Carbon Fibre Reinforced Polymer

Carbon fiber reinforced polymer (CFRP) is one of the composite materials which is
used in both repairing and strengthening of reinforced concrete structures.

From: Nanotechnology in Eco-efficient Construction (Second Edition), 2019

Related terms:

Delamination, Carbon Fiber, Acoustic Emission, Fiber-Reinforced Polymer, Polymer

Matrix Composite, Fibre Reinforced Polymer, Glass Fibre, Reinforced Concrete
Beam

View all Topics

Polymer Matrix Composites: Applica-

tions
F. Al-Mahmoud, in Comprehensive Composite Materials II, 2018

3.22.2.2.8 CFRP bars surface textures

CFRP bars are manufactured with a variety of surface textures, which strongly
affect their bond behavior as NSM reinforcement. Their surface can be smooth,
sand-blasted, sand-coated, or roughened with a peel-ply surface treatment. Round
bars can also be spirally wound with a fiber tow, or ribbed [10].

One type of FRP rods was used [8]: carbon-epoxy pultruded FRP with a diameter
of 12 mm. The modulus of elasticity and tensile strength of the CFRP rods were
determined by laboratory testing. To modify the surface configurations of the initially
smooth rods, two different surface treatments were applied:

Surface machining: A turning lathe and a grinding stone were used to create lugs
on the CFRP rods. The three configurations tested are presented in Fig. 7. The
dimensions of the lugs for the three configurations are given in Fig. 8.
Fig. 7. Sanded and machined CFRP rods used in the study.

Fig. 8. Lug dimensions for three types of machined CFRP rods.

Surface sanding: The smooth CFRP rods were coated with sand of various grain sizes
(GS 10/20, 20/30, and 30/40) by sprinkling the sand on to a thin layer of freshly
applied Eponal 376 epoxy resin (Fig. 7).

> Read full chapter

Design of auto body

J.R. Fekete, J.N. Hall, in Automotive Steels, 2017
1.5.3 Carbon fiber reinforced polymers
Carbon fiber reinforced polymers are composite materials which rely on the carbon
fiber to provide the strength and stiffness while the polymer provides a cohesive
matrix to protect and hold the fibers together and provides some toughness. Carbon
fibers provide highly directional properties much different than the metals most
commonly used for these automotive applications. They can be engineered to
achieve mass reductions not achievable by the metals. Since these are artificially
composited materials their properties and performance can be tailored to the appli-
cation through varying strength, length, directionality and amount of the reinforcing
fibers and in the selection of the polymer matrix. The largest drawbacks are the high
cost in producing the fibers and the low throughput rates at which components
can be manufactured. The cumulative time to place the fibers in a mold, inject the
polymer and allow the part to set is in the order of a few minutes.

Carbon fiber reinforced polymers are gaining popularity in the luxury, sport segment
for mass reduction. These materials, which may also be reinforced with glass or other
fibers, have high-price tags and are more suited for lower volume manufacturing
as a result of molding cycle times. New technology in reducing fiber cost and
panel processing has been in the news in the past few years and the increased
implementation is a good indicator progress is being made. More announcements
of improvements and applications over the next several years are expected. However,
application to higher volume, lower-cost vehicle segments is not expected any time
soon.

> Read full chapter

Advanced and conventional internal

combustion engine materials
L.L. Myagkov, ... I. Makhkamova, in Alternative Fuels and Advanced Vehicle Tech-
nologies for Improved Environmental Performance, 2014

11.3 Graphite/carbon and carbon/carbon fibre-reinforced poly-

mer composites (CFRPs)
Carbon-fibre reinforced polymers offer good potential for the manufacture of con-
rods. Figure 11.4 shows a compound conrod made of such material. These materials
are additionally reinforced by 5–20 micron dispersed ceramic nanoparticles, such as
Si3N4, Al2o3 and B4C which may make up between 5 and 30% of the material volume.
These materials have relatively high strength and stiffness at low density and the
anisotropic distribution of mechanical properties is advantageous in the manufac-
ture of certain engine components such as conrods. The mechanical properties of
carbon-fibre reinforced polymer used in the compound conrod of the R4 (82/71)
engine are compared with steel properties in Table 11.3. The material is made of
two monolayer polymers which differ in the orientation of their fibres. The main part
of the conrod and its cap are made of two monolayer polymers mixed in different
proportions.

11.4. A carbon-fibre-reinforced polymer conrod of R4 (82/71) engine.

Table 11.3. Mechanical properties of carbon fibreglass material

Properties Material
Monolayer Monolayer Steel Main part of the Cap of the conrod
(0°±90°) (0°±45°) conrod
, g/cm3 2.0 2.0 7.8 2.0 2.0
Ex × 10 −3 MPa 125 121 200 92 80

Ey × 10−3 MPa 8.4 8.35 – 34.2 46

μxy 0.29 0.29 0.3 0.133 0.1085

μyx – 0.0494 0.024
Gxy × 10− 3- 4.3 4.5 77 6.6 6.98
 MPa
+x/ −x MPa 1,050/950 900/850 1,000 226.7/666.6 195.6/592.2
+y/ −y MPa 20/120 30/118 – 82.1/258.5 111./348.3
xy MPa 60 60 800 111.2 –
Specific 53.0/48.0 4.6/4.3 9.3 12/34 10.0/30.2
strength, 
+x/ −x, m ×
10−3
Specifc 0.1/0.6 0.15/0.6 9.3 4.2/13.2 5.7/17.8
strength,  
+y/ −y m ×
10−3
Specifc stif- 6,400 6,200 2,600 4,700 4,100
fness, Ex/ *,
m × 10−3
Specific stif- 430 430 – 1,750 2.350
fness, Ey/ *, m
× 10−3

Significant efforts are currently being made to develop designs for pistons made
of graphite/carbon fibre or carbon/carbon fibre type composite materials. These
materials have a graphite or carbon matrix strengthened by carbon fibres. The
strengthening of the material usually takes place in four directions, giving the
material isotropic mechanical properties on the macro scale. The main advantage of
these materials is their combination of comparatively low density and high strength.
They also have stable mechanical properties over the typical temperature range for
internal combustion engines. The high cost of these materials currently prevents
their large scale application in engine production. Table 11.4 shows the mechanical
properties of a graphite/ carbon fibre composite material used in the production of
prospective piston designs.

