Femtosecond dynamics of metalloporphyrins: Electron transfer and energy redistribution

Article abstract of DOI:10.1016/S0009-2614(98)00753-2

Femtosecond-resolved, real-time dynamics of the low-lying excited state ²Q(π,π *) of cobalt(II) tetraphenylporphyrin are reported in different solvents of varying polarity and structure. The dependence of the dynamics on the probing wavelength and on the metal and porphyrin substitution was also investigated. The decay in the π system of the porphyrin occurs with a lifetime of 220 fs in benzene, while a second process with a lifetime of 1.8 ps is consistent with a proposed model for intramolecular charge separation: porphyrin a_2u(π)→Co(d_z²). The dynamics of the charge transfer state is influenced by the solvents' polarity and coordination ability.

Full text of DOI:10.1016/S0009-2614(98)00753-2

21 August 1998
Ž
.
Chemical Physics Letters 293 1998 1–8
Femtosecond dynamics of metalloporphyrins: electron transfer
and energy redistribution
)
H.Z. Yu, J.S. Baskin, B. Steiger, C.Z. Wan, F.C. Anson, A.H. Zewail
Laboratory for Molecular Sciences, Arthur Amos Noyes Laboratory of Chemical Physics, California Institute of Technology,
Pasadena, CA 91125, USA
Received 25 May 1998
Abstract
2
)
Ž
.
Ž .
Femtosecond-resolved, real-time dynamics of the low-lying excited state Q p,p of cobalt II tetraphenylporphyrin
are reported in different solvents of varying polarity and structure. The dependence of the dynamics on the probing
wavelength and on the metal and porphyrin substitution was also investigated. The decay in the p system of the porphyrin
occurs with a lifetime of 220 fs in benzene, while a second process with a lifetime of 1.8 ps is consistent with a proposed
Ž
.
Ž
.
model for intramolecular charge separation: porphyrin a2u p ™Co dz2 . The dynamics of the charge transfer state is
influenced by the solvents’ polarity and coordination ability. q 1998 Elsevier Science B.V. All rights reserved.
1. Introduction
presently available evidence, while reasonably com-
pelling, is entirely circumstantial. With the objective
of obtaining direct information about the dynamics
of intramolecular electron transfer within cobalt por-
phyrins, we undertook an investigation of the relax-
ation dynamics of electronically excited states on the
femtosecond to picosecond time scale.
The results of our studies revealed the presence of
femtosecond relaxation dynamics that were unob-
servable in previous studies of the same or similar
It has been shown in a series of recent investiga-
tions that cobalt porphyrin molecules modified by
attachment of ruthenium or osmium complexes to
the periphery of the porphyrin ring, exhibit high
catalytic activity for the electroreduction of O2 di-
w x
rectly to H2O 1 . The ability to catalyze the reduc-
tion of O2 by four electrons is unusual for monomeric
metalloporphyrins. For this reason, the mechanism
by which these ruthenated or osmiumated cobalt
porphyrins accomplish this feat is of considerable
interest. Evidence has been offered that back-bond-
ing from the attached ruthenium or osmium centers
into p orbitals of the porphyrin ring is responsible
for the observed enhancement in catalytic activity of
w
x
molecules 2–4 because they did not extend into the
first few picoseconds following the generation of the
excited states. The existence of these very rapid
relaxation pathways and a model for their origin are
the subjects of this report.
The ground-state absorption properties of por-
phyrin molecules have been well established both
w x
the modified cobalt porphyrins 1 . However, the
Ž .
experimentally and theoretically. Intense Soret B
)
bands around 420 nm and weaker Q-bands between
Corresponding author.
0009-2614r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 98 00753-2

(
)
2
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
TPPCoNO and TPPFeNO, by excitation to the B-
w
x
band 17,18 .
The femtosecond experiments we report here were
carried out using two independently tunable laser
pulses, with excitation to the Q-band, in order to
avoid complication from additional relaxation path-
ways. We were thereby able to examine the effect of
solvent polarity, structure, and the binding to the
metal center. We also examined the free base and
zinc porphyrins for the sake of comparison.
In the following sections, we describe our experi-
mental apparatus and the new results. We then dis-
cuss the results in terms of a model for the energy
redistribution and electron transfer pathways. The
Letter is concluded in Section 5.
Scheme 1.
)
Ž
.
