Molecular Modeling of Arsenic Species Adsorption on Clay Minerals and in the Presence of Organic Matter

Abstract

Arsenic (As) contamination of soils is a critical environmental and geochemical concern, with its mobility and bioavailability largely controlled by molecular-scale interactions with soil minerals. This study investigates the adsorption behavior of arsenate [As(V)] and arsenious acid [As(III)] on major clay minerals to elucidate fundamental controls on As retention in soil and sediment systems. Molecular modeling approaches were employed to investigate these interactions. Density functional theory (DFT) calculations were performed on cluster models of illite, chlorite, montmorillonite, and kaolinite to evaluate adsorption configurations and binding energies of arsenate and arsenious acid. In addition, semiempirical (PM6) and classical force-field (UFF) methods were used to examine the influence of vermicompost-derived organic matter on arsenate-mineral interactions. Multiple adsorption configurations, including atop atom, bridge, three-fold filled, and three-fold hollow sites, were evaluated, and binding energies were calculated with correction for basis set superposition error. The results indicate that three-fold hollow sites are the most favorable, with As(V) binding energies of 60–65 kcal mol⁻¹ on illite, chlorite, and montmorillonite, reaching 75 kcal mol⁻¹ on kaolinite at a surface distance of 2.7 Å. In contrast, As(III) shows weaker and energetically flatter adsorption, with binding energies of 28–54 kcal mol⁻¹ and larger equilibrium distances of 3.2–4.0 Å. Modeling of vermicompost addition suggests a substantial reduction in arsenate binding on most clay minerals, except illite, indicating competitive or disruptive interactions at mineral surfaces. These findings provide quantitative, atomistic insight into mineral- and amendment-specific controls on As stabilization and mobility in soil and sediment systems.

1. Introduction

Arsenic (As) contamination of soils and sediments poses a major global environmental and public health challenge, particularly in alluvial and deltaic regions where natural geogenic processes intersect with intensive agricultural activity [1]. Once mobilized in the soil–water system, As can enter the food chain through plant uptake, threatening food safety and human health [2,3]. The environmental fate of As is largely governed by its interactions with soil minerals, organic matter, and pore-water chemistry, which together regulate adsorption, desorption, transport, and bioavailability. Understanding these processes at a fundamental level is therefore critical for predicting As mobility and for designing effective remediation and management strategies. In soils, As primarily occurs as arsenate [As(V)] under oxic conditions and arsenite [As(III)] under reducing environments, with the two species exhibiting markedly different reactivities and toxicities. Their retention in soils is controlled by physicochemical interactions with clay minerals, iron and aluminum oxides, and organic constituents such as humic substances. The dominant aqueous forms of As are further controlled by pH and redox potential, with arsenate species prevailing under oxidizing conditions and arsenite species under reducing environments, a distinction that strongly influences adsorption behavior and mobility in soils. Experimental studies over the past decade have demonstrated that adsorption strength varies strongly with mineralogy, surface coordination environment, hydration state, and pH [4,5]. However, many macroscopic adsorption models rely on empirical fitting of sorption isotherms and surface complexation frameworks, which, while useful, often obscure the molecular mechanisms governing As–mineral binding.

Recent advances in spectroscopic techniques—including synchrotron-based X-ray absorption spectroscopy, vibrational spectroscopies, and high-resolution microscopy—have significantly improved our ability to probe As speciation and coordination at mineral surfaces [5,6]. Yet, interpretation of such data increasingly requires complementary atomistic-scale theoretical models. Density Functional Theory (DFT) and related ab initio approaches have therefore emerged as powerful tools for elucidating adsorption geometries, binding energetics, and electronic structure changes associated with As-mineral interactions [7,8,9].

Despite these advances, several important gaps remain. First, most theoretical studies have focused on iron oxides such as goethite and ferrihydrite. In contrast, layered aluminosilicate clay minerals-illite, chlorite, montmorillonite, and kaolinite-have received comparatively less attention, despite their abundance and high surface reactivity in soils and sediments. Second, the relative binding preferences of As(V) and As(III) across different clay mineral surfaces remain debated. Some studies suggest stronger inner-sphere complexation of As(V), while others highlight the role of surface heterogeneity, hydration, and protonation states in stabilizing As(III) [10,11]. These diverging hypotheses underscore the need for systematic, mineral-specific investigations under controlled theoretical frameworks.

Another emerging area of uncertainty concerns the role of organic amendments and soil organic matter in modulating As adsorption. While field and laboratory studies have reported both enhancement and suppression of As retention following application of composts or organic residues [3,12], the underlying mechanisms remain controversial. Organic molecules may compete with As for mineral surface sites, alter surface charge, or induce structural rearrangements at the mineral-organic interface. However, direct molecular-level evidence for these processes is still limited. Against this backdrop, quantum chemical modeling offers a unique opportunity to bridge experimental observations and mechanistic understanding. By explicitly resolving adsorption sites, coordination geometries, hydrogen-bonding networks, and energetic preferences, DFT and ab initio calculations can reveal trends inaccessible to macroscopic models alone. Such insights are particularly valuable for rationalizing why certain clay minerals act as stronger As sinks than others and how organic matter modifies these interactions. In natural soil systems, As adsorption does not occur in isolation but in the presence of other oxyanions such as phosphate (PO₄³⁻), silicate (SiO₄⁴⁻), and sulfate (SO₄²⁻), which are abundant in soil solution and may compete for similar surface hydroxyl and oxygen coordination sites on clay minerals. Experimental studies have shown that phosphate in particular can significantly suppress arsenate adsorption due to structural similarity and competition for inner-sphere complexation sites. Consequently, competitive adsorption processes may reduce effective As retention compared to single-component systems. The present study, therefore, focuses on intrinsic As–mineral interactions under controlled, theoretical conditions, providing a mechanistic baseline that can support future multicomponent adsorption investigations.

The present study applies an integrated molecular modeling framework to investigate the adsorption of As(V) and As(III) on representative soil clay minerals-illite, chlorite, montmorillonite, and kaolinite. Density functional theory (DFT) calculations were employed to evaluate adsorption geometries, binding energies, and site preferences on mineral cluster models. In addition, semiempirical (PM6) and classical force-field (UFF) approaches were used to examine the influence of vermicompost-derived organic matter on arsenate-mineral interactions. The study aims to (i) compare the adsorption behavior of As(V) and As(III) across major clay minerals, (ii) identify energetically favorable adsorption sites and coordination environments, and (iii) assess how simplified organic matter coatings modify As-mineral binding at the molecular scale. The results demonstrate consistent mineralogical trends in binding strength, preferred adsorption sites, and molecular orientations, offering mechanistic explanations that align with—but also refine—existing experimental observations. These findings may contribute to a deeper geochemical understanding of As behavior in soils and sediments and provide a theoretical basis for developing mineral- and amendment-based remediation strategies.

