Performance of Zeolite-Based Soil–Geopolymer Mixtures for Geostructures under Eccentric Loading

Abstract

Although soil stabilization with cement and lime is widely used to overcome the low shear strength of soft clay, which can cause severe damage to the infrastructures founded on such soils, such binders have severe impacts on the environment in terms of increasing emissions of carbon dioxide and the consumption of energy. Therefore, it is necessary to investigate soil improvement using sustainable materials such as byproducts or natural resources as alternatives to conventional binders—cement and lime. In this study, the combination of cement kiln dust as a byproduct and zeolite was used to produce an alkali-activated matrix. The results showed that the strength increased from 124 kPa for the untreated clay to 572 kPa for clay treated with 30% activated stabilizer agent (activated cement kiln dust). Moreover, incorporating zeolite as a partial replacement of the activated cement kiln dust increased the strength drastically to 960 and 2530 kPa for zeolite ratios of 0.1 and 0.6, respectively, which then decreased sharply to 1167 and 800 kPa with further increasing zeolite/pr to 0.8 and 1.0, respectively. The soil that was improved with the activated stabilizer agents was tested under footings subjected to eccentric loading. The results of large-scale loading tests showed clear improvements in terms of increasing the bearing capacity and decreasing the tilt of the footings. Also, a reduction occurred due to the eccentricity decreasing as a result of increasing the thickness of the treated soil layer beneath the footing.

1. Introduction

The construction of shallow foundations, pavements, dams, and other structures on weak soils may result in severe problems because such soil is characterized by low shear strength and high deformation. Moreover, the situation is worse in the case of footings positioned on soil subjected to movement in addition to vertical load, i.e., the eccentricity effect. Under such conditions, the contact stress decreases linearly from the eccentricity side, which subsequently decreases the bearing capacity of the soil. Therefore, it is necessary to enhance the engineering characteristics and performance of soil by incorporating additional materials [1,2,3,4,5].

Cement and lime are the most popular stabilizing agents or binders used in chemical soil stabilization to boost the geotechnical characteristics of weak soil [6,7,8,9]. However, an excessive reliance on cement and lime has led to significant CO₂ emissions, the depletion of natural resources, and dust pollution, among other environmental issues. The production of OPC uses a lot of energy (5000 MJ/t PC), which results in CO₂ emissions of 0.7–1.1 tonnes per tonne of OPC [10,11,12]. Moreover, cement production is linked to significant water consumption and the release of gasses including CO, CO₂, and SO₂ [13]. Therefore, sustainable materials are necessary as a replacement for OPC and lime to reduce their environmental effects in addition to improving the mechanical properties of stabilizing agents or binders [14,15]. Moreover, worldwide, industries create problematic waste. Industrial waste disposed of in an uncontrolled manner harms both the environment and people. In geopolymer synthesis, industrial waste such as fly ash, silica fume, steel slag, and cement kiln dust are employed.

Geopolymers are a more environmentally friendly alternative to traditional cement and lime. The CO₂ footprint of geopolymer production can be reduced by 80% compared with the manufacturing process of Portland cement [16,17,18]. In order for polymerization to occur, alkali earth and alkali-based cations are necessary. The resulting alkaline solution is characterized as a liquid with a higher pH than water [16,17,18,19,20]. These compounds can be classified as either homogeneous or heterogeneous. Typically, the production of geopolymers involves using an aluminosilicate binder with either hydroxides or alkali silicate [20,21,22,23,24]. Geopolymers are predominantly produced by synthesizing byproducts or natural materials such as calcined clay (CC), fly ash (FA), ground granulated blast furnace slag (GGBS), red gypsum, and palm oil fuel ash (POFA), which serve as aluminosilicate stabilizer agent materials that contain chains of silicon (SiO₄) and aluminum (AlO₄) tetrahedra [20,21,22,23,24,25]. Such composites are created by interacting alkali-activated silicate or hydroxide powder with binders to create a solid aluminosilicate substance [21,22,23,24,25,26]. The structure of geopolymers is commonly described as a network of interconnected mineral compounds and chains held together by hydrogen bonding.

