Application of High Voltage, High Frequency Pulsed ...eprints.qut.edu.au/53105/1/Hajarossadat_Asgarifar_Thesis.pdf · Application of High Voltage, High Frequency Pulsed Electromagnetic

Application of High Voltage, High Frequency

Pulsed Electromagnetic Field on Cortical Bone

Tissue

This thesis submitted as a requirement for the degree of

Master of Engineering

Hajarossadat Asgarifar

B.Eng (Electrical)

School of Biomedical Engineering and Medical Physics

Faculty of Science and Engineering

Queensland University of Technology

Brisbane, Australia

June 2012

II

Statement of Originality

The work contained in this thesis has not been previously submitted to meet requirements

for an award at this or any other higher education institution. To the best of my knowledge

and belief, the thesis contains no material previously published or written by another

person except where due reference is made.

Hajarossadat Asgarifar

III

Acknowledgments

My deep foremost gratitude to the creature of the world, Allah, who all what I have

is his blessing.

Next, I express my sincere thanks and gratitude to the following generous people

whom, the completion of this work was not possible without their support, patience,

encouragement and guidance:

My supervisors, Prof Kunle Oloyede and Associate Prof Firuz Zare for their

invaluable guidance and support

The many academic and technical staff and PhD students at IHIB for their

kind consultancies and assistances, in particular, Prof Christian Langton for

ultrasound facilities and medical engineering laboratory technicians and

research portfolio staff for their technical advices and continued helps

My friends and colleagues for sharing knowledge and providing a warm

research environment

The last but not the least, to my unique family, my beloved husband, Mehran,

for his most amazing support and great advices and my gorgeous favourite

twins, Hossein and Mahdi, for their kindness and patience all through my

study

And to my dear parents for their infinite love, spiritual support and

encouragement during my life and study even when I was too far from them

IV

Keywords

Pulsed Power

Cortical bone

High voltage, High frequency converter

Positive Buck-Boost Converter

Pulsed electromagnetic field

Electrical stimulation

Mechanical properties of bone

Bone functional behaviour

V

Abstract

Over the last few decades, electric and electromagnetic fields have achieved

important role as stimulator and therapeutic facility in biology and medicine. In particular,

low magnitude, low frequency, pulsed electromagnetic field has shown significant positive

effect on bone fracture healing and some bone diseases treatment. Nevertheless, to date,

little attention has been paid to investigate the possible effect of high frequency, high

magnitude pulsed electromagnetic field (pulse power) on functional behaviour and

biomechanical properties of bone tissue.

Bone is a dynamic, complex organ, which is made of bone materials (consisting of

organic components, inorganic mineral and water) known as extracellular matrix, and bone

cells (live part). The cells give the bone the capability of self-repairing by adapting itself to

its mechanical environment. The specific bone material composite comprising of collagen

matrix reinforced with mineral apatite provides the bone with particular biomechanical

properties in an anisotropic, inhomogeneous structure.

This project hypothesized to investigate the possible effect of pulse power signals on

cortical bone characteristics through evaluating the fundamental mechanical properties of

bone material. A positive buck-boost converter was applied to generate adjustable high

voltage, high frequency pulses up to 500 V and 10 kHz.

Bone shows distinctive characteristics in different loading mode. Thus, functional

behaviour of bone in response to pulse power excitation were elucidated by using three

different conventional mechanical tests applying three-point bending load in elastic region,

tensile and compressive loading until failure. Flexural stiffness, tensile and compressive

VI

strength, hysteresis and total fracture energy were determined as measure of main bone

characteristics. To assess bone structure variation due to pulse power excitation in deeper

aspect, a supplementary fractographic study was also conducted using scanning electron

micrograph from tensile fracture surfaces.

Furthermore, a non-destructive ultrasonic technique was applied for determination

and comparison of bone elasticity before and after pulse power stimulation. This method

provided the ability to evaluate the stiffness of millimetre-sized bone samples in three

orthogonal directions.

According to the results of non-destructive bending test, the flexural elasticity of

cortical bone samples appeared to remain unchanged due to pulse power excitation.

Similar results were observed in the bone stiffness for all three orthogonal directions

obtained from ultrasonic technique and in the bone stiffness from the compression test.

From tensile tests, no significant changes were found in tensile strength and total strain

energy absorption of the bone samples exposed to pulse power compared with those of the

control samples. Also, the apparent microstructure of the fracture surfaces of PP-exposed

samples (including porosity and microcracks diffusion) showed no significant variation

due to pulse power stimulation. Nevertheless, the compressive strength and toughness of

millimetre-sized samples appeared to increase when the samples were exposed to 66 hours

high power pulsed electromagnetic field through screws with small contact cross-section

(increasing the pulsed electric field intensity) compare to the control samples. This can

show the different load-bearing characteristics of cortical bone tissue in response to pulse

power excitation and effectiveness of this type of stimulation on smaller-sized samples.

These overall results may address that although, the pulse power stimulation can

influence the arrangement or the quality of the collagen network causing the bone strength

VII

and toughness augmentation, it apparently did not affect the mineral phase of the cortical

bone material. The results also confirmed that the indirect application of high power pulsed

electromagnetic field at 500 V and 10 kHz through capacitive coupling method, was

athermal and did not damage the bone tissue construction.

VIII

Contribution

High power pulsed electromagnetic field (Pulse Power), has been applied recently in

some fields of biology and medicine. However, the effect of pulse power on physical

characteristics of bone tissue has not yet been fully clarified. On the other hand, according

to various studies during last century, electrical stimulation using both constant and pulsed

electromagnetic field (PEMF) has had a drastic effect on bone growth and some bone

diseases healing. It was a good motivation for investigation of the possibility of applying

pulse power signals for stimulating bone.

The main contribution of the present thesis is to introduce a suitable, safe method

with controlled parameters for application of high power, pulsed electromagnetic fields on

bone tissue using capacitive coupling method. The basic biomechanical properties of

cortical bone material including stiffness, strength, toughness and brittleness have been

investigated (considering just extracellular fraction of the bone) in response to high

voltage, high frequency pulses up to 500V at 10 kHz. These have been achieved by:

The comparison and assessment of two pulse power application methods, direct

connection of bone with electrodes (which result in thermal effect and burning) and

capacitive coupling method through electrodes isolation (Chapter 4).

The determination and comparison of bone flexural elasticity before and after pulse

power excitation using the non-destructive three-point bending tests (in linear

elastic region) on both whole long bone and cortical bone strips (Chapter 4).

IX

The study of the bone fracture behaviour in response to high voltage, high

frequency pulsed electromagnetic field using tensile test until failure point by

investigation of fracture energy, hysteresis energy and strength of the samples

exposed to pulse power compared with those of the control samples and

supplementary fractograph study via scanning electron microscopy of the fracture

surfaces (Chapter 5).

The evaluation of the compressive strength and fracture energy of the millimetre-

sized cortical bone samples exposed to pulse power signals compared with the

control specimens (Chapter 6).

The application of ultrasonic technique as an alternative, non-destructive method

with the capability of measurement in different orthogonal directions for

determination and comparison of elastic property of cortical bone samples in

response to pulse power excitation (Chapter 7).

To author’s knowledge, this project was the first research investigating the effect of

high voltage, high frequency pulsed electromagnetic field on fundamental properties of

cortical bone structure.

Providing a basic information about the effect of pulse power excitation on bone

tissue structure, this study will contribute in further research on pulse power application on

live bone, investigating the bone growth enhancement potential of this kind of stimulation

for therapeutic purpose in musculoskeletal diseases.

Some of the results of this research were presented as accepted international

conference paper and item as below and other is going to submit as a journal paper:

H. Asgarifar, A. Oloyede, F. Zare, C. M. Langton “Evaluation of cortical bone

elasticity in response to pulse power excitation using ultrasonic technique” Ninth

X

IASTED International Conference on Biomedical Engineering (Biomed 2012), Feb.

2012. Innsbruck, Austria

H. Asgarifar, A. Oloyede, F. Zare “Investigation of high frequency, high voltage

pulses application on bending properties of bone” EPSM-ABEC Conference, Aug

2011, Darwin Australia.

XI

Table of Contents

Statement of Originality .................................................................................. II

Acknowledgments .......................................................................................... III

Keywords ........................................................................................................ IV

Abstract ............................................................................................................. V

Contribution ................................................................................................ VIII

Table of Contents ........................................................................................... XI

List of Figures ........................................................................................... XVIII

List of Tables .............................................................................................. XXV

List of Abbreviations and Symbols ........................................................ XXVII

Chapter 1: Introduction ................................................................................. 1

Chapter 2: Physical Behaviour and Electrical Stimulation of Bone .......... 8

2.1 Introduction................................................................................................ 9

2.2 Hierarchical architecture of bone............................................................. 10

2.3 Cortical bone structure ............................................................................. 12

2.3.1 Bone cells................................................................................................. 12

2.3.2 Extracellular matrix (ECM) architecture ................................................. 15

Collagen fibrils arrangement ....................................................... 17

Mineral crystals structure ............................................................. 18

The water content ......................................................................... 19

XII

2.4 Contribution of the bone constitutes at different hierarchical levels on its

mechanical competence .............................................................................................. 20

2.4.1 The bone basic elements (molecular level) ............................................. 22

Collagen fibrils ............................................................................ 22

The mineral crystals ..................................................................... 24

The bone water content ................................................................ 24

2.4.2 The mineralized collagen fibrils (nanoscale level) .................................. 26

2.4.3 The arrays of the collagen fibrils (mesoscale level) ................................ 26

2.4.4 The organization of the fibril arrays in lamellae and osteon (microscale

level) ........................................................................................................ 27

2.5 Bioelectric phenomena in bone ............................................................... 28

2.5.1 The origin of the stress generated potential (SGP) in bone ..................... 29

2.5.2 Electrical stimulation of bone with low intensity electromagnetic field . 31

Application of direct contact method for bone tissue stimulation31

Application of the pulsed electromagnetic field stimulation on bone

tissue ...................................................................................................... 33

Inductive coupling ............................................................................................. 34

Capacitive coupling ........................................................................................... 36

2.5.3 Influential factors in electrical stimulation methods ............................... 38

2.5.4 Some of the hypothesized mechanisms involved in bone generation due

to pulsed electromagnetic field ................................................................ 40

2.5.5 The effect of low intensity pulsed electromagnetic field on biomechanical

properties of bone .................................................................................... 41

2.5.6 Application of high intensity pulsed electromagnetic field on bone ....... 42

XIII

Chapter 3: Pulse Power Generator Based on Positive Buck-Boost Converter

44

3.1 Introduction.............................................................................................. 45

3.2 Topology of pulse power generator ......................................................... 47

3.2.1 General configuration of positive Buck-Boost Converter ....................... 47

3.2.2 Switching Modes ..................................................................................... 49

First state: charging inductor (S1: on, S2: on) ............................. 49

Second state: circulating the inductor current (S1: off, S2: on) ... 49

Third state: charging the capacitor and load supplying (S1: off, S2: off)

