Application of a Novel Bifunctionalized Magnetic Biochar to Remove Cr(VI) from Wastewater: Performance and Mechanism

: Biochar adsorption has emerged as a favorable and environmentally friendly approach for removing metals such as chromium (Cr) from wastewater. However, the use of pristine biochar (PBC) is limited due to its ﬁnite adsorptive capacity, selectivity, and potential for secondary pollution. In this study, a novel bifunctionalized magnetic biochar (BMBC) was fabricated by incorporating cystamine as a ligand and glutaraldehyde as a crosslinker into alkali-treated magnetic biochar (MBC). This chemical modiﬁcation introduced numerous amino groups and disulﬁde bonds onto the surfaces of BMBC. The biochar adsorbents’ surface morphologies, crystal structures, and texture properties were characterized using SEM, XRD, and N 2 adsorption-desorption techniques. The speciﬁc surface area was determined using the BET method. Furthermore, the surface functional groups and elemental compositions before and after adsorption were analyzed using FTIR and XPS, respectively. The results demonstrated higher Cr(VI) removal efﬁcacy of BMBC (100%) than MBC (72.37%) and PBC (61.42%). Optimal conditions for Cr(VI) removal were observed at a solution pH of 2, a temperature of 50 ◦ C, a reaction time of around 1440 min, and an initial adsorbate concentration of 300 mg/L. The sorption process followed a chemical mechanism and was controlled by monolayer adsorption, with a maximum adsorption capacity of 66.10 mg/g at 50 ◦ C and a pH of 2, as indicated by the larger ﬁtting values of the pseudo–second-order and Langmuir models. The positive ∆ H o and ∆ S o values and negative ∆ G0 values suggested a spontaneous and endothermic Cr(VI) adsorption process with high randomness at the solid/liquid interface. The removal of Cr(VI) was attributed to the reduction of Cr(VI) into Cr(III) facilitated by the introduced amino acids, sulfur, and Fe(II), electrostatic interaction between Cr(VI) in the solution and positive charges on the adsorbent surface, and complexation with functional groups. The presence of co-existing cations such as Cu(II), Cd(II), Mn(II), and K(I) had little effect on Cr(VI) removal efﬁciency. At the same time, the co-existence of anions of Cl − , NO 3 − , SO 42 − , and HPO 42 − resulted in a 7.58% decrease in the Cr(VI) removal rate. After ﬁve consecutive adsorption/desorption cycles, BMBC maintained a high Cr(VI) removal rate of 61.12%. Overall, this novel BMBC derived from rice straw shows great promise as a biosorbent for treating Cr(VI) in wastewater.


Introduction
Aquatic heavy metals control has gained significant attention worldwide due to the extreme toxicity of these metals to living organisms, their high potential for bioaccumulation and biomagnification, as well as their extensive use in various industries [1,2]. Chromium novel bifunctionalized magnetic biochar composite (BMBC) was developed for removing Cr(VI) from wastewater by incorporating cystamine as a ligand and glutaraldehyde as a crosslinker into alkali-treated MBC, and its properties, adsorption performance, and mechanisms of Cr(VI) removal were investigated in detail. For comparison, BMC was obtained from the co-pyrolysis of PBC and a mixed solution of Fe(III)/Fe(II), while the pyrolysis of dried rice straw powder prepared PBC. It was hypothesized that BMBC would exhibit superior Cr(VI) adsorption due to increased binding sites on its surface, stronger complexation, and electrostatic interaction between Cr(VI) and surface functional groups. The specific goals of this study were: (1) to examine the determining properties of BMBC for Cr(VI) adsorption; (2) to compare the adsorption performance of BMBC with that of BMC and PBC under varying solution pH, reaction time, temperature, initial adsorbate concentration, and the presence of co-existing anions and cations; and (3) to elucidate the mechanisms underlying Cr(VI) adsorption.

Fabrication of Biochar Adsorbents
Three biochar adsorbents, BMBC, MBC, and PBC, were prepared for subsequent experiments. PBC was fabricated by heating 20 g of feedstock in a box resistance furnace at 600 • C for 3 h and sieving the resulting materials through a 200-mesh sieve. MBC was prepared via pyrolysis as follows: 15 g of raw rice straw powder was dispersed in a mixture of 3 g FeCl 3 ·6H 2 O and 3 g FeSO 4 ·7H 2 O in 300 mL deionized water, shaken well at room temperature, then subjected to ultrasonic treatment for 2 h, immersed for 24 h, dried at 70 • C, and then pyrolyzed at 600 • C for 3 h in a box resistance furnace. The resulting MBC material was sieved through a 200-mesh sieve. Following the method described by Zhang et al. (2013), BMBC was obtained by chemically modifying alkali-treated BMC using cystamine and glutaraldehyde as ligands and crosslinkers, respectively. Initially, 10 g of MBC was mixed with 100 mL of 3 mol/L NaOH (pH = 13.78) for alkali treatment. After 2 h of reaction at 60 • C, the alkali-treated MBC particles were washed with ethanol and deionized water until the pH value was nearly neutral. The particles were then oven-dried at 70 • C. Subsequently, a 50 mL aqueous solution of glutaraldehyde (2%, v/v) was added to 7 g of the alkali-treated MBC, and the mixture was stirred for 4 h at room temperature. Then, 1.5 g of cystamine dihydrochloride was gradually added, and the stirring was continued for 12 h. Finally, the solid BMBC was separated using an external magnet and washed with ethanol and deionized water until the pH value was constant.

