AN EXPERIMENTAL PARAMETRIC STUDY ON THE 
 EFFICIENCY OF HYBRID FASTENING SYSTEM
 By Salina Ramli
    
      A DISSERTATION
 Submitted to
 Michigan State University
 in partial 
fulfillment of the requirements
 for the degree of
 Civil Engineering 
ΠDoctor of Philosophy
 2020  ABSTRACT
 AN EXPERIMENTAL PARAMETRIC STUDY ON THE EFFICIENCY OF 
 HYBRID FASTENING SYSTEM
 By Salina Ramli
 As the use of fiber
-reinforced polymer
 composites in ma
ss-produced structural components 
in all domains of industry has grown
, appropriate 
advancements in 
joining system
s are
 necessary. 
Bolted joints are a common method used due to the simplicity of the process. Drilling holes in 
composite
s for fastening results in delamination, creat
ion of 
locations for 
the 
onset of failure
, and 
can reduce load
-carrying capacities. 
Hybrid fastening techniques and 
other 
approaches for joining 
composite materials have been developed as a way to address proble
ms related to conventional 
bolted joints. 
One such technique is the h
ybrid fastening system
 in which a
 structural adhesive 
insert 
is placed in the bolt
-hole clearance
. This approach
 has 
been 
shown to eliminate bolt
-adherend slip, reduce delamination
, and i
ncrease load
-bearing capacit
ies
. Nevertheless, the 
extant 
work on such 
a hybrid fastening system is limited. 
 In this work, experimental characterization
s of hybrid fastening systems comprised of glass 
fiber reinforced polymer (GFRP) composite substrates f
astened using a fully threaded grade 5 steel 
bolts in ½fl
 (12.5 mm.)
 diameter with varying bolt hole clearance and 
three 
structural insert 
materials
 were performed. 
The GFRP substrates 
were 
manufactured using 
the 
vacuum
-assisted 
resin transfer molding (VART
M) process. The preload was maintained at 75% of the bolt yield 
strength for all joints. The 
objective of this work was to characterize/quantify the 
effect of
: a) 
the 
adhesive insert and 
b) 
the 
bolt
-hole clearances on the efficiency of the hybrid fastening
 system. 
For the first parameter,
 namely the effect of 
the 
adhesive insert, 
four configurations
, namely
 the
 hybrid fastening system with three different structural insert material
s and one 
control/conventional joint without any structural insert
 were studi
ed. Adhesive insert
 materials
 used 
were PRO
-SET epoxy resin, 
DEVCON epoxy carrying 
aluminum 
particles
, and 
polyurethane
. All resulting joints
 were
 cured at room temperature for 48h
 prior to testing
. For the 
second parameter,
 namely the effect of bolt clear
ance,
 four different bolt
-hole clearances: close fit 
(0.5mm), normal fit (1.0
 mm.), loose fit (1.5
 mm.), and extra loose fit (2.0
 mm.) were studied for 
each of the insert material
s along with the control/conventional joint without 
any 
structural insert
. The resulting joints were tested in a tensile
-shear configuration at a rate of 5 mm./min. 
 Hybrid joints were found to have 7 to 9 times higher load carrying capacities relative to 
slip
-loads for conventional joints. 
All hybrid fastening systems showed 
no bolt
-adherend slip along 
with delayed onset of delamination relative to conventional joints.
 Most of the joint
s with close fit 
(0.5mm) clearance were found to experience a catastrophic failure which resulted in bolt shearing 
failure. All other joints ex
perienced progressive delamination failure without any bolt
-shear 
failures. Further, t
he results indicate that the failure mechanism changed with the changes of bolt
-hole clearances. The larger the clearances, the 
more the 
bolt tilts/rot
ates and 
experience
s a combination of bending
 and
 shear. For small clearances, such 
as 
close
-fit, shear dominates and 
bolt
-shear occurs leading to bolt
 fracture. The combination of slightly larger clearance along with 
a structural adhesive insert allows tailoring the bolt
- joint performance, leading to 7
-9 times better 
performance than conventional joints with similar clearances. 
Future work should focus  on 

quantifying the stress
-concentrations and its reductions due to the addition of structural inserts. 
Overall, this work 
is novel and the first to report on the effect on clearance and varying adhesive 

insert materials for hybrid 
mechanical 
joint
s.     
 
 
 
   
 
 
 
 
 
 
 
   
 
 
 
Copyright by
 SALINA RAMLI
 2020 v   
 
 
 
   
 
 
 
 
 
 
 
   
 
 To beloved peoples who have meant and 
continue to mean so much to me.
 Nazrul Aizad, Naufal Syafi, Nia Sy
ifa, Abah & Mak
 Thank you for the prayers, sacrifices and supports
   vi ACKNOWLEDGEMENT
S  I wish to express my gratitude to Professor Dr. Mahmoodul Haq, 
faculty of Civil and 
Environmental 
Engineering, Michigan State University
 (MSU)
, for his constant guidance, 
encouragement, inspiration and support throughout this project. Professor Haq has been my role 
model in building my personality not only as a good researcher but also a better person 
in future. 

I am 
always greatly indebted him for his help, great support and guidance over these years.
 I wish to express my deepest gratitude to
 my committee members who were more than 
generous with their expertise and precious time. A special thanks to Dr
 Gary Cloud, Dr. Weiyi Lu 
and Dr. Nizar Lajnef 
for agreeing to serve on my committee.
 Not to forget, financial support provided by the Ministry of 
Higher 
Education Malaysia, 
and my employer Universiti Malaysia Pahang, Malaysia.  
Additionally, financial sup
port
 was also 
provided
 in the form of fellowship
 awards
 from the Department of Civil and Environmental 
Engineering
 at MSU
. All contributions are gratefully
 acknowledged.
 Help and support
, together with great suggestions and discussions by my colleagues Suh
ail 
Hyder Vattathurvalappil and Ben Swanson are gratefully acknowledged. Their presence and 

involvement have assisted me to develop and search the characteristics of this new 
hybrid fastening 

system
 less stressful. 
   vii
 TABLE OF CONTENTS
  LIST OF TABLES
––––––––––––––––––––––––––––––.
..ix
  LIST OF FIGURES–––––––––––––––––––––––––––––
–.x Chapter 1: Introduction
 ................................................................................................................... 1 1.1. Motivation
 ................................................................................................................................ 1 1.2. Objectives
 ................................................................................................................................ 4 1.3. Brief Literature Review
 ........................................................................................................... 5 1.3.1. Mechanisms in Mechanical Fastening
 ............................................................................... 8 1.3.2. Effect of Insert Material in Bolted Joint
 .......................................................................... 11 1.3.3. Effect of Tightening Torque Level in Mechanical Fastening System
 ............................. 15 1.3.4. Hybrid Fastening System
 ................................................................................................. 21  1.4. Scope
 ...................................................................................................................................... 22 Chapter 2: Materials and Manufacturing
 ...................................................................................... 24 2.1. Materials
 ................................................................................................................................ 24 2.2. Manufacturing
 of GFRP Substrate
......................................................................................... 24 2.3. Manufacturing of Hybrid Fastening System
 .......................................................................... 26 2.4. Project Approach and Parameters Studied
 ............................................................................. 29  
Chapter 3: 
Experimental Method & Approach
 ............................................................................. 33 3.1. Experimental Testing Parameters and Torque Calculations
 .................................................. 33 3.2. Experimental Test Setup
 ........................................................................................................ 36 3.3. Param
eter Definitions to Compare Joint Performance
 .......................................................... 36 3.3.1. Parameter 
Π1 (P
1): Comparing Onset of Delamination in Hybrid joints with 
Slip Load
 37 3.3.2. Parameter 
Π2 (P
2): Comparing Load at Maximum Slip Displacement in Hybrid Joints with 
Slip Loads in conventional 
Joint
 ................................................................................................ 38 3.3.3. Parameter 
Π3 (P
3): Comparing Peak Loads in Hybrid joints with Slip Load
 ................. 38  
Chapter 4: Effect of Varying Insert Material (Constant Clearance, 1.5 mm.) on the Efficiency of 

