Bioprocess product streams from bioethanol manufacture or other biomass conversions typically include 'sticky' polar molecules such as water, alcohols, amines, and organic acids that associate by hydrogen bonding. Today, process design efforts for such material streams require many laborious empirical measurements and expensive pilot plant studies. Accurate and adaptable predictive models are critically needed to reduce the time and resources spent on the pilot plant stage and to enable efficient design of renewables-based chemical manufacturing plants. Engineers and Scientists at Michigan State University, Dow Chemical, Fluor, and Honeywell/UOP are building new tools to simulate the equilibria behavior of solvents and solutions relevant to the emerging bioeconomy. This GOALI project addresses many of the challenges set forth in the National Bioeconomy Blueprint by facilitating the transition from petroleum feedstocks, supporting student interaction with industry, and educating students for the bioeconomy workforce. The model will be distributed as a plug-in to commercial software and hence will broadly benefit the biochemical and chemical industries. In addition to its practical value, this research aims to improve understanding of the physical nature of hydrogen bonds. The project will fund a PhD student and engage undergraduate students in research. Through collaboration with a high school chemistry teacher, the project will develop a secondary school module addressing engineering design, part of the Next Generation Science Standards.
This project addresses a need to link fundamental molecular measurements to parameter values used in the modeling of associating fluids and their equilibria. Industrial phase equilibria models using the Wertheim approach for association, such as Statistical Associating Fluid Theory (SAFT), Cubic Plus Association (CPA) and Elliott-Suresh- Donohue (ESD) equation of state models have become increasingly common. Despite the theoretical rigor of the framework, the association parameters are usually fitted to macroscopic phase equilibria data that relate only indirectly to true associations. As a result, the physical significance of the parameter values is unclear, hindering extensions to predict multicomponent or variable temperature behaviors. The project will include three interdependent features: (a) a model designed with the mathematical flexibility to represent liquid and vapor mixtures of associating compounds over temperature and pressure ranges relevant to reaction and separation processes; (b) spectroscopic analysis to quantify associations such as hydrogen bonding at the molecular level; and (c) quantum chemical simulations both to guide interpretation of the spectroscopy and provide insight into association energies. The novelty of this work is the proposed incorporation of spectroscopic data, informed by quantum chemical simulations, to determine the extent of association at the molecular level while simultaneously including thermodynamic balances. This three-way approach will improve parameter transferability and enable more accurate correlations and predictions of temperature dependence and multicomponent behavior for vapor-liquid equilibria, liquid-liquid equilibria, and simultaneous vapor-liquid-liquid equilibria.
