Synthesis and characterization of new trinitrofluorene-fullerene dyads as photosensitizers in photorefractive polymer materials. Redox behavior and charge-transfer properties

Article abstract of DOI:10.1016/j.tet.2005.07.124

Novel trinitrofluorene-C₆₀ dyads bearing a flexible polyether chain have been synthesized and characterized. UV-vis spectroscopy and cyclic voltammetric results indicate that there is no interaction in their ground states. NMR and electrochemic

Full text of DOI:10.1016/j.tet.2005.07.124

Tetrahedron 62 (2006) 2102–2109
Synthesis and characterization of new trinitrofluorene–fullerene
dyads as photosensitizers in photorefractive polymer materials.
Redox behavior and charge-transfer properties
a
Luis Martın-Gomis, Javier Ortiz, Fernando Fernandez-Lazaro, Angela Sastre-Santos,
a
a
a,
´
*
´
´
´
Bevan Elliottb and Luis Echegoyenb,
*
a
´ ´ ´ ´ ´
Division de Quımica Organica, Instituto de Bioingenierıa, Universidad Miguel Hernandez, Elche 03202 (Alicante), Spain
bDepartment of Chemistry, Clemson University, 219 Hunter Laboratories, Clemson, SC 29634, USA
Received 21 June 2005; revised 20 July 2005; accepted 22 July 2005
Available online 1 December 2005
Abstract—Novel trinitrofluorene–C60 dyads bearing a flexible polyether chain have been synthesized and characterized. UV–vis
spectroscopy and cyclic voltammetric results indicate that there is no interaction in their ground states. NMR and electrochemical studies of
dyads 2 and 3 in solution indicate the existence of a charge-transfer complex between the trinitrofluorene moieties and N-ethylcarbazol.
Preliminary results of the photorefractive performance of these compounds as sensitizers in polymer composites based on
poly(N-vinylcarbazole) are similar to that of C60, but with shorter response times and slightly lower gain coefficients.
q 2006 Elsevier Ltd. All rights reserved.
1. Introduction
The PR effect was observed for the first time in organic
crystals in 1990 by Sutter and Gu¨nter.5 Just one year later, in
1991, Ducharme et al. demonstrated PR effect in a polymeric
composite.6 Since then, several research groups have focused
experimental efforts on the development of PR polymers.7
The photorefractive (PR) effect can be defined as a spatial
modulation of the index of refraction induced by a light
intensity pattern.1 In order to display this effect, the material
must show both photoconductivity and a high nonlinear
response and/or birefringence. Under these conditions, an
internal redistribution of charges takes place, which induces
the variation on the refraction index. This effect was first
observed in 1966 in an inorganic crystal of LiNbO3.2 To
name a few, other examples of inorganic materials in which
the PR effect has been examined are BaTiO3, KNbO3,
Bi12SiO20 (BSO), GaAs.3 The applications of these
materials have been limited by the difficulty in crystal
growing and sample preparation. Organic PR materials,
however, exhibit refractive index modulations and gain
coefficients which are superior compared to their inorganic
counterparts, and due to their versatility, lower cost and
easier fabrication and processability are becoming promis-
ing candidates for applications in optical holographic data
storage, real-time image processing, and phase
conjugation.4
The physical phenomena necessary to achieve photo-
refractivity are photoinduced charge generation, charge
transport, trapping, and electrooptical nonlinearity. One of
the major approximations to achieve PR effect in organic
materials, and the most widely spread indeed, is the use of
functional polymers (which are responsible of one or two of
the functional properties needed to originate PR effect)
doped with molecules responsible for the rest of the
functional properties.8 The use of polymeric compositions
has led to materials with high gain coefficients (G), in the
order of 200 cmK1, and with diffraction efficiencies (h) of
100%.9 However, the most important drawback of organic
polymers is the low speed of grating formation, that is, the
high response time (t), although some compositions are
described having speeds in the range of 10 ms. Research has
been focused in the design of improved nonlinear optical
chromophores,10 in order to increase G while maintaining
high values of h, but little effort has been paid to increase the
photorefractive speed. Moerner and Twieg demonstrated
that, at least in certain cases, speed is limited by
photoconductivity (sph) rather than by chromophore
orientation.11 Due to the fact that photoconductivity
Keywords: Fullerene; Trinitrofluorene; Electrochemistry; Charge-transfer
complexes.
Corresponding authors. Tel.: C34 96 6658408; fax: C34 96 6658351
(A.S.); tel.: C1 864 6565017; fax: C1 864 6566613 (L.E.);
e-mail addresses: asastre@umh.es; luis@clemson.edu
*
0040–4020/$ - see front matter q 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.07.124

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2103
depends on charge generation efficiency, the improvement
of the latter should enhance the performance of the material.