Hybrid Fastening System
 .............................................................................................................. 39 4.1. Introduction
 ............................................................................................................................ 39 4.2. Performance of a Conventional Bolted Joint
 ......................................................................... 39 4.3. Effect of Insert Material on Hybrid Joints with 1.5 mm bolt
-hole clearance(L
-N-#)............ 41 4.4. Chapter Conclusion
 ................................................................................................................ 44  
Chapter 5: Effect of Varying Bolt Hole
-Hole Clearance on the Efficiency of Hybrid Fastening 
System
 ........................................................................................................................................... 45 5.1. Effect of Bolt
-hole Clearance on Conventional Joints
 .......................................................... 45 5.2. Effect of Aluminum
-filled Epoxy Insert on Hybrid Joints with Varying Bolt
-Hole Clearances
....................................................................................................................................................... 50 5.3. Effect of Proset Epoxy Insert on Hybrid Joints of Varying Bolt
-Hole Clearances
 ............... 52 viii
 5.4. Effect of Polyurethane Insert on Hybrid Joints of Varying Bolt
-Hole Clearances
................ 55 5.5. Chapter Con
clusion
 ................................................................................................................ 59  Chapter 6: Summary, Conclusion and Future Directions
 ............................................................. 60 6.1. Summary
 ................................................................................................................................ 60 6.2. Key Conclusions
 .................................................................................................................... 61 6.3. Future Work
 ........................................................................................................................... 63  
APPENDIX
––...––––––––––––––––––––––––––
–––––––
64 REFERENCES
–––––
.–...–––––––
.––––––––––––––––––––
66   ix LIST OF TABLES
  Table 2
-1: Clearance hole chart for inch
-fasteners according to ASME B18.2.8. Also highlighted 
is the ½fl diameter bolt used in this work and associate bolt 
clearances.
 ..................................... 31  Table 2
-2: Specimen Nomenclature Used.
 ................................................................................... 32 Table 3
-1: Calculation of maximum torque for different bolt grades (grade 2 & grade 5).
 ......... 34 Table 5
-1: Bolt
-adherend Clearance Characteristics.
 ................................................................... 45 Table 5
-2: Reduction in joint stiffness of conventional bolted joint as 
a function of bolt
-hole 
clearance.
 ...................................................................................................................................... 49  
Table 5
-3: Reduction in joint stiffness as a function of bolt
-hole clearance.
 ............................... 55 Table 5
-4: Reduction in joint stiffness as a function of bolt
-hole clearance.
 ............................... 58    x LIST OF FIGURES
  Figure 1
-1: Bearing stresses with inserts[6][19].
 ............................................................................ 3 Figure 1
-2: Ligament stresses with inserts[6][19].
 ......................................................................... 4 Figure 1
-3:. Schematic of parameters studied in this work.
 ........................................................... 5 Figure 1
-4: Failure modes in composite bolted joints.
 ................................................................... 8 Figure 1
-5: Single shear joint.
 ......................................................................................................... 9 Figure 1
-6: Bearing stress in bolted joint.
 ..................................................................................... 10 Figure 1
-
xx
/Sb 
for different insert material along the tension plane for CFRP 
laminates bolted joint [24].
 ........................................................................................................... 11  Figure 1
-
laminates bolted joint [24].
 ........................................................................................................... 12  
Figure 1
-9: Normalized radial stress distribution for CFRP laminates bolted joint [24].
............. 12 Figure 1
-10: Load
Œdisplacement curves for; (a) conventional bolted joints (B1
-B3), (b) bolted 
joints with inserts (BI1
-BI3) [17].
 ................................................................................................ 13  
Figure 1
-11: Load
Œdisplacement curves for a conventional bolted joint and a joint with inserts 
[17].
 ............................................................................................................................................... 14  
Figure 1
-12: Details of micrographs of specimens with and without inserts loaded to 7.5 kN [17].
....................................................................................................................................................... 15  
Figure 1
-13: Load
Œdisplacement curves. Friction coefficient equal to 0.1. Influence of torque [27].
....................................................................................................................................................... 16  
Figure 1
-14: Load
Œdisplacement curves. Friction coefficient equal to 0.5. Influence of torque [27].
....................................................................................................................................................... 16  
Figure 1
-15: Stress vs local strain curves for finger tightened bolt [13][14].
 ............................... 17 Figure 1
-16: Stress vs local strain curves for 20N
m torque tightened bolt [13][14].
 ................... 18 Figure 1
-17: Load
-displacement diagram of bolted joint with different tightening torques [28].
 19 Figure 1
-18: Effect of tightening torque on bearing strength of bolted joint [28].
 ....................... 20 xi Figure 1
-19: Applied load and clamping force versus the axial extension for case (a): specimen 3 
with an applied torque of 0 N m (finger tight) (a); specimen 4 with an applied 
torque of 6.4 N m 
(b); specimen 8 with an applied torque of 12.88 N m (c).[29].
 .................................................... 21  Figure 1
-20. Joint configuration of hybrid 
fastening system for studying variations of bolt
-hole 
clearances and insert materials.
..................................................................................................... 23  
Figure 2
-1: (a) Composite (GFRP) manufact
uring using VARTM process (b) Schematic of 
VARTM setup.
 .............................................................................................................................. 25  
Figure 2
-2: Layup for VARTM process.
 ...................................................................................... 26 Figure 2
-3: Dimension of GFRP specimen.
 .................................................................................. 27 Figure 2
-4: Configuration of single lap shear bolted joint specimen.
 ........................................... 27 Figure 2
-5: 3D printed centering washer
 ...................................................................................... 28 Figure 2
-6: Top and bottom view of centering disc attached to washer.
 ...................................... 28 Figure 2
-7: (a) Drilling process of a small channel through the washer, (b) Sideview of drilled 
channel
 .......................................................................................................................................... 29  
Figure 2
-8: Adhesive injection process.
 ........................................................................................ 29 Figure 2
-9: Drilling process using ½fl diamond core drill.
 ........................................................... 30 Figure 2
-10:.Interlaminate delamination around bolt
-hole.
 .......................................................... 31 Figure 3
-1: Process of cutting the cross section of bolted joints/
 ................................................. 35 Figure 3
-2: (a) Schematic of Test setup and (b) actual test setup.
 ................................................ 35 Figure 3
-3: Schematic of lap
-shear behavior of conventional and hybrid 
joints.
 ......................... 36 Figure 4
-1: Cross section of bolted joints with 1.5 mm clearance.
 ............................................... 40 Figure 4
-2: Representative Load
-displacement response for conventional bolted joint (specimen 
L-N).
.............................................................................................................................................. 41  
Figure 4
-3: Post failure images of conventional bolted joints.
 ..................................................... 41 Figure 4
-4: Load
-displacement response for hybrid fastening system with various structural inserts 
(bolt
-clearance = 1.5 mm.)
 ............................................................................................................ 42  
Figure 4
-5: Effect of different insert material on constant clearance (1.5 mm).
 .......................... 43 xii
 Figure 4
-6: Post fai
lure images of hybrid fastening system (a) Proset epoxy (b) Aluminum epoxy 
(c) Poly
-Urethane.
 ......................................................................................................................... 44  Figure 5
-1: Load
-displacement respo
nse of conventional bolted joint (without insert) for varying 
bolt
-hole clearances.
 ..................................................................................................................... 46  Figure 5
-2: Effect of clearance on convention
al joints (no inserts).
 ............................................. 47 Figure 5
-3: Post failure images of conventional bolted joints at various clearances.
 ................... 48 Figure 5
-4: Stiffness changes pre
- and post
-slip in conventional bolted joints with varying bolt
-hole clearances.
 ............................................................................................................................. 48  Figure 5
-5: Schematic of pre
- and post
-slip deformation in conventional bolted joints.
 ............. 49 Figure 5
-6: Load
-displacement response of hybrid joints with aluminum
-filled epoxy insert and 
varying bolt
-hole clearances.
 ........................................................................................................ 50  
Figure 5
-7: Effect of clearance on hybrid joints with aluminum
- epoxy inserts.
 ......................... 51 Figure 5
-8: Post failure images of with aluminum epoxy inserts at various clearances.
 .............. 51 Figure 5
-9: Load
-displacement response of hybrid joints with Proset epoxy insert and varying bolt
-hole clearances.
 ............................................................................................................................. 52  
Figure 5
-10: Effect of bolt
-hole clearance on hybrid joints with Proset epoxy inserts
 ................ 53 Figu
re 5
-11: Post failure images of with Proset epoxy inserts at various clearances.
 .................. 54 Figure 5
-12: Variation in Initial stiffness of 
hybrid joints with PRO
-SET epoxy as a function of 
bolt
-hole clearance
 ........................................................................................................................ 54  
Figure 5
-13: Load
-displacement response of hybrid joints wi
th polyurethane epoxy insert and 
varying bolt
-hole clearances.
 ........................................................................................................ 56  
Figure 5
-14: Effect of clearance on hybrid joints with 
poly
-urethane epoxy inserts.
 ................... 57 Figure 5
-15: Post failure images of with Polyurethane epoxy inserts at various clearances.
 ....... 57 Figure 5
-16: Variation in Initial stiffness of hybrid joints with Polyurethane insert as a function of 
bolt
-hole clearance
 ........................................................................................................................ 58   1 Chapter 1:
 Introduction
 1.1. Motivation
 Joining has 
been identified as one of the driving technologies for innovation and 
sustainable manufacturing
 [1]
,[2]
. Messler
 [3]
 defines joining to be: ‚‚The process used to bring 
separate parts of components together to produce a unified whole assembly or structural entity™™. 
Poor design of structur
al joints can lead to noise, vibration, problems with load transfer
, localized 
stress concentrations
, fatigue, and catastrophic failure. 
 Thus, the choice and design of fastening 
systems 
 is 
 important in order to increase manufacturing  efficiency 
[1, 3]
 and to 
maintain
 structural integrity
.  There are 
a great many
 joining techniques
, and these may be divided 
into 3 broad classes, 
namely:
 mechanical fastening, adhesive bonding, and welding. 
Some joints 
are
 permanent and 
cannot be dis
-assembled for maintenance
 or 
repair
. Examples of 
such 
permanent joining include 
adhesive bonding and welding.
 Other
 fastening devices, includi
ng bolts and cap screws,
 facilitate 
disassembly
 for inspection,
 structural
 repair
, or 
component replacement
. Enhancing the 
performance of these 
mechanical fastening systems is
 the subject of this dissertation. 
 Bolts 
and cap screws
 are 
the most 
commonly 
used
 mechanical 
techniques
 to join 
or secure 
component
s due to 
the 
simplicity of 
the 
pro
cess
es involved, their established dependability, and 
their long history of evolution and refinement
 [4]
. One of
 the biggest advantage
s of bolted joints 
is the
ir ease of assembly and disassembly
 in factory and field
 using simple tools. Furthermore, 
extensive surface preparation is not required.
  However, 
mechanical fastening
 has some limitations. The use of fasteners in joining the 
parts 
can
 limit weight savings
. One 
example from the past is instructive
: 18,000 mechanica
l fasteners are used in
 the
 F-18 fighter jet airplane
, and the
 weight of the fasteners alone is 
2 approximately 500 l
bs 
[6]
. The fastener
s account for 1/3 of the total cost of a
 typical commercial
 airplane, the same as the engines 
[6]
. Drilling holes
 for bolts 
 will in
tro
duce st
ress concentration
s, 
which 
can lead
 to 
strength degradation and 
introduce 
corrosion related problems
[6].  There is a continuous demand in
 the
 automotive, aerospace and marine industries for 
joining 
parts made of 
similar and dissimilar 
materials
 [5,7]. Some of the objectives involved in 
dissimilar material joints are to enhance product design flexibility and to enhance the structural 
and functional efficienc
ies
. For efficient
 manufacturing 
of 
lightweight 
structures
, it is essential 
to 
understand the p
otential of different joining techniques 
[5]
. There are several joining techniques
 that are used 
to join
 metal and 
polymer
-based
 materials.
 The
se techniques
 can be used individually 
or combined to
 address the flaws or challenges involved in conventional joints.
 However, it should 
be noted that these 
emerging techniques have their advantage
s and disadvantages
, and the most 
appropriate method depends on the application and service requirements. 
 Several
 parameters need to be considered in mechanical fastening to ensure 
efficient 
performance of
 the joints. This include
 material thickness, fast
ener type
s (bolt, cap screw, rivet), 
material properties (e.g. composite to composite / composite to metal), hybrid joining (mechanical 

plus adhesives)
, preload clamping, loss of preload 
due to 
creep and relaxation, fastener size and 
scaling effects, 
bolt
-hole clearance
, interference fits, tapered threads, washers (flat, spring, 
conical), hole shape (non
-round holes, tapered holes, etc.), fastener shape (non
-cylindrical, 
tapered, etc.), hole treatment (coldworked, peened), insert and bushings, fastener 
arrays (shape and 

spacing), edge distance, torque (shear) effect upon tightening (especially for laminates), blind 

fasteners (one
-side access)
 and manufacturing methods 
[6]. Hybrid fastening techniques and approaches for joining dissimilar materials have been 
developed to address problems 
arising in more
 traditional fastening techniques. 
Earlier work on a 
3 novel hybrid fastening system, which introduces a structural resin inser
t injected through a channel 
in the bolt shaft as a structural element, overcomes the effects of drilling, eliminates slip, reduces 
delamination and increases the load
-carrying capacities 
[6]. To name but fou
r examples, 
Herrera 
and Cloud
[19]
, Camanho, et al. 
[8]
, Koricho et al. 
[9]
 , and Kelly 
[10]
 in their studies show
ed that 
an 
insert used in 
a composite
 joint was able to reduce the stress concentration surrounding the bolt 
holes
 and improve fatigue behavior
. The
 basic concept of 
a structural insert fil
ling
 the
 bolt 
clearance is 
examined
 in th
ese
 stud
ies
. This structural insert fills the delaminations
 int
roduced 
during the drilling process
 and act
s as a 
protecti
ve 
layer  to 
arrest  
premature failure or disaster 
[6,9]. Figure
s 1-1 and 
Figure 
1-2 from the work of Herrara and Cloud 
[19]
 and Cloud 
[6]
 show the 
stress concentrations around the bolt hole and the 
role of
 inserts
 in reducing them. Nevertheless, 
the efficiency of hybrid fastening systems on the effect of bolt
-hole clearance and varying adhesive 
inser
t material have not been studied or fully characterized. 
  Figure 
1-1: Bearing stresses with inserts
[6]
[19]
.  4 Therefore, the efficiency of 
a hybrid fastening system with varying structural insert 
materials
 and varying
 bolt
-hole clearance with similar substrates was 
investigated 
in this study. 
Structural insert
s will be introduced 
to create a
 hybrid fastening system. The insert 
materials 
that 
have been proposed are 
two
-part 
epoxy
 (same as the resin used to manufacture the substrates)
, poly
-urethane and 
epoxy
 with alumin
um reinforcement
.   Figure 
1-2: Ligament
 stresses with inserts
[6]
[19]
. 1.2. Objectives
  The goal of this 
work
 is to investigate the 
improvements in performance, if any, ga
ined by 
a mechanical
 hybrid fastening system
 that incorporates an injected resin insert
. The
 effect
s of 
varying insert material and bolt hole clearances
 on joint static strength 
are
 investigated
.  Uniform 
5 substrate
 material 
(a glass fiber reinforced p
lastic 
(GFRP)
) was used to reduce the complexity 
of 
the study
.  The basic concept of 
the 
injected hybrid fastener system using 
resins or resins containing 
adjuncts
 as structural insert
s has been described in papers and patents by Cloud [
[6]
, [11]
,[12]
]. The tasks to be executed towards 
the 
accomplishment of th
is work
 are listed below and are also 
shown schematically in
 Figure 
1-3.   Figure 
1-3:. Schematic of parameters stud
ied
 in this work
. The overall objectives of the work can be summarized as follows.
  To study the effect
 on the strength of a composite hybrid fastening system
 of varying insert 
material with constant bolt diameter and 
various
 insert material thickness
es,
  To study the effect
 on the strength of a composite hybrid fastening system
 of varying bolt 
hole clearance / insert material thickness with constant bolt diamet
er 
and 
with respect to 
different insert material
s, 1.3. Brief Literature Review
 Composite materials 
are
 being used increasingly 
 in 
several
 applications 
in aerospace, ship 
building, autom
otive, bridge construction and 
other
 engineering 
domains 
where light
 weight
, c
GFRP
GFRP
Clearance to be filled with 
structural insert
Bolt
Washer with 
drilled channel
Nut
Washer
   
     
      


Effect of bolt
-clearance ficfl
6 stiffness and strength propert
ies
 are
 of primary concern
[14]
[15]
[16]
. Mechanical fastening is one 
of 
the
 commonly used joining technique
s because of its
 manufacturing eas
e and 
the 
confidence
 in 
its performance acquired
 through decade
s of experience. 
Join
ts are
 a critical part of a structure
 and 
it is important 
to consider all the design parameters to ensure safety
 [14]
. As mentioned, one of 
the 
main advantage
s of 
mechanical fastening is the ease of assembly and
 disassembly
. However, 
drilling holes in composite laminates 
may result in
 larger 
stress concentration
s around
 the hole
s by virtue of 
the 
material™s high degree of anisotropy
 and drilling induced defects
. Furthermore, the 
strength of composite laminates
 will 
be 
diminished
 due to the presence of 
a fastener hole
 [17]
[18]
.  It is known that mechanical fasten
ings
 are more efficient in metallic structures than in 
composite
s. Th
is can be attributed to the ductility of metallic materials
. The expected stres
s concentrations that occur at bolt holes can be substantially relieved through
 localized
 plastic 
yielding of the metal in the vicinity of the holes. By contrast, in composite structures
, there is only 
limited stress relief through plastic yielding. Fractu
re therefore typically occurs at a lower load and 
at a lower ratio of joint
-strength
-to-adherend
-strength.
 The design of
 a mechanical fastening 
for a composite 
is more complex 
than is the case for
 metal
lic materials
. The effects on performance of s
everal p
arameters related to 
the 
composite 
material 
need to be considered
. Th
ese
 include
, among others,
 fiber and matrix type, lay
-up, bolt
-hole clearance, stacking sequence, bolt type, joint geometry and torque
 applied
 [7]
. In a typical 
joint manufactured using composite materials, the strength 
efficiency is 25 % lesser than 
that of 
its metallic counterparts
[17]
[19]
. This 
can be attributed to
 the
 high stress concentration
s around 
the hole
 and 
the 
damage induced
 in composite material
s during the 
drilling process
. As mentioned 
above, i
n contrast to composite materials,
 yielding in 
metallic material
s reduces 
local
ized
 high 
stresses and 
redistribute
s the stresses
[17]
.  7 One of the several challenges involved in
 composite joint manufacturing is the drilling 
process. It introduces stress concentrations and delamination 
that 
reduce the overall load carrying 
capability. Furthermore, t
hese defects will be
come
 the points of failure during the loading process. 
Hence, it is important to 
devise 
novel techniques 
to relieve the stress concentration
s and to heal
 or limit the defects
. One solution is to incorporate bonded
 or unbonded 
insert
s into the hole
, these 
inserts filling 
the gap
 between the bolt shank and the inner surface of the hole. 
Such inserts
 act as a protective 
layer which 
favorably modifies
 the local stress distribution and 
increase
s the strength of
 the
 joint
 [17]
. Herrera et al. 
[19]
 measured strains and stress concentration in the
 vicinity of fastener holes  
in FGRP that carried bonded and unbonded inserts in  the form of thin bushings of aluminum and 

resins. They found that stresses were reduced to varying degrees, depending on the insert material 

