Screening electronic communication through ortho-, meta- And para-substituted linkers separating subphthalocyanines and C60

Article abstract of DOI:10.1002/chem.200800910

We have prepared two complementary series of SubPc-C₆₀ (SubPc=subphthalocyanine) electron/energy donor-acceptor systems, in which the two constituents are linked through ortho-, meta-, or para-substituted phenoxy spacers. In one of the series (

Full text of DOI:10.1002/chem.200800910

DOI: 10.1002/chem.200800910
Screening Electronic Communication through ortho-, meta- and para-
Substituted Linkers Separating Subphthalocyanines and C60
David Gonzµlez-Rodríguez,[b] Tomµs Torres,*[b] María ngeles Herranz,[c, d]
Luis Echegoyen,*[c] Esther Carbonell,[a] and Dirk M. Guldi*[a]
Abstract: We have prepared two com-
plementary series of SubPc–C60
and diphenylamino groups both influ-
ence the resulting oxidation potentials
of the electron-donating SubPc. They
also modulate the outcome of excited
state interactions, namely, energy and/
or charge transfer. In addition, we have
studied the impact that the substitution
pattern in the phenoxy spacer exerts
onto intramolecular processes in the
ground and excited states. Although
some of these processes are governed
by the spatial separation between both
components, the different electronic
coupling through ortho-, meta-, or
para- connections also plays decisive
roles in some cases.
(SubPc=subphthalocyanine) electron/
energy donor–acceptor systems, in
which the two constituents are linked
through ortho-, meta-, or para-substi-
tuted phenoxy spacers. In one of the
series (1a) the SubPc units bear iodine
atoms, while in the other series (1b) di-
phenylamino groups are linked to the
SubPc macrocycles. The iodine atoms
Keywords: donor–acceptor
sys-
tems · electron transfer · energy
transfer · fullerenes · subphthalo-
cyanines
Introduction
standing factors that ultimately influence photoinduced
energy- and electron-transfer reactions.[1] The importance of
energy and/or electron transfer is undisputed, since these
two primary events govern the conversion of solar energy
into profitable chemical energy in natural photoactive reac-
tion centers.[2] They also play a critical role in the perfor-
mance of, for example, organic solar cells.[3] Usually, well-de-
fined spacers are chosen to connect donors and acceptors.
Their nature, which may include covalent or noncovalent, p-
conjugated or nonconjugated, rigid or flexible, and so forth,
determines the degree of electronic communication, the dis-
tance, and the relative orientation.[4] In this context, p-con-
jugated systems stand out as spacers to afford excellent elec-
tronic couplings and, consequently, to accelerate intramolec-
ular events between the two active components.[5] Besides,
they may allow switching or fine-tuning of the electronic
communication.[6,7]
Testing molecular model systems—composed of electron-do-
nating and electron-accepting constituents—assists in under-
[a] Dr. E. Carbonell, Prof. Dr. D. M. Guldi
Department of Chemistry and Pharmacy &
Interdisciplinary Center for Molecular Materials (ICMM)
Friedrich-Alexander-Universität Erlangen-Nürnberg
91058 Erlangen (Germany)
Fax: ( +49)91318528307
E-mail: dirk.guldi@chemie.uni-erlangen.de
[b] Dr. D. Gonzµlez-Rodríguez, Prof. Dr. T. Torres
Departamento de Química Orgµnica
Facultad de Ciencias, Universidad Autónoma de Madrid
Cantoblanco, 28049 Madrid (Spain)
Fax: ( +34)91-497-3966
E-mail: tomas.torres@uam.es
Recent studies have illustrated the kinetic dependencies
of intramolecular processes on the ortho, meta, or para con-
nections in rigid, isomeric, phenyl spacers. These spacers
regulate the distance between the donor and acceptor and,
consequently, intramolecular processes that occur by means
of through-space mechanisms will be favored in the follow-
ing order: ortho>meta>para.[8] Appreciable differences
emerge for the electronic couplings in the isomeric forms. In
particular, in systems in which intramolecular events are
driven by through-bond interactions (i.e., the p and s bonds
[c] Dr. M. . Herranz, Prof. Dr. L. Echegoyen
Department of Chemistry, Clemson University
South Carolina 29634 (USA)
E-mail: luis@clemson.edu
[d] Dr. M. . Herranz
Current address:
Departamento de Química Orgµnica
Facultad de Ciencias Químicas
Universidad Complutense de Madrid, 28040 Madrid (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800910.
7670
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7670 – 7679

FULL PAPER
of the spacer) direct linkages across meta positions are less
effective than linkages across para positions. A reasonable
interpretation for this finding is lent from the superexchange
mechanism,[9] which leads to a stronger increase in coupling
at the ortho and para positions relative to that at the meta
connections. Leading examples include singlet–singlet and
triplet–triplet energy transfer,[10] photoinduced charge sepa-
ration and charge recombination,[11] charge transfer to elec-
trodes,[12] or electronic communication in conjugated den-
drimers[13] and intervalence metal complexes.[14]
In addition, some of these studies reveal that meta linkers
may provide enhanced electronic coupling in the excited
state, operating as a gated wire in p-conjugated systems.[13]
This effect has been demonstrated in conjugated donor–ac-
ceptor systems, in which the meta connection yielded faster
charge separation and slower charge recombination rates
than the para connection.[11e]
tion on the SubPc macrocycle. Donor groups, including
phenyl ether or diphenylamino groups, led to a considerable
lowering of the SubPc oxidation potentials and, in turn, as-
sisted in favoring electron transfer. However, only energy
transfer was operative when the SubPc macrocycle was en-
dowed with fluorine or iodine substituents.
Herein we describe the synthesis of two complementary
series of SubPc–C60 energy/electron donor–acceptor systems,
in which the two constituents are linked through ortho-,
meta-, or para-substituted phenoxy spacers. In one of the
series (1a) the SubPc units bear iodine atoms,[17a] while in
the other series (1b) diphenylamino groups are linked to
the SubPc macrocycle. A wide range of electrochemical and
photochemical experiments were employed to probe ground
and excited state interactions, as a function of substitution
pattern in the spacer, that range from energy- to electron-
transfer reactions. We demonstrate that, while some of these
processes are clearly regulated by the separation distance
between both components, the different electronic coupling
through ortho-, meta-, or para- connections can also play an
important role in some cases.
Subphthalocyanines[15] (SubPcꢁs) are nonplanar aromatic
macrocycles that exhibit interesting optical properties.[16]
Our recent work focused on the synthesis and physicochemi-
cal features of subphthalocyanine/fullerene, electron donor–
acceptor systems[17] (SubPc–C60). In this work the two
photo- and redox-active units were connected by meta phen-
oxy spacers.[17b] Selective photoexcitation of SubPc gave
rise—depending on the type of SubPc—to either an energy-
or charge-transfer mechanism to yield either a photoexcited
or one-electron reduced fullerene, respectively, as a reactive
intermediate. The outcome between these two competitive
processes was found to depend on the peripheral substitu-
Results and Discussion
Synthesis: The synthesis of 1a and 1b was carried out by
employing triiodoSubPc 4a[16b] as a common synthetic pre-
cursor (Scheme 1). The axial chlorine atom was first re-
placed by ortho-, meta-, or para-formylphenoxy ligands,
Scheme 1. Synthesis of the two sets of isomeric SubPc-C60 (1a and 1b).
Chem. Eur. J. 2008, 14, 7670 – 7679
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7671

D. M. Guldi, T. Torres, L. Echegoyen et al.
leading to o-2a, m-2a, and p-2a, or by phenol, yielding the
reference SubPc 3a. This set of triiodoSubPc compounds
was then subjected to a palladium-catalyzed amination reac-
tion[18] with diphenylamine that afforded o-2b, m-2b, p-2b,
and 3b in nice yields (75–87%). As an alternative to this
synthetic route we also examined the direct amination reac-
tion with 4a, but instead of observing the incorporation of
the diphenylamino substituents, the starting chloroSubPc de-
composed during these conditions. It seems that an axial
phenoxy group imparts chemical stability to the SubPc mac-
rocycle.[19] Finally, o-1a, m-1a, p-1a, o-1b, m-1b, and p-1b,
were prepared by 1,3-dipolar cycloaddition reactions[20] be-
tween the corresponding SubPc 2a and 2b compounds and
C60.[21]
Purification was performed by silica gel column chroma-
tography with toluene or toluene/hexane mixtures as elu-
ents. Unreacted fullerene eluted first, followed by the mono-
addition compound and a small amount of bisadducts. The
corresponding C3 and C1 regioisomers of the monoadducts
1a and 1b could be separated by column chromatography
and were individually characterized.[22] Moreover, in these
intrinsically chiral macrocycles, the presence of an additional
stereogenic center at the pyrrolidine ring gives rise to a 1:1
mixture of two diastereoisomers for each C1/C3 isomer. Only
the diastereoisomers of ortho-isomeric systems were found
to separate in some cases by column chromatography with
the conditions employed. Most probably, the restricted rota-
tion—the two subunits being held close by the spacer—is re-
sponsible for a greater differentiation between the diaste-
reoisomers.
