Transformations of heavy metals added to soil — application of a new sequential extraction procedure

Transformations of heavy metals added to soil — application of a new sequential extraction procedure

Geoderma 84 Ž1998. 157–168 Transformations of heavy metals added to soil — application of a new sequential extraction procedure Y.B. Ma ) , N.C. Uren...

147KB Sizes 0 Downloads 13 Views

Recommend Documents

No documents
Geoderma 84 Ž1998. 157–168

Transformations of heavy metals added to soil — application of a new sequential extraction procedure Y.B. Ma ) , N.C. Uren School of Agriculture, La Trobe UniÕersity, Bundoora, Bundoora, Vic. 3083, Australia Accepted 23 July 1997

Abstract A sequential extraction procedure, used to remove the heavy metals in specifically adsorbed and easily reducible manganese ŽMn. oxide fractions, was used to study the transformation of heavy metals added to an alkaline soil. Most of the endogenous Cu Ž86%. and Pb Ž79%. were found in the residual fraction ŽRES. which was considered to be mainly highly crystalline Fe oxides and silicate minerals. The recently added Cu, Pb and Cd existed mainly on the surfaces of the soil particles as reactive fractions Žwater-soluble plus exchangeable and NaCaHEDTA-extractable fractions. and as highly stable forms ŽRES fraction.. There was a particularly high affinity of Pb for Mn oxides. The concentrations of metals in the reactive fractions were in the order: Cd ) Cu ) Pb. When water-soluble heavy metals are added to the soil, they are rapidly retained by the soil. The reactive forms then slowly transform into highly stable forms. The processes associated with the transformation of added Cu and Pb can be described by a diffusion equation. The processes may be attributed mainly to the diffusion of the surface species into micropores and the entrapment in microporous solids. Unlike Cu and Pb, most of the exchangeable Cd transformed to the forms extracted with NaCaHEDTA and to residual forms. The slow processes of the transformation of Cd added to soil may be attributed to inner-sphere surface complexation via partial or complete dehydration of surface species. The relative diffusion rate coefficients Ž Drr 2 . were found to be of the order of 10y10 to 10y11 sy1. Addition of CaCO 3 decreased the reactivity or extractability of added heavy metals through the increase in pH. q 1998 Elsevier Science B.V. All rights reserved. Keywords: transformation; fractionation; heavy metals; soil

)

Corresponding author. Fax: q61 3 9479 2163; E-mail: [email protected]

0016-7061r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 1 6 - 7 0 6 1 Ž 9 7 . 0 0 1 2 6 - 2

158

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

1. Introduction The forms of heavy metals in soils strongly depend on their nature and origin. The endogenous heavy metals frequently exist in solid phases Ž Miller et al., 1986; Singh et al., 1988; Ramos et al., 1994. . The anthropogenic heavy metals mainly exist on the surfaces of soils as reactive forms Ž Hickey and Kittrick, 1984; Levy et al., 1992; McGrath and Cegarra, 1992; Ramos et al., 1994. . The heavy metals in soils may occur in the following forms: water-soluble, exchangeable, specifically adsorbed, associated with insoluble organic matter, carbonate, oxides of Fe, Al and Mn, and layer silicates Ž Beckett, 1989. . Sequential extraction procedures have been commonly used to determine the forms or associations of heavy metals in polluted or sludge-amended soils ŽChang et al., 1984; Hickey and Kittrick, 1984; Levy et al., 1992; Keller and Vedy, ´ 1994; Ramos et al., 1994.. Few schemes attempt to fractionate the heavy metals specifically adsorbed and those associated with easily reducible Mn oxides. As a result, high concentrations of Cd in contaminated soils which were extracted with 1 M NaOAc-HOAc at pH 5.0 following the extraction with 1 M MgCl 2 Ž Hickey and Kittrick, 1984; Ramos et al., 1994. probably came from not only the carbonate bound fraction but also from the specifically adsorbed ŽBelzile et al., 1989. and easily reducible Mn oxide bound fractions. The specifically adsorbed heavy metals were also extracted by pyrophosphate which was used for the selective extraction of organic matter bound fraction Ž Han et al., 1995. and by 0.1 M NH 2 OH P HCl in an acidic medium which was used for the Mn oxide fraction Ž Keller and Vedy, ´ 1994.. In these cases, it is difficult to precisely define whether the heavy metals are associated with either one or the other soil components. A new fractionation scheme that selectively removes the heavy metals in specifically adsorbed and easily reducible Mn oxide fractions has been developed to minimise these problems in the sequential extraction procedure ŽMa and Uren, 1995. . The anthropogenic heavy metals generally are adsorbed on the surfaces of soils via outer-sphere and inter-sphere surface complexation Ž Sposito, 1984. and then very slowly transform to highly stable forms with time. The mechanisms of decreasing extractability of heavy metals added to soils with time are unclear, although they are of importance in the evaluation of their toxicity and in remediation of contaminated soils. Hence, the new scheme of sequential extraction with fewer limitations was used to study the transformations of added Cu, Pb and Cd to highly stable forms in an alkaline soil as influenced by CaCO 3. 2. Materials and methods A topsoil sample Ž 0–20 cm. of dark kastanozem was collected from Hebei, North China. The soil is classified as a Mollisol according to Soil Taxonomy

