Mechanisms of concentration control alkali activated fly ash stabilized saline soil in seasonally frozen regions

Abstract

In the framework of sustainable development and environmental preservation, this research aims to improve the stability and frost resistance of sulfate saline soil by utilizing industrial solid waste. Geopolymer materials containing fly ash (FA) activated by different NaOH concentrations were studied for study on stabilized soil with saline soil, with NaOH concentrations used ranged from 0.1 to 0.9. This study investigates the impact of the molar concentration of NaOH and the number of freeze-thaw cycles on the microstructure and strength of stabilized soil incorporating FA geopolymer. The XRD, FTIR, and TG studies of NaOH-excited FA stabilized soil revealed that the FA gel material grew with increasing concentration. The strength and frost resistance of stabilized soil increased and then declined as NaOH concentration increased, with an optimum excitation concentration of 0.5 M. After 28 days of curing at 20 °C, its UCS and splitting strength were 7.18 MPa and 1.89 MPa, respectively. The residual values of UCS and splitting strength after 5 freeze-thaw cycles (12 hours of freezing followed by 12 hours of thawing at +20 °C) at the optimal concentration were 46.35% and 39.92%, respectively.

Introduction

Frozen soil is a widespread geological phenomenon at high latitudes that develops when temperatures fall below 0 °C^1,2. Typically, frozen soil is composed of four phases: soil particles, water, ice, and gas, exhibiting a non-homogeneous and anisotropic structure^3,4. Frozen soil is categorized into two types based on their freezing length of time: year-round frozen soil and seasonally frozen soil⁵. Moreover, a high concentration of salt in the soil can lead to the formation of frozen saline soil, which can pose engineering challenges such as salt expansion and erosion⁶. Prolonged temperature changes can produce freeze-thaw cycles (FTs), which can lead to saline expansion, frost heave⁷, and thaw-sinking⁸ from moisture condensation and thawing, leading to a reduction in the bearing capacity of the road base, resulting in problems such as cracks in the pavement and cracking of the road subgrade^9,10. The Xinjiang region is predominantly characterized by a seasonally saline frozen soil zone, creating an optimal external environment for the freezing and thawing of saline soil¹¹. The coupling between saline and frost in saline soil within the seasonally frozen zone is the main factor that weakens the strength of saline soil, leading to engineering issues. Consequently, in order to address these problems, it is essential to consolidate saline soil within the seasonally frozen zone¹².

Industrial solid wastes come from diverse sources such as: Fly ash (FA)¹³, silica fume, granulated blast furnace slag¹⁴, rice husk ash¹⁵, etc., and they have a wide range of uses^16,17 including their use in concrete¹⁸, bricks¹⁹ water-stabilized sub-base for roadways, and metal extraction^20,21 among others²². Currently, researchers are investigating the utilization of industrial solid waste reinforcement techniques to stabilize saline soil that are exposed to FTs. The inclusion of cement kiln dust (CKD) significantly enhances the unconfined compression strength, deformation modulus, and soaked California bearing ratio of the soil. However, it also reduces the ductility of the treated soil²³. Usama Khalid cured expansive soil by combining wheat straw with silica fume, which causes aggregation improving the mechanical response of expansive soil²⁴. Qin et al²⁵ found the experiment results revealed that, compared to unsolidified soil, the soil solidified with 8 % active MgO (8M0F) and a combination of 4 % active MgO and 4 % FA (4M4F) demonstrated a significant increase in ultimate strength by 14 and 12 times, respectively. Li et al²⁶ discovered a noteworthy decrease in the temperature sensitivity of stabilized soil through the application of composite curing agents during the curing process of saline soil. The addition of lime and FA to carbonate saline soil resulted in a transformation of the stress-strain curves from the strain hardening type to the strain softening type. Additionally, it altered the strain at which the peak deviatoric stress of the soil occurred²⁷. Usama Khalid²⁸ discovered that composite binary geopolymers, consisting of bagasse ash and quarry dust, have the potential to greatly enhance the compressive strength of roadbeds. Julphunthong²⁹ found that a Ca/(SiO₂ + Al₂O₃) ratio of 1.5 for a mixture of calcium carbide slag and FA gave the highest compressive strength of stabilized loess after prolonged curing. Temuujin³⁰ discovered that when CaO and Ca(OH)₂ were added to the FA based geopolymer at a mass ratio of 3%, respectively, the 7 day UCS increased from 11.8 MPa to 22.8 MPa and 29.2 MPa. Muhammad Hamza³¹ investigated the use of calcined eggshell powder for mitigation of fertilized clay geotechnical soil. He discovered that increasing the ESP content in fertilized clays resulted in a significant increase in unconfined compressive strength (UCS), elastic modulus, and California bearing ratio. With increasing powdered glass, the consistency limits, compression characteristics, swell characteristics, and optimum moisture content decreased while maximum dry unit weight, yield stress, California bearing ratio (CBR), and UCS increased³².