Table 11.4. Mechanical properties of a graphite-carbon fibre composite material

Bulk density, g/cm3 1.8–2.0

Young (elastic) modulus, GPa 13–16
Bending strength, MPa 80–120
Compressive strength, MPa 150–250
CTE, 10− 6 K− 1 5–7
Thermal conductivity, W/mK 40–60
Specific stiffness, km 6,800
Specific bending strength, km 5.0
Specific compressive strength, km 10.0

> Read full chapter

The impact of graphene oxide on ce-

mentitious composites
Alyaa Mohammed, ... Nihad T.K. Al-Saadi, in Nanotechnology in Eco-efficient
Construction (Second Edition), 2019

4.5 Some structural applications of GO/cement composites in

repairing of reinforced concrete
Carbon fiber reinforced polymer (CFRP) is one of the composite materials which
is used in both repairing and strengthening of reinforced concrete structures. The
usage of epoxy-based adhesives (and organic ones in general) incorporates disad-
vantages such as flammability of the resin or the issue of poisonous fumes (Täljsten
and Blanksvärd, 2007). Moreover, there are other issues such as susceptibility to
exposure to very low temperatures such as 70°C (Gamage et al., 2006), sunlight,
and UV radiations (Ombres, 2011). When the surface has some humidity or the
environmental temperature is lower than −10°C, the application of these composites
is problematic (D'Ambrisi and Focacci, 2011). Therefore, it seems necessary to use
an alternative adhesive such as polymer cement-based to overcome these issues.
There are some evidences that show the success of using these alternative mate-
rials in the method of strengthening by CFRP (Al-Saadi et al., 2016). Nonetheless,
polymer cement-based adhesives might have performance affected by the situation
in which the polymer is used such as its hydrothermal situation. Another type of
adhesive is a nonpolymer cementitious one, which has remarkable properties such
as suitable bonding and appropriate resistance to ecological situations. Moreover,
disadvantages such as flammability or release of poisonous fumes are not associated
with these adhesives (Hashemi and Al-Mahaidi, 2012). This section presents some
applications in which GO has been used to produce nonpolymer cementitious
adhesives.

Innovative high-strength self-compacting nonpolymer cement-based adhesive

(IHSSC-CA) is a mixture incorporating GO nanosheets and was used in a previous
study (Mohammed et al., 2016a). It is a very high strength mixture showing out-
standing tensile and compressive strength values of 18 and 116 MPa after 28 days
respectively (Mohammed et al., 2018c). Other properties of IHSSC-CA are given in
Table 4.1.

Table 4.1. Properties of the innovative cementitious adhesive (IHSSC-CA) (Mo-

hammed et al., 2016a)

Flow Initial set- Final set- 28-day 28-day Bond Permeability 

ting time ting time tensile compres- strength × 10−16 m2
(min) (min) strength sive (MPa)
(MPa) strength
(MPa)
7.5% 120 420 13.8 101 1.2 0.023

IHSSC-CA has been used in our previous studies and the results obtained for
all applications including stiffness, ductility, bond strength, etc. are promising.
Strengthening of fiber-reinforced concrete with the near-surface mounted CFRP
(NSM-CFRP) technique at different loading conditions has been studied. We showed
that it is more convenient to apply IHSSC-CA for NSM-CFRP than other common
adhesives such as epoxy-based and cement-based polymer ones. This is because
of the unique properties of IHSSC-CA including better flowability and workability
(Al-Saadi et al., 2018). Further, IHSSC-CA showed better pull-out and bond strength
results in NSM-CFRP applications than epoxy-based and cement-based polymer
adhesives (Al-Saadi et al., 2018). Fig. 4.4 shows the application of IHSSC-CA with
CFRP strip and Table 4.2 shows the pull-out test results.

Figure 4.4. Application of IHSSC-CA with CFRP strips (Al-Saadi et al., 2018).

Table 4.2. Pull-out test results (Al-Saadi et al., 2018)

Specimens ID Ultimate Ultimate Ultimate axial CFRP strip Failure mode

pull-out force bond stress (MPa) utilization
(kN) strength
(MPa)
Innovative 34.5 4.80 1233 0.34 Rupture of
cementitious CFRP strip
adhesive
(IHSSC-CA)
Polymer ce- 22.3 3.09 794 0.21 Pull-out of
ment adhesive CFRP strip
Epoxy adhesive 41.1 5.70 1467 0.39 Debonding of
CFRP strip

Another feature of using IHSSC-CA is its ability to reduce stress concentration.

Physical analysis (Fig. 4.5) of CFRP connected to reinforced concrete through the
application of IHSSC-CA showed a lower stress concentration than samples with
epoxy-based and cement-based polymer adhesives (Mohammed et al., 2017c). Af-
ter the pull-out test, a remarkable change in the pore network and structure of
IHSSC-CA was observed. The behavior is in contrast with epoxy-based and ce-
ment-based polymer adhesives which show slight change in the pore structure
under loading. This implies the suitable behavior of the applied IHSSC-CA adhe-
sive under the applied load (Fig. 4.6) (Mohammed et al., 2017c). Further, 3D laser
profilometry analysis results shown in Fig. 4.7 reveal that a very rough surface is
acquired for IHSSC-CA adhesive after pull-out tests rather than epoxy-based and
cement-based polymer adhesives (Mohammed et al., 2018b).
Figure 4.5. Images of the bond area after pull-out testing (Mohammed et al., 2017c).

