SEQUENTIAL EXTRACTIONS, FRACTIONATION STUDIES – WHAT ARE THEY DEFINING?

Karen Stead, Robert J. Hares & Neil I. Ward*

ICP-MS Facility, Department of Chemistry, School of Physics and Chemistry, University of Surrey, Guildford, GU2 7XH *n.ward@surrey.ac.uk

ABSTRACT

The analysis of the total heavy metal content of soil, sludge and sediment is insufficient when assessing the environmental impact the metals will have on the receiving media. Chemical forms in which the metal is associated will not only determine the metals behaviour but also the mobility and bioavailability within the environment. Sequential extraction procedures based upon Tessier et al., (1979) are one useful method in assessing the relative importance of geochemical fractions that may be present in the sample being tested. The main geochemical fractions that are commonly tested for are exchangeable, bound to carbonates, bound to iron and manganese oxides, bound to organic matter and residual. However, as this paper will illustrate the amount of soil used to determine these fractions, together with the chemical methods employed has varied considerably over 25 years. This paper will also outline the inherent problems found when using specific chemical methods in defining the geochemical fraction.

INTRODUCTION

Contamination of the terrestrial environment by heavy metals has been occurring for millions of years from the natural weathering of the parent rocks, which precipitate metals into the terrestrial system. Mankind has helped to vastly increase this contamination through various activities, e.g. metalliferous mining and smelting, biosolid disposal, fossil fuel combustion, traffic-related emissions, waste disposal, agricultural and horticultural materials, warfare and military training. To fully understand and appreciate the fate of heavy metal contamination in the terrestrial environment one must first understand the physico-chemical properties of the metal concerned and the media it is entering. Adequate knowledge of biological mechanisms is also essential if we are to understand the residence time, and impact on animals and ultimately their effect on man. Adsorption of heavy metals into the soil system will greatly depend on the form in which the metal enters the soil (solid particulate in wet or dry deposition or solution), range of metal species, charge on the metal entering the soil, the pH of the receiving environment, the percentage of organic matter in the soil, and the redox condition of the receiving environment. One must also be aware of the inherent changes that take place within the receiving media and the effects this will have on the mobility of the heavy metal concerned.

SEQUENTIAL EXTRACTION / FRACTIONATION PROCEDURES

The analysis of heavy metal species in soil, dust or sediment, can be undertaken by either acid digestion or sequential extraction techniques. Analysis using acid digestion allows the analyst to ascertain the total content of heavy metal contamination. However, it is insufficient when assessing the environmental impact of the contaminated soil or sediment, since the chemical form which the metal is in will determine its behaviour and hence mobility and bioavailability. Sequential extraction procedures, based upon Tessier et al., (1979) are useful in assessing the relative geochemical forms that may be present in the sample being tested. Sequential extraction techniques use successive chemical extractants of various types in order of greater destructive ability and therefore possess greater sensitivity than a single extraction procedure. Specifically defined ‘speciation’ of soils and sediments is difficult due to numerous environmental variables. Consequently, operationally defined ‘speciation’, using sequential extraction schemes have been developed for assessing geochemical forms in soil and sediment (Lagerwerff & Specht, 1970; Harrison et al., 1981; Ma & Rao, 1997; Zufiaurre et al., 1998). Fractionation by selective chemical extraction removes or dissociates a specific phase with the associated metal bonded to it. The geochemical fractions most commonly analysed for are: exchangeable, bound to carbonates, reducible, oxidisable and residual.

Problems with Sequential Extraction Techniques

Various sequential extraction procedures have been used as summarised briefly in Table 1. Most schemes are similar in their chemical extractants and procedures as initially implemented by Tessier et al., (1979). The analytical instrument used to analyse these geochemical fractions are Atomic Absorption Spectrometry (AAS), Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) or Inductively Coupled Plasma Mass Spectrometry (ICP-MS). It should be noted that AAS and ICP-AES do not use mass selective detection and are not therefore prone to polyatomic interferences, and are less affected by matrix or non-spectroscopic suppression (Churchman, 1997). However, analyte levels in the exchangeable and/or adsorbed fraction may be significantly low to be recognised only by ICP-MS.

The choice of chemical extractants when using ICP-MS must be considered very carefully. The use of 1M MgCl₂ presents a potential non-spectroscopic matrix suppression effect. Alternatives to MgCl₂ include 1M NaOAc and 1M NH₄OAc, but this leads to the effect of sodium suppression on the ionisation of heavy metals within the plasma. Care must also be taken when using ICP-MS when using NaOAc and HOAc. These chemicals can result in non-spectroscopic suppression and/or polyatomic (adduct) interferences, and organic carbon loading in the plasma (Churchman, 1997).

Chemical solvents can either be specific to a particular geochemical fraction or specific in their mode of action. However, sequential extraction techniques harbour many problems, namely the use of one specific chemical extractant for one specific geochemical fraction. The chemical used may only partially remove the metal defined in a geochemical fraction, and/or cause the metal to become redistributed onto another geochemical fraction.

Within published research there is a large unwillingness for researchers to detail the amount of soil used in the fractionation and the chemical methods employed. Typically masses of 1.0 g are used, although variations from 0.5 – 5.0 g have been reported. Also of interest is the apparent non-disclosure of the state of the soil, namely fresh or dry weight basis, whether the sample has been sieved and the particular fraction size of the sieve. Problems can arise if the soil is used fresh or oven-dried, if the sample has been homogenised, or if the sample has been sieved to a particular particle size.

