Photoisomerization of all-trans-1,6-diphenyl-1,3,5-hexatriene. Temperature and deuterium isotope effects

Article abstract of DOI:10.1021/jp021540g

Irradiation of all-trans-1,6-diphenyl-1,3,5-hexatriene (ttt-DPH) in acetonitrile (AN) gives ctt- and tct-DPH by relatively inefficient pathways but mainly via the singlet excited state. Assuming that twisted singlet excited state intermediates partition e

Full text of DOI:10.1021/jp021540g

3178
J. Phys. Chem. A 2003, 107, 3178-3186
Photoisomerization of all-trans-1,6-Diphenyl-1,3,5-hexatriene. Temperature and Deuterium
Isotope Effects†
Jack Saltiel,* Govindarajan Krishnamoorthy, Zhennian Huang, Dong-Hoon Ko, and
Shujun Wang
Department of Chemistry and Biochemistry, Florida State UniVersity, Tallahassee, Florida 32306-4390
ReceiVed: July 1, 2002; In Final Form: October 19, 2002
Irradiation of all-trans-1,6-diphenyl-1,3,5-hexatriene (ttt-DPH) in acetonitrile (AN) gives ctt- and tct-DPH
by relatively inefficient pathways but mainly via the singlet excited state. Assuming that twisted singlet excited
state intermediates partition equally to cis and trans ground state double bonds leads to the conclusion that
the major nonradiative decay process of the singlet excited state of ttt-DPH (1ttt-DPH*) is direct decay to the
ttt-DPH ground state, φnr ) 0.54. If CH stretching vibrations serve as accepting modes in this decay process,
deuterium substitution should profoundly attenuate it. We report a comparative study of perhydro-ttt-DPH
with di- and tetradeuterated ttt-DPH (ttt-DPH-dn, n ) 0, 2, and 4) involving deuteration of one and both
terminal triene double bonds. NMR and HPLC analyses give kH/kD ) 1.36 ( 0.1 for terminal bond isomerization
at 20.0 °C. The very small changes in τf and φf are consistent with such a kinetic deuterium isotope effect on
1
the rate constant for terminal bond isomerization. The temperature dependencies of τf, φctt, and φtct for ttt-
DPH-d0* give energy barriers for torsional relaxations that are well above the 2Ag-1Bu energy gap. The
major radiationless decay process is not related to trans f cis photoisomerization, is barrierless within
experimental uncertainty, is completely insensitive to deuterium substitution, and occurs in the ns time scale.
The mechanistic implications of these results are discussed.
Introduction
from the dependencies of photoisomerization quantum yields
and fluorescence lifetimes on temperature. In addition, we report
the synthesis and comparative study of derivatives of ttt-DPH
(ttt-DPH-dn, n ) 2 and 4) involving deuteration of one and both
terminal triene double bonds. Surprisingly, in contrast to the
photoisomerization decay channels, the major radiationless
decay pathway of ttt-DPH is insensitive to d-substitution and
to temperature.
Our interest in all-trans-1,6-diphenyl-1,3,5-hexatriene (ttt-
DPH) stems from its use as a model for long polyenes because
it is the first member of the R,ω-diphenylpolyene vinylogous
family whose lowest excited state is the forbidden doubly excited
2Ag state.1-3 This state couples vibronically4 with its precursor
state, the initially formed 1Bu state, and the two mixed states
exist in rapidly established5,6,7 thermal equilibrium.8 A crucial
role for the 2Ag state in the photoisomerization of the stilbenes
was proposed by Orlandi and Siebrand9 (OS) and adopted in a
generalized mechanism for the higher members of the R,ω-
diphenylpolyene family by Birks.10 A key assumption in the
Birks mechanism, that in ttt-DPH, as in the stilbenes, torsional
motion leading to photoisomerization is the only radiationless
decay channel competing with fluorescence has been shown to
be inconsistent with photoisomerization quantum yields mea-
sured in our laboratory.11,12 Torsional barrier heights, estimated
by Birks10 from the temperature dependencies of fluorescence
quantum yields and lifetimes,13 on the basis of this assumption,
are unreliable. Specifically, in degassed acetonitrile (AN)
solutions, fluorescence, intersystem crossing, and photoisomer-
ization account for only 34-44% of 1ttt-DPH* decay at 20 °C,12
depending on which φf literature value 0.1613 or 0.2614 is
employed. It follows that, absent a very unfavorable branching
ratio, the Birks analysis neglects a radiationless decay channel
that in AN accounts for more than half of 1ttt-DPH* deactivation
without contributing to photoisomerization.
Experimental Section
Materials. Sources and purification procedures for all-trans-
1,6-diphenyl-1,3,5-hexatriene (100.0%, HPLC), benzophenone,
trans-stilbene, anthracene, and acetonitrile were as previously
described.15 Reagents used in the syntheses were the best
available commercially and, unless noted otherwise, were used
as received. The syntheses of ttt-DPH-d2 (99.9%, HPLC) and
-d4 (98.0%, HPLC, 1.7% tcc-DPH-d4, major contaminant) were
accomplished by deuteration of the corresponding acetylenes
over Lindlar catalyst, followed by diphenyl diselenide catalyzed
isomerization to predominantly all-trans equilibrium mixtures
and purification. Synthetic details follow.
ttt-DPH-d2: Triphenylcinnamylphosphonium Bromide. Cin-
namyl bromide (1.97 g, 0.011 mol) and triphenyl phosphine
(3.15 g, 0.012 mol) were dissolved in 20 mL of anhydrous
benzene and stirred at room temperature for 24 h in a 100 mL
round-bottom flask. Precipitated triphenylcinnamylphosphonium
bromide was filtered, washed with ether to remove unreacted
triphenyl phosphine, and dried over P2O5.
The energetics for the torsional relaxations about the terminal
and central double bonds of ttt-DPH* in AN are derived here
1
trans,trans- and trans,cis-1,6-Diphenyl-1,3-dien-5-yne. Bu-
tyllithium (3 mmol, 2.5 mol in hexane) was added dropwise by
syringe over a 2 h period to a solution of triphenylcin-
namylphosphonium bromide (1.38 g, 3 mmol) in 70 mL of
† Part of the special issue “George S. Hammond & Michael Kasha
Festschrift”. J.S. dedicates this paper to his inspiring teacher George S.
Hammond with gratitude.
* To whom correspondence should be addressed.