2. Materials and Methods

2.1. Soil Mineral Selection and Arsenic Speciation

This study focuses on clay minerals that commonly dominate the reactive fraction of soils and sediments in As-affected regions, namely illite, chlorite, montmorillonite, and kaolinite. These minerals exhibit contrasting layer structures, surface hydroxyl densities, and cation substitution patterns, which collectively influence their geochemical reactivity and capacity to retain toxic trace elements. The structural differences among these clay minerals are expected to exert a strong control on As adsorption mechanisms. Illite and chlorite are non-expandable or weakly expandable 2:1 clay minerals in which interlayer potassium or brucite-like sheets limit access to internal surfaces, thereby restricting adsorption primarily to external basal and edge sites. In contrast, montmorillonite possesses an expandable 2:1 structure with significant isomorphic substitution, resulting in a permanent negative charge and enhanced surface reactivity that can promote electrostatic attraction and hydrogen-bond-assisted adsorption of As species. Kaolinite, a 1:1 non-expandable clay mineral, exhibits comparatively lower layer charge but a high density of structurally ordered surface hydroxyl groups, which can facilitate stable inner-sphere complex formation through ligand exchange reactions. These structural contrasts provide a useful framework for comparing adsorption energetics and coordination environments across mineral types and help rationalize mineral-specific differences in As retention observed in both experimental and theoretical studies. Their widespread occurrence and documented affinity for As species make them ideal model substrates for mechanistic adsorption studies [13].

Arsenic was considered in its two most environmentally relevant oxidation states: pentavalent As (As(V)), represented as the arsenate anion As(OH)₄⁻, and trivalent As (As(III)), modeled as arsenious acid As(OH)₃. These molecular forms are commonly reported in soil-water systems under oxidizing and reducing conditions, respectively, and are known to exhibit markedly different adsorption behaviors on mineral surfaces [14,15,16]. Arsenic speciation in natural soil–water systems is controlled primarily by pH and redox conditions, resulting in the occurrence of multiple protonation states of arsenate (e.g., H₂AsO₄⁻ and HAsO₄²⁻) and arsenite species. In oxidizing environments and at circumneutral to mildly alkaline pH, arsenate predominates and is commonly represented by the tetrahedral AsO₄ framework. In contrast, under reducing conditions, arsenite exists mainly as the neutral arsenious acid species (As(OH)₃). In the present study, As(OH)₄⁻ and As(OH)₃ were therefore selected as representative molecular models of As(V) and As(III), respectively, to enable systematic comparison of adsorption mechanisms associated with the two dominant oxidation states. This simplification is consistent with previous atomistic and density functional theory (DFT)–based adsorption studies that employ representative molecular forms of oxyanions to investigate adsorption mechanisms on mineral surfaces [8,17]. However, it does not explicitly account for the full range of aqueous protonation equilibria present in natural soil systems. Modeling both species enabled the assessment of speciation-dependent adsorption mechanisms at the mineral-water interface.

2.2. Development of Clay Mineral Cluster Models

To capture the essential physicochemical features of soil mineral surfaces, finite cluster models were constructed for illite, chlorite, montmorillonite, and kaolinite using crystallographic data derived from experimentally resolved X-ray diffraction (XRD) structures. Atomic coordinates and structural parameters for kaolinite and related 1:1 clay minerals were obtained from the crystallographic compilation of Jeans et al. [16]. Meanwhile, structural data for 2:1 clay minerals (illite and montmorillonite) were adopted from reported crystallographic refinements and theoretical surface studies [13]. Chlorite structural parameters were derived from standard crystallographic descriptions of clay minerals reported in the same compilations [16]. These experimentally determined datasets provide reliable layer stacking sequences, tetrahedral–octahedral coordination environments, and surface hydroxyl distributions required for realistic adsorption modeling.

Each mineral was represented by a four-layer (001) basal surface cluster constructed directly from the reported unit cell parameters. The final cluster sizes were as follows: illite (Si₁₆Al₈Mg₄O₅₆H₂₄; total 108 atoms), chlorite (Si₁₆Al₁₂Mg₈O₆₄H₃₂; total 132 atoms), montmorillonite (Si₁₆Al₆Mg₆O₅₆H₂₄; total 108 atoms), and kaolinite (Si₈Al₈O₂₄H₁₆; total 56 atoms). Peripheral dangling bonds were saturated with hydrogen atoms to maintain charge neutrality and preserve realistic surface hydroxyl configurations. These cluster sizes were selected to ensure representation of local coordination environments while maintaining computational tractability. Similar cluster dimensions have been widely adopted in quantum chemical investigations of clay–adsorbate interactions [13,18].

On each cluster surface, multiple adsorption environments were identified, including atop-atom sites, three-fold filled sites, three-fold hollow (vacant) sites, and bridge positions between adjacent oxygen atoms. These sites reflect the heterogeneity of real mineral surfaces and allow systematic evaluation of site-specific adsorption energetics. The constructed cluster models represent the basal (001) surfaces of illite, chlorite, montmorillonite, and kaolinite, which correspond to the most thermodynamically stable and commonly exposed crystallographic faces in layered aluminosilicate minerals. In natural soil systems, these basal surfaces constitute the dominant fraction of mineral surface area and therefore play a primary role in adsorption and surface complexation processes. Selection of the (001) surface also ensures preservation of realistic surface oxygen coordination and hydroxyl environments while minimizing artificial structural distortions that may arise from highly reactive edge terminations. This approach enables a systematic comparison of adsorption energetics across minerals under structurally consistent conditions and follows established practices in quantum-chemical studies of clay–adsorbate interactions.

2.3. Geometry Optimization of Arsenic Species

Prior to adsorption modeling, the molecular geometries of arsenate [As(OH)₄⁻] and arsenious acid [As(OH)₃] were optimized using Density Functional Theory (DFT) at the B3LYP level of theory. Geometry optimization refers to the iterative adjustment of atomic coordinates to obtain the minimum-energy configuration corresponding to a stationary point on the potential energy surface. The Stevens–Basch–Krauss–Jasien–Cundari (SBKJC) basis set, in combination with the LANL2DZ effective core potential (ECP), was employed to efficiently describe As-containing systems while maintaining computational feasibility. Effective core potentials replace chemically inactive inner-shell electrons with an equivalent potential, thereby reducing computational cost without compromising the accuracy of valence-electron interactions relevant to bonding and adsorption behavior. All geometry optimizations and adsorption calculations were performed using a consistent basis set framework across all As species and mineral cluster models to ensure methodological uniformity. The SBKJC effective core potential (ECP) basis set was employed for As atoms. Meanwhile, all other atoms (O, H, Si, Al, and Mg) were described using a double-zeta quality basis set of comparable flexibility.

