Introduction

Arsenic, a naturally occurring element, is ubiquitously distributed in the environment, including the atmosphere, hydrosphere, pedosphere, and lithosphere (Matschullat, 2000). Soil, in particular, serves as a prominent reservoir for arsenic, with concentration ranging from 1 to 40 mg/kg and averaging around 5 mg/kg (Sarkar & Paul, 2016). The United States Environmental Protection Agency (USEPA) has established a maximum limit of 24 mg/kg for arsenic in soils (Singh et al., 2015). Nevertheless, anthropogenic activities, e.g., mining, chemical manufacturing, and the discharge of metallurgical effluents, contribute to releasing arsenic into the environment (Raju, 2022), resulting in global soil arsenic contamination and posing health risks to ~ 200 million individuals (Wan et al., 2020). Arsenic is recognized as one of the most hazardous elements due to its persistence, refractory degradation, and bioaccumulation (Smedley & Kinniburgh, 2002). The USEPA classifies inorganic arsenic as a “human carcinogen” due to its damage to digestive, endocrine, renal, and reproductive systems upon prolonged exposure.

Understanding the adsorption behavior of arsenic in soils is of significance due to its universality and high toxicity. Arsenic adsorption mechanisms in soils can be broadly classified as nonspecific adsorption (outer-sphere surface complexation) and specific adsorption (inner-sphere surface complexation) (Aredes et al., 2013). Compared to outer-sphere complexes, inner-sphere complexes are more stable based on the integration by ligand covalent bonds and electrostatic interactions (Gorny et al., 2015). Extensive research has demonstrated that arsenic is readily adsorbed onto various soil colloids, such as iron oxides, aluminum oxides, manganese oxides, carbonates, and clays, forming stable surface complexes (Dousova et al., 2012). The pH conditions, which govern the ionization degree and chemical form of arsenic (Tahervand & Jalali, 2017), exert a significant influence on its adsorption. Organic carbon, a pivotal constituent of soil, competes with arsenic for adsorption sites. However, it also plays a crucial role in the redox transformation and migration of arsenic within the soil (Bauer & Blodau, 2006; Dobran & Zagury, 2006).

Surfactants, possessing both hydrophobic and hydrophilic properties, play pivotal roles in various applications. Figure 1 summarizes surfactants’ classification, properties, and applications (Bolan et al., 2022; Li et al., 2016; Ying, 2006). The global demand for surfactants has surged in recent years, reaching a consumption of 15 million tons in 2014, with an annual increase of 4.3% (Al Ghatta et al. 2022). Consequently, substantial amounts of surfactant residues have been discharged into the environment, ultimately distributed in water (Arora et al., 2023), sediment (Liang & Peng, 2017), and soil (Xiang et al., 2016). Moreover, surfactants in the aqueous environment readily bind to soil components (Ishiguro & Koopal, 2016). The fate of surfactants in soils has garnered increasing attention due to their widespread presence. The partitioning and adsorption of surfactants in surfactant-water-solid systems are influenced by various factors, including surfactant characteristics, pH, temperature, background ions, salinity, organic carbon, and minerals (Belhaj et al., 2021b; Liu et al., 2019, 2021; Wang & Keller, 2008). Potential mechanisms affecting surfactant adsorption behavior in soils encompass electrostatic interactions, ion association, hydrophobic bonds, π-electron polarization, dispersion force adsorption, etc. (Amirianshoja et al., 2013; Belhaj et al., 2020b, 2022; Groenendijk & van Wunnik, 2021; Liu et al., 2021). Electrostatic attraction between the charged head group and the solid surface is identified as the primary mechanism driving surfactant adsorption, facilitating the transfer of surfactant molecules from the solution phase to the surface interface (Belhaj et al., 2020c). However, the dehydration process of ionic head groups of surfactants may hinder adsorption (Zhang & Somasundaran, 2006). Additionally, the equilibrium between hydrophobic and hydrophilic forces of the surfactant tail and head groups predominantly determines the intensity of partitioning (Belhaj et al., 2020a). Monolayer and multilayer adsorption may simultaneously occur in various media systems (mineral, sediment, rock, etc.) for surfactants (Belhaj et al., 2021a; Saxena et al., 2019). Upon adsorption onto soil particles, surfactants instigate surface reactions altering colloid stability and various physicochemical properties of soils, including surface potential, permeability, pore structure, cation exchange capacity (CEC), and ionic strength (Ishiguro & Koopal, 2016; Kaya & Yukselen, 2005; Mejia-Avendaño et al., 2020; Mulligan et al., 2001). Given the prevalence of surfactants in soil environments, their potential to affect the fate of arsenic in soils and posing challenges for accurately assessing environmental risks and managing soil arsenic pollution. However, existing research on the influence of surfactants on arsenic adsorption behavior in soils remains incomplete.

