Toxaphene Technical Overview, Part 2
June, 1999

Overview

The pesticide toxaphene was banned from use and manufacture two decades ago. Produced in Brunswick, Georgia from about 1948 to 1980 the mixture gained popularity after DDT was outlawed. Toxaphene’s main use was as an agricultural crop treatment in the southern and western states. When toxaphene became a significant contaminant of the Great Lakes in the late 1960’s, its use was curtailed and finally eliminated in the U.S. In Brunswick, the manufacturing process resulted in release of as much as 300 pounds of pesticide per day into the Dupree and Terry Creek marshes. Natural erosion and dredging operations spread the contaminated material over a wide area of marsh.

Studies show that toxaphene increases tumors in laboratory animals. The toxicity of toxaphene is well known. The material is a biocide capable of adverse health effects ranging from organ damage to death.

Toxaphene is a complex mixture of related chemicals. Its manufacture began with pine tar treated to make camphene. The camphene was then treated with chlorine. The resulting mixture has over 600 different chemical species, depending on the source of camphene and the amount of chlorine added. Typically, about 200 major chemicals make up Technical grade toxaphene. In the environment some of these chemicals breakdown quickly, while some are very stable. The average component half-life (time to break down) is around 10 years. Its breakdown products are potentially harmful as well. Partially broken down toxaphene is still toxaphene and it still has the potential to produce tumors or kill. Further, the breakdown products often have lower chlorine content, with a greater potential for bioaccumulation.

Toxaphene Analysis

Although toxaphene is a “multi-component” mixture, it is only slightly more complicated to analyze than some single component chemicals. Toxaphene does not breakdown during analysis as some pesticides do, and does not require extensive derivatization or preparation. All of the routine Environmental Protection Agency (EPA) screening protocols, such as the CLP (Contract Laboratory Program) Pesticide protocol, are multi-component tests. So, no major changes are needed for routine toxaphene analysis.

Occasionally, other chemicals “overlap” during testing with one or more of the toxaphene components, but confirmation sampling— a routine and standard procedure— easily distinguishes these chemicals. It is rare for other pesticides to interfere with toxaphene analysis. Toxaphene is far more likely to mask other chemicals than to be confused with a different pesticide. In the Terry Creek Disposal Area, where there is a 50-year history of toxaphene in the drainage, it is clear that the chemical fingerprint is toxaphene rather than something else.

Chemicals are measured using a technique called “chromatography” where each individual chemical in the mixture is separated and measured. All environmental samples are complex mixtures. The only significant difference between multi-component analysis and single component analysis is that multi-component has two or more chemicals that are grouped together using simple addition.

Environmental testing is entirely automated. The sample is injected into an instrument, which separates and measures all chemicals in the sample, and information on the sample is printed by a computer (the report is a pesticide “chromatograph”). The instrument reports the “retention time” (or length of time each individual chemical stays on a “column” before entering a detector), and the “peak” (equal to the amount of each chemical found by the detector). Every environmental chemical, no matter how complex, can be characterized using these two factors. Before actually measuring toxaphene in environmental samples, the instrument is calibrated with a "standard" –  which is just a known amount of toxaphene. Then the unknown samples are run. In essence, the unknown sample chromatograph is compared to the known standard chromatograph. The instrument scores the sample for how similar it is to the known chemical. Then the information in the test is analyzed for the presence, and amount, of toxaphene. There are three basic methods for performing an analysis: the “Total Area” method, the “Toxaphene Task Force” method, and the “congener” method.

The basic concept of the “Total Area” method is that all chemicals in the unknown sample that match the toxaphene standard are added together as if they were one chemical, regardless of where the match occurs during the test. The CLP pesticide method generally follows this measuring protocol. In contrast, the Toxaphene Task Force (TTF) method has a human operator subjectively choosing about five toxaphene components in the last section of the test— the so-called “back-half” procedure. This is a highly biased methodology. Labs other than Hercules/EPA find the TTF method can under-report by as much as 400% compared to total area methods (ATSDR PHA, Terry Creek Dredge Spoil area, December 16, 1998). Another criticism of the “back-half” method is that the latter portion of the test is generally of the higher chlorine containing chemicals. In environmental samples, toxaphene chemicals have often lost chlorine, shifting to the front half of the test. Thus, in real samples the “back half” is least likely to match the standard, further biasing the test. 

