Automating the Optimization of Catalytic Reaction Mechanism Parameters Using Basin-Hopping: A Proof of Concept

Abstract

Parameter estimation is a crucial step for successful microkinetic modeling in catalysis. However, the large number of parameters to be optimized in order to match the experimental data is a bottleneck. In this regard, the global optimization algorithm Basin-Hopping is utilized to automate the typically time-extensive and error-prone task of manual fitting of kinetic parameters for a heterogeneous catalytic system. The stochastic approach of the Basin-Hopping algorithm to explore the kinetic parameter solution space coupled with local search methods makes it possible to screen the high-dimensional space for an optimal set of kinetic parameters giving the least residual between the simulated and the experimentally measured catalytic performance data. Our approach also ensures that only thermodynamically consistent solution candidates are explored at each optimization step. We utilize two example case studies in heterogeneous catalysis, namely, methane oxidation over a palladium catalyst and carbon monoxide methanation over a nickel catalyst, with corresponding detailed kinetic models to illustrate the applicability of the algorithm to efficiently fine-tune detailed kinetic models.

Interconnected Digital Solutions to Accelerate Modeling of the Reaction Kinetics in Catalysis

“…[58,59] Our previous paper showcases a proof-ofconcept for the same using the global optimization algorithm Basin-Hopping [60] to optimize high-dimensional microkinetic models for methanation of CO over Ni and methane oxidation over Pd catalysts. [61] Basin Hopping is a stochastic algorithm that searches for a potential global minimum by hopping within the parameter search space as determined by a user-defined step-size value. Once it has made a hop, it attempts to find a local minimum by leveraging a local-search technique such as the Nelder-Mead [62] method.…”

“…Gradient‐free global optimization algorithms offer a generalized methodology for optimizing these microkinetic parameters irrespective of the reactor configuration [58,59] . Our previous paper showcases a proof‐of‐concept for the same using the global optimization algorithm Basin‐Hopping [60] to optimize high‐dimensional microkinetic models for methanation of CO over Ni and methane oxidation over Pd catalysts [61] . Basin Hopping is a stochastic algorithm that searches for a potential global minimum by hopping within the parameter search space as determined by a user‐defined step‐size value.…”

Digitalization in Catalysis and Reaction Engineering: Automatizing Work Flows

“…Exemplarily, the Figure illustrates the communication between software tools that manage infor-CoRDI2023-25 mation on catalyst synthesis, characterization, and performance as well as drivers for numerical simulation of catalytic reactors using these catalytic materials. Recently developed optimization tools to speed up model development and scale-up will soon extend this picture [5,6]. The presentation is concluded by an overview of the work in the consortium NFDI4Cat, where digital tools, workflows and service offerings are developed in a community driven effort for catalysis related sciences including chemical engineering and process design.…”

Simple Rate Expression for Catalyzed Ammonia Decomposition for Fuel Cells

“…The lumped parameters in the simplified rate expression (Equation (6)) will be calibrated using rate data obtained from Detchem ® CSTR ® simulations run with the detailed NH 3 decomposition mechanism [ 8 ] shown above. Available online [ 13 ] with numerous applications, the Detchem ® software package is widely used [ 16 , 17 ] and quite versatile. This package executes detailed material, energy, and momentum balances using a user-supplied elementary reaction mechanism and required reactor input and parameter information.…”