Synthesis, conformational interconversion, and photophysics of tethered porphyrin-fullerene dyads with parachute topology

Article abstract of DOI:10.1002/chem.200900587

The synthesis of a porphyrin-fullerene dyad with "parachute" topology is reported. To determine whether the dyad is "flexing" at room temperature, low-temperature NMR experiments were used. Computational modeling has shown the low-energy conformation of the dyad to be nonsymmetric. Although, ¹H NMR spec-troscopy at room temperature is consistent with a molecule with C_2v symmetry, the spectrum changes on lowering the temperature consistent with "wind-shield wiper"-like motion, in which the porphyrin moiety rotates from one side of the C₆₀ sphere to the other. Nanosecond and picosecond fluorescence lifetime experiments show two components contribute to the fluorescence decay, also consistent with the presence of more than one conformer. 2009 Wiley-VCH Verlag GmbH & Co. KGaA.

Full text of DOI:10.1002/chem.200900587

DOI: 10.1002/chem.200900587
Synthesis, Conformational Interconversion, and Photophysics of Tethered
Porphyrin–Fullerene Dyads with Parachute Topology
Michael A. Fazio,[a] Alexander Durandin,[a] Nikolai V. Tkachenko,[b] Marja Niemi,[b]
Helge Lemmetyinen,[b] and David I. Schuster*[a]
Abstract: The synthesis of a porphy-
rin–fullerene dyad with “parachute”
topology is reported. To determine
whether the dyad is “flexing” at room
temperature, low-temperature NMR
experiments were used. Computational
modeling has shown the low-energy
conformation of the dyad to be non-
symmetric. Although, 1H NMR spec-
troscopy at room temperature is consis-
tent with a molecule with C2v symme-
try, the spectrum changes on lowering
the temperature consistent with “wind-
shield wiper”-like motion, in which the
porphyrin moiety rotates from one side
of the C60 sphere to the other. Nano-
second and picosecond fluorescence
lifetime experiments show two compo-
nents contribute to the fluorescence
decay, also consistent with the presence
of more than one conformer.
Keywords:
chromophores
·
conformation analysis · fullerenes ·
photophysics · porphyrinoids
Introduction
In porphyrin–fullerene systems that possess flexible linkers,
the two chromophores often assume a face-to-face confor-
mation due to p–p interactions.[1–5] Previously, we designed
and synthesized a dyad with the intent of enforcing a face-
to-face orientation of the porphyrin and fullerene moiet-
ies.[6–10] Due to its shape, in which the porphyrin lies above
the fullerene and is tethered by a strap attached at one
À
carbon carbon bond of the fullerene sphere, the dyad 1 re-
sembles a parachute attached to a ball. Photoinduced
charge separation (CS) dynamics in 1 were inconsistent with
through-bond electron transfer (ET) due to the distance the
electron would have to travel in this system, suggesting that
a through-space ET process was involved. The CS rate was
4.3ꢀ1010 sÀ1 for the free-base dyad in tetrahydrofuran
(THF) and was even higher (1.1ꢀ1011 sÀ1) for the zinc dyad
in toluene.[7] These rates are very similar to the CS rates de-
termined for a similar dyad with cyclophane topology 2, sug-
gesting the donor and acceptor moieties in 1 were indeed
adopting a face-to-face conformation.[11] However, for a
series of cyclophane analogues even higher CS rates (7.7–
14ꢀ1011 sÀ1) were measured via initial formation (100–230ꢀ
1011 sÀ1) of exciplex transient states.[12] Molecular modeling
of 1 suggested that in its lowest energy conformation the
porphyrin bends over the attaching trap to be in closer prox-
imity to the fullerene (Figure 1).[9] This conformation brings
the porphyrin in van der Waals contact distance with the
fullerene, which would facilitate through-space ET.
[a] M. A. Fazio, Dr. A. Durandin, Prof. D. I. Schuster
Department of Chemistry, New York University
100 Washington Square East (USA)
Fax : (+1)212-260-7905
E-mail: david.schuster@nyu.edu
[b] Prof. N. V. Tkachenko, Dr. M. Niemi, Prof. H. Lemmetyinen
Department of Chemistry and Bioengineering
Tampere University of Technology
Dyad 1 showed two exponential decay components by
time-resolved fluorescence studies carried out by the single-
photon counting method. The free-base dyad has lifetimes
33101 Tampere (Finland)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900587.
7698
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7698 – 7705

FULL PAPER
Figure 2. New parachute dyads 3 and 7.