For most of the testing at the Terry Creek site, the TTF method seems to be in use. EPA/Hercules argument for using a method that under-reports toxaphene is that other chlorinated chemicals may be counted as toxaphene in the Total Area methodology. However, this is irrelevant at Terry Creek where there is no doubt that toxaphene is present and very widespread. Estimates are that up to 55 tons of toxaphene were dumped each year for the 20+ year history of the product— possibly, over 1,000 tons of toxaphene went into the drainage. Obviously, there is no ambiguity over the presence of toxaphene in this marsh and, therefore, no need to bias the detection methodology.

From the scientific literature, it appears that only Region 4 EPA and Hercules use the biased TTF methodology. Labs other than EPA/Hercules, such as other Federal agencies, private labs, and Universities, invariably report toxaphene with higher concentrations than EPA/Hercules. The Toxaphene Task Force methodology appears to under-report toxaphene in both environmental samples and between toxaphene standards.

Another type of bias arises from standards used to calibrate toxaphene tests for both the TTF and Total Area methods. A paper in the Terry Creek Disposal Area Administrative Record shows that a variety of commercial standards run by a single laboratory showed wide differences (Carlin and Hoffman, undated; The Effect On Calculated Results of Analysis Caused by the Variability Among Toxaphene Reference Standards). However, the differences were more significant for the TTF method. In one experiment, the TTF method averaged 83% of the expected values, with a range of 37% to 113%; but the Total Area method averaged 93% of expected— an acceptable value. While differences in “standard” values are unusual, this paper shows it is not that great a problem when the Total Area method is used to test for toxaphene.

When various laboratories try to compare results using the Toxaphene Task Force methodology, the results have been equally unscientific. In one study between independent labs using the TTF method, the results showed wide variations for the same samples (Bennet Letter dated 6/4/93, titled Report of "Toxaphene Task Force").

The fact is that all of the available standards are toxaphene, as is environmental toxaphene. With carcinogens, it is reasonable and appropriate to use the most conservative standard and most conservative methodology. In EPA Region 4, this does not appear to be the case. The bottom line is: EPA and Hercules are using the TTF methodology knowing that it misses toxaphene in the environment.

The Terry Creek Expanded Site Inspection data evaluates the performance of analyses in EPA Region 4. These data compare environmental analysis using EPA’s standard CLP pesticide analysis program and the Toxaphene Task Force analysis format. In many samples, the TTF data is misses toxaphene while CLP analysis shows high levels of toxaphene. For example, a surface sample TC-SS-15 is blank using the TTF test, but 890 ppb using the CLP method. There were instances where the TTF data was slightly higher than the CLP readings. However, the critical factor is detection of toxaphene levels for the purpose of cleaning the site. Generally, the CLP test levels were as much as 10 times higher than the TTF methods. In no cases did the CLP method miss toxaphene where the TTF test found toxaphene.

The third analysis method, or “congener” method, does not use a standard of toxaphene as the starting point in the analysis. Instead, a “control” mixture made up of about 20 different known toxaphene chemicals is used for calibration. This has the advantage of greater reproducibility between tests and possibly improving resolution. At this time the Glynn Environmental Coalition has not been provided with supporting documentation for the assay, so it is impossible to determine if the method is a valid way of detecting environmental toxaphene. 

Overall, the combination of poor and unreliable methodology, little ability to compare results by different labs and impractical controls results in no confidence that Region 4 can cleanup toxaphene and leave the site safe for humans and the environment.

Conclusions

Manipulation of standards and analysis methods appears to have created a technical “loophole” that could leave substantial toxaphene in the Brunswick environment. To guide a cleanup it is essential that a) all of the waste is found, and, b) analysis during the cleanup ensures all the waste is treated. Unless these simple criteria are met, there can be no assurance that the cleanup was successful.

Hercules/EPA’s handling of toxaphene data suggests they may under-represent the scope of the Terry Creek cleanup. It is essential to get a laboratory outside of the EPA-Hercules axis to perform the analysis and assess the Terry and Dupree creek data. 

 Written by R. Kevin Pegg, Ph.D.; edited by Dr. Mary S. Saunders. Copies of the newsletter are available from the GEC, at the Glynn County library, or at www.NucleicAssays.com/tags  on the Internet.

"This project has been funded wholly or partly by the U.S. Environmental Protection Agency under Assistance Agreement Number 1984482-98-0 to The Glynn Environmental Coalition, Inc. The contents of this document do not necessarily reflect the views and policies of the U.S. Environmental Protection agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use."