Observation of proton-coupled electron transfer by transient absorption spectroscopy in a hydrogen-bonded, porphyrin donor - Acceptor assembly

Article abstract of DOI:10.1021/jp049296b

Proton-coupled electron transfer (PCET) kinetics of a Zn(II) porphyrin donor noncovalently bound to a naphthalene-diimide acceptor through an amidinium-carboxylate interface have been investigated by time-resolved spectroscopy. The S₁ singlet excited-state of a Zn(II) 2-amidinium-5,10,15,20-tetramesitylporphyrin chloride (ZnP-?2-AmH ⁺) donor is sufficiently energetic (2.04 eV) to reduce a carboxylate-diimide acceptor (??G?° = -460 mV, THF). Static quenching of the porphyrin fluorescence is observed and time-resolved measurements reveal more than a 3-fold reduction in the S₁ lifetime of the porphyrin upon amidinium-carboxylate formation (THF, 298 K). Picosecond transient absorption spectra of the free ZnP-?2-AmH⁺ in THF reveal the existence of an excited-state isosbestic point between the S₁ and T₁ states at ??_probe = 650 nm, providing an effective 'zero-kinetics' background on which to observe the formation of PCET photoproducts. Distinct rise and decay kinetics are attributed to the build-up and subsequent loss of intermediates resulting from a forward and reverse PCET reaction, respectively (k_PCET(fwd) = 9 ?? 10⁸ s ^-1 and k_PCET(rev) = 14 ?? 10⁸ s ^-1). The forward rate constant is nearly 2 orders of magnitude slower than that measured for covalently linked Zn(II) porphyrin-acceptor dyads of comparable driving force and D-A distance, establishing the importance of a proximal proton network in controlling charge transport.

Full text of DOI:10.1021/jp049296b

J. Phys. Chem. B 2004, 108, 6315-6321
6315
Observation of Proton-Coupled Electron Transfer by Transient Absorption Spectroscopy in
a Hydrogen-Bonded, Porphyrin Donor-Acceptor Assembly
Niels H. Damrauer, Justin M. Hodgkiss, Joel Rosenthal, and Daniel G. Nocera*
Department of Chemistry, 6-335, Massachusetts Institute of Technology, 77 Massachusetts AVenue,
Cambridge, Massachusetts 02139-4307
ReceiVed: February 16, 2004
Proton-coupled electron transfer (PCET) kinetics of a Zn(II) porphyrin donor noncovalently bound to a
naphthalene-diimide acceptor through an amidinium-carboxylate interface have been investigated by time-
resolved spectroscopy. The S1 singlet excited-state of a Zn(II) 2-amidinium-5,10,15,20-tetramesitylporphyrin
chloride (ZnP-â-AmH+) donor is sufficiently energetic (2.04 eV) to reduce a carboxylate-diimide acceptor
(∆G° ) -460 mV, THF). Static quenching of the porphyrin fluorescence is observed and time-resolved
measurements reveal more than a 3-fold reduction in the S1 lifetime of the porphyrin upon amidinium-
carboxylate formation (THF, 298 K). Picosecond transient absorption spectra of the free ZnP-â-AmH+ in
THF reveal the existence of an excited-state isosbestic point between the S1 and T1 states at λprobe ) 650 nm,
providing an effective ‘zero-kinetics’ background on which to observe the formation of PCET photoproducts.
Distinct rise and decay kinetics are attributed to the build-up and subsequent loss of intermediates resulting
from a forward and reverse PCET reaction, respectively (kPCET(fwd) ) 9 × 108 s-1 and kPCET(rev) ) 14 ×
108 s-1). The forward rate constant is nearly 2 orders of magnitude slower than that measured for covalently
linked Zn(II) porphyrin-acceptor dyads of comparable driving force and D-A distance, establishing the
importance of a proximal proton network in controlling charge transport.
Introduction
kinetics been ascertained directly from observed intermediates
of these types of tightly coupled PCET assemblies.
The challenge to directly measuring PCET kinetics by
transient absorption spectroscopy arises as a direct consequence
of the presence of a proton transfer network proximal to the
electron transfer network. In tightly coupled networks, the proton
significantly retards the ET rate so that charge transport is slow
with respect to the dynamics of a photoexcited donor. Most
notably for porphyrin donors, significant optical intensities
associated with S1 and T1 excited state interconversions
overwhelm the optical signatures of PCET intermediates.
Accordingly, measurement of charge transport in tightly coupled
PCET networks has been confined to monitoring the disappear-
ance of the S1 excited state by transient emission or absorption
spectroscopy. However, this approach can be problematic
because the dynamics of luminescence decay are not necessarily
specific to charge transport and PCET does not always proceed
from the emitting state. With regard to this latter issue, transient
EPR experiments of PCET assemblies involving porphyrins have
established the viability of the excited state donor to be T1 as
opposed to S1.23 In light of these complications, specific to
studies of tightly coupled PCET networks, it is desirable to
develop approaches that enable PCET kinetics to be determined
from directly observed charge transport intermediates. We now
report such an approach for a PCET network composed of a
porphyrin donor bridged to a diimide acceptor via an amidinium-
carboxylate salt bridge.
