Synthesis and luminescence properties of poly(p-phenylenevinylene) derivatives carrying directly attached carbazole pendants

Article abstract of DOI:10.1039/b102736k

New organic soluble poly(p-phenylenevinylene) (PPV) derivatives (polymers 1 and 2) that carry hole-transporting carbazole pendants were synthesized and their photo- and electro-luminescence properties were studied. The first one is poly[2-(carbazol-9-yl)-

Full text of DOI:10.1039/b102736k

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and luminescence properties of poly(p-phenylenevinylene)
derivatives carrying directly attached carbazole pendants
Kyungkon Kim,a Young-Rae Hong,a Seung-Wuk Lee,a Jung-Il Jin,*a Yongsup Park,b
Byung-Hee Sohn,c Woo-Hong Kimc and Jae-Kun Parkc
aDivision of Chemistry and Molecular Engineering, Center for Photo- and Electro-Responsive
Molecules, Korea University, 5-1, Anam-Dong, Seoul 136-701, Korea.
E-mail: jijin@mail.korea.ac.kr
bSurface Analysis Group, Korea Research Institute of Standards and Science, Yosung,
305-600 Taejon, Korea
cPolymer Lab, Samsung Advanced Institute of Technology, Taejon 305-380, Korea
Received 26th March 2001, Accepted 10th August 2001
First published as an Advance Article on the web 2nd October 2001
New organic soluble poly(p-phenylenevinylene) (PPV) derivatives (polymers 1 and 2) that carry
hole-transporting carbazole pendants were synthesized and their photo- and electro-luminescence properties
were studied. The first one is poly[2-(carbazol-9-yl)-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] (polymer 1)
which has both carbazole and 2-ethylhexyloxy pendant groups. The other one is poly[2-(carbazol-9-yl)-1,4-
phenylenevinylene] (polymer 2) which has only the carbazole pendant group attached to the main chain
phenylene ring. We fabricated single-layer EL devices using indium–tin oxide (ITO) coated glass anodes and
aluminium cathodes and investigated their electrical characteristics. The EL device of polymer 2 emits green
light (lmax~490 nm) with the EL efficiency being lower than that of PPV. But the EL device of polymer 1
emits bright yellow–green light (lmax~530 nm) and its external quantum efficiency was 550 times the efficiency
of polymer 2 and 60 times that of a PPV EL device with the same configuration. Especially, the EL device with
the configuration ITO coated glass/poly(3,4-ethylenedioxy-2,4-thienylene)/polymer 1/Ca/Al showed a low
turn-on electric field of 0.31 MV cm21 and a high photometric efficiency of 4.4 cd A21 with maximum
luminance being 30 390 cd m22 at an electric field of 1.50 MV cm21. Polymer 1 appears to perform better in EL
than MEH-PPV.
moieties attached directly to the phenylene rings in the back-
Introduction
bone of the PPV structure. The first one is poly[2-(carbazol-
9-yl)-5-(2-ethylhexyloxy)-1,4-phenylenevinylene] (polymer 1,
Scheme 1) which has both carbazole and 2-ethylhexyloxy
pendant groups. The other is poly[2-(carbazol-9-yl)-1,4-phenyl-
enevinylene] (polymer 2, Scheme 2) which has only the
carbazole pendant group attached to the main chain. We
previously reported briefly the luminescence properties of
polymer 2.8 The LED device of polymer 2 showed a rather poor
electroluminescence efficiency even when compared with
PPV. Attachment of a good hole-transporter may imbalance
the carrier mobilities even further. PPV itself is known to be a
much better hole-transporter than electron-transporter.23 This
factor could decrease the EL efficiency of the device of polymer
2. Another problem noted is that it possesses a high HOMO
(highest occupied molecular orbital) energy level (6.1 eV),
which requires a high drive voltage for hole injection from the
ITO anode to the emitting polymer layer. Moreover, the
solubility of polymer 2 was found unsatisfactory to form films
of good quality.
Polymer 1 (Scheme 1) carries additional electron-donating
alkoxy groups attached to the phenylene ring. The bulky
alkoxy group could reduce the possibility of formation of
interchain excitons24,25 by increasing the interchain distance
that is claimed to cause a lower EL efficiency26–28 and also
improves polymer solubility giving rise to a higher film quality.
In addition, we expected the electron donating alkoxy group to
reduce the HOMO energy level and lower the drive voltage due
to facilitated hole injection from the cathode.
Electroluminescent devices based on organic polymeric
materials have attracted a considerable amount of attention
due to possible applications in large-area light-emitting
displays.1–9 It is understood that in light emitting diode
(LED) devices electrons and holes are separately injected from
an anode and a cathode, respectively, under a bias voltage into
the light emitting polymer layer, where the injected negative
and positive carriers form excitons.1,10 The excitons can
disappear via various mechanisms; one of them is luminescent
decay or radiative decay. In order to improve the efficiency of
LEDs, there have been many attempts to balance the injection
of carriers from electrodes and also their mobility in the
polymer layer. Tang and Van Slyke11 demonstrated that the
use of a hole-transporting layer (HTL) for facile hole injection
from the anode into the emission layer significantly lowers the
drive voltage and improves the device performance. Poly-
(vinylcarbazole), a polymer which has the carbazole moieties as
pendants, is known to be an excellent hole-transporting
material12 and low molecular weight carbazole derivatives
and carbazole containing polymers have also recently been
used as hole-transporting materials.13–18
However most studies of carbazole polymers have been
limited to the use of the material as a hole-transporting layer or
as host polymer for doping with organic dyes. Recently, some
researchers reported luminescence properties of polymers that
have carbazole moieties in the main chain or as side
pendants.19–22
We have synthesized two polymers carrying carbazole
DOI: 10.1039/b102736k
J. Mater. Chem., 2001, 11, 3023–3030
This journal is # The Royal Society of Chemistry 2001
3023

View Article Online
Scheme 1 Synthetic route to monomer 1 and polymer 1.
5.15. Found: C, 50.28; H, 5.19%. 1H-NMR (acetone-d6;
d/ppm): 2.1 (s, 3H, -CH3), 2.3 (s, 3H, -CH3), 3.8 (s, 3H,
CH3O-), 6.9 (s, 1H, Ar-H), 7.2 (s, 1H, Ar-H). IR (KBr, cm21):
3051 (aromatic C–H stretch), 2957 (aliphatic C–H stretch),
1602, 1480 (Ar CLC stretch), 1205, 1101 (C–O stretch).
