enhancement of spike and stab resistance...

ENHANCEMENT OF SPIKE AND STAB RESISTANCE OF FLEXIBLE ARMOR

USING NANOPARTICLES AND A CROSS-LINKING FIXATIVE

by

Vincent Lambert

A Thesis Submitted to the Faculty of

The College of Engineering and Computer Science

In Partial Fulfillment of the Requirements for the Degree of

Master of Science

Florida Atlantic University

Boca Raton, Florida

April 2009

iii

ACKNOWLEDGEMENTS

I would like to thank Dr. H. Mahfuz for his direction, assistance and guidance in

the preparation of this thesis. I wish to thank the members of the supervisory committee,

Dr. M. Dhanak and Dr. F. Presuel-Moreno, for their valuable recommendations and

suggestions.

Financial support in the form of a research assistantship from the Army

Research Office under the Battlefield Capability Enhancement program, grant

W911NF0520006, is gratefully acknowledged.

I wish to thank my friends and lab mates for their valuable support. I would also

like to sincerely thank my family for their advice, support and unconditional love

although I am far from away from home.

iv

ABSTRACT

Author: Vincent Lambert

Title: Development of Flexible Body Armor using SiO2 nanoparticles and cross-linking fixatives

Institution: Florida Atlantic University

Thesis Advisor: Dr. Hassan Mahfuz

Degree: Masters of Science

Year: 2009

A novel approach has been introduced in making flexible armor composites.

Armor composites are usually made by reinforcing Kevlar fabric into the mixture of a

polymer and nanoscale particles. The current procedure deviates from the traditional

shear thickening fluid (STF) route and instead uses silane (amino-propyl-trimethoxy

silane) as the base polymer. In addition, a cross-linking fixative such as Glutaraldehyde

(Gluta) is added to the polymer to create bridges between distant pairs of amine groups

present in Kevlar and silated nanoparticles. Water, silane, nanoparticles and Gluta are

mixed using a homogenizer and an ultra-sonochemical technique. Subsequently, the

admixture is impregnated with Kevlar – by passing the heating and evaporating

processes involved with STF. The resulting composites have shown remarkable

improvement in spike resistance; at least one order higher than that of STF/Kevlar

composites. The source of improvement has been traced to the formation of secondary

amine C-N stretch due to the presence of Gluta.

v

TABLE OF CONTENTS

CHAPTER 1.INTRODUCTION ......................................................................................... 1

1.1 Literature Review ................................................................................................ 2

1.2 Scope of Thesis .................................................................................................... 9

CHAPTER 2.MATERIALS, EQUIPMENT AND SYNTHESIS ..................................... 11

2.1 KM-2 Kevlar® Fabric ....................................................................................... 11

2.2 Correctional Kevlar® ........................................................................................ 12

2.3 Spectra® ............................................................................................................ 14

2.4 Polyethylene Glycol (PEG) ............................................................................... 16

2.5 Silica Nanoparticles ........................................................................................... 16

2.6 Organosilane ...................................................................................................... 18

2.7 Glutaraldehyde .................................................................................................. 19

2.8 High Intensity Ultrasonic Liquid Processor ...................................................... 22

2.9 Synthesis of the Silated-Nanoparticles-Glutaraldehyde -Fabric Composites.... 23

CHAPTER 3.EXPERIMENTATION ............................................................................... 25

3.1 NIJ Stab Test ..................................................................................................... 25

3.1.1 Test Methodology .......................................................................................... 26

3.1.2 Procedure ....................................................................................................... 28

3.2 Scanning Electron Microscope (SEM) .............................................................. 30

3.3 Fourier Transform Infrared Spectroscopy (FTIR) ............................................. 31

vi

3.4 Mechanical Testing ........................................................................................... 31

3.4.1 Testing procedure .......................................................................................... 31

3.5 Flexibility Test ................................................................................................... 33

CHAPTER 4.RESULTS & DISCUSSION ....................................................................... 35

4.1 Introduction ....................................................................................................... 35

4.1.1 STF performances ......................................................................................... 35

4.1.2 Removal of Polyethylene Glycol (PEG) ....................................................... 36

4.1.3 Introduction of Silane-Silica-Glutaraldehyde Systems ................................. 39

4.2 NIJ Stab test ....................................................................................................... 39

4.2.1 NIJ Spike test ................................................................................................. 39

4.2.2 NIJ Knife test ................................................................................................. 41

4.2.3 Studies of various fabric ................................................................................ 43

4.2.4 Hybridization for Optimization ..................................................................... 47

4.2.5 Introduction of CaCo3 .................................................................................... 52

4.2.6 Failure Analysis ............................................................................................. 56

4.3 Microscopy ........................................................................................................ 74

4.3.1 Silane-Silica-Gluta/Kevlar Microscopy ........................................................ 74

4.3.2 Evolution of bonding ..................................................................................... 77

4.4 Chemical Analysis ............................................................................................. 80

4.5 Mechanical Testing ........................................................................................... 82

4.6 Flexibility test .................................................................................................... 90

4.7 Discussion .......................................................................................................... 92

CHAPTER 5. FINITE ELEMENT ANALYSYS OF SPIKE PENETRATION ............. 95

vii

5.1 The Finite Element Method ............................................................................... 95

5.2 Modeling of the spike penetration problem ....................................................... 96

5.3 Elements and mesh generation ........................................................................ 100

5.3.1 SOLID92 ..................................................................................................... 100

5.3.2 SOLID186 ................................................................................................... 103

5.3.3 CONTA178 ................................................................................................. 105

5.4 Determination of the materials properties ....................................................... 107

5.5 Boundary Conditions ....................................................................................... 111

5.6 Results ............................................................................................................. 112

5.7 Discussion ........................................................................................................ 114

CHAPTER 6.CONCLUSION ......................................................................................... 116

6.1 Summary .......................................................................................................... 116

6.2 Future Work ..................................................................................................... 117

viii

LIST OF FIGURES

Figure 1: (a) Molecular structure of Kevlar (b) aromatic ring (c) amide

group. ...................................................................................................................... 12

Figure 2: A x-ray linear dichroism microscopic view of the cross section

of a Kevlar fiber showing radial symmetry [52, 53]. .............................................. 12

Figure 3: Kevlar Correctional fabric under NIJ Spike test. [49] ...................................... 13

Figure 4: Spectra fibers are made out bright white polyethylene ..................................... 14

Figure 5: Molecular structure for Ultra high molecular weight

polyethylene (UHMWPE)....................................................................................... 15

Figure 6: Molecular formula for PEG where n denotes the molecular

weight. For the current research n=4.2 corresponds to a 200g/mol

average molecular weight. ...................................................................................... 16

Figure 7: Laser-induced CVC to produce nanostructured SiO2 ....................................... 17

Figure 8: Molecular structure of the silane coupling agent .............................................. 18

Figure 9: Molecular structure of the trialkoxysilane coupling agent ............................... 19

Figure 10: Silanol linkages between the polymer and the silica substrate

[57]. ......................................................................................................................... 19

Figure 11: Molecular formula of Glutaraldehyde ............................................................ 20

ix

Figure 12: Aldol Condensation Reaction ......................................................................... 20

Figure 13: Molecular reaction for cross-linking bonding of aldehyde

groups with silated silica particles .......................................................................... 21

Figure 14: VCX Series Ultrasonic Processor from Sonics [58]. ....................................... 22

Figure 15: The manufacturing procedures. Sonicating the particles and

infusing into the fabric using a sealed bag and finally, oven drying

the fabric composite. ............................................................................................... 24

Figure 16: NIJ115 drop mass [59]. ................................................................................... 26

Figure 17: NIJ115 threat weapons; (a) Engineered Knife Blade P1 (one

cutting edge) (b) Engineered Knife Blade S1 (two cutting edges)

(c) Engineered Spike [59]. ...................................................................................... 27

Figure 18: NIJ115 Composite Backing Material [59]. ..................................................... 28

Figure 19: NIJ115 drop tower and system setup; (a) drop apparatus (b)

drop mass (c) threat weapon (Spike) (d) velocity measurement

zone (e) backing material ........................................................................................ 29

Figure 20: (a) Kevlar composite after impact at 16 Joules (b) impacted

witness paper at various impact energies (c) magnified view of the

impacted witness paper. .......................................................................................... 30

Figure 21: Different directions of the testing samples ..................................................... 32

Figure 22: Samples cut from fabric composites ............................................................... 32

Figure 23: A typical tension test in Zwick ....................................................................... 33

x

Figure 24: Flexibility test developed by Lee et Al [64] ................................................... 34

Figure 25: NIJ Spike test of STF based composites [63] ................................................. 36

Figure 26: NIJ Spike test of STF based composites with and without

PEG [65] ................................................................................................................. 38

Figure 27: NIJ Spike test of Kevlar based fabrics [65] .................................................... 40

Figure 28: Failure of the backing material after reaching higher energy

level during NIJ Spike test ...................................................................................... 41

Figure 29: NIJ Knife test graph ........................................................................................ 42

Figure 30: NIJ Spike test for various fabrics .................................................................... 44

Figure 31: NIJ Knife test for various fabrics .................................................................... 45

Figure 32: NIJ Spike test for determination of the best hybrid composite ....................... 48

Figure 33: NIJ Knife test for optimum determination ...................................................... 49

Figure 34: Knife/Spike performance for a 15 layers Kevlar/Spectra

Hybrid ..................................................................................................................... 51

Figure 35: Nanoparticles of CaCO3 .................................................................................. 53

Figure 36: NIJ Spike test of CaCO3 based composite ...................................................... 54

Figure 37: NIJ Knife test of CaCO3 based composite ...................................................... 55

Figure 38: Spike penetration in Spectra composite .......................................................... 57

Figure 39: Melted fibers in Spectra composite ................................................................. 58

Figure 40: Spike penetration in Kevlar composite ........................................................... 59

xi

Figure 41: Gluta reinforcement being torn up during spike penetration in

Kevlar composite .................................................................................................... 60

Figure 42: Breakage of the reinforcement between fibers during spike

penetration in Kevlar composite ............................................................................. 61

Figure 43: Knife penetration in Spectra composite .......................................................... 62

Figure 44: Zoom in the area where the cut is done by the blade in the

Spectra composite ................................................................................................... 63

Figure 45: Cut profile of one fiber in the Spectra composite ........................................... 64

Figure 46: Knife penetration in Kevlar composite ........................................................... 65

Figure 47: Yarn cut in Kevlar composite ......................................................................... 66

Figure 48: Fiber cut in Kevlar composite ......................................................................... 67

Figure 49: A thin coating of silated SiO2 with glutaraldehyde mixture on

the surface of the Kevlar fabric ............................................................................... 70

Figure 50: Silane-Silica-Gluta/Kevlar bonding ................................................................ 71

Figure 51: Coating of the silated SiO2 with glutaraldehyde onto the

surface of the Spectra fabric. ................................................................................... 72

Figure 52: Coating of the silated SiO2 with glutaraldehyde mixture

wearing off the surface of the Spectra fabric after impact. ..................................... 73

Figure 42: A thin coating of the silated SiO2 with glutaraldehyde

mixture on the surface of the Kevlar fabric ............................................................ 75

xii

Figure 43: Agglomerated Silated SiO2 particles and Glutaraldehyde

mixture. ................................................................................................................... 76

Figure 44: Neat Kevlar. .................................................................................................... 77

Figure 45: Silane-Silica-PEG/Kevlar bonding ................................................................. 78

Figure 46: Silane-Silica-Gluta/Kevlar bonding ................................................................ 78

Figure 47: Silane-Calcium Carbonate-Gluta/Kevlar ........................................................ 79

Figure 48: FTIR of the different glutaraldehyde ratios added to the

silated silica mixture. Gluta1=0.225g, Gluta2=0.113g,

Gluta3=0.45g .......................................................................................................... 80

Figure 49: FTIR of different glutaraldehyde combinations present in the

final mixture Silane-Silica-Gluta/Kevlar ................................................................ 81

Figure 50: Warp direction tension test ............................................................................. 82

Figure 51: Fill direction tension test ................................................................................. 83

Figure 52: 45° direction tension tests ............................................................................... 84

Figure 64: a)45degree Kevlar composite before test. b) after test. .................................. 87

Figure 65: a)45degree Spectra composite before test. b) after test. ................................. 89

Figure 53: Flexibility Set-up ............................................................................................ 90

Figure 54: Geometry models of the spike and nylon mass .............................................. 96

Figure 55: Cross-sections of spike and nylon mass .......................................................... 97

Figure 56: Geometry model of backing materials ............................................................ 97

xiii

Figure 57: Backing materials geometry details ................................................................ 98

Figure 58: The spike-mass system is placed on the top of the backing

material on aligned on the center ............................................................................ 99

Figure 59: Global view of the geometry ........................................................................... 99

Figure 60: SOLID92 element geometry. 3-D 10-Node Tetrahedral

Structural Solid. .................................................................................................... 101

Figure 61: Meshing of the target and backing material with SOLID92 ......................... 101

Figure 62: SOLID92 stress output .................................................................................. 102

Figure 63: SOLID186 geometry ..................................................................................... 103

Figure 64: Meshing of the Spike and its mass with elements SOLID186 ...................... 104

Figure 65: Insertion of contact element CONTA178 in between the

spike and the target ............................................................................................... 105

Figure 66: CONTA178 element geometry. .................................................................... 106

Figure 67: Warp and weft yarn orientations in global coordinate system. ..................... 107

Figure 68: Local orthogonal coordinate system for warp layer. .................................... 107

Figure 69: Boundary conditions applied on the system ................................................. 112

Figure 70: Simulated penetration of the spike. VonMises stress result. ........................ 113

Figure 71: Comparison graph of Simulated Quasi-static impact

penetration to Experimental .................................................................................. 114

xiv

LIST OF TABLES

Table 1: The amount of glutaraldehyde, water and silane used to

functionalize 5.5g of 30 nm silica particles according the

manufacturers procedure ......................................................................................... 24

Table 2: Results table for NIJ Knife data of 12 layers combination of

reinforced Correctional Kevlar and Spectra .......................................................... 126

Table 3: Results table for flexibility test on the different armor

composites ............................................................................................................... 90

Table 4: Material properties table for the different constituents of the

model ..................................................................................................................... 111

Table 5: Results table for NIJ Spike data of 12 layers of independent

fabric neat and composite ..................................................................................... 125

Table 6: Results table for NIJ Spike data of 12 layers combination of



reinforced Kevlar and Spectra ............................................................................... 128


neat Correctional Kevlar and Spectra ................................................................... 129

xv


neat Kevlar and Spectra ........................................................................................ 130

Table 10: Results table for NIJ Knife data of 12 layers of independent

fabric neat and composite ..................................................................................... 131




reinforced Kevlar and Spectra ............................................................................... 133


neat Kevlar and Spectra ........................................................................................ 134

1

CHAPTER 1.INTRODUCTION

Most of body protection gears for soldiers have been aimed towards ballistic

protection. Because ballistic protection mainly consists of rigid panels (i.e. ceramic

plates) inserted into a fabric pouch or incorporated in hard formed gear, they are

restricted to the head and torso. It does not include the extremities such as hands, arms,

necks, and legs. The head and torso are extremely important to protect because these

areas consist of life supporting organs. Due to increased casualties as a result of

extremity injuries and the increase in soldiers carry on loads, further development in

lightweight and flexible warrior systems has been explored. The need for lightweight

and flexible armor is to maximize the maneuverability without sacrificing protection of

the soldier. In addition to ballistic resistance, attention, is also given to threats imposed

by sharp weapons. Protection against both threat systems incorporating the desired

lightweight and flexible characteristics is under continual investigation.

A case in point is nanotechnology. This allows doing away with the bulky and

heavy plates and replacing that with a nano-materials system that can dissipate the

impact energy effectively. Development of nanocomposites by nanoparticle infusion

into polymers has been around for quite sometime. The benefit of nanoparticle infusion

comes from the fact that with a low particle loading results in a relatively large

improvement in chemical, thermal, and mechanical properties. In addition to the

improved properties, the weight of the composite is significantly reduced compared to

2

conventional method of reinforcement with large objects (i.e. ceramic plates). The

increase in properties and performance is due to the increase in surface energy, which is

caused by the relative increase in atoms at the surface, facilitating the interaction with

the surrounding polymers. The goal of nanoparticle infusion is to achieve an optimal

dispersion of nanoparticles into the matrix, and preferably a strong bonding between the

particle and the surrounding polymer. Uniform dispersion is necessary to improve Van

der Waals interaction between the particles and polymer and is imperative for

enhancing the strength of the resulting nanocomposites. One of the ways to improve

such bonding is to use a coupling agent between the organic and inorganic phases. In

this investigation, we have used organosilanes to modify the surfaces of Silica (SiO2)

nanoparticles. Organosilanes has the ability to incorporate both organic- and inorganic-

compatible functionality within the same molecule. In order to achieve even greater

performances, investigations have shown that stronger bonding links could be obtained

by incorporating another chemical belonging to the formaldehyde family. This

formaldehyde type chemical called Glutaraldehyde increases the number of bonding

between silated particles. Improved impact resistance compared to neat fabric and

current flexible armor material systems has been achieved.

1.1 Literature Review

Extensive experimental work has been done in recent years on flexible material

for body protection applications which are not limited to surgical gloves,

geotextiles/geomembranes and impact resistant composites. The literature review

includes methods and research on stab resistant [2-5] (i.e. puncture due to a knife and/or

3

spike), cut resistant and ballistic resistant materials and technology that has improved

these characteristics in common materials used for body protection (i.e. Kevlar).

Ngyen et al. [1] researched the mechanisms of puncture in thin rubber

membranes (i.e. protective gloves, neoprene, nitrile, and natural rubber) following

ASTM F1342 [2]. They used a conical probe with different cone angles and varied

diameters to derive expressions to calculate the theoretical puncture force versus the

varied probe geometry based on the deformation of the membrane. It was concluded

that the penetration force was not dependent on the probe geometry but related to an

intrinsic material parameter.

Leslie et al. [3] studied puncture resistance of a variety of medical glove

material that is specific to resisting needle punctures (i.e. finger guards, glove liners and

thicker latex gloves). The results included a quantitative measurement of peak force

and work required for a needle to puncture the material. These two values were used to

compare surgical hand protection systems.

Erlich et al. [4] studied the quasi-static penetration of both a blunt and sharp end

penetrator through a single ply of Zylon® fabric. They used quasi-static tests to

compare their previous work on dynamic behavior. They realized that failure modes are

the same but just at different levels for static versus dynamic testing. The test setup

captured video of the entire load stepping up to failure. This allowed a side by side

comparison of the measured data with the captured failure modes occurring during

penetration. They report that the deformation and the failure modes present are the

same between the quasi-static and ballistic test taking into account what was previously

mentioned.

4

Koerner et al. [5], Narejo et al. [6] and Wilson-Fahmy et al. [7] studied the

puncture resistance of geomembranes in three sections; theory, experimental and

examples, which were individually published. The works discuss the effects of

geomembranes used as liners underground or underwater. They simulate puncture from

gravel and soil by studying the geometry of the protrusion to the deformation of the

geomembrane and includes hydrostatic and geostatic pressures during the load. They

compare the thickness of the geomembranes and the addition of geotextiles and their

thickness to give increase puncture resistance to geomembranes and to give a cost

effective approach.

In complement to the series of works mentioned above Ghosh et al. [8]

concentrates on the puncture resistance of geotextiles and uses a test method similar to

ASTM D4833 [9], but includes the pre-straining of the material.

Lara et al. [10, 11] in 1996 studied glove material (neoprene and Kevlar) for the

cutting effects of degradation of blade sharpness, blade speed, sample holder as flat

versus semi-circular, and the load applied to the specimen. After testing, the results

concluded that changing of the blade is required for each test due to dulling of the

blade, the blade speed and sample holder had little effect on the results, and a series of

test are recommended to obtain the range of load which is specific to each material.

The later is in replace of applying the same load to different materials and measuring

the varied blade cycle. Comparing low to high cut resistant material, more blade cycles

are required for the later. This adds to the degradation of the blade and inaccurate

results. In 1997, ASTM approved standard F1790 [12] which is similar to that of Lara

et al except it restricts the blade travel and measures the load which is allowed to vary,

5

due to the setup, instead of keeping it constant. In 2000, Lara et al. [13] evaluated

methods and standards of EN 388, ASTM 1790 and ISO 13997 and compared them to

their previous works in 1996. They continue to emphasize the point to use a constant

normal force to be applied to the fabric specimen that is to be cut. The faults of the

three standards were found to be optimal for different specimen thicknesses and the

results depended greatly on the coefficient of friction and hence are incomparable. Shin

et al. [14] developed a method and apparatus to measure cut resistance of yarn,

specifically tested Zylon®, under tension. His results are more detailed to include

visuals and quantitative results to analyze yarn failure modes. With their interest in

fragment impact, they also test at different slice angles. These methods and standards

are limited to cut resistance and do not include any other failure modes such as

puncture, but give an approach to critical thinking when comparing methods for testing

flexible protective materials.

An approach to spike impact has been developed in England, with the:NIJ

Standard 0115.00 [66]. This standard specifies the minimum performance requirements

for body armor that is resistant to attack by typical pointed and edged weapons; and it

describes the test methodology to be used for this assessment. The spike, or knife, is

held by a mass and dropped at various heights. A backing material supports the target

and is constituted with four neoprene layers, one polyethylene layer and two rubber

layers. Knowing the areal density of the target, it is then possible to calculate the energy

and normalized energy at impact. This allows the comparison and ranking of the

different personal body protections manufactured and put to the test.

6

Works from MIT by Deshmukh and McKinley [15] include the study of a liquid

infused fabric body armor which resists impact when under a magnetic field. This fluid

is called magnetorheological (MR) fluid which consists of iron particles. Works by

Gadow et al. [16] studied the stab resistance of a thermally sprayed ceramic and cermet

coatings on aramid fabrics, such as Twaron. The heated coating is sprayed onto the

fabric using an atmospheric plasma spray torch and is cooled at the point of contact

with the fabric to prevent fiber damage. The coating is 50 to 100 μm thick with

individual particles sizes of 10-22 μm. The coatings from the particles tend to have a

high density. This may give a more flexible and lightweight solution to the common

ceramic reinforced body armor, but the weight is still relatively high. Other coatings

have been explored using a shear thickening fluid, commonly known as a dilatant fluid,

which is impregnated into woven aramid fabrics. The rheology of the non-Newtonian

fluid behavior has been studied by Raghavan et al. [17, 18], Maranzano et al. [19] and

Lee et al. [20]. These works discuss the effects of the fluid’s increase in viscosity when

there is an increase in shear rate. The increase in fluid’s viscosity is the key in energy

dissipation of impacts (i.e. ballistic). More works of Lee et al. [21, 22], Wetzel et al.