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

159

ŽUSDA, 1988.. The soil pH Ž H 2 O, 1 : 2.5. was 8.28; the organic matter was 25.2 g kgy1 ; clay Ž- 1 mm. was 171 g kgy1 ; the concentrations of Fe, Al and Mn extracted with 0.175 M ŽNH 4 . 2 C 2 O4 –0.001 M H 2 C 2 O4 ŽpH 3.25. were 2.52, 1.51 and 0.26 g kgy1, respectively. CaCO 3 in the topsoil was not detected by volumetric calcimeter. The layer silicates are predominantly illite and smectite ŽXu, 1990.. The total concentrations of Cu, Pb and Cd were determined after digestion with a mixture of concentrated H 2 SO4 –HClO4 –HNO 3 –HF acids and were found to be 15.5, 14.5 and 0.78 mg kgy1, respectively. The soil was air-dried and passed through a 2.0 mm plastic sieve. In order to determine the effects of CaCO 3 on the forms and transformation of added heavy metals, CaCO 3 Ž powder. was added to 50 g soil at the rates of 0, 10, 30, and 50 g CaCO 3 kgy1. A 10 ml aliquot of 250 mg ly1 Cu, Pb and Cd as chloride was mixed with the soil at the rate of 50 mg kgy1. After incubation at 80% of field capacity Žwater content at 20%. and at 25 " 28C for 1, 15 Žonly for DTPA extraction., 30, and 60 days, samples were air-dried and passed through a 2.0 mm plastic sieve. Samples were fractionated in duplicate using the new sequential extraction procedure Ž Ma and Uren, 1995. with minor modifications. The chemical reagents, the conditions and the corresponding fractions determined are designated in Table 1. The EDTA-extractable metals were considered to be associated with the specifically adsorbed metals. In order to compare with other procedures, the samples were also fractionated using an old sequential extraction procedure ŽTable 2. which is similar to the scheme of Tessier et al. Ž 1979. .

Table 1 The new sequential extraction procedure and the corresponding forms Formrassociation

Abbr.

Step

Operational definition

Soil solution

WS

1

Distilled water, 1 : 5, shaking 30 min

Exchangeable

EXC

2

1 M MgCl 2 , pH 7.0, shaking 1 h

EDTA-extractable

EDTA

3

1% NaCaHEDTA in 1 M NH 4OAc, pH 8.3, 1 : 10, shaking 2 h

Easily reducible Mn

ERMn

4

0.2% quinol in 1 M NH 4OAc, pH 7.0, 1 : 10, shaking 1 h

Carbonate

CA

5

0.5 M NaOAc–0.5 M HOAc, pH 4.74, 1 : 10, soaking 15 h and shaking 3 h

Organic matter

OM

6

5 ml 30% H 2 O 2 , pH 4.74, digested twice at 858C and extracted by 0.5 M NaOAc–0.5 M HOAc for 1 h

