Synthesis and properties of oxygen-linked N-phenylcarbazole dendrimers

Article abstract of DOI:10.1021/ma202433y

A series of novel oxygen-linked N-phenylcarbazole (NPC) dendritic wedges (3-5) and triphenylamine-centered dendrimers (CBD-G0 to -G2) have been synthesized and extensively studied. The aryl-oxygen-aryl (Ar-O-Ar) linkages were established on 3,6-positions

Full text of DOI:10.1021/ma202433y

Article
pubs.acs.org/Macromolecules
Synthesis and Properties of Oxygen-Linked N-Phenylcarbazole
Dendrimers
Chung-Chieh Lee,† Man-kit Leung,*,†,‡ Pei-Yu Lee,∥ Tien-Lung Chiu,∥ Jiun-Haw Lee,§ Chun Liu,†
and Pi-Tai Chou*,†
§
†Department of Chemistry, ‡Institute of Polymer Science and Engineering, and Graduate Institute of Photonics & Optoelectronics
and Department of Electrical Engineering, National Taiwan University, 1 Roosevelt Road, Section 4, Taipei 106, Taiwan, R.O.C.
∥Department of Photonics Engineering, Yuan Ze University, Taoyuan 32003, Taiwan
S
* Supporting Information
ABSTRACT: A series of novel oxygen-linked N-phenyl-
carbazole (NPC) dendritic wedges (3−5) and triphenyl-
amine-centered dendrimers (CBD-G0 to -G2) have been
synthesized and extensively studied. The aryl−oxygen−aryl
(Ar−O−Ar) linkages were established on 3,6-positions of
carbazole and 4-position of N-phenyl group. UV−vis spectra
revealed that the Ar−O−Ar linkage would break down the π
conjugation and make NPC units manifest their individual
absorption moiety. Both steady-state luminescence and
fluorescence decay dynamics at room temperature and 77 K
in tetrahydrofuran suggested that the transfer of energy from the carbazole wedges to the triphenylamine (TPA) core operates in
CBD-G1, so that the luminescence mainly arises from the core unit. The quenching of the emission from carbazole wedges by
the TPA core becomes less effective as the generation develops into CBD-G2. The oxidation potential of the NPC derivatives
clearly revealed the mesomeric electron-donating effect of the oxygen atom and inductive electron-withdrawing effect of the
carbazole unit. The potential gradient could be established on those triphenylamine-centered dendrimers (CBD-G1 and CBD-
G2), such that the outer layer is electron-poor and the inner layer is electron-rich. It is remarkable that the HOMO level of the
NPC dendrimers is as high as −5.17 eV and triplet energy level is kept above 2.85 eV. These levels of HOMO and triplet energy,
together with good thermal stability and compatibility of solution processing, make NPC dendrimers ideal host materials for blue
triplet emitters. Using CBD-G2 as the host material, the FIrpic-doped solution-processed light-emitting diodes were fabricated,
and maximum luminous efficiency could be achieved at a record high of 24.7 cd/A at 484 cd/m2.
host10 materials of OLEDs. The CB dendrimers could be
categorized into two classes from the structural point of view:
The first class contains the carbazole units tethered to the
molecules has attracted considerable attention during the past
INTRODUCTION
The advance of the architecture for constructing dendritic
■
saturated C−C chains.5 The carbazole units therefore function
independently so that their original physical properties such as
electrochemical oxidation potential and/or photophysical
properties would be maintained. The second class has the
carbazole units chained through π-conjugation to form
dendritic wedges.11 Usually, this type of conjugated dendrimer
demonstrates fascinating optoelectronic properties that are
different from the carbazole monomer unit. Recently,
explorations about formation of the potential or redox gradient
have been reported.12 For examples, Yamamoto and Albrecht
reported the synthesis of N-linked rigid dendritic carbazoles
G4Fc and discovered the accumulative π-polarization sub-
stituent effect of the carbazole unit producing a potential
gradient in the carbazole dendron such that the outer layer
becomes electron-rich while the inner layers are electron-
two decades.1 The availability nowadays of a large variety of
dendrons with well-defined branches allows ample materials of
intriguing supramolecular properties for light harvesting,
exciton harvesting, electronic potential gradient, and molecular
recognition to be prepared. Applications including drug
delivery,2 organic electronics,3 and homogeneous catalysis4
have been reported. The maturity of synthetic architecture also
allows chemists to explore the manipulation of the relationships
between molecular structures and energy levels, one of the
most important topics in current chemistry.
Carbazole-based (CB) dendrimers have emerged as a new
and attractive family of advanced materials. Because of the
interesting photo- and electrochemistries, as well as distin-
guished electronic properties of the carbazole unit, the CB
dendrons and dendrimers have been utilized in nano-
composites,5 efficient light-harvesting antennae,6 phosphores-
cent iridium(III) dendrimers,7 and fluorescent dendritic
emitters8 for organic light-emitting diode (OLED). Besides,
the CB dendrimers hold promise as hole-transporting9 and
Received: November 3, 2011
Revised: December 3, 2011
Published: December 22, 2011
© 2011 American Chemical Society
751
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
poor.12a Blackstock and Selby reported redox-active polyaryl-
amine dendrimer TPA-G2, which exhibits contrary oxidation
gradients.12b The core, phenylenediamino (PD) group, is more
easily oxidized than the periphery, triarylamino (AA) group.
RESULTS AND DISCUSSION
■
Preparation of the Carbazole−Oxy−Carbazole Den-
dritic Wedge. In our strategy, we adopted the coupling
reaction of 1 with 2 to construct the dendron. To optimize the
reaction conditions, coupling 1 with phenol was first attempted
under different CuI promoted C−O coupling conditions to
give 3.19 Table 1 summarizes the results in which only under
the condition of CuI/N,N-dimethylglycine hydrochloride19a
and (DMGlyHCl)/1,4-dioxane did the reaction give a
satisfactory yield of 3 after prolonged heating for 72 h (entry
5). Other ligands such as 2-(hydroxymethyl)-2-methylpropane-
1,3-diol (TrisOH),19b ethyl 2-oxacyclohexanecarboxylate (CY-
COEt),19c and trans-N,N′-dimethylcyclohexyldiamine
(DMCHDA)19d were found to be less effective. In order to
increase the reaction efficiency, microwave (MW) irradiation
(200 W, open vessel), which has proved to be a practical
synthetic tool,20 was adopted to accelerate the C−O coupling
reaction. Under this condition, the reaction time could be
significantly reduced to 4 h with the yield maintained at 45%
(entry 6). The single crystal of 3 was grown by slow
evaporation from acetonitrile solution, and its structure was
supported by X-ray crystallographic analysis (Figure S-2).
However, the C−O coupling reaction of 1 with 2 was found
to be sluggish under similar conditions (entry 7), so that higher
reaction temperature was required to facilitate the process. This
may be due to the inductive electron-withdrawing properties of
the carbazole group12a that reduce the nucleophilic properties
of the phenol unit. After several attempts under various
conditions, we discovered that the coupling reactions could be
conducted in dipolar aprotic solvents at 150 °C by application
of microwave (MW) irradiation (entries 8−11). The
disubstituted product 4 was obtained in moderate yields. The
deiodo product 5 could also be isolated (31%) as a side
product. Deprotection of 4 was performed by demethylation in
the presence of pyridinium chloride in NMP under MW
irradiation21 for 7 h to give the key dendritic wedge 6 in 86%
yield.
As for our continuing efforts in arylamine derivatives,13 N-
phenylcarbazole (NPC) is another interesting monomeric unit
that can be used to construct the CB dendrimer.8c,e,14 Although
NPC and N-H-carbazole belong to the same family of
heterocyclic compounds and exhibit similar photophysical15
and electrochemical16 properties, pure NPC-based dendrimers
are seldom reported.17 Most carbazole-based dendrimers have
been prepared through the formation of C−N linkages,
coupling the nitrogen atom of an external carbazole unit to
the activated C3 and C6 sites of an inner carbazole moiety. In
the past decade, copper-catalyzed Ullmann condensations have
been extensively used for diaryl ether synthesis.18 Therefore, we
are interested in extending our study to the novel oxygen-linked
CB dendrimer CBD-G2 and its derivatives listed in Scheme 1.
We foresee that these compounds may possess intriguing
oxidation potential gradient in the designated structure.
However, achieving the C−O coupling at the C3 and C6 site
of carbazole moiety is challenging. We have thus developed a
novel synthetic approach for building up our Ar−O−Ar linkage,
instead of using the commonly adopted Ar−N−Ar linkage to
construct the molecular framework. We are also interested in
exploring whether the novel C−O−C linkage would bring any
interesting effects on the physical properties of the NPC units.
Synthesis of the CBD-G0, CBD-G1, and CBD-G2. Tris(4-
bromophenyl)amine (7)22 was employed as the precursor of
the electron-rich TPA core in the syntheses of CBD-G0, CBD-
G1, and CBD-G2 (Scheme 2). By applying our MW-assisted
Ullmann condensation protocol to the present synthesis, 8
could be smoothly coupled with phenol in benzonitrile at 150
°C to give CBD-G0 (57%). Similar conditions could be applied
to 2, giving CBD-G1 a 59% yield. The de-bromo disubstituted
product 9 (24%) could also be incidentally isolated. This
reaction yield of CBD-G1 is surprisingly high, considering that
three C−O couplings proceed in one step.
By employing the same strategy, CBD-G2 could be prepared
from 6 and 7 in 31% yield. Although the yield is as not
amazingly high as that of CBD-G1, it is still acceptable because
the procedure allows ones to prepare the desired CBD-G2 in
one pot. Furthermore, CBD-G2 could be purified conveniently
by liquid column chromatography. The structures of CBD-G1
1
and CBD-G2 were verified with H and 13C spectroscopy,
elemental analysis, and mass spectroscopy. In the gel-
permeation chromatography (GPC), the purity and mono-
dispersity of dendrimers were clearly confirmed (Figure 1).
