Mechanical Properties and Microscopic Study of Steel Slag–Fly Ash-Solidified Loess under Alkaline Conditions

: To address the geological hazard posed by unstable loess slopes prone to collapse and landslides, a high-strength geopolymer cementing material was developed utilizing green steel slag–fly ash as its primary constituent and activated through the application of sodium silicate alkalinity. The mechanical properties and microstructure changes of loess under varying dosages of steel slag–fly ash geopolymers and curing age were investigated through a series of tests, including unconfined compressive strength, direct shear, disintegration, electron microscope scanning, and X-ray diffraction. The findings indicate that the incorporation of geopolymers can significantly enhance the internal friction angle, cohesion, and unconfined compressive strength of loess, while mitigating the disintegration quantity and rate of stabilized soil. When 20% geopolymer is mixed into the solidified soil and cured for 28 days, the resulting solidified soil exhibits an internal friction angle of 31.12°, a cohesion of 81.09 kPa, and an unconfined compressive strength of 570.86 kPa. These values are 1.62 times, 1.76 times, and 3.36 times higher than those of loess, respectively. Moreover, the solidified soil shows minimal disintegration within 1800 s, with only 1.97% disintegration. The curing age of solidified soil has a significant impact on its curing effect. Enhancing the curing time can considerably enhance the mechanical properties of solidified soil. When the geopolymer content is 20% and the curing time is extended to 28 days, the internal friction angle, cohesion, and uncon-fined compressive strength increase by approximately 0.23 times, 0.48 times, and 1.61 times, respectively, compared to a curing time of 7 days. By analyzing SEM and XRD, it was found that the hydration of steel slag–fly ash geopolymer produces C-S-H and C-A-S-H cementing materials, which effectively fill the gaps between soil particles and enhance the mechanical properties of solidified soil. The research findings can serve as a theoretical foundation for the consolidation of loess sub-grade utilizing steel slag–fly ash geopolymer.


Introduction
China's loess is primarily distributed in the provinces of Shaanxi, Gansu, Ningxia, Qinghai, Shanxi, and Henan. In recent years, there has been a northward shift in the warm and humid line, leading to increased rainfall in the northwest and an extended flood season. Loess is highly susceptible to water and is prone to disintegrate, collapse, and liquefy when exposed to water. The frequency of loess landslide disasters is on the rise, posing a significant threat to the safety of people's lives and property [1,2]. Improving the strength of loess slopes and preventing loess landslides from causing disasters to humans holds significant practical engineering significance.
The consolidation methods for loess can be categorized into two main types: physical reinforcement methods, such as dynamic compaction, replacement, and structural support [3,4]; and chemical reinforcement methods, such as injection of cement, water glass, and other materials [5,6]. However, these methods still have certain limitations. Physical methods such as dynamic compaction can disrupt the soil and potentially compromise the stability of nearby structures [7], while many chemical methods are not environmentally friendly [8,9]. As a result, there is a need to explore green curing technology, which holds promising potential. In light of the worsening national climate, it is crucial to prioritize the use of green, low-carbon, low-energy, and cost-effective engineering materials to address the hazards posed by loess slopes and align with the national green development concept. However, there are still challenges when it comes to managing waste tailings. Steel slag, a bulk solid waste, has a relatively low utilization rate in iron and steel enterprises, with a comprehensive utilization rate of only about 20% in our country. This falls significantly short of the utilization rate exceeding 90% seen in many developed countries. Reports indicate that the accumulated amount of unused steel slag in our country is as high as 1 billion tons [10,11]. Steel slag is a by-product of the steelmaking process, possessing properties such as wear resistance, anti-skid capabilities, and high alkalinity. Due to these qualities, it can be considered a potential option for high-quality aggregates. Consequently, road construction has emerged as a primary application for steel slag. Notably, European countries utilize steel slag in road construction at rates ranging from 39% to 62%, highlighting its significant adoption. In contrast, China lags behind with a mere 8% utilization, indicating a substantial disparity with developed nations [12]. In 1978, French scientist Davidovits introduced geopolymer, a new environmentally friendly gelling material. This material utilizes fly ash and slag as raw materials and has found extensive applications in road repair and other industries [13][14][15]. Long et al. conducted a comparison of the dynamic mechanical properties and eco-economic benefits of ordinary Portland cement (OPC), limestone calcined clay cement (LC 3 ), and alkali-activated slag (AAS). The study revealed that LC 3 , when used in fiber-reinforced mortars (FRMs), exhibited remarkable resistance to dynamic