Mechanical properties and micro-mechanisms of soft soil stabilized with rice husk ash and multi-source solid waste-based cementitious materials

To address the issues of high energy consumption, significant carbon emissions, and suboptimal effectiveness associated with conventional cement-stabilized soft soil, this study proposes a novel binder composed of rice husk ash (RHA) and multi-source solid wastes for soft soil stabilization. Unconfined compressive strength (UCS) tests and scanning electron microscopy (SEM) were conducted to investigate the mechanical properties and micro-mechanisms of this composite stabilization system. The results indicate that: (1) For the composite stabilized soil across all curing ages, the order of significance of the factors influencing strength is RHA content > carbide slag content > cement-to-ground granulated blast furnace slag (GGBS) ratio. The optimal binder composition, based on UCS evaluation, is RHA:cement:GGBS:carbide slag = 3:6.4:9.6:1 when the mass ratio of composite binder to dry soil is 20%. (2) Compared to cement-stabilized soil with the same binder content, the optimal composite stabilized soil exhibited 24% and 39% higher UCS at 14 and 28 days, respectively. The stress-strain curves shifted rightward, with increased ultimate strain and enhanced toughness. (3) SEM analysis revealed significantly more dense honeycomb and network structures in the composite stabilized soil compared to cement-stabilized soil. Hydration-generated calcium silicate hydrate (C-S-H) gels connected and filled the pores, improving soil density and strength.

1 Introduction

The Yangtze River Delta region is characterized by extensive deposits of deep soft soils, which exhibit high water content and low strength, posing significant challenges to construction projects (He et al., 2020). Consequently, there is considerable interest in developing methods to stabilize and improve these soft soils to enhance their strength and stability.

Lime and cement are traditional materials widely used in soil stabilization (Jongpradist et al., 2011). However, these conventional binders often perform poorly when stabilizing soft soils with high water content, sometimes even failing to achieve effective stabilization in practice (Zhou et al., 2021). Furthermore, the production of these binders is associated with high energy consumption, substantial carbon emissions, and severe environmental pollution. For instance, producing one ton of cement emits 0.7–1.1 tons of CO₂. Given the escalating concerns over global warming, the continued use of such energy-intensive, high-carbon binders is unsustainable. There is an urgent need to develop novel, green, low-carbon binders for soil stabilization.

Recent research has focused on utilizing bulk industrial wastes such as slag, fly ash, carbide slag, and phosphogypsum, either individually or in combination, for soft soil stabilization. This approach not only achieves soil stabilization but also promotes the resource utilization of these wastes, reducing carbon emissions (Feng et al., 2024; Xue et al., 2024). Oormila and Preethi (2014) incorporated granulated blast furnace slag into soft soil to enhance its strength. Zhang et al. (2020) highlighted that slag contributes to strength improvement in stabilized dredged sludge by promoting the formation of hydration gels. Further studies have explored the synergistic use of a single industrial waste with cement. For example, Shi et al. (2022) used steel slag combined with cement for dredged sludge stabilization, while Zhu et al. (2022), Zeng et al. (2021), and Peng and Chen (2021) employed carbide slag, phosphogypsum, and blast furnace slag combined with cement, respectively. Ding et al. (2019) used a dual blend of cement and phosphogypsum to stabilize dredged sludge with high water content. Zhang et al. (2020) utilized cement-activated GGBS for sludge stabilization. Further advancing this line of inquiry, researchers have investigated the effectiveness of multiple industrial wastes combined with cement, or multiple industrial wastes used synergistically without cement, for soft soil stabilization. For instance, Liu et al. (2019) used a composite of cement, fly ash, and slag to stabilize river dredged sludge, finding that the shear strength of the stabilized soil initially increased and then decreased with increasing fly ash and slag content. Yi et al. (2016) employed slag combined with reactive magnesia for soil stabilization and explained the underlying mechanism. Phummiphan et al. (2018) studied the pavement performance of lateritic soil stabilized with high-calcium fly ash and granulated blast furnace slag. Wu et al. (2023) prepared a composite binder using carbide slag, granulated blast furnace slag, and fly ash, achieving UCS values for the stabilized soil that were 1.38–2.3 times those of cement-stabilized soil.

In addition to the aforementioned bulk industrial wastes, China produces approximately 3 million tons of rice husk ash (RHA) annually from biomass power generation, leading to increasing challenges related to its stockpiling and disposal. Due to its excellent pozzolanic activity, RHA has garnered significant attention for modifying cement-based materials. Studies have shown that incorporating RHA can significantly enhance the frost resistance (Zhang et al., 2023), impermeability (Kang et al., 2019), sulfate attack resistance (Chindaprasirt et al., 2007) and high-temperature resistance (Gencel et al., 2021) of concrete materials. However, research on the synergistic stabilization of soft soil using RHA in combination with multiple solid wastes remains limited.

