Adsorption of As(V) by the Novel and Efficient Adsorbent Cerium-Manganese Modified Biochar

Arsenic has become a global concern in water environment, and it is essential to develop efficient remediation methods. In this study, a novel adsorbent by loading cerium and manganese oxide onto wheat straw-modified biochar (MBC) was manufactured successfully aiming to remove arsenic from polluted water. Through scanning electron microscopy and energy-dispersive spectroscopy (SEM-EDS), X-ray diffractometer (XRD), X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectrometer (FT-IR), and other techniques, it was found the loading of cerium and manganese oxide on MBC played a significant role in As(V) adsorption. The results of the batch test showed that the adsorption of MBC followed the pseudo-second order kinetics and Langmuir equation. The adsorption capacity of MBC was 108.88 mg As(V)/g at pH = 5.0 (C0 = 100 mg/L, dosage = 0.5 g/L, T = 298 K) with considerable improvement compared to the original biochar. Moreover, MBC exhibited excellent performance over a wide pH range (2.0~11.0). Thermodynamics of the sorption reaction showed that the entropy (∆S), changes of enthalpy (∆H) and Gibbs free energy (∆G), respectively, were 85.88 J/(moL·K), 22.54 kJ/mol and −1.33 to −5.20 kJ/mol at T = 278~323 K. During the adsorption, the formation of multiple complexes under the influence of its abundant surface M-OH (M represents the Ce/Mn) groups involving multiple mechanisms that included electrostatic interaction forces, surface adsorption, redox reaction, and surface complexation. This study indicated that MBC is a promising adsorbent to remove As(V) from polluted water and has great potential in remediating of arsenic contaminated environment.


Introduction
Arsenic (As) is a carcinogenic trace metal that has long been identified as a major human health hazard and is a global concern [1,2]. Both natural and anthropogenic activities are sources of As pollution and include mining, fossil fuel combustion and pesticides. Both the World Health Organization (WHO) and United States Environmental Protection Agency (EPA) have recommended the standard guideline for drinking water at 10 µg/L.
Arsenic contamination is a global environmental problem and polluted groundwater is posing an increasing degree of health risk [3]. The long-term exposure to As in drinking water or foods is closely linked to several cancer types as well as significant mortality [4]. In the 1990s, large-scale groundwater poisoning by As in Bangladesh was considered to be the largest poisoning event in human history.

Biochar Modification
The BC was produced from wheat straws through anaerobic pyrolysis at 873 K for 2 h in a muffle furnace using a N 2 flow-in rate of 300 cm 3 /min. The obtained BC was ground to pass through a 0.15 mm sieve and 10 g was soaked in 50 mL of 1 M HCl solution for 12 h and then washed with deionized water until the pH was neutral and oven-dried at 343-353 K for a further 12 h. The pre-treated biochar (10 g) was successively immersed in 50 mL 0.5 M CeCl 3 and 50 mL 0.2 M KMnO 4 solutions with ultrasonic mixing for 2 h and then evaporated to dryness in a water bath at 368 K. The BC-CeCl 3 -KMnO 4 composite was heated at 873 K for 2 h under N 2 (anaerobic conditions) and then rinsed thoroughly with deionized water to remove impurities and dried at 343-353 K to obtain the Ce-Mn modified biochar (MBC).

Characterizations
Scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) measurements were used to observe the structural morphology of MBC using a JSM-7401 instrument (JEOL, Tokyo, Japan). Material composition was analyzed using D8 ADVANCE X-ray diffractometer (XRD) (Bruker, Karlsruhe, Germany). X-ray photoelectron spectroscopy (XPS) data was acquired using a PHI 5300 ESCA instrument (Perkin Elmer, Burlington, MA, USA) with Mg K α radiation (1253 eV) emitted from a double anode at 50 W and results were fitted using XPS PEAK 4.1. The Brunauer-Emmett-Teller (BET) surface areas and Barrette-Joynere-Halenda pore size distributions analysis were conducted using N 2 adsorption-desorption isotherms generated from an Autosorb-iQ2 Surface Area and Pore Size Analyzer (Quantachrome, Boynton Beach, FL, USA). The FTIR spectra of BC and MBC were generated from Fourier transform infrared spectra (FT-IR) collected using a Bruker TENSOR 27 MCT detector (Bruker, Ettingen, Germany) at a 4 cm −1 resolution. The pH of solution was analyzed by pH-meter with the model of FE28-Standard (Mettler Toledo, Zurich, Switzerland). The pHpzc (point of zero charge) of materials were measured by dynamic light scattering (DLS Zetasizer Nano ZS90, Malvern Instruments, Worcestershire, UK).

