Introduction

Arsenic (As) as a metalloid has been considered to be a proverbial carcinogen. Many anthropogenic activities including the use of insecticides, phosphate fertilizers, mining, and smelting operations have released As into soils and groundwater and thus caused global environment contamination (Kim et al. 2014; Xue et al. 2017). Arsenic-contaminated soils are harmful to food safety and groundwater quality, and eventually threaten human health in many regions of the world such as Mexico, USA, China, Bangladesh, and Pakistan (Jadhav et al. 2015). As compounds exist either as inorganic and organic species (Wu et al. 2016a). Generally, compared with organic As forms, inorganic As species are more detrimental to organisms (Wu et al. 2016b; Wu et al. 2017). Therefore, it is urgent to control the entrance of As especially inorganic As into natural environment. In order to solve the problem of As pollution, many studies are focusing on developing various kinds of adsorbed materials.

Red mud (RM), known as the popular name of bauxite residue, is the residue generated from the process of Al production by Bayer (Xue et al. 2016a; Zhu et al. 2016a). Due to the difficulty of disposal and large volumes (120 million tons per year), it is essential to find alternative options for red mud treatment and use (Xue et al. 2016b; Kong et al. 2017a). Generally, the mineralogical composition of red mud consists of diversified forms of iron and aluminum oxides, calcium and sodium aluminum silicates. Red mud has a fine, porous structural nature and its surface carries significant hydroxyl groups (Liu and Naidu 2014; Zhu et al. 2017). As a result, the material has great potential to remediate As contamination both in soils and groundwater. Previous studies confirmed that red mud, or neutralized red mud by different methods, has the ability to mitigate As in aquatic environment (Altundogan et al. 2000; Guo et al. 2014). Furthermore, it has been demonstrated that RM can inhibit the uptake of As by crops for it will transform from unstable As fractionation to stable forms in soils (Yan et al. 2013; Hua et al. 2017). The interaction mechanisms between As and red mud can be mainly attributed to electrostatic attraction/repulsion, chemical interaction (Castaldi et al. 2010).

Biochar is a stable solid derived from rich carbonaceous biomass which processing thermal or hydrothermal transformation under oxygen-limiting circumstances (Zhang et al. 2013). It has been used in numerous applications mostly as a soil amendment/conditioner, a waste management tool and to mitigate global warming (Khan et al. 2013). Furthermore, biochar is able to enhance soil luxuriance, elevate the bioavailability of soil nutrient, enhance agricultural productivity, and sequester carbon (Lone et al. 2015). However, there is evidence that biochar may also inhibit the bioavailability of essential plant micronutrients (Hartley et al. 2016). Additionally, biochar can effectively remove water contaminants (such as As, Cr and Cd) because of its favorable physicochemical properties (large surface area and porous structure and contains oxygen functional groups and aromatic compounds on its surface) (Tan et al. 2016). The properties of biochar, mainly the large surface area and cation exchange capacity (CEC), can stabilize several kinds of heavy metals like Pb, Zn, Cu (Lu et al. 2017).

Conversely, it has been demonstrated that biochar can elevate soil pH and thus mobilize As, which has limited its application in As remediation (Wang et al. 2017). Recently, investigations have been focusing on the application of biochar-based materials that have enhanced capabilities as adsorbents for removing As from water and soils. Biochar-based materials including ZnCl2-activated biochar, and Ca-/Fe-modified biochar have been applied to mitigate As from water (Samsuri et al. 2013; Xia et al. 2016). In spite of these good effects, biochar-based materials are comparatively complex and expensive to prepare.