Characterization of Biochar Adsorbents
The surface morphologies of biochar adsorbents were examined using a scanning electron microscope (SEM, Nova NanoSEM 450, FEI, Hillsborough, OR, USA). Their texture properties were evaluated by conducting N 2 adsorption-desorption measurements at 300 • C for 8 h. The SSA was determined by the Brunauer-Emmett-Teller (BET) method, and the porosities were analyzed using an automatic analyzer (Autosorb IQ3, Quantachrome, Boynton Beach, FL, USA). The pore size distribution and pore volume were determined using the Horvath-Kawazo and Barrett-Joyner-Halenda methods, respectively [30]. The crystal structures of the adsorbents were characterized using X-ray diffraction (XRD, Ulti-maIV, Neo Confucianism, Tokyo, Japan) in the range from 10 • to 80 • at a speed of 10 • per minute. The surface functional groups were analyzed using Fourier transform infrared spectroscopy (FTIR, Nicolet iS5, Thermo Electron, Waltham, MA, USA) with 32 scans over a 4000 to 400 cm −1 wavelength range and a resolution of 4 cm −1 . The elemental composition of the adsorbents before and after adsorption was analyzed using X-ray photoelectron spectroscopy (XPS, K-Alpha 1063, Thermo Fisher, Waltham, MA, USA), and the original binding energy was corrected based on the C1s peak at 1486.6 eV. The zeta potential of the suspension was measured using a Zetasizer nano series Nano ZS 90 instrument (Malvern Panaco, Malvern, UK). The magnetic properties were determined using a vibrating sample magnetometer (VSM, 7404, Lakeshore, Columbus, OH, USA) with a magnetic field range of ±2.

Preparation of Cr(VI) Aqueous Suspension
The Cr(VI) solution for the subsequent experiments was prepared by dissolving 2.8290 g of potassium dichromate (K 2 C 2 O 7 ) in 1000 mL of deionized water, resulting in a Cr(VI) stock solution with a concentration of 1000 mg/mL [34,38]. K 2 C 2 O 7 was first heated at 100 • C for 2 h and cooled to room temperature to prepare the stock solution. Subsequently, 2.8290 g of K 2 C 2 O 7 was diluted with deionized water and transferred into a 1000 mL volumetric flask to reach the desired volume. The solution was thoroughly mixed and stored in a cool, dark place for later use. The concentration of Cr(VI) in the solution was determined using the 1,5-diphenylcarbazide colorimetric method, which involved measuring the absorbance of the solution at the optimal wavelength of 540 nm using a UV-visible spectrophotometer. The concentration of Cr(VI) was found to be linearly correlated with the intensity of the absorbance signals (R 2 > 0.999).
where C 0 represents the initial concentration of Cr(VI) (mg/L) before adsorption, and C e denotes the final concentration of Cr(VI) (mg/L) after adsorption. V is the solution volume (L), and m is the mass (g) of the adsorbates.

Adsorption Kinetics
Time-dependent experiments were carried out to investigate the adsorptive rate of Cr(VI) onto biochar adsorbents. A total of 20 mg of biochar adsorbents was added to 10 mL of Cr(VI) solution at 50 mg/L with different contact times ranging from 5 to 2160 min.
The pH of the solution was adjusted to 2 using 1 mol/L HNO 3 and 1 mol/L NaOH. The kinetic data obtained was fitted using the pseudo-first-order model (Equation (3)), pseudo-second-order model (Equation (4)), and intraparticle diffusion model (Equation (5)) to quantitatively evaluate the capture of Cr(VI) on the different biochar adsorbents. The optimal kinetic model was selected according to the largest coefficient of determination (R 2 ).
where Q e and Q t represent the capacity (mg/g) to adsorb Cr(VI) at equilibrium and at time t, respectively. K 1 (1/min), K 2 (g/mg/min), and K p (g/mg/min 1/2 ) are the rate constants for pseudo-first-order, pseudo-second-order, and intraparticle diffusion models, respectively. The constant C is the intercept of the intraparticle diffusion model, which reflects the thickness of the boundary layer.

Equilibrium Isotherm
The Langmuir and Freundlich models are commonly used to describe sorption behavior based on different assumptions [15,39]. The Langmuir model assumes that the adsorbents do not interact with finite active sites within a monolayer of homogenous chemisorption. On the other hand, the Freundlich model considers the heterogeneous energetic distribution of active sites and accounts for the multilayer uptake of adsorbates. In this study, isotherm adsorption was investigated by mixing 20 mg of adsorbents with Cr(VI) solution at varying concentrations (10 mg/L, 25 mg/L, 50 mg/L, 100 mg/L, 200 mg/L, 300 mg/L, 400 mg/L, and 500 mg/L) at temperatures of 30 • C, 40 • C and 50 • C. After the adsorption process, the concentration of Cr(VI) in the solution was determined by shaking the mixture for 24 h. The obtained isotherm data were fitted using the Langmuir (Equations (6) and (7)) and Freundlich (Equation (8)) equations.
where Q e and Q max represent equilibrium and the maximum quantity (mg/g) of Cr(VI) adsorbed onto adsorbents, respectively. C e and C 0 denote the equilibrium and initial concentration (mg/L) of Cr(VI), respectively. The parameter K L (L/mg) is an energy constant, and n reflects the adsorption intensity. R L is a separation factor estimating the feasibility of adsorption, which is favorable when R L ranges from 0 to 1 [15,40]. K F (mg 1−n L n /g) is a coefficient related to the adsorption capacity.