Adsorption Experiments
Batch experiments were carried out to investigate adsorption performance of the adsorbents. The effects of different initial As(V) concentrations (5 to 200 mg/L), system temperatures (278, 288, 298, 308 and 323 K), MBC dosages (0.05 to 2.5 g/L) and ionic strength (0.005, 0.01, 0.1, 0.5 and 1 M NaNO 3 ) were investigated. The effects of different solution pH (2.0-11.0) at As(V) concentrations of 50 and 100 mg/L also have been studied. Besides, adsorption isotherm experiment (As(V) concentrations is from 5 to 200 mg/L) were performed at different pH values (pH = 3.0, 5.0, 7.0, 9.0 and 11.0) to determine the maximum adsorption capacity of MBC. Other experimental conditions were set respectively using the following standard levels: As(V) concentration (50 mg/L) in 0.01 M NaNO 3 , at 298 K, neutral pH, and MBC at 0.5 g/L.
MBC or BC batches were conducted with a shaking speed of 200 r.p.m. for 12 h, and then placed on centrifuge to separate the suspensions at 8000 r.p.m. for 5 min. The suspensions were filtered through a syringe filter (diameter 0.22 µm), the initial and final As concentrations in the filtrates were assumed by the hydride-generation atomic fluorescence spectroscopy using an AFS-9120 instrument (Titan Instruments, Beijing, China). The equilibrium adsorption capacity and removal efficiency were obtained according to the following formulas: Removal efficiency (%) = 100((C 0 − C e )/C 0 ) where C 0 is the initially loaded As(V) concentration (mg/L), and C e is As(V) concentration (mg/L) in solution at equilibrium time. V(L) is the solution volume, and m(g) is the adsorbent mass.

SEM-EDS Analysis
The morphology and structural properties of BC and MBC were characterized by SEM-EDS, and the BC modified with Ce-Mn produced particles with uneven surfaces compared with unmodified BC (Supplementary Figure S1). The clear and characteristic Ce and Mn peaks were observed in the MBC material, indicating that these elements were successfully impregnated into the BC and the Ce content of MBC was immensely enhanced (Figure 1b). increased slightly, and the adsorption hysteresis both appeared, that is hysteresis loop. When P/P0 > 0.8, the relative adsorption pressure is high, the adsorption amount of nitrogen by biochar increased sharply, which indicated that the mesopores were the main pore structure of biochar [37]. This is consistent with the corresponding pore size distribution. According to the classification made by the International Union of Pure and Applied Chemistry (IUPAC), the nitrogen adsorption isotherm in the figure belongs to type Ⅲ, H4 hysteresis loop.  Besides, the specific surface area increased from 5.525 to 6.881 m 2 /g, average pore width decreased from initial 36.000 to final 10.681 nm, and the pore volume increased from 0.005 to 0.024 cm 3 /g with cerium and manganese loading (Table 1), these changes are contributed to the adsorption of heavy metal. From N 2 adsorption/desorption isotherms in Supplementary Figure S2, it can be seen when the relative adsorption pressure is small, such as P/P 0 < 0.2, the adsorption amount of nitrogen by biochar is less. As P/P 0 continued to increase, the adsorption amount of nitrogen by biochar only increased slightly, and the adsorption hysteresis both appeared, that is hysteresis loop. When P/P 0 > 0.8, the relative adsorption pressure is high, the adsorption amount of nitrogen by biochar increased sharply, which indicated that the mesopores were the main pore structure of biochar [37]. This is consistent with the corresponding pore size distribution. According to the classification made by the International Union of Pure and Applied Chemistry (IUPAC), the nitrogen adsorption isotherm in the figure belongs to type III, H4 hysteresis loop.

