Long-range electron transfer in zinc-phthalocyanine-oligo(Phenylene- ethynylene)-based donor-bridge-acceptor dyads

Article abstract of DOI:10.1021/ic3013552

In the context of long-range electron transfer for solar energy conversion, we present the synthesis, photophysical, and computational characterization of two new zinc(II) phthalocyanine oligophenylene-ethynylene based donor-bride-acceptor dyads: ZnPc-OPE-AuP⁺ and ZnPc-OPE-C ₆₀. A gold(III) porphyrin and a fullerene has been used as electron accepting moieties, and the results have been compared to a previously reported dyad with a tin(IV) dichloride porphyrin as the electron acceptor (Fortage et al. Chem. Commun.2007, 4629). The results for ZnPc-OPE-AuP⁺ indicate a remarkably strong electronic coupling over a distance of more than 3 nm. The electronic coupling is manifested in both the absorption spectrum and an ultrafast rate for photoinduced electron transfer (k_PET = 1.0 × 10¹² s^-1). The charge-shifted state in ZnPc-OPE-AuP ⁺ recombines with a relatively low rate (k_BET = 1.0 × 10⁹ s^-1). In contrast, the rate for charge transfer in the other dyad, ZnPc-OPE-C₆₀, is relatively slow (k _PET = 1.1 × 10⁹ s^-1), while the recombination is very fast (k_BET ≈ 5 × 10¹⁰ s^-1). TD-DFT calculations support the hypothesis that the long-lived charge-shifted state of ZnPc-OPE-AuP⁺ is due to relaxation of the reduced gold porphyrin from a porphyrin ring based reduction to a gold centered reduction. This is in contrast to the faster recombination in the tin(IV) porphyrin based system (k_BET = 1.2 × 10¹⁰ s ^-1), where the excess electron is instead delocalized over the porphyrin ring.

Full text of DOI:10.1021/ic3013552

Article
pubs.acs.org/IC
Long-Range Electron Transfer in Zinc-Phthalocyanine-
Oligo(Phenylene-ethynylene)-Based Donor-Bridge-Acceptor Dyads
Erik Goransson,† Julien Boixel,‡ Jerome Fortage,‡ Denis Jacquemin,*,‡,§ Hans-Christian Becker,†
́
̂
̈
Errol Blart,‡ Leif Hammarstrom,*,† and Fabrice Odobel*,‡
̈
†Physical Chemistry, Department of Chemistry - Ångstrom, Uppsala University, Box 523, 751 20 Uppsala, Sweden
̈
‡LUNAM Universite,
6230; 2, rue de la Houssinier
́ ́ ́ ̀ ́
Universite de Nantes, CNRS, Chimie et Interdisciplinarite: Synthese, Analyse, Modelisation (CEISAM), UMR
̀
e − BP 92208 − 44322 Nantes Cedex 3, France
§Institut Universitaire de France, 103 blvd St Michel, 75005 Paris Cedex 5, France
S
* Supporting Information
ABSTRACT: In the context of long-range electron transfer
for solar energy conversion, we present the synthesis,
photophysical, and computational characterization of two
new zinc(II) phthalocyanine oligophenylene-ethynylene based
donor-bride-acceptor dyads: ZnPc-OPE-AuP+ and ZnPc-
OPE-C60. A gold(III) porphyrin and a fullerene has been
used as electron accepting moieties, and the results have been
compared to a previously reported dyad with a tin(IV) dichloride porphyrin as the electron acceptor (Fortage et al. Chem.
Commun. 2007, 4629). The results for ZnPc-OPE-AuP+ indicate a remarkably strong electronic coupling over a distance of more
than 3 nm. The electronic coupling is manifested in both the absorption spectrum and an ultrafast rate for photoinduced electron
transfer (kPET = 1.0 × 1012 s−1). The charge-shifted state in ZnPc-OPE-AuP+ recombines with a relatively low rate (kBET = 1.0 ×
109 s−1). In contrast, the rate for charge transfer in the other dyad, ZnPc-OPE-C60, is relatively slow (kPET = 1.1 × 109 s−1), while
the recombination is very fast (kBET ≈ 5 × 1010 s−1). TD-DFT calculations support the hypothesis that the long-lived charge-
shifted state of ZnPc-OPE-AuP+ is due to relaxation of the reduced gold porphyrin from a porphyrin ring based reduction to a
gold centered reduction. This is in contrast to the faster recombination in the tin(IV) porphyrin based system (kBET = 1.2 × 1010
s−1), where the excess electron is instead delocalized over the porphyrin ring.
recent years, and bridge structures with very small distance
dependencies have been found.12 However, most of these
investigations have been on donor-bridge-acceptor systems
with fairly weak electronic coupling between the three
components, and thus valuable information could be gained
from complementary studies of long-range electron transfer in
systems with high intercomponent electronic coupling.13
Furthermore, while electron transfer has been extensively
studied in zinc porphyrin based systems,2a,14 the use of zinc
phthalocyanines (ZnPc) is much less explored.15 Zinc
phthalocyanine is a fairly easily oxidized dye, which makes it
a good electron donor, and it displays a strong absorption band
in the red domain of the visible spectrum.15d,16 This feature
nicely complements the absorption of common light harvesting
dyes, which tend to absorb in the blue or green part of the
visible spectrum.17 Herein we present two new donor-bridge-
acceptor dyads, designed to undergo long-range photoinduced
electron transfer using zinc phthalocyanine as the chromophore
and electron donor. We have linked the zinc phthalocyanine to
established electron acceptors: a gold(III) porphyrin
(AuP+)13d,18 or a pyrrolidino functionalized fullerene (C60)
INTRODUCTION
■
Electron transfer is a central reaction in many biological
systems1 as well as in many molecular devices for solar energy
conversion.2 In biological systems, electron transfer occurs in
the initial steps of both Photosystem I and Photosystem II,3 in
the electron transport between iron−sulfur clusters in hydro-
genases4 and along amino acid side chains in proteins.1a,5 In
artificial systems photoinduced electron transfer is a funda-
mental step to convert solar energy into electricity or into fuels.
For example, organic bulk heterojunction solar cells,6 dye
sensitized solar cells,7 and artificial photosynthetic systems for
solar fuels production,8 all rely on light driven charge-
separation. Furthermore, in the emerging field of molecular
electronics the electron transfer is the key step that propagates
the electronic signal between the different components. There
are hopes that molecular systems can be designed to function
not only as electronic9 or photonic wires10 but also as
components such as diodes and switches11 for more advanced
electronic devices.
Investigations of photoinduced charge separation in model
systems is particularly valuable to better understand electron
transfer in biological and artificial systems and to assist the
design of new artificial systems. Understanding of long-range,
photoinduced electron transfer has been greatly advanced in
Received: June 25, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Chart 1. Structures of the Donor−Acceptor Dyads Described in This Study
(Chart 1).14e,g,19 An oligo(phenylene-ethylene) bridge (OPE)
was used to link the donor (ZnPc) and acceptor (C60 or AuP+).
This type of bridging unit has been shown to mediate electron
transfer over long distances14f−h,20 and when connected
through an ethynylene group to the donor and acceptor units
this occurs with surprisingly high rates.13c,d,21 These two new
dyads are compared with our earlier investigation, where the
acceptor was a tin(IV) dichloride porphyrin (SnP).13a Our data
strongly suggest that the reduction of a gold porphyrin first
occurs on the macrocycle (AuIIIP•) but that the system then
relaxes, resulting in a metal based reduction (AuIIP). This
particular feature of gold porphyrin has a profound impact on
the forward and backward electron transfer rates compared to
the dyads with C60 and SnP acceptors and can be used as a tool
to engineer rectifying molecular systems in which the kinetics
of the charge recombination is much slower than the forward
electron transfer.
kV acceleration voltage with 2,5-dihydroxybenzoic acid (DHB) or
dithranol as matrix. Thin-layer chromatography (TLC) was performed
on aluminum sheets precoated with Merck 5735 Kieselgel 60F254
.
Column chromatography was carried out either with Merck 5735
Kieselgel 60F (0.040−0.063 mm mesh) or with SDS neutral alumina
(0.05−0.2 mm mesh).
Compound 2. Compound 1 (ZnPc-OPE) (38 mg, 23.9 μmol)
was dissolved in THF (2.5 mL) and Bu4NF (57.3 μL, 1 M in THF)
was added. The solution was stirred at room temperature for 2 h. The
THF was removed and CH2Cl2 was added. The organic phase was
washed with water, dried over MgSO4, and concentrated to dryness,
affording a green solid (34 mg, 100%). 1H NMR δ (300 MHz, CDCl3,
25 °C): 0.84 (m, 6H, CH3), 1.24 (m, 36H, CH2), 1.67 (s, 27H, tBu),
1.78−1.82 (m, 4H, CH2), 3.12 (s, 1H, H ethynyl), 3.96 (m, 4H, CH2),
6.90−6.95 (m, 2H, H OPE), 7.40 (m, 4H, H OPE), 7.58 (m, 4H, H
OPE), 7.60−7.69 (m, 3H, ZnPc), 7.80−8.10 (m, 9H, ZnPc); MALDI-
TOF: m/z: calcd for C94H100N8O2Zn, 1437.733 [M+H]+; found
1437.62 [M+H]+. HR-MS: (ESI): m/z: calcd for C94H100N8O2Zn,
1437.7333 [M+H]+; found 1437.7286 [M]+.
OPE Spacer 7. Compounds 6 (240 mg, 1.86 mmol), 5 (1 g, 1.43
mmol), 4 (368 mg, 1.86 mmol), PdCl2 (50 mg, 0.28 mmol), PPh3
(112 mg, 4.3.10−5 mol), Cu(OAc)2 (14 mg, 7.1.10−5 mol),
diisopropylamine (0.5 mL), and triethylamine (13 mL) were dissolved
in dry THF (13 mL). The solution was degassed and stirred at 80 °C
for 20 h. After cooling down to room temperature, the crude reaction
mixture was diluted with CH2Cl2. The organic phase was washed with
water, dried over MgSO4 and evaporated to dryness. The product was
purified by column chromatography over silica (petroleum ether/
CH2Cl2: 1/1), to afford compound 7 as a yellow solid (342 mg, 31%).
