Homogeneous doping and semiconductor-to-"metal" transition in polyacetylene

The conductivity of iodine-doped cis- and trans-polyacetylenes was measured as a function of dopant concentration per C-H unit, $y$, from parts-per-million to the metallic regime. The sharp semiconductor-to-"metal" ($S-M$) transitions span over only 2 orders of magnitude of $y$ and the $log\sigma -\mathrm{vs}-logy$ relationships for the two isomers are identical over a large portion of the transitional region. The change in thermoelectric power $S$ occurs even more abruptly over only twofold increase in dopant concentration. The midpoints of the transition for $\sigma$ vs $y$ and $S$ vs $y$ are nearly the same at $y\sim 7\times {10}^{-4\mathrm{}}$. The number of Curie spins decreases rapidly during the transition; the EPR linewidth and line shape of cis-polyacetylene were not affected by doping. These behaviors are suggested as indicative of uniform doping. Polyacetylenes, both cis and trans, doped with As${\mathrm{F}}_5$ by a "slow" method yielded inhomogeneous materials devoid of a distinct $S-M$ transition. Using an improved "cyclic" doping procedure, we obtained homogeneously doped trans-${\left[\mathrm{CH}{\left(\mathrm{As}{\mathrm{F}}_5\right)}_y\right]}_x$ which exhibits $\sigma -\mathrm{vs}-y$ and $\sigma -\mathrm{vs}-S$ dependence nearly the same as iodine-doped polymers. The results are interpreted in terms of a phase transition. In lightly doped materials the charged soliton and polaron defects are pinned by the potential of one dopant ion, and these materials are said to be in a glassy state. Conversion to a liquid state takes place when the average separation of uniformly distributed dopant ions is less than the defect domain widths so that a defect interacts with two or more dopant ions. Conduction is probably via electron hopping between the charged soliton and polaron states in the transitional regime. This solition melting transition occurs at about 100-fold lower in dopant concentration than that for the onset of Pauli susceptibility. We were unable to achieve uniform doping of cis-polyacetylene with As${\mathrm{F}}_5$.