Bull. Mater. Sci., Vol. 39, No. 3, June 2016, pp. 769–775.


c Indian Academy of Sciences.
DOI 10.1007/s12034-016-1185-z

Influence of nanometric silicon carbide on phenolic resin composites

properties

GEORGE PELIN1,2 , CRISTINA-ELISABETA PELIN1,2 , ADRIANA ŞTEFAN1,∗ , ION DINCĂ1 ,

ANTON FICAI2 , ECATERINA ANDRONESCU2 and ROXANA TRUŞCĂ3
1 National Institute for Aerospace Research and Development ‘Elie Carafoli’, Materials Unit, 220 Iuliu Maniu Blvd,
061126 Bucharest, Romania
2 Politehnica University of Bucharest, Faculty of Applied Chemistry and Materials Science, Department of Science and

Engineering of Oxide Materials and Nanomaterials, 1-7 Gh. Polizu St., 011061 Bucharest, Romania
3 S.C. METAV Research & Development S.A., 31 C.A. Rosetti St., 020011 Bucharest, Romania

MS received 18 August 2015; accepted 8 November 2015

Abstract. This paper presents a preliminary study on obtaining and characterization of phenolic resin-based com-
posites modified with nanometric silicon carbide. The nanocomposites were prepared by incorporating nanometric
silicon carbide (nSiC) into phenolic resin at 0.5, 1 and 2 wt% contents using ultrasonication to ensure uniform disper-
sion of the nanopowder, followed by heat curing of the phenolic-based materials at controlled temperature profile up
to 120◦ C. The obtained nanocomposites were characterized by FTIR spectroscopy and scanning electron microscopy
analysis and evaluated in terms of mechanical, tribological and thermal stability under load. The results highlight
the positive effect of the nanometric silicon carbide addition in phenolic resin on mechanical, thermo-mechanical and
tribological performance, improving their strength, stiffness and abrasive properties. The best results were obtained
for 1 wt% nSiC, proving that this value is the optimum nanometric silicon carbide content. The results indicate that
these materials could be effectively used to obtain ablative or carbon–carbon composites in future studies.

Keywords. Phenolic resin; nanometric silicon carbide; nanocomposites; friction coefficient.

1. Introduction Even though the processing method involves facile routes,

the synthesis conditions and parameters are essential to
Phenolic resin composites have their applications in a wide achieve resins with particular properties [8]. Curing time and
range of fields from wood and adhesive industry to auto- temperature as well as mold materials influence the resulting
motive, aeronautics and aerospace industries. Phenolic resins homogeneity, glass transition temperature and mechanical
represent the first industrially produced synthetic polymer, properties.
first processed by Bayer in 1872 [1] and patented in 1907 Although phenolic resins have good thermo-oxidative
by Baekeland [2]. This class of resins is used in a wide resistance, research studies have been conducted to improve
range of applications in fields from construction and auto- their thermal properties through modification of their struc-
motive materials to high-tech industries such as aeronau- ture with introduction of inorganic additives, including boron
tics and aerospace [3]. This is sustained by their versatile [8], phosphorous [9,10], zirconium [11,12] and silicon [13]
properties, such as superior mechanical strength, heat resis- based compounds. Moreover, studies presented the enhance-
tance, chemical and dimensional stability, high resistance to ment of ablative properties by the addition of carbon
water, acids and solvents [3], supplemented by medium price nanotubes [14–16] and nanofibers [17], graphite [18],
and relatively simple processing techniques via heat curing montmorillonite clays and POSS compounds [19] and
[2]. In the field of advanced composite materials, phenolic nanographene [20].
resin-based composites are known for their excellent flame Owing to exceptional properties such as high temperature
resistance and the fact that they evolve low smoke upon oxidative and thermal shock resistance, as well as good fric-
incineration [3], being intensely used in the rocket industry tion characteristics, silicon carbide is a promising nanoaddi-
because of their ablative characteristics [4]. Other important tive for obtaining high temperature composites [21] as well
applications of phenolic resins are thermal insulation mate- as wear-resistant materials [22,23]. Silicon carbide in nano-
rials, molding compounds, foundry, coating materials, wood metric form has some advantages over its micrometric form,
products industry, adhesives, polymeric blends with other as being a nanofiller it can be added into much smaller con-
polymers and other composite materials [5–7]. tents relative to the polymeric matrix (down to less than 1 wt%)
compared with classical fillers (that require even up to
∗ Author for correspondence ([email protected]) 20 wt% [24]), fact that contributes to achieve minimum weight