and the bonding. In one case, the
 stress concentration att
ained a negative value
 near the hole
, a startling result
.  In order to increase
 bolted joint efficiency, several researchers have proposed the use of 
metallic inserts, either interference fit or bonded in between 
the 
bolt and the c
omposite laminate. 
Metallic insert
s ha
ve shown 
by many investigators to yield 
remarkable benefits
 in reduc
ing
 stress 
concentration
s around laminate hole
s [6]
[8]
[9]
[10]
[17]
[19]
[20]
[21]
[22]
[23]
[24]
. The insert
 helps 
to redistribute
 the stresses and strains around the hole
, particularly
 if the adhesive bond does not 
fail
. In all cases, 
the efficiency of bolted joints with inserts
 showed better performance when
 compared with conventional bolted joints
, and the increase in strength is reported to vary from
 10% to 100%
 [20]
[22]
. 8 1.3.1. Mechanism
s in Mechanical Fastening
 Drilling hole
s in 
a composite laminate creates stress concentr
ation
s, thus 
the 
overall load 
bearing capability of the laminate is severely reduced. The five common failure modes in 
a mechanical fastening system
, illustrated in Figure 1
-4, are tension, shear, cleavage, bearing and 
pull
-through
. Net tension failures oc
cur when 
the 
bolt diameter is 
an overly
 large fraction of
 the
 strip width. 
Acceptable
 fraction
s are
 depend
ent
 on the 
type of material and 
the 
lay
-up used. Bearing 
failures 
govern
 when
 the
 bolt diameter is 
a small fraction of plate width. This type of
 failure lead 
to elongation of 
the 
bolt hole. Shear failure can be classified as special kind of bearing failure. This 
failure can occur 
even with a
 very large end distance for highly orthotropic laminates. Cleavage 
failure
s are associated with both inadequ
ate end distance and too few transverse plies. Pull
-through 
failure
s can
 occur 
when countersunk fasteners are used and when 
significant bending loads are 
imposed
.  Figure 
1-4: Failure modes in composite bol
ted joints
. Figure 
1-5 shows 
the schematic of a 
single shear lap joint. The key dimensions are;
 d; 
fastener diameter
 t; thickness of the joint elements
 w; width of the plate
 e; edge distance
 Tension
Cleavage
Shear
Bearing
Pull
-through
9 P; load
 Dt; bearing area
 loaded in compression
 2et
; total shear
-out area loaded in shear
 (w Πd )t; net section
 area
 loaded in tension
 wt; gross section 
area 
loaded in tension
  Figure 
1-5: Single shear joint
. The 
important 
stresses in bolt
ed join
ts are categorized into bearing, shear
-out, net and gross 
stress.
 The
se stresses are calculated 
using
 the 
equations below;
 Shear
-out stress, 
= Net 
normal 
stress, 
=() Gross 
normal 
stress, 
= Bearing stress, 
= Bearing stress
 is the average
 compressive stress that
 is produced at 
the 
surface of contact 
between 
the bolt and the substrate
 when they
 are
 fastened together. It is developed due to the 
load
, P which ultimately affect
s the hole in the plate as shown 
in 
Figure 
1-6. Bearing stress 
is 
also known 
dweBearing 
section 
dtNet 
section 
(w-d)t
Gross 
section 
wtShear
-out 
section 
2et
tpp10 as crushing stress. Bearing failures are characterized by localized damage, 
such as delamination 
and matrix crazing around the hole. 
These
 failures occur by the fastener causing localized 
compression loading leading to buck
ling and kinking of the fibers followed by crushing of the 
matrix.
 The equation 
for bearing stress 
considers
 the 
stress
 not evenly distributed into the hole 
diameter.
 The 
stress distribution
 is more 
like 
an 
elliptical 
shape
, and varies drastically with hole 
clearance
  Figure 
1-6: Bearing
 stress in bolted joint
. For most applications, 
bolt
s are 
commonly 
manufactured 
of metal, typically steel
. When 
the bolt is tightened, it will stretch and behave like a tension spring
, causing
 a tensile 
stress in the 
bolt and a 
corresponding compressive 
clamping force in the struc
ture
. The behavior of the joint 
depends 
on how tightly the bolts clamp and how long they can maintain their preload
. If this stretch 
exceeds the elastic limit
 or if the joint is overloaded
, the bolt yields 
plastically, 
acquiring
 a permanent set. The result
 can be
 a loss in preload or clamp
ing force
.  The bolt itself can fail in one, two, or a combination of modes. If the loading is in pure 
shear fit is tight, th
en the bolt will exhibit shear failure. If bending loads are present or if the fit is 
loose so as
 to allow tilting of the bolt, then the failure might be in bending even in tension if the 
deformations are extreme.
 Bearing stress
PP11 1.3.2. Effect 
of Insert 
Material 
in Bolted Joint
 Filling of the clearance between bolt and hole is carried out through a small hole in the 
head 
or shank 
of 
a bolt or
 of a
 washer. After injection 
of the insert material, it is allowed to cure 
at room temperature or respective curing conditions.
 Koricho et al. 
[9]
[35]
 and 
Haq et al.
[36]
 stud
ied
 joint
s of GFRG to GFRP with constant bol
t diameter of 0.5fl. They use
d SC15 epoxy as 
the adhesive insert to fill the bolt
-hole clearance. Their stud
y showed
 that the j
oints with 
resin
 structural inserts 
eliminate slippage of
 the joints, and they were
 found to perform 
2-6 times
 better 
than
 convent
ional joints
.  Camanho et al. 
[24]
 in their study use
d adhesively bonded metallic insert
s to increase 
the 
efficiency of composite bolted joint
s. The ef
fect 
on joint strength 
of two different insert materials
, Aluminum 2000 seri
es and mild steel
, were determined. The results from their study
, reproduced 
in 
Figure
s 1-7, through 
Figure 
1-9, show
ed that the insert decrease
d the maximum stresses in 
the 
tension and bearing planes
. The stresses acting on 
the 
laminate surrounding 
the 
holes we
re redistributed. However, for
 the 
different
 configuration
s investigated, 
the 
adhesive failed before 
the 
full strength of 
the 
laminate 
was reached
. They recommended 
the 
use of thinner and more 
compliant insert
s of aluminum 
rather than
 steel. 
  Figure 
1-7: Distribu

xx/Sb for different insert material along the tension plane for CFRP laminates 
bolted joint 
[24]
.  12  Figure 
1-8: Distribution
 
laminates bolted joint 
[24]
.   Figure 
1-9: Normalized radial
 stress distribution for CFRP laminates bolted joint 
[24]
. Mara et al. 
[17]
  studied the effect of inserts on  bolt
-tension relaxation, the stiffness
, and
 the
 load bearing behavior of composite bolted joints.
 In their study, they 
used metal insert
s and 
compared the re
sults to conventional bolted joints. 
The s
tandard steel bolts used in their study were
 M10, class 8.8 (i.e. yield strength of 640 MPa and ultimate strength of 800 MPa). A clearance of 
1mm was chosen for the joint tests based on the common range of clearanc
es used in steel bridges 
for bolt diameters lower than 14 mm. As a result, a hole of 11mm was drilled in the 
joints having 
13 no inserts
, whereas, in the joints with inserts, the hole was 16mm in diameter in order to fit the 
insert.
  The results of their stud
y are shown in 
Figure 
1-10. Three (3) specimen 
were 
tested for 
each case
 and the joints were only finger
-tightened
. Slip due to the clearances 
was observ
ed. The 
difference in the amount of slip between the tests is due to the position of the bolt in relation to the 
hole, which was not controlled during the assembly of the joints. As can be seen, for 
a conventional 
bolted joint on
 the
 B1 specimen 
shown in 
Figure 
1-10 (a), slip occurred immediately after test
 initiation,
 whereas
 for 
specimen 
B2, the
 slip 
happened
 after some load 
was 
applied
 to the joint
. After slip, 
the 
load
-deflection
 curve 
increased
 linear
ly up until point A
 was reached
. The small 
drop of load at point A 
indicate
d damage initiation of 
the 
FRP laminate
. Further loading reduced 
the slope of the curve owing to the progressive damage within t
he composite substrate.
   (a)
 (b)
  Figure 
1-10: Load
Œdisplacement curves for; (a) conventional bolted joints (B1
-B3), (b) bolted joints with 
inserts (BI1
-BI3) 
[17]
. The load
Œdisplacement behavior of bolted joints with inserts (BI) can be seen as almost 
linear up to point A in 
Figure 
1-10 (b). The load at point A indicates the initiation of bearing 
damage in
 the
 laminate. 
Again, slip occurred because of the clearances. 
The joint
s with insert
s 14 manage
d to 
maintain joint stiffness until bearing damage was 
initiated
, and this occurred in all 
cases at a load higher than that corresponding to damage initiation in the joints having no inserts.
 The insert
 evidently provides
 lateral constraints, preventing the delamination
 of the FRP laminates 
and through
-thickne
ss expansion around the hole in the insert area. The damage in the
 laminate is 
therefore delayed until it splays outside the diameter of the insert lap, where an expansion of the
 -thickness of the laminate occurs.
 Figure 
1-11 shows the comparison of the load
Œdisplacement curves 
between
 a conventional 
bolted joint and a bolted joint with 
a metal insert. The initial slip in the tests was ignored for the
 sake of comparison. Th
is study show
ed that the load at initiation of damage for 
a bolted joint with
 an insert
 is more than twice 
that of
 conventional bolted joints. The initial stiffness is similar for 
both
 the
 joints
. However,
 the stiffness 
was significa
ntly reduced 
in the conventional bolted joints 
after the initial bearing damage. It should be 
noted
 that the bearing area of the joints with inserts 
are
 higher due to a larger hole diameter of 16 mm, compared with the conventional bolted joints 
which
 had a hole diameter of 11 mm. The maximum load that can be supported by the bolted joints 
with inserts is therefore expected to be higher.
  Figure 
1-11: Load
Œdisplacement curves for a conventional 
bolted 
joint 
and a joint with
 inserts 
[17]
. 15 Figure 
1-12 shows micro
photo
graphs of
 specimens with and without insert
s[21]
. In th
is 
study, bonded metal was use
d as structural insert for the bolted joint system. Damage can be seen 
in the
 laminate 
of 
the 
joint which was manufactured 
without
 an insert
, whereas
 no visible damage 
was evident
 on the 
laminate when insert
s were
 used. 
  Figure 
1-12: Details of
 micrographs of specimens with and without inserts loaded to 7.5 kN 
[17]
. 1.3.3. Effect of 
Tightening 
Torque 
Level in 
Mechanical
 Fastening 
System
 The 
magnitude
 of the
 torque
 applied 
to the bolt
 when tightening the joint
 direct
ly influence
s the stress fields in both the bolt and the plates. An
 increment in the 
increase of
 torque has 
major 
effects on
 the joint performance. The positive effect is the 
increase
 of friction
al forces
 between the 
substrates
 that lead to 
an increas
e in
 bearing strength
 and slip load
 as proved in
 [17]
[24]
. The
 negative effect is 
due to the 
out-of-plane 
crushing 
stresses
 that are developed when the torque is 
applied
. This can lead to the premature fail
ure of the joint.
 However, when joining laminates, some 
increase of this clamping load can prevent delamination. For conventional joining
 the maximum
 value of torque is limited by standards.
  Olmedo et
 al.
[27]
 studied the influence of friction coefficient and tightening torque for 
single
-lap bolted joints
 using
 computational models
. The model 
was
 initially 
validated through 
comparison with experimental result
s from lit
erature. The result
s showed that 
the joint strength 
16 increased with increase in
 friction coefficient
. For low friction coefficient
s, the 
load to produce 
joint 
displacement
 is very low
. In this case, 
joint strength
 cannot be improved with torque
 pre
load
. Figure
s 1-13 and 
Figure 
1-14 show the effect 
of 
friction coeffici
ent
 on joint strength
. The effect of 
torque level is depend
ent
 on friction coefficient. 
It can be 
clearly seen that for 0.5 friction 
coefficient, i
ncre
ment
s of tightening 
torque increase
d the 
joint stiffness
.    Figure 
1-13: Load
Œdisplacement
 curves. Friction coefficient equal to 0.1. Influence of torque 
[27]
.   Figure 
1-14: Load
Œdisplacement curves. Friction coefficient equal to 0.5. Influence of torque 
[27]
. 17 Girard et al.
 [13]
[14]
 studied the effect of clamping pressure on bearing failure of 
mechanically fastened carbon/epoxy 
laminated composite joints. T
hree different clamping 
methods were used. The first series of tests
 were
 done without any clamping pressure, using a 
dowel pin. The second series w
ere
 carried
 out with a finger
-tightened bolt, and the third 
series
 were
 carried
 out using a torque
-tightened
 bolt to exert a clamping pressure of at least 20 MPa.
 Figure 
1-15 and 
 1-16 show the stress vs local strain obtained 
by Girard et al.
 [13]
[14]
 for different 
clamping pressure
s. The
 bearing stress 
was defined as the load divided 
by the hole diameter 
times
 thickness of the
 laminate. 
Results from their studies show
ed that stiffness of 
the 
transversal
 surface 
is not influenced by clamping pressure. Nonetheless, the maximum stress and strain increase
d with 
clamping pressure.
   Figure 
1-15: Stress vs local
 strain curves for finger tightened bolt 
[13]
[14]
. 18  Figure 
1-16: Stress vs local
 strain curves for 20Nm torque tightened bolt 
[13]
[14]
. Khashaba et al
.[28]
 stud
ied
 the effects of torque on the strength of bolted joint
s in 
joints 
manufactured using 
glass fiber reinforced 
plastics
 (GFR
P) composite lamin
ates with 5.2 ± 0.1mm 
thickness.
  In their study
, they 
determined the
 strength of bolted joints with various torque
s (T = 
0, 5, 10, and 15 Nm) and washer sizes (outer diameter of washers, Dw = 14, 18, 22, and 27 mm).
 Representative results are shown in Fig
ure 1
-18.
 The experimental results show
ed that 
for a 
constant
 torque, the slope of 
the 
load
Œdisplacement 
curve
 (stiffness) increase
d with decreas
e in
 washer size. Bolted joint
s with 18 mm washer size and 15 Nm tightening torque 
manifested
 the 
maximum
 stren
gth.
 19  Figure 
1-17: Load
-displacement
 diagram of bolted joint with different tightening torque
s [28]. It 
was evident
 that
 for a constant bolt 
and washer 
diameter
s, the joint stiffness increased 
with increase in torque.
  The curve
 of the finger tight specimen
 (T = 0 Nm
) showed
 the lowest
 slope (stiffness) with several knees
 and nonlinear curve
s, an
 indication of
 uneven development of 
internal damage.
 Another study showed that t
he bearing strength also increased with torq
ue 
level 
as shown in Figure 1
-18. 20  Figure 
1-18: Effect of tightening torque on bearing
 strength of bolted joint 
[28]
. Tong 
[29]
 also s
tudied 
the effect of  torque 
(finger tight, 6.4Nm and 12.88Nm
) on carbon/epoxy composite bolted joint
s. His results, shown in
 Figure 
1-18, sho
w that 
the 
displacement
 increased linear
ly with the 
load
 until
 noticeable drop
 is seen
. This 
behavior 
indicate
s failure
s such as delamination or micro buckling. Similar to other studies, the stiffness 
of the 
joint
 increase
d with
 increase of torque level. It 
was clear that the initial failure loads increase
d as the 
applied
 torque 
was increased from finger tight to 12.88 N
-m. This 
was because 
of 
the 
relatively 
small 
unconstrained gap with a
 maximum radial clearance of 0.65 mm
. Thus,
 the join
t can still be
 significantly affected by the lateral constraint
s caused by
 the
 clamping forces applied
. After initial 
failure, the 
load 
Πdisplacement curve
s grew
 linear
ly with additional applied load 
but 
have a lower 
slope, meaning less stiffness
.   21    Figure 
1-19: Applied load and clamping force versus the axial extension for case (a): 
specimen 3 with an 
applied torque of 0 N m (finger tight) (a); specimen 4 with an applied torque of 6.4 N m (b); specimen 8 
with an applied torque of 12.88 N m (c).
[29]
. 1.3.4. Hybrid 
Fastening System
 One of the 
important
 parameters affecting the stiffness of bolted joints is the presence of 
clearances, which are inevitable to facilitate on
-site assembly.
 Clearances reduce the stiffness and 
the 
load to initial 
damage of join
ts. Further it
 induce
s slip between the connected members
, which 
can lead to misalignment of parts as well as the introduction of impact loads
. Slip due to clearances 
can also cause a reduction in the fatigue life of composite bolted joints, as demonstrate
d in the
 previous
ly cited
 literature
[18]
. In bolted joints manufactured using composite plates, the pre
-tension applied on the bolts
 may
 relax due to the viscoelastic properties of the composites.  
The 
use of
 high
-strength friction g
rips (HSFG) cannot be relied on 
for 
long term performance
. HSFG 
bolts transfer the connection force between contact surfaces by friction alone. Force transmitted in 
22 this way cannot be relied 
on, as was
 shown in a
 preliminary study by Mottram 
[31]
. The study
 showed that the
 bolts los
t their tightness over time due to FRP relaxation.
 An alternative to
 the
 close
-fitted or HSFG 
technique
s is 
to inject 
resi
n or resin carrying 
adjuncts such as metal particles
 into the bolt hole clearance cavity
, creating what
 is known as 
a hybrid fastening system
. Hybrid fastening system
s with resin injected into
 the
 bolt
-hole clearance 
offers various advantages
 including resistance 
to fatigue and shock loading
, resistan
ce to
 corrosion
, achieve
ment of
 acceptable slip and fatigue performance
, and low manufacturing costs
. As mentioned above, t
he basic concept of 
the 
injected hybrid fastener system using 
resins, 
adhes
ives, or liquid carriers containing adjuncts to form
 structural insert
s has been described in 
papers and patents by Cloud [6,
[11]
[12]
] In 
some other 
hybrid joining techniques,
 insert
-like shims 
are
 used to fill the c
learance 
holes between the bolt and substrate. Hühne et al
.[34]
 in their work on
 the
 effect
s of liquid shims 
on the 
behavior of carbon fiber laminate bolted joint sho
wed a decreas
e in stiffness 
when the
 substrate thickness
 was reduced.
 Haq et al
 [36]
 in their work show
ed that use of 
a resin insert with various reinforcements 
such
 as a
 carbon sleeve,
 a glass sleeve
, or a
 nano
particle
 filler will reduce slip
 and
 increase
 the
 load capacity
 and
 strength of composite joint
s while
 delay
ing
 delamination 
around
 the 
bolt 
holes.
 1.4. Scope
 In this 
report
, single
-lap joint strength 
using 
uniform
 substrate 
materials 
were
 evaluated 
experimentally for a novel hybrid join
ing technique
 that introduces a structural adhesive insert in 
the clearance 
between
 the bolt and the 
substrate
s. The effect 
of varying
 insert material and varying 
bolt hole clearance
 on the strength
 and performance
 of 
the 
resulting 
hybrid joint
s were
 studied
. 23 Glass fiber reinforced polymer (GFRP) composites w
ere
 used as the substrates
. Bolt
s having a 
nominal
 diameter of ½fl w
ere
 used 
in all the cases
  The test matrix incorporated 
3 diff
erent insert materials and 4 different bolt radial 
clearances.
 For comparison, joint specimens with standard bolted connections
 having no insert and
 having the same clearance
s were tested. 
Figure 1
-20 illustrates the joint configurations. 
Quantitative 
details of specimen design as well as the clearance and material parameters studied 
are given in Chapter 2 of this report.
 The peak loads, displacement at peak loads, loads at initial delaminations
, and slip loads 
were compared to 
assess
 the performance of
 the 
hybrid bolted joints. This study was intended to 
provide a benchmark for the better design of bolted joints using 
this 
novel hybrid joining technique.
  Figure 
1-20. Joint configuration
 of hybrid fastening system for 
studying 
variation
s of bolt
-hole
 clearances
 and insert materials.
    Clearance to be filled 
with structural insert
Variation of thickness for 
each insert material
a)Fit
tight
b)0.5 mm
c)1.0 mm
d)1.5 
mme)2.0 mm