[23], Egres et al. [24, 25] and Tan et al. [26] have studied the performance of the silica

infused fabrics. These three groups have mainly worked with Shear Thickening Fluid

(STF) which involves a centrifugation and exchange processes to transform a silica and

polyethylene glycol (PEG) mixture into STF. The development of shear thickening

fluid (STF) begins with micron size silica suspended in water. This material is made

into STF through centrifugation and exchange processes. In other approaches

polyethylene glycol (PEG) is added in incremental quantity to the silica suspension and

7

water is removed through evaporation. The process is repeated until the admixture

reaches the desired ratio of 55:45 for silica and PEG by weight. STF developed in this

manner is dissolved in ethanol and impregnated into Kevlar fabric and then dried to

remove ethanol. The resulting composite is Kevlar impregnated with a mixture of PEG

and silica. These studies report the ballistic performances of composite materials

composed of Kevlar fabric impregnated with a colloidal shear thickening fluid (micron

size silica particles dispersed in ethylene glycol). The impregnated Kevlar fabric yields

a flexible, yet penetration resistant composite material. Ballistic and stab penetration

measurements have demonstrated a significant improvement due to the addition of

shear thickening fluid to the fabric without any loss in material flexibility. Such

enhancement in the performance has been attributed to the increase in the yarn pullout

force upon transition of the STF to its rigid state during impact. Furthermore, STF

impregnated Kevlar has vastly superior stab resistance in addition to flexible ballistic

protection. While these preliminary studies establish clearly the viability of the

STF/fabric composite as a future flexible body armor system, the entire scope of

particle-polymer interaction along with the complexities associated with fabric

impregnation must still be addressed before an optimal, lightweight STF/fabric system

can be developed.

To evaluate the results, theoretical work has also been carried through. Works

on modeling ballistic impacts on fibrous materials, even force profiles of sewing

needles, and modeling sharp indenters on metals have been looked into.

The works of Phoenix et al. [27], Porwall et al. [28], and Taylor et al. [29] investigate

ballistic impacts on integral armor and fibrous composites. The ballistic studies focus

8

on the wave that occurs from the impact and temperature effects and yarn pullout. The

above works do not study impacts at low velocity where wave propagation and

temperature effects do not occur. Gu [30] didn’t consider the projectile geometry during

ballistic impact of plain-woven fabric, while it is believed that geometry of the

projectile plays an important role in penetration mechanics. Works of Luo et al. [31]

consider the fabric composite to be homogeneous orthotropic material comprised of

wavy fibers. Common among the above studies, deformation analysis is of great

importance in ballistic impact. In the present work of this thesis, the impact velocity is

below the ballistic range, and the projectile (in the case of the spike) is not a fragment

simulated projectile (FSP).

Stylios et al. [32] focused on the profile of a sewing needle point affecting the

penetration force. This work aims to optimize the profile to reduce the force of

penetration. The works of Suresh et al. [33, 34], Giannakopoulos et al. [35] and

Andrews et al. [36] from the mid nineties to 2001 were the most inspiring. Their work

focused on the analysis of sharp indenter impacts into rate dependent metals. The

motion of the indenter is described by Newton’s second law, considering the indenter

mass, the motion of the indenter and the impact force. Once the equation of motion and

the governing equation are formulated, it is solved with appropriate boundary

conditions to find the maximum depth of penetration. Penetration of an indenter

described in these references is similar to the penetration of a sharp spike considered in

our studies. Accordingly we have developed a numerical model using the finite element

method to simulate the motion of the indenter penetrating through the fabric.

9

1.2 Scope of Thesis

The current study is mainly focused on the enhancement in non-penetration

resistance. The STF/Kevlar composite previously developed, was only able to hold a

zero-penetration up to 11J-cm2/g. It is seen that the use of nanoscale silica particles

mixed with PEG does not fulfill the desired results expected in Shear Thickening

theories. Moreover the use of PEG does not allow the creation of good bonds between

the silica particles and the Kevlar fibers. It has consequently become crucial to find a

new way to improve the resistance to spike impacts.

The study of each component used before to manufacture the STF/Kevlar composite has

revealed that the inclusion of a Silane coupling agent would help increasing the bonds

between the silica particles and the polymer. However, despite an increase in

performance by the addition of the silane, it was observed that modestly higher level of

performance could be achieved. Therefore, by removing the PEG out of the equation of

the mixture, the new composite showed again an increase of performance.

It has then become clear that the use of nanoscale particles along with high bonding

ability chemicals was the way which would lead the resistance performance to a higher

level. Indeed the experiments show that by increasing the bonding between particles

and the Kevlar fibers, the performances in energy dissipation during spike impact has

significantly increased.

In order to achieve higher bonding performances, a new chemical has been

introduced into the mixture. The glutaraldehyde, a type of formaldehyde, creates strong

covalent bonds between the amino groups that cover the surface of the silica

nanoparticles. These amino groups are brought into the system by the addition of silane

10

into the mixture. By increasing the amount of bonds between the particles, the energy

dissipated will increase providing a good resistance to the composite.

By performing a Finite Element Model with ANSYS®, a predictive tool can be

used to determine the depth of penetration when NIJ Spike Tests are performed. It will

also indicate the amount of energy received during impact. This represents an important

tool in composite design as it allows the user to simulate any spike impact knowing the

material properties. This will also help in reducing the number of experiments.

In view of the discussion, it is therefore seen that a system of nanoparticle scale

silica dispersed ultrasonically into a mixture of silane and glutaraldehyde, and then

impregnated with ballistic grade fabric, such as Kevlar and Spectra constitute a

materials system which should be equivalent or better than the currently used PEG

based fabric composites for flexible body armor protection.

11

CHAPTER 2.MATERIALS, EQUIPMENT AND SYNTHESIS

2.1 KM-2 Kevlar® Fabric

The organic fiber was originally developed by DuPont in 1965 for steel belting

in vehicle tires [49]. In the early 1970’s it was considered to be a ballistic grade fabric

after NIJ tests revealed its ballistic resistant potential. This breakthrough started the

interest in new age flexible body armor based on its lightweight and high strength

characteristics. Multiple grades and styles are available and other companies have

produced similar textiles to Kevlar® such as, Twaron®, Spectra®, Zylon® and more

[51].

The Kevlar used in the current investigation is from JPS Composite Materials,

former textile manufacturing branch of Hexcel Schwebel. It is a plain-woven style 706,

also known as Kevlar KM-2 (600 denier). It is a ballistic grade, high performance

textile with an areal density of 180 g/m2 and fabric thickness of 0.23 mm. The

molecular structure of Kevlar, shown in Figure 1, contains a repeating unit of an amide

group and an aromatic ring. These molecular chains, making the polymer structure, are

further connected by hydrogen bonds. The two groups that makeup the repeating units

align parallel to the length of the fiber during the extrusion process and orientate

radially in a spoke like manner, shown in Figure 2. The regularity of the molecular

structure is what is responsible for its impact strength.

12

Figure 1: (a) Molecular structure of Kevlar (b) aromatic ring (c) amide group.

Figure 2: A x-ray linear dichroism microscopic view of the cross section of a Kevlar

fiber showing radial symmetry [52, 53].

2.2 Correctional Kevlar®

KEVLAR® CORRECTIONAL™ [49] was specifically designed to by Dupont

to prevent puncture threats. It is four times thinner than typical ballistic fibers (for an

O C

H

N

H : O

O : H

C

H

O

O

N

O

C O C

O : H

O C N C O

O

N C O C

O O

N

Hydrogen Bonding

Repeating Unit

N C

H O

C C

C C

C C

H H

H H

H H H

13

ultradense weave) and five times stronger than steel on an equal weight basis. As

explained by Dupont the fabric is woven so tightly together, that when struck by sharp

objects, such as spikes, awls or shanks, the fiber absorbs and dissipates the energy of the

puncture or penetration.

Figure 3: Kevlar Correctional fabric under NIJ Spike test. [49]

With similar chemical properties as KM-2 Kevlar®, the Correctional Kevlar®

stands out due to its lightweight characteristic.

14

2.3 Spectra®

Figure 4: Spectra fibers are made out bright white polyethylene

Spectra® fiber, developed by Honeywell [51], is one of the world's strongest

and lightest fibers. A bright white polyethylene, it is, pound-for-pound, fifteen times

stronger than steel, more durable than polyester and has a specific strength that is 40

percent greater than aramid fiber. Spectra® fiber is produced from ultrahigh molecular

weight polyethylene (UHMWPE) using a patented gel-spinning process. UHMWPE is a

remarkably durable plastic, and scientists at Honeywell have captured the tremendous

natural strength in the molecular backbone of this everyday plastic to create one of the

world's strongest and lightest fibers. The gel-spinning process and subsequent drawing

steps allow Spectra® fiber to have a much higher melting temperature (150°C or 300°F)

than standard polyethylene. With outstanding toughness and extraordinary visco-elastic

properties, Spectra® fiber can withstand high-load strain-rate velocities. Light enough

to float; it also exhibits high resistance to chemicals, water, and ultraviolet light. It has

excellent vibration damping, flex fatigue and internal fiber-friction characteristics, and

15

Spectra® fiber's low dielectric constant makes it virtually transparent to radar. Spectra®

fiber is used in numerous high-performance applications, including police and military

ballistic-resistant vests, helmets and armored vehicles, as well as sailcloth, fishing lines,

marine cordage, lifting slings, and cut-resistant gloves and apparel.

Figure 5: Molecular structure for Ultra high molecular weight polyethylene (UHMWPE).

Ultra high molecular weight polyethylene (UHMWPE) is a type of polyolefin,

and, despite relatively weak Van der Waals bonds between its molecules, derives ample

strength from the length of each individual molecule. It is made up of extremely long

chains of polyethylene, which all align in the same direction. Each chain is bonded to

the others with so many Van der Waals bonds that the whole can support great tensile

loads. The yield strength are as high as 2.4 GPa and density as low as 0.97 kg/l

When formed to fibers, the polymer chains can attain a parallel orientation greater than

95% and a level of crystallinity of up to 85%. In contrast, Kevlar derives its strength

from strong bonding between relatively short molecules

16

2.4 Polyethylene Glycol (PEG)

Polyethylene Glycol (PEG) is a water soluble condensation polymer of ethylene

oxide and water with a molecular structure given in Figure 6. The viscosity and melting

temperature increase (although melting temperatures are in the negative Celsius

temperature range) as the molecular weight increases [23, 54]. The current work uses a

low average molecular weight of 200g/mol with a density of 1.1239 g/mL.

Figure 6: Molecular formula for PEG where n denotes the molecular weight. For the current research n=4.2 corresponds to a 200g/mol average molecular weight.

Rheology research of PEG at varied molecular weight and mixed with varied

particle size, shape, type and weight percent [19, 23] has been compiled. Much of the

PEG and particle mixture research is that of Wetzel et al. [23] and the molecular weight

of choice is 200 g/mol. Wetzel et al. showed that when mixed with micron colloidal

silica particles the fluid transitioned from a shear thinning fluid at low shear rates to a

shear thickening fluid at high rates. PEG is used to couple the molecules of the silica

particles to produce non-ionic surfactants and bond to the Kevlar® fabric.

2.5 Silica Nanoparticles

30nm diameter silica particles from Sigma-Aldrich are used in the composite

fabric to give resistance against sharp impactors. They have a surface area in between

140 and 180 m2/g and a density of about 2.2 – 2.6 g/ml [55]. They are prepared using

the chemical vapor deposition (CVD) method utilizing either the plasma enhanced

method or the laser induced method [67]. The synthesis of nanostructured material by

HO – (CH2 – CH2 – O)n – H

17

chemical vapor condensation (CVC) or chemical vapor deposition (CVD) is well

established. In most of these systems, a chemical precursor is evaporated and undergoes

pyrolysis in a reduced pressure atmosphere to generate the nanostructured particles. The

products are then subjected to transport in a carrier gas and collected on a cold

substrate. Several conditions are required for successful processing: a low concentration

of precursor in the carrier gas, rapid expansion of the gas stream through a uniformly

heated reactor, and a rapid quenching of the gas phase-nucleated nanoparticles as they

deposit on the collector walls as shown in Figure 7.

Figure 7: Laser-induced CVC to produce nanostructured SiO2

The preparation of silicon dioxide (SiO2) comes from different sources.

Common sources of precursors include silane and oxygen, dichlorosilane (SiCl2H2) and

Precursor reservoirLaser Beam

Ultra sound nozzle

Laser plume

Cooled collector

Powder deposit To vacuum pump

Carrier gas

18

nitrous oxide (N2O), or tetraethylorthosilicate (TEOS; Si(OC2H5)4). The reactions are as

follows:

SiH4 + O2 → SiO2 + 2H2

SiCl2H2 + 2N2O → SiO2 + 2N2 + 2HCl

Si(OC2H5)4 → SiO2 + byproducts

Those reactions all have advantages and disadvantages in terms of additional

residues but the results produce similar SiO2 nanoparticles which are used into the

mixture.

2.6 Organosilane

In an attempt to further improve the penetration resistances of the composite,

silica particles were functionalized with a silane coupling agent. Such functionalization

was first tested with extruded Nylon 6 filaments in another research project [63]. This

resulted in a significant gain in Young’s Modulus and the Tensile Strength of the

filaments [38, 56]. Such improvements have been confirmed for stab impacts in the

early stage of this project [63]. The terms silated, functionalized and modified particles

will be used through out the text. The three terms are interchangeable and refer to the

addition of the silane coupling agent to the surface of the particles. The organosilane

used in our investigation is an amino-propyl-trimethoxy-silane and was procured from

Gelest®. The molecular formula for silane is

Figure 8: Molecular structure of the silane coupling agent

H

H2NCH2CH2NCH2CH2CH2Si(OCH3)3

19

We have used silane coupling agent to form durable bonds between inorganic

and organic materials, such as the silica nanoparticles and a polymer. The silane used in

our study was extremely effective with silica particles [57]. Molecular structure of

silane is shown below as it forms chemical bonds between the particles and the

polymer.

32 )( XSiCHR n −−−

Figure 9: Molecular structure of the trialkoxysilane coupling agent

Figure 10: Silanol linkages between the polymer and the silica substrate [57].

2.7 Glutaraldehyde

In addition to the organosilane, substances such as glutaraldehyde further

promote bonding between silated silica particles. The glutaraldehyde has been procured

from JT Baker® (222 Red School Lane Phillipsburg NJ 08865 U.S.A.). Our interests in

this chemical are in its property to create strong covalent bonds that bridges one silated

silica particles to another.

Organofunctional Linker Silicon

A Hydrolyzable Groups

20

Figure 11: Molecular formula of Glutaraldehyde

This is possible because monomeric glutaraldehyde polymerizes by aldol

condensation reaction. This reaction occurs at alkaline pH values and in an alcoholic

environment. It is a basic reaction in which an enolate ion reacts with a carbonyl

compound to form strong covalent bonds between two carbon atoms. The reaction is

followed by dehydratation as shown on Figure 7.

Figure 12: Aldol Condensation Reaction

The aldehyde group of the glutaraldehyde links with the silated silica particles

by creating bonds with amino groups as well as other aldehyde groups.

21

Figure 13: Molecular reaction for cross-linking bonding of aldehyde groups with silated silica particles

The impact performances of the fabric get incredibly improved when the

admixture is dry and used on the surface of another material as a coating. The network

is created by several bonds between the nano particles. It is due to the fact that each

single coated yarn is no longer independent but linked to the surrounding yarns through

the strong covalent bonds that the glutaraldehyde is providing. This new bonding is key

to excellent resistance to spike and stab threats.

22

2.8 High Intensity Ultrasonic Liquid Processor

Figure 14: VCX Series Ultrasonic Processor from Sonics [58].

This ultrasonic mixer from Sonics, as seen in Figure 14, converts voltage to high

frequency electrical energy and then to small mechanical vibrations [58]. These

mechanical vibrations are transmitted as ultrasonic waves into a liquid medium through

a titanium probe as seen above. These waves consist of alternate compressions and

rarefactions which create microscopic bubbles that implode causing shock waves. This

occurrence is also known as cavitation. The acoustic cavitation accelerates the

dispersion and emulsification of the mixture and in a liquid-solid mixture which can

alter the surface of the solid component (i.e. nanoparticles) in two ways. The first

causes from the asymmetric implosion due to the restrictions on the bubbles do to

spaces occupied by the solid material. The asymmetric implosion causes the liquid to

impact the surface at high energies. The second fact, which relates specifically with

nanoparticles, is that the cavitation causes the small particles or aggregates to collide at

Digital Processor

Medium

Probe

23

high speeds. Both of these cases erode and expose new surfaces to react further with

the surrounding medium.

2.9 Synthesis of the Silated-Nanoparticles-Glutaraldehyde -Fabric Composites

Actual fabrication procedures, shown in Figure 15, include the silica

nanoparticles dispersed, by sonication, directly into a mixture of Water, Silane and

Glutaraldehyde such that the ratio of water:silica:silane:gluta is about 24:22:10:1 by

weight. The modification of the silica nanoparticles with the use of silane is followed

according to the manufacturer’s procedures and is derived from the surface area of the

nanoparticles. Based on the fact that one gram of silane covers a surface area of 358

m2, silane was added to a 95% ethanol – 5% water mixture to yield a 2% final

concentration of silane. The actual amount of each chemical used is listed in Table 1.

After adding ethanol, the solution is homogenized with a high speed mechanical mixer

which will grind and break up major agglomerates. This 30 minutes step helps in

obtaining a uniform mixture which will be ready for the next step. The addition of

ethanol as a medium aides in the dispersion during sonication to breakup the silica

agglomerations as discussed in section 2.5. After sonication for about three hours, the

mixture was used to soak 12 layers of Kevlar fabric cut in dimensions of 30.48 cm x

30.48 cm (12 in x 12 in). To impregnate the fabric, the layers were placed in a sealed

plastic bag along with the sonicated mixture. The fabric was then let to rest for about

24 hours. Afterward, the fabric layer was placed in the furnace and heated at 110°C

until they were dry, i.e. all the ethanol had evaporated. The 12 layers of Kevlar

impregnated with the silica-silane-glutaraldehyde mixture resulted in an areal density of

24

approximately 0.224 g/cm2 (0.459psf). This fabrication procedure completely bypassed

the heating and centrifugation of the mixture and the addition of ethanol prior to

soaking as compared to the fabrication of STF. In opposition to the recent works, it also

bypasses the drying of the particles prior soaking the fabric relative, as it goes directly

from sonication to impregnation. The current procedure also eliminates the

functionalization of the silica particles prior to mixing with the polymer.

Figure 15: The manufacturing procedures. Sonicating the particles and infusing into the fabric using a sealed bag and finally, oven drying the fabric composite.

Particle Size Specific Surface Area Glutaraldehyde Water Silane Silica 30 nm 354 m2/g 0.2458g 6.022g 2.458g 5.5g Table 1: The amount of glutaraldehyde, water and silane used to functionalize 5.5g of

30 nm silica particles according the manufacturers procedure

25

CHAPTER 3.EXPERIMENTATION

3.1 NIJ Stab Test

The National Institute of Justice (NIJ) was established in 1968 [51]. NIJ

introduced its first standard for body armor performance, NIJ 0101.00 (NIJ101), in

1972, during a trend of increased homicide cases for law enforcement officers. The

publication of NIJ101 occurred during an initial breakthrough of a whim testing of

DuPont’s Kevlar® fabric, which was initially developed in 1965 to replace steel belting

in vehicle tires. It is not until 1976 that an improved Kevlar fabric was concluded to be

an effective ballistic material. This was the result of several tests conducted on the

effects of blunt trauma, comfort and psychological effects (i.e. officer’s confidence in

safety during a threatening situation). They have collaborated with Office of law

Enforcement Standards (OLES), National Law Enforcement and Correction

Technology Center (NLECTC), the U.S. Secret Service and the Police Scientific

Development Branch (PSDB) in the United Kingdom (UK) to develop this standard.

Just recently, in September 2000, NIJ introduced the NIJ Standard 0115.0 (NIJ115) for

stab and puncture resistant body armor. It gives the minimum performance

requirements for body armor to resist a sharp weapon [59]. The focus of the current

research is stab resistance which follows NIJ115. It quantifies three levels of protection

based on the impact energy, in Joules. The levels range from low-level to high-level

protection ranging in measurable impact energies of 24 joules to 43 joules of energy.

26

3.1.1 Test Methodology

This test method consists of a vertical cylindrical tube with a cylindrical drop

mass that can free fall inside the tube. The drop mass, shown in Figure 16, interfaces

with the threat weapon, i.e. knife or spike (Figure 17), and strikes the target, which is

the fabric composite being tested.

Below the fabric composite is the NIJ115 backing material shown in Figure 18,

starting from the bottom, consist of two rubber layers, one polyethylene foam and four

neoprene sponge layers with witness papers (not shown in figure below) in between

each neoprene sponge layer, as well as on top and bottom.

The target, which is placed on top of the NIJ115 backing material and secured

with two straps laid across the target, is impacted at different energy levels by changing

the drop height. At all impact energies the backing material is inspected for damaged

witness papers which results in the measured penetration depth. The standard gives a

certain depth of penetration with respect to the impact energy that is allowable for stab

resistant body armor.

Figure 16: NIJ115 drop mass [59].

27

(a)

(b)

(c)

Figure 17: NIJ115 threat weapons; (a) Engineered Knife Blade P1 (one cutting edge) (b) Engineered Knife Blade S1 (two cutting edges) (c) Engineered Spike [59].

28

Figure 18: NIJ115 Composite Backing Material [59].

3.1.2 Procedure

Kevlar composites were fabricated using the procedures previously mentioned

in 2.9, and were tested using a drop tower, built in-house. The drop tower, modified to

be short in size, was developed to conveniently setup inside and achieves lower impact

energies which maxed out around 16 J (approximately attaining a free fall of one

meter). The drop tower was constructed based on the Stab Resistance of Personal Body

Armor, NIJ Standard-0115.0 (NIJ115).

In addition to the impregnated fabric, the drop tower tests included; NIJ115

backing material and a 2.0kg nylon drop mass with the NIJ115 engineered spike (Figure

17c). The drop heights ranged from approximately 0.05 m to 1.0 m to produce

theoretical impact energies from 1 J to 16 J with an increment of 1 J. The velocities just

prior to impact were also recorded through a laser speed trap. Using the measured

impact velocity, the total mass of the spike and drop mass; the actual impact energy was

calculated.

