Using two simple methods of Ar-ArF self-assembly and isolation chromophores to further improve the comprehensive performance of NLO dendrimers

Article abstract of DOI:10.1002/chem.201202992

Herein, through a combination of divergent and convergent approaches, coupled with the utilization of the powerful Sharpless "click chemistry" reaction, two series of high-generation nonlinear optical (NLO) dendrimers have been conveniently prepared in high purity and satisfactory yields. Perfluoroaromatic rings and isolation chromophores were introduced to further improve their comprehensive performance. Thanks to the effects of Ar-Ar^F self-assembly and the isolation chromophores, coupled with perfect 3D spatial isolation from the highly branched structure of the dendrimer, G5-PFPh-NS displayed very large NLO efficiency (up to 257 pm V ^-1), which is, to the best of our knowledge, the new record highest value reported so far for simple azo chromophore moieties. High-quality wide optical transparency and good stability were also achieved. Copyright

Full text of DOI:10.1002/chem.201202992

DOI: 10.1002/chem.201202992
F
ꢀ
Using Two Simple Methods of Ar Ar Self-Assembly and Isolation
Chromophores to Further Improve the Comprehensive Performance of
NLO Dendrimers**
Wenbo Wu,[a] Gui Yu,[b] Yunqi Liu,[b] Cheng Ye,[b] Jingui Qin,[a] and Zhen Li*[a]
Abstract: Herein, through a combina-
tion of divergent and convergent ap-
proaches, coupled with the utilization
of the powerful Sharpless “click chem-
istry” reaction, two series of high-gen-
eration nonlinear optical (NLO) den-
drimers have been conveniently pre-
pared in high purity and satisfactory
yields. Perfluoroaromatic rings and iso-
lation chromophores were introduced
to further improve their comprehensive
branched structure of the dendrimer,
G5-PFPh-NS displayed very large
NLO efficiency (up to 257 pmVꢀ1),
which is, to the best of our knowledge,
the new record highest value reported
so far for simple azo chromophore
moieties. High-quality wide optical
transparency and good stability were
also achieved.
performance. Thanks to the effects of
F
ꢀ
Ar Ar self-assembly and the isolation
chromophores, coupled with perfect
3D spatial isolation from the highly
Keywords: isolation chromophores ·
click chemistry · dendrimers · non-
linear optics · self-assembly
Introduction
the shape of a spherical-shaped chromophore by the intro-
duction of isolation groups) was considered to be an effi-
cient approach for minimizing these interactions and en-
hancing the poling efficiency.[3a,4] Based on this principle,
various structures of chromophores and other materials
have been developed, from linear to various other shapes,
such as L-shaped, X-shaped, H-shaped, star-shaped, and so
on, to improve their topological structures.[5] Among these
topologies, dendritic NLO materials, including hyper-
branched polymers, dendronized polymers, and dendrimers,
have been considered to be a very promising molecular top-
ology for the next generation of highly efficient NLO mate-
rials, owing to their special 3D global highly branched struc-
ture (Figure 1).[6,7]
Recently, based on the excellent work mentioned above,
as well as our previous conclusions on “the concept of a
suitable isolation group”,[8] we designed and synthesized a
series of NLO dendrimers, G1–G5 (see the Supporting In-
formation, Schemes S1–S3),[9] that demonstrated excellent
NLO activities (the d33 value of G5 was up to 193.1 pmVꢀ1,
which was the highest value for simple azo chromophore
moieties at that time). On increasing the loading density of
the chromophore moieties, the tested NLO effects also in-
creased, which indicated that the frequently observed
asymptotic dependence of EO activity on the density of
chromophore loading may be overcome through rational
design, in accordance with the predictions of Sullivan, Rob-
inson, Dalton, and co-workers.[10]
Over the past decade, a great deal of effort has been devot-
ed to the search for—and preparation of—organic/polymeric
nonlinear optical (NLO) materials because of their promise
as candidates for future technological applications, such as
frequency doublers, optical-storage devices, and electro-
optic (EO) switches and modulators.[1] Large optical non
earity, low optical loss, and excellent temporal stability of
ACHTUNGTNERlNUNG in-
ACHTUNGTRENNUNG
the dipole orientation are important challenges that scien-
tists face in designing second-order NLO materials.[2,3] Of
these, the efficient translation of the large b values of the or-
ganic chromophores into high macroscopic NLO activities
of the polymers is the major problem, because the strong
dipole–dipole interactions between the chromophore moiet-
ꢀ ꢀ
ies with a donor–p–acceptor (D p A) structure would
make the pole-induced noncentrosymmetric alignment of
chromophore moieties (which is necessary for the materials
to exhibit a NLO effect) a daunting task during the poling
process under an electric field.[3] Fortunately, according to
the experimental results and theoretical analysis of Jen,
Dalton, and co-workers, the site-isolation principle (control
[a] W. Wu, Prof. J. Qin, Prof. Z. Li
Department of Chemistry, Hubei Key Laboratory on Organic and
Polymeric Opto-Electronic Materials, Wuhan University
Wuhan 430072 (P.R. China)
Fax : (+86)27-68756757
E-mail: lizhen@whu.edu.cn
However, was that the limit of NLO dendrimers and
could the NLO performance of dendrimers be enhanced fur-
ther? Thus, additional research is still needed. For chemists,
the general approach for improving the performance of ma-
terials is to modify their chemical structures on the molecu-
lar level. In most of the previous cases, based on the site-iso-
[b] Prof. G. Yu, Prof. Y. Liu, Prof. C. Ye
Organic Solids Laboratories, Institute of Chemistry
The Chinese Academy of Sciences, Beijing 100080 (P.R. China)
[**] NLO: nonlinear optical.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202992.
630
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

FULL PAPER
Figure 1. Our original design concepts for the NLO dendrimers.
lation principle, the isolation groups that were used were
non-polar and could only adjust the shape of the chromo-
phore.[3–9] In fact, the interactions between two isolation
groups and between a chromophore and an isolation group
should also be considered when designing NLO materials.
Very recently, two new ways to use these interactions to im-
prove the performance of NLO materials have been report-
ers consist of a nitro-based chromophore). The d33 value of
G5-PFPh-N was determined to be 206 pmVꢀ1, which was
higher than that of G5. Also, isolation chromophores should
be introduced into dendrimers to further improve their
NLO activities. Model hyperbranched polymers PS5 and
PS6 (see the Supporting Information, Chart S5) were syn-
thesized from an AB4-type monomer that contained two
chromophores with regular structures.[12a] The results
showed that placing the main chromophores at the periph-
ery would enhance its d33 value to a large degree, whereas
the NLO coefficient of the hyperbranched polymers that
contained isolation chromophores at the periphery exhibited
almost no improvement (see the Supporting Information,
Table S1). Thus, dendrimers G2-PFPh-NS–G5-PFPh-NS
(Figure 2; NS indicates that the nitro- and sulfonyl-based
chromophores are regular arranged, to differentiate them
from dendrimers G1-PFPh-N–G5-PFPh-N), which had
nitro-based azo chromophores at their periphery, were also
prepared. Excitingly, the d33 value of G5-PFPh-NS was de-
termined to be 257 pmVꢀ1, which is the new record highest
value reported so far for simple azo chromophore moieties.
Herein, we report the syntheses, characterization, and prop-
erties of these new dendrimers in detail.
ed: The first is the use of the aromatic/perfluoroaromatic
F
ꢀ
(Ar Ar ) self-assembly effect (see the Supporting Informa-
tion, Charts S1 and S2) between two isolation groups, as
proposed by Jen and co-workers (this method was also very
recently confirmed by ourselves);[11] the second is the intro-
duction of isolation chromophores (see the Supporting In-
formation, Chart S3), as reported by ourselves in 2012.[12]
However, reports on the application of these two methods
in the synthesis of high-generation dendrimers are scarce.[12b]
On the other hand, it is known that the preparation of
high-generation dendrimers is typically tedious and expen-
sive because of the need for repetitive protection, deprotec-
tion, and purification steps. Thus, it would be unwise to
modify the chemical structures of dendrimers directly with-
out having any preliminary experience. Herein, we use hy-
perbranched polymers, which have similar topological struc-
tures to dendrimers but are much easier to prepare,[13] as
model compounds. First, we designed and synthesized a
series of azo-chromophore-containing AB2-type hyper-
branched polytriazoles (PS1–PS4; see the Supporting Infor-
mation, Chart S4) that contained perfluoroaromatic rings in
different parts of these hyperbranched polymers.[14] The use
of pentafluorophenyl groups on the periphery produced
higher d33 values than “normal” phenyl groups in the same
positions, whereas the use of perfluoroaromatic rings in the
interior architecture produced slightly lower d33 values (see
the Supporting Information, Table S1). Inspired by this
result, we designed and synthesized a new series of NLO
dendrimers, G1-PFPh-N–G5-PFPh-N (Figure 2; PFPh indi-
cates that their end-capped groups are pentafluorophenyl
moieties to differentiate them from dendrimers G1–G5 that
we reported previously, and N indicates that these dendrim-
Results and Discussion
Synthesis: In general, dendrimers can be constructed
through divergent (a growing pattern from a multivalent
core) or convergent synthetic strategies (a dendron is graft-
ed onto the core).[15,16] However, these tedious, multistep
synthetic procedures often make the preparation of den-
drimers relatively costly and they often encounter several
difficulties, such as incomplete reaction of the end-groups in
the divergent approach and steric hindrance between the re-
active segments and the core molecule in the convergent
strategy.[16a] Fortunately, through a combination of both di-
vergent and convergent approaches (a “double-stage” meth-
od),[7b,17] as proposed by Frꢁchet and co-workers,[17a] the syn-
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
631

