A new hyperbranched poly(arylene-ether-ketone-imide): Synthesis, chain-end functionalization, and blending with a bis(maleimide)

Article abstract of DOI:10.1021/ma020066+

While aromatic polyimides have found widespread use as high-performance polymers, the present work addressed the need for organosoluble pre-imidized materials through the use of a hyperbranching scheme. The AB₂ monomer, N-[3,5-bis(4-hydroxybenz

Full text of DOI:10.1021/ma020066+

Macromolecules 2002, 35, 4951-4959
4951
A New Hyperbranched Poly(arylene-ether-ketone-imide): Synthesis,
Chain-End Functionalization, and Blending with a Bis(maleimide)
J on g-Beom Ba ek ,† Ha ih u Qin ,‡ P a tr ick T. Ma th er ,‡ a n d Loon -Sen g Ta n *,§
University of Dayton Research Institute, 300 College Park, Dayton, Ohio 45469; University of
Connecticut, Institute of Materials Science, 97 N. Eagleville Rd., U-136, Storrs, Connecticut 06269;
and Polymer Branch, AFRL/ MLBP, Materials & Manufacturing Directorate, Air Force Research
Laboratory, Wright-Patterson Air Force Base, Ohio 45433-7750
Received J anuary 14, 2002; Revised Manuscript Received April 16, 2002
ABSTRACT: While aromatic polyimides have found widespread use as high-performance polymers, the
present work addressed the need for organosoluble pre-imidized materials through the use of a
hyperbranching scheme. The AB2 monomer, N-[3,5-bis(4-hydroxybenzoyl)benzene]-4-fluorophthalimide,
was prepared from 4-fluoroisophthalic anhydride and 3,5-bis(4-hydroxybenzoyl)aniline. The latter was
synthesized in three steps starting from commercially available 5-nitroisophthalic acid. The AB2 monomer
was then polymerized via aromatic fluoride-displacement reaction to afford the corresponding hydroxyl-
terminated hyperbranched polymer, HT-PAEKI, which was then functionalized with allyl and propargyl
bromides as well as epichlorohydrin to afford allyl-terminated AT-PAEKI, propargyl-terminated PT-
PAEKI, and epoxy (glycidyl)-terminated ET-PAEKI, in that order. All hyperbranched poly(ether-ketone-
imide)s were soluble in common organic solvents. Intrinsic viscosities of HT-, AT-, PT-, and ET-PAEKI
in NMP were 0.13, 0.08, 0.08, and 0.08 dL/g, in that order. AT-PAEKI displayed an exotherm due to
Claisen rearrangement at 269 °C and allyl-based thermal-cure reaction at 343 °C. PT-PAEKI displayed
only a single, strong exotherm at 278 °C. Because of hydrogen bonding, HT-PAEKI displayed Tg of 224
°C while its derivatives exhibited lower Tg values ranging from 122 to 174 °C. Finally, AT-PAEKI was
blended with a bisphenol A-based bis(maleimide) (BPA-BMI) in various weight ratios. The results from
differential scanning calorimetric study indicated that the presence of AT-PAEKI (up to 32 wt %)
significantly affect the glass transition temperatures and cure behavior of BPA-BMI. Dynamic mechanical
analysis comparing cured BPA-BMI with the 5 wt % AT-PAEKI blend corroborates this increase in glass
transition temperature.
In tr od u ction
important advantages such as better solubility relative
to their linear counterparts, and easier syntheses than
their analogous dendrimers, which invariably involve
tedious, multistep synthetic schemes. Therefore, larger
quantities of hyperbranched polymers can be easily
produced from ABx (x g 2) monomers to facilitate wider
evaluation of the materials for their potential applica-
tions and commercialization.
Aromatic polyimides (PI’s) are well-known, high-
performance materials with widespread applications in
the aerospace and electronics industries due to their
excellent thermomechanical and dielectric properties.1
Recently, it was demonstrated that they are also useful
as optical materials based on their optical anisotropy
when cast in directions parallel (in-plane) and perpen-
dicular (out-of-plane) to the film surface.2 However,
when fully imidized, most aromatic PI’s have limited
solubility in common organic solvents, thus restricting
the choice in their processing options. Therefore, nu-
merous research efforts have been focused on organo-
soluble PI’s from the modification of the structure (a)
without substantially decreasing rigidity of their back-
bone,3,4 (b) to allow processing polymers with preformed
imide units, and (c) to avoid many problems associated
with handling poly(amic acid) (PAA) precursors.5,6
Furthermore, postpolymerization reactions of soluble
aromatic polyimides under homogeneous conditions
would also allow better control in the introduction of
desirable functional groups.
In comparison with other classes of wholly aromatic,
hyperbranched polymers,8 there are only few reports on
the syntheses of hyperbranched PI’s9-12 and their
utilization. The main reason is that amine and anhy-
dride functions are chemically incompatible to be present
in the same molecule. However, to circumvent this
problem, approaches such as A2 + B3 polymerization11
and preparation of hyperbranched poly(amic acid meth-
yl ester) followed by intramolecular imidization10 have
been successful. Another approach is to generate an ABx
monomer containing a preformed aromatic imide as a
chemically inert moiety together with A and B functions
that can react quantitatively under suitable conditions.
This approach is exemplified by the work of Moore et
al.,9 who developed a rapid, CsF-mediated synthesis of
aromatic hyperbranched poly(ether imide)s from AB2
monomers containing tert-butyldimethylsilyl ether moi-
eties. Wu and Shu have recently reported similar
work.12
Another viable alternative to attaining solubility in
aromatic PI’s is to change the traditional, linear geom-
etry of the macromolecules to three-dimensional, highly
branched (dendritic) architecture.7 As a subset of den-
dritic polymers, hyperbranched polymers have several
As part of our research program to explore and
develop niche applications for aromatic hyperbranched
polymers,13 we describe here the results on (i) the
synthesis of a new AB2 monomer containing a preformed
aromatic imide moiety and para-carbonyl functions to
* Corresponding author. E-mail: Loon-Seng.Tan@wpafb.af.mil.
† University of Dayton Research Institute.
‡ University of Connecticut.
§ Wright-Patterson Air Force Base.
10.1021/ma020066+ CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/18/2002

4952 Baek et al.
Macromolecules, Vol. 35, No. 13, 2002
Sch em e 1. Syn th esis of AB2 Mon om er (5)
a . SOCl2, reflux; b. AlCl3, anisole, rt; c. Py.HCl, reflux; d . H2 (60-65 psi), EtOH, rt; e. isoquinoline, NMP, reflux.
facilitate the formation of phenolate nucleophiles using
low-cost potassium carbonate, (ii) functionalization of
the resulting hyperbranched poly(ether-ketone-imide)
with reactive chain ends, and (iii) a preliminary thermal
study of the allyl ether-terminated hyperbranched
polymer derivative. The latter effort is to set the stage
for further evaluation of such reactive hyperbranched
polymer derivatives for uses as toughening additives14
for high-performance thermosets.