Figure 1. Low-energy conformation of parachute dyad 1.
Results and Discussion
of 14 and 57 ps in benzene, with amplitude factors of 0.904
and 0.095, respectively, and 23 and 80 ps in THF, with ampli-
tude factors of 0.819 and 0.158, respectively.[7,9] These num-
bers suggest more than one component contributes to the
fluorescence decay of the porphyrin singlet excited state.
Time-resolved electron paramagnetic resonance (TREPR)
data on dyad 1 were clearly inconsistent with the extended
conformation of this material, but could be explained by as-
suming that this dyad has the bent unsymmetrical conforma-
tion shown in Figure 1.[10]
Synthesis: Due to the necessary change in substituents on
the porphyrin ring, a different synthetic approach was un-
dertaken. While the synthesis of 3 was much simpler than
the method used for 1, the overall yield was lower. Thus, it
took six steps just to produce the pyrrole that was used to
make 1, with an overall yield of 45%.[13,14] This long synthe-
sis was necessary to produce a porphyrin with ethyl groups
at the beta positions. Since the porphyrin moiety in dyad 3
has only hydrogen at each beta position, it can be gained di-
rectly from commercially available pyrrole, thus eliminating
many steps. A malonate-linked strap 4 was used in both syn-
theses. Synthesis of 1 proceeded via a malonate-connected
bis-dipyrromethane, with an overall yield of 4.6%.[9,13] How-
ever, for the synthesis of 3 3,5-di-tert-butylphenyl-dipyrro-
methane (5) was prepared separately and combined with
the malonate strap to form the porphyrin.
The detailed route to the porphyrin unit 6 of dyad 3 is
shown in Scheme 1. Synthesis of bis-3-(2-formylphenoxy)-
propyl malonate (4) started with salicylaldehyde heated at
reflux with potassium hydroxide and 3-bromopropanol to
produce 2-(3-hydroxypropoxy)benzaldehyde that was then
coupled with malonic acid by using N,N’-dicyclohexylcarbo-
diimide (DCC). Dipyrromethane 5 was prepared from fresh-
ly distilled pyrrole and 3,5-di-tert-butylbenzaldehyde by a
previously published procedure.[15] A trichloroacetic acid
(TCA)-catalyzed condensation reaction of 4 and 5 followed
by oxidation with 2,3-dichloro-5,6-dicyano-p-benzoquinone
(DDQ) gave the desired porphyrin, which was purified by
column chromatography by using dichloromethane (DCM)
on silica gel. The yield of the purified porphyrin was only
1.3%. Zinc metal was inserted in quantitative yield by heat-
ing a solution of the porphyrin in DCM at reflux for 15 min
and then adding zinc acetate dihydrate to give 6. Purifica-
tion of the metalated porphyrin was accomplished by pass-
ing it over a pad of silica gel and eluting with DCM/triethyl-
amine.
In the low-energy conformation, the meso carbons on the
porphyrin ring occupy different chemical environments and
1
should be distinguishable in H and 13C NMR. However, at
room temperature they both have the same 1H chemical
shift of d=10.33 ppm. The NMR data can be explained by
assuming the conformation of the dyad is not static at ambi-
ent temperatures, but is moving, not unlike a windshield
wiper, from the low-energy conformation with the porphyrin
close to one hemisphere of the fullerene to that with the
porphyrin in the same position on the opposite hemisphere.
By this conformational interconversion, the porphyrin
would pass through the extended structure with C2v symme-
try, which might be a maximum or a shallow well on the po-
tential energy surface The windshield wiper-like conforma-
tional interconversion would place the meso carbons of the
porphyrin ring in the same chemical environment for the
same amount of time throughout each period of motion.
Slowing down the movement by reducing the temperature
should, in principle, reveal differences in chemical environ-
ment on a timescale measurable by NMR spectroscopy.
The windshield wiper proposal would also explain the two
exponential decay lifetimes, with the shorter lifetime attrib-
utable to the more populated low-energy conformation and
the longer lifetime attributable to the less populated extend-
ed conformation, assuming it lies in a shallow potential well.
1
To test this theory by variable-temperature H NMR spec-
troscopy, the dyad had to be structurally modified to allow
greater solubility at low temperatures. The new porphyrin
dyad 3 (see Figure 2) has substituted phenyl rings at the
meso positions of the porphyrin, which were unsubstituted
in the original parachute dyad 1.