To address this gap, this study proposes a composite binder composed of RHA, GGBS, cement, and carbide slag for soft soil stabilization. UCS tests and SEM analysis were conducted to investigate the strength improvement effect and micro-mechanisms of this novel composite binder.

2 Materials and methods

2.1 Materials

The soil used for testing is grayish-black in color and in a fluid-plastic state. Its fundamental physical properties are summarized in Table 1.

Table 1

Table 1. Basic physical properties of the soil.

The raw materials used for preparing the stabilizing binder included: rice husk ash (RHA), a solid waste obtained from low-temperature combustion (approximately 600 °C) in biomass power plants, presenting as a black powder; Grade 42.5 ordinary Portland cement, a grey powder; S95 grade ground granulated blast furnace slag (GGBS), a white powder; and carbide slag, a grey powder. The chemical compositions and respective percentages of these binder materials are presented in Table 2.

Table 2

Table 2. Chemical composition of the binder components (wt%).

2.2 Experimental program

This study aims to investigate the effectiveness of a composite binder comprising RHA, GGBS, carbide slag, and cement for soft soil stabilization, and to systematically analyze the mechanical properties and microstructural characteristics of the stabilized soil. An orthogonal experimental design was employed to optimize the binder composition. The specific scheme involved selecting three factors: RHA content, the cement-to-GGBS ratio, and carbide slag content. Each factor was designed with four levels, as detailed in Table 3. An L¹⁶(4⁵) orthogonal array was utilized. Without considering interactions between factors, columns 4 and 5 were assigned as error columns. The detailed orthogonal test plan is presented in Table 4. A control group was established using dredged soil stabilized with 20% cement content by weight. The testing protocol involved measuring the UCS of the stabilized soil with a total binder content of 20% at curing ages of 7, 14, and 28 days. Additionally, SEM was used to observe the microstructural morphology of the specimens. It should be pointed out that in all the following experiments, the mass ratio of the curing agent to dry soil is consistently 20%.

Table 3

Table 3. Factors and levels for the orthogonal test design of soil stabilization mixtures.

Table 4

Table 4. Orthogonal test scheme for soil stabilization mixtures.

2.3 Test methods

2.3.1 Specimen preparation and curing

Specimen preparation was conducted in accordance with the Chinese standard “Standard for Geotechnical Testing Method” (Ministry of Water Resources of the People’s Republic of China, 2019 GB/T 50,123–2019). The detailed procedure was as follows: First, the natural soil sample was dried, pulverized, and passed through a 2 mm standard sieve for subsequent use. Subsequently, a predetermined amount of the dried soil was mixed with the required quantity of water to achieve a water content of 50%. This mixture was stirred thoroughly for 5 min. Then, the binder was added uniformly at the specified dosage (all dosages are expressed as the mass percentage of each binder component to the dry soil mass) and mixed for 5–8 min to ensure homogeneity.

Following this, cylindrical PVC molds with dimensions of Φ100 mm × 50 mm were used. A small amount of Vaseline was applied to the inner walls of the molds. The mixed soil was then placed into the molds in small increments. Each time after adding soil, the mold was placed on a vibrating table and vibrated for 1–2 min. This process of adding soil and vibrating was repeated until the mold was completely filled, after which the surface was leveled. Finally, the prepared specimens were sealed and stored in a constant temperature and humidity curing chamber maintained at (20 ± 2) °C and a relative humidity of (95 ± 2)%. After 1 day of curing, the molds were removed, and the specimens continued to be stored under sealed conditions until testing.

2.3.2 UCS test

The UCS tests were performed using a strain-controlled unconfined compression apparatus. The specimens were loaded at a constant axial strain rate of 2 mm/min until failure. For each mix proportion and at each curing age, three replicate specimens were tested, and the average value was reported as the UCS.

2.3.3 SEM test

SEM observations were carried out using a German ZEISS Sigma 360 field emission scanning electron microscope. After the UCS tests, fresh fragments were taken from the middle section of the failed specimens. Small, regular-shaped fragments with a volume of approximately 0.5 cm³ were selected as samples for microstructural observation. These samples underwent vacuum freeze-drying, were stored in sealed containers in a desiccator, and finally, their surfaces were sputter-coated with a thin layer of gold prior to SEM examination.

3 Results and analysis

3.1 Analysis of unconfined compressive strength

The variation of UCS with curing age for all stabilized soil mixtures is presented in Table 5. As shown in the table, at the 7-day curing mark, the UCS of the control group (cement-stabilized soil) reached 1,125 kPa. In contrast, the maximum UCS achieved by the composite binder-stabilized soil was 592.8 kPa, significantly lower than that of the cement-stabilized soil. This is attributed to the relatively slower hydration reaction rate of the solid waste-based composite binder during the initial stage. As the curing time increased, the UCS of all mixtures gradually increased. The UCS of the cement-stabilized soil exhibited slow growth, indicating that the initial hydration process was largely complete. In stark contrast, the UCS of several composite binder-stabilized soils (specifically Nos. 1, 2, and 6) increased rapidly. Notably, the 14-day and 28-day UCS values of these Nos. surpassed those of the cement-stabilized soil, demonstrating the significant effectiveness of the solid waste-based composite binder in soft soil stabilization. Furthermore, the maximum UCS obtained from the orthogonal tests was 1956.6 kPa, corresponding to No. 6. This value is 1.39 times the strength of the cement-stabilized soil. The optimal binder composition for No. Six was determined to be RHA: Cement: GGBS: Carbide Slag = 3: 6.4: 9.6: 1.

