Release of cadmium in contaminated paddy soil amended with NPK fertilizer and lime under water management

Ecotoxicology and Environmental Safety 159 (2018) 38–45

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv

Release of cadmium in contaminated paddy soil amended with NPK fertilizer and lime under water management

T

⁎

Xiao-qing Han, Xi-yuan Xiao , Zhao-hui Guo, Ye-hua Xie, Hui-wen Zhu, Chi Peng, Yu-qin Liang School of Metallurgy and Environment, Central South University, Changsha 410083, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Soil moisture Agrochemicals Lime Cadmium Mobility

Agricultural soils contaminated with cadmium (Cd) pose a risk to receiving surface water via drainage or runoff. A 90-day laboratory incubation experiment was conducted to investigate the release characteristics and transformation of Cd from contaminated paddy soil amended with agrochemical (NPK fertilizer) and lime (L) under water management regimes of continuous flooding (F) and drying-wetting cycles (DW). The result showed that the dissolved Cd concentrations in overlying water of the fertilizer treatment under flooding (NPK+F) and drying-wetting (NPK+DW) reached up to 81.0 μg/L and 276 μg/L, and were much higher than that from the corresponding controls without NPK fertilizer addition at the end of experiment. The Cd concentration showed significantly negative correlation with overlying water pH, but positive correlation with soil redox potential and concentrations of dissolved total nitrogen, sulfate and manganese in overlying water (P < 0.05), indicating that drying-wetting cycles and N fertilizer addition may enhance soil Cd release. The Cd concentrations in overlying water from all treatments except NPK+L+F treatment exceeded the Cd threshold limit of Chinese Environmental Quality Standards for Surface Water (10 μg/L Grade V) and poses potential risk to surface water quality. Meanwhile, the proportion of Cd in the acid-soluble fraction from all incubated soil except NPK+L+F treatment increased compared to before incubation. The results indicated that continuous flooding was a reasonable water management candidate coupled with lime addition for immobilizing soil Cd.

1. Introduction Cadmium (Cd), a potential toxic element, has long been identified as major human health hazard (Ye et al., 2014; Xue et al., 2017). The main anthropogenic sources of Cd are activities such as non-ferrous metal mining and refining, but atmospheric sedimentation, wastewater irrigation, and application of chemical fertilizers and pesticides also result in high level of Cd accumulating in soils (Janoš et al., 2010; Liu et al., 2017). Agricultural soils contaminated with Cd may pose a risk to human health via the food chain due to its high mobility and toxicity (Zhu et al., 2014), and may also lead to contamination of surface water and groundwater through runoff and infiltration (Schipper et al., 2008; Wang et al., 2011). Unfortunately, soil Cd contamination occurs widely in paddy fields of subtropical China, especially in Hunan province (Zeng et al., 2015). This problem has recently received considerable attention because the increasing input flux of Cd poses a major threat to food safety and downstream surface water quality (Zhang et al., 2015). Cadmium availability and mobility in soil is influenced by many factors including water management, pH, redox potential (Eh), agricultural measures (Kashem and Singh, 2002, 2004; Ye et al., 2018). Soil

⁎

moisture regime is one of the most important factors affecting the physical, chemical, and biological properties of soil and may indirectly influence Cd transformation (Zheng and Zhang, 2011; Li and Xu, 2017a). For instance, the availability of Cd in soils decreased and redistributed from exchangeable fraction to iron-manganese (Fe-Mn) oxide-bound fraction after submergence due to its adsorption on hydrous Mn and Fe oxides (Zhu et al., 2012). Chemical fertilizer, as a basic agricultural input for crop growth, is another important factor affecting Cd mobility by directly reacting with Cd or altering soil properties such as pH and surface charge (Tu et al., 2000). Application of nitrogen (N), phosphorus (P) and potassium (K) fertilizer increased the solubility and exchangeable fraction of Cd in soil (Kashem and Singh, 2002; Chen et al., 2006). Lime, a calcareous material obtained very easily and inexpensively, is the most widely used stabilizer of soil Cd by increasing soil pH (Haddad et al., 2017). Wetland rice ecosystems, widely distributed in Asian countries such as China, India, Japan and Korea (Zeng et al., 2015), are generally in flooded and unflooded conditions by rotation to meet rice's needs for growth (Zhu et al., 2012). Paddy soil maintained under long-term flooding conditions before harvest may induce an array of abiotic and

Corresponding author. E-mail address: [email protected] (X.-y. Xiao).

https://doi.org/10.1016/j.ecoenv.2018.04.049 Received 22 December 2017; Received in revised form 19 April 2018; Accepted 23 April 2018 0147-6513/ © 2018 Elsevier Inc. All rights reserved.

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

weight to maintain an approximate 2.0 cm depth of surface water. In the drying-wetting experiments, the soils were incubated without water addition until the day before sampling when they were re-submerged under 2.0 cm depth water for 24 h. The overlying water (approximate 140 mL) in each treatment was removed carefully by pipette for analysis and collected on day 5, 15, 30, 60 and 90. Soil samples from each pot were collected at 30 and 90 days incubation to determine Cd fractions using sequential extractions. The incubation experiments were conducted in a thermostatic incubator maintained at 25 ± 1 °C and a relative humidity of 80%. The pH of each water sample was measured immediately, then the water was passed through a 0.45-μm membrane filter. Some of the filtrate was analyzed for concentrations of dissolved TN, TP and SO42-. The remainder of each filtered water sample was acidified to pH < 2.0 using concentrated nitric acid (HNO3) prior to being analyzed for dissolved Cd, Fe and Mn concentrations.

biotic reduction processes (Boivin et al., 2002). These processes largely modify soil physiochemical properties during periods of flooding and subsequent drainage (Caetano et al., 2003) and can affect the mobility of Cd directly via changes in its speciation or indirectly through related changes in pH, dissolved organic carbon (DOC), and the redox chemistry of Fe, Mn, and sulfur (S) (Frohne et al., 2011; Shaheen et al., 2014a, 2014b). Previous study using column leaching tests have focused on the influence of different water management regimes on soil Cd release to groundwater. Soil submersion decreased the Cd concentration in leachate due to the increased pH (Yun and Yu, 2015). Lime addition also can increase pH and accordingly reduce Cd concentration in column leachate (Ash et al., 2015). However, current knowledge is still limited about how the dynamics of redox chemistry combined with agro-measures to affect the release and transformation of Cd in soils during incubation experiments (Pan et al., 2016). In this study, a 90-day static laboratory incubation test was conducted using Cd contaminated paddy soil amended with NPK fertilizer and lime under water regimes of continuous flooding and alternating dryingwetting cycles. The objectives of the study were to: 1) investigate the release characteristics of dissolved Cd in soil to overlying water; 2) elucidate the relationships between Cd concentration and soil Eh, and concentrations of dissolved total N (TN), total P (TP), sulfate (SO42-), Fe, Mn and pH in the overlying water; and 3) determine the transformation characteristics of soil Cd affected by water management combined with application of NPK fertilizer and lime. This study will extend the knowledge to understand the potential risk of Cd release from soils to surface water as induced by agricultural measures.