Biochar has a porous structure that can be used to support the distribution of particles within its matrix (Wang et al. 2015c). Red mud shows some potential for As adsorption due to its distinctive oxide-rich constitution, especially iron and aluminum oxides, surface hydroxyl groups and its porous nature (Wang et al. 2008; Zhu et al. 2016b, c). Both rice straw and red mud are low-cost, abundant materials. Such is rarely concerned that composite materials composed of red mud and biochar seem to be effective to As remediation. Based on premises, the study attempts to develop a novel red mud-modified biochar, in order to (1) characterize its physicochemical properties, (2) compare the kinetics of As sorption by red mud-modified biochar and untreated biochar produced from rice straw, and (3) investigate adsorption mechanisms using X-ray absorption spectroscopy (XANES).

Materials and methods

Experiment materials

In the present experiment, sodium arsenite (NaAsO2) and sodium arsenate (Na2HAsO4·12H2O) were purchased from Aldrich Chemical Co. Rice straw was collected from a rice field located around the downtown of Wangcheng (Changsha city of Hunan, P.R. China). It was air-dried, milled into powder (0.15 mm), and then used for biochar production (Lu et al. 2017). Red mud (RM), obtained from the Guangxi Pingguo alumina refinery (Pingguo, Guangxi Province, China), was oven-dried overnight at 60 °C.

Biochar preparation

Red mud-modified biochar (RM-BC) was produced using the method adopted by Yao et al. (2014). The mixture solution was prepared by 2 g of raw RM and 500 mL deionized water and then stirred for 30 min to obtain stable RM suspensions. Subsequently, 10 g of rice straw was thoroughly mixed with the RM suspension and the mixture was stirred for 2 h. Finally, the RM-modified feedstock was obtained by filtering and then oven-drying at 80 °C.

Pyrolysis of the RM-modified feedstock took place in a muffle furnace (SX2-5-12, Yuandong Therm corporaion in Changsha, China) at 600 °C. In order to maintain biomass pyrolysis process preparation conditions, oxygen-free N2 was circulated into the system at a flowrate of 200 mL/min. The temperature increased at a rate of 5 °C/min until 600 °C. The raw materials were pyrolyzed in a muffle furnace for 6 h. After cooling to room temperature (25 °C), prepared materials were washed by deionized water for three times with the purpose of eliminating foreign substances, and oven dried at 80 °C. Rice straw without RM modification was prepared by the same process. Hereafter, red mud-modified biochar and untreated biochar are named RM-BC and BC, respectively.

Material characterization

BC and RM-BC were dried and sieved for instrument analysis. The surface areas were measured via surface area analyzer and calculated by the method of N2-BET. Variant morphologies and elements analysis of the BC and RM-BC were recorded by scanning electron microscope which equipped with energy-dispersive X-ray spectroscopy (SEM-EDS, Phenom XL). Furthermore, the crystal structure of these biochar-based materials was detected through X-ray powder diffraction (XRD).

The effects of solution pH on adsorption

10 mg/L As(V) and 10 mg/L As(III) as working solutions were obtained through diluting stock solutions prepared by dissolving Na2HAsO4·12H2O and NaAsO2 respectively. The working solutions contained 0.01 mol/L NaCl can be capable to keep a stable ionic strength. Approximately 0.12 g of BC or RM-BC was added to 30 mL As(V)/As(III) working solutions of the different pH values to make 4 g L−1 concentrations. Then, the solution pH was adjusted to 2, 4, 6, 8, 10, and 12, respectively, using 0.1 mol L−1 HCl/NaOH (Feng et al. 2015). Subsequently, these samples were shaken at a constant temperature (25 ± 1 °C) in an oscillator bath (DDHZ-300, Experimental facility corporation in TaiCang) at 200 rpm/min for 24 h. Samples were then immediately passed through syringe membrane filters (0.45 μm) to determine As concentrations. Total As in the filtrate was measured by the instrument of HG-AFS. Each treatment had three repetitions. The As adsorbed concentration q (μg g−1) was calculated as follows:

$$ \mathrm{q}=\frac{\left({\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{t}}\right)\mathrm{V}}{\mathrm{m}} $$
(1)

where q (μg g−1) means As adsorbed concentration at equilibrium; C0 and Ct (μg L−1) refer to the original and equilibrium solute concentrations individually; V (L) on behalf of solute volume and m (g) indicates the adsorbents amount.

