Effect of fulvic acid contamination on the shear strength and microstructural evolution of red clay

Abstract

The influence of fulvic acid (FA) contamination on the residual shear strength of red clay is an important issue in geological engineering. This study examines the effects of FA contamination on the shear strength, mineral composition, and microstructural changes in Guilin red clay using drained repeated direct shear test, along with X-ray diffraction (XRD), scanning electron microscopy (SEM), and particle and pore structure analysis system (PCAS). The results indicate that a low FA concentration (m_FA/m_RC = 2%) combined with short-term immersion (approximately 7 days) enhances the residual shear strength of red clay, primarily through increases in cohesion and internal friction angle. However, higher FA concentrations (m_FA/m_RC ≥ 4%) and prolonged immersion (≥ 21 days) lead to a significant reduction in shear strength, with the most notable decrease observed at m_FA/m_RC = 6% and t = 21 days, where cohesion decreased by approximately 42%, and the friction angle decreased by approximately 36%. XRD results show that FA contamination does not significantly alter the mineral composition, but it leads to the dissolution and transformation of calcite and illite, thereby altering their relative abundances. SEM observations suggest that short-term FA contamination promotes particle aggregation and pore filling, resulting in a denser microstructure and increased shear strength. In contrast, prolonged contamination causes a reduction in particle size, increased porosity, and a more loosely packed structure, ultimately weakening the soil’s shear strength. The study proposes a dual adsorption mechanism: at low FA concentrations, FA enhances soil aggregation through the formation of “organic–inorganic bridges,” while at high concentrations, it induces interparticle repulsion through the development of a hydrated organic film, leading to structural loosening and strength degradation. These findings provide insights into the complex mechanisms by which FA contamination affects the mechanical properties of red clay, offering a theoretical foundation for geotechnical design and safety assessments in FA-affected regions.

Introduction

Driven by continued industrialization, soil contamination has become a global environmental challenge, with acid–base pollution particularly prevalent^1–3. Frequent accidental releases of acidic pollutants pose persistent risks to the integrity of engineering structures and to surrounding ecosystems. In 2017, in Jinghai District, Tianjin, China, the non-compliant discharge of acidic wastewater above regulatory limits triggered extensive ground subsidence and, in turn, severe contamination of adjacent groundwater. Such incidents show that acidic pollution undermines geotechnical stability and propagates via the hydrological cycle, underscoring the urgent need for systematic prevention and control³. Evidence further indicates that acidic inputs markedly depress the soil’s chemical buffering capacity, thereby reducing its ability to retard contaminant migration. Sustained acidification of the soil environment promotes softening of interparticle cementation and mineral dissolution, leading to chemical degradation of the long-term strength and stability of foundations and slopes^4–7. Soil shear strength is highly sensitive to hydrochemical conditions; the macroscopic mechanical response is tightly coupled to the co-evolution of microstructure and the ionic composition of pore water^8,9. Therefore, elucidating the mechanisms and time-dependent evolution of acid-induced degradation in soil engineering properties is critical for evidence-based risk assessment and mitigation in related geotechnical projects.

In recent years, extensive research has examined the engineering behavior of acid–alkali–contaminated soils. Findings show that acidic and alkaline species react with soil minerals, inducing mineral dissolution and leaching of cementing agents, thereby altering soil shear strength^8,10–14. Bakhshipour et al.¹¹, for example, reported that interactions between acid rain and residual soils markedly reduce strength and maximum dry density, increase the liquid limit and optimum water content, and are accompanied by a decline in zeta potential, reorganization of particle fabrics toward face–face flocculated alignment, and dissolution of Fe and Al cations with associated bond breakage; collectively, these microscale changes loosen the soil structure and degrade strength. Other studies indicate that in kaolinite-rich clays, variable edge charges are highly pH-sensitive; acidic/alkaline conditions modify edge charges and diffuse double-layer interactions, shifting interparticle associations from edge–face to face–face and thereby influencing the evolution of compaction characteristics, Atterberg limits, and strength parameters^15,16. Yan et al.¹⁷, integrating microstructural observations, showed that the pore architecture of loess evolves dynamically under acid attack, with the extent governed primarily by pH and driven by progressive dissolution of carbonate and silicate cements. Chen et al.¹⁸ reviewed three core themes concerning loess disintegration under aqueous environments: experimental methodologies, the evolution of disintegration indices, and coupled hydrochemical–microstructural responses. They identified systemic shortcomings—non-standardized testing, poor interstudy comparability, limited specimen representativeness—and a lack of cross-scale frameworks linking microscale mechanisms to macroscale manifestations. Shear tests by Hassanlourad et al.¹⁵ further indicate that strong acids markedly reduce the strength of sands and blended soils, whereas at moderate acidity (pH ≈ 3) some systems exhibit a strengthening–then–weakening inflection.