Z. Li et al.
ꢁ
ꢁ
tions, G2- -PFPh-N
cates there was
(
indi-
a
terminal
alkyne group in this dendrimer)
could be easily prepared, with-
out the need for protection/de-
protection steps. However, this
approach was only appropriate
for the preparation of low-gen-
eration dendrimers because of
the large steric effects and the
difficulty of chromatographic
separation. For example, in the
ꢁ
synthesis of G2- -PFPh-N, the
yield was not higher than
57.3%, even though the azo-
coupling reaction was conduct-
ed over 5 days, whereas the
ꢁ
yield of G1- -PFPh-N was
79.6%. Thus, the yields of G3-
ꢁ
ꢁ
-PFPh-N and G4- -PFPh-N
in the subsequent steps would
be much lower. On the other
hand, the core compound (G2-
8N3-N, ꢀ8N3 indicates there
were eight azide groups on the
periphery of this dendrimer)
was synthesized through a di-
vergent approach from an AB2-
type monomer (Scheme 2). As
with the convergent approach,
this route was also not appro-
priate for the preparation of
high-generation dendrimers, be-
cause the end-groups did not
react completely, thus leading
to structural defects. However,
G5-PFPh-N could be obtained
through route 3 (Figure 3),
which is a combination of both
divergent and convergent ap-
proaches, from the click reac-
tion between G2-8N3-N and
ꢁ
G2- -PFPh-N. Thus, by using
this “double-stage” approach,
the number of required synthet-
ic steps for the preparation of
high-generation
dendrimers
could be decreased to a large
degree. Meanwhile, the purifi-
cation of G3-PFPh-N, G4-
PFPh-N, and G5-PFPh-N was
Figure 2. Structures of three series of dendrimers (G3 as examples) and their NLO coefficients.
thetic efficiency could be rapidly raised. Thus, this well-de-
signed “double-stage” approach was also used herein to syn-
thesize NLO dendrimers (Figure 3).
Schemes 1–3 show our synthetic routes to G5-PFPh-N. In
a convergent approach (Scheme 1), by taking advantage of a
combination of click chemistry[18] and azo-coupling reac-
very easy, through repeated precipitation of solutions in
THF with EtOAc, by taking advantage of the different solu-
bilities of the product and the reactant (Scheme 3). On the
contrary, if we used a “one-stage” approach, the purification
of high-generation dendrimers would be highly problematic.
Actually, even with this “double-stage” approach, there
632
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

Improving the Performance of NLO Dendrimers
FULL PAPER
Figure 3. Different approaches for the synthesis of the NLO dendrimers used here.
were still different possible routes for the synthesis of high-
thetic route is shown in the Supporting Information,
generation dendrimers (Table 1). For example, G4-PFPh-N
would be the product of “click” reactions between G2-8N3-
ACHTUNGTRENNUNG
Characterization and self-assembly effect: These newly pre-
pared dendrimers were well-characterized by spectroscopic
methods (for detailed results, see the Supporting Informa-
tion). Their FTIR spectra showed absorption bands that
were associated with the nitro groups at about 1515 and
1338 cmꢀ1, as well as a band that was associated with both
nitro and sulfonyl groups (1140 cmꢀ1), thus indicating the
successful introduction of the chromophore moieties into
the dendrimers (see the Supporting Information, Figures S1
and S2). Also, the peak for the azido groups (at about
2100 cmꢀ1) in the azido-containing core disappeared in the
spectra of high-generation dendrimers, whereas an absorp-
tion at 1740 cmꢀ1, which came from the carbonyl groups in
the end-capped chromophores, appeared. These results
showed that the target dendrimers were successfully pre-
pared from the click chemistry reaction by using this
“double-stage” approach. This conclusion was also con-
firmed by NMR spectroscopy. For example, in all of the
1H NMR spectra of the high-generation dendrimers, some
characteristic peaks of the chromophores and of the
COOCH2 protons (about d=4.6 ppm) in the end-capped
groups were observed, whilst the peak at about d=3.6 ppm,
Table 1. Different synthetic routes to dendrimers G3-PFPh-N, G4-PFPh-
N, and G5-PFPh-N.
Compound 1
G1-4N3-N
G2-8N3-N
ꢁ
G0- -PFPh-N
G3-PFPh-N[a]
G4-PFPh-N[a]
G5-PFPh-N[a]
ꢁ
G1- -PFPh-N
G3-PFPh-N
G4-PFPh-N
ꢁ
G2- -PFPh-N
G3-PFPh-N
[a] These routes were used herein.
ꢁ
ꢁ
N and G1- -PFPh-N or between G1-4N3-N and G2-
-PFPh-N. Herein, the first route was used because of the
large difference in molecular weight between the end-
ꢁ
capped dendron (G1- -PFPh-N) and the product (G4-
PFPh-N), which would afford easy purification. For the
same reason, the other high-generation dendrimers were
prepared through their chosen synthetic routes.
The synthetic route to dendrimers G2-PFPh-NS–G5-
PFPh-NS, in which two different chromophores (the sulfon-
yl- and nitro-based ones) are arranged in an orderly
manner, should be similar to that to G1-PFPh-N–G5-PFPh-
N, which only contain one type of chromophore; the syn-
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
633

Z. Li et al.
tures. However, unfortunately,
we failed to detect any signals
within the appropriate mass
range for high-generation den-
drimers (when the generation
number was higher than 3); one
possible reason for this result
might be that their larger mo-
lecular weight resulted in
much-lower peak intensity.
Thus, GPC measurements were
conducted and the results are
summarized in the Supporting
Information, Table S3. The
trends in GPC results were con-
sistent with the actual molecu-
lar weights: with each increase
in the generation number of the
dendrimers, the Mw values also
increased with low PDIs.
Herein, a very interesting phen-
AHCTUNGERTGoNNUN menon was observed. As we
know, GPC analysis by using
linear polystyrenes as calibra-
tion standards often underesti-
mates the molecular weight of
dendrimers with 3D branched
structures because of their quite
different hydrodynamic radii.[19]
On the contrary, if the differ-
ence between the tested value
and the true one were higher,
its hydrodynamic radius should
be more different from that of
linear polymers and its topolog-
ical structure might be closer to
spherical. Herein, on increasing
the generation number of the
dendrimers, the globular shape
improved and the difference
between the measured and cal-
culated values also increased
(Figure 4). In addition, the dif-
ference increased on the intro-
duction of perfluoroaromatic
rings (Figure 4). This indicates
that the topological structure became closer to ideal after
ꢁ
ꢁ
Scheme 1. Synthesis of end-capped dendrons G1- -PFPh-N and G2- -PFPh-N.
which was assigned to the protons in the CH2N3 group, dis-
appeared.
the perfluoroaromatic rings were introduced, especially in
F
ꢀ
Analysis of the dendritic growth by MS (MALDI-TOF)
and by GPC (see the Supporting Information, Table S3) was
performed after each step, as shown in Schemes 1–3 and the
Supporting Information, Schemes S4–S6. MS (MALDI-
TOF) is a good method to confirm the exact molecular
weight of dendrimers; the mass spectra of the prepared den-
drimers are shown in the Supporting Information, Fig-
high-generation dendrimers, possibly owing to an Ar Ar
self-assembly effect. Moreover, compared with Gn-PFPh-N,
Gn-PFPh-NS showed larger differences (Figure 4), which in-
dicated that Gn-PFPh-NS had a more perfect topological
structure. In our previous work, our experimental results
showed that NLO activity would only increase in polymers
with a regular AB structure (see the Supporting Informa-
tion, Chart S3, P4), that is, the isolation chromophore might
be next to the main chromophore.[12] Herein, the GPC re-
ACHTUNGTRENNUNG
634
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