500 and 600 nm, interpreted as p,p in origin, are
shown in the characteristic spectra of porphyrin
w x
molecules 5 , and these features have been success-
fully treated with Gouterman’s four-orbital model
2. Experimental section
w x
6 . The absorption spectra of the specific
Ž
.
5,10,15,20-tetraphenylporphyrins TPP investigated
2.1. Apparatus and procedures
Ž
.
in this work Scheme 1 , have also been reported
w
x
7–9 .
Transient absorption experiments were performed
in the Laboratory for Molecular Sciences facility at
During the past several years, ultrafast time-re-
Ž
.
solved spectroscopy studies have been performed on
a number transition metal porphyrins in order to
clarify the excited-state dynamics of these molecules.
Caltech Fig. 1 . An amplified titanium–sapphire
laser system produces 80 fs pulses with pulse ener-
gies of 2 mJ at 800 nm and 1 kHz repetition rate.
The output of this system was split into two 1 mJ
w
x
w
x
Among them, Zn 10–12 and Ni 13–15 porphyrins
are two of the most extensively investigated systems.
For Co porphyrin, the first work concerning the
optical properties of excited states was done by
Ž
Dzhagarov et al. in the end of the 1970s 10 ps
.
pulses . That work reported a -10 ps lifetime for
w x
the excited state of Co porphyrin 2 . Picosecond
Ž
.
kinetic studies 35 ps pulses of Co porphyrins in
different solvents have been performed by Tait et al.
w
x
3,4 . The transient absorption spectra they obtained
Ž .
suggest that the excited-state relaxation of Co II
2
)
Ž
.
porphyrins proceeds via the tripdoublet, T p,p
2Ž
and the p,d charge-transfer CT states.
.
Ž
.
In 1993, Loppnow et al., using 4 ps pulses,
reported a 12"4 ps lifetime for excited states of
Ž .
Ž
.
both Co II octaethylporphyrin OEP and a class of
donor-appended CoOEP molecules from ground-state
recovery. They suggested that the observed transient
q
Ž .
is not the Co I P CT state, but a dd excited state
Fig. 1. Experimental layout of femtosecond transient absorption
apparatus. OPA: optical parametric amplifier; p: prism polarizers;
lr2: half-wave plate; c: chopper; f: optical filters; l: monochro-
mator.
w
x
16 . Recently, Morlino et al. carried out a sub-pico-
second investigation of photoinduced deligation in
two transition metalloporphyrin-nitrosyl complexes,

(
)
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
3
Ž
.
pulse trains which were used to pump two optical
exp ytrti , using the minimum number of compo-
nents required to leave no systematic deviation of the
residual, except possibly near the rise of the signal,
attributable to the solvent response. In these cases
the data fits were confirmed by starting at a point
beyond the range of the measured solvent response.
When fitting the rise, convolution with a Gaussian
response function was included in the fitting proce-
dure. The reported uncertainties are estimates based
on reproducibility of the fit parameters from fitting a
number of independently measured transients.
Ž
.
parametric amplifiers OPA , providing pump and
probe pulses with independent wavelength tunability.
The pump pulse passed through a computer-con-
trolled optical delay line and a chopper before reach-
ing the sample cell. Pump and probe beams were
overlapped in the sample either by collinear recom-
bination at a beam splitter, or by crossing at a small
Ž
.
angle ;38 . Typical pulse energies at the sample,
with beam diameters of ;1 mm, were 3 mJ for the
pump and 0.5 mJ for the probe.
After traversing the sample, the probe pulse was
passed through a monochromator for wavelength
discrimination, and its energy measured by a photo-
diode. This value was normalized to the probe en-
ergy measured before the sample cell by a second
photodiode. The ratio between normalized transmit-
ted probe energies in the presence and absence of
excitation of the sample by the pump pulse formed
the basis of the D A transients, recorded as a function
of the time delay between passage of the pump and
probe pulses. The linear polarization of the probe
beam was set at an angle of 54.78 with respect to that
of the pump beam, in order to eliminate the influence
of molecular reorientation on the observed dynamics.
The pump pulse was tuned to the Q-band absorp-
Absorption spectra characterization of the sample
was performed before and after each sample run to
ensure the quality of the sample throughout the
course of the measurement. The absorption spectra
observed in our study for porphyrin molecules are
w
x
identical to those reported before 7–9 . Comparison
of results for samples with or without Ar bubbling
showed no noticeable effects that could be attributed
to oxygen binding or quenching in the present exper-
iments.
