OCR Text |
Show 2 Introduction Although the number of optical techniques used to probe reactive gas flows has proliferated in the past few years, few have been suitable for in-situ measurements in industrial combustion environments. A notable exception is coherent anti-Stokes Raman spectroscopy (CARS). Much recent work in the development of experimental procedure and modeling algorithms has established CARS as an alternative to more conventional diagnostic techniques. Moreover, CARS has proved viable in environments that preclude the use of standard methods. Applications of CARS to practical combustion systems are now numerous.^"6 A number of reviews on CARS is available.7~q Its application to a variety of combustion problems is summarized in a recent work by Hall and Eckbreth.10 Since CARS is currently the only technique that can make spatially and temporally precise, non-perturbing measurements in hot luminous media, it is the sole method available to profile the temperature in an MHD gas flow. Since the efficiency in the MHD process is a strong function of the gas conductivity and the conductivity, in turn, is a strong function of the peak temperature, the performance of an MHD generator is more accurately modeled if profiles are available. This obviates the use of line-of-sight optical techniques like sodium line reversal. This paper describes multiplex CARS temperature measurements in a flow that simulates the post-magnet gas stream of a coal-fired MHD generator. In this environment, a potassium seed is injected into a gas stream laden with coal fly ash particles. The medium, characterized by unstable and turbulent flow at temperatures near 2500 K, is one of the harshest environments in which CARS measurements have been attempted to date. It is also one of the first applications of CARS to coal combustion. Results are presented as illustrations of the temporal and spatial resolution possible in an extremely unstable flow when critical optical alignment is maintained over very long path lengths. Problems created by the presence of the fly ash are discussed. Temperatures are obtained by fitting a library of precalculated CARS spectra to the data. The comparison spectra were calculated using the recently introduced formalism of Kataoka et al.11 and Teets.12 The temperature error introduced by uncertainties in several of the experimentally measured parameters used in the model spectra is examined. |