Fig. 1
figure 1

The classification, properties, and applications of surfactants

Full size image

This study aims to investigate the adsorption of arsenic in soils in the presence of three surfactants, including sodium dodecyl sulfate (SDS), cetyltrimethylammonium bromide (CTAB), and polyethylene glycol anhydrous sugar alcohol monooleate (Tween 80), representing anionic, cationic, and nonionic surfactants, respectively. Batch experiments were conducted to analyze reaction time, pH levels, surfactant type and concentration, and reaction temperature on arsenic adsorption in soils. Then, primary and secondary factors influencing arsenic adsorption were identified through orthogonal experiments. Adsorption kinetic and isothermal adsorption models were discussed. A modified sequential extraction was utilized to analyze arsenic occurrences. The mechanism of arsenic adsorption in soils was elucidated by combining the findings of batch experiments and solid analyses, including Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), scanning electron microscopy (SEM), and contact angle (CA) analyses. This study will enhance our understanding of how surfactants impact the environmental behavior of arsenic in soils, facilitating a more precise assessment of arsenic risk in the presence of surfactants.

Experimental

Soil sample

Soil samples were collected from the surface soil layer (0–20 cm) at an uncontaminated site in Wuhan, China. The surface soil consists mainly of clay with a minor amount of gravel. The samples were dried at room temperature, followed by grinding and sieving through a 20-mesh sieve to remove coarse particles and impurities. Subsequently, the sieved samples underwent further grinding and were passed through a 100-mesh sieve to ensure homogeneity.

Materials and reagents

All the chemicals and reagents utilized in this study are at least analytical grade. Further details regarding the chemicals and reagents used can be found in Text S1.

Experimental procedure

Batch experiments were conducted in 50-mL tubes, each containing 20 mL of mixed solutions and 1.000 g (± 0.001 g) of soil samples, maintained at room temperature (25 ± 2 °C). The pH of As(V) solution, without any adjustment, was measured at 10.16 ± 0.05. Samples were collected at preset times. All experiments were replicated three times to ensure the reliability of data. Deionized water was utilized throughout the experiment and analysis.

Kinetic experiment

The 20 mL of 5 mg/L As(V) solution was mixed with the treated soil in a tube. The tubes were taken at preset times and centrifuged at 4000 rpm for 10 min to collect the supernatant for analysis. The same procedure was followed for kinetic experiments using various concentrations of As(V) (0, 10, 20, 30, 40, and 50 mg/L), with As(V) = 0 mg/L as the experimental blank. With the concentration of surfactant fixed at 200 mg/L, the influence of surfactants on As(V) adsorption was studied under various concentrations. Detailed information on the detection methods of different surfactants can be found in Text S2.

Isothermal adsorption

The soil sample was mixed with 20 mL of As(V) solution at various concentrations (5, 10, 20, 30, 40, and 50 mg/L) and subjected to rotary oscillation for 2880 min. After the reaction, the methods previously described were followed for analysis. Meanwhile, the surfactant concentration remained fixed at 200 mg/L, with varying arsenic concentrations employed for the surfactant experiments. The analysis was conducted using the same procedures as previously mentioned.

Effect of initial pH

The adsorption of arsenic in soil was investigated under various initial pH levels of 5, 6, 7, 8, 9, and 10, with a constant arsenic concentration of 30 mg/L. The reactions were carried out for preset times, and corresponding samples were collected for analysis using the aforementioned method. Additionally, surfactant experiments were conducted utilizing different surfactants at a consistent 200 mg/L concentration. The experimental procedures and testing methods were consistent with those previously described.

Effect of different surfactants

The mixed solution maintained a constant arsenic concentration of 30 mg/L, while the surfactant concentration varied at 20, 50, 100, 200, 500, and 1000 mg/L. The experimental procedures and testing methods were consistent with the aforementioned method.

Orthogonal experiment

The experiment maintains a constant concentration of As(V) at 30 mg/L and a reaction time of 2880 min. The selected factors were surfactant type, concentration, and solution pH, with the adsorption capacity of arsenic in the soil serving as the experimental research index. An L9(33) orthogonal experimental design was employed, and Table S1 displays the level of the orthogonal experimental variables.

Effect of reaction temperature

The mixed solution containing 30 mg/L of arsenic and 200 mg/L of surfactant was kept at different reaction temperature (10, 20, 30, 40, and 50 °C) for 2880 min. The experimental procedures and testing methods followed the aforementioned method.

Soil aging and sequential extraction of arsenic

The experiment maintained a constant arsenic concentration of 30 mg/L, surfactant concentration of 200 mg/L and the reaction time of 2880 min. After the reaction, solid–liquid separation was achieved through centrifugation at 4000 rpm for 10 min. The remaining soil samples were sealed and stored for 4 weeks for aging treatment. The arsenic sequential extraction procedure described by Wenzel et al. was conducted on the soil samples (Wenzel et al., 2001) (Supporting Text S3 and Table S2).

Test method for arsenic

The testing method for arsenic analysis was described in Text S4. Additionally, considering the potential for interference from different types and concentrations of surfactants in the testing process, a standard arsenic addition recovery experiment was conducted. Achieving a 95.8–103.9% recovery rate for the 30 mg/L As(V) solution demonstrates that the surfactants did not significantly interfere with the accurate measurement of arsenic.

