Nano-Scale Drinking Water Treatment Residuals A ﬀ ect Arsenic Fractionation and Speciation in Biosolids-Amended Agricultural Soil

: An incubation experiment was conducted to determine the e ﬀ ects of nanoscale drinking water treatment residuals (nWTRs) on arsenic (As) fractionation and speciation in agricultural soil amended with biosolids. The soils were treated with biosolids of 3% (w / w), along with nWTR application rates of 0, 0.25, 0.50, or 1.00% (w / w). The results revealed that the As adsorption rate increased with increasing the As treatment level from 50 to 800 mg / L. The maximum e ﬃ ciency of As adsorption was 95%–98% in the soil treated with nWTRs of 1%, while the least As adsorption was 53%–91% in the soil treated with nWTRs of 0.25%. The overall As bioavailability in the biosolids-amended soil followed a descending order of nWTRs treatment: (0%) > 0.25% nWTRs, > 0.50% nWTRs, and > 1% nWTRs. The addition of nWTRs signiﬁcantly changed As speciation in biosolids-amended soil. The X-ray absorption near-edge structure spectroscopy (XANES) and MINEQL + 4.6 analyses showed that most of As was in a oxidized form of As 5 + that likely incorporated in As pentoxide, and thus, with low mobility, bioavailability, and toxicity. This study demonstrated that nWTRs were e ﬀ ective in adsorbing and immobilizing As in biosolids-amended agricultural soils by forming stable As-nWTR surface complexes.


Introduction
Arsenic (As) is a common contaminant in soils and waters. The large environmental input of As has mainly been related to anthropogenic activities including those originated from agricultural land application of fertilizers, pesticides, and/or biosolids (treated sewage-sludge originating from municipal wastewater treatment plants). Thus, As pollution has been recognized as an important matter of concern for worldwide public health [1]. Thus, As remediation technologies have been developed to reduce As contents or limit As bioavailability in the environment [1][2][3][4]. Arsenic remediation can be expensive [5] and have application limitations [6,7]. In recent years some "green" remediation techniques have been proposed, and different As-adsorbents including nanomaterials have been evaluated to reduce As bioavailability in soil and water [8][9][10].

Soil Stabilization
The biosolids-and nWTR-treated soils were incubated at 25 ± 2 • C under the laboratory conditions for 60 days for stabilization. The biosolids application rate was 3% (w/w) in the experimental soils, and the soil nWTR treatments were 0, 0.25, 0.5, and 1.0% (w/w). The control soil was amended with 3% of biosolids but without nWTR treatment (i.e., 0%). The treated soil was homogenized and distributed into different jars. The soil water content in each jar was kept at 80% of the soil water holding capacity. Soil water loss was compensated daily during the experimental time period. At the end of the incubation time period, soil samples were collected from jars and stored at 4 • C for chemical analysis. A completely randomized experimental design was selected, and each treatment had three replicates.

Arsenic Sorption and Sorption Efficiency
The arsenic sorption capacity of each treated soil was determined, compared to the control soil, using the conventional batch equilibration technique. In brief, a unit mass of the treated soil sample (0.2 g) was mixed with 20 mL As solution of known concentration (0, 5, 50, 100, 200, 300, 400, 500, 600, 700, and 800 mg/L as NaAsO 2 ) in a flask, and the flask was placed on a shaker at 120 rpm at 25 • C for 24 h (i.e., the equilibrium time from kinetics experiment). Then, the treated soil samples with the As (III) solution was centrifuged at 5000× g for 15 min prior to sampling. The difference in As (III) sorption between un-amended and nWTRs-amended soils were quantified. At the end of this experiment, final As concentrations in the solutions were measured by Inductively-coupled Plasma/Mass Spectroscopy (ICP/MS). The amount of As adsorbed to the soil was calculated using the following equation: where q e is the amount of As sorbed by a unit mass of soil at equilibrium (mg/kg), V is the solution volume (mL), m is the mass of soil sorbent (g), and C 0 and C e are the initial and equilibrium concentrations of As, respectively. To describe the As sorption in treated soils, the isotherm classification concept was applied to distinguish four main types of isotherms corresponding to different solute adsorbent interactions, named as L, H, C, and S curves. Soil isotherms were characterized by a decreasing slope as equilibrium concentration increased. The As sorption efficiency of sorbent for arsenic was calculated using the following equation: Appl. Sci. 2020, 10, 5633 where E is the sorption efficiency, while C i and C e are the initial and equilibrium As concentrations, respectively.