Improving the Performance of NLO Dendrimers
FULL PAPER
peaks for the pentafluorophenyl
groups, which was consistent
with the GPC results that the
higher the generation number,
F
ꢀ
the stronger the Ar Ar self-as-
sembly effect. However, this
result did not confirm that
there was no self-assembly
effect in the lower-generation
dendrimers; perhaps, the self-
assembly effect was just a little
too weak to be observed by
19F NMR spectroscopy.
In our previous analysis of
dendrimers G1–G5, an increase
in generation number led to
better site-isolation effects and
wider optical-transparency win-
dows, which might have been
due to the better 3D topologi-
cal structure of high-generation
dendrimers than low-generation
dendrimers.[9] Presumably, the
exterior benzene moieties and
the interior triazole rings that
surrounded the azo chromo-
phore moieties also played a
key role in shielding the chro-
mophores from the solvatochro-
mic effect. Herein, we con-
F
ꢀ
firmed that the Ar Ar self-as-
sembly effect should make the
topological structure of the
dendrimers even better. Thus,
we wondered whether the pen-
tafluorophenyl groups at the
periphery of these dendrimers
would have an effect on their
transparency. To answer this
question, their UV/Vis spectra
were recorded (see the Sup-
Scheme 2. Synthesis of the dendrimer core, G2-8N3-N.
porting
Information,
Fig-
sults led us to a preliminary conclusion: The interactions be-
tween isolation chromophores and main chromophores
would improve the topological structure of the polymers or
dendrimers, which would further affect their poling behav-
ior.
ACHTUNGTRENuNUGN res S36–S47 and Table S4). As expected, in the solid films,
the dendrimers that contained perfluoroaromatic rings ex-
hibited blue-shifted absorption maxima compared with the
dendrimers that only contained normal phenyl groups in our
previous work (Figure 6), which demonstrated that the more
perfect 3D topological structure could directly impede the
strong intermolecular dipole–dipole interactions between
the chromophore moieties and greatly contribute to the or-
dered noncentrosymmetric alignment during the poling
process. Furthermore, owing to the presence of the sulfonyl
chromophore, all of the maximum absorptions of the den-
drimers that contained the isolation chromophore were
blue-shifted even further compared with those that only
contained one type of chromophore (Figure 6). Taking the
highest-generation dendrimers in this work as an example
To further confirm the GPC results, 19F NMR spectrosco-
py of all of the dendrimers was performed. From the chemi-
cal structure of these dendrimers, three types of F atoms
were distinguished for the pentafluorophenyl groups; corre-
spondingly, in their 19F NMR spectra, there should only be
F
ꢀ
three kinds of peaks. However, if an Ar Ar self-assembly
effect was operative in these dendrimers, some additional
peaks would be observed. The results were encouraging
(Figure 5): if the generation number was higher than three,
several new peaks appeared in addition to the original
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
635

Z. Li et al.
Scheme 3. Synthesis of dendrimers G3-PFPh-N, G4-PFPh-N, and G5-PFPh-N.
On the other hand, these dendrimers were thermally
stable (see the Supporting Information, Figures S47 and
S48); their temperatures for 5% weight loss are listed in
Table 2 (higher than 2308C). Their degradation tempera-
tures (Td) were a little lower than those of dendrimers G1–
G5, possibly owing to the presence of the not-so-stable pen-
tafluorophenyl group. However, this result is already suffi-
cient for NLO materials, because the temperature for their
application in “real” systems could be below 2008C. The
glass-transition temperatures (Tg) of the dendrimers were
also investigated by using differential scanning calorimetry
(DSC; Table 2). Similar to dendrimers G1–G5, the growth
of the dendrimers also resulted in an increase in their Tg
values and the Tg values of the dendrimers that contained
pentafluorophenyl groups were much higher than those that
Figure 4. Ratio between the measured and calculated molecular weights
for different dendrimers; Mc is the calculated molecular weight and Mw is
the measured value (determined by GPC).
only contained phenyl groups (1658C for G5-PFPh-N versus
F
ꢀ
1258C for G5), which further confirmed that the Ar Ar
self-assembly effect existed in these dendrimers. In addition,
higher Tg values would improve the temporal stability of the
dipole orientation of the chromophore in the dendrimers,
which could contribute greatly to their practical application.
(Figure 6B), the maximum absorption of compound G5 was
about 470 nm, which was already quite-largely blue-shifted
compared to the low-generation dendrimers. Moreover,
owing to its more ideal 3D topological structure, the maxi-
mum absorption of G5-PFPh-N was blue-shifted even fur-
ther to 460 nm, whilst the maximum absorption of G5-
PFPh-NS was only 443 nm. This blue-shifted absorption
maximum of the dendrimers would result in wide optical-
transparency windows and contribute to practical applica-
tions in the field of photonics.
NLO properties: All of the NLO dendrimers exhibited good
film-forming ability and their poled films were prepared for
the evaluation of their NLO activity. A convenient tech-
AHCTUNGTREGnNNNU ique to study their second-order NLO activity was to inves-
tigate the second-harmonic generation (SHG) processes,
which are characterized by d33, an SHG coefficient. The test
636
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

Improving the Performance of NLO Dendrimers
FULL PAPER
Figure 6. UV/Vis data of high-generation dendrimers: A) maximum ab-
sorption wavelengths of the dendrimers in thin films, and B) UV/Vis
spectra of the dendrimers in thin films.
In our previous work, the d33 values increased from den-
drimers G1 (100.0 pmVꢀ1) to G5 (193.1 pmVꢀ1), with an ac-
companying increasing in the loading density of the chromo-
phore moieties; moreover, the more perfect the 3D struc-
ture of the dendrimers, the higher the d33 values.[9] Herein, a
similar phenomenon appeared in the dendrimers that con-
tained pentafluorophenyl groups: Changing the loading con-
centration of the chromophore moieties from 0.326 in G1-
PFPh-N to 0.411 in G2-PFPh-N, 0.444 in G3-PFPh-N, 0.459
in G4-PFPh-N, and 0.466 in G5-PFPh-N led to an increase
in the NLO coefficients from 84 in G1-PFPh-N to 89 in G2-
PFPh-N, 108 in G3-PFPh-N, 187 in G4-PFPh-N, and
206 pmVꢀ1 in G5-PFPh-N. These encouraging results should
be also ascribed to their more perfect 3D structures
Figure 5. 19F NMR spectra of: A) dendrimers G1-PFPh to N-G5-PFPh-N,
and B) dendrimers G2-PFPh-NS to G5-PFPh-NS.
and to the isolation effect of the exterior penta-
fluorophenyl moieties and the interior triazole
rings, which could decrease the strong electronic in-
teractions between the chromophore moieties and
enhance their poling efficiency, according to the
“suitable isolation groups” concept; almost the
same result was observed previously with dendri-
Table 2. Physical and NLO data for the dendrimers.
[a]
[b]
[d]
[e]
Dendrimer
Tg
Td
Te[c]
d33
[pmVꢀ1
d33(1)
[pmVꢀ1
]
F[f]
N[g]
[oC]
[oC]
[oC]
U
]
E
G1-PFPh-N
G2-PFPh-N
G3-PFPh-N
G4-PFPh-N
G5-PFPh-N
G2-PFPh-NS
G3-PFPh-NS
G4-PFPh-NS
G5-PFPh-NS
59
78
103
135
165
62
119
127
142
252
243
237
236
233
241
233
228
241
85
105
125
125
135
100
125
125
140
84
89
108
187
206
95
109
179
257
15
16
21
38
42
23
26
44
65
0.12
0.16
0.22
0.28
0.34
0.16
0.26
0.28
0.39
0.326
0.411
0.444
0.459
0.466
0.432
0.464
0.482
0.488
AHCTUNGTRENNGmUN ers G1–G5. On the other hand, in comparison
with dendrimer G5, which only contained normal
phenyl groups at its periphery, G5-PFPh-N demon-
strated
a
higher NLO coefficient (206 pmVꢀ1
versus 193.1 pmVꢀ1), which should be attributed to
[a] The glass-transition temperatures (Tg) of the polymers were determined by DSC
analysis under an argon atmosphere at a heating rate of 108Cminꢀ1. [b] The tempera-
tures of the polymers for 5% weight loss were determined by TGA analysis under a
nitrogen atmosphere at a heating rate of 108Cminꢀ1. [c] The optimal poling tempera-
ture. [d] Second harmonic generation (SHG) coefficient. [e] The non-resonant d33
values were calculated by using the approximate two-level model. [f] Order parameter
F=1ꢀA1/A0, where A1 and A0 are the absorbance of the polymer film after and
before corona poling, respectively. [g] Loading density of the effective chromophore
moieties; this value contains both the main chromophore and the isolation chromo-
phore.
F
ꢀ
a strong Ar Ar self-assembly effect. However, in
the low-generation dendrimers, pentafluorophenyl
moieties made the d33 values lower, instead of
higher as in the high-generation dendrimers. This
result was understandable because it was well-
known that, in theory, under identical experimental
conditions, the d33 value is proportional to the den-
sity of chromophore moieties. The use of penta-
fluorophenyl groups instead of normal phenyl moi-
eties not only brought the self-assembly effect, but
procedure was similar to one that we had reported previous-
ly[8,9,12] and their d33 values were calculated from the experi-
mental data at the 1064 nm fundamental wavelength
(Table 2).
also made the loading concentration of the chromophore
moieties lower (Figure 7A, Table 2), because the atomic
mass of the F atom was much larger than that of the
H atom. From their 19F NMR spectra and GPC results, we
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
637

