Estimating Arsenic Mobility and Phytotoxicity Using Two Different Phosphorous Fertilizer Release Rates in Soil

Deficiencies in phosphorus (P), an essential factor for plant growth and aided phytostabilization, are commonly observed in soil, especially near mining areas. The objective of this study was to compare the effect of P-based fertilizer types on arsenic (As) extractability and phytotoxicity in As-contaminated soil after stabilizer treatment. Different treatments with respect to the P-releasing characteristics were applied to soil to determine As mobility and phytotoxicity in P-based fertilizers, with bone meal as a slow-releasing P fertilizer and fused superphosphate as a fast-releasing P fertilizer. In addition, P fertilizers were used to enhance plant growth, and two types of iron (Fe)-based stabilizers (steel slang and acid mine drainage sludge) were also used to reduce As mobility in As-contaminated soil under lab-scale conditions. A water-soluble extraction was conducted to determine As and P extractability. A phytotoxicity test using bok choy (Brassica campestris L. ssp. chinensis Jusl.) was performed to assess the elongation and accumulation of As and P. Within a single treatment, the As stabilization was higher in steel slag (84%) than in acid mine drainage sludge (27%), and the P supply effect was higher in fused superphosphate (24740%) than in bone meal (160%) compared to the control. However, a large dose of fused superphosphate (2%) increased not only the water-soluble P, but also the water-soluble As, and consequently, increased As uptake by bok choy roots, leading to phytotoxicity. In combined treatments, the tendency towards change was similar to that of the single treatment, but the degree of change was decreased compared to the single treatment, thereby decreasing the risk of phytotoxicity. In particular, the toxicity observed in the fused superphosphate treatments did not appear in the bone meal treatment, but rather the growth enhancement effect appeared. These results indicate that the simultaneous application of bone meal and stabilizers might be proposed and could effectively increase plant growth via the stabilization of As and supplementation with P over the long term.