10.1021/jp021540g CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/17/2002

Photoisomerization of ttt-DPH
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3179
anhydrous diethyl ether. The solution gradually became pink
in color with the simultaneous precipitation of white lithium
bromide. After stirring the mixture for 1 h at room temperature,
an equimolar amount of phenylpropynal in anhydrous ether was
added dropwise by syringe over a 1.5 h period. Stirring was
continued until starting materials were depleted (TLC). Pre-
cipitated triphenylphosphine oxide was removed by filtration,
and the filtrate was neutralized with acetic acid, washed with
water, and dried over sodium sulfate. A yellow oil was obtained
upon evaporation of diethyl ether. Addition of 20 mL of
methanol led to precipitation of pure trans,trans-1,6-diphenyl-
1,3-dien-5-yne as a light yellow solid (1H NMR,16 UV). Pure
trans,cis-1,6-diphenyl-1,3-dien-5-yne was obtained from the
filtrate by chromatography on silica gel with hexane as eluent
(1H NMR,16 UV).
photoproduct. A Nova 500 MHz NMR spectrometer was
employed. An analytic sample of ctt-DPH-d0 was obtained by
semipreparative HPLC of the mixture resulting on 366 nm
excitation of an air-saturated AN solution of ttt-DPH-d0 (1.0 ×
10-3 M).22 1H NMR peak assignments were based unambigu-
ously on a Phase Sensitive COSY spectrum: (500 MHz,
acetone-d6) δ 6.37-6.42 (dd, Jb′a′ ) 12.0 Hz, Jb′c′ ) 12.0 Hz,
Hb′), 6.47-6.51 (d, Ja′b′ ) 12.0 Hz, Ha′), 6.62-6.67 (dd,
Jcc′)15.0 Hz, Jcb)12.0 Hz, Hc), 6.68-6.72 (d, Jab ) 15.5 Hz,
Ha), 6.98-7.04 (dd, Jc′c ) 15.0 Hz, Jc′b′ ) 12.0 Hz, Hc′), 7.04-
7.09 (dd, Jba) 15.5 Hz, Jbc ) 12.0 Hz, Hb), 7.21-7.27 (m,
1H), 7.27-7.30 (m, 1H), 7.31-7.36 (m, 2H), 7.37-7.43 (m,
4H), 7.49-7.51 (m, 2H).
all-trans-1,6-Diphenyl-1,2-dideuterio-1,3,5-hexatriene (ttt-
DPH-d2). A mixture of trans,trans- and trans,cis-1,6-diphenyl-
1,3-dien-5-yne (1.6 g, 7 mmol) was dissolved in 60 mL of
anhydrous hexane (refluxed over sodium) containing Lindlar
catalyst (10% w/w). A syringe attached to a balloon filled with
deuterium gas was inserted into the reaction flask through a
rubber stopper, and the progress of the reaction was monitored
by GLC. Upon depletion of the starting materials, the catalyst
was removed by filtration, and the filtrate was reduced to half
its original volume under vacuum. Diphenyl diselenide (0.08
g, 5% w/w) was added, and the mixture was equilibrated by
refluxing in room light for 1 h.17 The ttt-DPH-d2 was crystallized
from the reaction mixture and isolated by filtration, 0.8 g, 50%.
1
ttt-DPH-d2: mp 198-199 °C; H NMR (300 MHz, CDCl3) δ
7.20-7.45 (10H, 2 Ph), 6.86s6.92 (1H), 6.58-6.62 (d, J )
16, 1H), 6.51-6.54 (2H); MS m/z 234 (M+, 100%), 232 (M+
- 2, 2.07%).
Ar bubbled ttt-DPH-d2, 1.0 × 10-3 M in AN, was irradiated
at 366 nm in a Hanovia reactor. The irradiation was stopped at
20% over all conversion (13.5% ctt, 6.8% tct). AN was removed
completely under reduced pressure. The products were separated
partially by column chromatography on activity I alumina with
30% n-pentane in benzene (v/v) as eluent and fractions rich in
ctt/ttc-DPH-d2 were subjected to semipreparative HPLC separa-
all-trans-1,6-Diphenyl-1,2,5,6-tetradeuterio-1,3,5-hexa-
triene (ttt-DPH-d4). A mixture of cis- (major component) and
trans-1,6-diphenyl-3-hexen-1,5-diynes was synthesized from cis-
1,2-dichloroethene following the procedure described by Voll-
hardt and Winn.18 Reduction of the 1,6-diphenyl-3-hexen-1,5-
diynes (1.3 g, 5.7 mmol) with deuterium gas over Lindlar
catalyst afforded a mixture of ttt- and tct-DPH-d4 which was
equilibrated with diphenyl diselenide as the radical source17 (see
above). The pure ttt-DPH-d4 (0.7 g) was isolated in 53% yield:
mp 198-199 °C; 1H NMR (300 MHz, CDCl3) δ 7.0-7.25 (tt,
2H, Ph), 7.30-7.35 (t, 4H, Ph), 7.41-7.44 (d, 4H, Ph), 6.524
(s, 2H); MS m/z 236 (M+, 100%), 232 (M+ - 4, 1.10%).
1
tion to yield pure ctt/ttc-DPH-d2 for H NMR analysis: (500
MHz, acetone-d6) δ 6.37-6.42 (dd, Jb′a′ ) 12.0 Hz, Jb′c′ ) 12.0
Hz, Hb′, ttc), 6.47-6.51 (d, Ja′b′ ) 12.0 Hz, Ha′, ttc), 6.62-
6.67 (m, Hc, ttc, Hc, ctt), 6.68-6.72 (d, Jab ) 15.5 Hz, Ha′, ctt),
6.98-7.04 (m, Hc′, ttc, Hc′, ctt), 7.04-7.09 (m, Hb, ctt), 7.21-
7.27 (m, 1H), 7.27-7.30 (m, 1H), 7.31-7.36 (m, 2H), 7.37-
7.43 (m, 4H), 7.49-7.51 (m, 2H). The spectrum is shown in
Figure 1.
Irradiation Procedure. Irradiations were carried out in a
Moses merry-go-round19 apparatus immersed in a thermostated
water bath. A heating coil connected to a thermoregulator
(Polyscience corporation) was used to control the temperature
to (0.1 °C. The benzophenone-sensitized photoisomerization
Fluorescence Measurements. Fluorescence spectra were
measured with a Hitachi F-4500 spectrophotometer equipped
with a 150-W Xe arc source and a Hamamatsu R3788
photomultiplier tube. The scan rate was 2400 nm/min and slit
widths were set at 2.5 nm for both excitation and emission
monochromators. (The Hitachi F-4500 employs horizontal
excitation and emission slits, instead of vertical.) Fluorescence
spectra were recorded for both air saturated and Ar bubbled
solutions of DPH in acetonitrile in standard 1 cm2 quartz cells.
Temperature was maintained at 20.0 ( 0.1 °C using a Neslab-
RTE 4DD constant temperature circulation bath and was
monitored continuously during each scan with an Omega
Engineering model 199 RTD digital thermometer in a reference
cell placed in the same constant temperature cell holder.
Absorption spectra were recorded with a Cary 300 UV-Vis
spectrophotometer at 20.0 °C. Fluorescence quantum yields were
determined using quinine sulfate in 1 N H2SO4 (φf ) 0.546 at
25.0 °C, adjusted to 0.553 for 20.0 °C)21,22 as reference standard.
Absorbances of samples and reference, As and Ar, respectively,
of trans-stilbene was used for actinometry, φtfc ) 0.55.20,21
A
Hanovia medium-pressure Hg lamp (200 W, Ace Glass, Inc.)
and Corning CS 7-37 and 0-52 filters were used for excitation
at 366 nm. Solutions, 3.0 mL, were pipetted into Pyrex ampules,
13 mm o.d., degassed, and flame-sealed at a constriction. Sample
preparation, degassing, and analysis were performed under
nearly complete darkness (red light).
Analytical Procedures. Actinometer solutions were analyzed
by GLC and DPH solutions by HPLC (λmon ) 350 nm) as
previously described,12,15 except that, on replacing AN with
hexanes prior to HPLC analyses, a stream of argon was used to
evaporate the solvent and care was exercised to avoid taking
samples to dryness. The ctt/ttc ratio from the terminal bond
photoisomerization of ttt-DPH-d2 in AN was determined from
1
the H NMR spectrum (500 MHz, acetone-d6) of the isolated

3180 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Saltiel et al.
Figure 1. 1H NMR of the ttc- and ctt-DPH-d2 mixture obtained on 366 nm irradiation of ttt-DPH-d2, see text.