2.2. Materials
The molecule 5,10,15,20-Tetraphenylporphyrin
Ž
.
w x
H2TPP was prepared according to Adler et al. 19 .
Ž
Ž
.
tion of the porphyrin systems under study 530–540
5,10,15,20-T etra 4-cyanophenyl porphyrin
.
Ž
.
w x
nm . The porphyrin concentrations were adjusted in
H2TCNPP was available from previous studies 20 .
order to obtain an absorbance near 0.5 in the Q-band
region. To minimize the transient signal from the
pure solvent, experiments to investigate sub-picosec-
ond dynamics were performed in fused silica cells
with 1 mm sample pathlengths. In this way, contami-
nation of the data by the solvent response could be
limited in magnitude and temporal extent to within
200 fs of the rise. The solvent effect was recorded in
conjunction with each sample measurement, since it
Both porphyrins were treated according to proce-
w
x
dures in the literature 21 to eliminate contamination
Ž
.
from the corresponding reduced porphyrin chlorin .
The molecules CoTPP, ZnTPP, and CoTCNPP were
prepared by reacting a 20% molar excess of cobalt
Ž
.
acetate tetrahydrate EM Science, GR or zinc ac-
Ž
etate dihydrate J.T. Baker, Crystals, Baker Analyzed
Reagent with 60 mg of the free porphyrin bases in 9
ml of refluxing N,N-dimethylformamide EM Sci-
ence, GR under argon for 30 min. 22 . The prod-
ucts were subsequently purified by chromatography
.
Ž
Ž
.
w x
depended sensitively on experimental conditions e.g.
.
alignment, focus, pulse energies . To use small quan-
Ž
tities of the samples and still avoid extended irradia-
tion of a fixed volume, the 1 mm cells were trans-
lated across the light path periodically during the
measurements. For the measurement of dynamics of
duration greater than a few picoseconds, cells with 5
mm pathlengths were sometimes used, and the sam-
ples were stirred.
on silica gel J.T. Baker, 40 mm Flash Chromatogra-
.
phy Packing using dichloromethane as eluent, and
crystallization from dichloromethane–cyclohexane.
Ž
.
Benzene EM Science, OmniSolv , cyclohexane
Ž
.
Ž
EM Science, OmniSolv , 1,2-dichloroethane EM
Science, OmniSolv DCE , and dichloromethane
EM Science, OmniSolv were passed through a
column of aluminum oxide Aldrich, activated, ba-
. Ž
.
Ž
.
Ž
The measured transients were analyzed by fitting
to a sum of exponential terms, D AsÝ Ai =
.
Ž
sic, Brockmann I before use. Pyridine EM Science,

(
)
4
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
.
Ž
OmniSolv and piperidine Aldrich, redistilled, 99.5
.
q% were used as received.
3. Results
Fig. 2 displays the femtosecond transient absorp-
tion profiles of CoTPP and ZnTPP in benzene and
the background response from pure benzene. The
profile of ZnTPP absorbance at 475 nm shows a
plateau following excitation at time zero by the 540
nm pump, while the profile of CoTPP presents a
rapid decay of two distinct components on the time
scale of a few picoseconds. H2TPP was also studied
for comparison, and, like ZnTPP, showed no short
time evolution. It should be noted that, although the
very short time behavior could be distorted by the
signal from the solvent, as shown in Fig. 2, it does
Fig. 3. Femtosecond transient absorbance of CoTPP in benzene
Ž
.
for two probe wavelengths. Note the different vertical scales.
Pump wavelengths: top, 530 nm; bottom 538 nm. Solid lines are
fits to the experimental data points.
not interfere with observation of the decay dynamics
of the transient absorption of either ZnTPP or CoTPP.
The total transient absorption signal of CoTPP in
Ž
.
benzene Figs. 3 and 4 can be well fit using three
exponential decay components with different life-
Ž
.
times and relative amplitudes Table 1 . The two
Ž
short lifetimes distinguishable in Fig. 2 220 fs and
1.8 ps are the main contributors, while a small but
.