Figure 4.6. Nitrogen adsorption isotherms for IHSSC-CA and PCA (Mohammed

et al., 2017c).
Figure 4.7. 3D and 2D images of the topographic surface of CFRP strip: (a) with
IHSSC-CA, (b) with PCA, and (c) with epoxy (Mohammed et al., 2018b).

High temperature testing of IHSSC-CA showed that this adhesive is able to main-
tain around 60% of its strength even at elevated temperatures (Mohammed et al.,
2016b). Fig. 4.8 shows the compressive and tensile strengths at different tempera-
tures for normal-strength concrete (NSC) and IHSSC-CA. The adhesive used in this
study had a thickness of 5 mm. Further experiments showed that this thickness is
insufficient to protect samples from the heat. An increased thickness of 20–25 mm
was then applied as the protective heat-resistance cover. Test results showed better
bond strength when the thickness increased (Fig. 4.9). It was revealed that higher
temperatures require higher thickness of the IHSSC-CA adhesive where for 600
and 800°C heat exposure, the application of at least 20 and 25 mm (respectively)
of adhesive is essential (Mohammed et al., 2016b).

Figure 4.8. Compressive and tensile strengths of tested specimens (Mohammed

et al., 2016b).
Figure 4.9. Average bond strength versus temperature relations (Mohammed et al.,
2016b).

The results of fatigue tests on concrete samples joined to NSM-CFRP through

the application of IHSSC-CA were better than samples with epoxy-based and ce-
ment-based polymer adhesives, especially at longer lives (Fig. 4.10). The analysis of
pore structure of specimens subject to fatigue loading shows a better composite
behavior for IHSSC-CA than other polymer-based adhesives (Al-Saadi et al., 2017a).
Physical examination after fatigue and post-fatigue tests (Fig. 4.11) shows uniform
stress distribution and therefore, suitable composite behavior is observed between
NSM-CFRP and IHSSC-CA (Mohammed et al., 2018c). This is against epoxy-based
and cement-based polymer adhesives where a nonuniform stress distribution was
observed, as shown in Fig. 4.12 (Al-Saadi et al., 2017a).

Figure 4.10. Load range-fatigue life relationships for IHSSC-CA (FR20IC) and poly-
mer cement-based (FR20 C) adhesives (Al-Saadi et al., 2017a).

Figure 4.11. Nitrogen adsorption isotherms for IHSSC-CA (R20IC) and polymer

cement-based (R20 C) adhesives (Al-Saadi et al., 2017a).
Figure 4.12. Images of CFRP strip and a bond area of tested specimens: (a) IHSSC-CA
and (b) polymer cement-based adhesive (Al-Saadi et al., 2017a).

In a previous study, deflection capacity (ductility) and flexural strength of reinforced

concrete beams strengthened by NSM-CFRP using IHSSC-CA adhesive showed
improvement compared to a control specimen (Al-Saadi et al., 2017b). Despite
IHSSC-CA, strengthening by using epoxy-based adhesives showed decreased duc-
tility and flexural strength as shown in Fig. 4.13 (Al-Saadi et al., 2017b). Fig. 4.14
shows the deformation behavior of reinforced concrete beams strengthened by
both epoxy-based and IHSSC-CA adhesives. This figure shows the suitability of
using IHSSC-CA adhesive in creating large deformations compared to epoxy-based
adhesive which caused abrupt rupture (Al-Saadi et al., 2017b). Further, as Fig. 4.13
shows, the application of epoxy-based adhesive to strengthen reinforced concrete
beams causes the appearance of no residual strength after the ultimate strength. The
post-cracking behavior of specimens strengthened by IHSSC-CA is different and a
large amount of residual strength (close to 87%) sustains after reaching its ultimate
strength (Al-Saadi et al., 2017b).
Figure 4.13. Load versus midspan deflection relations (Al-Saadi et al., 2017b).

Figure 4.14. Failure modes of tested beams: (a) with IHSSC-CA and (b) with epoxy
adhesive (Al-Saadi et al., 2017b).

Results of our previous studies show the effectiveness of using IHSSC-CA as ad-
hesive for repairing and strengthening of reinforced concrete beams. Values such
as deflection and strain are higher in structures strengthened by epoxy-based ad-
hesives. Further, the proposed IHSSC-CA adhesive has the capability for in-situ
applications (Al-Saadi et al., 2017c,d). Analysis of the fatigue behavior of reinforced
concrete beams strengthened by the IHSSC-CA adhesive shows the effectiveness of
this binder as better stress transfer and bond strength are attainable compared to
epoxy-based adhesives, as shown in Figs. 4.15 and 4.16. This behavior shows once
again that the utilized IHSSC-CA adhesive has the ability of maintaining compos-
ite action even under fatigue loads (Al-Saadi et al., 2017c,d). Besides mechanical
performance, IHSSC-CA is more convenient for on-site applications compared
to epoxy-based adhesives because its workability, flowability, and self-compacting
properties are better and create a uniform and smooth bonding layer (Al-Saadi et al.,
2017c).

Figure 4.15. Midspan deflection versus fatigue life relations for tested beams (-
Al-Saadi et al., 2017c).

Figure 4.16. Crack width versus fatigue life relations for tested beams (Al-Saadi et al.,
2017c).

