A novel two-step sequential extraction has been developed to assess the bioaccessibility of As, Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn in airborne particulate matter following inhalation and transport into the human gastrointestinal tract by mucociliary clearance. A new artificial mucus fluid (AMF) was used to determine the bioaccessible potentially toxic element (PTE) fraction in the upper airways, in sequence with the simplified bioaccessibility extraction test (SBET) or the stomach phase of the unified bioaccessibility method (gastric fluid only) (UBMG). Filter dynamic measurement system TX40 filters smeared with soil reference material (BGS RM 102) were used as test samples. Analysis was performed by ICP-MS. Comparison between results obtained for soil alone and when the soil was supported on TX40 filters indicated that the presence of the substrate did not affect the extraction efficiency, although a large Zn blank was detected. The sequential AMF→SBET extraction liberated similar amounts of Fe, Mn, Ni and Zn to the SBET alone; but significantly less Cd; and significantly more As, Cr, Cu and Pb. The sequential AMF→UBMG extraction liberated similar amounts of Cd, Cr, Mn and Zn to the UBMG alone, but significantly less As, Fe and Ni; and significantly more Cu and Pb. Enhanced extractability was due to the greater quantities of exchangeable ions and complexing agents present. Adoption of a two-step sequential extraction (AMF followed by either the SBET or the UBMG) is recommended because it is more representative of biological conditions and avoids overestimation or underestimation of bioaccessible PTE concentrations.

•High levels of Hg found in freshwater sediment from the Union Canal, Scotland, UK.•Lower mercury concentrations in sediment from the Forth & Clyde Canal, Scotland, UK.•Low methylmercury concentrations found in sediments from both canals.•Significant inverse relationship between total Hg and %MeHg concentrations.•Ethylmercury detected in sediment from the Union Canal.Mercury concentrations were investigated in freshwater sediment from two canals in Scotland, UK. High concentrations found in the Union Canal (35.3–1200 mg kg−1) likely originate from historical munitions manufacture, with lower levels in the Forth & Clyde Canal (0.591–9.14 mg kg−1). Concentrations of methylmercury (MeHg) were low – from 6.02 to 18.6 μg kg−1 (0.001–0.023% of total Hg) in the Union Canal and from 3.44 to 14.1 μg kg−1 (0.11–0.58% of total Hg) in the Forth & Clyde Canal – and there was a significant inverse relationship between total Hg concentration and %MeHg. Total Hg concentration was significantly negatively correlated with pH and positively correlated with Fe content (in the Union Canal only) but not with organic matter, S content or the proportion of clay present. The MeHg concentration was not correlated with any of the above sediment parameters. Ethylmercury was detected in the most highly contaminated sediments from the Union Canal.Download high-res image (294KB)Download full-size image

The simplified bioaccessibility extraction test (SBET) and the stomach phase of the unified bioaccessibility method (UBM) have been modified for use in assessing the bioaccessibility of potentially toxic elements in samples of PM10 collected during routine air quality monitoring. Arsenic, Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn were measured by inductively coupled plasma mass spectrometry in extracts of synthetic PM samples prepared by loading candidate soil reference material BGS102 onto TX40 (Teflon-coated borosilicate) filters widely used in standard commercial tapered element oscillating microbalance/filter dynamics measurement system (FDMS) ambient particulate monitors. Analysis of blanks revealed two important sources of contamination that had to be controlled in order to achieve reproducible results. The syringe filters used in the SBET released Cu and Zn into sample extracts during filtration and had to be washed with 80 mL of 0.4 M glycine at pH 1.5 immediately prior to use, whilst the FDMS filters used to collect PM10 were found to contain sufficient extractable Zn (ca. 3 μg per filter) to almost double the concentration released by the soil. The latter is a consequence of the use of Zn in filter manufacture and so could not be eliminated, but was successfully overcome by means of blank-subtraction. A ten-fold miniaturisation of the SBET and six-fold miniaturisation of the UBM allowed 0.1 g samples to be processed, with analyte recoveries generally within ±10% of those obtained when the conventional procedures were used. Comparison between results obtained when the modified procedures were applied to soil alone, and when soil was loaded onto FDMS filters, indicated that the presence of the filter had no effect on extraction efficiency, except for Fe, provided blank-correction was performed. Results obtained for As, Cd and Pb when the modified UBM was applied to BGS102 on FDMS filters were 4.40 ± 0.04, 0.224 ± 0.002 and 17.3 ± 0.8 mg kg−1, respectively (n = 3), all within recommended ranges.

