Ion-Ion Selectivity

Precise ion separation is highly desired for resource recovery, water reuse, and energy storage technologies. Yet, conventional membranes lack the ability to achieve high ion-ion selectivity. To guide the design of ion-selective membranes, we have introduced the transition-state theory (TST) to elucidate the molecular mechanisms governing ion transport through sub-nanometer pores. This energy barrier-based framework explains ion selectivity in terms of enthalpy and entropy changes during membrane transport. Using this theoretical approach, we systematically investigate how molecular-level physical phenomena, including ion dehydration, ion pairing, and ion-membrane electrostatic interaction, influence ion-ion selectivity.

Schematic showing the transition-state theory (TST) for membrane transport. The yellow-shaded area represents the transitional states of permeants (ions) during membrane transport, with the peaks in the potential energy profiles indicating the activation energy barriers.

Our mechanistic investigation reveals that precisely tailing membrane pore structures and ion-membrane interactions is crucial for designing highly selective membranes. Using coordination chemistry, we have developed a multilayer polymer membrane with highly selective ion-binding sites. This membrane achieves exceptional selectivity for copper over magnesium, tackling one of the most challenging separations between same-valence ions. Further enhancement of ion-ion selectivity requires reducing membrane thickness (pore length) and optimizing the spatial arrangement of ion-binding sites within membrane pores.

Solvent Transport

Reverse osmosis (RO) and nanofiltration (NF) are the most widely used pressure-driven membrane processes. Historically, two models have been proposed to describe water and solvent transport through their sub-nanometer pores: (i) the pore flow model, where solvent transport is driven by a pressure gradient, and (ii) the solution-diffusion (SD) model, which assumes solvent transport occurs via diffusion down a concentration gradient. For decades, the SD model has been the dominant explanation for transport in RO membranes.

However, our recent research, combining permeation experiments, physical modeling, and non-equilibrium molecular dynamics, has demonstrated that water and organic solvent transport in RO and NF membranes follows a pore flow mechanism, NOT solution diffusion. Crucially, the fundamental assumptions of the SD model, such as the presence of a concentration gradient for molecular diffusion, do not hold in these membranes. Instead, by adopting a pore flow mechanism and considering the frictions among water (solvent), solutes, and membrane, we have established the solution-friction (SF) model, which accurately describes the transport of solvent and solute in sub-nanometer pores. Additionally, our research has developed a transport theory for solvent permeation under membrane compaction, incorporating pore structure changes induced by solvent flow. The new transport framework based on the true physical mechanisms informs the development of next-generation membranes that achieve high permeability and selectivity through precise control of transport frictions. 

Schematic illustrating non-equilibrium molecular dynamics simulations of water transport in polyamide RO membranes. The pressure and water molecule density profiles during permeation confirm that water transport in RO membranes follows a pore flow mechanism.

Gas Transport

Membrane-based gas separation is an energy-efficient technology widely applied in industrial processes. However, gas transport mechanisms within polymeric membranes remain poorly understood, limiting the optimization of separation performance.

Using non-equilibrium molecular dynamics simulations, we have investigated gas permeation through aromatic polyamide membranes and compared it with the classical SD behavior observed in homogeneous liquid (water) films. Our results reveal that gas transport in polyamide membranes exhibits nanoporous material-like characteristics, including molecular sieving effects, where smaller gases such as helium and hydrogen exhibit markedly higher flux than larger gases like carbon dioxide and methane. Unlike the random walk behavior assumed in the classical SD model, gas molecules in polyamide membranes follow directional trajectories and exhibit non-uniform distributions, a consequence of the low fluidity of polyamide. This stabilized pore structure enables well-defined pathways for smaller gases, significantly influencing selectivity. Our findings highlight the need to integrate pore size distribution, connectivity, and porosity into gas transport models to better predict and enhance membrane-based gas separation performance.

Schematic illustrating non-equilibrium molecular dynamics simulations of gas transport in polyamide membranes. Molecular sieving effects, non-uniform gas distributions, and directional trajectories indicate that gas transport in polyamide membranes does not follow the classic Fickian diffusion.

Representative Publications

Epsztein, R., DuChanois, R. M., Ritt, C. L., Noy, A., & Elimelech, M. (2020). Towards single-species selectivity of membranes with subnanometre pores. Nature Nanotechnology, 15(6), 426-436. DOI: 10.1038/s41565-020-0713-6

DuChanois, R. M., Heiranian, M., Yang, J., Porter, C. J., Li, Q., Zhang, X., … & Elimelech, M. (2022). Designing polymeric membranes with coordination chemistry for high-precision ion separations. Science Advances, 8(9), eabm9436. DOI:10.1126/sciadv.abm9436

Wang, L., He, J., Heiranian, M., Fan, H., Song, L., Li, Y., & Elimelech, M. (2023). Water transport in reverse osmosis membranes is governed by pore flow, not a solution-diffusion mechanism. Science Advances, 9(15), eadf8488. DOI:10.1126/sciadv.adf8488

Qian, J., Wang, R., Wu, H., Wang, F., & Elimelech, M. (2025). Molecular Simulations Reveal Gas Transport Mechanisms in Polyamide Membranes. Under Review.