Sequential forward S2-S2 and back S 1-S1 (cyclic) energy transfer in a novel azulene-zinc porphyrin dyad

Article abstract of DOI:10.1021/jp0465175

A covalently tethered dyad containing the azulene (Az) and zinc tetraphenylporphyrin (ZnP) chromophores has been synthesized and its excited-state dynamics investigated, using the tether-substituted monochromophoric species as reference compounds. One pho

Full text of DOI:10.1021/jp0465175

10980
J. Phys. Chem. A 2004, 108, 10980-10988
Sequential Forward S2-S2 and Back S1-S1 (Cyclic) Energy Transfer in a Novel
Azulene-Zinc Porphyrin Dyad
Edwin K. L. Yeow,*,† Marcin Ziolek,‡ Jerzy Karolczak,‡, Sergey V. Shevyakov,§
Alfred E. Asato,§ Andrzej Maciejewski,‡,| and Ronald P. Steer*,†
Department of Chemistry, UniVersity of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5C9, Center for
Ultrafast Laser Spectroscopy, Adam Mickiewicz UniVersity, Umultowska 85, 61-614 Poznan, Poland, Faculty of
Physics, Adam Mickiewicz UniVersity, Umultowska 85, 61-614 Poznan, Poland, Faculty of Chemistry, Adam
Mickiewicz UniVersity, Grunwaldzka 6, 60-780 Poznan, Poland, and Department of Chemistry, UniVersity of
Hawaii, 2545, The Mall, Honolulu, Hawaii 96822
ReceiVed: August 3, 2004; In Final Form: September 23, 2004
A covalently tethered dyad containing the azulene (Az) and zinc tetraphenylporphyrin (ZnP) chromophores
has been synthesized and its excited-state dynamics investigated, using the tether-substituted monochromophoric
species as reference compounds. One photon excitation of the dyad at 270 nm results in selective population
of the S2 state of the azulene moiety, followed by near-quantitative electronic relaxation in the cycle S2(Az)-
S2(ZnP)-S1(ZnP)-S1(Az)-S0. Energy transfer from the S2 state of the azulene moiety to the S2 state of the
ZnP moiety is ultrafast (keet > 2 × 1012 s-1) and quantitative. The ZnP(S2) moiety subsequently undergoes
rapid (kic ) 3 × 1011 s-1), quantitative internal conversion to its S1 state. Thereafter, the excitation residing
on the S1 state of the ZnP is returned to the Az moiety via an efficient (ca. 90%) back S1-S1 energy transfer
process (keet ) 2.8 × 109 s-1). Ultimately the system returns to the electronic ground state via conical intersection
of the azulene S1 and S0 surfaces in ca. 1 ps. The cyclic interchromophoric energy transfer rates are nearly
the same in both acetonitrile and cyclohexane, suggesting that the conformation of the tether is similar in
both solvents. Fo¨rster theory is inadequate in explaining the efficient energy transfer dynamics, and other
processes such as the Dexter mechanism must be invoked.
1. Introduction
a zinc porphyrin.5 Due to significant spectral overlap between
the S2 emission spectrum of azulene and the S2 absorption
spectrum of zinc porphyrin, resonance S2-S2 energy transfer
from azulene to zinc porphyrin can take place. Furthermore, an
overlap between the S1 emission spectrum of zinc porphyrin
and the S1 absorption spectrum of azulene allows back S1-S1
energy transfer to occur after the S1 state of the zinc porphyrin
is produced by rapid internal conversion from its S2 level.
Recent advances in optoelectronics and molecular logic gates
have prompted great interest in the photophysics and photo-
chemistry of polyatomic molecules with relatively long-lived
highly excited electronic states.1 Azulene and metalloporphyrins
(e.g., zinc tetraphenylporphyrin), known to violate Kasha’s rule,
are photochemically stable and therefore act as potential
candidates for the design of efficient molecular logic devices.
Azulene exhibits relatively intense fluorescence from its S2 state,
while undergoing an ultrafast internal conversion from its S1 to
S0 states via conical intersection.2 On the other hand, many
diamagnetic metalloporphyrins display weak S2-S0 fluorescence
as a result of the modest energy gap (∼0.8 eV) between their
S2 and S1 states.3
Levine and co-workers have succinctly described the op-
portunity to utilize energy transfer between higher electronic
states to achieve more efficient molecular logic gates and
circuits.4 This arises from the need to use a greater number of
states per molecule (say, both S1 and S2 states) for logic
operations. We have previously proposed the design of an
efficient molecular logic circuit based on sequential forward
S2-S2 energy transfer and back S1-S1 energy transfer (i.e.,
cyclic energy transfer) for a system comprising an azulene and
S2-S2 energy transfer between azulene and zinc porphyrin
has been investigated in dichloromethane and cetyl trimethyl
ammonium bromide (CTAB) micelles.6 In dichloromethane, an
efficient transfer of energy, which cannot be rationalized using
Fo¨rster theory alone, is observed and was attributed to an
inhomogeneous distribution of the azulene donor molecules
surrounding the energy-accepting zinc porphyrins. Other mech-
anisms, such as short-range exchange interaction and higher
multipole interaction, were shown to be important. On the other
hand, the Fo¨rster mechanism was found to agree well with the
experimental energy transfer efficiency when azulene and zinc
porphyrin are encapsulated in CTAB micelles.
Herein, we report the synthesis of a covalently tethered
azulene-zinc porphyrin dyad (Az-ZnP) and determine its
photophysical properties in two solvents of different polarity,
acetonitrile and cyclohexane. The desired cyclic energy transfer
process is shown to occur with high efficiency. This study
provides useful insight into the factors that control the cyclic
energy transfer process and forms a platform for further studies
involving the incorporation of such species into molecular
devices.
* Corresponding authors. Fax: +1-306-966-4730. E-mail addresses:
eky341@duke.usask.ca (E. K. L. Yeow); ron.steer@usask.ca (R. P. Steer).
† University of Saskatchewan.
‡ Center for Ultrafast Laser Spectroscopy, Adam Mickiewicz University.
Faculty of Physics, Adam Mickiewicz University.
§ University of Hawaii.
| Faculty of Chemistry, Adam Mickiewicz University.
10.1021/jp0465175 CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/19/2004

Energy Transfer in Azulene-Zinc Porphyrin Dyad
J. Phys. Chem. A, Vol. 108, No. 50, 2004 10981
SCHEME 1a
spectrofluorimeters: a modified Edinburgh Instruments FL 9000
and a modified MPF-3 (Perkin-Elmer). The Edinburgh Instru-
ments spectrofluorimeter, with a laser as an excitation source,
ensured high intensity and monochromaticity of the excitation
beam, low levels of scattered light, and minimization of the
reabsorption effect.9 Emission quantum yields of the reference
compounds were obtained using the Edinburgh Instruments
spectrofluorimeter with quinine sulfate in 0.05 M aqueous
H2SO4 as the reference standard. The modifications introduced
into the MPF-3 spectrofluorimeter enabled single-photon count-
ing detection and data processing using a dual-photon counting
setup (Light-Scan). The MPF-3 spectrofluorimeter was also used
to measure the fluorescence excitation spectra.
