Investigating the Absorption of Quaternary Ammonium Salt-Functionalized Silica Towards Vanadium in Hydrochloric Acid Solution

ABSTRACT A quaternary ammonium salt-functionalized silica based on N-benzyl-N-methylethanolamine (BMEA-SD) was prepared for the removal and recovery of pentavalent vanadium [V(Ⅴ)] from hydrochloric acid solutions. The impact of pH, contact time, ionic strength, competitive metal ions, and recyclability of the adsorption factors were investigated. The results showed that BMEA-SD was an effective V(Ⅴ) adsorbent with a maximum adsorption capacity of 51.42 mg g−1 at solution pH 3, contact time of 30 minutes, and temperature of 298 K. Characterization of BMEA-SD materials by TG, FTIR, SEM, and XPS. In addition, the adsorption kinetics followed a pseudo-second-order model, with film diffusion being the determined step. The Langmuir isotherm suggested that monolayer chemical adsorption occurred on the adsorbent surface, and thermodynamics data indicated that the V(Ⅴ) adsorption process was spontaneous and endothermic. Selective separation studies (βV/Cr = 109.87, βV/Mn = 1171.40) highlighted its potential for selective adsorption of V(Ⅴ) over hexavalent chromium [Cr(Ⅵ)] and divalent manganese [Mn(II)]. Desorption was most effective using HCl aqueous solution, and this material maintained good adsorption capacity after five cycles. The experimental results indicated that the adsorption behavior might stem from the interplay of electrostatic forces, redox transformations, and ligand-exchange. This investigation revealed the potential of the adsorbent as a promising, efficient, and renewable for enriching and recovering V(Ⅴ) ions from hydrochloric acid solution.


Introduction
The rapid growth of modern industries has resulted in the accumulation of heavy metals, including vanadium (V), in water bodies, posing significant threats to ecosystems. [1,2]With an average concentration of 150 µg g −1 in the inorganic elements like iron (Fe), [28] aiming to enhance the adsorption performance of silica.
Previous studies [29,30] have demonstrated that neutral extractants probably exploit hydrogen bonding, while quaternary ammonium salt extractants utilize ion exchange to achieve sequestration of V. Furthermore, the ability of V to form complexes with monodentate or multidentate ligands in aqueous solutions has been established. [31,32]ence, we have developed a novel quaternary ammonium-functionalized silica, wherein the quaternary ammonium moiety exhibits a monodentate ligand structure incorporating an ethanolamine fragment.The presence of hydroxyl groups within the ligand structure may lead to heightened electron-deficient properties at the nitrogen center.This architectural modification is expected to facilitate the binding of V species with the quaternary ammonium salt in solution through multiple interplays.This research focuses on the synthesis and characterization of a novel quaternary ammonium salt-functionalized silica adsorbent for V (Ⅴ) ions, derived from N-benzyl-N-methylethanolamine, denoted as BMEA-SD.A comprehensive investigation was conducted to evaluate the influences of solution pH, contact time, ionic strength, competitive metals (manganese(Mn) and chromium(Cr)), and recyclability.Additionally, adsorption kinetics, thermodynamics, and isotherm studies were studied to gain insights into the adsorption process.

Analytical methods and materials
The Detailed reagent source and characterization information are available at supplementary.

Synthesis of BMEA-SD
A mixture of 10 g dry SiO 2 (SD), 12 mL of 3-chloropropyltriethoxysilane, and 50 mL of anhydrous toluene was refluxed for 24 hours.After cooling to room temperature, the mixture was centrifuged, and the silica was washed with a mixture of ethanol and ethyl acetate (1:1).The silica was then dried at 65°C for 8 hours, yielding 3.36 g 3-chloropropyl silica (CPTES-SD). [33] mixture of 3.0 g of CPTES-SD and 4.65 g N-benzyl-N-methylethanolamine and 10.0 mL of acetonitrile was refluxed for 8 hours.After cooling to room temperature, the mixture was centrifuged and washed with a mixture of ethanol and ethyl acetate (1:1).The resulting material was dried at 65°C for 8 hours, yielding 3.29 g quaternary ammonium salt-functionalized silica based on N-benzyl-N-methylethanolamine (BMEA-SD), as shown in Scheme.1.