Surface sediments collected from the Lagos Lagoon, Nigeria, and three adjoining rivers were analysed for their physicochemical properties and pseudo-total concentration of the potentially toxic metals (PTM) Cd, Cr, Cu, Pb and Zn. The concentration of the PTM varied seasonally and spatially. Odo-Iyaalaro was observed to be the most polluted river, with highest concentrations of 42.1 mg kg−1, 102 mg kg−1, 185 mg kg−1, 154 mg kg−1 and 1040 mg kg−1 of Cd, Cr, Cu, Pb and Zn, respectively, while Ibeshe River was the least contaminated, apart from a site affected by Cu from the textile industry. Some of the sediments were found to be above the consensus-based probable effect concentrations and Dutch sediment guideline for metals. Overall metal concentrations were similar to those reported for other tropical lagoon and estuarine systems affected by anthropogenic inputs as a result of rapid urbanisation. Due to the large number of samples, principal component analysis was used to examine relationships within the data set. Generally, sediments collected during the dry season were observed to have higher concentration of PTM than those collected during the rainy season. This means that PTM could accumulate over a prolonged period and then be released relatively rapidly, on an annual basis, into tropical lagoon systems.

Several physiologically based extraction procedures have been proposed to estimate the fraction of the potentially toxic element content that would be bioaccessible in the human gastro-intestinal tract following accidental ingestion of soil. Many of these procedures are complex, they have been applied to a very limited range of soils, and most work has focussed on arsenic and lead. In the present study, a simplified, two-stage extraction, simulating the human stomach and intestine, was developed and applied to urban soil samples from ten public-access areas in the City of Torino, Italy. The human oral bioaccessibility of chromium, copper, nickel, lead and zinc was estimated. Lead and zinc bioaccessibilities were found to be higher in the stomach, but chromium was more bioaccessible in the intestine. Analyte concentrations were higher in roadside soils than in soils from parks. A higher proportion of the soil metal content was found in bioaccessible forms at roadsides than in parks. Comparison of the current findings with results of earlier work involving sequential extraction of the same soils indicated that the sequential procedure gave a relative, but not an absolute, indication of bioaccessibility. Calculations based on the bioaccessible analyte concentrations suggest that ingestion of only 2–3 g of some of the roadside soil samples from Torino could deliver the tolerable daily oral intake of chromium, nickel and lead to a 20-kg child. The developed procedure is useful for preliminary screening of soils and prediction of whether their bioaccessible metal contents are likely to pose a risk to human health.

A simple, two-stage, physiologically based extraction has been applied to assess the human bioaccessibility of potentially toxic elements (PTE) in 20 urban soils from a major UK city. Chromium and iron bioaccessibilities were found to be markedly higher in the intestinal phase, whilst lead and zinc bioaccessibilities were higher in the stomach. Copper and manganese bioaccessibilities were generally similar under both extraction conditions. Principal component analysis was used to study relationships amongst bioaccessible element concentrations and land use. Distinctions could be observed between the distributions of the urban metals—copper, lead and zinc—and metals predominantly of geogenic origin, such as iron. There was no clear delineation between roadside soils and soils obtained from public parks. Bioaccessible analyte concentrations were found to be correlated with pseudototal (aqua regia soluble) analyte concentrations for all elements except iron. Results of the BCR sequential extraction did not, in general, provide a good indication of human bioaccessibility. Comparison of bioaccessible PTE concentrations with toxicological data indicated that lead is the element of greatest concern in these soils but that levels are unlikely to pose a health risk to children with average soil intake.

The revised Community Bureau of Reference (BCR) sequential extraction procedure has been applied to investigate the effectiveness of two soil remediation strategies to reduce the amounts of potentially toxic elements in three contrasting contaminated soils (soils A, B and C) from derelict chemical manufacturing sites in the UK. Soil A was from the 35–45 cm deep layer of a site used for the manufacture of sulfuric acid. Soils B and C were topsoils from a site used for the manufacture of explosives, nitric acid and nylon The remediation strategies were flushing with EDTA in a column experiment (applied to soils A, B and C) and EDTA enhanced phytoremediation with Taraxacum officinale in pots (applied to soil B). Soil B, which contained the least amounts of aqua regia extractable metals, except for lead, but higher proportions of analytes in non-residual (i.e. acid exchangeable, reducible and oxidisable) forms was found to release greater amounts of analytes when flushed with EDTA. Comparison of the BCR sequential extraction fractionation patterns obtained before and after flushing of soil B, suggested that EDTA removed calcium mainly from the acid exchangeable pool, manganese mainly from the reducible pool, zinc from both acid exchangeable and reducible pools, and copper and lead from acid exchangeable, reducible and oxidisable pools. The chelate enhanced phytoremediation pot experiment conducted using soil B showed that EDTA treatment was significantly positively, correlated (p < 0.05) with an increase in the proportion of analytes recovered from the soil in step 1 of the BCR extraction scheme, for all analytes, and also enhanced metal uptake by plants. The sum of the amounts of analyte released in the first three steps of the sequential extraction, commonly regarded as the maximum amount of elements potentially available for plant uptake, was not positively correlated with plant-uptake.