Experimental
A. Synthesis
Chemicals and instruments. All the chemicals were obtained
commercially from Aldrich or Junsei. They were of analytical
or higher grade and used without further purification. Solvents
to be used under anhydrous conditions were dried by standard
methods. Infrared spectra were recorded on an FT-IR Bomem
2-Bromo-5-methoxyterephthalic acid, 2. A mixture of 1
(40.6 g; 0.189 mol) and KMnO4 (300 g; 1.90 mol) was refluxed
for 24 h in water (2.5 L) with vigorous stirring. The hot
reaction mixture was filtered through a suction filter and the
cake of hydrated manganese dioxide on the filter was washed
with water. The combined filtrate was acidified cautiously by
adding 35% hydrochloric acid with continual agitation. The
white precipitate was collected on a filter and washed with cold
water (33.7 g; 64.8% yield), mp 275 uC. Anal. Calcd for
C9H7BrO5: C, 39.30; H, 2.57. Found: C, 39.27; H, 2.56%.
1H-NMR (acetone-d6; d/ppm): 3.0 (s, 2H, -COOH), 4.0 (s, 3H,
CH3O-), 7.5 (s, 1H, Ar-H), 8.0 (s, 1H, Ar-H). IR (KBr, cm21):
3350–2400 (O–H stretch of acid), 1700 (CLO stretch), 1600,
1475 (Ar CLC stretch), 1248, 1109 (C–O stretch).
1
Michaelson instrument. H-NMR spectra were recorded on a
Varian Gemini 300 spectrometer.
Synthesis of monomers and polymers. Synthesis of monomers
and polymers are outlined in Schemes 1 and 2. Synthetic details
are described below following the synthetic routes shown in the
schemes.
4-Bromo-2,5-dimethylanisole, 1. To a solution of 2,5-dimethyl-
anisole (60.3 g; 0.443 mol) dissolved in 600 mL of chloroform
was added dropwise at 0 uC a solution of bromine (70.2 g;
0.441 mol) in carbon tetrachloride (200 mL). After 7 h of
stirring at 0 uC, saturated aqueous NaOH solution (300 mL)
was added to the reaction mixture and the organic phase was
separated and washed with distilled water. After the solvent
was removed by evaporation, the impure product was
chromatographed on a silica gel column using methylene
chloride as an eluent to afford the title compound as an oil
(83.1 g; 87.2% yield). Anal. Calcd for C9H11BrO: C, 50.26; H,
2-(Carbazol-9-yl)-5-methoxyterephthalic acid, 3. A mixture
of 2 (10.0 g; 0.0364 mol), carbazole (12.0 g; 0.0718 mol),
K2CO3 (12.4 g; 0.0897 mol) and copper powder (0.301 g) in
N,N-dimethylformamide (300 mL) was refluxed for 48 h.29 The
reaction mixture was cooled and the solvent was evaporated.
Crude product was extracted with water followed by washing
3024
J. Mater. Chem., 2001, 11, 3023–3030

View Article Online
0.018 mol).31 The mixture was stirred for 4 h at 0 uC. The
mixture was then poured into a large excess of distilled water.
The organic phase was washed with distilled water. The organic
layer was separated and the solvent was evaporated. Chroma-
tography (silica gel, CH2Cl2 eluent) of the residue afforded the
title compound as a yellow solid (2.9 g; 87% yield), mp
80 uC. Anal. Calcd for C22H17NO5: C, 70.39; H, 4.56; N, 3.73.
Found: C, 70.31; H, 4.51; N, 3.71%. 1H-NMR (acetone-d6;
d/ppm): 3.2 (s, 3H, -CO2CH3), 3.9 (s, 3H, -CO2CH3), 7.1 (d,
2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H, Ar-H), 7.6 (s, 1H, Ar-
H), 7.9 (s, 1H, Ar-H), 8.2 (d, 2H, Ar-H), 10.7 (s, 1H, Ar-OH).
IR (KBr, cm21): 3210 (O–H stretch), 3071 (aromatic C–H
stretch), 2957 (aliphatic C–H stretch), 1727 (CLO stretch),
1595, 1452 (Ar CLC stretch), 1327 (C–N stretch), 1205, 1101
(C–O stretch).
Dimethyl 2-(carbazol-9-yl)-5-(2-ethylhexyloxy)terephthalate,
6. A mixture of 5 (2.91 g; 0.77561022 mol), K2CO3 (2.20 g;
0.0159 mol) and tetrabutylammonium bromide (0.1 g) in
acetone (300 mL) was refluxed for 24 h. The solid residue
was removed by filtration. Acetone in the filtrate was
evaporated. The product was recrystallized from methanol
(3.60 g; 95.2% yield), mp 138 uC. Anal. Calcd for C30H33NO5:
C, 73.90; H, 6.82; N, 2.87. Found: C, 73.85; H, 6.74; N, 2.81%.
1H-NMR (acetone-d6; d/ppm): 0.9–1.0 (m, 6H, -OCH2CH-
(CH2CH3)CH2CH2CH2CH3), 1.4 (m, 4H, -CH2CH2CH2CH3),
1.5–1.6 (m, 4H, -OCH2CH(CH2CH3)CH2CH2CH2CH3), 1.8
(m, 1H, -OCH2CH-), 3.2 (s, 3H, -CO2CH3), 3.8 (s, 3H,
-CO2CH3), 4.2 (d, 2H, -OCH2CH-), 7.1 (d, 2H, Ar-H), 7.2
(t, 2H, Ar-H), 7.3 (t, 2H, Ar-H), 7.7 (s, 1H, Ar-H), 7.8 (s, 1H,
Ar-H), 8.1 (d, 2H, Ar-H). IR (KBr, cm21): 3061 (aromatic C–H
stretch), 2948 (aliphatic C–H stretch), 1729 (CLO stretch),
1600, 1475 (Ar CLC stretch), 1310 (C–N stretch), 1223, 1101
(C–O stretch).
Scheme 2 Synthetic route to monomer 2 and polymer 2.
1,4-Bis(hydroxymethyl)-5-(2-ethylhexyloxy)-2-(carbazol-9-yl)-
benzene, 7. 32 To
a stirred mixture of LiAlH4 (0.94 g;
with ether. The aqueous phase was acidified cautiously by
adding 35% hydrochloric acid with continual agitation. The
yellow precipitate was collected on a filter and recrystallized
from chloroform (11.0 g; 83.6% yield), mp 298 uC. Anal. Calcd
for C21H15NO5: C, 69.80; H, 4.18; N, 3.88. Found: C, 69.86; H,
4.14; N, 3.83%. 1H-NMR (acetone-d6; d/ppm): 4.1 (s, 3H,
CH3O-), 7.1 (d, 2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H, Ar-H),
7.8 (s, 1H, Ar-H), 7.9 (s, 1H, Ar-H), 8.1 (d, 2H, Ar-H), 11.4 (s,
2H, COOH). IR (KBr, cm21): 3400–2400 (O–H stretch of
acid), 1700 (CLO stretch), 1598, 1460 (Ar CLC stretch), 1310
(C–N stretch), 1234, 1172 (C–O stretch).