Over the past few years, there has been growing interest in using fly ash-based (FA-based) geopolymers as a substitute for cement in improving soil strength and reducing harmful pollutants. According to [26,27], utilizing low-calcium fly ash with a specific activator solution can provide an affordable solution with positive results in soil stabilization for various purposes. It was found that building bricks made with low-calcium fly ash and geopolymers had a water absorption percentage of 19% and a compressive strength of 17 MPa [28]. Furthermore, the unconfined compressive strength (UCS) of soaked soil treated with fly ash-based geopolymer increased by 80%. Suksiripattanapong et al. (2021) [29] evaluated a fly ash-based geopolymer mixed with polyvinyl alcohol (PVA) to improve soft clay. The optimal content of fly ash was 40% and that of polyvinyl alcohol was 4–15%. The UCS was 40% greater than the ordinary FA-based alkaline binder. Thus, polyvinyl with fly ash is an efficient mixture for soft soil. When conducting FA-based geopolymer treatment on gypsum soil, it is essential to study the intrinsic sulfur content and examine the soil’s collapsibility. Alsafi et al. (2017) [30] found that fly ash activated using potassium reduced the probability of collapse and maintained permeability over a longer duration. Using traditional cement to stabilize gypsum soils will increase the sulfate reactivity because such a soil type has a high sulfate concentration. Therefore, combining alkaline activators with fly ash is a highly effective method for removing sulfates from these specific types of soils [31]. Soil with a high sulfate content is hard to remediate owing to its protracted heaving.

Phetchuay et al. (2016) [32] recommended treating soft and compressible marine soil with fly ash-based geopolymer to minimize carbon emissions. Arulrajah et al. (2018) [31] found a significant increase in the UCS of such soil after mixing it with 5% fly ash and 15% slag in an alkaline solution. According to Coudert et al. (2019) [33], the use of alkaline polymer combined with high-calcium fly ash can improve the performance of soft clay soil. This combination may also speed up the interaction between the soil and cement by forming calcium silicon links with reactive stages. Esaifan et al. (2015) [34] investigated the potential of alkali activators to enhance the stability and strength of kaolinitic materials. Dassekpo, Zha, and Zhan (2017) [35] discuss the chemical interaction between alkaline fly ash and soil with low strength. When considering the effects of geopolymers on particle contact, the percentage of soil and alkaline used for soil stabilization is determined based on both its long-term and short-term impacts. According to Yaghoubi et al. (2018) [36], fly ash improved the shear strength of geopolymer-stabilized soil.

In general, polymerization involves solubility, ion movement, gel formation between particles, the restructuring of molecules, and particle hardness. Every geopolymeric system consists of silica and alumina as its fundamental components. When in the gel state, the reaction between aluminosilicate and the alkaline solution intensifies significantly as a result of alkali metal copolymerization [37,38]. Xu et al. (2021) [39] mentioned that the activator dosage is a key feature of geopolymer materials. They also proposed that increasing the activator solution content and sodium silicate modulus may boost geopolymer strength. If the activator content is increased, it could potentially limit the movement of ion nanoparticles and result in a slower polymerization process [40]. In most circumstances, when substances containing aluminosilicate react with a high level of alkaline, they break down the bonds in the siliceous complex. This results in the creation of a new stage of aluminosilicate polymer, with an increased alumina amount [38]. Azevedo et al. (2021) [41] concluded that the incorporation of aluminosilicate stabilizer agents into geopolymerized samples led to improved characteristics in terms of increasing the strength and decreasing the deformation. This means that the combination of a stabilizer agent with high calcium content and another stabilizer agent with high aluminosilicate content can lead to an efficient geopolymerized matrix. It was observed that the best improvement relies on the amount of NaOH and Na₂SiO₃ solution [16,17,18]. It was found that despite the natural soil being cohesionless, the treatment resulted in clear cohesion that bonded the individual sand grains and produced a stable and stiff matrix due to the gel developed during the geopolymerization process.

