Soil Science Society of America Journal 64:609-612 (2000)
© 2000 Soil Science Society of America
DIVISION S-2-SOIL CHEMISTRY
A digestion method for trace metals recovery from oil and grease contaminated soils
Nicola Cooka,
Marie-Claude Turmelb and
William H. Hendershotb
a CSIRO Land and Water, Adelaide Lab., Glen Osmond, SA 5064, Australia, current)
b Dep. of Natural Resource Sci., McGill Univ. - Macdonald Campus, 21,111 Lakeshore Rd., Ste. Anne de Bellevue, QC, Canada H9X 3V9
williamh{at}nrs.mcgill.ca
ikkin3{at}hotmail.com
 |
ABSTRACT
|
|---|
Oil and grease contaminated soils are difficult to digest using the common HNO3–H2O2 digestion method. Even after an overnight pre-digestion with HNO3 and subsequent heating, the addition of H2O2 caused violent reactions in soils with oil and grease, resulting in the loss of the samples. At low metal concentrations at least a gram of soil needs to be digested to obtain concentrations measureable by flame atomic absorption spectrometry (FAAS) or inductively coupled plasma–atomic emission spectrometry (ICP-AES). We developed a modified procedure using HNO3–HClO4 for the analysis of total trace metals that can be used on all types of soils including those with oil and grease. Recovery rates of 99, 94, 114, 92, and 83% for Cd, Cu, Ni, Pb and Zn, respectively, were obtained for standard reference material (SRM) NIST 2710 (Montana Soil). Soils with 
1000 mg kg-1 oil and grease were successfully digested and gave extract concentrations suitable for analysis on FAAS or ICP-AES. Compared to the common HNO3–H2O2 soil digestion method, the proposed method was as effective (no significant difference) in extracting Cu, Pb, and Zn and was significantly better in liberating Ni from the soil. With the HNO3–H2O2 soil digestion method the Cd concentrations were often below the limit of detection by FAAS but were measureable in the HNO3–HClO4 digests. Variability of results using the proposed method was reduced in some cases.
Abbreviations: ETAAS, electrothermal atomic absorption spectrometry • FAAS, flame absorption spectrometry • FAES, flame atomic emission spectrometry • HF, hydrofluoric acid • ICP-AES, inductively coupled plasma-atomic emission spectrometry • ID, isotope dilution • INAA, instrumental neutron activation analysis • POLAR, plarography • RNAA, radiochemical neutron activation analysis • SRM, standard reference material • TIMS, thermal ionization mass spectrometry
|
INTRODUCTION
|
|---|
DETERMINING THE TOTAL TRACE ELEMENT CONCENTRATION of a soil is the first step in evaluating its potential health or ecological hazard. The literature dealing with contaminated soils is divided neatly into issues of organic or inorganic pollutants. This may be appropriate for contamination problems like oil spills (organic) or mine tailings (inorganic). In other situations, especially in urban or industrial soils, a mixture of contaminants at a site is likely to be found. Metals from the dumping of ash, scrap metal and leaded paint are often found with organics like oil and grease from railway and automotive activity and industrial disposal. The presence of oil and grease in soils complicates the use of standard soil digestion methods to measure the total soil metal concentrations, especially when metal concentrations are low. Oil and grease are complex organic compounds like the fats present in plant and animal tissue more commonly found in soil. Plant and animal derived organic matter may coat some soil mineral particles but is mainly found as a discrete part of the solid phase. Oil and grease in soils on contaminated urban–industrial land tends to coat the soil particles making the release of the bound metals more difficult. In the soils we were analyzing the metals and oil and grease components had been there for decades, and so had undergone an aging process, and were present in high concentrations. The trace metals of concern here are Cd, Cu, Ni, Pb, and Zn although other metals like Al, Ca, Fe, K, Mg, and Mn can be analyzed from the digests.