Fe and Al oxides

FeOx

7

0.175 M ŽNH 4 . 2 C 2 O4 –0.100 M H 2 C 2 O4 , pH 3.25, 1 : 10, soaking 15 h and shaking 2 h in daylight

Residual forms

RES

8

Total minus sum of the extractable

160

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

Table 2 The old sequential extraction procedure and the corresponding forms Formrassociation

Step

Operational definition

Soil solution ŽWS.

1

Distilled water, 1 : 5, shaking 30 min

Exchangeable ŽEXC.

2

1 M MgCl 2 , pH 7.0, shaking 1 h

Carbonate Ž‘CA’.

3

0.5 M NaOAc–0.5 M HOAc, pH 4.74, 1 : 10, soaking 15 h and shaking 3 h

Fe and Al oxides Ž‘FeOx’.

4

0.175 M ŽNH 4 . 2 C 2 O4 –0.100 M H 2 C 2 O4 , pH 3.25, 1 : 10, soaking 15 h and shaking 2 h in daylight

Organic matter Ž‘OM’.

5

5 ml 30% H 2 O 2 , pH 3.25, digested twice at 858C and extracted as for Fe and Al oxides fraction Ž2 h.

Residual forms Ž‘RES’.

6

Total minus sum of the extractable

The extractions were conducted in duplicate in 100 ml polypropylene centrifuge tubes. Between each successive extraction, the supernatant was obtained by centrifuging Ž 15 min, 3000 rpm., decantation and filtering on a Whatman 40 filter paper. The metals in the entrained solution in each extraction were accounted for by weighing the tube to estimate the volume of entrained liquid ŽMa and Xia, 1988. . The concentrations of Cu, Pb, and Cd in the supernatants were measured using an atomic absorption spectrophotometer. The recovery of the two sequential extraction procedures for Zn Ž sum of all fractionsrtotal Zn = 100%. has been found to be in the range of 94–105% Ž Ma and Xia, 1988; Ma and Uren, unpubl.. so that in this experiment, it was deemed satisfactory to determine the concentrations of heavy metals in residual fraction by calculation. The concentrations of added Cu, Pb and Cd in each fraction were calculated by the concentration in each fraction in soil with added heavy metals, minus the concentration in that fraction in soil without added heavy metals. The DTPA-extractable Cu, Pb and Cd in the soil samples Ž 10 g. were extracted by 20 ml of 0.005 M DTPA–0.1 M triethanolamine–0.01 M CaCl 2 at pH 7.3 by shaking for 2 h ŽLindsay and Norvell, 1978..

3. Results and discussion 3.1. EÕaluation of the fractionation scheme The new fractionation scheme was compared with an old sequential extraction procedure similar to the scheme of Tessier et al. Ž1979.; the latter does not estimate specifically adsorbed and easily reducible Mn oxide fractions. In the old method, specifically adsorbed heavy metals were released in the extractants

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

161

Table 3 The concentration of metals in individual fractions determined by the sequential extraction procedure with Žnew. and without Žold. provision for specifically adsorbed and easily reducible Mn oxide fractions Žmg kgy1 . a Method

Treatment

EDTA

ERMn

CA

OM

New Old

SoilqCu SoilqCu

31.5 –

1.0 –

0.1 1.1

New Old

SoilqPb SoilqPb

28.7 –

6.0 –

New Old

SoilqCd SoilqCd

11.5 –

0.9 –

a

FeOx

RES

1.5 0.6

2.1 36.3

29.1 27.5

0.6 2.5

0.6 10.1

4.5 14.5

22.9 36.1

0.4 3.9

0.9 0.6

0.4 1.7

10.0 17.8

See Tables 1 and 2 for abbreviations. Means of three samples with different incubation time.