Preparation of solutions
Stock solutions of NaVO 3 , Na 2 CrO 4 •4 H 2 O, MnCl 2 , and NaNO 3 were prepared with reagents of analytical grade to achieve concentrations of 0.1 mol L −1 for V(Ⅴ), Cr(Ⅵ), Mn(II), and 1 mol L −1 for NaNO 3 .Experimental solutions were prepared by diluting the stock solutions to the desired concentrations.The pH of the adsorption solution was adjusted using NaOH and HCl.HCl, HNO 3 and H 2 SO 4 were used to formulate the desorbent as 1 mol L −1 , respectively.

Points of zero charge experiment
The point of zero charge (pH PZC ) of BMEA-SD was determined using a salt titration method. [10,34]In a 50 ml centrifuge tube, 20 ml of 0.05 M NaNO 3 solution was added.The initial pH of each solution was adjusted to a range of 2-11 using 0.1 M HNO 3 and NaOH solutions, denoted as pH i .Subsequently, 70 mg of BMEA-SD was added to each solution and stirred for 24 hours.The final pH of the solution, denoted as pH f , was measured.Plotting the discrepancies between the pH f and pH i values (ΔpH) against the pH i yielded a curve, wherein the intersection point with the x-axis delineated the pH PZC .

Adsorption experiment
To investigate the adsorption performance of BMEA-SD on V(Ⅴ) and determine the optimal adsorption conditions, various experiments were conducted under different conditions.For each experiment, 70 mg of BMEA-SD was added to a 20 mL solution in a 50 mL centrifuge tube and placed on a magnetic stirrer (1000 rpm) at 298 K, except for exploring the effect of BMEA-SD dosage (10-100 mg) on adsorption.Throughout all experiments, the pH of the solution was maintained at 3, except for experiments investigating the effect of pH, which ranged from 1 to 7. For the kinetic experiments, samples were taken at intervals of 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 15, 20, 25, and 30  minutes, and all other batch experiments were conducted for a duration of 60 minutes.The initial concentration of the solution for all experiments was 2 × 10 −3 mol L −1 , except for adsorption isotherms, which were carried out at a concentration of solution between 4 × 10 −4 and 1 × 10 −2 mol L −1 .The thermodynamic experiments were performed at temperatures of 293, 298, 308, 318, and 328 K.In the separation experiments involving V(Ⅴ), Mn(II), and Cr(Ⅵ), the initial concentrations were 2 × 10 −3 mol L −1 , 1.6 × 10 −3 mol L −1 , and 1.1 × 10 −2 mol L −1 , respectively.After the adsorption process was completed, the solution was filtered through a 0.22 μm polyethersulfone (PES) membrane and the filtrate was analyzed for metal content by ICP-MS.Calculation of efficiency(A), capacity(q e ), and metal separation coefficients(β V/M ) for adsorption (Equation.1-4). [35]ere C o (mg L −1 ) and C e (mg L −1 ) represent the initial and equilibrium V (Ⅴ), Cr(Ⅵ) and Mn(II) concentrations, V (L) is the solution volume, and W (g) is the mass of the adsorbent.D (L g −1 ) is the distribution ratio, and D V and D M represent the distribution ratios of V(Ⅴ) and Cr(Ⅵ), Mn(II), respectively.

