Table 11 lists the conditions and summarizes the results of the validation studies. For more detail, see Sections 3.1.3 through 3.1.9. The classifications listed under ``Comments'' are subjective judgements to be used as a rough guide only for the specific mixture tested. Reference should be made to the orginal citation for a mechanism to determine if it has been tested for other situations.
The most useful mechanisms were the GRI Mech-2.11 [Frenklach et al. (1995)] and the modified form of [Miller and Bowman (1989)]. The GRI Mech-2.11 was the most accurate with reactions involving methane, as can be expected since it was designed to model natural gas combustion. However, it could not be used for ammonia combustion. The original form of the [Miller and Bowman (1989)] mechanism (actually a combination of separate hydrocarbon and ammonia mechanisms) was nearly as accurate for methane combustion but could also nominally be used with ammonia. Unfortunately it was particularly inaccurate with most ammonia tests. The earlier mechanism of [Miller et al. (1983)] proved better with ammonia but did not incorporate hydrocarbons. Finally, some reactions from [Fujii et al. (1981)], which were originally tested against shock tube data, were added to the [Miller and Bowman (1989)] mechanism to produce an overall reasonably accurate mechanism.