2.3. Sample analysis Selected soil properties were determined according to the general methods described by Lu (1999). Soil pH was measured using a glass electrode (Mettler Toledo 420) at a soil: water ratio of 1:2.5. Soil organic matter was determined using a volumetric method of potassium dichromate (K2Cr2O7) heating. The content of free iron oxide (Fe2O3) in soil was determined following extraction using a dithionite citrate system buffered with sodium bicarbonate; amorphous Fe2O3 was extracted using ammonium and oxalate under dark conditions. Soil Eh was measured using mini platinum-electrodes (Falkenberg, ELANA Ltd, Germany). The electrodes were horizontally installed and exactly positioned at the measurement depth of 5 cm below the soil surface. The pH of overlying water was determined using a glass electrode. The concentrations of dissolved TN and TP in water samples were digested with alkaline potassium persulfate and determined by UV spectrophotometric method, and the SO42- concentration was determined by barium chromate spectrophotometry as described by Lu (1999). The fractionation of Cd in soils was achieved using the sequential extraction procedure of European Community Bureau of Reference (BCR) method (Rauret et al., 1999). The method basically consisted of three extraction steps: (i) acid-soluble fraction (Aci-Cd) was extracted with acetic acid; (ii) the Cd combined with Fe/Mn oxides (reducible fraction, Red-Cd) was extracted with hydroxylamine hydrochloride; (iii) the Cd bound to organic matter and sulfides was extracted with hydrogen peroxide and ammonium acetate (oxidizable fraction, OxiCd). Soil samples and residual fractions (Res-Cd) were digested in a mixture of HNO3 and H2O2 to determine the content of Cd (USEPA, 1996). The concentrations of dissolved Cd, Fe and Mn in overlying water samples, extracting solution and digested solutions were determined using mass spectrometry with inductively coupled plasma (ICP-MS, Agilent 7500 Series). The accuracy of the digestion procedure and analytical method was checked with certified soil reference material (GBW-08303) obtained from China National Center, yielding analytical error < 10%. The recovery efficiency of the sequential extraction was calculated as: Recovery (%) = [(Step1 + Step2 + Step3 + Residual)/ (pseudo-total)] × 100, and the recovery values were found to be in the range of 90–105%.

2. Materials and methods 2.1. Soil physicochemical properties Surface soil (0–20 cm depth) was collected from an abandoned paddy field near a lead and zinc smelting plant in Hengyang city, Hunan Province, Southern China. The soil sample was air-dried, ground, and sieved through a 2-mm nylon screen before testing. The selected physiochemical properties of the soil were as follows: pH, 4.98; available nitrogen (N) content, 95.4 mg/kg; available phosphorous (P) content, 3.2 mg/kg; available potassium (K) content, 20.2 mg/kg; organic matter content, 27.3 g/kg. The contents of free Fe2O3 and amorphous Fe2O3 were 15.9 and 3.51 g/kg, respectively. The total Cd content was 11.5 mg/kg and far exceeded the recommended Cd content (0.3 mg/kg) for acid soil as described in the Chinese Environmental Quality Standard for Soils (GBl5618-1995) (Grade II for soil pH < 6.5) (MEPPRC, 1995). 2.2. Experimental design and laboratory incubation Air-dried soil samples (500 g) were placed in individual plastic pots (14 cm in height and 11 cm in diameter). The NPK fertilizer consisted of urea (CO(NH2)2), calcium dihydrogen phosphate (Ca(H2PO4)2) and potassium chloride (KCl) and was added to the soil with 0.2 g N/kg, 0.1 g P2O5/kg and 0.2 g K2O/kg soil. Lime (L) using calcium hydroxide (Ca(OH)2) was added at 2 g/kg soil. A treatment without fertilizer and lime addition was used as a control (CK). The treatment soils and controls were then incubated under two water regimes of continuous flooding (F) and alternating drying-wetting cycles (DW) using deionized water. Therefore, the treatments included (i) controls of CK+F and CK+DW, (ii) NPK fertilizer addition treatments of NPK+F and NPK+DW, and (iii) NPK fertilizer addition combined with lime treatments of NPK+L+F and NPK+L+DW. Each treatment was in duplicates. All chemicals of NPK fertilizer and lime were analytical reagent grade. On day 1 of an experiment, deionized water was added to all treatments to submerge the soil under 2.0 ± 0.5 cm depth of water. In the continuous flooding treatments, pots sprayed daily by measuring

2.4. Statistical analysis Statistical analyses were performed using Microsoft Excel 2013 and SPSS 19.0. One-way analysis of variance (ANOVA) was used to examine statistically significant differences among the Cd concentrations of overlying water and soils in different treatments. A probability level of 0.05 was considered to be statistically significant. 39

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

200

CK+F NPK+F NPK+L+F

soil Eh (mV)