769
770 George Pelin et al

increase. Besides this, lower contents required by nanomet-

ric SiC helps in an easier control of matrix viscosity increase
issues, consequently helping to obtain a better homogeniza-
tion with the aim of ultrasonication techniques.
There are several studies that show modified phenolic
resin-based composites processed using advanced thermal
treatments, used as final application ablative [25–27] or
carbon–carbon composites [11,28,29]. Few studies [20,30]
present the improvement and characterization of simple
cured phenolic resin without supplementary thermal treat-
ments to obtain further advanced composites. This study
focusses on presenting the basic characteristics of the ‘pre-
cursor’ material used for advanced composites, highlighting
the effect of nanometric SiC on the properties of the cured
phenolic resin. Therefore, the aim of this study represents
the confirmation that once these nanofilled composites are Figure 1. Temperature treatment programme for obtaining phenolic
fully characterized and the optimization of concentration is resin-based composites.
performed, they will be able to serve as matrix for advanced
materials (ablative or carbon–carbon composites) that will be being sonicated for a total of 12 min in 30 s sonication per
achieved in future studies. step, using Bandelin Sonopuls sonde equipment.
This paper presents the obtaining of phenolic resin-based The second stage consists curing process that took place in
nanocomposites, comprising a preliminary study regarding an oven up to 120◦ C. Both simple and nanomodified pheno-
resin properties modification by the addition of nSiC in dif- lic resin samples were cured in Teflon moulds with geometry
ferent contents. The processing technique involves heat cur- and dimensions specific for the mechanical tests. The inner
ing of the simple and nanomodified matrix, the latter being surface of the moulds was cleaned and coated with a silicone
developed via ultrasonication procedure prior to the curing oil [31] as release agent. The silicone oil was observed to be
stage. The obtained nanocomposites were tested in terms effective in decreasing void formation at the bottom surface
of mechanical, tribological and thermal stability under load of the sample during curing process. The mould was dried
properties and characterized by FTIR spectroscopy and SEM in an air oven at 75◦ C for 10 min to eliminate the existing
analyses. The results determine the optimum nanofiller con- water. The moulds were filled with simple and nanomodi-
tent selection for future studies involving the use of this kind fied resin samples and the curing process was performed fol-
of materials as matrix in advanced composites (carbon fibre lowing a controlled temperature schedule, according to the
reinforced nanofilled phenolic resin composites and ablative scheme in figure 1. Methanol solvent in the resin evaporates
materials). around 70◦ C generating gas bubbles in the composite mass.
The several trial sessions revealed that extended dwell time
2. Experimental at 70◦ C supports gas release from the resin solution. After
the curing cycle was completed, the samples were cooled in
2.1 Materials air down to room temperature. The following nomenclature
for the final nanocomposites was used: PR, PR+0.5% nSiC,
Matrix used was resole type phenolic resin ISOPHEN 215 PR+1% nSiC and PR+2% nSiC where 0.5, 1 and 2%
SM 57% provided by ISOVOLTA S.A. Bucharest, with represent the nSiC weight content relative to the phenolic
1.135 g cm−3 density. The nanofiller used was β type nanome- resin (PR).
tric silicon carbide purchased from Nanostructured & Amor-
phous Materials Inc., USA, with the following characteris- 2.3 Testing and characterization
tics: 97.5% purity, 34–40 m2 g−1 specific surface area and
3.22 g cm−3 true density. Nanocomposites were characterized in terms of mechanical,
thermo-mechanical, tribological and morphological proper-
2.2 Obtaining method ties. The phenolic-based nanocomposites were subjected to
spectroscopy analysis using a Spectrometer Nicolet iS50
The nanocomposites obtaining method involved two stages: (operated in ATR mode) and scanning electron microscopy
in the first stage dispersion of the nanometric silicon car- (SEM) using QUANTA INSPECT F microscope with field
bide (nSiC) powder into the phenolic resin (PR) occurred, emission gun and 1.2 nm resolution and energy dispersive
and the second stage was heat-curing process. The first stage X-ray spectrometer (EDS). The materials were subjected to
involved the addition of different contents (0.5, 1 and 2 wt%) mechanical tests by 3-point bending with INSTRON 5982
of nSiC into the resin, mechanical stirring for approximately mechanical testing machine according to SR EN ISO 178,
5 min for bulk homogenization followed by ultrasonication using 2 mm min−1 testing speed and nominal span length
technique for nanofiller optimum dispersion, each sample (16 × specimen thickness), on rectangular specimens
 Influence of nSiC on phenolic resin composites 771

