Elsevier

Chemosphere

Volume 54, Issue 1, January 2004, Pages 79-87
Chemosphere

Concentrations of dioxin-like PCB congeners in unweathered Aroclors by HRGC/HRMS using EPA Method 1668A

https://doi.org/10.1016/S0045-6535(03)00664-7Get rights and content

Abstract

We have determined the congener compositions of nine commercial Aroclor products of polychlorinated biphenyls (PCBs) to the sub-part-per-million level using high-resolution gas chromatography combined with high-resolution mass spectrometry according to US Environmental Protection Agency (EPA) Method 1668A. These Aroclor composition data should allow improved characterization and risk assessment of PCB contamination at hazardous waste sites, particularly for dioxin-like PCB congeners. By combining the data on the concentrations of each dioxin-like congener with its World Health Organization toxicity equivalency factor, we have established dioxin toxic equivalent concentrations for each pure Aroclor product.

Introduction

Polychlorinated biphenyls (PCBs) are a group of chlorinated aromatic hydrocarbons once widely used in industry as heat transfer fluids, hydraulic lubricants, flame retardants, plasticizers, and as dielectric fluids in electronic components such as capacitors and transformers. The thermal and chemical stability properties that made PCBs ideal for such applications also make them environmental contaminants that are slow to degrade and may bioaccumulate through the food chain (Safe, 1992). In addition, because PCBs are a mixture of up to 209 distinct congeners, laboratory analysis and risk assessment of PCBs is particularly challenging.

Most PCBs were commercially produced in the United States as standard mixtures bearing the brand name Aroclor. The reaction and separation conditions for production of each Aroclor favor the synthesis of certain congeners, giving each Aroclor a unique signature or pattern based on its congener composition. No Aroclor contains all 209 congeners; in fact, 110–120 congeners typically account for over 95% of the total mass in each Aroclor (Frame et al., 1996).

Determination of PCBs in the environment has traditionally focused on identifying and quantifying the Aroclor(s) by gas chromatography with an electron capture detector (GC/ECD). The analysis is both rapid and inexpensive and allows for estimation of total PCB concentration as the sum of the concentrations of Aroclors present. The Aroclor is identified by the retention times of the three to five highest peaks in the chromatogram, and is quantified by comparing the height or area of those peaks to those of a pure Aroclor standard (Erickson, 1997). However, the composition of Aroclors in the environment has been shown to change over time due to “weathering”––chemical and physical transformations that can alter the composition of a sample. Weathering occurs due to differences in volatilization, partitioning, chemical transformation, photo-degradation, biodegradation, or bioaccumulation of individual PCB congeners (Erickson, 1997). Weathering can make it difficult to match an environmental sample with an Aroclor pattern, leading to difficulties in identification and quantification of PCBs.

The presence of mixtures of Aroclors may also complicate the identification of Aroclors. Under these conditions, Aroclor analysis may provide a poor estimate of total PCB concentration. An alternative procedure involves quantifying the concentrations of individual congeners in PCB-contaminated media using GC/ECD or GC coupled with either low- or high-resolution mass spectrometry (GC/LRMS or GC/HRMS). Such techniques provide a more accurate estimate of total PCB concentrations, regardless of the extent of weathering or number of Aroclors present. However, GC/MS methods, particularly the HRMS version, are considerably more expensive than GC/ECD Aroclor analysis.

Congener analysis provides a wealth of data about the composition of PCBs in environmental samples, but risk assessors currently cannot assess human health risk from exposure to each individual congener because toxicity data are unavailable for most congeners. Historically, PCB risk assessments at Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA; “Superfund”) sites have focused on risks from exposure to total PCBs using the results of Aroclor analysis. More recently, however, focus has been on a subset of 12 PCB congeners that have demonstrated toxicity in mammals, including humans, similar to that of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, “dioxin”), and other chlorinated dibenzo-p-dioxins and dibenzofurans (CDDs/CDFs) (Safe, 1992; Van den Berg et al., 1998). Risk assessors evaluating sites with PCB contamination have begun to evaluate risks from these “dioxin-like” congeners separately from the risks associated with total PCBs, using toxicity equivalency factors (TEFs) to estimate a dioxin toxicity equivalence for the dioxin-like congeners (EPA, 1996). Thus, a congener analysis performed with appropriate specificity and accuracy is useful for improved estimates of total PCBs and determination of the 12 dioxin-like congeners.

The publication of TEFs (Ahlborg et al., 1994) has facilitated assessment of dioxin-like risk. However, in order to assess this risk directly, knowledge of the concentration of the dioxin-like congeners in an environmental PCB sample is necessary. Congener composition of the Aroclors has been studied extensively by Frame et al., 1996, Frame, 1997a, Frame, 1997b, Frame, 1999a, Frame, 1999b. These and other studies (Duinker et al., 1988; Schulz et al., 1989) established Aroclor composition to approximately the 0.01% (100 parts per million; ppm) level, but this level of scrutiny is insufficient for detecting and quantifying concentrations of certain dioxin-like congeners that are present at ppm and sub-ppm concentrations in pure Aroclor. Several groups have reported results to approximately the 5 ppm level: Hong et al. (1993) by GC/ECD; Schwartz et al. (1993) by GC/ECD and GC/LRMS; and Kannan et al. (1987) by GC/ECD and GC/LRMS for congeners 77, 126, and 169.

