Synthesis and electrooptical properties of copolymers derived from phenol-functionalized telechelic oligofluorenes

Article abstract of DOI:10.1021/ma0518353

This paper describes the synthesis, characterization, and electrooptical properties of a series of phenol-capped, discrete oligofluorenes with 2, 3, 5, and 7 fluorene units and a statistical oligomer with an average of about 10 fluorene units. Polymers were prepared from the oligomers by various linking reactions through the phenol groups. High molecular weight polycarbonate polymers derived from these oligomers were prepared using BPA-bis-chloroformate (BPABCF) as the linking group. Trends in the optical and electrical properties as a function of oligomer length are reported, Device data for this family of emissive copolymers indicates that charge mobility increases with conjugation length and can be better than that of an analogue homopolymer, A surprising shift in LUMO values due to CFs substitution at the ends of the oligomers was observed.

Full text of DOI:10.1021/ma0518353

Macromolecules 2005, 38, 10667-10677
10667
Synthesis and Electrooptical Properties of Copolymers Derived from
Phenol-Functionalized Telechelic Oligofluorenes
Timothy Burnell,† James A. Cella,* Paul Donahue, Anil Duggal, Thomas Early,
Christian M. Heller, Jerry Liu, Joseph Shiang, David Simon,
Katarzyna Slowinska,‡ Micah Sze, and Elizabeth Williams
General Electric Global Research Center, 1 Research Circle, Niskayuna, New York 12309
Received August 19, 2005; Revised Manuscript Received October 10, 2005
ABSTRACT: This paper describes the synthesis, characterization, and electrooptical properties of a series
of phenol-capped, discrete oligofluorenes with 2, 3, 5, and 7 fluorene units and a statistical oligomer with
an average of about 10 fluorene units. Polymers were prepared from the oligomers by various linking
reactions through the phenol groups. High molecular weight polycarbonate polymers derived from these
oligomers were prepared using BPA-bis-chloroformate (BPABCF) as the linking group. Trends in the
optical and electrical properties as a function of oligomer length are reported. Device data for this family
of emissive copolymers indicates that charge mobility increases with conjugation length and can be better
than that of an analogue homopolymer. A surprising shift in LUMO values due to CF3 substitution at
the ends of the oligomers was observed.
Introduction
impurities in the starting monomers or those introduced
during the polymerization process. For these polymers,
therefore, there is a conjugation length polydispersity,
not easily controlled, that leads to a spread in the energy
levels, carrier transport properties, and other photo-
physical properties that can affect OLED efficiency. In
such disperse systems broadened emission may result
or exciton energy transfer from high- to low-energy
states may lead to red-shifted emission emanating from
a very small population of conjugated segments.11
An approach that circumvents these conjugation
length dispersity effects is to prepare monodisperse
oligomers bearing functionality that enables linking of
the oligomers to form a polymer. In this way, the
integrity of the conjugated segment is maintained in the
polymerization process and is decoupled from the mo-
lecular weight polydispersity and from defects associ-
ated with the polymerization process. This approach
offers the possibility of tuning not only the electronic
properties by variation of the conjugated segment but
also the physical characteristics, such as morphology
and solubility, by varying the linking segment. Also, by
introducing comonomers it should be possible to tailor
the charge carrier transport properties of these materi-
als. These considerations are particularly important for
the fabrication of multilayer devices. Finally, the method
allows for rigorous purification of the oligomeric seg-
ments by chromatography or crystallization prior to
polymerization and should thus afford high molecular
weight polymers with good film-forming properties and
good thermal stability.
Significant advances in the performance of polymer-
based light-emitting diodes have been made since the
first demonstration of electroluminescence in conjugated
polymers more than a decade ago.1,2 The lure of polymer-
based devices compared to small molecule LED’s is the
possibility of using low-cost fabrication methods, such
as roll-to-roll processing or other printing methods
suitable for large area or pixilated displays. Polymers
that emit blue light efficiently are of special interest,
as they are needed for production of full color displays
and for white light illumination devices. Among the
blue-light-emitting polymers currently available, poly-
fluorenes have emerged as the most promising due to a
combination of high luminescence efficiency, good film-
forming properties, and good thermal stability.3-7
Efforts to improve the performance of polyfluorenes
in OLED devices, particularly with regard to minimizing
their tendency to aggregate in the solid state, often focus
on structural modification of substituents in the bridg-
ing 9-position. Bulky substituents in this position are
thought to diminish interchain association in the solid
state that leads to inefficient aggregate or excimer
emission in polyfluorenes.4,5,8-10 However, it is the
nature of the conjugated backbone, more specifically,
the conjugation length of segments in these polymers
that defines the band gap, HOMO-LUMO energy
levels, charge transport character, fluorescence quan-
tum yield, and the singlet/triplet exciton population
split, all of which are critical to efficient OLED device
performance. Polyfluorenes are prepared by aryl-aryl
coupling reactions of suitably functionalized monomers.
The conjugation length in these polymers is determined,
not by the average number of repeat units per chain
(DPave) but by solid state morphology, which determines
the distance between fluorene segments twisted out of
conjugation and by chain defects arising either from
The synthesis and characterization of monodisperse
oligofluorenes with various substituents at the 9-posi-
tion have been described in a number of publica-
tions.12-23 In general, optical and photophysical proper-
ties of these polymers have been reported but very little
device data is available. Several studies have examined
the use of discrete terfluorene or related oligomeric
molecules as emissive materials in small molecule
devices (SMOLEDs).24-26 Evaporated terfluorene films
have been shown to exhibit good carrier mobilities for
both holes and electrons in solid-state devices.27-28
* Corresponding author. E-mail: cella@crd.ge.com.
† Current address: 368 N Buck Rd, Dowingtown, PA 19335.
‡ Current address: Department of Chemistry & Biochemistry,
California State University at Long Beach, 1250 Bellflower Blvd.,
Long Beach, CA 90840.
10.1021/ma0518353 CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/19/2005

10668 Burnell et al.
Macromolecules, Vol. 38, No. 26, 2005
copious quantities of water, air-dried, and then recrystallized
from about 1.5 L of ethanol. The yield was 40 g (71%) and the
product contained 5-10% of 2,7-dibromofluorenone as indi-
cated by LC analysis. A small sample was purified by column
chromatography, but in general, the contaminated material
was carried on in subsequent steps.
To systematically evaluate the effect of conjugated
segment length on optical and photophysical properties
of polyfluorenes and to obtain data from working OLED
devices, discrete fluorene oligomers 1a (n ) 0, 1, 3, 5)
and 1b (n ) 0, 1) bearing phenol functionality at the
termini were prepared. The purified oligomers were
polymerized by reaction with BPA-bis-chloroformate
(BPAPCF) to obtain perfectly alternating BPA-oligo-
fluorene copolycarbonates 2a and 2b. The synthesis and
2-Bromo-9-[3,5-bis(trifluoromethyl)phenyl]-9-(4-hy-
droxyphenyl)fluorene, 5a. Solid 2-bromofluorenone (42 g,
161.5 mmol), dried in a vacuum oven overnight at 65 °C to
remove excess ethanol from crystallization, was added via a
solid addition funnel in portions to the Grignard reagent
prepared from 3,5-bis(trifluoromethyl)bromobenzene (50 g, 170
mmol), 1,2-dibromoethane (7.36 mL, 85 mmol), and magne-
sium (6.223 g, 256 mg-atom) in a three-neck 1-L flask filled
with 400 mL of anhydrous ether. After 30 min, the mixture
was quenched by adding an excess of saturated ammonium
chloride solution. The organic phase was washed three times
with equal volumes of water and saturated NaCl and was
passed through a cone of anhydrous MgSO4. Solvent was
removed on a rotary evaporator to afford a light brown solid
(4a). The total mass was dissolved in 800 mL of methylene
chloride in a one-neck 1-L flask along with phenol, (24 g, 256
mmol, 1.5 excess). Methanesulfonic acid (21.5 mL) was added
and the mixture was stirred at ambient temperature for 18 h.
When the reaction was complete, the organic phase was
washed with water and saturated NaCl. Evaporation of solvent
yielded an off-white solid. The product was further purified
by trituration with a minimal amount of methylene chloride
and hexanes (20/80) to give a final yield of 58.9 g (63.6%): MS
(FAB+) m/e 549 (M+); 1H NMR (CDCl3) δ 7.82-7.2 (m, 10,
ArH), 7.0 and 6.78 (AB doublet, 4, phenol-ArH), 4.78 ppm (s,
1, OH).