Each mineral was represented by a finite four-layer cluster constructed from crystallographic structural data. In the present context, the term four-layer refers to a truncated fragment that preserves the essential structural stacking of the clay mineral, including surface oxygen atoms, the outer tetrahedral sheet, the underlying octahedral sheet, and the adjacent tetrahedral framework (or equivalent structural layers depending on mineral type). This configuration maintains the local coordination environment of adsorption sites while limiting the cluster size to a computationally tractable level. Because finite-cluster models inevitably introduce edge terminations, terminal oxygen atoms at the cluster boundaries were hydrogen-saturated to remove artificial dangling bonds and prevent unrealistic charge accumulation or exaggerated surface reactivity. Proton saturation is a standard procedure in cluster-based quantum-chemical simulations of mineral surfaces. It ensures that the electronic structure near the adsorption region remains representative of the bulk mineral lattice [13,18]. The resulting clusters, therefore, represent localized fragments of the basal (001) surfaces of clay minerals, preserving the geometrical arrangement and coordination of surface oxygen atoms that govern adsorption processes.

This uniform basis set scheme, 6-311G**, was consistently applied to isolated adsorbates, mineral clusters, and adsorbate-mineral complexes. Selected benchmark calculations with larger triple-ζ basis sets confirmed that structural parameters differed by less than 2%, validating the chosen level of theory for systematic adsorption energy comparisons [19,20].

2.4. Quantum Chemical Framework and Computational Details

All electronic structure calculations were carried out within the Kohn–Sham Density Functional Theory (DFT) framework using the dispersion-corrected B3LYP functional with Grimme’s D3 correction and the 6-311G** basis set (B3LYP-D3/6-311G**). This level of theory was applied consistently to isolated adsorbates, mineral cluster models, and adsorbate–mineral complexes to ensure methodological consistency in adsorption energy comparisons. The B3LYP hybrid exchange-correlation functional, which combines Becke’s three-parameter exchange functional with the Lee–Yang–Parr correlation functional, was employed due to its well-established performance in describing hydrogen bonding, electrostatic interactions, and surface complexation phenomena in mineral-adsorbate systems. To ensure consistent treatment of dispersion interactions, Grimme’s D3 empirical dispersion correction was applied to the B3LYP functional (B3LYP-D3), as implemented in GAMESS-US (General Atomic and Molecular Electronic Structure System United States version) and NWChem (Northwest Computational Chemistry Package). Inclusion of the D3 correction improves the description of long-range van der Waals interactions, which may contribute to stabilizing weakly bound adsorption configurations and mineral-organic interfacial systems [21].

Core electrons of As were treated using Effective Core Potentials (ECPs), while valence electrons were explicitly described using double-zeta quality basis sets developed for accurate surface and cluster calculations [22]. Although the modeled systems are of moderate size, the use of ECPs ensures computational efficiency across the large number of adsorption configurations evaluated in this study. Since adsorption interactions are primarily governed by valence-electron redistribution, the ECP framework retains the chemically relevant electronic structure while avoiding the computational overhead of a full all-electron treatment of inner-core orbitals. Validation against selected all-electron calculations confirmed negligible deviation in optimized geometries and adsorption trends. The same basis set combination was employed consistently for isolated As species, mineral clusters, and adsorption complexes, thereby eliminating methodological inconsistencies in energy comparisons. Binding energies were corrected for basis set superposition error (BSSE) using the counterpoise method. The use of two independent program packages enabled cross-verification of optimized geometries, total energies, and adsorption trends, ensuring robustness and reproducibility of the computed results. Molecular structures, adsorption geometries, and electronic features were visualized using GaussView software (Version 5.0). The computed adsorption energies are interpreted comparatively across minerals and adsorption sites rather than as absolute thermodynamic quantities.

For systems involving large organic-mineral composites, such as the vermicompost-mineral models considered in this study, full geometry optimization using ab initio DFT becomes computationally demanding due to the large number of atoms and degrees of freedom. Therefore, preliminary geometry optimization of these larger systems was performed using computationally efficient semiempirical and molecular mechanics methods. Specifically, the Parametric Method 6 (PM6) semiempirical approach and the Universal Force Field (UFF) were employed to obtain stable initial configurations of the vermicompost-mineral complexes. These methods are widely used for large organic and biomolecular systems because they significantly reduce computational cost while preserving reasonable structural accuracy. The optimized geometries obtained from these methods were subsequently used as the starting configurations for the adsorption energy calculations described in Section 2.7.

For clarity and consistency, all adsorption (interaction) energies reported in this study are calculated using the B3LYP hybrid functional. Geometry optimizations were initially validated using DFT approaches; however, only BSSE-corrected adsorption energies obtained at the B3LYP level are reported and discussed. These energies, therefore, represent interaction energies calculated within the B3LYP framework and provide a basis for comparing adsorption stability across minerals and adsorption sites.

2.5. Adsorption Modeling Procedure

Adsorption of arsenate and arsenious acid on mineral surfaces was simulated by positioning each adsorbate above predefined surface sites and systematically varying the vertical distance between the molecule and the mineral surface. For each adsorption site, the adsorbate was moved incrementally along the surface normal until a minimum-energy configuration was obtained. Lateral movement of the adsorbate was restricted to isolate the intrinsic energetics of individual adsorption sites.

