Pyrazolino[60]fullerene-oligophenylenevinylene dumbbell-shaped arrays: Synthesis, electrochemistry, photophysics, and self-assembly on surfaces

Article abstract of DOI:10.1002/chem.200500089

Symmetrically substituted oligophenylenevinylene (OPV) derivatives bearing terminal p-nitrophenylhydrazone groups have been prepared and used for the synthesis of dumbbell-shaped bis(pyrazolino[60]fullerene)-OPV systems. In these triad arrays, the OPV-typ

Full text of DOI:10.1002/chem.200500089

FULL PAPER
Pyrazolino[60]fullerene–Oligophenylenevinylene Dumbbell-Shaped Arrays:
Synthesis, Electrochemistry, Photophysics, and Self-Assembly on Surfaces
Fernando Langa,*[a] Maria J. Gomez-Escalonilla,[a] Jean-Michel Rueff,[b]
Teresa M. Figueira Duarte,[b] Jean-FranÅois Nierengarten,*[b] Vincenzo Palermo,[c]
Paolo Samorì,*[c, d] Yannick Rio,[c] Gianluca Accorsi,[c] and Nicola Armaroli*[c]
Abstract: Symmetrically substituted
oligophenylenevinylene (OPV) deriva-
tives bearing terminal p-nitrophenylhy-
drazone groups have been prepared
and used for the synthesis of dumbbell-
the corresponding reference fullerene
molecule. The occurrence of electron
transfer in the multicomponent arrays
is evidenced by recovery of fullerene
fluorescence at 77 K in CH2Cl2 and in
toluene at 298 K. Under these condi-
tions the OPV!C60 energy transfer is
unaffected. The rate of this process
turns out to be higher for the OPV
trimer than for the corresponding pen-
tameric OPV arrays, in agreement with
energy-transfer theory expectations.
Scanning tunneling microscopy (STM)
and scanning force microscopy (SFM)
revealed that the bis(pyrazolino[60]ful-
lerene)–OPV can self-assemble into or-
dered layered crystalline architectures
on the basal plane of highly oriented
pyrolitic graphite.
shaped
bis(pyrazolino[60]fullerene)–
OPV systems. In these triad arrays, the
OPV-type fluorescence is dramatically
quenched as a consequence of ultrafast
OPV!C60 singlet energy transfer. In
its turn the fullerene singlet state is
quenched by pyrazoline!C60 electron
transfer, in line with the behavior of
Keywords: electrochemistry · elec-
tron transfer · fullerenes · oligophe-
nylenevinylenes · scanning probe
microscopy
Introduction
Owing to its relatively low reduction potential, symmetrical
shape, large size, and extended p-electron system, [60]fuller-
ene is an ideal candidate as an electron acceptor in photo-
and electroactive materials.[1,2] In particular, the electronic
properties of fullerenes make them attractive in the design
of new materials in which photoinduced electron transfer is
the key process for the conversion of light into chemical[3]
or electric energy.[4–9] The characteristics of C60 successfully
compare with those of electron acceptors with smaller size
such as benzoquinone.[1,10] Indeed, accelerated charge trans-
fer and decelerated charge recombination have been ob-
served in a fullerene-based donor–acceptor systems when
compared to similar arrays in which C60 is replaced by ben-
zoquinone.[10]
[a] Prof. F. Langa, Dr. M. J. Gomez-Escalonilla
Facultad de Ciencias del Medio Ambiente
Universidad de Castilla-La Mancha, 45071 Toledo (Spain)
Fax : (+34)-925-268-840
E-mail: fernando.lpuente@uclm.es
[b] Dr. J.-M. Rueff, T. M. Figueira Duarte, Dr. J.-F. Nierengarten
Groupe de Chimie des Fuller›nes et des Syst›mes ConjuguØs
Ecole EuropØenne de Chimie, Polym›res et MatØriaux (ECPM)
UniversitØ Louis Pasteur et CNRS, 25 rue Becquerel
67087 Strasbourg Cedex 2(France)
Fax : (+33)3-90-242-706
E-mail: jfnierengarten@chimie.u-strasbg.fr
[c] Dr. V. Palermo, Dr. P. Samorì, Dr. Y. Rio, Dr. G. Accorsi,
Dr. N. Armaroli
In principle, functionalization of fullerenes allows us to
tune the electron-acceptor properties of C60 derivatives;[11,12]
however, only a few C60 derivatives turned out to be better
electron acceptors than the parent molecule.[13–16] In this
regard, an important breakthrough was the synthesis of full-
eropyrazolines derivatives.[17,18] The inductive effect of the
pyrazoline nitrogen atom directly attached to the carbon
sphere tends to make them easier to reduce than other C60
Istituto per la Sintesi Organica e la Fotoreattività, Consiglio Nazio-
nale delle Ricerche, Via Gobetti 101 40129 Bologna (Italy)
Fax : (+39)051-639-9844
E-mail: armaroli@isof.cnr.it, samori@isof.cnr.it
[d] Dr. P. Samorì
Nanochemistry Lab
Institut de Science et dꢀIngØnierie SupramolØculaires (I.S.I.S.)
UniversitØ Louis Pasteur, 8 allØe Gaspard Monge
67083 Strasbourg (France)
Chem. Eur. J. 2005, 11, 4405 – 4415
DOI: 10.1002/chem.200500089
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4405

derivatives .[12] Moreover the pyrazoline–C60 unit was found
to act as a charge-separation module for the construction of
photoactive and tunable devices.[19]
In this paper we describe the synthesis, the electrochemis-
try, the photophysical properties and the self-assembly be-
havior in ultra-thin films of OPV–fulleropyrazoline com-
pounds. Two of them, F-3PV and F-5PV, are dyads in which
the C60 moiety is linked to a trimeric (3PV) and a penta-
meric (5PV) oligophenylenevinylene unit, respectively. The
other two, F-3PV-F and F-5PV-F, are triads made of an
OPV central unit (3PV and 5PV) bearing a fullerene frag-
ment at both extremities. Both the dyad and the triad sys-
tems show photoinduced energy- (OPV!C60) and electron-
transfer (pyrazoline!C60) processes. Detailed SFM and
STM studies on the largest system F-5PV-F reveal the spon-
taneous formation from solutions of highly ordered layer ar-
chitectures on surfaces and, on the molecular scale, the self-
assembly into polycrystalline structures characterized by a
few nanometer-sized crystalline domains. A preliminary
report on the preparation of F-3PV, F-5PV, F-3PV-F, and
F-5PV-F was published previously.[50]
To take advantage of their original electronic properties,
C60 and its derivatives have been linked to a variety of or-
ganic conjugated oligomers,[20] such as oligophenylenevinyl-
enes (OPV),[19,21–26] oligophenyleneethynylene (OPE),[27] oli-
gothiophenes (OTP),[28–31] and others.[25,32–36] These dyad and
triad systems exhibit interesting intramolecular photopro-
cesses between the conjugated linear backbone and the
carbon sphere, that is, electron or energy transfer. Some of
them are used as active materials in prototype photovoltaic
devices.[21,27,37] In the OPV–C60 arrays described so far, the
desired OPV!C60 electron-transfer process undergoes com-
petition from an efficient OPV!C60 singlet–singlet energy
transfer.[21,22,24,38,39] Practically, in solution, the OPV unit acts
as a light collecting (i.e., antenna) module, whilst electron
transfer, the extent of which can be modulated by solvent
polarity, occurs only from the lowest electronic singlet state
of the C60 moiety.[19,40]
The exploitation of the fullerene-based molecular systems
for the fabrication of optoelectronic devices requires the
processing of the molecular architecture in thin films. The
performance of these devices strongly depends on the inter-
play between electronic structure and arrangement at the
supramolecular level.[41] Achieving full control of the self-as-
sembly at surfaces of p-conjugated molecules into complex
supramolecular architectures is therefore of paramount im-
portance. Scanning probe microscopy techniques make it
possible to map a surface by probing different physicochem-
ical properties on the nano- and micrometer scale. Scanning
force microscopy (SFM) and scanning tunneling microscopy
(STM) studies of fullerenes have been performed on many
different surfaces, at both large and small length scales.