Photophysical Properties. Pertinent photophysical data
for 3, 4, 5, 8, CBD-G0, CBD-G1, and CBD-G2, including their
UV−vis absorption properties, as well as the PL behavior, were
752
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Scheme 1
Table 1. CuI-Catalyzed Cross-Coupling of 1 with ArOH under Different Ligand-Assisted Conditions
c
entry
ArOH
ligand
solvent
DMSO
MW
temp (°C)
time (h)
product (%)
a
1
PhOH
PhOH
PhOH
PhOH
PhOH
PhOH
2
TrisOH
no
120
100
120
reflux
reflux
100
100
150
150
150
150
48
48
24
72
96
4
3 (20)
3 (22)
3 (7)
a
2
CYCOEt
DMSO
no
a
3
DMCHDA
DMGlyHCl
DMGlyHCl
DMGlyHCl
DMGlyHCl
DMGlyHCl
DMGlyHCl
DMGlyHCl
DMGlyHCl
PhMe
no
a
4
1,4-dioxane
1,4-dioxane
1,4-dioxane
1,4-dioxane
PhCN
no
3 (60)
3 (58)
3 (45)
d
a
5
no
b
6
yes
yes
yes
yes
yes
yes
b
7
4
b
8
2
4
4 (49)
4 (51)
4 (46)
4 (47)
b
9
2
DMF
4
b
10
2
NMP
4
b
11
2
DMSO
4
a
b
1 (0.5 M, 1 equiv), phenol (2.4 equiv), CuI (20 mol %), ligand (40 mol %), Cs2CO3 (4 equiv), solvent (0.77 M). 1 (0.5 M, 1 equiv), 2 (2.4 equiv),
c
d
CuI (20 mol %), DMGlyHCl (60 mol %), Cs2CO3 (3 equiv), solvent (0.77 M). Isolated yield. No reaction.
collected and summarized in Table 2. In addition, we adopted
NPC and triphenylamine (TPA) as references for comparison.
UV−vis Absorption Behavior. The solution UV−vis spectra
shown in Figures 2 and 3 were collected in THF (Uvasol
spectral grade) at the concentration of ∼1 × 10−5 M. The UV−
vis spectra of all NPC derivatives show intense UV absorption
bands (ε ∼ 104 M−1 cm−1) due to the spin-allowed π−π*
transitions in the singlet manifold, ranging from 250 to 381 nm.
NPC shows two absorption bands at 292 and 340 nm,
assigned as S0 → S2 and S0 → S1, respectively.23 Characteristic
753
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Scheme 2. Synthesis of the CBD-G0 through Our Modified Ullman Coupling Conditions
Figure 2. Absorption spectra of NPC, 3−5, and 8 in THF. For details
of the absorptivity in terms of extinction coefficient, please see Table 2.
Figure 1. GPC analyses for the dendrimers CBD-G0 to CBD-G2.
fine structures were observed in both bands (Figure 2). 9-(4-
Methoxyphenyl)-9H-carbazole (8) shows similar spectral
behavior with the absorption only slightly red-shifted to 293
and 341 nm, indicating that the electronic perturbations arising
from the methoxy substituent on the N-phenyl ring are
minimal.15b
On the other hand, when aryloxy substituents are introduced
at the C3- and C6-positions of the carbazole ring, electronic
perturbation becomes so significant that 3 exhibits the
absorption maxima red-shifted substantially to 304 and 356
nm, respectively, along with the disappearance of the fine
vibronic structures. As extending the NPC from 3, the dendron
4 exhibits four sets of UV−vis absorption. The bands peaking at
λmax = 292 and 341 nm are attributed to the absorption arising
Figure 3. Absorption spectra of TPA, CBD-G0, CBD-G1, and CBD-
G2 in THF. See Table 2 for the corresponding absorption extinction
coefficient.
Table 2. Photophysical Data for NPC Derivatives
a
a
a
UV absorption
fluorescence
phosphorescence
c
c
c
d
b
c
e
λabs (ε × 10−4)
λonset
λmax
τ
QY
λ0,0
T1 level
NPC
292 (1.93)
293 (1.79)
304 (2.15)
292 (6.35)
340 (0.40)
341 (0.41)
356 (0.42)
341 (1.30)
361 (0.50)
341 (0.89)
352 (0.44)
348
352
381
380
347
356
387
386
6.44 (7.57)15c
0.29
0.25
0.21
0.19
409
411
437
438
3.03
3.02
2.86
2.85
8
3
4
6.18
4.25
4.19
f
f
f
304 (2.80)
5
292 (4.11)
367
372
4.59
0.22
425
2.92
f
300 (2.28)
TPA
299 (2.13)
303 (3.48)
292 (8.80)
340
375
375
358
389
392
355
385
2.31 (2.19)25
1.86
0.07
0.06
0.11
414
433
434
3.00
2.85
2.86
CBD-G0
CBD-G1
g
g
340 (2.07)
2.14
f
h
302 (4.70)
0.23
CBD-G2
292 (21.1)
340 (4.25)
362 (1.70)
379
2.1 (32%)
0.14
435
2.85
f
f
302 (13.1)
4.2 (68%)
a
b
Measured in THF. Quantum yield (QY): Quantified in THF against coumarin 1 (QY: 0.85) as standard.24 The value obtained by this method
c
d
e
f
would be within 30% accuracy. Wavelength (nm). Measured in THF at 298 K, nanosecond (ns). T1 level (eV) = 1240.8/λ0,0 (nm). Broad
shoulders. 392 nm: major emission band, 355 nm: very minor emission band. Monitored at 350 nm.
g
h
754
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 4. (a) Solution fluorescence spectra of NPC, 3−5, and 8 at room temperature. (b) Solution fluorescence spectra of TPA and CBD-G0 to
CBD-G2 for comparison. The emission spectra are normalized at each peak wavelength. The excitation wavelength is selected at the peak
wavelength of the lowest absorption band.
Table 3. Pertinent Data of Lifetime, kISC, and ΔES1−T1 for the Titled Compounds
a
b
c
d
d
e
τ(298 K)
τ(77 K)
τ(77 K)/τ(298 K)
kISC
λ0,0(S1)
λ0,0(T1)
ΔES1−T1
NPC
6.44
6.18
4.25
4.19
4.59
2.31
1.86
8.10
8.51
5.87
5.43
5.63
2.98
2.05
1.26
1.38
1.38
1.30
1.23
1.29
1.10
1.33
1.37
7.8
7.7
342
346
369
366
355
337
367
367
366
409
411
437
438
425
414
433
434
435
4790
4570
4220
4490
4640
5520
4150
4210
4330
8
3
12
4
14
13
31
46
30
11
5
TPA
CBD-G0
CBD-G1
CBD-G2
f
f
2.14
2.84
g
g
3.53
4.85
a
b
c
d
Measured in THF at 298 K, nanosecond (ns). Measured in THF at 77 K, nanosecond (ns). Derived from eq 4; units: ×107 S−1. Wavelength
e
(nm). Derived from the reciprocal of low-temperature (77 K) fluorescence λ0,0 (obtained from the crossing-point of the normalized absorption and
f
g
fluorescence spectra) and phosphorescence λ0,0, wavenumber (cm−1) at 77 K. Monitored at 380 nm. The weighted average of two components.
from the outer NPC units that possess similar chromophores
with 8, while the bands peaking at λmax = 304 and 361 nm are
attributed to the absorption arising from the inner NPC unit
analogous with 3. The half-dendron 5 demonstrates an
absorption spectrum slightly different from that of 4; the
inner monosubstituted carbazole displays the shoulder bands at
λmax = 300 and 352 nm. The degree of red shifts is apparently
lower in comparison to that of 3 and 4, manifesting a lesser
perturbation of the mono aryloxy (5) than that of the dual
aryloxy (3, 4) substitution.
Figure 3 depicts the UV−vis absorption spectra of TPA,
CBD-G0, CBD-G1, and CBD-G2. CBD-G0 has the absorption
peaking at 303 nm. The λmax is slightly red-shifted from that of
TPA. In addition, a small shoulder is observed at 350 nm with a
broad tail extending to 375 nm. The spectrum of CBD-G1
consists of a shoulder band at 302 nm ascribed to the TPA
moiety as well as the bands at λmax = 292 and 340 nm
originating from the NPC chromophore. As the generation
grows to CBD-G2, the contribution of dendron 4 could be
clearly observed so that the spectrum comprises the absorption
band at λmax = 340 nm for the outer carbazole units analogous
with 8 and the bands at λmax = 362 nm for the inner carbazole
units similar to that of 4. The well overlaying of the above
spectra implicates that the conformations of the Ar−O−Ar
linkage25 and the N-phenyl group26 are twisted so that the
electronic coupling between the NPC units is broken down. In
these situations, all carbazole groups are decoupled and behave
essentially as independent chromophores. Therefore, the
transitions of the core TPA as well as the carbazole groups at
the inner or outer site can be assigned unambiguously.
Table 2), low-temperature fluorescence and phosphorescence
(Ph) at 77 K in degassed THF (Uvasol spectral grade) glass
(Figures 7 and 8), and their associated relaxation dynamics
(Table 3) were performed to shed light on the effects of the
dendrons on the PL properties of the dendrimers. The
quantum yields were quantified in THF against coumarin 1
(QY: 0.85) as a standard according to the protocol published
elsewhere24a,b and announced by Jobin-Yvon Ltd. (Horiba
Scientific).24c Although the QY being measured by this method
is not as accurate as that obtained by an absolute method using
an integrating sphere, this approach has been widely used due
to the fact that the relative trends of the QY changes are still
reliable for a fair comparison. The spectra were collected within
the concentration range of 1 × 10−6−5 × 10−6 M, in which a
linear fluorescence response to the concentration was obtained
(Figure S-6). The observation of the linear response indicated
that the interferences arising from the self-quenching and self-
absorption problems are negligible in this concentration range.
Room Temperature FL Behavior. Introduction of the
aryloxy substituents onto either the TPA or the NPC cores
leads to the perturbation on their fluorescence spectra. NPC,
the reference compound, emits at 347 nm with vibronic fine
structures being observed. The quantum yield (QY) was found
to be 0.29 in THF (298 K), which is higher than that of the
carbazole dendrons 3 (0.21) and 4 (0.19). Red-shifts in the
emission spectra are observed in moving from the least aryloxy-
substituted 8 at λmax = 356 nm (Figure 4a) to the dendron 4 at
λmax = 386 nm, along with decreasing quantum yield. The trend
of the bathochromic shifts is parallel to the red shifts in their
UV spectra. The preservation of the vibronic fine patterns,
regardless of the significant red shift of the spectra, signifies that
all the emission of the dedrons originates from the rigid
Photoluminescence (PL) Studies. Room temperature
fluorescence (FL) emission in degassed THF (Figure 4 and
755
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
carbazole units. The decreasing of the QY as well as the spectral
red shifts may due to the electronic perturbations arising from
the electron-donating aryloxy substituent on the carbazole ring.
On the other hand, the aryloxy substituent might also introduce
the loose-bolt and free-rotor effects27the torsional motion of
which, in part, might induce the additional radiationless
deactivation pathway and hence quench the emission. Not
until the low-temperature fluorescence lifetime measurements
have been carried out, we are unable to identify the importance
of the loose-bolt and free-rotor effects on the quantum yield.