160

addition (NPK+F and NPK+DW) during the entire incubation period. Similarly, Eh in lime-treated soil decreased more rapidly than that in a control (non-limed) soil under waterlogging condition; this difference was attributed to a smaller void ratio and lower air penetration in the lime-treated soil (Yun and Yu, 2015). At the end of the incubation experiment, soil Eh in treatments NPK+DW and NPK+F was higher than that in the corresponding controls without fertilizer, indicating that NPK fertilizer addition can increase soil Eh regardless of water management. The increase in soil Eh caused by urea application was due to the high standard potential of NO3- to its reduced forms (Wang et al., 1992). The pH of overlying water from all treatments showed similar slight decreasing trends by day 15 (Fig. 1). Then, the water pH from treatments NPK+DW and NPK+L+DW was significantly lower than that from corresponding continuous flooding treatments after 15 days incubation. Meanwhile, the pH of overlying water from CK+DW was also lower compared to that for the CK+F treatment at the end of experiment, indicating that drying-wetting cycles favored release of soil H+ ions to surface water. There was a significant negative relationship between overlying water pH and soil Eh (P < 0.05) (Table 1), which was in agreement with previous research that showed lower soil solution pH from a treatment involving drying-wetting cycles as compared with flooding treatment (Zhu et al., 2012), perhaps due to the consumption of protons for the reduction of Mn and Fe in waterlogged soil (Pan et al., 2016). Moreover, the water pH in NPK+DW treatment was significantly lower than that in the other treatments after 30 days of incubation, reflecting that NPK fertilizer further decreased water pH due to greater nitrification under aerobic condition in the drying-wetting treatment (Pan et al., 2014; Dong et al., 2017). After being applied to soil, urea is rapidly hydrolyzed to ammonium (NH4+) by the enzyme urease (Agehara and Warncke, 2005), and then the NH4+ undergoes nitrification, releasing H+ ions and leading to a decline in soil pH (Tu et al., 2000; Xu et al., 2017). The overlying water pH from treatments NPK+L+F and NPK+L +DW was higher than that from corresponding treatments without lime addition after 60 days incubation. This could be attributed to lime as a representative proton acceptor causing an increase of OH- in overlying water (Yun and Yu, 2015). The water pH from the NPK+L+F treatment slightly decreased from 7.76 to 7.28 with increasing incubation time. In contrast, water pH from the NPK+L+DW treatment decreased sharply at day 30 of incubation, and then kept stable at 4.64–4.82. This result may be related to the fact that the significant increase in soil Eh with drying-wetting treatment during days 30–90 resulted in OH- ions consumed by Fe and Mn fixed in hydroxide form (Shaheen et al., 2014a; Yun and Yu, 2015).

CK+DW NPK+DW NPK+L+DW

120 80 40 0 0

15

30 45 60 Incubation time (day)

0

15

30 45 60 Incubation time (day)

75

90

9

overlying water pH

8 7 6 5 4 3 75

90

Fig. 1. The change in soil Eh and overlying water pH during 90 days incubation.

3. Results and discussion 3.1. Changes in soil Eh and overlying water pH Generally, soil Eh from all treatments showed the similar decreasing trend with increasing incubation time during the early period of 5–30 days (Fig. 1). This result may be attributed to deterioration of air penetration into the soil after submergence (Yun and Yu, 2015). In addition, oxygen in soil is consumed fast by biological and microbiological activities leading to Eh decrease (Abgottspon et al., 2015). Soil Eh in the drying-wetting treatments increased sharply to a peak on day 60 and then remained stable until the end of incubation (day 90). Soil Eh in continuous flooding treatments was constant after 30 days of incubation and significantly lower than that in drying-wetting treatments, which was in agreement with previous research (Shaheen et al., 2014a; Yun and Yu, 2015). Under continuous flooding conditions, O2 depletion occurs as a consequence of mineralization of organic matter (Pan et al., 2014). Soil Eh in lime treatments NPK+L+F and NPK+L+DW was significantly lower than in corresponding treatments without lime

3.2. Changes in concentrations of dissolved TN and TP in overlying water The dissolved TN concentrations from control treatments CK+F and CK+DW were close and increased from 0.16 to 6.51 mg/L as the incubation proceeded (Fig. 2). The concentrations of TN from treatments of NPK+DW and NPK+L+DW peaked on day 60 (51.0 and 66.9 mg/L,

Table 1 Pearson correlation coefficients between Cd concentration and soil Eh, overlying water pH and concentrations of TN, TP, SO42-, Fe and Mn in overlying water. Parameters Cd Mn Fe SO42TP TN pH

Eh

pH **

0.646 0.430* 0.079 − 0.067 − 0.424* 0.051 − 0.745**

TN

− 0.650 − 0.534* − 0.104 0.235 0.211 − 0.088 1

**

TP − 0.276 − 0.200 − 0.296 − 0.034 1

**

0.612 0.681** − 0.350 0.719** 0.125 1

* Correlation is significant at the 0.05 level (P < 0.05). ** Correlation is significant at the 0.01 level (P < 0.01). 40

SO42*

0.372 0.364* − 0.157 1

Fe

Mn

− 0.231 − 0.210 1

0.822** 1

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

CK+F NPK+F NPK+L+F

TN concentration (mg/L)

80

CK+DW NPK+DW NPK+L+DW

that large amounts of P may be lost when manure or fertilizer application is followed by a period of intense rainfall (Zeng et al., 2008). The TP concentrations from treatments NPK+F and NPK+L+F were higher than those of corresponding drying-wetting treatments on day 60, indicating that continuous flooding promoted the solubility of P from NPK fertilizer treatments whether applied alone or in combination with lime. The TP concentration was significantly positive with soil Eh (Table 1). A lower P sorption maximum and adsorption coefficient under anaerobic conditions compared with aerobic conditions were expected because of P desorption from Fe–P compounds due to the reduction of Fe(III) to Fe(II) (Pant and Reddy, 2001). Moreover, the TP concentration in water from the NPK+L+F treatment was higher than that from the other treatments, which may be related to the highest pH value resulting in P desorption (Hu et al., 2003) (Fig. 1). In general, phosphorus is regarded as an immobile nutrient in soil because it is fixed firmly onto soil colloids as insoluble inorganic phosphates in the presence of excess Al3+, Ca2+, Fe2+, and Mn4+ ions (McDowell et al., 2001). Therefore, the release of TP from soil was very low compared to the release of TN. Nevertheless, the total P concentration in almost 27% of samples exceeded 0.15 mg/L, which is accepted as the critical total P concentration for water eutrophication (Van der Molen et al., 1998). Phosphorus fertilization increased the concentration of P in overlying water, especially in the NPK+DW treatment with 0.24 mg/L observed at day 15 incubation. Thus, fertilization increased the potential for P transportation by surface runoff to streams and subsequent contamination of water bodies by eutrophication (Lee et al., 2011).

