ChE Assistant Professor Richard West was awarded a $140K NSF Grant to “Identify and resolve discrepancies in kinetic models of hydrocarbon combustion“.
This material is based upon work supported by the National Science Foundation under Grant No. 1403171.
Computational kinetic modeling of combustion chemistry has made significant progress in recent decades. Dozens of recent models, which describe tens of thousands of simultaneous reactions between thousands of intermediate species, are capable of explaining complicated combustion phenomena, allowing increasingly accurate engine simulations, and screening novel biofuels. However, these ever-proliferating detailed kinetic models are incompatible and inconsistent, are seldom compared directly, and often contain undetected mistakes.
The commonly used format to publish these models, devised in the 1970’s when input was limited by the width of 80-column punch-cards, forces model-builders to abbreviate species’ names, thereby losing their chemical identity, and to discard other metadata. The main challenge in comparing these models is in recognizing, for example, 
that the name “C3KET21” in one model represents 1-hydroperoxypropan-2-one, which another research group may have named “CH3COCH2O2H” in a different model.
This project develops tools to help identify the chemical species in a kinetic model, to facilitate comparison of models. The new tools will be built upon the open-source Reaction Mechanism Generator (RMG-Py) software, that we have been developing in a collaboration between Northeastern and the Massachussets Institute of Technology. A web-based user interface will make it easy for users to import models to the database, and provide instant reward for doing so (the ability to check the model, fill in gaps, and merge with other models). The proposed work will massively reduce the barriers to converting detailed kinetic models into a machine-readable format with clear and consistent species definitions. This will enable: (1) comparison between models, leading to more replicable science, (2) error identification, leading to higher quality data, (3) better rate estimates and a broader impact for mechanism generation software, and (4) wider use of process informatics tools. The unified database of all previous kinetic models will greatly assist progress in combustion modeling.