Figure 1. Models and estimated range of interchromophoric distances for
o-, m-, and p- isomeric SubPc-C60. Portion of the 1H NMR spectra
(CDCl3, 300 MHz) of o-1b, m-1b, and p-1b, showing the magnitude of
the upfield shift experienced by some of the pyrrolidine protons as a
function of their distance to the SubPc macrocycle.
in the spacer of the meta isomers seems to reflect an inter-
conversion between different conformations in a time scale
comparable to that of the 1H NMR measurements. This
must be due to the higher flexibility of the meta spacer,
which provides a wider distribution of interchromophoric
distances than the other two.
Ground state—electronic absorption spectra: The absorp-
tion spectra of all systems disclose features of both constitu-
ents (Figure S1a in the Supporting Information). For in-
stance, the presence of C60 is clearly seen through absorp-
tions in the 250–350 nm range, followed by a weak band
around 430 nm. The SubPc unit, on the other hand, domi-
nates the red part of the spectrum, in which the typical Q-
bands appear at around 570 or 615 nm for the a or b series,
respectively. This band is responsible for the marked
changes in color that ranges from the typical intense magen-
ta to green. In the b series, an additional broad band evolves
around 450 nm, which is attributed to n–p* transitions of
the amine group.
In the dyads, the SubPc Q-band shows a small red shift
when compared to the corresponding reference compounds
(i.e., 3a and 3b). This bathochromic shift (as illustrated in
Figure S1b in the Supporting Information for o-1b, m-1b,
and p-1b) slightly weakens as the distance between the two
constituents increases. This trend was qualitatively repro-
duced in several solvents, that is, toluene, CHCl3, THF, or
even benzonitrile, and seems to indicate a distance-depen-
dent perturbation of the p–p* SubPc transition by the fuller-
ene unit. Moreover, in the diphenylamine-substituted series
the magnitude of the red shift shows a solvent dependence,
and an increase in solvent polarity evokes an additional red
shift of the SubPc Q-bands (Figure S1c in the Supporting
Information).
Evaluation of the relative distance between the SubPc and
the C60 units: The nature of the phenoxy spacers, which
have two torsion angles around s-bonds that lead to multi-
ple conformations, impart some degree of flexibility to the
molecules. This, in turn, means that the C60 and SubPc units
are not restrained to any fixed relative position, but their
distance can shift within a small range. Structural modeling
of these systems, using molecular mechanics and semiempir-
ical methods,[23] allowed us to evaluate the maximum and
minimum distances between the two moieties in each
isomer. These values, defined as the distance between the
boron atom and the center of the carbon sphere, together
with a model of minimized conformation for each isomer,
are illustrated in Figure 1.
The calculated average distances agree quite well with the
upfield shifts observed for the signals of the pyrrolidine pro-
tons, especially when considering the signal of the proton on
the tertiary carbon atom (proton a, Figure 1). In fulleropyr-
rolidine 5 this proton signal appears as a singlet at d=
4.88 ppm. In o-1b, m-1b, and p-1b, on the other hand, the
same signal is shifted upfield to d=3.81, 4.39, and 4.44 ppm,
respectively. A distance-dependent interaction with the
SubPc ring-current is responsible for this trend. Similar ef-
fects were observed for the corresponding protons in o-1a,
m-1a, and p-1a at d=3.90, 4.52, and 4.60 ppm, respectively.
Furthermore, a significant broadening of the proton signals
7672
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7670 – 7679

Donor–Acceptor Systems
FULL PAPER
Ground state—electrochemistry: Cyclic voltammetry experi-
ments have been carried out with o-1a, m-1a, p-1a, o-1b,
m-1b, and p-1b, and compared to the references 3a, 3b and
5. In Table 1 the redoxpotentials (as measured with this
technique) are summarized, while Figure 2 displays the
cyclic voltammograms in THF.
maining compounds, this oxidation is placed close to the sol-
vent threshold and, in turn, barely observed. The cyclic vol-
tammograms of 3a also display an irreversible many elec-
tron wave reduction (ꢀ3e) at around À1900 mV (E2p,red). A
similar wave is observed for 1a (not shown).
In the case of the diphenylamino series, the first macrocy-
cle-based oxidation was detect-
Table 1. Potential data (mV vs. ferrocene) for SubPc-C60 dyads 1 and reference compounds 3 and 5 in THF.
ed at +0.37 V. It displays the
usual electrochemical irreversi-
bility seen during the oxidations
of SubPc. When scanning to
more positive potentials, the b
series exhibits a second irrever-
sible event around +0.6 V. A
three-electron quasi-reversible
wave suggests here an oxidation
of the diphenylamino groups.
[a]
[a]
[a]
E3ox
E2ox
E1ox
E11=2,red
E21=2,red
E31=2,red
HOMO–LUMO[b]
3a
+821
+822
+779
+854
+374
+436
+372
+378
À1323
À1016
À1004
À1037
À1579
À1025
À1017
À1028
À910
À1900[a] (=3e)
o-1a
m-1a
p-1a
3b
o-1b
m-1b
p-1b
5
À1393
À1560
1.84
1.78
1.89
+998
+1038
+616
+606
+586
+623
À1347
À1511
À1380
À1567
+824
+846
+840
+846
À2079
À2750[a] (=3e)
À2153
À1567 (2e)
À1532 (2e)
À1543 (2e)
À1435
1.46
1.39
1.41
À2162
À2180
À2019
[a] Electrochemically irreversible wave. Only peak potentials are given. [b] Estimated HOMO–LUMO energy
gap (in eV) as the difference between the first oxidative and reductive potentials of the dyads.
After
a full oxidation scan
strong adsorption at the work-
ing electrode evolves, which
leads to a green–blue film on the surface. This hampered
any further reductive electrochemistry from being detected.
The oxidative deposition, which very closely resembles sur-
face-confined behavior, is depicted in Figure S2 (Supporting
Information). This behavior was also observed in 1b and is
an interesting property that could be explored in the prepa-
ration of thin films.
As far as the cathodic part of 1b is concerned, the wave
at ꢀÀ1 V (E11=2,red), corresponding to one electron, was as-
signed to C60. The wave at ꢀÀ1.5 V (E21=2,red), a two-electron
process, seems to involve the first reduction of SubPc and
the second reduction of C60. The third reduction wave at ap-
proximately À2.1 V (one electron) could not definitely be
assigned to one or the other of the electroactive species. Fol-
lowing this third reduction process, the observation of the
multielectronic (=3e) irreversible reduction of the SubPc at
À2.7 V precluded further assignments from the cyclic vol-
tammograms. However, Osteryoung square-wave voltamme-
try (OSWV; see Figure S3) reveals the close spatial proximi-
ty between forming the tetra- and penta-anionic species in
these molecules. Adding a third electron to C60 and a second
electron to SubPc ring appears to happen nearly simultane-
ously. The order that these events follow is not clearly estab-
lished nor is the reversibility of the processes, due to the im-
minent presence of the irreversible multielectronic reduction
of the SubPc fragment at around À2.7 V.
Figure 2. Cyclic voltammograms (mV vs. ferrocene) of references 3a and
3b and o-1a, m-1a, p-1a, o-1b, m-1b, and p-1b in THF and 0.1m
TBAPF6 as supporting electrolyte.
Comparing the redoxpotentials of the ortho-, meta-, and
para- isomeric 1a and 1b revealed that, in both series, a
meta-configured spacer induces easier reduction and oxida-
tion of the two electroactive subunits. This is also reflected
in the corresponding HOMO/LUMO energy gaps. As
shown in Table 1, the lowest values in both series were seen
for the meta-substituted m-1a and m-1b, indicating the
weakest interactions between the donor and acceptor
groups in each series. Therefore, the electronic effects that
influence these ground-state redoxprocesses seem to take
place by means of through-bond rather than through-space
For the o-, m-, and p-dyads (1a), the cathodic waves at
ꢀÀ1 V (E11=2,red
)
and ꢀÀ1.5 V (E31=2, red)—one electron
each—are assigned to C60, while those seen at ꢀ+0.8 V
(E1p,ox) and ꢀÀ1.4 V (E12=2,red)—one electron each—are cen-
tered on the SubPc. All three reduction processes were
found to be fully reversible. On the other hand, the first and
the second anodic oxidation processes of m-1a and p-1a are
irreversible. The second oxidation wave at around +1 V
seems to correspond to the oxidation of the nitrogen atom
in the pyrrolidine ring attached to the C60 group. In the re-
Chem. Eur. J. 2008, 14, 7670 – 7679
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7673

D. M. Guldi, T. Torres, L. Echegoyen et al.
interactions, since they follow the expected (and similar) be-
havior for the ortho and para, and the weakest electronic
coupling for the meta isomer.