Introduction
Arsenic (As) contamination in soil by mining activity has been increasing continuously [1][2][3].High concentrations of As in soil are of particular concern because As is phytotoxic and accumulates in living organisms, including humans [4].Remediation technology has been developed so that it is possible to prevent the extent of soil As contamination through actual management practices, namely, the limitation of its mobility, capping, replacement, solidification, acid or alkaline wash out, addition of electrolytes, and stabilization [5,6].Of these various techniques, chemical stabilization has the advantages of high efficiency, low cost, and minimal environmental disturbance when used for remediation.Many recent related studies have been performed to account for the covalent bonding interactions responsible for As chemical stabilization using magnetite, zero-valent iron, limestone, red mud, and furnace slang [7][8][9][10].
The mechanism of As stabilization occurs through both adsorption and specific components in soil, which is observed in a straightforward case involving clay, organic matter, and Al/Fe/Mn/Ca carbonate oxide or hydroxide surfaces [11,12].The typical Fe oxides in As-contaminated soils are present in diverse functional formations, namely, goethite (α-FeOOH), hematite, (α-Fe 2 O 3 ), lepidocrocite (γ-FeOOH) and ferrihydrite (5Fe 2 O 3 •9H 2 O or Fe 5 HO 8 •4H 2 O) [12].Combinatory FeSO 4 applications to various plants led to decreased As in terms of both availability and uptake [13].Liang and Zhao [9] reported that As leaching was reduced by magnetite in As-spiked soil, as indicated by a batch test and a column test.Zero valent iron (ZVI), which acted as a precursor of Fe oxides, has been used for As stabilization in soil.After ZVI is applied to the soil, it produces amorphous and poorly crystalline Fe oxides followed by oxidation reactions [14].Based on a former process related to the oxidation of ZVI, Kim et al. [7] showed a 52% reduction in extractable As during the toxicity characteristic leaching procedures (TCLP) from mine tailings (1638 mg kg −1 of As) treated with ZVI.However, the fraction of amorphous and poorly crystalline hydrous oxides, including other iron compounds, was increased in the samples via sequential extraction method.Additionally, Koo et al. [15] found a decrease in As availability and an increase in microbial activity (i.e., in dehydrogenases and ß-glucosidases) when using various Fe compounds, including ZVI in mine tailings (1357-1496 mg kg −1 of As).Several combinatory treatments have been performed on samples of industrial wastes containing Fe compounds such as by-products of red mud and furnace slag from alumina smelters and steelworks [15,16].Lee et al. [8] observed reduced extractable As and a bioaccessible As fraction.In Korea, many researchers have used Fe-rich acid mine drainage sludge (AMDS) as Fe oxides and an Fe-based As stabilizer, which is also industrial waste from a mine drainage treatment facility used to immobilize trace elements in water or soil [17][18][19][20][21].
Gentle soil remediation options (GROs) are risk management strategies and technologies intended for the restoration of metalliferous mine waste-contaminated areas, and these methods have been supported by various scientific networks, including the International Phytotechnology Society, the Society of Environment Toxicology and Chemistry (SETAC), and the Society for Ecological Restoration (SER).GROs are known to be less invasive to soil structure, functions, and ecosystem services, and their ecological footprint is much lower than that of conventional soil remediation techniques [22,23].In particular, GROs are based on low-cost and sustainable tools, such as phytotechnologies and ecological restorations using plants for widespread contamination and/or the (phyto) management of brown fields [23][24][25][26].However, high concentrations of heavy metals and low concentrations of macronutrients (N and P) are the main limiting factors inhibiting plant growth in metalliferous mine wastes.In particular, deficiency in P is a very important issue pertaining to revegetation because available P easily forms insoluble, phosphate-heavy metal complexes in metalliferous mine waste areas that lack organic matter and clay [27].In alkali soils, the available P is easily bound to calcium and precipitates with the octocalcium phosphate or hydroxyapatiteforms [28], while the available P is not easily absorbed by the plant due to its complexation with iron and/or aluminum oxides in so-called wrap-occluded phosphorus within acid soils [29].Thus, treatment with P fertilizer has been added to the GROs as an indispensable option.
Treatment with P fertilizer enhanced the revegetation and early growth of plants after chemical stabilization to remediate the environment completely [30,31].The chemical fate and structure of both P and As are similar in highly As-contaminated soil, and P treatment resulted in an increase in the rates of As uptake by plants with increased As availability.Species-specific results are also available to consider the anomalous finding in which P absorption suppressed As uptake in Chinese brake (Pteris vittata L.) [32][33][34].The incorporation of Fe and P into As-spiked artificial soil increased root growth in lettuce (Lactuca sativa L.) but decreased As uptake, which enhanced the early-stage growth of plants in As-contaminated soil.The experiments were conducted without considering the dependence of pure As, P, and Fe reagents on the environmental conditions under the limited concentration treatment of 75-276 mg As kg −1 in an artificial soil [31].Kim et al. [35] showed that highly As-contaminated soil had potential correlations with the independent variables of P and Fe at low concentrations of 0.015%-0.053%P and 0.017%-0.095%Fe.
Practical approaches are key to developing an As stabilization methodology using current agricultural practices.This study compared the effect of P-based fertilizer types on As extractability and phytotoxicity in As-contaminated soil following stabilizer treatment to support GRO practices.Fe-based As stabilizers in steel slag (SS) and AMDS soil were evaluated according to their plant growth enhancement depending on different P dissolution rates with bone meal (BM) and fused superphosphate (FS).