............................................................................................................... 50

3.3 The pulse power generators applied in this study .................................... 52

3.4 Load modeling ......................................................................................... 55

Chapter 4: Physical Characterisation of Bone Exposed to Pulse Power in

Bending 57

4.1 Introduction.............................................................................................. 58

4.2 Factors influencing experimental measurement ...................................... 59

4.3 Materials and Methods ............................................................................ 62

4.3.1 Sample preparation .................................................................................. 62

4.3.2 Three-point bending test .......................................................................... 63

4.3.3 Data collection and calculation ................................................................ 65

4.3.4 Pulse Power excitation ............................................................................. 68

4.4 Experimental procedure and Results ....................................................... 69

4.4.1 Pulse power excitation with voltage up to 180V and 100 Hz frequency . 69

XIV

Whole bone stimulation with Pulses of 180 V at 100 Hz ............ 70

Bone strips stimulation with pulses of 180 V at 100 Hz ............. 71

4.4.2 Pulse power excitation with pulses up to 450 V magnitude at 10 kHz

frequency ................................................................................................. 73

Pulses up to 450 V at 340 Hz ....................................................... 73

Pulses up to 450 V at 10 kHz ....................................................... 75

4.5 Discussion ................................................................................................ 77

Chapter 5: Effect of Pulse Power Exposure on Functional Behaviour of

Cortical Bone in Tension .......................................................................................... 80

5.1 Introduction.............................................................................................. 81

5.1.1 Fractographic study ................................................................................. 83

5.2 Materials and Methods ............................................................................ 84

5.2.1 Practical consideration for tensile testing ................................................ 84

5.2.2 Sample preparation .................................................................................. 85

5.2.3 Pulse Power excitation ............................................................................. 88

5.2.4 Uniaxial quasi-static tensile test .............................................................. 90

5.2.5 Scanning electron fractograph ................................................................. 91

Sample preparation for SEM procedure ...................................... 91

5.3 Experimental procedure and Results ....................................................... 92

5.3.1 Dumbbell shape tensile test samples with round junction versus those

with sharp junction .................................................................................. 92

5.3.2 Hysteresis energy absorption for PP-exposed samples versus the control

samples .................................................................................................... 94

5.3.3 Tensile toughness and strength measurement.......................................... 96

XV

5.3.4 Fractographic examination using SEM ................................................... 98

5.4 Discussion .............................................................................................. 105

Chapter 6: Effect of Pulse Power Excitation on Basic Mechanical Properties of

Cortical Bone in Compression ............................................................................... 110

6.1 Introduction............................................................................................ 111

6.2 Materials and Methods .......................................................................... 112

6.2.1 Sample preparation ................................................................................ 112

6.2.2 Experimental Procedure......................................................................... 113

Bone samples stimulation with pulse power signals ................. 113

Compressive testing ................................................................... 115

6.3 Toughness and strength measurement (results) ..................................... 116

6.4 Discussion .............................................................................................. 120

Chapter 7: Evaluation of Cortical Bone Elasticity in Response to Pulse Power

Excitation Using Ultrasonic Technique ................................................................ 122

7.1 Introduction............................................................................................ 123

7.2 The theoretical consideration ................................................................. 125

7.3 Materials and Methods .......................................................................... 127

7.3.1 Sample preparation ................................................................................ 127

7.3.2 Density measurement............................................................................. 128

7.3.3 Experimental Procedure......................................................................... 129

Ultrasound velocity measurement ............................................. 129

Pulse Power excitation ............................................................... 132

XVI

7.4 Results ................................................................................................... 133

7.5 Discussion .............................................................................................. 135

Chapter 8: Effect of Pulse Power Stimulation on Functional and Physical

Characteristics of Cortical Bone (Discussion and Conclusion) .......................... 138

8.1 Introduction............................................................................................ 139

8.2 Research procedure description and justification .................................. 141

8.2.1 Introduction of a suitable pulse power application set up and evaluation

of the flexural elasticity of cortical bone through non-destructive 3-point

bending test ............................................................................................ 142

8.2.2 The effect of pulse power exposure on the tensile strength and total

fracture energy accompanying the microstructure analysis of the test

bone fracture surfaces ............................................................................ 143

8.2.3 The effect of the pulse power excitation on the compressive strength and

toughness of the small sized samples .................................................... 144

8.2.4 Application of ultrasonic technique to evaluate the effect of pulse power

on bone elasticity ................................................................................... 144

8.3 The effect of pulse power stimulation on functional behaviour of cortical bone

tissue 145

8.3.1 Results Interpretation ............................................................................. 145

8.3.2 Final results ............................................................................................ 152

8.4 Discussion and Conclusion .................................................................... 152

8.5 Research limitations............................................................................... 155

8.6 Future work and recommendation ......................................................... 156

XVII

References ...................................................................................................... 158

XVIII

List of Figures

Figure ‎2.1 Hierarchical structure of bone (a) Cortical and cancellous bone (b) Osteon

consist of haversian canal (c) Lamellae (d) Collagen fibers (e) Collagen

molecules and mineral crystals23

.................................................................. 11

Figure ‎2.2 Cross-section of a bone showing both cortical and cancellous bone

structure26

..................................................................................................... 12

Figure ‎2.3 Response pattern of the bone cells to extrinsic/intrinsic applied load27

... 15

Figure ‎2.4 Multi scale of bone architecture (a) Amino acid building block (the

smallest scale of bone) (b) Tropocollgen molecules made from three

polypeptide chains of over 1000 amino acid residues (c) Mineralized

collagen fibrils consisting of mineral crystallites embedded within and

between collagen fibrils (d) Fibrillar arrays, the arrangement of the

mineralized collagen fibrils (e) Different organizations of fibrillar arrays in

different bone types (f) The osteon which surrounds and protects the blood

vessels (g) Bone tissue level (h) Whole bone level21

.................................. 16

Figure ‎2.5 (a) Triple-helical structure of collagen molecule (tropocollagen molecule)

(b) The arrangement of the collagen molecules in the collagen fibrils , (the

staggered arrays of tropocollagen molecules assembles in collagen fibrils

which themselves organize into arrays. The neighboring collagen molecules

have the gap (G) of 40 nm and the overlap (O) of 27 nm relative to each

other29

.) ......................................................................................................... 18

XIX

Figure ‎2.6 Mineralization of the collagen fibrils during bone synthesis 33

.............. 19

Figure ‎2.7 Strain generated potential created on a femur under mechanical

deformation10

................................................................................................ 30

Figure ‎2.8 Four stimulatory techniques for application of electric current to the

tissue by direct contact of the electrodes (A) The cathode in the target site

and the anode on the skin (B) The cathode in target site and the anode in

some distance with the cathode implanted in soft tissue (C) Non-invasive

stimulation placing the electrodes on the skin (D) Both electrodes implanted

in the soft tissue, away from the target site 87

............................................. 32

Figure ‎2.9 Inductive coupling set up over a tibia fracture94

...................................... 34

Figure ‎2.10 Capacitive coupling set up over the fracture site94

................................. 36

Figure ‎3.1 Conversion of low power, long time input waveform to high power, short

time output waveform by a pulse power generator ...................................... 45

Figure ‎3.2 Typical diagram for pulse power generators ............................................ 46

Figure ‎3.3 A combination of current and voltage sources as a pulse power

generator135

................................................................................................... 47

Figure ‎3.4 Circuit diagram of positive buck-boost converter .................................... 48

Figure ‎3.5 First switching state, charging the inductor ............................................. 49

Figure ‎3.6 Second switching state, circulating the inductor current.......................... 50

Figure ‎3.7 Third switching state, charging the capacitor........................................... 51

Figure ‎3.8 Power delivery through the load switch ................................................... 51

Figure ‎3.9 Pulse power generator A (PGA) with NEC microcontroller.................... 53

XX

Figure ‎3.10 Pulse power generator B (PGB) with Digital Signal Controller (DSP) . 54

Figure ‎3.11 High-voltage high frequency pulses with 500V at 10 kHz generated by

GPB .............................................................................................................. 55

Figure ‎3.12 Topology of pulse generator B with load modeling ............................... 56

Figure ‎4.1 Two types of bending tests and the compression-tension relationship of

forces along the surfaces of the loaded specimens[3] .................................. 60

Figure ‎4.2 Three-cycle bending load in linear elastic region on the bone strip sample64

Figure ‎4.3 A small drop on the first cycle of bending test in elastic region that was

removed on the further cycles ...................................................................... 65

Figure ‎4.4 Three point bending test142

....................................................................... 66

Figure ‎4.5 Assumed elliptical cross-section for whole bone ..................................... 67

Figure ‎4.6 Cross-sectional area of whole long bone in ANSYS for determination the

area moment of inertia .................................................................................. 67

Figure ‎4.7 Bone strip obtained from the cortical diaphysis ....................................... 68

Figure ‎4.8 Variation of Young’s modulus of the ovine metatarsus exposed to 180V

and 100 Hz pulses over 5 days (PP-exposed sample) compared to that of the

control sample .............................................................................................. 71

Figure ‎4.9 Sketch of experimental set-up for pulse power stimulation of the cortical

bone strip sample .......................................................................................... 72

Figure ‎4.10 Variation of the Young's modulus of femoral cortical strips exposed to

180V at 100 Hz pulse power over 9 days compared with that of the same

samples without pulse power excitation ....................................................... 73

XXI

Figure ‎4.11 The pulse power waveform with 450V magnitude and 10 kHz frequency

applied on cortical bone samples.................................................................. 76

Figure ‎4.12 Elastic properties of the cortical bone samples exposed to pulse power

(450 V at 10 kHz) before and after excitation compared with those values of

the control samples ....................................................................................... 76

Figure ‎5.1Typical macroscopic tensile test fracture (A) ductile shear fracture (B)

moderately ductile fracture (C) brittle fracture 155

....................................... 84

Figure ‎5.2Dumbbell shape specimen with round junction (GL, GW and GT are gage

length, gage width and gage thickness respectively) ................................... 85

Figure ‎5.3 Sketch of partitioned tibia used for tensile test specimen preparation ..... 87

Figure ‎5.4 Dumbbell shape specimen with sharp junction (GL, GW and GT are gage

length, gage width and gage thickness respectively) ................................... 87

Figure ‎5.5 Top view of a sketch of experimental set up for Pulse Power excitation of

the bone tensile test specimens between two isolated aluminium strips ...... 89

Figure ‎5.6 Tensile testing of the cortical bone specimen .......................................... 90

