Siloxanes with pendent naphthalene diimides: Synthesis and fluorescence quenching

Article abstract of DOI:10.1021/ol070562o

Cyclic siloxanes with pendent naphthalene diimide groups were synthesized via hydrosilylation to form amorphous electron-accepting compounds. Photophysical measurements and >99.9% fluorescence quenching of well-known p-type polymers by the siloxanes demonstrate that these siloxanes form a new class of highly efficient n-type materials that provide some control over intermolecular interactions.

Full text of DOI:10.1021/ol070562o

ORGANIC
LETTERS
2
007
Vol. 9, No. 12
297-2300
Siloxanes with Pendent Naphthalene
Diimides: Synthesis and Fluorescence
Quenching
2
Palaniswamy Ganesan,†,‡ Barend van Lagen, Antonius T. M. Marcelis,
†
†
†
,†
Ernst J. R. Sudho
1
lter, and Han Zuilhof*
Laboratory of Organic Chemistry, Wageningen UniVersity, Dreijenplein 8, 6703 HB
Wageningen, The Netherlands, and Dutch Polymer Institute (DPI), P.O. Box 902,
5600 AX EindhoVen, The Netherlands
han.zuilhof@wur.nl
Received March 6, 2007
ABSTRACT
Cyclic siloxanes with pendent naphthalene diimide groups were synthesized via hydrosilylation to form amorphous electron-accepting compounds.
Photophysical measurements and >99.9% fluorescence quenching of well-known p-type polymers by the siloxanes demonstrate that these
siloxanes form a new class of highly efficient n-type materials that provide some control over intermolecular interactions.
5
Over the past decade, there has been a rapidly increasing
interest in the development of organic optoelectronic materi-
als. In the devices based on these materials the overall order,
any structural defect, π-π stacking, and (partial) crystal-
lization of the material pose constraints on the usefulness of
photovoltaic devices. Our interest is focused on the preven-
tion of large-scale crystallization and π-π stacking, and we
aim to obtain a homogeneous morphology of the photoactive
layer by developing amorphous materials. Recently, we
6
reported the synthesis and optoelectronic properties of a
novel n-type amorphous tetrahedral tetraphenylmethane-
based molecule, which yielded upon mixing with p-type
1
such a material. While nanoscopic phase separation is
essential for the proper performance of, e.g., organic pho-
tovoltaic solar cells, large-scale crystallization and stacking
are not desired. In this case, both phase separation between
n- and p-type materials and π-π overlap domains with
length scales on the order of the exciton diffusion length
(3) (a) Rittner, M.; Baeuerle, P.; Goetz, G.; Schweizer, H.; Calleja, F. J.
B.; Pilkuhn, M. H. Synth. Met. 2006, 156, 21-26. (b) Yang, H. C.; Shin,
T. J.; Yang, L.; Cho, K.; Ryu, C. Y.; Bao, Z. N. AdV. Funct. Mater. 2005,
1
5, 671-676. (c) Singh, T. B.; Gunes, S.; Marjanovic, N.; Sariciftci, N. S.;
2
Menon, R. J. Appl. Phys. 2005, 97, 114508-1-5.
4) (a) Bozano, L. D.; Carter, K. R.; Lee, V. Y.; Miller, R. D.; DiPietro,
are desired for an effective charge separation. There have
(
been many attempts to bring controlled order to organic
optoelectronic materials, and thus improve the morphology
R.; Scott, J. C. J. Appl. Phys. 2003, 94, 3061-3068. (b) Fichet, G.; Corcoran,
N.; Ho, P. K. H.; Arias, A. C.; MacKenzie, J. D.; Huck, W. T. S.; Friend,
R. H. AdV. Mater. 2004, 16, 1908-1912.
3
4
of the active materials used for organic FETs, LEDs, and
(
5) (a) Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.;
Durrant, J. R.; Bradley, D. D. C.; Giles, M.; McCulloch, I.; Ha, C. S.; Ree,
M. Nat. Mater. 2006, 5, 197-203. (b) Sun, X. B.; Zhou, Y. H.; Wu, W.
C.; Liu, Y. Q.; Tian, W. J.; Yu, G.; Qiu, W. F.; Chen, S. Y.; Zhu, D. B. J.
