Effects of Dry–Wet Cycles on Permeability and Shear Strength of Yuanmou Red Clay

Abstract

Investigating the properties of red clay under the action of dry–wet cycles is crucial for mitigating geological disasters and promoting the sustainable development of geotechnical engineering infrastructure. In this paper, red clay from the Yuanmou dry-hot valley in Yunnan Province was selected as the research subject. The investigation focused on examining the effects of dry–wet cycles on its permeability and shear strength. Samples were prepared by controlling the initial moisture content (8%, 11%, 14%, 17%, and 20% for permeability tests; 11%, 14%, and 17% for strength tests) and initial dry density (1.65 g/cm³, 1.70 g/cm³, 1.75 g/cm³, and 1.80 g/cm³). We conducted variable-head permeability tests and direct shear tests on samples undergoing 1–5 dry–wet cycles. The results demonstrated that (1) the saturated moisture content decreased with the increasing number of dry–wet cycles, with the first cycle showing the most significant decrease (decreasing by approximately 15–25% depending on initial conditions). (2) The permeability coefficient decreased continuously with the number of cycles, exhibiting a transition behavior around the optimum moisture content (14%). Samples with lower initial moisture content (8–14%) showed higher permeability reduction (up to 40% decrease) compared to those with higher initial moisture content (14–20%). (3) The dry–wet cycles lead to a significant attenuation of the shear strength, and the first cycle has the largest reduction. The shear strength parameters of red clay exhibit distinct attenuation patterns. The cohesion decreased exponentially with the number of cycles (total attenuation ≈55–60%), and the internal friction angle decreased linearly (total attenuation ≈20–25%). The total attenuation of cohesion was much larger than the internal friction angle. (4) The degradation mechanism is essentially a multi-scale coupling process of cementation dissolution, pore collapse, and fracture expansion of red clay internal structure. These findings provide critical insights for sustainable engineering design and disaster prevention in regions with similar soil conditions, contributing to the resilience and longevity of infrastructure under changing climatic conditions.

1. Introduction

Yuanmou is located in a typical low-latitude plateau monsoon climate zone. Affected by topography and climate, its dry and wet seasons are distinct and highly contrasting. During the rainy season, precipitation is intense and abundant, while the dry season is characterized by prolonged durations of strong sunshine and significant evaporation [1,2]. As a special soil, red clay exhibits distinct engineering properties, including high moisture content, significant plasticity, and elevated porosity, as well as strong water absorption capacity, large shrinkage deformation upon water loss, difficulty in compaction and a tendency to produce cracks [3,4,5,6]. After experiencing high-frequency and large-scale dry–wet cycles, red clay undergoes continuous physical and chemical changes, making the soil structure vulnerable to damage. The macroscopic performance is the significant deterioration of shear strength and bearing capacity and the possible accompanying volume deformation, especially in the dry season. Red clay is highly sensitive to drought [7,8]. These changes alter the soil structure and lead to a significant degradation of its engineering properties, which poses a serious threat to the long-term service performance and stability of the project with red clay as the main roadbed filler or in natural slope conditions, and leads to the occurrence of various engineering geological disasters [9,10,11]. Thus, a thorough and comprehensive investigation into the evolution patterns and underlying mechanisms affecting the physical and mechanical properties of Yuanmou red clay under dry–wet cycle conditions is crucial for providing a solid scientific foundation. This research can guide the development of disaster prevention and mitigation strategies in geotechnical engineering for this region, ultimately supporting more sustainable infrastructure growth in ecologically sensitive areas.

