Synthesis of thermosetting polyetherimide-containing allyl groups

Article abstract of DOI:10.1002/pola.26549

An allyl-containing diphenol, 1-(3-allyl-4-hydroxyphenyl)-1-(4- hydoxyphenyl)-1-(6-oxido-6H -dibenz[c,e][1,2]oxaphosphorin-6-yl)ethane (1), was prepared from a one-pot reaction of 9,10-dihydro-oxa-10-phosphaphenanthrene-10- oxide, 4-hydroxyacetophenone, and 2-allylphenol in the presence of p-toluenesulfonic acid monohydrate. Then, an allyl-containing dietheramine, 1-(4-(4-aminophenoxy)phenyl)-1-(3-allyl 4-(4-aminophenoxy)-phenyl)-1-(6-oxido- 6H-dibenz[c,e][1,2] oxaphosphorin-6-yl)ethane (3), was prepared from the nucleophilic substitution of (1) with 4-fluoronitrobenzene, followed by the reduction of the dinitro groups by Fe/HCl. A flexible polyetherimide (PEI) (4) with a curable characteristic was prepared from the condensation of (3) and 4,4′-oxydiphthalic anhydride (ODPA) in m-cresol in the presence of isoquinoline. Curing PEI (4) at 300°C leads to PEI (5), which exhibits much a higher T_g value (307°C) and a lower coefficient of thermal expansion (CTE) (29 ppm/°C) than PEI (4) (T_g = 253°C, CTE 52 ppm/°C).

Full text of DOI:10.1002/pola.26549

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Synthesis of Thermosetting Polyetherimide-Containing Allyl Groups
Ching Hsuan Lin,1,2 Sheng Lung Chang,1 Yu Ru Wang,1 Chin Kuang Hsu,1 Tzong Yuan Juang3
1Department of Chemical Engineering, National Chung Hsing University, Taichung, Taiwan
2Center of Nanoscience & Nanotechnology, National Chung Hsing University, Taichung, Taiwan
3Department of Applied Chemistry, National Chiayi University, Chiayi, Taiwan
Correspondence to: C. H. Lin (E-mail: linch@nchu.edu.tw) or T. Y. Juang (E-mail: tyjuang@mail.ncyu.edu.tw)
Received 14 October 2012; accepted 19 December 2012; published online 28 January 2013
DOI: 10.1002/pola.26549
ABSTRACT: An allyl-containing diphenol, 1-(3-allyl-4-hydroxy-
mide (PEI) (4) with a curable characteristic was prepared from
the condensation of (3) and 4,40-oxydiphthalic anhydride
(ODPA) in m-cresol in the presence of isoquinoline. Curing PEI
(4) at 300 ꢀC leads to PEI (5), which exhibits much a higher Tg
value (307 ꢀC) and a lower coefficient of thermal expansion
phenyl)-1-(4-hydoxyphenyl)-1-(6-oxido-6H -dibenz[c,e][1,2]oxa-
phosphorin-6-yl)ethane (1), was prepared from
a one-pot
reaction of 9,10-dihydro-oxa-10-phosphaphenanthrene-10-
oxide, 4-hydroxyacetophenone, and 2-allylphenol in the presence
of p-toluenesulfonic acid monohydrate. Then, an allyl-contain-
(CTE) (29 ppm/ꢀC) than PEI (4) (Tg ¼ 253 ꢀC, CTE 52 ppm/ꢀC).
C
V
ing
dietheramine,
1-(4-(4-aminophenoxy)phenyl)-1-(3-allyl
2013 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym.
4-(4-aminophenoxy)-phenyl)-1-(6-oxido-6H-dibenz[c,e][1,2] oxa-
phosphorin-6-yl)ethane (3), was prepared from the nucleophilic
substitution of (1) with 4-fluoronitrobenzene, followed by the
reduction of the dinitro groups by Fe/HCl. A flexible polyetheri-
Chem. 2013, 51, 1734–1741
KEYWORDS: allyl; high performance polymers; polyetherimide;
polyimides; thermosets
INTRODUCTION Aromatic polyimides are widely applied in
electronics and composites owing to their characteristics of
high glass transition temperature, high thermal stability, and
good mechanical properties.1–9 However, the good thermal
properties are usually accompanied by poor solubility in or-
ganic solvents, limiting their processability. Various
approaches have been developed to enhance their organosol-
ubility including (1) the incorporation of thermally stable
but flexible linkages, such as ether, methylene, sulfonyl, and
hexafluoroisopropylidene10–12; (2) introducing unsymmetri-
cal linkages13–19 or bulky pendants;15,20–23 and (3) copoly-
merization.24–29
inspired us to synthesize an allyl-containing polyimide that
can match the requirement of processability (before thermal
curing) and good thermal properties (after thermal curing).