which were used to remove the selectively of the heavy metals associated with soil components Ž Table 3. . For example, the concentration of Cu in the fraction ŽFeOx. which was extracted with 0.175 M ŽNH 4 . 2 C 2 O4 –0.100 M H 2 C 2 O4 following the extraction with 0.5 M NaOAc–0.5 M HOAc was 36.3 mg kgy1 in the old method, but it is only 2.1 mg kgy1 measured by the new fractionation method. The Cu extracted with 0.175 M ŽNH 4 . 2 C 2 O4 –0.100 M H 2 C 2 O4 in the soils without removal of the specifically adsorbed fraction was associated not only with Fe ŽAl. and Mn oxides but also with organic matter and layer silicates; it is difficult to define this Cu fraction. One option to minimise the non-selectivity in a sequential extraction procedure is to extract most of the adsorbed cations prior to dissolution of the soil solid phases. The NaCaHEDTA at high pH was chosen to extract the specifically adsorbed heavy metals. By comparison, the new sequential extraction procedure is found to be satisfactory for studying the transformation of heavy metals added to soils because it has greater discrimination than other sequential extraction procedures. 3.2. Form r association of heaÕy metals added to soil Most of the endogenous Cu Ž86%. and Pb Ž79%. were found predominantly in the residual fraction ŽRES. which is considered mainly to be highly crystalline Fe oxides and silicate minerals. The recently added Cu, Pb and Cd existed mainly on the surfaces of soil particles as reactive forms ŽEXC and EDTA fractions. and as highly stable forms RES fraction Ž Table 4. . The proportions of exchangeable metals were found in the order: Cd 4 Cu ) Pb. Similarly, McGrath and Cegarra Ž 1992. studied the chemical extractability of heavy metals in soils amended with sewage sludge and reported that the concentration of Cd extracted with 0.1 M CaCl 2 was much higher than those of

162

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

Table 4 The distribution Ž%. of recently added heavy metals in individual fractions a Metal

The order of distribution

Cu Pb Cd

EDTA Ž60. )RES Ž33. ) FeOx Ž3.6. )OM Ž2.7. ) others Ž0.9. EDTA Ž57. )RES Ž26. ) ERMn Ž10. ) FeOx Ž5.0. ) others Ž2.0. EXC Ž48. ) EDTA Ž26. )RES Ž21. ) others Ž5.6.

a

See Table 1 for abbreviations. Means of three samples with different incubation time.

Zn, Cu and Pb. These results reflect the high reactivity of added Cd. The species of exchangeable Cd were probably hydrated Cd 2q and adsorbed via outer-sphere surface complexation ŽNaidu et al., 1994. . The proportions of EDTA-extractable metals were found in the order: Cu G Pb ) Cd. The EDTA-extractable fraction was considered to be inner-sphere surface complexed forms. At the pHs of these systems, the predominant species of EDTA-extractable Cu and Pb were probably hydroxy-cations Žpartially or completely dehydrated. and sorbed via innersphere surface complexation. The proportion of Pb in the ERMn fraction was found to be higher than those of Cu and Cd. The particularly high affinity of Mn oxides for Pb has been studied using adsorption methods Ž McKenzie, 1980; Aualiitia and Pickering, 1987.. An explanation of this fact may be the residence of Pb in the tunnel structures of some Mn oxides. Large tunnels in Mn oxides, such as hollandite ŽBa,Pb,Na,K. 2yx ŽFe,Mn. 8 wOŽOH.x16 Ž or ideally, a-MnO 2 ., could accommodate large cations, such as Pb Ž Waychunas, 1991. . In a study of mimicked in situ remediation of soils contaminated with heavy metals, Mench et al. Ž 1994. showed that addition of hydrous Mn oxides caused decreases in concentrations of Cd and Pb in relatively reactive fractions and in ryegrass shoots. The Mn oxides may be a potential additive for in situ remediation of soils contaminated with heavy metals, especially with Pb. Very little Cu or Cd was tightly associated with Fe Ž Al. and Mn oxides in soils. In contrast, the synthetic Fe and Mn oxides have been shown to have a high affinity for Cu and Cd. For example, Gerth et al. Ž1993. showed that 28% of Cd which reacted with synthetic goethite associated with Si was not extracted with 0.7 M HNO 3 for 14 days, whereas 20% of the total Fe of the goethite and 39% of the associated Si had been dissolved. There is a remarkable difference between the extractability of heavy metals associated with Fe oxides from soils and synthetic ones. This fact may be explained by the different properties of the oxides, the adherence of organic matter on the oxides and the complexation of organic matter with Cu and Cd. Xue and Huang Ž1995. reported that the extractability of Zn Ž0.005 M DTPA at pH 7.3. adsorbed on Fe oxide, formed in the presence of citric acid, was higher than that formed in the absence of citric acid.

