Appendix A
The GRIMech Model
Approximately 25 years ago, a project supported by the Gas Research Institute (GRI) in Chicago led to a common kinetic model for the combustion of natural gas, which is dominated by methane and has small levels of other gaseous components such as ethane, propane, CO, and H2. A team of about a dozen combustion-chemistry experts collected extensive libraries of experimental and kinetic modeling papers and models of natural gas combustion. This team met several times a year, exchanged recommendations and opinions about which experiments were most reliable, carried out theoretical studies of important reactions and chemical species, and developed a fully detailed kinetic model—one that has been used for the past 20 years by researchers around the world. All of the information was assembled in a series of computer files, in a common format, with detailed descriptions of the data and methods used to optimize the resulting models, with significant curation of the entire system of data and evaluations.
The resulting model was called GRIMech (Frenklach, 2007), and over a period of years, several updates were released to the public. The benefits of the model included the consensus evaluations and recommendations of the panel of experts, the incorporation of modern data-curation and -evaluation processes in the model’s development, and the broad range of applicability of the resulting model. Its availability for no cost was also an obvious factor in its wide acceptance and common use.
The degree of acceptance of this process led to the widespread use of GRIMech. It was particularly valuable for researchers who knew that their work required a reliable kinetic mechanism for methane or natural gas combustion chemistry, but who were not personally experienced or knowledgeable about combustion kinetics. GRIMech gave these people a tool that was ready to use, had been thoroughly tested by combustion-chemistry experts, and was freely available in a common and convenient format.
Although details of GRIMech are no longer considered to be leading-edge kinetic expertise, the model continues to be used extensively, for exactly the same reasons that made it successful 20 years ago—namely, easy and free access; thorough testing, evaluation, and validation; and common acceptance. This mechanism provided a combustion simulation tool of significant value that unified the international combustion community and accelerated combustion progress for 20 years. It is interesting to note that GRIMech shared most of these attractive features with the CHEMKIN software, described in Appendix B of this report. For more than 10 years, a large fraction of the combustion community used one or the other or both of these research tools in their daily efforts, and the community prospered and made significant advances in all types of combustion research, experimental and theoretical, as well as in computer modeling. These were two essential parts of an effective combustion cyberinfrastructure, and they form a template for a possible new cyberinfrastructure based on more modern software tools and chemistry models for practical petroleum-based fuels and biofuels for future combustion systems.
The GRIMech panel of experts envisioned extending this approach to larger and more complex hydrocarbons and other fuels, but the disappearance of the GRI as a research funding agency and the lack of other continuing support commitments made further extensions impossible. A combustion cyberinfrastructure would make these extensions possible. If the GRIMech history is any predictor, such a set of tools, enabled by a combustion cyberinfrastructure, would again have the same type of generally unifying results. It would also significantly accelerate the pace of progress in combustion research and lead to greatly improved combustion systems.
REFERENCE
Frenklach, M. 2007. “Transforming Data into Knowledge—Process Informatics for Combustion Chemistry.” Proceedings of the Combustion Institute, Vol. 31, pp. 125-140.