0.2561022 mol) in dry tetrahydrofuran (THF)30 (300 mL)
was added dropwise a solution of 6 (4.0 g; 0.8261022 mol) in
dry THF30 (150 mL), and the mixture was refluxed for 4 h. The
reaction mixture was cooled and 0.9 mL of water, 0.9 mL of
15% NaOH and 2.7 mL of water were added successively.
Then, the solid residue was removed by filtration and the
filtrate was dried over anhydrous MgSO4, and the solvent was
evaporated to obtain the product. The product thus obtained
was utilized in the next step without further purification (3.4 g;
97% yield), mp 128 uC. Anal. Calcd for C28H33NO3: C, 77.93;
H, 7.71; N, 3.25. Found: C, 77.92; H, 7.70; N, 3.24%. 1H-NMR
(acetone-d6; d/ppm): 0.9–1.0 (m, 6H, -OCH2CH(CH2CH3)-
CH2CH2CH2CH3), 1.4 (m, 4H, -CH2CH2CH2CH3), 1.5–1.6
(m, 4H, -OCH2CH(CH2CH3)CH2CH2CH2CH3), 1.8 (m, 1H,
-OCH2CH-), 4.0 (d, 2H, -CH2OH), 4.1 (t, 1H, -CH2OH), 4.2
(d, 2H, -CH2OH), 4.2 (t, 1H, -CH2OH), 4.7 (d, 2H, -OCH2CH-),
7.0 (d, 2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H, Ar-H), 7.4 (s,
1H, Ar-H), 7.4 (s, 1H, Ar-H), 8.2 (d, 2H, Ar-H). IR (KBr,
cm21): 3367 (O–H stretch), 3054 (aromatic C–H stretch), 2913
(aliphatic C–H stretch), 1598, 1460 (Ar CLC stretch), 1310
(C–N stretch), 1223, 1014 (C–O stretch).
Dimethyl 2-(carbazol-9-yl)-5-methoxyterephthalate, 4. Com-
pound 3 (4.40 g; 0.0122 mol) was dissolved in 300 mL of
methanol. To this solution was slowly added purified thionyl
chloride30 (17.5 mL) with stirring over a period of 1 h. The
mixture was stirred for 12 h at room temperature. Then the
reaction mixture was poured into a large excess of distilled water.
To the slurry was added a 10% aqueous solution of NaHCO3
until the mixture became neutral. The precipitate formed was
collected on a filter and washed with distilled water. The
product was recrystallized from methanol (4.39 g; 92.4% yield),
mp 145 uC. Anal. Calcd for C23H19NO5: C, 70.94; H, 4.92; N,
1
3.60. Found: C, 70.96; H, 4.91; N, 3.58%. H-NMR (acetone-
1,4-Bis(chloromethyl)-5-(2-ethylhexyloxy)-2-(carbazol-9-yl)-
benzene, monomer 1. To a solution of 7 (1.2 g; 0.2961022 mol)
in dry chloroform30 (250 mL) was added with stirring a
solution of dimethylformamide (0.15 mL; 0.1961022 mol) and
phosphoryl chloride30 (0.18 mL; 0.1961022 mol) in dry
chloroform30 (150 mL) and stirring was continued for 24 h at
room temperature. The solvent was evaporated. The residue
was stirred with cold water and ether. The ethereal layer was
separated and was dried over calcium chloride. Ether was
removed by evaporation from the solution leaving behind the
crude product. Chromatography (silica gel, hexane) of the
d6; d/ppm): 3.2 (s, 3H, -CO2CH3), 3.8 (s, 3H, -CO2CH3), 4.0 (s,
3H, CH3O-), 7.1 (d, 2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H,
Ar-H), 7.7 (s, 1H, Ar-H), 7.9 (s, 1H, Ar-H), 8.2 (d, 2H, Ar-H).
IR (KBr, cm21): 3050 (aromatic C–H stretch), 2948 (aliphatic
C–H stretch), 1727 (CLO stretch), 1600, 1470 (Ar CLC stretch),
1300 (C–N stretch), 1223, 1106 (C–O stretch).
Dimethyl 2-(carbazol-9-yl)-5-hydroxyterephthalate, 5. Com-
pound 4 was dissolved in 300 mL of CH2Cl2. To this solution
was introduced dropwise at 0 uC 1 M BBr3 (17.8 mL;
J. Mater. Chem., 2001, 11, 3023–3030
3025

View Article Online
residue afforded the title compound as an oil (1.0 g; 76% yield).
Anal. Calcd for C28H31Cl2NO: C, 71.79; H, 6.67; N, 2.29.
Found: C, 71.78; H, 6.65; N, 2.27%. 1H-NMR (acetone-d6;
d/ppm):0.8–1.0(m,6H,-OCH2CH(CH2CH3)CH2CH2CH2CH3),
1.4 (m, 4H, -CH2CH2CH2CH3), 1.5–1.7 (m, 4H, -OCH2CH-
(CH2CH3)CH2CH2CH2CH3), 1.8 (m, 1H, -OCH2CH-), 4.2 (d,
2H, -OCH2CH-), 4.4 (s, 2H, CH2Cl), 4.8 (s, 2H, -CH2Cl), 7.0
(d, 2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H, Ar-H), 7.4 (s, 1H,
Ar-H), 7.5 (s, 1H, Ar-H), 8.2 (d, 2H, Ar-H). IR (KBr, cm21):
3050 (aromatic C–H stretch), 2934 (aliphatic C–H stretch),
1596, 1452 (Ar CLC stretch), 1325 (C–N stretch), 1234, 1185
(C–O stretch), 646 (C–Cl stretch).
2-(Carbazol-9-yl)-1,4-bis(chloromethyl)benzene, monomer 2.