Adsorption Thermodynamics
The spontaneity, extent of the driving force, and energy involved in the adsorption of Cr(VI) onto adsorbents were further investigated by analyzing the changes of Gibbs free energy (∆G o , KJ/mol), entropy (∆S o , J/(K·mol)), and enthalpy (∆H o , KJ/mol) according to Equations (9) and (10) [15,40].
where K a represents the constant related to thermodynamic equilibrium. R is a general gas constant set at 8.314 J/(K·mol), and T is the absolute temperature. A positive value of ∆H o indicates an endothermic reaction during the adsorption process, while a negative value of ∆H o indicates that the adsorption process occurs spontaneously and is feasible.

Stability and Re-Usability Assessment
A total of five consecutive adsorption/desorption cycles were performed to assess the stability and re-usability of adsorbents. In each cycle, the adsorbents were subjected to desorption by washing with 0.2 mol/L of NaOH after the adsorption step. The washed adsorbents were then filtered, washed with deionized water, and dried before being reused in the subsequent cycle. The removal rates of the adsorbents were determined according to Equation (2).

Biochar Characterization
As shown in Figure 1, the biochar adsorbents exhibit distinct surface morphologies. PBC displays smooth surfaces with apparent pores but lacks fine particles and polymers within the pores ( Figure 1A,a). On the other hand, the inner surface of BMC pores appears rough and contains noticeable aggregated particles ( Figure 1B,b). Notably, BMBC exhibits abundant particulate polymers closely adhering to the pore surface ( Figure 1C,c). The rougher inner surface of BMC pores compared to PBC may be attributed to loaded Fe 3 O 4 particles. In contrast, the presence of particulate polymers on BMBC could potentially enhance its capacity for adsorbing Cr(VI).  As presented in Table 1, the BET SSA of MBC (272.565 m 2 /g) is approximately twice that of PBC (123.079 m 2 /g), accompanied by an increase in its total pore volume to 0.173 m 3 /g. By contrast, BMBC exhibits a smaller BET SSA and total pore volume compared to the other two adsorbents, which can be attributed to the partial blockage of its pores by glutaraldehyde and cysteamine during chemical modification [41,42]. The average pore diameters ranging from 2.4-5.1 nm indicate that a mesoporous structure with a small presence of microporous material predominantly characterizes all three biochar adsorbents.

Biochar
Surface Properties Elemental Composition (wt.%) SSA (m 2 /g) TPV (m 3 /g) APD (nm) C 1s O 1s N 1s Fe 2p S 2p As presented in Table 1, the BET SSA of MBC (272.565 m 2 /g) is approximately twice that of PBC (123.079 m 2 /g), accompanied by an increase in its total pore volume to 0.173 m 3 /g. By contrast, BMBC exhibits a smaller BET SSA and total pore volume compared to the other two adsorbents, which can be attributed to the partial blockage of its pores by glutaraldehyde and cysteamine during chemical modification [41,42]. The average pore diameters ranging from 2.4-5.1 nm indicate that a mesoporous structure with a small presence of microporous material predominantly characterizes all three biochar adsorbents. SSA, specific surface area; TPV, total pore volume; APD, average pore diameter; -, no data.
The XRD spectra in Figure  particles on their surfaces. Fe-based materials are known for their excellent magnetic properties, which depend on shape anisotropy and crystallinity [43]. The hysteresis curves in Figure 2b demonstrate a stronger saturation magnetization of BMBC (3.95 emu/g) compared to MBC (2.87 emu/g). Additionally, the inset picture on the bottom right of Figure 2b shows that external magnets attracted more magnetic materials in BMBC systems. These qualitative and quantitative confirmations support the successful magnetization of BMBC. The superior magnetic properties of BMBC compared to MBC can be attributed, at least partially, to the difference in crystallinity and shape anisotropy of The FTIR analysis in Figure 3a revealed similar peaks among the three adsorbents. For instance, a peak at 3432 cm −1 corresponded to stretching vibrations of O-H and N-H, and peaks at 2924 cm −1 and 2856 cm −1 represented symmetrical and asymmetrical stretching vibrations of C-H in methylene [44,45]. Other similar peaks at 1594 cm −1 (C-H) and 1096 cm −1 (C-O) [46] were also observed. Notably, a band near 561 cm −1 , corresponding to Fe-O stretching vibrations, was only present in BMBC and MBC, confirming the successful loading of Fe3O4 [33,47]. The peaks at 1150 cm −1 (C-O) and 1110 cm −1 (C-N-H) were weaker in BMBC, potentially due to partial coverage of the surface by cystamine [48]. In addition, a weaker peak at 706 cm −1 in BMBC may be attributed to stretching vibrations of C-S [49].
As depicted in Figure 3f, peaks at 399.94 eV and 402.57 eV confirm the presence of functional groups of -NH 2 and N-C=O [52]. The S 2p spectrum of BMBC shows peaks at 164.16 eV, 165.35 eV, and 168.55 eV, confirming the presence of S-S, C-S, and S-O [50,53]. The increased Fe 2p signals in BMC and BMBC (Table 1) [54]. This is consistent with prior research data on se peaks and prior research data on Fe 3 O 4 [49,54]. Overall, the FTIR and XPS analyses confirmed the successful grafting of amino, disulfide, and ester functional groups onto the surface of BMBC, with Fe 3 O 4 magnetic particles adhering to MBC and BMBC.