Proton-coupled electron transfer (PCET) was launched as a
mechanistic field of study with the synthesis of donor-acceptor
(D-A) pairs assembled by hydrogen-bonding interfaces ([H+]).1-3
Initial D- - -[H+]- - -A assemblies employed dicarboxylic acid
dimers,4,5 Watson-Crick base pairs6,7 and related [H+] inter-
faces.8 Because charge redistribution within these interfaces is
negligible, the presence of the proton network has little effect
on the ET rate; the only available mechanism for PCET arises
from the small dependence of the electronic coupling on the
proton coordinate within the interface.9 Nonetheless, such cases
are unusual in nature where proton displacement within a
proximal network often strongly influences ET.2,10 For instance,
ET between oxygen-evolving complex and the primary chlo-
rophyll P680 of Photosystem II is governed by proton transfer
of an intervening tyrosine (YZ), which is connected to a proton-
transfer network via a histidine (D1-H190).11 In cytochrome c
oxidase, ET between the binuclear copper center and heme a is
proton-controlled by an essential glutamate (E286)12,13 as is the
redox activation of oxygen at heme a3 by a proximal tyrosine,
(Y288).14,15 We have shown that the control exerted by proximal
proton-transfer networks on ET and oxygen bond activation
events may be captured in D- - -[H+]- - -A assemblies possessing
[H+] interfaces capable of supporting proton transfer. The
pronounced effect of a proton-transfer network on catalytic Os
O bond activation has been inferred from reactivity studies of
redox cofactors appended with properly oriented [H+] inter-
faces.16,17 Its effect on charge transfer has been inferred from
the excited state decay of the photoexcited donor of D- - -[salt-
bridge]- - -A assemblies.18-22 In no case, however, have reaction
Experimental Section
Materials. Silica gel 60 (70-230 and 230-400 mesh, Merck)
and Merck 60 F254 silica get (Precoated sheets, 0.2 mm thick)
were used for column and analytical thin-layer chromatography,
10.1021/jp049296b CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/27/2004

6316 J. Phys. Chem. B, Vol. 108, No. 20, 2004
Damrauer et al.
respectively. Solvents for synthesis were of reagent grade or
better and were dried according to standard methods.24 Spec-
troscopic experiments employed CH2Cl2 and THF (spectroscopic
grade, Burdick & Jackson) which were dried using standard
methods and stored under high-vacuum. All other reagents were
used as received. 1H NMR spectra were collected in CDCl3 or
THF-d8 (Cambridge Isotope Laboratories) using either a Mer-
cury 300 or an Inova 500 spectrometer at 25 °C. All chemical
shifts are reported using the standard δ notation in parts-per-
million. Mass spectral and NMR analyses were performed in
the MIT Department of Chemistry Instrumentation Facility
(DCIF). Porphyrin 1 was prepared using previously reported
procedures.25
was accomplished in a manner identical to that for 2. 1H NMR
(300 MHz, CDCl3, 25 °C) 8.72 (b, 4H), 7.58 (d, J ) 3.3 Hz,
1H), 7.46 (dd, J1 ) 4.0 Hz; J2 ) 1.1 Hz, 1H), 7.00 (d, J ) 1.0
Hz, 1H), 4.73 (s, 2H), 3.35 (b, 12H), 1.31 (s, 9H), 1.27 (s, 9H).
ESIMS ([M-N(CH3)4 + 2H+]) for C30H30N2O6, m/z: calcd
513.20, found 513.20.
Physical Methods. Samples for steady-state spectroscopic
measurements were contained within high-vacuum cells consist-
ing of a 1 cm path length clear fused-quartz cell, which was
connected to a 10-cm3 solvent reservoir via a graded seal. The
two chambers were isolated from the environment and from
each other by high-vacuum Teflon valves. An aliquot of 1 was
added to the high-vacuum cell and an initial aliquot of diimide
acceptor was added to the solvent reservoir. The transferring
solvent was removed from both compartments on a high vacuum
manifold (<10-5 Torr). Dry THF was added to the cell by
vacuum transfer to make a 25 µM solution of 1. In this
configuration, spectroscopy on 1 was examined initially. Then,
by opening the valve between the two compartments, 1 was
mixed with the diimide acceptor to form the associated complex
while maintaining high vacuum and exactly the same amount
of solvent and compound. After each measurement, the solution
was transferred to the high-vacuum cell and isolated from the
solvent reservoir by closing the high-vacuum Teflon valve. The
solvent reservoir was then reopened and charged with an
additional aliquot of the diimide; each time the transferring
solvent was removed under high vacuum, after which it was
mixed with the solution contained in the high vacuum cell. All
spectroscopy was performed at 298 K.
Synthesis of [N-(2,5-Di-tert-butylphenyl)-N′-(4-carboxy-
phenyl)naphthalene-1,8:4,5-tetracarboxydimide][TMA] (2).
A solution of 1,8:4,5-naphthalenedianhydride (5.0 g, 18.64
mmol), 4-aminobenzoate (2.55 g, 18.64 mmol) and 2,5-di-tert-
butylaniline (3.83 g, 18.64 mmol) in DMF (100 mL) was
refluxed under a nitrogen atmosphere for 16 h. The solution
was then cooled to 0 °C and N,N′-(2,5-di-tert-butylphenyl)-
1,8:4,5-naphthalenetetracarboxydiimide was removed as a side
product by filtration and washed with cold CH2Cl2. The filtrate
was concentrated under reduced pressure, redissolved in CH2-
Cl2 and washed with 5% HCl. The organic layer was then
separated, and the aqueous layer was extracted twice with 100
mL of CH2Cl2. The organic layers were combined, dried over
Na2SO4 and concentrated under reduced pressure. The resultant
residue was chromatographed on silica gel, eluting with CH2-
Cl2 to remove nonpolar impurities and then with CH2Cl2 and
MeOH (75:1 v:v) to give the desired product, which was
recrystallized from CH2Cl2 and hexanes. The purified product
was filtered and washed with cold CH2Cl2 and hexanes (1:3
v:v) to afford the pure carboxylic acid diimide in 12% yield
(1.28 g). 1H NMR (300 MHz, CDCl3, 25 °C) 8.89 (s, 4H), 8.36
(d, J ) 8.1 Hz, 2H), 7.63 (d, J ) 8.7 Hz, 1H), 7.51 (dd, J1 )
8.7 Hz; J2 ) 2.4 Hz, 1H), 7.50 (d, J ) 8.1 Hz, 2H), 7.04 (d, J
) 2.4 Hz, 1H), 1.35 (s, 9H), 1.29 (s, 9H). ESI-MS (MNa+) for
C36H33N2O6Na, m/z: calcd 597.20, found 597.20.