The apparatus and procedures used for the time-resolved
emission measurements has been described in detail previ-
ously.10 Briefly, the train of 1-2 ps pulses from an argon ion-
pumped, tunable, mode-locked Ti:sapphire laser was pulse-
picked at a frequency of 4 MHz. Second and third harmonics
of these pulses were used as the excitation source for the
measurement of upper state fluorescence lifetimes by time-
correlated single-photon counting with an accuracy of ca. 300
fs. A graded, rotating neutral density filter was used to control
the intensity of the excitation beam, and a Fresnel double rhomb
was employed to rotate the plane of its polarization. Together
with a polarizer set to observe emission at the magic angle,
this arrangement eliminates rotational diffusion artifacts. The
detector is a Hamamatsu R3809U-05 microchannel plate pho-
tomultiplier.
a Reagents and conditions: (a) NaBH4, MeOH; (b) NaH, THF,
BrCH2C(O)Cl; (c) NaH, THF, acetyl chloride; (d) 3, Py, CH2Cl2; (e)
methyl bromoacetate, Py, CH2Cl2.
The equipment used for the transient absorption has previ-
ously been described in detail.11 The output of the laser system
(femtosecond titanium-sapphire) was set at a 1 kHz repetition
rate providing pulses of about 100 fs duration. The probe beam
passed through an optical delay line consisting of a retroreflector
mounted on a computer-controlled motorized translation stage
and was then converted to a white light continuum (in a 2 mm
rotating calcium fluoride plate), the diameter of which was 2-5
times smaller than that of the pumping beam. A grating
polychromator was used in conjunction with a thermoelectrically
cooled CCD camera to record the spectra. To improve the signal-
to-noise ratio, the transient absorption measurements were
performed in double-beam mode (probe and reference) with two
synchronized choppers in the pump and probe paths, which
substantially eliminated the influence of the laser intensity
fluctuations and, consequently, allowed measurements of much
lower values of the optical density changes. With this experi-
mental setup, the absorbance changes (∆A) can be measured
with an accuracy of (0.0005 in the 300-700 nm spectral range.
The thickness of the flowing sample was 2 mm. The pulse
energies for pump wavelengths of 270 and 400 nm were 2 and
10 µJ, respectively. All spectra were corrected for group velocity
dispersion effects according to the standard numerical scheme.12
The chirp of the white light continuum was obtained by
measuring two-photon absorption in a very thin (150 µm) BK7
glass plate. An additional contribution to dispersion, due to the
front window of the sample cell, was calculated from the
Sellmeier equation. The transient absorption signals originating
from the pure solvent were subtracted from the data collected.13
2. Experimental Section
2.1. Synthesis. Scheme 1 shows the synthetic route adopted
to synthesize the azulene-zinc porphyrin dyad 6 (Az-ZnP) and
the two reference compounds 4 (Az-ref) and 7 (ZnP-ref).
6-Formylazulene 1 was prepared according to the procedure of
Hafner starting from N-butyl-4-(1,3-dioxolan-2-yl)pyridinium
bromide.7 The reduction of aldehyde 1 was performed cleanly
by NaBH4 in MeOH at room temperature, and the resulting
alcohol 2 (yield ∼95%) was acylated with bromoacetyl chloride
in THF using NaH as a base to give the corresponding bromide
3 (80% yield). Reference compound 4 was obtained in the same
manner using acetyl chloride in the acetylation step (yield
∼96%).
Condensation of zinc tetrakis-5,10,15-triphenyl-20-(4-hy-
droxyphenyl)-21H,23H-porphyrin8 (5) with the bromo-derivative
of azulene 3 or with methyl bromoacetate under mild basic
conditions led to the desired dyad 6 (yield ∼75%) and the
reference compound 7 (yield ∼82%), respectively.
All final compounds were purified by flash chromatography
on Merck silica gel 60 (80-100 mesh) using CH2Cl2 as a
solvent. The NMR data of reference compounds 4 (Az-ref), 7
(ZnP-ref), and 6 (Az-ZnP dyad) are given in Table 1.
2.2. Spectroscopy. All measurements were performed at room
temperature in acetonitrile (Merck, HPLC grade) and cyclo-
hexane (Merck, HPLC grade). To achieve optimal experimental
signal-to-noise ratios, concentrations near the solubility limit
of the Az-ZnP dyad in these solvents were used. The
concentrations of all compounds were ca. 4 × 10-5 M in
acetonitrile and ca. 5 × 10-6 M in cyclohexane. No aggregate
formation was observed.
All our fits of the kinetic curves involved the temporal
instrumental function taking into regard the cell thickness and
hence the dispersion of the delay between the pump and the
probe, originating from different group velocities of pump and
probe pulses in the sample. The pump-probe cross-correlation
function, unaffected by dispersion, was determined based on
the two-photon absorption in the BK7 glass plate (excitation at
The stationary UV/vis absorption spectra were taken with a
UV-Vis-550 (Jasco) spectrophotometer. The steady-state fluo-
rescence emission spectra were recorded with two different

10982 J. Phys. Chem. A, Vol. 108, No. 50, 2004
Yeow et al.
TABLE 1: NMR Data for the Reference Compounds 4 (Az-ref) and 7 (ZnP-ref) and the Dyad 6 (Az-ZnP)
product
yield (%)
96
1H NMR, δ, J (Hz)a
4
2.16 (s, 3H, 6-CH2OC(O)CH3), 5.26 (s, 2H, 6-CH2OC(O)CH3), 7.20 (d, J ) 10.2, 2H, C5-H and C7-H),
7.40 (d, J ) 3.6, 2H, C1-H and C3-H), 7.92 (t, J ) 3.6, 1H, C2-H),
8.33 (d, J ) 10.2, 2H, C4-H and C8-H)
4.95 (s, 2H, -OCH2C(O)O-), 5.47 (s, 2H, -CH2C(O)O-),
7.20-7.32 (4H, C5-H, C7-H azulene and arom),
6
75
7.35 (d, J ) 3.9, 2H, C1-H and C3-H azulene), 7.68-7.84 (m, 9H, arom),
7.87 (t, J ) 3.9, 1H, C2-H azulene), 8.08-8.4 (10H, C4-H, C8-H azulene and arom),
8.86-9.06 (m, 8H, â-pyrrole)
7
82
3.95 (s, 3H, -OCH3), 4.94 (s, 2H, -OCH2C(O)-), 7.22-8.95 (27H, â-pyrrole and arom)
a
1H NMR spectra were recorded on Varian Mercury Plus 300 MHz with TMS as internal standard and CDCl3 as solvent.