Impacts of Reaction Time and Kinetic Study
As observed in Figure 4a, the removal rate of Cr(VI) by BMBC exhibits a rapid increase in the first 720 min, followed by a ground increase from 97.9% to 100.0% with longer reaction times. In comparison, BMBC demonstrated a higher removal rate (100.0%) compared to MBC (72.37%) and PBC (61.42%). Notably, in the BMBC system, the removal rate sharply rose to 62.87% within the first 5 min and steadily ascended to 100% around 720 min, indicating complete elimination of the adsorbates from the solution. These experiment results strongly supported the efficacy of BMBC, with its amino and sulfur functional groups, in efficiently removing Cr(VI) from wastewater.

Impacts of Reaction Time and Kinetic Study
As observed in Figure 4a, the removal rate of Cr(VI) by BMBC exhibits a rapid increase in the first 720 min, followed by a ground increase from 97.9% to 100.0% with longer reaction times. In comparison, BMBC demonstrated a higher removal rate (100.0%) compared to MBC (72.37%) and PBC (61.42%). Notably, in the BMBC system, the removal rate sharply rose to 62.87% within the first 5 min and steadily ascended to 100% around 720 min, indicating complete elimination of the adsorbates from the solution. These experiment results strongly supported the efficacy of BMBC, with its amino and sulfur functional groups, in efficiently removing Cr(VI) from wastewater. In line with previous literature [53], the pseudo-second-order model provided a better fit for kinetics data than the pseudo-first-order model, as indicated by the higher R 2 values for the tested adsorbents (Table 2). This suggests that the sorption process is predominantly driven by chemical adsorption [55], involving electron sharing and the formation of chemical bonds [56]. Typically, the adsorption of adsorbates from solution to In line with previous literature [53], the pseudo-second-order model provided a better fit for kinetics data than the pseudo-first-order model, as indicated by the higher R 2 values for the tested adsorbents (Table 2). This suggests that the sorption process is predominantly driven by chemical adsorption [55], involving electron sharing and the formation of chemical bonds [56]. Typically, the adsorption of adsorbates from solution to adsorbents involves three stages: (I) boundary layer diffusion, which transports the adsorbates from the solution onto the surface of adsorbents; (II) intra-particle diffusion, which transfers adsorbates from the surface to the interior particles of adsorbents; and (III) adsorption equilibrium, which repairs the adsorption reaction occurring on the active sites distributed within the interior particles of the adsorbents [45,55,57]. It is worth noting that the adsorption equilibrium usually affects little on the overall adsorption rate. Similar to the prior finding [15], the intra-particle diffusion model exhibited multi-curves in the kinetic data (Figure 4d), providing support for the existence of three distinct stages of Cr(VI) adsorption.  Table 3 consistently demonstrates a decline in the adsorption rate constant for Cr(VI) from the first stage (Kp 1 ) to the third stage (Kp 3 ) for all the test adsorbents. The highest Kp 1 values indicate a fast diffusion step related to the external mass transport of Cr(VI) from solution onto the adsorbent's surface, with diffusion rate constants of 0.8592 mg/g·min 1/2 (BMBC), 0.5167 mg/g·min 1/2 (BMC), and 0.5177 mg/g·min 1/2 (PBC), respectively. The slower diffusion at longer reaction times can be attributed to the mass transfer into the interior particles of the adsorbents, with respective Kp 2 values of 0.2511 mg/g·min 1/2 (BMBC), 0.3282 mg/g·min 1/2 (BMC), and 0.2450 mg/g·min 1/2 (PBC). The diffusion rate constants of Kp 2 decrease in the second stage due to increased resistance to the internal diffusion of Cr(VI). Subsequently, in the third stage, the diffusion rate constant of Kp 3 continues to decrease until reaching adsorption equilibrium due to the decreasing concentration of Cr(VI) in the solution. Notably, the non-zero value of C 1 obtained from the simulated results in the first stage suggests that boundary layer diffusion does not solely control this stage of adsorption. The higher values of R 2 than 0.93 indicate that intra-particle diffusion is suitable to describe the experimental data and acts as the rate-limiting step in the second stage. At the last stage, the intra-particle diffusion begins to slow down with a decrease in the concentration of Cr(VI) in the solution, resulting in an extremely low mass transfer rate [45]. In summary, our study found that the adsorption of Cr(VI) is not solely controlled by intra-particle diffusion as the rate-limiting step but may also be affected by boundary layer diffusion [57].