In the original synthesis of 1,[6,7] the four separate steps
needed to convert malonate 4 to the corresponding tetrae-
thylporphyrin product proceeded in 8.3% yield. The experi-
mentally simpler and less time consuming method used to
prepare 6 from 4 is a three-step sequence with a single pu-
Chem. Eur. J. 2009, 15, 7698 – 7705
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7699

D. I. Schuster et al.
Scheme 1. Strapped porphyrin synthesis.
1
rification step. However, while experimentally more effi-
cient, the overall yield is very low, only 1.3%. Attempts to
improve or optimize the yields were not pursued.
(see the Supporting Information). The H NMR spectrum of
dyad 3 is very similar to that of porphyrin 6 with some
peaks shifted downfield due to the deshielding effect of the
fullerene. Dyad 3 shows m/z: 1835.1 [M]+ (calcd m/z:
1834.5).
1
Porphyrin 6 was characterized by H NMR spectroscopy
1
and mass spectrometry. Peaks in the H NMR spectrum in-
dicative of porphyrin formation are the shift of the malonyl
a protons from d=3.4 ppm in the bis-aldehyde 4 to d=
1.1 ppm in porphyrin 6 due to the shielding effect of the
nearby porphyrin. The 3,5-di-tert-butylphenyl (di-tBuPh)
groups are too bulky to allow free rotation around the con-
necting bonds to the porphyrin. Thus, one tert-butyl group
lies above the porphyrin and the other lies beneath, as evi-
denced by the two peaks for the tBu protons at d=1.5 ppm
and d=1.6 ppm in the 1H NMR spectrum of 6. The aromatic
region has peaks at d=7.8, 8.0, and 8.2 ppm for the three
phenyl protons on the di-tBuPh groups present at the meso
positions of 6. Similarly, the two ortho protons on the
phenyl ring are in different environments, one above and
the other below the porphyrin ring, and are distinguishable
by NMR spectroscopy. Mass was determined by MALDI-
TOF mass spectrometry. In positive mode, porphyrin 6 has a
single peak at m/z: 1116.9 [M]+ (calcd m/z: 1116.5).
Photophysical studies: The UV/Vis spectrum of dyad 3 in
Figure 3 appears as a simple superimposition of the compo-
nent zinc porphyrin (ZnP) and fullerene, with a strong Soret
band at 423 nm in DCM and 427 nm in toluene, multiple Q
bands between 500 and 600 nm, and the fullerene absorption
Dyad 3 was made from porphyrin 6 utilizing the Bingel–
Hirsch reaction.[16,17] One equivalent of porphyrin 6 and five
equivalents of C60 were dissolved in toluene, carbon tetra-
bromide (1.2 equiv) and 1,8-diazabicycloACTHNUTRGNE[NUG 5.4.0]undec-7-ene
(DBU) (4 equiv) were added to this solution, and the mix-
ture was allowed to stir overnight. Thin-layer chromatogra-
phy using cyclohexane followed by DCM gave target dyad 3
in about 30% yield. The dyad was characterized by
MALDI-TOF mass spectrometry and NMR spectroscopy
Figure 3. UV/Vis spectra of C60 (solid gray line), porphyrin 6 (dashed
line) and dyad 3 (solid black line), all 4 mm in DCM.
7700
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7698 – 7705

Tethered Porphyrin–Fullerene Dyads with Parachute Topology
FULL PAPER
around 330 nm. The spectrum indicates that there is no ap-
preciable electronic communication between the two chro-
mophores in the ground state.
The steady-state fluorescence spectra of 6 and 3 were ob-
tained by exciting the samples at 424 nm in the region of the
Soret band and monitoring the emission from 450 to
800 nm. The dyad shows significant fluorescence quenching
in both toluene (see the Supporting Information) and DCM
(see Figure 4).
Figure 5. Fluorescence decay of 3 in toluene after excitation at 410 nm.
Figure 4. Fluorescence spectra of porphyrin 6 (dashed line) and dyad 3
(solid line), both 4 mm in DCM. Samples were excited at 424 nm.
Time-resolved absorption and fluorescence measurements
were performed in polar (benzonitrile, PhCN) and nonpolar
(toluene) solvents. The up-conversion (fluorescence) experi-
ments were carried out at an excitation wavelength of
410 nm, corresponding to the Soret band, which yields the
second ZnP singlet excited state. The fluorescence decay
was monitored at 615 nm where the ZnP S1 state emits. The
formation and decay curves of the dyad in toluene and
PhCN are presented in two different time-scales to better
resolve the formation of the S1 state of ZnP from the S2
state (shorter time scale) and to determine the fluorescence
lifetimes (longer timescale).
