DETERMINATION OF METHOD UNCERTAINTY A PRACTICAL EXAMPLE FOR THE ANALYSIS OF TRIHALOMETHANES IN DRINKING WATER USING PURGETRAP GC-MS

摘要

Johannesburg water is responsible for the supply and monitoring of drinking water supplied to the greater Johannesburg area. Part of this monitoring programme involves the determination of Trihalomethane (THM) levels and the comparison thereof against the limits stipulated in SANS 241. This paper demonstrates the approach used to determine the uncertainty of the analytical method associated with the determination of THM's, at ug/L levels, in drinking water using Purge and Trap (P&T) GC-MS. It indicates the five steps used namely: 1. Specification of the method; 2. Identifying the model; 3. Identification of sources of uncertainty using the Ishikawa approach; 4. Expanding and if needed combining the uncertainties of each of the chosen data sources; 5. Combining the expanded uncertainties and expressing the final method uncertainty. The paper also attempts an evaluation of the data obtained and recommends possible improvements, identified from the data obtained, which could be implemented to improve the uncertainty of the analytical method.