This leads to the need of a better photosensitizer.
covalently linked in order to benefit of a synergic
collaboration between them. Thus, C60–TNF dyads 1a–c
were synthesized in which the distance between the two
subunits was regulated by the position of the phenyl
substitution from ortho to meta to para. The photorefractive
properties of their compositions with PVK and 4-piper-
idindicyanostyrene (PDCST) were determined and
compared with those of the corresponding compositions
bearing C60. While the composite with C60 showed a
slightly higher gain coefficient, the compositions with 1a–c,
proved to be 2–3 times faster in grating formation, that is,
they had a response time 2–3 times lower when high electric
fields were applied.22
Charge photogeneration efficiency is governed by a
competition between recombination of a charge carrier
with its parent countercharge, termed geminate recombi-
nation, and electron–hole pair dissociation.12 In order to
favor the second possibility, a fast electron transfer from
another molecule must take place, that is, an adequate
relationship between HOMO and LUMO orbitals of
photosensitizer and charge transporting molecule must be
achieved. The best performance is obtained by optimizing
the charge-transfer (CT) properties between a given
sensitizer (an acceptor-like molecule or moiety) and its
parent transport molecule (a donor-like molecule or
moiety). To date, the most successful class of sensitizers
incorporated into photoconductor polymers have been C60
and 2,4,7-trinitro-9-fluorenone (TNF).
C60 is a unique molecule with a vast array of interesting
properties. Among them, it is an outstanding photosensitizer
due to its long-lived triplet state, high absorption, the
presence of various stable oxidation states and the low
degree of charge recombination.13 It has been found that the
photogeneration efficiency of poly(N-vinylcarbazole)
(PVK) based composites sensitized with C60 is higher
than when other sensitizers are used.14 On the other hand,
the synthesis of functionalized C60 is becoming more and
more interesting due to their remarkable optoelectronic
properties.15 C60 derivative [6,6]-phenyl-C61-butyric acid
methyl ester ([6,6]-PCBM) has been used as photosensitizer
in different photoconducting polymer-based composites
presenting a poor charge-carrier generation efficiency.16
This very positive result prompted us to redesign our
photosensitizers. Here we present the synthesis, characteri-
zation and electrochemical study of dyads 2 and 3 bearing a
flexible polyether chain to enhance the solubility in organic
solvents, to avoid phase segregation, to allow a better
interaction between the moieties, to introduce the possibility
of ion complexation, and prepared to work in the near
infrared region.
On the other hand, TNF is a strong electron acceptor, even
better than C60, a characteristic which allows the formation
of robust CT complexes with charge transporting molecules
of the carbazol series.17 Thus, the covalent linking of C60
and trinitrofluorenone would render a photosensitizer
combining the exceptional properties of each moiety, and
is expected to optimize the charge-transfer properties with
PVK increasing the charge generation efficiency.
Another point of interest related to the photosensitizer is the
laser wavelength needed for the photoexcitation. The election
depends on the absorption profiles of the photosensitizer and
the nonlinear chromophore. In fact, absorption due to the
chromophore should be avoided to reduce losses, while the
photosensitizer should show enough absorption to allow
charge generation, but not enough to minimize losses.
Therefore, photosensitizers in the near infrared domain are
interesting. Moreover, sensitizing in this region allows to use
2. Results and discussion
2.1. Synthesis of 2,5,7-trinitrofluorenone–C60 (TNF–C60)
and 2,5,7-trinitro-9-dicyanomethylenfluorene–C60
(TNDCF–C60) dyads
The hydroxyformyl derivative 4 was prepared by esterifi-
cation of 4-formylbenzoic acid with tetraethylene glycol
using N,N-dicyclohexylcarbodiimide (DCC) and
4-(dimethylamino)pyridine (DMAP). The reaction of 4
with 2,5,7-trinitrofluorenone-4-carboxylic acid chloride
(5)23 in CH2Cl2 in the presence of DMAP and triethylamine
afforded the corresponding trinitrofluorenone derivative 6.
commercially-available, cheap lasers, which is
a
major consideration if these materials are to be used in
technologically useful devices. While C60 and TNF are
irradiated at 633 nm, 2,4,7-trinitro-9-dicyanomethylen-
fluorene (TNDCF)18–20 operates at 830 nm. [6,6]-PCBM
was also utilized as a near infrared photosensitizer in
photorefractive polymer composites.21
Finally, the synthesis of C60–TNF derivative 2 was
accomplished by Prato’s reaction24 from aldehyde 6 with
sarcosine and C60 (Scheme 1).
With all these considerations in mind, we decided to
synthesize dyads containing C60 and TNF moieties

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2104
Scheme 1.
The synthesis of the 2,5,7-trinitro-9-dicyanomethylen-
fluorene–C60 dyad 3 was envisaged by two different vias
(Scheme 2): (a) Knoevenagel reaction of TNF–C60
derivative 2 and malononitrile, and (b) Prato reaction
between the trinitrodicyanomethylenfluorene aldehyde
derivative 7, sarcosine and C60. Thus, reaction of
TNF–C60 derivative 2 with an excess of malononitrile in
DMF at room temperature over a long duration afforded the
adduct TNDCF–C60 3 in a good (64%) yield.
following a modification of a previously reported procedure
(Scheme 3).
Scheme 3.
1
The new dyads TNF–C60 were fully characterized by H,
13C NMR, MALDI-TOF mass spectrometry and elemental
analysis.
2.2. Spectroscopic studies
Figure 1 shows the UV–vis spectra of 2 and 3, together with
that of C60 for comparison in toluene. The UV–vis spectra of
2 and 3 display strong absorptions due to both the TNF and
the fullerene moieties. Both compounds show a typical 1,2-
methanofullerene UV–vis spectrum, with a sharp absorption
at 430 nm and a broader band at around 700 nm24b showing
Scheme 2.