Soil characterization methods

To investigate the mechanism of the adsorption effect of surfactants on arsenic in soil, the soil samples were characterized by FTIR, XRD, SEM, CA, and Zeta potential. Detailed information on these soil sample characterization methods can be found in Text S5.

Data processing

  1. (1)

    The adsorption capacity of arsenic.

    The arsenic adsorption capacity C (As) was calculated by Eq. (1).

    $$C(As)=\frac{{(C}_{0}-{C}_{t})V}{m}$$
    (1)

    where V (L) is the volume of arsenic-containing solution added. m (g) is the mass of the soil sample added to the solution. \({C}_{0}\) and \({C}_{t}\) (mg/L) are concentrations of arsenic in the solution at the initial and t time, respectively.

  2. (2)

    The adsorption inhibition rate of arsenic.

    The arsenic adsorption inhibition rate I (As) was calculated by Eq. (2).

    $$I\left(As\right)=\frac{{(C}_{e}-{C}_{e}^{s})}{{C}_{e}}\times 100\%$$
    (2)

    where \({C}_{e}\) and \({C}_{e}^{s}\) (mg/kg) are the equilibrium adsorption capacity of arsenic in soil in the absence and presence of surfactant, respectively.

  3. (3)

    Adsorption kinetic

    Adsorption kinetic models, including pseudo-first order, pseudo-second order, intra-particle diffusion, and Elovich model, were used to evaluate the arsenic adsorption processes in soil. These kinetic models are represented by Eqs. (3)–(6), respectively. The corresponding information for kinetic models was provided in Text S6.

    $${q}_{t}={q}_{e}(1-{e}^{-{k}_{1}t})$$
    (3)
    $${q}_{t}=\frac{{k}_{2}{{q}_{e}}^{2}t}{1+{k}_{2}{q}_{e}t}$$
    (4)
    $${q}_{t}={k}_{i}{t}^\frac{1}{2}+C$$
    (5)
    $${q}_{t}=\frac{1}{{\beta }_{E}}ln\left(1+{\alpha }_{E}{\beta }_{E}t\right)$$
    (6)

    where qe (mg/kg) is the adsorption capacity at equilibrium and qt (mg/kg) is the adsorption capacity of the soil at time t (min). \({k}_{1}\), \({k}_{2}\), and \({k}_{i}\) are the pseudo-first order, pseudo-second order, and the intra-particle diffusion model rate constant, respectively. C is a constant that approximates the thickness of the boundary layer. αE and βE are the coefficients of the Elovich model, representing the initial adsorption rate constant and desorption rate constant, respectively.

  4. (4)

    Isothermal adsorption.

    Langmuir, Freundlich, Temkin, and D-R models were employed to fit the adsorption isotherms of arsenic in soil. These isotherm models are expressed by Eqs. (7)–(10), respectively. The corresponding information for isothermal adsorption models can be found in Text S7.

    $${q}_{e}=\frac{{{Q}_{max}K}_{L}{C}_{e}}{1+{K}_{L}{C}_{e}}$$
    (7)
    $${q}_{e}={K}_{F}{{C}_{e}}^\frac{1}{n}$$
    (8)
    $${q}_{e}=\frac{RT}{{K}_{T}}ln\left({b}_{T}{C}_{e}\right)$$
    (9)
    $${q}_{e}={Q}_{max}{e}^{-{K}_{D}{\varepsilon }^{2}}$$
    (10)

    where qe (mg/kg) represents the adsorption capacity at the equilibrium state and Qmax (mg/kg) is the saturated adsorption capacity. Ce (mg/L) corresponds to the arsenic concentration at the equilibrium state. KL, KF, KT, and KD are the adsorption constant of Langmuir, Freundlich, Temkin, and D-R isotherm model, respectively. n is the constant of adsorption reaction efficiency and \({b}_{T}\) is the Temkin isotherm equilibrium binding constant. \(\varepsilon\) (kJ/mol) is Polanyi potential D-R isotherm model, \(\varepsilon =RTln(1+\frac{1}{{C}_{e}})\). T (K) is the reaction temperature (298 K), R (J/(mol·K)) is the universal gas constant (8.314 J/(mol·K)).

  5. (5)

    Range analysis

    Range analysis is commonly used for analyzing orthogonal experimental data to identify the most sensitive factor affecting the target index. The range analysis of \({R}_{j}\) is defined as the difference between the maximum and minimum values of \(\overline{{K }_{ji}}\) in the j factor, indicating the importance of the factor’s influence (Zhao et al., 2013). \(\overline{{K }_{ji}}\) represents the average target value of each experimental factor at the same level i determined through orthogonal experiments. \({R}_{j}\) and \(\overline{{K }_{ji}}\) are calculated by Eqs. (11) and (12), respectively.

    $${R}_{j}=max\left\{\overline{{K }_{ji}}\right\}-min\left\{\overline{{K }_{ji}}\right\}$$
    (11)
    $$\overline{{K }_{ji}}={K}_{ji}/{k}_{j}$$
    (12)

    where the j (j = A, B, C) represents a certain factor while i (i = 1, 2, 3) corresponds to the level number. Kji refers to the sum of the targeting indexes of all levels in each factor j, and kj is the total levels of the corresponding factor.

