Prediction of the solubility of zinc, copper, nickel ... - Springer Link

Environ Monit Assess (2013) 185:10015–10025 DOI 10.1007/s10661-013-3309-x

Prediction of the solubility of zinc, copper, nickel, cadmium, and lead in metal-contaminated soils Nafiseh Rang Zan & S. P. Datta & R. K. Rattan & B. S. Dwivedi & M. C. Meena

Received: 1 September 2012 / Accepted: 11 June 2013 / Published online: 24 July 2013 # Springer Science+Business Media Dordrecht 2013

Abstract Risk assessment of metal-contaminated soil depends on how precisely one can predict the solubility of metals in soils. Responses of plants and soil organisms to metal toxicity are explained by the variation in free metal ion activity in soil pore water. This study was undertaken to predict the free ion activity of Zn, Cu, Ni, Cd, and Pb in metal-contaminated soil as a function of pH, soil organic carbon, and extractable metal content. For this purpose, 21 surface soil samples (0–15 cm) were collected from agricultural lands of various locations receiving sewage sludge and industrial effluents for a long period. One soil sample was also collected from agricultural land which has been under intensive cropping and receiving irrigation through tube well water. Soil samples were varied widely in respect of physicochemical properties including metal content. Total Zn, Cu, Ni, Cd, and Pb in experimental soils were 2,015±3,373, 236±286, 103 ±192, 29.8±6.04, and 141±270 mg kg−1, respectively. Free metal ion activity, viz., pZn2+, pCu2+, pNi2+, pCd2+, and pPb2+, as estimated by the Baker soil test was 9.37±1.89, 13.1±1.96, 12.8±1.89, 11.9±2.00, and 11.6±1.52, respectively. Free metal ion activity was predicted by pH-dependent Freundlich equation (solubility model) as a function of pH, organic carbon, and N. R. Zan (*) : S. P. Datta : R. K. Rattan : B. S. Dwivedi : M. C. Meena Division of Soil Science and Agricultural Chemistry, Indian Agricultural Research Institute, New Delhi 110012, India e-mail: [email protected]

extractable metal. Results indicate that solubility model as a function of pH, Walkley–Black carbon (WBC), and ethylenediaminetetraacetic acid (EDTA)-extractable metals could explain the variation in pZn2+, pCu2+, pNi2+, pCd2+, and pPb2+ to the extent of 59, 56, 46, 52, and 51 %, respectively. Predictability of the solubility model based on pH, KMnO4-oxidizable carbon, and diethylenetriaminepentaacetic acid-extractable or CaCl2-extractable metal was inferior compared to that based on EDTA-extractable metals and WBC. Keywords Solubility model . pH . Organic carbon . Extractable metals . Contaminated soils . Risk assessment

Introduction Solubility of metal in contaminated soils is a key factor which controls the phytoavailability and toxic effects of metals on soil flora and fauna (François et al. 2004; Gandois et al. 2010). Over the years, various approaches have been employed to estimate the bioavailability of metal in soil. Most studies focused on quantification of the solid phase reservoir of available metal—the “capacity” factor (Datta and Young 2005). For example, current legislation in most countries still utilizes the total soil metal concentration as a simple index of hazard in contaminated soils. However, total metal load takes no account of soil characteristics that modify the bioavailability of metal pollutants in contaminated soils. Meaningful risk assessment of metal-

10016

contaminated soils cannot be based solely on total metal content as poor correlations were generally found between the total metal content and the mobile fractions of metal in soil (François et al. 2004; Ivezic et al. 2012). The toxic effects of heavy metal have also been related to some operationally defined extractable fractions (Datta and Young 2005). Attempts have also been made to characterize the bioavailability of metals in soil using an “intensity factor,” i.e., metal concentrations or activity in soil solution (Datta and Young 2005; Almas et al. 2005, 2007). There is considerable experimental evidence to suggest that responses of plants and soil organisms to metal toxicity are explained by the variation in free metal ion activity in soil pore water (Jopony and Young 1993; Hough et al. 2005). In such studies, extraction of a representative soil aqueous phase to which, in theory, plant roots or microorganisms are continually exposed is of the utmost importance. Extraction of soil pore water, subsequent chemical analysis, and speciation may be difficult to adopt on a routine basis to assess ecotoxicity of metal in contaminated soil. Baker and Amacher (1981) developed a diagnostic soil testing program to measure the intensity parameter of different ions in soil. In this procedure, soil is extracted with the Baker soil test solution and total concentration of metals, including pH, is determined in solution analytically. Free ionic activity of different elements is calculated by successive iterations using computational algorithms as described by Baker and Amacher (1981). However, adoption of this soil test for estimating free metal ion activity in soil solution is also difficult on a routine basis because, usually, analytical measurement of total trace metal concentration even in contaminated soil in Baker soil extract requires flameless atomic absorption spectroscopy (AAS; graphite furnace) or inductively coupled plasma–optical emission spectroscopy. Solid–solution equilibria of metal in soil have often been examined using the batch equilibration procedure with large solution to solid ratios. These data may then be used to generate adsorption isotherms, which can be described by a range of curve-fitting algorithms (Travis and Etnier 1981). However, there are several problems associated with the wider application of sorption isotherm to predict metal ion concentration in soil pore water (Tye et al. 2003) because there is no provision for taking into account the soil properties that affect the distribution of metal between solid and solution phases. Jopony and Young (1994) developed

