Modulating Charge-Transfer Interactions in Topologically Different Porphyrin-C60 Dyads

Article abstract of DOI:10.1002/chem.200304995

Control over the interchromophore separation, their angular relationship, and the spatial overlap of their electronic clouds in several ZnP-C ₆₀ dyads (ZnP=zinc porphyrin) is used to modulate the rates of intramolecular electron transfer. For the first time, a detailed analysis of the charge transfer absorption and emission spectra, time-dependent spectroscopic measurements, and molecular dynamics simulations prove quantitatively that the same two moieties can produce widely different electron-transfer regimes. This investigation also shows that the combination of ZnP and C₆₀ consistently produces charge recombination in the inverted Marcus region, with reorganization energies that are remarkably low, regardless of the solvent polarity. The time constants of electron transfer range from the μs to the ps regime, the electronic couplings from a few tens to several hundreds of cm^-1, and the reorganization energies remain below 0.54 eV and can be as low as 0.16 eV.

Full text of DOI:10.1002/chem.200304995

FULL PAPER
Modulating Charge-Transfer Interactions in Topologically Different
Porphyrin C60 Dyads
Dirk M. Guldi,*[a] Andreas Hirsch,*[b] Michael Scheloske,[b] Elke Dietel,[b]
Alessandro Troisi,[c] Francesco Zerbetto,*[c] and Maurizio Prato*[d]
Dedicated to Dr. Pedastur Neta on the occasion of his 65th birthday
Abstract: Control over the interchro-
mophore separation, their angular rela-
tionship, and the spatial overlap of their
electronic clouds in several ZnP C60
dyads (ZnP ˆ zinc porphyrin) is used to
modulate the rates of intramolecular
electron transfer. For the first time, a
detailed analysis of the charge transfer
absorption and emission spectra, time-
dependent spectroscopic measurements,
and molecular dynamics simulations
prove quantitatively that the same two
moieties can produce widely different
electron-transfer regimes. This investi-
gation also shows that the combination
of ZnP and C60 consistently produces
charge recombination in the inverted
Marcus region, with reorganization en-
ergies that are remarkably low, regard-
less of the solvent polarity. The time
constants of electron transfer range
from the ms to the ps regime, the
electronic couplings from a few tens to
several hundreds of cmÀ1, and the reor-
ganization energies remain below
0.54 eV and can be as low as 0.16 eV.
Keywords: charge separation
charge transfer ¥ electronic coupling
fullerenes porphyrinoids
reorganization energy
¥
¥
¥
¥
particular interest are artificial model systems in which the
introduction of simple molecular changes is used to control
and tune the magnitude of the electron-transfer (ET)
parameters. Typically, to ensure rapid, intramolecular ET
processes, either in the form of energy or electron trans-
duction, synthetic chemistry exploits the formation of cova-
lent bonds[2] or the use of biomimetic methodologies, such as
the formation of p-stacks, hydrogen bonds, van der Waals
contacts, and so forth.[3] These and other effective strategies
are borrowed from the organization principle of the bacterial
photosynthetic reaction center[4] in which the electron-donor
is basically a porphyrinoid moiety. Both the arrangement
and respective orientation of the electron donor and accep-
tor are the result of the incorporation of light- and redox-
active components into well-defined transmembrane pro-
teins.
Introduction
The composition, the interchromophore separation/angular
relationship, the overall dynamic and stimulus-induced reor-
ganization, and the electronic coupling are crucial factors in
the development of charge transfer reaction centers.[1] Of
[a] Dr. habil. D. M. Guldi
Radiation Laboratory, University of Notre Dame
Notre Dame, IN 46556 (USA)
Fax : (‡1)574-631-8068
E-mail: guldi.1@nd.edu
[b] Prof. Dr. A. Hirsch, Dipl. Chem. M. Scheloske, Dr. E. Dietel
Institut f¸r Organische Chemie der Universit‰t Erlangen-N¸rnberg
Henkestrasse 42, 91054 Erlangen (Germany)
E-mail: hirsch@organik.uni-erlangen.de
[c] Prof. F. Zerbetto, Dr. A. Troisi
Dipartimento di Chimica ™G. Ciamician∫
In the artificial models, the donor acceptor link plays an
essential role and is a potent means to restrain the electron-
donor acceptor couples in their positions.[1] Synthetic chem-
istry tools can be used to create the linkage and introduce a
spacer between the donor and acceptor. Its effect is not
limited to the structure, but also determines the size of the
electronic coupling matrix element (V) between the frag-
ments it separates, ideally the donor and acceptor. An
important feature of a spacer is the possibility of introducing
a systematic alteration of separation, orientation, and overlap
without affecting the electronic nature of the donor acceptor
¡
Universita degli Studi di Bologna
V. F. Selmi 2, 40126 Bologna (Italy)
[d] Prof. Dr. M. Prato
Dipartimento di Scienze Farmaceutiche
¡
Universita di Trieste, Piazzale Europa 1
34127 Trieste (Italy)
E-mail: prato@univ.trieste.it
Supporting information for this article is available on the WWW under
http://www.chemeurj.org or from the author. Figure S1: top and side
views of trans-2-ZnP C60 (2), meta-ZnP C60 (4) and para-ZnP C60
(1); Figure S2: superposed structure of meta-ZnP C60 (4); Figure S3:
details regarding the determination of the thermodynamic parameters.
4968
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200304995
Chem. Eur. J. 2003, 9, 4968 4979

4968 4979
pair, whereby the coupling is proportional to the overlap of
their electronic clouds. In other words, the exponential decay
with the distance of the coupling may be modified by
contributions from an intervening medium, for example,
solvent.
After the donor and the spacer(s), suitable acceptors need
to be identified. Fullerenes have proved to be a rich platform
13]
on which to conduct charge-separation (CS) studies.[5
These three-dimensional building blocks are now readily
available and exhibit three characteristics of great relevance
for ETreactions:
1) Fullerenes are extremely good electron acceptors.[14]
2) Fullerenes undergo multiple derivatization processes and
can form a number of isomers in which two hexagon
hexagon edges have reacted to provide the attachment
point for other molecules to be tethered.[15]
3) Primarily because of its rigid structure, fullerenes possess a
remarkably low reorganization energy upon ETreac-
tions.[16] A small reorganization energy is one of the key-
factors in decreasing energy waste. In general, two-dimen-
sional electron acceptors have much larger reorganization
energies. The small reorganization energy of C60 renders it
an extremely useful building block for exploring the
modulation of rates and yields of charge separation (CS)/
charge recombination (CR), for which conventional elec-
tron acceptors, such as quinones, have proved unsatisfac-
tory.
It is notable that despite the number of synthetic model
systems so far investigated,[10a, 17] the detailed description of
geometrical factors that govern EThas remained unclear. In
this paper, we compare the effects that arise from structurally
different donor acceptor topologies and demonstrate that
the variation of distance and mutual orientation of a linked
zinc tetraphenylporphyrin (ZnP) and fullerene (C60) reactant
pair affects the charge-transfer (CT) parameters. The simplest
way to alter the relative orientation of a two-dimensional
planar electron donor (ZnP) with respect to a three-dimen-
sional spherical electron acceptor (C60) is through synthetic
modifications that are now available. An important issue, in
this context, is the intramolecular flexibility and the associ-
ated variation of distance and orientation that can occur on a
timescale comparable or faster than ET. Here, a combination
of molecular dynamics and quantum chemical calculations is
used to estimate these ™fast-factors∫ as a function of the
solvent. Despite different absolute rate constants for the
forward and backward ETprocesses, which reflect the degree
of interactions in the ground state and in the excited state, all
systems reveal a fundamental common feature: the charge
recombination kinetics are consistently located in the inverted
region of the ™Marcus parabola∫.
ized in Scheme 1. The precursor malonate 5 was synthesized
via the substituted 3-hydroxybenzaldehyde 6, with a malonate
terminus. The co-condensation of 6 with benzaldehyde and
pyrrole under Lindsey conditions[19] with Ph4PCl and
BF3 ¥ Et2O as a catalyst in CH2Cl2 afforded the porphyrin
precursor 7, which was subsequently metallated with
Zn(OAc)2 to give porphyrin malonate 5. This compound
1
was fully characterized by a variety of techniques: H NMR,
13C NMR, and UV-visible spectroscopy and fast-atom bom-
bardment mass spectrometry (FAB-MS). Finally, Bingel
cyclopropanation of 5 with C60 led to the formation of the
dyad 4.
First, the singly linked porphyrin fullerene ensembles shall
be discussed. In the two ensembles, the substitution pattern of
the phenyl groups of the porphyrin differs, that is they are
either meta- or para-linked. This imposes a notable impact on
the mutual orientation of porphyrin and fullerene. In partic-
ular, the para-substituted porphyrin system (1) is expected to
furnish closer contacts but poorer overlap with the fullerene
than the meta-analogue (4), because of steric constraints. For
top- and side-view illustrations see the Supporting Informa-
tion (S1). Despite the flexibility of the ethylene bridge,
molecular modeling leads to the conclusion that the upper
part of the cage of the fullerene is covered by the porphyrin
(see Supporting Information (S2)).
Results and Discussion
Synthesis and structural modeling: The synthesis and struc-
tural characterization of the dyads 1,[18a] 2,[18b,c] and 3[18d] has
been described previously. They were prepared by one- or
twofold cyclopropanation of C60 and the corresponding
bismalonates.[18] The synthesis of the new dyad 4 is summar-
To introduce more rigid packing into the porphyrin ful-
lerene ensembles, a double-linkage was pursued.[20, 21] Model-
ing of doubly linked porphyrin fullerene dyads suggests the
use of meta- rather than para-substituted adducts.
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4969

