The energy costs associated with the separation and purification of industrial commodities, such as gases, fine chemicals and fresh water, currently represent around 15 per cent of global energy production, and the demand for such commodities is projected to triple by 2050 (ref. 1). The challenge of developing effective separation and purification technologies that have much smaller energy footprints is greater for carbon dioxide (CO2) than for other gases; in addition to its involvement in climate change, CO2 is an impurity in natural gas, biogas (natural gas produced from biomass), syngas (CO/H2, the main source of hydrogen in refineries) and many other gas streams. In the context of porous crystalline materials that can exploit both equilibrium and kinetic selectivity, size selectivity and targeted molecular recognition are attractive characteristics for CO2 separation and capture, as exemplified by zeolites 5A and 13X (ref. 2), as well as metal-organic materials (MOMs). Here we report that a crystal engineering or reticular chemistry strategy that controls pore functionality and size in a series of MOMs with coordinately saturated metal centres and periodically arrayed hexafluorosilicate (SiF(2-)(6)) anions enables a 'sweet spot' of kinetics and thermodynamics that offers high volumetric uptake at low CO2 partial pressure (less than 0.15 bar). Most importantly, such MOMs offer an unprecedented CO2 sorption selectivity over N2, H2 and CH4, even in the presence of moisture. These MOMs are therefore relevant to CO2 separation in the context of post-combustion (flue gas, CO2/N2), pre-combustion (shifted synthesis gas stream, CO2/H2) and natural gas upgrading (natural gas clean-up, CO2/CH4).

We report MPM-1-TIFSIX, a molecular porous material (MPM) based upon the neutral metal complex [Cu2(adenine)4(TiF6)2], that self-assembles through a hydrogen-bonding network. This MPM is amenable to room-temperature synthesis and activation. Gas adsorption measurements and ideal adsorbed solution theory selectivity predictions at 298 K revealed enhanced CO2 separation performance relative to a previously known variant as well as the highest CO2 uptake and isosteric heat of adsorption yet reported for an MPM. MPM-1-TIFSIX is thermally stable to 568 K and retains porosity and capacity even after immersion in water for 24 h.

Reaction of biphenyl-3,4',5-tricarboxylate (H(3)BPT) and CdCl(2) in the presence of meso-tetra(N-methyl-4-pyridyl)porphine tetratosylate (TMPyP) afforded porph@MOM-10, a microporous metal-organic material containing CdTMPyP cations encapsulated in an anionic Cd(II) carboxylate framework, [Cd(6)(BPT)(4)Cl(4)(H(2)O)(4)]. Porph@MOM-10 is a versatile platform that undergoes exchange to serve as the parent of a series of porph@MOMs that exhibit permanent porosity and heterogeneous catalytic activity.

Grand canonical Monte Carlo (GCMC) studies of the mechanism of hydrogen sorption in an rht-MOF known as Cu-TPBTM are presented. The MOF is a decorated/substituted isostructural analogue to the unembellished rht-MOF, PCN-61, that was studied previously [ Forrest K. A. Forrest K. A. J. Phys. Chem. C20121161553815549]. The simulations were performed using three different hydrogen potentials of increasing complexity. Simulated hydrogen sorption isotherms and calculated isosteric heat of adsorption, Q st, values were in excellent agreement with the reported experimental data for only a polarizable model in one of four experimentally observed crystal structure configurations. The study demonstrates the ability of modeling to distinguish the differential sorption of distinct strucures; one configuration is found to be dominant due to favorable interactions with substrates. In addition, it was discovered that the presence of polar amide groups had a significant effect on the electrostatics of the Cu2+ ions and directs the low-pressure physisorption of hydrogen in the MOF. This is in contrast to what was observed in PCN-61, where an exterior copper ion had a higher relative charge and was the favored loading site. This tunability of the electrostatics of the copper ions via chemical substitution on the MOF framework can be explained by the presence of the negatively charged oxygen atom of the amide group that causes the interior Cu2+ ion to exhibit a higher positive charge through an inductive effect. Further, control simulations, taking advantage of the flexibility afforded by theoretical modeling, include artificially modified charges for both Cu2+ ions chosen equal to or with a higher charge on the exterior Cu2+ ion. This choice resulted in distinctly different hydrogen sorption characteristics in Cu-TPBTM with no direct sorption on the open-metal sites. Thus, this study demonstrates both the tunable nature of MOF platforms and the possibility for rational design of sorption/catalytic sites and characteristics through the active interplay of theory and experiment. The ability of accurate, carefully parametrized and transferable force fields to model and predict small molecule sorption in MOFs, even including open-metal sites, is demonstrated.

Pillar substitution in a long-known metal-organic material with saturated metal centres, [Cu(bipy)(2)(SiF(6))](n), has afforded the first crystallographically characterized porous materials based upon TiF(6)(2-) and SnF(6)(2-) anions as pillars. Gas adsorption studies revealed similar surface areas and adsorption isotherms but enhanced selectivity towards CO(2)vs. CH(4) and N(2).