International Journal of Adhesion and Adhesives 82 (2018) 254–262

Contents lists available at ScienceDirect

International Journal of Adhesion and Adhesives

journal homepage: www.elsevier.com/locate/ijadhadh

Polycyclic siloxanes: Base resins for novel high temperature resistant T

platinum curing transparent silicone adhesives
⁎
K. Indulekhaa, S. Monishab, Deepthi Thomasc, R.S. Rajeeva, , Dona Mathewa, K.N. Ninand,
C. Gourie
a
Polymers and Special Chemicals Division, Vikram Sarabhai Space Centre, Thiruvananthapuram 695022, Kerala, India
b
S.N. College, Varkala, Kerala, India
c
Analytical and Spectroscopy Division, Vikram Sarabhai Space Centre, Thiruvananthapuram 695022, Kerala, India
d
Propellants, Polymers, Chemicals and Materials Entity, Vikram Sarabhai Space Centre, Thiruvananthapuram 695022, Kerala, India
e
Polymers and Special Chemicals Group, Vikram Sarabhai Space Centre, Thiruvananthapuram 695022, Kerala, India

A R T I C L E I N F O A B S T R A C T

Keywords: There is increasing demand for polymers having optical transparency along with high thermal stability for
Polycyclic silicone specific applications. Herein, the synthesis of high performance, mainly ethylene bridged, polycyclic silicones
Flame retardancy possessing multiple functionalities like optical clarity, high temperature stability and inherent flame retardancy
High temperature resistant adhesive is reported which is formulated further as high temperature resistant polymeric adhesive. These specialty sili-
Hydrosilylation
cones were synthesized from cyclic siloxanes through hydrosilylation reactions. The reaction conditions were
optimized to synthesise the polycyclic silicone system with varying 2,4,6,8-tetramethyl-2,4,6,8-tetra-
vinylcyclotetrasiloxane (D4V): 1,3,5,7-tetramethylcyclotetrasiloxane (D4H) ratio and the effect of cyclomer ra-
tios on various physical, thermal, dynamic mechanical characteristics were explored. Among different cyclomer
feed ratios, the ratio of 1:1 resulted in effective hydrosilylation with improved thermal stability. High tem-
perature resistant adhesive formulations were made with a polycyclic silicone polymer as base resin, which
showed improved strength at 350 °C than at RT, and retained 25% of its strength at 450 °C. All conventional
silicone polymeric adhesives fail catastrophically at this temperature. The network formation in the polymer
resulted in higher decomposition temperature and increased flame retardancy, which are essential requirements
for high temperature resistant adhesives.

1. Introduction reactions from tetrafunctional siloxane cyclics have been reported

[9–12]. The membrane obtained from 1,3,5,7-tetravinyl, 1,3,5,7-tetra-
Silicone based adhesives are known for good adhesion to low energy methylcyclotetrasiloxane (D4V) and 1,4-bis(dimethylsilyl)benzene has
surfaces and outstanding thermal stability. Most commonly used sili- been reported to possess good permeability coupled with ideal se-
cone polymers are resins containing dimethylsiloxy and diphenylsiloxy lectivity relationships, especially for CO2/CH4 [9]. In this hybrid
groups along with a vinyl or acrylic functionality for crosslinking re- system, the introduction of the rigid 1,4-silphenylene group between
actions. In spite of several advantages, these adhesives fail for appli- the siloxane cyclics, in stoichiometric amounts, increased the Tg value
cations necessitating elevated temperatures. The thermal stability of the in relation to silicone membranes. On the other hand, polymeric net-
silicones can be improved by the insertion of rigid groups into the chain works obtained by free radical polymerization of divinylbenzene, DVB,
backbone, where the rigid structures hinder the intra-chain rearrange- are densely crosslinked and brittle materials, with residual non-reacted
ments that cause polymer degradation at higher temperatures [1–7]. double bonds, particularly with higher DVB contents [13]. In addition,
Another way to improve the thermal stability is through close-packed DVB-based networks obtained by the introduction of dimethylsiloxane
polymer network formation. Similar to the previous case, the usual of different chain lengths, showed clear boundaries between the soft
thermal degradation mechanisms are minimized in polymer networks, and hard phases in the material [14]. The product of platinum-cata-
due to higher restriction of the rearrangements and suppression of re- lyzed hydrosilylation reaction between 1,3,5,7- tetra-
active end blocks [8]. methylcyclotetrasiloxane (D4H) and D4V has been reported as a “hard
The development of different polycyclic silicones by hydrosilylation transparent glass” and the materials prepared from 1:1 and 1:2 M ratios

⁎
Corresponding author.
E-mail address: [email protected] (R.S. Rajeev).