To a solution of 9 (2.0 g; 0.6561022 mol) in DMF (150 mL)
30
was added dropwise at 15 uC purified SOCl2
(3.1 g;
0.026 mol). After 2 h of stirring at 10–15 uC, DMF and
excess SOCl2 were distilled out via vacuum distillation. Chro-
matography (silica gel, hexane–CH2Cl2~1 : 1 by volume) of
the residue afforded the title compound as an oil (2.1 g; 95%
yield). Anal. Calcd for C20H15Cl2N: C, 70.60; H, 4.44; N, 4.12.
Found: C, 70.11; H, 4.34; N, 4.05%. 1H-NMR (acetone-d6;
d/ppm): 4.4 (s, 2H, -CH2Cl), 4.9 (s, 2H, -CH2Cl), 7.1 (d, 2H,
Ar-H), 7.3 (t, 2H, Ar-H), 7.4 (t, 2H, Ar-H), 7.6 (s, 1H, Ar-H),
7.8 (d, 1H, Ar-H), 7.9 (d, 1H, Ar-H), 8.3 (d, 2H, Ar-H). IR
(KBr, cm21): 3050 (aromatic C–H stretch), 2950 (aliphatic
C–H stretch), 1590, 1447 (Ar CLC stretch), 1310 (C–N stretch),
680 (C–Cl stretch).
Poly[2-(carbazol-9-yl)-5-(2-ethylhexyloxy)-1,4-phenylenevinyl-
ene], polymer 1. To a solution of monomer 1 (1.0 g; 0.216
1022 mol) in purified THF30 (70 mL) was added dropwise at
0 uC a solution of 1 M potassium tert-butoxide in THF (12 mL)
via a syringe over a period of 10 minutes.33 The mixture was
stirred for 6 h at room temperature. The colorless solution
turned orange and became viscous. After reacting for an
additional 18 h at room temperature, the polymer formed was
precipitated by slow addition of methanol to the reaction
mixture. Subsequent dissolving in THF and precipitating in
methanol three times yielded an orange colored polymer. The
product was subjected to Soxhlet extraction for 24 h using
methanol (3.5 g; 44% yield). The molecular weight of this
polymer determined by GPC measurement with polystyrene as
Poly[2-(carbazol-9-yl)-1,4-phenylenevinylene], polymer 2.
The polymerization method for polymer 2 was the same as
polymer 1. Anal. Calcd for C20H13N: C, 89.86; H, 4.90; N, 5.24.
Found: 89.87; H, 4.89; N, 5.23%. IR (KBr, cm21): 3050
(aromatic C–H stretch), 2945 (aliphatic C–H stretch), 1598,
1476 (Ar CLC stretch), 1311 (C–N stretch).
B. Device fabrication and characterization
¯
¯
the calibration standard was Mw~72 000 and Mw~49 800 with
a polydispersity index (PDI) of 1.4.
ITO/polymer 1 or 2/Al and ITO/polymer 1 or 2/Li : Al
devices. Indium–tin oxide (ITO) coated glass slides with a sheet
resistance of 25 V cm22 were patterned by the vapor of a mixed
solution of HNO3 and HCl in a volume ratio of 3 : 1. The
patterned ITO-coated glass slides were cleaned by sequential
ultrasonication in acetone, methanol and propan-2-ol for
20 minutes, then dried in a stream of nitrogen.
The solution (1 wt%) of polymer 1 or 2 in purified 1,1,2,2-
tetrachloroethane30 was spin-coated on a 1.2 cm61.2 cm
ITO-coated glass in argon atmosphere using a Laurell (USA)
spin-coater. The spin-coated polymer layer was annealed under
1023 Torr at 150 uC for 2 hours to remove the residual 1,1,2,2–
tetrachloroethane solvent and to ensure complete dehydro-
chlorination.
Anal. Calcd for C28H29NO: C, 85.02; H, 7.39; N, 3.54.
Found: C, 85.50; H, 6.95; N, 3.43%. 1H-NMR (CDCl3; d/ppm):
0.83–1.92 (m, 15H, -CH(CH2CH3)CH2CH2CH2CH3), 3.98 (d,
2H, -OCH2CH-), 6.58 (d, 2H, -CHLCH-), 6.77–7.26 (m, 8H,
Ar-H), 7.89 (d, 2H, Ar-H). IR (KBr, cm21): 3058 (aromatic
C–H stretch), 2925 (aliphatic C–H stretch), 1727 (CLO stretch),
1598, 1476 (Ar CLC stretch), 1311 (C–N stretch), 1227, 1194
(C–O stretch), 969 (trans~CH out-of-plane bending).
2-(Carbazol-9-yl)terephthalic acid, 8. A mixture of 2-bromo-
terephthalic acid (5.0 g, 0.020 mol), carbazole (6.8 g;
0.041 mol), K2CO3 (1.4 g; 0.010 mol) and copper powder
(0.15 g) was mixed in DMF (250 mL) and the mixture was
refluxed for 48 h.29 The reaction mixture was worked up the
same way as in the preparation of compound 3. The crude
yellow precipitate was recrystallized from a mixture of water
and methanol (1 : 2 by volume) (5.0 g; 75% yield), mp
290 uC. Anal. Calcd for C20H13NO4: C, 72.50; H, 3.95; N,
˚
The Al or Li : Al alloy (Li 0.2 wt%) electrode 1500 A thick
was vapor deposited using a LEYBOLD L560 (Germany)
21
˚
apparatus at a deposition rate of 5 A s onto the polymer
layer at a pressure of 161026 Torr. Deposition of the cathode
electrode was conducted at the Korea Basic Science Institute –
Seoul Branch, Korea. The active area of the device was
4.9 mm2. The thickness of polymer was determined by a
TENCOR P-10 surface profiler.
1
4.23. Found: C, 72.31; H, 3.92; N, 4.21%. H-NMR (acetone-
d6; d/ppm): 7.1 (d, 2H, Ar-H), 7.2 (t, 2H, Ar-H), 7.3 (t, 2H,
Ar-H), 8.1 (d, 2H, Ar-H), 8.2 (s, 1H, Ar-H), 8.2 (d, 1H, Ar-H),
8.2 (d, 1H, Ar-H), 11.8 (s, 2H, COOH). IR (KBr, cm21): 3600–
2300 (O–H stretch of acid), 1700 (CLO stretch), 1591, 1454(Ar
CLC stretch), 1310 (C–N stretch).