They showed the formation of layers[42] or islands of mole-
cules, sometimes with a preferential island growth at surface
step edges,[43] exhibiting hexagonal packing.[44,45] When the
fullerene was linked to ferrocene, the molecule was found to
physisorb on graphite into ribbons, in which the single units
are packed into dimers, as ruled by the electric dipole of the
molecules.[46] On the other hand, making use of the Lang-
muir–Blodgett technique, a C60 molecule functionalized with
an aliphatic tail, bearing a polar head-group, was self-assem-
bled perpendicular to the substrate producing closely spaced
arrangements.[42] Thus, different types of molecular architec-
tures have been obtained by systematically changing several
parameters including the concentration of adsorbate at sur-
face, the substrate type, and the deposition processing
method and rate. Recently highly complex supramolecular
architectures of fullerenes that noncovalently interact with
discotic systems, such as phthalocyanine[47] or porphyrin,[48]
derivatives were visualized by STM studies performed in
ultra-high vacuum. Moreover, Bai and co-workers have
grown electrochemically ordered nanostructures of a C60 de-
rivative and a functionalized PPV, and characterized them
by STM.[49]
Results and Discussion
Synthesis: The strategy employed for the preparation of the
C60-OPV conjugates F-3PV, F-5PV, F-3PV-F, and F-5PV-F
is based upon the 1,3-dipolar cycloaddition of C60 with bis-
nitrilimine derivatives generated in a one-pot procedure
from the corresponding bis-hydrazones. The synthesis of the
OPV precursors bearing two hydrazone functions is shown
in Schemes 1 and 2. Compound 1 was prepared in three
steps from hydroquinone according to a previously reported
method.[51] Reaction of 1 with 2,2-dimethylpropane-1,3-diol
in refluxing benzene in the presence of a catalytic amount
of p-toluenesulfonic acid (TsOH) gave the protected alde-
hyde 2 in 87% yield (Scheme 1). Subsequent treatment with
Scheme 1. Reagents and conditions: i) 2,2-dimethylpropane-1,3-diol,
TsOH (cat.), benzene, D, Dean Stark trap, 48 h (87%); ii) nBuLi, Et2O,
08C, 1 h, then DMF, 08C to room temperature, 4 h (86%); iii) {[2,5-bis-
(octyloxy)-1,4-phenylene]bis(methylene)}bis(tetraethyl)phosphonate,
tBuOK, THF, 08C, 1 h, then I2 (cat.), toluene, D, 12h (80%); iv) CF 3-
CO2H, CH2Cl2, H2O, room temperature, 4 h (95%); v) p-nitrophenylhy-
drazine, AcOH, EtOH, D, 3.5 h (99%).
4406
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4405 – 4415

Dumbbell-Shaped Arrays
FULL PAPER
nBuLi in Et2O at 08C followed by quenching with dry DMF
afforded aldehyde 3 in 86% yield. Treatment of aldehyde 3
uene gave the all-E OPV pentamer 10 in 51% yield. Depro-
tection, accomplished by treatment with CF3CO2H in
CH2Cl2/H2O, followed by reaction of the resulting bis-alde-
hyde 11 with p-nitrophenylhydrazine in refluxing ethanol in
the presence of acetic acid, afforded bis-hydrazone 12 in
83% yield.
with
{[2,5-bis(octyloxy)-1,4-phenylene]bis(methylene)}bis-
(tetraethyl)phosphonate[52] in THF in the presence of tBuOK
afforded the targeted OPV trimer, but as an E:Z isomer
mixture. The reaction of benzylic phosphonates with aro-
matic aldehydes under Wittig–Horner conditions is general-
ly stereoselective, leading to the E isomer only. In the pres-
ent case, steric hindrance resulting from the presence of the
octyloxy group in the ortho-position of the reactive groups
in both reagents may explain the lack of E:Z selectivity. Iso-
merization to the all-E derivative 4 was easily achieved by
treatment with a catalytic amount of iodine in refluxing tol-
uene and trimer 4 was thus obtained in 80% yield. Subse-
quent deprotection with CF3CO2H in CH2Cl2/H2O afforded
bis-aldehyde 5 in 95% yield. The E configuration of the
double bonds in 5 was confirmed by a coupling constant of
about 17 Hz for the AB system corresponding to the vinylic
Bis-hydrazones 6 and 12 were characterized by analytical
and spectroscopic data (see Supporting Information for de-
1
tails). The H NMR spectra of both compounds confirm the
E configuration of the double bonds by a coupling constant
of approximately 17 Hz for the AB system corresponding to
the vinylic protons. Also, their structures were confirmed by
MALDI-TOF mass spectra, which show the expected
[M+H]+ peaks at m/z=1376.8 (6) and 2094.2 (12).
Treatment of bis-hydrazone 6 with N-chlorosuccinimide
(NCS) and pyridine in dry chloroform, followed by reaction
of the resulting bis-nitrilimine intermediate with C60 in
chlorobenzene in the presence of Et3N at room temperature
afforded the desired dumbbell-shaped derivative F-3PV-F in
27% yield (Scheme 3). Interestingly, in addition to F-3PV-F,
1
protons in the H NMR spectrum. Bis-hydrazone 6 was then
obtained in 99% yield from aldehyde 5 and p-nitrophenyl-
hydrazine in refluxing ethanol in the presence of acetic acid.
The preparation of the corresponding OPV pentamer 12
is depicted in Scheme 2. LiAlH4 reduction of aldehyde 3
Scheme 3. Reagents and conditions: i) NCS, pyridine, CHCl3, 08C to
room temperature, then C60, Et3N, chlorobenzene, room temperature, 1 h
(from 6: F-3PV-F (27%) and F-3PV (58%); from 12: F-5PV-F (10%)
and F-5PV (19%)).
the monosubstituted derivative F-3PV was also isolated
from the reaction mixture; the aldehyde group in F-3PV
probably originated by hydrolysis of unreacted hydrazone
moiety during the workup procedure. Following the same
procedure, the reaction of bis-hydrazone 12 with NCS and
subsequent reaction with C60 at room temperature gave the
bis-(pyrazolino[60]fullerene) OPV derivative F-5PV-F and
the monosubstituted compound F-5PV in 10 and 19%
yields, respectively.
Owing to the presence of the octyloxy groups, the C60–
OPV conjugates[50] F-3PV, F-5PV, F-3PV-F, and F-5PV-F
are well soluble in common organic solvents (CH2Cl2,
CHCl3, benzene, toluene, THF) and were thus easily charac-
terized by UV-visible, FT-IR, and 1H and 13C NMR spectros-
Scheme 2. Reagents and conditions: i) LiAlH4, THF, 08C, 4 h (80%);
ii) CBr4, PPh3, THF, 08C to room temperature, 4 h (85%); iii) P(OEt)3,
1508C, 4 h (99%); iv) 5, tBuOK, THF, 08C to room temperature, 2h,
then I2 (cat.), toluene, D, 12h (51%); v) CF 3CO2H, CH2Cl2, H2O, room
temperature, 3 h (88%); vi) p-nitrophenylhydrazine, AcOH, EtOH, D,
3.3 h (83%).
yielded alcohol 7 which, after treatment with CBr4/PPh3,
gave bromide 8. Phosphonate 9 was then obtained in 99%
yield by reaction of 8 with triethylphosphite under Michae-
lis–Arbuzov conditions. Reaction of bis-aldehyde 5 with
phosphonate 9 under Wittig–Horner conditions, followed by
treatment with a catalytic amount of iodine in refluxing tol-
Chem. Eur. J. 2005, 11, 4405 – 4415
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4407

F. Langa, J.-F. Nierengarten, P. Samorì, N. Armaroli et al.
copy (see Supporting Information). The structures of the
four C60–OPV arrays were also confirmed by their MALDI-
TOF mass spectra showing the expected molecular ion
peaks in each case (m/z): 2814.7 (F-3PV-F), 3530.9 (F-5PV-
F), 1959.8 (F-3PV), and 2676.5 (F-5PV).
The fulleropyrazoline and OPV derivatives shown here
have been also prepared. They were used as reference com-
Figure 1. Cyclic voltammogram of compound F-3PV-F.
As a general feature on the reduction side, triads F-3PV-
F and F-5PV-F give rise to six one-electron reduction
waves. Four of them (E1red, E2red, E4red, and E6red) are quasi-re-
versible and correspond to the successive reductions of the
fullerene core. The other two waves are irreversible and the
peaks E3red and Er5ed are assigned to the reduction of the p-ni-
trophenyl moiety by comparison with the model compounds
6 and 12. Comparable electrochemical behavior is shown by
F-3PV and F-5PV. Similar to other pyrazolino[60]fullerene
derivatives, the first reduction potentials (E1red) are analo-
gous to that of pristine C60 and shifted to more positive
values relative to other fullerene derivatives due to the ÀI
effect of the pyrazoline ring.[53,54]
On the oxidation side, an irreversible one-electron wave
was observed at around +0.44 V (F-3PV-F) and +0.35 V
(F-5PV-F), while these waves appear at +0.34 V and
+0.31 V in F-3PV and F-5PV, respectively. Similarly, bis-
hydrazones 6 and 12 show the same oxidation around
+0.37 V. These values can be assigned to the oxidation of
the OPV moiety, in agreement with previous reports.[19] The
second oxidation wave in the two bis-hydrazones must be at-
tributed to the hydrazone moiety;[53] an additional wave is
observed in F-3PV and F-5PV at around +0.6 V, assigned
to the oxidation of the aldehyde group.
pounds for multicomponent arrays in the photophysical in-
vestigations. Compounds F1 and F2 were prepared as previ-
ously reported.[19,53] The OPV derivatives 3PV and 5PV
were obtained by LiAlH4 reduction of the corresponding
bis-aldehydes 5 and 11, respectively.