The synthesis of 2,5,7-trinitro-9-dicyanomethylenfluorene-
4-carboxylic acid chloride (9) from 2,5,7-trinitro-9-dicyano-
methylenfluorene-4-carboxylic acid (8) and SOCl2, followed
by the condensation with the hydroxyaldehyde 4 to obtain
the formyldicyanomethylenfluorene derivative 7 in a one
pot reaction (Scheme 2), gave rise to a mixture of aldehydes
6 and 7 difficult to separate by column chromatography.
This hydrolysis of the dicyanomethylenefluorene aldehyde
7 to the fluorenone aldehyde 6 could be attributed to the
presence of acidic conditions while generating the acid
chloride derivative.
For comparison purposes, methyl 2,5,7-trinitrofluorenone-4-
carboxylate (10, MTNFC) and methyl 2,5,7-trinitrodicyano-
methylenfluorene-4-carboxylate (11, MTNDCM) were
prepared in a good yield from 2,5,7-trinitrofluorenone-
4-carboxylic acid in methanol using DCC and DMAP,
Figure 1. UV–vis spectra of compounds TNF–C60 2 and TNDCF–C60
(solid line) and C60 (dotted line) in toluene as solvent.
3

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2105
that both the TNF and the C60 moieties have little or no
electronic influence on each other in the ground state. It is
noteworthy that compound 3 has a stronger absorption in the
near IR than compound 2, because as a consequence, the
former could have application as near infrared
photosensitizer.
In the presence of an excess of alkaline metal cations (NaC,
KC and RbC) the UV–vis spectra of dyads 2 and 3 do not
show any shift in the absorption bands. A similar result was
observed in 1H NMR. Both observations suggest the
absence of a ground state interaction between the two
chromophores although the complex was formed.
We also studied the ability of the TNF–C60 dyads 2 and 3 to
form intermolecular CT complexes with donors in solution
using N-ethylcarbazole (ECZ) as the p-donor. The
appearance of additional bands in the visible region was
not detected maybe due to the overlapping with the signals
corresponding to the C60 derivative. However, addition of
different amounts of ECZ to a solution of dyads 2 and 3 in
CDCl3 produced pronounced changes both in chemical
shifts and splitting patterns for protons within the fluorene
Figure 3. 1H NMR spectra (CDCl3, 500 MHz, 298 K) of the TNDCF–C60
dyad 3 (A), a 1:1 mixture of 3:ECZ (B), a 1:10 mixture of 3:ECZ (C), a 1:30
mixture of 3:ECZ (D) and a 1:70 mixture of 3:ECZ (E).
1
ring moiety as observed by H NMR (Figs. 2 and 3).
excess. This result is attributable either to an interaction of
the ECZ with another fragment of the ball or to the absence
of any interaction at all between ECZ and the ball. To
provide insight into this question, we performed 13C NMR
in CDCl3 at room temperature of dyads 2 and 3 each in the
presence of a 100-fold excess of ECZ. As already pointed
out for 1H NMR, a dramatic effect on the chemical shifts of
the signals corresponding to the fluorene moieties was
observed (Ddz1.5 and Ddz4 ppm for complexes 2:ECZ
and 3:ECZ, respectively), while only minor changes could
be detected in the signals corresponding to the full-
eropyrrolidine moiety (Ddz0.4 ppm). For the rest of
atoms of the ball, the displacement was so small, if any,
that the close proximity of such a number of signals
precluded any assignation. Thus, it seems that the ECZ
interacts only with the fluorene moiety.
The association constants of the TNF–C60:ECZ and
TNDCF–C60:ECZ complexes were determined at room
temperature from the dependence of the dH8obs on the
Figure 2. 1H NMR spectra (CDCl3, 500 MHz, 298 K) of the TNF–C60 dyad
2 (A), a 1:1 mixture of 2:ECZ (B), a 1:10 mixture of 2:ECZ (C), a 1:30
mixture of 2:ECZ (D) and a 1:70 mixture of 2:ECZ (E).
Upon addition of the same excess of ECZ larger chemical
shifts were observed for the aromatic fluorene protons in
compound 3 than in compound 2. For example, addition of
70 equiv of ECZ give rise to a DdH8Z1.03 ppm for
TNDCF–C60 dyad 3 and to a DdH8Z0.36 ppm for
TNF–C60 dyad 2. These observations agree with the higher
electron withdrawing character of the TNDCF moiety
versus the TNF, leading to a better interaction with ECZ.
1
Interestingly, no change could be detected by H NMR in
the signals corresponding to the fulleropyrrolidine moiety in
both dyads 2 and 3 upon addition of ECZ, even to a 100-fold
Figure 4. Plot of dobs H8 of compound 3 as a function of the concentration of
N-ethylcarbazole, where [3]0Z0.83!10K3 M. Iterative fitting (solid line)
affords DdmaxZ1.73 and the association constant KaZ21.3G0.6 MK1
.

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2106
concentration of ECZ at a constant concentration of 10K3
M
of dyads 2 and 3 by the use of an iterative nonlinear
regression analysis (Fig. 4). The maximum complexation-
induced shifts Ddmax and the association constants Ka
1
(Table 1) were determined by the use of H NMR titration
experiments as described in the Section 4.