Alkali activation reactions are employed for the conversion of industrial wastes such as FA and slag into geopolymer-cured soil. In the production of geopolymers using different industrial solid waste raw materials, the use of an alkali activator is generally necessary³³. The most common alkali activator is a combination of NaOH and Na₂SiO₃ solutions, and the Na₂SiO₃/NaOH ratio is the most important factor influencing geopolymer mechanical characteristics³⁴. Wagdi Hamid et al³⁵ found that NaOH excited FA to de-cure sabkha soil, the UCS enhancement was greatly affected by the molar concentration of NaOH and the alkaline activation of sabkha soil incorporating FA had higher UCS and shear strength than the untreated and Ordinary Portland cement treated specimens. The study also discovered that the application of geopolymer treatment improves the durability of sabkha soil^36,37. Due to its low alkalinity, FA undergoes a restricted level of hydration. Therefore, the inclusion of an exciter is necessary to enhance the hydration reaction of FA³⁸. The content of alkali has an impact on the solubility of aluminum silicate, making it an important parameter for the activation of FA³⁹. Nematollahi et al⁴⁰ conducted experiments using activated mixtures of FA, slag, or slaked lime with three different grades of Na₂SiO₃, as well as combinations of Na₂SiO₃ and NaOH, at ambient temperature. The results showed that these mixtures produced geopolymers with low to medium compressive strength (~40 MPa) after 28 days. The solubility of the FA and slag is enhanced by increasing the OH⁻ content in the FA/slag solution, leading to an increase in compressive strength⁴¹. Hai-yan et al⁴² researched that the optimum conditions for dissolving SiO₂ and Al₂O₃ in geopolymers made from alkali-inspired saline soil combined with FA were 10 M NaOH, 60℃, and 24 hours. Previous research on alkali-excited FA for stabilized soil utilization has mostly focused on mechanical properties⁴³, and in order to achieve higher mechanical property indexes, compound alkali^44,45 and multiple solid wastes^46,47 are frequently used, complicating the excitation process and ignoring the alkali-excited process. The performance and mechanism of alkali-activated FA for stabilizing saline soil in the seasonally frozen zone have been given limited attention^48,49. Zhou et al⁵⁰ researched that an increase in the number of NaOH moles leads to an increase in the strength of stabilized soil, attributed to the fact that the increase in OH⁻ concentration accelerates the silica and aluminium reactions and promotes the formation of gelling products.

Recognizing the limitations of previous research, the main objective of this study was to assess the solidification mechanism and micro-macro characteristics during FA excitation and FTs with different alkali concentrations. The research examines the mechanical properties of FA-activated stabilized soil with saline soil and its behavior during the FTs, focusing on the microstructural changes caused by chemical processes. The study investigates the impact of alkali content and FTs using various methods such as UCS tests, splitting experiments, FTs experiments, X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM), Fourier Transform Infrared Spectroscopy (FTIR), X-ray Photoelectron Spectroscopy (XPS), and Thermogravimetric (TG-DSC) analysis. These techniques are used to analyze the macro-mechanical properties and micro-characterization of the soil. The research also explores the chemical composition, mineralogical composition, and micro-structural alterations of cured saline soil, as well as the mechanism of using NaOH-activated FA in the stabilization process.

Materials and methods

Materials

The saline soil was collected from the Boda Campus of Xinjiang University in Urumqi, Xinjiang, China, which is known for its high salinity levels. The SEM and XRD analyses of the saline soil and FA samples collected are presented in Fig. 1a, b the XRF results are shown in Table 1. While the basic physical properties of saline soil are summarized in Table 2. The XRD analysis of the saline soil, indicates the presence of mainly quartz (SiO₂), calcium carbonate (CaCO₃), and Margarite (KAl₂(AlSi₃O₁₀)(OH)₂). In order to ensure homogeneous salt content of the saline soil, the collected saline soil was desalted with distilled water prior to the creation of the cured soil, dried and then the desalted soil was crushed and sieved through a 2 mm sieve and then prepared for use. In addition, compaction tests showed that the maximum dry density and optimum moisture content of FA stabilized saline soil were 1.63 g/cm³ and 12.5%, respectively. The FA complies with Class F (EN450, ASTMC618) since it has high CaO content (>10% reactive CaO) and a content of “SiO₂+Al₂O₃+Fe₂O₃” near 70%, and the oxide contents shown in Table 1, with XRD and SEM as shown in Fig. 1d, e. General-Reagent Company of China provided the Na₂SO₄ (powder, purity >99.0%). Xinjiang Zhongtai Chemical Co Ltd provided the NaOH (flaky, purity ≥98%). The water was laboratory-grade deionized water.

Fig. 1.

Raw materials; (a) SEM of saline soil, (b) SEM of FA, (c) XRD of saline soil, (d) XRD of FA, (e) FTIR of FA.

Table 1.

Chemical compositions of saline soil and FA (wt%).

Materials	SiO2	Al2O3	CaO	Fe2O3	MgO	K2O	Na2O	Cl	SO3
Saline soil	61.78	15.63	8.57	4.97	3.39	2.79	1.66	0.17	0.09
FA	37.31	17.38	13.14	15.07	5.95	1.28	2.60	0.27	5.23

Table 2.

