Synthesis and photophysical properties of 3,6-diphenyl-9-hexyl-9H-carbazole derivatives bearing electron withdrawing groups

Article abstract of DOI:10.1007/s00706-007-0774-3

3,6-Diphenyl-9-hexyl-9H-carbazole derivatives bearing electron withdrawing groups, such as the formyl or the nitro-group in 4-positions of the phenyl substituents, were prepared and characterized. Their photophysical properties were evaluated and compared

Full text of DOI:10.1007/s00706-007-0774-3

Monatsh Chem 139, 223–231 (2008)
DOI 10.1007/s00706-007-0774-3
Printed in The Netherlands
Synthesis and Photophysical Properties of 3,6-Diphenyl-9-hexyl-9H-carbazole
Derivatives Bearing Electron Withdrawing Groups
Gabriele Kremser1, Oliver T. Hofmann2, Stefan Sax2;3, Stefan Kappaun1;3,
Emil J. W. List2;3, Egbert Zojer2, and Christian Slugovc1;ꢀ
1
Institute of Chemistry and Technology of Organic Materials, Graz University of Technology, Graz, Austria
2
Institute of Solid State Physics, Graz University of Technology, Graz, Austria
3
NanoTecCenter Weiz Forschungsgesellschaft mbH, Weiz, Austria
Received July 18, 2007; accepted July 23, 2007; published online February 12, 2008
# Springer-Verlag 2008
Summary. 3,6-Diphenyl-9-hexyl-9H-carbazole derivatives
bearing electron withdrawing groups, such as the formyl or
the nitro-group in 4-positions of the phenyl substituents, were
prepared and characterized. Their photophysical properties
were evaluated and compared with those of the unsubstituted
counterpart 3,6-diphenyl-9-hexyl-9H-carbazole. The electron
withdrawing groups bearing compounds exhibited consider-
able red shifts of the absorption and the emission maxima.
While 3,6-di(4-nitrophenyl)-9-hexyl-9H-carbazole emitted in
the orange region of the visible spectrum with its emission
maximum peaking at 585 nm, 3,6-di(4-formylphenyl)-9-hex-
yl-9H-carbazole gave a pure blue emission with a lumines-
cence quantum yield of 95% peaking at 450 nm. Observed
features were explained using quantum mechanical calcula-
tions and organic light emitting diodes using the formylphenyl
substituted compound as emissive layer were built demon-
strating the practical applicability of this class of compounds.
chemical stability. The carbazole unit can be easily
functionalized at its 3- and 6-positions and can be
covalently attached to other organic moieties via the
9 position, greatly improving their thermal stability
or glassy-state durability [2]. Thus, carbazole deriv-
atives are finding applications in a variety of fields,
such as organic light-emitting diodes (OLEDs) [3],
organic photorefractive materials [4], organic field
effect transistors (OFETs) [5], photoconductors [6],
and as two-photon absorption dyes for the photo-
deposition of silver [7]. For instance, the simple
9-alkyl-9H-carbazole has been used as charge trans-
porting plasticizer of photorefractive systems [8]
and 3,6-disubstituted carbazoles have been investi-
gated as charge transporting, electroluminescent host
materials for OLEDs [9]. In most applications, 3,6-
disubstituted carbazole derivatives are used, because
these materials are readily accessible starting from
3,6-dihalogenated precursor compounds. Additional-
ly, 2,7-disubstituted carbazole derivatives have at-
tracted considerable interest [10], but progress in
this field is somewhat hampered by the tedious syn-
thesis procedures like the Cadogan cyclization [11]
needed to prepare the corresponding 2,7-dihalocar-
bazole precursors.
Keywords. Dyes; Heterocycles; Fluorescence spectroscopy;
Absorption spectra; Density functional theory.
Introduction
In the last decade, amorphous and film-forming car-
bazole derivatives have attracted much attention as
molecular electronic materials [1] due to their good
hole-transporting capability, high charge carrier mo-
bility, and intense luminescence. Additionally, car-
bazoles exhibit high thermal, morphological, and
Although numerous OLED materials have been
developed, ongoing design of new carbazole based
compounds is a main issue in synthetic chemistry in
ꢀ
Corresponding author. E-mail: slugovc@tugraz.at

224
G. Kremser et al.
order to improve efficiencies, brightness, and life-
times of optoelectronic devices [12]. For green and
red electrophosphorescent devices, promising host-
guest systems are well-established [3a]. It is, how-
ever, still a challenge to find suitable host materials
for triplet-state blue-light emitters [13] due to the fact
that commonly used host molecules do not have suf-
ficiently high triplet energies and, hence, do not pre-
vent reverse energy transfer from the guest back to
the host [14]. Thus, host materials with triplet ener-
gies greater than 2.75 eV are required [15]. A further
challenge comprises stable blue light emitting chro-
mophores for OLEDs. Although polyfluorenes (PF)
are well known materials for blue light-emitting di-
odes, they suffer from the drawback that they are lia-
ble to oxidation on the 9-position of fluorene, leading
to a drastic loss in quantum yield and the appearance
of low-energy emission (‘‘Keto-defect’’) [16].
In this work, we focused on the functionalization
of the 3- and 6-position of carbazole with differently
substituted phenyl groups in order to provide larger
conjugated entities and, in particular, to tune the pho-
tophysical properties by incorporation of electron
withdrawing groups (EWGs) in the para-position of
these phenyl substituents. In this way, a push-pull
chromophore, based on the carbazole’s nitrogen and
the EWG is created. Although donor-acceptor func-
tionalized molecules have received considerably atten-
tion in, e.g., two photon absorption dyes [17], the use
of EWGs such as the formyl or nitro group in electro-
active materials is not particularly common [18].