TABLE 1: PSS Fractions and Photoisomerization Quantum
were matched at ∼0.1 at λexc ) 365 nm. Corrections for
differences in solvent refractive index, n, and absorbance at λexc
were applied using
Yields for ttt-DPH-dn in Degassed AN (366 nm, 20.0 °C)
PSS Fractions
compound
ttc + ctt
tct
tcc + cct
ttt
φfs ) φfr (AsArns2)/(ArAsnr2)
(1)
a
ttt-DPH-d0
ttt-DPH-d2
ttt-DPH-d4
0.239
0.206
0.183
0.106
0.106
0.115
0.020
0.027
0.025
0.635
0.667
0.661
where the fluorescence areas As and Ar of the sample and the
reference, respectively, are corrected for nonlinearity in instru-
mental response. Fluorescence lifetimes were measured with a
phase modulation Fluorolog-τ2 lifetime spectrofluorometer
(SPEX) equipped with a 450-W Xe arc source, a Hamamatsu
R928P photomultiplier tube, and a Lasermetrics BNC1072FW
Pockel cell (KD2PO4). The Fluorolog-τ2 modulates the fre-
quency of the excitation light from 0.5 to 300 MHz. A glycogen
solution (τ ) 0 ns, scattered light) was used as reference.
Temperatures were maintained and measured as described
above. DPH solutions (2.0 × 10-6 M) were degassed using 8
freeze-pump-thaw cycles to 4 × 10-6 Torr and flame-sealed
at a constriction. A 13 mm o.d. tube attached to a standard 1
cm2 quartz cell via a sidearm and a graded seal was employed.
Software (DMF 3000F Spectroscopy) provided by the manu-
facturer was used to analyze the data. Determinations of the
quality of the lifetimes were based on examination of the
statistics of a fit (a plot of the residual deviations vs frequency)
and reduced ø2 values that were all close to unity.
Quantum Yields
c
compound
φctt
φtct
φcct
kH/kD
ttt-DPH-d0
ttt-DPH-d2
ttt-DPH-d4
0.062
0.054
0.046
0.035
0.036
0.038
0.0051
0.0073
0.0056
1.35b
1.37
a From ref 12. b From the quantum yields and by H NMR. c No
back reaction correction was applied to this two-photon process.
1
also shown in Table 1. A value of 1.35 was calculated
from the H NMR spectrum in Figure 1 for the ttc/ctt ratio of
1
the terminal bond isomerization products from the irradia-
tion of ttt-DPH-d2 using the ratio of the average integrations of
the signals at δ 6.37-6.42 (Hb′, ttc) and 6.47-6.51 (Ha′, ttc) to
the area of the signal at δ 6.68-6.72 (Ha, ctt). The dependence
of photoisomerization quantum yields on temperature was
measured starting from ttt-DPH-d0 and the results are shown in
Table 2.
Fluorescence Measurements. Absorption and fluorescence
spectra of the set of ttt-DPHn in AN at 20 °C are shown in
Figure 2. Fluorescence quantum yields, measured for Ar- and
air-saturated AN solutions, and fluorescence lifetimes measured
for degassed and air-saturated solutions at 20.0 °C are shown
in Table 3. Fluorescence quantum yields for Ar-saturated
solutions and fluorescence lifetimes for degassed solutions at
different temperatures are included in Table 2. Excellent fits to
the single-exponential decay model were obtained (ø2 e 1.2),
and the lifetimes were independent of excitation and monitoring
wavelengths, within experimental uncertainty. Reproducibility
in independent measurements was better than (0.1 ns.
Results
Photochemical Observations. Photostationary states were
approached from ttt-DPH-dn (n ) 2 and 4). Degassed DPH
samples (1.0 × 10-3 M) in AN were irradiated for different
time intervals at 20.0 °C until HPLC analysis revealed no further
change in the isomer composition. Photostationary state fractions
are shown in Table 1 together with corresponding values
determined earlier10 for ttt-DPH-d0. Overall conversions for
quantum yield measurements were kept below 2.5% and were
corrected for back reaction as previously described.12 They are

Photoisomerization of ttt-DPH
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3181
TABLE 2: Temperature Effects on Photoisomerization and
Fluorescence Quantum Yields of ttt-DPH-d0 and of
Fluorescence Lifetimes of ttt-DPH-dn in Degassed AN (366
nm)
Discussion
The separation into singlet and triplet components of the
photoisomerization quantum yields of ttt-DPH-d0 in degassed
AN upon 366 nm has been based on the following mechanism12
T, °C φtttfttc φtttftct
φf,a
τd0,a ns
τd2, ns τd4, ns
hV
2.5 0.0301 0.0218
10.0
15.5 0.0491 0.0294
17.2 0.0511 0.0324
20.0 0.0616 0.0346 0.269 (0.16) 4.7 (3.9, 4.7)
30.0
30.1 0.0893 0.0424
35.0 0.0879 0.0552
40.0
44.4 0.1310 0.0562
50.0
57.5 0.1795 0.0747
57.7 0.2017 0.0793
60.0
1ttt 98 1ttt*
(2)
(3)
0.308 (0.22) 5.3 (4.7, 5.1)
5.3
5.4
kf
1ttt*
9
8 1ttt + hV
4.8
4.1
4.8
4.2
0.233 (0.13) 4.1 (3.3, 4.3)
knr
9
1ttt*
8 1ttt
(4)
(5)
0.201 (0.10) 3.6 (2.8, 3.8)
0.164 (0.085) 3.0 (2.4. 3.2)
3.6
3.2
3.8
3.4
kis
1ttt*
1ttt*
1ttt*
9
8 3eet*
k1
9
8 1ptt*
(6)
0.152 (0.22) 2.8 (2.1, 2.7)
2.9
3.1
k3
9
a The first values in parentheses are interpolated from ref 13; the
second values in the lifetime column are calculated from the rate
constants in Table 4 by assigning all T-dependence to the twisting rate
constants in eqs 6 and 7, see text.
8 1tpt*
(7)
kdl
9
1ptt*
8 R′1ctt + (1 - R′)1ttt
(8)
kd2
9
1tpt*
8 â′1tct + (1 - â′)1ttt
(9)
kd
9
3eet*
8 R1ctt + â1tct + (1 - R - â)1ttt
(10)
kttt
9
3eet* + 1ttt
8 γ1ctt + δ1tct + (γen + δen)3eet* +
[2(1 - γ - δ) + γq + δq]1ttt (11)
where “p” designates a twisted double bond (e.g., ptt indicates
twisting at a terminal double bond), “e” designates a double
bond that exists as an equilibrium mixture of planar geom-
etries,17 and the greek letters are decay fractions from twisted
singlets and from equilibrated planar triplets. Application of the
steady-state approximation to all excited species gives quantum
yield expressions 12 and 13 for ctt-DPH and tct-DPH formation:
Figure 2. Absorption and fluorescence spectra of ttt-DPH-dn (λexc
)
365 nm, 20.0 °C). The n ) 2 and 4 spectra are successively displaced
on the ordinate scale to facilitate comparison.