loads and displayed higher potential as green building materials compared to OPC and AAS. Additionally, AAS consumed 2.4% less energy and emitted 20.2% less CO2 during the production of alkali-activated slag concrete (AASC) in comparison to ordinary cement of the same strength grade. These findings indicate that AASC is not only more environmentally friendly but also more economical than OPC of the same strength grade. Furthermore, the inclusion of recycled rubber powder in alkali-activated slag mortar (AASM) was found to reduce the usage of natural aggregate and mitigate environmental pollution [16][17][18]. Previous research has demonstrated that steel slag possesses hydraulic properties and can form a gel-like substance when activated, serving as a mosaic and filling material. Consequently, steel slag can be utilized as a potential raw material for geopolymers [19]. Cui et al. discovered that substituting cement with steel slag can significantly enhance the compressive strength of solidified clay, thereby achieving low-carbon green soil consolidation [20]. Long et al. investigated the use of a fly ashslag geopolymer mortar for recycling waste cathode ray tube glass (CRT). Through their study, they determined that the optimal alkali content and silica powder modulus for geopolymer recycled CRT were 6% and 1.5, respectively [21]. Syahril et al. conducted a study where they utilized volcanic ash and tailings to solidify the road base soil. They discovered that the bearing capacity of the solidified soil, as determined through unconfined compression tests, was higher than that of the undisturbed soil [22]. Cristelo et al. and Liu et al. conducted research on the use of alkali-activated fly ash for solidifying loess and enhancing soft soil. The experiments demonstrated that the alkaline-activated fly ash was successful in enhancing both the strength of the soil and its microstructure [23,24]. Wu et al. observed that the microscopic morphology and physical phase of steel slag powder cement-solidified muddy soil undergo changes with erosion time. They found that C4AF and C2AF in steel slag powder can effectively reduce ion erosion. The lower steel slag powder demonstrates a good solidification effect and promotes early strength development in the solidified soil [25,26]. Deng et al. conducted a study on the use of steel slag base material for the reinforcement of soft soil. They discovered that when the steel slag base material content is 20%, the unconfined compressive strength of the solidified soft soil can reach 1.2 MPa after 28 days [27]. In their experimental study on the compressive strength of fly ash-steel slag geopolymers, Cao et al. discovered that the combination of fly ash and steel slag in preparing geopolymer composite cementitious materials resulted in improved early and later strength, thanks to their synergistic effect [28]. Tong et al. discovered that the vitreous body of fly ash in alkali-activated fly ash geopolymer undergoes a depolymerization-condensation reaction when exposed to alkaline solution, leading to erosion and damage. This reaction results in the formation of N-A-S-H gel, which fills the pores of the sample and enhances the growth of its unconfined compressive strength. The maximum value of 10.3 MPa for the unconfined compressive strength of the sample is achieved when the modulus is 1.1 and the curing age is 28 days [29]. Zhao et al. investigated the use of water glass solution to enhance the activity of fly ash for solidifying loess. The experimental study revealed that a decrease in the modulus of water glass and an increase in the Baume degree resulted in a continuous increase in the compressive strength of the solidified loess. When the molding condition was set at 1.5 and the Baume degree was 30, the compressive strength measured 3.42 MPa [30]. Li et al. discovered that by utilizing steel slag and desulfurized gypsum as soil stabilizers, the ultimate compressive strength of the stabilized soil can be enhanced and the curing period can be reduced when subjected to carbonization curing [31].
Currently, numerous studies utilize single steel slag, fly ash, or cement as the raw material for geopolymer, either individually or in combination. However, the use of these geopolymers as a single raw material or the absence of activators greatly restricts their ability to solidify soil. There is a scarcity of research on the solidification of loess using a mixture of steel slag and fly ash as admixtures, along with water glass as an activator. Fly ash has the ability to eliminate the unstable influence in steel slag. Additionally, water glass can effectively activate the hydration reaction of the combination of steel slag and fly ash, resulting in the formation of a material with gelling properties. This material can fill soil voids and enhance the overall soil structure [32,33]. This paper investigates the use of green and environmentally friendly steel slag-fly ash as raw material to create a highstrength geopolymer cementitious material for solidifying loess through alkali excitation of water glass. The mechanical properties and microstructure changes in loess are studied under different steel slag-fly ash geopolymer content (0%, 5%, 10%, 15%, 20%, 25%) and curing ages (7 d, 14 d, 28 d) using unconfined compressive strength test, direct shear test, disintegration test, scanning electron microscope test, and X-ray diffraction test.