Table 1. Comparison of Ddmax and Ka (MK1) for the formation of the
complexes between TNF–C60:ECZ and TNDCF–C60:ECZ in CDCl3 at
25 8C
Ddmax
Ka (MK1
)
TNF–C60
0.59 (H1)
1.14 (H3)
0.98 (H6)
1.36 (H8)
1.36 (H1)
0.50 (H3)
0.93 (H6)
1.73 (H8)
5.1G0.9
TNDCF–C60
21.3G0.6
Figure 6. OSWV of (a) compound 10 and (b) compound 2 (impurity
designated by *).
Job plot analyses were performed to determine the
stoichiometry of the complexes 2:ECZ and 3:ECZ. In the
case of the 3:ECZ complex, the plot of the mole fraction c
(cZ[ECZ]0/[3]0C [ECZ]0) versus the mole fraction
1
multiplied by the complexation-induced H NMR shift of
ECZ, cDdobs, shows a maximum at cZ0.5, which equates
to a 1:1 stoichiometry for the complex (Fig. 5).
Figure 7. OSWV of (a) compound 11 and (b) compound 3 (impurities
designated by *).
fullerene and the substituent. For compound 3, the second
TNF-based reduction overlapped with the first fullerene-
based reduction, resulting in a 2-electron wave.
The sequential addition of N-ethylcarbazole, up to a
100-fold excess, to the solution of compound 2 (Fig. 8)
caused a cathodic shift in the first TNF-based reduction of
Figure 5. The job plot of the complex TNDCF–C60:ECZ (proton H8 of 3 is
reported. The fitting line is calculated with the parameters: DdmaxZ1.73
and KaZ21.3G0.6 MK1
.
2.3. Square wave voltammetric studies
Electrochemical Osteryoung square wave voltammetric
(OSWV) studies of compounds 2 and 3, and their respective
reference compounds 10 and 11, showed reduction waves
corresponding to both the TNF and N-methylfulleropyrro-
lidine moieties (see Table 2, Figs. 6 and 7). In both cases,
the first reduction was slightly shifted anodically,
suggesting only a very small interaction between the
Table 2. Electrochemical reduction data for fluorene derivatives
Reduction (V) versus Fc
E1/2(I)
E1/2(II)
E1/2(III)
E1/2(IV)
K1.56
E1/2(V)
K1.94
2
K0.80
K0.82
K0.45
K0.46
K1.01
K1.17
K1.02
K1.03
K1.13
K2
K1.56
K1.81
10
3
11
K1.78
Figure 8. OSWV of compound 2 before (thin line) and after (thick line)
addition of 100 equiv of N-ethylcarbazole.

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2107
38 mV, thus confirming the NMR results suggesting that the
p-donor formed a complex with 2, which increased the
electron density on the TNF moiety. The first fullerene-
based reduction was also shifted by a comparable amount,
which is reasonable since C60 is also known to be a good
electron acceptor. As a result, the already reduced
fluorenone moiety would exhibit a strongly decreased
ability to interact with ECZ. In the case of compound 3,
after addition of 100 equiv of N-ethylcarbazole (Fig. 9), the
first TNF-based reduction was shifted cathodically by
70 mV, while the fullerene-based first reduction did not
shift. This observation can be rationalized on the basis of the
increased electron-accepting ability due to the two cyano
groups on the TNF moiety, which increases dramatically its
electron-accepting ability as compared to that of compound
2 to such an extent, that the reduced fluorene moiety still
interacts stronger with ECZ than the fullerene one does.
purification unless otherwise stated. All solvents were
purified using standard procedures. Column Chromato-
graphy: SiO2 (40–63 mm). TLC plates coated with SiO2
60F254 were visualized by UV light. UV–vis spectra were
recorded with a Helios Gamma spectrophotometer and IR
spectra with a Nicolet Impact 400D spectrophotometer.
NMR spectra were measured with a Bruker AC 300 and
with a Bruker AVANCE DRX-500. Mass spectra were
obtained from a Bruker Reflex III matrix-assisted laser
desorption/ionization time of flight (MALDI-TOF)
spectrometer. Elemental analysis were performed on a
Thermo Finnigan Flash 1112 CHN elemental analyzator.
The electrochemical measurements of compounds 2 and 3
including the model compounds methyl 2,5,7-trinitro-
fluorenone-4-carboxylate (10, MTNFC) and methyl 2,5,
7-trinitrodicyanomethylenfluorene-4-carboxylate (11,
MTNDCM), were performed in a one-compartment
electrochemical cell in oxygen-free tetrahydrofuran (THF)
using a CHI 660A electrochemical workstation. For all
experiments, a 0.7 mM sample concentration was analyzed.
A platinum disk (1 mm) was used as the working electrode
and a platinum wire as the counter electrode. A silver wire
immersed in the solvent mixture with 0.1 M supporting
electrolyte and 0.01 M AgNO3 separated from the bulk of
the solution by a Vycor glass served as the pseudo-reference
electrode. The supporting electrolyte used was tetrabuty-
lammonium hexafluorophosphate, TBAPF6 (0.1 M, Fluka,
O99%). Ferrocene was added as an internal standard, and
all redox waves are referenced to the Fc/FcC redox couple.
