Flame sampling for mass spectrometry

Abstract

Mass spectrometry is one of the most popular and fruitful methods available for the study of flames [1, 2]. This is a result of its versatility and the virtue of providing a uniform possibility of quantitative study for all molecular species, since they all can be ionized. The practice is, of course, much more complex than is implied by this naive point of view, but both stable and short-lived species can be studied. When properly applied, it is a powerful, rapid quantitative analytical technique. The applicability and limitations will be discussed with respect to the problems associated with delivering a representative sample into the ionization region of the mass spectrometer. The practical problems of analyzing complex mixtures will not be discussed in detail. The central problem in sampling flames or any system is to obtain a representative sample and to interpret this sample either qualitatively or quantitatively in terms of the desired information. The withdrawal of sample should either produce a quantitatively negligible disturbance of the system or one which can be accounted for quantitatively. In flame studies one is interested in qualitative information about which species are present and also in complete quantitative analysis of the system. Further, since flames are systems having strong concentration gradients, one is often interested in associating the analysis with a spatial position. Thus, the usual outcome of a flame study is not a simple analysis, but a profile (see Fig. 1) [1, 2]. Flames are very strongly coupled systems, and one of the major problems is the disturbance which the sampling processes introduce. This is a quantitative question, and we will attempt to outline the problems, their sources and magnitudes, and the areas of application. The paper will begin with a discussion of sampling techniques, followed by specific examples of flame systems which have been studied.