Test Material
The loess sample used in the experiment was obtained from a roadbed slope located in Qilihe District, Lanzhou. The soil has a yellowish-brown color and is classified as silty clay. Prior to the experiment, the soil was crushed, dried, and sieved through a 2 mm sieve. The study conducted an indoor soil test on the test soil, as per the regulations outlined in the Highway Soil Test Regulations (JTG-3430-2020) [34]. The basic physical properties of the test soil are presented in Table 1. The raw materials used in this study include loess, steel slag powder, fly ash, and a water glass solution, as illustrated in Figure 1. The particle gradation of loess, steel slag powder, and fly ash is presented in Figure 2, while their basic physical properties are summarized in Table 2. The mineral composition is depicted in Figure 3. The water glass solution used in this study was obtained from a reputable Wuxi company. The solution had a modulus of 2.31, with a mass fraction of 12.8% Na2O and 29.2% SiO2. The Baume degree of the solution was measured to be 50 Be'/(g·cm −3 ). For adjusting the modulus of the water glass solution during the experiment, NaOH from the laboratory chemical substance analysis pure reagent was used.

Sample Preparation
According to the research conducted by Hou et al. and Jiang et al., the compressive strength of fly ash-based polymer excited by the water glass increases with the decrease in water glass modulus and the increase in Baume degree. The optimal condition for the test is achieved when the mass ratio of steel slag to pulverized coal ash is 1:2 and the dosing of water glass solution is 20% of the total mass of the geopolymer [35,36]. Therefore, for the test, the required mass ratio of fly ash to steel slag in the geopolymer is 2:1, the dosage of water glass solution is 20% of the geopolymer mass, and the water glass solution has a modulus of 1.2 and a Baume degree of 30 Be'/(g·cm −3 ). The blending amount of geopolymer refers to the ratio of the total mass of the sample: 0%, 5%, 10%, 15%, 20%, and 25%. Figure 4 illustrates the compaction curves of the stabilized soils with different geopolymer contents, and Table 3 provides the maximum dry density and optimal water content. The sample preparation follows the dry mixing method. Initially, the geopolymer is mixed with the dry soil and stirred thoroughly. Then, a mixture of water glass solution and distilled water is evenly sprayed onto the mixture using a watering can. After completing the sample preparation, it should be placed in a polyethylene plastic bag and sealed for 24 h. The preparation process involves using the two-way static compaction method to create two different sizes of samples: Φ 5 cm × H 5 cm and Φ 6.18 cm × H 2 cm. The moisture content is maintained at 17% and the compaction degree is 93%. Once the preparation is finished, the sample should be immediately sealed with multiple layers of plastic wrap. Finally, the sample is placed in a constant-temperature and -humidity chamber with the temperature set to 20 ± 2 °C and relative humidity of 95%.

Test Scheme
At different ages (7 d, 14 d, 28 d), six types of solidified soil with varying geopolymer content (0%, 5%, 10%, 15%, 20%, 25%) were prepared. The theoretical density is presented in Table 4. The specimens included two different sizes: Φ 5 cm × H 5 cm and Φ 6.18 cm × H 2 cm. Table 5 provides a list of all the test groups. For the unconfined compressive strength test and disintegration test, each group was comprised of two test pieces, and the average value was recorded. If the error exceeded 3%, the test was repeated. Samples of Φ6.18 cm × H 2 cm were prepared with varying polymer content (0%, 5%, 10%, 15%, 20%, 25%) and different ages (7 d, 14 d, 28 d). A Z-J strain-controlled direct shear instrument was used to conduct the direct shear test. The test involved applying vertical pressures of 100 kPa, 200 kPa, 300 kPa, and 400 kPa and a shear rate of 0.08 mm/min. The shear strength of the solidified soil was measured under different curing ages and polymer content, as shown in Figure 5. By analyzing the relationship between the internal friction angle, cohesion, curing period, and amount of polymer added, relevant results were obtained.