4.1.1. Synthesis of 11-hydroxy-3,6,9-trioxaundecyl
p-formylbenzoate (4). A mixture of 4-formylbenzoic
acid (1.97 g, 13.1 mmol), tetraethylene glycol (13.47 g,
69.4 mmol), N,N-dicyclohexylcarbodiimide (2.79 g,
Figure 9. OSWV of compound 3 before (thin line) and after (thick line)
addition of 100 equiv of N-ethylcarbazole.
13.5 mmol),
4-(dimethylamino)pyridine
(160 mg,
1.3 mmol) and dry toluene (50 ml) was stirred at 65 8C
under argon atmosphere for 5 h. The reaction mixture was
cooled and a white solid was filtered off. The organic
solution obtained was washed several times with water,
dried, concentrated under vacuum, and the crude material
was purified by column chromatography (eluent CH2Cl2/
CH3OH 15:1) to afford the product as a colourless oil
(2.54 g, 59%). FT-IR (Ge-ATR) nmax 3457, 1738, 1367,
Preliminary results of the photorefractive performance of
these compounds as sensitizers in polymer composites based
on poly(N-vinylcarbazole) are similar to that of C60, but with
shorter response times and slightly lower gain coefficients.
A more detailed study of these experiments is out of the scope
of this article and will be published elsewhere.
1
1216, 1102 cmK1. H NMR (300 MHz, CDCl3) d 10.09
3. Conclusion
(s, 1H, CHO), 8.21 (d, 2H, JZ8.6 Hz, H-Ar), 7.94 (d, 2H,
JZ8.6 Hz, H-Ar), 4.51 (t, 2H, JZ4.8 Hz, O]C–OCH2),
3.84 (t, 2H, JZ4.8 Hz, O]C–OCH2–CH2), 3.58–3.71 (m,
13H, OH, 6!O–CH2). 13C NMR (125 MHz, CDCl3) d
191.5, 165.3, 138.9, 134.8, 130.0, 129.2, 72.3, 70.4, 70.3,
70.0, 68.8, 64.4, 61.4. FAB-MS: m/z: 327 ([MC1]C). Anal.
Calcd for C16H22O7: C, 58.89; H, 6.80. Found: C, 58.14; H,
7.04.
We have prepared and characterized the new C60–fluorene
adducts 2 and 3 bearing a flexible chain. UV–vis and
electrochemical studies indicate the absence of interaction
between both moieties in the ground state. However,
addition of an excess of the p-donor ECZ to solutions of
compounds 2 and 3 affords complexes in which the
carbazole interacts with the fluorene moiety, as detected
by NMR and OSWV.
4.1.2. Synthesis of 11-(20,50,70-trinitrofluorenone-40-
carboxylate)-3,6,9-trioxaundecyl p-formylbenzoate (6).
A mixture of 4 (790 mg, 2.1 mmol), 2,5,7-trinitro-
fluorenone-4-carboxylic acid chloride (690 mg, 2.1 mmol),
DMAP (20 mg, 0.2 mmol) and dichloromethane (25 ml)
was stirred for 48 h under argon at room temperature. The
reaction mixture was washed with sodium bicarbonate
solution and water. After the solvent was evaporated under
reduced pressure, the crude material was purified by column
4. Experimental
4.1. General
The chemical reagents were purchased reagent-grade from
Aldrich Corporation and were used without further

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2108
chromatography (eluent hexane/ethyl acetate 1:2), to yield
the pure product as an orange oil (500 mg, 35%). FT-IR
6!CH2–O), 2.81 (s, 3H, N–CH3). 13C NMR (125 MHz,
CDCl3) d 166.3, 164.1, 156.0, 153.8, 153.3, 152.9, 152.7,
149.2, 148.7, 147.3, 146.9, 146,5, 146.3, 146.2, 146.1,
145.9, 145.8, 145.7, 145.5, 145.4, 145.3, 145.2, 145.1,
144.7, 144.5, 144.4, 144.3, 143.1, 143.0, 142.6, 142.5,
142.2, 142.1, 142.0, 141.9, 137.6, 137.0, 136.4, 136.0,
135.9, 135.6, 132.5, 130.3, 130.0, 129.0, 124.6, 123.7,
123.0, 111.5, 111.4, 83.1, 70.7, 70.0, 69.2, 69.1, 68.6, 66.3,
65.8, 64.1, 40.0. MALDI-TOF (dithranol): m/z: 1463
([MC1]C), 743 ([MC1]CKC60). Anal. Calcd for
C95H30N6O13: C, 77.98; H, 2.07; N, 5.74. Found: C,
77.47; H, 2.36; N, 5.81.
1
(Ge-ATR) nmax 1724, 1540, 1347 1277, 1103 cmK1. H
NMR (300 MHz, CDCl3) d 10.07 (s, 1H, CHO), 8.94 (d, 1H,
JZ2.1 Hz, H–TNF), 8.89 (d, 1H, JZ2.2 Hz, H–TNF), 8.81
(d, 1H, JZ2.1 Hz, H–TNF), 8.75 (d, 1H, JZ2.2 Hz,
H–TNF), 8.18 (d, 2H, JZ8.3 Hz, H-Ar), 7.92 (d, 2H, JZ
8.3 Hz, H-Ar), 4.53–4.47 (m, 4H, 2!CH2–O–C]O),
3.88–3.83 (m, 4H, 2!CH2–CH2–O–C]O), 3.73–3.67 (m,
8H, 4!CH2–O). 13C NMR (125 MHz, CDCl3) d 191.3,
184.6, 165.0, 164.1, 149.2, 149.0, 146.2, 143.1, 139.3,
138.8, 138.5, 137.4, 134.6, 131.7, 130.4, 129.9,
129.1, 125.0, 122.1, 121.4, 70.3, 68.7, 68.2, 65.8, 64.3.