The fluorescence decay shown in Figure 5 is biexponen-
tial. The first decay component has a lifetime of 8 ps with an
amplitude factor of 0.661. The second decay component has
a lifetime of 29 ps with an amplitude factor of 0.339. It can
be seen that formation of the first singlet excited state on
excitation of ZnP at 410 nm takes place in 785 fs. This is
better resolved in the shorter time-scale study in Figure 6,
which shows the time for formation of the first ZnP excited
state by decay of S2 is 927 fs. These lifetime rates of 1.3ꢀ
1011 sÀ1 and 0.34ꢀ1011 sÀ1 correspond to the formation of a
CS state and are consistent with the formation rates of more
rigid cyclophane analogues.[12] Dyad 3 also displays two fluo-
rescence decay components in PhCN with lifetimes of 7.5 ps
and 26 ps, and amplitude factors of 0.49 and 0.51, respective-
Figure 6. Fluorescence formation of
410 nm.
3 in toluene after excitation at
ly. The first ZnP singlet excited state of 3 is formed from S2
in about 800 fs (see the Supporting Information).
The pump-probe (transient absorption) experiments on 3
were carried out at the excitation wavelength of 565 nm in
the Q band region of the porphyrin and were monitored in
the ranges of 490–710 nm and 850–1070 nm for the solution
in toluene, and 570–780 nm and 850–1070 nm for the solu-
tion in benzonitrile. The shift of the monitoring range to the
visible region in case of the benzonitrile solution was neces-
sary due to the excitation light causing too much disturb-
ance in measurements below 570 nm.
Figure 7 shows the calculated time-resolved component
spectra immediately after excitation (0 ps) and the decay
component spectrum in toluene, as obtained from global
fits. In both solvents, a two-exponential decay was required
to obtain reasonable fits to the data. Charge separation is
evident by the formation of porphyrin radical cation and C60
radical anion, as shown by negative amplitude in the signals
at 650 nm and 1040 nm, respectively. Extremely rapid
charge separation happens in both polar and non-polar sol-
Chem. Eur. J. 2009, 15, 7698 – 7705
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7701

D. I. Schuster et al.
presence of both the folded (low energy) and extended
(high energy) conformations. Conversely, raising the temper-
ature should also give rise to spectral changes: most notably
the tert-butyl peaks would be expected to merge as the
phenyl groups to which they are attached rotate more rapid-
ly around the connecting bond to the porphyrin, releasing
the tBu groups from being restricted to residing above or
below the porphyrin ring system, respectively.
[D8]Toluene was selected as solvent for this study because
of its higher boiling point as well as its lower melting point
compared to deuterated chloroform (CDCl3). The spectrum
of dyad 3 was first run in [D8]toluene at ambient tempera-
ture (300 K). Nearly all the peaks experienced a shift com-
pared to the same compound in CDCl3, and most peaks are
shifted downfield. The spectra of porphyrin 6 and dyad 3
both experienced similar shifts for all peaks, with the nota-
ble exception of the peaks corresponding to Hk (see spectra
and peak assignments in Figure 8). This peak was shifted up-
field only slightly for porphyrin 6 when switching from
CDCl3 to [D8]toluene, but was shifted downfield by nearly
0.5 ppm in the case of dyad 3.
Figure 7. Transient component spectra of dyad 3 in toluene. Sample was
~
&
excited at 565 nm. (c: 0 ps, calcd; : >2 ps; : 9 ps).
vents and occurs in nearly the same timescale (7.4 ps in
PhCN, 9 ps in toluene). These results are in quite good
agreement with values obtained for fluorescence decay by
up-conversion (7.5 ps in PhCN, 8 ps in toluene). The lifetime
of the charge-separated state in
PhCN solution was about
200 ps, while in toluene the CS
lifetime outlived the instru-
mentꢂs detection range of 2 ns.
The longer CS lifetime in the
less polar solvent is diagnostic
of back electron transfer
(charge recombination) in the
Marcus-inverted region.[18]
Limited studies were carried
out on the unsubstituted para-
chute dyad 7, namely structure
3 with R=H. This dyad was
made from 4 by the route de-
picted in Scheme 1 using the
unsubstituted dipyrromethane
instead of 5 (see the Supporting
Information for details). Dyad
7 was excited at 400 nm and the
fluorescence was monitored at
630 nm. Porphyrin 8 (porphyrin
6 with R=H and H2 in place of Figure 8. 1H NMR spectrum of 3 in [D8]toluene at 300 K.