Therefore, an organosoluble polyetherimide (PEI) (4) was
prepared through the solution polymerization of an allyl-
containing diamine (3) and 4,40-oxydiphthalic anhydride
(ODPA). Curing PEI (4) at 300 ꢀC led to thermosetting PEI
(5). The experimental data showed that (5) exhibited much
higher Tg value and lower thermal expansion than (4).
Detailed synthetic strategy and analysis are discussed in this
study.
EXPERIMENTAL
Although organosoluble polyimides exhibit processability,
they usually display decreased thermal properties compared
with insoluble polyimides. Many thermosetting polyimides
have been reported in the literature, most of them are termi-
nated with crosslinkable linkages, such as phenylmaleic,30
phenylethynyl,31,32 maleic,33 nadic,34,35 allylic,36 benzocyclo-
butene,37 and carboxylic acid.38 Thermosetting polyimide
with crosslinkable linkage in the repeat unit is rarely
reported.39 Ueda and coworkers40 prepared thermo-curable
polyphenylene ether through the oxidative coupling copoly-
merization of 2-allyl-6-methylphenol with 2,6-dimethylphe-
nol. The thermally cured copolymer showed excellent solvent
resistance, high thermal stability, high Tg, a low dielectric
constant, and a low dissipation factor. The above method
Materials
9,10-Dihydro-9-oxa-10-phosphaphenanthrene 10-oxide (DOPO,
from TCI), p-toluenesulfonic acid monohydrate (p-TSA, from
Showa), 4-hydroxyacetophenone (from Acros), 2-allylphenol
(from Showa), potassium carbonate (from Showa), 10% palla-
dium on carbon (Pd/C, from Acros), 4-fluoronitrobenzene
(from Acros), iron powder (Fe, from J. T. Baker), 37%
hydrochloric acid (HCl, from Scharlau), and calcium hydride
(from Acros) were used as received. ODPA (Chriskev) was
recrystallized from acetic anhydride. N,N-dimethylacetamide
(DMAc, from TEDIA) was purified by distillation under
reduced pressure over calcium hydride (Acros) and stored
over molecular sieves. The other solvents used are
Additional Supporting Information may be found in the online version of this article.
C
V
2013 Wiley Periodicals, Inc.
1734
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
commercial products (HPLC grade), which were used without
further purification.
mol), and 30 mL of DMAc were introduced to a round-bot-
tomed 100-mL glass flask equipped with a nitrogen inlet, a
condenser, andꢀa magnetic stirrer. The reaction mixture was
heated to 110 C and maintained at that temperature for 12
h. Then, the reaction mixture was filtered, and the filtrate
was poured into 450 mL of methanol/water (1/5, v/v) solu-
tion under stirring. The precipitate was filtered and recrys-
tallized from acetic anhydride, and then dried in a vacuum
oven at 110 ꢀC. Light yellow crystalꢀs (yield, 80%) were
obtained. Melting point (DSC) was 184 C.
Characterization
NMR measurements were performed using a Varian Inova
600 NMR in dimethyl sulfoxide (DMSO)-d6, and the chemical
shift was calibrated by setting the chemical shift of DMSO-d6
as 2.49 ppm. Differential scanning calorimeter (DSC) scans
were obtained by a Perkin-Elmer DSC 7 in a nitrogen atmos-
phere at a heating rate of 20 ꢀC/min. Dynamic mechanical
analysis (DMA) was performed by a Perkin-Elmer Pyris Dia-
ꢀ
ELEM. ANAL. for C41H31N2O8P: Calcd. C 69.29%, H 4.40%, N
3.94%; Found C 69.09%, H 4.52%, N 3.89%. 1H-NMR (ppm,
DMSO-d6), d ¼ 1.85–1.90 (3H, H14), 3.02–3.12 (2H, H20), 4.83–
mond DMA at a heating rate of 5 C/min. Thermomechanical
analysis (TMA) was performed by a SII TMA/SS6100 at a
ꢀ
heating rate of 5 C/min. Thermal gravimetric analysis (TGA)
4.95 (2H, H22), 5.65 (1H, H21), 6.88 (1H, H17), 6.95 (2H, H28),
was performed with a Perkin-Elmer Pyris1 at a heating rate
of 20 ꢀC/min in nitrogen and air atmosphere, respectively.