This issue will be discussed in the later section regarding
relaxation dynamics in the 77 K solid matrix. It is also
noteworthy that the emission spectra of 3 and 4 well overlay
with each other, indicating that the luminescence is mainly
arising from the inner carbazole fluorophore. Also, evidenced
by the absorption spectrum, the outer carbazole units on 4 do
not cause significant perturbation to electronic structures of the
inner carbazole unit (vide supra). Therefore, the lack of
emission from outer carbazole units indicates that the outer
carbazole units, upon excitation, must undergo fast energy
transfer to the inner carbazole units, giving rise to the observed
emission.
The dendrimers CBD-G0, CBD-G1, and CBD-G2 also
display red-shifted emission spectra at λmax ∼ 390 nm in
comparison with that of TPA (358 nm) (Figure 4b). As for
CBD-G0, since no NPC unit involves in the framework, the
emission originating from the intact tri-para-phenoxy-substi-
tuted TPA unit is unambiguous. As for CBD-G1, first of all, the
spectral profile and peak wavelength (390 nm) of the major
emission band are nearly identical with that of CBD-G0. Thus,
both CBD-G0 and CBD-G1 may share the same origin of the
emission from the core tri-para-aryloxy-substituted TPA unit.
This viewpoint is further supported by nearly the same
emission lifetime of 1.9 and 2.1 ns for CBD-G0 and CBD-
G1, respectively. However, as shown in Figure 4b, we notice
that for CBD-G1, a minor, short wavelength emission
maximized at ∼350 nm cannot be ignored. We tentatively
assign this emission to the outer mono aryloxy-substituted
NPC unit (see Scheme 1) due to its similar spectral profile and
peak position with respect to that of 8. However, upon
monitoring at 350 nm, the emission decay time is measured to
be 230 ps, which is much faster than the lifetime of 6.2 ns for 8.
Since the 350 nm emission overlaps with the absorption tail of
CBD-G0 (Figure 3), it is thus reasonable to propose the
occurrence of energy transfer from the outer mono aryloxy-
substituted NPC unit to the core tri-para-aryloxy-substituted
TPA moiety for CBD-G1.
Figure 5. Study of the energy transfer phenomenon from 8 to CBD-
G0 in polystyrene: (a) The blue lines show the absorption and
emission of 8 in PS. (b) The red lines show the emission and the
excitation spectra of the thin-film 8-CBD-G0-PS (10:4:100 by weight).
(c) The green lines show the emission and the excitation spectra of the
thin-film 8-CBD-G0-PS (10:8:100 by weight).
and transferred to CBD-G0 and finally contributed to the
emission at 392 nm.
Another question to be addressed is the possibility of having
self-quenching between the carbazole side-arms on CBD-G1,
which might also reduce of the FL intensity of carbazole part.
Although the linear response of the fluorescence intensity
versus the concentration of 8 in the QY measurements
indicated that the intermolecular self-quenching phenomenon
is not particularly serious in highly diluted conditions, it is still
difficult to justify if the self-quenching would occur when three
carbazole side-arms being grouped with the TPA core in a close
distance. Therefore, fluorescence experiments with various
doping concentration 8 in PS thin films were adopted to probe
this problem. In these measurements, front-face detection was
adopted to reduce the interference arising from the self-
absorption phenomenon.24d Figure 6 shows the fluorescence
Figure 6. Study of the fluorescence self-quenching phenomenon of 8
in PS at various doping levels. The samples were excited at 292 nm,
with the emission intensity being traced at 330 nm.
The energy transfer phenomenon was further investigated by
using a model system of 8 and CBD-G0 in polystyrene (PS)
(Figure 5). As mentioned before, compound 8 shows
characteristic absorption peaking at 330 and 345 nm. When
the film of 8 in PS (10 wt %) was irradiated at 321 nm, strong
PL at 355 and 370 nm were recorded. However, mixing CBD-
G0 (4 wt %) into the 8 (10 wt %)−PS blend led to significant
fluorescent quenching at 355 and 370 nm. Instead, a new
emission band from CBD-G0 was clearly observed. The
emission from 8 was completely quenched when the amount of
CBD-G0 was increased to 8 wt %. To further confirm the
energy transfer phenomenon, excitation spectra of the blends
were collected by monitoring the fluorescence at 392 nm. The
observation of the corresponding signals at 330 and 345 nm,
the signature of 8, strongly supports the energy transfer
mechanism, in which 8 would have the photoenergy absorbed
response of the films with different doping weight ratios of 8 in
PS. One can easily to perceive that the self-quenching problem
is insignificant even with high doping weight ratio up to 12%.
On the basis of the results discussed above, and by assuming
that the energy-transfer mechanism is the major pathway that
leads to the fluorescence decay, we take the lifetime of 230 ps
(350 nm band in CBD-G1) and 6.2 ns (for 8) to be the time
scale of energy transfer for NPC and energy-transfer-free NPC
moiety; the efficiency of energy transfer ET (%) can thus be
estimated from the expression
k
ET
ET (%) =
k
+ k
fl
(1)
ET
756
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Scheme 3. Energy Transfer Pathways of Three Representative Dendron and Dendrimers
where kET is the energy transfer rate constant and kfl is the
fluorescence decay rate constant with no energy transfer
process involved, which are (230 ps)−1 and (6.2 ns)−1,
respectively. As a result, ET (%) for the mono aryloxy-
substituted NPC unit to the core tri-para-aryloxy-substituted
TPA moiety in CBD-G1 is estimated to be only ∼96%. For
dendron 4, due to the lack of emission from the outer carbazole
units, ET is expected to be complete (100%) from the outer
carbazole units to the inner carbazole units; in other words, the
kET for CBD-G1 is presumably much slower that of 4, an
ultrafast one.
However, the rational explanations behind these observations
have not yet been completely disclosed. Therefore, it is
necessary to collect more evidence in order to outline the
mechanisms. One possibility leading to the major difference
between mono-aryloxy NPC → tri-aryloxy TPA (in CBD-G1)
and mono-aryloxy NPC → triaryloxy NPC (in 4) may lie in
that the former has relatively poor spectral overlap between
energy donor (emission of mono-aryloxy NPC) and energy
acceptor (absorption of tri-aryloxy TPA), as shown in Figures 3
and 4b. This is a fundamental requirement for having either the
transfer from the outmost mono-aryloxy NPC to the inner tri-
aryloxy NPC units with unity efficiency does occur. (2) The
energy transfer from the inner tri-aryloxy NPC units to the core
tri-aryloxy TPA is sluggish so that emissions from the inner tri-
aryloxy NPC units and the core tri-aryloxy TPA could be
simultaneously observed.
Of course, ineffective energy transfer in the latter case is
understandable due to the facts that (1) the core tri-aryloxy
TPA and the inner tri-aryloxy NPC units are electronically
decoupled (vide supra) and (2) both units have nearly identical
S0/S1 gap (Figures 2−4) so that the overlap between their
absorption (either acceptor) and emission (either donor) is
minimal. According to these reasons, one may perceive that
each excited-state relaxation can be treated independently,
giving dual emission covering nearly the same spectral range.
This hypothesis could also explain for the observation of the
high percentage weight for the longer lifetime carbazole
emission (τ = 4.2 ns): Dendron 4 and CBD-G0 have nearly
identical extinction coefficients of 3.1 × 104 and 3.5 × 104 M−1
cm−1 at 299 nm but very different fluorescence quantum yields
of 0.19 and 0.06, respectively. On the basis of extinction
coefficient of the dendron and the 3-fold symmetry of CBD-
G2, one might expect that around 70% of photons should be
harvested by the three dendritic branches on CBD-G2, while
about 30% of photons would be captured by the TPA core. If
each S1-state relaxation can be treated independently, and by
taking the QY factor into account, the contribution from the
carbazole dendrons in the emission intensity should be around
90%. Although one may expect to see some discrepancy based
on this qualitative model, the observation of nearly 70%
contribution from the longer lifetime carbazole clearly supports
that the energy transfer from carbazole dendron to the core
TPA must be minimal.
Dexter energy transfer or the Forster resonance energy transfer.
̈
A more clear-cut picture will be provided in a later discussion
on the basis of the HOMO−LUMO diagram.
For the largest generation dendrimer CBD-G2, as shown in
Figure 4b, the spectral coverage of emission is nearly the same
as that of CBD-G0 or CBD-G1. However, unlike the
structureless emission profile in CBD-G0 and CBD-G1, the
emission regains certain vibronic profile similar to that observed
in the dendrons 3 or 4. Moreover, the fluorescence decay,
obtained by excitation at 299 nm, cannot be well fitted by a
single-exponential component. Instead, the decay is fitted by
the combination of two single-exponential decay components,
indicating the existence of dual emissions. The lifetimes are
fitted to be 2.1 ns (32%) and 4.2 ns (68%), which, within
experimental error, well correlate with the lifetime of CBD-G1
core and dendron 3 (or 4), respectively. This, together with the
aforementioned results for all dendrons, CBD-G0 and CBD-
G1, leads us to the following conclusion about the dual
emission behavior of CBD-G2: (1) Similar to that of 4, energy
To sum up in brief, Scheme 3 depicts the energy transfer
pathways of three representative cases: dendron 4 and
dendrimers CBD-G1 and CBD-G2.
Low-Temperature FL and Ph Properties. NPC derivatives
are commonly used as the phosphorescent host materials for
OLEDs.10b,28 To further explore the steady-state emission
behavior of the dendrimers, especially the triplet state energy
757
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 7. Low-temperature PL behavior of 3−5 and 8: (a) Steady state photoluminescence spectra being collected in THF glass at 77 K. (b)
Correlation between Nphos/Nfl value and the number of ArO substituents on the carbazole unit.
Figure 8. Low-temperature PL behavior of CBD-G0, CBD-G1, CBD-G2, and dendron 4: (a) Steady state photoluminescence spectra being
collected in THF glass at 77 K. (b) Correlation between Nphos/Nfl value and the generation of the dendrimers.
for the later application, low-temperature (LT) steady-state PL
spectra of 3−5 and 8 and CBD-G0, CBD-G1, and CBD-G2 in
the organic glass of THF (Uvasol spectral grade) at 77 K were
collected and shown in Figures 7 and 8. The low-temperature
steady-state PL experiments were carried out using Hitachi F-
4500 spectrofluorometer equipped with a quartz Dewar flask
setup for handling the measurements in liquid nitrogen. The
data were collected at the fluorescence and phosphorescence
modes, respectively, in degassed THF glass. When the spectra
were collected at the fluorescence mode, both phosphorescence
and fluorescence emissions could be simultaneously observed.
The spectra of the phosphorescence part are identical with their
spectra measured in the phosphorescence mode, signifying that
these signals really originate from the phosphorescence. A
similar phenomenon in solid matrix has been reported by
Bonesi and Erra-Balsells for carbazole.15c The observation of
the phosphorescent signals attracted us to systematically
explore this phenomenon, and the related data are summarized
in Table 2.
