Solubility and Changes of Mercury Binding Forms in Contaminated


Solubility and Changes of Mercury Binding Forms in Contaminated...

0 downloads 194 Views 211KB Size

Environ. Sci. Technol. 1998, 32, 2755-2762

Solubility and Changes of Mercury Binding Forms in Contaminated Soils after Immobilization Treatment HARALD BIESTER* AND HOLGER ZIMMER Institute of Environmental Geochemistry, INF 236, 69120 Heidelberg, Germany

Mobility at different pH and binding forms of mercury (Hg) have been investigated in three Hg-contaminated soils after immobilization treatment with alkali-polysulfide (APS) and trimercapto-s-triazine trisodium salt solution (TMT). Changes of solid-phase Hg binding forms after immobilization were determined by Hg pyrolysis. Hg concentrations in the water extracts of all samples increased after treatments due to the formation of soluble mercury sulfides (APS treatment), and the mobilization of humic acid bound Hg at the high pH of the reagents. In contrast, Hg concentrations decreased sharply at low pH due to decomposition of soluble mercury sulfides and precipitation of humic acidbound Hg. Inorganic Hg compounds such as Hg0 or HgCl2 are effectively transformed to mercury sulfides by APS treatment, whereas TMT could transform HgCl2 but not Hg0. Both reagents were found to affect humic acid bound Hg by way of increasing Hg desorption temperatures, although APS was found not to desorb Hg completely from humic acids and TMT-Hg complexes are actually incorporated into humic acids.

Introduction Contamination of soils with mercury (Hg) is a serious problem due to the high toxicity of the metal and its compounds. Methods for remediation of Hg-contaminated sites include thermal treatment, leaching procedures, electrolysis (1-3), or a combination of these methods if additional contaminants are present. These methods are usually costly and timeconsuming. Therefore, in situ or on site immobilization of Hg in contaminated soils by chemical treatment has been considered as a stabilization treatment to reduce environmental risks (4) and as a chemical pretreatment of contaminated soils to maintain threshold values for landfill deposition. In the case of large contaminated areas or where the soil cannot be excavated, in situ immobilization of soluble Hg compounds is considered an attractive alternative to stop or prevent groundwater contamination. Due to the high affinity of Hg compounds for sulfur, the use of sulfurcontaining reagents such as alkali-polysulfides or organic compounds such as trimercapto-s-triazine salts have been offered by remediation companies for the immobilization of Hg in soil materials. Most of the available data concerning the reaction of Hg with immobilization reagents are based on the removal of Hg from wastewater by the formation of insoluble Hg compounds (5). In general, the aim of using sulfur-containing reagents is to reduce the mobility of soluble or volatile Hg compounds in soils by forming stable Hg * Corresponding author tel: +49-6221-544819; fax: +49-6221545228; e-mail: [email protected]. S0013-936X(97)00937-1 CCC: $15.00 Published on Web 08/11/1998

 1998 American Chemical Society

compounds such as metacinnabar (HgS) or insoluble organic Hg-S compounds. It is unclear how and to what extent Hg compounds in soils are transformed during such immobilization treatment. Although the magnitude of Hg mobility reduction can be determined by leaching tests or by the quantification of degassing volatile Hg compounds, there is no published data regarding the quality and longterm stability of Hg compounds formed during immobilization. It is also unclear how the immobilizing reagents react with different Hg compounds in soils contaminated with metallic Hg (Hg0), HgCl2, or humic acid (HA)-bound Hg. Here, we report that Hg binding forms in three Hg-contaminated soils after treatment with two different immobilizing reagents. Before and after treatment, solid-phase Hg binding forms were determined by a pyrolysis technique similar to that used in earlier studies to distinguish Hg binding forms in soils by thermal Hg desorption (6-10). The results of the solid-phase measurements were compared to those of standard Hg-S compounds obtained by the reaction of the reagents with HgCl2. It is known from numerous studies that coupling of Hg to humic substances is the predominant Hg binding form in most natural soils (11-13). Therefore, one objective of our study was to determine the changes of humic acid bound Hg binding characteristics during the immobilization process. We extracted HA from soils before and after treatment and compared changes in Hg content and Hg desorption temperatures. Additionally, we investigated aqueous-phase Hg mobility in the untreated and treated soils through leaching tests at different pH values.

