Competitive Electron Transfer from the S2 and S1 Excited States of Zinc meso-Tetraphenylporphyrin to a Covalently Bound Pyromellitimide: Dependence on Donor-Acceptor Structure and Solvent

Article abstract of DOI:10.1021/jp037176i

Zinc meso-tetraphenylporphyrin (ZnTPP) is known to have a relatively slow S₂ → S₁ internal conversion rate, (1-3 ps) ^-1, so that electron transfer from S₂ to a nearby acceptor is possible. We have synthesized tw

Full text of DOI:10.1021/jp037176i

J. Phys. Chem. A 2004, 108, 2375-2381
2375
Competitive Electron Transfer from the S2 and S1 Excited States of Zinc
meso-Tetraphenylporphyrin to a Covalently Bound Pyromellitimide: Dependence on
Donor-Acceptor Structure and Solvent
Ryan T. Hayes, Christopher J. Walsh, and Michael R. Wasielewski*
Department of Chemistry and Center for Nanofabrication and Molecular Self-Assembly,
Northwestern UniVersity, EVanston, Illinois 60208-3113
ReceiVed: October 20, 2003; In Final Form: January 28, 2004
Zinc meso-tetraphenylporphyrin (ZnTPP) is known to have a relatively slow S2 f S1 internal conversion
rate, (1-3 ps)-1, so that electron transfer from S2 to a nearby acceptor is possible. We have synthesized two
zinc porphyrins in which a pyromellitimide (PI) acceptor is attached to the para position of a meso-phenyl
group of the porphyrin (ZnTPP-PI) or to that of a â-phenyl group of the porphyrin (ZnTPP-â-PhPI). We
report here on competitive electron transfer from the S2 (excitation at 420 nm) and S1 states (excitation at 550
nm) of these zinc porphyrins to PI in 2-methyltetrahydrofuran (MTHF) and toluene. Our transient absorption
studies show that efficient electron transfer occurs from S2 of ZnTPP to PI in these porphyrins despite the
presence of the intervening phenyl between the zinc porphyrin macrocycle and the PI acceptor. Moreover,
the results show that while charge separation from S1 is about 6 times faster for PI attached to the meso-
phenyl than for PI attached to the â-phenyl, the opposite is observed for charge separation from S2. Subsequent
charge recombination is 2.6-2.8 times faster for PI attached to the meso-phenyl relative to that of the â-phenyl.
Comparisons of rates between MTHF and toluene show that the ordering of rates is the same for both solvents,
although the rate ratios are larger in MTHF than in toluene.
Introduction
true for the a2u orbital.11,12 Which of these two orbitals is the
zinc porphyrin HOMO depends on the nature and the position
The relatively slow S2 f S1 internal conversion rate, (1-3
ps)-1, of zinc meso-tetraphenylporphyrin (ZnTPP) has been the
subject of several recent reports, which explore the dynamics
of this process using mainly fluorescence upconversion
techniques.1-7 In some of these studies the ability of the S2 state
of zinc porphyrins to donate electrons to a covalently bound
acceptor has also been investigated.7-10 For example, LeGour-
rierec et al.8 have investigated electron transfer from the S2 state
of ZnTPP to a covalently bound Ru(bpy)3 (bpy ) 2,2′-
bipyridine) complex. In addition, Mataga and co-workers have
studied the energy gap dependence for photoinduced electron-
transfer rates from the S2 state of the zinc porphyrin within a
series of zinc 5-(aryleneimide (or arylenediimide))-10,20-
diphenylporphyrins in which the nitrogen atom of the imide
acceptor is directly bonded to the porphyrin macrocycle.7,10
Given the limited lifetime of the zinc porphyrin S2 state and
the short donor-acceptor distances of most previously studied
systems,7,10 we wish to determine whether increasing the
distance between the porphyrin macrocycle and a pyromelliti-
mide (PI) electron acceptor within these molecules has a
significant impact on the competition between electron transfer
from the S2 state of the zinc porphyrin and S2 f S1 internal
conversion. In addition, we wish to determine whether the rates
of electron transfer from S2 depend significantly on whether PI
is attached to either the meso or â position of the zinc porphyrin.
The two highest occupied molecular orbitals of zinc porphyrins
have a1u and a2u symmetry. It is well-known that the a1u orbital
has significant electron density at the â carbons of the porphyrin
with little density at the meso positions, while the opposite is
of the substituents attached to the porphyrin. For example, when
the meso positions (5,10,15,20) of the porphyrin are substituted
with electron-donating groups, e.g. phenyl rings or alkyl chains,
the a2u orbital is the HOMO. On the other hand, substitution of
the â positions with electron releasing groups, such as alkyl
groups, results in the a1u orbital being the HOMO. Generally,
increased electron density at particular peripheral carbon atoms
of the porphyrin results in stronger electronic coupling to
molecules attached to those positions. If the attached molecule
can accept energy and/or electrons from the porphyrin, the
increased electronic coupling generally leads to increased rates
of energy and/or electron transfer.13,14 Ab initio configuration
interaction calculations of porphyrin electronic states show that
the configuration of the S2 state has different contributions from
the a1u and a2u orbitals than does that of the S1 state.15 For
example, calculations on magnesium porphyrin show that its
S1 state has dominant contributions from its a2u (HOMO) and
a1u (HOMO-1) orbitals, where the a2u orbital has the larger
coefficient (0.68) relative to that for the a1u orbital (0.61). On
the other hand, its S2 state has a larger a1u orbital coefficient
(0.59) relative to that for the a2u orbital (0.56). Thus attachment
of the acceptor to the meso or â positions of the porphyrins
may have an impact on electron-transfer rates from S2. Last,
we are interested in whether solvent properties affect the rates
of electron transfer from S2. Solvents can act to selectively
stabilize or destabilize one or both of the a1u or a2u orbitals,
thus changing the characteristics of the S2 state itself.16
To examine these issues we have synthesized two zinc
porphyrins, Chart 1, in which a PI acceptor is attached to the
para position of a phenyl group at a meso-phenyl of the
* To whom correspondence should be addressed. E-mail: wasielew@chem.
northwestern.edu.
10.1021/jp037176i CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/05/2004

2376 J. Phys. Chem. A, Vol. 108, No. 13, 2004
Hayes et al.
CHART 1: Structures of Molecules Used in This Study
porphyrin (ZnTPP-PI) or to that of a â-phenyl group of the
porphyrin (ZnTPP-â-PhPI). We report here on competitive
electron transfer from the S2 (excitation at 420 nm) and S1 states
(excitation at 550 nm) of ZnTPP-PI and ZnTPP-â-PhPI in
2-methyltetrahydrofuran (MTHF) and toluene. In this paper the
S1 and S2 states of a given zinc porphyrin, ZnP, will be denoted
pyromellitimide (1.0 g, 2.1 mmol, 68%). 1H NMR (CDCl3): δ
8.37 (s, pyromellitimide, 2H); 7.66 (d, J ) 8.72 Hz, phenyl,
2H); 7.37 (d, J ) 8.68 Hz, phenyl, 2H); 3.75 (t, J ) 7.36 Hz,
n-methylene, 2H); 1.70 (t, J ) 6.88 Hz, methylene, 2H); 1.31
(m, methylenes, 10H); 0.86 (m, methyl, 3H). MALDI-MS:
483.1 (483.4, calcd).
1
1
as *ZnP and **ZnP, respectively.
N-[4-(Tributyltin)phenyl)]-N′-(n-octyl)pyromellitimide. Reac-
tion of (4-bromophenyl)pyromellitimide (1.6 g, 3.3 mmol) with
bis-tributyltin (5.77 g, 9.9 mmol) in dry refluxing toluene for
14 h under a nitrogen atmosphere using tetrakis(triphenylphos-
phine)palladium(0) as the catalyst gave N-[4-(tributyltin)-
phenyl)]-N′-(n-octyl)pyromellitimide, which was isolated using
Experimental Section
Synthesis. Proton nuclear magnetic resonance spectra were
recorded on a Mercury 400 or Inova 500 NMR spectrometer
using tetramethylsilane (TMS) as an internal standard. Laser
desorption mass spectra were obtained with a PE Voyager DE-
Pro MALDI-TOF mass spectrometer using 2-hydroxy-1-naph-
thoic acid or dithranol as a matrix. Solvents and reagents were
used as received except for tetrahydrofuran (THF), which was
purified on a basic alumina column. Flash and thin-layer
chromatography was performed using Sorbent Technologies
(Atlanta, GA) silica gel.
1
column chromatography (2.4 g, 67%). H NMR (CDCl3): δ
8.38 (s, pyromellitimide, 2H); 7.63 (d, J ) 8.21 Hz, phenyl,
2H); 7.39 (d, J ) 8.22 Hz, phenyl, 2H); 3.76 (t, J ) 7.36 Hz,
n-methylene, 2H); 1.72 (t, J ) 6.89 Hz, methylene, 2H); 1.55
(m, methylene, 6H); 1.34 (m, methylene, 16H); 1.09 (m,
methylene, 6H); 0.89 (m, methyl, 12H).
ZnTPP-PI. (5-Bromo-10,15,20-triphenylporphyrinato)zinc-
(II)17,18 was prepared from 5,10,15-triphenylporphyrin.19 Pd2dba3
(5 mg, dba ) dibenzylideneacetone) and P(o-tolyl)3 (50 mg)
were dissolved in 2 mL DMF. To the reaction mixture was
added 5-bromo-10,15,20-triphenylporphyrinatozinc(II) (32 mg,
0.047 mmol), and the flask was immersed in an oil bath at 70
°C. In a separate flask, N-[4-(tributyltin)phenyl)]-N′-(n-octyl)-
pyromellitimide (32 mg, 0.047 mmol) was dissolved in 2 mL
of DMF. The tin derivative was added to the reaction mix, and
10 mol % of CuI was added. The reaction was stirred overnight
under nitrogen. After 24 h, the reaction was cooled, DMF
removed under vacuum, and the residue deposited onto silica
gel. Column chromatography in 75/25 (v/v) chloroform/hexanes
gave pure ZnTPP-PI (25 mg, 52%). 1H NMR (CDCl3): δ 9.01
(m, porph, 8H); 8.39 (d, J ) 8.50 Hz, bridge phenyl, 2H); 8.34
(s, pyromellitimide, 2H); 8.24 (m, o-phenyl, 6H); 7.86 (d, J )
8.72 Hz, bridge phenyl, 2H); 7.78 (m, m,p-phenyl, 9H); 3.72
(t, J ) 7.36 Hz, n-methylene, 3H); 1.71 (m, methylene, 2H);
1.31 (m, methylene, 10H); 0.89 (m, methyl, 3H). MALDI-MS:
1004.2 (1004.5, calcd).
N-(n-Octyl)pyromellitic Monoanhydride (PIA). Benzene-
1,2,4,5-dicarboxyanhyride (2.0 g, 9.2 mmol) is placed in 35 mL
of dimethylformamide (DMF) and heated to 80 °C. n-Octyl-
amine (1.2 g, 9.2 mmol) in 15 mL of DMF is added dropwise
to the reaction flask over a period of 20 min. The reaction was
heated at 80 °C for another 10 h to ensure completion. The
reaction flask is removed from the heat, cooled to room
temperature, and then placed in a freezer for 4 h. Disubstituted
PI precipitates and is removed by filtration and centrifugation.
The DMF filtrate is then evaporated and the tacky residue is
redissolved in dichloromethane (DCM). Some unreacted starting
material precipitates from DCM as a white solid and is easily
filtered off. The filtrate is evaporated to the minimal amount of
DCM necessary to maintain solubility, and then the product is
precipitated by pouring the reaction mixture into a large volume
of hexane. The precipitate is recovered through centrifugation.
Further purification is achieved by flash silica chromatography
using a solvent gradient starting with 95/5 DCM/ethyl acetate
(v/v) and going to 80/20 DCM/ethyl acetate. Unreacted starting
material precipitates out on top of the silica column and does
not elute as long as the eluent is mainly DCM. After the column,
the tacky material is recrystallized from hot acetic anhydride
ZnTPP-â-PhPI. Pd2dba3 (5 mg) and P(o-tolyl)3 (50 mg) were
dissolved in 2 mL of DMF. To the reaction mix was added
(2-bromo-5,10,15,20-tetraphenylporphyrinato)zinc(II)20 (0.150
g, 0.198 mmol), and the flask was immersed in an oil bath at
70 °C. In a separate flask, N-[4-(tributyltin)phenyl)]-N′-(n-octyl)-
pyromellitimide (0.206, 0.297 mmol) was dissolved in 2 mL
of DMF. The tin derivative was added to the reaction mix, and
10 mol % of CuI was added. The reaction was stirred overnight
under nitrogen. After 24 h, the reaction was cooled, the DMF
was removed under vacuum, and the residue was deposited onto
silica gel. Column chromatography using 75/25 (v/v) chloroform/
1
to yield PIA (2.0 g, 33%). H NMR (CDCl3): δ 8.43 (d, J )
0.64 Hz, benzene, 2H); 3.76 (t, J ) 7.37 Hz, n-methylene, 2H);
1.71 (quint., J ) 7.13 Hz, methylene, 2H); 1.32 (m, methylene,
11H); 0.86 (t, J ) 6.76 Hz, methyl, 3H). MALDI-MS: 328.9
(329.4, calcd).
N-(4-Bromophenyl)-N′-(n-octyl)pyromellitimide. 4-Bromo-
aniline (0.522 g, 3.0 mmol) and PIA (1.0 g, 3.0 mmol) were
reacted in DMF at 120 °C for 4 h. Upon cooling to room
temperature, the product precipitated. The creamy-white solid
was filtered and dried to yield N-(4-bromophenyl)-N′-(n-octyl)-
1
hexanes gave pure ZnTPP-â-PhPI (133 mg, 62%). H NMR
(CDCl3): δ 8.91 (m, porph, 7H); 8.41 (s, pyromellitimide, 2H);

Transfer of Excited States of ZnTPP to PI
J. Phys. Chem. A, Vol. 108, No. 13, 2004 2377
TABLE 1: Absorption Maxima for the Zinc Porphyrin Soret and Q(0,0) Bands in MTHF and Toluene and Energies for S1, S2,
and the Radical Ion Pairs
Soret (nm)
Q(0,0) (nm)
ES1 (eV)
ES2 (eV)
∆GIP (eV)
compound
ZnTPP
ZnTPP-PI
ZnTPP-â-PhPI
MTHF
toluene.
MTHF
toluene
MTHF
toluene
MTHF
toluene
MTHF
toluene
423
423
426
424
424
427
595
595
598
589
589
591
2.08
2.07
2.06
2.09
2.09
2.07
2.92
2.91
2.88
2.90
2.90
2.88
1.63
1.63
2.06
2.04
8.21 (m, o-phenyl, 6H); 7.90 (d, J ) 7.51 Hz, o-phenyl, 2H);
7.76 (m, m,p-phenyl, 9H); 7.49 (d, J ) 8.45 Hz, bridge-phenyl,
2H); 7.38 (t, J ) 7.49 Hz, m-phenyl, 2H); 7.30 (t, J ) 7.14 Hz,
p-phenyl, 1H); 7.21 (d, J ) 8.34 Hz, bridge-phenyl, 2H); 3.77
(t, J ) 7.26 Hz, N-methylene, 2H); 1.75 (quintet, J ) 7.15 Hz,
methylene, 2H); 1.31 (m, methylene, 10H); 0.92 (m, methyl,
3H). MALDI-MS: 1079.9 (1080.6, calcd).
Spectroscopy. Femtosecond transient absorption measure-
ments were made using a regeneratively amplified titanium
sapphire laser system operating at a 2 kHz repetition rate
outfitted with a CCD array detector (Ocean Optics PC2000)
for simultaneous collection of spectral and kinetic data.21 The
frequency-doubled output from the laser was used to provide
420 nm, 130 fs pulses for excitation, while 550 nm, 130 fs
excitation pulses were produced using an optical parametric
amplifier.22 A white light continuum probe pulse was generated
by focusing the 840 nm fundamental into a 1 mm sapphire disk.