Results and Discussion
Starting from 3,6-dibromocarbazole, 3,6-dibromo-9-
hexyl-9H-carbazole (1) was obtained by a standard
alkylation reaction using 1-bromohexane as the al-
kylation reagent and NaOH as the base [19]. Target
molecules 3,6-diphenyl-9-hexyl-9H-carbazole (2),
3,6-di(4-formylphenyl)-9-hexyl-9H-carbazole (3),
and 3,6-di(4-nitrophenyl)-9-hexyl-9H-carbazole (4)
(cf. Scheme 1) were prepared via Suzuki cross-
coupling reactions of 1 with phenylboronic acid,
4-formylphenylboronic acid, and 4-nitrophenylbo-
ronic acid pinacol ester. A standard protocol using
[Pd(PPh3)4] as the catalyst in a mixture of toluene,
ethanol, and water and Na2CO3 as the base was used
[20]. Compound 2 was obtained in acceptable yield
of 63% upon purification by column chromatogra-
phy and subsequent washing with hot n-pentane. In
case of 3 and 4 yields were distinctly lower because
of the need to separate by-products with similar re-
tention indices. Compounds 3 and 4 were isolated
in 31 and 44% yield. Similar difficulties of the re-
moval of by-products were recently resolved by
using the alternative catalyst bis(dicyclohexylamine)
palladium acetate (DAPCy), ethanol as the solvent,
and KOH as base [21]. However, in the present cases
the above mentioned catalytic system failed. In all
cases, only yields below 10% could be obtained. Park
et al. reported the synthesis of 9-alkyl-3,6-diphenyl-
9H-carbazole derivatives via a nickel catalyzed
cross-coupling reaction of 3,6-dibromo-9-alkyl-9H-
Scheme 1

Synthesis and Photophysical Properties of 3,6-Diphenyl-9-hexyl-9H-carbazole
225
carbazole derivatives with the phenyl Grignard re-
agent [22] and obtained yields of 75–78%. This
synthesis route is certainly an alternative for the
preparation of 2, but cannot be directly adopted for
the synthesis of 3 and 4.
Additionally, 3,6-di-(4-hydroxymethylphenyl)-9-
hexyl-9H-carbazole (5) was prepared by reduction
of the two aldehyde functionalities in 3 with LiAlH4
in THF in a yield of 61%. Compounds 2–5 were
1
characterized by elemental analysis, 1H and 13Cf Hg
NMR as well as IR spectroscopy. NMR data were in
accordance with the proposed structures and did not
exhibit any special features. IR spectra of 2, 3, and
Fig. 1. TGA thermograms of compounds 2–5
4 are characterized by strong absorptions features
peaking at 1602, 1599, and 1513 cmꢁ1, which are
caused by an asymmetrical skeleton vibration of the
carbazole and the phenyl moieties [23]. Derivative
3 featured further characteristic peaks at 1697 and
1482 cmꢁ1, which were assigned to a CHO stretch-
ing vibration and a breathing mode of the pyrrole
core of the carbazole. In the nitro-group featuring
compound characteristic signals peaked at 1588 and
1337 cmꢁ1. These signals were assigned to an in-
plane bending vibration of the nitrogen atom of the
nitro group and to the NO2 stretching vibration.
Furthermore, thermal properties of 2–5 were de-
termined using combined DSC=TGA experiments.
Thermal stability is of particular interest, since it is
a crucial issue for device lifetime [24] and is needed
when thermal evaporation processes are employed.
While 2 showed a melting point at 123ꢂC, the melt-
ing points of 3 and 4 were found at higher tempera-
tures of 169 and 234ꢂC. Compound 5, on the other
hand, started to decompose at low temperatures of
about 50ꢂC as evidenced by the DSC as well as
by the TGA curve. The other compounds exhibited
thermal stability well above their melting points. As
a measure for the thermal stability, the temperature
at a weight loss of 5% was chosen. The correspond-
ing values are 326ꢂC for 2, 386ꢂC for 3, 346ꢂC for 4,
and 94ꢂC for 5. Figure 1 illustrates the results, which
reveal favourable thermal properties for 2, 3, and 4,
but which disqualifies 5 for the envisaged applica-
tion in OLEDs.
Table 1 and Fig. 2. To assign the various features in
the UV-Vis spectra, we performed time dependent-
density functional theory calculations using the hy-
brid B3LYP exchange-correlation functional together
with a SV(P) split valence polarised basis set. In
order to cross-check the predictions made by TD-
DFT, ZINDO calculations have been performed on
2, based on the DFT-optimized geometry. Results of
the calculations are presented in Table 2 and Kohn-
Sham orbitals are shown in Fig. 3.
Compound 2 exhibits its main absorption feature
with a maximum at 4.16 eV (298 nm), which is as-
signed to excitation into S3 dominated by a combi-
nation of HOMO ! LUMO þ 1 and HOMOꢁ1 !
LUMO transitions (cf. Fig. 3); The shoulder found
experimentally around 3.8 eV (see Fig. 2) is attribut-
ed to an excitation into S2 (containing also predom-
inantly HOMO ! LUMO þ 1 and HOMOꢁ1 !
LUMO contributions), for which a significantly
smaller oscillator strength is obtained. The two occu-
pied orbitals are delocalized over the whole molecule
with stronger contributions on the carbazole. The
LUMO is fully localized there and the LUMO þ 1
and LUMO þ 2 have their dominant contributions
on the phenyls, which will later turn out to be cru-
cial, when discussing substitution effects. For both
features, the calculations overestimate the transition
energy by about 0.2 eV.
This is also the case for S0 ! S1, which we ten-
tatively associate with the experimental features
around 3.5 eV (assigned to an electronic excitation
and its vibronic replica). This would mean a seri-
ous underestimation of the corresponding oscillator
strength in the simulations, which we do observe both
in the ZINDO as well as in the TD-DFT calculations;
To determine the photophysical properties of the
compounds under investigation, UV-Vis absorption
and photoluminescence (PL) measurements were
performed in diluted solutions of CHCl3 (in case of
2–4) and THF (in the case of 5) at room temperature
under ambient conditions. Results are presented in

226
G. Kremser et al.
the small oscillator strength associated with excita-
tion into S1 is insofar surprising as based on the sym-
metry of the involved orbitals, a pure HOMO !
