Designed Monomers and Polymers

ISSN: (Print) 1568-5551 (Online) Journal homepage: https://www.tandfonline.com/loi/tdmp20

Toughening of epoxy resins

K. P. Unnikrishnan & Eby Thomas Thachil

To cite this article: K. P. Unnikrishnan & Eby Thomas Thachil (2006) Toughening of epoxy resins,
Designed Monomers and Polymers, 9:2, 129-152, DOI: 10.1163/156855506776382664

To link to this article: https://doi.org/10.1163/156855506776382664

Published online: 02 Apr 2012.

Submit your article to this journal

Article views: 4654

View related articles

Citing articles: 105 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tdmp20
Designed Monomers and Polymers, Vol. 9, No. 2, pp. 129– 152 (2006)
 VSP 2006.
Also available online - www.vsppub.com

Review

Toughening of epoxy resins

K. P. UNNIKRISHNAN 1 and EBY THOMAS THACHIL 2,∗

1 Department of Chemistry, U.C. College, Aluva 2, India
2 Department of Polymer Science and Rubber Technology, Cochin University of Science and
Technology, Kochi 682 022, India

Abstract—Epoxy resins constitute an important class of thermosets which are used extensively in
the field of composites, adhesives surface coatings, etc. They are of aliphatic, cycloaliphatic or
aromatic backbones. Epoxy resins based on bisphenol A are commercially available and they have
good thermal and mechanical properties. They are cured by a variety of curatives, such as amines and
anhydrides, and the mechanism of curing largely depends on the type of curing agent used. They show
comparatively low cure shrinkage. Cured epoxies are brittle with poor resistance to crack initiation
and growth. Their mechanical properties, in particular toughness, can be modified by incorporating
a rubbery phase into the resin matrix. The epoxy matrix can also be modified by various other
methods such as incorporation of thermoplastics and particulate fillers. The toughening process leads
to improvement in fracture toughness and impact resistance. This review examines the various options
and the state-of-the-art vis-à-vis epoxy modification.

Keywords: Epoxy resin; modification; impact modification; toughening; bisphenol-A resins; phase
morphology; elastomer modification.

1. INTRODUCTION

Epoxies are versatile from the view point of chemical structure, as well as physical
and mechanical properties. They have been and will continue to be among the
mainstream plastic materials for applications like coatings, adhesives, laminates and
structural components. However, when the application is structural, unfortunately
epoxy resins suffer other engineering polymers; they are either brittle or notch
sensitive or both. As a result, considerable effort has been focused on toughness
improvement of epoxy resins [1].

∗ To whom correspondence should be addressed. E-mail: [email protected]

130 K. P. Unnikrishnan and E. Th. Thachil

2. EPOXY RESINS
Epoxy resins (also called epoxide resins and occasionally ethoxyline resins) are
characterised by the presence of more than one 1,2-epoxy group per molecule. This
group may lie within the body of the molecule but is usually terminal. The three-
membered epoxy ring is reactive to many substrates, thus leading to chain extension
and/or cross-linking through a rearrangement polymerisation process [2].
There are a large number of epoxy resins in use. The non-epoxy part of the
molecule may be aliphatic, cyclo-aliphatic or aromatic hydrocarbon. It may be
non-hydrocarbon and polar also. Treatment with curing agents gives insoluble
and intractable thermosets. To facilitate processing and modify resin properties
other constituents may be included in the compositions, e.g., solvents, diluents,
plasticizers and accelerators. High chemical and corrosion resistance, good thermal
and mechanical properties, outstanding adhesion to various substrates, flexibility,
low cure shrinkage, good electrical properties and ability to be processed under a
variety of conditions are the characteristic features of epoxy resins. The publication
of a German patent (No. 676117) in 1939 pointed to the commercial interest in
epoxy resins. Epoxy resins are used in protective coatings, structural applications
such as laminates and composites, tooling, construction bonding and adhesives apart
from several other less frequent applications.

2.1. Synthesis of epoxy resins

The first and still the most important class of commercial epoxy resins is the reaction
product of bisphenol A and epichlorohydrin in the presence of sodium hydroxide
(Scheme 1). It is called the diglycidyl ether of bisphenol A (DGEBA). An excess of
epichlorohydrin is commonly used to check the formation of high-molecular-weight
polymer. Many of the commercial liquid resins consist of the low-molecular-weight
DGEBA (I) with small amounts of high-molecular-weight polymers.

Scheme 1.
 Toughening of epoxy resins 131

Scheme 2.

Many of the liquid glycidyl ether resins have an average molecular weight in the
range 340–400. High-molecular-weight resins are formed by reducing the amount
of epichlorohydrin and by keeping the medium strongly alkaline. High-molecular-
weight resin (II, Scheme 2) is prepared by the chain-extension reaction of liquid
resin (crude diglycidyl ether of bisphenol A) with bisphenol A in presence of alkali
[3]. This is referred to as the ‘advancement process’ or ‘fusion process’ and is
widely used in commercial practice. Conventional advancement catalysts include
basic inorganic reagents such as hydroxides of sodium, potassium and lithium, soda
ash, amines and quaternary ammonium salts.

2.2. Types of epoxy resins

Depending on the chemical structure of the curing agent and conditions of curing, it
is possible to obtain toughness, chemical resistance, mechanical properties ranging
from extreme flexibility to high strength and hardness, high abrasive strength, good
heat resistance and high electrical resistance. Epoxy resins fall into different types
based on their structure and applications. In addition to DGEBA-type resins, there
are various other types of resins with epoxy functionality.

2.2.1. Miscellaneous glycidyl ether resins. Glycidyl ether resins are prepared
from polyhydric compounds and epichlorohydrin. Bisphenol F and bisphenol H are
used extensively. They have lower viscosity than the corresponding bisphenol A
resins.
Phenoxy resins are thermoplastic polymers derived from bisphenols and epichloro-
hydrin. They have the same repeating units as advanced epoxy resins and are clas-
sified as polyols or polyhydric ethers.
Epoxidised phenol novolac resins (EPN and ECN) resins are made by glycidyla-
tion of phenol/cresol-formaldehyde condensates (novolacs).This gives random or-
tho- and para-methylene bridges.
132 K. P. Unnikrishnan and E. Th. Thachil

2.2.2. Aromatic glycidyl amine resins. They are formed from aliphatic or
aromatic primary or secondary amines and are reactive to anhydride hardening
agents. Glycidyl amines are reported to provide stable products if the amine is
aromatic or, if aliphatic, the alkyl substituent has more than four carbon atoms.

2.2.3. Non-glycidyl ether epoxides. Non-glycidyl ether epoxides are of two

types: those with ring structures, as well as an epoxide group in the molecule (called
cyclic aliphatic resins) and those having linear structure on to which are attached
epoxide groups (called acyclic aliphatic epoxy resins). Vinyl cyclohexene dioxide,
dicyclopentadiene dioxide, etc., are commercially available resins of this type.
Among acyclic aliphatic resins, three types of resins can be identified; epoxidised
diene polymers, epoxidised oils and polyglycol diepoxides. Epoxidised drying oils
have been available for many years as stabilisers for polyvinyl chloride. Polyglycol
diepoxides are used as flexibilisers in commercial epoxy resins.

2.3. Characterisation
Uncured epoxy resins are mainly characterised by epoxy content, colour density,
hydrolysable chlorine and volatile content. The epoxy content of liquid resins is
expressed in epoxide equivalent or weight per epoxy (wpe) which is defined as the
weight of the resin containing 1 g equivalent of epoxide. A common method for
the analysis of the epoxy content of liquid and solid resins is either by using HBr
and acetic acid or by the pyridinium chloride method (ASTM D-1652–73 (1979)).
Typical commercial grade liquid epoxy resins have an average molecular weight of
about 370. Viscosity is 11–15 Pa·s at 25◦ C and weight per epoxide (wpe) is about
188. The epoxide functionality can be detected by spectroscopic methods such
as FT-IR and NMR. Thermal properties can be determined by thermo gravimetric
analysis (TGA) and differential thermal analysis (DTA). The curing behaviour and
glass transition can be studied by DSC and DMTA analysis.