Phys. Chem. B 2006, 110, 7702-7707. (c) Sivula, K.; Ball, Z. T.; Watanabe,
N.; Frechet, J. M. J. AdV. Mater. 2006, 18, 206-210. (d) Ma, W. L.; Yang,
C. Y.; Gong, X.; Lee, K.; Heeger, A. J. AdV. Funct. Mater. 2005, 15, 1617-
1622.
†
Wageningen University.
Dutch Polymer Institute.
‡
(1) (a) Special Issue on Organic Electronics. Chem. Mater. 2004, 16,
4
3
381-4846. (b) Katz, H. E.; Bao, Z. N.; Gilat, S. L. Acc. Chem. Res. 2001,
4, 359-369.
(2) Hoppe, H.; Sariciftci, N. S. J. Mater. Res. 2004, 19, 1924-1945.
1
0.1021/ol070562o CCC: $37.00
© 2007 American Chemical Society
Published on Web 05/15/2007

polymers a homogeneous morphology and a very high
conductivity in time-resolved microwave conductivity ex-
periments. While the resulting blend was amorphous in
nature, the structural rigidity of this molecule precluded
further optimization of the charge-transfer paths. Therefore
we aimed to develop amorphous n-type materials in which
the electron-accepting moieties are attached to an inert core
via a flexible linker. This should allow intra- and intermo-
lecular structural reorganization that may lead to better π-π
overlap between adjacent molecules, and thus to more
efficient charge-transfer channels, without compromising the
amorphous nature of the material. Siloxanes are attractive
for this purpose, as they have a flexible, yet photochemically
and electrically inert backbone. Cyclic oligomeric siloxanes
provide the additional benefit of a well-defined size-
dependent structure. While they have been used for ion
under argon at 70 °C (Scheme 1). The synthesis of starting
material 4 and the final products 1, 2, and 3 is described in
detail in the Supporting Information. Dimer 1 and tetramer
2 were obtained in a reasonable overall yield of around 50%,
but pentamer 3 could only be obtained in a very low yield
(∼1.5%), which we attribute to the previously reported
inaccessibility of the fifth hydrogen atom in the ring after
the initial four substitutions. Addition of more alkene and/
or catalyst or increase of the reaction temperature did not
help to drive the reaction further to completion.
1
0
Unlike compound 1, compounds 2 and 3 display more than
1
one signal for the naphthalene aromatic rings in H NMR
spectra (Figure 1a). Though the peak patterns of 2 and 3
7
8
transport, liquid crystalline materials, and organometallic
9
reactions, this class of materials has, as of yet, not been
used in optoelectronic materials. As part of our strategy we
therefore report on the synthesis, characterization, and
photophysical properties of three novel siloxanes with
pendent naphthalene diimide (NDI) chromophores (1-3;
Scheme 1), together with their fluorescence quenching of
p-type polymers.
1
Figure 1. (a) H NMR spectra of siloxanes in the aromatic region
(300 MHz); (b) possible stereoisomers of 2.
Scheme 1. Synthesis of Siloxanes Using Karstedt’s Catalyst
might look like a multiplet because of its symmetrical nature,
variation of the magnetic field strength (300 vs 400 MHz)
revealed a field-dependent splitting between the peaks. As
a result, the splitting patterns are not due to proton-proton
coupling, but are caused by a set of stereoisomers that display
singlet peaks with different chemical shifts from different
aromatic rings.
