Article pubs.acs.org/JPCA
Photophysics of Diphenylacetylene: Light from the “Dark State” Jack Saltiel* and V. K. Ratheesh Kumar Department of Chemistry and Biochemistry, Florida State University, Tallahassee, Florida 32306-4390, United States S Supporting Information *
ABSTRACT: A weak band at the tail of the known tolane (diphenylacetylene, DPA) fluorescence spectrum in several solvents is assigned to the forbidden 11Au → 11Ag transition on the basis of its lifetime (∼200 ps) and its fluorescence excitation spectra. The 11Au state, generally called the dark state, is not truly dark. We report the temperature (T) dependence of DPA fluorescence quantum yields (ϕf) in methylcyclohexane (MCH) solution and the fluorescence and phosphorescence quantum yields of DPA in glassy MCH at 77 K. Significant differences between fluorescence and phosphorescence excitation spectra reveal that, in addition to the 11B1u ← 11Ag transition, the first DPA absorption band includes a transition to another excited state, most probably the 11B2u state, from which intersystem crossing is more efficient. The T dependence of ϕf values in MCH solution is shown to be consistent with the previously reported T dependence of the lifetimes of transient DPA singlet excited state absorptions in the picosecond time scale. Transient absorption decay rate constants in hexane, methylcyclohexane and decalin as a function of T are retreated. Application of the medium enhanced barrier model shows that the medium is fully engaged with the molecular motion that is involved in the activated nonradiative decay path of the 11B1u state. In accord with theoretical calculations and experimental observations, that process is assigned to the diabatic internal conversion of the short-lived linear fluorescent π,π* (11B1u) state, over a low intrinsic energy barrier, to the longer lived weakly fluorescent trans-bent π,σ* (11Au), which is the precursor of the DPA triplet state. Absorption and fluorescence measurements in several solvents show that the 11B1u−11Ag energy gap decreases linearly with increasing medium polarizability. Our results allow a more definitive state order assignment for DPA.
■
INTRODUCTION Current interest in DPA stems from its use as a rod-like energy and electron conducting unit in conjugated polymers,1−3 dendrimers,4−6 and molecular photonic devices.7,8 Moreover, photoexcitation of derivatives of DPA leads to efficient doublestrand DNA cleavage and cancer cell apoptosis.9
and dubbed the dark state, was assigned to S1, and the longlived transient (500 ns in N2-outgassed acetonitrile based on time-resolved resonance Raman spectroscopy22) was assigned to T1.14 Study of the T dependence of the transient decay kinetics in several solvents led to the conclusion that the rate constant of nonradiative decay of the fluorescent state S2 to the longer lived “dark” state, although subject to a small energy barrier, is not significantly sensitive to changes in medium viscosity.14 It is generally agreed that the fluorescence corresponds to the strongly allowed 11B1u → 11Ag transition that also gives rise to the lowest energy transition, 11B1u ← 11Ag, in the absorption spectrum of DPA in solution. The mirror image relationship between absorption and fluorescence led others to consider that emitting state to be S1.12 Interestingly, the fluorescence excitation spectrum in solution is coincident with the DPA absorption spectrum for the lowest energy vibronic band, but loses intensity, strongly deviating from the absorption spectrum, at higher energies.14 Views vary concerning the geometry of the 200 ps excited singlet state,12,15−18 its relative S1/S2 state order, and its assignment in terms of state symmetry.
Photophysical results on DPA through 1999 and their interpretation were reviewed by Hirata.10 Fluorescence10−14 and transient absorption measurements14 on DPA solutions coupled with theoretical calculations,12,15−18 and fluorescence measurements in the gas phase under jet-cooled conditions19,20 and in inert gas matrixes at low T21 indicate that DPA exhibits unusual excited state behavior. The picosecond time-resolved transient absorption measurements14 are especially revealing. Immediately after the 295 nm dye laser excitation pulse, an absorption band appears with λmax at 500 nm whose decay time, τS ≈ 8 ps, at room temperature, corresponds to the fluorescence lifetime and to the rise time of absorption bands with λmax at 435 and 700 nm.14 Decay of the 435 and 700 nm transient absorptions occurs with a lifetime of ∼200 ps, τL, as a much longer-lived absorption appears at 415 nm. The 8 ps fluorescent state with transient absorption at 500 nm was assigned to S2, the 200 ps transient, considered nonfluorescent © 2012 American Chemical Society
Received: August 8, 2012 Revised: September 11, 2012 Published: October 9, 2012 10548
dx.doi.org/10.1021/jp307896c | J. Phys. Chem. A 2012, 116, 10548−10558
The Journal of Physical Chemistry A
Article
Figure 1. (a) DPA absorption (black), fluorescence excitation (λem = 323 nm, blue), and fluorescence (λexc = 293 nm, red) spectra in MCH at 25 °C, normalized to the intensity of the first vibronic band; (b) absorption and (c) fluorescence spectra of DPA in T (black), MCH (red), IP (blue), and PFH (green) (λexc = 293 nm).
and emission slits. Suprasil fluorescence quartz cells (1.00 cm2) were mounted in a thermostatted cell holder, and emission was recorded at right angle to the excitation beam. For solution measurements, excitation and emission slit widths were set at 5 and 2.5 nm, respectively. Temperature control was maintained as described above for the absorption measurements. Fluorescence and phosphorescence spectra at 77 K were recorded with samples in 2 mm i.d. Suprasil quartz tubes immersed in liquid N2 using the phosphorescence accessory Dewar of the Hitachi F-4500 spectrophotometer. The sample compartment was continuously purged with dry N2 or air to avoid interference by moisture condensation on the Dewar. Emission spectra were baseline corrected using the solvent as reference and corrected for nonlinearity of instrumental response. Fluorescence lifetimes were determined using a Horiba Fluoromax 4 fluorometer equipped with a time-correlated single-photon counting accessory and a R928 PMT detector (Hamamatsu). The light source was a 295 nm nanoLED (Horiba) having a pulse duration of 99% spectrophotometric grade) were used as received. Perfluorohexanes (Alfa Aesar 98+ %) were refluxed over KMnO4 for 24 h and distilled; the fraction boiling at 53−54 °C was collected and passed through a short plug of silica gel prior to use. Quinine sulfate (Matheson, Coleman & Bell) was recrystallized twice from deionized water and air-dried. Measurements. UV absorption spectra were recorded with a Cary 300B UV−vis spectrophotometer. Absorption spectra were measured as a function of T using a thermostatted cell holder for standard 1 cm quartz cells. Temperature control was provided by a Neslab RTE-4DD heating/cooling bath circulating water. Temperatures were constant to within ±0.2 °C, as measured with an Omega Engineering Model 199P1 RTD type digital thermometer. Fluorescence measurements in solution were recorded with a Hitachi F-4500 spectrophotometer equipped with a 150 W Xe arc source. The Hitachi F-4500 employs horizontal excitation
■
RESULTS Medium Effect on DPA Absorption and Fluorescence Spectra at RT. Room temperature (RT) DPA absorption, fluorescence, and fluorescence excitation spectra were measured in toluene (T), MCH, isopentane (IP), and perfluorohexanes (PFH). Solvent selection was intended to maximize the range of solvent polarizability. A stock solution of DPA was prepared in MCH, and 10 μL aliquots were diluted into 5 mL of MCH, IP, and PFH; 1.00 cm path length standard quartz cuvettes were used for spectral measurements. To reduce solvent background interference in the case of T, 100 μL of the stock solution was employed, and the absorption spectrum was measured in a quartz cuvette with 1.00 mm path length. Figure 1a shows normalized absorption, fluorescence, and fluorescence excitation spectra in MCH, and panels b and c show the solvent effect on absorption and fluorescence spectra, respectively.
10549
dx.doi.org/10.1021/jp307896c | J. Phys. Chem. A 2012, 116, 10548−10558
The Journal of Physical Chemistry A
Article
Figure 2. (a) DPA fluorescence spectra in T (black), MCH (red), IP (blue), and PFH (green) (λexc the first minimum in each absorption spectrum; areas are proportional to quantum yields), (b) fluorescence excitation (λem = 323 nm, black, 450 nm, red) in MCH at 25 °C, normalized to the same area, and (c) difference spectra showing the long λ fluorescence band; colors are the same as those in panel a.
Emission spectra were recorded for solutions at 25.0 °C, and absorption spectra were measured at ambient T (22.4−23.4 °C). The Beer−Lambert plot for absorbance at the first λmax in MCH yielded ε = 2.34 × 104 M−1 cm−1. The absorption spectra in Figure 1b are drawn relative to that value. As has been common practice, the fluorescence spectra in Figure 1a,c are forced to reach baseline at about 400 nm. However, an identical weak fluorescence band persists at longer wavelengths independent of the method used to purify DPA (recrystallization, sublimation, and zone refining). Full fluorescence spectra are shown in Figure 2a and fluorescence excitation spectra in MCH obtained by monitoring emission at 323 and 450 nm are compared in Figure 2b. Subtraction of the scaled long wavelength emission in PFH, Figure 2c, from the spectra in Figure 2a gives the spectra in Figure 1c. DPA Fluorescence Quantum Yields in Solution. Quinine bisulfate (QBS) in 1.0 N H2SO4 was used as the fluorescence standard.25−29 Because of the low DPA ϕf values, relatively high matched absorbances (A) of 0.251 ([QBS] = 2.83 × 10−4 M) at 268 nm and 0.222 ([QBS] = 1.04 × 10−4 M) at 293 nm were used. The spectra were corrected for nonlinearity in instrumental response. Corrections applied to the QBS ϕf = 0.546 value at 25 °C with the use of the Stern− Volmer constant of 14.5 M−1 for QBS self-quenching25 were almost negligible, yielding 0.544 and 0.545 values for 268 and 293 nm, respectively. The ϕf of QBS is known to be independent of λexc.29 Application of n2 index of refraction corrections30,31 yielded DPA ϕf values of 0.00284 ± 0.00004 and 0.00530 ± 0.00006 at 268 and 293 nm, respectively, at 25 °C, independent of whether the solutions were open to air or outgassed with Ar or O2. The T effect on ϕf was measured relative to the 25 °C values, Table 1. The ϕf values are
corrected for the change in DPA absorbance with T due to MCH density variation32 and spectral changes, and for the change in n2 values, see Tables 1S and 2S and Figures 1S and 2S in the Supporting Information. Fluorescence quantum yields for the short (11B1u → 11Ag) and long (11Au → 11Ag) wavelength DPA emissions in other solvents were measured at 25 °C for λexc = 293 nm relative to the value in MCH, Table 2. Table 2. Solvent Effect on DPA Fluorescence Quantum Yields, 25.0 °C, λexc = 293 nma solvent T MCH IP PFH
103 ϕfBb
103 ϕfBc
103 ϕfAd
103 ϕfAe
f vcf
280.35 288.15 293.15 298.15 323.15 343.15
3.66(8) 3.21(6) 3.01(7) 2.84(2) 2.02(6) 1.72(2)
6.99(10) 6.15(7) 5.67(6) 5.30(5) 3.58(5) 2.70(3)
0.65(2) 0.66(4) 0.70(4) 0.75(3) 0.78(2) 0.79(2)
0.51(6) 0.56(4) 0.63(7) 0.69(7) 0.74(10) 0.90(8)
0.52 0.52 0.53 0.54 0.56 0.64
4.57 5.30 4.44 1.96
(0.04) (0.06)c (0.08) (0.02)
103 ϕfA 0.66 0.75 0.60 0.35
(0.01) (0.02) (0.01) (0.01)
τfA (ps)
χ2
230 (9) 199 (2) 201 (3) 210
1.18−1.31 1.10−1.23 1.34−1.53 1.74
2.9 3.7 3.0 1.7
a
Values in parentheses are uncertainties based on reproducibility. Minimum values obtained from the ϕfA/τfA ratios in columns 3 and 4. c Determined independently and used as standard (see text). b
DPA Fluorescence Lifetimes. Air saturated DPA solutions in MCH (6.65 × 10−5 M), IP (5.51 × 10−5 M), T (7.91 × 10−5 M), and PFH (2.75 × 10−4 M) at 23.4 °C were excited with a 295 nm,