https://doi.org/10.1016/j.ijadhadh.2018.02.001
Received 8 October 2017; Accepted 26 January 2018
Available online 08 February 2018
0143-7496/ © 2018 Elsevier Ltd. All rights reserved.
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

of D4V and D4H were found to be significantly more thermally stable

than PDMS; this was attributed to the absence of a pathway to form the
cyclic trimer, which is the major thermal decomposition product of
PDMS [8]. The networks prepared from 1:2 and 2:1 M ratios of D4V and
D4H were reported to be amorphous by XRD [10]. The hydrosilylation
reaction between D4H and D4V was reported to be highly exothermic in
one case and burst into flames [10]. In another case, the exotherm and
reaction rate were controlled easily by adjusting the platinum con-
centration and reaction temperature and extremely cross-linked D4V/
D4H based polymers with good optical, mechanical and surface prop-
erties were obtained [12]. However, from the survey of literature re-
ports available, it became evident that the multiple qualities, properties
and potentials of both the reaction and the product have not been fully
explored so far by researchers.
Though it has been reported that the polycyclic silicone (with 1:2
ratio of D4V:D4H) exhibits good thermal stability along with high UV-
visible transparency [12], detailed investigations on the influence of
reaction time, temperature and mole ratios of D4V: D4H on the prop-
erties of the products have not been reported so far. Also, the effect of
cyclomer mole ratios on characteristics such as thermal, solvent re-
sistance and flame retardancy were not studied till now. Hence, herein
an attempt is made to optimise the reaction conditions to synthesise the
polycyclic silicone system with 1:2 M ratio of D4V: D4H and to employ
the optimised conditions for the synthesis of polycyclic silicones with
varying D4V: D4H ratios. Further, the effect of cyclomer ratios on var-
ious physical, thermal, dynamic mechanical and flame retardancy
characteristics were explored. A high temperature resistant adhesive
system also was developed based on the optimised polymer; and to the
best of our knowledge, this is the first time that a polycyclic silicone
based adhesive system is reported which possesses the capability of
withstanding very high temperatures, ~ 450 °C.

2. Experimental

2.1. Materials
Scheme 1. Synthesis of polycyclic silicones from tetrafunctional oligomers.
The precursors used were 2,4,6,8-tetramethyl-2,4,6,8-tetra-
vinylcyclotetrasiloxane (D4V) and 1,3,5,7-tetramethylcyclotetrasiloxane crosslinking with Pt catalyst. For this, the polymer was mixed thor-
(D4H), procured from Alfa Aesar. 0.05 M Pt (0) 1,3-divinyl 1,1,3,3-tet- oughly with 2 wt. % Pt catalyst for 2–3 min and cured under ambient
ramethyldisiloxane complex (Karstedt's catalyst) in vinyl terminated conditions for 7 days.
polydimethylsiloxane (V-PDMS) purchased from Aldrich was diluted to
0.01 M with V-PDMS, procured from Anabond, Chennai and used as
catalyst. Boron nitride (BN) with particle size < 10 µm purchased from 2.4. Characterization
CDH and ferric oxide (Fe2O3) (assay: 90%) procured from Fischer
Scientific, silica with Particle size < 20 µm purchased from CMET, Kerala The viscosity of the polymers was measured at 25 °C using a
and aluminum oxide (Al2O3, fused,− 325 mesh) from CDH were used as Brookfield Viscometer. FTIR spectra were taken using a Perkin Elmer
the filler materials. Spectrum GX A FTIR spectrophotometer. 1H NMR and 29Si NMR spectra
were recorded using a 400 MHz BRUKER nuclear magnetic resonance
2.2. Synthesis of polycyclic silicones spectrometer (Bruker, Germany), using CDCl3 as the solvent. Chemical
shifts (δ) are expressed in ppm, downfield using tetramethylsilane
Polycyclic silicones, T1, T2 and T3 were synthesized following the (TMS) as an internal standard. Refractive indices were determined
method reported by Zheng et al. [12]. The synthesis of T3 with 1:2 M using an Abbe refractometer (make: Atago/Japan, Model: DR-A1).
ratio of D4V and D4H is described here as a typical example. The tet- Perkin Elmer Lambda 950 UV/Vis spectrometer was employed for the
rafunctional oligomers, D4V [0.03 mol, 10.36 ml], D4H [0.06 mol, determination of UV-visible transparency of the samples.
14.56 ml], and 0.01 M Pt (0) 1,3-divinyl 1,1,3,3-tetramethyldisiloxane The cure reaction of the polymeric system was studied using TA
complex in V-PDMS [0.07 wt. % of total reactants/ 0.018 g] were instruments Q-20 differential scanning calorimeter (DSC). About
charged into a three necked, round bottom flask equipped with con- 10 ± 1 mg of sample was taken in aluminium pans and was heated from
denser, stirrer and thermometer. Reaction was carried out at 65 °C with room temperature to 200 °C at heating rate of 10 °C/min in an ultra-
vigorous stirring at 500 rpm for 4 h. The reaction pathway is depicted in pure nitrogen (99.999%) atmosphere at a flow rate of 50 mL/min. The
Scheme 1. The same procedure was adopted for the synthesis of other thermal decomposition behavior of the cross-linked polymeric systems
copolymers with varying ratios of D4V and D4H. was studied using a TA instruments SDT Q-600 simultaneous thermo-
gravimetric–differential scanning calorimetric analyzer (TG-DSC).
2.3. Cure procedure About 10 ± 1 mg of sample was taken in alumina crucible and the TG
analysis was carried out from room temperature to 900 °C at a heating
The cure procedure adopted for the composites in the study includes rate of 10 °C/min in an ultra-pure nitrogen atmosphere at a flow rate of
one day cure under ambient conditions followed by a second stage 100 mL/min. The glass transition temperature (Tg) of the cured