Environ Monit Assess (2013) 185:10015–10025

a pH-dependent Freundlich equation to predict metal ion activity in soil pore water based on easily measurable soil characteristics. Subsequently, several researchers validated this and other similar models, particularly in temperate regions where soil organic carbon is usually very high (Datta and Young 2005; Hough et al. 2005; Gandois et al. 2010; Groenenberg et al. 2010; Loncaric et al. 2010). Such model is yet to be extensively tested under tropical environment where soil organic carbon content is much lower. Moreover, the predictability of such model has not been evaluated in respect to free metal ion activity in soil as determined by the Baker soil test. Few more mechanistic models with greater input requirements have also been developed to predict the metal ion concentration in soil pore water for risk assessment of metal-contaminated soils (e.g., Lofts and Tipping 1998). But such models are not suitable for routine prediction of metal concentration or activity in soil pore water. The objectives of the present study were (1) to determine the free ion activity of metals, viz., zinc (Zn), copper (Cu), nickel (Ni), cadmium (Cd), and lead (Pb) in contaminated soils using the Baker soil test and (2) to assess the effectiveness of different chemical extractants and forms of organic carbon in predicting the free ion activity of these metals in contaminated soils under the modeling framework.

Materials and methods Collection, processing, and characterization of soil samples Twenty-two bulk surface soil samples (0–15 cm) were collected from various locations of India (Fig. 1; Table 1). Soil samples were air-dried, ground, and passed through a 2-mm sieve. The processed soil samples were used for characterization of initial soil properties. Soil analyses were carried out in triplicate. pH (soil/water ratio=1:2) and electrical conductivity (EC) (Jackson 1973), organic carbon (Walkley and Black 1934), KMnO4-oxidizable carbon (KMOC) (Blair et al. 1995), cation exchange capacity (Bower et al. 1952), and free iron and aluminum oxides (Jackson 1973) in experimental soils were determined using standard procedures. Mechanical composition (texture) of the soil was determined by the hydrometer method (Bouyoucos 1962). Soil samples were digested


10017

Fig. 1 Map showing the origin of the soil samples as used in this study

Table 1 Location and description of experimental soils Locations

No. of samples Description

Sonepat, Haryana

3

Surface soil samples (0–15 cm) were collected from agricultural field adjoining ATLAS cycle factory, Sonepat, which have been receiving industrial effluents for 15 years

Sonepat, Haryana

3

Surface soil samples were collected from the fields of plantation crop (Yamuna action plan) which have been receiving sludge and sewage effluents for 10 years

Keshopur, New Delhi 5

Surface soil samples were collected from Keshopur. These peri-urban agricultural lands have been irrigated with sewage effluents for about three decades under the Keshopur Effluent Irrigation Scheme of the Delhi Government

Okhla, New Delhi

3

Surface soil samples were collected from sewage effluent-irrigated agricultural lands of Madanpur Khadar. These agricultural lands have been irrigated with sewage effluents emanating from Okhla sewage treatment plan for five decades

Debari, Udaipur, Rajasthan

6

Surface soil samples were collected from agricultural lands which have been receiving irrigation through industrial effluents emanating from zinc smelter plants of Hindustan Zinc Limited, Udaipur, Rajasthan

IARI farm, New Delhi 2

One surface soil sample was collected from IARI farm which has been irrigated with sewage effluents for past five decades. Another surface soil sample was collected from IARI farm which has been irrigated with tube well water

10018


with aqua regia (75 % concentrated HCl+25 % HNO3) and analyzed for Zn, Cu, Ni, Cd, and Pb using flame atomic absorption spectrophotometer (Quevauviller 1998). Soil samples were also extracted with ethylenediaminetetraacetic acid (EDTA) (Quevauviller 1998), diethylenetriaminepentaacetic acid (DTPA) (Lindsay and Norvell 1978), and CaCl2 (Houba et al. 2000) and Zn, Cu, Ni, Cd, and Pb in filtrate were analyzed by flame or graphite AAS.

Sodium (Na) and potassium (K) were determined by flame photometry, magnesium (Mg) and calcium (Ca) were determined by AAS after the necessary dilution, manganese (Mn), iron (Fe), Cu, Cd, Ni, and Zn were determined by AAS, and aluminum (Al) was determined colorimetrically with aluminon (Page et al. 1982). Free ionic activity of metals was estimated by feeding these input data to a computer program (FORTRAN-77) (Datta et al. 2013).

