52 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
ISSN 2277 – 7164
Original Article
Studies on the effect of phenolic resin modifications on the adhesive
properties of epoxy resin
Najuma Abdul Razack*, Lity Alen Varghese
Department of Chemical Engineering
National Institute of Technology, Calicut-673601, Kerala, India.
*E-mail: [email protected].
Received 30 October 2014; accepted 25 November 2014 Abstract
In the present study, various modifications for epoxy resin system were done for metal to metal bonding. Epoxy adhesives
were synthesized from epichlorohydrin and bisphenol-A and the mole ratio of reactants (BPA/ECH) were optimized to be
1:3 from various shear and peel strength experimentations. In the first part modification was done by blending epoxy with
phenol formaldehyde(PF) resole resin and later, the epoxy was further modified by replacing it partially with
hyperbranched structures like epoxidised phenolic novolacs (EPN) and epoxidised cresol novolacs (ECN) . Phenol and
cresol novolacs were prepared by the reaction of phenol and cresol respectively with formaldehyde. The results indicated
that ECN/PF blends gave better strength properties when compared to EPN/PF and Epoxy/PF blends. Morphological
analysis using SEM gives an idea on the energy absorption during failure for the Epoxy/PF blend.
© 2014 Universal Research Publications. All rights reserved
Key Words: adhesive, epoxy, novolac, blends, hyperbranched.
1. Introduction
Epoxies are probably the most versatile family of adhesives
because they bond well to many substrates and can be
easily modified to achieve widely varying properties
[1].The most common epoxies used in adhesives are
derived from bisphenol A and epichlorohydrin and are
usually cured with reactive hardeners containing primary
and/or secondary amine groups[2]. In general, epoxy resins
are prepared by the reaction of compounds containing an
active hydrogen group with epichlorohydrin followed by
dehydrohalogenation. Epoxies dominate the field of
structural adhesives due to their better wetting ability,
excellent mechanical properties, high chemical and thermal
resistance [3].Because of the normally brittle nature, epoxy
adhesives have been toughened with many different resins
including thermoplastic particles, nylon and various
elastomers [4].Incorporation of thermoplastic poly(vinyl
butyral) into an epoxy- novolak combination was found to
enhance adhesive shear and peel strengths [5].
Combinations of epoxy and phenolic resins
provide structural adhesives with superior high temperature
resistance. Both novolac and resole type phenolics may be
used. The blends cure through reaction of the epoxy groups
with the phenolic hydroxyl groups. When resoles are used,
the epoxy may also react with the methylol groups.
Adhesives based on epoxy-phenolic blends are good for
continuous high temperature service in the 3500F range or
intermittent service as high as 5000F. They retain their
properties over a very high temperature range [6].
Resistance to weathering, oil, solvents and moisture is very
good. Because of the rigid nature of the constituent
materials, epoxy-phenolic adhesives have low peel and
impact strength and limited thermal-shock resistance.
These were developed primarily for bonding metal joints in
high temperature applications. Usually the phenolics used
are the resole type and often the epoxy is a minor
component [7].
Epoxidised phenol novolac resins and cresol
novolac resins are made by glycidylation of phenol/cresol
formaldehyde condensates (novolacs) obtained from acid
catalysed condensation of phenol/cresol and formaldehyde.
Epoxidation of novolacs with an excess of epichlorohydrin
minimizes the reaction of phenolic hydroxyl groups with
glycidylated phenol groups and prevents branching. An
increase in the molecular weight of the novolac increases
the functionality of the resin. EPN resins range from a high
viscosity liquid to a solid. The major applications of epoxy
novolac resins have been in heat resistant structural
laminates, chemical resistant filament wound pipes and
high temperature adhesives. The epoxy functionality is
between 2.2 and 3.8. ECN resins derived from o-cresol
novolacs have even higher functionalities (2.5 to 6).