In the case of tetramer 2 four stereoisomers are possible
(see Figure 1b) with a statistical distribution of 2:4:1:1. The
observed ratios could be different from the statistical
distribution, because of epimerization that allows equilibrium
1
1
distributions. The observance of six rather than of four
1
peaks in the H NMR spectrum can be explained by studying
the environment of the NDI rings in isomers 2a-d. In
compounds 2a every ring is oriented cis to one neighboring
ring, and trans to the other, while it is oriented trans with
respect to the aromatic ring across the siloxane ring. As a
result all aromatic rings are in the same environment, and
show up as one peak. Analogous arguments apply for 2c
and 2d, which yield peaks at different δ-values than 2a due
to differences in the environment. However, the NDI rings
in 2b are not all identical, and show up as three singlets in
a ratio of 1:2:1; given that 2b amounts statistically to half
Compounds 1-3 were synthesized via hydrosilylation
reactions catalyzed by Karstedt’s Pt -catalyst in dry toluene
0
(
6) Ganesan, P.; Yang, X. N.; Loos, J.; Savenije, T. J.; Abellon, R. D.;
Zuilhof, H.; Sudh o¨ lter, E. J. R. J. Am. Chem. Soc. 2005, 127, 14530-
4531.
7) Zhang, Z. C.; Lyons, L. J.; Jin, J. J.; Amine, K.; West, R. Chem.
Mater. 2005, 17, 5646-5650.
8) (a) Liu, L. M.; Zhang, B. Y.; He, X. Z.; Cheng, C. S. Liq. Cryst.
1
of the distribution, 6 signals are to be expected in the H
NMR with a signal ratio of 1:1:2:2:1:1, as is experimentally
observed. In fact, this is the first reported case in which the
1
(
(
2
1
004, 31, 781-786. (b) Lacey, D.; Mann, T. E. Liq. Cryst. 2003, 30, 1159-
(10) Harrod, J. F.; Pelletier, E. Organometallics 1984, 3, 1064-1069.
(11) (a) Unno, M.; Kawaguchi, Y.; Kishimoto, Y.; Matsumoto, H. J.
Am. Chem. Soc. 2005, 127, 2256-2263. (b) Pelletier, E.; Harrod, J. F.
Organometallics 1984, 3, 1070-1075. (c) Moore, C. B.; Dewhurst, H. A.
J. Org. Chem. 1962, 27, 693-694.
170.
(9) (a) Decken, A.; Passmore, J.; Wang, X. P. Angew. Chem., Int. Ed.
2
006, 45, 2773-2777. (b) Ramirez-Oliva, E.; Cuadrado, I.; Casado, C. M.;
Losada, J.; Alonso, B. J. Organomet. Chem. 2006, 691, 1131-1137.
2298
Org. Lett., Vol. 9, No. 12, 2007

ratios match exactly the statistical distribution, since epimer-
Nanosecond laser flash photolysis (LFP) studies on silox-
11a
ization is blocked by the bulky substituents at the Si atom.
anes 1, 2, and 3 in chloroform reveal the triplet spectrum of
11b
6
Analogously, for pentamer 3 four stereoisomers are possible,
the NDI moieties (Figure 3). Since the higher triplet states
giving rise to ten different environments for the NDI rings
due to splitting of peaks of three isomers into a 2:2:1 ratio
(see the Supporting Information). Since the addition of the
fifth substituent is highly activated (vide supra), no statistical
mixture is to be expected, in line with the observation of
1
seven peaks in the H NMR spectrum of 3 with an observed
peak ratio (%) of 7.7:9.8:11.6:31.8:14.0:14.5:10.6.
Absorption spectra of 1-3 obtained at micromolar con-
centrations showed a normal absorption of NDI with
6
vibrational features (Figure 2a). For 1, but not for 2 and 3,
Figure 3. Transient absorption spectra of 1 (blue), 2 (green), and
3
(red) in chloroform: (a) at low concentration [1, 100 µM; 2, 50
µM; and 3, 20 µM] and (b) at higher concentration [1, 0.2 mM; 2,
.1 mM; and 3, 79 µM (λexc ) 355 nm; fwhm ) 4 ns).