Excited states—steady-state and time-resolved fluorescence
spectroscopy: First insight into excited-state interactions be-
tween the photo- and redox-active constituents came from
steady-state and time-resolved fluorescence experiments, in
which SubPc was excited either at 570 (i.e., 1a) or 615 nm
(i.e., 1b). A strong decrease in the SubPc fluorescence in-
tensity is noted when contrasting the results for the dyads
(i.e., 1a and 1b, quantum yields on the order of 10À3) with
those of the references (i.e., 3a and 3b, quantum yields on
the order of 10À1). Both the fluorescence quantum yields
and the fluorescence lifetimes were typically two orders of
magnitude lower in the dyads (see Table 2).[24]
Figure 3. Steady-state fluorescence spectra of o-1a (dotted spectrum), m-
1a (dashed spectrum), and p-1a (solid spectrum) in toluene exhibiting
the same absorption of 0.15 at the 530 nm excitation wavelength.
discloses features that match
Table 2. Selected photophysical parameters of SubPc-C60 dyads 1a and 1b as a function of solvent and isomer-
the ground state absorption of
SubPc and C60. This provides
unambiguous evidence for the
origin of the C60 fluorescence.
Tests in solvents, which are
more polar than toluene, for
example, THF, o-dichloroben-
zene or benzonitrile, led to the
same intensity of the C60 fluo-
rescence in the 1a series. The
only differences that were de-
rived correspond to the quan-
tum yields of energy transfer:
when employing 5 as a stan-
dard with matching absorption
at the excitation wavelength
ic connection between the two chromophores.
Feature
Solvent o-1a
m-1a
p-1a
o-1b
m-1b
p-1b
fluorescence maximum [nm]
toluene 596
592
591
654
655
658
fluorescence quantum yield (SubPc) toluene 7.3”10À4 7.7”10À4 1.0”10À3 3.7”10À2 2.1”10À2 5.1”10À3
[a]
[a]
[a]
THF
1.1”10À3 1.2”10À3 1.3”10À3
fluorescence quantum yield (C60)
singlet lifetime (C60) [ns]
toluene 6.6”10À4 4.7”10À4 6.0”10À4 !10À4
!10À4
!10À4
THF
toluene 1.2
THF 1.0
toluene 4.2
THF 4.4
6”10À4
3.610À4
1.0
0.9
5.4”10À4 !10À4
1.1
1.0
!10À4
!10À4
[a]
[a]
[a]
[a]
[a]
[a]
singlet lifetime (SubPc) [ps]
10.9
13.9
15.1
9.8
11.7
[a]
[a]
1.0
1.7
2.9
[c]
[c]
[c]
energy transfer quantum yield
radical ion pair lifetime [ps]
toluene 1.0
toluene
0.8
0.85
[b]
[b]
[b]
4055
133
73
4350
192
111
>5000
319
235
[b]
[b]
[b]
[b]
[b]
THF
bzcn
[b]
[a] Not measured. [b] No charge separation was found. [c] No energy transfer was found.
We have previously demonstrated that the SubPc singlet
excited states in m-1a and m-1b give rise to different reac-
tivity patters in their excited states, namely, energy- versus
charge-transfer deactivation.[17b] The most notable results
came from solvent dependence. Solvents of different polari-
ty led in m-1a to marginal changes in the quantum yields.
This suggests a nearly activationless fluorescence deactiva-
tion. For m-1b, on the other hand, using a more polar envi-
ronment amplified the nonradiative deactivation pathway.
The aforementioned trends are reestablished in the two
remaining isomers. Noticeable is the impact that the change
in distance and electronic coupling exerts on the deactiva-
tion rates: ortho-substitution in o-1a, which imposes un-
equivocally the shortest distance between SubPc and C60,
leads to the strongest quenching within the 1a series, where-
as the weakest SubPc fluorescence quenching was seen in
the para-isomer p-1a (Figure 3). Such a trend parallels the
red-shifts registered in the ground state absorption (see Fig-
ure S1 in the Supporting Information).
the following trend emerges: o-1a>m-1a>p-1a (see
Table 2). Implicit is a distance-dependent singlet–singlet
energy transfer from the SubPc singlet excited state in 1a
(i.e., 2.1 eV) to C60 (i.e., 1.76 eV) that thermodynamically
outperforms any charge transfer (i.e., ca. 1.85 eV).
Quite different is the outcome of the fluorescence experi-
ments with o-1b, m-1b and p-1b. Although the strongest
quenching is still seen in o-1b (Figure 4), a considerable de-
crease of the fluorescence in the more polar media attests to
a different mechanism. In 1b the charge-transfer products
are energetically situated around 1.4 eV, which, in turn,
would favor the charge-transfer pathway evolving from the
SubPc singlet excited state (i.e., 1.9 eV) rather than the
energy-transfer scenario.[25]
Excited states—transient absorption spectroscopy: Time-re-
solved transient absorption measurements provided further
details about the deactivation mechanism. Upon 500 nm ex-
citation of 3a/1a and 3b/1b, we observe the singlet excited-
state characteristics of the SubPc constituents. In all cases a
bleach of the ground state is accompanied by a new transi-
tion that develops in the red. For the 3a/1a series maxima
and minima evolve around 640 and 574 nm, respectively,
Mechanistically, upon SubPc excitation in toluene a fluo-
rescence pattern evolves that bears close resemblance with
that known for C60 (i.e., 5), that is, a maximum at 720 nm.
Furthermore, the excitation spectra of the C60 fluorescence
7674
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7670 – 7679

Donor–Acceptor Systems
FULL PAPER
Figure 4. Steady-state fluorescence spectra of o-1b (dotted spectrum), m-
1b (dashed spectrum), and p-1b (solid spectrum) in toluene exhibiting
the same absorption of 0.24 at the 582 nm excitation wavelength.
Figure 6. Time-absorption profiles at 560 nm of the differential absorp-
tion measurements with o-1a (open circles), m-1a (crosses; see also
Figure 5), and p-1a (full circles) in toluene, monitoring the energy-trans-
fer processes.
while the corresponding features for the 3b/1b series are
seen at 756 and 615 nm. The intersystem-crossing dynamics
are rather slow (ꢀ108 sÀ1) in the two references (i.e., 3a and
3b). This is in sharp contrast to the ultrafast decays of the
SubPc singlet–singlet absorption in 1a and 1b with rates
ꢀ1010 sÀ1.
The C60 singlet excited state intersystem crosses to the
corresponding triplet manifold with dynamics that are virtu-
ally identical to that in 5. Despite this close kinetic agree-
ment, spectroscopically the differential absorption changes
are no match to that of the C60 triplet excited state (i.e.,
maxima at 360 and 700 nm). Maxima at 460 and 620 nm and
a minimum at 570 nm suggest that the C60 singlet decay af-
fords the SubPc triplet excited state. Likewise, the only com-
ponent detected in the nanosecond experiments was that of
the SubPc triplet excited state. Implicit is that the fullerene
triplet (1.5 eV), once formed by intersystem crossing, under-
goes a thermodynamically allowed, but kinetically not re-
solvable, transfer of triplet energy to generate the SubPc
triplet excited state (ꢀ1.45 eV). The underlying kinetics are
affected by the isomeric form and tend to be faster for m-1a
than for p-1a and o-1a. Such a trend contrasts the reactivity
of singlet–singlet energy transfer.
In the 1a series, instead of detecting the transient changes
that are associated with the SubPc triplet excited state, we
note spectral characteristics that match those seen for 5,
namely, the C60 singlet excited state (Figure 5). The most
prominent feature is a maximum at 880 nm. Since these C60
singlet features develop concomitantly with the decay of the
SubPc singlet excited state, we conclude a transduction of
singlet-excited-state energy. Appreciable kinetic differences
emerge for the different isomers in toluene and THF with
energy-transfer rate constants that are larger in o-1a than in
m-1a and p-1a (Figure 6). In other words, the underlying
distance-dependence resembles that seen in the fluorescence
experiments.
Different are the results with o-1b, m-1b, and p-1b (see
Figure 7 and 8). The SubPc singlet-excited-state features
change rapidly into a broadly absorbing species. In the visi-
ble region the one-electron-oxidized radical cation of the
SubPc donor evolves with maxima at 750 nm, while in the
near-infrared region the signature (1000 nm) of the one-
electron-reduced radical anion of the C60 acceptor. Taken
these attributes into concert, there is no doubt about the
successful formation of SubPc +–C60 . As already men-
tioned, better donating abilities in the 1b series enables the
highly efficient (>90%) separation of charges, which is ther-
modynamically prohibited in the 1a series. Figure 8 corrobo-
rates that in toluene the o-1b, m-1b, and p-1b isomers give
rise to appreciable differences in charge-transfer kinetics:
m-1b>p-1b>o-1b. Inductive effects are responsible for the
diverse rate constants. In fact, these values agree well with
the thermodynamic driving force (i.e., HOMO–LUMO
gaps), which were derived from the electrochemical experi-
ments. In THF, in which the thermodynamic driving forces
for the charge-transfer kinetics (i.e., around 0.5 eV) are
brought closer to the top of the Marcus parabola (i.e., on
C
CÀ
Figure 5. Differential absorption spectra (visible and near-infrared) ob-
tained upon femtosecond flash photolysis (500 nm, 200 nJ) of m-1a in
argon-saturated toluene with time delays of 1.1 ps (solid spectrum), 44 ps
(dashed spectrum) and 2472 ps (dotted spectrum) at room temperature,
illustrating the sequence of singlet–singlet energy transfer, intersystem
crossing, and triplet–triplet energy transfer.