Sample collection, preparation and storage can also lead to changes in chemical speciation and thus misinterpretation of results. The disturbance of equilibrium conditions, particularly during sample collection can be a major source of error. As mentioned above, the use of dried or fresh soil for sequential extraction techniques will greatly affect the natural heavy metal equilibrium existing in the soil before it was collected. Soil drying has been shown to reduce the proportion of iron extracted by reagents that remove amorphous iron oxides and thus suggests an increase in oxide crystallinity (Thomson et al., 1980). Hence, the use of dried soil or sediment to ascertain the bioavailability of a metal, through the use of a sequential extraction procedure, may become limited. The collection of sediment cores may result in the mixing of oxidised and anoxic layers within the core, which, will confuse the geochemical fractions (Luoma, 1995). However, it is virtually impossible to avoid changes in the natural equilibrium of these geochemical forms within a soil/sediment matrix, unless sequential extraction is undertaken immediately after sample collection. Thus comparisons of data can only be undertaken if the same sample preparation, analytical methods, etc., are employed.

To date there is only one certified reference material, which can be used in a three-step sequential extraction procedure – BCR/CRM 601 Lake Sediment.

Ideally, the use of certain chemical extractants will depend upon the aim of the study, the type of solid and the heavy metal of interest. Contact time and washing procedures between each geochemical defined phase is also important when considering the movement of metals within fractions.

FUTURE TRENDS

Since Tessier et al., (1979), research into the use of specific chemical extractants to better define geochemical fractionation have increased significantly. Researchers today are able to define further those heavy metals bound to non-crystalline Fe oxides and crystalline Fe oxides (Berti et al., 1996). Emphasis has also largely been placed on defining the ‘water soluble’, ‘plant available’ or ‘bioavailable’ metal fraction, since these geochemical fractions are the most important when assessing environmental impact.

CONCLUSION

There is a need for legislation to look at the metal fraction of soil that is ‘soluble’ or ‘plant available’. Setting contaminated land guidelines on ‘total’ metal content is unrealistic considering that the majority of the metal contained within the soil may be locked up, and therefore may be unavailable.

Limitations of sequential extraction procedures reveal the results are only representative of operationally defined speciation, rather than quantification of a specifically defined chemical species. Sequential extraction procedures should therefore be regarded as a process in which the sample is transferred into a small defined artificial environment where shifts in reaction equilibrium occurs releasing the heavy metal component in solution.

REFERENCES

· Berti W.R & Jacobs L.W 1996 Journal of Environmental Quality. 25:1025-1032

· Churchman, D 1997 The Analysis of Selenium in Human Blood Serum by ICP-MS. PhD Thesis. University of Surrey.

· Chu C.W, Poon C.S & Cheung R.Y.H 1998 Water Science Technology. 38, 2:25-32

· Harrison R.M, Laxen D.P.H & Wilson S.J 1981 Environmental Science & Technology. 15: 1378-1383

· Hickey M.G & Kittrick J.A 1984 Journal of Environmental Quality. 13, 3:372-376

· Lagerwerff J.V & Specht 1970 Environmental Science & Technology. 4, 7: 583-586

· Luoma S.N 1995 In: Metal Speciation and Bioavailability in Aquatic Systems. (Tessier A & Turner D.R, Eds), John Wiley & Sons, Chichester, pp609-659.

· Ma Y.B & Rao G.N 1997 Journal of Environmental Quality. 26: 259-264

· Tessier A, Campbell P.G.C & Bisson M 1979 Analytical Chemistry. 51, 7: 845-851

· Thomson et al., 1980 Water Air and Soil Pollution. 14: 215-233

· Zufiaurre R, Olivar A, Chamorro P, Nerin C & Callizo A 1998 Analyst. 123: 255-259

Table 1: Commonly used Fractionation Extractants

Exchangeable	Adsorbed, bound to carbonates	Bound to Fe-Mn oxides	Bound to organic matter	Residual bound	References
1M MgCl₂	1M NaOAc pH 5	0.04M NH₂OH.HCl/HOAc	0.02M HNO₃/30% H₂O₂/ 3.2M NH₄OAc	HF-HClO₄	Tessier et al., (1979)
1M MgCl₂	1M NaOAc pH 5	0.04M NH₂OH.HCl/HOAc	0.02M HNO₃/30% H₂O₂/ 3.2M NH₄OAc	HF-HClO₄	Hickey & Kittrick (1984)
1M MgCl₂	1M NaOAc pH 5	0.04M NH₂OH.HCl/HOAc	0.02M HNO₃/30% H₂O₂/ 3.2M NH₄OAc	HF-HCl/HNO₃	Ma & Rao (1997)
1M KNO₃	0.5M KF, 0.1M EDTA	N/A	0.1M Na₄P₂O₇	6M HNO₃	Chu et al., (1998)
0.5M MgCl₂	1M NaOAc pH 5	0.04M NH₂OH.HCl/HOAc	0.02M HNO₃/30% H₂O₂/ 3.2M NH₄OAc	Calculated as the difference between total metals and the sum of extracted metals	Zufiaurre et al., (1998)