Impacts of Initial Cr(VI) Concentration and Isotherm Models
The initial concentration of adsorbates is important to overcome the solid-liquid boundary's mass transfer resistance during the transfer of Cr(VI) from the solution to the adsorbents [45]. Figure 5a showed an ascending trend in the adsorption capacity as the initial Cr(VI) concentration increased until reaching the respective upper limits of BMBC (65.82 mg/g), MBC (36.63 mg/g), and PBC (31.56 mg/g). The higher adsorption capacity of MBC compared to PBC can be attributed, at least in part, to the additional binding sites provided by the increased SSA. Despite having approximately one-third the SSA of MBC, BMBC demonstrated a greater adsorption capacity for Cr(VI) due to the presence of amino groups and sulfur bonds on its surface, which enable electrostatic interaction and complexation, enhancing its ability to adsorb Cr(VI). According to Figure 5b-d and Table 4, the Langmuir model is more suitable for fitting the isotherm than the Freundlich model, as indicated by the larger values of R 2 . The Langmuir curve of BMBC shows a consistent maximum adsorption capacity of Cr(VI), which agrees well with experimental values (Figure 5b-d). The superior adsorption capacity of BMBC can be attributed to the reaction between Cr(VI) and the introduced functional groups, including iron, amino group, and disulfide bonds. The calculated RL of BMBC declines from 0.0201 to 0.0095 when the initial Cr(VI) concentration of Cr(VI) increases from 10 mg/L to 500 mg/L at 30 °C, 40 °C, and 50 °C (Table 4). This confirms that BMBC is the optimal adsorbent for removing Cr(VI). As described herein, the Langmuir model provides a better fit for experimental isotherm data in our study, and the availability of binding sites and functional groups on the surface of adsorbents limits the one-way chemical adsorption.   Table 4, the Langmuir model is more suitable for fitting the isotherm than the Freundlich model, as indicated by the larger values of R 2 . The Langmuir curve of BMBC shows a consistent maximum adsorption capacity of Cr(VI), which agrees well with experimental values (Figure 5b-d). The superior adsorption capacity of BMBC can be attributed to the reaction between Cr(VI) and the introduced functional groups, including iron, amino group, and disulfide bonds. The calculated R L of BMBC declines from 0.0201 to 0.0095 when the initial Cr(VI) concentration of Cr(VI) increases from 10 mg/L to 500 mg/L at 30 • C, 40 • C, and 50 • C (Table 4). This confirms that BMBC is the optimal adsorbent for removing Cr(VI). As described herein, the Langmuir model provides a better fit for experimental isotherm data in our study, and the availability of binding sites and functional groups on the surface of adsorbents limits the one-way chemical adsorption.

Impacts of Temperature and Thermodynamic Fitting
As observed in Figure 6a, the capacities for Cr(VI) adsorption rise with an increase in temperature. Higher temperatures facilitate the diffusion of Cr(VI) in the solution and mitigate the resistance to adsorption on the adsorbents, resulting in a slight increase in adsorption capacity. The linear relationship between lnK a and 1/T is observed in Figure 6b, and the fitted parameters are summarized in Table 5. The negative values of ∆G o at given temperatures indicate a spontaneous adsorption process, and the more negative ∆G o at higher temperatures suggests a greater degree of spontaneity. The positive values of ∆H o confirm an endothermic adsorption process, while the positive ∆S o reflects the increased randomness between the solid-liquid interface [45,58]. These findings indicate that Cr(VI) adsorption occurs spontaneously and endothermically. Moreover, adsorption capacities are enhanced at higher temperatures. As observed in Figure 6a, the capacities for Cr(VI) adsorption rise with an increase in temperature. Higher temperatures facilitate the diffusion of Cr(VI) in the solution and mitigate the resistance to adsorption on the adsorbents, resulting in a slight increase in adsorption capacity. The linear relationship between lnKa and 1/T is observed in Figure  6b, and the fitted parameters are summarized in Table 5. The negative values of ∆G o at given temperatures indicate a spontaneous adsorption process, and the more negative ∆G o at higher temperatures suggests a greater degree of spontaneity. The positive values of ∆H o confirm an endothermic adsorption process, while the positive ∆S o reflects the increased randomness between the solid-liquid interface [45,58]. These findings indicate that Cr(VI) adsorption occurs spontaneously and endothermically. Moreover, adsorption capacities are enhanced at higher temperatures.  Solution pH is a critical determinant for removing heavy metal irons as it affects both the solutionʹs ionic form and the adsorbents' surface properties [47,59]. As shown in Figure 7a, the maximum percentage of Cr(VI) was removed when the solution pH was 2.

Impact of pH in Solution
Solution pH is a critical determinant for removing heavy metal irons as it affects both the solution's ionic form and the adsorbents' surface properties [47,59]. As shown in Figure 7a, the maximum percentage of Cr(VI) was removed when the solution pH was 2. Among them, BMBC exhibits much higher Cr(VI) removal efficiency (98.81%) compared to MBC (69.13%) and PBC (57.06%). The removal efficiencies of Cr(VI) were negatively related to solution pH, which is attributed, at least in part, to the surface characteristics of adsorbents [44]. Figure 7b showed a rapid decrease in the positive zeta potential as the solution pH rose from 2 to 4, followed by a gentle decline until reaching the minimum value of 32.93 mV in the BMBC system, implying abundant positive charges on its surface. This suggests that the surface of BMBC presents abundant positive charges that can easily bind with Cr(VI) anions through electrostatic interactions. In an acidic environment, Cr(VI) is predominantly present in the forms of HCrO 4 − and Cr 2 O 7 2− , while it exists as CrO 4 2− in neutral and alkaline environments [60]. Alkaline functional groups on the surface of adsorbents can undergo protonation under acidic conditions, leading to the formation of positive charges [30], which are capable of binding solution Cr(VI) anion to the surface of adsorbents through electrostatic interaction. Similarly, the -NH 2 group on the BMBC surface can be protonated, creating additional sites [30]. On the other hand, disulfide bonds (S-S) can be cleaved, forming mercaptan groups that enable the coordination and immobilization of Cr(VI) ions [61]. This mechanism makes BMBC exhibit enhanced Cr(VI) removal at low pH levels. Despite having equal zeta potentials, the presence of Fe 3 O 4 particles on the surface of MBC contributes to a higher removal rate of Cr(VI) compared to PBC [62]. This suggests that the surface of BMBC presents abundant positive charges that can easily bind with Cr(VI) anions through electrostatic interactions. In an acidic environment, Cr(VI) is predominantly present in the forms of HCrO4 − and Cr2O7 2− , while it exists as CrO4 2− in neutral and alkaline environments [60]. Alkaline functional groups on the surface of adsorbents can undergo protonation under acidic conditions, leading to the formation of positive charges [30], which are capable of binding solution Cr(VI) anion to the surface of adsorbents through electrostatic interaction. Similarly, the -NH2 group on the BMBC surface can be protonated, creating additional sites [30]. On the other hand, disulfide bonds (S-S) can be cleaved, forming mercaptan groups that enable the coordination and immobilization of Cr(VI) ions [61]. This mechanism makes BMBC exhibit enhanced Cr(VI) removal at low pH levels. Despite having equal zeta potentials, the presence of Fe3O4 particles on the surface of MBC contributes to a higher removal rate of Cr(VI) compared to PBC [62].