In the case of arsenious acid, additional flexibility was introduced by allowing symmetric bending of the As-O-H bonds away from the surface once the optimal adsorption distance was identified. This approach accounted for the known tendency of As(III) species to adopt non-planar adsorption geometries and enabled identification of energetically favorable orientations [15]. During adsorption calculations, a constrained optimization strategy was adopted. The inner framework atoms of the mineral clusters (Si, Al, Mg, and subsurface oxygen atoms) were kept fixed to preserve the crystallographic integrity of the clay structures and maintain comparability among different mineral models. However, surface atoms directly involved in adsorption—including terminal oxygen atoms and hydroxyl groups located at or near the adsorption site—were allowed to relax together with the adsorbate during geometry optimization. This partial-relaxation approach enables realistic adjustments to local bonding environments and hydrogen-bonding interactions at the mineral-adsorbate interface while avoiding large structural distortions in the finite-cluster models. Such constrained optimization schemes are widely employed in cluster-based DFT studies of mineral surface adsorption to balance computational efficiency with physical realism [13,18]. This strategy permits local structural adjustment at the adsorption interface, including the reorientation of hydroxyl groups and modification of hydrogen-bonding geometry, while maintaining the overall structural stability of the cluster model. The approach follows established practices in quantum-chemical studies of mineral–adsorbate interactions and ensures physically meaningful adsorption geometries without introducing artifacts associated with full-cluster relaxation.

2.6. Binding Energy Calculations and Basis Set Superposition Error (BSSE) Correction

The strength of adsorption was quantified by calculating DFT interaction energies at the B3LYP-D3/6-311G** level for each As-mineral complex. Binding energies (BE) were calculated as the difference between the total energy of the optimized adsorption complex and the energies of the isolated fragments (adsorbate and mineral cluster). To minimize artificial stabilization arising from basis set superposition error (BSSE), all B3LYP binding energies were corrected using the counterpoise correction method of Boys and Bernardi. The magnitude of the BSSE correction was evaluated for each adsorption configuration and is reported explicitly in the corresponding tables. The corrected binding energy was obtained as follows:

Binding energy = total energy of the (adsorbate + substrate) − total energy of the adsorbate − total energy of the substrate − BSSE

Here, BSSE represents the counterpoise correction term. To avoid artificial overestimation of adsorption strength, the Basis Set Superposition Error (BSSE) was corrected using the counterpoise method, following standard quantum-chemical practice [18]. BSSE values were computed for representative adsorption configurations and averaged for each As-mineral system to maintain consistency across calculations. This correction is particularly important in cluster-based adsorption studies, where overlapping basis functions can significantly influence computed interaction energies [18,23].

For semiempirical PM6 calculations applied to the organic–mineral–arsenate systems, analogous BSSE estimates were evaluated to maintain consistency. The magnitude of BSSE corrections was generally small relative to the total interaction energies, indicating that the reported adsorption trends are not significantly affected by basis set artifacts.

2.7. Modeling the Influence of Vermicompost on Arsenic Adsorption

In the present study, vermicompost was represented using a simplified long-chain organic molecular model rather than an explicitly resolved chemical structure. This simplification was adopted because natural vermicompost and soil organic matter comprise highly heterogeneous macromolecular assemblies containing humic and fulvic substances, polysaccharides, proteins, and mineral-associated organic complexes that are computationally prohibitive to represent within ab initio DFT cluster calculations. The simplified model, therefore, aims to capture the dominant physicochemical effects relevant to As adsorption, including steric surface coverage, competition for mineral adsorption sites, and modification of local electronic environments, while maintaining computational tractability and consistency across mineral systems. Such reduced representations are commonly employed in quantum chemical studies to isolate first-order interaction mechanisms before incorporating higher levels of structural complexity. It is acknowledged that more advanced approaches, such as periodic DFT models, explicit solvation, or molecular dynamics simulations using chemically resolved humic structures, would better represent the dynamic and heterogeneous nature of organic matter in soils. However, the present approach provides a mechanistically interpretable framework for evaluating how organic coatings influence As-mineral binding at the molecular scale.

Following the computational strategy described in Section 2.4, the vermicompost model was initially optimized on mineral surfaces using the Universal Force Field (UFF) and the semiempirical Parametric Method 6 (PM6) [24,25].

In the present study, the analysis of the vermicompost–mineral interaction was restricted to arsenate [As(V)]. This choice was motivated by both environmental relevance and computational considerations. Under typical agricultural soil conditions where vermicompost is applied, As predominantly occurs in oxidized forms, and As(V) represents the major adsorbing species controlling As immobilization. Furthermore, due to its anionic nature, As(V) forms stronger inner-sphere complexes with mineral surfaces and is therefore more susceptible to competitive adsorption and surface masking by organic matter. In contrast, arsenious acid [As(OH)₃] is neutral at circumneutral pH. It exhibits weaker, orientation-dependent interactions dominated by hydrogen bonding, making systematic comparison within the simplified organic matter model less reliable. The focus on As(V) thus allows clearer mechanistic interpretation of organic amendment effects on mineral-bound As.

Following optimization of the mineral-vermicompost composite, As(V) was placed at its most favorable adsorption site on the underlying mineral surface, and binding energy calculations were carried out while keeping the composite structure fixed. For the organic-mineral-arsenate (O-M-As) systems, the binding energy of arsenate to the vermicompost–mineral composite was calculated according to the following expression:

BE = E(O-M-As) − [E(O-M) + E(As)] − BSSE

where E(O-M-As) is the total energy of the optimized organic-mineral-arsenate complex, E(O-M) represents the energy of the vermicompost-mineral composite in the absence of arsenate, and E(As) corresponds to the isolated arsenate species. BSSE corrections were applied using the counterpoise method to obtain reliable interaction energies. Importantly, the E(O-M) term was obtained after full relaxation of the organic molecule adsorbed on the mineral surface, ensuring that the interaction energy reflects adsorption on the stabilized organic–mineral composite rather than an unrelaxed configuration.

Prior to DFT optimization, initial geometries were pre-optimized using the PM6 semiempirical method. During this step, the heavy framework atoms of the clay mineral clusters (Si, Al, Mg, Fe, and structural O atoms) were kept fixed to preserve the crystallographic geometry of the finite cluster models and avoid unrealistic lattice distortions. Only the adsorbate molecules and surface hydroxyl groups were allowed to relax. The resulting structures were subsequently refined using DFT calculations. This constrained-cluster approach is commonly used in mineral surface modeling and ensures consistent comparisons of adsorption energies across different clay mineral systems.

Within this convention, negative binding energies represent stable adsorption, whereas positive values indicate destabilization of the adsorption complex or energetically unfavorable binding. This formulation was adopted to isolate the effect of organic coating on arsenate binding relative to the mineral–organic composite surface.

