Materials and Methods .1 SOILS

Speciation of Pb

XRD due to its impure and possibly also its poorly crystalline nature. In a study of soils contaminated by copper-mining, Pb was found to exist as magnetoplumbite (Pb(Fe/Mn)12O19) and plumferrite (PbFe4O7), which are likely to be untransformed slag from the smelting wastes; and in soils originating from the vicinity of a battery factory PbCO3,PbSO4, PbO and (PbCO3)2Pb(OH)2 were identified. These results were confirmed by SEM-EDX studies which in addition showed that most Pb was found in discrete particles of lead-compounds (Welter et al., 1999). Transformation of metallic Pb in a sub-surface lead-pipe into litharge, hydrocerussite and cerrussite was observed by XDR and confirmed by SEM-EDX in the crust of the pipe as well as in the surrounding soil (Essington et al., 2004). Formation of stable phosphates could not be verified although SEM-EDX proved presence of apatite, possibly for the same reasons as those given by (Cotter-Howells, 1996).

Sequential extractions showed that bonding of Pb at background levels ( 20mg/kg) mainly occur in the reducible and the residual fraction i.e. bound to oxides and the mineral matrix. In soils diffusely contaminated by industrial emissions however, the fractions of oxide bound, carbonate bound and organically bound lead, are increasing at the expense of residual lead. Generally only little exchangeable lead was found compared to other metals with the exception of acidic soils (pH < 5) (Chlopecka et al., 1996). Consistently, Pb was found to bind preferentially to organic matter in another diffusely contaminated soil (Miller and Mcfee, 1983) and in several other industrially polluted soils Pb was found to bind preferentially to oxides, carbonates and organics (Yarlagadda et al., 1995; Ma and Rao, 1997). Interpretation of sequential extractions of industrially contaminated soils should however be made with care, because sequential extraction procedures are based on the assumption that Pb interacts primarily with common soil constituents. In industrially contaminated soils other contaminating substances might play a key role. This was taken into consideration in a study of soil contaminated by mining and ore-processing, where a sequential extraction procedure was designed especially for extraction of the Pb-species expected to appear in such contamination, Pb was found preferentially in the residual fraction assumed to consist of sulfide (Cordos et al., 1995). This approach should however also be applied with care, because it might lead to wrong interpretations if the actual speciation differ from the expected.

3 Materials and Methods

Speciation of Pb AAS (graphite-furnace for Sn) after filtration through a 0.45µm filter. AAS analyses were for all metals validated through analysis of reference samples.

TABLE I Pb contaminated soils.

Soil Activity and probable Pb-source Activity

period 1 Ancient, burned down, church with Pb-roof 1627 2 Extraction of metals from scrap and ore and recycling of

lead-acid accumulators. Smelter waste products deposited on site.

1938-1985

3 Car painting (lead based paints). 1900’s

4 Harbor area, filled up with waste from porcelain production

and a gasworks. Breaking up and scrapping of locomotives. 1880-1988 5 Former waste-dump with mixed industrial waste and

household waste. Cowered with sewage sludge, ash and mould.

1913-1937

6 Former waste-dump with mixed industrial waste and household waste. Cowered with sewage sludge, ash and mould.

1913-1920

7 Gravel pit used as waste dump. Source unknown. 1900’s

8 Metal foundry 1921-1976

9 Harbor area filled up with harbor sludge and surface soil from central Copenhagen between 1780 and 1820. Site laid out as army ammunition site with chemical storage.

After 1780

10 Soil collected by contractor north of Copenhagen. Source

unknown. Unknown

Grain-size distribution was determined by wet-sieving approximately 100g natural wet soil with 0.002M Na4P2O7 through a 0.063mm sieve followed by separation by dry sieving of the larger fractions (>0.063 mm) and sedimentation velocity measured by XRD of the smaller fractions (<0.063 mm) on Micromeritics® SEDIGRAPH 5100. pH was measured by electrode MeterLab® CDM220 after shaking of 5.0g dry soil with 12.5mL 1M KCl constantly for 1 hour, followed by settling for 10min.