φois(R + γ kttt τTo[1ttt])
1 + (1 - γen - δen)kttt τoT[1ttt]
φctt ) R′k1τos +
φtct ) â′k3τoS +
(12)
(13)
TABLE 3: Fluorescence Quantum Yields and Lifetimes of
ttt-DPH-dn in AN, 20.0 °C
τ, ns
φois(â + δ kttt τTo [1ttt])
1 + (1 - γen - δen) kttt τoT[1ttt]
compound
φf
φfAr/φfAir
degassed
air
τ0/τ
ttt-DPH-d0
ttt-DPH-d2
0.269
0.271
1.29 (1.32)a
1.30
4.69
4.77
3.50
3.72
3.61
3.64
3.57
3.60
3.67
1.34
1.30
where τo ) (kf + knr + kis + k1 + k3)-1 is the singlet lifetime
S
of ttt-DPH and τoT ) kd is the lifetime of the equilibrated
-1
ttt-DPH-d4
0.279
0.277
1.30
4.76
4.76
4.75
1.33
1.32
1.29
triplet. The first terms on the right-hand side of the equal signs
in eqs 12 and 13 are the singlet contributions, φSxxt, and the
second terms are the triplet contributions, φTxxt, to the observed
quantum yields. The very low intersystem crossing yield in AN,
φ0is ) 0.01,12 and the previously defined parameters for DPH
triplets have led to the conclusion that, at 20.0 °C, photoisomer-
ization in this solvent is due primarily to the singlet pathway.12
This conclusion is consistent with the results in Table 1. The
new φtct value agrees exactly with our previous value,12 and
the new φctt value is 10% higher, probably because ctt-DPH
survives better under the milder conditions used to replace the
solvent prior to HPLC analysis. Nearly all terminal bond
photoisomerization (98.3%) and most central bond isomerization
(83.2%) occur in the singlet manifold.
a The value in parentheses is for a degassed instead of an Ar-saturated
solution; see also ref 12.
torsional barriers for twisting about the terminal and central
double bonds of ttt-DPH-d0, based on the above mechanism.
Examination of Table 2 shows that our lifetime values are
systematically higher than earlier values,13 probably because of
the difference between imperfect outgassing and our thorough
degassing. We proceed, initially, by neglecting the φTxxt contri-
butions, because they are small and their temperature depend-
encies are not known in AN. Plots of ln(φxxt/τoS) vs T-1, Figure
3, are, to a very good approximation (excellent in the case of
terminal bond isomerization), equivalent to Arrhenius plots of
the quantities R′2k1 (there are two equivalent terminal bonds)
and â′k3. Assuming, by analogy with stilbene,25 temperature
Energetics of Torsional Relaxation in ttt-DPH-d0. The
temperature dependencies of photoisomerization quantum yields
and fluorescence lifetimes in Table 2 can be used to estimate

3182 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Saltiel et al.
becomes less pronounced.30 Such effects were associated with
differences in vibrational overlap Franck-Condon factors, that
is, mainly the relative ability of CH and CD stretching vibrations
in the ground state to function as accepting modes for the
electronic energy.27-29 Recognition that d-substitution effects
on intersystem crossing rate constants depend also on the
position of substitution31-35 led to the proposal that specific out-
of-plane CH (CD less well) vibrations enhance spin-orbit
coupling between π and σ states.36
The comparative study of deuterated DPH derivatives was
initiated in order to determine (a) how the dominant radiationless
decay component, which is along coordinates that are unproduc-
tive with respect to trans f cis isomerization, responds to
d-substitution and (b) whether DIEs on photophysical and
photochemical processes in ttt-DPH are consistent with known,
highly position-dependent deuterium isotope effects on the
excited states of stilbene. Specifically, the lifetime of the
perpendicular triplet state of stilbene, 3p*, increases by 30% on
deuteration of the ethylenic positions37,38 but is not affected by
perdeuteration of the aromatic positions.37 Only in rigid media
at low temperatures where the trans-stilbene triplet maintains
a planar geometry is there a measurable increase in the triplet
lifetime on deuterium substitution of the aromatic positions.37,39
In solution, the deuterium isotope effect on the torsional
Figure 3. Temperature effect on twisting rate constants for terminal
(closed circles) and central (open circles) bonds of ttt-DPH-d0*.
1
independent decay fractions from the twisted intermediates, the
slopes of the lines in Figure 3 give Eptt ) 8.40 ( 0.20 kcal/mol
and Etpt ) 6.52 ( 0.20 kcal/mol as the activation energies for
torsional relaxation about the terminal and central double bonds
in 1ttt-DPH-d0*. Both of these values place the torsional barriers
well above the spectroscopically estimated 21Ag/11Bu energy
1
relaxation of the singlet excited state of trans-stilbene, t*, is
also specific to substitution of the ethylenic hydrogens.40,41
Simultaneous analysis of the temperature dependencies of τf
and φf values has led to the conclusion that the torsional barrier
is enhanced slightly in 1t*-d2 (0.2 and 0.3 kcal/mol in n-hexane
and n-tetradecane, respectively).41,42 At 20 °C in n-hexane,
where most stilbene singlets decay along the 1t* f 1p*
relaxation coordinate, the effect on the twisting rate constant is
nearly identical to the effect on the fluorescence quantum yields,
kH/kD ) 1.6.41 A factor of 2 attenuation of the radiationless decay
rate constant was observed earlier for the isolated molecules
under jet-cooled conditions on perdeuteration of trans-stilbene
for excitation above the torsional barrier.43 These DIEs are
accounted for satisfactorily by RRKM calculations which, for
the favored 11Bu f 21Ag crossing mechanism, predict an
activation energy increase of 0.28 kcal/mol on substitution of
the two vinyl-Hs,44 consistent with experiment.41 The magnitude
of this effect appears to be related to the torsional barrier height
because (a) there is no DIE on 1c* f 1p*, the barrierless
torsional relaxation of cis-stilbene in n-hexane,45 and (b)
d-substitution of the single vinyl H of trans-1-phenylcyclohex-
ene gives kH/kD ) 2.0 over a 12.1 kcal/mol trans f cis torsional
barrier.46 Theoretical calculations (ab initio MCSCF with 6-31
G* and/or 6-311 G** basis sets) indicate that the loss of zero-
point energy, primarily responsible for large predicted kinetic
DIEs in the thermal cis f trans isomerizations of alkenes,
involves a number of vibrational modes instead of a single
torsional mode.47 These calculations were bolstered by the
observation of kH/kD ) 1.5 at 287 °C for the thermal cis f
trans isomerization of cis-stilbene on d-substitution of its two
vinyl-Hs.47 Also relevant to our study are DIEs on the
interconversion of the cis- and trans-2,5-di-tert-butyl-1,3,5-
hexatienes.48 The small photoisomerization quantum yields in
diethyl ether, 0.046 and 0.052 in the trans f cis and cis f
trans directions, respectively, are attenuated by about 40% on
gap of 4.1 kcal/mol. Arrhenius A factors, can be estimated from
the intercepts of the plots in Figure 3 (30.77 ( 0.34 and 27.02
( 0.33 for terminal and central double bonds, respectively) by
assuming R′ = â′ = 0.5. The resulting values are Aptt ) (2.30
( 0.91) × 1013 and Atpt ) (1.08 ( 0.40) × 1012 s-1. We will
show below that adjusting the isomerization quantum yields by
subtracting approximated triplet contributions does not seriously
affect these parameters. Similar temperature effects, observed
on photoisomerization quantum yields in the hydrocarbon
solvents benzene and methylcyclohexane (MCH), are not as
easily interpreted because of much more significant triplet
contributions that cannot be neglected.26 However, one important
conclusion is inescapable. The selective, more than 20-fold
enhancement of terminal bond isomerization on increasing
solvent polarity from MCH to AN is consistent with the
proposed zwitterionic ptt-DPH twisted intermediate.11,12 It is
caused by a pronounced enhancement of the frequency factor.