Zn) has a single decay compo-
nent with a lifetime of 9.5 ns
and amplitude factor of 1.0. Dyad 7 has three decay compo-
nents with lifetimes of 136 ps, 970 ps, and 6.9 ns, and ampli-
tude factors of 0.958, 0.031, and 0.01 respectively.
As the temperature was raised to 323 K, identical changes
were seen in the spectra of both dyad 3 and porphyrin 6
(see the Supporting Information). As predicted, the peaks
associated with the tBu groups moved closer together, as did
the peaks associated with the protons on the phenyl groups
to which the tBu groups are attached. Both spectra also
showed movement of the peaks associated with Hj and Hk
in the porphyrin strap, specifically, they moved apart as the
temperature was raised. The peaks separated further in the
spectra of the dyad and the porphyrin when the temperature
was increased further to 348 K.
Temperature-dependent 1H NMR studies on dyad 3: As
mentioned earlier, temperature-dependent NMR spectra
should reveal conformational changes in the dyad with re-
spect to the relative positions of the porphyrin and fullerene
moieties. If the porphyrin is swinging from one hemisphere
of the fullerene to the other, lowering the temperature
might slow down the movement sufficiently to reveal the
7702
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7698 – 7705

Tethered Porphyrin–Fullerene Dyads with Parachute Topology
FULL PAPER
As the temperature was lowered from ambient tempera-
ture, a reverse of the trend mentioned above was noticed.
The most interesting changes occurred for the protons in the
strap of the dyad. When the dyad was cooled to 233 K,
there was only one peak for Hj and Hk, appearing at d=
3.09 ppm (see Figure 9, top). Upon further cooling to 223 K,
Hk was shifted further downfield than Hj (Figure 9, bottom).
At 223 K and even at 198 K, Hk is still upfield of Hj in
porphyrin 6 (see the Supporting Information). Clearly, the
proximity of the fullerene to the Hk protons is affecting the
Hk chemical shift. Although there is no clear splitting of
protons closer to C60 and protons further away from C60
when in the low energy conformation, such as the porphyrin
beta protons, the dramatic downfield shifting of the strap
protons Hk suggest that the dyad is increasingly in a bent
Figure 9. 1H NMR spectra of dyad 3 in [D8]toluene at 233 K (top) and 223 K (bottom).
Chem. Eur. J. 2009, 15, 7698 – 7705
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7703

D. I. Schuster et al.
conformation (however, see below). From the model shown
in Figure 10, we see that as the porphyrin moiety gets closer
to the fullerene, an axis perpendicular to the plane of the
show that these new dyads 3 and 7 exhibit significant
quenching of porphyrin fluorescence and extremely rapid
charge separation in both polar and non-polar solvents. The
high rate constants for charge separation in 3 and 7,
>1011 sÀ1, are very similar to those previously reported for
the original parachute dyad 1,[9] and are most readily ex-
plained by a through-space rather than a through-bond ET
mechanism. Porphyrin–fullerene dyads lacking this topology
typically display rate constants for charge separation that
are one to two orders of magnitude lower.[15,18] Charge re-
combination, for example back electron transfer, in 1, 3, and
7 is much slower by several orders of magnitude than charge
separation, and clearly occurs in the Marcus-inverted region
for electron transfer processes.[9,19] The meso-substituted as
well as unsubstituted dyads 3 and 7 show two fluorescence
decay components, attributed to the bent and the extended
conformations of these dyads. Temperature-dependent
1H NMR spectra, along with photophysical and TREPR
data of these dyads, are most consistent with conformational
interconversion by a windshield wiper-like motion of the
strapped porphyrin from one side of the fullerene sphere to
the other through the extended conformation, although
other explanations cannot be absolutely excluded.
Figure 10. In the low-energy form determined computationally, the por-
phyrin has less of a shielding effect on the connecting strap.
porphyrin and through its center is directed toward the top
of the fullerene and away from the malonate strap (see
Figure 10). The shielding effect of the porphyrin on the
strap is therefore diminished while the deshielding effect of
the fullerene shifts protons Hk downfield to a larger extent
than can be attributed to solvent or temperature alone.
Lower temperatures were attempted in the NMR experi-
ments (down to 208 K for dyad 3 and 183 K for porphyrin
6), at which point the spectra showed drastically broadened
peaks; the spectral range containing Hj and Hk became one
broad hump, making it impossible to distinguish between
the peaks.
