Introduction

In the construction of modern engineering projects, such as hydropower stations, roads, bridges, highways, and civil buildings, the number of slopes has increased significantly. The stability of slopes is of paramount importance. Unstable slopes may lead to geological disasters such as landslides and rockfalls1. Landslides are primarily associated with geological and geomorphological conditions, internal and external forces, and human factors2,3,4. The geological conditions of landslides are mainly characterized by the reduction in shear strength of rock masses, inconsistent hardness between rock layers, and the development of fractures and even faults within rock masses. Additionally, the erosive effects of hydrological conditions have the most significant impact on slope instability. The periodic hygrothermal cycles caused by the rise and fall of groundwater levels, natural precipitation, and solar drying can lead to slope instability5,6,7. The aqueous environment is usually not pure water but is predominantly a sulfate-rich environment. During the hygrothermal cycles, salt crystallization can lead to changes in rock mass porosity, which gradually evolves into fracture development and subsequently affects rock strength8,9,10,11. This is especially true for sandstone, a rock with a higher degree of porosity. Therefore, it is particularly important to study the evolution of pore structure in sandstone after hygrothermal cycles in a saline environment. Previous studies have indicated that under hygrothermal cycles in pure water-rock conditions, the damage to sandstone primarily occurs through hydrolytic corrosion4,12,13,14,15. In aqueous environments, calcite, albite, and orthoclase within sandstone undergo varying degrees of hydrolysis, leading to an increase in porosity16,17,18. When salts, such as sodium salts and sulfates, are present in the aqueous environment, the damage to sandstone is further exacerbated. During hygrothermal cycles in saline environments, crystallization evolves through discrete crystallization, circumferential crystallization, and crystallization expansion stages. Anhydrous sodium sulfate absorbs water to form decahydrate sodium sulfate, which upon heating, loses its crystallization water and reverts back to anhydrous sodium sulfate19. The phase transformation between anhydrous sodium sulfate and decahydrate sodium sulfate intensifies the porosity of sandstone, promoting the development of secondary pores from primary pores20. The pore structures of sandstone are predominantly tubular and columnar. During the wetting process, soluble salts can infiltrate through capillary and gravitational actions. When the crystallization pressure exceeds the tensile strength of the pore walls, microcracks begin to form. The re-crystallization of salt solutions within these microcracks further enlarges the crack dimensions, leading to the further development of secondary pores21,22,23. Currently, the observation of rock mass porosity mainly involves qualitative physical methods and quantitative analysis. Qualitative methods primarily utilize optical microscopy, CT scanning, and scanning electron microscopy. Quantitative analysis methods include nuclear magnetic resonance (NMR), nitrogen adsorption, and mercury intrusion porosimetry (MIP). Each of these quantitative methods has its own advantages. NMR allows for convenient and non-destructive testing of samples, but it relies on empirical formulas for pore size and type distribution, which introduces a degree of subjectivity and makes it difficult to characterize pores smaller than 65 nm24,25,26. Nitrogen adsorption experiments use nitrogen as a carrier to calculate pore sizes based on gas adsorption in the sample. This method is highly accurate for pores in the range of 2–100 nm, with effective measurement typically below 100 nm27,28,29. Mercury intrusion porosimetry measures pore sizes and distributions by applying external pressure to overcome the surface tension of the sample, allowing mercury to enter the pores. This method is generally used for samples with larger pores15,30.

In previous studies on the damage caused by dry-wet cycles to rock masses, researchers mostly focused on the microscopic changes of the rock masses, but failed to effectively integrate these microscopic changes with the damage mechanism. The relationship between the evolution of pore characteristics and the damage mechanism of sandstone by sodium sulfate under high cycling frequencies is still not clear. Therefore, in this experiment, sandstone samples were immersed in Na2SO4 solutions with concentrations of 28.01 mg/L, 40.38 mg/L, and 136.44 mg/L and subjected to 0, 10, 20, 30, 40, and 50 hygrothermal cycles. After every ten cycles, the sandstone samples were tested using nuclear magnetic resonance (NMR), mass, wave velocity, and hardness measurements to better understand the changes in pore distribution and basic physical properties of sandstone after hygrothermal cycles in a saline environment. This experimental study will help us better understand the pore characteristics of sandstone under hygrothermal cycles in a saline environment and provide an experimental basis for the construction and protection of slope engineering.