Results and discussion

Basic properties of soil samples

Basic properties of soil sample, including pH, density, organic matter content, and the content of heavy metals, were listed in Table S3. The results revealed that the mean concentrations of As, Cu, Cd, Cr(VI), Pb, Ni, and Hg remained below the corresponding risk control values in the Chinese standard limits for soil environmental quality (Du & Li, 2023), indicating that the soil is unpolluted.

Adsorption of arsenic by soil

Adsorption kinetic

After a considerable increment in qt during the initial 120 min, the qt slowly increases until reaching the adsorption equilibrium (Fig. 2a). This difference is ascribed to the quick consumption of active binding sites on particle surfaces through surface diffusion (Ociński et al., 2016). The insufficient binding site made the adsorption rate slower with the adsorption continuing (Das et al., 2014). The equilibrium was achieved at 2880 min. The qe was 70.19, 108.72, 145.49, 205.63, 208.27, and 206.78 mg/kg for initial arsenic concentration of 5, 10, 20, 30, 40, and 50 mg/L, respectively.

Fig. 2
figure 2

a Elovich model fitting and b isothermal adsorption models fitting

Full size image

Many kinetic models, including pseudo-first order, pseudo-second order, intra-particle diffusion, and Elovich models, were utilized to describe the adsorption kinetics of arsenic (Fig. S1). All fitting parameters were listed in Table S4. Based on the obtained correlation coefficients (R2), the Elovich model, with its superior fit (R2 = 0.993–0.999), was selected and depicted in Fig. 2a. These findings imply that the adsorption is primarily governed by heterogeneous chemical adsorption (Peralta et al., 2016), rather than solely relying on surface electrostatic interaction between particles and As(V) (Shen et al., 2017). This conclusion aligns with the findings that studied the adsorption kinetics of arsenic on different soils by Rahman et al. (2019) and Goh & Lim (2004). Notably, the initial adsorption rate (α) did not exhibit an increasing trend with escalating initial arsenic concentration (Table S4). This observation implies the existence of certain unstable behavior concerning the As(V) adsorption process by the soil. In contrast, the desorption constant (β) displayed a declining change with increasing arsenic concentrations, indicating facilitated arsenic desorption at low As(V) concentrations.

Adsorption isotherm

Isotherm analysis was conducted using Langmuir, Freundlich, Temkin, and D-R adsorption models. All fitting outcomes were depicted in Fig. 2b. Table S5 listed all related parameters. The Temkin model (R2 = 0.987) exhibited superior fitting compared to the Langmuir model (R2 = 0.932), the Freundlich model (R2 = 0.943), and the D-R model (R2 = 0.749). Therefore, the arsenic adsorption mechanism in this studied soil is controlled by chemical adsorption characterized by a homogeneous distribution of adsorption energy on the soil particle surface and the linear variation in the heat of surface molecule adsorption (Ramanayaka et al., 2023).

Initial pH

Figure 3 illustrates the impact of initial pH on arsenic adsorption. The qe decreases from 240.10 to 206.63 mg/kg as pH increases from 5 to 10. This decline can be attributed to the change in arsenic species at different pH ranges. Specifically, H2AsO4− is the primary specie within the pH range of 2.3–6.8, whereas HAsO42− is dominated within the range of 6.8–11.5 (Chen et al., 2021). The intensity of positive charge on soil surfaces is declined under the higher pH, which inhibits the adsorption of H2AsO4− and HAsO42− via electrostatic attraction (Yazdani et al., 2016). Additionally, the adsorption energies of H2AsO4−, HAsO42−, and AsO33− are sequentially higher (Chowdhury & Yanful, 2010), meaning that the adsorption of AsO33− at alkaline conditions requires more energy than the adsorption of H2AsO4− at acidic conditions. Therefore, the As(V) adsorption thus becomes increasingly challenging as the pH increases.