Desorption and regeneration experiment
To assess the durability of the BMEA-SD adsorbent material, adsorption experiments were conducted under optimal experimental conditions (pH 3, t = 60 min, T = 298 K) using 20 mL V(Ⅴ) solution at a concentration of 2 × 10 −3 mol L −1 .After each adsorption experiment, solid-liquid separation was achieved through centrifugation, and the adsorbent loaded with V(Ⅴ) was dried at 65°C for subsequent experiments.In the optimal desorption agent experiments, 20 mL of 1 M HCl, H 2 SO 4 , and HNO 3 solutions were individually studied.The experiments to investigate the effect of desorption agent concentration utilized 20 mL solutions with concentrations of 0.2, 0.4, 0.6, 0.8, 1, and 1.2 M HCl.For the 5 consecutive adsorption-desorption cycles, 20 mL of 1 M HCl was employed.Each experiment involved thorough mixing and stirring, followed by centrifugation after a 2-hour desorption period.In the adsorption-desorption cycle experiment, the adsorbent was washed with distilled water and centrifuged to a neutral pH, dried overnight at 65°C, and then reused for adsorption.After adsorption or desorption, the supernatant was filtered through a 0.22 μm polyethersulfone (PES) membrane, and the V (Ⅴ) content in the filtrate was determined using ICP-MS.Calculation of V (Ⅴ) desorption efficiency (Dsp) (Equation.5): where C d (mg L −1 ) is the equilibrium concentration of desorbed V(Ⅴ) (mg L −1 ), C o (mg L −1 ), and C e (mg L −1 ) represent the initial and equilibrium V(Ⅴ) concentrations.

Characterization
Thermogravimetric analysis (TGA) was conducted on SiO 2 (SD) and quaternary ammonium salt-functionalized silica based on N-benzyl-N-methylethanolamine (BMEA-SD) in the temperature range of 35°C to 900°C, with a heating rate of 10°C min −1 for 90 minutes.The results are presented in Figure 1.SD exhibited a weight loss of 5.21% below 100°C, attributed to physically adsorbed water in the surface pores.Between 100°C and 900°C, a weight loss of 5.25% indicated the condensation of silanol groups into silicon-oxygen-silicon (Si-O-Si) bonds. [36]BMEA-SD experienced a weight loss of 2.52% below 150°C due to physical adsorption of water.And followed by a 3.82% loss between 150°C and 230°C associated with partial decomposition of surface groups.As the temperature increased to 450°C, an additional weight loss of 8.23% indicated the continuous decomposition of organic matter. [36,37]igure 2 displays the FTIR spectra of SD, 3-chloropropyl silica (CPTES-SD), and BMEA-SD.Three absorption peaks at 1100 cm −1 , 810 cm −1 , and 470 cm −1 were observed in all SiO 2 samples, corresponding to the characteristic vibrations of silica (Si-O-Si) bonds.These peaks represent symmetric and asymmetric stretching vibrations of Si-O-Si bonds, along with symmetric stretching vibrations of Si-O bonds.[40] In the case of SD, the peak at 947 cm −1 is attributed to the bending vibration of Si-OH groups. [41]he presence of abundant surface silanol groups (Si-OH) on SiO 2 is crucial for subsequent functional group modification.After functionalization, absorption peaks at 3400 cm −1 and 947 cm −1 weakened or disappeared, probably due to a reduction in surface Si-OH groups caused by the modification of quaternary ammonium salt. [39]canning electron microscopy (SEM) was utilized to analyze the morphological changes of SiO 2 before and after modification, as well as after V loading.As shown in Figure 3, the surface of SD is smooth, with an irregular  shape and dispersed size distribution.After CPTES modification, the surface of SiO 2 became rougher and fragmented into smaller particles, although it maintained an irregular shape.Further functionalization with BMEA resulted in even smaller particle sizes, attributed to the presence of quaternary ammonium groups on the surface of BMEA-SD and the reduction of surface silanol groups (Si-OH). [40]Qualitative EDS analysis showed that the appearance of new elements (C, N) after modification (Table S1).The presence of V elements on the surface after adsorption confirmed the successful preparation of BMEA-SD and the adsorption of V(Ⅴ) on BMEA-SD.
The XPS spectrum of BMEA-SD (Figure 4a) exhibits peaks corresponding to C1s, N1s, O1s, and Si2p. Figure 4b shown the V2p spectrum, the emergence of a new peak at 531.61 eV may be related to the V-O bond at ~530 eV. [42][45] In Figure 4d, it can be observed that after V(Ⅴ) adsorption, there are slight changes in the positions of the C=C and C-C peaks, but the C-O peak is more pronounced.In the O1s spectrum (Figure 4e), the peak at 532.71 eV is attributed to the C-O bond of N-benzyl-N-methylethanolamine.The N1s spectrum (Figure 4g) shows that the two peaks at 399.17 eV and 402.33 eV are attributed to the nitrogen in BMEA associated with SD by non-bonded interaction and the nitrogen atoms of quaternary ammonium salt in modified BMEA-SD, respectively. [42,46]The XPS analysis further confirms the synthesis of BMEA-SD.