Thus, samples were withdrawn from the polymerization
mixture 10 min after the oil bath temperature had
reached the following temperatures: 180, 190, and 202
°C (NMP reflux temperature), in that order. The samples
(∼1-2 drops each from disposable pipets) were diluted
with about 2 mL of tetrahydrofuran (THF) before they
were subjected to GPC experiments. As shown in Figure
1, when the polymerization temperature exceeded 180
°C, the left shoulders on GPC curves were growing,
suggesting that the high-molecular-weight portion was
approaching a limit of approximately 2.2 million Da
with reference to polystyrene standards. Although the
high-molecular-weight portion could have arisen from
(a) the early stage of a cross-linking reaction initiated
by the attack of the imide function by the phenolate or
(b) the aggregation of the hyperbranched polymer salt
in THF/NMP mixture, we rule out possibility (a) because
the hyperbranched polymers were completely soluble in
NMP after acidic workup. Nevertheless, the peak value
of the maximum for each curve remained almost con-
stant at approximately 7000 Da. Thus, to control the
molecular weight and polydispersity, the polymerization
process was optimized at the reaction temperature of
160 °C and for a duration of 30 min. The resulting
hyperbranched polymer, HT-PAEKI (6), was isolated as
an off-white product that was quite soluble in most
common organic solvents such as polar aprotic solvents
(NMP, DMSO, DMF, DMAc, etc.), ether solvents (THF,
diethyl ether), and phenolic solvents (m-cresol, phenol),
displayed an intrinsic viscosity of 0.13 dL/g (NMP at
30 ( 0.1 °C), Mn of 2400 g/mol, Mw of 4800 g/mol, and
molecular weight distribution (MWD) of 2.0. Its glass
transition temperature (Tg) was determined to be 224
°C by differential scanning calorimetry (DSC). Although
we have not determined the degree of branching (DB)
for this hyperbranched polymer, we believe that its DB
should be within the range defined by similar hyper-
branched ether-imide polymers prepared by Moore et
al. (DB ∼ 67%)9 and by Wu and Shu (DB ∼ 50%).12
Resu lts a n d Discu ssion
Mon om er Syn th esis. The AB2 monomer (5) was
synthesized according to Scheme 1.
The sequence started with 5-nitroisophthalic acid
which was treated with thionyl chloride to afford
5-nitroisophthaloyl dichloride (1). Friedel-Crafts reac-
tion of 1 with anisole in the presence of aluminum
chloride yielded 3,5-bis(4-methoxybenzoyl)nitrobenzene
(2), which was subsequently demethylated with pyridine
hydrochloride to afford 3,5-bis(4-hydroxybenzoyl)ni-
trobenzene (3). Compound 3 was then reduced to 3,5-
bis(4-hydroxybenzoyl)aniline (4). Upon reacting with
4-fluoroisophthalic anhydride with catalytic amount of
isoquinoline, 4 was converted to the desired monomer
5, N-[3,5-bis(4-hydroxybenzoyl)benzene]-4-fluoroisoph-
thalimide. The identity and purity of 5 were ascertained
by conventional organic characterization prior to po-
lymerization experiments, including FT-IR, NMR, HPLC,
elemental analysis, and mass analysis.
P olym er iza tion . The monomer 5 was self-polymer-
ized in an N-methylpyrrolidinone (NMP)/toluene mix-
ture of varying ratio in the presence of potassium
carbonate to afford the hydroxyl-terminated hyper-
branched poly(arylene-ether-ketone-imide) (HT-PAE-
KI) after an acidic workup (Scheme 2). We observed that
as self-polymerization of 5 was progressing, the polym-
erization mixture set to a gellike state, most likely due
to the large number of phenolate salts at the ends,
causing partial insolubility of the growing polymer
chains. Therefore, we followed the polymerization kinet-
ics with the aid of gel-permeation (size-exclusion) chro-
matography (GPC) in terms of the semiquantitative
changes in the molecular weights and their distribution.
F u n ction a liza tion of HT-P AEKI. Hyperbranched
polymer 6 with hydroxyl end groups was treated with
allyl bromide, propargyl bromide, or epichlorohydrin to
afford allyl-terminated (AT-PAEKI, 7), propargyl-

Macromolecules, Vol. 35, No. 13, 2002
Hyperbranched Poly(arylene-ether-ketone-imide) 4953
Sch em e 2. P olym er iza tion of AB2 Mon om er a n d F u n ction a liza tion of th e Resu ltin g Hyp er br a n ch ed P olym er
a . K2CO3, NMP; b. H3O+; c. excess allyl bromide, K2CO3, NMP, 80-90 °C; d . excess propargyl bromide, K2CO3, NMP, 80-90
°C; e. excess epichlorohydrin, K2CO3, NMP, 80-90 °C.
F igu r e 2. 1H NMR (DMSO-d6) spectra of (1) HT-PAEKI (OH-
terminated), (2) AT-PAEKI (allyl-terminated), (3) PT-PAEKI
F igu r e 1. Influence of polymerization temperature of polymer
(propargyl-terminated), and (4) ET-PAEKI (epoxy-terminated)
molecular weight and distribution using size-exclusion liquid
(started from bottom). Proton residues from the solvent are
chromatography, indicating that higher molecular weight
denoted by asterisks.
portion increases with higher reaction temperature (polysty-
rene standards were used to estimate the molecular weights).
Th er m a l P r op er ties. The glass-transition temper-
terminated (PT-PAEKI, 8), and glycidyl (epoxy)-termi-
nated (ET-PAEKI, 9) poly(aryl-ether-ketone-imides),
in that order (Scheme 2). From their proton NMR (see
Figure 2) and FT-IR spectra, all three conversions were
essentially quantitative as evidenced by the absence of
the signal (δ 10.55-10.84 ppm) characteristic of the
hydroxyl group after the post-polymerization reactions.
Solu bility a n d Solu tion P r op er ties. All these
functionalized polymers were soluble in polar aprotic
solvents (DMF, DMAc, DMSO, NMP, sulfolane) and an
ether solvent (THF). In general, hyperbranched poly-
mers have greater solubility than their linear analogues
in a given solvent. The intrinsic viscosity values of AT-
PAEKI, PT-PAEKI, and ET-PAEKI were all found to
be 0.08 dL/g (NMP at 30 ( 0.1 °C). These values are
considerably lower than that of their parent hyper-
branched polymer (hydroxyl-terminated). This is most
likely due to the presence of intra- and intermolecular
hydrogen bonding in the latter.
atures of hyperbranched polymers were determined
using the DSC method. The thermograms were gener-
ated from their powder samples after they had been
previously heated to 200 °C and air-cooled to ambient
temperature. The Tg was defined as the midpoint of the
maximum baseline shift from the second run. As shown
in Table 1 and Figure 3, the HT-PAEKI displayed a Tg
at 225 °C. AT-PAEKI showed a Tg at 122 °C and two
exotherms with peak temperatures at 269 and 343 °C.
