1. Introduction
Chromium (Cr), a common toxic heavy metal, is extensively employed in industrial production, including electroplating, leather tanning, painting, dyeing, mining, etc [1, 2]. Chromium has a variety of oxidation states (from −2 to +6) while is predominantly stable in two oxidation states, i.e., trivalent [Cr(III)] and hexavalent [Cr(VI)] [1]. Compared with Cr(III), Cr(VI) (e.g. HCrO4−, CrO42− and Cr2O72−) exhibits higher toxicity, mobility and solubility [3]. Attentions have been attracted due to the high toxicity of Cr(VI), including teratogenicity, carcinogenicity and mutagenicity [3]. Therefore the reduction of Cr(VI) to the less toxic Cr(III) is the current predominant approach to the treatment of Cr(VI) contamination. Innovative, economical and effective technologies must be developed to effectively eliminate Cr(VI) in the aquatic environment. Nevertheless, weaknesses including high costs, high energy consumption and potential secondary pollution often accompanies with the strengths of most techniques [4]. With the aim of surmounting the challenges related to the most technologies, the removal of Cr(VI) via biochar adsorption and reduction has received considerable attention for its eco-friendliness and cost effectiveness [3].
Biochar, a carbonaceous solid produced by the pyrolysis of biomass under high temperature and anoxic conditions, has been widely used for climate change alleviation, soil rehabilitation, pollution treatment and energy production [5]. One of the most intriguing merits of biochar lies in its superior adsorption properties, such as large specific surface area, abundant porous structure and surface functional groups, and therefore has great potential to eliminate Cr(VI) [2, 6]. In addition to the emphasis on adsorption capacity, biochar is demonstrated as a reductant for Cr(VI) reduction [7]. Biochar could reduce Cr(VI) to Cr(III) and subsequently immobilize it further on the biochar, thus effectively separating chromium from the water or soil [7, 8].
However, the adsorption capacity of conventional biochar is relatively low compared to other adsorbents such as activated carbon [9]. We found in our previous study that the limited adsorption would then inhibit the further removal of Cr(VI) by biochar [8]. Hence it has attracted considerable attention to enhance its inherent properties such as increasing surface area, porosity and functional groups through biochar modification [10–12]. Among them, the modification of biochar with iron loading is one of the current research hotspots. Divalent iron (Fe2+) is one of the most effective reagents for Cr(VI) reduction [13]. However, directly dosing Fe2+-containing reagent for wastewater treatment usually suffers from the problems [13, 14], including the requirement for large excesses of Fe(II) over the stoichiometric amount, efficacy lose due to Fe(II) oxidation before use, alkaline consumption for the precipitation of Fe/Cr-containing products, as well as the production of substantial amount of toxic sludge during the Cr(III) precipitation. Combination of Fe and biochar would benefit to overcome the above shortcomings, through the enhanced adsorption of biochar, improved reduction via biochar-Fe interplay, as well as spontaneous removal of Fe/Cr-containing products from the bulk water. Inherent Fe in biochar was also proved in our previous report to play important roles in both reducing Cr(VI) and immobilizing Cr(III) [8]. The coupled adsorption-reduction processes by one-step application of Fe-modified biochar is expected to become a promising technique for Cr(VI)-containing wastewater treatment.
Intensive reports on the Fe-modified biochar could be found in literature, however, most were based on the nano zero-valent iron (nZVI) and biochar composites in which the biochar was employed as a carrier only for the stabilization of nZVI and majority of the Cr(VI) removal was contributed to nZVI [15, 16]. In some other reports of loaded iron salts, large amounts of iron were also consumed in the preparation process, although this significantly improved the performance of the biochar, which was attributed to a high accumulation of iron [17–20]. Complicated preparation procedures (e.g. for nZVI) or large mass of Fe were also required in these reports to obtain a Fe-modified biochar at a high cost. Additionally, some studies failed to take full advantage of the synergistic effects of the carbon matrix and the Fe-enriched components, and only unilaterally contributed the increased reduction to the carbon matrix [21] or the Fe-enriched components [22]. Furthermore, some other studies may suggest that enriched Fe in the Fe-modified biochar may exist in forms with less redox activities, e.g. crystalline form as γ-Fe2O3 which may only enhanced electrostatic adsorption [17, 18].