2-Bromo-9-(4-tert-butylphenyl)-9-(4-hydroxyphenyl)-
fluorene, 5b. Solid 2-bromofluorenone (5.18 g, 20 mmol) was
added in portions to the Grignard reagent prepared from
4-bromo-tert-butylbenzene (4.47 g, 21 mmol), 1,2-dibromo-
ethane (1.88 g, 10 mmol), and magnesium (0.753 g, 31 mg-
atom) in 75 mL of anhydrous ether. The reaction mixture was
refluxed for 1 h after all the solids had been added. The cooled
mixture was quenched by adding 50 mL of saturated am-
monium chloride. The organic phase was washed with equal
volumes of water and saturated NaCl and was passed through
a cone of anhydrous CaSO4. Solvent was removed on a rotary
evaporator to afford 10 g of an amber oil that was chromato-
graphed on silica gel (∼300 g) eluted with hexanes-ethyl
acetate to afford 5.64 g (72%) of the carbinol 4b as a colorless
oil: GC-MS (EI mode) (m/e) 392 (M+); 1H NMR (CDCl3) δ 7.7-
7.3 (m, 11, ArH), 1.3 ppm (s, 9, tert-butyl).
optical, photophysical, and charge transport properties
of these materials were determined, and the trends
observed as a function of oligomer length are reported
in this paper. The synthesis and characterization of a
range of condensation polymers derived from phenol-
functionalized terfluorenes and various difunctional
linkers are also described (Supporting Information).
Experimental Section
Measurements. 1H NMR spectra were recorded on a
Bruker 500 MHz instrument. UV spectra were recorded on a
Varian-Cary 300 Scan UV-vis spectrophotometer. Gel per-
meation chromatography was carried out using a Perkin-
Elmer Series 200 pump and a Perkin-Elmer 235 diode array
detector. Chloroform was used as the eluant at a flow rate of
1.0 mL/min through a MetaChem Technologies, 5 µm linear
300 × 7.8 mm column. Liquid chromatography was carried
out using a Perkin-Elmer Series 200 pump and UV-vis
detector on a Whatman Partisil 5 ODS-3 10 × 4.5 mm column
eluted with a water-acetonitrile (AN) linear gradient (50%-
95% AN) at a flow rate of 1.5 mL/min.
The carbinol 4b (6.0 g, 15 mmol) and phenol (2.2 g, 23.25
mmol) were dissolved in 10 mL of methylene chloride. Meth-
anesulfonic acid (200 µL, 3.08 mmol) was added, and the
mixture was stirred at ambient temperature. When the
reaction was complete (∼5 min), the organic phase was washed
with water and saturated NaCl. Evaporation of solvent af-
forded an oil that was chromatographed on 200 g of silica gel
(hexanes-ethyl acetate gradient) to afford 4.4 g (61%) of the
product as a colorless oil. The oil could be crystallized by slow
evaporation of an ether-hexane solution: MS (EI mode) m/e
470, 468 (M+)
Materials. 9,9-Dihexyl-2,7-dibromofluorene, 9,9-dihexyl-
fluorene-2,7-bis(trimethyleneboronate), fluorenone, palladium
acetate, tetraethylammonium hydroxide (20% aqueous solu-
tion), and N-bromosuccinimide were purchased from Aldrich
and used without further purification.
1H NMR (CDCl3) δ 7.8-6.7 (m, 15, ArH), 5.0 (s, 1, OH) and
1.3 ppm (s, 9, tert-butyl).
Tetrakis(triphenylphosphine)palladium(0) was either pur-
chased from Aldrich or prepared fresh according to the
literature.29
2-(1,1,2,2-Tetramethylethyleneboronato)-9-[3,5-bis(tri-
fluoromethyl)phenyl]-9-(4-hydroxyphenyl)fluorene, 6a.
A mixture of bromide 5a (5.5 g, 10 mmol), bis-pinacolatodi-
boron (3.0 g, 12 mmol), potassium acetate (2.9 g, 30 mmol),
and palladium(II) acetate (0.075 g, 0.3 mmol) in 25 mL of DMF
was stirred in a 75 °C bath for 3 h. The cooled mixture was
poured into 250 mL of water made slightly acidic with HCl.
The dark solids that separated were collected by filtration,
washed with water, and stirred with 150 mL of ethyl acetate.
The resulting slurry was filtered through Celite and the filtrate
was washed successively with water and brine then passed
through a cone of anhydrous CaSO4. Removal of solvent
afforded a dark residue that was chromatographed on 150 g
of silica gel eluted with 10-20% ethyl acetate in hexane to
n-Butyllithium (∼1.2 M) in hexane was obtained from
Aldrich and titrated prior to use using N-pivaloyl-o-toluidine.30
2-Bromofluorenone, 3. Fluorenone (41 g, 0.228 mol) was
dissolved in 225 mL of methanesulfonic acid and treated
portionwise at room temperature with solid N-bromosuccin-
imide (NBS) (38.55 g, 0.217 mol). The mixture became dark
and warm and the temperature was maintained below 60 °C
by controlling the rate of NBS addition. Upon completion of
the NBS addition, the mixture was allowed to cool to room
temperature and was poured into 1 L of ice water. The yellow
solid that separated was collected by filtration, washed with

Macromolecules, Vol. 38, No. 26, 2005
Copolymers Derived from Telechelic Oligofluorenes 10669
afford 6a as a colorless oil: MS (FAB+) m/e 596 (M+); 1H NMR
(CDCl3) δ 7.9-7.2 (m,10, ArH), 7.0 and 6.7 (AB doublets, 4,
phenol ArH), 1.35 ppm (s, 12, boronate CH3’s).
and brine (1×). The solution was then passed through a fritted
filter containing a 5-10 mm layer of 3-(diethylenetriamino)-
propyl-functionalized silica gel (SiliCycle) (200-400 mesh).
Evaporation of solvent afforded a residue that was dissolved
in ethyl acetate and passed through a short silica gel column
to afford 8 g of an orange solid. A 3.0 g sample of this material
was chromatographed on 120 g of silica gel (10-25% ethyl
acetate-hexane) to yield 1.96 g of the desired product: MS
(FAB+) m/e 1110 (M+); 1H NMR (CDCl3) δ 7.9-7.2 (m, 32,
ArH), 6.75 (d, 4, ArH o-phenol), 4.9 (s(br), 2, OH), 2.05 (br-s,
4, 9-R-CH2), 1.3 (s, 18, tert-butyl), 1.1 and 0.7 ppm (s(br), 22,
aliphatics).
Phenol Bromide Dimer 7. A mixture of 2.22 g (3.72 mmol)
of phenol borate 5a and 2,7-dibromo-9,9-dihexylfluorene (9.8
g, 20 mmol, 5.4 equiv), tetraethylammonium hydroxide (5.5
mL of a 20% aqueous solution), and 100 mL of toluene was
degassed with argon for 20 min and then tetrakis(tri-
phenylphosphine)palladium(0) was added and the mixture was
stirred under a positive nitrogen pressure and immersed in a
75 °C bath for 3 h. The cooled mixture was diluted with 50
mL of 1 N HCl, stirred for 10 min, and then filtered through
Celite. The aqueous phase was discarded and the organic
phase was washed successively with equal volumes of water
(3×) and brine (1×). The solution was then passed through a
fritted filter containing a 5-10 mm layer of 3-(diethylenetri-
amino)propyl-functionalized silica gel (SiliCycle) (200-400
mesh) and anhydrous CaSO4 and stripped to afford 12.6 g of
a residue that was chromatographed on 120 g of silica gel.
Elution with hexane afforded 7.9 g of recovered 2,7-dibromo-
9,9-dihexylfluorene (100%). Elution with 5-10% ethyl acetate/
hexane afforded 2.4 g (73%) of 7 as a white foam. MS (FAB+)
m/e 882 and 880 (M+); 1H NMR (CDCl3) δ 7.9-6.7 (m, 20, ArH),
4.9 (s, 1, OH), 2.0 (m(br), 4, 9-RCH2’s), 1.2-0.6 ppm (m(br),
22, aliphatic H’s).