Baker soil test equilibrium

Prediction of metal solubility

Intensity factor of metals in experimental soils was estimated according to the procedure of Baker and Amacher (1981).

Solubility of metals in soil was predicted using the following pH-dependent Freundlich equation (Jopony and Young 1994):

Soil test solution DTPA (1.5734 g) was dissolved in 300 ml of distilled water with gentle heating. The solution was transferred quantitatively to a 1-L volumetric flask; subsequently, 10 ml of 0.25 M KCl, 10 ml of 1.0 M MgCl2, and 25 ml of 2.0 M CaCl2·2H2O were added to the DTPA solution in a volumetric flask. The contents were made up to 1 L. Blank preparation A blank soil test solution was prepared by pipetting 40 ml of distilled water, 5 ml of soil test solution, and 5 ml of 0.0275 M triethanolamine (TEA) into a polyethylene container and mixed. The pH of this solution was adjusted to 7.30±0.05 by adjusting the amount of 0.0275 M TEA and distilled water in a final volume of 50 ml at the blank solution.

M2þ ¼

MC −n k M ðHþ Þ M

where (M2+) is the free metal ion activity in soil solution, MC is the labile pool of soil metal assumed to be exclusively adsorbed on humus (in moles per kilogram carbon), and kM and nM are empirical constants which express the pH dependence of the metal distribution coefficient. In the present study, 0.005 M DTPA, 0.05 M EDTA, and 0.01 M CaCl2-extractable Zn, Cu, Ni, Cd, and Pb were used as estimates of labile metal pool, whereas free metal ion activity as determined by the Baker soil test was used as an estimate of free metal ion activity in soil solution. In the case of soil organic carbon, both Walkley–Black organic carbon (WBC) and KMnO4-oxidizable organic carbon were used. The solubility model was parameterized using the Solver facility in Microsoft Excel 2007.

Procedure Results and discussion Five grams of processed soil (air-dried 2 mm sieved soil) was taken in a polyethylene container. The amount of distilled water needed to obtain a pH of 7.3±0.5 and the final volume of 50 ml with no soil present, as determined in the blank preparation step, was added. Then, 5 ml of soil test solution was added. The amount of TEA, as needed in the blank preparation step, was also added. The contents were shaken for 1 h and then allowed to stand for an additional 23 h. A small amount of the blank and sample solution was decanted for pH and EC determination. The remaining solution was filtered and used for analysis.

Initial properties of experimental soils Soil pH and EC were 7.63±0.61 and 1.36±1.72 dS m−1, respectively (Table 2). The WBC was 1.29±0.92 %, and the corresponding figure for KMOC was 1.56 ±1.14 mg g−1. Free Fe and Al oxides were 0.81±0.91 and 0.19±0.03 %, respectively. Cation exchange capacity across the soils was 17.3±3.89 cmol (p+)kg−1, with the clay content of 28.8±6.08 %. There was a wide variation in texture of experimental soils. Experimental soils belong to three textural classes, viz., clay loam,

EC electrical conductivity, CEC cation exchange capacity, WBC Wakley-Black carbon

Figures in parentheses indicate the number of samples that belong to the respective textural class

24.9±1.00 Sandy clay loam (2)

28.8±6.08 – 17.3±3.89

15.6±1.85 0.17±0.01

0.19±0.03 1.29±0.92 1.56±1.14 0.81±0.91 7.63±0.61 1.36±1.72 22 Overall

7.54±0.06 0.73±0.01 2 IARI

0.57±0.03 0.50±0.03 0.66±0.44

31.2±3.39 Clay loam (2), loam (1)

27.9±9.79 Clay loam (3), sandy clay loam (1), loam (2) 16.3±2.14

21.2±5.89 0.21±0.01

0.19±0.03 1.21±0.30 1.46±0.91 0.32±0.34 7.35±0.49 3.05±2.43 6 Debari

7.08±0.62 1.46±0.58 3 Okhla

2.77±1.50 2.61±1.46 0.48±0.02

31.2±1.88 Clay loam (3)

25.9±1.79 Sandy clay, loam (5) 15.3±2.36 0.18±0.04 5 Keshopur

8.35±0.93 0.98±0.51

7.98±0.33 0.26±0.06

1.04±0.76 0.97±1.06 0.34±0.14

33.2±0.94 Clay loam (3) 21.7±0.80

15.5±2.59 0.19±0.01

0.23±0.009 8.37±0.06 0.25±0.04

3

0.76±0.50 1.67±0.96 1.03±0.80

Total and extractable metal content in experimental soils

Sonepat (Yamuna action plan)

1.27±0.14 2.28±0.13 2.76±0.46

10019

loam, and sandy clay loam. Such variation in physicochemical properties of soil is a prerequisite of such type of modeling study.