The influence of dry–wet cycles on the engineering properties of soil is a significant focus of contemporary research, with numerous scholars conducting extensive studies in this area. In terms of permeability characteristics, Yuan discovered that an increase in the initial dry density resulted in a reduced initial permeability rate for loess. However, the permeability coefficient increased following dry–wet cycles [12]. Jing found that the dry–wet cycles altered the hydraulic pathways within loess, leading to a non-constant soil–water characteristic curve. It was recommended that this factor be incorporated into slope stability analyses. [13]. Zhou et al. explored the effects of dry–wet cycles on the release and migration of native colloidal particles by simulating the dry–wet cycles experiments of loess columns with different drying durations and cycles [14]. Ran et al. confirmed that the pore structure and permeability of granite residual soil undergo significant changes after dry–wet cycles [15]. Albrecht et al. found that the first cycle can lead to permanent cracking of clay and significantly increase permeability, and the subsequent impact was weakened [16]. Yang et al. investigated the effects of the number of dry–wet cycles and dry density on the permeability of Yunnan red clay using a variable-head permeability test. The findings indicated that an increase in the number of cycles and a higher dry density corresponded to a decrease in the permeability coefficient [17]. Chen et al. demonstrated that the quantitative relationship between the number of dry–wet cycles and the permeability coefficient of red clay can effectively be characterized by an exponential function. Additionally, they found that dry–wet cycles promote the development of surface cracks in saturated red clay [18]. Wan et al. noted that the permeability of compacted clay decreases as the pore structure changes following dry–wet cycles [19]. Ioke et al.’s study on expansive Yazoo clay demonstrated that an increase in the number of cycles leads to the expansion of dry cracks, resulting in a significant rise in the vertical permeability coefficient (Kv). While an increase in initial moisture content can reduce Kv, the impact of crack formation remains the predominant factor [20]. The study of Wang et al. on the four-layer soil of the collapsing hill further showed that the crack rate and permeability coefficient of the surface soil layer and the red soil layer increased with increasing numbers of cycles and then stabilized, and the two showed a quadratic function relationship. The permeability coefficient of the sand layer decreased, while the transition layer was relatively stable, and the influence of the crack on the permeability performance increased first and then decreased [21]. For compacted clay, Louati et al. found that the initial density significantly affected the early permeability through 12 cycles of tests combined with image analysis, SEM, and chemical detection. However, after seven cycles, the permeability coefficients of all samples tended to be consistent, which was related to about 7.5% of the fracture strength factor. The study also established a dual-pore fractal model to predict the permeability of fractured soils [22]. Previous studies have shown that dry–wet cycles typically result in an increase in the soil’s permeability coefficient, but in the case of Yunnan’s red clay, the permeability coefficient decreases after such cycles, suggesting that variations in soil types can lead to significant differences in outcomes.

In terms of the deterioration law and characterization of shear strength, a large number of studies have shown that dry–wet cycles lead to the general attenuation of soil shear strength parameters. Qi et al. observed through direct shear and triaxial tests that both the cohesion and internal friction angle of red clay significantly decreased with the number of cycles, with the first cycle showing the most severe reduction, and a plastic hardening model was established to characterize the stress–strain relationship after the cycle [23]. Chang et al. pointed out that under low stress conditions, the cohesion of shallow residual expansive soil can be attenuated to a very low range after dry–wet cycles and observed the phenomenon that the crack rate increases with the number of cycles. It was also noted that increasing the dry density helps to inhibit the development of cracks [24]. Using experimental data from granite residual soil, Ding et al. developed a power function prediction model for shear strength in relation to moisture content and the number of dry–wet cycles. The model showed that strength decreased as moisture content increased, with a rapid decline followed by stabilization as the number of cycles increased [25]. Xie et al. found that the crack rate of red clay increased with increased cycle times, and the volume shrinkage rate and cohesion decreased significantly with the increase in crack rate through synchronous shrinkage tests and direct shear tests [8]. Jimoh proposed that the strength parameters of laterite in the Ilorin area decrease linearly or parabolically with increasing moisture content and return to zero when the liquid limit is reached [26]. Chen et al. conducted suction-controlled direct shear tests on unsaturated compacted clay and found that the volume shrinkage effect caused by the first dry–wet cycle was the most significant. Under high normal stress, they observed a transition in the shear behavior of the soil, shifting from shear shrinkage to shear expansion as the number of cycles increased. This highlighted the influence of dry–wet cycle amplitude and vertical stress on the shear strength’s change behavior [27]. Liu et al. studied the cyclic mechanical behavior of silty clay and found that dry–wet cycles under low matric suction increased the cyclic stress and shear strength of soil, while this effect disappeared under high matric suction. The increase in loading frequency was also proved to be helpful to improve the cycle strength [28]. Luo et al. further revealed the key role of initial moisture content on the mechanical behavior of silty clay and determined that 17% was the critical initial moisture content of peak strength. It was observed that the dry–wet cycles promoted the stress–displacement curve to change from strain hardening to softening, and the strength parameters showed the evolution characteristics of first decreasing and then stabilizing [29]. In the study of Yili loess, Hao et al. found that the internal friction angle decreased significantly, the cohesion fluctuated, and the porosity increased [30]. The above studies generally show that the dry–wet cycles lead to the attenuation of cohesion and internal friction angle, but the attenuation mode and the stable threshold are affected by soil type, cycle number, moisture content, and other factors. Most of the existing research focuses on expansive soil and loess. There are few systematic studies on red clay in the special environment of dry-hot valley, especially on Yuanmou red clay in Yunnan under dry–wet cycles, and the properties of red clay in different regions are significantly different. In addition, research on the influence mechanism for the initial state of soil-on-soil characteristics under dry–wet cycles is not sufficient.

