Study on the Influence of Clay Content on the Freeze–Thaw Characteristics and Mechanisms of Solidified Low-Liquid-Limit Clay

Abstract

This study investigates the effects of clay content on the strength and microstructural mechanisms of artificially prepared low-liquid-limit clay solidified with SSGM binder, composed of salt sludge (SAS), steel slag (SS), ground granulated blast-furnace slag (GGBS), and light magnesium oxide (MgO), and the law of influence of viscous particles content on the strength of the solidified low-liquid-limit clay and its microscopic mechanism were investigated through a freeze–thaw cycle test and microscopic test. The results indicate that, under freeze–thaw cycles, both the mass and unconfined compressive strength of the solidified soil decrease with increasing cycle number. At the same number of cycles, samples with lower clay content exhibit smaller mass loss rates and unconfined compressive strength loss rates. Microstructural tests reveal that the hydration products of the binder, including C-S-H, C-A-S-H, C-A-H, and AFt, not only cement soil particles and fill internal pores but also interconnect to form a mesh-like structure, enhancing internal stability. However, as freeze–thaw cycles progress, the structure of the solidified soil deteriorates, with an increase in large pores and the formation of penetrating cracks and voids, leading to reduced strength. The SSGM binder demonstrates excellent freeze–thaw resistance for solidifying low-liquid-limit clay and improves the utilization rate of industrial waste, showing promising application potential in permafrost regions.

1. Introduction

Climatic conditions in cold regions, such as seasonal temperature changes and significant diurnal temperature variations, can cause some freeze–thaw-cycling effects on the soil. Low-liquid-limit clay, characterized by fine particles and a large surface area, exhibits strong water adsorption capacity. Additionally, its small and densely distributed pores make it particularly susceptible to the effects of water phase transitions during freeze–thaw processes [1,2], rendering this type of soil highly sensitive to freeze–thaw cycles [3,4]. In order to eliminate the undesirable effects of low-liquid-limit clay under freeze–thaw cycles, it is usually improved using solidifying binders [5], and the use of solid wastes, such as steel slag [6], fly ash [7], and ground granulated blast-furnace slag [8], can effectively improve its frost resistance.

Cement is the most commonly used binder. However, the production of cement is increasingly associated with significant energy consumption, resource depletion, ecological damage, and environmental pollution [9]. Salt sludge is a solid byproduct generated during the ammonia-soda process for alkali production [10]. Its disposal primarily involves stockpiling, landfill, or sea reclamation by damming. However, these methods can lead to the formation of “white seas”, causing soil and water alkalization [11]. Currently, the resource utilization of salt sludge primarily focuses on the recovery and extraction of NaCl [12], and limited research has been conducted on its application as an alkali activator. The CaO and MgO present in salt sludge can provide alkaline environments by releasing alkali metal cations such as Ca²⁺ and Mg²⁺ during hydration reactions [13]. Steel slag, a byproduct of the steel industry, contains cementitious components and exhibits potential as a cementitious material [14,15], but due to its own low early hydration activity [16] and constraints on its stability [17], its performance as a supplementary cementitious material in cement-solidified soil is not ideal when used alone [18]. However, studies have shown that steel slag, when combined with ground granulated blast-furnace slag as a cementitious material, can exhibit excellent synergistic hydration reactions [19,20].In the alkaline environment provided by soda residue [21], the alkali activator induces the dissociation of vitreous phases in steel slag and ground granulated blast-furnace slag during the early stages of hydration. This process significantly influences the structure of hydration products [22], further activating their reactivity and leading to the formation of cementitious materials [23]. Due to its insufficient alkalinity, salt sludge cannot fully activate the reactivity of steel slag–ground granulated blast-furnace slag. Adding chemical additives such as MgO can effectively enhance the alkalinity of the hydration environment. Furthermore, studies have found that reactive MgO-magnesium carbonate composite cementitious materials generate low-crystallinity brucite [24] during hydration, which effectively improves their strength [25,26]. Our research group has conducted a study on the solidifying of gold tailings with an SSG all-solid-waste binder prepared by using alkaline solid waste SAS as alkali exciter and active silica–alumina solid waste SS and GGBS as volcanic ash material. The results demonstrate that it can effectively solidify gold tailings and be used as a filling material for mining applications [27].