(80 × 10 × 4 mm), using 5 specimens per sample. Tribo- phenolic resin-based composites samples. The control cured
logical tests were performed using CETR UMT 3 (Universal sample was compared with the uncured sample (figure 2)
Macro Materials Tester) block-on ring module, on a 35 mm to evaluate if the curing process was completed, while the
diameter steel role, under 10 N load, for 60 s at two differ- nanofilled samples were compared with the cured unfilled
ent speeds: 1.75 and 2.62 m s−1 , testing 3 specimens for each sample (figure 3) to evaluate the existence of interactions
sample. HDT thermal stability was evaluated using Qualitest between nSiC and phenolic resin resulting in vibrations that
HDT1-heat deflection system, according to SR EN ISO 75, generate peak intensity modification or peak position shift-
testing 3 specimens for each sample using 2◦ C min−1 heat- ing. The used resin is a resole type that contains ether bridges
ing rate and 1.8 MPa flexural stress, in silicone oil immersion (−CH2 −O−CH2 −) that appear at approximately 1015 cm−1
environment. and methylene bridges (−CH2 −) that appear at 2900 and
1470 cm−1 as shown in figure 2. Curing takes place by
heating the resin and temperature processing generates
3. Results and discussion formaldehyde elimination from etheric bonds, so that the
3.1 FTIR spectroscopy final heat-treated cured resin will be based only on methylene
transversal bonds. This is demonstrated by the disappearance
FTIR spectroscopy analyses were performed on the liquid of the peak at 1015 cm−1 in the cured resin, proving that the
uncured phenolic resin sample and on the grinded cured curing was completed.

Figure 2. FTIR spectra of the uncured and cured phenolic resin.

Figure 3. FTIR spectra of simple and nSiC-based phenolic resin samples.

772 George Pelin et al

FTIR spectroscopy analysis were performed on the value increases with nSiC content increase, as illustrated in
grinded cured samples. Figure 3 shows the spectra of the sim- table 1. The position of the peak from ~815 cm1 is shift-
ple and silicon carbide nanofilled phenolic resin. All samples ing towards lower value once with the nSiC content increase.
show the characteristic peaks of the cured resin: OH stretch- It is also worth to mention that strong shift occur at 0.5
ing from resole appears at 3280 cm−1 , CH2 stretching at and 1% nSiC and only marginal shift for the 2% SiC which
2900 cm−1 , −C=C stretching in the aromatic ring results in means that the additional 1% SiC practically does not inter-
peaks in 1600–1470 cm−1 range, CH2 bending vibration is act with the matrix. This effect could be due to the interaction
visualized at 1437 and 1330 cm−1 and C–O stretching at between nSiC and phenolic resin, the intensity increase being
approximately 1200 cm−1 . The three peaks in the 900– a consequence of the higher nSiC content that interacts with
600 cm−1 range are due to the ortho-disubstituted, meta- the matrix. The noise between 1900 and 2300 cm−1 observed
disubstituted and mono-substituted benzenes [31,32]. because of the absorption of the diamond crystal and not due
Nanofilled samples spectra do not present major visible to the absorption of the sample and hence must be ignored.
differences compared to the cured resin. This is due to the
fact that the nSiC characteristic peak, assigned to Si–C vibra- 3.2 SEM analysis
tion, appears at approximately 800–815 cm−1 [33,34], over-
lapping with meta-disubstituted benzene peak from phenolic SEM microscopy analyses were performed in the fracture
resin. But the nSiC interaction with the phenolic resin can be cross-section of the mechanically tested sample. Figure 4
highlighted by the peak shifting from 805 (figure 3 inset) to presents SEM images of the PR-based materials. Figure 4b–d
~815 cm−1 . Evaluating the intensity ratio between the peaks shows nSiC-based samples illustrating the brittle nature of
from approximately 815 cm−1 and one of the main peaks the fracture. The nSiC presence is more visible as the content
of the phenolic resin (1595 cm−1 ), it can be noticed that its increases. Some agglomeration areas can be noticed, but the

Table 1. Intensity ratio between nSiC peak and one of the main peaks in PR.