Arsenic Fractionation
Fractionation analysis of As in the soil treated with biosolids was carried out before and after the nWTR-treatments by using the sequential extraction technique [28]. The sequential extraction method was used to divide soil As into five fractions: Exchangeable As, carbonate-bound As, Fe-Mn oxide absorbed As, organic matter-bound As, and As residual. Concentrations of As in extraction solutions were measured using ICP/MS.

Soil Solution Collection
Soil pore water samples were extracted by centrifuging soil samples for 20 min at relative centrifugal force (RCF) of 4.8 × 10 5 m/s 2 , according to the method described previously by Elkhatib et al. [29]. Soil pore water samples were analyzed for concentrations of As and other major ions, along with pH and temperature.

Speciation Modeling Using MINEQL+
In the pore water samples, soil reaction, electrical conductivity, major ions, toxic metals, and soluble As concentrations were determined. Using the measured As and other element concentrations, pH, temperature, log K, pCO 2 , pO 2 , pE, and calculated ionic strength, As speciation in soil pore water was calculated/modeled using the MINEQL+ (version 4.6) according to Schecher and McAvoy [30].

X-ray Absorption Near-Edge Structure (XANES) Analysis
XANES analysis was carried out at the Advanced Photon Source (APS) of Argonne National Laboratory in the U.S. XANES spectra of selected As standards and treated soil samples were collected at Beamline 13-ID-E. The incident beam and energy reading of As K edge was set to 12 keV. All spectra were collected at room temperature and normal atmospheric pressure. Arsenic standards included arsenic oxide (As 2 O 3 ), diarsenic pentoxide (As 2 O 5 ), arsenic sulfide (As 2 S 2 ), sodium arsenate (HAsNa 2 O 4 ), and cacodylic acid [(CH3) 2 AsO 2 H]. When needed, the chemicals were ground into fine powders and transferred on the sample holder. The prepared sample slides were mounted on a sample holder and attached to the detector for analysis. Two scans were collected for As standards, and three scans were collected for each soil sample. SixPack software [31] was used to average replicate spectra from each sample and fit the averaged spectra to those of As standards.

Statistical Analysis
The least significant difference (LSD) was used for comparison between the biosolids and the nWTR treatments. The level of significance (α) was 0.05. The statistical analysis was done using COSTAT Software [32].

Sorption Isotherms and Sorption Efficiency of Arsenic
The nanoscale drinking water treatment residuals significantly affected the sorption isotherm of As in calcareous sandy loam soil that was treated with different levels of biosolids ( Figure 1). Differences in As sorption capacity were observed between different nWTR-application rates of 0.25, 0.50, and 1.00% (w/w). The initial As concentration significantly affected the As adsorption to the soil treated with both biosolids and nWTRs. When the initial As concentration increased from 5 to 800 mg/L at room temperature, the amount of As sorbed per unit mass of the treated soil also increased. The soil As concentration increase would allow more As molecules to move from soil solution to the surfaces of adsorbents in soil. Thus, the As sorption in soil is highly dependent on the initial As concentration. At a lower initial As concentration, the As mass in solution is relatively small to the amount of available sorption sites, so it was highly sorbed with available sorotion sites [33]. Importantly, if without any nWTR treatments, the soils with or without biosolids treatments had shown moderate affinity for As sorption. With the nWTR treatments, As sorption isotherms changed from S-type to H-type isotherms ( Figure 1). In this study, the H-type isotherm became dominant especially at a high nWTR application rate of 1%. Therefore, the quantity of As sorbed by the soil followed a descending order of: 1% nWTR-treated soil, > 0.5% nWTR-treated soil, > 0.25% nWTR-treated soil, and > control soil (or no nWTR treatment) ( Figure 1). The nWTR treatments have contributed many more sorption sites such as aluminum, iron, and calcium hydroxides in the soils, in addition to high organic matter contents and high cation exchange capacity of biosolids-amended soils ( Table 1).
Appl. Sci. 2020, 10, x 5 of 13 treated with both biosolids and nWTRs. When the initial As concentration increased from 5 to 800 mg/L at room temperature, the amount of As sorbed per unit mass of the treated soil also increased. The soil As concentration increase would allow more As molecules to move from soil solution to the surfaces of adsorbents in soil. Thus, the As sorption in soil is highly dependent on the initial As concentration. At a lower initial As concentration, the As mass in solution is relatively small to the amount of available sorption sites, so it was highly sorbed with available sorotion sites [33]. Importantly, if without any nWTR treatments, the soils with or without biosolids treatments had shown moderate affinity for As sorption. With the nWTR treatments, As sorption isotherms changed from S-type to H-type isotherms ( Figure 1). In this study, the H-type isotherm became dominant especially at a high nWTR application rate of 1%. Therefore, the quantity of As sorbed by the soil followed a descending order of: 1% nWTR-treated soil, > 0.5% nWTR-treated soil, > 0.25% nWTR-treated soil, and > control soil (or no nWTR treatment) ( Figure 1). The nWTR treatments have contributed many more sorption sites such as aluminum, iron, and calcium hydroxides in the soils, in addition to high organic matter contents and high cation exchange capacity of biosolids-amended soils ( Table 1).

