Page 1
Dissertations and Theses
8-2016
Detecting Delamination in Carbon Fiber Composites Using Detecting Delamination in Carbon Fiber Composites Using
Piezoresistive Nanocomposites Piezoresistive Nanocomposites
Sandeep Chava
Follow this and additional works at: https://commons.erau.edu/edt
Part of the Aerospace Engineering Commons, and the Materials Science and Engineering Commons
Scholarly Commons Citation Scholarly Commons Citation Chava, Sandeep, "Detecting Delamination in Carbon Fiber Composites Using Piezoresistive Nanocomposites" (2016). Dissertations and Theses. 204. https://commons.erau.edu/edt/204
This Thesis - Open Access is brought to you for free and open access by Scholarly Commons. It has been accepted for inclusion in Dissertations and Theses by an authorized administrator of Scholarly Commons. For more information, please contact [email protected] .
Page 2
DETECTING DELAMINATION IN CARBON FIBER COMPOSITES USING
PIEZORESISTIVE NANOCOMPOSITES
A Thesis
Submitted to the Faculty
of
Embry-Riddle Aeronautical University
by
Sandeep Chava
In Partial Fulfillment of the
Requirements for the Degree
of
Master of Science in Aerospace Engineering
August 2016
Embry-Riddle Aeronautical University
Daytona Beach, Florida
Page 4
iii
DEDICATION
This thesis is dedicated to my father,
Venkatrao Chava
May his memory forever be a
comfort and a blessing
He was the best father
a kid could have
Page 5
iv
ACKNOWLEDGMENTS
It gives me immense pleasure to acknowledge all the people that supported me and
stood beside me in this journey. I would first like to thank my thesis advisor Dr. Sirish
Namilae, who consistently allowed this thesis to be my own work, but steered me and
motivated me in the right direction whenever he thought I needed it.
I would like to thank Dr. Kim for all of his advice and inputs, Dr. Habib Eslami for
being a constant source of encouragement and enthusiasm, Mr. Tony Sharp and my IT team
for their moral support for the last two plus years, Dr. David J Sypeck and Mr. Michael
Potash for sharing their knowledge, Mr. William Russo for all his assistance throughout
my research and all faculty that guided me during various stages of my masters.
I would like to extend my appreciation to my research group Jiukun Li, Sandeep
Choudhary, Bhanu Praketh Kota, Audrey Gbaguidi and all my friends for their support and
encouragement throughout my thesis work.
I would especially like to thank my mother Jhansi, my grandparents Lakshmi
Kanthamma & Krishnamurthy and my sister Keerthi. My hard-working parents have
sacrificed their lives for my sister and myself and provided unconditional love and care. I
love them so much, and I would not have made it this far without them.
Finally, I would like to thank Aerospace Engineering Department of Embry-Riddle
Aeronautical University and its entire staff and faculty for educating me and supporting me
throughout my masters. I always feel I took a right step by coming here to do my Master
of Science in Aerospace Engineering.
Page 6
v
TABLE OF CONTENTS
LIST OF TABLES ............................................................................................................ vii
LIST OF FIGURES ........................................................................................................ viii
SYMBOLS ......................................................................................................................... xi
ABSTRACT ...................................................................................................................... xii
1. Introduction ..................................................................................................................... 1
1.1 Significance .............................................................................................................. 1
1.2 Motivation................................................................................................................. 3
1.3 Problem statement and research objectives ............................................................ 6
Literature review ............................................................................................................. 8
2.1 Composites ............................................................................................................... 8
2.2 Defects in Aerospace Composite Structures: ....................................................... 10
2.2.1 Manufacturing Defects in Composite Structures .......................................... 11
2.2.2 In Service Defects in Composite Structures .................................................. 12
2.2.3 Boeing 787 Dreamliner Delamination Issue ................................................. 14
2.3 Carbon as Graphene and Carbon Nanotubes (CNTs) .......................................... 15
2.4 Properties of Carbon Nanotubes (CNTs) ............................................................. 18
2.5 Buckypaper (CNT Sheet) ...................................................................................... 19
2.6 Finite Element Models of Delamination .............................................................. 22
2.6.1 eXtended Finite Element Method (XFEM) ................................................... 23
2.6.2 Cohesive Zone Model (CZM) ........................................................................ 23
2.6.3 Virtual Crack Closure Technique (VCCT) ................................................... 25
2.6.4 Abaqus Software ............................................................................................. 27
Experimental Procedure ................................................................................................ 30
3.1 Fabricating Nanocomposite Sensor ...................................................................... 30
3.1.1 Materials .......................................................................................................... 31
3.1.2 Procedure ......................................................................................................... 32
3.2 Fabricating Carbon Fiber Prepreg Composite laminate ...................................... 36
Page 7
vi
3.2.1 Materials .......................................................................................................... 37
3.2.2 Fabrication Procedure ..................................................................................... 38
3.3 Preparing Samples for Testing .............................................................................. 42
3.4 Electro-Mechanical Measurement & Data Acquisition....................................... 44
3.5 Experimental Setup of Flexural Test .................................................................... 46
Experimental Results of Mechanical and Electrical Behavior of Composites ............. 50
4.1 Mechanical Properties of CNTs and Carbon Fiber Composites ......................... 50
4.1.1 Mechanical Properties of CNTs .................................................................... 50
4.1.2 Mechanical Properties of Carbon Fiber Composites ................................... 51
4.2 Experimental Results of Mechanical Properties .................................................. 52
4.3 Electrical Properties of CNTs and Carbon Fiber Composites............................. 55
4.4 Scanning Electron Microscopy ............................................................................. 58
Finite Element Model of Flexural Test ......................................................................... 60
5.1 Finite Element Model of Non-delaminated Sample ............................................ 60
5.2 Finite Element Model of Delaminated Sample using VCCT ............................. 62
5.3 Results of Finite Element Model (Abaqus/CAE) ................................................ 64
Analysis and Discussion ............................................................................................... 66
6.1 Electro-Mechanical Properties of Piezoresistive Nanocomposite ...................... 66
6.2 Comparing Results of Experiments and Simulation ............................................ 69
6.3 Strain in Composite Region (on laminate) in Abaqus/CAE ............................... 71
6.4 Strain Correlation of Piezoresistive Nanocomposite ........................................... 74
6.5 Conclusion .............................................................................................................. 76
Summary and Recommendations ................................................................................. 78
7.1 Summary ................................................................................................................. 78
7.2 Recommendations for Future Work ..................................................................... 79
REFERENCES ................................................................................................................. 80
Page 8
vii
LIST OF TABLES
Table 1.1 Estimated global carbon fiber consumption (Roberts, T 2010)…………...........3
Table 1.2 CNT and Graphene based sensors……………………………………………...5
Table 4.1 Young's Modulus and Tensile Strength of buckypaper/polymer nanocomposites
……………………………………………………………………………………………50
Table 4.2 Mech. properties of CF laminates cured for 90min at 275 oF by Autoclave (Hankuk
carbon co. LTD, CF3327-1 EPC: SE-019K)…………………………………………………51
Page 9
viii
LIST OF FIGURES
Figure 2.1 Relative importance of material development through history (Leny, 2009)…9
Figure 2.2 Strength to weight ratio of composites (©Cambridge Univ Eng. Dept,
2009)……………………………………………………………………………………..10
Figure 2.3 Showing damage in a 4-ply laminated plate………………………………….13
Figure 2.4 Flightglobal news on 5th Feb, 2012…………………………………………..15
Figure 2.5 Carbon allotropes (i) diamond; (ii) graphite; (iii) lonsdaleite; (iv) C60
(Buckminsterfullerene); (v) C540; (vi) C70; (vii) amorphous carbon; (viii) single-walled
carbon nanotube (Wikimedia. Created by Michael Ströck (mstroeck). CC BY-SA 3.0)..16
Figure 2.6 SWNT and MWNT…………………………………………………………..17
Figure 2.7 Buckypaper used in this research from Nano Tech Labs…………………….20
Figure 2.8 SEM micrograph of buckypaper……………………………………………..21
Figure 2.9 Schematic of cohesive zone model (CZM) (Scheider, I, 2006)……………...24
Figure 2.10 Abaqus/CAE main user interface…………………………………………...26
Figure 2.11 Model Tree with all modules and objects…………………………………...28
Figure 3.1 Buckypaper strip……………………………………………………………...31
Figure 3.2 BP strip with copper plates…………………………………………...………31
Figure 3.3 BP strips with copper plates on tooling plate……………………...…………32
Figure 3.4 Epoxy resin and Epoxy resin mixture………………………...……………...32
Figure 3.5 General vacuum bagging schematic……………………………………….....33
Figure 3.6 Showing BP strip, attached copper plate and final sensor……………………34
Figure 3.7 Peeled off nanocomposite sensors……………………………………………34
Figure 3.8 SEM micrograph of coarse graphene platelets and CNTs on the sensor…….35
Figure 3.9 Typical autoclave cure cycle (Hankuk Carbon)...……………………………37
Figure 3.10 Carbon fiber prepreg strip………...……………………………………..….37
Figure 3.11 Prepreg layup using a roller to avoid air gaps…...………………………….38
Page 10
ix
Figure 3.12 Wabash Press used in curing prepreg……………………………………….39
Figure 3.13 Digital temperature controls on Wabash press……………………………...39
Figure 3.14 Delamination in composite laminate sample (schematic)……………………40
Figure 3.15 Laminate samples and nanocomposite sensors side by side………………...41
Figure 3.16 Vacuum bagging setup for attaching nanocomposite sensor………………...42
Figure 3.17 Sample soldered with copper wires…………………………………………42
Figure 3.18 Schematic of voltage drop setup…………………………………………….43
Figure 3.19 LabVIEW code developed for data acquisition……………………………..44
Figure 3.20 ASTM D7264: Three-point loading diagram………………………………..45
Figure 3.21 Three-point setup for MTS testing system…………………………………..46
Figure 3.22 MTS Testing system…………………………………………………………46
Figure 3.23 Three-point setup for flexural test (ASTM D7264)…………………………47
Figure 3.24 Sample, under load and after deformation…………………………………...48
Figure 4.1 Stress - Strain plot of nanocomposites with coarse graphene platelets (5
wt.%)……………………………………………………………………………………..51
Figure 4.2 Stress - Strain plot of composite sample in Tension…………………………...52
Figure 4.3 Stress - Strain plot of non-delaminated composite sample (ASTM D7264)….53
Figure 4.4 Stress - Strain plot of delaminated composite sample (ASTM D7264)………53
Figure 4.5 Change in Resistivity - Strain plot of nanocomposites with coarse graphene
platelets (5 wt. %)……………………………………..…………………………………55
Figure 4.6 Change in Resistivity - Strain plot of non-delaminated composite sample…..56
Figure 4.7 Change in Resistivity - Strain plot of delaminated composite sample…….…56
Figure 4.8 SEM micrograph of fracture specimen……………………………..…………57
Figure 4.9 SEM micrograph showing CNTs on fracture specimen……………….……..58
Figure 5.1 Composite 8-ply layup in Abaqus/CAE……………………………….……..59
Figure 5.2 Assembly of three-point bend setup as experiments…………………………60
Page 11
x
Figure 5.3 Strain contour of deformed laminate sample………………………………….61
Figure 5.4 Strain contour of deformed delaminated sample……………………………...62
Figure 6.1 Stress – Strain and Resistivity – Strain response of a nanocomposite………..63
Figure 6.2 Stress – Strain and Resistivity – Strain response of nanocomposite attached to a
non-delaminated composite laminate…………………………………………………….64
Figure 6.3 Stress – Strain and Resistivity – Strain response of nanocomposite attached to a
delaminated composite laminate…………………………………………………………65
Figure 6.4 Force vs Displacement of non-delaminated laminate sample…………………66
Figure 6.5 Force vs Displacement of delaminated laminate sample from Abaqus/CAE…67
Figure 6.6 Comparing results of Simulation and Experiment in non-delaminated
sample…………………………………………………………………………………....68
Figure 6.7 Comparing results of Simulation and Experiment in a delaminated sample…..68
Figure 6.8 Cut out region of the nanocomposite section from assembly………………...69
Figure 6.9 Cut out region of the nanocomposite section from mesh….…………………69
Figure 6.10 Cut out region of the nanocomposite section from Abaqus visualization……70
Figure 6.11 Nanocomposite region and elements selected for strain calculation………..71
Figure 6.12 Strain vs Displacement plot of three elements in non-delaminated model….71
Figure 6.13 Strain vs Displacement plot of three elements in a delaminated model……..72
Figure 6.14 Correlation between the strain and resistivity of nanocomposite from three-
point bend test and tension test of a non-delaminated model……………………………73
Figure 6.15 Correlation between the strain and resistivity of nanocomposite from three-
point bend test and tension test of a delaminated model…………………………………..73
Figure 6.16 Stress – Strain and Resistivity – Strain response from flexural tests………..74
Figure 7.1 Composite repair patch with nanocomposite sensor for detecting the
effectiveness of the repair…………..……………………………………………………77
Page 12
xi
SYMBOLS
R resistance
I current
∆V voltage drop
ρ resistivity
D maximum deflection
L length of support span
b width of test beam
F load at any given point
d depth of tested beam
w width of nanocomposites
m slope of the secant of the force-deflection curve
t thickness of nanocomposites
f maximum flexural stress
f maximum strain
fE flexural modulus of elasticity
l length of nanocomposites
V applied voltage
R⁄Ro resistivity change
ɛ strain
Page 13
xii
ABSTRACT
Researcher: Sandeep Chava
Title: Detecting Delamination in Carbon Fiber Composites using Piezoresistive
Nanocomposites
Institution: Embry-Riddle Aeronautical University
Degree: Master of Science in Aerospace Engineering
Year: 2016
Carbon fiber prepreg composites are utilized successfully as structural materials for
different lightweight aerospace applications. Delamination is a critical failure mode in
these composite materials. As composite plies separate from each other, the composite
loses some of its ability for supporting expected loads. Therefore, detection of delamination
at right time is of foremost significance. This study presents a new way for detecting
delamination in composite plates using piezoresistive nanocomposites. This new procedure
is setup and studied through both experimental and computational investigations. In this
research, nanocomposites with 5% coarse graphene platelets are fabricated for detecting
delamination. 8-ply carbon fiber prepreg composite samples are fabricated through
compression molding. Delaminated composite samples are fabricated by placing a Teflon
film between layers of prepreg. Piezoresistive nanocomposites are attached on top of
prepreg laminate samples using epoxy resin. The change in electrical resistivity of these
nanocomposites due to the induced strain from flexural test (three point bend test) on
delaminated and neat composite laminates are monitored to demonstrate the delamination
detection method. A non-linear finite element model is developed using Abaqus software
suite to compliment the mechanical testing. Virtual Crack Closure Technique (VCCT) is
Page 14
xiii
used to model a delamination in the composite sample. Experimental results and the
simulations in this study indicate that piezoresistive nanocomposites can be used for
detecting delamination in carbon fiber composite materials.