Shear strength of contaminated soils—a key mechanical index—has long been central to engineering geology research^8,19–21. Liu et al.²², for example, showed that in slip-zone soils, increasing SO₄²⁻ concentration lowers residual shear strength and the strength parameters cohesion (c) and friction angle (φ); the shear plane becomes smoother, evidencing electrolyte-enhanced cementation breakdown and dissolution-driven degradation. Chen et al.²³ examined the effects of ionic valence and concentration on soil microstructure and mechanics, finding that metal-ion contamination reduces porosity under load and that the compression index increases with ionic concentration. The shear strength, friction angle, and cohesion of metal-ion–contaminated clays all decrease, with larger declines at higher ionic concentrations. Ye et al.²⁴ investigated diesel-contaminated clays under freeze–thaw cycling and found monotonic decreases in shear strength and strength parameters (cohesion c and friction angle φ) with increasing diesel content and number of cycles. The tests indicate that clay residual strength is jointly governed by pore-fluid properties and mineral assemblage. Classic studies further show that, under identical stress conditions, replacing deionized pore water with the NaCl solution has little effect on the residual strength of kaolinite but markedly increases that of bentonite²⁵.

Despite these advances, several gaps remain: (i) most studies focus on peak strength or the strain-hardening stage, leaving the evolution and mechanisms of residual strength under coupled acid–base and time-dependent effects unresolved-particularly for cementation-rich, micro-aggregated red clays^3,26; (ii) the literature largely employs inorganic strong acids or simple organic acids (e.g., acetic acid) as surrogates, with little attention to the weak organic acid, such as fulvic acid (denoted by FA hereinafter)²⁷, and (iii) although long-term acidification and acid-rain scenarios have been considered, systematic studies are lacking on how varying organic-acid concentrations and immersion durations jointly drive the coordinated evolution of microstructure, surface chemistry, and mineralogy, and thereby control residual strength^17,28. In contrast, the time-dependent creep deformation of red clay under mechanical loading has been well characterized under triaxial stress conditions²⁹. Accordingly, clarifying how FA alters the residual shear strength of red clay will not only fill a critical knowledge gap but also inform prevention and remediation of soil contamination in practice. In this study, a typical red clay is examined through controlled laboratory tests to quantify the effects of fulvic-acid concentration and immersion time on residual shear strength. Microstructural and mineralogical changes are characterized by X-ray diffraction (XRD) and scanning electron microscopy (SEM). By integrating macro- and microscale observations, we elucidate the interaction mechanisms between fulvic-acid contamination and red-clay structure, providing a theoretical basis and methodological guidance for related engineering applications.

Materials and methods

Materials

The red clay sample used in this study was collected from a construction site at Guilin University in Yanshan District, Guilin, Guangxi, China. The sampling depth was between 2.0 and 2.5 m below the ground surface, and the red soil, reddish-brown in color, exhibited a hard to plastic consistency and a blocky structure. The collected samples were oven-dried at 105°C, then ground manually into powder for grain size distribution analysis, and the results are shown in Fig. 1, indicating that the red clay consists of 77.31% sand (> 0.075 mm), 20.41% silt (0.005–0.075 mm), and 2.28% clay (< 0.005 mm). Furthermore, the mineral composition of the red clay sample was determined by X-ray diffractometer (X’ Pert PRO Powder, PANalytical, Netherlands) and the result was presented in Table 1. As shown in Table 1, the main mineral constituents of the sample are quartz, calcite, plagioclase, kaolinite, illite, chlorite and goethite. Among these, quartz, plagioclase, and calcite are the most abundant, accounting for 30.51%, 23.57%, and 19.14% of the composition, respectively. In addition, according to the standard method for determination of inorganic elements in sediments by wavelength-dispersive X-ray fluorescence spectrometry³⁰, the chemical composition of the red clay samples was determined using an X-ray fluorescence (XRF) spectrometer (ZSX Primus II, Rigaku Corporation, Japan) and the results are shown in Table 2. It can be concluded that the primary chemical constituents of the sample are alumina (Al₂O₃), ferric oxide (Fe₂O₃), and silica (SiO₂), together constituting approximately 82.15% of the total composition, which is consistent with previous findings³¹. A loss-on-ignition (LOI) value of 6.38% may arise not only from the combustion of organic matter, but also from the dehydroxylation/dehydration of Fe oxides (e.g., goethite) and the loss of structural water from certain clay minerals. Meanwhile, other main physical properties of the red clay were also performed and are summarized in Table 3. It can be seen from Table 3 that the liquid limit and plastic limit of the red clay sample are 54.8% and 28.6%, respectively. The red clay sample exhibits distinct characteristics of high iron-aluminum content and a low silicon-aluminum ratio, with generally high Fe₂O₃ content. These features reflect the intense weathering and laterization processes that the red clay in this region has undergone under warm and humid climatic conditions, demonstrating significant typicality and representativeness. Therefore, the clay employed in this investigation was categorized as clay of high plasticity (CH) in accordance with the Standard for engineering classification of soil³². The particle size distribution results for the red clay (Fig. 1), however, show that the sand sized fraction (> 0.075 mm) accounts for 77.31% of the sample. This result is clearly inconsistent with the classification of the red clay as high-plasticity clay (CH) and its liquid limit of 54.8%. This discrepancy may stem from incomplete disaggregation of clay aggregates during sample preparation. Specifically, when the specimen was manually crushed using a rubber hammer, some aggregated fine particles may not have been fully broken down; consequently, these fine aggregates could have been retained and counted as larger particles during sieving. This would artificially increase the measured proportion of particles larger than 0.075 mm, leading to the particle-size distribution shown in Fig. 1.