Effect of functionalization and BMEA-SD dosage
Figure 5a shows the results of the adsorption of V(Ⅴ) using different types of silica (3.5 g L −1 of SD, CPTES-SD, and BMEA-SD).It is evident that adsorption of V(Ⅴ) by BMEA-SD is 98.23% with a q e of 26.47 mg g −1 .The adsorption efficiency of BMEA-SD is significantly higher than that of SD and CPTES-SD, which is the best adsorbent.This can be attributed to the stronger affinity of the quaternary ammonium functional groups on the surface of BMEA-SD for V(Ⅴ), which indicates successful functionalization of SiO 2 . [47]Figure 5b explores the optimal dosage of BMEA-SD (0.5 to 5 g L −1 ).It is observed that a dosage of 3.5 g L −1 of the adsorbent achieves a remarkable 98.23% removal of V(Ⅴ) from the solution.However, the q e exhibits an opposite trend.When the adsorbent dosage increases from 0.5 g L −1 to 3.5 g L −1 , q e decreases from 52.59 mg g −1 to 26.49 mg g −1 .Similarly, Song et al. [20] and Erdem et al. [26] studied the adsorption of V(Ⅴ) using diatomite-chitosan composite (CS-DE-10%) and aminopropyl triethoxysilane-functionalized silica (3-APTES modified silica), respectively.They both found that the efficiency of adsorption increased with the increase of the adsorbent, while the adsorption capacity decreased.This is attributed to the insufficient occupation of the quaternary ammonium functional groups on the surface of BMEA-SD and q e decreased as the adsorbent's dosage increases. [45]The increased number of active sites ensures the thorough capture of V(Ⅴ) in the solution and the adsorption efficiency increases. [20]Therefore, 3.5 g L −1 was taken as the optimum adsorbent dose in further studies, considering factors such as removal efficiency and economic cost.

Effect of solution pH
As a key factor, the pH of the solution can affect the surface charge of the adsorbent and the species of metal ions. [48]Therefore, the study investigated the adsorption of V(Ⅴ) by BMEA-SD within the pH range of 1 to 7, as well as its pH PZC .As shown in Figure 6a, with an increase in solution pH from 1 to 3, the q e and adsorption rate sharply increased from 5.44 mg g −1 and 20.38% to the maximum of 26.49 mg g −1 and 98.36%, respectively.When the solution pH  further increased to 4 with q e (13.83 mg g −1 ) and adsorption rate (50.68%), the optimum pH occurs at pH 2-4.Further investigation within pH 2.3-3.9 showed that pH 2.3-3.0 exhibited an increasing adsorption capacity phase, while a decrease occurred within pH 3-3.9.This implies the existence of a narrow pH range from pH 2.7-3.0 centered on pH 3, where effective V(Ⅴ) adsorption occurs.This may be ascribed to the surface charge of BMEA-SD and the species of V(Ⅴ) in different pH aqueous solutions.It has been reported that at pH < 3, V primarily exists as the VO 2 + species in solution, along with a portion of H 3 V 10 O 28 3-.Within the pH range of 3 to 5, V(Ⅴ) cations transform into decamers (V 10 species) anions, where H 3 V 10 O 28 3-is the predominant species.At pH 5-7, V(Ⅴ) predominantly exists in the form of H 2 VO 4 −. [49] The pH PZC of BMEA-SD was found to be between 7 and 8 (Figure 6b), where the surface of BMEA-SD carries a positive charge at pH < pH PZC and a negative charge at pH > pH PZC . [10]Therefore, at pH < 2, the electrostatic repulsion between the positive charge on the surface of BMEA-SD and VO 2 + might lead to the lower adsorption of V(Ⅴ). [50]As the pH increases beyond 2, the growth of H 3 V 10 O 28 3concentration leads to increased adsorption capacity due to electrostatic attraction between the V(Ⅴ) cation and positive charge on the surface of BMEA-SD. [10]However, the decrease in adsorption capacity at pH > 3 may be caused by (1) The decrease in the affinity of the formed H 2 VO 4 − with the adsorbent [51] ; (2) With the increase of pH, and the positive charge on the adsorbent surface amount decreases [10,20] ; (3) Competition for adsorption of increased OH − in solution and V anionic groups on the active sites of the composite surfaces with increasing pH. [52,53]The observed enhancement in adsorption capacity, concurrent with the augmentation of positive surface charge density on the adsorbent.It is possible that a significant involvement of electrostatic interactions in the adsorption process of V(Ⅴ).