Basic physical properties of saline soil.

	Specific gravity	Plastic limit (%)	Liquid limit (%)	Optimum moisture content (%)	Maximum dry density	Optimum water content (%)	Soluble saline content(wt%)
							Na2SO4	NaCl	Na2CO3
Saline soil	2.65 g.cm−3	10.2	18.7	12.5	1.894 g.cm−3	12.5	0.86	0.11	0

Specimen preparation

Moisture and FA accounted for 12.5% and 13% of the total mass, respectively, while Na₂SO₄ and soil accounted for 1% and 99% of the remaining material mass, and NaOH was estimated using the concentration mass, as shown in Table 3. NaOH and Na₂SO₄ were dissolved in warm ionized water to produce an alkaline activator, which was set aside. The FA and desalinated soil were then combined and manually swirled for approximately 20 minutes to ensure a homogeneous mixture. The alkaline activator solution was added to the mixture and continuously stirred until a homogeneous mixture was achieved. The amount of alkaline solution used was estimated based on the optimum liquid content determined during the standard compaction experiment. The soil samples were packed in a bag with sealed to ensure uniform distribution of water and saline.

Table 3.

Ratio of material composition of stabilized soil subgroups.

	NaOH	FA (%)	NaSO4 (%)	Water (%)	Soil (%)
N1	0.1 M	13	0.75	12.5	73.75
N2	0.3 M	13	0.75	12.5	73.75
N3	0.5 M	13	0.75	12.5	73.75
N4	0.7 M	13	0.75	12.5	73.75
N5	0.9 M	13	0.75	12.5	73.75

The mixture was poured into the metal mold in three equal layers, and the tops of each layer were scraped and compacted statically. The stabilized soil was then compacted, demolded, and shaped into specimens measuring Φ100 mm*100 mm in height. These specimens were sealed with cling film, marked, and cured for 28 days at a temperature of 20 °C. The stabilized saline soil with NaOH concentrations of 0.1, 0.3, 0.5, 0.7, and 0.9 M were designated as N1, N2, N3, N4, and N5, respectively.

Unconfined compressive strength tests

In strength measurements, at least three specimens are tested and averaged. The tests were performed using a cement flexural and compressive integrated machine (YAW-SERIES) for UCS tests on stabilized saline soil after 28 days of cured and FTs, with a loading rate of 1.0 mm/min. Where the UCS of stabilized soil was shown in Eq. (1):

1

P-Maximum pressure (N) of the specimen at the time of destruction; d-Diameter of the specimen (mm).

Splitting strength tests

The tests were performed using a cement flexural and compressive integrated machine for splitting strength tests on stabilized saline soil after 28 days of cured and FTs, with a loading rate of 1.0 mm/min, and the peak load at the time of specimen damage was recorded. Where the splitting strength is used in the non-compacted sheet method, the formula shown in Eq. (2):

2

P-Maximum pressure (N) of the specimen at the time of destruction; d-Diameter of the specimen (mm); h-Height of the specimen (mm).

Freeze-thaw cycles test

To examine the impact of FTs on the mechanical properties of alkali activated stabilized saline soil, laboratory-based FTs tests were carried out. According to the “Hydraulic Concrete Test Specification (SL352-2006)”, experiments were conducted on FTs without water replenishment using a constant temperature box. Stabilized soil specimens, which had been cured for 28 days, were subjected to freezing at − 20°C for 12 hours followed by thawing at +20 °C for 12 hours to obtain FTs=N (where N = 1, 3, 5, 10, and 20). The FTs tested specimens are presented in Table 4. The residual strength ratio (DBR) was used to evaluate the frost resistance of stabilized soil as shown in Eq. (3):

3

Table 4.

Grouping of FTs specimens.

	FTs tested specimens
UCS	FTs = 1	FTs = 3	FTs = 5	FTs = 10	FTs = 20
Splitting tests	FTs = 1	FTs = 3	FTs = 5	FTs = 10	FTs = 20

where q_p is the peak strength; q_r is the residual strength.

Microstructural test

After fragmentation by split testing, internal soil samples or alkaline activated FA products were collected and dried at 60°C for 24 hours to end the hydration reaction before being tested for mineralogical compositions and microstructures. X-ray diffraction (XRD) was conducted using a Bruker D8 Advance equipment with a scanning range of 10–80° 2θ in 0.02° steps and a counting time of 2 seconds per step. The microstructure and morphology of the hydration products of FA were observed using a scanning electron microscope (Hitachi S 4800) for specimens cured for 28 days and representative specimens with FTs = N. The samples were gold-sprayed twice prior to testing and were tested at 15 KV, 10 mA. Fourier Transform Infrared Spectroscopy (FTIR) was performed using the KBr press method in a Great 10 spectrometer with a material-to-KBr mass ratio of 1:200 and 32 scans from 400 to 4000 cm⁻¹ with a resolution of 4 cm⁻¹. TG measurements were conducted using differential scanning calorimetry-thermogravimetric analysis (TG-DSC) on an DSC200F3 instrument. Approximately 15 mg of the material powder was used for the measurement. The temperature ranged from 20 to 1000 ℃, and the heating rate was 10 ◦C/min in a air atmosphere⁵¹. The binding energies of a representative sample of FA hydration products were determined using an X-ray photoelectron spectrometer (Thermo Fisher Scientific ESCALAB 250 Xi) with a monochromated Al target at an energy of 1486.6 eV, a full-spectrum transmission energy of 100 eV in steps of 1.0 eV, and a narrow-spectrum transmission energy of 30 eV in steps of 0.05 eV. The process of preparation, preparation, moulding and testing of the specimens is shown in Fig. 2.