D. M. Guldi, A. Hirsch, F. Zerbetto, M. Prato et al.
FULL PAPER
Scheme 1. Synthesis of the mono-linked 4: i) benzaldehyde, pyrrole, Ph4PCl, BF3*Et2O, CH2Cl2, RT; ii) p-chloranil, reflux, (silicagel, CH2Cl2/EtOAc (20:1));
iii) Zn(OAc)2 ¥ 2H2O, THF, reflux; iv) C60, DBU, toluene.
The dyads trans-2-ZnP C60 (2) and equatorial-ZnP C60
(3) were modeled by using MM ‡ force field.[22] The results
are shown in Figures 1 and 2, respectively. Molecular dynam-
ics calculations of six randomly selected conformers of both of
these dyads showed that:
one (99.8%). Both systems therefore possess strong
rigidity.
2) The face-to-face alignment in 2 with the shortest inter-
À
planar distance (Zn C: 3 ä) leads to appreciable p
p
stacking, whereas the face-to-edge alignment in 3 prohibits
effective p p interaction. The resulting distances (Ta-
ble 1) suggest that one has to consider the product of
Table 1. Ground-state absorption maxima and their extinction coefficients
of different ZnP C60 donor acceptor ensembles in CH2Cl2 at room
temperature.
2
4
3
meta-ZnP
1
para-ZnP
Ree [ä][a]
Soret-
band
3.0[b]
429
(5.44)
552
5.86
427
(5.44)
551
6.18
425
(5.46)
549
3.23
418
(5.63)
548
421
(5.76)
548
420
(5.66)
549
Q-
bands
(4.21)
591
(4.29)
588
(4.26)
584
(4.34)
587
(4.24)
587
(4.50)
589
Figure 1. Superimposed conformers of dyad 2.
(3.63)
(3.84)
(3.63)
(3.59)
(3.61)
(3.53)
À
[a] Edge-to-edge distances. [b] Zn C distance.
distance and overlap to assess the interaction between
porphyrin and fullerene. It can also be noted that close
separation but little overlap can be compensated by good
overlap and larger distances–compare, for example, para-
ZnP C60 (1) and meta-ZnP C60 (4) (Supporting Infor-
mation S1).
3) Dyad 3 can accommodate a solvent molecule, such as
toluene, between the two moieties.
Absorption spectroscopy: The close proximity of the por-
phyrin and fullerene p-systems gives rise to through-space[23]
and through-bond interactions, whose effects may be detect-
able in the shifts of some absorption bands or by the presence
of new CTbands in the spectra of the four dyads. We show in
the following that these features emerge here as sensitive
markers for assessing the electronic coupling in ZnP C60
ensembles. Red shifts of the Soret and Q-band transitions
are consistently observed in the dyads and are reported in
Table 1, in which similar data for the model porphyrin systems
are also given. To a good extent, the shifts track the
interchromophore separations (Ree) of the donor acceptor
ensembles: 2 (l ˆ 429, 552, 591 nm; Ree ˆ 3.0 ä) > 4 (l ˆ 427,
Figure 2. Superimposed conformers of dyad 3.
1) Due to the double linkage, the variability of relative
orientation of the two chromophores within each dyad is
very limited. There is very little conformational space
accessible. In 2, the face-to-face alignment is completely
locked, while in 3 only two types of face-to-edge orienta-
tions (vertical and horizontal) were found. NOE measure-
ments revealed, however, that the horizontal conformation
has a very low population (0.2%) relative to the vertical
4970
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4968 4979

Porphyrin C60 Dyads
4968 4979
551, 588 nm; Ree ˆ 5.86 ä) > 3(l ˆ 425, 549, 584 nm; Ree ˆ
6.18 ä). The reference values for the absorption of ZnP are
420 421, 548 549, and 587 589 nm. Deviations are ob-
served for the highest Q-bands, which probably interact with
other electronic states, and for 1, which is characterized by
bands only slightly shifted from the reference values, a feature
indicative of small interfragment interactions. The data
confirm the idea of the existence of a synergism between
separation and overlap, as illustrated in Figure S1 in the
Supporting Information.
The solvent effect on the CT band concurs with the
ordering of the couplings obtained from Equation (1). In-
creasing the solvent polarity from toluene to benzonitrile
leads to strong red shifts of the CTabsorption in 2 and 4–see
Figure 3. The effect is caused by the better solvation that
At longer wavelengths (ꢀ700 nm) only the weak fullerene
transitions (e ꢀ 200mÀ1 cmÀ1) should exist with maxima at
690 nm (mono-adduct), 696 nm (trans-2 bis-adduct) and
694 nm (equatorial bis-adduct).[24] However, subtraction of
the fullerene absorption from the spectra of the four dyads
reveals a new broad band in the 700 800 nm region. Since
neither ZnP nor C60 exhibit this feature, it is safe to assign this
additional band to a CTtransition between the porphyrin
donor and the electron-accepting fullerene. The presence of
the additional band, therefore, confirms electronic interaction
between the two chromophores in the dyads.
Figure 3. Charge-transfer absorption of dyad 2 in toluene (solid line), o-
dichlorobenzene (dashed line) and benzonitrile (dotted line) at room
temperature.
The intensity trend of the CT bands in toluene, reported in
Table 2, is reminiscent of the trend observed for the spectral
shifts described above. The most intense band (emax ˆ 1470
radical cations and/or radical anions experience in a more
polar environment. As an example, Figure 3 illustrates the CT
bands for 2 in toluene, o-dichlorobenzene, and benzonitrile,
which shift from 723 nm to 732 nm and finally to 749 nm. In
dyad 4 the maxima are also red-shifted upon increasing the
solvent polarity: 727 nm in toluene, 730 nm in chloroform,
734 nm in o-dichlorobenzene, and 754 nm in benzonitrile. The
weakness of the transition in 1 limited the validity of such an
analysis.
The spectral analysis carried out on 1) the band shifts of the
ZnP fragment, 2) the intensity of the newly formed CTbands,
and 3) their shifts upon variation of the solvent polarity proves
that the strength of electronic interaction between ZnP and
Table 2. Charge transfer absorption and electronic coupling in different
ZnP C60 donor acceptor ensembles in toluene at room temperature.
2
4
3
1
Rcc [ä][a]
6.6
1470
13831
961
436
9.5
740
13717
847
201
15.1
250
14084
900
7.9
10
13550
ꢀ 1000
30
emax [mÀ1 cmÀ1
]
nÄmax [cmÀ1
]
nÄ1/2 [cmÀ1
V [cmÀ1
]
]
76
[a] Center-to-center distances as measured from the zinc center.
MÀ1 cmÀ1) appears for 2, with its face-to-face structure, while
only a weak band (emax ˆ 250 MÀ1 cmÀ1) is present for 3 with its
face-to-edge orientation. As before, the data for 4 fall
between the other two, while the intensity of the new band
of 3 is the lowest. It should be further noted that the ordering
of the interactions deduced by the band shifts and the relative
intensities in the various dyads follows the anodic shifts of the
reduction of the fullerene moiety.[18] For instance, in dyad 2
the first reduction is anodically shifted by nearly 100 mV,
relative to that of 4, while in 3 it is shifted to higher energies
than for 2.
C
60 in the ground state is 2 > 4 > 3 > 1 and varies by more than
an order of magnitude.
Fluorescence spectroscopy: To study the electronic interac-
tions between the porphyrin and fullerene p-systems in the
excited state, emission measurements were carried out with
the ZnP C60 dyads in solvents of different polarity (i.e.,
toluene, chloroform, THF, dichloromethane, o-dichloroben-
zene, and benzonitrile). The 1*ZnP emission becomes
quenched in all solvents for all ZnP
yields (FF) usually of the order of 10À4 (Table 3)–in the ZnP
C60 dyads with quantum
The CT spectra are used to estimate the electronic coupling
element (V) by means of Equation (1):[25]
Table 3. Room temperature fluorescence quantum yields (1*ZnP) of
different ZnP
tion wavelength: 555 nm).
C60 donor acceptor ensembles in various solvents (excita-
1=2
2:06 Â 10À2ꢁemax
n
maxDn1=2
†
Vˆ
(1)
Rcc
2
4
3
1
toluene
chloroform
THF
1.26 Â 10À4 5.50 Â 10À4
1.31 Â 10À4 3.58 Â 10À4
15.9 Â 10À4
15.8 Â 10À4
in which Rcc denotes the shortest center-to-center separa-
tion, emax the extinction coefficient, nmax the energy, and n1/2 the
halfwidth of the CTtransition. The electronic coupling in
trans-2-ZnP C60 (2), meta-ZnP C60 (4), equatorial-ZnP
C60 (3), and para-ZnP C60 (1) is 436, 201, 76, and 30 cmÀ1,
respectively.
6.8Â10À4
1.16 Â 10À4 3.68 Â 10À4 12.0 Â 10À4 15.6 Â 10À4
dichloromethane
1.37 Â 10À4 1.81 Â 10À4
4.4 Â 10À4 14.9 Â 10À4
9.5 Â 10À4
o-dichlorobenzene 1.43 Â 10À4
benzonitrile
DMF
1.37 Â 10À4 3.42 Â 10À4
5.6 Â 10À4 12.6 Â 10À4
4.3 Â 10À4 12.1 Â 10À4
2.91 Â 10À4
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4971