Electrochemistry: The electrochemical behavior of bis-hy-
drazones 6 and 12, dyads F-3PV and F-5PV and triads F-
3PV-F and F-5PV-F was investigated by cyclic voltammetry
(CV) and Osteryoung square-wave voltammetry (OSWV)
techniques at room temperature in ODCB/CH3CN (4:1) as
solvent (ODCB=ortho-dichlorobenzene) and using Bu4N+
À
ClO4 (TBAP) as the supporting electrolyte. The redox po-
tentials F-3PV, F-5PV, F-3PV-F, and F-5PV-F are reported
in Table 1 and are compared to those of the precursors 6
and 12, and pristine C60 as reference. The cyclic voltammo-
gram of F-3PV-F is depicted in Figure 1.
As support for the electrochemical results, the structures
of both dimers, F-3PV-F and F-5PV-F, were calculated at
the semiempirical PM3 level.[55] This revealed torsion angles
of ~388 between the p-nitrophenyl moiety and the pyrazoli-
no rings, thus preventing conju-
gation between them, whereas
the OPV unit and the pyrazo-
line rings are almost coplanar
(the dihedral angle is lower
than 108; see Figure 2). The
LUMO levels of both dumbbell
compounds were found to be
lower than that of C60; these
calculations confirmed the ex-
perimentally determined first
reduction potential in F-3PV-F
and F-5PV-F. As expected, the
Table 1. Redox potentials (OSWV) of organofullerenes F-3PV-F, F-5PV-F, F-3PV, F-5PV, bis-hydrazones 6
and 12, and C60.[a]
Compound
E1red
E2red
E3red
E4red
E5red
E6red
E1ox
E2ox
E3ox
F-3PV-F
F-5PV-F
F-3PV
F-5PV
6
À0.95
À0.94
À0.96
À0.99
–
À1.37
À1.37
À1.39
À1.40
–
À1.71[b]
À1.70[b]
À1.70[b]
À1.72[b]
À1.67[b]
À1.69[b]
–
À1.84
À1.87
À1.86
À1.85
–
À2.09[b]
À2.07[b]
À2.02[b]
À2.06[b]
À2.09[b]
À2.06[b]
–
À2.27
À2.25
À2.21
À2.32
–
+0.44[c]
+0.35[c]
+0.34[c]
+0.31[c]
+0.38[c]
+0.37[c]
–
–
–
–
–
–
–
+0.61[c]
+0.62[c]
+1.17[c]
+1.18[c]
–
–
12
C60
–
–
–
–
–
–
À0.97
À1.38
À1.85
À2.31
[a] V vs Ag/AgNO3; GCE as working electrode; 0.1m TBAP; ODCB:MeCN (4:1); scan rate: 100 mVsÀ1
.
[b] Irreversible according to CV. [c] Measured by CV.
4408
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4405 – 4415

Dumbbell-Shaped Arrays
FULL PAPER
phenyl group as a peripheral unit, electron transfer is no
longer observed and the fullerene fluorescence is restored.
Both F1 and F2 exhibit virtually identical absorption spec-
tra, but F2 is a stronger emitter with a fluorescence quan-
tum yield of 5.0”10À4 in CH2Cl2 (Table 3). This value is
quite similar to that of fulleropyrrolidines,[38] which are
among the strongest emitters within the family of monosub-
stituted fullerenes.
The different behavior between F1 and F2 can be ration-
alized by some degree of delocalization of the positive
charge, upon oxidation, provided by the tridodecyloxyben-
zene ring of F1. This result is in agreement with those ob-
served in other fulleropyrazoline derivates with electron-
donor or electron-withdrawing susbtituents.[54]
À
Figure 2. Most stable conformation of F-5PV-F. The OC8H17 groups
À
have been replaced by OCH3 groups for calculations.
HOMO is centered in the OPV unit and the LUMO is
shared between the two C60 cages.
As shown in Table 2, and in line with the electrochemical
data, the calculated HOMO–LUMO gap is smaller than
that of C60. This gap is significantly lower in the 5PV system
(F-5PV-F) than in the 3PV derivative (F-3PV-F), suggest-
ing the possibility of an easier electron transfer process.
The absorption spectra of the oligophenylenevinylene
(OPV) model compounds 3PV and 5PV are depicted in
Figure 3. They exhibit the characteristic p–p* absorption
with maxima at 409 and 447 nm, respectively. The maximum
of the more extended p-conjugated 5PV lies at lower
energy relative to 3PV and exhibits a higher molar extinc-
tion coefficient, as a result of the larger number of phenyl
rings.[19,22]
Table 2. Calculated (PM3) HOMO and LUMO levels of F-3PV-F, F-
5PV-F, and C60.
EHOMO [eV]
ELUMO [eV]
DE [eV]
F-3PV-F
F-5PV-F
C60
À8.32
À7.84
À9.48
À3.13
À3.06
À2.88
5.19
4.78
6.60
Photophysical properties
Fullerene and OPV model compounds: Fulleropyrazoline
F1 undergoes intramolecular electron transfer from the lone
pair of the pyrazoline sp3 nitrogen atom to the carbon
sphere.[19] This process, which is signalled by the dramatic
quenching of the typical fullerene-type fluorescence around
700 nm, can be suppressed in acid solution in which the pro-
tonation of the sp3 nitrogen lone pair prevents electron mi-
gration.[19] However, it is interesting to point out the crucial
role played by the tridodecyloxybenzene unit attached to
the heteroatomic five-membered ring. In fact if one takes
the fulleropyrazoline F2, bearing a methyl rather than a
Figure 3. Electronic absorption spectra in CH2Cl2 of: a) 3PV (black
dotted line), F-3PV (grey full line), F-3PV-F (black solid line) and F1
(black dashed line). b) 5PV (black dotted line), F-5PV (grey full line),
F-5PV-F (black full line) and F1 (black dashed line). c) Fluorescence
spectra of 3PV (black dashed line) and 5PV (black full line); CH2Cl2,
lexc =320 nm, A=0.15.
Table 3. Fluorescence properties[a] and excited state singlet lifetimes in CH2Cl2.
OPV moiety (298 K)
C60 moiety (298 K)
OPV moiety (77 K)
C60 moiety (77 K)
lmax [nm][d] t [ns][c]
[b]
[d]
lmax [nm][b]
Ffl
t [ns][c]
lmax [nm][d]
Ffl
t [ns][c]
lmax [nm][b]
t [ns][c]
[e]
F1
F2
3PV
F-3PV
F-3PV-F
5PV
F-5PV
F-5PV-F
–
–
–
–
0.64
0.00026
0.00007
0.35
0.00170
0.00025
–
–
1.5
–
–
1.2
–
–
706
706
–
706
706
–
ꢀ0.00001
0.00055
–
ꢀ0.00001
ꢀ0.00001
–
–
–
–
504
516
516
600
–
–
698
710
–
696
694
–
1.5
1.5
–
1.5
–
466
526
480
526
580
570
1.4, 5.0
–
–
0.7, 2.5
–
–
[f]
[e]
[f]
–
–
–
–
1.7[g]
2.2[g]
–
[f]
[e]
[f]
[f]
[e]
[h]
[h]
706
706
ꢀ0.00001
–
–
690
690
1.3[g]
1.2[g]
[f]
[e]
[h]
[h]
ꢀ0.00001
–
[a] From emission spectra corrected for the photomultiplier response. [b] Excitation on the maximum of the OPV absorption band (see Figures).
[c] lexc =407 nm. [d] lexc =600 nm. [e] Not measured due to signal weakness. [f] Below the temporal resolution of our apparatus (200 ps); see text for an
estimation of these lifetimes based on Equation (1). [g] lem =800 nm, very weak and noisy signal due to low photomultiplier sensitivity and poor quality
of the rigid matrix. [h] Indistinguishable over the instrumental noise.
Chem. Eur. J. 2005, 11, 4405 – 4415
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4409

F. Langa, J.-F. Nierengarten, P. Samorì, N. Armaroli et al.
Some relevant photophysical data are collected in Table 3.