Fig. 2.

Specimen preparation and testing process.

Results

Compressive strength

The influence of alkali content on unconfined compressive strength

The UCS of saline soil stabilized with different amounts of NaOH-activated FA after 28 days of curing is presented in Fig. 3. The specimens were stabilized at room temperature using varying concentrations of NaOH-activated FA to stabilize the saline soil with sulfate. The strength variation observed was directly proportional to the concentration of NaOH.

Fig. 3.

UCS of FA-stabilized saline soil excited by various concentrations of NaOH for 28 days.

Figure 3 shows that there is a relationship between stabilized saline soil UCS and NaOH content. After 28 days of curing, the UCS values for N1, N2, N3, N4, and N5 stabilized saline soil were 4.85 MPa, 6.11 MPa, 7.18 MPa, 5.98 MPa, and 3.92 MPa, respectively. The UCS of stabilized saline soil initially increases and then decreases with increasing alkali concentration. The highest UCS value for N3 stabilized saline soil was achieved at a NaOH content of 0.5 M. These results indicate that a NaOH concentration of 0.5 M has the most favorable effect on the growth promotion of FA-stabilized saline soil. The reduced strengths of N1 and N5 can be attributed to two factors: the increased porosity of the stabilized soil and the decreased degree of polymerization of the alkali-activated FA gel. The strength of FA gel polymers mostly relies on the degree of polymerization⁵² which is influenced by the solubility of the silicate and aluminate monomers in the alkaline medium and their subsequent nucleation. Julphunthong et al²⁹ researched and found UCS 6.441 MPa after 28 days of curing in loess with 10 wt% of calcium carbide residue-FA stabilizer, which had similar results to the optimum strength of this experiment.

The influence of alkali content on splitting strength

Figure 4 shows the splitting strength of FA-stabilized saline soil after 28 d of cured at different NaOH-activation concentrations.

Fig. 4.

Splitting strength of FA-stabilized saline soil excited by various concentrations of NaOH for 28 days.

Figure 4 illustrates that the correlation between splitting strength and NaOH concentration was comparable to that between UCS and NaOH concentration. Both relationships exhibited a tendency to increase and then decrease. It was observed that N3 stabilized saline soil demonstrated the highest splitting strength. Furthermore, the impact of alkali-activated FA on the splitting strength of stabilized soil was more significant compared to the UCS test. The disparity between the UCS and splitting strength of stabilized saline soil primarily resulted from the formation of gel substances through the activation of sodium hydroxide-activated FA as a cementing agent. This process effectively enhanced the connection between soil particles and improved the splitting properties. The remaining gel materials fill the voids in the saline soil and improve the performance of the UCS. In order to increase the UCS of stabilized saline soil, it is necessary to use a cementitious material to bind the soil particles and ensure sufficient material to fill the pores. The results demonstrated that the impact of NaOH-activated fly ash on the splitting strength of stabilized soil was greater than that of UCS.

Freeze resistance experiments

Figure 5 shows the relationship between DBR and the number of FTs for UCS, as well as the splitting strength of FA-stabilized saline soil with varying NaOH-activation degrees.

Fig. 5.

Frost resistance of FA stabilized soil excited by different concentrations of NaOH; (a) DBR of UCS, (b) DBR of splitting strength.

Figure 5 shows that the DBR of stabilized saline soil steadily falls as FTs = N increases. After FTs = 5, the DBR of stabilized soil’s UCS and splitting strength experienced a significant decline, indicating a rapid deterioration stage. As the number of FTs increased , the DBR of stabilized soil’s UCS and splitting strength continued to decline, but at a slower rate, indicating a stable stage. Furthermore, the relative magnitude connections between UCS and DBR of splitting strength remained mostly unchanged among the groups, with the only difference being the degree of reduction in UCS and splitting strength in each group.

The DBR of N1, N2, N3, N4, and N5 stabilized saline soil UCS at FTs = 5 were 39.37%, 42.76%, 46.35%, 48.29%, and 42.82%, respectively. The DBR of stabilized soil UCS at FTs = 20 was approximately 25%. The splitting strength DBR at FTs = 5 were 51.26%, 48.15%, 39.9%, 52.30%, and 57.88%, respectively, whereas FTs = 20 stabilized saline soil splitting strength DBR by approximately 15%. The splitting strength decreased more in DBR than in UCS.