Adsorption kinetics

To investigate As adsorption mechanisms, adsorption kinetic studies were implemented at around 25 °C. Approximately 0.12 g (4 g L−1) of BC and RM-BC were added to 30 mL 10 mg L−1 As(V) solutions; the pH of these mixture were respectively adjusted to 6 and 2 based on the former study. In addition, 0.12 g (4 g L−1) of BC and RM-BC also were added to 30 mL 10 mg L−1 As(III) solutions. The pH of the mixture was regulated to 10 employing HCl or NaOH. All the treatments were then shaken at 200 rpm/min for predetermined different time periods of 0.5, 1, 2, 4, 8, 12, 16, and 24 h, respectively. The pH of all mixture solution changed lower than 0.5~0.8 unit. Samples were filtered for As determination via HG-AFS. Each treatment had three repetitions.

Kinetics data were fitted with four classical kinetics models to reveal the adsorption mechanism. The governing equations of the mathematical models are listed as follows:

$$ \mathrm{pseudo}-\mathrm{first}-\mathrm{order}:\kern0.5em \frac{\mathrm{dqt}}{\mathrm{dt}}=\mathrm{k}1\left(\mathrm{qe}-\mathrm{qt}\right) $$
(2)
$$ \begin{array}{cc}\hfill \mathrm{pseudo}-\mathrm{second}-\mathrm{order}:\hfill & \hfill \frac{\mathrm{dqt}}{\mathrm{dt}}=\mathrm{k}2{\left(\mathrm{qe}-\mathrm{qt}\right)}^2\hfill \end{array} $$
(3)
$$ \begin{array}{cc}\hfill \mathrm{Elovich}:\hfill & \hfill \frac{\mathrm{dqt}}{\mathrm{dt}}=\alpha \exp \left(-\upbeta \mathrm{qt}\right)\hfill \end{array} $$
(4)
$$ \begin{array}{cc}\hfill \mathrm{intraparticle}\ \mathrm{diffusion}:\hfill & \hfill {\mathrm{q}}_{\mathrm{t}}={\mathrm{K}}_{\mathrm{d}}{\mathrm{t}}^{\frac{1}{2}}+\mathrm{C}\hfill \end{array} $$
(5)

where q t and q e (μg/g) represent the adsorbent remove As concentrations at t time and equilibrium time, and k 1 (h−1), k 2 (g μg−1 h−1), and k d (μg g−1 min−1/2) are the rate of kinetic constants; C (μg g−1) is the y-intercept. Initial adsorption rate is regarded as α (μg g−1) and β (μg g−1) the desorption constant.

Adsorption isotherms

Adsorption equilibrium isotherms for As adsorption on RM-BC and BC were measured at various As level (1–50 mg/L As(V)/As(III)). Equilibrium adsorption investigations were performed for 24 h. The change of the solution pHs was small. Each treatment had three repetitions. The models of Langmuir and Freundlich were used to explain adsorption isotherms. Equations were as listed below:

$$ \begin{array}{cc}\hfill \mathrm{Langmuir}:\hfill & \hfill {\mathrm{q}}_e=\frac{{\mathrm{q}}_{\mathrm{m}}{\mathrm{k}}_{\mathrm{L}}{\mathrm{C}}_{\mathrm{e}}}{1+{\mathrm{K}}_{\mathrm{L}}{\mathrm{C}}_{\mathrm{e}}}\hfill \end{array} $$
(6)
$$ \begin{array}{cc}\hfill \mathrm{Freundlich}:\hfill & \hfill {\mathrm{q}}_{\mathrm{e}}={\mathrm{K}}_{\mathrm{F}}{{\mathrm{C}}_{\mathrm{e}}}^{\mathrm{n}}\hfill \end{array} $$
(7)

where q e (μg g−1) indicates the adsorbed As concentration at equilibrium; C e (mg L−1) represents the equilibrium concentration of As; q m (μg g−1) denotes the maximum adsorption capacity and n refers parameter; K L (L mg−1)and K F represent the constants of Langmuir and Freundlich, respectively.