Model compound 3PV exhibits an intense fluorescence
band (Figure 3) in CH2Cl2 both at 298 K (lmax =466 nm,
ffl =0.64) and at 77 K (lmax =504 nm). The singlet decay
time is monoexponential at room temperature (t=1.5 ns),
whereas at 77 K a biexponential decay is observed (t1 =
1.4 ns, t2 =5.0 ns), probably as a consequence of the pres-
ence of two different frozen rotameric forms in the solid
matrix.[21,56] The fluorescence properties of 5PV in CH2Cl2
are similar to 3PV. An intense emission band is observed
both at room temperature (lmax =526 nm , ffl =0.35, t=
1.2ns; Figure 3) and at 77 K in a rigid matrix ( lmax =564 nm,
t1 =0.7 ns, t2 =2.5 ns). The comparison of the fluorescence
properties of 3PV and 5PV leads to the same observations
described for other similar systems: the singlet excited-state
lifetime and the photoluminescence quantum yield decrease
with increasing conjugation length.[57]
to singlet energy transfer to the carbon sphere, which, once
excited, undergoes pyrazoline!C60 electron transfer. The
lack of fullerene sensitization is also confirmed by NIR lu-
minescence. We could not observe any sensitized singlet
oxygen emission at 1270 nm for F-3PV and F-5PV, confirm-
ing the lack of fullerene triplet formation.[19]
The occurrence of intramolecular electron transfer in F-
3PV and F-5PV, in line with the behavior of the reference
molecule F1, is supported by the recovery of the fluores-
cence in CH2Cl2 at 77 K (Table 3, Figure 4) and in a less
Dimeric compounds F-3PV and F-5PV: The absorption
spectra of F-3PV and F-5PV are shown in Figure 3. If one
compares them with the sum of the spectra of the compo-
nent units (F, 3PV, 5PV) some differences can be noticed;
in particular a red-shift of the OPV absorption band is
found. This difference is likely to be due to the more ex-
tended p-conjugation of the OPV-type moiety in F-3PV and
F-5PV compared to 3PV and 5PV, which possess CH2OH
terminal groups.
For any dyad, it is not possible to excite either the C60 or
the OPV moiety with 100% selectivity (Figure 3). Neverthe-
less, most of the light is absorbed by the OPV unit when ex-
citing in the range 450–520 nm. On the other hand, a sub-
stantial part of the light is absorbed by the fullerene moiety
for l<350 nm and l>630 nm, at which it exhibits a very
weak absorption tail, (e~300mÀ1 cmÀ1 at 650 nm).[19]
Figure 4. Fluorescence spectra of F-3PV (black full line), F1 (grey full
line) and F2 (black dashed line) in: a) 298 K, CH2Cl2, solution; b) 77 K,
CH2Cl2, rigid matrix. c) Fluorescence spectra of F-3PV in solvents of dif-
ferent polarity, toluene (TOL), dichloromethane (DCM) and benzonitrile
(BN). For all compounds lexc =330 nm and A=0.80.
polar solvent as toluene at 298 K (Figure 4). Under these
conditions the solvent repolarization around the charge-sep-
arated couple is prevented or strongly limited, thus disfavor-
ing the electron-transfer process.
The OPV-type fluorescence bands of F-3PV and F-5PV
are both red-shifted (as the absorption spectra) and dramati-
cally quenched relative to 3PV and 5PV (Table 3). Upon
excitation at 470 nm, at which the OPV moiety prominently
absorbs, a 2400- and 210-fold fluorescence intensity decrease
is observed relative to the model compounds F-3PV and F-
5PV, respectively. In F-3PV and F-5PV sensitization of the
fullerene fluorescence cannot be used as a probe for OPV!
C60 energy transfer, as done earlier for similar dyads con-
taining fulleropyrrolidines.[21] In the present case, as pointed
out above, the fullerene fluorescence is intrinsically
quenched by intramolecular electron transfer from the pyra-
zoline group to the carbon sphere. Practically, in CH2Cl2
both the potentially fluorescent moieties are quenched at
room temperature in F-3PV and F-5PV. The very weak
fluorescence band observed for the reference fullerene F1
(Table 3) is completely obscured by the tail of the residual
OPV emission when excitation is addressed to the OPV
moiety. On the other hand, excitation of the C60 subunits re-
sults in a very weak fluorescence band, identical to those of
F1 (Table 3). In analogy with a similar fulleropyrazoline–C60
array investigated earlier,[19] we assume that ultrafast
quenching of the OPV moiety in F-3PV and F-5PV is due
However, under these same conditions, quenching of the
OPV fluorescence is still active. This provides evidence for
two distinct photoprocesses in F-3PV and F-5PV, such as
OPV!C60 energy transfer and pyrazoline!C60 electron
transfer, for which the latter can be controlled by tempera-
ture and solvent polarity.[19] From Equation (1)[22] it is possi-
ble to estimate the rate constant of the OPV!C60 energy-
transfer process, which turns out to be 1.6”1012 and 1.7”
1011 sÀ1, corresponding to OPV quenched lifetimes of 0.6
and 5.9 ps respectively, that is, below our instrumental reso-
lution. In Equation (1), F and Fref are the OPV fluorescence
quantum yield of the multicomponent arrays and of the ref-
erence compounds, whereas tref is the singlet lifetime of the
latter.
ðFref=FÞÀ1
ð1Þ
kEnT
¼
tref
Trimeric compounds F-3PV-F and F-5PV-F: The absorp-
tion spectra of F-3PV-F and F-5PV-F in CH2Cl2 are report-
ed in Figure 3. Similarly to the dimeric systems, a red-shift
4410
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4405 – 4415

Dumbbell-Shaped Arrays
FULL PAPER
of the OPV moiety absorption band is observed for F-3PV-
F and F-5PV-F with respect to the OPV model compounds
3PV and 5PV, which can be rationalized as for the dyads
(see above).
The excited state behavior of F-3PV-F and F-5PV-F in
CH2Cl2 is practically the same as the dyad analogues. In par-
ticular we observe: 1) strong quenching of the OPV fluores-
cence in solvents of any polarity due to OPV!C60 energy
transfer, 2) quenching of fullerene fluorescence, 3) recovery
of C60 fluorescence by cooling down the samples to 77 K, 4)
recovery of fullerene fluorescence by decreasing the solvent
polarity (toluene), 5) no substantial effects of solvent polari-
ty on the extent of OPV fluorescence quenching, and 6) no
evidence of formation of fullerene triplet under any condi-
tion. These results suggest the occurrence of OPV!C60
energy transfer and pyrazoline!C60 electron transfer for F-
3PV-F and F-5PV-F.
The quenching factors for the OPV fluorescence band in
the triads are 9600 and 1400 for F-3PV-F and F-5PV-F, re-
spectively. According to Equation (1), the rate of energy
transfer (kEnT) can be calculated as 6.4”1012 and 1.1”
1012 sÀ1, respectively, corresponding to lifetimes shorter than
1 ps. It is interesting to note (see Table 3) that OPV!C60
singlet energy transfer, signaled by OPV fluorescence
quenching, is more efficient for the pentameric rather than
the trimeric OPV (see F-5PV-F vs F-3PV-F and F-5PV vs
F-3PV) and also on increasing the number of fullerenes
units (see F-3PV vs F-3PV-F and F-5PV vs F-5PV-F).
These results can be rationalized in terms of energy-transfer
theory both under the Fçrster and the Dexter approach,
which can be applied for OPV–C60 arrays as already dis-
cussed.[21] In this frame the energy-transfer rate is increased
upon enlargement of the overlapping area between the
emission spectra of the sensitizer (OPV in our case) and the
absorption spectra of the quencher (C60). As shown in
Figure 3, the molar extinction coefficient of fulleropyrazo-
line constantly decreases above 400 nm, while, on the other
hand, the OPV fluorescence bands are red-shifted when in-
creasing the conjugation length. As a consequence, the over-
lap integral is larger for 3PV- than 5PV-based systems and
increases by passing from one to two fullerene units. Ac-
cordingly, the quenching factor is the highest for F-3PV-F
and the smallest for F-5PV, corresponding to a 40-fold dif-
ference in the kEnT value.
Figure 5. a) STM current images of F-5PV-F physisorbed at the solution–
HOPG interface. b) Zoom-in of a monocrystalline area. Arrows indicate
dark spots, which are ascribed to fullerenes. Molecule shape and dimen-
sions have been obtained from energy minimization of a single isolated
molecule using commercial software. For sake of clarity, the alkyl chains
are not shown. The 1”1 unit cell is drawn with a solid line and the 2”1
superstructure with a dashed line. Tunneling parameters: Tip bias (Utip)=
550 mV, average tunnelling current (It)=15 pA.
crystal is depicted. The unit cell of the packing was deter-
mined from 2D Fourier analysis as a=3.2ꢁ0.2nm b=2.7ꢁ
0.2nm, and a=79ꢁ48 and confirmed by auto-correlation
averaging (see the Supporting Information). Molecular me-
chanics modeling revealed a molecular length of ~4.5 and a
width of 2.2 nm, in the case in which both the OPV chain
and the alkyl side chains adopt a fully extended conforma-
tion.
