Identification of sustainable carbon capture and utilization pathways using superstructure approach

Abstract

Carbon capture and utilization (CCU) is a realistic solution to mitigate greenhouse gas emission (GHG). Despite the technological advances in decades, many of the CCU processes are not certain to guarantee truly carbon reducing and economic benefits as they consume a significant amount of energy and additional raw materials. The sustainability of the CCU processes are affected by not only the capture and utilization efficiency but also the sources of the energy and raw materials. To identify the sustainable CCU pathway among the multiple candidates, a well-constructed framework is required which contains reliable life cycle inventory (LCI) database, incorporation of CCU technologies with their technological features, and methodological framework to evaluate the best pathways. A superstructure approach has been widely adapted to identify the promising CCU processes

however, no work has been conducted which encompasses all these issues enumerated above. Throughout this dissertation, methodologies for identification of truly sustainable CCU pathways using superstructure approach are studied. In the first study, a computer-aided tool, ArKaTAC3 Ver. 3, is developed. This new version succeeds a previously published version but is distinguished from the old one as the new tool contains LCI database, parameterization of CCU processes by generic linear model, pathway identification function by mathematical programming, and provides graphic user interface that serves as intuitive visualization of superstructure. Two case studies are performed using this tool: (1) identification of potentially sustainable carbon utilization pathways under various scenarios with uncertainties, and (2) optimization of CCU supply chain network in Middle East. In the second study, surrogate modeling for CCU processes are introduced. The purpose of the surrogate modeling is to construct alternative models capable to be incorporated into a CCU superstructure so that the true sustainable processes can be identified considering first principles of the chemical processes as nonlinear surrogate equations instead of rigorous models. The superstructure optimization problem with surrogate models can be formulated as mathematical programming provided that the surrogate models composed of simple mathematical expressions. In this dissertation, how the surrogate models can be constructed are discussed and the two case studies are conducted as an illustration of the methodology: (1) amine scrubbing CO2 capture processes and (2) vacuum swing adsorption process for CO2 capture. Finally, a noble framework is proposed that incorporates the nonlinear surrogate models into a CCU superstructure to identify sustainable CCU pathways. The framework adapts state-task network (STN) representation of CCU processes for flexible representation of all types of chemical processes with nonlinearities. In addition, a superstructure network is modified so that selection of materials from different sources (e.g. gray, blue, and green hydrogen) and products into different markets can be considered. To solve the nonconvex mixed-integer programming (NC-MIP), logic-based outer approximation (LOA) is introduced. A case study demonstrates that the noble framework can identify the sustainable pathways.