Appendix B
CHEMKIN Chemical Kinetics Software
To provide insight into the role of various parameters and components of idealized combustion systems, software tools have been developed to simulate these systems. For example, models for premixed laminar flames started to appear in the 1960s at about the same time that research was revealing insights about chemical kinetic reaction mechanisms. Much of the new understanding about kinetics at the time was a result of the give-and-take between laminar premixed flame experiments and models.
The initial laminar flame models were inefficient and difficult to use, until a group of researchers (experimental and theoretical chemists and applied mathematicians) at the Sandia National Laboratories in Livermore, California, developed a significantly better flame model. That group also developed models for other common, idealized zero-dimensional combustion problems, including the plug-flow reactor, the jet-stirred reactor, the constant-volume reactor, and the rapid-compression machine. The group’s full software package was entitled “CHEMKIN—A Chemical Kinetics Software Package.”
The CHEMKIN software package provided a very convenient interface for specifying chemical kinetic reaction mechanisms, thermochemical data, and transport parameters, together with the computer software necessary to evaluate the various properties needed in the governing equations. Each of the individual system models was as good as, or significantly better than, any comparable models available at the time, and the authors made the entire package available to the general combustion community at no cost. In addition to the broad functionality of the
software, distribution of the source code for CHEMKIN made it easy for others to modify the codes to suit any special needs of individual groups, greatly accelerating the growth in combustion-simulation capabilities in new directions.
This attractive combination of significant functionality, convenience, and zero cost rapidly made CHEMKIN the software tool of choice throughout the technical combustion world, and it rapidly became effectively an industry standard. This result had far-reaching and somewhat unexpected effects. It was no longer necessary for each research group to develop its own computational tools, especially the highly specialized models required for the challenging, stiff differential equations characteristic of chemical kinetic systems. It became easy for researchers anywhere to reproduce computational results carried out elsewhere. In addition, CHEMKIN enabled researchers with virtually no modeling expertise to build a computational component into their projects almost overnight. Most importantly, an individual researcher could go to another research organization anywhere in the world and quickly become completely functional because the new organization was using exactly the same modeling tools. Model improvements and new capabilities within the CHEMKIN family of models simultaneously provided the entire research community with the same new tools, with no cost or development expenses to anyone outside the core development group.
After 16 years, with minimal specified funding for its development, the CHEMKIN project at the Sandia National Laboratories ended, and the code itself was taken over by Reaction Design, a commercial company that has continued to improve the functionality and computational solvers in the CHEMKIN package (Reaction Design, 2009). However, Reaction Design started selling licenses for the software at costs that began to discourage some groups that were using these models. In addition, the codes were no longer open-source codes, so their flexibility in facilitating modifications was lost. As a result, although the CHEMKIN codes are still widely used, some users have changed to different modeling tools, while others continue to use the last free versions, which are now more than 10 years old.
The CHEMKIN history represents a case study of the enormous and far-reaching benefits made possible by a combustion cyberinfrastructure. There is enormous value in high-quality, practical, easily used software tools that simulate most of the common problems of importance to most research projects and that are constantly being refined and improved. In many cases, graduate students would learn how to use these computational tools during their master’s and doctoral research and then carry that new expertise with them into their jobs, and current professionals could change jobs and continue to use the same software tools with-
out further training. The same history also illustrates that the easy, free availability of the software is an essential piece of any cyberinfrastructure—the CHEMKIN codes are no longer an industry standard owing to increased cost, even though they have actually been improved following their privatization.
REFERENCE
Reaction Design. 2009. Chemkin MFC-3.5. San Diego, Calif.