Page 15
1
Introduction
1.1 Significance
Composite materials have been widely used in for structural applications. Light
weight, high specific strength, resistance to corrosion and flexibility in design, etc. are
some of the properties displayed by these materials that have benefited many industries
such as aerospace, automotive and marine. For example, commercial aircraft such as
Boeing 787 and Airbus 380 are two of the first commercial aircraft to feature composites
in fuselage and other primary structures (Garnier, Pastor, Eyma, & Lorrain, 2011).
Despite these benefits the susceptibility of composite materials to impact damage
is high and creates a major concern related to structural integrity (Abrate, 2005). In
aerospace structures, low-velocity impacts are often caused by tool drops during
manufacturing and servicing or runway stones during landing or take off. Such impacts
may result in various forms of damage such as indentation, matrix cracking, delamination
or fiber fracture, leading to severe reduction in strength and also reduction in integrity of
composite structures.
Although structures designed with fail-safe principles can withstand in theory,
impact damage detection is an important issue in maintenance of aircraft and aerospace
structures. While visible damage can be easily detected and remedial action can be taken
to maintain structural integrity, a major concern to end-users is the growth of undetected,
hidden damage caused by low-velocity impacts and fatigue. This internal damage is also
known in aerospace applications as Barely Visible Impact Damage (BVID), and failure to
detect BVIDs may result in a catastrophic collapse of the structure (Aymerich &
Staszewski, 2010).
Page 16
2
Hence, there is an increasing need for damage detection technologies on primary
structural components of the aircraft (such as the fuselage or the wings) so that we can
repair the damage rather than replacement of these components as a first-time solution (Hu
& Soutis, 2000).
Common damage on an aircraft can arise from accidental impact, bird strike,
hailstones and lightning strike or from deterioration caused by the absorption of moisture
or hydraulic fluid (Cheng, Gong, Hearn, & Aivazzadeh, 2011). Because of the laminated
layers in composite structures, damage often manifests as delamination between plies.
Delamination is one of the major failure modes and it may cause structural failure leading
to catastrophic consequences.
Development of an early damage detection method for delamination is an important
requirement for maintaining the integrity and safety of composite structures. Many
detection techniques have been proposed for structural health monitoring (SHM) and some
of the non-destructive evaluation approaches that utilize advanced technologies, such as
X-ray imaging (Tillack, Nockemann, & Bellon, 2000), ultrasonic scans (Rose, 2007),
infrared thermograph (Meola & Carlomagno, 2004), and eddy current (Grimberg, Savin,
& Rotundu, 2001), can identify damages.
However, most of these approaches are difficult to implement for in-service aircraft
testing and in situ space structures. Almost all of the above techniques require that the
vicinity of the damage is known in advance and the portion of the structure being inspected
is readily accessible (Qiao, Lu, Lestari, & Wang, 2006). This thesis investigates
delamination in laminates composite structures with the intention of developing a method
for damage detection approach using piezoresistive nanocomposites. The nanocomposites
Page 17
3
can be embedded in the structure facilitating potential in situ damage detection and
monitoring
The piezoresistive nanocomposites used in this work are developed based on
Carbon Nanotube sheets and graphene platelets. This thesis attempts to understand the
change in electrical behavior of the piezoresistive nanocomposites under applied strain
when they are embedded in composite laminates with and without delamination.
Additional applications are in the area of composite repair, wherein these piezoresistive
nanocomposite sensors have the potential to study the effectiveness of the repair on
composite laminates.
1.2 Motivation
Carbon fiber is generally defined as a fiber that contains at least 92 wt % carbon.
On the other hand, the fiber with 99 wt % carbon is called a graphite fiber (Fitzer, 1989).
Due to their excellent tensile properties and low densities, these are used in composites in
the form of woven textiles, prepregs, and chopped fibers. The demand for the carbon fiber
is increasing due to its use in many industries, such as aerospace, military and transport.
The estimated prediction of global carbon fiber consumption is shown in Table 1 as of
2010 (Roberts, 2006).
Table 1.1. Estimated global carbon fiber consumption (Roberts, 2006)
1999 (tons) 2004 (tons) 2006 (tons) 2008 (tons) 2010 (tons)
Aerospace 4,000 5,600 6,500 7,500 9,800
Page 18
4
Industrial 8,100 11,400 12,800 15,600 17,500
Sporting Goods 4,500 4,900 5,900 5,700 6,900
Total 16,600 21,900 25,200 29,800 34,200
These carbon fiber composites offer a combination of strength and modulus that
are either comparable to or better than many traditional metallic materials (Harris, 1991).
In addition to these properties, there are number of other differences between metallic
structures and carbon fiber composites. Due to inherent heterogeneity and anisotropy,
failure of composites involves many mechanisms. Whereas fatigue failure of isotropic and
homogeneous materials such as metals is the result of initiation and propagation of a single
dominant crack, fatigue failure of composites is characterized by initiation and
multiplication of many cracks in the weak phase (Shokrieh & Taheri-Behrooz, 2009).
For example, metals exhibit plastic deformation while most fiber reinforced
composites are elastic in their tensile stress-strain characteristics. Mechanisms of damage
development and growth in metalic and composite structures are also quite different. While
carbon fiber composites are used extensively in these many industries, there is always a
need for early damage detection in composite structures. Piezoresistive nanocomposite
sensor can potentially monitor the delamination in composite structures.
There are many carbon nanotube (CNT) based sensors that are currently used in
fields like biomedical, automotive and food industry etc. (see Table1.2). Sensors in
Biomedical Field like CNT Implantable Nanosensor is used in detecting diseases or
hazardous radiation exposure in early stages (Sinha & Yeow, 2005). In automotive
Page 19
5
industry, sensors like antitheft sensor that prevents stealing of vehicles are also made based
on CNTs (Weinberg, 2002). In food industry, sensors like pH sensors are used to determine
pH value in water for survival and growth of fishes (Xu, Chen, Qu, Jia, & Dong, 2004).
Sensors like CNT-based acoustic and optical sensors used for breath alcohol detection at
room temperature (Penza, Cassano, Aversa, Antolini, Cusano, Cutolo, & Nicolais, 2004).
Table 1.2 CNT and Graphene based sensors
Sensor area Sensor type Sensor uses Ref
Biomedical
CNT Nano biosensor Detecting DNA sequences in body R1
CNT Chemical sensor Blood analysis (Detecting ‘Na’ etc.) R2
CNT Pressure sensor Eye surgery, respiratory devices etc. R3
Automotive
CNT Force sensor Determining if an air filter is bad R2
CNT Pressure sensor Controlling dampers in suspensions R4
CNT Rollover sensor Finding roll rate to prevent tipping over R5
Food
CNT Gas sensor Monitor meat freshness during shipping R6
CNT Humidity sensor Monitoring humidity changes R7
CNT CO2 sensor Monitoring CO2 for plants growth R8
R1: Gao, 2010, R2: Kauffman, 2010, R3: So, 2013, R4: Gau, 2009, R5: Yang, 2013, R6:
Yoon, 2011, R7: Cao, 2011, R8: Shivananju, B. N, 2013.
Page 20
6
1.3 Problem statement and research objectives
The primary motivation is to develop a new damage detection sensor and detect
delamination in carbon fiber composites using the developed sensor. In this thesis we
explore and study CNT based piezoresistive nanocomposites and their application in
detecting delamination through both experiments and finite element model. Experiments
of tensile testing and flexural deformation of carbon fiber composites with and without
delamination are used to characterize the embedded nanocomposite sensors. A finite
element model is developed to further analyze the results of the experiments. The specific
research objectives are identified as follows:
a) Design and fabricate nanocomposites using Buckypaper and 5 wt% coarse
graphene platelets mixed with epoxy matrix. This composition has been
suggested to exhibit highest peizoresistivity through prior work in our group
[Li & Namilae, 2016].
b) Design and fabricate carbon fiber composite laminate samples with 8 ply layup
using carbon fiber prepreg.
c) Tension test using MTS testing machine and flexural deformation test using the
3 point bend setup following ASTM standards on the carbon fiber prepreg
laminate sample attached with piezoresistive nanocomposite on top of them.
The results of the tests are stress-strain and resistivity-strain curves obtained
through LabVIEW code developed.
d) Flexural test on a delaminated sample following ASTM standard and collecting
similar results using LabVIEW code developed.
e) Developing Finite Element Model (FEM) using Abaqus CAE and doing the
Page 21
7
same flexural test without delamination and with delamination using VCCT in
Abaqus CAE.
f) Analyzing the results from experiments and finite element model and
understanding the difference in test results between delaminated and baseline
specimen.
g) Suggesting how the piezoresistive nanocomposite can be used to detect
delamination or to identify the effectiveness of a repair in a repair patch on
carbon fiber composite laminates.
By determining the change in electrical resistivity of a piezoresistive
nanocomposite on a delaminated and non-delaminated sample, this thesis demonstrate
the potential of nanocomposite utilization in damage/delamination sensing in aircraft
structures.
Page 22
8
Literature review
2.1 Composites
Composites are among the most adaptable advanced engineering materials known
till date. Composites in general are heterogeneous materials and consist of two or more
constituents such as fiber and a matrix (Lubin, 1982). In contrast to metallic alloys, each
constituent material retains its separate chemical, physical, and mechanical properties.
The matrix of composites can be metal (Metal Matrix Composite), ceramic
(Ceramic Matrix Composite) or carbon based (Carbon-Carbon or polymeric). These
different matrix give composites their shape, appearance, environmental tolerance and
durability while the fibers carry most of the structural load and thus, making these materials
strong and stiff (Mark, Bikales, Overberger, Menges, & Kroschwitz, 1985)
Ashby (Ashby, Bush, Swindells, Bullough, Ellison, Lindblom, & Barnes, 1987)
presents a chronological variation of the relative importance of each group of materials
from 10,000 B.C. and extrapolates their importance through the year 2020. The information
contained in Ashby’s article has been partially reproduced in Figure 2.1. From the figure it
can be noticed that the importance of composites increased steadily from 1960 and is
projected that it will increase through the next several decades as composites replace
metallic materials in many applications.
Page 23
9
Figure 2.1 Relative importance of material development through history (Leny, 2009)
One of the biggest benefits of composites is that the properties are adjustable per
design parameters, for example, the mecchancial properties, content, orientation and fiber
architecture, and the properties of the matrix are all materails design parameters.
Composite materials play a key role in industries like aerospace and automobile because
of their outstanding strength to weight ratio and modulus to weight ratio (Figure 2.2). Some
of these composites like graphite, Kevlar, boron or silicon carbide fibers in polymeric
matrices have been studied extensively because of their applications in aerospace and space
vehicle technology (Nielsen, 1972; Woods, 1994; Sohn, 2001; Rajeev, 2003)
Page 24
10
Figure 2.2 Strength to weight ratio of composites (©Cambridge University Engineering
Department, 2009)
2.2 Defects in Aerospace Composite Structures:
Defects in carbon fiber composite structures are produced either during the
manufacturing process or in the course of normal service life of the structure. There are
multiple types of defects that are caused during manufacturing process. Porosity is one of
the important defect in manufacturing process which is the presence of small voids in the
Page 25
11
matrix (Kastner, Plank, Salaberger, & Sekelja, 2010). This can be caused by incorrect cure
parameters such as pressure, temperature or duration of cure. Sandwich structures with
honeycomb cores can suffer due to poor bonding of skin to the core. Dis-bonding can occur
at these skin-to-adhesive or adhesive-to-core interactions (Fruehmann, Wang, Dulieu-
Barton, & Quinn, 2011).
Service defects are mostly due to impacts. These impacts lead to matrix cracking
and delamination of the ply layers. Delamination is a critical failure mode in these
composite materials. In some cases the damage can be only internal referred as barely
visible impact damage (BVID) (Garnier, Pastor, Eyma, & Lorrain, 2011). Sandwich
structures can suffer from same matrix cracking and delamination in the skin during an
impact.
2.2.1 Manufacturing Defects in Composite Structures
There are multiple methods in manufacturing composite structures. All of these
methods aim to combine the fiber and matrix into one product. They can be separate before
manufacturing or can already be combined like prepreg materials. All of these different
methods selected to manufacture composites depend on the size and quality of the products
required. Higher quality structures are usually used in aerospace applications to minimize
weight, which are manufactures using hot pressing method or autoclaving.
During all these manufacturing processes, defects can be introduced into the
structure and their effects depend on the process used to manufacture them and their
applications. Multiple defect types have been identified including the following (Smith,
2009):
i. Porosity: Voids are created due to improper curing.
Page 26
12
ii. Foreign bodies: Knowingly/unknowingly adding foreign bodies.
iii. Fiber vol. fraction: Incorrect fiber volume fraction due to excess or
insufficient resin.
iv. Bonding: Bonding defects during manufacturing.
v. Fiber misalignment: This causes local changes in volume faction by
preventing ideal packing of fibers.
vi. Ply misalignment: Mistakes made in layup of plies cause this defect.
This alters overall stiffness and strength of the laminate.
vii. Incomplete cure: Incompletely cured matrix during curing cycle.
viii. Wavy fibers: These are produced by in-plane kinking of the fibers
and can seriously affect laminate strength.
ix. Fiber defects: The presence of defects in fibers themselves is one
of the limiting factor in determining strength, these defects are considered
as one of the basic material properties.
2.2.2 In Service Defects in Composite Structures
Composite structures can degrade over time in service due to many mechanisms
and most of them are due to environment experienced defects. Some of these mechanisms
are static overload, impact, fatigue, overheating and lightning strike etc. The main detects
that are found in service are as following (Smith, 2009):
a. Delamination
b. Bond failures
c. Cracks
Page 27
13
d. Moisture entry
e. Fracture or buckling of fibers
f. Failure of interface between the fibers and matrix
All these mechanisms vary depending on the type of loading, and mechanical
properties of the constituents. Failure mechanisms are same in most composites but their
mode of occurrence vary depending on the type of loading and properties of the
constituents (Atiqullah, 2011). The major in service defect requiring detection in the
presence of delamination. Delamination can be produced by fatigue, bearing damage,
impact, etc (Xiong, 2010). Dis-bonding can also be found due to impacts. Cracks are the
ones that usually lead to delamination in a critical stage.
Moisture degrades the strength properties of composites that are matrix dependent
and also reduces residual strain (Smith, 2009). It may be possible to measure moisture
content nondestructively. Fracture of the fibers and failure of interface between the fibers
and matrix are due to impact (Kachanov, 2012). All these differ depending on application
and composite type.
Figure 2.3 Showing damage in a 4-ply laminated plate (Iowa State Univ.)
Page 28
14
Figure 2.3 is a photograph, reproduced from Iowa State University’s research on
nondestructive evaluation showing damage in a 4-ply laminated plate that was subjected
to thermal shock. One can see two types of cracks: cracking of the matrix material within
a ply, and separation cracks (delamination) at the boundary between plies. The carbon
fibers give the material the bulk of its strength, but rarely large cracks can be developed
without breaking many fibers.
