Silicon-containing trifunctional and tetrafunctional cyanate esters: Synthesis, cure kinetics, and network properties

Article abstract of DOI:10.1002/pola.27052

The synthesis and physical properties of new silicon-containing polyfunctional cyanate ester monomers methyl[tris(4-cyanatophenyl)]silane and tetrakis(4-cyanatophenyl)silane, as well as polycyanurate networks formed from these monomers are reported. The higher crosslinking functionality compared to di(cyanate ester) monomers enables much higher ultimate glass transition temperatures to be obtained as a result of thermal cyclotrimerization. The ability to reach complete conversion is greatly enhanced by cocure of the new monomers with di(cyanate ester) monomers such as 1,1-bis(4-cyanatophenyl)ethane. The presence of silicon in these polycyanurate networks imparts improved resistance to rapid oxidation at elevated temperatures, resulting in char yields as high as 70% under nitrogen and 56% in air in the best-performing networks. The water uptake in the silicon-containing networks examined is 4-6 wt % after 96 h of immersion at 85°C, considerably higher than both carbon-containing and/or di(cyanate ester) analogs.

Full text of DOI:10.1002/pola.27052

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Silicon-Containing Trifunctional and Tetrafunctional Cyanate Esters:
Synthesis, Cure Kinetics, and Network Properties
Andrew J. Guenthner,1 Vandana Vij,2 Timothy S. Haddad,2 Josiah T. Reams,2
Kevin R. Lamison,2 Christopher M. Sahagun,3 Sean M. Ramirez,2 Gregory R. Yandek,1
Suresh C. Suri,1 Joseph M. Mabry1
1Aerospace Systems Directorate, Air Force Research Laboratory, Edwards AFB, California 93524
2ERC Incorporated, Air Force Research Laboratory, Edwards AFB, California 93524
3National Research Council/Air Force Research Laboratory, Edwards AFB, California 93524
Correspondence to: A. J. Guenthner (E-mail: andrew.guenthner@edwards.af.mil)
Received 25 October 2013; accepted 22 November 2013; published online 18 December 2013
DOI: 10.1002/pola.27052
ABSTRACT: The synthesis and physical properties of new
silicon-containing polyfunctional cyanate ester monomers
methyl[tris(4-cyanatophenyl)]silane and tetrakis(4-cyanatophe-
nyl)silane, as well as polycyanurate networks formed from
these monomers are reported. The higher crosslinking func-
tionality compared to di(cyanate ester) monomers enables
much higher ultimate glass transition temperatures to be
obtained as a result of thermal cyclotrimerization. The ability
to reach complete conversion is greatly enhanced by cocure of
the new monomers with di(cyanate ester) monomers such as
1,1-bis(4-cyanatophenyl)ethane. The presence of silicon in
these polycyanurate networks imparts improved resistance to
rapid oxidation at elevated temperatures, resulting in char
yields as high as 70% under nitrogen and 56% in air in the
best-performing networks. The water uptake in the silicon-
containing networks examined is 4–6 wt % after 96 h of immer-
sion at 85 ꢀC, considerably higher than both carbon-containing
C
V
and/or di(cyanate ester) analogs. 2013 Wiley Periodicals, Inc.
J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 767–779
KEYWORDS: curing of polymers; cyanate ester; differential scan-
ning calorimetry (DSC); high performance polymers; polycya-
nurate; silane; thermosets
blends and modifiers,22–24 silica nanoparticles,25–34 clay nano-
platelets,35–47 or other silicon-containing nanostructures48–53
including oligomeric silsesquioxanes.54–66 While the addition of
silicon in fully oxidized form (i.e., as SiO2) can provide an in-
place barrier to oxygen permeation, the use of unoxidized orga-
nosilicon moieties or partly oxidized silicates allows for the in
situ generation of a protective barrier during high-temperature
oxidation or exposure to energetic forms of oxygen (such as
atomic oxygen in low Earth orbit).66–68 Such in situ generation
from the bulk has the advantage of spontaneous regeneration in
the event that a protective layer is breached or damaged during
further exposure. Furthermore, the incorporation of silicon at
the molecular level by the use of organosilicon small-molecule
monomers (either as single component substitutes or as cocura-
tives) can preserve desirable processing characteristics such as
a low viscosity in the uncured state.
INTRODUCTION Polycyanurate networks formed from the
cyclotrimerization of cyanate ester monomers have proven to
be an especially useful class of thermosetting material, with
numerous high-performance applications in the microelec-
tronics and aerospace industries.1–4 Some of the more interest-
ing example applications include magnet casing for nuclear
fusion reactors5,6 and the Large Hadron Collider,7 brush seals
for turbine engines,8,9 and spacecraft structures such as frame-
works for telescope mirrors,10 and solar panel supports.11 In
many applications, protection against oxidation during short-
term exposure to temperatures at or above 500 ꢀC is highly
desirable. Unfortunately, the onset of thermochemical degrada-
tion in the triazine crosslinks of cyanurate networks occurs
12
ꢀ
well below 500 C, thus thermooxidative protection must be
provided by the incorporation of more stable materials.
A strategy used to provide thermooxidative protection in cya-
nurate polymer networks that has been explored through
much recent investigation is the incorporation of silicon or
siloxy moieties in the form of monomers and cocuratives,13–21
To date, there have been several reports relating to organosi-
licon monomers for epoxy,13,14,69–73 benzoxazine,74 and bis-
maleimide18,75–77 thermosetting networks. In cyanate esters,
Additional Supporting Information may be found in the online version of this article.
C
V
2013 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
767

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
a silicon-containing analog (named SiMCy) to the well-
known dicyanate ester of bisphenol A (BADCy) was first syn-
thesized by Wright et al. in 2003.20 Compared to BADCy,
SiMCy exhibited improved thermooxidative resistance at high
temperatures along with lower water uptake, and improved
processing and cure characteristics.19 Two cyanate ester
organosilane monomers have also been described recently
for use in the formation of microporous structures.15
chloroform, hexane, and toluene were dried by passage
through columns of activated alumina under a nitrogen
atmosphere and then degassed before use. Anhydrous
dichloromethane and acetone were purchased from Aldrich
and used as received. Methyltrichlorosilane and tetrachlorosi-
lane were purchased from Gelest and distilled before use.
Triethylamine was purchased from Aldrich Chemical and dis-
tilled before use. 4-Benzyloxybromobenzene was obtained
from Aldrich and recrystallized from ether before use. Cyano-
gen bromide, n-butyllithium, nonylphenol, and 10% palladium
on carbon (wet, Degussa type) were obtained from Aldrich
Chemical and used as received. Copper(II) acetylacetonate
was purchased from ROC/RIC and used as received. All
manipulations of compounds and solvents for the preparation
of compounds (1)–(6) were carried out using standard
Schlenk line techniques. Hydrogenation was done using a Parr
Though the incorporation of silicon at the molecular level in
thermosetting polymers offers improved thermooxidative
resistance, it can have negative effects on other aspects of
performance. Because the carbon–silicon bond is in general
longer and more flexible than a corresponding carbon–car-
bon bond, the molecular segments of networks containing
such bonds are more flexible, resulting in a lower glass tran-
sition temperature at full cure, when silicon is incorporated
into the network.17,19 Approaches to the molecular level
incorporation of silicon that enable the networks to retain
high glass transition temperatures at full cure are therefore
of considerable interest. To date, however, there has been
very little work devoted to demonstrating means of incorpo-
rating molecular-level silicon in a manner that facilitates the
retention of high glass transition temperatures at full cure.
R
Hydrogenator equipped with pressure safe vessels and VitonV
seals. Selected cyanate ester samples were purified using a
W-Prep2XY Yamazen chromatographic column.
Preparation of Methyl[tris(4-benzyloxyphenyl)]silane (1)
A
chilled (278 ꢀC) THF (400 mL) solution of 4-
bromophenyl benzyl ether (20.00 g, 76.0 mmol was treated
with 2.5M n-BuLi (30.4 mL, 76 mmol) and allowed to react
with stirring for 2 h at 278 C. This mixture, now heteroge-
ꢀ
In this article, we report the synthesis, cure kinetics, and net-
work properties of trifunctional and tetrafunctional silicon-
containing cyanate esters in which the silicon serves as a net-
work junction point after cure. It has been shown that the
incorporation of network junctions, whether they arise from
crosslinks or from branch points in tri- (or higher) functional
monomers with functional end groups, provide for an
increased glass transition temperature at full cure in thermo-
setting composite resins.78 By incorporating silicon into tri-
functional and tetrafunctional monomers in such a way as to
create a network junction coincident with each incorporated
silicon atom, we show that networks are obtained that exhibit
glass transition temperatures higher than those of the corre-
sponding networks formed from difunctional cyanate esters
without silicon. We further show that these advantages are
retained when the silicon-containing monomers are cocured
neous, was treated with slow addition of methyltrichlorosi-
lane (3.787 g, 25.33 mmol, diluted with THF) and the
cooling bath removed. The mixture was allowed to react
with stirring overnight and the solvents were removed
under reduced pressure. Chloroform (300 mL) was added,
and the mixture was stirred for additional hour. This was
filtered to remove LiCl salt, and the solvent from the filtrate
was removed under reduced pressure on a rotary evapora-
tor. The off-white crude product was precipitated into
methanol (400 mL). This was stirred overnight, filtered, and
dried under nitrogen to afford 1 as a white solid (13 g,
87% yield).
1H NMR (acetone-d6) d: 7.48–7.32 (m, 21H), 7.03 (d, J 5 8
Hz, 6H), 5.12 (s, 6H), 0.75 (s, 3H). 13C NMR (acetone-d6) d:
160.94 (C4), 138.32 (C6), 137.47 (C2), 136.86 (C1), 129.35
(C8), 128.70 (C9), 128.50 (C7), 115.38 (C3), 70.23 (C5),
22.66 (C10, SiCH3). 29Si NMR (acetone-d6) d: 212.32 (s).
R
with di(cyanate ester)s such as PrimasetV 1,1-bis(4-cyanato-
phenyl)ethane (LECy) in order to achieve complete conver-
sion to cyanurates. These networks thus offer both high glass
transition temperatures as well as increased thermooxidative
stability while retaining good processing characteristics.
EXPERIMENTAL
General Considerations
R
R
PrimasetV BADCy and PrimasetV LECy were purchased from
Lonza and used as received. SiMCy, that is, bis(4-cyanatophe-
nyl)dimethylsilane, was synthesized according to a previously
published procedure.17 ESR-255, that is, tris(4-cyanatophe-
nyl)ethane, was synthesized according to a previously pub-
lished procedure79 by Dr Matthew Davis at the Naval Air
Warfare Center, Weapons Division, China Lake, CA, and used
as received. Tetrahydrofuran (THF), diethyl ether (ether),
768
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Combustion analysis calculated (Found): % C, 81.04 (80.77);
% H, 6.12 (6.13).
Preparation of Methyl[tris(4-hydroxyphenyl)]silane (2)
A THF (200 mL) solution containing 1 (10.00 g, 16.87
mmol) and 10 wt % palladium on carbon (400 mg), was
placed in a 1000-mL pressure-safe vessel equipped with
R
VitonV seals and connected to a hydrogenator. This was pres-
surized with hydrogen (35 psi) and allowed to react with
stirring for 48 h. The catalyst was removed by filtration
through Celite, and the solvent was removed under reduced
pressure to afford 4.60 g (85% yield) of 2, as an off-white
solid. For purification, compound 2 was dissolved in THF
and precipitated out in hexane. The white product was fil-
tered, washed with ether, and dried under dynamic vacuum.
Preparation of Tetrakis(4-benzyloxyphenyl)silane (4)
A chilled (278 ꢀC) THF (400 mL) solution of 4-bromophenyl ben-
zyl ether (20.00 g, 76.0 mmol was treated with 2.5M
n-BuLi (30.4 mL, 76 mmol) and allowed to react with stirring for
1H NMR (acetone-d6) d: 8.47 (s, 3H), 7.34 (d, J 5 8 Hz, 6H),
6.87 (d, J 5 8 Hz, 6H), 0.70 (s, 3H). 13C NMR (acetone-d6) d:
159.43 (C4), 137.52 (C2), 127.37 (C1), 115.92 (C3), 22.51 (C5,
SiCH3). 29Si NMR (acetone-d6) d: 212.57 (s). Combustion anal-
ysis calculated (Found): % C, 70.78 (70.35); % H, 5.63 (5.69).
2 h at 278 C. This mixture, now heterogeneous, was treated
ꢀ
with a slow addition of tetrachlorosilane (3.22 g, 19 mmol,
diluted with THF), and the cooling bath removed. The mixture
ꢀ
was allowed to reflux at 55 C with stirring for two nights. The
solvents were removed under reduced pressure. Chloroform
(300 mL) was added, and the mixture was stirred for additional
hour. This was filtered to remove LiCl salt, and the solvent from
the filtrate was removed under reduced pressure on a rotary
evaporator. The off-white crude product was precipitated into
methanol (400 mL). This was stirred overnight, filtered, and dried
under nitrogen to afford 4 as a white solid (11 g, 76% yield).
1H NMR (CDCl3) d: 7.54–7.38 (m, 28H), 7.05 (d, J 5 9 Hz,
8H), 5.12 (s, 8H). 13C NMR (CDCl3) d: 160.54 (C4), 137.84
(C2), 136.98 (C6), 128.66 (C8), 128.06 (C9), 127.59 (C7),
126.42 (C1), 114.46 (C3), 69.80 (C5). 29Si NMR (CDCl3) d:
214.60 (s). Combustion analysis Calculated (Found): % C,
82.07 (81.49); % H, 5.83 (5.90).
Preparation ofꢀMethyl[tris(4-cyanatophenyl)]silane (3)
A chilled (220 C) ether (200 mL) solution containing meth-
yl[tris(4-hydroxyphenyl)]silane, 2, (4.0 g, 12.41 mmol) and
cyanogen bromide (4.88 g, 46.53 mmol) was treated with
triethylamine (4.71 g, 46.53 mmol) in a drop-wise manner.
This mixture was allowed to react for 2 h with stirring at
220 ꢀC. Diethyl ether (500 mL) was added to the reaction
mixture and stirred overnight. The mixture was filtered to
remove the hydrobromide salt, and the organic layer was
washed with (2 3 100 mL) DI water, followed by a brine
wash and then dried over MgSO4. The solvents were
removed under reduced pressure, and crude product (3.2 g,
65% yield) was precipitated out from ether to afford 2.9 g
(59% yield) of 3 as white crystalline solid (mp 118 ꢀC).
Some batches of cyanate ester 3 were further purified by
flash chromatography using silica–gel (40 lm, 60 Å) univer-
sal column size M (20 ID 3 80-mm packed length) with a
dichloromethane mobile phase.
Preparation of Tetrakis(4-hydroxyphenyl)silane (5)
A THF (200 mL) solution containing 4 (10.00 g, 13.14 mmol)
and 10 wt % palladium on carbon (400 mg) was placed in a
1H NMR (acetone-d6) d: 7.74 (d, J 5 9 Hz, 6H), 7.48 (d, J 5 9
Hz, 6H), 0.98 (s, 3H). 13C NMR (acetone-d6) d: 154.48 (C4),
137.37 (C2), 134.11 (C1), 115.13 (C3), 108.20 (C6, OCN),
24.43 (C5, SiCH3). 29Si NMR (acetone-d6) d: 210.51 (s).