3. Results and Discussion

3.1. Validation of Cluster Models and Computational Framework

The reliability of adsorption energies and geometric descriptors obtained from ab initio calculations depends strongly on the realism of the mineral surface models used. In the present study, four-layer cluster models of illite, chlorite, montmorillonite, and kaolinite were constructed based on crystallographic data reported in standard clay mineralogy literature [16]. These models preserve the essential surface coordination environment, including surface oxygen atoms, hydroxyl groups, and octahedral and tetrahedral cationic sites that dominate adsorption processes in natural soils. The identification of distinct adsorption sites-atop atom, bridge, three-fold filled, and three-fold hollow sites-allows systematic evaluation of As binding preferences (Figure 1a–d). It should be noted that the structures illustrated represent finite cluster fragments of the basal surfaces rather than complete mineral layers. Consequently, only a portion of the ideal crystallographic unit cell is shown, which may visually resemble a partial layer (e.g., in the chlorite model). This truncation is necessary to maintain computational feasibility while preserving the local coordination geometry of adsorption sites. To avoid artificial reactivity associated with dangling bonds at the cluster edges, terminal oxygen atoms were saturated with hydrogen atoms during model construction. This treatment stabilizes the electronic structure of the cluster and ensures that adsorption interactions occur primarily at interior surface oxygen sites rather than at artificial edge defects. Similar proton-saturated cluster models have been widely used in quantum chemical studies of mineral-adsorbate systems.

The atoms shown in Figure 1a–d (O, Si, Al, and Mg) play distinct but complementary roles in controlling As adsorption on clay mineral surfaces. Surface oxygen atoms constitute the primary reactive sites for adsorption in the present cluster models. In natural clay minerals, additional hydroxyl-terminated surfaces (for example, the Al-OH octahedral surface of kaolinite) can also participate in adsorption processes; however, such hydroxyl-terminated basal surfaces were not explicitly represented in the simplified cluster models used in this study. The spatial distribution and coordination of surface oxygen atoms determine the availability of bridge, atop, and three-fold hollow adsorption sites identified in the present study. Silicon atoms form the tetrahedral framework of the clay structure and do not directly participate in adsorption; however, they influence adsorption indirectly by stabilizing the siloxane surface and governing the geometric arrangement and electronic environment of surface oxygen atoms. Aluminum and magnesium atoms, located mainly within octahedral sheets, contribute indirectly by modifying layer charge, local electrostatic potential, and surface acidity, which in turn affect the polarization and reactivity of adjacent oxygen atoms. Consequently, As adsorption in aluminosilicate clays is controlled primarily by oxygen-mediated interactions, while Si, Al, and Mg regulate adsorption strength through structural and electronic effects on the mineral surface.

Although the present investigation is computational in nature, the structural features represented in the cluster models are consistent with experimentally characterized clay mineral surfaces reported in the literature. X-ray diffraction (XRD) studies have established the layer structures, basal spacings, and tetrahedral–octahedral coordination environments of illite, chlorite, montmorillonite, and kaolinite, which form the structural basis of the present models. In addition, vibrational spectroscopic investigations, such as FTIR and ATR-FTIR, have demonstrated the presence and reactivity of surface hydroxyl groups and Si–O–Al linkages, which actively participate in adsorption and surface complex formation. Previous combined FTIR–theoretical studies on clay–adsorbate systems have shown good agreement between calculated adsorption geometries and experimentally observed vibrational shifts, supporting the reliability of cluster-based DFT approaches for describing adsorption mechanisms at mineral surfaces [13,26]. Thus, the computational framework adopted here remains consistent with experimentally observed mineral surface chemistry.

Geometry optimizations of isolated arsenate [As(OH)₄⁻] and arsenious acid [As(OH)₃] were performed at the B3LYP level using the SBKJC/LANL2DZ basis set. Structural validation against larger basis sets confirmed the reliability of the optimized geometries within acceptable error limits (<2%). This validation justifies fixing adsorbate geometries during surface scan, thereby reducing computational cost without compromising accuracy. Similar cluster-based approaches have been widely adopted for studying adsorption on aluminosilicate and iron oxide surfaces [8,18]. Furthermore, the use of finite-cluster models enables the explicit treatment of localized surface heterogeneities, edge effects, and coordination asymmetries that are often averaged out in periodic models. Such resolved representations are particularly important for clay minerals, where surface terminations and hydroxyl configurations exert strong control over contaminant binding. By systematically probing multiple adsorption sites on each mineral surface, the present approach provides a comprehensive and internally consistent framework for comparing As-mineral interactions across contrasting clay mineralogies. Collectively, this ensures that the computed adsorption energies and geometrical descriptors are both physically meaningful and environmentally relevant. While dispersion interactions may contribute to secondary stabilization, the dominant adsorption mechanism identified here involves short-range coordination and hydrogen bonding, suggesting that the qualitative adsorption trends remain robust despite the known limitations of dispersion in conventional hybrid functionals.

In addition to mineral composition, the microstructural characteristics of mineral surfaces play a decisive role in determining As adsorption strength. At the atomistic scale, adsorption is governed by the spatial distribution of surface oxygen atoms, hydroxyl group density, and the local coordination environment created by tetrahedral and octahedral sheets. Minerals possessing closely spaced surface oxygen atoms enable multidentate binding configurations, allowing arsenate species to form stable inner-sphere complexes through simultaneous interactions with multiple surface oxygens. This configuration lowers the total system energy and results in higher binding energies observed at three-fold hollow or filled sites. In contrast, surfaces with lower hydroxyl density or sterically constrained coordination environments restrict the formation of such multidentate interactions, leading to weaker adsorption as observed for arsenious acid. Furthermore, differences between 1:1 and 2:1 clay structures influence adsorption through variations in surface charge distribution, isomorphic substitution, and electrostatic heterogeneity, which collectively regulate hydrogen bonding strength and ligand exchange processes at the mineral–water interface [8,18]. The present results, therefore, demonstrate that adsorption strength is not solely a function of mineral chemistry but emerges from the microstructural arrangement of reactive surface sites that control accessibility, coordination geometry, and the electronic stabilization of adsorbed As species. Recalculation of representative adsorption configurations using the dispersion-corrected B3LYP-D3 functional resulted in moderate increases in absolute binding energies (typically 3%–8%). Meanwhile, the relative stability ordering among minerals and adsorption sites remained unchanged. This confirms that the mechanistic trends discussed herein are robust with respect to dispersion treatment.

3.2. Adsorption Site Preference and Binding Energies on Illite

The adsorption behavior of As species on illite reveals a clear dependence on surface coordination environment. As shown in

Table 1. Adsorption distances (R_┴) and BSSE-corrected binding energies calculated at the B3LYP-D3/6-311G** level for As(OH)₄⁻ and As(OH)₃ on a four-layer cluster model of illite.

Adsorption Site of As(OH)4−	R┴ (Å)	Binding Energy (kcal mol−1)		BSSE (kcal mol−1)
Atop atom	3.73	48.8		5.4
Bridge	3.24	36.1		4.1
Three-fold hollow	3.08	59.8		6.2
Three-fold filled	3.46	35.3		4.1
Adsorption Site of As(OH)3	R┴ (Å)	Binding Energy (kcal mol−1)	Angle of Tilt of As-O-H away from Surface (°)	BSSE (kcal mol−1)
Atop atom	4.01	30.5	20.7	3.0
Bridge	3.86	31.6	14.7	3.1
Three-fold hollow	3.53	33.6	19.6	3.4
Three-fold filled	3.92	27.8	12.1	2.9

, BSSE-corrected adsorption energies calculated at the B3LYP level indicate that As(V) exhibits its strongest binding at the three-fold hollow site, with a BSSE-corrected binding energy of 59.8 kcal mol⁻¹ and a vertical separation of approximately 3.08 Å from the surface. The atop atom site is the second most favorable, while the three-fold filled site shows the weakest interaction (Figure 2a,b). This hierarchy indicates that adsorption is dominated by multidentate interactions involving surface oxygen atoms rather than direct coordination to underlying cations.

Arsenious acid shows comparatively weaker binding on illite, with maximum binding energies limited to 33.6 kcal mol⁻¹ at the three-fold hollow site. Additionally, As(OH)₃ prefers a tilted adsorption geometry, with As-O-H bonds oriented away from the surface (Table 1 and Figure 2c). This geometry minimizes steric repulsion and indicates weaker hydrogen bonding than in arsenate. The flatter energy landscape for As(OH)₃ across different sites suggests higher surface mobility, which has important implications for As bioavailability under reducing soil conditions [15,25].

These results underline the critical role of surface oxygen topology in governing As retention on illite, where hollow sites provide optimal coordination geometry for inner-sphere complexation. The pronounced difference in binding strength between arsenate and arsenious acid reflects speciation-dependent electrostatic and hydrogen-bonding interactions at the mineral-water interface. The weaker and more flexible adsorption of As(OH)₃ implies a greater susceptibility to desorption and transport, particularly under anoxic conditions where As(III) dominates. Consequently, illite-rich soils may act as less effective sinks for As in reduced environments, enhancing the risk of As migration into groundwater and the food chain.

3.3. Comparative Adsorption Behavior on Chlorite

Results for chlorite closely mirror those observed for illite, reinforcing the robustness of the observed adsorption trends. As summarized in

Table 2. Adsorption distances (R_┴) and BSSE-corrected binding energies calculated at the B3LYP-D3/6-311G** level for As(OH)₄⁻ and As(OH)₃ on a four-layer cluster model of chlorite.

Adsorption Site of As(OH)4−	R┴ (Å)	Binding Energy (kcal mol−1)		BSSE (kcal mol−1)
Atop atom	3.65	51.8		6.4
Bridge	3.42	36.9		4.3
Three-fold hollow	3.04	62.1		6.8
Three-fold filled	3.78	40.6		4.7
Adsorption Site of As(OH)3	R┴ (Å)	Binding Energy (kcal mol−1)	Angle of Tilt of As-O-H away from Surface (°)	BSSE (kcal mol−1)
Atop atom	3.61	33.5	15.6	2.9
Bridge	3.54	34.7	14.5	3.1
Three-fold hollow	3.37	37.2	17.2	3.3
Three-fold filled	3.96	29.9	11.3	2.8

, As(V) again binds most strongly at the three-fold hollow site (62.1 kcal mol⁻¹), followed by the atop atom site, as also evident from Figure 3a,b. The adsorption distances remain within 3.04–3.78 Å, consistent with inner-sphere surface complexation.

For arsenious acid, binding energies range from 29.9 to 37.2 kcal mol⁻¹, with the three-fold hollow site being the most stable (Figure 3c). The tendency of As(OH)₃ to adopt tilted configurations persists, with tilt angles between 11° and 17°. These findings suggest that chlorite surfaces, despite their mixed octahedral composition, exhibit adsorption characteristics similar to those of illite, primarily governed by surface oxygen coordination rather than by specific cation identity.

The close similarity between chlorite and illite adsorption behavior indicates that the fundamental mechanism of As retention is largely independent of variations in octahedral cation composition. Instead, the availability and spatial arrangement of surface oxygen and hydroxyl groups are decisive in stabilizing As species at the mineral-water interface [27]. The consistent preference for three-fold hollow sites reflects enhanced multidentate interactions and hydrogen bonding possibilities. Consequently, chlorite-rich soils are expected to exhibit As retention capacities comparable to illitic systems under similar geochemical conditions.

3.4. Enhanced Arsenic Binding on Montmorillonite

Montmorillonite exhibits stronger interactions with both species of As relative to illite and chlorite, highlighting the role of expandable 2:1 clay structures in contaminant retention. As shown in

Table 3. Adsorption distances (R_┴) and BSSE-corrected binding energies calculated at the B3LYP-D3/6-311G** level for As(OH)₄⁻ and As(OH)₃ on a four-layer cluster model of montmorillonite.

Adsorption Site ofAs(OH)4−	R┴ (Å)	Binding Energy (kcal mol−1)		BSSE (kcal mol−1)
Atop atom	3.42	49.6		5.4
Bridge	3.18	57.3		5.9
Three-fold hollow	3.37	60.7		6.2
Three-fold filled	2.90	65.4		6.3
Adsorption Site of As(OH)3	R┴   (Å)	Binding Energy (kcal mol−1)	Angle of Tilt of As-O-H away from Surface (°)	BSSE (kcal mol−1)
Atop atom	3.67	43.7	13.9	3.9
Bridge	3.52	46.2	12.2	4.1
Three-fold hollow	3.37	60.7	12.4	6.3
Three-fold filled	3.21	49.5	11.9	4.8

, arsenate binding energies reach up to 65.4 kcal mol⁻¹ at the three-fold filled site, with adsorption distances as low as 2.90 Å. The enhanced binding can be attributed to the high surface reactivity of montmorillonite, greater density of hydroxylated oxygen atoms, and the presence of isomorphic substitutions that generate localized negative charges. These features collectively promote stronger electrostatic and hydrogen-bonding interactions with arsenate species (Figure 4a–d).