Fig. 1.

Particle size distribution curve of the red clay.

Table 1.

Mineral composition of the red clay.

Sample	Relative content of the mineral constituents/%
	Quartz	Calcite	Plagioclase	Kaolinite	Illite	Chlorite	Goethite
Red clay	30.51	19.74	23.57	10.62	5.65	3.28	7.23

Table 2.

Chemical compositions of red clay and mass fraction of each component.

Sample	Chemical compositions and mass fraction of each component (%)
	Al2O3	Fe2O3	SiO2	K2O	MgO	CaO	Na2O	MnO	P2O5	TiO2	Loss on ignition
Red clay	26.72	22.86	32.57	2.04	0.76	3.59	0.68	0.65	0.24	2.51	6.38

Table 3.

Basic engineering properties of the red clay.

Engineering property indicators	Value	As per
Maximum dry density (ρd, max), g/cm3	1.54	The Standard for Geotechnical Testing Methods34
Optimum moisture content (wopt), %	22.6
Specific gravity (Gs)	2.71
Median particle size d50, mm	0.34
Liquid limit (wL), %	54.8
Plastic limit (wp), %	28.6
Plasticity index	26.2
Soil classification	Clay of high plasticity

Humic substances are amorphous, dark-brown organic mixtures formed through the microbial degradation and transformation of plant and animal residues in soil. Their primary components include humic acid, FA, and humin. This study employed a commercial FA standard, obtained from Shanghai DiBo Chemical Technology Co., Ltd., presented as a brown, amorphous dry powder. As a ubiquitous and vital component of humic substances, FA constitutes a core fraction of dissolved organic matter. It was selected as the reactive material in this study due to its molecular structure rich in active functional groups (e.g., carboxyl, phenolic hydroxyl, and carbonyl), which impart notable complexation capacity and redox activity³³. To ensure data accuracy and reproducibility, all FA reagents were stored in a desiccator at 4 °C, protected from light, prior to unsealing.

Sample preparation

To avoid potential surface reactions or chemical adsorption between FA molecule and the cutting ring, which may affect the experimental results of the red clay samples and compromise experimental accuracy, a custom cutting ring made of polyether ether ketone (PEEK) was employed. It should be emphasized that PEEK, a high-performance specialty engineering plastic, was selected for its exceptional mechanical strength, temperature resistance, and, most critically, its superior chemical corrosion resistance and anti-hydrolysis properties. These characteristics effectively prevent metal ion leaching or interfacial reactions, thereby eliminating potential interference with FA activity and the overall test system. The custom cutting ring was fabricated with a diameter of 61.8 mm and a height of 20 mm to ensure compliance with the dimensional specifications of a standard direct shear apparatus. It has been reported that the organic matter content in Guilin red clay generally ranges from 0% to 8%³¹. FA, a major constituent of soil organic matter, is an amorphous natural macromolecular organic compound characterized by aromatic ring structures. It is formed through the microbial-mediated decomposition, transformation, and polymerization of animal and plant residues³⁵, possesses a high ion exchange capacity and notable adsorption capability³⁶.