29

Figure 19: NIJ115 drop tower and system setup; (a) drop apparatus (b) drop mass (c) threat weapon (Spike) (d) velocity measurement zone (e) backing material

Along with the impact energy, the penetration depth is also a factor in

classifying stab resistant body armor. The penetration depth was measured by damaged

witness papers placed immediately underneath the fabric specimen and underneath

consecutive sponge layers that compose the backing material shown in Figure 20. In

Figure 20c, there is a tear and a small hole in the middle. The tear may occur due to the

impact causing the material to deform but does not constitute penetration. Only the

hole counts were considered depth of penetration. The impact energy along with the

penetration depth is used to compare fabric composite performance.

target

witness paper

neoprene sponge

polyethylene foam

rubber

a

b

c

d

e

30

(a) (b) (c)

Figure 20: (a) Kevlar composite after impact at 16 Joules (b) impacted witness paper at various impact energies (c) magnified view of the impacted witness paper.

3.2 Scanning Electron Microscope (SEM)

The improved performance of the Kevlar® composite is attributed to the

infusion of the silated silica and glutaraldehyde mixture coating the fabric. The

distribution over the fabric as well as covering the surface area of the tows can only be

seen by viewing it at a microscopic level. In addition to the distribution of particles, it

is important to see the agglomerations that are expected to occur when dispersing the

silica nanoparticles into the Ethanol. Although the clusters will occur naturally, the

objective is to still maintain uniform dispersion as much as possible. To observe these

features, a Field Emission Scanning Electron Microscope (FESEM) JEOL JSF-6330F

was used. In addition to the coating, the FESEM makes it possible to observe the impact

damage of the fabric composite. The samples were fixed to the stage with copper tape

and were coated with gold. The gold coating prevented charge build-up by the

electrons absorbed by the specimen. It gives the non-conductive specimen electrical

conductivity to reduce the ability to attain an electrostatic charge. This enabled the use

of a higher voltage to increase magnification.

tear

hole

31

3.3 Fourier Transform Infrared Spectroscopy (FTIR)

The Fourier Transform Infrared Spectroscopy uses the infrared spectrum formed

by the absorption of electromagnetic radiation at frequencies that associate to the

vibration of particular groups of chemical bonds from a molecule [60]. FTIR spectrum

versus wavelength absorption presents peaks referring to higher absorption of light

energy necessary for excitation. At these specific wave-numbers, the types of

functional bonds occurring in the specimen are indicated, giving rise to its molecular

structure or the so-called fingerprint [60]. The functional groups assist in the

understanding the stability and chemical bond characteristics in each type of bonds.

The samples were analyzed in powdered form.

3.4 Mechanical Testing

In order to have a better understanding and overview of the performance of

various composites, more specific tests were performed. Mechanical tests based on

tension were conducted in three different configurations. Those configurations depend

on the fabric architecture: the direction along the roll which folds when unrolling the

fabric is called warp and the orthogonal direction in which the yarns are sewed in

between the previous ones is called fill.

Finally a tension test at 45° to the yarn configuration was conducted to measure

the in-plane shear strength.

3.4.1 Testing procedure

32

Samples are cut out from the various composites manufactured, into stripes of

152.4mm x 25.4mm (6in x 1in). The direction of cut corresponds to the direction of the

yarn as mentioned previously: fill, warp or 45°.

Figure 21: Different directions of the testing samples

Figure 22: Samples cut from fabric composites

25.4x25.4 mm2 Carbon/epoxy tabs are glued on the edges of each sample to

allow the clamps of the universal Zwick/Roëll testing machine to have a good grip on

the sample. The tabs are glued with Loctite®, a fast curing epoxy. After twenty four

hours of curing, the samples are ready to be tested. Each sample is placed inside the

universal testing machine equipped with a 50KN load cell.

1in

6in

4in = gage length Clamped zone

Warp Fill Shear 45˚

33

Once inserted and locked into position, the samples are pre-tensioned up to 10N

in order to avoid any slip in the tab and grip. When the pre-tension is attained, the actual

test begins at a uniform speed of 0.2mm/s until the fabric ruptured.

The data recorded from the test performed provide load displacement and stress-

strain curves.

Figure 23: A typical tension test in Zwick

3.5 Flexibility Test

Lee et Al [64] proposed a two-dimensional drape test in order to determine the

flexibility of each composite manufactured. In all cases a 20g weight was used and

encapsulated targets were used as test specimens. The bending angle is reported as a

34

measure of the target flexibility, with larger indicating greater flexibility. The target

thickness at the center of the composite is also measured with a micrometer.

Various composites used in this study were tested to ensure that the composites

developed keep its flexibility throughout the process.

Figure 24: Flexibility test developed by Lee et Al [64]

13.9 cm 1.3 cm

α

Aluminium mass ~20g

Composite

35

CHAPTER 4.RESULTS & DISCUSSION

4.1 Introduction

4.1.1 Functionalized Silica with PEG

Traditionally shear thickening fluid _STF_ reinforced with Kevlar has been used

to develop flexible armor. At the core of the STF-Kevlar composites is a mixture of

polyethylene glycol _PEG_ and silica particles. This mixture is often known as STF and

is consisted of approximately 45 wt % PEG and 55 wt % silica. During rheological

tests, STF shows instantaneous spike in viscosity above a critical shear rate. Fabrication

of STF-Kevlar composites requires preparation of STF, dilution with ethanol, and then

impregnation with Kevlar. The nanoscale silica particles were dispersed directly into a

mixture of PEG and ethanol through a sonic cavitation process. Two types of silica

nanoparticles were used in the previous investigation: 30 nm crystalline silica and 7 nm

amorphous silica. The admixture was then reinforced with Kevlar fabric to produce

flexible armor composites. From this point, further improvements have been made: the

silica particles were functionalized with a silane coupling agent to enhance bonding

between silica and PEG. The performance of the resulting armor composites improved

significantly. As evidenced by National Institute of Justice spike tests, the energy

required for zero-layer penetration _i.e., no penetration_ jumped twofold: from 12 to 25

J-cm2/ g which can be observed in Figure 24.

36

Figure 25: NIJ Spike test of STF based composites [63]

The source of this improvement has been traced to the formation of siloxane Si-

O-Si bonds between silica and PEG and superior coating of Kevlar filaments with

particles [63]. In summary, the experiments have demonstrated that functionalization of

silica particles followed by direct dispersion into PEG resulted in superior Kevlar

composites having higher spike resistance.

4.1.2 Removal of Polyethylene Glycol (PEG)

Although the performances of the STF-based Kevlar obtained were improved,

the stab resistance was not sufficient enough to stop a stab impact in which the energy

level would attain 45J/g/cm² or more. Also the idea of Shear Thickening is STF

Neat Kevlar

Sonicated PEG/Kevlar

Sonicated 7nm Silated silica PEG/Kevlar

Sonicated 7nm silica PEG/Kevlar

Sonicated 30nm silated silica PEG/Kevlar

STF Kevlar (Decker et Al)

Sonicated 30nm silica PEG/Kevlar

37

composite during impact did not seem practical since the fabric composite was already

dry. In absence of a fluid, shear thickening and formation of hydroclusters are

unrealistic. On the other hand, it became evident that increasing the bonding strength in

between the particles as well as with the fabric itself was more appropriate in order to

stop a spike or knife impact. The intent of the current investigation was to come up with

appropriate components of the armor to maximize stab performances. The outcome is a

completely new approach regarding the fabrication route and constituents. The new

fabrication route does not use PEG as one of the constituents. It has been observed that

the PEG was reacting with the silane coupling agent which was leading to a decreased

number of bonds between the particles and the fabric as the silane role in this particular

reaction was to link the silica particles with the fabric. High energy dissipation has been

proved to have a good correlation with resistance performances. The energy dissipation

mechanism reaches higher level when the number of bonds increases. By taking the

polyethylene glycol out the system, it is observed that the number of bonds actually

increases and higher energy level was achieved for stab impacts as shown in Figure 26.

38

Figure 26: NIJ Spike test of STF based composites with and without PEG [65]

Original STF/Kevlar curve shows “zero-layer” penetration up to 12J-g/cm2. It

rapidly reaches the depth of 3 layers as the energy increases. The introduction of

functionalization with the silane coupling agent clearly shows an improvement compare

to the STF/Kevlar composite. The energy level reached before penetration goes up to

33J-g/cm2. The introduction of silane helps binding the polymer to the silica substrate

which increases the number of bonds. The examination of the present result led us to

believe that PEG was not conducive to the formation of bonds that functionalization

intends to do. Therefore the PEG has been removed and tests results exposed on the

curve (Silane-silica/Kevlar) show a significant increase in energy dissipation: the “zero-

layer” penetration goes up to 43J-g/cm2. Consequently increasing bonding is the goal

that we are trying to achieve in order to increase the “zero-layer” penetration.

NIJ Spike Test

0

1

2

3

0,00 10,00 20,00 30,00 40,00 50,00 60,00Normalized Energy (J/g/cm2)

Laye

rs

PEG-Silane-Silica/Kevlar

STF

Silane-SiliSilica/Kevlar

39

4.1.3 Silane-Silica-Glutaraldehyde Systems

As mentioned previously, the Glutaraldehyde targets essentially amino-groups

in which it will create strong covalent bonds. The Glutaraldehyde is then incorporated

according to the amount of silane agent present in the solution. Different types of silane

can be used but the key component within the silane which determines the amount of

glutaraldehyde, is the amino-groups. The amino-groups are functional groups that

contain a basic nitrogen atom. In our case, the silane possesses di-amino-groups so the

Glutaraldehyde can make the link in between the different amino-groups of the silane.

Kevlar fabric also has amide groups which are believed to connect with the admixture

when in presence of glutaraldehyde.

As the silane attaches to the silica particles included into the mixture, it results

in the creation of a particle network or a mesh in which the glutaraldehyde acts as the

linker. Performances of the composite show much higher resistance to penetration than

any previous composite systems.

4.2 NIJ Stab test

4.2.1 NIJ Spike test

According to the NIJ standard, a series of spike impact tests have been

performed to determine the level of resistance to penetration for the various composites.

Those series of tests clearly shows how well the silane-silica-glutaraldehyde system

performed under a puncture threat.

40


(density:0.236 g/cm2)

Silane-Silica/Kevlar(density:0.217

g/cm2)

Silane-Silica-Gluta/Kevlar


Failure of the backing material

PEG-Silica(STF)/Kevlar


0

1

2

3

0,00 50,00 100,00 150,00 200,00 250,00 300,00

Normalized Energy (J-cm2/g)

Witn

ess

Lay

ers

Figure 27: NIJ Spike test of Kevlar based fabrics [65]

The impact energies were normalized by dividing them by the areal density of

the tested fabric composites and plotted against the penetration depth, as seen in Figure

26. It is noticed that with the regular STF/Kevlar composites the spike penetrates the

fabric after 12 J-cm2/g. It is quickly followed by the penetration of the second layer at

18J-g/cm2 and finally reaches the third layer at 63 J-cm2/g. As seen previously, it is

important to notice the removal of PEG out of the equation which brought the PEG-

Silica-Silane/Kevlar to Silane-Silica/Kevlar from 33 J-cm2/g to 43 J-cm2/g.

The results of the spike test shown on Figure 27 clearly demonstrate the benefit

of addition of glutaraldehyde into the composite structure. The performances obtained

with the Silane-Silica-Gluta/Kevlar reaches a significant energy level. The “zero-layer”

penetration is brought up to an energy level of 211 J-cm2/g. It is even more impressive

as above that energy level, the composite is still not failing. However the backing

41

material starts failing and gets penetrated by the spike and the fabric reaching 2 layers

of penetration at around 240 J-cm2/g as seen in Figure 27.

The gluta-system clearly shows high improvements compared to the STF fabric

that has been already improved by removing the PEG. The ‘0-level penetration’

increases by 4.5 times the Silane-Silica/Kevlar composite and by 10 times than that of

STF system.

Figure 28: Failure of the backing material after reaching higher energy level during NIJ

Spike test

The gluta-system is consequently a major improvement in this research leading

to the manufacture of a flexible body armor. Such increase in energy level for the “zero-

layer” penetration is made possible by the addition of glutaraldehyde which increases

the number and quality of bonds between the different constituents of the composites.

4.2.2 NIJ Knife test

Once the spike tests were completed, NIJ Knife tests were performed. These

tests were conducted in identical manners of those of spike tests.

42

The knife impact is considered to be more penalizing threat for the manufacture

of the composites. It combines both puncture and shear when impacting the fabric.


10mil-COEX-Kevlar 6layers

3mil-PE-Kevlar 10layers Neat/Kevlar

0

1

2

3

4

5

0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 8,00 9,00 10,00Normalized Energy (J/g/cm2)

Laye

rs

Figure 29: NIJ Knife test graph

Investigations regarding the knife tests showed interesting results. In figure 29 it

is shown that neat Kevlar as well as silane-silica-gluta/Kevlar composite get penetrated

directly after impact and do not hold any “zero-layer” resistance to the knife. The

difference between the reinforced fabric and the neat fabric stands on the penetration of

layers one to three as the composite system dissipate more energy at those layers than

neat Kevlar. For neat Kevlar the penetration of the first two layers is achieved at 2.21 J-

cm2/g compare to the gluta system which energy level for the first layer is obtained at

3.66 J-cm2/g and the second layer at 5.61 J-cm2/g. The neat Kevlar linearly fails from

43

the second to the fourth layer at about 6.55 J-cm2/g, and the silane-silica-gluta/Kevlar

composite fails at 7.64 J-cm2/g.

Knife results obtained from Tuskegee University show slight improvement in

stab resistance as the 6 layesr of 10mm-COEX-Kevlar and 10 layers of 3mm-PE-Kevlar

hold the “zero-layer” penetration up to around 0.65 J-cm2/g. For the first one the knife

goes straight through reaching the fourth layer at the energy level of 3.36 J-cm2/g. The

3mm-PE-Kevlar composite reaches the third layer at 3.13 J-cm2/g where the energy is

confined at this layer for up to 3.65 J-cm2/g and then reaches the fourth layer at the

energy of 4.13 J-cm2/g showing slight improvements compare to the 10mm-COEX-

Kevlar composite.

The observation of the results makes evident the fact that the newly developed

Gluta System with Kevlar cannot offer any knife stab resistance as the “zero-layer”

penetration is not achieved at any time.

To address this problem, it is believed that enlarging the spectrum of fabrics

used would help reaching higher energy level. Therefore other fabrics have been

introduced such as Spectra® and Kevlar Correctional®.

4.2.3 Studies of various fabric

In order to obtain better performances regarding the knife impact but also to

reduce the areal density of the composite for the spike test, two new fabrics have been

introduced: the Correctional Kevlar® and the Spectra®. The density of the correctional

Kevlar® is much less than the Kevlar KM-2®, 132g/m2 to 180 g/m2. It also has a lot

more yarns with 70 in both directions to 34 for the Kevlar KM-2.

44

The introduction of Spectra® has shown significant improvements in stab and

ballistic resistance [50]. It is already used for making bulletproof vests as well as anti-

cut working gloves. We are expecting that the incorporation of the glutaraldehyde-based

mixture to those fabrics will show improvements regarding spike and knife impacts.



Silane-Silica/Kevlar(density:0.217

g/cm2)




PEG-Silica(STF)/Kevlar


Silane-Silica-Gluta/Correctional

Kevlar(density:0.158

g/cm2)

Silane-Silica-Gluta/Spectra (density:0.255

g/cm2)

0

1

2

3

0,00 50,00 100,00 150,00 200,00 250,00 300,00 350,00

Normalized Energy (J-cm2/g)

Witn

ess

Lay

ers

Figure 30: NIJ Spike test for various fabrics

Results of the NIJ spike tests are shown in Figure 30. The various composites

performances are plotted for comparison but we will mainly focus on the performances

of the Spectra and correctional Kevlar composites. The spectra fabric infused with

silane, silica and glutaraldehyde presents a “zero-layer” penetration up to 22.4 J-cm2/g

reaching the first layer penetration at 30.5 J-cm2/g. This drop of performance, in

comparison to the Kevlar composite performance, is attributed to the fabric itself. It

seems that the mixture used to impregnate the fabric does not bond as well as on

Kevlar. This is confirmed by the examination of the Correctional Kevlar performances

as it achieves the “zero-layer” penetration up to 232 J-cm2/g. The high energy level is

45

obtained for two reasons: the use of Kevlar fabric in the composite and the low density

of the neat Correctional Kevlar fibers: 0.152g/cm2. Similarly to conventional Kevlar

KM-2, the Correctional Kevlar composite does not fail. It reaches an energy level in

which the backing material starts to fail and progressively reaches the two layers

penetration.

NIJ Knife tests have been performed with Correctional Kevlar and Spectra

based composites. The results of their performance are given below in comparison with

the knife tests completed earlier.

Silane-Silica-Gluta/Spectra



KevlarNeat/Spectra


3mil-PE-Kevlar 10layers

0

1

2

3

4

5


Laye

rs

Figure 31: NIJ Knife test for various fabrics

The knife test results of Figure 31 show poor resistance to stab impacts for the

various composites. Silane-Silica-Gluta systems with Kevlar or Correctional Kevlar

composites do not show any resistance to knife penetration. The 10mm-COEX-Kevlar

46

and 3mm-PE-Kevlar from Tuskegee University obtain some resistance with a “zero-

layer” penetration up to 0.65 J-cm2/g. The observation of neat spectra show that “zero-

layer” penetration is also not achieved as it goes straight to the second layer for an

energy level of 3.5 J-cm2/g. The energy dissipation stays on the second layer level for

about 1.5 J-cm2/g but quickly goes up to the third layer at 6.8 J-cm2/g. On the third layer

a plateau is observed until the energy dissipation reaches 11.7 J-cm2/g, after which the

composite fails to the last layer at 14.7 J-cm2/g. This plateau shows the significance of

Spectra regarding stab impacts. When infused with Silane, Silica and Glutaraldehyde

the resistance of the spectra based composite increases significantly. The “zero-layer”

penetration is achieved up to an energy level of 6.4 J-cm2/g. The composite fails to first

layer at 9.9 J-cm2/g and rapidly to the second layer at 12.4 J-cm2/g. The plateau

observed with the neat spectra at the third layer is now shortened with the reinforced

spectra as it only goes up to 17.1 J-cm2/g. The composite finally fails to the fourth layer

at 19.4 J-cm2/g.

The examination of Correctional Kevlar composite results shows that it does not

provide any resistance to a stab impact. The composite fails and the different layers are

linearly penetrated at 2.6 J-cm2/g, 5.2 J-cm2/g, 7.7 J-cm2/g and 9.9 J-cm2/g. This

indicates clearly that the choice of the fabric is crucial in the protection of armor

composite. It is believed that Spectra based composites will provide good resistance to

stab impacts whereas Kevlar based composites will demonstrate significant

performance to spike impacts.

47

From this statement a straightforward conclusion has been made: hybridization

of composite i.e. mixing Spectra layers and Kevlar layers would offer the best

compromise in the resistance to both puncture and stab threats.

4.2.4 Hybridization of Kevlar and Spectra fabrics

The concept of hybridization comes from the intention to provide an armor

composite which would offer reasonable resistance to both puncture and stab threats. In

this study we will attempt to optimize one single composite in which both advantages

will be present. Hybridization of a mixing of Kevlar layers with Spectra is considered.

The goal of optimization is to determine the adequate combination of Kevlar and

Spectra layers in order to keep a good resistance to Spike as well as to Knife resistance.

Several series of tests have been performed in order to determine the right

combination. The tests are performed with 12 layers composites. The hybridization

combinations vary from Spectra to Kevlar or from Spectra to Correctional Kevlar. It

starts from 12 layers of Spectra and when a new composite is manufactured one layer of

Spectra is replaced by one layer of Kevlar or Correctional Kevlar keeping the layer

count to 12 layers.

Various stacking sequence have been operated. The first stacking sequence to be

performed was an alternate combination of Kevlar and one layer of Spectra. Further

stacking sequences have been tried, such as having all the Kevlar layers on top and all

the Spectra layers at the bottom and reversely. Those stacking sequences have

demonstrated no variation in results to spike or knife penetration.

48

Silane-Silica-Gluta/[12]Spectra

Silane-Silica-Gluta/[12]Kevlar

Silane-Silica-Gluta/[12]Correction

al Kevlar

Silane-Silica-Gluta/[7]Correctional Kevlar - [5]Spectra

Silane-Silica-Gluta/[8]Kevlar -

[4]Spectra

0

1

2

3

4

0,00 50,00 100,00 150,00 200,00 250,00 300,00 350,00Normalized Energy (J/g/cm2)

Laye

rs

Figure 32: NIJ Spike test for determination of the best hybrid composite

Results in Figure 32 show the determination of the number of layers useful for

an hybrid composites to be resistant to a spike impact. The results of Silane-silica-

glutaraldehyde infused with Spectra, Kevlar and Correctional Kevlar are taken as

element of reference in order to compare with the newly hybrid composites tested.

During the tests two hybrid composites showed significant performances in spike

penetration. The Silane-silica-glutaraldehyde/[8]Kevlar-[4]Spectra composite hold the

“zero-layer” penetration up to 32.4 J-cm2/g but rapidly fails to the fourth layer at 33.7 J-

cm2/g. The silane-silica-glutaraldehyde/[7]Correctional Kevlar-[5]Spectra composite

also achieves a “zero-layer” penetration up to an energy level of 37.6 J-cm2/g. Over this

energy level the spike goes straight through all four layers at about 40.9 J-cm2/g. The

addition of Spectra layers into the hybrid composite increases the density of the entire

49

composite which is why the performances of the hybrid composites are not as high as

for Kevlar based composite.

In figure 33, the results of NIJ Knife tests of Hybrid composites are shown.

Silane-silica-gluta/[12]Spectra

Silane-silica-gluta/[12]Kevlar

Silane-silica-gluta/[12]Correctional

Kevlar

Silane-silica-gluta/[6]Correctional

Kevlar-[6]Spectra

Silane-silica-gluta/[3]Kevlar-

[9]Spectra

0

1

2

3

4

5


Laye

rs

Figure 33: NIJ Knife test for optimum determination

Previous results of Spectra, Kevlar and Correctional Kevlar composites are

shown as references to compare the hybrid composites. The Silane-silica-

glutaraldehyde/[6]Correctional Kevlar-[6]Spectra composite presents a “zero-layer”

penetration zone up to 2.1 J-cm2/g. The composite fails to the second layer at the energy

level of 6.5 J-cm2/g. It is confined on the second layer for about 1.6 J-cm2/g and then

reaches the third layer at 10.2 J-cm2/g. After 1.8 J-cm2/g on the third layer, it finally

penetrates to the fourth layer at 14.2 J-cm2/g.

50

The silane-silica-glutaraldehyde/[3]Kevlar-[9]Spectra also shows a “zero-layer”

penetration up to the energy level of 1.8 J-cm2/g. Similarly to the previous hybrid

composite, the knife goes straight to the second layer at the energy level of 5.4 J-cm2/g.