Fig. 4.17(a) shows that reinforced concrete beams with IHSSC-CA adhesive upon the
rupture of CFRP strip. This happens once the concrete cover in the compression zone
crushes and reinforced steel bars reach to their yield strength. In other words, there
is no degradation between the adhesive and CFRP strip under fatigue loads (Al-Saadi
et al., 2017d). The results for CFRP strips joined to reinforced concrete through
epoxy-based adhesive in Fig. 4.17(b) show an abrupt failure after the crushing of
concrete cover and the yielding of steel bars. In other words, because of bond
degradation between CFRP and epoxy-based adhesive, this type of failure occurs
(Al-Saadi et al., 2017d). Ductility of beams reinforced by IHSSC-CA also increased
compared to epoxy-based adhesives (Al-Saadi et al., 2017d). Less deformation of
the surface of CFRP strips was proved by studying the results through 3D laser
profilometry, as shown in Fig. 4.18. The less deformation means higher capacity of
repaired and strengthened reinforced concrete to resist fatigue loading (Al-Saadi
et al., 2017d).

Figure 4.17. Failure modes of tested post-fatigue beams: (a) with IHSSC-CA and

(b) with epoxy adhesive (Al-Saadi et al., 2017d).
Figure 4.18. 2D and 3D images of the topographic surface of CFRP strip (a) with
IHSSC-CA and (b) with epoxy adhesive (Al-Saadi et al., 2017d).

> Read full chapter

Haptic-based virtual reality system to

enhance actual aerospace composite
panel drilling training
Sivadas Chandra Sekaran, ... Razman Shah Rajab, in Structural Health Monitoring
of Biocomposites, Fibre-Reinforced Composites and Hybrid Composites, 2019

7.1.1 Drilling carbon fiber-reinforced composites

Carbon fiber reinforced polymers (CFRPs) are increasingly replacing conventional
metallic alloys in aircraft structures. This is mainly due to tensile strength carried by
CFRP, which falls between 1500 and 3500 MPa, whereas its metallic counterparts
such as aluminum and steel only possess tensile strength of 450–600 MPa and
750–1500 MPa, respectively. Moreover, a relatively lower density at around 1.5–2.0-
 g/cm3 of CFRP makes it the ideal choice for being a major structural component
of aircrafts [1]. However, machining CFRP presents its own set of difficulties
such as delamination, fiber pullout, and fiber breakout [2]. Due to the unpredictive
nature of composites, it is important to avoid these issues from the machining stage
itself to guarantee their strength against fatigue and assembly tolerance [3]. Drilling
operation of aerospace-grade CFRP, being the major cutting process involved in
CFRP machining [4], faces delamination as the major defect mode.

Multiple studies have been conducted on identifying the causes and solutions for
CFRP during delamination. Several factors that impact the products of drillings
are still being investigated. The important variables are identified as thrust force,
drilling speed, and feed rate. Among them, increased thrust force is identified to
be the significant contributor to tool wear as well as composite's delamination
[4]. Although there are the latest computer numerical controlled (CNC) drilling
machines that can set and perform drilling at fix thrust force, drilling speed, and
feed rate, a great portion of CFRP drilling operation in aerospace industry is still
being conducted using hand-held pneumatic drill machines (Fig. 7.1). Despite the
existence of automated drilling, larger aircraft panels, especially irregular shaped,
that make up parts such as spoilers, fairings, and flight control surfaces, require
technicians to use a hand-held drill that enables them to perform drilling from
different angles. Most often, the drill feed rate and spindle of the machine can be
fixed, however, the driller is still responsible for maintaining the thrust force within
the threshold, unique to the material that is being drilled. Exceeding the thrust force
during drilling of composites causes significant delamination damage to the final
layer of the panel, called fiber pushout.

Figure 7.1. Pneumatic drill.

Moreover, when it comes to hand-held drilling, drilling angles play a vital role is
producing a perfect hole. Any deviation in angle would cause the hole to be elliptical
instead of a perfect circle, which in turn would cause the part's assembly tolerance
to be compromised. Therefore, it is crucial for the driller to perform the drilling,
perpendicular to the surface, and maintain the required thrust force throughout the
drilling.

Since it is impossible to quantitatively teach technicians on drilling at a certain

thrust force and perpendicular angle, developing motor skills is vital. In large
aerospace manufacturing facilities, technicians are trained in-house to develop basic
aerospace composite drilling skills. Performing a proper drilling not only produces
a high-quality product close to a set standard but also increases the life of the tool
itself. At higher-than-required thrust force and cutting speed, change in geometry
of drill bit due to tool wear is clearly observed [3].

Based on the Dreyfus model of skill acquisition as shown in Fig. 7.2, training and
acquiring a physical skill would require the technician to go through five stages,
namely novice, advanced beginner, competence, proficiency, and expertise. The
transition from one stage to another requires theoretical knowledge and experience
until it ultimately becomes intuitional at the expert stage [5].

Figure 7.2. Dreyfus model of skill acquisition.

Therefore, traditionally, in order to develop expert level drillers, aircraft manufac-

turing facilities have to train them for an extensive amount of time using actual
specimens, which in this case are the carbon fiber composites. Unfortunately, this
is a major challenge due to 9–10 years of backlog and the need to lower the
operation and manufacturing cost in aircraft manufacturing facilities. To aggravate
the situation, the attrition rate among aerospace technicians is high. Most facilities
use walk-in interview methods to quickly employ more technician to fill the man
power, however, the skill lost due to attrition is not sustained. Training these new
employees to perform CFRP drilling will take a tremendous amount of time and
resources. With spiking global demand on aircraft, and composites to satisfy the
requirement [6], alternative methods need to be adopted to train technicians quicker
with lower cost.
The traditional training system that measures the performance of a trainee via
postprocessed, product-based, qualitative evaluation contributes very little to skill
attainment. Current practice in drilling training facility is to allow the trainees to
drill a fixed number of holes on CFRPs while being monitored for posture and drill
orientation by a supervisor. The drilling quality is then tested by observing the quality
of the hole using equipment called a go-no-go gauge as well as visual and tactile
inspection on the final layer of the CFRP. Error due to angle will cause the panel to fail
the go-no-go gauge test if the hole exceeds a few microns (10−6 m) of tolerance. The
error done due to not maintaining the optimal thrust force is observed by overshoot
made by the drill bit and physical damage caused on the last layer of the composite
panel. Although the evaluation method is effective in predicting the quality of the
machined product, they play a minimal role in training the technician. With no
information on their mistakes done during (real-time) the task, it will require a much
longer time to correct themselves and improve.