LUMO transition (which is the most dominant con-
tribution for this excitation) is allowed. Therefore,
the small oscillator strength has to be related to
the influence of other contributions, which is more
obvious for the ZINDO-calculations, where the
HOMO ! LUMO (CI-coefficent: 0.54) derived tran-
sition dipole is largely cancelled by other contri-
butions from other excited determinants. Here, for
example, transitions from HOMOꢁ1 to LUMO þ 1
and LUMO þ 5 (CI: 0.29 and 0.20, respectively) as
well as from HOMOꢁ2 to LUMO þ 1 and LUMO þ
Table 1. Absorption (labs) and emission wavelengths (lem) as
well as molar coefficients of extinction (") and luminescence
quantum yields (F) in solution of 2, 3, and 4 (recorded in
CHCl3) and 5 (recorded in THF)
Compound
l
abs=nm
("=10ꢁ3 Mꢁ1 cmꢁ1
lem
nm
=
F=%
3 ꢃ 2
)
2
298 (42.6), 350 (4.1),
364 (2.6)
395
3
4
315 (33.7), 362 (32.1)
298 (15.9), 363 (25.0),
390 (23.2)
450
585
94 ꢃ 5
26 ꢃ 10
5
301 (41.8), 347 (3.3),
363 (2.1)
397
3 ꢃ 2
Fig. 2. Absorption and emission spectra of 2, 3, 4, and 5
recorded in solution
Fig. 3. Kohn-Sham orbitals of 2, 3, and 4 and their absolute energies in eV

Synthesis and Photophysical Properties of 3,6-Diphenyl-9-hexyl-9H-carbazole
227
7 (CI: ꢁ0.17 and 0.12, respectively) are also charac-
terized by relatively large coefficients [25].
LUMO þ 1 and LUMO þ 2 of 2 are much more af-
fected by the substitution. The former is stabilized
by 1.2 eVand the latter even by nearly 1.4 eV render-
ing the corresponding orbitals the new LUMO and
LUMO þ 1 in 3. This is fully consistent with the
strong contributions of these on the phenyl substitu-
ents. In fact, the substitution results in a stronger lo-
calization of these two unoccupied orbitals in the
vicinity of the acceptor groups. Also the red-shift in
the absorption spectra and the fact that the HOMO !
LUMO dominated excitation becomes strongly op-
tically allowed in 3 (in contrast to 2) is simply a
consequence of the different nature and strong stabi-
lization of the LUMO as a result of the substitution.
The fluorescence spectra recorded for 2–5 are
shown in Fig. 2. The spectral position of the emis-
sion maximum is distinctly dependent on the sub-
stituent of the phenyl groups. While the parent 2
exhibited its emission maximum at 395 nm (3.14 eV),
a red-shift towards 450 nm (2.76 eV) in case of 3 and
towards 585 nm (2.12 eV) in case of 4 was observed
(It is well known that the nitro group is a much
stronger acceptor than an aldehyde. A convenient
measure is for example the empirically determined
Hammett-parameter, which is considerable larger for
NO2 (ꢁp ¼ 0.78) than for CHO (ꢁp ¼ 0.42) [26]).
The UV-Vis spectra of 3 and 4 differ significantly
from 2, while that of 5 is very similar. The latter is
not surprising, considering that the conjugation be-
tween the ꢀ-electron system and the OH groups is
broken by the CH2 linkers. The spectral signatures of
3 and 4, however, deserve further attention.
Compound 3 exhibits two clearly separated ab-
sorption maxima with similar extinction coefficients
at 3.94 eV (315 nm) and 3.43 eV (362 nm), which are
both shifted to lower energies compared to the main
peak in 2. This trend is fully reproduced by the cal-
culations, where the lowest energy peak is associat-
ed with an excitation into S1 (largely HOMO !
LUMO), while the higher peak is a superposition of
absorption into S6 (largely HOMO ꢁ 1 ! LUMO þ
1) and S7 (HOMO ꢁ 1 ! LUMO and HOMO !
LUMO þ 2).
The effect of the acceptor substituents is best seen
in the molecular orbitals in Fig. 3. Both the shapes as
well as the energies of the occupied orbitals are only
weakly affected by the formyl group (stabilization
by about 0.4 eV). The same is true for the LUMO
of 2, which is not surprising considering its local-
ization on the carbazole. In contrast to that, the
Table 2. B3LYP=SV(P) and ZINDO calculated S0 ! Sn transition energies (E), corresponding wavelengths (labs,calc), and
oscillator strengths (f) for the most relevant excited states in 2, 3, and 4 in their ground-state equilibrium conformations.