GFRP
GFRP
Bolt
Washer with 
drilled channel
NutWasher
cccc24 Chapter 2:
 Materials
 and
 Manufacturing 
 2.1. Materials
 The joints were made of 16
-layer S
-2 plain weave glass fiber reinforced plastic (GFRP) 
substrates, with an 
area
-specific weight of 832 g/m
2. These substrates were bonded using PRO
-SET INF
-114 infusion epoxy and PRO
-SET INF
-210 fast infusion hardener with a mix ratio of 
3.65:1 by weight. 
Threaded hex bolts having13 threads per inch and a bolt diameter of 0.5 inc
h were used in the experiment. These bolts were made of
 medium strength 
grade 5 steel
. Furthermore, zinc
-plated medium strength grade 5 steel hex nut
s and zinc yellow
-chrome plated 
grade 8 steel washers having  0.531fl internal diameter and 1.062fl outer dia
meter were chosen for 
the study. A total of three structural adhesive inserts (injected into bolt
-hole clearances) namely 
PRO
-SET INF
-114 infusion epoxy, polyurethane from Scotch
-Weld
, and DEVCON aluminum
-filled
 resin
 were used in this study. 
 2.2. Manufacturin
g of 
GFRP Substrate
 Large plates of 16
-layer GFRP were manufactured using 
Vacuum Assisted Resin Transfer 
Molding 
(VARTM) and then cut into substrates. A detailed description of the manufacturing 
process has been provided in 
Figure 
2-1 (process diagram showing the experimental layout
 and 
layup sequence
) and 
Figure
 2-2. Initially, 16 layers of glass fabric were laid on the mold. 
Permeable ply was then placed over the glass fabric on both sides followed by the application of 
transfer media which is used to en
hance the resin flow and ensure easy separation of layers to 
create 
a satisfactory surface finish. The resin was infused using three inlets, positioned such that it 
wetted the entire fabric. T
he mo
ld was finally
 sealed
 using a vacuum bag
. Tacky tape was us
ed on 
the edges of the vacuum bag to ensure there 
was
 no leakage. 
The impregnated plate was cured at 
room temperature
 for 24 hours. The
 resulting thickness of the cured GFRP plate was 3/8fl (9.53 
25 mm). After room temperature curing, the plate was further cur
ed for 8 hours at 82
oC inside a 
convection oven. The fiber volume fraction of GFRP was 0.55 as shown by the calculations below.
 

=6979 
=5084 
=


=1895 
=

=50842.49=2042 

=



 =1.23
 
=11+

 =11+1.23=0.45 =1

 =10.45 = 0.55
   (a)
 (b)
 Figure 
2-1: (a) Composite (GFRP) manufacturing using VARTM process (b) Schematic of 
VARTM setup
.  Transfer media 
16 layers
glass fiber
Impermeable ply
Vacum
bag
Inlet
Resin
Resin
Resin
Tacky tape
Mol
dSpiral tubing
Permeable ply
26  Figure
 2-2: Layup
 for VARTM process
. 2.3. Manufacturing of Hybrid 
Fastening System
 The dimensions of the joints 
were designed
 as a function of
 bolt diameter
 (DB) used
 for the 
experiment
. Figure
 2-3 shows the dimensions of 
the 
GFRP substrates. As can be seen from the 
figure, the e
dge distance (E
1) and end distance (E
2) were designated as multiples of the bolt 
diameter (2.5D
B and 4D
B respe
ctively). 
The specimen geometry was in conformation to the 
minimum requirements of bolt geometries specified in 
the 
ASCE pre
-standard for pultruded 
(building frame) structures.
 The experimental program using the single lap shear bolted connection 
configura
tion 
is 
shown in
 Figure 
2-4. To facilitate an equal thickness of the adhesive insert around the bolt, a special polylactic 
centering disc was made
 as shown in 
Figure 
2-5. The centering disc, which was manufactured 
using 3D printing, was glued to the washer to ensure central positioning of the bolt a
nd an equal 
clearance around its shaft (Please see 
Figure 
2-6). Once all joints were put together and tightened 
by hand, they were subjected to a preloa
d torque of 38 ft
.lb as calculated in Chapter 
3. The adhesive 
insert
 was injected by drilling a 
channel
 through the washer
 as shown in 
Figure 
2-7 (a). 
The drilling 
process to form the channel was very intricate and involved multiple drill sizes.
 27  Figure
 2-3: Dimension
 of GFRP specimen
.  Figure 
2-4: Configuration of
 single lap shear bolted joint specimen
. Figure 
2-7 (b)
 shows 
a side view of 
the 
drilled channel through the washer. 
This channel 
was formed by using  