The UV–vis absorption and luminescence spectra were
respectively recorded on an HP8452A Diode Array spectro-
photometer and an AMINCO-Bowman Series 2 luminescence
spectrometer at room temperature. The current and lumines-
cence intensity as a function of applied field were measured
using an assembly consisting of a PC-based DC power supply
(HP 6623A) and a digital multimeter (HP 34401). A light power
meter (Newport Instruments, Model 818-UV) was used to
measure the device light output in microwatts. Luminance was
measured by a MINOLTA LS-100 luminance meter. The
ultraviolet photoelectron spectroscopy (UPS) data were
acquired at room temperature with a VG ESCALab 220i
spectrometer (UK) with a VG UV lamp. UPS analysis was
performed using He I (21.2 eV) photons. The base pressure of
the analysis chamber was lower than 1610210 Torr and the
combined instrumental resolution was about 0.1 eV. Atomic
force microscopy (AFM) was conducted on a AutoProbe CP
(Park Scientific Instruments, USA) at the Korea University
Engineering Laboratory Center, Seoul, Korea.
2-(Carbazol-9-yl)-1,4-bis(hydroxymethyl)benzene, 9. To
a
stirred mixture of LiAlH4 (1.8 g; 0.048 mol) in dry THF30
(200 mL) was added dropwise a solution of 8 (3.2 g; 0.976
1022 mol) in dry THF30 (200 mL) and the mixture was refluxed
for 2 h. The reaction mixture was treated as the same way as in
the preparation of compound 7. Chromatography (silica gel,
hexane–ethyl acetate~1 : 1 by volume) of the crude residue
afforded the title compound as an oil (1.6 g, 55% yield). Anal.
Calcd for C20H17NO2: C, 79.19; H, 5.65; N, 4.62. Found: C,
1
78.94; H, 5.54; N, 4.59%. H-NMR (acetone-d6; d/ppm): 4.2
(t, 1H, -CH2OH), 4.3 (d, 2H, -CH2OH), 4.5 (t, 1H, -CH2OH),
4.8 (d, 2H, -CH2OH), 7.1 (d, 2H, Ar-H), 7.3 (t, 2H, Ar-H), 7.4
(t, 2H, Ar-H), 7.4 (s, 1H, Ar-H), 7.6 (d, 1H, Ar-H), 7.9 (d, 1H,
Ar-H), 8.3 (d, 2H, Ar-H). IR (KBr, cm21): 3342 (O–H stretch),
3051 (aromatic C–H stretch), 2914 (aliphatic C–H stretch),
1590, 1448 (Ar CLC stretch), 1312 (C–N stretch).
3026
J. Mater. Chem., 2001, 11, 3023–3030

View Article Online
ITO/polymer 1/Ca/Al and ITO/PEDOT/polymer 1/Ca/Al
devices. The patterned ITO-coated glass slides were cleaned
by sequential ultrasonication in acetone, propan-2-ol and
distilled water for 13 minutes, respectively, and finally cleaned
by ultraviolet–ozone for 13 minutes.34
The conducting polymer solution of poly(3,4-ethylene-
dioxythiophene-2,5-diyl) doped with polystyrene sulfonate
(PEDOT–PSS) (Bayer) was spin-coated with 3300 rpm for
1 min and dried at 150 uC for 30 minutes. The electrical
conductivity of this film measured by the four line probe
method was 10 S cm21. The solution (1 wt%) of polymer 1
in purified 1,1,2,2-tetrachloroethane30 was spin-coated at
1000 rpm for 1 min and the film was subjected to thermal
treatment at 150 uC for 1 h under vacuum. The Ca cathode
˚
2000 A thick was vacuum deposited from the tungsten boat at a
Fig. 1 UV–vis absorption and PL spectra of thin films of polymers 1
and 2 (lEX~340 nm).
deposition rate of 2 A s21 under a pressure of 3.061027 Torr.
˚
An Al capping layer was then evaporated to protect the Ca
21
˚
cathode at a deposition ratio of 4 A s
under the same
the alkoxy pendants. These values are very close to the value
(2.4 eV) of PPV reported earlier by others37 and us.8 The
absorption positions by the carbazole pendant groups in
the polymers are slightly red-shifted when compared with the
absorption position of the corresponding low molar mass
compound. Carbazole itself shows absorption at 300–345 nm
in ethanol with lmax~323 nm.38 This suggests that there is a
slight electronic interaction between the main chain and the
carbazole pendant. Similar observations were made by us
earlier for the PPV derivatives carrying styryl pendants.39
When compared with the absorption of unsubstituted PPV
(lmax~430 nm), the absorption (lmax~461 nm) by polymer 1
is red-shifted whereas the absorption (lmax~433 nm) position
pressure. The active area of the EL devices was 4 mm2.
Photoluminescence (PL) and electroluminescence (EL)
spectra were recorded on a PC1 photon counting spectro-
fluorometer (ISS Inc., USA). Current–voltage (I–V) character-
istics and the intensities of EL emission were simultaneously
measured with a Keithley 238 SMU electrometer and a BM7
luminance meter (Topcon Technologies, Inc., USA). Device
fabrication and all the measurements were performed in a dry
argon filled glove box without exposing to air.
Results and discussion
by polymer
ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) absorbs
at 350–570 nm with max~503 nm. The only structural
2 is similar to PPV. Poly[2-methoxy-5-(2’-
Synthesis of monomers and polymers
Polymers 1 and 2 were synthesized by the Gilch and
Wheelwright method33 widely used in the synthesis of organic
soluble PPV derivatives. The structures of intermediates and
monomers were confirmed by elemental analysis, and IR and
1H-NMR spectroscopy.
Polymer 1 is readily soluble at room temperature in organic
solvents such as tetrahydrofuran and 1,1,2,2-tetrachloro-
ethane. However, polymer 2 shows a poorer solubility in
these solvents. The molecular weight of polymer 1 determined
by gel-permeation chromatography in tetrahydrofuran against
l
difference between polymer 1 and MEH-PPV is that polymer
1 bears a carbazole moiety instead of the methoxy group in
MEH-PPV. The carbazole group appears to be less efficient in
electron-donating than the methoxy group according to the
UV–vis absorption spectra of polymer 1 and MEH-PPV. In
fact, comparison of the UV–vis absorption spectrum of
polymer 2 with that of PPV leads to the conclusion that the
carbazole group causes little change in the absorption position.
It is conjectured that the electron-donation character of the
carbazole group is nullified by the partial destruction of the
coplanarity of the main chain by the steric effect brought about
due to its bulkiness.
Photoluminescence (PL) spectra of the films (600 nm thick)
of polymers 1 and 2 obtained for the excitation wavelength of
340 nm are compared in Fig. 1. Polymer 1 emits light between
490–650 nm (maximum emission at about 530 nm) and
polymer 2 at 450 nm–630 nm (maximum emission at about
490 nm). The PL emission of polymer 1 again is red-shifted
when compared with the emission by polymer 2. The PL
spectra of polymers 1 and 2 are composed of two main bands,
being slightly simpler than the spectral feature of PPV
emission.