It is not consistent with the usual explanation of a lowering of
the torsional barrier by the polar solvent.3 If anything, the barrier
1
for terminal bond twisting in ttt-DPH-d0* is higher in AN.
Previous derivations of overall effective torsional barriers (3.4
kcal/mol10 and 5.3 kcal/mol13 in AN) based only on fluorescence
measurements yielded significantly underestimated values be-
cause of the erroneous assumption that all radiationless decay
is along photoisomerization coordinates.
Deuterium Isotope Effects. Triplet lifetime enhancement
upon perdeuteration of aromatic hydrocarbons was important
in the development of theories of radiationless transitions.27-29
As the T1-S0 energy gap decreases, intersystem crossing rate
constants increase and the kinetic deuterium isotope effect (DIE)
deuteration of the central CC bond, consistent with kH/kD
)
1.6 in both directions. Strong involvement of out-of-plane CH
vibrations in the path to photoisomerization was suggested.49
We consider first the DIE on the isomerization of the terminal
1
bonds in ttt-DPH*. Because the first term in eq 12 accounts

Photoisomerization of ttt-DPH
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3183
TABLE 4: Photophysical Parameters of ttt-DPH-dn in AN,
20.0 °C
expected, the minor correction for photoisomerization in the
triplet state does not affect significantly the rate constants for
terminal bond twisting nor does it affect the DIE for that process,
k1H/k1D ) 1.37, Table 4. The decrease in the rate constant for
terminal bond twisting, alone, accounts for the small changes
in fluorescence quantum yields and lifetimes. Use of the average
rate constants in Table 4 predicts (τf, φf) pairs of (4.67 ns, 0.268),
(4.75 ns, 0.273), and (4.83 ns, 0.278) for ttt-DPH-d0, ttt-DPH-
d2, and ttt-DPH-d4, respectively. The agreement with the
observed values, Table 3, is excellent.
a
a
10-7kf, 10-7k1H
,
10-7k1D
,
10-7k3,b 10-7kis,c 10-7knr,d
compound
s-1
s-1
s-1
s-1
s-1
s-1
ttt-DPH-d0
ttt-DPH-d2
ttt-DPH-d4
5.74
5.68
5.84
2.60/2
1.28
1.24
1.26
1.34
1.28
0.21
0.21
0.21
0.21
11.5
11.6
11.7
11.6
0.94
1.89/2
0.94
Ave. 5.75
1.29
a Based on quantum yields in Table 1, adjusted for triplet component
(0.001, see ref 12); these quantities represent 2k1H, k1H + k1D, and 2k1D
for ttt-DPH-d0, ttt-DPH-d2, and ttt-DPH-d4, respectively. b Based on
quantum yields in Table 1, adjusted for triplet component (0.006, see
ref 12). c Based on φis ) 0.01 from ref 14. d The difference between
τf-1 and ∑ki, where the ki are the entries in the preceding columns.
Corrected Torsional Relaxation Energy Barriers. Because
the temperature effect on τf for ttt-DPH-d0, Table 2, is accounted
for, almost exactly, if one assumes that the isomerization
channels are the only temperature-dependent processes, we were
encouraged to further refine the derivation of the Arrhenius
parameters for twisting by calculating approximate net isomer-
ization quantum yields along the singlet excited-state pathways.
The known triplet contributions to the photoisomerizations at
20 °C were assumed to vary with temperature in the same way
as the quantities φisR and φisâ. The temperature dependence of
the intersystem crossing quantum yield, φis, was obtained from
the measured fluorescence lifetimes by assuming that kis (Table
4) is T-independent and that the triplet decay fractions, R and
â, have the same temperature dependencies in AN as they do
in benzene.17 Arrhenius plots (not shown) of torsional relaxation
rate constants based on the estimated singlet pathway isomer-
ization quantum yields give Eptt ) 8.48 ( 0.20 kcal/mol and
Etpt ) 7.38 ( 0.25 kcal/mol as the activation energies for
torsional relaxation about the terminal and central double bonds
in 1ttt-DPH-d0*. The corresponding Arrhenius frequency factors
are Aptt ) (2.60 ( 1.07) × 1013 and Atpt ) (3.84 ( 2.0) × 1012
s-1. The activation parameters for terminal bond twisting are
identical within experimental uncertainty to those derived above
without this correction. Changes in the activation parameters
for central bond twisting are more pronounced, but the overall
conclusions are not affected.
for more than 98% of φctt at 20 °C,12 the ratio k1H/k1D can be
obtained from
d0 d4
ctt ctt
k1H/k1D ) (R′ /R′ )(φ /φ )(τd4/τd0)
(14)
H
D
by assuming that the decay fraction is not affected by d-
substitution (R′H ) R′ ), as is the case for stilbene singlets and
D
triplets.49 Substitution of the ratios of the observed quantum
yields (Table 1) and fluorescence lifetimes (Table 3) for ttt-
DPH-d0 and ttt-DPH-d4 into eq 14 gives k1H/k1D ) 1.37, in
excellent agreement with 1.35, the value obtained by 1H NMR
analysis of the ttc- and ctt-DPH photoproduct ratio from ttt-
DPH-d2. The observation that φctt for ttt-DPH-d2 exactly equals
the average of the corresponding quantum yields for ttt-DPH-
d0 and ttt-DPH-d4 provides further confirmation of these
independent inter- and intramolecular determinations of k1H
/
k
1D. The effect is smaller than the corresponding value for the
1t* f 1p* process in trans-stilbene but is consistent47 with the
proposed torsional relaxation mechanism for terminal bond
isomerization.11,10
Our fluorescence quantum yield for ttt-DPH-d0 in AN at 20.0
°C, Table 2, is higher than the first reported value13 but agrees
with 0.26, the value reported by Schael and Lo¨hmannsro¨ben.14
Our lifetimes are systematically higher than those reported by
Cehelnik et al.,13 Table 2, and the 4.1 ns value measured at
20.0 °C by Schael and Lo¨hmannsro¨ben.14 The reproducibility
of our values (to (0.1 ns in independent experiments, Table 3)
and the difference in deaerating procedures suggests that residual
oxygen was responsible, at least in part, for the lower quantum
yields and lifetimes reported earlier.
Assuming that torsional relaxations along photoisomerization
coordinates are the only T-dependent processes and using the
refined Arrhenius parameters to calculate the rate constants
improves the agreement between calculated and observed
fluorescence lifetimes slightly, Table 2. Precisely stated, agree-
ment between observed and calculated lifetimes is attained by
assuming that the sum of the rate constants kf, knr, and kis is
T-dependent. We conclude that if one of these rate constants
actually varies with T, complementary changes in the other two
must compensate for this variation. Furthermore, because kis is
negligibly small, the sum of kf + knr is insensitive to T.