2.2.3 Boeing 787 Dreamliner Delamination Issue
As an example of these type of manufacturing defect one can refer to Boeing
Dreamliner 787 delamination issue caused in 2012. According to Flightglobal, the
structural stiffeners were found to be improperly joined to the composite skin in the aft
sections of the aircraft, causing parts of the aircraft's carbon fiber to delaminate.
Boeing has found that incorrect shimming was performed on support structure on
the aft fuselage on certain airplanes in their facility in Everett. Flightglobal has confirmed
there are at least three affected airframes, Airplanes 56, for All Nippon Airways, where the
problem was first discovered, and Airplanes 57 and 58, were the first two aircraft for Qatar
Airways.
News articles (reference) mention that the stiffeners, or longerons that run along
the length of the aircraft, delaminate around the rear opening of the Section 48 section
above and below the cutout known as the "bird's mouth" that holds the Alenia Aeronautica-
built horizontal stabilizer.
When the longerons are installed on the wound carbon fiber barrel, frames and
longerons are secured to the skin of the structure to give it strength. When natural variations
in the fit of parts exists, aerospace mechanics will install shims, or spacers, which
Page 29
15
compensate for variations and wedge into structure to create a tighter fit. Without the
shims, damage can be sustained to the composite when fasteners are installed by pulling
the structure together, which cause damaging of the layers of carbon fiber. Over the long-
term composite delamination can decrease the fatigue life of the aircraft's structure.
Figure 2.4 Flightglobal news on 5th Feb, 2012
Boeing has faced manufacturing quality issues before, most notably in the June
2010 inspection, teardown and reinstallation of many Alenia Aeronautica-built horizontal
stabilizers were assembled without proper shimming, creating gaps in the structure that
threatened the fatigue life of the empennage, according to Flightglobal.
2.3 Carbon as Graphene and Carbon Nanotubes (CNTs)
We now discuss the properties of carbon nanotubes, graphene and nanocomposites
that are used in this work for damage detection application. Carbon is one of the most
studied elements in the periodic table. The versatility of chemical bonds enables many
Page 30
16
carbon allotropes. In three-dimensional form, carbon can exist as graphite and diamond,
which comprise of sp2 and sp3 covalent bonds, respectively. In the 1980s and 1990s,
another two types of carbon allotropes, the zero-dimensional fullerene (Kroto., 1991) and
one-dimensional carbon nanotubes, were discovered.
Figure 2.5 Carbon allotropes (i) diamond; (ii) graphite; (iii) lonsdaleite; (iv) C60
(Buckminsterfullerene); (v) C540; (vi) C70; (vii) amorphous carbon; (viii) single-walled
Page 31
17
carbon nanotube (Wikimedia. Created by Michael Ströck (mstroeck). CC BY-SA 3.0).
Ever since their discovery, their contribution to development of studies in the field
of physics, chemistry and material sciences is huge. There are many research studies on
the structure (Dresselhaus, Dresselhaus, & Saito, 1995), properties (Mintmire & White,
1995) and their various applications (Ajayan, 1997). Graphene is composed of a
honeycomb lattice of carbon atoms. Structurally, graphene is related to many carbon
allotropes (Figure 2.5). For example, carbon nanotubes can be formed by rolling graphene
along certain axes, and graphite can be formed by stacking graphene vertically (Geim &
Novoselov, 2007).
Carbon nanotubes are basically rolled up graphene sheets (hexagonal structures)
into cylindrical form and capped with half shape of fullerene structure. Many of the
properties of CNTs are due to the way the graphene sheets are wrapped around. There are
two types of carbon nanotubes:
(a) Single walled carbon nanotubes (SWNTs), formed by rolling a single graphene
sheet into a cylinder (Figure 2.6). SWNTs with their high length to diameter
ratio, atomic strength and chemical stability constitute one-dimensional
molecules (Gommans, Alldredge, Tashiro, Park, Magnuson, & Rinzler, 2000)
Figure 2.6 SWNT and MWNT
Page 32
18
(b) Multi walled carbon nanotubes (MWNTs), can be considered as stacking of
several layers of graphene in the form of cylinders with an interspacing of
around 0.36nm (Figure 2.6). The length and diameter of MWNTs differ a lot
from SWNTs and, of course, their properties are also very different.
Carbon nanotubes have unique electronic and mechanical properties which are
achieved by chemically processing these CNTs in order to purify and get required property
in them. CNTs can be metallic or semiconducting depending upon the atomic
arrangements. A large number of these nanocomposites are produced in various methods
such as arc evaporation method, electrolysis, laser ablation, chemical vapor deposition, etc.
(Ying, Salleh, Yusoff, Rashid, & Razak, 2013). Production of carbon nanotubes in a
controlled way in large quantities could potentially encounter problems that remain to be
solved.
2.4 Properties of Carbon nanotubes (CNTs)
Carbon nanotubes have very unique thermal, electrical and mechanical properties.
In spite of no direct methods to prove their properties, several experimental tests like SEM,
AFM, TEM, nanoindentation, etc. and theoretical methods like molecular dynamics,
continuum model, etc. are used to describe the mechanical properties of carbon nanotubes
(Coleman, Blau, Dalton, Munoz, Collins, Kim, ... & Baughman, 2006).
Experimental studies of Georgakilas et al., found that high stiffness, high modulus
and low density carbon nanotubes can be ideal material for fabrication of different
composites (Georgakilas, Perman, Tucek, & Zboril, 2015). Yu et al. showed that only outer
layer in a MWNT was able to withstand higher loadings while inner layers were observed
to be very weak (Yu, 2000). These ultimate measurements were carried out and managed
Page 33
19
to perform stress-strain measurements on individual arc-MWNTs inside an electron
microscope and for a range of tubes the modulus ranges from 0.27 to 0.95 TPa. Fracture of
MWNTs occurred at strains up to 12% and width strengths in the range 11-63 GPa (Prato
& Maurizio, 2009).
Along with these mechanical properties, electronic properties of carbon nanotubes
are also studied. Theoretical studies conclude that depending upon the diameter and
chirality of the tube, carbon nanotubes may be either metallic or semiconducting (Terrones,
2013). In several experiments using scanning tunneling microscope (STM), it is observed
that the tunneling conductance is a direct measure of local electron density of states of
carbon nanotubes (Mittal & Garima, 2015).
The electrical resistivity of metallic carbon nanotubes was observed to be around
10-8 to 10-7 ohm-m (Charlier & Issi, 1996) and the electrical conductivity of individual
MWNT is measured by four probe measurements using lithographic deposition of tungsten
to be in the range of 107 to 108 S/m (Ebbesen, Lezec, Hiura, Bennett, Ghaemi, & Thio,
1996). Along with these properties, carbon nanotubes show high thermal conductivity also.
New studies show that ultra-small SWNTs have shown superconductivity below 20oK and
the high value of 6000 W/mK was shown by isolated nanotubes, which is comparable to
graphene monolayer and diamond (Berber, Kwon, & Tománek, 2000). The small diameter
and high aspect ratio of CNTs is favorable for field emission, which results from the
tunneling of electrons from metal tip into vacuum under application of strong electric field.
2.5 Buckypaper (CNT Sheet)
Buckypaper (BP) is an outstanding material which contains entangled networks of
CNTs formed by Van der Waals interactions (Baughman, 1999), which is an effective way
Page 34
20
of introducing carbon nanotubes into composites (Endo, Muramatsu, Hayashi, Kim,
Terrones, & Dresselhaus, 2005). Buckypaper can be produced in large sizes and provides
ease of handling, and improves the safety of using CNTs in industrial manufacturing
facilities. Buckypaper can be fabricated by using double-walled CNTs (Gong, 2007),
SWCNTs (Teague, 2007), and MWCNTs (Xu, 2008).
Figure 2.7 Buckypaper used in this research from Nano Tech Labs
Due to its high CNT concentration, buckypaper provides great advantages to
enhance electrical properties (Cheng, 2010), actuation (Chen, 2010), fire retardancy (Wu,
2011), and electromagnetic interference shielding properties in composites (Gnidakouong,
Kim, Park, Park, Jeong, Jung, & Park, 2013).
A number of techniques for fabricating buckypaper have been proposed. Some of
which are:
Page 35
21
i. A vacuum filtration method for fabricating large-area buckypaper less than 200
nm thick (Hennrich, Lebedkin, Malik, Tracy, Barczewski, Rösner, & Kappes,
2002).
ii. Fabrication of buckypaper by liberation of electrophoretically deposited carbon
nanotubes (Rigueur, Hasan, Mahajan, & Dickerson, 2010).
iii. Fabrication of highly oriented buckypaper made of aligned carbon nanotubes
(Zhang, Jiang, & Peng, 2014).
Figure 2.8 SEM micrograph of buckypaper
Due to the component material (CNTs), microstructure and properties of
buckypaper in the past decade, buckypaper and buckypaper composites have been
extensively studied. It is believed to be an excellent material for many engineering
applications, such as electrodes, actuators, sensor, and heat conductors and as
reinforcement for polymer composites (Chen, 2013). Some of the studies which
Page 36
22
demonstrated sensors made of CNTs are:
a. Li et al. demonstrated the potential of carbon nanotube films in measuring strain at
the macro scale (Li & Dharap, 2004).
b. Kang et al. developed a composite electrical resistance strain sensor based on
SWNTs, and it was used to measure the strain of a structure at the macro scale
(Kang, Schulz, Kim, Shanov, & Shi, 2006).
c. Li et al. have studied the possibility of using multiwall carbon nanotube
(MWCNTs) films as strain sensors (Li, Levy, & Elaadil, 2008).
d. Gao et al. reported a simple approach to deposit multi-walled carbon nanotube
(MWNTs) networks onto glass fiber surfaces achieving semi conductive MWNTs-
glass fibers, along with application of fiber/polymer interphase as in situ
multifunctional sensors (Gao, Zhuang, Zhang, Liu, & Mäder, 2010).
These features make buckypaper an excellent candidate for manufacturing large-
scale composite samples with carbon fiber prepreg to achieve high CNT loading. This
thesis also develops a finite element model to explain the experimental observations using
Abaqus/CAE. This next section gives an introduction to finite element models for
delamination and fracture.
2.6 Finite element models of Delamination
Delamination brings significant material degradations in both stiffness and strength
under compression, tension and flexural loading. It occurs under any combinations of
mixed Mode I, Mode II, and Mode III. Different methods have been used to model and
simulate these delamination. Some of the widely used methods are Extended Finite
Page 37
23
Element Method (XFEM), Cohesive Zone Model (CZM) and Virtual Crack Closure
Technique (VCCT).
2.6.1 eXtended Finite Element Method (XFEM)
The extended finite element method (XFEM), also known as generalized finite
element method (GFEM) or partition of unity method (PUM) has been used very
successfully to model cracks because the finite element mesh can be created independent
from the crack geometry, and in particular the domain does not have to be remeshed as the
crack propagates (Richardson, Hegemann, Sifakis, Hellrung, & Teran, 2009). Richardson
et al. used XFEM method for modelling geometrically elaborate crack propagation in
brittle materials. This method was developed to reduce difficulties in solving problems
with localized features that are not efficiently resolved by mesh refinement.
One of the initial applications was the modeling of fractures in a material. A key
advantage of XFEM is that the finite element mesh does not need to be updated to track
the crack path (Jiang, 2013). In recent years, the extended finite element method (XFEM)
has emerged as a powerful numerical procedure for the analysis of fracture problems. It
has been widely acknowledged that the method eases fracture growth modeling under the
assumptions of linear elastic fracture mechanics (LEFM).
Since the introduction of the method in 1999, many new extensions and
applications have appeared in the scientific literature (Karihaloo & Xiao, 2003). XFEM
has been used in the study of composite delaminations (Motamedi, 2014; Motamedi, 2013;
Hulton, 2015; Sosa, 2012).
2.6.2 Cohesive Zone Model (CZM)
Page 38
24
Cohesive zone model (CZM) is one of the most versatile evolutions in the area of
fracture mechanics. These partition of the surfaces involved in the cracks takes place across
an extended crack tip, or cohesive zone, and is resisted by cohesive tractions (Liu, 2013).
The concept of cohesive zone ahead of the crack tip, which was introduced by Dugdale
(Dugdale, 1960) and Barenblatt (Barenblatt, 1962) has become a guiding idea for a class
of crack propagation models. Figure 2.9 shows the schematic of cohesive zone model for
various failure phenomena: damage is localized in an interface.
Figure 2.9 Schematic of cohesive zone model (CZM) (Scheider, 2006)
A cohesive model in combination with finite elements was first used for concrete
(Hillerborg, Modéer, & Petersson, 1976) and, more than ten years later, also for metals
(Needleman, 1990). Interface elements obeying a cohesive law are introduced between the
continuum elements. New applications cover a variety of phenomena like viscoplastic
(Corigliano, 2001) and viscoelastic (Rahu, 1999) separation behavior, the modelling of
Page 39
25
fragmentation (Repetto, Radovitzky, & Ortiz, 2000) and fiber de-bonding, failure under
dynamic and cyclic loading (Roe & Siegmund, 2003).
The respective cohesive law has to be chosen in dependence on the
micromechanical damage mechanism leading to fracture. Commonly, two material
parameters, namely a cohesive strength, and a critical separation, are chosen to characterize
the cohesive behavior. Finite element simulations with cohesive elements run numerically
stable up to large amounts of crack extension and yield very good results for structures
with different size and constraint conditions.
CZM can be applied to both 2D (Cornec, 2003) and 3D (Gao, 2006) structures.
Some studies of CZM to composite delamination include Milad saeedifar’s delamination
growth prediction (Saeedifar, 2015), Qiang Ye’s cohesive strength predication for
composite delamination (Ye, 2011), Libin Zhao’s simulation of delamination using
cohesive elements (Zhao, 2014) and Jalal Yousefi’s CZM to simulate delamination growth
(Yousefi, 2015).
2.6.3 Virtual Crack Closure Technique (VCCT)
The virtual crack closure technique (VCCT) is widely used for computing energy
release rates based on results from continuum (2D) and solid (3D) finite element (FE)
analyses to supply the mode separation required when using the mixed mode fracture
criterion (Jimenez, 2004). Lately, an increased interest in using a fracture mechanics–based
approach to assess the damage tolerance of composite structures in the design phase and
during certification has also renewed the interest in the virtual crack closure technique
(Tay, 2003).
Page 40
26
The VCCT can be used to analyze delamination in laminated materials using a
fracture mechanics approach. The method implements linear elastic fracture mechanics
(LEFM). This LEFM method is used for delamination analysis in composite laminates
which determines the total strain energy release rate (GT) which is the sum of individual
components GI, GII, and GIII (Mohammed, 2014).
The virtual crack closure technique (VCCT) is the most popular and powerful tool
to approximately compute G's values. It was first introduced by Rybicki and Kanninen
(Rybicki, 1977) for 2D crack problems and was extended to 3D crack problems by
Shivakumar (Shivakumar, Tan, & Newman, 1988).
