Acta Chim. Slov. 2004, 51, 317-324. 317 Short Communication NOVEL APPROACHES FOR DETERMINATION OF THE CHEMICAL AVAILABILITY OF METAL(LOID)S IN SOIL BASED ON THE KD CONCEPT Johannes T. van Elteren National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia Received 22-10-2003 Abstract A theoretical framework is presented to introduce several novel approaches for deduction of the chemically available metal(loid) concentrations in soil. The framework is based on the KD concept assuming a linear relationship between the extractable metal(loid) concentration in the solid phase and the metal(loid) concentration in the liquid phase. The approaches introduced include both exact approaches using (radioactive or stable) tracers and approximate approaches using more conventional extraction techniques. Key words: sol, metal(loid)s, chemical availability, partitioning coefficient Introduction The metal(loid) concentration in a soil is generally assessed by complete digestion or destructive analytical procedures. These procedures measure the total concentration of a metal(loid) in a soil; however, the concentration actually available to contribute to the exposure of organisms in soil is quite different. For reasons of clarity we distinguish between the parameters chemical availability and bioavailability.u The chemical availability of metal(loid)s in soils depends on adsorption-desorption processes, i.e. the dynamic exchange between the metal(loid) in soil particles and soil pore water under environmental conditions. This parameter is mostly estimated by extraction protocols3 using extractants simulating the soil pore water composition and is governed by the partitioning coefficient, KD4 The bioavailability describes the proportion of a metal(loid) in soil which is available for uptake by biota under environmental conditions and depends on the chemical availability, the physico-chemical form (speciation) of the metal(loid), transfer mechanism across membranes, etc. Once in the bioorganism a toxic effect may be expressed in the form of e.g. a growth deficiency. In this communication we will focus on novel approaches to accurately determine the chemical availability of metal(loid)s in soil. Most extraction protocols assume J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… 318 Acta Chim. Slov. 2004, 51, 317-324. quantitative recovery of chemically available metal(loid)s for a given extractant; however, this is questionable in many instances.5 Only when we are able to extrapolate to e.g. an infinite extractant volume, are quantitative recoveries found. Furthermore, it is important to note that the actual chemical availability is only given by the metal(loid) fraction directly related to the solution in the soil pores; extractants having a chemical composition dissimilar from the soil pore solution yield an estimate of the actual chemical availability. Some novel approaches to describe the (actual) chemical availability will be theoretically derived by adopting the KD concept. Theory In soil the total metal(loid) concentration may be expressed by at =a0+a(š where at is the total metal(loid) concentration, a0 the chemically available concentration and a® the chemically unavailable concentration; ali concentrations in mol kg"1. The chemically available concentration a0 is directly related to the metal(loid) concentration c0 (in mol L"1) in the soil pore solution via the in situ KD: with KD in L kg"1. This is true when we assume a steady-state equilibrium and linear adsorption isotherm behaviour on one soil adsorption site only. The schematic diagram in Figure 1 reflects this situation in undisturbed soil systems and is given in the form of a 2-compartmental model. In practical measurement situations it is unavoidable that we disturb the systems measured, but with well-chosen approaches the deviations may be minimized and a pseudo-equilibrated system may be assumed. An exact approach using tracers and more conventional approximate approaches using extraction techniques are discussed below for retrieval of the chemical availability. Where possible the approaches will be illustrated with appropriate examples. J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… Acta Chim. Slov. 2004, 51, 317-324. 