However, as pointed out by a referee, the change with
temperature of the 1H NMR spectra of the flexible poly-
ether linkers in toluene could be some sort of solvation
effect and is not in itself proof of the proposed windshield
wiper-like motion of dyads 1 and 3. However, it must be em-
phasized that the photophysical and low-temperature
TREPR data[10] are inconsistent with a predominantly ex-
tended conformation of these dyads, even though the
1H NMR data at and above room temperature are not what
one would expect for the low-energy bent conformation
shown in Figure 1. Thus, some kind of conformational
motion through a potential barrier must be occurring in
these systems, of which the windshield wiper proposal seems
to us the most reasonable explanation consistent with all the
experimental evidence determined thus far.
Experimental Section
General procedure: Materials not synthesized were purchased from Al-
drich and used without further purification unless otherwise noted.
1H NMR spectra were obtained by using a 400 MHz Bruker-AV spec-
trometer or a Varian Gemini 200 MHz spectrometer. They were mea-
sured in CDCl3 unless otherwise noted, and are reported relative to a tet-
ramethylsilane internal standard (0.1% v/v). Mass spectra were obtained
by using a Bruker Omni-Flex MALDI-TOF mass spectrometer. Mass
spectra were obtained without using a matrix. UV/VIS spectra were re-
corded by using a Perkin–Elmer Lambda-35 UV/VIS spectrophotometer.
Steady-state fluorescence spectra were obtained by using a Hitachi F-
2500 fluorescence spectrophotometer. Infrared spectra were recorded by
using a Thermo Nicolet Avatar 360 FTIR spectrophotometer. Pump-
probe and up-conversion techniques for time-resolved absorption and
fluorescence, respectively, were used to detect the fast processes with a
time resolution shorter than 0.2 ps. The instrumentation and the data
analysis procedure have been described earlier.[20,21]
2-(3-Hydroxypropoxy)benzaldehyde:
Potassium
hydroxide
(3.8 g,
56.3 mmol, 85% purity) was dissolved in EtOH (60 mL). The flask was
attached to a reflux condenser with a rubber septum on top and purged
with nitrogen. Salicylaldehyde (6 mL, 56.3 mmol) was added through the
septum while stirring the solution, which turned yellow and then formed
crystals. 3-Bromo-1-propanol (6 mL, 67.6 mmol) was also added through
the septum. A nitrogen filled balloon was fitted to the septum and the re-
action mixture was heated at reflux for 19 h. The mixture was cooled to
room temperature and then filtered through a vacuum funnel, and the
collected solids were washed with acetone. The liquid filtrate was washed
with a saturated aqueous solution of potassium carbonate and then ex-
tracted with CH2Cl2. The organic layer was washed with water and then
dried over sodium sulfate. The product was purified by column chroma-
tography (silica gel, EtOAc/petroleum 2:1) to afford the desired product
Conclusions
A new route to synthesizing parachute-shaped porphyrin–
fullerene dyads has been developed, which allows more effi-
cient synthesis and incorporation of a wider variety of sub-
stituents on the porphyrin ring. As with the previously stud-
ied dyads with this special topology,[7] photophysical studies
1
as a yellow oil (7.4 g, 73%). H NMR (200 MHz, CDCl3): d=10.4 (s, 1H;
CHO), 7.8 (d, 1H; ArH), 7.5 (m, 1H; ArH), 7.0 (m, 2H; ArH), 4.3 (t,
2H; ArO-CH2), 3.9 (s, 2H; CH2-OH), 3.0 (s, 1H; OH), 2.1 ppm (m, 2H,
CH2-CH2-CH2); MALDI-TOF MS: m/z calcd for C10H12O3: 180.08 [M+];
found 180.16; calcd for mass fragment C7H5O2: 121.03; found 120.40.
7704
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 7698 – 7705

Tethered Porphyrin–Fullerene Dyads with Parachute Topology
FULL PAPER
Bis-3-(2-formylphenoxy)propyl malonate (4): 2-(3-Hydroxypropoxy)ben-
zaldehyde (0.27 g, 1.5 mmol), malonic acid (7.8 mg, 0.75 mmol), and
DCC (0.31 g, 1.5 mmol) were stirred in CH2Cl2 under nitrogen overnight.