Elemental analysis (C, H, and N) was performed on an Ele-
mentar Vario EL III. The infrared spectra were recorded
using a Nicolet Avatar 320 FTIR spectrophotometer and 32
0
7.08 (4H, H25,28 ), 7.15 (1H, H10), 7.21 (2H, H8,16), 7.32(1H,
H23), 7.38 (1H, H9),7.47 (3H, H4,24), 7.52 (1H, H3),07.76 (1H, H2),
8.07 (1H, H7), 8.18 (1H, H1), 8.26–8.30 (4H, H29,29 ).
Synthesis of (3)
scans were collected with a spectral resolution of 1 cmꢁ1
.
A mixture of 1.5 mL of 37% HCl and 3 mL of 50% aqueous
ethanol was slowly added to a mixture of 5.0 g of (2) (7
mmol), 1.96 g of iron powder (35 mmol), and 15 mL of 50%
aqueous ethanol in a three-necked, 100-mL, round-bottomed
flask equipped with a nitrogen inlet and a magnetic stirrer.
The mixture was stirred under reflux for 3 h, and then 1.3
mL of ammonium hydroxide solution (10 wt %) was slowly
added to this mixture over a 1-h period. The mixture was
hot-filtered, and the filtrate was poured into water. The pre-
cipitate was filtered and dried in a vacuum oven to afford
diamine (3). Light orange crystals ꢀ (yield, 70%) were
obtained. Melting point (DSC) was 103 C.
The flame retardancy of the PEIs was determined by a
UL-94VTM vertical thin test. In that test, an 800 ꢂ 200 sample
was wrapped around a 1/200 mandrel, and then taped on
one end. The mandrel was removed, leaving a cone-shaped
sample that was relatively rigid. The two flame applications
took 3 s each for the UL-94 VTM vertical thin film. After the
first ignition, the flame was removed and the time for the
polymer to self-extinguish (t1) was recorded. Cotton ignition
was noted if polymer dripping occurred during the test.
After cooling, the second ignition was performed on the
same sample and the self-extinguishing time (t2) and
dripping characteristics were recorded. If t1 þ t2 < 10 s
without any dripping, then the polymer was considered to
be a VTM-0 material. If t1 þ t2 was in the range of 10–30 s
without any dripping, then the polymer was considered to
be a VTM-1 material.
ELEM. ANAL. for C41H35N2O4P: Calcd. C 75.68%, H 5.42%, N
4.31%; found C 75.44%, H 5.57%, N 4.28%. 1H-NMR (ppm,
DMSO-d6), d ¼ 1.70 (3H, H14), 4.23 (2H, H20), 4.99 (6H,
0
H20,NH2), 5.82 (1H, H21), 6.37 (1H, H17), 6.57 (4H, H29,29 ),
0
6.65 (4H, H25,28), 6.73 (2H, H28 ), 6.97 (1H, H16), 7.08 (1H,
H10), 7.20(4H, H8,23,24), 7.33 (1H, H9), 7.37 (1H, H4), 7.43
(1H, H3), 7.70(1H, H2), 7.95 (1H, H7), 8.07(1H, H1).