255
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

elastomers was measured using a TA instruments DMA Q800. The Table 1

analysis was done in tension mode at a frequency of 1 Hz and an am- Physical states of polymers formed under different reaction conditions (molar ratio of 1:2
for D4V:D4H, 0.07 wt. % catalyst).
plitude of 5 μm. The samples were cooled to −150 °C and then heated
from −150 °C to 50 °C at a heating rate of 3 °C/min. Temperature (°C) Time of reaction (h) Physical nature of the product
Thermogravimetry-coupled with mass spectrometry (TG–MS) was done
using a Perkin-Elmer Pyris-1 TGA-SQ-8T quadrupole mass spectro- 30 24 Liquid
50 6 Oily
meter. TG analysis was done at a heating rate of 10 °C/min under a
65 4 Resinous
helium atmosphere in the temperature range 100–950 °C. The flamm- 80 1.5 Solid
ability of the polymers was determined by measuring its limiting
oxygen index (LOI) value. LOI of all the samples was measured using a
Dynisco LOI analyzer according to ASTM D-2863. The samples of size Transparent polymers resulted under all the conditions employed;
150 mm x 6 mm x 3 mm were subjected to a burning test. The top however, they were having varying consistencies as given in Table 1.
portion of the sample in a pre-set oxygen- nitrogen atmosphere was An increase in temperature accelerated the crosslinking of cyclic oli-
ignited using a flame and was tested for its flame sustainability. The gomers. Even when the reaction was conducted at low temperature, i.e,
percentage of oxygen present in the O2 - N2 atmosphere in which the 30 °C for 24 h continuously, the product was obtained in a very low
flame sustains for 180 s was recorded as the LOI value of the sample. viscous consistency indicating that almost no reaction had taken place.
Lap shear strength (LSS) on stainless steel (SS) was measured using When the temperature was increased to 50 °C, an oily product was
UTM-Instron 5500 K5251 at a crosshead speed of 1 mm/min as per obtained after 6 h. At 65 °C, a resinous product was obtained in 4 h;
ASTM D 1002 using test specimens of dimensions 100 mm x 25 mm x whereas at 80 °C, the acceleration was at its peak and a transparent
1.6 mm. Surfaces of the samples were cleanined with TCE and acetone. solid product resulted within 1.5 h. From this, it is evident that the
After drying the surface, a thin coat of primer, a low viscous epoxy- temperature has a profound role to play on the hydrosilylation reaction
silicone hybrid (proprietory item of VSSC synthesised from epoxy and a proper control of the reaction temperature and duration is es-
functional monomer and alkoxyfunctional silanes) was applied, and sential to realize desired polycyclic networks. Since it is desirable to
kept undisturbed for 90 min at ambient conditions. The hydrolysed have the polymers in a resinous consistency for practical applications
primer acts as a coupling agent between the adhesive and SS surface. where it can be easily converted to a usable form by further cross-
The adhesive was then applied over the primed surface and allowed to linking, the reaction condition for synthesis of polycyclic silicones in
cure. For determining LSS at higher temperature, the bonded specimens this study was optimized as 65 °C for 4 h. Accordingly, the polycyclic
were soaked for 10 min at the desired temperature and then tested at silicones with varying molar ratios of D4V:D4H were also synthesized
that temperature. under the optimized conditions. Since the polymer contains residual Pt
catalyst, the cross-linking reaction proceeds slowly while being main-
2.5. Solvent swelling and cross-link density measurement tained under ambient conditions with solidification generally occurring
within 2 days. Hence, to get a pot life with a minimum of 2 weeks to be
For carrying out swelling studies in toluene, samples were cut into useful as a material for practical applications, the synthesized polymer
10 mm x 10 mm x 3 mm (length x breadth x thickness) and weighed was immediately stored in a sealed container under low temperature
(W0). The samples were then immersed in toluene (25 ml) in a 50 ml (0–3 °C) in a deep freezer until further use.
beaker and soaked for 10 min. The swollen sample was removed from For realizing polycyclic silicones with varying D4V and D4H con-
the beaker and excess solvent removed from the surface of the sample tents, D4V to D4H mole ratios in the reaction mixture were varied from
using tissue paper and the mass of the swollen sample (W) determined. 1:0.5 to 1:2 and the reactions were carried out under the optimized
The experiment was repeated with an interval of 10 min. The percen- conditions of 65 °C, 4 h. Table 2 summarises the different mole ratios
tage of swelling was calculated as, studied and the data on viscosity and refractive index of the resulting
polymers, determined 24 h after completion of synthesis. From the
% swelling = (W − W0)/W0 × 100 (1)
viscosity data, it is clear that network formation is in the initial stages
FR
The cross-link density (n ) was estimated using the Flory–Rehner for T1 and T2 and there is no significant increase in viscosity with molar
equation, ratio change. T3 shows a substantial build-up in viscosity indicating
that pronounced network formation has happened in this case, where
nFR = −[v2 + v22χ1 + ln(1 − v2)]/V1(v21/3 − 0.5v2) (2)
D4V /D4H is 1:2 mol ratio. RI values of the copolymers are influenced
where, v2 corresponds to the polymeric volume fraction in the swollen mainly by two factors: (i) contribution from the cyclomers present in
mass, V1 is the solvent molar volume and χ1 is the Flory–Huggins varying contents, and (ii) extent of network formation in the polymer,
polymer–solvent interaction parameter. The parameter v2 was calcu- with higher RI for the densely cross-linked system where pronounced
lated by Eq. 3. hindrance to the path of light occurs. From Table 2, it can be seen that
T1 and T3 showed a higher RI compared to T2. In the case of T1, as per
v2 = m1d s /(m1 (d s − d2) + m2 d2) (3)
the feed ratio, D4V having a higher RI (1.4319 for D4V vs. to 1.3846 for
where, m1 and m2 are the masses of original sample and sample swollen D4H) is predominant in the system. From the viscosity data it is clear
to equilibrium, respectively, and ds and d2 are the densities of solvent that cyclomers are almost in the unreacted state with minimum ad-
and cured samples, respectively. vancement of cross-linking and hence, D4V decides the RI of resulting