A simpler protocol with less consumption of chemical reagent to produce Fe-modified biochar is then to be developed. Additionally, a higher efficacy is also expected due to the synergistic effects of the carbon matrix and Fe-enriched components. Therefore, we propose to obtain the Fe-enriched biochar by the simple methods, i.e. pre-soaking feedstock with low-cost Fe salt solution and then pyrolysis. High removal efficiency of Cr(VI) from real wastewater was aimed to achieve with as few Fe dosage as possible. In this study, experiments were conducted (1) to synthesize and characterize Fe-modified biochars derived from different Fe salts (common iron salt reagents, i.e. FeCl3, FeSO4 and (NH4)2Fe(SO4)2) and to compare their removal capacity for Cr(VI); (2) to explore the underlying mechanisms of Cr(VI) removal by Fe-modified biochar; (3) to compare the economics of other reports; (4) to examine the effectiveness of Fe-modified biochar in practical wastewater applications.
2. Materials and Methods
2.1. Chemicals and Materials
Rice straw was selected as a precursor for the preparation of biochars. The precursor was collected from farmland after harvest and were then washed and cut into pieces before use. Potassium dichromate (K2Cr2O7), diphenylcarbohydrazide, acetone, ferric chloride hexahydrate (FeCl3•6H2O), ferrous sulfate heptahydrate (FeSO4•7H2O), ferrous ammonium sulfate hexahydrate ((NH4)2 Fe(SO4)2•6H2O), ammonium acetate, phenanthroline hydrate, hydrochloric acid (HCl), phosphoric acid (H3PO4), sulfuric acid (H2SO4), sodium hydroxide (NaOH) were all of analytical purity and purchased from Kermel Chemical Reagent Company (Tianjin, China). Concentrated Cr(VI) containing wastewater was collected from the spent liquid of an electrolyser, which contained about 20.56 g/L Cr(VI) and 18.99 g/L SO42−.
2.2. Preparation of Biochars (BCs)
Pre-modification was employed to prepare Fe-modified BCs [23]. In brief, 10 g of rice straw was immersed into 500 mL FeCl3•6H2O, FeSO4•7H2O, or (NH4)2Fe(SO4)2•6H2O solution, respectively, and stirred for 24 h with a magnetic stirrer. To compare the different dosing contents of Fe, 0.04–0.4 g Fe in total was prepared in then 500 mL of Fe solutions. Then the biomass was separated from the solution, washed several times with deionized water, dried at 80°C for 24 h and then pyrolyzed for 2 h in a tube furnace filled with N2 atmosphere at a heating rate of 10°C/min to 500°C. For the temperature selection, a lower temperature is often unfavored for the carbon matrix development and the accumulation for electron donation, while a higher temperature would usually result in declined content of oxygen-containing functional groups which is important for Cr(VI) detoxification [24, 25]. The obtained biochar was ground to pass through a 60-mesh sieve (~ 250 μm). These samples were denoted as Fe-BC, FeS-BC and FeSN-BC, respectively, where Fe, FeS and FeSN represents the salts of FeCl3, FeSO4, or (NH4)2Fe(SO4)2 in the spiking solution. If not stated otherwise, the Fe-BC, FeS-BC and FeSN-BC refers to the modified BCs with 0.4 g-Fe spiking. Biochar derived from rice straw without pretreatment was denoted as RBC.
2.3. Removal of Cr(VI) by the Prepared BCs
To investigate the removal of Cr(VI) by the prepared BCs, BC was added into 40 mL Cr(VI)-containing solution in each polypropylene tube at room temperature (25 ± 2°C). The initial pH value of Cr(VI) solution was adjusted using 0.1 M HCl or 0.1 M NaOH solution. Details of the batch tests were described in Text S1, Supplementary Information and SI). These polypropylene tubes were placed flat on a shaker and shook at 130 rpm for 24 h. Then, the solid-liquid separation was carried out by centrifuging at 3600 rpm for 5 min. The mixture was filtered through a polyether sulfone filter with a pore size of 0.45 μm for Cr determination. The measurements and calculations were described in detail in Text S2, SI.
2.4. Characterization of BCs and Density Functional Theory Calculation
To verify the physicochemical properties of the sample before and after the reaction, several characterizations including elemental composition, pH value, ζ-potential, surface morphology by scanning electron microscopy (SEM), surface area and pore volume, surface functional groups by Fourier transform infrared spectroscopy (FTIR) and Raman spectrometry, element binding state by X-ray photoelectron spectroscopy (XPS), crystalline by X-ray diffraction (XRD), as well as persistent free radicals by Electron paramagnetic resonance (EPR), were carried out. Details can be found in Text S3, SI.