Impacts of Co-Existing Ions
Other co-existing ions in wastewater can compete for surface binding sites with Cr(VI), reducing adsorption capacity [63]. To investigate the impacts of these ions on Cr(VI) removal, additional experiments were carried out using a prepared solution containing a mixed concentration of 50 mg/L for Cr(VI), Cu(II), Mn(II), Cd(II), and K(I), with a separate pure solution of 50 mg/L Cr(VI) as a reference (Cr(VI)-0). As shown in Figure  8a, BMBC exhibits an excellent Cr(VI) removal efficiency of 96.37% but much lower eliminate rates for other co-existing cations such as Cu(Ⅱ) (15.27%), Cd(Ⅱ) (10.50%), Mn(Ⅱ) (9.76%), and K(Ⅰ) (4.60%), suggesting its better specificity for Cr(VI) removal. However, there is a decrease in the percentage of Cr(VI) removal in MBC (8.17%) and PBC (6.79%) compared to that of BMBC (3.63%) in mixed systems, indicating the existence of competitive adsorption between Cr(VI) and other metal cations. At a pH level of 2, the positive charges on the BMBC surface can produce electrostatic repulsion with other cations and selectively bind with the Cr(VI) anion.
The impacts of four common anions (i.e., Cl − , NO3 − , SO4 2−, and HPO4 2− ) on Cr(VI) removal were illustrated in Figure 8b. Although the Cr(VI) removal rate of BMBC decreased by 7.58% in the presence of co-existing anions compared to the reference, it remained at

Impacts of Co-Existing Ions
Other co-existing ions in wastewater can compete for surface binding sites with Cr(VI), reducing adsorption capacity [63]. To investigate the impacts of these ions on Cr(VI) removal, additional experiments were carried out using a prepared solution containing a mixed concentration of 50 mg/L for Cr(VI), Cu(II), Mn(II), Cd(II), and K(I), with a separate pure solution of 50 mg/L Cr(VI) as a reference (Cr(VI)-0). As shown in Figure 8a, BMBC exhibits an excellent Cr(VI) removal efficiency of 96.37% but much lower eliminate rates for other co-existing cations such as Cu(II) (15.27%), Cd(II) (10.50%), Mn(II) (9.76%), and K(I) (4.60%), suggesting its better specificity for Cr(VI) removal. However, there is a decrease in the percentage of Cr(VI) removal in MBC (8.17%) and PBC (6.79%) compared to that of BMBC (3.63%) in mixed systems, indicating the existence of competitive adsorption between Cr(VI) and other metal cations. At a pH level of 2, the positive charges on the BMBC surface can produce electrostatic repulsion with other cations and selectively bind with the Cr(VI) anion.

Mechanism Underlying Sorption
XPS spectra (Figure 9a) after adsorption revealed the emergence of a new band corresponding to Cr 2p. The Cr 2p spectra of BMBC (Figure 9b) exhibited two peaks at 587.28 eV (Cr 2p1/2) and 577.28 eV (Cr 2p3/2), with the Cr 2p3/2 peak comprising Cr(VI) (578.45 eV) and Cr(III) (577.05 eV) [38,60]. These peaks provide evidence that the adsorbed Cr(VI) on the BMBC surface was partially reduced to Cr(III). The Fe 2p spectra of BMBC (Figure 9c) exhibit two peaks at 723.63 eV and 709.63 eV that can be assigned to Fe2O3. However, Fe2O3 was not observed in the XRD analysis. One possible explanation is that the generated Fe2O3 was very low and thus could not be detected [6,64]. The peaks of Fe 2p3/2 were shifted to 710.86 eV (Fe(II)) and 712.90 eV (Fe(III)), and a new peak appeared at 711.30 eV, along with a satellite peak at 718.08 eV, which corresponds to FeCr2O4 and Fe(III), respectively [6,54]. This result suggests that a portion of Fe(II) was oxidized to Fe(III). Accordingly, an oxidation-reduction reaction occurred between Cr(VI) and Fe3O4 on the adsorbent surface, resulting in a partial reduction of Cr(VI) to Cr(III).
Comparison of the C 1s spectra of BMBC before ( Figure 3d) and after (Figure 9d) adsorption showed that the percentage of C-H/C decreased from 48.31% to 44.84% and that of N-C=O decreased from 16.43% to 15.70%. Conversely, the percentage of C-O increased from 35.27% to 39.46%. The phenomenon can be attributed to the participation of active surface groups such as ester, which are capable of reducing Cr(VI) to Cr(III) by providing electrons [65]. The N 1s spectrum of BMBC after Cr(VI) absorption (Figure 9e) exhibited a slight shift in the peaks corresponding to -NH2 and HN-C=O, especially a 22.28% reduction in the -NH2 peak. This peak shift and percentage reduction can be explained by the protonation of surface amino groups into -NH3 + and the subsequent formation of a chromium complex (Cr(III)-N) through electrostatic interaction with Cr(VI) [60]. Additionally, a new peak assigned to nitrate was observed at 406.72 eV, indicating the reaction of BMBC surface groups with Cr(VI) and the formation of a nitrate-like substance to (Cr(III)-N-O) [52]. The S 2p spectrum of BMBC after absorbing Cr(VI) (Figure 9f) provided evidence that the disulfide bond (S-S) can reduce Cr(VI) into Cr(III) via redox reaction, which shifted the S-S peak and lessened its proportion [66].  Figure 8b. Although the Cr(VI) removal rate of BMBC decreased by 7.58% in the presence of co-existing anions compared to the reference, it remained at 92.42% due to its remarkable specificity to Cr(VI). Co-existing anions may hinder the electrostatic interaction between Cr(VI) adsorbents by competing for active sites with Cr(VI).