D. M. Guldi, A. Hirsch, F. Zerbetto, M. Prato et al.
FULL PAPER
reference FF ˆ 0.04. Although the yield drops, the ZnP
emission patterns are not affected by the presence of the
fullerene fragment.
A close inspection of the fluorescence data reveals a
number of interesting features:
1) The fluorescence quantum yield of dyad 2 does not change
appreciably with the solvent. This is attributed to the
intramolecular charge-transfer reaction, which is mediated
through the overlapping p-orbitals and does not use
bridging solvent molecules.
2) The two dyads with the largest edge-to-edge separation
(Ree), that is 4 (5.86 ä) and 3 (6.18 ä), show a similar
solvent dependence, that is, appreciably stronger fluores-
cence quenching in the polar solvents than in the nonpolar
media.
The emission spectra therefore indicate that at short
donor acceptor separations (Ree ꢀ 3 ä) ETtakes place even
in toluene and that a larger distances transfer of singlet
excited state energy may occur when the solvent polarity is
sufficiently low. In more polar media the picture is somewhat
different; ETdominates, regardless of the separation.
Charge transfer emission: Apart from the ZnP and the
occasional C60 fluorescence, the dyads show rather broad
emission in the region located around 800 nm–see Fig-
ure 4.[29] The new band is assigned to CT emission in analogy
3) Dyad 1 has a small interchromophore distance (3.23 ä)
but a poor spatial overlap and reveals the strongest 1*ZnP
fluorescence.
Interestingly, the greater the fluorescence quenching, the
larger is the electronic coupling that was determined from the
ground-state absorption spectra. This indicates that if major
rearrangements occur in the excited state, they are already
present in the spectra.
Not all dyads emit uniquely from ZnP: emission was also
observed from C60 in 3 at ꢀ765 nm with FF ˆ 2.2  10À4 in
toluene, that is, about half of that measured for the C60
reference,[24] while in chloroform, the same fullerene emission
was seen with a lower quantum yield (1.1 Â 10À4). Further
increase of the solvent polarity, for example, use of THF led to
a complete cancellation of the C60 emission through enhance-
ment of the ETprocess. Emission from C60 after excitation of
ZnP implies energy transfer, which was confirmed by a set of
excitation spectra.[26] For a Coulombic energy transfer, the
distance dependence (Ree) is defined by Equation (2):
Figure 4. Charge transfer emission of ZnP C60 (4: solid line) and (2:
dashed line) in toluene at 77 K (ꢀ1.0 Â 10À5 m); excitation wavelength
560 nm.
with the CTabsorption bands discussed earlier (vide supra).
Similar interpretations were reported previously for two
different ZnP C60 systems.[25, 30] At low temperature (77 K)
the emission bands, which are rather broad at room temper-
ature, are resolved, in accordance with the assumption that its
width is determined by (4lSkBT)1/2. Illustrative examples are
shown in Figure 4, with the maximum at 800 nm for 2 and
785 nm for 4. The former is characterized by a p-stacked
system and has a CTabsorption at slightly higher energies.
The energy difference between absorption and emission, or
Stokes shift, is quite small (0.17 eV for 2 and 0.12 eV for 4). In
turn, this gives vibrational reorganization energies (lV) of
0.09 Æ 0.02 and 0.06 Æ 0.01 eV. Support for these remarkably
small lV values also derives from an earlier theoretical
calculation of the vibrational reorganization energy of C60
(lV ˆ 0.06 eV).[31] The p-stacked dyad reveals a notably larger
value, which is indicative of the perturbation due to the
overlapping p-systems. This conclusion is in agreement with
the relatively strong coupling estimated for this donor ac-
ceptor ensemble (vide supra).
The position of the CT emission band (1.58 eV and 1.54 eV)
can be further used to determine the solvent reorganization
energy (lS), since the excitation energy is given by À(lS ‡
DGCR8). Thus, lS values of 0.1 Æ 0.01 eV and 0.02 Æ 0.01 eV
result for 4 and 2, respectively. The different values of lS
suggests that solvation is, therefore, more efficient in the
former than in the latter, inducing stronger solvent interac-
tions with the ZnP and C60 moieties. As noted earlier, in dyad
2, most of the porphyrin surface is covered by the fullerene.
ꢀ
ꢁ
RC
6
k ˆ kD
(2)
Ree
in which kD ˆ 4.0  108 sÀ1 is the intrinsic deactivation of the
1
photoexcited *ZnP, and RC is the critical transfer radius at
which k equalizes the intrinsic deactivation of the donor.[27]
Considering Ree and RC equal 6.18 and 9.78 ä, respectively, a
rapid energy transfer with k ˆ 7.2  109 sÀ1 is estimated; this is
in good agreement with the experimental value (ꢀ1010 sÀ1).
In the two closer packed dyads, namely, 1 and 2, the C60
fluorescence is not observed even in toluene. This implies that
the shorter donor acceptor separation alters the deactivation
1
mechanism of *ZnP, that is, ETtakes over with respect to
energy transfer. Forasmuch as the small reorganization energy
of, for example, 2 in toluene (vide infra), which also infers
negligible activation barriers for ET, the distance dependence
can be formulated in a first approximation within the frame-
work of the Dexter double electron exchange mechanism
[Eq. (3)]:
2
4pꢁV0exp‰À0:5ꢁRee À R0†Š†
k ˆ
J
(3)
ꢀh
in which V is the electronic coupling element and J is the
pertinent spectral overlap of orbitals involved in ETreactions
(5.04 Â 10À4 cmÀ1).[28] This leads to rate constants ꢂ1012 sÀ1.
4972
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4968 4979

Porphyrin C60 Dyads
4968 4979
Charge-separation dynamics: The fate of 1*ZnP and the
identity of the photoproducts in the dyads were examined by
pico-/nano-/microsecond transient absorption spectroscopies.
Representative picosecond time-resolved absorption spec-
tra, taken after an 18 ps laser pulse at 532 nm[32] for a solution
of ZnP in toluene as a reference, are displayed in Figure 5a.
compared with what was seen for the ZnP reference. At early
times (i.e., 50 -100 ps), these are practically identical to those
of the ZnP reference, disclosing strong bleaching at 550 nm
(similar to Figure 5a/b). However, at a delay time of 200 ps, a
new transition around 640 nm grows-in (Figure 6a) accom-
Figure 5. Picosecond transient absorption spectrum recorded 50 ps upon
flash photolysis of ZnP (5.0 Â 10À5 m) at 532 nm in deoxygenated a) toluene
and b) benzonitrile, indicating the ZnP singlet singlet absorption. The
insert to a) illustrates the intersystem crossing process in photoexcited ZnP.
The differential spectrum recorded immediately after the
laser pulse (50 ps time delay) is characterized by bleaching of
the porphyrin Q-band absorption at 550 nm and broad
absorption between 570 and 750 nm. These spectral attributes
Figure 6. Picosecond transient absorption spectrum recorded a) 400 ps and
b) 50 ps upon flash photolysis of 4 and 2 (5.0 Â 10À5 m), respectively at
532 nm in deoxygenated toluene, indicating the ZnP p-radical cation
absorption. Inserts show the decay of meta-1*ZnP C60 (@610 nm) and
are indicative of the ZnP singlet excited state (ESinglet
ˆ
2.00 eV), which decays slowly (4.0 Â 108 sÀ1) to the energeti-
cally lower-lying triplet excited state (ETriplet ˆ 1.53 eV) pre-
dominantly through intersystem crossing (ZnP triplet quan-
tum yield, FTriplet: 0.88). The formation of a penta-coordinate
Zn environment in coordinating solvents, such as THF,
benzonitrile (Figure 5b)[13] and DMF, shifts the Q-band
slightly to the red, to 565 nm, with respect to the non-
coordinating solvents toluene and dichloromethane.
.
.
‡
trans-2-ZnP
C60 À (@630 nm).
panied by another absorption in the near-infrared. Based on a
spectral comparison, we ascribe the former to the zinc
.
‡
porphyrin p-radical cation (ZnP ), while the latter band
.
belongs to the fullerene p-radical anion (C60 À).[24] In accord-
Next, the transient absorption changes of ZnP
C
60 donor
ance with these results, we propose that CS, from the ZnP
acceptor ensembles were recorded in toluene with several
time delays after the 532 nm picosecond laser pulse and
singlet excited state to the electron-accepting fullerene,
.
.
À
‡
creates the ZnP
C60 state, which is responsible for the
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4973