The hybrid composite dissipates the impact energy on the second layer to about 8.7 J-

cm2/g, then reaches the third layer at 10.6 J-cm2/g. A plateau is observed on the third

layer dissipating the energy up to 16.7 J-cm2/g which is very close to the level achieved

by the twelve layers of spectra composite. The fourth layer is reached at the energy

level of 19 J-cm2/g.

The constitution of the hybrid is made by taking the least number of layers from

each fabric determined during the NIJ tests. It has been concluded that Kevlar based

composite were providing the spike resistance whereas the Spectra based composite

were providing the knife resistance. With the NIJ Spike test seven layers of Kevlar were

sufficient to obtain a “zero-layer” penetration up to 37 J-cm2/g. In the NIJ knife test, 8

layers of Spectra showed a “zero-layer” penetration up to 2 J-cm2/g. In order to obtain

both advantages the 7 layers of Kevlar and 8 layers of Spectra will be mixed producing

a 15 layers composite which differs from the usual 12 layers composite. Both NIJ Knife

and Spike are performed in Figure 34 and compared with known results from Kevlar

composites and Spectra composites.

51

0

1

2

3

4

5


Laye

rs

Knife: Silane-Silica-Gluta/Hybrid

Spike:Silane-Silica-Gluta/Spectra

Spike: Silane-Silica-Gluta/Kevlar

Knife: Silane-Silica-Gluta/Spectra

Knife: Silane-Silica-Gluta/Kevlar

Spike: Silane-Silica-Gluta/Hybrid

Figure 34: Knife/Spike performance for a 15 layers Kevlar/Spectra Hybrid

Several tests have been conducted in order to obtain the best compromise to

manufacture the hybrid. Those results are put in annex and the methodology is

explained for the NIJ Spike tests on Correctional Kevlar/Spectra determination. To

quickly reference the composite, a code name has been given: C-Kevlar stands for

Correctional Kevlar and SPC for Spectra. The respective number of layers is mentioned

in brackets such as [11]C-Kevlar/[1]SPC. A precision on whether the hybrid is neat or

composite is added to the title [11]C-Kevlar/[1]SPC/NEAT.

Here is an example on how to read the following tables and the ones in annex.

[#Layers]Fabric1/[#Layers]Fabric2

Dimension (in2) Layers Mass

(grams)

Areal Density

Velocity Depth of

penetration (#Layers)

(psf) (g/cm2) Energy (J) Normalized

Energy (J-cm2/g)

52

On the tests recorded in the table 2, it is possible to retrieve the data from which

the determination of the hybrid composite has been possible. The velocity are recorded

and then computed to determine the amount of energy. This energy is normalized by the

areal density in order to compare all the results from different fabrics which would have

had more or less density.

As an example in the table below we can clearly see how important is the

Spectra in the hybrid composite. The more layers the hybrid possesses, the higher

energy it reaches on NIJ Knife test. The removal of one Spectra layer and replacement

with a Correctional Kevlar layer instead decrease slowly the resistance of the hybrid to

the stab threat.

For instance the [1]C-Kevlar/[11]SPC/composite goes up to an energy level of

about 18.99J-cm2/g whereas the [11]C-Kevlar/[1]SPC/composite only goes up to 10.74

J-cm2/g.

By doing so we were able to determine the right number of layers from each

fabric to produce the hybrid composite.

4.2.5 Introduction of CaCo3

It has been shown that resistance to spike and stab threats were in part obtained

by the addition of nanoparticles. Up to now, various forms of silica nanoparticles have

been utilized. But only silica has been utilized in order to correlate with the STF theory.

As the STF has been discarded throughout the investigation, introduction of new

53

nanoparticles has been explored and calcium carbonate has been found to be a good

candidate as these particles are biodegradable.

Calcium carbonate particles were purchased from Reade, Winnofil®, now part

of Solvay Chemicals (Solvay Chemicals, Inc., Headquarters 3333 Richmond Avenue,

Houston, Texas 77098), the particles have around a 100nm diameter size and a high

specific surface area: about 20m2/g.

Figure 35: Nanoparticles of CaCO3

54

The CaCO3 particles were then replaced with SiO2 particles into the mixture.

Keeping the same ratios for the components of the mixture, similar tests have been

performed with the new manufactured composite.


density:0.236g/cm2

Silane-Silica/Kevlar

density:0.217g/cm2


density:0.224g/cm2


PEG-Silica/Kevlar density:0.268g/cm2


Kevlar density:0.158g/cm2


density:0.255g/cm2Silane-CaCo3-

gluta/Kevlardensity:0.226g/cm2

Silane-gluta/Kevlar

Gluta/Kevlar

0

1

2

3

4


Laye

rs

Figure 36: NIJ Spike test of CaCO3 based composite

In figure 36, it can be seen that Silane, Calcium carbonate and glutaraldehyde

infused with Kevlar shows a very similar response to spike impact as the Silane-Silica-

Gluta/Kevlar composite. The calcium carbonate composite achieves the “zero-layer”

penetration up to an energy level of 224.5 J-cm2/g. Over this energy level, the backing

material starts to fail following the same trend as before with the Correctional Kevlar

reaching the second layer at 280.6 J-cm2/g. This differs completely the STF theory as

silica particles are no longer used, but replaced with calcium carbonate particles.

55

Samples of Kevlar impregnated with glutaraldehyde only and with silane and

glutaraldehyde have been tested in order to better understand the implication of the

glutaraldehyde in the composite manufacture. Both composites do not show any

resistance to spike impacts. The spike goes directly through all four layers at an energy

level of 12.2 J-cm2/g for the gluta/Kevlar composite and 17.5 J-cm2/g for the silane-

gluta/Kevlar composite.

Figure 37 shows NIJ Knife test results for Silane-calcium carbonate-

glutaraldehyde/Kevlar composite.


Silane-Silica-Gluta/Kevlar Silane-Silica-

Gluta/Correctional Kevlar

Neat/Spectra


3mil-PE-Kevlar 10layers

Silane-CaCO3-Gluta/Kevlar

0

1

2

3

4

5


Laye

rs

Figure 37: NIJ Knife test of CaCO3 based composite

The silane, calcium carbonate and glutaraldehyde with Kevlar composite shows

pour resistance to stab impacts as it does not achieve the “zero-layer” penetration. The

penetration goes straight to the first layer at 3.2 J-cm2/g and gradually goes through all

56

the layers respectively at 4.7 J-cm2/g for the second, 5.7 J-cm2/g for the third and 6.7 J-

cm2/g for the fourth.

4.2.6 Failure Analysis

A failure analysis has been performed in order to establish the differences

between the knife and the spike penetration. To observe the fundamental differences

between those two modes of failure, some electron microscopy pictures have been taken

of the composites after impact. The fabrics studied are Kevlar reinforced with silane-

silica-glutaraldehyde and the Spectra with the same mixture. In all cases, the first layer

of each composite has been chosen to perform the analysis. Therefore a higher level of

damages can be observed which would lead to a better understanding of the failure

mechanisms involved during those impacts.

57

Figure 38: Spike penetration in Spectra composite

In figure 38, it is possible to see that during the spike penetration, the fabric

seems to have its matrix damaged when dissipating the energy received. We can

observe several blocks around the impacted point, but no failure seems to be present.

58

Figure 39: Damaged matrix in Spectra composite

In figure 39 it is clearly shown that fibers get damaged together to form a single

block.. During an impact the energy received is sufficient enough, around 20J-cm2/g to

produce by dissipation on one single fiber, the matrix is getting damaged.

59

Figure 40: Spike penetration in Kevlar composite

The impact penetration in Kevlar composite differs significantly from the

penetration mechanism happening in Spectra composite. In figure 40, the fibers are

pushed on the side and pulled out from their original position. During impact, the fibers

are pulled out and reinforcement gets torn up. It is clearly seen that the reinforcement is

the key in spike penetration as it keeps the fibers together.

Figure 41 shows the fibers reinforcement to be torn up as it gets penetrated.

60

Figure 41: Gluta reinforcement being torn up during spike penetration in Kevlar

composite

In figure 41 the reinforcement impregnating the Kevlar fibers gets peeled off the

fibers as the fibers get penetrated by the spike. The penetration forces the different

fibers to separate from each other inducing the gluta reinforcement to have its bonds

broken as seen in Figure 41. The energy received is also very high, to the order of 200J-

cm2/g, which implies that the dissipation through bonds breaking between fibers is a

significant mechanism in the spike resistance of the composite.

61

Figure 42: Breakage of the reinforcement between fibers during spike penetration in

Kevlar composite

Figure 42 clearly shows the breakage happening during impact. The

reinforcement is peeled off the fibers as they are moved by the spike. The mechanisms

of failure for the spike penetration is more than likely associated with the material

properties for the Spectra composite and the bonding strength for the Kevlar composite.

Those mechanisms observed during the spike impact are expected to be present in the

knife impact. However due to the results obtained during the several tests performed,

the knife penetration seems to have a completely different mechanism. The results show

62

that Spectra performs better than Kevlar. Study of the failure mechanism should give an

indication regarding this wide variation in performance.

Figure 43: Knife penetration in Spectra composite

Figure 43 shows a good picture of a knife penetration. We can see the puncture

point performed at impact on the top of the opening. Lower to this triangular shape

impact, it is possible to see the cut done by the blade of the knife. Similar comments can

be done regarding the triangular impact hole observed at the top of the penetration.

However the cut obtained by the blade shows a different mechanism. The fibers seem to

be cut and not moved or pulled out. A zoom in to this area shows that the fibers have

63

both mechanisms involved. The energy received is dissipated through either a matrix

damage as observed in the bottom of Figure 44, or a fiber cut at the top.

Figure 44: Zoom in the area where the cut is done by the blade in the Spectra

composite

The examination of the fibers cut shows that an important extension occurred

before rupture. This can be seen on the fibers in figure 44 where a thin piece of fabric is

curling at the end of the fibers showing reduction of cross section occurring during

tension and which happened during the impact. A closer look at those fiber cut can be

seen in figure 45.

64

Figure 45: Cut profile of one fiber in the Spectra composite

The cut observed on one fiber of the Spectra composite indicates that the cut has

been taking place with an angle. It is possible to see the different lamellae plans that

have been cut. The coating is nearly inexistent which indicates a possible ability to not

adhering onto the fibers of UHMWPE.

The performances of the Spectra still stands up compare to the Kevlar

performances. In figure 46, the cut of a knife impact can be observed on a Kevlar

composite.

65

Figure 46: Knife penetration in Kevlar composite

The impact penetration observed in the Kevlar composite differs from the

Spectra composite. It has a triangular shape where most of the fibers seem to be cut by

the blade.

A better visualization of the cut that is occurring in the yarns is given in Figure

47.

66

Figure 47: Yarn cut in Kevlar composite

Contrary to the Spectra composite, the Kevlar composite shows a cleaner

straight cut as seen in figure 47. The fibers are all cut at the same time and do not show

any type of deformation behavior such as the one seen with Spectra fibers. Figure 48

confirms the fact that each individual fiber fails quickly after impact.

67

Figure 48: Fiber cut in Kevlar composite

It appears that during impact the fiber is being cut without offering much

resistance, not a very high elastic deformation. The figure shows that the fiber keeps its

dimensions until the area of impact where a swelling is observed at the end of the fiber.

This can be attributed by the elastic reaction that occurs after failure: the fiber retracts

on itself forming those little bulbs at the tip of the cut fiber. The reinforcement after

impact is still well present which indicates that the failure mode occurring during knife

impacts are mainly due to the fabric properties rather than the present reinforcement.

68

To summarize the analysis, we can explain that after the observation of the spike

and knife penetration in both cases: Kevlar and Spectra composites. The mechanisms

operating during the penetration is slightly different in each case and would then

explain the wide variation happening in resistance performances. During spike

penetration the Kevlar composite shows high resistance to the penetration due to the

reinforcement bonding to the fibers. On the other hand, Spectra shows more difficulties

for the mixture to bond onto the fibers and would then explain the drop of

performances. Also when impacted the energy dissipating into the fabric is damaging

the matrix that can be shown around the impact location. This mechanism is also

occurring during knife penetration, but it is believed to that the increase in resistance to

knife for the Spectra composite compared to the Kevlar composite is mainly due to the

deformation of each fiber. The fibers get stretched and in that way are slowing down the

fall of the knife. This can be correlated with some mechanical testing performed later

on, where tension in shear direction shows high strain for the Spectra composite. On the

opposite, the Kevlar composite, show nearly no resistance to the knife penetration.

Despite the reinforcement the knife cut all the fibers in a clean cut with no other

mechanism involved.

4.2.7 Difference between Spike and Knife impact

The wide variation observed between the spike and the knife performances with

Kevlar and Spectra composite are believed to be mainly due to both the ability of the

bonding of the reinforcement and the fabric material properties. The question occurring

after the NIJ tests and mechanical testing performed on the Spectra and Kevlar

69

composites is: why Kevlar is performing better under spike impact than Spectra and

reversely? To answer this problem further microscopy and material properties studies

have been completed. Figure 49 shows the different scenario of penetration for spike

and knife impacts.

Figure 49: Scenarios of impacts on fabric

70

Figure 50 shows the silane-silica-glutaraldehyde bonding onto the Kevlar.

Figure 50: A thin coating of silated SiO2 with glutaraldehyde mixture on the surface of

the Kevlar fabric

The reinforcement is well attached to the fibers and a uniform coating is

provided onto the fibers. Few agglomerates are presents and bonding between fibers is

observed. It is believed that such bonding is providing the resistance to the spike

penetration for the Kevlar composite. The study of single reinforced fibers shows that

the coating is not only deposited on the surface of the fabric but is uniformly

71

encompassing each single fiber with the mixture. This can be clearly observed in Figure

50.

Figure 51: Silane-Silica-Gluta/Kevlar bonding

On the other hand, the impregnation of Spectra fabric with the silane-silica-

glutaraldehyde mixture does not seem to bond as well. Figure 51 shows that the

reinforcement coated onto the Spectra fibers does not bond uniformly. Large areas of

fibers are not covered and more agglomerates are present. The study based on the NIJ

Spike test show that both neat Kevlar and neat Spectra have similar performance.

However after the impregnation of silane-silica-glutaraldehyde onto the Kevlar fabric

the resistance to the spike penetration is significantly improved. This is simply not the

case regarding the Spectra fibers which still show poor performances after being

treated. The observation of the coating of the reinforcement onto the fibers is an

explanation to this variation in spike resistance and is further explained by the

72

difference in terms of material composition. The Kevlar fabric shows an amide group

which is believed to link with the glutaraldehyde. Spectra fibers made out of UHMWPE

do not have such goup, therefore the bonding to the fibers is not occurring and only

connections in between silica particles are present. Those connections help the

reinforcement to surround the spectra fibers but do not hold the energy received during

impact as the particles will wear off after impact as seen in Figure 51.

Figure 52: Coating of the silated SiO2 with glutaraldehyde onto the surface of the

Spectra fabric.

73

Figure 53: Coating of the silated SiO2 with glutaraldehyde mixture wearing off the

surface of the Spectra fabric after impact.

The presence of reinforcement deposited onto the Spectra fibers still helps

improving the performances of the composite. The examination of the NIJ knife test

results shows that reinforced Spectra fabric performs better than neat Spectra. This

believed to results in part from the reinforcement deposition at the surface of the fibers.

Even though the reinforcement does not bond to the fibers the thin coating existing

provides the first line of defense to the knife impact. Further studies on NIJ knife test

show that Spectra presents at the third layer penetration, a plateau of resistance which

74

differs greatly from the Kevlar composite. The explanation to this variation can be

found in the material properties of the fabrics.

The examination of the Shear moduli of the fabrics shows great differences.

Spectra fibers have a shear modulus of 25.57 MPa compare to Kevlar fibers which only

have a shear modulus of 0.146MPa. This variation explains the difference observed

during the mechanical tests performed as well as the difference in knife resistance. The

resistance of Spectra is then believed to be due to the reinforcement addition but

moreover by its mechanical properties which demonstrate higher shear moduli than the

Kevlar fabric.

Fiber Density (ρy) Kg/m3

Yield strength

(σymax) GPa

Elastic Modulus (Ey)

Gpa

Maximum strain (εy) %

Shear Modulus (G) GPa

Kevlar 1440 2.9 74 3.4 0.15 Spectra 970 3.25 116 2.9 27

Table 2: Table of major material differences between Spectra and Kevlar fibers

4.3 Microscopy

4.3.1 Silane-Silica-Gluta/Kevlar Microscopy

Extensive microscopic studies were performed to observe failure mechansms

and coating of fibers by the particles. This coating is present both at the top and bottom

of the fabric encompassing the entire area of the laminate as shown in figure 53. It is

believed that this coating offers the first line of resistance during the spike penetration.

75

The coating consists of agglomerated silica particles embedded in the body of the

matrix as in Figure 55.

Figure 54: A thin coating of the silated SiO2 with glutaraldehyde mixture on the surface

of the Kevlar fabric

76

Figure 55: Agglomerated Silated SiO2 particles and Glutaraldehyde mixture.

The agglomerated particles are relatively large in size. They are formed due to

the presence of silane and glutaraldehyde. Because of the high concentration (22%) of

the silica particles, they could not be dispersed fully within the matrix. The

agglomerated particles are in the micrometer range. Once this thin layer is penetrated,

the subsequent resistances are offered by the impregnated fiber tows. It is seen in

Figure 42 that a large number of agglomerated particles are adhering to the fiber tows

especially in the region where they are bonded with the neighboring tows. The

presence of the particles at this inter-tows area also offering resistance should the spike

penetrate through the coating. Therefore, it is observed that the mixture of Silated

77

Silica with Glutaraldehyde incorporate multiple phases of resistances on to the Kevlar

fabric.

4.3.2 Evolution of bonding

We can now notice the evolution of the bonding of the nanoparticles to the

Kevlar fibers showing then that the gluta system has more bonding ability than previous

STF mixtures.

Figure 56: Neat Kevlar.

78

Figure 57: Silane-Silica-PEG/Kevlar bonding

Figure 58: Silane-Silica-Gluta/Kevlar bonding

79

Figure 59: Silane-Calcium Carbonate-Gluta/Kevlar

In figures 55 to 58, the amount of nanoparticles bonded to the fibers becomes

clearly more important when using the gluta system. Both Silica and Calcium carbonate

particules bond uniformly well on the fibers. It shows how strong the bonding ability of

the gluta system is. This bonding ability is believed to come from the specific chemical

links between atoms.

Further studies on links between atoms are being investigated in a chemical

analysis.

80

4.4 Chemical Analysis

500 1000 1500 2000 2500 30000

20

40

60

80

100

120

140

160

% T

rans

mis

ion

Wave number (cm-1)

Gluta 2 Gluta 1 Gluta 3 Silane Silica H2O

Figure 60: FTIR of the different glutaraldehyde ratios added to the silated silica

mixture. Gluta1=0.225g, Gluta2=0.113g, Gluta3=0.45g

FTIR is very useful in identifying chemical bonding, either organic or inorganic,

by the energy transmission at various wave numbers. FTIR was performed on different

sets of glutaraldehyde mixtures which ratios were different: Gluta1=0.225g,

Gluta2=0.113g, Gluta3=0.45g. It is observed that at a specific glutaraldehyde ratio the

nanoparticles transmit less energy. This is shown at the lowest peak wave number near

one thousand, representing the secondary amine C-N stretch at 1140cm-1 which occurs

with the addition of glutaraldehyde.

81

Figure 61: FTIR of different glutaraldehyde combinations present in the final mixture


The relative absorbance of the different glutaraldehyde combinations shows the

various bonds present in the armor composite. As mentioned previously, the secondary

amine stretch C-N is playing the most important part in the bonding ability of the silica

to the Kevlar fabric which is certainly giving the resistance to the armor composite. The

absorption energy at 944cm-1 corresponds to the C-H bend vibrations. The one at

1408cm-1 corresponds to the C-C stretch (in-ring). Finally another bond present in the

armor composite is the methyl C-H stretch at 2975cm-1.

An energy-absorption spectrum has been observed at a particular wavenumber

1710cm-1 on the Kevlar with glutaraldehyde sample only. The peak observed comes

from the glutaraldehyde in presence of the Kevlar and represents the weak bands

overtones reflecting the substitution pattern on the ring. The small energy absorption of

82

this bond noticed in the entire solution suggests that it has been masked by the addition

of the other chemicals.

4.5 Mechanical Testing

Testing the different Kevlar based composites according to the procedure

explained earlier brings further evidence of the advantage of the addition of

glutaraldehyde into the mixture.

Figure 62: Warp direction tension test

In figure 61 the results of the warp direction test are shown. It is first observed

that there differences in stresses between the various fabrics are very less. This is due to

the way the fabric is manufactured. The fibers in the warp direction are pre-tensioned

whereas the fibers in the fill direction are sewed in between the warp ones and do not

get pre-tensioned.

0

100

200

300

400

500

600

0 2 4 6 8 10 12Strain (%)

Stre

ss (M

Pa)

Silane-silica-gluta/kevlar

Neat/kevlar

PEG-Silane-silica-

Silane-silica/Kevl

83

During the tests a pre-tension load of 10N is applied before starting in order to

align the fibers. The neat Kevlar and PEG-silane-silica Kevlar composite are going up

to similar stresses of about 485 MPa. The neat Kevlar reaches this maximum stress at

8.4% strain before failure occurs. The PEG-silane-silica Kevlar composite reaches a

strain of 9% before failing. The silane-silica Kevlar composite reaches 9.7% of strain

before failing at a stress of 508MPa. The examination of the silane-silica-glutaraldehyde

Kevlar indicates that the addition of glutaraldehyde into the mixture enhance the

mechanical properties of the composite as it fails 533MPa at 8.1% of strain. This

indicates that the reinforced fibers hold higher stresses when impregnated with a

mixture of glutaraldehyde.

Figure 63: Fill direction tension test

0

100

200

300

400

500

600

700

0 2 4 6 8 10 12Strain (%)

Stre

ss (M

Pa)


Neat/kevlar PEG-Silane-

silica-

Silane-silica/Kevl

84

The results in Figure 62 show the fill direction tension test. Those results show a

clear comparison of the various fabrics as this is the fiber direction which has not been

pre-tensioned. Therefore neat Kevlar show the lowest stress as it goes up to 413MPa at

strain of 5.1%. Then we can notice that the PEG-silane-silica Kevlar composite does not

do better than the neat Kevlar as it only reaches 384MPa at a strain of 5.4%. This

clearly indicates that STF based composites do not enhance the mechanical properties

of the composite. The silane-silica Kevlar composite shows better resistance to tension

as it 575MPa at a strain of 6.7% indication that removal of PEG helps increasing the

bonding of the particles and therefore increases the resistance of the fibers. The silane-

silica-glutaraldehyde Kevlar composite confirms the warp direction tension test as it

reaches a stress of 670MPa before failure occurs at a strain of 6.2%. The addition of

glutaraldehyde is again confirmed to be responsible in the composite resistance.

