AmericanChemicalSociety.com

A library of porous polymers was successfully synthesized through the reactions of various aryl amines with boron sources (borane or boron trihalide) by thermolysis reactions and under catalytic conditions.  Amine-borane adducts are known to undergo dehydrocoupling processes to form cyclic aminoboranes or borazines.  Extensive studies have been done on these reactions using homogeneous and heterogeneous catalysis.  In addition to thermolysis, we have employed homogeneous catalytic systems in which catalytic residuals can be rinsed out of the pores after polymer formation is complete as well as heterogeneous catalysts whereby the metal is incorporated into the porous structure.  These polymers resulted in the formation of borazine rings—the first time that borazine has been incorporated in a polymeric form. 

What makes borazine such an attractive building block compared to their porous polymer analogs is their tunable chemical nature which can be accessed by careful selection of the building blocks.  The impact of the building block choice on the chemical nature of the polymer is illustrated through the selection of the aryl amine.  Choosing one amine or another has resulted in polymers with pore decoration utilizing either three atoms or six.  Additionally, the type of atom in the boron source starting material affects the resulting polymer.  A number of different halides as well as hydrogen can be connected to the boron atom.  As a result, an induced size and electronic effect on the pores of the polymer emerges.  This combination between boron source and amine building blocks provides a simple way to tune the hydrophobicity of the channels.  In addition, post-synthesis modification could take advantage of the reactive boron-halide bonds to yield yet more polymers in this library.

Models were generated using space groups and vertex positions obtained from the Reticular Chemistry Structure Resource (http://rcsr.anu.edu.au) for the expected structures. The triangular vertices were replaced by a borazine (B₃N₃) ring with the nitrogen atoms pointing along the means of extension. The isolation and handling of all products were performed under nitrogen using either glove-box or Schlenk line techniques with dry nitrogen. Before synthesis of each polymer, solvents were distilled over the appropriate drying agent. To establish the chemical connectivity, geometry, and composition of the polymers, a battery of tests was performed to fully characterize these materials including FT-IR, solid-state NMR, and elemental analysis.  In each polymer’s case, the FT-IR spectra showed depletion of the characteristic starting material peaks including the amine hydrogen stretches which occur around 3300 cm^-1, coupled with the formation of borazine rings in both the stretching and bending bands. In general, a dominant stretch occurs at around 1400 cm^-1 which is consistent with all previously reported data on the free borazine ring.  Solid-state NMR showed that each boron atom possessed the intended trigonal shift and symmetry value and were in sharp contrast to tetrahedrally-bound boron.  Also, the elemental analysis showed excellent agreement between the intended products and the actual results in all cases.  Phase purity was investigated for each sample using scanning electron microscopy revealing unique morphology for samples prepared by thermolysis.  Particles were spherical in shape with diameters of approximately 150 – 200 nm. SEM images of catalyzed materials show similar morphologies, yet the phase of particles is not completely homogeneous. Unlike microcrystalline COFs, the amorphous nature of our polymers precluded their investigation by powder-XRD technique, however we have confirmed that crystallinity is not a prerequisite for high porosity.  To assess each polymer’s thermal stability, thermogravimetric analysis was performed.  Each polymer showed significant thermal stabilities ranging from 250 to 500 °C thus illustrating their sustainability at mainstream application temperatures.  Additionally, nitrogen sorption experiments (Figure 1, as an example) were performed to investigate the polymers’ permanent porosity following an activation process. 

All polymers exhibited isotherms related to polymers of microporous nature, which is consistent with their theoretical computer-generated models.  The synthesized polymers yielded BET surface areas between 503 and 1364 m²g^-1.  Surface areas of that magnitude are attractive for a wide range of applications including gas storage and catalysis.  Additionally, the pore size distributions, as calculated from Non-Local Density Functional Theory, showed that these polymers did indeed conform to one of the two possible topologies. The calculated NLDFT isotherm also agreed closely with experimental N₂ adsorption isotherms.

 We also investigated the potential use of borazine-linked polymers in gas storage capabilities by performing sorption experiments on other gases such as hydrogen, carbon dioxide, and methane. One of the advantages of hydrogen physisorption in porous polymers is the readily reversible loading process due to a relatively low heat of sorption compared to other storage media such as metal hydrides.  However, because of this low heat of adsorption, cryogenic temperatures are required to reach high storage capacities. Hence, our goal was to reach a maximum isosteric heat of adsorption, Q_st.  Hydrogen Q_st values at zero coverage was calculated. Though lower than required to reach goals for use at ambient temperature and pressure, in several cases, these values surpassed that of most studied hydrogen storage media to date. These results along with the gas storage isotherms showed promising capabilities for these polymers as media for gas storage as well as gas separation in processes such as carbon dioxide capture and sequestration.

Much of this work has been submitted for publication already, and another publication is currently in preparation.  Additionally, the natural extension of this work was submitted via proposal to the U.S. Department of Energy Office of Basic Energy Sciences and was accepted.