2.4. Cross-linking
The resin has to be cross-linked into a three-dimensional insoluble and infusible
network. For this the choice of cross-linking agent depends upon the processing
method and conditions and the properties required [4, 5].
Two types of curing agents are widely used: catalytic systems and polyfunctional
cross-linking agents that link the epoxide resin molecules together (co-reactive
systems). Some systems may involve both catalytic and condensation systems. A
catalytic curing agent functions as an initiator for the resin homo-polymerisation
process.
Amine curing agents can cross-link epoxy resin either by a catalytic mechanism or
by bridging across epoxy molecules. In general, primary and secondary amines act
as co-reactive hardeners while tertiary amines are essentially catalytic. As a class
primary aliphatic amines form fast curing agents at room temperature. Primary and
 Toughening of epoxy resins 133

Table 1.
Commercial amine curing agents

Formula Name Abbreviation

NH2 CH2 CH2 NHCH2 CH2 NH2 Diethylenetriamine DETA

NH2 CH2 CH2 NHCH2 CH2 NH2 NHCH2 CH2 NH2 Triethylenetetramine TETA

Poly(oxypropylene) diamine

1,2-diamino cyclohexane DAC

N-amino ethyl piperazine AEP

Methylene dianiline MDA

4,4 -diaminodiphenylsulfone DDS

m-phenylene diamine MPDA

secondary aliphatic polyamines give good room temperature cures with DGEBA-
type resins. Modified aliphatic polyamines are better curing agents to achieve
properties such as lower toxicity, lower exotherms in large castings, improved
flexibility and longer pot life. Cycloaliphatic amines are found to increase pot life.
Piperidine is fast reacting and allows beneficial effects in rubber modification of
epoxy resins [6]. N-aminoethyl piperazine is fast curing and gives flexibility and
longer pot life. Aromatic polyamines are slower to react than aliphatic polyamines
and require high temperature cure. The nature of the curing agent used can influence
the final properties of the modified epoxy in several ways [7].
Compared to amine cured systems, acid hardening systems give lower exotherms
upon cure and are less skin sensitive. In practice, acid anhydrides are preferred to
acids since the latter release more water upon curing which leads to foaming of the
product. Both esterification and etherification occur. Curing reaction by anhydrides
is catalysed by tertiary amine by a zwitterion mechanism [8]. Epoxy–anhydride
systems show low viscosity, long pot life and little shrinkage when cured at high
temperatures. The cured system exhibits good mechanical and electrical properties.
The different types of amine and anhydride curing agents used commonly are listed
in Tables 1 and 2, respectively.
134 K. P. Unnikrishnan and E. Th. Thachil

Table 2.
Commercial anhydride curing agents

Name Structure

Phthalic anhydride

Tetrahydrophthalic anhydride

Methyl tetrahydrophthalic anhydride

Nadic methyl anhydride

2.4.1. Other curing agents. Among these, polyamides form the basis of some
domestic adhesive systems. Dicyandiamide (dicy), even though insoluble in
common resins at room temperature, dissolves at higher temperatures, forming the
basis of a one-pack system. Complexes of BF3 with amines (e.g., monoethyl amine)
are found to enhance pot life of the resin. Isocyanates act on epoxy resins through
the epoxy group to form oxazolidone structure, or with a hydroxyl group to form a
urethane linkage. Diisocyanates or polyisocyanates can lead to cross-linked systems
with polyepoxides.

3. TOUGHENING OF EPOXY RESINS

A polymer whose failure requires the application of large stress and the absorption
of more energy will be more useful than a polymer that fails under less vigorous
 Toughening of epoxy resins 135

conditions [9]. Within the context of failure mechanics [10] toughness of a specimen
refers to the total energy required to cause failure, i.e., the total area under the stress–
strain curve. Toughening of the resin by the addition of suitable toughening agents
or chemical modification is a means to improve the energy absorption capacity.
Apart from toughness, this results in enhancement of impact resistance, elongation
and resistance to crack propagation. Since high strength and modulus are required
for engineering polymers, the toughening process must not lead to any serious
deterioration in these properties.
Toughness modification of epoxy resins can be carried out in different ways.
These can be classified into (i) elastomer modification, (ii) particulate modifica-
tion, (iii) thermoplastic modification and (iv) miscellaneous methods. The most
successful toughening method has been found to be the incorporation of a second
elastomeric phase into the glassy epoxy matrix through in situ phase separation [11].

3.1. Elastomer modification

The various types of elastomeric materials which have been studied with a view
to rubber modification of epoxy resins are the following: (i) poly(siloxane)s [12],
(ii) fluoro-elastomers [13], (iii) acrylated elastomers [14], and reactive butadiene–
acrylonitrile solid [15] and liquid rubbers. Sultan and McGarry [16] in their studies
used carboxy-terminated acrylonitrile (CTBN) liquid rubber for modifying Epon
828 diglycidyl ether of bisphenol A epoxy resin with excellent results.
Toughening can be achieved by dispersing a small amount of elastomer as discreet
phase or by incorporation of reactive liquid polymers [17]. Epoxy resin, curing
agent and curing accelerator are mixed at room temperature at a definite ratio. To
this liquid rubber is added and mixed carefully until the mixture is homogeneous
and clear [18]. The mixture is degassed in vacuum at about 60◦ C and poured into
a RTV silicone mould with proper sample shape. The moulding is then cured in
an oven at high temperature. The curing time and temperature vary with respect
to the system which can be optimized by DSC measurements. In most cases to
ensure completion of cross-linking, heating is done in two stages; i.e., a pre-cure
heating followed by a post-cure heating. In the case of carboxy-terminated liquid
rubbers like CTBN, chemical modification (esterfication) is done by refluxing a
mixture of the rubber and the epoxy resin in a three-necked flask at 80◦ C for 5 h.
Triphenyl phosphine (0.25 g/100 g resin) is used as catalyst [19]. Acid value of the
resin should be less than 1 mg KOH/g resin. The epoxy–rubber adduct formed this
way behaves as epoxy-terminated rubber, which is more compatible with the epoxy
matrix. Solid rubber can be used as a solution (e.g., in methyl ethyl ketone) which
can be mixed with the resin and the hardener. The blend can be kept in vacuum to
allow evaporation of the solvent and curing simultaneously [15].
Butadiene–acrylonitrile rubbers comprise a relatively low-molecular-weight back-
bone with reactive groups in the terminal positions. In addition, rubbers having
reactive groups pendant to the main polymer chain have been developed. Several
variants of these elastomers have been studied with respect to the acrylonitrile con-
136 K. P. Unnikrishnan and E. Th. Thachil

Table 3.
Effect of functionality on the toughening ability of butadiene–acrylonitrile rubbers [22]

Elastomer Functionality Fracture energy (kJ/m2 )

CTBN Carboxyl 2.8
PTBN Phenol 2.6–3.0
ETBN Epoxy 1.8–2.5
HTBN Hydroxyl 0.9–2.6
MTBN Mercaptan 0.2–0.4

tent [20], molecular weight and terminal functionality [21]. In rubber modification
the rubber is initially miscible with the resin and the hardener. When the reac-
tion starts, the rubber first forms a copolymer with the resin, then phase-separates.
Thus, the cured thermoset possesses a dispersed rubbery phase. The type of rubber
modifier used will depend on the chemical characteristics of the matrix polymer at
least in terms of terminal functionality. Several functionalities have been studied
including epoxy, phenol, vinyl, hydroxyl, mercaptan, amine and carboxyl with rea-
sonable success (Table 3) [22]. However, the greatest benefit has been reported for
elastomers with carboxyl functionality (CTBNs). The rubbery domains are precip-
itated in situ during cure, yielding toughened epoxy materials. Numerous reports
have been devoted to this area in the last two decades [23, 24]. Toughness en-
hancement of epoxy matrix with elastomers demands a reaction (e.g., esterification)
between the elastomer and the resin, which leads to an adequate bond between the
elastomeric and epoxy phases. Attempts to employ hydroxyl-terminated elastomers
for toughness improvement have encountered difficulties because the conditions re-
quired to promote the necessary hydroxyl–epoxy reaction generally lead to self-
polymerisation of the epoxy, the latter occurring at the expense of the former and,
thus, limiting the extent of elastomer–epoxy reaction. A recent approach to this
problem is to employ a co-reactant such as toluene diisocyanate (TDI), capable of
reacting with both epoxide and hydroxyl functionalities resulting in the formation
of both urethane [25] and oxazolidone groups between the elastomeric and epoxide
components. Silane coupling agents can also be used to improve the epoxy–rubber
interface [26].