Fig. 3
figure 3

Effect of initial pH of solution on arsenic adsorption by soil

Full size image

Effect of surfactant on arsenic adsorption in soil

Type and concentration of surfactants

The effects of three surfactants, SDS, CTAB, and Tween 80, on arsenic adsorption in soil were investigated at different concentrations (Fig. 4a). All three surfactants inhibit arsenic adsorption, with the inhibition getting stronger as the surfactant concentration increases. The qe is 205.63 mg/kg without surfactant, while it decreases to 180.83, 194.44, and 183.20 mg/kg at 20 mg/L of concentrations of SDS, CTAB, and Tween 80, respectively; the values further decrease to 145.20, 164.04, and 161.40 mg/kg when the surfactant concentration is 1000 mg/L. The inhibition primarily arises from the competitive adsorption between As(V) and surfactants on soil surfaces (Zhang et al., 2011). The aggregation of surfactants on soil surfaces, driven by the hydrophobic interaction, can lead to the competition with As(V) for adsorption sites on soil surfaces (Sharma et al., 2011). Hydrophobic interaction can also facilitate the formation of micelle structures between surfactant molecules and As(V), enhancing the solubility of arsenic and reducing its adsorption on soil surfaces. Apart from hydrophobic interaction, different types of surfactants may have distinct adsorption mechanisms. SDS/CTAB (ionic surfactants) can associate with the exchangeable anions/cations within clay matrices, forming a monolayer on the external surface, especially at low surfactant concentrations. However, this monolayer can turn into bilayer (or more) with surfactant concentrations increasing, where surfactant molecules (micelles) are bound together via hydrophobic interaction (Taffarel & Rubio, 2010). In contrast, the adsorption of nonionic surfactants (e.g., Tween 80) is governed by hydrogen bonding (Shen, 2000), considering the lack of ionic and chemisorption groups. Moreover, surfactants can effectively reduce the adsorption equilibrium time from 48 to 36 h (Fig. S2). Figure 4b suggests that the inhibition of surfactants on arsenic adsorption follows the order of SDS > Tween 80 > CTAB. The observed results are further supported by the measured zeta potential values (Fig. S2). The adsorption of SDS led to an increase in the number of negative charges on the soil surface, consequently causing a decrease in zeta potential. Considering the positive charge of CTAB, the zeta potential changed only from − 17.2 to − 16.9 mV, indicating less adsorption of CTAB. For Tween 80, which is uncharged upon adsorption onto soil, the observed change in zeta potential from − 17.2 to − 16.6 mV due to the reduction of arsenic-containing anions adsorption on the soil.

Fig. 4
figure 4

The effect of surfactants on the adsorption of arsenic in soil: a arsenic concentration variation with surfactant concentration in 20–1000 mg/L; b equilibrium adsorption capacity change under different surfactant concentrations (all data is provided in Fig. S3)

Full size image

Adsorption kinetic

The fittings of all studied kinetic models were shown in Fig. S3, and the related parameters were provided in Table S6. Only the fitting of Elovich model was depicted in Fig. 5a–c due to its R2 (0.980–0.997) significantly higher than other models (R2 = 0.704–0.956). The finding implies a limited influence of three surfactants on arsenic adsorption mechanism. According to the fitting parameters of the Elovich model, the α exhibits an enhanced value in the presence of different surfactants. This suggests that surfactants can elevate the initial adsorption rate of arsenic, despite the diminished qe when arsenic adsorption reaches the equilibrium. This phenomenon can be further elucidated by the equilibrium time reducing from 48 to 36 h in the presence of surfactants. Furthermore, the β generally increases with the increase in the surfactant concentration. This indicates that surfactants intensify the desorption of arsenic from the soil, with the desorption intensity escalating with higher surfactant concentrations.

Fig. 5
figure 5

a, b, and c Elovich model fitting with SDS, CTAB, and Tween 80, respectively. d, e, and f Isothermal fitting with SDS, CTAB, and Tween 80, respectively

Full size image

Isothermal adsorption

The effects of SDS, CTAB, and Tween 80 on arsenic isothermal adsorption were shown in Fig. 5d, e, and f, respectively. Notably, the Langmuir (0.978–0.981) and Temkin (0.976–0.987) models exhibit superior fitting performance with higher R2 values compared to Freundlich (0.941–0.955) and D-R (0.731–0.745) models (Table S7) for arsenic adsorption in soil with surfactants. This implies that soil arsenic adsorption is primarily driven by chemical adsorption, with both monolayer and multilayer adsorption taking place within a heterogeneous adsorption system affected by surfactants (Peralta et al., 2016; Bankole et al., 2017; Belhaj et al., 2021a). The behavior is closely related to the hydrophobic effect of surfactants, which occupy the adsorption sites on the surface of soil particles, thereby causing variations in the isothermal adsorption process compared to conditions without surfactants.

Initial pH

The qe increases as pH rises in the presence of surfactants, similar to the finding without surfactants. With increasing pH from 5 to 10, the qe decreases from 223.83 to 149.88 mg/kg with SDS, decreases from 201.83 to 164.99 mg/kg with Tween 80, and decreases from 208.47 to 167.46 mg/kg with CTAB (Fig. 6a), respectively. Figure 6b shows the change in qe as functions of the pH and surfactants. The qe with SDS is remarkably higher than that with CTAB and Tween 80 at pHs < 7, while an opposite tendency is observed at pHs > 7. This difference is attributed to the basicity of SDS, which leads to neutralization reaction occurred in acidic environments. This facilitates the protonation of the negatively charged surface of micelles produced by SDS, resulting in deactivation. The inhibition of SDS on arsenic adsorption thus decreases. On the other side, soil particles, arsenic ions, and SDS are all electronegative in alkaline environments, which might strengthen the inhibition of arsenic adsorption due to the electrostatic repulsion. However, there is little distinction between Tween 80 and CTAB, indicating that hydrophobic interaction plays a more prominent role in arsenic adsorption in soil compared to electrostatic interaction.