" and 
 
" drill bit
s at an angle towards the bolt.
 At the entry point for the 
resin, the channel was drilled with 
the
 larger bit. The size of this drill bit was chosen so that the 
hole would create an airtight seal wi
th the head of the syringe
 used to inject the resin
. This 
prevented air bubbles from entering the cavity and
 assured
 a uniform distribution of 
adhesive
 throughout. Prior to injecting the resin, a similar channel was drilled on the other side of the bolt 
E1=1.25fl
E2=2fl
L=5.75fl
DB=0.5fl
W=2.5fl
E1=1.25fl
E2=2fl
L=5.75fl
DB=0.5fl
W=2.5fl
t=0.375fl
1.6mm
28 head as an escape passage for air and excess 
adhesive
 once the cavity was completely full. This 
permitted 
an insertion
 for the 
adhesive
 as it was pushed through the channel by 
using
 a syringe
 as 
shown in 
Figure 
2-8. Once the 
adhesive
 was
 injected into the hybrid joints, they 
were cured for 48 
hours at room temperature.
   (a)
 (b)
 Figure
 2-5: 3D printed 
centering washer
    (a)
 (b)
 Figure 
2-6: Top
 and bottom view of centering disc attached to washer
.  29   (a)
 (b)
 Figure 
2-7: (a) Drilling
 process of a small channel through the washer, (b) Sideview of drilled channel
     (a)
 (b)
 Figure
 2-8: Adhesive injection process
. 2.4. Project Approach and Parameters Studied
 Experiments
 were carried out using 
multiple adhesives 
(detailed in section 2.1
) as structural 
insert
s as detailed in 
Table 
2-1 below. In addition to that, specimens
 without any structural inserts 
30 (conventional bolted joint
s) were also tested
 to develop a baseline. For each configuration, three 
specimens were prepared.
   Figure 
2-9: Drilling
 process using ½fl diamond core drill.
 Four different bolt
-hole clearances were used in this study, in accordance with 
ASME 
B18.2.8.
 The bolt
-hole clearance for fit
-class close, normal and loose were set to be around 0.5mm, 
1.0mm and 1.6mm respectively for the ½fl diameter bolt
. Specimens with an fiextra
-loosefl fit were 
added to the experimental schedule as mentioned below. 
For 
the 
fit t
ight bolted joint, the holes in 
the 
composite plates were cut using a ½fl diamond core drill (
Figure 
2-9). For other fit
-class
es, holes were cut using a 
½fl diamond core drill first, and then
 the clearance 
was created by using 
cobalt steel drill bit
s of sizes 9/16fl, 17/32fl, 39/64fl and 21/32fl respectively 
to drill through 
the 
substrate. This procedure was performed to avoid severe delamination of
 the
 composi
te during 
drilling. 
Due to the drilling process, 
initial delamination was 
observed
 along the interior 
of the 
bolt 
hole
 as shown in 
Figure 
2-10 (a). This delamination occurs due to the localized pushing out of the 
substrate layers in the vicinity of the drill bit
. To ensure a smooth hole in the substrate, some of 
the delaminated porti
on was removed using a cutter as shown in 
Figure 
2-10(b).
 The corresponding 
specimen numbering is shown i
n Table 
2-2. 31   (a)
 (b)
 Figure 
2-10:.Interlaminate
 delamination around bolt
-hole
.  Table 
2-1: Clearance 
hole chart for inch
-fasteners according to ASME B18.2.8
. Also highlighted is the 
½fl diameter bolt used in this work and associate bolt clearances
. Clearance Holes for Inch Fasteners
 Nominal 
Screw 
Size
 Fit Class 
- Normal
 Fit Class 
- Close
 Fit Class 
- Loose
 Nominal 
Drill 
Size
 Hole 
Diameter
 Nominal 
Drill 
Size
 Hole 
Diameter
 Nominal 
Drill Size
 Hole Diameter
 Min.
 Max.
 Min.
 Max.
 Min.
 Max.
 #0 #48 0.076 0.082 #51 0.067 0.071 3/32
 0.094 0.104 #1 #43 0.089 0.095 #46 0.081 0.085 #37 0.104 0.114 #2 #38 0.102 0.108 3/32
 0.094 0.098 #32 0.116 0.126 #3 #32 0.116 0.122 #36 0.106 0.110 #30 0.128 0.140 #4 #30 0.128 0.135 #31 0.120 0.124 #27 0.144 0.156 #5 5/32
 0.156 0.163 9/64
 0.141 0.146 11/64
 0.172 0.184 #6 #18 0.170 0.177 #23 0.154 0.159 #13 0.185 0.197 #8 #9 0.196 0.203 #15 0.180 0.185 #3 0.213 0.225 #10 #2 0.221 0.228 #5 0.206 0.211 B 0.238 0.250 1/4
 9/32
 0.281 0.290 17/64
 0.266 0.272 19/64
 0.297 0.311 5/16
 11/32
 0.344 0.354 21/64
 0.328 0.334 23/64
 0.359 0.373 3/8
 13/32
 0.406 0.416 25/64
 0.391 0.397 27/64
 0.422 0.438 7/16
 15/32
 0.469 0.479 29/64
 0.453 0.460 31/64
 0.484 0.500 1/2
 9/16
 0.562 0.572 17/32
 0.531 0.538 39/64
 0.609 0.625 5/8
 11/16
 0.688 0.698 21/32
 0.656 0.663 47/64
 0.734 0.754 3/4
 13/16
 0.812 0.824 25/32
 0.781 0.789 29/32
 0.906 0.926 7/8
 15/16
 0.938 0.950 29/32
 0.906 0.914 1-1/32
 1.031 1.051 1 1-3/32
 1.094 1.106 1-1/32
 1.031 1.039 1-5/32
 1.156 1.181 1-1/8
 1-7/32
 1.219 1.235 1-5/32
 1.156 1.164 1-5/16
 1.312 1.337 1-1/4
 1-11/32
 1.344 1.360 1-9/32
 1.281 1.291 1-7/16
 1.438 1.463 1-3/8
 1-1/2
 1.500 1.516 1-7/16
 1.438 1.448 1-39/64
 1.609 1.634 1-1/2
 1-5/8
 1.625 1.641 1-9/16
 1.562 1.572 1-47/64
 1.734 1.759   32 Table 
2-2: Specimen
 Nomenclature Used
. Adhesive Insert
 Material 
  Clearance Size 
 Proset
 Epoxy
 Polyurethane
 Aluminum 
Liquid
 Conventional 
(without 
insert)
 Fit Class 
- Close
 C ΠP - 1 C ΠPU 
- 1 C ΠAL 
- 1 C ΠN - 1 C ΠP - 2 C ΠPU 
- 2 C ΠAL 
- 2 C ΠN - 2 C ΠP - 3 C ΠPU 
- 3 C ΠAL 
- 3 C ΠN - 3 Fit Class 
- Normal
 N ΠP - 1 N ΠPU 
- 1 N ΠAL 
- 1 N ΠN - 1 N ΠP - 2 N ΠPU 
- 2 N ΠAL 
- 2 N ΠN - 2 N ΠP - 3 N ΠPU 
- 3 N ΠAL 
- 3 N ΠN - 3 Fit Class 
- Loose
 L ΠP - 1 L ΠPU 
- 1 L ΠAL 
- 1 L ΠN - 1 L ΠP - 2 L ΠPU 
- 2 L ΠAL 
- 2 L ΠN - 2 L ΠP - 3 L ΠPU 
- 3 L ΠAL 
- 3 L ΠN - 3 L ΠP - 4 L ΠPU 
- 4 L ΠAL 
- 4 L ΠN - 4 Fit Class 
ΠExtra 
Loose
 XL 
ΠP - 1 XL 
ΠPU 
- 1 XL 
ΠAL 
- 1 XL 
ΠN - 1 XL 
ΠP - 2 XL 
ΠPU 
- 2 XL 
ΠAL 
- 2 XL 
ΠN - 2 XL 
ΠP - 3 XL 
ΠPU 
- 3 XL 
ΠAL 
- 3 XL 
ΠN - 3 Table 2
- 2 enumerates all the joint samples in this work for varying clearances and insert 
material type. The nomenclature used is provided in equation (1) as follows:
 A ΠB - #       (1)
 wherein ‚A™ rep
resents the clearance type, ‚B™ represents the insert material type and ‚#™represents 
the specimen number.  As expla
ined in chapter 2, four clearances were studied
, and hence 
parameter A in equation 1 has four values
 namely 
(i) close
-fit
 (C)
, (ii
) normal
-fit (N)
, (iii
) loose
e-fit
 (L)
 and 
(iv
) extra
-loose
 fit
 (XL)
. Similarly the parameter B has four values corresponding to 
the insert material type, namely: (i) Proset epoxy (P), (ii) Polyurethane (PU), (iii) Aluminum filled 
epoxy (AL), and (iv) normal/convent
ional joint without an insert (N).  
The next chapters 
focus
 on the experimental testing and results from the characterization of all the joint specimens enumerated 

in 
Table 2
- 2.  33 Chapter 3:
 Experimental Method & Approach
 This chapter discusses the experimental setup, torque applied to the bolts and associated 
calculations, and the definition of parameters to compare the results of the 
various case studies 
performed in this work.
 3.1. Experimental Testing Parameter
s and Torque Calculations
 The final phase in the manufacturing process of conventional joints is the application of 
torque load. 
 As mentioned, the clamping load in a joint that is 
developed by the bolting torque is 
critical to its performance. Clamping force was not selected as a variable parameter in this study, 

so a single value of bolting torque was selected on the basis of standard design practice. 
The 
bolting 
torque was
 calcula
ted using the following equation and procedure
.  =
       Equation 
(1) where,
 T = Torque (ft.lbs)
 K = Coefficient of friction 
 
P = Desired clamp load tension (lbs)
 D = Nominal diameter (inches)
 In this study, grade 5 bolt
s of diameter ½™™ and coarse threa
d of 
13tpi were used
. The torque 
load
 was calculated using
 equation 1 
in accordance with the
 design tables provided by the bolt 
manufacturer 
[26]
. This torque was further reduced by 33% as the joints were tested in shear.
  Table 2
- 1 shows the comparison of the torque calculation for grade 2 and grade 5 bolts based on 
the design table and Equation 1.
 
  
 34 Table 
3-1: Calculation of maximum torque for different bolt grades 
(grade 2 & grade 5)
.  GRADE 2
 GRADE 5
 Design 
Table
 From table provided by manufacturer 
 T = 49 ft.lbs
  Adjustments to torque setting based on 

thread treatment, for cadmium plating 

reduce 25%
 T = 36.75lbs
  
(reduced torque 33% for bolt in shear);
 T = 25 
ft.lbs
  From table provided by manufacturer 
 T = 75 ft.lbs
  
Adjustments to torque setting based on 

thread treatment, for cadmium plating 

reduce 25%
 T = 56.25lbs
  
(reduced torque 33% for bolt in shear);
 T = 38 ft.lbs
  Calculation
 A=0.1419in S=55000psi
 P=0.1419x0
.75x55000
=5853
.375lb
  T=KPD
 =0.20 x 5853
.375lb
 x 0.5inch
12inch
x ft =49ft
.lb  (Adjustments to torque setting based on 

thread treatment, for cadmium plating 

reduce 25%
 T=36.75ft
.lb  (reduced torque 33% for bolt in shear);
 T=25ft
.lb   A=0.1419in S=85000psi
 P=0.1419x0
.75x85000
=9046
.125lb
  T=KPD
 =0.20 x 9046
.125lb
 x 0.5inch
12inch
x ft =75ft
.lb  (Adjustments to torque setting based on 

thread treatment, for cadmium plating 

reduce 25%
 T = 56.25lbs
  
(reduced torque 33% for bolt in shear);
 T = 38 ft.lbs
   Based on the
 thread treatment,
 the 
torque setting 
was
 reduce
d by
 25%
, and the m
aximum 
torque 
considered
 in this study
 was 38 ft.lbs.
 Prior to injecting
 the
 resin
, the bolts were preloaded to 
the torque level
 as outlined above
. After res
ting injection, a
ll the specimens were then cured for 48 hours before being tested. 
Samples 
‚L Œ P - 2, L Œ PU - 2, L Œ AL 
- 2 and 
L Œ N Œ 2™ were selected to be cut using 
a metal cutting 
bandsaw as shown 
in Figure 3
- 1. This was done to observe the cross
-sectional view 
showing the 
bolt and clearances filled with the adhesive.
 This was done to ensure that the assembly process 
35 was not introducing any air
-bubbles or voids in the clearance cavity. The assembled joints were 
then 
quasi
-statically test
ed in tensile
-shear configuration
 as described in the next section
  Figure 
3-1: Process
 of cutting the cross section o
f bolted joints
/    (a)
 (b)
 Figure 
3-2: (a) Schematic of Test setup and (b) actual test setup
.  PPLaser 
tabs to measure 
relative displacement 
between substrates
36 3.2. Experimental
 Test Setup
 Blocks
 of similar 
material and 
dimensions as the substrates 
were attached to 
the 
assembled 
joints 
in 
the grip
-locations 
where 
the 
joints are clamped in the MTS and 
loads are applied. Thi
s ensures that the 
joint 
is tested 
in pure
 shear configuration. Th
e testing was performed using
 a uniaxial testing machine (MTS 810) with a load cell capacity of 100 kN
. A 
loading rate of 5 
mm/min was maintained for all joints
, and
 they
 were tested 
until failure. A
n external
 laser 
extensometer was used to measure the relative displacements between t
he substrates in addition to 
the MTS crosshead displacement.
 The schematic 
of testing 
and actual test setup 
is shown in 
Figure 
3- 2. 3.3. Parameter Definitions to 
Compare Jo
int Performance
 The performance of the hybrid fastening system was 
compared to
 conventional joints
 with 
similar clearances
. A schematic
 of typical
 load 
- displacement 
response
s of conventional and 
hybrid joints 
is provided
 in Figure 3
- 3.      Figure 
3-3: Schematic of 
lap
-shear behavior of 
conventional and hybrid joints
.  PP37 From 
Figure 3
- 3, it is obvious that the conventional joint experiences a bolt
-adherend slip 
characterized by a sudden increase in displacement without any significant increase in load. In 
applications 
wherein
 small displacements are important, this bolt
-adherend slip can be considered 
as the failure of the resulting joint. The load corresponding to this bolt
-adherend slip is termed as 
the ‚Slip
-load.™ 
The maximum displacement occurring due to this bolt
-adheren
d slip prior to joint 
re-align
ment
 to carry load is termed as the fimaximum slip displacement.fl 
It should be noted that 
all the hybrid joints tested in this work did not experience any bolt
-adherend slip, thereby the term 
fislip loadfl is exclusively represen
tative of conventional bolted joints. In hybrid joints, 
the first 
sign of failure is in the form of a small drop in load corresponding to onset/start of delamination. 

The load corresponding to this 
point is 
termed as the fiOnset of Delamination Load (ODL).fl
 It 
should be noted that the onset of delamination 
also occurs in conventional joints after the 

occurrence of bolt
-adherend slip.
 Finally, the maximum load experienced by the joint is termed as 
the fiPeak load (PL).fl
 Since ‚slip load™ in conventional joints
 is considered the ‚failure criterion,™ three parameters 
were framed to compare the performance of hybrid joints. The definitions of these three parameters 

are provided as follows. 
 3.3.1. Parameter 
Π1 (P1): Comparing Onset of Delamination 
in Hybrid joints with 
Slip Load
 Parameter 
Π1 (P
1) is defined as the ratio of 
‚the load at onset of delamination in hybrid 
joints
™ to 
‚the slip load of conventional joints with equivalent clearance
™ and is represented in 
equation (2).
  =
_
_

_

_
=
_
   Eq. (2)
 The load corresponding to 
P1 is represented in 
Figure 3
- 3 with the letter 
‚a™. 
P1 quantif
ies
 the performance of hybrid joints relative to slip loads of conventional joints. In other words, 
(P1 = 38 1) indicates that the slip load and the load at onset of delamination of hybrid joints are the same. 
Similarly, 
(P1>1) will indicate 
that 
the hybrid joints hav
e onset of delamination loads higher than 
the slip load. Lastly 
(P1<1) will indicate the hybrid joints having onset of delamination at loads 
lower than the occurrence of slip loads. 
 3.3.2. Parameter 
Π2 (P
2): Comparing Lo
ad at Maximum Slip Displacement in Hybrid 
Joints with 
Slip Load
s in conventional Joint
 Parameter 
Π2 (P
2) is defined as the ratio of the ‚load in hybrid joints corresponding to the 
maximum slip displacement in 
equivalent 
conventional joints™ to ‚the slip l
oad of conventional 
joints with equivalent clearance
,™ and is represented by equation (3).
 =
_
_
_

_
_
_

_
=
@
_
__

  Eq. (3)
 The load corresponding to P
2 is represented in 
Figure 3
- 3 with a letter ‚b™. In applicatio
ns where 
displacement is critical, P
2 provides the load carrying capacity of hybrid joints at displacements 
corresponding to the maximum slip in conventional joints. This parameter P
2 is thus not valid for 
conventional joints and only valid for comparing h
ybrid joints with equivalent conventional joints.
 P2 will always be greater than 1.
 3.3.3. Parameter 
Π3 (P
3): Comparing Peak Loads in Hybrid joints with Slip Load
 Parameter 
Π3 (P
3) is defined as the ratio of the fip
eak load 
in hybrid jointsfl to the fi
slip 
load in
 conventional joints of equivalent clearance
fl and is represented in by equation 4.
 =
_