3.1.1. Phase morphology development. To provide toughness enhancement,

phase separation followed by the slow development of a two-phase morphology is
critically important. To achieve this, the rubber must initially dissolve and become
dispersed on a molecular level in the epoxy and is precipitated when epoxy cross-
linking takes place. This gives the required two-phase morphology with the for-
mation of rubbery particles dispersed and bonded to the cross-linked epoxy matrix.
A Transmission Electron Micrograph (TEM) showing the presence of the second
phase in DGEBA–CTBN–amine system is shown in Fig. 1. The factors likely to
influence morphology include elastomer concentration [22], molecular weight and
acrylonitrile content [20], the type and extent of hardener used and cure conditions
 Toughening of epoxy resins 137

Figure 1. Electron micrograph showing liquid HYCAR rubber precipitated in the epoxy resin
(magnification 4800×).

[27]. The morphological factors considered important for optimum toughening in

rubber modified epoxies are the rubber volume fraction, particle size and particle
size distribution [28 –30]. The early work of Sultan and McGarry [16] suggested
that the key factor was particle size, with large particles (>1 µm diameter) pro-
moting crazing within the epoxy matrix and smaller particles (0.01 µm) leading
to shear deformation. Rowe et al. [20] determined the fracture energy of a liquid
DGEBA–piperidine system with 12% and 18% acrylonitrile content of CTBN and
found a general decrease in average particle size from 3 µm at 12% bound AN to
0.2 µm at 5% bound AN. Siebert and Riew [21] were able to explain the chem-
istry of rubber particle formation in a DGEBA–CTBN–piperidine system. Studies
with butadiene–acrylonitrile rubbers show that both the functionality of end-groups
and acrylonitrile content have strong influence on the microstructure and interfacial
bonding of the modified epoxy resin. The size of rubber particles decreases with in-
creasing acrylonitrile content of CTBN and practically no phase separation occurs
in epoxies modified with CTBN containing a high amount (about 26%) of acryloni-
trile [31]. Toughening is found only when the rubber forms a separate phase in the
epoxy matrix with a particle size of the order of micrometers.

3.1.2. Bimodal particle systems. The role of particle size and in particular
particle size distribution has to be emphasised particularly with respect to some
rubber-toughened commercial epoxy formulations where a deliberate attempt is
made to promote bimodal particle size morphology. The toughening influence
of bisphenol A in rubber modified epoxy systems was illustrated by Levita et al.
[32]. Riew et al. demonstrated a bimodal distribution of rubber particles based on
CTBN–bisphenol A–DGEBA liquid epoxy resin–piperidine [22]. The substantial
improvement in toughness achieved by the inclusion of bisphenol A to rubber-
modified epoxy resin was related to a bimodal particle size morphology. Figures 2
and 3 are micrographs (OsO4 -stained TEM) showing the bimodal distribution of
rubber particles based on the DGEBA–bisphenol A–CTBN–piperidine system.
138 K. P. Unnikrishnan and E. Th. Thachil

Figure 2. Negative of an electron micrograph of an OsO4 -stained microsection of a bisphenol-A-

modified CTBN-toughened epoxy resin (magnification 29 700×).

Figure 3. Electron micrograph of a bisphenol-A-modified CTBN-toughened epoxy system (magnifi-

cation 7000×).

3.1.3. Volume fraction of dispersed phase. An increase in rubber volume fraction

(vf ) of dispersed rubbery phase generally promotes toughness of a multiphase
thermoset. With a rubbery volume fraction of 0.1 to 0.2 the fracture energy
(GIC ) may be increased 10–20-fold. The main toughening mechanisms initiated
by the presence of rubber particles are localized shear deformation in the form of
shear bands running between rubber particles [11, 33] and internal cavitation or
debonding of rubber particles [34] with subsequent plastic growth of voids in the
epoxy matrix. Bucknall and Yoshii [35] found a linear relation between fracture
energy (GIC ) and the volume fraction of rubber phase for different epoxy hardener
systems using a CTBN rubber. Kunz et al. [36] found that once a vf of about
0.1 had been achieved in the matrix, further increase in volume fraction gave only
minor increases in fracture energy (Table 4). Kinloch and Hunston [37] were able
to show that in the case of unimodal distribution of rubbery particles at low test
temperature, GIC increases only slowly with increase in volume fraction and then
reaches a plateau value. At high temperature the relationship is almost linear and
the rate of increase of GIC with vf is far greater than that found at low temperature.
 Toughening of epoxy resins 139

Table 4.
Influence of elastomer concentration (CTBN or ATBN) on the fracture resistance of an epoxy [36]

Elastomer CTBN ATBN

concentration KIC GIC KIC GIC
(wt%) (M/Nm3/2 ) (kJ/m2 ) (M/Nm3/2 ) (kJ/m2 )
0 0.8 0.3 0.8 0.3
5 1.9 1.5 2.0 1.7
10 2.0 2.0 1.9 1.8
15 1.9 2.0 2.0 2.1

3.1.4. Size of rubber particles and toughness. The effect of the rubber particle
size on deformation mechanism in epoxy system was studied by Sultan and
McGarry [16]. Shear mechanism is enhanced by the presence of small rubber
particles while crazing appears as the main reason for toughness improvement with
large rubber particles (1.5–5 µm), and the fracture energy of such a material is
five times more than that with smaller rubber particles. It is believed that optimal
toughening is obtained under conditions of combined shear and craze deformations,
which are obtainable under both small and large rubber particles present in bimodal
distribution [38]. Kunz et al. prepared a series of toughened epoxy polymers with
bimodal distribution of very small (<0.1 µm) and larger (0.2–1 µm) particles and
they found that the value of fracture energy did not increase with increasing value
of volume fraction as much as had been expected. They concluded that larger
particles are not as effective as smaller ones in improving toughness. However,
Kinloch and Hunston [37] reported that a bimodal distribution of particle sizes (0.1
and 1.3 µm) resulted in higher values than resulting from unimodal distribution
(1.2 µm). This enhancement of GIC by bimodal distribution was found only over a
range of temperatures and rates, indicating that matrix effects also have an important
role. The effect of particle size distribution has still to be conclusively established.

3.1.5. Mechanism of phase separation. Two mechanisms have been proposed

to explain the process of phase separation. The nucleation-growth mechanism
[30, 39, 40] is derived from the fact that the morphologies consist of spherical
domains dispersed in a continuous matrix and this is widely accepted. On the other
hand, the origin of the phase separation process is explained based on spinodal
decomposition [41]. Yamanaka [42] gave experimental evidence for the presence
of spinodal decomposition in fast reacting systems with high rubber concentration.
According to Yamanaka the nucleation-growth mechanism is unlikely because it is
a very slow process.
The development of morphology and phase separation can be of two different
ways; a polymerisation-induced phase separation (PIPS) [43, 44] and a thermally-
induced phase separation (TIPS) [45, 46]. Both polymerisation and phase separation
occur simultaneously and the mechanism and kinetics of PIPS are quite complex.
140 K. P. Unnikrishnan and E. Th. Thachil

As per PIPS the driving force of the system is the progressive increase of molec-
ular weight of the polymerisation monomers. Nucleation-growth mechanism may
take place in PIPS along with spinodal decompositions. In the case of rubber mod-
ified epoxies, as the temperature reaches the curing temperature, the homogeneous
mixture starts to demix due to the increase in the molecular weight of epoxy. The
results furnished by Yamanaka et al. [47] show that curing-induced phase separa-
tion proceeds mainly by spinodal decomposition. During curing, the system passes
through the meta stable region before it is forced into the unstable region. Thus,
phase decomposition by a nucleation growth mechanism is likely. Some recent
studies have revealed the possibility of nucleation growth in such a polymerisation
induced phase separation [48, 49]. Chen et al. [48] found in a system of epoxy–
ETBN that the appearance of a precipitated phase is much earlier than the cloud
point time. Nucleation growth was believed to take place well before the spinodal
decomposition. The morphology of a DGEBA–HTBN system cured with tetrahy-
drophthalic anhydride was explained based on Chen’s observations [50]. In the first
stage of reaction (0–9 min) the system undergoes nucleation growth, which takes a
long time (here the phase morphology would be heterogeneous), and then a rapid
change occurs, indicating fast spinodal decomposition. Thus, the mechanism is nu-
cleation growth coupled with spinodal decomposition. The various aspects of TIPS
have been studied theoretically [51] and experimentally [52].