Sample PR PR+0.5% nSiC PR+1% nSiC PR+2% nSiC

Peak 1 position 1594.76 1594.99 1594.84 1593.99

I1 (peak 1 intensity) 0.0536 0.0145 0.0163 0.0306
Peak 2 position 818.04 817.9 814.77 814.29
I2 (peak 2 intensity) 0.103 0.0292 0.038 0.0833
I2 /I1 1.92 2.01 2.33 2.72

Figure 4. SEM images of (a) nSiC nanopowder and PR-based materials cross-
section of (b) PR+0.5% nSiC; (c) PR+1% nSiC and (d) PR+2% nSiC.
 Influence of nSiC on phenolic resin composites 773

Table 2. Mechanical and thermo-mechanical properties of PR-based materials.

Flexural Flexural Heat deflection

Sample strength (MPa) modulus (GPa) Extension (%) temperature (◦ C)

PR 111.1 ± 8.76 5.3 ± 0.12 2.91 104.2

PR+0.5% nSiC 125.5 ± 13.8 5.8 ± 0.35 2.34 126.7
PR+1% nSiC 148 ± 5.6 6.9 ± 0.15 2.28 132.1
PR+2% nSiC 126.2 ± 7.8 6.1 ± 0.06 2.21 129.7

nanoparticles dimensions can still be distinguished. As nSiC

SEM image (figure 4a) shows the irregular shaped nanoparti-
cles’ diameter is in the range of 35–140 nm. The nanoparticles
average dimension in PR samples is approximately in the
same range showing that the nanoparticles did not form
large aggregates, but they are distributed in the whole mass-
forming areas with high concentration of nanoparticles.

3.3 Mechanical testing

Mechanical testing consisted of 3-point bending tests. Flex-
ural tests were performed both on the simple cured resin
and on the nanofilled resin samples. The test was set to be
carried out until conventional deflection, but all specimens
broke before this value was reached, as phenolic resins are Figure 5. Deflection–temperature curves during HDT test.
brittle materials. Maximum flexural stress represents flexural
strength. The most substantial increase in thermal stability under
For each tested sample, an average of optimum speci- load is presented by the samples with 1 wt% nanofiller as
men replicas was calculated. Comparing flexural strength it can be seen in table 2. In terms of thermal stability under
and modulus results (table 2), it is observed the positive load, adding 0.5 and 2 wt% nSiC resulted in a 22% increase
effect generated by nSiC addition in the phenolic resin. The compared to the simple resin, while the addition of 1 wt%
most substantial increase in strength and modulus is pre- nSiC generated an increase of thermal stability under load
sented by the samples with 1 wt% nanofiller. In terms of flex- greater than 27%.
ural strength, adding 0.5 and 2 wt% nSiC resulted in a 13% These results are in accordance with flexural properties
increase compared to the simple resin, while the addition of obtained from the 3-point bending tests that show that nSiC
1 wt% nSiC-generated an increase of flexural strength greater content increase up to 1% leads to an increase of properties,
than 33%. Regarding flexural modulus, 0.5 wt% nSiC-based while increasing the content up to 2% generates a decrease
samples showed an average value around 10% higher than of strength and stiffness compared with prior content. This
the simple PR, followed by the 2 wt% based samples with an trend can be correlated with FTIR spectroscopy analysis that
increase of 15% and 1 wt% based sample with an increase suggest that the additional nSiC added in 2%-based sam-
of approximately 30%. The obtained results could be due to ples does not interact with the matrix (in comparison with
the fact that the nanoparticles are embedded in the pheno- 0.5 and 1%) and also with the presence of higher degree of
lic resin and uniformly covered, creating a strong mechani- agglomerations in the 2% nSiC-based phenolic resin. These
agglomeration are most likely owing to the higher concentra-
cal interface between the phases. A possible reason could be
tion that generated higher viscosity mixtures, that were more
the fact that the embedded nanoparticles could contribute to
difficult to uniformly disperse and create homogenous mate-
reduce the crack propagation effect in the resin, acting as a
rials, but also to the additional nSiC that did not interact with
reinforcing agent, increasing the PR strength and stiffness.
the matrix. The inhomogeneity in PR+2% nSiC act as stress
Flexural strain average values were in accordance with
concentration sites that negatively affect mechanical proper-
strength and modulus values, all nanofilled samples showed a
ties at room temperature and higher temperature, as they act
decrease of strain compared to the simple resin. Lower strain
as defect points. When a load is applied, in these agglomer-
values generate higher strength and modulus.
ation sites, the load distribution and transfer is significantly
affected, generating premature failure of the materials and
3.4 Thermal stability under load lower strength and stiffness.
Figure 5 illustrates the deflection that the specimens suf-
3.5 Tribological testing
fered during HDT testing, showing the similar trend of the
samples. It can be noticed that nSiC presence enhances the Likewise, tribological tests results highlighted the positive
phenolic resin thermal stability under load. effect of nSiC addition. Figure 6 illustrates the friction
774 George Pelin et al

