Predicted observable fluorescent lifetimes of ... - ACS Publications

Department of Chemistry, Montana State University, Bozeman, Montana 59715 (Received January 2, 1974). Predicted observable fluorescent lifetimes of si...
1 downloads 0 Views 263KB Size
Nancy J . L. Roth and Arnold C,Craig

It5

bservable Fluorescent Lifetimes of Several Cyanines aney J. b. Roth and Arnold C. Craig” Department

ofChemistry, Montana State University, Boreman, Montana 59715 (Received January 2, 1974)

Predicted observable fluorescent lifetimes of six cyanine dyes have been obtained from intrinsic fluorescent lifetime and quantum yield measurements. The predicted lifetimes vary by nearly three orders of magnitude, from 1 nsec to 1.7 psec, and appear to be more sensitive to chain length variation than heteroatom differences. In particular the shortest chain cyanines appear to have a facile nonradiative pathway from the excited singlet.

Tpred

The determination d fluorescence lifetimes is critical to the understanding of the “energy-handling” properties of excited states. Although there have been recent advances in measuring short lifetimes by pulse distortion,l single. photon counting,2 anal picosecond pulse laser^,^ they are not readily applicable to all compounds. It would be desirable to be able to determine indirectly, but with a good degree of confidence, observable fluorescent lifetimes. Such determinations can be made if the intrinsic fluorescent lifetimet; and fluorescent quantum yields are measured with reasonable accuracy. Because cyanine dyes are of current interest as photographic sensitizer~,~ dye lasers, and mode locks for lasers3 and can be readily obtained in various structural modifications with respect, for example, to chain length, heterocyclic nuclei, and ring and chain s u b ~ t i t u t i o n ,we ~ proposed to obtain predicted observable fluorescent lifetimes for several serie4 of cyanines. The results for the first two such series are reported here. The lifetime of dye 3 has been measured by two photon absorption and subsequent fluorescence decay analysis as 1.6 f 0.1 n ~ e c .Even ~ though the experimental conditions were not identical with the present work the agreement is rather good. Another group has estimated the observable fluorescent lifetime of 3 ais 1.24 nsec Lasing a similar analysk6 Their data for b, (0.49) and T~~~~ (2.54 nsec) compare favorably with our data.

Experimental Soetion The dyes 1-6 were prepared by standard methods4 and were analytically pure. Determination of absorption spectra were macle an a Cary Model 14 spectrophotometer. Fluorescent spectra and quantum yields were determined on a custom apparatus utilizing a tungsten source, two Bausch & Lomh 500-mm monochromators, and an EM1 9558Q photonnultipher tube.? Predicted observable lifetimes ( T ) were ~ calculated ~ ~ ~ from intrinsic fluorescent lifetimes ( ~ , ~and t ~fluorescent ) quantum yields (6)by The Journal ofPhysical Chemistry, Vol. 78, No. 12, 1974

Tint,$

Intrinsic lifetimes were calculated from absorption and fluorescent parameters by the Birks and Dyson modification of the Strickler-Berg relationship.8,9 l h i n t r = 2.880X10-9n,,fi3/n~,,.

x

Fluorescent quantum yields were determined by comparison with known standards,lO by methods generally described by Parker.ll The fluorescent standards used were quinine bisulfate, 3-aminophthalimide, 3-nitro-N,N-dimethylaniline,4-dimethylamino-4’-nitrostilbene, and Rhodamine B. The cyanine dyes were dissolved in absolute methanol; 10-6 to 10-5 M for absorption and fluorescence spectra and 10- to 10- M for quantum yield measurements. All measurements were carried out a t room temperature. Action spectra for fluorescence of the cyanine dyes indicated the main species was responsible for emission.

Results and Discussion The predicted observable fluorescent lifetimes of several cyanine dyes are presented in Table I. We believe they represent a realistic assessment of the observable fluorescent lifetimes of these dyes. The data show differences in ‘Tpred of almost three orders of magnitude. The most striking observation is that the shortest chain length compounds, both the monomethine oxygen and sulfur cyanines, 1 and 4,have extremely short predicted lifetimes, 6.4 and 1.7 psec, respectively. A facile nonradiative pathway from excited singlet must exist for these two dyes. That the counterion, iodide, is not providing that pathway i s supported by the high (0.43

TABLE I: Luminescence Properties of Six Cyanine Dyes Dse

1 2 3 4 5 6

Q

0.0037 0.053 0.43 0.00059 0.048 0.33

rintr(l09, sec

1.72 2.02 2.37 2.94 3.09 2.87

rpred(in’O),

sec

0.064 1.1 10 0.017 1. . 5 9.5

Theory of

Saturation and Double Resonance

in Esr Spectra

and 0.33) fluorescent quantum yields of the longest chain pair, 3 and 6. The somewhat longer intrinsic lifetimes of the sulfur compounds ,are almost compensated by their lower fluorescent quantum yields SQ that the oxygen and sulfur compounds of comparable chain length have nearly the same predicted lifetimes. Since t:hese predicted observable lifetimes were determinedl in dilute solutions without the presence of obvious quenchers or energy receptors and since the composite rate of all decay processes from the excited state is the inverse of the observable lifetime, a minimum may be placed on the rate of energy transfer from the first excited sing1e.t state of these dyes to quenchers or acceptor molecules. 1nvest.igzitionof the energy transfer rates .to acceptors in dilute solutioins is a logical step which we plan to investigate.

