Effects of copper and lead on the sorption and desorption behaviors of benzene onto humic acids and black carbons

BTEX (benzene, toluene, ethylbenzene, and xylene) is a high-priority pollutant of concern in urban soils (Eom et al. 2023). They are widely used in industrial applications and can be discharged into soil and groundwater through improper practices or accidental spills (Shahrokhi-Shahraki et al. 2020). The strong volatility and high water solubility of BTEX are critical factors influencing their fate in soil and groundwater, posing potential health risks to humans and ecological risks to the environment. Notably, benzene is a known human carcinogen that causes leukemia and is therefore regulated as a priority pollutant by the United States Environmental Protection Agency (Smith et al. 2000). Given that many BTEX-polluted sites also contain large amounts of heavy metals (Mohan et al. 2020; Shahrokhi-Shahraki et al. 2020), it is crucial to understand the effects of heavy metals on the fate of BTEX and to further determine the effects on their environmental risks in soils.

Sorption plays a critical role in the transport and transformation of BTEX in soils (Wang et al. 2022). Soil organic matter (SOM) is widely acknowledged as the primary sorption medium for pollutants, including BTEX and heavy metals (Zhao et al. 2019). The primary sorption mechanism for BTEX by SOM is believed to be the hydrophobic partitioning (Shih & Wu 2002; Tang et al. 2019), while complexation is considered the principal mechanism for heavy metals (Li et al. 2010). However, the distinct sorption mechanisms of BTEX and heavy metals may mutually interfere, ultimately affecting their fate and bioavailability in soils. Previous studies showed that the presence of heavy metals can lead to varied effects on the sorption behaviors of different organic pollutants. Some studies suggested that heavy metals may inhibit the sorption of certain organic pollutants onto SOM. For instance, Wang et al. (2009) reported that Cu²⁺ and Pb²⁺ ions significantly reduced the sorption of 2,4,6-trichlorophenol onto black carbon (BC) and two humic acids (HAs) due to the competitive sorption for the carboxylic, hydroxylic, and phenolic groups. Liu et al. (2012) similarly found that the hydration shells formed by the surface complexation of Cu²⁺ ions had a diminishing effect on the sorption of mefenacet by HAs. Furthermore, it was found that the pore blockage played a crucial role in reducing the sorption capacity of p-nitrophenol onto BC in the presence of Cu²⁺, Pb²⁺, and Zn²⁺ ions (Wang et al. 2011). On the other hand, the coexistence of heavy metal cations can also enhance the sorption of organic pollutants through other mechanisms, such as cation–π interaction between benzene rings and metal cations (Ali et al. 2022), or by decreasing the hydrophobicity of BC with soft cations like Ag⁺ ions (Chen et al. 2007). Therefore, the effects of coexisting heavy metal cations on the sorption behaviors of organic pollutants are often a result of diverse mechanisms.

Nevertheless, the effects of coexisting heavy metal cations on the sorption and desorption behaviors of BTEX onto SOM are not fully understood, which introduces significant uncertainty to the environmental risk assessment of BTEX in soil. Benzene serves as the basic unit of BTEX and can represent certain characteristics of other aromatic compounds. Copper (Cu²⁺) and lead (Pb²⁺) are prevalent heavy metals that often occur in high concentrations and may significantly affect the interactions between SOM and benzene. In this study, we investigated the sorption and desorption behaviors of benzene onto several SOM in the presence of Cu²⁺ and Pb²⁺ ions. This research aims to improve the understanding of the fate and bioavailability of BTEX in soil, providing valuable insights for environmental risk assessment. Fourier transform infrared spectroscopy (FTIR) and the site energy distribution (SED) model were also employed to elucidate the potential mechanisms through which Cu²⁺ and Pb²⁺ ions affect the sorption and desorption behaviors of benzene onto SOM.