Z. Li et al.
To further explore the alignment of the chromophore
moieties in these dendrimers, we measured their order pa-
rameter, F (Table 2 and the Supporting Information, Figur-
es S50–S58). According to the equation F=1ꢀA1/A0, the F
values of the dendrimers were calculated and the results are
listed in Table 2. The trend in F value was the same as that
in the d33 values of the dendrimers, which further supported
our conclusion.
De-poling experiments of the dendrimers were also con-
ducted, in which the real-time decay of their SHG signals
was monitored as the poled films were heated from 35 to
1508C in air at a rate of 48Cminꢀ1. Figure 8 displays the
Figure 7. A) Effective chromophore densities, and B) d33 values of differ-
~
&
*
ent types of dendrimers; Gn ( ), Gn-PFPh-N ( ), Gn-PFPh-NS ( ).
F
ꢀ
observed that the Ar Ar self-assembly effect in low-gener-
ation dendrimers was weak and that the improvement in
their 3D topological structure from this effect could not
compensate for the negative effects of the decreased chro-
mophore-loading densities. Thus, the d33 values of Gn-
PFPh-N in low-generation dendrimers were a little lower
than those that only contained normal phenyl groups as the
end-capped groups. On increasing the generation number,
F
ꢀ
the Ar Ar self-assembly effect became stronger and the d33
values also increased (Figure 7B). As expected, the d33
value of G4-PFPh-N (187 pmVꢀ1) was higher than that of
G4 (177 pmVꢀ1) and the NLO coefficient of G5-PFPh-N
was even as higher (206 pmVꢀ1).
Figure 8. Decay curves of the SHG coefficients of the dendrimers as a
function of temperature.
Based on dendrimers G1-PFPh-N to G5-PFPh-N, the
structures of compounds G2-PFPh-NS to G5-PFPh-NS
were further improved by the introduction of isolation chro-
mophore moieties and this improvement could be reflected
in their NLO coefficients: Almost all of the d33 values of
Gn-PFPh-NS were higher than those of Gn-PFPh-N in the
same generation (Figure 7B, Table 2). As an example, the
d33 value of G5-PFPh-NS was as high as 257 pmVꢀ1, which
was more than 50 pmVꢀ1 higher than that of G5-PFPh-N. It
is worth pointing out that this value is the new record high-
est value reported so far for simple azo chromophore moiet-
ies to the best of our knowledge.
decay of the SHG coefficient of the dendrimers as a func-
tion of temperature. Owing to their high Tg values, which
F
ꢀ
were derived from the Ar Ar effect, the temporal stability
of the dipole orientation of these dendrimers was also im-
proved. For G5-PFPh-N, the temperature for the decay was
up to 1118C, which is promising for practical NLO applica-
tions.
Conclusion
Because the films of these dendrimers still showed some
absorption at 532 nm (the doubled frequency of the 1064 nm
fundamental wavelength), the NLO properties of the den-
drimers should be smaller (d33(1); Table 2), owing to the res-
onant-enhancement effect.[20] As mentioned above, the max-
imum absorption of these new dendrimers was blue-shifted
compared to that of dendrimers G1–G5; thus, the increasing
trend in the d33(1) values of these dendrimers should be
more obvious. In fact, the d33(1) value of G5-PFPh-NS was
65 pmVꢀ1, which was almost twice that of dendrimer G5
(34 pmVꢀ1), thus making these dendrimers suitable for po-
tential applications in the field of optics. This phenomenon
also means that the introduction of pentafluorophenyl moi-
eties or isolation chromophores into dendrimers (or poly-
mers) is a new way of solving the contradictions between
nonlinearity and transparency in some way.
In summary, by using a combination of divergent and con-
vergent approaches, coupled with the powerful click-chemis-
try reaction, two new series of dendrimers were successfully
F
ꢀ
obtained in which two new effects, that is, Ar Ar self-as-
sembly and isolation chromophores, were used to further
improve their comprehensive performance. These dendri-
AHCTUNGTRENNUNG
mers exhibited very large d33 values (up to 257 pmVꢀ1,
which is the highest value reported so far for simple azo
chromophore moieties), excellent optical transparency (the
maximum absorption was only 443 nm in a film), and good
stability. Thus, these successful examples might open up new
avenues to synthesize more dendrimers with good NLO
properties; moreover, we believe that, by modifying the pol-
ymeric structure more intelligently, better performance
could be achieved by utilizing these very simple azo chro-
mophore moieties.
638
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