Since their introduction in the late 1970s, sequential extraction procedures have experienced a rapid increase in use. They are now applied for a large number of potentially toxic elements in a wide range of sample types. This review uses evidence from the literature to consider the usefulness and limitations of sequential extraction and thereby to assess its future role in environmental chemical analysis. It is not the intention to provide a comprehensive survey of all applications of sequential extractions or to consider the merits and disadvantages of individual schemes. These aspects have been covered adequately in other, recent reviews. This review focuses in particular on various key issues surrounding sequential extractions such as nomenclature, methodologies, presentation of data and interpretation of data, and discusses typical applications from the recent literature for which sequential extraction can provide useful and meaningful information. Also covered are emerging developments such as accelerated procedures using ultrasound- or microwave energy-assisted extractions, dynamic extractions, the use of chemometrics, the combination of sequential extraction with isotope analysis, and the extension of the approach to non-traditional analytes such as arsenic, mercury, selenium and radionuclides.

The revised (four-step) BCR sequential extraction procedure has been applied to fractionate the chromium, copper, iron, manganese, nickel, lead and zinc contents in urban soil samples from public-access areas in five European cities. A preliminary inter-laboratory comparison was conducted and showed that data obtained by different laboratories participating in the study were sufficiently harmonious for comparisons to be made between cities and land types (e.g. parks, roadside, riverbanks, etc.). Analyte recoveries by sequential extraction, with respect to direct aqua regia digestion, were generally acceptable (100 ± 15%). Iron, nickel and, at most sites, chromium were found mainly in association with the residual phase of the soil matrix. Copper was present in the reducible, oxidisable and residual fractions, whilst zinc was found in all four sequential extracts. Manganese was strongly associated with reducible material as, in some cities, was lead. This is of concern because high lead concentrations were present in some soils (>500 mg kg−1) and the potential exists for remobilisation under reducing conditions. As would be expected, extractable metal contents were generally highest in older, more heavily industrialised cities. Copper, lead and zinc showed marked (and often correlated) variations in concentrations between sites within the same city whereas manganese and, especially, iron, did not. No overall relationships were, however, found between analyte concentrations and land use, nor between analyte partitioning and land use.

The revised, four-step BCR sequential extraction for soil or sediment has been compared with an alternative procedure in which 0.2 mol l−1 ammonium oxalate (pH 3) replaced 0.5 mol l−1 hydroxylammonium chloride (pH 1.5) in step 2, the reducible step. A variety of substrates were studied: BCR CRM601, a sewage sludge amended soil, two industrial soils, and a steel manufacturing by-product (basic oxygen furnace filter cake). Greater amounts of iron were recovered in step 2 when acid ammonium oxalate was used, for all substrates. Similar trends were observed for copper. Manganese and zinc were not strongly affected by the procedural modification, except for zinc in the two industrial soils, where oxalate extraction proved more efficient than use of hydroxylammonium chloride. A large proportion of the calcium and lead isolated in step 2 of the BCR procedure was not released until step 3 when the alternative procedure with oxalate in step 2 was used. This is probably due to rapid precipitation of analyte oxalates from solution. Thus, whilst oxalate offers superior dissolution of iron-containing matrix components, it should not be used if calcium or lead concentrations are to be measured. Selection of the most appropriated sequential extraction protocol for use in a particular study must always be carried out on the basis of “fitness for purpose” criteria. However, the revised BCR protocol, involving use of 0.5 mol l−1 NH2OH·HCl in the reducible step, appears to be more generally applicable than procedures involving acid ammonium oxalate.

An ultrasonic bath and an ultrasonic probe have been used to develop rapid versions of the three-stage Community Bureau of Reference (BCR, now the Standards, Measurement and Testing Programme) sequential extraction procedure. The effect of the ultrasonic treatments on the extraction of copper, iron, manganese and zinc from a sewage sludge-amended soil has been assessed. Recoveries similar to those of conventional shaking (i.e., conventional value, ±30%) could generally be obtained for copper, manganese and zinc, but not for the important matrix element iron. With the use of compromise sonication conditions, steps 1, 2 and 3 of the sequential extraction (excluding the hydrogen peroxide digestion in step 3, which was not performed with sonication) could be completed in 3, 5 and 1 min, respectively, using the ultrasonic probe, and in 3, 1 and 1 h, respectively, using the bath. The extraction procedures developed using the soil performed well when applied to lake sediment BCR CRM 601. Analyte partitioning was generally similar to that obtained with mechanical shaking, and overall metal recoveries were 84–98% of those obtained with the conventional BCR protocol, except for copper extracted with the probe (74%). Poorer performance (analyte recoveries, 58–104%) was obtained when the methods were applied to an intertidal sediment. This highlights the difficulty of developing a version of the BCR extraction, with ultrasonic assistance, which gives a performance equivalent to conventional shaking when applied to different substrates.