Many of the flame speed target and validation computations for GRI-Mech 1.2
were performed with an erroneous sign for the thermal diffusion term.  The 1.2
predictions presented thus may have been too low in value by a few percent.