Reaction models from reactive molecular dynamics and high-level kinetics predictions

The design and optimization of complex chemical processes is a key challenge in chemical engineering and requires knowledge of the underlying kinetic model. This information can be obtained from experiments by inverting the reaction mechanism, which needs to be known therefore. Solving this inverse problem, however, is mathematically challenging, if not impossible, and the reaction mechanism is mostly unknown for novel compounds. Both of these challenges are adressed in the present thesis by proposing a novel methodology for forward reaction model development, which is based on exploring chemical space without the need for prior knowledge.