Take your time and get your first COF single crystal diffraction structure and a Science

The synthetic polymers ubiquitous in our life, such as nylon, polyester, Teflon, polyethylene, etc., are almost always made of long linear polymer chain structures entangled with each other, much like a tangled mass of spaghetti. In contrast to this more random microstructure, covalent organic framework (COF), with its regular and ordered lattice structure, is a promising polymeric material that has attracted a lot of attention in the last decade or so in the fields of gas adsorption and separation, catalysis, sensing, and energy storage. Compared with the earlier developed metal organic framework (MOF), COF materials have a number of obvious advantages, such as metal-free, more stable, lighter weight and lower cost. At present, although many COF materials have been reported, due to the difficulty of controlling the nucleation and crystal growth process, the final products are basically polycrystalline or amorphous, and few large-size, single-crystalline and stable COF materials have been reported. Since these COF materials with polycrystalline or amorphous structures have shown remarkable performance, people can't help but look forward to the surprises that large-size, single-crystal COF materials will bring!

In two recent papers published back-to-back in Science, two research teams have used different strategies to obtain large-size, high-quality COF single-crystal structures. In this paper, Prof. Wei Wang of Lanzhou University and Prof. Junliang Sun of Peking University, in collaboration with Prof. Omar M. Yaghi of UC Berkeley, the inventor of COF materials, used excess aniline as a nucleation inhibitor and competitive regulator to achieve the first large-size, high-quality single crystal growth and X-ray diffraction (SXRD) of three-dimensional COF based on imine bonds. -ray diffraction (SXRD) structure analysis. Another paper, by Prof. William R. Dichtel's group at Northwestern University, used a "two-step" approach to separate the nucleation and crystallization processes to obtain high-quality borate ester-based single crystals. Today, I will focus on the work of the LANU-PU-UC Berkeley team, and an explanation of the work of Prof. Dichtel's team will be released tomorrow.

Prof. Wei Wang (left), Prof. Junliang Sun (center), and Prof. Omar M. Yaghi (right). Photo credit: Lanzhou University / Peking University / UC Berkeley.

Although the polycrystalline structure of COF has been widely studied, a "more realistic" single-crystal SXRD structure has not been available, which has become a major challenge for researchers in this field. High-quality single-crystal growth requires dynamic covalent bond formation and breakage, i.e., reversibility of bonding. The authors argue that the main reason why previous COF structures are prone to amorphous materials or very small nanocrystals and polycrystals is that the bonding is too fast. If the bonding speed is slow enough, it is expected to increase the reversibility and the defects in the growth of COF crystals will have a chance to be "self-corrected", resulting in large size and high-quality single crystals. So how can bonding speed be controlled?
The third generation COFs use imine bonds to link two ligands (typically polyamines and polyalkaldehydes), and the authors envision a clever "imine exchange" strategy to control the reaction rate. Specifically, an excess of aniline is added to the polyamine and polyaldehyde ligands as a competitive modulator and nucleation inhibitor. Aniline is comparable in reactivity to the COF ligand polyamine and contains only one amino group, which competes with the polyamine ligand; in addition, the imine bond is more susceptible to nucleophilic attack by amines rather than water (Figure A below). Therefore, the addition of excess aniline is theoretically promising to increase the reversibility of imine bond formation, improve the self-correction process of crystal defects, and ultimately optimize crystallization. The experimental results confirmed their vision, and they obtained a variety of large-size, high-quality single crystals of COF materials based on imine bonds, including COF-300, COF-303, LZU-79 and LZU-111 (Figure B below), with single crystal sizes ranging from 10-100 μm and optical microscopy to identify their crystalline states. Moreover, the crystal size can be controlled by adjusting the amount of aniline added; in the case of COF-300, for example, different amounts of aniline added can yield COF-300 single crystals in the range of 10-60 μm.

Aniline regulates the single crystal growth of COF materials based on imine bonds. Image source: Science

In the case of COF-300, the ligand used is tetrakis(4-aminophenyl) methane (TAM) with terephthalaldehyde (BDA). In the previous synthesis method without the addition of aniline, only polycrystals of about 500 microns were obtained (Figure B, first row, far left, above). With the addition of excess aniline, high quality COF-300 single crystals of up to 100 µm were obtained after 30-40 days of "slow" growth. With the high-quality single crystals in hand, the authors obtained the single crystal structure of this COF material at 0.85 Å resolution by SXRD (Figure A below). Exposing COF-300 to water, the water-absorbed material structure was further distorted with one-dimensional chain-like water clusters within the pore channels of its crystal structure. The distortion of the pore channels is thought to be related to the hydrogen bonding between the imine group and the guest molecular water (Figure B/C below).

SXRD structures of COF-300 and COF-300 containing guest molecular water. Image source: Science

When interchanging the amino and aldehyde groups in the COF-300 ligand, a similar structure COF-303 can be obtained (Figure A below). Since COF-303 has the same topology, space group and seven-fold interpolation as COF-300, the conventional powder diffraction technique simply cannot distinguish COF-300 from COF-303, while the single-crystal-based SXRD technique can. When the two crystal structures of COF-300 and COF-303 are overlapped together, the difference in the distortion angle of the imine bond can be clearly discerned (Figure B below).

The SXRD structure of COF-303 and the overlapping structure map with COF-303. Image source: Science

To verify the universality of this strategy, the authors synthesized a new chiral COF, LZU-111, which consists of two tetrahedral-type ligands TAM and tetra(4-formylphenyl) silane (TFS) to form a three-dimensional structure. Excess aniline was added during the synthesis, and the final LZU-111 single crystal was a hexagonal prismatic single crystal of approximately 60 μm (Figure A below). To improve the resolution, the authors used a synchrotron light source for SXRD, which increased the resolution to 1.2 Å. Nitrogen adsorption experiments showed that LZU-111 has a high specific surface area of 2120 m2 g-1 and a pore volume of 0.918 cm3 g-1, which is generally consistent with the theoretical predictions based on the SXRD model (Figure B below). Finally, the authors also characterized its chemical shifts by solid-state NMR (Figure C below).

SXRD structure and characterization of LZU-111. Image source: Science