Figure ‎5.7 Cortical bone samples mounted on the SEM stubs, place for gold coating91

Figure ‎5.8 Tensile Stress-Strain responses until failure of dumbbell shape samples

with round junction ( ) versus those of dumbbell shape samples with

sharp junction ( ).................................................................................. 93

Figure ‎5.9 Comparison of the strength and toughness of dumbbell shaped samples

with round junction and those of samples with sharp junction .................... 93

XXII

Figure ‎5.10 Hysteresis loops in tensile loading-unloading cycle for a bone specimen

exposed to pulse power before and after 145 hours excitation .................... 94

Figure ‎5.11 Hysteresis loops in the tensile loading-unloading cycle for a control

bone sample before and after 145 hours being in similar environmental

condition as PP-exposed samples ................................................................. 95

Figure ‎5.12 Mean hysteresis energy of the control samples versus the samples

exposed to pulse power before and after 145 hours excitation .................... 96

Figure ‎5.13 Tensile stress-strain graphs of the cortical bone samples in four groups

up to failure .................................................................................................. 97

Figure ‎5.14 SEM micrographs from the top and side views of the control samples

(unexposed to pulse power) with their corresponding stress-strain graphs 100

Figure ‎5.15 SEM micrographs from top and side views of cortical bone samples

exposed to 500Vand 10 kHz pulse power for 145 hours with their

corresponding stress-strain graphs ............................................................. 101

Figure ‎5.16 SEM micrographs from top and side views of the cortical bone samples

exposed to pulse power, A and B for28 hours, C and D for 35 hours with

their equivalent stress-strain graphs ........................................................... 102

Figure ‎5.17 Details of scanning electron micrographs of fracture surface in higher

magnification (A) Dimpled, irregular appearance of fracture surface (B)

Microcrack diffusion (C) Microvoids (D) Crack bridging by collagen fibrils104

Figure ‎5.18 Higher magnification of scanning micrographs of the fracture surfaces

of the representative samples from each group (A) Control sample (B)

XXIII

Samples exposed to pulse power for 28 hours (C) Sample exposed to pulse

power for 35 hours (D) Sample exposed to pulse power for 145 hours ..... 105

Figure ‎6.1 Position and directions of the rectangular specimen obtained from the

tibial cortical dyaphysis .............................................................................. 113

Figure ‎6.2 Sketch of experimental set-up for pulse power stimulation of millimetre-

sized cortical bone samples ........................................................................ 115

Figure ‎6.3 Compressive testing of cortical bone specimen ..................................... 116

Figure ‎6.4 Compressive stress-strain responses for the control specimens ( )

verse those for the samples exposed to pulse power ( ) .............. 117

Figure ‎6.5 The total strain fracture energy of the samples exposed to 500V, 10 KHz

electromagnetic field compared to that of the control samples .................. 118

Figure ‎6.6 The strength of the samples exposed to 500V, 10 KHz electromagnetic

field compared to that of the control samples ............................................ 118

Figure ‎6.7 Comparison of the stiffness of the samples exposed to pulse power with

that of the control samples.......................................................................... 119

Figure ‎7.1 Ultrasound wave propagation in a bone specimen142

............................. 125

Figure ‎7.2 Ultrasound velocity measurement set up inside water tank ................... 130

Figure ‎7.3 Ultrasound wave propagation trough the sample and time delay

measurement on Lab view Signal Express ................................................. 131

Figure ‎8.1 The elastic modulus of the normal specimens compared with the samples

exposed to pulse power for 144 hours obtained from ultrasonic technique146

XXIV

Figure ‎8.2 Comparison of the flexural elastic modulus of the control and the PP-

exposed samples before and after pulse power stimulation ....................... 147

Figure ‎8.3 Comparison of the bone mineral density of the control and PP-exposed

samples before and after pulse power excitation ........................................ 148

Figure ‎8.4 Comparison of the hysteresis energy dissipated by the control and the PP-

exposed samples before and after excitation .............................................. 149

Figure ‎8.5 Comparison of the tensile strength and total failure strain energy of the

samples exposed to pulse power for 145 hours with those of the control

samples ....................................................................................................... 150

Figure ‎8.6 The strength and total fracture energy absorption of the samples exposed

to pulse power for 66 hours compared with those parameters of the control

samples ....................................................................................................... 151

Figure ‎8.7 Comparison of the Young’s modulus of the samples exposed to pulse

power with that of the control samples obtained from compression tests.. 152

XXV

List of Tables

Table ‎4.1 Comparison of the area moment of inertia of the whole bone samples

obtained from ANSYS and calculation ........................................................ 68

Table ‎4.2 Mean value± standard deviation for Young’s modulus of cortical bone

before and after pulse power excitation (450V at 340Hz) in three days ...... 75

Table ‎5.1 Mean value ±standard deviation for the toughness and strength of the

tensile bone samples in four treated groups ................................................. 98

Table ‎7.1 Comparison between the conventional mechanical tastings and the

ultrasonic technique161, 162

.......................................................................... 124

Table ‎7.2 Mean values ± standard deviation for the specimens’ dimensions .......... 127

Table ‎7.3 Mean density ± standard deviation for cortical bone specimens before and

after pulse power excitation ....................................................................... 129

Table ‎7.4 Mean value± standard deviation for ultrasound velocity and Young’s

modulus of PP-exposed samples before and after pulse power excitation in

longitudinal, radial and tangential directions respectively ......................... 134

Table ‎7.5 Mean value± standard deviation for ultrasound velocity and Young’s

modulus of control samples before and after pulse power excitation period

in longitudinal, radial and tangential directions respectively ..................... 134

XXVI

Table ‎7.6 Mean value ± standard deviation of ultrasound velocity and Young's

modulus in PP-exposed groups after pulse power excitation compared with

those of the control group in the same time ............................................... 135

XXVII

List of Abbreviations and Symbols

AC Alternating Current

ANOVA Analysis of Variance

ASTM American Society for Testing and Materials

CCPEF Capacitive Coupling Pulsed Electromagnetic Field

BMD Bone Mineral Density

DC Direct Current

E Elastic Modulus or Young’s Modulus

ECM Extra Cellular Matrix

F Applied Force

I Area moment of Inertia

K Bulk Modulus

G Shear Modulus

PBB Positive Buck-Boost Converter

PEMF Pulsed Electromagnetic Field

PGA Pulse Generator A

PGB Pulse Generator B

PP Pulsed Power

P value Probability, with a value ranging from zero to one

XXVIII

SGP Stress Generated Potential

u Modulus of Toughness

v Velocity

σ Nominal Stress

ε Nominal strain

ν Poisson’s ratio

ρ Density

Chapter 1: Introduction


2

Electrical phenomena play an important and effective role in biophysics, biology and

medicine. There is a strong evidence that human and animal bodies can generate

endogenous electric signals with large and stable gradient1. Also, the research indicates

that all organisms from bacteria to mammals respond to electromagnetic fields in different

ways for example cell division, tissue growth, wound repair 2. Observation of the effect of

this endogenous electrical current on the tissue growth and repair has induced interest in

the study and application of exogenous electrical stimulation in the field of orthopaedics.

For instance, over the last four decades the application of time-varying, weak magnetic

field, known as Pulsed Electromagnetic Field (PEMF), has opened a new, exciting gateway

to the connective tissue research and treatment for musculoskeletal disorders 3, 4

. However,

the first investigation in this field dates back to 160 years ago 5, 6

.

Bone is a dynamic tissue comprised of primarily cells including osteocyte, osteoblast

and osteoclast ensconced in an extensive matrix called extracellular matrix (ECM). ECM is

a composite consists of both organic (mostly type I collagen fibrils) and inorganic material

(mineral part, mostly hydroxyapatite). This particular composition results in a living,

complicated hierarchical structure, which has different physical, solid-state and electro-

mechanical properties7. These properties give the bone the capacity to respond to physical

stimulation by generating a very small electric current relating to bone formation 4. A

direct relationship between the mechanical deformation of bone and the generation of

endogenous electrical currents (caused by Stress Generated Potential) in bone has been

well indicated in different studies 7, 8

.

According to Wolf’s law, physical loadings on bone alter the bone structure and

leads to adapting the bone to its mechanical environment. Although the mechanism under

which bone responds to the applied physical loading is not fully understood, it has been


3

suggested that the stress generated potential in bone is related to the piezoelectric effect on

collagen fibers with non-centrosymmetric structure in dry bone and deformation of fluid-

flow in the small channels between osteocytes (canaliculi) in wet bone 2, 7

. This

endogenous electric signals and the mechanical strain in the cellular level generating

through loading the bone are proposed as the possible stimuli that cause the cellular

response leading to bone growth and osteogenesis.

There are various in vivo studies which have reported the advantages of both direct

current and PEMF stimulation on tissue growth 3, 4, 9

. This effect led to the utilisation of

electromagnetic fields stimulation for bone generation and as an accepted remedy for

some bone disorders such as delayed-union bone fracture and failed joint 3, 10

. In addition, a

few in vitro experiments stated the beneficial effects of electrical stimulation on

osteogenesis with both constant and pulsed electromagnetic field application 11, 12

.

However, a few studies reported the contradictory outcomes in the application of PEMF on

cell proliferation and differentiation13

. These diverse results are likely attributed to

different PEMF parameters and experimental conditions.

Three main parameters are involved in the application of any kind of electrical

stimulation that influence the results: the magnitude of the applied energy, the amplitude of

the stimulus and the frequency of application. The tissues appear to respond differently to

these factors. To achieve the desired outcome, the choice of appropriate parameters in a

suitable manner is essential.

A review of the advantages of PEMF stimulation on connective tissue in both animal

and clinical studies and on the other hand, the observation of the lack of studies in the

application of high power, high frequency electromagnetic fields, spurred interest in the


4

investigation of the possibility of applying pulsed power signals, a subset of PEMF, for

bone stimulation.

Pulse Power (PP) systems convert low power, long-time input to high power, short

time output. These systems typically store energy within an electrostatic field (i.e.

capacitors) or a magnetic field (i.e. inductors) over a comparatively long time and releases

it very quickly (in microseconds or less) which results in the delivery of larger amount of

instantaneous power (several kilowatts) in a very short time, though the total energy is the

same 14

. To generate such electromagnetic fields, high voltage and high current sources are

required. For prevention of the thermal effect the pulse interval, needs to be very short 15

.

Pulse Power technology has been used variously in biology and medicine, especially

at intercellular scale. Some of its established/demonstrated applications are controlling the

ion transport processes across membranes, prevention of biofauling, bacterial

decontamination of water and liquid food, delivery of chemotherapeutic drugs into tumour

cells, gene therapy, transdermal drug delivery, programmed cell death which can be used

for cancer treatment and intracellular electro manipulation for gene transfer into cell

nuclei16

. However, no published work has reported its utilization in skeletal system for

stimulation purposes.