All-reflective optics were used to focus the 840 nm pulse into
the sapphire and recollimate the white light output, thus limiting
the chirp on the white light pulse to less than 200 fs from 450
to 800 nm. Cuvettes with a 2 mm path length were used, and
the samples were irradiated with 0.5 µJ/pulse focused to a 200
µm spot. The optical density at the excitation wavelength was
typically 0.3-0.5. The samples were stirred during the experi-
ment using a wire stirrer to prevent thermal lensing and sample
degradation. The total instrument response for the pump-probe
experiments was 180 fs. Steady-state absorption and emission
spectra were performed on a Shimadzu 1601 UV/vis spectro-
photometer and PTI instruments single-photon-counting spec-
trofluorimeter, respectively. A 10 mm quartz cuvette was used
with optical density at 425 nm maintained at 0.1 ( 0.05 to avoid
reabsorption artifacts. All solvents were spectrophotometric
grade. Toluene was purified by passing it through series CuO
and alumina columns (GlassContour), while 2-methyltetrahy-
drofuran (MTHF; Aldrich) was purified by passing it through
a basic alumina column immediately prior to use.
Kinetic analyses were performed at several wavelengths using
a nonlinear least-squares fit to either a general sum-of-
exponentials function or to a series A f B f C kinetic
mechanism using the Levenberg-Marquardt method23 account-
ing for the presence of the finite instrument response. The
wavelength for these analyses centers on the relatively narrow
PI- band at 720 nm.24 Kinetic simulations were carried out using
the CKS program developed by the IBM Almaden Research
Center.
Electrochemical measurements were performed using a CH
Instruments Model 622 electrochemical workstation. The solvent
was butyronitrile (PrCN) containing 0.1 M tetra-n-butylammo-
nium perchlorate electrolyte. A 1.0 mm diameter platinum disk
electrode, a platinum wire counter electrode, and a Ag/AgxO
reference electrode were employed. The ferrocene/ferrocinium
couple (Fc/Fc+, 0.52 vs SCE) was used as an internal reference
for all measurements. The energies of the radical ion pairs given
in Table 1 were calculated25 using the one-electron oxidation
potentials of the zinc porphyrins (0.85 V vs SCE), the one-
electron reduction potential of PI (-0.71 vs SCE), ionic radii
of 7 Å and 5 Å, respectively, for the zinc porphyrins and PI,
Figure 1. Transient absorption spectra for ZnTPP-PI in MTHF with
(A) 550 and (B) 420 nm excitation.
and donor-acceptor distances of 13.1 Å (ZnTPP-PI) and 13.5
Å (ZnTPP-â-PhPI). The total reorganization energies for the
electron-transfer reactions λCS ) λCR ) 0.33 eV in toluene and
0.86 eV in MTHF were calculated using the dielectric continuum
expression of Marcus.26
Results
The advantages of employing PI as an electron acceptor have
been well-documented.24,27-34 The sharp transient absorption
feature of PI- near 720 nm (ꢀ ) 41 700 M-1 cm-1 24
allows
)
for kinetic analyses of the electron-transfer reactions with
minimal interference from the zinc porphyrin features. It is well-
known that strong, overlapping optical absorptions of 1*ZnTPP
and ZnTPP+ occur mainly at 460-470 nm and that ZnTPP+
has a broad absorption feature around 650 nm as well, which
is more pronounced when the zinc porphyrin is an octaalkyl-
porphyrin.12,28,32 While PI- also has a second, lower intensity
absorption at 652 nm (ꢀ ) 9800 M-1 cm-1), we will not rely
on this feature because it overlaps with the absorption of both
1
1
ZnTPP+ and *ZnTPP. The onset of *ZnTPP fluorescence at
650 nm, which appears as a stimulated emission feature (an
absorbance bleach) in the transient absorption experiments, can
also be used to monitor the lifetime of S2.
The optical absorption maxima of the Soret and Q(0,0) bands
for each compound and ZnTPP in MTHF and toluene are listed
in Table 1. The Soret band maxima are nearly solvent-
independent, while the Q(0,0) band maxima show consistent,
small red shifts of their absorption maxima in MTHF relative
to toluene. These shifts are most likely due to ligation of the
MTHF oxygen atom to the zinc atom in each case. The S1 and
S2 energy levels of each molecule were determined by averaging
the energies of the Q(0,0) and Soret band absorptions of the
zinc porphyrin with those of the corresponding emission
maxima.
Excitation of ZnTPP-PI in MTHF at 550 nm produces the
transient absorption spectra shown in Figure 1A. The spectrum
1
at 1 ps shows the presence of *ZnTPP alone, while by 30 ps
ZnTPP+-PI- has formed, as evidenced by the small peak at
720 nm. The PI- band appears with τ ) 17 ps and is assigned
to the formation of ZnTPP+-PI-, which subsequently decays
with τ ) 140 ps. The weak 480 nm feature at 1000 ps seen in
Figure 1A is due to 3*ZnTPP formed by radical pair intersystem
crossing, which, although not unprecedented,35 is somewhat

2378 J. Phys. Chem. A, Vol. 108, No. 13, 2004
Hayes et al.
TABLE 2: Summary of Transient Absorption Kinetic Dataa
compound
solvent
λ
EXC (nm)
λ
PR (nm)
τ
R1 (ps)
AR1 (%)
τ
R2 (ps)
AR2 (%)
τ
D1 (ps)
AD1 (%)
τ
D2 (ps)
AD2 (%)
ZnTPP-PI
MTHF
MTHF
MTHF
MTHF
MTHF
MTHF
toluene
toluene
toluene
toluene
550
420
550
420
550
420
550
420
550
420
720
720
480
480
720
720
720
720
720
720
17
100
100
90
80
100
50
100
57
100
100
140
18
0.3
0.6
370
380
700
610
>6 ns
2.0 ns
100
80
15
15
85
ZnTPP-PI
1.4
0.3
0.5
96
0.7
120
0.6
150
140
160
>6 ns
>6 ns
20
85
85
15
5
ZnTPP-PI
ZnTPP-PI
ZnTPP-â-PhPI
ZnTPP-â-PhPI
ZnTPP-PI
ZnTPP-PI
ZnTPP-â-PhPI
ZnTPP-â-PhPI
18
18
10
20
9.7
50
43
95
100
100
100
100
127
0.2
0.4
a τR1, τR2, AR1, and AR2 are the time constants and their respective amplitudes for the rise of the PI- absorption, while τD1, τD2, AD1, and AD2 are
the time constants and their respective amplitudes for the decay of the PI- absorption. The 0.2 ps instrument response function has been deconvoluted
from the data. The S2 lifetimes of ZnTPP in MTHF and toluene are 2.3 and 1.1 ps, respectively.