D. M. Guldi, A. Hirsch, F. Zerbetto, M. Prato et al.
FULL PAPER
fast deactivation of the photoexcited chromophore. The
A final comment should address the photophysical proper-
ties of 2. The rate of intramolecular CS is essentially within the
time-resolution of our apparatus (ꢂ20 ps). Instead, the
differential absorption changes, typically noted immediately
after the 20 ps laser pulse, are already composed of the
individual contributions stemming from the oxidized donor
(Figure 6b) and reduced acceptor.
.
.
À
‡
absorption of the charge-separated ZnP
C60 pair is
persistent on the picosecond timescale and decays in the
nano-/microseconds regime (vide infra).
.
.
À
‡
Similar spectral characteristics of ZnP and C60 were
found for all the donor acceptor ensembles in THF, dichloro-
methane, benzonitrile, and DMF.[33] Table 4 indicates that the
Charge recombination dynamics: The charge recombination
(CR) dynamics were analyzed by following the absorption
changes of the one-electron reduced form of the electron
Table 4. Singlet excited state lifetimes of 1*ZnP [in ps] in different ZnP
C
60 donor acceptor ensembles in various solvents (excitation wavelength:
532 nm).
.
acceptor (C60 À) and that of the one-electron oxidized form of
4
3
1
.
‡
the electron donor (ZnP ) (as an example, see Figure 7).
In oxygen-free solutions, the decays were well fitted by a
single exponential expression. Charge separation (CS) and
charge recombination (CR) rates, quantum yields, and
thermodynamic parameters are shown in Tables 5 and Ta-
ble 6; see also the Supporting Information (Section S3) for
[a]
[a]
toluene
chloroform
THF
dichloromethane
benzonitrile
DMF
52
36
35
132
115
130
107
103
97
149
47
55
28
[b]
[b]
35
[a] Energy transfer. [b] Not measured.
.
.
À
‡
Table 6. Lifetimes (t) and quantum yields (F) of the ZnP
C60 radical pair
formed in different ZnP C60 donor-acceptor ensembles (excitation wavelength:
532 nm).
lifetimes of the 1*ZnP substantially decreased with the solvent
polarity and ranged from 149 to 28 ps. It should be noted that
the different functionalization pattern of the fullerene core
leads to different lmax values for the corresponding C60 p-
radical anion fingerprints: 900 nm (2), 1000 nm (1) and
1060 nm (3) (Figure 7). The values are in excellent agreement
2
4
3
1
t [ps] t [ns]
F
t [ms]
F
t [ns]
F
[a]
toluene (e ˆ 2.38)
THF (e ˆ 7.6)
619
385
395
215
195
113
99
0.19
0.28
0.47[b]
0.42
0.36
0.25
0.45[b] 2.6
0.28 236
1.85[c] 0.36 207
dichloromethane (e ˆ 9.08) 121
0.44
0.27
0.17
benzonitrile (e ˆ 24.8)
DMF (e ˆ 36.7)
38
1.1
0.21
0.20 149
0.06 133
[d]
[a] Energy transfer. [b] In chloroform. [c] In o-dichlorobenzene (e ˆ 9.98). [d] Not
measured.
details on their estimation. As far as the thermodynamics are
concerned, the larger the solvent polarity the less negative is
the free energy associated with CR, while the opposite trend
holds for CS.
Inspection of Table 6 reveals that the differences in
.
.
À
‡
electronic coupling affect the lifetimes of the ZnP
C60
radical pair, with actual values ranging from the picosecond
and nanosecond to the microsecond regime. Regardless of the
solvent, the ordering of the lifetimes is: 2 < 4 < 1 < 3. For all
dyads, the higher the solvent polarity, the shorter the lifetime
of the charge-separated state. A decrease of ÀDG8 and faster
CR kinetics is typical of Marcus inverted region, whereby the
ETrates start to decrease with increasing free-energy
change.[34] From the best fits of the driving force dependence
on CR rates [Eq. (4)]:
Figure 7. Nanosecond transient absorption spectrum (visible/near-infrared
part) recorded 50 ns upon flash photolysis of 3 (5.0 Â 10À5 m, dashed line)
and 4 (5.0 Â 10À5 m, solid line) at 532 nm in deoxygenated benzonitrile,
indicating the ZnP p-radical cation and C60 p-radical anion absorption.
with those found in an independent pulse radiolysis study,
which focused on the free radical-induced reduction of a
series of methanofullerene derivatives.[24]
ꢀ
ꢁ
ꢀ
ꢁ
4p3
h2lkBT
ÀDGo
kBT
kET
ˆ
1/2V2exp
(4)
Table 5. Thermodynamic parameters (ÀDGCo S, ÀDGCo R† for different ZnP C60 donor-acceptor ensembles.
2
4
3
1
À DGCo S [eV] À DGoCR [eV] À DGCo S [eV] À DGCo R [eV] À DGoCS [eV] À DGCo R [eV] À DGCo S [eV] À DGCo R [eV]
toluene (e ˆ 2.38)
THF (e ˆ 7.6)
0.44
0.54
0.55
0.57
1.56
1.46
1.45
1.43
0.32
0.78
0.82
0.94
0.96
1.68
1.22
1.18
1.06
1.04
0.46
0.56
0.57
0.59
0.60
1.54
1.44
1.43
1.41
1.40
0.61
0.67
0.79
0.82
1.39
1.33
1.21
1.18
dichloromethane (e ˆ 9.08)
benzonitrile (e ˆ 24.8)
DMF (e ˆ 36.7)
4974
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4968 4979

Porphyrin C60 Dyads
4968 4979
the following values for the reorganization energies (l) and
the electronic couplings (V) were derived: l ˆ 0.71 eV and
Vˆ 313 cmÀ1 for 2; l ˆ 0.54 eV and Vˆ 10 cmÀ1 for 3.[35]
Interestingly, the trend seen for V derived from the fit of
ETrates further substantiates the values calculated based on
the CTabsorption, although the absolute magnitude is slightly
different.
Finally, the close p-stacking in 2 seems to lead to a
significant perturbation of the overlapping p-systems. This is,
for example, expressed by the somewhat larger reorganization
energy (l) found for 2 relative to 3 and 4. Note that the
energetic difference between the CTabsorption and CT
emission (vide supra), as an independent measure for l, led to
the same conclusion.
Dynamical modeling of the electronic coupling: Molecular
dynamics trajectories for the two limiting cases of 2 and 3 were
calculated in the gas phase and in three different solvents (i.e.,
toluene, dichloromethane, and benzonitrile). The solvent was
treated explicitly to allow for the possibility of a molecule
inserting between the acceptor and the donor.[36] The calcu-
lations were run with the TINKER package[37] with the MM3
force field,[38] which has found a wide range of applications in
one of our laboratories.[39] The force field was corrected to
.
.
À
‡
account for the charge distribution on the ZnP
C60
Figure 8. Molecular dynamics simulation of the electronic coupling matrix
element: a) gas phase, ground state; b) gas phase, CTstate; c) in toluene,
ground state; d) in toluene, CTstate; e) in CH 2Cl2, ground state; f) in
CH2Cl2, CTstate; g) in benzonitrile, ground state; h) in benzonitrile, CT
state; solid line 3, dashed line 2.
species by adding point charges, which were calculated as
the difference between the Mulliken charges of the CTstate
and the ground state (see also below). The simulation box,
with periodic boundary conditions, gave a minimum distance
between solute molecules of 16 ä; the cutoff for the
van der Waals interactions was set to 8 ä. All solvents were
added using room temperature densities. The solvents were
considered explicitly to include the effect of specific solvation.
The molecular dynamnics (MD) calculations were run at a
constant temperature of 300 K, within the canonical ensem-
ble, using the Berendsen algorithm.[40] Hydrogen atoms were
constrained to their ideal bond length by using the RATTLE
algorithm[41] and the time step was 1 fs. After an initial
equilibration of 30 ps, snapshots for the evaluation of the
electronic coupling matrix element were taken every 0.1 ps.
The coupling V was calculated, with the INDO/S plus
configuration interaction singles model,[42] for the CR process.
The CT state is mainly described as the one-electron
excitation from the HOMO of ZnP and the LUMO of C60.
In this simple picture, the coupling between the ground state
Table 7. Calculated electronic coupling (V) in cmÀ1
.
2
3
ground state
CTstate
ground state
CTstate
gas phase
toluene
dichloromethane
benzonitrile
339 Æ 20
257 Æ 17
280 Æ 21
314 Æ 22
358 Æ 24
364 Æ 21
360 Æ 26
387 Æ 22
37.6 Æ 3.6
34.5 Æ 2.8
36.0 Æ 4.0
41.2 Æ 3.7
54.5 Æ 4.4
36.7 Æ 3.8
25.7 Æ 2.8
37.3 Æ 3.5
values of V together with the calculated standard deviation
calculated for the dynamics of the dyad in the ground and with
the charges of its CTstate.
The couplings depend weakly on the solvent, a finding that
justifies this assumption in the fitting procedure of the
experimental data. For 2, they range from 257 cmÀ1 in the
ground state in toluene to 387 cmÀ1 in the CTstate in
benzonitrile, while for 3 they vary from 34 cmÀ1 in the ground
state in toluene to 54 cmÀ1 for the CTstate in the gas phase.
These values can be compared with 436 cmÀ1 obtained from
the spectral analysis of 2 for which the fitting of the CR data
gave 313 cmÀ1. Also a value of 76 cmÀ1 was obtained from the
spectral analysis of 3, while the fitting of the CR data gave
10 cmÀ1.
The calculations therefore confirm the order of magnitude
of difference for V in the two dyads. The larger couplings
found in the molecular dynamics simulations that used the
charges of the CTstate are due to the presence of increased
Coulomb forces, which bring the two moieties closer. Ideally,
[43]
and the CTstate can be formulated as Equation (5):
Vˆ hf0HOMOꢁZnP† j F 0 j fL0 UMOꢁC60†
i
(5)
in which F is the Fock operator and the suffix zero indicates
that the orbitals are unperturbed and localized either on ZnP
or on C60. The approach does not include through-bond
coupling, which is less important and roughly similar for the
two dyads. Illustrative plots of the coupling V versus time are
given in Figure 8, whereby V undergoes large fluctuations on
the timescale of CTreactions. In hindsight, the fluctuations
seen in Figure 8 further assert the need to carry out dynamic
simulation. Importantly, the MD was run until the time
average value of V achieved convergence. Table 7 gives the
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4975