Fig. 6
figure 6

The effect of initial pH on the adsorption of arsenic by soil in the presence of surfactants: a arsenic concentration changes in the range of pH 5–10; b equilibrium adsorption capacity (all data is provided in Fig. S4)

Full size image

Orthogonal experimental analysis

The impact of various factors on soil adsorption of arsenic was further evaluated using orthogonal experiments. The L9(33) orthogonal experimental design was conducted to investigate the influence of surfactant type, surfactant concentration, and initial pH on the soil adsorption of arsenic. Table 1 presents the design of the orthogonal experiments and the experimental results of the soil adsorption capacity index for arsenic. The calculated results in Table 1 demonstrate that the initial pH of the solution has the most significant impact on the soil adsorption of arsenic, followed by the effect of surfactant type and the effect of surfactant concentration, as indicated by the range analysis of the three factors RC > RB > RA. These findings provide valuable insights into the mechanism of arsenic adsorption in soil, emphasizing the significant influence of solution pH. Moreover, the observation of RB > RA indicates the dominant role of surfactant’s hydrophobic interaction in affecting arsenic adsorption in soil compared to the electrostatic interaction.

Table 1 Table of orthogonal experimental design and calculation result
Full size table

Reaction temperature

The adsorption behavior of arsenic and surfactants in soils was found to be temperature dependent (Liu et al., 2021; Marzi et al., 2023). Figure 7a illustrates the impact of varying reaction temperatures on arsenic adsorption by soil, demonstrating an increase in adsorption with rising temperatures. Specifically, the soil’s arsenic concentration rose from 141.46 to 250.58 mg/kg as the temperature increased from 10 to 50 °C in the absence of surfactant. Similar trends were observed with SDS, Tween 80, and CTAB, showing increases from 108.04 to 195.64 mg/kg, 120.50 to 226.74 mg/kg, and 124.07 to 248.39 mg/kg, respectively. These increases were attributed to enhanced diffusion rates of arsenic into soil pores and the creation of additional adsorption sites, resulting from bond breakage within soil particles at elevated temperatures (Genç-Fuhrman et al., 2004; Partey et al., 2008). Conversely, the adsorption inhibition rate of different surfactants exhibited variability (Fig. 7b). The adsorption inhibition rate of SDS initially increased (10–30 °C) and then decreased (30–50 °C), possibly due to SDS destabilization at excessively high temperatures leading to decreased inhibition (Liu et al., 2021). Yekeen et al. (2017) and Liu et al. (2019) observed a similar phenomenon where the adsorption of anionic surfactants by clay initially increases and then decreases with increasing temperature. Tween 80 and CTAB showed a decrease in inhibition with increasing temperature, likely due to the reduced viscosity and activity of surfactants, limiting their adsorption onto soil particle surfaces (Belhaj et al., 2022; Bera et al., 2013). Remarkably, the impact of elevated temperature was particularly pronounced in CTAB, with the inhibition rate decreasing linearly from 12.19% at 10 °C to just 0.88% at 50 °C. This phenomenon may be due to higher temperatures leading to dehydration of CTA B (Gürses et al., 2009).

Fig. 7
figure 7

The effect of temperature on the adsorption of arsenic by soil in the presence of surfactants: a and b, arsenic concentration and adsorption inhibition rate at equilibrium, respectively

Full size image

Aging and sequential extraction analysis

The sequential extraction procedure (SEP) is commonly conducted to investigate the occurrence of heavy metals, describing the migration and transformation, toxicity, and bioavailability of arsenic in soils (Zhuang et al., 2023). The influence of surfactants on the occurrence and stability of arsenic in soil after aging was further evaluated using SEP (Fig. 8). The main conclusions are listed as below:

  1. 1)

    The fractions of F1 and F2 (characterized by high mobility, toxicity, and availability) are evidently reduced in soil after four-week aging, despite the unchanged concentration of adsorbed arsenic. This suggests the transformations of F1 and F2 into F3, F4, and F5.

  2. 2)

    The addition of SDS, CTAB, and Tween 80 can mainly reduce the fractions of F3 and F4, with the fraction of F2 in soil increasing from 7.06% to 7.90%, 8.95%, and 11.64%, respectively. The F2 displays the higher bioavailability than F3 and F4 due to its higher mobility (Kim et al., 2014). Simultaneously, the residual arsenic (F5) increased by 2.42%, 1.10%, and 1.08%, respectively, compared to 4.52% without surfactants.

  3. 3)

    Compared to the aged soil, the fractions of F4 increased by 6.02%,7.65%, and 6.31%, respectively, compared to 10.32% without surfactants. The fractions of F5 increased by 3.85%,1.94%, and 1.89%, respectively, compared to 9.08% without surfactants. Notably, due to the F4 and F5 with higher stability (Kumpiene et al., 2021), the surfactants can exacerbate the instability of arsenic during soil aging.