Available online at http://www.urpjournals.com
Advances in Polymer Science and Technology: An International Journal
Universal Research Publications. All rights reserved
53 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
Novolac epoxy resin, being multifunctional, can
produce a more tightly crosslinked three dimensional
network and hence give better adhesive strength. Also, they
combine the reactivity of the epoxy group and the thermal
resistance of the phenolic backbone.In this paper, we are
reporting the adhesive properties of epoxy resin, EPN and
ECN in blend with phenol formaldehyde for metal to metal
(Al-Al) bonding.
2. Materials and methods
2.1. Materials
Phenol, m-cresol, formaldehyde (40% solution),
sodium hydroxide (All from Merck India Ltd.),
Epichlorohydrin (Sisco Research laboratories Pvt. Ltd.,
Mumbai, India) and Bisphenol A (Loba Chemie Pvt. Ltd.,
Mumbai, India) was used in this investigation.
2.2. Synthesis of Epoxy resin
Bisphenol A (1 mole) was dissolved in a mixture
of an excess of epichlorohydrin (6 moles) and 50 cc water
in a one litre three necked flask. The flask was equipped
with a mechanical stirrer, thermometer and a Liebig's
condenser. The mixture was heated gently over a water
bath till the epichlorohydrin began to boil. Heating was
stopped and caustic soda (2 moles) was added in two
pellets at a time down the condenser. The reaction was
allowed to subside before more alkali was added. When all
the caustic soda pellets had been added, the reaction
mixture was heated strongly for 45 minutes. Heating was
stopped as the reaction mixture turned viscous. The excess
epichlorohydrin was removed by vacuum distillation. The
remaining mixture was extracted with benzene to
precipitate sodium chloride which was removed by
filtration under vacuum. The filtrate was distilled in
vacuum to remove benzene and dried in vacuum for about
3hours.
2.3. Synthesis of novolac resins
The novolacs were prepared by reacting phenol
with formaldehyde in the molar ratio 1:0.8 in presence of
oxalic acid catalyst in a 3-necked flask fitted with a
mechanical stirrer, water condenser and thermometer. The
reaction mixture was heated and allowed to reflux at about
100°C for 2-3 hours. When the resin is separated from the
aqueous phase the reaction was stopped. The resin was
neutralised with sodium hydroxide, filtered, washed with
water and vacuum dried. The novolac resin contains 4-6
benzene rings per molecule. The same procedure was used
to synthesise novolac resin from m-cresol (ECN) [8].
2.4. Synthesis of Epoxidised phenolic novolacs (EPNs)
1 mole of the novolac resin (1:0.8) was dissolved
in 6 moles of epichlorohydrin and the mixture heated in a
boiling water bath. The reaction mixture was stirred
continuously for 4 hours while 1mole of sodium hydroxide
in the form of 30 %aqueous solution was added drop wise.
The rate of addition was maintained such that the reaction
mixture remains at a pH insufficient to colour
phenolphthalein. The resulting organic layer was separated
and then fractionally distilled under vacuum. Epoxidised
novolac resin was similarly prepared from m-cresol
novolac (ECN) using the same procedure.
2.5. Synthesis of Phenol formaldehyde (PF) resole resins
The resole resin were prepared by reacting phenol
with formaldehyde in the molar ratio 1:1.7 in presence of
33% NaOH solution in a 3-necked flask fitted with a
mechanical stirrer, condenser and thermometer. The
reaction mixture was heated and allowed to reflux at about
100°C for 1 hour. The resin was further neutralised with
oxalic acid in minimum amount of water, washed with
water and vacuum dried.
2.6. Bonding and adhesive performance
Aluminium strips of size 100x25 mm were
machined from 0.8 mm thick sheets to serve as metal
substrates for peel studies on metal-to-metal bonds. Strips
of 100x25x1 mm were used for shear strength. Surface
preparation is necessary before the application of adhesive.
Solvent degreasing was done with trichloroethylene, to
remove dust and traces of oily impurities on the surface.