Synthesis of (1)
In brief, 10.81 g of DOPO (0.05 mol), 6.81 g of 4-hydroxyace-
tophenone (0.05 mol), 0.432 g of p-TSA (4 wt % based on
the weight of DOPO), and 6.71 g of 2-allylphenol (0.15 mol)
were introduced into a 100-mL round-bottomed glass flask
equipped with a nitrogenꢀ inlet and a magnetic stirrer. The
mixture was stirred at 60 C for 12 h. After that, the reaction
mixture was cooled to room temperature. The precipitate
was filtered, washed with ethyl acetate, recrystallized by
methanol, and dried at 100 ꢀC in a vacuum oven. White
crystꢀal (yield, 70%) was obtained. Melting point (DSC) was
227 C.
Preparation of (4) and (5)
PEI (4) was prepared by reacting (3) with an equal mole of
ODPA. To
a 100-mL three-neck round-bottomed flask
equipped with a magnetic stirrer, nitrogen inlet, 1.301 g of
(3) (2 mmol), 0.436 g of ODPA (2 mmol), 8.7 g of m-cresol,
and two drops of isoquinoline were added. The mixture was
reacted at 200 ꢀC for 8 h. Then, the viscous solution was
poured into methanol. The precipitate was filtered, dried,
and dissolved in DMAc to afford a 20 wt % solution. The
solution was casted on glass by an automatic film applicator,
and dried at 60 ꢀC (12 h), 100 ꢀC (1 h), 150 ꢀC (1 h), and
200 ꢀC (1 h). Thermosetting PEI (5) was obtained by the
ELEM. ANAL. for C29H25O4P: Calcd. C 74.35%, H 5.38%; Found
1
C 74.25%, H 5.57%. H-NMR (DMSO-d6), d ¼ 1.58 (3H, H14),
3.09 (2H, H20), 4.91 (2H, H22), 5.78 (1H, H21), 6.57 (3H,
H17,25), 6.96 (1H, H16), 7.07–7.12 (4H, H10,23,24), 7.14–7.20
(2H, H4,8), 7.32–7.36 (2H, H3,9), 7.66 (1H, H2), 7.95 (1H, H7),
8.07 (1H, H1), 9.33–9.41 (2H, OH).
ꢀ
thermal curing of (4) at 300 C (1 h).
RESULTS AND DISCUSSION
Synthesis of (1)
Synthesis of (2)
Briefly, 4.68 g of (1) (0.01 mol), 3.10 g of 4-fluoronitroben-
zene (0.022 mol), 3.04 g of potassium carbonate (0.022
The allyl-containing diphenol (1) was prepared in a one-pot
procedure by the reaction of DOPO, 4-hydroxyacetophenone
in excess 2-allylphenol using p-TSA as catalyst (Scheme 1).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741
1735

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
into two peaks at 51.4 and 52.0 ppm with a coupling
constant of 90 Hz. For the same reason, the signals of C12
split into two peaks at 122.3 and 123.1 ppm with a coupling
constant of 110 Hz. The allyl signals at 136.9 (C21), 115.4
(C22), and 33.8 (C20) were clearly observed. The peak
patterns of ArAC, assigned by the assistance of the correla-
tions shown in the 1HA13C HETCOR spectra (Supporting
Information Fig. S3) also confirmed the structure of (1).
SCHEME 1 Synthesis of (1).
Supporting Information Scheme S1 shows the proposed
mechanism for the synthesis of (1). In the mechanism, DOPO
attacks the carbonyl group via a nucleophilic addition, form-
ing a tertiary hydroxyl group. The tertiary hydroxyl group is
then protonated by acid. Then, spontaneous dissociation of
the protonated hydroxyl group occurs to yield a carbocation
intermediate plus water. The carbocation can be stabilized
by the electron-donating phenolic OH. Finally, 2-allylphenol
reacts with the carbocation via an electrophilic substitution,
yielding (1). Figure 1 shows the 1H-NMR spectrum of (1).
The signals of the methyl group (H14) were split into two
The structure of (1) was further confirmed by the single
crystal diffractogram (Fig. 2 and Supporting Information
Table S1). The crystal of (1) belongs to the triclinic system
with a ¼ 8.7431 Å, a ¼ 94.015ꢀ; b ¼ 14.4670 Å, b ¼
102.431ꢀ, and c ¼ 19.8130 Å, c ¼ 98.761ꢀ. The bond length
for C(21)AC(22) is 1.272 Å, which is shorter than those of
C(20)AC(21) (1.535 Å) and C(13)AC(14) 1.455 Å. The
shorter bond length supports the finding that the
C(21)AC(22) is a double bond, and no isomerization of the
allyl chains to propenyl moieties was found in the synthesis
(to be discussed later).