References and Notes ( I ) K, Osada, Rev. Sei. instrum.. 44,656 (1973). (2) (a) R. Schuyier. I . iaenberg, and R D. Dyson, Photochem. Photo-

1155 biol., 15, 395 (1972);(b) W. R. Ware, L. J . Doemeny, and T. L. Nemzek, J . Phys. Chem., 77, 2038 (1973):(c) C. Lewis, W. R . Ware, L. J. Doemeny, and T. L. Nemzek, Rev. Scl. Insfrum., 44,

107 (1973). (3) (a) P. M. Rentzepis, Advan. Chem. Phys., 23, 189 (1973);(b) T. A. Erdmann, H. Figger, and H. Walther, Opt. Commun., 6, 166 (1972). (4) L. G. S. Brooker in "The Theory of the Photographic Process," 3rd

ed, C. E. K. Mees and T. H. James, Ed:, MacMlllan, New York, N. Y., 1966,Chapter 11 and references cited therein. (5) H . Cirkel, L. Ringwelski, and F. P. Schaeffer, Z. Phys. Chem. (Frankfurt am Main), 81,158 (1972). (6) D. N. Dempster, T. Morrow, R. Rankin, and G. F Thompson, J . Chem. Soc., Faraday Trans. 2, 68,1479 (1972). (7) The authors wish to thank Dr. Patrik R . Callis for use of hi$ equipment and for helpful discussions. (8) (a) S. J. Strickler and R. A . Berg, J. Chem. Phys., 37, 814 (1962); (b) J. B. Birks and D. J. Dyson, Proc. Roy. SOC.,Ser. A, 275, 135

(1963). (9)The cyanines meet very nicely. the criterion established by Strickier and Bergaa for those compounds for which tile equation would give good predictions. (IO) (a) E. Lippert, W. Nageie, I. Seibold-Biankenstein, U. Staiger, and W . Voss, Z. Anal. Chem., 170, 1 (1954);(b) R. J. Argauer and C . E: White, Anal. Chem., 36, 368 (1964);(c) R. J. Argauer, Ph.D. Thesis, University of Maryland, June 1963,pp 12-42. (11) C . A. Parker, "Photoluminescence of Solutions," Elsevier, New York, N. Y., 1968,pp252-261.

Theory off Saturation and Double Resonance in Electron Spin Resonance Spectra. V I . Saturation Recoveryq Jack H. Freed Department of Chemistry, Cornel; University, lthaca, New York 74850 (Received December 27, 7973) Putriication costs assisted by the Petroleum Research Fund

The general theory of Freed for steady-state saturation and double resonance in esr spectra of free radicals is extended to cover time-dependent experiments. The solution is again found to depend on the same matrix representations developed in the earlier work. Particular attention is paid to saturation recovery in the light of recent such experiments. It is shown that while the general solutions yield sums of many exponential decays, the dominant observation may, to a first approximation in many cases, be described in terms of single T I = l/zW,, where We is the electron-spin flip rate, in agreement with recent observations. This T I is characteristically the slowest decay constant, and is either well separated from the much faster decays (due to nuclear spin flip, exchange, and/or reorientational effects), or may be difficult to distinguish from decay constants of comparable magnitude (due to the same type of effects). Both conventional saturation recovery and eldor-type recoveries are discussed from this point of view. The general approach given is equally adaptable to cases of esr spectra in the motional narrowing region as well as esr spectra characteristic of slow tumbling. Both cases are discussed in detail with several examples given. In particular, in the slow tumbling region for the case of a radical with hyperfine structure (eg., a nitroxide), it is shown that, in general, both direct reorientational effects as well as nuclear spin flip processes contribute directly to the relaxation modes. The analysis given emphasizes analytic aspects although the general expressions, appropriate for accurate computer simulation, are given.

I. Introductioni Recently, there bas been growing interest in pulsed esr experiments on free radicals in liquids, in particular, saturation recovery-type experiments.2 The relaxation behavior of such systems in, in general, rather complex, and has

been the subject of a series of papers (I-V) analyzing steady-state saturation and double resonance b e h a ~ i o r . ~ - ~ A review of some of these aspects has recently been given.1° In the future, one may expect to see an increase in importance of pulsed techniques, so it was deemed apThe Journalof Physical Chemisfry, Val. 78. No. 12, 7974