The other potential source for the geopolymer is cement kiln dust, which is an accumulation of clinker-entrained particles, raw material, and certain calcined raw materials. Cement kiln dust is a particulate substance derived from the exhaust emissions of Portland cement kilns [19,20,21]. Cement facilities have become increasingly concerned about the production of cement kiln dust in recent years. This increase may be attributed to the direct costs associated with producing this waste material or to government-imposed environmental restrictions on the cement industry. Moreover, the estimated proportion of cement kiln dust that is caused by the production of cement is 25–45%, resulting in hundreds of millions of metric tonnes of cement kiln dust being produced annually on a global scale [22,23]. If these substantial quantities are not recycled within the cement industry or put to use in alternative industrial contexts, they are deposited in landfills, leading to adverse environmental impacts. Calcium carbonate and silica oxides make up a significant portion of the chemical composition of cement kiln dust, which is remarkably similar to that of the raw materials utilized in feeding kilns that produce OPC. Consequently, numerous investigations have been carried out to effectively handle cement kiln dust waste and minimize its ecological impacts and disposal expenses by employing it in diverse industrial uses [24,25,26].

Although many studies have been carried out on geopolymers, previous researchers have recommended that extensive efforts should be concentrated on the performance of geopolymers using new precursors under different conditions. Thus, this study aims to use a new byproduct aluminosilicate precursor to form alkali-activated binders to improve the bearing capacity of weak soil.

To the best of our knowledge, no previous study has investigated the application of a geopolymer consisting of a combination of cement kiln dust and zeolite, which represents the novelty of this research. This study is the first to investigate and validate the potential application of a new zeolite-based geopolymer as a stabilizer for clayey soil-supported footings under various eccentric loadings.

2. Materials and Methods

2.1. Materials

2.1.1. Native Soil

The soil used in this study is a high plastic clay—CH.

Table 1. (a) Some properties of the native soil, cement kiln dust, and zeolite used in this study. (b) Chemical analysis of the cement kiln dust and zeolite.

(a)
	Passing Sieve No. 200, %		Liquid Limit, %		Plastic Limit, %	Specific Gravity		Maximum Dry Density, gm/cm3
Clayey soil	77		51		22	2.74		1.75
Cement kiln dust	94		30		4.8	2.87		1.64
Zeolite	100		45		20	2.3		1.46
(b)
	SiO2	Al2O3	Fe2O3	CaO	MgO	SO3	K2O	Na2O	Cl
Cement kiln dust	11.5	4.8	2.1	46.1	3.1	7.2	1.98	2.9	0.84
Zeolite	64.2	14.2	2.1	8.5	0.71	0.53	Nil	3.1	Nil

(a) summarizes some of the clay’s properties, while Figure 1 and Figure 2a show the particle size distributions of the materials and a photo of the clay.

2.1.2. Stabilizer Agents

Figure 1 shows the particle size distributions of the two types of stabilizer agents used in this study—cement kiln dust and zeolite. Photos of the stabilizer agents are shown in Figure 2b,c, while their chemical analyses are illustrated in Table 1 (b).

2.1.3. Alkali Activator

The alkali activator has a crucial role in geopolymerization reactions and in activating the stabilizer materials. In this study, potassium hydroxide KOH and potassium silicate K₂SiO₃ were the main components of an alkali solution (Figure 2d,e). The mix design specified that 10 M of KOH and K₂SiO₃ were to be combined in order to produce the alkali activator.

2.2. Method and Testing Program

2.2.1. Unconfined Compressive Test

In this study, two experiments were carried out. The first experiment consisted of extensive tests of unconfined compressive strength (UCS) in order to investigate the effect of the stabilizer agents (cement kiln dust and zeolite) on the strength of the clayey soil. Unconfined compressive strength (UCS) testing was carried out according to ASTM D2166-16 [42] and ASTM D1633-17 [43] to evaluate the efficiency of geopolymer ingredients, where a higher UCS value indicates a more effective geopolymer mixture. Figure 2f,g show the UCS test and the mixture samples.