Usually, in STM current images the bright areas (corre-
sponding to high tunnelling current) can be ascribed to the
p-conjugated part of the molecule, since the energy differ-
ence between their frontier orbital and the Fermi level of
the substrate is relatively small.[60] In the present case semi-
STM and SFM investigations of F-5PV-F at surfaces: An
STM study at the solid–liquid interface made it possible to
cast light onto the self-assembly on highly oriented pyrolitic
graphite (HOPG) with a nanometer scale resolution. Fig-
ure 5a displays an STM current image recorded at the solid–
liquid interface. It shows a polycrystalline structure charac-
terized by small crystalline domains. The limited size of the
crystals, being in the range of a few tens of square nanome-
ters, can be explained in view of the competition between
the tendency of the OPV to assemble on HOPG[58] and the
poor inclination of C60 to pack on graphite[59] at the solid–
liquid interface. In Figure 5b a zoom-in showing a single 2D
Chem. Eur. J. 2005, 11, 4405 – 4415
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4411

F. Langa, J.-F. Nierengarten, P. Samorì, N. Armaroli et al.
empirical PM3 calculations performed with Hyperchem 5.1
revealed a distribution of the electronic states (HOMO and
LUMO) as reported in Figure 6. These levels need to be
correlated with the Fermi level of HOPG which amounts to
À4.7 eV.
In this case the different contrast of the two adjacent C60
groups could be due to the different distance of the mole-
cule from the substrate along the z axis. In view of this inter-
pretation, a 2”1 superstructure describes correctly the
supramolecular motif, whereby the correct unit cell bridges
four equivalent larger dark
spots, as indicated with
a
dashed line in Figure 5b. In
other words, the rows parallel
to the b vector of the unit cell
consist of a periodic structure
which conveys two C60 moieties
packed at the same height from
the surface that are intercalated
by a third C60 moiety slightly
shifted vertically with respect to
the two neighboring ones. This
packing motif allows a reducion
in the steric hindrance, similar
to the case of the supramolec-
ular staircase of functionalized
hexa-peri-hexabenzocoronene
films observed by STM at the
graphite–solution interface.[63]
Unfortunately the spatial reso-
Figure 6. Scheme of the electron affinities (Ea) and ionization potential (Ip) for different molecular systems (F-
3PV-F and F-5PV-F) and moieties (3PV, 5PV, C60, alkyl chains) in vacuum as estimated with semiempirical
calculations in vacuo. Since the experimental results have been recorded in the condensed phase, for the inter-
pretation of the STM contrasts they have to be considered only for the trend in the energy differences of the
levels.
By recording STM current images with a tunneling tip
that has a positive polarity, the obtained contrast is mainly
dominated by the HOMO of the adsorbate. Therefore one
should expect a brighter contrast for the oligo-PV segment,
a darker one for the alkyl chains, and an even darker one
yet for the C60 moiety. The black circles in the STM images,
marked with white arrows in Figure 5b, have sizes similar to
the fullerene moiety.
Correlating the STM measurements obtained in a mono-
crystalline areas with the results of molecular modeling, we
propose a molecular packing as the one depicted in Fig-
ure 5b. The dark spots can be ascribed to the fullerene moi-
eties, while the OPV segments are tilted respect to the full-
erene rows.
While both the Fourier analysis and the auto-correlation
averaging yielded the dimension of the smaller lattice cell,
that is, 3.2”2.7 nm, careful observation of the STM images
revealed the presence of smaller dark spots alternated with
the bigger ones along the C60 rows, as marked in Figure 5b
with grey and white arrows, respectively. Given the size of
the unit cell and the type of contrast observed in Figure 5,
one model of packing can be proposed. This is based on the
consideration that the b vector spacing of 2.7 nm is filled
with two fullerenes in a row. This distance is larger than
twice the 1.004 nm hard-sphere diameter of fullerenes.[61]
This might be explained with the effect of the steric hin-
drance of the nitro-functionalized pyrazoline linked to the
C60. The thermodynamically favored interactions between
two electron acceptors,[62] such as nitro and fullerene, could
promote the co-adsorption of neighboring C60 and NO2 units
along the dark rows, leading to the observed spacing of
2.7 nm.
lution attained here in the STM imaging did not allow us to
elucidate the packing at the molecular scale of the PV moi-
eties and the alkyl chains. This is due to the high conforma-
tional mobility at surfaces as reflected by the small size of
the crystalline domains.
The SFM measurements at the solid–air interface allowed
us to study molecular packing on the (sub-)micrometer
scale. Figure 7 displays topographical images of a dry film of
F-5PV-F prepared by drop-casting a 5”10À3 mgmLÀ1 solu-
tion of the sample in toluene on HOPG. It reveals a self-as-
sembled layer of F-5PV-F extended on the micrometer
scale. It conveys holes with various lateral sizes, that is, from
one micrometer down to a few nanometers. Tracing topo-
graphical profiles across these holes it was possible to deter-
mine the film thickness as 1.3 nm. Taking into account the
1.004 nm hard-sphere diameter of fullerene,[61] the measured
thickness does not correspond to a single nor to a double
layer, but rather indicates that, due to the tight packing, not
all the fullerenes moieties are at the same height on graph-
ite, but that some of them are shifted vertically, in good
agreement with the 2”1 supramolecular pattern model de-
scribed above.
The layer structure on the micron scale appears similar to
that obtained from the assembly on mica of a fullerene mol-
ecule in which the bulky ball is linked to an aliphatic chain
with a terminal carboxyl unit.[42] Nevertheless the two cases
are different. The one described in Langmuir–Blodgett pro-
cessing of the molecules on the hydrophilic mica surface[42]
gave an edge-on packing, with the aliphatic chains arranged
perpendicular to the basal plane of the surface, giving a
3 nm thick layer. In contrast, in the system studied here, the
molecules lie flat on the graphite surface forming the self-as-
4412
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4405 – 4415

Dumbbell-Shaped Arrays
FULL PAPER
Figure 7. a,b) SFM images of F-5PV-F in toluene, 5”10À3 mgmLÀ1 showing a self-assembled layer adsorbed on the graphite. White arrows indicate
graphite steps on the surface. c) Anisotropic molecular ad-layer adsorbed at a graphite step, indicated by an arrow. z ranges : a) 12nm; b) 3.4 nm;
c) 5.5 nm.
sembled layer. Because of the limit in the resolution of non-
contact mode SFM being on the order of a few nanometers,
one cannot identify within the layer the different thickness
of the central backbone of the molecule and the side groups
of the C60 moieties. Whilst the overall root-mean square
roughness (Rrms) of this film over areas of 1”1 mm is very
small (~1.5 nm), on the nanometers scale it is governed by
the different moieties that comprise the single molecule.
The boundaries in the self-assembled monolayers (indi-
cated by white arrows in Figure 7a) are dictated by the
structure of the crystalline HOPG underneath. In some
lower coated areas of the surface, a preferential adsorption
at the step edges has been observed, forming long thin mo-
lecular rows (arrow in Figure 7c) with widths of tens of
nanometers. A more diluted solution (5”10À5 mgmLÀ1)
leaded to an adsorbed layer similar to that described above,
but with larger holes (images not shown).
transfer. The former process, signaled by quenching of the
OPV moieties, is the fastest for F-3PV-F and the slowest in
the case of F-5PV, in agreement with energy-transfer theory
predictions.
A fine tuning of the processing conditions from solutions
allowed the production of ordered layer architectures of the
F-5PV-F on HOPG surfaces. An SFM study at the solid–air
interface revealed a pretty smooth surface in the microme-
ter scale. On the other hand, molecular scale explorations
performed with STM have shown the tendency of this
hybrid molecular system to form polycrystalline structures
characterized by very small crystalline domains. This can be
explained in view of the competition between physisorption
at surfaces and solvation of the different units composing
the molecule.
The self-assembled layer extended on the scale of several
micrometers provides evidence for the strong tendency of
the molecules to aggregate due to p–p stacking.[64] As the
growth of such ordered films of p-conjugated molecules on
graphite is a kinetically controlled process,[65] a higher
degree of order was achieved exploiting a slow deposition
process such as drop casting, when compared to the less-or-
dered thin films produced by spin coating a similar hybrid
molecular system.[21] The paramount role played by p–p in-
teractions in the self-assembly of F-5PV-F at surfaces was
evidenced also by the tendency of films prepared from a
more concentrated solution to give large disordered agglom-
erates with diameters of hundreds of nanometers (images
not shown).