Sonepat (ATLAS cycle factory) 3

No. of pH samples Location

Table 2 Initial properties of experimental soils

Free Fe oxide Free Al oxide CEC (cmol Clay (%) EC (dS m−1) WBC (%) KMnO4 (%) (p+)kg−1) oxidizable (%) carbon (mg g−1)

Textural class


The aqua regia-extractable (pseudo total) Zn, Cu, Ni, Cd, and Pb contents across the experimental soils were 2,015±3,373, 236±286, 103±192, 29.8±6.04, and 141 ±270 mg kg−1, respectively (Table 3). Extraordinarily high Zn and Cd contents as recorded in Debari soils is attributed to the fact that these soils have been receiving industrial effluents from a zinc smelter industry for a long period. Usually, Zn and Cd have common ores which is reflected in the contents of both of these elements in Debari soils. Zinc, Ni, and Cd contents were unusually high in Sonepat soils which have been receiving industrial effluents from the ATLAS cycle factory. Although elevated levels of Zn, Cu, and Ni were observed in sewage effluent-irrigated soils of Keshopur and Okhla, the magnitude of accumulation of these metals was far lower as compared to that of industrial effluent-irrigated soils. The EDTA-extractable Zn, Cu, Ni, Cd, and Pb in soils were 193±158, 100±126, 31.2 ±48.6, 12.9±22.5, and 65.1±113 mg kg−1, respectively (Table 3). The DTPA-extractable Zn, Cu, Ni, Cd, and Pb in soils were 167±178, 39.8±54.1, 5.07±7.77, 7.30 ±13.2, and 4.28±1.98 mg kg−1, respectively (Table 4). The CaCl2-extractable metals were far less than that extracted by EDTA and DTPA. In general, soil samples were highly polluted with heavy metals, particularly those collected from industrial effluent-irrigated lands. Total Cd content in Sonepat and Debari soils exceeded the permissible limit (3 mg kg−1) as prescribed in the European Council Directive (1986). Total Zn content in soils was above the permissible limit, i.e., 300 mg kg−1, not only in industrial effluent-irrigated soils but also in a few sewage effluent-irrigated soils (e.g., Okhla soils). In the case of Pb, all the experimental soils exhibited a total Pb content well within the permissible limit (300 mg kg−1), except three soils from Debari. In the case of Cu, industrial effluent-irrigated soils of both the locations (Sonepat and Dabari) exhibited far higher total Cu content than what was prescribed (140 mg kg−1) in the European Council Directive (1986). Total Ni content exceeded the permissible limit, i.e., 75 mg kg−1, only in industrial effluent-irrigated soils of Sonepat.

6 2 22

Debari

IARI

Overall

114±94.4

No. of samples

3 3 5 3 6 2 22

Location

Sonepat (ATLAS cycle factory)


Keshopur

Okhla

Debari

IARI

Overall

2.11±1.69

31.3±27.2 133±133

531±108

Zn

19.4±8.27 102±78.8

407±406

103±192

Pb

167±178

8.96±6.74

367±196

118±73.2

43.4±38.3

52.9±57.3

242±18.6

39.8±54.1

5.76±4.63

22.5±17.2

17.5±8.74

16.1±8.98

38.5±45.3

166±46.9

5.07±7.77

0.94±0.29

0.86±0.63

1.44±0.70

2.73±2.13

6.19±7.22

22.6±1.37

7.30±13.2

0.56±0.05

24.4±15.5

0.93±0.23

0.73±0.08

0.77±0.32

1.40±0.73

4.28±1.98

2.71±0.98

5.46±2.58

4.75±1.96

4.03±0.28

2.72±1.03

4.51±0.31

5.07±11.1

0.53±0.41

15.9±17.1

1.70±1.37

0.85±0.84

0.55±0.10

1.18±0.21

0.55±0.58

0.19±0.06

0.29±0.14

1.10±1.17

0.62±0.21

0.77±0.63

0.46±0.13

Cu

Zn

Cu

4.36±3.47 41.9±26.5 174±167

0.13±0.16

0.01±0.009

0.02±0.02

0.31±0.22

0.15±0.12

0.21±0.13

0.23±0.12

Ni

1.42±3.94

0.11±0.01

4.26±6.75

0.23±0.22

0.29±0.24

0.53±0.43

0.05±0.01

Cd

1.36±0.71

0.44±0.31

1.63±0.46

1.49±0.48

1.73±0.70

0.69±0.26

1.33±0.79

Pb

31.2±48.6 12.9±22.5 65.1±113

14.2±2.80 1.89±0.28 0.66±0.03 5.53±3.13 100±126

Zn

Cd

32.0±23.9 5.55±3.29 0.99±0.23 8.57±1.54

CaCl2 (mg kg−1)

903±158

Pb

5.41±0.73 75.2±9.78

Cd

49.6±27.2 4.59±2.81 1.51±0.44 14.1±11.4

DTPA (mg kg−1) Ni

29.8±6.04 141±270

130±10.8

Ni

79.5±99.3 40.5±53.3 1.78±1.57 21.1±25.3

326±147

Cu

2,806±2,037 108±94.6

26.2±2.45 8.77±2.22 0.88±0.02 14.4±3.40 12.0±8.37

2,015±3,373 236±286

113±37.3

Pb

8.01±2.02 133±15.8

Cd

46.3±32.3 16.0±5.87 1.40±0.36 14.9±3.10 72.7±61.8

119±148

549±100

Ni

EDTA (mg kg−1)