Besides time, providing structured training to technicians also requires extra costs
in terms of materials, instructors, and tools. Technicians should be trained to drill
several types of panels, starting from flat shaped to curved. Actual aerospace CFRP
panels are too expensive for training purposes, therefore most facilities leverage
on scrapped CFRP panels or different materials for the training. This irregularity in
the training regime compromises the effectiveness of the training since the shape,
strength, toughness, and texture of the material greatly differs from the actual panel
that the technicians are required to work on in the future.

> Read full chapter

Virtual testing of impact in fiber rein-

forced laminates
S. Sádaba, ... J. LLorca, in Structural Integrity and Durability of Advanced Compos-
ites, 2015

10.1 Introduction
Carbon fibre–reinforced polymers are nowadays extensively used in applications
where outstanding mechanical properties are necessary in combination with weight
saving. Good examples are the Airbus A350 XWB and the Boeing 787 Dreamliner,
where fibre composites are used for up to ≈50% of wings, fuselage sections and
tail surfaces. Although the demanding in-plane loads are perfectly covered by the
high strength and stiffness of the carbon laminates, it is well known that such
laminates are very sensitive to out-of-plane loads coming from impacts during
maintenance (i.e. tools drops) and service operations (e.g. runway debris or bird and
lightning impacts) (Abrate, 1998). The traditional strategy to guarantee structural
integrity is based on extensive and costly experimental campaigns in which the
burden of testing is immense (up to ≈104 tests are required for certification of
an airframe structure (Cox & Yang, 2006; MIL-HDBK-17-1F, 2002)). Experimental
tests include those ranging from simple coupons (tension, compression, shear) to
small components (damage tolerance) up to the final composite structure (fuselage
barrel or wing). Recent developments in material modelling, together with increased
computational power and improvements in computing tools, are rapidly changing
the aforementioned scenario of costly and time-consuming testing. Nowadays it
is becoming possible to accurately predict the behaviour until failure of composite
materials and small components by computer-assisted virtual testing (LLorca et al.,
2011; LLorca, González, Molina Aldareguía, & Lopes, 2013).

Structural composites are manufactured by stacking different composite layers

(unidirectional or woven) into a laminate with a given stacking sequence. Different
damage mechanisms co-exist during composite deformation and can lead to lami-
nate failure. Fracture due to tensile stresses parallel to the fibres is controlled by the
tensile fracture of fibres, while compressive stresses along the fibres lead to kinking.
Stresses perpendicular to the fibre induce failure by matrix cracking in tension or
shear and interface decohesion. In addition, interply delamination is a typical failure
mechanism in laminates due to the thermoelastic mismatch between adjacent plies.
Accurate models to predict the failure of composite laminates should include all
these failure modes as well as the complex interactions among them.

Modelling of impact damage presents additional difficulties because it is im-

portant to take into account the dynamic structural behaviour and the local effects
at the impact point. As a result, the use of analytical models often results in an
oversimplification of the problem and thus inaccurate predictions. On the contrary,
numerical simulations based on the finite element model can account for the
complex constitutive behaviour of the composite material at the local (element)
level together with the simulation of complex structural behaviour under complex
external loads and boundary conditions.

The present chapter presents a virtual testing methodology to simulate the mechan-
ical performance of composite laminates under impact by means of computational
mesomechanics. This strategy is based on fully coupled modelization of intralaminar
and interlaminar failure mechanisms in composite laminates, and is described in
detail in Section 10.2. Intralaminar failure is addressed within the framework of con-
tinuum damage mechanics (CDM). Interply damage by delamination is taken into
account by a cohesive crack approach that uses either cohesive elements or cohesive
surfaces between adjacent plies. Two applications of this modelling strategy devoted
to low- and high-velocity impact are presented in Sections 10.3 and 10.4 respectively.
Future trends and expected developments in the field are briefly outlined in the last
section of this chapter.

> Read full chapter

Pultrusion of advanced fibre-rein-

forced polymer (FRP) composites
Ramôa CorreiaJ. , in Advanced Fibre-Reinforced Polymer (FRP) Composites for
Structural Applications, 2013

9.9.3 Strengthening strips

As already mentioned, in most civil engineering strengthening applications, FRP
strips are composed of carbon fibres embedded in epoxy resin. Table 9.5 presents
typical ranges of variation for the physical and mechanical properties of CFRP
strips. Compared to the alternative traditional strengthening solutions of concrete
jacketing and, particularly, steel plate bonding (the main competitor), the main
advantages of CFRP strips are very high tensile strength, high stiffness, comparable
to steel, relatively high deformation capacity, lightness, which facilitates on-site
application, reduces labour costs and lessens the increase of weight in the structure
to be strengthened, and finally non-corrodibility.