In order to eliminate the adverse effects of freeze–thaw cycles on low-liquid-limit clay and, at the same time, to solve the problem of environmental pollution and waste of resources caused by a large number of industrial solid waste piles and landfills, in this study, low-liquid-limit clay with varying clay contents was prepared by mixing pure clay soil, pure silt soil, and pure sandy soil at different mass ratios. The soil samples were solidified using the SSGM binder, composed of salt sludge (SAS), steel slag (SS), ground granulated blast-furnace slag (GGBS), and light magnesium oxide (MgO). Freeze–thaw cycle tests were conducted to investigate the freeze–thaw characteristics and mechanisms of the solidified low-liquid-limit clay with different clay contents. The results of this study can provide a theoretical basis for the application of SSGM-solidified soil in actual projects, and, at the same time, it has certain practical significance and practical value for engineering environmental protection and sustainable economic and social development.

2. Materials and Methods

2.1. Materials

2.1.1. Test Soil

In this study, commercial kaolin, Dongying silt, and commercial quartz sand were used as pure clay soil, pure silt soil, and pure sandy soil. The particle size of commercial kaolin is less than 0.005 mm; Dongying silt was collected from a depth of 30~50 cm below the surface in the Yellow River Estuary wetland (118°43′30′′ E, 37°35′47′′ N), screened, dried, crushed, and milled through a 0.075 mm sieve, with a particle size distribution of 0.005~0.075 mm; the commercial quartz sand was purchased from Jinan Zhengtai Company (Jinan, China), with a particle size distribution of 0.075~1 mm.

2.1.2. Binder Components and Ratios

The SSGM binder employed in this experiment was formulated by blending alkali-activated salt industrial waste residue (SAS), pozzolanic steel slag (SS), and ground granulated blast-furnace slag (GGBS) at a mass ratio of 1:5:4, supplemented with 2% lightweight magnesium oxide (MgO) by mass of SAS as an additive. SAS was taken from Shandong Aluminum Chlor-Alkali Plant (Shandong, China); SS was taken from Shandong Laigang Yongfeng Steel Company (Shandong, China); GGBS was taken from a steel plant in Jinan, Shandong Province; and MgO was a chemically pure reagent of Sinopharm Group (Figure 1). The particle size grading curves of each component of SSGM binder are shown in Figure 2, the main chemical compositions and contents are listed in

Table 1. Chemical composition of test solid waste.

Solid Waste Type	Chemical Composition and Content (%)
	SiO2	CaO	Al2O3	SO3	Fe2O3	MgO	K2O	Na2O
SAS	1.18	30.96	0.31	0.81	0.85	24.16	8.64	0.59
SS	14.84	48.64	3.72	0.64	27.55	1.21	0.56	0.04
GGBS	30.87	41.86	14.97	2.50	0.34	8.05	0.27	0.32

, and the XRD patterns are shown in Figure 3. The chemical composition of SAS is dominated by CaO and MgO, with the main mineral phases being calcium carbonate and brucite, with a pH of 9.76. The chemical composition of SS is dominated by CaO, Fe₂O₃, and SiO₂, with the main mineral phases being C₂S and RO phases. The chemical composition of GGBS is dominated by SiO₂, CaO, and Al₂O₃, with no distinct crystalline mineral phase diffraction peaks observed, indicating that its mineral phase is predominantly an amorphous glassy phase.

2.2. Test Schemes and Test Methods

2.2.1. Schemes

Pure clay soil, pure silt soil, and pure sandy soil were mixed at different mass ratios to prepare artificially prepared soil with varying particle size compositions. Figure 4 shows the particle size composition and particle size distribution curves of the pure clay soil, pure silt soil, pure sandy soil, and artificially prepared soil.

Table 2. Basic physical properties of test soil and artificially prepared soil.