_
_
_

_
=
_
_

_   Eq. 
(4)
 The load 
corresponding to P
3 is represented in 
Figure 3
- 3 with a letter ‚c™.
 P3 will 
always be greater 
than 1, as the peak loads are higher than the slip loads. 
The next two ch
apters cover the 
experimental results corresponding to the varying insert material
s and hole clearances.
  39 Chapter 4:
 Effect of Varying Insert Material 
(Constant Clearance
, 1.5 mm.)
 on 
the Efficiency 
of Hybrid Fastening System
 4.1. Introduction
 The work on hybrid fa
stened joints similar to the 
that
 performed 
in this thesis,
 wherein an 
adhesive structural insert is used to 
fill the cavity/gap between the bolt and the adherend is limited 
to the work performed in the research group of Prof. Haq and Prof. Cloud. 
In an 
earlier work
 [35]
, an SC
-15 epoxy was used as an adhesive structural insert in a hybrid joint consisting of 
glass fiber 
reinforced p
lastic
 (GFRP) substrate
s, ½fl diame
ter grade 5 bolt, and a constant clearance of ~1.5 
mm. The insert material and the clearances were kept constant in this work 
[35]
. It is prudent to 
compare our stud
y on the same clearance (1.5 mm) 
with the 
earlier work 
[35]
. Hence, this chapter 
compares the effect of insert materials for a constant clearance of 1.5 mm
 (loose
-fit
). Figure 4 
- 1shows the cross section of 
loose-fit (
1.5 mm
) hole clearance bolted joints for 
both the 
conventional and
 the 
hybrid fastening system for each 
of the 
adhesive insert
s used
 in this 
study
. It 
can
 be observed from the figure that the cavity between bolt shank and hole 
was 
completely
 filled
 by the 
insert material
. Air
-voids or improper filling was not observed indicating 
a good manufacturing process.
 4.2. Perfor
mance of a Conventional Bolted Joint
 Figure 
4-2 shows
 the 
representative 
load
-displacement 
response of conventional bolted 
joint with l
oose-fit (1.5 mm) clearance. As e
xplained in section 3.3, the joint is tested in shear 
loading. When the compressive forces and associated friction forces due to the application of the 

torque are overcome by the applied shear load, a bolt
-adherend slip is observed. The corresponding 
avera
ge slip load for these samples was ~8 kN. In the slip event, the load does not significantly 
increase, but the displacement increases until the bolt tilts and is resting/bearing on the substrates
. 40   (a)
 Conventional joint
 (b)
 Hybrid joint with PRO
-SET epoxy
   (c)
 Hybrid joint with aluminum liquid
 (d)
 Hybrid joint with polyurethane
 Figure 
4-1: Cross section of bolted joint
s with 1.5 mm clearance
. Further increase in applied load results in onset of delamination in 
the 
GFRP substrates at 
the locations of contact of 
the 
bolts with 
the 
substrates. Delamination and fiber breakage a
t bolt 
substrate contact surfaces are represented by a small drop 
in the 
load displacement curves. This is 
followed by a reduction in stiffness u
ntil the peak load is reached. After the slip
-load, the bolt
-shank undergoes a combination of bending and shear loads. As the load increases the delamination 

increases until the peak load is reached. Beyond the peak 
load, 
the extent of delamination increas
es 
with increase in displacement and reduction in load carrying capacity. 
 41    Figure 
4-2: Representative Load
-displacement response for conventional bolted joint (specimen L
-N). Figure 4
-3 shows the 
post 
failure 
images of conventional bolted joints. Severe delamination 
was
 observed 
in the substrate around the bolt
-hole along with plastic deformations in the washers
. This can be attributed to the bearing load introduced by the bolt on the hole surface. The 
dominant 
mode of final failure was 
bolt 
pull
-through in all the 
samples of this case (L
-N-#).      Figure 
4-3: Post failure images of conventional bolted joints
.  4.3. Effect of Insert Material on 
Hybrid 
Joints
 with 1.5 mm bolt
-hole clearance
(L-N-#)  Figure 4
-4 shows
 the 
representative
 load
-displacement 
response
 of 
each of the 
hybrid joints
 corresponding to the 
three 
different structural insert
 material. For comparison, the response of the 

conventional joint without any insert material
 is also included. 
 PP42    Figure 
4-4: Load
-displacement response for hybrid fastening 
system with various structural inserts 
(bolt
-clearance = 1.5 mm.)
  As explained in section 3.3,
 bolt
-adherend 
slip was not observed in any of the hybrid joints 
tested in this case study and is shown in 
Figure 4
-4. However, for joints with polyurethane (ductile, 
low
-stiffness) insert, a reduction in stiffness was observed in the loading region equivalent to 
the 
slip load in conventional joint. This shows that the insert material is activated after the 
slip 
load 
and acts as a structural element to transfer the load from the substrate to the bolt and 
vice
-versa
, The onset of delamination loads of joints with SC
-15, Aluminum
-filled epoxy and Proset epoxy 
are approximately 5.5 times better than the slip loads of the conventional joints. The joints with 
polyurethane inserts showed an improvement of 5 times better properties than conventional joints.
  Figure 4
-5 compares the
 parameters P
1, P2 and P
3 (defined in section 3.3) for the three 
insert materials in this work and the SC
-15 epoxy from earlier work 
[35]
. P1, P2 and P
3 represent 
the delamination onset load, the load corresponding to the maximum slip displacement and the 
peak load relative to the slip load of conventional joints (section 3.3). 
It is evident that the hybrid 
joints with all the four inserts outperformed the conventional joints. 
 PP43  Figure 
4-5: Effect of different insert material on constant clearance (1.5
 mm). P2 (load corresponding 
to the maximum slip displacement in equivalent control
) parameter 
showed an
 average
 improvement of 
~8 times that 
compared to that of the slip
-loads of conventional 
joints.
 Interestingly, the parameter P
2 was statistically constant for all the insert materi
als in this 
work. This indicates that while the bolt
-adherend slip is eliminated by the presence of adhesive 
insert, the stress
-transfer from the bolt to 
the 
substrate
s is efficient
 for all insert materials
. Hence, 
the loads corresponding to onset of delam
ination are the same for all the hybrid joints.
 Hybrid joints with SC
-15 as structural insert showed the highest P
3 (representation of peak 
loads) values
 of all joints tested in this case study. This can be attributed to change not only in the 
insert material but also in the substrate. In the earlier work, 
the 
substrates were also made of SC
-15 along with the insert material. SC
-15 is a two
-part epoxy with
 an additional toughening agent
 and
 thereby has higher performance compared to the 
inserts
 used in this work. 
 Figure 4
-6 shows the 
extent of
 failure i
n hybrid joint
s with 
different
 adhesive insert
 material
s. Compared to the conventional bolted joints 
(see
 Figure 4
-3), the extent of damage in joints with 
structural inserts was relat
ively minimal and bolt pull
-throug
h was not obse
rved. 44    (a)
 (b)
 (c)
 Figure 
4-6: Post f
ailure 
images
 of hybrid fastening system
 (a) Proset epoxy (b) Aluminum epoxy (c) Poly
-Urethane
. 4.4. Chapter 
Conclusion
 In this 
chapter, the effect of insert material was studied for hybrid joints with a constant 
clearance of 1.5 mm (loose
-fit). Bolt
-adherend slip was eliminated in all 
the 
hybrid joints.
 Depending on the insert material, the performance of the hybrid joints
 could be tailored. 
Polyurethane insert joints had lower stiffness but higher ductility.
 On average, the performance 
(P1, P2, P3 combined) of all hybrid joints was 
~ 4-8 times higher/better than the slip loads of the 
equivalent conventional bolted joint. A
dditionally, the extent of damage in hybrid joints was less 
severe relative to the conventional joints. 
Similar comparison
s for 
the 
effect
s of insert material can 
be made for other constant clearances in this study. For brevity, those are not included and 
instead, 
the effect of clearance on 
the 
efficiency of hybrid joints is discussed in detail in the next chapter. 
 
  45 Chapter 5:
 Effect of Varying Bolt Hole
-Hole Clearance on 
the Efficiency of 
Hybrid Fastening System
 This chapter showcases the results of the experimenta
l characterization of the hybrid bolted 
joints with varying bolt clearances. As described in section 
2.4, four different bolt
-hole clearances 
were studied for each o
f the insert material
s. These bolt
-hole 
clearances were 
characterized
 as close fit (
C, 
0.5mm), normal fit (
N, 
1.0mm), loose fit (
L, 
1.5mm) and extra loose fit
-class (
XL-2.0mm).
 The 
aforementioned bolt
-clearances are shown in 
Table 
5-1. Table 
5-1: Bolt
-adherend Clearance Characteristics
. Clearance 
Type / 
 Fit
-Class
 Designation/ 
Symbol used
 Clearance size 
(mm.)
 Close 
 C 0.5 Normal 
 N 1.0 Loose 
 L 1.5 Extra Loose 
 XL 2.0 In the following sections, the effect of bolt hole clearance on conventional joints and each 
of the insert material
s are presented and discussed.
 5.1. Effect of Bolt
-hole Clearance on C
onventional Joints
 Figure 5
-1 shows the 
representative 
load
-displacement 
responses
 of the conventional 
bolted joint
s corresponding to each of the bolt
-hole clearance studied int his work. 
The initial 
segment in the load
-displacement curve 
reveals
 a linear
-elastic
 behavior up to 
~10kN for all joints
. At this point, a bolt
-substrate slip occurs in all joints. As explained in the previous chapter, the 
compressive force holding the joints toget
her is provided by the torque/clamping load applied on 
the joint. When the sliding forces (applied) loads are greater than this compressive force, the bolt
-adherend slip occurs. Since the torque was maintained a constant for all joints in this work, the sl
ip 
loads remain approximately constant for all joints. But, the maximum slip displacement is 
46 controlled by the bolt
-hole clearance. 
The higher the clearance values, the more the slip
 displacement
. This maximum slip distance is the distance at which the bol
t comes into contact 
(bearing) 
onto the 
substrates to an extent that it can transfer 
the 
applied loads. 
At the end of 
the 
slip phase, the 
value of the 
load 
increase
s linearly until the onset of delamination in
 the
 composite 
substrates. Th
ese
 are progressive 
delamination
™s, and 
they 
can 
fur
ther 
reduce the
 stiffness. 
As the 
applied load increases, the peak load is achieved
, after which a
 rapid increase in 
delamination/failure with 
a continuous drop in load
 is observed
.     Figure 
5-1: Load
-displacement response of conventional bolted joint (without insert)
 for varying bolt
-hole clearances
. Figure 5
-2 shows the 
comparison of
 parameters 
P1 (onset 
of 
delamination) and P
3 (Peak 
load)
 as a function of bolt
-hole clearance for conventional joints. It should be noted that parameter 
P2 compares the load in hybrid joints at the maximum slip loads to its corresponding slip load in 
conventional joint. Hence this parameter P
2 does not apply for conventional bolted joints and 
is 
not shown in Figure 5
-2. For parameter P
3, it is evident that t
he best properties were obtained f
or joints with 
normal (1
 mm) 
and l
oose
 (1.5 mm) clearances
. This suggests that there is a need for an 
optimal clearance in GFRP substrates to not only efficiently transfer the loads but also to mitigate 
PP47 the progression of 
failure. For instance, in extra l
oose clearances, upon bolt
-adherend slip there 
will be a considerable tilt/rotation of the bolt along with locations of high
-stress concentration 
where the bolt is resting/bearing on the GFRP substrate. This will lead to ea
rly onset of 
delamination (P1) and low peak loads (P3), and was observed in this case as shown in 
Figure 5
-2.  Figure 
5-2: Effect of clearance on conventional joints (no inserts)
. Figure 5
-3 shows the post
-failure images of conventional bolted joints 
having
 different 
bolt
-hole clearances.
 It is clear that the mode of failure changes with respect to bolt
-hole clearance. 
When the clearance is small, there is not enough space for the bolt to tilt or rotate
, and 
it 
experiences pure shear leading to bolt
 fracture. When the clearance increases, 
the bolt tilts/rotates 
and experiences a combination of bending and shear along with stress
-concentrations at the bearing 
points. Hence the delamination continues at the bearing points until the bolt is pulled
 out of the 
joint.
  48     (a)
 (b)
 (c)
 (d)
 Close fit (0.5mm),
 Normal fit (1.0 mm)
 Loose fit (1.5mm)
 Extra loose fit (2 mm)
 Figure 
5-3: Post failure images of conventional bolted joints
 at various clearances.
    (a)
 0.5mm clearance (Close fit)
 (b)
 1.0mm 
clearance (Normal fit)
    (c)
 1.5mm clearance (Loose fit)
 (d)
 2.0 mm clearance (Extra loose fit)
 Figure 
5-4: Stiffness 
changes pre
- and post
-slip in 
conventional bolted joint
s with varying
 bolt
-hole 
clearances
. The effect of bolt
-hole clearance is not only evident at the ultimate failure point, but also 
in the entire load
-displacement response. First, the extent of slip displacement increases with 
increasing bolt
-substrate clearance. Second, the bolt comes into c
ontact with the substrates after 
49 the slip and re
-engages to carry the loads. The second stiffness after the slip event is also different 
for every bolt
-hole clearance, and is indicated by slope 2 in 
Figure 5
- 4. As the bolt
-clearance increases, the bolt
-tilts and experiences a combination of bending 
and shear
 (see 
Figure 5
- 5). The higher the dominance of bending, the 
larger the reduction in the
 slope of the load
-displacement curve.
 Also, an
 increase in bolt
-hole clearance means smaller 
contact surface
 for bearing
, thereby higher stress leading to early failures.
  Figure 
5-5: Schematic of pre
- and post
-slip deformation in
 conventional bolted joint
s. As 
observed
 in 
Figure 5
-4, a reduction in slope (stiffness) due to increasing bolt
-hole 
clearance is clearly visible. The joint stiffness (
marked as slope 2
) was calculated
 to be 
8.0 kN/mm, 
8.0 kN/mm, 7.2
 kN/mm and 7
.0 kN/mm 
for 
close, normal
-, loose
-, and extra loose
- fit clearance
s. Table 
4-2 show the change in 
joint 
stiffness 
(after the slip, slope 2) 
as the clearance
 is varied fr
om 
close fit
 (0.5mm
) to 
extra loose fit
 (2.0mm
). Additionally, the decrease in slope 
of all joints 
relative 
to the close
-fit has also been highlighted
. Table 
5-2: Reduction in joint stiffness 
of conventional bolted joint 
as a function of bolt
-hole clearance
.  Clearance
  Close
 Normal
 Loose
 Extra loose
 Stiffness (kN/mm)
 8 8 7.2 7 Percentage change from close to extra 
loose fit clearance
 - 0% -10%
 -12.5%
     