Mechanism Underlying Sorption
XPS spectra (Figure 9a) after adsorption revealed the emergence of a new band corresponding to Cr 2p. The Cr 2p spectra of BMBC (Figure 9b) exhibited two peaks at 587.28 eV (Cr 2p 1/2 ) and 577.28 eV (Cr 2p 3/2 ), with the Cr 2p 3/2 peak comprising Cr(VI) (578.45 eV) and Cr(III) (577.05 eV) [38,60]. These peaks provide evidence that the adsorbed Cr(VI) on the BMBC surface was partially reduced to Cr(III). The Fe 2p spectra of BMBC ( Figure 9c) exhibit two peaks at 723.63 eV and 709.63 eV that can be assigned to Fe 2 O 3 . However, Fe 2 O 3 was not observed in the XRD analysis. One possible explanation is that the generated Fe 2 O 3 was very low and thus could not be detected [6,64]. The peaks of Fe 2p 3/2 were shifted to 710.86 eV (Fe(II)) and 712.90 eV (Fe(III)), and a new peak appeared at 711.30 eV, along with a satellite peak at 718.08 eV, which corresponds to FeCr 2 O 4 and Fe(III), respectively [6,54]. This result suggests that a portion of Fe(II) was oxidized to Fe(III). Accordingly, an oxidation-reduction reaction occurred between Cr(VI) and Fe 3 O 4 on the adsorbent surface, resulting in a partial reduction of Cr(VI) to Cr(III).
Comparison of the C 1s spectra of BMBC before ( Figure 3d) and after (Figure 9d) adsorption showed that the percentage of C-H/C decreased from 48.31% to 44.84% and that of N-C=O decreased from 16.43% to 15.70%. Conversely, the percentage of C-O increased from 35.27% to 39.46%. The phenomenon can be attributed to the participation of active surface groups such as ester, which are capable of reducing Cr(VI) to Cr(III) by providing electrons [65]. The N 1s spectrum of BMBC after Cr(VI) absorption ( Figure 9e) exhibited a slight shift in the peaks corresponding to -NH 2 and HN-C=O, especially a 22.28% reduction in the -NH 2 peak . This peak shift and percentage reduction can be explained by the protonation of surface amino groups into -NH 3 + and the subsequent formation of a chromium complex (Cr(III)-N) through electrostatic interaction with Cr(VI) [60]. Additionally, a new peak assigned to nitrate was observed at 406.72 eV, indicating the reaction of BMBC surface groups with Cr(VI) and the formation of a nitrate-like substance to (Cr(III)-N-O) [52]. The S 2p spectrum of BMBC after absorbing Cr(VI) (Figure 9f) provided evidence that the disulfide bond (S-S) can reduce Cr(VI) into Cr(III) via redox reaction, which shifted the S-S peak and lessened its proportion [66]. According to the results above, the mechanisms underlying Cr(VI) adsorption in the BMBC system can be attributed to three aspects ( Figure 10). Firstly, solution Cr(VI) can transfer rapidly to the surface of BMBC through boundary-layer and intra-particle diffusion and be adsorbed onto the BMBC surface through electrostatic interactions. Secondly, a portion of adsorbed Cr(VI) can be reduced to Cr(III) and captured by a weaker acid, thus facilitating simultaneous reduction and adsorption processes for both Cr(VI) and Cr(III). Thirdly, the Cr(III) species forms complexes with surface functional groups, such as amino and ester groups, present on the BMBC surface. According to the results above, the mechanisms underlying Cr(VI) adsorption in the BMBC system can be attributed to three aspects ( Figure 10). Firstly, solution Cr(VI) can transfer rapidly to the surface of BMBC through boundary-layer and intra-particle diffusion and be adsorbed onto the BMBC surface through electrostatic interactions. Secondly, a portion of adsorbed Cr(VI) can be reduced to Cr(III) and captured by a weaker acid, thus facilitating simultaneous reduction and adsorption processes for both Cr(VI) and Cr(III). Thirdly, the Cr(III) species forms complexes with surface functional groups, such as amino and ester groups, present on the BMBC surface.