Figure 64: 45° direction tension tests

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 10 20 30 40 50 60 70Displacement [mm]

Load

[N]


Neat/kevlar

PEG-Silane-silica-

Silane-silica/Kevla

Silane-CaCO3-

85

The [45degre] shear test (ASTM D3518-76) involves uniaxial tension testing of

±45degree composites; usually laminates. The method is used for measuring in-plane

shear properties, such as the shear modulusG12 and the ultimate shear strength τ12 of

unidirectional fiber-reinforced composites. The method used is described earlier in

section 3.4, where samples of 1inch by 6 inches (25.4mmx152.4mm) are tested in

tension with a Zwick/Roell universal testing machine. The gage length for each sample

is about 4inches (101.6mm). The equation that relates the shear strength to the

longitudinal stress is related by the equation: xxστ21

12 = . The shear strain equation is

given by yyxx εεγ −=12 where σxx, εxx, and εyy represent tensile stress, longitudinal

strain and transverse strain respectively. By knowing the samples parameters such as

the thickness, the length and the width, the data obtained during the tension test can be

related to the shear strength.

The shear strength will be crucial in our test as it gives important information on

the bonding between the fibers yarns.

In figure 63, the neat Kevlar shows very little performance as it has no

reinforcement what so ever, meaning no bonding between fibers. The load to fail the

neat Kevlar goes up to 233.6N to a displacement of 12.4mm before failure. The test of

neat nylon also shows pour performance as it reaches a load of 214.2N for a

displacement of 22.3mm. The slow failure of nylon is mainly due to the reorganization

mechanism of the molecular chains within the fibers. The PEG-silane-silica-Kevlar

composite shows higher performance, indicating that reinforced fabric does help

86

increasing the resistance of the fibers. It goes up to 1131.3N and to a displacement of

28.2mm prior failure.

Following a similar trend as the previous mechanical tests, the silane-silica-

kevlar composite is performing better than the PEG-based composite as it reaches a

load of 1282.4N at a displacement of 27mm. The bonding between the fibers is

confirmed to be increased by the removal of the PEG out of the composite manufacture.

The observation of the silane-silica-glutaraldehyde-kevlar composite shows the

highest resistance to the tensile load as it reaches a load of 1844.5N at a displacement of

37.8mm which confirms the high bonding between the fiber yarns. This high bonding

ability is obtained by the addition of glutaraldehyde.

The silane-silica-glutaraldehyde-spectra composite is slowly increasing to

855.6N where it extends its length to about 35.5mm where a rupture partially occurs

within the sample. The resistance continues as the loading goes up and finally reaches

its failure point at 958.8N after an elongation of 44.4mm over its original length.

To further understand the difference occurring during the 45° degree test

between the Kevlar and Spectra composite, some microscopy has been performed.

Figure 64 shows the difference between Kevlar composite before and after the tension

test in a 45 degree direction.

87

Figure 65: a)45degree Kevlar composite before test. b) after test.

A)

B)

88

The difference before and after the test is performed resides mostly in the

orientation of the fibers between each other. The angle starts with a 90degree angle and

slowly decreases as the tension goes on. At the failure point the alignment between the

fibers reaches an angle of approximately 50degree. It is also observed that the

reinforcement has gone away during the test, especially at the intersection between

yarns.

Figure 65 shows the difference between before and after the tension test in a

45degree direction of Spectra composites. The deformation shows a similar yarn

configuration between post and prior tension test. The yarn rotation is around 70degree

which is less than for Kevlar. It also shows that a lot less reinforcement is present at the

surface of the composite. This highlight the fact impregnation of Gluta system on

Spectra fibers do not work as efficiently as it does with Kevlar. This can then be

explained by the fact that Spectra do not exhibit an Amide group in its molecular

formula, which prevents the glutaraldehyde to bond to the surface.

89

Figure 66: a)45degree Spectra composite before test. b) after test.

A)

B)

90

4.6 Flexibility test

According to the flexibility test procedure described in 3.5, several composites

have been tested. The results have been recorded in the following table:

Figure 67: Flexibility Set-up

Although the surface of the armor composites appears to be cracked at some

locations, the flexibility is not compromised. In order to compare flexibility, procedures

outlined in section 2.6 are followed. Twelve 15.2x15.2 cm2 layers of various categories

of fabrics were stacked together and clamped at one end. Clamping width was

maintained at 1.3 cm. The rest of the length, i.e., about 13.9 cm was protruding off from

the clamped end. A total weight of 19.44 g was applied at the free end using an

aluminium strip. Each category of fabric was clamped and loaded in identical manner.

Flex angle for each category are then measured.

TYPE NEAT\Kevlar

NEAT/Correctional Kevlar

Spectra PEG-Silane-Silica\Kevlar

PEG-Silica\Kevl

arANGLE

(deg) 42 30 31 50 31.5

Table 3: Results table for flexibility test on the different armor composites

Silane-Silica\Kevlar

Silane-Silica-gluta\Kevlar

Silane-Silica-gluta\Correctional Kevlar

Silane-CaCo3-

gluta\Kevlar

Silane-Silica-gluta\Spectra

45 40 32.5 33 28

91

Results of flexibility tests show that any Gluta systems stay very flexible. It

slightly varies when compared to the neat fabric. Neat Kevlar has a flex angle of about

42° and the reinforced Kevlar with glutaraldehyde has a 40° angle. Similarly neat

correctional Kevlar has a 30° angle whereas the reinforced one as a 32.5° angle. The

neat Spectra fabric has a 31° angle and the reinforced Spectra fabric has a 28° angle.

We can easily conclude that the reinforcement applied to the fabric has no consequence

on the flexibility of the various composites.

The introduction of new particles such as Calcium carbonate do not have any

influence on flexibility as it is seen that reinforced Kevlar with a mixture of silane,

calcium carbonate and glutaraldehyde shows a flex angle of about 33°.

The PEG-silane-silica/Kevlar composite is the only one showing less flexibility as it has

a 50° angle. This can be explained by the addition of silane which is stiffening the

composite.

The PEG-silica/Kevlar composite has a low angle of 31.5° which can be due to

the high density of PEG into the composite.

The Silane-silica/Kevlar composite stays relatively close to neat Kevlar with a 45°

angle.

A remark can be made regarding the correlation with the density of the

composite. The higher is the density of the composite, the smaller is the flex angle

which is then translated into higher flexibility according to the present standard. A more

comprehensive flexibility test should be found in order to measure more accurately the

flexibility of any composite manufactured.

92

4.7 Discussion

Overall, it is observed that a sonicated mixture of Glutaraldehyde and silated

silica nanoparticles impregnated into Kevlar or Spectra fabric offers significant

resistance to spike penetration as well as to knife penetration. This conclusion has been

drawn from the investigation on STF composite in which PEG has been removed from

the equation. The outcome composite, silane-silica-kevlar showed significant

improvements and a better coating has been observed on SEM pictures indicating

higher number of bonds between particles. This observation led us to the introduction of

glutaraldehyde. This type of formaldehyde was believed to increase the amount of

bonds between silica particles as well as with the Kevlar fibers. By creating strong

covalent bonds with the amino-groups contained in the amino-propyl-trimethoxy-silane

and the amide groups present in the Kevlar fabric, the glutaraldehyde enhances

considerably the adherence of the particles on to the Kevlar. Extensive microscopic

studies revealed this uniform coating on the fibers and chemical studies gives more

details on which atomic linkage is responsible for this improved bonding ability. FTIR

has been performed and strongly reveals the secondary amine stretch C-N at the

wavelength of 1140cm-1 to be the cause of the increased amount of bonds within the

composite. The increased number of bonds brought by the addition of glutaraldehyde

into the mixture is believed to be translated into higher mechanical properties of the

impregnated composite. Several tensile tests have been conducted in warp, fill and 45°

degree direction. The outcome of those tests show that glutaraldehyde based composites

have their mechanical performances greatly increased from neat Kevlar or previous

93

silane-silica-kevlar composites. The gluta-based composite reaches the stresses up to

533MPa in warp direction compared to 508MPa. The difference is more evident in fill

direction tension test where the gluta-based composite reaches a maximum stress of

670MPa compare to the silane-silica-kevlar composite which only goes up to 575MPa.

The study of the 45° degree test gives important strength information on how strong the

bonds created by the glutaraldehyde are. The analysis of the different impact patterns

adds up more details on the failure mechanism of such reinforced fabric: reinforcement

of various fabrics bonds the fibers together and less yarns pullout are observed during

spike impact. On the other hand knife impact patterns show a clean cut through the

fibers as the reinforcement restricts the movement of the fibers when the blade is

penetrating. It indicates that fiber properties play an important role in stab resistance.

This is confirmed when NIJ spike and knife tests are performed on the various selection

of fabrics used. The investigation has concluded that Kevlar based composites

performed very well under spike impacts and Spectra based composites under knife

impacts. For instance the silane-silica-glutaraldehyde-Correctional Kevlar composite

achieves the “zero-layer” penetration up to 232 J-cm2/g for the spike test whereas the

silane-silica-gluta-spectra composite only goes up to 30.5 J-cm2/g. This is the opposite

when NIJ knife tests are performed: spectra-based composite goes up to 6.4 J-cm2/g

whereas the Kevlar-based composites do not even achieve a “zero-layer” penetration.

The concept of hybridization is then introduced to address this problem and offer

protection from both puncture and stab. A Kevlar-Spectra hybrid has been developed

showing relatively good performances to both threats in comparison with the full

composites tested. In other word, the hybrid composed of 7 layers of reinforced Kevlar

94

and 8 layers of reinforced Spectra, is the best compromise to provide resistance from

both the spike and the knife. The hybrid composite achieves the “zero-layer”

penetration up to 4 J-cm2/g for the knife and up to 26.7 J-cm2/g for the spike.

In the attempt to increase the resistance performances the Kevlar fabric, calcium

carbonates nano particles have been introduced in place of the silica particles. Identical

results to that of silica based composites have been observed. This discards further more

the STF theory and opens the researches to new type of particles that could enhance the

fabric.

By reinforcing the fabrics, one could easily believe that flexibility is lost along

the process. However, flexibility is measured by recording the angle with the vertical

for each composite. Most of the composites show no variations in comparison to their

neat fabric.

95

CHAPTER 5. FINITE ELEMENT ANALYSYS OF SPIKE PENETRATION

5.1 The Finite Element Method

The finite element method is a numerical technique for finding approximate

solutions of partial differential equations as well as of integral equations. The solution

approach is based either on eliminating the differential equation completely (steady

state problems), or rendering the PDE into an approximating system of ordinary

differential equations, which are then numerically integrated using standard techniques

such as Euler's method, Runge-Kutta, etc.

In solving partial differential equations, the primary challenge is to create an

equation that approximates the equation to be studied, but is numerically stable,

meaning that errors in the input data and intermediate calculations do not accumulate.

There are many ways of doing this, all with advantages and disadvantages. The Finite

Element Method is a good choice for solving partial differential equations over complex

domains (like cars and oil pipelines), when the domain changes (as during a solid state

reaction with a moving boundary), when the desired precision varies over the entire

domain, or when the solution lacks smoothness. For instance, in the spike penetration

simulation, increased precision of the mesh can be operated over the spike and

decreased over the backing material.

In order to recreate the actual problem using the finite element approach, a

commercial modeler, solver and post-processor called Ansys® is utilized.

96

5.2 Modeling of the spike penetration problem

The NIJ spike test has been modeled using Ansys®: a 2000g spike and a nylon

mass assembly has been modeled with the target and backing materials as they are used

in actual NIJ test set-up. In Figures 67 to 72, various components of the setup are

shown.

Figure 68: Geometry models of the spike and nylon mass

The geometry is modeled by first defining the keypoints of the structure. As it

presents an axis of rotation, we will only position the keypoints of a cross-section of the

structure, and then rotate the cross section to create the 3D geometries. Dimensions

used to model both geometries are detailed in figure 68.

1

ANSYS 11.0SP1

Nylon mass

Spike

97

Figure 69: Cross-sections of spike and nylon mass

Figure 70: Geometry model of backing materials

180mm

100mm

0.2mm

50mm

510mm

Axis of rotation

2mm

1

X

Y

Z

ANSYS 11.0SP1

2 Rubber layers 1 Polyethylene foam layer

4 layers of neoprene

Target (Kevlar Composite)

98

The backing materials are modeled similarly using keypoints positioned at the

dimensions measured on the actual backing material. Dimensions are referenced in

figure 70. Once the section is drawn using the keypoints, the volumes are generated

according to the respective height of each layer.

Figure 71: Backing materials geometry details

Finally the spike and backing materials are modeled together in order to obtain

the geometry which we will use to simulate the problem. Figure 71 shows the actual

placement of the spike to the backing material. The axis of rotation of the spike and

mass system is aligned with the center of the backing material plan.

304.8mm

304.8mm

Stacking sequence

Target

Neoprene

PE foam

Rubber 12.6mm

33mm

20mm

5mm

Top view

99

1

X

Y

Z

ANSYS 11.0SP1

Figure 72: The spike-mass system is placed on the top of the backing material on

aligned on the center

1

X

Y

Z

ANSYS 11.0SP1

Figure 73: Global view of the geometry

100

The aim of the simulated problem is to study the penetration of a spike into the

composite.

5.3 Elements and mesh generation

After modeling the geometry, the mesh was created using the following

elements. Various elements utilized to mesh the structure are as follows:

SOLID92 is the element type chosen to mesh the plate which is

constituted of 8 layers. 1 layer with the Kevlar properties, 4 layers with

neoprene material, 1 layer with polyethylene foam and 2 layers with

rubber.

SOLID186 is the element type chosen to mesh the spike. It has been

chosen for its characteristics to handle complex sharp shapes as it

presents a high number of nodes.

CONTA178 is an element type called contact element. It allows the

displacement of the spike toward the target.

5.3.1 SOLID92

SOLID92 has a quadratic displacement behavior and is well suited to model

irregular meshes. The element shown in figure 73, is defined by ten nodes having three

degrees of freedom at each node: translations in the nodal x, y, and z directions.

101

Figure 74: SOLID92 element geometry. 3-D 10-Node Tetrahedral Structural Solid.

The element also has plasticity, creep, swelling, stress stiffening, large

deflection, and large strain capabilities

1

X

Y

Z

ANSYS 11.0SP1

Figure 75: Meshing of the target and backing material with SOLID92

102

To mesh the volumes constituting the backing materials, an element size was

chosen. The size had to be not too small in order to avoid too long calculations for the

simulation and not too big in order to provide accurate results. A size of 20mm has been

chosen for the edges of the element. Ansys meshed all volumes based on this element

size.

The solution output associated with element includes as part of other things the

nodal displacement and the stresses in every direction. The element stress directions are

parallel to the element coordinate system. The surface stress outputs are in the surface

coordinate system and are available for any face. The coordinate system for face J-I-K

is shown in figure 75. The other surface coordinate systems follow similar orientations.

Figure 76: SOLID92 stress output

Some restrictions concern the use of this element. The element must not have a

zero volume. Elements may be numbered either as shown in figure 61or may have node

L below the I-J-K plane.

An edge with a removed midside node implies that the displacement varies

linearly, rather than parabolically, along that edge.

103

5.3.2 SOLID186

SOLID186 is a higher order 3-D 20-node solid element that exhibits quadratic

displacement behavior. The element is defined by 20 nodes having three degrees of

freedom per node: translations in the nodal x, y, and z directions. The geometry, node

locations, and the element coordinate system for this element are shown in figure 76. A

prism-shaped element may be formed by defining the same node numbers for nodes K,

L, and S; nodes A and B; and nodes O, P, and W. A tetrahedral-shaped element and a

pyramid-shaped element may also be formed as shown in figure 76.

In addition to the nodes, the element input data includes the anisotropic material

properties. Anisotropic material directions correspond to the element coordinate

directions. The element coordinate system orientation is as described in figure 63.

Figure 77: SOLID186 geometry

104

The element supports plasticity, hyperelasticity, creep, stress stiffening, large

deflection, and large strain capabilities. It also has mixed formulation capability for

simulating deformations of nearly incompressible elastoplastic materials, and fully

incompressible hyperelastic materials.

SOLID186 Structural Solid is well suited to modeling irregular meshes. The

element can also have any spatial orientation.

Figure 78: Meshing of the Spike and its mass with elements SOLID186

Some assumptions and restrictions are applied to this element as follows:

The element must not have a zero volume. Also, the element may not be twisted

such that the element has two separate volumes (which occurs most frequently

when the element is not numbered properly). Elements may be numbered either

as shown in figure 76 or may have the planes IJKL and MNOP interchanged.

An edge with a removed midside node implies that the displacement varies

linearly, rather than parabolically, along that edge.

When degenerated into a tetrahedron, wedge, or pyramid element shape, the

corresponding degenerated shape functions are used. Degeneration to a

pyramidal form should be used with caution. The element sizes, when

1

ANSYS 11.0SP11

ANSYS 11.0SP1

105

degenerated, should be small to minimize the stress gradients. Pyramid elements

are best used as filler elements or in meshing transition zones.

Also stress stiffening is always included in geometrically nonlinear analyses.

5.3.3 CONTA178

CONTA178 represents contact and sliding between any two nodes of any types of

elements. The element has two nodes with three degrees of freedom at each node with

translations in the X, Y, and Z directions. It can also be used in 2-D and axisymmetric

models by constraining the UZ degree of freedom. The element is capable of supporting

compression in the contact normal direction and Coulomb friction in the tangential

direction. The element may be initially preloaded in the normal direction or it may be

given a gap specification. A longitudinal damper option can also be included.

Figure 79: Insertion of contact element CONTA178 in between the spike and the target

1

MX

X

Y

Z

0

.119E+09.237E+09

.356E+09.474E+09

.593E+09.711E+09

.830E+09.949E+09

.107E+10

ANSYS 11.0SP1

CONTA 178

106

Figure 80: CONTA178 element geometry.

As shown in figure 78, the element CONTA178 is used to connect the spike tip

to the target. The element is defined by two nodes, each node belonging to either the

spike or the target. The element is used so that the displacement of the spike towards the

target is computed to close the gap. The element also transfers the applied load coming

from the spike and deforms then the target.

The restrictions applied to this element are as follows:

The element operates bilinearly only in the static and the nonlinear transient

analyses. If used in other analysis types, the element maintains its initial status

throughout the analysis.

The element is nonlinear and requires an iterative solution.

Nonconverged substeps are not in equilibrium.

Unless the contact normal direction is specified by (NX, NY, NZ), nodes I and J

must not be coincident or overlapped since the nodal locations define the

107

interface orientation. In this case the node ordering is not an issue. On the other

hand, if the contact normal is not defined by nodal locations, the node ordering

is critical.

The element maintains its original orientation in either a small or a large

deflection analysis unless the cylindrical gap option is used.

5.4 Determination of the materials properties

Material properties have been implemented in ANSYS® using the models of

equation developed at Purdue University [67].

The strain in the yarn direction is generally small. The relative rotation between yarns

could be large as shown in figure 80. Yarn rotation angles are recorded at each

incremental deformation step. The warp and weft layers are modeled separately and the

interaction between them is considered.

Figure 81: Warp and weft yarn orientations in global coordinate system.

Figure 82: Local orthogonal coordinate system for warp layer.

α β

x

y

Weft direction

Warp direction

αx

y Warp direction x2 x1

108

Considering the warp layer, a local orthogonal coordinate system x1-x2 is set up

with x1 parallel to the yarn direction as shown in Figure 81. The stress-strain relations

for the warp layer are formulated as

11 1 11

22 2 22

12 12 12

0 00 00 0

EE

G

α

α

α

σ εσ εσ γ

⎡ ⎤Δ Δ⎧ ⎫ ⎧ ⎫⎢ ⎥⎪ ⎪ ⎪ ⎪Δ = Δ⎨ ⎬ ⎨ ⎬⎢ ⎥

⎪ ⎪ ⎪ ⎪⎢ ⎥Δ Δ⎩ ⎭ ⎩ ⎭⎣ ⎦

(1)

where 1Eα is curve-fitted using the data from the tension test of the fabric in the warp

direction. It is a function of yarn strain along the warp direction. 2Eα and 12Gα are several

orders smaller than 1Eα . In this study, 12Gα is set to be a constant, and 2Eα is a highly

nonlinear function of nondimensional yarn spacing between warp yarns. The stress-

strain relations in global coordinate system x-y are obtained by coordinate

transformation:

11 12 16

12 22 26

16 26 66

xx xx xx

yy yy yy

xy xy xy

Q Q QQ Q Q QQ Q Q

α α α

α α α α

α α α

σ ε εσ ε εσ γ γ

⎡ ⎤⎧ ⎫ ⎧ ⎫ ⎧ ⎫Δ Δ Δ⎢ ⎥⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎡ ⎤Δ = Δ = Δ⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎢ ⎥ ⎣ ⎦

⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎢ ⎥Δ Δ Δ⎩ ⎭ ⎩ ⎭ ⎩ ⎭⎣ ⎦

(2)

( )

( ) ( )( ) ( )

4 2 2 411 1 12 2

2 212 1 2 12

4 2 2 422 1 12 2

3 316 1 12 2 12

3 326 1 12 2 12

66 1 2

cos 4 sin cos sin

4 sin cos

sin 4 sin cos cos

2 sin cos 2 sin cos

2 sin cos 2 sin cos

Q E G E

Q E E G

Q E G E

Q E G E G

Q E G E G

Q E E

α α α α

α α α α

α α α α

α α α α α

α α α α α

α α α

α α α α

α α

α α α α

α α α α

α α α α

= + +

= + −

= + +

= − − −

= − − −

= +( ) ( )2 2 4 412 122 sin cos sin cosG Gα αα α α α− + +

(3)

109

Following the same procedures, we obtain the stress-strain relations for the weft

layer in the global coordinate system as

11 12 16

12 22 26

16 26 66

xx xx xx

yy yy yy

xy xy xy

Q Q QQ Q Q QQ Q Q

β β β

β β β β

β β β

σ ε εσ ε εσ γ γ

⎡ ⎤⎧ ⎫ ⎧ ⎫ ⎧ ⎫Δ Δ Δ⎢ ⎥⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎡ ⎤Δ = Δ = Δ⎨ ⎬ ⎨ ⎬ ⎨ ⎬⎢ ⎥ ⎣ ⎦

⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎢ ⎥Δ Δ Δ⎩ ⎭ ⎩ ⎭ ⎩ ⎭⎣ ⎦

(4)

( )

( ) ( )( ) ( )

4 2 2 411 1 12 2

2 212 1 2 12

4 2 2 422 1 12 2

3 316 1 12 2 12

3 326 1 12 2 12

66 1 2

cos 4 sin cos sin

4 sin cos

sin 4 sin cos cos

2 sin cos 2 sin cos

2 sin cos 2 sin cos

Q E G E

Q E E G

Q E G E

Q E G E G

Q E G E G

Q E E

β β β α

β β β β

β β β β

β β β β β

β β β β β

β β β

β β β β

β β

β β β β

β β β β

β β β β

= + +

= + −

= + +

= − − −

= − − −

= +( ) ( )2 2 4 412 122 sin cos sin cosG Gβ ββ β β β− + +

(5)

To simplify the formulation, we assume that both warp and weft layers occupy

the same space and have the same thickness 0.23 mmT = as one fabric layer. This

thickness is set to be constant over the fabric undergoing a large deformation. To be

consistent, material properties for the warp and weft layers are defined using thickness

T. Thus the stiffness matrix for the fabric is obtained as

[ ]Q Q Qα β⎡ ⎤ ⎡ ⎤= +⎣ ⎦ ⎣ ⎦ (6)

110

The material properties 1Eα and 1Eβ are obtained from the respective uniaxial

tension tests of the fabric. Bilinear functions of yarn strain are used to fit the stress-

strain curves with the result for kevlar composite:

{ }{ }

( )( ) ϑε

ϑε

ε

ε

αβ

βα

β

β

β

α

α

α

sin1

sin1

067.0,9.101

081.0,1.111

11

11

111

111

+=

+=

≤=

≤=

L

L

GPaL

E

GPaL

E

(7)

where θ α β= − is the angle between warp and weft yarns; 11αε and 11

βε are yarn strains

in warp and weft directions, respectively; Lα and Lβ are nondimensional yarn spacings

for the warp yarns and weft yarns, respectively.