Experimental Section
General: Reagents and solvents were purchased as reagent grade and
used without further purification. The preparation of compounds 1–12 is
described in the Supporting Information. All reactions were performed
in standard glassware under an inert Ar atmosphere. Evaporation and
concentration were done at water aspirator pressure and drying in vacuo
at 10À2 Torr. Column chromatography: silica gel 60 (230–400 mesh,
0.040–0.063 mm) was purchased from E. Merck. Thin-layer chromatogra-
phy (TLC) was performed on glass sheets coated with silica gel 60 F254
purchased from E. Merck, visualization by UV light. NMR spectra were
recorded on
a Bruker AC 200 (200 MHz) or a Bruker AM 400
(400 MHz) with solvent peaks as reference. Mass measurement were car-
ried out on a Brucker BIFLEXTM matrix-assisted laser desorption time-
of-flight mass spectrometer (MALDI-TOF). A saturated solution of
1,8,9-trihydroxyanthracene (dithranol ALDRICH EC: 214–538–0) in
CH2Cl2 was used as a matrix. FAB mass spectra were obtained on a VG
AutoSpec instrument, using m-nitrobenzyl alcohol as a matrix. FT-IR
spectra were recorded on a Nicolet Impact 410 spectrophotometer using
KBr disks. Elemental analyses were performed by the analytical service
at the Institut Charles Sadron, Strasbourg.
Conclusion
Symmetrically substituted oligophenylenevinylene (OPV)
derivatives bearing terminal p-nitrophenyl-hydrazone
groups have been prepared and used for the synthesis of
dumbbell-shaped bis(pyrazolino[60]fullerene)–OPV systems
with one (F-3PV, F-5PV) or two (F-3PV-F, F-5PV-F) C60
terminal units. All of these arrays exhibit intramolecular
OPV!C60 energy transfer and pyrazoline!C60 electron
General procedure for the preparation of the C60-OPV derivatives: Pyri-
dine (5 mL) was added under argon to a solution of the appropriate hy-
drazone (1 equiv) in dry chloroform (10 mL). The mixture was cooled to
08C, NCS (4.1 equiv) was added, and the mixture was stirred for 1 hour;
[60]fullerene (2.2 equiv) in dry chlorobenzene (30 mL) and Et3N
(1 equiv) were then added. The solution was allowed to reach room tem-
perature and stirred for 1 hour. The solvent was removed under reduced
pressure and the two reaction products were purified by silica gel flash
Chem. Eur. J. 2005, 11, 4405 – 4415
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4413

F. Langa, J.-F. Nierengarten, P. Samorì, N. Armaroli et al.
chromatography, with toluene as eluent. Centrifugation with methanol
(4 times) achieved further purification of the solid.
1700 nm range). Corrected spectra were obtained on a calibration curve
supplied with the instrument. Emission lifetimes were determined with
the Edinburgh FLS920 spectrometer equipped with a laser diode head as
excitation source (1 MHz repetition rate, lexc =407 nm, 200 ps time reso-
lution) and the above-mentioned Hamamatsu R928 PMT as detector.
F-3PV-F: Yield: 27%; 1H NMR (CDCl3): d=8.34 (d, J=9.5 Hz, 4H),
8.26 (d, J=9.5 Hz, 4H), 7.57 (B of AB, d, J=17 Hz, 2H), 7.48 (A of AB,
d, J=17 Hz, 2H), 7.28 (s, 2H), 7.21 (s, 2H), 7.14 (s, 2H), 4.05–3.99 (m,
12H), 1.90–1.71 (m, 12H), 1.57–1.25 (m, 60H), 0.96–0.75 ppm (m, 18H);
13C NMR (CDCl3): d=151.9, 151.0, 150.3, 150.0, 147.6, 147.0, 146.9,
146.6, 146.2, 146.1, 145.9, 145.8, 145.5, 145.3, 145.2, 145.1, 144.6, 144.3,
143.9, 143.1, 142.8, 142.7, 142.3, 142.1, 141.9 (2C), 141.8, 140.6, 139.2,
137.1, 135.0, 130.3, 127.2, 125.3, 123.1, 119.1, 119.0, 116.3, 110.7, 89.3,
84.3, 69.9, 69.5, 32.2, 30.0, 29.9, 29.8, 29.7, 26.7, 26.6, 26.5, 23.0, 14.5,
14.4 ppm; IR (KBr): n˜ =2925.4, 1638.6, 1616.7, 1587.6 (CH=N), 1493.0,
1328.1, 1197.1, 1022.5, 845.4, 527.7 cmÀ1; MALDI-TOF: m/z: 2814.7,
2092.6 [M+ÀC60].
STM and SFM measurements: Scanning tunneling microscopy was per-
formed at the solid–liquid interface employing a Multimode (Veeco) op-
erating with a picoAmp preamplifier and 0.25 mm thick mechanically cut
Pt/Ir tips. A drop of an almost saturated solution in 1,2,4-trichloroben-
zene (Aldrich) was applied to the basal plane of a freshly cleaved highly
oriented pyrolitic graphite (HOPG) surface. Trichlorobenzene was
chosen because it possesses a low volatility which permits to carry out
the STM studies for several hours at the solid–liquid interface with the
STM tip immersed in the solution. Moreover it is pretty good for solubi-
lizing conjugated molecules due to both its chlorination and its aromatici-
ty. By changing the tunneling parameters it was possible to visualize
either the organic adsorbate with a sub-molecular resolution or the
HOPG substrate with atomic resolution. Typical tunneling parameters
employed to visualize the organic adsorbates are bias voltage (Ut)=200–
600 mV and average tunneling current (It)=5–15 pA. Image processing
was done exploiting a commercial software SPIP version 2.000, Image
Metrology ApS.
F-5PV-F: Yield: 10%; 1H NMR (CDCl3): d=8.34 (d, J=9.5 Hz, 4H),
8.26 (d, J=9.5 Hz, 4H), 7.58 (B of AB, d, J=17 Hz, 4H), 7.48 (A of AB,
d, J=17 Hz, 4H), 7.29 (s, 2H), 7.28 (s, 2H), 7.21 (s, 2H), 7.17 (s, 2H),
7.15 (s, 2H), 4.10–4.00 (m, 20H), 1.92–1.82 (m, 20H), 1.58–1.27 (m,
100H), 0.89–0.81 ppm (m, 30H); 13C NMR (CDCl3): d=151.9, 150.9,
150.1, 147.6, 147.0, 146.3, 146.1, 146.0, 145.8, 145.5, 145.2, 145.1, 144.7,
144.3, 143.1, 142.9, 142.3, 142.2, 142.0, 140.7, 139.2, 137.1, 135.2, 130.4,
127.9, 126.8, 125.4, 123.3, 114.2, 110.5, 97.8, 89.3, 83.5, 69.9, 69.5, 32.2,
29.8, 29.7, 26.6, 26.5, 23.8, 23.0, 14.5, 14.4 ppm; IR (KBr): n˜ =2921.3,
1636.3, 1616.5, 1588.1 (CH=N), 1495.5, 1319.9, 1262.9, 1020.9, 800.2,
527.3 cmÀ1; MALDI-TOF: m/z: 3530.9, 2810.9 [M+ÀC60].
Scanning force microscope (SFM) studies in the noncontact mode[66]
were carried out using an Autoprobe CP Research (Thermomicroscope)
recording the height signal (output of the feedback signal). SFM was run
in an air environment at room temperature with scan rates of 1–1.5 Hz
per line. Images with scan lengths ranging from 60 mm down to 1 mm
were recorded with a resolution of 512”512 pixels by using the 100 mm
scanner and noncontact Si ultralevers with a spring constant of k in the
F-3PV: Yield: 58%; 1H NMR (CDCl3): d=10.4 (s, 1H; CHO), 8.32(d,
J=9.5 Hz, 2H), 8.24 (d, J=9.5 Hz, 2H), 7.55–7.44 (m, 4H), 7.26 (s, 2H),
7.20 (s, 2H), 7.13 (s, 2H), 4.05–4.02 (m, 12H), 1.86–1.71 (m, 12H), 1.58–
1.25 (m, 60H), 0.88–0.79 ppm (m, 18H); 13C NMR (CDCl3): d=189.1
(CHO), 156.2, 152.0, 151.3, 151.1, 150.6, 150.4, 150.2, 147.7, 147.1, 147.0,
146.6, 146.3, 146.2, 146.1, 145.9, 145.6, 145.4, 145.3, 145.2, 145.1, 144.7,
144.4, 144.0, 143.2, 143.1, 143.0, 142.9, 142.8, 142.4, 142.2, 142.1, 142.0,
141.9, 140.7, 139.2, 137.1, 135.2, 130.3, 127.7, 127.0, 125.3, 125.2, 124.0,
123.0, 119.0, 116.3, 110.6, 110.0, 89.2, 84.2, 69.7, 69.3, 69.1, 69.0, 58.5, 31.8,
31.8, 31.7, 30.9, 29.7, 29.5, 29.4, 29.3, 29.2, 26.4, 26.3, 26.2, 26.1, 22.6, 18.4,
14.1, 14.0 ppm; IR (KBr): n˜ =2852.5, 1733.0 (C=O), 1677.4, 1591.9 (CH=
N), 1467.9, 1322.6, 1198.6, 852.3, 526.9 cmÀ1; MALDI-TOF: m/z: 1959.8,
1239.8 [M+ÀC60].
range 2.1–17.0 NmÀ1
.