The last column lists the dominant contributions to the excited states as computed by TD-DFT
Compound
State
labs,calc (E)=nm (eV)
f
Contributions (c2 ꢄ 100=%)
2
S1
S2
S3
S1
332 (3.73)
310 (4.00)
284 (4.37)
332 (3.73)
0.01
0.21
0.85
0.03
H ! L (90.7)
H ! L þ 1 (80.9), H-1 ! L (17.6)
H-1 ! L (64.3), H ! L þ 1 (15.4)
H-2 ! L þ 1 (5.5), H-1 ! L þ 1 (16.3), H-1 ! L þ 5 (8.2),
H ! L (58.9)
2 (ZINDO)
S2
S3
S1
S2
S6
S7
S8
S1
S2
S4
S7
S9
3.02 (4.10)
291 (4.26)
368 (3.37)
346 (3.58)
317 (3.91)
311 (3.99)
283 (4.38)
398 (3.11)
380 (3.27)
342 (3.62)
332 (3.85)
289 (4.29)
0.15
0.82
0.52
0.12
0.25
0.28
0.28
0.42
0.12
0.13
0.13
0.20
H-6 ! L (5.5), H-1 ! L (46.7), H-1 ! L þ 2 (7.7), H ! L þ 7 (9.4)
H-1 ! L þ 2 (7.6), H ! L þ 1 (69.7)
H ! L (98.1)
3
H ! L þ 1 (94.4)
H-1 ! L þ 1 (92.4)
H-1 ! L (61.1), H ! L þ 2 (33.0)
H-1 ! L þ 2 (75.7), H ! L þ 5 (10.5)
H ! L (98.9)
4
H ! L þ 1 (98.6)
H-1 ! L þ 1 (97.4)
H ! L þ 2 (71.1), H-1 ! L (15.6)
H-2 ! L þ 1 (35.8), H-1 ! L þ 2 (25.8), H-10 ! L þ 1(12.2),
H-9 ! L (11.9)
S15
267 (4.46)
0.14
H ! L þ 3 (42.5), H-3 ! L (15.5), H ! L þ 4 (11.4),
H-4 ! L þ 1 (11.0)

228
G. Kremser et al.
Table 3. B3LYP=SV (P) calculated lowest energies (E), emis-
sion wavelength (lem,calc), and oscillator strength (f) for the
lowest lying excited state of 2, 3, and 4 in their TD-DFT
optimized S1 equilibrium geometries
was fabricated in a multilayer architecture on an
indium-tin-oxide (ITO) covered glass substrate act-
ing as a transparent anode followed by a spin-coated
poly(3,4-ethylenedioxythiophene):poly(styrenesul-
fonate) (PEDOT:PSS) film. Compound 3 was spin-
coated as emissive layer and finally a Ca electrode
capped with an Al layer was deposited.
Compound
lem,calc=nm (E=eV)
f
2
3
4
358 (3.47)
411 (3.01)
468 (2.65)
0.01
0.39
0.14
The obtained electroluminescence spectra are
shown in the inset of Fig. 4. As representative exam-
ples, the data for operation at 8.0, 10.0, 13.0, and
16.0 V are shown. At these operating voltages, an
EL maximum at 465 nm was observed, resulting
in a pure and bright blue emission of the device.
Compared to the PL spectrum of 3, the EL maxi-
mum is slightly red-shifted (450 vs. 465 nm). The
luminance-voltage characteristic of the device ITO=
PEDOT:PSS=3=Ca=Al is also given in Fig. 4. The
onset voltage was approximately 7.5 V and lumi-
nance intensities of ꢅ60 cd mꢁ2 were obtained at a
driving voltage of 11 V. Although this confirms the
applicability of 3 as a blue-light emitting material in
OLEDs, it should be noted that the herein presented
data were obtained from a non-optimized device
structure.
Photoluminescence quantum yields for all studied
compounds were determined and results are pre-
sented in Table 1. While 2, 4, and 5 exhibited low
to medium quantum yields, 3 offered bright blue
emission with 94 ꢃ 5% quantum yield. This can be
well correlated with the computed oscillator strengths
of the first vertical transition of the S1 optimized
geometries (see Table 3). It is also a striking feature
of the spectra that the Stokes shift of 4 is larger than
that of 3 by 61% (1.06 and 0.65 eV). This is qualita-
tively reproduced by our calculations, which predict a
77% larger change of the transition energies for 4 than
for 3 when comparing the S0 ! S1 transitions for S0
and S1 equilibrium geometries. Interestingly, for the S1
geometry of 4 a breaking of the pseudo-C2 symmetry
is observed. The absolute values of the Stokes shifts
are, however, severely underestimated (0.36 eV for 3
and 0.46 eV for 4).
Conclusion
The preparation of carbazoles bearing 4-formyl and
4-nitro-substituted phenyl rings was achieved by
Suzuki cross-coupling reactions of 3,6-dibromo-9-
hexyl-9H-carbazole with the corresponding boronic
acid derivatives. Resulting EWG bearing 3,6-diphe-
nyl-9-hexyl-9H-carbazole derivatives exhibited pro-
nounced red shifts of their absorption and emission
properties when compared to the parent compound
3,6-diphenyl-9-hexyl-9H-carbazole. The electronic
effects of the EWGs on the photophysical properties
of the compounds were described using quantum
mechanical calculations. 3,6-Di(4-formylphenyl)-
9-hexyl-9H-carbazole showed particularly interest-
ing photophysical properties, exhibiting an emission
maximum peaking at 450 nm and a luminescence
quantum yield of 95% in solution. The capability
of this derivative to act as the emissive layer in
OLEDs was confirmed.
The electroluminescent (EL) properties of 3 were
investigated because of the high PL quantum yield
and the favourable blue emission maximum of this
compound. For this purpose, an EL device using 3
Experimental
Fig. 4. Bias=electroluminescence characteristics and nor-
malized electroluminescence spectra at different voltages
for OLEDs prepared from 3
Unless otherwise noted, materials were obtained from com-
mercial sources (Aldrich, Fluka or Lancaster) and were used

Synthesis and Photophysical Properties of 3,6-Diphenyl-9-hexyl-9H-carbazole
229
without further purification. 3,6-Dibromo-9-hexyl-9H-carba-
zole (1) was prepared according to the publication of Liu
et al. [19]. Solvents for reactions were freshly distilled over
appropriate drying agents prior to use. Reactions were carried
out under inert atmosphere of Ar using standard Schlenk tech-
TURBOMOLE 5.7 package [30]. B3LYP has been reported
to correctly reproduce the ordering of states [31], and was
recently very successful in explaining the optical properties
of various classes of molecules [32], even when small basis
sets were used. For geometry optimizations, no symmetry
constraints were employed in order to prevent any bias of
the results. For all converged geometries, a vibrational analy-
sis was performed (i) to ensure that a minimum on the respec-
tive potential energy surface has been reached and (ii) to
assign the measured IR spectra. TD-DFT results were sup-
ported using Zerner’s Intermediate Neglect of Differential
Overlap (ZINDO) method as implemented in the Gaussian03
package [33], with a CI active space of 25 occupied and 25
unoccupied orbitals. The transition and Kohn-Sham orbital
energies are all given for the ground state equilibrium geome-
tries for absorption and for S1-excited state optimized geome-
tries for emission.
1
niques. H NMR spectra were recorded on a Varian INOVA
1
500 MHz Spectrometer at 500 MHz, 13Cf Hg NMR spectra
were recorded at 125 MHz. Assignment of the peaks was done
by DEPT and COSY NMR spectroscopy. Solvent residual
peaks were used for referencing the NMR-spectra to the
corresponding values given in literature [27]. FT-IR spectra
were obtained from films on NaCl windows with a Perkin
Elmer Spectrum One and a DTGS-detector. Elemental analy-
ses (C and H) were conducted for the novel compounds 2–5
with results that were found to be in good agreement (ꢃ0.3%)
with the calculated values. UV-Vis absorption spectra were
recorded on a Cary 50 Bio UV-Vis Spectrophotometer, fluores-
cence spectra on a Perkin Elmer Luminescence Spectrometer
LS50B. PL quantum yields were measured using a Shimadzu
RF-5301PC Spectrofluorimeter (corrected for its spectral re-
sponse) and Perkin Elmer Lambda 9 UV=VIS=NIR Spectro-
photometer using quinine sulfate dihydrate in 0.1 N H2SO4 as
standard [28]. Combined differential scanning calorimetry=
thermogravimetric analysis measurements were performed
with a Polymer Laboratories simultaneous thermal analyzer
STA 625. The experiments were carried out in aluminum pans
with a heat rate of 10ꢂC=min in a nitrogen flow of approxi-
mately 35 cm3=min. Organic light emitting devices (OLEDs)
were fabricated on glass substrates covered with transpar-
ent indium-tin-oxide (ITO), acting as anode in a common-
ly used sandwich structure (ITO=PEDOT:PSS=3=Ca=Al).
First, the samples were cleaned in an ultrasonic bath for
ten minutes in isopropanol followed by 10 min in toluene.
Subsequently, they were etched for 10 min in oxygen
plasma. Then poly(3,4-ethylenedioxythiophene):poly(styrene-
sulfonate) (PEDOT:PSS) was spin cast under ambient con-
ditions (approximate thickness 50 nm). The samples were
annealed in a vacuum oven at a temperature of 150ꢂC for at
least 60 min, followed by cooling to room temperature over
40min. 3 was spin-coated from CHCl3 solutions with a con-
centration of 8 mg=cm3. Calcium top electrodes and an alumi-
num capping layer were evaporated in an evaporation chamber
located in an argon glove box. Devices were tested under inert
atmosphere of argon. Device characteristics were recorded
using a customized setup. While current=voltage characteris-
tics were measured using a Keithley 236 Source Measure
Unit, a photodiode mounted in an integrating sphere and con-
nected to a Keithley 181 Nanovoltmeter was used to measure
the emission intensity. To obtain luminance values (for de-
vices operated in cw-mode), the set-up was calibrated using
a Luminance Meter LS100 (Minolta) near the maximum emis-
sion intensity. Electroluminescence spectra were obtained with
a DB401-UV CCD detector (Andor) mounted on an Oriel
Multispec grid spectrometer. The spectral sensitivity was cali-
brated with an Ocean Optics LS-1-CAL tungsten-halogen
lamp. Calculations have been performed using (time de-
pendent)-density functional theory. The exchange-correlation
hybrid functional B3LYP [29] together with a SV(P) double
valence basis set was employed as implemented in the
3,6-Diphenyl-9-hexyl-9H-carbazole (2, C30H29N)
3,6-Dibromo-9-hexyl-9H-carbazole (100.4mg, 0.25 mmol),
105 mg phenylboronic acid (0.86 mmol), and 136 mg Na2CO3
(1.28 mmol) were dissolved in a mixture of 4 cm3 degassed
toluene, 0.5 cm3 EtOH and 1 cm3 H2O. After heating the
reaction mixture to 90ꢂC, 20mg Pd(PPh3)4 (0.02 mmol,
8 mol%) was added and the mixture was stirred for 20 h. Water
(10 cm3) and 10 cm3 Et2O were subsequently added. The wa-
ter layer was separated and extracted with 3ꢆ10cm3 Et2O.
The combined organic layers were dried over Na2SO4 and the
solvent was evaporated under reduced pressure. The residue
was purified by column chromatography on silica (cy:ee¼
10:1) sampling the band at Rf ¼ 0.86 (cy:ee¼ 3:1). Subsequent
washing of the residue with hot n-pentane afforded a white
microcrystalline powder, which was dried in vacuum. Yield:
61.7mg (63%); mp 123ꢂC; 1H NMR (500MHz, CDCl3):
3
ꢂ ¼ 8.38 (s, 2H, carb4,5), 7.75–7.74 (d, 4H, JHH ¼ 7.3 Hz,
3
ph2,6), 7.61–7.60 (d, 2H, JHH ¼ 7.6 Hz, carb2,7), 7.51–7.48
(m, 6H, ph3,5, carb1,8), 7.37–7.35 (m, 2H, ph4), 4.37–4.34
(t, 2H, hex1), 1.94–1.91 (p, 2H, hex2), 1.37–1.30 (m, 6H,
hex3,4,5), 0.90–0.88 (t, 3H, hex6) ppm; 13C NMR (125 MHz,
CDCl3): ꢂ ¼ 142.2, 140.5 (4C, carb3,6, ph1), 132.5 (2C,
carb8a,9a), 128.9 (4C, ph3,5), 127.4 (4C, ph2,6), 126.6 (2C,
ph4), 125.5 (2C, carb4,5), 123.6 (2C, carb4a,4b), 119.1
(2C, carb2,7), 109.2 (2C, carb1,8), 43.5 (1C, hex1), 31.8 (1C,
hex4), 29.2 (1C, hex2), 27.2 (1C, hex3), 22.7 (1C, hex5), 14.2
(1C, hex6) ppm; IR (Film on KBr): ꢃꢀ¼ 2954 (w), 2927 (m),
1600 (m), 1477 (s), 1436 (w), 1352 (w), 1283 (w), 1267 (w),
1155 (w), 809 (w), 762 (s), 697 (m) cmꢁ1; UV-Vis (CHCl3):
labs ¼ 298 nm (" ¼ 42600Mꢁ1 cmꢁ1); fluorescence (CHCl3):
lem ¼ 395 nm.