The shear strength and permeability of soil are two paramount properties in geotechnical engineering practice. The former dictates the resistance to failure, while the latter governs the seepage process and the subsequent evolution of pore water pressure during rainfall events. This study focuses on these two properties because the Yuanmou red clay, with its high porosity and susceptibility to cracking, is highly sensitive to moisture changes. The degradation of its shear strength, especially the cohesion component, serves as a direct and sensitive indicator of the structural damage induced by cyclic wetting and drying. Meanwhile, understanding the evolution of permeability is crucial for predicting the depth and intensity of the effects of dry–wet cycles. In the context of the Yuanmou dry-hot valley, where seasonal rainfall is intense and evaporation is strong, studying the evolution of these two indicators has an irreplaceable guiding role for disaster prevention and mitigation and engineering design. This paper focuses on red clay from the dry-hot valley of Yuanmou, Yunnan Province, as the research subject, conducting indoor experiments to investigate the effects of dry–wet cycles on the clay’s permeability and strength characteristics. By controlling the initial moisture content and dry density of the red clay, the study uses the South-55 variable-head permeameter and strain-controlled direct shear instrument to analyze these effects. The findings aim to provide a theoretical basis for engineering disaster prevention and the rational utilization of resources in this region.

2. Materials and Methods

2.1. Test Material

The red clay samples utilized in the study were sourced from Yuanmou County, Yunnan Province. The study area belongs to Yuanmou County, located in Santai Mountain and in gorges and mountains on both sides of the Jinsha River. The elevation is 1900~2835 m, and most of them form cliff steep slope topography. The slope gradient is mostly 35~45°, and the erosion structure developed in the east–west valley is high mountain landform. The exposed strata in the area are mainly Mesozoic ‘red beds in central Yunnan‘, followed by Proterozoic Kunyang Group metamorphic rocks, Sinian limestone, and a small amount of Jinning intrusive rocks. Cenozoic lacustrine and alluvial–diluvial strata are distributed in the basin. There are a large number of fault planes and fracture zones in the fault zone and its vicinity. The main structures are Yuanmou large fault, Yangjie syncline, and Huidan syncline. The linear structure of the fault is relatively simple. The overall trend is north–south and tends to the west. The fault is mainly developed along the boundary between Proterozoic (Pt) and Triassic (T) and strata. It is composed of several faults that are basically parallel and zonally distributed. The width is 2 ~ 4 km. The Proterozoic strata have obvious thrust uplift and are in fault contact with the caprock. The study area is located in the Longchuan River Valley, a tributary of the Jinsha River. It is a typical dry-hot valley area with less leeward rain and a warming effect. The geographical location and sampling points of the study area are shown in Figure 1. Refer to

Table 1. Basic physical property index of red clay.

Natural Density (g·cm−3)	Liquid Limit (%)	Plastic Limit (%)	Soil Classification	Specific Gravity	Optimum Moisture Content (%)	Maximum Dry Density (g·cm−3)
1.94	37	24	CL	2.707	14	1.85

for its fundamental physical and mechanical characteristics and Figure 2 for the particle size distribution curve.

2.2. Testing Program

2.2.1. Test Procedure

We crushed and sieved the red clay samples through a 2 mm sieve. Subsequently, we dried them at a constant temperature of 60 °C for 24 h. Then, we prepared the test red clay samples to achieve the predetermined initial moisture contents. A reconstituted sample, 40 mm in height and 61.8 mm in diameter, was created using the Proctor compaction test. The variable-head permeability test and direct shear test were carried out by controlling the initial water content and initial dry density of the sample. The testing program matrix is shown in

Table 2. Testing program matrix.

Test Type	Initial Dry Density (g/cm3)	Initial Moisture Content (%)	Number of Dry–Wet Cycles
Variable-Head Permeability Test	1.65, 1.70, 1.75, 1.80	8, 11, 14, 17, 20	0, 1, 2, 3, 4, 5
Direct Shear Test	1.65, 1.70, 1.75, 1.80	11, 14, 17	0, 1, 2, 3, 4, 5

. The specific test procedures are shown in Figure 3. To further elucidate the microstructural changes within the soil samples, the void ratio (e) and initial degree of saturation (Sr) were calculated for each test set based on the controlled initial dry density (ρ_d) and initial moisture content (ω). The calculation formulas of e and Sr are shown in (1) and (2), where Gs is the specific gravity of the soil solids, and ρ_w is the density of water (1.0 g/cm³). The calculated values are presented in

Table 3. Calculated initial void ratios and degrees of saturation for the testing program.