A standard, relatively safe, method that allows for the recovery of at least approximately 90% of soil bound metals is required in most laboratories working with trace metal contaminated soil. The most common methods used to digest trace metal contaminated soils and sediments are concentrated HNO3 followed by 30% H2O2 — USEPA Method 3050 (Miller and McFee, 1983; Edgell, 1988) or aqua regia (Agemian and Chau, 1976; Mench et al., 1994). Agemian and Chau (1976) showed that aqua regia was generally less effective than HNO3–HClO4 at extracting elements from a Lake Ontario sediment. Due to the very strong efficiency of HClO4 to oxidize (Burau, 1982), it is capable of extracting some of the metal from the silicate lattice (Agemian and Chau, 1976), whereas aqua regia and the USEPA Method 3050 will dissolve all metals except those in the silicate structure (Agemian and Chau, 1976; Sheppard and Stephenson, 1995). Nitric acid acts as a strong oxidizer and H2O2 destroys organic matter and releases the associated metals. Hydrogen peroxide (H2O2) oxidizes organic matter to form CO2 and H2O, during which the oxidation state of H is reduced from +2 to +1. Relative to H2O2, HClO4 has a much greater capacity to oxidize organic matter, gaining eight electrons per mole compared to one electron per mole for H2O2. Perchloric acid is also a strong dehydrator and leaching agent and its use requires a special fume hood.
For a more accurate estimate of the total metal, hydrofluoric acid (HF), usually in combination with HNO3 and HClO4 must be used, which will recover metals that are part of the silicate structure (Agemian and Chau, 1976; Tessier et al., 1979). It is relatively dangerous for daily laboratory use and requires specialized equipment (i.e., teflon digestion vessels). Soils spiked with trace metals and digested with HF are reported to yield recoveries between 92 and 104% of the added metal (Gupta and Chen, 1975). All of these acids are more effective at dissolving metals when boiling (Agemian and Chau, 1976). Other methods like spark source and isotope dilution mass spectroscopy as well as neutron and particle activation analysis may be more precise, but most laboratories are not equipped for them (Stoeppler, 1985).
During the digestion process of urban soils contaminated with trace metals and oil and grease, the introduction of H2O2 to soils, partially digested by hot HNO3, caused a violent reaction when more than 0.2 g soil was used, resulting in the loss of the sample from some tubes and contamination of other samples being digested. A small soil sample size reduced this problem but then the metal concentrations in the digests were below the detection limit of FAAS. Solution concentrations higher than the FAAS detection limit of 0.02 mg Cd L-1, 0.08 mg Cu L-1, 0.07 mg Ni L-1, 0.19 mg Pb L-1, and 0.01mg Zn L-1 are required. Detection limits for ICP-AES are similar.
A search of the literature yielded no information on how to digest soils contaminated with oil and grease to measure the trace metal concentrations. Therefore the objective of this study was to develop a standardized digestion method that can be used on a range of soil types including those contaminated with oil and grease that will yield adequately high metal concentrations in the digests for analysis on FAAS or ICP-AES.
|
Materials and Methods
|
|---|
The Soils
Samples were collected from the surface horizon (0–0.20 m) of five contaminated sites (Soils 3–7) in Montréal, QC, Canada. Metals in these soils are associated with transportation corridors and decades of fallout and dumping of local industrial scrap. A comparitively uncontaminated soil from the Morgan Arboretum, Ste. Anne de Bellevue, Québec, was also analyzed (Soil 9). Two-kilogram subsamples of each soil were air dried to constant weight, homogenized, and ground using a mullite mortar and pestle to pass a 500-µm sieve (<500-µm soil). Subsamples were homogenized again prior to digestion. Oil and grease concentrations were previously determined by the Ville de Montréal and found to contain 1000 mg kg-1 (Table 1)
(Anonymous, 1992). Organic matter content (Nelson and Sommers, 1982), pH in H2O (Hendershot et al., 1993a), CEC (Hendershot et al., 1993b), particle-size distribution (McKeague, 1976), extractable oxides of Al, Fe, and Mn (McKeague and Day, 1993) and total carbonates (as % CaCO3) (Bundy and Bremner, 1972) were determined on air-dried subsamples. Total Ca, K, Mg, and P was measured by ICP-AES (Table 2) . Throughout the study the trace metals Cd, Cu, Ni, Pb, and Zn were analyzed on FAAS. A SRM was analyzed for quality control: a highly contaminated soil, NIST 2710 Montana Soil (NIST,1993). The total concentrations in the quality control, NIST SRM, were obtained by digestion with HNO3–HClO4–HF and analyzed by isotope dilution (ID), ICP-MS and radiochemical neutron activation analysis (RNAA) for Cd; RNAA, flame atomic emission spectrometry (FAES) and ICP for Cu; ID, thermal ionization mass spectrometry (TIMS), polarography (POLAR) and ICP for Pb and ID, ICP-MS, electrothermal atomic absorption spectrometry (ETAAS) and instrumental neutron activation analysis (INAA) for Ni; ID, TIMS, ICP, INAA, and POLAR for Zn.