Adsorption kinetics
As shown in Figure 7, in the initial stage (t <4 min) of V(Ⅴ) adsorption onto BMEA-SD, the rate increased rapidly to 6.52 mg g −1 min −1 , which is a result of the abundant unoccupied active sites on the adsorbent surface. [50,54]ubsequently, the adsorption rate slowed down to 0.29 mg g −1 min −1 between 4 and 8 min.And after t >8 min, the average equilibrium q e remained stable at 27.35 mg g −1 due to the gradual saturation of active sites over time. [55]o analyze and evaluate the adsorption kinetics of V(Ⅴ) on BMEA-SD, the experimental data were tested using non-linear regression of the pseudo-firstorder (PFO) model, pseudo-second-order (PSO) model, and Elovich model (Equation.6-8). [56,57]here q e (mg g −1 ) and q t (mg g −1 ) represent the adsorption capacity at equilibrium and different times t (min), respectively.K 1 (min −1 ) and K 2 (g mg −1 min −1 ) are the PFO and PSO model rate constants, respectively.α (mg g −1 min −1 ) is the initial adsorption rate constant, and β (g mg −1 ) is a constant related to the coverage of the adsorption surface and the activation energy of chemical adsorption.
The PSO model resulted in a calculated q e (27.81 mg g −1 ) closer to the experimental q e,exp (27.41 mg g −1 ), and a higher R 2 value (0.955) compared to the PFO and Elovich models (R 2 = 0.694 and 0.908, respectively) (Table 1).The results show that the PSO model can better describe the adsorption kinetics of V(Ⅴ) onto BMEA-SD and chemical adsorption may be the ratelimiting step in the adsorption process. [50]he application of the intraparticle diffusion model (Equation.9) and Boyd's model (Equation.10 and 11) [58] further revealed the mechanism and limiting steps of the adsorption process on BMEA-SD. [56]The Boyd's model assumes that if the plot of Bt against t is a straight line passing through the origin, the adsorption process is controlled by intraparticle diffusion.Otherwise, it is controlled by film diffusion (chemical adsorption).Where K id (mg g −1 min 1/2 ) is the intraparticle diffusion rate constant, and "C" is a constant estimating the thickness of the boundary layer.Bt is the Boyd's number, and F(t) = q/q e is a function of time t.
The two linear segments with different slopes in the intraparticle diffusion model (Figure 8a) (6.390 and 0.113, respectively) disclosed that intraparticle diffusion is one of the rate-limiting steps.The segment with a larger slope possibly corresponds to ion exchange (chemical adsorption) between BMEA-SD and the adsorbate, which predominantly controls the adsorption process.- [50]The plot of Bt against t (Figure 8b) exhibited a nonlinear trend that does not pass through the origin, which indicates that film diffusion was ratelimiting in the adsorption of V(Ⅴ) onto BMEA-SD.