The first exotherm stemmed from the Claisen rear-
rangement reaction,15 during which o-allylphenols were
formed from the allyl phenyl ether end groups (see
Scheme 3b). The latter exotherm was ascribable to the
thermal reaction cure via the allyl groups. Similar DSC
observation was reported for fluorinated polybenzox-
azoles with allyl ether pendants.16 PT-PAEKI exhibited
a Tg at 134 °C and an exotherm with the onset
temperature about 200 °C and peak temperature at 278
°C. This exotherm (∆Hexo ) 451 J /g) is attributable to

4954 Baek et al.
Macromolecules, Vol. 35, No. 13, 2002
Ta ble 1. Th er m a l P r op er ties of Hyp er br a n ch ed P oly(a r ylen e-eth er -k eton e-im id e) a n d th e F u n ction a lized Der iva tives
TGA
in helium
in air
Td5%c (°C)
Td5%c (°C)
char (%)
char (%)
at 850 °C
b
PAEKI
[η]a (dL/g)
Tg (°C)
Texo1 (°C)
∆H1 (mJ )
Texo2 (°C)
∆H2 (mJ )
in
at 850 °C
in
HT-
AT-
PT-
ET-
0.13
0.08
0.08
0.08
225
122
134
174
411
431
422
330
55
58
50
46
416
408
412
318
0.6
0.1
3.3
1.7
269
98.2
343
278
350
114.7
450.5
294.2
a
b
Intrinsic viscosity measured in NMP at 30.0 ( 0.1 °C. Inflection in baseline on DSC thermogram obtained in N2 with a heating rate
of °C/min. c Temperature at which 5% weight loss occurred on TGA thermogram obtained with a heating rate of 10 °C/min.
In the case of HT-PAEKI, the lower degradation (weight
loss) temperature in helium than that in air was
probably due to the presence of a large number of
hydroxyl groups at the chain ends, thus making the
material hygroscopic and capable to absorb moisture (up
to ∼2%) between 25 and 100 °C when its TGA experi-
ment was conducted in air. Among the polymer deriva-
tives, the propargyl-terminated one was the most ther-
mally stable and the epoxy-terminated the least (Figure
4).
AT-P AEKI/BP A-BMI Blen d s. BMI resins18 are at-
tractive in numerous aerospace applications because of
their easy processing similar to epoxies but higher use-
temperature capability than the latter and relatively
low cost. However, their main drawbacks are brittleness
and proclivity to microcrack under stresses in cross-ply
composites.19 Therefore, many approaches have been
reported in the literature to address these problems: (i)
chain extension via Michael addition or cycloaddition
reactions (ene and Diels-Alder reactions) with amine-
terminated oligomers and thermoplastics,20 diallyl com-
pounds,21 or bis(benzocyclobutene) compounds;22 (ii)
reactive liquid rubbers;23-25 (iii) thermoplastics pow-
ders;26 (iv) engineering thermoplastics and their deriva-
tives (with reactive pendants or end groups);27 (v)
hyperbranched aliphatic polyester.28 The challenge of
all of these approaches is to inhibit microcracking while
maintaining high-temperature utility.
In our blend studies, we selected a bisphenol A-based
bis(maleimide) (BPA-BMI)29 as the thermoset compo-
nent primarily because it has much lower melting
temperature (84 °C, DSC) than more commonly used
methylenedianiline (MDA)-based BMI (mp 162 °C,
DSC). Among the three functionalized hyperbranched
polymers synthesized, the allyl-terminated AT-PAEKI
was chosen on the basis of the consideration that there
F igu r e 3. DSC thermograms of (a) HT-PAEKI; (b) (i) AT-
PAEKI, (ii) PT-PAEKI, and (iii) ET-PAEKI.
is a large number of allyl groups at its chain ends, and
the ene reaction between diallyl and bis(maleimide)
groups is well established.21 In fact, we found that AT-
PAEKI was soluble in molten BPA-BMI. To carry out
this study at the lab scale, however, it was more
convenient to prepare the intimate mixtures of BPA-
BMI and AT-PAEKI in THF.
The DSC scans of virgin BPA-BMI showed a Tg at
70.5 °C, a Tm at 84 °C, and a polymerization exotherm
with peak temperature (Texo) at 200 °C and 173.5 J /g
for the heat of reaction (∆Hexo); see Table 2. No initial
Tg was observed for all AT-PAEKI/BPA-BMI blends
except the one with AT-PAEKI content of 32 wt %. The
latter has a single initial Tg that was 22 °C higher than
that of pure AT-PAEKI. This indicates that AT-PAEKI
and BPA-BMI form a compatible blend. Furthermore,
we suspect that the origin of compatibility goes beyond
the presence of imide functions in both components
(physical interactions). Since maleimide is a strong
the formation of chromene from the propargyl group
followed by the curing of the latter.17 ET-PAEKI showed
a Tg at 174 °C and an exotherm (∆Hexo ) 294 J /g) with
onset and peak temperatures at 300 and 350 °C,
respectively. This exotherm is likely the result of the
ring-opening of the epoxy functions and subsequent
reactions as well as decomposition as indicated by some
weight loss in its TGA (helium) around 350 °C. It is
noteworthy that the sum of heat of reaction (∆Hexo
)
detected for AT-PAEKI is appreciably smaller than the
values for PT-PAEKI and ET-PAEKI. We suspect that
the thermal conversion of Claisen rearrangement in AT-
PAEKI might not be quantitative and that the loss of
some allyl groups might have occurred at such high
temperatures.
From the powder samples of HT-PAEKI, AT-PAEKI,
PT-PAEKI, and ET-PAEKI, temperatures at which a
5% weight loss was observed were in the range of 330-
431 °C in helium and 318-416 °C in air, respectively.

Macromolecules, Vol. 35, No. 13, 2002
Hyperbranched Poly(arylene-ether-ketone-imide) 4955
Sch em e 3. (a ) En e Rea ction betw een a n Allylp h en ol a n d a Ma leim id e; (b) Cla isen Rea r r a n gem en t Rea ction of a n
Allyl Eth er ; (c) P r op osed Th er m a l Rea ction s Occu r r in g in BP A-BMI a n d AT-P AEKI Blen d sa
a
Temperatures indicated in parentheses are the approximate onset and peak values.
dienophile (electron-deficient) and the O-allyl group is
a relatively electron-rich olefin, the donor-acceptor
interactions are very likely. In addition, it is known that
ene reaction (see Scheme 3a) takes place in the tem-
perature range 130-150 °C.21a Hence, in the 32 wt %
AT-PAEKI/BPA-BMI blend, a sufficient amount of the
adduct must have formed from the ene reaction to allow
the detection of initial Tg, which is higher than those of
both components. One would expect the initial Tg of a
physically compatible blend to be somewhere in between
those of the components.
