Mechanism of Carbon Dioxide Adsorption in a Highly Selective Coordination Network Supported by Direct Structural Evidence

Understanding the interactions between adsorbed gas molecules and a pore surface at molecular level is vital to exploration and attempts at rational development of gasselective nanoporous solids. Much current work focuses on the design of functionalized metal–organic frameworks (MOFs) or coordination networks (CNs) that selectively adsorb CO2. [1–9] While interactions between CO2 molecules and the p clouds of aromatic linkers in MOFs under ambient conditions have been explored theoretically, no direct structure evidence of such interactions are reported to date. Here we provide the first structural insight of such interactions in a porous calcium based CN using single-crystal X-ray diffraction methods, supported by powder diffraction coupled with differential scanning calorimetry (DSC-XRD), in situ IR/Raman spectroscopy, and molecular simulation data. We further postulate that such interactions are responsible for the high CO2/N2 adsorption selectivity, even in the case of a high relative humidity (RH). Our data suggest that the key interaction responsible for such selectivity, the room-temperature stability and the relative insensitivity to the RH of the CO2-CN adduct, is between two phenyl rings of the linker in the CN and the molecular quadrupole of CO2. The specific geometry of the linker molecule results in a “pocket” where carbon from the CO2 molecule is placed between two centroids of the aromatic ring. Our experimental confirmation of this variation on theoretically postulated interactions between CO2 and a phenyl ring will promote the search for other CNs containing phenyl ring pockets. Selective adsorption and sequestration of CO2 from sources of anthropogenic emissions, such as untreated waste from flue gas and products of the water gas shift reaction, is important to mitigate the growing level of atmospheric CO2. [10] Current separation methods use absorption in alkanolamine solutions, which are toxic, corrosive, and require significant energy for their regeneration. Hence microporous solid-state adsorbents, such as zeolites, hybrid zeolite–polymer systems, porous organic materials, and MOFs are proposed as alternatives, especially in combination with pressure swing processes. Rather than relying solely on tuning the pore diameters of microporous materials to select between gases based on size (the kinetic diameters of CO2, CH4 and N2 are 3.30, 3.76 3.64 , respectively ) selective separation relies on differences in electronic properties, such as the quadrupole moment and polarizability. Attempts to produce MOFs or CNs with adsorption properties competitive with those of commercially established aluminosilicate zeolites, relies on strategies that include pore surface modification with strongly polarizing functional groups, such as amines 7, 9,15] and desolvating metals centers 8, 16] to produce low-coordinated sites suitable for CO2 adsorption. The amine-functionalized materials offer a high selectivity toward CO2 adsorption, but a low effective surface area and thus, a low total uptake capacity. Strong interactions with polarizing functional groups, as well as with open metal sites presents other drawbacks including an increase in the costs for material regeneration. Furthermore, water effectively competes with CO2 at low-coordinated cation sites, impeding the performance of frameworks in commercial flue gas. We recently described a porous framework, CaSDB (SDB: sulfonyldibenzoate, compound 1) with a high CO2/N2 selectivity. At 0.15 bar of CO2 and 0.85 bar of N2, a typical composition of flue gas mixture from power plants, the selectivity is in the range of 48 to 85 at 298 K. CaSDB shows a reversible uptake of CO2 of 5.75 wt% at 273 K and 1 bar pressure and 4.37 wt% at room temperature, with heats of adsorption for CO2 and N2 of 31 and 19 kJmol , respectively. The as-synthesized compound contains not coordinated water molecules and is easily activated for gas adsorption by heating to 563 K in vacuum; remarkably the activated framework does not readsorb water, even if exposed to a RH greater than [*] A. M. Plonka, W. R. Woerner, Prof. Dr. J. B. Parise Department of Geosciences, Stony Brook University Stony Brook, NY 11794-2100 (USA) E-mail: john.parise@stonybrook.edu