Experimental Set-Up
As-contaminated soil was collected from the Gangwon (GW) mining area (37 • 19 19.6 N, 128 • 48 47.4 E) in Jeongseon-gun, Gangwon Province, Korea.The obtained soil sample was air-dried and passed through a 2 mm sieve.Stabilizers with SS and AMDS were chosen for this study.The SS and AMDS were obtained from a steel plant (Taeseo Industry, Taebaek-si, Korea) and the sludge was taken from an acid mine drainage (AMD) treatment facility at the Hamtae mine in Gangwon Province, Korea.Two types of P fertilizer were also studied, namely, bone meal (BM) (Kyoung Lim Co., Yeongcheon-si, Korea) and fused superphosphate (FS) (KG Chemical Co., Sungnam-si, Korea).The total P contents of the two fertilizers are 19.6% and 20.1%, respectively.All the amendments were sieved to smaller than 0.5 mm before adding them to the soil.The amendments were applied to each soil in various combinations at a 2% w/w ratio (Table 1).For example, SS 2 indicates that 2% of the SS was used as treatment and BM 2 indicates that 2% of the BM was used, while SS 1 BM 1 indicates that both 1% of SS and 1% of BM were used as a treatment.The amount of P fertilizer, 1-2% was exceeded in actual farming practice, but the amount of P was determined to equalize the total doses of amendments when co-treated with As stabilizers.The soil and amendments were mixed thoroughly for homogeneity; the samples were equilibrated for 4 weeks while the soil moisture was maintained at approximately 60% of the soil water holding capacity (WHC).

Soil Analysis
The soil pH and electrical conductivity (EC) were determined in a 1:5 soil:water suspension with a combination pH-EC meter (Thermo Orion 920A, Thermo Fisher Scientific, Waltham, MA, USA).The weight loss-on-ignition (LOI) was detected to determine the water content at 120 • C and the carbon content at 550 • C over 24 h [36].Total P was determined by HClO 4 digestion, and P in filtrates was determined using an ultraviolet (UV) spectrophotometer (UV-1650PC, Shimadzu, Kyoto Japan) via modified molybdenum blue method [37].The oxalate-extractable Al, Fe, and Mn concentrations were determined by ammonium oxalate extraction methods [38] and the samples were analyzed using an inductively coupled plasma optical emission spectrometer (ICP-OES, 730 Series, Agilent, Santa Clara, CA, USA).The total trace element (As, Cd, Cu, Pb, and Zn) concentrations were determined by digesting the samples with aqua regia, a mixture of HNO3/HCl (v:v = 1:3), in accordance with ISO 11466 [39], and they were analyzed using an ICP-OES.The accuracy of the analytical data with regards to the trace elements was assessed using certified reference material (NIST 2711a, Montana II Soil, NIST, Gaithersburg, MD, USA).The detection limits of As, Cd, Pb, and Zn were 0.001, 0.001, 0.001, and 0.003 mg L −1 , respectively.
The effects of the stabilizers and fertilizers on the As and P extractability were evaluated according to the content of deionized water (DW)-soluble forms.For water soluble-As and P, 1 g soil was added to 20 mL DW in 50-mL polypropylene centrifuge tubes, which were then shaken for 1 h on a wrist-action shaker, centrifuged for 20 min at 10,000 g (Relative Centrifugal Force, RCF), and then filtered through a 0.45 µm filter [40].All the remaining As and P contents in the filtrates were determined using ICP-OES after acidification [41].

Phytotoxicity Test with Bok Choy
A growth test on bok choy (Brassica campestris L. ssp.chinensis Jusl.) was used as a phytotoxicity test.Twelve seeds were placed in 60 × 15 mm plastic petri dishes containing 35 g of treated soil after 4 weeks of aging and were subsequently cultivated for 3 weeks.The moisture content was maintained at approximately 60% of the soil WHC.Three weeks after sowing, the plants were harvested and washed with distilled water and their elongation was immediately measured using an image analyzer program.The elongation of the bok choy roots and shoots was calculated for each seedling relative to the germination success rate (80%) in the control group and expressed in cm seedling −1 .After this determination, the plants were separated into roots and shoots and dried in an oven at 70 • C. The dried samples were digested with HNO 3 and H 2 O 2 by the hot-block digestion procedure at 120 • C and immediately filtered.The filtrates were used to determine the As concentrations in an ICE-OES.The P concentrations of the filtrates were determined using an ultraviolet (UV) spectrophotometer (Shimadzu, UV-1650PC; Shimadzu, Kyoto Japan) via a modified molybdenum blue method.The accuracy of the As measurement was assessed using a certified reference material (BCR (Community Bureau of Reference) -402, white clover).