Combustion analysis Calculated (Found): % C, 66.48 (66.01);
% H, 3.80 (3.68); % N, 10.57 (10.68).
R
1000-mL pressure-safe vessel equipped with VitonV seals and
connected to a hydrogenator. This was placed under an atmos-
phere of hydrogen (35 psi) and allowed to react with stirring
for 48 h. The catalyst was removed by filtration through Celite,
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
769

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
and the solvent was removed under reduced pressure to afford
4.80 g (91% yield) of 5 as an off-white solid. For purification,
compound 5 was dissolved in THF and precipitated out in hex-
ane. The white product was filtered, washed with ether, and
dried under dynamic vacuum.
1H NMR (acetone-d6) d: 7.77 (d, J 5 8, 8H), 7.54 (d, J 5 8, 8H).
13C NMR (acetone-d6) d: 155.73 (C4), 139.72 (C2), 132.38
(C1), 116.40 (C3), 109.02 (C5, OCN). 29Si NMR (acetone-d6) d:
214.95 (s). Combustion analysis Calculated (Found): % C,
67.19 (67.21); % H, 3.22 (3.27); % N, 11.19 (10.84).
1H NMR (acetone-d6) d: 8.63 (s, 4H), 7.38 (d, J 5 6 Hz, 8H),
6.81 (d, J 5 6 Hz, 8H). 13C NMR (acetone-d6) d: 159.54 (C4),
138.59 (C2), 125.98 (C1), 115.92 (C3). 29Si NMR (acetone-
d6) d: 214.64 (s). Combustion analysis calculated (Found):
% C, 71.97 (68.89); % H, 5.03 (5.55).
Preparation of Thermally Cured Networks
Cured networks were prepared from the newly synthesized
cyanate esters (3) and (6), blends of (3) and (6) with the
R
commercial dicyanate ester PrimasetV LECy (1,1-bis(4-cyana-
tophenol)ethane), and three different “control” materials: (1)
R
PrimasetV BADCy (2,2-bis(4-cyanatophenyl)propane), (2)
SiMCy (bis(4-cyanatophenyl)dimethylsilane), and (3) ESR-
25589 (1,1,1-tris(4-cyanatophenyl)ethane). The properties of
SiMCy16,17,19,20 and ESR-25579,80 have been described in
numerous previous publications. To assess the effects of
purity, some batches of (3) and SiMCy were purified by pass-
ing solutions in dichloromethane through a W-Prep2XY Yama-
zen flash chromatograph, followed by rotary evaporation.
To mix monomers for cocuring of blended networks, equal
weights of each component were weighed out into a vial and
then heated to 10–20 ꢀC above the highest melting point of any
of the system components and stirred together. An analogous
procedure was used for single-component networks except that
no stirring was required. Once prepared in a homogeneous,
molten state, networks were de-gassed for 30 min under
reduced pressure (ca. 300 mmHg), then poured into cylindrical
molds measuring 12 mm in diameter 3 about 3 mm deep com-
posed of silicone rubber. Details of the mold-making procedure
have been published elsewhere.81 The samples were then
Preparation ofꢀTetrakis(4-cyanatophenyl)silane (6)
A chilled (220 C) ether (200 mL) and THF (5 mL) solution
containing tetrakis(4-hydroxyphenyl)silane, 5, (4.0 g, 10
mmol) and cyanogen bromide (5.22 g, 50.mmol) was treated
with triethylamine (5.06 g, 50 mmol) in a drop-wise manner.
This mixture was allowed to react for 2 h with stirring at
220 ꢀC. Dichloromethane (200 mL) was added to the reac-
tion mixture and stirred for an hour. The mixture was fil-
tered to remove the hydrobromide salt, and the organic
layer was washed with (2 3 100 mL) DI water, followed by
a brine wash and then dried over MgSO4. The solvents were
removed under reduced pressure, and crude product (3.8 g,
76% yield) was precipitated from ether to afford 3.5 g (70%
ꢀ
heated under dry nitrogen to 150 C for 1 h, followed by 210
ꢀC for 24 h, using ramp rates of 5 K min21. Once cooled, the
samples were de-molded to form homogeneous discs.
Characterization
1H, 13C, and 29Si NMR measurements were performed using a
Bruker AC 300 or Bruker 400 MHz instrument. 1H and 13C NMR
chemical shifts are reported relative to the deuterated solvent
peak (1H, 13C: acetone-d6, d 2.05 ppm, d 29.9 ppm or CDCl3, d
7.28 ppm, d 77.23 ppm). 29Si NMR chemical shifts are reported
relative to external tetramethylsilane at 0 ppm. Elemental analy-
ses were obtained from Atlantic Microlabs or performed on a
Perkin Elmer EA2400 Series II combustion analyzer.
ꢀ
yield) of 6 as white crystalline solid (mp 169 C).
Differential scanning calorimetry (DSC) was carried out on a
TA Instruments Q Series 200 instrument under 50 mL
min21 dry nitrogen purge. Several different programs were
used for analysis. Standard nonisothermal analysis involved
heating samples at 10 K min21 to 350 ꢀC, cooling at 10 K
min21 to 100 ꢀC, and then reheating to 350 ꢀC at 10 K
min21 to establish a baseline. For cure kineꢀtics studies, sam-
ples were first heated at 5 K min21 to 130 C to achieve any
required melting, then heated as quickly as possible (ꢁ100
K min21) to the desired isothermal cure temperature and
held for 30 min isothermally. Following the isothermal cure,
the samples were quenched (ꢁ100 K min21 cooling) to 120
ꢀC and then heated to 350 ꢀC at 10 K min21. The samples
were then cooled at 10 K min21 to 100 ꢀC, and the entire
770
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
procedure was repeated with the cured sample in place to
establish a baseline. The details of this procedure, including the
method of applying baseline corrections, have been published
previously.80 Cure kinetics were determined using the Kamal
model82 with a variant of Kenny’s graphical method;83 complete
details of this procedure have also been published previously.80
Thermogravimetric analysis (TGA) was carried out using a TA
Instruments Q5000 under 50 mL min21 sample gas purge
(dry nitrogen or air), and 10 mL min21 balance gas purge
(with gas type matched to the sample). TGA samples were
comprised of small chunks of cured discs, and were heated at
10 K min21 to 600 C. Oscillatory thermomechanical analysis
ꢀ
(TMA) was performed using a TA Instruments Q400, with a
50 mL min21 dry nitrogen purge. Cured discs were subjected
to a mean compressive force of 0.1 N, oscillating with an
amplitude of 0.1 N and a frequency of 0.05 Hz. Samples were
preloaded with a 0.2-N force to help achieve good contact
with the probe, which consisted of the 5-mm diameter cylin-
drical “standard” quartz probe supplied by TA Instruments.
For “dry” TMA analysis, samples were cycled between 100
and 200 ꢀC twice at 10 K min21 to measure thermal lag
(details of this procedure are published elsewhere),81 then
heated to 350 ꢀC at 10 K min21, cooled to 100 ꢀC at 10 K
SCHEME 1 Synthesis of new cyanate ester derivatives SiCy-3
(3) and SiCy-4 (6).
RESULTS AND DISCUSSION
Synthesis of New Monomers (3) and (6)
The new cyanate ester monomers (3) and (6) were synthe-
sized following the procedures similar to those published for
production of SiMCy17 (See Scheme 1). Lithium-bromo
exchange on p-bromophenyl benzyl ether with butyllithium
produces a nucleophile, which can be used to replace all the
chlorides on tetrachlorosilane or methyltrichlorosilane with
benzyloxyphenyl groups to produce compounds (1) and (4).