Table 9.5. Typical physical and mechanical properties of CFRP strengthening strips

Property Test standards Standard modu- High modulus Ultra-high modu-

lus lus
Density (g/cm3) ISO 1183, ASTM D 1.50–1.60 1.50–1.60 1.50–1.60
792
Fibre content (%) ISO 1172, ASTM D 65–70 65–70 65–70
3171
Nominal thickness – 1.2–1.9 1.2–1.4 1.4
(mm)
Width (mm) – 50–120 50–120 50
Axial tensile 2690–2800 1290–2800 1800
strength (MPa)
Axial elasticity 155–165 210–300 400
modulus (GPa)
Axial ultimate 1.8 1.35 0.45
strain (%)

However, CFRP strengthening systems have certain limitations (some of them

similar to those already described for FRP profiles and bars, intrinsic to their nature):

• Linear-elastic behaviour up to failure, which can limit the ductility of the

strengthened element
• Limited exploitation of material performance due to the occurrence of prema-
ture adherence failure mechanisms, which can be prevented to some extent by
using anchorage devices
• The strength, stiffness and bond properties of CFRP systems are severely de-
teriorated at moderately elevated temperatures – if the strengthening systems
are to be exploited when designing for the fire load combination, appropriate
thermal insulation is needed (Williams et al., 2008; Ahmed and Kodur, 2011;
Firmo et al., 2012).

> Read full chapter

Development of service life prognosis

systems for hydrogen energy devices
P.E. Irving, in Gaseous Hydrogen Embrittlement of Materials in Energy Technolo-
gies: Mechanisms, Modelling and Future Developments, 2012

13.4.8 Wrapped pressure vessels

Wrapped cfrp filament wound storage vessels with steel, aluminium or polymer
autofrettaged liners will suffer from more complex failure mechanisms than mono-
lithic steel vessels. These pressure vessels are currently preferred for on vehicle use
because of their relatively light weight. Design and fatigue performance of these
vessels have been investigated by a number of workers [e.g. 58–62].

The cyclic loading inputs to a SIPS for on vehicle hydrogen storage, determined by
the pressure cycling will be the same as the monolithic structures. The failure mode
primarily responsible for hydrogen leakage will be once again fatigue crack initiation
in the inner skin. Fatigue cracks may initiate here permitting leakage without
propagating into the filament wound composite region. In addition pressure cycling
may cause delamination development within the filament wound layer, leading to
inadequate support of the skin and promoting local fatigue crack initiation. A further
possible damage mode of this type of storage would be external mechanical damage
to the filament wound region, once again leading to delamination of the filament
winding and local initiation of fatigue cack in the storage vessel lining. It is probable
that because the cylinder liner is relatively thin, the number of fatigue pressure cycles
occupied in growing the initiated crack to a size causing leakage will be reduced.
Consequently, the period of service for which the crack is detectable will be reduced,
and there will be a smaller warning period before hydrogen leakage starts than in the
monolithic vessels. The possibility that aluminium alloy cracking is not enhanced by
H2 is considered in Chapter 19 in Volume 1.
> Read full chapter

Volume or Inside Heating Eddy Cur-

rent Thermography
Yunze He, ... Ruizhen Yang, in Transient Electromagnetic-Thermal Nondestructive
Testing, 2017

9.3.2 Impacted Sample

A CFRP sample with impact damage was tested. It consisted of 12 layers of 5 HS
carbon fiber, woven with balanced woven fabric. The polymer matrix was composed
of polyphenylene sulfide, a thermoplastic resin system. The size of this plate was
100 × 150 mm2 and the average thickness was 3.78 ± 0.05 mm. The sample had a
volume ratio of 0.5 ± 0.03 and a density of 1460 kg/m3. The plates were manu-
factured by TenCate Advanced Composites, Netherlands. Impact damage of 12 J
was fabricated in the middle of the sample [229]. This impact led not only to a
broken surface structure, but also caused inside delamination [145]. The existence
of delamination can be predicted and identified by thermography testing [145]. In
this experiment, the coil and camera were located on the back and front sides of
the sample, respectively. The heating time was set as 1 s and the cooling time as
0.5 s. The acquisition frequency (fs) was 100 Hz and there were 150 sampling points
(N). The frequency resolution (fr = fs/N) was 0.6667 Hz. Fig. 9.8A shows the VHST
thermal image at 50 ms in the heating phase. Carbon fiber (hot area) and matrix
(cooler area) can be separated. Fig. 9.8B shows the VHPT thermal image for the
front side at 1.2 s in the cooling phase. The circle-shaped impact area can be seen
as a hot area but the expected dark area (delamination caused by an impact) cannot
be observed. Fig. 9.8C shows temperature responses for three points. As shown in
Fig. 9.8A–B, P1 is located on a sound part of carbon fiber, P2 is located on carbon
fiber at the impact edge, and P3 is located in the middle of the impact area. The
temperature–time curve is not linear because of broken structures in the sample and
it decreases slowly in the cooling phase. P3 immediately shows a lower temperature
than P1 and P2, indicating delamination is very shallow. Phase spectra for the three
points are shown in Fig. 9.8D. Phase spectra were modulated by a period (fm) which
is 1 Hz, the reciprocal of heating time (1 s). The phasegrams for the front side at
0.3906 Hz (the peak of phase spectra) are shown in Fig. 9.8E. It is found that both
the nonuniform heating effect and the blurring effect have been reduced. Broken
surface damage can be detected in the red area, and invisible damage caused by the
impact can be detected in the blue area (as marked by ellipses). The phasegrams for
the front side at 2.7344 Hz are shown in Fig. 9.8F. It can be seen that although both
the nonuniform heating effect and the blurring effect have been reduced, periodic
carbon fiber structures including broken carbon fiber structures can be observed.
Comparing Fig. 9.8E with Fig. 9.8F, useful information (inside delamination) and
background noise (periodic structure) can be divided into different phasegrams. It
should be noted that it is difficult to image defects in a woven composite. VHPPT
is a solution for this situation. Two-dimensional FFT and frequency-domain filters
are other solutions to eliminate periodic woven structures.

Figure 9.8. (A) VHST thermal image at 50 ms (unit: DL), (B) VHPT thermal image
at 1.2 s (unit: DL), (C) temperature responses for three points (unit: DL), (D) phase
spectra for three points (unit: radian), (E) phase image at 0.3906 Hz (unit: radian),
and (F) phase image at 2.7344 Hz (unit: radian) [143].