Press sample Demold
Maintain Direct shear test

Unconfined Compressive Strength Test
The unconfined compressive strength test was conducted using the YTW-2 strain type unconfined pressure instrument, as depicted in Figure 6: Φ 5 cm × H 5 cm test samples were prepared at different geopolymer contents (0%, 5%, 10%, 15%, 20%, 25%) and at each age (7 days, 14 days, 28 days). After applying a thin layer of Vaseline to both ends of the sample, it is positioned in the center of the pressure plate. The handle should be rotated clockwise to bring the sample into contact with the upper pressure plate, and then returned to zero to adjust the scale. The motor then moves the pressurized plate at a speed of 2.4 mm per minute until the soil sample is damaged, and the axial stress is determined based on the reading on the dial indicator. The study focused on analyzing the relationship between the compressive strength of the solidified soil and the curing period, as well as the steel slag-fly ash geopolymer.

Disintegration Test
In order to determine the disintegration speed of solidified soil in water, the SHY-1 was used for this test. To ensure consistent sample production, a sample of Φ 5 cm × H 5 cm was placed in the center of the stencil. A mesh plate was hung under the buoy, and the sample was quickly immersed into the water cylinder by holding the neck of the buoy. Electronic equipment was used to record the disintegration of the solidified soil. Data were recorded 1 min before the start of the test, every 10 s, and then every 1 min thereafter. The time interval of recording can be adjusted based on the disintegration speed of the solidified soil. The test is considered complete when the sample falls completely through the grid, or when the sample does not collapse significantly for a long time and the buoy scale does not change significantly within the 30 min time interval. The situation of the sample on the grid and in the water should be described and recorded, and real-time photos should be taken. The trial is terminated at this point [37]. The disintegration calculation can be seen in Formula (1).
In the formula, t A represents the amount of disintegration at time t in percentage, 0 R represents the scale reading of the buoy at the water surface at the beginning of the test, t R represents the scale reading of the buoy at the water surface at time t, and e R represents the scale reading of the buoy at the water surface at the end of the test. The disintegration rate, expressed in min −1 , indicates the speed of specimen disintegration over time. It serves to reflect the rate of disintegration transformation and, consequently, the erodibility of the soil. The disintegration rate calculation can be seen in Formula (2).
In the formula, t v represents the average disintegration rate of the test piece during a specific time period from ti to ti+1, measured in min −1 , 1 i t A + represents the amount of disintegration at time ti+1, i t A represents the amount of disintegration at time ti.

XRD and SEM Tests
X-ray diffraction (XRD) is a valuable method for qualitative phase analysis. X-rays can penetrate thick materials and their short wavelength allows for the examination of crystal's periodic structure. By utilizing diffraction effects, precise data on the crystal structure can be obtained. Scanning electron microscopy (SEM) is another important tool in experimental analysis. In SEM, a processed sample is placed under a microscope and bombarded with a high-speed electron beam. The detector collects the emission information of secondary electrons for further processing. The surface morphology of the sample can be observed on a fluorescent screen and photographs can be taken [38,39]. The test piece was dried at 60 ℃ in an oven. After the treatment, select a piece from the center of the test piece and cut it into a cube with a diameter of less than 1 cm using a knife. Apply a spray of gold onto the cube. Additionally, collect the middle fragments of the samples and pulverize them into powders smaller than 75 μm. Perform an XRD scanning test on the powders at an angle 2θ ranging from 5° to 90°. During the SEM test, it can be challenging for the conductive adhesive to adhere due to the accumulation of dust on the surface of the test piece. Therefore, prior to the test, it is necessary to securely fix the test piece onto a relatively flat surface using resin, and then proceed with scanning it using an electron microscope. Table 6 demonstrates that the internal friction angle and cohesion of consolidated soil are significantly influenced by different geopolymer content and curing age. The utilization of geopolymers for stabilizing loess can enhance its shear strength, particularly cohesion. At the age of 28 days, the lower solidified soil mixed with 20% geopolymer exhibits a maximum cohesion of 81.09 kPa, representing a 75.32% increase compared to plain soil. The lower solidified soil mixed with 25% geopolymer demonstrates a maximum internal friction angle of 31.20°, which is 64.04% higher than that of plain soil.  