Statistical Analysis
All the measurements were performed in triplicate.One-way analysis of variance (one-way ANOVA) tests were used to compare the means of different treatments.When significant p-values (p < 0.05) were obtained, the differences between the means were evaluating using Duncan's test.The relationships among the experimental results were evaluated using Pearson's correlation analysis.The data were analyzed using an SAS program (SAS 9.4, SAS Institute Inc., Cary, NC, USA).

Basic Properties of Soil and Amendsments
The concentration of total As in the soil was much higher (1853 mg kg −1 ) than the threshold measurements listed in Korean soil regulations (25, 50, and 200 mg kg −1 for Areas I, II, and III, respectively), which makes it ~9 times more contaminated with As than Area III (Table 2).Other trace elements were also detected, but their levels were not as high as the As contamination, as indicated by the regulatory limits.Although trace elements were detected in the amendments, they were used in this experiment to account for the low treatment dose (2%), the low heavy metal extractable properties, and the high As adsorption capacity from the previous study [18,35] (Table 3).   .c Loss on ignition (%).d Total trace element concentration (mg kg −1 ).e Below detection limit.

Effects of the Amendments on Soil Chemical Characteristics
The initial soil pH was 8.21, indicating an alkaline soil.The amendments changed the pH and the EC of the soil (Figure 1).In a single treatment, only AS significantly increased the soil pH from 8.21 to 8.53 (AS 2 ).In combined treatments, a significant increase in the soil pH was observed only in AS 1 BM 1 , from 8.21 to 8.54, indicating that the ability to increase the soil pH was highest for AS, although the pH of SS (12.4) was higher than that of AS (8.4).The soil EC was greatly increased in the FS 2 , SS 1 FS 1 , and AS 1 FS 1 treatments, in which FS fertilizer was applied in single or combined treatments.This result was attributed to the high EC (60.3 ds m −1 ) of the FS and the value of the EC, which was greater than 4 ds m −1 (FS 2 , SS 1 FS 1 , and AS 1 FS 1 ), after treatment, which indicated the potential for plant growth inhibition by the water potential or salt stress [42].Therefore, it seemed that the osmotic stress due to high salinity, as well as the toxicity of As in FS, treatments might have affected the plant growth and toxicity results.
An investigation was conducted to evaluate the effects of the amendments on the extractability of As and P according to the water-soluble As (WS-As) and water-soluble P (WS-P) (Figure 1).Both the SS and AS stabilizers decreased As mobility in the soil from 6.94 to 1.11 and 5.09 mg kg −1 , respectively.SS is rich in Fe and Ca oxides with reactive surface sites that can bind As, and it caused a decrease in WS-As.AS, which also contains a large amount of Fe oxides, caused efficient As stabilization [18,35].
The difference in the WS-As result between BM and FS was also found in the WS-P.P application significantly increased the WS-P in treatments FS2, SS1FS1, and AS1FS1 from 5 to 1237 mg kg −1 , 315 mg kg −1 , and 327 mg kg −1 , respectively, in which FS fertilizer was applied in single or combined treatments.There are two possible reasons for the relatively low change in P and As mobility for BM compared to FS.First, the P supply rate of BM is slow, so there may not have been enough time (4 weeks) to elute the phosphorus from the BM.Second, since the rate of P elution is slow and the amount of eluted P is low, the slowly supplied P might be adsorbed to various adsorption sites on the soil, SS, and AS, rather than desorbing the As from the soil.
Various types of P fertilizer, including FS, have been used as stabilizers in lead (Pb)-contaminated soil via the formation of pyromorphite [43][44][45].By contrast, because of the similar structure and behavior of As and P in soil, these elements compete for adsorption sites, resulting in increased As mobility when P fertilizer is applied to As-contaminated soil [34].Cao and Ma [32] reported that P amendments significantly increased the water-soluble As and the As uptake by Pteris vittata L. Kwak et al. [46] also reported that the As uptake by Pteris vittata L. was increased when phosphate was applied to As-spiked soil.Therefore, FS significantly increased the WS-As, from 6.97 to 20.07 mg kg −1 (FS 2 ).Among the combined treatments, only FS increased the WS-As from 6.94 to 12.10 mg kg −1 (SS 1 FS 1 ) and 16.65 mg kg −1 (AS 1 FS 1 ).It is well known that the P-releasing ability of FS, based on inorganic P, is high [47,48], resulting in the increases in WS-As during the 4 weeks of aging time.However, BM, a slow-releasing fertilizer based on organic P, could not increase the WS-As significantly in either the single or combined treatments (BM 2 , SS 1 BM 1 , and AS 1 BM 1 ).
The difference in the WS-As result between BM and FS was also found in the WS-P.P application significantly increased the WS-P in treatments FS 2 , SS 1 FS 1 , and AS 1 FS 1 from 5 to 1237 mg kg −1 , 315 mg kg −1 , and 327 mg kg −1 , respectively, in which FS fertilizer was applied in single or combined treatments.There are two possible reasons for the relatively low change in P and As mobility for BM compared to FS.First, the P supply rate of BM is slow, so there may not have been enough time (4 weeks) to elute the phosphorus from the BM.Second, since the rate of P elution is slow and the amount of eluted P is low, the slowly supplied P might be adsorbed to various adsorption sites on the soil, SS, and AS, rather than desorbing the As from the soil.Through these results, we could reconfirm that the efficiency of the P supply was higher for FS than BM.However, the results also implied the possibility of spreading soil contamination due to the increase in As mobility when using FS and the need for a follow-up study on the longevity and capacity of BM for a P supply through long-term experimentation over 4 weeks.