The collected red clay samples were dried in an oven at 105 °C, ground into powder, and then passed through a 2 mm sieve. Based on preliminary physical property tests (Table 3), the target dry density for all FA contaminated red clay specimens were set at 1.5 g/cm³, and the fulvic acid (FA) to red clay mass ratios (denoted by m_R=m_FA/m_RC hereinafter) was established at 0%, 2%, 4%, and 6%. According to these ratios, the required masses of FA and red clay powder were weighed separately. After being thoroughly mixed, the weighed red clay and FA powder mixture was statically compacted into a custom PEEK cutting ring (61.8 mm in diameter, 20 mm in height) for specimen preparation. Subsequently, the specimens from each m_FA/m_RC group were placed in butterfly saturators and subjected to vacuum saturation with distilled water (pH = 7). During saturation, the water level was maintained above the specimen surface, and an identical mass of distilled water was added to each specimen to ensure consistent conditions. Subsequently, the containers were sealed with plastic film to prevent contamination of the specimens by atmospheric carbon dioxide. Following this, the samples were immersed in acidic solutions containing different FA concentrations. Their macro- and micro-properties were tested at contamination durations of 0, 7, 14, and 21 days.

Drained repeated direct shear test

Accordance with the Chinese Standard for Geotechnical Testing Methods (GB/T 50123 − 2019)³⁴, the drained repeated direct shear test of the FA contaminated red clay specimens was performed using a strain-controlled four-unit direct shear apparatus (manufactured by Nanjing Soil Instrument Factory, China) as shown in Fig. 2, with the applied vertical pressures (σ_v) of 100, 200, 300, and 400 kPa, and the shear rate of 0.5 mm/min. The shear program was then initiated at a constant rate of 0.5 mm/min and continued until the maximum shear displacement was reached. Immediately afterward, the reverse reset procedure was activated, returning the lower shear box to its original position aligned with the upper box at the same rate (0.5 mm/min). A fixing pin was inserted to lock the assembly in place. Following a 2 min rest period, the same shearing process was repeated. This cycle was continued until the shear stress readings stabilized. Upon completion of the shearing process, the final vertical displacement was recorded, free water in the shear box was removed, and the displacement transducer, vertical load, pressure plate, and shear box components were dismantled in sequence. The failure characteristics of the specimen shear surface were then observed and described.

Fig. 2.

Schematic diagram of a strain-controlled four- unit direct shear apparatus.

XRD tests

To characterize the evolution of mineral composition in red clay after FA immersion, specimens from the shear surface were collected after the drained repeated direct shear tests and freeze-dried in liquid nitrogen. After complete drying, the samples were ground and passed through a 0.075 mm soil sieve. Subsequently, the sieved powder samples were compressed into uniform thin slices and mounted on the sample stage of an X-ray diffractometer (X’ Pert PRO Powder, PANalytical, Netherlands). The measurement was performed using a Cu target (Kα radiation) at 40 kV and 40 mA, with a scanning speed of 2°/min, and the diffraction patterns were record. Finally, the diffraction data were analyzed to obtain the sample’s crystal structure, unit cell parameters, and other structural characteristics.

SEM measurement

To characterize the particle morphology of red clay before and after FA contamination, scanning electron microscopy (SEM) was employed to observe both pristine red clay and samples subjected to immersion in solutions with varying FA concentrations and durations. To maintain structural integrity, the samples were snap-frozen in liquid nitrogen and lyophilized. The dried specimens were then affixed to aluminum stubs with conductive adhesive and sputter-coated with a carbon layer to mitigate electron charging. Microstructural examination was conducted using a field-emission scanning electron microscope (Gemini SEM300, UK) at an accelerating voltage of 5 kV, a working distance of 8 mm, and under high vacuum conditions. Analysis of the surface morphology provided further insight into the interaction mechanisms between FA and red clay particles.

Factors affecting residual shear strength

Effect of fulvic acid content on the residual shear strength

Drained repeated direct shear tests were conducted on red clay specimens under different vertical pressures. After the initial shear stage reached the peak shear stress, the lower shear box was slowly moved in reverse to realign with the upper box, followed by insertion of a fixing pin and a 2-minute rest period. Subsequently, a second shearing stage was performed, and this process was repeated until the shear stress stabilized. Based on the experimental data, the relationship between shear strength and vertical pressure for specimens with a FA content of 2% under 200 kPa vertical pressure is plotted in Fig. 3. The shear stress versus shear displacement curves were plotted for the specimens, from which the final stable values were taken as the residual strength. Subsequently, the relationship between shear strength and vertical pressure was graphically represented, enabling the determination of the cohesion and internal friction angle of the specimens.

Fig. 3.

Shear stress versus shear displacement curves of red clay (m_R=2%,σ_v = 200 kPa).

The variations in cohesion of red clay induced by different FA contents under various immersion durations are presented in Fig. 4.

Fig. 4.

Variations in cohesion of red clay induced by different FA contents.