The decrease in DBR of UCS and splitting strength of stabilized saline soil specimens at a FTs of 20 was similar. However, it should be noted that the N3 stabilized saline soil initially had higher UCS and splitting strength compared to the other groups. This indicates that the N3 stabilized saline soil not only exhibited superior mechanical properties within the group, but also demonstrated exceptional resistance to frost.

Microanalysis

XRD analysis of FA hydration with time

Figure 6 shows the XRD of cured sulfate materials with different concentrations of NaOH activated FA for different times.

Fig. 6.

XRD of alkali-activated FA at different concentrations; (a) N1 activation 1, 3, 7 days, (b) N3 activation 1, 3, 7 days, (c) N5 activation 1, 3, 7 days, (d) N1, N3, N5 activation 28 days.

Figure 6 illustrates that the intensities of the diffraction peaks exhibit minimal variation, indicating that the phase compositions remain nearly identical at different ages of alkali activation. The activation of FA with different concentrations of NaOH results in noticeable diffuse peaks within the 2θ range of 20–40° at various ages, which correspond to the presence of amorphous C-A-H and C-S-H gels⁴⁰. However, the intensity of SiO₂ diffraction peaks decreased significantly over time. Figure 6a, b, and c shows diffraction peaks of C-S-H, C-A-H gels, Na₂SO₄, gypsum, unnamed zeolite, and quartz that were observed for all alkali-activation of FA at 1 day. Some of the primary peaks in FA have diminished, indicating that they have dissolved in the polymerisation reaction, resulting in an amorphous geopolymer gel. As the activation time was prolonged, the C-A-H and C-S-H diffraction peaks initially increased and then decreased, while the sodium sulfate diffraction peak disappeared. This indicates that NaOH activation of FA produces C-S-H and C-A-H gels, in which C-A-H reacts with Na₂SO₄ to form ettringite (AFt). These hydration products can fill the pore structure of the cured soil, raise the compactness of the cured soil to improve its UCS, and efficiently connect the cured soil particles to boost splitting strength^41,53. At 28 days of NaOH activation of FA, the diffraction peaks of SiO₂ decreased as the alkali concentration increased, while the amorphous gelation remained relatively unchanged. It is possible that the higher concentration of NaOH led to the dissolution of SiO₂ and Al₂O₃, but hindered the polymerization process of [SiO₄]⁴ and [AlO₄]⁴, resulting in a decrease in the SiO₂ diffraction peaks and minimal changes in the gel peaks. A concentration of 0.5 M NaOH facilitates the breaking and reorganization of the covalent bonds of silica-aluminum oxides, as well as the polycondensation reaction, leading to the continuous production of gel substances. However, when the NaOH concentration reaches 0.9 M, it slows down the activated FA process and inhibits the polymerization of [SiO₄]⁴ and [AlO₄]⁴. Figure 6d shows that N3 has the most pronounced decrease in SiO₂ diffraction peaks as well as the broadest gel peaks, indicating that the most gels were formed.N3 has more gel-encapsulated soil particles, which promotes particle cohesiveness and resistance to external loading during damage and FTs⁵³.

SEM analysis of FA hydration with time

Figure 7 shows SEM of N1, N3, and N5 at 1, 3, 7 and 28 days after FA activation. After one day of stimulation, N1, N3, and N5 (Fig. 7a, b, and c) exhibit a high number of needle-like AFt and scale-like C-(A)-S-H gel⁵⁴, as well as a modest amount of lamellar Ca(OH)₂⁵⁵. N1 shows more needle-like AFt, whereas N3 and N5 show more flocculent gels material. Consistent with the XRD observations in Fig. 6a, the concentration of 0.5 M rapid the production of gel material⁴¹.

Fig. 7.

SEM of alkali-activated N1, N3, and N5 at 1, 3, 7 and 28 days.

Figure 7d, e, and f displays the SEM images of N1, N3, and N5 after 3 days of FA activation at different alkali concentrations. It is evident that N1 exhibits a higher presence of needle-like AFt⁴⁹, whereas N3 and N5 exhibit a higher presence of Ca(OH)₂ and flocculent materials. The earlier emergence of Ca(OH)₂ indicates that larger concentrations have the effect of speeding up the hydration reaction⁵⁶. The SEM images of N1, N3, and N5 after 7 days of FA activation indicate an increase in short rod-like AFt for N1 and N3, while N5 show an increase in Ca(OH)₂ and flocculent materials. Furthermore, the SEM images of FA activation of N1, N3, and N5 over a period of 28 days demonstrate a thinning of the gel layer and a significant amount of needle-like AFt formation. Low NaOH concentrations have more short thick rods of AFt, as shown by the white circles in Fig. (j, k).

Figure 7a, d, g, and j illustrates that N1 exhibits the highest quantity of heterogeneous gels on the first day, which subsequently decreases over time. However, N3 and N5 display comparable conditions. The primary factor was the transformation of C-S-H and C-A-H into diminutive crystals, as well as the creation of AFt from the combination of C-A-H and sulfate. These findings align with the XRD results.