3
peaks with a coupling constant of 17.4 Hz owing to a JPAH
coupling. Two phenol peaks at around 9.4 ppm, and the allyl
signals at 5.8 (H21), 4.9 (H22), and 3.1 (H20) were clearly
observed. The peak patterns of ArAH, assigned by the assis-
tance of correlations shown in the 1HA1H COSY spectra
(Supporting Information Fig. S1) confirmed the structure
of (1). In the 13C-NMR spectrum (Supporting Information
Synthesis of (2)
Dinitro compound (2) was synthesized from the nucꢀleophilic
substitution of (1) and 4-fluoronitrobenzene at 110 C using
potassium carbonate as catalyst. Hamerton and coworkers41
reported the isomerization of allyl chains to propenyl moi-
eties in the reaction of 2-allylphenol and bis(4-fluorophenyl)-
sulfone in the presence of potassium carbonate. Therefore,
1
Fig. S2), owing to the JPAC coupling, the signals of C13 split
FIGURE 1 1H-NMR spectrum of (1) in DMSO-d6.
1736
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 2 Single crystal diffractogram of (1).
0
spectrum of (1) (Fig. 1), the signals from H28AH28 and
0
we believe that the nucleophilic substitution of (1) and
4-fluoronitrobenzene in the presence of excess potassium car-
bonate might lead to two products, (2) and (20) (Scheme 2).
H29AH29 indicate that (2) was successfully prepared. Owing to
the inductive and anisotropic deshielding effect of the nitro
0
group, the signals of H29 and H29 (ortho to nitro group) show
the highest chemical shift at aro0und 8.3 ppm. In contrast, the
signals of H17, H25, H28, and H28 (ortho to ether group) show
low chemical shift at 6.9–7.1 ppm because of the shielding
effect of the electron-donating ether group. Other spectroscopic
data such as 13C-NMR, 1HA1H COSY, and 1HA13C HETCOR
spectra are shown in Supporting Information Figures S5–S7.
Figure 3 shows the 1H-NMR spectra of the reaction product
prepared with several potassium carbonate-to-(1) ratios at
110 ꢀC for 12 h. Small propenyl peaks at 1.8, 5.9, and 6.2
ppm were found when the ratio of potassium carbonate-
to-(1) was higher than 1.0 (Note that we found that the
isomerization of the allyl chains to propenyl is strongly
dependent on the reactant. Obvious isomerization was found
in other systems, and the result will be published else-
where.). For the molar ratio of ꢃ1, [Figs. 3(a,b)], no signals
from the isomerization of the allyl chains to propenyl moi-
eties were found. In addition, the allyl signals at 5.8 (H21),
4.9 (H22), and 3.1 (H20) clearly remained, suggesting that (2)
was successfully prepared under the reaction condition.
Supporting Information Figure S4 shows the detailed
1H-NMR spectrum of (2). Compared with the 1H-NMR
Synthesis of (3)
Initially, we reduced (2) via the hydrogen catalytic reduction
in the presence of Pd/C. An amino peak at 5.0 ppm was
clearly observed (Supporting Information Fig. S8). However,
the signals from the allyl group disappeared completely, and
new signals for propyl linkage at 0.8 (CH3), 1.4 (CH2), and
2.4 (PhACH2) ppm appeared. This indicates that the reduc-
tion occurred not only on the nitro groups, but also on the
SCHEME 2 Synthesis of (2) and (20).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741
1737

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1
FIGURE 3 H-NMR spectra of the reaction product prepared with several potassium carbonate-to-(1) ratios at 110 ꢀC for 12 h.
FIGURE 4 1H NMR spectrum of (3) in DMSO-d6.
1738
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 3 Synthesis of PEI (4).
allyl group. We then used Fe/HCl as the reduction agent.42
We found that the nitro was successfully reduced to amine,
whereas the allyl group was preserved (Fig. 4). Because of
polymerization at high temperature, as supported by the
exothermic peaks at around 252 and 280 ꢀC. Therefore, a
thermal curing of (4) at 300 ꢀC resulted in a thermosetting
(5). Figure 5 shows the enlarged IR spectra of (4) and (5).