Different percentages of cement kiln dust as a stabilizer agent (5, 10, 15, 20, 25, and 30%) were considered to investigate its effects on the strength of the stabilized soil. Moreover, in order to determine the impact of incorporating zeolite as a partial replacement of cement kiln dust, another series of mixtures was prepared, where the soil was mixed with different percentages of cement kiln dust and zeolite as stabilizer agents (5, 10, 15, 20, 25, and 30%). The ratios of zeolite to the total stabilizer agent were 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0. Each mixture of the stabilizer agent consisted of a specific percentage of cement kiln dust and zeolite, as shown in

Table 2. (a) Testing program of the unconfined compressive strength test. (b) Testing program of the loading test.

(a)
Series	Test Type	Stabilizer Agent, %	Zeolite/Stabilizer Agent	Activator Ratio
Series 1	Unconfined Compressive Test	5	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0	0.5
		10	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0
		15	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0
		20	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0
		25	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0
		30	0.0, 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0
(b)
Series	Test Type	Eccentricity Ratio (e/B)	Thickness of Treated Layer
Series 2	Loading Tests	0.1	H/B = 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75
		0.2	H/B = 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75
		0.3	H/B = 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75
		0.4	H/B = 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75
		0.5	H/B = 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75

.

The samples of soil–stabilizer agent mixtures were prepared by mixing proportions of cement kiln dust with the dry clay and/or zeolite based on the percentages detailed in Table 2. Then, the alkaline solution and optimum water content were added to the dry clay–stabilizer agent mixtures and mixed until a homogeneous mixture was obtained. Based on previous studies conducted by the authors, the ratio of the activator (K₂SiO₃ and KOH) was 0.5, which was kept constant for all mixtures. The prepared mixture was manually compacted based on the maximum dry density using a hammer. Following compaction, the geopolymer samples underwent a 24 h maintenance period before immersion in water for curing. A duration of seven days was selected as the mean period for the curing process.

2.2.2. Loading Test

The second experiment consisted of performing extensive loading tests according to ASTM D 1196-21 [44] on a large-scale model of a square footing (B = 50 cm) positioned on soil treated with different thicknesses of geopolymer layers. Figure 2h shows the physical model of the loading test. The soil box of the physical model had dimensions of 2.1 m × 2.1 m × 2.1 m.

This section aims to investigate the improvement in the bearing capacity of a footing positioned on soil treated with different thicknesses of geopolymer layers and subjected to eccentric loads. This means that soils treated with different geopolymer thickness ratios (H/B = 0, 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75) and various eccentric loads (e/B = 0.1, 0.2, 0.3, 0.4, and 0.5) were considered. It should be noted that only one percentage of the stabilizer agents was selected to be adopted based on the results of the UCS test.

Laboratory models were prepared by placing soil inside the soil box and compacting it as layers. Then, the top layer was treated with the activated cement kiln dust and zeolite. The soil was used in its natural state. The soil was placed in the soil box and tamped by dividing the box into controlled weight/volume layers to achieve the required natural density. After completing the last layer, the soil was leveled to achieve a flat surface. In each case, the top layer was treated with geopolymer considering different layer thicknesses. After completing the preparation of the treatment layer, the boxes were covered with plastic wrap to prevent moisture loss. The improved layer of soil–geopolymer was allowed to cure for 28 days before being loaded with model footings. After 28 days, the load was applied gradually to the foundation at different eccentricity ratios (e/B = 0.1, 0.2, 0.3, 0.4, and 0.5).

3. Results and Discussion

3.1. Effect of Stabilizer Agents (Cement Kiln Dust and Zeolite) on the Stress–Strain Characterization of Clayey Soil

The stress–strain characteristics of clay treated with 5, 10, 15, 20, 25, and 30% stabilizer agents considering different zeolite ratios (0.1, 0.2, 0.4, 0.6, 0.8, and 1.0) are presented in Figure 3a–f.

