Porous crystals constructed from the assembly of organic linking units with metal ions or metal clusters are proving to be an exciting class of materials with unprecedented properties and high potential for use in applications such as catalysis, gas storage, and separations. The underlying structure obtained from a given secondary building unit (SBU) with a multitopic ligand can be analyzed, at least with hindsight, as belonging to various nets. In some cases this approach has led to well-defined series with preserved topologies whose members vary in metrics or functionalities. At the same time, it is clear that even a single linker with a given metal under the same reaction conditions can give rise to considerable structural diversity, as is illustrated by our recent application of polymer-induced heteronucleation to the discovery of three new phases based on the simple terephthalic acid linker/zinc nitrate system. Perhaps one lesson to be learned is that there is currently no single structure-predicting design scheme but rather sets of empirically derived default behaviors that can often be used to rationalize the outcome of an experiment. In the case of employing two different linkers possessing the same coordinating functionality, experimental data are lacking, and there is no basis for answering even the most basic question of phase composition. In broad terms we can expect two different behaviors in a crystalline mixed-linker coordination polymer. The default behavior for two components combined and allowed to crystallize is segregation; this behavior forms the basis of purification by crystallization. By contrast, the default behavior when monomers of similar reactivity are combined is random copolymerization. Therefore, it is interesting to contemplate whether in a porous crystal in which strong bonds reversibly assemble the framework, copolymerization patterns will dominate or if self-sorting crystallization will prevail. The former would represent an expeditious route to discover new porous solids. To investigate this question we undertook the synthesis of porous crystals based on two organic linkers of different topologies, namely, terephthalic acid (H2BDC) and 1,3,5-tris(4-carboxyphenyl)benzene (H3BTB). H2BDC is the organic linker that, when combined with Zn, yields MOF-5, a stable cubic structure that is the best studied of the metal–organic frameworks. Using H3BTB under essentially identical synthetic conditions affords MOF177. This trigonal framework exhibits exceptional porosity and surface area and boasts the highest reported uptake of hydrogen gas in a physisorptive material. Combining H2BDC with H3BTB, two ligands possessing aryl carboxylic acid coordinating groups, in the presence of zinc nitrate incorporates both components into a completely new type of structure that can be obtained to the exclusion of materials derived from either pure linker. Figure 1 illustrates the porous crystals produced by heating various ratios of H2BDC and H3BTB in the presence of excess Zn(NO3)2·4H2O at 85 8C for 2 days. Three distinct crystalline phases are observed as the mole fraction of H3BTB is increased. At low H3BTB concentrations, only MOF-5 crystals are formed; however, at a mole ratio of 4:1 (H2BDC:H3BTB), a new needle-shaped phase is formed along with MOF-5. Increasing the H3BTB concentration leads to exclusive formation of the needleshaped phase, which according to powder X-ray diffraction (XRD) data is different from MOF-5 and MOF-177 (Figure S4 in the Supporting Information). A further increase of the H3BTB concentration results in MOF-177 forming as well. Finally, at H2BDC:H3BTB mole ratios greater than or equal to 2:3, MOF-177 is the first product to crystallize out of solution. A single-crystal X-ray diffraction study of the needleshaped crystals revealed a structure with one-dimensional hexagonal channels. The product crystallizes in the space group P63m and dramatically differs from the structures derived from the pure linkers. The framework of the material consists of Zn4O clusters linked together by two BDC and four BTB linkers arranged in an octahedral geometry (Figure 2a). Two BDC linkers are adjacent, leaving the other four positions occupied by BTB linkers, and these octahedra assemble into a structure containing both micropores and mesopores. This product is denoted as UMCM-1 (University of Michigan Crystalline Material-1). The micropores are found in cage-like structures constructed from six BDC linkers, five BTB linkers, and nine Zn4O clusters, and with an internal dimension of approximately 1.4 nm@1.7 nm (subtracting the van der Waals radii of the atoms, Figure 2b). Six such microporous cages assemble together in an edgesharing fashion to define the diameter of the mesopore, a 1D hexagonal channel 2.7 nm@ 3.2 nm (measured between pore walls, Figure 2c). When van der Waals radii of the atoms are taken into account, the mesopore is 2.4 nm@ 2.9 nm. Comparison of the bulk powder XRD pattern to that simulated [*] Dr. K. Koh, Dr. A. G. Wong-Foy, Prof. A. J. Matzger Department of Chemistry and the Macromolecular Science and Engineering Program University of Michigan 930 North University Ave, Ann Arbor, MI 48109-1055 (USA) Fax: (+1)734-6715-8853 E-mail: matzger@umich.edu
Search from 250 M+ research papersSearch from 250 M+ research papers
Open Access
Apr 25, 2024
Su-Fen Li + 5