Ž
.
discernible long lifetime component 16 ps is in the
tail of the whole decay profile. As shown in Fig. 3
and indicated in Table 1, the relative amplitudes of
the components can be altered by changing the probe
wavelength from 475 to 585 nm. In CoTPPrbe-
nzene, the amplitude of the second component in-
creases by a factor of ;3 relative to that of the fast
component, while that of the third component in-
creases about 20 times. In contrast, no changes have
been detected in the lifetimes, beyond the range of
experimental uncertainties indicated in Table 1. For
the system of CoTPP in DCE, the probe wavelength
Fig. 2. Femtosecond transient absorbance of two metallopor-
phyrins in benzene. Pump wavelength: 540 nm for ZnTPP and
530 nm for CoTPP. Probe wavelength: 475 nm. The transient of
the pure solvent was recorded immediately before the CoTPP
measurement under identical conditions.

(
)
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
5
probe wavelength is changed to 585 nm. No mea-
surements were carried out for CoTPPrpiperidine at
585 nm, while for CoTPP in pyridine, the lifetimes
measured at 585 nm were distinctly different than
Ž
those measured at 475 nm see footnote ‘c’ in Table
.
1 .
The properties of solvents, such as polarity and
coordinating propensity, may alter the absorption as
well as the decay characteristics of the excited states,
due to interactions of solvent and porphyrin
molecules. In particular, heterocyclic nitrogenous
ligands such as pyridine and piperidine can bind to
the cobalt center to form five andror six coordinated
species. A comparison of the decay dynamics at
Ž
.
Ž
.
fixed pump 530 nm and probe 475 nm wave-
lengths in different solvents is shown in Fig. 4. The
insert shows the long time scan of CoTPP in benzene
Ž
.
in a 5 mm cell , by which we could accurately
determine the third lifetime despite its relatively
small amplitude.
One may easily note in Fig. 4 the systematic
reduction in the amplitude of the second decay com-
ponent in the order of benzene)DCE)piperidine
)pyridine, as indicated in Table 1 for a probe
wavelength of 475 nm. The first two lifetimes are
affected only moderately, if at all, by the solvent, but
the third lifetime is clearly solvent dependent.
A preliminary investigation of another cobalt por-
Fig. 4. Solvent dependence of CoTPP transient absorbance. Pump
wavelength: 530 nm. Probe wavelength: 475 nm. Solid lines are
fits to the experimental data points. A 1 mm pathlength was used
for the data in the main figure, while that in the inset was
measured with a 5 mm pathlength.
Ž
.
dependence is similarly dramatic; the amplitude of
the first component dominates at 475 nm, but it
cannot be resolved in the decay profile when the
phyrin CoTCNPP in DCE was also performed. The
Ž
.
short time transient -17 ps did not differ percepti-
bly from that of CoTPP in DCE. Further studies of
Table 1
Decay lifetimes and relative amplitudes of the transient absorption dynamics of CoTPP
Solvent
Lifetimes
Relative amplitudesa
t1
t2
t3
Probe: 475 nm
Probe: 585 nm
Ž .
ps
Ž .
ps
Ž .
ps
A1
A2
A3
A1
A2
A3
benzene
DCE
piperidine
pyridinec
0.22"0.05
0.23"0.05
0.28"0.05
0.30"0.05
1.8"0.2
1.7"0.2
2.0"0.2
2.1"0.3
16"3
1
1
1
1
0.8"0.4
0.6"0.3
0.1"0.05
yd
0.03"0.01
0.02"0.01
0.04"0.02
0.02"0.01
1
2.2"0.8
1
0.5"0.2
0.2"0.1
23"3b
190"40
40"10
a The Ai values have been normalized to the amplitude of the shortest lifetime for each solvent and probe wavelength.
b
Ž
.
There is an additional very long 01 ns component, the fraction of which is 30 times smaller than that of t3.
c When probing at 585 nm in pyridine, we observed different lifetimes: t1 s1.5"0.2 ps and t2 s11"2 ps with A2rA1 s0.14"0.04.
d
The amplitude of A2 varied with pump wavelength.

(
)
6
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
Ž2
.
G
the dependence of the dynamic behavior on molecu-
lar structure are underway in our laboratory.
electronic configuration, the mixing of the state
resulting from the unpaired electron in the dz2 or-
bital, with the normal porphyrin singlet and triplet
1
3
Ž1
Ž
.
Ž
) ..
)
states A ground state, Q p,p , and T p,p
produces doublets and quartets. In other words, the
4. Discussion
ground state becomes a doublet and the lowest
)
Ž
.
p,p
excited states become the singdoublet
2
2
)
)
Ž
.
Ž
.