The previous studies in the field of the electrical stimulation of the connective tissue

have mostly considered the use of low energy, weak electromagnetic fields or high

intensity electromagnetic field at low frequency. Nevertheless, the physical characterestics

of bone at high-energy levels have not received adequate attention. Additionally, there is a

limited number of researches investigating the frequency dependence of the electrical

properties of bone. The motivation for this research project was to explore the safe and


5

controlled application of pulse power on bone tissue as a proof-of-concept for potential in

future clinical application.

Structurally, bone is a complex tissue with unique mechanical properties which are

well adapted to bone functions. These properties, which are dependent on both bone

material quality and its micro and macro structure, are essential for bone to perform its

vital duties in the body. Hence, the investigation of the effect of pulse power stimulation

on biomechanical properties of bone is crucial as a primary step for safe and controlled

application of high voltage, high frequency pulsed electromagnetic field on

musculoskeletal system.

On the other hand, although there are many reported research regarding to the

application of electrical stimulation of the bone cells proliferation, differentiation and some

bone diseases treatment, very little studies investigated the effect of pulsed electromagnetic

fields on biomechanical properties of bone tissue 17-19

. For that reason, this research aims to

investigate the possible effect of high-power pulses at high frequency relative to changes in

the biomechanical/functional properties of the cortical bone samples. Along this way,

before animal or clinical study, assurance of the safe application of high power signals on

bone tissue is necessary to prevent any thermal effect or extra loading which can disturb

the quality of the bone composite material. Therefore, this pilot study is established to

investigate the controlled application of pulse power signals on bone tissue.

Chapter 2 reviews the basic structure and the biomechanical properties and the

electrical phenomenon of bone tissue and presents some of the previous in vitro and in vivo

researches in the electrical stimulation of bone tissue.

The pulse power generator was designed and fabricated based on the topology of the

positive buck-boost converter (a subset of DC-DC converters). The output voltage was


6

adjustable in magnitude, frequency and duty cycle for determination of pulse power

parameters. The timing of pulse power stimulant in the experimental protocol was also

considered in order to evaluate the possible effect of this factor on bone response. This

study establishes a new attempt in the field of pulse power technology to determine the

safe application method and controlled limits of parameters (pulse width, magnitude, and

frequency) for bone excitation. The details of the principles of pulse power generators and

the positive buck-boost converter topologies, which was utilized in this research, are

presented in chapter 3.

To evaluate the behaviour of bone in response to pulse power excitation,

determination of the functional properties of bone is required. The primary function of

bone is to be stiff to bear the loads applied to it through both internal and external forces.

In addition, it should be strong enough to resist breakage and remain stiff. The effect of

pulse power stimulation on bone stiffness, strength, the total strain failure energy

absorbance (bone toughness) and the hysteresis energy and the ductility of cortical bone

tissue were therefore evaluated. These properties were determined conducting three

conventional mechanical testing including three-point bending, tensile and compressive

tests. A non-invasive ultrasonic technique was also applied for evaluation of the cortical

bone stiffness. To investigate the effect of pulse power exposure on the microstructure of

cortical bone tissue (its porosity and the diffusion of microcracks), the fracture surfaces of

the bone specimens were evaluated using scanning electron microscopy (SEM). The results

and analysis of these four methods were presented in chapters 4 to 7.

The flowchart diagram presented in the next page, demonstrates the general research

procedure that was traversed regarding to research hypothesis approach:


7

Chapter 2: Physical Behaviour and Electrical

Stimulation of Bone

Chapter 2: Physical Behaviour and Electrical Stimulation of Bone

9

2.1 Introduction

Bone is a rigid connective tissue with unique mechanical properties that forms the

basis of the skeleton of vertebrates. It acts as both a mechanically skeletal structure and a

physiological unit with close relation. With a complex structure, bones form the

lightweight but hard and protective load-bearing framework for the body. Bone has to bear

up a combination of different loading including compressive, tensile, bending and torsion

during everyday activity, which strongly influences its structure and function. For example,

the high continuous loading on the sport people’s bones have increased bone mass around

the muscle attachment points while significant reduction in bone mass was observed after

long period of bed rest or for astronauts after prolonged space flights which is caused by

decreased loading of bone20

.

From the biological aspect, bone is a connective tissue, which exists in different

shape and size and provides a variety of mechanical, synthetic and metabolic functions in

the body. Beyond giving support and shape to the body, bones work in concert with the

muscular system to assist the body with movement and enables sound transduction in the

ear, serve as storage of minerals (calcium, phosphorous, etc.) and provide blood production

and stem cells from bone marrow for healing and cell growth7, 21

.

From the structural aspect, bone is a dynamic, hierarchical structure, which has a

unique capacity for self-repair, and adaptation to respond external mechanical loading with

continuous remodeling. Hence, an understanding of the micro and macro structure of bone

from molecular level and the mechanical properties of its constituents and their

relationship at different levels of the hierarchical structure is useful for realizing the

biomechanical behaviour of bone tissue in response to pulse power stimulation.


10

This chapter firstly provides a brief description about structure and biomechanics of

bone tissue (in particular cortical bone) and then reviews some past in vitro and in vivo

researches on the use of electrical stimulation on connective tissue.

2.2 Hierarchical architecture of bone

As mentioned earlier, bone has a complex and hierarchical structure which shows

various physical, solid-state and electro-mechanical properties22

. This special architecture

makes the bone a highly anisotropic and inhomogeneous material differing in component

distribution and spatial arrangement, which results in different mechanical properties in

each direction. The hierarchical organization of bone can be arranged in five levels with

particular mechanical properties coherent to each level which are interrelating together 23,

24.

1) Macrostructure or tissue level, containing i) trabecular bone (also known as

cancellous or spongy bone) which has unorganized lamellae arrangement with very high

porosity in spongy nature and accounts for approximately 20% of the bone mass and fill

the interior layer and two ends of long bones ii) cortical or compact bone which is a solid,

dense material with less porous that makes up the hard outer layer of long bones and

accounts for 80% of the bone mass25

2) Microstructure level (from 10 to 500 µm) including single osteons or trabeculae

3) Sub-microstructure (1-10 µm) lamellar level

4) Ultrastructure or nanostructure level (from a few hundred nanometres to 1 µm)

consisting of collagen fibril and mineral components of bone


11

5) sub-nanostructure, molecular level (less than one hundred nanometre) including

collagen and noncollagen protein molecules and mineral crystals 23, 24

Figure 2.1 shows a schematic diagram from this structural concept.

Figure ‎2.1 Hierarchical structure of bone (a) Cortical and cancellous bone (b) Osteon consist of haversian

canal (c) Lamellae (d) Collagen fibers (e) Collagen molecules and mineral crystals23

The cortical bone which is the focus of this thesis, is comprised of dense osteons.

The dense osteons themselves constitute of concentric lamellae in a layered structure with

porosities namely lacunae that are regularly diffused between layers and contain osteocytes

(a type of bone cells). The lacunas are connected with several canals containing osteocyte

fingers called canaliculi. They carry nutrients to and waste from osteocytes to the blood

vessels embedded in the haversian canals. As illustrated in Figure 2.2 the outer surface of

the bone cortex is the periosteum where the bone cells are laid down and act as the growth

source in the bone width. The next layers are cortical and trabecular bones adjacent to


12

marrow cavity. The inner surface is endosteal which is covered by cells that remove the

bone tissue20

.

Figure ‎2.2 Cross-section of a bone showing both cortical and cancellous bone structure26

2.3 Cortical bone structure

Compact bone primarily consists of 2% cells (by volume) ensconced in an extra

cellular matrix (ECM) of organic (collagen fibers) and inorganic (hydroxyapatite)

components 20

. Although cells and the matrix are working separately, their functions

interrelate to each other providing the bone growth and its dynamic behaviour and

adaptation to different internal and external stimuli.

2.3.1 Bone cells

There are three special types of cells that are found only in the bone: Osteoblasts,

Osteoclasts and Osteocytes, which work continuously to maintain bone tissue through

modeling and remodeling process. These bone cells which are responsible for bone

http://en.wikipedia.org/wiki/File:Illu_compact_spongy_bone.jpg


13

generation and growth, can change their activities and thus the name of a cell pertains to its

function at the time20

.

Osteoblasts secret and deposit bone extracellular matrix and are responsible for bone

formation and manufacture of hormones. Osteoclasts remove bone tissue by eliminating its

mineralized matrix and breaking up the organic bone (bone resorption) and Osteocytes

which are mature bone cells, originate from Osteoblasts 7.

Osteoblasts create bone through a two-step process: first, they secret an initial

collagen meshwork known as osteoid, which produce the basic framework of the bone

tissue. Then, they mineralize the collagen matrix by embedding needle, rod or plate-form

mineral crystallites within and between the collagen fibers27

.

The bone cells play the key role in sensing the intrinsic and extrinsic external

stimulation and responding appropriately to them with various biological signals, which

lead to bone growth and its continuous adaptation to its environment20

.

There are two processes in bone known as modeling and remodeling. Modeling is

either formation or resorption of bone at bone periosteal and endosteal surfaces which

cause bone mass augmentation and therefore its strength enhancement. By contrast,

remodeling is a coupled process of resorption followed by replacement of bone with little

change in shape on haversian and trabecular surfaces which repeatedly occurs during life

and can reduce bone mass and strength. This process which is controlled by extracellular

stimulation (e.g. applied loading) as well as hormones, calcium, vitamin D and genes,

regulate the balance of essential minerals in serum, repair micro-damaged bones (created

in bone by everyday stresses). It also provides a mechanism for bone adaptation to its

mechanical environment and hence shape and sculpture the skeleton during growth28

.

Osteoblasts and Osteoclasts are coupled to do this process and the balance change between

http://en.wikipedia.org/wiki/Bone_tissue

http://en.wikipedia.org/wiki/Calcium_homeostasis

http://en.wikipedia.org/wiki/Microdamage_in_bone


14

them results in some bone mass loss which is the main cause of some diseases like

osteoporosis (the most common bone disorder especially in women after menopause)7.