D. M. Guldi, A. Hirsch, F. Zerbetto, M. Prato et al.
FULL PAPER
the coupling in the Marcus treatment should be calculated at
the crossing of the potential-energy surfaces. In the present
case, an average of the values for the two states should be a
rather good approximation to such a value.
The reorganization energy (l) was also estimated with an
™instantaneous∫ approach. At each of the snapshots of the
dynamics the partial atomic charges were switched from those
of one state to those of the other and the energy was
recalculated–see Figure 9. In the dynamics of the ground
spectral analysis of 2, while the fitting of the CR data in
several solvents gave 0.74 eVand b) 0.54 eV obtained from
the fitting of the CR data of 3. The disagreement between
the two analysis of the experimental data is solved in favor
of the results of the fitting of the CR kinetics by the
molecular dynamics calculations.
3) Dichloromethane has the largest l value (0.77 eV for 2 and
0.61 eV for 3). This is in keeping with the notion that the
phenyl rings of toluene and benzonitrile match the shapes
of C60 and ZnP by forming p-stack complexes. Atomic
charge variations affect such stacks only mildly. On the
contrary, the shape of dichloromethane does not fit with
that of the two charged moieties, and even small charge
variation may cause a large rotational reorganization of
this solvent.
4) The effect of the large reorganization energy in dichloro-
methane finds experimental confirmation. Table 5 shows
that the free energy for CS of 3 in this solvent from the
excited state to the charge-separated state is 0.67 eV, which
is very similar to the calculated reorganization energy,
0.61 eV. In turn, this implies that the potential-energy
curve of the CTstate crosses the excited state near its
minimum. The resulting effect is very fast conversion of
the excited state to the CTstate. Table 6 shows that upon
excitation the yields of the charge-separated state of 3 in
dichloromethane is the largest of the three solvents
studied.
5) The similarity of the reorganization energy for 2 and 3
makes the large difference in electronic coupling even
more relevant to the issue of this work, and supports the
proposal that ZnP C60 systems are excellent choices to
study the effect of topological variations on ETrates.
Figure 9. Molecular dynamics simulation of the reorganization energy (l):
Conclusion
a) in toluene, ground state; b) in toluene, CTstate; c) in CH Cl2, ground
2
state; d) in CH2Cl2, CTstate; e) in benzonitrile, ground state; f) in
benzonitrile, CTstate; solid line 3, dashed line 2.
In conclusion, the photo-induced ETin four different ZnP
C60 dyads has been studied. The distance and orientation
between electron acceptor and electron donor varies sub-
stantially, in a consistent way, as a function of the mechanical
link. This feature makes them a good case for verifying the
generally accepted rules on the distance dependence of ET.
For the first time, spectral analysis was carried out compre-
hensively on 1) the band shifts of the ZnP fragment, 2) the
intensity of newly formed CTbands, and 3) their spectral
shifts upon variation of the solvent polarity. All the results
agree quantitatively and show that the strength of interaction
between ZnP and C60 varies more than an order of magnitude
and scales with the ZnP C60 distance (i.e., trans-2-ZnP C60
(2) > meta-ZnP C60 (4) > equatorial-ZnP C60 (3) > para-
ZnP C60 (1)). Also the quantum yields of formation of
1*ZnP were found to be a function of the same distance.
The different strength in electronic coupling affects sensi-
tively both CS and CR kinetics–no change in mechanism
state, the sudden appearance of the charges of the CTstate
increases the energy (the same happens in the dynamics of the
CTstate when the atomic charges used are those of the
ground state). Time averaging the energy differences over the
dynamics gives the results, listed in Table 8.
Table 8. Estimated reorganization energy (l) in eV.
2
3
toluene
dichloromethane
benzonitrile
0.26 Æ 0.01
0.77 Æ 0.02
0.43 Æ 0.01
0.34 Æ 0.01
0.61 Æ 0.02
0.38 Æ 0.01
Several points are worth noticing:
1) l is solvent dependent, but does not vary too markedly.
This justifies the assumptions in the fitting of the exper-
imental data, at least as a first approximation.
2) For 2, l ranges from 0.26 to 0.77 eV, while for 3 l varies
from 0.34 to 0.61 eV. These values can be compared with
.
.
À
‡
occurs. Most importantly, the lifetime of the ZnP
C60
radical pair, product of an intramolecular CS event evolving
1
1
from *ZnP (i.e., 532 nm excitation) or *C60 (i.e., 355 nm
excitation), exhibits values that range from the picosecond
and nanosecond to the microsecond regime.
a) lV
‡ lS ˆ 0.09‡0.02 ˆ 0.11 eV obtained from the
4976
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4968 4979

