Using one-photon excitation we studied photoluminescence of C60 saturated with molecular hydrogen over a temperature range from 10 to 230 K. Saturation of samples was done at 30 atm and at temperatures low enough (T < 250 °C) to exclude chemical sorption. The samples were saturated during periods of varied duration τ to reach different occupancy levels. To check the reliability of our luminescence results and their interpretation, our spectra for pure C60 were compared with data known in the art, demonstrating good compatibility. The luminescence spectra were attributed according to the approach of Akimoto and Kan’no by separating the total spectra into two components of different origin. The A-type spectra, associated with exciton transport to deep traps, become prevalent over the B-type emission above 70 K. The integrated intensity I as a function of the temperature T of the luminescence measurements I(T) remained at a constant level up to the orientational vitrification point of about 100 K when the saturation times exceeded a certain value (for one, 50 h for a saturation temperature of 200 °C); then I(T) went down rather steeply with increasing T. However, at longer τ the intensity I(Τ) persisted consistently to higher T (the higher, the longer τ) and then dropped with increasing T. This finding made us reexamine the lattice parameter vs. saturation time dependence for saturation temperatures of 200 and 230 °C. As a result, additional evidence allowed us to infer that after the completion of the single-molecule filling of O-voids (specifically, after roughly 50 h at Tsat = 200 °C) a slower process of double filling sets in. Double filling entails an anisotropic deformation of the octahedral cage, which modifies rotational dynamics more than single filling. Further, we argue that singlet exciton transport to traps (which is responsible for the A-type emission) can be crucially hampered by rotational jumps of one of the molecules, over which a travelling exciton is spread. Such jumps break coherence, and the exciton stops, thereby increasing the probability of emissionless deactivation. If so, then the temperature at which the rotational jumps occur sufficiently frequently may be by inference considered to be the unfreezing point for the orientational glass state (essentially coinciding with the inverse critical point Tg, where the rotational system freezes into the orientational glass). This treatment of Tg differs from that existing in the art, according to which the glass state is destroyed owing to the increased density of phonon states. Keeping to our reasoning, we conclude that the orientational glass state does not disappear but, instead, is conserved almost unchanged under one-molecule filling and persists to appreciably higher temperatures in the case of double filling, which has a stronger effect on exciton dynamics.
Read full abstract