60

40

20

0 0

15

30 45 60 Incubation time (day)

75

90

0

15

30 45 60 Incubation time (day)

75

90

TP concentration (mg/L)

0.3

0.2

0.1

0.0

3.3. Changes in concentrations of dissolved SO42-, Fe and Mn in overlying water

Fig. 2. The change in concentrations of dissolved TN, TP in overlying water released from paddy soils during 90 days incubation.

The SO42– concentration in overlying water from NPK+DW treatment increased throughout the incubation period, while those from other treatments fluctuated without an obvious trend (Fig. 3). Generally, the concentration of SO42– in water from the NPK+L+F treatment was higher than that from NPK+L+DW treatment during the incubation period of 5–30 days; however, a significantly contrary result was observed after 60 days of incubation. The SO42– concentrations from treatments CK+DW and NPK+DW were higher than those from corresponding continuous flooding treatments, and it was in agreement with a previous research that a transformation from reducing to oxidizing conditions in soil results in the release of SO42– into soil solution (De et al., 2011). This phenomenon may be related to the fact that reducing solutions generally contain H2S and an important S-containing ion in oxidizing solutions is SO42– (Tai et al., 2017). The SO42– concentrations in treatments with lime were higher than those from corresponding treatments without lime addition, which may be explained by the positive relationships between pH and water-soluble SO42– concentration (Xu et al., 2017). The SO42– concentration was significantly positively correlated with TN concentration (P < 0.05) (Table 1), which was consistent with previous research that N fertilizer can enhance soil water-soluble SO42– concentration (Xu et al., 2017). The Fe concentrations in overlying water from the controls CK+F and CK+DW were significantly higher than those from treatments involving both NPK and NPK+L during the incubation period of 5–30 days (Fig. 3). The Fe concentrations from CK+F and NPK+F treatments sharply increased in contrast to that from CK+DW treatment which decreased after 30 days incubation. These results indicated that flooding favored Fe release from soil, which was consistent with the result that soluble Fe concentration was negatively correlated with soil Eh (Shaheen et al., 2016). Fe-Mn oxides and oxyhydroxides are more dissolved under flooding and more precipitated during oxidation in the drying-wetting cycles (De et al., 2011). The Fe concentration from the NPK+L+F treatment was significantly lower than those from the other continuous flooding treatments throughout the experiment due to lime addition causing the increase in soil pH and Fe hydroxide (Shaheen et al., 2014a).

respectively), then decreased sharply at the end of incubation. Similarly, the TN concentrations from corresponding flooding treatments also increased during the early incubation period (days 5–30) and then decreased. The sharp increase in TN concentrations in overlying water during the early period of incubation may be related to N release from the exogenous N fertilizer. Tomar and Soper (1981) have reported that urea hydrolysis after 4 weeks of incubation averaged 81% in 11 different soils. Generally, the TN concentrations from treatments NPK+F and NPK+L+F were higher than those from corresponding dryingwetting treatments at the early incubation stage of 5–30 days, but showed the opposite relationship as the incubation proceeded. This result was consistent with previous finding that greater soil N loss occurred from soil under alternating flooding and drying than under continual submergence (Hanif et al., 1986). In addition, the TN concentration from the NPK+L+DW treatment was higher than that in NPK+DW treatment throughout the incubation period. A similar research has showed that the release of NH4+-N from soils increased with increasing lime application rates (Haddad et al., 2017) due to the increase of hydrolysis of some amines in soil organic matter (Senwo and Tabatabai, 1998). The TN concentrations in overlying water released from controls and treated soils were higher than the Dutch surface water quality standard of 2.2 mg/L total N, which is generally accepted as the critical total N concentration that causes water eutrophication (Van der Molen et al., 1998). Therefore, runoff or drainage water from contaminated paddy soils was an important non-point source of surface water eutrophication in this region. The dissolved TP concentrations in overlying water from treatments NPK and NPK+L under the two water management regimes increased sharply at the early incubation state (days 5–15), and thereafter decreased markedly with increasing incubation time (Fig. 2). These patterns may be related to the release of exogenous P from fertilizer to the overlying water (Chen et al., 2017). Previous study has demonstrated 41

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

SO4 2- concentration (mg/L)

100

CK+F NPK+F NPK+L+F

CK+DW NPK+DW NPK+L+DW

0.4

incubation (5–30 days), but peaked at 1.39 and 1.15 mg/L, respectively on day 90. In general, the concentrations of Mn from the treatments of CK+F, NPK+F and NPK+L+F were higher than those from the corresponding treatments under drying-wetting cycles during the early incubation period of 5–30 days, which was in agreement with previous study (Shaheen et al., 2014a), and due to the increase of the Mn hydrous oxides caused by Eh decrease (Kashem and Singh, 2004). However, the Mn concentrations from treatments under continuous flooding were lower than those from corresponding drying-wetting treatments at the end of incubation. This result may be linked to increasing soil microbial populations during favorable aerobic conditions in the drying phase, which produces large amounts of soluble Mn when the soil rewets and becomes saturated (Hardie et al., 2007). The Mn concentration in NPK+L treated soil under both water management regimes decreased significantly compared with the corresponding NPK treatment, which may be related to the fact that both flooding and lime significantly increased soil pH (Fig. 1), then Mn availability subsequently decreased due to the formation of Mn hydroxide (Yun and Yu, 2015). The Mn concentration showed negative correlation with pH, but positive correlation with soil Eh, TN and SO42- concentrations in overlying water (Table 1).

0.2

3.4. Characteristics of soil Cd release to overlying water

80 60 40 20 0 0

15

30 45 60 Incubation time (day)

75

90

Fe concentration (mg/L)

0.8

0.6

The dissolved Cd concentrations in overlying water from all treatments varied widely during soil incubation period. The Cd concentrations from treatments NPK+F (102 μg/L) and NPK+DW (62.6 μg/L) were high at 5 days incubation, and significantly higher than those from controls CK+F and CK+DW throughout the incubation except on day 30 (Table 2). This result indicated that fertilizer application can threaten water quality, in agreement with the result that commercial NPK fertilizer enhances the mobility of Cd even in soils with high heavy metal retention capacity (Ammar et al., 2016). Previous research showed that phosphate fertilizers increased the Cd concentrations in soil water extracts by 87% and 80% in field and laboratory experiments, respectively compared to the control (Lambert et al., 2007), which may be due to the reduced pH resulting from application of phosphate fertilizers (Grant et al., 2002). In addition, the application of N fertilizer releases NH4+ cations that can displace Cd from adsorption sites (Lorenz et al., 1994). During days 60–90, the Cd concentrations from treatments CK, NPK and NPK+L under drying-wetting cycles were higher than those from corresponding treatments under continuous flooding. The Cd concentrations from CK+DW and NPK+DW treatments were especially high and peaked at 94.8 and 276 μg/L on day 90, respectively, indicating that drying-wetting cycles may induce Cd release. Drying of acid soil (pH < 6) can promote Cd desorption (Tang et al., 2011), which may be related to the fact that the sulfide in soil is oxidized under drying-wetting cycles resulting in dissolution of cadmium sulfide (CdS) (Rizwan et al., 2018; Shaheen et al., 2014b). Moreover, the pH of acid soils can increase after flooding, causing an increase in the adsorption of Cd to reactive surfaces of soil organic