50 5.2. Effect of Aluminum
-filled Epoxy
 Insert on Hybrid Joints 
with 
Varying Bolt
-Hole 
Clearances
 Figure 5
-6 shows the 
representative 
load
-displacement 
responses
 of 
hybrid
 joint
s with 
aluminum
-filled epoxy 
insert
 for each of the bolt
-hole 
clearances.
 The initial segment in the load
-displacement curve 
was
 linear
ly elastic u
p to 
~ 10kN
. This is the region wherein the compressive 
forces (clamping loads) are governing the behavior of the joint. Hence this region is the same for 
all joints irrespective of the insert. 
As explained in section
 3.3 and section 
4.3
, hybrid joints do not 
exhibit bolt
-adherend slip
. However, a change in stiffness was visible 
around
 the 
slip loa
d (~10 
kN)
, indicating transfer of stresses from the substrate
-to-insert
-to-bolt and 
vice
 versa
. Hence, 
around the slip
-load the compressive forces are overcome by the applied shear forces and force 
transfer occurs through the insert to the substra
te as indicated by the change in slope. 
Further 
application of load exhibited a non
-linear response 
until the onset 
of 
delamination 
was reached. 
Once delamination starts, a progressive failure starts
, and is
 characterized by a large increase in 
displacemen
t with minimal increase in load until peak load is reached. Beyond the peak loads, the 
displacements increase while the load carrying capacity reduces until failure. 
    Figure 
5-6: Load
-displacement 
response of hybrid joints with aluminum
-filled epoxy insert and varying 
bolt
-hole clearances
. PP51 Figure 5
-7 shows the effect of clearances 
on P1 (onset of delamination), 
P2 (load based on 
maximum slip displacement) and P
3 (Peak load) as a function of slip load in conventional joint
s for the hybrid joints 
with ‚
aluminum
-filled
 epoxy 
inserts
.™ It 
was observed 
that the best 
performance in all parameters was 
obtained from 
loos
e (1.5 mm) clearance
 and the worst 
performance of all parameters was in the close
-fit clearance
. Overall, on average (P
1, P2 and P
3 combined) the hybrid joints with 
the 
Aluminum filled epoxy insert
s were ~3
-8 times better than 
the slip load of equivalent c
learance conventional bolted joint
s.  Figure 
5-7: Effect of clearance on hybrid joints with aluminum
- epoxy
 inserts
.       (a)
 (b)
 (c)
 (d)
 Close fit (0.5mm)
 Normal fit (1mm),
 Loose fit (1.5mm)
 Extra 
loose fit (2 mm)
 Figure 
5-8: Post failure images of 
with aluminum epoxy inserts at various clearances
. 52 Figure 5
-8 shows the post
-failure images of hybrid joints with aluminum
-filled 
epoxy as 
a structural insert in the bolt
-hole clearances. Similar to that of conventional joints, the mode of 
failure change
s with respect to bolt
-hole clearance. 
If the clearance is small, the failure happens 
by bolt shear. When the clearance increases, the joints failed by pull through.
    Figure 
5-9: Load
-displacement 
response of hybrid joints with Proset epoxy insert and varying bolt
-hole clearances
. 5.3. Effect of Proset
 Epoxy
 Insert on Hybrid Joints of Varying Bolt
-Hole Clearances
 Figure 5
-9 shows the representative load
-displacement responses of hybrid joints with 
Proset
 epoxy insert for each of the bolt
-hole clearances.
 Similar to hybrid joints with aluminum
-filled epoxy inserts
 (section 5.2)
, the 
initial segment in the load
-displacement curve 
was
 linear
 elastic u
p to 
~ 10kN. This is the region wherein the compressive forces (clamping loads) are 
governing the behavior of the joint. Hence this region is the same for all joints irr
espective of the 
insert. The elimination of slip and the transfer of loads explained in 
the 
previous section for 
the 
‚aluminum
-filled epoxy inserts™ joint is also valid for this joint
, and for brevity purposes not 
repeated here. One observation that was un
ique to the Proset epoxy insert joints was a significant 
drop in loads at ~30 kN. This was observed for 
Proset 
joints with all clearances. This drop is 
PP53 attributed to the brittle failure of 
the 
Proset epoxy insert. Further increase in applied load revealed 
onset and progressive delamination of the substrate until 
a peak load was reached
, followed by 
failure similar to other joints in this work.
  Figure 
5-10: Effect of bolt
-hole clearance on hybrid joints with 
Proset epoxy inserts
 Figure 5
- 10 shows the effect of clearances on P
1 (onset of delamination), P
2 (load based 
on maximum slip displacement) and P
3 (Peak load) as a function of slip load in conventional joint
s for the hybrid joints with ‚Proset epoxy inserts™. Similar to that of aluminum filled epoxy inserts 
(section 5.2), it was observed that the best performance in all parameters was obtained from 
loose 

(1.5 mm) clearance and the worst performance of all parameters was in the close
-fit clearance. 
Overall, on average (P
1, P2 and P
3 combined) the hybrid joints with Proset epoxy insert were ~3
-8 times better than the slip load of equivalent clearance c
onventional bolted joint
s.  Figure 5
-11 shows the post failure images of hybrid joints with 
Proset epoxy as structural 
insert in the bolt
-hole clearances. Similar to that of conventional joints, the mode of failure changes 
with resp
ect to bolt
-hole clearance. 
For close
-fit clearance
, bolt
-shear failure occurs
, and for all 
other clearances the joints fail with excessive delamination and bolt pull
-through. 
 54      (a)
 (b)
 (c)
 (d)
 Close fit (0.5mm)
 Normal fit (1mm),
 Loose fit 
(1.5mm)
 Extra loose fit (2 mm)
 Figure 
5-11: Post failure images of with 
Proset epoxy inserts at various clearances
.    (a)
 0.5 mm clearance (Close fit)
 (b) 1.0
 mm clearance (Normal fit)
   (c) 1.5
 mm clearance (Loose fit)
 (d) 2.0
 mm clearance (Extra loose fit)
 Figure 
5-12: Variation in I
nitial stiffness 
of 
hybrid joint
s with PRO
-SET epoxy
 as a function of bolt
-hole 
clearance
 In terms of joint stiffness, joints with 
close fit clearance 
had higher stiffness relative to 
other joints in this case study. This is expected as the load gets transferred from the bolt to the 
55 substrate instantaneously
. As the bolt clearance increases, th
e effect of the insert material on stress 
transfer increases. The insert material (matrix/adhesive) is an order of magnitude less stiff than the 
GFRP substrate and thereby supporting the reduction in stiffness observed in this study. Hence, 

from 
Figure 5
- 12, it can be concluded that the joint stiffness decrease
s as clearance increase
s. 
Similar to other joints, o
nce the delamination 
start
s, the stiffness of the joint redu
ces gradually 
until it reaches its maximum load
 followed by failure. The decrease in joint stiffness as the bolt
-clearance increases is quantified in
 table 5
-3. It was observed that the joint stiffness reduced by 
~33% in extra loose fit clearance joints re
lative to close
-fit clearance joints.
 Table 
5-3: Reduction in joint stiffness as a function of bolt
-hole clearance
.  Close
 Normal
 Loose
 Extra loose
 Stiffness (kN/mm)
 13 10 10 8.7 Percentage change from close
 to extra 
loose fit clearance
 - -23%
 -23%
 -33%
 5.4. Effect of Poly
urethane Insert on Hybrid Joints of Varying Bolt
-Hole Clearances
 Figure 5
-13 shows the representative load
-displacement responses of hybrid joints with 
polyurethane 
insert for each of the bolt
-hole clearances.
 Similar to hybrid joints with 
proset epoxy
 (Section 5.3)
 and 
aluminum
-filled epoxy inserts (section 5.2), the 
initial segment 
of the load
-displacement curve 
was
 linear
ly elastic u
p to 
~ 10kN
. This is the region wherein the compressive 
forces (clamping loads) are governing the behavior of the joint. Hence this r
egion is the same for 
all joints irrespective of the insert. The elimination of slip and the transfer of loads explained in
 
the
 previous section for ‚epoxy inserts™ joint
s is also valid for this joint and for brevity purposes 
not repeated here. One observa
tion that was unique to the 
polyurethane
 insert joints 
was the rapid 
change in slope/stiffness after the ~10 kN mark. As explained earlier, this is the load corresponding 

to bolt
-adherend slip wherein the compressive forces from clamping loads are overtake
n by the 
applied shear forces. Hence, the transfer forces through the ductile polyurethane leads to this 

change in slope. Both the epoxy inserts (Aluminum
-filled and Proset) are stiff and do not exhibit 
56 this slope change observed in polyurethane insert joi
nts. Similar to all other joints, f
urther increase 
in applied load 
resulted in the 
onset and progressive delamination of the substrate until peak load 
was reached followed by failure similar to other joints in this work.
     Figure 
5-13: Load
-displacement response of hybrid joints with polyurethane epoxy insert a
nd var
ying
 bolt
-hole clearances
. Figure 5
- 14 shows the effect of clearances on P
1 (Onset delamination), P
2 (Load based on 
max. slip distance) and P
3 (Peak load) as a function of slip load in conventional joint
s for the hybrid 
joints with
 ‚Polyurethane inserts.™
 Unlike 
epoxy inserts (section 5.2 & 5.3)
, the
 best 
performance 
in all parameters was
 obtained from
 joints with
 normal
-fit
 (1mm) and
 loose
-fit
 (1.5 mm) 
clearances
 and the least performance was observed in joints with extra loose
-fit
 clearance
. Overall, on average 
(P1, P2 and P
3 combined) the hybrid joints with Polyurethane insert were
 ~5 times better than the 
slip load of equivalent clearance conventional bolted joint.
 PP57  Figure 
5-14: Effect of clearance on hybrid joints with poly
-urethane epoxy inserts
. Figure 5
-15 shows the post failure images of hybrid joints with Polyurethane as 
the 
structural insert in the bolt
-hole clearances. Similar to that of conventional joints, the mode of 
failure changes with 
respect to bolt
-hole clearance. For close
-fit clearance, the bolt
 shear failure 
occurs
, and for all other clearances the joints fail with excessive delamination and bolt pull
-through.
      (a)
 (b)
 (c)
 (d)
 Close fit (0.5mm)
 Normal fit (1mm),
 Loose fit
 (1.5mm)
 Extra loose fit (2 mm)
 Figure 
5-15: Post failure images of with 
Polyurethane epoxy inserts at various clearances
.  58   0.5mm clearance (Close fit)
 (b) 1.0mm clearance (Normal fit)
    (c) 1.5mm 
clearance (Loose fit)
 (d) 2.0mm clearance (Extra loose fit)
 Figure 
5-16: Variation in Initial stiffness of hybrid joints with Polyurethane insert as a function of bolt
-hole clearance
 The change in stiffness after the 10 kN load (slip load) was calculated similar to that in 
the 
conventional joint (section 5.2) and 
the 
Proset epoxy insert (section 5.4). It can be observed from 
Figure 5
-16 and 
Table 5
-4 that the joint 
stiffness decreased with increase in the 
bolt
-substrate 
clearance values. 
It was observed that the jo
int stiffness reduced by ~48% in extra loose fit 
clearance joints relative to close
-fit clearance joints
. Table 
5-4: Reduction in joint stiffness as a function of bolt
-hole clearance
.  Close
 Normal
 Loose
 Extra
 loose
 Stiffness (kN/mm)
 10 9.7 6.2 6.2 Percentage change from close to extra 
loose fit clearance
 - -3% -38%
 -48%
 59 5.5. Chapter 
Conclusion
 In this 
chapter, the effect of bolt
-hole clearance on hybrid joints was studied for 
four 
different bolt
-hole clearances 
for each of the three insert materials.
 Bolt
-adherend slip was 
eliminated in all hybrid joints. Depending on the insert material, the performance of the hybrid 
joints could be tailored. All the joints 
having 
close
-fit clearances experienced bolt
 shear fail
ure. 
Similarly, all the joints with extra loose fit performed lower than normal and loose
-fit joints. In 
other words, normal
-fit and loose
-fit joints had the best performance in both peak loads and onset 
of delamination relative to the slip load of equival
ent conventional joints. 
Polyurethane insert 
joints had lower stiffness but higher ductility. Hybrid joints with epoxy inserts (aluminum filled 

and Proset) had a good balance of stiffness and toughness. 
Overall, the hybrid joints were 
4 to 8 

times better r
elative to their equivalent conventional joints. Finally, depending on the application, 
the insert material can be selected to tailor the joint behavior
 to meet application requirements
.   60 Chapter 6:
 Summary, Conclusion and Future Direction
s 6.1. Summary
 In this wo
rk, experimental characterizations of hybrid fastening systems comprised of glass 
fiber reinforced polymer (GFRP) composite substrates fastened using a fully threaded grade 5 steel 
bolts in ½fl (12.5 mm.) diameter with varying (four) bolt hole clearance and
 three structural insert 
materials
 were performed. The GFRP substrates were manufactured using the vacuum
-assisted 
resin transfer molding (VARTM) process. The preload was maintained at 75% of the bolt yield 

strength for all joints. 
 The objective of this w
ork was to characterize/quantify the effect of: a) the adhesive insert 
and b) the bolt
-hole clearances on the efficiency of the hybrid fastening system. For the first 
parameter, namely the effect of the adhesive insert, three adhesive insert materials were
 used. The 
adhesive insert materials used were PRO
-SET epoxy resin, DEVCON epoxy filled with aluminum 
particles, and Polyurethane. All resulting joints were cured at room temperature for 48h prior to 

testing. For the second parameter, namely the effect of 
bolt clearance, four different bolt
-hole 
clearances: close fit (0.5mm), normal fit (1.0 mm.), loose fit (1.5 mm.), and extra loose fit (2.0 

mm.) were studied for each of the insert materials along with the control/conventional joint without 

any structural 
insert. All joints were tested in tensile lap
-shear configuration at a loading rate of 5 
mm / min. A total of 64 joints were experimentally tested.
 The performance of the hybrid fastening system was compared to conventional joints with 
similar clearances. 
All conventional joints (without insert) experienced bolt
-adherend slip which 
was characterized by sudden increase in displacement without any increase in load. The load 

corresponding to this phenomenon of bolt
-adherend slip was termed the ‚Slip Load.™ In 
applications wherein small displacements are important, this bolt
-adherend slip can be considered 
61 as the failure of the resulting joint. Hence, the performance of hybrid joints was compared with 
the slip load of conventional joint with equivalent clearance
. Three parameters were defined to 
compare the performances of the hybrid joints with respect to the slip load. Parameter
-1 (P
1) compared the ‚load corresponding to onset of delamination (ODL)™ with slip load of conventional 
joints of equivalent clearance.
 Parameter
-2 (P
2) compared the ‚load in hybrid joints corresponding 
to the maximum slip displacement in equivalent conventional joints™ with slip load of conventional 
joint of equivalent clearance. Parameter
-3 (P
3) compared the ‚
peak load in hybrid joints
™ to the slip 
load in conventional joints of equivalent clearance
. In addition, the type and extent of damage of 
hybrid joints was compared with equivalent conventional bolted joints. 
 6.2. Key Conclusions
   Conventional bolted joints experienced ‚bolt
-adherend™ 
slip wherein the compressive forces 
and associated friction forces due to clamping loads were overcome by the applied shear forces. 