3.1.6. Phase inversion. It has been reported that as the amount of CTBN was
increased in a cycloaliphatic epoxy resin cured by hexahydrophthalic anhydride,
phase inversion occurred [53]. Their TEM micrographs show that the CTBN–epoxy
adduct becomes a continuous phase with domains of epoxy resin. DMA analysis of
epoxy–rubber polyblends show a phase reversal at a volume fraction of 0.5 in which
an intermediate compound is formed [54]. A similar phenomenon was observed by
Siebert [55] in a CTBN–liquid DGEBA–amine system.

3.2. Particulate modification

Incorporation of particulate fillers such as silica and alumina trihydrate can enhance
toughness of cross-linked epoxies. Particulates contribute to a greatly enhanced
modulus which is a significant advantage over elastomeric modification where a
reduction in modulus is observed [56]. The degree of toughness improvement
was found to depend upon the volume fraction as well as the particle size and
the shape of the filler [57]. It has been reported that in the case of filled
polymers, fracture energy can attain a maximum at a specific volume fraction of
the added particles which may also decrease the elongation at break and impact
resistance [58, 59]. The mechanism of particulate toughening is somewhat different
from that of elastomeric modification. Although some modest improvements
in toughness have been achieved with particulate reinforcement, these have not
been of the same magnitude as elastomeric modification. A substantial modulus
increase accompanying toughness enhancement is a major advantage of particulate
 Toughening of epoxy resins 141

modification. Several studies have been conducted exploring the possibilities of

combining both particulate and rubber modification of epoxies [60, 61]. The
benefits of hybrid structures (filler and rubber simultaneously) in toughening
epoxies have been illustrated by Dusek et al. [62]. Epoxy–hybrid composites with
glass beads and α,ε-oligo butyl methacrylate diol were found to have improved
mechanical properties [63].

3.3. Thermoplastic modification

Attempts at thermoplastic modification of epoxies have been conducted using
primarily polyethersulphone and polyetherimide. The relationship between the
structure and the mechanical properties of polyethersulphone modified epoxies
has been studied by Bucknall and Partridge [64] with respect to a trifunctional
epoxy and tetraglycidyl diamino diphenyl methane (TGDDM). They found that the
morphology was dependent on both the type and concentration of the resin and
the curing agent employed. Unfortunately polyethersulphone addition was found
to have only minor advantages in fracture energy; addition of up to 40 phr of
resin in epoxy resulted in less than a 100% increase in fracture energy. McGrath
et al. [65] used hydroxyl- and amino-terminated polysulphones as modifiers for
DGEBA type resins. In comparison with rubber modification the level of toughness
improvement achieved by thermoplastic modification is generally poor especially
with relatively low-Tg systems based on difunctional resins. Venderbrosch et al.
[66] used polyphenylene ether (PPE), and Pearson and Yee [67] used polyphenylene
oxide (PPO) as modifier of epoxy resin. Polyether esters were also used to toughen
epoxy resin [68]. High-performance epoxy modified by poly(etherimide) was
found to have increased fracture toughness [69, 70]. Characterisation of epoxy
that makes decomposition easy by blending with thermoplastics has been done
in detail [71]. A toughened DGEBA–DDS system containing tetra-functional
epoxy and thermoplastic was shown to possess improved thermal and mechanical
properties [72].

3.4. Other methods of modification

Epoxidised soyabean oil [73] has been used to toughen epoxy resins cured with
ambient temperature hardener. Strengthening of epoxy resins using liquid crys-
talline epoxies [74] has been studied extensively with reference to reinforcement of
DGEBF by the liquid crystalline diglycidyl ether of 4,4-dihydroxybiphenol (DGE-
DHBP). Polyesters prepared by direct polycondensation from bisphenol A and
alphatic dicarboxylic acids (adipic acid, sebacic acid and dodecanoic acid) have
been used to improve toughness of DGEBA and diamino diphenyl methane epox-
ies [75]. Modification of epoxies with cresol novolacs catalysed by triphenyl phos-
phine [76] has been reported. Blends of methylene dianiline (MDA)–cured DGEBA
epoxy oligomer and ethylene–vinyl acetate co-polymer rubber (EVA) [77] have
been shown to have tensile, flexural, thermal and hardness properties. A series
142 K. P. Unnikrishnan and E. Th. Thachil

of poly(butyl acrylate)–poly(methyl methacrylate) core–shell elastomer particles

(CSEP) were used as toughening modifiers for the general epoxy resin E 44 (Epon)
and the impact and shear strengths of the modified epoxy resin were apparently en-
hanced [78]. The impact properties and morphology of MDA-cured epoxy resin
toughened with acrylate based liquid rubbers have been studied [79]. The effect
of electron beam radiation in the toughening process of CTBN-modified epoxy has
been identified by Park et al. [80]. Blending of two thermosets through interpene-
trating network (IPN) has been extensively studied [81]. Toughened interpenetrating
network (IPN) materials based on unsaturated polyester (UPR) and DGEBA-type
resin have been prepared using meta-(xylene diamine) (MXDA) and benzoyl per-
oxide curing agents [82]. Toughening of epoxy resin by blending with thermotropic
hydroxy ethyl cellulose has also been reported [83].
Toughness modification of epoxies by polyurethane pre-polymer based on hy-
droxyl terminated polyesters by IPN grafting [84] resulted in superior mechani-
cal properties. It was found that hydroxyl aliphatic polyester improved fracture
toughness more than amine- or acid-terminated polyester due to effective molecular
weight build-up by a chain-extension reaction. In fact hydroxyl-terminated poly-
esters are considered as a special case of polyols which are shown to be effective
in enhancing the impact strength, as well as fracture toughness of epoxy resin [85].
The use of polycarbonate polyurethane [86] as epoxy modifier has also been re-
ported. A great deal of literature has been devoted to the toughening of epoxies
using polyurethane as second phase to form IPNs [87, 88].
Aromatic polyesters and hydroxyl-terminated polyesters [89] have been used as
modifiers of epoxy resins, Further, studies on chain extended ureas as curing agents
and toughness modifiers for epoxies have been conducted with reasonable success
[90]. Plastisols based on PVC and diethyl hexyl phthalate have been used as modi-
fiers for epoxy resins [91]. Schroder et al. [92] have used telichelic methacrylates as
toughening agents for epoxy resins. Isocyanate-terminated polybutadiene has also
been employed as toughness modifier in epoxy resins [93].

4. MECHANISMS OF TOUGHENING
Several mechanisms have been proposed to account for the toughness improvements
in rubber-modified epoxy resins. These mechanisms explain the improved fracture
energy or fracture toughness which may result when a thermosetting polymer
possesses a multiphase microstructure of dispersed rubber particles. It will be
relevant to consider each of these mechanisms and to assess their application to
rubber-modified epoxies.