Following the solvent wiping, the surface was abraded with
emery paper (P 100) and is again wiped with solvent to
remove completely the debris of abrasion. An even contact
pressure is applied throughout the joining surface after the
application of the adhesive and it is then kept in the oven at
the required temperature for the specified time.
Peel strength and lap shear strength of metal-to-
metal specimens were determined on a universal testing
machine as per ASTM D903 [9] and ASTM D1002 [10]
respectively with a grip separation rate of 50 mm/min at
room temperature.
2.7. Formulation of adhesive blends
Epoxy resin were synthesized in varying molar
ratios of Epichlorohydrin(ECH) to Bisphenol-A(BPA) viz,
1.57:1, 3:1, 3.5:1 and 4:1 and the mole ratio were optimised
based on lap shear and peel strength experimentations. The
epoxy was further modified by replacing it with epoxy/PF
blend in varying ratios. The same procedure is repeated for
EPN/PF and ECN/PF. All the blends were tested for lap
shear and peel strength in the Universal testing machine.
3. Resin Characterisation
3.1. Epoxide Equivalent weight (EEW)
The epoxy content of epoxy resins is an important variable
in determining their reactivity and the properties of
coatings made from them. Knowing the EEW, the required
amount of the crosslinking agent can be calculated [11].
EEW is usually defined as the weight of resin containing
one gram equivalent of epoxide. The epoxy content of
liquid resins, is frequently expressed as weight per epoxide
(wpe) or equivalent/kg of the resin. A common method of
analysis of epoxide content of liquid resins involves the
opening of the epoxy ring by hydrogen halides
(hydrohalogenation) [12]. Weight per epoxide values of the
synthesised epoxy, EPN and ECN resins were determined
by the pyridinium chloride method as per ASTM D 1652-
73 [13].
0.1 to 0.2 gm of the epoxy resin was mixed with
2ml HCl in 25 ml pyridine. The mixture was heated to
reflux on a water bath for 45 minutes. The solution was
cooled to room temperature and the unreacted acid present
was estimated by back titration with standard NaOH
solution (0.1 N) using Phenolphthalein indicator. A blank
was also carried out under same reaction conditions.
Epoxide equivalent = N x V/w, where N is the
strength of alkali, V is the volume of alkali used up and w
54 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
is the weight of the resin. Epoxide equivalent can be
obtained as eq/kg from which wpe value of the resin can be
calculated.
3.2. Gel time determination of blends
This method covers the determination of the time from the
initial mixing of the reactants of a thermoset composition to
the time taken when soldification commences, under
conditions approximating the conditions of use.
About 2 gm of the sample is filled into a test tube
and dipped into a water bath at 1000C. The increase in
viscosity is monitored with a glass rod used roughly like a
reciprocating piston. As soon as the gel point is reached the
glass rod gets trapped in the resin. This period of time is
recorded as the gel time. The test is done as per DIN 16945
[14].
3.3. Spectroscopic analysis Fourier transform infra-red (FTIR) spectra are generated by
the absorption of electromagnetic radiation in the frequency
range 400 to 4000 cm-1 by organic molecules. Different
functional groups and structural features in the molecule
absorb at characteristic frequencies. The frequency and
intensity of absorption are indicative of the bond strengths
and structural geometry in the molecule. FTIR spectra of
the samples were recorded on a Shimadzu, IR Prestige
21Spectrometer by KBr pellet method. Nuclear magnetic
resonance (NMR) spectroscopic method has been used to
good effect by authors investigating the structure of phenol-
formaldehyde resins [15]. Here 1HNMR spectroscopy is
used for the characterization of the synthesized
resins(Epoxy, EPN, ECN and PF).The spectra were
recorded on a 500MHz Bruker Avance DPX spectrometer
using acetone as solvent.