R L (the equilibrium parameter in the Langmuir model) used to be employed to determine whether adsorption is favorable or not, is as follows (Sun et al. 2015):

$$ {\mathrm{R}}_{\mathrm{L}}=\frac{1}{1+{\mathrm{K}}_{\mathrm{L}}{\mathrm{C}}_0} $$
(8)

where C 0 (mg L−1) suggests the incipient concentration of As and K L (L mg−1) represents Langmuir constant.

XANES analysis

Mechanism of As(V) sorption on RM-BC was detected by X-ray absorption spectroscopy (XANES). All minerals (hematite, magnetite, goethite, and gibbsite) were obtained from Sigma primarily aimed at reference spectra. Approximately 4 g L−1 of each mineral was applied to 30 mL of As(V) solution (10 mg·L−1) for 24 h using 0.01 mol·L−1 NaCl to maintain ionic strength. The suspension was centrifuged and then filtered. Solid samples were washed with ultrapure water, and then freeze-dried for subsequent analysis. The data were acquired at beamline BL14W at the Shanghai Synchrotron Radiation Facility (SSRF). Data was processed by Athena IFEFFIT package and SixPack to quantitatively determine As species.

Data analysis

The data in the paper was processed by Microsoft Excel 2010 and SPSS 19.0 and photographs were fabricated by Origin 9.0.

Results and discussion

BC and RM-BC characterization

The BET N2 surface areas of BC and RM-BC were 210.29 and 186.95 m2/g. The morphology images and EDX of RM and RM-BC are shown in Fig. 1. The EDX spectrum and SEM confirmed that the surface of biochar was loaded with red mud. The EDX analysis of RM-BC revealed extra peaks for calcium, iron, aluminum, titanium and sodium, which are the elemental compositions of red mud when compared to that of BC (Fig. 1b). The surface of RM-BC appeared smooth without adsorption (Fig. 1b) and then became coarser after As(V) adsorption process (Fig. 1c) and As(III) removal process (Fig. 1d). The EDS spectra analysis showed that As was coated on the surface of the RM-BC, demonstrating that adsorption had occurred (Fig. 1c, d).

Fig. 1
figure 1

SEM-EDS of biochar (BC) and red mud-modified biochar (RM-BC). a BC, b RM-BC before adsorption, c RM-BC after adsorption of As(V), d RM-BC after adsorption of As(III)

Full size image

The structures of RM-BC and BC were clearly different, when detected by XRD. The crystalline phases of RM-BC were probed and identified as hematite, magnetite, goethite, calcite, gibbsite, and perovskite which were not detected in the crystalline phases of BC (Fig. 2). The results further confirmed that red mud had been successfully loaded on the surfaces of the biochar matrix.

Fig. 2
figure 2

XRD spectra of biochar (BC) and red mud-modified biochar (RM-BC)

Full size image

The effect of solution pH

Figure 3a, b respectively showed the effect of different pH on the As adsorption process of each biochar (BC/RM-BC). The quantity of adsorbed As(V) (q, μg g−1) decreased with increasing pH on RM-BC (Fig. 3a). The maximum adsorption density for As(V) at pH = 2 was 1622.51 μg g−1. Adsorption capacity of BC for As(V) raised in the wake of pH from 2 to 6, but then reduced when the pH ranged from 6 to 12. The maximum adsorption of BC was q = 481.61 μg g−1 at pH = 6. Biochar had a somewhat lower As(V) removal capacity, contrasted with that of RM-BC.