4.1. Rubber tear (particle deformation)

This mechanism was put forward initially by Mertz, Claver and Baer in 1956
to explain the deformation and tearing of the rubber particles present in a two-
phase system. According to this mechanism, rubber particles simply hold the
 Toughening of epoxy resins 143

opposite faces of a propagating crack together and the toughness of such a system is
dependent on the energy required to rupture the particles together with that required
to fracture the glassy matrix. Although this mechanism has been regarded as
irrelevant to toughened thermoplastics, a number of microscopic investigations have
proved its validity in the case of rubber modified epoxies by providing evidence of
stretched rubber particles spanning loaded cracks. From such observations, Kunz-
Douglas et al. [94] proposed that the toughness enhancement provided by rubber
particle inclusion was dependent primarily on the degree of elastic energy stored
in the rubber particles during loading of the two-phase system. According to them
it is the principal deformation mechanism in the matrix, enhanced by the presence
of the second phase which improves toughness. The mechanism put forward by
Beaumont and co-workers [95] emphasizes the role of deformation and fracture of
the rubber particles. Stretching and tearing of rubber particles embedded in the
epoxy matrix result in high energy absorption during failure [96]. Rubber tearing
was shown to be the main contributor to the failure energy of high-molecular-weight
NBR-modified epoxies [15]. However, the rubber tear theory does not explain the
existence of stress whitening frequently observed in rubber-modified epoxies. It
does not account for yielding and plastic flow contribution to toughness.

4.2. Multiple crazing

A craze appears to be similar to a crack owing to its lower refractive index than its
surroundings. It actually contains fibrils of polymer drawn across, normal to the
craze surface in an interconnecting void network. The multiple crazing theory, due
to Bucknall and Smith [97] proposes that toughness improvement is attributed to the
generation and efficient termination of crazes by rubber particle. This process has
been demonstrated with thermoplastics, such as high-impact polystyrene, through
optical microscopic studies. Subsequent studies [98, 99] on various rubber-modified
polymers confirmed that crazes frequently initiate from the rubber particles at
regions of high stress concentration at the equatorial region normal to the applied
stress direction. Craze termination occurs when the craze encounters another rubber
particle. This stabilises the craze and prevents it from growing into a large crack
like structure. Thus, a greater amount of energy can be absorbed by the system
prior to failure thereby leading to an effective improvement in the toughness of the
polymer. According to Sultan and McGarry [16] crazing is the dominant toughening
mechanism in rubber-modified epoxies.
The detailed mechanism of craze initiation, growth and break down around rubber
particles has been studied by Donald and Kramer [100, 101]. According to them an
optimum particle size of about 2–5 µm is responsible for maximum toughness and
crazes are rarely mediated from particles smaller than 1 µm. Craze initiation from
rubber particles follows the condition that the initial elastic stress concentration at
the particle must exceed the stress concentration at a static craze tip and the spatial
extent of stress enhancement changes with the particle diameter. This explains the
inability of smaller rubber particles to initiate crazes. The possibility of crazing as a
144 K. P. Unnikrishnan and E. Th. Thachil

toughening mechanism in rubber-modified epoxies has been suggested by Bucknall

and Yoshi [35] in addition to shear deformation. They observed increases in
specimen volume in addition to longitudinal extension during tensile creep tests on
rubber modified epoxies. This volume increase could be associated with a massive
crazing within the specimen. However, studies on thermoplastics have indicated an
apparent transition from a crazing to a shear yielding mechanism as the length of
polymer chain decreases. Thus, with thermosets where cross-link density is high
and, hence, chain length between cross-links short, crazing would be suppressed
[102]. The frequently observed stress whitening phenomenon was attributed to the
generation of crazes.

4.3. Shear yielding

This theory was proposed by Newman and Strella [103] following work on
acrylonitrile–butadiene–styrene thermoplastics. The main proposal of this theory
is that shear deformation taking place either as shear bands or as diffuse form of
shear yielding is initiated at stress concentrations resulting from the presence of
rubber particles [104]. This serves as the main source of energy absorption and
hence toughness improvement. Plastic shear yielding in the resin matrix is enhanced
by stress concentrations associated with the embedded soft rubber particles [105].
The function of the rubber particles is to produce sufficient tri-axial tension in the
matrix, so as to increase the local free volume and, hence, enable shear yielding
and drawing to initiate. According to Donald and Kramer the dispersed rubber
phase initiates micro-shear bands at an angle of 55 to 64 degrees to the direction of
the applied stress [104]. In materials having smaller rubber particles crazing was
not initiated but shear deformation, promoted by rubber particle cavitation was the
major toughening mechanism. The presence of rigid particulate fillers like glass
beads, silica, etc., may also cause enhanced shear yielding of the matrix. However,
they are not as effective as rubber particles in increasing toughness.
A major drawback of shear yielding theory is its inability to account for stress
whitening, since shear yielding is a constant volume deformation. Therefore, it
has been suggested that crazing and shear yielding occur simultaneously in many
polymers with the former accounting for the stress whitening effect [16]. It is
generally recognised that crazing rarely occurs in epoxies and the toughening
mechanism now regarded by many workers as being the most applicable to rubber
modified epoxies can be a dual-mode mechanism based on rubber particle cavitation
and matrix shear yielding [106].

4.4. Cavitation-shear yielding

This toughening mechanism, developed independently by Kinloch et al. [106]
and by Pearson and Yee [107] is regarded as the most consistent in terms of
experimental data generated in recent years. The correlation between toughness and
the extent of plastic deformation found on fracture surfaces has led to a mechanism
 Toughening of epoxy resins 145

based on yielding and plastic shear flow of the matrix as the primary source of
energy absorption in rubber-modified epoxies. Enhanced plastic deformation in
the matrix has been found to accompany the inclusion of rubber particles and
the stress distribution existing around rubber particles located in the vicinity of
a stressed crack tip becomes important. Initially the development of a tri-axial
stress dilates the matrix and along with this, the tri-axial stresses inherent in the
rubber particles (due to differential thermal contraction effects during initial curing)
provide the necessary conditions for cavitation of the rubber particles. Rather
than crazing of the epoxy matrix, it is the cavitation process which is considered
responsible for the stress whitening effects usually observed in rubber-modified
epoxies. The increasing stress concentrations around rubber particles during loading
would promote shear yield deformation zones in the matrix. Since the particles
would also act as sites of yield terminations, yielding would remain localized in the
vicinity of the crack tip. It is reasonable to assume that both cavitation and shear
yielding would occur during the early stages of load application. Once initiated,
the rubber particle cavitation would enhance further shear yielding in the matrix.
Crack-tip blunting would increase extensively resulting in increased development
of the plastic zone at the crack tip. Thus, toughness would be enhanced as has
been observed in practice. Bascom et al. [108] accredited the toughness of CTBN-
modified epoxy resin to an increase in the plastic zone size. At the same time
Yee and Pearson attributed an order of magnitude increase in toughness to the
cavitation of rubber particles followed by shear yielding of the epoxy matrix [107].
When dispersed acrylic rubber (DAR) is used to modify the epoxy matrix, the DAR
particles are found to be cavitated around the crack tip and the crack wake.
The absence of cavitation in solid rubber-modified epoxies should result from the
high molecular weight of solid rubber (e.g., NBR) which imparts greater tensile
strength to the rubber. High tensile strength of rubber can eliminate premature
cavitation [15]. It is postulated that the low-molecular-weight CTBN rubber with
low tensile strength is readily cavitated at early stages of loading and it is interesting
to note that the fracture toughness of the rubber-modified epoxy is dependent on
the loading rate [109]. The process of cavitation during fracture diminishes the
importance of the rubber tearing mechanism because the failed rubber particles
require little or no tearing energy [110, 111].

4.5. Crack pinning

This mechanism was developed by Lange and Radford [112] to explain modification
by particulate fillers. Lange showed that by inclusion of alumina trihydrate as
particulate filler fracture energy of epoxy matrix could be increased. Although
particulate reinforcement could impose stress concentrating effects on the epoxy
matrix, this is not considered significant. The crack-pinning mechanism based on
the impeding characteristics of the particles proposes that a propagating crack front,
when encountering an inhomogeneity, becomes temporarily pinned at that point
[113]. As the load increases the degree of bowing between pinning points increases
146 K. P. Unnikrishnan and E. Th. Thachil

which results in both a new fracture surface and an increase in the length of the crack
front. Lange proposed that when the bowed crack front attains a radius of dp/2,
where dp is the inter-particle distance, it breaks away from the pinning positions
and creates characteristic tails. It is seen that the fracture energy of the particulate
composite will increase as the particle spacing decreases. The degree of toughening
depended on both the volume fraction and particle size of the filler.