ArO Substituent Effects on the Phosphorescence Emission.
Most of them exhibit phosphorescence at the region of 410−
440 nm, corresponding to triplet state energy of 3.03−2.81 eV.
Introduction of the ArO substituents at the C3 and C6
positions leads to a small red-shift effect on the triplet state
emission. The first band of the phosphorescence emission of 8,
5, and 4 is therefore recorded as 411, 425, and 441 nm,
respectively. It is worth to emphasize here the phosphorescence
emissions of 3 and 4 are very similar and well-overlaying with
each others, indicating that the inner carbazole core is the key
phosphor in these cases. According to the above data, we
conclude that the dendrons and the corresponding dendrimers
possess substantially high energy for their lowest lying triplet
state and therefore are suitable host materials for blue-light
triplet emitters that would be discussed in later sections.
Intensity Ratio for the Phosphorescence versus Fluo-
rescence. The intensity ratio reported in Figures 7 and 8 were
estimated based on the integration of the spectral signals and
denoted as Nphos/Nfl. As shown in Figure 7b, the least aryloxy-
substituted 8 shows strong fluorescence over the phosphor-
escence emission at 77 K with the Nphos/Nfl value of 0.81 was
found. When one ArO substituent is introduced at the C3
position to form 5, the intensity ratio for phosphorescence
versus fluorescence was significantly enhanced to Nphos/Nfl =
2.1. When two ArO substituents are introduced onto the
carbazole ring, the relative intensity of Nphos/Nfl = 2.3 was
obtained for 4. Compound 3 also shows a value of Nphos/Nfl =
2.0, indicating that aryloxy substituent effects are salient
features in the presence system.
On the other hand, the dendrimers exhibit photophysical
behaviors that are different from the dendritic wedges.
Although CBD-G0 shows PL in a similar region, the observed
relative phosphorescence intensity is much stronger, and hence
a substantially larger Nphos/Nfl value of 13.1 was recorded
(Figure 8a). When three NPC units are introduced in CBD-
G1, the Nphos/Nfl value drops significantly to 4.9. Since this
value is still far larger than that of 2.3 for dendron 8, we
therefore believe that the emission from the TPA core might be
overwhelming in this case. This is consistent with the early
conclusion about the NPC → TPA energy transfer process in
CBD-G1. In CBD-G2, the Nphos/Nfl value of 2.3 is nearly
identical to that of 4 (Nphos/Nfl = 2.3) (Figure 8). It is again
consistent with our previous suggestion that the carbazole
dendrons on CBD-G2 is the major group responsible for the
observed emission behavior.
Although the above systematic analyses revealed the
ratiometric phosphorescence/fluorescence changes of the
compounds, any justifications or conclusions simply attributed
to the structural change are not immediately obvious. Many
758
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 9. Correlation analysis about kISC: (a) Plot of kISC as a function of 1/(ΔES1−T1)2. (b) Plot of Nphos/Nfl against kISC
.
parameters about the fluorescence and the intersystem crossing
(ISC) process may vary as a result of the variances of the
singlet−triplet energy gap. Therefore, further understanding
about the relaxation dynamics becomes essential for our study.
Time-Resolved PL Spectroscopy and the Associated
Relaxation Dynamics. To elucidate the correlation between
the ratiometric phosphorescence/fluorescence changes and the
ISC (S1 → Tn) rate constants among the titled dendrons and/
or dendrimers, their fluorescence properties were characterized
by time-resolved PL spectroscopy.15c,d,27a,b
suppressed at low temperature in an organic glass, which leads
to a longer lifetime. Among the data we collected, the τ(77 K)/
τ(298 K) = 1.26 for NPC, which serves as a reference molecule,
is recorded. Since the carbazole unit is relatively rigid, and only
the N-phenyl ring is allowed to rotate freely in NPC, we
therefore suggest that the thermally activated deactivation
process may arise from the free rotation of the phenyl unit.
Further introduction of RO groups to the carbazole ring might
also lead to some enhancement of the τ(77 K)/τ(298 K) value
even though the change is not that remarkable. For example, 3
and 8 actually show a higher value of τ(77 K)/τ(298 K) = 1.38.
Although CBD-G1 exhibits slightly slower decay time of τ =
2.84 ns than that of CBD-G0, this value is much shorter than
that of 8. The τ(77 K)/τ(298 K) value of 1.33 for CBD-G1 is
close to that of 3 and 8, revealing that similar extents of the
loose-bolt and free rotor effects are operating in CBD-G1.
CBD-G2 exhibits a decay time constant of τ = 5.4 ns (79%)
and 2.8 ns (21%), which well correlate with the lifetime of
dendron 4 and CBD-G1, respectively. Both results are
consistent with the resulting trend at 298 K and are in good
agreement with the dual emission hypothesis. Again, this
observation supports the argument that energy transfer from
the outmost mono-aryloxy NPC to the inner tri-aryloxy NPC
units may be more efficient than energy transfer from inner
triaryloxy NPC units to TPA core (CBD-G2) so that dual
emission could be observed.
Pertinent data measured in THF at 77 K are summarized in
Table 3. Among the titled compounds studied, the longest
lifetime τ = 8.51 ns (λfl
= 356 nm) was obtained for
max
carbazole 8. Having the aryloxy substituents at the C3 and/or
C6-positions, carbazole dendrons 3, 4, and 5 show similar
fluorescence lifetime of τ = 5.87, 5.43, and 5.63 ns, respectively.
As for the dendrimers, CBD-G0 exhibits the shortest
fluorescence decay time of τf = 2.05 ns.
By assuming that the intermolecular quenching processes are
insignificant in highly diluted solutions, the equations that
could be used to estimate the overall deactivation rate in S1, i.e.
kobs, and the corresponding fluorescence quantum yield can be
expressed as
1
τobs
kobs = kf + kISC + knr(T) =
(2)
(3)
Theoretical Implications. Under the assumption that the
intersystem crossing takes place solely via the S1 → T1 pathway,
the corresponding rate constant, kISC, could thus be expressed
as
k
k
f
k
obs
f
Φ (T) =
=
f
k + k + k (T)
f
ISC
nr
The radiative decay rate constant, kf, is an intrinsic property,
which is proportional to the square of the transition moment
and is thus independent of the temperature. A similar
standpoint can be applied for kISC if S1 → T1 is the only
intersystem crossing channel. Assuming that the thermally
activated deactivation process knr(T) in eqs 2 and 3 is
drastically suppressed in the 77 K THF solid matrix and can
thus be neglected, kISC could be deduced as
2
⟨Ψ |H |Ψ ⟩
T
SO S
1
1
k
∝
ISC
2
(ΔE
)
1
S −T
(5)
1
where HSO is the Hamiltonian for spin−orbit coupling (SOC)
and ΔES1−T1 is the energy difference between S1 and T1 states
due to the difference in electron correlation energy,27d which
can be calculated by the gap of the first vibronic peak energy
between fluorescence and phosphorescence (see Table 3). For
compounds having the same type of fluorophore, their SOC
term should be similar and hence their kISC should be inversely
proportional to (ΔES1−T1)2. We then made an attempt by
plotting the kISC value as a function of 1/(ΔES1−T1)2. First, as
shown in Figure 9, the plotted data of the carbazole dendrons
3−5, 8, NPC, and CBD-G2 are lined up to give a linear
correlation. Knowing that the slope should be proportional to
spin−orbit coupling constant expressed in eq 5, the linear
correspondence of the carbazole dendrons and CBD-G2
indicates that this family of compounds may have a similar
Φ (298 K)
τ(298 K)
1
f
k
= k
− k =
−
ISC
(77K)
f
τ(77 K)
(4)
Knowing the lifetime in 77 and 298 K, together with
fluorescence QY estimated at 298 K, the values of kISC
among the titled compounds are listed in Table 3.
As one can see in Table 3, the fluorescence lifetimes for all
dendrons and dendrimers at 77 K increase. The ratio of τ(77
K)/τ(298 K) falls into a range of 1.10−1.38, dependent on the
chemical structure. This observation is consistent with our
assumption that the thermally activated deactivation process is
759
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 10. Electrochemical study of the dendrons: (a) Cyclic voltammograms of 3−5. (b) Differential pulse voltammograms of 3−5, 8. Solvent:
CH2Cl2.
Figure 11. Electrochemical behavior of the dendrimers: (a) Cyclic voltammograms of CBD-G0, CBD-G1 and CBD-G2. (b) Differential pulse
voltammograms of CBD-G0, CBD-G1, CBD-G2, and 4 for comparison. Solvent: CH2Cl2.
Table 4. Thermal and Electrochemical Data for N-Phenylcarbazole Derivatives
a
c
c
thermal analysis (°C)
Eonset (V) (CV)
Eox (V) (DPV)
energy level (eV)
d
e
f
g
Tg
Td
E1
E2
Epa1
E1/2
HOMO
Eg
LUMO
NPC
1.32
1.26
1.08
1.18
1.14
1.01
0.76
0.82
0.82
1.38
1.31
1.15
1.24
1.20
1.07
0.81
0.87
0.87
1.41
1.34
1.18
1.27
1.23
1.10
0.84
0.90
0.90
−5.68
−5.61
−5.45
−5.54
−5.50
−5.37
−5.11
−5.17
−5.17
3.57
3.53
3.25
3.38
3.27
3.65
3.31
−2.11
−2.08
−2.20
−2.16
−2.23
−1.72
−1.80
8
3
86
378
351
505
5
65
1.28
1.21
4
129
TPA
b
CBD-G0
CBD-G1
CBD-G2
−
314
571
617
1.34
1.28
1.22
129
198
a
b
c
d
Scan rate: 10 deg/min. No Tg observed, Tc = 108 °C. In CH2Cl2 with Bu4NClO4 (0.1 M) as supporting electrolyte vs Ag/AgCl. E1/2 = Epa1
ΔE/2, where ΔE = 0.05 V. HOMO = −(E1/2 − 0.53) − 4.8 (eV). Band gap =1240.8/absorption λonset (nm). LUMO = HOMO + band gap (eV).
+
e
f
g
SOC term and their emissions mainly originate from the NPC
chromophore.
and differential pulse voltammetry (DPV).29 The HOMO
values are determined from the estimated E1/2 with respect to
ferrocene (Fc). Selected CV and DPV diagrams are displayed in
Figures 10 and 11. Pertinent data of the oxidation potentials
(versus Ag/AgCl couple) are summarized in Table 4.
On the other hand, slower kISC rate was obtained with the
generation growth of dendrimers. The kISC of CBD-G0, CBD-
G1, and CBD-G2 vary dramatically even their ΔES1−T1 values
are confined within a narrow range of 4150−4350 cm−1,
indicating that CBD-G0, CBD-G1, and CBD-G2 should
possess a very different SOC term. This result implies that
very different intersystem crossing and relaxation mechanisms
are operating in these compounds; this picture is consistent
with our previous conclusion being shown in Scheme 3.