3. Results and discussion Table 2

Mole ratio of feed along with viscosity and RI data (determined after 24 h of synthesis) of
3.1. Synthesis and characterization of polycyclic silicones polycyclic silicones.

Polymer D4V / D4H Viscosity at Refractive index Physical state

Polycyclic silicones were synthesized using hydrosilylation ap- reference (mole 25 °C (Pa.s) at 25 °C of polymer
proach by following Scheme 1. Preliminary studies were carried out to ratio)
optimize the reaction time and temperature using a 1:2 M ratio of
T1 1: 0.5 20 × 10−3 1.4357 Liquid
D4V:D4H (the best molar ratio reported by Zheng et al. [12]), with
T2 1: 1 30 × 10−3 1.4263 Liquid
0.07 wt. % catalyst. Table 1 summarises the physical nature of the T3 1: 2 310 × 10−3 1.4389 Resinous
products formed under different conditions of temperature and time.

256
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

polymer. In T2, it is the combined effect of both the cyclomers (in 1:1
mol ratio) and the low level of cross-linking that resulted in the low RI
value. For T3 with a higher fraction of D4H in the feed, the RI was
expected to be lower than T1 and T2. However, RI is found to be the
highest (1.4389) for T3. This is attributed to the substantial reaction
that has proceeded between the cyclomers forming networks that dis-
rupts the path of light and hence causes an increase in the RI value
[15,16]. This is corroborated by the fairly high viscosity value of T3
also.
T2 with 1:1 ratio of cyclomers is expected to give the maximum
extent of reaction. However, from the viscosity and RI data, it can be
concluded that a molar excess of D4H in the system (T3) promotes the
extent of reaction (hydrosilylation). With the bulky vinyl groups in D4V,
its reactivity is low compared to D4H. In T1, with a molar excess of D4V,
steric hindrance offered by bulky vinyl groups retards the extent of
hydrosilylation and results in a low viscosity for the polymer. The
conditions are most favorable in the case of T3, with a molar excess of
D4H that dilutes the reaction medium also and results in more extensive
network formation. The RI values obtained here for the optically clear
polycyclic silicones are quite high when compared to conventional
vinyl terminated polydimethylsiloxane (V-PDMS) (RI = 1.4034). The
effect of storage for 2 weeks under low temperature, on the RI at 25 °C Fig. 2. 1H NMR spectrum of T1.
was evaluated and found as 1.4367, 1.4301 and 1.4401 for T1, T2 and
T3, respectively. It can be seen that there is a slight increase in the RI of Table 3
all the systems due to an increase in the extent of crosslinking in the Signal integration data from 1H NMR of polycyclic silicones.
systems.
Polymer Feed ratio From 1H NMR
Fig. 1 shows FTIR spectra of the polymers taken after 24 h of
synthesis. The characteristic peaks of Si-O-Si are present at 1081 cm−1 Vinyl Si-H Methyl Vinyl Si-H Methyl -CH2-CH2- -CH(CH3)
in all the polymers. The peaks at 1250 and 2165 cm−1 show the pre-
sence of Si–C and Si–H stretching vibrations, respectively. A peak due to T1 33 17 50 27.4 9.9 54.8 6.2 1.6
the vinyl group -CH=CH2 is seen at 1593 cm−1 while the one at T2 25 25 50 21.3 17.7 54.6 4.9 1.3
T3 17 33 50 9.5 25.5 55.8 7.4 1.7
3055 cm−1 is assigned to the C–H stretching vibrations of the -C=C–H
units in the vinyl groups [17]. As evident from Fig. 1, peaks due to Si-H
and Si-CH=CH2 are visible in the spectrum of all polymers, attributed in network formation. Table 3 summarises the signal integration data
to the residual functionalities in the polymers. The polymers were from the 1H NMR spectra of the polymers. From this, it is evident that
further characterized by 1H NMR spectroscopy and Fig. 2 represents the the increase in the signal intensity in the C–H region occurs due to the
spectrum obtained in the case of T1 taken after 24 h of synthesis, as a increase in the network formation as more ethylene bridges are formed
typical example. The signal at 5.80–6.20 ppm corresponds to the vinyl between the cyclomers. Further, it is clear that the decrease in con-
group in the polymer and that at 4.70 corresponds to the proton of Si–H centration of vinyl and Si-H in polycyclic silicones are compensated by
bond. The signal at 0.50 ppm is due to proton of Si-CH2-CH2-Si groups the increase in -CH2-CH2- signal intensity, which indirectly indicates
(produced due to addition of Si-H to vinyl against Markovnikov rule) the extent of the hydrosilylation reaction. The signal intensity in the -C-
and that at 1.20 ppm corresponds to the methyl of Si-CH(CH3) (pro- H region is higher in T3, due to increased network formation, which
duced due to Markovnikov addition). The signals at 0.10 and 0.20 ppm supports the viscosity and RI results discussed earlier.
correspond to the methyl protons of Si–CH3. The peak at 1.60 ppm The polymers were further characterised by 29Si NMR spectroscopy
could be due to trapped moisture. The increased signal intensity at 0.50 and the spectrum obtained for T1 is given as a representative example,
ppm indicates the presence of more ethylene bridges due to the increase in Fig. 3. The spectrum shows the presence of Si in two different che-
mical environments. The signal appearing in the range −15 to −20
ppm corresponds to Si attached to CH3 and -CH2-CH2- groups, and that
at −32 to −37 ppm is due to the presence of Si(CH3)(CH2=CH2) and
silicon attached to hydride groups. This is in accordance with the ex-
pected structure of polycyclic silicone with -CH2-CH2- linkages between
the silicon atoms as well as having unreacted residual moieties attached
to silicon atoms.

