MOBILISATION OF LEAD – FIELD MEASUREMENTS AT A TRAP RANGE COMPARED TO SOLUBILITY EXPERIMENTS

Mattias Bäckström^*, Stefan Karlsson and Bert Allard

Man–Technology–Environment Research Centre, Örebro University, 701 82 Örebro, Sweden

^*Corr. author: Phone +46 19 30 39 65; Fax +46 19 30 31 69; mattias.backstrom@nat.oru.se

Abstract

The studied skeet and trap range is situated in the county of Örebro in the centre of Sweden and has been in use since the 1920s. The soil is a peat bog with a groundwater table that varies from the surface to 20-30 cm below. A small brook intersects the range. The principal forms of lead in ground- and surface waters were determined by filtration/ion-exchange in order to elucidate possible mobilisation processes.

In the groundwater some 75-95% of the lead was retained already by the 1.0 mm filter. Approximately 60-70% of the filterable lead consisted of anion-exchangeable forms. It is reasonable to conclude that the predominant forms consist of rather large agglomerates, most likely of organic matter. The total concentration of lead in the brook increased with app. 50% when passing the shooting range. The empirical speciation of aqueous lead forms and mobilisation rates at the field site are compared to those found in controlled experiments.

Introduction

Lead has been extensively used for shooting for several centuries due to its excellent ballistic properties. However, the negative impact on both the shot and the environment has been reported during the last few decades. Metallic lead was believed to resist corrosion at natural pH through the formation of a passivating precipitation on the surface. This theory has been questioned and it is now known that the weathering rate is highly related to the composition of the aqueous phase. The present study concerns the distribution of lead species in surface and groundwater at a shooting range with an organic rich soil that is close to saturation with groundwater most of the year. The field measurements were compared to lead concentrations found in dissolution experiments performed in the laboratory.

Experimental

Field site. Kvastbo is a skeet and trap range situated in Örebro county in the centre of Sweden. The range has been in use since the 1920s. According to the authorities and the shooting club approximately 8-9 metric tons of lead pellets have been deposited at an approximate area of 100 x 100 m. The shooting range makes up a small and well defined watershed. The soil consists of app. 1 m thick peat bog on moraine and the groundwater table reaches a minimum level at 30 cm during sommer. Due to the high content of organic matter and saturated conditions, the post-oxic zone is found just a few cm below the surface. A small brook (about 1 m wide and a torrent of about 5 l/s) intersects the shooting range.

Field sampling and chemical analysis. Four groundwater pipes (polypropylene) with filtertips were placed orthogonal to the brook and parallel to the shooting direction. GW1 and GW2 are situated between the shooting positions and the brook while GW3 and GW4 are situated on the opposite side of the brook. Surface water samples were taken at the source of the brook (SW1), tributary at the shooting range (SW2), directly after the shooting range (SW3) and 200 m after the shooting range (SW4).

Water was withdrawn from the pipes and the surface water using a polyethylene syringe connected with a plastic tubing. Three fractions were taken from each sample point: (i) unfiltered (total concentration), (ii) filtered 1.0 mm (Osmonics; PC; 47 mm) and (iii) filtered 0.4 mm (Osmonics; PC; 47 mm). The filtered fractions were divided in the field into three subfractions: (i) total, (ii) anionic and (iii) cationic. The method is a modification of the technique used by Pettersson et al. (1993). Anionic and cationic forms were defined as the fraction retained by an anion exchange resin (QAE SephadexÔ A-25, Amersham Pharmacia Biotech AB, Sweden) and a cation exchange resin (SP SephadexÔ C-25, Amersham Pharmacia Biotech AB, Sweden), respectively. All samples collected for metal analysis were kept in acid washed polyethylene bottles, acidified with nitric acid (Merck; suprapur) in the field. Electrical conductivity and pH were measured in the field. Samples for determination of anions, organic carbon and alkalinity were taken in non-acid washed bottles. At the arrival at the laboratory the ion exchangers were separated from the solutions using centrifugation: 40 ml was centrifuged at 3000 rpm for 10 minutes using a Sorvall Super T21. The supernatant was decanted and acidified for metal analysis. The metals were analysed using an ICP-MS (HP 4500 Series 200) fitted in a class 1000 cleanroom. Anions were analysed using capillary electrophoresis (HP ^3DCE) after filtering through 0.4 mm (Osmonics; PC; 25 mm). Total organic carbon (TOC) was analysed using a TOC-5000 Total Organic Carbon Analyzer (Shimadzu).