0
lie close to the excited singlet states, the intersystem crossing
is very efficient and rapidly deactivates the excited singlet
states. This leads to low fluorescence quantum yields and
1
3
short singlet lifetimes. These triplet spectra were vibra-
tionally (456 and 485 nm) resolved up to a certain concentra-
tion. Above this concentration (for 1, 0.1-0.2 mM; for 2,
0
.005-0.01 mM; for 3, 0.008-0.01 mM) the spectra were
red-shifted with loss of vibrational resolution. The red shift
in the transient absorption maxima is more pronounced in
dimer 1 (510 nm) than in 2 and 3 (505 nm). At these
concentrations, an extra shoulder was seen in the UV-vis
spectrum of 1 indicating the formation of ground-state
aggregates as shown in Figure 2b. Such a shoulder was not
seen in UV-vis spectra of 2 and 3. However, for these
compounds, it is also expected that intermolecular π-π
stacking of chromophores becomes more significant at higher
concentrations, resulting in both changes in the vibrational
Figure 2. (a) Absorption and emission (λexc ) 381 nm) spectra of
dimer 1 (blue, 5.04 µM), tetramer 2 (green, 2.54 µM), and pentamer
3
(red, 1.96 µM) in CHCl
equal chromophore concentrations. (b) Absorption spectra of dimer
at higher concentrations (2-10 mM). (c) Emission spectra of 1
solid line) and 2 (dotted line) in ODCB (∼8 µM).
3
; concentrations were chosen to yield
1
(
an extra absorption band around 450 nm appears upon
increasing the concentration to the millimolar range, which
we attribute to ground state aggregation (Figure 2b). In
contrast, the steady-state emission spectra of all three
compounds displayed, over the whole measurable concentra-
tion range, two distinct features, namely, the monomer (410
nm) and exciplex (510 nm) emission. This kind of emission
1
4
features and the red shift in the triplet absorption maxima.
As this was indeed observed in pure films of both 1 and 2
(see the Supporting Information), we hypothesize that
intermolecular complexes are ubiquitously present.
The presence of intermolecular complexes at higher
concentrations was also confirmed by time-resolved nano-
second fluorescence spectroscopy. Immediately after the laser
pulse, at low concentrations no excimer/exciplex peak was
seen (4, 5, and 10 µM, for 3, 2, and 1, respectively), whereas
at 50-fold higher concentrations an excimer component was
clearly observed (Figure 4; λexc ) 355 nm; fwhm ) 4 ns).
Measurement of the fluorescence 6 ns after the pulse yields
-
7
was observed even at submicromolar concentrations (10
M), indicating that the emission at 510 nm is due to
intramolecular excimer formation.
The position and intensity of the fluorescence of 1-3 are
significantly influenced by the solvent (Figure 2c). On going
from chloroform to o-dichlorobenzene (ODCB), the mono-
mer fluorescence of 2 enhanced with a shift from 410 to
(12) (a) Barros, T. C.; Brochsztain, S.; Toscano, V. G.; Berci, P.; Politi,
430 nm, while the excimer fluorescence decreased in
M. J. J. Photochem. Photobiol. A 1997, 111, 97-104. (b) Chatterjee, S.;
Basu, S.; Ghosh, N.; Chakrabarty, M. Spectrochim. Acta, Part A 2005, 61,
intensity. This phenomenon is attributed to π-π stacking
1
887-1891.
13) Ganesan, P.; Baggerman, J.; Zhang, H.; Sudh o¨ lter, E. J. R.; Zuilhof,
H. Submitted for publication.
14) Chen, Z. J.; Stepanenko, V.; Dehm, V.; Prins, P.; Siebbeles, L. D.
1
2
of the aromatic solvent with the chromophores, which
diminishes excimer formation for 2. For low concentrations
of compound 1 this π-π stacking in ODCB even leads to
disappearance of the excimer peak.
(
(
A.; Seibt, J.; Marquetand, P.; Engel, V.; Wurthner, F. Chem.-Eur. J. 2007,
13, 436-449.
Org. Lett., Vol. 9, No. 12, 2007
2299

Figure 5. Fluorescence quenching of MDMO-PPV (a) and P3HT
b) by 1 (blue), 2 (green), and 3 (red) in 1:1 wt % films spin coated
(
from o-dichlorobenzene at 2000 rpm for 60 s.
indicates a highly efficient electron transfer from the p-type
material to the NDI moieties. In fact, MDMO-PPV fluores-
cence was quenched by 1 and 2 (>99.9% in 1:1 films) even
more efficiently than previously reported for the tetrahedral
Figure 4. Time-resolved emission spectra of siloxanes in CHCl
3
at low (black line) (1, 10 µM; 2, 5 µM; and 3, 4 µM) and high (red
line) (1, 0.5 mM; 2, 0.25 mM; and 3, 0.2 mM) concentrations,
immediately (a-c) and 6 ns (d-f) after the pulse.