3.2. Characterisation of self-cross linked systems

The self-crosslinked polymers (kept as such after preparation under

cold storage conditions for 30 days) were analysed by FTIR spectro-
scopy, thermal and dynamic mechanical properties. Fig. 4 represents
the FTIR spectra of the cured polymers. The peaks in the cured systems
are broader than that of the uncured systems. The peak at 1260 cm−1
shows the presence of Si–C. The peaks at 2168 cm−1 and 1597 cm−1
show the presence of unreacted Si-H and vinyl groups respectively, in
all the three polymers. The functional groups, from network precursor
Fig. 1. FTIR spectra of as-prepared, polycyclic silicones.
units, were not totally incorporated in the network structure by the

257
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

Fig. 5. Tan delta vs. temperature profiles of crosslinked polymers.

phase, on keeping undisturbed under cold conditions, all the three

polymers crosslinked to the same extent, with no considerable differ-
ence in the network structure formed after crosslinking, as evidenced
Fig. 3. 29
Si NMR spectrum of T1. from the FTIR spectral and Tg data.
The self-crosslinked systems obtained were optically clear and a
representative UV-visible spectral data of T2 given as supporting in-
formation, S1 shows that there is no considerable absorption in the
entire visible range, similar to that of the neat, uncrosslinked system.

3.3. Thermal properties

Thermogravimetric analysis was done to evaluate the thermal stabi-

lity of the crosslinked polycyclic silicones. Fig. 6 shows the thermograms
of polymers formed with different feed ratios. T1 and T2 show initial
weight loss of < 1% up to 300 °C; whereas, T3 shows higher loss of 3%
up to 300 °C, attributed to the evolution of unreacted precursors present
in the cured network. As T3 exhibited higher viscosity build-up at the
initial phase due to increased hydrosilylation (as evidenced from visc-
osity, RI and NMR data), the possibility of inclusion of unreacted cyclo-
mers in the network structures are more pronounced in T3, which sub-
sequently lead to a mass loss of 3% on heating up to 300 °C. The polymers
T1, T2 and T3 (after evolution of unreacted precursors) show high onset
Fig. 4. FTIR spectra of self cross-linked polymers.
decomposition temperatures; thus, the temperature for 5% weight loss
(Td5) is in the range 600 − 630 °C with ceramic residues of 82%, 85%,
hydrosilylation reaction, mainly due to the increase of the steric hin- and 84.4%, respectively observed at 900 °C. The increased thermal sta-
drance around these groups in the course of the reaction. After gelation bility of these polymers when compared to conventional polysiloxane
of the system, the mobility of the reactants and catalyst get diminished PDMS, which depolymerises at 300–350 °C, is attributed to the highly
which further makes the reaction difficult. In a polyfunctional system, cross-linked polycyclic structure, which retards the possibility of the
even after annealing, a considerable amount of unreacted, but reactive, unzipping of siloxane chains, and hence, depolymerisation.
groups still remain unable to cure, because of steric separation from the
corresponding reactive groups needed to effect cure. The inequality in
the reactivity of functional groups (i.e. when two or three groups of a
molecule have already reacted, the reactivity of the last one should be
lower due to steric hindrance) increasing the probability of the for-
mation of linearly linked clusters has also been reported [18].
Dynamic mechanical analysis was performed to find out the glass
transition temperature (Tg) of the self-crosslinked polycyclic silicones.
Fig. 5. shows the tan delta profiles obtained for the polymers. In the
DMA profile of the polymers, multiple transitions were observed with
less predominant sub-transitions in T3. The multiple transitions ob-
served can be attributed to different molecular architectures like
polycyclic, cyclo-linear, etc, attributed to the inequality in the re-
activities of functional groups. These polycyclic silicones have sub-
stantially high Tg values (−20 to −10 °C) when compared to the
conventional silicones (around −120 °C). This is attributed to the in-
creased network formation resulting in rigid molecular architectures.
Though T3 showed a higher extent of hydrosilylation at the initial Fig. 6. Thermograms of cross-linked polycyclic silicones.