Pentamer via Coupling of 7 with 2,7-Bis(trimethyl-
eneboronato)-9,9-dihexylfluorene, 1a (n ) 3). Bromophe-
nol 7 (0.651 g, 0.739 mmol) and 2,7-bis(trimethyleneboronato)-
9,9-dihexylfluorene (0.178 g, 0.355 mmol) were dissolved in
50 mL of toluene and 15 mL of 2 M K2CO3 solution. The
solution was degassed with argon for 20 min and then
hexaethylguanidinium chloride (40 µL, 0.075 mmol) was added
along with tetrakis(triphenylphosphine)palladium(0) (0.008 g,
0.70 mmol). Reaction was heated at 110 °C under nitrogen for
24 h. The cooled mixture was diluted with 20 mL of 1 N HCl,
stirred for 10 min, and then filtered through Celite. The
aqueous phase was discarded and the organic phase was
washed successively with equal volumes of water (3×) and
brine (1×). The solution was then passed through a fritted
filter containing a 5-10 mm layer of 3-(diethylenetriamino)-
propyl-functionalized silica gel (SiliCycle) (200-400 mesh) and
anhydrous CaSO4 and stripped to afford a residue that was
chromatographed on silica gel (10-25% EtOAc/Hexanes) to
afford 0.45 g of 1a (n ) 3): MS (FAB+) m/e 1936 (M+); 1H NMR
(CDCl3) δ 8.0-7.25 (m, 38, ArH), 7.1 and 6.8 (AB doublets, 8,
phenol-ArH), 4.80 (s, 2, OH), 2.1 (m, 12, 9-R-CH2), 1.1 and 0.8
ppm (m, 66, hexyl-H’s)
2-Bromo-7-(trimethylsilyl)-9,9-dihexylfluorene, 8. 2,7-
Dibromo-9,9-dihexylfluorene (50 g, 101.63 mmol) was dissolved
in dry THF (150 mL) at ambient temperature. The mixture
was cooled to -78 °C, at which point n-butyllithium (78.5 mL
of a freshly titrated 1.3 M solution, 102.05 mmol) was added
dropwise. The metalation reaction was monitored by quench-
ing aliquots with saturated ammonium chloride followed by
LC analysis. After 30 min at -78 °C, 15 mL of trimethylsilyl
chloride (15% excess, d ) 7.88 mmol/mL) was added. The
reaction mixture was allowed to warm to room temperature,
diluted with 100 mL of ether, and washed successively with
equal volumes of water (3×) and brine (1×). The organic phase
was dried over MgSO4 and stripped to afford 8 as a colorless
oil: 1H NMR (CDCl3) δ 7.7-7.37 (m, 6 (Ar-H), 2.0 (m, 4, 9-R-
CH2), 1.18, 0.82, and 0.70 (m, 22, hexyl-H’s), 0.38 ppm (s, 9,
Si-CH3). The main impurity was found to be 2,7-bis(trimeth-
ylsilyl)-9,9-dihexylfluorene.
2-(1,1,2,2-Tetramethylethyleneboronato)-9-(4-tert-bu-
tylphenyl)-9-(4-hydroxyphenyl)fluorene, 6b. A mixture of
5b (0.469 g, 1.0 mmol), bis(pinacolato)diboron (0.279 g, 1.1
mmol), potassium acetate (0.294 g, 3.0 mmol), palladium(II)
acetate (0.0075 g, 0.03 mmol), and DMF (10 mL) was stirred
at 80 °C under a positive nitrogen pressure for 5 h. Additional
bis(pinacolato)diboron (0.075 g) and palladium(II) acetate
(0.002 g) was added, and heating and stirring were continued
for an additional 1.5 h. The mixture was poured into 100 mL
of cold water made slightly acidic with HCl. Solids were
collected by filtration and stirred with ethyl acetate (50 mL).
The solution was filtered through Celite, washed with brine,
and dried by passage through a cone of anhydrous CaSO4.
Evaporation of solvent afforded a residue that was chromato-
graphed on 50 g of silica gel eluted with 10-20% ethyl acetate/
hexanes. The main fractions crystallized slowly from an ether-
1
hexane solution: MS (FAB+) m/e 516 (M+); H NMR (CDCl3)
δ 7.8-7.2 (m, 11, ArH), 7.15 and 6.7 (AB doublets, 4, phenol-
ArH), 4.8 (br-s, 1 OH), 1.3 (s, 12, pinacol-CH3), 1.25 ppm (s, 9,
tert-butyl CH3’s).
Phenol-Functionalized Bifluorene 1a (n ) 0). A mixture
of phenol borate 6a (0.596 g, 1.0 mmol), bromophenol 5a (0.549
g, 1.0 mmol), toluene (15 mL), and tetraethylammonium
hydroxide (0.275 g, 1.4 g of a 20% aqueous solution) was
degassed with argon for about 20 min and then tetrakis-
(triphenylphosphine)palladium(0) (0.0035 g, 0.022 mmol) was
added and the mixture was stirred under a positive nitrogen
pressure at 75 °C for 18 h. The cooled mixture was diluted
with ethyl acetate (25 mL) and 1.0 N HCl (25 mL), stirred
briefly, and then filtered through Celite. The organic phase
was washed with water (2×) and brine (1×) and then passed
through a funnel containing Drierite on top of a 1 cm layer of
3-(diethylenetriamino)propyl-functionalized silica gel (Sili-
Cycle) (200-400 mesh). Removal of solvent in vacuo afforded
a white powder: MS (FAB+) m/e 938 (M+); 1H NMR (CDCl3) δ
8.0-6.75 (m, 28, ArH) and 5.05 ppm (s, 2, OH).
Phenol-functionalized bifluorene 1b (n ) 0) was pre-
pared using the same procedure described for the preparation
of 1a from phenol borate, 6b, and bromophenol 5b: MS (FAB+)
m/e 778 (M+); 1H NMR (CDCl3) δ 7.8-6.7 (m, 30, ArH), 4.7
(s(br), 2, OH), 1.25 ppm (s, 18, tert-butyl).
Phenol-Capped Terfluorene 1a (n ) 1). A mixture of
phenol bromide 5a (3.288 g, 6 mmol), 2,7-bis(trimethylenebo-
ronato)-9,9-dihexylfluorene (1.266 g, 3 mmol), hexaethylgua-
nadinium chloride (HEG) (80 µL, 0.15 mmol), 150 mL of
toluene, and 2 M K2CO3 (25 mL) was degassed with argon for
20 min then tetrakis(triphenylphosphine)palladium(0) (75 mg)
was added and the mixture was immersed in a 110 °C bath.
The mixture was stirred under a positive nitrogen pressure
for 4.5 h then was allowed to cool to room temperature, at
which point only one spot was evident via TLC. 3-(Diethylene-
triamino)propyl-functionalized silica gel (SiliCycle) (75 mg,
200-400 mesh) was added and the mixture allowed to stir for
half an hour. The mixture was then filtered through Celite.
The filtrate was washed successively with equal volumes of
10% HCl (1×), water (2×), and brine (1×). Evaporation of
solvent afforded a residue that was dissolved in ethyl acetate
and passed through a short silica gel column to afford a white
solid (10-25% ethyl acetate-hexane): MS (FAB+) m/e 1271
1
(M+); H NMR (CDCl3) δ 7.95-7.2 (m, 26, ArH), 7.1 and 6.8
(AB doublets, 8, phenol ArH), 4.96 (s, 2, OH), 2.05 (m, 4, 9-R-
CH2) and 1.03 and 0.7 ppm (m, 22, hexyl-H’s) yield, 85%.