Soil Classification Nomenclature		Clay/Silt/Sand	GS	wP (%)	wL (%)	IP
Test Soil	Clay	-	2.74	33.9	55.1	21.2
	Silt	-	2.71	22.7	31.6	8.9
	Sand	-	2.67	-	-	-
Artificially prepared soil	CSS011	0∶50∶50	2.68	12.0	17.1	5.1
	CSS122	20∶40∶40	2.69	7.0	23.5	16.5
	CSS111	33.3∶33.3∶33.3	2.70	16.2	24.7	8.5
	CSS100	100∶0∶0	2.74	33.9	55.1	21.2

presents their basic physical properties. The pure clay soil, pure silt soil, and pure sandy soil are classified as high-liquid-limit clay, low-liquid-limit silt, and poorly graded sand, respectively. CSS011, CSS122, and CSS111 belong to low-liquid-limit clay, while CSS100 is classified as high-liquid-limit clay [28].

The artificially prepared soils listed in Table 2 were mixed with the SSGM binder to prepare solidified soil samples, which were then subjected to freeze–thaw cycles curing.

Table 3. Sample preparation scheme for solidified soil.

Soil Type	Binder	Mixing Ratio of Binder (%)	Curing Age (d)	Freeze–Thaw Cycle Times (Cycles)	Water Admixture (%)
CSS011 CSS122 CSS111 CSS100	SSGM P.O 42.5	15	28	0, 2, ,4, 6, 8, 10	1.2wL

shows the preparation scheme of the solidified soil specimens.

2.2.2. Methods

• Sample preparation methods;

According to the preparation scheme in Table 3, the soil samples were mixed with the binder, and the corresponding mass of deionized water was added and mixed uniformly. The mixture was then placed into molds with a diameter of 50 mm and a height of 100 mm. The molds were placed in a curing chamber for standard curing (20 ± 2 °C, humidity ≥ 95%). After 3 days of curing, the samples were de-molded and further cured until 28 days, after which subsequent tests were conducted.

Refer to ASTM D560-03 [29] for freeze–thaw cycles. The specimen of standard maintenance for 28 d is put into the automatic freeze–thaw cycles test equipment for maintenance, and the design of freeze–thaw cycles temperature is −20 °C~20 °C. Firstly, the specimens are put into the freezing chamber of −20 °C for 24 h, and then they are put into the thawing chamber of 20 °C to melt for 24 h as one freeze–thaw cycle. Then, 0, 2, 4, 6, 8, 10 freeze–thaw cycles are carried out. The unconfined compressive strength test and microscopic test were performed after reaching a set number of cycles.

2.

Unconfined compressive strength test

For the specimens that reached the designated curing age, unconfined compressive strength tests were conducted using the WCY-1 unconfined compression tester manufactured by Nanjing Soil Instrument Company. The test was performed under strain-controlled loading at a rate of 1%/min, and the test was terminated upon specimen failure.

3.

Microscopic examination

An XRD analysis was performed using a Rigaku Miniflex600 X-ray diffractometer to determine the mineral phase composition. The samples were dried at 105 °C to constant weight, powdered through a 0.075 mm sieve, and a copper rotary anode target was used. Scanning method: 2θ goniometer, accuracy 0.002°; scanning angle 5–75°.

An SEM-EDS analysis was conducted using a German Zeiss Gemini500 field-emission scanning electron microscope in combination with an X-ray energy spectrometer from Oxford Instruments ULTIM MAX100 to analyze the microscopic morphology and elemental composition. The samples were lyophilized and treated with gold spraying on the surface, and the magnification was set at 20,000×.

An MIP analysis was performed using a Micro Active AutoPoreV9600 porosimeter, with a measurable pore size range of 0.006 to 300 μm.

3. Clay Content on Freeze–Thaw Characteristics of Solidified Soils

3.1. Mass Loss Rate

The rate of mass loss of the specimen is calculated according to Equation (1):

$$K_m={\displaystyle \frac{m_0-m_n}{m_0}}\times 100\%$$

(1)

where K_m is the mass loss rate, %; m₀ is the initial mass of the specimen before the freeze–thaw cycle, g; m_n is the mass of the specimen after the n cycles, g.