In the slip event, the load does not significantly increase, but the displacement increases until 

the bolt tilts and is rest
ing/bearing on the substrates.
  Hybrid bolted joints of all insert materials do NOT exhibit ‚bolt
-adherend™ slip. 
  The initial slope for all joints (hybrid and conventional) was similar. Additionally, the average 
slip load for all conventional joints (irres
pective of clearance) was also similar. This is the 
region which is controlled by the clamping load on the bolt wherein the compressive forces 

and associated friction forces are larger than the applied shear loads. Since the torque applied 

was the same for
 all joints, the slip
-loads and the initial slope of the load
-displacement response 
was same for all joints irrespective of insert material or bolt
-hole clearance.
 62  In conventional bolted joints, as expected, the maximum slip displacement increased with 
increase in bolt clearance. As the bolt clearance increases, there is more space available for the 
bolt to tilt/rotate and hence larger slip displacements.
  In conventional bolted joints, after the bolt
-adherend slip, the increase in applied loads leads 
to 
onset of delamination at the locations of contact of the bolt with substrates.
  This delamination continues until the peak load is reached followed by progressive 
delamination until failure.  
  In conventional joints, the stiffness of the joint after the sli
p event remains the same irrespective 
of the bolt clearance. 
  For hybrid joints, once the applied load was larger than the equivalent slip load, the effect of 
insert material and the bolt
-hole clearance was evident. 
  For joints with polyurethane (ductile, 
low
-stiffness) insert, a reduction in stiffness was 
observed in the loading region equivalent to the slip load in conventional joint. This shows that 
the insert material is activated after the slip load and acts as a structural element to transfer the 
load
 from the substrate to the bolt and vice
-versa
. Such reduction in stiffness was not observed 
for joints with both Proset and Aluminum filled epoxies. In short, P
olyurethane insert joints 
had lower stiffness but higher ductility.
  Hybrid joints with epoxy in
serts (aluminum filled and Proset) had a good balance of stiffness 
and toughness.
  All the joints having close
-fit clearances experienced bolt shear failure. 
  Similarly, all the joints with extra loose fit performed lower than normal and loose
-fit joints.
  The failure mechanism of all conventional bolted joints exhibited severe delamination 
culminating with a bolt pull
-through failure.
 63  The failure mechanisms of all hybrid joints were less severe than equivalent conventional 
joints. 
  Overall, on average (P
1, P2 and P
3 combined) the hybrid joints were ~5
-8 times better than the 
slip load of equivalent clearance conventional bolted joint.
  Finally, depending on the application, the insert material can be selected to tailor the joint 

behavior to meet application re
quirements.
  Overall, this work is novel, and the first to report on the effect on clearance and varying 
adhesive insert materials for hybrid fastening joint.
 6.3. Future Work 
 This work was the first to report on the effect on clearance and varying adhesive ins
ert 
materials for hybrid fastening joint. While
 the merits of hybrid fastening were clear, there is a 
need to quantify the stress
-concentrations and its reductions due to the addition of these 
structural inserts. Novel measurement techniques such as embedd
ed fiber
-optic sensors can be 
used to provide accurate stress measurements at the appropriate locations of the hybrid joint. 

Such precise stress measurements will also allow for the selection of the right type of insert 

material based on the applications. 
Secondly, it is impractical and infeasible to perform 
experimental characterization for all the design parameters of hybrid fastened joints. Instead 

this work can be used to develop and validate numerical simulations. The experimentally 

validated simulatio
ns can then be used to explore the design space without the costly trial
-and
-error approach. Lastly, this work tested three samples per case. For industrial applications, a 

more detailed statistical testing needs to be performed prior to field applications
. Overall, the 
hybrid fastened joints are tailorable based on the application needs and hence have a great 

potential in a wide range of applications. 
  64   
   
 
       
 APPENDIX
    65 APPENDIX
: List of Publications by the Candidate at the Time of Dissertation 
Submission
  [1].
 Balachandran, S. Ramli, N. Abdol, P. Soroushian, Transformation of Landfilled Ash into a 

Hydraulic Binder with Concurrent Capture of Carbon Dioxide for Effective Hazardous Waste 

Immobilization, September 2018
 [2].
 K. Zhu, S. Ramli, A.G.N.D. Darsanas
iri, A. 
Balachandran, Production and Characterization of 
a Hydraulic Cement with Landfilled Coal Ashes of Different Disposal Durations, Using 

Carbon Dioxide 
as a Raw Material, September 2018
 [3].
 A.G.N.D. Darsanasiri, F. Matalkah, S. Ramli, P. Soroushian, Terna
ry Alkali Aluminosilicate 
Cement Based on Rice Husk Ash, Slag and Coal Fly Ash, April 2018
 [4].
 K. Zhu, F. Matalkah, S. Ramli, A. 
Balachandran, Carbon Dioxide Use in Beneficiation of 
Landfilled Coal Ash for Hazardous Waste Immobilization, March 2018
 [5].
 X. Wang, F.
 Matalkah, N. Abdol, S. Ramli, P. Soroushian, A. Balachandran, Effects of the 
duration of landfill disposal on the physicochemical, minerology and toxicity characteristics 

of coal ash. March 2018
 [6].
 X. Wang, K. Zhu, S. Ramli, P. Soroushian, A. Balachandran, C
onversion of Landfilled Ash 
into Hydraulic Cements under Different Environments, November 2017
 [7].
 S. Ramli, K. Zhu, F. Matalkah, P. Soroushian, A. 
Balachandran, Development of Refined 
Chemistries and Processing Methods for Integration of Carbon Dioxide into a
 Hydraulic 
Binder for Effective Heavy Metal Immobilization, August 2017
   
  66   
   
 
       REFERENCES
  67 REFERENCES
  [1].
 Abbas, M., N. Sakundarini, and I. Kong. Optimal Selection for Dissimilar Materials using 
Adhesive Bonding and Mechanical Joining. in IOP 
Conference Series: Materials Science 
and Engineering. 2019. IOP Publishing.
  [2].
 Martinsen, K., S. Hu, and B. Carlson, Joining of dissimilar materials. CIRP Annals, 2015. 

64(2): p. 679
-699.  
[3].
 Messler, R.W., Joining of materials and structures: from pragmatic 
process to enabling 

technology. 2004: Butterworth
-Heinemann.
  
[4].
 Campbell, F.C., Joining: understanding the basics. 2011: ASM International.
  
[5].
 Hwang, H.
-Y., Bolted joint torque setting using numerical simulation and experiments. 
Journal of Mechanical Science a
nd Technology, 2013. 27(5): p. 1361
-1371.
  
[6].
 Cloud, G., 2012 William M. Murray lecture: some curious unresolved problems, 

speculations, and advances in mechanical fastening. Experimental Mechanics, 2013. 53(7): 

p. 1073-1104.  
[7].
 Kah, P., et al., Techniques for 
joining dissimilar materials: metals and polymers. Reviews on 

Advanced Materials Science, 2014. 36(2).
  [8].
 Camanho, P. and F. Matthews, Bonded metallic inserts for bolted joints in composite 

laminates. Proceedings of the Institution of Mechanical Engineers, P
art L: Journal of 
Materials: Design and Applications, 2000. 214(1): p. 33
-40.  
[9].
 Koricho, E.G., et al., Fatigue behavior of novel hybrid fastening system with adhesive inserts, 

in Fracture, Fatigue, Failure and Damage Evolution, Volume 8. 2016, Springer. p. 
269-276.   
[10].
 Kelly, G., Quasi
-static strength and fatigue life of hybrid (bonded/bolted) composite single
-lap joints. Composite structures, 2006. 72(1): p. 119
-129.  
[11].
 Cloud, G.L., Hybrid Fastener, U.S. Patent 9,464,658. Issued 11 October 2016. Assigned to 

Mic
higan State University.
  
[12].
 Cloud, G.L., Hybrid Fastener, U.S. Patent 10,400,813 B2. Issued 3 September 2019. 

Assigned to Michigan State University
  [13].
 Girard, C., et al., Bearing Behavior of Mechanically Fastened Joints in Composite 
Laminates
--Part I: Strength 
and Local Strains. Mechanics of Advanced Materials and 
Structures, 2003. 10(1): p. 1
-21. 68 [14].
 Girard, C., et al., Bearing Behavior of Mechanically Fastened Joints in Composite 
Laminates
--Part II: Failure Mechanisms. Mechanics of Advanced Materials and Structure
s, 
2003. 10(1): p. 23
-42.  
[15].
 Zhou, S., et al., Experimental and numerical investigation on bolted composite joint made 
by vacuum assisted resin injection. Composites Part B: Engineering, 2013. 45(1): p. 1620
-1628.  [16].
 Ekh, J. and J. Schön, Effect of secondary 
bending on strength prediction of composite, single 

shear lap joints. Composites Science and Technology, 2005. 65(6): p. 953
-965.
  
[17].
 Mara, V., R. Haghani, and M. Al
-Emrani, Improving the performance of bolted joints in 
composite structures using metal insert
s. Journal of Composite Materials, 2016. 50(21): p. 

3001-3018.  
[18].
 Mirabella, L. and S. Galea, Strength of composite laminates might be degraded due to the 

presence of fastener hole. Iccm
-12: Processing, Joints, Normalisation, Durability/Ageing and 
Repair, 19
97. 6: p. 148.
  
[19].
 Herrera
-Franco, P.J. and G.L. Cloud, Strain
-relief inserts for composite fasteners
-an 
experimental study. Journal of Composite Materials, 1992. 26(5): p. 751
-768.
  
[20].
 Raju, K.P., et al., Improving load sharing in hybrid bonded/bolted composite
 joints using an 
interference
-fit bolt. Composite Structures, 2016. 149: p. 329
-338.  
[21].
 Camanho, P., et al., Increasing the efficiency of composite single
-shear lap joints using 
bonded inserts. Composites Part B: Engineering, 2005. 36(5): p. 372
-383.  
[22].
 Risple
r, A.R., G.P. Steven, and L. Tong, Photoelastic evaluation of metallic inserts of 
optimi
zed shape. Composites science and technology, 2000. 60(1): p. 95
-106.
  
[23].
 Nilsson, S., Increasing strength of graphite/epoxy bolted joints by introducing an adhesively 

bonded metallic insert. Journal of composite materials, 1989. 23(7): p. 642
-650.  
[24].
 Camanho, P. and F. Matthews, Bonded inserts for cfrp bolted joints. EUROPEAN SPACE 

AGENCY
-PUBLICATIONS
-ESA SP, 2000. 468: p. 207
-212.  [25].
 Egan, B., et al., Stress analysis of 
single
-bolt, single
-lap, countersunk composite joints with 
variable bolt
-hole clearance. Composite Structures, 2012. 94(3): p. 1038
-1051.
  [26].
 Titanium, T. Fasteners 101 
ΠTorque Setting. 2017; Available from: 
http://www.ti64.com/v/vspfiles/assets/docs/fastene
rs%20101%20torque.pdf.
  [27].
 Olmedo, Á. and C. Santiuste, On the prediction of bolted single
-lap composite joints. 
Composite Structures, 2012. 94(6): p. 2110
-2117. 69 [28].
 Khashaba, U., et al., Effect of washer size and tightening torque on the performance of bolted 
joints in composite structures. Composite Structures, 2006. 73(3): p. 310
-317.
  [29].
 Tong, L., Bearing failure of composite bolted joints with non
-uniform bolt
-to-washer 
clearance. Composites Part A: Applied science and manufacturing, 2000. 31(6): p. 609
-615.  [30].
 Qureshi, J. and J.T. Mottram, Resin injected bolted connections: a step towards achieving 
slip
-resistant joints in FRP bridge engineering. Proceedings FRP bridges, 2012: p. 55
-66.  
[31].
 Mottram, J., Friction and load transfer in bolted joints of pultruded fibre
 reinforced polymer 
section. 2 nd Inter. Conf. on FRP Composites in Civil Engineering (CICE04). 2005, Taylor 

& Francis plc, London.
  
[32].
 Yang, B., E. Pan, and F. Yuan, Three
-dimensional stress analyses in composite laminates 
with an elastically pinned hole. In
ternational Journal of Solids and Structures, 2003. 40(8): 

p. 2017-2035.  
[33].
 Padhi, G., M. McCarthy, and C. McCarthy, BOLJAT: a tool for designing composite bolted 

joints using three
-dimensional finite element analysis. Composites Part A: Applied Science 
and 
Manufacturing, 2002. 33(11): p. 1573
-1584.  
[34].
 Hühne, C., et al., Progressive damage analysis of composite bolted joints with liquid shim 

layers using constant and continuous degradation models. Composite Structures, 2010. 

92(2): p. 189
-200.  
[35].
 Koricho, E.G., M
. Haq, and G.L. Cloud, Numerical and Experimental Characterization of 
Hybrid Fastening System in Composite Joints. 2017: p. 71
-80.  
[36].
 Haq, M. and G. Cloud. Flexible hybrid fastening system. in SEM 2013 Annual Conference 

& Exposition on Experimental and Appli
ed Mechanics, Chicago. 2013.