0.0 0

15

30 45 60 Incubation time (day)

75

90

0

15

30 45 60 Incubation time (day)

75

90

Mn concentration (mg/L)

1.5 1.2 0.9 0.6 0.3 0.0

Fig. 3. The change in concentrations of dissolved SO42-, Fe and Mn in overlying water released from paddy soils during 90 days incubation.

The Mn concentration in overlying water from the NPK+F treatment decreased with increasing incubation time and was significantly higher than those from the NPK+L+F treatment as well as from the controls CK+F and CK+DW (Fig. 3). The Mn concentrations from NPK +DW and NPK+L+DW treatments were low at the early stage of

Table 2 The concentration of Cd in overlying water released from contaminated soil during 90 days incubation. Treatments

CK+F CK+DW NPK +F NPK+DW NPK +L+F NPK +L+DW

Cd concentration (μg/L) 5d

15d

23.8 ± 1.22cC 17.4 ± 2.43dC 102 ± 1.06aA 62.6 ± 1.88bC 4.55 ± 1.17eBC 2.59 ± 0.01eD

40.5 19.4 87.0 60.5 11.0 9.11

30d ± ± ± ± ± ±

0.33bcB 1.17cBC 10.2aA 2.48abC 0.14cA 1.19cD

58.3 27.9 29.9 60.0 9.14 26.2

± ± ± ± ± ±

0.91aA 4.06bBC 12.1bB 9.62aC 1.30cA 7.00bcC

60d

90 d

23.2 ± 2.49dC 31.0 ± 1.10cB 25.8 ± 0.55dB 219 ± 1.20aB 8.09 ± 2.83eAB 190 ± 1.20bA

9.09 ± 0.80dD 94.8 ± 9.90cA 81.0 ± 1.25cA 276 ± 20.9aA 4.03 ± 0.15dC 160 ± 1.63bB

Data are presented as mean values ± stdev. Means followed by the same lower case letter within the same column are not significantly different (P > 0.05). Means followed by the same capital letter within the same horizontal line are not significantly different (P > 0.05). 42

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

Table 3 The chemical fractionations of Cd in soil after 30 and 90 days incubation. Incubation time

Treatments

Cd content (mg/kg) Aci-Cd

30 days

90 days

Original soil CK+F CK+DW NPK+F NPK+DW NPK+L+F NPK+L+DW Original soil CK+F CK+DW NPK+F NPK+DW NPK+L+F NPK+L+DW

6.91 6.32 6.66 6.20 7.29 6.23 6.80 6.91 7.53 7.83 7.31 7.74 6.58 7.25

± ± ± ± ± ± ± ± ± ± ± ± ± ±

Red-Cd 0.09b 0.15c 0.18b 0.23c 0.07a 0.02c 0.11b 0.09cd 0.09ab 0.14a 0.10bc 0.20a 0.31d 0.20bc

1.13 1.53 1.31 1.33 1.27 1.44 1.34 1.13 1.44 1.38 1.71 1.84 1.29 1.62

± ± ± ± ± ± ± ± ± ± ± ± ± ±

Oxi-Cd 0.02d 0.03a 0.03c 0.04c 0.05c 0.01b 0.06c 0.02e 0.06bcd 0.16cde 0.23ab 0.07a 0.06de 0.01abc

0.36 0.41 0.40 0.59 0.41 0.37 0.37 0.36 0.31 0.38 0.36 0.38 0.36 0.38

± ± ± ± ± ± ± ± ± ± ± ± ± ±

Res-Cd 0.01d 0.03bc 0.01bcd 0.02a 0.00b 0.01cd 0.02d 0.01a 0.01b 0.02a 0.01a 0.03a 0.01a 0.01a

3.10 3.24 3.13 3.38 2.53 3.39 3.00 3.10 2.22 1.91 2.13 1.54 3.28 2.26

± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.08ab 0.20ab 0.14ab 0.24a 0.12c 0.11a 0.03b 0.08a 0.03b 0.32bc 0.32b 0.16c 0.25a 0.18b

Data are presented as mean values ± stdev. Means followed by the same lower case letter within the same column are not significantly different (P > 0.05). Table 4 Pearson correlation coefficients between soil Cd fractions and soil pH, Eh and overlying water Cd. Parameters Aci-Cd Red-Cd Oxi-Cd Res-Cd Overlying water Cd Soil pH

Red-Cd 0.417 1

*

Oxi-Cd

Res-Cd

− 0.412 − 0.191 1

*

Overlying water Cd

− 0.956 − 0.625** 0.318 1 **

**

0.571 0.734** − 0.026 − 0.704** 1

Soil pH

Soil Eh

− 0.452 − 0.277 − 0.191 0.494* − 0.480* 1 *

0.617** 0.451* 0.107 − 0.683** 0.836** − 0.731**

* Correlation is significant at the 0.05 level (P < 0.05). ** Correlation is significant at the 0.01 level (P < 0.01).

