Polymers can be assembled either via covalent or noncovalent bonds. The more common covalent polymers have the advantage of being stable over a broad range of temperatures and chemical environments. While noncovalent materials are less robust, they have the distinct advantage that their formation is reversible and therefore they can be readily assembled and dissambled in situ. It is our aim to create materials that combine the robustness of covalent polymers with the flexibility and reversibility of noncovalent materials.

To accomplish this, we will exploit the ability of polycyclic aromatic hydrocarbons to undergo [4+4]-cycloaddition reactions. For example, anthracene, when irradiated with visible light, forms a covalent dimer. This dimer, in turn, can be reconverted to its monomeric form either at elevated temperatures or by irradiation with UV light (see below).

The ability to control the assembly of molecules in this manner can be readily exploited in polymer synthesis by covalently linking two photodimerizable groups together. Upon irradiation with visible light, such bifunctional monomers should readily polymerize:

We can also extend this approach to the modification of existing polymers. For example, by exploiting the tendency of some aromatic molecules to form heterodimers rather than homodimers, it should be possible to reversibly modify the side-chains on an existing anthracene-containing polymer:

This may prove a useful way to reversibly modify a surface in situ. For example, by creating a surface that is decorated with a photodimerizable unit, it should be possible to reversibly attach various functional groups to a surface. In this manner, we hope to create surfaces with "tunable" macroscopic and microscopic properties. For example, receptors could be reversibly appended to the surface to change its affinity for a specific substrate: