Hyperbranched polyphenylquinoxalines from self-polymerizable AB 2 and A2B monomers

Article abstract of DOI:10.1021/ma0489467

A self-polymerizable AB₂ monomer, 2,3-bis(4-hydroxyphenyl)-6-fluoroquinoxaline, and an A₂B monomer, 2,3-bis(4-fluorophenyl)-6-(4-hydrcixyphenoxy)quinoxaline) were prepared and polymerized to afford phenol-terminated and aryl fluoride terminated, hyperbranched polyphenylquinoxalines (HPPQs), respectively. MALDI-TOF analysis showed that intramolecular cyclization was a dominant process for the low molecular weight portion during the polymerizations. After isolation and complete dryness, the phenol-terminated HPPQ 1 was only soluble in strong organic acids, while the aryl fluoride terminated HPPQ 2 was soluble in most common organic solvents. HPPQ 1 was treated with allyl bromide to afford an allyl ether terminated HPPQ 3, which was also soluble in most organic solvents. Intrinsic viscosity measurements and SEC analysis indicated that HPPQ 2 had a much higher M_w and a much broader molecular weight distribution (PDI ～ 60) than HPPQ 1 and HPPQ 3 (PDI ～ 4). The results also suggested that HPPQ 1 formed aggregates in solution and that HPPQ 2 had a much more extended and open conformation. All the HPPQs, which were highly fluorescent, had UV absorption maxima near 375 nm in THF. However, the wavelength of their emission maxima, which ranged from 424 to 466 nm, depended on their end groups.

Full text of DOI:10.1021/ma0489467

Macromolecules 2005, 38, 297-306
297
Hyperbranched Polyphenylquinoxalines from Self-Polymerizable AB2
and A2B Monomers
Jong-Beom Baek*,†
School of Chemical Engineering, Chungbuk National University, Cheongju,
Chungbuk 361-763 South Korea
Frank W. Harris*,‡
The Maurice Morton Institute and Department of Polymer Science, The University of Akron,
Akron, Ohio 44325-3909
Received May 28, 2004; Revised Manuscript Received August 20, 2004
ABSTRACT: A self-polymerizable AB2 monomer, 2,3-bis(4-hydroxyphenyl)-6-fluoroquinoxaline, and an
A2B monomer, 2,3-bis(4-fluorophenyl)-6-(4-hydroxyphenoxy)quinoxaline, were prepared and polymerized
to afford phenol-terminated and aryl fluoride terminated, hyperbranched polyphenylquinoxalines
(HPPQs), respectively. MALDI-TOF analysis showed that intramolecular cyclization was a dominant
process for the low molecular weight portion during the polymerizations. After isolation and complete
dryness, the phenol-terminated HPPQ 1 was only soluble in strong organic acids, while the aryl fluoride
terminated HPPQ 2 was soluble in most common organic solvents. HPPQ 1 was treated with allyl bromide
to afford an allyl ether terminated HPPQ 3, which was also soluble in most organic solvents. Intrinsic
viscosity measurements and SEC analysis indicated that HPPQ 2 had a much higher Mw and a much
broader molecular weight distribution (PDI ∼ 60) than HPPQ 1 and HPPQ 3 (PDI ∼ 4). The results
also suggested that HPPQ 1 formed aggregates in solution and that HPPQ 2 had a much more extended
and open conformation. All the HPPQs, which were highly fluorescent, had UV absorption maxima near
375 nm in THF. However, the wavelength of their emission maxima, which ranged from 424 to 466 nm,
depended on their end groups.
Introduction
cular weight. The PPQ obtained displayed excellent
properties.
Polyphenylquinoxalines (PPQs) are a well-known
class of high performance thermoplastics that have
many desirable properties such as good processability,
excellent tensile and adhesion properties, and high glass
transition temperatures (Tgs).1 Recently, they have
received attention for possible use in photonic applica-
tions such as electron transport layers in light emitting
diodes.2 In fact, a PPQ containing thiophene segments
has been shown to be far superior to other electron
transport layers in PPV-based light emitting devices.2c
PPQs are prepared by two different methods. The
classical method involves the condensation polymeri-
zation of aromatic tetraketones with aromatic tet-
raamines in m-cresol.3 In a more recent approach,
aromatic bisphenolate salts are polymerized with acti-
vated aromatic dihalides4 via aromatic nucleophilic
substitution reactions in polar aprotic solvents. In this
approach, a preformed quinoxaline ring is incorporated
in either the bisphenolate salt or the dihalide. This
method is currently favored because it is more versatile
and potentially less expensive than the first. As part of
an effort to further reduce cost, a self-polymerizable AB
PPQ monomer was synthesized in this laboratory.5 The
monomer contains a phenolic group and a fluorine atom
activated for nucleophilic substitution by a pyrazine
ring. The monomer, which exists as two isomers, 3-(4-
hydroxyphenyl)-2-phenyl-6-fluoroquinoxalne and 2-(4-
hydroxyphenyl)-3-phenyl-6-fluoroquinoxaline, was self-
polymerized in an NMP/toluene mixture to high mole-
The overall objective of this work was to use similar
chemistry in the preparation of hyperbranched PPQs
(HPPQs). Hyperbranched polymers are a new class of
materials that display properties distinctly different
from those of their linear analogues.6 For example,
hyperbranched systems display lower viscosities and
better solubility than their linear analogues with similar
molecular weights.6f,6i When repeating units themselves
are optically active chromophores and incorporated into
the hyperbranched polymer molecules, they also display
enhanced optical properties. For example, the fluores-
cent quantum yields of hyperbranched systems are
relatively high compared to their linear analogues,
presumably due to the efficient physical isolation of the
chromophores in polymer molecules by preventing self-
quenching.7 Dendritic system has been developed for
light-emitting devices.8 Related dendrimer systems have
been found to be very efficient in light harvesting and
energy transfer.9
The specific objectives of this work were to synthesize
the AB2 and A2B monomers 2,3-bis(4-hydroxyphenyl)-
6-fluoroquinoxaline10 and 2,3-bis(4-fluorophenyl)-6-(4-
hydroxyphenoxy)quinoxalines.10b The monomers were
to be self-polymerized to afford the corresponding
phenol-terminated and aryl fluoride terminated hyper-
branched systems. The new materials were to be
thoroughly characterized. In particular, the effects of
the terminal groups on the material solubilities, solution
viscosities, and their thermal and optical properties
were to be determined.
* To whom correspondence should be addressed.
† Telephone: +82-43-261-2489. Fax: + 82-43-262-2380. E-
mail: jbbaek@chungbuk.ac.kr.
Experimental Section
Reagents and Solvents. All chemicals, unless otherwise
mentioned, were purchased from Aldrich Chemical Inc and
‡ Telephone: + 1-330-972-7511. Fax: + 1-330-972-5704. E-
mail: fharris@uakron.edu.
10.1021/ma0489467 CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/30/2004

298 Baek and Harris
Macromolecules, Vol. 38, No. 2, 2005
were used as received. N-Methyl-2-pyrrolidinone (NMP) (Al-
drich) was distilled from phosphorus pentoxide under reduced
pressure. All other solvents were purchased from Fisher
Scientific Inc and used as received. Dithranol was purchased
from ICN Biomedicals, Inc., Aurora, OH.
CH), 5.85 (d, 1H, OH), 6.85 (dd, 4H, Ar), 7.20 (d, 2H, Ar), and
7.90 ppm (d, 2H, Ar).
1,2-Bis(4-methoxyphenyl)ethanedione (2).12 Method I.
Into a 1 L, three-necked, round-bottom flask equipped with a
reflux condenser, an overhead stirrer, and a nitrogen inlet were
placed desoxyanisoin (103.2 g, 0.4025 mol), copper(II) bromide
(200.0 g, 0.8955 mol), 300 mL of DMSO, and 300 mL of ethyl
acetate. The mixture was heated at reflux for 7 h. After the
solution was allowed to cool to room temperature, it was
poured into 3 L of ice water. Methylene chloride (1 L) was then
added. The aqueous layer was separated and discarded. The
organic layer was washed several times with water until the
aqueous layer was no longer green. The organic layer was
again separated and filtered to remove the silverlike solid that
precipitated. The solvent was removed on a rotary evaporator.
