1.Introduction

In general, expansive soils are usually widely distributed all over the world [1]. This type of soil is very sensitive to changes in its moisture content and due to these large changes in volume occur [2]. Because of that, possibly inflicts considerable damage to the structures, especially for lightweight buildings and pavements, etc. [3] This performance is commonly due to containing montmorillonite in clay minerals. In practice situations, the swelling problems stabilized using various materials and additives to improve its engineering properties [4, 5]. One of the new and sustainable materials used as a new generation for soil improvement is geopolymer, which combines industrial waste materials including for example fly ash [6, 7].

A number of previous studies included the use of fly ash with different activators for soil stabilization. Bose [8] studied the soil stabilization use of fly ash with different percentages (0, 20, 40, 60, 80, and 90 %), the addition of fly ash can reduce the free swell index, and values of swelling pressure as well as change the grain size of treated soils, also reducing the plasticity of expansive soil. Hasan [9] found that adding 20 % of fly ash can reduce free swelling by 33 % and swelling pressure by 40 %. This fly ash percentage reduced the expansive soil's liquid limit, while the plastic limit decreased by 18 %, and the plasticity index was 28 %. Phani Kumar & Radhey [10] used the same proportions of fly ash that were already used by Hasan [9], but the potential of swelling of the soil was less, where it was found that both swelling pressure and free swelling were reduced by 50 % when 20 % of fly ash was used. As for the liquidity limit, it decreased by 12.5 %, and the plasticity limit increased by 36 %. Salim [11] added different percentages of fly ash (5, 10, and 15 %) of the dry weight to expansive soil containing different percentages of bentonite (30, 50, and 70 %), where the optimum percentage was 5 % of fly ash, which results in less swelling pressure and less swelling with better workability. Das & Parhi [12] used alkaline activated fly ash (AAFA) in different proportions (0, 5, 10, and 15 %) of the dry weight of the soil, to improve the properties of the expansive soil, where it was observed that the free swelling index (FSI) decreased to one-third of the initial soil FSI at 15 % AAFA content. Sharma & Sivapullaiah [13] used a mixture of fly ash and ground granulated blast furnace slag (GGBS) in different proportions to improve the properties of the expansive soil, including the potential swelling, FSI, and swelling pressure. The results revealed that the swelling behavior decreases with increasing the concentration of the binder and adding 1 % lime to the binder, increasing the ability to reduce swelling. Phanikumar [14] compared the addition of two pozzolanic materials, lime, and fly ash, at different percentages (0, 2, 4, and 6 % and 0, 10, and 20 %, respectively), as the results indicated that their effects on the swelling potential and swelling pressure are close. Radhakrishnan et al. [15] studied the effect of mixing chlorides (Aluminum Chloride AlCl3, Magnesium Chloride MgCl2) with fly ash in different proportions (0, 0.5, 1, 1.5, and 2 % and 0, 5, 10, 15, and 20 %, respectively), on the swelling properties (swell potential, FSI, and swell pressure) on expansive soil, the effect of 1 % AlCl3 with 10 % fly ash was found to be more effective than the other two. Sabat & Pradhan [16] discussed the suitability of mixtures of expansive soil reinforced with fibers and stabilized with fly ash as subgrade materials for flexible pavement. The results showed that 1 % of propylene fibers with a length of 12 mm is the optimal ratio for strengthening expanded soil stabilized with an ideal amount of 20 % fly ash, as the swelling pressure decreased to 17 kN/m2.

Fly ashes are microscopic particles made up primarily of silica, alumina, and iron [17]; it is a pozzolanic substance produced by coal burning in thermal power stations [18]. It is simply a non-plastic fine silt whose composition changes depending on the type of coal burned. Currently, the production of fly ash much outnumbers its consumption.

Based on previous studies, it can seem that the application of treated soil with geopolymers in laboratory models is very limited with different thicknesses. Therefore, this study involves optimizing layers of different thicknesses of expansive soil using a small-scale laboratory model.

2.Methods and Materials

2.1.Expansive Soil

The expansive soil was prepared by mixing 80 % bentonite and 20 % sand, with adding the optimum water content, and mixing it with the soil until the mixture becomes homogeneous. Table 1 displays the expansive soil's mechanical and physical characteristics.

Table 1. Summary of the engineering properties of the soil used.

Standard

Soil Property

Value

ASTM D 4318

USCS classification

CH

ASTM D 854

Specific gravity

2.78

ASTM D-422

Liquid limit, %

98

Plastic limit, %

45

Plasticity index, %

53

ASTM D-3084

Free swelling, %

13

Swelling pressure, kPa

230

Compression index (Cc)

0.16

ASTM D-1557

Maximum dry unit weight, kN/m³

14.45

Optimum moisture content, %

26

ASTM D-2216

Unconfined compressive strength(qu), kPa

245

2.2.Fly Ash and Alkali Activator

In this study, fly ash class (F). Its chemical properties are shown in Table 2, sodium hydroxide, and sodium silicate were used as alkaline activators to accelerate the reaction. Sodium hydroxide (NaOH) is a chemical compound that dissolves in water and is used in wide fields. Sodium silicate (Na2SiO3) is an important chemical compound also known as water glass.

Table 2. Fly Ash chemical composition.

