4380
J. Phys. Chem. 1983, 87, 4360-4361
hydrated material the cation forms a cluster with water molecules. On dehydration the cation presumably comes nearer to aluminum. This causes a change in the distribution of charges around aluminum and possibly a distortion of the aluminum-oxygen tetrahedra. Both effects, the change in charge distribution and distortion of the tetrahedron, can give rise to an increase electric field gradient at the aluminum nucleus. However, from the observation that the increase in quadrupole interaction on dehydration is stronger for Na-ZSM-5 than for H-ZSM-5, we tend to believe that the distortion of the tetrahedra is more important for the increased electric field gradient than the redistribution of charges, although of course the latter effect may induce the local distortions. In the case of dehydrated H-ZSM-5 it is often indicated that the H+ ion exists in a Al-O-H+ band. For a perfect tetrahedron with an Al-0 distance of 1.59 A and an 0-H distance of 1 A, the dipolar interaction between Al and H would give rise to a broadening of the Al NMFt line of 2300
Hz. Such a broadening, however, is not observed, suggesting either that the Al-H+ distance is much larger or that the H+ ion is mobile. Finally, we want to comment on the statement made earlier in this paragraph that in hydrated H-ZSM-5 the nAl line width is caused by a distribution of nAl chemical shifts. Preliminary 27Alspin-lattice relaxation (T,) experiments clearly reveal several species with distinct T1 values and distinct chemical shift. Further investigations are needed to establish the origin of these different 27Al sites which together form the line shapes discussed in this paper.
Acknowledgment. We thank DSM, Geleen, for providing the ZSM-5 samples and Prof. Dr. E. de Boer, Dr. G. P. M. van der Velden, Ir. P. Frenken, Ir. J. de Haan, and Dr. E. Duynstee for discussions. This work is supported by DSM Research, Geleen, The Netherlands. Registry No. Aluminum, 7429-90-5.
Exponential Fluorescence Decay from Single Rovibratlonal Levels of S1 Acetaldehyde Marcus Noble and Edward K. C. Lee' Department of Chemistry, Unlverslty of Califomla, Itvhe, Callfornla 92717 (Received: Ju& 19, 1983)
Single rotational level S1emission has been observed for jet-cooled acetaldehyde and a-monodeuterated acetaldehyde. All the levels studied (up to Evib = 1760 cm-') yielded single exponential decays, the vibronic origin being the longest-lived levels (CHSCHO, 220 ns;CHSCDO, 1.52 ps). The data are correlated with previous room temperature studies.
Introduction Acetaldehyde is a molecule of atmospheric significance' as well as a possible model for the photochemical and kinetic behavior of larger aldehydes.2 Because it is a seven-atom molecule with 15 vibrational degrees of freedom and small rotational constants (A" = 1.8870, B" = 0.3387, C"= 0.3033 cm-')? the room temperature electronic spectrum is too congested to obtain detailed rovibrational information. In spite of this congestion, the weak nature of the S1 Sotransition, and the less than unity quantum yield for fluorescence, several attempts have been made to study its S1 dynamics by absorption and emission spectroscopic measurement^.^*^,^ A recently reported experiment by Speiser et al. is the closest to a single vibrational level collision-freestudy! In this experiment an unnarrowed frequency doubled dye laser (fwhm > 1 cm-') was used to excite low-pressure (2 25 mtorr) acetaldehyde vapor at 320 nm. This wavelength was arbitrarily chosen to lie below the predissociative threshold (131250 ~ m - ' ) . ~These authors reported the finst observance of biexponential decay for S1acetaldehyde +
-
(at 25 mtorr: Tfast 29 ne, 7,lOw94 ns) and estimated the fluorescence quantum yield as 7.5 X whereas a prior study2in our laboratory found the longest lifetime at 10 torr to be 3.7 ns at 325 nm. The biexponential decay observed by Speiser et al. was found to be pressure dependent: the slow component was preferentially quenched upto 400 mtorr (self-quenching rate coefficient: 6.7 X lo7 torr-' s-') at which pressure the decay appeared single exponential. Above 400 mtorr the decay continued to be self-quenched at the same rate. The authors attributed this overall behavior to the process of rapid reversible intersystem crossing followed by internal conversion. This is a model which has been introduced and used to explain the same behavior in a-dicarbonyls by Kommandeur and co-workers.8 As part of our overall studies into the dynamics of small molecules in electronically excited states, we have observed the laser-induced fluorescence decay from single rovibrational levels of S1 acetaldehyde and a-monodeuterated acetaldehyde. The assignment of these levels was known from our acetaldehyde (CH,CDO) rovibronic analysis of the ?r* n transition of acetaldehyde and its deuterated derivative~.~J~ The fluorescence lifetimes of these levels show interesting mode specificity and energy dependence which will be reported in detail elsewhere.'O We report +
(1)See, for an example, B. J. Finlayson and J. N. Pitts, Jr., Science, 192,111 (1976). (2)D. A. Hansen and E. K. C. Lee, J.Chem. Phys., 63,3272(1975). (3) R. W. Kilb, C. C. Lin, and E. B. Wilson, J.Chem. Phys., 26, 1695 (1957). (4)C. S.Parmenter and W. A. Noyes, Jr., J.Am. Chem. SOC.,85,416 (1963). (5) A. Gandini and P. A. Hackett, Chem. Phys. Lett., 52,107(1977). (6) S. Speiser, W. F. Pfeiffer, and G. H. Atkinson, Chem. Phys. Lett., 93,480 (1982). (7) R. J. Gill and G. H. Atkinson, Chem. Phys. Lett., 64,426 (1979). 0022-365418312087-4360$01.5010
(8)(a) R. van der Werf, E. Schutten, and J. Kommandeur, Chem. Phys., 11, 281 (1975). (b) R.van der Werf and J. Kommandeur, ibid.,
16, 125 (1976). (9)M.Noble, E.C. Apel, and E. K. C. Lee, J. Chem. Phys., 78,2219 (1983). (10)M. Noble and E. K. C. Lee, results on the spectroscopy and dynamics of CHSCHO and deuterated molecules to be published.