The organic reagents, including benzene, methanol, and n-hexane, were purchased from Aladdin BioChem Technology Company Limited (Shanghai, China). The benzene has a purity exceeding 99.7%, a vapor pressure of 12.7 kPa, and water solubility of 1,780 mg L⁻¹ at 25 °C, which are critical physical properties affecting its sorption and desorption behaviors. The methanol and n-hexane solvents used in this study are of chromatographically pure quality. The inorganic reagents, such as heavy metal salts (CuCl₂·2H₂O and PbCl₂), NaOH, and HCl, were of analytical grade and were procured from Sinopharm Chemical Reagent Company Limited (Shanghai, China).

Two experimental HAs, namely, THA (obtained from Tianjin Guangfu Fine Chemical Research Institute, Tianjin, China) and SHA (obtained from Shanghai Yuanye Biotechnology Company Limited, Shanghai, China), were purified following the method of the International Humic Substances Society (IHSS) and Johnson et al. (2002). BCs were produced from rice straw (DBC) and corn straw (YBC) through oxygen-limited pyrolysis at 600 °C for 4 h. The products were treated with 1 mol L⁻¹ HCl four times, before purified in the mixture of HCl and HF (1:1). Finally, the BCs were washed with distilled water to remove soluble salts.

The element contents (C, H, O, N, and S) of the experimental SOM were determined using the elemental analyzer (Vario MACRO, Elementar, Germany), while the ash contents were analyzed by burning them in a muffle furnace at 800 °C for 4 h (Jin et al. 2018). The specific surface area (SSA) and pore structure of the SOM were determined by nitrogen adsorption–desorption experiments conducted at 77 K using an AutoChem 2950 HP Chemisorption Analyzer (Micromeritics, USA). In addition, the FTIR spectra of the experimental SOM at different sorption processes were scanned in the transmission mode from 4,000 to 400 cm⁻¹ (Nicolet 6700, Thermo Corp., USA).

The batch experiments were conducted in 40 mL screw-cap brown glass vials with silicone rubber septum-Teflon liners. One hundred milligrams of the experimental SOM were mixed with 40 mL background solution (containing 0.01 M CaCl₂ and 0.1% methanol). The concentration of metal cations was held constant (0.01 M), while the amounts of benzene varied during sorption experiments. The mixed solution was maintained at 40 mL in the vials, which had a negligible headspace volume. In addition, the pH was set to 5 ± 0.1 by adding HCl or NaOH solution to eliminate the effects of hydrolytic acidification caused by the metal cations. Subsequently, all vials were sealed tightly before being rotated end-over-end for 4 h to reach sorption equilibrium (determined by preliminary experiments) at 20 rpm and 25 °C.

Desorption experiments were conducted using a sequential decant–refill technique immediately after the completion of sorption experiments. Specifically, 15 mL aliquots of the supernatant were quickly replaced with the same volume of fresh background solution. The resealed vials were then shaken for an additional 4 h. This desorption procedure was repeated twice.

Sorption kinetic experiments were performed in vials containing 100 mg L⁻¹ benzene and 0.01 M metal cations. The sealed vials were rotated for 5, 15, 30, 60, 120, 240, 360, 480, and 600 min, respectively.

All experiments were repeated three times with a control group (lacking the SOM) employed for each trial. After each reaction process, the vials were centrifuged at 2,000 rpm for 10 min, and the benzene in the supernatant was extracted by n-hexane for further analysis.

The amount of benzene was determined by GC-FID (6890A, Agilent Technologies, USA). The injection and detection temperatures were set at 200 and 280 °C, respectively. The column temperature was initially maintained at 40 °C for 4 min and subsequently increased to 100 °C at 30 °C min⁻¹. The concentration of benzene was quantified by peak area integration.