Fig. 3
figure 3

Effects of pH on the concentration of a As(V) and b As(III) adsorbed by BC and RM-BC

Full size image

The influence of pH on As(III) adsorption are presented in Fig. 3b. Clearly, BC and RM-BC showed similar trends on the pH-dependent adsorption curves. The quantity adsorbed As(III) was on the rise following the solution pH from 2 to 10, but when pH was elevated from 10 to 12, the adsorbed amount for As(III) of both BC and RM-BC reduced. Overall, the results indicated that As(V) is adsorbed in acidic pH conditions, but As(III) tends to be adsorbed in alkaline pH circumstances. In addition, red mud modification significantly enhanced the sorption ability of biochar both on As(V) and As(III).

pH is considered as a crucial factor influencing As species distribution and the surface charge of BC/RM-BC. Anionic species of arseneous and arsenic acids are dictated by pH. Ionization equations are as listed below:

$$ \begin{array}{cc}\hfill {\mathrm{H}}_3{\mathrm{AsO}}_3\leftrightarrow {\mathrm{H}}_2{{\mathrm{AsO}}_3}^{-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_1=9.23\hfill \end{array} $$
(9)
$$ \begin{array}{ccc}\hfill \mathrm{As}\left(\mathrm{III}\right)\hfill & \hfill {\mathrm{H}}_2{{\mathrm{AsO}}_3}^{-}\leftrightarrow {{\mathrm{H}\mathrm{asO}}_3}^{2-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_2=12.10\hfill \end{array} $$
(10)
$$ \begin{array}{cc}\hfill {{\mathrm{H}\mathrm{asO}}_3}^{2-}\leftrightarrow {{\mathrm{AsO}}_3}^{3-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_3=13.41\hfill \end{array} $$
(11)
$$ \begin{array}{cc}\hfill {\mathrm{H}}_3{\mathrm{AsO}}_4\leftrightarrow {\mathrm{H}}_2{{\mathrm{AsO}}_4}^{-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_1=2.24\hfill \end{array} $$
(12)
$$ \begin{array}{ccc}\hfill \mathrm{As}\left(\mathrm{V}\right)\hfill & \hfill {\mathrm{H}}_2{{\mathrm{AsO}}_4}^{-}\leftrightarrow {{\mathrm{H}\mathrm{AsO}}_4}^{2-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_2=6.94\hfill \end{array} $$
(13)
$$ \begin{array}{cc}\hfill {{\mathrm{H}\mathrm{AsO}}_4}^{2-}\leftrightarrow {{\mathrm{AsO}}_4}^{3-}+{\mathrm{H}}^{+}\hfill & \hfill {\mathrm{pKa}}_3=11.50\hfill \end{array} $$
(14)

Given pH < pHPZC (point of zero charge), positive charges will form on the surface of biochar. At this time, As(V) predominantly as H2AsO4 − form and adsorption occurs by electrostatic and chemical attraction (Cheng et al. 2016). At higher pH values, the major As species exist as HAsO4 2− and AsO4 3−. However, the availability of binding sites declines with increasing pH. When the pH is above the pHPZC adsorbent surfaces develop negative charges, which repulse anions including HAsO4 2− and AsO4 3−. In the meantime, OH− will compete with As anion for positively charged surface sites, and as a result weaken the adsorption process (Feng et al. 2015). However, the pHpzc value of red mud could not be accurately determined since some components of red mud were solubilized during the potentiometric titration (Altundogan et al. 2000). Furthermore, some studies reported that the properties and mineralogical composition of RM will be changed by pH and specific sorption locations of RM will decrease with increasing pH (Castaldi et al. 2010; Kong et al. 2017b).

The adsorbed volume of As(III) reached its maximum between pH 8 and 10 (Fig. 3b). This is agreed with the consequence of Manju et al. (1998) which used copper impregnated activated carbon as a sorbent, and discovered that As(III) adsorption elevated with the increasing of pH (Manju et al. 1998). Baig et al. (2014), using magnetic biochar, who noted that As(III) adsorption reached a maximum at pH = 8 (Baig et al. 2014). However, Yu et al. (2015) demonstrated that As(III) reached a maximum at pH = 3 with manganese oxide-modified biochar (Yu et al. 2015). This variation may be the result of the diverse adsorbents used in the investigations.