319 Figure 1. Schematic representation of an environmental soil system. Qs is the chemically available metal(loid) amount (=m-a0) in the soil and g, the corresponding metal(loid) amount in the soil pore solution (=V.c0), both in moles. m is the mass of the soil (in kg) and F is the volume of the soil pore solution (in L). Exact approach An isotopic exchange approach6'7 yields the closest resemblance to the actual chemical availability. At the sampling site representative soil samples are put into containers and a tracer (stable or radioactive) containing the labelled metal(loid) under study is added and thoroughly mixed. It is essential that the tracer has a “labelled/unlabelled” ratio which is so high that a negligible amount of the unlabelled metal(loid) is added with the tracer. After several weeks of equilibration, and, depending on the sample type, several wet/dry cycles, the soil samples are centrifuged/filtered to separate the phases. The label has distributed proportionally over the accessible fractions (the chemically unavailable metal(loid) is excluded) as follows: 1 Q ,9l Qi (3) with qs and qi the “label amounts” (in arbitrary units depending on the tracer properties) in solid and liquid, respectively, and Qs and Qi the related “unlabelled amounts” (in moles) in solid and liquid, respectively. Since QilV and QJM mimic c0 and a0, respectively, KD may be written as KD = a, Q?V q?V ---- =---------=-------- c0 Ql ?m ql?m (4) thereby giving a quantitative measure of the in situ KD via measurement of the “label amounts” qs and qb knowing volume V (in L) and soil mass m (in kg). When c0 is measured as well, the chemical availability a0 may be derived: J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s... 320 Acta Chim. Slov. 2004, 51, 317-324. ql-m When e.g. 10000 counts (= count rate in cpm X counting tirne in min) of a metal(loid) radiotracer are added to a system where m is 0.001 kg and F is 0.001 L with a measured c0 of 0.0002 mol L"1, and we measure a qs of 8333 counts and a qi of 1667 counts (after an appropriate equilibration tirne), KD is 5 L kg"1 and a0 is 0.001 mol kg"1. Approximate approaches The in situ KD and actual chemical availability a0 can only be properly estimated by isotopic exchange methods when applied very carefully as described above. Extraction approaches are approximate because the soil pore solution has to be replaced with an extractant which has to mimic the composition of the soil solution (pH, Eh, ionic strength, etc.) as closely as possible. Often aqueous extractants such as e.g. 0.01 mol L"1 CaCl2 are used for that purpose. Then, a fairly accurate estimate of the in situ KD and a0 may be obtained under various experimental approaches. After appropriate equilibration and removal of the solid phase by centrifugation/filtration the metal(loid) concentration in the liquid phase is measured. Variable volume/mass ratio To be able to extract the chemically available metal(loid) concentration from a soil, a high V/m ratio may be required, depending on the KD. In that čase the resulting metal(loid) concentrations in the extract may become undetectably low. To circumvent this problem the metal(loid) concentration in the extract may be measured as a function of the Vlm ratio from which relationship we may deduce the chemically available metal(loid) concentration in the soil as described below.8 When an amount of soil (m) containing an chemically available metal(loid) concentration (a0) is contacted with an extractant, after equilibration the following mass balance exists: M-a0=c1-F+M-a1 J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… Acta Chim. Slov. 2004, 51, 317-324. 321 where ax is the remaining metal(loid) concentration (in mol kg"1) in soil and a the extracted metal(loid) concentration in the extractant (in mol L"1). Substitution of the relationship ax=KD-cx in Eq. (6) gives 1 \ V K =------+ ^- cl a0 m a0 00 Thus, by measuring \lcx as a function of the Vlm ratio a linear relationship is obtained with a slope of l/a0 and an intercept of KD/a0 (see Figure 2A). Initial successful experiences with this approach have been reported.9 Repetitive extraction Another approach to obtain insight into the chemically available concentration is by performing repetitive extractions. If we perform enough extractions, fmally the chemically available fraction will be completely stripped from the soil. In order not to end up with an extremely high number of extraction steps, in the čase of high KD values, a relationship between the extraction step and the corresponding metal(loid) concentration in the extract may be established to derive the chemically available concentration by extrapolation. A similar mass balance as given in Eq. (6) is obeyed, but now for each individual extraction step p: m-ap_l=cp-V+m-ap with p = 1, 2, 3, ... For each extraction step/? the following relationship is valid: Kd=— (9) Substitution of Eq. (9) in Eq. (8) for the individual extraction steps gives c = a°'K^ (10) P J T V + KD m Thus, by measuring cp as a function of the repetitive extraction step p, KD and a0 may be retrieved with a non-linear curve fitting protocol (see the discrete graph in Figure 2B). Initial successful experiences with this approach have been reported.9 J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… 322 Acta Chim. Slov. 2004, 51, 317-324. 