MALDI-TOF (dithranolCNaI): m/z: 690 ([MCNa]C).
Anal. Calcd for C30H25N3O15: C, 53.98; H, 3.78; N, 6.29.
Found: C, 53.9; H, 4.13; N, 5.72.
4.1.5. Synthesis of methyl 2,5,7-trinitrofluorenone-4-
carboxylate (10). To a mixture of 2,5,7-trinitro-
fluorenone-4-carboxylic acid (400 mg, 1.1 mmol), DCC
(280 mg, 1.4 mmol), DMAP (10 mg, 0.1 mmol) in 10 ml of
dry toluene, 0.45 ml (11.1 mmol) of dry methanol were
slowly added. The mixture was heated at 65 8C for 20 h
under argon atmosphere. After the solvent was evaporated,
the crude material was purified by column chromatography
(eluent hexane/ethyl acetate 4:1), to afford the pure product
as a pale yellow solid (210 mg, 51%). FT-IR (KBr) nmax
4.1.3. Synthesis of 11-(20,50,70-trinitrofluorenone-40-
carboxylate)-3,6,9-trioxaundecyl p-(N-methyl-3,4-fullero-
pyrrolidin-2-yl)benzoate (2). A mixture of C60 (180 mg,
0.3 mmol), 6 (500 mg, 0.7 mmol), sarcosine (70 mg,
0.8 mmol) and 1,2-dichlorobenzene (30 ml) was heated
under argon for 7 h. After flash chromatography (eluent
toluene/diethyl ether 9:1) the desired product was isolated as
a brown solid (110 mg, 31%). 1H NMR (300 MHz, CDCl3)
d 8.95 (d, 1H, JZ2.1 Hz, H–TNF), 8.90 (d, 1H, JZ2.2 Hz,
H–TNF), 8.82 (d, 1H, JZ2.1 Hz, H–TNF), 8.75 (d, 1H, JZ
2.2 Hz, H–TNF), 8.10 (d, 2H, JZ8.1 Hz, H-Ar), 7.90 (m,
2H, H-Ar), 5.01 (d, 2H, JZ9.6 Hz, CHHN), 5.00 (s, 1H,
CHN), 4.53 (t, 2H, JZ4.7 Hz, TNF–CO2–CH2), 4.43 (t, 2H,
JZ4.9 Hz, Ar-CO2–CH2), 4.29 (d, 1H, JZ9.6 Hz, CHHN)
3.91–3.63 (m, 12H, 6!CH2–O), 2.81 (s, 3H, N–CH3).
13C NMR (125 MHz, CDCl3) d 184.6, 166.2, 164.3, 155.9,
153.7, 152.8, 152.6, 149.4, 149.1, 147.3, 147.2, 146.4, 146.3,
146.2, 146.1, 146.0, 145.9, 145.7, 145.6, 145.4. 145.3, 145.2,
145.1, 144.6, 144.5, 144.3, 143.2, 143.1, 142.9, 142.6, 142.5,
142.4, 142.1, 142.0, 141.9, 141.8, 141.7, 141.6, 141.5, 140.1,
139.8, 139.4, 139.3, 138.5, 137.4, 136.9, 136.3, 135.9, 135.6,
131.8, 130.5, 130.0, 129.0, 125.0, 122.2, 21.6, 83.0, 70.6,
70.0, 69.0, 68.5, 66.1, 65.8, 64.1, 40.0. FT-IR (KBr) nmax
1736, 1613, 1594, 1538, 1343, 1093 cmK1. MALDI-TOF
(dithranol): m/z: 1415 ([MC1]C), 695 ([MC1]CKC60).
Anal. Calcd for C92H30N4O14: C, 78.08; H, 2.14; N, 3.96.
Found: C, 78.12; H, 2.55; N, 3.90.
1
1736, 1616, 1594, 1543, 1345 cmK1. H NMR (300 MHz,
CDCl3) d 8.97 (d, 1H, JZ2.1 Hz, H–TNF), 8.89 (d, 1H,
JZ2.2 Hz, H–TNF), 8.83 (d, 1H, JZ2.1 Hz, H–TNF), 8.78
(d, 1H, JZ2.2 Hz, H–TNF), 4.00 (s, 3H, CO2CH3).
4.1.6. Synthesis of methyl 2,5,7-trinitro-9-dicyano-
methylenfluorene-4-carboxylate (11). A mixture of 10
(100 mg, 0.3 mmol), malononitrile (100 mg, 1.5 mmol) and
dry DMF (5 ml), was stirred and heated at 50 8C, under
argon, for 14 h. The reaction mixture was diluted with ethyl
acetate and the residue was washed with brine and water.