258
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

Fig. 7. TG-MS trace analysis profile of [a] T1, [b] T2, and [c] T3.

Since the polymers exhibited good thermal stability, it is quite in- 660 °C in T1. This fraction is very minimum in the decomposition
teresting to find out the products of decomposition at higher tem- products of T2 and T3, attributed to increased network formation that
peratures. The decomposition products were analysed by TG-MS. Fig. 7. hinders depolymerisation and volatilization. The benzene fraction in
[a], [b] and [c] show the TG-MS trace analysis data of T1, T2 and T3, polymers might have been formed by the recombination reactions of
respectively. The onset decomposition temperature is 500 °C in T1 and vinyl fragments at high temperatures [19]. This fraction is also max-
T3, whereas decomposition starts at a higher temperature of 570 °C in imum in the case of T1, attributed to the increased amount of unreacted
T2. The decomposition products in each case comprised of a major vinyl due to the molar excess of D4V in the reactants.
fraction of methane along with other volatile species such as ethane, Since T2 exhibited the highest char yield at 900 °C (85%), the flame
propene, benzene, toluene, hexamethylcyclotrisiloxane (D3) and octa- resistance characteristics of the system was further evaluated. The LOI
methylcyclotetrasiloxane (D4), which are produced as a result of frag- of the system was determined as 34%, which is quite high for an un-
mentation and recombination reactions taking place at higher tem- filled siloxane system without any hetero atoms in the back-bone.
peratures. The peak intensity for all the species is in the order Conventional PDMS based silicones exhibit an LOI around 18% only
T1 > T3 > T2. For a better comparison, the TG-MS profiles of D4 and and they will burn in air unless compounded with suitable fire resistant
benzene for all the three polymers are compared and shown in Fig. 8 [a] additives and/or fillers. For the present system, it is to be noted that the
and [b]. The evolution of D4 starts at around 500 °C and prevails up to flame retardancy is achieved inherently through structural modification

Fig. 8. TG-MS trace analysis profile of polymers corresponding to [a] D4 fraction, and [b] benzene fraction.

259
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

Fig. 10. DSC cure profiles of T2 with different catalyst concentrations.

Fig. 9. Swelling behavior of T2 and V-PDMS in toluene.

of the silicone resin, alleviating the need for external additives/fillers.

Here, due to the modification of the siloxane back-bone with a poly-
cyclic network, the possibility of depolymerisation by unzipping in the
cyclomers is not there. This leads to the formation of high yield of
ceramics, which prevents the propagation of flame and hence, resulted
in high flame retardancy.

3.4. Assessment of crosslink density and structural features

Since the polycyclic silicones synthesised here exhibited greater ri-

gidity when compared to conventional silicones, swelling experiments
in toluene were carried out to assess the resistance of the crosslinked
networks towards swelling/decomposition [20]. All three systems ex-
hibited the same swelling behavior in toluene with maximum swelling
of 90% at the end of 2 h. The swelling behavior of polycyclic silicones
was compared with that of the conventional silicone, that is, V-PDMS.
Fig. 9 depicts the swelling behavior of T2, as a representative example Fig. 11. Cure behavior of T2 with 2% catalyst after different durations of cure.
vs. V-PDMS. From the figure it is evident that the polycyclic silicone
system swells to < 50% of the conventional silicone. V-PDMS showed
continuous swelling and reached 200% after 2 h immersion; whereas, Fig. 11, which found that even after 2 weeks, complete curing is not
after initial immersion to 40 min, T2 did not show any sign of further effected. Also, it is seen that though there is a considerable decrease in
swelling. The cross-link density of V-PDMS and T2 was calculated using ΔH after cure from day-1 onwards (75.1 vs. 424.5 J/g), even after 2
the Flory-Rehner equation, by using a χ value of 0.48 for the poly- weeks it shows residual cure with ΔH of 71 J/g. It is evident that day-1
dimethylsiloxane–toluene system, for the estimation [21]. The cross- onwards the rate of decrease in ΔH is very low, which can be attributed
link density of V-PDMS was 1.1 × 10−4 mol/cm3; whereas, it was 4.32 to the hindrance for further curing subsequent to the formation of
× 10−4 mol/cm3 for T2. The increased solvent resistance shown by the networks within a day. Also, there is a systematic shift in the cure
polycyclic silicone, T2 can thus be attributed to the increased cross-link exotherm peak from 96 °C to 146.5 °C with respect to time of curing.
density (4 times higher than conventional V-PDMS) due to the network This can also be attributed to the increased barrier for further
structure.