Phenol-Capped Terfluorene 1b (n ) 1). A mixture of
phenol bromide 5b (7.74 g, 16.5 mmol), 2,7-bis-trimethylenebo-
ronato-9,9-dihexylfluorene (4.07 g, 8.1 mmol), hexaethylgua-
nadinium chloride (40 mg, 0.15 mmol), and 2 M K2CO3 (50
mL) was degassed with argon for 20 min and then tetrakis-
(triphenylphosphine)palladium(0) (0.26 g, 0.225 mmol) was
added and the mixture was immersed in a 110 °C bath. The
mixture was stirred under a positive nitrogen pressure for 23
h and then filtered through Celite. The filtrate was washed
successively with equal volumes of 10% HCl (1×), water (2×),
2-(Trimethylsilyl)-9,9-dihexylfluorene-7-pinacolatobo-
rate, 9. A solution of bromide 8 (4.85 g, 10 mmol), bis-

10670 Burnell et al.
Macromolecules, Vol. 38, No. 26, 2005
(pinacolato)diboron (2.54 g, 10 mmol), potassium acetate (2.00
g, 20.0 mmol), and palladium(II) acetate (0.100 g, 0.45 mmol)
in N,N-dimethylformamide (25 mL) was stirred and heated
at 75 °C under a positive nitrogen pressure for 2 days. The
cooled mixture was poured into 150 mL of water slightly
acidified with HCl. The precipitated solids were collected by
filtration and then stirred with ethyl acetate and filtered again
through Celite. The ethyl acetate solution was washed with 1
N HCl (2×), water (3×), and brine (1×) and then passed
through a cone of anhydrous CaSO4. Removal of solvent
afforded a residue that was chromatographed on silica gel
eluted with hexane-5% EtOAc/hexane to afford a purified
residue that was crystallized from methanol to afford pure 9:
MS (EI+) m/e 532 (M+); 1H NMR (CDCl3) δ 7.85-7.5 (m, 6,
ArH), 2.0 (t, 4, 9-R-CH2), 1.4 (s, 12, pinacol-CH3), 1.1 (m, 12,
hexyl-CH2), 0.8 (t, 6, hexyl-CH3), 0.61(m, 4, hexyl-CH2), 0.35
ppm (s, 9, Si-CH3).
r,ω-2,2′-Diiodo-9,9-dihexylfluorene Pentamer 13. Bis-
(trimethylsilyl) pentamer 12 (0.95 g, 0.525 mmol) was dis-
solved in 20 mL of carbon tetrachloride. The solution was
cooled to 0 °C and iodine monochloride (1.08 mL of a 1 M
solution in methylene chloride, 2.05 mmol) was added dropwise
over 15 min. The reaction mixture was allowed to warm to
ambient temperature over 30 min and was then quenched with
saturated sodium sulfite solution. The mixture was transferred
to a separatory funnel and the organic phase was washed with
water (3×) and brine (1×) then dried over MgSO4. Removal of
solvent in vacuo afforded a light yellow solid (1.0 g, 99.5%):
MS (MALDI-TOF) m/e 1915 (M+); 1H NMR (CDCl3) δ 7.9-7.5
(m, 30, ArH), 2.1 (m(br), 20, 9-R-CH2), 1.2 and 0.8 ppm (m,
110, hexyl-H); for quantitative 13C NMR (CDCl3), see the
Supporting Information.
12 + 5a: Phenol-Functional Heptamer 1a (n ) 5).
Diiodo compound 13 (0.3635 g, 0.1898 mmol) and borate 6a
(0.25 g, 0.4176 mmol) was added to 25 mL of toluene, 1 mL of
40% tetraethylammonium hydroxide, and 1 mL of water. The
mixture was degassed with argon for 20 min and then tetrakis-
(triphenylphosphine)palladium(0) (0.0225 g, 0.019 mmol) was
added and the mixture was stirred under a positive nitrogen
pressure at 75 °C for 24 h. The cooled mixture was diluted
with 20 mL of 1 N HCl, stirred for 10 min, and then filtered
through Celite. The aqueous phase was discarded and the
organic phase was washed successively with equal volumes
of water (3×) and brine (1×). The solution was then passed
through a fritted filter containing a 5-10 mm layer of
3-(diethylenetriamino)propyl-functionalized silica gel (Sili-
Cycle) (200-400 mesh) and anhydrous CaSO4. Removal of
solvent in vacuo afforded a residue that was chromatagraphed
on silica gel (0-50% EtOAc/hexanes) to afford pure 1a (n ) 5)
(0.300 g, 61%) as a gum: MS (MALDI-TOF) m/e 2600 (M+);
1H NMR (CDCl3) δ 8.0-7.3 (m, 50, ArH), 7.2 and 6.8 (AB
doublets, 8, phenol-ArH), 4.75 (s, 2, OH), 2.2 (m, 20, 9-R-CH2),
1.2 and 0.8 ppm (m, 110, hexyl-H’s).
Statistical Oligomer 14. A mixture of 9,9-dihexylfluorene-
2,7-bis(trimethyleneborate) (0.301 g, 0.6 mmol), 2,7-dibromo-
9,9-dihexylfluorene (0.197 g, 0.4 mmol), 5a (0.220 g, 0.4 mmol),
tetramethylammonium hydroxide (1.0 mL of a 20% aqueous
solution), and 15 mL of toluene was degassed with argon for
20 min and then tetrakis(triphenylphosphine)palladium(0)
(0.03 g, 0.026 mmol) was added. The mixture was stirred under
a positive nitrogen pressure and immersed in a 75 °C bath for
3 h. The cooled mixture was diluted with 25 mL of 1 N HCl,
stirred for 10 min, and then filtered through Celite. The
aqueous phase was discarded and the organic phase was
washed successively with equal volumes of water (3×) and
brine (1×). The solution was then passed through a fritted
filter containing a 5-10 mm layer of 3-(diethylenetriamino)-
propyl-functionalized silica gel (SiliCycle) (200-400 mesh) and
anhydrous CaSO4 and stripped to afford a tan solid that was
dissolved in CH2Cl2 and precipitated into methanol. The
collected solid was dried in a vacuum oven to afford ∼0.4 g of
oligomer 14: 1H NMR (CDCl3) δ 7.9-7.2 (m, 57, ArH), 7.1 and
6.8 (AB doublets, 8, phenol ArH), 4.8 (s, 2, OH), 2.1 (m(br),
25, 9-R-CH2), 1.1 and 0.8 ppm (m(br), 167, hexyl-H); 31P NMR
endgroup quantitation; Mn (calcd ) 3761, n ) 8.50); MS
(MALDI-TOF) (major signals) m/e (rel intensity), 1270 (14, n
) 0) (32), 543 (phenyl cap, n ) 0) (50), 1935 (14, n ) 1) (100),
2208 (n ) 1, phenyl cap) (40), 2600 (14, n ) 2) (65), 2872 (n )
2, phenyl cap) (15), 3266 (14, n ) 3) (30%), 3930 (14, n ) 4)
(11), 4596 (14, n ) 5) (6), 5259 (14, n ) 6) (3).
2-Bromo-7-(4-hydroxycumyl)-9,9-dihexylfluorene, 15.
Butyllithium (6.25 mL of 1.6 M in hexane, 10.0 mmol) was
added over 20 min to a solution of 2,7-dibromo-9,9-dihexyl-
fluorene (5.0 g, 10.16 mmol) in dry THF at -78 °C. This
solution was stirred at -78 °C for 1.5 h and then quenched
with 5 mL (excess) of dry acetone. The mixture was stirred
with saturated NH4Cl and transferred to a separatory funnel.
The organic phase was washed with water and brine and then
solvent was removed on a rotary evaporator. The residue was
dissolved in CH2Cl2, and phenol (1.9 g, 20 mmol) was added
followed by 0.5 mL of methanesulfonic acid. After stirring for
2 h at room temperature, the mixture was transferred to a
r,ω-2,2′-Bis(trimethylsilyl)-9,9-dihexylfluorene Trimer
10. A mixture of bromide 8 (9.00 g, 15.96 mmol), 2,7-bis-
(trimethyleneboronato)-9,9-dihexylfluorene (3.78 g, 7.98 mmol),
tetraethylammonium hydroxide (7.5 mL of a 40% aqueous
solution), and toluene (150 mL) was degassed with argon for
20 min, and then tetrakis(triphenylphosphine)palladium(0)
(0.075 g, 0.065 mmol) was added. The reaction mixture was
heated at 75 °C under a positive nitrogen pressure for 24 h.
The cooled mixture was diluted with 20 mL of 1 N HCl, stirred
for 10 min, and then filtered through Celite. The aqueous
phase was discarded and the organic phase was washed
successively with equal volumes of water (3×) and brine (1×).
The solution was then passed through a fritted filter contain-
ing a 5-10 mm layer of 3-(diethylenetriamino)propyl-func-
tionalized silica gel (SiliCycle) (200-400 mesh) and anhydrous
CaSO4 and stripped to afford a residue that was chromato-
graphed on silica gel (10-25% EtOAc/hexanes) to afford the
crude product, which was again purified by column chroma-
tography eluting with hexanes to afford pure 10: MS (FAB+)
1
m/e 1144 (M+); H NMR (CDCl3) δ 7.9-7.6 (m, 18, ArH), 2.0
(m(br), 12, 9-R-CH2), 1.2 and 0.8 (m(br), 66, hexyl-H), 0.3 ppm
(s, 18, Si-CH3).
r,ω-2,2′-Diiodo-9,9-dihexylfluorene Trimer 11. Tri-
methylsilylated trimer 10 (6.54 g, 5.72 mmol) was dissolved
in carbon tetrachloride (35 mL) at room temperature. The
solution was cooled to 0 °C and iodine monochloride (11.73 mL
of a 1 M solution in methylene chloride, 11.44 mmol) was
added dropwise over 15 min. The reaction was stirred vigor-
ously for another 35 min and then was allowed to warm to
ambient temperature. A concentrated solution of sodium sulfite
was added until the dark color of the solution disappeared.