treatments under continuous flooding significantly decreased by 8.5–10.3%, while that from the NPK+DW treatment significantly increased by 5.4% as compared with original soil (P < 0.05). At the end of incubation (day 90), the Aci-Cd contents from all treatments except NPK+L+F increased by 4.9–13.3% compared with that of the original soil, indicating that the potential risk of soil Cd contamination to water increased after long-term incubation. The Cd concentration in overlying water showed significant positive correlation with the Aci-Cd content in soil (P < 0.01) (Table 4). After 30 and 90 days of incubation, the contents of Aci-Cd from treatments CK+DW, NPK+DW and NPK+L +DW were higher than those from the corresponding flooding treatments, while contents of Res-Cd showed the opposite relationship, suggesting that drying-wetting cycles greatly enhance the Cd mobility potential to the overlying water. This observation is consistent with the results that water Cd concentrations from treatments under dryingwetting cycles were higher than those under flooding condition (Table 2). In addition, the content of Aci-Cd, as well as the water Cd concentration from treatment NPK+L+F, was the lowest among all treatments at the end of incubation. There was significant negative correlation between Aci-Cd and Res-Cd (P < 0.01), and Res-Cd showed significant positive correlation with soil pH and negative correlation with Eh (P < 0.01) (Table 4), suggesting that lime application and continuous flooding can directly transform the highly availability fraction of Cd to a more stable fraction (Res-Cd) due to the higher soil pH. The higher soil pH under continuous flooding combined with lime addition caused an increase in the variable negative charges of soils and increased adsorption of Cd onto colloids, ultimately resulting in lower exchangeable Cd (Li and Xu, 2017b).

matter, dissolved organic matter and metal-(hydr)oxides (Pan et al., 2016). Meanwhile, there was a significantly negative correlation between Cd concentration and pH in overlying water, but positive relationships between Cd concentration and soil Eh, as well as concentrations of TN, SO42- and Mn in overlying water (Table 1), further indicating that alternating drying and wetting of soil combined with application of N fertilizer may enhance Cd release. Generally, the Cd concentrations in overlying water from all treatments without lime addition were higher than the Cd threshold limit of Chinese Environmental Quality Standards for Surface Water (GB38382002) (10 μg/L Grade V) (MEPPRC, 2002) (Table 2) throughout the experiment. The Cd concentration from NPK+L+DW treatment was less than 10 μg/L before 15 days of incubation, but significantly increased as incubation proceeded, indicating that the Cd-polluted paddy soil would threaten surface water quality through runoff and drainage. However, the Cd concentration from the NPK+L+F treatment was almost lower than 10 μg/L throughout the incubation, suggesting that lime addition combined with flooding would lower the potential risk of Cd release to water by increasing the soil pH, resulting in lower solubility of Cd absorbed to Fe- Mn hydroxide (Janoš et al., 2010). 3.5. Transformation of Cd fractions in soil The mobility of Cd in soil strongly depends on its chemical fractionation or speciation (Qu et al., 2017). In the non-incubated (original) soil the proportion of acid-soluble Cd (Aci-Cd), the most mobile fraction of Cd, was high and accounted for 60.1% of total Cd, followed by the residual fraction (Res-Cd)(26.9%) and the reducible fraction (Red-Cd) (9.8%) (Table 3). This fractionation agreed with previous demonstration in other study that more than 50% of total Cd was Aci-Cd in contaminated soils (Cui et al., 2016). The Cd transformation among all fractions in soil may occur after incubation. At day 30 of incubation, the contents of Aci-Cd in soil from

4. Conclusions Water management and agro-measures significantly affect soil release to water body. This study showed that the Cd concentration has a 43

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

significantly negative correlation with overlying water pH, but positive correlation with soil Eh and concentrations of TN, SO42- and Mn in overlying water. Addition of N fertilizer and alternating drying-wetting cultivation of Cd-contaminated paddy soil enhance Cd release from soil to water. The Cd concentration in overlying water released from all treatments except NPK+L+F treatment exceeded the Cd threshold limit of Chinese Environmental Quality Standards for Surface Water (10 μg/L Grade V) and poses potential risk to surface water quality. The proportion of Cd in acid-soluble fraction from all treatments except NPK +L+F increased after 90 days incubation, and the residual fraction Cd from NPK+L+F treatment was significantly higher than those from other treatments. Continuous flooding cultivation combined with lime addition is a good management strategy for decreasing the potential risk of Cd pollution from paddy soil to a surrounding water body. Cdcontaminated paddy soil needs more careful management and attention to prevent surface water quality pollution.