1H NMR δ (300 MHz, CDCl3, 25 °C): 0.22 (s, 9H, TMS), 0.89 (m,
EXPERIMENTAL SECTION
■
Synthesis. Compounds ZnPc-OPE 1,13a halogeno gold porphyrin
3,13d 2,5-didodecyl-1,4-diiodobenzene 5,22 1-ethynyl-4-
(trimethylsilylethynyl)benzene 4,13d 4-ethynyl benzaldehyde 623 and
iodo-zinc phthalocyanine 924 were prepared according to literature
methods. Chemicals were purchased from Aldrich, Acros and used as
received. Air sensitive reactions were carried out under argon in dry
solvents and glassware.
1H and 13C NMR spectra were recorded on a Bruker ARX 300
MHz or AMX 400 MHz Bruker spectrometer. Chemical shifts for H
NMR spectra are referenced relative to residual protium in the
1
3
1
6H, CH3), 1.27 (m, 32H, CH2), 1.55 (quint., J = 6.3 Hz, 4H, CH2),
deuterated solvent (CDCl3 δ = 7.26 ppm for H and δ = 77.16 ppm
3
3
for 13C). EI mass spectra were recorded on an EI-MS HP 5989A
spectrometer. MALDI-TOF analyses were performed on a Bruker
Ultraflex III, micrOTOF Q spectrometer in positive linear mode at 20
1.83 (quint., J = 6.8 Hz, 4H, CH2), 4.04 (t, J = 6.3 Hz, 4H, CH2),
7.01 (s, 2H, H OPE), 7.45 (m, 4H, H OPE), 7.66 (d, 3J = 7.9 Hz, 2H,
H OPE), 7.86 (d, 3J = 7.9 Hz, 2H, H OPE), 10.02 (s, 1H, CHO). HR-
B
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
MS (ESI): m/z: calcd for C52H70O3Si, 770.5099 [M]+; found 770.5170
[M]+.
with argon before the measurements. All potentials are quoted relative
to SCE. In all the experiments the scan rate was 0.1 V/s.
Spectroscopy. Absorption and Emission Spectroscopy. UV−vis
absorption spectra were recorded on a Cary 5000 absorption
spectrometer, and steady-state emission spectra were measured on a
Horiba SPEX Fluorolog 3 fluorimeter. During emission measurements
the optical density was kept below 0.1 at the excitation wavelength.
The benzonitrile (PhCN) (99%, Fluka) was distilled over phosphorus
pentaoxide and filtered through K2CO3(s). Using nondistilled
benzonitrile resulted in aggregation, which could be observed as a
strong perturbation of the ZnPc Q-band.
OPE Spacer 8. The spacer 7 (340 mg, 0.44 mmol) and K2CO3
(607 mg, 4.4 mmol) were dissolved in a mixture of CH2Cl2 (30 mL)
and MeOH (40 mL). The solution was stirred at room temperature
for 45 min. CH2Cl2 was added, and the organic phase was washed with
water, dried over MgSO4, and concentrated to dryness, affording 8 as a
1
yellow solid (262 mg, 85%). H NMR δ (300 MHz, CDCl3, 25 °C):
0.85 (m, 6H, CH3), 1.24 (m, 32H, CH2), 1.56 (quint., 3J = 6.1 Hz, 4H,
3
CH2), 1.87 (quint., J = 6.3 Hz, 4H, CH2), 3.16 (s, 1H, H ethynyl),
4.05 (m, 4H, CH2), 7.02 (s, 2H, H OPE), 7.47 (s, 4H, H OPE), 7.66
(d, 3J = 8.1 Hz, 2H, H OPE), 7.86 (d, 3J = 8.1 Hz, 2H, H OPE), 10.02
(s, 1H, CHO). HR-MS (ESI): m/z: calcd for C49H62O3, 698.4693
[M]+; found 698.4682 [M]+.
Absorption spectra of the reduced and oxidized states of the dyads
were recorded with a HP diode array spectrometer. Tri-p-
+
−
́
bromophenyl aminium hexafluorophosphate (TpBPA PF6 , E° ≈ 1.1
vs SCE)25 was used as oxidant to form the ZnPc+•-OPE-acceptor. The
Compound 10. In a Schlenk flask were dissolved zinc tri-tert-butyl-
iodophthalocyanine 9 (160 mg, 0.18 mmol) and the connector 8 (128
mg, 0.18 mmol) in dry piperidine (10 mL), and the mixture was
degassed. Triphenylarsine (110 mg, 0.36 mmol) and Pd2(dba)3.CHCl3
(55 mg, 54 μmol) were added, and the solution was degassed again.
The reaction mixture was stirred at 40 °C overnight under argon. The
solvent was removed and CH2Cl2 was added. The organic phase was
washed with water (twice), dried over MgSO4, and concentrated to
dryness. The product was purified by column chromatography over
silica gel (petroleum ether/dioxane: 8/2), to afford a green-blue solid
aminium salt was prepared according to the procedure of Rhile et al.26
25
́
As a reductant we used cobaltocene (CoCp2, E° ≈ −1.0 V vs SCE),
allowing for the spectra of the reduced dyads. The cobaltocene was
purchased from Sigma-Aldrich (Sweden) as a 10% solution in hexane
and used as received. Both the oxidation and the reduction of the
dyads were done in a glovebox and deaerated PhCN was used as
solvent.
Time-resolved fluorescence was measured with a time-correlated
single photon counting setup (TCSPC), using 400 or 405 nm for
excitation. The system has previously been described by Habenicht et
al.27 In summary: The 400 nm excitation light was obtained from the
frequency doubled output of a Ti-sapphire laser (Coherent RegA 900;
200 kHz repetition rate, λ = 800 nm, fwhm 180 fs) (IRF ∼80 ps). For
405 nm excitation a 405 nm diode laser at 5 MHz repetition rate
(fwhm 100 ps) was used (IRF ∼120 ps). The emission was detected
using a Hamamatsu MCP. Magic angle setup was ensured with two
polarizers, and scattered excitation light and filter emission was
blocked using a combination of a 600 nm cutoff filter and a 751 nm
interference filter. The instrument response function (IRF) was
measured using a scattering solution of Ludox silica sol in water. That
the 400 or 405 nm excitation populates a higher singlet excited state
does not affect the observed data since the internal conversion to the
lowest 1ZnPc is expected to be much faster than the TCSPC
resolution (∼40 ps). This is in agreement with the lack of a detectable
1
(134 mg, 52%). H NMR δ (300 MHz, CDCl3, 25 °C): 0.89 (m, 6H,
CH3), 1.29 (m, 36H, CH2), 1.62 (s, 27H, tBu), 1.84−1.94 (m, 4H,
CH2), 4.06 (m, 4H, CH2), 6.98−7.02 (m, 2H, H OPE), 7.47 (m, 4H,
H OPE), 7.64 (m, 4H, H OPE), 7.75−8.05 (m, 3H, H ZnPc), 8.10−
8.80 (m, 9H, H ZnPc), 10.1 (s, 1H, CHO). HR-MS (ESI): m/z: calcd
for C93H100N8O3Zn 1440.7210, found 1440.7214 [M]+.
Dyad ZnPc-OPE-C60. Compound 10 (65 mg, 0.045 mmol),
fullerene (65 mg, 0.09 mmol), and N-methylglycine (40 mg, 0.44
mmol) were dissolved in dry toluene (65 mL), and the mixture was
heated at 140 °C overnight. The solvent was removed and CH2Cl2 was
added. The organic phase was washed twice with water, dried over
MgSO4, and concentrated to dryness. The product was purified by
column chromatography over silica gel (toluene/dioxane: 9/1), to
afford a green-blue solid (30 mg, 30%). 1H NMR δ (300 MHz, THF-
d8, 25 °C): 0.89 (m, 6H, CH3), 1.31 (m, 59H, 27H tBu + 32H
dodecane), 1.85 (m, 4H, CH2), 1.95 (m, 4H, CH2), 2.81 (s, 3H, CH3
pyrrolidinyl), 4.06 (t, 3J = 6 Hz, 4H, CH2), 4.26 (d, 3J = 9.6 Hz, 1H, H
pyrrolidinyl), 5.01 (m, 2H, H pyrrolidinyl), 7.08 (s, 1H, H OPE), 7.11
(s, 1H, H OPE), 7.61 (m, 4H, H OPE), 7.66 (m, 4H, H OPE), 8.30
(m, 4H, H ZnPc), 9.30 (m, 4H, H ZnPc), 9.49 (m, 4H, H ZnPc).
MALDI TOF: m/z: calcd. for C155H105N9O2Zn 2187.7677, found
2187.7713 [M]+.
1
rise time for the observed ZnPc emission at 751 nm.
Femtosecond Transient Absorption (fs-TA). For a detailed
description see Petersson et al.28 Briefly, the output from a Coherent
Legend (1 kHz, λ = 800 nm, fwhm 80 fs) was split into a pump and a
probe part. Using a TOPAS (Light Conversion) the pump beam was
transformed into the desired wavelengths. A 500 Hz chopper was used
to block every second pump pulse, and the pump energy was
attenuated with neutral density filters so that the pulse energy at the
sample was kept between 400 and 600 nJ/pulse (energy variation of
the pump energy within a measurement was less than 10%). The
timing between the pump and probe pulse was controlled by two
different delay lines. For measurements that required subpicosecond
time resolution a 30 cm long (∼time window ∼2 ns) delay line from
Physik Intrumente was used, while a 1.5 m (∼10 ns) delay line from
H2W Technologies was used when a longer time window was
necessary. For probing in the UV−visible (350−750 nm), the probe
pulse was transformed into a white-light continuum by focusing the
800 nm light into a CaF2 plate. Residual 800 nm and NIR light was
removed by a KG3 filter. For probing in the NIR region (860−1150
nm) a sapphire plate was used to create a broad white light continuum.
Visible light was removed using an 800 nm cutoff filter. Detection
between 750 and 860 nm is prevented because of excess light from the
laser fundamental at 800 nm. The probe light was detected with a
silicon photodiode array spectrometer (512 points in a 328 nm wide
spectral window). The silicon diodes have a good response at
wavelengths below 1000 nm but lose sensitivity at wavelengths longer
than 1000 nm, which explains the small signal-to-noise ratio in the
NIR measurements. The polarization of the pump was set to ensure
magic angle compared to the probe, and the two beams were then
focused and overlapped on a ∼0.1 mm2 area inside a 1 mm quartz
cuvette.
