“…Initially, the dye was incorporated into a 1 ym thickpolymer film [36] (often polyimide), spin-coated onto the I T 0 electrodes; this was not without its problems, as the resistive polymer layer dictated that a higher voltage was required to cause dielectric reorientation of the smectic phase. It was also calculated that if the dye could be dissolved in the liquid crystal itself, rather than coated onto the substrate, the efficiency could almost be doubled [37] and the line width reduced; a large proportion of the heat generated is dissipated into the substrate and wasted in heating the substrate, rather than the liquid crystal film, and causing line broadening [23]. However, when traditional dyes were tested it was found that they had poor solubility in the liquid crystal and poor stability; the first attempts were abandoned.…”