To verify the adsorption behavior, the interaction of Cr(VI) with the Fe species from biochar was modeled using the Castep model based on DFT calculation. The functions were described by generalized gradient approximation (GGA) implemented with PW91 with ultra-fine-quality. The tolerance of SCI was 5.0*e−7 with 100 Max. SCF cycles. All the atoms were allowed to relax until the residue forces on each atom were less than 0.01 eV/Å. The k-point set was fine with 1*1*1. All the configurations were built by the Material studio.
where Eads is the adsorption energy corresponding to the interaction between Fe and Cr(VI), EFe-Cr(VI) is the total energy of the system after termination of the reaction between Fe and Cr(VI), EFe represents the energy of Fe species, and ECr(VI) is the energy of the Cr(VI) ion. The higher the absolute value of Eads, the stronger the adsorption.
3. Results and Discussion
3.1. Characteristics of RBC and Fe-modified BCs
The presence of many particles on the surfaces of the Fe-modified BCs (Fig. S1a–d), compared to the smooth surface of the RBC, suggested that Fe-salt was successfully loaded onto the surface, which was consistent with the elemental composition of BCs measured by EDS (Fig. S1e–h). It can be clearly seen that the element Fe was present in all the Fe-modified BCs but absent in the RBC. The peak height of Fe in the Fe-modified BCs followed the order of Fe-BC > FeS-BC > FeSN-BC, which was consistent with the elemental analysis by digestion method (Table S1). Whilst only trace amounts of S can be detected on the surface of FeS-BC (Fig. S1g), indicating that the efficiency of loading S with FeSO4 or (NH4)2Fe(SO4)2 was very low. This may be due to the fact that S escaped in the form of SO2 during pyrolysis [26]. N was failed to be captured by EDS (Fig. S1e–h), though it could be more accurately detected by elemental analysis (Table S1).
Successful Fe modification on BCs could also be confirmed by the XPS spectra, as shown in Fig. 1a and Table S2. There was no peak in the spectra of RBC (Fig. 1a1), which suggested that no Fe was present before modification. In the spectra of Fe-modified BCs, the bands at 710–714 and 723–727 eV attributing to the peaks of Fe 2p1/2 and Fe 2p3/2, respectively, could be obviously noted [8]. In the spectra of Fe-modified BCs, Fe existed in three valence states (i.e. Fe(0), Fe(II) and Fe(III)). Although lower valences were not directed introduced by the spiking salts, the presence of Fe(0) or Fe(II) could be attributed to the reduction of Fe(II) or Fe(III) by the char during the anoxic pyrolysis [26]. In FeS-BC and FeSN-BC, however, Fe(III) was present, which may be attributed to the oxidization during the pretreatment of Fe loading. XRD patterns suggested new peaks appeared in the Fe-modified biochars at 35.6°, 57.3°, 63.0°(Fig. 1b), corresponding to Fe2O3, indicating that some of the Fe was attached to the surface of the biochar were in the form of oxide crystals.
The surface area (SA) (and micropore surface area, i.e. MicroSA) as well as pore volume (PV) (and micropore volume, i.e. MicroPV) of the biochar increased significantly after the Fe-modification (Table S1). It had been proved that the reaction between Fe and biomass may promote the release of low molecular weight organic matters and the formation of more mesopores [27, 28]. In addition, the types of iron salts may also influence the formation of the pore structure, with FeSN-BC exhibiting the largest SA and PV, which may be due to the formation of a multi-stage pore structure (Fig. S2).
Fe-modification also changed the surface charges of BCs. As noted in Fig. 2a, all the BCs in this study were negatively charged in the tested pH range and the surface ζ-potential further decreased to be more negatively charged as the pH elevated. In comparison, the Fe modification could lower the negative charge, and Fe-BC had the highest surface ζ-potential at low pH, i.e., the least negative charge.
For the functional groups on the surface of BCs, as revealed by FTIR spectra in Fig. 2b, the broad adsorption peak of FTIR at ~3400 cm−−1 indicated the presence of hydroxyl (-OH) [29]. The peak of aliphatic -C=O at 1700 cm−1 only present in Fe-BC and FeS-BC [30]. The peaks at 1620, 1100 and 800 cm−1 corresponded to the -C=O bond of the conjugated quinone, the -C-O stretching vibration of the alcoholic groups and aromatic compounds, respectively [31]. The XPS spectra of four samples presented in Fig. 2c suggested similar percentage distributions among the C-C, C-O, C=O and carbonates groups while slightly higher C-O could be found on Fe-modified BCs (Table S3).