The observation of multiple exponential decay
components in CoTPP is most easily explained as
detection of transient absorption by multiple
excited-state species. For CoTPP in benzene and
DCE, this picture is supported by the fact that the
relative amplitudes of the components, but not their
lifetimes, display a pronounced probe wavelength
dependence. The nature of the spectral variation of
the absorption strength of each component provides
an important clue to identifying the origin of the
dynamics.
Q p,p , the tripdoublet T p,p , and the quar-
4
)
Ž
.
tet T p,p , with the quartet state calculated to be
lower in energy than the tripdoublet state by several
w
x
hundred wavenumbers 23 . Moreover, both iterative
Ž
.
extended Huckel IEH calculations and near IR
¨
absorption spectra indicate the existence of a
Ž .
Ž
. w x
low-lying CT state, porphyrin a2u p ™Co dz2 24 .
The picosecond experimental results of Tait et al.
for cobalt porphyrins in toluene revealed two distinct
time scales of transient absorption decay with differ-
ent spectral signatures following their 35 ps excita-
tion flash. The two lifetimes, which were -10 ps
Fig. 5 shows the energy diagram for CoTPP,
Ž
based on both experimental and theoretical data from
and 10 – 20 ps for C oO E P
O E P s
7
w
x
Ž .
.
Gouterman’s group 23,24 . Since Co II has a d
octaethylporphyrin , were attributed to the deactiva-
2
2
)
Ž
.
Ž
.
tion of the T p,p and CT p,d states, respec-
tively, consistent with the nature of the spectra. For
CoTPP, they report that their transients appear to
w
x
have even shorter lifetimes 3,4 .
These properties of Co porphyrins can be con-
trasted with those of the H2TPP and ZnTPP. There
are no unoccupied d orbitals in Zn, so in ZnTPP the
electronic states remain singlet and triplet as for the
free base porphyrin, and there is no low-lying CT
state in either one. The only excited state lower than
1
3
)
)
Ž
.
Ž .
Q p,p
is T p,p , and no fast relaxation is
w
x
expected upon excitation into the Q-band 10 . The
1
)
Ž
.
lifetimes for decay of Q p,p , primarily to
3
)
Ž
.
T p,p , in ZnTPP and H2TPP were measured to
be 1.9 and 14.7 ns, respectively, with fluorescence
w
x
quantum yields of several percent 25 , whereas,
experiments on Co porphyrins place an upper limit
on the fluorescence yield of 0.0005 upon excitation
w
x
of either the B- or Q-band 26 .
The present femtosecond results clearly show that
there is no ultrafast relaxation pathway from the
1
)
Ž
.
Q p,p state of ZnTPP, in contrast with the case
of CoTPP. The femtosecond process in CoTPP,
which could not be resolved in previous picosecond
experiments, represents the deactivation of the ini-
Fig. 5. Energy level diagram of CoTPP. Energy levels indicated
2
)
Ž
.
tially excited doublet state Q p,p . However, be-
cause of the presence of nearby T p,p
w
x
by black bars have been determined experimentally 24 , while
2
)
Ž
.
w
x
and
those indicated by hatched bars are calculated 23 .

(
)
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
7
4
2
)
Ž
.
states, we view the deactivation to the
T p,p
evidence that the primary three-step relaxation pro-
cess proposed for CoTPPrbenzene is modified by
ligand complexation. Further studies will be needed
to understand the role of deligation.
Third, our results support the notion of a charge
separation on the ultrafast time scale. This picture
Ž
.
CT p,d state as involving the coupled porphyrin p
system. It is known that in the case of low energy
Ž
w x.
gap see, e.g. Ref. 27 , this coupling leads to a
quasi-exponential dephasing of the initial state. Ac-
cordingly, the 220 fs component represents the time
for ‘equilibration’ in the p system and the 1.8 ps
component represents the rate at which the popula-
tion leaves to enter the CT state.
With this in mind, several points can be made.
First, taking into account the difference in absolute
absorption for the two probe wavelengths we investi-
gated, as seen in Fig. 3, the two short lifetime
w x
was invoked in the previous picosecond studies 3
w
x
but questioned by more recent work 16 , where the
involvement of a dd state was suggested. The latter
interpretation was based on the observation that the
picosecond decay of cobalt porphyrin was unaffected
by changes in the nature of attached donor groups. In
our study of CoTCNPP, we also did not observe
significant change in the femtosecond dynamics
compared with CoTPP, but we do not believe that
this behavior constitutes evidence for lack of in-
volvement of the CT state. The electrochemical data
components are stronger at 475 nm, as is expected
)
Ž
.
w x
for transient absorption of p,p excited states 3 .