In a single experiment, VHPPT can observe delamination/impact and woven struc-

tures using phasegrams with different frequencies. It can be predicted that volume
lock-in thermography can obtain phasegrams to show delamination/impact and
woven structures, respectively. However, it requires repetitions of experiments under
different lock-in frequencies, which is more time consuming than VHPPT.

In this chapter, VHT and IHT have been proposed through electromagnetic excita-
tions like high-frequency eddy currents, microwaves, and terahertz waves for the
quantified inspection and characterization of advanced composite materials. More
specifically, three VHT methods have been built: VHST, VHPT, and VHPPT. After
basic theory analysis, both simulation and experimental studies have been carried
out to verify the performance of characterizing defects in CFRP. The conclusions
are: (1) with VHST, the linear temperature–time curve and separation time can
be used to detect and quantify defects in MUT; (2) with VHPT, the temperature
response can be fitted as a power function and it is a linear line in logarithm domain.
Separation time is still valid to quantify the depth variance; (3) with VHPPT, depth
variance can be quantified by specific features like conventional surface heating pulse
phase thermography. Nonuniform heating effects and carbon fiber structures can
be eliminated from phasegrams, which is useful for characterizing inside defects in
woven composites; and (4) the above conclusions are valid under both reflection and
transmission modes, and the results under both modes are very similar. As expected,
VHT and IHT have a potential application for quantified inspection and character-
ization of composites, polymers, biomaterials, and the structures and components
made of these materials [230].

However, there are a few limitations. First, the simulation and the experiment are
not the same. The reason is that the 3D model of CFRP, which considers microscopic
parameters has not been built. Second, absolute quantification information about
depth variance has not been obtained, and the maximum detectable depth has
not been evaluated. Third, due to the very slow development of contrast between
sound and defective regions (when the contrast emerges from the noise) during
experiments, uncertainty of defect depth is very high. As a matter of fact, this
problem is identical in the case of surface heating thermography. This problem has
not been investigated in this section, but the early detection method is a potential
solution for this problem [231]. Fourth, comparison between volume heating and
surface heating (flash lamp or laser stimulation) has not been drawn. Fifth, volume
lock-in thermography has not been investigated.

> Read full chapter

Design and testing of crashworthy aero-

space composite components
A.F. Johnson, ... M.W. Joosten, in Polymer Composites in the Aerospace Industry,
2015

10.3.1 Energy absorbing mechanisms in composite structural

elements
Although CFRP are generally brittle in their tensile behaviour with low failure
strains, it has been shown that composite structural elements that fail in other
failure modes such as local compression, shear, microbuckling or delamination
have the ability to absorb energy through a controlled progressive crush failure. The
key is to initiate a progressive crush mode of failure where the absorbed energy
is considerably higher than that measured in brittle failures. This is achieved by
trigger mechanisms (also known as crush initiators) introduced into a structural
element, which reduce peak crush forces and invoke a progressive failure mode that
propagates through the structure. Many different progressive failure modes have
been observed that depend significantly on fibre type, matrix type, fibre–matrix
interface, fibre stacking sequence and fibre orientation, specimen geometry, mode
of loading, design of trigger, strain rate, temperature effects etc. Research on energy
absorbing failure modes has continued for over 30 years covering all aspects from
composites micromechanics, through macromechanics at the ply/laminate level, up
to trigger concepts and design at the structural level.

Early work by [32–37] showed that glass or CFRP tubes loaded in axial compres-
sion through a bevelled or chamfered tube end (a chamfer trigger) could fail in a
progressive crush mode, provided tube wall thickness was high enough to prevent
premature global or local buckling failures. Through this work, the main crush
failure mechanisms were identified as a splaying or fragmentation mode. Splaying
is initiated at a chamfer trigger by delaminations at the crush front that cause plies
to bend or splay outward leading to axial splitting and bending or transverse shear
failures. Depending on the ductility of the composite, the splayed plies or fronds
will be short due to fragmentation, or long. Fragmentation of the shorter fronds at
the crush front leads to a debris wedge, which then drives the delamination failures
and leads to a steady crush failure mode with high EA. This is a basic mechanism
of composite tube absorbers under axial crush loads. An important step in the
development of crashworthy composite structures was to show that similar failure
mechanisms could be triggered in nontubular structural elements, such as open
circular segments and sine wave webs [27,36], and box sections [38].

To measure and understand the EA and failure mechanisms of composite structural

elements requires the development of new test procedures and appropriate test
specimens. Feraboli [9] reviews current test procedures and discusses the character-
istics of flat plate and corrugated specimens for axial crush tests. The EA properties to
be measured are obtained from specimens that fail in a steady crush mode, requiring
specially designed support fixtures or self-supporting specimens such as tubes and
cylindrical profiles. DLR has developed chamfered tube segment specimens, see
[28], which are easier to fabricate than tubes, are self-stabilising without lateral sup-
ports and give reproducible axial crush failures under both quasi-static and dynamic
loading conditions. Quasi-static tests on the chamfered tube segment specimens
are discussed here, and dynamic crush tests are described further in [39]. The test
specimens consist of a half-circle cylindrical segment with flanges at the longitudinal
free edges to inhibit local buckling in axial compression. They were manufactured
from Hexply M18/1/43%/G939-220gsm 4HS carbon fabric/epoxy prepreg with eight
fabric plies and a [0/90/0/90]s layup. Dimensions are length 75 mm, diameter 50 mm
at flanges, nominal thickness 2.0 mm. The trigger mechanism was a 45° outside
chamfer, which initiates a steady crush process at the chamfer tip. The base of the
segment specimens is fixed in a support clamp to a lower steel platen mounted on
the load cell of a quasi-static test machine. The chamfered edge is loaded in axial
compression by a steel plate attached to the crosshead, which loads the specimen at
1 mm/s. Axial force time and crosshead displacement time are measured in the test
for typical crush distances of 50 mm.