The organic phase was separated, washed (3×) with water and
brine, and dried over MgSO4, and solvent was removed under
vacuum to afford a white solid residue of 11: MS (FAB+) m/e
1250 (M+); 1H NMR (CDCl3) δ 7.9-7.5 (m, 18, ArH), 2.0 (m(br),
12, 9-R-CH2), 1.18 and 0.8 ppm (m, 66, hexyl-H).
r,ω-2,2′-Bis(trimethylsilyl)-9,9-dihexylfluorene Penta-
mer 12. A mixture of diiodo compound 11 (2 g, 1.6 mmol),
borate 9 (1.70 g, 3.2 mmol), toluene (50 mL), and 2 M K2CO3
(15 mL) was degassed with argon for 20 min and then
hexaethylguanadinium chloride (40 µL, 0.075 mmol) and
tetrakis(triphenylphosphine)palladium(0) (8 mg, 0.007 mmol)
was added. The mixture was heated at 110 °C under a positive
nitrogen pressure for 24 h. The cooled mixture was diluted
with 20 mL of 1 N HCl, stirred for 10 min, and then filtered
through Celite. The aqueous phase was discarded and the
organic phase was washed successively with equal volumes
of water (3×) and brine (1×). The solution was then passed
through a fritted filter containing a 5-10 mm layer of
3-(diethylenetriamino)propyl-functionalized silica gel (Sili-
Cycle) (200-400 mesh) and anhydrous CaSO4 and stripped
to afford a residue that was chromatagraphed on silica gel
(10-25% EtOAc/hexanes) to afford pure 12 (1 g) as a white
foamy solid with a green-blue tinge: MS (MALDI-TOF) m/e
1915 (M+); 1H NMR (CDCl3) δ 8.0-7.5 (m, 30, ArH), 2.1 (m(br),
20, 9-R-CH2), 1.2 and 0.8 (m, 110, hexyl-H), 0.4 ppm (s, 18,
Si-CH3).

Macromolecules, Vol. 38, No. 26, 2005
Copolymers Derived from Telechelic Oligofluorenes 10671
Table 1. Polycarbonates Derived from Phenol-Capped
Oligofluorenes and BPABCF
separatory funnel, washed with water (3×) and brine (1×),
and then passed through a cone of anhydrous CaSO4. Evapo-
ration of solvent followed by chromatography on 150 g of silica
gel afforded bromophenol 15 as an oil (4.45 g, 80%): MS
(FAB+) m/e 546, 548 (M+); 1H NMR (CDCl3) δ 7.6-6.7 (m, 10,
ArH), 5.2 (s, 1, OH), 1.9 (m, 4, 9-RCH2’s), 1.75 (s, 6, gem-CH3),
1.2-0.5 (m, 22, aliphatics).
Cumylphenol-Functionalized Trimer 16. A mixture of
bromophenol 14 (0.460 g, 0.84 mmol), 9,9-dihexylfluorene-2,7-
bis(trimethyleneborate) (0.200 g, 0.40 mmol), tetraethylam-
monium hydroxide (0.5 g of a 40% aqueous solution), water (1
mL), and toluene (10 mL) was degassed for 20 min with argon.
Tetrakis(triphenylphosphine)palladium(0) (0.015 g, 0.014 mmol)
was added, and the mixture was stirred under a positive
nitrogen pressure and heated at 75 °C for 3 h. The cooled
mixture was stirred with 10 mL of 1 N HCl for 15 min and
then filtered through Celite. The organic phase was washed
successively with water (2 × 25 mL) and brine (1 × 25 mL)
and then passed through a 1 mm column of 3-(diethylenetri-
amino)propyl-functionalized silica gel (SiliCycle) (200-400
mesh). Evaporation of solvent afforded 0.7 g of a residue that
was chromatographed on 25 g of silica gel eluted with hexane
to 20% ethyl acetate/hexane to afford pure 15: MS (FAB+) m/e
Eg
Eg
Mw Mn
UVmax, (UV), HOMO LUMO (CV),
polymer (103) (103) Mw/Mn
eV
eV
(CV)
(CV)
eV
2a (n ) 0) 29.9 10.3 2.9
2a (n ) 1) 50.8 15.3 3.31
2a (n ) 3) 15.2 7.6 1.99
2a (n ) 5) 56.1 24.3 2.31
2b (n ) 0) 42.9 9.3 4.61
2b (n ) 1) 18.9 9.2 2.05
3.75
3.49
3.34
3.29
3.75
3.49
3.36
3.14
3.05
3.02
3.35
3.10
5.86
5.71
5.69
5.69
5.92
5.61
2.55 3.40
2.47 3.24
2.38 3.31
2.34 3.35
2.17 3.75
2.18 3.43
and then recorded the I-Vs of the stable pixels with a standard
source/measuring unit. The values reported are for the polarity
corresponding to hole injection from the PEDOT contact and
electron injection from the top contact.
For the bipolar results the structure was glass/ITO/
PEDOT:PSS followed by about a 65-nm-thick emissive poly-
mer, a vacuum deposited cathode of 4 nm of sodium fluoride
(NaF), and about 100-nm-thick aluminum.
Cyclic Voltammetry. The solution used for electrochemical
measurements was prepared by dissolving 0.1 M dry tetrabu-
tylammonium tetrafluoborate (98% min., GFS Chemicals) in
acetonitrile (HPLC grade, J.T. Baker). Acetonitrile was dis-
tilled and freeze-dried before use. All cyclic voltammetric
curves were obtained using a standard electrochemical setup
consisting of a CH Instruments model 660A potentiostat and
the three-electrode cell. The cell was placed in a glovebox
(water below 1 ppm, oxygen below 2 ppm). A platinum disk
(0.2 cm2) was polished with 1 µm alumina before every
experiment and used as a working electrode. The polymeric
films were deposited by spin coating (about 3000 rpm, spin
coater made in house) outside the glovebox. The reference
electrode, Ag/Ag+ [0.1 M AgNO3 (99.9%, Alfa-Ventron) in
acetonitrile], was calibrated vs a Fc/Fc+ (98%, Aldrich) elec-
trode (0.017 V vs Ag/Ag+). Platinum mesh was used as a
counter electrode. Voltammograms were recorded at 0.1 V/s
always starting at the open circuit potential. The procedure
to calculate HOMO and LUMO values from oxidation and
reduction potentials respectively follows ref 32.
1
1266 (M+); H NMR (CDCl3 δ 7.9-7.3 (m, 18, fluorene-ArH),
7.2 and 6.8 (AB doublets, 8, phenol ArH’s), 4.8 (s, 2, OH), 2.2
(m, 4, middle fluorene R-CH2), 2.0 (m, 8, end fluorene R-CH2),
1.75 (s, 12, cumyl-CH3), 1.2 and 1.5 (m, 66, aliphatics).
Polymers. Alternating Polycarbonates from Fluorene
Oligomer Bis-phenols and BPA-Bis-chloroformate. A
typical procedure for the preparation of perfectly alternating
oligomer-BPA copolycarbonates is as described for the prepa-
ration of the copolymer derived from BPABCF and phenol-
functional heptafluorene 1a (n ) 5). A dry reaction vessel
equipped with a magnetic stirring bar and a septum fitted with
a syringe leading to a dry nitrogen bubbler was charged with
the bis-phenol (MW ) 2600) (178.2 mg, 0.0685 mmol), BPAB-
CF (24.2 mg, 0.0685 mmol), and 1.5 mL of dry CH2Cl2. The
resulting solution was immersed in an ice-salt bath for 15 min
and then charged with 25 µL (0.179 mmol) of dry triethyl-
amine. The mixture was maintained at 0-5 °C with stirring
for 1 h, allowed to warm to room temperature, and stirred for
an additional hour. Then the mixture was diluted with 1.0 mL
of CH2Cl2, 1.0 mL of 10% NaHCO3 was added, and the mixture
was stirred for 10 min and then transferred to a separatory
funnel. The aqueous phase was discarded and the organic
phase was washed successively with equal volumes of 1 N HCl
(1×) and water (2×). The solution was concentrated to about
two-thirds of its original volume and then precipitated into
40 mL of methanol. The collected polymer was redissolved in
CH2Cl2 and this solution was added slowly to 100 mL of
boiling, deionized water. The solids were again collected, air-
dried, redissolved in fresh CH2Cl2 and reprecipitated again into
50 mL of methanol. The resulting polymer was dried at 50 °C
in a vacuum oven for 18 h. Gel permeation chromatography
(relative to polystyrene) indicated Mw ) 56.2K, Mn ) 24.3K,
Mw/Mn ) 2.31.