3.5. Cure studies by DSC

The polycyclic silicone system with a 1:1 mol ratio of D4V: D4H (T2)
was found to be promising in terms of thermal stability. However, since
the catalyst concentration in the polymer is very low, for effecting the
physical crosslinking, the system required a long duration of curing
under cold storage conditions of nearly a month. Hence, a second stage
of curing was carried out to increase the rate of crosslinking, by mixing
of the resin with additional quantities of Pt catalyst. Initial experiments
were carried out to optimize the catalyst concentration. The catalyst
concentrations chosen were 2, 6 and 10% (by weight) and the corre-
sponding DSC profiles obtained are shown in Fig. 10. The system with
the lowest (2%) Pt catalyst concentration exhibited a peak cure tem-
perature at 96 °C which got shifted to 82.8 °C with 10% catalyst con-
centration, with a ΔH of 425 J/g in all cases.
The system with 2% catalyst concentration was checked for residual
Fig. 12. Cure profile of T2 with 10% catalyst.
cure by carrying out DSC after regular intervals of cure, as shown in

260
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

Fig. 13. [a] Dynamic and [b] isothermal thermograms of T2/composite.

Table 4 system with 5 times higher amount of catalyst (10%) and observed an
LSS data of T2 based adhesive with different catalyst concentrations. improved strength at RT, attributed to increased cross linking at a much
lower temperature as shown in Fig. 10. However, its higher tempera-
Catalyst LSS at RT LSS at 350 °C ( LSS at 450 °C (
concentration (%) (Average, MPa) Average, MPa) Average, MPa) ture performance was not promising, and the adhesive strength at RT
was reduced considerably to 1.0 MPa when temperature was increased
2 2.0 (0.6) 2.4 (0.6) 0.5 (0.05) to 350 °C. This can be attributed to the increased brittleness on in-
10 2.8 (0.4) 1.0 (0.3) – corporating higher catalyst concentration, due to randomly oriented
network formation. Hence, it can be concluded that the T2 based ad-
hesive system with a low amount of catalyst (2%) is a promising ad-
crosslinking, due to extensive network formation.
hesive capable of withstanding a high temperature environment, with
The system with the highest Pt catalyst concentration (10%) was
proven capability up to 450 °C, where other conventional silicone based
also checked for residual cure after 2 weeks and the cure profile ob-
adhesives fail.
tained is shown in Fig. 12. There is a considerable decrease in ΔH from
425.8 J/g to 118.6 J/g, but with a broadened cure exotherm shifted to a
4. Conclusions
higher temperature regime. Again, this is due to increased network
formation with the higher catalyst concentration, which might have
Optically clear polycyclic silicone polymers were synthesised
resulted in randomly crosslinked structures with a considerable amount
through a hydrosilylation approach by varying the D4V:D4H mole ratios
of residual functionalities.
and were characterized in detail by spectroscopic and thermal methods.
The onset decomposition temperatures of the polycyclic silicone sys-
3.6. Polycyclic silicone as base resin for high temperature adhesive tems were very high when compared to conventional silicones, which
was attributed to increased network formation that hinders the depo-
Since T2 exhibited high temperature stability with a peak decom- lymerisation of the siloxane back bone, which subsequently resulted in
position temperature at 600 °C, it was further evaluated as the base higher flame retardancy. The high temperature resistant silicone poly-
resin for a high temperature resistant adhesive. The adhesive was for- mers thus synthesized were formulated in to an adhesive with added
mulated by mixing T2 with inorganic fillers (a mixture of 22% BN, 18% catalyst, and showed improved strength at 350 °C than at RT, and re-
silica, 1.3% Fe2O3, and 13% Al2O3) followed by mixing with 2% pla- tained 25% of its strength at 450 °C, where conventional silicone
tinum complex catalyst to form T2/composite. The composite system polymeric adhesives fail catastrophically. Thus, the synthesised poly-
exhibited thermal stability up to 570 °C without any weight loss, as cyclic silicones are proven to be high performance polymeric systems
evidenced from the thermogram shown in Fig. 13 [a]. Further, iso- possessing high temperature stability with inherent flame retardancy
thermal TGA of the composite was carried out at 300 °C, 400 °C and and base resin for polymeric adhesive possessing very high thermal
500 °C for 1 h and no appreciable weight loss is observed at 300 °C and stability.
400 °C. A weight loss of only 3.5% is observed at high temperature of
500 °C for a residence time of 1 h, illustrating the high temperature Acknowledgements
stability of the polymer based composite. Fig. 13 [b] shows the iso-
thermal thermograms. Experimentally it is proven by conducting LSS of The authors thank Director, Vikram Sarabhai Space Centre (VSSC),
adhesive on suitably primed SS substrates at different temperatures, Thiruvananthapuram, India for permission to publish this work and
after curing them at RT for 1 week. Table 4 summarizes the LSS results colleagues in Polymers and Special Chemicals Division and Analytical
obtained at different temperatures. The results shown are the average of Spectroscopy Division, VSSC for their support.
4 experimental results in each case. The standard deviation from the
mean is provided in brackets along with the average values. Appendix A. Supplementary material
The adhesive system with 2% catalyst exhibited good strength at
higher temperatures. The adhesive exhibited higher strength at 350 °C Supplementary data associated with this article can be found in the
than at RT, attributed to further crosslinking of residual functionalities online version at http://dx.doi.org/10.1016/j.ijadhadh.2018.02.001.
at higher temperatures. Beyond 350 °C, the adhesive started to lose its
strength. However, at 450 °C, it retains 25% of its strength at RT, which References
is quite high for a polymeric adhesive. Beyond this temperature, the
system showed debonding. In order to determine the effect of cross- [1] Dvornic PR, Lenz RW. Exactly alternating silarylene-siloxane polymers. 10.
Synthesis and characterization of silphenylene-siloxane polymers containing
linking on LSS, an adhesive was formulated from the same compounded