The Procedure
The standard method was adapted from Mench et al. (1994). We used 0.2-g samples of <500-µm soil in 100 mL Tecator digestion tubes. Fifteen milliliters of trace metal grade HNO3 was added and the tubes allowed to stand overnight. Tubes were then placed in a block digestor and heated to 150°C for 3 to 8 h (until brown vapors stopped). They were then removed from the block and allowed to cool before the addition of 5 mL of 30% H2O2. Tubes were then heated at 120°C. Additions of 1 mL H2O2 were continued until the suspension stopped bubbling. The total digestion time was 10 h. The proposed new method was adapted from several studies including Miller and McFee (1983) and Agemian and Chau (1976) and must be performed in a fume hood appropriate for HClO4 use. We used 1.0-g samples of <500-µm soil in 100 mL Tecator digestion tubes. Fifteen milliliters of trace metal grade HNO3 was added and the tubes allowed to stand overnight. Due to the potential for violent reactions when HClO4 comes into contact with organic matter, this pre-digestion step is essential. The tubes were then placed in a block digestor and heated to 120°C; the heat was then gradually raised to150°C and maintained until there were no more brown fumes (3–6 h). To prevent the tubes from running dry, HNO3 was added as needed to maintain the volume between 10 and 15 mL. The tubes were then removed from the block and allowed to cool before the addition of 5 mL of trace metal grade HClO4. The tubes were then placed back on the block, the temperature raised gradually (over 2 h) to 160°C. The digestion was complete when the white smoke stopped and the soil residue was white with small black specks (3–6 h). The solution, especially for soils with oil and grease, was not always clear, but the residues were as described. The tubes were removed from the block, allowed to cool, brought to volume with deionized H2O, covered with parafilm and inverted several times to homogenize. After suspended particles had settled out (16 h) the top 70 to 80 mL of solution was decanted without filtering and used for metal analysis.
Statistical Analysis
Paired t-tests were performed on mean soil metal values using Genstat 4.1 for Windows (Lawes Agricultural Trust, 1990) to determine significant differences (
= 0.05, unless otherwised indicated) between the two methods. This test indicates whether one method produces consistently higher or lower results than another (test of bias between methods).
|
Results and Discussion
|
|---|
The efficiency of the method on a typical, highly contaminated soil (SRM NIST 2710) is presented in Table 3
. This soil does not contain oil or grease; no such SRM was available. These results show that the method is effective at removing Cd, Cu, Ni, Pb, and Zn. Differences in recovery rates may be explained by the geochemistry of the sample. The type of mineral and the type of organic matter present will influence the ability of the acid to extract metals (Agemian and Chau, 1976). The data we are comparing our results with is for true totals obtained from a variety of analytical techniques. Our technique may be able to extract some metals from silicates but would not completely liberate all bound metals. Therefore we did not expect to obtain 100% recovery if the samples contained trace metals in silicate minerals.
View this table:
[in this window]
[in a new window]
|
Table 3 Quality control: Metal concentrations recovered using SRM NIST 2710 digested in HNO3–HClO4 compared to certified reference values (mean values in mg kg-1)
|
|
Table 4
shows the results obtained using two different digestion methods on the contaminated soils from Montréal (3–7) and the clean soil (9). Soils 3 and 4 contain oil and grease with concentrations of 1000 mg kg-1 soil. The two methods are comparable in terms of recovery when metals are not associated with oil and grease and are not in low concentrations, for example, Cd. In many of the cases the variability was reduced using the proposed method (2), in part because 1 g of soil was used in the proposed method instead of 0.2 g. Statistical analyses comparing the two methods showed that the proposed method was as effective (no siginificant difference) as the standard method for Cu, Pb, and Zn, but was significantly more effective at extracting Ni ( = 0.01). The paired t-test used indicates whether one method measures more than the other across the six samples studied. With the HNO3–H2O2 soil digestion method the Cd concentrations were often below the limit of detection by FAAS but were measurable in the HNO3–HClO4 digests. For individual samples, the two methods do not always give the same result but these differences may be due to many factors including errors due to sample homogenization or contamination.