Results and Discussion
Synthesis of Phenol-Functionalized Oligofluo-
renes. The synthesis of discrete fluorene oligomers has
been reported by several authors.12-19 Most relevant to
this study is the divergent/convergent approach used
by Chen and co-workers to prepare a series of discrete
9,9-dialkylfluorene oligomers through a hexadecamer.17
Although this approach suited our purposes as a general
strategy for building up fluorene units, the need to link
these oligomeric segments into high molecular weight
polymers required a capping agent suitable for Suzuki
coupling to the ends of the conjugated segments (an aryl
halide or boronic acid derivative) that also bore func-
tionality for linking the oligomeric segments in a
subsequent polymerization reaction. Bromophenols 5a
and 5b, readily obtained in three steps from fluorenone
as outlined in Scheme 1, fulfilled this requirement.
Bromophenols 5a,b were also converted to the corre-
sponding pinacol borates 6a,b, either by a lithiation/
borylation sequence or by palladium-catalyzed boryla-
tion with bis(pinacolato)diboron (Scheme 1).
Other polycarbonates were prepared similarly. Pertinent
characterization data is presented in Table 1.
Device Fabrication and Measurement. For hole-only
devices, indium-tin oxide (ITO) coated glass was cleaned,
exposed to UV/ozone for 10 min, coated with about 60 nm of
poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PE-
DOT:PSS), baked in air at about 170 °C for at least 10 min,
and then coated with the copolymer by spin-coating from a 20
mg/mL solution in m-xylene. A shadow mask with eight 0.03
cm2 holes was used to vacuum-deposit top electrodes. For the
hole-only samples, about a 60-nm-thick film of gold was
evaporated at a rate of about 0.8 nm/s. For the electron-only
samples, we used aluminum as the bottom electrode and a thin
layer of sodium fluoride plus aluminum for the top electrode.
Several samples with different spin speeds were made to allow
interpolating for a target value of 50 nm. Electrical contact
was made with a thin gold wire mounted to a microprobe. For
the initial characterization we used a Textronix curve tracer

10672 Burnell et al.
Macromolecules, Vol. 38, No. 26, 2005
Scheme 1. Synthesis of Phenol-Functionalized Bromides and Boronates
With these phenol-functionalized capping agents in
hand, the synthesis of discrete phenol-capped oligomers
1a,b (n ) 0, 1) and 1a (n ) 3) (pentamer) was
accomplished as depicted in Scheme 2. Heptamer 1a (n
) 3) was prepared as outlined in Scheme 3.
After making adjustments for retained solvents, the
oligomers were pure enough in most cases to afford high
molecular weight polymers on reaction with difunctional
linking agents.
In choosing a structural motif to introduce functional-
ity into the oligofluorenes, a 4-tert-butyl phenyl group
was initially selected as the “spectator” group residing
at the 9-position that would also bear the phenol. It was
noticed rather early, however, that oligomers bearing
this “spectator” group were often contaminated with
trace amounts yellow-colored impurities that proved to
be fluorenone derivatives in which both the phenol and
the aryl group were replaced with a carbonyl. These
fluorenone contaminants were effective even in trace
amounts at quenching the natural blue fluorescence of
the oligomers and polymers derived from them, afford-
ing products that exhibited a red-shifted and less
efficient photoluminescence. Treatment of the mono-
mers or polymers with sodium borohydride restored
efficient blue emission, but the long wavelength emis-
sion returned quickly when devices fabricated with
these reduced materials were operated under ordinary
conditions. This behavior is similar to what has been
observed for polyfluorenes that have been exposed to
typical photooxidation conditions and is thought to be
due to fluorenone defects that develop when these
polymers are oxidized.33-35 Given the stability of
9-arylfluorenyl cations, anions, and radicals, loss of
phenol, phenoxide, or a phenyl radical from the 9,9-
diaryl end group of these oligomers should be facile
under the influence of heat and acidic, basic, or transi-
tion metal reagents used in synthetic operations involv-
ing these substrates. Indeed, the FAB+ mass spectra of
oligomers bearing the 9-(4-hydroxyphenyl)-9-(4-tert-
The oligomers were, in general, obtained as viscous
gums, foams, or oils after purification by silica gel
column chromatography. These materials tended to
retain solvents rather tenaciously, particularly ethyl
acetate from chromatographic purification. The purity
of the oligomers was generally assessed by liquid
chromatography, gel permeation chromatography, and
1H NMR and mass spectroscopy. The aromatic protons
1
on the end-group phenol substituents in the H NMR
spectrum appeared upfield from the rest of the aromatic
protons. Integration of these protons relative to the
hexyl protons R to the C9 fluorene position provided a
good method of purity assessment. Gel permeation
chromatographic (GPC) traces for discrete oligomers 1a
(n ) 0, 1, 3, 5) and of a statistical oligomer 14 with an
average length of nominally 10.5 fluorene units are
presented in Figure 1. For the dimer two peaks corre-
sponding to the D, L, and meso isomers are observed.
The average of the two dimer peaks and the peaks of
the trimer, pentamer, and heptamer follow a linear
trend vs the reciprocal fluorene oligomer length, which
can be used to calibrate the GPC time axis. For the
statistical oligomer, only lengths of an odd number of
fluorene units are expected according to the method of
preparation. Even though some peaks are slightly
shifted, the contribution of oligomers with three, five,
seven, and nine fluorenes are clearly resolved, and the
longer ones are visible in the initial tail.

Macromolecules, Vol. 38, No. 26, 2005
Copolymers Derived from Telechelic Oligofluorenes 10673
Scheme 2. Synthesis of Phenol-Capped Dimers 1a (n ) 0) and 1b (n ) 0), Trimers 1a,b (n ) 1), and Pentamer 1a
(n ) 3)
butylphenyl) end groups exhibited peaks at (M+ - 94)
and (M+ - 134) resulting from loss of the 9-aryl
substituents. In an effort to minimize or avoid this
fragmentation, a more electron-deficient aryl “spectator”
group, 3,5-bis(trifluoromethyl)phenyl was employed to
lessen the tendency for loss of phenol from the 9-position
by destabilizing the 9-arylfluorenyl cation. Phenol-
capped oligomers prepared with this substituent at the
9-position were typically not contaminated with yellow
fluorenone impurities. Moreover, the FAB+ mass spectra
of these oligomers did not exhibit the prominent (M -
aryl)+ signals noted before. Accordingly, this substituent
was used to prepare the majority of oligomers used in
this study. Changing the “spectator” group in the
9-position from tert-butylphenyl to bis(trifluoromethyl)-
phenyl also had a surprisingly large effect on the LUMO
values (vide infra) for the lower members of the oligomer
series. This result is somewhat surprising, since the
9-position in the oligofluorenes is remote from the
p-phenylene segments that make up the conjugated
system.
In addition to the series of discrete phenol-capped
oligomers, statistical oligomers 14 were also prepared
by condensing a stoichiometric imbalance of bis-bor-
onate and dibromide monomers in the presence of
bromophenol chain stopper 5a under typical Suzuki
polymerization conditions and varying the ratio of bis-
boronate to dibromide to achieve the desired oligomer

10674 Burnell et al.
Macromolecules, Vol. 38, No. 26, 2005
Scheme 3. Synthesis of Phenol-Capped Heptamer 1a (n ) 5)
Scheme 4. Synthesis of Phenol-Capped Statistical Oligomers
length (Scheme 4). After some optimization this proce-
dure produced oligomers that, for the most part, were
phenol-capped at both ends.
analysis of the phenolic-OH content of these oligomers
by derivatization with 1,3-phenylene phosphorochlorid-
ite in the presence of a known amount of 2,4,6-trichlo-
rophenol as an internal standard, followed by 31P NMR
analysis of the resulting mixture, enabled adjustment
of the stoichiometry in polymerizations so that high
molecular weight polymers could be obtained from these
oligomers.37 For comparative purposes, poly-9,9-dihexyl-
fluorene was also prepared by Suzuki condensation of
equimolar amounts of 9,9-dihexylfluorene-2,7-bis(tri-
methyleneborate) and 2,7-dibromo-9,9-dihexylfluorene.