In the case of assembly 1:3, Ka was determined from
luminescence quenching of the S1 exited state as 3 was added
to solution. Steady-state fluorescence spectra were recorded on
an automated Photon Technology International QM 4 fluorom-
eter that is equipped with a 150-W Xe arc lamp and a
Hamamatsu R928 photomultiplier tube. In this experiment the
excitation wavelength was 560 nm and emission was collected
from 580 to 720 nm. Association constants (Ka) were determined
from Benesi-Hildebrand plots of the luminescence intensity
as a function of carboxylate concentration where the asymptotic
fluorescence intensity limit was obtained for 3 in a large
excess.26
The carboxylic diimide was converted to the corresponding
carboxylate salt by dissolving 123 mg (0.214 mmol) of the
acceptor in 25 mL of 2% MeOH in CH2Cl2 and cooling the
resultant solution to -78 °C. In dropwise fashion, 90.1 µL of
tetramethylammonium hydroxide (TMAOH) (25% in MeOH)
was added to the cold stirring solution, which was stirred for
an additional 5 min at -78 °C and then warmed to room
temperature. Concentration of the solution under reduced
pressure yielded a solid, which was dried for 14 h under vacuum.
Electrochemical experiments were carried out using a Bio-
analytical Systems (BAS) Model CV-50W potentiostat/gal-
vanostat. Cyclic voltammetry was performed in a two-
compartment cell using a platinum disk as the working electrode,
a Ag/AgCl reference electrode, and a platinum wire auxiliary
electrode. The supporting electrolyte used for electrochemistry
experiments was either 0.1 M n-tetrabutylammonium hexafluo-
rophosphate (TBAPF6) or perchlorate (TBAP). The solution in
the working compartment of the cell was deaerated by a nitrogen
stream. Background cyclic voltammograms of the electrolyte
solution were recorded prior to addition of the solid sample.
Redox couples were referenced to SCE by using a ferricenium-
ferrocene internal standard of 0.307 V vs SCE.27
1
A pure product was obtained in quantitative yield. H NMR
(300 MHz, CDCl3, 25 °C) 8.78 (b, 4H), 8.24 (d, J ) 8.4 Hz,
2H), 7.56 (d, J ) 8.7 Hz, 1H), 7.47 (dd, J1 ) 8.7 Hz; J2 ) 2.1
Hz, 1H), 7.26 (d, J ) 8.4 Hz, 2H), 7.00 (d, J ) 2.1 Hz, 1H),
3.33 (s, 12H), 1.30 (s, 9H), 1.23 (s, 9H). ESI-MS ([M -
N(CH3)4 + 2H+]) for C36H34N2O6, m/z: calcd 575.22, found
575.22.
Synthesis of [N-(2,5-Di-tert-butylphenyl)-N′-(carboxy-
methyl)naphthalene-1,8:4,5-tetracarboxydimide][TMA] (3).
Preparation of the methylene-spaced carboxylic acid naphthalene
diimide acceptor was analogous to the foregoing synthesis of
2, except that glycine was used in place of 4-aminobenzoate.
1H NMR (300 MHz, CDCl3, 25 °C) 8.83 (b, 4H), 7.60 (d, J )
5.4 Hz, 1H), 7.49 (dd, J1 ) 3.6 Hz; J2 ) 1.5 Hz, 1H), 7.00 (d,
J ) 1.2 Hz, 1H), 5.01 (s, 2H),1.31 (s, 9H), 1.25 (s, 9H). ESIMS
(MH+) for C30H30N2O6, m/z: calcd 513.20, found 513.20.
Spectroelectrochemical studies were performed in an optically
transparent 1 mm thin-layer cell using a platinum mesh working
electrode, a Ag/AgCl reference electrode, and a platinum wire
auxiliary electrode. The supporting electrolyte used was 0.1 M
TBAPF6 in THF. Oxidation was carried out at 1200 mV versus
the reference electrode and absorption spectra were recorded at
selected time intervals. Extinction coefficients for the oxidized
porphyrin (1+) were calculated based on the disappearance of
initial porphyrin absorbance. Comparison of the spectra for 1
before the oxidation process and after a short recovery period
Deprotonation of the methylene spaced carboxylic diimide

PCET in a Porphyrin Assembly
J. Phys. Chem. B, Vol. 108, No. 20, 2004 6317
following oxidation showed that virtually no porphyrin had
decomposed during the experiment.
SCHEME 1
Computational Methods. Density functional theory (DFT)
calculations were executed at the local density approximation
(LDA) level of theory using the Amsterdam Density Functional
program.28,29 The calculations were performed on a home-built
Linux cluster consisting of 36 processors running in parallel.
Gradient corrections were introduced by using the Becke (B)
exchange functional30 and the Lee-Yang-Parr31 (LYP) cor-
relation functional. Relativistic corrections were included by
using the scalar zero-order regular approximation (ZORA).32,33
C and H were described by a Slater-type orbital double-ê basis
set augmented by one set of polarization functions. N, O, and
Zn were described by a Slater-type orbital triple-ê basis set,
also augmented by one set of polarization functions. Non-
hydrogen atoms were assigned a relativistic frozen core
potential, treating as core the shells up to and including 2p for
Zn, and 1s for C, N, and O.
reference spectrum was taken at negative time. Typically, 3000
shots were accumulated per exposure and each time point was
visited 30 times. Spectra taken from forward and reverse scans
were averaged.
For single wavelength TA kinetics measurements, the pump
beam was mechanically chopped at ω ) 500 Hz. The probe
beam was spectrally resolved in the monochromator, and a
single wavelength ((2 nm) was measured on an amplified
photodiode. This signal served as the input for a digital lock-in
amplifier (Stanford Research Systems SR830) locked to ω. The
cross correlation of the excitation beam with a single wavelength
of the probe beam has a fwhm ) 300 fs for the employed
monochromator configuration. Typically, the time range sampled
was divided into 80 steps, including several steps at negative
time. Ten forward and reverse scans were averaged.