It is noteworthy to point out that the value of Nphos/Nfl being
obtained in the last section is positively correlated to the kISC
value (Figure 9b). This relationship clearly explains for the fact
that higher intersystem crossing rate from S1 to T1 would lead
to stronger the relative phosphorescence emission in the low-
temperature PL experiments.
The first point to note from the electrochemistry about the
dendrons is that there appears to be a trend for the first wave
shifting to less positive potentials when more ArO substituents
are introduced to the carbazole unit (Figure 10). NPC
displayed irreversible oxidation wave that is attributed to the
oxidation of the carbazole ring with the onset voltage at 1.32 V.
The irreversibility is arising from the oxidative NPC
dimerization.30 Because of the π-electron-donating ability of
the MeO group, 8 shows lower onset oxidation voltage at 1.26
V, which is about 0.06 V earlier than that of NPC. When the π-
electron donating PhO substitutents were introduced at the C3
and C6 positions of 3, further destabilization on the HOMO
occurs so that the CV wave exhibits a reversible pattern with
the onset oxidation at 1.08 V, which coincides with the
Electrochemical Oxidation Properties. The electro-
chemical properties were studied by cyclic voltammetry (CV)
760
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 12. Distribution of the oxidation potential and mimicked energy diagrams of CBD-G1 and CBD-G2.
reported result of 9-ethyl-3,6-dimethoxy-9H-carbazole.16a An
explanation for the shift to the less oxidative potential is that
the electron density on the carbazole unit would be
substantially enriched due to the presence of the PhO
substituents. On the other hand, irreversible oxidations were
observed for 4 and 5, with the onsets at 1.14 and 1.18 V,
respectively. It is necessary to point out that the oxidation
potential of 3 is lower than that of 4 by 0.06 V, reflecting the
fact that the outer carbazole units on 4 are more electron-
withdrawing than the phenyl substituent on 3. This is
consistent with the Yamamoto’s conclusion on G4Fc,12a in
which the carbazole units are electron-withdrawing through
inductive or π-polarization effects.
The irreversible oxidative process on the NPCs was also
examined by DPV method and could be partially resolved.
From the DPV (see Figure 10b), it was estimated that the first
oxidation of 8 was at ∼1.31 V while that of 3 was at ∼1.15 V.
On the other hand, 5 demonstrates two waves peaking at 1.24
and 1.42 V. The first wave is assigned to the electrochemical
oxidation of the inner 3-ArO-substituted carbazole while the
second wave is assigned to the outer one. Dendron 4 shows
three waves peaking at 1.20, 1.35, and 1.49 V that are
reasonably attributed to the oxidations of the inner carbazole as
well as two outer carbazoles, respectively.
On the other hand, both cyclic voltammograms of CBD-G1
and CBD-G2 exhibit two oxidation waves: one corresponds to
the TPA core oxidation while the other is relevant to the
oxidation of the carbazole dendron (Figure 11a). Both of the
first waves of CBD-G1 and CBD-G2 are reversible and have
the E1/2 appearing at 0.90 V. They are unambiguously assigned
to the single electron oxidation of the TPA core. However, the
E1/2 value is higher than that of CBD-G0, supporting the
assumption that the dendritic wedge is electron-withdrawing as
elaborated in the early sections.
The second set of irreversible oxidation for CBD-G1 and
CBD-G2 onsets at 1.28 and 1.22 V, respectively. Although the
waves are complicated and could only be partially resolved even
by DPV methods (Figure 11b), we believe that they are
corresponding to the oxidation of the NPC dendrons due to
the close Eonset values of the CBD-G1, CBD-G2, 4, and 5.
Close examination revealed that the Eonset of CBD-G2 is almost
identical to the Eonset of 4. This result implies that the dendritic
wedge oxidation would not be repelled by cation formed on the
TPA+ core in the first oxidation. In addition, the Eonset of CBD-
G2 is slightly lower than those of CBD-G1. This is reasonable
due to the electron rich environment of the inner carbazole of
the CBD-G2.
Summarizing the above electrochemistry results, we conclude
that the central core of CBD-G2 is the most electron-rich, the
inner-carbazole units form the second electron-rich interlayer,
while the outmost carbazoles construct the least electron-rich
shell (Figure 12). Therefore, both CBD-G1 and CBD-G2
possess oxidation potential gradients such that the outer layer is
electron-poor and the inner layer is electron-rich. This result is
similar to the phenylazomethine dendrimers,32 but opposite to
the dendrimers with pure N-H-carbazole backbone.12a
The second group of molecules worthy to compare with is
CBD-G0, CBD-G1, and CBD-G2. Unlike CBD-G0 that has
the simple electrochemical oxidation behavior, the electro-
chemical properties of CBD-G1 and CBD-G2 are intricate by
integration of the dendritic wedges to the TPA core. The waves
could only be partially resolved even by DPV methods (Figure
11b).
First of all, two oxidation processes onset at 0.76, and 1.34 V
are respectively observed in the CV of CBD-G0. The first
oxidation process is reversible with the onset lower than that of
TPA by 0.25 V, implying that CBD-G0 is more electron-rich
due to the presence of the π-donating ArO substituents. The
Estimation of the HOMO−LUMO Levels. The HOMO
levels of the compounds were estimated according to the E1/2
values obtained from the DPV experiments. Herein the HOMO
levels were calculated on the basis of the equation of HOMO =
−(E1/2 − 0.53) − 4.8 (eV). For simple compounds such as 3−
5, 8, NPC, TPA, and CBD-G0, their LUMO levels could be
estimated directly from the equation of LUMO = HOMO +
band gap (eV), provided that the corresponding optical band
gaps could be easily obtained from the UV−vis absorption
E1/2 of 0.84 V was recorded accordingly. On the other hand, the
second oxidation is irreversible with the onset at 1.34 V, which
might be due to the oxidation of the radical cation (CBD-G0•+)
to dication (CBD-G02+), giving rise to the quinoidal forms.31
761
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
experiments. However, due to the multichromophoric behavior
of the dendrimers of CBD-G1 and CBD-G2, evaluation of their
electronic levels is relatively complicated. Fortunately, our early
results elucidated that the subgroups in each shell of the
dendrimers are only weakly coupled so that they can be
considered as electronically isolated subgroups. Therefore, we
were able to build their energy level diagrams respectively, as
shown in Figure 12, based on the HOMO−LUMO values of
CBD-G0, 3, and 8. From the diagram one can easily perceive
that the HOMO of CBD-G1 and CBD-G2 are located at the
TPA core, while the corresponding LUMO falls onto the
carbazole units.
Figure 13. Thermal gravimetric analyses of CDB-G0 to CBD-G2.
On the basis of these HOMO−LUMO diagrams, our energy
transfer mechanism proposed in Scheme 3 could be explained:
(1) Transfer of the energy from the outer carbazole to the inner
carbazole in CBD-G2 is exothermic. In addition, while the
LUMO level of the inner carbazole is lower than the outer one,
the HOMO level of the inner carbazole is higher than that of
the outer one. This type of HOMO−LUMO alignment would
be beneficial for the Dexter energy transfer mechanisms, and
therefore fast energy transfer is allowed to occur in this case.
This explanation could also be applicable to the dendron 4 due
to the fact that all subgroups could be treated as electronically
isolated ones. (2) On the other hand, the energy transfer
process from the outer carbazole to the core TPA in CBD-G1
would be relatively slow in comparison to the first one. In this
case, although energy transfer process is still exothermic due to
the fact that the energy gap of the outer carbazole is larger than
that of the core TPA unit, the LUMO energy level of the core
TPA is higher than that of the outer carbazole by 0.28 eV (+27
kJ). The endothermic alignment of the LUMO energy levels
would slow down the energy transfer process. This could help
to understand why residual emission from the outer carbazole
could be observed in the PL spectrum of CBD-G1 in Figure 4b.
(3) Energy transfer from the inner carbazole of CBD-G2 to the
core TPA unit is unfavorable. The energy transfer process is
slightly endothermic by 0.06 eV (+6 kJ). In addition, the
LUMO energy level alignment is endothermic by 0.4 eV (+39
kJ). These two factors prohibit any effective energy transfer
from the inner carbazole to the core TPA. On the other hand,
the reverse energy transfer from the core TPA to the inner
carbazole is also unfavorable due to the endothermic alignment
of the HOMO by 0.34 eV (+33 kJ). This would lead to the
observation of the dual-emission phenomenon.
It is noteworthy to point out the interesting LUMO energy-
well features in CBD-G2; either the LUMO of the outer-shell
carbazole units or the central TPA core would have higher
LUMO energy levels in comparison to the LUMO of the inner-
shell carbazole units. This feature would play important role in
the electroluminescence behavior that would be discussed in
the later sections.
Thermal Properties of the Carbazole Dendrons and
Dendrimers. All O-linked-NPC shows high thermal stability
in the TGA analysis (Figure 13). In particular, CBD-G1 and
CBD-G2 can stand for high temperature up to 500 °C or above
under nitrogen atmosphere. Except CBD-G0, the dendritic
NPCs form amorphous organic glass. No crystallization takes
place from the corresponding amorphous glasses when heated
after Tg. This is a key criterion to fabricate thermally stable thin
film materials for optoelectronic applications.
HOMO levels, which matches the Fermi level of PEDOT:PSS
(−5.1 eV).33 The hole injection between the interfaces of
PEDOT/host would become more efficient. More importantly,
the triplet energy of CBD-G1 and CBD-G2 is higher than that
of FIrPic measured to be ∼2.75 eV. Therefore, the reverse
energy transfer from the triplet emitter to host34 would be
prohibited. Along with their good thermal stability, CBD-G1
and CBD-G2 are potential candidates of solution-processable
dendritic hosts for blue-electrophosphorescent device. To
evaluate the performance of CBD-G1 and CBD-G2 as host
materials, a device consisting of ITO (100 nm)/Host-FIrPic
(10 wt %) (40 nm)/TAZ (50 nm)/LiF (1.2 nm)/Al (120 nm)
was employed for the preliminary study (Figure 14). The
maximum luminous efficiency of CBD-G2 could be achieved
up to 24.7 cd/A at 484 cd/m2. It is quite appealing that the
CBD-G2 device could maintain good luminous efficiency at
such high brightness. Another comparable result10b was
reported in pure carbazole-based conjugated dendritic host at
lower brightness. As a reference, small molecule 3 has also been
applied as host material for FIrPic by thermal deposition
processes. However, only ∼2 cd/A maximum luminous
efficiency could be obtained on the multilayer devices.