homopolymerization of BPA-BMI (see Figure 5). The
total heats of reaction (∆Hexo) were in the range 143-
157 J /g with a gradual increase as AT-PAEKI content
increased. On closer examination of the thermograms,
the exotherms of the blends with AT-PAEKI content of
2-16 wt % were bimodal, suggesting the overlap of two
or more thermal reactions. In addition, when the sample
with largest content of AT-PAEKI (32 wt %) was heated
to 400 °C, the characteristic exotherms associated with
the thermal reactions, namely, Claisen rearrangement
(∼269 °C) and allyl radical cure (∼343 °C) of allyl ether
groups, were not detected. Since the initial Tg of this
blend was significantly increased (∆ ) 73 °C) in
comparison to the neat BPA-BMI, this may have such
a delaying effect kinetically on the onsets of both the
As the content of AT-PAEKI increased from 2 to 32
wt %, the exotherm peak shifted proportionally and
gradually from 214 to 278 °C, indicating that the
hyperbranched polymer was effectively suppressing the

4956 Baek et al.
Macromolecules, Vol. 35, No. 13, 2002
storage and loss modulus for both BPA-BMI and the 5
wt % AT-PAEKI/BPA-BMI blend. At room temperature,
both samples feature a substantial tensile storage
modulus of 3.6-3.7 GPa, with the blend showing a
slightly larger loss modulus attributed to a weak
â-relaxation centered at 85 °C. At elevated temperatures
a clear distinction between the two samples is observed,
with the pure BPA-BMI sample undergoing a softening
(glass-rubber) transition beginning at 175 °C and
featuring a loss modulus maximum at 215 °C, both
substantially below the postcure temperature of 250 °C.
Continued heating of this sample beyond 300 °C results
in an obvious onset of further cure as evidenced by a
significant increase in storage modulus with tempera-
ture, although samples invariably fracture at this point
in the DMA apparatus. In contrast, the 5 wt % AT-
PAEKI/BPA-BMI blend maintains a high storage modu-
lus (>2 GPa) until the tests end at 350 °C. An increase
in the loss modulus near 325 °C suggests the onset of
Tg in these samples, but we note that the exact shape
of the loss modulus curve in this temperature region is
somewhat variable from sample to sample. This sug-
gests the alternate explanation of prolonged microc-
racking at high temperature.
Con clu sion
The results of our thermal studies showed that
compatible blends were formed when the allyl-termi-
nated poly(arylene-ether-ketone-imide), AT-PAEKI,
was added to a bisphenol A-based bis(maleimide) (BPA-
BMI) in various weight percents up to 32 wt %. We
believe that the thermodynamic driving force for the
compatibility is most likely provided by the combined
effect of the similarity in the chemical structures and
the chemical interactions of the components. Further-
more, it is possible to modulate the thermal reactions,
namely the ene reaction between the O-allyl and ma-
leimide groups to provide covalent bonding between the
hyperbranched-polymer additive and the BMI host as
well as the radical homopolymerization of BMI, by
simply controlling the amount of AT-PAEKI in the
blends. The enhancing effect on the fracture toughness
of thermoset host via such covalent bonding, similar to
the interfacial adhesion30,31 between the thermoplastic
additives and the host, is expected for our AT-PAEKI/
BPA-BMI blend system. To this end, we are currently
investigating the cure kinetics, morphology, and me-
chanical properties of these blends.
F igu r e 4. TGA thermograms for the hyperbranched poly-
(arylene-ether-ketone-imides): (a) in air, (b) in helium.
radical homopolymerization of BPA-BMI and the ene
reaction that they take place in the same temperature
range, resulting in a single exotherm. This explanation
is supported by the fact that the magnitude of the shift
(∆ ) 78 °C) between the exotherm peaks for BPA-BMI
and the 32 wt % blend is about the same. Thus, we
postulate that in the 2-16 wt % BPA-BMI/AT-PAEKI
blends the first exotherm can be attributed to the ene
reaction of the allyl ether and maleimide groups, fol-
lowed by radical addition of maleimide to the adduct,
and the second exotherm is the result of the residual
radical homopolymerization of the excess BMI (see
Scheme 3c).
Exp er im en ta l Section
Finally, when BPA-BMI and all the blends were
rescanned after previously heating to 400 °C, none of
the samples showed any Tg or cure exotherm, indicative
of complete cure under DSC conditions.
Ma ter ia ls. 4-Fluorophthalic anhydride was purchased from
Matrix Scientific (Columbia, SC). 2,2-Bis[4-(4-maleimidophe-
noxy)phenyl]propane (BPA-BMI) was obtained via a custom
synthesis by University of Dayton Research Institute. All other
chemicals were reagent grade and purchased from Aldrich
Chemical Inc. and used as received, unless otherwise specified.
N-Methyl-2-pyrrolidinone (NMP) was distilled under reduced
pressure over phosphorus pentoxide. Other solvents were used
as received.
In str u m en ta tion . Proton and carbon nuclear magnetic
resonance (1H NMR and 13C NMR) spectra for intermediates,
monomer, and polymers were measured at 270 and 50 MHz
on a Tecmag-270 spectrometer. Infrared (IR) spectra were
recorded on a Beckman FT-2100 Fourier transform spectro-
photometer or with Mattson Galaxy Series FTIR 5000 spec-
trophotometer. Elemental analysis and mass spectral analysis
were performed by System Support Branch, Materials Direc-
torate, Air Force Research Lab, Dayton, OH. The melting
points (mp) of all compounds were determined on a Mel-Temp
While DSC analysis on the AT-PAEKI/BPA-BMI
blends indicated the absence of either Tg or further cure
during a second heating in the instrument, we have
employed dynamic mechanical analysis (DMA) to in-
vestigate the postcure thermal behavior in a more
sensitive manner. Thus, we prepared multiple DMA
bars of BPA-BMI and the 5 wt % blend (solvent-blended
as for DSC) by melting the materials into a silicone mold
at 150 °C and precuring under vacuum for 16 h. This
process yielded transparent amber bars of dimension
20 × 5 × 2.25 mm. Such samples were then postcured
at 250 °C in a convection oven for 2 h, resulting in dark
and rigid bars characteristic of BMI thermosets. Figure
6 shows the temperature dependence of the tensile

Macromolecules, Vol. 35, No. 13, 2002
Hyperbranched Poly(arylene-ether-ketone-imide) 4957
Ta ble 2. Th er m a l P r op er ties of AT-P AEKI/BP A-BMI Blen d s
composition (wt %)
BMI
AT-PAEKI
Tg (°C)
Tm (°C)
∆Hf (J /g)
Texo1 (°C)
∆Hexo1 (J /g)
Texo2 (°C)
∆Hexo2 (J /g)
100
98
96
92
84
68
0
0
2
4
70.5
84.0
19.5
200.0
214.7
173.5
109.4
87.0
253.4
261.4
269.2
276.4
278.0
343.0
33.4
56.7
82.5
115.9
152.6
114.7
225.0
8
∼233a
∼237a
∼58.1
16
32
100
∼32.6
144.3
133.5
269.0
98.2
a
The first exotherm appears as a shoulder with approximate peak value as indicated.
melting point apparatus and are uncorrected. Intrinsic viscosi-
ties were determined with Cannon-Ubbelohde No. 150 vis-
cometer. Flow times were recorded for NMP solution with 1%
lithium bromide and polymer concentrations of approximately
0.5-0.10 g/dL at 30.0 ( 0.1 °C. Differential scanning calorim-
etry (DSC) analysis were performed in nitrogen with a heating
rate of 10 °C/min using a Perkin-Elmer model 2000 thermal
analyzer equipped with differential scanning calorimetry cell.
