Self-modeling mixture analysis of Raman microspectrometric investigations of dust emitted by lead and zinc smelters

The application of SIMPLe-to-use Interactive Self-modeling Mixture Analysis (SIMPLISMA) is reported for analysis of huge Raman image data of dust samp les collected in the vicinity of lead and zinc smelters. The Raman data wer e carried out using Raman microprobe and automated mapping stage at 1 mum s tep size over at most 80 mum x 70 mum sample area for each experiment. Nume rous significant pure spectra were extracted by the SIMPLISMA method. Resul ts provide the number of major as well as minor chemical species. Concentra tion maps of the resolved components demonstrate the heterogeneity of indiv idual particles at level of 1 mum spatial resolution. A discussion is given addressing the analysis aspects of SIMPLISMA and the relative merits and d rawbacks of the technique in comparison with conventional analysis from two real Raman image data. (C) 2001 Elsevier Science B.V. All rights reserved.