during the 3 wetting-drying cycles, indicating minimal changes.Whereas wetting is defined as the process of increasing the water content in the soil, drying is defined as the process of reducing the water content and degree of soil saturation due to evaporation and evapotranspiration.Since the drying process takes a very long time to return to its initial condition, the focus is primarily on the changes that occur during the wetting phase in each cycle.Most of the pores in the expansive soil filled with FA result in stronger chemical compounds such as (C-S-H) and (N-A-S-H), which form cementitious products, thereby increasing the soil strength [18].Chemical composition testing of the soil was conducted on natural soil and soil with +9%FA+15%WFS at cycle 0 and cycle 3.The purpose was to obtain information regarding the changes in chemical composition that occur due to the wetting-drying cycles.In natural soil, the wetting-drying cycles do not cause changes in the chemical composition.In the soil+9%FA+15%WFS, the wetting-drying cycles lead to a decrease in SiO2, an increase in CaO, and Fe2O3.

Conclusion
Industrial waste (fly ash and waste foundry sand) as additives for stabilizing expansive soil have a positive effect on improving the bearing capacity and reducing the swelling potential, particularly under high moisture conditions.The addition of fly ash and waste foundry sand to expansive soil increases the CBR value and decreases the swelling potential.Seasonal changes, especially during the rainy season, negatively affect the bearing capacity, but in subsequent cycles, there is no significant change observed.The large decrease in CBR values in the first cycle is due to the soil grains absorbing water and becoming saturated.This can reduce the interaction between soil particles and increase the pore pressure in the soil.High pore pressure can weaken the soil structure and reduce its bearing capacity, which in turn can result in lower CBR values.

Figure 3 .
Figure 3.The influence of wetting-drying cycles on the CBR.

Figure 4 .
Figure 4. Influence of wetting-drying cycles on soil swelling potential.

Table 1 .
Physical properties of natural soil.

Table 2 .
Mechanical properties of natural soil.

Table 3 .
The results of the swelling and CBR during cyclic wetting-drying cycles.

Table 4 .
Chemical composition of soil.Table4shows that the SiO2 and CaO content in the soil with +9%FA+15%WFS is greater than in natural soil.The microstructure of the soil improves due to the formation of hydration products[15].The higher availability of silica, alumina, and calcium supports a greater increase in strength[16][17].