Wang et al. made significant contributions to improve the accuracy and enhance
the capability of the approach (Wang & Raju, 1996). Some applications of VCCT include
the delamination of composites (Krueger & O'Brien, 2001), the de-bonding of skin-
stiffeners (Krueger, Paris, O'Brien, & Minguet, 2002), and the failure of adhesively bonded
joints (Xie, Chung, Waas, Shahwan, Schroeder, Boeman, & Klett, 2005). VCCT based
crack-growth simulation can be created involving the following assumptions (Reeder,
Song, Chunchu, & Ambur, 2002):
a. Crack growth occurs along a pre-defined crack path.
b. The path is defined via interface elements.
c. The analysis is quasi-static and does not account for transient effects.
d. The material is linearly elastic and can be one of isotropic, orthotropic
or anisotropic material.
Abaqus/CAE has been used in this research to create the finite element model for
delamination analysis. Delamination in aerospace structures at the Boeing Company have
Page 41
27
been extensively investigated using VCCT implementation in Abaqus.
2.6.4 Abaqus software
Figure 2.10 Abaqus/CAE main user interface
The core of the Abaqus are Abaqus/Standard and Abaqus/Explicit which are the
analysis modules tools integrated into Abaqus. Abaqus/Standard is a general purpose finite
element module. It is used in analyzing many types of problems including nonstructural
applications. On the other hand Abaqus/Explicit is an explicit dynamics finite element
module in Abaqus. Abaqus/CAE (Complete Abaqus Environment) incorporates the
analysis modules for modeling, managing and monitoring Abaqus analyses and results.
Abaqus/CAE can be customized to create application specific systems. It integrates
modeling, analysis, job management and result evaluation seamlessly. It also provides
Page 42
28
complete interface with Abaqus solver programs available.
Abaqus/CAE has one of the best modern graphical user interface (GUI) with icons,
menus and dialog boxes. This GUI provides access to all capabilities, accelerate access to
frequently used features and to select various other options.
Abaqus/CAE has various modules that are easily accessible. Each module contains
a logical subset of the overall functionality. It also has a Model Tree with a graphical view
of the model created, and all the objects that it contains, as shown in Figure 2.11. It acts
like a convenient, centralized tool for moving between modules and for managing objects.
Page 43
29
Figure 2.11 Model Tree with all modules and objects
Page 44
30
Experimental Procedure
There were three considerations in detecting delamination in carbon fiber
composites that were addressed:
(i) Fabricating piezoresistive nanocomposite sensor
(ii) Fabricating carbon fiber prepreg composite sample (with and without
delamination)
(iii) Embedding nanocomposite on composite sample, and mechanical testing
of the samples (Three point bend flexural test)
3.1 Fabricating nanocomposite sensor
Buckypaper (BP), which is a thin sheet of carbon nanotubes (CNTs) show a great
promise in fabricating multifunctional nanocomposites. One of the serious problems to the
use of CNTs in engineering applications is the inability to synthesize long nanotubes which
is why buckypaper is used in this research to fabricate the sensor. Buckypaper can be
considered as a composite by two ways. One by infusing with resin and the other one by
incorporating into conventional fiber reinforced composites (Wang, Liang, Wang, Zhang,
& Kramer, 2004). Unlike the CNTs directly added into matrix, buckypaper based
composites have much higher concentration of CNTs and high conductivity.
This research uses epoxy resin as matrix material in fabricating buckypaper based
composite sensor (Piezoresistive Nanocomposite). Epoxy modified by adding coarse
graphene platelets, is utilized in fabricating these nanocomposites. Tensile tests and
simultaneous electrical resistivity measurements are performed on these nanocomposite
samples which are used for further analysis in detecting delamination in carbon fiber
composites.
Page 45
31
3.1.1 Materials
The buckypaper (multiwall CNT sheet) was procured from NanoTech Labs
which consists of 100% free standing nanotubes. This buckypaper has an area
density of 21.7 g/m2 and surface electrical resistivity of 1.5 W/m2. The electrical
resistivity was measured independently through experiments.
The graphene sheet (6 inch x 6 inch) supplied by Graphene Supermarket
has low resistance of 2.8x10-2 W/m2. It is used as coarse graphene platelets after
finely chopping the graphene sheet in to size between 300–1000mm. This graphene
sheet is made out of multiple layers of nanoscale fine graphene platelets adhesively
bonded together.
The silver epoxy resin supplied by MG chemicals has high conductivity and
high adhesive properties. This epoxy has a 1:1 mix ratio of epoxy and hardener and
a 4 hour working time. This conductive epoxy is used in attaching electrodes to
nanocomposites.
The regular epoxy resin is a West System # 105 Epoxy Resin with West
System # 206 Slow Hardener. This epoxy has a 5:1 mix ratio of epoxy and hardener,
and a 20 minute working time. This epoxy is a light amber, low-viscosity liquid
epoxy resin specifically formulated as functions of wetting out, bonding with fiber
glass, carbon fiber and other materials.
The other materials required for the fabrication include copper plates,
peelply, breather film, aluminum tooling plate, epoxy mixing cups, vacuum bag
and complete vacuum set-up.
Page 46
32
3.1.2 Procedure
Figure 3.1 Buckypaper strip
The buckypaper is cut into strips of size 6.35cm x 1.27cm using a laser blade
as shown in Figure 3.1 and copper plates gauging 32 with dimension 1.27cm x 1.27
cm are attached to both sides of cut buckypaper strips using the conductive silver
epoxy paste as shown in Figure 3.2.
Figure 3.2 Buckypaper strip with copper plates
These attached copper plates are used for conductivity measurement during
experiments. These buckypaper strips with copper plates are placed on a peelply
which is again placed on a flat aluminum tooling plate as shown in Figure 3.3.
Page 47
33
Figure 3.3 Buckypaper strips with copper plates on tooling plate
The coarse graphene platelets (5 wt. %) are mixed into the epoxy resin
evenly without mixing hardener as shown in Figure 3.4. Hardener is mixed later
before applying on the buckypaper strips. This increases the working time before
the resin solidifies. The weight of graphene is calculated before to ensure right
weight of graphene in the final mixture (Li & Namilae, 2016).
Figure 3.4 Epoxy resin and Epoxy resin mixture
Page 48
34
This epoxy mixed with 5 wt. % graphene platelets after adding hardener is applied
to both sides of buckypaper strip samples on the tooling plate. The set-up is now covered
with peelply followed by breather film (removes excess epoxy). The final set-up is covered
with a vacuum bag of pressure 88.05 KPa. This vacuum bagging procedure helps the
breather film to absorb extra epoxy.
Figure 3.5 Vacuum bagging schematic
The general vacuum bagging setup is shown in Figure 3.5 for fabricating the
piezoresistive nanocomposite. These nanocomposite samples are peeled from the peelply
after curing the resin for 12 hours at room temperature. The curing time and curing
temperature will vary from one epoxy mixture to another.
To follow up, the buckypaper is cut into strips and copper plates are attached to the
strips which act as electrodes followed by adding graphene platelets and curing them by
applying vacuum which gives the required piezoresistive nanocomposite sensor. Figure 3.8
shows the SEM micrograph of coarse graphene platelets and CNTs on the nanocomposite
sensor.
Vacuum Bag Vacuum
Breather
Sealing
Tape
Release Film
Peel ply
Nanocomposite Strip
Page 49
35
Figure 3.6 Showing Buckypaper strip, attached copper plate and final sensor
Figure 3.7 Peeled off nanocomposite sensors
Figure 3.7 shows some of the peeled off piezoresistive nanocomposites
which are ready for further testing.
Page 50
36
Figure 3.8 SEM micrograph of coarse graphene platelets and CNTs on the sensor
3.2 Fabricating carbon fiber prepreg composite laminate
Carbon fiber prepreg is conventional carbon fiber that has been pre-impregnated
with partially cured resin during manufacture. Because the resin has already been mixed
with hardener, this carbon fiber prepreg needs to be stored at very low temperatures to
prevent the resin from curing before it is used. The strength of carbon fiber is its weave.
The more complex the weave, the more durable the composite will be.
The angle of the weave and the type of resin used with the fiber will determine the
overall strength of composite. The resin is commonly the epoxy that is applied to the carbon
fiber fabric by precisely calibrated machinery with required ratio of resin to reinforcement.
This material has a wide range of applications, as it can be formed at various densities in
unlimited sizes and shapes.
Page 51
37
The advantages of carbon fiber composites are their high stiffness, high strength to
weight ratio, excellent fatigue endurance, corrosion resistance, impact resistance and, as
discussed earlier, their flexibility in design adapt them to any design requirements.
Despite the many advantages, carbon fiber composites also have few disadvantages
such as high material cost, high fabrication cost, and requirements for nondestructive
inspection techniques to detect flaws and damages. This research helps in that direction by
introducing a new sensor for in-situ detection of delamination in composites.
3.2.1 Materials
The carbon fiber prepreg (CF3327-1 EPC: Se-019K) supplied by Hankuk
Carbon is based on 250oF (121oC) curing, consists of a carbon fabric impregnated with
epoxy resin which includes Carbon 3K as a warp and Carbon 3K as a fill. The Fiber
area weight (FAW) of this prepreg is 200 g/m2. The weave of this prepreg is 2X2 twill.
The fiber volume fraction and resin contents are 50% and 40%, respectively. The shelf
life of this prepreg is 6 months at storage temperature of below -18oC. This material is
suitable for fabricating high performance composite structures according to Hankuk
Carbon.
The genesis series hydraulic compression press supplied by Wabash MPI is
ideal for compression molding of rubber, plastic, composites and laminating with
clamp force calculated from 15 to 150 tons of weight. This press features steel platens,
programmable controller, automatic transition from closing to pressing speed, pressure
relief valve with analog pressure gauge, internal hydraulic system with high efficiency
motor, reservoir & water –cooled heat exchanger, digital temperature controls and
many more. This hydraulic compression composite molding press is used to replicate
Page 52
38
the pressure – temperature cycle of an autoclave cure.
Figure 3.9 Typical autoclave cure cycle (Hankuk Carbon)
The other materials required for the fabrication of carbon fiber composite
laminate using Wabash hydraulic compression molding press are tooling plate, electric
scissors, ruler, protractor, a roller, markers, Teflon tape and a 650X tile wet saw.
3.2.2 Fabrication Procedure
Figure 3.10 Carbon fiber prepreg strip
Page 53
39
The carbon fiber prepreg is cut into multiple 15.24x15.24x10-2m strips with 0o
and 45o alignment using electric scissors as shown in Figure 3.10. A clean tooling plate
is taken and eight of these strips are used to layup an 8-ply layup with (0o/45o)4 layup.
A roller is used to roll after laying up each layer to prevent air gaps as shown
in Figure 3.11. Once the layup is completed, this setup is now moved to Wabash
composite compression molding press which is preheated to 250oF. The press is set to
a pressure of 35 psi which is recommended by Hankuk Carbon for curing the prepreg.
A small program is written in the press to cure the prepreg for 90 minutes at 250oF.
This can be seen in Figures 3.12 and 3.13.
Figure 3.11 Prepreg layup using a roller to avoid air gaps
Page 54
40
Figure 3.12 Wabash Press used in curing prepreg
Figure 3.13 Digital temperature controls on Wabash press
Page 55
41
Once cured, an 8 ply composite carbon fiber laminate of dimensions 15.24cm x
15.24cm (6”X6”) is fabricated. This fabricated laminate is later cut into six pieces of
2.54cm x 15.24cm (1”X6”) samples using a 650XT tile wet saw. The similar procedure is
followed for fabricating delaminated composite samples.
For delaminated composite laminate fabrication, after laying up four layers of
prepreg on tooling plate, one layer of Teflon (PTFE) tape of dimension 7.62cm x 15.24cm
is laid on the fourth layer from half to one end. Remaining four layers of prepreg are laid
and the setup is cured as earlier. Once cured a composite laminate with known delamination
fabrication is completed. Using a 650XT tile wet saw six samples of dimension 2.54cm x
15.24cm (1”X6”) are cut from the laminate. These sample have 2.54cm x 7.62cm
delamination in them as shown in Figure 3.14.
Figure 3.14 Delamination in composite laminate sample (schematic)
Page 56
42
3.3 Preparing samples for testing
After the fabrication of piezoresistive nanocomposite and making carbon fiber
composite laminate samplse, the next step is to attach the piezoresistive nanocomposite
sensor on top of the carbon fiber composite laminate sample. Materials required for
this process are primarily nanocomposite sensor and composite laminate sample,
followed by the regular epoxy resin which is a West System # 105 Epoxy Resin with
West System # 206 Slow Hardener.
Figure 3.15 Laminate samples and nanocomposite sensors side by side
Page 57
43
This epoxy has a 5:1 mix ratio of epoxy and hardener, a 20 minute working
time. Vacuum bagging setup is used to cure the epoxy under pressure to remove any
air gaps between sensor and laminate sample.
Figure 3.16 Vacuum bagging setup for attaching nanocomposite sensor
Peelply and breather are used during vacuum bagging to peel off the final
samples and to absorb excess epoxy, respectively. This setup as shown in Figure
3.16 is left for 12 hours to cure in room temperature.
Figure 3.17 Sample soldered with copper wires
Page 58
44
Once cured, the finals samples are peeled off and copper wires are soldered to
the copper tabs on nanocomposite sensor to facilitate stable resistance measurement as
shown in Figure 3.17.
3.4 Electro-Mechanical measurement & Data acquisition
The resistance measurement of nanocomposite sensor is obtained by four point
probe testing method to IEEE and ASTM standard testing methods (ASTM, 2004;
IEEE, 2005; ASTM, 2005). This four point testing technique is specially designed to
measure sheet resistance of thin films. This technique is designed to use separate pairs
of current carrying and voltage sensing electrodes to make accurate measurements than
two terminal testing method which is simpler and more common. Hence, this method
is used in this research for resistance measurements.
This method works by forcing a current through the nanocomposite and
measuring voltage using a four-wire Kelvin-connection scheme. The resistance of the
sample is calculated using Ohm’s Law by passing a controlled current (0.5 Amperes)
and recording a voltage drop (ΔV) which is shown in Figure 3.18. The change in
resistance can be monitored by the LabVIEW.
Voltage drop using Ohm’s law: V
RI
Figure 3.18 Schematic of voltage drop setup
I
c
Power supply
DAQ
R
Page 59
45
National Instruments LabVIEW system design software with a graphical
programming syntax that makes it simple to visualize, create, and code engineering
systems, is unmatched in helping engineers translate their ideas into reality, reduce test
times, and deliver business insights based on collected data. A LabVIEW code is developed
to monitor the voltage drop with a data acquisition system (DAQ) as shown in Figure 3.19.
Figure 3.19 LabVIEW code developed for data acquisition
A tensile test on the nanocomposite is performed using CS-225 Digital Force
Tester. A constant head speed of 0.16 mm/sec is applied to the nanocomposite samples and
the resistance change is recorded as the sample is subject to loading simultaneously.