Figure 2. Transient absorption kinetics of ZnP-PhPI in MTHF probed
at (A) 720 and (B) 480 nm. Solid lines are exponential fits.
Figure 4. Transient absorption kinetics of ZnTPP-â-PhPI in MTHF
(A) at 720 nm using 420 and 550 nm excitation and (B) at 650 nm
using 420 nm excitation. Solid lines are exponential fits.
ps, Figure 3B, which occurs substantially earlier than is observed
following 550 nm excitation. Figure 4A shows kinetic traces
for both excitation wavelengths. Electron transfer from S2 is
competitive with internal conversion, and the kinetic traces show
that PI- is formed rapidly (τ ) 0.7 ps) relative to that produced
from S1 (τ ) 96 ps). At 650 nm, Figure 4B, the exponential fit
of the initial, fast decay process reveals that the stimulated
emission onset occurs with τ ) 0.7 ps in agreement with the
faster of the two time constants for the formation of PI-. Note
that the stimulated emission onset is the decaying transient
feature in the kinetic trace, i.e., the appearance of the bleach
due to stimulated emission. The initial rise of the signal is due
to the formation of the S2 f Sn absorption convolved with the
0.2 ps instrument response function.
The slower τ ) 9.7 ps component for the PI- formation
observed at 720 nm may be due to spectral narrowing resulting
from vibrational relaxation. Interestingly, kinetic components
readily assignable to vibrational and/or solvent relaxation are
not observed for ZnTPP-PI in either solvent or for ZnTPP-
â-PhPI in toluene. While it is possible that these relaxation
processes are very fast in these cases, it is more likely that the
kinetic components due to vibrational and solvent relaxation
occur with similar time constants as do charge separation and
recombination, so that the relative contributions of the various
processes cannot be easily separated. This is evident from the
kinetic components given in Table 2. For example, the electron-
transfer rates from S1 for both ZnTPP-PI and ZnTPP-â-PhPI
in either solvent are all slow, making it difficult to discern an
additional component on the 10 ps time scale, especially if it
has a low amplitude.
Figure 3. Transient absorption spectra of ZnTPP-â-PhPI in MTHF
with (A) 550 and (B) 420 nm excitation.
unusual given that this process is generally inefficient for short-
lived radical ion pairs.36 The corresponding spectra for 420 nm
excitation are given in Figure 1B. The spectrum at 2 ps already
reveals an intense PI- feature at 720 nm. The transient
absorption kinetic traces at 720 nm shown in Figure 2A reveals
that this absorption appears with τ ) 1.4 ps and decays
biexponentially with τ ) 18 and 140 ps. Moreover, the transient
absorption kinetics at 480 nm requires 18 ps rise and 140 ps
decay components to fit the data, Figure 2B. The kinetic
components and their amplitudes are listed in Table 2. For
completeness, transient absorption spectra of ZnTPP in MTHF
at 0.4 ps (S2 f Sn) and 20 ps (S1 f Sn) following 420 nm
excitation are given in the Supporting Information, Figure S1.
The transient absorption spectra for ZnTPP-â-PhPI in MTHF
following 550 nm excitation are shown in Figure 3A. At 20 ps
the transient spectrum is due to 1*ZnTPP (see Figure S1), while
by 200 ps a feature due to PI- has appeared at 720 nm. After
3
1600 ps, the only species remaining is *ZnTPP. Transient
absorption kinetics obtained at 720 nm, Figure 4A and Table
2, show that the charge separation reaction: 1*ZnTPP-â-PhPI
f ZnTPP+-â-PhPI- is about six times slower than the
corresponding reaction for ZnTPP-PI. Once again, the forma-
tion of 3*ZnTPP is evident at long times. The spectrum
following 420 nm excitation shows a significant PI- peak at 5
Excitation of ZnTPP-PI in toluene with 550 nm pulses
produces transient absorption spectra similar to those shown
for ZnTPP-PI in MTHF, Figure 5A. The transient spectrum at
10 ps is dominated by 1*ZnTPP and evolves by 150 ps into the
spectrum of ZnTPP+-PI-, which decays more slowly to give
an intense spectrum due to 3*ZnTPP by 4.5 ns. The correspond-

Transfer of Excited States of ZnTPP to PI
J. Phys. Chem. A, Vol. 108, No. 13, 2004 2379
Figure 5. Transient absorption spectra of ZnTPP-PI in toluene with
(A) 550 and (B) 420 nm excitation.
Figure 8. Transient absorption kinetics for ZnTPP-â-PhPI in toluene
using 420 nm excitation and probed at (A) 720 and (B) 650 nm. The
solid lines are exponential fits.
TABLE 3: Electron-Transfer Rate Constants
compound
solvent kCS1 (s-1
)
kCS2 (s-1 φCS2 (%) kCR (s-1
) )
ZnTPP-PI
MTHF 5.8 × 1010 2.8 × 1011
39
69
46
64
7.1 × 109
ZnTPP-â-PhPI MTHF 1.0 × 1010 9.9 × 1011
2.7 × 109
1.4 × 109
5.0 × 108
ZnTPP-PI
ZnTPP-â-PhPI toluene
toluene 8.3 × 109 7.6 × 1011
1.6 × 1012
1
8B, both show that charge separation from **ZnTPP occurs
very rapidly with τ ) 0.4 ps (following deconvolution of the
0.2 ps instrument response function).
