A wide-band based weighted-sum-of-gray-gases model for participating media: Application to H2O-CO2 mixtures with or without soot

The presence of participating gases and soot in combustion processes often makes the thermal radiation the main mechanism of heat transfer due to the elevated temperatures involved. The accurate solution of the radiative transfer equation (RTE) through the line-by-line (LBL) integration in engineering applications is in general still impracticable due to the high computational effort required to account for the millions of spectral lines that characterize the absorption coefficient of a gas. Global spectral models, such as the weighted-sum-of-gray-gases (WSGG) model, are attempts to replace the LBL integration with low computational cost and satisfactory accuracy. In the present study, an improvement of the WSGG model through a partition of the spectrum is proposed. The spectrum is divided into a set of ranges in which the standard methodology of the WSGG model is applied to determine the gray gases coefficients in these bands. This is the central aspect of what is named here a wide-band based WSGG (WBW) model, which consists firstly of determining the emittance of each band from the HITEMP 2010 database, then obtaining the pressure-absorption coefficients and the temperature-dependent coefficients. The total radiative heat flux and radiative heat source are obtained by the summation of the contribution of each band individually. For soot, a gray gas approach is considered for each band via Planck-mean absorption coefficients. A set of 1D problems consisting of two infinite parallel black walls, filled with a non-isothermal and non-homogeneous gaseous mixture of water vapor, carbon dioxide and soot, is solved. The results show that, especially in cases where the thermal radiation of soot is dominant, the maximum deviations obtained with the proposed method in relation to the LBL solution reduce by almost three times compared to other global WSGG models in the literature, in addition to decreasing by 40% the number of equations that need to be solved.