The fluorescence quantum yield of triphenylamine is smaller
than that of N-phenylcarbazole due to its less rigid structure.
This also indicates faster nonradiative deactivation processes
and shorter lifetime of singlet state in triphenylamine. In light
of this viewpoint, triphenylamine should be a poor candidate
for host materials since those fast nonradiative deactivation
processes would compete with the energy transfer process and
lower the efficiency of singlet-state harvesting. The singlet
excited state is mainly populated on the TPA core in CBD-G1
and on the inner tri-aryloxy NPC units in CBD-G2 (vide
supra). Therefore, the high luminous efficiency of CBD-G2
could be attributed to two combination factors. On the one
hand, the singlet-state harvesting is facilitated via energy
transfer from the inner tri-aryloxy NPC units to the triplet
emitters. On the other hand, as evidenced by both CV and
DPV data (vide supra), HOMO of CBD-G2 is strictly located
at the TPA core, whereas LUMO is mainly confined at the
inner NPC sites. The spatial separation of HOMO and LUMO
energy level may, in part, lead to high efficiency of the device
performance.35
CONCLUSION
■
In summary, a series of novel oxygen-linked N-phenylcarbazole
(NPC) dendritic wedges (3−5) and triphenylamine-centered
dendrimers (CBD-G0 to -G2) have been synthesized and
extensively studied. The analyses of UV−vis spectra revealed
that the Ar−O−Ar linkage would break down the π
Performance of NPC Derivatives as Host in the
Phosphorescent OLED. According to the results above, the
NPC-based dendrimers CBD-G1 and CBD-G2 possess high
762
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
Figure 14. (a) Brightness and current density versus applied voltage characteristics for CBD-G1 (squares) and CBD-G2 (triangles). (b) Luminous
and power efficiency versus current density for CBD-G1 (squares) and CBD-G2 (triangles).
13C NMR (100 MHz, CD2Cl2): δ 158.39, 152.81, 144.35, 130.27,
conjugation and make NPC units manifest their individual
125.69, 123.50, 120.63, 118.82. HRMS (EI) calcd for C36H27NO3
521.1991 (M+), obsd 521.1990. Anal. Calcd for C36H27NO3: C, 82.90;
H, 5.22; N, 2.69. Found: C, 83.04; H,4.93; N, 2.63.
absorption peaks. The PL properties were fully characterized by
luminescence spectroscopy at both room temperature and 77 K
in THF. The results reveal the operation of energy transfer
between NPC and core TPA as well as the NPCs between
different generations. As a result, the origin of emission in both
singlet and triplet manifolds can be ascribed to two categories,
namely the TPA, CBD-G0, and CBD-G1 from the TPA moiety
and the rest of titled compounds from NPC chromphore. Note
that CBD-G2 may possess both but keen on the major NPC
moiety. From the measurement of the oxidation potential of
the NPC derivatives, the mesomeric electron-donating effect of
the oxygen atom and inductive electron-withdrawing effect of
the carbazole unit were revealed. The potential gradient could
be established on those triphenylamine-centered dendrimers
(CBD-G1 and CBD-G2) such that the outer layer is electron-
poor and the inner layer is electron-rich. Standing on these
intriguing physical properties, using CBD-G2 as the host
material, the FIrpic-based electrophosphorescent devices were
successfully fabricated and the maximum luminous efficiency
could be achieved up to a record high of 24.7 cd/A at 484 cd/
m2.
Tris(4-(4-(9H-carbazol-9-yl)phenoxy)phenyl)amine (CBD-
G1). Elution: hexane/CH2Cl2 = 2/1; white solid; yield 59%; mp
1
176 °C, Tg = 129 °C, Td = 571 °C. H NMR (400 MHz, CD2Cl2): δ
8.15 (d, J = 8.0 Hz, 6H), 7.53 (d, J = 8.7 Hz, 6H), 7.44−7.38 (m,
12H), 7.30−7.25 (m, 12H), 7.21 (d, J = 8.9 Hz, 6H), 7.11 (d, J = 8.9
Hz, 6H). 13C NMR (100 MHz, CD2Cl2): δ 157.75, 152.48, 144.75,
141.73, 132.83, 129.17, 126.49, 125.92, 123.72, 121.14, 120.76, 120.38,
119.76, 110.21. HRMS (FAB) calcd for C72H48N4O3 1016.3726 (M+),
obsd 1016.3730. Anal. Calcd for C72H48N4O3: C, 85.02; H, 4.76; N,
5.51. Found: C, 85.19; H, 4.78; N, 5.32.
Tris(4-(4-(3,6-bis(4-(9H-carbazol-9-yl)phenoxy)-9H-carbazol-
9-yl)phenoxy)phenyl)amine (CBD-G2). Elution: hexane/CH2Cl2 =
1/1; white solid; yield 31%; mp 231−233 °C, Tg = 198 °C, Td = 617
1
°C. H NMR (400 MHz, CD2Cl2): δ 8.13 (d, J = 7.8 Hz, 12H), 7.90
(d, J = 2.3 Hz, 6H), 7.58 (d, J = 2.1 Hz, 6H), 7.29−7.44 (m, 18H),
7.41−7.36 (m, 24H), 7.32−7.20 (m, 42H), 7.12 (d, J = 9.0 Hz, 6H).
13C NMR (200 MHz, CD2Cl2): δ 159.05, 157.96, 152.38, 150.33,
144.79, 141.74, 139.47, 132.61, 132.28, 129.10, 126.46, 125.94, 124.27,
123.66, 121.21, 120.73, 120.31, 119.85, 118.93, 112.36, 111.66, 110.20.
MS (MALDI) calcd for 13CC179H114N10O9 2559.9 (M+), obsd 2559.2.
Anal. Calcd for C180H114N10O9: C, 84.42; H, 4.49; N, 5.47. Found: C,
84.64; H, 4.44; N, 5.25.
EXPERIMENTAL SECTION
■
General Procedure of Microwave-Assisted Synthesis. The
reaction flask was charged with corresponding aryl halides (1 equiv),
aryl phenols (2.4 equiv for wedges and 3.6 equiv for dendrimers),
copper iodide (0.2 equiv), N,N-dimethylglycine hydrochloride (0.6
equiv), Cs2CO3 (3 equiv for wedges and 4 equiv for dendrimers), and
benzonitrile (0.77 M). The reaction mixtures were then put into
microwave reactor and reacted at 200 W, open vessel, 150 °C, for 4 h.
The reaction mixture was quenched with water and extracted with
CH2Cl2. The organic layer was separated and dried over anhydrous
MgSO4. The solution was filtered through Celite and distilled under
vacuum to remove the high polar solvent. The crude product was
purified thorough chromatography.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental section, UV−vis and PL spectra, NMR spectra,
crystal structure, cif file, and several additional data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*Phone +886-2-3366-1673, Fax +886-2-23636359, e-mail
mkleung@ntu.edu.tw (M.-k.L.). Phone 886-2-3366-3894, Fax
886-2-2369-5208, e-mail chop@ntu.edu.tw (P.-T.C.).
9,9′-(4,4′-(9-(4-Methoxyphenyl)-9H-carbazole-3,6-diyl)bis-
(oxy)bis(4,1-phenylene))bis(9H-carbazole) (4). Elution: hexane/
CH2Cl2 = 1/1; white solid; yield 49%; mp 145 °C, Tg = 129 °C, Td =
1
505 °C. H NMR (400 MHz, CD2Cl2): δ 8.14 (d, J = 7.7 Hz, 2H),
ACKNOWLEDGMENTS
■
7.91 (d, J = 2.3 Hz, 2H), 7.54−7.49 (m, 6H), 7.42−7.38 (m, 10H),
7.32−7.22 (m, 6H), 7.17 (d, J = 8.9 Hz, 2H), 3.93 (s, 3H). 13C NMR
(100 MHz, CD2Cl2): δ 159.78, 159.15, 150.16, 141.77, 139.74, 132.23,
130.51, 129.11, 128.98, 126.47, 124.11, 123.67, 120.73, 120.30, 118.90,
115.78, 112.33, 111.70, 110.23, 56.21. HRMS (FAB) calcd for
C55H37N3O3 787.2835 (M+), obsd 787.2823. Anal. Calcd for
C55H37N3O3: C, 83.84; H, 4.73; N, 5.33. Found: C, 83.63; H, 4.74;
N, 5.21.
The present work was supported by the Ministry of Education
and National Taiwan University, Academia Sinica Thematic
Project, National Science Council of Taiwan (NSC-98-2119-M-
002-006-MY3, NSC-98-2221-E-002-038-MY3, NSC-99-2218-
E-155-003, NSC-99-2221-E-155-092, and NSC-99-2622-E-155-
010-CC3).
Tris(4-phenoxyphenyl)amine (CBD-G0). Elution: hexane/
REFERENCES
CH2Cl2 = 2/1; white solid; yield 57%; mp 131 °C, Tc = 108 °C, Td
■
1
= 373 °C. H NMR (400 MHz, CD2Cl2): δ 7.33 (t, J = 8.0 Hz, 6H),
(1) (a) Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev. 2010, 110,
1857. (b) Caminade, A.-M.; Majoral, J.-P. Chem. Soc. Rev. 2010, 39,
7.10−7.06 (m, 9H), 7.01 (d, J = 8.0 Hz, 6H), 6.93 (d, J = 8.9 Hz, 6H).
763
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
2034. (c) Franc, G.; Kakkar, A. K. Chem. Soc. Rev. 2010, 39, 1536.
(d) Newkome, G. R.; Shreiner, C. Chem. Rev. 2010, 110, 6338.
(2) (a) Li, Y.; Hou, T. Curr. Med. Chem. 2010, 17, 4482. (b) Mintzer,
M. A.; Grinstaff, M. W. Chem. Soc. Rev. 2011, 40, 173. (c) Roeglin, L.;
Lempens, E. H. M.; Meijer, E. W. Angew. Chem., Int. Ed. 2011, 50, 102.
(3) (a) Percec, V.; Glodde, M.; Bera, T. K.; Miura, Y.; Shiyanovskaya,
I.; Singer, K. D.; Balagurusamy, V. S. K.; Heiney, P. A.; Schnell, I.;
Rapp, A.; Spiess, H. W.; Hudson, S. D.; Duan, H. Nature 2002, 419,
384. (b) Lo, S.-C.; Burn, P. L. Chem. Rev. 2007, 107, 1097.
(4) (a) Astruc, D.; Chardac, F. Chem. Rev. 2001, 101, 2991.
(b) Astruc, D. Tetrahedron: Asymmetry 2010, 21, 1041. (c) Astruc, D.;
Ornelas, C.; Diallo, A. K.; Ruiz, J. Molecules 2010, 15, 4947.