More significant alteration in the carbon matrix may be observed in the graphitic structure revealed by Raman spectra (Fig. 2d). Two characteristic bands at about 1350 cm−1 (D-band) and 1580 cm−1 (G-band) could be attributed to the carbon sp3 and sp2 structures, respectively [32]. The intensity ratio of D-band and G-band (ID/IG) was a parameter to estimate the ordered and disordered graphitic structure in carbon [33]. The ID/IG ratios of Fe-BC, FeS-BC and FeSN-BC were 0.74, 0.71 and 0.64, respectively, lower than that of RBC (0.81), indicating that the possession of more graphite- like structures in the Fe-modified BCs.
As for persistent free radicals (PFRs) in biochar was considered to be potentially important in facilitating the redox activities [34], the BCs were characterized by EPR (Fig. 2e). The highest EPR intensity of RBC indicated that the highest concentration of PFRs were found in RBC. This could possibly be explained by the consumption of PFRs by the Fe modification, since PFRs can act as the electron donor for the reduction of Fe(III) [34, 35]. According to our results, the g-factors for both RBC and Fe-BC were 2.0040, and the g-factors for FeS-BC and FeSN-BC were 2.0041, which were characteristic of carbon-centered radicals with adjacent oxygen atoms and oxygen-centered radicals, respectively [28].
3.2. Comparison of Cr(VI) Removal by RBC and Fe-modified BCs
The Cr(VI) removal experiments were conducted with RBC, Fe-BC, FeS-BC and FeSN-BC, respectively (Fig. 3a&d). As noted, the Cr(VI) removal efficiency of the Fe-modified BCs can reach 100% at most, which was about 18 folds higher than RBC (5.5%). Therefore, the Fe loading to the biochar exhibited considerable enhancement of Cr(VI) removal. As the spiked Fe content increased (x from 0.04 to 0.4), the removal of Cr(VI) by Fe-modified BCs increased. More significant differences among the three Fe-modified BCs could be observed at higher pH condition, i.e. pH = 4. Generally, the removal of Cr(VI) by FeS-BC was slightly higher than that by FeSN-BC, while Fe-BC was significantly higher than the others (Fig. 3d).
Biochar was often an alkaline material [2], so the pH values increased after the reaction from 2.00 to 7.28 or from 4.00 to 9.87 (RBC), respectively (Fig. 3b&e). In addition, the reduction of Cr(VI) to Cr(III) was a proton consuming reaction [15, 23], which was another reason for the increase in the pH value. According to our previous finding, the self-releasing of alkalinity would restrict the primary adsorption of Cr(VI) and thus inhibit the afterward reduction and overall removal [8]. While, the increase in pH values during the reaction process caused by Fe-modified BCs was much less than that by RBC, and as the Fe content in the biochar increased, the pH value further decreased. Therefore, the introduction of Fe to the BCs successfully helped to reduce the alkalinity of the reaction system [23, 28]. From this aspect, the studied 0.4Fe-BC with the lowest pH (Table S1) would be the most favored one for the Cr(VI) treatment.
The reduced Cr(III) existed in two forms, including Cr(III) in solution (Cr(III)aq) and Cr(III) adsorbed on biochar (Cr(III)s), whereas Cr(III)s account for the major fractions (Fig. 3c&f). EDS spectra and high resolution spectra of Cr 2p by XPS indicated the adsorption of Cr on the biochar surfaces (Fig. S3). Compared with the Cr 2p spectra before reaction, obvious bands appeared after the reaction, with binding energies of 575–580 and 585–590 eV, corresponding to the Cr 2p3/2 and Cr 2p1/2 orbitals, respectively [36]. Thereinto, Fe-BC had the most significant peaks for Cr 2p, while only trace signals could be identified on FeS-BC or FeSN-BC. According to the chemical measurements, the Cr(VI) adsorbed on the biochar was almost 100% reduced to Cr(III), since no significant concentration of adsorbed Cr(VI) could be recovered by the extraction of reacted biochar samples. The final step of Cr(III) immobilization with biochar was critical for retaining Cr-containing contaminants rather than releasing them back into solution [7], and was important for reducing Cr toxicity at contaminated sites [37]. The surfaces of Fe-modified BCs were all negatively charged in the tested pH range, enabling electrostatic adsorption with Cr3+.