Initial Dry Density (g/cm3)	Initial Moisture Content (%)	Void Ratio e	Degree of Saturation Sr (%)
1.65	8	0.641	33.8
1.65	11	0.641	46.5
1.65	14	0.641	59.1
1.65	17	0.641	71.8
1.65	20	0.641	84.5
1.70	8	0.592	36.6
1.70	11	0.592	50.3
1.70	14	0.592	64.0
1.70	17	0.592	77.8
1.70	20	0.592	91.5
1.75	8	0.547	39.6
1.75	11	0.547	54.4
1.75	14	0.547	69.3
1.75	17	0.547	84.1
1.75	20	0.547	99.0
1.80	8	0.504	43.0
1.80	11	0.504	59.1
1.80	14	0.504	75.2
1.80	17	0.504	91.3
1.80	20	0.504	107.4

to facilitate a more comprehensive interpretation of the test results.

$$\mathrm{e}={\displaystyle \frac{Gs\times {\rho }_{\mathrm{\omega }}}{{\rho }_d}}-1$$

(1)

$$Sr={\displaystyle \frac{\mathrm{\omega }\times Gs}{e}}$$

(2)

2.2.2. Dry–Wet Process

This study investigated the impact of dry and wet cycles on red clay samples with varying initial moisture content and dry density. The simulation of the dry–wet cycles process was achieved through artificial humidification and oven dehumidification. The humidification process involves immersing the samples in water for 24 h after placing filter paper on both sides of each sample and securing a permeable stone on top. Subsequently, the samples are subjected to a drying process by baking them at 60 °C for 24 h. Figure 4 illustrates the fluctuation in moisture content throughout the dry–wet cycles (saturated moisture content is expressed by ω_sat).

3. Results and Analysis

3.1. Variation Law of Saturated Moisture Content Under Dry–Wet Cycles

Figure 5 illustrates the characteristics of saturated water content and dry–wet cycles of red clay under different initial water content conditions. It is evident that as the number of dry–wet cycles increases, the corresponding saturated moisture content decreases significantly. This suggests that the dry–wet cycles induce irreversible changes in the structure, thereby reducing the water-holding capacity in the saturated state. The primary cause of the decrease in saturated moisture content is an irreversible alteration of the pore structure. The drying process causes red clay to shrink due to water loss. During the subsequent wetting process, the softening and destruction of cementing materials between particles, along with the reorganization of the particle arrangement, prevent the red clay from fully recovering to its original loose state, resulting in irreversible volume shrinkage. As a result, the red clay becomes more compact, the total pore volume decreases, and the available space for water storage is reduced, directly leading to a decrease in saturated moisture content. Additionally, larger pores are more likely to collapse or transform into smaller pores with lower water-holding capacity during shrinkage and expansion, further exacerbating the reduction in saturated moisture content.

Figure 6 shows the characteristics of saturated water content and dry–wet cycles of laterite under different initial dry densities. The saturated moisture content is notably affected by the initial dry density and the corresponding void ratio. With the same number of dry–wet cycles, a lower initial dry density leads to an increase in the saturated moisture content. Additionally, as the number of dry–wet cycles increases, the reduction in saturated moisture content becomes more significant with a lower initial dry density. Samples with a low initial dry density have a relatively loose initial structure, larger pore space, and better connectivity. Consequently, irreversible compaction and macropore collapse are more likely to occur during the dry–wet cycles, leading to a more significant decrease in saturated moisture content. In contrast, samples with high initial dry density possess a compact initial structure, tightly arranged particles, low porosity, and finer pores. These samples exhibit higher structural stability and greater resistance to dry and wet disturbances, resulting in a smaller decrease in saturated moisture content.

For all samples with varying initial dry densities, the first complete dry–wet cycles resulted in the most significant decrease in saturated moisture content. This indicates that the structure is most sensitive to the initial dry–wet cycles, with the initial water loss and absorption processes causing the most substantial structural failure and reorganization. As the number of cycles increases, the saturated moisture content continues to decrease, but the rate of decrease gradually slows down. By the fourth and fifth cycles, the curve begins to flatten, suggesting that after several cycles, the structural changes become less pronounced, and the structure gradually reaches a new, relatively stable equilibrium. Consequently, the ability to undergo further structural changes diminishes with an increasing number of dry–wet cycles.