Dyad ZnPc-OPE-AuP+. The compound 2 (23 mg, 19.7 μmol), the
gold porphyrin 3 (34 mg, 23.7 μmol), Pd(dppf)Cl2 (3 mg, 4 μmol),
and CuI (1 mg, 4 μmol) were dissolved in dry Et3N (1.1 mL) and dry
DMF (4.8 mL). The solution was degassed and stirred under argon at
60 °C for 15 h. CH2Cl2 was added, and the organic phase was washed
with water (3 times), dried over MgSO4, and concentrated. The
product was purified by column chromatography over silica (CH2Cl2
100% then MeOH/CH2Cl2: 2/98), to afford a green solid (22 mg,
44%). 1H NMR δ (300 MHz, CDCl3, 25 °C): 0.84 (m, 6H, CH3), 1.24
(m, 36H, CH2), 1.59 (s, 36H, tBu), 1.67 (s, 27H, tBu), 1.78−1.85 (m,
4H, CH2), 4.00−4.30 (m, 4H, CH2), 7.00−7.20 (m, 2H, H OPE),
7.40−8.25 (m, 22H, 3H ZnPc + 4H ortho +2H para +5H phenyl +8H
OPE), 8.30−8.70 (m, 9H, H ZnPc), 9.00−9.45 (m, 6H, AuP Hβ),
10.15−10.17 (m, 2H, AuP Hβ). HR-MS (ESI): m/z: calcd for
C148H154N12O2ZnAu, 2392.1275 [M]+; found 2392.1278 [M]+.
Electrochemistry. Electrochemical measurements were performed
with a potentiostat-galvanostat AutoLab controlled by resident GPES
software (General Purpose Electrochemical System 4.9) using a
conventional single-compartment three-electrode cell. The working
electrode was a Pt electrode, the auxiliary was a Pt wire of 10 mm long,
and the reference electrode was the saturated potassium chloride
calomel electrode (SCE). The supported electrolyte was 0.1 N
Bu4NPF6 in dichloromethane, DCM, and the solutions were purged
C
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 1. Synthesis of the Dyad ZnPc-OPE-AuP+
a
a) Bu4NF, THF, RT, 100%; b) Pd(dppf)Cl2, CuI, Et3N, DMF, 60 °C, 43%.
Data analysis of the TCSCP data was done in MATLAB 7, using a
model based on a sum of one or two exponential decays, reconvoluted
with the experimentally recorded response function. Data from the fs-
TA measurements were analyzed with Igor Pro 5.29 Kinetic data were
globally fitted to a sum of exponentials reconvoluted with a Gaussian
function to account for the instrument response function. The
wavelengths chosen in the global fit were evenly spaced over the
observed spectra (every 10th nm for ZnPc-OPE-AuP+ and every 30th
nm for ZnPc-OPE-C60). The width of the response function and
lifetimes were linked between different wavelengths, while time zero
and amplitudes of the pre-exponential factors were allowed to vary
freely. Each analysis started out from a model of two consecutive
reactions (photoinduced electron transfer (PET) followed by back
electron transfer (BET)). A third exponential (τ > 100 ns) accounts
simply removed the ZnPc side but conserved the full oligo(phenylene-
ethynylene) bridge.
Computational Protocol. To perform DFT and TD-DFT
simulations of the structures and properties of the new dyads, the
latest commercial version of the Gaussian program has been selected.30
Default algorithms, parameters, and thresholds have been used, except
when otherwise indicated. The computational strategy recently applied
to unravel the excited-state properties of inorganic and organic dyes31
has been adopted. This procedure consists of two sequential steps:
1. The ground-state geometrical parameters have been determined
at the PBE032/6-31G(d) level through a force-minimization process.
Note that LanL2DZ pseudo-potential and basis set have been used for
metallic atoms. The bulk solvent (benzonitrile) effects have been
included by means of the Polarizable Continuum Model (PCM) in its
Integral Equation Formalism (IEF).33
3
for the presence of long-lived ZnPc, formed in parallel to the ET
2. The first fifteen to twenty lowest-lying excited-states have been
determined within the vertical TD-DFT approximation using the
CAM-B3LYP34/6-31G(d) [LanL2DZ] level of approximation. The
choice of CAM-B3LYP, a range-separated hybrid, is dictated by the
charge-transfer/long-range nature of the phenomena under inves-
tigation. For such states, it is well-recognized that a range-separated
hybrid such as CAM-B3LYP offers a physically sound description.35 As
in the first step, the solvent effects have been taken into account at the
PCM level of approximation, using a linear-response nonequilibrium
model,36 that is adequate when the reorganization of the solvent is
limited to its electronic component (fast processes).
transition (ISC competing with PET or BET partially resulting in
3ZnPc). However, to achieve a satisfactory global fit a fourth
exponential had to be added to the model to account for the presence
of unquenched 1ZnPc. The quality of a fit was judged from the
residual. Igor Pro reported a standard deviation of the fitted lifetimes
to be less than 5%. However by forcibly changing a lifetime, it was
noticed that satisfactory fits could be obtained also after changing a
lifetime as much as 25% so the overall uncertainty is probably closer to
25%. Examples of fits and residuals can be found in the Supporting
Information (SI Figure S4 and SI Figure S5). Following a satisfactory
global fit, the obtained response function and lifetimes were used to fit
all of the 512 traces, and a decay associated difference spectrum
(DADS) was created by plotting the pre-exponential factors against
the wavelength for each of the obtained lifetimes. The obtained
variation of time zero with wavelength was used to chirp correct the
observed difference spectra. To clarify spectral features a smoothing
function (second order Savitsky-Golay, 15 points) was applied to
difference spectra. Note that no smoothing function was applied on
data before kinetic analysis.
DFT Simulations. Models. The calculations have been performed
on the compounds shown in Chart 1, which have only been slightly
simplified by removing the tBu groups of the Pc and replacing the
OC12H25 lateral chains by methoxy groups. As we investigate
electronic properties and take into account all π-electrons these
simplifications should have only a trifling impact on the computed
transition energies. For the AuP+ system, the impact of the counterion
has been neglected as well. The optimized Cartesian coordinates for
the three dyads can be found in the SI. In addition, we have also used
specific systems of the acceptor moieties to simulate what follows ET
(see discussion around Figure 7). In these acceptor structures, we have
The contour threshold used to represent the molecular orbitals was
set to 0.020 au, and a broadening Gaussian presenting a fwhm of 0.35
eV was used to simulate UV/vis spectra from the vertical TD-DFT
estimates.
RESULTS AND DISCUSSION
■
Synthesis of the Dyads. The synthetic route of the dyad
ZnPc-OPE-AuP+ is based on an extension of an earlier
procedure previously developed for ZnPc-OPE-SnP.13a Start-
ing with the known intermediate ZnPc-OPE-TIPS 1,13a the
trisisopropylsilyl group (TIPS) was cleaved with tetrabutylam-
monium fluoride. The gold porphyrin 313d was appended to the
OPE spacer of the above compound 1 according to a
Sonogashira cross-coupling reaction in 43% yield using the
catalytic system previously described to prepare bisporphyrin
dyads containing a gold porphyrin (Scheme 1).13a,d The
synthesis of the dyad ZnPc-OPE-C60 required the intermediate
spacer 8, whose synthesis is depicted in Scheme 2. Compound
D
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 2. Synthesis of the Dyad ZnPc-OPE-C60
a
a) PdCl2, PPh3, Cu(OAc)2, THF, Et3N, 80°C, 31%; b) K2CO3, CH2Cl2, MeOH, RT, 85%; c) Pd2dba3-CHCl3, AsPh3, piperidine, 40 °C, 52%; d)
sarcosine, C60, toluene, reflux, 30%.
Table 1. Electronic Absorption and Emission Properties in Benzonitrile (PhCN) at Room Temperature and Electrochemical
Properties in Dichloromethane (DCM) (1 mM TBAPF6)
emission
E1/2 (V vs SCE)
OPE+/0
a
b
c
d
absorption λmax (nm)
λmax (nm)
φrel
τem (ns)
ZnPc+/0
0.46
Accn/n‑1
e
f
ZnPc-OPE
364, 617, 641, 679, 696
322, 429, 545, 592, 676
388, 549, 597, 681, 698
353, 617, 641, 679, 696
355, 438, 579, 618, 642, 679, 696
705
1.00
h
1.8
1.30
1.34
1.36
1.28
1.30
g
h
h
OPE-AuP+
−0.56
−0.71
−0.65
−0.82
i
j
i
ZnPc-OPE-AuP+
706
705
705
0.22
<0.04 (1.8)
0.61 (1.8)
0.45
0.45
0.46
ZnPc-OPE-C60
0.26
e
i
j
i
ZnPc-OPE-SnP
0.05
<0.04 (1.8)
a
b
c
d
Excitation at 680 nm. Relative to ZnPc-OPE emission. Excitation at 400 nm. Accn/n‑1 is the first reduction potential of the electron acceptor
moiety. Data from Fortage et al.13a Corrected from Fortage et al.13a From Fortage et al.13d Not observed. Most of the observed emission comes
from a fraction of ZnPc that is not quenched at all and has the same emission lifetime as the ZnPc-OPE reference, see main text. Lifetime faster
e
f
g
h
i
j
1
than time resolution of TCSPC system.
7 was synthesized in 31% yield by the mixed Sonogashira cross-
coupling reaction between the 2,5-didodecyl-1,4-diiodobenzene
5,22 the 1-ethynyl-4-(trimethylsilylethynyl)benzene 4,13d and
the 4-ethynyl benzaldehyde 6.23 The desired compound 7 was
easily separated by column chromatography from the sym-
metrical side-products owing to the large differences of polarity
of the three compounds. The TMS group was removed with
potassium carbonate, and the spacer 8 was then reacted with
iodo-zinc phthalocyanine 924 using classical Sonogashira
conditions with this type of reagents (Scheme 2).13a,d Finally,
the fullerene moiety was grafted on the spacer using Prato
reaction following a 1,3-dipolar cyclo-addition between the
fullerene and the azomethane ylide formed by the reaction of
the aldehyde function of 10 and N-methylglycine.23,37
Electrochemistry. The reduction potentials of the dyads
were measured with cyclic voltammetry, and the processes were
assigned according to the potentials recorded in the reference
compounds (Table 1). The first oxidation process occurs on
zinc phthalocyanine at around +0.45 V vs SCE, while the
oxidation of the OPE spacer is shifted far away into the anodic
region at around +1.3 V vs SCE. The three electron acceptors
have reduction potentials in a quite narrow potential window,
with pyrrolidino-fullerene as the most easily reducible system
(−0.65 V vs SCE), followed by the gold porphyrin (−0.71 V vs
SCE) and finally by the tin porphyrin (−0.83 V vs SCE). The
reduction of the spacer was not accessible in the electroactive
window of our experimental conditions (<−1.6 V vs SCE).
