Engineering Inner-Surface Electrostatic Potential of Robust B←N Dative Cages for Efficient Benzene/Cyclohexane Separation.

Benzene (Bz) and cyclohexane (Cy) pose formidable separation challenges owing to their nearly identical physicochemical properties, while their inherent electronic structural differences offer a viable avenue for selective discrimination. Herein, we report an efficient separation strategy via precise electrostatic potential engineering of confined cage cavities to realize electronic structure matching with the aromatic π-system of Bz. Guided by this principle, a series of robust trigonal prismatic B←N dative cages with continuously tunable inner-cavity electrostatic potential distributions were fabricated by synergistically modulating tritopic and ditopic building subunits. Dynamic breakthrough experiments verified the preferential adsorption of Bz over Cy for all synthesized cages, with SNNU-735 delivering a superior breakthrough time of 104 min g-1. Liquid-phase competitive adsorption tests further validated the selective Bz recognition capability, demonstrating the universal applicability of the strategy in both vapor and liquid phases. Combined single-crystal structural analysis of Bz-loaded cages and theoretical calculations reveal that cavity electrostatic potential engineering optimizes electronic matching between the cage inner surface and Bz aromatic π-surface, which enhances host-guest interactions and achieves specific molecular recognition.