The yellow residue was dissolved in hot ethanol. The solution
was filtered to remove any insoluble solid and then allowed
to cool to room temperature to afford 101.2 g (93%) of bright
yellow needles: mp 133-134 °C (lit.12c mp 132-134 °C); FT-
IR (KBr, cm-1) 1655 (carbonyl); 1H NMR (CDCl3) δ 3.88 (s,
3H, OCH3), 6.97 (d, 4H, Ar), and 7.97 ppm (d, 4H, Ar).
Method II. A solution of anisoin (100.0 g, 0.3593 mol) in
350 mL of DMSO was stirred in a 1 L, three-necked, round-
bottom flask equipped with an overhead stirrer, a reflux
condenser, a nitrogen inlet, and an addition funnel. Hydro-
bromic acid (48%, 200 mL) was then added slowly over 1 h.
The reaction mixture was heated to 50 °C for 4-5 h. During
this period, the exothermic reaction evolved a large amount
of gas bubbles. The reaction flask was occasionally cooled in a
cold water bath. After the evolution of bubbles subsided, the
reaction mixture was slowly heated to 90 °C and maintained
at that temperature overnight. After the reaction mixture was
allowed to cool to room temperature, it was poured into a 2 L
slurry of ice and water. The precipitate that formed was
collected by filtration, washed with a large amount of water,
and dried in air. The yellow solid was dissolved in boiling
methanol or ethanol. The solution was filtered and then
allowed to cool to room temperature to give 90.4 g (91%) of
bright yellow needles: mp 133-134 °C (lit.12c 132-134 °C);
FT-IR (KBr, cm-1) 1655 (carbonyl); 1H NMR (CDCl3) δ 3.88
(s, 3H, OCH3), 6.97 (d, 4H, Ar), and 7.97 ppm (d, 4H, Ar).
5-Fluoro-2-nitroaniline (6).13 Into a 3 L, three-necked,
round-bottom flask equipped with an overhead stirrer, a reflux
condenser, an addition funnel, and a nitrogen inlet were placed
2,4-difluoronitrobenzene (100.0 g, 0.629 mol) and NMP (475
mL). After ammonium hydroxide (127 mL) was added dropwise
over 1 h, the solution was stirred for 24 h. Thin-layer
chromatography (TLC) showed that only a trace of 2,4-
difluoronitrobenzene remained. The slurry was cooled in an
ice bath and diluted to 2.5 L by dropwise addition of deoxy-
genated water. The resulting yellow solid was collected by
suction filtration and washed with deoxygenated water. The
compound was recrystallized from deoxygenated aqueous
2-propanol to give 94.2 g (96% based on 2,4-difluoronitroben-
zene) of bright yellow needles: mp 95-97 °C (lit.13 mp 97 °C);
FT-IR (KBr, cm-1) 1255 (Ar-F), 1571 (Ar-NO2), 3368 (Ar-
NH2); 1H NMR (CDCl3) δ 6.20 (s, 2H, NH2), 6.40-6.55 (m, 2H,
Ar), and 8.17 ppm (dd, 1H, Ar).
Instrumentation. Infrared (IR) spectra were obtained with
an ATI Mattson Genesis Series FT-IR 5000 spectrophotometer.
Solid samples were imbedded in KBr disks. Proton and carbon
nuclear magnetic resonance (1H NMR and 13C NMR) spectra
were obtained at 200 and 50 MHz on a Varian Gemini-200
NMR spectrometer. Fluorine nuclear magnetic resonance (19F
NMR) spectra were obtained with a Varian Gemini XL-400
nuclear magnetic spectrometer. Elemental analysis and mass
spectral analysis were performed by Galbraith Laboratories,
Inc., Knoxville, TN. Melting points (mp) were measured using
a Mel-Temp melting point apparatus and are uncorrected.
Intrinsic viscosities were determined with Cannon Ubbelohde
No. 150 and 200 viscometers. The solutions were filtered
through a 0.45 µm syringe filter prior to the measurement.
Flow times were recorded for m-cresol or methanesulfonic acid
(MSA) solutions with polymer concentrations of approximately
0.5-0.25 g/dL at 30.0 ( 0.1 °C. Differential scanning calorim-
etry (DSC) analyses were performed in nitrogen with a heating
rate of 20 °C/min using a DuPont model 2000 thermal analyzer
equipped with differential scanning calorimetry cell. The
thermograms were obtained on powder samples after they had
been heated to 350 °C and air-cooled to ambient temperature.
Glass transition temperatures (Tgs) were taken as the mid-
point of the baseline shift. Themogravimetric analyses (TGA)
were obtained in nitrogen (N2) and air atmospheres with a
heating rate of 20 °C/min using a TA Hi-Res TGA 2950
themogravimetric analyzer.
A Bruker-Franzen Analytik GMBH, MALDI time-of-fight
(TOF) mass spectrometer was employed to determine masses
using a reflection mode. Dithranol and potassium trifluoro-
acetate were used as the UV-absorbing matrix and cationizing
salt, respectively. In the case of HPPQ 1, size exclusion
chromatography (SEC) was carried out on a Waters 490E
equipped with UV detector. 0.01 mL samples of the NMP
solution were diluted with THF to 2 mL and filtered through
a 0.45 µm syringe filter before they were subjected to SEC
analysis. In the case of HPPQ 2, SEC was carried out on a
Waters 150-CV equipped with refractive index, viscosity, and
light scattering detectors. Tetrahydrofuran (THF) was used
as the elution solvent. UV-visible spectra were obtained on a
Hewlett-Packard 8435 UV-visible spectrophotometer. Photo-
luminescence measurements were performed with a Shimadzu
RF-5301PC Spectrofluorophotometer. The excitation wave-
length used was that of the UV absorption maximum of each
sample. The energy minimized structures and dihedral angles
of monomers were carried out by CS Chem 3D Std computa-
tional package (Verson 5.0, CambridgeSoft Corporation, Cam-
bridge, MA 02140).
1,2-Bis(4-methoxyphenyl)-2-hydroxyethanone (1).11 Into
a 2 L, three-necked round-bottom flask equipped with an
overhead stirrer, a reflux condenser, and a nitrogen inlet were
placed 4-methoxybenzaldehyde (209.8 g, 1.541 mol), ethanol
(1 L), and a solution of potassium cyanide (105.0 g, 1.622 mol)
in water (300 mL). The reaction mixture was heated at reflux
for 4 h. The solution changed from light yellow at the initial
stage to deep yellow at the final stage. The reaction mixture
was allowed to cool to room temperature and poured into 2 L
of water. The mixture was very slowly acidified with 200 mL
of concentrated HCl (Caution! HCN gas was evolved) in an
ice bath. Methylene chloride (500 mL) was then added to aid
in the workup. The aqueous layer was separated and dis-
carded. The organic layer was washed several times with a
large amount of water, separated again from the aqueous
layer, and then taken to dryness on the rotary evaporator. The
viscous residue was taken up in acetone, filtered, and again
taken to dryness on a rotary evaporator. The yellow oil was
dissolved in hot 95% ethanol or glacial acetic acid and then
stored in a refrigerator to obtain 140.6 g (67%) of yellow
crystals: mp 111-112 °C (lit.11d mp 111-112 °C); 1H NMR
(CDCl3) δ 4.75 (s, 3H, OCH3), 4.80 (s, 3H, OCH3), 4.58 (d, 1H,
4-Fluoro-1,2-phenylenediamine (4).14 Method I. A solu-
tion of 4-fluoro-2-nitroaniline (33.0 g, 0.211 mol) in 150 mL of
deoxygenated ethyl acetate containing 5% palladium on acti-
vated carbon (0.53 g) was placed in a hydrogenation apparatus.
Hydrogen was charged and discharged five times. The mixture
was agitated under hydrogen (55-65 psi) at room temperature
12 h. The solution was filtered through Celite to remove the
catalyst. The filtrate was reduced to dryness on a rotary
evaporator. After the gray solid residue was dissolved in hot
toluene, the solution was allowed to cool to room temperature
to give 25.5 g (96%) of gray crystals: mp 89-91 °C (lit.14 mp
89-91 °C).