The “protecting” benzyl group is easily removed by low-
pressure hydrogenation catalyzed with 10% palladium on
carbon to produce phenols (2) and (5). These phenols react
in high yield with cyanogen bromide in the presence of trie-
thylamine to produce the desired cyanate ester monomers
(3) and (6). All six compounds were isolated in high yield
and fully characterized by 1H, 13C, and 29Si NMR spectros-
copies and combustion analysis. Their spectra are simple
and unremarkable with chemical shifts all in the expected
regions and ratios. The data are presented in the Experimen-
tal section. Additional verification of the structures of (3)
and (6) was provided by X-ray diffraction of single crystals.
The resultant structures are shown in Figures 1 and 2. All
bond angles and bond lengths are in the expected ranges for
such compounds.
min21, and then reheated to 350 C at 10 K min21. For “wet”
ꢀ
TMA analysis, samples wereꢀ first heated to 350 ꢀC, then
cycled between 100 and 200 C to measure thermal lag, then
reheated to 350 ꢀC, all at 20 K min21. The procedure for
“wet” TMA is designed to minimize drying of the sample
before measurement of the glass transition temperature.
Crystal data for compounds (3) and (6) were collected at
T 5 100.0 (K) using Kusing Bruker 3-circle, SMARTAPEX CCD
with c-axis fixed at 54.748, running on SMART V 5.625 pro-
gram (Bruker AXS: Madison, 2001). Graphite monochro-
mated MoKa ( k 5 0.71073 Å) radiation was employed for
data collection and corrected for Lorentz and polarization
effects using SAINT V 6.22 program (Bruker AXS: Madison,
2001), and reflection scaling (SADABS program, Bruker AXS:
Madison, WI, 2001). Structures were solved by direct meth-
ods (SHELXL-97, Bruker AXS: Madison, 2000) and all nonhy-
drogen atoms refined anisotropically using full-matrix least-
squares refinement on F2. Hydrogen atoms were added at
calculated positions when necessary. CDCC 960730 (SiCy-3)
and CDCC 960731 (SiCy-4) contain the supplementary crys-
tallographic data for this article. The data can be obtained
free of charge from the Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Figure 3 provides the structures of all the monomers uti-
lized. ESR-255 is the carbon-containing analog of the silicon-
containing tricyanate (3). SiMCy is the dicyanate analog of
(3) in which one cyanate ester functional arm has been
replaced by a methyl group, and BADCy is the carbon-
containing analog of SiMCy. The carbon-containing analog of
(6) has been reported previously,84 but was difficult to cure
into a network, thus substantial physical property data are
unavailable. Previous work on tricyanate esters has shown
that preparing networks comprising 50 wt % of the dicya-
nate with 50 wt % LECy can result in synergistic improve-
ments in performance.85 Thus, networks were also prepared
from mixtures of equal parts by weight of (3) with LECy and
(6) with LECy, as well as with equal parts of ESR-255 and
LECy for comparison. For all investigations, no catalysts
were added to the monomers before network formation.
Densities were determined by immersing cured discs in solu-
tions of CaC12 and varying the CaCl2 concentration until neu-
tral buoyancy was achieved. The density of the fluid needed
to achieve neutral buoyancy was then determined by weigh-
ing 10.00 mL of the fluid in a volumetric flask. Fourier trans-
form infrared spectroscopy (FTIR) was carried out using a
Thermo Corporation Nicolet 6700 spectrometer in attenu-
ated total reflectance mode with the ZnSe crystal attachment.
A total of 32 scans were completed on these surfaces with a
resolution of 4 cm21 to obtain spectra.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
771

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1 ORTEP representation of SiCy-3 (3) collected at 100
K. Thermal ellipsoids shown at 50%. Hydrogen atoms omitted
for clarity. Black, C; blue, Si, red, O; purple, N. [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
FIGURE 3 Chemical structure of monomers utilized for com-
parative studies.
Cure Characteristics of Monomers (3) and (6)
with cure is indicative of incomplete conversion, as further
indicated by an enthalpy of cyclotrimerization of 94 6 4 kJ
per cyanate molar equivalent, versus an expected value of
around 110 kJ equiv21. In fact, the enthalpies of cyclotrime-
rization for SiCy-3 and ESR-255 are not significantly different
when measured by the same nonisothermal DSC protocol,
indicating that the more flexible C–Si linkages are ineffective
at relieving the steric constraints that prevent complete
cyclotrimerization of these tricyanate monomers. An “L”-
shaped exotherm with incomplete conversion is also a good
indication that, due to vitrification, a sample is in the glassy
state at the end of the DSC scan. For SiCy-3, then, the glass
transition temperature (TG) exceeds 350 ꢀC even for the
The nonisothermal DSC scan, shown in Figure 4, for SiCy-3
(3) after purification by chromatography, is typical for highly
pure cyanate ester monomers, with recrystallization of the
supercooled melt at 50–100 ꢀC followed by melting at 118
ꢀC, a very wide processing window, and cyclotrimerization
onset near 300 ꢀC. Both the melting temperature and
enthalpy are very similar to that of 1,1,1-tris(4-cyanatophe-
nyl)ethane (known as ESR-255), the carbon-containing ana-
log of SiCy-3. (A full set of comparative nonisothermal DSC
data for all monomers studied is provided in Supporting
Information Section S2.) The “L”-shaped exotherm associated
FIGURE 2 ORTEP representation of SiCy-4 (6) collected at 100
K. Thermal ellipsoids shown at 50%. Hydrogen atoms omitted
for clarity. Black, C; blue, Si, red, O; purple, N. [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
FIGURE 4 Nonisothermal DSC of SiCy-3 (3). [Color figure can
be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
772
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
3 (Fig. 4), resulting in a very prominent “L”-shaped exo-
therm. The total enthalpy of cyclotrimerization (at 96 6 10
kJ equiv21) is not significantly different from SiCy-3.
Isothermal DSC analysis of the cyclotrimerization of SiCy-3
produced a pattern of conversion as a function of time and
temperature similar to most cyanate ester monomers, as
seen in Figure 6. The cure kinetics up to 50% conversion
were described well by the Kamal model82 using analysis
procedures that have been described in detail elsewhere.80
(A complete listing of model parameter values and conver-
sion as a function of time and temperature graphs for SiCy-3
and its many analogs, including analysis of batch to batch
reproducibility and effects of purity level, are provided in
Supporting Information Section S3.) Arrhenius analysis of
the cyclotrimerization of SiCy-3 after purification by chroma-
tography (Fig. 7) resulted in an estimated activation energy
of 115 6 1 kJ mol21. Note that this uncertainty estimate
reflects only the precision of the Arrhenius model fit to the
data. When two separate batches of SiCy-3 were analyzed,
the computed activation energy differed by about 7% (see
Supporting Information Fig. S23). Thus, the value of 107 6 1
kJ mol21 for ESR-255 reported previously80 does not reflect
a significant difference. In neither the tricyanates nor their
dicyanate analogs (see Supporting Information Section S3)
were any significant difference in activation energy observed
when considering the effect of substitution of Si for C in
highly pure samples. Note that the absolute conversion rates
observed did vary significantly among monomers and
batches with differing purities (but not among highly puri-
fied batches of the same monomer), a reflection of the fact
that the cyclotrimerization of cyanate esters are primarily
determined by the nature of impurities.
FIGURE 5 Nonisothermal DSC of SiCy-4 (6). [Color figure can
be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
incompletely cured network. Thermochemical degradation of
SiCy-3 networks begins at 350–400 ꢀC according to TGA
data (see Fig. 12). Thus, although these networks cannot be
fully cyclotrimerized, sufficient conversion can be attained to
prevent the significant mechanical softening associated with
the TG before the onset of thermochemical degradation. In
other words, chemical, rather than mechanical, stability con-
siderations will limit the high-temperature performance of
these networks in many situations.