Transient Spectroscopy. Picosecond transient absorption
(TA) and emission (TE) measurements were performed on a
chirped pulse amplified Ti:sapphire laser system. Sub-100 fs
laser pulses were generated in a mode-locked Ti:sapphire
oscillator (Coherent Mira 900), which was pumped by a 5 W
cw Coherent Verdi solid-state, frequency-doubled Nd:YVO4
laser. The 76-MHz output was amplified in a regenerative
amplifier cavity (BMI Alpha-1000) pumped by a 12 W Thales
Jade Nd:YLF solid-state laser to generate a 1 kHz pulse train
with central wavelength λo ) 810 nm and a power of ∼900
µJ/pulse. Pulse durations of 100 fs were measured with a
Positive Light SSA single-shot autocorrelator. The output beam
was split, with the majority component frequency doubled in a
BBO crystal to produce 405 nm excitation pulses.
Results and Discussion
Our efforts to observe PCET by direct spectroscopic means
have focused on porphyrins bearing an amidinium directly fused
to porphyrin macrocycles.25 The presence of this amidinium
moiety on the porphyrin macrocycle permits the formation of
the supramolecular D- - -[H+]- - -A systems shown in Scheme
1. The photoexcitable Zn(II) porphyrin donor is placed proximal
to a diimide carboxylate acceptor via the well-defined two-point
hydrogen bond of the amidinium-carboxylate salt bridge, which
features a favorable electrostatic interaction augmented by two
favorable secondary electrostatic interactions.35 The confluence
of these two effects engenders significant stability on supramo-
lecular complexes 1:2 and 1:3. A signature of the salt bridge
formation is the downfield shift of amidinium protons involved
in hydrogen bonding (NHax) and an insensitivity of the chemical
shift for the amidinium protons external to the salt bridge (NHeq).
As exemplified by the 1H NMR spectrum for the Ni(II)
derivative of the 1:3 complex shown in Figure 1, the chemical
shift of the NHax protons moves downfield by nearly 1.8 ppm
upon their hydrogen bonding association to the carboxylate,
whereas the chemical shift of the NHeq protons changes by
<0.05 ppm. In principle, the shift of the NHax protons permits
the association constant (Ka) of the supramolecular complex to
be determined. However, the stability of the supramolecular
complex in THF is so high that association constants from
binding isotherms cannot be reliably determined by NMR
spectroscopy.36 Under such high binding conditions, Ka is better
determined from steady-state and time-resolved luminescence
measurements (vide infra).
Transient emission kinetics were measured on a Hamamatsu
C4334 Streak Scope streak camera which has been described
elsewhere.34 In this experiment, the excitation power was
attenuated to ∼200 nJ/pulse, and the emission was collected
from 580 to 720 nm with a 20 ns timebase. The 610-620 nm
range was integrated for fitting. Samples were prepared in an
identical way to those prepared in steady-state spectroscopic
experiments except that the cells had a 2 mm path length and
solutions of 1 were 100 µM. All spectroscopy was performed
at 298 K.
Transient absorption experiments also employed a 405-nm
excitation pulse whose power was typically attenuated to 2 µJ/
pulse as higher powers resulted in thermal lensing problems. A
continuum probe pulse was generated by focusing 2 µJ/pulse
of the λo ) 810 nm output onto a sapphire substrate. The probe
pulse carried a chirp of ca. 1 ps (over 450-700 nm), which
was characterized by cross-correlation of the pump and probe
in a 5% solution of benzene in methanol. Time resolution was
achieved by propagating the excitation beam along a computer-
controlled, 1.70 m long optical delay line at 1 µm precision
(Aerotech ATS 62150). The polarization of pump and probe
pulses were set at the magic angle (θm ) 54.7°) and focused
collinearly in the sample, which was stirred in the axis of beam
propagation using a mini-magnetic stirbar. This procedure was
implemented to miminize the effects of thermal lensing while
maintaining small sample volumes.
The NMR results also establish that π-stacking between the
amidinium porphyrin and diimide acceptor is not an important
association mechanism. The chemical shift of the â-proton
adjacent to the amidinium group at 8.85 ppm, and the broad
multiplet resulting from the remaining seven â-protons (8.5-
8.7 ppm) are virtually invariant with the addition of 3. The slight
upfield shift of roughly 0.04 ppm of all eight â-protons upon
carboxylate binding is presumably due to an increase in electron
density within the porphyrin π-system. As the â-protons are a
sensitive measure of π-stacking in porphyrin systems,37 the
Transient absorption spectra were recorded at discrete times
after excitation over the wavelength range 475-675 nm. The
probe beam exiting the sample was coupled into a liquid light
guide to spatially homogenize the beam. The spectrum was then
resolved in the monochromator (ISA Instruments, TRIAX 320)
and recorded on a CCD camera (Andor technology). The

6318 J. Phys. Chem. B, Vol. 108, No. 20, 2004
Damrauer et al.
Figure 2. Singlet (S1) emission kinetics for a 6 × 10-5 M solution of
1 in THF (solid) and in the presence of two equivalents of 3 (dashed).
The inset shows static quenching of the emission spectrum for 1 (solid)
(2.5 × 10-5 M) upon addition of three equivalents of 3 (dashed).
Figure 1. 1H NMR spectra of the Ni(II) derivative of 1 (bottom) and
associated to 3 (top) in THF-d8. The displayed spectral range captures
the amidinium protons internal (NHax) and external (NHeq) to the salt
bridge interface. Labels (a) and (b) represent the resonances of the
â-pyrrole protons adjacent to and distal from the amidinium functional-
ity, respectively. The large resonance labeled (c) corresponds to the
aromatic naphthalene protons of 3. Resonances (a) and (b) display
negligible shifts upon amidinium-carboxylate binding to form the
supramoelculae complex.
port to such a degree that it cannot effectively compete with
the internal relaxation dynamics of the electronically excited
porphyrin donor.