Figure 1.
The amount of sorbed arsenic (As) in the calcareous soil amended with biosolids (3%, w/w) and with different levels of nWTRs of 0%, 0.25%, 0.5%, and 1% (w/w) at different equilibrium concentrations of As. Data shown are means and standard deviation (n = 3).The error bars were too small to be visible in the figure.
The amount of sorbed As in the biosolids-amended (3%) soil was 8000, 66,100, 68,200, and 76,700 mg As/kg or 8, 66.1, 68.2, and 76.7 mg As/g soil when the soil was treated with nWTRs of 0%, 0.25%, 0.5%, and 1% (w), respectively (Figure 1). High specific surface areas of nWTRs (125 m 2 /g) mainly containing Al-, Ca, and/or Fe (hydr) oxides significantly enhanced the anion sorption due to nanoscale nWTR particles and the amorphous nature of metal hydroxides [24]. Arsenic sorption on metal hydroxide surfaces replaces singly coordinated OH − groups, and then undergoes a re-arrangement into a more stable binuclear bridge-type bond between cations [34]. The results from our study (desorption rate was about 1%-5% at all treatments for 800 mg As(III)/L-treated soil) generally agree with those of Mahmoud et al. [35] and Elkhatib et al. [24], showing that nWTR-adsorbed As are stable or immobilized over a long time period [26,36]. The amount of sorbed As in the biosolids-amended (3%) soil was 8000, 66,100, 68,200, and 76,700 mg As/kg or 8, 66.1, 68.2, and 76.7 mg As/g soil when the soil was treated with nWTRs of 0%, 0.25%, 0.5%, and 1% (w), respectively (Figure 1). High specific surface areas of nWTRs (125 m 2 /g) mainly containing Al-, Ca, and/or Fe (hydr) oxides significantly enhanced the anion sorption due to nanoscale nWTR particles and the amorphous nature of metal hydroxides [24]. Arsenic sorption on metal hydroxide surfaces replaces singly coordinated OH − groups, and then undergoes a re-arrangement into a more stable binuclear bridge-type bond between cations [34]. The results from our study (desorption rate was about 1%-5% at all treatments for 800 mg As(III)/L-treated soil) generally agree with those of Mahmoud et al. [35] and Elkhatib et al. [24], showing that nWTR-adsorbed As are stable or immobilized over a long time period [26,36].
Increasing the treatment level of nWTRs from 0.25% to 1% also likely increased the efficiency of As retention in the soil (Figure 2). For examples, at 0.25% nWTRs application rate the removal efficiency significantly (p < 0.05) increased from 67.5% at As concentration 50 mg/L to 82.6% at 800 mg As/L-treated soil (Figure 2). In similar, at 0.50% nWTRs application rate the removal efficiency of As significantly (p < 0.05) increased from 58.8% at As concentration 50 mg/L to 85.3% at 800 mg As/L. In addition, at 1% nWTRs application rate the removal efficiency significantly (p < 0.05) increased from 95.6% at As concentration 50 mg/L to 95.6% at 800 mg As/L (Figure 2).
Appl. Sci. 2020, 10, x 6 of 13 Increasing the treatment level of nWTRs from 0.25% to 1% also likely increased the efficiency of As retention in the soil (Figure 2). For examples, at 0.25% nWTRs application rate the removal efficiency significantly (p < 0.05) increased from 67.5% at As concentration 50 mg/L to 82.6% at 800 mg As/L-treated soil (Figure 2). In similar, at 0.50% nWTRs application rate the removal efficiency of As significantly (p < 0.05) increased from 58.8% at As concentration 50 mg/L to 85.3% at 800 mg As/L. In addition, at 1% nWTRs application rate the removal efficiency significantly (p < 0.05) increased from 95.6% at As concentration 50 mg/L to 95.6% at 800 mg As/L (Figure 2).