0 1983 American Chemical Society
The Journal of Physical Chemlstty, Vol. 87, No. 22, 1983
Letters
lo3
-
h
In
c
.-
C
3
n 102Y
c
.-C
10°1
;
I
I
I
200
400
600
I 800
t ( nsec 1 Figure 1. Plots of fluorescence decay tom CHSCHOlevels of pattiwlar signlflcance: (a) 0' (origin), l,, (J'K;K;); (b) 10'14'15' (Ai), lii: (c) unknown vibration at 31 232 cm-', llo.
here our observation that all of the single rovibrational levels under study decayed with single exponential decay behavior, and show how these results can be reconciled with the room temperature "bulb" study.6
Experimental Section The free jet expansion apparatus has been described elsewheres and only the briefest details of the specifics of this experiment are given here. Premixed CHBCHOor CH3CD0 in helium (-300 ppm) was expanded through a 7bpm-diameter nozzle at 19 atm. The frequency doubled, dye laser beam (fwhm: 0.08 cm-', 7 ns) crossed the jet 5-6-mm downstream where the rotational temperature was -1 K. This temperature permitted excitation of several rotational levels but with minimal overlap of lines. The resultant fluorescence was focussed onto a fast photomultiplier (RCA 8575) through a laser blocking filter (Corning 4-96) without spatial filtering. The photomultiplier signal was processed with a 100-MHz waveform recorder (Biomation 8100) and signal averager system (Tracor Northern 1500/Nova-3). Up to 2500 laser pulses were summed and a corresponding number of background signals collected. The decay curves shown in Figure 1are semilog plots of the summed signal minus summed baseline.
Results and Discussion As stated in the Introduction, all of the S1CH3CH0 rovibronic levels studied (with excess vibrational energies up to 1770 cm-')showed single exponential decay behavior. The lifetimes were in the range 95-230 ns and are considered accurate to *lo%. Figure 1shows typical decay plots of several levels. The highest assigned energy level pumped was the l,, (JLdK;)level of the 101141154(A,) vibration at 30 749 cm-' (or E,& = 978 cm-'). This has a lifetime of 131ns vs. the 220 ns observed for the l,, level of the S1vibronic origin at 29771 cm-'. If we assume a radiative lifetime of 5 p s , these represent fluorescence
-
4361
quantum yields of 0.026 and 0.044, respectively. A value of 5 p s seems a reasonable, rough estimate considering we have seen single exponential decay from low-lying CH3CDO levels with lifetimes up to 1.52 p s and previous estimates of the radiative lifetime are in the range of 2.2-12 p ~ The . ~variation of CH3CD0 s1lifetimes with E.& indicates that even the low-lying excited-state levels of CH3CD0 undergo nonradiative transitions.1° Therefore, 1.5 p s merely fixes a lower limit on the radiative lifetime. There are several vibronic bands in the jet spectrum at 320 nm (the wavelength used by Speiser et a1.% We arbitrarily chose the llolevel of the moderately strong, unassigned vibration at 31 232 cm-I for comparison. The fluorescence decay from this level followed a single exponential with Tf = 98 ns. This is almost identical with the "slow" component of the room temperature biexponential decay (94 f 2 nsh6 The fact that all of the decays are single exponential along with the agreement in lifetime between the two results recorded at the same wavelength strongly suggests an alternative explanation for the room temperature results. At room temperature, levels with much higher J and/or K and with vibrational energy upto a few hundred wavenumbers higher than this could be populated by 320-nm radiation. It seems highly probable that Speiser et al. excited several rovibronic levels and the "fast" and "slow" components of their observed decays show the presence of several different single level exponential decays. The observation that the quenching of the shorter-lived component is not significant below 400 mtorr is therefore merely a reflection of the shorter lifetime of that particular rovibronic level. To throw further light on the "fast" (29 ns) decay, we studied various levels up to 31 530 cm-'. All decayed with a single exponential having lifetimes 2 95 ns. This suggests that the short-lived level(s) are higher angular momentum and/or free internal rotor levels neither of which can be accessed in the jet due to rotational and torsional level populations at 1K as well as selection rules. The ongoing study shows that there is no implicit rotational dependence for the internal conversion process but higher rotational levels may be shorter lived due to an increased density of available states. We also have tentative evidence that levels above the 650-cm-l potential barrier for methyl rotation (d16)may be shorter lived than those below.1° In conclusion we have shown that low-lying (Edb I1750 cm-') levels of the S1state of acetaldehyde undergo slower internal conversion than previously estimated and do not undergo rapid reversible intersystem crossing. This behavior is similar to that of S1formaldehyde and firmly establishes that the lower region of the S1state of acetaldehyde belongs to the "small"/"intermediate" molecule class. Acknowledgment. This research has been supported by the National Science Foundation Grant CHE-82-17121. Registry No. CH,CHO, 75-07-0; CH,CDO, 4122-13-8.
