Sampling-Expansion Approach Enabling Efficient and Accurate Steady-State Solutions for Descriptor-Based Microkinetic Modeling

Abstract Microkinetic modeling (MKM) is a pivotal computational tool that bridges potential energy surfaces and catalytic activity, while descriptor-based MKM enables rational catalyst design through computational methods. However, the use of descriptors requires repeated MKM calculations across numerous conditions, increasing the overall cost by orders of magnitude and imposing stringent demands on solver stability. To address these challenges, we herein report a robust and efficient Sampling-Expansion Approach (SEA) for accelerated descriptor-based MKM, which is incorporated into our in-house developed multiscale kinetic modeling platform (MuKiP). SEA maximizes computational efficiency by strategically controlling the solving strategy, leveraging converged solutions as initial guesses, and employing a parallel computing framework. Compared with conventional approaches, SEA achieves significantly higher efficiency and reduces reliance on solver stability, thereby exhibiting strong robustness. The application of the proposed method and framework is expected to substantially accelerate the rational design of catalysts.