Polymer Synthesis. The purified phenol-function-
alized oligomers could be converted to polymers using
a variety of linking chemistries to obtain polycarbonates,
The MALDI-TOF spectra of these oligomers (see
Figure 2, Supporting Information), however, revealed
that in addition to the expected phenol-capped oligo-
mers, a series of oligomers having phenols at one end
and phenyl caps at the other end were produced. The
phenyl end caps are likely derived from the phosphine
ligand,36 since use of tritolylphosphine in place of
triphenylphosphine, produced tolyl-capped rather than
the phenyl-capped byproducts. With some optimization
of the procedure, these monofunctional byproducts could
be minimized but not completely avoided. Quantitative

Macromolecules, Vol. 38, No. 26, 2005
Copolymers Derived from Telechelic Oligofluorenes 10675
Figure 3. HOMO and LUMO values for polymers derived
from 1a (n ) 0-5), 1b (n ) 0,1 closed squares), 14, and poly-
9,9-dihexylfluorene (closed square).
densation of these monomers in methylene chloride at
0-5 °C in the presence of a slight excess of triethyl-
amine afforded high molecular weight, colorless, trans-
parent, and highly emissive polymers with excellent
solubility and film-forming properties. As quantities of
these polymers were limited and needed for device
fabrication, thermal analysis generally was not carried
out. For the few materials that were subjected to DSC
analysis, no discernible transitions were noted in the
range from about 50 to 350 °C. Molecular weight, optical
properties, and CV data for all the polymers are
presented in Table 1.
Figure 1. Gel permeation chromatograms of discrete oligo-
mers 1a (n ) 0, 1, 3, 5) and statistical oligomer 14.
Optical absorption and cyclic voltammetry (CV) were
measured for all polymers to gauge the energy levels of
the excitons and charge carriers. The optical gap of
polymers 2a (n ) 0, 1, 3, 5) decreases with chain length,
as demonstrated in earlier works,38 and quickly ap-
proaches the value for the polymer. In addition to the
series for 2a, the figure also shows values for polycar-
bonates derived from the dimer and trimer 2b, showing
only a minor shift of the absorption onset due to this
change in substitution.
Figure 2. Optical absorption onset in methylene chloride vs
the length of the fluorene blocks. The values for the tert-butyl-
substituted dimer and trimer and for the hexylfluorene ho-
mopolymer are shown with closed symbols.
polyethers, polyformals, polysilyl ethers, and polyether
siloxanes (see Supporting Information). This versatility
was demonstrated for oligomer 1b (n ) 1) as depicted
in Scheme 5.
To assess trends in electronic properties as a function
of oligomer length, polycarbonates 2a and 2b, derived
from reaction of the phenol-capped oligomers with
BPA-bis-chloroformate (BPABCF), were prepared. Con-
Figure 3 shows the LUMO and HOMO values as
measured by CV of polymers derived from the 2a series
of oligomers with n ) 2, 3, 5, 7, the statistical n ≈ 8.5
oligomer, and those derived from oligomers 2b with tert-
butyl-substituted end groups (n ) 0, 1) and of poly-9,9-
dihexylfluorene made by Suzuki polymerization. In
general, with increasing conjugation length, the HOMO
levels typically decrease and the LUMO levels increase39
Scheme 5. Various Polymers Derived from Terfluorene 1b (n ) 1)

10676 Burnell et al.
Macromolecules, Vol. 38, No. 26, 2005
Table 2. Device Operating Characteristics (turn-on
voltage, operating voltage, and device lifetime) for
Polymers Made from Selected Phenol-Capped Oligomers
turn-on
voltage
V at
life (h) at
oligomer
10 mA/cm2
∼4 mA/cm2
1a (n ) 1)
1b (n ) 1)
15
1a (n ) 3)
1a (n ) 5)
14 (n ) 8.5)
4.9
6.5
a
a
0.1
<0.1
0.13
1.7
5.1
4.5
3.7
3.9
6.4
5.4
4.6
4.2
a Less than 1 min.
The electroluminescence of bipolar devices was in-
vestigated for all mentioned materials using ITO/
PEDOT:PSS as the anode, a vacuum-deposited cathode
of 4 nm sodium fluoride (NaF), and about 100 nm
aluminum. Similar to the behavior of hole-only devices,
the bipolar voltage also drops with conjugation length,
with the voltages about equal for the heptamer and
homopolymer. The external quantum efficiency at a
current density of 10 mA/cm2 and a target thickness of
65 nm of the emissive layer ranged from 0.12% to 0.7%
and was similar to that obtained with commercial40
samples of poly-9,9-dihexylfluorene and poly-9,9-
dioctylfluorene homopolymers. The CF3-substituted co-
polymers with n ) 3 and 7 exhibited the best efficiency,
which might simply be due to their superior purity or
to uncontrolled details of the electron hole charge
balance. For the operational stability (Table 2), a drastic
increase was observed with increasing conjugation
length. The reason for this is not fully understood but
might be due to instability of the phenolic and aryl
substituent at the 9 position of the terminal fluorene
groups.
Figure 4. The voltage at a current density of 1 mA/cm2
(diamonds) and 10 mA/cm2 (squares) for hole-only devices for
50-nm-thick copolymer films derived from oligomers with n )
1, 3, 5 and statistical oligomer 14 (n ∼ 8.5). The two dashed
lines depict the voltages measured for a poly-9,9-dihexylfluo-
rene homopolymer.
as the HOMO-LUMO difference is equal to the de-
creasing optical band gap minus the exciton binding
energy. One can see that the behavior of the HOMO
levels of the 2a copolymer series follows this expectation
for n ) 0 and 1 and then flattens, approaching the value
for the homopolymer within experimental uncertainty.
However, the LUMO values for this series significantly
decrease in magnitude, contrary to expectation. To
understand whether this trend was due to substituent
effects, we compare values for the dimer and trimer of
structure 2b. Note that due to the change in substitu-
tion, the LUMO of the dimer is shifted by almost 0.4
eV. This trend of the LUMO for series 2a may be due
to the large electron-withdrawing effect of the CF3
substituents on the terminating fluorene units. As this
substitution only affects the end groups of the fluorene
oligomers, the LUMO gradually recovers with length
and seems to approach that of the hexylfluorene ho-
mopolymer. This result also implies a significant reduc-
tion in exciton binding energy via CF3 substitution,
which can trivially be derived from the difference of
HOMO, LUMO, and optical band gap.
Motivated by the suspicion that functionality at the
9-position of the oligomer termini might affect opera-
tional stability, terfluorene 16 was prepared (see Ex-
A major motivation of this study was to understand
the effect of the conjugation length on the charge
transport properties. We suspected that for short conju-
gated segments, the sizable insulating linking groups
might impede current flow. To separate the effects for
holes and electrons, we used ITO/PEDOT bottom con-
tacts with gold top contacts for hole-only devices, and
aluminum bottom contacts with sodium fluoride/alumi-
num top contacts for electron-only devices. Figure 4
compares the saturated hole-only voltages for 1 and 10
mA/cm2 (interpolated for a thickness of 50 nm) for the
polycarbonates derived from oligomers 1a and 14, with
the values for the poly-9,9-dihexylfluorene. The voltages
drop significantly with length and starting with the
pentamer are below the values for the homopolymer.
As all the presented materials have similar hole-in-
jection barriers vs PEDOT, we attribute the increase of
hole currents with oligomer length to an increase in hole
mobility. The comparison with an analogue homopoly-
mer shows that the mobility of such copolymers can be
superior above a certain oligomer length. Electron-only
devices showed a less pronounced decrease of voltage
with length with values of about 3.5 to 5 V at 1 mA/cm2
for the copolymers and 7.5 V for the homopolymer.
perimental Section) and converted to a polymer by
reaction with BPABCF. The stability of a standard
device made with this version of a linked terfluorene
oligomer compares favorably with devices made from
1a or 1b with n ) 1, indicating that the phenol in the
9-position may be detrimental to stability.
Summary
A series of discrete (n ) 2, 3, 5, and 7) and statistical
phenol-capped fluorene oligomers have been synthe-
sized. High molecular weight polycarbonate polymers
derived from these oligomers were prepared using
BPA-bis-chloroformate (BPABCF) as a linking group.