Adsorption isotherms
To elucidate the equilibrium performance of V(Ⅴ) adsorption on BMEA-SD and the distribution of the adsorbate in the system, [59] the nonlinear Langmuir and Freundlich isotherm models [60] (Equations.12 and 13, respectively) were used to study the V(Ⅴ) adsorption isotherms on BMEA-SD.The Langmuir model assumes monolayer chemical adsorption occurring on the uniform surface of the adsorbent, while the Freundlich model describes non-equilibrium adsorption on all saturated adsorption sites. [61]ble 1.Kinetic model parameters for the adsorption of V(Ⅴ) on BMEA-SD (20 mL solution of V (Ⅴ), C o = 2 × 10 −3 mol L −1 , pH = 3, BMEA-SD = 3.5 g L −1 , T = 298 K).
As shown in Figure 9 and Table 2, the Langmuir isotherm model yielded better fitting parameters (R 2 = 0.991) compared to the Freundlich isotherm model (R 2 = 0.978).This suggests that it is suitable for describing the adsorption process, as well as the monolayer chemisorption of V(Ⅴ) on BMEA-SD.This is consistent with the results observed by functionalizing silica with 2methyl-8-quinolinol [25] and aminopropyl triethoxysilane. [26]In addition, Table 3 reveals that the adsorbent BMEA-SD demonstrates an intermediate theoretical maximum adsorption capacity (q m ) compared to reported adsorbents.
The efficiency of V(Ⅴ) adsorption on BMEA-SD can be represented by the dimensionless separation factor (R L) associated with the Langmuir model [63] (Equation.14).R L varies between 0 and 1, reflecting that the adsorption process is irreversible, favorable, linear, or unfavorable with RL = 0, 0 < R L <1, R L = 1, and R L >1, respectively.Where C o (mg g −1 ) is the initial concentration of V(Ⅴ), and K L (L mg −1 ) is the Langmuir constant related to the adsorption energy.
The effect of C o on adsorption and the R L shown in Figure S1.The results showed that the equilibrium q e increased with an increase in the initial concentration of V(Ⅴ), possibly due to a higher proportion of strongly affinitive H 3 V 10 O 28 3-. [10]This suggests that higher initial concentrations of V (Ⅴ) are more favorable for adsorption.Furthermore, R L (0.41-0.037) decreased with an increase in C o , which implies that the adsorption process is favorable.

Adsorption thermodynamics
The effect of V(Ⅴ) adsorption on BMEA-SD was investigated in the temperature range of 293-328 K, and the results are shown in Figure 10a.The temperature increases favored adsorption, with the maximum q e (26.93 mg g −1 ) reached at 328 K.It may be that the reduction in the boundary layer thickness on the BMEA-SD surface caused by temperature increases resulted in decreased mass transfer resistance of the adsorbate on the surface or within the pores of the adsorbent. [51]This is consistent with the results observed in Song et al. [20] using Diatomite-chitosan composite (CS-DE-10%) for V removal, but different from Ghanim et al. [10] preparation of KOH modified seaweed hydrochar for adsorption of V(Ⅴ).

Langmuir
Freundlich q m (mg g −1 ) K Trypsin 72.4 [27]   iron (II) acetylacetonate 58.75 [28]   glycidyl trimethyl ammonium chloride modified pine bark 35.0 [42]   Zr(IV)-loaded orange juice residue 51.09 [22]   red mud modified saw dust biochar 16.45 [62]   Sawdust-Derived Cellulose Nanocrystalline 47.2 R (8.314 J mol −1 K −1 ) is the gas constant, and T (K) is the temperature of the adsorption process.The K C and ΔG° can be calculated from Equation.15 and 17, respectively, while ΔH° and ΔS° were calculated from the slope and intercept of the lnK C versus 1/T curve (Equation.16) Based on Figure 10b and the values provided in Table 4, the thermodynamic parameters of V(Ⅴ) adsorption on BMEA-SD were obtained.ΔH° (47.00 kJ mol −1 ) > 0 suggests an endothermic adsorption process, and q e increased with rising temperature. [51]ΔS° (187.89J mol −1 K −1 ) > 0 indicates an increase in randomness at the solid-liquid interface system and a strong affinity of BMEA-SD for V(Ⅴ). [65]ΔG° < 0 and decreases with increasing temperature, indicating the feasibility and spontaneity of V(Ⅴ) adsorption on the adsorbent, which is more favorable at higher temperatures. [10]A similar phenomenon was observed in the adsorption of V by 8-hydroxyquinoline immobilized cellulose. [66]