Porphyrin C60 Dyads
4968 4979
119.48, 118.73 (meso-C), 114.20 (CHaromat), 65.86, 63.85 (CH2), 52.54
Apart from the spectral analysis, the interactions between
ZnP and C60 were quantitatively evaluated in two other ways:
by fitting the CR rates and by molecular dynamics calcula-
tions. The three approaches gave a rather consistent picture.
For instance, molecular dynamics calculations for 2 gave
257 ˆ Vˆ 387 cmÀ1 as compared with 313 ˆ Vˆ 436 cmÀ1
obtained by fitting of the experimental data, while for 3 they
gave 34 ˆ Vˆ 54 cmÀ1 as compared with 10 ˆ Vˆ 76 cmÀ1
obtained by fitting of the experimental data. Similar calcu-
lations gave for 2 0.26 ˆ l ˆ 0.77 eV as compared with 0.11 ˆ
l ˆ 0.74 eV obtained by fitting of the experimental data, while
for 3 they gave 0.34 ˆ l ˆ 0.61 eV as compared with 0.54 eV
from fitting of experimental data.
À
À
(OCH3), 41.17 ppm (CO CH2 CO); UV/Vis (CH2Cl2): lmax (loge) ˆ 399
(sh, 4.87), 418 (5.65), 514 (4.21), 549 (3.77), 591 (3.63), 645 nm (3.42);
FT-IR (KBr): nÄ ˆ 3316, 3054, 3024, 2950, 1736, 1596, 1575, 1471, 1440,
1349, 1178, 1152, 964, 801, 730, 700 cmÀ1; MS (MALDI-TOF): m/z: 775
‡
[M‡H] .
Synthesis of 5-{3-[2-(methoxymalonyloxy)ethoxy]phenyl}-10,15,20-triphe-
nylporphyrinatozinc(ii) (5): A solution of 7 (250 mg, 0.32 mmol) and
Zn(OAc)2 ¥ 2H2O (50 mg, 0.37 mmol)in THF (50 mL) was refluxed for one
hour. The product was separated by flash chromatography (SiO2, CH2Cl2/
EtOAc 20:1). Yield: 260 mg (0.31 mmol), 97% referred to 7. 1H NMR
(400 MHz, CDCl3): d ˆ 8.94 (m, 8H; b-CH), 8.21 (AA'BB', 6H;
ArH),7.85(d, 1H; ArH), 7.74 (m, 10H; ArH), 7.63 (t, 1H; ArH), 7.28 (d,
1H; ArH), 4.51 (t, 3J ˆ 4.64 Hz, 2H; CH2), 4.32 (t, 3J ˆ 4.76 Hz, 2H; CH2),
3.65 (s, 3H; OCH3), 3.36 (s, 2H; CO CH2 CO); 13C NMR (100.5 MHz,
À
À
ˆ
CDCl3): d ˆ 166.71, 166.45 (C O), 156.66 (Caromat), 150.26, 150.03 (a-C),
144.31, 142.81 (Caromat), 134.42 (CHaromat), 132.03, 131.99, 131.86 (b-CH),
128.50, 127.46, 126.55 (CHaromat), 121.18 (meso-C), 120.98 (CHaromat), 120.48
(meso-C), 114.11 (CHaromat), 65.86, 63.83 (CH2), 52.50 (OCH3), 41.08
Experimental Section
À
À
(CO CH2 CO); UV/Vis (CH2Cl2): lmax (loge) ˆ 399 (sh, 4.54), 417 (5.68),
‡
548 (4.24), 587 nm (3.47); MS (FAB): m/z: 838 [M‡H] .
General: Picosecond laser-flash photolysis experiments were carried out
with 532 nm laser pulses from a mode-locked, Q-switched Quantel YG-501
DP Nd:YAG laser system (18 ps pulse width, 2 3 mJ per pulse). Nano-
second laser-flash photolysis experiments were performed with laser pulses
from a Quanta-Ray CDR Nd:YAG system (532 nm, 6 ns pulse width) in a
front face excitation geometry.
Synthesis of dyad 4: A solution of the porphyrin malonate 5 (200 mg,
0.24 mmol) was stirred together with C60 (186 mg, 0.26 equiv) at RTfor 24 h
in toluene (150 mL) in the presence of 1,8-diazabicyclo[5,4,0]undec-7-en
(DBU; 100 mL, 0.67 mmol) and I2 (61 mg, 0.24 mmol). Chromatographic
separation was performed on silica gel with toluene/ethylacetat (20:1) as
eluent. Yield 134 mg, 36% referred to 5. 1H NMR (400 MHz, CDCl3): d ˆ
8.81 (m, 4H; b-CH), 8.77 (m, 4H; b-CH), 8.19 (AA'BB', 1H; ArH), 8.12
(AA'BB', 1H; ArH), 8.08 (AA'BB', 1H; ArH), 8.01 (AA'BB', 2H; ArH),
7.80 (d, 1H; ArH), 7.68 (m, 10H; ArH), 7.40 (m, 1H; ArH), 7.25 (d, 1H;
ArH), 4.78 (t, 3J ˆ 4.76 Hz, 2H; CH2), 4.41 (t, 3J ˆ 4.76 Hz, 2H; CH2),
3.93 ppm (s, 3H; OCH3); 13C NMR (100.5 MHz, CDCl3): d ˆ 162.42, 162.03
Fluorescence lifetimes were measured with a laser strope fluorescence
lifetime spectrometer (Photon Technology International) with 337 nm laser
pulses from a nitrogen laser fiber coupled to a lens-based T-formal sample
compartment equipped with a stroboscopic detector. Details of the Laser
Strobe systems are described on the manufacture×s web site (http://
www.pti-nj.com).
ˆ
(C O), 156.09 (Caromat), 149.78, 149.53 (a-C), 144.43, 144.30, 144.16, 144.04,
Emission spectra were recorded with a SLM 8100 spectrofluorometer. The
experiments were performed at room temperature. Each spectrum
represents an average of at least 5 individual scans, and appropriate
corrections were applied whenever necessary.
144.01, 143.88, 143.83, 143.81, 143.78, 143.74, 143.71, 143.60, 143.36, 143.16,
143.05, 142.90, 142.46, 142.39, 141.86, 141.82, 141.55, 141.50, 141.28, 140.98,
140.79, 140.48, 140.40, 139.61, 139.57, 138.81, 137.47 (C60 sp2-C/Caromat),
134.18, 134.14 (CHaromat), 131.82, 131.78, 131.75, 131.70 (b-CH), 129.68,
127.85, 127.47, 127.27, 127.18, 127.14, 126.26 (CHaromat), 122.05 (meso-C),
120.92 (CHaromat), 120.81, 119.89 (meso-C), 113.83 (CHaromat), 70.54 (C60 sp3-
C), 66.18, 64.57 (CH2), 53.09 (OCH3), 51.11 ppm (cyclopropan-C); UV/Vis
(CH2Cl2): lmax (loge) ˆ 257 (5.15), 328 (4.73), 401 (sh, 4.60), 427 (5.44), 551
Synthesis of 3-[2-(methoxymalonyloxy)ethoxy]benzaldehyde (6): A solu-
tion of 3-[2-hydroxyethoxy]benzaldehyde (12.0 g, 0.072 mol) and pyridine
(6.5 mL, 0.081 mol) in dichloromethane (100 mL) was cooled down to 08C.
Subsequently methyl malonyl chloride (8.0 mL, 0.084 mol) was added and
the reaction mixture was stirred for 1 hour at 08C and another hour at RT.
The precipitate was dissolved in water and the organic phase was extracted
twice with 0.1n HCl, once with NaHCO3 and dried over MgSO4. After
evaporation of the solvent a yellow oil was obtained. Yield: 15.98 g
(0.060 mol), 83% referred to 3-[2-hydroxyethoxy]benzaldehyde. FT-IR
(film): nÄ ˆ 3068, 3004, 2956, 2845, 1738, 1447, 1386, 1262, 790, 684 cmÀ1; MS
‡
(4.29), 588 nm (3.84); MS (FAB): m/z: 1557 [M‡H] ; FT-IR (KBr): nÄ ˆ
3049, 3022, 2988, 2923, 1748, 1596, 1438, 1339, 1232, 1003, 796, 700,
526 cmÀ1
.
The correct structural assignment of the trans-2 addition pattern carried out
in reference [18b] was further corroborated: 1) The dyad was hydrolyzed
according to reference [44] to give the corresponding bis-malonic acid
‡
(EI): m/z (%): 266 (10) [M] , 166 (35), 145 (90), 121 (100).
derivative. According to the same procedure the known bis-adducts
Synthesis of 5-{3-[2-(methoxymalonyloxy)ethoxy]phenyl}-10,15,20-triphe-
[18f]
trans-2-C62(COOEt)4 and trans-3-C62(COOEt)4
were hydrolyzed to
nylporphyrin 7:
A solution of 6 (6.65 g, 25.0 mmol), benzaldehyde
give the bis-malonic acid derivatives trans-2-C62(COOH)4 and trans-3-
C62(COOH)4. The comparison with the corresponding trans-3 derivative
was carried out because this is the only other possible C2-symmetric
trans addition pattern and the C2-symmetry within the dyad is unambig-
uously proven. After hydrolysis of the Zn dyad trans-2-C62(COOH)4
and not trans-3-C62(COOH)4 was formed as demonstrated by the
dentity of the HPLC retention times: k' (HPLC, SiO2, methanol, flow
rate 1.4 mLminÀ1) ˆ 1.47 and the electronic absorption spectra (e.g.,
characteristic trans-2 absorption at 435 nm). 2) Molecular Modeling: To
account for the flexibility of the alkyl chains within the free base model
dyads with trans-2 and trans-3 addition patterns, MD simulations at 300 K
for a simulation time of 10 ps (time step 0.001 ps) have been performed. For
each adduct five independent simulations were carried out and the
corresponding equilibrium structures were used as starting geometries for
(2.53 mL, 25.0 mmol), pyrrole (3.50 mL, 50.0 mmol) and Ph4PCl (58 mg,
0.15 mmol) in dichloromethane (500 mL) was stirred for 10 min under a
nitrogen atmosphere before BF3 ¥ Et2O (628 mL, 5.0 mmol) were added.
After one hour para-chloranil (9.25 g, 37.6 mmol) was added under light
protection and the reaction mixture was refluxed for another hour. The
solvent was evaporated and the mono-functionalized porphyrin was
separated by flash chromatography as the second fraction (SiO2, CH2Cl2/
EtOAc 20:1). Yield: 527 mg (0.68 mmol), 5.4% referred to pyrrole.
1H NMR (400 MHz, CDCl3): d ˆ 8.84 (m, 8H; b-CH), 8.21 (AA'BB',
6H; ArH), 7.85 ppm (d, 1H; ArH); 1H NMR (400 MHz, CDCl3): d ˆ 9.98
(s, 1H; CHO), 7.48 (m, 2H; ArH), 7.40 (s, 1H; ArH), 7.21 (m, 1H; ArH),
4.54 (t, 3J ˆ 4.64 Hz, 2H; CH2), 4.26 (t, 3J ˆ 4.76 Hz, 2H; CH2), 3.73 (s, 3H;
OCH3), 3.45 ppm (s, 2H; CO CH2 CO); 13C NMR (100.5 MHz, CDCl3):
À
À
ˆ
d ˆ 191.84 (CHO), 166.65, 166.34 (C O), 158.90, 137.78 (Caromat), 130.15,
‡
energy minimization at 0 K optimization using the MM force field. The
123.94, 121.89, 112.77 (CHaromat), 65.82, 63.44 (CH2), 52.48 (OCH3), 41.05
(1H; ArH), [7.78 (s, 1H; ArH), 7.75 (m, 9H; ArH), 7.63 (t, 1H; ArH),
7.30 (d, 1H; ArH), 4.57 (t, 3J ˆ 4.64 Hz, 2H; CH2), 4.36 (t, 3J ˆ 4.64 Hz, 2H;
free bases were chosen as models because parameters for zinc are not
‡
unavailable within this force field, MM . The averaged energies computed
thereafter were used to compare the stabilities of the trans-2 and trans-3
derivatives. The energy data clearly show that trans-2 isomer (HOF ˆ
À
À
CH2), 3.68 (s, 3H; OCH3), 3.44 (s, 2H; CO CH2 CO), À2.57 ppm
(brs, 2H; NH)]; 13C NMR (100.5 MHz, CDCl3): d ˆ 166.76, 166.50
668.6 kcalmÀ1
)
is much more stable than trans-3 isomer (HOF ˆ
ˆ
(C O), 156.77, 143.63, 143.06, 142.13 (Caromat), 134.55 (CHaromat), 131.15
(b-CH), 128.17, 127.72, 127.61, 126.68, 121.12 (CHaromat), 120.28,
861.9 kcalmÀ1).
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4977