XRD
The x-ray diffraction patterns for BC and MBC were then compared with samples of MnO 2 and CeO 2 in ( Figure S3). The peaks for CeO 2 were well defined and consistent with good crystallinity and purity. In contrast, MnO 2 possessed poor crystallinity and no obvious diffraction peaks. This compound is an amorphous or weakly crystalline mineral. BC and MBC displayed the characteristic peak at 26.7 • , while MnO 2 and CeO 2 all appeared to show same peaks as that of MBC at 36.7 • , 28.6 • , and 47.6 • , respectively, which indicated MBC achieved peak stacking and possessed more peaks than BC deducing MBC was effectively loaded with MnO 2 and CeO 2 . The 4 matrices also displayed different sorption capacities for As(V).

FT-IR
Adsorption of As(V) to MBC was monitored using FT-IR ( Figure 2). It identified a surface rich in hydroxyl groups (M-OH, M represents the Ce/Mn) for MBC with strong adsorption at 1124-1062 cm −1 due to bending vibrations of metal associated surface hydroxyls [38]. This type of metal bonded surface is important in anion adsorption [38,39]. The FT-IR spectra also showed broad adsorption peaks at 3300-3600 cm −1 and 1622 cm −1 for MBC that represent stretching and vibration regions of H-O and H-O-H, respectively [38,40]. In contrast, the adsorption strength of BC was relatively weak.
peaks at 3300-3600 cm and 1622 cm for MBC that represent stretching and vibration regions of H-O and H-O-H, respectively [38,40]. In contrast, the adsorption strength of BC was relatively weak.
Following As(V) adsorption, the intensity of peaks situated at 1622 cm −1 and 3442 cm −1 were dramatically decreased and the M-OH bending band at 1106 cm −1 was sharply weakened for MBC. This indicated an association of As(V) on the MBC surface that might be assisted by hydroxyl groups [20,38]. In addition, the MBC spectrum after adsorption of As(V) induced the appearance of new peak at 865 cm −1 that was not present before adsorption. This peak most likely represents the formation of an As-O bending band that was in agreement with other studies [41][42][43]. The formation of As-O bond showed that the As(V) adsorption onto MBC might follow the inner-sphere complex mechanism.

XPS
The chemical and electronic states of the As(V) adsorption to MBC was assessed using XPS. A certain electron transfer occurred during the modification of biochar, and meanwhile the characteristic Ce3d and Mn2p3 peaks were observed as expected for MBC ( Figure 3a). After As(V) adsorption, As3d peak occurred at 45.4 eV as illustrated in Figure 3b-d [30], confirming its presence.
The high-resolution O 1s spectra of MBC and MBC-As can be divided into three peaks ( Figure [38]. After As(V) adsorption, a new peak appeared at 533.3 eV that was identified as an As-O bond with 20.13% of the relative area [30]. Moreover, the area ratio for M-OH was also sharply decreased from 43.90% to 30.47%. These data indicated that the oxygen constituents of MBC were linked to the adsorption process. After incorporating Ce and Mn element Following As(V) adsorption, the intensity of peaks situated at 1622 cm −1 and 3442 cm −1 were dramatically decreased and the M-OH bending band at 1106 cm −1 was sharply weakened for MBC. This indicated an association of As(V) on the MBC surface that might be assisted by hydroxyl groups [20,38]. In addition, the MBC spectrum after adsorption of As(V) induced the appearance of new peak at 865 cm −1 that was not present before adsorption. This peak most likely represents the formation of an As-O bending band that was in agreement with other studies [41][42][43]. The formation of As-O bond showed that the As(V) adsorption onto MBC might follow the inner-sphere complex mechanism.