Figure 10.4 shows images from the film of the crush test showing three phases,
crushing down of trigger, splaying of the laminate into inner and outer laminae
with a central delamination crack, followed by axial splitting of laminae bundles into
petal-like fronds, as described in more detail in [40] and seen in the final crushed
specimen in Figure 10.5(a). The steady-state crush phase consists of frond formation
driven by the central delamination crack with a debris wedge, Figure 10.5(b), which
shows a high resolution computed tomography (HRCT) X-ray scan of the segment
wall cross-section at the crush front. The measured crush force–displacement curve
of the segment specimen is plotted in Figure 10.6 up to a displacement of 40 mm.
The curve indicates a rapidly increasing load with peak at about 2 mm displacement,
which corresponds to loading the chamfer trigger. This is followed by a sharp drop
in load as the trigger fractures with load falling to a constant mean value in the
steady-state crushing phase. This steady crush is the basic mechanism used in the
development of composites energy absorbing structures. Key parameters for design
of such devices are the steady-state crush force (SSCF), the EA, defined as the area
under the crush force–displacement curve and SEA, which is the energy absorbed per
unit mass of fractured material. These data are shown in the first row of Table 10.1
with SSCF of 24 kN and SEA of 87.1 kJ/kg, which are typical values for quasi-static
DLR segment tests with M18/1 carbon fabric prepreg materials. In an extended
study of these specimens with different laminate layups and manufacturing variants
reported in [41]; the SEA range of 79.4–90.6 kJ/kg was measured as layup was varied
with different proportions of 0° and 45° fabric plies to the load axis.

Figure 10.4. Quasi-static crush test sequence on carbon fabric/epoxy tube segment.

Figure 10.5. Carbon fabric/epoxy segment specimen after quasi-static crush test. (a)
Crushed segment; (b) High resolution computed tomography scan of crush front.
Figure 10.6. Crush force–displacement curves for carbon fabric/epoxy segment
specimens at different loading rates.

Table 10.1. Energy absorption performance of carbon fibre reinforced polymer

segments at different loading rates

Loading rate Peak crush SSCF (kN) Crush dis- Energy ab- SEA (kJ/kg)
(m/s) force (kN) tance (mm) sorbed (kJ)
0.001 31.86 24.03 58.48 1.420 87.11
2.00 31.01 17.71 49.29 0.891 67.89
5.00 27.88 17.80 45.67 0.825 67.53
10.00 35.69 18.04 41.62 0.741 67.04

Figure 10.6 also includes equivalent force–displacement curves on similar segment

specimens tested dynamically at higher loading rates at velocities 2 m/s, 5 m/s and
10 m/s in the Instron VHS 100/20 High Strain Rate Test machine at the DLR, see
[39] for more details. This shows that the dynamic specimens exhibited a steady
crush response after triggering at an initial peak load. The most interesting feature
of the curves is that the three dynamic curves were very similar with no obvious
dependence on the test velocity, and all were significantly below the quasi-static
test curve. Table 10.1 summarises the main results of the quasi-static and dynamic
tests. This shows a reduction of 22% in SEA and 26% in SSCF at 2 m/s compared with
the quasi-static values. These results are in line with previous tests done at the DLR
on crash elements and are explained by the more brittle behaviour of the composite
at higher rates causing shorter fronds at the crush front, which fragment sooner,
absorbing less energy. The data are important for the designer since composite
absorbers are designed for crash loads, and it is apparent that reliance on quasi-static
tests for development of EA structures is nonconservative.

The chamfered half-tube segments exhibit high SEA but are not suitable themselves
as structural energy absorbers, due to their idealised shape. The segment crush tests
 are relatively easy to perform with high reproducibility, and the tests are mainly
used for screening new composite materials and lay-ups for use in crash energy
absorbers. They are a basic element of sine wave and corrugated webs that are used
in crashworthy subfloor structures as described by [27,36]. Current interest is now
to study EA behaviour for a range of composite channel shapes, which can then be
the basis of more advanced EA structures. A detailed study of carbon fabric/epoxy
profiles with corrugated webs [9] and several channel sections cut from box sections,
including the full box, C-channels and right angle corner sections [42] has been
undertaken in the CMH-17 Crashworthiness Working Group [8]. Quasi-static tests
on the channel sections fabricated from a Toray T700/2510 woven fabric prepreg
composite indicated SEA values ranging from 80 kJ/kg for the corrugated webs,
62 kJ/kg for a small right angle corner section, down to 37 kJ/kg for a box section
and large C-channel. This follows from the different failure modes that are shape
dependent. The corrugated web crushed down similar to the DLR segments and
had high SEA, whereas in the C-channel the unsupported sidewalls bend outward
and caused a global splaying failure mode with energy absorbed by tearing along
the channel corners.

To increase the level of complexity, a variety of integrated composite structures,

involving circular and trapezoidal segments bonded and/riveted to backing plates,
or used as core with sandwich skins, were investigated by CRC-ACS, as reported in
[29,43–45]. These test articles were designed to generate information on the effect of
integrating crush elements with the surrounding structure. Tests were conducted to
study both the difference between bonded and mechanically fastened components,
and the effects of different load introduction paths. This led to the development of an
EA web segment that is discussed further in Section 10.3.3 and is the basic element
of the concept subfloor structure described in Section 10.4.

> Read full chapter

ScienceDirect is Elsevier’s leading information solution for researchers.

Copyright © 2018 Elsevier B.V. or its licensors or contributors. ScienceDirect ® is a registered trademark of Elsevier B.V. Terms and conditions apply.