The mass loss rate of the SSGM-solidified soil versus the number of cycles is shown in Figure 5. After 10 freeze–thaw cycles, the mass loss rates of specimens CSS011, CSS122, CSS111, and CSS100 were 0.77%, 0.94%, 0.97%, and 1.83%, respectively. As shown in the figure, the mass loss rate of solidified soil increases with the number of freeze–thaw cycles, indicating that greater freeze–thaw cycles lead to higher mass loss. Under the same number of cycles, the mass loss rate of solidified soil increases with higher clay content, suggesting that lower clay content results in less mass loss.

3.2. Stress–Strain Curves

The stress–strain curves of the SSGM-solidified soil are shown in Figure 6, where FT represents the number of freeze–thaw cycles.

As shown in the figure, the stress of the solidified soil increases with strain and reaches a distinct peak stress. The destructive strain is concentrated between 1.0% and 1.2%, and the curve exhibits strain-softening behavior, indicating brittle damage to the specimens. After freeze–thaw cycles, the peak stress decreases, but the destructive strain and the form of the stress–strain curve remain largely unchanged [30]. The pore structure damage of high-clay-content (>40%) soils increases after freeze–thaw and becomes more fragile with the increase in the number of cycles, with the damage strain shifted significantly to the right in the figure. For the soil with high content of silt-sand, due to its coarse particles, the friction between particles is stronger, and in the process of the freeze–thaw cycle, it will not be as fine-grained due to significant damage or deformation, so the damage strain is lower, and with the increase in the number of cycles, the damage strain does not increase significantly.

3.3. Strength Loss Rate

The peak stress was taken as the unconfined compressive strength of the specimens, and the cement-solidified soil under the same condition of maintenance was used as the control specimen. The relationship between the unconfined compressive strength and the number of freeze–thaw cycles of the SSGM-solidified soil and cement-solidified soil is shown in Figure 7.

As shown in Figure 7a, the unconfined compressive strength of the SSGM-solidified soil gradually decreases with an increasing number of freeze–thaw cycles. Additionally, the rate of strength reduction diminishes as the clay content decreases, resulting in flatter curves. The unconfined compressive strength of the SSGM-solidified soil decreases from 7041.8 kPa to 6978.6 kPa, 5680.0 kPa to 5450.1 kPa, 5052.1 kPa to 4853.8 kPa, and 1205.7 kPa to 1052.8 kPa, respectively. From Figure 7b, the unconfined compressive strength of cement-solidified soil decreased from 6224.6 kPa to 6115 kPa, 5747.3 kPa to 5609.7 kPa, 5435.0 kPa to 4969.7 kPa, and 3494.5 kPa to 957.7 kPa, respectively. The strength of the SSGM-solidified soil after undergoing freeze–thaw cycles was not lower than the strength of the cement-solidified soil under the same conditions.

The process of freeze–thaw cycles caused a significant pore water migration redistribution effect in clay particles due to their high water absorption and strong adhesion; the high-clay-content soil, in the process of freezing and thawing, has a larger ice crystal expansion and contraction amplitude. The extrusion of soil particles leads to the destruction of the internal pore structure of the soil body, showing a more pronounced reduction in strength. By fitting the relationship between unconfined compressive strength of SSGM-solidified soil and the number of cycles, the following equation is obtained:

$$q_{un}=q_{u0}+{a*b}^n$$

(2)

where q_un is the unconfined compressive strength of the specimen after n times of freeze–thaw cycles, kPa; q_u0 is the unconfined compressive strength of the specimen before freeze–thaw cycles, kPa; a, b are the fitting parameters.

The fitted parameters are shown in

Table 4. Table of fitting parameters.

Soil Type	a	b	R2
CSS011	76.4	0.85	0.9828
CSS122	289.6	0.85	0.9949
CSS111	229.9	0.82	0.9991
CSS100	359.9	0.94	0.9710

, from which it can be seen that the fitted coefficients R² are all greater than 0.95, and the correlation of the fitted relational equation is good.

