MULTIQUANTUM STATES DERIVED FROM DAVYDOVS VERTICAL-BAR-D-1] ANSATZ - II - AN EXACT SPECIAL CASE SOLUTION FOR THE SU-SCHRIEFFER-HEEGER HAMILTONIAN AND ITS RELATION TO THE VERTICAL-BAR-PHI(2)] STATE

Authors
Citation
W. Forner, MULTIQUANTUM STATES DERIVED FROM DAVYDOVS VERTICAL-BAR-D-1] ANSATZ - II - AN EXACT SPECIAL CASE SOLUTION FOR THE SU-SCHRIEFFER-HEEGER HAMILTONIAN AND ITS RELATION TO THE VERTICAL-BAR-PHI(2)] STATE, Journal of physics. Condensed matter, 10(12), 1998, pp. 2631-2662
Citations number
97
Categorie Soggetti
Physics, Condensed Matter
ISSN journal
09538984
Volume
10
Issue
12
Year of publication
1998
Pages
2631 - 2662
Database
ISI
SICI code
0953-8984(1998)10:12<2631:MSDFDV>2.0.ZU;2-5
Abstract
We present the derivation of an exact special case solution (for a cla ssical lattice) for the Su-Schrieffer-Heeger model for the calculation of soliton dynamics in trans-polyacetylene. Our solution is exact, in the sense that the ansatz state yields an exact solution provided tha t the equations of motion for its parameters are obeyed. However, thes e equations can be solved only numerically tin principle to any desire d accuracy), not analytically. The model is applied to time simulation s of neutral solitons as a function of temperature. We find agreement of the results of our time simulations with experimental data on the m obility of neutral solitons in the system. Comparative calculations us ing the completely adiabatic model indicate that the results of this m odel are at variance both with experiment and with those of our model. A simple consideration of the potential barriers for soliton displace ment leads to an overestimation of the soliton mobility for low temper atures and an underestimation for higher ones. In an appendix we discu ss in some detail the relationship of this exact solution with the \Ph i(2)] state ansatz as presented in our previous paper. We find that th e ansatz state and the exact solution yield identical results for latt ice momenta, displacements and site occupancies, but differ in a time dependent phase factor. Thus spectra computed with the dynamics result ing from the exact solution for the classical lattice on one hand and from the ansatz state on the other would differ from each other.