Key points are not available for this paper at this time.
Artificial stimulus-responsive membranes, particularly those responsive to different solvents, have important applications in complex and graded separation systems. Inspired by natural lipid membrane that alters mass transport behavior in response to interactions with various solvents, we report that incorporating porous graphene (PG) into graphene oxide (GO) membrane enables smart and switchable molecular sieving reversibly responsive to solvent types. The membrane shows high permeance for water and methanol, 45.52 and 13.56 L m−2 h−1 bar−1, respectively, and its molecular weight cut-off (MWCO) at ~319 g mol−1 in water, similar to pristine GO membrane, reversibly switches to 960 g mol−1 in methanol which is not observed in either pristine GO or graphene membrane. We accounted this switching to the change of transport pathways. In water, the GO-GO nanochannel is dominant, providing similar molecular sieving to pristine GO. In methanol, the GO-PG nanochannel becomes favorable because a strong solvent adsorption on the nanochannel surface, coupled with a weak solvent network under nanoconfinement, promotes a significant interlayer expansion, reducing the transport resistance and enabling larger, switched MWCO. This switchable sieving behavior is further demonstrated for efficient graded separation of ternary solution of solutes with various molecular weights. Porous graphene integrated into graphene oxide membranes yields smart molecular sieves that switch separation based on the solvent. This allows for graded separations of different-sized molecules.
Li et al. (Mon,) studied this question.