Table 5

Table 5. Unconfined compressive strength of stabilized soils.

3.2 Range analysis

3.2.1 14-day curing period

The range analysis results for the 14-day strength are presented in Table 6. As shown in Table 6, for the 14-day UCS of the stabilized soil, the order of influence of the factors is: RHA content (X) > carbide slag content (Z) > cement-to-GGBS ratio (Y). Under standard 14-day curing conditions, the degree of influence of RHA content on the strength of the stabilized soil is greater than that of the cement-to-GGBS ratio.

Table 6

Table 6. Range analysis results (14-day).

The influence of component content on the 14-day UCS of the stabilized soil is shown in Figure 1. It can be observed from Figure 1 that the UCS of the stabilized soil initially increases and then decreases significantly with increasing RHA content. It first increases, then decreases, and increases again with the increasing cement-to-GGBS ratio. A significant decrease is observed with increasing carbide slag content. The optimal contents for RHA, the cement-to-GGBS ratio, and carbide slag are X2, Y2, and Z1, respectively. Therefore, based on the range analysis results of the 14-day UCS of the stabilized soil, the optimal mix proportion for the 14-day curing age is determined to be X2Y2Z1.

Figure 1

Figure 1. Influence of component content on the 14-day UCS of stabilized soil.

3.2.2 7-day curing period

The range analysis results for the 7-day strength are presented in Table 7. As shown in Table 7, for the 7-day UCS of the stabilized soil, the order of influence of the factors is: RHA content (X) > carbide slag content (Z) > cement-to-GGBS ratio (Y). Under standard 7-day curing conditions, the degree of influence of RHA content on the strength of the stabilized soil is greater than that of the cement-to-GGBS ratio.

Table 7

Table 7. Range analysis results (7-day).

The influence of component content on the 7-day UCS of the stabilized soil is shown in Figure 2. It can be observed from Figure 2 that the UCS of the stabilized soil initially increases and then decreases significantly with increasing RHA content, increases gradually with the increasing cement-to-GGBS ratio, and decreases gradually with increasing carbide slag content. The optimal contents for RHA, the cement-to-GGBS ratio, and carbide slag are X2, Y4, and Z1, respectively. Therefore, based on the range analysis results of the 7-day UCS of the stabilized soil, the optimal mix proportion for the 7-day curing age is determined to be X2Y4Z1.

Figure 2

Figure 2. Influence of component content on the 7-day UCS of stabilized soil.

3.2.3 28-day curing period

The range analysis results for the 28-day strength are presented in Table 8. As shown in the table, for the 28-day UCS of the stabilized soil, the primary and secondary order of the influencing factors is: RHA content (X) > carbide slag content (Z) > cement-to-GGBS ratio (Y). Under standard 28-day curing conditions, the degree of influence of RHA content on the soil strength is greater than that of the cement-to-GGBS ratio.

Table 8

Table 8. Range analysis results (28-day).

The effect of component content on the 28-day UCS is illustrated in Figure 3. It can be observed that the UCS initially increases and then decreases significantly with increasing RHA content, shows an initial increase followed by a decrease with increasing cement-to-GGBS ratio, and decreases significantly with increasing carbide slag content. The optimal content levels for RHA, cement-to-GGBS ratio, and carbide slag are X2, Y2, and Z1, respectively. Consequently, based on the range analysis results of the 28-day UCS, the optimal mix proportion for the 28-day curing age is determined to be X2Y2Z1.

Figure 3

Figure 3. Influence of component content on the 28-day UCS of stabilized soil.

3.3 Analysis of variance

Analysis of variance (ANOVA) was performed based on the UCS test results. The ANOVA results for the 14-day strength are presented in Table 9. According to the ANOVA, the critical F-values obtained from the statistical table are F_0.10(2,2) = 9.00, F_0.05(2,2) = 19.0, and F_0.01(2,2) = 99.0. It can be concluded that for the 14-day UCS of the stabilized soil, the effects of RHA content and carbide slag content are significant, whereas the effect of the cement-to-GGBS ratio is not significant. Furthermore, the influence of RHA content on the 14-day strength is greater than that of carbide slag content, and substantially greater than that of the cement-to-GGBS ratio. A larger F-value indicates a higher degree of influence of the corresponding factor on the test results. The order of significance of the factors based on the F-values is consistent with the results obtained from the range analysis, confirming the validity of the orthogonal experimental analysis.

