2024 AIChE Annual Meeting

(306b) Surrogate Optimization of Liquid Organic Hydrogen Carrier Processes

Authors

Kitchin, J., Carnegie Mellon University
Akhade, S., Lawrence Livermore National Laboratory
Liquid organic hydrogen carriers (LOHCs) have been proposed to economically distribute hydrogen for use as fuel. This concept requires two catalytic processes to hydrogenate and dehydrogenate the LOHC. To make such processes economically feasible, it is necessary to optimize over many kinds of variables, including those related to the catalyst material, LOHC transportation method, and process design. This is further complicated by variables which pertain to multiple aspects of the problem. For instance, reactor operating conditions affect the catalyst activity at the nanoscale and the operating cost at the macroscale. We introduce an approach to simultaneously optimize the process design and catalyst material for these processes. In particular, we employ surrogate models for 1) a microkinetic model of catalytic properties, 2) a reactor process model, and 3) a costing model based on economic analysis. The surrogate models are directly embedded into rigorous global optimization methods. By co-optimizing the catalyst and process, we demonstrate how to systematically select an economically viable design from a range of LOHCs, catalysts, catalytic reactor designs, and other key variables. We demonstrate a case study where carbon dioxide is used as the LOHC (in hydrogenated form, formic acid). By co-optimizing these models, we achieve significant process cost savings compared to optimizing at each scale independently.