Method II. The compound was prepared from 5-fluoro-2-
nitroaniline (60.0 g, 0.384 mol) using the procedure described
for method I. The gray solid was recrystallized from toluene
to give 47.5 g (98%) of white crystals: mp 89-91 °C (lit.14 mp
89-91 °C).

Macromolecules, Vol. 38, No. 2, 2005
Hyperbranched Polyphenylquinoxalines 299
4,4′-Dihydroxybenzil (3). Into a 1 L, three-necked, round-
bottom flask equipped with an overhead stirrer, a reflux
condenser, and a nitrogen inlet were placed 50.0 g (0.184 mol)
of 4,4-dimethoxybenzil, 250 mL of acetic acid, and 250 mL of
48% hydrobromic acid or 300 g of freshly prepared pyridine
hydrochloride (py‚HCl). The suspension was heated at reflux
with vigorous stirring until it became homogeneous. This
usually took about 5-6 h. The crystals that formed during
cooling was collected by filtration and washed with water to
afford 32.8 g (85%) of yellow needles: mp 254-256 °C (lit.15
The reaction mixture was then allowed to cool to room
temperature and poured into a 1 L slurry of ice and water
containing 50 mL of concentrated hydrochloric acid. The
precipitate that formed was collected by suction filtration,
washed with water, and dissolved in hot aqueous ethanol
containing charcoal. The solution was filtered and then allowed
to cool to room temperature to give 15.6 g (95% based on 4,4′-
dihydroxybenzil) of yellow crystals: mp 142 °C and 238 °C
(DSC) (lit.10a mp not reported); FT-IR (KBr, cm-1) 1209 (Ar-
F), 3259 (Ar-OH). 1H NMR (acetone-d6, ppm) δ 6.80-6.86 (d,
4H, Ar), 7.40-7.46 (dd, 4H, Ar), 7.58-7.73 (m, 2H, Ar), 8.06-
8.13 (t, 1H, Ar), and 8.83 ppm (s, 2H, OH); 13C NMR (acetone-
d6, ppm) δ 114.54, 114.97, 117.70, 121.79, 122.31, 133.21,
133.36, 133.84, 134.05, 134.24, 140.82, 156.73, 161.06, 161.24,
162.74, and 167.68. Anal. Calcd for C20H13FN2O2: C, 72.28;
H, 3.94; N, 8.43. Found: C, 72.18; H, 4.06; N, 8.29. Mass
spectrum (m/e): 332 (M+, 100% relative abundance).
2,3-Bis(4-fluorophenyl)-6-(4-hydroxyphenoxy)quinox-
aline (11). The monomer was prepared from 4,4′-difluorobenzil
(20.0 g, 81.2 mmol) and 3,4-diamino-4-hydroxydiphenyl ether
(17.6 g, 81.3 mmol) using the procedure described for 2,3-bis-
(4-hydroxyphenyl)-6-fluoroquinoxaline. The yellow product was
recrystallized from aqueous acetic acid containing charcoal to
give 28.8 g (83% based on 4,4′-difluorobenzil) of bright yellow
crystals: mp 223 °C (DSC); FT-IR (KBr, cm-1) 1229 (Ar-F),
3429 (Ar-OH). 1H NMR (DMSO-d6, ppm) δ 6.83-6.91 (d, 2H,
Ar), 7.06-7.12 (d, 2H, Ar), 7.14-7.22 (m, 5H, Ar), 7.43-7.51
(m, 4H, Ar), 7.60-7.62 (d, 1H, Ar), 8.12-8.15 (d, 1H, Ar), 9.54
(s, 1H, Ar). Anal. Calcd for C26H16F2N2O2: C, 73.23; H, 3.78;
N, 6.57. Found: C, 73.45; H, 4.35; 6.23. Mass spectrum (m/e):
426 (M+, 100% relative abundance).
Preparation of HPPQ 1 and HPPQ 2. A typical synthesis
of a HPPQ was conducted in a three-necked, round-bottom
flask equipped with an overhead stirrer, a Dean-Stark trap
with a reflux condenser, and a nitrogen inlet and outlet. The
flask was charged with monomer (20 wt %) and potassium
carbonate (20 mol % excess to hydroxyl group). The solids were
carefully washed in with a mixture of toluene and NMP. The
mixture was heated until the toluene began to reflux, and then
it was maintained at 150-160 °C until water could no longer
be observed in the Dean-Stark trap, which typically took
about 3-5 h. The dried solution was slowly heated to 180 °C
over 1 h under a strong nitrogen flow. The solution was then
heated to reflux and maintained at reflux until the solution
viscosity began to noticeably increase. In the case of HPPQ
1, the solution became elastic gels and was no longer able to
be efficiently stirred due to the large number of phenolate
salts. This usually took approximately 30-40 min at NMP
boiling temperature. The gel was added to large excess of water
containing 5 wt % hydrochloric acid (1 L) to precipitate. In
the case of HPPQ 2, the reaction mixture was diluted with
NMP, allowed to cool to room temperature, and poured into a
large quantity (1 L) of water containing 5 wt % hydrochloric
acid (1 L) to precipitate. The polymer that precipitated was
collected by filtration and air-dried overnight. The polymers
were dissolved in MSA or NMP and passed through a pres-
surized filter to remove any insoluble salts. The filtrates were
again poured into methanol/water (9/1, v/v) mixture and boiled
for several h to remove any trapped salts and Soxhlet extracted
for a week with water (HPPQ 1) and 3 days with water and
4 days with methanol (HPPQ 2). The polymers were collected
and dried at 200 °C over phosphorus pentoxide at reduced
pressure (1 mmHg) for approximately 48 h. The yields were
98%+: HPPQ 1 [η] ) 0.60 dL/g (MSA, 30 ( 0.1 °C) and Tg )
308 °C (DSC). Anal. Calcd for C20H12N2O2: C, 76.91; H, 3.87;
N, 8.97. Found: C, 71.14; H, 4.37; N, 8.04; HPPQ 2 [η] ) 1.13
dL/g (m-cresol, 30 ( 0.1 °C) and Tg ) 225 °C (DSC). Anal. Calcd
for C26H15FN2O2: C, 76.84; H, 3.72; N, 6.89. Found: C, 76.55;
H, 3.77; N, 6.74.
1
mp 248-250 °C); H NMR (acetone-d6) δ 6.97-7.01 (dd, 4H,
Ar), 7.80-7.85 (dd, 4H, Ar), and 9.65 ppm (s, 2H, OH).
1,2-Bis(4-fluorophenyl)-2-hydroxyetnanone (9). The com-
pound was prepared from 4-fluorobenzaldehyde (250.0 g, 2.014
mol) according to the procedure described for the synthesis of
4,4-dimethoxybenzoin. The yellow solid was recrystallized from
90% ethanol or aqueous acetic acid to yield 163.0 g (65%) of
yellow crystals: mp 81-83 °C (lit.16 mp 80-82 °C). FT-IR (KBr,
cm-1): 1235 (Ar-F), 1682 (carbonyl), 3478 (Ar-OH).
4,4′-Difluorobenzil (10). The compound was prepared from
4,4′-difluorobenzoin (60.0 g, 0.242 mol) using the procedure
described for the oxidation of 4,4′-dimethoxybenzoin. The
yellow solid was recrystallized from methanol or ethanol to
give 57.3 g (96%) of bright yellow flakes: mp 121.5-123 °C
(lit.17 mp 121-122 °C); FT-IR (KBr, cm-1) 1231 (Ar-F), 1670
(carbonyl). 1H NMR (CDCl3, ppm) δ 7.16-7.26 (td, 4H, Ar),
7.99-8.06 (td, 4H, Ar); 13C NMR (CDCl3, ppm) δ 116.31,
116.63, 129.41, 132.75, 132.90, 164.98, 168.78, 192.22.