400 nm); its fwhm is 150 fs. The real instrumental function
used for the convolution with the kinetic exponential functions
was determined separately for each wavelength, according to
the theory presented in ref 14. It should be noted that the
dispersion in the sample cell is especially important when 270
nm pump pulses were used. For example, the delay between
pump and probe transit times after passage through the 2 mm
sample cell filled with acetonitrile is 540 fs for a probe
wavelength of 350 nm and 840 fs for a probe wavelength of
500 nm. This is much greater than the duration of the dispersion-
free pump-probe cross-correlation function (150 fs). In such
cases the temporal instrumental function is affected mostly by
the dispersion delay in the sample cell, not by the pulse duration,
and assumes a characteristic hatlike shape.14 Thus, the temporal
resolution in our transient absorption experiments was better
for 400 nm excitation (about 100 fs) than for 270 nm excitation
(about 500 fs).
3. Results and Discussion
3.1. Steady-State Measurements. The steady-state absorp-
tion and fluorescence spectra (excitation at 270 nm) of the
reference compounds Az-ref and ZnP-ref in acetonitrile are
shown in Figure 1, panels a and b. The maximum of the Soret
absorption band for ZnP-ref occurs at 422 nm, while the maxima
of the visible Q-bands are located at 557 and 597 nm. The
emission spectrum for ZnP-ref is characterized by band maxima
at 440 (S2 emission), 603, and 656 nm (S1 emission). The
absorption spectrum of Az-ref shows a broad S0-S1 absorption
band in the 500-700 nm region, a S0-S2 absorption band with
a maximum at 344 nm, and a strong S0-S3 absorption band
with two distinctive peaks at 282 and 277 nm. A fluorescence
maximum corresponding to emission from the S2 state of Az-
ref is observed at 380 nm.
Figure 1c shows the steady-state absorption and fluorescence
spectra (excitation at 270 nm) of the Az-ZnP dyad in
acetonitrile. The absorption spectrum is essentially a sum of
the spectra of the individual reference compounds (i.e., Az-ref
and ZnP-ref). This indicates the absence of a strong ground-
state electronic interaction between the two chromophores in
the dyad molecule. Upon excitation at 270 nm, the fluorescence
spectrum of the Az-ZnP dyad displays emission maxima at
380, 440, 603, and 656 nm. The emission band at 380 nm is
ascribed to S2 emission from the Az moiety, and the rest of the
bands to either S2 (λem ) 440 nm) or S1 (λem ) 603 and 656
nm) emission from the ZnP moiety. The fluorescence spectra
of Az-ref, ZnP-ref, and Az-ZnP presented in Figure 1 were
obtained from sample solutions of similar concentrations (ca. 4
× 10-5 M), and are on the same relative scale.
Figure 1. The normalized absorption (;) and emission (‚‚‚) spectra
for (a) Az-ref, (b) ZnP-ref, and (c) Az-ZnP dyad in acetonitrile. Abs
is the absorption band, while em is the emission band. The absorption
spectra of Az-ref and ZnP ref are independently normalized and the
emission intensity scale is the same in all three panels, a, b, and c.
band. Furthermore, the intensities at maxima 603 and 656 nm
are the same when measured with reabsorption minimized on
the Edinburgh Instruments spectrofluorimeter. Regretably,
quantitative results could not be obtained from this instrument
due to laser photolysis of the dyad molecules. The emission
spectra and fluorescence decay times (vide infra) of the solutions
that exhibit photodecomposition contain features that are similar
to those of the individual reference compounds, suggesting that
the dyad undergoes simple scission of the tether connecting the
two chromophores. A semiquantitative analysis suggests that
the quantum yield of photodecomposition is small, but the rate
of decomposition is nevertheless significant with the laser
excitation system used on this instrument. Fortunately, accurate
The emission intensities of bands due to the ZnP chromophore
in either the ZnP-ref or dyad are diminished by the reabsorption
effect. This is especially true of the S2 emission band at 440
nm, which overlaps strongly with the strong Soret absorption

Energy Transfer in Azulene-Zinc Porphyrin Dyad
J. Phys. Chem. A, Vol. 108, No. 50, 2004 10983
relative emission intensity and quantum yield measurements,
comparing the dyad and the reference compounds, could be
obtained using the MPF-3 spectrofluorimeter (polychromatic
source) in which the rate of photodecomposition was small. In
these experiments, fluorescence reabsorption occurs to the same
extent when solutions of similar concentration are employed,
so that accurate relative intensities can still be measured.
The molar extinction coefficients of Az-ref and ZnP-ref in
acetonitrile at 270 nm were determined to be 3.82 × 104 and
1.62 × 104 M-1 cm-1, respectively. At the same concentrations
(ca. 4 × 10-5 M), the Az and ZnP moieties of the dyad absorb
a smaller fraction of the incident light intensity (at 270 nm)
than their respective reference compounds. The fraction of
intensity of the light absorbed by the individual chromophores,
X, of the dyad is obtained using the following equation:
Figure 2. Excitation spectra (λem ) 660 nm) of ZnP-ref (‚‚‚) and Az-
ZnP (;) in acetonitrile.
S2 energy transfer from Az to ZnP, followed by an S2-S1
internal conversion within the ZnP moiety, before a back S1-
S1 energy transfer from ZnP to Az occurs.
AX
Further evidence of energy transfer from Az to ZnP in the
dyad is provided from the fluorescence excitation spectra of
Az-ZnP and ZnP-ref (see Figure 2). The excitation spectrum
of ZnP-ref at λem ) 660 nm is characterized by principal bands
with maxima at 598, 558, and 422 nm and a relatively weak
and broad band between 250 and 350 nm associated with upper
(n > 2) excited states of ZnP. The normalized fluorescence
excitation spectrum of Az-ZnP (Figure 2), monitored at λem
) 660 nm, reveals bands arising from ZnP and an additional
well-defined band with a maximum at 280 nm. The latter is
attributed to the Az moiety, whereby excitation into the S3 state
of the Az leads to emission from the S1 state of the ZnP, which
can only be rationalized from the transfer of excitation energy
from Az to ZnP in the dyad. The band in the excitation spectrum
of the dyad associated with the S2 state of Az, near 344 nm, is
masked by the ZnP band and is therefore not apparent in the
excitation spectrum of Az-ZnP. The ratio of the band intensities
at 270 nm for Az-ZnP and ZnP-ref, after equalizing the incident
light intensity absorbed by the two ZnP chromophores, is ∼3.4
(cf. the ratio of the molar extinction coefficients of Az-ZnP
and ZnP-ref at 270 nm is also ∼3.4). This again suggests that
following rapid, efficient Az S3-S2 internal conversion, efficient
energy transfer (close to 100%) occurs from the S2 state of Az
to the S2 state of ZnP in the dyad. The excited molecule then
relaxes to the S1 state of the ZnP moiety, which emits at 660
nm.