5. EFFECT ON OTHER PROPERTIES

Toughening refers to mechanical and thermal stress resistance and it does not affect
the other properties such as thermal stability, strength and hardness. Table 5 shows
the physical properties of unmodified resins cured with amines and anhydrides.
The room temperature mechanical properties of epoxy systems with different
percentages of rubber (CTBN) are given in Table 6 [30].
Incorporation of low level of liquid rubbers to a normally brittle epoxy matrix
significantly improves the crack resistance and impact strength without much re-
Table 5.
Physical properties of DGEBA cured with amines and anhydride

Property By aminea By amineb By anhydridec

Tensile strength (psi) 4730 8000 12 000
Modulus (psi × 10−6 ) 0.29 0.458 0.45
Flexural strength (psi) 9110 19 000 19 000
Impact strength (Izod, J/cm) 0.23 0.41 0.70
Deflection temperature (◦ C) 155 109
Hardness, Rockwell L87 M112 M100
Coeff. of thermal expansion (in/(in)◦ C) 0.00048 4.8 × 10−5
a Poly methylene diamine (24 h at 23◦ C and 16 h at 120◦ C).
b m-Phenylene diamine (MPDA).
c Phthalic anhydride.

Table 6.
Room temperature mechanical properties of epoxy systems with different percentages of CTBN [30]

CTBN Tensile test Compressive test Charpy impact test

(wt%) E σR εR σy εy Rs (σ ) Wi
(GPa) (MPa) (%) (MPa) (%) (kJ/m2 ) (%)
0 2.88 40 1.8 130 10.6 17 (5) 65
5 2.78 71 3.9 109 9.5 21 (5) 63
10 2.40 70 4.7 95 7.9 16 (6) 64
15 2.15 60 6.6 76 7.1 33 (1) 74
20 1.70 46 5.1 65 7.5 36 (6) 82
25 1.05 14 7.0 23 4.9 33 (20) 63
 Toughening of epoxy resins 147

Table 7.
Mechanical properties of Bisphenol A modified CTBN–epoxy system [114]

Property Unmodified epoxy resin Bisphenol-A-modified

CTBN–epoxy system
Tensile strength (MPa) 65.5 64.1
Elongation at break (%) 4.8 9.0
Modulus (GPa) 2.8 2.7
Fracture energy (kJ/m2 ) 0.18 5.3–8.8
Izod impact (J/m of notch) 0.68 3.5

duction in other thermal and mechanical properties. The enhanced crack resistance
of amine-cured DGEBA resin containing CTBN has been studied by fracture sur-
face energy measurements Riew et al. [114] developed a model system based on
CTBN–bisphenol A–DGEBA liquid resin–piperidine in which toughness synergism
appears through the inclusion of a diphenol. This resulted in a bimodal distribution
of rubber particles. The physical properties of the system are given in Table 7.
The mechanical properties of blends of DGEBA-based epoxy resin and internally
epoxidised hydroxyl-terminated polybutadiene rubber have been studied by Bussi
et al. [115]. The dependence of modulus and yield stress on rubber content reveals
two types of behaviour. For medium epoxidised rubber (at low rubber content)
both modulus and yield stress were independent of rubber content. DMA analysis
showed that most of the medium epoxidised rubber dissolved into the epoxy phase
[116]. The tensile test results indicated that dissolved rubber does not affect the
properties of the epoxy phase. For the low epoxidised rubber, modulus and yield
stress decreased continuously with rubber content.
With the incorporation of rubber into the epoxy matrix, impact strength, fracture
toughness and fracture energy are increased, while Young’s modulus and yield
strength are decreased slightly. The influence of rubber content on the stress–strain
curve for the rubber-modified epoxy system has been studied in detail [117]. The
slight decrease in modulus is due to an increase in the dissolved rubber in the epoxy
matrix. The elastic modulus decreases with rubber content [118, 119]. However,
the rate of increase of fracture energy with rubber content varies at different test
conditions [37]. Sankaran et al. [120] have compared the properties of hydroxyl-
terminated poly(butadiene-co-acrylonitrile)/toluene diisocyanate-toughened epoxy
with a physically blended HTBN/epoxy mixture. Table 8 illustrates the results.
Generally, toughness property was found to increase when the HTBN content is
below 12%. Also, inclusion of up to 6 wt% rubber had toughened the system
without much reduction in tensile strength and tensile modulus and rubber above
6 wt% showed significant drop in strength and modulus when the system was
toughened. The poor toughening effect of physical blending was explained by
the relative ease with which debonding can occur between the two phases during
fracture. TG-DTA analysis of the rubber-toughened and neat epoxies showed that
the percentage weight loss is slightly more for the rubber-toughened system. This
148 K. P. Unnikrishnan and E. Th. Thachil

Table 8.
Mechanical properties of HTBN-toughened epoxy formulations [120]

Formulation HTBN Tensile Tensile Elongation-at- Flexural Toughnessa

(wt%) strength modulus break strength (J × 106 )
(MPa) (GPa) (%) (MPa)
Epoxy neatb — 67.88 2.71 3.18 78.6 1.60
Epoxy/CRR 1.5c 1.5 60.92 2.98 7.27 — 3.12
Epoxy/CRR 3 3.0 59.61 3.12 9.10 107.6 4.57
Epoxy/CRR 6 6.0 50.91 2.84 8.62 81.8 3.31
Epoxy/CRR 9 9.0 46.60 2.30 7.90 72.10 2.94
Epoxy/CRR 12 12.0 40.22 2.06 10.00 37.40 3.00
Epoxy/PBR 5d 5.0 45.91 2.82 4.50 64.74 1.57
a Area under the stress–strain curve.
b LY 556.
c Chemically reacted rubber.
d Physically blended rubber.

was due to the high susceptibility of the aliphatic chain of the rubber network in the
molecular backbone for thermal and oxidative degradation.

5.1. Glass transition temperature

The Tg of the matrix will fall by the incorporation of liquid rubbers due to the
gain in ductility. Thus, increased ductility and toughness are frequently observed
only at the expense of high temperature properties. A DGEBA based epoxy resin
containing 15% epoxy-terminated butadiene acrylonitrile (ETBN) when cured with
a cycloaliphatic diamine showed a decrease of 25◦ C in Tg compared to that of the
neat matrix [121]. Rubber that did not phase separate remains dissolved and could
plasticize the epoxy glass transition temperature [122]. Tg generally decreases with
increasing rubber content and increases as the compatibility of rubber and epoxy
decreases.

5.2. Influence of epoxy formulation

The three aspects of formulation detail which can play a major role in rubber mod-
ification, as well as in the final properties, are the molecular weight of the resin,
type and strength of the curing agent and finally the inclusion of bisphenol A. Sev-
eral studies (for example, Ref. [123]) have demonstrated the difficulties involved
in toughening highly cross-linked epoxies by rubber modification. Pearson and
Yee [123] have furnished data showing that the inherent toughness of unmodified
epoxies is only moderately affected by the initial molecular weight of the resin How-
ever, the toughness of rubber modified epoxy is strongly influenced by molecular
weight, thus showing that toughenability of an epoxy resin by elastomers will be de-
pendent largely on the cross-link density of the epoxy matrix. The lower the cross
link density, the greater will be the toughenability. Curing agents that contribute
 Toughening of epoxy resins 149

to a high cross-link density will generally result in cured formulations showing low
fracture toughness values. Incorporation of bisphenol A into rubber-modified epoxy
system [107] has been shown to provide substantial toughening benefits as illus-
trated by Yee and Pearson [124].