3-Amino-4-nitro-4′-hydroxydiphenyl Ether (7). Into a
1 L, three-necked, round-bottom flask equipped with an
overhead stirrer, a Dean-Stark trap with a reflux condenser,
and a nitrogen inlet and outlet were placed hydroquinone
(110.1 g, 1.000 mol), potassium hydroxide (56.1 g, 1.00 mol),
75 mL of toluene, and 300 mL of DMAc. The reaction mixture
was heated at 160 °C overnight. During this period, the water
that was generated was removed as a toluene azeotrope. After
cooling to 130 °C, 5-fluoro-2-nitroaniline (40.0 g, 0.256 mol)
was added. The mixture was heated at 130 °C for 4 h with
stirring, allowed to cool to room temperature, and poured into
a 1.5 L slurry of ice water containing 100 mL of concentrated
hydrochloric acid. The precipitate that formed was collected
by filtration, washed with water, and dissolved in hot ethanol.
The solution was filtered to remove insoluble solid and then
allowed to cool to room temperature to afford 54.8 g (87% based
on 5-fluoro-2-nitroaniline) of brown crystals: mp 210-211 °C
(lit.18 mp not reported); 1H NMR (DMSO-d6, ppm) δ 6.22-6.28
(dd, 2H, Ar), 6.80-6.84 (d, 2H, Ar), 6.95-6.99 (d, 2H, Ar), 7.45
(s, 2H, NH2), 7.94-7.99 (d, 1H, Ar), and 9.54 (s, 1H, OH). Anal.
Calcd for C12H10N2O4: C, 58.54; H, 4.09; N, 11.38. Found: C,
58.33; H, 4.33; N, 11.27. Mass spectrum (m/e): 246 (M+, 100%
relative abundance).
3,4-Diamino-4′-hydroxydiphenyl Ether (8). A solution
of 3-amino-4-nitro-4′-hydroxydiphenyl ether (18.5 g, 7.51
mmol) in 150 mL of deoxygenated ethanol containing 5%
palladium on activated carbon (0.80 g) was placed in a
hydrogenation apparatus. The mixture was agitated under
hydrogen (55-65 psi) at room temperature overnight. The
solution was filtered through Celite to remove the catalyst.
Deoxygenated water was added to the filtrate, which was
stored in a refrigerator, to give 15.9 g (98%) of brown
crystals: mp 220-221 °C (lit.19 mp not reported). FT-IR (KBr,
cm-1): 1213 (Ar-O-Ar), 3347 (Ar-NH2), 3410 (Ar-OH). 1H
NMR (DMSO-d6, ppm): δ 4.81 (s, 2H, NH2), 5.08 (s, 2H, NH2),
6.20-6.42 (m, 4H, Ar), 6.61-6.74 (m, 3H, Ar), and 9.55 (s, 1H,
OH). Anal. Calcd for C12H12N2O2: C, 66.65; H, 5.59; N, 12.95.
Found: C, 66.33; H, 5.63; N, 12.77. Mass spectrum (m/e): 216
(M+, 100% relative abundance).
2,3-Bis(4-hydroxyphenyl)-6-fluoroquinoxaline (5).10 A
500 mL, round-bottom flask equipped with a magnetic stirring
bar, a reflux condenser, a Dean-Stark trap, and a nitrogen
inlet was charged with 4,4′-dihydroxybenzil (12.0 g, 4.95 mol),
4-fluoro-1,2-phenylenediamine (6.25 g, 4.96 mmol), toluene (75
mL), and deoxygenated acetic acid (250 mL). The reaction
mixture was stirred and gently heated at reflux overnight. The
water that was generated was removed as a toluene azeotrope.
Preparation of HPPQ 3. Into a 50 mL one-necked, round-
bottomed flask equipped with a magnetic stirrer and nitrogen
inlet were placed HPPQ 1 (1.0 g, 3.2 mmol, repeat unit),
potassium carbonate (1.0 g, 7.2 mmol), allyl bromide (0.5 g,
4.1 mmol), and NMP (20 mL). The reaction mixture was then
heated and maintained at 90 °C for 10 h. During this time

300 Baek and Harris
Scheme 1. Synthesis of AB2 Monomer 5a
Macromolecules, Vol. 38, No. 2, 2005
Figure 1. Energy minimized structures (CS ChemBats3D Std
V5.0): (a) 5; (b) 11.
Scheme 2. Synthesis of A2B Monomer 11a
a
Key: (a) KCN, aqueous EtOH, reflux; (b) HBr (48%),
DMSO, 50 °C; (b′) CuBr2, EtOAc/DMSO, reflux; (c) HBr (48%),
AcOH, reflux or pyridine hydrochloride (py‚HCl), reflux; (d)
H2 (65-70 psi), Pd-C, EtOAc, room temperature; (e) AcOH/
toluene, reflux.
period, the red mixture became light yellow and homogeneous.
After the solution had been allowed to cool to room tempera-
ture, it was filtered through Celite 545 to remove any insoluble
salts. The filtrate was poured into water containing 5%
hydrochloric acid (200 mL). The resulting light yellow powder
was collected by filtration and dried. The sample was precipi-
tated twice from chloroform with methanol, Soxhlet extracted
with water for 3 days and methanol for 4 days, and dried under
reduced pressure (1 mmHg) over phosphorus pentoxide at 100
°C for 48 h. The yield was essentially quantitative. [η] ) 0.31
dL/g (NMP at 30.0 ( 0.1 °C). Anal. Calcd for C23H16N2O2: C,
78.39; H, 4.58; N, 7.95; O, 9.08. Found: C, 77.94; H, 4.83; N,
7.93, O, 9.14. 1H NMR (CDCl3, δ in ppm): 4.73 (-CH2-CHd
CH2), 5.27-5.46 (-CH2-CHdCH2), 6.05 (-CH2-CHdCH2),
and 7.06-8.13 (Ar-H).
a Key: (a) NH4OH, NMP, room temperature; (b) Excess
hydroquinone, K2CO3, NMP/toluene, 140-150 °C; (c) H2 (65-
70 psi), Pd-C, EtOH, room temperature; (d) KCN, aqueous
EtOH, reflux; (e) HBr (48%), DMSO, 50 °C; (f) AcOH/toluene,
reflux.
Results and Discussion
Synthesis of Self-Polymerizable AB2 and A2B
Monomers. The AB2 monomer 2,3-bis(4-hydroxyphe-
nyl)-6-fluoroquinoxaline (5) was prepared by the se-
quence shown in Scheme 1. Thus, the benzoin conden-
sation of 4-anisaldehyde gave 4,4′-dimethoxybenzoin (1),
which was oxidized with hydrobromic acid in DMSO to
afford 4,4′-dimethoxybenzil (2). Intermediate 2 was also
prepared by the oxidation of 4-methyoxybenzyl-4′-meth-
oxyphenyl ketone (desoxyanisoin) with copper(II) bro-
mide in a DMSO/ethyl acetate mixture. 4,4′-Dimethoxy-
benzil was demethylated by treatment with 48% HBr
in acetic acid or by treatment with pyridine hydrochlo-
ride (py‚HCl) to give 4,4′-dihydroxybenzil (3). This
intermediate was treated with 4-fluoro-1,2-phenylene-
diamine (4), which was prepared by reduction of 4-fluoro-
2-nitroaniline, to give the AB2 monomer. The structure
of the monomer, which was recrystallized from aqueous
ethanol, was ascertained by FT-IR, 1H NMR, 13C NMR,
elemental analysis, and mass spectroscopy. The DSC
thermogram of 5 contained two melting endotherms
with minima at 142 and 238 °C. This behavior, which
has been observed previously with analogous AB mono-
mers, has been attributed to two different crystalline
forms. Energy minimization of a space-filling molecular
model showed that the pendent phenyl groups are
twisted with regards to each other so that they are
approximately 40° out of the plane of the quinoxaline
ring (Figure 1). This AB2 monomer was also prepared
and polymerized by Hedrick and co-workers during the
course of this research.10a
The new extended A2B monomer 2,3-bis(4-fluorophe-
nyl)-6-(4-hydroxyphenoxy)quinoxaline (11) was pre-
pared by the synthetic route shown in Scheme 2. Thus,
the benzoin condensation of 4-fluorobenzaldehyde in
aqueous ethanol in the presence of potassium cyanide
gave 4,4′-difluorobenzoin (9), which was oxidized to 4,4′-
difluorobenzil (10) with hydrobromic acid in DMSO in
over 90% yield. 3,4-Diamino-4′-hydroxydiphenyl ether
(8) was prepared by catalytic reduction of 3-amino-4-
nitro-4′-hydroxydiphenyl ether (7) with hydrogen over
palladium on activated carbon. To synthesize the in-
termediate 7, 2,4-difluoronitrobenzene was initially
treated with ammonium hydroxide in NMP to give
5-fluoro-2-nitroaniline (6) in >95% yield13 and the
compound 6 was treated with a 3 M excess of hydro-
quinone in the presence of potassium carbonate to give
3-amino-4-nitro-4′-hydroxydiphenyl ether (7).18 Conden-
sation of 8 with 10 in an acetic acid/toluene mixture
gave the A2B monomer 11.