The nonisothermal DSC of SiCy-4 (Fig. 5) indicates that simi-
lar steric constraints on cyclotrimerization exist for the tetra-
cyanate analog of SiCy-3. Furthermore, the high melting
point precludes many low-cost processing operations. Owing
to its low potential for use, further work with the neat nmo-
nomer, including purification by chromatography, was not
attempted, although promising results were obtained when
In the case of dicyanates, the substitution of Si for C in the
bridge linking the aryl cyanate groups in fully cyclotrimer-
ized networks results in a significant decline in the glass
transition temperature (from ca. 320 to 260 ꢀC for BADCy
and SiMCy, respectively, see Supporting Information Section
S4). As a result, the ability to utilize SiMCy for very high
R
cocured with PrimasetV LECy (vide infra). As a result of the
lower purity, the cyclotrimerization exotherm in SiCy-4 (Fig.
5) is shifted to much lower temperatures compared to SiCy-
FIGURE 7 Arrhenius plot showing maximum autocatalytic
reaction rates of SiCy-3 (3, purified by chromatography) and
ESR-255. [Color figure can be viewed in the online issue, which
is available at wileyonlinelibrary.com.]
FIGURE 6 Isothermal cure kinetics of SiCy-3 (3) after purifica-
tion by flash chromatography. [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 2014, 52, 767–779
773

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 9 Nonisothermal DSC of SiCy-4 (6) cocured with an
equal weight of LECy. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
FIGURE 8 Nonisothermal DSC of SiCy-3 (3) cocured with an
equal weight of LECy. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
24 h at 210 ꢀC. Peaks near 2250 cm21 due to residual
unreacted cyanate ester groups are still clearly present in
the neat SiCy-3 network, while in the cocured system the
same peaks are almost completely absent. The conversion to
cyanurate is confirmed by the presence of the peaks near
temperature applications is more limited compared to BADCy.
For the corresponding tricyanates, however, the mechanical
softening points of sufficiently cyclotrimerized networks are
so high that they are not a factor in limiting high temperature
service, and thus substitution of more flexible SiAC bonds for
CAC bonds in the bridges connecting aryl cyanates does not
adversely affect service temperatures. However, the tricyanate
architecture studied comes with the significant drawback of
an inability to achieve complete cyclotrimerization due to
steric hindrance, which typically leads to decreased thermal
and hydrolytic stability in cyanurate networks.
1365 and 1550 cm21
.
Physical Properties of Cured Networks
As mentioned previously, the absence of a discernible glass
transition in DSC scans of SiCy-3 and SiCy-4 (and their mix-
tures with LECy), suggests that a glass transition tempera-
ture in excess of 350 ꢀC is possible for sufficiently
cyclotrimerized networks. Previous studies of cyanate esters
have shown that when such high glass transition tempera-
tures are possible, the actual glass transition temperatures
obtained during cure are primarily controlled by the cure
temperature rather than the glass transition temperature of
the fully cured network.80 This principle is illustrated in Fig-
ure 11, which compares DSC traces of BADCy, SiMCy, ESR-
One way to overcome the aforementioned limitation is to
cocure monomers such as SiCy-3 with other cyanate esters
R
that can fully cyclotrimerize. PrimasetV LECy, a low-viscosity
liquid at room temperature featuring a more flexible ethyli-
dene bridge between two aryl cyanates, has long been recog-
nized as an excellent choice for formulating multicomponent
cyanate ester mixtures,85 though formulation with other liq-
uid dicyanates, such as RTX-366,86,87 is also possible. We
investigated the cocuring of SiCy-3 and its tetracyanate ana-
log SiCy-4 with LECy. Figures 8 and 9 show nonisothermal
DSC scans of SiCy-3 and SiCy-4, respectively, cocured with an
equal weight of LECy. (For these preliminary investigations,
purification by chromatography was not attempted.)
The scans show clearly that the crystallization of the tricya-
nate or tetracyanate component is greatly curtailed, and the
melting points lowered, by mixing with LECy. In addition,
the cyclotrimerization exotherms are more symmetric, partic-
ularly for the SiCy-3/LECy mixture, indicating more complete
cyclotrimerization. The enthalpies of cyclotrimerization were
128 6 13 and 113 6 11 kJ equiv21 for the SiCy-3 and SiCy-4
mixtures, respectively, indicative of complete cyclotrimeriza-
tion. Additional evidence for dramatic improvements in the
ability to fully cyclotrimerize the mixtures is presented in
Figure 10, which shows the FTIR spectra of cured SiCy-3
and cocured SiCy-3/LECy (mixed in equal weights) after
FIGURE 10 FTIR spectra of SiCy-3 (3) and a mixture of equal
parts SiCy-3 and LECy, after cure for 24 h at 210 ꢀC. [Color fig-
ure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
774
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
networks formed by cocuring with LECy, both full conversion
and a very high glass transition temperature are possible,
making these networks good candidates for dry high-
temperature applications. Analysis of moisture uptake, how-
ever, revealed a significant weakness. The water uptake of
the SiCy-3 network was 5.5%, while that of the SiCy-3/LECy
and SiCy-4/LECy networks were 4.7 and 4.4%, respectively,
ꢀ
after 96 h immersion at 85 C of networks cured for 24 h at
210 ꢀC). These values are significantly higher than either
ESR-255 (3.5%) or ESR-255 cocured with an equal weight of
LECy (2.7%). In order to determine the extent to which sam-
ple purity played a role in these results, the same water
uptake test was performed on a sample of SiCy-3 further
purified by chromatography. The resultant water uptake of
5.0% represents only a modest improvement, and is still
much higher than the corresponding value for ESR-255. This
result contrasts with earlier results for dicyanates19 (con-
firmed by us in catalyzed networks)17 that show significantly
lower water uptake for SiMCy compared to BADCy under
identical conditions.
FIGURE 11 DSC scans after cure for 12 h at 210 ꢀC of SiCy-3
(3) compared to its carbon-containing analog ESR-255, and the
respective silicon- and carbon-containing dicyanate analogs
SiMCy and BADCy. The onset of residual cure at around 275–
280 ꢀC provides an indication of the TG after the 12 h cure for
the tricyanates, while for the dicyanates, a step change in heat
capacity (combined with an endothermic event for BADCy)
indicates the TG. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
In the dicyanate case, the lower glass transition temperature
of SiMCy compared to BADCy is believed to be the key factor
that drives the difference in water uptake. The lower glass
transition temperature of SiMCy allows for greater relaxation
of the free volume formed by cyclotrimerization during cure
in the vitreous state, and thereby limits the capacity for water
255, and SiCy-3 upon quenching and reheating after 12 h at
210 ꢀC in the DSC. For the dicyanates, the step change in
heat capacity associated with the glass transition is clearly
visible, and occurs between the cure temperature of 210 ꢀC
and the glass transition temperatuꢀre at full cure, which is
ꢀ
uptake. In fact, when cured at 250 C (very close to the glass
transition temperature for fully cured SiMCy but not for fully
cured BADCy), networks of uncatalyzed SiMCy showed half
the water uptake compared to networks of BADCy.19 When
catalyzed networks were cured at 210 ꢀC (well below the
glass transition temperature for both networks), the water
uptake of SiMCy was only about 15% lower than that of
BADCy.17 Because the water uptake values of fully cured cya-
nurate networks are typically quite similar whether catalyzed
or not, the relaxation of free volume near the glass transition
temperature appears to be the key factor. As mentioned ear-
lier, the glass transition temperatures achieved by cure at 210
ꢀC were slightly higher for SiCy-3 than for ESR-255, but both
were much higher than the cure temperature. ESR-255, how-
ever, spent less time in the vitreous state, and may have
ꢀ
roughly 260 C for SiMCy and 320 C for BADCy. The differ-
ence in the “as cured” glass transition temperatures, which
were 213 ꢀC for SiMCy and 247 ꢀC for BADCy, principally
reflect the difference in the glass transition temperatures of
the fully cured networks. In contrast, for both ESR-255 and
SiCy-3, the “as cured” glass transition temperatures are
between 275 and 280 ꢀC, and are indicated by the onset of
significant residual cure (as a result of devitrification) with
the step change in heat capacity masked by the cure exo-
therm. The SiCy-3 appears to have a slightly higher devitrifi-
cation temperature, which may simply reflect overall faster
cure kinetics during the isothermal stage compared to ESR-
255. Supporting Information Figures S26 and S27 illustrate
the approximate cure kinetics of ESR-255 and SiCy-3 during
undergone
a
lesser extent of cyclotrimerization after
ꢀ
isothermal cure at 210 C for 12 h, as gleaned from isother-
mal DSC data. These data show that SiCy-3 reaches the vitri-
fication point in about 2 h, allowing for 10 h of cure in the
glassy state, whereas ESR-255 vitrifies after about 4.5 h,
allowing only 7.5 h of cure in the glassy state. We speculate
that the longer time available for SiCy-3 to cure in the glassy
state may enable SiCy-3 to undergo more cyclotrimerization
in the glassy state, which in turn may enable the glass tran-
sition in SiCy-3 to climb to a slightly higher temperature.