Ultra thin films for SFM analysis have been prepared by drop casting on
freshly cleaved HOPG surfaces a solution of F-5PV-F in toluene (HPLC
grade, Aldrich). Different concentrations of 5”10À1, 5”10À3, and to 5”
10À5 mgmLÀ1 were used. After deposition, the solvent was removed by
heating the sample for 60 min at 708C.
F-5PV: Yield: 19%; 1H NMR (CDCl3): d=10.4 (s, 1H; CHO), 8.32(d,
J=9.5 Hz, 2H), 8.26 (d, J=9.5 Hz, 2H), 7.66–7.44 (m, 8H), 7.33 (s, 2H),
7.21 (s, 2H), 7.18 (s, 2H), 7.17 (s, 2H), 7.15 (s, 2H), 4.07–4.04 (m, 20H),
1.89–1.86 (m, 20H), 1.59–1.27 (m, 100H), 0.90–0.85 ppm (m, 30H);
13C NMR (CDCl3): d=189.2 (CHO), 156.2, 152.1, 152.0, 151.2, 151.1,
151.0, 150.6, 150.3, 150.2, 147.7, 147.1, 147.0, 146.4, 146.3, 146.1, 146.0,
145.9, 145.6, 145.4, 145.3, 145.2, 145.1, 145.1, 144.8, 144.7, 144.4, 144.3,
144.0, 143.8, 143.2, 143.1, 143.0, 142.9, 142.8, 142.3, 142.2, 142.1, 142.0,
141.9, 140.7, 139.2, 139.1, 137.1, 135.2, 125.3, 119.0, 110.8, 110.6, 110.5,
110.3, 110.0, 89.4, 89.2, 84.2, 69.4, 69.3, 31.9, 31.8, 31.8, 29.7, 29.5, 29.4,
29.4, 29.3, 29.3, 29.2, 26.4, 26.2, 26.1, 22.6, 14.2, 14.1, 14.0 ppm; IR (KBr):
n˜ =2925.3, 2850.4, 1717.3 (C=O), 1636.7, 1618.5, 1587.3 (CH=N), 1496.3,
1322.1, 1257.1, 1197.3, 802.1, 526.6 cmÀ1; MALDI-TOF: m/z: 2676.5,
1956.5 [M+ÀC60].
Acknowledgements
We thank the European Commission for financial support through the
RTN project FAMOUS (Contract HPRN-CT-2002–00171). We also
thank Italian MIUR (contract FIRB RBNE019H9K, Molecular Manipu-
lation for Nanometric Machines), the ESF-SONS-BIONICS project, the
Spanish Ministry of Education and Science (Proyect CTQ2004–00364/
BQU) and Feder funds.
[1] H. Imahori, Y. Sakata, Eur. J. Org. Chem. 1999, 2445–2457.
[2] D. M. Guldi, Chem. Commun. 2000, 321–327.
[3] D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2001, 34, 40–
48.
[4] G. Yu, J. Gao, J. C. Hummelen, F. Wudl, A. J. Heeger, Science 1995,
270, 1789–1791.
[5] A. Cravino, N. S. Sariciftci, J. Mater. Chem. 2002, 12, 1931–1943.
[6] J. K. J. van Duren, X. N. Yang, J. Loos, C. W. T. Bulle-Lieuwma,
A. B. Sieval, J. C. Hummelen, R. A. J. Janssen, Adv. Funct. Mater.
2004, 14, 425–434.
[7] N. Armaroli, Photochem. Photobiol. Sci. 2003, 2, 73–87.
[8] J. F. Nierengarten, New J. Chem. 2004, 28, 1177–1191.
[9] J. F. Nierengarten, Sol. Energy Mater. Sol. Cells 2004, 83, 187–199.
[10] H. Imahori, K. Hagiwara, T. Akiyama, M. Aoki, S. Taniguchi, T.
Okada, M. Shirakawa, Y. Sakata, Chem. Phys. Lett. 1996, 263, 545–
550.
Electrochemistry: Cyclic voltammetry measurements were carried out on
an Autolab PGSTAT 30 potentiostat by using a BAS MF-2062 Ag/0.01m
AgNO3, 0.1m TBAP in ODCB (ortho-dichlorobenzene)/ACN (acetoni-
trile) reference electrode, an auxiliary electrode consisting of a Pt wire,
and a Metrohm 6.0805.010 conventional glassy carbon electrode (3 mm
o.d.) as a working electrode, which was directly immersed in the solution.
A 10 mL electrochemical cell from BAS (Model VC-2) was also used.
The reference potential was shifted by 290 mV towards a more negative
potential relative to the Ag/AgCl scale. E1/2 values were taken as the
average of the anodic and cathodic peak potentials. Scan rate:
100 mVsÀ1
.
Photophysical measurements: Experimental procedures were as de-
scribed previously.[19] Emission spectra were obtained with an Edinburgh
FLS920 spectrometer (continuous 450 W Xe lamp), equipped with a pelt-
ier-cooled Hamamatsu R928 photomultiplier tube (185–850 nm) or a Ha-
mamatsu R5509–72supercooled photomultiplier tube (193 K, 800–
[11] M. Carano, T. Da Ros, M. Fanti, K. Kordatos, M. Marcaccio, F. Pao-
lucci, M. Prato, S. Roffia, F. Zerbetto, J. Am. Chem. Soc. 2003, 125,
7139–7144.
4414
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 4405 – 4415

Dumbbell-Shaped Arrays
FULL PAPER
[12] F. Langa, P. de la Cruz, J. L. Delgado, M. J. Gomez-Escalonilla, A.
Gonzalez-Cortes, A. de la Hoz, V. Lopez-Arza, New J. Chem. 2002,
26, 76–80.
[13] M. Prato, Q. C. Li, F. Wudl, V. Lucchini, J. Am. Chem. Soc. 1993,
115, 1148–1150.
[39] G. Accorsi, N. Armaroli, J. F. Eckert, J. F. Nierengarten, Tetrahedron
Lett. 2002, 43, 65–68.
[40] J. F. Nierengarten, N. Armaroli, G. Accorsi, Y. Rio, J. F. Eckert,
Chem. Eur. J. 2003, 9, 37–41.
[41] F. Cacialli, J. S. Wilson, J. J. Michels, C. Daniel, S. C. , R. H. Friend,
N. Severin, P. Samorì, J. P. Rabe, M. J. OꢀConnell, P. N. Taylor, H. L.
Anderson, Nat. Mater. 2002, 1, 160–164.
[14] D. M. Guldi, S. Gonzalez, N. Martin, A. Anton, J. Garin, J. Orduna,
J. Org. Chem. 2000, 65, 1978–1983.
[15] T. Da Ros, M. Prato, M. Carano, P. Ceroni, F. Paolucci, S. Roffia, J.
Am. Chem. Soc. 1998, 120, 11645–11648.
[42] S. Z. Kang, S. L. Xu, H. M. Zhang, L. B. Gan, C. Wang, L. J. Wan,
C. L. Bai, Surf. Sci. 2003, 536, L408–L414.
[16] N. Martin, L. Sanchez, B. Illescas, I. Perez, Chem. Rev. 1998, 98,
2527–2547.
[43] K. Yase, N. Ara-Kato, T. Hanada, H. Takiguchi, Y. Yoshida, G.
Back, K. Abe, N. Tanigaki, Thin Solid Films 1998, 331, 131–140.
[44] Y. Kim, L. Jiang, T. Iyoda, K. Hashimoto, A. Fujishima, Surf. Sci.
1997, 385, L945–L951.
[45] T. Sakurai, X. D. Wang, Q. K. Xue, Y. Hasegawa, T. Hashizume, H.
Shinohara, Prog. Surf. Sci. 1996, 51, 263–408.
[46] P. Byszewski, Z. Klusek, S. Pierzgalski, S. Datta, E. Kowalska, M.