The shear moduli 12Gα and 12Gβ control the initial shear rigidity when the warp and

weft yarns are orthogonal to each other. We have

MPaGGG 4.241212 === βα (8)

The transverse moduli 2Eα and 2Eβ are related to the highly nonlinear shear rigidity

of the fabric under large shear deformation. Physically 2Eα and 2Eβ prevent the parallel

yarns from being too close to each other. We have

111

2 1

2 1

7.80min ,

: Kevlar with par7.80min ,

SilicaLock

SilicaLock

E E MPaL L

E E MPaL L

α α

α

β β

β

⎫⎛ ⎞= ⎪⎜ ⎟−⎝ ⎠ ⎪

⎬⎛ ⎞ ⎪= ⎜ ⎟ ⎪⎜ ⎟−⎝ ⎠ ⎭

ticles

(9)

where 0.70SilicaLockL = indicate the angles of the rotation locking for the fabric with

reinforcement. When Lα and Lβ approach these values, the transverse module increases

rapidly to prevent further relative yarn rotations.

Results obtained during the mechanical testing experiments were implemented

into Ansys following this procedure. Table 4 shows the material properties used.

Properties Nylon mass Spike Kevlar

composite Neoprene Polyethylene Rubber

EX (Pa) 3.2E9 207E9 124E9 105E3 565E3 2.24E9 Ey (Pa) 112.4E9 147E3 83E3 1.34E9 EZ (Pa) 124E9 105E3 565E3 2.24E9 υxy 0.33 0.3 0.36 0.3 0.35 0.35 υyz 0.36 0.3 0.35 0.35 υxz 0.36 0.3 0.35 0.35

Gxy (Pa) 28.4E9 27.2E3 0.117E9 24.4E9 Gyz (Pa) 28.4E9 27.2E3 0.117E9 24.4E9 Gxz (Pa) 28.4E9 27.2E3 0.117E9 24.4E9 Density 1.15 7850 1.44 1.5 3.45 1.52

Table 4: Material properties table for the different constituents of the model

5.5 Boundary Conditions

The Finite Element problem is designed to simulate the fall of a spike and its

impact onto the target. Therefore it is shown in figure 82, that nodes from the elements

associated with the spike will be assigned a gravitational acceleration of 9.81m.s-2. The

backing material has to stop the spike from falling. All the nodes from the last layer of

112

rubber will be constrained in all degrees of freedom. This will allow the rest of the

backing material to deform when the spike will impact.

Figure 83: Boundary conditions applied on the system

5.6 Results

Once the geometry and mesh were created and boundary conditions applied. The

finite element problem was solved by involving Ansys solver. The solution of the Finite

Element problem yielded displacement and stresses at nodes of elements both at the

spike and at the target. Penetration of the spike was then simulated using a time

X

Y

Z

113

resolved transient analysis. The goal of the transient analysis was to simulate the actual

penetration as well as to predict the depth of penetration.

1

MN

MX

X

Y

Z

0

.119E+09.237E+09

.356E+09.474E+09

.593E+09.711E+09

.830E+09.949E+09

.107E+10

ANSYS 11.0SP1

Figure 84: Simulated penetration of the spike. VonMises stress result.

The maximum stress is recorded at the center of the target and is about 1.07GPa.

The maximum displacement is also observed at the center of the target and backing

material setup, right under the tip of the spike. It has been recorded to be about 48mm.

Comparisons between the simulation and experiment has been done and shown

in figure 84. The validity of the model is confirmed since the simulated curve follows

relatively the experimental curve.

114

Figure 85: Comparison graph of Simulated Quasi-static impact penetration to

Experimental

The experimental curve goes up to 1.10e3 N and depth of 50mm whereas the

finite element problem shows a force going up to about 1.20e3 N and a depth of 48mm.

Despite the differences observed, the model shows good prediction of the depth of

penetration and forces involved.

5.7 Discussion

The finite element analysis performed has shown very similar results to that of

the experimental work. The choice of proper elements has helped obtaining a stable

mesh. The choice of SOLID92 for the backing material and the target has shown good

results due to its quadratic displacement behaviour. It is the best suited to model meshes

that will undergo large deflections. Utilizing SOLID186 to mesh the spike was also the

more appropriate choice as it is an element that is perfectly suited for irregular shapes. It

also has mixed formulation capability for simulating deformations of nearly

0.00E+00

2.00E+02

4.00E+02

6.00E+02

8.00E+02

1.00E+03

1.20E+03

1.40E+03

0 10 20 30 40 50 60Depth of Penetration [mm]

Forc

e [N

]Ansys

Experimental

115

incompressible elastoplastic materials, and fully incompressible hyperelastic materials.

Finally CONTA178 was chosen mostly because it is a 3D node-to-node contact, which

means that it only requires 2 nodes to be defined. The geometries being centered on the

same axis it become more practical to define this type of contact element. Once, the

geometries are meshed, the material properties defined and the boundary conditions set,

the solution is computed through the Ansys solver. In order to compare the simulation

output to the experimental work, the force and the depth of penetration are plotted.

Comparison of the curves shows a good correlation between experimental and finite

element model. Therefore it is possible to utilize the finite element model in order to

predict the force and depth of penetration of a spike test.

116

CHAPTER 6.CONCLUSION

6.1 Summary

The following is the summary of the present investigation:

1. A flexible armor system using a cross-linking fixative (Glutaraldehyde) with

silane, silica and Kevlar has been developed. The new system demonstrates

spike resistance almost 10 times more than that of the traditional PEG-silica-

Kevlar system without any loss of flexibility.

2. An unprecedented material resistance to spike penetration has been observed as

Gluta creates strong chemical bonds between distant pairs of amine groups

present in functionalized silica particles, and Kevlar.

3. Formation of C-N stretches as a result of interaction between Gluta and amide

functional group is responsible for such high resistance to spike. This seems to

happen in presence of Kevlar, but not Spectra which does not have amide

functional groups.

4. When silica particles are replaced in the new system with biodegradable CaCO3

particles, spike performance remains unchanged.

5. Knife performances have been improved significantly by hybridizing Kevlar and

Spectra fabrics – by doing so one has to sacrifice some of the spike resistances,

but overall the new system is superior to previous PEG based systems

117

6. Preliminary work using Finite Element Analysis has been completed to model

the penetration of the spike into the fabric. Geometries, material properties and

boundary conditions have been defined to be as close as possible to the actual

NIJ spike test. The results obtained show a good correlation between the

experimental and the simulation.

6.2 Future Work

1. In our recent studies, we have noticed that Spectra perform well in knife

experiments while Kevlar does that with spike. This prompted us to hybridize

the armor construction with Kevlar and Spectra. However, the exact differences

in the failure mechanisms with knife and spike are not yet fully known. We

intend to perform failure studies with SEM studies and a follow-up numerical

modeling

2. We will also investigate the role of fiber during penetration - as to why Spectra

is good for knife while Kevlar is for spike. This will require design of new

experiments to monitor spike or knife penetration in-situ using a SEM or a high

speed camera

3. Role of fiber in absorbing energy is well known. Although used for ballistic and

puncture applications, the elastic storage energy capacity of commercially

available Spectra and Kevlar fibers are still very low. Nanoscale reinforcement

of these fibers will certainly enhance such capacity. We will reinforce ultra high

molecular polyethylene (UHMWPE, polymer precursor for Spectra fiber) with

118

carbon nanotubes (CNT) through a solution spinning process to increase yield

strength, modulus and fracture strain of the fiber.

.

119

REFERENCES

1 C.T. Nguyen, T. Vu-Khanh, and J. Lara. 2004. "Puncture characterization of rubber membranes." Theoretical and Applied Fracture Mechanics 42: 25-33.

2 "Standard Test Method for Protective Clothing Material Resistance to

Puncture." ASTM Standard F1342-05. 3 L.F. Leslie, J. A. Woods, J.G. Thacker, R.F. Morgan, W. McGregor, and R.F.

Edlich. 1996. "Needle puncture resistance of surgical gloves, finger guards, and glove liners." Journal of Biomedical Materials Research - Applied Biomaterials 33: 41-46.

4 D.C. Erlich, D.A. Shockey, and J.W. Simons. 2003. "Slow penetration of

ballistic fabrics." Textile Research Journal 73, no. 2: 179-184. 5 R.M. Koerner, R.F. Wilson-Fahmy, and D. Narejo. 1996. "Puncture protection

of geomembranes, Part III: Examples." Geosynthetics International 3, no. 5: 655-670.

6 D. Narejo, R.M. Koerner, and R.F. Wilson-Fahmy. 1996. "Puncture protection

of geomembranes, Part II: Experimental." Geosynthetics International 3, no. 5: 629-653.

7 R.F. Wilson-Fahmy, D. Narejo, and R.M. Koerner. 1996. "Puncture protection

of geomembranes, Part I: Theory." Geosynthetics International 3, no. 5: 605-628.

8 T.K. Ghosh. 1998. "Puncture resistance of pre-strained geotextiles and its

relation to uniaxial tensile strain at failure." Geotextiles and Geomembranes 16, no. 5: 293-302.

9 "Standard Test Method for Index Puncture Resistance of Geotextiles,

Geomembranes, and Related Products." ASTM Standard D4833-00e1. 10 J. Lara, D. Turcot, R. Daigle, and J. Boutin. 1996. "New test method to

evaluate the cut resistance of glove materials." ASTM Special Technical Publication 1237: 23-31.

120

11 J. Lara, D. Turcot, R. Daigle, and F. Payot. 1996. "Comparison of two methods to evaluate the resistance of protective gloves to cutting by sharp blades." ASTM Special Technical Publication 1237: 32-42.

12 "Standard Test Method for Measuring Cut Resistance of Materials Used in

Protective Clothing." ASTM Standard F1790-05. 13 J. Lara and S. Masse. 2000. Evaluating the cutting resistance of protective

clothing materials. Proceedings of 1st European Conference on Protective Clothing, Stockholm, Sweden, May 7-10, 2000, edited by Kalev Kuklane and Ingvar Holmer,145-149.

14 H.S. Shin, D.C. Erlich, and D.A. Shockey. 2003. "Test for measuring cut

resistance of yarns." Journal of Materials Science 38: 3603-3610. 15 S.S. Deshmukh and G.H. McKinley. 2003. Magnetorheological suspensions:

Rheology and applications in controllable energy absorption. Proceedings of the Society of Rheology, October 2003.

16 R. Gadow and K. von Niessen. 2006. “Lightweight Ballistic with Additional

Stab Protection Made of Thermally Sprayed Ceramic and Cermet Coatings on Aramide Fabrics.” Journal of Applied Ceramic Technology 3, no. 4: 284-292.

17 S.R. Raghavan and S.A. Khan. 1995. “Shear-induced microstructural changes in

flocculated suspensions of fumed silica,” Journal of Rheology 39, no. 6: 1311-1325.

18 S.R. Raghavan and S.A. Khan. 1997. “Shear-Thickening Response of Fumed

Silica Suspensions under Steady and Oscillatory Shear,” Journal of Colloid and Interface Science 185: 57-67.

19 B.J. Maranzano and N.J. Wagner. 2001. “The effects of particle size on

reversible shear thickening of concentrated colloidal dispersions,” Journal of Chemical Physics 114, no. 23: 10514-10527.

20 Y.S. Lee and N.J. Wagner. 2003. “Dynamic properties of shear thickening

colloidal suspensions,” Rheology Acta 42, no. 3: 199-208. 21 Y.S. Lee, Wetzel, E.D. and N.J. Wagner. 2003. “The ballistic impact

characteristics of Kevlar® woven fabrics impregnated with a colloidal shear thickening fluid,” Journal of Materials Science 38, no. 13: 2825-2833.

121

22 Y.S. Lee, E.D. Wetzel, R.G. Egres Jr., and N.J. Wagner. 2002. Advanced Body Armor Utilizing Shear Thickening Fluids. Proceedings of the 23rd Army Science Conference, 2002.

23 E.D. Wetzel, Y.S. Lee, R.G. Egres, K.M. Kirkwood, J.E. Kirkwood, and N.J.

Wagner. 2004. Proceedings of NUMIFORM, 2004. 24 R.G. Egres, Y.S. Lee, J.E. Kirkwood, K.M. Kirkwood, E.D. Wetzel and N.J.

Wagner. 2003. Proceedings of the 14th International Conference on Composite Materials, July 2003.

25 R.G. Egres Jr., M.J. Decker, C.J. Halbach, Y.S. Lee, J.E. Kirkwood, K.M.

Kirkwood, E.D. Wetzel, N.J. Wagner. 2004. Stab Resistance of Shear Thickening Fluid (STF)-Kevlar Composites for Body Armor Applications. In Proceedings of the 24th Army Science Conference, Orlando, Florida, November 29-December 2, 2004.

26 V.B.C. Tan, T.E. Tay and W.K. Teo. 2005. “Strengthening fabric armour with

silica colloidal suspensions,” International Journal of Solids and Structures 42: 1561-1576.

27 S.L. Phoenix and P.K. Porwal. 2003. “A new membrane for the ballistic impact

response and V50 performance of multi-ply fibrous systems,” International Journal of Solids and Structures 40: 6723-6765.

28 P.K. Porwal and S.L. Phoenix. 2005. “Modeling system effects in ballistic

impact into multi-layered fibrous materials for soft body armor,” International Journal of Fracture 135: 217-249.

29 W.J. Taylor and J.R. Vinson. 1989. “Modeling Ballistic Impact into Flexible

Materials,” American Institute of Aeronautics and Astronautics 28, no. 12: 2098-2103.

30 B. Gu. 2003. “Analytical modeling for the ballistic perforation of planar plain-

woven fabric target by projectile,” Composites: Part B 34: 361-371. 31 S. Luo and T. Chou. 1990, “Finite Deformation of Flexible Composites,”

Proceedings of the Royal Society of London 429, no.1877: 569-586. 32 G. Stylios and Y.M. Xu. 1995. “An Investigation of the Penetration Force

Profile of the Sewing Machine Needle Point,” Journal of Textile Instruments 86, no. 1: 148-163.

122

33 S. Suresh, A.E. Giannakopoulos and J. Alcala. 1997. “Spherical Indentation of Compositionally Graded Materials: Theory and Experiments,” Acta Metallurgica 45, no.4: 1307-1321.

34 S. Suresh and A.E. Giannakopoulos. 1998. “A New Method for Estimating

Residual Stresses by Instrumented Sharp Indentation,” Acta Metallurgica 46, no. 16: 5755-5767.

35 A.E. Giannakopoulos and S. Suresh. 1999. “Determination of Elastoplastic

Properties by Instrumented Sharp Indentation,” Scripta Materialia 40, no. 10: 1191-1198.

36 E.W. Andrews; A.E. Giannakopoulos; E. Plisson, and S. Suresh: 2002.

“Analysis of the impact of a sharp indenter,” International Journal of Solids and Structures 39, no. 2: 281-295.

37 C. Nathaniel, H. Mahfuz, V. Rangari, A. Ashfaq and S. Jeelani. 2005.

“Fabrication and Mechanical Characterization of Carbon/Epoxy Nanocomposites,” Composite Structures 67: 115-124.

38 H. Mahfuz, A. Adnan, V.K. Rangari, M.M. Hasan, S. Jeelani, W.J. Wright and

S.J. DeTeresa. 2006. “Enhancement of Strength and Stiffness of Nylon 6 Filaments through Carbon Nanotubes Reinforcements,” Applied Physics Letters 88, no. 1.

39 R. Rodgers, H. Mahfuz, V. Rangari, N. Chisholm, and S. Jeelani. 2005.

“Infusion of Nanoparticle into SC-15 Epoxy; an Investigation of Thermal and Mechanical Response,” Macromolecular Materials & Engineering 290, no. 5: 423-429.

40 V.M. Harik and M. Salas, eds. 2003. Trends in Nanoscale Mechanics: Analysis

of Nanostructured Materials and Multi-Scale Modeling. Dordrecht, The Netherlands: Kluwer Academic Publishers.

41 M.W. Urban and E.M. Salazar-Rojas. 1988. “Ultrasonic PTC Modification of

poly(vinylidene fluoride) Surfaces and Their Characterization,” Macromolecules 21: 372-378.

42 Hesheng Xia, Qi Wang, and Guihua Qiu. 2003. “Polymer-Encapsulated Carbon

Nanotubes Prepared through Ultrasonically Initiated In Situ Emulsion Polymerization, Chemistry of Materials 15, no. 20: 3879-3886

43 Toshio Sato, Takeyoshi Uchida, Akito Endo, Shinichi Takeuchi, Naimu

Kuramochi and Noricmichi Kawashima. 2002. Study on dispersion and

123

surface modification of diamond powders by ultrasound exposure. In Proceedings of the IEEE Ultrasonics Symposium 1: 521-525.

44 S. Ramesh, Y. Koltypin and A. Gedanken. 1997. “Ultrasound driven

aggregation and surface silanol modification in amorphous silica microspheres,” Journal of Materials Research 12, no. 12: 3271-3277.

45 R. Birringer and H. Gleiter. 1998. In: R.W. Cahn, Editor, Advance in Materials

Science, Encyclopedia of materials science and engineering, Volume 1. Pergamon Press, Oxford: 339.

46 H. Gleiter. 1989. “Nanocrystalline Materials,” Progress in Materials Science 33:

223-315. 47 R. Dagani. 1992. “Nanostructural materials promise to advance range of

technologies,” Chemical Engineering & News: 18-24. 48 Hyunsook Kim, Myoung-soon Kim, Hyesun Paik, Yeon-Sook Chung, In Seok

Hong and Junghun Suh. “Effective Artificial Proteases Synthesized by Covering Silica Gel with Aldehyde and Various Other Organic Groups”. Bioorganic & Medicinal Chemistry Letters 12 (2002) 3247–3250

49 2001. Kevlar Home Page. http://www.dupont.com/kevlar/. 50 Spectra Home Page, http://www51.honeywell.com/sm/afc/products-

details/fiber.html 51 Selection and Application Guide to Personal Body Armor, NIJ Guide – 100-01,

November 2001. 52 A.P. Smith and H. Ade. 1996. “Quantitative orientational analysis of a

polymeric material (Kevlar® fibers) with x-ray microspectroscopy,” Applied Physics Letters 69, no. 25: 38833-3835.

53 H. Ade and B. Hsiao. 1993. “X-ray Linear Dichroism Microscopy,” Science

262, no. 5138: 1427-1429. 54 MSDS: Sigma-Aldrich – Polyethylene Glycol av. Mol. Wt. 200, P3015. 55 Sigma Aldrich Product Information, Silica nanopowder 637238. 56 H. Mahfuz. 2006. “Functionalized nanoparticles and their influence on the

properties of nanocomposites,” Presented at the Annual Review meeting for ONR, Washington, D.C.

124

57 Gelest, “Gelest Silane Coupling Agents: Connecting Across Boundaries,” Gelest, Inc., 2006.

58 Sonics, “High Intensity Ultrasound Liquid Processors,” Sonics & Materials,

Inc., 2004. 59 K. Rice. 2000. National Institute of Justice: Stab Resistance of Personal Body

Armor: NIJ Standard – 0115.0, September 2000. 60 John Coates. 2000. “Interpretation of Infrared Spectra, A Practical Approach,”

Encyclopedia of Analytical Chemistry. R.A. Meyers. Chichester: John Wiley & Sons Ltd: 10815-10837.

61 “Department of Defense Test Method Standard: V50 Ballistic Test for Armor,” MIL-STD-662F, 1997.

62 National Institute of Justice: Ballistic Resistance of Personal Body Armor: NIJ

Standard – 0101.4, Revision A, June 2001. 63 Floria Clements, “development of flexible puncture resistant materials system

using nanoparticles”. College of Engineering and Computer Science, 2007

64 Lee et Al, “The ballistic impact characteristics of Kevlar woven fabrics

impregnated with a colloidal shear thickening fluid”. Journal of Materials Science 38 (2003) 2825 – 2833

65 H.Mahfuz, V.Lambert, P.Bordner. V.Rangari “Development of Stab Resistant

Body Armor Using Silated SiO2 Nanoparticles Dispersed into Glutaraldehyde”. NSTI 2008 paper 788

66 NIJ Standard 0115.00 “Stab resistance of personal body armor” National

Institute of Justice, Sept 2000. 67 C.T.Sun, V.Dong, ARO Report 2007

125

APPENDIX

Below are the tables where all the different series of tests have been performed.