Replacement of the phenylene spacer with a methylene group
results in a considerable shortening of the edge-to-edge through
bond distance while maintaining a comparable driving force for
a PCET reaction (-460 mV).38 Under these conditions, the S1
emission of porphyrin 1 in THF is quenched upon its association
to 3; static quenching of the S1 fluorescence of 1 by ∼40% is
observed in the limit of salt-bridge formation (Figure 2 inset).
The Ka for the 1:3 complex may be determined from the
concentration dependence of this luminescence quenching. A
Benesei-Hildebrand fit of the intensity of the S1 luminescence
band with added carboxylate yields Ka(1:3) ) (2.4 ( 0.6) ×
104. A time-resolved fluorescence decay trace of the quenching
process is shown in Figure 2. Compound 1 exhibits single-
exponential decay kinetics with an observed lifetime of τobs(1)
) 2.19 ( 0.07 ns (298 K, THF), corresponding directly to the
lifetime of the S1 state. A marked decrease in lifetime of the S1
state is observed and decay traces become biexponential upon
1:3 formation. The minor decay component corresponds to the
excited-state relaxation of unbound species 1 and the major
decay component is attributed to a quenched S1 excited state
of 1 associated to 3 (τobs(1:3) ) 0.72 ( 0.03 ns). A quantitative
analysis of the ratio of these two components leads to an
alternative measure of Ka (1:3) ) (2.8 ( 0.6) × 104, which is
consistent with the result obtained from the steady-state titration
described above.40 The quenching rate is found to be (9.5 (
0.6) × 108 s-1 (k ) 1/τobs(1:3) - 1/τobs(1:benzoate)). This
calculation employs the natural lifetime of 1 bound to benzoate
(τo ) 2.3 ns) since, as mentioned above, the S1 lifetime is
slightly perturbed by salt-bridge formation, even in the absence
of a redox-inactive quencher.
constancy of their chemical shift with supramolecular formation
indicates that that π-stacking is not an important association
mechanism. DFT calculations of the supramolecular complex
support this contention. Energy optimized geometry calculations
confirms association via the two-point hydrogen bond (Figure
S1); a π-stacked complex was not observed. Moreover, geom-
etry optimization shows that there is no opportunity for the
acceptor to achieve face-to-face contact with the donor while
maintaining the two-point hydrogen bond. The geometry
optimized structure shows one other notable featuresthe ami-
dinium is rotated by ∼76° with respect to the porphyrin
macrocycle. This canting leads to the electronic decoupling of
the amidinium functionality from the porphyrin chromophore.
Consequently, the absorption spectrum of 1 in THF, typical of
a standard Zn(II) porphyrin four-orbital model (λabs,max(B) )
428 nm (ꢀ ) 4.0 × 104 M-1 cm-1); λabs,max(Q1,0) ) 562 nm (ꢀ
) 1.4 × 104 M-1 cm-1); λabs,max(Q0,0) ) 601 nm (ꢀ ) 6.8 ×
103 M-1 cm-1)), is not perturbed upon association of the
amidinium to the carboxylate acceptors. Owing to the insensi-
tivity of these spectral features to the protonation state of the
amidinium, we are only afforded optical signatures pertinent to
the electron-transfer component of the PCET reaction.
There exists a -500 mV driving force for electron transfer
from the singlet excited state (S1) to the diimide acceptor of
the 1:2 complex.38 Nonetheless, PCET is not observed. Elec-
tronic excitation of 1 affords S1 luminescence (λem,max(Q0,0) )
612 nm and λem,max(Q1,0) ) 665 nm), which is not quenched
upon titration of 0-12 equivalents of 2 per equivalent of 1.
Conversely, time-resolved emission experiments show that the
S1 lifetime monotonically increases from 2.19 ns (THF, 298
K) for unbound 1 to 2.3 ns for 1:2. This result is comparable to
the lifetime increase observed when 1 is titrated with benzoate,
a base that binds but cannot be reduced. Transient absorption
measurements of the 1:2 complex reveal that the S1 excited-
state relaxes to the triplet excited state (T1) with the normal
intersystem-crossing rate of kisc ) (4.3 ( 0.6) × 108 s-1. A
450 mV decrease in reducing power accompanies this relaxation
to T1,39 thereby precluding PCET from occurring. Thus, charge
transport from S1 is too slow to compete with T1 r S1
intersystem crossing. The proton-network retards charge trans-
Transient absorption spectroscopy was undertaken with the
aim of defining the quenching mechanism within the 1:3
complex. For these studies, THF was chosen as the solvent
because of its ability to axially bind the Zn(II) center of 1,
thereby inhibiting molecular aggregation at the relatively high
concentrations needed to acquire transient absorption spectra.41
Figure 3 shows transient absorption spectra of 1 at 500 ps, 1 ns
and 2 ns following laser excitation. The 500 ps spectrum consists
of broad transient absorption of the S1 porphyrin excited state
that dominates the visible spectral region.42 Superimposed on
this absorption envelope are two prominent bleach features at
562 and 602 nm, corresponding to loss of ground-state Q-band
absorptions. As shown in Figure 4a (solid circles), the TA signal
of S1 at 615 nm decays with a rate constant of 4.6 × 108 s-1
;

PCET in a Porphyrin Assembly
J. Phys. Chem. B, Vol. 108, No. 20, 2004 6319
anion of the diimide is known to absorb at λmax ) 475 and 610
nm.43 Because the quenching process is in kinetic competition
with intersystem crossing and the optical signatures for the S1
and T1 excited states are strong, the transient signals of the
charge separated intermediates are overwhelmed. This can be
observed from the decay trace of the transient signal of the 1:3
complex at 615 nm (open circles, Figure 4a). The overall
transient decay for 1 is accelerated in the presence of 3, and
the yield of T1 is attenuated, consistent with a quenching process.