Note that the AuP+/0 potential in ZnPc-OPE-AuP+ is shifted
0.15 V negative compared to the reference, OPE-AuP+. This is
a sign of strong electronic interaction in the ground state of this
dyad (vide infra).
Electronic Absorption Spectra. The electronic absorp-
tion spectra of the new dyads were recorded in benzonitrile
(PhCN), where no aggregation occurs, and are shown together
with the relevant reference compounds in Figure 1. The
absorption spectrum of ZnPc-OPE-SnP can be found in the
Supporting Information (SI Figure S1).13a The absorption
spectrum of ZnPc-OPE-C60 mainly exhibits features of the
ZnPc-moiety (Figure 1a). However, the presence of the C60
unit can be seen as an additional feature in the UV region,
E
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Electronic absorption spectra in benzonitrile at 298 K of a)
ZnPc-OPE-C60 (solid line) and its reference compounds ZnPc-OPE
(dotted line) and C60 (dashed line); b) ZnPc-OPE-AuP+ (solid line)
and its reference compounds ZnPc-OPE (dotted line) and OPE-AuP+
(dashed line). Spectra are normalized to unity intensity of the Q
bands.
Figure 2. a) Difference spectra of oxidized (blue straight line) and
reduced (red dotted line) form of ZnPc-OPE-AuP+. b) Simulated
difference spectra for the charge shifted state (CSS) of ZnPc-OPE-
AuP+, created as a sum of the difference spectra of the oxidized and
reduced form of ZnPc-OPE-AuP+. Overlaid is the decay associated
difference spectra (DADS) of the 1.0 ns lifetime observed upon 555
nm excitation of ZnPc-OPE-AuP+. The similarity between the two
spectra confirms that the 1.0 ns lifetime corresponds to back electron
transfer.
causing an apparent blue-shift of the Soret-band compared to
that in ZnPc-OPE. The splitting of the ZnPc Q-band is most
likely caused by the ethynylene linkage, which induces an
asymmetry of the otherwise degenerate Qx- and Qy-transitions
of the ZnPc.13a,38
It is surprising that the interaction between the donor and
acceptor is so strong over such a long distance (∼23 Å, edge-to-
edge). Ethynylene as linking group has indeed been shown to
mediate electronic coupling but usually over much shorter
distances.13c,39 Furthermore, in a system with the same bridge
and acceptor, but with a zinc porphyrin instead of zinc
phthalocyanine as electron donor, we did not observe an
interaction of this magnitude.13d
Steady-State and Time-Resolved Emission. The
emission spectra of ZnPc-OPE-C60 and ZnPc-OPE-AuP+
(λexc = 680 nm) are very similar to that of the ZnPc-OPE
reference, with a λmax ≈ 705 nm, but with significantly lower
quantum yields (SI Figure S2), indicating that electron transfer
or some other process is quenching the excited state (vide
infra). The emission quantum yields relative to ZnPc-OPE
(φrel) were determined to 0.26 and 0.22 for ZnPc-OPE-C60
and ZnPc-OPE-AuP+ respectively (Table 1). OPE-AuP+ has
no detectable emission at room temperature, and there is no
indication of C60 emission in the ZnPc-OPE-C60 dyad.
The emission lifetime of the lowest singlet excited state
(1ZnPc), was determined with time-correlated single photon
counting (TCSPC), exciting at 400 nm or at 405 nm. For
ZnPc-OPE-C60 two exponential components, τ1 = 0.61 ns
(90%) and τ2 = 1.8 ns (10%), were necessary to achieve a
satisfactory fit of the data. The rate constant for the quenching
reaction, kq, was calculated using eq 1 and the lifetime of the
major component, yielding kq = 1.1 × 109 s−1 for ZnPc-OPE-
ZnPc-OPE-AuP+ has a more complicated absorption
spectrum (Figure 1b). In the Q-band region (500−700 nm)
the spectrum of ZnPc-OPE-AuP+ can be reasonably well
described by a sum of the spectra of references ZnPc-OPE and
OPE-AuP+, with only a small red-shift and broadening of the
ZnPc Q-band (λmax 698 nm compared to 696 nm in ZnPc-
OPE). In contrast, the Soret-band region differs significantly
from a sum of the reference compounds. Instead of two distinct
peaks from the ZnPc and AuP+ Soret-bands there is a strong
and very broad peak centered around 400 nm. This shows a
strong interaction between the ZnPc and AuP+ moieties,
possibly due to excitonic coupling between the Soret-bands or a
direct donor−acceptor charge transfer transition (vide infra).
Chemical oxidation of ZnPc-OPE-AuP+, using tri-p-
bromophenyl aminium hexafluorophosphate as oxidant, results
in the expected disappearance of ground state features and the
concomitant growth of ZnPc+• features at 525 nm, 745 nm, and
845 nm around. There is also a sharp absorption peak growing
in around 435 nm (Figure 2a), that is not seen in oxidized
ZnPc-OPE,13a but is very similar to the Soret band of OPE-
AuP+. It thus appears that the perturbation of the gold
porphyrin is much weaker when the zinc phthalocyanine is
oxidized so that the “normal” AuP+-Soret band is restored.
Similarly, one electron reduction of ZnPc-OPE-AuP+, using
cobaltocene as reductant, results in the appearance of a ground
state bleach feature at 420 nm, 550, and 594 nm, and
concurrently a new band grows in at 455 nm and the ZnPc Q-
bands becomes stronger (Figure 2a). The features at 420 and
455 nm can be explained by a slight red shift of the Soret band,
often seen for reduction of AuIIIP+ to AuIIP.18d,e The reduction
also causes the interaction between gold porphyrin and zinc
phthalocyanine to be weaker, explaining the disappearance of
the broad Soret band and the slight increase in the ZnPc Q-
band.
C
60 in PhCN. The 1.8 ns component is attributed to a fraction
1
of unquenched ZnPc-OPE.
−1
kq = τobs − τ01
(1)
For ZnPc-OPE-AuP+ a single 1.8 ns lifetime was observed
13a
1
which is equal to the lifetime of the ZnPc-OPE reference.
However, the much lower emission yield in ZnPc-OPE-AuP+
indicates that the majority of ZnPc is quenched by a rate faster
F
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. fs-TA measurements on ZnPc-OPE-C60, excitation at 680 nm. a) Chirp corrected difference spectra at different time points. Inset: Chirp
corrected difference spectra in the NIR, showing a band indicative of C60•‑. The color legend for different time points is valid for both the visible and
the NIR spectra. b) Kinetic traces at different wavelengths. Markers are measured data, and solid lines are the results of the global fit to the data (see
text).40 The trace at 700 nm has been scaled down by a factor of 3 to better fit the graph. Inset shows the rise and decay feature of C60•‑, around 1025
nm (average of 1010−1040 nm) on a logarithmic time scale.
Figure 4. fs-TA measurements on ZnPc-OPE-AuP+, excitation at either 680 nm (a and b) or at 555 nm (c and d). Spectra in a and c correspond to
transient spectra at a different times after excitation. Kinetic traces at selected wavelengths are shown in b and d. Markers correspond to experimental
data, and the solid line is the kinetic fit (see text). The trace at 700 nm has been scaled down by a factor of 3 to better fit the graph.
than the time resolution of the instrument (∼40 ps). This
means that most of the observed steady-state emission from the
ZnPc-OPE-AuP+ sample comes from a smaller fraction of
1ZnPc (possibly an impurity) which is not quenched at all. This
is confirmed by the femtosecond experiments below.
Femtosecond Transient Absorption. Femtosecond
pump−probe experiments were performed to follow the
photoinduced electron transfer events in ZnPc-OPE-AuP+
and ZnPc-OPE-C60.
ZnPc-OPE-C60. Figure 3 shows transient absorption data for
ZnPc-OPE-C60 at a few time points after a 150 fs excitation
pulse at 680 nm (Q-band of ZnPc). The spectrum at early
G
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
times (t < 2 ps) is in very good agreement with that of the
lowest singlet exited state (1ZnPc) of ZnPc-OPE (SI Figure
S3).13a 1ZnPc is dominated by a broad absorption band
extending over most of the visible region, and superimposed on
this band is the ground-state bleach, which creates distinct
valleys at 615, 680, and 696 nm. At later delay times (t > 1000
ps) there is a decrease in absorption in the region around 600−
670 nm, resulting in a net bleach. This is indicative of ZnPc
triplet state (3ZnPc) formation, but the overall rate and yield is
not the same as in ZnPc-OPE. In a global analysis of ten
wavelengths, evenly spaced over the observed spectra (Δλ ≈ 30
nm), four lifetimes were needed to fit the observed data; τ = 20
ps, 0.66 ns, 1.8 ns, and ∞ (see SI Figure S4 for example of fits
and residuals).40
1.1 ns, and ∞ (held fixed) (see SI Figure S5 for example of fits
and residuals). The majority of the ZnPc Q-band bleach
recovers with a lifetime of 1.1 ns, but this does not match the
emission lifetime (1.8 ns), nor is the magnitude of the ZnPc
ground state bleach recovery occurring through the 1.1 ns time
1
component consistent with ZnPc decay, because that would
have resulted in a larger fluorescence yield than observed. Thus
this kinetic process is not the decay of 1ZnPc but rather
attributed to back electron transfer from the charge shifted
state, ZnPc•+-OPE-AuIIP, to the ground state. This assignment
is further supported by the good match between a simulated
charge separated state and the decay associated difference
spectrum (DADS) of the 1.1 ns component as shown in Figure
2b.44 From inspection of the decay associated difference spectra
of the three remaining time components (SI Figure S7) it was
concluded that the 1.2 ps lifetime corresponds to photoinduced
A complementary measurement, probing in the near-infrared
(NIR) showed the growth and decay of a broad absorption
band between 1000 and 1050 nm.41 This matches the signature
1
electron transfer from ZnPc to AuP+. Four lines of evidence
19
•‑
of C60
,
which would indicate that the quenching process is
support this assignment. 1) The corresponding DADS show
changes in the region between 400 and 500 nm, which would
be expected from a process that affects AuP+.43 2) The 1.2 ps
time component barely affects the ground state bleach from the
ZnPc Q-band. This is consistent with photoinduced electron
transfer since this is a process that does not change the
concentration of ground state ZnPc, while it at the same time
lowers the concentration of ground state AuP+. 3) Further-
more, the spectral changes in the region between 400 and 500
nm are nicely mirrored in the decay associated difference
spectrum of the 1.1 ns component, which makes it very likely
that the species created with τ = 1.2 ps (photoinduced electron
transfer) disappears with τ = 1.1 ns (back electron transfer), see
reaction scheme in Figure 5b. 4) Lastly, observed emission
lifetimes and steady-state emission yield suggest that the
photoinduced electron transfer forming the charge-separated
state (ZnPc•+-OPE-C60•‑). The rise and decay features in the
NIR region were globally fitted at 16 wavelengths to lifetimes of
∼15 ps and ∼0.70 ns and ∞, which are similar to the lifetimes
observed in the visible region.