85.1±43.8 19.6±3.39 2.19±0.51 22.8±10.3 240±189

5,717±4,236 256±212

467±281

193±232

710±297

Cu

Table 4 DTPA-extractable and CaCl2-extractable metal contents in experimental soils

5 3

Okhla

255±278

Keshopur

1,091±212

3


Zn

No. of samples Total (mg kg−1)

Sonepat (ATLAS cycle factory) 3

Location

Table 3 Total (aqua regia-extractable) and EDTA-extractable metal contents in experimental soils

10020 Environ Monit Assess (2013) 185:10015–10025


10021

The highest amount of metal was extracted by EDTA, followed by DTPA and CaCl2. These results concurred with the findings of Hooda et al. (1997). It is established that neutral salts like CaCl2 cannot extract metal from more tightly bound pools, such as specifically adsorbed, oxide-bound, and organically bound pools (Miller et al. 1986). This extractant is able to extract weakly sorbed metal species, particularly those retained on the soil surface by relatively weak electrostatic interaction and those that can be released through exchangeable processes (Gleyzes et al. 2002). On the other hand, EDTA and DTPA are capable of extracting water-soluble, exchangeable and organically bound metals (O’Connor 1988). Free ion activity of metals in experimental soils Across the experimental soils, free metal ion activity, viz., pZn2+, pCu2+, pNi2+, pCd2+, and pPb2+ was 9.37±1.89, 13.1±1.96, 12.8±1.89, 11.9±2.00, and 11.6±1.52, respectively (Table 5). Activity of Zn was far higher in industrial effluent-irrigated soils of Sonepat and Debari as compared to other soils. An almost similar trend was observed in the case of Cd and Pb. In the case of Cu and Ni, the maximum activity of these metals was observed in intensively contaminated soils of Sonepat. Simple correlation coefficients (r) between free metal ion activity and extractable metals in experimental soils indicated that EDTA-extractable metals had negative and significant correlations with pZn2+ (r = −0.51), pCu2+ (r = −0.79), pNi2+ (r = −0.79), pCd2+ (r = −0.62), and pPb2+ (r = −0.61) (Table 6). Similarly, DTPA-extractable metals showed a negative relationship with pZn2+, pCu2+, pNi2+, pCd2+, and pPb2+, whereas CaCl2-extractable metals did not show any relationship with the intensity of these metal ions. This essentially means that metals extracted by both the chelating agents positively contributed to the free metal

ion activity in soils. The lack of relationship between CaCl2-extractable metals and free metal ion activity may be attributed to the poor extracting ability of metal by CaCl2. Several researchers attempted to characterize the bioavailability of metals in soil using an “intensity factor,” i.e., metal concentration or activity in soil solutions (e.g., Baker and Amacher 1981; François et al. 2004; Datta and Young 2005: Almas et al. 2007). In such studies, extraction of a representative soil aqueous phase is the most important. Ranges of techniques and extraction procedures have been reported in the literature for the extraction of soil solution, which include (1) extraction of soil with water and other weak extractants maintaining soil to solution ratio from 1:2 to 1:10, (2) centrifugation technique to extract soil pore water, (3) squeezing of moist soil, and (4) using Rhizon sampler. However, extraction of soil pore water, subsequent chemical analysis, and speciation using appropriate ion speciation model are tedious, whereas in the Baker soil test, preparation of soil solution extract and subsequent computation of free ion activity of metals are relatively easier. Prediction of metal solubility Model parameters for predicting free ion activity of Zn, Cu, Ni, Cd, and Pb in soil as a function of pH, WBC, and extractable metals were computed (Table 7). Results indicate that variability in free ion activity of Zn, Cu, Ni, Cd, and Pb could be explained by pH, WBC, and EDTA-extractable metal to the extent of 59, 56, 46, 52, and 51 %, respectively (Fig. 2). In the case of Zn and Cu, more or less similar values of prediction coefficient (R2) were obtained for the solubility model based on pH, WBC, and DTPA-extractable metals. This model was less effective in predicting (Ni2+) and (Cd2+) as compared to that based on pH, WBC, and EDTA-

Table 5 Activity of heavy metals in solution (intensity parameter) as determined by the Baker soil test Location

No. of samples

pZn2+

pCu2+

pNi2+

pCd2+

pPb2+

Sonepat (ATLAS cycle factory)