Elongation of Bok Choy Roots and Shoots
According to the previous results, the P fertilizers increased both P and As availability in the soil, which might be helpful or harmful for plant growth by affecting the nutrient or toxic material availability.Figure 2 shows the results of the root and shoot elongation used to investigate the effects of the amendments on the growth of bok choy.For bok choy shoots, significant differences were observed in some treatments; however, it was difficult to detect definitive differences in the shoots among the treatments because the soil As did not affect the shoots as greatly as the roots [31,35].
Agronomy 2019, 9, x FOR PEER REVIEW 7 of 13 increase in As mobility when using FS and the need for a follow-up study on the longevity and capacity of BM for a P supply through long-term experimentation over 4 weeks.

Elongation of Bok Choy Roots and Shoots
According to the previous results, the P fertilizers increased both P and As availability in the soil, which might be helpful or harmful for plant growth by affecting the nutrient or toxic material availability.Figure 2 shows the results of the root and shoot elongation used to investigate the effects of the amendments on the growth of bok choy.For bok choy shoots, significant differences were observed in some treatments; however, it was difficult to detect definitive differences in the shoots among the treatments because the soil As did not affect the shoots as greatly as the roots [31,35].1; SS2, steel slag (SS) 2%; AS2, acid mine drainage sludge 2%; BM2, bone meal (BM) 2%, FS2, fused superphosphate (FS) 2%; SS1BM1, SS 1% + BM 1%; SS1FS1, SS 1% + FS 1%; AS1BM1, AMDS 1% + BM 1%; and AS1FS1, AMDS 1% + FS 1%.
Treatment with AS alone (AS2) did not significantly affect root growth of bok choy, but SS (SS2) enhanced the root growth in the soil compared to the control.The phytotoxicity of As seemed to be decreased by As stabilization via As adsorption by the Fe oxides in SS, and similar results occurred in other studies [49].In general, the As stabilization increased with increases in the Fe dose; however, Mench et al. [14] reported that the soil physics and structure worsened when the amount of ZVI was over 5% by weight, causing cementation and changes in porosity.Notably, we did not find any similar phenomena in this study (maximum of 2% for SS and AS).
For the treatments of FS2, SS1FS1, and AS1FS1, when the WS-As and WS-P exhibited high concentrations (Figure 1), the root growth was significantly decreased compared to the control.This result demonstrated that the FS treatment increased the WS-As, leading to phytotoxicity, and it was greater than the increase in WS-P, leading to enhanced plant growth.By contrast, treating with BM, a slow-release fertilizer, either alone or in combination with other stabilizers (BM2 and SS1BM1), caused no significant differences in root growth, except for AS1BM1.