Figure 7 illustrates the trend of the internal friction angle and cohesion of the stabilized soil with the inclusion of polymer. The graph demonstrates that as the amount of polymer added increases, the internal friction angle initially rises and subsequently stabilizes. Similarly, the cohesive force exhibits an initial increase followed by a decrease. The addition of steel slag fine powder with rich edges and corners in the solidified soil enhances the interaction force between soil particles. As the steel slag content increases, the contact area between steel slag and loess also increases, leading to an increase in the internal friction angle of solidified loess. Additionally, the combination of steel slag and fly ash, activated by alkaline water glass, produces C-S-H and C-A-S-H gels that fill the gaps between soil particles and strengthen the connections between them, thereby improving the soil's cohesion. However, when the addition of geopolymer exceeds a certain threshold, it leads to a decrease in the proportion of net loess content. This results in a reduction in loess skeleton and the aggregation of cementitious materials, forming a lump. However, this is insufficient to effectively fill the voids between soil particles; it merely adheres to the surface of the soil mass. Consequently, while the internal friction angle remains stable, the cohesion decreases. When the local polymer content remains constant, the internal friction angle and cohesion increase with longer curing time. For instance, with a geopolymer content of 20%, the internal friction angle and cohesion at 28 days of curing time are 1.23 times and 1.48 times higher than at 7 days of curing time, respectively. Similarly, at 14 days of curing time, the internal friction angle and cohesion are 1.15 times and 1.36 times higher. When the geopolymer content is 5% and the curing time is 28 days, the internal friction angle and cohesion are 1.07 times and 1.47 times higher than at 7 days of curing time, and 1.03 times and 1.44 times higher than at 14 days of curing time. The results of the direct shear test indicate that both curing time and geopolymer content significantly influence cohesion, while the internal friction angle is primarily affected by the geopolymer content, with little impact from curing time.

Analysis of Unconfined Compressive Strength Results
The relationship between the amount of geopolymer incorporated and the unconfined compressive strength in the solidified soil is illustrated in Figure 8. It is evident that the compressive strength of the solidified soil with different amounts of geopolymers added during different curing periods is higher compared to that of the plain soil. The gelatinous substances C-S-H and C-A-S-H, formed through the hydration reaction of geopolymers, have a cementing effect on solidified soil. They fill the gaps between soil particles, resulting in improved soil stability and strength. During the initial stage, the hydration reaction of the geopolymer is not fully developed, resulting in a gradual increase in solidification strength. However, as the curing age progresses, the geopolymer continues to undergo a hydration reaction, leading to a further enhancement in the compressive strength of the solidified soil. The compressive strength of the consolidated soil is highest at a curing age of 28 days and a geopolymer content of 20%, reaching 570.86 kPa. Compared to the plain soil and the stabilized soil mixed with 25% geopolymer at the same age, the strength increased by 400.82 kPa and 28.34 kPa, respectively. The compressive strength of the solidified soil initially increases and then decreases with the addition of geopolymers at the same age. This is because initially, the amount of geopolymer added is small and the gelled substances produced by the hydration reaction of geopolymer are relatively small. If the amount is insufficient, the particle gap filling is inadequate, and the strength of the solidified soil increases as the amount of geopolymer added increases. However, the strength of the solidified soil decreases when the amount of geopolymer added is 25% compared to when the addition amount is 20%. This decrease is attributed to the increase in mass proportion of geopolymer in the solidified soil, which leads to a decrease in the net content proportion of loess. Consequently, the loess skeleton is reduced. Additionally, the gelled material, which forms clumps and covers unreacted steel slag and fly ash, is unable to effectively