6
tetraphenylmethane derivative with NDI moieties, which
make them highly promising materials for further (photo-)-
electrical studies. The quenching efficiencies of all three
siloxanes were almost equal upon 1:1 (wt %) mixing with
P3HT (Figure 5b). These quenching experiments in film with
two different well-known donor polymers point to the
versatility of these siloxanes.
a clear excimer emission already at the lower concentrations.
At higher concentrations, the chromophores form ground
state dimers, which are excited and yield the exciplex
emission immediately after the pulse. At lower concentrations
the formation of ground state aggregates is less likely.
Femtosecond transient absorption and picosecond fluores-
cence spectroscopies show the singlet lifetime of NDI to be
In conclusion, we have synthesized new cyclic and
amorphous siloxanes, which consist of sets of stereoisomers
bearing electron-accepting NDI moieties. The size of these
NDI moieties prevents the interconversion of these isomers,
via, e.g., epimerization, which for the first time allows the
use of cyclic siloxanes to obtain a stable and controllable
morphology. Photophysical measurements display that in-
termolecular interactions of acceptor chromophores occur at
higher concentrations. This opens up the possibility to create
continuous pathways for charge transport in appropriate
films, which is under current investigation. A major stimulus
for such further studies is the highly efficient quenching of
the fluorescence of various p-type polymers by the NDI
moieties, which points to the potential of these amorphous
siloxanes as n-type materials in applications such as organic
solar cells.
∼
10 ps.13 This is confirmed by picosecond time-correlated
single photon counting measurements (TCSPC) of 1-3 in
chloroform, which show a multiexponential decay with
lifetimes ranging from 13 ps to 18 ns (see the Supporting
Information). The shortest observed lifetime is in line with
the ∼10 ps singlet lifetime of NDI. Therefore, excimer
formation by diffusion also is not possible. Since NDIs form
triplets efficiently, the excimer fluorescence is likely due to
triplet-triplet annihilation, in which the diffused molecules
15
stay together. This also explains why the exciplex intensity
increases over time from directly after the pulse to 6 ns
thereafter.
Polarized optical microscopy shows that dropcasted neat
films of 1-3 are amorphous, and this is also the case for
films in which these siloxanes are mixed with well-known
polymeric p-type materials such as poly(2-methoxy-5-{3′,7′-
dimethyloctyloxy}-p-phenylene vinylene) [MDMO-PPV] and
poly-(3-hexylthiophene) [P3HT]. Next, the fluorescence
quenching of these p-type polymers was tested, both in
solution and in films. Figure 5 shows the fluorescence
quenching in a 1:1 (wt %) mixture of siloxanes 1-3 with
MDMO-PPV and P3HT, respectively.
Acknowledgment. The authors thank the Dutch Polymer
Institute (Functional Polymer Systems, Project 324) for
financial support, Dr. T. Savenije (TU Delft) for P3HT
polymer and helpful discussions, Dr. M. Koetse (TNO
Industries) for MDMO-PPV, and A. van Hoek (WUR) for
help with the TCSPC. Finally, we thank two reviewers of
an earlier version of this paper for helpful comments.
Supporting Information Available: Synthesis and NMR
data of siloxanes 1, 2, and 3 and the starting alkene 4, as
well as TCSPC data for 1 and 2. This material is available
free of charge via the Internet at http://pubs.acs.org.
All three siloxanes 1-3 were very efficient in quenching
the singlet excited states of the p-type materials. This
(15) Lee, S. K.; Zu, Y. B.; Herrmann, A.; Geerts, Y.; Mullen, K.; Bard,
A. J. J. Am. Chem. Soc. 1999, 121, 3513-3520.
OL070562O
2300
Org. Lett., Vol. 9, No. 12, 2007