But rise and decay associated with charge separation and
recombination is not clearly discernible; rather a weak shoulder
at early times in the decay trace is observed.
We attempted to minimize the contribution of T1 r S1
dynamics to the transient signal by moving to longer wave-
lengths, where the S1 and T1 absorbance is weaker. However,
the transient spectrum at λ > 650 nm is complicated by S1
stimulated emission. S1 stimulated emission leads to an increased
photon flux at the detector. This is manifested in an apparent
decrease in the absorption of the S1. Consequently, as shown
in Figure 4b (solid circles), a rise in ∆OD at λprobe ) 660 nm
is observed as S1 is lost and T1 is formed. We note that the
intensity of the stimulated emission is governed by the S1
population. Hence, the growth of the TA signal in 4b occurs
with the same kinetics as the disappearance of the TA signal in
Figure 4a (kobs ) (4.6 ( 0.6) × 108 s-1). Owing to the smaller
absorbance of the S1 and T1 excited states in this redder spectral
region, transient signals of the Zn(II) porphyrin cation radical
(ꢀ660 ) 2750 M-1 cm-1) are more pronounced upon the
formation of the 1:3 complex. A rise and decay of the transient
signal (open circles) is now clearly observed. However, the
analysis is complicated by the superposition of the TA signals
for the forward and reverse charge-transfer events within the
1:3 complex on the rising TA signal of 1.
Figure 3. Pump-probe transient spectra of 1 (50 µM) in THF at 298
K collected at 500 ps, 1 ns, and 2 ns.
The PCET dynamics of 1:3 are greatly simplified when they
are probed at the specific wavelength of 650 nm. The combina-
tion of rise and decay of ∆OD at the same rate results in an
isosbestic point at 650 nm. Because the transient absorption
spectrum of 1 at 650 nm does not change as a function of time,
kinetics collected on 1 at this wavelength appear as a step
function as shown in Figure 4c (solid circles). For our purposes,
this steplike behavior of the transient absorption profile of 1 is
extremely useful for two reasons. First, it provides a kinetics-
free or flat background on which to examine PCET product
formation in 1:3. Second, we are afforded the fortunate situation
that the porphyrin cation radical, formed as a result of PCET,
has significant intensity at this wavelength (∆ꢀ650 ≈ 3000 M-1
cm-1).44
Pump-probe kinetics collected for 1:3 at 650 nm are shown
in Figure 4c (open circles). In contrast to the featureless transient
collected for 1, the time profile of 1:3 shows a prominent rise
in ∆OD followed by subsequent decay corresponding to the
appearance and disappearance, respectively, of the porphyrin
cation radical. Quantitatively, PCET rate constants can be
determined from a model that accounts for the kinetics of S1,
T1, and the PCET photoproduct. The trace in Figure 4c was fit
(solid line) by a sum of two variable exponentials with rate
constants (kisc + kPCET(fwd)) and kPCET(rev) (where kPCET(rev)
refers to the rate constant with which ground state is reformed
as a result of back PCET); intersystem crossing dynamics for
unbound porphyrin species in the sample cell can be ignored
as they do not contribute to the temporal change of ∆OD at
this wavelength. This analysis yields forward and reverse PCET
Figure 4. Pump-probe kinetics for 1 (solid circles) and 1:3 (open
circles) probed at (a) 615 nm, (b) 660 nm and (c) the S1-T1 isosbestic
point at 650 nm. The supramolecular complex was formed from addition
of two equivalents of 3 to 1 (50 µM) in THF. Fits to the data are shown
by solid lines. See text for details of the fitting procedure.
previous studies of Zn(II) porphyrins shows that this decay is
predominantly due to intersystem crossing to the T1 excited
state.42 Unfortunately, the TA features associated with these S1
dynamics overlap with the optical signatures of the PCET
products; spectroelectrochemical measurements of the porphyrin
cation radical yields λmax ) 412 and 637 nm and the radical
rates constants of kPCET(fwd) ) (9 ( 1) × 108 s-1 and kPCET
-
(rev) ) (14 ( 1) × 108 s-1, respectively. It is noted that the

6320 J. Phys. Chem. B, Vol. 108, No. 20, 2004
Damrauer et al.
(7) Shafirovich, V. Y.; Courtney, S. H.; Ya, N.; Geacintov, N. E. J.
Am. Chem. Soc. 1995, 117, 4920-4929.
forward PCET rate constant obtained from transient absorption
measurements is in excellent agreement with quenching rate
constant measured by time-resolved fluorescence (vide supra).
This result establishes that PCET occurs exclusively from the
S1 porphyrin excited state. It is noteworthy that the forward
PCET rate constants are nearly 2 orders of magnitude slower
than that measured for covalently linked Zn(II) porphyrin-
acceptor dyads of comparable driving force in solvents of
comparable dielectric constant.45,46 Although these systems
cannot be rigorously compared owing to possible differences
in electronic coupling, the significant disparity between the D-A
kinetics of covalent and salt-bridge networks speak directly to
the pronounced effect that a proximal proton-transfer network
can have in mediating electron-transfer events.
(8) Ghaddar, T. H.; Castner, E. W.; Isied, S. S. J. Am. Chem. Soc.
2000, 122, 1233-1234.
(9) Cukier, E.; Daniels, S.; Vinson, E.; Cave, R. J. J. Phys. Chem. A
2002, 106, 11240-11247.
(10) Hammes-Schiffer, S. Acc. Chem. Res. 2001, 34, 273-281.
(11) Tommos, C.; Babcock, G. T. Biochim. Biophys. Acta 2000, 1458,
199-219.
(12) A¨ ldelroth, P.; Karpefors, M.; Gilderson, G.; Tomson, F. L.; Gennis,
R. B.; Brzezinski, P. Biochim. Biophys. Acta 2000, 1459, 533-539.
(13) Namslauer, A.; Aagaard, A.; Katsonouri, A.; Brzezinski, P.
