Publication
The ultimate thinness of graphene pores makes them ideal as a high-flux selective layer for manipulating molecular transport. However, due to the limited understanding of the formation of ångstrom (Å)-scale pores, controlling pore size and understanding gas transport across realistic graphene pores remains challenging. This thesis focuses on understanding pore formation in graphene to develop membranes with a high density of pores. Building on this understanding, it then seeks to reveal gas transport mechanisms across realistic pores to guide experimental design.
A novel photonic gasification method to form gas-sieving pores is introduced. It is first demonstrated that during oxidation, oxygen clusters form and then evolve into a core/shell structure, with an ether core surrounded by an epoxy group. This organization is driven by an attempt to minimize lattice strain. The core of the oxygen cluster was then selectively gasified at room temperature using 3.2 eV light, leading to pore at the cluster's center. By leveraging oxidation temperature, the feasibility of maintaining a narrow pore size distribution, independent of pore density, is demonstrated. A simultaneous increase in gas permeance and gas pair selectivity was observed, overcoming a common trade-off and long-standing challenge in the field. Ultrahigh H2 permeance of 12000 gas permeation units (GPU), along with a highly attractive H2/N2 selectivity of 50, was achieved.
Building on the understanding of the 2D pore formation mechanism mentioned above, molecular dynamics (MD) simulations were conducted to explore CO2/O2 mixture separation across realistic graphene pores prepared by oxidation where semiquinone groups are present at the pore edge. Herein, using molecular dynamics (MD) simulations, we show that C=O displays a remarkable molecular-interaction-dependent dynamic motion, leading to a distribution in PLD comparable to the size differences between CO2, O2, and N2. Dynamic open and closed pore states are observed in small pores, making impermeable pores CO2-permeable. The strong molecular interaction eliminates effusive transport, resulting in selective gating of CO2 from O2 and N2, even from large PLD pores expected to be nonselective. Finally, transition-state-theory calculations validated against MD simulations reveal the immense potential of porous graphene for carbon capture beyond the state-of-the-art membranes. These insights will inspire improved graphene membrane design, pushing the carbon capture frontier.
Finally, the selective transport of NH3 from graphene pores was explored. Attractive separation performance from centimeter-scale porous graphene was achieved with NH3 permeance of 16000 GPU accompanying NH3/N2 selectivity up to 105. MD simulations revealed competitive adsorption between NH3 at graphene pores, where NH3 transports solely by an adsorbed-phase pathway blocking the access for N2. Both simulations and experiments are showing that 2D graphene pores are pro