According to Wolff’s law, bone’s shape and orientation is changed by interaction

with its mechanical environment. On the other word, as mentioned earlier bone is a self-

repair and adaptive tissue in response to the applied intrinsic or extrinsic mechanical

loading. This means that if the loading on a particular bone increases, the bone will

remodel itself over time to become stiffer to resist such a load. This load can occur by

muscle or with an external mechanical load. In fact, the balance between bone formation

and bone resorption is largely controlled by mechanical stresses. For example, when a

compression strain is adapted to a long bone, bone formation occurs in the condensed side

and bone resorption on the tension side 10, 22

. According to this theory, a gradient potential,

stress generated potential (SGP), is produced along the collagen fibers by applying the

load, which imposed a local stimuli for cells involved in osteogenesis. Figure 2.3

demonstrates a schematic diagram of the basic response pattern of bone cells to the

extrinsic and intrinsic stimulus.

http://en.wikipedia.org/wiki/Menopause


15

Figure ‎2.3 Response pattern of the bone cells to extrinsic/intrinsic applied load27

It was therefore proposed that the interference of external agents for example

applying mechanical stress or electro stimulation could artificially control the growth

process of bone which is the main motivation for application of electromagnetic fields for

bone therapeutic purposes 7.

Although bone cells play particularly significant role inside the bone tissue regarding

sense and response to external stimulus and more importantly in bone growth and

osteogenesis through modeling and remodeling process, research on their behaviour in

response to pulse power excitation is beyond the scope of this thesis and will remain for

future research.

2.3.2 Extracellular matrix (ECM) architecture

ECM, which is mainly the origin of the mechanical properties of bone, is a

nanocomposite consists of about 70% mineral components mainly of calcium

hydroxyapatite Ca5(PO3 CO3)3(OH) with small percentage of some impurities like citrate,


16

fluoride and magnesium, 22% organic materials mostly comprise of type-I collagen with a

small amount of noncollagenous acidic proteins such as various growth factors and 8%

water by weight21, 29, 30

. The particular high level of incorporation of the very small mineral

and organic molecules provides the bone with its unique mechanical properties. Figure 2.4

shows the multi-scale structural levels of extra cellular matrix in bone tissue.

Figure ‎2.4 Multi scale of bone architecture (a) Amino

acid building block (the smallest scale of bone) (b)

Tropocollgen molecules made from three polypeptide

chains of over 1000 amino acid residues (c) Mineralized

collagen fibrils consisting of mineral crystallites

embedded within and between collagen fibrils (d)

Fibrillar arrays, the arrangement of the mineralized

collagen fibrils (e) Different organizations of fibrillar

arrays in different bone types (f) The osteon which

surrounds and protects the blood vessels (g) Bone tissue

level (h) Whole bone level21


17

The collagen fibrils arrangement

More than 20 types of known collagen are present in the different connective tissue

including bone, cartilage, ligament, tendon and skin31

. Among them, type I collagen is the

most abundant fibril protein in the body which constitute the most important structural

protein in the bone as well. It is composed of tropocollagen molecules (TC) which

themselves are made from three polypeptide chains, arranged in triple helical geometry and

built from over 1000 amino acid molecules connecting by hydrogen bonding. The

tropocollagen molecules aggregate in a staggered repetitive pattern with 67 nm linear shift

between neighbouring molecules to form the collagen fibrils. The fibrils themselves

arrange in arrays to create a network of collagen fibers, acting as the basic structural

framework for bone synthesis (Figure 2.5) 21, 29

. In bone formation, firstly, collagen is laid

down and then mineralized by hydroxyapatite crystal.


18

Figure ‎2.5 (a) Triple-helical

structure of collagen molecule

(tropocollagen molecule) (b) The

arrangement of the collagen

molecules in the collagen fibrils ,

(the staggered arrays of

tropocollagen molecules assembles

in collagen fibrils which themselves

organize into arrays. The

neighboring collagen molecules have

the gap (G) of 40 nm and the overlap

(O) of 27 nm relative to each

other29

.)

The mineral crystals structure

As the bone tissue grows and matures, the tiny crystals of carbonated hydroxyapatite

(dahllite) assemble in the gap region as well as in the overlap region between the layers of

collagen molecules and mineralize the randomly oriented collagen fibrils. The bone

mineral crystals, which are the smallest biologic crystals, grow to 30-50 nm width, 60-100

nm length and 2-6 nm thick in the same direction as the collagen fibrils 21, 29, 32

. The main

role of these mineral components is to stiffen the collagen fibrils by increasing the

crosslinking density and decreasing the crosslink length. Figure 2.6 shows the

mineralization process during bone formation.


19

Figure ‎2.6 Mineralization of the collagen fibrils during bone synthesis 33

The mineralized collagen fibrils are the fundamental building block of the extra

cellular matrix which their quality and spatial arrangement determine the functional

properties of bone tissue in nanoscale. For example, the stiffness of bone caused by the

mineral phase provides resistance to compressive stresses, while the collagen provides

bone with toughness and resistance to tensile stresses 34

.

The arrangement of these fibrils varies in different bone type which results in

differences in their functional properties. Figure 2.4(e) illustrates the randomly oriented

fibrils, in parallel, titled or woven bundle-patterns in bone.

The water content

As mentioned earlier, in extra cellular matrix, in addition to osteoid (pure collagen

fibrils) and mineral contents, water, contributes a volume fraction of bone tissue (about 8-

12%) which exists as embedded molecules in collagen matrix and mineral crystals as well

as mobile water in haversian canals, lacunae and canalacunie35, 36

. The water content in

unmineralized collagen matrix is higher (up to 60% of volume fraction) than calcified


20

collagen (reduced down to 20% volume fraction) in which the osteoid water is dislocated

by apatite crystals. This fraction is reduced to 10 % in old bones which result in the

reduction in the bone impact strength20, 37

.

Water molecules have interaction with the collagen and mineral molecules in several

ways. Because of the water molecule polarity, it has bonding with hydrophilic groups (e.g.

glycine, hydroxyproline ) in collagen molecules and the charged portion (e.g. Ca+ and PO4

-) in mineral crystals

38. The water bonding with collagen molecules and mineral phase are

at two levels: structural water that has hydrogen bonding inside the triple helix of collagen

molecules and mineral lattice structure and needs more energy to remove (between 200˚C

to 400˚C temperature) and loosely bound water at the surfaces of the molecules which

required less removal energy (below 200˚C heating)39

.

2.4 Contribution of the bone constitutes at different hierarchical levels on its

mechanical competence

As mentioned earlier, bone has an anisotropic and inhomogeneous structure and

therefore, its mechanical response is different in each direction. It is composed of an

organo-mineral nanocomposite material whit unique mechanical properties, which are well

adapted to bone functions. Several factors involve in creation of the specific mechanical

attributes of bone tissue such as particular structural properties of the mineral and organic

components and their special hierarchical arrangement at different levels. In fact, the

functional properties of bone tissue at macroscale reveal the intrinsic material properties of

its components and their spatial arrangement and interaction in nano and micro scales30, 32

.

Stiffness, strength and toughness are three main functional properties of bone to

perform its vital duties in the body. They are important to help bone to resist any


21

deformations in response to combination of internal and external forces in different

directions and prevent its breakage. These mechanical properties are associated to both

quantity and quality of bone tissue. In addition to bone mineral density (BMD) (showing

the quantity of bone tissue), which definitely affects bone strength, there are some other

important factors showing the quality of bone tissue contribute in bone strength and its

susceptibility to fracture such as: bone micro and macro architecture, inherent bone

material properties including porosity and crystallinity and the potential existence

microcracks and their repairs in bone tissue32

.

Nevertheless, these properties are changed over the lifetime because of age, diseases,

etc. For example, bone strength can be increased by adding bone mass or changing bone

geometry to distribute the applied loads (stress) or by variation in bone microstructure via

processes such as formation and deformation (modeling and remodeling) interaction with

its mechanical environment. It is well established that external stimulus like mechanical or

electrical stimulation can also affect bone functions27

.

The functional behaviour of bone at different hierarchical levels have been evaluated

through several conventional mechanical tests which simulate the mechanical loads applied

on bone throughout everyday activities in vitro. These mechanical tests are including three

or four-point bending, tensile and compressive tests that are employed in this study for

determination of stiffness, strength and the capacity energy absorption of bone tissue in

response to pulse power stimulation. There are also some non-destructive methods for

determination of bone mechanical properties such as ultrasonic technique. This method

was also applied for evaluation of variation in bone elasticity in three orthogonal directions

due to pulse power excitation. More details about the applied methods and the required

information for them are presented in the following corresponding chapters.


22

2.4.1 The bone basic elements (molecular level)

Although the specific combination of the organic (collagen matrix) and inorganic

(mineral) phases in a specific architecture generally provides the unique mechanical

properties for bone tissue, they play their own different roles for this purpose. The collagen

matrix provides the bone with its plasticity and the ability to absorb energy (e.g. bone

toughness) and sustain the tensile strain while the mineral phase provides the bone with its

stiffness and effective resistance to the compressive loading 21, 30, 32, 40

. It has been shown

that the mineral contents have more effect on Young’s modulus than ultimate strength of

cortical bone tissue41, 42

whereas the collagen matrix has poor relation with Young’s

modulus of bone tissue and direct strong effect on its toughness32

.

The collagen fibrils

The quality and the orientation of collagen fibrils can influence the structural quality

and the mechanical properties of cortical bone tissue30, 43

. Although, researchers has not

reported an integrated mechanism for the plastic deformation and the energy dissipation

process in bone, it was mostly attributed to breaking or reforming of hydrogen bonds

inside individual tropocollagen molecules and between hydroxyapetite and tropocollagen

molecules21

. In collagen fibrils, stretching, unwinding and intermolecular sliding of

individual collagen molecules that involve hydrogen–bond breaking cause their

deformation under progressive strain and gives the bone the ability to withstand the large

plastic strain and bone ductility21

.

The orientation of collagen fibers according to the direction of loading has

significant effect on the bone strength along with the interaction between the mineral and

collagen molecules 32

. This effect causes different susceptibility to load-bearing capacity of

long bones in different directions30

. For example, the femur can resist easily the


23

longitudinal compressive load without significant damage whereas it will break with

similar load in transverse direction. The woven bone with unorganized collagen fibrils

showed lower mechanical properties compared to the lamellar bone with organized

collagen fibrils32

. In addition, reorganization of collagen fibrils through exercise resulted in

the maintenance of the mechanical properties of bone tissue including its strength,

although bone mineral density was decreased44

. These findings can illustrate the significant

effect of the collagen fibers orientation in functional and biomechanical characteristics of

bone tissue.

The orientation and arrangement of collagen fibrils can be affected by several factors

such as an electromagnetic field exposure. This effect was used as a most common method

for collagen fibrils alignment in synthesis of scaffolds that mimic the aligned collagen

fibrils in very regular tissue like tendon and ligament or as an aligned sheets in bone and

corneal tissue45, 46

.