Column chromatography (silica gel, CH2Cl2/EtOAc 10:1) gave pure 4 as
a dark yellow oil (0.24 g, 74%). 1H NMR (200 MHz, CDCl3): d=10.5 (s,
2H; CHO), 7.8 (d, 2H; ArH), 7.5 (t, 2H; ArH), 7.0 (m, 4H; ArH), 4.4
(t, 4H; ArO-CH2), 4.1 (t, 4H; CH2-O-CO), 3.4 (s, 2H; CO-CH2-CO),
2.2 ppm (m, 4H; CH2-CH2-CH2); MALDI-TOF MS: m/z calcd for
C23H24O8: 428.15 [M+]; found 428.9.
Acknowledgements
The work at NYU was supported by grants from the National Science
Foundation, the Petroleum Research Fund of the American Chemical
Society, and the Research Challenge Fund of New York University.
[1] P. A. Liddell, J. P. Sumida, A. N. Macpherson, L. Noss, G. R. Seely,
K. N. Clark, A. L. Moore, T. A. Moore, D. Gust, Photochem. Photo-
biol. 1994, 60, 537–541.
[2] H. Imahori, K. Hagiwara, M. Aoki, T. Akiyama, S. Taniguchi, T.
Okada, M. Shirakawa, Y. Sakata, J. Am. Chem. Soc. 1996, 118,
11771–11782.
[3] E. Dietel, A. Hirsch, E. Eichhorn, A. Rieker, S. Hackbarth, B.
Rçder, Chem. Commun. 1998, 1981–1982.
[4] J. P. Bourgeois, F. Diederich, L. Echegoyen, J. F. Nierengarten, Helv.
Chim. Acta 1998, 81, 1835–1844.
[5] For a recent review of interaction of the convex surface of [60]-full-
erene with concave molecular receptors, see: E. M. Pꢃrez, N.
Martꢄn, Chem. Soc. Rev. 2008, 37, 1512–1519.
[6] P. Cheng, S. R. Wilson, D. I. Schuster, Chem. Commun. 1999, 89–90.
[7] 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, 11599–11600.
3,5-Di-tert-butylphenyl-substituted dipyrromethane (5): 3,5-Di-tert-butyl-
benzaldehyde[15] (5 g, 22.9 mmol) was dissolved in pyrrole (39.8 mL, 0.57
mol) with stirring. Trifluoroacetic acid (0.18 mL, 2.3 mmol) was added
dropwise and the mixture was stirred for 30 min. Triethylamine (0.33 mL,
2.3 mmol) was added to quench the reaction. Excess pyrrole was re-
moved under vacuum. Crude product was dissolved in DCM, extracted
with a 0.1m solution of sodium hydroxide and then water, and dried over
sodium sulfate. The solvent was removed under vacuum and the product
was purified by column chromatography (silica gel, cyclohexane/EtOAc/
triethylamine 85:14:1) to give 5 as a dark brown viscous liquid (6.5 g,
85%). 1H NMR (200 MHz, CDCl3): d=8.0 (s, 2H; NH), 7.3 (s, 1H;
ArH), 7.1 (s, 2H; ArH), 6.7 (d, 2H; pyrrole-H), 6.2 (t, 2H; pyrrole-H),
6.0 (d, 2H; pyrrole-H), 5.5 (s, 1H; benzylic-H), 1.3 ppm (s, 18H, tBuH).
Strapped porphyrin 6: Malonate strap 4 (1.27 g, 2.96 mmol) and dipyrro-
methane 5 (1.9 g, 5.7 mmol) were dissolved in DCM (1000 mL) and the
flask was purged with nitrogen. The flask was protected from light for
the duration of the reaction. TCA (0.14 g, 0.83 mmol) was added to the
solution, and the reaction was allowed to proceed for 1 h, after which
time DDQ (2 g, 5.9 mmol) was added to the flask and the reaction was
allowed to stir overnight. The porphyrin was purified by column chroma-
tography (silica gel/DCM) to give the free-base porphyrin. Solvent
volume was reduced under vacuum and zinc acetate dihydrate (1 g in
10 mL methanol) was added and the solution was heated at reflux for
15 min. The product was purified by column chromatography (silica gel,
DCM/triethylamine 100:7) to give compound 6 as a purple solid (42 mg,
1.3%). 1H NMR (400 MHz, CDCl3, 258C): d=8.9 (d, 4H; bH), 8.6 (d,
4H; bH), 8.3 (d, 2H; ArH), 8.25 (s, 2H; ArH), 8.0 (s, 2H; ArH), 7.8 (s,
2H; ArH), 7.7 (t, 2H; ArH), 7.5 (t, 2H; ArH), 7.3 (d, 2H; ArH), 3.6 (t,
4H, ArO-CH2), 2.2 (m, 4H; CH2-CH2-CH2), 1.55 (s, 18H; tBu), 1.50 (s,
18H; tBu), 1.1 (s, 2H; CO-CH2-CO), 1.0 ppm (t, 4H; CH2-O-CO);
MALDI-TOF MS: m/z calcd for C69H72N4O6Zn: 1116.47 [M+]; found
1116.88.