In this study, it seemed that water stress and oxidative stress were the main reasons for root growth inhibition.In the treatments of FS2, SS1FS1, and AS1FS1 root growth inhibition was observed and the samples had high EC values (6.41-7.76ds m −1 ) and WS-As concentrations (12.10-20.07mg kg −1 ).Similarly, according to Bernstein [42], plant growth was inhibited in the 4-8 ds m −1 range for the EC.An increased salt concentration might increase the EC and decrease the osmotic potential,  Treatment with AS alone (AS 2 ) did not significantly affect root growth of bok choy, but SS (SS 2 ) enhanced the root growth in the soil compared to the control.The phytotoxicity of As seemed to be decreased by As stabilization via As adsorption by the Fe oxides in SS, and similar results occurred in other studies [49].In general, the As stabilization increased with increases in the Fe dose; however, Mench et al. [14] reported that the soil physics and structure worsened when the amount of ZVI was over 5% by weight, causing cementation and changes in porosity.Notably, we did not find any similar phenomena in this study (maximum of 2% for SS and AS).
For the treatments of FS 2 , SS 1 FS 1 , and AS 1 FS 1 , when the WS-As and WS-P exhibited high concentrations (Figure 1), the root growth was significantly decreased compared to the control.This result demonstrated that the FS treatment increased the WS-As, leading to phytotoxicity, and it was greater than the increase in WS-P, leading to enhanced plant growth.By contrast, treating with BM, a slow-release fertilizer, either alone or in combination with other stabilizers (BM 2 and SS 1 BM 1 ), caused no significant differences in root growth, except for AS 1 BM 1 .
In this study, it seemed that water stress and oxidative stress were the main reasons for root growth inhibition.In the treatments of FS 2 , SS 1 FS 1 , and AS 1 FS 1 root growth inhibition was observed and the samples had high EC values (6.41-7.76ds m −1 ) and WS-As concentrations (12.10-20.07mg kg −1 ).Similarly, according to Bernstein [42], plant growth was inhibited in the 4-8 ds m −1 range for the EC.An increased salt concentration might increase the EC and decrease the osmotic potential, causing an increase in the osmotic pressure, resulting in a reduction of the available water content, causing plant water stress [50].Furthermore, a high salt concentration might produce oxygen species, which would disturb the cell membrane and cause cell death [51].Another possible reason is that As that was absorbed into the plant root might cause inhibited root elongation via the production of reactive oxygen species (ROS), including O 2− , OH•, and H 2 O 2 , which could disrupt cellular activity via lipid peroxidation or interrupt plant cell metabolism [52].
With combined treatments of a stabilizer and fertilizer, bok choy root growth mainly depended on the type of fertilizer rather than the type of the stabilizer in As-contaminated soil.Compared to FS, which supplied P rapidly, BM with P in the organic form did not show any growth inhibition in its roots (negative effect); specifically, the root growth was enhanced by an increase in the WS-P in the soil.However, the result showing that the root growth was highest in the SS-alone treatment (SS 2 ) indicated that the effect of the reduced As phytotoxicity, due to the difference in the single stabilizer dose (2%) compared to the combined treatment (1%), was stronger than the effect of the root growth enhancement by a P supplement.However, Kim et al. [35] reported that the effect of the P supply on root elongation of lettuce was significantly greater than that of Fe under a second-order central composite rotation design (CCRD) based on weight levels (g kg −1 ).Therefore, to determine the main factor, we need to evaluate the amounts of As and P absorbed by plants.