In fact, the general feature of the emission spectra of polymer
1 and 2 is very similar to that of MEH-PPV. The strongest
emission peaks of the present polymers may be the results of
overlapped emissions both by the carbazole pendants and the
alkoxy-substituted (polymer 1) or unsubstituted (polymer 2)
main chain. Alternatively carbazole groups may be involved
only in energy transfer processes between themselves and the
backbone. Carbazole itself emits light over 340–420 nm in
ethanol.38 Poly(vinylcarbazole) is known to emit at 360–
540 nm with maximum emission at 418 nm.3 The excitation
spectra of polymer 1 collected for the emission at the
wavelengths of 530, 560, and 620 nm are given in Fig. 2 and
are compared with its absorption spectrum. First of all, we note
not only that the excitation spectra are independent of emission
¯
¯
a polystyrene standard is Mw~72 000 and Mw~49 700 with a
polydispersity index (PDI) of 1.4. The relatively narrow
molecular weight distribution must be brought about by
removal of the low molar mass portion by the extraction
process. Since polymer 2 is not readily soluble in solvent, we
measured only its solution viscosity values at 30 uC for a
solution (0.2 g dL21) in 1,1,2,2-tetrachloroethane; its solution
inherent viscosity value is 1.12 dL g21. Both polymers readily
form free standing films when cast from 1,1,2,2-tetrachloro-
ethane solution.
UV–vis absorption and photoluminescence characteristics
Fig. 1 compares UV–vis absorption spectra and photolumines-
cence (PL) spectra of polymers 1 and 2. Both polymers 1 and 2
show an absorption from the carbazole moiety with lmax at
340 nm. For polymer
1 the absorption at 350–540 nm
originates from the p–p* transition of the main chain. Polymer
2 absorbs at about 350–500 nm for the same transition. The
red-shift for the absorption by polymer 1 when compared to
polymer 2 can be explained by the presence of the electron
donating alkoxy pendant group, since the alkoxy group is well
known to be an electron donor.35,36 The absorption edges of
the absorption spectra for polymers 1 and 2 are 530 and
500 nm, respectively, corresponding to the optical band gap of
2.4 eV and 2.5 eV. As is well known,35,36 the presence of the
electron-donating alkoxy groups in polymer 1 reduces the band
gap energy when compared with polymer 2 which does not bear
J. Mater. Chem., 2001, 11, 3023–3030
3027

View Article Online
Fig. 2 Comparison of excitation spectra of polymer 1 at different
emission wavelength.
Fig. 4 Comparison of electric field vs. current density and light
intensity (inset) curves of ITO/polymer 1 or 2 (60 nm)/Al devices.
wavelength and are not exactly same as the absorption
spectrum, but also that absorption by the carbazole pendants
at about 320–350 nm makes a strong contribution to the
emission by the backbone, which can be interpreted by a facile
excited state energy transfer from pendants to the backbone.
This must be the reason why we do not observe the occurrence
of a separate emission by the carbazole pendants. Lee et al.40
where the carriers are injected through triangular barriers
caused by the offset between the bands of the polymer and the
electrode.41 We obtained energy levels of polymers 1 and 2
from ultraviolet photoelectron spectra (UPS)8 and their optical
band gaps. The HOMO and LUMO energy levels of polymers
were evaluated as described by Schmidt et al.42 The HOMOs of
polymers 1 and 2 are 5.5 and 6.1 eV, respectively. Their LUMO
levels were evaluated to be 3.2 and 3.6 eV, respectively. In
general, the LUMO level of a PPV derivative is significantly
lowered by the alkoxy group attached to the phenylene ring as
is observed for the present polymers.35,36 According to the
band diagram it is expected that injection of holes from the ITO
anode to polymer 1 will be more favorable than to polymer 2.
This indicates that, at the same field, the current density of
polymer 1 will be higher than that of polymer 2 as shown in
Fig. 4. As a result the field required to obtain the same current
density is lower for polymer 1 than for polymer 2. The inset in
Fig. 4 also tells us that the brightness of emitted light by
polymer 1 increases much more rapidly than that by polymer 2.
The external quantum efficiencies of the devices fabricated
from the two polymers are compared in Fig. 5. For the sake of
comparison, data for PPV and MEH-PPV prepared by us8 are
included in Fig. 5. The maximum external quantum efficiency
of the ITO/polymer 1/Al device is 0.01% and is 550-fold higher
than that of the ITO/polymer 2/Al device (1.861025%), and
60-fold higher than ITO/PPV/Al (1.761024%). This value is
about 5.6 times the efficiency reported for MEH-PPV devices.41
The higher external quantum efficiency of polymer 1 than that
of polymer 2 may partly be due to an improved balance of the
carrier injection and transport. Also, it is probable that the
bulky alkoxy pendant groups in polymer 1 keep the conjugated
polymer backbones apart from each other and prevent the
formation of interchain excitons more efficiently than polymer
2.24,25 Formation of interchain excitons is known to cause
recently reported
a similar energy transfer process for
poly(carbazolylacetylenes). If this is the case, the emission
peaks should not contain the contribution from the carbazole
pendants and they are mainly from the p-systems in the main
chain. In addition, we observed that the PL spectra of polymers
1 and 2 show exactly the same features regardless of the
excitation wavelength, indicating that the emissive state is the
same even with high-energy excitation.
Electroluminescence and I–V characteristics of light-emitting
devices
Fig. 3 compares electroluminescence (EL) and photolumines-
cence (PL) spectra of polymers 1 and 2. The general appearance
of the EL spectra is not much different from the corresponding
PL spectra with the EL spectra being broader. It implies that
the carbazole moiety just changes the electronic state of the
main chain and does not emit light. This also indicates that
emissive excitons are formed from combination of the injected
holes and electrons mainly along the main chain, and if any
excitons are formed in the pendants, they are in a higher energy
state and, thus, migrate very fast to the lower energy states of
the main chain.
The characteristics of current density and light intensity
versus electric field of ITO/polymer 1 or 2/Al devices are
compared in Fig. 4. The threshold field required to obtain a
current density of 0.1 mA mm22 for polymer 1 was 0.6 MV cm21
and is lower than that of polymer 2 (0.8 MV cm21). These I–V
characteristics could be explained using a band–bend model
Fig. 5 Comparison of external quantum efficiencies of polymer 1, 2 and
PPV devices.