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

163

Table 5 The concentration of the recently added metals Ž50 mg kgy1 . in individual fractions in different incubation time Žmg kgy1 . a Metal

Time Ždays.

WS

EXC

EDTA

ERMn

CA

OM

FeOx

RES

Cu

1 30 60

0.20 nd b nd

0.24 nd nd

31.0 29.5 29.1

0.18 0.29 0.31

0.09 0.06 0.05

1.09 1.39 1.67

1.07 2.34 1.97

16.4 16.7 17.1

LSD 0.05 c

0.12

0.21

0.18

0.16

0.36

0.51

1 30 60

0.14 nd nd

0.56 nd nd

4.65 5.52 5.61

0.48 0.20 0.58

0.24 0.46 0.70

1.83 2.93 2.67

LSD 0.05

0.09

0.31

0.37

0.42

0.49

0.51

1 30 60

0.10 nd nd

0.99 1.34 1.45

0.46 0.56 0.71

0.51 0.78 0.90

0.22 0.23 0.27

LSD 0.05

0.11

0.36

0.35

0.39

0.21

Pb

Cd

26.9 22.4 22.6 0.46

0.24 29.6 28.3 27.1 0.42 11.4 14.0 13.8 0.51

1.54 12.5 13.0 13.5 1.32 9.6 10.9 10.4 1.34

a

See Table 1 for abbreviations. Means of four samples with different CaCO 3 contents. nd s not detected Ž - 0.05 mg kgy1 .. c The least significant difference at the 0.05 level of probability.

b

3.3. Transformations of heaÕy metals added to soil The water-soluble heavy metals added to soils were rapidly retained, then the reactive forms slowly transformed or transferred into highly stable forms ŽTable 5.. The results Ž Table 5. show that the WS q EXC-Cu and EDTA-Cu trans-

Fig. 1. The change of concentrations of Cu, Pb and Cd in different fractions with time Žsee Table 1 for abbreviations and Table 5 for data..

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

164

formed into the OM-Cu and FeOx-Cu and that the WS q EXC-Pb and EDTA-Pb transformed into the ERMn-Pb and FeOx-Pb. Unlike Cu and Pb, the EXC-Cd transformed mainly to EDTA-Cd Ž Fig. 1. . Mann and Ritchie Ž1994. used a sequential extraction procedure to study the changes in the forms of Cd with time Ž 8 days. in some soils and showed that the concentration of water-soluble Cd or Cd exchangeable with BaCl 2 decreased with time and transformed to less reactive forms. For Cu, the proportion of applied Cu as fertilizer and extractable in 0.005 M EDTA in a long-term trial on a lateritic sandy soil decreased with time for 20 years Ž McLaren and Ritchie, 1993.. However, the mechanisms of such processes have not been documented. When the overall rate of the processes which lead to decreases in the extractability of heavy metals added to soils are controlled by diffusion into micropores of soils and the diffusion is considered to be radial assuming that the aggregates are spherical, then the diffusion equation for a constant diffusion coefficient takes the form:

EC

sD

ž

E 2C 2

2 EC q

/

Et Er r Er where C is concentration of diffusing substance, D is the diffusion coefficient and r is the radius of the spherical particle. If the surface concentration is constant, the diffusion equation in the early stages can be expressed as follows ŽCrank, 1975; Karger and Ruthven, 1992. : ¨ X Xm

s A q B't

Bs

6

'p



D r2

/

where X is the concentration of heavy metals which diffuse into micropores, Xm is the total concentration of heavy metals added to soils, t is time, and A is a constant which may present the proportion of the heavy metals tightly retained by surface complexation but unextractable with one EDTA extraction. Thus a plot of XrXm versus 6t is approximately linear as far as XrXm s 1r2 ŽCrank, 1975.. The Drr 2 expresses the relative diffusion rate coefficient. In a strict sense the condition of constant concentration cannot be fulfilled in these incubation experiments. However, if the experimental data permits the calculation of a linear regression line according to the diffusion equation, then this can be taken as an indication that the process or processes are controlled by diffusion of heavy metals into micropores of soil ŽBruemmer et al., 1988. . If the heavy metals in the reactive forms Ž EXC q EDTA. are considered to exist on the surfaces of soils as exchangeable and specifically adsorbed fractions, the heavy metals which are transformed or transferred into the unreactive forms ŽERMn, CA, OM, FeOx and RES. are considered to be associated mainly with the forms of heavy metals in micropores. The processes associated with the transformation of reactive Cu and Pb Ž EXC q EDTA. to unreactive Cu and Pb can be described by the diffusion equation ŽFig. 2.. The results suggest that the

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

165

Fig. 2. The proportions of added Cu, Pb and Cd in the unreactive fraction ŽERMn, CA, OM, FeOx and RES. as a function of square root of time Ždiffusion equation.. The symbol ) indicates the correlation coefficient Ž R . significant at 0.05 level of probability.

slow processes of transformations of Cu and Pb in the EXCq EDTA fractions to unreactive forms may be attributed mainly to the diffusion of the surface species into micropores and the entrapment in microporous solids. The relative diffusion rate coefficients Ž Drr 2 . were found to be of order of 10y10 to 10y11 sy1. The transformation of Cu and Pb retained by soils may be mainly related to the ‘groovy structure’ on the surfaces of oxides such as MnO 2 andror to the microporosity of soil components such as layer silicates. Unlike Cu and Pb, most exchangeable Cd transformed to the EDTA-Cd. Compared with Cu and Pb, there is low linear correlation coefficients for the transformation of Cd described by the diffusion equation Ž Fig. 2. . The slow

Fig. 3. The proportion Ž%. of metals added to soil in DTPA-extractable fraction as a function of time Žmeans of four samples with different CaCO 3 contents..

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

166

Fig. 4. The proportions of added Cu and Pb in the DTPA-unextractable fraction Žnon-DTPA. as a function of square root of time Ždiffusion equation.. The symbols ) and ) ) indicate the correlation coefficient Ž R . significant at 0.05 and 0.01 levels of probability, respectively.

processes of the transformation of Cd added to soil may be attributed to inner-sphere surface complexation via partial or complete dehydration of the surface species ŽFig. 1.. The proportions of recently added Cu and Pb in the DTPA-extractable fraction decreased markedly with time, while that of Cd decreased barely with time ŽFig. 3.. The processes associated with the decrease in extractability in DTPA of Cu and Pb except for Cd added to soil can be described also by the

Table 6 The concentration of the recently added metals Ž50 mg kgy1 . in individual fractions as affected by CaCO 3 Žmg kgy1 . a Metal Cu