Biochemistry 2003, 42, 1488-1498.
(14) Karpefors, M.; A¨ ldelroth, P.; Namslauer, A.; Zhen, Y.; Brzezinski,
P. Biochemistry 2000, 39, 14 664-14 669.
(15) Gennis, R. B. Biochim. Biophys. Acta 1998, 1365, 241-248.
(16) Chang, C. J.; Chng, L. L.; Nocera, D. G. J. Am. Chem. Soc. 2003,
125, 1866-1876.
(17) Chng, L. L.; Chang, C. J.; Nocera, D. G. Org. Lett. 2003, 5, 2421-
2424.
Concluding Remarks
(18) Roberts, J. A.; Kirby, J. P.; Nocera, D. G. J. Am. Chem. Soc. 1995,
117, 8051-8052.
The propensity of proton transfer networks to retard charge
transfer rates has practical consequences for mechanistic studies
of PCET reactions. Attenuated rates translate to low yields of
PCET intermediates. For this reason, it has been difficult to
measure PCET reaction kinetics directly by time-resolved
methods. Charge transfer has only been observed across
hydrogen-bonded interfaces in which charge redistribution is
negligible.4,47 For the systems described here, the proton network
strongly perturbs charge transport. In 1:2, PCET is not observed;
in 1:3, PCET intermediates are spectrally uncovered only when
the transient difference signal between S1 and T1 excited states
is minimized. This procedure, which is similar to one previously
exploited in studies of D-A dyads48 and heme protein-protein
complexes,49 opens the door to a host of future experiments
designed to directly monitor rates of electron transfer that are
strongly coupled to proton motion.
(19) Deng, Y. Q.; Roberts, J. A.; Peng, S. M.; Chang, C. K.; Nocera,
D. G. Angew. Chem., Int. Ed. Engl. 1997, 36, 2124-2127.
(20) Roberts, J. A.; Kirby, J. P.; Wall, S. T.; Nocera, D. G. Inorg. Chim.
Acta 1997, 263, 395-405.
(21) Kirby, J. P.; Vandantzig, N. A.; Chang, C. K.; Nocera, D. G.
Tetrahedron Lett. 1995, 36, 3477-3480.
(22) Kirby, J. P.; Roberts, J. A.; Nocera, D. G. J. Am. Chem. Soc. 1997,
119, 9230-9236.
(23) Berman, A.; Izraeli, E. S.; Levanon, H.; Wang, B.; Sessler, J. L. J.
Am. Chem. Soc. 1995, 117, 8252-8257.
(24) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals, 4th ed.; Butterworth-Heinmann: Oxford, 1996.
(25) Yeh, C.-Y.; Miller, S. E.; Carpenter, S. D.; Nocera, D. G. Inorg.
Chem. 2001, 40, 3643-3646.
(26) Connors, K. A. Binding Constants; Wiley: New York, 1987.
(27) Bard, A. J.; Faulkner, L. R. Electrochemical Methods. Fundamentals
and Applications; John Wiley: New York, 1980.
(28) ADF2000.02, Vrije Universiteit Amsterdam: Amsterdam, The
Netherlands, 1999.
(29) Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra,
C.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem.
2001, 22, 931-967.
In closing, the importance of a proton network in controlling
ET rates is becoming more apparent as the structural details of
biological systems are revealed.50-64 The foregoing results
establish that this emerging principle in biological charge
transport can be modeled and studied directly in D- - -[H+]- - -
A systems featuring proton-transfer networks.
(30) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100.
(31) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(32) Van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1993, 99, 4597-4610.
(33) Van Lenthe, E.; Ehlers, A.; Baerends, E. J. J. Chem. Phys. 1999,
110, 8943-8953.
(34) Loh, Z.-H.; Miller, S. E.; Chang, C. J.; Carpenter, S. D.; Nocera,
D. G. J. Phys. Chem. A 2002, 106, 11 700-11 708.
(35) Pranata, J.; Wierschke, S. G.; Jorgensen, W. L. J. Am. Chem. Soc.
1991, 113, 2810-2819.
(36) To assert the existence of a chemical equilibrium, the analytical
response must be a nonlinear function of the ligand concentration (see
Person, W. B. J. Am. Chem. Soc. 1965, 87, 167-170). This is most
prominent when the equilibrium concentration of the bound complex is on
the same order of magnitude as the concentration of the more dilute of the
two unbound components. However, the high concentration conditions
required for NMR spectroscopy on 1:3 mean that over 99% of 1 is bound,
and the nonlinear region of the binding isotherm cannot be observed with
any fidelity.
Acknowledgment. We thank Dr. Chen-Yu Yeh and Dr.
Scott Miller for synthesis and initial experiments on 1:2. N.H.D.
gratefully acknowledges the NIH for support of a postdoctoral
fellowship. J.R. thanks the Fannie and John Hertz Foundation
for a pre-doctoral fellowship. We kindly thank O.T.P. This work
was supported by a grant from the National Institutes of Health
(GM 47274).
Supporting Information Available: Structure and xyz
coordinates for the gas-phase density functional calculation of
the 1:3 complex. This material is available free of charge via
the Internet at http://pubs.acs.org.
(37) Kobuke, Y.; Ogawa, K. Bull. Chem. Soc. Jpn. 2003, 76, 689-708.
(38) The S1 excited-state of 1 in THF is 2.04 eV (298 K); cyclic
voltammetry yields one-electron reduction potentials for amidinium por-
phyrin donor 1 and carboxylate diimide acceptor 2 of E1/2(12+/+) ) 1060
mV and E1/2(2-/2- ) -480 mV vs Ag/AgCl in THF, respectively. The
one-electron reduction potentials for carboxylate diimide acceptor 3 is
References and Notes
1/2(3-/2-) ) -520 mV vs Ag/AgCl in THF.