Nylon + Gluta 36 12 73,86 0,651 0,318 V n/a 0 1,35 2 2,37D 2 0 2 3 4E #VALUE! 0,00 1,63 3,58 5,03

NE #VALUE! 0,00 2,50 5,50 7,72SPC/gluta 36 12 59,316 0,523 0,255 V 1,28 1,68 2,32 2,35 2,74 2,9 3,1

D 0 0 0 0 1 1 4E 1,70 2,93 5,58 5,73 7,79 8,72 9,97

NE 6,65 11,46 21,86 22,42 30,48 34,15 39,02R-Kevlar/comp 36 12 53,4 0,471 0,230 V 1,69 2,04 2,62 3,01 3,34 3,64 3,93 4,2 4,63

D 0 0 0 0 0 0 0 0 0 0 0 0 0 2E 2,96 4,32 7,12 9,40 11,57 13,74 16,02 18,29 22,23

NE 12,88 18,77 30,96 40,86 50,32 59,76 69,66 79,56 96,69 116,81 127,55 144,75 210,56 238,27C-Kevlar/Comp 36 12 36,66 0,323 0,158 V 1,32 1,61 1,9 2,12 2,36 2,74 3,05 3,43 3,75 4,23 4,51

D 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2E 1,81 2,69 3,74 4,66 5,78 7,79 9,65 12,20 14,58 18,55 21,09

NE 11,45 17,03 23,72 29,53 36,59 49,32 61,12 77,29 92,39 117,55 133,63 160,36 178,18 231,81 292,94C-Kevlar/NEAT 36 12 35,28 0,311 0,152 V 0 1 1,72 2,22 2,65 2,97 3,27 3,64 3,93 4,05 4,43

D 0 0 0 1 1 1 1 1 1 1 1 1 1 4E 0,00 0,90 2,65 4,41 6,29 7,89 9,57 11,86 13,82 14,68 17,56

NE 0,00 2,88 8,51 14,18 20,20 25,38 30,76 38,12 44,43 47,19 56,46 61,59 67,75 75,27R-Kevlar/NEAT 36 12 46,896 0,414 0,202 V 0 0,97 1,4

D 0 3 4E 0,00 0,84 1,75

NE 0,00 2,04 4,24SPECTRA/NEAT 36 12 58,2 0,513 0,251 V 0 0,97 1,39 1,59

D 0 1 3 4E 0,00 0,84 1,73 2,26

Table 5: Results table for NIJ Spike data of 12 layers of independent fabric neat and composite

126

[1]C-Kevlar/[11]SPC/c 36 12 57.854 0.510 0.249 V 0 1.01 1.38 1.78 1.98 2.2 2.34 2.68 2.8 3.01 3.12 3.29D 0 0 0 0 1 1 2 2 3 3 3 4E 0.00 0.91 1.70 2.84 3.51 4.33 4.90 6.43 7.02 8.11 8.71 9.69

NE 0.00 1.79 3.34 5.56 6.88 8.49 9.61 12.60 13.75 15.89 17.08 18.99[2]C-Kevlar/[10]SPC/c 36 12 55.342 0.488 0.238 V 0 1 1.41 1.74 1.97 2.22 2.36 2.59 2.8 3.12

D 0 0 0 0 1 1 2 3 3 4E 0.00 0.90 1.78 2.71 3.47 4.41 4.98 6.00 7.02 8.71

NE 0.00 1.83 3.65 5.55 7.12 9.04 10.21 12.30 14.38 17.85[3]C-Kevlar/[9]SPC/c 36 12 53.793 0.474 0.232 V 0 1 1.42 1.74 1.93 2.16 2.39 2.57 2.77 2.97

D 0 0 0 0 1 2 3 3 3 4E 0.00 0.90 1.80 2.71 3.33 4.18 5.11 5.91 6.87 7.89

NE 0.00 1.89 3.80 5.71 7.03 8.80 10.78 12.46 14.48 16.64[4]C-Kevlar/[8]SPC/c 36 12 52.244 0.461 0.225 V 0 1.03 1.42 1.72 1.95 2.2 2.39 2.51 2.77 3.01

D 0 0 0 1 1 2 3 3 3 4E 0.00 0.95 1.80 2.65 3.40 4.33 5.11 5.64 6.87 8.11

NE 0.00 2.06 3.92 5.75 7.39 9.40 11.10 12.24 14.91 17.60[5]C-Kevlar/[7]SPC/c 36 12 50.695 0.447 0.218 V 0 0.99 1.43 1.72 1.97 2.2 2.39 2.65 2.8

D 0 0 1 2 2 2 3 3 4E 0.00 0.88 1.83 2.65 3.47 4.33 5.11 6.29 7.02

NE 0.00 1.96 4.09 5.92 7.77 9.69 11.44 14.06 15.70[6]C-Kevlar/[6]SPC/c 36 12 49.146 0.433 0.212 V 0 1 1.45 1.77 1.98 2.22 2.41 2.62

D 0 0 1 2 2 3 3 4E 0.00 0.90 1.88 2.80 3.51 4.41 5.20 6.14

NE 0.00 2.07 4.34 6.47 8.10 10.18 11.99 14.18[7]C-Kevlar/[5]SPC/c 36 12 47.597 0.420 0.205 V 0 1.01 1.43 1.73 1.98 2.22 2.39 2.62

D 0 1 2 2 3 3 3 4E 0.00 0.91 1.83 2.68 3.51 4.41 5.11 6.14

NE 0.00 2.18 4.36 6.38 8.36 10.51 12.18 14.64[8]C-Kevlar/[4]SPC/c 36 12 46.048 0.406 0.198 V 0 1 1.41 1.72 1.98 2.21 2.41

D 0 1 2 2 3 3 4E 0.00 0.90 1.78 2.65 3.51 4.37 5.20

NE 0.00 2.20 4.38 6.52 8.64 10.76 12.80[9]C-Kevlar/[3]SPC/c 36 12 44.499 0.392 0.192 V 0 0.98 1.39 1.75 2 2.14

D 0 1 2 3 3 4E 0.00 0.86 1.73 2.74 3.58 4.10

NE 0.00 2.19 4.41 6.98 9.12 10.44[10]C-Kevlar/[2]SPC/c 36 12 42.95 0.379 0.185 V 0 0.97 1.38 1.75 1.93 2.18

D 0 1 2 3 3 4E 0.00 0.84 1.70 2.74 3.33 4.25

NE 0.00 2.22 4.50 7.24 8.80 11.23[11]C-Kevlar/[1]SPC/c 36 12 36.66 0.323 0.158 V 0 1.01 1.41 1.74 1.97

D 0 1 2 3 4E 0.00 0.91 1.78 2.71 3.47

NE 0.00 2.82 5.50 8.38 10.74

Table 6: Results table for NIJ Knife data of 12 layers combination of reinforced

Correctional Kevlar and Spectra

127

[1]C-K/[11]SPC/C 36 12 57,854 0,510 0,249 V 0 1,02 1,4 1,65 1,97 2,24 2,37 2,46 2,74 3,01 3,21 3,64 4,975 54,725D 0 0 0 0 0 0 0 0 0 1 1 4E 0,00 0,93 1,75 2,44 3,47 4,49 5,03 5,42 6,72 8,11 9,22 11,86

NE 0,00 1,83 3,44 4,78 6,81 8,80 9,85 10,62 13,17 15,89 18,08 23,24[2]C-K/[10]SPC/c 36 12 55,342 0,488 0,238 V 0 1,03 1,5 1,8 2,07 2,32 2,58 2,8 2,97 3,01 3,38

D 0 0 0 0 0 0 0 0 0 1 4E 0,00 0,95 2,01 2,90 3,83 4,82 5,96 7,02 7,89 8,11 10,22

NE 0,00 1,95 4,13 5,94 7,86 9,87 12,21 14,38 16,18 16,62 20,95[3]C-K/[9]SPC/c 36 12 53,793 0,474 0,232 V 0 1,15 1,65 2,24 2,57 2,71 2,94 3,05 3,25

D 0 0 0 0 0 1 1 1 4E 0,00 1,18 2,44 4,49 5,91 6,57 7,74 8,33 9,45

NE 0,00 2,50 5,14 9,47 12,46 13,86 16,31 17,55 19,93[4]C-K/[8]SPC/c 36 12 52,244 0,461 0,225 V 0 0,95 1,68 2,14 2,59 2,77 2,94 3,05 3,21 3,48 3,53 3,75

D 0 0 0 0 0 0 0 0 0 0 0 4E 0,00 0,81 2,53 4,10 6,00 6,87 7,74 8,33 9,22 10,84 11,15 12,59

NE 0,00 1,75 5,48 8,90 13,03 14,91 16,79 18,07 20,02 23,53 24,21 27,32[5]C-K/[7]SPC/c 36 12 50,695 0,447 0,218 V 0 0,95 1,68 2,2 2,49 2,94 3,25 3,48 3,58 3,69 4,06

D 0 0 0 0 0 0 0 0 0 0 4E 0,00 0,81 2,53 4,33 5,55 7,74 9,45 10,84 11,47 12,19 14,75

NE 0,00 1,81 5,65 9,69 12,41 17,30 21,15 24,24 25,66 27,26 33,00[6]C-K/[6]SPC/c 36 12 49,146 0,433 0,212 V 0 1 1,68 2,2 2,54 2,86 3,25 3,58 3,69 3,93

D 0 0 0 0 0 0 0 0 0 4E 0,00 0,90 2,53 4,33 5,77 7,32 9,45 11,47 12,19 13,82

NE 0,00 2,07 5,83 10,00 13,32 16,89 21,81 26,47 28,12 31,90[7]C-K/[5]SPC/c 36 12 47,597 0,420 0,205 V 0 0,97 1,73 2,16 2,5 2,97 3,29 3,53 3,75 3,87 4,02 4,13 4,2 4,38

D 0 0 0 0 0 0 0 0 0 0 0 0 0 4E 0,00 0,84 2,68 4,18 5,59 7,89 9,69 11,15 12,59 13,40 14,46 15,27 15,79 17,17

NE 0,00 2,01 6,38 9,95 13,33 18,81 23,08 26,57 29,99 31,94 34,46 36,37 37,61 40,91[8]C-K/[4]SPC/c 36 12 46,048 0,406 0,198 V 0 0,96 1,69 2,16 2,62 2,94 3,29 3,48 3,87 4,2

D 0 0 0 0 0 0 0 0 0 4E 0,00 0,82 2,56 4,18 6,14 7,74 9,69 10,84 13,40 15,79

NE 0,00 2,03 6,29 10,28 15,13 19,05 23,86 26,69 33,01 38,88[9]C-K/[3]SPC/c 36 12 44,499 0,392 0,192 V 0 1,01 1,66 2,1 2,57 2,94 3,25 3,64 3,87 4,13 4,28 4,4

D 0 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,91 2,47 3,95 5,91 7,74 9,45 11,86 13,40 15,27 16,39 17,33

NE 0,00 2,33 6,28 10,06 15,06 19,71 24,09 30,22 34,16 38,90 41,78 44,16[10]C-K/[2]SPC/c 36 12 42,95 0,379 0,185 V 0 0,97 1,69 2,14 2,57 3,05 3,25 3,64 3,88 4,11 4,37

D 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,84 2,56 4,10 5,91 8,33 9,45 11,86 13,47 15,12 17,09

NE 0,00 2,22 6,75 10,82 15,61 21,98 24,96 31,31 35,57 39,92 45,13[11]C-K/[1]SPC/c 36 12 36,66 0,323 0,158 V 0 0,96 1,67 2,1 2,57 3,02 3,21 3,65 3,9 4,08 4,36

D 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,82 2,50 3,95 5,91 8,16 9,22 11,92 13,61 14,90 17,01

NE 0,00 2,55 7,72 12,21 18,29 25,25 28,53 36,88 42,11 46,08 52,63

Table 7: Results table for NIJ Spike data of 12 layers combination of reinforced

Correctional Kevlar and Spectra

128

[11]R-K/[1]SPC/c 36 12 54,04 0,477 0,233 V 0 0,99 1,72 2,11 2,62 2,97 3,29 3,62 3,91 4,08 4,35

D 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,88 2,65 3,98 6,14 7,89 9,69 11,73 13,68 14,90 16,94

NE 0,00 1,84 5,56 8,36 12,89 16,57 20,33 24,61 28,71 31,26 35,54[10]R-K/[2]SPC/c 36 12 54,44 0,480 0,234 V 0 1,01 1,7 2,2 2,59 2,98 3,21 3,58 3,87 4,12 4,43

D 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,91 2,59 4,33 6,00 7,95 9,22 11,47 13,40 15,19 17,56

NE 0,00 1,90 5,39 9,02 12,51 16,56 19,21 23,89 27,92 31,64 36,59[9]R-K/[3]SPC/c 36 12 54,84 0,484 0,236 V 0 0,96 1,65 2,18 2,59 2,94 3,23 3,53 3,87 4,15 4,38

D 0 0 0 0 0 0 0 0 0 0 0E 0,00 0,82 2,44 4,25 6,00 7,74 9,34 11,15 13,40 15,41 17,17

NE 0,00 1,71 5,04 8,80 12,41 16,00 19,31 23,06 27,72 31,87 35,50[8]R-K/[4]SPC/c 36 12 55,24 0,487 0,238 V 0 0,96 1,75 2,13 2,61 2,95 3,28 3,62 3,92 4,2 4,28

D 0 0 0 0 0 0 0 0 0 0 4E 0,00 0,82 2,74 4,06 6,10 7,79 9,63 11,73 13,75 15,79 16,39

NE 0,00 1,69 5,63 8,34 12,52 15,99 19,77 24,08 28,23 32,41 33,66[7]R-K/[5]SPC/c 36 12 55,64 0,491 0,240 V 0 0,97 1,76 2,12 2,61 2,96 3,25 3,63 3,93 4,13

D 0 0 0 0 0 0 0 0 0 4E 0,00 0,84 2,77 4,02 6,10 7,84 9,45 11,79 13,82 15,27

NE 0,00 1,72 5,65 8,20 12,43 15,98 19,27 24,04 28,17 31,11[6]R-K/[6]SPC/c 36 12 56,04 0,494 0,241 V 0 0,97 1,65 2,19 2,56 2,84 3,25 3,57 3,65 3,95

D 0 0 0 0 0 0 0 0 0 4E 0,00 0,84 2,44 4,29 5,87 7,22 9,45 11,41 11,92 13,96

NE 0,00 1,70 4,93 8,69 11,87 14,61 19,13 23,08 24,13 28,26[5]R-K/[7]SPC/c 36 12 56,44 0,498 0,243 V 0 0,98 1,65 2,18 2,76 2,8

D 0 0 0 0 0 4E 0,00 0,86 2,44 4,25 6,82 7,02

NE 0,00 1,73 4,90 8,55 13,70 14,10[4]R-K/[8]SPC/c 36 12 56,84 0,501 0,245 V 0 0,96 1,64 2,19 2,59 2,8 3,01

D 0 0 0 0 0 0 4E 0,00 0,82 2,41 4,29 6,00 7,02 8,11

NE 0,00 1,65 4,80 8,56 11,98 14,00 16,18[3]R-K/[9]SPC/c 36 12 57,24 0,505 0,246 V 0 0,97 1,65 2,17 2,62 2,83

D 0 0 0 0 0 4E 0,00 0,84 2,44 4,21 6,14 7,17

NE 0,00 1,67 4,83 8,35 12,17 14,20[2]R-K/[10]SPC/c 36 12 57,64 0,508 0,248 V 0 0,96 1,64 2,18 2,6 2,92 3,01

D 0 0 0 0 0 0 4E 0,00 0,82 2,41 4,25 6,05 7,63 8,11

NE 0,00 1,62 4,74 8,37 11,90 15,01 15,95[1]R-K/[11]SPC/c 36 12 58,04 0,512 0,250 V 0 1 1,68 2,16 2,57 2,77

D 0 0 0 0 0 4E 0,00 0,90 2,53 4,18 5,91 6,87

NE 0,00 1,75 4,94 8,16 11,55 13,42

Table 8: Results table for NIJ Spike data of 12 layers combination of reinforced Kevlar

and Spectra

129

[11]C-K/[1]SPC/N 36 12 56,92 0,502 0,245 V 0 1,02 1,39 1,72 2 2,2 2,46 2,62 2,77 2,97 3,16 3,25 3,43 3,64 3,87 3,93 4,2 4,28

D 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1E 0,00 0,93 1,73 2,65 3,58 4,33 5,42 6,14 6,87 7,89 8,94 9,45 10,53 11,86 13,40 13,82 15,79 16,39

NE 0,00 1,86 3,45 5,27 7,13 8,63 10,79 12,24 13,68 15,73 17,80 18,83 20,98 23,62 26,70 27,54 31,45 32,66[10]C-K/[2]SPC/N 36 12 57,2 0,504 0,246 V 0 1,02 1,34 1,73 2,01 2,19 2,46 2,62 2,74 2,99 3,15 3,25 3,42 3,65 3,87 3,95 4,2 4,29

D 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1E 0,00 0,93 1,61 2,68 3,62 4,29 5,42 6,14 6,72 8,00 8,88 9,45 10,47 11,92 13,40 13,96 15,79 16,47

NE 0,00 1,85 3,19 5,31 7,17 8,51 10,74 12,18 13,32 15,86 17,61 18,74 20,75 23,64 26,57 27,68 31,30 32,65[9]C-K/[3]SPC/N 36 12 57,48 0,507 0,247 V 0 0,96 1,35 1,75 2 2,19 2,45 2,63 2,74 2,99 3,15 3,25 3,43 3,66 3,88 3,95 4,16 4,3

D 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1E 0,00 0,82 1,63 2,74 3,58 4,29 5,37 6,19 6,72 8,00 8,88 9,45 10,53 11,99 13,47 13,96 15,49 16,55

NE 0,00 1,63 3,22 5,41 7,06 8,47 10,60 12,21 13,26 15,79 17,52 18,65 20,77 23,65 26,58 27,55 30,56 32,65[8]C-K/[4]SPC/N 36 12 57,76 0,509 0,249 V 0 0,96 1,35 1,75 2 2,19 2,45 2,63 2,74 3 3,25 3,64

D 0 0 0 0 0 0 0 1 1 3 3 4E 0,00 0,82 1,63 2,74 3,58 4,29 5,37 6,19 6,72 8,06 9,45 11,86

NE 0,00 1,62 3,20 5,38 7,03 8,43 10,55 12,15 13,19 15,81 18,56 23,28[7]C-K/[5]SPC/N 36 12 58,04 0,512 0,250 V 0 1 1,36 1,74 1,95 2,24 2,46 2,62

D 0 0 1 1 1 2 3 4E 0,00 0,90 1,66 2,71 3,40 4,49 5,42 6,14

NE 0,00 1,75 3,23 5,29 6,65 8,77 10,58 12,00[6]C-K/[6]SPC/N 36 12 58,32 0,514 0,251 V 0 0,98 1,38 1,71 1,97

D 0 1 1 2 4E 0,00 0,86 1,70 2,62 3,47

NE 0,00 1,67 3,31 5,09 6,75[5]C-K/[7]SPC/N 36 12 58,6 0,517 0,252 V 0 1,01 1,35 1,7 1,95

D 0 0 0 1 4E 0,00 0,91 1,63 2,59 3,40

NE 0,00 1,77 3,16 5,01 6,59[4]C-K/[8]SPC/N 36 12 58,88 0,519 0,254 V 0 0,97 1,39 1,72

D 0 0 3 4E 0,00 0,84 1,73 2,65

NE 0,00 1,62 3,33 5,10[3]C-K/[9]SPC/N 36 12 59,16 0,522 0,255 V 0 0,98 1,4 1,75

D 0 0 3 4E 0,00 0,86 1,75 2,74

NE 0,00 1,65 3,36 5,25[2]C-K[10]SPC/N 36 12 59,44 0,524 0,256 V 0 0,99 1,42 1,72

D 0 1 3 4E 0,00 0,88 1,80 2,65

NE 0,00 1,67 3,44 5,05[1]C-K/[11]SPC/N 36 12 59,72 0,527 0,257 V 0 0,99 1,38 1,69

D 0 1 2 4E 0,00 0,88 1,70 2,56

NE 0,00 1,67 3,24 4,85

Table 9: Results table for NIJ Spike data of 12 layers combination of neat Correctional

Kevlar and Spectra

130

[1]R-K/[11]SPC/N 36 12 61,613 0,543 0,265 V 0 0,98 1,38 1,73D 0 0 2 4E 0,00 0,86 1,70 2,68

NE 0,00 1,58 3,14 4,93[2]R-K/[10]SPC/N 36 12 61,186 0,540 0,263 V 0 0,99 1,37 1,65

D 0 1 3 4E 0,00 0,88 1,68 2,44

NE 0,00 1,63 3,11 4,52[3]R-K/[9]SPC/N 36 12 60,759 0,536 0,262 V 0 1,02 1,4 1,72

D 0 1 3 4E 0,00 0,93 1,75 2,65

NE 0,00 1,74 3,27 4,94[4]R-K/[8]SPC/N 36 12 60,332 0,532 0,260 V 0 1 1,42 1,72

D 0 1 3 4E 0,00 0,90 1,80 2,65

NE 0,00 1,68 3,39 4,98[5]R-K/[7]SPC/N 36 12 59,905 0,528 0,258 V 0 0,96 1,38 1,7

D 0 1 3 4E 0,00 0,82 1,70 2,59

NE 0,00 1,56 3,23 4,90[6]R-K/[6]SPC/N 36 12 59,478 0,525 0,256 V 0 0,99 1,41 1,72

D 0 1 3 4E 0,00 0,88 1,78 2,65

NE 0,00 1,67 3,39 5,05[7]R-K/[5]SPC/N 36 12 59,051 0,521 0,254 V 0 0,99 1,42

D 0 2 4E 0,00 0,88 1,80

NE 0,00 1,68 3,47[8]R-K/[4]SPC/N 36 12 58,624 0,517 0,252 V 0 0,99 1,4

D 0 2 4E 0,00 0,88 1,75

NE 0,00 1,70 3,39[9]R-K/[3]SPC/N 36 12 58,197 0,513 0,251 V 0 0,99 1,41

D 0 3 4E 0,00 0,88 1,78

NE 0,00 1,71 3,47[10]R-K/[2]SPC/N 36 12 57,77 0,509 0,249 V 0 0,99 1,41

D 0 3 4E 0,00 0,88 1,78

NE 0,00 1,72 3,49[11]R-K/[1]SPC/N 36 12 57,343 0,506 0,247 V 0 1,02 1,42

D 0 3 4E 0,00 0,93 1,80

NE 0,00 1,84 3,57

Table 10: Results table for NIJ Spike data of 12 layers combination of neat Kevlar and

Spectra

131

Nylon + Gluta 36 12 73,86 0,651 0,318 V 0 0 1,35 2 2,37D 0 0 2 3 4E 0,00 0,00 1,63 3,58 5,03

NE 0,00 0,00 2,50 5,50 7,72Spectra/Gluta/SiO2 36 12 58,86 0,519 0,253 V 0 1,35 1,92 2,39 2,68 3,15 3,35

D 0 0 0 1 2 3 4E 0,00 1,63 3,30 5,11 6,43 8,88 10,04

NE 0,00 3,14 6,36 9,85 12,38 17,11 19,35R-Kevlar/Gluta/Si02 36 12 50,484 0,445 0,217 V 0 1,35 1,67 1,75 1,95