Followed by this tensile test, a three point bending test is performed on the final laminate
samples.
Page 60
46
3.5 Experimental setup of flexural test
The sample soldered with copper wires (as in Figure 3.17) is marked for flexural
test according to ASTM D7264: Standard test method for Flexural Properties of Polymer
Matrix Composite Materials.
Figure 3.20 ASTM D7264: Three-point loading diagram
In the three-point configuration, the maximum flexural stress is located directly
under the center force application member unlike in four-point configuration the bending
moment is constant between the central force application members. The resultant vertical
shear force in the three-point configuration is present everywhere in the beam except right
under the mid-point force application member. The equations for the three-point setup are
as follows:
a. Maximum flexural stress 2
3( )
2f
FL
bd
b. Maximum strain 2
6( )f
Dd
L
c. Flexural modulus of elasticity 3
3( )
4f
L mE
bd
Page 61
47
Where, L is the support span, b is the width of test beam, d is the depth of test beam,
F is the load at any given point and D is the maximum deflection. Once marked with the
appropriate dimensions from ASTM D7264 standard, the samples are ready for three-point
flexural test.
Figure 3.21 Three-point setup for MTS testing system
Figure 3.22 MTS Testing system
Page 62
48
A standard flexural setup for ASTM standard as shown in Figure 3.21 is used in a
MTS testing system shown in Figure 3.22 for doing the test.
Figure 3.23 Three-point setup for flexural test (ASTM D7264)
The three-point setup is fixed into the MTS testing machine and the sample is kept
between the top and bottom fixture of the setup such that the ASTM markings are aligned
with the roller pins on the three-point setup. The wires of the sample are connected to the
power source as well as the data acquisition system (DAQ) and this whole setup is
controlled using a MTS controller. The DAQ is connected to a computer with LabVIEW
installed in it. A current of certain value is applied to the nanocomposite on the sample.
LabVIEW code developed (shown in Figure 3.19) monitors and controls the
voltage on the sample and MTS testing machine respectively. LabVIEW measures the
voltage drop and corresponding applied load along with the displacement on the on the
sample. The measured voltage is voltage drop resulting from the resistance of the
nanocomposite. The measured load is the load applied to bend the sample for flexural test.
Page 63
49
The measured displacement is the deflection of the center of the sample. The voltage drop
is unstable at the beginning when the current start to flow through samples and then
stabilizes to a constant value.
Figure 3.24 Sample, under load and after deformation
LabVIEW measures the drop in voltage and force-displacement-time
simultaneously while the sample is deformed. The voltage drop data is used to calculate
change in resistance in the nanocomposite while force-displacement data can be used to
calculate stress-strain data using ASTM standard equations. Same three-point flexural
setup is setup for a delaminated sample as well. The voltage drop and force-displacement
data for the delaminated sample are also recorded in LabVIEW. From that data, change in
resistance and stress-strain data is calculated.
Page 64
50
Experimental results of Mechanical and Electrical Behavior of
Composites
Results of mechanical properties and stress-strain plots of the piezoresistive
nanocomposites and the composite laminate samples are presented later in this chapter.
4.1 Mechanical properties of CNTs and carbon fiber composites
4.1.1 Mechanical properties of CNTs
From the moment carbon nanotubes are discovered, it was expected that they would
display good mechanical properties like graphite. Graphite had an in-plane modulus of 1.06
TPa and transverse elastic modulus is 36GPa (Palaci, 2012). CNTs are expected to display
similar stiffness. It is estimated that tensile strength of graphene is as high as 130 GPa and
elastic modulus of graphene is determined to be 1000 GPa (Charles & Gilmore, 2014) from
the properties of C-C bonds.
The calculated Young’s modulus of SWNT using ab initio local density
calculations to determine the parameters in a Keating potential was 1500 Gpa (Overney,
1993), similar to that of graphite. The first direct measurement of Young’s modulus of arc-
MWNTs pinned at one end using an atomic force microscope (AFM) which gave an
average value of 1.28 TPa (Wong, Sheehan, & Lieber, 1997).
Yu et al. in 2000 performed stress-strain measurements on individual arc-MWNT
inside an electron microscope, for a range of tubes they obtained modulus values of 0.27 –
0.95 TPa (Yu, 2000). They also showed fracture of MWNT at strains of up to 12% and
with strengths in the range of 11 to 63 GPa.
The mechanical properties of buckypaper (CNT sheet), with van der Waals bonds
between CNTs, are much lower than those of single CNT. Because of the weak Van der
Page 65
51
Waals Force in buckypaper, the stress cannot be effectively transferred between CNTs.
Measured Young’s moduli of this porous fibrous material reach maximum value of 2 GPa
(Yeh, 2007), which is approximately 0.2 % of the modulus of SWCNT. Compared to
values reported in the literature, Young’s Modulus and tensile strength achieved in our
experiments are lower because the buckypaper used in our work consists of randomly
oriented multiwall nanotubes.
Table 4.1 Young's Modulus and Tensile Strength of buckypaper/polymer nanocomposites
Young's
modulus (GPa)
Tensile
strength (Mpa)
Average tube
diameter (nm)
Average rope
diameter(nm) Reference
8 30 0.8 10~50 (Sreekumar et al.,
2003)
6.9 57 0.8 10~50 (Coleman et al.,
2003)
2.3 6.29 0.8 10~50 (Baughman et al.,
1999)
1.1 17.7 0.8 (Pham et al.,
2008a)
4 32.3 0.8 (Pham et al.,
2008a)
1.5 13.5 1.36 (Pham et al.,
2008a)
2.7 33.2 1.36 (Pham et al.,
2008a)
4.1.2 Mechanical properties of carbon fiber composites
Carbon fiber composite materials in practice can be subjected to a wide variety of
different loading conditions in the form of mechanical stresses and environmental effects
that are related to temperature and moisture (Friedrich, 1989). Mechanical stresses occur
under different types of loading, such as tension, compression, and fatigue in structural
Page 66
52
components. The mechanical properties of carbon fiber composite material used in this
research provided in seller data sheet (Cure 275 oF – Autoclave) are as follows:
Table 4.2 Mech. properties of CF lamina cured for 90min at 275 oF by Autoclave
(Hankuk carbon co. LTD, CF3327-1 EPC : SE-019K)
Specimen Units Results Temp (oC) Test Method
Tensile Strength 0o MPa 716 23 25
Tensile modulus 0o GPa 66 23 25
Poisson’s Ratio 0o - 0.053 23 4
Comp. Strength 0o MPa 547 23 0.20
Comp. modulus 0o GPa 60 23 440
Flex. Strength 0o MPa 930 23 425
Flex. modulus 0o GPa 56 23 440
Inter laminar shear strength 0o MPa 68 23 425
4.2 Experimental results of mechanical properties
Figure 4.1 Stress - Strain plot of nanocomposites with coarse graphene platelets (5 wt. %)
0
2
4
6
8
10
0 0.01 0.02 0.03 0.04 0.05 0.06
Str
ess
(MP
a)
Strain
Stress - Strain
Page 67
53
The stress-strain plot of piezoresistive nanocomposite with 5 wt. % coarse graphene
platelets is shown in Figure 4.1. The maximum stress on the nanocomposite is observed to
be 9.83 MPa and the maximum strain observed to be 0.06. The Young’s modulus is
calculated to be 163.83 MPa.
Similarly, the stress-strain plot for a tension test is shown in Figure 4.2 and Flexural
stress-strain plot for the non-delaminated carbon fiber composite sample calculated from
the force-displacement data from LabVIEW which is plotted in Figure 4.3. The maximum
flexural stress is observed to be 408.26 MPa and the maximum flexural strain is observed
to be 0.0194 mm/mm. The flexural modulus is calculated to be 21.04 GPa. On the other
hand, the stress-strain plot for delaminated sample is also plotted in Figure 4.4. The
maximum flexural stress is observed to be 272.14 MPa and the maximum flexural strain is
observed to be 0.028 mm/mm. The flexural modulus is calculated to be 9.7 GPa.
Figure 4.2 Stress - Strain plot of composite sample in Tension
Page 68
54
Figure 4.3 Stress - Strain plot of non-delaminated composite sample (ASTM D7264)
Figure 4.4 Stress - Strain plot of delaminated composite sample (ASTM D7264)
0
50
100
150
200
250
300
350
400
450
0 0.5 1 1.5 2
Fle
xu
ral
Str
ess
(MP
a)
Flexural Strain 10-2
Stress - Strain
0
50
100
150
200
250
300
0 0.5 1 1.5 2
Fle
xura
l S
tres
s (M
Pa)
Flexural Strain 10-2
Stress - Strain
Page 69
55
4.3 Electrical properties of CNTs and carbon fiber composites
Several studies have focused over the past decade on piezoresistive polymers made
by dispersing CNTs into a polymer to form a conductive matrix (Karimov, 2012; Kang,
2006; Vemuru, 2009; Thostenson, 2008; Grow, 2005; Alamusi & Hu N, 2011; Dharap,
2004; Hu N, 2008; Hu N, 2010; Hu B, 2013). The conductive polymer can be molded to
any desired shape. The conductive nature of the carbon fibers allow their use in sensing. In
this research, the piezoresistive nanocomposite developed by addition of coarse graphene
platelets is used to detect the delamination by using the change in electrical propriety
(resistance) while applying load.
CNTs have high electrical conductivities and extremely large length to diameter
ratios (aspect ratio) and can improve the conductivity of the polymer matrix with only very
low content (Martin, Sandler, Windle, Schwarz, Bauhofer, Schulte, & Shaffer, 2005). They
are widely used in the production of conductive composites, electromagnetic shielding
materials and antielectrostatic materials (Mahapatra, 2008). In addition CNT based
electronics is one of the potential uses of nanotubes. The flexibility of nanoscale design
and the availability of both semiconducting and metallic nanotubes enable a wide variety
of device configurations, starting with an early prototypical devices utilized the surface on
which a nanotube was deposited as a gate (Tans, 1998; Martel, 1998).
Moreover individual CNT have excellent conductivity of about 105 – 108 S/m and
reaches a high aspect ratio up to 100 – 1000 (Laurent, Flahaut, Peigney, & Rousset, 1998).
It has been established that electrical conductivity of buckypapers and mechanical
characteristics decrease with increasing molecular mass of CNTs (Boge, Sweetman, &
Ralph, 2009). The results of electrical resistivity measurements of nanocomposite samples
Page 70
56
with deformation and without deformation are shown here. The electrical resistivity of
piezoresistive nanocomposite is obtained using the following expression:
( )R w t
l
Where, R is the calculated resistance by Ohm’s Law, w and t are the width and
thickness of piezoresistive nanocomposite and l is the length of composite strip.
Figure 4.5 Change in Resistivity - Strain plot of nanocomposites with coarse graphene
platelets (5 wt. %)
Figure 4.5 shows the change in resistivity vs strain plot of piezoresistive
nanocomposite with 5 wt. % coarse graphene platelets. All these values reported are
averaged from tests on five identical nanocomposite samples. From Figure 4.4 at maximum
strain of 0.06 mm/mm, the change in resistivity is observed to be 11.68x10-5 ohm-m.
0
2
4
6
8
10
12
0 0.01 0.02 0.03 0.04 0.05 0.06
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Strain
Resistivity - Strain
Page 71
57
Figure 4.6 Change in Resistivity - Strain plot of non-delaminated composite sample
Figure 4.7 Change in Resistivity - Strain plot of delaminated composite sample
0
0.5
1
1.5
2
2.5
3
3.5
4
0 0.5 1 1.5 2
Ch
ange
in R
esis
tiv
ity 1
0-5
(oh
m-m
)
Flexural Strain 10-2
Resistivity - Strain
0
0.5
1
1.5
2
2.5
3
0 0.5 1 1.5 2
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Flexural Strain 10-2
Resistivity - Strain
Page 72
58
Figure 4.6 shows the change in resistivity against strain plot of a non-delaminated
composite sample from three-point bend flexural test. From the plot at a maximum strain
of 1.94x10-2, a change in resistivity of 3.68x10-5 ohm-m is observed.
Similarly from Figure 4.7, which shows the change in resistivity against strain plot
of a delaminated sample, at the maximum strain of 2.08x10-2, a change in resistivity of
2.67x10-5 ohm-m is observed and maximum change in resistivity of 2.89x10-5 ohm-m is
observed at a strain of 1.87x10-2.
4.4 Scanning Electron Microscopy
The scanning electron microscope micrographs of the fracture specimen is shown
in Figure 4.8 and Figure 4.9.
Figure 4.8 SEM micrograph of fracture specimen
Page 73
59
Figure 4.9 SEM micrograph showing CNTs on fracture specimen
These SEM micrographs of fracture specimen indicate that the nanocomposite is
completely integrated into the composite layup and does not peel off after composite
deformation.
Page 74
60
Finite Element Model of Flexural Test
5.1 Finite element model of non-delaminated sample
The three-point bend flexural test (ASTM D7264) of the composite laminate is
modeled in the general purpose finite element software Abaqus using the standard module
for static analysis. Abaqus standard is used in analyzing many types of problems including
nonstructural applications. Abaqus/CAE (Complete Abaqus Environment) incorporates the
analysis modules for modeling, managing and monitoring Abaqus analyses and results.
The composite laminate is modeled using shell elements with 8-ply layup as in the
experimental setup as shown in Figure 5.1
Figure 5.1 Composite 8-ply layup in Abaqus/CAE
Page 75
61
The lamina properties (E11 = E22 = 66 GPa) are obtained from the supplier data
sheet for the composite. The assembly is modeled similar to three-point bend setup from
the flexural test as shown in Figure 5.2. The diameter of the rollers in the assembly is the
same as the diameter of the roller pins of the three-point bend setup from the experiments.
Figure 5.2 Assembly of three-point bend setup as experiments
Shell planar elements are used to model the laminate and solid extrude elements for
rollers. Contact properties with tangential and normal behaviors are given for the contact
between laminate and rollers. These contact properties are given with surface-to-surface
interactions. The location of rollers are as per ASTM standard as used in experiments.
After assembly, load is applied on the laminate model as in experimental setup and
all the results like applied load and displacement of the center of the laminate are collected
Page 76
62
and plotted in Abaqus/CAE. There are a total of 15808 linear quadrilateral elements of type
S4R in the laminate model with 16165 nodes. Figure 5.3 shows contour of the deformed
laminate when load is applied.
Figure 5.3 Strain contour of deformed laminate sample
5.2 Finite element model of delaminated sample using VCCT
Virtual Crack Closure Technique (VCCT) for Abaqus/CAE is a capability within
Abaqus that provides delaminated / debonding analysis capabilities for structures
containing bonded surfaces.