The organic layer was dried, concentrated and washed with
methanol to give the product as a yellow-green solid
(70 mg, 62%). FT-IR (KBr) nmax 2232, 1734, 1602, 1536,
1350 cmK1. 1H NMR (300 MHz, CDCl3) d 9.70 (d, 1H, JZ
1.9 Hz, H–TNDCF), 9.62 (d, 1H, JZ2.0 Hz, H–TNDCF),
9.00 (d, 1H, JZ1.8 Hz, H–TNDCF), 8.91 (d, 1H, JZ
2.0 Hz, H–TNDCF), 4.00 (s, 3H, CO2CH3).
Determination of Ka: 1H NMR titration method
4.1.4. Synthesis of 11-(20,50,70-trinitro-90-dicyano-
methylenfluorene-40-carboxylate)-3,6,9-trioxaundecyl
p-(N-methyl-3,4-fulleropyrrolidin-2-yl)benzoate (3). A
mixture of 2 (30 mg, 0.02 mmol), malononitrile (20 mg,
0.3 mmol) and dry DMF (20 ml), was stirred under argon at
room temperature for 72 h. The reaction mixture was diluted
with ethyl acetate and washed with brine and water. The
organic layer was dried, concentrated and the residue was
washed with methanol and diethyl ether to give the product
as a brown solid (20 mg, 64%). FT-IR (KBr) nmax 2199,
The uncomplexed dyad (Host, H) and ethylcarbazole
(Guest, G) are in equilibrium with the 1:1 complex (C)
H CG%C
The association constant Ka is defined by Eq. 1
½CŠ
Ka Z
(1)
(2)
½HŠ½GŠ
and can be rewritten as Eq. 2
1
1720, 1611, 1537, 1342, 1104 cmK1. H NMR (300 MHz,
½CŠ
Ka Z
CDCl3,) d 9.69 (d, 1H, JZ1.8 Hz, H–TNDCF), 9.60 (d, 1H,
JZ2.0 Hz, H–TNDCF), 8.98 (d, 1H, JZ1.9 Hz,
H–TNDCF), 8.93 (d, 1H, JZ2.1 Hz, H–TNDCF), 8.06 (d,
2H, JZ8.9 Hz, H-Ar), 7.91 (m, 2H, H-Ar), 4.99 (d, 2H,
JZ9.6 Hz, CHHN), 4.99 (s, 1H, CHN), 4.48 (t, 2H, JZ
4.8 Hz, TNDCF–CO2–CH2, Ar-CO2–CH2), 4.35 (t, 2H, JZ
5.1 Hz), 4.28 (d, 1H, JZ9.6 Hz, CHHN) 3.84–3.62 (m, 12H,
ð½H0ŠK½CŠÞð½G0ŠK½CŠÞ
where [H0] and [G0] are the starting concentrations of the
dyad and N-ethylcarbazole, respectively.
Under fast-exchange conditions the observed chemical shift
dobs of the dyad will be an averaged value between free (dH)

´
L. Martın-Gomis et al. / Tetrahedron 62 (2006) 2102–2109
2109
9. Meerholz, K.; Volodin, B. L.; Sandalphon, B.; Kippelen, B.;
Peyghambarian, N. Nature (London) 1994, 371, 497–500.
10. Wu¨rthner, F.; Wortmann, R.; Meerholz, K. Chem. Phys. Chem.
2002, 3, 17–31.
and complexed host (dC), as we can see in Eq. 3
½HŠ
½CŠ
dobs
Z
dH C
dC
(3)
½HŠ C½CŠ
½HŠ C½CŠ
11. (a) Wright,D.;Diaz-Garcia,M.A.;Casperson,J.D.;DeClue,M.;
Moerner, W. E.; Twieg, R. J. Appl. Phys. Lett. 1998, 73,
1490–1492.(b)Diaz-Garcia,M.A.;Wright, D.;Casperson, J.D.;
Smith, B.; Glazer, E.; Moerner, W. E.; Sukhomlinova, L. I.;
Twieg, R. J. Chem. Mater. 1999, 11, 1784–1791.
Combination of Eqs. 2 and 3 leads to Eq. 4, that gives the
observed shift for the host as a function of [H0], [G0], dH, dC
and Ka
ꢂꢀ
ꢁ
dH KdC
½G0Š
C
1
dobs ZdH K
1 C
12. Pope, M.; Swenberg, C. E. In Electronic Processes in Organic
Crystals and Polymers; Oxford University Press: New York,
1999.
2
½H0Š ½H0ŠKa
3
5
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢀ
ꢁ
2
½G0Š
1
½G0Š
½H0Š
K
1 C
C
K4
ð4Þ
´
´
´
½H0Š ½H0ŠKa
13. Martın, N.; Sanchez, L.; Illescas, B.; Perez, I. Chem. Rev.
1998, 98, 2527–2548 and references cited therein.
14. Hendrickx, E.; Zhang, Y. D.; Ferrio, K. B.; Herlocker, J. A.;
Anderson, J.; Armstrong, N. R.; Mash, E. A.; Persoons, A. P.;
Peyghambarian, N.; Kippelen, B. J. Mater. Chem. 1999, 9,
2251–2258.
The NMR studies were performed on a Bruker Avance DRX
500 MHz at 298G1 K in CDCl3, the concentration of the
dyad was fixed at 0.99!10K3 and 0.83!10K3 M for dyad
2 and dyad 3, respectively. Increasing amounts of ECZ were
added while the total volume of the resulting solution was
kept constant (0.5 ml), and the complexation-induced shift
of the protons of the fluorene moiety (dobs) were observed.