Chemical Composition

Percent (%)

SiO2

47.67

Al2O3

27.73

Fe2O3

18.42

CaO

5.11

MgO

2.65

TiO2

1.3

K2O

0.6

2.3.Experimental Model

In this study, a container used with dimensions 450 × 450 × 500 mm made of iron with a thickness of 40 mm, shown in Fig. 1. It is supplied with water by an elevated small tank. The clay soil thickness is 20 cm, and the upper soil layers were treated starting from the soil surface divided into three stages, the first stage is 2 cm the second is 3 cm, and the last stage is 4 cm. A square aluminum plate 10 × 10 cm with a thickness of 1 cm was used as a foundation and placed in the center of the container. This foundation is designed to carry 30 kPa. After applying the surface load, the soil is supplied with water until the soil reaches a fully saturated state, and then a dial gauge reading is taken with respect to load and free soil surface every 24 hours.

Figure 1. Experimental model.

3.Result and Discussion

In this part, the results include the basic tests (i.e., Atterberg limits, swelling pressure, and free swelling) that were carried out for the expansive soil before and after improvement. When adding the fly ash-based geopolymer at five different percentages (0.5, 1, 2, 4, and 6 %) from the weight of dry soil. Based on the results, it can be concluded that the best percentage was 2 %, which shows high improvement and means transferring the soil stat from high to very low according to potential expansion (PE) (ASTM-D4829). Depending on that, this percentage has been selected as an additive in the small-scale laboratory model.

Fundamental Tests Results

Fig. 2 shows the effect of fly ash-based geopolymer percentages on the Atterberg limits, where it was observed that they gradually decrease with the increase of the fly ash-based Geopolymer content. The liquid limit value decreased from 98 to 65 %, i.e., a decrease of 34 %, the plasticity index value also decreased from 53 to 35 % at 4 % geopolymer. The interpretation of the above results is that, as a result of replacing coarse fly ash particles with fine soil particles, the addition of fly ash-based geopolymer reduces the volume fraction of clay in the soil, induces the flocculation of the clay particles, and raises the number of coarse particles [19], decreasing the soil's liquid limit and plasticity index.

Figure 2. The effect of fly ash-based geopolymer percentages on the Atterberg limits.

By performing swell-consolidation tests with samples that were 75 mm in diameter, and 16 mm in thickness, the swelling potential and swelling pressure of the blended samples were calculated at various percentages of fly ash-based geopolymer, according to ASTM-D4546, the free swell method. With a rise in geopolymer content, swelling potential and swelling pressure dropped and reached more than 90 % as illustrated in Fig. 3. As a result of reduced clay minerals' capacity to absorb water due to ion exchange with geopolymer, pozzolan, which fills gaps, is produced.

Figure 3. Effect of geopolymer on swelling percentage and swelling pressure.

Results of Laboratory Model

In this part, the results obtained from the laboratory model have been presented and discussed. Fig. 4 shows the amount of swelling under the foundation and the amount of swelling in the free surface for the untreated soil over a 16-day time period. It was observed that the swelling of the free surface of the soil reached 26 mm, which means 13 % of the total height of expansive soil. While the amount of swelling under load was 10.25 mm, which represents 5 %. According to the ASTM-D4829, the swelling of the free surface of the soil can be categorized as high PE. On the other hand, soil swelling under load is about two and a half times less than the free swelling of the soil surface. The differences between the swelling potential (i.e., free and under load) are a result of the influence of load that usually reduces the swelling potential.

Figure 4. Swelling of free soil and load vs. time for untreated soil.

Fig. 5 shows the free swelling of the untreated soil surface and the free swelling of the treated soil surface in the form of layers of different thicknesses starting from the soil surface to a depth of 4 cm. The free swelling of the treated soil surface decreases at 2, 3, and 4 cm depth by 33.5, 61, and 75 %, respectively, of the free swelling of the untreated soil surface, according to ASTM-D4829, the swelling of the free surface of the soil treated at 2 cm can be categorized as medium PE, and at 3, 4 cm – as low PE.

Figure 5. Swelling vs. time for treated and untreated soil for free surface.

Fig. 6 shows swelling under load for the untreated soil and swelling under load for the treated soil. The amount of swelling under load was less when treating the soil at a depth of 2, 3, and 4 cm by 40, 70, and 88.5 %, respectively, than swelling under load for the untreated soil, according to ASTM-D4829, the swelling of the load of the soil treated at 2 cm can be categorized as low PE and at 3, 4 cm – as very low PE.

Figure 6. Swelling vs. time for treated and untreated soil for load.

Fig. 7 shows the relationship between free swelling of the soil surface and swelling under load with the improved soil layers. It can seem that the thickness of improved soil increased the swelling decreased in all cases. The reason for this is that swelling occurs more in the surface layers compared to the lower layers because the weight of the soil itself prevents swelling with depth. In addition, these improvements occur due to the activity of the geopolymer as it forms a pozzolanic material that fills the gaps and causes flocculation of the clay particles by the exchange cation [20].

Figure 7. Prediction of swelling with an Increase in thickness of the improved layer.

4.Conclusion

Based on the results, it can seem that the use of geopolymer material is good results as expansive soil improvement, the following conclusion is developed:

The best percentage for improving soil properties such as liquid limit, plastic limit, and plasticity index was 4 %, both the liquid limit and plasticity index decreased by 34 %, and the liquid limit decreased by 33 %, but it is possible to use 2 % because the difference in percentage improvement was convergent. The addition of 4 % reduced free swelling and swell pressure by 96 and 95 %, respectively. While the 2 % can reduce free swelling and swell pressure to 90 and 92 %, respectively. That is, it is possible to use 2 % geopolymer instead of 4 % because the results of free swelling and swelling pressure at those percentages of geopolymer are very close. For free soil, the soil transfers from high swelling potential to a low state at 3 and 4 cm improved layer. While, for the soil under load, the soil transfers from a high to low state at 2 cm, and to very low at both 3 and 4 cm.