Effect of ionic strength
Vanadium-contaminated water often contains other co-existing ions that may affect the adsorption of V on adsorbents. [15,52]Therefore, evaluating the effect of ionic strength on adsorption may facilitate a better understanding of adsorbent performance.When NaNO 3 <0.03M (q e = 27.17mg g −1 ), the adsorption of V(Ⅴ) on BMEA-SD was minimally affected, and a significant reduction was observed from 0.05 to 0.15 M NaNO 3 (Figure 11).Adsorption decreased with increasing ionic strength of the aqueous solution may be attributed to: (i) changes in activity of V(Ⅴ) or double-layer properties [67] ; (ii) competition between high concentrations of NO 3 − and V(Ⅴ) for adsorption sites on the adsorbent surface. [68]Similar results were observed on goethite and birnessite for V(Ⅴ) adsorption. [68]

Effect of Cr(Ⅵ) and Mn(II)
Manganese (Mn) and chromium (Cr) are often coexisting with V in minerals, and the similar physicochemical properties of V(Ⅴ) and Cr(Ⅵ) make their separation difficult. [69]The Mn(II) has reducing properties, which may affect Table 4. Thermodynamic parameters for the adsorption of V(Ⅴ) on BMEA-SD (20 mL solution of V (Ⅴ), C o = 2 × 10 −3 mol L −1 , pH = 3, BMEA-SD = 3.5 g L −1 , t = 60 min).the speciation of V(Ⅴ) in solution.Therefore, the adsorption selectivity of BMEA-SD in the solution of Cr(Ⅵ), Mn(II), and V(Ⅴ) under acidic conditions was assessed, and the results are shown in Figure 12.At pH 6, the highest separation factor for the adsorption of V(Ⅴ) and Cr (Ⅵ) on BMEA-SD was observed (β V/Cr = 109.87).This is possible attributed to vanadate polymers usually possessing more negative charge numbers and  larger thermodynamic radii than chromate polymers, which results in higher polarizability and nucleophilicity of vanadate polymers. [22]Additionally, at pH < 2, the adsorption efficiency of V(Ⅴ) and Cr(Ⅵ) differed significantly from other pH values, probably due to the influence of pH on the metal species in solution.At low V(Ⅴ) concentrations, V(Ⅴ) mainly exists as V 10 species at pH 2.5-6, while it transforms to VO 2 + below pH 2.5. [12]Furthermore, Cr(Ⅵ) exists mainly as HCrO 4 − at pH 1-6, when its concentration is less than 2 mM. [22,70]However, due to the higher affinity of V 10 species to BMEA-SD, preferential adsorption of V 10 species likely occurs. [71]Therefore, at pH 2-5, V 10 species occupies more adsorption sites, limiting the adsorption of Cr(Ⅵ) on the adsorbent.At pH 2, most of V(Ⅴ) exists as VO 2 + has weak affinity to the adsorbent, so that HCrO 4 − can access more adsorption sites.For Mn(II), low adsorption efficiency was observed at pH 1-6, with the highest separation factor at pH 4 (β V/Mn = 1171.40).The positive surface charge of BMEA-SD may be unfavorable for chemisorption with Mn 2+ at pH < 6, as Mn 2+ is the dominant species under these conditions. [72,73]