Figure 2.
Arsenic sorption efficiency in the calcareous soil amended with biosolids (3%, w/w) and treated with different levels of nWTRs of 0%, 0.25%, 0.5%, and 1% (w/w). Data shown are means and standard deviation (n = 3). The error bars were too small to be visible in the figure.

Investigation of As Removal Mechanism
To investigate the role of nWTRs in As retention in biosolids-treated soil, a Fourier Transmission Infrared Spectroscopy (FTIR) analysis was performed ( Table 2).
The FTIR spectrum of nWTRs/biosolids-treated soil before and after As saturation are presented in Table 2. Table 2 showed the shifting and disappeared of some functional groups revealed that changes in sorbent properties occurred after As sorption. These shifts are a strong evidence for chemical complexation between As and functional groups of nWTRs and biosolids. Before As saturation (Table 2), most changes between 1700 and 500 cm −1 refers to complexation of As ions with COOH and O-H groups present in phenolic and carboxylic compounds present in biosolids and nWTRs [24,37]. In conclusion, the FTIR spectra of the biosolids-amended soil that was treated with different levels of nWTRs during a 60-day incubation period showed initial HOH stretching and bending, as well as a bending vibration of hydroxyl groups on metal oxides, and then some peaks completely disappeared or shifted, suggesting that those OH groups on the WTR particle surface reacted with As.  Figure 3 shows fractionation changes of As in biosolids-amended soil treated with different levels of nWTRs. The As fractionation result in the control soil was significantly different from those in the soils treated with biosolids and nWTRs. In the control soil, exchangeable As was 4.1%, carbonate-bound As was 22.8%, Fe-Mn oxides-bound As was 8.9%, organic matter-bound As was 8.1%, and residual As was 56.1%. In the biosolids-amended soil, exchangeable As was 8.1%, carbonate-bound As was 22.1%, Fe-Mn oxides-bound As was 12.5%, organic matter-bound As was 30.9%, and residual As was 26.5%. However, when the biosolids-amended soil was treated with nWTRs, As fractionation has been significantly affected (Figure 3). The nWTR treatment of 0.25% significantly increased the residual fraction of As (RS) up to 73.7%, while other As fractions decreased in comparison with the soils with or without biosolids treatments. Further, increasing the nWTR application rate from 0.5% to 1% drastically increased the stable and immobilized residual fraction (RS-As) of As from 82.5% to 87.7%, respectively (Figure 3), due its accumulation in silicates [28].  The relatively high percentage of carbonate-bound As in the soils was mainly due to high calcium carbonate content (326 g/kg) in the studied soil ( Table 1). The overall mobility and potential bioavailability of As in the studied soil amended with 3% biosolids and treated with different levels of nWTRs followed a descending order of: control soil without nWTR treatment, > 0.25% nWTR treated soil, > 0.5% nWTR-treated soil, and > 1% nWTR-treated soil (Figure 3). These results were generally in accordance with previous research findings [37][38][39], showing the effect of nWTR treatments on the bioavailability of some heavy metals. It was concluded that nWTRs can significantly reduce the bioavailability of As, Cr, and Hg in contaminated calcareous soil. The relatively high percentage of carbonate-bound As in the soils was mainly due to high calcium carbonate content (326 g/kg) in the studied soil ( Table 1). The overall mobility and potential bioavailability of As in the studied soil amended with 3% biosolids and treated with different levels of nWTRs followed a descending order of: control soil without nWTR treatment, > 0.25% nWTR treated soil, > 0.5% nWTR-treated soil, and > 1% nWTR-treated soil (Figure 3). These results were generally in accordance with previous research findings [37][38][39], showing the effect of nWTR treatments on the bioavailability of some heavy metals. It was concluded that nWTRs can significantly reduce the bioavailability of As, Cr, and Hg in contaminated calcareous soil. Table 3 shows the dominant chemical forms of As in the solution of soils with different treatments via the MINEQL+ modeling calculations that determine the soluble chemical As forms. The As speciation analysis showed that arsenous acid (H 3   The application of nWTRs to the biosolids-amended soil significantly affected As speciation results, depending on the application rate of nWTRs ( Table 3). The 0.25% nWTR treatment resulted in approximately 45% amorphous arsenic hydroxide or As (OH) 5 and 36% arsenous acid (As(III)) ( Table 3), and also increased the proportions of cacodylic acid and di-arsenic pentoxide. However, with a higher level of nWTR treatment (0.5%), amorphous arsenic hydroxide increased to 56% but arsenous acid decreased to 19.6%. Cacodylic acid and di-arsenic pentoxide also increased with increasing the level of nWTR treatment. The WTR application of 1% in the biosolids-amended soil increased the percentage of amorphous arsenic hydroxide, but arsenous acid percentage decreased (Table 3). Overall, the nWTR treatment was efficient in immobilizing As in the biosolids-amended soil. These research findings also support those reported previously by Elkhatib et al. [24] and Moharem et al. [40], showing reduced solubility of Cr and Hg in biosolids that were treated with Al 2 (SO 4 ) 3 or FeCl 3 compounds that are commonly used in drinking water treatment. Different As species have different physiochemical properties that determine their relative tendency to precipitate, adsorb, or dissolve in soil solution. Thus, evaluating nWTR effects on As speciation in agricultural soil is essential for developing the best environmental management practices with As-contaminated soil and water. Choi et al. [41] reported that over 80% of As(V) and Zn(II) were steadily removed at solution pH 6 by the hybrid ion-exchange fabric/ceramic membrane system. Reduced flow rate and removal capacity were recovered through a backwashing process during continuous treatment with the hybrid ion-exchange fabric/ceramic membrane system. Hwang et al. [42], in their study about how arsenic leaches from different soil and sediment types and responds to hydrologic conditions to identify areas susceptible to arsenic contamination, revealed that most arsenic-loaded sediments accumulated in the vicinity of a reservoir inlet, and arsenic in reservoir beds has a higher leaching potential than that from agricultural land and stream beds. Arsenic deposited at the bottom of reservoirs had higher mobility than that in the other soils and sediments, and arsenic leaching was closely associated with the acidity of water.