3.2. Variation Law of Permeability Coefficient Under Dry–Wet Cycling

Figure 7 illustrates the characteristics of saturated water content and dry–wet cycles of laterite under different initial water content conditions. It is clear that for a fixed initial dry density, the permeability coefficient of red clay steadily decreases with an increasing number of dry–wet cycles. Notably, the permeability before any cycle is highly dependent on the initial moisture content. The sample with the lowest moisture content (ω = 8%) shows the highest initial permeability coefficient, which is attributed to its strong tendency to desiccation and the subsequent formation of initial fracture networks upon preparation and initial drying. The degree of this reduction is affected by both the initial moisture content and the initial dry density. Taking the optimum moisture content (14%) as the cut-off point, the permeability coefficient values of samples with lower initial moisture content (8~14%) are higher and more dispersed, while those with higher initial moisture content (14~20%) are lower and more concentrated. The above phenomena are due to the different shrinkage deformation mechanisms of red clay under different moisture contents. The red clay with low moisture content has strong shrinkage potential, and it easily forms macro-fracture networks in the drying stage. In the subsequent humidification process, water migrates in the fracture and carries free fine particles, resulting in local blockage of pore channels, thus significantly reducing the permeability of the sample. On the contrary, high moisture content samples approach a saturation state, shrinkage deformation is restrained by pore water pressure, structural adjustment is mainly reflected in elastic compression of microscopic pores, and particle migration space is limited, so the permeability coefficient decreases relatively gently. With increasing cycle times, the repeated action of humidification and dehumidification causes continuous dissolution and recrystallization of soil particles, which weakens the effective cementation between particles and reduces the effective contact between particles. At the same time, the irreversible cumulative damage of the soil skeleton caused by cyclic action intensifies particle reorganization, resulting in the continuous degradation of pore connectivity and finally leads to the continuous decline in permeability.

Figure 8 shows characteristics of the permeability coefficient and dry–wet cycles of laterite under different initial dry densities. The permeability coefficient decreases as the initial dry density increases. In the lower initial dry density range (1.65–1.70 g/cm³), the permeability coefficient curve is scattered, with a steep slope, indicating that permeability is more sensitive to the changes in cycle times within this density range, and exhibits larger fluctuations. This is because soils with low dry densities have stronger water absorption capacities, making them more prone to water accumulation, which induces pore shrinkage and a decline in connectivity, resulting in a rapid decrease in permeability. When the initial dry density increases to a higher range (1.75–1.80 g/cm³), the permeability coefficient curve becomes more concentrated and smoother. This is primarily due to the high compactness and low permeability coefficient of high-density soils, which limits the effectiveness of particle migration and reorganization in the subsequent dry–wet cycles on the pore structure and permeability.

The analysis indicates that dry and wet cycles have a significant impact on the permeability of red clay samples. Specifically, the permeability coefficient decreases as the number of dry–wet cycles increase. The effect of dry density on permeability is most evident in the sample without dry and wet cycles, where the permeability coefficient decreases sharply as the dry density increases. However, as the number of dry and wet cycles increases, the decrease in permeability due to higher dry density becomes less pronounced. This suggests that a higher dry density has a mitigating effect on the reduction in permeability caused by dry and wet cycles. This is likely due to the compaction of the clay structure, which makes it less susceptible to the cyclical wetting and drying processes. As a result, the soil becomes more resistant to changes in permeability, maintaining a relatively stable structure despite repeated moisture fluctuations.

3.3. The Law of Changes in Strength Characteristics Under the Action of Dry and Wet Cycles

3.3.1. Characteristic Law of Shear Strength

Figure 9 illustrates how the shear strength of red clay changes as it undergoes multiple dry and wet cycles under varying initial moisture content and vertical stress conditions. The data show that shear strength in all samples with varying initial moisture contents significantly decreased after the first cycle. As the number of cycles increased, the rate of attenuation diminished. On average, each additional cycle caused a decrease of about 5–10% in shear strength. The first three dry and wet cycles had the most pronounced effect on strength deterioration, while the attenuation became noticeably weaker after the fourth and fifth cycles, suggesting that the shear strength had reached a relatively stable state. The drastic changes in moisture content during the first cycle triggered substantial reorganization of the fine structure, leading to considerable damage to the cementation between particles. This damage was evident in the macroscopic cracking observed in the specimens.

