Excited singlet states of Molybdenum(II) Chloride Clusters

Compounds containing the molybdenum cluster anion $[Mo$₆X₁₄]^2- are known to show fast light-induced electron transfer reactions, which have made them candidates for specific photochemical reactions utilizing long-lived metastable states with powerful oxidizing properties. High quantum efficiencies for radiative emission have been reported. Theoretical calculations predicted the existence of several adjacent excited states, lying above the Fermi level. Long-lived luminescence, i.e. `phosphorescence', from an excited triplet state has been experimentally confirmed; its lifetime was determined to lie in the range of a few tenths of a millisecond. No emissions from excited singlet states have thus far been observed, although these transitions should be spin-allowed.

Figure 1: Time-resolved luminescence spectra of $[N(C$₄H₉)₄]₂Mo₆Cl₁₄. a) 40 to 200ps after laser pulse, b) 360 to 520ps after laser pulse, c) time integrated.

We have measured time-resolved photoluminescence on well-defined single crystal samples of $[N(C$₄H₉)₄]₂Mo₆Cl₁₄ with time-resolution down to the picosecond regime. This special technique has enabled us to detect the predicted excited states and to determine their energies and their lifetimes. Fig. 1 shows the luminescence spectra at room temperature for several time intervals after excitation with lambda = 532nm (pulse width 25ps). Three emission bands are observed, identified by their maxima in the luminescence spectra and designated with the conventional terminology: A: Maximum at $600\; +/-10\; nm$; B: Maximum at $630\; +/-3\; nm$; C: Maximum at $690\; +/-10\; nm$. Band C shows a pronounced red shift to 760 +/- 10 nm at temperatures below 100K, while the maxima of A and B do not change between 5 and 290K.

Figure 2: Arrangement of molybdenum octahedra in $[N(C$₄H₉)₄]₂Mo₆Cl₁₄; view along the crystallographic b-axis.

Emission bands A and B can only be detected with picosecond spectroscopy. Band C is negligible in the first picoseconds after the excitation but dominant in spectra detected with integration times of a microsecond or more. The evaluated lifetime of Band A ranges from $630\; +/-\; 30\; ps$ at room temperature to $1000\; +/-\; 100ps$ at 5K. The lifetime of band B is found to be $50\; +/-\; 30$ps.

Energy-transfer experiments have proven luminescence C to be an emission from the lowest excited triplet state of the cluster ion. The transition is spin-forbidden; consequently, the excited triplet state has a rather long lifetime. On the other hand, the newly discovered bands A and B must be due to spin-allowed transitions as they are visible on a picosecond time-scale, i.e. they must have large oscillator strengths. Band A can be identified as an emission from excited singlet states of the cluster ion. Both A and C are broad due to strong electron-phonon coupling, a feature well known from luminescence spectra of the coordination compounds of transition metal ions. In contrast, B is much narrower and there is no relaxation from A to B, indicating that B does not originate from transitions in the individual cluster units.

A remarkable feature is that the intensity of band B depends on the orientation of the single crystal relative to the incident beam. The intensity change cannot be due to surface effects as the penetration depth at our excitation wavelength is a few tenths of a millimeter and no symmetric dependence on the angle between surface and incident beam is observed. The intensity of B, relative to emission A, shows a maximum if the exciting beam is coincident with the crystallographic b-axis, along which the cluster units align as shown in Fig. 2. A possible explanation for this effect is a long-range interaction in the crystalline state. A theoretical model for this phenomenon has yet to be developed.

(N.Perchenek, U.Strauß, W.W.Rühle, H.J.Queisser, A.Simon)

From the yearbook of the institute ("Wissenschaftlicher Tätigkeitsbericht") 1991

webmaster