Kashem, M.A., Singh, B.R., 2002. The effect of fertilizer additions on the solubility and plant-availability of Cd, Ni and Zn in soil. Nutr. Cycl. Agroecon. 62, 287–296. Kashem, M.A., Singh, B.R., 2004. Transformations in solid phase species of metals as affected by flooding and organic matter. Commun. Soil Sci. Plan. 35, 1435–1456. Lambert, R., Grant, C., Sauvé, S., 2007. Cadmium and zinc in soil solution extracts following the application of phosphate fertilizers. Sci. Total Environ. 378, 293–305. Lee, C.H., Hong, C.O., Kim, S.Y., Schumacher, T., Kim, P.J., 2011. Reduction of phosphorus release by liming from temporary flooded rice rotational system in greenhouse upland soil. Ecol. Eng. 37, 1239–1243. Li, J.R., Xu, Y.M., 2017a. Use of clay to remediate cadmium contaminated soil under different water management regimes. Ecotoxicol. Environ. Saf. 141, 107–112. Li, J.R., Xu, Y.M., 2017b. Immobilization remediation of Cd-polluted soil with different water condition. J. Environ. Manag. 193, 607–612. Liu, Y.N., Guo, Z.H., Xiao, X.Y., Wang, S., Jiang, Z.C., Zeng, P., 2017. Phytostabilisation potential of giant reed for metals contaminated soil modified with complex organic fertiliser and fly ash: a field experiment. Sci. Total Environ. 576, 292–302. Lorenz, S.E., Hamon, R.E., Mcgrath, S.P., Holm, P.E., Christensen, T.H., 1994. Applications of fertilizer cations affect cadmium and zinc concentrations in soil solutions and uptake by plants. Eur. J. Soil Sci. 45, 159–165. Lu, R.K., 1999. Soil and Agro-chemistry Analytical Methods. Agriculture Science and Technology Press of China, Beijing. McDowell, R.W., Sharpley, A.N., Condron, L.M., Haygarth, P.M., Brookes, P.C., 2001. Processes controlling soil phosphorus release to runoff and implications for agricultural management. Nutr. Cycl. Agroecon. 59, 269–284. MEPPRC (Ministry of Environmental Protection of the People's Republic of China), 1995. Environmental Quality Standard for Soils, GBl5618-1995. China Environmental Science Press, Beijing. MEPPRC (Ministry of Environmental Protection of the People's Republic of China), 2002. Environmental Quality Standards for Surface Water, GB3838-2002. China Environmental Science Press, Beijing. Pant, H.K., Reddy, K.R., 2001. Phosphorus sorption characteristics of estuarine sediments under different redox conditions. J. Environ. Qual. 30, 1474–1480. Pan, Y.Y., Bonten, L.T.C., Koopmans, G.F., Song, J., Luo, Y., Temminghoff, E.J.M., Comans, R.N.J., 2016. Solubility of trace metals in two contaminated paddy soils exposed to alternating flooding and drainage. Geoderma 261, 59–69. Pan, Y.Y., Koopmans, G.F., Bonten, L.T.C., Song, J., Luo, Y.M., Temminghoff, E.J.M., Comans, R.M.J., 2014. Influence of pH on the redox chemistry of metal (hydr)oxides and organic matter in paddy soils. J. Soil Sediment. 14, 1713–1726. Qu, L., Xie, Y.Y., Lu, G., Yang, C.F., Zhou, J.N., Yi, X.Y., Dang, Z., 2017. Distribution, fractionation, and contamination assessment of heavy metals in paddy soil related to acid mine drainage. Paddy Water Environ. 15, 553–562. Rauret, G., Lopez-Sanchez, J.F., Sahuquillo, A., Rubio, R., Davidson, C., Ure, A., Quevauviller, 1999. Improvement of the BCR three step sequential extraction procedure prior to the certification of new sediment and soil reference materials. J. Environ. Monit. 1, 57–61. Rizwan, M., Ali, S., Abbas, T., Adrees, M., Zia-ur-Rehman, M., Ibrahim, M., 2018. Residual effects of biochar on growth, photosynthesis and cadmium uptake in rice (Oryza sativa L.) under Cd stress with different water conditions. J. Environ. Manag. 206, 676–683. Schipper, P.N.M., Bonten, L.T.C., Plette, A.C.C., Moolenaar, S.W., 2008. Measures to diminish leaching of heavy metals to surface waters from agricultural soils. Desalination 226, 89–96. Senwo, Z.N., Tabatabai, M.A., 1998. Amino acid composition of soil organic matter. Biol. Fert. Soils 26, 235–242. Shaheen, S.M., Rinklebe, J.R., Frohne, T., White, J.R., DeLaune, R.D., 2016. Redox effects on release kinetics of arsenic, cadmium, cobalt, and vanadium in Wax Lake Deltaic freshwater marsh soils. Chemosphere 150, 740–748. Shaheen, S.M., Rinklebe, J.R., Rupp, H., Meissner, R., 2014a. Lysimeter trials to assess the impact of different flood-dry-cycles on the dynamics of pore water concentrations of As, Cr, Mo and V in a contaminated floodplain soil. Geoderma 228–229, 5–13. Shaheen, S.M., Rinklebe, J.R., Rupp, H., Meissner, R., 2014b. Temporal dynamics of pore water concentrations of Cd, Co, Cu, Ni, and Zn and their controlling factors in a contaminated floodplain soil assessed by undisturbed groundwater lysimeters. Environ. Pollut. 191, 223–231. Tai, Y.P., Li, Z.A., McBride, M.B., Yang, Y., 2017. Dry cultivation enhances cadmium solubility in contaminated soils but minimizes cadmium accumulation in a leafy vegetable. J. Soil Sediment. 17, 2822–2830. Tang, X.Y., Katou, H., Suzuki, K., Ohtani, T., 2011. Air-drying and liming effects on exchangeable cadmium mobilization in contaminated soils: a repeated batch extraction study. Geoderma 161, 18–29. Tomar, J.S., Soper, R.J., 1981. An incubation study of nitrogen added as urea to several Manitoba soils with particular reference to retention of nitrogen. Can. J. Soil Sci. 61, 1–10. Tu, C., Zheng, C.R., Chen, H.M., 2000. Effect of applying chemical fertilizers on forms of lead and cadmium in red soil. Chemosphere 41, 133–138. USEPA (US Environmental Protection Agency), 1996, pp. 1–12. Acid digestion of sediments sludge and soils. 〈http://www.epa.gov/sw-846/pdfs/3050b.pdf〉. Van der Molen, D.T., Breeuwsma, A., Boers, P.C.M., 1998. Agricultural nutrient losses to surface water in the Netherlands: impact, strategies, and perspectives. J. Environ. Qual. 27, 4–11. Wang, Z., Delaune, R.D., Lindau, C.W., Patrick Jr, W.H., 1992. Methane production from anaerobic soil amended with rice straw and nitrogen fertilizers. Fertil. Res. 33, 115–121. Wang, Z.X., Chai, L.Y., Wang, Y.Y., Yang, Z.H., Wang, H.Y., Wu, X., 2011. Potential health risk of arsenic and cadmium in groundwater near Xiangjiang River, China: a case study for risk assessment and management of toxic substances. Environ. Monit.