Improving the Performance of NLO Dendrimers
FULL PAPER
68.54, 69.20, 83.08, 107.47, 109.17, 111.77, 116.30, 117.33, 122.27, 125.99,
135.91, 139.35, 141.66, 143.73, 145.03, 146.59, 147.10, 148.31, 149.94,
Experimental Section
ꢁ
155.11, 158.80 ppm; IR (KBr): n˜ =3290 (C CH), 1740 (C=O), 1516,
Materials and instrumentation: THF was dried over and distilled from a
K-Na alloy under an atmosphere of dry nitrogen. DMF was dried over
1338 cmꢀ1 (NO2); MS (MALDI-TOF): m/z calcd for C91H66F20N18O17:
2085.5 [M+Na]+; found: 2085.9.
ꢁ
and distilled from CaH2. Compound 1, diazonium salts 2 and 3, G0-
Synthesis of dendrimer G2-PFPh-N: The procedure was similar to the
synthesis of dendrimer G1-PFPh-N with G1- -PFPh-N (454.0 mg,
-PFPh-N, and dendronized cores G1-4N3-N, G2-8N3-N, G1-4N3-S, and
G2-8N3-NS were prepared according to our previously reported proce-
dures.[9,12a,b,14] N,N,N,N,N-pentamethyldiethylenetriamine (PMDETA) and
pentafluorobenzoic acid were purchased from Alfa Aesar. All other re-
agents were used as received.
ꢁ
0.22 mmol) and N,N-bis(2-azidoethyl)aniline (1; 23.1 mg, 0.10 mmol). The
crude product was purified by column chromatography on silica gel
(THF/CHCl3, 1:1 v/v) to afford G2-PFPh-N as a red solid (416 mg,
95.5%). 1H NMR (300 MHz, CDCl3, 298 K): d=2.21 (br s; CH2), 2.93
(br s; CH2C), 3.64 (br s; NCH2), 3.74 (br s; NCH2), 3.91 (br s; NCH2),
4.11 (br s; NCH2), 4.40 (br s, 4H; OCH2), 4.59 (br s; COOCH2), 6.54
(br s; ArH), 6.70 (br s; ArH), 6.85 (br s; ArH), 7.14 (br s; ArH), 7.28 (m;
ArH and C=CH), 7.45–7.90 ppm (m; ArH); 13C NMR (75 MHz, CDCl3,
298 K): d=21.73, 28.37, 47.24, 49.62, 51.17, 63.33, 68.59, 109.23, 111.81,
116.33, 117.36, 122.35, 126.02, 129.63, 136.03, 139.52, 141.66, 143.85,
145.06, 146.63, 147.11, 148.34, 149.98, 155.16, 158.82 ppm; IR (KBr): n˜ =
1738 (C=O), 1519, 1337 cmꢀ1 (NO2); MS (MALDI-TOF): m/z calcd for
1H, 13C, and 19F NMR spectra were measured on Varian Mercury300,
Varian Mercury600, or Bruker ARX400 spectrometers by using tetra-
ACHTUNGTRENNUNGmethACHTUNGTRENNUNGylsilane (TMS; d=0 ppm) as an internal standard. FTIR spectra
were recorded on a PerkinElmer-2 spectrometer in the region 3000–
400 cmꢀ1. UV/Vis spectra were obtained on a Shimadzu UV-2550 spec-
trometer. MS (MALDI-TOF) was performed on a Voyager-DE-STR
MALDI-TOF mass spectrometer (ABI, American) that was equipped
with a 337 nm nitrogen laser and a 1.2 m linear flight path in positive-ion
mode. Elemental analysis was performed on
a CARLOERBA-1106
C
192H145F40N43O34: 4379.0 [M+Na]+; found: 4379.9; elemental analysis
calcd (%) for C192H145F40N43O34: C 52.91, H 3.35, N 13.82; found: C 53.63,
H 3.05, N 13.44.
micro-elemental analyzer. Gel-permeation chromatography (GPC) was
used to determine the molecular weights of the polymers. GPC analysis
was performed on a Waters HPLC system that was equipped with a
2690D separation module and a 2410 refractive-index detector. Polysty-
ꢁ
Synthesis of dendrimer G2- -PFPh-N: The procedure was similar to the
synthesis of dendrimer G1- -PFPh-N with diazonium salt 2 (50.0 mg,
0.156 mmol) and dendrimer G2-PFPh-N (340.0 mg, 0.078 mmol). The re-
action time was 5 days. The crude product was purified by column chro-
matography on silica gel (THF/CHCl3, 1:1 v/v) to afford G2- -PFPh-N
as a red solid (205.3 mg, 57.3%). H NMR (400 MHz, CDCl3, 298 K): d=
2.01 (s; C CH), 2.20 (br s; CH2), 2.92 (br s; CH2C), 3.76 (br s; NCH2),
3.91 (br s; NCH2), 4.10 (br s; NCH2), 4.28 (br s, 4H; OCH2), 4.40 (br s,
4H; OCH2), 4.59 (br s; COOCH2), 6.51 (br s; ArH), 6.60 (br s; ArH),
6.84 (br s; ArH), 7.30 (br s; ArH), 7.42–7.58 (m; ArH and C=CH), 7.67–
7.84 ppm (m; ArH); 13C NMR (150m, CDCl3, 298 K): d=21.95, 28.19,
28.60, 29.93, 47.53, 49.89, 51.45, 63.60, 68.71, 107.58, 109.39, 112.03,
116.62, 117.60, 122.70, 126.32, 137.11, 138.80, 142.81, 144.58, 144.93,
ꢁ
ACHTUNGTRENNUNGrene standards were used as calibration standards, THF was used as an
eluent, and the flow rate was 1.0 mLminꢀ1. Thermogravimetric analysis
(TGA) was performed on a NETZSCH STA449C thermal analyzer at a
heating rate of 108Cminꢀ1 under a flow of nitrogen gas (flow rate:
50 cm3 minꢀ1). The thermal transitions of the polymers were investigated
by using a METTLER differential scanning calorimeter DSC822e under
a nitrogen atmosphere at a scanning rate of 108Cminꢀ1. The melting
points were uncorrected. The thickness of the films was measured on an
Ambios Technology XP-2 profilometer.
ꢁ
1
ꢁ
ꢁ
Synthesis of dendrimer G1-PFPh-N: Chromophore G0- -PFPh-N
(1.60 g, 2.0 mmol), N,N-bis(2-azidoethyl)aniline (1; 208 mg, 0.9 mmol),
CuSO4·5H2O (10 mol%), NaHCO3 (20 mol%), and ascorbic acid
(20 mol%) were dissolved in THF (30 mL)/water (6 mL) under a nitro-
gen atmosphere in a Schlenk flask. After the mixture was stirred at 25–
308C under a nitrogen atmosphere for 8 h, the reaction was stopped by
the addition of water, extracted with CHCl3, and washed with brine. The
organic layer was dried over anhydrous magnesium sulfate and purified
by column chromatography on silica gel (EtOAc/CHCl3 (2:1, v/v) as
eluent to afford G1-PFPh-N as a red solid (1.42 g, 86.6%). 1H NMR
(300 MHz, CDCl3, 298 K): d=2.24 (br s; CH2), 2.94 (br s; CH2C), 3.63
(br s; NCH2), 3.92 (br s; NCH2), 4.16 (br s; NCH2), 4.31 (br s, 4H;
OCH2), 4.60 (br s; COOCH2), 6.52 (d, J=7.8 Hz; ArH), 6.73 (t; ArH),
6.86 (d, J=8.4 Hz; ArH), 7.14 (m; ArH and C=CH), 7.63 (m; ArH),
7.79–7.90 ppm (m; ArH); 13C NMR (75 MHz, CDCl3, 298 K): d=21.69,
28.44, 47.28, 49.57, 51.48, 63.31, 68.54, 109.16, 111.80, 112.73, 116.28,
117.33, 118.34, 122.08, 125.98, 129.62, 135.95, 139.29, 143.71, 145.10,
145.80, 146.66, 146.84, 148.36, 149.94, 155.16, 158.77 ppm; IR (KBr): n˜ =
1740 (C=O), 1516, 1338 cmꢀ1 (NO2); MS (MALDI-TOF): m/z calcd for
145.26, 146.62, 146.82, 147.36, 148.53, 149.58, 150.22, 155.37, 159.13 ppm;
ꢀ1
ꢁ
IR (KBr): n˜ =3310 (C CH), 1739 (C=O), 1519, 1338 cm (NO2); MS
(MALDI-TOF): m/z calcd for C203H154F40N46O37: 4612 [M+Na]+; found:
4613.
General procedure of the synthesis of G3-PFPh-N, G4-PFPh-N, and G5-
PFPh-N: A mixture of G2-8N3-N (1.00 equiv), different end-capped den-
drons (9.00 equiv), and CuBr (8.00 equiv), was dissolved in DMF (0.02m
N3) under a nitrogen atmosphere in a Schlenk flask and N,N,N,N,N-pen-
tamethyldiethylenetriamine (PMDETA; 8.00 equiv) was added. After the
mixture had been stirred at 25–308C for 6 h, the reaction was stopped by
the addition of water. The precipitate was washed repeatedly with water
and further purified by repeated precipitation of their solutions in THF
or CHCl3 with EtOAc, filtered, washed with a large volume of EtOAc,
and dried under vacuum at 408C to a constant weight.
ꢁ
G3-PFPh-N: G2-8N3-N (30.1 mg, 0.010 mmol) and G0- -PFPh-N
(72.1 mg, 0.090 mmol). G3-PFPh-N was obtained as
a
red powder
C80H57F20N15O14
:
1854.4 [M+Na]+; found: 1853.8; elemental analysis
(80.0 mg, 85.0%). 1H NMR (300 MHz, CDCl3, 298 K): d=1.8–2.4 (CH2),
2.8–3.1 (CH2C and CH2), 3.6–4.2 (NCH2), 4.3–4.7 (OCH2 and COOCH2),
6.4–6.6 (ArH), 6.7–6.9 (ArH), 7.2–7.9 ppm (ArH and C=CH); 13C NMR
(150 MHz, CDCl3, 298 K): d=22.05, 28.66, 29.88, 47.56, 49.90, 51.34,
63.62, 69.03, 107.65, 109.64, 112.13, 116.58, 117.61, 122.74, 126.24, 137.10,
138.79, 142.71, 144.90, 145.34, 146.62, 148.55, 149.69, 150.33, 155.41,
159.04 ppm; IR (KBr): n˜ =1737 (C=O), 1518, 1338 cmꢀ1 (NO2); elemen-
tal analysis calcd (%) for C416H321F80N99O74: C 53.09, H 3.44, N 14.74;
found: C 51.81, H 3.23, N 14.57.
calcd (%) for C80H57F20N15O14: C 52.44, H 3.14, N 11.47; found: C 52.97,
H 3.22, N 11.30.
ꢁ
Synthesis of dendrimer G1- -PFPh-N: Diazonium salt
2 (181.8 mg,
0.57 mmol) and dendrimer G1-PFPh-N (696.3 mg, 0.38 mmol) were dis-
solved in DMF (5 mL) at 08C. The reaction mixture was stirred for 40 h
at 08C, treated with water, extracted with CHCl3, and washed with brine.
The organic layer was dried over anhydrous sodium sulfate. After remov-
al of the organic solvent, the crude product was purified by column chro-
ꢁ
ꢁ
matography on silica gel (EtOAc/CHCl3, 2:1 v/v) to afford G1- -PFPh-N
as a red solid (624.3 mg, 79.6%). H NMR (300 MHz, CDCl3, 298 K): d=
G4-PFPh-N: G2-8N3-N (15.0 mg, 0.005 mmol) and G1- -PFPh-N
1
(92.9 mg, 0.045 mmol). G4-PFPh-N was obtained as
a
red powder
2.02 (s; C CH), 2.16 (br s; CH2), 2.24 (br s; CH2), 2.48 (br s; CH2),2.95
(70.4 mg, 72.1%). 1H NMR (300 MHz, CDCl3, 298 K): d=1.4–2.0 (CH2),
2.1–2.3 (CH2), 2.8–3.1 (CH2C and CH2), 3.6–4.2 (NCH2), 4.3–4.7 (OCH2
and COOCH2), 6.4–6.6 (ArH), 6.7–6.9 (ArH), 7.2–7.9 ppm (ArH and C=
CH); 13C NMR (150 MHz, CDCl3, 298 K): d=22.14, 28.65, 43.10, 47.62,
49.89, 63.63, 68.85, 107.58, 109.46, 112.04, 116.54, 117.53, 122.89, 126.29,
137.08, 138.73, 142.80, 144.93, 145.23, 146.63, 148.44, 150.27 ppm; IR
ꢁ
(br s; CH2C), 3.76 (br s; NCH2), 3.92 (br s; NCH2), 4.15 (br s; NCH2),
4.30 (br s, 4H; OCH2), 4.42 (br s, 4H; OCH2), 4.60 (br s; COOCH2), 6.58
(d, J=6.9 Hz; ArH), 6.86 (d, J=7.5 Hz; ArH), 7.31 (br s; ArH), 7.59 (m;
ArH and C=CH), 7.74–7.87 ppm (m; ArH); 13C NMR (75 MHz, CDCl3,
298 K): d=14.99, 21.70, 27.92, 28.34, 47.18, 49.58, 51.20, 63.29, 68.00,
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
639