leading to higher mechanical and thermo-mechanical (heat

deflection temperature) properties. Therefore, the same trend
noticed in both flexural and thermal stabilities under load
tests is most likely owned to the fact that additional nSiC
in 2% samples does not interact with the matrix and also to
the significant agglomerations degree in the 2% nSiC-based
phenolic resin generated by the higher nSiC content that
increased the mixtures viscosity, and consequently hindered
the uniform dispersion and homogeneity of the materials.
nSiC being a popular abrasive material, the tribological
properties improvement was also confirmed, the best result
was obtained for 2% content, but 1 wt% nSiC also deter-
mined substantial friction coefficient increase. Therefore, the
Figure 6. Friction coefficient variation function of nSiC content, optimum nSiC content can be considered for 1 wt% as it
evaluated using 2 speed rate values (1.75 and 2.62 m s−1 ). leads to several properties improvements.
The results represent the preliminary stage experimen-
coefficient average values obtained for simple and nanofilled tal study regarding phenolic resin properties modification
resin. For both speed values, friction coefficient increases with the aim of using it as matrix for advanced composite
when nanocarbide was added. For the 1.75 m s−1 speed materials, such as carbon preform (fibre, fabric, felt)-based
value, friction coefficient increased with the increase of nSiC materials for aerospace applications.
content, up to 1 wt% nSiC, and at higher content (2 wt%),
friction coefficient showed a minor decrease compared to Acknowledgements
1 wt% content. For the 1.75 m s−1 speed value, the friction
coefficient of 0.5, 1 and 2 wt%-based samples increased by This work has been funded by the Sectorial Operational Pro-
71, 99 and 94% compared to the simple resin. gramme Human Resources Development 2007–2013 of the
For 2.62 m s−1 speed, friction coefficient increased uni- Ministry of European Funds through the financial agreement
formly with nSiC content increase, showing 91, 122 and POSDRU/159/1.5/S/134398.
119% higher values than PR.
The 2 wt% nSiC-based samples’ friction coefficient slight References
decrease could be due to small size nonuniformities of
the nanopowders dispersion in the matrix, as 2 wt% con- [1] Knop A and Pilato L A 1985 Phenolic resins (Berlin:
tents present higher agglomeration tendency compared to SpringerVerlag)
lower contents. [2] Mucha H, Kim Y E and Wielage B 2007 Approaches to reveal
Overall, these results show that nanometric silicon carbide porosity in phenolic resin derived CFRP and C/C composites.
can substantially improve tribological performance of phe- International Committee on Composite Materials (ICCM),
nolic resin, supporting them as potential candidates for high 8–13 July 2007, Kyoto, Japan
friction materials to use in abrasive applications. [3] Shuhua Q, Chunlua L and Ying H 2006 Preparation of SiC
hybrid phenolic resin composites. 57th International Astro-
nautical Congress, 2–6 October 2006, Valencia, Spain
4. Conclusions [4] Srebrenkoska V, Bogoeva-Gaceva G and Dimeski D 2009
J. Serb. Chem. Soc. 74 441
The preliminary results presented in this paper highlight [5] Park B, Riedl B, Hsu E W and Shields J 1999 Polymer 40
the positive effect of nSiC addition in phenolic resin 1689
on mechanical, thermo-mechanical and tribological perfor- [6] Lenghaus K, Qiao G G and Solomon D H 2000 Polymer 41
1973
mances, improving their strength, stiffness and abrasive
[7] Okoroafor E U, Villemaire J P and Agassant J F 1992 Polymer
properties. Following a trial established thermal treatment,
33 5264
different PR/nSiC formulations were obtained, character-
[8] Kawamoto A M, Pardini L C, Diniz M F, Lourenço V L,
ized by FTIR spectroscopy and SEM and tested in terms of
Ferreira M and Takahashi K 2010 J. Aerosp. Technol. Manag.
mechanical, thermo-mechanical and tribological properties. 2 169
Results show that the best performance in all conducted tests [9] Gardziella A, Pilato L A and Knop A 2013 Phenolic resins:
are obtained when using 1 wt% nSiC. This content is the opti- Chemistry, applications, standardization, safety and ecology
mum as it leads to a solid embedding of the nanoparticles into (Berlin, Germany: Springer Science & Business Media) p 566
the resin, as they interact completely with the matrix, as FTIR [10] Chi-Chang G and Tao L 2013 Phosphorus modified phe-
analysis shows, without generating high agglomeration areas nolic resin and preparation method thereof. Patent no. CN
that form at higher nSiC content (2 wt%) as shown by SEM 103467685 A, China
analysis. This allows the nanoparticles to act as a reinforc- [11] Naderi A, Mazinani S, Ahmadi S J, Sohrabian M and Arasteh R
ing agent, probably by reducing crack propagation effect and 2014 J. Therm. Anal. Calorim. 117 393
 Influence of nSiC on phenolic resin composites 775