Figure 6. Transient absorption kinetics of ZnTPP-PI in toluene (A)
at 720 nm with 420 and 550 nm excitation and (B) at 650 nm with
420 nm excitation. Solid lines are exponential fits.
Discussion
The rate constants for electron transfer from 1*ZnTPP to PI,
kCS1, given in Table 3 are obtained using the following
expression: kCS1 ) (1/τobs) - (1/τS1), where τobs is the observed
time constant for the formation of the radical ion pair and τS1
1
is the lifetime of *ZnTPP, which is 2.4 ns. To obtain the
corresponding rate constants for electron transfer from 1**ZnT-
PP to PI, we assume that S2 is initially and selectively populated
by one-photon absorption at 420 nm. The observed time
constant, kobs, for the decay of the stimulated emission band at
650 nm and the appearance of the PI- band at 720 nm is kobs
)
Figure 7. Transient absorption spectra of ZnTPP-â-PhPI in toluene
kCS2 + kIC, where kCS2 is the rate constant for charge separation
from S2 and kIC is the rate constant for S2 f S1 internal
conversion. This expression assumes that vibrational relaxation
and solvent dynamics can be distinguished from the initial
charge separation event. In general, however, as has been
discussed above, this assumption is not always valid. The values
of kIC for ZnTPP in toluene and MTHF are (1.1 ps)-1 and (2.3
ps)-1, respectively, as determined from the onset of stimulated
emission at 650 nm following 420 nm excitation. Fluorescence
upconversion studies have shown that the S2 f S1 internal
conversion rates for ZnTPP in benzene and THF are similar to
those we obtain in toluene and MTHF.7,38
with (A) 550 and (B) 420 nm excitation.
ing data for ZnTPP-PI excited at 420 nm results in formation
of ZnTPP+-PI- from the S2 state of ZnTPP as evidenced by
the appearance of the PI- band in the transient absorption
spectrum at 5 ps, Figure 5B. A comparison between the transient
kinetics for PI- formation monitored at 720 nm is shown in
Figure 6A. It is evident from the data in Figure 6A that 420
nm excitation results in a significant increase in radical ion pair
production at early times relative to that produced by 550 nm
excitation. Figure 6B compares the kinetics of the onset of
stimulated emission for ZnTPP and ZnTPP-PI in toluene
monitored at 650 nm. The onset for ZnTPP occurs with τ )
1.1 ps, while that for ZnTPP-PI occurs with τ ) 0.6 ps. The
latter value matches the fast component for PI- formation
following 420 nm excitation of ZnTPP-PI.
Figure 9 shows that the radical ion pair energy levels for
ZnTPP-PI and ZnTPP-â-PhPI are nearly isoenergetic with that
of 3*ZnTPP in MTHF, while they are nearly isoenergetic with
1
that of *ZnTPP in toluene. The observed rate constants for
Excitation of ZnTPP-â-PhPI in toluene with 550 nm laser
pulses produces transient absorption spectra that show no
indication of ion pair formation, Figure 7A, while excitation
with 420 nm pulses shows the presence of PI- at 5 ps. Direct
charge separation from S1 of both ZnTPP-PI and ZnTPP-â-
PhPI, Table 3, increase as the solvent polarity increases and
the energy levels of the radical ion pairs are stabilized in MTHF
relative to toluene. For example, the charge separation rate for
ZnTPP-PI in MTHF is about 7 times faster than in toluene.
The free energies for charge separation from S1 in MTHF and
toluene are about -0.45 and -0.05 eV, respectively, while the
corresponding values of the total reorganization energies are
0.86 and 0.33 eV, so that the observed increase in rate as the
free energy of reaction becomes more negative is consistent
with both charge separation reactions being in the normal region
of the Marcus rate vs free energy profile.26,39 A similar
comparison for ZnTPP-â-PhPI cannot be made because the
1
formation of *ZnTPP using 550 nm excitation is followed by
spin-orbit-induced intersystem crossing to yield 3*ZnTPP over
the 6 ns time course of the experiment as indicated by the strong
480 nm feature37 in the spectrum at 4.5 ns. Figure 7B shows
that ZnTPP+-â-PhPI- undergoes radical ion pair intersystem
crossing followed by charge recombination to ultimately yield
3*ZnTPP as well. The transient absorption kinetics for the
formation of PI- at 720 nm, Figure 8A, and the onset of
stimulated emission due to formation of S1 at 650 nm, Figure

2380 J. Phys. Chem. A, Vol. 108, No. 13, 2004
Hayes et al.
Figure 9. Energy level diagrams for ZnP-PI systems in (A) MTHF and (B) toluene. The energy levels for all four compounds are within (0.05
V of the energies indicated.
charge separation rate from S1 in toluene is too slow to compete
with the decay of S1. On the other hand, making the same
comparisons for charge separation from S2 in ZnTPP-PI and
ZnTPP-â-PhPI shows that the rates of charge separation are
2-3 times faster in toluene for each molecule relative to MTHF,
just the opposite from that observed for charge separation from
S1. The free energies of reaction for charge separation from S2
for both ZnTPP-PI and ZnTPP-â-PhPI in MTHF and toluene
are about -1.3 and -0.8 eV, respectively, while the corre-
sponding values of the total reorganization energies are 0.86
and 0.33 eV, so that the rate data are consistent with both charge
separation reactions from S2 being in the Marcus inverted region.
smaller differences. The results strongly suggest that the a2u
orbital with significant electron density at the meso positions
dominates the electronic coupling matrix elements for both
charge separation from S1 and charge recombination. These
results are consistent with earlier observations concerning the
role of the porphyrin a1u and a2u orbitals.11,12,15 Yet, this effect
is reversed for charge separation from the S2 state. Configura-
tional mixing within the S2 state produces an electronic
distribution that places more electron density on the â rather
than the meso position.15,16 This shift in electronic distribution
results in faster rates of charge separation from the zinc
porphyrin S2 state for PI attached to the â position, which has
not been observed previously.