(39) Estimated based on the shifts in phosphorescence spectra compared
(1) Chang, C. J.; Brown, J. D. K.; Chang, M. C. Y.; Baker, E. A.;
Nocera, D. G. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-
VCH: Weinheim, Germany, 2001; Vol. 3.2.4, pp 409-461.
(2) Cukier, R. I.; Nocera, D. G. Annu. ReV. Phys. Chem. 1998, 49,
337-369.
E
with fluorescence spectra of related Zn porphyrins (see Gradyushko, A.
T.; Tsvirko, M. P. Opt. Spectrosc. 1971, 31, 291-295).
(40) Ka ) Q(Q + 1)/([3]0(Q + 1) - [1]0Q), where Q is the ratio of the
pre-exponential factors of the short (0.72 ns) and long (2.2 ns) components
which also corresponds to the ratio [1:3]/[1].
(3) Chang, C. J.; Chang, M. C. Y.; Damrauer, N. H.; Nocera, D. G.
Biochim. Biophys. Acta 2003, in press.
(4) Turro´, C.; Chang, C. K.; Leroi, G. E.; Cukier, R. I.; Nocera, D. G.
(41) In noncoordinating solvents such as CH2Cl2, aggregation is observed
as indicated by the appearance of a green solution, resulting from a shift in
the Q-bands from 556 and 592 nm to 573 and 612 nm, respectively.
(42) Rodriguez, J.; Kirmaier, C.; Holten, D. J. Am. Chem. Soc. 1989,
111, 6500-6506.
J. Am. Chem. Soc. 1992, 114, 4013-4015.
(5) de Rege, P. J. F.; Williams, S. A.; Therien, M. J. Science 1995,
269, 1409-1413.
(6) Sessler, J. L.; Wang, B.; Springs, S. L.; Brown, C. T. In
ComprehensiVe Supramolecular Chemistry; Murakami, Y., Ed.; Pergamon
Press: Oxford, 1996; Vol. 4; p 311-336 and references therein.
(43) Rogers, J. E.; Kelly, L. A. J. Am. Chem. Soc. 1999, 121, 3854-
3861.

PCET in a Porphyrin Assembly
J. Phys. Chem. B, Vol. 108, No. 20, 2004 6321
(44) Neither the singlet (unpublished result) nor triplet (see ref 43)
excited states of 3 exhibit absorption at 650 nm. These intermediates are
the products of an energy transfer quenching mechanism, which in this
case is not plausible: energy transfer from the porphyrin S1 excited state
to produce 3* is energetically uphill and the production of 3* is a spin-
forbidden process.
(45) Wasielewski, M. R.; Niemczyk, M. P.; Svec, W. A.; Pewitt, E. B.
J. Am. Chem. Soc. 1985, 109, 1080-1082.
(46) Osuka, A.; Yoneshima, R.; Shiratori, H.; Okada, S.; Taniguchi, S.;
Mataga, N. Chem. Commun. 1998, 1567-1568.
(53) Siegbahn, P. E. M.; Eriksson, L.; Himo, F.; Pavolv, M. J. Phys.
Chem. B 1998, 102, 10 622-10 629.
(54) Carra, C.; Iordanova, N.; Hammes-Schiffer, S. J. Am. Chem. Soc.
2003, 125, 10 429-10 436.
1
3
(55) Westphal, K. L.; Tommos, C.; Cukier, R. I.; Babcock, G. T. Curr.
Opin. Plant Biol. 2000, 3, 236-242.
(56) Tommos, C. Philos. Trans. R. Soc. London B 2002, 357, 1383-
1394.
(57) Cukier, R. I.; Seibold, S. A. J. Phys. Chem. B 2002, 106, 12 031-
12 044.
(58) Aubert, C.; Vos, M. H.; Mathis, P.; Eker, A. P. M.; Brettel, K.
Nature 2000, 405, 586-590.
(59) Hille, R.; Anderson, R. F. J. Biol. Chem. 2001, 276, 31 193-31 201.
(60) Knapp, M. J.; Rickert, K.; Klinman, J. P. J. Am. Chem. Soc. 2002,
124, 3865-3874.
(61) Cohen, A.; Klinman, J. P. Acc. Chem. Res. 1998, 31, 397-404.
(62) Crofts, A. R.; Guergova-Kuras, M.; Kuras, R.; Ugulava, N.; Li, J.
Y.; Hong, S. J. Biochim. Biophys. Acta-Bioenerg. 2000, 1459, 456-456.
(63) Berry, E. A.; Guergova-Kuras, M.; Huang, L.-S.; Crofts, A. R. Annu.
ReV. Biochem. 2000, 69, 1005-1075.
(64) Graige, M. S.; Paddock, M. L.; Bruce, J. M.; Feher, G.; Okamura,
M. Y. J. Am. Chem. Soc. 1996, 118, 9005-9016.
(47) Sessler, J. L.; Sathiosatham, M.; Brown, C. T.; Rhodes, T. A.;
Wiederrecht, G. J. Am. Chem. Soc. 2001, 123, 3655-3660.
(48) Greenfield, S. R.; Svec, W. A.; Gosztola, D.; Wasielewski, M. R.
J. Am. Chem. Soc. 1996, 118, 6767-6777.
(49) Liang, N.; Mauk, A. G.; Pielak, G. J.; Johnson, J. A.; Smith, M.;
Hoffman, B. M. Science 1988, 240, 311-313.
(50) Stubbe, J.; Nocera, D. G.; Yee, C. S.; Chang, M. C. Y. Chem. ReV.
2003, 103, 2167-2201.
(51) Chang, M. C. Y.; Yee, C. S.; Stubbe, J.; Nocera, D. G. Proc. Natl.
Acad. Sci. U.S.A. 2003, submitted for publication.
(52) Yee, C. S.; Chang, M. C. Y.; Ge, J.; Nocera, D. G.; Stubbe, J. J.
Am. Chem. Soc. 2003, asap.