Chem. Eur. J. 2008, 14, 7670 – 7679
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7675

D. M. Guldi, T. Torres, L. Echegoyen et al.
Please note the rather large driving force for the charge re-
combination, which is on the order of 1.4 eV. Secondly, de-
creasing the distance between donor and acceptor decreases
the SubPc +–C60 lifetime despite the opposite effects in the
charge separation—exceeding a factor of two when contrast-
ing o-1b and p-1b.
C
CÀ
Conclusion
The current work demonstrates the impact that a partially
conjugated spacer exerts on the ground- and excited-state
communication between electron/energy donor and acceptor
units in two complementary sets of SubPc–C60 systems.
Perturbations of the electronic transitions, caused by the
proximity of the two p-conjugated units, dominate the
ground state. Additional electrochemical experiments reveal
that the HOMO–LUMO gaps tend to be smaller in the
meta-isomers when compared to those in the corresponding
ortho- and para-isomers. In summary, through-bond interac-
tions play an important role in the electronic communica-
tion between the donor and the acceptor in the ground
state.
Figure 7. Differential absorption spectra (visible and near-infrared) ob-
tained upon femtosecond flash photolysis (500 nm, 200 nJ) of m-1b in
argon-saturated toluene with time delays of 1.1 ps (solid spectrum) and
24 ps (dashed spectrum) at room temperature, illustrating the charge
transfer.
In the excited state, the underlying mechanism, energy
(i.e., iodine substituents: 1a) versus electron transfer (i.e.,
diphenylamino substituents: 1b), determines the interac-
tions. Distance matters when dipole–dipole interactions are
operative. These govern the transduction of singlet-excited-
state energy in 1a. As a consequence, o-1a reacts much
faster than any of the remaining isomers (i.e., p-1a and m-
1a). Triplet–triplet energy-transfer reactions, on the other
hand, follow a double electron-transfer mechanism, in which
inductive effects exert a stronger impact: m-1a reacts seem-
ingly faster than o-1a or p-1a. This conclusion finds inde-
pendent support in the electrochemical experiments, in
which the SubPc oxidation potentials reveal isomer-depen-
dent shifts that suggest through-bond electronic effects.
In the 1b series, for the initial charge separation a resem-
bling trends evolves, namely, m-1b undergoes faster charge
transfer than o-1b or p-1b. It is only for the strongly exo-
thermic charge recombination, again for which distance ef-
fects seem to dominate, and the rate constants exhibit the
following order: ortho>meta>para.[28]
Figure 8. Time-absorption profiles at 750 nm of the differential absorp-
tion measurements with o-1b (open circles), m-1b (crosses; see also
Figure 7), and p-1b (full circles) in toluene, monitoring the charge-trans-
fer processes.
the order of 0.6 eV as in phthalocyanine–C60 and porphryin–
C60),[4,26] the rate constants increase relative to those mea-
sured in toluene.
Charge recombination leads to the recovery of the ground
state for o-1b, m-1b, and p-1b, since the SubPc +–C60
The reason why this trend differs from previous stud-
ies,[11e] in which meta isomers showed the slowest charge re-
combination rates, probably resides in the nature of the
phenoxy spacer. This spacer offers more flexibility and
poorer electronic coupling than fully p-conjugated spacers.
C
CÀ
energy levels (i.e., 1.4–1.2 eV) are presumably situated
below those of both triplets (i.e., SubPc and C60). The kinet-
ics of charge recombination are strongly dependent on the
solvent polarity and on the isomeric form. Determining the
SubPc +-C60 lifetimes in the different solvents for o-1b, m-
C
CÀ
1b, and p-1b (Table 2) brought the following trends to light:
+
C
Firstly, increasing the solvent polarity decreases the SubPc
Experimental Section
CÀ
–C60 lifetime despite a better solvation of the radical ions:
several thousands of picoseconds in toluene and several
hundreds of picoseconds in THF, and about hundred pico-
seconds in benzonitrile. This is a clear attribute of the in-
verted region of the Marcus parabola, the thermodynamic
maximum of which is located between 0.5 and 0.6 eV.[4,27]
General methods: Melting points (m.p.) were determined in a Büchi
504392-S equipment and are uncorrected. UV/Vis spectra were recorded
with a Hewlett–Packard 8453 instrument. IR spectra were recorded on a
Bruker Vector 22 spectrophotometer. LSI-MS and HRMS spectra were
determined on a VG AutoSpec apparatus and MALDI-TOF-MS spectra
were obtained from a Bruker ReflexIII instrument equipped with a ni-
7676
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7670 – 7679

Donor–Acceptor Systems
FULL PAPER
trogen laser operating at 337 nm. NMR spectra were recorded with a
Bruker AC-300 instrument. Elemental analyses were performed with a
Perkin–Elmer 2400 CHN equipment. Column chromatography was car-
ried out on silica gel Merck-60 (230–400 mesh, 60 Š), and TLC on alumi-
num sheets precoated with silica gel 60 F254 (E. Merck). Chemicals were
purchased from commercial suppliers and used without further
purification.
139.0, 132.1, 129.7, 131.6, 131.5, 131.4, 130.3, 123.6, 123.5, 118.8,
96.5 ppm; MS (FAB, m-NBA): m/z: 894 [M]+, 773 [MÀaxial group]+;
HRMS: m/z calcd for C31H14N6O2BI3: 893.8405; found: 893.8438; UV/Vis
(CHCl3): lmax (loge)=572 (4.5), 531 (sh), 326 (3.9), 274 nm (4.2); FT-IR
À
(KBr): n˜ =1687 (C=O), 1597, 1508, 1459, 1429, 1266, 1176, 1039 (B O),
818, 779, 755, 701 cmÀ1; elemental analysis calcd (%) for C31H14N6O2BI3:
C 41.65, H 1.58, N 9.40; found: C 42.07, H 1.50, N 9.40.
Electrochemistry: Electrochemical measurements were performed at
room temperature in a home-built one-compartment cell with a three-
electrode configuration, containing 0.1m tetrabutylammonium hexafluor-
ophosphate (TBAPF6) as the supporting electrolyte, which was recrystal-
lized twice from ethanol and dried under vacuum. A glassy carbon or
gold electrode (3 or 1.5 mm 1, respectively) was used as the working
electrode, and a platinum mesh and a silver wire, separated from the so-
lution by a vycor tip, were employed as the counter and the reference
electrodes, respectively. Prior to each voltammetric measurement the cell
was degassed and pumped to 10À6 mmHg. The solvent, THF (5 mL),
which had also been degassed and pumped to the same pressure, was
then vapor-transferred into the cell, directly from Na/K. The electro-
chemical measurements were performed using a concentration of approx-
imately 0.5 mm of the corresponding compound, and ferrocene was
added as an internal reference.
Standard procedure for the synthesis of SubPcs 2b and 3b: An oven-
dried 25 mL flask was charged with finely ground Cs2CO3 (294 mg,
0.9 mmol), [Pd2ACTHRE(UNG dba)3] (8 mg, 0.009 mmol), BINAP (5.6 mg, 0.009 mmol),
diphenylamine (152 mg, 0.9 mmol) and the corresponding SubPc (2a or
3a (1:3 mixture of C3/C1 regioisomers; 0.1 mmol). The flask was then
purged with argon and dry toluene (10 mL) was added through a syringe.