The shear strength increases with increasing vertical stress. However, the strength attenuation caused by dry–wet cycles is negatively correlated with vertical stress. Under low vertical stress, the strength attenuation is greater. Under the condition of high vertical stress, the attenuation amplitude is relatively small. This phenomenon is because of the high vertical stress. On the one hand, the initial strength is improved by increasing the contact stress and friction resistance between soil particles. However, on the other hand, it induces greater tensile stress during the drying shrinkage process, which easily causes crack expansion and aggravates damage accumulation. Therefore, although the initial strength is high under high stress, its ability to resist the deterioration of dry–wet cycles is relatively weak.

Under the same vertical stress and under identical dry–wet cyclic conditions, an increase in initial moisture content from 11% to 17% results in a significant reduction in shear strength, accompanied by an increase in its attenuation. As the primary factor contributing to the expansion and softening of red clay, moisture markedly weakens the interparticle bonding force. This leads to an increase in soil shrinkage during the drying and desiccation process, thereby intensifying the degree of cracking. Collectively, these factors enhance the degradation of shear strength.

3.3.2. Characteristic Law of Cohesion

Table 4. Cohesion and its attenuation rates with different initial moisture contents.

Number of Dry–Wet Cycles	Cohesion (KPa)			Percentage of Attenuation Compared with No Dry–Wet Cycle
	ω0 = 11%	ω0 = 14%	ω0 = 17%	ω0 = 11%	ω0 = 14%	ω0 = 17%
0	114.55	108	100.71	0%	0%	0%
1	82.84	77.5	71.27	27.68%	28.24%	29.23%
2	69.85	66.89	60.88	39.02%	38.06%	39.55%
3	58.95	56.59	52.64	48.54%	47.60%	47.73%
4	54.64	48.56	44.93	54.92%	55.96%	55.39%
5	47.12	43.52	40.52	58.87%	59.70%	59.77%

and Figure 10 show the numerical values, attenuation ratios, and change relationships of different initial moisture contents and dry and wet cycles on the cohesion of red clay under the initial dry density in the natural state (1.75 g/cm³). Table 4 illustrates that the cohesion consistently decreases as the number of dry and wet cycles increases, with the rate of decline gradually slowing down as the cycles progress, eventually reaching a stable level. The cohesion is significantly reduced during the first cycle, with an average decrease of 28.4%, accounting for 48% of the total attenuation of the five cycles. This suggests that most cohesion loss occurs early in the cycle process, with the rate of deterioration becoming less pronounced as the cycles progress. This phenomenon is primarily attributed to dehydration and dry shrinkage, which induce pore expansion, coupled with the limited hydrolysis of minerals within the soil sample. These factors collectively lead to the fracture of cementitious bonds, weakening the cohesive forces, and causing brittle failure. As the number of cycles increases, the decay rate from the second to the fourth dry–wet cycle gradually decreases, with the loss rate per cycle dropping from 15% to 5%. This fracture network has a double adverse effect. On the one hand, it provides a channel to accelerate water penetration, promotes the hydrolysis of clay minerals, and further weakens the cementation bond. On the other hand, it hinders the effective transmission path of stress through the soil matrix. The synergistic effect of these two processes results in the weakening of chemical bonds and the physical destruction of stress transfer, which further leads to a further significant decrease in cohesion, which is the reason for the secondary decay observed in the middle of the cycle. After five dry–wet cycles, the attenuation rate stabilizes at around 4% per cycle. At this point, the crack network propagation rate slows down, the pore distribution reconstruction is completed, and the mineral composition tends to become stable, which jointly promotes the attenuation of cohesion into a stable state.

Figure 10 clearly demonstrates the exponential decay pattern of cohesion as the number of dry and wet cycles increases. To quantitatively describe this trend, an exponential model fit was applied. The results of the data analysis indicate high correlation between the model and the experimental data, with R² values greater than 0.98 for both, suggesting that the selected model accurately captures the influence pattern between the two variables.

3.3.3. Characteristic Law of Internal Friction Angle

Table 5. Internal friction angle and its attenuation rate with different initial moisture contents.

Number of Dry–Wet Cycles	Internal Friction Angle (°)			Percentage of Attenuation Compared with No Dry–Wet Cycle
	ω0 = 11%	ω0 = 14%	ω0 = 17%	ω0 = 11%	ω0 = 14%	ω0 = 17%
0	24.64	22.67	20.53	0%	0%	0%
1	22.62	21.83	19.54	3.63%	8.21%	8.21%
2	21.83	20.41	18.05	11.40%	9.97%	12.08%
3	21.05	19.57	18.19	14.57%	13.67%	11.40%
4	19.56	18.66	17.69	20.62%	17.69%	13.83%
5	18.65	18.11	16.33	24.31%	20.11%	20.46%