Trends in the optical and electrical properties were
determined. Device data for this family of emissive
copolymers indicates that charge mobility increases with
conjugation length and is better than that of an ana-
logue homopolymer. We attribute this improvement to
a number of features, including reduction of accidental
chemical defects, a better-defined distribution of con-
jugation length, and possibly a better local alignment.
A surprising shift in LUMO values due to CF3 substitu-

Macromolecules, Vol. 38, No. 26, 2005
Copolymers Derived from Telechelic Oligofluorenes 10677
(17) Katsis, D.; Geng, Y. H.; Ou, J. J.; Culligan, S. W.; Trajkovska,
A.; Chen, S. H. Rothberg, L. J. Chem. Mater. 2002, 14, 1332.
(18) Geng, Y.; Katsis, D.; Culligan, S. W.; Ou, J. J.; Chen, S. H.;
Rothberg, L. J. Chem. Mater. 2002, 14, 463.
(19) Geng, Y.; Trajkovska, A.; Katsis, D.; Ou, J. J.; Culligan, S.
W.; Chen, S. H. J. Am. Chem. Soc. 2002, 124, 8337.
(20) Tirapattur, S.; Belleteˆte, M.; Drolet, N.; Leclerc, M.; Durocher,
G. Macromolecules 2002, 35, 8889.
(21) Beaupre´, S.; Ranger, M.; Leclerc, M. Macromol. Rapid
Commun. 2000, 21, 1013.
(22) Belleteˆte, M.; Ranger, M.; Beaupre´, S.; Leclerc, M.; Durocher,
G. Chem. Phys. Lett. 2000, 316, 101.
(23) Mastrorilli, P.; Nobile, C. F.; Grisorio, R.; Rizzuti, A.; Sur-
anna, G. P.; Acierno, D.; Amendola, E.; Iannelli, P. Macro-
molecules 2004, 37, 4488.
(24) Wu, C.-C.; Lin, Y.-T.; Wong, K.-T.; Chen, R. -T.; Chien, Y.-Y.
Adv. Mater. 2004, 16, 61.
tion at the ends of the oligomers was observed. This
feature may reduce the barrier for electron injection
from common cathodes for these materials. In the future
we hope this flexible approach will lead to light-emitting
devices with improved efficiency and will give greater
flexibility in tuning physical properties required for
multilayered devices.
Supporting Information Available: Preparative proce-
dures and characterization data for polymers depicted in
Scheme 5; Quantitative 13C NMR spectrum of diiodopentafluo-
rene 13 and MALDI-TOF spectrum of statistical oligomer 14.
This information is available free of charge via the Internet
at http://pubs.acs.org.
(25) Wong, K.-T.; Chien, Y.-Y.; Chen, R.-T.; Wang, C.-F.; Lin, Y.-
T.; Chiang, H.-H.; Hsieh, P.-Y.; Wu, C.-C.; Chou, C. H.; Su,
Y. O.; Lee, G.-H.; Peng, S.-M. J. Am. Chem. Soc. B 2002, 124,
11576.
(26) Kim, Y.-H.; Shin. D.-C.; Kim, S.-H.; Ko, C.-H.; Yu, H.-S.; Chae,
Y.-S.; Kwon, S.-K. Adv. Mater. 2001, 13, 1690.
(27) Wu, C.-C.; Liu, T.-L.; Lin, Y.-T.; Hung, W.-Y.; Ke, T.-H.; Wong,
K.-T.; Chao, T.-C. Appl. Phys. Lett. 2004, 85, 1172.
(28) Wu, C.-C.; Liu, T.-L.; Hung, W.-Y.; Lin, Y.-T.; Wong, K.-T.;
Chen, R.-T.; Chen, Y.-M.; Chien, Y.-Y. J. Am. Chem. Soc.
2003, 125, 3710.
(29) Coulson, D. R. Inorg. Synth. 1990, 28, 107-9.
(30) Suffert, J. J. Org. Chem. 1989, 54, 509.
(31) Hwang, S.; Chen, Y. Polymer 2000, 41, 6581.
(32) Janietz, S.; Bradley, D. C.; Grell, M.; Giebeler, C.; Inbaseka-
ran, M.; Woo, E. P. Appl. Phys. Lett. 1998, 73, 2453.
(33) Gong, X.; Iyer, P. K.; Moses, D.; Bazan, G. C.; Heeger, A. J.;
Xiao, S. S. Adv. Mater. 2003, 13, 325.
(34) Gaal, M.; List, E. J. W.; Scherf, U. Macromolecules 2003, 36,
4236.
(35) List, E. J. W.; Guentner, P. S. dF.; Scherf, U. Adv. Mater.
2002, 14, 374.
(36) Goodson, F. E.; Wallow, T. I.; Novak, B. M. J. Am. Chem.
Soc. 1997, 119, 12441.
(37) Williams, E. A.; Donahue, P. E.; Early, T. A. Unpublished
results. Details of the procedure will be published separately.
(38) Cornil, J.; Beljonne, D.; Heller, C. M.; Campbell, I. H.;
Laurich, B. K.; Smith, D. L.; Bradley, D. D. C.; Mu¨llen, K.;
Bredas, J. L. Chem. Phys. Lett. 1997, 278, 139.
(39) Gebhardt, V.; Bacher, A.; Thelakkat, M.; Stalmach, U.; Meier,
H.; Schmidt, H.-W.; Haarer, D. Adv. Mater. 1999, 11, 119.
(40) Commercial polyfluorene samples used in this study were
obtained from Cambridge Display Technologies (CDT), Cam-
bridge, UK, or American Dye Source, Inc. (ADS), Baie D’urfe,
Quebec, Canada.
References and Notes
(1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Phys. 1987, 51, 913.
(2) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks,
R. N.; MacKay, K.; Friend, R. H.; Burn, P. L.; Holmes, A. B.
Nature 1990, 347, 539.
(3) Yu, W.-L.; Pei, J.; Huang, W.; Heeger, A. J. Adv. Mater. 2002,
12 828.
(4) Setayesh, S.; Grimsdale, A. C.; Weil, T.; Enkelmann, V.;
Mu¨llen, K.; Meghdadi, F.; List, E. J. W.; Leising, G. J. Am.
Chem. Soc. 2001, 123, 926.
(5) Marsitzky, D.; Vestberg, R.; Blainey, P.; Tang, B. T.; Hawker,
C. J.; Carter, K. R. J. Am. Chem. Soc. 2001, 123, 6965.
(6) Bernius, M.; Inbasekaran, M.; Woo, E.; Wu, W.; Wujkowski,
L. Thin Solid Films 2000, 363, 55.
(7) Wu, W.; Inbasekaran, M.; Hudack, M.; Welsh, D.; Yu, W.;
Cheng, Y.; Wang, C.; Kram, S.; Tacey, M.; Bernius, M.;
Fletcher, R.; Kiszka, K.; Munger, S.; O’Brien, J. Microelectron.
J. 2004, 35, 343.
(8) Chou, C.-H.; Shu, C.-F. Macromolecules 2002, 35, 9673.
(9) Tang, H.-Z.; Fujiki, M.; Zhang, Z.-B.; Torimitsu, K.; Motonaga,
M. Chem. Commun. 2001, 2426.
(10) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc. 1990, 112,
7638.
(11) Menon, A.; Dong, H.; Niazimbetova, Z. I.; Rothberg, L. J.;
Galvin, M. E. Chem. Mater. 2002, 14, 3668.
(12) Jo, J.; Chi, C.; Hoger, S.; Wegner, G.; Yoon, D. Y. Chem. Eur.
J. 2004, 10, 2681.
(13) Chen, A. C. A.; Culligan, S. W.; Geng, Y.; Chen, S. H.; Klubek,
K. P.; Vaeth, K. M.; Tang, C. W. Adv. Mater. 2004, 16, 783.
(14) Geng, Y.; Trajkovska, A.; Culligan, S. W.; Ou, J. J.; Chen, H.
M. P.; Katsis, D.; Chen, S. H. J. Am. Chem. Soc. 2003, 125,
14032.
(15) Geng, Y.; Chen, A. C. A.; Ou, J. J.; Chen, S. H.; Klubek, K.;
Vaeth, K. M.; Tang, C. W. Chem. Mater. 2003, 15, 4352.
(16) Geng, Y.; Culligan, S. W.; Trajkovska, A.; Wallace, J. U.;
Chen, S. H. Chem. Mater. 2003, 15, 542.
MA0518353