XPS
The chemical and electronic states of the As(V) adsorption to MBC was assessed using XPS.  [38]. After As(V) adsorption, a new peak appeared at 533.3 eV that was identified as an As-O bond with 20.13% of the relative area [30]. Moreover, the area ratio for M-OH was also sharply decreased from 43.90% to 30.47%. These data indicated that the oxygen constituents of MBC were linked to the adsorption process. After incorporating Ce and Mn element into BC, abundant M-OH could be introduced on the surface of MBC, this was especially true for hydroxyl groups (OH-) that have been shown to be the major factor involved in As(V) adsorption [38,44,45]. Above analyses indicated that the As(V) adsorption onto MBC might follow the formation of multiple complexes (As-O-M) involving multiple mechanisms.
The XPS spectra of the Ce3d region of MBC before and after As(V) revealed that binding energies could be assigned to Ce(IV) and Ce(III) [29]. Moreover, the Ce(IV) content (80.22%) was much larger than Ce(III) (19.78%) in accordance with our expectations (Figure 3g,h). The Mn2p could be deconvoluted into three peaks at 641.6, 641 and 645 eV and represented the characteristic of Mn(IV), Mn(II/III) and Mn2p regions, respectively. The primary oxide found by analysis was Mn(IV) (50.02%) (Figure 3i,j). Additionally, the element content percentage of different valence of Mn and Ce changed after As sorption, which suggested that the redox reaction occurred during the adsorption process. could be assigned to Ce(IV) and Ce(III) [29]. Moreover, the Ce(IV) content (80.22%) was much larger than Ce(III) (19.78%) in accordance with our expectations (Figure 3g,h). The Mn2p could be deconvoluted into three peaks at 641.6, 641 and 645 eV and represented the characteristic of Mn(IV), Mn(II/III) and Mn2p regions, respectively. The primary oxide found by analysis was Mn(IV) (50.02%) (Figure 3i,j). Additionally, the element content percentage of different valence of Mn and Ce changed after As sorption, which suggested that the redox reaction occurred during the adsorption process.

Contact Time
As(V) sorption onto BC and MBC was divided into two stages; the kinetic adsorption curve initially rose sharply, and MBC achieved almost 70% of equilibrium sorption capacity within the first 60 min. This deduced an abundance of available adsorption sites that were gradually occupied with As(V). The adsorption rate then slowed and reached equilibrium saturation in second stage. The relatively fast surface adsorption and slow intra-particle diffusion reflected the entire adsorption process (Figure 4).
Using the pseudo-first-order (Equation (3)) and pseudo-second-order models (Equation (4)), kinetics of As(V) sorption were fitted. The two models are as follows: pseudo-first-order model: 7 of 15 pseudo-second-order model: where Q e is the equilibrium As adsorption capacity (mg/g), t is the time of adsorption (min), Q t is the As adsorption capacity (mg/g) at time t, and k 1 and k 2 are adsorption rate constants. The pseudo-second-order model provided a better fit and this model is based on the assumption that the adsorption rate is controlled by the chemisorption mechanism (Supplementary Table S1). This indicated that the sorption process of two materials might be chemisorption. Pseudo-second-order model has been more reliable for describing the adsorption process [46]. Our results indicated that the surface functional groups of MBC were chemically adsorbed with arsenate ions resulting in the formation of inner sphere complexes. In addition, MBC possessed more As(V) surface binding sites due to BC thermochemical conversion [30,37]. We also found that the content of Ce and Mn release from MBC is limited during As(V) adsorption, so the environmental risks can be neglected.
As(V). The adsorption rate then slowed and reached equilibrium saturation in second stage. The relatively fast surface adsorption and slow intra-particle diffusion reflected the entire adsorption process (Figure 4).
Using the pseudo-first-order (Equation (3)) and pseudo-second-order models (Equation (4)), kinetics of As(V) sorption were fitted. The two models are as follows: pseudo-first-order model: pseudo-second-order model: t/Qt = (k2 Qe 2 ) −1 + t/Qe (4) where Qe is the equilibrium As adsorption capacity (mg/g), t is the time of adsorption (min), Qt is the As adsorption capacity (mg/g) at time t, and k1 and k2 are adsorption rate constants. The pseudosecond-order model provided a better fit and this model is based on the assumption that the adsorption rate is controlled by the chemisorption mechanism (Supplementary Table S1). This indicated that the sorption process of two materials might be chemisorption. Pseudo-second-order model has been more reliable for describing the adsorption process [46]. Our results indicated that the surface functional groups of MBC were chemically adsorbed with arsenate ions resulting in the formation of inner sphere complexes. In addition, MBC possessed more As(V) surface binding sites due to BC thermochemical conversion [30,37]. We also found that the content of Ce and Mn release from MBC is limited during As(V) adsorption, so the environmental risks can be neglected.