3,6-Di-(4-formylphenyl)-9-hexyl-9H-carbazole
(3, C32H29NO2)
3,6-Dibromo-9-hexyl-9H-carbazole (300.5mg, 0.73 mmol),
384 mg 4-formylphenylboronic acid (2.56 mmol) and 390.9 mg
Na2CO3 (3.69 mmol) were dissolved in a mixture of 8 cm3
degassed toluene, 1 cm3 EtOH and 2 cm3 H2O. After heating
the reaction mixture to 90ꢂC, 60 mg Pd(PPh3)4 (0.05 mmol,
7 mol%) was added and the mixture was stirred for 20 h. H2O

230
G. Kremser et al.
3,6-Di-(4-hydroxymethylphenyl)-9-hexyl-9H-carbazole
(20 cm3) and 20 cm3 Et2O were subsequently added. The wa-
ter layer was separated and extracted with 3ꢆ30cm3 Et2O.
After drying the combined organic layers over Na2SO4 and
evaporation of the solvent the resulting residue was purified
by column chromatography on silica (cy:ee¼ 8:1) sampling
the band at Rf ¼ 0.24 (cy:ee¼ 5:1). The product was further
purified by washing with hot n-pentane six times to give a
(5, C32H33NO2)
Compound 3 (50.2mg, 0.11mmol) was dissolved in 3 cm3
absolute THF and cooled to 0ꢂC under inert atmosphere of
argon. LiAlH4 (0.436cm3, 0.44mmol, 1.0 M in THF) was
added drop wise by which the reaction mixture turned cloudy.
The reaction mixture was allowed to warm to room tempera-
ture and was stirred for further 4 h. After cooling to 0ꢂC,
excess LiAlH4 was hydrolyzed by careful, drop wise addition
of 10% HCl (0.5cm3). Extraction with Et2O (3ꢆ) from a
saturated NaHCO3 solution and drying over Na2SO4 gave a
pale yellow solid after removing the solvent under reduced
pressure. The product was purified by column chromato-
graphy on silica (cy:ee¼ 1:1) sampling band at Rf ¼ 0.15
(cy:ee¼ 1:1) and washing with hot n-pentane. Yield: 30.8mg
(61.0%); mp not observed, decomposition starting at 50ꢂC;
1
yellow solid. Yield: 103.9 mg (30.9%); mp 169ꢂC; H NMR
(500 MHz, CDCl3): ꢂ ¼ 10.08 (s, 2H, CHO), 8.45 (s, 2H,
carb4,5), 8.00–7.99 (d, 4H, 3JHH ¼ 8.0 Hz, ph3,5), 7.91–7.90
(d, 4H, JHH ¼ 7.8 Hz, ph2,6), 7.82–7.80 (d, 2H, JHH
¼
3
3
8.5 Hz, carb2,7), 7.54–7.52 (d, 2H, JHH ¼ 8.5 Hz, carb1,8),
4.39–4.36 (t, 2H, hex1), 1.95–1.92 (p, 2H, hex2), 1.37–1.30
(m, 6H, hex3,4,5), 0.90–0.87 (t, 3H, hex6) ppm; 13C NMR
(125 MHz, CDCl3): ꢂ ¼ 192.1 (2C, CHO), 148.1 (2C, ph1),
141.2 (2C, carb3,6), 134.7, 131.1 (4C, carb8a,9a, ph4), 130.5
(4C, ph3,5), 127.7 (4C, ph2,6), 125.8 (2C, carb4,5), 123.7
(2C, carb4a,4b), 119.5 (2C, carb2,7), 109.7 (2C, carb1,8),
43.6 (1C, hex1), 31.7 (1C, hex4), 29.2 (1C, hex2), 27.1
(1C, hex3), 22.7 (1C, hex5), 14.2 (1C, hex6) ppm; IR (Film
on KBr): ꢃꢀ¼ 2926 (m), 2854 (w), 2731 (w), 1697 (s), 1599
(s), 1563 (w), 1482 (s), 1390 (w), 1353 (w), 1308 (m), 1273
3
1H NMR (500MHz, CDCl3): ꢂ ¼ 8.70 (s, 2H, carb4,5),
3
7.93–7.88 (m, 6H, carb2,7, ph2,6), 7.79–7.78 (d, 2H, JHH
¼
5.6 Hz, carb1,8), 7.60 (d, 4H, JHH ¼ 2.4 Hz, ph3,5), 4.82 (s,
4H, CH2OH), 4.62–4.61 (t, 2H, hex1), 1.94–1.91 (p, 2H,
hex2), 1.37–1.30 (m, 6H, hex3,4,5), 0.90–0.88 (t, 3H, hex6)
ppm; 13C NMR (125MHz, CDCl3): ꢂ ¼ 142.5, 142.3, 142.2
(6C, ph1, ph4, carb3,6), 133.8 (2 C, carb8a,9a), 128.9 (4C,
ph3,5), 128.4 (4C, ph2,6), 126.8 (2C, carb4,5), 125.4 (2C, car-
3
(m), 1215 (m), 1171 (s), 837 (w), 804 (m), 719 (w) cmꢁ1
;
UV-Vis (CHCl3): labs ¼ 362 nm (" ¼ 32100 Mꢁ1 cmꢁ1);
b
4a,4b), 120.4 (2C, carb2,7), 111.3 (2C, carb1,8), 65.4 (2C,
fluorescence (CHCl3): lem ¼ 450 nm.