VCCT for Abaqus utilizes the convergence and stabilization algorithms and the
existing load incrementation capabilities. Supporting active delamination between bonded
surfaces, calculating crack growth based on fracture mechanics, inclusion of mixed mode
crack growth, computing intermediate crack shape, allowing VCCT to be performed during
Page 77
63
nonlinear analysis, etc., are some of key advantages of VCCT for Abaqus/CAE. In this
research, the delamination in the finite element model is incorporated using VCCT in
fracture criterion in the contact property. Direction of crack growth relative to local 1 –
direction is taken as maximum tangential stress direction. A tolerance of 0.2 with zero
viscosity is used in VCCT. The strain energy release rate (G1C) of 600 J/m2 is used as
VCCT input.
Figure 5.4 shows the strain contour of deformed delaminated sample modeled in
Abaqus/CAE using VCCT. The delaminated composite laminate is also modeled using
shell elements (element type) in Abaqus. The results of the simulation are captured and
plotted in Abaqus.
Figure 5.4 Strain contour of deformed delaminated sample
Page 78
64
5.3 Results of Finite Element Model (Abaqus/CAE)
Force and displacements values of the three-point bend flexural test simulation are
collected for both non-delaminated and delaminated (VCCT) model. Figure 6.4 shows the
force vs displacement plot of a non-delaminated composite laminate model from
Abaqus/CAE. This result is again plotted using the data from Abaqus.
Figure 6.4 Force vs Displacement of non-delaminated laminate sample
From the Figure 6.4, a maximum force of 255.53 N is observed for a displacement
of 0.017 m. The values of this plot are obtained from Abaqus/CAE. A similar plot is plotted
for a delaminated sample. Figure 6.5 shows force vs displacement plot of a delaminated
composite laminate model obtained from Abaqus/CAE. The delamination is created in
0
50
100
150
200
250
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018
Forc
e (N
)
Displacement (m)
Simulation
Page 79
65
Abaqus using Virtual Crack Closure Technique (VCCT) in Abaqus/CAE. A maximum
force of 79.27 N is observed at a displacement of 5.39 mm.
Figure 6.5 Force vs Displacement of delaminated laminate sample from Abaqus/CAE
0
10
20
30
40
50
60
70
80
90
0 1 2 3 4 5 6
Forc
e (N
)
Displacement (mm)
Simulation
Page 80
66
Analysis and Discussion
The final results and discussion of experimental and finite element model are
presented in this chapter. The experiments measure the electrical resistivity of the
piezoresistive nanocomposite sensor when the attached composite laminate (delaminated
and non-delaminated) is subject to mechanical deformation from the three-point bend
flexural test. The finite element simulation measures the strain created in the
nanocomposite region on the modeled composite laminate (delaminated using VCCT and
non-delaminated).
6.1 Electro-mechanical properties of piezoresistive nanocomposite
Figure 6.1 Stress – Strain and Resistivity – Strain response of a nanocomposite
0
2
4
6
8
10
12
0
2
4
6
8
10
0 0.01 0.02 0.03 0.04 0.05 0.06
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Str
ain (
MP
a)
Strain
Stress - Strain
Resistivity - Strain
Page 81
67
Figure 6.2 Stress – Strain and Resistivity – Strain response of nanocomposite attached to
a non-delaminated composite laminate
Piezoresistive nanocomposite sensor exhibit change in resistance when subject to
mechanical deformation. As the composite laminate is deformed in three-point flexural
test, the strain is transferred to the nanocomposite attached on top of the laminate. This
strain created on the nanocomposite results in the change in resistivity.
Figure 6.1 shows the Stress – Strain and Resistivity – Strain response of a
piezoresistive nanocomposite in tension, the nanocomposite is used in this research to act
as a sensor in order to detect delamination. Figure 6.2 shows the Stress – Strain response
of composite laminate and Resistivity – Strain response of piezoresistive nanocomposite
in three-point flexural test.
0
0.5
1
1.5
2
2.5
3
3.5
4
0
50
100
150
200
250
300
350
400
450
500
0 0.5 1 1.5 2
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Fle
xu
ral
Str
ess
(MP
a)
Flexural Strain 10-2
Stress - Strain
Resistivity - Strain
Page 82
68
Change in resistivity of 11.68x10-5 ohm-m can be observed in the piezoresistive
nanocomposite from Figure 6.1 and a maximum stress of 9.83 MPa is also observed on it.
Change in resistivity of 3.68x10-5 ohm-m is observed in the piezoresistive nanocomposite
with application of deformation on the composite laminate in Figure 6.2. Maximum
flexural stress of 408.26 MPa is observed on the composite laminate sample (Non-
delaminated).
Figure 6.3 Stress – Strain and Resistivity – Strain response of nanocomposite attached to
a delaminated composite laminate
Figure 6.3 shows the Stress – Strain response of composite laminate and Resistivity
– Strain response of piezoresistive nanocomposite. Change in resistivity of 2.89x10-5 ohm-
m is observed in the piezoresistive nanocomposite with application of deformation on the
composite laminate. Maximum flexural stress of 272.14 MPa is observed on the composite
laminate sample (Delaminated).
0
0.5
1
1.5
2
2.5
3
0
50
100
150
200
250
300
0 0.5 1 1.5 2
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Fle
xura
l S
tres
s (M
Pa)
Flexural Strain 10-2
Stress - Strain
Resistivity - Strain
Page 83
69
6.2 Comparing results of experiments and simulation
The results of experimental tests and Abaqus simulations are compared by plotting
them together for both non-delaminated and delaminated laminate samples.
Figure 6.6 shows the result of experiment and simulation of a non-delaminated
sample. The maximum force that is observed in this plot is 233N in experiments and 255N
in simulation while the maximum displacement observed is 1.77x10-2 m in experiments
and 1.7x10-2 m in the simulations.
Figure 6.7 shows the result of experiment and simulation of a delaminated sample.
The maximum force that is observed in this plot is 60N in experiments and 79N in
simulation while the maximum displacement observed is 5.85x10-3 m in experiments and
5.39x10-3 m in the simulations.
Both of these plots clearly show that there is only a small difference (they are very
close) in the values recorded for both delaminated and non-delaminated cases. This proves
that the modeling of composite laminate and the Virtual crack closure technique (VCCT)
applied in Abaqus/CAE is correct. The next thing will be finding the strain on the
nanocomposite region of the composite laminate model from Abaqus and compare the
strain to the actual strain created on the nanocomposite during tension.
Page 84
70
Figure 6.6 Comparing results of Simulation and Experiment in non-delaminated sample
Figure 6.7 Comparing results of Simulation and Experiment in a delaminated sample
0
50
100
150
200
250
300
0 0.005 0.01 0.015 0.02
Fo
rce
(N)
Displacement (m)
Simulation
Experiment
0
10
20
30
40
50
60
70
80
90
0 1 2 3 4 5 6
Forc
e (N
)
Displacement (mm)
Experiment
Simulation
Page 85
71
6.3 Strain in composite region (on laminate) in Abaqus/CAE
Nanocomposite region is defined on the composite laminate modeled in
Abaqus/CAE. This region helps calculating the strain created in the region when load is
applied for three-point flexural test.
Figure 6.8 Cut out region of the nanocomposite section from assembly
Figure 6.9 Cut out region of the nanocomposite section from mesh
Page 86
72
Figure 6.8 and Figure 6.9 show the cut out region of the nanocomposite section
separated from the assembly and mesh, respectively. After defining and solving the model
for three-point bend flexural simulation, the strain values are collected from the
nanocomposite region defined. Figure 6.10 shows the cut out region of nanocomposite
section from visualization after simulation.
Figure 6.10 Cut out region of the nanocomposite section from Abaqus visualization
The strain in the nanocomposite region of the overall sample is partitioned and
calculated for three elements, one at the center and two at left and right sides of the
nanocomposite region as shown in Figure 6.11. This calculation is done for both non-
delaminated and delaminated laminate sample models. The calculated strain at these three
points of a non-delaminated sample model are plotted against displacement in Figure 6.12.
Page 87
73
Figure 6.11 Nanocomposite region and elements selected for strain calculation
Figure 6.12 Strain vs Displacement plot of three elements in non-delaminated model
From Figure 6.12, maximum strain is identified at the center element which is
13.97x10-3. At left most and right most elements the maximum strain observed is 5.56x10-
3 and 6.64x10-3, respectively. Similarly the strain is plotted for a delaminated laminate as
well, as shown in Figure 6.13.
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0 5 10 15
Str
ain
Displacement (m) 10-3
Left
Center
Right
Page 88
74
Figure 6.13 Strain vs Displacement plot of three elements in a delaminated model
From Figure 6.15, maximum strain is identified at the center element which is
21.61x10-3. At left most and right most elements the maximum strains observed are
7.51x10-3 and 7.96x10-3, respectively. Once the strain in the nanocomposite region is
obtained from Abaqus/CAE, the next step is to correlate this strain to the tension test results
obtained from the piezoresistive nanocomposite as shown in Figure 6.1.
6.4 Strain correlation of piezoresistive nanocomposite
The strain observed on the nanocomposite region in Abaqus/CAE for both non-
delaminated model and delaminated model are correlated to the strain created on the
piezoresistive nanocomposite during the tension test (Figure 6.1). Figures 6.14 and 6.15
show the correlation plots of strain and resistivity of a nanocomposite for non-delaminated
model and delaminated model.
0
0.005
0.01
0.015
0.02
0.025
0 1 2 3 4 5 6
Str
ain
Displacement (mm)
Left
Right
Center
Page 89
75
Figure 6.14 Correlation between the strain and resistivity of nanocomposite from three-
point bend test and tension test of a non-delaminated model
Figure 6.15 Correlation between the strain and resistivity of nanocomposite from three-
point bend test and tension test of a delaminated model
Page 90
76
The strain correlated to the tension test of the nanocomposite and corresponding
electrical resistivity is compared to that of experimentally obtained in the flexural test. A
one-to-one correlation between strain in the nanocomposite and electrical resistivity in both
tension and bending is observed. This proves that the simulations and the experiments in
studying the piezoresistive nanocomposite are comparable.
6.5 Conclusion
Based on all the results above and plotting the experimental results of non-
delaminated and delaminated samples together we have Figure 6.16
Figure 6.16 Stress – Strain and Resistivity – Strain response from flexural tests
0
0.5
1
1.5
2
2.5
3
3.5
4
0
50
100
150
200
250
300
350
400
450
0 0.5 1 1.5 2
Chan
ge
in R
esis
tivit
y 1
0-5
(ohm
-m)
Fle
xura
l S
tres
s (M
Pa)
Flexural Strain 10-2
Stress - Strain Baseline
Stress - Strain 3" Delamination
Resistivity - Strain Baseline
Resistivity - Strain 3" Delamination
Page 91
77
From Figure 6.16 it can be observed that, the change in resistivity at maximum
strain for a non-delaminated laminate is 3.68x10-5 ohm-m while in the delaminated sample
it is 2.67x10-5 ohm-m. This indicates that the piezoresistive nanocomposite sensor that is
developed in this research can be effectively used to detect strain changes caused by
delamination and other defects in composite structures.
Page 92
78
Summary and Recommendations
7.1 Summary
In this research, piezoresistive nanocomposite sensors are fabricated from
buckypaper with 5 % wt graphene nanoplatelets added with resin by vacuum assisting
process, and cured in room temperature. Resistance of these nanocomposite sensors are
measured using the four-point probe testing method under applied tensile loading.
Composite laminates composed of 8 plys are also fabricated using compression press with
and without delamination for testing the nanocomposite sensors.
The final sample is prepared by attaching the nanocomposite sensor on top of the
composite laminate and marked with ASTM D7264 markings. A three-point bend flexural
test is carried out on these samples using the MTS testing machine connected to a computer
with LabVIEW code developed (Figure 3.18) in it. The sensors are subject to electrical
current and LabVIEW code is used to monitor the voltage drop of the nanocomposites
along with measuring the stress and strain on the laminate by flexural test. The resistance
can then be calculated using Ohm’s law.
The resistance of the nanocomposite changes due to the mechanical strain created
due to flexural test and is measured through these experiments. The resisitivity – strain of
the nanocomposite sensor attached on top of delaminated and non-delaminated sample
shows a great variation in both cases. For a non-delaminated composite laminate sample,
the resistivity occured at maximum strain is 3.68x10-5 W-m, while in the delaminated
composite laminate sample it is 2.67x10-5 W-m. Seeing all these results, it is believed that
the piezoresistive nanocomposite sensor can be used to detect delamination in carbon fiber
composite structures.
Page 93
79
The finite element models are used to correlate strains from the three-point flexural
tests of carbon fiber composite laminates – nanocomposite sensor assembly to the strains
and resistivity changes in the tension tests of standalone nanocomposites indicating
applicability of these sensors under multiple loading conditions.
7.2 Recommendations for Future Work
The piezoresistive nanocomposite sensors developed in this research can be
employed to detect the effectiveness of a composite repair patch as shown in Figure 7.1.
Future work along these lines will advance the concepts developed here. Also these kinds
of studies can be expanded to include more advanced structures such as stiffened panels,
wing-ribs, fuselage panels etc.
Figure 7.1 Composite repair patch with nanocomposite sensor for detecting the
effectiveness of the repair
Page 94
80
REFERENCES
Abrate, S. (2005). Impact on composite structures. Cambridge university press.
Ajayan, P. M., & Ebbesen, T. W. (1997). Nanometre-size tubes of carbon. Reports on
Progress in Physics, 60(10), 1025.
Ashby, M. F., Bush, S. F., Swindells, N., Bullough, R., Ellison, G., Lindblom, Y., &
Barnes, J. F. (1987). Technology of the 1990s: Advanced materials and predictive
design [and discussion]. Philosophical Transactions of the Royal Society of London
A: Mathematical, Physical and Engineering Sciences, 322(1567), 393-407.
ASTM – Standard (2004). Test Method for D-C Resistance or Conductance of Moderately
Conductive Materials.
ASTM – Standard (2005). Test Methods for DC Resistance or Conductance of Insulating
Materials.
Atiqullah, M. M. Effect of Defects on Mechanical Properties of Composites:
Undergraduate Research on Materials.
Aymerich, F., & Staszewski, W. J. (2010). Impact damage detection in composite
laminates using nonlinear acoustics. Composites Part A: Applied Science and
Manufacturing, 41(9), 1084-1092.
Aymerich, F., & Staszewski, W. J. (2010). Impact damage detection in composite
laminates using nonlinear acoustics. Composites Part A: Applied Science and
Manufacturing, 41(9), 1084-1092.
Barenblatt, G. I. (1962). The mathematical theory of equilibrium cracks in brittle fracture.
Advances in applied mechanics, 7, 55-129.
Baughman, R. H., Cui, C., Zakhidov, A. A., Iqbal, Z., Barisci, J. N., Spinks, G. M., ... &
Jaschinski, O. (1999). Carbon nanotube actuators. Science, 284(5418), 1340-1344.
Berber, S., Kwon, Y. K., & Tománek, D. (2000). Unusually high thermal conductivity of
carbon nanotubes. Physical review letters, 84(20), 4613.
