<t>p branching ratio

8.5 x 10-4"

1.3 x IO-3"

<x(S„) (cm2) <r(S,) (cm2)

1.57 x 10~17'

9.22 x 10"18'

a(T,) (cm2)

2.87 x 10"18'

«moi(S,)@920 nm (M-'cm"1)



amoi(^i)@747 nm (M-'cm"1)



5, lifetime (ns)



T, lifetime (ms)



"[13.30]; 6 [13.38]; c[13.39]; d[13.40]; c[13.41]; /[13.37]; «[13.42]; A[13.31); '[13.43]; >[13.44]; *[13.13]; '[13.45],

"[13.30]; 6 [13.38]; c[13.39]; d[13.40]; c[13.41]; /[13.37]; «[13.42]; A[13.31); '[13.43]; >[13.44]; *[13.13]; '[13.45], excited states. In experimental determinations of the decay times for the various excited states, it is important to take into account the very large decrease in the measured decay times of excited states caused by rapid relaxation via impurities, especially oxygen.

In Fig. 13.8 we show the transient absorption spectra for C60 and C70 in toluene solution obtained for different delay times between the pump and probe pulses [13.41]. In these experiments, the "pump" laser emitted 0.8 ps pulses at 587 nm (0.3 mJ/pulse at 20 Hz), and the "probe" pulse continuum was generated by focusing the same pump pulse into water. The arrows in Fig. 13.8(a) indicate whether an increase (t) or decrease (4-) of excited state absorption occurs with increasing time delay between the pump and probe pulses. Two peaks are apparent in the spectra; one at 920 nm (1.35 eV) and the other at the 747 nm (1.66 eV). As a function of pump-probe delay time, the 747 nm peak is seen to grow at the expense of the 920 nm peak, indicating that the 920 nm peak should be identified with 5] S„ absorption, since the decrease in absorption is consistent with the depopulation of the Sj state by the 5, -*■ Tx intersystem crossing. Accordingly, the 747 nm peak is identified with Tx Tn absorption. From fits to the data in Fig. 13.8(a) the lifetime of the state was found to be 1.3 ±0.2 ns, as obtained from the time evolution of the 920 nm peak above [13.37,41], in good agreement with other determinations of the lifetime of the 5! state [13.47-50], including single photon counting fluorescence decay studies yielding a lifetime of 1.17 ns for the state [13.48]. The Si lifetime should be approximately equal to the intersystem crossing time (which has been reported as 1.2 ns [13.37]), since the quantum efficiency for this process is nearly unity, and almost complete Sj -* 7\ transfer has been observed after a time of 3.1 ns [13.41], The energy separation between the Si and S2 levels [see Fig. 13.8(a)] has been reported by different groups as 1.35 eV [13.41] and as 1.40 eV [13.37], as summarized in Table 13.3. In this table we see that the peak molar absorption coefficient am0, for the triplet 7, —► T2 transition is found to be twice as large as for the Si -»■ S2 transition. Several groups have also assigned the 747 nm (1.66 eV) transient absorption feature to T„ transitions [13.37, 42, 48, 51] and transitions to higher-lying triplet and singlet states have also been reported (see Table 13.3). We return to a discussion of the transient spectroscopy of C60 in the solid state in §13.7.1.

Whereas there is general agreement that the lifetime of the singlet 5, state is 1.2-1.3 ns, and that the lifetime of the triplet Tx state is much longer (by several orders of magnitude) than that for Su a wide range of values has been quoted for the triplet lifetime. Analysis of Raman intensities for oxygen-free C60 thin films has indicated values of 55 ± 5 ms for the triplet lifetime, falling to 30 ± 5 ms for oxygenated C60 films [13.45], in

Was this article helpful?

0 0

Post a comment