3.4. Gel Permeation Chromatography (GPC)
GPC, also called size exclusion chromatography, makes
use of a chromatographic column filled with a gel or porous
solid beads having a pore size similar to that of the polymer
molecules [16]. A dilute solution of the polymer is
introduced into a solvent stream flowing through the
column. Smaller molecules of the polymer will enter the
beads while the larger ones will pass on. Thus the larger
molecules have a shorter retention time than the smaller
molecules.
The synthesized epoxy resin was subjected to
GPC analysis with a view to estimate the molecular weight
of the resin. The analysis was carried out using Waters
GPC equipped with an autoinjector with an injection
volume of 50 µl and Tetrahydrofuran (THF) as the eluent.
3.5. Morphological studies
Scanning electron microscope (SEM) is a very useful tool
in polymer research for studying morphology and
microstructure [17]. Hitachi SU6600 Variable pressure
field emission SEM (FESEM) was used to investigate the
morphology of the failed adhesive surfaces. In this
technique, an electron beam is scanned across the specimen
resulting in back scattering of electrons of high energy,
secondary electrons of low energy and X-rays. These
signals are monitored by detectors (photo multiplier tube)
and magnified. An image of the investigated microscopic
region of the specimen is thus observed in a cathode ray
tube and photographed using black and white film. The
SEM observations reported in the present study were made
on the fracture surfaces of failed peel specimens.
4. Results and discussion
4.1. Mechanical properties
The average lap shear and peel strengths of epoxy
resins synthesized in varying molar ratios of BPA/ECH is
given in table 1.
Table 1 Variation of lap shear and peel strengths with
BPA/ECH molar ratio
Sample
(BPA/ECH)
Average lap
shear strength
(Nmm-2)
Average Peel
strength
(Nmm-1)
1: 1.57 - -
1: 3 4.402 0.550
1: 3.5 1.412 0.513
1: 4 0.627 0.497
In the case of 1:1.57, the glycidyl ether formation is less
than 10% and the resin formed is in solid form which is
uncoatable. On increasing the ratio beyond 1:4, the
viscosity of the resin formed was found to decrease
considerably. Consequently, the more viscous liquid
epoxides are usually preferred to the less viscous members
of the series for adhesive applications[18].According to the
experimental data, a relatively good adhesion property is
achieved at the molar ratio of BPA/ECH 1:3 and hence we
optimised the resin for further studies. This epoxy is further
blended with PF resin. The PF content in the blend is varied
from 10 phr to 100 phr. Similarly EPN and ECN resin
systems were blended with PF in varying ratios from 10 phr
to 100 phr. The variations in lap shear strengths and peel
strengths of these systems can be observed from Figure 1
and Figure 2 respectively.
From the plot, it was observed that the lap shear
and peel strengths of almost all the systems increased with
PF content. The results indicated that ECN/PF blends gave
better strength characteristics when compared to EPN/PF
and Epoxy/PF. Phenol formaldehyde, due to its brittle
nature, increases the cross linking sites in the system,
thereby increasing the flexibility and lap shear strength. m-
cresol, has 3 reactive positions, which can react with PF, to
give a cross linked structure, resulting in an increase in
strength.
Figure 1: Variations of Lap shear strengths of various
systems.
55 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
Figure 2: Variations of Peel strengths of various systems.
4.2. Resin Characterisation
4.2.1. Epoxide Equivalent weight (EEW)
The epoxide equivalents of the synthesized resins were
determined by the titration method. The epoxy equivalents
for Epoxy, EPN and ECN resins were found to be 4.2, 2.6
and 3.1 eq/kg respectively. Epoxidised cresol novolac was
found to have greater epoxide content than epoxidised
phenol novolac. The high epoxide functionality, results in a
higher density of crosslinks.
4.2.2. Gel time determination of blends
From the experiment, it was found that the gel time for
Epoxy/PF blend was 40 min, while the EPN/PF and
ECN/PF blends were stable after 1 hr. ie, the gel time was
found to increase when the epoxy was replaced with
novolac resin in the blend. Also, it is evident that Epoxy/PF
blend have a short gel time and often a short shelf life
compared to the other blends. This is attributed to the low
reactivity of novolac based blends.