Table 1 presents the elemental composition, ash contents, SSA, and pore capacity of the experimental SOM used in this study. The properties of the experimental SOM are evidently different. YBC and DBC have higher carbon contents than THA and SHA. The H/C atomic number ratio is usually applied to indicate the degree of aromaticity, with a smaller H/C ratio corresponding to a higher aromaticity of SOM (Hur et al. 2009). Therefore, the order of aromaticity among the four SOM samples was YBC > DBC > THA > SHA. Polarity is also an important property of SOM. YBC had the lowest polarity due to the lowest atomic ratios of O/C and (N + O)/C (Wang et al. 2017). Moreover, YBC had the fewest hydrophilic groups among the experimental SOMs, as indicated by its lowest atomic of O/C (Wang et al. 2011). The SSA, total pore volume (V_T), and micropore pore volume (MC) of the two BCs were much higher than those of the two HAs, while their average pore size (APZ) was smaller. The SSA, V_T, and MC of four SOM were ranked as YBC > DBC > THA > SHA, while the APZ showed THA > SHA > DBC > YBC. THA exhibited higher MC and APZ than SHA. This difference may be attributed to the greater number of macropores in THA, as evidenced by V_T. These differences in the properties of four SOM would lead to different sorption characteristics for benzene.

The fitting results (Table 2) demonstrated that the Langmuir, Freundlich, and DA models provided a good fit to the experimental data. According to the fitting results, the sorption characteristics of benzene onto four SOM were different. YBC with the higher aromaticity and lower polarity showed the stronger affinity for benzene sorption, indicating that the sorption capacity was positively correlated with aromaticity and negatively correlated with the polarity of the sorbent.

The sorption isotherms were best fitted by the DA model, indicating that the Dubinin–Polanyi theory (microporous filling theory) best explained the sorption process of benzene onto SOM. Based on the fitting results of the DA model, YBC and DBC exhibited significantly higher Q⁰ for benzene, which could be attributed to their more abundant pore structures. The Q⁰ predicted by the DA model was smaller than Q_m predicted by the Langmuir model, indicating that the sorption mechanisms of benzene onto SOM included not only pore filling but also other mechanisms such as surface adsorption.

The fitting results from the Freundlich model indicated that all four SOM exhibited nonlinearity in benzene sorption, and YBC showing the highest nonlinearity, followed by DBC, THA, and SHA. The nonlinearity coefficient N is associated with both the energy distribution of sorption sites and the physical structure and maturity of the SOM surface (Weber et al. 1999). Therefore, the smallest N value of YBC suggests that it has the most inhomogeneous energy distribution of the sorption sites and may have a higher maturity and denser structure (Zhang et al. 2009). According to Table 2, YBC had the highest K_F, also indicating that it had the highest affinity for benzene. This may be attributed to the larger SSA and greater pore structure of YBC, thereby offering more sorption sites for benzene.

The Langmuir model is often employed to estimate the maximum sorption capacity (Q_m) of the sorbent onto sorbate. The results indicated that SOM with a large SSA and more porous structure had a higher Q_m for benzene. And the order of polarity and hydrophilic functional group content of four SOM is exactly opposite to the order of Q_m, indicating that the benzene sorption onto SOM might occur in nonpolar carbon and hydrophobic regions.

Among them, K (L mol⁻¹) is the equilibrium constant of the sorption process, which is converted from the Langmuir model parameter K_L (L mg⁻¹).

The ΔG value below 0 indicates that the sorption process is spontaneous, and the smaller value reflects a more energetically favorable sorption process (Pavlović et al. 2022). The calculated results of the ΔG were ranged from −14.15 to −12.73 kJ mol⁻¹, confirming that Cu²⁺ and Pb²⁺ ions did not change the thermodynamically favorable and spontaneous of the sorption processes. The calculated results fell within the range of −20 to 0 kJ mol⁻¹, indicating that the physical adsorption of benzene occurred onto the four SOMs in the presence of Cu²⁺ and Pb²⁺ ions (Xu et al. 2021). The reduction in Gibbs free energy with the addition of Cu²⁺ and Pb²⁺ ions also revealed that these cations inhibited the adsorption of benzene onto organic matter.