The increases in peak stress resulting from increases in the percentage of activated stabilizer agent (cement kiln dust) are plotted in Figure 4. Notably, the increases in peak stress with 5% activated stabilizer agent (cement kiln dust) were minor, while the rate of improvement with 25% activated stabilizer agent (cement kiln dust) was low. For example, the peak stress increased from 124 kPa for untreated clay to 150 kPa for treated soil with 5% activated stabilizer agent, constituting a 20% improvement in the unconfined compressive strength. However, by increasing the activated stabilizer agent to 10, 15, 20, 25, and 30%, the UCS increased to 236, 360, 440, 530, and 572 kPa, respectively. This means that the peak stress increased by 20% with the inclusion of 5% activated cement kiln dust, compared with 327% for soil with 25% cement kiln dust. Moreover, as the cement kiln dust percentage was increased from 25 to 30%, the peak stress increased from 327 to only 361 kPa.

On the other hand, it is clear that the improvement in UCS was not the same for specific stabilizer agent contents, and this is linked mainly to the fact that the ratio of zeolite plays an important role in shearing resistance. Therefore, the results reveal that incorporating zeolite content in the stabilizer agent significantly affects the stress–strain characteristics of clayey soil. This influence can be divided into two ranges—zeolite ratio = 0.1–0.6 and zeolite ratio = 0.8–1.0 (Figure 5). For the zeolite ratio = 0.1–0.6 range, the developed stress–strain relationship changed such that its slope was steeper than that of untreated clay. The developed stress–strain relationships for a different ratio of zeolite/stabilizer agent almost coincided in the first part, and then significant differences occurred, where the UCS stress increased with the incorporation of zeolite.

The peak stress increased significantly for zeolite ratios ≤ 0.6. However, there was a different trend for zeolite ratios ≥ 0.8, where the peak stress decreased sharply for all stabilizer agent percentages. For example, with 5% stabilizer agent, the peak stress increased from 124 kPa for untreated soil to 172, 188, 198, and 214 kPa for geopolymerized clay with zeolite ratios = 0.1, 0.2, 0.4, and 0.6, respectively, constituting a 38–72% improvement in peak stress. However, increasing the zeolite ratio to 0.8 and 1.0 resulted in the peak stress decreasing to 200 and 198 kPa, respectively. For larger stabilizer agents (30%), the variation in the peak stress due to the incorporation of zeolite became clearer. Notably, the variation in peak stress becomes clearer with >10% stabilizer agent. For example, the peak stress for mixtures with 30% stabilizer agent increased to 960, 1560, 2424, and 2530 kPa by increasing the zeolite ratio to 0.1, 0.2, 0.4, and 0.6, respectively; however, the peak stress decreased sharply to 1167 and 800 kPa upon further increasing the zeolite ratio to 0.8 and 1.0, respectively. This behavior is related to the fact that the geopolymer requires free calcium and aluminosilicate, which are provided by the cement kiln dust and zeolite, respectively. Figure 6 compares the UCS obtained in the present study with the results obtained in previous work. It should be mentioned that although extensive studies on geopolymerized soil have been carried out, most of these studies used fly ash, GGBFS, metakaoline, and POFA to produce the geopolymer. To the best of our knowledge, no studies have been conducted on geopolymers produced using cement kiln dust and/or zeolite.

3.2. Effect of Eccentricity on the Bearing Capacity and Settlement of Footings Positioned on Treated Layers

Footings subjected to eccentric loading exhibited different performance results. This type of loading significantly influences the bearing capacity and the associated settlement. Under eccentric loads, footings positioned on untreated clay steadily sunk into the clayey layer, with visible rotation towards the eccentricity part. This effect significantly influences the performance of the footing in terms of rapidly decreasing the bearing pressure and significantly increasing the settlement. This result can be related to the failure of the soil surface being mostly located under the footing and not symmetrically on both sides, which occurs when foundations are subjected to a central load. The variation in the ultimate bearing capacity with an eccentricity ratio of e/B is plotted in Figure 7.