4.2.3. Spectroscopic data
The FTIR spectrum of epoxy resin is given in Figure3. The
C-H stretching in epoxies is at 2964cm-1.Symmetrical
stretching or ring breathing frequency is observed at
1242cm-1and this is characteristic of the epoxy ring. The
band at 915 cm-1(asymmetric ring stretching in which C-C
stretches during contraction of C-O bond), 829cm-1and 750
cm-1are typical of the epoxy group [19].
Figure 3: FTIR spectra of Epoxy resin.
In Figure 4, which shows the FTIR of PF resin,
characteristic absorptions were found around 3250 cm-1
(hydroxyl group, broad band), 1504cm-1(phenyl ring) and
1479cm-1(CH2 bending).The bands around 1220cm-1 and
1019cm-1are characteristic of C-O stretching in phenol and
methylol groups respectively. The bands at 881.48cm-1 and
831.10 cm-1 are characteristic of ortho-para substitution.
Figure 4: FTIR spectra of PF resin.
Figure 5: FTIR spectra of EPN resin.
The Figure 5 shows the FTIR spectra of
synthesised EPN resin. Similar to Figure 3, here also we
can see peaks which show the epoxide functionality.The
band at 1242 cm-1 denotes symmetrical C-O stretching
(ring breathing frequency) in epoxides. Further, the bands
at 2931 cm-1, 815cm-1 and 748 cm-1 are also typical of
epoxides. The less intense broad band at 3250-3500 cm-1
range indicates the phenolic hydroxyl group, which also
has a considerable involvement in epoxidation. FTIR
spectra of ECN in Figure 6 also shows peaks similar to
EPN, which confirms the epoxide functionality.
Figures 7, 8, 9 and 10 shows the 1HNMR spectra
of Epoxy, PF, EPN and ECN resins respectively. In the
proton NMR for PF resin, the spectra displayed peaks
of aromatic protons (δ 7.2), methoxy protons (δ 3.7),
-CH=CH- protons (δ 5.35), -CH2-Ar protons (δ 2.55) and
-CH2-CH=CH- protons (δ 2.01).The proton NMR of EPN,
56 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
Figure 6: FTIR spectra of ECN resin.
Figure 7: NMR spectra of Epoxy resin.
+
Figure 8: NMR spectra of PF resin.
57 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
Figure 9: NMR spectra of EPN resin.
Figure 10: NMR spectra of ECN resin.
Figure 11: Gel Permeation Chromatogram of Epoxy resin.
58 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
Figure 12: Scanning electron micrograph of Epoxy/PF blend.
displayed strong peaks of Aryl protons (δ 2.1,δ 2.45) and
–CH3-C (δ 0.98). Peaks for aromatic protons (δ 6.8, δ 7.3)
and –CH2 protons (δ 3.7) are also evident.
In proton NMR of ECN, the doublet signal at 6.8 ppm is for
aromatic protons. The spectra also displayed peaks of Aryl
protons (δ 2.3), CH-X protons (δ 2.9), Alkenyl H (δ 4.1)
and -CH3-C protons (δ 0.95). The weak singlet at 3.8 ppm
is for –CH2 protons.
4.2.4. Gel Permeation chromatography This is a reliable, precise, and fast method to measure the
molar mass averages, the polydispersity index, PDI and the
complete molar mass distribution of polymers. Figure 11
shows the Gel Permeation Chromatogram of the
synthesized epoxy resin. The observed number average
molecular weight and the polydispersity index were found
to be 400 and 2.375 respectively.
4.2.5. Morphological studies
Fracture surfaces of the failed adhesive bonds were
subjected to Scanning electron microscopy (SEM) to
observe morphological features.
Figure 12 is a view of the fracture surface of an
Epoxy/PF blend after the peel test. A deepening of the
crevices is seen in the figure. This is an indication of
greater energy absorption or toughness. The striations on
the surface caused by layer by layer failure as well as
more prominent valleys and peaks indicate a more
emphatic cohesive failure of the substrate.