Fig. 8
figure 8

The effect of arsenic fractions on the soil aging in the presence of surfactant (F1: exchangeable, F2: specifically sorbed, F3: amorphous oxide-bound, F4: crystalline oxide-bound, F5: residual). S0-As and AS0-As, original and aging soil after arsenic adsorption; SS-As and ASS-As, no aging and aging soil after arsenic adsorption with SDS; SC-As and ASC-As, no aging and aging soil after arsenic adsorption with CTAB; ST-As and AST-As, no aging and aging soil after arsenic adsorption with Tween 80

Full size image

Consequently, the presence of surfactants reduces the proportion of stable arsenic fractions, enhancing arsenic bioavailability, thereby increasing the potential risk of soil arsenic pollution.

Mechanism analysis

To elucidate the influence of surfactants on arsenic adsorption in soil, characterization methods, including FTIR, XRD, SEM, and CA, were employed on soil samples. Among the experimental groups, the one containing the surfactant SDS exhibited the most significant impact on arsenic adsorption in the soil, posing the highest potential environmental risks. Consequently, the SDS experimental group was characterized as providing a robust scientific foundation for environmental management (samples information, S0: original soil; S0-As: soil after arsenic adsorption; AS0-As: aging soil after arsenic adsorption; SS-As: arsenic adsorption in soil with SDS; ASS-As: aging soil after arsenic adsorption with SDS).

FTIR analysis

The FTIR results demonstrate modifications in functional groups in the soil due to aging and surfactant addition (Fig. 9a). Compared to samples without aging, the aging exhibited increased peak intensity near 3621 cm−1, related to Si–O-Si, Si–O, -OH, H–O-H, and O–H/Si–OH groups (Huang et al., 2017). This alteration implies that the arsenic adsorption onto the soil is inclined towards forming outer-sphere surface complexes with As-O ligands after aging (Gao et al., 2020). Notably, the alkyl groups (-CH) stretching vibrations emerge at 2958, 2929, and 2857 cm−1 (Fig. 9b) (Stockenhuber et al., 2001; Tunç et al., 2012), and the peak intensity of samples containing surfactant is stronger than non-surfactant, evidencing the adsorption of surfactant in soil. The peak near 1384 cm−1 is assigned to the Fe-polymer conjugate (Tang et al., 2022). It is well evidenced that the arsenic can integrate with Fe–O(H) bonds to form inner-sphere complexations, so the intensity of this peak in AS0-As and ASS-As samples are higher than in S2 and S4 (Fig. 9c), which can be attributed to soil aging. In addition, there is a negligible difference in peak intensity between the surfactant and non-surfactant samples, such as S0-As and SS-As, AS0-As, and ASS-As, indicating that SDS presents minimal interference in the iron-arsenic interaction in the soil. The peak near 977 cm−1 is referring to the asymmetric stretching vibrations of Si–O (Zhang et al., 2018). However, this observed peak shifts to ~ 970 cm−1 in aged samples, accompanied by an elevation in peak intensity (Fig. 9d). This shift indicates the asymmetric stretching vibration attributed to Si–O-Al (Cao et al., 2020), signifying the interaction between aluminum and silicate during aging, culminating in the generation of aluminosilicate precursor.

Fig. 9
figure 9

Characterization of soil samples, (a) FTIR data of samples. (b), (c) and (d) Normalized spectrum of different soil samples. (d) XRD data of samples

Full size image

XRD analysis

As depicted in Fig. 9e, the soil (S0) consists of quartz, mica, albite, muscovite, dolomite, and silicates such as, iron magnesium silicate, potassium aluminum silicate, and calcium aluminum silicate. No new characteristic peak was found after arsenic adsorption. The interaction between potassium-bearing minerals (e.g., mica, muscovite, potassium aluminum silicate) and adsorbed arsenic during the aging facilitates the formation of potassium arsenate in AS0-As. However, no new peak was detected in aged SDS-containing samples, indicating the inhibition of crystallizing potassium arsenate by SDS addition. This phenomenon can be attributed to the alteration of interface properties (e.g., surface tension and interface charge distribution) in soil particles. Consequently, there is a reduction in the interaction between arsenic and mineral components in the soil, ultimately delaying or inhibiting the formation of arsenic-containing substances. This conclusion aligns with SEP results, where the fractions of F4 and F5 are reduced when the aging occurs with surfactants.

SEM analysis

Generally, the formation of rough surfaces is attributed to the instability structure and the surface attachment in soils (Fig. 10a and b). The addition of surfactants and the aging could decompose or dissolve these attachments, leading to a smooth surface of soil particles (Fig. 10g, j and m). This finding is more evident at higher magnification (Fig. 10k and n), where some flaky structures and irregular particles (Fig. 10b, e and h) on soil surfaces were diminished. Meanwhile, a denser surface structure in ASS-As was observed (Fig. 10n). This may be due to the hydrophobicity of surfactants, which can induce the aggregation of soil particles (Conte et al., 2005). Additionally, the strengthening bonds formed by soil composition during the aging due to the strengthened binding formed by soil components during the aging process, contributing to the formation a more compact soil surface structure (Vlčková & Hofman, 2012).