fill the voids between the soil particles and adhere to the soil surface. As a result, the overall strength of the solidified soil is diminished. The cured compressive strength increases with the extension of the curing age and reaches its maximum value at 28 days. When the geopolymer content is 25%, the compressive strength of the solidified soil is 2.12 times that of the 7-day curing period and 1.73 times that of the 14-day curing period at a curing age of 28 days. Similarly, the unconfined compressive strength of the solidified soil at a curing age of 28 days is 2.11 times that of the 7day curing age and 1.6 times that of the 14-day curing age. These test results demonstrate that both the curing age and the amount of geopolymer significantly affect the strength of the consolidated soil. During the 28-day curing period, Li et al. discovered that when recycled fine powder (RFP) was added at a rate of 15%, the solidified loess exhibited the highest compressive strength, reaching 357.7 kPa [40]. Wang et al. conducted a reinforcement test using an F1 ion-curing agent and determined that the maximum unconfined compressive strength of the tested loess was achieved when the dosage of the F1 curing agent was 0.3 L/m 3 , reaching 280 kPa [41]. Tian et al. utilized a wood calcium-source EICP solution to solidify roadbed loess and observed an increase in unconfined compressive strength to 232 kPa [42]. Wu et al. employed a 20 be composite modified water glass for loess reinforcement, resulting in an increase in unconfined compressive strength to 287 kPa [43]. Compared to recycled fine powder (RFP), F1 ion-curing agent, wood calcium-source EICP solution, and composite modified water glass, geopolymer-solidified loess offers several advantages. Firstly, the material is easily accessible and cost-effective, meeting the requirements of green emission reduction. Moreover, the solidification effect of geopolymers on loess is remarkably significant. This study found that when the geopolymer addition is 20%, the consolidated soil's unconfined compressive strength reaches its peak value at 570.86 kPa. This strength is 1.59 times, 2.03 times, and 1.98 times higher than that of loess solidified by F1 ion-curing agent, wood calcium source EICP solution, and composite modified water glass, respectively.

Analysis of Disintegration Test Results
The change in disintegration form is the most noticeable indication of the disintegration of solidified soil. As shown in Figure 9, the remolded loess disintegrates after 28 days of curing, which can be roughly divided into three stages. The first stage is the water absorption and exhaust stage. During this stage, air trapped in the particle pores is released into the water, resulting in the formation of pore water. The most noticeable manifestation of this stage is the upward escape of bubbles on the outer surface of the specimen. The second stage is the disintegration stage. The water on the specimen's surface becomes turbid, and due to concentrated stress, spalling initially occurs at the edges of the soil. As the moisture content gradually increases and the soil becomes saturated, the internal friction angle and cohesion decrease, leading to widespread disintegration of the soil from the specimen's surface to its interior, causing large-scale peeling. Once the large-scale disintegration ends and the soil reaches an equilibrium state, the disintegration enters the third stage. During this stage, the disintegration rate decreases until it stabilizes, indicating the completion of the disintegration process. Figure 9. Disintegration phenomenon of primary soil at the 28th day of maintenance. Figure 10 presents the disintegration test curves of loess and solidified soil with varying geopolymer content at different curing ages. The figure demonstrates that the curing period has a notable influence on the disintegration of loess after solidification with geopolymer. On the 7th day of curing, the geopolymer content had minimal impact on the disintegration of the solidified soil. At 726 s, the remodeled loess had completely disintegrated. The disintegration of the solidified soil was the lowest (86.73%) when the geopolymer content was 25%. As the curing age increased, the addition of geopolymers effectively reduced the disintegration of the solidified soil. The smallest disintegration of the solidified soil (36.02%) was observed at a curing time of 14 days and 1800 s with a 20% geopolymer content. This value was 0.362 and 0.782 times the disintegration of plain soil and 20% geopolymer-solidified soil, respectively. When the curing period reached 28 days, the solidified soil with 20% and 25% geopolymer content exhibited minimal collapse, with disintegration of 1.97% and 4.72%, respectively, at 1800 s. The