The mixture was heated to reflux under argon atmosphere with continu-
ous stirring for 8 h in both cases. The solution was cooled to room tem-
perature, diluted with toluene (20 mL), filtered, and concentrated to
about 2–3 mL. The crude product was then purified by flash chromatog-
raphy on silica gel by using a mixture of hexane/THF (4:1 for 3b and 3:1
for 2b) as eluent. The C3 and C1 regioisomers of 2b were not separated
and are characterized below as 3:1 mixtures. The resulting deep green
solids could be further purified by precipitation from cold hexane. The
characterization of SubPcs 3b and m-2b has been recently reported.[17b]
Phototophysics: Steady-state emission spectra were recorded on a Fluo-
roMaxꢂ 3 Fluorometer (HORIBA). The measurements were carried out
at room temperature. Fluorescence lifetimes were measured with a Fluo-
SubPc o-2b (1:3 mixture of C3 and C1 isomers): Compound o-2b was ob-
tained as a green solid: 70 mg (69%); m.p. >2508C; H NMR (300 MHz,
1
CDCl3): d=9.07 (s, 1H), 8.55 (m, 3H), 8.35 (m, 3H), 7.53 (m, 3H), 7.4–
7.0 (m, 32H), 6.69 (dd, Jo’ =Jo’’ =7.6 Hz, 1H), 5.18 ppm (d, Jo’’ =7.6 Hz,
1H); 13C NMR (75.5 MHz, CDCl3): d=189.7, 156.5, 152.3, 151.5, 150.5,
150.32, 150.27, 150.0, 147.3, 147.2, 135.8, 135.2, 129.7, 129.6, 127.3, 125.5,
125.41, 125.36, 125.2, 124.4, 124.3, 124.2, 123.0, 122.9, 122.8, 121.2, 118.9,
114.4, 114.3, 114.2, 114.1 ppm; MS (FAB, m-NBA): m/z: 1018 [M+H]+;
HRMS: m/z calcd for C67H44N9O2B: 1017.3711; found: 1017.3745; UV/
Vis (CHCl3): lmax (loge)=623 (4.5), 577 (sh), 426 (4.2), 303 (4.6), 255 nm
(4.5); FT-IR (KBr): n˜ =1670 (C=O), 1609 (C=N), 1487, 1260, 1195, 1154,
À
rologꢂ TCSPC system (HORIBA). The samples were excited by
a
NanoLED-405 LH (peak wavelength 403 nm) and the signal was detect-
ed by a Hamamatsu MCP photomultiplier (type R3809U-50). Femtosec-
ond transient absorption studies were performed with 500 nm laser
pulses (1 kHz, 150 fs pulse width, 200 nJ) from amplified Ti:Sapphire
laser system (Clark-MXR, Inc.).
Standard procedure for the synthesis of SubPcs 2a and 3a: In a 25 mL
round-bottomed flask, equipped with a condenser and a magnetic stirrer,
the corresponding phenol (2.5 mmol) and SubPc 4a (0.5 mmol; 404 mg)
were heated under refluxin toluene (2 mL) for 40 h ( m-2a and p-2a) or
16 h (3a), depending on the reactivity of the starting SubPc and phenol.
In the case of o-2a the reaction was performed in a 10 mL round-bot-
tomed flask, equipped with a magnetic stirrer and rubber seal, in which
o-hydroxybenzaldehide (1 mmol) and SubPc 4a were placed. The mix-
ture was heated to the melting point of the phenol and stirred at that
temperature for 3 h. Then, in all cases, the reaction mixture was cooled
down to room temperature, the solvent was evaporated, and the resulting
residue was washed with a 4:1 mixture of methanol/water. The dark ma-
genta solid was then subjected to column chromatography on silica gel
using toluene (3a) or mixtures of toluene/THF (50:1 for o-2a or 30:1 for
m-2a and p-2a). The C3 and C1 regioisomers of 2a were not separated
and are characterized below as 3:1 mixtures. All the compounds were
further purified by recrystallization from CH2Cl2/hexane mixtures. The
characterization of SubPcs 3a and m-2a has been recently reported.[17b]
1044 (B O), 750, 697 cmÀ1
;
elemental analysis calcd (%) for
C67H44N9O2B: C 79.05, H 4.36, N 12.38; found: C 78.47, H 4.69, N 12.22.
SubPc p-2b (1:3 mixture of C3 and C1 isomers): Compound p-2b was ob-
1
tained as a green solid: 80 mg (79%); m.p. >2508C; H NMR (300 MHz,
CDCl3): d=9.63 (s, 1H), 8.55 (m, 3H), 8.35 (m, 3H), 7.54 (m, 3H), 7.4–
7.0 (m, 32H), 5.48 ppm (AA’XX’ system, 2H); 13C NMR (75.5 MHz,
CDCl3): d=190.8, 159.1, 152.5, 151.6, 151.0, 150.6, 150.3, 150.1, 150.0,
149.2, 147.3, 147.23, 147.17, 133.2, 133.0, 132.6, 131.4, 129.74, 129.68,
125.5, 125.4, 125.3, 125.2, 124.4, 124.3, 124.2, 123.0, 122.9, 122.8, 118.7,
114.4, 114.3, 114.2, 114.1 ppm; MS (FAB, m-NBA): m/z: 1017 [M]+;
HRMS: m/z calcd for C67H44N9O2B: 1017.3711; found: 1017.3706; UV/
Vis (CHCl3): lmax (loge)=626 (4.4), 484 (3.8), 436 (3.9), 298 nm (4.5); FT-
IR (KBr): n˜ =1696 (C=O), 1601 (C=N), 1480, 1264, 1196, 1156, 1034 (B-
O), 751, 697 cmÀ1; elemental analysis calcd (%) for C67H44N9O2B: C
79.05, H 4.36, N 12.38; found: C 78.66, H 4.52, N 11.97.
Standard procedure for the synthesis of SubPc-C60 dyads 1a and 1b: A
solution of C60 fullerene (70 mg, 0.1 mmol), N-methylglycine (25 mg,
0.27 mmol), and the corresponding (formylphenoxy)SubPc 2a or 2b
(0.09 mmol) in dry toluene (50 mL) was heated to refluxunder argon at-
mosphere for 20 h (for m-2a, p-2a, m-2b, and p-2b) or 30 h (for o-2a
and o-2b). The solution was then cooled to room temperature and con-
centrated under vacuum to about 10 mL. The resulting mixture was
poured onto a silica gel column and eluted with toluene/hexane 5:1 (for
compound o-1a) or toluene (for all the rest of products), in order to sep-
arate the monoaddition products from the bisadducts and unreacted full-
erene. The C3 and C1 regioisomers of all the compounds 1a and 1b were
separated during chromatography and their individual characterization
data is listed below following their elution order. Further purification of
the monoadducts was achieved by thoroughly washing the solid products
with acetone, methanol and hexane. The characterization of SubPc-C60
dyads m-1a and m-1b has been recently reported.[17b]
SubPc o-2a (1:3 mixture of C3 and C1 isomers): Compound o-2a was ob-
tained as a pink solid: 246 mg (55%); m.p. >2508C; H NMR (300 MHz,
1
CDCl3): d=9.20 (brs, 3H), 8.94 (s, 1H), 8.55 (d, Jo =8.2 Hz, 3H), 8.22 (d,
Jo =8.2 Hz, 1H), 7.29 (dd, Jo =8.1 Hz, Jm =1.8 Hz, 1H), 7.03 (ddd, Jo =
8.1 Hz, Jo’ =7.3 Hz, Jm’ =1.8 Hz, 1H), 6.74 (dd, Jo’ =7.3 Hz, Jo’’ =7.6 Hz,
1H), 5.13 ppm (dd, Jo’’ =7.6 Hz, Jm’ =1.8 Hz, 1H); 13C NMR (75.5 MHz,
CDCl3): d=188.9, 155.4, 151.5, 152.4, 151.2, 150.4, 150.2, 150.0, 139.0,
135.2, 132.2, 129.8, 131.6, 127.9, 123.7, 123.6, 122.1, 119.3, 96.6 ppm; MS
(FAB, m-NBA): m/z: 894 [M]+, 773 [MÀaxial group]+; HRMS: m/z
calcd for C31H14N6O2BI3: 893.8405; found: 893.8517; UV/Vis (CHCl3):
lmax (log e)=572 (4.5), 532 (sh), 325 (3.9), 273 nm (4.2); FT-IR (KBr):
À
n˜ =1684 (C=O), 1593, 1546, 1433, 1383, 1291, 1260, 1229, 1143, 1041 (B
O), 817, 752, 702 cmÀ1; elemental analysis calcd (%) for C31H14N6O2BI3:
C 41.65, H 1.58, N 9.40; found: C 41.24, H 1.67, N 9.36.
