The importance of reaction energy in predicting chemical reaction barriers with machine learning models

Improving our fundamental understanding of complex heterocatalytic processes increasingly relies on electronic structure simulations and microkinetic models based on calculated energy differences. In particular, calculation of activation barriers, usually achieved through compute-intensive saddle point search routines, remains a serious bottleneck in understanding trends in catalytic activity for highly branched reaction networks. Although the well-known Brønsted-Evans-Polyani (BEP) scaling – a one- dimensional linear regression model – has been widely applied in such microkinetic models, they still rely on calculated reaction energies and may not generalize beyond a single facet on a single class of materials, e.g., a terrace sites on transition metals. For highly branched and shallow reaction networks, such as electrochemical CO2 reduction or waste remediation, calculating even reaction energies on many surfaces can become computationally intractable due to the combinatorial explosion of states that must be considered. Here, we report a new database of over 500 dehydrogenation activation barriers calculated on a range of molecules and transition metal surfaces using a one- dimensional potential energy surface scan, benchmarked to a more rigorous transition state search routine. We investigate the importance of the reaction energy when predicting activation barriers using linear and nonlinear machine learning (ML) models and find that inclusion of the reaction energy significantly improves both classes of ML models. By excluding the reaction energy as a feature, even our most complex nonlinear model fails to significantly outperform the simplest BEP scaling when predicting activation barriers on new systems. Our results suggest that the reaction energy is a critical feature to consider when building models to predict activation barriers, indicating that efforts to reliably predict reaction energies reliably through, e.g., the Open Catalyst Project, will be an important route to effective model development for more complex systems.