TABLE 1 Key Thermomechanical Properties of Networks Contain-
ing SiCy-3 (3) and SiCy-4 (6) Cured for 24 h at 210 ꢀC
“Fully
Water
Density
Cured” TG
(by TMA, ꢀC) (wt %) (by TMA, ꢀC)
Uptake
“Wet” TG
Networka
(g cc21
)
SiCy-3
1.245
1.226
>360
>360
5.5
4.4
202
226
SiCy-3/LECy
(50/50 wt %)
SiCy-4/LECy
(50/50 wt %)
Analysis of the SiCy-3, SiCy-3/LECy, and SiCy-4/LECy net-
works was also undertaken by TMA (see Table 1 and Sup-
porting Information Section S5). The TMA scans confirmed
that all three networks exhibited glass transition tempera-
1.258
>360
4.7
247
a
Density, water uptake, and “wet” TG are for networks after 24 h of
ꢀ
ꢀ
tures of at least 360 C after heating to 350 C. Thus, for the
cure at 210 ꢀC.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
775

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 14 TGA traces of cured SiCy-4 (6)/LECy (mixed in
equal parts by weight before cure) in nitrogen and air. [Color
figure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
FIGURE 12 TGA traces of cured SiCy-3 (3) in nitrogen and air.
[Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com.]
vitrification. These factors suggest the water uptake for the
SiCy-3 networks should be higher than for ESR-255. Although
the development of free volume during cure in the vitreous
state may play an important role in determining the differen-
ces in water uptake88 between networks containing SiCy-3
and ESR-255, a more detailed investigation would be needed
to establish the cause with certainty.
exploration of the relationship between network functional-
ity and moisture resistance in cyanurate networks could pro-
vide valuable insights that would enhance performance.
A final consideration is the thermochemical stability of the
networks. In the dicyanate case, incorporation of Si into the
network chains appeared to provide significant protection
from oxidation at high temperatures, raising char yields in
air to around 50%. As seen in Figures 12–14, TGA traces in
nitrogen and air for SiCy-3, SiCy-3/LECy (equal parts by
weight), and SiCy-4/LECy (equal parts by weight) networks,
respectively, also show char yields in excess of 50% in both
nitrogen and air, with the SiCy-4/LECy network providing
the best char yield in nitrogen (70%) and similar char yields
between 50 and 56% in air for all three networks. A full tab-
ulation of all comparative data (along with graphical data for
the ESR-255/LECy networks) is provided in Supporting
Information Section S6. Interestingly, ESR-255 networks
exhibited higher degradation onset temperatures and higher
char yields compared to SiCy-3 networks, while the SiCy-3/
LECy networks showed lower degradation onset tempera-
tures but higher char yields than the ESR-255/LECy net-
works. It is important to note, however, that many factors,
such as interactions among network components in multi-
component networks,85 as well as the extent of cure at the
onset of degradation, can affect the thermal decomposition
process, so caution is needed when attempting to attribute
changes in thermal degradation to changes in the chemical
structure of networks.
As expected the higher water uptake for SiCy-3 and SiCy-3/
LECy networks leads to lower “wet” glass transition temper-
atures (202 and 226 ꢀC, respꢀectively) compared to their
ESR-255 analogs (224 and 242 C, respectively). The SiCy-4/
LECy network, however, equals the performance of the ESR-
255/LECy network, with a “wet” glass transition temperature
(TG) of 247 ꢀC. The higher water uptake for the silicon-
containing networks is expected to lead to faster hydrolytic
degradation of the cyanurate networks. The unexpectedly
high value of the “wet” TG for the SiCy-4/LECy network,
however, is presently unexplained. It suggests that further
Nonetheless, taken together, these results indicate that in
cases where there is a large difference in the thermochemi-
cal degradation of a network in air as compared to nitrogen,
the presence of Si appears to provide a consistent improve-
ment in char yield in air. The magnitude of the improvement
at 600 ꢀC is much larger than can be accounted for simply
by the formation of nonvolatile SiO2 in place of volatile CO2
with all else being equal whenever Si is substituted for C in
the network structure, suggesting a true protective effect.
FIGURE 13 TGA traces of cured SiCy-3 (3)/LECy (mixed in
equal parts by weight before cure) in nitrogen and air. [Color
figure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
776
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
However, when thermal degradation in air and under nitro-
gen are very similar, as is the case for many cyanurate net-
works with high char yields, then the presence of Si does
not improve char yield significantly. Thus, the protective
effect appears to be important only if oxidation provides a
significant contribution to thermal degradation. Another
important point to note is that, in comparison to cyanurate
networks in which Si offers a significant protective effect,
even higher char yields in air are readily obtained in cyanu-
rate networks with no Si simply by avoiding functional
groups (such as isopropylidene bridges between phenyl
rings) that promote degradation.80
is gratefully acknowledged. The ESR-255 was prepared by
Matthew Davis at the Naval Air Warfare Center, Weapons Divi-
sion (China Lake, CA). C. M. Sahagun performed this work as
part of the National Research Council’s Research Associateship
Program.
REFERENCES AND NOTES
1 Chemistry and Technology of Cyanate Ester Resins; I. Hamer-
ton, Ed.; Chapman & Hall: London, 1994.
2 C. P. R. Nair, D. Mathew, K. N. Ninan, In New Polymerization
Techniques and Synthetic Methodologies; A. A. Abe,-C.
Albertsson, H. J. Cantow, Eds.; Advances in Polymer Science
155; Springer: New York, 2001; Chapter 1, pp 1–99.
CONCLUSIONS
3 T. Fang, D. A. Shimp, Prog. Polym. Sci. 1995, 20, 61–118.
The synthesis of two new cyanate ester monomers, methyl[t-
ris(4-cyanatophenyl)]silane and tetrakis(4-cyanatophenyl)si-
lane, was achieved in overall yields of around 50% with
good purity using a straightforward extension of a three-step
method originally developed for the synthesis of bis(4-cyana-
tophenyl)dimethylsilane from dimethyldichlorosilane. The
primary advantage of the new trifunctional and tetrafunc-
tional cyanate ester monomers was the ability to achieve
much higher glass transition temperatures (at least 350 ꢀC)
at the maximum practical extent of cure. As single-
component networks, however, both monomers could not be
completely converted to cyanurates due to steric hindrance
of the uncured cyanate ester groups. Cocuring with an equal
4 I. Hamerton, J. N. Hay, High Perform. Polym. 1998, 10, 163–
174.
5 P. Fabian, M. Haynes, H. Babcock, M. Hooker, IEEE Trans.
Appl. Supercond. 2013, 23, No. 7700204.
6 N. A. Munshi, J. K. Walsh, M. W. Hooker, H. K. Babcock, A.
H. Haight, S. R. Durso, A. Kawaguchi, P. Hough, IEEE Trans.
Appl. Supercond. 2013, 23, No. 7700104.