Fig. 3 Comparison of EL and PL spectra of polymer 1 and 2.
3028
J. Mater. Chem., 2001, 11, 3023–3030

View Article Online
nonradiative decay lowering the external quantum efficiency of
LED devices.26–28 As a result of these factors, it seemed that
polymer 1 shows a much higher efficiency than polymer 2.
Moreover, it was noted that the film quality of polymer 1 was
much better than polymer 2. Spin-casting of polymer 1 gave a
balance of electron and hole injection could be improved and
the radiationless quenching of excitons at the metal electrode/
polymer interface could be reduced by shifting the recombination
zone away from the metal electrode.47,48
As a result, the external quantum efficiencies of the devices A
and B are significantly higher than that of the ITO/polymer
1/Al device. Fig. 6b clearly demonstrates that device B
performs best among the three devices as far as brightness of
emitted light and photometric efficiency are concerned. The
device fabricated from polymer 1 and PEDOT exhibits the
highest luminance for the same current density values. This tells
us that opposite carriers form singlet excitons most effectively
in device B indicating that injection and mobility of the carriers
are most balanced. The maximum photometric efficiencies of
devices A and B were 0.3 and 4.4 cd A21, respectively, with the
maximum luminance attainable being 1740 cd m22 for device A
at an electric field of 1.0 MV cm21 and 30 390 cd m22 for
device B at an electric field of 1.50 MV cm21. The device
fabricated with MEH-PPV showed a maximum photometric
efficiency of 1.1 cd A21 with the maximum luminance being
20 540 cd m22 at 1.38 MV cm21. The driving electric fields for
200 cd m22 were 0.59 MV cm21 for device A and 0.45 MV cm21
for device B, respectively. In comparison, the maximum
luminance value observed for the devices of ITO/polymer
1/Li : Al was 1300 cd m22. It also should be noted that polymer
1 is a green light emitter whereas MEH-PPV is an orange–red
light emitter. As far as device stability is concerned, it took
70 hours for the device B to reduce its brightness from
1000 cd m22 to 500 cd m22. However, detailed studies are
required on the long-term stability of the devices described.
˚
film of an average roughness of 12.4 A, whereas the average
˚
roughness (29.2 A) of the polymer 2 film was much worse. The
difference results from poorer solubility of polymer 2 in organic
solvents than polymer 1. In addition, our preliminary results43
suggest that polymer 1 may reveal an improved balance in
carrier transport, which would result in a better EL device
performance.
Fig. 5 also shows that a device prepared using a Li : Al alloy
exhibits much higher external efficiencies than the ones having
the Al cathode. Evidently, the low work function of lithium
metal (2.9 eV) improves electron injection.44 In addition,
insertion of the conducting PEDOT [poly(3,4-ethylenedioxy-
2,5-thiophene)] layer between the ITO anode and the light
emitting layer further improves the efficiency.45 Fig. 5, how-
ever, implies that the devices with the Li cathode have a poor
stability. This may be due to the higher sensitivity of the lithium
metal toward air including water, oxygen and carbon dioxide.
It is also very possible that aluminium was not effectively co-
deposited to protect the lithium metal. The presence of the con-
ducting layer did not change the EL spectrum at all in the
overall position and shape. This is a strong indication that the
conducting layer simply acts as a carrier transporting medium.
In order to facilitate carrier injection, in addition to the use
of a conducting polymer layer on the ITO electrode, we
fabricated a device using a low work function calcium cathode
instead of aluminium. We again coated the doped PEDOT
layer, as an organic conducting hole injection layer, on the ITO
electrode. It not only makes the hole injection easier, but also
is expected to mitigate the adhesion problem between the
electrode and organic emitting layer.46 In order to protect the
calcium cathode, aluminium was vacuum deposited onto it. We
compared the characteristics of two LED devices; ITO/polymer
1/Ca/Al (device A) and ITO/PEDOT/polymer 1/Ca/Al (device
B). The electric field versus current density and light intensity
characteristics of devices are compared in Fig. 6. For the sake
of comparison, data for the device fabricated using MEH-PPV
prepared by us also are included in the figure.
Acknowledgements
This research was supported by the Ministry of Science and
Technology, Republic of Korea and the Korea Science and
Engineering Foundation, through the Center for Electro-
and Photo-Responsive Molecules, Korea University. K. Kim
wishes to acknowledge the support by the Ministry of
Education through the Brain Korea 21 program. Y. Park
was partially supported by the MOST and the KOSEF through
the National Research Laboratory Program and the Atomic-
Scale Surface Science Research Center.
It is known that23 the major carrier for PPV and its
derivatives is positive holes that govern the threshold electric
field and that minor carriers are electrons that control the
device efficiency. As described above, the PEDOT layer (see
Fig. 6a) helps hole (major carrier) injection to the polymer layer
References
1
J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. Mackay, R. H. Friend, P. L. Burn and A. B. Holmes, Nature,
1990, 347, 539.
and reduces turn-on electric field of device B (0.31 MV cm21
)
compared to device A (0.43 MV cm21). Because the Ca cathode
facilitates electron injection to the emitting polymer layer, the
2
3
4
5
6
7
8
9
G. Gustafsson, Y. Cao, G. M. Treacy, F. Klavetter, N. Colaneri
and A. J. Heeger, Nature, 1992, 357, 477.
J. Kido, K. Hougawa and K. Nagai, Appl. Phys. Lett., 1993, 63,
2627.
N. C. Greenham, S. C. Moratti, D. D. C. Bradley, R. H. Friend
and A. B. Holmes, Nature, 1993, 365, 628.
Z. Yang, I. Sokolik and F. E. Karasz, Macromolecules, 1993, 26,
1188.
M. Berggern, O. Lnganas, G. Guslafsson, J. Ramusson,
M. R. Anderson and T. Hjertberg, Nature, 1994, 372, 444.
W. J. Feast, J. Tsibouklis, K. L. Pouwer, L. Groenendaal and
E. W. Meijer, Polymer, 1996, 37, 5017.
S.-J. Chung, K.-Y. Kwon, S.-W. Lee, J.-I. Jin, C. H. Lee, C. E. Lee
and Y. Park, Adv. Mater., 1998, 10, 1112.
S.-J. Chung, J.-I. Jin and K. K. Kim, Adv. Mater., 1997, 9, 551.
10 N. F. Colaneri, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. Mackay, R. H. Friend, P. L. Burn, A. B. Holmes and
C. W. Spangler, Phys. Rev. B, 1990, 42, 11671.
11 C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett., 1987, 51, 913.
12 A. Tsuchida, A. Nagata, M. Yamomoto, H. Fukui, M. Sawamoto
and T. Higashimura, Macromolecules, 1995, 28, 1285.