The 19F NMR spectra of 5 and 11 are shown in Figure
2. The single absorption peak in the spectrum of 5
appears at 109.85 ppm, very close to the absorption peak
in the spectrum of the corresponding AB monomer
isomer (-109.50 ppm, Figure 2a).5g Thus, the chemical

Macromolecules, Vol. 38, No. 2, 2005
Hyperbranched Polyphenylquinoxalines 301
5 wt % hydrochloric acid to precipitate the product. A
small portion of the polymerization mixture was re-
moved, acidified, collected, air-dried, dissolved in NMP,
and diluted with THF prior to precipitation. The sample
was used for SEC and spectroscopic analysis. The
solution of HPPQ 2, which became quite viscous but
did not gel, was also added to water containing 5 wt %
hydrochloric acid to precipitate the product. Both samples
were further purified by Soxhlet extraction for a week
with water and/or methanol.
The polymerization rates of the monomers were much
faster than that of an AB analogue. High molecular
weight polymer was obtained in approximately 0.5-1
h, in refluxing NMP, while it took approximately 3-5
h for a linear polymer under similar conditions.5 This
is because the number of B functional groups (DP + 1)
per molecules increased as the polymerization pro-
ceeded.
Figure 2. 19F NMR spectra of monomers: (a) 5; (b) 11.
Initial attempts to dry HPPQ 1 at 200 °C under
reduced pressure (1 mmHg) were unsuccessful. TGA
showed that the polymer tenaciously retained ap-
proximately 5 wt % NMP (Figure 3). Samples containing
residual solvent were soluble in ether solvents such as
THF and most polar aprotic solvents. To remove the
residual NMP, the following procedure was carried out.
The polymer was reprecipitated twice from THF with
hexane. After it was extracted with water for a week, it
was dried over phosphorus pentoxide at 200 °C under
reduced pressure (1 mmHg) for approximately 48 h. The
orange polymer particles, which showed no weight
losses below 390 °C when subjected to TGA (Figure 3),
could only be swollen in organic solvents. This behavior
is in marked contrast to that of the previously described
sample of this material, which displayed solubility in
N,N-dimethylpropyleneurea (DMPU).10a It is postulated
that the sample may not have been dried sufficiently
and, thus, retained residual DMPU. This behavior is
also much different than that of the linear PPQ ana-
logue of HPPQ 1, which is soluble in chlorinated and
polar aprotic solvents. It is postulated that once all the
residual solvent was completely removed, strong inter-
and intramolecular hydrogen bonds formed between the
shift is not significantly affected by the hydroxyl group
on the phenyl substitutent in the quinoxalines 3-posi-
tion. This group cannot participate in resonance that
directly affects the 6-position. The spectrum of 11
(Figure 2b) contains two peaks at -112.93 and -113.23
ppm due to differences in the electronic environments
of the fluorine atoms on the aromatic rings in the
quinoxaline’s 2- and 3-positions. Since the fluorine atom
on the 2-aryl substituent can be deshielded by resonance
with the phenoxy substitutent at the 6-position, the
peak at -113.23 ppm can be assigned to this substitu-
tion pattern.
Self-Polymerization of the AB2 and A2B Mono-
mers. Monomers 5 and 11 were self-polymerized in an
NMP/toluene mixture in the presence of potassium
carbonate to afford the phenol-terminated HPPQ 1 and
the aryl fluoride terminated HPPQ 2, respectively
(Scheme 3). The solution of HPPQ 1 became very
viscous and then set to a dark elastic gel. The gelation
phenomenon could be related to ionic interaction be-
tween the large number of phenolate salts at chain ends.
The gel was added to a large excess of water containing
Scheme 3. Synthesis of HPPQ 1 and HPPQ 2a
a Key: (a) K2CO3, NMP/toluene, 150-160, 180, and 202°C.

302 Baek and Harris
Macromolecules, Vol. 38, No. 2, 2005
Figure 4. MALDI-TOF mass spectrum of HPPQ 2.
Figure 3. TGA thermograms of HPPQ 1 with hearing rate
of 20 °C/min.
Table 1. Properties of HPPQs
TGAc
terminal hydroxyl groups, which prevented solvation.
HPPQ 1 could only be dissolved in strong acids such
as methanesulfonic acid (MSA) and sulfuric acid at room
temperature. High molecular weight, phenol-termi-
nated, hyperbranched poly(aryl ether sulfones) are also
insoluble in organic solvents, while their aryl fluoride
terminated analogues display excellent solubility.20
The light yellow powder sample of HPPQ 2 was dried
over phosphorus pentoxide at 200 °C under reduced
pressure (1 mmHg) for approximately 48 h. Although
TGA indicated that all the NMP was removed by this
treatment, to be consistent with the handling of HPPQ
1, the sample was extracted with methanol and water
for 1 week and then dried over phosphorus pentoxide
at 200 °C for 48 h. The aryl fluoride terminated HPPQ
2 was soluble in polar aprotic solvents (DMF, DMAc,
DMSO, NMP, sulfolane), ether solvents (THF), chlori-
nated solvents (methylene chloride, chloroform), phe-
nolic solvents (m-cresol), and strong acids (sulfuric acid,
MSA).
MALDI)TOF Analysis of HPPQ 2. Considerable
research has been carried out to determine the presence
or absence of a unique focal point group in hyper-
branched polymers.21 MALDI-TOF analysis of a previ-
ously prepared sample of HPPQ 1 indicated the absence
of a fluorine at focal point.10a Thus, intramolecular ring
closure was a dominant process in the material prepa-
ration at lower molecular region. A single hydroxyl
group at the focal point of the HPPQ 2 should have also
been present if no intramolecular cyclization occurred
during its polymerization. The MALDI-TOF mass
spectrum of HPPQ 2 is dominated by one series of
peaks separated by 406 amu and containing exclusively
protonated ions due to the high basicity of the quinoxa-
line moiety of polymer. The value of 20 amu correspond-
ing to a loss of HF is lower than that expected based on
the repeat unit molecular weight up to 8000 m/z (Figure
4). This means all HPPQ 2 oligomers have undergone
intramolecular cyclization and lost their unique hy-
droxyl focal point. Acidic hydrogen atom from hydroxyl
group at focal point is reacted with potassium carbonate
producing phenolate salt and potassium bicarbonate,
which attacks one of activated DP + 1 fluorines at chain
ends forming macrocyclic oligomer and loosing focal
point. A series of [C26H15N2O2]nH+ macrocyclic oligomers
were formed with n ) 3-10. A minor series of peaks
appeared 70 amu below the major series of peaks is
proposed to originate from Na+ adducts of defective
GPCa
Mw
char
% at
b
[η]
Mn
Tg
Td5% in
polymer (dL/g) (g/mol) (g/mol)
HPPQ 1 0.60d 87100e
322000
PDI (°C) N2 (°C) 800 °C
3.70 308 511
70
68
73
HPPQ 2 1.13f 44000g 2643000 60.07 225 575
HPPQ 3 0.31h 55700g
243000 4.36 523 (He)
a Data from refractive index response. b Inflection in baseline
on DSC thermogram obtained in N2 with a heating rate of 20 °C/
min. c Temperature at which 5% weight loss occurred on TGA
thermogram obtained with a heating rate of 20 °C/min. d Intrinsic
viscosity measured in MSA at 30.0 ( 0.1 °C. e Determined in THF
containing less than 1 vol % NMP. f Intrinsic viscosity measured
in m-cresol at 30.0 ( 0.1 °C. g Determined in THF. h Determined
in NMP at 30.0 ( 0.1 °C.
oligomers, containing one monomer unit without the
4-hydroxyphenyl substituent. The formation of many
hyperbranched systems has been found to proceed with
the formation of cyclics. Since the sensitivity of the
MALDI-TOF measurements were limited to molecular
weights less than 10000 amu, the possibility that the
hyperbranched macromolecules with higher molecular
weights had a focal point cannot be ruled out.