SEM of specimens with FTs

Figure 8 shows representative SEM images of N1, N3 and N5 after the stabilized soil FTs tests. As depicted in Fig. 8a, b, and c, the soil particles were tightly enclosed by the flocculated gel of alkali-activated FA, resulting in complete agglomerates. By filling the voids between the stabilized soil with FA and flocculated ground polymer gel products, the concentration of voids and exposed soil particles decreased. When FTs = 1, no cracks were observed in the SEM images of all stabilized soil samples (Fig. 8d, e, and f). However, when FTs = 3, microcracks started to appear in all stabilized soil samples⁵⁷, with N1 exhibiting the highest number of fractures (Fig. 8g), while N3 (Fig. 8h) and N5 (Fig. 8i) showed fewer internal cracks and less damage compared to N1. During the FT tests, the stabilized soil undergoes volume expansion due to the condensation of free water into ice crystals, resulting in significant pressure. This pressure dissipates when the ice crystals melt⁵⁸, leading to the fragmentation of some large soil aggregates, changes in the arrangement of soil particles, and cracks in the gel film and bonding materials. N1, having the lowest quantity of gel material, the least coagulation, and the largest gaps, is the most susceptible to freeze-thawing damage⁵⁹.

Fig. 8.

SEM of N1, N3, and N5 on FTs = 0, 1, 3 and 20 cycles.

Cracks in the stabilized soil continued to form as a result of FTs, and the soil sample particles adjusted to their equilibrium state. At FTs = 20, extensive gaps and cracks occurred in the region shown by the red circles in Fig. 8j, k, and l. The hydrated cementitious material continued to degrade, and cementitious material was shed from the surface of the particles in the area depicted by the red circle in Fig. 8J.

FTIR analyses of FA hydration with time

Figure 9 shows the FTIR of FA over time for various concentrations of NaOH activation. The alkali-activated FA at 28 days increased as the NaOH concentration increased, and the absorption peak at 3440 cm⁻¹ corresponded to an increase in -OH absorption peaks in C-S-H and C-(A)-S-H. This suggests that C-A-H and C-S-H increased with higher NaOH concentration. The absorption peaks ranging from 1390 to 1420 cm⁻¹ were associated with the expansion vibrations of C-O in carbonates⁶⁰. This is related to the carbonation of alkaline substances, some of which produce carbonates. The absorption peak observed at 1640 cm⁻¹ is associated with the bending vibration of the O-H bonds in molecular water. Additionally, this peak also corresponds to the presence of bound water in the reaction products C-A-H and C-S-H⁶¹. The absorption peaks around 1000 cm⁻¹ were asymmetric stretching vibrations of Si-O-T, which were mainly found in C-S-H and C-A-S-H gel⁴⁷. The Si-O-T absorption band shifts to higher bands with increasing excitation time, which indicates an increase in the degree of gelation polymerization. The absorption peak at 1005 cm⁻¹ shifted to higher wavelengths with increasing activation time, indicating a higher Si/Al ratio in the C-A-S-H gel⁶². The peaks observed at 528 cm⁻¹ and 469 cm⁻¹ are attributed to the bending vibration of the Si-O-Al (Mg) bond. SiO₂ exhibits significant absorption peaks in the range of 755 cm⁻¹ to 800 cm⁻¹. The absorption peaks observed at 660 cm⁻¹ to 670 cm⁻¹ correspond to the bending vibration of the Si-O-Si bond⁶³. As the concentration of NaOH increases, the absorption peaks of OH, H-O-H, Si-O-T, and Si-O become more pronounced³⁸. The Si-O-T band narrows, and the peak intensity increases. This could be because FA produces more coexisting aluminosilicate gels during the alkali reaction^64,65.

Fig. 9.

FTIR of alkali-activated N1, N3, and N5 at 1, 3, 7, and 28 days; (a) N1, (b) N3, (c) N5.

TG analyses of FA hydration

Figure 10a, b, and c illustrates the TG-DSC curves for N1, N3, and N5 activation over a period of 28 days. The mass loss was approximately 10%, and this mass loss increased as the NaOH concentration increased. As depicted in Fig. 10, the mass percentages of N1, N3, and N5 decreased by 1.34%, 1.08%, and 1.55%, respectively, within the temperature range of 20–100 °C. This decrease was primarily attributed to the loss of water resulting from FA adsorption and hydration products⁵³. The mass losses for the dehydration processes of the hydration products C-S-H, C-A-H, and AFt⁶⁶, which are bound to water, were 1.20%, 1.10%, and 1.53%, respectively, at temperatures ranging from 100 to 380 °C. Furthermore, it was observed that the first two mass losses of N1, N3, and N5 increased with higher NaOH concentrations^67,68. At 380–550 °C, exothermic oxidation of organic materials in FA resulted in a mass loss of around 5.50%. At 550–620 °C, a mass loss of 0.42%, 0.72% and 0.66% was detected, primarily due to the carbonate decomposition reaction⁶⁹.

Fig. 10.

TG-DSC of alkali-activated N1, N3, and N5 with 28 days; (a) N1, (b) N3, (c) N5.