|
Conclusions
|
|---|
The proposed method allows the digestion of soils contaminated with oil and grease. It is practical for routine laboratory analysis as 40 samples can be run at one time if a block digestor is available. It permits the use of 1 g of soil and so yields extract concentrations in the range required for FAAS or ICP-AES analysis. It works well on a range of soil types and metal concentrations.
|
ACKNOWLEDGMENTS
|
|---|
This work was supported by FCAR (doctoral scholarship to Nicola Cook and Team grant to William H. Hendershot) and the Environmental Innovations Program, Canada.
Received for publication April 9, 1998.
|
REFERENCES
|
|---|
- Agemian H., Chau A.S.Y. Evaluation of extraction techniques for the determination of metals in aquatic sediments. The Analyst 1976;101:761-767.
- Anonymous. 1992. Etude de caracterisation des sols cour Mile-end nord du Canadien Pacifique, Etude No. 91F 158-D. Laboratoire module genie de l'environnement service des travaux publics, Ville de Montréal.
- Bundy L.G., Bremner J.M. A simple titrimetric method for determination of inorganic carbon in soils. Soil Sci. Soc. Am. Proc. 1972;36:273-275.
- Burau R.G. Lead. In: Page A.L., et al. , ed. Methods of soil analysis. Part 2, 2nd ed Madison, WI: ASA and SSSA, 1982:347-348.
- Edgell, K. 1988. USEPA Method Study 37—SW-846 Method 3050 Acid digestion of sediments, sludges and soils. EPA Contract no. 68-03-3254.
- Gupta S.K., Chen K.Y. Partitioning of trace metals in selective chemical fractions of nearshore sediments. Environ. Lett. 1975;10:129-158.[ISI][Medline]
- Lawes Agricultural Trust. 1990. Genstat 4.1. IACR, Rothamsted Experimental Stn., Harpenden, UK.
- McKeague, A., and J.H. Day. 1993. Acid ammonium oxalate extraction method 25.3. In M.R. Carter (ed.) Soil sampling and methods of analysis. Can. Soc. for Soil Sci. Lewis Publ., Chelsea, MI.
- Mench M.J., Didier V.L., Löffler M., Gomez A., Masson P. A mimicked in situ remediation study of metal-contaminated soils with emphasis on Cd and Pb. J. Environ. Qual. 1994;23:58-63.[Abstract/Free Full Text]
- Miller W.P., McFee W.W. Distribution of cadmium, zinc, copper and lead in soils of industrial Northwestern Indiana. J. Environ. Qual. 1983;12:29-33.
- National Institute of Standards and Technology. National institute of standards and technology, certificate of analysis, standard reference material 2710. Gaithersburg, MD: Dep. of Commerce, 1993.
- Nelson D.W., Sommers L.E. Total carbon, organic carbon and organic matter. In: Page A.L., et al. , ed. Methods of soil analysis. Part 2, 2nd ed Madison, WI: ASA and SSSA, 1982:539-579.
- Sheppard, M.I., and M. Stephenson. 1995. Critical evaluation of selective extraction methods for soils and sediments. Contaminated soils. p. 69–97. In R. Prost (ed.) Proceedings of the Third International Conference on the Biogeochemistry of Trace Elements, Paris, France. 15–19 May 1995. INRA, Paris, France.
- Stoeppler, M. 1985. Analytical aspects of the determination and characterization of metallic pollutants In H.W. Nürnberg (ed.) Pollutants and their ecological significance. John Wiley & Sons, New York.
- Tessier A., Campbell P.G.C., Bisson M. Sequential extraction procedure for the speciation of particulate trace metals. Anal. Chem. 1979;51:844-851.
TOP
ABSTRACT
INTRODUCTION
