SITE INFORMATION AND CORRESPONDENCE

Home Browse Search

LLISTLine Bulletin 43 • Mach 2003 ■MTBE Acid Hydrolysis To very briefly summarize their a*any potential limitations to that from page 19 results, using the standards of U.S. data.That information should serve EPA's performance evaluation proto- you well,and your corrective action col criteria, all three labs performed decision-making should improve. A Multi-Lab Performance within acceptable limits of accuracy However,at most sites you prob- Study Validation of Methods and precision for all four compounds ably do not need to use a heated P&T 5030/8260 for the entire range of concentrations extraction to generate good data for As mentioned earlier,the recent arti- studies (i.e., 1 to 3,000µg/L).MTBE MTBE and TBA.If heating is shown ides and TBA recoveries were almost to improve method performance, ves Evans and Colsman an excellent summary of pro- always within the acceptable range then temperatures of 40°to 45°C are study that addresses many of thea v (i.e., between 75 and 125 percent likely to be sufficient, and at these issues raised about the data quality recovery). In all cases, the desired temperatures, MTBE hydrolysis provided by P&T/GC-MS analysis. low ppb detection limits were should not compromise data quality. The authors note that the U.S. EPA reached. Purge-and-trap extraction of water Office of Solid Waste is a strong The authors conclude that the samples at ambient or slightly ele- advocate of performance-based ana- results reflect a successful demon- vated temperatures (combined with lytical methodology, which is a stration of method applicability good calibration procedures)should requirement for several RCRA pro- (P&T/GC-MS) for the compounds adequately recover any TBA present grams. Briefly defined, "the perfor- studied. This single study of three in samples. mance-based approach requires that labs does not, of course, mean that Furthermore, it is probably methods be selected and approved each and every one of the almost unnecessary in most situations to use based upon their ability to meet the 1,000 environmental labs nationwide TSP as your field preservative, data quality goals of a given project routinely performs as well.However, instead of acid.While both low acid- in the actual matrix to be sampled." the authors conclude, "The re- ity (pH) and high temperature are (Readers looking for more informa- sults...appear to be a successful known to enhance the rates of MTBE tion this topic are referred to EPA's demonstration...that...purge-and-trap hydrolysis,an evaluation of these fac- site at:httpd/www.epa.gov/epaoswer/ GC-MS is an acceptable technology tors and a review of the standard hazwaste/test/pbms.hbn. for the analysis of MTBE and its oxy- analytical procedures employed at The authors compared the per- genated breakdown products." many commercial laboratories indi- formance of three different California There is a lot of other good infor- cate that it is unlikely that significant commercial labs for MTBE and TBA mation and discussion in that well- MTBE hydrolysis occurs.For condi- that use routine P&T/GC-MS analy- written article, and I strongly tions used at most UST release sites— sis. They wanted to make sure that encourage you to take a closer look using standard acid preservation of the data they were collecting for their for yourself. It provides a good tem- samples, followed by P&T/GC-MS site (the world-famous Port Huen- plate for the kind of information you extraction and analysis—hydrolysis eme)would be adequate to quantify should be able to obtain from your of MTBE to TBA will be very limited MTBE at 5µg/L and TBA at 12µg/L laboratory regarding the quality of its and will not significantly influence (2-propanol and acetone were two analytical performance. data quality.Hydrolysis of MTBE to other target analytes). TBA should not be a concern during The authors do not indicate that With Good Data and storage and analysis of acid-pre- they had concerns regarding possible Attention to Protocol... served samples. hydrolysis of MTBE to TBA,but they We know a lot more today than we Data users should have a good do note that some critics question the did a few years back about the proper understanding of the basic perfor- purging efficiency of these methods approaches for oxygenate analysis. mance of the sampling and analytical for nonpolar compounds like MTBE We also know a lot about MTBE acid procedures they use to generate oxy- and especially TBA. hydrolysis and its potential effects on genate data. If they have concerns They submitted a variety of Port groundwater analysis,but that's not about data quality for ongoing evalu- Hueneme groundwater samples to to say we know it all. I suspect that ations of corrective action at UST the three labs(concentrations ranged this topic will continue to be of high sites where gasoline oxygenates are from 1 µg/L to about 3,000 µg/L in interest and that we will learn a lot present in groundwater,they should the samples submitted to each lab) more in the coming year. review the protocols in use,the list of and asked them to run their standard The studies cited in this article target oxygenate analytes, and their method. Two labs used EPA 5030 / and in LL#42 collectively provide a performance criteria(e.g.,recoveries 8260 and one used EPA 5242, a simi- lot of interesting information, but and detection limits)to determine if lar P&T/GC-MS method. All three more information will be required to any changes might improve perfor- labs used a purge cycle of 11 minutes. conclusively document the perfor- mance.I would also recommend that Two labs purged at ambient tempera- mance of the analytical methods cur- you talk about these issues with some ture,and the other used 40°C.A vari- rently in common use. To be sure, of the commercial labs in your area to ety of quality assurance and quality following the recommended protocol determine what they think are the control procedures were also provided in LL#42 would help you best approaches to maintaining or employed to confirm method perfor- become more aware of the quality of improving data quality.■ mance. the data for the sites you manage, 20