degree of hydration reaction of the geopolymer increases with the curing age. After 28 days of curing, the gelatinous substance produced by the geopolymer's hydration reaction tightly cements the soil particles, preventing large-scale disintegration in a short period of time. However, when a small amount of geopolymer is added, the solidified soil may partially disintegrate because the cementing material cannot fully cement the soil particles. The disintegration of the solidified soil with 25% geopolymer content is higher than that with 20% geopolymer content. This is because the gelatinous substance resulting from the hydration reaction of the geopolymer partially adheres to the specimen's surface and dissolves in water, leading to the phenomenon of "disintegration and rise" in the solidified soil with 25% geopolymer content.   Figure 11 shows the disintegration rate curve of loess and solidified soil with varying content of geopolymers at different curing ages. The disintegration rate of solidified soil initially increases and then decreases, eventually stabilizing at the 7th day of curing. This trend is less influenced by the amount of geopolymer used. The disintegration rate of solidified soil with varying geopolymer content ranged from 3.74 min −1 to 4.16 min −1 within the first minute. The disintegration rate of solidified soil reached its peak at approximately the eighth minute, with a maximum value of 11.26 min −1 . As the geopolymer content increased, the disintegration rate of stabilized soil initially decreased and then stabilized. After 14 days of curing, the average disintegration rate during the first 1800 s was 2.12 min −1 . This was a decrease of 3.44 min −1 and 0.38 min −1 compared to remolded loess and solidified soil with 25% geopolymer content, respectively. The disintegration rate of solidified soil decreased by varying degrees with increasing curing age. On the 7th day of curing, the average disintegration rates for remolded loess and 20% geopolymer were 7.69 min −1 and 4.06 min −1 , respectively. On the 28th day of curing, the average disintegration rates for remolded loess and 20% geopolymer were 4.28 min −1 and 0.065 min −1 , respectively. When the geopolymer content was 20%, the solidified soil showed minimal disintegration and the disintegration rate fluctuated only slightly. Only a small amount of disintegration occurred in the surface soil of the specimen when exposed to water. Please refer to Figure 12 for the disintegration phenomenon of solidified soil with 20% geopolymer content after 300 min. The disintegration rate of solidified soil is influenced by the internal friction angle, cohesion, and overall strength of the specimen. Longer curing time enhances the cementation effect of the geopolymer, thereby improving the overall stability of the specimen.   Figure 13 shows the XRD pattern of remolded loess in a subgrade slope in Qilihe District, Lanzhou. The main mineral composition is quartz (SiO2), In addition, there are albionite (NaAlSi3O8), KAlSi3O8, calcite (CaCO3), dolomite (CaMg(CO3)2), muscovite (KAl3Si3O10(OH)2), kaolinite (Al2Si2O5(OH)4), and chlorite ((Mg, Al, Fe)6(Si, Al)4O10(OH)8), and other crystal phases exist. Figure 14 displays the XRD patterns of the solidified soil with various polymers. Figure 13 shows that the primary component of the solidified soil remains quartz (SiO2) after the addition of geopolymers, which is consistent with the diffraction pattern of reshaped loess. However, there is a decrease in the diffraction intensity of minerals, particularly quartz (SiO2), albite (NaAlSi3O8), and calcite (CaCO3), which can be attributed to the increased proportion of geopolymers. Additionally, the presence of dense and low amorphous peaks suggests that the geopolymer has undergone hydration reactions and formed new substances. Under the stimulation of a sodium silicate alkaline agent, steel slag powder releases a significant amount of Ca 2+ ions. These ions then combine with the silicon (aluminum) oxygen tetrahedron formed through the hydration of fly ash, resulting in the formation of calcium silicate hydrate (C-S-H) and calcium aluminosilicate hydrate (C-A-S-H) gels in an amorphous state. The specific reaction formulae are shown in (3) and (4)

XRD X-ray Diffraction Analysis
Over time, the gelling substance continues to increase. This is especially evident when water glass is used as an alkaline activator, as it accelerates the hydration reaction of steel slag micropowder. This reaction promotes the formation of C-S-H and C-A-S-H at a faster rate. The gelatinous substances formed continuously fill the pores between the loess particles, resulting in a more compact specimen. Ultimately, this improves the overall stability and compressive strength of the solidified soil.