Poplawska, J. Electron Spectrosc. Relat. Phenom. 2003, 130, 25–32.
[47] M. de Wild, S. Berner, H. Suzuki, H. Yanagi, D. Schlettwein, S. Ivan,
A. Baratoff, H. J. Guentherodt, T. A. Jung, ChemPhysChem 2002, 3,
881–885.
[17] F. Langa, P. de la Cruz, E. Espildora, A. de la Hoz, J. L. Bourde-
lande, L. Sanchez, N. Martin, J. Org. Chem. 2001, 66, 5033–5041.
[18] P. de la Cruz, A. Diaz-Ortiz, J. J. Garcia, M. J. Gomez-Escalonilla,
A. de la Hoz, F. Langa, Tetrahedron Lett. 1999, 40, 1587–1590.
[19] N. Armaroli, G. Accorsi, J. P. Gisselbrecht, M. Gross, V. Krasnikov,
D. Tsamouras, G. Hadziioannou, M. J. Gomez-Escalonilla, F. Langa,
J. F. Eckert, J. F. Nierengarten, J. Mater. Chem. 2002, 12, 2077–2087.
[20] D. M. Guldi, C. P. Luo, A. Swartz, R. Gomez, J. L. Segura, N.
Martin, J. Phys. Chem. A 2004, 108, 455–467.
[21] J. F. Eckert, J. F. Nicoud, J. F. Nierengarten, S. G. Liu, L. Echegoyen,
F. Barigelletti, N. Armaroli, L. Ouali, V. Krasnikov, G. Hadziioan-
nou, J. Am. Chem. Soc. 2000, 122, 7467–7479.
[22] E. Peeters, P. A. van Hal, J. Knol, C. J. Brabec, N. S. Sariciftci, J. C.
Hummelen, R. A. J. Janssen, J. Phys. Chem. B 2000, 104, 10174–
10190.
[23] D. M. Guldi, A. Swartz, C. P. Luo, R. Gomez, J. L. Segura, N.
Martin, J. Am. Chem. Soc. 2002, 124, 10875–10886.
[24] E. H. A. Beckers, P. A. van Hal, A. Schenning, A. El-ghayoury, E.
Peeters, M. T. Rispens, J. C. Hummelen, E. W. Meijer, R. A. J. Jans-
sen, J. Mater. Chem. 2002, 12, 2054–2060.
[48] D. Bonifazi, H. Spillmann, A. Kiebele, M. de Wild, P. Seiler, F. Y.
Cheng, H. J. Guntherodt, T. Jung, F. Diederich, Angew. Chem. 2004,
116, 4863–4867; Angew. Chem. Int. Ed. 2004, 43, 4759–4763.
[49] M. J. Han, L. J. Wan, S. B. Lei, H. M. Li, X. L. Fan, C. L. Bai, Y. L.
Li, D. B. Zhu, J. Phys. Chem. B 2004, 108, 965–970.
[50] M. J. Gomez-Escalonilla, F. Langa, J. M. Rueff, L. Oswald, J. F. Nier-
engarten, Tetrahedron Lett. 2002, 43, 7507–7511.
[51] J. F. Nierengarten, T. Gu, G. Hadziioannou, D. Tsamouras, V. Kras-
nikov, Helv. Chim. Acta 2004, 87, 2948–2966.
[52] H. Detert, D. Schollmeyer, E. Sugiono, Eur. J. Org. Chem. 2001,
2927–2938.
[25] E. H. A. Beckers, P. A. van Hal, A. Dhanabalan, S. C. J. Meskers, J.
Knol, J. C. Hummelen, R. A. J. Janssen, J. Phys. Chem. A 2003, 107,
6218–6224.
[26] M. Gutierrez-Nava, G. Accorsi, P. Masson, N. Armaroli, J. F. Nieren-
garten, Chem. Eur. J. 2004, 10, 5076–5086.
[53] J. L. Delgado, P. de la Cruz, V. Lopez-Arza, F. Langa, D. B. Kimball,
M. M. Haley, Y. Araki, O. Ito, J. Org. Chem. 2004, 69, 2661–2668.
[54] E. Espildora, J. L. Delgado, P. de la Cruz, A. de la Hoz, V. Lopez-
Arza, F. Langa, Tetrahedron 2002, 58, 5821–5826.
[55] Hyperchem 5.1 Package
[27] T. Gu, D. Tsamouras, C. Melzer, V. Krasnikov, J. P. Gisselbrecht, M.
Gross, G. Hadziioannou, J. F. Nierengarten, ChemPhysChem 2002,
3, 124–127.
[56] U. Mazzucato, F. Momicchioli, Chem. Rev. 1991, 91, 1679–1719.
[57] E. Peeters, A. M. Ramos, S. C. J. Meskers, R. A. J. Janssen, J. Chem.
Phys. 2000, 112, 9445–9454.
[28] S. Knorr, A. Grupp, M. Mehring, G. Grube, F. Effenberger, J. Chem.
Phys. 1999, 110, 3502–3508.
[29] P. A. van Hal, E. H. A. Beckers, S. C. J. Meskers, R. A. J. Janssen, B.
Jousselme, P. Blanchard, J. Roncali, Chem. Eur. J. 2002, 8, 5415–
5429.
[58] A. Gesqui›re, P. Jonkheijm, A. Schenning, E. Mena-Osteritz, P.
Bꢂuerle, S. De Feyter, F. C. De Schryver, E. W. Meijer, J. Mater.
Chem. 2003, 13, 2164–2167.
[59] S. Uemura, P. Samorì, M. Kunitake, C. Hirayama, J. P. Rabe, J.
Mater. Chem. 2002, 12, 3366–3367.
[30] H. Kanato, K. Takimiya, T. Otsubo, Y. Aso, T. Nakamura, Y. Araki,
O. Ito, J. Org. Chem. 2004, 69, 7183–7189.
[60] R. Lazzaroni, A. Calderone, J. L. BrØdas, J. P. Rabe, J. Chem. Phys.
1997, 107, 99–105.
[31] T. Nakamura, M. Fujitsuka, Y. Araki, O. Ito, J. Ikemoto, K. Taki-
miya, Y. Aso, T. Otsubo, J. Phys. Chem. B 2004, 108, 10700–10710.
[32] J. L. Segura, R. Gomez, N. Martin, C. P. Luo, D. M. Guldi, Chem.
Commun. 2000, 701–702.
[61] Y. Z. Li, M. Chander, J. C. Patrin, J. H. Weaver, L. P. F. Chibante,
R. E. Smalley, Phys. Rev. B 1992, 45, 13837–13840.
[62] C. A. Hunter, K. R. Lawson, J. Perkins, C. J. Urch, J. Chem. Soc.
Perkin Trans. 2 2001, 651–669.
[33] Y. Obara, K. Takimiya, Y. Aso, T. Otsubo, Tetrahedron Lett. 2001,
42, 6877–6881.
[63] P. Samorꢃ, A. Fechtenkotter, F. Jꢂckel, T. Bohme, K. Mꢄllen, J. P.
Rabe, J. Am. Chem. Soc. 2001, 123, 11462–11467.
[34] J. J. Apperloo, C. Martineau, P. A. van Hal, J. Roncali, R. A. J. Jans-
sen, J. Phys. Chem. A 2002, 106, 21–31.
[35] C. Martineau, P. Blanchard, D. Rondeau, J. Delaunay, J. Roncali,
Adv. Mater. 2002, 14, 283-+.
[36] K. I. Yamanaka, M. Fujitsuka, Y. Araki, O. Ito, T. Aoshima, T. Fu-
kushima, T. Miyashi, J. Phys. Chem. A 2004, 108, 250–256.
[37] N. Negishi, K. Takimiya, T. Otsubo, Y. Harima, Y. Aso, Chem. Lett.
2004, 33, 654–655.
[64] V. Georgakilas, F. Pellarini, M. Prato, D. M. Guldi, M. Melle-
Franco, F. Zerbetto, Proc. Natl. Acad. Sci. USA 2002, 99, 5075–
5080.
[65] P. Samorì, M. Keil, R. Friedlein, J. Birgerson, M. Watson, M.
Mꢄllen, W. R. Salaneck, J. P. Rabe, J. Phys. Chem. B 2001, 105,
11114–11119.
[66] P. Samorì, J. P. Rabe, J. Phys. Condens. Matter 2002, 14, 9955.
[38] N. Armaroli, F. Barigelletti, P. Ceroni, J. F. Eckert, J. F. Nicoud, J. F.
Nierengarten, Chem. Commun. 2000, 599–600.
Received: January 25, 2005
Published online: April 29, 2005
Chem. Eur. J. 2005, 11, 4405 – 4415
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4415