Initial As(V) Concentration and Temperature
Adsorption is also dependent upon the concentration of adsorbing species in solution. As the As(V) concentrations increased, the adsorbent showed a consistent pattern initial sharp rise and a gradual stabilization. What is more, this pattern was affected by temperature and higher temperatures led to a higher adsorption capacity (Supplementary Figure S4). These experimental data were further fitted by Langmuir (Equation (5)) and Freundlich models (Equation (6)) to express the adsorption process, and the fitted parameters are presented in Table 2. The two models are expressed as follows:

Initial As(V) Concentration and Temperature
Adsorption is also dependent upon the concentration of adsorbing species in solution. As the As(V) concentrations increased, the adsorbent showed a consistent pattern initial sharp rise and a gradual stabilization. What is more, this pattern was affected by temperature and higher temperatures led to a higher adsorption capacity (Supplementary Figure S4). These experimental data were further fitted by Langmuir (Equation (5)) and Freundlich models (Equation (6)) to express the adsorption process, and the fitted parameters are presented in Table 2. The two models are expressed as follows: Freundlich model: where Q e (mg/g) is defined as above, C e (mg/L) represents the As concentration in solution at equilibrium, respectively; Q m (mg/g) is the maximum Langmuir adsorption capacity, k l (L/mg) is the sorption equilibrium constant, and n and k f (mg 1−n ·L n /g) are the Freundlich adsorption affinity parameters related to the adsorption intensity, respectively. The fitted Langmuir adsorption isotherms of As(V) at various temperatures indicated that the As(V) adsorption capacity was temperature dependent and this was best described by the Langmuir model. This demonstrated that monolayer adsorption played a major role in the As(V) adsorption and is consistent with similar modeling of As(V) adsorption using cerium oxide modified activated carbon [30]. It can be known that the maximum adsorption capacity of MBC for As(V) is 89.97 mg/g through the Langmuir model(C 0 = 200 mg/L, pH = 7.0, T = 323 K). Meanwhile, the 1/n values were <1 in the Freundlich model that conventionally represent good adsorption [11,47].
The thermodynamic behavior for As(V) adsorption was estimated using the Gibbs-Helmholtz equation as follows: ln Kd = −∆H/RT + ∆S/R (7) where Kd (Q e /C e ) is the distribution coefficient [48], ∆H (kJ/mol) is the standard enthalpy change, ∆S (kJ/(mol·K)) is the change of entropy, ∆G (kJ/mol) is the Gibbs free energy change, T (K) is the temperature, R is the gas molar constant (8.314 J/(mol·K)). The thermodynamic analysis and parameters are shown in Supplementary Figure S5 and Table  S2. The ∆H calculated at 278 to 323 K were all positive indicating that As(V) adsorption by MBC is endothermic. A high ∆H value ensured a strong interaction between the arsenate anion and the adsorbent. The standard entropy change ∆S > 0 indicated that the degree of freedom of the solid-liquid interface increased during As(V) adsorption by MBC [49]. Moreover, negative ∆G values (−1.33 to −5.20 kJ/mol) indicated that As(V) adsorption was a spontaneous process. The ∆G value can reflect the driving force of the adsorption process and the absolute ∆G value of MBC increased with increasing temperature. This generated a larger adsorption driving force that facilitated the reaction [50]. In general, MBC showed an outstanding adsorption performance with large enthalpy and entropy increases and these collectively contributed to the strong spontaneous As(V) adsorption reaction of MBC.