Rate Constants for the Major Decay Paths. The T-
dependence of the effective radiative rate constant kf was
obtained from the τf and φf values in Table 2. Normally,
radiative rate constants for allowed transitions decrease slightly
with increasing temperature because of their dependence on the
index of refraction. For instance, for the allowed 1Bu f 1Ag
transition in trans-stilbene, kf decreases by less than 4% for a
temperature increase of 60 °C in n-alkane solvents.41 In ttt-
DPH, the situation is much more complex because (a) the major
portion of the emission originates from the 2Ag state and is
forbidden,1,2 (b) there is competing fluorescence from the nearby
1Bu state (the two states, 1Bu and 2Ag, are vibronically coupled
and exist as an equilibrium mixture in the all-s-trans conformer
of ttt-DPH),8 (c) the degree of coupling and, therefore, the
magnitudes of the radiative rate constants kfA and kfB depend
on the 1Bu/2Ag energy gap, which in turn is T-dependent
because the energy of the 1Bu state decreases with increasing
Except for a small loss in vibronic resolution in the
fluorescence spectra, the spectra of ttt-DPH are nearly inde-
pendent of d-substitution, Figure 2. Fluorescence quantum yields
and lifetimes, Tables 2 and 3, reveal little enhancement on
deuteration. Much bigger changes would apply if knr, the rate
1
constant for the major decay process of ttt-DPH*, eq 4, were
affected. Effective radiative rate constants kf (φf/τf ratios at 20.0
°C, Table 3) are nearly identical, Table 4. The average value,
5.75 × 107 s-1, is assumed to apply to the three members of
our ttt-DPH-dn series. Rate constants for the other unimolecular
decay processes, also derived from quantum yield-to-lifetime
ratios, are collected in Table 4. Calculation of the torsional
relaxation rate constants of the excited singlet state was based
on the isomerization yields corrected for the φTxxts, the triplet
contributions. The latter, taken from ref 12, were assumed to
be independent of d-substitution. Assignment of the unaccounted
for portions of the decay rates results in knr values that are
independent of d-substitution. Multiplied by the observed
lifetimes, they give nearly identical φnr values of 0.54, 0.55,
and 0.56 for ttt-DPH-d0, ttt-DPH-d2, and ttt-DPH-d4, respec-
tively, the increase reflecting the slight change in lifetime. As

3184 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Saltiel et al.
polarizability (decreasing T), and (d) fluorescence also stems
from the s-cis,s-trans conformer and its contribution increases
with increasing T (excitation at the red edge of the absorption
spectrum leads to enhanced emission at longer wavelengths as
has been reported for methylcyclohexane).50 Our φf values are
larger than earlier values,13 Table 2. We consider our quantum
yields more reliable because our value at 20 °C agrees with a
subsequent determination14 and because ref 13 also reports small
τf values. Effective radiative rate constants (kf) obtained from
the φf/τf ratios increase from 5.42 × 107 at 60 °C to 5.81 × 107
s-1 at 10 °C. The 7% change is marginally larger than the
change experienced by the radiative rate constant of trans-
stilbene. It suggests that the small decrease in kfA with increasing
T (because of the increase in the 1Bu/1Ag energy gap) is nearly
exactly canceled by expected increases that should accompany
the larger fractions of fluorescence from the two higher energy
components.
Complementary changes in the rate constant knr of the major
radiationless decay process predict a very modest 3.4% increase
over the 10-60 °C T range, corresponding to a negligible
activation energy, Enr, of 0.13 kcal/mol. This activation energy
is much smaller than ∆EBA, the 11Bu/21Ag energy gap, ensuring
that the radiationless process in eq 4 originates in S1. The rate
constant for eq 4 is enhanced from 2.2 × 107 to 1.15 × 108 s-1
on changing the solvent from MCH to AN.10 In both solvents,
Figure 4. Possible torsional relaxation pathways leading to 1ttt-DPH*
photoisomerization (schematic, not to scale).
an allyl cation-methylene anion intermediate, analogous to our
ptt intermediate, to account for the regioselectivity60 (zwitterionic
1,3-diene excited-state intermediates were first proposed by
Dauben61). The very low trans f cis photoisomerization yield
of trans-H in solution (0.016)62 is consistent with theoretical
computations53c which predict that, following S2 (Bu) f S1 (Ag)
decay in the fs time scale, S1 f S0 radiationless decay occurs
very fast to the original ground state via a barrierless process
involving a “nontotally symmetric deformation of the molecular
backbone” that provides access to a CI.53c Such decay is
predicted to be slow in trans,trans-1,3,5,7-octatetraene because
the analogous transition state is calculated to be less accessible
energetically.53c Two recent studies of internal conversion in
all-trans-â-carotene may be relevant. The effect of 13C-
substitution on the rate of radiationless decay of all-trans-â-
carotene was found to be significantly larger than that of
d-substitution leading to the conclusion that the in-phase CdC
normal stretching mode plays a major role in the 2Ag f 1Ag
internal conversion.63 Similarly, a recent CARS spectroscopic
study of the population dynamics in vibrational modes that
accompany the S1 f S0 radiationless decay of all-trans-â-
carotene bolsters the conclusion that nuclear motions of CC
bonds along the polyene backbone are good candidates for the
bottleneck of population flow.64
the radiationless transition is faster than knr ) 1.2 × 106 s-1
,
the value expected from Siebrand’s energy-gap law51 for
polynuclear aromatic hydrocarbons with the same S1-S0 energy
gap as DPH. This process originates in the lowest excited singlet
state and returns the system to the original ground state. It
corresponds to a 2Ag f 1Ag transition in which olefinic CH
vibrations fail to function as either promoting or accepting
modes. Other possibilities, such as aborted photochemical
pathways involving new CC bond formation, come to mind.
Analogy with unsubstituted 1,3,5-hexatriene (H) may be instruc-
tive. Irradiation of cis-H at 20 K in an argon matrix gives a
thermally unstable product (IR) that was tentatively identified
as exo-2-vinylbicyclo[1.1.0]butane.52 The absence of such a
product on irradiation of trans-H under the same conditions was
rationalized on the basis of the expectation that the analogous
process in the lowest excited singlet state of trans-H would have
to give the sterically congested endo isomer.52 Theoretical
calculations favor 1,3-bond formation in the 2Ag states of
polyenes at the conical intersection (CI) geometry for ultrafast
radiationless decay to the ground state,53,54 and it has been
argued, on both theoretical53,54 and experimental grounds,55 that
such structures play a role in cis-trans photoisomerization (the
Hula-Twist mechanism56). The possible involvement of cyclo-
propylmethylene intermediates in the photoisomerization of 1,3-
butadienes was proposed by one of us long ago as an alternative
to allylmethylene intermediates.57 Their presence was considered
earlier by Srinivasan to account for the formation of cyclopro-
penes and the related butadiene photodimer.58 The small
calculated torsion angle at the CI for 1,3-butadiene has been
thought54 to explain the supposed low isomerization efficiencies
of the 1,3-pentadienes (the 0.1 values cited57a,59 are for 254 nm
excitation which is selectively absorbed by s-cis conformers;60
excitation of s-trans,trans-1,3-pentadiene at 229 nm results in
an even lower trans-cis photoisomerization quantum yield59 of
0.025). This explanation, however, does not account for the
highly efficient photoisomerization of the 2,4-hexadienes57a and
for the regioselective cis f trans photoisomerization of the C1C2
bond of trans-1,3-pentadiene, revealed by Squillacote’s study
of 1-cis-deuterio-trans-1,3-pentadiene.60 Squillacote proposed
Conclusions
Photoisomerization in the singlet excited state of ttt-DPH
occurs by competing torsional relaxation pathways involving
the terminal and central bonds of the triene moiety leading to
twisted intermediates, 1ptt* and 1tpt*, with perpendicular
geometries.11,12 These two processes are differentially enhanced
on increasing solvent polarity (MCH to AN). Terminal bond
isomerization is enhanced more than 20-fold, whereas central
bond isomerization is enhanced only 3-fold.12 The solvent effect
and substituent effects11 are consistent with the proposal that
1
1ptt* and tpt* are zwitterionic and biradicaloid, respectively,
at least in polar solvents.11 Because these isomerization pathways
are subject to large torsional barriers that far exceed ∆EBA, both
of these states and higher energy states are viable intermediates
for the photoisomerization. Two possible pathways, one for each
of the lowest two excited states, are shown in Figure 4, and
one may apply preferentially to the central bond and the other
to the terminal bonds. The involvement of the higher Ag state
is consistent with the modification of the OS mechanism

Photoisomerization of ttt-DPH
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3185
proposed by Hohlneicher and Dick,65 and the small torsional
barrier in the 1Bu state is the pathway favored by Troe and
Weitzel for stilbene photoisomerization.66 Other possibilities can
be envisioned.