As shown in Fig. 4, except for the specimens immersed for 21d, the cohesion of all other specimens initially increased and then decreased with rising FA content. Specifically, compared to the specimens with 0% FA content, the cohesion of samples immersed for 0, 7, and 14 days increased by 2.55%, 20.63%, and 8.49%, respectively, at a FA content of 2%. In contrast, for specimens immersed for 21 days, the cohesion decreased by 7.64%. When the FA content increased to 6%, the cohesion under all immersion durations showed a significant decline, with reductions of 1.82% (0 d), 3.22% (7 d), 23.17% (14 d), and 42.08% (21 d), respectively. In summary, when the fulvic acid content exceeds 2%, the cohesion of red clay exhibits a negative correlation with the content, indicating that higher content leads to lower cohesion.

The variations in internal friction angle of red clay induced by different FA contents under various immersion durations are illustrated in Fig. 5.

Fig. 5.

Variations in internal friction angle of red clay induced by different FA contents.

As shown in Fig. 5, for specimens with 0 days of immersion, the internal friction angle showed no significant changes across different FA contents. At a FA content of 2%, compared to the uncontaminated sample, the internal friction angle of specimens immersed for 7 and 14 days increased by 11.54% and 6.46%, respectively, while that of the specimen immersed for 21 days decreased by 9.45%. When the FA content increased from 2% to 6%, the internal friction angle exhibited a continuous declining trend. At the 6% content level, the internal friction angles of specimens immersed for 7, 14, and 21 days were reduced by 5.74%, 28.67%, and 35.79%, respectively, compared to the uncontaminated sample.

Effect of contamination duration on the residual shear strength

The variations in cohesion of red clay induced by different contamination durations under different FA contents are presented In Fig. 6.

Fig. 6.

Variations in cohesion of red clay induced by different contamination duration.

As shown in Fig. 6, for uncontaminated specimens (0% FA content), the cohesion remained stable at approximately 16 kPa across different immersion durations. At FA contents of 2% and 4%, the cohesion exhibited an initial increase followed by a decreasing trend with prolonged immersion time: after 7 days of immersion, the cohesion increased by 11.1% and 6.83% respectively compared to initial values; while after 21 days of contamination duration, it decreased by 13.8% and 22.3% respectively. It is noteworthy that specimens with 6% FA content showed a continuous decline in cohesion with extended immersion time, demonstrating a clear negative correlation.

The variations in internal friction angle of red clay induced by different contamination durations are illustrated in Fig. 7.

Fig. 7.

Variations in internal friction angle of red clay induced by different contamination duration.

As shown in Fig. 7, for the uncontaminated specimen (0% FA content), the internal friction angle remained stable at approximately 16.5° across all durations. In contrast, the contaminated specimens with FA contents of 2%, 4%, and 6% all exhibited an initial increase followed by a subsequent decrease in their internal friction angles: reaching peak values after 7 days of contamination with increases of 17.9%, 12.4%, and 0.31% respectively compared to initial values. Subsequently, when the contamination duration extended to 21 days, the internal friction angles decreased by 20.2%, 31.3%, and 33.1% respectively, demonstrating a negative correlation with contamination time.

Results and discussion

Mineralogical characteristics

To investigate the mechanism by which FA influences the shear strength of red clay, changes in mineral composition were analyzed. After the drained repeated direct shear tests, specimens with FA contents of 2%, 4%, and 6% and contamination durations of 0, 7, 14, and 21 days were subjected to XRD analysis (Fig. 8). All specimens exhibited comparable diffraction patterns, with no emergence or disappearance of major peaks, indicating no fundamental change in mineral type. Since chemical interactions may be more significant at higher FA contents, the specimens with 6% FA were further analyzed by Rietveld refinement to quantify the changes in the relative contents of mineral phases.

Fig. 8.

X-ray diffraction pattern at different contamination durations (a) m_R=2%; (b) m_R =4% and (c) m_R =6%.

As shown in Table 1, the red clay is mainly composed of quartz, calcite, plagioclase, kaolinite, illite, chlorite, and goethite. Table 4 summarizes the XRD results of red clay contaminated with FA at an m_R of 6% for different contamination durations (0, 7, 14, and 21 days). The results indicate that the relative contents of the mineral phases changed markedly with increasing contamination time. Compared with the uncontaminated specimen (0 days), the FA-contaminated specimens exhibited noticeable increases in the relative contents of quartz, chlorite, and kaolinite. After 21 days of contamination, the contents of quartz, chlorite, and kaolinite increased by 17.3%, 58.5%, and 54.6%, respectively, whereas the illite content decreased by 34.7%. Meanwhile, the calcite content showed a gradual decline with increasing contamination duration, with a total reduction of 46.4% after 21 days. The goethite content decreased initially and then increased; specifically, it dropped sharply by 65.1% after 14 days of contamination, followed by a marked increase with further extension of the contamination period. In addition, the plagioclase content showed an increasing trend at the early stage and then decreased as contamination time progressed.

Table 4.