6. CONCLUSIONS

The brittle nature of epoxy resin prevents it from many end-use applications.
Blending of epoxies with elastomers is the most common method of impact
modification or toughening. The development of a multi-phase in these resins
based on the inclusion of a rubbery second phase has led to major improvement
in their toughness. Significant toughening is possible only when the rubber can
form a separate phase inside the epoxy matrix with a particle size of the order of
micrometers. Generally 10–15 phr of rubber is necessary to generate toughness
properties in epoxy systems. In an initially miscible blend, curing may cause
phase separation. Curing reaction and phase separation are competitive. Chemical
toughening leads to an increase in mechanical and thermal properties due to
the formation of strong covalent bonds whereas physical blending produces only
marginal toughening effect. Toughening of epoxies by particulates improves the
modulus of the cured system which is uncommon in elastomer toughening. With the
advent of new telechelic liquid rubbers with reactive terminal functional groups, the
technology of toughened epoxies assumes new dimensions and it will be interesting
to look ahead for developments in this field in the years to come.

REFERENCES
1. C. K. Riew (Ed.), Adv. Chem. Ser. 222, 1 (1989).
2. J. A. Brydson, Plastic Materials, 4th edn, Chapter 27. Butterworth, London (1982).
3. N. Hata and J. Kumanotani, J. Appl. Polym. Sci. 17, 3545 (1973).
4. C. A. May, in: Epoxy Resins; Chemistry and Technology, 2nd edn, p. 551. Marcel Decker, New
York, NY (1988).
5. W. G. Potter, in: Epoxide Resins, p. 211. Iliffe, London (1970).
6. A. A. Collyer (Ed.), in: Rubber Toughened Engineering Plastics, p. 167. Chapman & Hall,
London (1994).
7. A. C. Meeks, Polymer 15, 675 (1994).
8. L. Matejka, J. Lovy, S. Pokorny and K. Dusek, J. Polym. Sci. Polym. Chem. Edn. 21, 2873
(1983).
9. M. F. Ashby, Materials Selection in Mechanical Design. Pergamon, New York, NY (1992).
10. A. J. Kinloch and R. J. Young, Fracture Behavior of Polymers. Elsevier Applied Science,
London (1983).
11. A. J. Kinloch, Adv. Chem. Ser. 222, 67 (1989).
12. E. M. Yorkgits, C. Trau, N. S. Eiss, T. Y. Hut, I. Yilgor, G. L. Wilkes and J. E. McGrath, Adv.
Chem. Ser. 208, 137 (1984).
13. J. Mijovic, E. M. Pearce and C. C. Foun, Adv. Chem. Ser. 208, 293 (1984).
14. S. L. Krishenbaum, S. Gazit and J. P. Bell, Adv. Chem. Ser. 208, 163 (1984).
150 K. P. Unnikrishnan and E. Th. Thachil

15. M. Frounchi, M. Mehrabzadeh and M. Parvary, Polym. Int. 49, 163 (2000).
16. J. N. Sultan and F. McGarry, Polym. Eng. Sci. 13, 29 (1973).
17. C. K. Riew, Adv. Chem. Ser. 222, 208 (1989).
18. E. Yilgor and I. Yilgor, Polymer 39, 1691 (1998).
19. P. Bartlet, J. P. Pascault and H. Sautereau, J. Appl. Polym. Sci. 30, 2955 (1985).
20. E. H. Rowe, A. R. Siebert and R. S. Drake, Mod. Plast. 47, 110 (1985).
21. A. R. Siebert and C. K. Riew, Org. Coat. Plast. Chem. 31, 552 (1971).
22. C. K. Riew, E. H. Rowe and A. R. Siebert, Adv. Chem. Ser. 154, 326 (1976).
23. C. K. Riew and A. J. Kinloch (Eds), Adv. Chem. Ser. 233 (1993).
24. C. K. Riew and J. K. Gilham (Eds), Adv. Chem. Ser. 208 (1984).
25. S. Sankaran and M. Chanda, J. Appl. Polym. Sci. 39, 1459 (1990).
26. C. Kayanak, C. Celikbilek and G. Akovali, Eur. Polym. J. 39, 1125 (2003).
27. S. J. Shaw and D. A. Tod, J. Adhesion 28, 331 (1989).
28. H. N. Nae, J. Appl. Polym. Sci. 31, 15 (1986).
29. A. J. Kinloch and D. L. Hunston, J. Mater Sci. Lett. 6, 131 (1987).
30. S. Montarnal, J. P. Pascault and H. Sautereau, Adv. Chem. Ser. 222, 193 (1989).
31. Y. Huang, A. J. Kinloch, R. J. Bertsch and A. R. Siebert, Adv. Chem. Ser. 233, 195 (1993).
32. G. Levita, A. Marchetti, A. Lazzeri and V. Frosini, Polym. Composit. 8, 141 (1987).
33. H. J. Sue, Polym. Eng. Sci. 31, 275 (1991).
34. A. Lazzeri and C. B. Bucknall, J. Mater. Sci. 28, 6799 (1993).
35. C. B. Bucknall and T. Yoshii, Br. Polym. J. 10, 53 (1978).
36. S. C. Kunz, J. A. Sayre and R. A. Assink, Polymer 23, 1897 (1982).
37. A. J. Kinloch and D. L. Hunston, J. Mater. Sci. Lett. 6, 137 (1987).
38. W. D. Bascom, R. Y. Ting, R. J. Moulton, C. K. Riew and A. R. Siebert, J. Mater. Sci. 16, 2657
(1981).
39. A. Vazquez, A. J. Rojas, H. E. Adabbo, J. Borrajo and R. J. J. Williams, Polymer 28, 1156
(1987).
40. L. T. Manzione, J. K. Gillham and C. A. McPherson, J. Appl. Polym. Sci. 26, 907 (1987).
41. H. S. Y. Hsieh, Proc. 34th Int. SAMPE Symp, p. 884 (1984).
42. K. Yamanaka, Y. Takagi and T. Inoue, Polymer 60, 1839 (1989).
43. T. Inoue, Prog. Polym. Sci. 20, 119 (1995).
44. P. A. Oyanguren, P. M. Frontini, R. J. J. Williams and J. P. Pascault, Polymer 37, 3087 (1996).
45. T. Kyu and J. H. Lee, Phys. Rev. Lett. 76, 3746 (1996).
46. J. Zhang, H. Zhang, D. Yan, H. Zhou and Y. Yang, Sci. China Ser. B 40, 15 (1997).
47. K. Yamanaka, Y. Takagi and T. Inoue, Polymer 30, 1840 (1989).
48. D. Chen, J. P. Pascault and H. Sautereau, Polym. Int. 32, 369 (1993).
49. M. Okada, K. Fujimoto and T. Nose, Macromolecules 28, 1795 (1995).
50. J. Zhang, H. Zhang and Y. Yang, J. Appl. Polym. Sci. 72, 59 (1999).
51. H. Zhang, J. Zhang and Y. Yang, Macromol Theory Simul. 4, 1001 (1995).
52. T. Hashimoto, M. Itakura and H. J. Hosegawa, Chem. Phys. 85, 6118 (1986).
53. A. S. Burhans and A. C. Soldatos, Adv. Chem. Ser. 99, 531 (1971).
54. N. K. Kalfoglou and H. L Williams, J. Appl. Polym. Sci. 17, 1377 (1973).
55. A. R. Siebert, J. Elastomer. Plast. 8, 177 (1976).
56. A. C. Moloney, H. H. Kausch and H. R. Stieger, J. Mater. Sci. 19, 1125 (1984).
57. A. C. Moloney, H. H. Kausch, T. Kaiser and H. R. Beer, J. Mater. Sci. 22, 381 (1987).
58. Y. Nakamura, M. Yamaguchi, A. Kitayama, M. Okubo and T. Matsumoto, Polymer 32, 2221
(1991).
59. E. Urbaczewski-Espuche, J. F. Gerard, J. P. Pascault, G. Reiffo and J. Sautereau, J. Appl. Polym.
Sci. 47, 991 (1993).
60. D. Maxwell, R. J. Young and A. J. Kinloch, J. Mater. Sci. Lett. 3, 9 (1984).
61. A. J. Kinloch, D. Maxwell and R. J. Young, J. Mater. Sci. 20, 4169 (1985).
 Toughening of epoxy resins 151