3

6.48±0.92

9.42±1.04

0.97±1.08

9.65±0.97

9.27±1.07


3

10.5±0.84

13.8±0.96

13.4±0.94

13.2±0.42

12.6±0.44

Keshopur

5

10.5±0.59

14.0±0.44

13.6±0.53

13.2±0.27

12.4±0.24

Okhla

3

9.33±0.91

13.7±0.61

13.5±0.62

12.8±0.52

12.1±0.21

Debari

6

8.29±1.67

12.8±1.90

12.7±1.55

10.1±1.95

11.2±1.79

IARI

2

11.3±0.35

14.9±0.30

14.3±0.15

14.1±0.55

12.7±0.25

Overall

22

9.37±1.89

13.1±1.96

12.8±1.89

11.9±2.00

11.6±1.52

10022


Table 6 Simple correlation coefficients (r) between free metal ion activity and extractable metals in experimental soils Metal

Extractant EDTA

DTPA

CaCl2

−0.51a

−0.77b

−0.39

2+

−0.78

−0.81b

0.06

2+

b

pZn2+ pCu

b

−0.79

−0.79

−0.34

pCd2+

−0.62b

−0.60b

−0.31

pPb2+

−0.61b

−0.60b

−0.29

pNi

b

a

Values of r are significant at the 5 % probability level

b

Values of r are significant at the 1 % probability level

extractable metals. Among these metals, the solubility model based on pH, WBC, and CaCl2 extractable metals could successfully predict the free ion activity of Zn only. The KMOC-based models were far less effective in predicting free metal ion activity as compared to those based on WBC (Table 7). The predictability of the solubility model based on extractable metal by chelating agents, i.e., EDTA and DTPA, was far better than that based on CaCl2-extractable metal. This may be attributed to the fact that

these chelating agents extract metal from exchangeable and organically bound pools, whereas CaCl2 is capable of extracting metals only from the exchangeable pool, and those metals which are weakly bounded on soil colloids. These findings may derive indirect support from the literature where several researchers could successfully use chelating agents in assessing the bioavailability of metals in contaminated soils (Hooda et al. 1997; Datta and Young 2005). In case of Zn and Cu, models based on EDTA-extractable and DTPA-extractable metals were almost equally effective in predicting the free metal ion activity in soil, while in the case of other metals, the EDTA-based solubility model was more effective as compared to that based on DTPA. Lindsay and Norvell (1978) developed the DTPA soil test to assess the deficiency of Zn, Cu, Fe, and Mn in near-neutral and calcareous soils with a 2:1 soil to solution ratio; the capacity of DTPA to complex each of micronutrient cations (express in parts per million of metal on dry weight soil basis) is ten times its atomic weight and ranges from 550 to 650 ppm depending on micronutrient cations. Hence, this extractant might have been saturated with metals and become less effective, particularly in the case of Ni, Cd, and Pb in intensively metal-contaminated

Table 7 Model parameters for the prediction of free ion activity of Zn, Cu, Ni, Cd, and Pb in soil as a function of pH, organic carbon, and extractable metals Metal

Extractant EDTA

DTPA

Model parameters Log kM

R2

nM

CaCl2


nM

R2


nM

R2

WBC Zn

8.58

0.02

0.59

7.05

0.17

0.62

7.69

−0.14

0.31

Cu

12.5

−0.09

0.56

10.8

0.08

0.56

10.1

−0.03

0.02

Ni

10.4

0.04

0.46

10.1

0.02

0.31

11.7

−0.42

0.11

Cd

6.28

0.40

0.52

6.81

0.29

0.31

8.47

−0.01

0.03

Pb

8.84

0.09

0.51

7.72

0.15

0.02

7.22

0.15

0.000008

KMOC Zn

9.66

0.008

0.42

8.13

0.16

0.39

8.77

−0.15

0.12

Cu

13.6

−0.10

0.34

11.9

0.07

0.34

11.1

−0.04

0.11

Ni

11.5

0.03

0.29

11.1

0.01

0.13

12.8

−0.43

0.04

Cd

7.36

0.38

0.36

7.89

0.28

0.18

9.55

−0.03

0.003

Pb

9.92

0.08

0.35

8.80

0.14

0.03

8.30

0.13

0.05

The values of R2 >0.17 are significant at the 5 % probability level


10023

12

15

11

14

10

13 Predicted pCd2+

Predicted pZn2+

R² = 0.59

9 8 7

R² = 0.52

12 11 10 9

6

8

5 5

6

7

8

9

10

11

8

12

9

10

11

12

13

14

15

Observed pCd2+

Observed pZn2+

16

14 R² = 0.51

R² = 0.56

15

13 12

13 Predicted pPb2+

Predicted pCu2+

14

12 11 10

11 10 9

9

8

8 8

9

10

11

12

13

14

15

16

8

9

10

11

12

13

14

Observed pPb2+

Observed pCu2+

15 R² = 0.46 14

Predicted pNi2+

13 12 11 10 9 8 8

9

10

11

12

13

14

15

Observed pNi2+

Fig. 2 Comparison of observed and predicted free ion activity of metals in soil on 1:1 line; (M2+) was predicted by the solubility model as a function of soil pH, WBC, and EDTA-extractable metals

soils as contents of these metals were far lower than Zn and Cu in most of the experimental soils. On the other hand, ten times higher concentration of EDTA than DTPA was used in the EDTA soil test, and this extractant was specially designed for estimating the bioavailability of metal in contaminated soil by the European Commission (Quevauviller 1998).