Mineral composition of contaminated red clay at different contamination durations (m_R=6%).

t/d	Relative content of the mineral constituents/%
	Quartz	Calcite	Plagioclase	Kaolinite	Illite	Chlorite	Goethite
0	28.26	19.32	14.84	10.76	10.54	8.06	8.22
7	28.64	18.07	19.36	11.23	7.31	10.92	4.47
14	30.72	12.52	21.83	14.38	6.47	11.21	2.87
21	33.17	10.24	16.07	16.64	7.02	12.78	4.08

The results demonstrate that the variation in the shear strength parameters of soils is primarily governed by interparticle interactions. Among these, the contribution of cementing agents such as iron oxides, aluminum oxides, and calcium carbonate is particularly critical; alterations in the morphology of these cementitious materials directly weaken or enhance interparticle bonding, thereby determining the evolutionary trend of shear strength³⁷. The chemical composition analysis of the aforementioned specimens indicates that the primary chemical component of the red clay are alumina (Al₂O₃), ferric oxide (Fe₂O₃), and silica (SiO₂), together constituting approximately 82.15% of the total composition (Table 2), and the experimental results indicate that the notable enhancement in soil cohesion can be primarily attributed to the cementation provided by these oxides³⁸. Similar findings have shown that Al₂O₃ and Fe₂O₃ can effectively remediate contaminated soils and enhance their shear strength by forming cementitious bonds between particles³⁹. The mineral composition test results presented in Table 4 demonstrate that FA contamination induces significant alterations in the relative content of minerals within the red clay. Combined with the relationship curve between cohesion and contamination duration, it is evident that the cohesion of contaminated specimens undergoes corresponding changes, indicating a close correlation between cohesion variation and the formation and transformation of specific minerals in the specimens. As an organic acid, fulvic acid interacts with red clay particles and promotes the dissolution of kaolinite, releasing aluminum ions (Al³⁺). These ions subsequently undergo hydrolysis (Reaction 1), thereby increasing the content of soluble aluminum in the soil and compressing the thickness of the double diffusion layer on the surface of red clay particles, ultimately influencing the cohesion behavior of the soil^3,40,41. It should be emphasized that, red clay is a regionally specific soil formed through the laterization of parent rocks in carbonate-rich regions, where calcite typically serves as the key cementing agent between particles. These carbonate cements effectively enhance the connectivity between soil particles; however, in acidic environments, they undergo chemical reactions as shown in Eq. (2)~(6), thereby leading to the degradation of the cementation structure^3,28,42. Therefore, when red clay is immersed in a FA solution, extending the immersion period from 7 to 21 days leads to a progressive reduction in calcium carbonate content (Table 4). This reduction directly weakens interparticle cementation, thereby ultimately resulting in decreased soil shear strength. Additionally, the continuous dissolution of minerals in the soil under the action of FA leads to an increase in cation concentration in the solution of the sample. This compresses the thickness of the diffuse double layer on the surface of the soil particles, weakening the interparticle repulsive forces and thereby promoting their mutual approach and association, ultimately resulting in the formation of a flocculated structure³⁷. Additionally, as shown in Table 4, prolonged FA contamination significantly increases the contents of kaolinite and chlorite, while decreasing the content of illite. This pattern aligns with the well-established illite-to-kaolinite transformation mechanism in acidic environments³, as illustrated by the reaction in Eq. (7):

1

2

3

4

5

6

7

It should also be noted that FA constitutes a significant component of humic substances, characterized by a complex molecular structure abundant in diverse functional groups including carboxyl, phenolic hydroxyl, methoxyl, and ketonic carbonyl groups ³³. These functional groups not only facilitate the dissolution of specific minerals within red clay but also engage in ligand exchange with hydroxyl groups on the surfaces of iron and aluminum oxides (e.g., goethite) and their hydrous oxides, forming inner-sphere surface complexes. Alternatively, they can establish hydrogen bonds with oxygen atoms and hydroxyl groups on mineral surfaces, thereby influencing the physicochemical properties of the red clay.