CH2OH), 44.6 (1C, hex1), 33.3 (1C, hex4), 30.9 (1C, hex2),
28.4 (1C, hex3), 24.2 (1C, hex5), 15.2 (1C, hex6); UV-Vis
(THF): labs ¼ 301 nm (" ¼ 41800 Mꢁ1 cmꢁ1); fluorescence
(THF): lem ¼ 397 nm.
3,6-Di-(4-nitrophenyl)-9-hexyl-9H-carbazole
(4, C30H27N3O4)
To a mixture of 3,6-dibromo-9-hexyl-9H-carbazole (200.4mg,
0.49 mmol), 4-nitrophenylboronic acid pinacol ester (425.9mg,
1.71mmol) and Na2CO3 (260mg, 2.44mmol) was added a
mixture of 8 cm3 degassed toluene, 1 cm3 EtOH, and 2 cm3
H2O. After heating the reaction mixture to 90ꢂC, 40mg
Pd(PPh3)4 (0.03 mmol, 6 mol%) was added and the mixture
was stirred for 20h. H2O (15 cm3) and 15 cm3 Et2O were sub-
sequently added. The water layer was separated and extracted
with 3ꢆ20cm3 Et2O. The combined organic layers were
dried over Na2SO4 and the solvent was removed under re-
duced pressure. The residue was purified by column chroma-
tography on silica (cy:ee¼ 15:1 to 10:1) sampling the band at
Rf ¼ 0.57 (cy:ee¼ 3:1). Recrystallization from CH2Cl2 and
methanol gave an orange solid. Yield: 106.9mg (44.3%);
mp 234ꢂC; 1H NMR (500MHz, CDCl3): ꢂ ¼ 8.43 (s, 2H,
(d, 4H, 3JHH ¼ 8.8 Hz, ph2,6), 7.81–7.79 (d, 2H, 3JHH ¼ 8.5 Hz,
carb2,7), 7.56–7.54 (d, 2H, 3JHH ¼ 8.5 Hz, carb1,8), 4.40–4.37
(t, 2H, hex1), 1.95–1.92 (p, 2H, hex2), 1.37–1.30 (m, 6H,
hex3,4,5), 0.90–0.87 (t, 3H, hex6) ppm; 13C NMR (125 MHz,
CDCl3): ꢂ ¼ 148.5, 146.7 (4C, ph4, ph1), 141.5 (2C, carb3,6),
130.4 (2C, carb8a,9a), 127.8 (4C, ph2,6), 126.0 (2C, carb4,5),
124.5 (4C, ph3,5), 123.7 (2C, carb4a,4b), 119.8 (2C, carb2,7),
110.0 (2C, carb1,8), 43.8 (1C, hex1), 31.8 (1C, hex4), 29.2 (1C,
hex2), 27.2 (1C, hex3), 22.8 (1C, hex5), 14.3 (1C, hex6) ppm;
IR (Film on KBr): ꢃꢀ¼ 2923 (s), 2853 (m), 1588 (s), 1513 (s),
1477 (m), 1380 (w), 1337 (s), 1306 (w), 1275 (w), 1246 (w),
Acknowledgements
Financial support by the Austrian Science Fund in the frame-
work of the Austrian Nano Initiative RPC ISOTEC – RP 0701
and 0702 is gratefully acknowledged.
References
[1] Strohriegl P, Grazulevicius JV, Pielichowski J,
Pielichowski K (2003) Prog Polym Sci 28: 1297
[2] Joule JA (1984) Adv Heterocycl Chem 35: 83
[3] a) Adachi C, Baldo MA, Thompson ME, Forrest SR
(2001) J Appl Phys 90: 5048; b) Zaami N, Slugovc C,
Pogantsch A, Stelzer F (2004) Macromol Chem Phys
205: 523; c) Thomas KRJ, Lin JT, Tao Y-T, Ko C-W
(2001) J Am Chem Soc 123: 9404
[4] Ostroverkhova O, Moerner WE (2004) Chem Rev 104:
3267
[5] Dimitrakopoulos CD, Malenfant RL (2002) Adv Mater
14: 99
[6] Law KY (1993) Chem Rev 93: 449
3
carb4,5), 8.34–8.33 (d, 4H, JHH ¼ 8.8 Hz, ph3,5), 7.88–7.86
[7] Halik M, Wenseleers W, Grasso C, Stellacci F, Zojer E,
´
Barlow S, Bredas J-L, Perry JW, Marder SR (2003)
Chem Commun 1490
1185 (w), 1111 (w), 856 (w), 804 (w), 755 (m), 689 (w) cmꢁ1
;
[8] Choi DH, Jun W, Oh KY, Kim JH (2002) Polym Bull
49: 173
[9] Forrest SR (2003) Org Electron 4: 45
UV-Vis (CHCl3): labs ¼ 390 nm (" ¼ 23200 Mꢁ1 cmꢁ1); fluo-
rescence (CHCl3): lem ¼ 585 nm.