35 30 25 20 15 10 5 0 10 15 V /m (l kg-1 ) 20 25 70 60 50 40 30 20 10 ¦-. '-"¦--- ¦ »¦¦¦» 0 2 4 6 8 10 450 400 350 300 250 200 150 100 50 0 100 200 300 400 coµmol l"1) 500 600 Figure 2. Graphical illustration of the different approaches to derive KD and a0: variable volume/mass ratio approach (A), repetitive extraction approach (B) and titration approach (C). For ali examples KD was set to 5 L kg1 and a0 was set to 0.001 mol kg1; for (B) and (C) the V/m ratio was set to 10. 0 5 B 0 p 0 J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… Acta Chim. Slov. 2004, 51, 317-324. 323 Titration The method of choice for determination of KD is by varying the concentration of the metal(loid) (c0) in the extractant during equilibration with a known amount of soil.4 Only low c0 concentrations may be used so as not to jeopardize the linear adsorption isotherm concept; when adsorption sites become saturated the KD concept no longer holds. However, this approach may also be applied for retrieval of the chemical availability assuming that the following mass balance exists under equilibrium conditions: c0-V + m-a0=cl-V + m-al (11) Substitution of the relationship ax=KD-cx in Eq. (11) gives V c ¦ m m Thus, by measuring cx as a function of c0 a linear relationship is obtained with a slope of (V/m)/(V/m+KD) and an intercept of a0/(V/m+KD) (see Figure 2C). Conclusions Some novel approaches are introduced to derive the chemical availability of metal(loid)s in soils. It is shown that application of the KD concept allows for calculation of the chemically available fraction relying upon various experimental approaches. The exact approach using (radioactive or stable) tracers is the closest to estimating the actual chemical availability, whereas the approximate extraction approaches (variation of volume/mass ratios, repetitive extraction and titration) may deliver suitable results as well when the extractant composition matches the soil pore solution as closely as possible. Work is in progress to experimentally evaluate ali approaches developed in this communication. Acknowledgements The author would like to thank Dr. A.R. Byrne for constructive comments on the manuscript. J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s… 324 Acta Chim. Slov. 2004, 51, 317-324. References 1. I. K. Iskandar, M. B. Kirkham, Trace Elements in Soil: Bioavailability, Flux and Transfer, Lewis Publishers, Inc., Boca Raton, Florida, United States, 2001. 2. J. D. Wolt, Soil Solution Chemistry: Applications to Environmental Science and Agriculture, John Wiley and Sons, New York, United States, 1994. 3. V. H. Kennedy, A. L. Sanchez, D. H. Oughton, A. P. Rowland, Analyst 1997, 122, 89R–100R. 4. USEPA, Understanding variation in partitioning coefficient, KD, values, Volume I and II, Washington, United States, 1999. 5. J. T. van Elteren, B. Budič, Anal. Chim. Acta 2004, 514, 137–143. 6. L. Blake, N. Hesketh, S. Fortune, P. C. Brookes, Soil use manage. 2002, 18, 199–207. 7. J. T. van Elteren, K. J. Kroon, Z. I. Kolar, Z. Šlejkovec, T. G. Verburg, Radiochim. Acta, submitted. 8. H. Hellmann, Fresenius Z. Anal. Chem. 1984, 319, 267–271. 9. J. T. van Elteren, Z. Šlejkovec, R. Milačič, Int. J. Environ. Anal. Chem. 2003, 83, 389–396. Povzetek Članek obravnava teoretične osnove različnih eksperimentalnih pristopov za določitev okoljsko razpoložljivih koncentracij kovin in polkovin v tleh. Ideja je osnovana na konceptu KD, ki predpostavlja linearno adsorpcijsko izotermo, to je linearno razmerje med razpoložljivo koncentracijo kovine/polkovine v trdni fazi in koncentracijo kovine/polkovine v tekoči fazi. Opisane experimentalne tehnike vključujejo tako ekzaktni pristop z uporabo radioaktivnih ali stabilnih sledilcev kot tudi neekzaktni pristop z uporabo konvencionalnih ekstrakcijskih tehnik. J. T. van Elteren: Novel Approaches for Determination of the Chemical Availability of Metal(Loid)s…