The rate of loss of strength of the specimen is calculated according to Equation (3):

$$K_q={\displaystyle \frac{q_{u0}-q_{un}}{q_{u0}}}\times 100\%$$

(3)

where K_q is the strength loss rate, %; q_u0 is the strength of the specimen before cycling, kPa; q_un is the strength of the specimen after the n cycles, kPa.

A smaller strength loss rate K_q indicates stronger freeze–thaw resistance of the solidified soil. The relationship between the strength loss rate and the number of cycles is shown in Figure 8.

The rate of strength loss versus the number of cycles for SSGM-solidified and cement-solidified soils is shown in Figure 8. As can be seen in Figure 8, after 10 freeze–thaw cycles, the strength loss rates of SSGM-solidified soil specimens CSS011, CSS122, CSS111, and CSS100 were 18.37%, 22.70%, 23.49%, and 29.09%, respectively. The strength loss rates of cement-solidified soil specimens CSS011, CSS122, CSS111, and CSS100 were 20.59%, 22.64%, 25.69%, and 78.01%, respectively. Experimental results demonstrate that a lower clay content corresponds to reduced strength loss rates and enhanced freeze–thaw resistance in solidified soils. SSGM-solidified soil exhibits comparable freeze–thaw durability to cement-solidified soil while significantly outperforming cement solidification under high clay content conditions.

Freeze–thaw cycling reduces the inter-particle bonding in soils with high clay content, resulting in a significant reduction in the strength of the soil mass. When the clay content is low, the strength of the soil body is relatively high due to the lower swelling and moisture adsorption capacity, as well as the tighter particle structure of the soil body, and the freeze–thaw cycle results in less damage, and therefore the loss of strength of the soil body is low. In addition, freeze–thaw cycles show significant inhibition of alkali-stimulated cementation reaction, and its inhibition effect is positively correlated with the number of cycles. The weakening of alkali-stimulated cementation reaction will lead to the reduction in hydration product generation, and at the same time, the freezing and thawing of pore water will also damage the structure of the hydration product that has been generated by the reaction, which is an important reason for the decrease in the strength of solidified soils.

3.4. Variation in Deformation Modulus

The relationship between the deformation modulus E₅₀, the number of freeze–thaw cycles, and the unconfined compressive strength is shown in Figure 9. As shown in the figures, the deformation modulus of the specimens decreases with an increasing number of cycles, indicating that freeze–thaw cycles reduce the ability of the solidified soil to resist deformation.

In addition, the deformation modulus E₅₀ of the solidified soil showed a good linear relationship with the unconfined compressive strength q_u, viz:

E₅₀ = 69.09 q_u + 53.42

(4)

With a 95% confidence interval, it can be observed that the deformation modulus E50 has a good linear relationship with the unconfined compressive strength q_u. Furthermore, as the clay content in the solidified soil decreases, E₅₀ shows an increasing trend, with the deformation modulus shifting toward the upper right. This indicates that the ability of the specimens to resist deformation gradually increases with lower clay content, and their brittleness becomes more pronounced.

3.5. Microscopic Properties

3.5.1. XRD Analysis

Figure 10 shows the XRD pattern of SSGM-solidified soil after 10 freeze–thaw cycles of curing versus standard curing for 48 d.

As illustrated in the figure, under standard curing conditions, free CaO and MgO in the alkali-activated SAS rapidly react with water to form Ca(OH)₂ and Mg(OH) ₂, releasing heat and elevating the system pH. This alkaline environment promotes pozzolanic reactions, activating the vitreous phases in slag and C₂S (dicalcium silicate) in steel slag under high pH conditions. Primary alkali-activated products include AFt (ettringite), CH (portlandite), C-S-H (calcium silicate hydrate), C-A-H (calcium aluminate hydrate), and C-A-S-H (calcium aluminosilicate hydrate) gels. Products after freeze–thaw cycling are approximately the same as in standard conservation conditions. Compared to the solidified soil under standard curing conditions, freeze–thaw cycles caused the microstructure of the hydration products to change, and the diffraction peaks were slightly shifted to the left. After 10 freeze–thaw cycles, the intensity of the amorphous phase diffraction peaks in the 2θ range of 25~35° significantly decreases, indicating that the expansion and contraction forces generated by freeze–thaw cycles lead to the formation of more micropores within the specimens [31]. This disrupts the gel structures of amorphous phases such as C-S-H, C-A-H, and C-A-S-H.