Pb

Cd

a b

CaCO 3 Ž%. 0 1 3 5 0 1 3 5 0 1 3 5

pH CaCl 2 7.08 7.55 7.68 7.68 7.08 7.55 7.68 7.68 7.08 7.55 7.68 7.68

WS 0.1 0.1 0.1 0.1 0.1 nd nd 0.1 0.1 nd nd nd

EXC b

nd 0.1 0.1 0.1 0.5 nd nd nd 26.5 24.1 23.0 22.5

EDTA

ERMn

CA

OM

FeOx

RES

30.9 30.1 29.4 29.0 28.7 28.9 28.1 27.6 11.4 13.1 13.9 13.9

1.0 nd nd nd 5.9 5.1 4.9 5.2 0.9 1.1 1.2 1.8

0.1 0.1 0.1 0.1 0.4 0.6 0.4 0.3 0.4 0.6 0.6 0.7

1.2 1.6 1.4 1.3 0.6 0.9 0.4 nd 0.7 0.9 0.7 0.6

1.3 1.7 2.1 2.1 2.3 3.2 2.4 1.9 0.2 0.2 0.2 0.3

15.7 16.5 17.1 17.6 11.5 11.4 13.9 15.2 10.0 10.2 10.5 10.4

See Table 1 for abbreviations. Means of three samples at different incubation time. nd s not detected Ž - 0.1 mg kgy1 ..

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

167

diffusion equation which gives the proportion of added Zn in the non-DTPA fraction as a function of the square root of incubation time Ž Fig. 4. . The results also suggest that the diffusion of Cu and Pb may be a rate-limiting process in their transformations in soils. 3.4. Effect of CaCO3 With addition of CaCO 3 , the soil pH ŽCaCl 2 . increased from 7.08 to 7.68. Except for a decrease in EXC-Cd, there was little effect of the addition of CaCO 3 on the distribution of metals in the individual fractions in the soil Ž Table 6.. There were low concentrations of Cu, Pb and Cd in the carbonate bound fraction even though CaCO 3 was added to the soil. 4. Conclusions The new sequential extraction procedure, which fractionated the specifically adsorbed heavy metals and the heavy metals associated with easily reducible Mn oxides, was deemed to be satisfactory for studying the transformations of heavy metals added to soils. Most of the added Cu and Pb were found in the specifically adsorbed and residual fractions while Cd was primarily found in the exchangeable fraction and secondly in the specifically adsorbed and residual fractions. A particularly high affinity of Pb for Mn oxides was found. The concentrations of metals in the reactive fractions Žwater-soluble, exchangeable, and EDTA-extractable fractions. were in the following order: Cd ) Cu ) Pb; they decreased and transformed to highly stable forms with time. The processes associated with the transformations of added Cu and Pb may be attributed mainly to the diffusion of the surface species into micropores and their entrapment in microporous solids, while the processes of the transformation of added Cd may be attributed to the inner-sphere surface complexation via partial or complete dehydration of the surface species. The effect of CaCO 3 on metal transformations was only minor.

References Aualiitia, T.U., Pickering, W.F., 1987. The specific sorption of trace amounts of Cu, Pb and Cd by inorganic particulates. Water, Air Soil Pollut. 35, 171–185. Beckett, P.H.T., 1989. The use of extractants in studies on trace metals in soils, sewage sludges, and sludge-treated soils. Adv. Soil Sci. 9, 143–176. Belzile, N., Lecomte, P., Tessier, A., 1989. Testing readsorption of trace elements during partial chemical extractions of bottom sediments. Environ. Sci. Technol. 23, 1015–1020. Bruemmer, G.W., Gerth, J., Tiller, K.G., 1988. Reaction kinetics of the adsorption and desorption of nickel, zinc and cadmium by goethite, I. Adsorption and diffusion of metals. J. Soil Sci. 39, 37–52.