Arsenious acid also shows enhanced binding on montmorillonite compared to other 2:1 clays, with binding energies exceeding 49 kcal mol⁻¹ at favorable sites (Figure 4e). However, the relatively small energy differences among adsorption sites suggest a flatter potential energy surface for As(OH)₃, implying higher surface mobility and sensitivity to environmental conditions. This behavior reflects the weaker electrostatic interactions of neutral arsenious acid compared to those of anionic As(V). Overall, the results are consistent with experimental reports of enhanced As retention in smectitic soils, particularly under neutral to slightly alkaline pH conditions [6,28], reinforcing the geochemical significance of montmorillonite-rich soils in As sequestration.

3.5. Strongest Arsenic Retention on Kaolinite Surfaces

Among all clay minerals investigated, kaolinite exhibits the strongest affinity for As(V). As shown in

Table 4. Adsorption distances (R_┴) and BSSE-corrected binding energies calculated at the B3LYP-D3/6-311G** level for As(OH)₄⁻ and As(OH)₃ on a four-layer cluster model of kaolinite.

Adsorption Site ofAs(OH)4−	R┴   (Å)	Binding Energy (kcal mol−1)		BSSE (kcal mol−1)
Atop atom	2.95	51.6		5.1
Bridge	2.81	68.8		6.9
Three-fold hollow	2.84	59.2		6.1
Three-fold filled	2.70	75.4		7.3
Adsorption Site of As(OH)3	R┴ (Å)	Binding Energy (kcal mol−1)	Angle of Tilt of As-O-H away from Surface (°)	BSSE (kcal mol−1)
Atop atom	3.58	49.7	16.6	4.9
Bridge	3.41	50.6	17.0	5.1
Three-fold hollow	3.39	52.1	18.4	5.3
Three-fold filled	3.26	53.5	18.0	5.5

, the three-fold filled site yields a binding energy as high as 75.4 kcal mol⁻¹, accompanied by short adsorption distances (~2.70 Å) which are characteristic of strong inner-sphere surface complexes. This enhanced binding can be attributed to the high density of surface hydroxyl groups and the structurally ordered, non-expandable 1:1 layer configuration of kaolinite, providing well-defined and energetically favorable adsorption sites (Figure 5a,b). Such surface characteristics promote stable coordination and limit As desorption under varying environmental conditions.

Arsenious acid, although more weakly bound than arsenate, still shows appreciable binding (Figure 5c) on kaolinite, with energies ranging from 49.7 to 53.5 kcal mol⁻¹. The pronounced tilt of the As-O-H bonds away from the mineral surface reflects steric accommodation and the reorganization of hydrogen-bonding networks at the interface. Despite this reduced binding strength, the overall retention of As(III) on kaolinite remains significant relative to other clay minerals. The superior As sequestration capacity of kaolinite observed in this study lends strong theoretical support to its application in As filtration and remediation systems, particularly in As-affected regions [29,30].

3.6. General Trends in Arsenic Speciation and Mineralogical Control

Although four adsorption environments (atop atom, bridge, three-fold filled, and three-fold hollow sites) were identified and evaluated for all mineral cluster models, only selected representative adsorption geometries are illustrated for illite, chlorite, and kaolinite in Figure 2, Figure 3 and Figure 5 to maintain clarity and avoid redundancy. All adsorption energies discussed in this section correspond to BSSE-corrected interaction energies obtained using the B3LYP-D3/6-311G** level, indicating that As(V) exhibits its strongest binding functional, ensuring consistent comparison across mineral surfaces, as reported in Table 1, Table 2, Table 3 and Table 4. In contrast, montmorillonite exhibits a comparatively heterogeneous surface coordination environment and smaller energetic differences among adsorption sites; therefore, all four adsorption configurations are explicitly shown in Figure 4 to illustrate the diversity of stable adsorption geometries. The difference in graphical representation thus reflects a choice in visualization rather than differences in adsorption site availability or computational treatment across minerals.

Across all mineral surfaces, several consistent trends emerge that clarify the mechanistic controls on As retention in soils. Arsenate [As(V)] exhibits systematically stronger adsorption than arsenious acid [As(III)], primarily due to its higher negative charge and enhanced ability to form multidentate hydrogen-bonded surface complexes. These trends further emphasize that mineral microstructure—particularly surface oxygen topology and hydroxyl arrangement—controls the energetic stabilization of As species by governing the availability of multidentate adsorption geometries across different clay minerals. The universal preference for three-fold hollow sites highlights the dominant role of coordinated surface oxygen atoms in stabilizing As species through inner-sphere complexation. Furthermore, the stronger retention observed on kaolinite and montmorillonite compared to illite and chlorite reflects the influence of mineral structure, surface hydroxyl density, and layer reactivity in regulating As mobility and persistence in soil environments.

These atomistic trends are in strong agreement with spectroscopic and macroscopic adsorption studies reported for iron oxides and clay minerals, which identify inner-sphere complex formation as the primary mechanism of As immobilization [26,31]. The close correspondence between computational predictions and experimental observations confirms that the DFT-based cluster models employed here successfully capture the essential physicochemical processes governing As-soil interactions. It should be noted, however, that the adsorption energies reported here represent intrinsic binding strengths derived from single-adsorbate systems. In natural soils, the presence of competing oxyanions, such as phosphate, silicate, and sulfate, may reduce the effective adsorption of As by occupying reactive surface sites or altering local electrostatic environments. Therefore, while the computed trends reliably describe relative mineral affinities and adsorption mechanisms, absolute adsorption stability in field conditions may be lower due to competitive sorption effects.

3.7. Effect of Vermicompost on Arsenic Adsorption

Incorporation of a simplified vermicompost model reveals a pronounced modification of arsenate-mineral interactions, with a clear tendency toward destabilization of As(V) binding on most clay surfaces [28]. The simplified representation of vermicompost used here captures dominant competitive and electronic effects but does not account for the full chemical heterogeneity and dynamic behavior of natural soil organic matter. As shown in

Table 5. Binding energy of arsenate interacting with vermicompost–mineral composite systems (O-M-As). For these large organic-mineral-arsenate cluster models, binding energies were calculated using the semiempirical PM6 method.

Soil Crystallite	Binding Energy of Arsenate on Vermicompost-Soil Composite (kcal mol−1)	BSSE (kcal mol−1)
Chlorite	+1036.67	0.28
Illite	−8.44	0.35
Kaolinite	+1063.40	0.33
Montmorillonite	+822.40	0.41

and Figure 6, the organic model molecule effectively disrupts arsenate adsorption on chlorite, kaolinite, and montmorillonite, resulting in positive binding energies that signify desorption and enhanced mobility. The persistence of weak As(V) binding only on illite (−8.44 kcal mol⁻¹) indicates that the response to organic amendments is strongly mineral-specific and governed by surface structural characteristics.