3.3. Cr(VI) Removal Behavior by the Selected Fe-BC
Based on the comparison above, Fe-BC was the most favored by the Cr(VI) removal. Thus the following study for removal behavior, mechanisms and real application was conducted with Fe-BC.
3.3.1. Effect of solution pH on Cr(VI) removal
The acidic environment was more beneficial for the removal of Cr(VI), and the removal efficiency of Cr(VI) decreased with increasing initial pH (Fig. 4a). The highest Cr(VI) removal efficiency of 100% was achieved at an initial pH of 2.0. However, when the initial pH increased to 6.0, the removal efficiency declined to only 43.4% and 35.3% for initial Cr(VI) concentrations of 50 and 100 mg/L, respectively. It was thus clear that high pH was not favorable for Cr(VI) removal.
3.3.2. Effect of biochar dosage on Cr(VI) removal
In general, it is often expected that the removal efficiency of Cr(VI) increases with increasing biochar dosage. However, the alkalinity of biochar would possibly make the pH elevated, and excessive dosage of biochar might cause undesired adverse impact to the removal of Cr(VI). Thus, the effect of biochar dosage on Cr(VI) adsorption was examined. It could be clearly seen from Fig. 4b that as the dosage of Fe-BC increased from 1.25 to 5 g/L, the removal efficiency significantly increased from 41.0% and 22.4% to 99.5% and 87.5% for initial Cr(VI) concentrations of 50 and 100 mg/L, respectively. At the dosage of less than 10 g/L as tested, no significant adverse effect of too much pH elevation was caused. The increase of active sites with increasing biochar dosage led to an efficient increase in the removal of Cr(VI) [38].
3.3.3. Adsorption kinetics
As shown in Fig. 4c, the Cr(VI) removal efficiency increased rapidly at the beginning and then slowly until equilibrium was reached. The adsorption kinetics fitting results of the four conventional kinetic models (Text S4, SI) were shown in Fig. S4 and Table S4, respectively. The correlation coefficient (R2) of the pseudo-second- order kinetic model was the highest (R2 = 0.9968), indicating that the adsorption was controlled by the chemisorption mechanism, for example, the chemical reactions such as redox reactions between the adsorbent and the adsorbate [39].
A multi-linear fit of the intra-particle diffusion kinetic for adsorption of Cr(VI) on Fe-BC was observed (Fig. S4d). The straight line of the model did not pass through the origin, which suggested that intra-particle diffusion was not the only rate-limiting step, but was also controlled by surface diffusion [40]. Three phases could be identified and followed in order of rate constants: kid,1 > kid,2 > kid,3 (Table S4), confirming a rate decrease during the Cr(VI) adsorption process.
3.3.4. Adsorption isotherm
The adsorption isotherm and the fitting results of the three conventional isotherm models (Text S5, SI) were shown in Fig 4d, Fig. S5 and Table S5. As shown in Fig 4d, the adsorption of Cr(VI) by Fe-BC increased with the initial concentration and tended to increase more slowly, eventually reaching saturation. Adsorption sites on the biochar were gradually fully utilized as the initial concentration increased. As presented in Table S5, Freundlich isotherm produced the most satisfactory fit to the experimental data (R2 = 0.9268). This indicated that the adsorption of Cr(VI) on the surface of Fe-BC was heterogeneous rather than homogeneous [35]. In the adsorption isotherm data, the maximum adsorption capacity (qm) was 21.98 mg/g, which was similar to the adsorption kinetics results. A comparison of the adsorption capacity of Cr(VI) on other adsorbents reported in previous studies was listed in Table S6. The prepared Fe-BC in this study showed a promising adsorption capacity.
3.3.5. Effects of co-existing anions
In general, co-existing ions in wastewater have an effect on Cr(VI) removal by biochar. Therefore, seven types of anions were introduced into solution to investigate the effects of co-existing anions on adsorption efficiency. As shown in Fig. 4e, when the concentration of co-existing anions was lower (50 mg/L), the addition of Cl−, NO3−, and HCO3− increased the removal of Cr(VI), this promoting effect can be attributed to the compression of the electrostatic double layer [36]. But the opposite occurred for CO32−, SO42−, PO43−, and HPO42−, with a decrease in the removal of 1.38%, 3.20%, 8.96%, and 9.62%, respectively. At a higher concentrations (i.e. 100 mg/L) of co-existing anions, all co-existing anions inhibited Cr(VI) removal, with SO42−, PO43−, and HPO42− showing the strongest inhibitory effects. It indicated that co-existing anions could compete with Cr(VI) for adsorption sites and thus inhibit the adsorption of Cr(VI). CO32−, PO43−, and HPO42− could be adsorbed to the surface through the formation of inner-sphere surface complexes that consumed iron ions and occupied the reaction sites [41]. SO42− was the purely competitive adsorption anion and could compete with Cr(VI) for adsorption sites [41].