3.5.2. SEM-EDS Analysis

Figure 11 and Figure 12 show the SEM-EDS patterns of SSGM-solidified soil after standard maintenance for 48 d and 10 freeze–thaw cycles. To further investigate the hydration products, EDS tests were conducted on the points marked in the figure. Combined with EDS spectral analysis, it was found that a large number of cementitious materials, such as needle-like AFt, flocculent C-S-H, C-A-H, and C-A-S-H, were generated in the solidified soil under standard curing conditions. These products filled the internal pores of the specimens, and a significant amount of CaCO₃ and Ca(OH)₂ was observed attached to the surface. Compared with the standard maintenance conditions, the specimens showed a large number of obvious cracks and holes after the freeze–thaw cycles, the structure of the massive CaCO₃ was destroyed, and the cementation between the cementitious materials gradually weakened, which made the strength of the solidified soil decrease.

3.5.3. MIP Analysis

Figure 13a shows the pore size distribution curves of soil samples under different numbers of cycles. As shown in the figure, the pore size distribution of the specimens measured by MIP tests is mainly concentrated in the range of 25~130 nm. With an increasing number of cycles, the peak value and peak width of the pore size distribution curves gradually decrease.

The specimen pores were classified into three major categories based on the characteristics of the pore distribution curves: small pores (<10 nm), medium pores (10~100 nm), and large pores (>100 nm), as shown in Figure 13b.

As the number of freeze–thaw cycles increases, the number of pores smaller than 100 nm gradually decreases, while the number of pores larger than 100 nm gradually increases. This indicates that the freeze–thaw cycles alter the pore distribution in the soil. The pore size of the soil sample enlarges with the increasing number of cycles, which is consistent with the observation from SEM images that interconnected cracks and voids appear in the solidified soil after 10 cycles. The formation of these interconnected cracks and voids is the primary reason for the reduction in the strength of the solidified soil.

4. Conclusions

This study utilizes salt sludge (SAS) as an alkali activator, steel slag (SS) and ground granulated blast-furnace slag (GGBS) as pozzolanic materials, and light magnesium oxide (MgO) as an external admixture to prepare SSGM binder for solidifying artificially configured soil. The research investigates the influence of different clay contents on the freeze–thaw characteristics and mechanisms of SSGM-solidified low-liquid-limit clay under freeze–thaw-cycling conditions. The main conclusions obtained are as follows:

(1) Under the action of freeze–thaw cycles, the mass loss rate of solidified soil increases with the number of freeze–thaw cycles, and at the same number of cycles, the mass loss rate increases with the increase in clay content.

(2) As the number of freeze–thaw cycles increases, the unconfined compressive strength of the solidified soil gradually decreases, following a pattern consistent with an exponential function model. When the clay content is 0%, 20%, 33%, and 100%, the unconfined compressive strength loss rates of the solidified soil after 10 cycles are 18.37%, 22.70%, 23.49%, and 29.09%, respectively. The higher the clay content, the greater the strength loss rate.

(3) Microscopic test results indicate that the hydration products generated by the binder mainly include C-S-H, C-A-H, C-A-S-H, and AFt. These products cement soil particles and fill internal pores while also forming an interconnected mesh-like structure, thereby enhancing the internal stability of the solidified soil. However, freeze–thaw cycles cause the migration and redistribution of moisture within the sample, leading to structural damage in the solidified soil. This results in an increase in the number of large pores and the formation of interconnected cracks and voids, ultimately reducing its strength.