Effect of pH on As(V) Adsorption
Solution pH often plays a large role in adsorption process so the effects of the initial solution pH on As(V) adsorption was examined ( Figure 5). The adsorption capacity of As(V) for MBC was biphasic and increased from pH 2.0-5.0 and then gradually decreased from pH 5.0-11.0 at the same As(V) concentrations, and the point of zero charge (pH zpc ) of MBC was approximately 5.28. A similar trend was observed in a previous study using cerium oxide-modified activated carbon [30]. These results indicated that at lower pH the MBC adsorbent is protonated, and this expanded the electrostatic interaction between adsorbent and the negatively charged As(V) [51]. In contrast, hydroxyl ions and the anion produced by arsenic hydrolysis can compete for the adsorption sites on the adsorbent surface under alkaline conditions [50]. In addition, hydroxyls, carbonyls, and amides of MBC will be negatively charged at alkaline conditions and constitute repulsive forces resulting in decreased adsorption [52]. The repulsive forces between the negatively charged sorbent and anionic species of arsenic thereby causing a decrease in adsorption efficiency. Therefore, MBC exhibited good performance for a wide range of pH values (pH 2.0-11.0), especially at pH = 5.0 for MBC, the adsorption capacity reached 108.88 mg/g (C 0 = 100 mg/L, dosage = 0.5 g/L, T = 298 K). The fitted adsorption isotherms of As(V) at various pH values indicated that the As(V) adsorption was perfectly described by the Langmuir model (Supplementary Figure S6). When pH = 5.0, the maximum adsorption capacity for As(V) is 104.58 mg/g calculating by Langmuir model (Supplementary Table S3). These results suggested that the electrostatic attraction was one of the dominant mechanisms for As(V) adsorption on MBC.
adsorption capacity for As(V) is 104.58 mg/g calculating by Langmuir model (Supplementary Table  S3). These results suggested that the electrostatic attraction was one of the dominant mechanisms for As(V) adsorption on MBC.
A comparison of other systems used for As(V) adsorption indicated that our MBC composite possessed a high maximum adsorption capacity indicating higher affinity between MBC and As(V) than for other systems (Table 3).   A comparison of other systems used for As(V) adsorption indicated that our MBC composite possessed a high maximum adsorption capacity indicating higher affinity between MBC and As(V) than for other systems (Table 3). Table 3. Comparison of the As(V) adsorption capacity with some reported materials.

Experimental Conditions
Adsorption Capacities (mg/g)

Reference pH
Dosage (g/L)

Effect of Ionic Strength
The effect of ionic strength on the As(V) adsorption in this study was investigated by changing the concentration value of NaNO 3 from 0.005 to 1 M, and the result is shown in Supplementary  Figure S7. It can be seen that the adsorption of As(V) obviously increased for MBC with the increase of ionic strength in the concentration range of 0.005-0.01 M. In term of the whole trend, as the NaNO 3 concentration increased, just little effect of ionic strength on As(V) adsorption on MBC was observed. An analogous phenomenon has also appeared in a novel study [42]. When the ionic strength increases, if the anion adsorption capacity increases or does not change, this manifested that the inner layer surface composite were formed during the process of As(V) adsorption [32]. Because the oxyacid ions adsorbed by the formation of the inner-sphere complexes association is either less sensitive to ionic strength or has greater adsorption for higher ionic strength. This is consistent with the fact that the main surface interaction between As(V) species and metal oxides was the inner sphere complex in nature [54]. Therefore, combining the effects of pH and ionic strength on arsenic adsorption, it suggests that the inner surface complex was formed on the surface of the MBC that involved in multiple mechanisms, such as electrostatic interaction force and surface complexation.

Adsorbent Dosage
The removal efficiency by an adsorbent is also dependent on its dosage ( Figure S8). As the amount of MBC increased at constant As(V) concentration levels, As(V) removal plateaued at 1 g/L and this surpassed that of a Mn-modified pine wood biochar for As adsorption [55]. This plateau indicated a stabilization of removal efficiency and a stable removal rate can be achieved consistent with results obtained for As(V) adsorption on a Fe-Mn-La-impregnated biochar and diclofenac adsorption by pig manure and pine wood biochars [31,56].

Possible Adsorption Mechanism
As(V) adsorption by MBC was investigated using various experimental techniques and the SEM-EDS results indicated that the high efficiency of As(V) adsorption on this modified biochar (MBC) was closely related to the loading of cerium and manganese oxide. As(V) sorption capacity of original biochar (BC) is low (0.79 mg/g), the specific surface area is increased after the modification which is beneficial for the As(V) adsorption, and thus the increased sorption for MBC should be attributed to the presence of the composite metal oxide. At the same time, the abundant hydroxyl groups which played a key role during the As(V) adsorption also ascribed to the loading of cerium and manganese oxide. Besides, the batch experiments as well as FT-IR and XPS analyses indicated that the appearance of As-O bond showed that the As(V) adsorption onto MBC might follow the formation of multiple complexes involving multiple mechanisms that included electrostatic interaction forces, surface adsorption, redox reactions and surface complexation. The schematic diagram of the possible mechanism for the As(V) adsorption was described in Figure 6.