3.4. Cr(VI) Removal Mechanisms by the Fe-BC
3.4.1. pH dependence and electrostatic interactions
It is intensively reported that the Cr(VI) removal by biochar was often pH sensitive [6, 33, 40] and pH was one of the most limiting factors which determined the overall removal of Cr(VI) [8]. Besides of the proton consuming process of Cr(VI) reduction which prefers acidic environment [15, 23], the electrostatic interactions may be more influential in the primary adsorption of Cr(VI) onto biochar surface [6, 7, 17]. The surface was all biochars negatively charged in the tested pH range and became even more negative as the pH increased (Fig. 2a), which would cause electrostatic repulsion to the anions of Cr(VI).
On the other hand, the species of Cr(VI) alters in different pH (Fig. S6). At pH 2.0 – 6.4, the dominant form of Cr(VI) was chromate (HCrO4− and Cr2O72−), which gradually converted to dichromate (CrO42−) as the pH rose above 4. The adsorption free energy of HCrO4− (−2.5 to −0.6 kcal/mol) is lower than that of CrO42− (−2.1 to −0.3 kcal/mol) [42]. Therefore, in addition to the effects of protons and electrostatic forces mentioned above, when the pH rose above 4, the conversion of the species form made it more difficult for Cr(VI) to be adsorbed on biochar surface.
Compared among the three Fe-modified BCs, Fe-BC did not have the largest SA and PV, but had the highest removal efficiency of Cr(VI), while FeS-BC and FeSN-BC had a significant difference in SA and PV, but almost the same removal efficiency of Cr(VI), indicating that other factors besides of surface area and pores determines the overall performance in Cr(VI) removal. Surface charge and acidity would be the more crucial factors. Fe-modified BCs had less negative charge and higher acidity (low pH) than RBC (Fig. 2a and Table S1), which made them much favored by Cr(VI) removal, especially Fe-BC outplayed the other two modified BCs with the highest surface charge and lowest pH value.
3.4.2. Roles of the enriched Fe
Fe-BC contained considerable amount of Fe, and a significant portion were reductive Fe species (i.e. Fe(0) and Fe(II)). As shown in Fig. 1a2, the XPS spectra of Fe 2p suggesting that Fe(0), Fe(II) and Fe(III) accounted for 24%, 38% and 38% of the total Fe, respectively. After the reaction, the proportion of Fe(0), Fe(II) and Fe(III) in Fe-BC became 18%, 37% and 45%, respectively (Fig. 5a2). The percentages of Fe(0) decreased and that of Fe(III) increased significantly, indicating that oxidation of Fe had occurred. However, the decrease in Fe(II) was not significant, which may be attributed to the oxidation of Fe(0) to produce Fe(II). Similar transformation also occurred in FeS-BC and FeSN-BC (Fig. 5a3–4). Fe(0) and Fe(II) could serve as the important reductants for Cr(VI) reduction. The reaction equations that may be involved were as follows [15, 23].
Cr(VI) was reduced to Cr(III) by Fe(0) and Fe(II) (Eqs. (2–5) at low pH, Eqs. (6–7) at high pH), and additionally at high pH Cr(III) could co-precipitated with Fe(III) to form Fe(III)/Cr(III)(oxy) hydroxides (Eqs. (8–9)). As dissolved Cr(III) could be reoxidized to Cr(VI) in an oxidizing environment, (Fe,Cr)(OH)3 co-precipitation could prevent the reoxidation of Cr(III), and was therefore essential for the successful remediation and removal of Cr.
The interaction of Fe species with Cr(VI) anions was verified with DFT simulation. HCrO4−, CrO42− and Cr2O72− were selected as the model anions. The adsorption energies (Fig. 5b) corresponding to their interaction with Fe(0) and Fe(II) on biochar were calculated as −0.22 – 3.74 eV and 3.11 – 4.42 eV, respectively. Overall, the theoretical calculations verify the Eqs. (1–6). It indicated that Cr(VI) anions had good combining capacity with both Fe(0) and Fe(II). In comparison, Fe(II) may contribute more to the adsorption of Cr(VI) than Fe(0), as it had relatively higher adsorption energies and thus stronger interaction with Cr(VI) anions.