Dissolution experiments. Granulated lead pellets (0.5 g; 0.5-2 mm; 2-3 cm²; pa; Merck) were placed in centrifugation tubes (40 ml; PPCO; Nalgene) and covered with water (35 ml; from SW1 and SW3).

Table 1: The surface waters used in the dissolution experiments

	pH	Cond (uS/cm)	Abs 254 nm	Alk (meq/l)	TOC (ppm)	TC (ppm)	IC (ppm)	Chloride (ppm)	Sulphate (ppm)	Nitrate (ppm)
SW1	6.51	126	0.058	0.63	3.22	16.46	13.24	3.29	10.34	0.39
SW3	6.73	131.2	0.124	-	4.49	15.47	10.98	16.19	8.09	0.33

Both unfiltered and filtered (0.40 mm; PC; Poretics) water were used in triplicates for each system, giving a total of 12 replicate systems. After 3, 6, 24, 92 and 296 h the aqueous phase was changed by centrifugation at 10 000 rpm for 10 minutes (Sorvall Super T21). The water was decanted from the tubes and replaced with fresh water. The concentration of lead was measured with flame AAS (Aanalyst 800, Perkin Elmer), as well as the change in pH.

Results and discussion

Field study. The total concentration of lead in the brook increased with approximately 50% when passing the shooting range (0.88 ppb for SW1 compared to 1.25 ppb for SW3). The low contribution of lead to the brook indicates that it is retained by the organic soil. The total concentrations found in the groundwater are somewhat higher (between 6 and 50 ppb) and between 96% and 100% of the lead was found to be anionic (probably bound to large organic structures such as humic and fulvic acid). However, 60% of the lead appeared to be cationic as well. This discrepancy can be explained by the fact that the cation exchanger used contains sulphopropyle groups (i.e. a very strong cation exchanger). If the solution contains lead coordinated to organic matter or inorganic colloids with weaker forces of attraction than to the cation exchanger, the lead ions will be attached to the cation exchanger instead and be perceived as cationic. The results imply that lead is attached to organic groups with 60% of them with weaker association constants than sulphopropyle.

Table 2: Analytical parameters for the different sampling points at the shooting range

Point	pH	Cond (uS/cm)	Cl^- (ppm)	SO4^2-(ppm)	Alk (meq/l)	TOC (ppm)	TC (ppm)	IC (ppm)
SW1	6.7	135	3.2	3.5	0.63	3.22	16.46	13.24
SW2	7.1	128	14.1	2.6	0.51	7.862	17.50	9.638
SW3	6.73	131.2	16.19	8.09	-	4.49	15.47	10.98
SW4	6.98	156	14.2	1.9	0.39	6.49	19.15	12.66
GW1	6.74	323	3.19	-	2.39	16.96	45.70	28.74
GW2	7.03	329	4.59	2.64	2.23	16.99	55.26	38.27
GW3	6.75	129	21.61	3.96	1.138	-	-	-
GW4	6.76	117	-	-	0.57	6.73	18.57	11.84

Figure 1: The fractions of anionic (dotted bar) and cationic (striped bar) lead at the different sampling sites and the total concentration of lead (solid line)

Dissolution experiment. Already after 3 hours the concentration of lead in all systems reached at least 800 ppb. After 296 h (5 volumes of 35 ml) the total amount leached is around 200 mg which represents approximately 0.4% of the original amount.

Figure 2: The accumulated leached amount of lead (mg) at the different sample times.