Regeneration and Re-Usability Study
Re-usability is a significant consideration in determining the application potential of an adsorbent for wastewater treatment. As revealed in Figure 11, the Cr(VI) removal rate gradually decreased with an increase in consecutive adsorption/desorption cycles for all tested adsorbents. However, BMBC demonstrated superior performance in Cr(VI) removal compared to BMC and PBC. In the BMBC system, the Cr(VI) removal rate remained at 61.12% after the fifth regeneration cycle, which is comparable to the maximum Cr(VI) removal rate of PBC. Accordingly, this novel BMBC is stable and reusable for treating Cr(VI) in wastewater.

Comparing Adsorption Capacity of Cr(VI)
As summarized in Table 6, the maximum adsorption capacity of BMBC is 66.10 mg/g, which is nearly double that of MBC and higher than the capacities reported for many other chemically modified crop straw adsorbents in previous studies (Table 6). These findings suggest that BMBC is an efficient and promising magnetic adsorbent for Cr(VI). This observation is consistent with the results of other studies [32,66,67], which have also demonstrated that surface modification of biochar using cystamine is an effective method to fabricate super adsorption materials for Cr(VI).

Regeneration and Re-Usability Study
Re-usability is a significant consideration in determining the application potential of an adsorbent for wastewater treatment. As revealed in Figure 11, the Cr(VI) removal rate gradually decreased with an increase in consecutive adsorption/desorption cycles for all tested adsorbents. However, BMBC demonstrated superior performance in Cr(VI) removal compared to BMC and PBC. In the BMBC system, the Cr(VI) removal rate remained at 61.12% after the fifth regeneration cycle, which is comparable to the maximum Cr(VI) removal rate of PBC. Accordingly, this novel BMBC is stable and reusable for treating Cr(VI) in wastewater.

Regeneration and Re-Usability Study
Re-usability is a significant consideration in determining the application potential of an adsorbent for wastewater treatment. As revealed in Figure 11, the Cr(VI) removal rate gradually decreased with an increase in consecutive adsorption/desorption cycles for all tested adsorbents. However, BMBC demonstrated superior performance in Cr(VI) removal compared to BMC and PBC. In the BMBC system, the Cr(VI) removal rate remained at 61.12% after the fifth regeneration cycle, which is comparable to the maximum Cr(VI) removal rate of PBC. Accordingly, this novel BMBC is stable and reusable for treating Cr(VI) in wastewater.

Comparing Adsorption Capacity of Cr(VI)
As summarized in Table 6, the maximum adsorption capacity of BMBC is 66.10 mg/g, which is nearly double that of MBC and higher than the capacities reported for many other chemically modified crop straw adsorbents in previous studies (Table 6). These findings suggest that BMBC is an efficient and promising magnetic adsorbent for Cr(VI). This observation is consistent with the results of other studies [32,66,67], which have also demonstrated that surface modification of biochar using cystamine is an effective method to fabricate super adsorption materials for Cr(VI).

Comparing Adsorption Capacity of Cr(VI)
As summarized in Table 6, the maximum adsorption capacity of BMBC is 66.10 mg/g, which is nearly double that of MBC and higher than the capacities reported for many other chemically modified crop straw adsorbents in previous studies (Table 6). These findings suggest that BMBC is an efficient and promising magnetic adsorbent for Cr(VI). This observation is consistent with the results of other studies [32,66,67], which have also demonstrated that surface modification of biochar using cystamine is an effective method to fabricate super adsorption materials for Cr(VI).

Conclusions
Modification of BMC using cystamine, which contains abundant amino groups and disulfide bonds, improved Cr(VI) removal efficacy. The results confirmed that BMBC exhibited a higher percentage of Cr(VI) removal (100%) compared to MBC (72.37%) and PBC (61.42%). The maximum removal of Cr(VI) was achieved at a solution pH of 2 for all three biochar adsorbents. The optimal reaction time and initial adsorbate concentration for Cr(VI) removal were determined to be 1440 min and 300 mg/L, respectively. The higher fitting values obtained from the Langmuir model indicated that the adsorption followed a monolayer adsorption process, with a maximum adsorption capacity of 66.10 mg/g at 50 • C and pH of 2. The better-fitting values obtained for the pseudo-second-order kinetics suggested a chemical adsorption process. The positive values of ∆H o and ∆S o and the negative value of ∆G o indicated that the Cr(VI) adsorption was spontaneous, endothermic, and characterized by high randomness at the solid/liquid interface. The potential mechanisms underlying the adsorption involved electrostatic interactions between Cr(VI) in the solution and the positive charges on the adsorbent surface. Additionally, the introduced amino, sulfur, and Fe(II) groups facilitate the reduction of adsorbed Cr(VI) to Cr(III). Subsequently, complexation occurs between the reduced Cr(III) species and functional groups present on the adsorbent surface. In the presence of co-existing cations such as Cu(II), Cd(II), Mn(II), and K(I), the Cr(VI) removal efficacy was minimally affected. In contrast, the presence of certain anions, such as Cl − , NO 3 − , SO 4 2− , and HPO 4 2− , resulted in a 7.58% decrease in the removal rate of Cr(VI). After undergoing five consecutive adsorption/desorption cycles, BMBC still exhibited a high Cr(VI) removal rate of 61.12%. Overall, this novel BMBC derived from rice straw shows promising potential as a biosorbent for treating Cr(VI) in wastewater.
Author Contributions: All authors contributed to this work. Specifically, J.L. and M.L. designed the experiments. J.W. conducted the batch experiments, and Q.Z. performed the characterization. X.C. interpreted the data and worked out the manuscript. J.H., X.H., and C.L. advised on the conception and design of the study, as well as on the revision of the manuscripts. All authors have read and agreed to the published version of the manuscript.