Similar types of quenching behavior were also observed in
the nonpolar cyclohexane solvent. Upon excitation of Az-ZnP
at 270 nm, a drastic reduction in the S2 fluorescence intensity
of the dyad Az (band maximum at 340 nm) and a concomitant
increase in the S2 fluorescence intensity of the dyad ZnP (band
maximum at 430 nm) are observed when compared with their
respective reference compounds. The S2 emission quantum yield
of the dyad Az is reduced ∼120-fold as a result of energy
transfer to the S2 state of the ZnP moiety. The S1 emission
intensity of the dyad ZnP chromophore is quenched ∼2.3-fold
compared with the reference compound (maxima at 600 and
645 nm). The normalized fluorescence excitation spectrum of
Az-ZnP (measured at λem ) 642 nm) shows bands that are
identical to the reference ZnP and an extra well-defined band
at 280 nm arising from the S3 state excitation of Az. The ratio
of the intensity of the excitation spectra at 270 nm for Az-
ZnP and ZnP-ref is ∼3.6, which corresponds well with the ratio
of their molar extinction coefficients at the same absorption
wavelength. When the number of excited dyad ZnP molecules,
obtained from either direct excitation at 270 nm or energy
transfer from Az, is matched with those of the ZnP-ref, the S1
ZnP emission quantum yield of the dyad is found to be reduced
7.7-fold compared to the reference compound. This is in
Az+AZnP
)
FX )
(1 - 10-(A
)
(1)
AAz + AZnP
where X is either Az or ZnP, and AAz (ZnP) is the absorbance of
Az (ZnP) at 270 nm. For quantitative comparison purposes, the
emission band intensities associated with each of the chro-
mophores in Az-ZnP were corrected accordingly so that the
dyad chromophore and its reference compound absorb equal
incident light intensity.
Upon excitation at 270 nm, the fluorescence intensity of the
dyad Az (λem ) 380 nm) is greatly reduced compared with the
reference compound (cf. Figure 1). The S2 emission quantum
yield of Az in the dyad is ∼160 times smaller than Az-ref (ΦS -
2
(Az-ref) ) 0.032) in acetonitrile. The most likely process by
which the excited S2 state of the Az moiety of the dyad is
efficiently quenched is electronic energy transfer to the S2 state
of the ZnP chromophore. Indeed a ∼3.2-fold enhancement in
the S2 fluorescence intensity of the dyad ZnP (at 440 nm in
Figure 1c) is observed compared with ZnP-ref (Figure 1b).
The S1-S0 fluorescence, in the 550-750 nm range, of the
dyad ZnP (Figure 1c) is quenched with respect to that of the
ZnP-ref (Figure 1b). At λex ) 270 nm, the S1 emission intensity
of ZnP-ref is reduced 1.8-fold when connected to an Az
chromophore in the dyad. At the same sample concentrations,
a larger proportion (∼3.2 times from the emission intensity ratio
of S2 dyad to S2 ZnP-ref) of excited dyad Az-ZnP molecules
is created from both direct excitation (at 270 nm) and energy
transfer from the Az moiety compared with the ZnP-ref. When
considering an equivalent number of excited ZnP molecules,
the S1 ZnP emission quantum yield for the dyad is reduced 5.8-
fold (1.8 × 3.2) compared with the reference compound
(ΦS (ZnP-ref) ) 0.029). This is in agreement with the S1
1
emission quenching effects observed at λex ) 556 nm where
the S1 state of the dyad ZnP is exclusively excited, and no S2-
S2 energy transfer exists between the two chromophores. In this
case, a 5.6-fold reduction in the S1 fluorescence quantum yield
for the dyad ZnP is observed compared with ZnP-ref. A
deactivation pathway for the excited S1 state of ZnP in the dyad
involves S1-S1 energy transfer to Az. The electronic energy
transfer efficiency, E, calculated using
ΦS (dyad-ZnP)
1
E ) 1 -
(2)
ΦS (ZnP-ref)
1
where ΦS (dyad-ZnP) and ΦS (ZnP-ref) are the S1 emission
1
1
quantum yields of the dyad ZnP and ZnP-ref, respectively, is
83%. A cyclic energy transfer process is therefore shown to
occur in the dyad at λex ) 270 nm. This entails a forward S2-

10984 J. Phys. Chem. A, Vol. 108, No. 50, 2004
Yeow et al.
TABLE 2: Fluorescence Decay Lifetimes for ZnP-ref,
Az-ref, and Az-ZnP Obtained at Two Different Excitation
a
Wavelengths, λEx
λex ) 400 nm
em (nm) τ (ps)
λex ) 270 nm
λem (nm) τ (ps)
λ
ZnP-ref
424
600, 660
1.7
1600
428
603
380
603
1.9
1600
690
290 (0.7),
1600 (0.3)
Az-ref
Az-ZnP
660, 660
290 (0.7),
1600 (0.3)
a The numbers in parentheses are the fractions of the total emission
having the corresponding lifetime, τ.
Figure 3. Transient absorption spectra of ZnP-ref (‚‚‚) and Az-ZnP
(;) in acetonitrile with 270 nm excitation and a 7 ps probe delay.
agreement with the results obtained at λex ) 546 nm where only
the S1 state of ZnP is excited. The S1-S1 back energy transfer
efficiency between ZnP and Az is calculated to be 87% from
eq 2.
3.2. Time-Resolved Fluorescence Measurements. To in-
vestigate the excited-state dynamics, the time-resolved fluores-
cence lifetimes of Az-ZnP and the reference compounds were
evaluated at several excitation and emission wavelengths in
acetonitrile. The single-exponential fluorescence decay of Az-
ref (λex ) 270 nm and λem ) 380 nm) yielded a lifetime of 690
ps. The S1 fluorescence decay profiles of ZnP-ref measured at
At λex ) 400 nm, the Az-ZnP dyad displays dual fluorescence
lifetimes of 290 ps and 1.6 ns when emission was monitored at
600 nm (or 660 nm), suggesting no change in the energy transfer
rate at the two excitation wavelengths (i.e., 270 and 400 nm).
In nonpolar cyclohexane, ZnP-ref decays with a monoexpo-
nential lifetime of 1.9 ns (λex ) 270 nm and λem ) 639 nm).
The forward S2-S2 energy transfer rate in the dyad is beyond
the time resolution of our instrument and was therefore not
measurable. The S1 fluorescence decay of dyad-ZnP (λex ) 270
nm and λem ) 639 nm) is fitted satisfactorily to a double-
exponential decay function with lifetimes of 370 ps and 1.9 ns.