Z. Li et al.
(KBr): n˜ =1737 (C=O), 1517, 1339 cmꢀ1 (NO2); elemental analysis calcd
(%) for C864H673F160N211O154: C 53.18, H 3.48, N 15.14; found: C 52.99,
H 3.62, N 14.47.
suitable end-capped dendrons, and CuBr (1.00 equiv) was dissolved in
DMF (0.02m N3) under a nitrogen atmosphere in a Schlenk flask and
N,N,N,N,N-pentamethyldiethylenetriamine (PMDETA; 1.00 equiv) was
added. After the mixture had been stirred at 25–308C for 6 h, the reac-
tion was stopped by the addition of water. The precipitate was washed re-
peatedly with water and further purified by repeated precipitation of
their solutions in THF or CHCl3 with EtOAc, filtered, washed with a
large volume of EtOAc, and dried under vacuum at 408C to a constant
weight.
ꢁ
G5-PFPh-N: G2–8N3-N (6.0 mg, 0.002 mmol) and G2- -PFPh-N
(82.6 mg, 0.018 mmol). G5-PFPh-N was obtained as
a
red powder
(63.1 mg, 79.5%). 1H NMR (300 MHz, CDCl3, 298 K): d=1.6–2.5 (CH2),
2.6–3.1 (CH2C and CH2), 3.4–4.2 (NCH2), 4.3–4.8 (OCH2 and COOCH2),
6.2–6.9 (ArH), 7.0–7.9 ppm (ArH and C=CH); 13C NMR (150 MHz,
CDCl3, 298 K): d=22.05, 28.66, 49.87, 63.64, 68.82, 107.65, 109.38, 111.98,
116.59, 117.53, 126.32, 137.11, 138.78, 144.92, 146.66, 148.42, 155.37,
159.09 ppm; IR (KBr): n˜ =1737 (C=O), 1516, 1337 cmꢀ1 (NO2); elemen-
tal analysis calcd (%) for C1760H1377F320N435O314: C 53.22, H 3.49, N 15.34;
found: C 53.35, H 3.88, N 14.83.
ꢁ
G3-PFPh-NS: G1-4N3-N (10.2 mg, 0.0088 mmol) and G1- -PFPh-NS
(80.0 mg, 0.0387 mmol). G3-PFPh-NS was obtained as a red powder
(58.2 mg, 70.2%). 1H NMR (300 MHz, [D6]DMSO, 298 K): d=1.7–2.2
(CH2), 2.7–3.1 (CH2C), 3.6–3.8 (NCH2), 3.8–4.0 (NCH2), 4.1–4.2 (NCH2),
4.2–4.6 (CH2O and CH2COO), 6.4–6.8 (ArH), 6.9–7.2 (ArH), 7.3–
8.2 ppm (ArH); IR (KBr): n˜ =1738 (C=O), 1516, 1332 (NO2), 1142 cmꢀ1
(SO2); elemental analysis calcd (%) for C416H325F80N95O70S4: C 53.03,
H 3.48, N 14.12; found: C 52.71, H 3.68, N 13.66.
ꢁ
Synthesis of dendrimer G1- -PFPh-NS: The procedure was similar to
ꢁ
the synthesis of dendrimer G1- -PFPh-N with diazonium salt
3
(173.9 mg, 0.54 mmol) and dendrimer G1-PFPh-N (659.7 mg, 0.36 mmol).
The reaction time was 2 days. The crude product was purified by column
ꢁ
G4-PFPh-NS: G1-4N3-S (4.1 mg, 0.0035 mmol) and G2- -PFPh-NS
ꢁ
chromatography on silica gel (EtOAc) to afford G1- -PFPh-NS as a red
solid (654.1 mg, 86.6%). H NMR (300 MHz, CDCl3, 298 K): d=1.98 (m;
1
(70.8 mg, 0.0154 mmol). G4-PFPh-NS was obtained as a red powder
(55.8 mg, 81.6%). 1H NMR (300 MHz, [D6]DMSO, 298 K): d=1.7–2.2
(CH2), 2.8–3.0 (CH2C), 3.8–4.0 (NCH2), 4.1–4.2 (NCH2), 4.2–4.6 (CH2O
and CH2COO), 6.5–6.8 (ArH), 6.9–7.0 (ArH), 7.4–8.0 (ArH). IR (KBr):
n˜ =1735 (C=O), 1516, 1337 (NO2), 1141 cmꢀ1 (SO2); elemental analysis
calcd (%) for C864H673F160N201O144S10: C 53.09, H 3.52, N 14.40; found:
C 53.63, H 3.86, N 13.94.
ꢁ
CH2C), 2.05 (s; C CH), 2.24 (br s; CH2), 2.33 (br s; CH2), 2.96 (br s;
CH2C), 3.26 (br s; SCH2) 3.75 (br s; NCH2), 3.92 (br s; NCH2), 4.14 (br s;
NCH2), 4.39 (br s; OCH2), 4.60 (br s; COOCH2), 6.56 (d; ArH), 6.86 (d,
J=7.2 Hz; ArH), 7.23 (s; C=CH), 7.61–7.96 ppm (m; ArH); 13C NMR
(75 MHz, CDCl3, 298 K): d=13.92, 17.00, 21.54, 28.27, 47.03, 49.44, 50.96,
54.80, 60.15, 63.24, 68.42, 70.03, 81.62, 107.22, 109.00, 111.67, 116.16,
117.15, 122.20, 122.73, 125.88, 128.81, 135.88, 139.24, 141.57, 143.64,
ꢁ
G5-PFPh-NS: G2-8N3-NS (4.5 mg, 0.0015 mmol) and G2- -PFPh-NS
144.86, 146.46, 146.99, 148.13, 149.33, 149.94, 155.00, 155.60, 158.68 ppm;
(60.7 mg, 0.0132 mmol). G5-PFPh-NS was obtained as a red powder
(43.2 mg, 72.4%). 1H NMR (300 MHz, [D6]DMSO, 298 K): d=1.4–2.0
(CH2), 2.6–2.8 (CH2C), 3.6–4.0 (NCH2), 4.1–4.6 (NCH2, CH2O and
CH2COO), 6.3–6.9 (ArH), 7.1–8.0 ppm (ArH); IR (KBr): n˜ =1738 (C=
O), 1517, 1332 (NO2), 1141 cmꢀ1 (SO2); elemental analysis calcd (%) for
C1760H1397F320N415O294S20: C 53.13, H 3.54, N 14.61; found: C 53.57, H 3.87,
N 13.99.
ꢀ1
ꢁ
IR (KBr): n˜ =3289 (C CH), 1716 (C=O), 1514, 1338 (NO2), 1140 cm
(SO2); MS (MALDI-TOF): m/z calcd for C91H67F20N17O16S: 2088.4
[M+Na]+; found: 2088.8.
Synthesis of dendrimer G2-PFPh-N: The procedure was similar to the
ꢁ
synthesis of dendrimer G1-PFPh-N with dendrimer G1- -PFPh-NS
(454.7 mg, 0.22 mmol) and N,N-bis(2-azidoethyl)aniline (1; 23.1 mg,
0.10 mmol). The crude product was purified by column chromatography
on silica gel (THF/CHCl3, 1:1 v/v) to afford G2-PFPh-NS as a red solid
(425.8 mg, 97.5%). 1H NMR (300 MHz, CDCl3, 298 K): d=1.83 (br s;
CH2), 2.05 (br s; CH2), 2.23 (br s; CH2C), 2.78 (br s; CH2C), 2.94 (br s;
CH2C), 3.13 (br s; SCH2), 3.62 (br s; NCH2), 3.73 (br s; NCH2), 3.92
(br s; NCH2), 4.14 (br s; NCH2), 4.39 (br s; OCH2), 4.59 (br s; COOCH2),
5.59 (s; NCH2), 6.59 (br s; ArH), 6.84 (br s; ArH), 7.16 (s; C=CH), 7.62
(m; ArH), 7.77–8.00 ppm (m; ArH); 13C NMR (75 MHz, CDCl3, 298 K):
d=21.63, 22.00, 22.43, 23.64, 27.64, 28.33, 47.11, 49.53, 51.02, 55.10, 63.31,
68.45, 109.16, 111.78, 112.81, 116.29, 117.29, 122.28, 122.80, 125.97, 128.88,
129.69, 135.94, 139.37, 143.75, 144.40, 144.98, 145.84, 146.61, 147.07,
148.24, 149.34, 150.00, 155.07, 158.78 ppm; IR (KBr): n˜ =1738 (C=O),
1504, 1332 (-NO2), 1142 cmꢀ1 (SO2); MS (MALDI-TOF): m/z calcd for
C192H147F40N41O32S2: 4387 [M+Na]+; found: 4388; elemental analysis
calcd (%) for C192H147F40N41O32S2: C 52.48, H 3.39, N 13.16; found:
C 52.90, H 3.56, N 12.76.
Preparation of the polymer thin films: The dendrimers were dissolved in
THF (concentration: 4 wt.%), the solutions were filtered through syringe
filters, and the films were spin-coated onto indium-tin-oxide (ITO)-
coated glass substrates, which were cleaned sequentially with DMF, ace-
tone, distilled water, and THF in an ultrasound bath before use. Residual
solvent was removed by heating the films in a vacuum oven at 408C.
NLO measurements of the poled films: The second-order optical non
ACHTUNGTRENNUNGlin-
AHCTUNGTRENNUNG
eration (SHG) experiments in a closed temperature-controlled oven with
optical windows and three needle electrodes. The films were kept at an
angle of 458 to the incident beam and poled inside the oven; the intensity
of SHG was monitored simultaneously. The poling conditions were as fol-
lows: T: different for each polymer (Table 2); V=7.5 kV at the needle
point; gap distance: 0.8 cm. The SHG measurements were carried out by
using a Nd:YAG laser that was operating at a repetition rate of 10 Hz
and a pulse width of 8 ns at 1064 nm. A Y-cut quartz crystal served as the
reference compound.
ꢁ
Synthesis of dendrimer G2- -PFPh-NS: The procedure was similar to
ꢁ
the synthesis of dendrimer G1- -PFPh-N with diazonium salt 2 (51.2 mg,
0.16 mmol) and dendrimer G2-PFPh-NS (349.2 mg, 0.08 mmol). The re-
action time was 6 days. The crude product was purified by column chro-
ꢁ
matography on silica gel (THF/CHCl3, 2:1 v/v) to afford G1- -PFPh-NS
as a red solid (229.1 mg, 62.3%). H NMR (400 MHz, CDCl3, 298 K): d=
Acknowledgements
1
ꢁ
2.01 (s, C CH), 2.10 (m; CH2 and CH2C), 2.22 (m; CH2), 2.48 (m;
We are grateful to the National Science Foundation of China (21034006)
for financial support.
CH2C), 2.82 (m; CH2C), 2.94 (m; CH2C), 3.13 (t, J=4.0 Hz; SCH2), 3.73
(br s; NCH2), 3.91 (br s; NCH2), 4.12 (br s; NCH2), 4.30 (br s; CH2), 4.44
(br s; OCH2), 4.59 (t, J=8.0 Hz; COOCH2), 6.54 (d, J=8.0 Hz; ArH),
6.69 (d, J=8.0 Hz; ArH), 6.84 (d, J=8.0 Hz; ArH), 7.61 (m; ArH and
C=CH), 7.70–7.90 ppm (m; ArH); 13C NMR (150 MHz, CDCl3, 298 K):
d=21.99, 23.98, 28.63, 47.48, 49.89, 51.43, 63.61, 68.75, 107.58, 109.42,
112.10, 116.68, 117.66, 122.59, 126.37, 138.83, 144.93, 146.69, 147.41,
[1] a) D. M. Burland, R. D. Miller, C. A. Walsh, Chem. Rev. 1994, 94,
31–75; b) Y. Bai, N. Song, J. P. Gao, X. Sun, X. Wang, G. Yu, Z. Y.
Wang, J. Am. Chem. Soc. 2005, 127, 2060–2621; c) T. J. Marks,
M. A. Ratner, Angew. Chem. 1995, 107, 167–187; Angew. Chem. Int.
Ed. Engl. 1995, 34, 155–173; d) D. Yu, A. Gharavi, L. P. Yu, J. Am.
Chem. Soc. 1995, 117, 11680–11686; e) S. R. Marder, B. Kippelen,
A. K. Y. Jen, N. Peyghambarian, Nature 1997, 388, 845–851; f) M.
Lee, H. E. Katz, C. Erben, D. M. Gill, P. Gopalan, J. D. Heber, D. J.
McGee, Science 2002, 298, 1401–1403; g) Y. Shi, C. Zhang, H.
ꢁ
148.60, 150.22, 155.39, 159.13, 160.87 ppm; IR (KBr): n˜ =3289 (C CH),
1738 (C=O), 1514, 1335 (NO2), 1142 cmꢀ1 (SO2); MS (MALDI-TOF): m/
z calcd for C203H156F40N44O35S2: 4618 [M+Na]+; found: 4618.
General procedure of the synthesis of G3-PFPh-NS, G4-PFPh-NS, and
G5-PFPh-NS: A mixture of the corresponding azide-containning core,
640
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 630 – 641

