Investigations into the Vertical Distribution of PCDDs and Mineralogy in Three Ball Clay Cores from the United States Exhibiting the Natural Formation Pattern DAMIEN GADOMSKI,* ,²,‡ MATS TYSKLIND, ² ROBERT L. IRVINE, ‡ PETER C. BURNS, ‡ AND ROLF ANDERSSON ² Department of Civil Engineering and Geological Sciences, University of Notre Dame, Notre Dame, Indiana, 46556, and Environmental Chemistry, Department of Chemistry, Umea˚ University, SE-901 87 Umea˚, Sweden In this study, we report the PCDD and mineralogical results from the analyses of 27 different samples from three ball clay cores from different locations in Kentucky and Tennessee. One goal of this study was to determine if there is a correlation between the mineralogy of the ball clay samples and the PCDD concentrations and/or homologue profiles in each sample. Samples from each of the three cores exhibited the natural formation profile with extremely high PCDD concentrations with low and mostly undetectable levels of polychlorinated dibenzofurans (PCDFs). The maximum toxic equivalents (TEQs) for Cores C-E were 2500, 440, and 15 000 pg WHO-TEQ/g, respectively. Although there does not seem to be a direct correlation between mineralogy and PCDD concentrations or homologue profiles, the mineralogy of Core C is substantially different than that of Cores D and E, which may in part explain the differences in congener patterns we observed among the three cores. Introduction In recent years, there have been many reports of an unusual PCDD/F profile in certain environmental matrixes, which cannot be traced back to any known anthropogenic source(s). These include wetland and marine sediment samples (1-5) as well as ball clay and kaolin samples from different parts of the United States, Germany, and Spain (6-11). All these samples contained elevated levels of PCDDs, dominated in most cases by OCDD and decreasing concentration with decreasing chlorination. PCDFs were found at low or even undetectable levels in all samples. Moreover, the 1,2,3,7,8,9- HxCDD peak (which is coeluted with 1,2,3,4,6,7-HxCDD on a DB-5 column but will be referred to as 1,2,3,7,8,9-HxCDD throughout this paper) dominates the 2,3,7,8-substituted HxCDD isomers, which are not found in known sources, such as combustion related samples. One general difference between the profiles reported for the sediment and clay samples exhibiting the natural formation profile is that the lower chlorinated 2,3,7,8-substituted PCDD congeners are found in much higher concentrations in the clay samples. Although it has been suggested by many researchers that the dioxins found in these particular environmental samples are of natural origin, no definitive evidence has been brought forward to explain the mechanisms and precursors involved in the formation and transformation of these compounds. In 2000, Ferrario et al. reported results from the analysis of eight raw and four processed ball clay samples from Sledge, MS (6). Each of the raw ball clay samples was dominated by OCDD followed by 1,2,3,4,6,7,8-HpCDD, or in two cases, 1,2,3,7,8,9-HxCDD among the 2,3,7,8-substituted PCDDs. 2,3,7,8-TCDD and 1,2,3,7,8-PeCDD were found at extremely high concentrations in the raw ball clay samples with averages of 711 and 508 pg/g dry weight (dw), respectively. Average PCDF values were lower than what was reported for an urban background sample, indicating the clear dominance of PCDDs. There was a distinct change in homologue profiles from the raw to processed samples analyzed by Ferrario et al. (6). Although OCDD was also found in the highest concentration in the processed clay samples, its absolute and relative concentration was much higher than in the raw ball clay samples, with an average OCDD concentration exceeding 200 000 pg/g. Additionally, there was a dramatic decrease in the concentration of TCDD in the processed samples. The average 2,3,7,8-TCDD concentration of the raw ball clay samples was 711 pg/g, whereas the processed samples had an average of 50 pg/g. The authors could not confirm if the extreme differences in the homologue profile of the raw clay samples to that of the processed ones were due to the processing or if the changes reflected the loss and/or formation of specific isomers. Later, Rappe et al. reported PCDD/F results from the analysis of U.S. and German kaolin samples and U.S. ball clay samples (8). The data from the study indicated that the German kaolin and U.S. ball clay both contained elevated concentrations of PCDDs and low or undetectable levels of PCDFs. In addition, the tetra-, penta-, and hexaCDD patterns were similar in both the U.S. ball clay and the German kaolin. Although the patterns were similar, the absolute concentra- tions were much higher in the U.S. ball clay. The mean WHO- TEQ (WHO-TEQ values established by Van den Berg et al. (12)) of the German kaolin was 128 pg/g, whereas the mean of the U.S. ball clay samples was 1035 pg WHO-TEQ/g dm. This study clearly demonstrated that the distinct non- anthropogenic pattern in ball clays and kaolin is not limited to the southern U.S. In 2002, Ferrario et al. reported additional results from the analysis of processed ball clay mined in the U.S. (10). The 2,3,7,8-substituted PCDD concentrations of the processed ball clay samples were extraordinarily high with an average WHO-TEQ value of 3200 pg/g. Congener distribution and the homologue profile of the samples were remarkably similar to previously analyzed samples with the elevated levels of the tetra- to octa-PCDDs. Results demonstrated the stable and reproducible nature of the isomer patterns in the ball clay (10). Recently, Abad et al. reported the results from a com- prehensive study of PCDD contents in binder and anticaking agent feed additives from Spain (11). The samples analyzed included minerals such as sepiolite, bentonite, zeolite, vermiculite, and kaolin. All 16 samples reported except kaolin had TEQ levels lower than the European Union (EU) limit of 0.5 pg WHO-TEQ/g. The results from the two kaolin * Corresponding author phone: +46 90 7869241; fax: +46 90 128133; e-mail: dgadomsk@nd.edu. ² Umeå University. ‡ University of Notre Dame. Environ. Sci. Technol. 2004, 38, 4956-4963 4956 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004 10.1021/es049579h CCC: $27.50  2004 American Chemical Society Published on Web 09/02/2004