The fitting results showed that the sorption rates (k₁ and k₂) of benzene on both two HAs decreased after adding Cu²⁺ or Pb²⁺ ions (Table 3). As for HAs, the pseudo-first-order kinetic model fitted better, indicating that the hindrance of the diffusion process by Cu²⁺ or Pb²⁺ ions was the main reason for the inhibiting effects on benzene sorption. In contrast, the fitting results reflected that the sorption processes of benzene onto the two BCs were governed by both two stages of fast and slow sorption, as evidenced by the better fitness for the pseudo-second-order kinetic model. The addition of Cu²⁺ or Pb²⁺ ions inhibited the diffusion and pore-filling processes, thereby reducing the sorption rate of benzene on BCs.

The results indicated that the desorption of benzene exhibited hysteresis in all cases. The changing trends of HI values suggested that the desorption process was less hysteretic when benzene concentration increased gradually, but became more hysteretic at high concentration. These trends can be explained by the different sorption processes of benzene onto solid SOM (Chang et al. 1997; Wang et al. 2009). Briefly, a large percentage of benzene molecules preferentially tend to bind with the high-energy sites onto the surface of SOM, resulting in weaker desorption at lower concentrations. As the concentration gradually increases, more benzene molecules would be desorbed easily since they tend to occupy the low-energy binding sites through forming multilayer adsorption or partitioning into SOM. At the highest concentration, some benzene molecules may penetrate the SOM matrix and adsorb onto additional high-energy sites of the interior surface, which makes the desorption of benzene difficult again.

The effects of Cu²⁺ and Pb²⁺ ions on the benzene desorption from four SOM were similar. Cu²⁺ and Pb²⁺ ions could promote the desorption of benzene as indicated by the HI values. This suggested that Cu²⁺ and Pb²⁺ ions could shield the high-energy sites for benzene sorption. Furthermore, the HI values of HAs were generally higher than those of BCs, indicating that the benzene desorption from HAs was relatively easier. This could be attributed to the less high-energy sites on HAs.

The vibrational sorption peaks of the benzene ring C = C skeleton around 1,580 cm⁻¹ for four SOMs were shifted after the sorption experiments. The sorption peaks of the SOM samples after the benzene sorption were red shifted, possibly due to the π–π electron donor–acceptor (π–π EDA) interaction between the benzene molecule and the benzene ring of SOM (Dai et al. 2018). The sorption peaks of the SOM samples after the sorption of Cu²⁺ and Pb²⁺ ions were also shifted, likely due to the cation–π bonding (Nzediegwu et al. 2021), which can affect the electron cloud distribution of the benzene ring skeleton structure (Kawamoto 2017). In addition, the sorption peaks of the aromatic ring C-H bending vibration around 790 cm⁻¹ in the SOM samples adsorbed with benzene, Cu²⁺, and Pb²⁺ ions were blue shifted, further confirming that benzene, Cu²⁺, and Pb²⁺ ions were successfully sorbed onto the SOM samples (Ringwald & Pemberton 2000).

The FTIR spectra of all four SOM samples had the broadband around 3,400 cm⁻¹ corresponding to the stretch of the -OH group. The intensity of the band (-OH group) of four SOM samples after benzene sorption did not show significant changes, while the sorption peaks of the SOM samples with adsorbed heavy metals were weakened, indicating that the -OH group was involved in the sorption process of Cu²⁺ and Pb²⁺ ions (Cwielag-Piasecka et al. 2018). Additionally, the enhanced peaks (1,380 cm⁻¹) related to the symmetric stretching vibrations of the ⁻COO⁻ group were observed in two HAs, suggesting that the carboxyl groups also played a role in the sorption process of Cu²⁺ and Pb²⁺ ions (Zhao et al. 2019). The sorption peaks of the four SOM samples before and after benzene sorption showed variations in the region concentrated at 1,580 and 790 cm⁻¹, indicating that the sorption sites for benzene were mainly in the aromatic carbon region and the sorption forces included hydrophobic interactions (Dai et al. 2018).