(12) Saltiel, J.; Wang, S.; Watkins, L. P.; Ko, D.-H. J. Phys. Chem. A
2000, 104, 11443-11450.
(13) Cehelnik, E. D.; Cundall, R. B.; Lockwood, J. R.; Palmer, T. J. J.
Phys. Chem. 1975, 79, 1369-1380.
(14) (a) Schael, F. W. Dissertation; Technische Universita¨t: Carlo-
Wilhelmina zu Braunschweig, Germany, 1995. (b) Schael, F.; Lo¨hm-
annsro¨ben, H.-G. Chem. Phys. 1996, 206, 193-210.
(15) Saltiel, J.; Wang, S.; Ko, D.-H.; Gormin, D. A. J. Phys. Chem. A
1998, 102, 5383-5392.
Terminal bond twisting is sensitive to d-substitution, k1H/k1D
) 1.37 for both olefinic positions, consistent with analogous
observations on trans-stilbene,37-43 and with theoretical expecta-
tions.36,44,47 Whether the central bond is similarly sensitive to
d-substitution remains to be established. The relatively low
Arrhenius frequency factor Atpt associated with crossing the
torsional barrier for the central bond suggests that the process
may be diabatic as in a variant of the OS mechanism for stilbene
(e.g., a partially avoided crossing between the 1Bu state and a
higher nAg state, n > 2).9 The polar solvent AN enhances the
Arrhenius frequency factor Aptt for the terminal bonds to the
point where the possibility that this process has been rendered
adiabatic must be entertained.
In view of the tendency in vogue to explain the radiationless
deactivation of polyenes by invoking ultrafast crossing of CIs,
perhaps the most unexpected conclusion is that decay from 1ttt-
DPH* is dominated by eq 4, an essentially barrierless radia-
tionless pathway, which is insensitive to d-substitution at one
or both terminal bonds of the triene moiety. Nuclear motions
of CC bonds along the triene backbone may be involved in this
process as is proposed for polyenes generally53,54 and as has
been found recently for what may be the analogous transition
in all-trans-â-carotene.63,64 However, branching between pho-
toisomers and the original ground state is supposed to occur in
the ground state after internal conversion through a CI.53,54 Such
a model clearly does not apply to DPH. The major radiationless
decay process is completely divorced from trans f cis photoi-
somerization, is barrierless within experimental uncertainty,
shows no DIE, and occurs in the ns time scale. If ultrafast decay
characterizes crossing a CI, then a CI is not involved here. For
a CI via partial 1,3-bond formation, or CC bonding on the path
to a bicyclobutane to play a role in eq 4, it would have to be
neither ultrafast nor along cis-trans photoisomerization reaction
coordinates. More work is required to elucidate the mechanism
of this highly effective radiationless decay channel.
(16) Nu¨chter, U.; Zimmermann, G.; Francke, V.; Hopf, H. Liebigs Ann.
Recl. 1997, 1505-1515.
(17) Saltiel, J.; Crowder, J. M.; Wang, S. J. Am. Chem. Soc. 1999, 121,
895-902.
(18) Vollhardt, K. P. C.; Winn, L. S. Tetrahedron Lett. 1985, 26, 709-
12.
(19) Moses, F. G.; Liu, R. S. H.; Monroe, B. M. Mol. Photochem. 1969,
1, 245-249.
(20) Hammond, H. A.; DeMeyer, D. E.; Williams, J. L. R. J. Am. Chem.
Soc. 1969, 91, 5180-5181.
(21) Valentine, D., Jr.; Hammond, G. S. J. Am. Chem. Soc. 1972, 94,
3449-3454.
(22) Saltiel, J.; Sears, D. F., Jr.; Ko, D.-H.; Park, K. M. In Handbook
of Organic Photochemistry; Horspool, W. M., Song, P.-O., Eds.; CRC
Press: London, 1995; Section 1, pp 3-15.
(23) Melhuish, W. H. J. Phys. Chem. 1961, 65, 229-235.
(24) Meech, S. R.; Phillips, D. J. Photochem. 1983, 23, 193-217.
(25) Saltiel, J.; Charlton, J. L. In Rearrangements in Ground and Excited
States; de Mayo, P., Ed.; 1980; Vol. III, pp 25-89.
(26) (a) Ko, D.-H. Ph.D. Dissertation, Florida State University, Tal-
lahassee, FL, 1997. (b) Wang, S. Ph.D. Dissertation, Florida State University,
Tallahassee, FL, 1998.
(27) (a) Robinson, G. W.; Frosch, R. P. J. Chem. Phys. 1962, 37, 1962.
(b) Robinson, G. W.; Frosch, R. P. J. Chem. Phys. 1963, 38, 1187.
(28) Lower, S. K.; El-Sayed, M. A. Chem. ReV. 1966, 66, 199.
(29) Jortner, J.; Rice, S. A.; Hochstrasser, R. M. AdV. Photochem. 1969,
7, 149-309.
(30) Siebrand, W. The Triplet State; Beirut Symposium, 1967; Zahlan,
A. B., Ed.: Cambridge University Press: Cambridge, 1967; p 31.
(31) Hirota, N.; Hutchison, C. A., Jr. J. Chem. Phys. 1967, 46, 1561.
(32) (a) Martin, T. E.; Kalantar, A. H. J. Chem. Phys. 1968, 48, 4996.
(b) Martin, T. E.; Kalantar, A. H. Chem. Phys. Lett. 1968, 1, 623.
(33) (a) Watts, R. J.; Strickler, S. J. J. Chem. Phys. 1968, 49, 3967. (b)
Burke, T. D.; Watts, R. J.; Strickler, S. J. J. Chem. Phys. 1969, 50, 5425.
(34) Simpson, J. D.; Offen, H. W.; Burr, J. D. Chem. Phys. Lett. 1969,
2, 383.
(35) Henry, B. R.; Charlton, J. L. J. Am. Chem. Soc. 1973, 95, 2782-
2785.
(36) Henry, B. R.; Siebrand, W. J. Chem. Phys. 1971, 54, 1072. (b)
Henry, B. R.; Siebrand, W. Chem. Phys. Lett. 1969, 3, 327.
(37) Saltiel, J.; D’Agostino, J. T.; Herkstroeter, W. G.; Saint-Ruf, G.;
Buu-Ho¨ı, N. P. J. Am. Chem. Soc. 1973, 95, 2543-2549.
(38) Caldwell, R. A. Pure Appl. Chem. 1984, 56, 1167-1177.
(39) Heinrich, G.; Holzer, G.; Blume, H.; Schulte-Frohlinde, D. Z.
Naturforsch. B 1970, 25, 496.
Acknowledgment. The National Science Foundation, most
recently by Grant CHE 9985895, supported this research.
(40) Courtney, S. H.; Balk, M. W.; Phillips, L. A.; Web, S. P.; Yang,
D.; Levy, D. H.; Fleming, G. R. J. Chem. Phys. 1988, 89, 6697.
(41) Saltiel, J.; Waller, A. S.; Sears, D. F., Jr.; Garrett, C. Z. J. Phys.
Chem. 1993, 97, 2515-2522.
(42) The corresponding effects on the transmission coefficients are small
(10-15%), but the changes are in opposite directions in the two solvents.30
(43) Felker, P. M.; Zewail, A. H. J. Phys. Chem. 1985, 89, 5402-5411.
(44) Negri, F.; Orlandi, G. J. Phys. Chem. 1991, 95, 748-757 and
references cited.
References and Notes
(1) Hudson, B. S.; Kohler, B. E. Annu. ReV. Phys. Chem. 1974, 25,
437-460.
(2) Hudson, B. S.; Kohler, B. E.; Schulten, K. In Excited States; Lim,
E. C., Ed.; Academic Press: New York, 1982; Vol. 6, pp 1-95.
(3) Saltiel, J.; Sun, Y.-P. In Photochromism, Molecules and Systems;
Du¨rr, H., Bouas-Laurent, H., Eds.; Elsevier: Amsterdam, 1990; pp 64-
164.
(4) (a) Pfeiffer, M.; Werncke, W.; Hogiu, S.; Kummrow, A.; Lau, A.
Chem. Phys. Lett. 1998, 295, 56-62. (b) Hogiu, S.; Werncke, W.; Pfeiffer,
M.; Lau, A. Chem. Phys. Lett. 1999, 303, 218-222. (c) Werncke, W.; Hogiu,
S.; Pfeiffer, M.; Lau, A.; Kummrow, A. J. Phys. Chem. A 2000, 104, 4211-
4217.
(5) Hilinski, E. F.; McGowan, W. M.; Sears, D. F., Jr.; Saltiel, J. J.
Phys. Chem. 1996, 100, 3308-3311.
(6) Yee, W. A.; O’Neil, R. H.; Lewis, J. W.; Zhang, J. Z.; Kliger, D.
S. Chem. Phys. Lett. 1997, 276, 430-434.
(7) Hogiu, S.; Werncke, W.; Pfeiffer, M.; Lau, A.; Steinke, T. Chem.
Phys. Lett. 1998, 287, 8-16.
(8) Itoh, T.; Kohler, B. E. J. Phys. Chem. 1987, 91, 1760-1764.
(9) Orlandi, G.; Siebrand, W. Chem. Phys. Lett. 1974, 30, 352-354.
(10) Birks, J. B.; Birch, D. J. S. Chem. Phys. Lett. 1975, 31, 608-610.
(b) Birks, J. B.; Tripathy, G. N. R.; Lumb, M. D. Chem. Phys. 1978, 33,
185-194. (c) Birks, J. B. Chem. Phys. Lett. 1978, 54, 430-434.
(11) Saltiel, J.; Ko, D.-H.; Fleming, A. S. J. Am. Chem. Soc. 1994, 116,
4099-5000.
(45) Saltiel, J.; Waller, A. S.; Sears, D. F., Jr. J. Am. Chem. Soc. 1993,
115, 2453-2464.
(46) Caldwell, R. A.; Misawa, H.; Healy, E. F.; Dewar, M. J. S. J. Am.
Chem. Soc. 1987, 109, 6869-6870.
(47) Olson, L. P.; Niwayama, S.; Yoo, H.-Y.; Houk, K. N.; Harris, N.
J.; Gajewski, J. J. J. Am. Chem. Soc. 1996, 118, 886-892.
(48) Brouwer, A. M.; Cornelisse, J.; Jacobs, H. J. C. J. Photochem.
Photobiol. A: Chem. 1988, 42, 313-320.
(49) Saltiel, J. J. Am. Chem. Soc. 1968, 90, 6394-6400.
(50) Saltiel, J.; Sears, D. F., Jr.; Sun, Y.-P.; Choi, J.-O. J. Am. Chem.
Soc. 1992, 114, 3607-3612.
(51) (a) Siebrand, W. J. Chem. Phys. 1967, 46, 440; 47, 2411. (b)
Klessinger, M.; Michl, J. Excited States and Photochemistry of Organic
Molecules; VCH Publisher: New York, 1995; p 254.
(52) Datta, P.; Goldfarb, T. D.; Boikess, R. S. J. Am. Chem. Soc. 1971,
93, 5189-5193.
(53) (a) Olivucci, M.; Bernardi, F.; Celani, P.; Ragazos, I. N.; Robb,
M. A. J. Am. Chem. Soc. 1994, 116, 1077-1085. (b) Celani, P.; Garavelli,
M.; Ottani, S.; Bernardi, F.; Robb, M. A.; Olivucci, M. J. Am. Chem. Soc.

3186 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Saltiel et al.
1995, 117, 11584-11585. (c) Garavelli, M.; Celani, P.; Bernardi, F.; Robb,
M. A.; Olivucci, M. J. Am. Chem. Soc. 1997, 119, 11487-11494.
(54) For a recent review of polyene photophysics, see: Fuss, W.; Haas,
Y.; Zilberg, S. Chem. Phys. 2000, 259, 273-295.
(59) Vanderlinden, P.; Boue´, S. J. Chem. Soc. Chem. Commun. 1975,
932-933.
(60) Squillacote, M. E.; Semple, T. C. J. Am. Chem. Soc. 1987, 109,
892-894.
(61) (a) Dauben, W. G.; Ritscher, J. S. J. Am. Chem. Soc. 1970, 92,
2925-2926.
(62) (a) Jacobs, H. J. C.; Havinga, E. AdV. Photochem. 1979, 11, 305.
(b) Vroegop, P. J.; Lugtenburg, J.; Havinga, E. Tetrahedron 1973, 29, 1393.
(63) Nagae, H.; Kuki, M.; Zhang, J.-P.; Sashima, T.; Mukai, Y.; Koyama,
Y. J. Phys. Chem. A 2000, 104, 4155-4166.
(55) Mu¨ller, A. M.; Lochbrunner, S.; Schmid, W. E.; Fu¨ss, W. Angew.
Chem., Int. Ed. 1998, 37, 505-507.
(56) (a) Liu, R. S. H.; Asato, A. E. Proc. Natl. Acad. Sci. U.S.A. 1985,
82, 259-263. (b) Liu, R. S. H.; Hammond, G. S. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 11153-11158. (c) Liu, R. S. H. Acc. Chem. Res. 2001,
34, 555-562.
(57) (a) Saltiel, J.; Metts, L.; Wrighton, M. J. Am. Chem. Soc. 1970,
92, 3227-3229. (b) Saltiel, J.; D′Agostino, J. T.; Megarity, E. D.; Metts,
L.; Neuberger, K. R.; Wrighton, M.; Zafiriou, O. C. Org. Photochem. 1973,
3, 1-113.
(64) Siebert, T.; Schmitt, M.; Engel, V.; Materny, A.; Kiefer, W. J. Am.
Chem. Soc. 2002, 124, 6242-6243.
(65) Hohlneicher, G.; Dick, B. J. Photochem. 1984, 27, 215.
(58) Srinivasan, R. J. Am. Chem. Soc. 1968, 90, 4498-4499.
(66) Troe, J.; Weitzel, K.-M. J. Chem. Phys. 1988, 88, 7030.