Fitting of these data to the 1:1 binding isotherm by iterative
methods25 delivered the parameters Ka and DdmaxZdHKdC.
´
15. Guldi, D. M.; Martın, N. In Fullerenes: From Synthesis to
Optoelectronic Properties; Kluwer Academic: The Nether-
lands, 2002.
16. Mecher, E. H.; Brauchle, C.; Horhold, H. H.; Hummelen, J. C.;
Meerholz, K. Phys. Chem. Chem. Phys. 1999, 1, 1749–1756.
17. Perepichka,D.F.;Bryce,M.R.;Perepichka,I.F.;Lyubchik,S.B.;
Christensen, C. A.; Godbert, N.; Batsanov, A. S.; Levillain, E.;
McInnes, J. L. E.; Zhao, J. P. J. Am. Chem. Soc. 2002, 124,
14227–14238 and references cited therein.
Acknowledgements
This work was financially supported by the Spanish
Government (CICYT Grant no. BQU2002-04513-C02-01),
by the Spanish Generalitat Valenciana (Grant no. CTIDIA/
2002/33) and by the US National Science Foundation (Grant
CHE-0135786).
18. Kippelen, B.; Marder, S. R.; Hendrickx, E.; Maldonado, J. L.;
Guillemet, G.; Volodin, B. L.; Steele, D. D.; Enami, Y.;
Sandal-phon; Yao, Y. J.; Wang, J. F.; Rockel, H.; Erskine, L.;
Peyghambarian, N. Science 1998, 279, 54–57.
19. Wu¨rthner, F.; Yao, S.; Schilling, J.; Wortmann, R.;
Redi-Abshiro, M.; Mecher, E.; Gallego-Gomez, F.;
Meerholz, K. J. Am. Chem. Soc. 2001, 123, 2810–2824.
20. Ostroverkhova, O.; Wright, D.; Gubler, U.; Moerner, W. E.;
He, M.; Sastre-Santos, A.; Twieg, R. Adv. Funct. Mater. 2002,
12, 621–629.
References and notes
1. Solymar, L.; Webb, D. J.; Grunnet-Jepsen, A. In The Physics
and Applications of Photorefractive Materials; Clarendon:
Oxford, 1996.
´
¨
21. Mecher, E.; Gallego-Lopez, F.; Tillmann, H.; Horhold, H.-H.;
Hummelen, J. C.; Meerholz, K. Nature (London) 2002, 418,
959–964.
2. Ashkin, A.; Boyde, G. D.; Dziedzic, J. M.; Smith, R. G.;
Ballman, A. A.; Levenstein, J. J.; Nassau, K. Appl. Phys. Lett.
1966, 9, 72–74.
´ ´
22. Ortiz,J.;Fernandez-Lazaro,F.;Sastre-Santos,A.;Quintana,J.A.;
´
´
Villalvilla, J. M.; Boj, P.; Dıaz-Garcıa, M. A.; Rivera, J.;
Stepleton, S.; Cox, C.; Echegoyen, L. Chem. Mat. 2004, 16,
5021–5026.
3. Gu¨nter, P.; Huignard, J.-P. In Photorefractive Materials and
Their Applications I and II; Springer: Berlin, 1988.
4. Ostroverkhova, O.; Moerner, W. E. Chem. Rev. 2004, 104,
3267–3314.
23. (a)Sulzberg, T.;Cotter, R. J.J.Org. Chem. 1970, 35, 2762–2769.
(b)Perepichka,I.F.;Popov,A.F.;Orekhova,T.V.;Bryce,M.R.;
Vdovichenko, A. N.; Batsanov, A. S.; Goldenberg, L. M.;
Howard, J. A. K.; Sokolov, N. I.; Megson, J. L. J. Chem. Soc.,
Perkin Trans. 2 1996, 2453–2469. (c) Perepichka, D. F.;
Bryce, M. R.; Batsanov, A. S.; McInnes, E. J. L.; Zhao, J. P.;
Farley, R. D. Chem. Eur. J. 2002, 8, 4656–4669.
5. Sutter, K.; Gu¨nter, P. J. Opt. Soc. Am. B 1990, 7, 2274–2278.
6. Ducharme, S.; Scott, J. C.; Twieg, R. J.; Moerner, W. E. Phys.
Rev. Lett. 1991, 66, 1846–1849.
7. (a) Moerner, W. E.; Silence, S. M. Chem. Rev. 1994, 94,
127–155. (b) Moerner, W. E.; Grunnet-Jepsen, A.; Thompson,
C. L. Ann. Rev. Mater. Sci. 1997, 27, 585–623.
24. (a) Maggini, M.; Scorrano, G.; Prato, M. J. Am. Chem. Soc.
1993, 115, 9798–9799. (b) Prato, M.; Maggini, M. Acc. Chem.
Res. 1998, 31, 519–526.
8. (a) Wang, Y.; West, R.; Yuang, C.-H. J. Am. Chem. Soc. 1993,
115, 3844–3845. (b) Wang, Y. Nature (London) 1992, 356,
585–587. (c) Silence, S. M.; Walsh, C. A.; Scott, J. C.;
Moerner, W. E. Appl. Phys. Lett. 1992, 61, 2967–2969.
25. A nonlinear regression analysis of Eq. 4 (see Section 4) was
performed by the program OriginPro 7.5.