Likewise, the enhanced absorption at 585 nm of the
third component, with a lifetime of 16 ps in benzene,
is consistent with the porphyrin-cation-radical-like
absorption expected for the CT state. This spectral
dependence and assignment conforms to the results
w
x
in Ref. 16 indicate very little dependence of half-
wave potentials on the nature of substituents, which
implies that they may not interact strongly with the
porphyrin system. Furthermore, to make a transition
to a dd state requires a two-event process which
2
w x
of Ref. 3 .
Second, in changing the solvent from benzene,
with a dielectric constant of 2.2, to DCE, with a
dielectric constant of 10.8, no pronounced change in
the lifetimes or relative amplitudes of the first two
components was found, but the lifetime of the
longest-lived component, which is believed to be the
transient absorption of the CT state, becomes longer.
This change is consistent with stabilization of the CT
state by the increase in solvent polarity. It is also
reasonable that the fastest relaxation process is not
sensitive to the solvent environment, since it is cen-
tered within the p system of the porphyrin.
should be less likely. Iterative extended Huckel cal-
culation predicts that in CoTPP, the dd absorption
¨
from dz to dx yy is at even higher energy than the
2
2
2
Soret band, and transitions between dxy ,yz,zx and dz
2
w
x
are of much lower energy than the CT 23 . It is
important to note, as pointed out by Holten and
w
x
colleagues 3,4 , that the transient absorption of a dd
state is expected to be similar to the porphyrin
ground state absorption, which was not the case for
Ž
.
their longer 10–20 ps lifetime component in CoOEP
w x
3 .
The third lifetimes which are measured when
probing at 475 nm become even longer in the strongly
We have made attempts to excite the CT state
directly and succeeded in observing the long life-
time. For CoTPP in benzene, the lifetime for 800 nm
femtosecond excitation was found to be consistent
with the 16"3 ps value obtained by exciting the
Q-band. The femtosecond transient could not be
resolved because the transition is extremely weak
and the solvent influence is unavoidable. Moreover,
Ž
.
coordinating solvents pyridine and piperidine .
These slow processes need not correspond to decay
of the CT state, however, since axial ligand binding
to CoTPP raises the possibility that photodissociation
Ž
.
of the ligand deligation complicates the overall
scheme for the deactivation of the excited states.
2
because the CT band overlaps with the Q and 2 T
w x
Based on their picosecond spectra, Tait et al. 4
Ž
.
assign a long-lived 130"20 ps transient absorp-
tion decay in CoTPPrpiperidine to the recombina-
tion of the ligand with the ground-state cobalt por-
phyrin following dissociation of the ligand in the CT
state. In the case of CoTPP in pyridine, the 11 ps
lifetime we observed when probing at 585 nm is
absorptions at 800 nm, the ;2 ps decay is also
present. Finally, our results are consistent with those
of Morlino et al, in particular their inference of 200
fs or less as the lifetime for 2Q to 2 T in the system of
w
x
TPPFeNO 18 , and their 2.1 ps lifetime reported for
1
w
x
S to CT in the system of TPPCoNO 17 .

(
)
8
H.Z. Yu et al.rChemical Physics Letters 293 1998 1–8
w x
6
Ž
.
M. Gouterman, G.H. Wagniere, J. Mol. Spectrosc. 11 1963
`
5. Conclusion
108.
w x
7
G.D. Dorough, J.R. Miller, F.M. Huennekens, J. Am. Chem.
In this Letter, we reported our first study of the
femtosecond dynamics of metalloporphyrins. Using
femtosecond-resolved transient absorption tech-
niques, the dynamics in cobalt porphyrins were ex-
Ž
.
Soc. 73 1951 4315.
w x
8
Ž
.
.
D.W. Thomas, A.E. Martell, J. Am. Chem. Soc. 78 1956
1338.
w x
9
Ž
D.J. Quimby, F.R. Longo, J. Am. Chem. Soc. 97 1975
2
5111.
)
Ž
.
amined upon excitation to the Q p,p state. The
decay profiles and effect of solvent and substitution
indicate that the p system relaxation occurs on the
time scale of 220–300 fs depending on the system,
while the formation of the CT state, which originates
as the result of intramolecular electron transfer from
w
w
w
x
10 J. Rodriguez, C. Kirmaier, D. Holten, J. Am. Chem. Soc. 111
Ž
.