The results of the tests consisting of footings positioned on untreated clayey soil revealed that the bearing capacity decreased drastically from 140.4 kPa under centric loading to 113.2, 83.2, 75.4, 61.1, and 50.5 kPa when the load was subjected to eccentricity ratios e/B of 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75, respectively (Figure 7). These results consisted of reduction ratios of 19.2, 40.7, 46.4, 56.4, and 63.8%, respectively, as shown in Figure 8. Such eccentricity was associated with the visible rotation of the footing, where a larger settlement was observed on the side of the applied load. Figure 9 shows the rotation of the footing that developed during the application of eccentric loading. It was observed that the rotation of the footing was negligible under centric loading, while it developed to 6.0, 15.1, 21.4, 25.6, and 28.3° under eccentricity ratios of e/B of 0.1, 0.15, 0.3, 0.45, 0.6, and 0.75, respectively.

Treating the soil layer beneath the footing is a common and efficient method utilized in many geostructures. This section evaluates the efficiency of using activated stabilizer agents produced using cement kiln dust and zeolite to enhance the strength of shallow soil layers in terms of increasing the bearing capacity and reducing the developed settlement. Under centric load, the treated layer with activated cement kiln dust and zeolite resulted in significant increases in the ultimate bearing capacity for all values of depth ratio, H/B. The enhancement was clear, even with an H/B value of 0.1, where the ultimate bearing capacity increased from 140.4 kPa for untreated clay to 196.5 kPa, constituting a 40.0% increase in the bearing capacity (Figure 8 and Figure 9). Increasing the stabilized soil’s height resulted in greater failure load values and stiffer stress–displacement responses. Consequently, treated soil covers a larger part of the failure zone, which can improve the foundation’s bearing capacity. This improvement increased to 64.1, 72.3, 97.1, 145.3, and 161.2% when the depth ratio, H/B, of the treated layer increased to 0.15, 0.3, 0.45, 0.6, and 0.75, respectively. Notably, the rate of improvement in ultimate bearing capacity beyond H/B = 0.6 was slow.

Despite this noticeable improvement, soil treated with an activated stabilizer agent still exhibits relative dependency on the eccentric loading, as represented by the change in the ultimate bearing capacity corresponding to the variation in the eccentricity ratio, e/B. Footings positioned on the treated soil layers showed very significant increases in peak stress, and increasing the depth of the treated layer resulted in larger increases in the ultimate bearing capacity, while decreases in the variation in strength occurred due to variation in the eccentricity ratio, e/B. For example, under e/B = 0.0, 0.1, 0.2, 0.3, 0.4, and 0.5, footings on untreated soil layers had bearing capacities of 140.4, 113.2, 83.2, 75.4, 61.1, and 50.5 kPa, respectively, while for footings positioned on the treated soil layers with H/B = 0.1, the bearing capacity increased to 196.5, 160.7, 130.4, 104.5, 90.9, and 78.6 kPa, respectively. With H/B = 0.75, the bearing capacity under such eccentricity ratios increased to 366.4, 346.0, 330.6, 320.5, 290.2, and 260.8 kPa, respectively. This means that the variation in the bearing capacities caused by increasing the eccentricity ratio, e/B, from 0 to 0.5 was significant (64.0%) for H/B = 0.0 and decreased significantly to 31% and 28% for H/B = 0.6 and 0.75, respectively. This means that to address the decrease in the bearing capacity of the untreated soil from 140.4 kPa to 50.5 kPa caused by changing the eccentricity ratio from e/B = 0.0 to e/B = 0.5, the treated layer with a thickness of H/B = 0.45 can be used as it provides a bearing capacity (even under e/B = 0.5) close to that of untreated soil under centric loading.

Regarding the developed rotation of the footing, the ultimate pressure of the untreated soil (140.4 kPa) was selected as a reference. This means that the tilt of the footing under eccentric loading was calculated at a similar reference pressure of 140.4 kPa (the ultimate capacity of untreated soil). The results showed that although the soil replacement under the footing with a geopolymerized layer generally enhances the settlement performance of the footing, there is some variation associated with increasing the eccentricity of the applied load.