A consequence of the fact that the energy of ZnTPP+-PI-
3
is nearly isoenergetic with that of *ZnTPP in MTHF is the
Conclusions
observed biexponential nature of the decay of ZnTPP+-PI-,
Figure 2A. The kinetic data suggest that there is a rapid
equilibrium between ZnTPP+-PI- and the nearly isoenergetic
3*ZnTPP state. Simulation of the kinetic data using the
mechanism outlined in Figure 9A reveals that the dominant short
component results from rapid radical ion pair recombination to
3*ZnTPP with kCRT ) 5.9 × 1010 s-1, while the longer
component is due to repopulation of the radical ion pair state
with kCST ) 3.3 × 1010 s-1 followed by recombination of the
Our transient absorption studies show that efficient electron
transfer occurs from S2 of ZnTPP to PI in these porphyrins
despite the presence of the intervening phenyl between the zinc
porphyrin macrocycle and the PI acceptor. This significantly
increases the overall distance of the charge separation reaction
relative to systems in which the PI is bound directly to the
porphyrin macrocycle.7 Our results also show that the relative
rates of charge separation to an electron acceptor can be
modified in an opposing fashion through attachment of the
acceptor to either the meso- or â-pyrrole positions of the
macrocycle, where the relative ordering of rates depends on
whether the electron transfer proceeds from S1 or S2. This
ordering correlates well with the larger calculated contribution
of the metalloporphyrin a2u orbital to S1 and the a1u orbital to
S2.15 In addition, the decrease in the relative rate ratios observed
in toluene relative to those in MTHF is most likely a
consequence of solvent modulation of the a2u and a1u orbital
energies.16 This diversity of properties will allow us to design
more complex donor-acceptor systems in which wavelength
selective formation of S1 or S2 will produce different ion pair
states with different spectroscopic observables. This type of
behavior may prove useful in developing systems for informa-
tion processing using ultrafast electron-transfer reactions within
donor-acceptor molecules.
radical pair leading to ground state with kCRS ) 7.0 × 109 s-1
.
When 550 nm excitation is used, the buildup of the ZnTPP+-
PI- population from S1 is relatively slow (τ ) 17 ps), while
the recombination rate to 3*ZnTPP is comparable (τ ) 18 ps),
so that the concentration of ZnTPP+-PI- during the time course
of the reaction is relatively low, as is indicated by the transient
spectra and kinetics in Figures 1A and 2A, respectively. On
the other hand, the much more rapid formation of ZnTPP+-
PI- following 420 nm excitation results in a large initial
concentration of the radical ion pair that is revealed by the
spectra and kinetics in Figures 1B and 2A, respectively. The
3
formation of *ZnTPP with τ ) 18 ps is reflected in the rise
component necessary to fit the transient absorption data obtained
at 480 nm, where 3*ZnTPP absorbs strongly, Figure 2B.40 The
corresponding energy gap between ZnTPP+-PI- and 3*ZnTPP
is much larger in toluene, Figure 9B, so that biexponential decay
kinetics are not observed.
Acknowledgment. This work was supported by the National
Science Foundation (Grant CHE-0102351) and the Office of
Naval Research (Grant N00014-02-1-0381).
Comparing the charge separation rate constants from S1 for
ZnTPP-PI vs ZnTPP-â-PhPI in MTHF, the rate for PI at the
meso position is 5.8 times faster than that for PI at the â position.
On the other hand, the corresponding comparison of charge
separation rates from S2 show that the rate for ZnTPP-PI is
3.5 times slower than that for ZnTPP-â-PhPI. Comparing the
charge recombination rates to ground state for ZnTPP+-PI-
vs ZnTPP+-â-PhPI- in MTHF, the CR rate for ZnTPP+-PI-
is 2.6 times faster than that of ZnTPP+-â-PhPI-. The corre-
sponding comparisons in toluene show similar trends albeit with
Supporting Information Available: Transient absorption
spectra of ZnTPP in MTHF at 0.4 and 20 ps following a 420
nm laser pulse (Figure S1). This material is available free of
charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Aaviksoo, J. F. A.; Savikhin, S.; Stelmakh, G. F.; Tsvirko, M. P.
Chem. Phys. Lett. 1984, 111, 275-278.

Transfer of Excited States of ZnTPP to PI
J. Phys. Chem. A, Vol. 108, No. 13, 2004 2381
(2) Chosrowjan, H. T. S.; Okada, T.; Takagi, S.; Arai, T.; Tokumaru,
K. Chem. Phys. Lett. 1995, 242, 644-649.
(22) Greenfield, S. R.; Wasielewski, M. R. Opt. Lett. 1995, 20, 1394-
1396.
(3) Gurzadyan, G. G.; Tran-Thi, T. H.; Gustavsson, T. J. Chem. Phys.
1998, 108, 385-388.
(23) Marquardt, D. W. J. Soc. Ind. Appl. Math. 1963, 11, 431-441.