Experimental materials and methodology

Experimental materials

The un-weathered sandstone used in this study is obtained from a slope in Wanzhou District, Chongqing, China. The sandstone has dense fine-grained clastic structure, good homogeneity and no obvious fractures. X-ray diffraction analysis shows that the main minerals of this sandstone are quartz (73.6%) and albite (21%), as shown in Fig. 1. The mineral particle diameter of sandstone is between 0.25 mm and 0.5 mm, which is defined as medium grained quartz sandstone. The sandstone is processed into cylinder with a size of 50 mm (diameter)× 25 mm (height) according to the recommended code of ISRM (Fig. 2). Simultaneously, 5 water samples are collected adjacent to the sandstone sampling site and analyzed for their ionic composition. The content of anions and cations in the obtained water sample is shown in Table 1. This article selects Na2SO4 solutions that meet the natural background conditions for hygrothermal cycle experiments, and prepares three different concentrations of Na2SO4 solutions with concentrations of 28.01 mg/L, 40.38 mg/L, and 136.44 mg/L, respectively. In addition, distilled water is selected as the control group.

Fig. 1
Fig. 1
Full size image

X-ray diffraction pattern and mineral compositions of sandstone.

Fig. 2
Fig. 2
Full size image

Macroscopic and microscopic photos of sandstone samples.

Table 1 Characteristics of water samples.
Full size table

Experimental methodology

The main cyclic hygrothermal process in this experiment is shown in Fig. 3. Firstly, the sandstone samples were placed into beakers containing different concentrations of Na2SO4 solutions and distilled water at room temperature for 1 h. Subsequently, the soaked samples were placed in an electrically heated air-drying oven for drying. Considering that the temperature of the rocks under natural sunlight is around 60℃, the target temperature was set at 60℃ and the drying time was 1 h. Finally, the samples were naturally cooled to room temperature. The above process is defined as one hygrothermal cycle. In this experiment, 50 hygrothermal cycles were set up, in which the weight, P-wave velocity and NMR of the samples were tested after every 10 cycles to reveal the effect of hygrothermal environment on the damage and degradation of sandstone pore structures in the condition of sulfate solution.

Fig. 3
Fig. 3
Full size image

Experimental scheme of sandstone subjected to cyclic hygrothermal process.

The pore structure characteristics of sandstone after different hygrothermal cycles can be obtained through the \({T_2}\)spectrum of nuclear magnetic resonance (NMR). Equation (1) shows the relationship between the pore radius of rock and its transverse relaxation time \({T_2}\)31:

$$r=\rho {F_s}{T_2}=C{T_2}$$
(1)

In the formula, r represents the pore radius, in micrometers (µm); \({T_2}\) is the transverse relaxation time, in milliseconds (ms); \(\rho\) is the transverse surface relaxation intensity, in µm/ms; \({F_s}\) is the pore shape factor; C is the conversion coefficient, in µm/ms, with a value range of 0.01–0.15 μm/ms. In this paper, C is taken as 0.02 μm/ms.

The porosity of saturated rock samples is directly proportional to the NMR signal. According to the research theory of Morriss et al. 2020, the porosity variation of different samples can be obtained through Eq. (2):

$$\varphi =\frac{{M(0)}}{{{M_{100\% }}(0)}}=\sum {\frac{{{M_{0i}}}}{{{M_{100\% }}(0)}}} =\Sigma {\varphi _i}$$
(2)

In the formula: \(\varphi\)represents the total porosity; \(M(0)\) is the total signal intensity of the sandstone sample; \({M_{100\% }}(0)\) is the signal intensity produced by the same volume of pure water; \({M_{0i}}\) is the signal intensity of the i-th type of pore; \({\varphi _i}\) is the porosity of the i-th type of pore.

Results and discussion

Influence of salt content on sandstone mass under hygrothermal cycles

Figure 4 shows the variation of the relative mass of sandstone samples with the number of hygrothermal cycles. It can be clearly seen from the figure that the overall change patterns of the curves in three different concentrations of Na2SO4 solutions are relatively close. With the increase in the number of cycles, the mass loss of each rock sample gradually increases, and the mass variation amplitude also increases with the increase in the concentration of Na2SO4 solution.

When the number of hygrothermal cycles reached 10, the mass of sandstone samples increased by 0.2% after exposure to a 28.01 mg/L Na2SO4 solution, decreased by 0.19% after exposure to a 40.38 mg/L Na2SO4 solution, and decreased by 1.32% after exposure to a 136.44 mg/L Na2SO4 solution. When the number of hygrothermal cycles reached 50, the mass loss rate of sandstone samples exposed to a 136.44 mg/L Na2SO4 solution reached 4.58%. This indicates that after 50 cycles of hygrothermal action, the internal fissures of the sandstone samples continuously expanded and extended, leading to the phenomenon of fine debris spalling. This is mainly due to the hydrolysis of K-Na feldspar, the hydrosol swelling of clay minerals such as kaolinite, and the dissolution of soluble minerals16.