Under low FA concentrations, FA molecules-functioning as negatively charged polyelectrolytes—specifically adsorb onto positively charged sites at clay particle edges (e.g., broken bonds in the aluminol octahedral sheets of kaolinite and illite) via abundant functional groups such as carboxyl and phenolic hydroxyl groups, thereby forming stable “organic–inorganic bridges.” These bridges not only strengthen interparticle connections but also neutralize positive edge charges, thereby suppressing the formation of unstable flocculated structures induced by edge–face attraction between heterogeneously charged particles⁴³. This promotes particle reorganization into denser face–face configurations, which collectively enhances the soil shear strength through increased cohesion and internal friction angle. However, when the FA content exceeds a critical threshold, excess FA molecules extend beyond edge-specific adsorption and form a thick, hydrated organic film over the entire clay particle surface via hydrophobic interactions and hydrogen bonding. This film introduces a dual weakening effect: on one hand, it increases the interparticle distance through steric hindrance and double-layer repulsion, effectively replacing short-range chemical bonding with long-range weak repulsion; on the other hand, its high hydration capacity promotes the adsorption of bound water, forming a lubricating film at particle interfaces that significantly reduces frictional resistance⁴⁴. Simultaneously, the persistent negative charges carried by FA further enhance the electronegativity of particle surfaces, intensifying interparticle electrostatic repulsion and promoting a transition toward a dispersed state. Ultimately, under high FA content, red clay shifts from a dense aggregated state to a dispersed suspended state, with interparticle interactions transitioning from attraction-dominated to repulsion-dominated, resulting in a marked reduction in shear strength. This turning point—from enhancement to attenuation—corresponds to the critical saturation point at which the adsorption behavior of FA shifts from monolayer chemical adsorption to multilayer physical adsorption on the particle surface.

To characterize the internal morphology of FA contaminated red clay, specimens with a FA content of 4% and immersion durations of 0, 7, 14, and 21 days were selected. After the drained repeated direct shear test, samples from the shear surface were freeze-dried in liquid nitrogen and subsequently examined using scanning electron microscopy (SEM). The results are shown in Fig. 9.

Fig. 9.

SEM images and corresponding binary images of FA contaminated red clay at different contamination durations (a) 0d, (b) 7d, (c) 14d and (d) 21d.

As shown in Fig. 9, the particles of uncontaminated red clay (0 d) are predominantly arranged in a face-to-face contact pattern, forming a relatively dense structure with predominantly elongated pores. When the contamination duration extends to 7 d, the pore morphology gradually shifts to being dominated by small pores, with some pores being filled or interconnected, resulting in reduced overall connectivity and tighter particle aggregation. As the contamination duration further increases to 14 d, the action of fulvic acid leads to a significant increase in surface roughness and the number of micropores on the red clay. The interparticle contact mode gradually transitions to being dominated by edge-to-face contacts, particle size tends to decrease, and both the number and size of pores increase, progressively forming a more connected macropore structure. When contamination continues to 21 d, edge-to-face contacts become further dominant, pore size continues to increase, and a more developed macropore system can be observed in the SEM images.

To quantitatively investigate the microstructural alterations in red clay, scanning electron microscopy (SEM) images were processed using the Particles (Pores) and Cracks Analysis System (PCAS) for binarization, and the corresponding pore structure parameters were subsequently extracted⁴⁵. Before SEM image identification, a threshold-segmentation method was applied to distinguish pores and soil particles based on grayscale values. An appropriate threshold was selected to convert the SEM images into binary images. During image processing, a relatively high initial threshold was first set, causing regions that originally belonged to soil particles to be misclassified as pores and displayed in white. The threshold was then gradually reduced until the fine soil particles within pores could be accurately identified. To reduce subjective bias, the threshold selection was repeated three times; when the mean difference among the three selected grayscale levels was no greater than 2 units, the average of the three values was adopted as the final threshold. If an SEM image contains too few pixels, the true geometric parameters of pores cannot be reliably represented. In such cases, a minimum pore-area threshold can be specified in the PCAS software to remove noise and small artifacts. The results are summarized in Table 5. Further analysis revealed that the pore structural characteristics of the specimens experienced significant changes with increasing contamination duration.

Table 5.

Pore measurements in FA-contaminated red clay from PCAS analysis.

Contamination duration/d	Fractal dimension	Porosity/%
0	1.2725	0.4611
7	1.3164	0.3538
14	1.2563	0.4995
21	1.2643	0.5537

As presented in Table 5, when the fulvic acid contamination duration extended from 0 day to 7 days, the porosity of the red clay decreased by 23.27%, indicating an improvement in soil densification, which subsequently resulted in an increase in the residual shear strength. The fractal dimension of the soil particles also increased by 3.31%. However, as the contamination duration was further prolonged to 14 and 21 days, the porosity increased by 41.18% and 56.50%, respectively, signifying significant development of the pore structure. This, in turn, led to a marked reduction in residual shear strength, as evidenced by decreases in both cohesion and the internal friction angle. It should be noted, however, that the PCAS-based quantitative analysis was conducted only for the 4% dosage, aiming to representatively reveal the age-dependent evolution of pore structure and aggregation characteristics of red clay under a typical fulvic acid dosage. For the other dosages (0%, 2%, and 6%), comparable PCAS quantitative statistics have not yet been performed; therefore, a systematic microstructural comparison across different dosages should be identified as a key direction for future work.