3.4.3. Contribution of carbon matrix and oxygen-containing groups
It was also noted that EPR intensity of BCs decreased after the reaction (Fig. 6a), indicating the participation of PFRs. Because of the free electron nature of PFRs [43], PFRs on biochar could potentially provide electrons directly to reduce Cr(VI) to Cr(III) [36]. However, since the Fe-modified BCs had much less PFRs, the changes in the EPR intensity for the Fe-modified BCs after reacted with Cr(VI) were insignificant. In addition, even though the RBC had a high content of PFRs, the removal of Cr(VI) was low. These two observations suggested that PFRs might not the main contributor in the treatment of Cr(VI) by BCs produced in this study.
As the FTIR spectra shown in Fig. 6b, the peaks of -OH at ~3400 cm−1, -C=O at 1620 cm−1, -C-O at 1100 and 800 cm−1 decreased after the reaction (Fig. 2b), representing the involvement of these functional groups. High resolution C 1s spectra by XPS could provide more sensitive comparisons, as shown in Fig. 2c & 6c and Table S3. Carbonates declined for all BCs after reaction, possibly due to the dissolution to the bulk solution. These may provide anion exchange sites for Cr(VI). Percentage of C-O and C=O also decreased, which was consistent with the FTIR results. To exclude the interferences by the changes of carbonates, the ratio of C=O to C-O could provide clearer information of the oxidative transformation from C-O to C=O, as for the ratios increased for all BCs after reacted with Cr(VI) (Table S3). It is commonly believed that the oxygen-containing groups such as phenolic and carboxylic groups played crucial roles in reduction and complexation [23, 44].
Besides, after the reduction of Cr(VI) by Fe(II) to produce Fe(III), the reductive groups were able to transfer electrons to Fe(III) in the biochar to reduce it to Fe(II) [45–47], which could then continue to reduce the remaining Cr(VI), thus forming an electron transfer chain. Here, biochar could act as an electron shuttle, with reductive groups in the biochar promoting the Fe(III)/Fe(II) cycle through electron transfer [46, 47], and thus increased the reduction rate of Cr(VI). According to the Raman spectra (Fig. 2d), ID/IG decreased after the Fe modification, indicating a change in carbon matrix to more graphite-like structures. As the graphitic structures were often believed could facilitate electron transfer [33, 34], this electron transferring pathway could be more efficient in Fe-modified biochar. Consequently, the Fe-modification could further enhance the redox performances of biochar via facilitating the structural changes of carbon matrix.
3.4.4. Adsorption-reduction-immobilization processes
As noted, after the reaction with Cr(VI) at pH=4, Cr(VI) was not completely removed from the solution (Fig. 3d), but neither Fe(0) nor Fe(II) was completely oxidized in the biochar (Fig. 5a), indicating that the reductive capability was not exhausted. In addition, it was observed that the reaction took a long time to reach equilibrium (Fig. 4c), while the reduction reaction was usually a rapid process [48]. These phenomena indicated that the reduction of biochar may not be the predominantly governing process in the removal of Cr(VI). As reported previously, adsorption played a determining role in the overall removal of Cr(VI), the reduction of Cr(VI) was subject to the adsorption in the previous step [8]. From this point of view, the pH and surface ζ-potential of the material were decisive factors in the overall removal efficiency.
To sum up, the removal mechanism of Cr(VI) by the prepared Fe-BC was an adsorption-reduction-immobilization process [7, 8]. As present in Fig. 7, Cr(VI) was firstly adsorbed on the surface of Fe-BC not only influenced by the electrostatic forces and sensitive to pH (1), but also through binding to Fe(0) and Fe(II) in the biochar (2); Then, Fe-BC acted as an electron donor with the contribution from reductive Fe species (3), oxygen-containing functional groups and PFRs (4), to reduce the Cr(VI) to Cr(III), with reductive Fe species playing the main electron donating roles; The electron shuttling effects facilitated by the carbon matrix, especially by the graphitic structure, further enhanced the electron transferring efficiency (5). Finally, Cr(III) produced by reduction co-precipitated with Fe(III) (6) or was immobilized to the surface of biochar through electrostatic adsorption (7) and complexation (8).