1989 6501.
x
11 G.E. O’Keefe, G.J. Denton, E.J. Harvey, R.T. Phillips, R.H.
Ž
.
Friend, H.L. Anderson, J. Chem. Phys. 104 1996 805.
x
12 G.G. Gurzadyan, T.-H. Tran-Thi, T. Gustavsson, J. Chem.
Ž
.
Phys. 108 1998 385.
w
w
x
Ž
.
13 J. Rodriguez, D. Holten, J. Chem. Phys. 91 1989 3525.
Ž . Ž .
porphyrin a2u
p
to Co dz2 , occurs on the time
x
14 H.S. Eom, S.C. Jeoung, D. Kim, J.-H. Ha, Y.-R. Kim, J.
scale of about 2 ps. Experiments in different solvents
and at different probe wavelengths give further sup-
port for the model of charge separation. Further
studies on these and related systems are in progress.
Ž
.
Phys. Chem. A 101 1997 3661.
w
x
15 C.M. Drain, S. Gentemann, J.A. Roberts, N.Y. Nelson, C.J.
Medforth, S.L. Jia, M.C. Simpson, K.M. Smith, J. Fajer, J.A.
Ž
.
Shelnutt, D. Holten, J. Am. Chem. Soc. 120 1998 3781.
w
w
w
w
x
16 G.R. Loppnow, D. Melamed, A.R. Leheny, A.D. Hamilton,
Ž
.
T.G. Spiro, J. Phys. Chem. 97 1993 8969.
x
17 E.A. Morlino, L.A. Walker II, R.J. Sension, M.A.J. Rodgers,
Ž
.
J. Am. Chem. Soc. 117 1995 4429.
Acknowledgements
x
18 E.A. Morlino, M.A.J. Rodgers, J. Am. Chem. Soc. 118
Ž
.
1996 11798.
This work was supported by the National Science
x
19 A.D. Adler, F.R. Longo, J.D. Finarelli, J. Goldmacher, J.
Ž
.
Foundation Laboratory for Molecular Sciences .
Ž
Ž
.
.
Assour, L. Korsakoff, J. Org. Chem. 32 1967 476.
w
w
x
20 B. Steiger, F.C. Anson, Inorg. Chem. 33 1994 5767.
x
21 G.H. Barnett, M.F. Hudson, K.M. Smith, Tetrahedron Lett.
Ž
.
1973 2887.
References
w
x
22 A.D. Adler, F.R. Longo, F. Kampas, J. Kim, J. Inorg. Nucl.
Ž
.
Chem. 32 1970 2443.
w x
1
Ž
.
F.C. Anson, C. Shi, B. Steiger, Acc. Chem. Res. 30 1997
437, and references therein.
w
w
x
Ž
.
23 R.L. Ake, M. Gouterman, Theoret. Chim. Acta 15 1969 20.
x
Ž
.
24 D. Antipas, M. Gouterman, J. Am. Chem. Soc. 105 1983
w x
2
B.M. Dzhagarov, Y.V. Timinskii, V.S. Chirvonii, G.P. Guri-
4896.
Ž
.
novich, Dokl. Biophys. 247 1979 138.
w
x
25 S. Gentemann, N.Y. Nelson, L. Jaquinod, D.J. Nurco, S.H.
w x
3
C.D. Tait, D. Holten, M. Gouterman, Chem. Phys. Lett. 100
Leung, C.J. Medforth, K.M. Smith, J. Fajer, D. Holten, J.
Ž
.
1983 268.
Ž
.
Phys. Chem. B 101 1997 1247.
w x
4
C.D. Tait, D. Holten, M. Gouterman, J. Am. Chem. Soc. 106
w
w
x
Ž
.
26 D. Eastwood, M. Gouterman, J. Mol. Spectrosc. 35 1970
Ž
.
1984 6653.
359.
w x
5
Ž
.
M. Gouterman, in: D. Dolphin Ed. , The Porphyrins, Vol. 3,
Academic Press, New York, 1978, pp. 1–156.
x
Ž
.
27 A.H. Zewail, in: J. Manz, L. Woste Eds. , Femtosecond
¨
Chemistry, Vol. 1, VCH, Weinheim, 1995, pp. 18–23.