SubPc p-2a (1:3 mixture of C3 and C1 isomers): Compound p-2a was ob-
1
tained as a pink solid: 299 mg (67%); m.p. >2508C; H NMR (300 MHz,
SubPc-C60 o-1a; C3 isomer (1:1 mixture of diastereoisomers): Compound
o-1a (C3) was obtained as a dark red solid: 14 mg (9%); m.p. >2508C;
1H NMR (300 MHz, CDCl3/CS2 (1:5)): d=9.22, 9.18 (2d, Jm =1.8 Hz,
3H), 8.57, 8.54 (2d, Jo =8.2 Hz, 3H), 8,21 (m, 3H), 7.40 (m, 1H), 6.72
CDCl3): d=9.61 (s, 1H), 9.15 (m, 3H), 8.50 (m, 3H), 8.18 (m, 3H), 7.31
(AA’XX’ system, 2H), 5.44 ppm (AA’XX’ system, 2H); 13C NMR
(75.5 MHz, CDCl3): d=190.6, 158.1, 151.6, 151.4, 151.2, 150.3, 150.1,
Chem. Eur. J. 2008, 14, 7670 – 7679
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7677

D. M. Guldi, T. Torres, L. Echegoyen et al.
(m, 2H), 4.84 (m, 1H), 4.70 (m, 1H), 3.96 (d, 2J=9.7 Hz, 1H), 3.91, 3.89
(2s, 1H), 2.32 ppm (s, 3H); MS (MALDI-TOF, dithranol): m/z: 1642
[M+H]+; UV/Vis (CHCl3): lmax (loge)=575 (4.4), 525 (sh), 430, 320
(4.2), 262 nm (4.6); FT-IR (KBr): n˜ =2934, 2922, 1626, 1451, 1344, 1276,
151.84, 151.77, 151.2, 151.0, 150.21, 150.16, 149.8, 149.6, 149.4, 149.3,
148.7, 148.66, 148.5, 146.5, 146.3, 146.2, 146.18, 145.8, 145.6, 145.4, 145.0,
149.7, 144.9, 144.8, 144.73, 144.69, 144.4, 144.2, 143.7, 143.6, 143.5, 143.3,
143.0, 142.6, 142.3, 142.0, 141.6, 141.34, 141.26, 141.0, 140.6, 140.5, 140.4,
139.6, 139.3, 139.0, 138.7, 136.11, 136.06, 135.2, 135.1, 147.5, 147.42,
147.37, 137.0, 133.3, 133.2, 129.7, 129.5, 129.0, 128.6, 125.7, 125.5, 125.4,
124.8, 124.7, 124.5, 124.3, 124.2, 124.1, 124.0, 123.1, 123.0, 121.8, 117.2,
114.33, 114.27, 114.2, 77.6, 69.9, 39.7 ppm; MS (MALDI-TOF, dithranol):
m/z: 1766 [M+H]+; UV/Vis (CHCl3): lmax (loge)=629 (4.5), 581 (sh), 436
(4.1), 401 (3.9), 306 (4.7), 259 nm (4.9); FT-IR (KBr): n=2957, 1596,
1060 cmÀ1 (B O); elemental analysis calcd (%) for C93H19N7OBI3: C
À
68.04, H 1.17, N 5.97; found: C 67.16, H 1.32, N 6.34.
SubPc-C60 o-1a; C1 isomer: Compound o-1a (C1) was obtained as a dark
red solid: 50 mg (34%). The two diastereoisomers of o-1a (C1: a and b)
could be isolated and characterized independently by 1H NMR and UV/
Vis spectroscopy. For the measuring of melting points, 13C NMR and FT-
IR spectroscopy, MS, and elemental analysis, the mixture of diastereoiso-
mers was instead employed. m.p. >2508C; 1H NMR (300 MHz, CDCl3/
CS2 (1:5)): diastereoisomer a: d=9.20 (m, 3H), 8.55 (m, 3H), 8,21 (m,
3H), 7.39 (m, 1H), 6.71 (m, 2H), 4.82 (m, 1H), 4.69 (d, 2J=9.3 Hz, 1H),
3.96 (d, 2J=9.3 Hz, 1H), 3.91 (s, 1H), 2.31 ppm (s, 3H); 1H NMR
(300 MHz, CDCl3/CS2 (1:5)): diastereoisomer b: d=9.18 (m, 3H), 8.54
(m, 3H), 8.20 (m, 3H), 7.39 (dd, Jo =6.9 Hz, Jm =2.4 Hz, 1H), 6.71 (m,
2H), 4.80 (m, 1H), 4.68 (d, 2J=9.3 Hz, 1H), 3.94 (d, 2J=9.3 Hz, 1H),
3.90 (s, 1H), 2.30 ppm (s, 3H); 13C NMR (75.5 MHz, CDCl3/CS2 (1:5)):
d=138.5, 138.4, 131.6, 131.5, 129.2, 128.6, 123.5, 123.4, 121.8, 116.9, 96.9,
96.5, 78.1, 69.8, 39.6 ppm. The low solubility of this dyad prevented the
clear detection of most of the quaternary carbon atom signals, even after
36 h. MS (MALDI-TOF, dithranol): m/z: 1642 [M+H]+; UV/Vis
(CHCl3): lmax (loge)=575 (4.4), 525 (sh), 430, 320 (4.2), 262 nm (4.6);
1480, 1453, 1279, 1185, 1050 (B O), 751, 697 cmÀ1; elemental analysis
À
calcd (%) for C129H49N10OB: C 87.75, H 2.80, N 7.93; found: C 86.92, H
2.91, N 8.21.
SubPc-C60 p-1b; C3 isomer (1:1 mixture of diastereoisomers): Compound
p-1b (C3) was obtained as a dark green solid: 16 mg (11%); m.p.
>2508C; 1H NMR (300 MHz, CDCl3): d=8.44, 8.42 (2d, Jo =8.8 Hz,
3H), 8.37, 8.29 (2d, Jm =1.8 Hz, 3H), 7.43 (m, 3H), 7.35–7.0 (m, 32H),
5.40 (m, 2H), 4.75 (d, 2J=9.4 Hz, 1H), 4.50, 4.49 (2s, 1H), 4.03 (d, 2J=
9.4 Hz, 1H), 2.56, 2.55 ppm (2s, 3H); MS (MALDI-TOF, dithranol): m/
z: 1766 [M+H]+; UV/Vis (CHCl3): lmax (loge)=624 (4.4), 578 (sh), 433
(4.0), 308 (4.7), 260 nm (4.8); FT-IR (KBr): n˜ =2977, 1597, 1489, 1260,
1179, 1044 (B O), 818, 751, 697 cmÀ1; elemental analysis calcd (%) for
À
C129H49N10OB: C 87.75, H 2.80, N 7.93; found: C 86.19, H 2.92, N 7.53.
SubPc-C60 p-1b; C1 isomer (1:1 mixture of diastereoisomers): Compound
p-1b (C1) was obtained as a dark green solid: 57 mg (35%); m.p.
>2508C; 1H NMR (300 MHz, CDCl3): d=8.42 (m, 3H), 8.33, 8.26 (2m,
3H), 7.42 (m, 3H), 7.35–7.0 (m, 32H), 5.37 (m, 1H), 4.73 (d, 2J=9.4 Hz,
1H), 4.49 (brs, 1H), 4.00 (d, 2J=9.4 Hz, 1H), 2.56, 2.54 ppm (2s, 3H);
13C NMR (75.5 MHz, CDCl3): d=155.5, 155.2, 153.3, 153.2, 151.8, 151.7,
151.1, 151.0, 150.2, 150.1, 149.9, 149.7, 149.6, 149.4, 149.3, 148.9, 148.7,
148.5, 146.3, 146.14, 146.08, 145.61, 145.56, 145.3, 145.1, 145.0, 144.94,
144.89, 144.7, 144.5, 144.4, 144.1, 143.9, 143.5, 143.0, 142.3, 142.0, 141.6,
141.31, 141.26, 141.1, 140.9, 140.6, 140.4, 139.3, 139.2, 139.1, 138.8, 136.2,
136.0, 135.0, 134.9, 147.1, 147.0, 137.5, 133.5, 133.4, 133.2, 129.8, 125.5,
125.4, 125.2, 124.7, 124.6, 124.4, 124.2, 124.1, 123.2, 123.1, 123.0, 121.9,
114.5, 114.4, 114.2, 82.6, 69.7, 68.2, 39.6 ppm; MS (MALDI-TOF, dithra-
nol): m/z: 1766 [M+H]+; UV/Vis (CHCl3): lmax (loge)=624 (4.5), 578
(sh), 433 (4.1), 308 (4.8), 260 nm (4.9); FT-IR (KBr): n˜ =2975, 2923, 1599,
both diastereoisomers
a
and
b
presented the same features. FT-IR
(KBr): n˜ =2934, 2920, 1628, 1449, 1345, 1280, 1059 cmÀ1 (B O); elemen-
tal analysis calcd (%) for C93H19N7OBI3: C 68.04, H 1.17, N 5.97; found:
C 67.30, H 1.26, N 5.72.
À
SubPc-C60 p-1a; C3 isomer (1:1 mixture of diastereoisomers): Compound
p-1a (C3) was obtained as a dark red solid: 20 mg (14%); m.p. >2508C;
1H NMR (300 MHz, CDCl3/CS2 (1:5)): d=9.13–9.07 (2m, 3H), 8.48, 8.43
(2d, J0 =8.2 Hz, 3H), 8,15 (m, 3H), 7.18 (m, 2H), 5.37 (m, 2H), 4.83 (d,
2
2J=9.7 Hz, 1H), 4.60 (s, 1H), 4.10 (d, J=9.7 Hz, 1H), 2.59 ppm (s, 3H);
MS (MALDI-TOF, dithranol): m/z: 1642 [M+H]+; UV/Vis (CHCl3): lmax
(loge)=572 (4.4), 522 (sh), 432, 320 (4.2), 260 nm (4.6); FT-IR (KBr): n˜ =
2974, 2919, 1625, 1453, 1345, 1276, 1061 cmÀ1 (B O); elemental analysis
À
calcd (%) for C93H19N7OBI3: C 68.04, H 1.17, N 5.97; found: C 67.29, H
1.19, N 5.59.