The primary constituents of alkali-activated FA hydration products were C-S-H, C-A-H, and AFt. The strength of stabilized soil increased with a higher amount of gel and a greater degree of polymerization. The initial and secondary mass losses of N1 were lower compared to those of N3 and N5, indicating that the amount of C-S-H and C-A-H in the hydration products of FA activated with 0.1 M NaOH was lower than that of N3 and N5. Additionally, it was observed that N5 exhibited the fastest loss of the first two masses, suggesting that the overall polymerization of C-S-H and C-A-H formed by FA stimulated by a high NaOH concentration was moderate. This finding was consistent with the results obtained from XRD and FTIR analyses. The low gel volume for N1 and poor gel polymerization for N5 aligned with the low UCS and splitting strength observed in N1 and N5 stabilized soil.

XPS analyses of FA hydration

The XPS spectra of the alkali-activated FA gel are shown in Fig. 11. As can be observed from the spectra, the primary chemical element compositions were consistent with XRF. After NaOH activation of FA, the binding energies of O 1s, Ca 2p, S 2p, Si 2p, and Al 2p decreased before increasing with alkali concentration. This suggests that the activity of FA hydration products initially increases and subsequently diminishes as the concentration increases..

Fig. 11.

XPS of alkali-activated N1, N3, and N5 with 28 days; (a) full spectrum of N1, N3 and N5, (b) O 1s, (b) Ca 2p, (c) S 2s, (e) Si 2p, (f) Al 2p.

The Ca 2p orbital activation energy has two distinct peaks that correspond to the Ca 2p _3/2 orbitals (left) and Ca 2p _5/2 orbitals (right). The breakdown of the calcium-rich phase in FA results in a huge amount of free Ca²⁺, which interacts with groups like [SiO₄], [AlO₄] and SO₄²⁻ to form a new calcium-containing phase. The Ca 2p orbital activation energy has two distinct peaks that correspond to the Ca 2p _3/2 orbital (left) and the Ca 2p _5/2 orbital (right).The dissolution of the calcium-rich phase in the FA produces a considerable quantity of free Ca²⁺, which reacts with [SiO₄], [AlO₄] and SO₄²⁻, and other groups to form a new calcium-containing phase.

The activation energy of Si 2p and Al 2p orbitals exhibits an initial increase followed by a decrease due to similar reasons. There are two primary factors contributing to this trend. Firstly, the activation of silicon in FA leads to insufficient hydrolysis of Si-O and Al-O bonds due to low concentration, resulting in a limited decrease in binding energy. Secondly, an increase in the polymerization degree of silicon/aluminates also leads to higher activation energies of the orbitals. This suggests that both low and high concentration activation of hydrated silicon/aluminates are likely to result in the formation of longer polymerization chains or even shelf silicates. Consequently, the binding energies of Si 2p and Al 2p can be used to characterize the degree of polymerization of Si-O and Al-O. Therefore, the XPS analysis of the chemical bonding information on the FA surface partially reflects the surface’s structural information, although it is not entirely reactive⁷⁰.

Discussion

When NaOH come into touch with FA, the reliant SiO₂ and Al₂O₃ on its surface dissolve under alkaline circumstances, and the Si-O-Si, Si-O-Al, and Al-O-Al bonds break⁷¹, breaking down into monomers like Al(OH)₄⁻, Si(OH)₃⁻. The monomers are connected through -OH groups, subsequently dehydrated and condensed to produce oligomeric gel. These gels are then linked by cations to form a hydrated layer of FA. The presence of this hydrated layer on the FA surface decreases the contact between FA and the alkali solution. The OH⁻ ions diffuse through the pores of the hydrated layer to sustain the reaction with FA. The resulting monomers then polymerize on the surface of the hydrated layer, utilizing the pores within the hydrated layer⁷⁰. During the polymerization reaction, certain hydration products reacted with SO₄²⁻ to produce needle-like AFt. Additionally, the floating gel gradually dissipated. It was observed that in the initial stages of the reaction, dissolution became the rate-controlling step. However, after the FA surface was completely covered by hydration products, diffusion became the rate-controlling phase⁶⁷.

The curing process of saline soil using NaOH-activated FA is illustrated in Fig. 12. The alkali action caused damage to the macroscopic manifestation of the FA surface structure. The monomers produced during this process polymerized into gels such as C-S-H and C-A-H⁷². Gradually, the FA became enveloped in C-S-H and C-A-H gel materials, and it was also covered by hydration products. SEM analysis detected the presence of “hairy” phenomena. As equilibrium was established, the hydration cycle slowed down, and the C-S-H and C-A-H reacted with sulfate to form needle-like AFt.

Fig. 12.

Schematic diagram of the activation changes of alkali-activated FA.