Table 9

Table 9. Three-factor analysis of variance (ANOVA) results (14-day).

The ANOVA results for the 7-day and 28-day strengths of the stabilized soil are presented in Tables 10, 11, respectively. For the 7-day UCS, the effect of RHA content is significant, whereas the effects of the cement-to-GGBS ratio and carbide slag content are not significant. For the 28-day UCS, the effects of both RHA content and carbide slag content are significant, while the effect of the cement-to-GGBS ratio remains not significant.

Table 10

Table 10. Three-factor analysis of variance (ANOVA) results (7-day).

Table 11

Table 11. Three-factor analysis of variance (ANOVA) results (28-day).

3.4 Stress-strain relationship of stabilized soil

3.4.1 Stress-strain curves of stabilized soil

The stress-strain curves of the stabilized soils are shown in Figure 4, where Figure 4a corresponds to the cement-stabilized soil and Figure 4b to the composite binder-stabilized soil (No. 6). The stress-strain curves of the cement-stabilized soil exhibit a consistent nonlinear growth trend across different curing ages. The stress-strain curve of cement-stabilized soil can be divided into five stages: (1) Compaction Stage: During initial loading, a horizontal compaction plateau appears on the stress-strain curve. This is attributed to voids between the specimen and loading platen, as well as inherent voids within the specimen. Under uniaxial compression, internal micro-defects gradually close, the void ratio of the stabilized soil decreases, and unhydrated cement particles and soil particles are compacted, resulting in increased stiffness. (2) Elastic Stage: With continued loading, stress increases predominantly in this phase. The curve rises nearly linearly with minimal change in the elastic modulus, indicating elastic deformation within the specimen. (3) Plastic Development Stage: As axial pressure further increases, the stress growth rate diminishes, reaching the peak stress. The strain of the cement-stabilized soil increases, elasticity weakens, plasticity enhances, and deformation intensifies. (4) Softening Stage: Following the plastic development stage, the stress-strain curve peaks, after which the stress declines gradually. This post-peak descending branch signifies plastic failure. During this stage, cracks and fissures in the cement-stabilized soil propagate, and the specimen damage becomes irreversible. (5) Residual Strength Stage: In this phase, the stress-strain curve of the cement-stabilized soil becomes essentially horizontal, maintaining a constant stress level. This indicates that the specimen retains some load-bearing capacity after failure, providing residual strength. At this point, plastic deformation is substantial, stress remains nearly constant, and deformation continues to increase.

Figure 4

Figure 4. Stress-strain curves of stabilized soils: (a) cement-stabilized soil; (b) composite binder-stabilized soil.

The stress-strain curves of the soil stabilized with the composite binder are shifted rightward as a whole. This phenomenon may be attributed to the micro-filling and pozzolanic effects of RHA and GGBS within the stabilized soil, which fill pores and generate gel substances. These processes enhance the bonding strength of the stabilized soil, making it less prone to deformation under external load and increasing the elastic and deformation moduli. When the load increases to a certain level, soil particles, cement hydration products, and gel substances all undergo compressive deformation. Due to differences in the deformation moduli of these components, their lateral extrusion deformations under load vary, generating tensile stresses within the material that lead to crack formation and eventual failure.

At 7 days of curing, the stress-strain curves of the composite binder-stabilized soil are similar to those of the cement-stabilized soil. As the curing age increases to 14 and 28 days, the ascending branch of the stress-strain curve gradually steepens, the peak stress increases, the peak shape transitions from gentle to sharp, and the descending branch becomes steeper, indicating a shift in failure mode from plastic to brittle. This evolution in the stress-strain behavior is likely due to the increasing formation of Ca(OH)₂ from cement hydration with prolonged curing. The incorporated RHA and GGBS react with this Ca(OH)₂ to produce calcium silicate hydrate (C-S-H) gel, strengthening the bonds between soil particles. Concurrently, the reduction of Ca(OH)₂ leads to a denser interfacial transition zone, increasing the elastic and deformation moduli of the stabilized soil. Under higher loads, cracks initiate and, once formed, propagate rapidly, resulting in brittle failure of the specimen.

3.4.2 Ultimate strain of stabilized soil

The ultimate strain, defined as the strain value corresponding to the peak stress on the stress-strain curve, serves as an indicator for evaluating material toughness and compressive deformation. Material toughness is directly proportional to the ultimate strain; a greater ultimate strain signifies better material toughness. The ultimate strain values of the stabilized soils are presented in Table 12. As shown in Table 12, the ultimate strain of cement-stabilized soil ranges from 1.1% to 2.4%, while that of the composite binder-stabilized soil ranges from approximately 1.7%–3.3%. Overall, the ultimate strain of the stabilized soils at 7 days of curing is greater than that at 14 and 28 days. Comparative analysis reveals that the incorporation of RHA, GGBS, and carbide slag increases the ultimate strain of the stabilized soil, with values generally higher than those of traditional cement-stabilized soil. This demonstrates that the addition of RHA, GGBS, and carbide slag enhances the toughness of the stabilized soil.