D 0 1 2 3 4E 0,00 1,63 2,50 2,74 3,40

NE 0,00 3,66 5,61 6,16 7,64C-Kevlar/Gluta/Si02 36 12 39,852 0,351 0,172 V 0 1 1,43 1,74 1,97

D 0 1 2 3 4E 0,00 0,90 1,83 2,71 3,47

NE 0,00 2,55 5,21 7,71 9,88C-Kevlar/NEAT 36 12 35,28 0,311 0,152 V 0 1,01 1,41 1,56

D 0 2 3 4E 0,00 0,91 1,78 2,18

NE 0,00 2,93 5,72 7,00R-Kevlar/NEAT 36 12 46,896 0,414 0,202 V 0 1,01 1,4 1,74

D 0 2 3 4E 0,00 0,91 1,75 2,71

NE 0,00 2,21 4,24 6,55SPECTRA/NEAT 36 12 58,2 0,513 0,251 V 0 1 1,42 1,7 1,98 2,18 2,41 2,59 2,9

D 0 1 2 2 3 3 3 3 4E 0,00 0,90 1,80 2,59 3,51 4,25 5,20 6,00 7,53

NE 0,00 1,74 3,52 5,04 6,84 8,29 10,13 11,70 14,67 Table 11: Results table for NIJ Knife data of 12 layers of independent fabric neat and

composite

132

[1]C-Kevlar/[11]SPC/c 36 12 57,854 0,510 0,249 V 0 1,01 1,38 1,78 1,98 2,2 2,34 2,68 2,8 3,01 3,12 3,29D 0 0 0 0 1 1 2 2 3 3 3 4E 0,00 0,91 1,70 2,84 3,51 4,33 4,90 6,43 7,02 8,11 8,71 9,69

NE 0,00 1,79 3,34 5,56 6,88 8,49 9,61 12,60 13,75 15,89 17,08 18,99[2]C-Kevlar/[10]SPC/c 36 12 55,342 0,488 0,238 V 0 1 1,41 1,74 1,97 2,22 2,36 2,59 2,8 3,12

D 0 0 0 0 1 1 2 3 3 4E 0,00 0,90 1,78 2,71 3,47 4,41 4,98 6,00 7,02 8,71

NE 0,00 1,83 3,65 5,55 7,12 9,04 10,21 12,30 14,38 17,85[3]C-Kevlar/[9]SPC/c 36 12 53,793 0,474 0,232 V 0 1 1,42 1,74 1,93 2,16 2,39 2,57 2,77 2,97

D 0 0 0 0 1 2 3 3 3 4E 0,00 0,90 1,80 2,71 3,33 4,18 5,11 5,91 6,87 7,89

NE 0,00 1,89 3,80 5,71 7,03 8,80 10,78 12,46 14,48 16,64[4]C-Kevlar/[8]SPC/c 36 12 52,244 0,461 0,225 V 0 1,03 1,42 1,72 1,95 2,2 2,39 2,51 2,77 3,01

D 0 0 0 1 1 2 3 3 3 4E 0,00 0,95 1,80 2,65 3,40 4,33 5,11 5,64 6,87 8,11

NE 0,00 2,06 3,92 5,75 7,39 9,40 11,10 12,24 14,91 17,60[5]C-Kevlar/[7]SPC/c 36 12 50,695 0,447 0,218 V 0 0,99 1,43 1,72 1,97 2,2 2,39 2,65 2,8

D 0 0 1 2 2 2 3 3 4E 0,00 0,88 1,83 2,65 3,47 4,33 5,11 6,29 7,02

NE 0,00 1,96 4,09 5,92 7,77 9,69 11,44 14,06 15,70[6]C-Kevlar/[6]SPC/c 36 12 49,146 0,433 0,212 V 0 1 1,45 1,77 1,98 2,22 2,41 2,62

D 0 0 1 2 2 3 3 4E 0,00 0,90 1,88 2,80 3,51 4,41 5,20 6,14

NE 0,00 2,07 4,34 6,47 8,10 10,18 11,99 14,18[7]C-Kevlar/[5]SPC/c 36 12 47,597 0,420 0,205 V 0 1,01 1,43 1,73 1,98 2,22 2,39 2,62

D 0 1 2 2 3 3 3 4E 0,00 0,91 1,83 2,68 3,51 4,41 5,11 6,14

NE 0,00 2,18 4,36 6,38 8,36 10,51 12,18 14,64[8]C-Kevlar/[4]SPC/c 36 12 46,048 0,406 0,198 V 0 1 1,41 1,72 1,98 2,21 2,41

D 0 1 2 2 3 3 4E 0,00 0,90 1,78 2,65 3,51 4,37 5,20

NE 0,00 2,20 4,38 6,52 8,64 10,76 12,80[9]C-Kevlar/[3]SPC/c 36 12 44,499 0,392 0,192 V 0 0,98 1,39 1,75 2 2,14

D 0 1 2 3 3 4E 0,00 0,86 1,73 2,74 3,58 4,10

NE 0,00 2,19 4,41 6,98 9,12 10,44[10]C-Kevlar/[2]SPC/c 36 12 42,95 0,379 0,185 V 0 0,97 1,38 1,75 1,93 2,18

D 0 1 2 3 3 4E 0,00 0,84 1,70 2,74 3,33 4,25

NE 0,00 2,22 4,50 7,24 8,80 11,23[11]C-Kevlar/[1]SPC/c 36 12 36,66 0,323 0,158 V 0 1,01 1,41 1,74 1,97

D 0 1 2 3 4E 0,00 0,91 1,78 2,71 3,47

NE 0,00 2,82 5,50 8,38 10,74

Table 12: Results table for NIJ Knife data of 12 layers combination of reinforced Correctional Kevlar and Spectra

133

[11]R-Kevlar/[1]SPC/c 36 12 54.04 0.477 0.233 V 0 1.01 1.38 1.73 1.85 2.2D 0 1 2 3 3 4E 0.00 0.91 1.70 2.68 3.06 4.33

NE 0.00 1.92 3.58 5.62 6.43 9.09[10]R-Kevlar/[2]SPC/c 36 12 54.44 0.480 0.234 V 0 0.99 1.4 1.75 2.24 2.38 2.62

D 0 1 2 2 3 3 4E 0.00 0.88 1.75 2.74 4.49 5.07 6.14

NE 0.00 1.83 3.65 5.71 9.35 10.56 12.80[9]R-Kevlar/[3]SPC/c 36 12 54.84 0.484 0.236 V 0 1.12 1.59 1.95 2.16 2.49 2.59

D 0 1 2 2 3 3 4E 0.00 1.12 2.26 3.40 4.18 5.55 6.00

NE 0.00 2.32 4.68 7.04 8.63 11.47 12.41[8]R-Kevlar/[4]SPC/c 36 12 55.24 0.487 0.238 V 0 0.97 1.27 1.47 1.87 2 2.3 2.45 2.9

D 0 1 1 2 2 3 3 3 4E 0.00 0.84 1.44 1.93 3.13 3.58 4.73 5.37 7.53

NE 0.00 1.73 2.96 3.97 6.42 7.35 9.72 11.03 15.45[7]R-Kevlar/[5]SPC/c 36 12 55.64 0.491 0.240 V 0 1.13 1.58 1.83 2.07 2.3 2.62 2.82

D 0 1 2 3 3 3 3 4E 0.00 1.14 2.23 3.00 3.83 4.73 6.14 7.12

NE 0.00 2.33 4.55 6.11 7.82 9.65 12.52 14.51[6]R-Kevlar/[6]SPC/c 36 12 56.04 0.494 0.241 V 0 1.01 1.4 1.75 1.98 2.22 2.44 2.83

D 0 1 2 2 2 3 3 4E 0.00 0.91 1.75 2.74 3.51 4.41 5.33 7.17

NE 0.00 1.85 3.55 5.55 7.10 8.93 10.78 14.50[5]R-Kevlar/[7]SPC/c 36 12 56.44 0.498 0.243 V 0 1 1.43 1.75 2.22 2.46 2.97 3.02

D 0 1 1 2 3 3 3 4E 0.00 0.90 1.83 2.74 4.41 5.42 7.89 8.16

NE 0.00 1.80 3.68 5.51 8.86 10.88 15.86 16.40[4]R-Kevlar/[8]SPC/c 36 12 56.84 0.501 0.245 V 0 1.02 1.41 1.75 1.97 2.22 2.34 2.65 2.83 3.05

D 0 1 1 2 2 3 3 3 3 4E 0.00 0.93 1.78 2.74 3.47 4.41 4.90 6.29 7.17 8.33

NE 0.00 1.86 3.55 5.47 6.93 8.80 9.78 12.54 14.30 16.61[3]R-Kevlar/[9]SPC/c 36 12 57.24 0.505 0.246 V 0 1.02 1.4 1.74 2.01 2.22 2.44 2.57 2.83 2.97 3.07 3.27

D 0 0 1 2 2 2 3 3 3 3 3 4E 0.00 0.93 1.75 2.71 3.62 4.41 5.33 5.91 7.17 7.89 8.44 9.57

NE 0.00 1.84 3.48 5.37 7.16 8.74 10.56 11.71 14.20 15.64 16.71 18.96[2]R-Kevlar/[10]SPC/c 36 12 57.64 0.508 0.248 V 0 1.03 1.45 1.81 1.98 2.2 2.41 2.65 2.8 3.05 3.34

D 0 0 1 2 2 2 3 3 3 3 4E 0.00 0.95 1.88 2.93 3.51 4.33 5.20 6.29 7.02 8.33 9.98

NE 0.00 1.87 3.70 5.77 6.90 8.52 10.23 12.37 13.80 16.38 19.64[1]R-Kevlar/[11]SPC/c 36 12 58.04 0.512 0.250 V 0 1.03 1.46 1.82 2 2.28 2.49 2.54 2.87 3.16 3.22 3.48

D 0 0 0 1 2 2 2 2 3 3 3 4E 0.00 0.95 1.91 2.96 3.58 4.65 5.55 5.77 7.37 8.94 9.28 10.84

NE 0.00 1.86 3.73 5.79 6.99 9.09 10.84 11.28 14.40 17.46 18.13 21.18

Table 13: Results table for NIJ Knife data of 12 layers combination of reinforced

Kevlar and Spectra

134

[1]R-Kevlar/[11]SPC/NEAT 36 12 61.613 0.543 0.265 V 0 1 1.48 1.64 1.98 2.22 2.41 2.57 2.87D 0 1 2 2 3 3 3 3 4E 0.00 0.90 1.96 2.41 3.51 4.41 5.20 5.91 7.37

NE 0.00 1.65 3.61 4.43 6.46 8.12 9.57 10.88 13.57[2]R-Kevlar/[10]SPC/NEAT 36 12 61.186 0.540 0.263 V 0 1.01 1.4 1.74 2 2.32 2.57

D 0 1 2 2 3 3 4E 0.00 0.91 1.75 2.71 3.58 4.82 5.91

NE 0.00 1.69 3.25 5.02 6.63 8.93 10.96[3]R-Kevlar/[9]SPC/NEAT 36 12 60.759 0.536 0.262 V 0 1.02 1.41 1.74 2 2.16 2.44

D 0 2 2 3 3 3 4E 0.00 0.93 1.78 2.71 3.58 4.18 5.33

NE 0.00 1.74 3.32 5.06 6.68 7.79 9.94[4]R-Kevlar/[8]SPC/NEAT 36 12 60.332 0.532 0.260 V 0 1.01 1.43 1.77 1.98 2.22 2.37 2.59

D 0 2 2 3 3 3 3 4E 0.00 0.91 1.83 2.80 3.51 4.41 5.03 6.00

NE 0.00 1.72 3.44 5.27 6.59 8.29 9.45 11.28[5]R-Kevlar/[7]SPC/NEAT 36 12 59.905 0.528 0.258 V 0 1.02 1.45 1.77 1.98 2.24 2.4

D 0 2 2 3 3 3 4E 0.00 0.93 1.88 2.80 3.51 4.49 5.16

NE 0.00 1.76 3.56 5.31 6.64 8.50 9.76[6]R-Kevlar/[6]SPC/NEAT 36 12 59.478 0.525 0.256 V 0 1.03 1.45 1.78 2 2.22 2.41

D 0 2 2 3 3 3 4E 0.00 0.95 1.88 2.84 3.58 4.41 5.20

NE 0.00 1.81 3.59 5.41 6.83 8.41 9.91[7]R-Kevlar/[5]SPC/NEAT 36 12 59.051 0.521 0.254 V 0 1.04 1.44 1.74 2.07 2.26

D 0 2 3 3 3 4E 0.00 0.97 1.86 2.71 3.83 4.57

NE 0.00 1.86 3.56 5.20 7.36 8.78[8]R-Kevlar/[4]SPC/NEAT 36 12 58.624 0.517 0.252 V 0 1.03 1.42 1.73 2 2.12

D 0 2 3 3 3 4E 0.00 0.95 1.80 2.68 3.58 4.02

NE 0.00 1.84 3.49 5.18 6.92 7.78[9]R-Kevlar/[3]SPC/NEAT 36 12 58.197 0.513 0.251 V 0 1.01 1.43 1.77 2

D 0 2 3 3 4E 0.00 0.91 1.83 2.80 3.58

NE 0.00 1.78 3.57 5.46 6.98[10]R-Kevlar/[2]SPC/NEAT 36 12 57.77 0.509 0.249 V 0 1.01 1.47 1.75 1.97

D 0 2 3 3 4E 0.00 0.91 1.93 2.74 3.47

NE 0.00 1.79 3.80 5.38 6.82[11]R-Kevlar/[1]SPC/NEAT 36 12 57.343 0.506 0.247 V 0 1.02 1.41 1.72 1.97

D 0 2 3 3 4E 0.00 0.93 1.78 2.65 3.47

NE 0.00 1.84 3.52 5.24 6.87

Table 14: Results table for NIJ Spike data of 12 layers combination of neat Kevlar and

Spectra

135

ANSYS CODE

/PREP7 /OUTPUT,AMD03-OUTPUT,TXT !!!!!!!!!!!!!!!! !!!ELEMENT CHOICE !!!!!!!!!!!!!!!!!!!!! ET,1,92 !SOLID92 FOR PLATE ET,2,SOLID186 !SOLID186 FOR SPIKE KEYOPT,2,6,1 !KEYOPT,2,10,1 ET,3,CONTA178,0,,0,0,5,,1,,0,3 R,3,,0.002,1.0 !ET,4,SOLSH190 !KEYOPT,4,8,0 !!!!!!!!!!!!!!!! !!!MATERIAL CHOICE !!!!!!!!!!!!!!!!!!!!! MP,EX,1,207E9 !MATERIAL TYPE1 FOR SPIKE MP,NUXY,1,.3 MP,DENS,1,7850 MP,EX,2,6.5E9 !MATERIAL TYPE2 FOR PLATE MP,EY,2,5.5E9 !RUBBER MP,EZ,2,6.5E9 MP,PRXY,2,.35 MP,PRYZ,2,.35 MP,PRXZ,2,.35 MP,GXY,2,24.4E9

136

MP,GYZ,2,24.4E9 MP,GXZ,2,24.4E9 MP,DENS,2,1.52E-6 MP,EX,3,565E3 !CROSS LINKED POLYETHYLENE MP,EY,3,83E3 MP,EZ,3,565E3 MP,PRXY,3,.35 MP,PRYZ,3,.35 MP,PRXZ,3,.35 MP,GXY,3,0.117E9 MP,GYZ,3,0.117E9 MP,GXZ,3,.117E9 MP,DENS,3,45E-6 MP,EX,4,1050E3 !NEOPRENE MP,EY,4,147E3 MP,EZ,4,1050E3 MP,PRXY,4,.3 MP,PRYZ,4,.3 MP,PRXZ,4,.3 MP,GXY,4,27.2E3 MP,GYZ,4,27.2E3 MP,GXZ,4,27.2E-6 MP,DENS,4,192 MP,EX,5,124E9 !MATERIAL TYPE2 FOR PLATE MP,EY,5,112.4E9 !KEVLAR MP,EZ,5,124E9 MP,PRXY,5,.36 MP,PRYZ,5,.36 MP,PRXZ,5,.36 MP,GXY,5,24.4E9 MP,GYZ,5,24.4E9 MP,GXZ,5,24.4E9 MP,DENS,5,1.44E-6 !!!!!!!!!!!!!!!! !!!GEOMETRY !!!!!!!!!!!!!!!!!!!!! !BACKING MATERIAL K,1,-152.4,0,-152.4 !KEYPOINTS FOR THE PLATE K,2,-152.4,0,0

137

K,3,0,0,0 K,4,0,0,-152.4 A,1,2,3,4 !CREATE VOLUME LAYER 1 ASEL,S,AREA,,1 VOFFST,1,6.3 MAT,2 ALLSEL,ALL ASEL,S,AREA,,2 !VOLUME 2 VOFFST,2,6.3 MAT,2 ALLSEL,ALL ASEL,S,AREA,,7 !VOLUME 3 VOFFST,7,33 MAT,3 ALLSEL,ALL ASEL,S,AREA,,12 !VOLUME 4 VOFFST,12,5 MAT,4 ALLSEL,ALL ASEL,S,AREA,,17 !VOLUME 5 VOFFST,17,5 MAT,4 ALLSEL,ALL ASEL,S,AREA,,22 !VOUME 6 VOFFST,22,5 MAT,4 ALLSEL,ALL ASEL,S,AREA,,27 !VOUME 7 VOFFST,27,5 MAT,4 ALLSEL,ALL ASEL,S,AREA,,32 !VOUME 8 kevlar VOFFST,32,5 MAT,5 ALLSEL,ALL

138

!SPIKE HI=68 H01=HI K,128,0,H01,0 !KEYPOINTS FOR THE SPIKE K,129,2E-1,H01,0 K,130,2E-1,(H01+1),0 K,131,2,(H01+80),0 K,132,2,(H01+180),0 K,133,50,(H01+180),0 !KEYPOINTS FOR THE MASS K,134,50,(H01+690),0 K,135,0,(H01+690),0 A,128,129,130,131,132,133,134,135 !CREATE AREA 37 !!!!!!!!!!!!!!!! !!!MESHING !!!!!!!!!!!!!!!!!!!!! !PLATE MESH TYPE,1 !DEFINE TYPE1 AND MESH THE PLATE VSYMM,X,1,8,1,,0,0 ALLSEL,ALL VSEL,ALL VSYMM,Z,ALL,,,,0,0 ALLSEL,ALL NUMMRG,KP,1E-4,,,LOW !MERGE KEYPOINTS AND AREAS VADD,1,8,15,22 VADD,2,9,16,23 VADD,3,10,17,24 VADD,4,11,18,25 VADD,5,12,19,26 VADD,6,13,20,27 VADD,7,14,21,28 VADD,8,15,22,29 ESIZE,100

139

VMESH,ALL ALLSEL,ALL ESEL,S,TYPE,,1 NSLE CM,PLATE,NODE ALLSEL,ALL NSEL,S,LOC,Y,0 !SELECT BOTTOM OF PLATE AND CREATE COMPONENT BOTTOM CM,BOTTOM,NODE ALLSEL,ALL NSEL,S,LOC,Y,68 CM,TOP,NODE ALLSEL,ALL !VSEL,S,VOLU,8,32,8 !VCLEAR,8,32,8 !TYPE,4 !VSEL,S,VOLU,8,32,8 !ALLSEL,ALL !ESIZE,100 !VMESH,ALL !ALLSEL,ALL !SPIKE MESH TYPE,2 !DEFINE TYPE2 + CREATE CYLINDER + MESH THE SPIKE MOPT,TETEXPND,2 ASEL,S,AREA,,42 MSHKEY,0 MSHAPE,1,3D VROTAT,42,0,0,0,0,0,128,135,,4 ALLSEL,ALL ESIZE,20

140

VMESH,ALL ESEL,S,TYPE,,2 !SELECT AND CREATE COMPONENT SPIKE NSLE MAT,1 !DEFINE MATERIAL FOR SPIKE CM,SPIKE,NODE ALLSEL,ALL ESEL,S,TYPE,,1 NSLE CM,PLATE,NODE ALLSEL,ALL NUMMRG,NODES,1E-4,,,LOW !MERGE NODES !GAP MESH TYPE,3 !CONTAC178 NSEL,S,NODE,,929 !PLATE NODE NSEL,S,NODE,,3656 !SPIKE NODE E,929,3656 !CREATE GAP ELEMENT %%%%% !DIRECTION IS IMPORTANT FIRST NODE CORRESPOND TO THE MOVING NODE ALLSEL,ALL TYPE,1 !ESEL,S,TYPE,,1 !NSLE NSEL,S,NODE,,SPIKE !DEFINE BOUNDARY CONDITIONS FOR THE SPIKE D,ALL,ACCY,-9.81E3 ALLSEL,ALL !NSEL,S,NODE,,SPIKE !DEFINE BOUNDARY CONDITIONS FOR THE SPIKE !D,ALL,VELY,-5 !0.08333 !ALLSEL,ALL

141

NSEL,S,NODE,,BOTTOM !DEFINE BOUNDARY CONDITIONS FOR THE PLATE D,ALL,ALL ALLSEL,ALL !NSEL,S,NODE,,1915 !D,ALL,UY,-25 !ALLSEL,ALL FINISH !EXIT PROCESSOR !!WORK!! c****************** c**** SOLUTION c****************** /SOLU !ENTER SOLUTION PROCESS !SOLCONTROL,ON NLGEOM,ON TIMINT,ON KBC,0 TIME,300 DELTIM,10 !SOLVE !AUTOTS,ON !TIMINT,ON OUTRES,ALL,ALL SOLVE

142

FINISH c****************** c**** POST26 c****************** /POST1 NSOL,1,929,U,Y PRNSOL,U,Y ESOL,2,495,929,S,EQV PRESOL,S, !/GRID,1 !XVAR,2 !NSOL,3,929,U,Y !DISPLACEMENT OF MIDDLE PLATE NODE Y !ESOL,4,495,929,S,EQV !EQUIVALENT STRESS !ADD,2,2,,,DISP,,,-1 !/AXLAB,X,DISPLACEMENT [M] !PLOT DISPLACEMENT ON X !/AXLAB,Y,EQV STRESS [Pa] !PLOT EQUIVALENT STRESS !PLVAR,3 !FINISH !/POST26 !NSOL,2,929,U,Y,IMPACT_POINT !ESOL,3,530,929,S,EQV,VON_MISES !PROD,4,2,,,,,,-1 !/AXLAB,X,DISPLACEMENT !/XRANGE,0,3

143

!/AXLAB,Y,STRESS !/YRANGE,0,1.2E9 !/AUTO,1 !XVAR,4 !PRVA,3 !FINISH

enhancement of spike and stab resistance...

Documents