The footing rotations according to the H/B and e/B values are presented in Figure 10. From this figure, it can be observed that the eccentricity ratio has less influence on the settlement of the treated layer with H/B = 0.45–0.75. For example, footings positioned on untreated soil rotated by 6, 15, 21, 25, and 28° under eccentric loading with e/B values of 0.1, 0.2, 0.3, 0.4, and 0.5, respectively. However, under the largest eccentricity (e/B = 0.5), the rotation of the footing positioned on the treated soil layers with H/B values of 0.45, 0.6, and 0.75 was reduced to 12.3, 11.6, and 8.7°, respectively.

4. Discussion Based on the Microstructures

The results of the experiments revealed that incorporating zeolite in the stabilizer agent significantly increased the shear strength of the soil, i.e., the unconfined compressive strength. Moreover, geopolymer-treated soil beneath the footing enhanced the strength of the shallow soil layers in terms of increasing the bearing capacity, reducing the developed settlement, and reducing the decrease in bearing capacity and the rotation of the footing under eccentric loading. This behavior is related to the enhancement of the microstructures of the soil due to the development of the geopolymer gel between particles.

The microstructure and chemical composition of untreated and treated soils were analyzed using FESEM to better understand the stabilization mechanism and to determine the exact changes in the microstructure of the clay soil–geopolymer-stabilized samples. SEM revealed the enhanced morphology of the cement kiln dust and zeolite-based geopolymer in terms of decreasing voids, increasing the solidarity, and increasing homogeneity, and its microstructure was improved by increasing the zeolite ratio, which indicated the formation of the gel stage and a fibrous structure (Figure 10). The chains formed in the geopolymer improve soil strength. This development of the microstructure of zeolite-based geopolymers was associated with high compressive soil strength. Due to cross-linking, the geopolymer created a harder, denser soil morphology. Compared with cement, the geopolymer gel bond has a longer network. Moreover, the aluminum in the geopolymer improves soil-bearing capacity.

The improvement in the microstructure is attributed to the addition of an alkaline solution to aluminosilicate stabilizer agents, which causes them to dissolve. This process leads to the condensation of soil molecular particles into a gel, which is subsequently activated. Aluminosilicate is reduced by absorbing the water used while blending; thus, aluminate and silicate are produced. The process of alkaline activation involves the dissolution of polymeric silica and alumina, which, in turn, activates the conversion of untreated soil particles. In alkaline solutions, silica can generate a highly concentrated aluminosilicate mixture in an amorphous state with a high pH value. During this process, a saturated solution forms a gel from an oligomer network. Water in the soil matrix contributes to the formation of biphasic gel pores through alkaline activators.

The structure of geopolymers is commonly described as a network of interconnected mineral compounds and chains held together by hydrogen bonding. It is critical to understand the phases involved in the geopolymer creation process. Geopolymers are typically described as the product of a chemical reaction between alkali silicate and hydroxides, which activates aluminosilicate binders with a high pH concentration. The geopolymer chemical process is based on polysilicates and aluminosilicates. By exchanging oxygen atoms and linking the framework tetrahedra to negatively charged particles containing Al₃ ions, the silicate structural network establishes connections between AlO₄ and SiO₄ [50].

Polymeric silicates are formed as chains through the interaction of Al₃ and Si₄ components with a source of oxygen. However, this step is only sometimes employed in the production of polymers. Alternatively, iron oxide-containing polysilicates may develop during geopolymer alkaline activation. Iron is found in the ferrosilicate network in different configurations, such as being incorporated into the main chain, balancing charges at interstitial sites, or forming Fe₃O₄ oxide that is connected by secondary bonds [51]. In geopolymer structures, iron can substitute [Si(OH)₄]¯ and generate [Fe(OH)]₂ instead of [Al(OH)₄]¯. The synthesis mechanism of alkali-activated polymers is characterized by calcium oxide concentration. Geopolymers with low calcium content are largely aluminum silicate, whereas materials rich in calcium (Ca/(Si + Al) > 1) exhibit an alkali-activated response akin to Portland cement, resulting in the creation of geo-gels that are referred to as tobermorite.