On the other hand, the alteration in collagen fibrils (both quality and orientation) can

affect bone mechanical attributes like its toughness and overall strength32, 47

. For example,

the deterioration of the bone collagen matrix by ionizing radiation (producing crosslinking)

was reported to reduce the strength and energy absorption capacity of the bone samples

with no effect on their Young’s modulus48, 49

. Similarly, it was demonstrated that the

denaturation of the collagen network by heating without changing in mineral content

significantly decrease the toughness and strength of the bone tissue whereas its elastic

modulus remains almost constant50

. These results reinforce previous studies suggesting

that the collagen network play key role in bone toughness and overall strength while

having minimal effect on the bone elasticity. This can show the positive involvement of the

collagen fibrils on increasing the amount of energy required for bone fracture20

. This


24

mechanical property (the total strain failure energy) is defined as the area under the stress-

strain curve until failure and is known as the modulus of toughness.

The concentration, pattern and specific structure of collagen crosslinks were also

reported to play key role in the bone strength and material deformation so that the

distribution of the collagen crosslinks in osteoporosis patient bone was significantly

different compared to healthy bone21, 51

. In addition, the reduction of crosslinks

concentration may decrease the bone strength52

.

The mineral crystals

In addition to collagen matrix, the size and distribution of mineral crystals influence

the mechanical properties of bone53

. As stated earlier, the presence of the mineral crystals

provides the bone its rigidity and stiffness. The small dimension of plate-like

hydroxyapetite crystallites gives them the strength of a perfect crystal. The anisotropic

behaviour of these crystals results in their different deformation under different load

directions which can involve in the total anisotropic property of the bone tissue29

.

The size and the orientation of the mineral crystals are associated to the structure and

organization of collagen fibrils and other noncollagenous proteins as well as diseases,

drugs and aging54

. For example, in the aged bones, the average size of the crystals and the

crosslink density between the collagen molecules increase compared with those in the

young bone structure, resulting in the general deterioration of strength, stiffness and

toughness of the old bone tissue21, 30

.

The bone water content

The water content plays key role in the mechanical properties of bone tissue and

appears both as mobile water in pores and with interact by bone constitutes at different

energy levels. At first level, the mobile water needs less energy to remove from tissue,


25

after that the water molecules trapped in collagen molecules with loss hydrogen bounding

requires little removal energy and finally the water molecules imbedded in hydroxyapetite

lattice needs the highest energy20

.

Bone dehydration will raise by increasing of drying temperature and affect the bone

strength and toughness. Drying even at room temperature may cause water loss from some

phases of extracellular matrix which lead to augmentation of bone strength. A nonlinear

relationship was observed between bone strength and its dehydration55

. Furthermore, water

loss causes reduction of collagen molecular diameter 56

and increasing the collagen

stiffness38

.

As the collagen fibrils play the predominate role in bone toughness, bone

dehydration (which affect collagen properties) also decreases bone toughness and work to

fracture (because of decrease in both strength and strain at failure point)20

. In addition, the

observation of less water in more mineralised bone has been suggested that the reduction in

bone energy absorption capacity with increase in bone mineralization is also associated to

reduction in water-mineral interaction57

.

Although in normal hydrated situation, the collagen fibrils apparently have less effect

on bone strength compared to the bone mineral and its porosity, bone dehydration with

enough energy removing both collagen and mineral molecules interaction with water, will

decrease the bone strength. Hence, it appears that bone dehydration at low temperature will

increase bone strength (due to collagen stiffening) whereas the more bone water loss at

higher temperatures cause bone strength reduction because of the variation in the bone

mineral content20

. Additionally, the bone stiffness which has a linear relationship with

water loss, will decrease by bone dehydration58

.


26

The above mentioned effect of water loss on bone mechanical properties can suggest

that the variation in the bone strength and toughness at aging could be related to decrease

of water content in the bone constitutes (e.g. water bounded to collagen) and its increase in

the pores20

. Furthermore, it confirms the importance of prevention of the bone dehydration

when its mechanical properties is explored under an external stimuli.

2.4.2 The mineralized collagen fibrils (nanoscale level)

It has been well established that the mechanical properties of the mineralized

collagen fibrils are significantly different from those of the pure collagen fibrils. This

effect appears to be because of the direct interaction between the mineral and the

tropocollagen molecules29

.

As mentioned earlier in mineralized collagen fibrils the collagen molecules fastened

between parallel mineral plates which reduces their flexibility in the lateral direction and

stiffen the organic phase. It has been shown that the stiffness of the collagen fraction of the

mineralized matrix is much higher than that of nonmineralized tissue while the strain of the

collagen fibrils in the mineralized matrix is less than that of in the nonmineralized network

59. The mineralized collagen fibrils have also anisotropic behaviour which is related to the

different stiffness of the mineral crystals in different directions60

as well as the differences

of elastic modulus of the collagen fibrils in transverse compared with the longitudinal

planes61

and the relative involvement of the mineral and the organic phase to the

mechanical properties of the collagen fibrils in different orientations62

.

2.4.3 The arrays of the collagen fibrils (mesoscale level)

At this level, the interaction between the adjacent collagen fibrils and the

extrafibrillar components (including the mineral crystals and the noncollagenoous

molecules) between them play important role in determining mechanical response of the


27

tissue. Similar to previous level, here, the sliding of the fibrils along each other, under

tensile load cause the tissue strain. Nevertheless, this strain increases hierarchically from

nano- to mesoscale in bone tissue. In general, a combination of stiff mineralized collagen

fibrils with soft noncollagenous acidic protein between them, provides a specific network

with unique stiffness and toughness which can dissipate large amount of energy under

mechanical loading and reform when the load is removed63

. However, a small changes in

the mineral contents inside and outside of the collagen fibrils in such a network can

influence the total mechanical properties of the tissue. Similar to the previous levels, the

mineralized fibril arrays have a high mechanical and structural anisotropic behaviour with

the highest modulus value for tension. They show the highest resistance to compressive

loading in direction along the collagen fibrils 64

. This behaviour can therefore confirm that

the changes in fibrils orientation alter the mechanical properties of the tissue in specific

direction.

2.4.4 The organization of the fibril arrays in lamellae and osteon (microscale level)

In the microscopic level, the collagen fibrillar arrays arrange differently in different

tissues adapting to a unique function. For example, in human bone, the co-aligned

arrangement of the collagen fibrils creates a parallel lamellar bone with osteonal structure.

In lamellar bone, the circular arrangement of the neighbouring layers of co-aligned fibrils

(lamellae) creates a microlamminate composite, called rotated plywood structure. Each

lamellae itself composed of several rotated sublayers. This complex hierarchical structure

causes high resistant to the crack propagation in normal direction across the lamellae plane

and low resistance along the lamellae plane65

.

The osteonal bone is constituted of several osteons that are cylindrical concentric

layers (lamellae) which surround the blood and nerve vessels in haversian canals. The


28

collagen fibrils in each lamellae have specific directions which are different from those in

the adjacent lamellae. This special arrangement helps the bone to sustain the applied

loading in different directions30

.

The osteonal bone shows a specific pattern of microcracks propagation combined of

a few micron short radial cracks and larger circumferential microcracks under compression

loads. This unique type of microcrack diffusion results in a large amount of energy

dissipation and preventing of severe failure and gives the tissue the capacity to maintain its

high strength and resilience even in plastic deformation66

.

As mentioned earlier, bone has a high structural and mechanical anisotropy in all

hierarchical levels which is reduced from nano- to macro levels, so that it shows less

anisotropy at higher level. This important characteristic provides the bone with the ability

to withstand to various loading, applied in different directions and patterns 64

.

2.5 Bioelectric phenomena in bone

Dense connective tissues like bone, which are nanocomposites of collagen fibrils

reinforced by the mineral crystals (mainly hydroxyapatite) reveal some special

bioelectrogenic events such as piezoelectricity and electrokenetic potential. The structure

and biochemical composition of bone, which is altered by age, gender, anatomical location

and hydration, can affect these electrical properties of bone8. As mentioned earlier, these

electrical attributes, which are strongly associated to applied mechanical loading, can

mediate the biological processes like bone growth and its remodeling by coordinating of

bone cells, hormones and enzymes2, 20

. This phenomenon happens by generating an

electrical potential differences called stress generated potential (SGP) along the collagen

fibrils following the mechanical deformation of the tissue which provides a local stimulus

for bone-generating cells proliferation and extracellular secretion67

. This shows the close


29

interaction between bone cells and extracellular constituents. Observation of the

contribution of this endogenous electrical signals on bone tissue growth and healing, has

suggested the use of exogenous electrical stimulation to induce bone formation in clinical

studies since early 1840s, when a tibial non-union fracture was treated successfully using

direct current5, 6, 68

. However, the interest in this area has arisen since the 1950s when

Fukada and Yasuda demonstrated the piezoelectric property of collagen inside bone and

the positive effect of electricity on bone healing 69, 70

.

2.5.1 The origin of the stress generated potential (SGP) in bone

Two predominant mechanisms were introduced for creation of the stress generated

potential in bone: piezoelectricity and the streaming potential which are the properties of

bone material 20, 67

. However, some other researchers indicated that the induced electrical

potential in bone is also relying on migration of inorganic ions within the bone and the

corresponding cellular realignment and relocation71, 72

. It can suggest that an electric

charge produced in the living bone is different from that generated by the dead bone and

this latter mechanism act as a secondary origin of the electric generated potential in the live

tissue 73

.

The piezoelectric effect illustrates that an electric potential (SGP) creates in bone

while undergoing a mechanical deformation. This mechanism has been recognized as the

main mechanism for SGP in dry bone and is dominantly attributed to collagen fibrils which

has a lack centre of symmetry in their structure 20, 67

. It was demonstrated that removing the

hydroxyapetite crystals from the bone tissue matrix did not affect the electrical gradient

generated by the stress. It confirms that hydroxyapetite crystal is not the basis of induced

electrical potential in bone70

. Furthermore, it has been reported that the piezoelectric

coefficient decrease by increasing the water content (attached to collagen molecules or


30

mobile water in haversian canals) in the tissue74

. It can suggest that hydration of the bone

tissue can influence electrical properties of bone as well as the bone mechanics55

.

However, piezoelectricity has gained less importance as a possible mechanism for

SGP in wet bone 67

. According to several studies, the stress induced electrical polarization

in wet bone is generated by streaming potential caused by movement of the ions in fluid

flow upon deformation. The movement of the charged fluid through the haversian and

Volkmann channels creates an electric current which therefore causes a potential

differences between two points in the channel2, 20

.

The amplitude of this generated potential waveform is strongly associated to the

frequency and the magnitude of the loading while the polarity as presented in figure 2.7 is

determined by the direction of the bone deformation 75, 76

. Figure 2.7 demonstrates that the

electrons movement under the mechanical deformation of bone, causes a negative charge

and bone generating on the compression side and an equal positive charge accompanying

with bone resorption on the tension side which creates a potential differences that

disappears when the force is removed77, 78

.