Table 12

Table 12. Ultimate strain of stabilized soils.

3.4.3 Deformation modulus of stabilized soil

In practical engineering applications, such as calculating settlement of stabilized soil foundations, deformation of supporting structures, and establishing constitutive models, the deformation modulus serves as a key parameter. Therefore, statistical analysis of the deformation modulus of stabilized soils possesses significant engineering and practical relevance. The parameter E₅₀ is commonly used in engineering practice to quantify the deformation modulus. E₅₀ is defined as the secant slope from the origin to the point on the stress-strain curve corresponding to 50% of the peak stress, expressed as:

$E_{50}=\frac{\Delta \sigma }{\Delta \epsilon }=\frac{0.5q_u}{{\epsilon }_{0.5}}(1)$

where q_u is UCS of the stabilized soil, and ε_0.5 is the strain value corresponding to the stress equal to 0.5q_u.

The deformation modulus values of the stabilized soils derived according to Equation 1 are summarized in Table 13. As shown in the table, the deformation modulus of the composite binder-stabilized soil at 7 days is lower than that of the cement-stabilized soil. In contrast, at 14 and 28 days, the deformation modulus of the composite stabilized soil is significantly greater than that of the cement-stabilized soil.

Table 13

Table 13. Deformation modulus of stabilized soils.

3.5 Microstructural analysis

Figure 5 presents SEM micrographs of specimens from No. 6 (experimental group) and the control group at curing ages of 7, 14, and 28 days. At a magnification of 2000×, the evolution of the pore structure within the stabilized soil can be observed. When magnified to 5,000×, the morphological development of the hydration products can be seen more clearly. From the images, it can be seen that the stabilized soil primarily consists of soil particles of various morphologies cemented together by gel substances, forming an integrated matrix with specific structural characteristics. The internal structure of the stabilized soil is mainly composed of soil particles, hydration products, and pores. The soil particles are arranged randomly, with pores of varying sizes existing between them. Hydration products are distributed on the surfaces of the soil particles and/or fill the pores. Simultaneously, these hydration products form connections between different particles, thereby enabling the stabilized soil to develop an integrated structure.

Figure 5

Figure 5. SEM micrographs of specimens from No. 6 (experimental group) and the control group at diffrent curing ages. (a) No. 6, 1:2000, 7 days (b) No. 6, 1:5,000, 7 days (c) No. 6, 1:2000, 14 days (d) No. 6, 1:5,000, 14 days (e) No. 6, 1:2000, 28 days (f) No. 6, 1:5,000, 28 days (g) Control group, 1:2000, 7 days (h) Control group, 1:5,000, 7 days (i) Control group, 1:2000, 14 days. (j) Control group, 1:5,000, 14 days. (k) Control group, 1:2000, 28 days. (l) Control group, 1:5,000, 28 days.

Under 2000× magnification, the following microstructural evolution was observed in No. 6: At 7 days of curing, the gel substances formed in the stabilized soil primarily consisted of plate-like and short columnar structures, with a minor amount of needle-like formations. The structure contained numerous pores, with pore sizes distributed in the range of 2.82–5.19 μm. At 14 days of curing, a noticeable increase in longer needle-like, columnar, and flocculent structures was observed within the gel products. The needle-like minerals thickened and interwove, filling the pores. This resulted in a denser overall structure, with a reduction in both pore quantity and size, primarily ranging from 2.55 to 4.14 μm. At 28 days of curing, the quantity of gel substances increased further. Needle-like and columnar structures became enveloped and filled by honeycomb-like and three-dimensional network structures. This led to tighter connections between individual particles, forming a more coherent mass. The pore sizes decreased further, concentrating mainly in the range of 2.43–3.59 μm. Observations at 5,000× magnification revealed the following detailed morphological changes with increasing curing time: At 7 days, a certain quantity of needle-like and columnar minerals was present, but they were predominantly scattered and relatively short, with sizes approximately in the range of 0.74–0.91 μm. Additionally, a significant amount of plate-like minerals was distributed over the surfaces of these needle/columnar minerals and the soil particles. At 14 days, a greater volume of hydration products was formed. The increase in needle-like and columnar minerals was more pronounced, showing evident growth in length and thickness, with sizes mostly between 0.95 and 2.74 μm. Localized agglomeration of these needle-like and columnar minerals, filling pores, was observed. At 28 days, the needle-like and columnar hydration products were largely encapsulated and filled by honeycomb-like and three-dimensional network structures. This led to tighter interparticle connections and a further enhancement of the overall structural integrity.