Materials and Methods Soil Samples. Soil samples were collected from three Hgcontaminated sites having different soil types and Hg pollution histories. Sample CAP was taken from the top soil layer of a former chlor alkali plant (Bitterfeld, Germany), where Hg0 was directly spilled into the soil. The dark brown soil consists mostly of sand and humic materials together with small amounts of silty and clayey components (9.8%). KYA soil was sampled from soils of a wood pressure treatment site in Bad Krozingen, SW Germany. The site was contaminated by HgCl2 washed off from treated timber. This loess soil consists of high amounts of carbonates (27%), clayey and silty components (61.8%), and comparatively small amounts of organic matter. Sample REC was taken as a mixed sample from a 2.0-2.9-m section of soil core taken from a former Hg recycling site in Frankfurt a.M., Griesheim, where large amounts of metallic Hg were spilled into the top soil. This soil consists of well-sorted medium grained sands. Clayey components account for only 6.3% and the content of organic carbon is low. Immobilization. Immobilization tests were carried out using either a sodium polysulfide solution (APS) (AC 2000/ aqua control) containing 5.29 mol/L sulfur or an aqueous solution (15%) of trimercapto-s-triazine trisodium salt (Na3C3N3S3) (TMT 15, Degussa) containing 1.85 mol/L sulfur. For each test, 200 g of sample and 600 mL of demineralized water were placed into a 1-L centrifuge bottle. After ultrasonic dispersion (KLN System 582) of the samples, the reagents were added at a molar concentration double the content of Fe, Mn, Cu, Pb, Zn, Cd, Sn, As, and Hg in the sample. The reagent concentrations were calculated to produce equimolar amounts of sulfur and metals. The samples were shaken end to end for 2 h and centrifuged (1 h/4300 rpm). The supernatants were decanted and filtered through a 0.45-µm nitrate cellulose filter. The solid residues were thoroughly homogenized and frozen (-18 °C) until analysis. The pH of VOL. 32, NO. 18, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2755

TABLE 1. Concentrations of Metals with Affinity to Sulfur (Fe, Mn, Cu, Pb, Zn, Cd, Sn, Hg, As), Total Hg Content, and Content of Organic Carbon (Corg) in Hg-Contaminated Soil Samples (CAP, KYA, REC) sample

total Hg (mg/kg)

∑ Fe, Mn, Cu, Pb, Zn, Cd, Sn, Hg, As (mol/kg)

Corg (%)

CAP KYA REC

1717 ( 331 161.3 ( 15.1 23.05 ( 2.44

0.22591 0.29166 0.31779

1.4 ( 0.08 0.7 ( 0.05 0.1 ( 0.008

the solutions was determined directly after centrifugation by means of a glass electrode. Hg concentrations of the solutions were determined by cold vapor atomic absorption spectrometry (CV-AAS) after digesting 5 mL of the solution in 20 mL of aqua regia (2 h/160 °C) and reducing Hg2+ with stannous chloride using an automated Hg analysis system (TSP mercury monitor 3200). Standard Substances. Standard substances for mercury sulfur compounds were derived from the reaction of Hg2+ with APS or TMT using a 2-fold excess of the reagents to a 0.01 M HgCl2 solution. The solutions were shaken for 2 h, centrifuged, and decanted. The precipitates were washed four times, freeze-dried, and stored frozen until analysis. Polysulfides are generally described to react with mercury according to

Hg + Sn2- f HgS + Sn-12-

(n ) 3-6)

(1)

The reaction of Hg with TMT leads to the formation of insoluble organic complexes according to S– 2+

Hg

+

S–

N

S–

N

S Hg S

S–

N

2

N

N S–

N

N S–

N

N S– (2)