Improving the Performance of NLO Dendrimers
FULL PAPER
Zhang, J. H. Bechtel, L. R. Dalton, B. H. Robinson, W. H. Steier,
L. R. Dalton, Science 2000, 288, 119–122; h) C. V. Mclaughlin, M.
Hayden, B. Polishak, S. Huang, J. Luo, T.-D. Kim, A. K. Y. Jen,
Appl. Phys. Lett. 2008, 92, 151107; i) L. R. Dalton, P. A. Sullivan,
D. H. Bale, Chem. Rev. 2010, 110, 25–55; j) J. D. Luo, X. H. Zhou,
A. K. Y. Jen, J. Mater. Chem. 2009, 19, 7410–7424; k) J. D. Luo, S.
Huang, Z. W. Shi, B. M. Polishak, X. H. Zhou, A. K. Y. Jen, Chem.
Mater. 2011, 23, 544–553.
Polymer 2006, 47, 7881; i) Z. Li, W. Wu, G. Qiu, G. Yu, Y. Liu, C.
Ye, J. Qin, Z. Li, J. Polym. Sci. Part A 2011, 49, 1977–1987; j) Z. Li,
G. Yu, S. Dong, W. Wu, Y. Liu, C. Ye, J. Qin, Z. Li, Polymer 2009,
50, 2806–2814; k) Z. Li, L. Wang, B. Xiong, C. Ye, J. Qin, Z. Li,
Dyes Pigm. 2010, 84, 134–139; l) Z. Li, W. Wu, C. Ye, J. Qin, Z. Li,
J. Phys. Chem. B 2009, 113, 14943–14949.
[9] a) Z. Li, G. Yu, W. Wu, Y. Liu, C. Ye, J. Qin, Z. Li, Macromolecules
2009, 42, 3864–3868; b) Z. Li, W. Wu, Q. Li, G. Yu, L. Xiao, Y. Liu,
C. Ye, J. Qin, Z. Li, Angew. Chem. 2010, 122, 2823–2827; Angew.
Chem. Int. Ed. 2010, 49, 2763–2767.
[10] P. A. Sullivan, H. Rommel, Y. Liao, B. C. Olbricht, A. J. P. Akelaitis,
K. A. Firestone, J.-W. Kang, J. Luo, J. A. Davies, D. H. Choi, B. E.
Eichinger, P. J. Reid, A. Chen, A. K. Y. Jen, B. H. Robinson, L. R.
Dalton, J. Am. Chem. Soc. 2007, 129, 7523–7530.
[11] a) T. D. Kim, J. Kang, J. Luo, S. Jang, J. Na, N. Tucker, J. B. Bene-
dict, L. R. Dalton, T. Gray, R. M. Overney, D. H. Park, W. N.
Herman, A. K.-Y. Jen, J. Am. Chem. Soc. 2007, 129, 488–489; b) X.
Zhou, J. Luo, S. Huang, T. Kim, Z. Shi, Y. Cheng, S. Jang, D. B.
Knorr, R. M. Overney, A. K.-Y. Jen, Adv. Mater. 2009, 21, 1976–
1981; c) W. Wu, Q. Huang, G. Qiu, C. Ye, J. Qin, Z. Li, J. Mater.
Chem. 2012, 22, 18486–18495.
[2] a) S.-H. Jang, J. Luo, N. M. Tucker, A. Leclercq, E. Zojer, M. A.
Haller, T.-D. Kim, J.-W. Kang, K. Firestone, D. Bale, D. Lao, J. B.
Benedict, D. Cohen, W. Kaminsky, B. Kahr, J.-L. Bredas, P. Reid,
L. R. Dalton, A. K.-Y. Jen, Chem. Mater. 2006, 18, 2982–2988;
b) S. R. Marder, L. T. Cheng, B. G. Tiemann, A. C. Friedli, M. Blan-
chard-Desce, J. W. Perry, J. Skindhøj, Science 1994, 263, 511–514;
c) Q. Li, Z. Li, F. Zeng, W. Gong, Z. Li, Z. Zhu, Q. Zeng, S. Yu, C.
Ye, J. Qin, J. Phys. Chem. B 2007, 111, 508–514.
[3] a) B. H. Robinson, L. R. Dalton, J. Phys. Chem. A 2000, 104, 4785–
4795; b) B. H. Robinson, L. R. Dalton, H. W. Harper, A. Ren, F.
Wang, C. Zhang, G. Todorova, M. Lee, R. Aniszfeld, S. Garner, A.
Chen, W. H. Steier, S. Houbrecht, A. Persoons, I. Ledoux, J. Zyss,
A. K. Y. Jen, Chem. Phys. 1999, 245, 35–50.
[4] L. R. Dalton, W. H. Steier, B. H. Robinson, C. Zhang, A. Ren, S.
Garner, A. Chen, T. Londergan, L. Irwin, B. Carlson, L. Fifield, G.
Phelan, C. Kincaid, J. Amend, A. K. Y. Jen, J. Mater. Chem. 1999, 9,
1905–1920.
[12] a) W. Wu, C. Ye, G. Yu, Y. Liu, J. Qin, Z. Li, Chem. Eur. J. 2012, 18,
4426–4434; b) W. Wu, C. Li, G. Yu, Y. Liu, C. Ye, J. Qin, Z. Li,
Chem. Eur. J. 2012, 18, 11019–11028; c) W. Wu, L. Huang, L. Xiao,
Q. Huang, R. Tang, C. Ye, J. Qin, Z. Li, RSC Adv. 2012, 2, 6520–
6526; d) Z. Li, J. Qin, A. Qin, C. Ye, Polymer 2005, 46, 363–368;
e) Z. Li, W. Gong, J. Qin, Z. Yang, C. Ye, Polymer 2005, 46, 4971–
4978; f) Z. Li, J. Hua, Q. Li, C. Huang, A. Qin, C. Ye, J. Qin, Poly-
mer 2005, 46, 11940–11948; g) Z. Li, C. Huang, J. Hua, J. Qin, Z.
Yang, C. Ye, Macromolecules 2004, 37, 371–376; h) Z. Li, J. Qin, S.
Li, C. Ye, J. Luo, Y. Cao, Macromolecules 2002, 35, 9232–9235; i) Z.
Li, Q. Zeng, Z. Li, S. Dong, Z. Zhu, Q. Li, C. Ye, C. Di, Y. Liu, J.
Qin, Macromolecules 2006, 39, 8544–8546.
[13] a) Y. H. Kim, J. Polym. Sci. Part A 1998, 36, 1685–1698; b) A.
Sunder, J. Heinemann, H. Frey, Chem. Eur. J. 2000, 6, 2499–2506;
c) B. Voit, J. Polym. Sci. Part A 2000, 38, 2505–2525; d) M. Scholl,
Z. Kadlecova, H.-A. Klok, Prog. Polym. Sci. 2009, 34, 24–61;
e) C. J. Hawker, R. Lee, J. M. J. Frꢁchet, J. Am. Chem. Soc. 1991,
113, 4583–4588.