In this study, we determined the concentration of individual congeners in each of nine different Aroclors to the sub-ppm level using high-resolution gas chromatography combined with high-resolution mass spectrometry (HRGC/HRMS). The analysis uses silica and carbon cleanup columns and multiple GC columns to separate co-eluting congeners, and the most modern HRMS to achieve low detection levels. This analytical system facilitates the isolation and quantification of all of the dioxin-like congeners, including non-ortho congeners 77, 81, 126, and 169; and mono-ortho congeners 105, 114, 118, 123, 156, 157, 167, and 189. The HRGC/HRMS method is significantly more sensitive than LRMS methods, and eliminates many interferences that LRMS and ECD analytical techniques do not. The method achieves a high degree of accuracy by use of isotope dilution for determination of the dioxin-like congeners, particularly at low concentrations. Results from this analysis of commercial Aroclor products can help to address important issues related to the risk assessment of PCB contamination at Superfund and other hazardous waste sites.

Section snippets

Use of HRGC/HRMS

Because of its high sensitivity and specificity, HRGC/HRMS is the analytical technique of choice for determining the concentrations of individual PCB congeners, especially the dioxin-like congeners. EPA has developed EPA Method 1668A for this purpose (EPA, 1999). EPA Method 1668A uses 13C12-labeled PCB congeners for isotope dilution quantification of the dioxin-like PCB congeners and of congeners with the earliest and latest retention time at each level of chlorination (LOC). Other congeners

Congener concentrations

Results of analyses of nine commercial Aroclors by HRGC/HRMS are presented in Table 2. For each of the nine Aroclors, Table 2 provides the concentration of each of the 12 dioxin-like congeners in μg/g (ppm). Each congener is identified by its IUPAC number, as listed in EPA Method 1668A (Guitart et al., 1993). The bottom (bolded) row in Table 2 shows the TEQ for each Aroclor, which is calculated from the concentration of dioxin-like congeners in the Aroclor and the 1998 WHO TEFs (Van den Berg et

Discussion

This section provides a perspective on the current results by comparing them to results of previous determinations of Aroclor composition, and discusses major sources of uncertainty associated with the analysis.

Application

The results reported in this paper can be used to help risk assessors evaluate the extent of PCB weathering in environmental samples, which may be useful for evaluating risks posed both by PCBs in total and by dioxin-like PCB congeners specifically. Physical and chemical transformations of PCB contamination in environmental samples may lead to enhancement of dioxin-like congener concentrations. This enhancement could result in added risk relative to that estimated from Aroclor data. The results

Acknowledgements

The authors acknowledge the contributions of Val Scott, Yvonne Fried, Kathie Coffey, and Todd Fisher at Axys Analytical for analyses of the Aroclors; Alfred Mayo, Ward Jacox, Pornkeo Chinyavong, and Michael Walsh at DynCorp I&ET for data review and calculations; Emily Levin and Adena Greenbaum at Industrial Economics for assistance with risk calculations; and David Bennett, Jim Cogliano, Phil Cook, and Tala Henry of the US Environmental Protection Agency for useful insights.

Funding for this

References (22)

  • R. Guitart et al.

    Chemosphere

    (1993)
  • T.R. Schwartz et al.

    Chemosphere

    (1993)
  • U.G. Ahlborg et al.

    Chemosphere

    (1994)
  • Brown, J.F., Jr., Silkworth, J.B., Mayes, B.A., 1997. Characterization of PCB Composition, Tissue Accumulation, and...
  • D.E. Burgin et al.

    Environ. Health Perspect.

    (2001)
  • J.C. Duinker et al.

    Anal. Chem.

    (1988)
  • M.D. Erickson

    Analytical Chemistry of PCBs

    (1997)
  • EPA, 1996. PCBs: Cancer Dose–Response Assessment and Application to Environmental Mixtures, EPA/600/P-96/001F,...
  • EPA, 1999. Method 1668, Revision A: Chlorinated Biphenyl Congeners in Water, Soil, Sediment, and Tissue by HRGC/HRMS,...
  • EPA, 2001. Data Review Guidelines for Chlorinated Biphenyl Analysis by EPA Method 1668, Revision A: Draft, available...
  • EPA, 2002. See Supporting Information...
  • Cited by (92)

    View all citing articles on Scopus

    The views expressed in this paper are those of the authors and do not represent the views of any US Government Department or Agency.

    1

    Present address: MicroMass, 4901 Lockside, Sidney, British Columbia, Canada V8Y 2E6.

    View full text