The Failure of Conventional Models

In the following chapter, only the range T 0.4 K shall be discussed. The prerequisites listed in the previous chapter for a model of charge transport that could explain the experimental data eliminate most of the conventional models. 

So the temperature dependence cannot be explained by phonon scattering alone as in many conventional metals, as the conductivity is increasing with increasing temperature instead of decreasing. 

This temperature dependence is also expected in the case of granular metals, agglomerates of mesoscopic metallic particles /49/. These metallic particles could possibly be identified with small, highly-conducting regions (e.g. domains of high dopant concentration) in polyacetylene. However, these mechanisms do not fulfil the condition a finite conductivity for T- 0. 

On the other hand, it is not necessarily only one mechanism of conductivity which is responsible for the a(T) behaviour above 0.4 K. Therefore it has to be examined if these models are able to describe the temperature dependence of c at least in certain parts of this temperature range. In this case, the experimental y(T) curve could possibly be described by a superposition of different mechanisms which dominate y(T) at different temperatures. For this reason, the same experimental data are plotted in fig. 11 in five different ways the plots of figs, llb-e would give a straight line for the a(T) curve if the corresponding model could be applied, i.e. different models of 

Even in the low temperature range (14 mK T 300 mK) none of the models discussed above could be fitted to the experimental data. Because of the weak temperature dependence, however, the formula o = Oq + Oi log(T/lK) also gives an agreeable fit to the experimental data. 

---