5. Conclusion
Adhesive behaviour of different epoxy/phenolic
blends has been investigated by lap shear and peel test
using aluminium substrates. It is observed from the results
that, the modifications by phenolic resin systems are very
effective for improving the strength properties of epoxy
resins. The 3:1 ratio of ECH/BPA is found to give optimum
lap shear and peel strengths. The strengths of various
systems were found to increase by increasing the
incorporation of PF. It is also observed that ECN blending
gave better mechanical strength properties compared to
others. Further modifications for this system to obtain
comparable shear strengths with the other systems are
possible by trying with various adhesion promoters. The
fracture mechanisms were also studied using FESEM.
References
1. E. M. Petrie, Handbook of adhesives and Sealants.
McGraw Hill., New York, 2000, pp.355-373.
2. Manilal Savla, Epoxy resin adhesives, in: Handbook of
Adhesives, I. Skiest (Ed.), Van Nostrand Reinhold
Company, New York, 1977, pp.434-445.
3. Silvia G. Prolongo, Gilberto del Rosario and Alejandro
Urena, Comparative study on the adhesive properties
of different epoxy resins, Int. J. Adhes. Adhes. 26
(2006) 125-132.
4. Sun.Luyi, L.G.Warren, D. Davis, and H.J.Sue, Nylon
Toughened Epoxy/SWCNT Composites, J. Mater. Sci.
46 (2011) 207-214.
5. A. Salimi, H. Omidian and M. J. Zohuriaan-Mehr,
Mechanical and thermal behavior of modified epoxy–
novolak film adhesives, J. Adhes. Sci. Technol., 17
(2003) 1847-1861.
6. Lity Alen Varghese, Beena T Abraham and Eby
Thomas Thachil, Adhesive properties of neoprene-
phenolic-epn blends, Progress in Rubber, Plastics and
recycling technology. 22 (2006) 225-240.
7. E.M. Petrie, Epoxy hybrid adhesives, in: Special Chem
Adhesives and Sealants (2005).
8. K.P. Unnikrishnan, Studies on the toughening of epoxy
resins. Ph.D thesis, Cochin University of Science and
Technology (2006).
9. Annual Book of ASTM standards: D 903-98.
10. Annual Book of ASTM standards: D 1002-05.
11. G. Bianchini , R. Dyer, A. Fream, P. Heffer, S.P.
McOwan and P. Oldring, Waterborne and solvent
based epoxies and their end user application. Wiley,
New York, 1997.
12. H. R. Kricheldorf, O. Nuyken and G. Swift, Handbook
of Polymer Synthesis.Marcel Dekker, NewYork, 2005.
13. Annual Book of ASTM standards: D 1652-73.
59 Advances in Polymer Science and Technology: An International Journal 2014; 4(4) : 52-59
14. DIN 16945 standard.
15. N. A. Novikov, A. P. Mozoleva, A.V. Uvarov, S. G.
Alekseyeva and I. Y.A. Slonim, Analysis of the 13C
and 1H NMR spectra of phenol-formaldehyde resole
oligomers, Polymer Science. 28 (1986) 1187-1196.
16. V. R. Gowariker, N. V. Viswanathan and Jayadev
Sreedhar, Polymer Science, First ed., New age
international Publishers, New Delhi, 1986.
17. L. Engel, H. Klinglee, G.W. Ehrenstein and H.
Schaper, An atlas of polymer damage, Wolfe Science
Books, Munich, 1981.
18. Manilal Savla, Epoxy resin adhesives, in: Handbook of
Adhesives, I. Skiest (Ed.), Van Nostrand Reinhold
Company, New York, 1977, pp.434-445.
19. R.M. Silverstein and F. X. Webster, Spectroscopic
identification of organic compounds. John Wiley and
Sons Inc., New York, 2002.
Source of support: Nil; Conflict of interest: None declared