The 0.66 ns (UV−vis) and 0.7 ns (NIR) lifetimes matches
the fluorescent lifetime (0.61 ns) and is thus assigned to
photoinduced charge separation (kPET = 1.1 × 109 s−1). Since
the shortest lifetime (∼20 ps) also is linked to the C60•‑ signal,
it is reasonable to assign this to the charge recombination
reaction (kBET ≈ 5 × 1010 s−1). Thus, assuming a reaction
scheme with kPET = 1.1 × 109 s−1 and kBET ≈ 5 × 1010 s−1 (see
Figure 5a), one would expect a maximum concentration of the
charge-separated state, ZnPc•+-OPE-C60•‑, at ∼80 ps corre-
sponding to only ∼1% of the original excited state population
(see SI for details on calculations). The small ratio kPET/kBET
1
majority of the ZnPc in the ZnPc-OPE-AuP+ sample reacts
with a lifetime that is faster than 40 ps.
•‑
explains why the C60 signal is so low in the NIR region and
For the remaining two time components the 0.11 ps lifetime
had too small and unspecific features to be attributed to a
specific process, while the “infinite” time component was
attributed to a small fraction of unquenched 1ZnPc states
why there are no clear ZnPc•+ signals in the visible region.42
The latter causes the 0.66 ns lifetime (charge separation) to
appear to mainly result in ground state recovery.
45
3
Other quenching mechanisms, such as energy transfer from
the 1ZnPc state to the fullerene or enhanced intersystem
crossing to the 3ZnPc state, are energetically possible, but based
on the absence of transient signals from the long-lived fullerene
triplet state43 and low yield of 3ZnPc these cannot be
significant.
evolving into long-lived ZnPc states.
Like in our previous study on ZnPc-OPE-SnP13a we were
also interested in the reaction induced by selective excitation of
the AuP+ Q(1,0)-band at 555 nm. In the OPE-AuP+ reference a
strong absorption band appears in the region above 600 nm (SI
Figure S6). For symmetric gold porphyrins, the lowest singlet
excited state (1AuP+) undergoes very rapid intersystem crossing
to the lowest triplet state (3AuP+), and this triplet state has a
ZnPc-OPE-AuP+. Excitation of ZnPc-OPE-AuP+ at 680 nm
results in an initial spectrum (Figure 4a, t = 0.2 ps) which
displays a broad transient absorption band, peaking at 520 nm,
similar to the band of oxidized ZnPc-OPE-AuP+, Figure 2a.
Overlaid with this band there are ground state bleach features
from the ZnPc Soret band (λ < 450 nm) and the Q-band
(650−720 nm). Unexpectedly, there are also additional ground
state bleach features from the AuP+ moiety at 400−500 nm,
555 nm, and 620 nm. Combined with the 520 peak, the AuP+
bleach features indicate strong coupling between ZnPc and AuP
also in the initial excited state and/or possibly a partial charge
transfer absorption.
strong absorption above 600 nm.46 In contrast, no AuP+
1
intermediate could be distinguished in OPE-AuP+, indicating
that the intersystem crossing is faster than the time resolution
of this experiment (∼100 fs) or that the spectral difference
1
3
between AuP+ and AuP+ state of OPE-AuP+ is very small.
Two time components are necessary for a good fit of the
transient absorption data; a 2.5 ns component corresponding to
3
the lifetime of the AuP state and a 10 ps component, with
small spectral changes. Comparing to other gold porphyr-
ins,46,47 it seems unlikely that the intersystem crossing of OPE-
AuP+ is as slow as 10 ps, so it is more probable that this
component corresponds to solvent and/or vibrational
relaxation.48
The initial excited state evolves on a few ps time scale, and a
new absorption feature grows in around 420 nm and the bleach
at 460 nm becomes more pronounced. Subsequently these new
features disappear on the same time scale as the bleach features
from the ZnPc and AuP+ Q-bands. A detailed analysis, using a
global fit of 30 wavelengths evenly spaced over the observed
spectra (Δλ ≈ 10 nm), resulted in four lifetimes: 1.2 ps, 0.11 ns,
Excitation of ZnPc-OPE-AuP+ at 555 nm results in the same
initial absorption features above 600 nm as in the OPE-AuP+
reference (Figure 4c). However, within the duration of the laser
pulse these features start to transform into the same spectra as
H
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. a) State diagram and rates of the reaction pathways in ZnPc-OPE-C60. b) State diagram showing the reaction pathways in ZnPc-OPE-
AuP+.
Figure 6. Representation of the calculated molecular orbitals for three donor−acceptor dyads. From bottom to top: HOMO-2, HOMO-1, HOMO,
LUMO, LUMO+1, LUMO+2, and LUMO+3. See the SI for a representation of all relevant orbitals.
those observed immediately after excitation at 680 nm. The
rapid (<1 ps) change is most easily seen as a fast decay of the
AuP+ excited state absorption at λ > 600 nm and a rapid
formation of the ZnPc bleach around 680−710 nm (Figure 4d).
A global fit analysis resulted in five time components: τ = 0.2
ps, 1.0 ps, 0.14 and 1.0 ns, and ∼∞ (held fixed). The 0.2 ps
time component is assigned to a shift of excited state energy
from a state that is mainly localized on the AuP+-moiety to the
same excited state as that produced upon 680 nm excitation.
This is supported by the fact that the four longer time
components and their corresponding decay associated differ-
ence spectra are the same as for ZnPc excitation at 680 nm (SI
Figure S7). Summarily, excitation of ZnPc (at 680 nm) induces
a very fast electron transfer (1 ps) followed by a slower charge
recombination (1 ns) to the ground state, while excitation of
AuP (at 555 nm) is accompanied by an extremely fast energy
transfer to ZnPc (0.2 ps) and subsequently to charge separation
and recombination to the ground state as upon ZnPc excitation
(Figure 5b).
TD-DFT Calculations. To further investigate the strong
interaction between the donor and acceptor moieties in ZnPc-
OPE-AuP+ we turned to computational chemistry. Calculations
of the absorption spectra, performed with time-dependent
density functional theory (TD-DFT), managed to qualitatively
I
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reproduce the absorption spectra of the investigated dyads. In
the red part of the ZnPc-OPE-AuP+ spectrum, two intense
vertical transitions were computed at 654 nm (f = 1.1) and 631
nm (f = 0.7). Both transitions are dominated by electronic
excitation from HOMO (mainly ZnPc) to LUMO+2 and
LUMO+3 (also ZnPc centered), see Figure 6. The two less
intense transitions at 574 nm (f = 0.2) and 563 nm (f = 0.5)
are related to the AuP+ moiety. The latter transition is mainly
from HOMO-2 to LUMO but also contains a small
contribution related to a partial charge transfer from the bridge
(HOMO-1) to the AuP+ (LUMO). The situation is similar for
ZnPc-OPE-SnP, but the SnP Q-band transition, computed at
593 nm (f = 0.4), has some contribution of charge transfer
from the OPE bridge (HOMO-2) to the SnP (LUMO) (see SI
for more detailed information on orbital contributions to the
calculated transitions).
There are multiple bands in the Soret band region. In ZnPc-
OPE-SnP the strongest ones, 432 nm (f = 4.0) and 398 nm (f
= 1.4), are dominated by SnP localized π−π* transitions, but
there is also a significant contribution of charge transfer from
OPE to SnP. The same pattern can be seen in ZnPc-OPE-
AuP+, where the strongest transitions at 437 nm (f = 3.1) and
401 nm (f = 1.1) are mainly localized on the AuP+ moiety.
However, in this case there are not only contributions from
OPE-to-acceptor charge transfer but also a smaller, but not
insignificant, contribution from direct charge transfer from
ZnPc to AuP+ (HOMO to LUMO). These charge transfer
interactions could be the reason for the strong broadening of
the AuP+ Soret band that is observed in the experimental
spectra. Finally, as expected for ZnPc-OPE-C60, the calcu-
lations indicate ZnPc based transitions in the visible range.
Nevertheless, it is noticeable that the LUMO remains well
centered on the C60 accepting side. The break in conjugation
between the fullerene and bridge introduced by the pyrrolidino
functionality can also be seen in the molecular orbitals. In both
ZnPc-OPE-AuP+ and ZnPc-OPE-SnP, there are high-lying
occupied orbitals that cover both the OPE bridge and the
accepting porphyrin (HOMO-2 for the former, HOMO-1 for
the latter), whereas no such mixed orbitals can be found for
ZnPc-OPE-C60.
Table 2. Summary of Rate Constants and Driving Forces for
the Three Dyads
a
a
ΔG°PET
ΔG°BET
kPET (s−1
)
(eV)
kBET (s−1
)
(eV)
ZnPc-OPE-C60
ZnPc-OPE-SnP
1.1 × 109
7.6 × 1010
1.0 × 1012
−0.89
−0.68
−0.61
∼5 × 1010
1.2 × 1010
1.0 × 109
−0.88
−1.09
−1.16
b
ZnPc-OPE-AuP+
a
Driving forces for photoinduced electron transfer (ΔG°PET) and the
subsequent back electron transfer (ΔG°BET) were estimated with the
Weller equation, see the main text and the Supporting Information for
more information. Data from Fortage et al.13a
b
donor and acceptor and the fairly polar solvent (PhCN: ε =
25)51 we expect a larger reorganization energy (λ > 0.7 eV, see
the SI for further discussion). This estimate of the
reorganization energy suggests that the trend in electron
transfer rates is not simply due to the Marcus inverted region
effect.