Under 2000× magnification, the following microstructural evolution was observed in Control Group (Cement-Stabilized Soil): At 7 days of curing, the structures of the gel products formed in the cement-stabilized soil were consistent with those in No. 6. However, the quantity of needle-like and short columnar structures was greater compared to No. 6. Pore sizes ranged from 3.21 to 4.86 μm. At 14 days of curing, the amount of longer needle-like, columnar, and flocculent structures increased, but the extent of this increase was less pronounced than in No. 6. Some interweaving and pore-filling by these structures was observed. Pores decreased in number and size, primarily concentrated in the range of 2.43–4.8 μm. At 28 days of curing, hydration products increased further. In addition to needle-like and columnar structures, a certain amount of honeycomb-like and three-dimensional network structures appeared, but their quantity was less than that in No. 6. Pore sizes distributed in the range of 2.9–4.06 μm. Observations at 5,000× Magnification are as follows: At 7 days, relatively short needle-like and columnar minerals were observed filling pores and coating soil particle surfaces. A certain degree of interconnection between these needle-like and columnar minerals was present, with sizes ranging approximately from 0.56 to 1.26 μm. At 14 days, hydration products increased further. Needle-like and columnar minerals became longer and thicker, with sizes mostly between 1.17 and 1.66 μm. The growth in the quantity and size of hydration products was weaker than in No. 6, with only localized, minor agglomeration observed. At 28 days, some honeycomb-like and three-dimensional network structures also appeared, enveloping needle-like and columnar minerals, but their quantity was less than in No. 6. A significant number of needle-like and columnar minerals remained exposed, and the connections between particles were less tight. These microstructural observations explain why the strength of the cement-stabilized soil was higher than that of the composite stabilized soil at 7 days—due to the greater quantity of hydration products formed initially. Conversely, the superior strength of the composite stabilized soil at 14 and 28 days is attributed to its subsequently exceeding the cement-stabilized soil in both the quantity and formation rate of hydration products.

3.6 Analysis of stabilization mechanism

Based on the aforementioned SEM observations and UCS test results, the stabilization mechanism can be analyzed as follows: Cement, carbide slag, and GGBS are rich in reactive CaO. A substantial amount of CaO reacts with water, creating an alkaline environment and simultaneously generating Ca(OH)₂. Dicalcium silicate (C₂S) and tricalcium silicate (C₃S) in cement rapidly react with water to form calcium silicate hydrate (C-S-H). The reactive SiO₂ and Al₂O₃ present in RHA and GGBS are activated in the alkaline environment and undergo hydration reactions to form calcium aluminosilicate hydrate (C-A-S-H). These processes are referred to as hydration reactions, with the former (cement hydration) proceeding at a faster rate than the latter (pozzolanic reaction). During the hydration reactions, OH⁻ ions are released. Si²⁺ ions (or more accurately, reactive silica) from the clay soil react with Ca²⁺ and OH⁻ ions to form calcium silicate hydrate gel. This process is identified as an ion exchange reaction. Concurrently, a significant amount of reactive SiO₂ reacts with Ca(OH)₂ to produce additional calcium silicate hydrate gel, which is characterized as a pozzolanic reaction. The relevant chemical reactions are expressed as follows:

3.6.1 Hydration reactions

The primary hydration reactions involved in the stabilization process are as follows:

$2\left(3\text{CaO}·{\text{SiO}}_2\right)+6{\mathrm{H}}_2\mathrm{O}\to 3\text{CaO}·2{\text{SiO}}_2·3{\mathrm{H}}_2\mathrm{O}+3\text{Ca}{\left(\text{OH}\right)}_2(2)$

$2\left(2\text{CaO}·{\text{SiO}}_2\right)+4{\mathrm{H}}_2\mathrm{O}\to 3\text{CaO}·2{\text{SiO}}_2·3{\mathrm{H}}_2\mathrm{O}+\text{Ca}{\left(\text{OH}\right)}_2(3)$

$3\text{CaO}·{\text{Al}}_2{\mathrm{O}}_3+6{\mathrm{H}}_2\mathrm{O}\to 3\text{CaO}·{\text{Al}}_2{\mathrm{O}}_3·6{\mathrm{H}}_2\mathrm{O}(4)$

3.6.2 Ion exchange and pozzolanic reactions

The stabilization process further involves ion exchange and pozzolanic reactions, which contribute to the formation of additional cementitious gels. These reactions can be represented as follows:

${\text{SiO}}_2+\text{Ca}{\left(\text{OH}\right)}_2+{\mathrm{H}}_2\mathrm{O}\to \text{CaO}·{\text{SiO}}_2·{\mathrm{H}}_2\mathrm{O}(5)$

${\text{SiO}}_2+2{\text{OH}}^{-}\to {\mathrm{H}}_2{\text{SiO}}_4^{2-}(6)$

Equations 2–4 represent the hydration reactions. Equations 5, 6 illustrate the ion exchange and pozzolanic reactions. Concurrently, RHA contains a substantial amount of nano-sized SiO₂ particles, which effectively fill the internal pores of the stabilized soil. This physical filling action significantly enhances the compactness of the soil matrix, reduces porosity, and consequently improves the overall performance of the stabilized soil. Furthermore, the unique nano-porous structure of RHA endows it with excellent water retention capacity. During the later stages of hydration, the water stored within these pores is gradually released, continuously participating in the hydration reactions. The porous structure of RHA may possess a certain moisture-retention capacity, which could help provide water for ongoing hydration reactions during later curing stages, thereby ensuring the steady development of long-term strength in the stabilized soil. Thus, a comprehensive stabilization mechanism is achieved: the cement and carbide slag in the composite binder create an alkaline environment, with the cement providing early strength through its rapid reaction, while the RHA and GGBS contribute to the medium-to long-term strength development.