Bhowmick, A. K., De, S. K., & Bandyopadhyay, S. (2003). Short melamine fiber filled
nitrile rubber composites. Journal of applied polymer science, 90(2), 544-558.
Boge, J., Sweetman, L. J., & Ralph, S. F. (2009). The effect of preparation conditions and
biopolymer dispersants on the properties of SWNT buckypapers. Journal of
Materials Chemistry, 19(48), 9131-9140.
Cao, C. L., Hu, C. G., Fang, L., Wang, S. X., Tian, Y. S., & Pan, C. Y. (2011). Humidity
Page 95
81
sensor based on multi-walled carbon nanotube thin films. Journal of
Nanomaterials, 2011, 5.
Charlier, J. C., & Issi, J. P. (1996). Electrical conductivity of novel forms of carbon. Journal
of Physics and Chemistry of Solids, 57(6), 957-965.
Chen, I. W. P., Liang, Z., Wang, B., & Zhang, C. (2010). Charge-induced asymmetrical
displacement of an aligned carbon nanotube buckypaper actuator. Carbon, 48(4),
1064-1069.
Chen, Y., Cheng, C. Y., Miao, H. Y., Zhang, M., Liang, R., Zhang, C., ... & Wang, B.
(2013). Application of response surface methodology in the optimization of laser
treatment in buckypaper lighting for field emission displays. The International
Journal of Advanced Manufacturing Technology, 64(1-4), 515-536.
Cheng, P., Gong, X. J., Hearn, D., & Aivazzadeh, S. (2011). Tensile behaviour of patch-
repaired CFRP laminates. Composite structures, 93(2), 582-589.
Cheng, Q., Wang, B., Zhang, C., & Liang, Z. (2010). Functionalized Carbon‐Nanotube
Sheet/Bismaleimide Nanocomposites: Mechanical and Electrical Performance
Beyond Carbon‐Fiber Composites. Small, 6(6), 763-767.
Coleman, J. N., Blau, W. J., Dalton, A. B., Munoz, E., Collins, S., Kim, B. G., ... &
Baughman, R. H. (2003). Improving the mechanical properties of single-walled
carbon nanotube sheets by intercalation of polymeric adhesives. Applied Physics
Letters, 82(11), 1682-1684.
Coleman, J. N., Khan, U., Blau, W. J., & Gun’ko, Y. K. (2006). Small but strong: a review
of the mechanical properties of carbon nanotube–polymer composites. Carbon,
44(9), 1624-1652.
Corigliano, A., & Ricci, M. (2001). Rate-dependent interface models: formulation and
numerical applications. International Journal of Solids and Structures, 38(4), 547-
576.
Cornec, A., Scheider, I., & Schwalbe, K. H. (2003). On the practical application of the
cohesive model. Engineering fracture mechanics, 70(14), 1963-1987.
Dharap, P., Li, Z., Nagarajaiah, S., & Barrera, E. V. (2004). Nanotube film based on single-
wall carbon nanotubes for strain sensing. Nanotechnology, 15(3), 379-82.
Dresselhaus, M. S., Dresselhaus, G., & Saito, R. (1995). Physics of carbon nanotubes.
Carbon, 33(7), 883-891.
Dugdale, D. S. (1960). Yielding of steel sheets containing slits. Journal of the Mechanics
and Physics of Solids, 8(2), 100-104.
Page 96
82
Ebbesen, T. W., Lezec, H. J., Hiura, H., Bennett, J. W., Ghaemi, H. F., & Thio, T. (1996).
Electrical-conductivity of individual carbon nanotubes. Nature, 382(6586), 54-56.
Endo, M., Muramatsu, H., Hayashi, T., Kim, Y. A., Terrones, M., & Dresselhaus, M. S.
(2005). Nanotechnology: ‘Buckypaper’from coaxial nanotubes. Nature, 433(7025),
476-476.
Fitzer, E. (1989). Pan-based carbon fibers—present state and trend of the technology from
the viewpoint of possibilities and limits to influence and to control the fiber
properties by the process parameters. Carbon, 27(5), 621-645.
Flightglobal: https://www.flightglobal.com/news/articles/delamination-prompts-boeing-
to-inspect-787-fleet-367793/
Friedrich, K. (1989). Fractographic analysis of polymer composites. Application of
fracture mechanics to composite materials, 6.
Fruehmann, R. K., Wang, W., Dulieu-Barton, J. M., & Quinn, S. (2011). The application
of thermoelastic stress analysis to evaluate debond damage in composite sandwich
structures. In Applied Mechanics and Materials (Vol. 70, pp. 470-475). Trans Tech
Publications.
Funazaki, N., Hemmi, A., Ito, S., Asano, Y., Yano, Y., Miura, N., & Yamazoe, N. (1995).
Application of semiconductor gas sensor to quality control of meat freshness in
food industry. Sensors and Actuators B: Chemical, 25(1), 797-800.
Gao, S. L., Zhuang, R. C., Zhang, J., Liu, J. W., & Mäder, E. (2010). Glass fibers with
carbon nanotube networks as multifunctional sensors. Advanced Functional
Materials, 20(12), 1885-1893.
Gao, X., & Kim, J. (2006). Modeling of ductile fracture: significance of void coalescence.
International Journal of Solids and Structures, 43(20), 6277-6293.
Gardiner, G. (2011). Primary structure repair: the quest of quality. High-Performance
Composites, 19(6), 48-54.
Garnier, C., Pastor, M. L., Eyma, F., & Lorrain, B. (2011). The detection of aeronautical
defects in situ on composite structures using Non Destructive Testing. Composite
structures, 93(5), 1328-1336.
Gau, C., Ko, H. S., & Chen, H. T. (2009). Piezoresistive characteristics of MWNT
nanocomposites and fabrication as a polymer pressure sensor. Nanotechnology,
20(18), 185503.
Geim, A. K., & Novoselov, K. S. (2007). The rise of graphene. Nature materials, 6(3), 183-
Page 97
83
191.
Georgakilas, V., Perman, J. A., Tucek, J., & Zboril, R. (2015). Broad family of carbon
nanoallotropes: classification, chemistry, and applications of fullerenes, carbon
dots, nanotubes, graphene, nanodiamonds, and combined superstructures.
Chemical reviews, 115(11), 4744-4822.
Gilmore, C. (2014). Materials science and engineering properties, SI Edition, ISBN-13:
978-1-111-98861-6:344-345
Gnidakouong, J. R. N., Kim, M., Park, H. W., Park, Y. B., Jeong, H. S., Jung, Y. B., ... &
Park, J. M. (2013). Electromagnetic interference shielding of composites consisting
of a polyester matrix and carbon nanotube-coated fiber reinforcement. Composites
Part A: Applied Science and Manufacturing, 50, 73-80.
Gommans, H. H., Alldredge, J. W., Tashiro, H., Park, J., Magnuson, J., & Rinzler, A. G.
(2000). Fibers of aligned single-walled carbon nanotubes: polarized Raman
spectroscopy. Journal of Applied Physics, 88(5), 2509-2514.
Gong, T., Zhang, Y., Liu, W., Wei, J., Li, C., Wang, K., ... & Zhong, M. (2007). Connection
of macro-sized double-walled carbon nanotube strands by bandaging with double-
walled carbon nanotube films. Carbon, 45(11), 2235-2240.
Grimberg, R., Premel, D., Savin, A., Le Bihan, Y., & Placko, D. (2001). Eddy current
holography evaluation of delamination in carbon-epoxy composites. Insight, 43(4),
260-264.
Grimberg, R., Savin, A., & Rotundu, C. R. (2001). Eddy current microscopy applied to
graphite-epoxy composite. Sensors and Actuators A: Physical, 91(1), 73-75.
Grow, R. J., Wang, Q., Cao, J., Wang, D., & Dai, H. (2005). Piezoresistance of carbon
nanotubes on deformable thin-film membranes. Applied Physics Letters, 86(9),
093104.
Halmshaw, R. (1995). Development of industrial radiography technique over the last fifty
years. Insight, 37(9), 684-687.
Hamada, N., Sawada, S. I., & Oshiyama, A. (1992). New one-dimensional conductors:
graphitic microtubules. Physical Review Letters, 68(10), 1579.
Harris, B. (1991). A perspective view of composite materials development. Materials &
Design, 12(5), 259-272.
Hennrich, F., Lebedkin, S., Malik, S., Tracy, J., Barczewski, M., Rösner, H., & Kappes,
M. (2002). Preparation, characterization and applications of free-standing single
walled carbon nanotube thin films. Physical Chemistry Chemical Physics, 4(11),
Page 98
84
2273-2277.
Hillerborg, A., Modéer, M., & Petersson, P. E. (1976). Analysis of crack formation and
crack growth in concrete by means of fracture mechanics and finite elements.
Cement and concrete research, 6(6), 773-781.
Hu, B., Hu, N., Cai, Y., Yuan, W., Yan, C., Cai, Y., ... & Matsushita, M. (2013).
Performance characterization of VGCF/epoxy nanocomposite sensors under static
load cycles and in static structural health monitoring. Smart Materials and
Structures, 22(4), 045008.
Hu, F. Z., & Soutis, C. (2000). Strength prediction of patch-repaired CFRP laminates
loaded in compression. Composites Science and Technology, 60(7), 1103-1114.
Hu, N., Fukunaga, H., Atobe, S., Liu, Y., & Li, J. (2011). Piezoresistive strain sensors made
from carbon nanotubes based polymer nanocomposites. Sensors, 11(11), 10691-
10723.
Hu, N., Karube, Y., Arai, M., Watanabe, T., Yan, C., Li, Y., & Fukunaga, H. (2010).
Investigation on sensitivity of a polymer/carbon nanotube composite strain sensor.
Carbon, 48(3), 680-687.
Hu, N., Karube, Y., Yan, C., Masuda, Z., & Fukunaga, H. (2008). Tunneling effect in a
polymer/carbon nanotube nanocomposite strain sensor. Acta Materialia, 56(13),
2929-2936.
Hulton, A. W. (2015). Modeling Unidirectional Composite Laminates Using XFEM (No.
NUWC-NPT-TR-12-178). NAVAL UNDERSEA WARFARE CENTER DIV
NEWPORT RI RANGES ENGINEERING AND ANALYSIS DEPT.
IEEE Standard (2005). Test Methods for Measurement of Electrical Properties of Carbon
Nanotubes,
Jiang, Y., Tay, T. E., Chen, L., & Sun, X. S. (2013). An edge-based smoothed XFEM for
fracture in composite materials. International Journal of Fracture, 179(1-2), 179-
199.
Jimenez, M. A., & Miravete, A. (2004). Application of the finite-element method to predict
the onset of delamination growth. Journal of Composite Materials, 38(15), 1309-
1335.
Joseph, H. (1997). MEMS in the medical world. Sensors-the Journal of Applied Sensing
Technology, 14(4), 47-51.
Kachanov, L. (2012). Delamination buckling of composite materials (Vol. 14). Springer
Science & Business Media.
Page 99
85
Kang, I., Schulz, M. J., Kim, J. H., Shanov, V., & Shi, D. (2006). A carbon nanotube strain
sensor for structural health monitoring. Smart materials and structures, 15(3), 737.
Karihaloo, B. L., & Xiao, Q. Z. (2003). Modelling of stationary and growing cracks in FE
framework without remeshing: a state-of-the-art review. Computers & Structures,
81(3), 119-129.
Karimov, K. S., Khalid, F. A., & Chani, M. T. S. (2012). Carbon nanotubes based strain
sensors. Measurement, 45(5), 918-921.
Kastner, J., Plank, B., Salaberger, D., & Sekelja, J. (2010, November). Defect and porosity
determination of fibre reinforced polymers by X-ray computed tomography. In 2nd
International Symposium on NDT in Aerospace (pp. 1-12).
Kauffman, D. R., & Star, A. (2010). Graphene versus carbon nanotubes for chemical sensor
and fuel cell applications. Analyst, 135(11), 2790-2797.
Kelly, A., & Macmillan, N. H. (1986). Strong Solids. Oxford University Press, Walton
Street, Oxford OX 2 6 DP, UK, 1986.
Kelly, B. T. (1981). Physics of graphite.
Kroto, H. W. (1991). Allaf, AW, Balm, SP. Chem. Rev, 91(6), 1213.
Krueger, R., & O'Brien, T. K. (2001). A shell/3D modeling technique for the analysis of
delaminated composite laminates. Composites Part A: applied science and
manufacturing, 32(1), 25-44.
Krueger, R., Paris, I. L., O'Brien, T. K., & Minguet, P. J. (2002). Comparison of 2D finite
element modeling assumptions with results from 3D analysis for composite skin-
stiffener debonding. Composite Structures, 57(1), 161-168.
Laurent, C., Flahaut, E., Peigney, A., & Rousset, A. (1998). Metal nanoparticles for the
catalytic synthesis of carbon nanotubes. New Journal of Chemistry, 22(11), 1229-
1237.
Leny, M., & Narayanankutty, S. K. (2009). Development of Elastomeric Hybrid
Composite Based on Synthesised Manosilica and Short Nylon Fiber (Doctoral
dissertation, Cochin University of Science and Technology).
Li, J., & Namilae, S. (2016, February). Nanotube Sheet‐Graphite Hybrid Nanocomposite
for Damage Detection. In TMS 2016: 145 Annual Meeting & Exhibition:
Supplemental Proceedings (pp. 69-76). John Wiley & Sons, Inc.,
Li, X., Levy, C., & Elaadil, L. (2008). Multiwalled carbon nanotube film for strain sensing.
Page 100
86
Nanotechnology, 19(4), 045501.
Li, Z., Dharap, P., Nagarajaiah, S., Barrera, E. V., & Kim, J. D. (2004). Carbon nanotube
film sensors. Advanced materials, 16(7), 640-643.
Liu, P. F., & Islam, M. M. (2013). A nonlinear cohesive model for mixed-mode
delamination of composite laminates. Composite Structures, 106, 47-56.
Lourie, O., & Wagner, H. D. (1998). Evaluation of Young's modulus of carbon nanotubes
by micro-Raman spectroscopy. Journal of Materials Research, 13(09), 2418-2422.
Lubin, G. (1982). Hand book of Composites.
Mahapatra, S. P., Sridhar, V., Chaudhary, R. N. P., & Tripathy, D. K. (2008). AC
conductivity and positive temperature coefficient effect in microcellular EPDM
vulcanizates. Polymer Composites, 29(10), 1125-1136.
Mallick, P. K. (2007). Fiber-reinforced composites: materials, manufacturing, and design.
CRC press.
Mallick, P. K. (2007). Fiber-reinforced composites: materials, manufacturing, and design.
CRC press.
Mark, H. F., Bikales, N., Overberger, C. G., Menges, G., & Kroschwitz, J. I. (1987).
Encyclopedia of polymer science and engineering, Vol. 10: Nonwoven fabrics to
photopolymerization.
Martel, R., Schmidt, T., Shea, H. R., Hertel, T., & Avouris, P. (1998). Single-and multi-
wall carbon nanotube field-effect transistors. Applied Physics Letters, 73(17),
2447-2449.
