Formatted Title:

Enhancing the Precision and Accuracy of EPA Method 8082 for PCBs

Background/Objectives:

Polychlorinated biphenyls (PCBs) have been analyzed by EPA Method 8082 since the mid-1980s when the method was first developed and it is still widely used today. To understand changes in PCB concentrations in the environment, precise and accurate measurements are needed over time. The Hudson River PCBs Superfund site has a robust laboratory quality assurance/quality control (QA/QC) program to provide data that are precise and comparable through time. In addition to standard QA/QC, the program includes the use of National Institute of Standards and Technology (NIST) Standard Reference Materials (SRMs), analyzed both by Method 8082 and Method 1668. Use of the NIST SRMs and the QA/QC data together provided an opportunity to improve on the precision and accuracy of Method 8082 data based on surrogate recovery.

Approach/Activities:

Our approach involved examining whether the recovery of Method 8082 TCMX surrogate could be used to explain observed differences in: (1) NIST results with known concentrations, (2) paired analysis using Method 8082 and Method 1668, and (3) laboratory control samples. Datasets evaluated included both sediment and fish over multiple years of analysis.

Results/Lessons Learned:

Our analysis revealed a strong positive correlation between surrogate recovery and Method 8082 quantification, explaining up to 85% of the variation in some cases. In particular, analyses of NIST standards and of laboratory control standards, where sample concentrations were known, provided the strongest evidence for the correlation. During routine application of Method 8082, the allowable range of TCMX recovery is 60 to 140%. Based on the strong observed correlation, this wide range of allowable recovery translates nearly directly to variability in the Method 8082 PCB result. Further examination of the data showed that adjustment of the Method 8082 results to account for TCMX recovery significantly reduced variance and improved agreement (i.e., accuracy) with Method 1668 results over the entire study period. These observations led to two important recommendations for the use of Method 8082. 1) Where surrogate data are available, consideration should be given to linearly adjusting Method 8082 results based on TCMX surrogate recovery. This can be done for both current and historical data and should reduce year-to-year variance due to laboratory differences. This approach is similar to the corrections developed for standard surrogate recovery methods such as Method 1668. The efficacy of this correction for a given data set can be demonstrated by examining the laboratory control standard results that are typically run with each sample delivery group. 2) Consideration should be given to improving surrogate recovery wherever possible, narrowing the range of acceptable laboratory performance.

Edward Garvey

Affiliation

Refine Results

Enhancing the Precision and Accuracy of EPA Method 8082 for PCBs