The intensity of allyl absorption at 920 cmꢁ1 clearly
decreased,40 supporting the curing characteristic of the allyl
group at high temperatures.
the shielding effect of the electron-donating amino groups,
0
signals of H29 and H29 upshifted from 8.3 ppm in (2) to 6.6
ppm in (3), demonstrating the successful reduction of (2).
The other spectroscopic data, such as 13C-NMR, 1HA1H
COSY, and 1HA13C HETCOR spectra (Supporting Information
Figs. S9–S11), support the structure and purity of (3).
Film Quality and Thermal Properties
Supporting Information Figure S14 shows a image of (4)
and (5), in which both PEIs are flexible. As both films are
flexible, DMA and TMA were applied to evaluate their ther-
mal mechanical properties and dimensional stability. Figure
6 shows the DMA thermograms of (4) and (5). The Tg
obtained from the peak temperature of tan(d) increased
from 253 C for (4) to 307 C for (5). An increase in modu-
lus after glass transition was observed for (4). The increase
is thought to be related to the increased rigidity of (4)
owing to the curing of the allyl groups. The increased rigid-
ity can also be supported by the smaller tan(d) height of (5)
compared to that of (4). Figure 7 shows the TMA curves of
PEI Synthesis
PEI (4) was prepared by the solution polymerization of (3)
with ODPA in m-cresol in the presence of isoquinoline
(Scheme 3). Supporting Information Figure S12 shows the
1H-NMR spectrum of (4). Signals for the allyl group were
clearly observed, suggesting that the allyl group is stable at
this stage.
ꢀ
ꢀ
Supporting Information Figure S13 shows the DSC thermo-
gram of (1). The allyl group, although relatively stable for
free radical polymerization in mild conditions, can undergo
FIGURE 5 IR spectra of PEIs (4) and (5).
FIGURE 6 DMA thermograms of (4) and (5).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741
1739

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Data of UL-94 VTM Test for (4), (5), and Kapton
First
Second
Burning
Time (s)
Sample
Code
Burning
Time (s)
UL-94
Grade
Dripping
(4)
1.6
1.2
4.6
0.4
0.3
1.8
No
No
No
V-0
V-0
V-0
(5)
Kapton
The flame retardancy of both systems was measured by a
UL-94VTM vertical thin test (Table 2). The t1 þ t2 took <3 s
for both systems. Therefore, they belong to the VTM-0 grade.
In contrast, the t1 þ t2 took more than 6 s for Kapton.
Although Kapton also belongs to the VTM-0 grade, the
shorter total burning time of (4) and (5) demonstrates the
excellent flame retardancy provided by the phosphorus
element.
FIGURE 7 TMA thermograms of (4) and (5).
CONCLUSIONS
(4) and (5). The Tg obtained from the onset of dimension
transition increased from 225 ꢀC for (4) to 277 ꢀC for (5).
Both figures demonstrate the advantage of crosslinking in
enhancing the Tg. Figure 7 also shows that the coefficient of
thermal expansion (CTE) in the range of 50–150 ꢀC
decreased dramatically from 52 ppm/ꢀC for (4) to 29
ppm/ꢀC for (5). The CTE value is a relatively low one com-
pared with other polyimides. This result points out the
advantage of crosslinking in enhancing dimensional stability.
We have successfully prepared an allyl-containing diphenol
(1) based on the reaction of DOPO, 4-hydroxyacetophenone,
and 2-allylphenol in the presence of p-TSA. A dietheramine
(3) with an allyl group was prepared from a potassium car-
bonate-catalyzed nucleophilic aromatic substitution of (1)
with 4-fluoronitrobenzene, followed by the reduction of the
dinitro compound (2). We found that the molar ratio of po-
tassium carbonate-to-(1) influenced the purity of (2). No
isomerization of the allyl chains to propenyl moiety was
found when the ratio was ꢃ1. We also found that Fe/HCl is a
better reducing agent than the Pd/C, H2 system, which leads
to the reduction of the allyl bond simultaneously. PEI (4) pre-
pared from (3) and OPDA in m-cresol shows a curable char-
acteristic, as supported by the reduction of the IR absorption
of the allyl group at 920 cmꢁ1 after thermal treatment. After
The thermal stability of both PEIs was evaluated by TGA
(Supporting Information Figs. S15 and S16 and Table 1). The
5 wt % degradation temperature is 443 ꢀC for (4), and
447 ꢀC for (5) in a nitrogen atmosphere, and is 440 ꢀC for
ꢀ
(4) and 443 C and (5) in an air atmosphere. The char yield
is 54 wt % for (4) and 60 wt % for (5) in a nitrogen atmos-
phere, and 20 wt % for (4) and 41 wt % for (5) in an air
atmosphere. Although significant enhancement in Tg and CTE
was observed after thermal curing, the difference in the ini-
tial decomposition temperature for (4) and (5) is not as
obvious in the case of the TGA thermograms. This result is
reasonable as PAC is the weakest bond in both systems.
