Answers are grounded in this paper and cite their sources.
Localized states are characterized kinetically (via rate constants) and separated into transport states and traps according to whether the states actually enable transport (mobilize a particle) or disable transport (or act as temporary immobilizers). A technique is developed for effecting the separation uniquely. It is based on the principle that the mobility of a particle should be maximum under trap-free conditions. An alternative percolation technique is also described which is believed to produce the same result. Either technique is general and can be applied to any array of states linked by a set of transition probabilities, such as those which define the Pauli master equation (PME). The separated states and certain approximations then make it possible to transform the PME into a generalized form of trapping equations, which reduce to the conventional trapping equations for extended-state transport in the limit in which the mean hopping time between transport states tends to zero. The generalized trapping equations are further shown to be exactly equivalent to the generalized master equation for continuous-time random walks (CTRW). Thus a new physical meaning of random walks emerges which is curiously more general and more specific than the previous meaning. It is more general in the sense that it now includes random stopping (trapping) as well as random motion (diffusion and drift). It is more specific in the sense that the rate constants which define the trapping equations also uniquely determine the waiting-time density function ψ(t) for a corresponding walk. Our separation of states also leads naturally to distinctly different mechanisms of diffusion, trapping and dielectric polarization (dipole relaxation), whereas these same phenomena necessarily appear indistinguishable when all localized states are treated equally. Dielectric polarization, or a.c. conduction, is identified as a Debye relaxation process within isolated pairs or clusters of defect states which act collectively as complex polarizable traps, the dynamics of which has been treated by Pollak and Geballe (1961) and Pollak (1964, 1965). The trapping equations, on the other hand, are dependent only on diffusion, drift and trapping. Exclusion of the dipole processes from the trapping equations, whose solutions are also frequency independent, and the exact equivalence between the trapping equations and simple random walks (characterized by a single ψ(t)), rules out the possibility of explaining the frequency dependence of a.c. conduction via simple random walks. This result is consistent with Tunaley (1974) but inconsistent with Scher and Lax (1973 a, b) (SL) and Lax and Scher (1977). The basic reason for this inconsistency is traced to the facts that SL treated all localized states equally and the emission events from them as being statistically independent, neither of which are valid in the view of the author. Application of the formalism to impurity hopping also leads to the result that hopping random distances between states of the same energy cannot account for the anomalous dispersion of transit times. The reason is that traps (slow states) isoenergetic with the transport states can never capture enough particles to obscure the transit of untrapped particles. The uniqueness of ψ(t) constructed from rato constants combined with the principle of detailed balance makes it possible to determine trap distributions relative to the concentration of transport states by fitting transient current data. The unique role of traps in producing dispersive transport also de-emphasizes the importance of long-range disorder (LRD) as a dispersive agent. The primary trap sites are believed to originate at point defects, while LRD plays a secondary role of modulating the capturability of traps.
No takes yet. Share an insight, caveat, or question.
F.W. Schmidlin (1980) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: