shah2019

Composite Structures 217 (2019) 100–121
Contents lists available at ScienceDirect
Composite Structures
journal homepage: www.elsevier.com/locate/compstruct
Review
Impact resistance and damage tolerance of fiber reinforced composites: A T

review
⁎
S.Z.H. Shah, S. Karuppanan, P.S.M. Megat-Yusoff , Z. Sajid
Mechanical Engineering Department, Universiti Teknologi PETRONAS, 32610 Bandar Seri Iskandar, Perak, Malaysia
A R T I C LE I N FO A B S T R A C T
Keywords: LVI has become a serious threat to the composite industry because it induces complex failure mechanisms and
Impact resistance internal damages which significantly reduce the structural properties of composites. The impact resistance and
Damage tolerance damage tolerance of FRC are affected by various factors such as fabric architecture, resin toughness, impactor
Thermoplastic geometry, extreme conditions, stacking sequence, and fiber/matrix hybridization. It has been identified that
Fabric architecture
among these factors fabric architecture and resin toughness are the most important factors in improving the
Damage prediction
impact resistance and damage tolerance of composites. The objective of this paper is to review the effect of
different factors on LVI performance of FRC, identify the gap in the literature and suggest directions for future
work in this area. In addition to this, different damage modeling strategies used to predict the impact resistance
and damage tolerance of FRC will be discussed in detail.
1. Introduction impact resistance of FRC by changing the fabric architecture and usage
of toughened resin system [3–5]. 2D composites (unidirectional and
Nowadays, composite materials are very popular among aerospace, bidirectional) have been used in the composite industry for more than
automotive, defense and wind power industries due to their high four decades; however, they exhibit poor transverse properties. With
strength/stiffness to weight ratio and improved mechanical properties the advancement in the weaving technology, 3D composites have been
as compared to metals. However, composite materials are sensitive to developed which have the unique characteristic of through the thick-
LVI and it has become a serious threat to the composite structures be- ness reinforcement. These composites have been used in the landing
cause it produces large internal damages such as delamination, matrix gears and in the engine structure to improve the impact resistance of
cracking, and fiber/yarn breakage, while leaving a small indent on the FRC [6]. For more than four decades the composite industries have
impact surface, see Fig. 1. Such internal damages which are not visible been using TS resin system for the manufacturing of composites parts
on the surface of composites, degrade the material properties and re- and structures due to their ease of manufacturing and improved me-
duce the load carrying capacity of the composite structure by more than chanical/thermal properties. However, they have poor transverse
50% [1]. To mitigate these problems, large safety margins (extra plies) properties and non-recyclability due to the irreversible exothermic
are added to composite structures. However, these safety margins in- chemical reaction during the curing process [7]. The above-mentioned
crease the weight and reduce the competitiveness of the composite drawback associated with TS resin system, draw the attention of the
structures with metals. These issues raised the demand to improve the composite industry to use TP resin system, which has improved
impact resistance and damage tolerance of the composite structures toughness and recycling capability due to the physical change in the
under LVI. shape upon heating. In addition to this, TP composite offers other ad-
Two promising factors that improve the impact resistance and da- vantages such as rapid manufacturing, better chemical stability, higher
mage tolerance of FRC are fabric architecture and resin system [2]. modulus to weight ratio, low moisture absorption and less degradation
Based on these factors many studies were conducted to improve the due to ultra-violet radiations [8–10]. TP resins system have an infinite
Abbreviations: UD, unidirectional; BD, bidirectional; TS, thermoset; TP, thermoplastic; FE, finite element; OWC, orthogonal woven composites; LVI, low-velocity
impact; FRC, fiber reinforced composite; RTM, resin transfer molding; CAI, compression after impact; SEA, specific energy absorption; PPS, polypropylene; TAI,
tension after impact; FAI, flexure after impact; CDM, continuum damage mechanics; RVE, representative volume element; CDA, critical damage area; TTT, through
the thickness; AIWC, angle interlock woven composite; NWC, non-woven composite; PEEK, polyether-ether ketone; ASTM, American Society for Testing and
Material; AiTM, airbus industries test method; BVID, barely visible indentation depth; LaRC, Langley research center; GLARE, glass laminated reinforced epoxy
⁎
Corresponding author.
E-mail address: puteris@utp.edu.my (P.S.M. Megat-Yusoff).
https://doi.org/10.1016/j.compstruct.2019.03.021
Received 1 November 2018; Received in revised form 15 February 2019; Accepted 6 March 2019
Available online 07 March 2019
0263-8223/ © 2019 Elsevier Ltd. All rights reserved.
S.Z.H. Shah, et al. Composite Structures 217 (2019) 100–121
Fig. 1. Schematic drawing of internal damages under LVI.
shelf life and good thermal stability under extreme environment, which
make them suitable for defense, marine, and airplane engine applica-
tions [11]. Also, TP composite parts can be assembled through spot
welding, which is a suitable alternative to the mechanical fasteners and
gives equivalent load bearing capability and excellent joint stiffness
[12].
The prepreg based FRC requires manual layup and autoclave for
curing which is major resistance in high volume production of these
composites. Also, the cost of a prepreg is very high. To deal with this
problem the composite industry is looking for dry fiber preforms and TP Fig. 3. Schematic diagram of force-time, energy-time and deflection-time plot.
resin systems which can be used with conventional manufacturing
processes such as resin infusion or RTM process. Through a literature force-time, energy-time and deflection-time plots of a typical LVI. The
review, it has been observed that only a few studies are available in the force-time plot shows oscillations due to transient stress waves in the
case of 3D composites with TP due to their manufacturing difficulties. transverse direction. The y-axis shows the maximum peak force “Fmax ”
However, recently developed TP resin by Arkema called Elium® re- experienced by the FRC material; whereas, the x-axis shows the max-
solved these issues, as it is in liquid form at room temperature and can imum contact time “OA” between the FRC material and impactor. The
be used with conventional manufacturing techniques. velocity of the impactor at impact can be calculated using Eq. (1) [13],
Impact resistance and damage tolerance of FRC are affected by where, “V (t ) ”, “Vi ”, “F (t ) ”, “m” and “g” are velocity of the impactor at a
several factors, as shown in Fig. 2. In this review paper, these factors are time “t”, initial velocity, force at time “t”, the mass of the impactor and
divided into primary and secondary factors. The primary factors are the gravitational acceleration respectively.
fabric architecture and resin toughness which has the most significant
⎛ F (t ) ⎞ dt
t
effect on impact resistance and damage tolerance of FRC. Whereas, V (t ) = Vi + gt − ∫0 ⎝ m ⎠ (1)
secondary factors are hygrothermal conditions, stacking sequence, fiber
hybridization, matrix hybridization, impactor geometry (size, shape, The deflection in FRC with respect to time can be calculated using
and mass) and repeated impact. In the following section, the mechanics Eq. (2), where, “δ (t ) ” and “δi ” are deflection at a time “t” and initial
of LVI, the effect of the above mentioned primary and secondary factors deflection. Fig. 3 shows the schematic diagram of deflection-time plot
on impact resistance and damage tolerance of FRC will be discussed in obtained using Eq. (2) [13]. It is important to note that the maximum
detail. deflection “δmax ” does not occur at maximum force; however, it occurs
at the nose of force-deflection plot, shown in Fig. 4.
1.1. Mechanics of LVI gt 2 ⎛ F (t ) ⎞ dt ⎞ dt
t t
δ (t ) = δi + Vi t +
2
+ ∫0 ⎛∫0 ⎜
⎝ m ⎠ ⎠
⎟
(2)
⎝
The LVI test provides basic information in a form of force-time data.
The other important information such as energy absorbed by the Similarly, the energy absorbed by the FRC can be calculated using
composite, deflection in composite and the velocity of the impactor are Eq. (3) [13], where, E (t ) is the energy absorbed by the composite at any
evaluated from force-time data. Fig. 3 shows the schematic diagram of time “t”. The schematic diagram of the energy-time plot is calculated
Fig. 2. Factors affecting impact resistance and damage tolerance of FRC.
101
combination of damage initiation energy “Eini ” and damage propaga-

tion energy “Eprop ”. In “F (δ ) ” diagram the point “D” indicates the point
where impactor stops due to friction experienced by the impactor
during perforation, and the force becomes zero.
2. Effect of primary factors on impact resistance and damage

tolerance
2.1. Impact resistance of FRC
LVI events are very common in composite structures during their

service life, which produces internal damages such as matrix cracking,
fiber damage, and fiber/matrix interface de-cohesion. These internal
damages degrade material properties, which may lead to structural
failure. Therefore, it is essential to improve the impact resistance of the
composites so that damage caused by the LVI can be reduced. The
amount of damage caused by the LVI depends upon the fabric archi-
Fig. 4. Schematic diagram of force-deflection plot.
tecture, stacking sequence, fiber-matrix interface properties and shape
of the impactor. Kim [3] proposed various methods for improving the
using Eq. (3) is shown in Fig. 3. The energy-time plot shows the max- impact resistance of composites structures such as using tough matrix
imum impact energy “Emax ” experienced by the composite, which is the system, through thickness stitching, woven fabric, and hybrid fibers.
sum of elastic energy “Eel ” and absorbed energy “Ea ”. The rebounding The main objective of all the methodologies mentioned above is to
of the impactor in LVI is due to stored elastic energy, which indicates improve the interlaminar fracture toughness, reduce damaged area and
that only a part of the total energy is absorbed by the composite and the increase the residual strength of composites. Among many factors, two
rest of the energy is used by the impactor to rebound. major contributors in improving the impact resistance of composite are
fabric architecture and a resin system. In the following sections, these
E (t ) = m (Vi2 − V 2 (t )) + mgδ (3) factors will be discussed in detail.
Another important plot in the mechanics of LVI is the force-de- 2.1.1. Effect of fabric architecture on impact resistance
flection plot, which gives the amount of energy absorbed by the com- UD composites have been widely used in the aerospace industry in
posite (area under the force-deflection curve). The slope of the force- the manufacturing of wing box, fuselage, tail fin and wings due to their
deflection diagram gives bending stiffness of the composites, as shown excellent in-plane properties. However, they are poor under transverse
in Fig. 4. It also gives important information about the permanent de- loadings and undergo large delamination, fiber failure and matrix da-
flection (indentation) of the composite, which is represented by the mage [1,14]. Delamination in laminated composites is an alarming
points “G” and “H”. The Fig. 4 also shows different damage scenario defect as it is difficult to detect and it degrades the residual strength of
under LVI i.e. rebounding of the impactor, the maximum deflection of the composites by large amount [15]. To address the above-mentioned
composite without perforation, perforation of the impactor and the limitations of UD composites, 2D woven/bidirectional composites have
penetration of impactor, which are represented by the points “G”, “H”, been developed, which show better impact resistance due to the un-
“I” and “J”, respectively. In the case of rebound, the maximum de- dulation of yarns in the fabric architecture. However, their in-plane
flection is represented by point “E”. properties are lower due to stress concentration caused by the crimping
Fig. 5 shows the comparison between force-deflection “F (δ ) ” and of yarns [16]. To reduce crimping of yarns, NCF has been developed
energy-deflection “E (δ ) ” plots. The point “B” in “E (δ ) ” plot indicates which has improved impact and delamination resistance [17].
the energy at damage initiation (matrix cracking); whereas, the line To deal further with delamination problem in laminated compo-
“BA” represents the energy absorption during damage propagation sites, various methodologies have been adopted such as stitching, z-
(matrix crack, plasticization and fiber failure). Due to fiber failure at pinning, and weaving (introducing yarn in the z-direction). Stitching is
point ‘A” perforation of the impactor starts, as a result after this point widely used in the industry to improve transverse properties such as
the energy remains constant. The perforation energy “Eperf ” is the delamination resistance and interlaminar fracture toughness [18].
Francesconi and Aymerich [19] investigated the effect of through the
thickness stitching on the impact resistance and found that delamina-
tion has been reduced due to increased Mode-II fracture toughness
under compression. However, the stitching process produced manu-
facturing defects such as resin rich areas due to fish-eye generation,
which reduced the overall in-plane properties of the laminate. Tan et al.
[20] reported that through the thickness stitching reduces delamination
under LVI due to improved interlaminar fracture toughness. Z-pinning
is another promising method to improve transverse properties of la-
minates such as delamination resistance and interlaminar fracture
toughness in Mode-I and Mode-II [21–23]. However, more than 25%
reduction in the in-plane properties (compressive and tensile) are re-
ported due to Z-pinning [24].
With the development in weaving technology, 3D fabrics have been
developed, which have through the thickness reinforcement in the
fabric architecture. In 3D fabric, the warp and weft yarn mains in-plane
properties; whereas, transverse properties are improved by the in-
troduction of the reinforcements in the z-direction. This reinforcement
Fig. 5. Schematic diagram of force-deflection and energy-deflection plot. gives the fabric enhanced stability and excellent transverse properties
102
[25–27]. The through-thickness reinforcement not only increases the damage and possesses the highest damage tolerance, see Fig. 6(b). The
structural stability and delamination resistance but also increases the reported research on the effect of fabric architecture on impact re-
energy absorption capability of these materials due to higher intra-layer sistance is given in Table 1.
shear strength. Also, 3D composites increase ease in the manufacturing
as they are available in near net shape and in various thicknesses. 2.1.2. Effect of resin toughness on impact resistance
Hence, less manufacturing time is required in the layup of thick lami- Another important factor in improving the impact resistance of FRC
nates. In 3D composites, out of plane reinforcement in addition to in- is the resin toughness. In comparison with TS resin, TP resin possesses
plane reinforcement controls the overall energy dissipation and damage high toughness and strain to failure, which are desired properties in
evolution. Hence by changing the properties of through-thickness re- various applications i.e. impact, transverse loading, bolted joints and
inforcement, the performance of 3D composites improves significantly. flexible oil/gas pipelines. TP composites exhibit superior impact re-
In 3D composites, impact performance is dominated by fabric ar- sistance; however, they are more prone to permanent indentation. The
chitecture and arrangement of yarns. Selrzer et al. [28] found that 3D permanent indentation in TP composites is due to fiber breakage and
composites possess more than two times energy absorption capability as matrix plasticization, which absorb a large amount of impact energy. In
compared to 2D composites. Wang et al. [29] studied the 3D hybrid addition to this, fiber bridging in TP composites improves the Mode-I
basalt/Kevlar composite under LVI and found that layer by layer failure fracture toughness and reduces crack propagation. Also, TP resin keeps
in interlayer hybrid composites leads to high energy absorption; fiber orientations and distributes stresses among fibers which reduce
whereas, in intralayer composites, a sudden drop in the load leads to fiber damage [32]. As a result, damage extension and delamination in
lower ductility and energy absorption. Potluri et al. [30] proposed that TP-based composites are reduced.
the CAI strength of 3D composites can be improved further by reducing Various studies have been conducted on 2D composites with TP. It
the tow waviness in 3D fabric architecture. Bandaru et al. [31] studied was observed that the toughness of TP composites is ten times higher
the effect of fabric architecture in FRC and observed that by changing than those of TS composites [33], which improved their performance
the fabric architecture from 2D to 3D composite the tensile and impact under impact loading. Tan and Falzon [34,35] studied the crash-wor-
properties are improved due to better structural integrity. thiness of TP composites and found that TP composites demonstrated
These studies revealed that optimizing fabric architecture is the key 57.3% higher SEA as compared to TS composites. In another study,
to improve the impact resistance of FRC. Changing the fiber volume Striewe et al. [36] reported that the SEA of the TP composite is 20%
fraction, the density of Z-yarn, introducing off-axis yarns in addition to higher than that of steel. Santiago et al. [37] studied the TP-based fiber
on-axis yarns (warp and weft) not only improves the in-plane properties metal laminates and reported a 75% higher impact energy absorption as
(tensile, compressive and shear) but also, enhance the impact resistance compared to conventional TS based fiber metal laminates such as
and damage tolerance of FRC. Fig. 6 shows the effect of fabric archi- GLARE. Chen et al. [38] studied the indentation response of TP and TS
tecture on the impact resistance of FRC. Under LVI the UD composites laminates and found that TP composites experienced much higher
experience higher damage area as compared to BD and 3D composites, contact force and time as compared to brittle TS composites. Also, the
see Fig. 6(a). Hence, it has the lowest damage tolerance among different TP laminate dent depth has increased under LVI, which increased the
fabric architectures. Whereas, the 3D FRC experiences the lowest visibility during visual inspection. Kim et al. [39] reported that dela-
mination reduces the bending stiffness of the plate and increases the
deflection. The bending stiffness of TP laminate is low as compared to
TS laminate because TS laminate is more rigid. However, the TP com-
posite carries more load and undergoes large deformation as compared
to TS laminate [40]. This indicates that toughened composites exhibit
higher impact resistance due to maximum contact force and time.
Vieille et al. [41] investigated the impact performance of TP (PEEK and
PPS) and TS (Epoxy) composites at different impact energies. The au-
thors observed that fiber bridging in TP composites prevents crack from
opening in mode-I and slows down crack propagation due to mode II
and mode III. This fiber bridging reduces delamination and improves
the impact performance of composites.
Literature review reveals that a few studies are available for 3D TP
composites [42–44] as compared to 3D TS composites
[25–27,30,45,46] under LVI. Bandaru et al. [42,43] studied the impact
properties of polypropylene-based 3D hybrid composites in detail and
found that 3D TP composites absorb 26% more impact energy as
compared to 2D TP composites. Zhang et al. [44] studied polyethylene
based UD, 2D and 3D composites. The authors found that among all
configurations, 3D TP composites possess the highest impact resistance
and damage tolerance. Hart and Chia [25,45] studied the impact re-
sistance and residual strength of TS based 2D and 3D composites and
found that 3D composites show better impact resistance, less delami-
nation area, and stable residual strength due to through the thickness
reinforcement. Also, the less damaged area was found. Umer et al. [26]
studied different types of 3D TS composites and found that 3D ortho-
gonal composites exhibited better impact resistance. Gao and Kim [47]
studied the effect of cooling rate in TP composites and found that the
cooling rate effects the matrix ductility and Mode-II fracture toughness.
Fast cooling gives carbon/PEEK composites a higher impact resistance
than slow cooling rate.
Fig. 6. Effect of fabric architecture on impact resistance [30] (a) Damage area Fig. 7 shows the damage caused by the LVI in TP (carbon/PEEK) and
vs impact energy, (b) Impact resistance vs impact energy. TS (carbon/epoxy) based FRC. Due to higher toughness and strain to
103
Table 1
Reported research on the effect of fabric architecture on impact properties.
Reference year Material Comment
Richardson et al. [1] 1996 UD Composites Poor transverse properties under LVI
Baker et al. [16] 2004 2D woven composites Lower in-plane properties
Greve et al. [17] 2006 NCF composites Produce fisheye defects
Seltzer et al. [28] 2013 3D composites 2.5 times higher impact energy absorption
Francesconi et al. [19] 2017 Stitched Composites Produce resin rich areas and fisheye defects
Yasaee et al. [24] 2017 Z-pinned composites 25% reduction in the in-plane properties
Hart et al. [25,45] 2017 2D/Epoxy and 3D/Epoxy 3D composites show better impact resistance
Umer et al. [26] 2017 2D/Epoxy and 3D/Epoxy 3D composites show impact resistance
1978. Since then damage tolerance has been one of the main subjects of
investigation in the composite industry. Damage tolerance is the cap-
ability of composites structures to maintain their undamaged or initial
strength and modulus. Damage tolerance in composites is determined
by measuring the residual strength of the composite. The most common
method to determine damage tolerance of composites structure is to
produce BVID by LVI tests, followed by residual strength tests such as
CAI, FAI, and TAI tests. BVID is the minimum permanent indentation
that can be detected through the visual inspection, and the BVID limits
are 0.25 mm < BVID < 0.5 mm. Also, the permanent indentation
must be detected with the accuracy of 0.01 or less [48]. The two
standards used to measure the compressive residual strength (Com-
pression after impact) are ASTM D7137 and Airbus Method AiTM 1-
0010. Both these standards are used to measure the residual strength
after face-on impact. However, in the cases of edge impact (impact at 0°
and 45°) these standards are insensitive to damage. For such applica-
Fig. 7. LVI damage in TS and TP composites [47].
tions, combined loading compression fixture (ASTM D6641) is used
[49]. In the following section, the effect of primary factors on damage
tolerance will be discussed in detail.
failure of TP resin, less damage occurs under LVI in these composites.
Hence, carbon/PEEK possesses the highest damage tolerance. It is
concluded that higher the toughness of the resin system, the more en- 2.2.1. Effect of fabric architecture on damage tolerance
ergy will be absorbed under impact loading. This was confirmed by The fabric architecture has a significant effect on damage tolerance
comparing the SEA of TP and TS-based composites. It is important to of FRC. In the case of 2D FRC (UD and BD), LVI with reasonable impact
note that, all the studies were focused on the impact behaviour of 2D TP energies produces permanent indentations at impact location and de-
composites; whereas, TP based 3D composites have not been studied in lamination around the indentation, which divides the laminates into
detail, their damage development, failure mechanisms and energy ab- sub-laminates. These sub-laminates under compressive load produce
sorption modes are unclear. The reported research on the effect of resin global buckling [50]. Hence, the composite laminate fails due to the
toughness on impact properties is given in Table 2. propagation of transverse cracks, perpendicular to the loading direction
as shown in Fig. 8. These cracks travel from permanent indentation to
the specimen edge. It was observed that these transverse cracks origi-
2.2. Damage tolerance of FRC nate when the load reaches 80% of the failure load [51]. Among 2D
FRC, the UD composites experience higher damage area in the form of
The extensive use of composite materials in civil airplane structures matrix cracking and delamination, which results in the low residual
has raised the concern of safety certification agencies about the limited strength as shown in Fig. 9.
knowledge of composite materials and their behaviour. Damage toler- Several authors performed damage tolerance studies in 3D compo-
ance terminology was defined by the safety certification agencies in sites. Chiu et al. [52] determined damage tolerance of 3D angle
Table 2
Reported research on the effect of resin toughness on impact properties.
Gao and Kim. [47] 2001 Carbon/Epoxy, Carbon/PEEK TP composite improves impact resistance.
Chen et al. [38] 2006 Carbon/5222, Carbon/5228 TP has higher contact force and dent depth
Reyes et al. [54] 2010 PPS/Glass TP exhibits higher energy absorption
Zhang et al. [44] 2013 Polyethylene/Glass (2D and 3D) 3D composites show high impact resistance
Vieille et al. [41] 2013 Carbon/PEEK, Carbon/PPS Bridging effect improves impact resistance
Tan and Falzon [34,35] 2016 Carbon/PEKK TP composite has 57.3% higher SEA
Bandaru et al. [42,43] 2016 Basalt/Kevlar/Glass/PPS TP has 26% more energy absorption
Jung et al. [190] 2017 Polypropylene/E-Glass Less delamination area and impact resistance
Sorrentino et al. [58] 2017 PPS/Epoxy/E-Glass Multilayer TS/TP increases impact resistance
Santiago et al. [37] 2017 TP based fiber metal laminate TP has 75% higher energy absorption
Dogan and Arikan [38] 2017 TS/TP sandwich composite Impact on TP face improves impact resistance
Hart et al. [25,45] 2017 2D/Epoxy and 3D/Epoxy 70% reduction in FAI, 20% reduction of CAI
Umer et al. [26] 2017 2D/Epoxy and 3D/Epoxy 3D TP shows higher impact resistance
Striewe et al. [36] 2018 Glass/Polyamide TP has 20% higher SEA than steel
104
composites possess superior impact resistance, delamination resistance,

and damage tolerance properties. It was observed that, the compressive
residual strength of TP composites is twofold as compared to TS com-
posites. Reyes and Sharma [54] investigated the TP composite under
different impact velocities and found a 27% reduction in the flexure
strength. The authors found that in LVI, it is very common for TS
composites to undergo matrix cracking; however, in TP composites no
matrix cracking was observed. Vieille et al. [55] studied, PEEK and PPS
based TP composites under LVI followed by residual strength test. The
authors found that the residual strength of C/PEEK is 10% higher in
comparison to C/PPS and 40% higher in the case of C/epoxy. Fig. 9
shows the effect of resin toughness on damage tolerance of FRC. Due to
the brittle nature of TS composite, higher damage occurs at low impact
energies as compared to TP composites. Hence, the load bearing ca-
Fig.8. Global buckling under CAI test.
pacity of TS composites decreases and they failed at much lower stress
level. On the other hand, TP composites undergo less delamination and
damage area due to their higher interlaminar fracture toughness and
strain to failure which increases their residual strength. Fig. 9 shows,
the compressive residual strength in the case of TP composite decreased
by around 50%; whereas, in the case of TS composites it is decreased by
70%. Hence, the TP composite possesses two times higher residual
compressive strength as compared to TS composites.
Bull et al. [56] studied the effect of resin toughness in laminated
composites and observed that toughened FRC suppresses delamination
and exhibits higher damage tolerance as predicted by CAI test. Increase
in damage tolerance of toughened laminates is due to particle bridging,
which plays an important role in delamination resistance and delays
fiber buckling. In performing CAI, it was observed that under com-
pression this delamination starts to grow towards undamaged area of
the cone and creates sub laminates. These sub laminates failed upon
further loading due to fiber micro-buckling. This fiber buckling is due to
a sudden increase in the unsupported length due to delamination
growth which significantly reduced the load-bearing capacity of sub
Fig. 9. Effect of fabric architecture and resin toughness on damage tolerance laminates, see Fig. 10. Ishikawa et al. [57] performed CAI test on
(CAI) [189]. carbon/PEEK and compared the results with carbon/epoxy. In TS
composites, the CAI test showed that the buckling deflection reversed
interlock woven composites and compared the results with 2D com- its direction just before the failure and reduced the load bearing cap-
posites. They found that in 3D composites, cracks were arrested by the ability of the laminate. Whereas, in TP composites, the maximum stress
through-thickness yarns, which were followed by buckling/kinking. at buckling initiation is higher than TS composites. Also, the buckling
Whereas, in the 2D specimens, cracks were initiated at impact zone and propagation is very small in TP composites, which results in higher CAI
propagated through the specimen. The residual strength of the 3D strength.
composite was 16% higher than that of 2D composites. Chen and Sorrentino et al. [58] investigated the polypropylene/E-glass based
Hodgkinson [53] studied the CAI strength of UD, NCF and 3D compo- TP composites under LVI followed by FAI test. The authors optimized
site and found that 3D composites exhibited the highest damage tol- the stacking sequence of the hybrid polypropylene matrix to get better
erance among all three composites. Potluri et al. [30] investigated the flexure properties and impact resistance. Gao and Kim [47] investigated
CAI behaviour of different FRC and found that 3D composites exhibited the effect of cooling rate in TS and TP composites. The authors found
the highest residual strength. The authors observed that if damage size that slow cooling rate improves the fiber/matrix interface properties,
is below the critical value, then there is no noticeable difference in the strength, and modulus. Whereas, fast cooling rate introduces matrix
residual strength of 3D composites. Hart et al. [25] performed FAI and ductility, and increases the interlaminar fracture toughness which gives
CAI test on 2D and 3D composites and found a large reduction in the composites a high damage resistance and damage tolerance. TP com-
flexure strength/modulus as compared to compressive strength/mod- posites have high toughness, but the carbon/PEEK cylinder shows low
ulus which were 70% and 20%, respectively. However, in 3D compo- residual strength after impact. This indicates that manufacturing
sites, less reduction in the residual strength was observed due to parameters must be optimized to take advantage of the toughness of TP
through-thickness reinforcement which resisted delamination and composites to improve the residual strength of composite structures
cracks propagation. This indicates that measuring the residual strength [59]. Based on the experimental data, the authors correlated the re-
using FAI test is very important as it is more sensitive to impact da- sidual strength with the impact energy given by Eq. (4), where, “σr ” and
mage. Damage tolerance of 3D composites can be improved further by “σ0 ” are the residual strength and undamaged strength, “U0 ” and “U” are
reducing the tow waviness in 3D fabric architecture. The reported re- the threshold and impact energy. The exponent “β ” represents the re-
search on the effect of fabric architecture on damage tolerance is given duction in the residual strength.
in Table 3.
σr U β
= ⎛ 0⎞
σ0 ⎝U ⎠ (4)
2.2.2. Effect of resin toughness on damage tolerance
The impact damage performance of FRC depends on two main It can be concluded that TP based composites exhibit higher damage
properties i.e. resin toughness (strain to failure) and Mode-II fracture tolerance due to the bridging effect. This phenomenon resists Mode-I
toughness (interlaminar fracture toughness). Due to this reason the TP opening and crack propagation, which reduces damage extension and
105
Table 3
Reported research on the effect of fabric architecture on damage tolerance.
Chiu et al. [52] 2004 2D and 3D Composites 3D composites have 16% higher residual strength
Chen and Hodgkinson [53] 2009 UD, NCF & 3D composites 3D composite has the highest damage tolerance
Gonzalez et al. [93] 2011 UD composites Buckling of sub laminates
Potluri et al. [30] 2012 UD, 2D & 3D Composite 3D composite has better damage tolerance.
Hart et al. [25,45] 2017 2D/Epoxy and 3D/Epoxy 70% reduction in the residual strength of 2D composites and 20% reduction in 3D composites
interlaminar fracture toughness of composites [34,62]. Baere et al. [63]

studied the Mode-I fracture toughness of TP and TS composites and
found that the interlaminar fracture toughness of TP-based composites
is two times higher. Mode-I and Mode-II interlaminar fracture tough-
ness of TP (PEEK) and TS (epoxy) was investigated in detail by Frie-
drich et al. [64] and it was found that Mode-I and Mode-II fracture
toughness of TP composites is ten times higher. Kim and Ye [65] stu-
died the effect of temperature on Mode-I and Mode-II fracture tough-
Fig. 10. Schematic drawing of LVI followed by CAI test, (a) Permanent in- ness of TP composites and found that fracture toughness increases with
dentation due to LVI, (b) Fiber micro-buckling under compression. temperature. Hinkley [66] studied the effect of resin toughness and
observed that the interlaminar fracture toughness was improved by the
toughened resin.
delamination in TP composites. In the case of TS composites, debris
In addition to this, the accuracy of FE simulations depends on ac-
produced due to fiber and matrix damage prevents the crack and de-
curate values of intralaminar and interlaminar fracture toughness for
lamination to close upon unloading. It is important to note that TP
progressive damage modeling based on CDM and fracture mechanics
based composites are more prone to permanent indentation. These
[67]. To accurately predict delamination’s behaviour of laminated
permanent indentations facilitate the impacted specimen to buckle
composites, different ASTM standards are used to measure interlaminar
under compressive load. This is the main reason why TP-based com-
fracture toughness such as ASTM D5528-01 for Mode-I (Double canti-
posites do not possess significantly high CAI strength as compared to TS
lever beam) test, ASTM D7905/D7905M for Mode-II (End node flexure)
composites. The reported research on the effect of resin toughness on
test and Mode-III (Mixed mode bending) test. Intralaminar damage
damage tolerance is given in Table 4.
modes are classified into fiber dominant and matrix dominant damage
modes. For accurate prediction of fiber dominant damage modes,
3. Effect of secondary factors on impact resistance and damage longitudinal tensile and compressive fracture energies are required.
tolerance These fracture energies can be determined from compact tension test
ASTM E399 and compact compression test ASTM E1820 which are
In the following section, the effect of secondary factors on impact normally used for metals. For matrix and shear dominant failure modes,
resistance and damage tolerance are discussed in detail. since the behaviour of composites is highly nonlinear, and the com-
posites undergo plastic deformation especially in the case of transverse
3.1. Effect of fracture toughness impact loading, shear fracture toughness is required which can be
calculated using the V-notched shear test according to ASTM D7078/
Laminated composites are manufactured using stacking of plies in D7078M-12.
different orientations followed by curing and consolidation process, Tan and Falzon [35] observed that the determination of shear
resulting plies sticking together and showing excellent in-plane prop- fracture toughness is the key to accurately predict nonlinear shear da-
erties. However, they are weak under transverse loading and undergo mage modes. Tan and Falzon [67] also found that interlaminar and
separation of plies called delamination. Such behaviour of laminated intralaminar fracture toughness is very important for the accuracy of FE
composites is due to their poor interlaminar fracture toughness, which simulation. The authors used experimentally determined fracture en-
reduces the service life of composite structures. Over the decade various ergies for interlaminar and intralaminar damage prediction and ob-
techniques have been adopted to increase the interlaminar fracture served that the accuracy of the results was improved. Fiber compressive
toughness of laminated composites through reinforcement in the failure is one of the main intralaminar damage modes under LVI and
transverse direction such as stitching, z-pinning and 3D weaving compression after impact test. Francesconi and Aymerich [19] reported
[24,60,61]. It was observed that such modification in the fabric archi- that to accurately predict delamination area and size, exact values of
tecture increases the interlaminar fracture toughness by two to three Mode-II fracture energies are required because through the thickness
times as compared to unmodified fabric. compression improves the fracture energies under LVI. This phenom-
It is important to note that not only the change in fabric architecture enon reduces delamination area in composites and improves the impact
but also the use of a toughened resin system will improve the resistance and residual strength. The effect of compressive fracture
Table 4
Reported research on the effect of resin toughness on damage tolerance.
Gao and Kim [47] 2001 Carbon/epoxy, PEEK Fast cooling increases matrix ductility & residual strength
Reyes et al. [54] 2010 PPS/Glass 12–27% reduction in FAI
Vieille et al. [55] 2014 Carbon/PEEK, PPS Residual strength is 10% higher for PPS, 40% for PEEK
Bull et al. [56] 2014 Carbon/TS/TP 30% higher damage tolerance due to improved toughness.
Sorrentino et al. [58] 2017 PPS/Epoxy/E-Glass Increases damage tolerance and impact resistance.
Sonnenfeld et al. [104] 2017 PPS/Epoxy/E-Glass Multilayer TS and TP composite increases impact resistance.
106
energies in fiber compressive failure was discussed in detail by Hon- repeated impact location and found that damage progress more quickly
gkarnjanakul et al. [68]. The authors observed that compressive frac- when the distance is small.
ture energies must be determined to achieve a good correlation be- Another important factor is the effect of temperature on the re-
tween experimental and simulation results. The authors determined peated impact because it changes the matrix behaviour and overall
fiber compressive fracture energies using energy release rate per unit failure mechanisms of composite structures. Atas and Dogan [72] in-
volume of a continuum element, which is given by Eq. (5), where, vestigated the effect of thermal aging (humidity and temperature) on
“G1fC ”, “ε ”, “ε1”, “S” and “V” are the fiber fracture toughness, strain along the repeated impact performance of glass/epoxy composites. The au-
fiber direction, strain at final failure along fiber direction, element thors observed that increasing the aging time decreases the amount of
cross-section normal to fiber direction and element volume. energy absorbed and fatigue life of the composites. Also, the repeated
1 ε1 impact of aged specimens shows a decrease in the perforation energy by
∫V (∫0 )
σdε . dV = SG1fC
(5) 3 times when the specimens are exposed to the extreme environment
for 1300 h. The total amount of energy absorbed was 919 J as compared
It can be concluded that the experimentally determined inter- to 292 J by the aged sample. Icten et al. [73] determined the effect of
laminar and intralaminar fracture energies are key to accurate predic- low temperature on repeated impact. The authors found that at low-
tion of damage in FRC materials. Specifically, in the case of LVI and temperature, matrix behaves in a brittle manner and produces higher
crashworthiness of composites structures, where not only fracture en- contact forces, lower contact duration and higher energy absorption as
ergies along longitudinal and transverse directions are important but, compared to room temperature. In addition to this, Karthikeyan et al.
the shear fracture energies play a critical role in predicting such com- [74] simulated the multiple impact events, where the velocities of
plex behaviour. In addition to this, the fracture energies may vary by multiple impacts were varied to establish a relationship between impact
the change in environmental conditions such as a change in the tem- energies (survival and penetration of projectile). Arikan and Sayman
perature, humidity, and moisture. The effects of such extreme condi- [75] performed repeated impact tests at 50 J on TS and TP composites.
tions on the fracture energies must be considered for better prediction. The authors reported that TP composites improve the impact perfor-
The reported research on the effect of fracture toughness on impact mance of composites due to their lower bending stiffness and capability
resistance and damage tolerance is given in Table 5. to undergo large deformation as compared to TS resins. Due to this
large deformation in TP composites, stresses are transferred over a large
3.2. Effect of repeated impact area, hence they absorb more energy.
In actual scenarios, impact in composites structures under LVI are
The repeated impact events are very common in the marine and random by nature and depends on various factors such as impact en-
aerospace structures. Yacht and offshore structures experience repeated ergy, impact velocity, shapes of objects/tools and mass. Hence, com-
impact from different hard objects in addition to waves. In the case of posite structures may face repeated impacts, which are a combination
repeated impact, each impact event enhances damage in composite of above-mentioned factors. Whereas, all the studies mentioned above
structures in the form of matrix damage, fiber damage and delamina- are based on the repeated impact at the same energy level which is an
tion. ideal case. It is important to study the effect of above-mentioned
Baucom et al. [69] studied the effect of repeated impact on different random factors on the impact and post-impact performance of FRC.
fabric architecture. The authors found that the dominant failure me-
chanisms in 2D composites are delamination, fiber tensile failure, and
matrix cracking. Whereas, in the case of 3D composites, the dominant 3.3. Effect of impact geometry
failures are due to matrix cracking and fiber/matrix debonding which
distribute damage over a large area and absorb more energy. Under The geometry of the impactor has a significant effect on the impact
repeated impact, the main damage mode in 3D composites is the performance of composites in terms of impact resistance and damage
straining of Z-yarn in addition to pulling out of weft yarn from the tolerance. Literature review reveals that the majority of the LVI tests
surface, which is absent in 2D composites. This indicates that by were conducted using hemisphere impactor; however, in practical
changing the fabric architecture, the impact resistance can be im- scenarios, the object or flying fragments that impact the composite
proved. Santos et al. [70] investigated the effect of the inclined hole and surface vary in size, shape, and mass. Hence it is important to consider
a vertical hole under repeated impact and reported that the presence of these effects in predicting LVI performance of composites. The change
vertical or inclined holes degrades the impact properties. However, the in the diameter of the impactor changes the contact area between the
inclined hole has a more adverse effect due to high-stress concentration. impactor and specimen, which affects the LVI performance. Icten et al.
In another study, Santos et al. [71] studied the effect of distance from [76] investigated the effect of impactor diameter on impact resistance
Table 5
Reported research on the effect of fracture toughness on impact resistance and damage tolerance.
Reference Year Comments
Friedrich et al. [64] 1988 Mode-II fracture toughness of TP is ten times higher
Hinkley [66] 1990 TP improves the fracture toughness and impact resistance
Kim and Ye [65] 2004 Fracture toughness of TP increases with temperature
Pankow et al. [61] 2011 Higher Mode-II fracture toughness improves impact resistance
Baere et al. [63] 2012 Fracture toughness of TP is two times higher than TS
Hingkarnjanakul et al. [68] 2013 Compressive fracture toughness effects the residual strength
Tan et al. [67] 2016 Accurate values of fracture toughness improve prediction
Tan and Falzon [34] 2016 TP improves the fracture toughness and crashworthiness
Tan and Falzon [35] 2016 Nonlinear shear damage depends on the shear fracture toughness
Francesconi and Aymerich [19] 2017 Transverse compression improves fracture toughness and impact resistance
Yasaee et al. [24] 2017 Z-pinning improves fracture toughness and delamination
Shiino et al. [60] 2017 Stitching improves Mode-I fracture toughness
Abir et al. [177] 2017 Mode-I and Mode-II effects the residual strength of composites
Jung et al. [190] 2017 Mode-I and Mode-II fracture energies are key to accurate damage prediction
Liu et al. [62] 2018 TP improves the fracture toughness which increases impact resistance
107
Fig. 11. Different types of indenter/impactor used for LVI test.
and damage tolerance of composites under LVI. The authors reported Fig. 12. Cone damage formation in laminated composites under LVI.
that by decreasing the diameter of the impactor, the impact resistance
of composites is decreased, and more damage occurred at the same
plate as compared to “C” channel geometry. This large damage in flat
energy. Whereas, the perforation limit of the specimen is increased by
plate could be due to low bending stiffness as compared to “C” channel
increasing the diameter of the impactor. At same impact energy, com-
geometry. The authors also studied the effect of stacking sequence and
plete perforation was observed with 12.7 mm impactor; whereas, no
found that the geometry of the specimen is still dominant and produces
perforation was reported with 38.1 mm impactor. Damage tolerance of
large damages in a flat plate, irrespective of the layup. This lower da-
the composites will be increased by increasing the diameter of the
mage area indicates that c-shaped geometry will possess higher damage
impactor, however, it decreases with increasing impact energy. The
tolerance. It is concluded that size and shape of the geometry (impactor
different shape of impactor geometries used for LVI is given in Fig. 11.
and specimen) affects the impact resistance and damage tolerance of
Another important factor is the shape of the impactor. Sevkat et al.
FRC; whereas, changing the mass of the impactor has a negligible effect
[77] studied the effect of impactor shape on the impact performance of
on the impact performance and residual strength of FRC.
hybrid composites. The authors considered four different types of im-
pactor i.e. spherical, cylindrical with a flat head, diamond shape, and
Charpy straight line impactor. Impact properties were affected by the
3.4. Effect of stacking sequence
geometry of the impactor in terms of peak force, contact duration, and
initial peak force. The authors reported that the contact area is the key
In laminated composites, LVI produces internal damages in the form
parameter which affects the energy absorption of the composites i.e.
of cones due to matrix cracking and delaminations. Matrix cracking
impactor with a large contact surface produced the highest initial peak,
occurs due to high transverse stresses under the impactor; whereas,
large delamination, and maximum force. Similarly, Mitrevski et al. [78]
delaminations are initiated at the boundaries of the cone as shown in
investigated the effect of shape and reported that composite materials
Fig. 12. It is well known that delamination occurs in between plies
can absorb more energy when the conical impactor is used.
having different orientations. As discussed, the impact resistance and
Artero et al. [79] studied the effect of impactor mass on the LVI
damage tolerance can be improved by improving interlaminar fracture
performance of composites. The authors used three different impactor’s
toughness, which depends on the stacking sequence and interface
mass i.e. 1.15, 1.59 and 2.04 kg. It was reported that the performance of
properties. Hence, changing the stacking sequence suppress or delay the
composite laminates under LVI is insensitive to the mass of the im-
interlaminar fracture by changing the load from tensile to compressive.
pactor. Hence, the maximum force, residual strength and maximum
Several researchers studied the effect of stacking sequence under LVI
displacement are functions of impact energies only and they do not
and observed that by changing the stacking sequence, the impact re-
depend on the velocity or mass of the impactor. Several other re-
sistance and damage tolerance of the laminated composite can be im-
searchers studied the effect of mass under LVI and draw the similar
proved [2,87–90]. Wang et al. [91] studied delamination behaviour of
conclusion that damage caused by the LVI depends on the impact en-
different stacking sequences and found that accurate prediction of de-
ergy only and it does not depend on the mass of impactor [80–82]. Also,
lamination is the key in predicting damage tolerance of composites
the change in the mass of the impactor has no effect on the residual
structures. Bouvet et al. [92] found that damage initiation was due to
strength of composites. All these studies revealed that mass has no ef-
high energy release rate in Mode-I; whereas, delamination’s propaga-
fect on the impact resistance and damage tolerance of the composite
tion was governed by the Mode-II energy release rate due to excessive
structures. However, some researchers observed the effect of impactor
shear stresses at the interface.
mass under LVI, when the impact energy is constant [83,84]. In these
Many authors studied the effect of stacking sequence on the impact
studies, it was observed that the mass affects the amount of damage and
resistance of composites [28,93–95]. Jang et al. [96] reported that by
delaminations. This difference in the effect of mass is due to the strain
changing the stacking sequence, delamination and plastic deformation
rate sensitivity of the laminated composites where interlaminar damage
can be reduced. Sevkat et al. [77] studied the effect of stacking se-
mode is dominant. When intralaminar damage such as fiber failure was
quence in hybrid composites and observed that the layup sequence
initiated, the effect of mass was nullified. Borea et al. [85] studied the
plays an important role in improving the impact resistance of FRC. The
effect of mass on self-reinforced composites and observed that high
authors found that GP/GL/GP has the highest initial peak and low
velocity and low mass produced higher vibrations in the specimen as
maximum force. Benli and Sayman [97] studied the effect of stacking
compared to high mass and low velocity. In addition to this, increasing
sequence and temperature under LVI and observed that both these
the mass increases the amount of energy absorbed by the composites.
factors have a catastrophic effect in terms of damage area and energy
Apart from the geometry of impactor, the impact resistance and
absorption. Similarly, Strait et al. [98] observed that the stacking se-
damage tolerance also depend on the geometry of the specimen. The
quence has a significant effect on impact resistance at higher impact
ASTM standards used to measure impact resistance (ASTM D7136) does
energies. Freitas et al. [99] found that damage area in laminated
not consider geometry effects and use standard specimen geometry i.e.
composites is highly dependent on the stacking sequence, which pro-
flat plate of (100 × 150 mm). Recently, Gliszcynski et al. [86] studied
duces variation in the stress distribution. Aktas et al. [100] investigated
the effect of specimen geometry and boundary conditions on the LVI
single and multiple layered glass composites and found that the main
performance of FRC. The authors studied flat-plate and “C” channel
parameters decreasing the CAI strength of the composite are the impact
geometry and found that damaged area due to LVI is strongly depen-
energy and compressive residual strength, which decrease when the
dent on the geometry of the specimen i.e. more damage is found in flat-
impact energy is increased.
108
In composites, orientation and stacking sequence of plies can be

optimized to obtain the best stiffness and load-bearing capacity; how-
ever, these laminates may have poor impact resistance or damage tol-
erance. Lopes et al. [101] carried out a detailed investigation on the
effect of stacking sequence and proposed a methodology to improve
impact resistance and damage tolerance while keeping the same
bending stiffness. The orientation angle between two consecutive plies
was restricted which reduced the interlaminar stresses and damage area
and improved the impact resistance and damage tolerance. Other au-
thors such as Dost et al. [102] found that stacking sequence has a sig-
nificant effect on damage tolerance. Reis and Freitas [103] observed
that increasing the number of the interface increases damage tolerance
of laminated composites. Rivallant et al. [90] studied the effect of
stacking sequence on damage tolerance and found that by optimizing
the stacking sequence damage tolerance can be improved. Sonnenfeld
et al. [104] proposed a methodology to improve damage tolerance of
composite by optimizing the stacking of TS and TP plies. The authors Fig. 13. Effect of stacking sequence on impact resistance [97], L1 [90/0/0/90]
observed that the addition of compatible TP interface layers between TS s, L2 [0/90/45/45]s and L3 [0/90/45/-45]s.
and TP material increases the fracture energies by 15 times and reduces
delamination area by 2.5 times. A similar conclusion was drawn by
3.5. Effect of environmental conditions
Freitas et al. [99], i.e. by changing the stacking sequence and impact
energy, the damaged area is increased which decreases damage toler-
The impact resistance and damage tolerance of FRC are affected by
ance of the FRC.
the extreme conditions i.e. low temperature, high temperature, hu-
A detailed study of the coupling effect of stacking sequences on
midity and hygrothermal aging etc. These factors affect the perfor-
damage tolerance was carried out by Baker et al. [105]. The authors
mance of the composite especially in the case of impact loading.
used benchmarking configurations (symmetric, anti-symmetric and
Therefore, it is important to study the performance of FRC materials
non-symmetric) studied by York et al. [106] and found that the sym-
under extreme conditions before they are considered for such applica-
metric laminates are not a good choice in the case of damage tolerance
tions. The extreme conditions change the overall failure mechanisms
design; however, anti-symmetric laminates give the highest damage
and performance of composite structures. It was observed that the in-
tolerance among all three laminate designs. Also, damage tolerances of
crease in the temperature reduces the thermal stresses. In FRC (TS and
non-symmetric and anti-symmetric configurations are close to each
TP) as the temperature reaches the curing temperature, it results in the
other, which indicates that non-symmetric laminates are not a bad
weak fiber/matrix interface properties and improves absorbed energy.
choice for damage tolerance designs. It was found that adding the
The weak fiber/matrix interface properties result in fibers pull out
number of plies does not guarantee higher damage tolerance and it
under transverse loading. This phenomenon dissipates a large amount
reduces the load-bearing capacity of the composites due to large Pois-
of energy and improves the LVI performance. However, at low tem-
son’s ratio effect. This attributes to damage initiation and delamination
perature, in TS composites the fiber/matrix interface becomes stronger
propagation in laminated composites, which depends on the Mode-I
due to matrix contraction, which makes these composite brittle. As a
and Mode-II energy release rate. This energy release rate is significantly
result, it produces a higher contact force, lower energy absorption and
affected by the ply orientation especially in the case of mechanically
contact duration as compared to impact at room temperature. Hence,
coupled laminates. The standard ASTM standards i.e. ASTM D7905 and
more damage will occur, and impact resistance of composite will be
ASTM D5528 used for measuring the fracture toughness do not consider
reduced. Whereas, in TP composites at low temperature, the impact
the effect of geometry. Samborski [107,108] studied the effect of ply
resistance is improved due to increase in the matrix ductility. Fig. 14
orientation on the Mode-I and Mode-II fracture toughness and found
shows the effect of temperature on the damage initiation and damage
large variations in the values due to the coupling effect (bending-ex-
propagation energy in TS composite. As the temperature decreased
tension/extension-twisting/shearing-bending coupling). Hence, the
from 90 °C to −50 °C the energy required to initiate damage is reduced
fracture toughness measured for 0° ply laminate is not valid for angle-
due to brittle behaviour of FRC at low temperature. Hence, more da-
ply laminate.
mage will occur at low impact energy.
Fig. 13 shows the effect of stacking sequence on the damaged area
Hygrothermal aging degrades the material properties and affects the
of FRC. The symmetric cross-ply (L1) laminate has the highest damage
LVI performance of FRC. Parvatareddy et al. [109] studied the impact
area, when the impact energy exceeds 25 J. Hence damage tolerance of
symmetric-cross ply laminate will be the lowest. Whereas, the anti-
symmetric (L3) laminate has the lowest damaged area and it will pos-
sess the highest damage tolerance. The symmetric angle-ply has an
intermediate damage area. It is important to note that, damage modes
are also affected by the thickness of plies. In the case of thin ply la-
minate, damage starts from the bottom plies due to bending stresses,
which travels upward to the impact face. Whereas, in the case of thick
ply laminate, the matrix cracks first initiate at the top plies due to high
contact stresses and travels downward to form a pine tree. It is con-
cluded that the stacking sequence, orientations and thickness of plies
affect delamination and damaged area in laminated composites, which
play a significant role in the impact resistance and damage tolerance
performance of FRC.
Fig. 14. Effect of temperature on damage initiation energy [97].
109
resistance and damage tolerance of FRC exposed at 150 °C and observed strength, as stress concentrations and overstress were reduced. Russoa
that the residual strength was decreased by 70–75%. Li et al. [110] et al. [113] studied the effect of low temperature on the polyurethane-
investigated the effect of hygrothermal aging and observed that da- based composite under LVI. The authors observed that at −50 °C (lower
maged area was increased under LVI which decreased the post-impact than the glass transition temperature of the matrix i.e. −40 °C) the
performance of composites. The authors observed that the combination ductility of the matrix was increased, and no delamination occurred. In
of humidity and temperature decreased the residual strength (TAI) another study, Russoa et al. [114] observed that TP composite at low
exponentially. Mortas et al. [111] investigated the aging of composites temperature showed better damage tolerance due to the absence of
through the corrosive solution at elevated temperature, which affects delamination, which is a common damage mechanism in TS composite.
the residual strength of the FRC. Atas and Dorgan [72] studied the ef- Similarly, Boumbimba et al. [115] investigated TP (Elium®) composites
fect of hygrothermal aging under repeated LVI and found that in- under LVI at ± 80 °C and observed that at low temperature the impact
creasing the aging time decreases the absorbed energy and fatigue life resistance of TP composites had been increased as compared to room
of composite by three times. Icten et al. [73] found that at −50 °C, temperature. Recently, Wang et al. [116,117] studied PPS/carbon
epoxy showed brittle behaviour which results in higher contact force composite at elevated temperatures. It was observed that the increase in
and damage area. the temperature decreases the stiffness, damaged areas, and delami-
It is important to note that a combination of stacking sequence and nation. At elevated temperatures, delamination was decreased by 57%;
temperature may have an adverse effect on LVI performance. The however, the permanent indentation was increased by 15–40% due to
coefficient of thermal expansion along the longitudinal direction is very matrix ductility. Dubary et al. [2] found that at 150 °C delamination
small compared to the transverse direction, which results in different was reduced by 3 times and the impact energy required to produce
thermal stresses in laminates. Benli and Sayman [97] studied the effect BVID was reduced by 24% which increased damage detection. The
of temperature and stacking sequence on LVI. At low impact energies, main reason for the above observations is the material ductility and
matrix cracking, laminate bending, and matrix plastic deformation toughness at an elevated temperature. Sorrentino et al. [118] in-
were reported; whereas, at high impact energies, fiber breakage was the vestigated the LVI performance of different TP (Polyethylene 2,6
dominant failure. Fig. 15 shows the effect of temperature on the impact naphtholate) composites. The authors found that TP/basalt composite
resistance of FRC at different impact energies. Fig. 15(a) shows, at low possesses the highest impact resistance.
temperature i.e. −50 °C FRC undergoes higher damages due to brittle It is concluded that, in the case of TS and TP composites, at high
behaviour. Hence, at this temperature, they exhibited the lowest impact temperature the matrix ductility increases which improve the impact
resistance, as compared to impact at room temperature or at 90 °C as resistance; however, this increase in the matrix ductility increases the
shown in Fig. 15(b). BVID and decreases damage tolerance of FRC. At low temperature, the
TP composites have better resistance against temperature and hu- TS composites show brittle behaviour and undergo more damage area
midity. Vieille et al. [112] studied the effect of extreme conditions which degrades the impact resistance and damage tolerance of FRC.
(temperature and moisture) on the TS and TP composites. It was ob- Whereas, in TP composites at low temperature, the impact resistance
served that TP composite had better capability to retain their me- and damage tolerance are improved as compared to room temperature.
chanical properties after hygrothermal aging and showed improved This attributes to matrix plasticity and the absence of delamination. It is
important to note that, at elevated temperature the behaviour of com-
posites changes, which affects the Mode-I and Mode-II fracture tough-
ness. The fracture toughness must be investigated at elevated tem-
perature to observe the effect of temperature on them. In addition to
this to the best of authors’ knowledge, no study has been reported re-
garding the effect of extreme conditions on LVI performance of 3D TP
woven composites, which is an interesting area to explore. The reported
research on the effect of environmental conditions on LVI is given in
Table 6.
3.6. Effect of fabric hybridization
Another important factor in improving damage tolerance and im-

pact resistance of FRC is fabric hybridization because impact resistance
is dominated by fabric architecture and the arrangement of yarns in it.
Composite structures are made of highly stiffed yarns such as Kevlar,
carbon, glass and graphite. Each type of yarns has advantages and
disadvantages e.g. graphite and basalt have high stiffness but lacks in
toughness, Kevlar has high toughness, glass is the most commonly used
fiber due to low cost etc. Recently Safri et al. [119] published a detailed
review about natural and synthetic hybrid composite focusing on the
impact resistance and penetration. In hybrid FRC, two or more fibers
are combined through stacking or weaving process, normally a high
stiffness/cost fiber is combined with low stiffness/cost fibers. The main
objective of hybridization is to combine the load carrying capability of
high modulus fibers with the toughness of low modulus fibers. Hy-
bridization improves damage tolerance, impact resistance and reduces
the material cost.
Several researchers investigated the 2D hybridized composites. Naik
and Meduri [120] studied the interlayer hybridization of UD and woven
composites. The authors observed that the impact resistance of hybrid
Fig. 15. Effect of temperature on FRC at different impact energies [97]. (a) laminate was improved. Sevkat et al. [121] investigated the glass/
damage area vs impact energy, (b) impact resistance vs impact energy. graphite hybrid composite and found that glass layer on the outer side
110
Table 6
Reported research on the effect of environmental conditions on impact resistance and damage tolerance.
Reference Year Environmental conditions Comments
Parvatareddy et al. [109] 1996 +150 °C, Ambient Residual strength was decreased by 70–75%
Li et al. [110] 2000 +50 °C, +100 °C Residual strength decreases exponentially
Benli and Sayman [97] 2011 −50 °C, +20 °C, +90 °C Decreasing temperature increases impact energy
Vieille et al. [112] 2012 85% Humidity, 120 °C TP improves the impact resistance
Mortas et al. [111] 2014 HCL, NaOH, +60 °C Residual strength and elastic energy decreases
Atas and Dogan [72] 2015 70% Humidity, +95 °C Energy absorption is decreased by 3 times
Icten et al. [73] 2015 −50 °C Perforation limit was increased
Russoa et al. [113] 2016 −50 °C, +25 °C At low temperature no delamination was found
Ma et al. [191] 2016 199 K, 100 K Small damage and poor energy absorption
Russoa et al. [114] 2017 −50 °C, +25 °C Lower temperature increases the stiffness
Boumbimba et al. [115] 2017 −80 °C, +80 °C Low temperature improves impact resistance
Dubary et al. [2] 2017 +150 °C Delamination is reduced by 3 times
Sorrentino et al. [118] 2017 +60 °C, +100 °C Flexure strength was reduced
Wang et al. [27] 2017 +180 °C, +90 °C At +90 °C, impact resistance was improved by 6.4%
Wang et al. [116,117] 2018 +125 °C BVID was increased by 40.2%, delamination decreased by 57.5%
improved the impact properties of graphite; however, it increased the Table 7

weight of the laminate. Dehkordi et al. [122,123] studied the basalt/ Reported research on the effect of fabric hybridization on impact resistance and
nylon hybrid composites and found that by changing the nylon content, damage tolerance.
the failure mode changed from fiber breakage to delamination. At low Reference Year Hybridization Configuration
impact velocities, varying the content of nylon has no effect on the
impact performance of hybrid composites. However, at high impact Wang et al. [29] 2008 Basalt/Kevlar 3D Composites
Sevkat et al. [121] 2009 Glass/Graphite 3D Composites
energies, damage area and energy absorption depend on nylon/basalt
Sun et al. [192] 2009 Glass/Kevlar 3D Composites
content. Sarasini et al. [124] investigated synthetic and natural hybrid Dehkordi et al. [122,123] 2010, 2013 Basalt/Nylon 2D Composites
composites and found that the hybrid composites improved the impact Sarasini et al. [124] 2016 Carbon/Flax 2D Composites
resistance and the residual strength. The authors reported that flax/ Bandaru et al. [42,43] 2016 Glass/Basalt/Kevlar 3D Composites
carbon/flax hybrid laminates possessed the highest impact resistance Bandaru et al. [31] 2017 Glass/Basalt/Kevlar 3D Composites
Dubary et al. [2] 2017 Carbon/Glass 2D Composites
and damage tolerance measured through FAI test. Dubary et al. [2]
studied damage tolerance of PEEK-based carbon/glass hybrid compo-
sites. The authors observed that permanent indentation has a negligible achieved by these modifications is limited. One of the promising
effect on residual strength, and specimens failed due to global buckling. methods to improve impact properties is through fabric hybridization.
These studies revealed that hybridization has not always ensured im- However, fabric hybridization is a complicated process to optimize
provement in the impact resistance. Therefore, the properties of in- fabric architecture and to get better results. In recent years, a new
dividual yarns and arrangement of yarns in the architecture must be approach was introduced in which hybridization of the matrix
carefully considered. [126–128] was used instead of fabric hybridization. In this method,
In the hybridization of 3D composites, Wang et al. [29] investigated different fillers (micro or Nano) are added into TS or TP resin to im-
intralayer and interlayer basalt/Kevlar hybridized 3D fabric under LVI. prove the resin toughness, which will improve the fracture toughness
The authors found that interlayer hybrid composites showed higher and impact resistance of FRC. The main advantage of this approach is
ductility and specific energy absorption due to layer by layer failure that the fabric configuration and design factors remained unchanged.
mechanisms of the composite laminate as compared to intralayer hy- In this context, Boumbimba et al. [115] studied the effect of tri-
bridization where failure was abrupt. Bandaru et al. [125] studied the block copolymer on the impact resistance of FRC. The acrylic triblock
polypropylene-based basalt/Kevlar hybrid composites and found that copolymer (Nanostrength) was added to the TP acrylic resin (Elium®) to
hybridization increased the modulus by 8–9%, tensile strength by improve the toughness. The authors observed that adding Nanostrength
8–23% and failure strain by 8–23%. In another study, Bandaru et al. in TP resin improved the peak load, energy absorption and damage in
[43] reported that the Kevlar/basalt hybrid composite has 7.67–48.49% composites. The impact resistance was increased by 24% at 80 °C;
higher energy absorption in comparison with homogeneous basalt and whereas, at low temperature (−80 °C) penetration threshold was in-
Kevlar fabric. creased by 20%, and less damage area was found. Nash et al. [129]
As discussed, fabric architecture is the primary factor in improving studied bulk resin modification by introducing the TP phase into TS
the impact and post-impact performance of composites. Hence, opti- composites. It was observed that damage resistance and damage tol-
mizing the fabric architecture through fabric hybridization is an ef- erance is improved. Borea et al. [85] investigated the self-reinforced TP
fective way to improve the impact resistance and damage tolerance of composites and found that unlike TS composites, TP composites showed
composites under LVI. However, more research is needed to understand ductile behaviour and extended plasticity which increased the energy
the LVI behaviour and damage development in TP based 3D woven absorption of self-reinforced composites.
hybrid composites. The reported research on the effect of fabric hy- Sorrentino et al. [58] studied matrix hybridization using poly-
bridization on LVI and residual strength is given in Table 7. propylene films made up of polypropylene MA712 (Non-compatible
“N”) and polypropylene with maleic anhydride (Compatible “C”). The
authors observed that grading of fiber/matrix interface was an effective
3.7. Effect of matrix hybridization
method to improve damage tolerance under LVI, which is confirmed
from the impact tests at 6 and 27 J. Among all configurations, asym-
In composites, stiff fibers are used due to their better strengths;
metric matrix arrangement HCN (50% C and 50% N) exhibited the
however, they have less toughness and poor under transverse loading.
highest impact resistance. Bull et al. [56] studied different TP laminates
To improve transverse properties especially in the case of LVI, various
manufactured using a different percentage of TP particles in TS resin.
methodologies have been adopted such as the layup sequence, fabric
The CAI results showed approximately 30% higher damage tolerance as
architecture, laminae thickness. However, the degree of improvement
111
compared to untoughened laminates. Hence, resin modification (in- modeling based on fracture mechanics and CDM to predict the beha-
troducing TP in TS resin) and resin hybridization (combination of TS viour and damage modes in LVI [133–138]. Lannucci and Willows
and TP plies) is an effective way to improve the toughness of FRC, [133,139] proposed a damage model based on CDM. The authors used
which reduced damage and delamination caused by the matrix damage strain-based damage evolution law in laminated composites. The in-
and fiber/matrix interface de-cohesion. This not only increases the elastic strains were considered in this model based on the cyclic tensile
impact resistance of composites but also improves damage tolerance of test. This model was implemented with a shell element and did not
FRC. consider the through-thickness effects. Johnson et al. [95] proposed a
CDM based model, where cyclic tests data was used to obtain stiffness
4. Impact damage prediction in FRC reduction as damage grows. Damage initiation was based on damage
model proposed by Williams et al. [140], in which the authors con-
Analytical solutions developed to predict LVI behaviour of FRC have sidered both the in-plane and transverse stresses. However, in this
limitations, due to the complex damage state and oversimplified as- model permanent deformation, mesh sensitivity and strain rate effect
sumptions. These issues were addressed by finite element methods, were not considered. Maimi et al. [141] proposed a CDM based damage
which enabled us to solve impact problems with realistic boundary model in which LaRC04 failure criteria was used for damage initiation,
conditions. Also, the development of interlaminar and intralaminar which is implemented in Abaqus through user subroutine UMAT. Lopes
damage models coupled with CDM, the theory of plasticity and fracture et al. [142] extended damage model proposed by Maimi et al. to 3D ply
mechanics enhanced the capabilities of finite element methods to pre- model and successfully predicted different failure modes such as dela-
dict the progressive damage and nonlinear behaviour of composites. mination area, fiber failure, and matrix cracking. However, the accu-
Most of the progressive damage models available for composites are racy of the results decreased by the increase in impact energy. In this
based on shell elements, where the out of plane components (normal model, nonlinear transverse shear effects, strain rate effects and de-
and shear) are neglected. This assumption is due to the small thickness pendency of fracture energy on fiber orientation were not considered.
of composites as compared to the in-plane dimensions. However, this is Sevkat et al. [121] observed that the stress/strain curve of glass/gra-
not valid in the case of transverse/impact loading. This is because the phite showed nonlinearity and Chang-Chang model [143] (linear
impact load is parallel to the normal direction of shell elements, and model) is not appropriate to predict such nonlinearity. The authors
delamination is one of the main damage modes, which depends on the implemented a nonlinear orthotropic material model which is a non-
normal and shear components. Hence, three-dimensional finite element linear version of Chang-Chang model using user subroutine. The pro-
models give more realistic results under transverse loading conditions. posed model improved damage prediction and behaviour of compo-
Damage modeling under LVI is divided into four different categories sites.
i.e. failure criteria-based damage model, continuum damage me- Several researchers used Hashin 3D failure criteria to predict in-
chanics-based models, fracture mechanics-based models and theory of tralaminar damage in FRC under LVI [144,145]. Guo et al. [146] used
plasticity or yield surface-based damage models. In failure criteria- Hashin 3D failure criteria and exponential damage evolution function
based models, stress or strain based polynomial expressions are used to to avoid a sudden decrease in the stiffness which may cause stiffness
define the failure envelope of damage initiation. However, this ap- matrix singularities. This proposed damage model predicted LVI be-
proach cannot identify the position of crack, size of cracks and pro- haviour; however, this model did not consider the progressive damage
gressive damage in composites. These problems were addressed by the in the laminate. Kim et al. [39] used Hashin 3D failure criteria and
fracture mechanics-based models in which the energy required to create strain energies to accommodate the nonlinear effects in damage in-
a new hole was used. However, this approach has limitations as it re- itiation. Damage evolution based on Weibull distribution and material
quires an initial flaw/crack. Damage models based on plasticity are softening law proposed by Lannucci [139] were considered. Similarly,
used for ductile composites such as 3D woven composites and TP-based Maio et al. [94] used Hashin 3D failure criteria for intralaminar damage
composites. CDM approach originally proposed by Kachanov [130] and initiation and exponential law proposed by Matzenmiller [147] for
Rabotnov [131] to study creep failure in metals is used in predicting damage evolution. The proposed damage model predicted delamination
progressive damage in composites. This approach can be combined in a form of peanut shape and size, aligned along the fiber direction.
with failure criteria and fracture mechanics to predict progressive da- Zhang et al. [148] used 3D failure criteria proposed by Hashin and Hou
mage in composites. However, it requires a reliable testing data as an et al. [149] to predict failure in braided composites. Topac et al. [150]
input. In the following section, different modeling strategies to predict used CDM based damage model, in which the authors used LaRC04
intralaminar and interlaminar damage in FRC due to LVI will be dis- [151] failure criteria for damage initiation and damage evolution based
cussed in detail. on equivalent stress/equivalent displacement proposed by Matzen-
miller. The LaRC04 was implemented in Abaqus using user subroutine
4.1. Intralaminar damage prediction VUMAT. Numerical simulation predicts that damage process consists of
the following steps i.e. a) Matrix failure (Primary cracks) due to impact,
Intralaminar damage model consists of fiber dominant damage b) delamination failure initiation from primary cracks tips, c) diagonal
modes (fiber failure under longitudinal tension/compression), matrix crack initiation (Secondary cracks), d) delamination connects with
dominant failure modes (matrix failure under longitudinal tension/ secondary cracks to complete propagation of damage. The reported
compression) and shear failure mode. The schematic diagram of in- research on the effect of different CDM used for damage prediction in
tralaminar damage modes in FRC is shown in Fig. 16. Several re- LVI is given in Table 8.
searchers used failure criteria-based damage model to predict in-
tralaminar damage in FRC. Freitas et al. [99] used Tsai Hill failure 4.1.1. Nonlinear damage modeling in FRC
criteria to predict the matrix cracking due to in-plane transverse As discussed LVI produces permanent indentations on the surface of
stresses. Similarly, Naik and Meduri [120] used Tsai Hill failure criteria composites with non-visible internal damages. These permanent in-
to predict matrix cracking in 3D composites. Menna et al. [132] used dentations are responsible for the degradation in the post-impact
Tasi-Wu failure criteria for intralaminar damage initiation. However, properties. To predict such behaviour a nonlinear damage model is
the predicted results were inaccurate because these models are linearly required. Bouvent et al. [92] proposed a plastic-like damage model to
elastic and did not consider the progressive damage of composites. predict permanent indentation in UD composites. The authors observed
Unlike in metals, damage in FRC is not abrupt; however, it is a that matrix crack debris were responsible for permanent indentation.
combination of different damage modes which may occur as damage The proposed plastic-like damage model captures the permanent in-
grows in composites. Several researchers used progressive damage dentation and damage modes i.e. matrix cracking, delamination and
112
Fig. 16. Schematic diagram of intralaminar damage modes.
fiber failure under LVI. However, this model requires some parameters captured with matrix plasticity, other failure modes must be considered
from LVI data, which limits its application to different cases. to capture permanent indentation using simulations. Das et al. [154]
The LVI and quasi-static indentation response were studied by predicted the indentation response of 3D non-woven fabric. A simpli-
Abdallah et al. [152]. The authors observed that damage process in fied model was used, which only considered the effect of fibers along
laminated composites starts with the primary matrix cracks on the non- the loading direction.
impacted side of the specimens. The secondary cracks start and pro- To predict permanent indentation in FRC, a nonlinear damage
pagate at 45° along the transverse direction. These primary and sec- model is required. The nonlinear behaviour of fiber-reinforced com-
ondary cracks cause failure in the bottom plies and permanent in- posites can be represented by two ways i.e. a) The nonlinear shear
dentation at the impact location. Caprino et al. [153] proposed an based damage model, where a polynomial expression is developed by
empirical law based on indentation and penetration energy to predict the curve fitting of experimental shear stress/shear strain data, b)
the indentation depth as given by Eq. (6) Considering the theory of plasticity and plastic flow rule proposed by
Sun and Chen [155], where a single parameter is used to introduce
β
U nonlinear behaviour. This single parameter is determined by the off-
I = I0 ⎜⎛ ⎟⎞ axis tensile test [156,157].
U
⎝ p⎠ (6)
where, “I”, “U” and “Up ” are indentation depth, incident energy and a) Nonlinear shear-based damage models
penetration energy. The two constants “I0 ” and “β ” were determined
experimentally. However, this model is valid when the contact condi- It has been established that, under transverse or shear loading, the
tions between the indenter and material are described by the Hertzian UD composites undergo large nonlinear shear deformation due to fiber/
contact law. Chen et al. [38] observed that permanent indentation in matrix interface de-cohesion, matrix cracking, and matrix plasticity.
laminated composites is a coupling between permanent indentation and The behaviour of composites under transverse loadings such as crushing
fiber failure. The authors suggested that, since this coupling cannot be of composites, low-velocity impact and bolted joints lead to the shear
Table 8
Reported research on the continuum damage modeling to predict LVI.
Reference year Damage initiation Damage evolution
Lannucci et al. [133,139] 2006 Strain-based criteria Strain-based linear softening

Maimi et al. [141] 2007 LaRC04 failure criteria Based on damaged activation function
Johnson et al. [95] 2009 Strain-based criteria Strain-based damage evolution
Lopes et al. [142] 2009 Stress-based damage activation Based on damaged activation function
Sevkat et al. [121] 2009 Max stress-based criteria Strain-based criteria for element deletion
Guo et al. [146] 2013 Hashin 3D failure criteria Exponential damage evolution law
Kim et al. [39] 2013 Hashin 3D failure criteria Exponential damage evolution law
Maio et al. [94] 2013 Hashin 3D failure criteria Exponential damage evolution law
Zang et al. [148] 2015 Hashin 3D failure criteria Linear softening based on Disp. equ.
Hassoon et al. [144] 2017 Hashin 3D failure criteria Linear softening based on Disp. equ.
Topac et al. [150] 2017 LaRC04 failure criteria Linear softening based on Disp. equ.
113
predict the nonlinear shear behaviour of FRC. Shi et al. [138,162]

proposed a CDM based model using Hashin [163] and Puck [164]
failure criteria, and nonlinear shear behaviour was predicted by the
semi-empirical relationship, given by Eq. (8), which was originally
developed by Berbinau et al. [165], where, “Sij ” and “Gij0 ” represent the
ultimate strength and initial shear modulus. The authors found that
matrix cracking and delamination were the main failure modes ob-
served in LVI and the finite element simulation efficiently predicted
both damage modes. However, this model underpredicted the results at
low impact energies.
0
⎡ ⎛ Gij γij ⎞ ⎤
τij (γij ) = Sij ⎢1 − exp ⎜− , no sum over i, j… i, j= 1, 2, 3
Sij ⎟ ⎥
⎣ ⎝ ⎠⎦ (8)
Feng and Aymerich [166] studied delamination area, size, and or-
ientation in laminated composites. The authors considered a strain-
based nonlinear damage model, where nonlinear shear behaviour was
Fig. 17. Shear stress vs shear strain curve. included by curve fitting of a cubic polynomial. The proposed pro-
gressive damage model will be able to predict delamination at the in-
behaviour [35,135,145,158]. terface of each ply at all impact energies. Chiu et al. [167,168] studied
To predict the shear behaviour of FRC, a damage model based on the crushing behaviour of laminated composite by considering shear
plasticity and damage theories was presented by Ladeveze and nonlinearities and proposed a single damage variable to control shear
LeDantec [159], in which stability criteria was used for damage in- damage under transverse compression. Tan et al. [158] used the above
itiation. Based on the cyclic test data, the authors proposed a damage model proposed by Chiu et al. to predict the LVI behaviour and residual
evolution law to represent nonlinear shear behaviour of composites. strength of laminated composites. The proposed model successfully
Donadon et al. [134] improved the nonlinear shear model proposed by captured the impact behaviour and permanent indentation which play a
Ladeveze and LeDantec. The shear stress “τij ” vs shear strain “γij ” re- significant role in predicting residual strength.
sponse was presented by a cubic polynomial given by Eq. (7). In another study, Tan and Falzon [34,35] proposed a nonlinear
damage model in which degradation in the shear modulus was coupled
τij (γij ) = a1 γij + a2 γij2 + a3 γij3, no sum over i, j … i, j= 1, 2, 3 (7) with plastic deformation which was then used in the constitutive re-
where, the constants a1, a2 and a3 are determined by fitting polynomial lationship. The authors used the exponential model to represent the
expressions to experimentally determined stress/strain curve. The shear stress/strain relationship and degradation in shear modulus “Gij∗”
polynomial expression representing the nonlinear shear behaviour is is given by Eqs. (9) and Eq. (10), where, “τijY ” is the yield strength. The
shown in Fig. 17. The new model was based on CDM and fracture constants “α ”, “β ” and “P1 − 4 ” control the strain hardening, elastoplastic
mechanics which required few factors and the use of single damage transition and degradation in the shear modulus, respectively.
variable to control the nonlinear shear behaviour. In addition to this, Y
⎧ τij [exp(αγij ) − exp(βγij )], γij > 0
the authors addressed the strain localization and mesh dependencies τij (γij ) = no sum over i
problem due to material softening in progressive damage modeling. ⎨ τijY [−exp(−αγij ) + exp(−βγij )], γij < 0
⎩
Van Paepegem et al. [160,161] investigated the shear behaviour using
, j i, j = 1, 2, 3, (9)
off-axis ( ± 45°) tensile test of laminate and developed a damage model
to predict nonlinear shear behaviour. The off-axis configuration to Gij∗ (γij ) = P1exp(P2 γij ) + P3exp(P4 γij ) no sum over i, j i, j = 1, 2, 3 (10)
predict shear behaviour in 2D FRC is shown in Fig. 18. Based on ex-
perimentally determined nonlinear shear behaviour, the degradation in The proposed damage model accurately captures the crushing be-
the shear model and accumulation of inelastic shear strain were in- haviour of TP composites by combining nonlinearities during shear
troduced in the damage model which efficiently predicted the shear deformation. Liu et al. [62] used nonlinear shear model proposed by
nonlinearity. Tan et al. [158] and predicted the behaviour of UD and BD hybrid
Faggiani and Falzon [135] modified the incremental damage composites under LVI. The authors used three failures criterions for
scheme used by Donadon et al. [134] and proposed a two-phase non- matrix compression i.e. Hashin, Puck, and Chang-Chang and found that
linear shear model based on plasticity and CMD to predict nonlinearity Puck failure criteria showed better prediction among all three criteria
and progressive damage in FRC. The shear stress vs shear strain re- due to the accommodation of nonlinear shear effects in damaged
sponse was presented by a cubic polynomial given by Eq. (7). The model.
constants in the equation were determined by the curve fitting of ex- The scope of previous studies in the literature has been dedicated to
perimentally determined shear stress/shear strain data, as shown in the nonlinear shear behaviour of 2D composites (TS and TP resin) to
Fig. 17. This proposed model successfully captured interlaminar da- predict their LVI impact behaviour. However, a few experimental stu-
mage, intralaminar damage, and permanent indentation. In another dies have been reported to determine the nonlinear shear behaviour of
studied Falzon and Apruzzese [136,137] used implicit formation to 3D TS composite. Lomov et al. [169] compared the off-axis damage
mechanisms of 2D and 3D composites. The authors observed that 2D
composites showed large nonlinear shear behaviour due to “scissoring”
effect of warp and weft yarns. Whereas, in 3D composites, rotation of
warp and weft yarns at Z-yarn locations, which act as a hinge is re-
sponsible for large shear strains. Also, 3D composites possess 25–30%
higher strain to failure as compared to 2D composites. Saleh et al. [170]
investigated the off-axis response of different types of 3D woven com-
posites and found that 3D OWC possesses the highest failure strain and
strength. Recently, Labanieh et al. [171] introduced two biased yarns
Fig. 18. Off-axis configuration of 2D FRC to predict shear behaviour. ( ± θ°) in conventional 3D OWC and studied their effect under tensile
114
− − −
F (σij, εP ) = 3f (σij ) − R (εP ) (12)
Several researchers used the theory of plasticity based damage

model to predict the nonlinear behaviour of composites under LVI to
calculate inelastic strains and strain rates [156,157,172,173]. Hao et al.
[172] used critical damage area for damage initiation and elastoplastic
model to include strain rate effect and successfully predicted the be-
haviour of 3D composites under high strain rate loading. Singh et al.
[156,174] used Hashin and Puck failure criteria for damage initiation
and theory of plasticity proposed by Sun and Chen for nonlinear effects.
The authors found that, damaged induced plasticity model showed
better damage prediction as compared to the elastic damaged model.
Liao and Liu [157] used Puck failure criteria for damage initiation and
elastoplastic model to predict the nonlinear behaviour. They found that
at low impact energies i.e. 2–5 J both elastic damage model and plastic
damage model prediction were the same; however, when the impact
energy was increased the elastic model cannot predict the sudden load
Fig. 19. Off-axis configuration of 3D FRC to predict shear behaviour.
drop. Noticeable plastic deformation and fiber damage were reported at
higher impact energies, which indicated that plastic-damage model
loadings. The authors performed loading/unloading tests along warp, must be used while considering high impact energies. Ren et al. [175]
weft and off-axis specimens to predict their stress/strain behaviour. used the theory of plasticity and Hill failure criteria to predict damage
The off-axis configuration to predict shear behaviour of 3D FRC is modes in 3D composites under LVI, i.e. matrix cracking, fiber damage,
shown in Fig. 19. Among the three specimens, off-axis specimens and interface failure. Das et al. [154,176] coupled theory of plasticity
showed higher nonlinear behaviour and strains to failure. To predict and predicted the compression and indentation of non-woven compo-
high strain to failure and nonlinear behaviour in 3D composites it is sites. However, the model was oversimplified which overpredicted the
recommended to predict such nonlinear behaviour of off-axis specimens results. The reported research on the theory of plasticity-based damage
and incorporate them in damage model to predict damage accurately. model is given in Table 10.
To the best of authors’ knowledge, no study is available in the literature
regarding the nonlinear shear behaviour of 3D orthogonal TP woven
4.2. Interlaminar damage prediction
composites. The reported research on the nonlinear shear damage
models used to predict the nonlinear behaviour of FRC is given in
Delamination is one of the main damage modes in laminated com-
Table 9.
posites under LVI. In 3D composites, this delamination occurs in the
form of fiber/yarns interface decohesion. Delamination modeling in
b) Theory of plasticity-based nonlinear damage models
FRC is divided into three categories i.e. failure criteria-based model,
fracture energy-based model, and cohesive zone modeling.
The nonlinear behaviour of FRC can be determined using the theory
Delamination modeling using failure criteria is computationally very
of plasticity. Damage in FRC starts with the formation of microcracks.
efficient. On the other hand, the fracture energy-based approach re-
These microcracks redistribute the stresses to undamaged area, which
quires knowledge of the crack position and it is computationally very
starts to deform elastically or plastically. A flow rule based on the single
expensive especially in the case of the 3D model. Cohesive zone mod-
parameter was proposed by Sun and Chen [155] to represent the
eling is the most widely used delamination modeling approach in which
nonlinear/plastic behaviour of FRC. Eq. (11) gives the plastic potential
− − the interface is modeled independently, and it does not require the
function “ f (σij ) ” in a simplified form; where, “σij ” represents the stress
knowledge of the crack position. In cohesive zone modeling, both da-
tensor of impregnated tows in the local coordinate system (1, 2, 3). The
mage initiation and damage propagation are modeled separately. The
parameter “a 66 ” is sensitive to plastic deformation and represents the
failure criteria in terms of interface stresses are used to predict dela-
nonlinear shear behaviour of FRC under transverse compression or
mination/decohesion initiation; whereas, the fracture mechanics-based
shear loading. Eq. (12) gives the yield function “F” in terms of plastic
− approach is used to predict delamination evolution.
potential function and equivalent plastic strain “εP ”.
Several researchers used surface-based (zero-thickness) cohesive
− 1 − − −2 −2 −2 zones between plies to predict delamination/decohesion
f (σ ) = [(σ22 − σ33 )2 + 4σ23 + 2a66 (σ13 + σ12)]
2 (11) [34,62,148,150,177]. This modeling scheme was based on master/slave
Table 9
Reported literature to predict nonlinear shear behaviour of composites.
Reference Year Material Intralaminar damage model
Van Paepegem [160,161] 2006 E-glass/Epoxy –

Donadon et al. [134] 2008 E-glass/Epoxy Strain-based criteria, Puck and Shurmann
Faggiani and Falzon [135] 2010 Carbon/Epoxy prepreg Strain-based criteria, Puck and Shurmann
Falzon and Apruzzese [136,137] 2011 Carbon/Epoxy prepreg Stress-based criteria, Puck and Shurmann
Shi et al. [138] 2012 Carbon/Epoxy prepreg Hashin, Puck, and Shurmann
Shi et al. [162] 2014 Carbon/Epoxy prepreg Hashin, Puck, and Shurmann
Feng and Aymerich [166] 2014 Graphite/Epoxy Strain-based criteria
Chiu et al. [167] 2015 Carbon/Epoxy prepreg Strain-based criteria, Puck and Shurmann
Tan et al. [158] 2015 Carbon/Epoxy prepreg Strain-based criteria, Puck and Shurmann
Chiu et al. [168] 2016 Carbon/Epoxy prepreg Strain-based criteria, Puck and Shurmann
Tan and Falzon [34] 2016 Carbon/PEKK Strain-based criteria, Puck and Shurmann
Tan and Falzon [35] 2016 AS4/PEKK Strain-based criteria, Puck and Shurmann
Liu et al. [62] 2018 Carbon/Epoxy Hashin, Puck, and Shurmann
115
Table 10
Reported literature to predict nonlinear behaviour using the theory of plasticity.
Reference Year Material Intralaminar damage model
Hao et al. [172] 2008 3D Glass OWC Critical damage area

Chen et al. [173] 2014 AS4/PEEK Hashin Failure criteria
Singh et al. [156,174] 2015 Graphite/Epoxy Hashin, Puck, and Shurmann
Liao and Liu [157] 2017 Graphite/Epoxy Strain-based criteria, Puck and Shurmann
Ge et al. [193] 2018 3D Braided carbon/Epoxy Hashin 3D
Ren et al. [175] 2018 3D angle interlock/Epoxy Hill failure criteria
Das et al. [154,176] 2018 3D non-woven composites Hill failure criteria
surfaces, which obeys bilinear traction separation/displacement law macro-level models, 3D composites are modeled with orthotropic ma-
and it is computationally very efficient, as given by Eq. (13). terial model. These models are computationally efficient but they are
less accurate. Meso level models give more detailed predictions because
ti = Ki δi no sum over i i = n, s, t (13)
in these models, fiber, matrix and fiber/matrix interface are considered
where, “ti ”, “δi ” and “Ki ” represent the traction stress vector, separation separately. The matrix is considered as linear elastic or elastoplastic.
vector and interface stiffness matrix, respectively. The symbols n, s and However, these models are computationally very expensive. The micro-
t represent the normal, shear and tangential direction. This traction/ macro model is an efficient way to model 3D composites. In this ap-
displacement law was defined in terms of interface stiffness (normal proach micro model in a form of the unit cell (RVE) is used to calculate
and shear) which controls the displacement between two surfaces. Prior effective properties and global compliance matrix of the unit cell, which
to damage initiation, to eliminate any relative displacement at the in- is considered at each integration point. This approach is computation-
terface, the interface/penalty stiffness was initially set to a large value ally very efficient and provides good accuracy. The micro-macro model
to avoid this problem [178]. can be coupled with the theory of plasticity or nonlinear shear model to
Now there are different choices for damage initiation and evolution predict the nonlinear behaviour of 3D composites. This damage model
criteria. Damage initiation based on quadratic stress-based failure cri- can be used for various loading conditions where the composites un-
teria was proposed by Phino and Davila [151], in which the normal and dergo large deformations such as LVI, bolted joints, off-axis notched/
traction stresses at the interface were used to evaluate damage initia- unnotched and on-axis notched/unnotched specimens.
tion. Several researchers used quadratic stress failure criteria for dela- The main failure mode in 3D composites is fiber/matrix interface
mination initiation [166,177]; whereas, maximum stress failure criteria debonding, which decreases the rigidity of the structures. In FE simu-
for damage initiation was proposed by Singh and Mahajan [156]. The lation, the decrease in the rigidity is introduced through stiffness re-
quadratic and maximum stress failure criteria are given by Eqs. (14) duction, hence it is very important to use proper damage evolution law.
and (15), respectively, where, “tn∗”, “ts∗” and “tt∗” represents the peak Elias et al. [46] used Onera damage model to predict the LVI behaviour
stresses. Damage is initiated, when one of the stress ratios reaches 1. of 3D woven composites. This proposed model accurately captures
2 2 2 damage pattern in terms of indentation depth and indentation area,
⎧ 〈tn 〉 ⎫ + ⎧ ts ⎫ + ⎧ tt ⎫ = 1 which shows good agreement with experimental data. However, in this
∗ ∗ ∗
⎨
⎩ tn ⎬ ⎭ ⎨
⎩ ts ⎬
⎭ ⎨
⎩ tt ⎬⎭ (14) damage model, 3D composites are modeled as orthotropic material
hence no fiber/matrix debonding and cracks can be observed. Bandaru
〈t 〉 t t
max ⎧ n∗ , s∗ , t∗ ⎫ = 1 et al. [43] used Chang-Chang model [143] to predict the behaviour of
⎨
⎩ tn ts ⎬
tt ⎭ (15) different 3D TP hybrid composites and found that 3D Kevlar/basalt
Once the damage is initiated, the interface stiffness is decreased hybrid fabric has the highest energy absorption capability. The authors
according to Eq. (16). Where, the constants K i0 and “D” represent the proposed that the behaviour of 3D TP composites under LVI is unclear
undamaged stiffness and scalar damage variable. The scalar damage and needs to be investigated further in terms of testing and FE simu-
variable gives the overall damage at the interface due to normal and lations. It is important to note that, the Chang-Chang model is a linear
shear stresses. elastic damage model; whereas, the behaviour of 3D TP is nonlinear. To
predict such behaviour nonlinear damage model with appropriate da-
Ki = (1 − D) K i0 i = n, s, t (16) mage evolution scheme must be used.
The interlaminar damage evolution criteria are divided into two Hao et al. [172] predicted the behaviour of 3D composites using
categories, (a) damage variable is defined in terms of relative dis- critical damage area and theory of plasticity to predict nonlinear be-
placement [138,146] and (b) damage variable is defined in terms of haviour. The CDA is calculated based on critical damaged length “δ ”
energy required for delamination to grow. For transverse impact given by Eq. (17)
loading, the energy-based approach gives a better prediction, in which k 1
k+1
the area under traction/displacement curve must be equal to the frac- ⎛ Xf ⎞ (k + 1) L ⎞ k + 1
δ = rf ⎜ Y ⎟ ⎜⎛ ⎟
ture energy [156,173]. The critical energy at complete failure and 4τ 2rf (17)
⎝ f ⎠ ⎝ ⎠
failure initiation involved in the formulation can be evaluated using
mixed mode theory. Two widely used methods based on mixed mode where, “rf ”, “ Xf ”, “τ fY ”, “k” and “L” are tow radius, fiber tensile
theory are Benzeggagh-Kenane law [179] and power law [180]. More strength, fiber yield strength, Weibull constant and length of tow. The
details on cohesive zone modeling for the prediction of the impact proposed model cannot predict different damage modes occurring in 3D
behaviour of composites can be found in the detailed review by Abrate composites due to simplified damage model. Munoz et al. [182] used
et al. [181]. maximum stress failure criteria to predict the failure in warp and weft
direction. The authors used an embedded element technique to model
5. LVI damage prediction in 3D FRC TTT yarns. However, this model did not consider the fiber/matrix in-
terface and resin rich areas. Turner et al. [183] used Hashin failure
Damage prediction in 3D composites requires a detailed finite ele- criteria [163] for damage initiation and damage evolution proposed by
ment model. The different approaches used to model 3D composites are Matzenmiller et al. [147] to predict the behaviour of 3D composites
i.e. macro-level model, meso-level model and, micro-macro models. In under LVI. The authors considered yarns as shell element, and tie
116
constraints were used to avoid the solid element model, which is very found that the results are in good agreement as reported by Tan et al.
expensive in terms of computational time. The proposed 3D FE model [158]. Abir et al. [177] used a single FE model to perform LVI followed
captured the dominant failure modes and deflection in 3D OWC. by CAI test. The authors used maximum stress and Tsai Wu failure
It is established that 3D composites possess high strain to failure as criteria for damage initiation. It was observed that failure under CAI
compared to 2D composites and show nonlinear behaviour. Potluri was due to local buckling and delamination growth. Also, the important
et al. [30] performed residual strength tests and found plasticity in 3D parameters that affect the residual strength of composites were the
composites under compression. This indicates that the elastoplastic Mode-II interlaminar fracture toughness and fiber compressive fracture
material model is required to predict such nonlinear behaviour. To the toughness. Increase in the Mode-II interlaminar fracture toughness re-
best of authors’ knowledge, no study is reported for the predictions of duces delamination size and increases damage tolerance.
such nonlinear behaviour in 3D TP composites. Also, all the available Rivallant et al. [90] and Hongkarnjanakul et al. [68] improved the
models for 3D composites are based on the theory of plasticity. How- progressive damage model proposed by Bouvent et al. [92] in which
ever, no study is reported on nonlinear shear-based damage models for they successfully captured the permanent indentation. Fiber failure
3D composites. Such models are expected to accurately predict the under compression which is the most important failure mode under CAI
behaviour of 3D TP composites under LVI. was added to the damaged model. CAI simulations efficiently captured
the crack propagation and buckling of sub laminates which were pro-
6. Damage tolerance prediction in FRC duced due to impact. The FE predictions were compared with experi-
mental data in terms of failure modes, impact behaviour and residual
The accurate prediction of damage tolerance in FRC plays a sig- strength. Tan et al. [158] proposed a damage model based on nonlinear
nificant role in reducing the weight of aerospace structures, especially shear. The authors coupled the matrix tensile and compressive failure
in civil aircraft. These weight margins are included in the aerospace criteria proposed by Puck and Schurmann [164] and Catalanoti et al.
structures to sustain the degradation of the material properties under [188] to predict compressive residual strength. The authors used three
LVI events. To deal with this problem, designers consider strength de- steps process to perform CAI simulation i.e. perform LVI simulation,
sign. The strength design of composite structures mainly depends upon stabilize the specimen and modify boundary conditions and perform
the compressive residual strength properties. The prediction of com- CAI simulation. The proposed damage model captures the permanent
pressive residual strength properties enables the designers to optimize indentation, damage modes and residual strength with high accuracy.
their designs. It is important to note that this residual strength test is performed on
Only a few authors are able to simulate the residual strength tests a small part (sample) and is difficult to be applied to bigger parts. Also,
due to the simulation difficulties [2,177,184,185]. Gonzalaz et al. [93] there is no upscaling approach available to relate this residual strength
simulated the LVI and CAI test using interlaminar and intralaminar to a bigger part due to complex damage state. The numerical simulation
damage model. The authors used LaRC04 failure criteria [151] for in- could be one possible option to establish this scaling relationship from
tralaminar damage initiation. This proposed model predicted the CAI small test samples to a full-scale structure. However, at present, this is
strength within a 20% error. However, this model is very expensive in not possible due to limitations of numerical modeling such as complex
terms of computational time. Caputo et al. [186] used a single step damage state, damage progression and strength reduction during im-
analysis to simulate LVI and CAI tests in ABAQUS. This model captures pact. Therefore, damage tolerance assessment is still depending on the
damage modes that occurred during residual strength test. Elias et al. experimental evaluation. Also, accurate prediction of fracture energies
[46] proposed a two-step process to perform CAI simulations. In the and the critical energy release rate, Gi,jC , are the most important para-
first step, LVI simulation will be performed to obtain damage pattern meters to predict impact resistance and damage tolerance of composite
(permanent indentation depth and damage area) and damage indices of materials. The critical energy release rate must be measured experi-
each element. In the second step, compression of pre-impacted speci- mentally through standard tests and related to strain energy density,
mens will be performed using deformed shape and damage indices of gi,jC , through element characteristic length. Therefore, it is important to
each element. measure the element characteristic length accurately, so that energy
A simplified damage model was proposed by Rozylo et al. [187] to dissipated in the numerical simulation is consistent with the experi-
predict CAI strength of laminated composites. The authors developed mental predictions. The reported research on the prediction of residual
the simple relationship between ply thickness and impact energies and strength is given in Table 11.
calculated the minimum thickness of first ply and successive plies for
any impact energy using the following relationships given by Eqs. (18) 7. Concluding remarks
and. (19)
The objective of this review paper is to provide a detailed review on
y1 = 2 − (X1 E ) (18) the effect of different factors on impact resistance and damage tolerance
of fiber reinforced composites, with the main emphasis on primary
yn = yn − 1 − (X2 E ) (19) factors (fabric architecture and resin toughness). The concluding re-
where, “ y1”, “ yn ” and “E” are the thickness of first ply, successive plies marks are summarized below:
and impact energy. The constants “ X1” and “ X2 ” represent the decrease
in the thickness of plies at any impact energy “E” in the damage region. • The TP resin has superior impact resistance, damage tolerance and
Using the methodology above, the CAI simulation was performed on improved performance under extreme conditions (hygrothermal
impact damaged laminate to predict residual strength. The authors conditions). Also, the use of TP resin increases the probability to
Table 11
Reported literature to predict damage tolerance (residual strength).
References year Damage model Comments
Gonzalaz et al. [93] 2012 LaRC04 failure criteria Predicts CAI behaviour but computationally very expensive
Rivallant et al. [90] 2013 Strain based criteria Predicts indentation and CAI behaviour
Caputo et al. [186] 2015 Hashin failure criteria CAI results show good correlation with experiments
Tan et al. [158] 2015 Strain-based failure criteria, Puck Predicts indentation and CAI behaviour with good accuracy
Abir et al. [177] 2017 Tasi Wu, Max stress Predicts CAI strength and damage mechanisms
117
detect the surface damage caused by the LVI. The effects of hygro- [6] Tong L, Mouritz AP, Bannister M. 3D fibre reinforced polymer composites.
thermal condition (temperature and humidity) have been studied in Elsevier; 2002.
[7] Components NRCCoTCaS, Board NRCNMA. The place for thermoplastic compo-
detail on 2D composites with both TS and TP resin. However, their sites in structural components: report of the committee on thermoplastic compo-
effects on 3D TP composites need to be investigated in detail before sites as structural components: National Academies; 1987.
their application to the real structures. [8] Bazhenov S. Dissipation of energy by bulletproof aramid fabric. J Mater Sci
• To the best of authors’ knowledge, only a few studies are available

1997;32:4167–73.
[9] Liu X, Yu W, Pan N. Evaluation of high performance fabric under light irradiation.
for 3D composites and TP resin due to their manufacturing diffi- J Appl Polym Sci 2011;120:552–6.
culties and poor interface strength. In addition, no studies are [10] Vodicka R. Thermoplastics for airframe applications: a review of the properties
and repair methods for thermoplastic composites. Citeseer 1996.
available for 3D composites with TP resin “Elium®”. It is expected [11] Sorrentino L, Silva de Vasconcellos D, D'Auria M, Tirillò J, Sarasini F. Flexural and
that the usage of TP resin in 3D composites will provide a perfect low velocity impact characterization of thermoplastic composites based on PEN
combination to achieve excellent impact resistance and damage and high performance woven fabrics. Polym Compos 2017;39:2942–51.
[12] Zhao T, Palardy G, Villegas IF, Rans C, Martinez M, Benedictus R. Mechanical
tolerance by taking the advantage of through-thickness reinforce-
behaviour of thermoplastic composites spot-welded and mechanically fastened
ment of 3D woven fabric and toughness of TP resin. joints: a preliminary comparison. Compos B Eng 2017;112:224–34.
• In the real scenarios, impact events are random in nature and they [13] ASTM D7136/D7136M-05. Standard test method for measuring the damage re-
sistance of a fiber-reinforced polymer matrix composite to a drop-weight impact
vary in terms of impact energies, the mass of impactor and impact
event. West Conshohocken, PA: ASTM International; 2005.
velocities. To study the effect of such multiple random impact [14] Cantwell W, Morton J. The impact resistance of composite materials—a review.
events, there is a need to investigate the multiple impacts on the Compos Mater 1991;22:347–62.
composite structure and develop a method to simulate them to re- [15] Wisnom M. The role of delamination in failure of fibre-reinforced composites. Phil
Trans R Soc A. 2012;370:1850–70.
duce the computational cost. [16] Baker AAB. Composite materials for aircraft structures. AIAA 2004.
• Another important area is to evaluate the residual strength of impact [17] Greve L, Pickett A. Delamination testing and modelling for composite crash si-
mulation. Compos Sci Technol 2006;66:816–26.
damaged specimen and to correlate them with the visible dent (dent
[18] Mouritz A. Ballistic impact and explosive blast resistance of stitched composites.
area or depth). This will help the inspection team to take a quick Compos B Eng 2001;32:431–9.
decision, whether to repair the composite structure or allow it to [19] Francesconi L, Aymerich F. Numerical simulation of the effect of stitching on the
continue its operation. delamination resistance of laminated composites subjected to low-velocity impact.
• The residual strength is measured on a small specimen and it is very Compos Struct 2017;159:110–20.
[20] Tan KT, Watanabe N, Iwahori Y. Impact damage resistance, response, and me-
difficult to conclude whether this residual strength represents the chanisms of laminated composites reinforced by through-thickness stitching. Int J
strength of the whole composite structure. There is a strong need to Damage Mech 2012;21:51–80.
[21] Tan KT, Watanabe N, Iwahori Y. Experimental investigation of bridging law for
develop an upscaling approach to correlate local and global residual
single stitch fibre using Interlaminar tension test. Compos Struct
strength in composite structures. 2010;92:1399–409.
• To predict the high strain to failure and the nonlinear behaviour of [22] Wood MD, Sun X, Tong L, Luo Q, Katzos A, Rispler A. A new ENF test specimen for
the mode II delamination toughness testing of stitched woven CFRP laminates. J
3D composites, three step methodology is proposed. In the first step,
Compos Mater 2007;41:1743–72.
tensile tests (loading/unloading) of off-axis and on-axis specimens [23] Dransfield KA, Jain LK, Mai Y-W. On the effects of stitching in CFRPs Mode I
are performed. Secondly, the data analysis and extraction of the delamination toughness. Compos Sci Technol 1998;58:815–27.
most sensitive factors representing such nonlinear behaviour of 3D/ [24] Yasaee M, Bigg L, Mohamed G, Hallett SR. Influence of Z-pin embedded length on
the interlaminar traction response of multi-directional composite laminates. Mater
TP composites are carried out. Lastly, these factors and relationships Des 2017;115:26–36.
are included in the damage model. This will help in predicting the [25] Hart KR, Chia PXL, Sheridan LE, Wetzel ED, Sottos NR, White SR. Comparison of
nonlinear behaviour of composites under various loading conditions compression-after-impact and flexure-after-impact protocols for 2D and 3D woven
fiber-reinforced composites. Compos A Appl Sci Manuf 2017;101:471–9.
where shear behaviour is dominant i.e. bolted joints, bearing pro- [26] Umer R, Alhussein H, Zhou J, Cantwell W. The mechanical properties of 3D woven
blems and low-velocity impact. composites. J Compos Mater 2017;51:1703–16.
[27] Wang M, Cao M, Wang H, Siddique A, Gu B, Sun B. Drop-weight impact behaviors
of 3-D angle interlock woven composites after thermal oxidative aging. Compos
Acknowledgment
Struct 2017;166:239–55.
[28] Seltzer R, González C, Muñoz R, LLorca J, Blanco-Varela T. X-ray micro-
The authors would like to acknowledge the financial support pro- tomography analysis of the damage micromechanisms in 3D woven composites
under low-velocity impact. Compos A Appl Sci Manuf 2013;45:49–60.
vided by Universiti Teknologi PETRONAS.
[29] Wang X, Hu B, Feng Y, Liang F, Mo J, Xiong J. Low velocity impact properties of
3D woven basalt/aramid hybrid composites. Compos Sci Technol 2008;68:444–50.
Conflicts of interest [30] Potluri P, Hogg P, Arshad M, Jetavat D, Jamshidi P. Influence of fibre architecture
on impact damage tolerance in 3D woven composites. Appl Compos Mater
2012;19:799–812.
The authors declare no conflict of interest with respect to the re- [31] Bandaru AK, Sachan Y, Ahmad S, Alagirusamy R, Bhatnagar N. On the mechanical
search or publication of this work. response of 2D plain woven and 3D angle-interlock fabrics. Compos B Eng
2017;118:135–48.
[32] Bhatnagar A. Lightweight ballistic composites: military and law-enforcement ap-
Appendix A. Supplementary data plications. Woodhead Publishing; 2016.
[33] Béland S. High performance thermoplastic resins and their composites. William
Supplementary data to this article can be found online at https:// Andrew; 1990.
[34] Tan W, Falzon BG. Modelling the crush behaviour of thermoplastic composites.
doi.org/10.1016/j.compstruct.2019.03.021. Compos Sci Technol 2016;134:57–71.
[35] Tan W, Falzon BG. Modelling the nonlinear behaviour and fracture process of AS4/
References PEKK thermoplastic composite under shear loading. Compos Sci Technol
2016;126:60–77.
[36] Striewe J, Reuter C, Sauerland K-H, Tröster T. Manufacturing and crashworthiness
[1] Richardson M, Wisheart M. Review of low-velocity impact properties of composite of fabric-reinforced thermoplastic composites. Thin-Walled Struct
materials. Compos A Appl Sci Manuf 1996;27:1123–31. 2018;123:501–8.
[2] Dubary N, Taconet G, Bouvet C, Vieille B. Influence of temperature on the impact [37] Santiago R, Cantwell W, Alves M. Impact on thermoplastic fibre-metal laminates:
behavior and damage tolerance of hybrid woven-ply thermoplastic laminates for experimental observations. Compos Struct 2017;159:800–17.
aeronautical applications. Compos Struct 2017;168:663–74. [38] Chen P, Shen Z, Xiong J, Yang S, Fu S, Ye L. Failure mechanisms of laminated
[3] Kim JK. Methods for improving impact damage resistance of CFRPs. Key en- composites subjected to static indentation. Compos Struct 2006;75:489–95.
gineering materials: Trans Tech Publ. 1998. p. 149–68. [39] Kim E-H, Rim M-S, Lee I, Hwang T-K. Composite damage model based on con-
[4] Kim JK, Sham ML. Impact and delamination failure of woven-fabric composites. tinuum damage mechanics and low velocity impact analysis of composite plates.
Compos Sci Technol 2000;60:745–61. Compos Struct 2013;95:123–34.
[5] Bibo G, Hogg P. The role of reinforcement architecture on impact damage me- [40] Dogan A, Arikan V. Low-velocity impact response of E-glass reinforced thermoset
chanisms and post-impact compression behaviour. J Mater Sci 1996;31:1115–37. and thermoplastic based sandwich composites. Compos B Eng 2017;127:63–9.
118
[41] Vieille B, Casado VM, Bouvet C. About the impact behavior of woven-ply carbon [73] Icten BM. Low temperature effect on single and repeated impact behavior of
fiber-reinforced thermoplastic-and thermosetting-composites: a comparative woven glass-epoxy composite plates. J Compos Mater 2015;49:1171–8.
study. Compos Struct 2013;101:9–21. [74] Karthikeyan K, Russell BP, Deshpande VS, Fleck NA. Multi-hit armour character-
[42] Bandaru AK, Chavan VV, Ahmad S, Alagirusamy R, Bhatnagar N. Low velocity isation of metal-composite bi-layers. J Mech Mater Struct 2012;7:721–34.
impact response of 2D and 3D Kevlar/polypropylene composites. Int J Impact Eng [75] Arikan V, Sayman O. Comparative study on repeated impact response of E-glass
2016;93:136–43. fiber reinforced polypropylene & epoxy matrix composites. Compos B Eng
[43] Bandaru AK, Patel S, Sachan Y, Alagirusamy R, Bhatnagar N, Ahmad S. Low ve- 2015;83:1–6.
locity impact response of 3D angle-interlock Kevlar/basalt reinforced poly- [76] Icten BM, Kıral BG, Deniz ME. Impactor diameter effect on low velocity impact
propylene composites. Mater Des 2016;105:323–32. response of woven glass epoxy composite plates. Compos B Eng 2013;50:325–32.
[44] Zhang D, Sun Y, Chen L, Pan N. A comparative study on low-velocity impact re- [77] Sevkat E, Liaw B, Delale F. Drop-weight impact response of hybrid composites
sponse of fabric composite laminates. Mater Des 2013;50:750–6. impacted by impactor of various geometries. Mater Des 2013;52:67–77.
[45] Hart KR, Chia PXL, Sheridan LE, Wetzel ED, Sottos NR, White SR. Mechanisms and [78] Mitrevski T, Marshall I, Thomson R. The influence of impactor shape on the da-
characterization of impact damage in 2D and 3D woven fiber-reinforced compo- mage to composite laminates. Compos Struct 2006;76:116–22.
sites. Compos A Appl Sci Manuf 2017;101:432–43. [79] Artero-Guerrero JA, Pernas-Sánchez J, López-Puente J, Varas D. Experimental
[46] Elias A, Laurin F, Kaminski M, Gornet L. Experimental and numerical investiga- study of the impactor mass effect on the low velocity impact of carbon/epoxy
tions of low energy/velocity impact damage generated in 3D woven composite woven laminates. Compos Struct 2015;133:774–81.
with polymer matrix. Compos Struct 2017;159:228–39. [80] Ambur D, Kemmerly H. Influence of impactor mass on the damage characteristics
[47] Gao SL, Kim JK. Cooling rate influences in carbon fibre/PEEK composites. Part III: and failure strength of laminated composite plates. 39th AIAA/ASME/ASCE/AHS/
impact damage performance. Compos A Appl Sci Manuf 2001;32:775–85. ASC structures, structural dynamics and materials conference and exhibit. 1998. p.
[48] Alderliesten R. Damage tolerance of bonded aircraft structures. Int J Fatigue 1784.
2009;31:1024–30. [81] Feraboli P, Kedward KT. A new composite structure impact performance assess-
[49] Thorsson SI, Sringeri SP, Waas AM, Justusson BP, Rassaian M. Experimental in- ment program. Compos Sci Technol 2006;66:1336–47.
vestigation of composite laminates subject to low-velocity edge-on impact and [82] Bucinell RB, Nuismer RJ, Koury JL. Response of composite plates to quasi-static
compression after impact. Compos Struct 2018;186:335–46. impact events. Composite materials: fatigue and fracture. ASTM International;
[50] González E, Maimí P, Camanho P, Lopes C, Blanco N. Effects of ply clustering in 1991.
laminated composite plates under low-velocity impact loading. Compos Sci [83] Karakuzu R, Erbil E, Aktas M. Impact characterization of glass/epoxy composite
Technol 2011;71:805–17. plates: an experimental and numerical study. Compos B Eng 2010;41:388–95.
[51] Kinsey A, Saunders D, Soutis C. Post-impact compressive behaviour of low tem- [84] Zabala H, Aretxabaleta L, Castillo G, Urien J, Aurrekoetxea J. Impact velocity ef-
perature curing woven CFRP laminates. Compos Mater 1995;26:661–7. fect on the delamination of woven carbon–epoxy plates subjected to low-velocity
[52] Chiu C, Lai M, Wu C. Compression failure mechanisms of 3-D angle interlock equienergetic impact loads. Compos Sci Technol 2014;94:48–53.
woven composites subjected to low-energy impact. Polym Polym Compos [85] Boria S, Scattina A, Belingardi G. Impact behavior of a fully thermoplastic com-
2004;12:309–20. posite. Compos Struct 2017;167:63–75.
[53] Chen F, Hodgkinson J. Impact behaviour of composites with different fibre ar- [86] Gliszczynski A, Kubiak T, Rozylo P, Jakubczak P, Bieniaś JJCS. The response of
chitecture. Proc Inst Mech Eng, Part G: J Aerospace Eng 2009;223:1009–17. laminated composite plates and profiles under low-velocity impact load. Compos
[54] Reyes G, Sharma U. Modeling and damage repair of woven thermoplastic com- Struct 2018.
posites subjected to low velocity impact. Compos Struct 2010;92:523–31. [87] Bandaru AK, Ahmad S, Bhatnagar N. Ballistic performance of hybrid thermoplastic
[55] Vieille B, Casado VM, Bouvet C. Influence of matrix toughness and ductility on the composite armors reinforced with Kevlar and basalt fabrics. Compos A Appl Sci
compression-after-impact behavior of woven-ply thermoplastic-and thermosetting- Manuf 2017;97:151–65.
composites: a comparative study. Compos Struct 2014;110:207–18. [88] Luan K, Gu B. Energy absorption of three-dimensional angle-interlock woven
[56] Bull D, Spearing S, Sinclair I. Observations of damage development from com- composite under ballistic penetration based on a multi-scale finite element model.
pression-after-impact experiments using ex situ micro-focus computed tomo- Int J Damage Mech 2015;24:3–20.
graphy. Compos Sci Technol 2014;97:106–14. [89] Agrawal S, Singh KK, Sarkar P. Impact damage on fibre-reinforced polymer matrix
[57] Ishikawa T, Sugimoto S, Matsushima M, Hayashi Y. Some experimental findings in composite–a review. J Compos Mater 2014;48:317–32.
compression-after-impact (CAI) tests of CF/PEEK (APC-2) and conventional CF/ [90] Rivallant S, Bouvet C, Hongkarnjanakul N. Failure analysis of CFRP laminates
epoxy flat plates. Compos Sci Technol 1995;55:349–63. subjected to compression after impact: FE simulation using discrete interface
[58] Sorrentino L, Sarasini F, Tirillò J, Touchard F, Chocinski-Arnault L. Damage tol- elements. Compos A Appl Sci Manuf 2013;55:83–93.
erance assessment of the interface strength gradation in thermoplastic composites. [91] Wang HR, Long SC, Zhang XQ, Yao XH. Study on the delamination behavior of
Compos B Eng 2017;113:111–22. thick composite laminates under low-energy impact. Compos Struct
[59] Davies P, Riou L, Mazeas F, Warnier P. Thermoplastic composite cylinders for 2018;184:461–73.
underwater applications. J Thermoplast Compos Mater 2005;18:417–43. [92] Bouvet C, Rivallant S, Barrau J-J. Low velocity impact modeling in composite
[60] Shiino MY, Pelosi TS, Cioffi MOH, Donadon MV. The role of stitch yarn on the laminates capturing permanent indentation. Compos Sci Technol
delamination resistance in non-crimp fabric: chemical and physical interpretation. 2012;72:1977–88.
J Mater Eng Perform 2017;26:978–86. [93] González E, Maimí P, Camanho P, Turon A, Mayugo J. Simulation of drop-weight
[61] Pankow M, Salvi A, Waas A, Yen C, Ghiorse S. Resistance to delamination of 3D impact and compression after impact tests on composite laminates. Compos Struct
woven textile composites evaluated using End Notch Flexure (ENF) tests: experi- 2012;94:3364–78.
mental results. Compos A Appl Sci Manuf 2011;42:1463–76. [94] Maio L, Monaco E, Ricci F, Lecce L. Simulation of low velocity impact on com-
[62] Liu H, Falzon BG, Tan W. Experimental and numerical studies on the impact re- posite laminates with progressive failure analysis. Compos Struct 2013;103:75–85.
sponse of damage-tolerant hybrid unidirectional/woven carbon-fibre reinforced [95] Johnson H, Louca L, Mouring S, Fallah A. Modelling impact damage in marine
composite laminates. Compos B Eng 2018;136:101–18. composite panels. Int J Impact Eng 2009;36:25–39.
[63] De Baere I, Jacques S, Van Paepegem W, Degrieck J. Study of the Mode I and Mode [96] Jang B, Chen L, Wang C, Lin H, Zee R. Impact resistance and energy absorption
II interlaminar behaviour of a carbon fabric reinforced thermoplastic. Polym Test mechanisms in hybrid composites. Compos Sci Technol 1989;34:305–35.
2012;31:322–32. [97] Benli S, Sayman O. The effects of temperature and thermal stresses on impact
[64] Friedrich K, Gogeva T, Fakirov S. Thermoplastic impregnated fiber bundles: damage in laminated composites. Math Comput Appl 2011;16:392–403.
manufacturing of laminates and fracture mechanics characterization. Compos Sci [98] Strait L, Karasek M, Amateau M. Effects of stacking sequence on the impact re-
Technol 1988;33:97–120. sistance of carbon fiber reinforced thermoplastic toughened epoxy laminates. J
[65] Kim K-Y, Ye L. Interlaminar fracture toughness of CF/PEI composites at elevated Compos Mater 1992;26:1725–40.
temperatures: roles of matrix toughness and fibre/matrix adhesion. Compos A [99] De Freitas M, Silva A, Reis L. Numerical evaluation of failure mechanisms on
Appl Sci Manuf 2004;35:477–87. composite specimens subjected to impact loading. Compos B Eng
[66] Hinkley J. Interface effects in interlaminar fracture of thermoplastic composites. J 2000;31:199–207.
Reinf Plast Compos 1990;9:470–6. [100] Aktaş M, Balcıoğlu HE, Aktaş A, Türker E, Deniz ME. Impact and post impact
[67] Tan W, Falzon BG, Price M, Liu H. The role of material characterisation in the behavior of layer fabric composites. Compos Struct 2012;94:2809–18.
crush modelling of thermoplastic composite structures. Compos Struct [101] Lopes C, Seresta O, Abdalla M, Gurdal Z, Thuis B, Camanho P. Stacking sequence
2016;153:914–27. dispersion and tow-placement for improved damage tolerance. 49th AIAA/ASME/
[68] Hongkarnjanakul N, Bouvet C, Rivallant S. Validation of low velocity impact ASCE/AHS/ASC structures, structural dynamics, and materials conference. 2008.
modelling on different stacking sequences of CFRP laminates and influence of fibre p. 1735.
failure. Compos Struct 2013;106:549–59. [102] Dost EF, Ilcewicz LB, Avery WB, Coxon BR. Effects of stacking sequence on impact
[69] Baucom J, Ma Zikry, Rajendran A. Low-velocity impact damage accumulation in damage resistance and residual strength for quasi-isotropic laminates. Composite
woven S2-glass composite systems. Compos Sci Technol 2006;66:1229–38. materials: fatigue and fracture. ASTM International; 1991.
[70] Santos RAM, Reis PNB, Silva FGA, de Moura MFSF. Influence of inclined holes on [103] Reis L, De Freitas M. Damage growth analysis of low velocity impacted composite
the impact strength of CFRP composites. Compos Struct 2017;172:130–6. panels. Compos Struct 1997;38:509–15.
[71] Santos RAM, Reis PNB, Santos MJ, Coelho CACP. Effect of distance between im- [104] Sonnenfeld C, Mendil-Jakani H, Agogué R, Nunez P, Beauchêne P. Thermoplastic/
pact point and hole position on the impact fatigue strength of composite laminates. thermoset multilayer composites: a way to improve the impact damage tolerance
Compos Struct 2017;168:33–9. of thermosetting resin matrix composites. Compos Struct 2017;171:298–305.
[72] Atas C, Dogan A. An experimental investigation on the repeated impact response of [105] Baker N, Butler R, York CBJCS. Technology damage tolerance of fully orthotropic
glass/epoxy composites subjected to thermal ageing. Compos B Eng laminates in compression. Compos Sci Technol 2012;72:1083–9.
2015;75:127–34. [106] York CJJOAE. Unified approach to the characterization of coupled composite
119
laminates benchmark configurations and special cases. J Aerospace Eng composite structures Part II: Applications. Compos Struct 2011;93:1047–53.
2009;23:219–42. [138] Shi Y, Swait T, Soutis C. Modelling damage evolution in composite laminates
[107] Samborski SJCS. Numerical analysis of the DCB test configuration applicability to subjected to low velocity impact. Compos Struct 2012;94:2902–13.
mechanically coupled Fiber Reinforced Laminated Composite beams. Compos [139] Iannucci L. Progressive failure modelling of woven carbon composite under im-
Struct 2016;152:477–87. pact. Int J Impact Eng 2006;32:1013–43.
[108] Samborski SJCS. Analysis of the end-notched flexure test configuration applic- [140] Williams KV, Vaziri R, Poursartip A. A physically based continuum damage me-
ability for mechanically coupled fiber reinforced composite laminates. Compos chanics model for thin laminated composite structures. Int J Solids Struct
Struct 2017;163:342–9. 2003;40:2267–300.
[109] Parvatareddy H, Tsang PW, Dillard D. Impact damage resistance and tolerance of [141] Maimí P, Camanho PP, Mayugo J, Dávila C. A continuum damage model for
high-performance polymeric composites subjected to environmental aging. composite laminates: Part I-constitutive model. Mech Mater 2007;39:897–908.
Compos Sci Technol 1996;56:1129–40. [142] Lopes C, Camanho P, Gürdal Z, Maimí P, González E. Low-velocity impact damage
[110] Li G, Pang SS, Helms JE, Ibekwe SI. Low velocity impact response of GFRP la- on dispersed stacking sequence laminates. Part II: numerical simulations. Compos
minates subjected to cycling moistures. Polym Compos 2000;21:686–95. Sci Technol 2009;69:937–47.
[111] Mortas N, Er O, Reis P, Ferreira J. Effect of corrosive solutions on composites [143] Chang F-K, Chang K-Y. A progressive damage model for laminated composites
laminates subjected to low velocity impact loading. Compos Struct containing stress concentrations. J Compos Mater 1987;21:834–55.
2014;108:205–11. [144] Hassoon O, Tarfaoui M, El Moumen A. Progressive damage modeling in laminate
[112] Vieille B, Aucher J, Taleb L. Comparative study on the behavior of woven-ply composites under slamming impact water for naval applications. Compos Struct
reinforced thermoplastic or thermosetting laminates under severe environmental 2017;167:178–90.
conditions. Mater Des 2012;35:707–19. [145] Kim D-H, Jung K-H, Lee I-G, Kim H-J, Kim H-S. Three-dimensional progressive
[113] Russo P, Langella A, Papa I, Simeoli G, Lopresto V. Low-velocity impact and failure modeling of glass fiber reinforced thermoplastic composites for impact si-
flexural properties of thermoplastic polyurethane/woven glass fabric composite mulation. Compos Struct 2017;176:757–67.
laminates. Procedia Eng 2016;167:190–6. [146] Guo W, Xue P, Yang J. Nonlinear progressive damage model for composite lami-
[114] Russo P, Langella A, Papa I, Simeoli G, Lopresto V. Thermoplastic polyurethane/ nates used for low-velocity impact. Appl Math Mech 2013;34:1145–54.
glass fabric composite laminates: low velocity impact behavior under extreme [147] Matzenmiller A, Lubliner J, Taylor R. A constitutive model for anisotropic damage
temperature conditions. Compos Struct 2017;166:146–52. in fiber-composites. Mech Mater 1995;20:125–52.
[115] Boumbimba RM, Coulibaly M, Khabouchi A, Kinvi-Dossou G, Bonfoh N, Gerard P. [148] Zhang C, Li N, Wang W, Binienda WK, Fang H. Progressive damage simulation of
Glass fibres reinforced acrylic thermoplastic resin-based tri-block copolymers triaxially braided composite using a 3D meso-scale finite element model. Compos
composites: low velocity impact response at various temperatures. Compos Struct Struct 2015;125:104–16.
2017;160:939–51. [149] Hou JP, Petrinic N, Ruiz C, Hallett S. Prediction of impact damage in composite
[116] Wang Y, Zhang J, Fang G, Zhang J, Zhou Z, Wang S. Influence of temperature on plates. Compos Sci Technol 2000;60:273–81.
the impact behavior of woven-ply carbon fiber reinforced thermoplastic compo- [150] Topac OT, Gozluklu B, Gurses E, Coker D. Experimental and computational study
sites. Compos Struct 2018;185:435–45. of the damage process in CFRP composite beams under low-velocity impact.
[117] Wang Y, Zhang J, Zhang J, Zhou Z, Fang G, Wang S. Compressive behavior of Compos A Appl Sci Manuf 2017;92:167–82.
notched and unnotched carbon woven-ply PPS thermoplastic laminates at dif- [151] Pinho ST, Davila CG, Camanho PP, Iannucci L, Robinson P. Failure models and
ferent temperatures. Compos B Eng 2018;133:68–77. criteria for FRP under in-plane or three-dimensional stress states including shear
[118] Sorrentino L, de Vasconcellos DS, D'Auria M, Sarasini F, Tirillò J. Effect of tem- non-linearity. 2005. NASA/TM-2003-213530.
perature on static and low velocity impact properties of thermoplastic composites. [152] Abdallah EA, Bouvet C, Rivallant S, Broll B, Barrau J-J. Experimental analysis of
Compos B Eng 2017;113:100–10. damage creation and permanent indentation on highly oriented plates. Compos Sci
[119] Safri SNA, Sultan MTH, Jawaid M, Jayakrishna K. Impact behaviour of hybrid Technol 2009;69:1238–45.
composites for structural applications: a review. Compos B Eng 2017:112–21. [153] Caprino G, Langella A, Lopresto V. Indentation and penetration of carbon fibre
[120] Naik N, Meduri S. Polymer-matrix composites subjected to low-velocity impact: reinforced plastic laminates. Compos B Eng 2003;34:319–25.
effect of laminate configuration. Compos Sci Technol 2001;61:1429–36. [154] Das S, Kandan K, Kazemahvazi S, Wadley HN, Deshpande VS. Indentation response
[121] Sevkat E, Liaw B, Delale F, Raju BB. Drop-weight impact of plain-woven hybrid of a 3D non-woven carbon-fibre composite. J Mater Res 2018;33:1–13.
glass–graphite/toughened epoxy composites. Compos A Appl Sci Manuf [155] Sun C, Chen J. A simple flow rule for characterizing nonlinear behavior of fiber
2009;40:1090–110. composites. J Compos Mater 1989;23:1009–20.
[122] Dehkordi MT, Nosraty H, Shokrieh MM, Minak G, Ghelli D. Low velocity impact [156] Singh H, Mahajan P. Modeling damage induced plasticity for low velocity impact
properties of intra-ply hybrid composites based on basalt and nylon woven fabrics. simulation of three dimensional fiber reinforced composite. Compos Struct
Mater Des 2010;31:3835–44. 2015;131:290–303.
[123] Dehkordi MT, Nosraty H, Shokrieh MM, Minak G, Ghelli D. The influence of hy- [157] Liao B, Liu P. Finite element analysis of dynamic progressive failure of plastic
bridization on impact damage behavior and residual compression strength of in- composite laminates under low velocity impact. Compos Struct 2017;159:567–78.
traply basalt/nylon hybrid composites. Mater Des 2013;43:283–90. [158] Tan W, Falzon BG, Chiu LN, Price M. Predicting low velocity impact damage and
[124] Sarasini F, Tirillò J, D'Altilia S, Valente T, Santulli C, Touchard F, et al. Damage Compression-After-Impact (CAI) behaviour of composite laminates. Compos A
tolerance of carbon/flax hybrid composites subjected to low velocity impact. Appl Sci Manuf 2015;71:212–26.
Compos B Eng 2016;91:144–53. [159] Ladeveze P, LeDantec E. Damage modelling of the elementary ply for laminated
[125] Bandaru AK, Patel S, Sachan Y, Ahmad S, Alagirusamy R, Bhatnagar N. Mechanical composites. Compos Sci Technol 1992;43:257–67.
characterization of 3D angle-interlock Kevlar/basalt reinforced polypropylene [160] Van Paepegem W, De Baere I, Degrieck J. Modelling the nonlinear shear stress–-
composites. Polym Test 2016;55:238–46. strain response of glass fibre-reinforced composites. Part I: experimental results.
[126] Simeoli G, Acierno D, Meola C, Sorrentino L, Iannace S, Russo P. The role of in- Compos Sci Technol 2006;66:1455–64.
terface strength on the low velocity impact behaviour of PP/glass fibre laminates. [161] Van Paepegem W, De Baere I. Modelling the nonlinear shear stress–strain response
Compos B Eng 2014;62:88–96. of glass fibre-reinforced composites. Part II: model development and finite element
[127] Russo P, Simeoli G, Sorrentino L, Iannace S. Effect of the compatibilizer content on simulations. Compos Sci Technol 2006;66:1465–78.
the quasi-static and low velocity impact responses of glass woven fabric/poly- [162] Shi Y, Pinna C, Soutis C. Modelling impact damage in composite laminates: a si-
propylene composites. Polym Compos 2016;37:2452–9. mulation of intra-and inter-laminar cracking. Compos Struct 2014;114:10–9.
[128] Boccardi S, Meola C, Carlomagno G, Sorrentino L, Simeoli G, Russo P. Effects of [163] Hashin Z. Failure criteria for unidirectional fiber composites. J Appl Mech
interface strength gradation on impact damage mechanisms in polypropylene/ 1980;47:329–34.
woven glass fabric composites. Compos B Eng 2016;90:179–87. [164] Puck A, Schürmann H. Failure analysis of FRP laminates by means of physically
[129] Nash N, Young T, McGrail P, Stanley W. Inclusion of a thermoplastic phase to based phenomenological models. Compos Sci Technol 2002;62:1633–62.
improve impact and post-impact performances of carbon fibre reinforced ther- [165] Berbinau P, Soutis C, Goutas P, Curtis P. Effect of off-axis ply orientation on 0-fibre
mosetting composites: a review. Mater Des 2015;85:582–97. microbuckling. Compos A Appl Sci Manuf 1999;30:1197–207.
[130] Kachanov LM. Time of the rupture process under creep conditions, Izy Akad. Nank [166] Feng D, Aymerich F. Finite element modelling of damage induced by low-velocity
SSR Otd Tech Nauk. 1958;8:26–31. impact on composite laminates. Compos Struct 2014;108:161–71.
[131] Rabotnov YN. Creep rupture. Applied mechanics. Applied mechanics: Springer; [167] Chiu LN, Falzon BG, Boman R, Chen B, Yan W. Finite element modelling of
1969. p. 342–9. composite structures under crushing load. Compos Struct 2015;131:215–28.
[132] Menna C, Asprone D, Caprino G, Lopresto V, Prota A. Numerical simulation of [168] Chiu LN, Falzon BG, Chen B, Yan W. Validation of a 3D damage model for pre-
impact tests on GFRP composite laminates. Int J Impact Eng 2011;38:677–85. dicting the response of composite structures under crushing loads. Compos Struct
[133] Iannucci L, Willows M. An energy based damage mechanics approach to modelling 2016;147:65–73.
impact onto woven composite materials Part I: numerical models. Compos A Appl [169] Lomov SV, Bogdanovich AE, Ivanov DS, Mungalov D, Karahan M, Verpoest I. A
Sci Manuf 2006;37:2041–56. comparative study of tensile properties of non-crimp 3D orthogonal weave and
[134] Donadon M, Iannucci L, Falzon BG, Hodgkinson J, de Almeida SF. A progressive multi-layer plain weave E-glass composites. Part 1: materials, methods and prin-
failure model for composite laminates subjected to low velocity impact damage. cipal results. Compos A Appl Sci Manuf 2009;40:1134–43.
Comput Struct 2008;86:1232–52. [170] Saleh MN, Wang Y, Yudhanto A, Joesbury A, Potluri P, Lubineau G, et al.
[135] Faggiani A, Falzon B. Predicting low-velocity impact damage on a stiffened Investigating the potential of using off-Axis 3D woven composites in composite
composite panel. Compos A Appl Sci Manuf 2010;41:737–49. joints’ applications. Appl Compos Mater 2017;24:377–96.
[136] Falzon B, Apruzzese P. Numerical analysis of intralaminar failure mechanisms in [171] Labanieh AR, Liu Y, Vasiukov D, Soulat D, Panier S. Influence of off-axis in-plane
composite structures Part I: FE implementation. Compos Struct 2011;93:1039–46. yarns on the mechanical properties of 3D composites. Compos A Appl Sci Manuf
[137] Falzon B, Apruzzese P. Numerical analysis of intralaminar failure mechanisms in 2017;98:45–57.
120
[172] Hao A, Sun B, Qiu Y, Gu B. Dynamic properties of 3-D orthogonal woven com- [184] Wang C, Venugopal V, Peng L. Stepped flush repairs for primary composite
posite T-beam under transverse impact. Compos A Appl Sci Manuf structures. J Adhes 2015;91:95–112.
2008;39:1073–82. [185] Sun X, Wisnom M, Hallett S. Interaction of inter-and intralaminar damage in
[173] Chen J-F, Morozov EV, Shankar K. Simulating progressive failure of composite scaled quasi-static indentation tests: Part 2–Numerical simulation. Compos Struct
laminates including in-ply and delamination damage effects. Compos A Appl Sci 2016;136:727–42.
Manuf 2014;61:185–200. [186] Caputo F, De Luca A, Sepe R. Numerical study of the structural behaviour of im-
[174] Singh H, Namala KK, Mahajan P. A damage evolution study of E-glass/epoxy pacted composite laminates subjected to compression load. Compos B Eng
composite under low velocity impact. Compos B Eng 2015;76:235–48. 2015;79:456–65.
[175] Ren C, Liu T, Siddique A, Sun B, Gu B. High-speed visualizing and mesoscale [187] Rozylo P, Debski H, Kubiak TJCS. A model of low-velocity impact damage of
modeling for deformation and damage of 3D angle-interlock woven composites composite plates subjected to Compression-After-Impact (CAI) testing. Compos
subjected to transverse impacts. Int J Mech Sci 2018;140:119–32. Struct 2017;181:158–70.
[176] Das S, Kandan K, Kazemahvazi S, Wadley H, Deshpande V. Compressive response [188] Catalanotti G, Camanho P, Marques A. Three-dimensional failure criteria for fiber-
of a 3D non-woven carbon-fibre composite. Int J Solids Struct 2018;136:137–49. reinforced laminates. Compos Struct 2013;95:63–79.
[177] Abir M, Tay T, Ridha M, Lee H. Modelling damage growth in composites subjected [189] Spamer G, Brink N. Investigation of the CAI properties of C/PPS and C/APC-2 TP
to impact and compression after impact. Compos Struct 2017;168:13–25. materials. Proceedings of the 33rd international SAMPE symposium: materials-
[178] Song K, Dávila CG, Rose CA. Guidelines and parameter selection for the simulation pathway to the future. 1988.
of progressive delamination. Compos Mater 2008:1–15. [190] Jung K-H, Kim D-H, Kim H-J, Park S-H, Jhang K-Y, Kim H-S. Finite element
[179] Benzeggagh M, Kenane M. Measurement of mixed-mode delamination fracture analysis of a low-velocity impact test for glass fiber-reinforced polypropylene
toughness of unidirectional glass/epoxy composites with mixed-mode bending composites considering mixed-mode interlaminar fracture toughness. Compos
apparatus. Compos Sci Technol 1996;56:439–49. Struct 2017;160:446–56.
[180] Wu EM, Reuter Jr. R. Crack extension in fiberglass reinforced plastics. University [191] H-l Ma, Jia Z, K-t Lau, Leng J, Hui D. Impact properties of glass fiber/epoxy
of Illnois; 1965. composites at cryogenic environment. Compos B Eng 2016;92:210–7.
[181] Abrate S, Ferrero J, Navarro P. Cohesive zone models and impact damage pre- [192] Sun B, Liu Y, Gu B. A unit cell approach of finite element calculation of ballistic
dictions for composite structures. Meccanica 2015;50:2587–620. impact damage of 3-D orthogonal woven composite. Compos B Eng
[182] Muñoz R, Martínez-Hergueta F, Gálvez F, González C, LLorca J. Ballistic perfor- 2009;40:552–60.
mance of hybrid 3D woven composites: experiments and simulations. Compos [193] Ge J, He C, Liang J, Chen Y, Fang D. A coupled elastic-plastic damage model for
Struct 2015;127:141–51. the mechanical behavior of three-dimensional (3D) braided composites. Compos
[183] Turner P, Liu T, Zeng X. Dynamic response of orthogonal three-dimensional woven Sci Technol 2018;157:86–98.
carbon composite beams under soft impact. J Appl Mech 2015;82:121008.
121

shah2019

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

shah2019

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

shah2019

Uploaded by

Copyright:

Available Formats

Composite Structures 217 (2019) 100–121

Contents lists available at ScienceDirect

Impact resistance and damage tolerance of ﬁber reinforced composites: A T

Fig. 1. Schematic drawing of internal damages under LVI.

Fig. 2. Factors aﬀecting impact resistance and damage tolerance of FRC.

combination of damage initiation energy “Eini ” and damage propaga-

2. Eﬀect of primary factors on impact resistance and damage

2.1. Impact resistance of FRC

LVI events are very common in composite structures during their

composites possess superior impact resistance, delamination resistance,

interlaminar fracture toughness of composites [34,62]. Baere et al. [63]

Fig. 11. Diﬀerent types of indenter/impactor used for LVI test.

In composites, orientation and stacking sequence of plies can be

Fig. 14. Eﬀect of temperature on damage initiation energy [97].

3.6. Eﬀect of fabric hybridization

Another important factor in improving damage tolerance and im-

improved the impact properties of graphite; however, it increased the Table 7

Fig. 16. Schematic diagram of intralaminar damage modes.

Lannucci et al. [133,139] 2006 Strain-based criteria Strain-based linear softening

predict the nonlinear shear behaviour of FRC. Shi et al. [138,162]

Several researchers used the theory of plasticity based damage

Van Paepegem [160,161] 2006 E-glass/Epoxy –

Hao et al. [172] 2008 3D Glass OWC Critical damage area

• To the best of authors’ knowledge, only a few studies are available

You might also like