Integrated with the aforementioned SEM observations, the microstructure of the red clay exhibits a stage-wise evolution with prolonged contamination: the initially dense face-to-face contact structure with elongated pores transitions, after short-term contamination, into aggregated particles forming small, densely distributed pores. As corrosive effects intensify, particles undergo gradual refinement, edge-to-face contacts become dominant, the number of micropores increases, pore sizes expand and interconnect, ultimately resulting in a well-developed macropore system that leads to overall structural loosening and reduced stability of the soil mass. This study, however, has not yet conducted FTIR characterization on the surface of red clay contaminated with fulvic acid, and therefore no spectral data are available to directly support the “dual adsorption mechanism.” As a result, we are unable to confirm the binding characteristics of the functional groups of fulvic acid molecules with the edges or active sites of clay particles, which would provide direct evidence for the mechanism. Therefore, in future research, it will be essential to employ techniques such as FTIR, XPS, and TG-DSC to reveal the binding modes of fulvic acid functional groups with the edges or active sites of clay particles, further strengthening the evidence for the proposed mechanism. Furthermore, the “fulvic acid adsorption–microstructural evolution–strength variation” relationship identified in this study has a certain degree of general significance at the theoretical level. However, the magnitude of its effects may vary with factors such as mineralogical composition, the contents of Fe–Al oxides, and the initial pore structure of different red clays. Therefore, caution is required when extrapolating these conclusions to other types of red clay, and site-specific soil characteristics should be taken into account.

Conclusions

This study investigated the effects of FA contamination on the residual shear strength of Guilin red clay through drained repeated direct shear tests, X-ray diffraction (XRD), scanning electron microscopy (SEM), and particle and pore structure analysis system (PCAS). Based on the experimental results, the following main conclusions can be drawn:

1. The impact of FA contamination on the shear strength of red clay exhibits a biphasic trend: initial enhancement followed by subsequent reduction. Under conditions of low FA content (m_R=2%) and short-term immersion (approximately 7 days), both the cohesion and internal friction angle of the red clay increased, resulting in a shear strength improvement of approximately 10–20%. However, with higher FA content (m_R≥4%) or prolonged immersion (≥ 21 days), the cohesion and internal friction angle decreased significantly, leading to a marked decline in shear strength. Specifically, under conditions of m_R = 6% and t = 21 days, cohesion decreased by approximately 42%, and the friction angle decreased by approximately 36%.

2. The shear strength evolution of red clay is governed by the combined effects of FA content and contamination duration. Within the FA content range of 2–4%, both cohesion and the internal friction angle exhibit an initial increase followed by a decrease with prolonged immersion time, peaking at approximately 7 days before declining. In contrast, under high FA content (m_R= 6%), cohesion demonstrates a continuous decrease over time, while the internal friction angle shows a slight initial increase followed by a sharp reduction, reflecting a clear negative correlation with prolonged FA contamination.

3. The transformation of minerals induced by FA contamination has been identified as a primary mechanism governing the variation in cohesion. XRD analyses indicate that while FA contamination does not alter the types of minerals present in the red clay, it significantly modifies their relative abundances. Particularly at m_R= 6%, FA immersion led to the dissolution of calcite (approximately 46% reduction) and triggered notable changes in clay mineral composition, such as an increase in kaolinite and chlorite content accompanied by a decrease in illite. These observations suggest the occurrence of an illite-to-kaolinite transformation in the acidic environment. Such mineralogical alterations weaken interparticle cementation, consequently resulting in the degradation of cohesion.

4. The adsorption mechanism of FA reveals a stage-wise evolution in the structural behavior of red clay. Short-term FA contamination promotes particle aggregation and pore filling, leading to a denser structure and enhanced shear strength. However, with prolonged immersion, particle size decreases, edge-to-face contacts increase, and porosity rises, resulting in a more loosely packed soil fabric and a significant reduction in residual shear strength.

Author contributions

X.D . and P.Z spearheaded the study’s conception, designed the methodology and wrote the original draft. J.W and W.C performed extensive formal analysis. K.Y conducted the investigation, carried out data curation, and validated the findings. JF. W contributed to the investigation, developed key software, and assisted in writing the original draft. C.D developed supporting software, performed validation tasks, wrote the original draft and performed a critical review and editing of the manuscript. W.C secured the necessary funding for the project. L. L provided supervisory support and additional formal analysis. All authors approved the final version of the manuscript.

Funding

This work was financially supported by the 2025 Annual Henan Provincial Key Scientific and Technological Research Project of China (No. 252102321018).

Data availability

The data is included in the manuscript. For clarification or questions, the corresponding author may be contacted.

Declarations

Competing interests

The authors declare no competing interests.