Themogravimetric analysis (TGA) was conducted in nitrogen
(N2) and air atmospheres at a heating rate of 10 °C/min using
a TA Hi-Res TGA 2950 themogravimetric analyzer. Gel per-
meation chromatography (GPC) was carried out on a Waters
150-CV equipped with 254, 280, and 330 nm UV detectors.
Tetrahydrofuran (THF) was used as the eluting solvent.
Dynamic mechanical analysis was performed using a TA
Instruments 2980. The apparatus was run in controlled strain
mode and with a three-point bending fixture with a controlled
deflection amplitude of 10 µm for our 20 mm span. A heating
rate of 5 °C/min was employed, and a purge of N2 gas was
used.
5-Nitr oisop h th a loyl Dich lor id e (1). Into a 500 mL one-
necked round-bottomed flask equipped with a magnetic stirrer
and nitrogen inlet, 5-nitroisophthalic acid (25.0 g, 0.12 mol)
was dissolved freshly distilled thionyl chloride (80 mL) con-
taining DMF (3 drops). The mixture was stirred at room
temperature for 2 h and gently heated under reflux for 6 h.
Excess amount of thionyl chloride was distilled off, and the
mixture was then chilled in an ice-and-salt bath. Freshly
distilled hexane was added into the light yellow residue with
vigorous stirring. The resulting white needles are collected by
suction filtration and dried under reduced pressure to give 29.1
g (99.1% yield) of white needles; mp 59-61.5 °C. FT-IR (KBr,
cm-1): 1536, 1349 (Ar-NO2), 1757 (carbonyl). Mass spectrum
(m/e): 248 (M+, 100% relative abundance). 1H NMR (CDCl3, δ
in ppm): 8.96 (s, 2H, Ar), 9.12 (s, 1H, Ar). 13C NMR (CDCl3, δ
in ppm): 127.89, 133.33, 136.12, 148.25, 165.35.
F igu r e 5. DSC thermograms of AT-PAEKI/BPA-BMI
blends: (i) BPA-BMI, (ii) 2 wt % AT-PAEKI, (iii) 4 wt % AT-
PAEKI, (iv) 8 wt % AT-PAEKI, (v) 16% AT-PAEKI, (vi) 32 wt
% AT-PAEKI 32. (a) is the first heating scan, and (b) is the
second heating scan.
3,5-Bis(4-m eth oxylben zoyl)n itr oben zen e (2). Into a 250
mL three-necked, round-bottomed flask equipped with
a
magnetic stirrer, nitrogen inlet, and dropping funnel, alumi-
num chloride (25.4 g, 0.19 mol) and anhydrous anisole (60 mL)
were introduced. After the mixture was cooled to 15 °C in an
ice-water bath, a solution of 5-nitroisophthaloyl dichloride
(15.0 g, 60 mmol) in anhydrous anisole was then added
dropwise for 20 min. The mixture was allowed to warm to room
temperature. After 8 h of stirring, the mixture was poured into
5% hydrochloric acid. The organic layer was diluted with
methylene chloride, separated with the aid of a separatory
funnel, and rotavapped to dryness. The resulting off-white
solid residue was dissolved in hot ethanol and allowed to cool
to room temperature to give 11.2 g (47.7% yield) of off-white
solids; mp 181-182 °C. Anal. Calcd for C22H17NO6: C, 67.52%;
H, 4.38%; N, 3.58%; O, 24.53%. Found: C, 67.56%; H, 4.35%;
N, 3.45%; O, 24.76%. FT-IR (KBr, cm-1): 1538 (Ar-NO2), 1262,
1325 (ether), 1598, 1655 (carbonyl). Mass spectrum (m/e): 391
(M+, 100% relative abundance). 1H NMR (DMSO-d6; δ in
ppm): 3.89 (s, 6H, OCH3), 7.12-7.15 (d, 2H, Ar), 7.85-7.88
(d, 2H, Ar), 8.26 (t, 1H, Ar), 8.61 (d, 2H, Ar). 13C NMR (DMSO-
d6; δ in ppm): 55.61, 114.15, 126.27, 128.09, 132.47, 135.00,
139.18, 147.67, 163.57, 191.53.
3,5-Bis(4-h yd r oxyben zoyl)n itr oben zen e (3). Into a 250
mL three-necked, round-bottomed flask equipped with
a
magnetic stirrer, a condenser, and nitrogen inlet, 3,5-bis(4-
methoxyphenylcarbonyl)nitrobenzene (6.2 g, 15.8 mmol) and
freshly prepared pyridine hydrochloride (100 g) were placed.
The mixture was heated under reflux until the solution became
homogeneous. It took about 4 h. After being cooled to 120 °C,
the mixture was poured into water. The resulting precipitate
was collected and dried. The yellow solid was slurred in boiling
toluene and collected by suction filtration to give 5.5 g (96%
yield); mp 230-231.8 °C. Anal. Calcd for C20H13NO6: C,
66.12%; H, 3.61%; N, 3.86%; O, 26.42%. Found: C, 66.07%;

4958 Baek et al.
Macromolecules, Vol. 35, No. 13, 2002
Hyp er br a n ch ed P olym er Der ived fr om N-[3,5-Bis(4-
h yd r oxyben zoyl)ben zen e]-4-flu or op h th a lim id e. Into a
100 mL three-necked, round-bottomed flask equipped with a
magnetic stirrer, nitrogen inlet and outlet, and Dean-Stark
trap with a condenser, N-[3,5-bis(4-hydroxybenzoyl)benzene]-
4-fluorophthalimide (1.5 g, 3.1 mmol), potassium carbonate (1.0
g, 7.2 mmol), and a mixture of NMP (30 mL) and toluene
solvent were placed. The reaction mixture was then heated
and maintained at 140-150 °C for 4 h. During this time period,
the water formed was removed by toluene azeotropic distilla-
tion via a Dean-Stark trap. After complete removal of toluene
by an increased the flow of nitrogen, the orange solution was
then heated at 160 °C for 3 h. The solution became brown in
color and viscous. Some precipitate was observed 30 min after
reaction temperature had reached 160 °C. After being allowed
to cool on its own, the mixture was poured into a beaker
containing 5% hydrochloric acid (300 mL). The resulting
precipitate was collected by suction filtration and air-dried.
Off-white powder was dissolved in NMP again and passed
through a cake of Celite 545 to remove any insoluble salts.
