A Shortcut Modeling Platform for Optimizing Catalyst Distribution of the CO2-to-Methanol Process

Abstract

Methanol synthesis via CO2 hydrogenation is increasingly recognized as a pivotal strategy in CO2 utilization, addressing both environmental concerns and the demand for sustainable chemical production. However, the exothermic nature of methanol synthesis often leads to thermal management issues in reactors, particularly near the inlet. This study proposes an optimized three-stage reactor system with a distinctive approach to catalyst distribution. Each reactor is divided into two zones, culminating in six zones with varied catalyst loadings, specifically designed to alleviate hot spots. The advanced modeling platform developed in this study significantly reduces the computational intensity typically required in such processes. This platform integrates a process simulator with computational fluid dynamics (CFD) modeling, allowing for efficient computation of reaction rates and heat release as well as detailed temperature profiles in both axial and radial directions. Using an efficient Gaussian process Bayesian optimization (GPBO) algorithm, we optimized the catalyst loading with the minimum number of iterations. Our findings indicate a significant decrease in temperature variances in reactors of various sizes: 67.5% in a 2-inch tube reactor, 55.4% in a 1-inch reactor, and 67.3% in a 3/8-inch reactor, when compared to conventional methanol reactor. The maximum temperature differences were notably reduced by up to 15.7 K in the axial direction for 3/8-inch tubes and up to 2.5 K in the radial direction for 2-inch tubes. Additionally, the study provides insights into the promotional effects of CO. This platform not only optimizes reactor performance but also provides a substantial advancement in the computational modeling of methanol synthesis from CO2.