[5] a) H. Yamamoto, S. Katogi, T. Watanabe, H. Sato, S. Miyata, Appl.
Phys. Lett. 1992, 60, 935–939; b) H. Kang, P.-W. Zhu, Y. Yang, A.
Facchetti, T. J. Marks, J. Am. Chem. Soc. 2004, 126, 15974–15975;
c) C. R. Moylan, S. Ermer, S. L. Lovejoy, I. H. McComb, D. S.
Leung, R. Wortmann, P. Krdmer, R. J. Twieg, J. Am. Chem. Soc.
1996, 118, 12950–12955; d) E. A. Weiss, L. E. Sinks, A. S. Lukas,
E. T. Chernick, M. A. Ratner, M. R. Wasielewski, J. Phys. Chem. B
2004, 108, 10309–10316; e) B. R. Cho, S. J. Lee, S. H. Lee, K. H.
Son, Y. H. Kim, J.-Y. Doo, G. J. Lee, T. I. Kang, Y. K. Lee, M. Cho,
S.-J. Jeon, Chem. Mater. 2001, 13, 1438–1440; f) P. Gopalan, H. E.
Katz, D. J. McGee, C. Erben, T. Zielinski, D. Bousquet, D. Muller, J.
Grazul, Y. Olsson, J. Am. Chem. Soc. 2004, 126, 1741–1747; g) Z.
Li, P. Hu, G. Yu, W. Zhang, Z. Jiang, Y. Liu, C. Ye, J. Qin, Z. Li,
Phys. Chem. Chem. Phys. 2009, 11, 1220–1226; h) W. Wu, L. Huang,
C. Song, G. Yu, C. Ye, Y. Liu, J. Qin, Q. Li, Z. Li, Chem. Sci. 2012,
3, 1256–1261; i) Q. Li, G. Yu, J. Huang, H. Liu, Z. Li, C. Ye, Y. Liu,
J. Qin, Macromol. Rapid Commun. 2008, 29, 798–803; j) Z. Li, G.
Yu, Z. Li, Y. Liu, C. Ye, J. Qin, Polymer 2008, 49, 901–913; k) Z. Li,
W. Wu, G. Yu, Y. Liu, C. Ye, J. Qin, Z. Li, ACS Appl. Mater. Interfa-
ces 2009, 1, 856–863.
[6] a) M. J. Cho, D. H. Choia, P. A. Sullivan, A. J.-P. Akelaitis, L. R.
Dalton, Prog. Polym. Sci. 2008, 33, 1013–1058; b) Y. V. Pereverzev,
K. N. Gunnerson, O. V. Prezhdo, P. A. Sullivan, Y. Liao, B. C. Ol-
bricht, A. J. P. Akelaitis, A. K.-Y. Jen, L. R. Dalton, J. Phys. Chem.
C 2008, 112, 4355–4363.
[7] a) H. Ma, S. Liu, J. Luo, S. Suresh, L. Liu, S. H. Kang, M. Haller, T.
Sassa, L. R. Dalton, A. K. Y. Jen, Adv. Funct. Mater. 2002, 12, 565–
574; b) W. Shi, J. Luo, S. Huang, X.-H. Zhou, T.-D. Kim, Y.-J.
Cheng, B. M. Polishak, T. R. Younkin, B. A. Block, A. K.-Y. Jen,
Chem. Mater. 2008, 20, 6372–6377.
[14] W. Wu, Y. Fu, C. Wang, C. Ye, J. Qin, Z. Li, Chem. Asian J. 2011, 6,
2787–2795.
[15] a) J. M. J. Frꢁchet, Science 1994, 263, 1710–1715; b) S. M. Grayson,
J. M. J. Frꢁchet, Chem. Rev. 2001, 101, 3819–3867.
[16] a) F. Zeng, S. C. Zimmerman, Chem. Rev. 1997, 97, 1681–1712;
b) A. W. Bosman, H. M. Janssen, E. W. Meijer, Chem. Rev. 1999, 99,
1665–1688; c) S. C. Lo, P. L. Burn, Chem. Rev. 2007, 107, 1097–
1116; d) L. Crespo, G. Sanclimens, M. Pons, E. Giralt, M. Royo, F.
Albericio, Chem. Rev. 2005, 105, 1663–1681.
[17] a) K. L. Wooley, C. J. Hawker, J. M. J. Frꢁchet, J. Am. Chem. Soc.
1991, 113, 4252–4261; b) J.-L. Wang, J. Yan, Z. M. Tang, Q. Xiao, Y.
Ma, J. Pei, J. Am. Chem. Soc. 2008, 130, 9952–9962; c) H.-T. Chang,
C.-T. Chen, T. Kondo, G. Siuzdak, K. B. Sharpless, Angew. Chem.
1996, 108, 202–206; Angew. Chem. Int. Ed. 1996, 35, 182–186.
[18] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew.
Chem. 2002, 114, 2708–2711; Angew. Chem. Int. Ed. 2002, 41, 2596–
2599.
[19] a) Z. Muchtar, M. Schappacher, A. Deffieux, Macromolecules 2001,
34, 7595–7600; b) M. W. Weimer, J. M. J. Frꢁchet, I. Gitsov, J.
Polym. Sci. Part A 1998, 36, 955–970; c) Y. Kim, O. W. Webster,
Macromolecules 1992, 25, 5561–5572.
[20] M. Kauranen, T. Verbiest, C. Boutton, M. N. Teeren, K. Clays, A. J.
Schouten, R. J. M. Nolte, A. Persoons, Science 1995, 270, 966–969.
[8] a) Z. Li, Z. Li, C. Di, Z. Zhu, Q. Li, Q. Zeng, K. Zhang, Y. Liu, C.
Ye, J. Qin, Macromolecules 2006, 39, 6951–6961; b) Q. Zeng, Z. Li,
Z. Li, C. Ye, J. Qin, B. Z. Tang, Macromolecules 2007, 40, 5634–
5637; c) Q. Li, Z. Li, C. Ye, J. Qin, J. Phys. Chem. B 2008, 112,
4928–4933; d) Z. Li, P, Li, S. Dong, Z. Zhu, Q. Li, Q. Zeng, Z. Li,
C. Ye, J. Qin, Polymer 2007, 48, 3650–3657; e) Z. Li, S. Dong, G.
Yu, Z. Li, Y. Liu, C. Ye, J. Qin, Polymer 2007, 48, 5520–5529; f) Z.
Li, S. Dong, P. Li, Z. Li, C. Ye, J. Qin, J. Polym. Sci. Part A 2008, 46,
2983–2993; g) Z. Li, Q. Li, J. Qin, Polym. Chem. 2011, 2, 2723–
2740; h) Z. Zhu, Z. Li, Y. Tan, Z. Li, Q. Li, Q. Zeng, C. Ye, J. Qin,
Received: August 22, 2012
Published online: November 23, 2012
Chem. Eur. J. 2013, 19, 630 – 641
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
641