NSF Award
2404048
NON-TECHNICAL SUMMARY<br/><br/>Separating different kinds of molecules in a liquid or vapor stream lies at the heart of most industrial chemical processes. Traditional separations which boil and condense liquids to separate them are energy-intensive. Polymer membranes are increasingly used to perform these separations more efficiently. Membranes also find increasing applications in producing drinkable water from brackish or sea water, as well as in fuel cells and chemical batteries, which will play an important role in an energy future powered by wind and solar. <br/><br/>Polymer membranes separate molecules by size and by affinity. Small pores allow small molecules to pass but block large molecules; membranes decorated with charges can encourage or discourage the passage of charged or polar molecules. Evidently, there are tradeoffs in membrane design: bigger pores allow molecules to pass easily, but degrade the ability of the membrane to discriminate between different molecules. Membrane design has proceeded slowly, by trial and error. An atomic-scale view of membrane structure, and how it affects the entry and passage of different molecules, would enable better designs.<br/><br/>Experimental probes of membranes on the atomic scale are vital but limited. An alternative approach is to use molecular simulations, in which movies are made of the molecules in a small region of a membrane. Simulations can in principle show how molecules pass through a membrane, and allow us to measure the affinity of the membrane for different species. But for these movies to accurately represent real molecular motion, multiple challenges must be met:<br/>1) realistic molecular arrangements of membranes must be constructed;<br/>2) forces between atoms must be realistic, particularly for strongly interacting charged species;<br/>3) simulations must cover enough time that molecules explore the membrane;<br/>4) special techniques must be developed to “encourage” molecules normally repelled by the membrane to enter, so that rejection efficiency can be measured; <br/>5) simulations in which one species is pulled through the membrane must be performed to measure the resistance experienced by molecules as they move.<br/>This project aims to meet all these challenges, and thereby enable simulations to assess the performance of membranes of different structures and compositions, which will help design better membranes for myriad applications in a sustainable future.<br/><br/>The principal investigator (PI) for this project emphasizes broader impacts in undergraduate and graduate education, including: 1) extensive online simulation tutorials; 2) a newly developed course in writing and presenting for scientists and engineers; and 3) a recently written book based on the course, which is unique in its teaching of writing and presenting together, in the broader context of the scientific enterprise.<br/><br/><br/>TECHNICAL SUMMARY<br/><br/>Aqueous membranes for reverse osmosis, ion exchange membranes for chemical flow batteries, and nanofiltration membranes for lithium recovery all face the common design challenge of readily transporting some species while strongly rejecting others. This inevitably involves tradeoffs: bigger pores improve transport but decrease selectivity, and stronger species binding promotes selectivity but impedes transport. Transport depends on the pore space geometry, membrane flexibility, and species binding to membrane moieties. Ion selectivity can be manipulated in several ways, including 1) high concentrations of bound ions that attract counterions and repel like-charge ions; 2) narrow pores too small for ions to be well solvated; and 3) favorable interactions with bound polar species.<br/><br/>Because membrane permeability and ion selectivity both depend on Angstrom-scale structure and kinetics, atomistic simulations have the potential to advance understanding and aid design of aqueous membranes for reverse osmosis, chemical flow batteries, and ion recovery. This project develops and exploits new approaches to unlock this potential, by addressing key challenges in membrane simulations:<br/>1. validated ion potentials, so that mobile ions stick properly to bound ions;<br/>2. fast atomistic simulations, to thoroughly equilibrate membrane structures;<br/>3. transfer free energies, which quantify how well a membrane excludes ions;<br/>4. full transport measurements, to predict all fluxes in response to any gradients.<br/>Atomistic simulations are well suited to explore transport and selectivity, providing unique insight to complement experimental results, if the key challenges listed above are met. Simulations can also help the membrane science community to revise and refine conflicting traditional models of transport (free-volume mediated diffusion, versus percolative flow) and selectivity (Donnan exclusion, versus size exclusion), which have persisted for decades.<br/><br/>Polymer membranes are essential elements for sustainable technologies, including reverse osmosis to supply fresh water, chemical flow batteries to store wind and solar energy, and nanofiltration to recover lithium for electric vehicles. Membrane-based separations are much more energy efficient than traditional alternatives based on phase transitions.<br/><br/>The PI’s research program emphasizes broader impacts in university education, including: 1) an extensive set of simulation tutorials online, to which new techniques developed under this project will be added; 2) a recently developed 3-credit course in writing and presenting for scientists and engineers; and 3) a recently book based on my course, which is unique in its teaching of writing and presenting together, in the broader context of the scientific enterprise.<br/><br/><br/>STATEMENT OF MERIT REVIEW<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