The filtrate was poured in a beaker containing 5% hydrochloric
acid (300 mL) and warmed to around 60-70 °C for 2 h. The
white powder was collected and dried under the reduced
pressure over phosphorus pentoxide at 100 °C for 48 h. The
yield was essentially quantitative. [η] ) 0.13 dL/g; Tg ) 224
°C. Anal. Calcd for C28H15NO6: C, 72.88%; H, 3.28%; N, 3.06%.
Found: C, 68.55%; H, 3.93%; N, 3.01%. 1H NMR (DMSO-d6; δ
in ppm): 6.89-8.39 (Ar-H) and 10.55-10.84 (Ar-OH).
Allyla tion of Hyp er br a n ch ed P oly(a r ylen e-eth er -
im id e). Into a 50 mL three-necked, round-bottomed flask
F igu r e 6. Comparison of the temperature dependence of the
storage (upper traces) and loss (lower traces) modulus for BPA-
BMI (solid line) and a 5% BPA-BMI/AT-PAEKI blend. The
cured samples were tested in three-point bend mode at an
oscillation frequency of 1 Hz and a heating rate of 5 °C/min.
H, 3.64%; N, 3.67%; O, 17.01%. FT-IR (KBr, cm-1): 1321, 1538
(Ar-NO2), 1602, 1648 (carbonyl), 3420 (Ar-OH). Mass spec-
trum (m/e): 363 (M+, 100% relative abundance). 1H NMR
(DMSO-d6, δ in ppm): 6.93-6.96 (d, 4H, Ar), 7.76-7.79 (d,
4H, Ar), 8.23 (s, 1H, Ar), 8.58 (s, 2H, Ar), 10.66 (s, 2H, OH).
13C NMR (DMSO-d6, δ in ppm): 115.48, 126.02, 126.65, 132.84,
134.83, 139.44, 147.61, 156.80, 162.70, 191.36.
equipped with
a magnetic stirrer, nitrogen inlet, and a
condenser, hydroxyl-terminated hyperbranched poly(arylene-
ether-ketone-imide) (0.5 g, 1.08 mmol), potassium carbonate
(0.4 g, 2.9 mmol), allyl bromide (0.30 g, 2.48 mmol), and NMP
(10 mL) were placed. The reaction mixture was then heated
and maintained at 80-90 °C for 10 h. During this time period,
the orange solution became light yellow in color and homoge-
neous. After it had been allowed to cool on its own, the mixture
was filtered through a cake of Celite 545 to remove any
insoluble salts. The filtrate was poured into a beaker contain-
ing 5% hydrochloric acid (300 mL), and the mixture was
warmed around 60-70 °C for 2 h. The white powder was
collected and dried under the reduced pressure in the presence
of phosphorus pentoxide at 50 °C for 48 h. The yield was
essentially quantitative. [η] ) 0.08 dL/g; Tg ) 134 °C. Anal.
Calcd for C31H19NO6: C, 74.10%; H, 4.01%; N, 2.79%; O,
19.10%. Found: C, 73.10%; H, 4.17%; N, 2.77%, O, 17.76%.
1H NMR (DMSO-d6, δ in ppm): 4.71 (-CH2-CHdCH2), 5.24-
5.43 (-CH2-CHdCH2), 6.01 (-CH2-CHdCH2), and 7.11-
8.21 (Ar-H).
Pr op a r gyla t ion of H yp er b r a n ch ed P oly(a r ylen e -
eth er -im id e). Into a 50 mL three-necked, round-bottomed
flask equipped with a magnetic stirrer, nitrogen inlet, and a
condenser, hydroxyl-terminated hyperbranched poly(arylene-
ether-ketone-imide) (0.5 g, 1.08 mmol), potassium carbonate
(0.4 g, 2.9 mmol), propargyl bromide (0.30 g, 2.52 mmol), and
NMP (10 mL) were placed. The reaction mixture was then
heated and maintained at 80-90 °C for 10 h. During this time
period, the orange solution became light yellow in color and
homogeneous. After it had been allowed to cool on its own,
the mixture was filtered through a cake of Celite 545 to remove
any insoluble salts. The filtrate was then poured into a beaker
containing 5% hydrochloric acid (300 mL), and the mixture
was warmed to around 60-70 °C for 2 h. The white powder
was collected and dried under the reduced pressure in the
presence of phosphorus pentoxide at 50 °C for 48 h. The yield
was essentially quantitative. [η] ) 0.08 dL/g; Tg ) 122. Anal.
Calcd for C31H17NO6: C, 74.55%; H, 3.43%; N, 2.80%; O,
19.22%. Found: C, 73.07%; H, 3.82%; N, 2.70%; O, 17.99%.
1H NMR (DMSO-d6, δ in ppm): 3.64 (-CH2-CtCH), 4.96
(-CH2-CtCH), and 7.11-8.21 (Ar-H).
3,5-Bis(4-h yd r oxyben zoyl)a m in oben zen e (4). Into a 500
mL high-pressure bottle, 5-nitroisophthalic acid (4.8 g, 13
mmol), palladium on activated carbon (10%, 0.5 g), and ethanol
(100 mL) were charged. The bottle was placed on the hydro-
genation vessel. Hydrogen was charged and discharged five
times and agitated at 60-65 psi for 24 h. After the reaction
mixture had been filtered through a cake of Celite 545 to
remove catalyst, the solvent of the filtrate was removed on a
rotary evaporator. The light yellow solid was recrystallized
from deoxygenated 20% ethanol to give 4.4 g (>99% yield) of
yellow powder; mp 249.5-250.5 °C. Anal. Calcd for C20H15
-
NO4: C, 72.06%; H, 4.54%; N, 4.20%; O, 19.20%. Found: C,
72.04%; H, 4.89%; N, 3.91%; O, 18.60%. FT-IR (KBr, cm-1):
763 (Ar-NH2), 1598 1634 (carbonyl), 3378 (Ar-NH2). Mass
spectrum (m/e): 333 (M+, 100% relative abundance). 1H NMR
(DMSO-d6; δ in ppm): 5.71 (s, 2H, NH2), 6.90-6.93 (d, 4H,
Ar), 6.70 (s, 1H, Ar), 7.14 (s, 2H, Ar), 7.70-7.71 (d, 4H, Ar),
10.44 (s, 1H, OH). 13C NMR (DMSO-d6; δ in ppm): 115.10,
117.09, 117.55, 127.95, 132.35, 138.57, 148.91, 161.87, 194.21.
N-[3,5-Bis(4-h yd r oxyben zoyl)ben zen e]-4-flu or op h th a l-
im id e (5). Into a 250 mL three-necked, round-bottomed flask
equipped with a magnetic stirrer and nitrogen inlet and outlet,
3,5-bis(4-hydroxyphenylcarbonyl)aminobenzene (4.3 g, 13 mmol)
was completely dissolved in NMP (50 mL). 4-Fluorophthalic
anhydride (2.1 g, 12 mmol) was then added. The mixture was
then heated, and when temperature was approaching 170-
180 °C, isoquinoline (5 drops) was added. The mixture was
heated at 200 °C with stirring for 18 h. After cooled to room
temperature, the mixture was poured into 5% hydrochloric
acid, and the resulting precipitate was collected by suction
filtration and dried under the reduced pressure. The off-white
solid was dissolved in hot toluene and allowed to cool to room
temperature to give 4.3 g (72% yield) of off-white solid; mp
274-276 °C. Anal. Calcd for C28H16FNO6: C, 69.86%; H, 3.35%;
N, 2.91%. Found: C, 69.90%; H, 3.90%; N, 2.66%. FT-IR (KBr,
cm-1): 1644, 1601 (imide), 1724 (carbonyl) 3413 (Ar-OH).