The above discussion can suggest that the bone electrical activity and hence, its

growth and remodeling can be influenced and controlled by external stimulation (electrical

or mechanical).

Figure ‎2.7 Strain generated potential created on a femur under mechanical deformation10


31

2.5.2 Electrical stimulation of bone with low intensity electromagnetic field

Electrical stimulation has been applied widely in different animal and clinical

studies. To date, three major methods have been introduced for electrical stimulation of

bone tissue using: i) Direct contact electrode method (electric current stimulation) ii)

Inductively coupled fields (electromagnetic field stimulation) and iii) Capacitive coupling

(electric field stimulation) 16

. A brief review of the history of the in vitro and in vivo

applications of electromagnetic field as stimulator for bone growth and some bone diseases

treatment is presented at the following sections.

Application of direct contact method for bone tissue stimulation

As Figure 2.8 shows, four different stimulatory techniques for electric current

stimulation of bone has been applied via direct placement of the electrodes at the target site

which results in passing electric current through the tissue. In the two first methods which

were used mostly in different studies, up to four electrodes (cathode, the negative pole)

placed in the target site and the other electrode (anode) is placed either on the skin surface,

outside the body or in the target site with some distance close to the cathode. The

electrochemical interaction can be occurred in both of those methods, due to direct contact

of the electrodes with the tissue79, 80

. Nevertheless, in two other methods, the electric

current was introduced into the desired region where the osteogenesis aimed to happen

without direct placement of the electrodes in the site and thus the electrochemical effect

appears weakly or is omitted completely. As illustrated in Figure 2.8C, the electrodes are

located on the skin and hence, this method is non-invasive81

. In method D, the electrodes

implanted in the soft tissue with distance from the target site82

.

The direct electrical current generally acts through occurrence of an electrochemical

reaction at the cathode which leads to the collagen and proteoglycan synthesis. It is

believed that a decrease in oxygen concentration occurs at the cathode which enhance


32

osteoblastic activity and reduce osteoclastic activity83, 84

. This electrochemical reaction also

appears to increase ph and generate hydrogen peroxide which may excite macrophages to

release endothelial growth factor (An crucial factor for osteogenesis)85, 86

.

Figure ‎2.8 Four stimulatory techniques for application of electric current to the tissue by direct contact of the

electrodes (A) The cathode in the target site and the anode on the skin (B) The cathode in target site and the

anode in some distance with the cathode implanted in soft tissue (C) Non-invasive stimulation placing the

electrodes on the skin (D) Both electrodes implanted in the soft tissue, away from the target site 87

Another possible mechanism according to the application of direct electric current on

live tissue is the realignment of the osteogenic cells (in particular osteoblast) in the electric

field which affects significantly bone regeneration and remodeling on the cathode site.

According to this mechanism osteoblasts migrate toward cathode due to Ca+

influx to

anodal side of the cell ( termed galvanotaxis of the osteoblasts)71

.

Electric current stimulation through direct contact electrode system has been reported

to accelerate non-union bone healing with a success range of greater than 70% 16

and also

has had good effect on reducing the pain of considerable number of patients suffering from

osteonecrosis 88

. Nevertheless, the dose of delivered constant current is important for the

osteogenesis effect. The currents smaller than 5 µA and greater than 20 µA showed no

effect and cell necrosis respectively79

.


33

However, direct contact method is totally invasive (except method C) and needs to

implant the electrode wires at precise locations inside the intended tissue during an open

surgical procedure which may result in electrolysis and local thermal effect and therefore

tissue damage89

. For the case of large bone, several electrodes are required for bone

generation in a reasonable duration 79

. The weaknesses of this method have led to the use

of the non-invasive field coupling methods through indirect contact of electrodes

surrounding the target tissue including the capacitive and inductive coupling.

Application of the pulsed electromagnetic field stimulation on bone tissue

The pulsed electromagnetic field has several advantages as an effective stimulus for

bone growth in particular for the therapeutic purposes compared to direct current method

as follow:

The electrodes do not have contact and hence electrochemical

interaction with the target tissue. Therefore, the application of these

non-invasive, athermal methods appears to have no known risk or

discomfort.

It is very easy to use without any open surgical operation (can be done

in an office setting) and the equipment can be portable especially in

capacitive coupling method90

Treatment expenses are low compared to the cost of surgery

It seems that the existence of implanted metals does not influence

their remedial properties 91

.

According to above advantages, pulsed electromagnetic field has been applied

successfully as a reliable treatment for various diseases including orthopaedic and

rheumatologic disorders, spinal fusion, soft tissue regeneration, neurological disorders and


34

cancer during last half decade 4, 10, 92

. It was utilized through two capacitive and inductive

coupling methods.

Inductive coupling

This method has employed an air coil system which places with no direct contact

along the target tissue. A pulsed current passing through the wires of the coil generates

time changing magnetic field perpendicular to the flow of the current. This time varying

electromagnetic field induced an electric field which generates a small current in the target

tissue and simulates the normal response of bone cells to applied mechanical loading 93

acting as a local stimuli for the bone-generating cells and causes the bone growth and

remodeling 10, 94

. The commercial device using this method has been applied for fracture

healing since about 30 years ago. The therapeutic equipment generally includes a portable,

battery-powered pulse generator and a coil of wire which is placed externally, without any

contact, over the intended tissue 3. Figure 2.9 shows an inductive coupling set up for a tibia

fracture healing.

Figure ‎2.9 Inductive coupling set up over a tibia fracture94


35

Basset et al. reported the first satisfactory application of inductive-coupled PEMF

stimulation by two low frequency, low intensity fields one with 2mv/cm, 1.5 ms at 1 Hz

and the other with 20 mV/cm, 0.15 ms at 65 Hz on beagle dogs. They recognized that

using the higher frequency is more effective in producing new bone tissue compared to the

lower frequency 95, 96

. They also reported the first successful clinical therapeutic use of

PEMF in humans increasing the bone formation, using quasi rectangular asymmetric

waves with 300 pulse width and 75 Hz frequency for 12 hours daily over 3-4 months 97

.

The age of the patient, the site of fracture, type of non-union and presence of infection

affect the result of the therapy with PEMF stimulation 91

. In addition, the magnitude of

pulsed magnetic field and duration and length of treatment influence the effectiveness of

PEMF stimulation 98

. Based on this finding and also further reports of success rate up to

80% of PEMF stimulation on non-union fracture healing, it is anticipated that PEMF

stimulation of ordinary fractures could decrease the period of healing and cast wearing 10

.

The PEMF stimulation also has been suggested for other skeletal disease like

osteoporosis (the most common bone disorder that is defined by decreased bone mass,

microarchitectural deterioration of bone tissue and increase the susceptibility to fracture).

Tabrah et al. reported an initial increase in bone density after 12 weeks of 10 hours daily

PEMF exposure (72 HZ frequency,2.85 mT peak and 380 quasirectangular wave

followed by 6 ms quasitriangular wave) on post-menopausal women. However, bone

densities had steady decline 36 weeks after treatment. It is indicated that PEMF is helpful

for osteoporosis treatment immediately after exposure and the effect will be reduced

following its removal and hence can show the influence of the period of stimulation on its

effect 99

.


36

Capacitive coupling

Capacitor stimulator is composed of a power supply and two opposing capacitor

electrodes, which are attached directly to skin surface surrounding the intended bone

tissue. A pulsed or alternative voltage is therefore applied between the capacitor electrodes

to generate a pulsed electric field within the target tissue. Despite the prior methods, here

the precise localization of the electrodes which cover a large area of the target tissue is not

required 10

. Furthermore, this method is more competent in generating electric field

compared to inductive coupling technique where the time varying magnetic field create the

major effect and the electric field produces the minor effect in the tissue20

. Figure 2.10

demonstrates a capacitive coupling set up on a fracture bone.

Figure ‎2.10 Capacitive coupling set up over the fracture site94

Brighton and Pullack successfully used capacitive coupled electromagnetic field

method with 60 KHz, 5 VP-P Sine wave for treatment of chronic non-union fractures in

humans applying stainless-steel capacitor plates placed in direct contact with the skin


37

surface surrounding the non-union fracture. They noted that wearing a cast, the presence of

osteomyelits and placing implants in the bone did not influence the outcomes90

.

In 1991, Behari et al. designed and applied a pulsed radio frequency electric field

stimulator with the ability to generate high frequencies up to 14 MHz pulsating frequency

(modulated at extremely low frequencies of 16-76Hz) and 10 VP-P amplitude in square

wave form for the purpose of fracture healing. They adapted the output signal by

capacitive coupled with the aid of a couple of stainless steel electrodes at fracture site of rat

bone for 30 days, 2 hours/day. To investigate the results of stimulation, the bone mass

estimated by measurement of the cortical thickness and ultrasonic attenuation. The results

showed that the capacitive coupling with mentioned electrical parameters is a useful,

reliable method for accelerating bone fracture healing 16

.

The capacitive coupled electromagnetic field with saw tooth pulses of 100V at 16 Hz

was also effective in the bone cell proliferation and differentiation in vitro and improving

extracellular matrix formation and maturation100

A recent research investigated the effect of electrical stimulation on organic,

inorganic and macrostructural properties of varectomized rat bones using a capacitive

coupling pulsed electric field (CCPEF) generator similar to previous study (with 10 VP-P,

pulsed square wave at 16 Hz modulated frequency and 14 MHz career frequency). It was

concluded that the bone mineralization, collagen deposition and microstructural

compactness of bone are increased after PEMF exposure therapy to the osteoporotic bones

34. In a similar study, they also reported that CCPEF is beneficial in reducing the

ovarioctomy-induced bone mineral loss in rats which can be used as a prevention for

osteoporosis especially in post-menopausal women 101

.


38

2.5.3 Influential factors in electrical stimulation methods

In direct contact technique, both pulsed and alternative current (time varying) (AC)

and direct constant electric (DC) current have been applied with no substantial differences

between the results102, 103

. However, Black et al. indicated that direct current causes a little

more bone generation compared to a number of pulsed currents with similar amplitude104

.

It was also demonstrated that a low frequency electrical current could reduce bone loss by

decreasing the osteoclast differentiation and increase bone generation so that maintain the

bone and preserve its integrity 105

.

Similarly, for the field coupling methods, both static and time varying electric and

electromagnetic fields have been applied in different studies. The comparison of the

constant and pulsed electromagnetic fields on synthesis of either organic or inorganic

components of bone demonstrated that an alternating electromagnetic field increased the

production of both constituents of bone while the static field did not show significant effect

in both collagen and the mineral contents106

.

The effectiveness of the electrical stimulation depends on the electrical signal

characteristics such as the magnitude, duration, frequency and waveform shape 107, 108

.

