Part I. Analysis of aromatic amines in banned azo dyes by SPME and GC/MS/MS. Part II. Analysis of residual ignitable fluids in fire debris by GC/MS/MS and development of an expert system.

摘要

The first part of the dissertation focused on analysis of target aromatic amines originating from banned azo dyes in textiles, leather products, and dyes. Four analytical approaches were applied: GC/MS and GC/MS/MS of the organic extract, SPME-GC/MS on the aqueous solution after reductive cleavage, and SPME-GC/MS/MS of the aqueous solution. GC conditions, MS/MS conditions, and SPME conditions were optimized. The effects of thermal stress and UV-radiation on the stability of the 20 target aromatic amines were also investigated. Calibration and quantitation parameters were established. Target compound analysis of spiked extract demonstrated very high selectivity for the GC/MS/MS method. The newly developed SPME-based methods provided simpler chromatograms than conventional GC/MS analysis of the organic extract. The SPME techniques did not require sample clean-up and preconcentration steps. Sample preparation time for the SPME-based methods was reduced to 10 minutes, compared to one to two hours for the conventional method.; The second part of the dissertation dealt with analysis of ignitable fluids in fire debris. The first task involved the investigation of the feasibility of increasing target compound specificity by adding another stage of mass spectrometry, performed on an ion trap GC/MS/MS instrument. The second task required the development of an expert system for assisting forensic scientists in making a decision regarding the presence and the class of potential accelerant in fire debris samples. The MS/MS conditions under which meaningful responses can be obtained for different classes of accelerants varied widely. Aromatic compounds produced intense molecular ions and prominent daughter ions. They can be monitored with excellent selectivity and sensitivity. Alkanes, on the other hand, do not generate adequate molecular ions and fragment extensively. The combination of both MS/MS and SIS permits the detection of both classes of compounds with high selectivity and sensitivity. The optimized conditions were successfully applied for the detection of ignitable fluids in simulated arson debris samples. An expert system for pattern recognition of ignitable fluids in fire debris samples has been successfully developed on the Varian Saturn GC/MS using Procedure Language.