Physiosorption of large molecules onto surfaces has been an area of longstanding interest for the Clarke group. Simple Van der Waals forces act to keep the molecules attached to the surface, while a variety of intermolecular interactions can lead to spontaneous self-assembly in the plane parallel to the surface. Confinement to two dimensions leads to novel phase behaviour, as well as allowing intermolecular interactions to be more easily characterised.
Recent work has focused on halogen bonds, the comparatively understudied cousin of the hydrogen bond. Halogen bonding is a function of the anisotropic charge distribution around a halogen atom (X) bonded to a substituent (Y). If Y is sufficiently electron withdrawing, a σ-hole of partial positive charge develops on X, on the opposite face to that facing Y. This can be considered as being due to the relatively low energy of the σ* antibonding molecular orbital (MO).
Such a σ-hole can then accept electrons from a lone pair, leading to a favourable halogen bonding interaction. These interactions are of the same order of magnitude as hydrogen bonds, with greater than 20 kJ mol-1 per bond being typical for the systems studied. The details of the charge distribution around the halogen also mean that such bonds are much more directional than their hydrogen bond equivalents, a key advantage for crystal engineering.
A variety of linear co-crystals have already been studied, using a combination of computer simulations, and experimental diffraction methods. The long infinite chains formed between 1,4-diiodotetrafluorobenzene (DITFB) and 4,4’- bipyridine (BPY) are pictured above.
Future work is focused towards the design of porous structures. Such structures must overcome the general close packing principle, but have been observed in hydrogen bonded species. The comparable strength and greater directionality of the halogen bond should aid design of analogous structures.
Initial studies considered a triiodo analogue (TITFB) of the previous system with BPY. Such a system would be hoped to contain hexagonal pores (pictured). Diffraction data indicates a co-crystal does form, but that it isn’t porous. Computer simulations indicate that the formation of a halogen bond increases the electron density in the TITFB, thus weakening subsequent bonds. The third halogen bond is thus destabilising and doesn’t form, leading to a close packed 2 coordinate structure.