(24) Gosztola, D.; Niemczyk, M. P.; Svec, W.; Lukas, A. S.; Wasielews-
ki, M. R. J. Phys. Chem. A 2000, 104, 6545-6551.
(25) Greenfield, S. R.; Svec, W. A.; Gosztola, D.; Wasielewski, M. R.
J. Am. Chem. Soc. 1996, 118, 6767-6777.
(26) Marcus, R. A. J. Chem. Phys. 1965, 43, 679.
(27) Wiederrecht, G. P.; Niemczyk, M. P.; Svec, W. A.; Wasielewski,
M. R. J. Am. Chem. Soc. 1996, 118, 81-88.
(28) Yamada, K.; Imahori, H.; Yoshizawa, E.; Gosztola, D.; Wasielews-
ki, M. R.; Sakata, Y. Chem. Lett. 1999, 3, 235-236.
(29) Yamazaki, T.; Yamazaki, I.; Osuka, A. J. Phys. Chem. B 1998,
102, 7858-7865.
(30) Osuka, A.; Marumo, S.; Maruyama, K.; Mataga, N.; Tanaka, Y.;
Taniguchi, S.; Okada, T.; Yamazaki, I.; Yoshinobu, N. Bull. Chem. Soc.
Jpn. 1995, 68, 262-276.
(31) Osuka, A.; Nakajima, S.; Maruyama, K.; Mataga, N.; Asahi, T.
Chem. Lett. 1991, 6, 1003-1006.
(32) Imahori, H.; Yamada, K.; Hasegawa, M.; Taniguchi, S.; Okada,
T.; Sakata, Y. Angew. Chem., Int. Ed. Engl. 1997, 36, 2626-2629.
(33) Harrison, R. J.; Pearce, B.; Beddard, G. S.; Cowan, J. A.; Sanders,
J. K. M. Chem. Phys. 1987, 116, 429-448.
(4) Vlcek, A. Chemtracts 2000, 13, 776-779.
(5) Gurzadyan, G. G.; Tran-Thi, T.; Gustavsson, T. In New Trends in
Atomic and Molecular Spectroscopy; Gurzadyan, G. G., Karmenyan, A.
V., Eds.; SPIEsThe International Society of Optical Engineers: Bellingham,
WA, 2000; Vol. 4060.
(6) Mataga, N.; Shibata, Y.; Chosrowjan, H.; Yoshida, N.; Osuka, A.
J. Phys. Chem. B 2000, 104, 4001-4004.
(7) Mataga, N.; Chosrowjan, H.; Taniguchi, S.; Shibata, Y.; Yoshida,
N.; Osuka, A.; Kikuzawa, T.; Okada, T. J. Phys. Chem. A 2002, 106,
12191-12201.
(8) LeGourrierec, D.; Andersson, M.; Davidsson, J.; Mukhtar, E.; Sun,
L.; Hammarstrom, L. J. Phys. Chem. A 1999, 103, 557-559.
(9) Andersson, M.; Davidsson, J.; Hammarstrom, L.; Korppi-Tommola,
J.; Peltola, T. J. Phys. Chem. B 1999, 103, 3258-3262.
(10) Mataga, N. C. H.; Shibata, Y.; Yoshida, N.; Osuka, A.; Kikuzawa,
T.; Okada, T. J. Am. Chem. Soc. 2001, 123, 12422-12423.
(11) Weiss, C.; Kobayashi, H.; Gouterman, M. J. Mol. Spectrosc. 1965,
16, 415-450.
(12) Fajer, J.; Borg, D. C.; Forman, A.; Dolphin, D.; Felton, R. H. J.
Am. Chem. Soc. 1970, 92, 3451-3459.
(13) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002,
35, 57-69.
(34) Debreczeny, M. P.; Svec, W. A.; Wasielewski, M. R. Science 1996,
274, 584-587.
(14) Tsai, H.; Simpson, M. C. Chem. Phys. Lett. 2002, 353, 111-118.
(15) Petke, J. D.; Maggiora, G. M.; Shipman, L. L.; Christoffersen, R.
E. J. Mol. Spectrosc. 1978, 71, 64-84.
(35) Wasielewski, M. R.; Johnson, D. G.; Svec, W. A.; Kersey, K. M.;
Minsek, D. W. J. Am. Chem. Soc. 1988, 110, 7219-7221.
(36) Wiederrecht, G. P.; Svec, W. A.; Wasielewski, M. R.; Galili, T.;
Levanon, H. J. Am. Chem. Soc. 2000, 122, 9715-9722.
(37) Pekkarinen, L.; Linschitz, H. J. Am. Chem. Soc. 1960, 82, 2407-
2411.
(16) Davis, M. S.; Forman, A.; Fajer, J. Proc. Natl. Acad. Sci. U.S.A.
1979, 76, 4170-4174.
(17) Therien, M. J.; Dimagno, S. G. PCT Int. Appl. 1994, 62 pp.
(18) Callot, H. J.; Schaeffer, E. J. Chem. Res. 1978, 2, 51.
(19) Yoshida, N.; Shimidzu, H.; Osuka, A. Chem. Lett. 1998, 1, 55-
56.
(38) Yu, H.-Z.; Baskin, S.; Zewail, A. H. J. Phys. Chem. A 2002, 106,
9845-9854.
(39) Jortner, J. J. Chem. Phys. 1976, 64, 4860-4867.
(40) Pekkarinen, L.; Linschitz, H. J. Am. Chem. Soc. 1960, 82, 2407-
2411.
(20) Callot, H. J. Bull. Soc. Chim. Fr. 1974, 7-8, 1492-1496.
(21) Lukas, A. S.; Miller, S. E.; Wasielewski, M. R. J. Phys. Chem. B
2000, 104, 931-940.