Solution Properties of HPPQ 1 and HPPQ 2. The
intrinsic viscosities of HPPQ 1 and HPPQ 2 were 0.60
dL/g (MSA at 30.0 (0.1 °C) and 1.13 dL/g (m-cresol at
30.0 (0.1 °C), respectively (Table 1). Although a direct
comparison cannot be made due to the different solvents
used, the viscosity of HPPQ 2 was almost twice that of
HPPQ 2. While this is most likely due to a large
difference in molecular weight, this may also be an
indication of the differences in the conformations of the
polymers in solution. Intramolecular hydrogen bonding
superior to structural rigidity in HPPQ 1 and the
interaction between polymer and solvent could lead to
a compact structure, while the van der Waals interac-
tion between the terminal dipoles in HPPQ 2 inferior
to the structural rigidity of polymer chain and the
interaction between polymer and solvent could lead to
an open, solvated, extended structure. The large viscos-
ity of HPPQ 2 is unusual in that hyperbranched
systems typically display viscosities well below one.
Thermal Properties of HPPQ 1 and HPPQ 2. The
HPPQ 1 displayed a Tg of 308 °C (Figure 5a, Table 1).
This is more than 50 °C higher than that of the
analogous linear PPQ (252-256 °C).5 It is postulated
that this is due to strong inter- and intramolecular
hydrogen bonding between the hydroxyl chain ends.

Macromolecules, Vol. 38, No. 2, 2005
Hyperbranched Polyphenylquinoxalines 303
Figure 6. FT-IR (KBr) spectra: (a) HPPQ 1; (b) HPPQ 3.
Figure 5. DSC thermograms of polymers with heating rate
of 20 °C/min: (a) HPPQ 1; (b) HPPQ 2.
with FT-IR by the appearance of the absorption bands
at 2918 and 3064 cm-1 (Figure 6).
Surprisingly, the Tg was also more than 50 °C higher
than that reported for a previously prepared sample of
HPPQ 1 (255 °C),10a which is similar to that of the
linear analogue.5 This may be further evidence that the
previous sample contained residual solvent (see Figure
3), which could have been difficult to completely remove,
served as a plasticizer, and lowered the Tg. The other
reason may be due to a lower molecular weight polymer.
The Tg of HPPQ 2 was 225 °C (Figure 5b, Table 1),
which was not significantly different from that of the
analogous linear PPQ (220 °C) with the same repeating
unit.22 The much lower Tg of HPPQ 2 compared to
HPPQ 1 can be attributed to the extra flexible units in
the repeating unit, to no hydrogen bonding, and also to
its apparent extended, open conformation. The Tgs of
phenol-terminated, hyperbranched poly(aryl ether sul-
fones) are considerably higher than those of their aryl
fluoride terminated counterparts.20
Powder samples of HPPQ 1 and HPPQ 2 underwent
5% weight losses at 511 and 575 °C, respectively, when
they were subjected to TGA in N2 with heating rate of
20/min. The reduced stability of HPPQ 1 is most likely
due to the lower stability of the terminal phenol groups.
Phenol-terminated, hyperbranched poly(aryl ether sul-
fones) are also less thermally stable than their aryl
fluoride terminated counterparts.20
In contrast to its parent,HPPQ 3 was soluble in polar
aprotic solvents (DMF, DMAc, DMSO, NMP, sulfolane),
chlorinated solvents (dichloromethane, chloroform), and
ether solvents (diethyl ether, THF). It displayed an
intrinsic viscosity of 0.31 dL/g (NMP at 30 ( 0.1 °C),
which is considerably lower than that of its parent ([η]
) 0.60 dL/g in MSA at 30 ( 0.1 °C). This is most likely
due to the change in the end groups from hydroxyl
groups, which could form hydrogen bonds with the MSA
solvent and promote aggregation, to relatively nonpolar
hydrocarbon residues.
A DSC thermogram was obtained on a powder sample
of HPPQ 3 after it had been heated to 200 °C and air-
cooled to ambient temperature. The thermogram con-
tained a major exotherm with a maximum at 303 °C
(∆Hexo ) 88 J/g) and a very minor exotherm with a
maximum at 379 °C (∆Hexo1 ) 2.6 J/g). The major
exotherm is most likely due to the Claisen rearrange-
ment of the allyl ether end groups and their subsequent
polymerization. Similar behavior has been observed
with polybenzoxazoles containing pendent oxyallyl sub-
stitutents.23 TGA analysis of powder samples of HPPQ
3 showed 5% weight losses at 523 °C in helium and at
512 °C in air.
SEC Analysis of HPPQs. To obtain a solution of the
phenol-terminated HPPQ 1 for SEC analysis, an aliquot
of the NMP polymerization solution was diluted with
THF. The trace obtained showed a multimodal molec-
ular weight distribution with a weight-average molec-
ular weight (Mw) of 322000 and a polydispersity index
(PDI) of 3.70 (Figure 8a, Table 1). It appears that a
portion of the very high molecular weight fractions were
Functionalization of HPPQ 1. HPPQ 1 was treated
with allyl bromide in NMP to afford an allyl ether-
terminated HPPQ 3 (Scheme 4). Although dried HPPQ
1 was only swollen in NMP, it gradually went into
solution as the functionalization proceeded. The conver-
sion were followed with 1H NMR by the disappearance
of the hydroxyl proton absorption at 8.4-9.5 ppm and
Scheme 4. Functionalization of HPPQ 1 To Afford HPPQ 3a
a Key: (a) CH2dCHCH2-Br, K2CO3, NMP, 90 °C.

304 Baek and Harris
Macromolecules, Vol. 38, No. 2, 2005
Figure 8. GPC traces: (a) HPPQ 1 and HPPQ 3; (b) HPPQ
2 (i) refractive index response, (ii) viscosity response, and (iii)
light scattering response.
Figure 7. DSC thermograms of HPPQ 1 and HPPQ 3 with
heating rate of 20 °C/min: (a) first heating scan; (b) second
heating scan.
agglomerates, because when this polymer was function-
alized with allyl groups, most of these fractions could
not be detected. Although the dilute solution of HPPQ
1 was passed through a 0.45 µm filter prior to injection
into the SEC, it is possible that some physical ag-
gregates remained. It is also possible that the molecules
simply aggregated due to hydrogen bonding. High
molecular weight polymers that contain large amounts
of hydroxyl groups have a tendency to aggregate in poor
solvents. The Mw of the allyl ether-terminated HPPQ
3 was 243000 and the PDI was 4.36 (Figure 8a, Table
1). The functionalization not only removed the terminal
hydroxyl groups but also afforded solubility in THF.
HPPQ 2, which was very soluble in THF, was
analyzed with SEC using a refractive index detector, a
viscosity detector, and a light scattering detector (Figure
8b, Table 1). The trace obtained with the refractive
index detector (top curve, Figure 8b) showed a very
broad molecular weight distribution (PDI ∼ 60). The
number-average molecular weight (Mn) and weight-
average molecular weight (Mw) were 44000 and 2643000
g mol-1, respectively. Although a PDI of approximately
60 has been predicted for hyperbranched polymers from
AB2 monomers,24 there have been few previous reports
of PDIs approaching such a high value.25 The use of the
viscosity detector resulted in the number-average mo-
lecular weight (Mn) and weight-average molecular weight
(Mw) were 68000 and 2992000 g mol-1, respectively with
a PDI of 44. Analysis with the light scattering detector
indicated a much higher molecular weights with the
number-average molecular weight (Mn) of 1876000 g
mol-1, weight-average molecular weight (Mw) of 24388000
g mol-1, and PDI of HPPQ 2. The radius of gyration
was 14.07 nm. The narrower molecular weight distribu-
tion is most likely due to the better sensitivity of the
light scattering detector in the higher molecular weight
regions. The high molecular weight response in all these
analyses is postulated to be due to an extended molec-
ular conformation due to van der Waals interaction of
the aryl fluoride groups at the chain ends is not strong
enough to overcome the structural rigidity of polymer
chains and the interaction between polymer and solvent.