Instead the trend in photoinduced electron transfer rates
could be explained by differences in electronic coupling. Trends
in coupling strengths can be rationalized on the basis of both
molecular structure and distribution of relevant orbitals. The
break in conjugation between bridge and acceptor can explain
why ZnPc-OPE-C60 appears to have the weakest electronic
coupling. Similarly, that the PET rate is higher in ZnPc-OPE-
AuP+ than in ZnPc-OPE-SnP can possibly be explained by the
fact that the latter has a vinyl group instead of an ethynyl group
as the linking unit between bridge and acceptor.13b Steric
interaction between the vinyl group and the hydrogen at the β-
position causes a ∼40° dihedral angle between the double bond
and the porphyrin plane, something which is expected to
induce a weaker π-conjugation and hence a weaker electronic
coupling (see the SI for calculated geometries).13b,c,52 However,
it should be noted that for both ZnPc-OPE-SnP and ZnPc-
OPE-AuP+ the LUMO is partly spread out from the acceptor
moiety onto the OPE bridge (Figure 6), indicating that both
have a fairly strong electronic coupling between the acceptor
and bridging moieties. Perturbation of the ground state
absorption spectrum is another way to assess the magnitude
of electronic coupling. We already know that interaction
between donor and acceptor strongly perturb the absorption
spectrum and the reduction potentials of ZnPc-OPE-AuP+
(vide supra). The effect of electronic coupling is less apparent in
the absorption spectrum of ZnPc-OPE-SnP, but a small extra
absorption can be seen in the region between 500 and 600 nm
(SI Figure S1a). In contrast, the observed spectrum for ZnPc-
OPE-C60 is an almost perfect match to the sum of the ZnPc-
OPE and C60 references, which indicates fairly little electronic
communication between ZnPc and C60. This trend is further
supported by the computational results above. The correlation
between electron transfer rates and degree of perturbation of
the absorption spectrum agrees with the hypothesis that it is
the electronic coupling that is the main reason for the
difference in reactivity between the three donor−acceptor
dyads.
Comparison of Photoinduced Electron Transfer Rates.
According to semiclassical Marcus theory (eq 2) electron
transfer rates can be explained and predicted by four
parameters: temperature (T), driving force (ΔG°), reorganiza-
tion energy (λ), and electronic coupling (VDA).1b,49 We will
discuss the observed trends in electron transfer rates between
the three dyads within this framework.
2
⎛
⎞
−(ΔG° + λ)
4λkBT
π
kET
=
VD2A exp
⎜
⎟
⎠
ℏ2λkBT
⎝
(2)
Driving forces for the three dyads were calculated from the
values in Table 1 using the Weller equation50 (Table 2, see SI
for details). ZnPc-OPE-C60 has the largest driving force for
photoinduced electron transfer (ΔG°PET), followed by ZnPc-
OPE-SnP and then ZnPc-OPE-AuP+, while the trend for the
rates for photoinduced electron transfer (kPET) goes in the
opposite direction, namely kPET, ZnPc‑OPE‑AuP+ > kPET, ZnPc‑OPE‑SnP
> kPET, ZnPc‑OPE‑C60 (Table 2). The observation of opposite
trends for rates and driving forces would be consistent if the
reorganization energy were small (λ = ∼0.6 eV), placing the
electron transfer processes in the Marcus inverted region (i.e., λ
< −ΔG°).1b,2b,49 However, due to the long distance between
Comparison of Back Electron Transfer Rates. The rates
for back electron transfer (kBET) do not follow the same trend
as the photoinduced electron transfer rates. ZnPc-OPE-C60,
which has the lowest forward rate, has the fastest back electron
transfer rate. Faster back electron transfer rate compared to
forward electron transfer is a fairly common feature in many
donor−acceptor dyads,13c,53 but it is difficult to explain why
J
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
this behavior only appears in the fullerene based dyad and not
in the two porphyrin based systems. In an extensive review by
G. Bottari et al.15d all the reported ZnPc-C60 dyads have charge
separation rates that are faster than the charge recombination
rates,54 so ZnPc-OPE-C60 is a surprising exception also in that
context.
A strong electronic coupling for forward electron transfer is
often associated with a strong coupling for the back electron
transfer, but ZnPc-OPE-AuP+ does not follow this pattern.
Studies by Kadish and co-workers on the reduction of gold(III)
porphyrins suggest a possible explanation.18d,e They found that
depending on the substitution pattern of the gold porphyrin,
the first reduction of an AuIIIP+ can occur on either the metal or
on the porphyrin ring. In contrast, tin(IV) porphyrins, like
most regular porphyrins, are expected to undergo ring based
reductions.55 Our hypothesis is that in both ZnPc-OPE-AuP+
and ZnPc-OPE-SnP photoinduced electron transfer results in
an initial reduction of the porphyrin ring, However, for ZnPc-
OPE-AuP+ the initial charge shifted state rapidly relaxes so that
the gold ion is reduced instead of the porphyrin ring.56 This
would influence the back electron transfer in two ways.
The first effect of metal based reduction of AuP+ is a change
in reorganization energy. Kadish and co-workers reported that
the reductive quenching of the triplet states of pyrene and
phenanthrene by a gold porphyrin was associated with an
unusual large reorganization energy (1.2 eV in PhCN), which
can be compared to the typical value for metalloporphryins
with ring centered reductions (∼0.6 eV).18e However, for
ZnPc-OPE-AuP+, a similar increase in reorganization energy
would actually be expected to result in a faster electron transfer
rate (since λ ≈ −ΔG°), so this effect is not the major factor
behind the slow back electron transfer.
The second effect is due to changes in the electronic
coupling. If the excess electron is mainly situated in metal based
orbitals it will interact less with the OPE bridge compared to if
it were situated on the porphyrin ring. To confirm this
hypothesis we utilized DFT to calculate the spin densities of
the reduced state of the three different OPE-acceptor
references. First, we optimized the geometries for the
nonreduced bridge-acceptor fragments. Second, the nuclei
were kept in the same position, but an extra electron was
introduced. The spin density distribution that followed from
this should approximate where the excess electron is situated
directly after the electron is transferred (“unrelaxed” in Figure
7). Finally, the reduced bridge-acceptor fragment was relaxed to
its lowest energy geometry. The resulting spin density should
now show where the electron is situated in the relaxed charge-
separated or charge-shifted state (“relaxed” in Figure 7). In the
case of ZnPc-OPE-AuP+ one can clearly see the change from a
ring-based reduction (unrelaxed) to a metal-centered reduction
(relaxed). This strongly supports our hypothesis that the
photoinduced electron transfer in ZnPc-OPE-AuP+ is first to a
charge-shifted state that is situated on the rings of the donor
and acceptor (ZnIIPc+•-OPE-AuIIIP•), which allows for a high
coupling in the forward direction. The charge-shifted state then
rapidly relaxes to a state where the excess electron in AuP is
mainly on the gold ion (ZnIIPc+•-OPE-AuIIP) where the
electron has been “trapped’’. The same kind of stabilization is
not possible in ZnPc-OPE-SnP, which may explain why the
back electron transfer is about an order of magnitude faster
than in ZnPc-OPE-AuP+. This “trapping” mechanism might be
part of the explanation to why dyad systems with AuP+ tend to
have long-lived charge shifted states.13d,14c,18a,g
Figure 7. Spin density (contour threshold 0.0004 au) of the radical of
the accepting side, using an unrelaxed or relaxed geometry (see text for
more details). From top to bottom: OPE-AuP+, OPE-SnP, and the
OPE-C60. Note how the spin density of relaxed AuP is centered on the
metal, while for relaxed SnP it is spread out over the π-conjugated
system.
CONCLUSIONS
■
We have shown that both ZnPc-OPE-AuP+ and ZnPc-OPE-
C60 in PhCN undergo photoinduced electron transfer. For
ZnPc-OPE-C60 the back electron transfer is much faster (kBET
≈ 5 × 1010 s−1) than the photoinduced forward reaction (kPET
= 1.1 × 109 s−1), which prevents build up of the desired charge-
separated state. In contrast, ZnPc-OPE-AuP+ undergoes
ultrafast, long-range photoinduced electron transfer over an
edge-to-edge distance of 23 Å with kPET = 1.0 × 1012 s−1,
followed by recombination with kBET = 1.0 × 109 s−1. The high
forward rate is due to strong electronic coupling, which is also
reflected in a strong perturbation of the steady-state absorption
spectrum. The comparably long-lived charge-shifted state is
formed because the reduced gold porphyrin relaxes from a
AuIIIP• radical to a AuIIP species, with the excess electron
density localized on the metal. The metal-centered reduced
species interacts less with the bridge and the electronic
interaction with the hole on ZnPc becomes smaller. ZnPc-
OPE-SnP, which is structurally very similar, does not undergo
the same kind of metal based reduction. In this case, the rate for
back electron transfer is an order of magnitude larger. This
information may be useful for the design of multichromophoric
systems that undergoes very fast photoinduced electron transfer
and still have a significant lifetime for the charge-separated
state.
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional spectroscopic data, details on calculations of rate
constants, driving forces and reorganization energies, Cartesian
coordinates for the three dyad systems, simulated UV/vis
spectra, list of the main vertical transitions and corresponding
orbital composition, and representation of all relevant
molecular orbitals. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: +33 (0)2 51 12 54 29. Fax: +33 (0)2 51 12 54 02. E-
mail: Denis.Jacquemin@univ-nantes.fr (D.J.) and Fabrice.
■
K
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Nature 2007, 445, 414−417. (f) Chen, J.; Reed, M. A.; Rawlett, A. M.;
Tour, J. M. Science 1999, 286, 1550−1552.
Odobel@univ-nantes.fr (F.O.). Phone: +46-18-471 3648; Fax:
+46-18-471 6844; E-mail: Leif.Hammarstrom@fotomol.uu.se
(L.H.).
(12) (a) Winters, M. U.; Pettersson, K.; Martensson, J.; Albinsson, B.
̊
Chem.–A Eur. J. 2005, 11, 562−573. (b) de la Torre, G.; Giacalone,
F.; Segura, J. L.; Martín, N.; Guldi, D. M. Chem.–Eur. J. 2005, 11,
1267−1280. (c) Giacalone, F.; Segura, J. L.; Martín, N.; Guldi, D. M. J.
Am. Chem. Soc. 2004, 126, 5340−5341. (d) Giacalone, F.; Segura, J. L.;
Martín, N.; Ramey, J.; Guldi, D. M. Chem.–Eur. J. 2005, 11, 4819−
4834. (e) Davis, W. B.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem.
Soc. 2001, 123, 7877−7886. (f) Davis, W. B.; Svec, W. A.; Ratner, M.
A.; Wasielewski, M. R. Nature 1998, 396, 60−63. (g) Odobel, F.;
Fortage, J. C. R. Chim. 2009, 12, 437−449.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Susanne Karlsson for providing valuable input to
this project while working at Uppsala University. Financial
support from the following agencies is gratefully acknowledged:
Knut and Alice Wallenberg foundation, the Swedish Reaseach
Council, the Agence Nationale de la Recherche (ANR)
“PhotoCumElec” program, ACI jeune chercheur, and the
European Community for the COST D35 program. D.