Acknowledgements This research was supported by the National Key Technology Research and Development Program of China (2015BAD05B02). References Abgottspon, F., Bigalke, M., Wilcke, W., 2015. Fast colloidal and dissolved release of trace elements in a carbonatic soil after experimental flooding. Geoderma 259–260, 156–163. Agehara, S., Warncke, D.D., 2005. Soil moisture and temperature effects on nitrogen release from organic nitrogen sources. Soil Sci. Soc. Am. J. 69, 1844–1855. Ammar, R., Kanbar, H.J., Kazpard, V., Wazne, M., El Samrani, A.G., Amacha, N., Saad, Z., Chou, L., 2016. Role of phosphogypsum and NPK amendments on the retention or leaching of metals in different soils. J. Environ. Manag. 178, 20–29. Ash, C., Tejnecký, V., Šebek, O., Houška, J., Chala, A.T., Drahota, P., Drábek, O., 2015. Redistribution of cadmium and lead fractions in contaminated soil samples due to experimental leaching. Geoderma 241–242, 126–135. Boivin, P., Favre, F., Hammecker, C., Maeght, J.L., Delarivière, J., Poussin, J.C., Wopereis, M.C.S., 2002. Processes driving soil solution chemistry in a flooded rice-cropped vertisol: analysis of long-time monitoring data. Geoderma 110, 87–107. Caetano, M., Madureira, M.J., Vale, C., 2003. Metal remobilisation during resuspension of anoxic contaminated sediment: short-term laboratory study. Water Air Soil Pollut. 143, 23–40. Chen, M.S., Ding, S.M., Zhang, L.P., Li, Y.Y., Sun, Q., Zhang, C.S., 2017. An investigation of the effects of elevated phosphorus in water on the release of heavy metals in sediments at a high resolution. Sci. Total Environ. 575, 330–337. Chen, S., Sun, T.H., Sun, L.N., Zhou, Q.X., Chao, L., 2006. Influences of phosphate nutritional level on the phytoavailability and speciation distribution of cadmium and lead in soil. J. Environ. Sci. 18, 1247–1253. Cui, L.Q., Pan, G.X., Li, L.Q., Bian, R.J., Liu, X.Y., Yan, J.L., Quan, G.X., Ding, C., Chen, T.M., Liu, Y., Yin, C., Wei, C., Yang, Y., Hussain, Q., 2016. Continuous immobilization of cadmium and lead in biochar amended contaminated paddy soil: a five-year field experiment. Ecol. Eng. 93, 1–8. De, L.J., McLaughlin, M.J., Hettiarachchi, G.M., Kirby, J.K., Beak, D.G., 2011. Cadmium solubility in paddy soils: effects of soil oxidation, metal sulfides and competitive ions. Sci. Total Environ. 409, 1489–1497. Dong, W.J., Guo, J., Xu, L.J., Song, Z.F., Zhang, J., Tang, A., Zhang, X.J., Leng, C.X., Liu, Y.H., 2017. Water regime-nitrogen fertilizer incorporation interaction: field study on methane and nitrous oxide emissions from a rice agroecosystem in Harbin, China. J. Environ. Sci. 64, 289–297. Frohne, T., Rinklebe, J., Diaz-Bone, R.A., Laing, G.D., 2011. Controlled variation of redox conditions in a floodplain soil: impact on metal mobilization and biomethylation of arsenic and antimony. Geoderma 160, 414–424. Grant, C.A., Bailey, L.D., Harapiak, J.T., Flore, N.A., 2002. Effect of phosphate source, rate and cadmium content and use of penicillium bilaii on phosphorus, zinc and cadmium concentration in durum wheat grain. J. Sci. Food Agr. 82, 301–308. Haddad, S.A., Tabatabai, M.A., Loynachan, T.E., 2017. Effects of liming and selected heavy metals on ammonium release in waterlogged agricultural soils. Biol. Fert. Soils 53, 153–158. Hanif, M., DeMooy, C.J., Lindsay, W.L., 1986. The effect of alternate flooding and drying on the pH and redox of submerged rice soil. J. Agron. Crop Sci. 156, 253–259. Hardie, A.M., Heal, K.V., Lilly, A., 2007. The Influence of pedology and changes in soil moisture status on manganese release from upland catchments: soil core laboratory experiments. Water Air Soil Pollut. 182, 369–382. Hu, Q.H., Zhu, Y.M., Song, J., Li, Z.Y., Wen, J.J., 2003. Effects of pH and Eh on release of nitrogen and phosphorus from sediments of West Lake. J. Zhejiang Univ. Sci. 4, 358–362. Janoš, P., Vávrová, J., Herzogová, L., Pilařová, V., 2010. Effects of inorganic and organic amendments on the mobility (leachability) of heavy metals in contaminated soil: a sequential extraction study. Geoderma 159, 335–341.

44

Ecotoxicology and Environmental Safety 159 (2018) 38–45

X.-q. Han et al.

contamination in rice-producing soils of hunan province, China and potential health risks. Int. J. Environ. Res. Public Health 12, 15584–15593. Zeng, S.C., Su, Z.Y., Chen, B.G., Wu, Q.T., Ou, Y.Y., 2008. Nitrogen and phosphorus runoff losses from orchard soils in South China as affected by fertilization depths and rates. Pedosphere 18, 45–53. Zhang, X.Y., Zhong, T.Y., Liu, L., Ouyang, X.Y., 2015. Impact of soil heavy metal pollution on food safety in China. PLoS One 10 (8). http://dx.doi.org/10.1371/journal.pone. 0135182. Zheng, S.N., Zhang, M.K., 2011. Effect of moisture regime on the redistribution of heavy metals in paddy soil. J. Environ. Sci. 23, 434–443. Zhu, N.M., Li, Q., Guo, X.J., Zhang, H., Deng, Y., 2014. Sequential extraction of an aerobic digestate sludge for the determination of partitioning of heavy metals. Ecotoxicol. Environ. Saf. 102, 18–24. Zhu, Q.H., Huang, D.Y., Liu, S.L., Zhou, B., Luo, Z.C., Zhu, H.H., 2012. Flooding-enhanced immobilization effect of sepiolite on cadmium in paddy soil. J. Soil Sediment. 12, 169–177.

Assess. 175, 167–173. Xu, C., Xiang, Q., Zhu, Q.H., Zhu, H.H., Huang, D.Y., Yang, Z.Z., 2017. Effects of controlled-release ureas on heavy-metal mobility in multimetal-contaminated soil. Pedosphere 10.1016/ S1002-0160(17)60467-3. Xue, S.G., Shi, L.Z., Wu, C., Wu, H., Qin, Y.Y., Pan, W.S., Hartley, W., Cui, M.Q., 2017. Cadmium, lead, and arsenic contamination in paddy soils of a mining area and their exposure effects on human HEPG2 and keratinocyte cell-lines. Environ. Res. 156, 23–30. Ye, X.X., Li, H.Y., Ma, Y.B., Wu, L., Sun, B., 2014. The bioaccumulation of Cd in rice grains in paddy soils as affected and predicted by soil properties. J. Soil Sediment. 14, 1407–1416. Ye, X.X., Li, H.Y., Zhang, L.G., Chai, R.S., Tu, R.F., Gao, H.J., 2018. Amendment damages the function of continuous flooding in decreasing Cd and Pb uptake by rice in acid paddy soil. Ecotoxicol. Environ. Saf. 147, 1708–1714. Yun, S.W., Yu, C., 2015. The leaching characteristics of Cd, Zn, and As from submerged paddy soil and the effect of limestone treatment. Paddy Water Environ. 13, 61–69. Zeng, F.F., Wei, W., Li, M.S., Huang, R.X., Yang, F., Duan, Y.Y., 2015. Heavy metal

45