Carbonate content was determined volumetrically by the Scheibler-method when reacting 3g of soil with 20mL of 10% HCl. The amount was calculated assuming that all carbonate was present as calciumcarbonate. Organic matter was determined by loss of ignition in a heating furnace at 550ºC for 1 hour. CEC was determined after ion exchange of 10g dry soil with NH4+, followed by exchange of NH4+ for Na⁺. The ammonium concentration of the supernatant was measured by spectrophotometer via flow-injection. Conductivity was measured by electrode MeterLab® CDM210 in solution prepared by constantly mixing of 10g soil and 25ml distilled water for 30min, followed by settling for 20min. Phosphate was measured after digestion of 0.2-0.5g sample at 550ºC followed by boiling with HCl, reaction with ammonium molybdate to form yellow phosphor-molybden acid, and reduction by ascorbic acid in the presence of antimony. The strong blue color was measured by spectrophotometer Shimadzu UV-1601. XRD analysis: All soils were subjected to XRD-analysis to reveal the mineral composition of the bulk soil as well as the clay fraction. The bulk-minerals were quantified on basis of peak-height while the clay-bulk-minerals were

Speciation of Pb

quantified on basis of peak-area after ethylene-glycolation and heating to first 350ºC and then 550ºC.

3.3 SPECIATION ANALYSIS

Pb distribution in soil fractions was determined upon analysis of the grain size distribution. Samples of each fraction were crushed, and the Pb concentration in each fraction was measured by AAS. The < 2 m fraction, which was separated out for XRD-analysis of clay minerals was additionally analyzed for Pb. For SEM-EDX analysis carbon coated, polished pucks were analyzed by a JEOL scanning electron microscope, JSM 5900. EDX results were treated by the software: Oxford Instruments INCA version 4.02. One specimen of each soil was analyzed (two specimens of soils 2 and 3), looking for bright spots (back-scatter-mode), which might contain Pb. When such a spot was found, it was analyzed for total elements, and in case it contained Pb, the relative amounts (atomic %) of all elements (except C, O and H, which could not be determined due to the sample preparation) in the spot were determined.

Mobilization of Pb due to pH changes was quantified after extraction of 5.00g dry, crushed soil with 25.00ml reagent at 200rpm for 7 days. The reagents were as follows:

1.0M NaOH, 0.5M NaOH, 0.1M NaOH, 0.05M NaOH, 0.01M NaOH, distilled water, 0.01M HNO3, 0.05M HNO3, 0.1M HNO3, 0.5M HNO3, 1.0M HNO3. pH was measured after 10min settling, after which the liquid was filtered through a 0.45 m filter for subsequent measurement on AAS. Non acidic samples were preserved with one part of conc. HNO3 to four parts of liquid in autoclave at 200 kPa and 120ºC for 30 minutes prior to AAS measurement. Sequential extraction was performed according to the method from the Standards, Measurements and Testing Program of the European Union (former BCR) (Mester et al., 1998). 0.5g of dry, crushed soil was treated in four steps as follows: I) Extraction with 20.0ml 0.11M acetic acid pH 3 for 16 hours. II) Extraction with 20.0 ml 0.1M NH2OH HCl pH2 for 16 hours. III) Extraction with 5.0ml 8.8M H2O2 for one hour and heating to 85ºC for one hour with lid followed by evaporation of the liquid phase at 85ºC until it had reduced to < 1ml by removal of the lid. The addition of 5.0 ml 8.8M H2O2 was repeated followed by resumed heating to 85ºC for one hour and removal of the lid for evaporation until almost dryness. After cooling down the sample, 25.0 ml 1M NH4OOCCH3 pH 2 was added, and extraction took place for 16 hours. IV) Digestion according to DS 259.

Between each step the sample was centrifuged at 3000rpm for 15min, and the supernatant was decanted and stored for AAS. Before addition of the new reagent the sample was washed with 10.0ml distilled water for 15min, centrifuged at 3000rpm for 15min and the supernatant was decanted. All extractions were performed at room temperature and shaking at 100rpm unless otherwise mentioned. Sequential extraction of pure phases was made in order to study the extraction of common contaminating Pb-species, and facilitate interpretation of extraction results. Metallic Pb and Pb bound in solder were obtained commercially. Lead-sulphate and lead-chromate were both precipitated in the lab by mixing a concentrated lead-nitrate solution with sulfuric acid and potassium dichromate respectively followed by filtration through a 0.45µm filter and three times washing with distilled water.

Speciation of Pb