Synthesis and Photophysical Properties of 3,6-Diphenyl-9-hexyl-9H-carbazole
231
[10] a) Morin J-F, Leclerc M (2001) Macromolecules 34:
4680; b) Sigwalt P, Wegner G, Morin J-F, Leclerc M,
Ades D, Siove A (2005) Macromol Rapid Commun
26: 761
[11] a) Cadogan JIG (1962) Quart Rev 16: 208; b) Cadogan
JIG, Cameron-Wood M (1962) Proc Chem Soc 361;
c) Freeman AW, Urvoy M, Criswell ME (2005) J Org
Chem 70: 5014
[12] Mu¨llen K, Scherf U (eds) (2006) Organic Light Emitting
Devices. Wiley VCH, Weinheim
[13] Coppo P, Plummer EA, De Cola L (2004) Chem
Commun 1774
[14] Adachi C, Kwong RC, Djurovich P, Adamovich V,
Baldo MA, Thompson ME, Forrest SR (2001) Appl
Phys Lett 79: 2082
[15] Holder E, Langeveld BMW, Schubert US (2005) Adv
Mater 17: 1109
[16] a) Hubber J, Mu¨llen K, Salbeck J, Schenk H, Scherf U,
Stehlin T, Stern R (1994) Acta Polym 45: 244; b)
Lemmer U, Heun S, Mahrt RF, Scherf U, Hopmeier
state description is 90% HOMO ! LUMO, but in this
context it has to be kept in mind that such percentage
values are proportional to the square of the coefficients
with which certain configurations enter the description
of certain excited states, while the transition dipole to a
certain state is linearly proportional to the sum of the
transition dipoles to the individual configurations times
their coefficients. Moreover, these coefficients are not
normalized! I.e., the sum of the norms of the coefficients
is typically larger than 1 and only their squares are
normalized. This means that also in the TD-DFT calcu-
lations other excitations will make appreciable contri-
butions to the transition dipole, which has to be held
responsible for the very small oscillator strength of S1
[26] Hansch C, Leo A, Taft RW (1991) Chem Rev 91:
165
[27] Gottlieb HE, Kotylar V, Nudelman A (1997) J Org Chem
62: 7512
[28] Demas JN, Crosby GA (1971) J Phys Chem 991
[29] a) Dirac PAM (1929) Proc Royal Soc A 123: 714;
b) Slater JC (1951) Phys Rev 81: 385; c) Vosko SH,
Wilk L, Nusair M (1980) Can J Phys 58: 1200; d) Becke
AD (1988) Phys Rev A 38: 3098; e) Lee C, Yang W, Parr
RG (1988) Phys Rev B 37: 785; f) Becke AD (1993)
J Chem Phys 98: 5648
¨
¨
¨
M, Sieger U, Gobel EO, Mullen K, Bassler H (1995)
Chem Phys Lett 240: 373; c) Grell M, Bradley DDC,
Ungar G, Hill J, Whitehead KS (1999) Macromolecules
32: 5810; d) Scherf U, List EJW (2002) Adv Mater
14: 477
¨
¨
¨
[30] Ahlrichs R, Bar M, Haser M, Horn H, Komel C (1989)
[17] a) Marder SR (2006) Chem Commun 131; b) Albota M,
Beljonne D, Bredas J-L, Ehrlich JE, Fu J-Y, Heikal AA,
Hess SE, Kogej T, Levin MD, Marder SR, McCord-
Maughon D, Perry JW, Rockel H, Rumi M,
Subramaniam G, Webb WW, Wu X-Li, Xu C (1998)
Science 281: 1653
Chem Phys Lett 162: 165
[31] Bauernschmitt R, Ahlrichs R (1996) Chem Phys Lett
256: 454
[32] a) Norton J, Houk K (2005) J Am Chem Soc 127: 4162;
b) Zade S, Bendikov M (2007) Chem Eur J 13: 3688; c)
Medina BM, Beljonne D, Egelhaaf H-J, Gierschner J
(2007) J Chem Phys 126: 111101; d) Pogantsch A,
Heimel G, Zojer E (2002) J Chem Phys 117: 12, 5921
[33] Gaussian 03, Revision C.02, Frisch MJ, Trucks GW,
Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR,
Montgomery JA Jr, Vreven T, Kudin KN, Burant JC,
Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci
B, Cossi M, Scalmani G, Rega N, Petersson GA,
Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R,
Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O,
Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross
JB, Adamo C, Jaramillo J, Gomperts R, Stratmann RE,
Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski
JW, Ayala PY, Morokuma K, Voth GA, Salvador P,
Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels
AD, Strain MC, Farkas O, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul
AG, Clifford S, Cioslowski J, Stefanov BB, Liu G,
Liashenko A, Piskorz P, Komaromi I, Martin RL, Fox
DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A,
Challacombe M, Gill PMW, Johnson B, Chen W,
Wong MW, Gonzalez C, Pople JA (2004) Gaussian
Inc., Wallingford, CT
¨
[18] Kappaun S, Sax S, Eder S, Moller KC, Waich K,
Niedermair F, Saf R, Mereiter K, Jacob J, Mu¨llen K,
List EJW, Slugovc C (2007) Chem Mater 19: 1209
[19] Liu X-M, Xu J, Lu X, He C (2006) Macromolecules 39:
1397
[20] Li Y, Ding J, Day M, Tao Y, Lu J, D’iorio M (2004)
Chem Mater 16: 2165
[21] a) Tao B, Boykin DW (2004) J Org Chem 69: 4330;
b) Kappaun S, Zelzer M, Bartl K, Saf R, Stelzer F,
Slugovc C (2006) J Polym Sci Part A Polym Chem
44: 2130; c) Kappaun S, Sovic T, Stelzer F, Pogantsch
A, Zojer E, Slugovc C (2006) Org Biomol Chem 4:
1503; d) Zelzer M, Kappaun S, Zojer E, Slugovc C
(2007) Monatsh Chem 138: 453
[22] Park M, Buck JR, Rizzo CJ (1998) Tetrahedron 54: 12707
[23] All assignments are based on vibration analyses per-
formed during the quantum chemical calculations pre-
sented below
[24] Kim YH, Ahn JH, Shin DC, Kwon SK (2004) Polymer
45: 2525
[25] At a first glance, it might seem surprising that this can
also explain the low oscillator strength in the TD-DFT
calculations, where according to Table 2, the excited