Based on the results of FTIR analysis, the mechanisms of benzene sorption onto SOM include π–π EDA interaction and hydrophobic partitioning. Cu²⁺ and Pb²⁺ ions would interact with the functional groups of SOM, thus affecting the interaction between SOM and benzene.

Previous experiments have shown that Cu²⁺ and Pb²⁺ ions predominantly form inner complexes through interactions with carboxyl or hydroxyl groups (Wang et al. 2009). The binding of Cu²⁺ and Pb²⁺ ions to the acidic functional groups of SOM can act as a bridge bond, thereby reducing particle and pore size of SOM in the reaction system. Combining the results of FTIR and the SED model, it was speculated that the hydrated layer of adsorbed Cu²⁺ and Pb²⁺ ions would occupy the surface of SOM and prevent the interactions between the sorption sites and benzene molecules, thereby inhibiting the sorption process of benzene (Chen et al. 2008; Wang et al. 2011).

Moreover, the hydrated Cu²⁺ and Pb²⁺ ions may occupy and block the pores of SOM, thus inhibiting the pore-filling process of benzene. The diameter of the benzene molecule is 0.59 nm (Lu et al. 2020), while the diameters of the hydrated Cu²⁺ and Pb²⁺ ions are 0.40 and 0.42 nm, respectively (Nightingale 1959). Based on the results of the pore size analysis, it can be concluded that most of the pores in the four SOM can accommodate benzene molecules, as well as hydrated Cu²⁺ and Pb²⁺ ions. Therefore, if some pores on the SOM surface are occupied by hydrated Cu²⁺ or Pb²⁺ ions, the sorption of benzene may be inhibited because benzene molecules can hardly penetrate through the hydration shell of the metal cations (Liu et al. 2013).

Based on the aforementioned discussions, it can be speculated that the inhibition mechanisms of coexisting Cu²⁺ and Pb²⁺ ions on the benzene sorption by SOM are as follows: (1) Cu²⁺ and Pb²⁺ ions adsorb by the acidic function groups of SOM, reducing the particle and pore size of SOM via the bridging effect, (2) the hydrated layer of adsorbed Cu²⁺ and Pb²⁺ ions occupy the surface of SOM and hinder the interaction between benzene molecules and the sorption sites of SOM, and (3) the adsorbed Cu²⁺ and Pb²⁺ ions block the pores of SOM, thereby inhibiting the pore-filling process of benzene.

This study elucidated that the coexistence of Cu²⁺ and Pb²⁺ ions can inhibit benzene sorption and make desorption less hysteretic onto HA and BC. Combined with the results of FTIR and the SED model, it was speculated that the unfavorable structure for benzene sorption was formed by Cu²⁺ and Pb²⁺ ions. The findings of this study suggested that the presence of Cu²⁺ and Pb²⁺ ions may enhance the mobility of benzene and increase its environmental risks in soils. This study contributes to a better understanding of the environmental fate of BTEX in soil.

The authors gratefully acknowledge the State Key Laboratory of Biogeology and Environmental Geology of China University of Geosciences (Wuhan) for data testing during the experiments.

Zhi Tang: entire experimental works, data evaluation, writing, editing, and funding acquisition. Sen Yang: consulting, data evaluation, and writing. Yilian Li: consulting, project administration, and funding acquisition. Juan Du: data evaluation and writing. Yangfu Xiong and Shengbo Fu: consulting and project administration. All authors read and approved the final manuscript.

This research was supported by the National Natural Science Foundation of China (NSFC, No. 42077186), the Hubei Provincial Natural Science Foundation of China (No. 2023AFD222), the Science and Technology Project of Hubei Geological Bureau (No. KJ2023-30), and the Science and Technology Project of Six Geological Team of Hubei Geological Bureau (No. DKJ2022-02).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.