1489, 1261, 1181, 1045 (B O), 818, 750, 699 cmÀ1; elemental analysis
À
SubPc-C60 p-1a; C1 isomer (1:1 mixture of diastereoisomers): Compound
p-1a (C1) was obtained as a dark red solid: 61 mg (42%); m.p. >2508C;
1H NMR (300 MHz, CDCl3/CS2 (1:5)): d=9.09 (m, 3H), 8.46 (m, 3H),
8,15 (m, 3H), 7.18 (m, 2H), 5.38 (m, 2H), 4.82 (d, 2J=9.3 Hz, 1H), 4.59
calcd (%) for C129H49N10OB: C 87.75, H 2.80, N 7.93; found: C 87.09, H
2.83, N 7.69.
2
(s, 1H), 4.09 (d, J=9.3 Hz, 1H), 2.59 ppm (s, 3H); 13C NMR (75.5 MHz,
CDCl3/CS2 (1:5)): d=138.5, 131.6, 131.5, 129.6, 123.3, 119.0, 96.9, 82.4,
81.8, 68.2, 69.5, 39.5 ppm. The low solubility of this dyad prevented the
clear detection of most of the quaternary carbon atom signals, even after
36 h. MS (MALDI-TOF, dithranol): m/z: 1642 [M+H]+; UV/Vis
(CHCl3): lmax (loge)=572 (4.5), 520 (sh), 432, 320 (4.3), 260 nm (4.7); FT-
Acknowledgements
Financial support from the Ministerio de Ciencia
y Tecnología
(CTQ2005–08933/BQU, Consolider-Ingenio 2010 CSD2007–00010 Nano-
ciencia Molecular, ESF-MEC MAT2006–28180-E, SOHYDS), and Co-
munidad de Madrid (MADRISOLAR, S-0505/PPQ/0225) is gratefully ac-
knowledged (T.T.). This work was also partially supported by the U.S.
National Science Foundation, grant DMR-0809129 (L.E.), Deutsche
Forschungsgemeinschaft (SFB 583), FCI, and the Office of Basic Energy
Sciences of the U.S. Department of Energy (D.M.G.).
IR (KBr): n˜ =2934, 2920, 1628, 1449, 1345, 1280, 1059 cmÀ1 (B O); ele-
À
mental analysis calcd (%) for C93H19N7OBI3: C 68.04, H 1.17, N 5.97;
found: C 67.11, H 1.29, N 5.77.
SubPc-C60 o-1b; C3 isomer (1:1 mixture of diastereoisomers): Compound
o-1b (C3) was obtained as
a dark green solid: 15 mg (9%); m.p.
>2508C; 1H NMR (300 MHz, CDCl3): d=8.52, 8.49 (2d, Jo =8.6 Hz,
3H), 8.39, 8.37 (2d, Jm =1.8 Hz, 3H), 7.50 (dd, Jo =8.6 Hz, Jm =1.8 Hz,
3H), 7.29 (m, 13H), 7.15 (m, 18H), 6.70 (m, 2H), 5.10 (m, 1H), 4.73 (d,
2J=9.4 Hz, 1H), 4.05 (d, 2J=9.4 Hz, 1H), 3.83, 3.79 (2s, 1H), 2.40 ppm
(s, 3H); MS (MALDI-TOF, dithranol): m/z: 1766 [M+H]+; UV/Vis
(CHCl3): lmax (loge)=629 (4.5), 452 (4.1), 434 (4.1), 312 nm (4.8); FT-IR
(KBr): n˜ =2932, 1596, 1480, 1453, 1279, 1185, 1050, 751, 697 cmÀ1; ele-
mental analysis calcd (%) for C129H49N10OB: C 87.75, H 2.80, N 7.93;
found: C 86.66, H 2.89, N 7.56.
[1] Electron Transfer in Chemistry, Vol. I–V, (Ed.: V. Balzani), Wiley-
VCH, Weinheim, 2001.
[2] Molecular Mechanisms of Photosynthesis (Ed.: R. E. Blankenship),
Blackwell Science, Oxford, 2002.
[3] a) C. Winder, N. S. Sariciftci, J. Mater. Chem. 2004, 14, 1077–1086;
b) S. Guenes, H. Neugebauer, N. S. Sariciftci, Chem. Rev. 2007, 107,
1324–1338.
SubPc-C60 o-1b; C1 isomer (1:1 mixture of diastereoisomers): Compound
o-1b (C1) was obtained as a dark green solid: 49 mg (31%); m.p.
>2508C; 1H NMR (300 MHz, CDCl3): d=8.52 (m, 3H), 8.38 (m, 3H),
7.57 (m, 3H), 7.4–7.0 (m, 31H), 6.70 (m, 2H), 5.10 (m, 1H), 4.74 (d, 2J=
9.4 Hz, 1H), 4.07 (d, 2J=9.4 Hz, 1H), 3.84, 3.80 (2s, 1H), 2.41 ppm (s,
3H); 13C NMR (75.5 MHz, CDCl3): d=155.6, 155.4, 155.2, 153.1, 153.0,
[4] a) D. M. Guldi, N. Martín, J. Mater. Chem. 2002, 12, 1978–1992;
b) D. M. Guldi, H. Imahori, J. Porphyrins Phthalocyanines 2004, 8,
976–983; c) D. M. Guldi, F. Zerbetto, V. Georgakilas, M. Prato, Acc.
Chem. Res. 2005, 38, 38–43; d) D. M. Guldi, G. M.A. Rahman, C.
Ehli, V. Sgobba, Chem. Soc. Rev. 2006, 35, 471–487; e) D. M. Guldi,
Phys. Chem. Chem. Phys. 2007, 9, 1400–1420.
7678
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7670 – 7679

Donor–Acceptor Systems
FULL PAPER
[5] a) J. M. Tour, Chem. Rev. 1996, 96, 537–554; b) M.A. Reed, C.
Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science 1997, 278, 252–
254; c) J. Chen, M.A. Reed, A. M. Rawlett, J. M. Tour, Science
1999, 286, 1550–1552; d) D. K. James, J. M. Tour, Top. Curr. Chem.
2005, 257, 33–62.
[6] T. Hayes, M. R. Wasielewski, D. Gosztola, J. Am. Chem. Soc. 2000,
122, 5563–5567.
[7] W. B. Davis, W. A. Svec, M.A. Ratner, M. R. Wasielewski, Nature
1998, 396, 60–63.
2565; d) T. Fukuda, J. R. Stork, R. J. Potucek, M. M. Olmstead, B. C.
Noll, N. Kobayashi, W. S. Durfee, Angew. Chem. 2002, 114, 2677–
2680; Angew. Chem. Int. Ed. 2002, 41, 2565–2568; for nonlinear op-
tical properties see: e) A. Sastre, T. Torres, M.A. Díaz-García, F.
Agulló-López, C. Dhenaut, S. Brasselet, I. Ledoux, J. Zyss, J. Am.
Chem. Soc., 1996, 118, 2746–2747; f) B. del Rey, U. Keller, T.
Torres, G. Rojo, F. Agulló-López, S. Nonell, C. Martí, S. Brasselet, I.
Ledoux, J. Zyss, J. Am. Chem. Soc. 1998, 120, 12808–12817; g) G.
de la Torre, P. Vµzquez, F. Agulló-López, T. Torres, J. Mater. Chem.
1998, 8, 1671–1683; h) G. de la Torre, P. Vµzquez, F. Agulló-López,
T. Torres, Chem. Rev. 2004, 104, 3723–3750.
[8] a) M. Sirish, R. Kache, B. G. Maiya, J. Photochem. Photobiol. A
1996, 93, 129–136; b) N. V. Tkachenko, H. Lemmetyinen, J. Sonoda,
K. Ohkubo, T. Sato, H. Imahori, S. Fukuzumi, J. Phys. Chem. A
2003, 107, 8834–8844; c) H. Nishikawa, S. Kojima, T. Kodama, I.
Ikemoto, S. Susuki, K. Kikuchi, M. Fujitsuka, L. Hongxia, Y. Araki,
O. Ito, J. Phys. Chem. A 2004, 108, 1881–1890.
[9] a) H. A. Kramers, Physica 1934, 1, 182–192; b) M. Bixon, J. Jortner,
M. E. Michel-Beyerle, A. Ogrodnik, Biochim. Biophys. Acta 1989,
977, 273–286.
[10] a) D. Gust, T. A. Moore, A. L. Moore, C. Devadoss, P. A. Liddell, R.
Hermant, R. A. Nieman, L. J. Demanche, J. M. DeGraziano, I.