and Figure 11 show the influence values, attenuation ratios, and variation laws of different initial moisture contents and dry–wet cycles times on the internal friction angle of red clay under the natural dry state of initial dry density. Table 5 demonstrates that the internal friction angles of the three soil sample groups with varying moisture contents decrease as the number of dry–wet cycles increases, with the rate of decline becoming smaller as the cycles continue. After the first cycle, the internal friction angle decreases most significantly, and the attenuation amplitude of the subsequent cycle gradually slows down. After five cycles, the internal friction angle decreases by 20~25%, and still shows a downward trend. The internal friction angle is mainly affected by the roughness and arrangement of particles. During the dry–wet cycles, water invades the soil skeleton gap, and its lubrication reduces the surface roughness of particles and promotes the adjustment of particle arrangement. The smooth effect of water flow on the surface roughness of particles and the adjustment of arrangement reduce the internal friction angle, but the rigid mineral skeleton in the soil maintains the effective occlusion between particles and limits the decrease in internal friction angle. Compared with the significant attenuation of cohesion, the decrease in internal friction angle is relatively small. In summary, the deterioration in shear strength of red clay under dry–wet cycles is mainly attributed to the sharp loss of cohesion.

Figure 11 clearly illustrates that the internal friction angle follows an approximately linear attenuation trend as the number of dry–wet cycles increases. To accurately predict this change, a linear model was applied for fitting. Data analysis confirmed a significant linear relationship between the two variables.

3.4. Analysis of Red Clay Degradation Mechanism Under Dry–Wet Cycles

The damage of red clay induced by dry–wet cycles is essentially a coupling process of multi-scale irreversible structural deterioration and cement decay, as conceptually illustrated in Figure 12. Mineral composition and pore water significantly affect the shear strength of red clay [31]. During the initial humidification process, due to the invasion of water, the cemented material between soil particles is partially dissolved. With the dissolution of the cemented material, the pores in the soil particles expand, and the free ions in the mineral composition migrate with water, resulting in mineral–fluid interaction [32]. After the drying process, the free ions migrate and are lost with the water, resulting in irreversible degradation of the cementation. With increasing numbers of cycles, the effective contact area of cementation decreases, the cementation material continues to be lost, and the macroscopic performance is the attenuation of cohesion. Correspondingly, in the drying stage, the pore–fracture network undergoes an irreversible evolution, as depicted in the upper part of Figure 12. The drying shrinkage stress causes the large pores to collapse into micropores, blocking the particle stress transfer path, and the macroscopic performance is the attenuation of the internal friction angle. At the same time, the expansion of the fracture accelerates the migration of small particles, providing sufficient space for the migration of small particles, resulting in the blockage of the pore channel and a decrease in the permeability coefficient [23]. With the accumulation of the cycle, the internal structure of the soil is irreversibly reorganized. As visualized in the lower section of Figure 12, the large soil particles are broken into small particles, resulting in a decrease in strength. The particles are reorganized into a new equilibrium structure, and the strength tends to become stable. At the same time, in this process, the network pore channel is formed inside the soil particles, which weakens the influence of the number of cycles on the change in pore structure. This process is affected by the initial state. The initial structure of the low dry density sample is loose, the distribution of macropores increases, and the pores become more sensitive to drying shrinkage. The structural reorganization space is large, and the anti-deterioration ability is relatively weak, while the initial structure of high dry density is dense and the anti-deterioration ability is strong. When the initial moisture content is high, the corresponding pore water pressure is significant, which inhibits the shrinkage of soil particles and slows down the damage. When the initial moisture content is low, the shrinkage potential of soil particles is strong, which is conducive to the development of cracks.

3.5. Comparative Analysis with Previous Studies and Limitations

The findings of this study on the evolution of permeability and shear strength of Yuanmou red clay under dry–wet cycles both align with and diverge from the results of previous research on other clayey soils, highlighting the significant influence of soil type and initial state.