As and P Uptakes by Bok Choyelongation of Bok Choy Roots and Shoots
Figure 3 shows the As and P concentrations in bok choy grown for 3 weeks.The elemental concentrations measured in the plant roots and shoots represent both the bioavailability and absorption, translocation, and accumulation abilities of plants for trace elements [53,54].As a result of the high concentration of As in the soil, As uptake by bok choy was higher.For bok choy shoots, both the As and P concentrations and the changes in the As and P uptake trends were similar to the results in the roots, but the tendency was lower than it was in the roots.Above all, the shoots are located far from the root so the shoots were not exposed directly to the trace elements; hence, a lower accumulation of trace elements occurred [59].When observed closely, unlike for P, bok choy showed that As was mainly absorbed and allocated in the root rather than the shoot because As has low mobility for translocation from the root to shoot [60].Under aerobic soil conditions, in non-hyperaccumulators such as bok choy, arsenate (V) that is absorbed into the root cells via phosphate transporters is often reduced to arsenite (III) and is consequently sequestrated into the vacuole in the cytoplasm rather than being transferred to the aboveground part of the plant via xylem in the form of arsenite [60].This result indicated that bok choy is a poor translocator of As  The As accumulation in the roots was significantly decreased by SS 2 and AS 2, from 300 mg kg −1 in the control to 61 and 99 mg kg −1 , respectively.This result is similar to the results of Gutierrez et al. [55], who used steel-making slag (SMS) in As-contaminated soil and observed that the As uptake was decreased by 50% in the radish roots.However, when SS was used with fertilizers (SS 1 BM 1 and SS 1 FS 1 ), the As uptake was increased from 61 to 110 and 146 mg kg −1 , respectively, due to the increase in WS-As from 1.11 to 2.81 mg kg −1 nd 12.10 mg kg −1 , respectively (Figure 1).These trends were also similar under the AS treatments.
Taken together, the FS fast releasing fertilizer increased both WS-As and As uptake, but the increase in As uptake (1.12 times greater) was not as high as the increase in WS-As uptake (2.89 times greater) in FS 2 .And the anomalous result showing that the As uptake was not as high in the FS treatment was related to the high concentration of P taken up by the roots.P was absorbed through the P transporters in the cell membrane, and As also used the same pathway to absorb into the roots, indicating that P and As compete for a common absorption pathway [56][57][58].For example, in soil treated with FS alone (FS 2 ), WS-As increased from 6.94 to 20.07 mg kg −1 (2.9-fold), but WS-P greatly increased, from 5 to 1237 mg kg −1 (247-fold), so the large amount of P suppressed As absorption.Koo et al. [31] and Kim et al. [35] reported that the P and As concentrations in lettuce roots were negatively correlated and that the As uptake was decreased by P uptake.In the combined treatments (SS 1 BM 1 , SS 1 FS 1 , AS 1 BM 1 , and AS 1 FS 1 ), when the P concentration in the roots was high, the trend in As absorption decreased in the soil.
For bok choy shoots, both the As and P concentrations and the changes in the As and P uptake trends were similar to the results in the roots, but the tendency was lower than it was in the roots.Above all, the shoots are located far from the root so the shoots were not exposed directly to the trace elements; hence, a lower accumulation of trace elements occurred [59].When observed closely, unlike for P, bok choy showed that As was mainly absorbed and allocated in the root rather than the shoot because As has low mobility for translocation from the root to shoot [60].Under aerobic soil conditions, in non-hyperaccumulators such as bok choy, arsenate(V) that is absorbed into the root cells via phosphate transporters is often reduced to arsenite(III) and is consequently sequestrated into the vacuole in the cytoplasm rather than being transferred to the aboveground part of the plant via xylem in the form of arsenite [60].This result indicated that bok choy is a poor translocator of As to aboveground plant tissues and is characterized as a "root trap" for As [61].On the other hand, P, which is a major macronutrient, is distributed evenly in the root and the above shoot without any special sequestration or defense mechanisms.
In addition, when P and As absorbed by plant roots were calculated based on equivalents, it was confirmed that absorbed P was 22-62 mmol kg −1 , while As was 0.8-4.8mmol kg −1 , indicating 5 to 30 times less As molecules being absorbed than P. In the case of shoots, P was absorbed by 19-49 mmol kg −1 , while As was 0.02-0.18mmol kg −1 , indicating that 200-2000 times less As molecules were absorbed than P.These equivalent results revealed that P absorbed by roots was smoothly allocated or redistributed to shoots.And even if As was introduced through P-transport proteins, the proportion and amount of As was very small.