(5) (a) Taranekar, P.; Park, J.-Y.; Patton, D.; Fulghum, T.; Ramon, G.
J.; Advincula, R. Adv. Mater. 2006, 18, 2461. (b) Kaewtong, C.; Jiang,
G.; Felipe, M. J.; Pulpoka, B.; Advincula, R. ACS Nano 2008, 2, 1533.
(c) Park, Y.; Taranekar, P.; Park, J. Y.; Baba, A.; Fulghum, T.;
Ponnapati, R.; Advincula, R. C. Adv. Funct. Mater. 2008, 18, 2071.
(d) Jiang, G.; Ponnapati, R.; Pernites, R.; Grande, C. D.; Felipe, M. J.;
Foster, E.; Advincula, R. Langmuir 2010, 26, 17629. (e) Kaewtong, C.;
Jiang, G.; Ponnapati, R.; Pulpoka, B.; Advincula, R. Soft Matter 2010, 6,
5316.
(6) (a) Loiseau, F.; Campagna, S.; Hameurlaine, A.; Dehaen, W. J.
Am. Chem. Soc. 2005, 127, 11352. (b) Li, S.; Zhu, W.; Xu, Z.; Pan, J.;
Tian, H. Tetrahedron 2006, 62, 5035. (c) Xu, T. H.; Lu, R.; Qiu, X. P.;
Liu, X. L.; Xue, P. C.; Tan, C. H.; Bao, C. Y.; Zhao, Y. Y. Eur. J. Org.
Chem. 2006, 4014. (d) Kwon, T.-H.; Kim, M. K.; Kwon, J.; Shin, D.-Y.;
Park, S. J.; Lee, C.-L.; Kim, J.-J.; Hong, J.-I. Chem. Mater. 2007, 19,
3673. (e) Xu, T.; Lu, R.; Liu, X.; Zheng, X.; Qiu, X.; Zhao, Y. Org. Lett.
2007, 9, 797. (f) Xu, T.; Lu, R.; Liu, X.; Chen, P.; Qiu, X.; Zhao, Y.
Eur. J. Org. Chem. 2008, 1065.
T. Tetrahedron Lett. 2006, 47, 8949. (f) Promarak, V.; Ichikawa, M.;
Sudyoadsuk, T.; Saengsuwan, S.; Jungsuttiwong, S.; Keawin, T. Synth.
Met. 2007, 157, 17. (g) Albrecht, K.; Kasai, Y.; Kimoto, A.; Yamamoto,
K. Macromolecules 2008, 41, 3793. (h) Promarak, V.; Ichikawa, M.;
Sudyoadsuk, T.; Saengsuwan, S.; Jungsuttiwong, S.; Keawin, T. Thin
Solid Films 2008, 516, 2881. (i) Usluer, O.; Demic, S.; Egbe, D. A. M.;
Birckner, E.; Tozlu, C.; Pivrikas, A.; Ramil, A. M.; Sariciftci, N. S. Adv.
Funct. Mater. 2010, 20, 4152.
(10) (a) Tsai, M.-H.; Hong, Y.-H.; Chang, C.-H.; Su, H.-C.; Wu, C.-
C.; Matoliukstyte, A.; Simokaitiene, J.; Grigalevicius, S.; Grazulevicius,
J. V.; Hsu, C.-P. Adv. Mater. 2007, 19, 862. (b) Ding, J.; Zhang, B.; Lu,
̈
J.; Xie, Z.; Wang, L.; Jing, X.; Wang, F. Adv. Mater. 2009, 21, 4983.
(c) Yang, J.; Ye, T.; Zhang, Q.; Ma, D. Macromol. Chem. Phys. 2010,
211, 1969. (d) Soh, M. S.; Santamaria, S. A. G.; Williams, E. L.; Perez-
Morales, M.; Bolink, H. J.; Sellinger, A. J. Polym. Sci., Part B: Polym.
Phys. 2011, 49, 531. (e) Yang, J.; Ye, T.; Ma, D.; Zhang, Q. Synth. Met.
2011, 161, 330.
(11) (a) Hameurlaine, A.; Dehaen, W. Tetrahedron Lett. 2003, 44,
957. (b) Kimoto, A.; Cho, J.-S.; Higuchi, M.; Yamamoto, K. Chem.
Lett. 2003, 32, 674. (c) McClenaghan, N. D.; Passalacqua, R.; Loiseau,
F.; Campagna, S.; Verheyde, B.; Hameurlaine, A.; Dehaen, W. J. Am.
Chem. Soc. 2003, 125, 5356.
(12) (a) Albrecht, K.; Yamamoto, K. J. Am. Chem. Soc. 2009, 131,
2244. (b) Selby, T. D.; Blackstock, S. C. J. Am. Chem. Soc. 1998, 120,
12155. (c) Xu, Z.; Moore, J. S. Acta Polym. 1994, 45, 83. (d) Tada, T.;
Nozaki, D.; Kondo, M.; Yoshizawa, K. J. Phys. Chem. B 2003, 107,
14204. (e) Wang, J.; Yan, J.; Tang, Z.; Xiao, Q.; Ma, Y.; Pei, J. J. Am.
Chem. Soc. 2008, 130, 9952. (f) Bronk, K.; Thayumanavan, S. J. Org.
Chem. 2003, 68, 5559. (g) Holzapfel, M.; Lambert, C. J. Phys. Chem. C
2008, 112, 1227. (h) Lambert, C.; Schelter, J.; Fiebig, T.; Mank, D.;
Trifonov, A. J. Am. Chem. Soc. 2005, 127, 10600.
(13) (a) Lin, K.-R.; Chien, Y.-H. C.; Chang, C.-C.; Hsieh, K.-H.;
Leung, M.-k. Macromolecules 2008, 41, 4158. (b) Chiang, C. C.; Chen,
H.-C.; Lee, C.-s.; Leung, M.-k.; Lin, K.-R.; Hsieh, K.-H. Chem. Mater.
2008, 20, 540. (c) Leung, M.-k.; Chang, C.-C.; Wu, M.-H.; Chuang,
K.-H.; Lee, J.-H.; Shieh, S.-J.; Lin, S.-C.; Chiu, C.-F. Org. Lett. 2006, 8,
2623. (d) Chien, C.-H.; Leung, M.-k.; Su, J.-K.; Li, G.-H.; Liu, Y.-H.;
Wang, Y. J. Org. Chem 2004, 69, 1866. (e) Leung, M.-k.; Chou, M.-Y.;
Su, Y. O.; Chiang, C. L.; Chen, H.-L.; Yang, C. F.; Yang, C.-C.; Lin, C.-
C.; Chen, H.-T. Org. Lett. 2003, 5, 839.
(7) (a) Namdas, E. B.; Ruseckas, A.; Samuel, I. D. W.; Lo, S.-C.;
Burn, P. L. Appl. Phys. Lett. 2005, 86, 091104. (b) Lo, S.-C.; Namdas,
E. B.; Shipley, C. P.; Markham, J. P. J.; Anthopolous, T. D.; Burn, P. L.;
Samuel, I. D. W. Org. Electron. 2006, 7, 85. (c) Ding, J.; Gao, J.; Cheng,
Y.; Xie, Z.; Wang, L.; Ma, D.; Jing, X.; Wang, F. Adv. Funct. Mater.
2006, 16, 575. (d) Jung, K. M.; Kim, K. H.; Jin, J.-I.; Cho, M. J.; Choi,
D. H. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 7517. (e) Ding, J.;
Lu, J.; Cheng, Y.; Xie, Z.; Wang, L.; Jing, X.; Wang, F. Adv. Funct.
̈
Mater. 2008, 18, 2754. (f) Knights, K. A.; Stevenson, S. G.; Shipley, C.
P.; Lo, S.-C.; Olsen, S.; Harding, R. E.; Gambino, S.; Burn, P. L.;
Samuel, I. D. W. J. Mater. Chem. 2008, 18, 2121. (g) Liu, Q.-D.; Lu, J.;
Ding, J.; Tao, Y. Macromol. Chem. Phys. 2008, 209, 1931. (h) Ding, J.;
(14) Wong, K.-T.; Lin, Y.-H.; Wu, H.-H.; Fungo, F. Org. Lett. 2007,
9, 4531.
Lu, J.; Cheng, Y.; Xie, Z.; Wang, L.; Jing, X.; Wang, F. J. Organomet.
(15) (a) Mo, Y.; Bai, F.; Wang, Z. J. Photochem. Photobiol. A 1995, 92,
25. (b) Zaion, S. M.; Hashim, R.; Taylor, A. G.; Phillips, D. J. Mol.
Struct.: THEOCHEM 1997, 401, 287. (c) Bonesi, S. M.; Erra-Balsells,
R. J. Lumin. 2001, 93, 51. (d) De, A. K.; Ganguly, T. J. Lumin. 2001,
92, 255.
(16) (a) Ambrose, J. F.; Carpenter, L. L.; Nelson, R. F. J. Electrochem.
Soc. 1975, 122, 876. (b) Lamm, V. W.; Pragst, F.; Jugelt, W. J. Prakt.
Chem. 1975, 317, 995. (c) Desbene-Monvernay, A.; Lacaze, P. C.;
Dubois, J. E. J. Electroanal. Chem. 1981, 129, 229.
̈
Chem. 2009, 694, 2700. (i) Ding, J.; Wang, B.; Yue, Z.; Yao, B.; Xie, Z.;
Cheng, Y.; Wang, L.; Jing, X.; Wang, F. Angew. Chem., Int. Ed. 2009,
48, 6664. (j) Iguchi, N.; Pu, Y.-J.; Nakayama, K.-i.; Yokoyama, M.;
Kido, J. Org. Electron. 2009, 10, 465. (k) Jung, K. M.; Lee, T. W.; Kim,
K. H.; Cho, M. J.; Jin, J.-I.; Choi, D. H. Chem. Lett. 2009, 38, 314.
(l) Kim, H.-J.; Kim, M.-J.; Park, H.-D.; Lee, J.-H.; Noh, S. T.; Lee, Y.-
C.; Kim, J.-J. Synth. Met. 2010, 160, 1994.
(8) (a) Pan, J.; Zhu, W.; Li, S.; Zeng, W.; Cao, Y.; Tian, H. Polymer
2005, 46, 7658. (b) Lu, J.; Xia, P. F.; Lo, P. K.; Tao, Y.; Wong, M. S.
Chem. Mater. 2006, 18, 6194. (c) Liu, Q.-D.; Lu, J.; Ding, J.; Day, M.;
Tao, Y.; Barrios, P.; Stupak, J.; Chan, K.; Li, J.; Chi, Y. Adv. Funct.