XRD examination revealed that the primary hydration products of NaOH-excited FA were C-A-H, C-S-H, AFt, and carbonate. The findings of the TG-DSC investigation revealed that as the NaOH concentration grew, so did the gel products. After the stabilized soil had been cured for 28 days, alkali-activated FA was used to C-S-H and C-A-H gels. These gels connected the saline soil particles and filled the gaps between them. When the soil froze and thawed, the free water converted into ice crystals, which enlarged the gaps between the soil particles and disrupted the gel substance. These pores remained even after the ice crystals melted, leading to a continuous decrease in the strength of the stabilized soil. However, as the number of FTs increased, the arrangement of soil particles gradually approached equilibrium and the bonding between particles became more stable, resulting in a slower rate of strength loss. Consequently, the gel produced by alkali-activated FA improved the UCS, splitting strength, and frost resistance of the stabilized saline soil by filling the voids and reacting with Na₂SO₄.

Conclusions

In this paper, the macroscopic and microscopic characteristics of alkali-activated FA-stabilized saline soil in the seasonally frost zone are researched. By using alkali-activated FA materials to stabilize saline soil, the stabilized soil’s UCS, splitting strength, and FTs tests were researched, and the mechanical and frost resistance properties were analyzed. The microstructure of alkali-activated FA materials and stabilized soil was analyzed by XRD, SEM, FTIR, TG, and XPS experiments, and the stabilization mechanism was discussed. The key findings were as follows:

1. Alkali-activated FA significantly influenced the UCS and splitting strength of stabilized saline soil, with a greater impact on the latter. Lower alkali concentrations of activated FA led to the formation of fewer gel materials, while higher alkali concentrations of activated FA resulted in reduced polymerization of gel materials and decreased strength. The best mechanical properties of the stabilized soil were observed when the NaOH concentration was 0.5 M for the activated FA. The concentration of alkali-activated FA-cured saline soil has a significant impact on their microstructure, which in turn affects the mechanical properties and frost resistance of the cured soil specimens.

2. The freeze-thaw cycle reduced the strength of all samples. The mechanical strength DBR of stabilised soil steadily declined as FTs increased, and eventually stabilized at FTs = 5. The DBR of UCS and splitting strength at the same number of FTs first increased and then gradually decreased with increasing concentration, with N3 stabilized soil exhibiting the highest frost resistance. The use of NaOH-activated FA materials can effectively enhance the frost resistance of saline soil in regions with seasonal frost.

3. XRD demonstrated that the presence of alkali stimulated the formation of C-S-H, C-A-H, and hydrated gel products. These products developed over time and reacted with sulfate to form AFt. SEM analysis revealed that the C-S-H and C-A-H gels enhanced the splitting strength of the saline soil by encapsulating and connecting the soil particles. The gel filled the cavities in the soil, resulting in a denser structure and higher UCS.

4. The XRD, SEM, and FTIR data revealed that the addition of NaOH stimulated the formation of amorphous C-A-H and C-S-H gels in the stabilized soil, resulting in increased strength. XRD and TG-DSC analysis indicated that the N5 gel was the most abundant and had a lower degree of polymerization. XPS results demonstrated that the N3 gel exhibited the highest reactivity and performance in the cured sulfate.

5. The Xinjiang region possesses a significant amount of thermal power, resulting in a large quantity of FA that is not utilized effectively and has a detrimental impact on the environment. Choosing the appropriate concentration of alkali material to enhance FA stabilized of saline soil in the seasonally frozen zone would not only increase the viability of saline soil in the seasonally frozen zone, but will also lower the associated road construction expenses. This is critical for reducing solid waste emissions, protecting the environment, and utilizing saline soil in the seasonal frozen zone.

According to research, the concentration of alkali and FTs have a substantial influence on the microstructure of alkali-activated FA-stabilized saline soil, thereby modifying the mechanical properties of stabilized soil specimens. However, in the domain of chemically stabilized saline soil in the seasonally frost zone, several crucial factors such as stabilization time, saline content, freezing and thawing temperatures, and water content play a significant role in determining the effectiveness of stabilization. Although these factors have not been examined in this paper, we intend to investigate them in future research.

The datasets generated and/or analyzed during the current study are not publicly available due to the research is ongoing, but are available from the corresponding author on reasonable request.

Author contributions

Sining Li: Writing—review & editing, Writing—original draft. Yong Huang: Resources, Funding acquisition, Writing—review. Jian Sun: Supervision, Resources, Writing—review & editing. Qiushuang Cui: Visualization, Data curation. Rui Yu: Formal analysis, Data curation. Yubin Liu: Investigation, Formal analysis, Data curation.

Funding

This research was funded by the Major Science and Technology Project of the Xinjiang Production and Construction Corps Science and Technology Bureau [No. 204AA007], the Scientific and Technological Research Programs in key Areas of Xinjiang Production and Construction Corps Science and technology Bureau [No. 2023AB013-01], the Science and Technology Development Plan Project of the Innovation-driven Development Experimental Zone of the Silk Road Economic Belt and the National Independent Innovation Demonstration Zone of Urumqi-Changji-Shihezi [No.2023LQ03002], the Major Science and Technology Special Projects in Xinjiang Uygur Autonomous Region [No.2023A03004-04] and the Xinjiang Uygur Autonomous Region Science and Technology Department [No.2023B03011-3].

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to the research is ongoing, but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.