However, it is assumed that the Hg binding is intermolecular as well as intramolecular with the two S ligands of the TMT molecule (14). Moreover, it is assumed that TMT-Hg molecules can exist as monomers or polymers (15). Solid-phase standards for HgCl2 and cinnabar were obtained by mixing 0.001 M HgCl2 (Merck) and 0.001 M HgS (red cinnabar, Merck) with 20 g of quartz powder for dilution. Carboniferous schists bearing visible droplets of metallic Hg were used as a standard of unbound metallic Hg. Hg0 incubated iron oxyhydroxides were prepared by incubating dry iron oxyhydroxides in a sealed container for 14 d/40 °C in a Hg0-saturated atmosphere. Humic acids were extracted from untreated and treated samples CAP and KYA using 0.1 M NaOH according to the standard procedure of Calderoni and Schnitzer (16). Due to the low amount of organic carbon in REC (Table 1), we could not extract sufficient amounts of solid HA from this soil to determine the Hg content of the HA-bound Hg fraction or to analyze Hg desorption characteristics. The HA fraction was precipitated by acidifying the extracts to pH KYA > REC. These results indicate that the formation of SMS is the predominant process causing the increase in Hg mobility after APS treatment, whereas TMT mainly increases the Hg mobility by mobilization of HA at high pH. We conclude that the strong decrease of the Hg concentrations in the acid extracts is attributed to the precipitation of HA and the decomposition of mercury polysulfides to HgS, S0, and H2S in the APS-treated samples. The decomposition of the polysulfides could be observed by the precipitation of white sulfur (S0) and the smell of H2S during the acid leaching. Conversely, the lower Hg concentrations in the extracts after the acid leaching of the TMT-treated samples CAP and KYA are caused by precipitation of the Hg-bearing HA mobilized during the treatment. Different from CAP and KYA, Hg concentrations in the water extracts of treated REC were not higher than Hg concentration in the pH 3 extracts (Figure 8). Moreover, Hg concentrations in the extracts of the TMT- and APS-treated REC do not show the same high differences as found for the other samples. The comparatively low Hg concentrations in the water extract of the APS-treated sample REC are attributed to the fact that more than 46% of the total Hg in this sample was previously extracted during the APS treatment. The TMT-treated sample generally does not show large differences in the Hg content of the extracts obtained by leaching at different pH, whereas the APS-treated sample shows increasing Hg concentrations in the extract after the pH 3 leaching (Figure 8). As the absence of easily reducible Hg in the extracts confirms that metacinnabar or TMT-Hg were not dissolved during the acid leaching, we assume that one reason for the high Hg concentrations in the pH 3 extracts (Figure 8) is the low amount of gleyey and organic matrix components in this soil where HgS or TMT-Hg complexes could be adsorbed. Moreover, no coprecipitation of HgS or TMT-Hg by precipitation of HA occurs due to the lack of HA in this sample. Despite the fact that Hg release curves after APS and TMT treatment indicate successful transformation of all Hg in REC (Figure 7), which is assumed to be due to the lack of organic materials, Hg compounds formed in this sample seem to be more easily remobilized at low pH for the same reason.

2762

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 18, 1998

Acknowledgments This research was funded by the Environmental Ministry of Baden Wu ¨rttemberg/Germany Projekt Wasser, Abfall, Boden (PWAB) No. PD 95172.