Mass spectrum (m/e): 481 (M+, 100% relative abundance). 1H
NMR (DMSO-d6, δ in ppm): 6.93-6.96 (d, 4H, Ar), 7.71-7.75
(d, 1H, Ar), 7.78-7.82 (d, 4H, Ar), 7.89-7.93 (dd, 1H, Ar),
7.96-7.97 (t, 1H, Ar), 8.05-8.07 (d, 1H, Ar), 8.09-8.10 (d, 2H,
Ar), 10.58 (s, 2H, OH). 13C NMR (DMSO-d6, δ in ppm): 111.44,
115.36, 121.38, 126.30, 127.11, 127.71, 128.95, 130.45, 132.70,
134.65, 138.43, 162.36, 163.97, 165.47, 165.73, 167.71, 192.45.
Ep oxid a tion of Hyp er br a n ch ed P oly(a r ylen e-eth er -
k eton e-im id e). Into a 50 mL three-necked, round-bottomed
flask equipped with a magnetic stirrer, nitrogen inlet, and a
condenser, hydroxyl-terminated hyperbranched poly(arylene-

Macromolecules, Vol. 35, No. 13, 2002
Hyperbranched Poly(arylene-ether-ketone-imide) 4959
(11) Fang, J .; Kita, H.; Okamoto, K. Macromolecules 2000, 33,
4639.
(12) Wu, F.-I.; Shu, C.-F. J . Polym. Sci., Part A: Polym. Chem.
2001, 39, 2536.
(13) (a) Baek, J .-B.; Tan, L.-S. Polym. Prepr. 2001, 42 (2), 468.
(b) Baek, J .-B.; Mather, P. T.; Tan, L.-S. Polym. Mater. Sci.
Eng. 2001, 84, 742. (c) Baek, J .-B.; Ferguson, J . B.; Mather,
P. T.; Tan, L.-S. Polym. Mater. Sci. Eng. 2001, 84, 724. (d)
Baek, J .-B.; Simko, S. R.; Tan, L.-S. Polym. Prepr. 2001, 42
(2), 502. (e) Baek, J .-B.; Tan, L.-S. Polym. Prepr. 2002, 43
(1), 514. (f) Baek, J .-B.; Tan, L.-S. Polym. Prepr. 2002, 43
(1), 533.
(14) (a) Boogh, L.; Pattersson, B.; Kaiser, P.; Manson, J .-A.
SAMPE J . 1997, 33, 45. (b) Boogh, L.; Pettersson, B.; Manson,
J . A. E. Polymer 1999, 40, 2249.
(15) Tarbell, D. S. Org. React. 1944, 2, 1.
(16) Dang, T. D.; Hudson, L. S.; Feld, W. A.; Arnold, F. E. Polym.
Prepr. 2000, 41 (1), 103.
(17) (a) Dirlikov, S. K.; Feng, Y. Polym. Mater. Sci. Eng. 1988,
59, 990. (b) Dirlikov, S. K.; Feng, Y. Polym. Mater. Sci. Eng.
1989, 61, 835. (c) Feng, Y.; Dirlikov, S. K. Polym. Mater. Sci.
Eng. 1989, 60, 618. (d) Dirlikov, S. K. High Perform. Polym.
1990, 2, 67.
(18) Stenzenberger, H. D. Adv. Polym. Sci. 1994, 117, 165.
(19) Scola, D. A. Thermoset Composites, Advanced. In Lee, S. M.,
Ed.; International Encyclopedia of Composites; VCH Publish-
ers: New York, 1991; Vol. 6, pp 34-48.
(20) (a) Crivello, J . V. J . Polym. Sci., Polym. Chem. Ed. 1985, 11,
1973. (b) Guilio, C. D.; Ganther, M.; J esse, B. J . Appl. Polym.
Sci. 1984, 29, 1771. (c) Verma, I. K.; Fohlen, G. M.; Parker,
J . A. J . Polym. Sci., Polym. Chem. Ed. 1982, 20, 283.
(21) (a) Carduner, K. R.; Chattha, M. S. Polym. Mater. Sci. Eng.
1987, 56, 660. (b) Chaudhari, M. A.; Galvin, T. J .; King, J . J .
Proc. 29th Natl.SAMPE Conf. 1985. (c) Zahir, A. C.; Renner,
A. U.S. Pat. 4,100,140, 1978. (d) King, J . J .; Chaudhari, M.
A.; Zahir, S. A. Proc. 29th Natl. SAMPE Conf. 1984. (e)
Barrett, K. A.; Chaudhari, M. A.; Lee, B. H. SAMPE J . 1989,
25 (2), 17.
(22) Tan, L.-S.; Arnold, F. E. J . Polym. Sci., Part A: Polym. Chem.
1988, 26, 3103.
(23) Rakutt, D.; Fitzer, E.; Stenzenberger, H. D. High Perform.
Polym. 1991, 3, 59.
(24) (a) Kinloch, A. J .; Shaw, S. J .; Tod, D. A. In Rubber-
Toughened Polyimides; Riew, C. K., Gillham, J . K., Eds.; Adv.
Chem. Ser. No. 208; American Chemical Society: Washing-
ton, DC, 1984; Chapter 8, pp 101-115. (b) Shaw, S. J .;
Kinloch, A. J . Int. J . Adhes. Adhes. 1985, 5, 123.
(25) St. Clair, A. K.; St. Clair, T. L. Int. J . Adhes. Adhes. 1981, 1,
249.
(26) (a) Wei, G.; Sue, H.-J . J . Appl. Polym. Sci. 1999, 74, 2539.
(b) Blair, M. T.; Streiner, P. A.; Willis, E. N. Int. SAMPE
Symp. 1988, 33, 524.
ether-ketone-imide) (0.1 g, 0.22 mmol), potassium carbonate
(0.2 g, 14.5 mmol), epichlorohydrin (0.30 g, 2.52), and NMP
(10 mL) were placed. The reaction mixture was then heated
and maintained at 80-90 °C for 10 h. During this time period,
the orange solution became light yellow in color and homoge-
neous. After it had been allowed to cool on its own, the mixture
was filtered through a cake of Celite 545 to remove any
insoluble salts. The filtrate poured in a beaker containing 5%
hydrochloric acid (300 mL), and the mixture was warmed
around 60-70 °C for 2 h. The white powder was collected and
dried under the reduced pressure in the presence of phospho-
rus pentoxide at 50 °C for 48 h. The yield was essentially
quantitative. [η] ) 0.08 dL/g; Tg ) 174 °C. Anal. Calcd for
C
31H17NO6: C, 74.55%; H, 3.43%; N, 2.80%; O, 19.22%.
