Why boron nitride polymers?

Polymer semiconductors are becoming an increasingly attractive, versatile and cheap alternative to conventional solid-state semiconductors [29,30,31]. The key property of both conventional and polymer semiconductors is the energy band gap for electrons: combining materials with different band gaps lies at the heart of many modern electronic devices. The idea of this work came from the observation that the most popular conventional semiconductors (e.g. IV, III-V and II-VI semiconductors such as Si, GaAs and ZnSe, as well as ternary mixtures such as Al$_x$Ga$_{1-x}$As) are characterized by essentially the same crystalline structure, but made from different chemical elements. This results in different electronic properties, which can be further tuned by alternating thin layers of materials with different band gaps (e.g. GaAs and AlAs) in superlattices and quantum wells, where electrons and holes are confined within the narrow gap material. This process of engineering the band gap by hand opens up a huge number of possible choices when building electronic devices. In conjugated organic polymers (i.e. polymers with a delocalized $\pi$-electron system along the carbon backbone) the band gap usually depends on the the structural details; hence it can be varied by modifying the structure, for example by adding side chains to the carbon backbone. However, the results can be complex with potential difficulties for processing. A different route, more akin to that of conventional semiconductors, might also be possible: can we tune the electronic properties by using and combining polymers with similar structures but made from atoms other than carbon? In other words, can we engineer the band gap in a manner similar to what is routinely done in conventional semiconductors? Many carbon structures can also be made from BN: from diamond, to graphite and even nanotubes [32,33]. Replacing every pair of carbon atoms by one boron and one nitrogen atom results in stable structures, with generally larger band gaps. Would a similar substitution also hold for the organic polymer structures? Benzene, which is an essential building block of conjugated organic polymers, does indeed have a BN analogue, borazine, which is easily synthesized. In terms of dimensionality, conjugated polymers can be ranked somewhere between benzene and graphite, both characterized by $\pi$-bonding. Since these two extremes exist for BN compounds, we can argue that BN polymers could also be made. Indeed borazine-based polymers, with structures similar to the carbon polymer poly( p-phenylene), have been recently synthesized [34,35]. Experiments have been motivated so far by the search for good precursors to BN ceramics. However, we believe that BN polymers might be interesting in their own right because of their electronic properties. Making a BN polymer out of a carbon one is equivalent to making a III-V compound out of a group-IV semiconductor: using adjacent elements in the periodic table results in an unchanged structure, accompanied by an increase in the ionicity of the bonds which, in turn, increases the band gap and modifies the electronic properties. This also allows us to investigate possible routes to extend (or break) the process of conjugation in polymers. Our goal is to see how we can tailor the electronic properties by combining BN polymers with their organic counterparts.