7 P. Abramian, F. de Aragon, J. Calero, J. de la Gama, L.
Garcia-Tabares, J. L. Gutierrez, M. Karppinen, T. Martinez, E.
Rodriguez, I. Rodriguez, L. Sanchez, F. Toral, C. Vazquez, IEEE
Trans. Appl. Supercond. 2013, 23, No. 4101204.
8 H. C. Chen, K. Shivakumar, CMC-Comput. Mat. Contin. 2008,
8, 33–42.
R
9 K. N. Shivakumar, H. Chen, G. Holloway, J. Reinf. Plast. Com-
pos. 2009, 28, 675–689.
weight of the dicyanate monomer PrimasetV LECy, however,
enabled practically complete conversion while maintaining a
glass transition temperature in the fully cured state of over
350 ꢀC. The higher melting points of the monomers, how-
ever, did limit the available processing window, to a modest
extent for the neat trifunctional monomer (melting point
10 P. C. Chen, C. W. Bowers, D. A. Content, M. Marzouk, R. C.
Romeo, Opt. Eng. 2000, 39, 2320–2329.
11 P. D. Wienhold, D. F. Persons, SAMPE J. 2003, 39, 6–17.
12 R. E. Lyon, R. N. Walters, S. Gandhi, Fire Mater. 2006, 30,
89–106.
ꢀ
118 C), and severely for the tetrafunctional monomer (melt-
13 H. X. Yan, M. M. Zhang, C. Liu, J. P. Zhang, Polym. Bull.
2013, 70, 2923–2933.
ing point 169 ꢀC). Cocured systems, however, retained the
favorable processing characteristics of di(cyanate ester)
monomers. As with networks formed from di(cyanate ester)
monomers, a significant improvement in thermooxidative
stability compared to carbon-containing analogs was
achieved for silicon-containing networks, but only in cases
where oxidation contributed significantly to the thermal deg-
radation. The water uptake for the silicon-containing net-
works was, in all cases, significantly higher compared to
carbon-containing analogs, perhaps due to differences in the
evolution of free volume during cure. The higher water
uptake leads to a lower “wet” TG in the silicon-containing
networks when compared to their carbon-containing analogs.
14 Z. Y. Zhang, A. J. Gu, G. Z. Liang, L. Yuan, D. X. Zhuo, Soft
Matter 2013, 11, 346–352.
15 H. Yu, C. J. Shen, M. Z. Tian, J. Qu, Z. G. Wang, Macromo-
lecules 2012, 45, 5140–5150.
16 J. T. Reams, A. J. Guenthner, K. R. Lamison, V. Vij, L. M.
Lubin, J. M. Mabry, ACS Appl. Mater. Interfaces 2012, 4, 527–
535.
17 A. J. Guenthner, K. R. Lamison, V. Vij, J. T. Reams, G. R.
Yandek, J. M. Mabry, Macromolecules 2012, 45, 211–220.
18 D. X. Zhuo, A. J. Gu, G. Z. Liang, J. T. Hu, L. Cao, L. Yuan,
Polym. Degrad. Stab. 2011, 96, 505–514.
19 A. J. Guenthner, G. R. Yandek, M. E. Wright, B. J. Petteys,
R. Quintana, D. Connor, R. D. Gilardi, D. Marchant, Macromole-
cules 2006, 39, 6046–6053.
ꢀ
Nonetheless, the “wet” TG, at 247 C, of the best performing
networks, produced by cocuring the tetrakis(cyanate ester)-
R
silane with PrimasetV LECy, represents outstanding “wet”
20 M. E. Wright, Polym. Prepr. (Am. Chem. Soc. Div. Polym.
Chem.) 2004, 45 (2), 294.
performance in an easily processed thermosetting resin.
21 E. M. Maya, A. W. Snow, L. J. Buckley, Macromolecules
2002, 35, 460–466.
22 R. J. Zaldivar, J. Salfity, G. Steckel, B. Morgan, D. Patel, J.
P. Nokes, H. I. Kim, J. Compos. Mater. 2012, 46, 1925–1936.
ACKNOWLEDGMENTS
Support from the Air Force Office of Scientific Research and the
Air Force research Laboratory, Aerospace Systems Directorate
23 S. K. Dai, A. J. Gu, G. Z. Liang, L. Yuan, Polym. Adv. Tech-
nol. 2011, 22, 262–269.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
777

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
24 S. K. Pollack, Z. Fu, Polym. Prepr. (Am. Chem. Soc. Div.
Polym. Chem.) 1998, 39, 452–453.
52 Y. Z. Pan, Y. Xu, L. An, H. B. Lu, Y. L. Yang, W. Chen, S.
Nutt, Macromolecules 2008, 41, 9245–9258.
53 R. H. Lin, C. W. Lin, A. C. Lee, Y. H. Chen, F. S. Yen, J. Appl.
Polym. Sci. 2007, 103, 1356–1366.
25 S. Devaraju, M. R. Vengatesan, M. Selvi, J. K. Song, M.
Alagar, Microporous Mesoporous Mater. 2013, 179, 157–164.
54 Z. P. Zhang, G. Z. Liang, X. L. Wang, S. Adhikari, J. Z. Pei,
High Perform. Polym. 2013, 25, 427–435.
26 G. L. Wu, K. C. Kou, M. Chao, L. H. Zhuo, J. Q. Zhang,
J. Wuhan Univ. Technol.-Mat. Sci. Edit. 2013, 28, 261–264.
55 S. Rakesh, C. P. S. Dharan, M. Selladurai, V. Sudha, P. R.
Sundararajan, M. Sarojadevi, High Perform. Polym. 2013, 25,
87–96.
27 W. X. Sun, W. Z. Sun, M. R. Kessler, N. Bowler, K. W.
Dennis, R. W. McCallum, Q. Li, X. L. Tan, ACS Appl. Mater.
Interfaces 2013, 5, 1636–1642.
56 S. Jothibasu, S. Devaraju, M. R. Venkatesan, A. Chandramohan,
A. A. Kumar, M. Alagar, High Perform. Polym. 2012, 24, 379–388.
28 E. A. Taha, J. T. Wu, K. Gao, L. Guo, Chin. J. Polym. Sci.
2012, 30, 530–536.
57 A. Chandramohan, K. Dinkaran, A. A. Kumar, M. Alagar,
High Perform. Polym. 2012, 24, 405–417.
29 J. T. Hu, A. J. Gu, G. Z. Liang, D. X. Zhuo, L. Yuan, Polym.
Adv. Technol. 2012, 23, 454–462.
58 J. T. Hu, A. J. Gu, Z. J. Jiang, G. Z. Liang, D. X. Zhuo, L.
Yuan, B. J. Zhang, X. X. Chen, Polym. Adv. Technol. 2012, 23,
1219–1228.
30 X. Sheng, M. Akinc, M. R. Kessler, Polym. Eng. Sci. 2010,
50, 1075–1084.
31 Q. L. Zhang, X. Y. Ma, G. Z. Liang, X. H. Qu, Y. Huang, S. H.
Wang, K. C. Kou, J. Polym. Sci. Part B: Polym. Phys. 2008, 46,
1243–1251.
59 Z. P. Zhang, J. Z. Pei, G. Z. Liang, L. Yuan, J. Appl. Polym.
Sci. 2011, 121, 1004–1012.
60 Y. Lin, J. Jin, M. Song, S. J. Shaw, C. A. Stone, Polymer
2011, 52, 1716–1724.
32 W. K. Goertzen, X. Sheng, M. Akinc, M. R. Kessler, Polym.
Eng. Sci. 2008, 48, 875–883.
61 K. Liang, H. Toghiani, C. U. Pittman, J. Inorg. Organomet.
Polym. Mater. 2011, 21, 128–142.