Literature Cited (1) Stepan, D. J.; Fraley, R. H.; Charlton, D. S. Remediation of Mercury-Contaminated Soils: Development and Testing of Technologies; Topical Reports of the Gas Research Institute; GRI-94/0402; GSI: Palo Alto, 1995; 40 pp. (2) Pedroso, A. C. S.; Gomes, L. E. R.; De Carvalho, J. M. R. Environ. Technol. 1994, 15, 657-667. (3) Charlton, D. S.; Harju, J. A.; Stepan, D. J.; Ku ¨ hnel, V.; Schmit, C. R.; Butler, R. D.; Henke, K. R.; Beaver, F. W.; Evans, J. M. In Mercury Pollution-Integration and Synthesis; Watras, C. J., Huckabee, J. W., Eds.; Lewis Publishers: Chelsea, MI, 1994; pp 595-600. (4) Tho¨ming, J.; Sobral, L. G. S.; Santos, R. L. C.; Hempel, M.; Wilken, R. D. Abstracts of the Fourth International Conference on Mercury as a Global Pollutant, Hamburg, Germany; 1996; p 161. (5) Findlay, D. M.; McLean, R. A. N. Environ. Sci. Technol. 1981, 15, 1388-1390. (6) Azzaria, L. M.; Aftabi, A. Water, Air Soil Pollut. 1991, 56, 203217. (7) Bombach, G.; Bombach, K.; Klemm, W. Fresenius J. Anal. Chem. 1994, 66, 18-20. (8) Windmo¨ller, C.; Wilken, R. D.; Jardim W. Water, Air Soil Pollut. 1996, 89, 399-416. (9) Biester, H.; Scholz, C. Environ. Sci. Technol. 1997, 31, 233-239. (10) Biester, H.; Nehrke, G. Fresenius J. Anal.Chem. 1997, 358, 44464452. (11) Meili, M. Water, Air Soil Pollut. 1991, 56, 333-348. (12) Schuster, E. Water, Air Soil Pollut. 1991, 56, 667-680. (13) Stein, E. D.; Cohen, Y.; Winer, A. M. Crit. Rev. Environ. Sci. Technol. 1996, 26 (1), 1-43. (14) Nakamura, Y. Patent Ger. Offen. 2,240,733 (Cl. A 01n), March 8, 1973; Japan Appl. 7163740, August 20, 1971, 17 pp. (15) Feher, F.; Hirschfeld, D.; Linke, K. H. Z. Naturforsch. 1962, 17b, 624. (16) Calderoni, G.; Schnitzer, M. Geochim. Cosmochim. Acta 1984, 48, 2045-2051. (17) Mu ¨ ller, G., Gastner, M. N. Jb. Miner. Mh. 1971, 10, 466-469. (18) Meili, M.; Iverfeld, A° .; Ha˚kanson, L. Water, Air Soil Pollut. 1991, 56, 439-453. (19) Hintelmann, H.; Welbourn, P. M.; Evans, R. D. Environ. Sci. Technol. 1997, 31, 489-495. (20) Skyllberg, U. L.; Bloom, P. R.; Nater, E. A.; Xia, K.; Bleam, W. F. Proceedings of Extended Abstracts from the Fourth International Conference on the Biogeochemistry of Trace Elements, Berkeley, CA, June 23-26, 1997; pp 285-286. (21) Kerndorf, H.; Schnitzer, M. Geochim. Cosmochim. Acta 1980, 44, 1701-1708. (22) Allard, B.; Arsenie, I. Water Air Soil Pollut. 1991, 56, 457-464. (23) Weber, J. H.; Reisinger, K.; Stoeppler, M. Environ. Technol. Lett. 1985, 6, 203-208. (24) Weber, J. H. Binding and Transport of Metals by Humic Materials. In Humic Sustances and their Role in the Environment; Frimmel, F. H., Christman, R. F., Eds.; John Wiley & Sons: New York, 1988; pp 165-178. (25) Paquette, K. E.; Helz, G. Environ. Sci. Technol. 1997, 31, 21482153. (26) Andersson, A. Mercury in Soils. In The Biogeochemistry of Mercury in the Environment; Nriagu, J. O., Ed.; Elsevier/Holland Biomedical Press: Amsterdam, 1979; pp 79-112. (27) Hempel, M.; Wilken, R. D.; Miess, R.; Hertwich, J.; Beyer, K. Water, Air Soil Pollut. 1995, 80, 1089-1098.

Received for review October 23, 1997. Revised manuscript received June 9, 1998. Accepted June 25, 1998. ES9709379