This result is opposite to that for the fluoride terminated
hyperbranched poly(ether ketone).26
Optical Properties of Monomers and HPPQs.
The monomers and polymers prepared in this study
were highly fluorescent. Monomer 5 displayed a major
UV absorption peak with a maximum at 376 nm and a
fluorescence emission peak with a maximum at 441 nm
in THF. HPPQ 1 and HPPQ 3 had absorption maxima
of 378 and 376 nm, respectively and emission maxima
of 466 and 458 nm, respectively (Figure 9a, Table 2).
The absorption maximum of the phenol-terminated
HPPQ 1 was almost identical to that of its monomer,
but its emission peak maximum was red-shifted 25 nm.
The emission maximum of HPPQ 3 was slightly blue-
shifted (7-8 nm) from its parent. The absorption and
emission maxima of monomer 11 were at 369 and 416
nm, respectively (Figure 9b, Table 2). The absorption
and the emission maxima of the HPPQ 2 prepared from

Macromolecules, Vol. 38, No. 2, 2005
Hyperbranched Polyphenylquinoxalines 305
quenching. Meanwhile, emission intensities of HPPQ
2 and HPPQ 3 were approximately twice as strong as
HPPQ 1 at the same concentration.
Surprisingly, the aryl fluoride terminated HPPQ 2
emitted a strong blue-green fluorescence when excited
at 360 nm in the solid state, while HPPQ 1 was not
active.28 This is further evidence that the hydroxyl-
terminated system is much more compact and subject
to internal aggregate excimer formation and self-
quenching due to intramolecular hydrogen bonding
between the end groups.
Conclusions
There are limited examples in the literature showing
that the polarity of surface groups on hyperbranched
polymers with similar structure is of great influence on
properties of hyperbranched polymers.26 To study the
influence of surface groups on hyperbranched polymer,
known AB2 and novel A2B monomers can be prepared
and self-polymerized via aromatic nucleophilic substi-
tute (SNAr) reactions to very high molecular weight
HPPQs. Corresponding HPPQs are fully worked-up
until no residual solvents are detected and used for
characterization. Some properties of known HPPQ 1
displays quite different Tg and solubility compared to
reported HPPQ 110a and HPPQs prepared in this
manner dramatically depend on the nature of end
groups. Thus, our work firmly suggests several points
of fundamental significance. First, it indicates that
thoughtful workup is very important to get rid of
unnecessary factors that can affect properties of new
synthesized polymer, although a lot of researchers often
ignore this point. In particular, we believe extra atten-
tion is required for dealing with polar amorphous
polymers. Their properties could be more significantly
influenced by residual impurity such as solvents. It is
noteworthy that the interaction between polar hyper-
branched polymer and polar solvent is too strong,
resulting in removal of the residual solvent entrapped
in polymer matrix being not such a simple process, and
thus one needs to pay careful attention to final workup
to purify the polymer. Second, depending on the surface
group nature of hyperbranched polymers, their proper-
ties such as Tg, degradation temperature, solution
behavior, molecular weight, molecular weight distribu-
tion, and optical properties have been also greatly
influenced, since the number [DP + (1)] of periphery
groups on hyperbranched polymer are additional major
parameters beside molecular weight, molecular weight
distribution, and the backbone nature of polymer. Last
but not least, a rare example of solid-state fluorescence
from fluoride-terminated HPPQ 2 is also demonnstrat-
ed whereas hydroxyl-terminated HPPQ 1 does not. This
fact is currently being investigated.
Figure 9. UV-absorption and emission spectra in THF
solutions: (a) 5, HPPQ 1, HPPQ 3; (b) 11, HPPQ 2.
Table 2. Optical Properties of Monomers and Polymers
compound
λaba (nm)
λema (nm)
monomer 5
monomer 11
HPPQ 1
376
369
378b
371
376
441
416
466b
424
460
HPPQ 2
HPPQ 3
a Determined in THF with a concentration of 10-4-10-6 g/L.
b Determined in THF containing 8% of HPPQ 1 solution in NMP
with a concentration of 10-3-10-5 g/L.
this monomer 11 were at 371 and 424 nm, respectively
(Figure 9b). Thus, the absorption maximum of the
polymer was also almost identical to that of its mono-
mer, while the emission maximum was red-shifted.
Interestingly, the absorption maxima of the monomers
and the polymers prepared from them occurred at
almost identical wavelengths, while the fluorescence
emission maxima varied depending on the end group.
Similarly, an emission peak shift depending on solvent
polarity was reported. The emission maxima of organic
dyes attached on dendrimer varied by solvent polarity.27
Both the absorption and emission intensities of the
HPPQs in THF were linearly dependent on the polymer
concentrations from 1.00 × 10-6 to 1.00 × 10-4 g/L. At
concentration above 1.00 × 10-4 g/L the intensity of the
fluorescence emission was off the scale of the instru-
mentation. This behavior indicates that the quinoxaline
chromophores in each repeating unit are physically well
isolated from each other and not subject to π-π electron
quenching via aggregate excimer formation and self-
Acknowledgment. This work was supported in part
by NASA-Langley under NASA Contract NAG-1-448
with Dr. Brian J. Jensen as the contract monitor. We
are grateful to professor Chrys Wesdemiotis (Depart-
ment of Chemistry, The University of Akron) for MAL-
DI-TOF analysis and detailed peak assignment.
Supporting Information Available: Color photographs
showing the sample vials containing HPPQ 1 and HPPQ 2.
This material is available free of charge via the Internet at
http://pubs.acs.org.

306 Baek and Harris
Macromolecules, Vol. 38, No. 2, 2005
(8) Freeman, A. W.; Koene, S. C.; Malenfant, P. R. L.; Thompson,
M. E.; Fre´chet, J. M. J. J. Am. Chem. Soc. 2000, 122, 12385.
(9) Adronov, A.; Gilat, S. L.; Fre´chet, J. M. J.; Ohta, K.; Neuwahl,
F. V. R.; Fleming, G. R. J. Am. Chem. Soc. 2000, 122, 1175.
(10) (a) Srinivasan, S.; Twieg, R.; Hedrick, J. L.; Hawker, C. J.
Macromolecules 1996, 29, 8543. (b) Baek, J.-B.; Harris, F.
W. Polym. Prepr. 2000, 41 (1), 157.
(11) (a) Gomberg, M.; Bachmann, W. E. J. Am. Chem. Soc. 1927,
49, 236. (b) Gomberg, M.; Bachmann, W. E. J. Am. Chem.
Soc. 1927, 49, 2584. (c) Gomberg, M.; Von-Natta, F. J. J. Am.
Chem. Soc. 1929, 51, 2239. (d) Buck, J. S.; Jenkins, S. S. J.
Am. Chem. Soc. 1929, 51, 2163.
(12) (a) Armesto, D.; Horspool, W. M.; Ortiz, M. J.; Perez-Ossorio,
R. Synthesis 1988, 799. (b) Clennan, E. L.; Speth, D. R.;
Bartlett, P. D. J. Org. Chem. 1883, 48, 1246. (c) McKillop,
A.; Ford, M. E. J. Org. Chem. 1980, 45, 3433. (d) Yonemura,
H.; Nishino, H.; Kurosawa, K. Bull. Chem. Soc. Jpn. 1986,
59, 3153. (e) Irvine; Moodie J. Chem. Soc. 1907, 91. (f) Irvine,
J. C.; Moodie, A. M. J. Chem. Soc. 1907, 536.
(13) Hodgson, H. H.; Nicholson, D. E. J. Chem. Soc. 1941, 766.