(13) (a) Fortage, J.; Goransson, E.; Blart, E.; Becker, H.-C.;
̈
Hammarstrom, L.; Odobel, F. Chem. Commun. 2007, 4629−4631.
̈
(b) Odobel, F.; Suresh, S.; Blart, E.; Nicolas, Y.; Quintard, J.-P.;
Janvier, P.; Le Questel, J.-Y.; Illien, B.; Rondeau, D.; Richomme, P.;
Haupl, T.; Wallin, S.; Hammarstrom, L. Chem.–A Eur. J. 2002, 8,
̈
̈
Jacquemin acknowledges the Reg
the European Research Council (ERC) for financial support in
the framework of a recrutement sur poste strategique and a
Starting Grant (Marches -278845), respectively. The theoretical
part of this research used resources from the following: 1) the
Interuniversity Scientific Computing Facility located at the
University of Namur, Belgium, which is supported by the
F.R.S.-FNRS under convention No. 2.4617.07, 2) the GENCI-
CINES/IDRIS (Grant c2011085117), and 3) the CCIPL
(Centre de Calcul Intensif des Pays de Loire).
́
ion des Pays de la Loire and
3027−3046. (c) Redmore, N. P.; Rubtsov, I. V.; Therien, M. J. J. Am.
Chem. Soc. 2003, 125, 8769−8778. (d) Fortage, J.; Boixel, J.; Blart, E.;
Hammarstrom, L.; Becker, H. C.; Odobel, F. Chem.–Eur. J. 2008, 14,
3467−3480. (e) Lembo, A.; Tagliatesta, P.; Guldi, D. M.; Wielopolski,
M.; Nuccetelli, M. J. Phys. Chem. A 2009, 113, 1779−1793.
́
̈
(14) (a) Fukuzumi, S. Electron Transfer Chemistry of Porphyrins
and Metalloporphyrins. In Porphyrin Handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: New York, USA, 1999; pp
115−152; (b) Mataga, N.; Chosrowjan, H.; Taniguchi, S. J. Photochem.
Photobiol., C 2005, 6, 37−79. (c) Wiberg, J.; Guo, L.; Pettersson, K.;
Nilsson, D.; Ljungdahl, T.; Martensson, J.; Albinsson, B. J. Am. Chem.
̊
Soc. 2006, 129, 155−163. (d) Holten, D.; Bocian, D. F.; Lindsey, J. S.
Acc. Chem. Res. 2001, 35, 57−69. (e) Kuciauskas, D.; Lin, S.; Seely, G.
R.; Moore, A. L.; Moore, T. A.; Gust, D.; Drovetskaya, T.; Reed, C. A.;
Boyd, P. D. W. J. Phys. Chem. 1996, 100, 15926−15932. (f) Pettersson,
REFERENCES
■
(1) (a) Gray, H. B.; Winkler, J. R. Proc. Natl. Acad. Sci. U.S.A. 2005,
102, 3534−3539. (b) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta,
Rev. Bioenerg. 1985, 811, 265−322.
(2) (a) Wasielewski, M. R. Chem. Rev. 1992, 92, 435−461. (b) Closs,
G. L.; Miller, J. R. Science 1988, 240, 440−447.
(3) (a) Tommos, C.; Babcock, G. T. Acc. Chem. Res. 1998, 31, 18−
25. (b) Ort, D. R.; Yocum, C. F.; Heichel, I. F. Oxygenic Photosynthesis:
The Light Reactions; Kluwer Academic Publisher: Dordrecht, The
Netherlands, 1996; Vol. 4.
(4) (a) Lubitz, W.; Reijerse, E.; van Gastel, M. Chem. Rev. 2007, 107,
4331−4365. (b) Pandelia, M.-E.; Nitschke, W.; Infossi, P.; Giudici-
Orticoni, M.-T.; Bill, E.; Lubitz, W. Proc. Natl. Acad. Sci. U.S.A. 2011.
(5) Gray, H. B.; Winkler, J. R. Biochim. Biophys. Acta, Bioenerg. 2010,
1797, 1563−1572.
K.; Wiberg, J.; Ljungdahl, T.; Martensson, J.; Albinsson, B. J. Phys.
̊
Chem. A 2005, 110, 319−326. (g) Guldi, D. M. Chem. Soc. Rev. 2002,
31, 22−36. (h) Imahori, H. Org. Biomol. Chem. 2004, 2, 1425−1433.
(i) Osuka, A.; Tanabe, N.; Kawabata, S.; Yamazaki, I.; Nishimura, Y. J.
Org. Chem. 1995, 60, 7177−7185. (j) Nakano, A.; Osuka, A.;
Yamazaki, T.; Nishimura, Y.; Akimoto, S.; Yamazaki, I.; Itaya, A.;
Murakami, M.; Miyasaka, H. Chem.–A Eur. J. 2001, 7, 3134−3151.
(k) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 1993, 26,
198−205. (l) Kim, D.; Osuka, A. Acc. Chem. Res. 2004, 37, 735−745.
(15) (a) Lo, P.-C.; Leng, X.; Ng, D. K. P. Coord. Chem. Rev. 2007,
251, 2334−2353. (b) Maligaspe, E.; Kumpulainen, T.; Lemmetyinen,
H.; Tkachenko, N. V.; Subbaiyan, N. K.; Zandler, M. E.; D’Souza, F. J.
Phys. Chem. A 2009, 114, 268−277. (c) Martinez-Diaz, M. V.; de la
Torre, G.; Torres, T. Chem. Commun. 2010, 46, 7090−7108.
(d) Bottari, G.; de la Torre, G.; Guldi, D. M.; Torres, T. Chem. Rev.
2010, 110, 6768−6816.
(16) (a) Gonzalez-Cabello, A.; Vazquez, P.; Torres, T. J. Organomet.
Chem. 2001, 637−639, 751−756. (b) Sutton, J. M.; Boyle, R. W.
Chem. Commun. 2001, 2014−2015. (c) Garcia-Frutos, E. M.;
O’Flaherty, S. M.; Maya, E. M.; De La Torre, G.; Blau, W.;
Vazquez, P.; Torres, T. J. Mater. Chem. 2003, 13, 749−753.
(d) Imahori, H.; Umeyama, T.; Ito, S. Acc. Chem. Res. 2009, 42,
1809−1818.
(6) (a) Bred
́
as, J.-L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc.
Chem. Res. 2009, 42, 1691−1699. (b) Inganas, O.; Zhang, F.;
̈
Andersson, M. R. Acc. Chem. Res. 2009, 42, 1731−1739. (c) Roncali, J.
Acc. Chem. Res. 2009, 42, 1719−1730.
(7) (a) Gratzel, M. Acc. Chem. Res. 2009, 42, 1788−1798.
̈
(b) O’Regan, B. C.; Durrant, J. R. Acc. Chem. Res. 2009, 42, 1799−
1808. (c) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737−740.
̈
(d) Hagfeldt, A.; Gratzel, M. Acc. Chem. Res. 2000, 33, 269−277.
̈
(8) (a) Youngblood, W. J.; Lee, S.-H. A.; Maeda, K.; Mallouk, T. E.
Acc. Chem. Res. 2009, 42, 1966−1973. (b) Gust, D.; Moore, T. A.;
Moore, A. L. Acc. Chem. Res. 2009, 42, 1890−1898. (c) Magnuson, A.;
Anderlund, M.; Johansson, O.; Lindblad, P.; Lomoth, R.; Polivka, T.;
(17) (a) Hardin, B. E.; Hoke, E. T.; Armstrong, P. B.; Yum, J.-H.;
Comte, P.; Torres, T.; Frechet, J. M. J.; Nazeeruddin, M. K.; Gratzel,
M.; McGehee, M. D. Nat. Photonics 2009, 3, 406−411. (b) Brown, M.
D.; Parkinson, P.; Torres, T.; Miura, H.; Herz, L. M.; Snaith, H. J. J.
Phys. Chem. C 2011, 115, 23204−23208. (c) Gonzalez-Cabello, A.;
Vazquez, P.; Torres, T.; Guldi, D. M. J. Org. Chem. 2003, 68, 8635−
8642.
Ott, S.; Stensjo, K.; Styring, S.; Sundstrom, V.; Hammarstrom, L. Acc.
̈
̈
̈
Chem. Res. 2009, 42, 1899−1909.
(9) Tour, J. M. Acc. Chem. Res. 2000, 33, 791−804.
(10) (a) Wagner, R. W.; Lindsey, J. S. J. Am. Chem. Soc. 1994, 116,
9759−9760. (b) Wagner, R. W.; Lindsey, J. S.; Seth, J.; Palaniappan,
V.; Bocian, D. F. J. Am. Chem. Soc. 1996, 118, 3996−3997.
(18) (a) Harriman, A.; Heitz, V.; Sauvage, J. P. J. Phys. Chem. 1993,
97, 5940−5946. (b) Brun, A. M.; Harriman, A.; Heitz, V.; Sauvage, J.
(11) (a) Gust, D.; Moore, T. A.; Moore, A. L. Chem. Commun. 2006,
1169−1178. (b) Balzani, V. Photochem. Photobiol. Sci. 2003, 2, 459−
476. (c) Aviram, A.; Ratner, M. A. Chem. Phys. Lett. 1974, 29, 277−
283. (d) Debreczeny, M. P.; Svec, W. A.; Wasielewski, M. R. Science
1996, 274, 584−587. (e) Green, J. E.; Wook Choi, J.; Boukai, A.;
Bunimovich, Y.; Johnston-Halperin, E.; DeIonno, E.; Luo, Y.; Sheriff,
B. A.; Xu, K.; Shik Shin, Y.; Tseng, H.-R.; Stoddart, J. F.; Heath, J. R.
P. J. Am. Chem. Soc. 1991, 113, 8657−8663. (c) Kilsa, K.; Kajanus, J.;
̊
Macpherson, A. N.; Martensson, J.; Albinsson, B. J. Am. Chem. Soc.
̊
2001, 123, 3069−3080. (d) Kadish, K. M.; E, W.; Ou, Z.; Shao, J.;
Sintic, P. J.; Ohkubo, K.; Fukuzumi, S.; Crossley, M. J. Chem. Commun.
2002, 356−357. (e) Ou, Z.; Kadish, K. M.; E, W.; Shao, J.; Sintic, P. J.;
L
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Ohkubo, K.; Fukuzumi, S.; Crossley, M. J. Inorg. Chem. 2004, 43,
2078−2086. (f) Fortage, J.; Boixel, J.; Blart, E.; Becker, H. C.; Odobel,
F. Inorg. Chem. 2009, 48, 518−526. (g) Andersson, M.; Linke, M.;
emission measurements (1.8 ns). The “infinite” component comes
3
from the very long-lived ZnPc, which forms in a significant amount
1
from the ZnPc state due to the slow electron transfer rate, which
allows the intersystem crossing to compete.