Fig. 6 Comparison of electric field vs. current density (a) and current
density vs. luminance (b) curves of device A (ITO/polymer 1(100 nm)/
Ca/Al), B (ITO/PEDOT(25 nm)/polymer 1(80 nm)/Ca/Al) and MEH-
PPV (ITO/PEDOT(25 nm)/MEH-PPV(80 nm)/Ca/Al).
13 X.-T. Tao, Y.-D. Zhang, T. Wada, H. Sasabe, H. Suzuki,
T. Watanabe and S. Miyata, Adv. Mater., 1998, 10, 226.
14 D. Romero, F. Nu¨esch, T. Benazzi, D. Ades, A. Siove and
L. Zuppirali, Adv. Mater., 1997, 9, 1158.
J. Mater. Chem., 2001, 11, 3023–3030
3029

View Article Online
15 Y. Yang and A. J. Heeger, Appl. Phys. Lett., 1994, 64, 1245.
16 H.-K. Shim, H.-J. Kim, T. Ahn, I.-N. Kang and T. Zyung, Synth.
Met., 1997, 91, 289.
17 I. D. Parker, Q. Pei and M. Marrocco, Appl. Phys. Lett., 1994, 65,
1272.
18 Y. Yang, Q. Pei and A. J. Heeger, J. Appl. Phys., 1996, 79, 934.
19 H. Meng, Z.-K. Chen, W.-L. Yu, J. Pei, X.-L. Liu, Y.-H. Lai and
W. Huang, Synth. Met., 1999, 100, 297.
A. J. Heeger, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.),
1994, 35, 185.
36 X.-C. Li and S. C. Moratti, in Photonic Polymer Systems, ed.
D. L.Wise, G. Wnek, D. J. Trantolo, T. M. Cooper and
J. D. Gresser, Marcel Dekker, New York, 1998, pp. 335–371.
37 W. Rieß, in Organic Electroluminescence Materials and Devices;
ed. S. Miyata and H. S. Nalwa, Gordon and Breach, London,
1997, pp. 73–146.
20 S.-H. Jin, W.-H. Kim, I.-S. Song, S.-K. Kwon, K.-S. Lee and
E.-M. Han, Thin Solid Films, 2000, 363, 255.
21 Z. Peng, Z. Bao and M. E. Galvin, Chem. Mater., 1998, 10, 2086.
22 S.-Y. Song, M. S. Jang, H.-K. Shim, D.-H. Hwang and T. Zyung,
Macromolecules, 1999, 32, 1482.
23 C. Adachi, T. Tsutsui and S. Saito, Appl. Phys. Lett., 1990, 56, 799.
24 Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Solid State
Commun., 1991, 80, 605.
25 M. R. Andersson, G. Yu and A. J. Heeger, Synth. Met., 1997, 85,
1275.
38 I. B. Berlman, in Handbook of Fluorescence Spectra of Aromatic
Molecules, 2nd edn., Academic Press, New York, 1971, p. 206.
39 J.-I. Jin, Y.-H. Lee, C.-K. Park and B.-K. Nam, Macromolecules,
1994, 27, 5329.
40 P. P.-S. Lee, Y. Geng, H. S. Kwok and B. Z. Tang, Thin solid films,
2000, 363, 149.
41 I. D. Parker, J. Appl. Phys., 1994, 75, 1656.
42 A. Schmidit, M. L. Anderson and N. R. Armstrong, J. Appl.
Phys., 1995, 78, 5619.
43 Y. R. Hong, D. W. Lee, K. Kim, J.-I. Jin, C. E. Lee, H. M. Lee,
Y. Park, B.-H. Shon and J.-K. Park, presented at the World
Polymer Congress, Warsaw, Poland, July 9–14, 2000; Macromol.
Symp.. in press. The ratio of hole mobility against electron
mobility of polymer 1 determined by the time-of-flight method was
about 5. Up to the present moment, there are no data reporting
electron mobility in MEH-PPV, although there are some reports
only on hole mobility.
44 S. Naka, N. Tamekawa, T. Terashita, H. Okada, H. Anada and
H. Onnagawa, Synth. Met., 1997, 91, 129.
45 S. A. Carter, M. Angelopoluos, S. Karg, P. J. Brock and J. C. Scott,
Appl. Phys. Lett., 1997, 70, 2067.
26 T.-Q. Nguyen, I. B. Martini, J. Liu and B. J. Schwartz, J. Phys.
Chem. B, 2000, 104, 237.
27 M. Yan, L. J. Rothberg, F. Papadimitrakopoulos, M. E. Galvin
and T. M. Miller, Phys. Rev. Lett., 1994, 72, 1104.
28 J. M. Leng, S. Jeglinski, X. Wei, R. E. Benner, Z. V. Vardeny,
F. Guo and S. Mazumdar, Phys. Rev. Lett., 1994, 72, 156.
29 S. C. Creason, J. Wheeler and R. F. Nelson, J. Org. Chem., 1972,
37, 4440.
30 D. D. Perrin and W. L. F. Armarego, in Purification of Laboratory
Chemicals, 3rd edn., Pergamon Press, New York, 1988.
31 M. Demuynck, T. Eda, K. Sakaki and T. Maeshima, Synthesis,
1980, 746.
32 R. F. Nystorm and W. G. Brown, J. Am. Chem. Soc., 1947, 69,
1197.
33 H. G. Gilch and W. L. Wheelwright, J. Polym. Sci., Part A-1,
1966, 4, 1337.
34 K. Sugiyama, H. Ishii, Y. Ouchi and K. Seki, J. Appl. Phys., 2000,
87, 295.
35 J. L. Bredas, D. A. D. Santos, C. Quattrocchi, R. H. Friend and
46 J. S. Kim, R. H. Friend and F. Cacialli, Appl. Phys. Lett., 1999, 74,
3084.
47 A. Kraft, P. L. Burn, A. B. Holmes, D. D. C. Bradley, R. H. Friend
and J. H. F. Martens, Synth. Met., 1993, 57, 4163.
48 N. C. Greenham, R. H. Friend, A. R. Brown, D. D. C. Bradley,
K. Pichler, P. L. Burn, A. Kraft and A. B. Holmes, Proc. SPIE –
Int. Soc. Opt. Eng., 1993, 1910, 84.
3030
J. Mater. Chem., 2001, 11, 3023–3030