(14) Smith, Jr. W. T.; Steinle, Jr. E. C. J. Am. Chem. Soc. 1953,
75, 1292.
References and Notes
(1) Hergenrother, P. M. J. Polym. Sci., Part A1 1968, 6, 3170.
(2) (a) Lee, B. L.; Yamamoto, T. Macromolecules 1999, 32, 1375.
(b) Baek, J.-B.; Chien, L.-C. J. Polym. Sci., Part A; Polym.
Chem. 2004, 42, 3587. (c) Cui, Y.; Zhang, X.; Jenekhe, S. A.
Macromolecules 1999, 32, 3824. (d) Justin Thomas, K. R.; Lin,
J. T.; Tao, Y. T.; Chuen, C.-H. Chem. Mater. 2002, 14, 2796.
(e) Thlelakkat, M. M.; Posch, P.; Schmidt, H.-W. Macromol-
ecules 2001, 34, 7441. (f) Jonforsen, M.; Johansson, T.;
Ingana¨s, O.; Andersson, M. R. Macromolecules 2002, 35,
1638. (g) Zhan, X.; Liu, Y.; Wu, X.; Wang, S.; Zhu, D.
Macromolecules 2002, 35, 2529. (h) Bangcuyo, C. G.; Ellsworth,
J. M.; Evans, U.; Myrick, M. L.; Bunz, U. H. F. Macromol-
ecules 2003, 36, 546.
(3) Hergenrother, P. M.; Levine, H. H. J. Polym. Sci., Part A1
1967, 5, 1453.
(4) (a) Bass, R. G.; Waldbauer, Jr. R. O.; Hergenrother, P. M.
Polym. Prepr. 1988, 29 (1), 292. (b) Connell, J. W.; Hergen-
rother, P. M. Polym. Prepr. 1988, 29 (1), 172. (c) Connell, J.
W.; Hergenrother, P. M. Polymer 1992, 33, 3739. (c) Labadie,
J. W.; Hedrick, J. L.; Hofer, C. D. Polym. Prepr. 1987, 28 (1),
69. (d) Hedrick, J. L.; Labadie, J. W. Macromolecules 1988,
21, 1883. (e) Hedrick, J. L.; Labadie, J. W. Macromolecules
1990, 23, 1561. (f) Hedrick, J. L.; Labadie, J. W.; Matray, T.;
Carter, K. Macromolecules 1993, 26, 4833.
(5) (a) Harris, F. W.; Korleski, J. E. Polym. Mater. Sci. Eng. 1989,
61, 870. (b) Harris, F. W.; Korleski, J. E. US Patent 5,030,-
704, 1991. (c) Kim, B. S.; Korleski, J. E.; Zhang, Y.; Klein, D.
J.; Harris, F. W. Polymer 1999, 40, 4553. (d) Baek, J.-B.;
Harris, F. W. Polym. Prepr. 1999, 40 (2), 882. (e) Klein, D.
J.; Baek, J.-B.; Harris, F. W. Polym. Prepr. 1999, 40 (2), 886.
(f) Ooi, I. H.; Hergenrother, P. M.; Harris, F. W. Polymer
2000, 41, 5095. (g) Klein, D. J.; Modarelli, D. A.; Harris, F.
W. Macromolecules 2001, 34, 2427.
(15) Schonberg, A.; Kramer, O. Chem. Ber. 1992, 55, 1174.
(16) Miyashita, A.; Suzuki, Y.; Iwamoto, K. I.; Higashino, T. Chem.
Pharm. Bull. 1994, 42, 2633.
(17) Seko, N.; Yoshino, K.; Yokota, K.; Tsukamoto, G. Chem.
Pharm. Bull. 1991, 39, 651.
(18) Loewe, H.; Urbanietz, J.; Kirsch, R.; Duewel, D. Ger. Offen.
DE 2,443,297, 1985.
(19) Loewe, H.; Urbanietz, J.; Kirsch, R.; Duewel, D. Ger. Offen.
DE 2,164,690, 1979.
(20) Mart´ınez, C. A.; Hay, A. S. J. Polym. Sci., Part A: Polym.
Chem. 1997, 35, 2015.
(21) (a) Percec, V.; Chu, P.; Kawasumi, M. Macromolecules 1994,
27, 4441. (b) Hawker, C. J.; Chu, F. Macromolecules 1996,
29, 4370.
(6) (a) Majoral, J. P.; Caminade, A. M. Chem. Rev. 1999, 99, 845.
(b) Kim, Y. H. J. Polym. Sci., Part A: Polym. Chem. 1998,
36, 1685. (c) Matthews, O. A.; Shipway, A. N.; Stoddart, J.
F. Prog. Polym. Sci. 1998, 23, 1. (d) Malmstrom, E.; Hult, A.
J. Macromol. Sci.sRev. Macromol. Chem. Phys. 1997, 37,
555. (e) Bochkarev, M. N.; Katkova, M. A. Russ. Chem. Rev.
1995, 64, 1035. (f) Fre´chet, J. M. J. Science 1994, 263, 1710.
(g) Hawker, C. J.; Fre´chet, J. M. J. Macromolecules 1990,
23, 4726. (h) Wooley, K. L.; Hawker, C. J.; Fre´chet, J. M. J.
Angew. Chem., Int. Ed. Engl. 1994, 33, 82. (i) Tomalia, D.
A.; Durst, H. D. Top. Curr. Chem. 1993, 165, 194 and
references therein. (j) Tomalia, D. A.; Hedstrand, D. M.;
Wilson, L. R. Encyclopedia of Polymer Science & Engineering,
2nd ed.; Wiley and Sons Inc.: New York, 1990; p 46. (k)
Burchard, W. Adv. Polym. Sci. 1983, 48, 1. (l) Turner, S. R.;
Voit, B. I.; Mourey, T. H. Macromolecules 1993, 26, 4617. (j)
Daoud, M.; Family, F.; Jannink, G. J. Phys. Lett. 1984, 45,
L199.
(22) (a) Baek, J.-B.; Harris, F. W.Polym. Prepr. 2001, 42 (1), 566.
(b) Baek, J.-B.; Harris, F. W. J. Polym. Sci., Polym. Chem.
2005, 43, 78.
(23) Dang, T. D.; Hudson, L. S.; Feld, W. A.; Arnold, F. E. Polym.
Prepr. 2000, 41 (1), 103.
(24) (a) Flory, P. J. J. Am. Chem. Soc. 1952, 74, 2718. (b)
Burchard, W. Macromolecules 1972, 5, 604.
(25) Turner, S. R.; Voit, B. I.; Mourey, T. H. Macromolecules 1993,
26, 4617.
(26) Chu, F.; Hawker, C. J. Polym. Bull. (Berlin) 1993, 30, 265.
(27) (a) Mishra, A.; Behera, R. K.; Behera, P. K.; Mishra, B. K.;
Behera, G. B. Chem. Rev. 2000, 100, 1973 and references
therein. (b) Buss, C. E.; Mann, K. R. J. Am. Chem. Soc. 2002,
124, 1031. (c) Furukawa, K. Acc. Chem. Res. 2003, 36, 102.
(d) Cheng, Q.; Yamamoto, M.; Stevens, R. C. Langmuir 2000,
16, 5333. (e) Gonza´les-Ronda, L.; Martin, D. C. Macromol-
ecules 1999, 32, 4558. (f) Cleij, T. J.; Jenneskens, L. W. J.
Phys. Chem. B 2000, 104, 2237.
(7) (a) Peng, H.; Cheng, L.; Luo, J.; Xu, K.; Sun, Q.; Dong, Y.;
Salhi, F.; Lee, P. P. S.; Chen, J.; Tang, B. Z. Macromolecules
2002, 35, 5349. (b) Li, B.; Bo, Z. Macromolecules 2004, 37,
2013. (c) Adronov, A.; Robello, D. R.; Fre´chet, J. M. J. J.
Polym. Sci., Part A: Polym. Chem. 2001, 39, 1366.
(28) Color photographs showing the sample vials containing
HPPQ 1 and HPPQ 2 are provided as Supporting Informa-
tion.
MA0489467