Chambron, J.-C.; Davidsson, J.; Heitz, V.; Hammarstrom, L.; Sauvage,
̈
(41) Since we used a cutoff filter to block visible light, we can exclude
that the absorption in the NIR is an artifact due to second order
diffraction of visible light.
(42) Detection of the ZnPc+ radical signal at ∼850 nm was
precluded due to instrument limitations to detect in around 800 nm
(see the Experimental Section).
J.-P. J. Am. Chem. Soc. 2002, 124, 4347−4362.
(19) Prato, M.; Maggini, M. Acc. Chem. Res. 1998, 31, 519−526.
(20) (a) Donhauser, Z. J.; Mantooth, B. A.; Kelly, K. F.; Bumm, L. A.;
Monnell, J. D.; Stapleton, J. J.; Price, D. W.; Rawlett, A. M.; Allara, D.
L.; Tour, J. M.; Weiss, P. S. Science 2001, 292, 2303−2307. (b) Weber,
H. B.; Reichert, J.; Weigend, F.; Ochs, R.; Beckmann, D.; Mayor, M.;
(43) Guldi, D. M.; Prato, M. Acc. Chem. Res. 2000, 33, 695−703.
(44) The spectral changes in the DADS are blue-shifted about 20 nm
compared to the features in the simulated charge separated state
(Figure 2b and SI Figure S7c). Most likely this mismatch is due to the
strong interaction between ZnPc and AuP+, which perturbs the bands
in the Soret band region. Since the Soret band region in the ground
state could not be fully described by a sum of ZnPc and AuP+ it is
reasonable to expect that a ZnPc•+-OPE-AuIIIP and ZnPc-OPE-AuIIP
could not fully describe the charge separated state, ZnPc•+-OPE-AuIIP.
(45) The small fraction of unquenched 1ZnPc and the spectral
similarity beteween the 1ZnPc and 3ZnPc makes it difficult to resolve
the 1.8 ns time component with fs-TA. This is in contrast to the
TCSPC measurement where there are no interfering signals from the
ZnPc units that undergo photoinduces electron transfer (i.e., the
majority of ZnPc).
Ahlrichs, R.; Lohneysen, H. v. Chem. Phys. 2002, 281, 113−125.
̈
(21) Ik Yang, S.; Lammi, R. K.; Prathapan, S.; Miller, M. A.; Seth, J.;
Diers, J. R.; Bocian, D. F.; Lindsey, J. S.; Holten, D. J. Mater. Chem.
2001, 11, 2420−2430.
(22) Gonzalez-Rojano, N.; Arias-Marin, E.; Navarro-Rodriguez, D.;
Weidner, S. Synlett 2005, 1259−1262.
(23) Chaignon, F.; Torroba, J.; Blart, E.; Borgstrom, M.;
̈
Hammarstrom, L.; Odobel, F. New J. Chem. 2005, 29, 1272−1284.
̈
(24) Maya, E. M.; Vazquez, P.; Torres, T. Chem.–Eur. J. 1999, 5,
2004−2013.
(25) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.
(26) Rhile, I. J.; Markle, T. F.; Nagao, H.; DiPasquale, A. G.; Lam, O.
P.; Lockwood, M. A.; Rotter, K.; Mayer, J. M. J. Am. Chem. Soc. 2006,
128, 6075−6088.
(27) Habenicht, A.; Hjelm, J.; Mukhtar, E.; Bergstrom, F.; Johansson,
̈
(46) Andrea
́
sson, J.; Kodis, G.; Lin, S.; Moore, A. L.; Moore, T. A.;
L. B. Å. Chem. Phys. Lett. 2002, 354, 367−375.
Gust, D.; Martensson, J.; Albinsson, B. Photochem. Photobiol. 2002, 76,
̊
(28) Petersson, J.; Eklund, M.; Davidsson, J.; Hammarstrom, L. J.
̈
47−50.
Phys. Chem. B 2010, 114, 14329−14338.
(29) WaveMetrics, Igor Pro 5.
́
̊
(47) Eng, M. P.; Ljungdahl, T.; Andreasson, J.; Martensson, J.;
Albinsson, B. J. Phys. Chem. A 2005, 109, 1776−1784.
(30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Lyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M. a. K, M.; a. K, J. E.; a. C, J. B.; a. B, V.; a. A, C.; a. J, J.;
a. R. Gaussian 09 Revision A.02; Gaussian Inc.: Wallingford, CT, 2009.
(48) (a) Mizutani, Y.; Kitagawa, T. J. Mol. Liq. 2001, 90, 233−242.
(b) Pigliucci, A.; Duvanel, G.; Daku, L. M. L.; Vauthey, E. J. Phys.
Chem. A 2007, 111, 6135−6145. (c) Petersson, J.; Eklund, M.;
Davidsson, J.; Hammarstrom, L. J. Phys. Chem. B 2010, 114, 14329−
̈
14338.
(49) Marcus, R. A. J. Chem. Phys. 1965, 43, 679−701.
(50) (a) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259−271.
(b) Weller, A. Z. Phys. Chem. 1982, 133, 93−98.
(51) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. Handbook of
Photochemistry, 3rd ed.; CRC Press, Taylor & Francis: Boca Raton, FL,
USA, 2006.
(31) Jacquemin, D.; Perpet
Res. 2009, 42, 326−334.
(32) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6170.
(33) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3094.
(34) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393,
51−56.
(35) (a) Peach, M. J. G.; Benfield, P.; Helgaker, T.; Tozer, D. J. J.
Chem. Phys. 2008, 128, 044118. (b) Goerigk, L.; Grimme, S. J. Chem.
̀
e, E. A.; Ciofini, I.; Adamo, C. Acc. Chem.
(52) (a) Wallin, S.; Hammarstrom, L.; Blart, E.; Odobel, F.
̈
Photochem. Photobiol. Sci. 2006, 5, 828−834. (b) Eng, M. P.;
Albinsson, B. Chem. Phys. 2009, 357, 132−139.
(53) (a) Fortage, J.; Goransson, E.; Blart, E.; Becker, H.-C.;
̈
Hammarstrom, L.; Odobel, F. Chem. Commun. 2007, 4629−4631.
̈
(b) Kirmaier, C.; Yang, S. I.; Prathapan, S.; Miller, M. A.; Diers, J. R.;
Bocian, D. F.; Lindsey, J. S.; Holten, D. Res. Chem. Intermed. 2002, 28,
719−740. (c) Palacios, R. E.; Kodis, G.; Herrero, C.; Ochoa, E. M.;
Gervaldo, M.; Gould, S. L.; Kennis, J. T. M.; Gust, D.; Moore, T. A.;
Moore, A. L. J. Phys. Chem. B 2006, 110, 25411−25420.
̀
Phys. 2010, 132, 184103. (c) Jacquemin, D.; Perpete, E. A.; Ciofini, I.;
Adamo, C. J. Chem. Theory Comput. 2009, 5, 2420−2435.
(36) Cossi, M.; Barone, V. J. Chem. Phys. 2001, 115, 4708−4717.
(37) Tagmatarchis, N.; Prato, M. Synlett 2003, 768−779.
́
(54) (a) Gouloumis, A.; de la Escosura, A.; Vazquez, P.; Torres, T.;
Kahnt, A.; Guldi, D. M.; Neugebauer, H.; Winder, C.; Drees, M.;
Sariciftci, N. S. Org. Lett. 2006, 8, 5187−5190. (b) Guldi, D. M.;
́
Gouloumis, A.; Vazquez, P.; Torres, T.; Georgakilas, V.; Prato, M. J.
(38) (a) Gonzal
́
ez-Rodríguez, D.; Claessens, C. G.; Torres, T.; Liu,
S.; Echegoyen, L.; Vila, N.; Nonell, S. Chem.–A Eur. J. 2005, 11, 3881−
Am. Chem. Soc. 2005, 127, 5811−5813. (c) Guldi, D. M.; Gouloumis,
A.; Vazquez, P.; Torres, T. Chem. Commun. 2002, 2056−2057.
(d) Isosomppi, M.; Tkachenko, N. V.; Efimov, A.; Vahasalo, H.;
Jukola, J.; Vainiotalo, P.; Lemmetyinen, H. Chem. Phys. Lett. 2006, 430,
36−40. (e) Kahnt, A.; Guldi, D. M.; de la Escosura, A.; Martinez-Diaz,
M. V.; Torres, T. J. Mater. Chem. 2008, 18, 77−82. (f) Quintiliani, M.;
Kahnt, A.; Vazquez, P.; Guldi, D. M.; Torres, T. J. Mater. Chem. 2008,
18, 1542−1546.
(55) Kalyanasundaram, K. Photochemistry of Polypyridine and
Porphyrin Complexes; Academic Press: London, UK, 1992.
(56) DFT indicated no significant barrier, so we assume that this
relaxation process occurs on a subps time scale. The unknown liftime
(∼120 ps), which was found in the global analysis, is to slow to be a
plausible candidate for this reaction.
́
́
3893. (b) Jimenez, A. J.; Spanig, F.; Rodríguez-Morgade, M. S.;
̈
Ohkubo, K.; Fukuzumi, S.; Guldi, D. M.; Torres, T. Org. Lett. 2007, 9,
2481−2484.
(39) (a) Lin, V.; DiMagno, S.; Therien, M. Science 1994, 264, 1105−
1111. (b) Lin, V. S. Y.; Therien, M. J. Chem.–A Eur. J. 1995, 1, 645−
651. (c) Shediac, R.; Gray, M. H. B.; Uyeda, H. T.; Johnson, R. C.;
Hupp, J. T.; Angiolillo, P. J.; Therien, M. J. J. Am. Chem. Soc. 2000,
122, 7017−7033. (d) Winters, M. U.; Karnbratt, J.; Eng, M.; Wilson,
̈
C. J.; Anderson, H. L.; Albinsson, B. J. Phys. Chem. C 2007, 111,
7192−7199.
(40) Two longer lifetimes, 1.8 ns and ∞ (both held fixed), were
needed for a satisfactory global fit of the visible data. The 1.8 ns
component corresponds to the decay of the singlet excited state of the
small unquenched fraction of ZnPc observed in the time-resolved
M
dx.doi.org/10.1021/ic3013552 | Inorg. Chem. XXXX, XXX, XXX−XXX