Regarding permeability, the observed continuous decrease in the permeability coefficient of Yuanmou red clay with increasing dry–wet cycles contrasts with the common trend reported for many other soils, such as loess [12], expansive Yazoo clay [20], and surface layers of collapsing hills [21]. This divergence can be primarily attributed to the unique initial structure and mineral composition of the tested red clay. Unlike highly crack-prone expansive soils or loess, the structural adjustment in this red clay during cycles is dominated by pore compression and particle reorganization rather than macroscopic cracking, leading to a reduction in pore connectivity and permeability. This result is consistent with findings by Yang et al. [17] on Yunnan red clay and Wan et al. [19] on compacted clay, suggesting that for certain compacted or specific clay types, densification and pore clogging due to particle migration can outweigh the crack-induced increase in permeability. Furthermore, the inhibitory effect of high initial dry density on permeability attenuation observed in this study echoes the findings of Louati et al. [22], who noted that initial density significantly influences early permeability, but its effect diminishes after multiple cycles as the soil structure reaches a new equilibrium. Concerning shear strength degradation, the significant attenuation, particularly the exponential decay of cohesion and linear reduction in the internal friction angle (Figure 8 and Figure 9), is a consensus shared by numerous studies on various soils under dry–wet cycles [8,23,24,25]. The most pronounced strength reduction occurring during the first cycle aligns with observations by Chen et al. [27] on compacted clay and Albrecht et al. [16] on natural clays, underscoring the irreversibility of initial structural damage. The established exponential model for cohesion decay and linear model for internal friction angle reduction provide quantitative relationships that are valuable for prediction. However, the extent of attenuation observed here (cohesion reduction of 55–60%) is notably severe compared to some studies on other red clays or residual soils [25,29], which may be attributed to the particularly strong shrinkage potential and cementation sensitivity of the Yuanmou red clay from the dry-hot valley environment. In summary, while the general trend of strength degradation aligns with the broader literature, the permeability response of Yuanmou red clay demonstrates a distinct behavior, decreasing with cycles rather than increasing. This highlights the critical role of soil specificity and initial conditions. Our study systematically quantifies these effects specifically for Yuanmou red clay, considering the coupling effect of initial moisture content and dry density; this is an aspect less discussed in previous studies of similar materials. The multi-scale degradation mechanism proposed integrates these findings and provides a conceptual model that can help explain the divergent behaviors observed in different soils under similar cyclic conditions.

However, this study has some limitations. Due to the rapidity of the direct shear test, it was impossible to distinguish between true and apparent cohesion. The mechanism is inferred from the macroscopic experimental results and their consistency with the established literature, and the research on the mechanism lacks the verification of microscopic experiments. The properties of red clay in five cycles were studied, and the change law of more cycles was not studied. If more cycles are performed, its physical and mechanical properties may exhibit other characteristics.

4. Conclusions

In this study, red clay samples were subjected to multiple dry–wet cycles, and variable-head permeability and direct shear tests were performed. The initial moisture content and dry density of the samples were carefully controlled. The study systematically examined the impact of the number of dry–wet cycles on the saturated moisture content, permeability coefficient, shear strength, and other related parameters. The key conclusions are as follows:

(1) The dry–wet cycles lead to decreases in the saturated moisture content of red clay, and the magnitude of this decrease decreases with increasing initial dry density. The first cycle decreases significantly, and the later stages gradually tend to become stable.

(2) The dry–wet cycles lead to a continuous decrease in the permeability coefficient of red clay. Taking the optimal moisture content (14%) as the mutation point, the decrease is slow when the moisture content is 14%. This change is due to the difference in void ratio and channel structure under different moisture contents. The permeability coefficient decreases with increasing initial dry density, and the increase in initial dry density has an inhibitory effect on the permeability attenuation caused by dry–wet cycles.

(3) The dry–wet cycles significantly deteriorate the shear strength of red clay, and the deterioration continues to accumulate with increasing cycle times. The first cycle attenuation is the largest, and then gradually slows down and tends to become stable. The strength parameters decrease with the increase of dry–wet cycles. The cohesion shows an exponential decay law and finally tends to become stable, with a total attenuation of about 55~ 60%. The internal friction angle shows a linear attenuation and a continuous attenuation trend, with a total attenuation of about 20–25%.

(4) The damage mechanism is essentially a multi-scale coupling process of cementation dissolution–pore collapse–fracture expansion in the red lay internal structure. Cementation dissolution causes the attenuation of cohesion, and the combined effect of pore collapse and fracture network leads to attenuation of the internal friction angle and deterioration of permeability. The damage sensitivity of samples with low dry density and low moisture content is significantly increased due to their strong shrinkage potential and high degree of fracture development.

(5) In the future, undrained triaxial testing can be carried out to obtain shear stress and strain data, and the simulation study of slope can also be carried out. In the meantime, the number of cycles can be extended to reveal the long-term behavioral effects.

The insights gained from this study are instrumental for promoting sustainability in geotechnical engineering. By understanding the degradation mechanisms of red clay under dry–wet cycles, engineers can develop more resilient and durable infrastructure designs that require less maintenance and fewer resources over their lifecycle. This contributes directly to the sustainable development goals of reducing environmental impact and enhancing the adaptability of infrastructure to climate change.