261
K. Indulekha et al. International Journal of Adhesion and Adhesives 82 (2018) 254–262

fluoroalkyl and hydrido side groups. Macromolecules 1994;27(20):5833. Technol 2002;123:232.

[2] Dvornic PR; Lenz RW. “High Temperature Polysiloxane Elastomers”; Huthig & [12] Zheng P, McCarthy TJ. Rediscovering silicones: molecularly smooth, low surface
Wepf: Basel; 1990. energy, unfilled, uv/vis-transparent, extremely cross-linked, thermally stable, hard,
[3] Wang S, Mark JE. Computational analysis of some conformational properties of poly elastic PDMS. Langmuir 2010;26:18585.
(tetramethyl-p-silphenylene) siloxane. Comput Polym Sci 1993;3:33. [13] Matsumoto A. Free-radical crosslinking polymerization and copolymerization of
[4] Wang S, Mark JE. An experimental study of the unperturbed chain dimensions of multivinyl compounds. Adv Polym Sci 1995;123:41.
poly(tetramethyl-p-silphenylene-siloxane). Polym Bull 1993;31:205. [14] Hill DJT, Perera MCS, Pomery PJ, Toh HK. Dynamic mechanical properties of
[5] Zhang R, Pinhas AR, Mark JE. Synthesis of Poly(tetramethyl-m-silphenylenesi- networks prepared from siloxane modified divinyl benzene pre-polymers. Polymer
loxane), an Elastomer of Enhanced High-Temperature Stability. Macromolecules 2000;41:9131.
1997;30:2513. [15] Al-Ma’adeed MA, Al-Qaradawi IY, Madi N, Al-Thani NJ. The effect of gamma ir-
[6] Kawakita T, Oh H-S, Moon J-Y, Liu Y, Imae I, Kawakami Y. Synthesis, character- radiation and shelf aging in air on the oxidation of ultra-high molecular weight
ization and thermal properties of phenylene–disiloxane polymers obtained from polyethylene. Appl Surf Sci 2006;252:3316.
catalytic cross-dehydrocoupling polymerization of bis(dimethylsilyl)benzene iso- [16] Chandrashekara MN, Ranganathaiah C. Chemical and photochemical degradation
mers and water. Polym Int 2001;50:1346. of human hair: a free-volume microprobe study. J Photoch Photobio B
[7] Li Y, Kawakami Y. Synthesis and Properties of Polymers Containing Silphenylene 2010;101:286.
Moiety via Catalytic Cross-Dehydrocoupling Polymerization of 1,4-Bis(di- [17] Smith AL. Analysis of silicones 41. John Wiley & Sons, Inc; 1974.
methylsilyl)benzene. Macromolecules 1999;32(26):8768. [18] Kidera A, Higashira T, Ikeda Y, Urayama K, Kohjiya S. GPC analysis of polymer
[8] Michalczyk MJ, Farneth WE, Vega AJ. High temperature stabilization of crosslinked network formation. Polym Bull 1997;38:461.
siloxanes glasses. Chem Mater 1993;5:1687. [19] Jones BM, Zhang F, Kaiser AI, Jamal A, Mebel AM, Cordiner MA, Charnley SB.
[9] Redondo SUA, Radovanovic E, Torriani IL, Yoshida IVP. “Polycyclic silicone Formation of benzene in the interstellar medium. PNAS 2011;108:2452.
membranes. Synthesis, characterization and permeability evaluation. Polymer [20] Chen D, Nie J, Yi S, Wu W, Zhong Y, Liao J, Huang C. Thermal behaviour and
2001;42:1319. mechanical properties of novel RTV silicone rubbers using divinyl-hexa[(tri-
[10] Stein J, Lewis LN, Gao Y, Scott RA. In Situ determination of the active catalyst in methoxysilyl)ethyl]-POSS as cross-linker. Polym Degrad Stabil 2010;95:618.
hydrosilylation reactions using highly reactive Pt(0) catalyst precursors”. J Am [21] Jose NM, Prado LASA, Yoshida IVP. Synthesis, characterization, and permeability
Chem Soc 1999;121:3693. evaluation of hybrid organic–inorganic films. J Polym Sci B Polym Phys
[11] Schiavon MA, Radovanovic E, Yoshida IVP. Microstructural characterisation of 2004;42:4281.
monolithic ceramic matrix composites from polysiloxane and SiC powder. Powder

262