Fundamental problems related to validation of analytical data elaborated on the example of TCDD†

Abstract

Abstract Problems of sample selection, contamination, and degradation, background noise, interferences, signal detection and measurement, identification, confirmation, and corroboration are highlighted in a review of efforts to determine 2, 3, 7, S‐tetrachloro‐dibenzo‐p‐dioxin (2, 3, 7, 8‐TCDD) at the limit of detection in a variety of matrices. Recently developed methods using lengthy and painstaking procedures can separate, isolate, detect, and measure this material without interference from its isomers at parts per trillion levels. This particular achievement is helpful in putting earlier work in perspective and illustrates the general nature of difficulties encountered when using modern analytical methods and generating data close to the measurable limits of the method. A need for simple criteria for assuring the soundness of data obtained near the limit of detection is made clear. These criteria are currently being developed by an ACS subcommittee on Environmental Analytical Chemistry.