Generally the water from SW3 has solubilized more lead from the pellets than the water from SW1. The only significant chemical differences (table 1) is the higher concentrations of chloride and organic matter in SW3 (Drane, 1976). SW1 also has higher concentrations of sulphate and inorganic carbon which can have a passivating impact on the corrosion (Graedel, 1994). The slightly higher concentrations of lead in the filtered samples might be induced by lead that is sorbed to colloids (<0.40 mm) and particles (>0.40 mm) that are removed from the solution during centrifugation of the unfiltered samples. In an attempt to elucidate possible solubility limiting stoichiometric phases the log SI was calculated for some relevant lead minerals (table 3). A value around 0±0.5 indicate equilibrium with that mineral (Jurinak and Tanji, 1993). The results indicate equilibrium with PbCO₃(s) and/or Pb₃(CO₃)₂(OH)₂(s) for all systems. The concentrations are well below saturation for all other minerals.

Table 3: Calculated^a log SI at 296 h

	PbO	Pb(OH)2	PbCl2	PbSO4	PbCO3	PbOHCl	Pb3(CO3)2 -(OH)2	Pb(OH)1.5-(SO4)0.25	PbOH-(SO4)0.5
SW1	-5.39	-0.79	-11.32	-3.56	0.54	-4.32	0.28	-2.38	-5.94
SW1 filt	-5.33	-0.73	-11.26	-3.50	0.60	-4.26	0.45	-2.32	-5.88
SW3	-5.37	-0.77	-9.92	-3.64	0.48	-3.61	0.18	-2.38	-6.08
SW3 filt	-5.33	-0.73	-9.88	-3.60	0.51	-3.57	0.29	-2.35	-6.04

^aThe calculations are based on dissociation constants found in Karlsson (1987) and Lindsay (1979). The calculations were performed at pH 7.

Despite the fact that the laboratory experiments indicate a high dissolution of metallic lead in the natural waters very low concentrations are found in the surface waters at the shooting range. The reason for the low mobility in the field system as compared to the laboratory experiments can have several explanations; (i) the soil is to impermeable for dissolved species, (ii) the corrosion of the lead pellets is suppressed and (iii) the dissolved species are retained by the soil. The soil is not impermeable for water and thus not for dissolved species since the shooting range is a discharge area. Since the lead pellets are deposited on the soil surface it will be in contact with both rain- and soil water. The laboratory studies have shown that the corrosion is rapid and that the aqueous phase probably equilibrates with PbCO₃(s) and/or Pb₃(CO₃)₂(OH)₂(s). There is hence no reason to believe that the pellets will be passivated by carbonate even if the partial pressure of CO₂ is higher in the soil. The dissolved lead species are thus probably retained by the soil through hydrophobic interactions since the soil is highly organic and most of the lead is coordinated to organic matter.

Conclusions

Considerably lower concentrations of lead were found in the surface waters compared to the groundwater. In the groundwater all lead was found to be anionic and it is assumed that the organic soil retains the lead by hydrophobic interactions. The change from cationic lead to anionic lead in the surface water when passing the shooting range supports the hypothesis that the lead is leached from the shooting range as organically coordinated. The dissolution experiments showed that metallic lead in contact with natural water would rapidly give very high concentrations and that probably PbCO₃(s) and/or Pb₃(CO₃)₂(OH)₂(s) are limiting the concentration.

References

Drane, C.W., (1976). Natural waters. In: L.L. Shreir (Editor), Corrosion. Metal/Environment reactions. Newnes-Butterworths, London, pp. 2:38-2:50.

Graedel, T.E., (1994). J. of Electrochem. Soc. 141(4): 922-927.

Jurinak, J.J. and Tanji, K.K., (1993). J. of Irr. and Drainage Eng. 119(5): 848-867.

Karlsson, S. (1987) Ph.D. Thesis. Linköping Studies in Arts and Science, 10, Linköping University, Sweden

Lindsay, W.L., (1979). Chemical equilibria in soils. John Wiley & Sons, New York.

Pettersson, C., Håkansson, K., Karlsson, S. and Allard, B., (1993). Wat. Res. 27(5): 863-871.