Triplet-exciton annihilation and triplet spin relaxation in molecular crystals

The anisotropy of the delayed fluorescence in a magnetic field of most molecular crystals differs significantly from the well-known anisotropy of anthracene. So far it was assumed that Suna's theory, which explains the behavior in anthracene very well, is not able to describe the anisotropy in these other molecular crystals. We have measured the delayed fluorescence anisotropy of naphthalene and $p$-terphenyl in a high magnetic field at very low triplet density. For the first time a thorough comparison of the experimental results and Suna's theory has been made in a molecular crystal other than anthracene. We demonstrate that, contrary to expectations, a good fit to the data for naphthalene can be achieved, if only spin relaxation is properly taken into account. This leads to a discussion of the theory and experiments dealing with the ESR linewidths of paramagnetic triplet excitons in molecular crystals. It seems that for naphthalene the theory by Reineker for the ESR linewidths, which is based on the Haken-Strobl model, does not fully agree with the experimental findings. By respecting spin relaxation the fit of Suna's theory yields a few important kinetic parameters of triplet excitons in naphthalene such as the out-of-plane hopping rate and the nearest-neighbor annihilation rate.