In general, the solubility model based on WBC had higher values of prediction coefficients as compared to KMOC. Soil organic matter is a heterogeneous mixture of materials ranging from fresh plant and microbial residues to relatively inert humic compounds. The recovery of organic carbon in the case of the Walkley–Black method is much higher as compared to that of the KMnO4

10024

oxidizing method. This is ascribed to the fact that chromic acid is a more powerful oxidizing agent at higher temperature (influenced by heat of dilution) as compared to the neutral KMnO4. Therefore, soil organic carbon oxidized by 333 mM KMnO4 has been considered as a useful index of labile soil organic carbon (Blair et al. 1995; Datta et al. 2010). This fraction encompasses all readily oxidizable organic components, including humic materials and polysaccharides. This pool of soil organic carbon has been proved to be very important as far as nutrient-supplying capacity of soil is concerned (Verma et al. 2010). In the present study, there is an indication that some portion of carbonized/recalcitrant materials is also important for adsorption of metal in addition to the labile pool. Hence, the predictability of the solubility model based on WBC is higher as compared to that based on KMOC. In several studies, it is established that the solubility of metals in soil mainly depends on soil pH and organic carbon (Jopony and Young 1994; Datta and Young 2005; Hough et al. 2005; Rattan et al. 2005; Gandois et al. 2010; Ivezic et al. 2012). In most of these cases, soil solution was extracted with centrifugation or Rhizon sampling technique and free ion activity of metals was determined using appropriate ion speciation model (GEOCHEM, WHAM, etc.). In the present study, free ion activity of metals was estimated using the Baker soil test which was far easier to adopt on a routine basis.

Conclusions It can be concluded that pH and organic carbon are among the important soil properties which control the solubility of metals in contaminated soils. The EDTAextractable metals were more useful as an input of solubility model as compared to DTPA-extractable and CaCl2-extractable metals. WBC was more useful as a predictor variable of the solubility model than the labile pool of organic carbon. The solubility model as a function of pH, organic carbon, and EDTA-extractable metals was reasonably effective in predicting metal ion activity in contaminated soils. References Almas, A. R., Lombnaes, P., Song, T. A., & Mulder, J. (2005). Speciation of Cd and Zn in contaminated soils assessed by DGT-DIFS, and WHAM/Model VI in relation to uptake by spinach and ryegrass. Chemosphere, 62, 1647–1655.

Environ Monit Assess (2013) 185:10015–10025 Almas, A. R., Lofts, S., Tipping, E., & Mulder, J. (2007). Solubility of major cations and Cu, Zn and Cd in soil extracts of some contaminated agricultural soils near zinc smelter in Norway: Modeling with a multisurface extensions of WHAM. European Journal of Soil Science, 58, 1074–1086. Baker, D.E. & Amacher, M.C. (1981). Development and interpretation of a diagnostic soil testing program. Pennsylvania Agricultural Experiment Station Bulletin No. 826, Pennsylvania Agricultural Experiment Station, University Park, Pennsylvania, 20 pp. Blair, G. J., Lefroy, R. D. B., & Lisle, L. (1995). Soil carbon fractions based on their degree of oxidation, and the development of a carbon management index for agricultural systems. Australian Journal of Soil Research, 46, 1459–1466. Bouyoucos, G. J. (1962). Hydrometer method improved for making particle size analysis of soil. Agronomy Journal, 54, 464– 465. Bower, C. A., Reitemeier, R. F., & Firemen, M. (1952). Exchangeable cation analysis of saline and alkaline soils. Soil Science, 73, 251–261. Datta, S. P., & Young, S. D. (2005). Predicting metal uptake and risk to the human food chain from leaf vegetables grown on soils amended by long-term application of sewage sludge. Water, Air, and Soil Pollution, 163, 119–136. Datta, S. P., Rattan, R. K., & Chandra, S. (2010). Labile soil organic carbon, soil fertility and crop productivity as influenced by manure and mineral fertilizers in the tropics. Journal of Plant Nutrition and Soil Science, 173, 715–726. Datta, S. P., Meena, B. L., & Rattan, R. K. (2013). Development of a computer program for calculating metal ion activity using Baker soil test. Journal of the Indian Society of Soil Science, 61, 47–50. European Council Directive. (1986). European Community Council Directive of 12 June 1986 on the protection of the environment, and in particular the soil, EEC (OJ L 181, 4.7.1986, pp. 6–12). François, M., Dubourguier, H. C., Li, D., & Douay, F. (2004). Prediction of heavy metal solubility in agricultural topsoils around two smelters by the physico-chemical parameters of the soils. Aquatic Sciences, 66, 78–85. Gandois, L., Probst, A., & Dumat, C. (2010). Modeling trace metal extractability and solubility in French forest soils by using soil properties. European Journal of Soil Science, 61, 271–286. Gleyzes, C., Tellier, S., & Astruc, M. (2002). Fractionation studies of trace elements in contaminated soils and sediments: A review of sequential extraction procedures. Trend in Analytical Chemistry, 21, 451–467. Groenenberg, J. E., Romkens, P. F. A. M., Comans, R. N. J., Luster, J., Pampura, T., Shotbolt, L., Tipping, E., & de Vries, W. (2010). Transfer functions for solid-solution partitioning of cadmium, copper, nickel, lead and zinc in soils: Derivation of relationship for free metal ion activities and validation with independent data. European Journal of Soil Science, 61, 58–73. Hooda, P. S., McNulty, D., Alloway, B. J., & Aitken, M. N. (1997). Plant availability of heavy metals in soils previously amended with heavy applications of sewage sludge. Journal of the Science of Food and Agriculture, 73, 446–454. Houba, V. J. G., Temminghoff, E. J. M., Gaikhorst, G. A., & van Vark, W. (2000). Soil analysis procedures using 0.01 M calcium chloride as extraction reagent. Communications in Soil Science and Plant Analysis, 31, 1299–1396.

Environ Monit Assess (2013) 185:10015–10025 Hough, R. L., Tye, A. M., Crout, N. M. J., McGrath, S. P., Zhang, H., & Young, S. D. (2005). Evaluating a “free ion activity model” applied to metal uptake by Lolium perenne L. grown in contaminated soils. Plant and Soil, 270, 4–12. Ivezic, V., Almas, A. R., & Singh, B. R. (2012). Predicting the solubility of Cd, Cu, Pb and Zn in uncontaminated Croatian soils under different land uses by applying established regression models. Geoderma, 170, 89–95. Jackson, M. L. (1973). Soil chemical analysis. New Delhi: Prentice Hall of India Private Limited. Jopony, M., & Young, S. D. (1993). Assessment of lead availability in soils contaminated by mine spoil. Plant and Soil, 151, 273–278. Jopony, M., & Young, S. D. (1994). The solid solution equilibria of lead and cadmium in polluted soils. European Journal of Soil Science, 45, 59–70. Lindsay, W. L., & Norvell, W. A. (1978). Development of a DTPA soil test for zinc, iron, manganese and copper. Soil Science Society of America Journal, 42, 421–428. Lofts, S., & Tipping, E. (1998). An assemblage model for cation binding by natural particulate matter. Geochimica et Cosmochimica Acta, 62, 2069–2625. Loncaric, Z., Popovic, B., Karalic, K., Rekasi, M. & Kovacevic, V. (2010). Regression model for prediction availability of essential heavy metals in soils. In: Gilkes, RJ., Prakongkep, N. (Eds.), Proceedings of the 19th World Congress of Soil Science, Soil Solutions for a Changing World. IUSS (International Union of Soil Sciences), Brisbane, Australia. ISBN: 978-0-646-53783-2, pp. 92–95. Published on DVD. Retrieved from http://www.iuss.org. Miller, W. P., Martens, D. C., & Zelazny, W. (1986). Effect of sequence in extraction of trace metals from soils. Soil Science Society of America Journal, 50, 598–601.

10025 O'Connor, G. A. (1988). Use and misuse of DTPA soil test. Journal of Environmental Quality, 17, 715–718. Page, A.L., Miller, R.H. & Keeney, D.R. (1982). Methods of soil analysis, part 2. Chemical and microbiological properties. Agronomy Monograph, 2nd ed. Madison: ASA and SSSA. Quevauviller, P. (1998). Operationally defined extraction procedures for soil and sediment analysis. Trends in Analytical Chemistry, 17, 289–298. Rattan, R. K., Datta, S. P., Chhonkar, P. K., Suribabu, K., & Singh, A. K. (2005). Long-term impact of irrigation with sewage effluents on heavy metal content in soils, crops and groundwater, a case study. Agriculture, Ecosystems and Environment, 109, 310–322. Travis, C. C., & Etnier, E. L. (1981). A survey of sorption relationships for reactive solutes in soil. Journal of Environmental Quality, 10, 8–17. Tye, A. M., Young, S. D., Crout, N. M. J., Zhang, H., Preston, S., Barbosa-Jefferson, V. L., Davison, W., Mcgrath, S. P., Paton, G. I., Kilham, K., & Resende, L. (2003). Predicting the activity of Cd2+ and Zn2+ in soil pore water from the radio-labile metal fraction. Geochimica et Cosmochimica Acta, 67, 375–385. Verma, B. C., Datta, S. P., Rattan, R. K., & Singh, A. K. (2010). Monitoring changes in soil organic carbon pools, nitrogen, phosphorus and sulfur under different agricultural management practices in the tropics. Environmental Monitoring and Assessment, 171, 579–593. Walkley, A., & Black, I. A. (1934). An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Science, 37, 29–38.