[8] D. I. Schuster, Carbon 2000, 38, 1607–1614.
[9] D. I. Schuster, P. Cheng, P. D. Jarowski, D. M. Guldi, C. Luo, L.
Echegoyen, S. Pyo, A. R. Holzwarth, S. E. Braslavsky, R. M. Wil-
liams, G. Klihm, J. Am. Chem. Soc. 2004, 126, 7257–7270.
[10] T. Galili, A. Regev, H. Levanon, D. I. Schuster, D. M. Guldi, J. Phys.
Chem. A 2004, 108, 10632–10639.
[11] D. M. Guldi, C. Luo, M. Prato, E. Dietel, A. Hirsch, Chem.
Commun. 2000, 373–374.
[12] V. Chukharev, N. V. Tkachenko, A. Efimov, D. M. Guldi, A. Hirsch,
M. Scheloske, H. Lemmetyinen, J. Phys. Chem. B 2004, 108, 16377–
16385.
[13] P. Cheng, PhD Thesis, New York University (USA), 2001.
[14] T. D. Lash, J. R. Bellettini, J. A. Bastian, K. B. Couch, Synthesis
1994, 170–172.
[15] D. I. Schuster, S. MacMahon, D. M. Guldi, L. Echegoyen, S. E. Bras-
lavsky, Tetrahedron 2006, 62, 1928–1936.
[16] C. Bingel, Chem. Ber. 1993, 126, 1957–1959.
Dyad 3: C60 (9.7 mg, 0.0135 mmol) was dissolved in toluene (5 mL) by
stirring under nitrogen. Porphyrin 6 (3 mg, 0.0027 mmol), carbon tetra-
bromide (1.1 mg, 0.0032 mmol), and DBU (1.6 mg, 0.0108 mmol) were
each dissolved in toluene and then added through syringe in the order as
listed. The flask was covered by foil to keep light out and stirred over-
night. The product was purified by prep-TLC first using cyclohexane as
eluent to remove excess C60 and then CH2Cl2 to move the dyad past the
byproducts that stay at the baseline. Dyad 3 was isolated as a brown solid
(1.5 mg, ꢀ30%). 1H NMR (400 MHz, CDCl3, 258C): d=9.05 (d, 4H;
bH), 8.9 (d, 4H; bH), 8.4 (d, 2H; ArH), 8.3 (s, 2H; ArH), 8.06 (s, 2H;
ArH), 7.8 (s, 2H; ArH), 7.7 (t, 2H; ArH), 7.5 (t, 2H; ArH), 7.2 (d, 2H;
ArH), 3.8 (t, 4H; ArO-CH2), 2.2 (m, 4H; CH2-CH2-CH2), 1.57 (s, 18H;
tBu), 1.50 (s, 18H; tBu), 0.86 ppm (t, 4H; CH2-O-CO); MALDI-TOF
MS: m/z calcd for C129H70N4O6Zn: 1834.46 [M+]; found 1835.1.
[17] X. Camps, A. Hirsch, J. Chem. Soc. Perkin Trans. 1 1997, 1595–
1596.
[18] S. A. Vail, P. J. Krawczuk, D. M. Guldi, A. Palkat, L. Echegoyen,
J. P. C. Tomꢃ, M. A. Fazio, D. I. Schuster, Chem. Eur. J. 2005, 11,
3375–3388.
[19] R. A. Marcus, Angew. Chem. 1993, 105, 1161–1172; Angew. Chem.
Int. Ed. Engl. 1993, 32, 1111–1121.
[20] N. V. Tkachenko, L. Rantala, A. Y. Tauber, J. Helaja, P. H. Hynni-
nen, H. Lemmetyinen, J. Am. Chem. Soc. 1999, 121, 9378–9387.
[21] V. Vehmanen, N. V. Tkachenko, H. Imahori, S. Fukuzumi, H. Lem-
metyinen, Spectrochim. Acta Part A 2001, 57, 2229–2244.
Received: March 4, 2009
Published online: June 30, 2009
Chem. Eur. J. 2009, 15, 7698 – 7705
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7705