SEM Analysis
The microstructure of solidified soil specimens with different geopolymer content (0%, 5%, 10%, 15%, 20%, 25%) after curing for 14 d and 28 d is depicted in Figure 15 and Figure 16. The magnification used for these photos is 1000 times. Analyzing the microstructure of solidified soil provides qualitative and quantitative insights into the effectiveness of soil solidification. The variation in soil structure significantly impacts its mechanical properties. Comparing the microstructure of polymer-solidified soil allows for a direct evaluation of the solidification effect of loess under different cycles and polymer content. Figure 15a depicts the microstructure of the remolded loess after a 14-day maintenance period. The remolded soil primarily consists of small particles, which are randomly arranged and in contact with each other. The gaps between the soil particles are irregularly distributed and lack a specific shape. Upon the addition of geopolymers, the hydration products C-S-H and C-A-S-H fill some of the pores between the soil particles. The floc in the red box represents its hydration product. Additionally, unreacted fly ash glass spherical particles are randomly dispersed within the pores of the soil particles, as indicated by the black circles. As the geopolymer content increases, the amount of hydrated flocculated gelatinous substances also increases, leading to a continuous reduction in the gap between the soil particles. When the geopolymer addition reaches 25%, the experimental results reveal the presence of numerous unhydrated fine fly ash glass ball particles on the surface of the solidified soil. This occurrence can be attributed to the short curing time, where some steel slag-fly ash geopolymers have not fully undergone hydration.  Figure 16 presents a microstructure diagram of solidified soil mixed with different geopolymers after 28 days of curing. The figure shows that compared to the 14-day maintenance period, the structure of the solidified soil becomes more compact, indicating better hydration of the geopolymer. Additionally, the remaining amount of geopolymer decreases as the maintenance period increases. This is because during the 14-day curing age, most of the silicon-aluminum oxides in the geopolymer raw materials dissolve and the formed hydration gel fills the pores. However, a portion of the geopolymer remains unhydrated, resulting in less full soil particle gaps in the microstructure compared to the 28-day period. On the other hand, during the 28-day curing period, the silicon-aluminum oxides hydrate to a greater extent, and the fly ash balls are enveloped by the generated C-S-H and C-A-S-H gelling substances. This significantly fills the pores, leading to a more compact microstructure and improved mechanical properties. SEM observation confirms that the steel slag-fly ash geopolymer test's macroscopic properties mainly derive from the steel slag-fly ash substance in the geopolymer. Under the alkali excitation of water glass, a depolymerization-condensation reaction occurs, forming an amorphous phase C-S-H and C-A-S-H flocculent gel that combines with the vitreous body in the geopolymer. This flocculent gel material strengthens the binding force, enhancing the compactness and compressive strength of the solidified soil.

Conclusions
This study aimed to investigate the feasibility of solidifying loess by using steel slagfly ash as the main raw material in an alkaline environment of water glass for hydration to form gelled material. The study included various tests, such as unconfined compressive strength, direct shear, compaction, disintegration, and XRD and SEM analysis. The effects of different geopolymer additions and curing ages on the solidified soil are considered. Based on the test results, the following conclusions can be drawn.
(1) The addition of fly ash-steel slag geopolymer as an admixture to loess has been found to effectively enhance the mechanical properties of the solidified soil and improve its internal structure. After a curing period of 28 days, the largest cohesion value of the solidified soil, reaching 81.09 kPa, is achieved with the addition of 20% geopolymer, which is 75.32% higher than that of plain soil. The internal friction angle is also increased to 31.12°, surpassing that of plain soil by 62.08%. Furthermore, the unconfined compressive strength of the consolidated soil is enhanced to 570.86 kPa, which is 236% higher than that of the plain soil. Incorporating 20% geopolymer into the stabilized soil not only effectively improves its internal friction angle and cohesion but also aligns with the principles of green environmental protection and economic rationality.
(2) Compared to remolded loess, the hydration of geopolymer becomes better with an increase in curing age. This leads to an effective reduction in the disintegration of solidified soil. For instance, when the geopolymer content is 20%, the solidified soil shows minimal disintegration in water even after 300 min. In scenarios involving rainfall and water accumulation, the addition of geopolymers can effectively decrease the likelihood of loess collapse, thereby reducing the occurrence of engineering accidents.
(3) The analysis of XRD composition and SEM structure reveals that under the alkaline activation condition of a water glass solution, steel slag and fly ash exhibit synergistic effects. Steel slag powder releases a significant amount of Ca 2+ ions, which then combine with the silicon (aluminum) oxygen tetrahedron formed by the hydration process in fly ash. This combination results in the formation of amorphous gels such as calcium silicate hydrate (C-S-H) and calcium aluminosilicate hydrate (C-A-S-H), which effectively fill the pores in loess particles. As the curing time increases, the hydration products of geopolymers become more abundant, and the gelled substances of C-S-H and C-A-S-H are more plentiful. Consequently, the geopolymer vitreous body becomes enveloped by C-S-H and C-A-S-H gelling substances, effectively filling the pores and creating a more compact microstructure. This ultimately enhances the mechanical properties of the solidified soil. Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.