62. K. Dusek, L. Matejka, J. Plesti and J. Somvarsky, Extended Abstracts 5th European Symposium
on Polymer Blends, Maastricht, May 12–15, Preprints, p. 322 (1996).
63. N. Schroder, L. Konczol, W. Doll and R. Mulhaupt, J. Appl. Polym. Sci. 88, 1040 (2003).
64. C. B. Bucknall and I. V. Partridge, Polymer 24, 639 (1983).
65. J. L. Hedrick, I. Yilgor, G. L. Wilkes and J. E. McGrath, Polym. Bull. 13, 201 (1985).
66. R. W. Venderbrosch, G. L. Nelisson and P. J. Lemstra, Macromol. Chem. Macromol Symp. 75,
73 (1993).
67. R. A. Pearson and A. F. Yee, Polymer 34, 3658 (1993).
68. F. Hofflin, W. Bohne and R. Mulhaupt, European Symposium on Impact and Dynamic Fracture
and Composites, Porto Cervo, September 20–22, Preprints (1993).
69. S. Seungham and J. Jyongsik, J. Appl. Polym. Sci. 65, 2237 (1997).
70. J. Diamont and R. S. Moulton, 29th Natl. SAMPE Symp. 29, 422 (1984).
71. K. Mimura and H. Ita, J. Appl. Polym. Sci. 89, 527 (2003).
72. I. Blanco, G. Cicala, C. Lofaro and A. Recca, J. Appl. Polym. Sci. 89, 268 (2003).
73. D. Ratna, Polym. Int. 50, 179 (2001).
74. P. Prakaipetch, B. Witold and N. A. D’Souza, Soc. Plast. Eng. 58, 2147 (2000).
75. S. M. Shin, D. K. Shin and D. C. Lee, J. Appl. Polym. Sci. 78, 2464 (2000).
76. L.-G. Sheng, Cresol Novolac-Epoxy Networks; Synthesis, Properties and Process, Dissertation,
ETD-etd-04262001-142525 (2001).
77. S. K. Siddhamali, J. Vinyl Addit. Technol. 6, 211 (2000).
78. H. Fan, J. Wang and Z. Chen, Chem. Abstr. 37, 277 (2000).
79. S. Kar, D. Gupta, A. K. Banthia and D. Ratna, Polym. Int. 52, 1332 (2003).
80. J. S. Park, P. H. Kang, Y. C. Nho and D. H. Suh, J. Macromol. Sci. Pure Appl. Chem. A40, 641
(2003).
81. S. C. Kim and L. H. Sperling, IPNs Around the World. Wiley, New York, NY (1997).
82. M. S. Lin, C. C. Liu and C. T. Lee, J. Appl. Polym. Sci. 72, 585 (1999).
83. Q. Dai, J. Chen and Y. Huang, J. Appl. Polym. Sci. 70, 1159 (1998).
84. H. Harani, S. Fellahi and M. Bakar, J. Appl. Polym. Sci. 70, 2603 (1998).
85. H. Harani, M.Sc. Thesis. IAP, Boumerdes, Algeria (1997).
86. T. Chen, H. Li, Y. Gao and M. Zhang, J. Appl. Polym. Sci. 69, 887 (1998).
87. K. F. Hsieh, Y. C. Chern, C. C. M. Ma and Y. G. Gong, SPE ANTEC Tech. Papers 38, 1488
(1992).
88. H. H. Wang and J. C. Chen, Polym. Eng. Sci. 35, 1468 (1995).
89. H. Harani, S. Fellahi and M. Bakar, J. Appl. Polym. Sci. 71, 29 (1999).
90. B. Zhang, H. Q. Zhang, Y. C. You, D. Z. Ding, P. W. Tao and H. J. Fu, J. Appl. Polym. Sci. 69,
339 (1998).
91. J. Lopez, S. Gisbert, S. Ferrandiz, J. Vilaplana and A. Jemenez, J. Appl. Polym. Sci. 67, 10
(1998).
92. N. Schroder, L. Koenczoel, W. Doell and R. Muelhaupt, J. Appl. Polym. Sci. 70, 785 (1998).
93. F. L. Barcia, M. A. Abrahao and B. G. Soares, J. Appl. Polym. Sci. 83, 838 (2002).
94. S. Kunz-Douglas, P. W. R. Beaumont and M. F. Ashby, J. Mater. Sci. 15, 1109 (1980).
95. S. Kunz and P. W. R. Beaumont, J. Mater. Sci. 16, 3141 (1981).
96. S. Bandyopadhyay, Mater. Sci. Eng. A 125, 157 (1990).
97. C. B. Bucknall and R. R. Smith, Polymer 6, 437 (1965).
98. J. D. Moore, Polymer 12, 478 (1971).
99. P. Beahan, A. Thomas and M. Bevis, J. Mater. Sci. 11, 1207 (1976).
100. A. M. Donald and E. J. Kramer, J. Appl. Polym. Sci. 71, 1123 (1982).
101. A. M. Donald and E. J. Kramer, J. Mater. Sci. 17, 2351 (1982).
102. A. M. Donald and E. J. Kramer, J. Mater. Sci. 17, 1871 (1982).
103. S. Newman and S. Strella, J. Appl. Polym. Sci. 9, 2297(1965).
104. A. M. Donald and E. J. Kramer, J. Mater. Sci. 17, 1765 (1982).
152 K. P. Unnikrishnan and E. Th. Thachil

105. Y. Huang and A. J. Kinloch, Polym. Mater. Sci. Eng. 63, 564 (1990).
106. A. J. Kinloch, S. J. Shaw and D. L. Hunston, Polymer 24, 1355 (1983).
107. R. A. Pearson and A. F. Yee, Polym. Mater. Sci. Eng. 49, 316 (1983).
108. W. D. Bascom, R. L. Cottingham, R. L. Jones and P. J. Peyser, J. Appl. Polym. Sci. 19, 2545
(1975).
109. B. J. Cardwell and A. F. Yee, Polymer 34, 1695 (1993).
110. Y. Huang and A. J. Kinloch, J. Mater. Sci. Lett. 11, 484 (1992).
111. R. A. Pearson and A. F. Yee, J. Mater. Sci. 26, 3828 (1991).
112. F. F. Lange and K. C. Radford, J. Mater. Sci. 6, 1197 (1971).
113. D. J. Green, P. S. Nicholson and J. D. Emberg, J. Mater. Sci. 14, 1657 (1979).
114. C. K. Riew, E. H. Rowe and A. R. Siebert, Adv. Chem. Ser. 154, 326 (1976).
115. P. Bussi and H. Ishida, J. Appl. Polym. Sci. 53, 441 (1994).
116. P. Bussi and H. Ishida, Polymer 35, 956 (1994).
117. D. Verchere, J. P. Pascault, H. Sautereau, S. M. Moschiar, C. C. Riccardi and R. J. J. Williams,
J. Appl. Polym. Sci. 43, 293 (1991).
118. J. F. Hwang, J. A. Manson, R. W. Hertzberg, G. A. Miller and L. H. Sperling, Polym. Eng. Sci.
29, 1466 (1989).
119. L. T. Manzione, J. K. Gillham and C. A. McPherson, J. Appl. Polym. Sci. 26, 907 (1981).
120. S. Sankaran and M. Chanda, J. Appl. Polym. Sci. 39, 1635 (1990).
121. S. M. Moschiar, C. C. Riccardi, R. J. J. Williams, D. Verchere, H. Sauterau and J. P. Pascault,
J. Appl. Polym. Sci. 42, 717 (1991).
122. J. K. Gillham, C. A. Glandt and C. A. McPherson, Am. Chem. Sci. Div. Org. Coat. Plast. Chem.
37, 195 (1997).
123. R. A. Pearson and A. F. Yee, J. Mater. Sci. 24, 2571 (1989).
124. A. F. Yee and R. A. Pearson, J. Mater. Sci. 21, 2462 (1986).