4 Conclusion

This study developed a composite binder using RHA, GGBS, carbide slag, and cement for soft soil stabilization. The effects of the binder composition on the strength development and microstructure of the stabilized soil at different curing ages were investigated through UCS tests and SEM. The main conclusions are as follows:

1. For the composite stabilized soil across all curing ages, the order of significance of the factors influencing its properties is: RHA content > carbide slag content > cement-to-GGBS ratio. The influences of RHA content and carbide slag content are significant. Based on the 14-day and 28-day UCS as evaluation criteria, the optimal mix proportion for the composite binder was determined as RHA: Cement: GGBS: Carbide Slag = 3: 6.4: 9.6: 1.

2. Under the optimal composite binder mix proportion, the q_u value of the stabilized soil at 14 days reached 2.51 times that at 7 days, and at 28 days, it reached 1.31 times that at 14 days. This strength growth pattern indicates that the reactions within the composite binder-soft soil system persist for an extended duration.

3. Compared to cement-stabilized soil, the composite stabilized soil incorporating RHA, GGBS, and carbide slag exhibited a rightward shift in the stress-strain curve, an increase in ultimate strain, and enhanced toughness.

4. SEM images revealed an abundance of dense honeycomb-like and network structures in the 28-day composite stabilized soil, significantly more than in the cement-stabilized soil. C-S-H gels connected and filled the pores, increasing soil density and strength, and resulting in better integrity and cementation of the soil skeleton.

5. The stabilization mechanisms of the composite soil primarily involve hydration reactions, ion exchange, pozzolanic reactions, the filling effect of RHA, and its water storage-release mechanism. The C-S-H gel produced by hydration reactions is the primary source of strength for the stabilized soil.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

RX: Validation, Methodology, Writing – original draft. YX: Data curation, Investigation, Writing – original draft. YC: Writing – review and editing, Methodology. ZL: Conceptualization, Writing – review and editing, Supervision.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was supported by the Science and Technology Project of State Grid Jiangsu Electric Power Co., Ltd. (Grant No. JCSQ202506). The authors declare that this study received funding from State Grid Jiangsu Electric Power Co., Ltd.. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article, or the decision to submit it for publication.

Conflict of interest

Author RX was employed by Suqian Sunshine Transmission and Distribution Engineering Co., Ltd.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Chindaprasirt, P., Kanchananda, P., Sathonsaowaphak, A., and Sata, V. (2007). Sulfate resistance of blended cements containing fly ash and rice husk ash. Constr. Build. Mater. 21 (6), 1356–1361. doi:10.1016/j.conbuildmat.2005.10.005

CrossRef Full Text | Google Scholar

Ding, J., Wan, X., Feng, X., Xu, G., Qian, S., and Ji, F. (2019). Compression model for solidified soil of dredged clays with high water content. J. Southeast Univ. Nat. Sci. Ed. 49 (2), 369–374. (in Chinese).

Google Scholar

Feng, S., Zhang, G., and Ren, Y. (2024). Properties and microstructure of soil solidified by titanium slag-flue gas desulfurized gypsum-Portland cement composites as solidifiers. Constr. Build. Mater. 438, 137061. doi:10.1016/j.conbuildmat.2024.137061

CrossRef Full Text | Google Scholar

Gencel, O., Benli, A., Bayraktar, O. Y., Türkmen, H., Sutcu, M., and Elabade, W. A. T. (2021). Effect of waste marble powder and rice ash on the microstructural, physico-mechanical and transport properties of foam concretes exposed to high temperatures and freeze-thaw cycles. Constr. Build. Mater. 291, 123374. doi:10.1016/j.conbuildmat.2021.123374

CrossRef Full Text | Google Scholar

He, J., Shi, X., Li, Z., Zhang, L., Feng, X., and Zhou, L. (2020). Strength properties of dredged soil at high water content treated with soda residue, carbide slag, and ground granulated blast furnace slag. Constr. Build. Mater. 242 (5), 118126. doi:10.1016/j.conbuildmat.2020.118126

CrossRef Full Text | Google Scholar

Jongpradist, P., Youwai, S., and Jaturapitakkul, C. (2011). Effective void ratio for assessing the mechanical properties of cement-clay admixtures at high water content. J. Geotechnical Geoenvironmental Eng. 137 (6), 621–627. doi:10.1061/(asce)gt.1943-5606.0000462