33 W. K. Goertzen, M. R. Kessler, Compos. Part A 2008, 39,
761–768.
62 M. E. Wright, B. J. Petteys, A. J. Guenthner, G. R. Yandek, L.
C. Baldwin, C. Jones, M. J. Roberts, Macromolecules 2007, 40,
3891–3894.
34 T. J. Wooster, S. Abrol, D. R. MacFarlane, Macromol. Mater.
Eng. 2005, 290, 961–969.
35 Y. Lin, M. Song, C. A. Stone, S. J. Shaw, Thermochim. Acta
2013, 552, 77–86.
63 K. W. Liang, G. Z. Li, H. Toghiani, J. H. Koo, C. U. Pittman,
Chem. Mater. 2006, 18, 301–312.
36 L. Yuan, A. J. Gu, G. Z. Liang, X. Y. Ma, C. Lin, F. Chen,
Polym. Eng. Sci. 2012, 52, 2443–2453.
64 T. L. Lu, G. Z. Liang, Z. Guo, J. Appl. Polym. Sci. 2006, 101,
3652–3658.
37 B. John, C. P. R. Nair, K. N. Ninan, Mater. Sci. Eng. A 2010,
527, 5435–5443.
65 M. E. Wright, B. J. Petteys, A. J. Guenthner, S. Fallis, G. R.
Yandek, S. J. Tomczak, T. K. Minton, A. Brunsvold, Macromole-
cules 2006, 39, 4710–4718.
38 D. A. Kissounko, J. M. Deitzel, S. P. Doherty, A. Shah, J. W.
Gillespie, Eur. Polym. J. 2008, 44, 2807–2819.
66 S. W. Duo, M. M. Song, T. Z. Liu, C. Y. Hu, M. S. Li, J. Nano-
sci. Nanotechnol. 2013, 13, 1356–1359.
39 S. Nagendiran, C. K. Chozhan, M. Alagar, I. Hamerton, High
Perform. Polym. 2008, 20, 323–347.
67 T. K. Minton, M. E. Wright, S. J. Tomczak, S. A. Marquez, L.
H. Shen, A. L. Brunsvold, R. Cooper, J. M. Zhang, V. Vij, A. J.
Guenthner, B. J. Petteys, ACS Appl. Mater. Interfaces 2012, 4,
492–502.
40 G. I. Anthoulis, E. Kontou, A. Fainleib, I. Bei, Y. Gomza,
J. Polym. Sci. Part B: Polym. Phys. 2008, 46, 1036–1049.
41 V. A. Bershtein, A. M. Fainleib, P. Pissis, I. M. Bei, F.
Dalmas, L. M. Egorova, Y. P. Gomza, S. Kripotou, P. Maroulos,
P. N. Yakushev, J. Macromol. Sci. Part B: Phys. 2008, 47, 555–
575.
68 X. W. Zhang, H. Y. Ren, J. H. Wang, Y. Zhang, Y. Y. Shao,
Mater. Lett. 2011, 65, 821–824.
69 X. E. Cheng, W. F. Shi, J. Therm. Anal. Calorim. 2011, 103,
303–310.
42 Z. P. Fang, H. H. Shi, A. J. Gu, Y. Feng, J. Mater. Sci. 2007,
42, 4603–4608.
70 G. H. Hsiue, W. J. Wang, F. C. Chang, J. Appl. Polym. Sci.
1999, 73, 1231–1238.
43 D. H. Yu, B. Wang, Y. Feng, Z. P. Fang, J. Appl. Polym. Sci.
2006, 102, 1509–1515.
71 S. J. Park, F. L. Jin, J. R. Lee, Macromol. Res. 2005, 13, 8–
44 I. Mondragon, L. Solar, A. Nohales, C. I. Vallo, C. M. Gomez,
Polymer 2006, 47, 3401–3409.
13.
72 V. Thulasiraman, M. Sarojadevi, High Perform. Polym. 2009,
21, 437–454.
45 T. J. Wooster, S. Abrol, D. R. MacFarlane, Polymer 2005, 46,
8011–8017.
73 W. J. Wang, L. H. Perng, G. H. Hsiue, F. C. Chang, Polymer
2000, 41, 6113–6122.
46 T. J. Wooster, S. Abrol, D. R. MacFarlane, Polymer 2004, 45,
7845–7852.
74 Y. L. Liu, C. W. Hsu, C. I. Chou, J. Polym. Sci. Part A: Polym.
Chem. 2007, 45, 1007–1015.
47 S. Ganguli, D. Dean, K. Jordan, G. Price, R. Vaia, Polymer
2003, 44, 6901–6911.
75 C. Y. Chiang, R. S. Tsai, W. J. Shu, e-Polymer 2007, 7, 1499–
48 S. Devaraju, M. R. Vengatesan, M. Selvi, A. A. Kumar, I.
Hamerton, J. S. Go, M. Alagar, RSC Adv. 2013, 3, 12915–12921.
1512.
76 H. Y. Tang, N. H. Song, X. F. Chen, X. H. Fan, Q. F. Zhou, J.
Appl. Polym. Sci. 2008, 109, 190–199.
49 D. X. Zhuo, A. J. Gu, Y. Z. Wang, G. Z. Liang, J. T. Hu, L.
Yuan, W. Yao, Polym. Adv. Technol. 2012, 23, 1121–1128.
77 R. S. Tsai, Z. Y. Wang, W. J. Shu, Int. Polym. Proc. 2010, 25,
50 G. Z. Zhan, X. L. Tang, Y. F. Yu, S. J. Li, Polym. Eng. Sci.
2011, 51, 426–433.
125–131.
78 A. J. Guenthner, M. C. Davis, K. R. Lamison, G. R. Yandek,
L. R. Cambrea, T. J. Groshens, L. C. Baldwin, J. M. Mabry,
Polymer 2011, 52, 3933–3942.
51 S. Jothibasu, A. A. Kumar, M. Alagar, High Perform. Polym.
2011, 23, 11–21.
778
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
79 D. A. Shimp, S. J. Ising, J. R. Christenson, High Tempera-
ture Polymers and Their Uses; Society of Plastics Engineers:
Cleveland, OH, 1989; pp 127–140.
84 J. T. Reams, Ph.D. Dissertation, South Dakota School of
Mines, Rapid City, SD, 2011.
85 A. J. Guenthner, J. T. Reams, K. R. Lamison, S. M. Ramirez,
D. D. Swanson, G. R. Yandek, C. M. Sahagun, M. C. Davis, J.
M. Mabry, ACS Appl. Mater. Interfaces 2013, 5, 8772–8783.
80 A. J. Guenthner, M. C. Davis, M. D. Ford, J. T. Reams, T. J.
Groshens, L. C. Baldwin, L. M. Lubin, J. M. Mabry, Macromole-
cules 2012, 45, 9707–9718.
86 S. L. Simon, J. K. Gillham, J. Appl. Polym. Sci. 1993, 47,
461–485.
81 A. J. Guenthner, G. R. Yandek, J. M. Mabry, K. R. Lamison, V.
Vij, M. C. Davis, L. R. Cambrea, Insights into moisture uptake and
processability from new cyanate ester monomer and blend stud-
ies. SAMPE International Technical Conference; SAMPE Interna-
tional Business Office: Covina, CA, 2010; paper 42ISTC-119.
87 D. A. Shimp, S. J. Ising, Advanced Materials: the Challenge
for the Next Decade. In 35th International SAMPE Symposium
and Exhibition; G. Janicki, V. Bailey, H. Schjelderup, Eds.;
SAMPE International Business Office, Covina, CA, 1990; p 1045.
82 M. R. Kamal, S. Sourour, Polym. Eng. Sci. 1973, 13, 59–64.
83 J. M. Kenny, J. Appl. Polym. Sci. 1994, 51, 761–764.
88 O. Georjon, J. Galy, Polymer 1998, 39, 339–345.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 767–779
779