Fig. 10
figure 10

SEM and CA Characterization of soil samples: a, b, and c for S0; d, e, and f for S0-As; g, h, and i for AS0-As; j, k, and l for SS-As; m, n, and o for ASS-As

Full size image

CA analysis

The CA characterization results (Fig. 10c, f, I, l, and o) are evident that adding surfactant and aging treatment reduce CA of soils. The surfactant-induced reduction of CA can be attributed to two factors. Firstly, SDS possesses hydrophilic properties. The hydrophilic groups interact with water molecules upon adding to the soil, promoting water wetting on soil particles and decreasing the contact angle (Ogunmokun et al., 2020). Secondly, the surfactants compete with arsenic for adsorption, occupying the adsorption sites to form an adsorption layer of surfactant molecules on soil surfaces, lowering the surface tension, and decreasing CA. After aging, the decrease in CA is attributed to potential alterations in the chemical and physical properties of the soil particle surfaces, such as the decomposition of organic matter and mineral recombination. These changes may lead to an increase in the surface energy of the soil particles, promoting easier wetting of the soil surface and resulting in a reduction in CA (Bachmann & McHale, 2009). Moreover, the soil aging treatment can induce changes in soil structure, including pore size, connectivity, and surface roughness. The previous SEM results indicate that both aging and surfactants contribute to reducing the roughness of the soil particles. Generally, the lower roughness of soil particles indicates a lower CA (Bryant et al., 2007).

Environmental implications

Soil arsenic pollution is a severe and widespread problem globally. Because of their low toxicity, surfactants have not yet received sufficient attention, despite their prevalent distribution in soils, leading to an underestimation of their influence on arsenic behavior in soils. The findings here indicate that the addition of surfactants can restrict the arsenic adsorption in studied soil and shorten the time to reach adsorption equilibrium. Considering the temperature dependence of arsenic and surfactant adsorption in soils, it is crucial to consider seasonal variations when evaluating arsenic adsorption behavior. Moreover, surfactants can impede the formation of stable arsenic forms, such as crystalline oxide-bound arsenic and residual arsenic, thus increasing the bioavailability of arsenic with the higher health risk for humans. Additionally, surfactants effectively decrease the surface tension of soils, enhance its hydrophilicity, facilitate the infiltration of precipitation into soils, and promote the migration of arsenic through runoff and seepage, thereby expanding the diffusion of arsenic pollution. Therefore, recognizing the potential impacts of surfactants and re-evaluating the soil’s adsorption capacity for arsenic are of great environmental significance for arsenic pollution remediation and risk assessment.

Conclusion

The study explores the influence of three representative surfactants (SDS, CTAB, and Tween 80) on arsenic adsorption in soil. The findings demonstrate that surfactants shorten the adsorption equilibrium time of arsenic in soil from 48 to 36 h, while the adsorption process still adheres to Elovich kinetic. However, the adsorption isotherm of arsenic in soil changes from a single Temkin model without to an excellent fit with both the Langmuir and Temkin models. Moreover, the Elovich kinetic parameters reveal that surfactants can enhance the initial adsorption rate and promote arsenic desorption in studied soil. The inhibition of surfactants on arsenic adsorption in soil primarily stems from competitive adsorption, electrostatic interaction, and hydrophobic interaction. The SDS exhibits the most significant inhibitory impact among the surfactants, followed by Tween 80 and CTAB. The changes in zeta potential changes under various surfactant further elucidate this phenomenon. Additionally, within the pH range of 5 to 10, the inhibitory effect of surfactants increases with rising pH. Based on the result of L9(33) orthogonal experiment, the primary factor that influences arsenic absorption in soil is solution pH, with surfactant concentration and surfactant type being the secondary factors, respectively. Thus, the pH exerts a dominant influence on arsenic adsorption, and the hydrophobic interaction of surfactants outweighs the impact of the electrostatic interaction. CTAB showed a particularly pronounced effect with increasing temperature, with the inhibition rate decreasing linearly from 12.19% at 10 °C to just 0.88% at 50 °C. The SEP reveals that surfactant decrease the proportion of stable arsenic (crystalline oxide-bound arsenic and residual arsenic), enhancing arsenic bioavailability. The FTIR, XRD, SEM, and CA characterization analyses further confirm that the addition of surfactant alter the number and abundance of functional groups (Si–O, Si–O-Al, and -CH), inhibit the formation of arsenic-containing minerals, improve soil particle density, surface smoothness, and hydrophilicity, ultimately influencing the adsorption of arsenic in soils. This study provides valuable insights into the effects of surfactants on arsenic adsorption in soils, highlighting their potential role in increasing the mobility and bioavailability of arsenic, allowing for a more accurate assessment of arsenic risk in the presence of surfactants.