Excited states in poly-diacetylene chains: A Density-matrix-renormalization-group study

TL;DR

This study uses the density-matrix renormalization group method to analyze excited states in poly-diacetylene chains, providing detailed theoretical insights that align with experimental data.

Contribution

It applies DMRG to the Peierls--Hubbard-Ohno model for poly-diacetylene chains, accurately predicting excitation energies and identifying dark in-gap states, with a comprehensive parameter fitting approach.

Findings

Accurate calculation of ground state and excitations for chains up to 102 sites.

Identification of dark in-gap states consistent with experimental observations.

Determination of a coherent parameter set fitting experimental gap energies.

Abstract

We study theoretically poly-diacetylene chains diluted in their monomer matrix. We employ the density-matrix renormalization group method (DMRG) on finite chains to calculate the ground state and low-lying excitations of the corresponding Peierls--Hubbard-Ohno Hamiltonian which is characterized by the electron transfer amplitude t0 between nearest neighbors, by the electron-phonon coupling constant \alpha, by the Hubbard interaction U, and by the long-range interaction V. We treat the lattice relaxation in the adiabatic limit, i.e., we calculate the polaronic lattice distortions for each excited state. Using chains with up to 102 lattice sites, we can safely perform the extrapolation to the thermodynamic limit for the ground-state energy and conformation, the single-particle gap, and the energies of the singlet exciton, the triplet ground state, and the optical excitation of the triplet ground state. The corresponding gaps are known with high precision from experiment. We determine a coherent parameter set (t0*=2.4 eV, \alpha*=3.4 eV/\AA, U*=6 eV, V*=3 eV) from a fit of the experimental gap energies to the theoretical values which we obtain for 81 parameter points in the four dimensional search space (t0, \alpha, U, V). We identify dark in-gap states in the singlet and triplet sectors as seen in experiment. Using a fairly stiff spring constant, the length of our unit cell is about one percent larger than its experimental value.