The former is assigned to the quenched lifetime of intact ZnP
in the dyad molecule, while the latter is again attributed to
emission from free ZnP produced from laser photolysis of the
dyad. The back S1-S1 energy transfer efficiency from ZnP to
Az in the dyad is 81%, and the transfer rate is 2.2 × 109 s-1
(eq 3). This energy transfer rate is only slightly smaller than
the one in acetonitrile and implies little solvent dependence on
the keet values. The lack of solvent dependence suggests that
the conformation of the dyad is similar in the two solvents.
3.3. Transient Absorption Measurements. The transient
absorption spectrum of Az-ref recorded by exciting the com-
pound at 270 nm was too weak to provide an analyzable signal.
This stems from the small extinction coefficients of the transient
species of azulene. In this case, the S2-S2 energy transfer
process can be detected as a potential rise time in the ZnP
transient spectrum, provided the transfer rate is slower than the
temporal resolution of the instrument. Though Foggi et al.15
have obtained femtosecond transient spectra of azulene using
sample concentrations ranging from 10-1 to 10-3 M, such high
concentrations could not be employed in this study due to the
limited solubility of the dyad in the solvents used and the need
to avoid dyad aggregation.
Figure 3 shows the transient absorption spectrum of ZnP-ref
recorded 7 ps after excitation at 270 nm (2 µJ/pulse). The
spectrum measured at a pump wavelength of 400 nm (10 µJ/
pulse) shows the exact same features. Only the excited S1 state
of ZnP-ref persists after 7 ps, and the structure of the spectrum
is characteristic of those previously reported for S1-Sn absorp-
tions of zinc tetraphenylporphyrin.16-18 In particular, the bleach-
ing of the Q(1,0) absorption band is seen at 560 nm, while the
bleaching of the Q(0,0) band at 600 nm suffers from low signal
intensity (Figure 3). However, this bleaching is more prominent
when a higher pump pulse energy at 400 nm is used. It should
be noted that in the 410-425 nm region the spectrum suffers
from diminished probe light intensity due to the high absorbance
of the ZnP-ref Soret band.
λem ) 600 and 660 nm exhibited a lifetime of 1.6 ns (λex
)
400 nm). At λex ) 270 nm, the S1 fluorescence lifetime of ZnP-
ref (λem ) 603 nm) is also 1.6 ns. The S2 fluorescence lifetimes
of ZnP-ref determined using 400 and 270 nm laser excitations
are 1.7 ps (λem ) 424 nm) and 1.9 ps (λem ) 428 nm),
respectively, and are equal within experimental error. The
various fluorescence lifetime values for the reference compounds
are summarized in Table 2.
The temporal fluorescence decay profile of Az-ZnP mea-
sured at λex ) 270 nm and λem ) 380 nm is dominated by
azulene-containing photoproducts, making reliable analysis
difficult in this particular case. The absence of a rise time
component in the S2 fluorescent profile of the ZnP chromophore
of the dyad (λex ) 270 nm and λem ) 428 nm), which is best
fitted to a monoexponential decay function with lifetime
component similar to that of ZnP-ref, implies that the intramo-
lecular S2-S2 electronic energy transfer rate between Az and
ZnP in the dyad is too fast to be detected by the time resolution
of our current experimental setup (i.e., <1 ps).
The S1-S1 energy transfer dynamics from ZnP to Az was
studied by exciting Az-ZnP at 270 nm and measuring the
emission decay at 603 nm. The decay profile was analyzed as
a sum of two exponentials that yielded a short lifetime
component of 290 ps (70%) and a long lifetime component of
1.6 ns (30%). The latter is attributed to emission from free ZnP
produced by laser photolysis of a fraction of the dyad. The short
decay time of 290 ps is assigned to emission from the excited
S1 state of intact dyad ZnP quenched by intramolecular energy
transfer to the S1 state of the Az moiety. The energy transfer
efficiency is calculated from eq 2 by replacing ΦS (dyad-ZnP)
1
with τS (dyad-ZnP) and ΦS (ZnP-ref) with τS (ZnP-ref) where
1
1
1
τS (dyad-ZnP) and τS (ZnP-ref) are the S1 emission lifetimes of
1
1
the dyad ZnP (290 ps) and ZnP-ref (1.6 ns), respectively. A
value of E ) 82% is obtained from the time-resolved measure-
ments, which is in excellent agreement with the steady-state
result. The rate of S1-S1 electronic energy transfer is calculated
to be 2.8 × 109 s-1 using the following expression:
The transient absorption profiles of ZnP-ref obtained at pump
wavelengths of 270 and 400 nm show similar kinetics when
probed at the same wavelength. Three representative profiles
1
1
keet
)
-
(3)
τS (dyad-ZnP) τS (ZnP-ref)
1
1

Energy Transfer in Azulene-Zinc Porphyrin Dyad
J. Phys. Chem. A, Vol. 108, No. 50, 2004 10985
which again corresponds to the S1 decay of ZnP-ref. This
suggests that the extinction coefficients for S2-Sn and S1-Sn
transitions are similar at this probe wavelength, and a short
lifetime component corresponding to the S2-S1 internal conver-
sion is therefore not observed.
Transient absorption measurements were also performed on
the dyad Az-ZnP in acetonitrile, and the spectrum at 7 ps after
excitation with pump wavelength 270 nm is shown in Figure
3. We note from Figure 3 that at similar sample concentrations
(ca. 4 × 10-5 M), the signal intensity for Az-ZnP is larger
than that for ZnP-ref such that the average ratio at the maxima
of the positive bands is ∼2.3. This ratio value increases to 3.2
when the fraction of incident light intensity absorbed by the
dyad is corrected based on eq 1. This is also seen at different
delay times and is indicative of a larger number of populated
dyad ZnP S1 states compared to the reference compound (i.e.,
∼3.2 times, in agreement with steady-state measurement results
discussed in section 3.1) resulting from intramolecular S2-S2
energy transfer within Az-ZnP. We also noted that laser
photolysis of the dyad is significantly reduced by the use of a
flow cell in these transient absorption measurements.
The transient absorption kinetic profile for Az-ZnP at a pump
wavelength of 270 nm and a probe wavelength of 450 nm
(Figure 5a) requires 3.3 ps rise and ∼300 ps decay components
to adequately fit the data. When the probe wavelength is tuned
to 370 nm (Figure 5b), the kinetic profile displays a single-
exponential decay with a time constant of ∼300 ps. The 3.3 ps
time constant observed at probe wavelength 450 nm is ascribed
to the time taken for internal conversion to occur from the S2
state of the dyad ZnP, while the ∼300 ps time constant at both
probe wavelengths (370 and 450 nm) is attributed to a faster
(than for ZnP-ref) relaxation of the excited S1 state of dyad
ZnP via energy transfer to the S1 state of dyad Az. This is
consistent with the time-resolved fluorescence measurement
results discussed in section 3.2. We note that no temporal
components in these kinetic profiles of these transients can be
assigned to the rapid S2-S2 energy transfer process within the
dyad. This suggests that the energy transfer from the Az moiety
to the ZnP moiety occurs in a time scale shorter than the
temporal resolution of our transient absorption instrument (i.e.,
<500 fs).
The transient absorption kinetic profile of Az-ZnP at a pump
wavelength of 400 nm and a probe wavelength of 450 nm is
shown in Figure 5c. At this excitation wavelength, S2-S2 energy
transfer does not take place. The kinetic trace is described by a
short 2.3 ps time component and a long ∼300 ps time
component, consistent with the lifetime values obtained at a
pump wavelength of 270 nm and probing at the same wave-
length (Figure 5a). This confirms that when the dyad is excited
at 270 nm, an ultrafast forward S2-S2 energy transfer process
(<500 fs) occurs in the dyad, followed by a much slower S1-
S1 back energy transfer. The latter is independent of the
excitation wavelength and is shown to take place within
∼300 ps.
Figure 4. Transient absorption kinetics for ZnP-ref in acetonitrile: (a)
pump wavelength ) 270 nm and probe wavelength ) 450 nm; (b)
pump wavelength ) 400 nm and probe wavelength ) 450 nm; (c)
pump wavelength ) 400 nm and probe wavelength ) 370 nm. The
early portion of the transient is shown in greater detail in the inset.
recorded at probe wavelengths 450 and 370 nm are displayed
in Figure 4. In each case, only the first 100 ps of the profile is
shown in detail. The transient absorption trace at a pump
wavelength of 270 nm and probe wavelength of 450 nm (Figure
4a) shows a rise characterized by a 3.0 ps time constant; the
long-lived decay component (not shown) is arbitrarily fixed to
the fluorescence lifetime of the S1 state of ZnP-ref (i.e., 1.6
ns). The transient absorption trace at pump wavelength 400 nm
and probe wavelength 450 nm (Figure 4b) displays similar
features with a rise time of 2.5 ps, followed by a slow decay,
which is again fixed to 1.6 ns. The short component seen in
Figure 4, panels a and b, tracks the changes in the transient
absorption as the molecule undergoes S2-S1 internal conversion,
and the two values obtained (2.5 and 3.0 ps) are within
experimental error. This is consistent with the S2 emission
lifetime value measured using time-resolved fluorescence
measurements (see section 3.2).
The transient absorption spectra of the Az-ZnP dyad and
ZnP-ref in cyclohexane, measured with a pump wavelength of
270 nm, have similar features to those in acetonitrile. The
positive Az-ZnP signal is ca. 3.5 times larger than that of the
reference compound and supports the notion that extra dyad
ZnP chromophores are excited by energy transfer from the Az
moiety. In these experiments, in contrast to those in which
acetonitrile was used as a solvent, the lower solubility of the
dyad resulted in a significantly smaller transient absorption
signal. Quantitative kinetic fits were only obtained in the region
The time evolution profile of the transient absorption at a
probe wavelength of 370 nm following excitation with a 400
nm laser pulse (Figure 4c) does not change significantly in the
first few picoseconds. Instead it decays slowly over 1-2 ns,

10986 J. Phys. Chem. A, Vol. 108, No. 50, 2004
Yeow et al.
attributed to the S2-S1 internal conversion in cyclohexane and
the longer component to the back S1-S1 energy transfer process.
In cyclohexane, the forward S2-S2 energy transfer is again too
rapid (>2 × 1012 s-1) to be detected and occurs in a time scale
of <500 fs.
3.4. Energy Transfer Dynamics. The rate of singlet-singlet
energy transfer between the donor D and the acceptor A is often
written in the form derived by Fo¨rster:19
6
Ro
1
τD
kFeet
)
(4)
( )( )
R
where τD is the fluorescence lifetime of D in the absence of A,
R is the distance between D and A, and Ro is the critical distance
where the transfer probability equals the emission probability.
Ro is defined by
9000(ln 10)κ2ΦDJ(ν˜)
6
Ro
)
(5)
128π5Nη4
where ΦD is the fluorescence quantum yield of D, N is
Avogadro’s number, η is the refractive index of the solvent,
J(ν˜) is the spectral overlap integral of the emission of D with
the absorption of A, and κ2 is the orientation factor defined by
κ2 ) (cos θT - 3 cos θD cos θA)2
(6)
where θT is the angle between the transition moments of D and
A and θD (A) is the angle between the transition moments of D
(A) and the line connecting their centers. Mårtensson’s method20
2
was employed to calculate the average orientation factor, κ ,
for a system consisting of a chromophore of 4-fold symmetry
(i.e., the ZnP moiety). By assuming that the orientation of the
various transition moments of azulene are not significantly
affected by the tethering substituent, the S0-S1 and S0-S2
transitions are taken to be polarized along the short and long
axis of the azulene molecule, respectively. Using the known
geometries of ZnP and Az and the energy-minimized structure
of the tether connecting these two moieties (Hartree-Fock with
a 3-21G* basis set using Spartan ‘04, Wavefunction Inc.), the
calculated average orientation factors for the forward S2-S2 and
back S1-S1 energy transfer processes are 1.56 and 0.035,
respectively. The center-to-center separation of the chro-
mophores R is taken to be 14.4 Å.
Figure 5. Transient absorption kinetics for Az-ZnP in acetonitrile:
(a) pump wavelength ) 370 nm and probe wavelength ) 450 nm; (b)
pump wavelength ) 270 nm and probe wavelength ) 450 nm; (c)
pump wavelength ) 400 nm and probe wavelength ) 450 nm. The
early portion of the transient is shown in greater detail in the inset.
The rate of Fo¨rster S1-S1 energy transfer from dyad ZnP to
dyad Az, in acetonitrile, determined from eqs 4 and 5 (using
PhotochemCAD21) is 7.5 × 107 s-1 (3.9 × 10-15 cm6/mmol).
The experimental energy transfer rate (i.e., keet ) 2.8 × 109
s-1) is ca. 40 times faster than predicted assuming the Fo¨rster
mechanism. One simple rationale is that the conformation of
the active electronic energy transfer species is not the extended
one but rather one with R ≈ 8 Å that is independent of solvent
polarity. However, if the extended conformation is dominant,
then this suggests that other energy transfer mechanisms are
also operative in the dyad system. One such mechanism,
proposed by Dexter,22 involves the short-range orbital overlap
interaction between the energy donor and acceptor chro-
mophores, and the energy transfer rate is given by23
Figure 6. Transient absorption kinetics for Az-ZnP in cyclohexane
with pump wavelength ) 270 nm and probe wavelength ) 420 nm.
The early portion of the transient is shown in greater detail in the inset.
of the bleaching band where the signal was not affected by the
high Soret band absorbance of the ZnP chromophore. A
representative transient absorption kinetic profile of Az-ZnP
at a probe wavelength of 420 nm (presented in Figure 6) decays
with a time constant of 1.7 ps followed by a recovery with a
time constant of ∼300 ps. The 1.7 ps component is again
2π
p
kDeet
)
KJD exp(-2Re/L)
(7)
where L is the average Bohr radius and JD is the integral overlap
between donor fluorescence and acceptor absorption spectra.
The edge-to-edge displacement (Re) between the Az and ZnP

Energy Transfer in Azulene-Zinc Porphyrin Dyad
J. Phys. Chem. A, Vol. 108, No. 50, 2004 10987
moieties in the dyad is <7 Å, which means that they are close
enough to bring the Dexter mechanism into effect. Furthermore,
JD (1.6 × 10-4 cm) is relatively high, which encourages the
augmentation of the energy transfer rate by the Dexter mech-
anism. Another orbital-overlap-dependent interaction term in-
volving the coupling between ionic charge transfer configura-
tions and locally excited states (i.e., through-configuration
interaction) has also been shown to be important.24
When the separation between D and A approaches the size
of the chromophores, as is the case for Az-ZnP, higher order
Coulombic terms and the contributions described by the
distributed transition-monopole theory may also play a signifi-
cant role in accounting for the energy transfer rate.25,26 The
parameters required for the evaluation of these two mech-
anisms are often not easily extracted from measured spectro-
scopic data, and this renders any quantitative analysis difficult.
Superexchange through-bond-mediated energy transfer via the
spacer between the ZnP and Az moieties can also contribute
significantly to the energy transfer process. Its importance will
be further tested by using spacer groups of different length and
structure.
The Fo¨rster S2-S2 energy transfer rate from dyad Az to dyad
ZnP in acetonitrile is calculated (from PhotochemCAD) to be
2.2 × 1011 s-1 (J(ν˜) ) 1.0 × 10-13 cm6/mmol), while the actual
transfer rate is >2 × 1012 s-1. The Fo¨rster rate suggests that
the S2 lifetime of the quenched dyad Az is 4.4 ps, which cannot
be experimentally resolved from the S2 lifetime of the dyad ZnP.
However, the absence of a rise component of ca. 4 ps in the
time-resolved fluorescence profile of the S2 state of dyad ZnP,
and the absence of a ca. 4 ps component in the transient
absorption kinetic trace of Az-ZnP at pump wavelength 270
nm and probe wavelength 370 nm also strongly indicate that
Fo¨rster mechanism does not adequately describe the actual
energy transfer dynamics. The several mechanisms present in
the S1-S1 energy transfer process work in parallel and are also
in operation here. The Fo¨rster critical distance in eq 4 does not
change significantly in the two solvents employed. For the S2-
S2 energy transfer process, Ro is 33.5 Å in both acetonitrile and
cyclohexane, and for the S1-S1 energy transfer process, Ro is
10.1 Å in acetonitrile and 9.3 Å in cyclohexane (kFeet ≈ 4 × 107
s-1). These similarities in the Ro values, together with the similar
dyad relaxation rates observed in cyclohexane and acetonitrile,
suggest that the conformation of the dyad is independent of the
polarity of the aprotic solvent and that the same relaxation
mechanism is operative in both solvents. However, in both
solvents, the rapid energy transfer rates observed cannot be
rationalized using the Fo¨rster mechanism, and it is therefore
necessary to invoke other processes (e.g., the Dexter mechanism)
to account for the discrepancy.
Figure 7. Energy levels (nts) of the Az and ZnP moieties in the Az-
ZnP dyad and the major EET and decay pathways.
term (Ne2/(ꢀR)) containing the relative permittivity parameter
(ꢀ) is 0.03 eV, and the -∆G° value evaluated from eq 8 is
therefore 1.25 eV in acetonitrile. A fast electron-transfer reaction
is known to take place in the activationless area when -∆G° is
close to the total reorganization energy λ (i.e., -∆G° ≈ λ). For
rigid molecules (e.g., porphyrins) in polar solvents, the λ value
is often smaller than 1 eV. Therefore, the electron-transfer
process from Az to ZnP in the dyad should fall in the Marcus
inverted region (-∆G° > λ) and thus would be too slow to
account for the efficient quenching of the S2 state of dyad Az.
Makinoshima et al.27 and Kesti et al.28 have also reported slow
electron-transfer procesess (inverted region) from either S2
azulene or S2 zinc porphyrin to C60, based on the calculated
∆G° values. In both studies, a purely charge-separation mech-
anism was found to be inadequate to explain the quenching
dynamics of the donor molecule.
The driving force for electron transfer from S1 ZnP to Az in
acetonitrile is calculated to be ∆G° ) 0.32 eV, using the
parameters Eox ) 0.80 V for ZnP-ref,29 Ered ) -1.65 V for
Az-ref,30 and Eoo ) 2.1 eV for the energy of the S1 ZnP-ref.
The positive ∆G° indicates that electron transfer from S1 dyad
ZnP to Az is not thermodynamically feasible.
The ∆G° values for electron transfer in cyclohexane are
evaluated using eq 8 with a correction term included to account
for the different dielectric constant of cyclohexane from the
reference solvent (acetonitrile). The positive ∆G° evaluated for
charge separation between S2 Az and ZnP and between S1 ZnP
and Az are 0.43 and 2.0 eV, respectively. Therefore, in nonpolar
cyclohexane, intramolecular electron-transfer cannot take place
from either the S2 state of dyad Az or the S1 state of dyad ZnP.
This complements the finding that the observed excited state
decay rates are independent of solvent polarity and are attribut-
able exclusively to electronic energy transfer.
The possible role of electron transfer in the quenching of the
S2 and S1 states of dyad Az and ZnP, respectively, in acetonitrile
can be assessed by calculating the Gibbs free energy change
(∆G°) involved in the charge separation process:
4. Conclusion
We have investigated the dynamics of cyclic energy transfer
in a covalently tethered azulene-zinc porphyrin dyad. Upon
excitation of the S2 state of the Az moiety, ultrafast energy
transfer (keet > 2 × 1012 s-1) to the S2 state of the ZnP moiety
takes place in acetonitrile (see Figure 7). The ZnP in its S2 state
subsequently undergoes rapid internal conversion to its S1 state.