Adsorption kinetics

As(V) and As(III) kinetic experiments revealed that the sorption occurred rapidly firstly and then became slowly, finally reached equilibrium within 24 h (Fig. 4). First-order and second-order models were applied to elucidate mononuclear and binuclear adsorption, while the Elovich model considered the influence of desorption. Compared with BC, RM-BC showed a greater sorption capacity for As. Table 1 listed kinetic models and best-fit parameter values for As adsorption data. Compared with other models, the data were much better conformed to the pseudo-second-order pattern for BC and RM-BC adsorbed As(V) (Table 1). However, for As(III) adsorption on BC and RM-BC, the fit is different. Between the four kinetic models, the former is more in favor of the pseudo-second-order model. When As(III) was adsorbed on RM-BC, the Elovich model agrees with the data more accurately.

Fig. 4
figure 4

Adsorption kinetics and modeling for As on adsorbents: a and b are As(V) adsorption on BC and RM-BC, respectively. c and d are As(III) adsorption on BC and RM-BC, respectively

Full size image
Table 1 Kinetic models and best-fit parameters for As adsorption data
Full size table

The results showed that (1) the number of BC/RM-BC surface active sites probably impact the reaction rate in part and the rate- determining step will possibly due to chemical sorption in line with the pseudo-second-order kinetic model assumption. (Taty-Costodes et al. 2003; Mohan et al. 2011), (2) RM-BC adsorbed As(III) was a heterogeneous process and may be controlled by multiple mechanisms including chemisorption (Yao et al. 2014), and (3) both BC and RM-BC removed As(V) more effectively than As(III). Moreover, the adsorption amount of As(V) on RM-BC is more than that of BC.

Adsorption isotherms

Langmuir and Freundlich isotherms for As adsorbed onto BC and RM-BC are displayed in Fig. 5. With the increase of As concentration, As adsorption gradually rose until equilibrium was reached. In this study, all sorption isotherms were “L” shaped with higher R 2 compared with Freundlich (Table 2). The calculated Langmuir maximum adsorption capacity (Qmax) of RM-BC for As(V) and As(III) were 5923.8 and 520.0 μg g−1 respectively. In contrast, the Qmax of BC for As(V) and As(III) were only 552.0 and 447.6 μg g−1, respectively.

Fig. 5
figure 5

Adsorption isotherm data and modeling for As onto adsorbents: a and b are As(V) adsorption on BC and RM-BC, respectively. c and d are As(III) adsorption on BC and RM-BC, respectively

Full size image
Table 2 Isotherm models and best-fit parameters for As adsorption data
Full size table

Langmuir isotherm is used to describe monolayer adsorption on sorbent surfaces. Both BC and RM-BC obey the adsorption mechanisms of Langmuir, which is in accordance with many carbonaceous adsorbents (Zhang et al. 2013; Wang et al. 2015a). Red mud-modified biochar exhibited improved removal for As(V) and As(III) compared to BC, suggesting that it may enhance biochar adsorption sites (Wang et al. 2015a; Li et al. 2016).

In this work, RL values for As(V) adsorption on BC and RM-BC and As(III) adsorption on BC and RM-BC were in the range of 0.0663–0.825, 0.303–0.970, 0.136–0.919, and 0.0778–0.860, respectively, indicating that adsorption was favorable.

Table 3 The maximum As(V) adsorption amount differed in various adsorbents
Full size table

Table 3 suggested that the maximum As(V) adsorption amount differed in various adsorbents. The Qmax of RM-BC for As(V) is inferior to the Qmax of Fe coated biochars (Samsuri et al. (2013)) and La loaded biochar (Feng et al. (2015)). Nevertheless, production of biochar treated with pure chemical reagents (Fe and La) would be expensive at an industrial level. Moreover, the Qmax of red mud modified biochar was much higher than that of other three biochar-based materials such as hematite modified biochar, MnCl2·4H2O modified biochar and birnessite modified biochar. Additionally, RM-BC is significantly better than pure red mud (Table 3).

