On the accuracy of micro Winkler titration procedures: a case study

Accuracy data (expressed as precision and trueness) presented by the authors of three different micro modifications of the Winkler titration procedure for dissolved oxygen concentration determination are critically evaluated. Tentative uncertainty estimates are extracted from the data based on the single-laboratory validation approach (originally published in the Nordtest Technical Report 537) and they lead to expanded uncertainty (k = 2) estimates in the range from 0.13 to 0.27 mg l(-1) for the three procedures. It is demonstrated that, in all cases, the authors have presented the accuracy and/or precision estimates of the procedures in a way that can lead to too optimistic conclusions about the uncertainty of their procedures. This case study demonstrates the usefulness and flexibility of the single-laboratory validation approach to uncertainty estimation, even in the case of insufficient data, and can be of interest to laboratory workers dealing with measurement procedures from the literature. It is also expected to be of interest to university instructors of analytical chemistry and metrology in chemistry as a real-life example of the critical evaluation of the literature data.