D. M. Guldi, A. Hirsch, F. Zerbetto, M. Prato et al.
FULL PAPER
M. G. Ranasinghe, M. J. Shephard, M. N. Paddon-Row, Chem. Phys.
Lett. 1997, 268, 223.
Acknowledgement
[12] Leading example: a) F.-P. Montforts, O. Kutzki, Angew. Chem. 2000,
112, 612; Angew. Chem. Int. Ed. 2000, 39, 599.
[13] Leading examples: a) F. D×Souza, G. R. Deviprasad, M. S. Rahman, J.-
P. Choi, Inorg. Chem. 1999, 38, 2157; b) F. D×Souza, G. R. Deviprasad,
M. E. El-Khouly, M. Fujitsuka, O. Ito, J. Am. Chem. Soc. 2001, 123,
5277.
This work was supported by the Stiftung Volkswagenwerk, the European
Community (HPRNT-CT-2002 00177 program WONDERFULL), SFB
583 (Redoxaktive Metallkomplexe Reaktivit‰tssteuerung durch moleku-
lare Architekturen), the Office of Basic Energy Sciences of the US
Department of Energy and MIUR (cofin 2002, prot. 2002032171). This is
contribution NDRL-4477 of the Radiation Laboratory.
[14] L. Echegoyen, L. E. Echegoyen, Acc. Chem. Res. 1998, 31, 593.
[15] a) M. Prato, M. Maggini, Acc. Chem. Res. 1998, 31, 519; b) F.
Diederich, R. Kessinger, Acc. Chem. Res. 1999, 32, 537; c) ™Fullerenes
and Related Structures∫: Top. Curr. Chem. 1999, 199, whole volume;
d) Lecture Notes on Fullerene Chemistry, (Ed.: R. Taylor), Imperial
College Press, London, 1999; e) S. R. Wilson, D. I. Schuster, B. Nuber,
M. S. Meier, M. Maggini, M. Prato, R. Taylor, in Fullerene, Chemistry,
Physics and Technology (Eds.: K. M. Kadish, R. S. Ruoff), Wiley, New
York, 2000.
[16] a) H. Imahori, K. Hagiwara, T. Akiyama, M. Aoki, S. Taniguchi, T.
Okada, M. Shirakawa, Y. Sakata, Chem. Phys. Lett. 1996, 263, 545;
b) D. M. Guldi, K.-D. Asmus, J. Am. Chem. Soc. 1997, 119, 5744;
c) D. M. Guldi, M. Prato, Acc. Chem. Res. 2000, 33, 695.
[17] For orientation effects: a) J. L. Sessler, M. R. Johnson, T.-Y. Lin, S. E.
Creager, J. Am. Chem. Soc. 1988, 110, 3659; b) Y. Sakata, S.
Nakashima, Y. Goto, T. Asahi, M. Hagihara, S. Nishikawa, T. Okada,
N. Mataga, J. Am. Chem. Soc. 1989, 111, 8979; c) A. M. Brun, A.
Harriman, V. Heitz, J.-P. Sauvage, J. Am. Chem. Soc. 1991, 113, 8657;
d) A. Helms, D. Heiler, G. McLendon, J. Am. Chem. Soc. 1991, 113,
4325; e) H. A. Staab, S. Nikolic, C. Krieger, Eur. J. Org. Chem. 1999,
1459; f) I. Read, A. Napper, R. Kaplan, M. B. Zimmt, D. H. Waldeck,
J. Am. Chem. Soc. 1999, 121, 10976; g) R. W. Kaplan, A. M. Napper,
D. H. Waldeck, M. B. Zimmt, J. Am. Chem. Soc. 2000, 122, 12039;
h) N. R. Lokan, M. N. Paddon-Row, M. Koeberg, J. W. Verhoeven, J.
Am. Chem. Soc. 2000, 122, 5075; i) T. D. M. Bell, K. A. Jolliffe, K. P.
Ghiggino, A. M. Oliver, M. J. Shephard, S. J. Langford, M. N. Paddon-
Row, J. Am. Chem. Soc. 2000, 122, 10661; j) A. Ambroise, R. W.
Wagner, P. D. Rao, J. A. Riggs, P. Hascoat, J. R. Diers, J. Seth, R. K.
Lammi, D. F. Bocian, D. Holten, J. S. Lindsey, Chem. Mater. 2001, 13,
1023; k) M. Goes, M. de Groot, M. Koeberg, J. W. Verhoeven, N. R.
Lokan, M. J. Shephard, M. N. Paddon-Row, J. Phys. Chem. A 2002,
106, 2129.
[1] a) F. L. Carter, Molecular Electronic Devices, Dekker, New York,
1987; b) Photoinduced Electron Transfer (Eds.: M. A.Fox, M. Chan-
non), Elsevier, Amsterdam, 1988; c) Electron Transfer in Chemistry
(Ed.: V. Balzani), Wiley-VCH, Weinheim, 2001.
[2] a) M. D. Newton, Chem. Rev. 1991, 91, 767; b) M. R. Wasielewski,
Chem. Rev. 1992, 92, 435; c) D. Gust, T. A. Moore, A. L. Moore, Acc.
Chem. Res. 1993, 26, 198; d) M. N. Paddon-Row, Acc. Chem. Res.
1994, 27, 18; e) I. R. Gould, S. Farid, Acc. Chem. Res. 1996, 29, 522;
f) V. Balzani, A. Juris, M. Venturi, S. Campagna, S. Serroni, Chem.
Rev. 1996, 96, 759; g) I. Willner, Acc. Chem. Res. 1997, 30, 347; h) P.
Piotrowiak, Chem. Soc. Rev. 1999, 28, 143; i) H. Kurreck, M. Huber,
Angew. Chem. 1995, 107, 929; Angew. Chem. Int. Ed. Engl. 1995, 34,
849.
[3] Leading examples of noncovalent donor acceptor assemblies: a) P.
Tecilla, R. P. Dixon, G. Slobodkin, D. S. Alavi, D. H. Waldeck, A. D.
Hamilton, J. Am. Chem. Soc. 1990, 112, 9408; b) P. J. F. de Rege, S. A.
Williams, M. J. Therien, Science 1995, 269, 1409; c) J. P. Kirby, J. A.
Roberts, D. G. Nocera, J. Am. Chem. Soc. 1997, 119, 9230; d) S. L.
Springs, D. Gosztola, M. R. Wasielewski, V. Kral, A. Andrievsky, J. L.
Sessler, J. Am. Chem. Soc. 1999, 121, 2281; e) K. Yamada, I. Imahori,
E. Yoshizawa, D. Gosztola, M. R. Wasielewski, Y. Sakata, Chem. Lett.
1999, 235; f) M.-J. Blanco, M. C. Jimenez, J.-C. Chambron, V. Heitz,
M. Linke, J.-P. Sauvage, Chem. Soc. Rev. 1999, 28, 293; g) J. L. Sessler,
B. Wang, S. L. Springs, C. T. Brown, Compr. Supramol. Chem. Oxford,
1996, 4, 311.
[4] The Photosynthetic Reaction Center (Eds.: J. Deisenhofer, J. R.
Norris), Academic Press, San Diego, 1993.
[5] a) H. Imahori, Y. Sakata, Adv. Mater. 1997, 9, 537; b) M. Prato, J.
¬
[18] a) E. Dietel, A. Hirsch, J. Zhou, A. Rieker, J. Chem. Soc. Perkin Trans.
2 1998, 1357; b) E. Dietel, A. Hirsch, E. Eichborn, A. Rieker, S.
Hackbarth, B. Rˆder, Chem. Commun. 1998, 1981. c) D. M. Guldi, C.
Luo, M. Prato, E. Dietel, A. Hirsch, Chem. Commun. 2000, 373;
d) D. M. Guldi, C. Luo, M. Prato, A. Troisi, F. Zerbetto, M. Scheloske,
E. Dietel, W. Bauer, A. Hirsch, J. Am. Chem. Soc. 2001, 123, 9166;
e) X. Camps, E. Dietel, A. Hirsch, S. Pyo, L. Echegoyen, S. Hackbarth,
B. Rˆder, Chem. Eur. J. 1999, 5, 2362; f) A. Hirsch, I. Lamparth, H. R.
Karfunkel, Angew. Chem. 1994, 106, 453; Angew. Chem. Int. Ed. Engl.
1994, 33, 437; g) A. Hirsch, I. Lamparth, T. Groesser, H. R. Karfunkel,
J. Am. Chem. Soc. 1994, 116, 9385; h) A. Herzog, A. Hirsch, O.
Vostrowsky, Eur. J. Org. Chem. 2000, 171; i) A. Hirsch, O. Vostrowsky,
Eur. J. Org. Chem. 2001, 829.
Mater. Chem. 1997, 7, 1097. c) N. MartÌn, L. Sanchez, B. Illescas, I.
¬
Perez, Chem. Rev. 1998, 98, 2527; d) F. Diederich, M. Gomez-Lopez,
Chem. Soc. Rev. 1999, 28, 263; e) H. Imahori, Y. Sakata, Eur. J. Org.
Chem. 1999, 2445. f) D. M. Guldi, Chem. Commun. 2000, 321; f) D.
Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2001, 34, 40.
[6] Leading examples: a) H. Imahori, K. Hagiwara, M. Aoki, T. Akiyama,
S. Taniguchi, T. Okada, M. Shirakawa, Y. Sakata, J. Am. Chem. Soc.
1996, 118, 11771; b) S. Fukuzumi, K. Ohkubo, H. Imahori, J. Shao, Z.
Ou, G. Zheng, Y. Chen, R. K. Pandey, M. Fujitsuka, O. Ito, K. M.
Kadish, J. Am. Chem. Soc. 2001, 123, 10676.
[7] Leading examples: a) P. A. Liddell, D. Kuciauskas, J. P. Sumida, B.
Nash, D. Nguyen, A. L. Moore, T. A. Moore, D. Gust, J. Am. Chem.
Soc. 1997, 119, 1400; b) D. Kuciauskas, P. A. Liddell, S. Lin, T. E.
Johnson, S. J. Weghorn, J. S. Lindsey, A. L. Moore, T. A. Moore, D.
Gust, J. Am. Chem. Soc. 1999, 121, 8604.
[19] J. S. Lindsey, I. C. Schreiman, H. C. Hsu, P. C. Kearney, A. M.
Marguerettaz, J. Org. Chem. 1987, 52, 827.
[20] For the synthesis and photophysical investigations of similiar dyads
inlcuding those with a trans-1 addition pattern see: J.-P. Bourgois, F.
Diederich, L. Echegoyen, J.-F. Nierengarten, Helv. Chim. Acta 1998,
81, 1835.
[8] Leading examples: a) N. Armaroli, F. Diederich, C. O. Dietrich-
Buchecker, L. Flamigni, G. Marconi, J.-F. Nierengarten, J.-P. Sauvage,
Chem. Eur. J. 1998, 4, 406; b) J.-F. Nierengarten, C. Schall, J.-F.
Nicoud, Angew. Chem. 1998, 110, 2037; Angew. Chem. Int. Ed. 1998,
37, 1934.
[9] Leading examples: a) P. S. Baran, R. R. Monaco, A. U. Khan, D. I.
Schuster, S. R. Wilson, J. Am. Chem. Soc. 1997, 119, 8363; b) D. I.
Schuster, P. Cheng, S. R. Wilson, V. Prokhorenko, M. Katterle, A. R.
Holzwarth, S. E. Braslavsky, G. Klihm, R. M. Williams, C. Luo, J. Am.
Chem. Soc. 1999, 121, 11599
[10] Leading examples: a) N. V. Tkachenko, L. Rantala, A. Y. Tauber, J.
Helaja, P. H. Hynninen, H. Lemmetyinen, J. Am. Chem. Soc. 1999,
121, 9378; b) N. V. Tkachenko, E. Vuorimaa, T. Kesti, A. S. Alekseev,
A. Y. Tauber,; P. H. Hynninen, H. Lemmetyinen, J. Phys. Chem. B
2000, 104, 6371.
[21] For leading references on tether directed bisfunctionalization of C60
see: a) J.-F. Nierengarten, T. Habicher, R. Kessinger, F. Cardullo, F.
Diederich, V. Gramlich, J. P. Gisselbrecht, C. Boudon, M. Gross, Helv.
Chim. Acta 1997, 80, 2238; b) U. Reuther, T. Brandm¸ller, W.
Donaubauer, F. Hampel, A. Hirsch, Chem. Eur. J. 2002, 8, 2261.
[22] HYPERCHEM, Hypercubes, http://www.hyper.com; Heating time:
10 ps, run time 15 ps, cooling time: 50 ps; starting temperature 08K,
simulation temperature: 12008K, final temperature: 08K. For refer-
ences see: a) W. F. van Gusteren, H. J. C. Berendsen, Angew. Chem.
1990, 102, 1020; Angew. Chem. Int. Ed. Engl. 1990, 29, 992; b) G. H.
Grant, W. G. Richards, Computional Chemistry, Oxford University
Press 1955, p. 51; c) J. Helaja, A. Y. Tauber, Y. Abel, N. V. Tkachenko,
H. Lemmetyinen, I. Kipelainen, P. H. Hynninen, J. Chem. Soc. Perkin
[11] Leading examples: a) M. J. Shephard, M. N. Paddon-Row, Aust. J.
Chem. 1996, 49, 395; b) T. D. M. Bell, T. A. Smith, K. P. Ghiggino,
4978
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4968 4979