Therefore, the effect of crosslinking is not obvious in the ini-
tial thermal decomposition. However, the effect of crosslink-
ing in enhancing char yield is obvious.
ꢀ
ꢀ
curing (4) at 300 C for 1 h, Tg shifted from 253 to 307 C,
and CTE decreased from 52 to 29 ppm/ꢀC. This study pro-
vides an approach for transferring soluble PEIs to high Tg,
high dimensional stability thermosetting PEIs.
ACKNOWLEDGMENT
The authors thank the National Science Council of the Republic
of China for financial support.
TABLE 1 Thermal Properties of (4) and (5)
Char Yielde
T
d5%(ꢀC)d
(wt %)
Sample
Film Quality
Tg (ꢀC) (DMA)a
Tg (ꢀC) (TMA)b
CTE (ppm/ꢀC)c
In N2
In air
In N2
In air
(4)
(5)
a
Flexible
Flexible
253
307
225
277
52
29
443
447
440
443
54
60
20
41
The peak temperature of tan d, measured by DMA at a heating 5 ꢀC/min.
Measured by TMA at a heating rate of 5 ꢀC/min.
b
c
Coefficient of thermal expansion in the range of 50–150 ꢀC.
d
e
Temperature corresponding to 5% weight loss, as recorded by TGA at a heating rate of 20 ꢀC/min.
Residual weight percent at 800 ꢀC.
1740
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
22 A. S. Mathews, D. Kim, Y. Kim, I. Kim, C.-S. Ha, J. Polym.
Sci. Part A: Polym. Chem. 2008, 46, 8117–8130.
REFERENCES AND NOTES
1 D.-J. Liaw, K.-L. Wang, Y.-C. Huang, K.-R. Lee, J.-Y. Lai, C.-S.
Ha, Prog. Polym. Sci. 2012, 37, 907–974.
23 S. Wu, T. Hayakawa, M. A. Kakimoto, H. Oikawa, Macromo-
lecules 2008, 41, 3481–3487.
2 C.-L. Chung, C.-P. Yang, S.-H. Hsiao, J. Polym. Sci. Part A:
Polym. Chem. 2006, 44, 3092–3102.
24 Y. Zhuang, X. Liu, Y. Gu, Polym. Chem. 2012, 3, 1517–1525.
25 P. R. Buch, A. V. R. Reddy, Polymer 2005, 46, 5524–5532.
3 H.-S. Li, J.-G. Liu, J.-M. Rui, L. Fan, S.-Y. Yang, J. Polym. Sci.
Part A: Polym. Chem. 2006, 44, 2665–2674.
26 G. Maglio, R. Palumbo, A. Schioppa, D. Tesauro, Polymer
1997, 38, 5849–5856.
4 S. M. Kwak, J. H. Yeon, T. H. Yoon, J. Polym. Sci. Part A:
Polym. Chem. 2006, 44, 2567–2578.
27 C. Lee, N. P. Iyer, K. Min, H. Pak, H. Han, J. Polym. Sci. Part
A: Polym. Chem. 2004, 42, 137–143.
5 T. Ogura, T. Higashihara, M. Ueda, J. Polym. Sci. Part A:
Polym. Chem. 2010, 48, 1317–1323.
28 H. Behniafar, S. Habibian, Polym. Int. 2005, 54, 1134–1140.
29 T. Iijima, N. Hayashi, T. Oyama, M. Tomoi, Polym. Int. 2004,
53, 1417–1425.