Correlation Analysis
The relationships among the results from this study were analyzed (Table 4).However, in this study, no significant correlations were observed in the growth of the shoots and the As/P uptake by the shoots, because the roots were more sensitive to As than the shoots [62].At first, a highly positive relationship between WS-As and WS-P (r = 0.864) was found.The increases in WS-P inhibited bok choy root growth by increasing the WS-As rather than supplying the P nutrient, resulting in a negative relationship between root elongation and both WS-P (r = −0.830)and WS-As (r = −0.942).This result conflicted with the result of Koo et al. [31].These conflicting results could have occurred because the previous study used artificial soil, which mimics mine tailings that rarely contain nutrients, and added P could serve as a nutrient.However, in this study, the soil contained appreciable P, and additional P was applied.The RE had a negative relationship with the P concentration in the roots (r = −0.780),which could indicate that the increase in P content did not inhibit root growth, but the effect of the increase in WS-As that inhibited the root growth was much greater than the effect of the increase in the WS-P, which increased the nutrient supply.These results showed that if the P supply was above a certain level, the increased phytotoxicity of As might overwhelm the effect of the nutrient supply.The increasing or decreasing trends in the WS-As and WS-P due to the amendments did not affect As and P uptake by the roots.This result could be explained because the WS-P concentrations were similar (1-5 mg kg −1 ) in the treatments without added P (control, SS 2 and AS 2 ) and the treatments containing both stabilizer and BM (SS 1 BM 1 and AS 1 BM 1 ).Nevertheless, because the amount of P uptake by the roots was extremely high in the BM and FS treatments, a high correlation coefficient was not observed after statistical analysis, indicating that the water-soluble assessment has some limits in estimating or predicting the real amount of uptake by the plant.

Conclusions
The present study was conducted to investigate the effects of various amendments, including stabilizers and fertilizers, on As mobility and, and to compare the effects of both P fertilizers.The stabilization of As by SS was higher than that of AMDS, and the solubility of the P supply was higher with FS than with BM.However, a large dose of P over a short time by FS increased not only the WS-P, but also the WS-As, and consequently, the As uptake by the roots, causing phytotoxicity.Although the FS was treated with stabilizers simultaneously, phytotoxicity symptoms were also observed.By contrast, BM showed positive effects of phytotoxicity reduction and growth enhancement when treated with stabilizers.In addition, some data showed that the results of the chemical and biological assessments did not coincide, suggesting that an equivalent molecular concept would be needed for further study on As, Fe, and P. Nevertheless, the present study confirmed the applicability of combining chemical stabilizers with P fertilizers for the amelioration of As-contaminated soil.These results indicated that for the successful gentle remediation of agricultural soil or abandoned mining areas, the simultaneous application of stabilizers and slow-releasing P fertilizer might be proposed and could effectively increase crop productivity and the early growth of plants via the stabilization of As and supplementation with P over the long term.

Table 2 .
Basic properties and trace element concentrations of the soil.
a Electrical conductivity.b Loss on ignition.c Total phosphorus concentration.d Oxalate-extractable metal concentration.e Total trace element concentration.

Table 3 .
Basic properties and trace element concentrations of the amendments.