3.4.5. Fe-carbon synergistic effects
Quantitative analysis showed that at pH = 2 the removal of Cr(VI) by the RBC and pure Fe components was 5.5% (Fig. 3a) and ~30.3% (obtained from calculation), respectively. If the Fe component of Fe-BC and the carbon matrix were to work separately, the sum of the removal efficiency was much less than that of Fe-BC (100%) (Fig. 3a). It was thus clear that the synergistic effects of Fe-enriched components and carbon matrix were crucial, contributing to more than half of the total removal rate. The synergistic effects includes the pH adjustment benefited from the Fe-hydrolysis acidity, surface improvement mainly owing to the elevated surface charge, and mostly important the electron shuttling that facilitated the redox reactions for Cr(VI) reduction, as discussed above.
Benefited from these Fe-carbon synergistic effects, the Cr(VI) removal could be achieved to higher efficiency with minimum addition of Fe-species. In Table S6, Fe/C represented the weight of iron required to prepare each gram of adsorbent in previous and the present studies. It can be seen that in the present study Fe/C was significantly lower than that in other literatures, indicating that the dosage of iron consumed for the preparation of biochar was remarkably lower. The qm/(Fe/C) represented the removal capacity of Cr(VI) contributed by each unit of consumed iron dosage in each unit of adsorbent prepared. It was obvious that in our work the value of qm/(Fe/C) was remarkably higher than that in other literatures. Therefore, the Fe-modified biochar prepared in the present paper could achieve high Cr(VI) removal efficiency even with the consumption of low dosage of Fe, which promised both economic cheapness and environmental safety.
3.5. Application in Cr(VI)-containing Wastewater Treatment
To further evaluate the potential of Fe-BC in practical applications, Cr(VI)-contained electroplating wastewater was used as a treatment target. The effect of the initial pH of the solution and the dosage of biochar on the removal of Cr(VI) from the wastewater was investigated. As presented in Fig. 8, the removal of Cr(VI) decreased and increased with increasing initial pH and dosage, respectively, which was consistent with the trend in pure Cr(VI) solution (Fig. 4a&b). As could be seen in Fig. 8a, when the initial pH is less than 2, the removal efficiency of Cr(VI) can reach 100% in the wastewater containing Cr(VI) with initial concentration of 50 or 100 mg/L. The initial pH of the original electroplating wastewater containing 50 and 100 mg/L Cr(VI) without pH adjustment was 2.8 and 2.5, respectively, and the removal efficiency reached 97.5% and 96.9% when treated with 10 g/L Fe-BC.
Noted from Fig. 8b, the removal efficiency of Cr(VI) in wastewater was 4.30% lower than in pure solution. In section 2.1 it was mentioned that Cr(VI) and SO42− were present in the electroplating wastewater at a concentration ratio close to 1:1 and in section 3.3.3 the removal was decreased by 3.20% when the concentration ratio of Cr(VI) and SO42− in solution was 1:1. Thus it could be concluded that the slightly lower removal efficiency in wastewater was attributed to the inhibition of SO42−. The removal of wastewater Cr(VI) with unadjusted pH (pH = 2.5) by Fe-BC was almost the same as that of the wastewater adjusted to 2 at different biochar dosages. Generally, the electroplating wastewater was acidic, which was beneficial for the removal of Cr(VI), and the removal efficiency was still high even without adjusting the initial pH, which could lead to savings in treatment costs in practical applications. However, to achieve better effluent quality (Cr(VI) concentration less than 0.5 mg/L as required by quality standards for integrated wastewater discharge of China [49]), pretreatment of acidification was still required to improve the removal efficiency.
4. Conclusions
In this study, three kinds of Fe-modified biochars were synthesized by pre-immersing the feedstock in FeCl3, FeSO4 or (NH4)2Fe(SO4)2, respectively. The FeCl3 modified biochar (Fe-BC) with the lowest pH, highest surface charge and highest content of Fe achieved the best performance in Cr(VI) removal. The adsorption-based removal was very pH-sensitive and could be well described by the pseudo-second-order kinetic model and Freudlich model. The pH-governing adsorption plays a decisive role, followed by Cr(VI) reduction and Cr(III) immobilization. The reductive Fe species played crucial roles in donating electrons for Cr(VI) reduction, which could be further facilitated by the graphitic structure. Owing to the synergistic effects of the biochar matrix and Fe enrichment, Fe-BC with high efficiency of Cr(VI) could be obtained with minimum dosage of Fe salts via a simple pyrolysis procedure. In conclusion, Fe-BC is expected to be an efficient, cost-effective and promising material for wastewater treatment.
PDF Links
PubReader
Full text via DOI
Download Citation 
Supplement
Print 