6 G.-S. Liou, P.-H. Lin, H.-J. Yen, Y.-Y. Yu, W.-C. Chen, J.
Polym. Sci. Part A: Polym. Chem. 2010, 48, 1433–1440.
7 C.-Y. Wang, G. Li, J.-M. Jiang, Polymer 2009, 50, 1709–1716.
30 G. W. Meyer, J. L. Heidbrink, J. G. Franchina, R. M. Davis,
S. Gardner, V. Vasudevan, T. E. Glass, J. E. McGrath, Polymer
1996, 37, 5077–5088.
8 L. Tao, H. Yang, J. Liu, L. Fan, S. Yang, Polymer 2009, 50,
6009–6018.
31 T. Kuroki, A. Shibuya, M. Toriida, S. Tamai, J. Polym. Sci.
Part A: Polym. Chem. 2004, 42, 2395–2404.
9 T. Kurose, V. E. Yudin, J. U. Otaigbe, V. M. Svetlichnyi, Poly-
mer 2007, 48, 7130–7138.
32 T. Sasaki, H. Moriuchi, S. Yano, R. Yokota, Polymer 2005,
46, 6968–6975.
10 C. P. Yang, W. T. Chen, Macromolecules 1993, 26,
4865–4871.
33 J. A. Mikroyannidis, J. Polym. Sci. Part A: Polym. Chem.
1994, 32, 2403–2411.
11 C.-P. Yang, W.-T. Chen, J. Polym. Sci. Part A: Polym. Chem.
1993, 31, 1571–1578.
12 J. Hao, M. Jikei, M. A. Kakimoto, Macromolecules 2003, 36,
3519–3528.
34 A. J. Hu, J. Y. Hao, T. He, S. Y. Yang, Macromolecules 1999,
32, 8046–8051.
13 I. S. Chung, S. Y. Kim, Macromolecules 2000, 33, 3190–3193.
35 T. Pascal, D. Audigier, R. Mercier, B. Sillion, Polymer 1991,
32, 1119–1125.
14 H. Choi, I. S. Chung, K. Hong, C. E. Park, S. Y. Kim, Polymer
2008, 49, 2644–2649.
36 L. Bes, A. Rousseau, B. Boutevin, R. Mercier, J. Polym. Sci.
Part A: Polym. Chem. 2001, 39, 2602–2619.
15 Y.-T. Chern, J.-Y. Tsai, J.-J. Wang, J. Polym. Sci. Part A:
Polym. Chem. 2009, 47, 2443–2452.
37 L.-S. Tan, F. E. Arnold, J. Polym. Sci. Part A: Polym. Chem.
1988, 26, 1819–1834.
16 W. G. Kim, A. S. Hay, Macromolecules 1993, 26, 5275–5280.
17 Y.-T. Chern, M.-H. Ju, Macromolecules 2008, 42, 169–179.
18 Y.-T. Chern, J.-Y. Tsai, Macromolecules 2008, 41, 9556–9564.
38 K. Xu, J. Economy, Macromolecules 2004, 37, 4146–4155.
39 N. Furukawa, M. Yuasa, Y. Yamada, Y. Kimura, Polymer
1998, 39, 2941–2949.
19 S.-H. Hsiao, K.-H. Lin, J. Polym. Sci. Part A: Polym. Chem.
2005, 43, 331–341.
40 T. Fukuhara, Y. Shibasaki, S. Ando, M. Ueda, Polymer 2004,
45, 843–847.
20 Y. Liu, Y. Xing, Y. Zhang, S. Guan, H. Zhang, Y. Wang, Y.
Wang, Z. Jiang, J. Polym. Sci. Part A: Polym. Chem. 2010, 48,
3281–3289.
41 J. M. Barton, A. Chaplin, I. Hamerton, B. J. Howlin, Polymer
1999, 40, 5421–5427.
21 Y.-C. Kung, G.-S. Liou, S.-H. Hsiao, J. Polym. Sci. Part A:
Polym. Chem. 2009, 47, 1740–1755.
42 K. Xie, S. Y. Zhang, J. G. Liu, M. H. He, S. Y. Yang, J.
Polym. Sci. Part A: Polym. Chem. 2001, 39, 2581–2590.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2013, 51, 1734–1741
1741