Porphyrin C60 Dyads
4968 4979
Trans. 1 1999, 16, 2403; d) P. Bonnet, C. Jaime, L. Morin-Allory, J. Org.
Chem. 2001, 66, 689.
[31] S. Larsson, A. Klimkans, L. Rodriguez-Monge, G. Duskesas, J. Mol.
Struct. 1998, 425, 155.
[23] a) M. M. Olmstead, D. A. Costa, K. Maitra, B. C. Noll, S. L. Phillips,
P. M. van Calcar, A. L. Balch, J. Am. Chem. Soc. 1999, 121, 7090;
b) P. D. W. Boyd, M. C. Hodgson, C. E. F. Rickard, A. G. Oliver, L.
Chaker, P. J. Brothers, R. D. Bolskar, F. S. Tham, C. A. Reed, J. Am.
Chem. Soc. 1999, 121, 10487; c) D. V. Konarev, I. S. Neretin, Y. L.
Slovokhotov, E. I. Yudanova, N. V. Drichko, Y. M. Shul'ga, B. P.
Tarasov, L. L. Gumanov, A. S. Batsanov, J. A. K. Howard, R. N.
Lyubovskaya, Chem. Eur. J. 2001, 7, 2605.
[24] D. M. Guldi, K.-D. Asmus, J. Phys. Chem. A 1997, 101, 1472.
[25] N. Armaroli, G. Marconi, L. Echegoyen, J.-P.Bourgeois, F. Diederich,
Chem. Eur. J. 2000, 6, 1629.
[26] The excitation spectrum of the C60 emission is an exact match of the
ZnP excitation spectrum and the ZnP ground state features (i.e.,
Q-bands).
[32] Upon excitation of the fullerene core, for example, with a 355 nm laser
pulse, CS also occurred with the exception of toluene. This confirms
that both singlet states (1*ZnP: 2.00 eV; 1*C60: 1.76 eV) are sufficiently
energetic to power intramolecular CS processes.
[33] From Table 6, it appears that the quantum yields are largest for THF
and dichloroethane.
[34] a) R. A. Marcus, N. Sutin, Biochim. Biophys. Acta 1985, 811, 265;
b) R. A. Marcus, Angew. Chem. 1993, 105, 1161; Angew. Chem. Int.
Ed. Engl. 1993, 32, 1111.
[35] l ˆ 0.59 eV and Vˆ 51 cmÀ1 (meta-ZnP C60).
[36] P. Vath, M. B. Zimmt, D. V. Matyushov, G. A. Voth, J. Phys. Chem. B
2000, 103, 9130.
[37] a) J. Ponder, F. Richards, J. Comput. Chem. 1987, 8, 1016; b) C.
Kundrot, J. Ponder, F. Richards, J. Comput. Chem. 1991, 12, 402.
c) M. J. Dudek, J. Ponder, J. Comput. Chem. 1995, 16, 791.
[38] a) N. L. Allinger, Y. H. Yuh, J.-H. Lii, J. Am. Chem. Soc. 1989, 111,
8551. b) J.-H. Lii, N. L. Allinger, J. Am. Chem. Soc. 1989, 111, 8566;
c) J.-H. Lii, N. L. Allinger, J. Am. Chem. Soc. 1989, 111, 8576.
[39] a) D. A. Leigh, A. Murphy, J. P. Smart, M. S. Deleuze, F. Zerbetto, J.
Am. Chem. Soc. 1998, 120, 6458; b) R. Caciuffo, A. Degli Esposti,
M. S. Deleuze, D. A. Leigh, A. Murphy, B. Paci, S. Parker, F. Zerbetto,
J. Chem. Phys. 1998, 109, 11094; c) M. S. Deleuze, D. A. Leigh, F.
Zerbetto, J. Am. Chem. Soc. 1999, 121, 2364; d) V. Bermudez, N.
Capron, G. Torsten, F. G. Gatti, F. Kajzar, D. A. Leigh, F. Zerbetto, S.
Zhang, Nature 2000, 406, 608; e) M. Cavallini, M. R. Lazzaroni, R.
Zamboni, F. Biscarini, D. Timpel, F. Zerbetto, G. J. Clarkson, D. A.
Leigh, J. Phys. Chem. B 2001, 10, 10826.
[40] H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, A. Di Nola,
J. R. Haak, J. Chem. Phys. 1984, 81, 3684.
[41] H. C. Andersen, J. Comput. Phys. 1983, 52, 24.
[42] ™Semi Empirical Molecular Orbital Methods∫: M. C. Zerner, Rev.
Comput. Chem. 1999, 2, 313.
[43] a) G. Orlandi, A. Troisi, F. Zerbetto, J. Am. Chem. Soc. 1999, 121,
5392; b) A. Troisi, G. Orlandi Chem. Phys. Lett. 2001, 344, 509.
[44] I. Lamparth, A. Hirsch, Chem. Commun. 1994, 1727.
[27] T. Fˆrster, Discuss. Faraday Soc. 1959, 27, 7; the overlap integral (J ˆ
2.3 Â 10À14 cm6 molÀ1) was evaluated on the basis of Equation (6). For
calculations of the critical distance Rc [Eq. (7)], K2 the geometry
factor was chosen as 2/3, N the Avogardo constant as 6.022 Â
1023 molÀ1 and n as the refractive index of the solvent.
R
4
~
R
~
~
~
‰Fꢁn†eꢁn†Š=n dn
J ˆ
(6)
(7)
~
~
Fꢁn†dn
9000ꢁln 10†K2F
128 p5Nn4
Rc6
ˆ
J
[28] D. L. Dexter, J. Chem. Phys. 1953, 21, 836 The overlap integral (J ˆ
5.0 Â 10À4 cmÀ1) was evaluated on the basis of Equation (8):
R
~
~
R
~
Fꢁn†eꢁn† dn
J ˆ R
(8)
~
~
~
~
Fꢁn† dn eꢁn† dn
[29] The CTemission is found upon excitation into 1) the fullerene features
(i.e., 325 nm), 2) the porphyrin features (i.e., 425 nm), and 3) the CT
absorption (i.e., <700 nm).
[30] a) H. Imahori, N. V. Tkachenko, V. Vehmanen, K. Tamaki, H.
Lemmetyinen, Y. Sakata, S. Fukuzumi, J. Phys. Chem. A 2001, 105,
1750; b) V. Vehmanen, N. V. Tkachenko, H. Imahori, S. Fukuzumi, H.
Lemmetyinen, Spectrochim. Acta Part A 2001, 57, 2229.
Received: March 26, 2003 [F4995]
Chem. Eur. J. 2003, 9, 4968 4979
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4979