The calculated energies correspond to the interaction of arsenate with an already formed organic-mineral composite system (O-M). Binding energies were therefore computed relative to the organic-coated mineral surface using the expression described in Section 2.6. Consequently, positive binding energies represent energetically unfavorable adsorption of arsenate on the organic-coated surface, indicating that the organic layer disrupts or weakens the intrinsic mineral-arsenate interaction. The magnitude of the positive interaction energies partly reflects steric constraints imposed by the simplified organic model and the fixed mineral framework used during initial PM6 pre-optimization, which can amplify destabilization when arsenate is forced into sterically crowded adsorption environments. Therefore, these values should be interpreted primarily in a comparative and qualitative sense, as indicators of destabilized adsorption rather than absolute thermodynamic adsorption energies, reflecting relative disruption of mineral–arsenate interactions by organic coatings.

Binding energies were calculated as BE = E(O-M-As) − [E(O-M) + E(As)] with BSSE correction. For the large organic-mineral-arsenate composite systems, energies were evaluated using the semiempirical PM6 method, as the system size rendered full B3LYP-D3 calculations computationally impractical. Positive values indicate destabilization of arsenate adsorption on the organic-coated mineral surface.

The vermicompost influence was evaluated only for As(V), as this species dominates under oxidizing soil environments typically associated with organic amendment application. Because arsenate exhibits stronger electrostatic and multidentate interactions with mineral surfaces, any disruption caused by organic coatings becomes more clearly quantifiable. The weaker, largely hydrogen-bond-driven adsorption of As(III), as demonstrated in earlier sections, would be less sensitive to competitive site occupation in the simplified organic model used here. These observations suggest that organic matter can competitively inhibit As adsorption by masking reactive surface sites, modifying hydrogen-bonding networks, or altering local electrostatic environments at the mineral interface [32]. Such behavior is in agreement with experimental reports documenting increased As release and transport following organic matter enrichment in soils and sediments. Consequently, the results emphasize the need for mineral-organic interactions when designing and implementing remediation strategies based on vermicompost amendments, particularly in As-contaminated environments.

The atomistic insights generated in this study clarify the fundamental mechanisms controlling As retention and transport at soil mineral interfaces. A notable outcome of the simulations is the distinctly different response of illite compared to chlorite, montmorillonite, and kaolinite following vermicompost addition. While organic coating substantially destabilized arsenate adsorption on most minerals, illite retained weakly negative binding energy, indicating partial preservation of adsorption stability. This difference can be attributed to mineral structural characteristics and surface charge distribution. Illite, a non-expandable 2:1 clay with fixed interlayer potassium and localized negative charge from tetrahedral substitution, restricts penetration and rearrangement of organic molecules at reactive surface sites, thereby limiting surface masking. In contrast, expandable clays such as montmorillonite allow greater organic accessibility to adsorption sites, leading to stronger disruption of multidentate arsenate coordination. Kaolinite, despite its strong intrinsic affinity for arsenate, contains abundant surface hydroxyl groups that readily interact with organic functional groups, resulting in competitive adsorption and weakened arsenate binding. Chlorite exhibits intermediate behavior but remains susceptible to organic-induced electrostatic screening.

Mechanistically, vermicompost-derived organic matter likely modifies As adsorption through the following: (i) competitive occupation of surface oxygen sites by organic functional groups, (ii) alteration of local electrostatic conditions at the mineral interface, and (iii) disruption of hydrogen-bonding networks stabilizing inner-sphere arsenate complexes. The persistence of limited arsenate binding on illite, therefore, suggests that structural rigidity and localized charge stabilization can partially counteract organic competition.

The markedly stronger adsorption of As(V) on kaolinite and montmorillonite indicates their potential to act as long-term geochemical sinks for As under oxidizing conditions. In contrast, the relatively weaker and flatter binding-energy landscape of arsenious acid suggests enhanced mobility and bioavailability, particularly in reduced soil and sediment environments. Furthermore, the disruption of As adsorption in the presence of organic amendments highlights the complex and sometimes counterproductive role of organic matter in remediation strategies. Collectively, these findings underscore that the fate of As in soils is governed by the interplay of mineralogical composition, chemical speciation, and organic interactions, demonstrating the strength of DFT-based methods in resolving environmentally relevant geochemical processes at the molecular scale.

4. Conclusions

This study provides a detailed atomistic understanding of As speciation-dependent interactions with major soil clay minerals via ab initio quantum-chemical calculations. Density functional theory-based adsorption analyses reveal that both As(V) and As(III) preferentially bind to three-fold hollow surface sites across illite, chlorite, montmorillonite, and kaolinite, with As(V) consistently exhibiting stronger binding energies than As(III). Among the minerals studied, kaolinite and montmorillonite demonstrate the highest affinity for As species, highlighting the critical role of mineralogical composition in governing As immobilization in soils and sediments. The results further indicate that adsorption is dominated by short-range chemisorptive interactions, hydrogen bonding, and surface coordination rather than deep lattice penetration, offering mechanistic insight consistent with experimental spectroscopic observations reported in the literature. However, the present work is constrained by the use of finite cluster models, static adsorption configurations, and simplified representations of soil organic matter and vermicompost, which do not fully capture long-range electrostatics, surface heterogeneity, competitive ion effects, or dynamic soil-water interactions. Furthermore, As speciation was represented using simplified molecular forms of As(V) and As(III), and future studies incorporating explicit pH-dependent protonation states and aqueous speciation equilibria would further improve environmental realism. In the computational framework, dispersion interactions were explicitly included by applying the Grimme D3 correction to the B3LYP functional (B3LYP-D3). This approach accounts for long-range van der Waals interactions, thereby enhancing the reliability of weak adsorption configurations and mineral–organic interface modeling. Meanwhile, it also preserves the previously identified mineralogical and speciation-dependent adsorption trends. Furthermore, competitive adsorption effects involving coexisting oxyanions, such as phosphate, silicate, and sulfate, were not explicitly modeled and may reduce As binding strength under natural soil conditions. Future studies should therefore integrate periodic DFT calculations, explicit solvation, molecular dynamics simulations, and more chemically realistic organic matter models to better represent natural soil systems. Coupling such theoretical approaches with advanced spectroscopic validation and multicomponent adsorption scenarios will be essential for translating atomistic insights into predictive frameworks for As mobility, risk assessment, and the design of mineral-based remediation strategies in contaminated soil and sediment environments.