Fig. 4
Fig. 4
Full size image

Variation of weight under hygrothermal cycling.

Influence of salt content on sandstone hardness under hygrothermal cycles

Figure 5 shows the average surface hardness of sandstone samples after 50 hygrothermal cycles. It can be observed from the figure that the hardness of sandstone subjected to Na2SO4 solution can be divided into two stages: a slow change stage (0–20 cycles) and a rapid decrease stage (20–50 cycles). In the first stage (0–20 cycles), the hardness of sandstone subjected to different concentrations of Na2SO4 solution changes differently. For example, the hardness of sandstone subjected to 28.01 mg/L Na2SO4 solution decreases slowly with the increase in the number of hygrothermal cycles, and it decreases by about 1.5% at the 20th cycle. The hardness of sandstone subjected to 40.38 mg/L Na2SO4 solution reaches its maximum at the 10th cycle, increasing by 0.35% compared with the initial value, and then gradually decreases. The hardness of sandstone subjected to 136.44 mg/L Na2SO4 solution reaches its maximum at the 20th cycle, increasing by about 0.6% compared with the initial value. In the second stage (20–50 cycles), the hardness of sandstone decreases rapidly. When the number of cycles reaches 50, the hardness of sandstone subjected to the three concentrations of Na2SO4 solution decreases to 89.7%, 90.24%, and 92% of the initial value, respectively, all of which are lower than that subjected to pure water (97.9%). The main factors causing the change in the surface hardness of Wanzhou sandstone are the crystallization stress caused by mineral crystallization and the softening effect of water, which leads to the surface32 looseness33 and hardness34 reduction of mudstone35.

Fig. 5
Fig. 5
Full size image

Variation of hardness change rate under hygrothermal cycling.

Influence of salt content on sandstone pores under hygrothermal cycles

The hygrothermal cycles in saline environments can alter the pore structure characteristics of sandstone, such as pore size, quantity, and distribution, thereby affecting its strength. The T2 distribution of nuclear magnetic resonance (NMR) can reflect the pore structure of sandstone through the shape of the curve. The relationship between the transverse relaxation time T2 and the pore radius r can effectively analyze the pore structure characteristics of sandstone. Combining previous studies and this research, the pores in sandstone are divided into three size ranges: micropores (r < 0.1 μm), mesopores (0.1 μm < r < 1 μm), and macropores (r > 1 μm)36.

Fig. 6 show the pore size distribution curves of sandstone after 50 hygrothermal cycles under different concentrations of Na2SO4 solution. It can be observed from the figures that the pore size distribution curves of sandstone have three main peaks, with Peak 1 in the micropore range (r < 0.1 μm), Peak 2 in the mesopore range (0.1 μm < r < 1 μm), and Peak 3 in the macropore range (r > 1 μm). Under the conditions of 28.01 mg/L, 40.38 mg/L, and 136.44 mg/L Na2SO4 solution, as the number of cycles increases, the number of pore size distribution peaks increases from three to four. This indicates that under higher concentrations of Na2SO4 solution, new pores are formed and continuously develop within the sandstone as the number of hygrothermal cycles increases. Small pores gradually expand into larger pores. Through rock porosity testing, it is found that there is a negative correlation between porosity and hardness. When the internal structure of the rock is loose, that is, when the internal structure is full of pores and fractures, the plasticity is enhanced, the elasticity is weakened, and the hardness is reduced.

Fig. 6
Fig. 6
Full size image

T2 distribution and pore diameter distribution curves of sandstone under different hygrothermal cycles.

Figure 7 more intuitively shows that with the increase in the number of cycles, the number of micropores in sandstone gradually decreases, while the number of mesopores and macropores increases. This trend is most evident under the condition of 136.44 mg/L Na2SO4 solution. In Fig. 7(c), in the initial state, the proportions of mesopores and macropores in sandstone are 23.96% and 12.78%, respectively, while micropores account for 63.26%. After 50 cycles, the proportions of mesopores and macropores increase to 24.28% and 15.74%, respectively, while the proportion of micropores decreases to 59.98%. After 50 cycles, the higher the concentration of Na2SO4 solution, the smaller the proportion of micropores in sandstone. (a) 28.01 mg/L Na2SO4, (b) 40.38 mg/L Na2SO4, (c) 136.44 mg/L Na2SO4.

Fig. 7
Fig. 7Fig. 7
Full size image

T2 distribution and pore volume proportion of sandstone under different hygrothermal cycles. (a) 28.01 mg/L Na2SO4, (b) 40.38 mg/L Na2SO4, (c) 136.44 mg/L Na2SO4, (d) distilled water.