Mater. 2007, 17, 1028. (d) Adhikari, R. M.; Duan, L.; Hou, L.; Qiu, Y.;
Neckers, D. C.; Shah, B. K. Chem. Mater. 2009, 21, 4638. (e) Zhang,
H.; Wang, S.; Li, Y.; Zhang, B.; Du, C.; Wan, X.; Chen, Y. Tetrahedron
2009, 65, 4455. (f) Zhao, Z.; Li, J.-H.; Chen, X.; Wang, X.; Lu, P.;
Yang, Y. J. Org. Chem. 2009, 74, 383. (g) Jin, H.; Xu, Y.; Shen, Z.; Zou,
D.; Wang, D.; Zhang, W.; Fan, X.; Zhou, Q. Macromolecules 2010, 43,
8468. (h) Zhao, Z.-H.; Jin, H.; Zhang, Y.-X.; Shen, Z.; Zou, D.-C.; Fan,
X.-H. Macromolecules 2011, 44, 1405.
(17) (a) Zhu, Z.; Moore, J. S. Macromolecules 2000, 33, 801. (b) Zhu,
Z.; Moore, J. S. J. Org. Chem. 2000, 65, 116. (c) Feng, G.-L.; Ji, S.-J.;
Lai, W.-Y.; Huang, W. Synlett 2006, 2841.
(18) (a) Monnier, F.; Taillefer, M. Angew. Chem., Int. Ed. 2008, 47,
3096. (b) Monnier, F.; Taillefer, M. Angew. Chem., Int. Ed. 2009, 48,
6954. (c) Das, P.; Sharma, D.; Kumar, M.; Singh, B. Curr. Org. Chem.
2010, 14, 754. (d) Niu, J.; Zhou, H.; Li, Z.; Xu, J.; Hu, S. J. Org. Chem.
2008, 73, 7814. (e) Xia, N.; Taillefer, M. Chem.Eur. J. 2008, 14,
6037. (f) Jammi, S.; Sakthivel, S.; Rout, L.; Mukherjee, T.; Mandal, S.;
Mitra, R.; Saha, P.; Punniyamurthy, T. J. Org. Chem. 2009, 74, 1971.
(g) Larsson, P.-F.; Correa, A.; Carril, M.; Norrby, P.-O.; Bolm, C.
Angew. Chem., Int. Ed. 2009, 48, 5691. (h) Naidu, A. B.; Jaseer, E. A.;
Sekar, G. J. Org. Chem. 2009, 74, 3675. (i) Sreedhar, B.; Arundhathi,
R.; Reddy, P. L.; Kantam, M. L. J. Org. Chem. 2009, 74, 7951.
(j) Zhang, Q.; Wang, D.; Wang, X.; Ding, K. J. Org. Chem. 2009, 74,
7187.
(9) (a) Du, P.; Zhu, W.-H.; Xie, Y.-Q.; Zhao, F.; Ku, C.-F.; Cao, Y.;
Chang, C.-P.; Tian, H. Macromolecules 2004, 37, 4387. (b) Kimoto, A.;
Cho, J.-S.; Higuchi, M.; Yamamoto, K. Macromol. Symp. 2004, 209, 51.
(c) Kimoto, A.; Cho, J.-S.; Ito, K.; Aoki, D.; Miyake, T.; Yamamoto, K.
Macromol. Rapid Commun. 2005, 26, 597. (d) Albrecht, K.; Yamamoto,
K. J. Photopolym. Sci. Technol. 2006, 19, 175. (e) Promarak, V.;
Ichikawa, M.; Meunmart, D.; Sudyoadsuk, T.; Saengsuwan, S.; Keawin,
(19) (a) Ma, D.; Cai, Q. Org. Lett. 2003, 5, 3799. (b) Chen, Y.-J.;
Chen, H.-H. Org. Lett. 2006, 8, 5609. (c) Lv, X.; Bao, W. J. Org. Chem.
764
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765

Macromolecules
Article
2007, 72, 3863. (d) Jogdand, N. R.; Shingate, B. B.; Shingare, M. S.
Tetrahedron Lett. 2009, 50, 4019.
Ueda, M.; Akiike, T.; Miyamoto, H.; Kajita, T.; Kakimoto, M.-a. Chem.
Mater. 2008, 20, 2532. (d) Ge, Z.; Hayakawa, T.; Ando, S.; Ueda, M.;
Akiike, T.; Miyamoto, H.; Kajita, T.; Kakimoto, M.-a. Adv. Funct.
Mater. 2008, 18, 584.
(20) (a) Loupy, A. Microwaves in Organic Synthesis, 2nd ed.; Wiley-
VCH: Weinheim, 2006. (b) Kappe, C. O.; Dallinger, D.; Murphree, S.
S. Practical Microwave Synthesis for Organic Chemists; Wiley-VCH:
Weinheim, 2009. (c) He, H.; Wu, Y.-J. Tetrahedron Lett. 2003, 44,
3445. (d) Zhu, X.-H.; Chen, G.; Ma, Y.; Song, H.-C.; Xu, Z.-L.; Wan,
Y.-Q. Chin. J. Chem. 2007, 25, 546.
(21) (a) Lamba, M. S.; Makrandi, J. K. J. Chem. Res. 2007, 585.
(b) Mason, J. J.; Janosik, T.; Bergman, J. Synthesis 2009, 3642.
(22) Cao, X.; Wen, Y.; Guo, Y.; Yu, G.; Liu, Y.; Yang, L.-M. Dyes
Pigm. 2010, 84, 203.
(23) Sarkar, A.; Chakravorti, S. J. Lumin. 1998, 78, 205.
(24) (a) Rusalov, M.; Druzhinin, S.; Uzhinov, B. J. Fluoresc. 2004, 14,
193. (b) Williams, A. T. R.; Winfield, S. A.; Miller, J. N. Analyst 1983,
108, 1067. (c) For standard procedure announced by Jobin-Yvon Ltd,
see “A Guide to Recording Fluorescence Quantum Yields” at http://
www.jobinyvon.com/usadivisions/Fluorescence/applications/
quantumyieldstrad.pdf. (d) For the procedure and setup for the front
face detection announced by Jobin-Yvon Ltd, see http://www.horiba.
com/fileadmin/uploads/Scientific/Documents/Fluorescence/Small_
Samples_FL-6.pdf.
(25) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules, 2nd ed.; Academic Press: New York, 1971.
(26) (a) Rettig, W.; Zander, M. Chem. Phys. Lett. 1982, 87, 229.
(b) Galievsky, V. A.; Druzhinin, S. I.; Demeter, A.; Mayer, P.;
Kovalenko, S. A.; Senyushkina, T. A.; Zachariasse, K. A. J. Phys. Chem.
A 2010, 114, 12622.
(27) (a) Lakowicz, J. R. Principles of Fluorescence Spectrscopy, 2nd ed.;
Springer: New York, 1999. (b) Chou, P. T.; Chen, Y. C.; Yu, W. S.;
Chou, Y. H.; Wei, C. Y.; Cheng, Y. M. J. Phys. Chem. A 2001, 105,
1731. (c) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern
Molecular Photochemistry of Organic Molecules; University Science:
Sausalito, CA, 2005. (d) Turro, N. J. Modern Molecular Photochemistry:
University Sciences: Sausalito, CA, 1991; p 170. (e) McGlynn, S. P.;
Azumi, T.; Kinoshita, M. Molecular Spectroscopy of the Triplet State;
Prentice-Hall: Englewood Cliffs, NJ, 1969.
́
(28) (a) Marsal, P.; Avilov, I.; da Silva Filho, D. A.; Bredas, J. L.;
Beljonne, D. Chem. Phys. Lett. 2004, 392, 521. (b) Shih, P.-I.; Chiang,
C.-L.; Dixit, A. K.; Chen, C.-K.; Yuan, M.-C.; Lee, R.-Y.; Chen, C.-T.;
Diau, E. W.-G.; Shu, C.-F. Org. Lett. 2006, 8, 2799. (c) Sapochak, L. S.;
Padmaperuma, A. B.; Cai, X.; Male, J. L.; Burrows, P. E. J. Phys. Chem.
C 2008, 112, 7989. (d) Cosimbescu, L.; Koech, P.; Polikarpov, E.;
Swensen, J.; Von Ruden, A.; Rainbolt, J. E.; Padmaperuma, A. Dig.
Tech. Pap.Soc. Inf. Disp. Int. Symp. 2010, 41, 1887.
(29) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, 2nd ed.;
Wiley: New York, 2001.
(30) (a) Heinze, J.; Hinkelmann, K.; Dietrich, M.; Mortensen, J. Ber.
Bunsen-Ges. Phys. Chem. 1985, 89, 1225. (b) Oyama, M.; Matsui, J.
Bull. Chem. Soc. Jpn. 2004, 77, 953.
(31) (a) Wu, X.; Davis, A. P.; Lambert, P. C.; Kraig Steffen, L.; Toy,
O.; Fry, A. J. Tetrahedron 2009, 65, 2408. (b) Amthor, S.; Noller, B.;
Lambert, C. Chem. Phys. 2005, 316, 141. (c) Sreenath, K.; Thomas, T.
G.; Gopidas, K. R. Org. Lett. 2011, 13, 1134.
(32) (a) Higuchi, M.; Shiki, S.; Ariga, K.; Yamamoto, K. J. Am. Chem.
Soc. 2001, 123, 4414. (b) Yamamoto, K.; Higuchi, M.; Shiki, S.;
Tsuruta, M.; Chiba, H. Nature 2002, 415, 509. (c) Yamamoto, K.;
Imaoka, T. Bull. Chem. Soc. Jpn. 2006, 79, 511.
(33) Brunner, K.; van Dijken, A.; Borner, H.; Bastiaansen, J. J. A. M.;
̈
Kiggen, N. M. M.; Langeveld, B. M. W. J. Am. Chem. Soc. 2004, 126,
6035.
(34) (a) Holmes, R. J.; Forrest, S. R.; Tung, Y.-J.; Kwong, R. C.;
Brown, J. J.; Garon, S.; Thompson, M. E. Appl. Phys. Lett. 2003, 82,
2422. (b) Sudhakar, M.; Djurovich, P. I.; Hogen-Esch, T. E.;
Thompson, M. E. J. Am. Chem. Soc. 2003, 125, 7796.
(35) (a) Fan, C.; Chen, Y.; Jiang, Z.; Yang, C.; Zhong, C.; Qin, J.;
Ma, D. J. Mater. Chem. 2010, 20, 3232. (b) Ge, Z.; Hayakawa, T.;
Ando, S.; Ueda, M.; Akiike, T.; Miyamoto, H.; Kajita, T.; Kakimoto,
M.-a. Org. Lett. 2008, 10, 421. (c) Ge, Z.; Hayakawa, T.; Ando, S.;
765
dx.doi.org/10.1021/ma202433y | Macromolecules 2012, 45, 751−765