Table 4 SPOIL values of references spectra of As(V) adsorption on RM-BC
Full size table

Analysis of adsorption mechanisms

Considering RM-BC exhibited a better As(V) adsorption capacity than BC based on the above research results and As(III) is unstable that easy to be oxidized (Adra et al. 2015). Therefore, we investigate the mechanism of adsorption As(V) on RM-BC.

K-edge XANES analyses for As standards and RM-BC-As(V) are presented in Fig. 6. The excitation energies are displayed at 11872.4 and 11,876.5 eV corresponding to As(III) and As(V). It was revealed that both As species were adsorbed on the surface of the red mud by XANES spectra. Arsenate was the major species and accounted for 97.6% of the total As (Table 4). In addition, LCF results revealed three main As phases on RM-BC including hematite-As(V) (53.5%), magnetite-As(V) (33.8%), and gibbsite-As(V) (13.6%), which probably formed during the sorption process (Table 4).

Fig. 6
figure 6

Normalized As k-edge XANES spectra for As(V) adsorbed onto RM-BC a the percentages of As species; b the percentages of As adsorbed on different minerals

Full size image

Previous studies demonstrated that the existence of iron and alumina oxides was indispensable to As sorption (Ding et al. 2017; Bai et al. 2016; Chen et al. 2017). It is revealed by XRD that hematite, goethite, magnetite, and gibbsite existed on RM-BC which is consistent with that reported in other studies which showed that the high temperature influence the components of red mud such as hematite and goethite (Liu and Naidu 2014). According to correlational research that high-temperature-treated hematite may lead some hematite turn to maghemite (Wang et al. 2015a). The diffraction peak of goethite has been reduced by reason of the chemical reaction 2FeOOH → Fe2O3 + H2O. In other words, the mineral phrases transform from goethite to hematite gradually (Yang et al. 2015). The mechanism of gibbsite can be effectively bound to Arsenate mainly through forming inner sphere complexes (Ladeira et al. 2001). Hematite, magnetite, and goethite effectively reduced As in aquatic environments, especially hematite demonstrating greatest adsorption capacity in acidic pH conditions (Gimenez et al. 2007; Mamindy-Pajany et al. 2011). Iron oxides can be able to effectively bound As mainly via ligands exchange forming inner sphere complexes (Mamindy-Pajany et al. 2011).

According to XANES results, possible adsorption mechanisms for As(V) on RM-BC are presented in Fig. 7. The structure M-O-As(V) formed may substantially enhanced the adsorption capacity of As(V). The possible mechanisms responsible for the adsorption of As on RM-BC may be similar to previous studies demonstrating that the mechanisms responsible for As adsorption on biochar are through electrostatic interactions, surface complexation (inner sphere surface monodentate and bidentate complexes), and chemical interactions between As and surface functional groups (Li et al. 2016; Zhang et al. 2013).

Fig. 7
figure 7

Mechanism of As(V) adsorption on red mud-modified biochar, where M is the metal oxide composition of the red mud surface

Full size image

Conclusion

Red mud and rice straw biochar are promising materials for As-contaminated soil remediation due to their improved adsorbent capacity. A red mud-modified biochar (RM-BC) was created from red mud suspension treating biomass through pyrolysis. The kinetic processes of As(V) adsorption on RM-BC fitted the pseudo-second-order model, while that of As(III) favored the Elovich model. Moreover, all sorption isotherms produced improved fits for the Langmuir model. RM-BC revealed improved adsorption capacity of As(V) and As(III), with the maximum adsorption capacity ten times greater than that of the untreated biochar. XANES results indicated that the As(V) adsorbed onto RM-BC was possibly due to surface complexation and electrostatic interactions.