Co-firing of hydrogen and natural gas in a practical DLN combustor model

To reduce carbon dioxide emissions, the combustion of natural gas-hydrogen blends in a lean premix gas turbine combustor has been investigated. Previous studies have mostly investigated the fuel blends at relatively low pressure (up to 5 bar) with relatively low hydrogen concentrations (up to 50vol%) on lab-scale or generic burner configurations. However, the influence of higher pressure and higher hydrogen content (over 50vol%) has not been widely studied, particularly on a practical industry-scale lean premixed burner as presented in this study. Such an operation is more challenging as it increases the turbulent flame speed gradient, which is an important factor in determining the likelihood of boundary layer flashback. A preliminary RANS-based Computational Fluid Dynamics (CFD) study has been conducted using ANSYS Fluent 2021R1, employing the Realizable K-Epsilon turbulence model and the Flamelet-Generated Manifold (FGM) combustion model. The combustor consists of a diffusion pilot and premixed main fuel nozzles. Methane-hydrogen blends of up to 90vol% hydrogen were investigated at a fixed fuel energy input. 100vol% methane at 15 bar pressure was taken as the baseline reference case. To investigate the flame characteristics, contour plots of OH mass fraction, equivalence ratios and temperatures (at different planes) are presented.

This study shows that the expected reduction in the flame length with increasing hydrogen concentration occurs up to 40vol%. Significantly different flame shapes (as indicated by OH contours) were seen at higher hydrogen content. For this model, the flashback occurred at 90vol% H2 as indicated by a premature development of the flame within the nozzle of the main fuel burner. NOx emissions are shown to progressively rise with increasing hydrogen content up to 60vol% but reduce as the hydrogen content increases to 70vol%. The decrease appears to be related to an improvement in the quality of fuel-air mixing. It is important to note that the apparent rate of increase in NOx with increasing hydrogen is dependent on the reporting approach used. When reporting conventionally (parts per million by volume corrected to 15% O2 on a dry basis) the increase is significantly greater than when reporting on a mass per fuel energy input basis (gram per Joule). Reporting in the conventional manner disadvantages hydrogen because of the impact of oxygen consumption and water production on the corrections.