Gouni, J. Am. Chem. Soc. 1992, 114, 3590–3603; b) A. Harriman,
F. M. Romero, R. Ziessel, J. Phys. Chem. A 1999, 103, 5399–5408.
[11] a) H. Imahori, K. Hagiwara, M. Aoki, T. Akiyama, S. Taniguchi, T.
Okada, M. Shirakawa, Y. Sakata, J. Am. Chem. Soc. 1996, 118,
11771–11782; b) S. Shinoda, H. Tsukube, Y. Nishimura, I. Yamaza-
ki, A. Osuka, Tetrahedron 1997, 53, 13657–13666; c) A. Stockmann,
J. Kurzawa, N. Fritz, N. Acar, S. Schneider, J. Daub, R. Engl, T.
Clark, J. Phys. Chem. A 2002, 106, 7958–7970; d) S. Shaakov, T.
Galili, E. Stavitski, H. Levanon, A. Lukas, M. R. Wasielewski, J.
Am. Chem. Soc. 2003, 125, 6563–6572; e) A. L. Thompson, T.-S.
Ahn, K. R. J. Thomas, S. Thayumanavan, T. J. Martínez, C. J.
Bardeen, J. Am. Chem. Soc. 2005, 127, 16348–16349.
[17] SubPcs are the subject of increasing attention as electron and/or
energy donors/acceptors in molecular systems. a) D. Gonzµlez-Ro-
dríguez, T. Torres, D. M. Guldi, J. Rivera, L. Echegoyen, Org. Lett.
2002, 4, 335–338; b) D. Gonzµlez-Rodríguez, T. Torres, M.A. Her-
ranz, J. Rivera, L. Echegoyen, D. M. Guldi, J. Am. Chem. Soc. 2004,
126, 6301–6313; c) D. Gonzµlez-Rodríguez, C. G. Claessens, T.
Torres, S. Liu, L. Echegoyen, N. Vila, S. Nonell, Chem. Eur. J. 2005,
11, 3881–3893; d) D. Gonzµlez-Rodríguez, T. Torres, M. M. Olm-
stead, J. Rivera, M.A. Herranz, L. Echegoyen, C. Atienza Castella-
nos, D. M. Guldi, J. Am. Chem. Soc. 2006, 128, 10680–10681;
e) R. S. Iglesias, C. G. Claessens, G. M.A. Rahman, M.A. Herranz,
D. M. Guldi, T. Torres, Tetrahedron 2007, 63, 12396–12404; f) M. E.
El-Khouly, S. H. Shim, Y. Araki, O. Ito, K.-Y. Kay, J. Phys. Chem. B
2008, 112, 3910–3917; g) A. Medina, C. G. Claessens, G. M.A.
Rahman, A. M. Lamsabhi, O. Mo, M. YaÇez, D. M. Guldi, T. Torres,
Chem. Commun. 2008, 1759–1761.
[18] a) J. F. Hartwig, Angew. Chem. 1998, 110, 2154–2177; Angew. Chem.
Int. Ed. 1998, 37, 2046–2067; b) B. H. Yang, S. L. Buchwald, J. Orga-
nomet. Chem. 1999, 576, 125–146.
[19] C. G. Claessens, D. Gonzµlez-Rodríguez, B. del Rey, T. Torres, G.
Mark, H.-P. Schuchmann, C. von Sonntag, J. G. MacDonald, R. S.
Nohr, Eur. J. Org. Chem. 2003, 2547–2551.
[12] J. J. Sunner, S. E. Creager, J. Am. Chem. Soc. 2000, 122, 11914–
11920.
[13] a) K. M. Gaab, A. L. Thompson, J. Xu, T. J. Martinez, C. J. Bardeen,
J. Am. Chem. Soc. 2003, 125, 9288–9289; b) A. L. Thompson, K. M.
Gaab, J. Xu, C. J. Bardeen, T. J. Martinez, J. Phys. Chem. A 2004,
108, 671–682.
[20] a) M. Prato, M. Maggini, Acc. Chem. Res. 1998, 31, 519–526; b) N.
Tagmatarchis, M. Prato, Synlett 2003, 6, 768–779.
[21] Carrying out the palladium-catalyzed amination reaction on SubPc-
C60 compounds 1a did not yield any product, presumably by inacti-
vation of the catalyst due to coordination to the C60 unsaturated
sphere; see: A. L. Balch, M. M. Olmstead, Chem. Rev. 1998, 98,
2123–2165.
[22] The electrochemical and photophysical studies were performed on a
1:3 mixture of C3/C1 regioisomers, due to their almost identical phys-
ical properties.
[23] Structural modeling was performed using the Hyperchem 6.03 pack-
age (Hypercube, Inc.) for Windows.
[24] Detection of the subphthalocyanine fluorescence was hampered
in some cases (dyads 1a) by the instrumental time resolution
(< 100 ps).
[25] When turning to time-resolved fluorescence measurements the only
detectable component was that of C60 with lifetimes that in o-1a, m-
1a, and p-1a are superimposable to that of 5; however, no such C60
component was seen in o-1b, m-1b, and p-1b.
[26] a) D. M. Guldi, Pure Appl. Chem. 2003, 75, 1069–1075; b) H. Ima-
hori, Org. Biomol. Chem. 2004, 2, 1425–1433; c) S. Fukuzumi,
Funct. Org. Mater. 2007, 465–510.
[27] a) D. M. Guldi, A. Gouloumis, P. Vµzquez, T. Torres, V. Georgakilas,
M. Prato, J. Am. Chem. Soc., 2005, 127, 5811–5813; b) M. Quintilia-
ni, A. Kahnt, T. Wçlfle, W. Hieringer, P. Vµzquez, A. Gçrling, D. M.
Guldi, T. Torres, Chem. Eur. J. 2008, 14, 3765–3775.
[28] Notable is, nevertheless, that the charge recombination is not quan-
titative. For example, at the end of the femtosecond timescale a re-
sidual absorbance persists that lasts well into the nanosecond
regime. The overall yield of this long-lived product is about 10%.
[14] a) D. E. Richardson, H. Taube, J. Am. Chem. Soc. 1983, 105, 40–51;
b) V. Marvaud, J.-P. Launay, C. Joachim, Chem. Phys. 1993, 177, 23–
30; c) B. P. Paulson, L.A. Curtiss, B. Bal, G. L. Closs, J. R. Miller, J.
Am. Chem. Soc. 1996, 118, 378–387; d) C. Patoux, C. Coudret, J.-P.
Launay, C. Joachim, A. Gourdon, Inorg. Chem. 1997, 36, 5037–
5049; e) C. Patoux, J.-P. Launay, M. Beley, S. Chodorowsky-Kimmes,
J.-P. Collin, S. James, J.-P. Sauvage, J. Am. Chem. Soc. 1998, 120,
3717–3725; f) C. Rovira, D. Ruíz-Molina, O. Elsner, J. Vidal-Gance-
do, J. Bonvoisin, J.-P. Launay, J. Veciana, Chem. Eur. J. 2001, 7, 240–
250; g) V. Lloveras, J. Vidal-Gancedo, D. Ruíz-Molina, T. M. Fig-
ueira-Duarte, J.-F. Nierengarten, J. Veciana, C. Rovira, Faraday Dis-
cuss. 2006, 131, 291–305.
[15] a) A. Meller, A. Ossko, Monatsh. Chem. 1972, 103, 150–155; semi-
nal work in this area was done by N. Kobayashi, see, for example:
b) N. Kobayashi, R. Kondo, S. Nakajima, T. Osa, J. Am. Chem. Soc.
1990, 112, 9640–9641; relevant reviews can be found in: c) M.
Geyer, F. Plenzig, J. Rauschnabel, M. Hanack, B. del Rey, A. Sastre,
T. Torres, Synthesis 1996, 1139; d) C. G. Claessens, D. Gonzµlez-Ro-
dríguez, T. Torres, Chem. Rev. 2002, 102, 835–853; e) “Synthesis and
Spectroscopic Properties of Phthalocyanine Analogs”: N. Kobayashi,
in The Porphyrin Handbook Vol. 15, (Eds.: K. M. Kadish, K. M.
Smith, R. Guilard), Academic Press, San Diego, CA, 2003, pp. 161–
262; f) T. Torres, Angew. Chem. 2006, 118, 2900–2903; Angew.
Chem. Int. Ed. 2006, 45, 2834–2837; g) G. de la Torre, C. G. Claes-
sens, T. Torres, Chem. Commun. 2007, 2000–2015.
[16] a) N. Kobayashi, T. Ishizaki, K. Ishii, K. Konami, J. Am. Chem. Soc.
1999, 121, 9096–9110; b) C. G. Claessens, T. Torres, Tetrahedron
Lett. 2000, 41, 6361–6365; c) C. G. Claessens, T. Torres, Angew.
Chem. 2002, 114, 2673–2677; Angew. Chem. Int. Ed. 2002, 41, 2561–
Received: May 13, 2008
Published online: July 15, 2008
Chem. Eur. J. 2008, 14, 7670 – 7679
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7679