CrossRef Full Text | Google Scholar

Kang, S., Hong, S., and Moon, J. (2019). The use of rice husk ash as reactive filler in ultra-high performance concrete. Cem. Concr. Res. 115, 389–400. doi:10.1016/j.cemconres.2018.09.004

CrossRef Full Text | Google Scholar

Liu, F., Zhu, C., Yang, K., Ni, J. F., Hai, J., and Gao, S. H. (2019). Effects of fly ash and slag content on the solidification of river-dredged sludge. Mar. Georesources Geotechnol. 39 (1), 65–73. doi:10.1080/1064119x.2019.1677827

CrossRef Full Text | Google Scholar

Ministry of Water Resources of the People’s Republic of China (2019). GB/T 50123-2019 standard for geotechnical testing method. Beijing: China Planning Press.

Google Scholar

Oormila, T. R., and Preethi, T. V. (2014). Effect of stabilization using flyash and GGBS in soil characteristics. Int. J. Eng. Trends Technol. 11 (6), 284–289. doi:10.14445/22315381/ijett-v11p254

CrossRef Full Text | Google Scholar

Peng, L., and Chen, B. (2021). Investigation on rapid solidification of waste dredged sludge with magnesium phosphate cement modified by GGBFS. Eur. J. Environ. Civ. Eng. 26 (2), 1–26. doi:10.1080/19648189.2021.1987333

CrossRef Full Text | Google Scholar

Phummiphan, I., Horpibulsuk, S., Rachan, R., Arulrajah, A., Shen, S. L., and Chindaprasirt, P. (2018). High calcium fly ash geopolymer stabilized lateritic soil and granulated blast furnace slag blends as a pavement base material. J. Hazard. Mater. 341, 257–267. doi:10.1016/j.jhazmat.2017.07.067

PubMed Abstract | CrossRef Full Text | Google Scholar

Shi, J., Wang, S., Cao, W., Su, J., and Zhang, S. (2022). Mechanical properties and strengthening mechanism of dredged silty clay stabilized by cement and steel slag. Materials 15 (11), 3823. doi:10.3390/ma15113823

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, Y., Yang, J., and Chang, R. (2023). The design of ternary all-solid-waste binder for solidified soil and the mechanical properties, mechanism and environmental benefits of CGF solidified soi. J. Clean. Prod. 429, 139439. doi:10.1016/j.jclepro.2023.139439

CrossRef Full Text | Google Scholar

Xue, Z., Zhang, Y., Luo, J., Li, S., Li, J., and Li, Y. (2024). Analysis of compressive strength, durability properties, and micromechanisms of solidified loess using industrial solid waste:Slag-white mud-calcium carbide residue. J. Build. Eng. 84, 108511. doi:10.1016/j.jobe.2024.108511

CrossRef Full Text | Google Scholar

Yi, Y., Liska, M., Jin, F., and Al-Tabbaa, A. (2016). Mechanism of reactive magnesia-ground granulated blastfurnace slag (GGBS) soil stabilization. Can. Geotechnical J. 53 (5), 773–782. doi:10.1139/cgj-2015-0183

CrossRef Full Text | Google Scholar

Zeng, L., Bian, X., Zhao, L., Wang, Y., and Hong, Z. (2021). Effect of phosphogypsum on physiochemical and mechanical behaviour of cement stabilized dredged soil from Fuzhou, China. Geomechanics Energy Environ. 25, 100195. doi:10.1016/j.gete.2020.100195

CrossRef Full Text | Google Scholar

Zhang, W., Zhao, L., Brian, A. M. C., Han, Y., Yang, M., Liu, Y., et al. (2020). Dredged marine sediments stabilized/solidified with cement and GGBS: factors affecting mechanical behaviour and leachability. Sci. Total Environ. 733, 138551. doi:10.1016/j.scitotenv.2020.138551

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, W., Liu, C., Liu, H. W., Lin, X., and Zhang, Z. N. (2023). Freeze-thaw damage deterioration mechanism of rice husk ash concrete based on pore volume fractal dimension. Acta Mater. Compos. Sin. 40 (8), 4733–4744.

Google Scholar

Zhou, H., Wang, X., Wu, Y., Wang, J., and Hu, X. (2021). Mechanical properties and micro-mechanisms of marine soft soil stabilized by different calcium content precursors based geopolymers. Constr. Build. Mater. 305, 124722. doi:10.1016/j.conbuildmat.2021.124722

CrossRef Full Text | Google Scholar

Zhu, X., Niu, F., Ren, L., Jiao, C., Jiang, H., and Yao, X. (2022). Effect of calcium carbide residue on strength development along with mechanisms of cement-stabilized dredged sludge. Materials 15 (13), 4453. doi:10.3390/ma15134453

PubMed Abstract | CrossRef Full Text | Google Scholar