The need for speed - optimal CO<inf>2</inf> hydrogenation processes selection via mixed integer linear programming

Abstract

The race towards mitigating carbon emissions led to the development of countless processes, product chemicals, and catalysts for carbon capture, storage, and utilization. The abundance of data and potential pathways for decarbonization thus require the development of models that could easily adapt to the increasing boundaries of evaluation. The model presented herein offers a rapid evaluation tool to find the optimal choice(s) amongst reported thermocatalytic CO2 hydrogenation processes. Using mixed-integer linear programming (MILP), the model sizes and costs 4 major types of chemical units (compressor, heat exchanger, reactor, and distillation column) in a network optimization of 29 literature processes. The model results show excellent agreement with results of its nonlinear program (NLP) equivalent (max 2% objective function deviations), with a two order of magnitude reduction in CPU solving time. The findings highlight the potential to enhance decision making via the evaluation of myriad carbon utilization processes at low computational times.