Polymerising pores

For a recent paper from my JSPS Post-Doctoral Fellowship in the Furukawa group, we managed to turn a random thought in a group meeting [1] into a Nature Communications article (DOI: 10.1038/s41467-018-04834-0). More importantly, we found a way to control porosity in amorphous materials.
Polymerising pores

The paper in Nature Communications is here: go.nature.com/2LlMEhY

Kyoto University is well known for research into porous materials based on coordination bonds, often called porous coordination polymers or metal-organic frameworks (PCPs/MOFs). For most coordination chemists working with these porous materials, seeing an amorphous pattern in an X-ray diffractogram means bad news. Generally, the first step in assessing PCPs/MOFs is by looking at their crystallinity - the crystalline network often provides the stability required to withstand the desolvation process that unblocks the pores and makes them accessible. While crystallinity helps us to understand a material´s performance, it can be difficult to achieve crystallinity at the same time as tailoring the morphology of the final product, which is an important consideration for materials processing. Thus, we started investigating alternative ways of controlling porosity in coordination materials that could bypass the need for crystallinity. By the end of my Fellowship in the Furukawa group, few things made us happier than seeing that our reactions yielded fully amorphous materials.

The common approach to synthesize porous coordination polymers is based on linking metal ions or clusters and organic linkers into extended crystalline networks, in which the pores come from the architecture of the network. In other words, the pores (or potential pores) only exist once the whole network is built. We thought we could start by building from the pore. By introducing the porous unit at the beginning of the reaction as a building block, their presence would be ensured by the components of the network and not only by the overall structure. That way, we would have more freedom to think about the final morphology of our materials and how to shape it, without worrying about crystallinity and long-range order. To do that, first we needed pre-formed, porous molecules to use as building blocks, or as we began to call them porous monomers, in the coordination reactions. We then needed a molecule to link our monomers into polymers.

The porous molecules had to be able to withstand solution-phase assembly into polymers, and also withstand the desolvation process. These requirements ruled out many porous molecules based on coordination bonds, which are often assembled with labile coordination bonds such as the Cu(II)-carboxylate bond. These bonds have a tendency to break, which would cause the decomposition of our molecule and the loss of our pore. In fact, it turned out that we already had the ideal candidates for porous monomers in our lab: rhodium-based metal-organic polyhedra (Rh-MOPs). These molecules are robust thanks to the strong rhodium–carboxylate bond in the equatorial site, but the axial sites are more labile. In addition, changes in the coordination on these axial sites cause clear changes in the colour of the solutions. These features allowed us to use simple spectroscopic techniques to follow what was happening at the axial sites of our MOPs. By considering the affinity of N-donors for Rh(II) ion centers, we came to use imidazole based ligands to link our monomers. A key part of the work turned out to be the use of dynamic light scattering (DLS), which we used to track the growth of our polymers. More significantly, we could use the results of the DLS and the spectroscopic measurements to understand how our polymers were growing. It was this insight into the mechanism behind the polymerisation that let us control it and target the different morphologies we describe in the article, coordination polymer particles and gels.

We believe that this work is a first step. Coordination chemistry provides an endless combination of linkers and porous monomers to explore. We hope that these initial results will trigger new research efforts focused on understanding how porous molecules self-assemble from the molecular scale to macroscopic shapes, which can provide a real advance in the current challenge that is the synthesis of soft matter that is both permanently porous and amenable to materials processing.


[1] “Can we polymerise MOPs?”, S. Furukawa, 2014, iCeMS, Kyoto University. 


We thank the Hayanon’s Science Manga Studio for the image illustrating this post.