1
Found: C, 73.07%; H, 3.82%; N, 2.70%; O, 17.99%. H NMR
(DMSO-d6, δ in ppm): 3.33 (s, 2H, -CH2-CH(O)CH2), 3.70
(s, 1H, -CH2-CH(O)CH2), 3.95 (s, 2H, -CH2-CH(O)CH2), and
7.16-8.38 (m, 14H, Ar-H).
Solu tion Blen d in g of AT-P AEKI w ith BP A-BMI. A stock
solution was first prepared by dissolving AT-PAEKI (0.1 g)
completely in THF (10 mL). It was then added to the vials
containing predetermined amounts of BPA-BMI (0.098-0.068
g) so as to prepare the solutions with AT-PAEKI content
ranging from 2 to 32 wt %. Each vial was then diluted with
additional THF until the mixture became homogeneous over-
night. The solvent was then removed from the blend solutions
via rotary evaporation. The resulting solid blends (all formed
a coating on the vial walls) were pulverized with a spatula.
The power samples were then dried under reduced pressure
at 70 °C for 48 h prior to thermal analysis. The DSC
thermograms were recorded on the powder samples after they
had been heated to 150 °C, cooled to 20 °C (first cycle), heated
again to 320 °C, cooled again to 20 °C (second cycle), finally
heated to 350 °C and cooled to room temperature (third cycle),
all with heating or cooling rate of 10 °C/min. The thermal data
are summarized in Table 2. Figure 5a shows the initial scans
(after first and second cycles) for BMI-BPA and its blends
with AT-PAEKI. No thermal transition was detected for all
samples during the third cycle scans (Figure 5b).
Ack n ow led gm en t. We are grateful to Charles Ben-
ner (University of Dayton Research Institute) for GPC
data, Marlene Houtz (University of Dayton Research
Institute) for TGA data, and Dr. Kevin Church (Depart-
ment of Chemistry, University of Dayton) for NMR
spectra.
(27) (a) Stenzenberger, H. D.; Romer, W.; Herzog, M.; Konig, P.
Int. SAMPE Symp. 1988, 33, 1546. (b) Stenzenberger, H. D.;
Romer, W.; Hergenrother, P. M.; J ensen, B. J . Int. SAMPE
Symp. 1989, 34, 2054. (c) Stenzenberger, H. D.; Romer, W.;
Hergenrother, P. M.; J ensen, B. J .; Breiigam, W. Int. SAMPE
Symp. 1990, 35, 2175. (d) Stenzenberger, H. D.; Romer, W.;
Hergenrother, P. M.; J ensen, B. J . Int. SAMPE Symp. 1989,
34, 2054. (e) Lin, C.-R.; Lin, W. L.; Hu, J . T. Int. SAMPE
Symp. 1989, 34, 1803. (f) Wilkinson, S. P.; Ward, T. C.;
McGrath, J . E. Polymer 1993, 34, 870. (g) Iijima, T.; Hirano,
M.; Fukuda, W.; Tomoi, M. Eur. Polym. J . 1993, 29, 1399.
(h) Pascal, T.; Mercier, R.; Sillion, B. Polymer 1989, 30, 739.
(i) Pascal, T.; Sillion, B.; Grosjean, F.; Grenier-Loustalot, M.
F.; Grenier, P. High Perform. Polym. 1990, 2, 95. (j) Pascal,
T.; Mercier, R.; Sillion, B. Polymer 1990, 31, 78.
(28) Gopala, A.; Wu, H.; Xu, J .; Heiden, P. J . Appl. Polym. Sci.
1999, 71, 1809.
(29) BPA-BMI (2,2-bis[4-(4-maleimidophenoxy)phenyl]propoane)
was reported to have mp 119-120 °C, an exotherm with onset
and peak temperatures at 203 and 302 °C, respectively, and
complete cure at 344 °C; a cure at Tg 312 °C (determined by
TMA) after it was cured for 10 h at 280 °C. See: Takeda, S.;
Akiyama, H.; Kakiuchi, H. J . Appl. Polym. Sci. 1988, 35,
1341.
Refer en ces a n d Notes
(1) (a) Wilson, D.; Stenzeberger, H. D.; Hergenrother, P. M.
Polyimides; Chapman and Hall: New York, 1990. (b) Sroog,
C. E. Prog. Polym. Sci. 1991, 16, 561.
(2) Lin, S.-H.; Li, F.; Cheng, S. Z. D.; Harris, F. W. Macromol-
ecules 1998, 31, 2080.
(3) Harris, F. W.; Hsu, S. L.-C. High Perform. Polym. 1989, 1, 3.
(4) Harris, F. W.; Hsu, S. L.-C.; Tso, C. C. Polym. Prepr. 1990,
31, 342.
(5) Boise, A. I. J . Appl. Polym. Sci. 1986, 32, 4043.
(6) Hoffman, D. A.; Ansari, H.; Frank, C. W. In Material Science
of High-Temperature Polymers for Microelectronics; Grubb,
D. T., Mita, I., Yoon, D. Y., Eds.: MRS Symposium Series
227; Material Research Society: Pittsburgh, PA, 1991; p 125.
(7) Matthews, O. A.; Shipway, A. N.; Stoddart, J . F. Prog. Polym.
Sci. 1998, 23, 1.
(8) (a) Kim, Y. H. J . Polym. Sci., Part A: Polym. Chem. 1998,
36, 1685. (b) Malmstro¨m, E.; Hult, A. J . Macromol. Sci., Rev.
Macromol. Chem. Phys. 1997, 37, 555. (c) Hult, A.; J ohansson,
M.; Malmstro¨m, E. Adv. Polym. Sci. 1999, 143, 1. (d) Inoue,
K. Prog. Polym. Sci. 2000, 25, 453. (e) Voit, B. J . Polym. Sci.,
Part A: Polym. Chem. 2000, 36, 2505. (f) Hult, A.; Malm-
stro¨m, E.; J ohansson, M. In Polymeric Materials Encyclope-
dia; Salamone, J ., Ed.; CRC Press: Boca Raton, FL, 1996;
Vol. 5 H-L, p 3171.
(9) Thompson, D. S.; Markoski, L. J .; Moore, J . S. Macromolecules
1999, 32, 4764.
(10) Yamanaka, K.; J ikei, M.; Kakimoto, M. Macromolecules 2000,
33, 6937.
(30) (a) Yee, A. F.; Pearson, R. A. J . Mater. Sci. 1986, 21, 2462.
(b) Pearson, R. A.; Yee, A. F. J . Mater. Sci. 1986, 21, 2475.
(31) Illjima, T.; Nishina, T.; Fukuda, W.; Tomoi, M. J . Appl. Polym.
Sci. 1998, 67, 769 and references therein.
MA020066+