168

Y.B. Ma, N.C. Uren r Geoderma 84 (1998) 157–168

Chang, A.C., Page, A.L., Warneke, J.E., Grgurevic, E., 1984. Sequential extraction of soil heavy metals following a sludge application. J. Environ. Qual. 13, 33–38. Crank, J., 1975. The Mathematics of Diffusion. Oxford Univ. Press, London. Gerth, J., Brummer, G.W., Tiller, K.G., 1993. Retention of Ni, Zn and Cd by Si-associated ¨ goethite. Z. Pflanzenernahr. ¨ Bodenk. 156, 123–129. Han, F.X., Hu, A.T., Qi, Y.H., 1995. Transformation and distribution of forms of zinc in acid, neutral and calcareous soils of China. Geoderma 66, 121–135. Hickey, M.G., Kittrick, J.A., 1984. Chemical partitioning of cadmium, copper, nickel and zinc in soils and sediments containing high levels of heavy metals. J. Environ. Qual. 13, 372–376. Karger, J., Ruthven, D.M., 1992. Diffusion in Zeolites and other Microporous Solids. John Wiley, ¨ New York. Keller, C., Vedy, ´ J.C., 1994. Distribution of copper and cadmium fractions in two forest soils. J. Environ. Qual. 23, 987–999. Levy, D.B., Barbarrick, K.A., Siemer, E.G., Sommers, L.E., 1992. Distribution and partitioning of trace metals in contaminated soils near Leadville, Colorado. J. Environ. Qual. 21, 185–195. Lindsay, W.L., Norvell, W.A., 1978. Development of a DTPA soil test for zinc, iron, manganese and copper. Soil Sci. Soc. Am. J. 42, 421–428. Ma, Y.B., Uren, N.C., 1995. Application of a new fractionation scheme for heavy metals in soils. Commun. Soil Sci. Plant Anal. 26, 3291–3303. Ma, Y.B., Xia, R.J., 1988. Research on soil microelements, I. Fractionation method of zinc in calcareous soils Žin Chinese.. Acta Agric. Univ. Pekinensis 14, 149–155. Mann, S.S., Ritchie, G.S.P., 1994. Changes in the forms of cadmium with time in some Western Australian soils. Aust. J. Soil Res. 32, 241–250. McGrath, S.P., Cegarra, J., 1992. Chemical extractability of heavy metals during and after long-term applications of sewage to soil. J. Soil Sci. 43, 313–321. McKenzie, R.M., 1980. The adsorption of lead and other heavy metals on oxides of manganese and iron. Aust. J. Soil Res. 18, 61–73. McLaren, R.G., Ritchie, G.S.P., 1993. The long-term fate of copper fertilizer applied to a lateritic sandy soil in Western Australia. Aust. J. Soil Res. 31, 36–50. Mench, M.J., Didier, V.L., Loffler, M., Gomez, A., Masson, P., 1994. A mimicked in-situ ¨ remediation study of metal-contaminated soils with emphasis on cadmium and lead. J. Environ. Qual. 23, 58–63. Miller, W., Martens, D.C., Zelazny, L.W., 1986. Effect of sequence in extraction of trace metals from soils. Soil Sci. Soc. Am. J. 50, 598–601. Naidu, R., Bolan, N.S., Kookana, R.S., Tiller, K.G., 1994. Ionic strength and pH effects on surface charge and Cd sorption characteristics of soils. Eur. J. Soil Sci. 45, 419–429. Ramos, L., Hernandez, L.M., Gonzalez, M.J., 1994. Sequential fractionation of copper, lead, cadmium and zinc in soils from or near Donana National Park. J. Environ. Qual. 23, 50–57. ˜ Singh, J.P., Karwasra, S.P.S., Singh, M., 1988. Distribution and forms of copper, iron, manganese, and zinc in calcareous soils of India. Soil Sci. 146, 359–366. Sposito, G., 1984. The Surface Chemistry of Soils. Oxford University Press, New York. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the speciation of particulate trace metals. Anal. Chem. 51, 844–851. USDA, 1988. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Robert E. Krieger Publ. Comp., FL. Waychunas, G.A., 1991. Crystal chemistry of oxides and oxyhydroxides. Rev. Mineral. 25, 11–68. Xu, J.Q., 1990. Clay minerals in soils. In: Soils of China Žthe Engl. ed... Institute of Soil Science, Academia Sinica, Science Press, Beijing, pp. 434–459. Xue, J., Huang, P.M., 1995. Zinc adsorption and desorption on short-range ordered iron oxide as influenced by citric acid during its formation. Geoderma 64, 343–356.