Effect of nWTRs on As Speciation in Biosolids-Amended Soil
Arsenic speciation analyses in the soils were also directly measured using XANES, a non-destructive analytical method (Table 4 and Figure 4) that determine the total chemical forms of As in soil sample. The XANES of control soil sample show that the dominant As chemical form was sodium arsenate of 74.8%, followed by arsenic pentoxide of 25.2%. Moreover, The XANES of original nWTRs show that the dominant As chemical form was sodium arsenate of 28.8%, followed by cacodylic acid of 14.5%, and arsenic pentoxide (56.7%). On the contrary, in the biosolids-amended soil, arsenic pentoxide was 61.5%, sodium arsenate was 22.4%, cacodylic acid was 14.9%, and arsenic sulfide was 1.2% (Table 4). Similarly, the application of 1% nWTRs to the biosolids-amended soil changed the percentage of As species such as arsenic pentoxide was 68.0%, sodium arsenate was 10.0%, and cacodylic acid was 22.0% (Table 4). Wang et al. [43] indicated that the XANES is efficient in determining arsenic oxidation states of the sediments with low arsenic contents and multiple arsenic species. Table 4. Speciation fraction (%) of arsenic (As) in the soils that were treated with biosolids (3%) and nanoscale WTRs (1%) using X-ray absorption near-edge structure spectroscopy (XANES) analysis.

As Compounds Control Soil
Original nWTRs

Conclusions
The initial As concentration significantly affected the As adsorption to biosolids and nWTRs-soil mix. Increasing the application rate of nWTRs increased the efficiency of As retention in soil. The overall mobility and potential bioavailability of As in the studied soil treated with biosolids

Conclusions
The initial As concentration significantly affected the As adsorption to biosolids and nWTRs-soil mix. Increasing the application rate of nWTRs increased the efficiency of As retention in soil. The overall mobility and potential bioavailability of As in the studied soil treated with biosolids and/or nWTRs follows the following order: biosolids amended soil, > nWTRs-biosolids amended soil. Moreover, application of nWTRs affected the soil As speciation depending on the application rates of biosolids and nWTRs. It has been found that the dominant chemical form of As was in an oxidized form, nominally As 5+ . It was likely to be incorporated in arsenic pentoxide, and this phase generally retained As in a state of low mobility, toxicity, and bioavailability. The high capability of nWTRs to immobilize As in biosolids-treated soil and the stability of As-nWTRs surface complexes suggest the strong affinity of nWTRs as economic and effective sorbents for removal of As from contaminated sites. In conclusion, nano-scale particles of drinking water treatment have proven to be effective in As removal from biosolids-treated soil.