Desorption and regeneration
To develop an economically efficient and reliable adsorbent for removing metal pollutants from wastewater, the desorption efficiency and regeneration capacity should be examined during the adsorption-desorption cycle. [47]esorption of metal ions is influenced by pH and desorption agents. [61]ince vanadium-bearing slags or waste streams are commonly leached in HCl or H 2 SO 4 solutions. [74,75]Besides, vanadium-bearing waste streams generated during the production of metal products also contain significant amounts of HCl or H 2 SO 4 . [76,77]Therefore, 20 mL of 1 M HNO 3 , HCl, and H 2 SO 4 were used as desorption agents, in view of the possible application prospects of BMEA-SD adsorbent.As shown in Figure 13a, the results indicate that HCl (97.15%) is effective in eluting V(Ⅴ) compared to H 2 SO 4 and HNO 3 .Subsequently, by evaluating the influence of HCl concentration on desorption efficiency.It was found that V(Ⅴ) loaded on the adsorbent was essentially eluted completely when the concentration was ≥1 M. Therefore, the reusability of BMEA-SD was assessed through five cycles of regeneration experiments using 1 M HCl (Figure 13b).During the initial two adsorption cycles, the adsorption capacity of BMEA-SD (>97%) remained largely unaffected.Even after five consecutive adsorption cycles, the q e and adsorption rate was maintained at 18.39 mg g −1 and 66.03%, respectively.

Adsorption mechanisms
In XPS analysis, Figure 4a shows that BMEA-SD, after adsorbing V(Ⅴ), displays peaks corresponding to C1s, N1s, O1s, and Si2p, but no distinct V2p peak is observed.The V2p spectrum (Figure 4b) reveals a weak V2p 2/3 peak at a binding energy of 516.60 eV, which indicates V(IV) on the surface of BMEA-SD and V(Ⅴ) is reduced to V(IV) in adsorption process. [20]This aligns with observations in the study of V adsorption utilizing modified chitosan beads. [52]The O1s spectrum (Figure 4f) undergoes significant changes in peak positions after adsorption, and the O-H peak area increases by 26.29%, which may be due to the redox reaction with V(Ⅴ). [20,45]The N1s peak undergoes a noticeable change in position after adsorption (Figure 4h), and it is assumed that quaternary nitrogen can effectively remove V(Ⅴ) anionic pollutants through electrostatic attraction. [42]The experimental findings presented herein suggest that the adsorption mechanism of V species employing quaternary ammonium salt-functionalized silica is considerably more intricate than initially envisioned.The observed adsorption behavior possibly arises from the synergistic interplay of electrostatic interactions, redox reactions and ligand-exchange processes.

Conclusion
A novel quaternary ammonium salt-functionalized silica (BMEA-SD) was successfully synthesized and effectively employed for the removal of V(Ⅴ) ions from hydrochloric acid solutions.The synthesis of BMEA-SD and the effective adsorption of V(Ⅴ) were confirmed through EDS and XPS analyses.BMEA-SD exhibited a 20.80 mg g −1 higher adsorption capacity compared to unmodified SiO 2 , and factors such as solution pH and contact time significantly influenced the adsorption efficiency.Adsorption kinetics indicated that membrane diffusion was the main controlling step, following a pseudo-second-order kinetic model.The adsorption isotherms followed the Langmuir model and suggested a monolayer chemical adsorption occurring on the adsorbent's surface.In the research on selective adsorption, the maximum separation coefficients for V(Ⅴ)/Cr(VI) and V(Ⅴ)/Mn(II) were found to be β V/Cr = 109.87and β V/Mn = 1171.40.Adsorption-desorption experiments demonstrated that after five cycles, this material still retained certain adsorption performance.The point of zero charge, EDS, and XPS analyses showed that electrostatic interactions, redox reactions, and ligand-exchange collectively contribute to the adsorption mechanism.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Table 3 .
Comparison of the adsorption capacity with other V(Ⅴ) adsorbents.