Figure 8 shows the variation of average porosity of sandstone samples with the number of cycles under different concentrations of Na2SO4 solution. It can be observed from the figure that the porosity of sandstone in pure water hardly changes, while under different concentrations of Na2SO4 solution, the porosity generally shows a phased change of first decreasing and then increasing with the increase in the number of cycles. The porosity of sandstone subjected to 28.01 mg/L, 40.38 mg/L, and 136.44 mg/L Na2SO4 solution reaches its minimum at the 10th, 20th, and 30th hygrothermal cycles, respectively, and then gradually increases. After 50 cycles, the porosity of sandstone has increased compared with the initial value, and the higher the concentration of Na2SO4 solution, the more significant the increase in porosity. This indicates that sandstone in Na2SO4 solution has undergone significant salt crystallization, recrystallization, and water-salt reactions that cause the decomposition of some minerals6. The damage mechanism is shown in Fig. 937,38,39,40.

Fig. 8
Fig. 8
Full size image

Variation of sandstone porosity with number of cycles under different solutions.

Fig. 9
Fig. 9
Full size image

Damage mechanism of sandstone after hygrothermal cycles under salt erosion.

Influence of salt content on sandstone wave velocity under hygrothermal cycles

The P-wave velocity of sandstone is closely related to its internal microstructure. Under saline conditions, the pore and fracture structures of rocks change with the concentration of the salt solution and the number of cycles of salt solution action. Therefore, analyzing the variation of P-wave velocity in sandstone after different concentrations of Na2SO4 solution and different hygrothermal cycles can help further understand the characteristics of sandstone’s pore and fracture structures.

Figure 10 shows the relationship between the P-wave velocity of sandstone samples and the number of hygrothermal cycles under different concentrations of Na2SO4 solution. It can be seen from Fig. 10 that the variation of P-wave velocity shows two distinct stages with the increase in the number of hygrothermal cycles: an increasing stage (0–20 cycles) and a decreasing stage (20–50 cycles). In the increasing stage, the P-wave velocity of sandstone treated with 28.01 mg/L Na2SO4 solution increased from 2.632 km/s to 2.966 km/s, an increase of 12.7%; that treated with 40.38 mg/L Na2SO4 solution increased from 2.427 km/s to 3.018 km/s, an increase of 24.3%; and that treated with 136.44 mg/L Na2SO4 solution increased from 2.294 km/s to 3.192 km/s, an increase of 39.1%. It can be concluded that with the increase in the concentration of Na2SO4 solution, the P-wave velocity of sandstone correspondingly increases, indicating that during this process, the Na2SO4 solution in the sandstone pores crystallizes into salt and fills the pores during the drying process, increasing the compactness of the sandstone. As the number of hygrothermal cycles further increases, the P-wave velocity of sandstone begins to gradually decrease. At this time, the degree of salt crystallization gradually intensifies, accumulating inside the sandstone and generating increasing pressure41. When the surface minerals are eroded away, the salt solution seeps into the interior of the sandstone and undergoes deeper crystallization, continuously exacerbating the damage and destruction of the rock, thereby leading to a decrease in wave velocity21, such as Fig. 9.

Fig. 10
Fig. 10
Full size image

Distribution graph of wave velocity with respect to the number of cycles.

s

Conclusion

This study employs testing methods such as wave velocity, hardness, colorimetry, and low-field nuclear magnetic resonance to analyze the surface characteristics and pore structure changes of sandstone after hygrothermal cycles under different concentrations of Na2SO4 solution. The following main conclusions are drawn:

  • The mass loss of sandstone mainly shows an increasing trend with the increase of sodium sulfate solution concentration and the increase of the number of cycles. Affected by the salt concentration, the mass loss rates are 4.58%, 1.18%, and 1.06% for Na2SO4 solution concentrations of 136.44 mg/L, 40.38 mg/L, and 28.01 mg/L, respectively.

  • The hardness can be divided into two stages: a slow decrease (0–20 cycles) and a rapid decrease (20–50 cycles). There is a negative correlation between hardness and porosity. Porosity is affected by salt content. High Na2SO4 content promotes the development of rock pores, leading to a decrease in hardness.

  • The P-wave velocity exhibits two distinct stages of variation, reaching its peak at the 20th cycle, with increases of 12.7%, 24.3%, and 39.1% corresponding to the different salt concentrations.

  • The rock damage caused by the hygrothermal cycle mainly affects sandstone through saltwater erosion and stress concentration resulting from recrystallization. Its characteristic is a transition from micropores to more mesopores and macropores, thereby increasing the overall porosity.