Synthesis of bisphenol A-free oligomeric phthalonitrile resins with sulfone and sulfone-ketone containing backbones

Article abstract of DOI:10.1002/pola.28020

Two new oligomeric sulfone and sulfone-ketone containing phthalonitrile (PN) resins with excellent processability have been developed. The PN monomers were prepared from the reaction of an excess amount of bisphenol S with 4-(chlorophenyl)sulfone or 4,4-dichlorobenzophenone in the presence of a base in a solvent mixture (dimethylsulfoxide/toluene), followed by end-capping with 4-nitro-PN in a two-step, one-pot reaction. These PN resins exhibited good viscosities and cure times for molding into various shapes. After being thermally cured to yield crosslinked polymers, these polymers demonstrated superb mechanical properties, thermo-oxidative stability, and maintained good dielectric properties.

Full text of DOI:10.1002/pola.28020

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Synthesis of Bisphenol A-Free Oligomeric Phthalonitrile Resins
with Sulfone and Sulfone-Ketone Containing Backbones
1
1
1
1
Matthew Laskoski, Jadah S. Clarke, Arianna Neal, Holly L. Ricks-Laskoski,
2
1
W. Judson Hervey, Teddy M. Keller
1
Chemistry Division, Materials Chemistry Branch, Naval Research Laboratory, Washington, District of Columbia 20375-5320
Center for Biomolecular Science and Engineering, Code 6910, Naval Research Laboratory, Washington, District of Columbia
2
20375-5320
Correspondence to: M. Laskoski (E-mail: matthew.laskoski@nrl.navy.mil)
Received 19 January 2015; accepted 1 December 2015; published online 8 January 2016
DOI: 10.1002/pola.28020
ABSTRACT: Two new oligomeric sulfone and sulfone-ketone con-
taining phthalonitrile (PN) resins with excellent processability
have been developed. The PN monomers were prepared from
the reaction of an excess amount of bisphenol S with 4-(chloro-
phenyl)sulfone or 4,4-dichlorobenzophenone in the presence of a
base in a solvent mixture (dimethylsulfoxide/toluene), followed
by end-capping with 4-nitro-PN in a two-step, one-pot reaction.
These PN resins exhibited good viscosities and cure times for
molding into various shapes. After being thermally cured to yield
crosslinked polymers, these polymers demonstrated superb
mechanical properties, thermo-oxidative stability, and maintained
†
good dielectric properties. Published 2016. J. Polym. Sci., Part A:
Polym. Chem. 2016, 54, 1639–1646
KEYWORDS: resins; thermal properties; thermosets
INTRODUCTION Polysulfone thermoplastics containing sul-
fonyl linkages are widely used in high temperature applica-
tions due to their high glass transition temperatures (T )
g
have excellent processability and the thermoset polymers
exhibit outstanding thermal and mechanical properties, low
water absorption (<1.5%), low flammability and have the
potential to replace many current high temperature resins
(e.g., PMR-15) so long as the cost of the resin is competi-
tive with current technologies.
and outstanding thermal, oxidative, hydrolytic, and mechani-
cal properties. They have several important applications
such as membranes in fuel cells and in the manufacture of
1
3
1
–5
various composite structures.
However, the presence of a
PN resins can be grouped into two distinct classes. The first
generation PN resins, consisting of small crystalline molecules
Tg limits their full thermal potential for composite structural
applications to around 250 8C.
0
synthesized from the phenolate salts of bisphenols (e.g. 4, 4 -
High temperature thermosetting resins, which garner inter-
biphenol, bisphenol A, bisphenol S, bisphenol A6F, etc.) and 4-
nitrophthalonitrile, exhibited high melting points (195–230
est due the their high T and ease of processing from small
g
1
4–19
molecules, have been recently gaining popularity as replace-
8C) and were difficult to process effectively.
The second
6
ments for metals in aerospace and marine applications.
generation PN resins greatly improve upon the processing
limitations of the first generation PN by using oligomeric aro-
matic ether spacers between the terminal PN groups. The PN
resins are prepared through a two-step, one-pot nucleophilic
displacement reaction between activated or unactivated
Three classes of resins have emerged from this research;
7
,8
9,10
cyanate esters,
polyimides, and phthalonitriles (PN).
Cyanate esters are a class of resins that are easy to process,
cure at low temperatures, and can exhibit high T s, but suffer
g
from long term hydrolytic stability issues and have upper
temperature limits around 400 8C. Polyimides, on the other
dihaloaromatic compounds and an appropriate bisphenol, fol-
1
1
10,20–23
lowed by end-capping with 4-nitrophthalonitrile.
More
hand, are more difficult to process than other resins,
although they have a high T , long-term thermal stability and
g
recently, we have developed an improved synthesis of a
sulfonyl-containing PN resin composition by reacting 4-(chlor-
ophenyl) sulfone with an excess of bisphenol A followed by
end-capping with 4-nitrophthalonitrile. Although this PN
resin had excellent properties, the inclusion of bisphenol A
limited the thermal stability to 400 8C. In addition, bisphenol
good mechanical properties. Typically, polyimides absorb a
fair amount of water (>5%), which limits their use in some
high temperature applications by effecting their structural
2
3
1
2
integrity and mechanical properties. PN resins, however,
†
Published 2016. This article is a U.S. Government work and is in the public domain in the USA.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646
1639

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
was allowed to equilibrate for 1–4 min depending on the
experiment.
Dielectric measurements were performed on precast films
using a HP 4291A Network Analyzer sweeping between 1
MHz and 1.8 GHz and calibrated with HP Kit 4291A. Sample
thicknesses were 1.0794 mm (6a), 1.543 mm (6b), and
FIGURE 1 Structure of PN resin 1 containing sulfonyl linkages.
1.653 mm (6c) while a parallel plate (HP 16453A) cell was
calibrated to Teflon.
A is a known endocrine disruptor, which may limit its future
usage in some applications.
Mass spectrometry data were recorded on a QTRAP 5500
mass spectrometer (AB Sciex, Foster City, CA) under control
of the Analyst 1.6.2 software. Polymers were reconstituted
in a solution of 0.1% trifluoracetic acid in acetonitrile and
infused via a TurboV electrospray ionization source (ESI) at
Due to the excellent physical properties exhibited by the
incorporation of sulfonyl linkages into PN resins, we sought
to determine what effect replacing bisphenol A with bisphe-
nol S had on the overall properties of the resin and resulting
thermosetting polymer (Fig. 1). Additionally, knowing that a
polysulfone backbone has a high affinity for self-interaction
between the sulfonyl linkages, we also incorporated a ketone
linkage to disrupt the symmetry and reduce the interaction
between the polar sulfonyl units. The synthesis of these two
new resins systems will be reported in this paper along with
a comparison of the rheological properties. The effect of
structural changes within the PN resin on the processing
parameters as well as the thermal, mechanical, and other
physical properties will be discussed in detail.
2
1
a flow rate of 2 mL min . Liquid connections between the
on-board syringe pump and the ionization source were
constructed of 100 mm i.d. fused silica coupled to an in-line
filter fitting (M-520, IDEX/Upchurch) as previously
2
4
described. Mass spectra were acquired in positive mode
ESI.
Synthesis of the PN Resin 1a
To a 1000 mL, three-necked flask fitted with a thermometer,
a Dean-Stark trap with condenser and a nitrogen inlet, were
added bisphenol S 2 (30 mol % excess relative to 3a)
(
0
50.0 g, 0.200 mol), bis(4-chlorophenyl) sulfone 3a (20.0 g,
.070 mol), sodium hydroxide (14.4 g, 0.360 mol), potassium
EXPERIMENTAL
carbonate (11.0 g, 0.080 mol), toluene (50 mL), and dime-
thylsulfoxide (DMSO) (300 mL). Toluene was added to con-
trol the refluxing azeotropic removal of water and to control
the temperature of the reaction content. The resulting mix-
ture was degassed with nitrogen at ambient temperature
and the Dean-Stark trap was filled with toluene. The mixture
was refluxed at 150 - 160 8C under a nitrogen atmosphere
for 12 h or until no more water was observed being col-
lected in the Dean-Stark trap. The toluene was then slowly
distilled off causing the temperature to rise in the reaction
vessel, which caused an enhancement of the yield, and high
conversion of the intermediate hydroxyl salt 4a. After the
mixture was cooled to 50 8C, 4-nitrophthalonitrile 5 (46.2 g,
All starting materials were of reagent grade and used with-
out further purification. Differential scanning calorimetric
(
2
DSC) analysis was performed on a TA Instruments DSC
920 modulated thermal analyzer at a heating rate of 10 8C
21
3
21
min and a nitrogen purge of 50 cm min Thermogravi-
metric analysis (TGA) was performed on a TA Instruments
2
1
TGA Q50 at a heating rate of 10 8C min under a nitrogen
3
21
or air purge of 100 cm min . Infrared (IR) spectra were
recorded on films deposited on NaCl plates using a Nicolet
iS50 FTIR spectrometer. Solution NMR was performed on a
Br u€ ker ADVANCE 300 spectrometer.
Rheometric measurements from 25 to 400 8C were per-
formed on a TA Instruments AR-2000 Rheometer, in conjunc-
tion with an environmental testing chamber for temperature
control and torsion fixtures to monitor the response of poly-
meric samples (50 3 13 3 2 mm ) to oscillatory testing.
The measurements were made under a nitrogen atmosphere
0
.267 mol) was added in one portion and the reaction mix-
ture was heated at 80 8C for 6 - 8 h. The mixture was
allowed to cool to ambient temperature and poured into a
5% (w/w) aqueous HCl solution resulting in the formation
3
of a solid. The material was broken up and collected using a
B u} chner funnel. The solid was washed with 200 mL portions
of distilled water mixed with 20% ethanol until a neutral pH
was reached. The isolated solid was vacuum dried to yield
the 2:1 oligomeric sulfone containing PN composition 1a
(88.8 g, 97%) as an off-white amorphous solid, which is a
over the temperature range of ꢀ25–400 8C. A temperature
2
1
ramp of 3 8C min was used to determine the storage mod-
ulus and damping factor (tan d) of the material at a fre-
2
2
quency of 1 Hz and a strain of 2.5 3 10 %. Normal force
control was used throughout the tests to keep the samples
taut. Viscosity measurements were performed at various
temperatures using a 25 mm top plate and 40 mm diameter
cup shaped bottom plate at a frequency of 1 Hz and a strain
1
mixture of ꢀ70% n 5 1 and ꢀ30% n 5 0. H NMR (300
MHz, CDCl , d): 7.93 (m, aromatic-H), 7.85 (m, aromatic-H),
3
7.70 (d, aromatic-H), 7.24 (d, aromatic-H), 7.16 (s, aromatic-
H), 7.10 (m, aromatic-H), 7.01 (d, aromatic-H), 7.01 (s, aro-
matic-H). IR (NaCl): k 3059 (s; C@C), 2236 (s; CN), 1596 (w;
C@C), 1507 (s; aromatic), 1490 (s; aromatic), 1315 (w,
AOA), 1125 (s, SO ). Calc. For M1 (C H N O S) 502.50
2
2
of 2.5 3 10 %. The rheometric chamber was heated to the
desired temperature and the resin placed onto the bottom
plate, melted and the gap was set to 100 lm. The chamber
2
28 14 4 4
1
640
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
(
n 5 0), (C H N O S ) 967.01 (n 5 1). Observed ESI-MS
Sample Preparation for Dielectric Measurements
5
2 30 4 10 3
m/z ratios for the singly-charged n50 and n51 species
Samples for dielectric measurements were prepared by
were 503.0 and 967.9, respectively.
degassing 1a or 1b under vacuum at 200 8C for 4 h in a
25 mm round metal pan. m-BAPS (2.7 mol %) was added
with stirring and the samples were cooled, placed in an
oven, and heated to gelation under argon at 200 8C for 16 h
Synthesis of the PN Resin 1b
To a 2000 mL, three-necked flask fitted with a thermometer,
a Dean-Stark trap with condenser and a nitrogen inlet, were
added bisphenol S 2 (30 mol % excess relative to 3b)
(
overnight), then heated to 375 8C for 8 h (ramped at 3 8C
21
min ). The cured polymer 6 were removed from the mold
and sanded to a uniform thickness (ꢀ 1 mm) using 220 and
(
0
100 g, 0.400 mol), 4, 4’-dichlorobenzophenone 3b (35.12 g,
.140 mol), potassium carbonate (115.80 g, 0.839 mol), tolu-
600 grit sandpapers.
ene (200 mL), and DMSO (1000 mL). The nitrogen degassed
mixture was refluxed at 150–160 8C under a nitrogen atmos-
phere for 16 h or until no more water was observed being
collected in the Dean-Stark trap. The toluene was then
slowly distilled off causing the temperature to rise in the
reaction vessel, which resulted in an enhancement of
the yield, and high conversion of the intermediate hydroxyl
salt 4b. After the mixture was cooled to 50 8C, 4-
nitrophthalonitrile 5 (91.5 g, 0.529 mol) was added in one
portion and the reaction mixture was heated at 80 8C for 6–
Sample Preparation for Water Uptake
and Oxidative Aging Studies
Samples for water uptake studies measured ꢀ20 3 13 3
3
2
mm . The previously dried samples were heated in dis-
tilled water for 24 h at 100 8C before weighing.
Samples for oxidative aging studies measured ꢀ50 3 13 3
3
2 mm for all experiments. The samples were placed in a
330 8C muffle furnace under a flow of air and periodically
removed and weighed.
8
h. The mixture was allowed to cool to ambient tempera-
RESULTS AND DISCUSSION
ture and the aqueous workup was performed identical to 1a.
The isolated solid was convection dried at 50 8C to yield the
The synthesis of the amorphous multiple aromatic ether
sulfonyl-linked PN resin 1a was achieved via a nucleophilic
displacement reaction between bisphenol S 2 (30% excess)
and bis(4-chlorophenyl) sulfone 3a under basic conditions in
dimethylsulfoxide (DMSO) using toluene as an azeotropic
solvent to remove the water formed during the reaction
(Scheme 1). In this reaction, the chlorine group in compound
3a is extremely easy to displace due to the highly activating
nature of the sulfonyl moiety. We also used a mixed base
system of 90 mol % sodium hydroxide to 20 mol % potas-
2
:1 oligomeric sulfone-ketone containing PN composition 1b
(
ꢀ
186.6 g, 98%), which is a mixture of ꢀ70% n 5 1 and
1
30% n 5 0. H NMR (300 MHz, CDCl , d): 8.05 (s, aro-
3
matic-H), 8.01 (d, aromatic-H), 7.95 (d, aromatic-H), 7.88 (d,
aromatic-H), 7.84 (d, aromatic-H), 7.78 (m, aromatic-H), 7.32
(
7
s, aromatic-H), 7.30 (s, aromatic-H), 7.29 (s, aromatic-H),
.24 (s, aromatic-H), 7.20 (s, aromatic-H), 7.15 (m, aromatic-
H), 7.09 (s, aromatic-H). IR (NaCl): k 3060 (w; C@C), 2235
s; CN), 1652 (s, C@O), 1597 (s; C@C), 1508 (s; aromatic),
492 (s; aromatic), 1315 (w, AOA), 1174 (w, CAO), 1126 (s,
SO ). Calc. For M1 (C28 S) 502.50 (n 5 0),
) 930.96 (n 5 1). Observed ESI-MS m/z ratios
were for the singly charged n 5 0 and n 5 1 species were
03.0 and 931.2, respectively.
(
1
sium carbonate (K CO ) relative to each reactive hydroxyl
2 3
2
H
14
N
4
O
4
group on 2. This first reaction was heated at 150–160 8C for
16 h and contained an excess amount of 2 relative to 3a
yields a dimetallic phenolate intermediate 4a, which is com-
posed of a mixture of oligomeric sulfonyl-containing pheno-
lates (n 5 1) and bisphenol S diphenolate (n 5 0) in an
53 30 4 9 2
(C H N O S
5
ꢀ
70:30 ratio. Formation of this intermediate was considered
complete when H-NMR spectroscopy confirmed that all of
a had reacted. However, the synthesis of the sulfonyl-
Preparation of Monomer/m-BAPS Mixtures for DSC
Analysis
To the melt of 1a or 1b at 200 8C was added 2.7 or 5.0 mol
1
3
ketone containing PN resin 4b was not as simple. First
attempts using the NaOH and K CO3 mixed base system
failed. This is presumably due to the chlorine group in com-
pound 3b being more difficult to displace than the one on
%
of bis[4-(3-aminophenoxy)phenyl]sulfone (m-BAPS). Once
the curing additive had been evenly dispersed by stirring for
min, the sample was cooled and used in the DSC studies.
2
2
3
a, In addition, the hydrolytic side reactions seen previously
Sample Preparation for Rheometric Measurements
Samples for rheometric measurements were prepared by
degassing 1a or 1b under vacuum at 200 8C for 4 h in an
in the synthesis of high molecular weight polysulfones
appears to be a factor that alters the reaction stoichiometry
involving 3b. This was determined by the presence of a
2
5
2
1
aluminum mold with cavity dimensions of 50 3 13 mm . m-
doublet in the H-NMR around 6.9 ppm, which has been
BAPS was added with stirring. The samples were cooled,
placed in an oven, and heated to gelation under argon at
identified as the protons a to the OH group on an aromatic
phenol. Upon cooling to below 50 8C, 4-nitrophthalonitrile 5
was slowly added to the dimetallic phenolate intermediate
mixture 4 producing the PN 1a and 1b in nearly quantitative
yields after heating at 80 8C for 6–8 h, in which there is the-
oretically around 70% of n 5 1 (oligomeric PN) and 30% of
n 5 0 (bisphenol S PN). The PN resin compositions 1a and
2
00 8C for 16 h (overnight), then heated to 375 8C for 8h
2
1
(
ramped at 3 8C min ). Selected samples were postcured
for an additional 2 h at 415 8C. The cured polymers 6 were
removed from the mold and sanded to a thickness of
ꢀ
2 mm before testing using 220 and 600 grit sandpapers.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646
1641

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
the n 5 0 which significantly alters the viscosity profile by act-
ing as a reactive plasticizer, should also make the thermoset
more rigid by increasing the crosslinking density. The viscos-
ity values for both PN resin compositions 1 rival resins such
as cyanate ester and polyimides that are currently used in the
manufacture of composites by resin transfer molding (RTM)
under similar thermal conditions.
Polymerization studies of PN resins 1a and 1b were
achieved by DSC analyses up to 400 8C in the presence of
either 2.7 or 5.0 wt % of bis(3-[4-aminophenoxy]phenyl)sul-
fone (m-BAPS) to afford thermoset 6. Since the molecular
weights of both resins are very similar, the use of wt % ver-
sus mol % of catalyst did not affect the results. Upon melt-
ing, 1a had a higher viscosity than 1b which affected the
polymerization rate. The PN resin mixture 1a containing 2.7
wt % of m-BAPS exhibited two T s around 100 and 145 8C
g
(
Fig. 2), followed by an exothermic transition peaking at 245
8
C. The appearance of two T s transitions are likely due to
g
the two different components (n 5 1 and n 5 0) contained
within the resin. Samples containing 5.0 wt % of m-BAPS
showed a two stage T at around 100 and 135 8C followed
g
by a stronger exothermic transition peaking at 235 8C. The
lower viscous PN resin mixture 1b containing 2.7 wt % of
m-BAPS, exhibited T s around 80 and 138 8C (Fig. 3), fol-
g
lowed by a rather broad complex exothermic transition peak-
ing at 235 8C. Similarly, samples of 1b containing 5.0 wt %
of m-BAPS showed slightly higher T s at 92 and 145 8C,
g
SCHEME 1 Synthesis of PN monomers 1a and 1b and poly-
mers 6.
likely due to some reaction of the curing additive with 1b
during mixing, followed by a much stronger but similarly
shaped exothermic transition peaking at 235 8C. These
results indicate that a higher amount of m-BAPS appears
necessary to cure 1a at a faster rate than 1b, which could
be a result of 1a having a higher viscosity and less mobility
than 1b at all temperatures.
1b were soluble in common organic solvents such as tolu-
ene, acetone, methylene chloride, and DMSO. The structures
1
of resins 1a and 1b were confirmed by FTIR, H-NMR, and
mass spectroscopy. The mass spectrometry of both resins
indicated the composition contained both n 5 0 and n 5 1
with no higher oligomers. The ratio of n 5 1 to n 5 0 can be
readily varied by changing the amount of excess 2 relative to
Rheometric viscosity analysis of resin 1a and 1b containing
2
2
.7 and 5.0 wt % of m-BAPS was performed at 200 and
25 8C. As with most high temperature thermosetting resins,
3
but for the scope of this article, we chose to study the
composition containing ꢀ70% of the n 5 1 PN resin and
0% of the n 5 0 PN resin.
the PN resin pot-life can be controlled as a function of
3
The viscosity of the PN resin compositions 1 at various tem-
peratures is an important factor in determining the processing
parameters needed to mold the resin into various shapes.
Upon heating 1a to 150 8C, a viscosity of 180,000 cP was
observed, which quickly dropped with further heating to
7,000 cP at 175 8C, 800 cP at 200 8C, and 250 cP at 225 8C.
As a comparison, we performed viscosity measurements on
samples of 1b, which contained a ketone moiety instead of a
sulfone moiety in the oligomeric structure. In this case, viscos-
ity values of 7,400, 1,100, 475, and 175 cP were observed at
150, 175, 200, and 225 8C, which is markedly lower than 1a
at any temperature. This difference is most likely a result of
the ketone group in 1b being less polar relative to the highly
polar, self-interacting sulfone groups in 1a, which apparently
is strong enough to act as a physical crosslinking center.
These newly synthesized compositions containing an excess of
FIGURE 2 DSC curves of 1a with 2.7 wt % (top) and 5.0 wt %
m-BAPS (Bottom).
1
642
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 4 Rheometric viscosity plot at 200 8C with 2.7 wt % of
m-BAPS; resin 1a (A) and 1b (B) and with 5.0 wt % of m-BAPS;
resin 1a (C) and 1b (D). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
FIGURE 3 DSC curves of 1b with 2.7 wt % (top) and 5.0 wt %
m-BAPS (Bottom).
heating time at various temperatures (Figs. 4 and 5). As
shown previously, the initiation reaction (early/flat part of
curve) and the propagation reaction (part of curve where
the viscosity increases exponentially) were clearly visible
Rectangular shaped samples were prepared for 6a and 6b
post-cured at a maximum temperature of 375 8C and ana-
lyzed mechanically using a rheometer from 25 to 400 8C. We
observed (Fig. 6) that the storage modulus for 6a cured with
either 2.7 (Plot A) or 5.0 (Plot C) wt % of m-BAPS showed a
2
6
(
Fig. 4). Heating above 150 8C initiates a chemical cross-
linking reaction resulting in a liquid to rubbery phase
0
00
change. The intersection of storage (G ) and loss (G ) moduli,
which we take as the gel point, corresponds to ꢀ100,000 cP
glass transition (T ) at approximately 325 and 375 8C,
g
2
3
respectively. This transition is also observed as a spike in
the tan d values near those temperatures (Fig. 7). The higher
on the viscosity plots of 1a and 1b (Figs. 4 and 5).
The time to gelation for 1a and 1b were found to be a func-
tion of the curing temperature. Samples 1a and 1b contain-
ing 2.7 and 5.0 wt % of m-BAPS reached the gel point at
amount of m-BAPS did result in a higher T . Unlike 6a, 6b
g
showed a gradual decreased in the storage modulus from
ambient temperature to 400 8C indicating some flexibility
within the thermoset with a rise in the temperature. How-
ever, this mobility does not result in complete loss of
mechanical stability. When heated from 25 to 400 8C, the
storage modulus values for 6b that had been post-cured to
200 8C after 45 and 25 min and 33 and 25 min, respectively.
For example, 1a took longer to gel at the lower concentra-
tion (2.7 wt %) of catalyst relative to 1b, which parallels the
observed lower intensity of the exotherm in the DSC analysis
indicating a slower rate of curing (Figs. 2 and 3). However,
at the higher concentration (5.0 wt %), the gel times are the
same for 1a and 1b but the shape of the exothermic curves
are different. For 1a, the initiation step is just a few minutes
longer than 1b but they both eventually catch up to one
another and gel at around 25 min. At 225 8C, all gel times
are significantly shorter for both 1a and 1b. Samples con-
taining 2.7 and 5.0 wt % of m-BAPS, reaching the gel points
at 225 8C after around 15 and 7 min for 1a and around 6
and 3.5 min for 1b, respectively. Again, we observe a longer
gel time at both catalyst concentrations for 1a when com-
pared to 1b. This phenomenon can be explained by the
higher viscosity of 1a vs. 1b, where the catalyst and resin
are less mobile and also due to the strong self-interactions
of the sulfone linkages in 1b that inhibits the movement of
the resin and curing additive to find neighboring molecules
in which to react. As seen in the data for both systems, the
resin can be exploited and utilized using various processing
conditions and catalyst concentrations to achieve the desired
cure times depending on the application. Overall, resins 1a
and 1b demonstrate excellent viscosity parameters and rea-
sonable cure times for potential component fabrication.
375 8C gradually changed from 1230 to 500 MPa and 1300
FIGURE 5 Rheometric viscosity plot at 225 8C with 2.7 wt % of
m-BAPS; resin 1a (A) and 1b (B) and with 5.0 wt % of m-BAPS;
resin 1a (C) and 1b (D). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646
1643

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 6 Rheometric analysis of polymers 6 with 2.7 wt % of
m-BAPS; 6a (A) and 6b (B) and with 5.0 wt % of m-BAPS; 6a
FIGURE 8 Rheometric analysis of polymers 6 with 2.7 wt % of
m-BAPS; 6a (A) and 6b (B) and with 5.0 wt % of m-BAPS; 6a
(C) and 6b (D) post-cured to 415 8C. [Color figure can be viewed
in the online issue, which is available at wileyonlinelibrary.
com.]
(
C) and 6b (D) post-cured to 375 8C. [Color figure can be viewed
in the online issue, which is available at wileyonlinelibrary.
com.]
a full cure and the absence of a Tg for 6a. However, the
increased amount of curing additive (5.0 wt % of m-BAPS)
affected the mechanical integrity of the thermoset 6a and
to 400 MPa for the 2.7 and 5.0 wt % of m-BAPS cured sam-
ples, respectively. From these data, we can conclude that 6b,
regardless of the amount of curing additive and thermal cur-
ing conditions, is more fully cured than 6a reflecting what
was observed during the DSC studies.
6b, especially after curing above 400 8C, indicated by the
lower storage modulus for both samples.
The thermal and thermo-oxidative properties were investi-
gated between 25 and 1000 8C in a TGA chamber. Figure 9
shows the TGA thermograms for polymers 6a and 6b cured
to 375 8C with 2.7 mol % of m-BAPS. The thermal and oxida-
tive stabilities were determined on powdered samples, which
represent a worst case scenario in thermal performance but
is the most accurate way to compare the two systems since
thermal stability is highly dependent on sample surface area.
Upon postcuring to 415 8C for 2h, the storage modulus val-
ues for 6a gradually changed from 1230 to 500 MPa and
885 to 300 MPa for the 2.7 and 5.0 wt % of m-BAPS cured
samples, respectively (Fig. 8). Polymer 6b, however, gradu-
ally changed from 1285 to 620 MPa and 1125 to 530 MPa
for the 2.7 and 5.0 wt % m-BAPS cured samples, respec-
tively. In both samples, the tan d plots were featureless over
the entire temperature range. From this data, we can con-
clude that post-curing above 400 8C is necessary to achieve
FIGURE 7 Tan d plot of polymer 6 with 2.7 wt % of m-BAPS;
FIGURE 9 TGA of polymer 6a and 6b with 2.7 wt % of m-BAPS;
6a (A) and 6b (B) under nitrogen and 6a (C) and 6b (D) under air
after postcuring to 375 8C. [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
6a (A) and 6b (B) and with 5.0 wt % of m-BAPS; 6a (C) and 6b
(
D) post-cured to 375 8C. [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
1
644
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Polymers 6a and 6b retained 95% weight at 450 and 480 8C
and exhibited char yields of 60 and 52%, respectively, when
heated under inert conditions. Both polymers showed weight
retention in air of 95% at 450 8C, with catastrophic decom-
position occurring slowly between 500 and 700 8C. Addi-
tional postcuring of 6a to 415 8C improved the weight
retention to 480 8C and the char yield to 58%. This postcur-
ing had little effect on the properties of 6b. These values are
consistent with what is seen for most PN resin systems,
which is a significant improvement over traditional thermo-
plastic polysulfones, which illustrates the unique effects that
the PN crosslinking unit has on the thermal and oxidative
and 1b exhibit excellent viscosity parameters under process-
ing conditions that rivals current technologies with resin 1b
showing the lower viscosity of the two systems at any given
temperature. The thermal, oxidative, and mechanical proper-
ties of polymers 6 can be controlled as a function of the cur-
ing additive concentration and the post-curing conditions.
Although, the substitution of bisphenol A with bisphenol S in
these resin systems has a detrimental effect on most of the
physical properties of the reported resins and polymers, these
systems exhibit properties that are still far superior than cur-
rent resin systems and the benefit of removing bisphenol A
can outweigh the slight reduction in performance.
2
7,28
properties.
When the PN resins 1a and 1b are fully cured, polymers 6
displays many important attributes including the absence of
To simulate the use of polymer 6 under real environmental
conditions, we assessed the effect of long-term heating or
aging at high temperatures under air. By subjecting polymers
a T when fully cured, which translates to minimal loss of
g
structural integrity at elevated temperatures. Additionally, 6
displays excellent thermo-oxidative properties, low dielectric
constants and relatively low water absorption. One aspect of
all oligomeric PN resins including 1a and 1b is that B-staged
prepolymers exhibit indefinite shelf-life under ambient con-
ditions but flow above 150 8C and cure above 200 8C, which
enhances its importance in the fabrication of composite com-
ponents by emerging cost effective manufacturing methods
such as RTM, resin infusion molding (RIM), filament winding,
etc. Due to their physical properties, polymers 6 have the
potential to be used in structural components for ships, mis-
siles, aircraft, high temperature tooling, and many other mili-
tary and domestic applications where the properties
exhibited by the PN polymers are essential.
6
a and 6b to continuous heating at 329 8C (625 8F), a cumu-
lative weight loss of 10 and 8.5 % and 16 and 18 % was
observed for polymers 6a and 6b, postcured at 375 8C, for
100 and 150 h, respectively. These values are exceptional
considering these are all organic polymeric systems. These
weight loss values would be expected to be considerable bet-
ter for fiber reinforced composites where the resin content
is around 30%.
High temperature polymeric materials have potential use in
many electronic applications including radomes, which
require the polymer to be transparent to microwaves meas-
ured by a low dielectric constant (D ), low dissipation factor
f k
or tan d (D ), and low water absorption. The D values for
k
2
9
samples of 6 postcured at 375 8C were measured from 1
MHz to 1.8 GHz and found to be 3.4 with the D in the range
f
of 0.01. These values are similar to other resins currently
being used in the fabrication of radomes and other electrical
ACKNOWLEDGMENTS
2
4
components.
reported PN polymers,
presence of the highly polar SO groups in the backbone of
They are slightly higher than previously
The authors would like to thank SERDP for financial support of
this project. The authors thank the Office of Naval Research and
the ONR/ASEE Science and Engineering Apprenticeship Pro-
gram (SEAP) for financial support of this project.
3
0,31
which were closer to 3.0 but the
2
6
a and 6b could account for the higher D values. Addition-
k
ally, samples of thermosets 6 were heated in distilled water
at 100 8C for 24 h, dried, and weighed. The maximum
amount of water absorption for polymers 6a and 6b over
that time was around 4.8 and 3.9% by weight, respectively.
The water absorption for these systems is higher than other
PN polymers, which are closer to 1 to 2% by weight, but is
not surprising due to the presence of many polar groups in
the backbone. In a composite, the overall water absorption
would be far less.
REFERENCES AND NOTES
1
C. Camacho-Zuniga, R. F. A. uiz-Trevino, S. Hernandez-Lopez,
M. G. Zolotukhin, F. H. J. Maurer, A. Gonzalez-Montiel, J.
Membr. Sci. 2009, 340, 221–226.
2 F. Wang, M. Hickner, Y. S. Kim, T. A. Zawodzinski, J. E.
McGrath, J. Membr. Sci. 2002, 197, 231–242.
3 B. Staude, L. Breitbach, J. Appl. Polym. Sci. 1991, 43, 559–
5
66.
CONCLUSIONS
4 F. Lufrano, G. Squadrito, A. Patti, E. Passalacqua, J. Appl.
Polym. Sci. 2000, 77, 1250-2–1256.
The synthesis of two PN resins compositions 1a and 1b con-
taining sulfone and sulfone-ketone groups in the backbone
has been achieved in high yield by reaction of 4-(chloro-
5
M. Esanu, M. Rusu, V. Cozan, E. Butucm, Mater. Plast. 1998,
3
5–225–230.
6
2
A. Badshah, M. R. Kessler, Z. Heng, A. Hasan, Polym. Int.
014, 63, 465–469.
0
phenyl) sulfone 3a or 4, 4 -dichorobenzophene 3b with an
excess of bisphenol S 2. The excess bisphenol S creates resins
in which there is approximately 30 mol % of a bisphenol S
PN resin, which acts as a reactive plasticizer and lowers the
overall viscosity of the two resin systems. Both PN resins 1a
7
J. J. Cash, M. C. Davis, M. D. Ford, T. J. Groshens, A. J.
Guenthner, B. G. Harvey, K. R. Lamison, J. M. Mabry, H. A.
Meylemans, J. T. Reams, C. M. Sahagun, Polym. Chem. 2013,
4, 3859–3865.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646
1645

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
8
B. G. Harvey, A. J. Guenthner, W. W. Lai, H. A. Meylemans,
21 M. Laskoski, D. D. Dominguez, T. M. Keller, J. Polym. Sci.
Part A: Polym. Chem. 2013, 51, 4774–4778.
M. C. Davis, L. R. Cambrea, J. T. Reams, K. R. Lamison, Macro-
molecules 2015, 48, 3173–3179.
2
2 D. D. Dominguez, T. M. Keller, High Perform. Polym. 2006,
18, 283–304.
3 M. Laskoski, T. M. Keller, H. Ricks-Laskoski, C. Giller, Macro-
mol. Chem. Phys. 2015, 17, 1808–1815.
4 J. Hervey, G. Khalsa-Moyers, P. K. Lankford, E. T. Owens, C.
9
5
M. O. Abdalla, D. Dean, S. Campbell, Polymer 2002, 43, 5887–
893.
2
1
0 M. Laskoski, A. Neal, T. M. Keller, D. D. Dominguez, C. A.
Klug, A. P. Saab, J. Polym. Sci. Part A: Polym. Chem. 2014, 52,
1
2
662–1668.
K. McKeown, T. Y. Lu, L. J. Foote, K. G. Asano, J. L. Morrell-
Falvey, W. H. McDonald, D. A. Pelletier, G. B. Hurst, J. Pro-
teome Res. 2009, 8, 3675–3688.
1
1 L. J. Kasehagen, I. Haury, C. W. Macosko, D. A. Shimp, J.
Appl. Polym. Sci. 1997, 64, 107–113.
2 W. P. Paplham, R. A. Brown, I. M. Salin, J. C. Seferis, J.
Polym. Sci. 1995, 75, 133–137.
1
2
5 R. Viswanathant, B. C. Johnson, J. E. McGrath, Polymer
984, 25, 1827–1836.
1
13 W. Xie, W. P. Pan, K. C. Chuang, Thermochim. Acta 2001,
2
6 T. Fuchs, W. Richtering, W. Burchard, K. Kajiwara, S.
3
67, 143–153.
Kitamura, Polym. Gels Networks 1997, 5, 541–559.
7 D. Parker, Bussink, J. H. T. van de Grampel, G. W.
14 T. M. Keller, C. M. Roland, U. S. Patent 1992, 7, 945.
2
1
5 T. M. Keller, In Proceedings of the 18th International SAMPE
Wheatley, E. U. Dorf, E. Ostlinning, K. Reinking, Ullmann’s
Encyclopedia of Industrial Chemistry, 2000.
Technical Conference, Las Vegas, NV, 1986; Vol. 31, pp 528–
5
1
1
34.
2
8 X. G. Li, M. R. Huang, React. Funct. Polym. 1999, 42, 59–64.
6 T. M. Keller. Chemtech. 1988, 18, 635–639.
2
9 C. P. R. Nair, D. Mathew, K. N. Ninan, Adv. Polym. Sci.
7 J. F. Giuliani, T. M. Keller, Sens. Mater. 1989, 1, 2247–2252.
2001, 155, 1–99.
1
1
8 S. B. Sastri, J. P. Armistead, T. M. Keller, Polym. Compos.
996, 17, 816–822.
30 M. Laskoski, A. Neal, M. B. Schear, T. M. Keller, H. L. Ricks-
Laskoski, A. P. Saab, J. Polym. Sci. Part A: Polym. Chem. 2015,
53, 2186–2191.
1
9 S. B. Sastri, J. P. Armistead, T. M. Keller, U. Sorathia,
Polym. Compos. 1997, 18, 48–54.
31 M. Laskoski, M. B. Schear, A. Neal, D. D. Dominguez, H.
L. Ricks-Laskoski, J. Hervey, T. M. Keller, Polymer 2015, 67,
185
20 M. Laskoski, D. D. Dominguez, T. M. Keller, J. Polym. Sci.
Part A: Polym. Chem. 2005, 43, 4136–4143.
1
646
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1639–1646