Thereafter, the excitation residing on the S1 state of the ZnP is
returned to the Az via an efficient back S1-S1 energy transfer
(keet ) 2.8 × 109 s-1). The cyclic energy transfer rates (S2-S2
and S1-S1 rates are >2 × 1012 and 2.2 × 109 s-1, respectively)
do not vary significantly in cyclohexane. In both solvents,
Fo¨rster theory is inadequate in explaining the efficient energy
∆G° ) Eox - Ered - Eoo - (Ne2/(ꢀR))
(8)
We first examine the electron-transfer process from Az to ZnP
in the dyad. The redox potentials of azulene and zinc tetraphe-
nylporphyrin in acetonitrile were assigned to Az-ref and ZnP-
ref, respectively, without loss of generality. The oxidation
potential of Az-ref is thus Eox ) 0.88 V, while the reduction
potential of ZnP-ref is Ered ) -1.30 V.6 The energy of the Az-
ref S2 state (Eoo ) 3.4 eV) is estimated by averaging the energies
of the absorption and corresponding emission maxima. The work

10988 J. Phys. Chem. A, Vol. 108, No. 50, 2004
Yeow et al.
(9) Karolczak, J.; Kowalska, D.; Lukaszewicz, A.; Maciejewski, A.;
Steer, R. P. J. Phys. Chem. A 2004, 108, 4570.
transfer dynamics, and other processes such as the Dexter
mechanism must be invoked.
(10) Karolczak, J.; Kubicki, J.; Komar, D.; Wrozowa, T.; Dobek, K.;
Ciesielska, B.; Maciejewski, A. Chem. Phys. Lett. 2001, 344, 154.
(11) Maciejewski, A.; Naskrecki, R.; Lorenc, M.; Ziolek, M.; Karolczak,
J.; Kubicki, J.; Matysiak, M.; Szymanski, M. J. Mol. Struct. 2000, 555, 1.
(12) Nakayama T.; Amijima Y.; Ibuki K.; Hamanoue K. ReV. Sci.
Instrum. 1997, 68, 436.
Acknowledgment. We thank James Hutchison and Professor
Ken Ghiggino (University of Melbourne) for doing preliminary
experiments on this system. Spectroscopic measurements were
performed at the Center for Ultrafast Laser Spectroscopy, A.
Mickiewicz University, Poznan, Poland. R.P.S. and E.K.L.Y.
thank the Natural Sciences and Engineering Research Council
of Canada for financial support. A.M., J.K., and M.Z. thank
the State Committee for Scientific Research, Poland, Project 4
T09A 166 24, for financial support.
(13) Lorenc, M.; Ziolek, M.; Naskre¸cki, R.; Karolczak, J.; Kubicki, J.;
Maciejewski, A. Appl. Phys. B 2002, 74, 19.
(14) Ziolek, M., Lorenc, M.; Naskrecki, R. Appl. Phys. B 2001, 72, 843.
(15) Foggi, P.; Neuwahl, F. V. R.; Moroni, L.; Salvi, P. R. J. Phys.
Chem. A 2003, 107, 1689.
(16) Rodriguez, J.; Kirmaier, C.; Holten D. J. Am. Chem. Soc. 1989,
111, 6500.
(17) Hayes, R. T.; Walsh, C. J.; Wasielewski, M. R. J. Phys. Chem. A
2004, 108, 2375.
(18) Enescu, M.; Steenkeste, K.; Tfibel, F.; Fontaine-Aupart, M.-P. Phys.
Chem. Chem. Phys. 2002, 4, 6092.
(19) Fo¨rster, Th. Discuss. Faraday Soc. 1959, 27, 7.
(20) Mårtensson, J. Chem. Phys. Lett. 1994, 229, 449.
(21) Du, H.; Fuh, R.-C. A.; Li, J.; Corkan, L. A.; Lindsey, J. S.
Photochem. Photobiol. 1998, 68, 141.
References and Notes
(1) Burdzinski, G.; Kubicki, J.; Maciejewski, A.; Steer, R. P.; Velate,
S.; Yeow, E. K. L. In Molecular and Supramolecular Photochemistry -
Organic Photochemistry Part II; Ramamurthy, V., Schanze, K. S., Eds.;
Marcel Dekker: New York, 2004, in press.
(2) Te´treault, N.; Muthyala, R. S.; Liu, R. H. S.; Steer, R. P. J. Phys.
Chem. A 1999, 103, 2524.
(3) Kobayashi, H.; Kaizu, Y. In Porphyrins: Excited States and
Dynamics; Gouterman, M., Rentzepis, P. M., Straub, K. D., Eds.; American
Chemical Society: Washington, DC, 1986; p 105.
(4) Remacle, F.; Speiser, S.; Levine, R. D. J. Phys. Chem. B 2001,
105, 5589.
(22) Dexter, D. L. J. Chem. Phys. 1953, 21, 836.
(23) Speiser, S. Chem. ReV. 1996, 96, 1953.
(24) Yeow, E. K. L.; Ghiggino, K. P. J. Phys. Chem. A 2000, 104, 5825.
(25) Scholes, G. D.; Ghiggino, K. P. In AdVances in Multiphoton
Processes and Spectroscopy; Lin, S. H., Villaeys, A. A., Fujimura, Y., Eds.;
World Scientific: Singapore, 1996; Vol. 10, p 95.
(26) Karvarnos, G. J. Fundamentals of Photoinduced Electron Transfer;
VCH: New York, 1993.
(5) Yeow, E. K. L.; Steer, R. P. Chem. Phys. Lett. 2003, 377, 391.
(6) Yeow, E. K. L.; Steer, R. P. Phys. Chem. Chem. Phys. 2003, 5,
97.
(27) Makinoshima, T.; Fujitsuka, M.; Sasaki, M.; Araki, Y.; Ito, O.;
Ito, S.; Morita, N. J. Phys. Chem. A 2004, 108, 368.
(28) Kesti, T.; Tkachenko, N.; Yamada, H.; Imahori, H.; Fukuzumi, S.;
Lemmetyinen, H. Photochem. Photobiol. Sci. 2003, 2, 251.
(29) Morandeira, A.; Engeli, L.; Vauthey, E. J. Phys. Chem. A 2002,
106, 4833.
(7) (a) Hafner, K.; Bernhard, C. Angew. Chem. 1957, 69, 393. (b)
Estdale, S. E.; Bzettle, R.; Dunmur, D. A.; Marson, C. M. J. Mater. Chem.
1997, 7 (3), 391.
(8) (a) Alder A. D.; Longo F. R.; Finarelli J. D. J. Org. Chem. 1967,
32, 476. (b) Little R. G.; Anton J. A.; Loach P. A.; Ibers J. A. J. Heterocycl.
Chem. 1975, 12 (2), 343. (c) Semeikin A. S.; Koyfman O. I.; Berezin B.
D. Khim. Heterotz. Soed. (Russ.) 1982, 10, 1354.
(30) Muller, P.-A.; Vauthey, E. J. Phys. Chem. A 2001, 105, 5994.