Multi-Scale Modeling for Ionic Liquid Enabled Separation Systems

Humanity's energy consumption is enormous, accompanied by high costs and CO2 emissions. Separation processes account for 45-55% of industrial energy consumption in the United States. There exists ample opportunity for reducing and replacing high-energy separation processes with innovative separation agents. This thesis explores methods to aid the search for novel materials integrating data science and process systems engineering tools to find property targets for separation agents and contemplate which data is more valuable to screen these separation agents. Two frameworks are proposed in this thesis:

1. Molecular design targets and optimization framework of low-temperature thermal desalination systems. The framework focuses on finding molecular design targets for directional solvents for water desalination.
2. Assessing what data are most valuable to screen ionic liquid entrainers for extractive distillation. The framework focuses on screening ILs as entrainers for separating high Global Warming Potential (GWP) mixtures of hydrofluorocarbons (HFCs).

Multi-scale optimization frameworks are proposed to facilitate and accelerate solvent discovery. Using molecular-systems engineering, the proposed frameworks aim to identify properties that significantly influence the cost of the separation process. Once the impactful properties are identified, the framework can be used to study which data is more valuable for multi-scale modeling and what accuracy in the data is necessary for entrainer screening. Finally, an added value metric is included to understand the economics of a separation process.