[12] Liu C, Li K, Hejun Li, Zhang S and Zhang Y 2014 Polym. [23] Nakada M 1994 Tribol. Int. 27 3
Degrad. Stab. 102 180 [24] Balaji R, Sasikumar M and Elayaperumal A 2014 Inter. J. Res.
[13] NPCS Board of Consultants & Engineers 2007 Phenolic Eng. Technol. 3 609
resins technology handbook (Niir Project Consultancy [25] Srikanth I, Daniel A, Kumar S, Padmavathi N, Singh V,
Services, India) p 260 Ghosal P, Kumara A and Rohini Devi G 2010 Scripta Mater.
[14] Tate J S, Gaikwad S, Theodoropoulou N, Trevino E and Koo 63 200
J H 2013 J. Compos. (online) Article ID 403656 [26] Natali M, Monti M, Puglia D, Kenny J M and Torre L 2012
[15] Wang Z J, Kwon D J, Gu G Y, Lee W I, Park J K, Devries Compos. Part A – Appl. S 43 174
K L and Park J M 2011 Dispersion evaluation and interfa- [27] Wang Z J, Kwon D J, Gu G Y, Lee W, Park J K, Devries K L
cial sensing of carbon fiber/Cnt-phenolic composites using and Park J M 2014 Compos. Part B: Eng. 60 597
electro-micromechanical technique. 18th International Con-
[28] Manocha L M 2003 Sadhana 28 349
ference on Composite Materials, 21–26 August 2011, Jeju
Island, South Korea [29] Bacos M P 1993 J. de Phys. IV 3 1895
[16] Gau C, Chen S-Y, Tsai H L, Jenq S-T, Lee C-C, Chen Y-D [30] Taheri-Behrooz F, Memar Maher B and Shokrieh M M 2015
and Chien T H 2009 Jpn. J. Appl. Phys. (online) 48 06FF10-1-4 Comp. Mater. Sci. 96 411
[17] Patton R D, Pittman Jr C U, Wang L and Hill J R 1999 [31] Parameswaran P S 2010 Modification of phenol formalde-
Compos. Part A – Appl. S 30 1081 hyde resin for improved mechanical properties (PhD The-
[18] Bafekrpour E, Simon G P, Habsudab J, Naebea M, Yang C sis). Department of Polymer Science & Rubber Technology,
and Fox B 2012 Mat. Sci. Eng. A 545 123 Cochin University of Science and Technology
[19] Koo J H, Natali M, Tate J S and Allcorn E 2013 Int. J. Energ. [32] Perez J M, Echeverria J M, Oliet M, Alonso M V and
Mater. Chem. Propuls. 12 119 Rodriguez F 2007 BioResources 2 270
[20] Bansal D, Pillay S and Vaidya U 2013 J. Reinforce. Plast. [33] Janz S 2006 Amorphous silicon carbide for photovoltaic
Compos. 32 955 applications (Masters Thesis). Fakultät für Physik Universität
[21] Kang P H and Yang H S 1998 Korean. J. Chem. Eng. 15 580 Konstanz
[22] Atarian M, Salehi H R, Atarian M and Shokuhfar A 2012 Iran. [34] Prabhakaran P V, Sreejith K J, Swaminathan B, Packirisamy S
Polym. J. 21 297 and Ninan K N 2008 J. Mater. Sci. (online) 44 528