Martin, C. A., Sandler, J. K. W., Windle, A. H., Schwarz, M. K., Bauhofer, W., Schulte,
K., & Shaffer, M. S. P. (2005). Electric field-induced aligned multi-wall carbon
nanotube networks in epoxy composites. Polymer, 46(3), 877-886.
Meola, C., & Carlomagno, G. M. (2004). Recent advances in the use of infrared
thermography. Measurement science and technology, 15(9), R27.
Mintmire, J. W., & White, C. T. (1995). Electronic and structural properties of carbon
nanotubes. Carbon, 33(7), 893-902.
Mittal, G., Dhand, V., Rhee, K. Y., Park, S. J., & Lee, W. R. (2015). A review on carbon
nanotubes and graphene as fillers in reinforced polymer nanocomposites. Journal
of Industrial and Engineering Chemistry, 21, 11-25.
Mohammed W & Kiran K. N. (2014). “Finite Element Modelling for Delamination
Page 101
87
Analysis of Double Cantilever Beam Specimen.” SSRG International Journal of
Mechanical Engineering (SSRG – IJME) Volume 1 Issue 5
Motamedi, D., & Milani, A. S. (2013). 3D nonlinear XFEM simulation of delamination in
unidirectional composite laminates: a sensitivity analysis of modeling parameters.
Open Journal of Composite Materials, 3(04), 113.
Motamedi, D., Milani, A. S., Komeili, M., Bureau, M. N., Thibault, F., & Trudel-Boucher,
D. (2014). A stochastic XFEM model to study delamination in PPS/glass UD
composites: effect of uncertain fracture properties. Applied Composite Materials,
21(2), 341-358.
Needleman, A. (1990). An analysis of decohesion along an imperfect
interface. International Journal of Fracture, 42(1), 21-40.
Nielsen L.E., & Fields, J.E (1994) U.S. 3652491 (march28,1972)) (Woods D.W., Ward
I.M., J. Mater. Sci., 29, 2572
Ong, K. G., Zeng, K., & Grimes, C. A. (2002). A wireless, passive carbon nanotube-based
gas sensor. IEEE Sensors Journal, 2(2), 82-88.
Overney, G., Zhong, W., & Tomanek, D. (1993). Structural rigidity and low frequency
vibrational modes of long carbon tubules. Zeitschrift für Physik D Atoms,
Molecules and Clusters, 27(1), 93-96.
Palaci, I., Fedrigo, S., Brune, H., Klinke, C., Chen, M., & Riedo, E. (2012). Radial elasticity
of multi-walled carbon nanotubes. arXiv preprint arXiv:1201.5501.
Penza, M., Cassano, G., Aversa, P., Antolini, F., Cusano, A., Cutolo, A., & Nicolais, L.
(2004). Alcohol detection using carbon nanotubes acoustic and optical sensors.
Applied Physics Letters, 85(12), 2379-2381.
Pham, H., Yu, H., & Laski, F. A. (2008). Cofilin/ADF is required for retinal elongation
and morphogenesis of the Drosophila rhabdomere. Developmental biology, 318(1),
82-91.
Prato, M. (2009). Fullerene polymers: synthesis, properties and applications. N. Martín, &
F. Giacalone (Eds.). John Wiley & Sons.
Qiao, P., Lu, K., Lestari, W., & Wang, J. (2007). Curvature mode shape-based damage
detection in composite laminated plates. Composite Structures, 80(3), 409-428.
Rahul-Kumar, P., Jagota, A., Bennison, S. J., Saigal, S., & Muralidhar, S. (1999). Polymer
interfacial fracture simulations using cohesive elements. Acta materialia, 47(15),
4161-4169.
Page 102
88
Reeder, J. R., Song, K., Chunchu, P., & Ambur, D. R. (2002, April). Postbuckling and
growth of delaminations in composite plates subjected to axial compression. In The
43rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and
Materials Conference.
Repetto, E. A., Radovitzky, R., & Ortiz, M. (2000). Finite element simulation of dynamic
fracture and fragmentation of glass rods. Computer Methods in Applied Mechanics
and Engineering, 183(1), 3-14.
Richardson, C. L., Hegemann, J., Sifakis, E., Hellrung, J., & Teran, J. M. (2009). An
XFEM method for modeling geometrically elaborate crack propagation in brittle
materials. Institute for Computational and Applied Mathematics, 90095, 1555.
Richardson, T. (1987). Composites: a design guide. Industrial Press Inc, 200 Madison Ave,
New York, New York 10016, USA, 1987. 343.
Rigueur, J. L., Hasan, S. A., Mahajan, S. V., & Dickerson, J. H. (2010). Buckypaper
fabrication by liberation of electrophoretically deposited carbon nanotubes.
Carbon, 48(14), 4090-4099.
Roberts, T. (2006). The carbon fibre industry: Global strategic market evaluation 2006-
2010. Materials Technology Publication; pp. 10, 93–177, 237
Roe, K. L., & Siegmund, T. (2003). An irreversible cohesive zone model for interface
fatigue crack growth simulation. Engineering fracture mechanics, 70(2), 209-232.
Rose, J. L. (2007, December). An introduction to ultrasonic guided waves. In 4th Middle
east NDT conference and exhibition.
Rybicki, E. F., & Kanninen, M. F. (1977). A finite element calculation of stress intensity
factors by a modified crack closure integral. Engineering Fracture Mechanics, 9(4),
931-938.
Saeedifar, M., Fotouhi, M., Najafabadi, M. A., & Toudeshky, H. H. (2015). Prediction of
delamination growth in laminated composites using acoustic emission and
Cohesive Zone Modeling techniques. Composite Structures, 124, 120-127.
Scheider, I., & Brocks, W. (2006). Cohesive elements for thin-walled structures.
Computational Materials Science, 37(1), 101-109.
Shivakumar, K. N., Tan, P. W., & Newman, J. C. (1988). A virtual crack-closure technique
for calculating stress intensity factors for cracked three dimensional bodies.
International Journal of Fracture, 36(3), R43-R50.
Shivananju, B. N., Yamdagni, S., Fazuldeen, R., Kumar, A. S., Hegde, G. M., Varma, M.
M., & Asokan, S. (2013). CO2 sensing at room temperature using carbon nanotubes
Page 103
89
coated core fiber Bragg grating. Review of Scientific Instruments, 84(6), 065002.
Shokrieh, M. M., & Taheri-Behrooz, F. (2009). Progressive fatigue damage modeling of
cross-ply laminates, I: Modeling strategy. Journal of composite materials.
Sinha, N., & Yeow, J. W. (2005). Carbon nanotubes for biomedical applications. IEEE
transactions on nanobioscience, 4(2), 180-195.
Smith, R. A. (2009). Composite defects and their detection. Materials Science and
Engineering, 3, 103-143.
So, H. M., Sim, J. W., Kwon, J., Yun, J., Baik, S., & Chang, W. S. (2013). Carbon nanotube
based pressure sensor for flexible electronics. Materials Research Bulletin, 48(12),
5036-5039.
Sohn, M. S., Hu, X. Z., Kim, J. K., & Walker, L. (2000). Impact damage characterisation
of carbon fibre/epoxy composites with multi-layer reinforcement. Composites Part
B: Engineering, 31(8), 681-691.
Sosa, J. C., & Karapurath, N. (2012). Delamination modelling of GLARE using the
extended finite element method. Composites Science and Technology, 72(7), 788-
791.
Sreekumar, T. V., Liu, T., Kumar, S., Ericson, L. M., Hauge, R. H., & Smalley, R. E.
(2003). Single-wall carbon nanotube films. Chemistry of Materials, 15(1), 175-178.
Tans, S. J., Devoret, M. H., Dai, H., Thess, A., Smalley, R. E., Georliga, L. J., & Dekker,
C. (1997). Individual single-wall carbon nanotubes as quantum wires. Nature 386
(6624), 474-477.(1997).
Tans, S. J., Verschueren, A. R., & Dekker, C. (1998). Room-temperature transistor based
on a single carbon nanotube. Nature, 393(6680), 49-52.
Tay, T. E. (2003). Characterization and analysis of delamination fracture in composites: an
overview of developments from 1990 to 2001. Applied Mechanics Reviews, 56(1),
1-32.
Teague, L. C., Banerjee, S., Wong, S. S., Richter, C. A., Varughese, B., & Batteas, J. D.
(2007). Effects of ozonolysis and subsequent growth of quantum dots on the
electrical properties of freestanding single-walled carbon nanotube films. Chemical
physics letters, 442(4), 354-359.
Terrones, M. (2013). Carbon nanotubes: synthesis and properties, electronic devices and
other emerging applications. International materials reviews.
Thostenson, E. T., & Chou, T. W. (2008). Real-time in situ sensing of damage evolution
Page 104
90
in advanced fiber composites using carbon nanotube networks. Nanotechnology,
19(21), 215713.
Tillack, G. R., Nockemann, C., & Bellon, C. (2000). X-ray modeling for industrial
applications. NDT & E International, 33(7), 481-488.
Treacy, M. J., Ebbesen, T. W., & Gibson, J. M. (1996). Exceptionally high Young's
modulus observed for individual carbon nanotubes.
Vemuru, S. M., Wahi, R., Nagarajaiah, S., & Ajayan, P. M. (2009). Strain sensing using a
multiwalled carbon nanotube film. The Journal of Strain Analysis for Engineering
Design, 44(7), 555-562.
Wang, J. T., & Raju, I. S. (1996). Strain energy release rate formulae for skin-stiffener
debond modeled with plate elements. Engineering Fracture Mechanics, 54(2), 211-
228.
Wang, J., Liu, G., & Jan, M. R. (2004). Ultrasensitive electrical biosensing of proteins and
DNA: carbon-nanotube derived amplification of the recognition and transduction
events. Journal of the American Chemical Society, 126(10), 3010-3011.
Wang, Z., Liang, Z., Wang, B., Zhang, C., & Kramer, L. (2004). Processing and property
investigation of single-walled carbon nanotube (SWNT) buckypaper/epoxy resin
matrix nanocomposites. Composites Part A: Applied Science and Manufacturing,
35(10), 1225-1232.
Warner, S. B. (1995). Fiber science, A division of Simon & Schuster, Inc.
Weinberg, H (2002), Available: http://www.sensorsmag.com/ articles/0202/36/
Westbrook, M. H., & Turner, J. D. (1994). Automotive sensors. IOP.
Wong, E. W., Sheehan, P. E., & Lieber, C. M. (1997). Nanobeam mechanics: elasticity,
strength, and toughness of nanorods and nanotubes. Science, 277(5334), 1971-
1975.
Wu, Q., Bao, J., Zhang, C., Liang, R., & Wang, B. (2011). The effect of thermal stability
of carbon nanotubes on the flame retardancy of epoxy and bismaleimide/carbon
fiber/buckypaper composites. Journal of thermal analysis and calorimetry, 103(1),
237-242.
Xie, D., Chung, J., Waas, A. M., Shahwan, K. W., Schroeder, J. A., Boeman, R. G., ... &
Klett, L. B. (2005). Failure analysis of adhesively bonded structures: from coupon
level data to structural level predictions and verification. International journal of
fracture, 134(3-4), 231-250.
Page 105
91
Xiong, J., Ma, L., Wu, L., Wang, B., & Vaziri, A. (2010). Fabrication and crushing
behavior of low density carbon fiber composite pyramidal truss structures.
Composite Structures, 92(11), 2695-2702.
Xu, G., Zhang, Q., Zhou, W., Huang, J., & Wei, F. (2008). The feasibility of producing
MWCNT paper and strong MWCNT film from VACNT array. Applied Physics A,
92(3), 531-539.
Xu, Z., Chen, X., Qu, X., Jia, J., & Dong, S. (2004). Single-wall carbon nanotube-based
voltammetric sensor and biosensor. Biosensors and Bioelectronics, 20(3), 579-584.
Yang, Z., Nakajima, M., Shen, Y., Wang, P., Ru, C., Zhang, Y., ... & Fukuda, T. (2013,
November). Test of A CNT gyroscope based on field emission. In The 7th IEEE
International Conference on Nano/Molecular Medicine and Engineering (pp. 59-
62). IEEE.
Ye, Q., & Chen, P. (2011). Prediction of the cohesive strength for numerically simulating
composite delamination via CZM-based FEM. Composites Part B: Engineering,
42(5), 1076-1083.
Yeh, C. S. (2007). A study of nanostructure and properties of mixed nanotube buckypaper
materials: fabrication, process modeling characterization, and property modeling.
ProQuest.
Ying, L.S. M.A.M. Salleh, H.B.M. Yusoff, S.B.A. Rashid, & J.B.A. Razak, J. (2013). Ind.
Eng. Chem. 17 367
Yoon, H. J., Yang, J. H., Zhou, Z., Yang, S. S., & Cheng, M. M. C. (2011). Carbon dioxide
gas sensor using a graphene sheet. Sensors and Actuators B: Chemical, 157(1), 310-
313.
Yousefi, J., Mohamadi, R., Saeedifar, M., Ahmadi, M., & Hosseini-Toudeshky, H. (2015).
Delamination characterization in composite laminates using acoustic emission
features, micro visualization and finite element modeling. Journal of Composite
Materials, 0021998315615691.
Yu, M. F., Files, B. S., Arepalli, S., & Ruoff, R. S. (2000). Tensile loading of ropes of
single wall carbon nanotubes and their mechanical properties. Physical review
letters, 84(24), 5552.
Yu, M. F., Lourie, O., Dyer, M. J., Moloni, K., Kelly, T. F., & Ruoff, R. S. (2000). Strength
and breaking mechanism of multiwalled carbon nanotubes under tensile load.
Science, 287(5453), 637-640.
Yu, M. F., Lourie, O., Dyer, M. J., Moloni, K., Kelly, T. F., & Ruoff, R. S. (2000). Strength
and breaking mechanism of multiwalled carbon nanotubes under tensile load.
Page 106
92
Science, 287(5453), 637-640.
Zhang, J., Jiang, D., & Peng, H. X. (2014). A pressurized filtration technique for fabricating
carbon nanotube buckypaper: structure, mechanical and conductive properties.
Microporous and Mesoporous Materials, 184, 127-133.
Zhang, Y., Yu, K., Xu, R., Jiang, D., Luo, L., & Zhu, Z. (2005). Quartz crystal
microbalance coated with carbon nanotube films used as humidity sensor. Sensors
and Actuators A: Physical, 120(1), 142-146.
Zhao, L., Gong, Y., Zhang, J., Chen, Y., & Fei, B. (2014). Simulation of delamination
growth in multidirectional laminates under mode I and mixed mode I/II loadings
using cohesive elements. Composite Structures, 116, 509-522.
Zhu, Z., Song, W., Burugapalli, K., Moussy, F., Li, Y. L., & Zhong, X. H. (2010). Nano-
yarn carbon nanotube fiber based enzymatic glucose biosensor. Nanotechnology,
21(16), 165501.