This investigation contributes to the formulation of a cement kiln dust–zeolite geopolymer, which promises to be an efficient and enduring stabilizer for improving soils for academic research and engineering applications. It is anticipated that the results of this study will be especially crucial given that soil treatment with sustainable materials has emerged as a potentially effective method for changing and engineering the behavior of soils. Furthermore, it is expected that the test results will provide valuable and enlightening data that contribute to the comprehension of the behavior of stabilized soil and the potential utilization of zeolite-based geopolymers as stabilizers for various soils in a variety of civil applications. By providing essential data regarding the mechanical properties and performance of geopolymer-treated soil, this study’s findings ultimately contribute to the advancement of databases utilized in subsequent investigations about analogous soils. A significant advantage of this new cement kiln dust–zeolite geopolymer material compared with traditional soil–cement mixtures is its environmentally friendly nature, since no CO₂ is generated to obtain the stabilizer agent. The eco-friendliness and cost of geopolymers also add to their attractiveness for use in engineering applications. In other words, the use of geopolymers in civil engineering applications has great potential to benefit our society in terms of reducing demands on natural materials, reducing environmental problems, and conserving energy. Finally, a limitation of this investigation is that a long-term assessment of the geopolymerized soil was not conducted. Moreover, the effects of the molarity of the activator, curing time, and durability were not assessed.

5. Conclusions

The objective of this study was to investigate the benefits of using the byproduct cement kiln dust and a natural zeolite-based geopolymer in soil improvement as an alternative to the conventional binders—cement and lime—in terms of decreasing severe environmental impacts, increasing compressive strength, and improving bearing capacity under eccentric loading. The zeolite-based geopolymer was also evaluated regarding its ability to mitigate the reduction in the bearing capacity that occurs due to eccentric loading. To this end, the selected percentage of the stabilizer agent was used with the activator to treat clay soil layers with different thicknesses (H/B) under a square footing (B = 50 cm) subjected to eccentric loadings of e/B = 0.1, 0.2, 0.3, 0.4, and 0.5.

In light of the analysis results, the following conclusions were drawn:

• Treatment with cement kiln dust-based geopolymers resulted in an improvement in the performance of the clayey soil in terms of increasing the unconfined compressive strength, q_u. This improvement became clearer with >15% stabilizer agent, as the compressive strength of the clay (124.0 kPa) increased by 254, 327, and 360% for cement kiln dust percentages of 20, 25, and 30%, respectively.
• Incorporating zeolite as a partial replacement for cement kiln dust led to a significant increase in the strength of the clay. The compressive strength of geopolymerized clayey soil increased as the zeolite ratio increased, reaching a maximum value at around zeolite/stabilizer agent = 0.6, and then decreased as the zeolite ratio increased to 1.0. For the 30% stabilizer agent, the compressive strength increased to 960, 1560, 2424, and 2530 kPa as the zeolite ratio (zeolite/stabilizer agent) increased to 0.1, 0.2, 0.4, and 0.6, respectively, compared with 124 kPa for untreated soil and 572 kPa for soil treated with only cement kiln dust. Then, the strength decreased to 1167 and 800 kPa as the zeolite ratio increased to 0.8 and 1.0, respectively.
• The ultimate bearing capacity of the untreated clay decreased significantly over the range e/B = 0.1–0.5, in which the reduction ratio reached 64%. However, activated cement kiln dust and zeolite improved the characteristics of clayey soil and reduced the eccentricity effect. The ultimate bearing pressure, when e/B = 0, increased from 140.4 kPa for untreated clay to 366.0 kPa when the geopolymerized layer beneath the footing had H/B = 0.75. This ultimate bearing pressure did not change significantly over the range e/B = 0.1–0.3, in which its reduction did not exceed 12.0%; however, beyond this range of e/B, the reduction in the ultimate bearing capacity reached 28.0% for e/B = 0.5.
• Eccentric loading resulted in a visible tilt of footings positioned on untreated clay, developing to 6, 15, 21, 25, and 28° corresponding to increasing e/B to 0.1, 0.2, 0.3, 0.4, and 0.5, respectively. However, in the case of footings positioned on the treated layer with H/B = 0.45–0.75, the tilt was reduced significantly.