Conclusions
In this study, a wheat straw biochar modified with Ce and Mn (MBC) was successfully synthesized that functioned as an effective As(V) adsorbent. The Langmuir adsorption isotherm and the pseudo-second-order kinetic models correlated satisfactorily with the adsorption results that were perfectly fitted. And the adsorption capacity was 108.88 mg/g at pH = 5.0 (C0 = 100 mg/L, dosage = 0.5 g/L, T = 298 K). These results also suggested that the As(V) was chemisorbed onto MBC. MBC performed better for As(V) removal compared with unmodified biochar and thermodynamic parameters depicted the spontaneous endothermic nature for the As adsorption onto MBC (ΔH = 22.54 kJ/mol, ΔG = −1.33 to −5.20 kJ/mol at 278-323 K). Significantly, it can be inferred that the loading of cerium and manganese oxide on biochar played a major role in As(V) adsorption via SEM-EDS analysis. The batch experiments, FTIR and XPS demonstrated that the high As(V) adsorption capacity was closely related to the formation of multiple complexes under the influence of abundant surface M-OH groups that involved multiple mechanisms including electrostatic interactions, surface adsorption, redox reactions and surface complexation. These results highlight the significant potential of MBC as adsorbent for As(V) removal from natural water and groundwater.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: SEM images of (a) BC and (b) MBC. Figure S2: N2 adsorption/desorption isotherms of (a) BC and (b) MBC, respectively. Inset: the corresponding pore size distribution of the aforementioned two adsorbents. Figure S3. X-ray diffraction pattern of MnO2, CeO2, BC and MBC. Figure S4. Adsorption isotherm under different temperatures. Figure S5. Overall these results indicate that MBC is an effective As(V) adsorbent and can be employed in the remediation of arsenic pollution in water.

Conclusions
In this study, a wheat straw biochar modified with Ce and Mn (MBC) was successfully synthesized that functioned as an effective As(V) adsorbent. The Langmuir adsorption isotherm and the pseudo-second-order kinetic models correlated satisfactorily with the adsorption results that were perfectly fitted. And the adsorption capacity was 108.88 mg/g at pH = 5.0 (C 0 = 100 mg/L, dosage = 0.5 g/L, T = 298 K). These results also suggested that the As(V) was chemisorbed onto MBC. MBC performed better for As(V) removal compared with unmodified biochar and thermodynamic parameters depicted the spontaneous endothermic nature for the As adsorption onto MBC (∆H = 22.54 kJ/mol, ∆G = −1.33 to −5.20 kJ/mol at 278-323 K). Significantly, it can be inferred that the loading of cerium and manganese oxide on biochar played a major role in As(V) adsorption via SEM-EDS analysis. The batch experiments, FTIR and XPS demonstrated that the high As(V) adsorption capacity was closely related to the formation of multiple complexes under the influence of abundant surface M-OH groups that involved multiple mechanisms including electrostatic interactions, surface adsorption, redox reactions and surface complexation. These results highlight the significant potential of MBC as adsorbent for As(V) removal from natural water and groundwater.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4441/12/10/2720/s1, Figure S1: SEM images of (a) BC and (b) MBC. Figure S2: N 2 adsorption/desorption isotherms of (a) BC and (b) MBC, respectively. Inset: the corresponding pore size distribution of the aforementioned two adsorbents. Figure S3. X-ray diffraction pattern of MnO 2 , CeO 2 , BC and MBC. Figure S4. Adsorption isotherm under different temperatures. Figure S5. Thermodynamic analysis of As(V) adsorption by MBC. Figure S6. Adsorption isotherm at different initial pH values. Figure S7. Effect of ionic strength on the adsorption of As(V) by MBC. Figure S8. Effect of MBC dosage on As(V) adsorption. Table S1: Kinetics parameters for As(V) adsorption onto two different adsorbents. Table S2. Parameters for As(V) adsorption by MBC. Table S3. Equilibrium adsorption isotherm fitting parameters of As(V) onto MBC.

Conflicts of Interest:
The authors declare no conflict of interest.