However, this selective action, which is the advantage of pulsed electromagnetic field

application, is not available with a constant field. For instance, Brighton et al.

demonstrated that the electric field intensity and also pulse configuration and pulse width

at appropriate field strength are significant factors in the bone cells proliferation 109

. They

found that the electric field strength of 0.1-10 mV/cm enhance the cell proliferation while

the field less than 0.1 mV/cm did not affect the proliferation. In addition, the lowest

applied or induced current density which seems to be effective on bone formation was

reported in the range of 0.005 µA/mm2 while the highest was about 25 µA/mm2. However,

the densities above 1 µA/mm2 appeared to cause necrosis with the electrodes. The optimum


39

current magnitude pertains on the application technique, the model system and the nature

of the electrodes87

. For example, the platinum electrodes are more osteogenic at lower

current densities while the stainless steel electrodes appears to be more active at higher

current densities110

.

The asymmetric, quasi-rectangular or quasi-triangular and sinusoidal waveforms

were accepted as the appropriate waveforms for PEMF stimulation 92

. It has also been

observed that the intermittent use of PEMF stimulation has far greater outcome responses

to the continuous application 10

.

Although, different studies demonstrated the positive effect of low frequency pulsed

electromagnetic field on the bone generation and growth, a few studies investigated the

bone stimulation at high frequency. This is because that the actual physiological actions are

limited to extremely low frequencies20

and hence, most researchers tried to mimic the in

vivo situations. Nevertheless, there are some studies investigating the frequency

dependence of the electrical properties of animal and human bone especially in high

frequencies. Reddy and Saha reported that the impedance of bovine compact bone is

almost independent of the frequency up to 70 KHz and after that it reduces with rising

frequency 111

. According to Kosterich et al. conductivity of fresh and fixed rat bone is

independent of the frequency under 100 KHz and there after it increased with the

frequency 112

. Singh and Behari investigated the effect of various frequencies of PEMF

exposure in the range of 0.5-108 MHz, on impedance and phase angle of bone, using

human femur at room condition. They found that the resistivity, dielectric constant and

impedance of bone decreased with increasing frequency 113

. In addition, as mentioned

earlier, the effectiveness of the capacitive coupling electromagnetic fields at high


40

frequencies on mineralization, collagen deposition and bone compactness of osteoporotic

rat bones was demonstrated in recent studies 34, 101

.

In a recent PhD thesis, the combination of an electrical and mechanical stimulations

on bone cells proliferation and differentiation was investigated. It was found that the

synergistic effect of these to stimulator was useful in bone cells development while the

application of PEMF alone lead to depression of the cell proliferation. Also, the

mechanical strain alone did not show significant effect on bone cells proliferation and

differentiation114

.

2.5.4 Some of the hypothesized mechanisms involved in bone generation due to

pulsed electromagnetic field

As bone cells predominantly contribute in the bone generation and growth, it is clear

that the positive response to PEMF stimulation should be cell specific. Because it is not

clear how biophysical mechanisms detect and convert electromagnetic field to a biological

signal, the cellular mechanisms involved in the success of PEMF stimulation on

osteogenesis and bone growth are not fully understood. However, there are some

hypothesis on the cellular mechanisms by which PEMF stimulation has effect on bone

growth and remodeling. As stated earlier, the first proposition was that the time-changing

magnetic fields induce an electric field (Faraday’s law of induction) which generates a

small current in connective tissue 10, 34

. In addition, it is suggested that PEMF stimulation

has an effect on the calcification of fibro cartilage in the space between the bony segments

and since it raises the blood supply by affecting the calcium channels and improving bone

healing. It is also stated that PEMF may affect osteoblast and cause augmentation of the

rate of bone formation 115-117

. According to Berg and Zhang, PEMF increases the

transmembrane voltage and cause augmentation of electromagnetic conductivity of cell

membrane protein and hence, lipid and expression of genes are altered and result in the cell


41

proliferation 118

. Spadro and Bergstorm suggested that PEMF exposure depolarizes the cell

membrane of osteoblast to alter the uptake of calcium ions and increase the concentration

of intercellular free calcium in osteoblast cytoplasm 119

.

The two latter hypothesis for the mechanisms involved in electrical stimulation of the

bone cells are related to electrical activity of the cell, considering the cell as a conductive

object (cytoplasm) which surrounded by the surface membrane assumed as a dielectric

layer. Exposing the cell to an electromagnetic field (e.g. applying an unipolar voltage pulse

to two electrodes on two sides of the cell) causes electric concentration on cell membrane

which results in the augmentation of membrane permeability and cause a voltage across

the membrane. However, if the membrane voltage exceeds a critical value (in the range of

-95mV to -60mV for different cells), a structural changes would happen on the membrane

surface during a process called electroporation or cell depolarization which forms some

transmembrane pores in the cell membrane. In fact, the cellular effect of the applied pulsed

electromagnetic field is dependent on its frequency, field amplitude and pulse duration15,

120. For example, if the membrane voltage remains below the critical value (applying not

too high field) or the applied pulse duration is limited ( not too long pulses), the increase in

the membrane permeability can be reversed and the cell therefore stays alive. This effect

has been used as one of the medical application of the pulse power technology for

electrochemotrapy and gene and drug delivery into cells121, 122

.

2.5.5 The effect of low intensity pulsed electromagnetic field on biomechanical

properties of bone

Numerous studies investigated the effect of electrical stimulation on the bone growth

and bone healing. However, the effect of electromagnetic fields on structure and

biomechanical quality of bone has received scant attention. Two recent in vivo studies


42

investigated the effect of long-term (one for 45 days and other for10 months) extremely

low frequency magnetic field (ELF-MF) with100 µT, 500 µT and 1 mT at 50 Hz

parameters on strength, energy absorption capacity, ultimate stress, ultimate strain, elastic

modulus and toughness in rat bone17, 18

. The results showed that the ELF-MF can affect the

geometric and biophysical properties of bone in particular bone quality and strength. They

in general concluded that the long-term application of ELF-MF deteriorates bone quality

by influencing bone mineralization and collagen integrity. For example, they found that the

elastic modulus of the bones exposed to 500 µT was higher compared to the control

samples while their toughness decreased. It can show the bone samples become brittle due

to this magnetic field stimulation18

.

In contrast, another study demonstrated the effectiveness of PEMF exposure with 1

and 2 mT at 15 Hz with 5 ms pulse width on increasing the bone mineral density (BMD) ,

the maximum load bearing and the structural rigidity of rabbit bone 19

.

2.5.6 Application of high intensity pulsed electromagnetic field on bone

As mentioned in previous sections, several researchers have reported the beneficial

effect of the low-intensity pulsed electromagnetic field on the bone cell proliferation123-125

,

differentiation126, 127

, and in general osteogenesis development and some bone disease

treatment16, 128, 129

. Due to different PEMF parameters such as its frequency and magnitude,

there were also some negative reports for PEMF application on bone tissue 130-133

.

However, the effect of high level electromagnetic field on bone behaviour has received less

attention in previous studies and there were a few reported published research in this area.

Brighton et al. reported that the constant electric field with 1500 V/cm through

capacitive coupled electrodes increase the epiphysal plate growth12

. In contrast, a very

recent study investigated the effect of high-intensity pulsed electromagnetic field with 50


43

and 400 kV/m at 0.5 Hz frequency and 350 ns pulse-width, on bone formation and

osteoblast like cells proliferation. The results indicated that the high-level pulsed

electromagnetic field but with low frequency, suppressed the bone cells proliferation and

differentiation and mineralization and therefore, appeared to be harmful for the bone

generation13

.

Nevertheless, the effect of high frequency and high intensity pulsed electromagnetic

field simultaneously on bone tissue behaviour including its structural and biomechanical

properties seems not to have been tried until now.

Chapter 3: Pulse Power Generator Based on

Positive Buck-Boost Converter


45

3.1 Introduction

Pulse power technology in general is characterized by accumulated energy that is

released in an instance pulse. Hence, all pulse power generators aims to convert a low

power, long time input to a high power, short time output. Figure 3.1 illustrates the pulse

compression by a pulse power generator.

Figure ‎3.1 Conversion of low power, long time input waveform to high power, short time output waveform

by a pulse power generator

These systems typically store energy within electrostatic field (i.e. capacitors) or

magnetic field (i.e. inductors) over a comparatively long time and releases it very quickly

(in microseconds or less) which results in the delivery of larger amount of instantaneous

power (several kilowatts) in a very short time, though the total energy is the same 14

. To

generate such electromagnetic fields, high voltage and high current sources are required.

For prevention of thermal effect the pulse interval need to be very short 15

. Figure 3.2

shows the typical diagram for pulse power generators.


46

Figure ‎3.2 Typical diagram for pulse power generators

Pulse power system, with single shot and high peak power has initially been applied

for nuclear fusion studies and military defence applications. However, pulse power

systems generating repetitive pulses with moderate peak power have been developed

recently for industrial application 134

. Pulse power technology has also been used variously

in biology and medicine, especially at intercellular scale. Some of its applications are

controlling the ion transport processes across membranes, prevention of biofauling,

bacterial decontamination of water and liquid food, delivery of chemotherapeutic drugs

into tumour cells, gene therapy, transdermal drug delivery, programmed cell death which

can be used for cancer treatment and intracellular electro manipulation for gene transfer

into cell nuclei16

. Nevertheless, no published work has reported the utilization of high

voltage pulses with high frequency in skeletal system for stimulation purpose. This study

therefore aimed to investigate the effect of pulse power stimulation on bone material

properties.

Preventing the bone specimen from dehydration, it was wrapped in saline soaked

gauze during the excitation, which resulted in a resistive-capacitive load. Hence, a

combination of current –voltage sources was suggested as the required pulse power

generator 135

. Figure 3.3 reveals a general design for the combination of current and

voltage sources applied for pulse power generation.


47

Figure ‎3.3 A combination of current and voltage sources as a pulse power generator135

In this design, the inductor and the capacitor play the role of the current and voltage

sources respectively, supply the energy and generate the appropriate voltage level135

.

Positive buck-boost converter (a subset of DC converters) provides an appropriate

configuration for the proposed design.

3.2 Topology of pulse power generator

3.2.1 General configuration of positive Buck-Boost Converter

All of the conventional DC-DC converters use single stage and one transistor as a

switch. Therefore, the output power of these converters is generally limited to tens of

watts, because the single transistor has limitation in current handling. In addition, at a

greater current magnitude, the size of other components (inductor and capacitor) increases

and this results in higher losses and reduction of efficiency. For these reasons, multistage

converters are employed for high power application.

One of the proposed topologies is positive buck-boost (PBB) or non-inverting buck-

boost converter which is cascade combination of both buck and boost converters by

reducing a capacitor and an inductor 136

. Then, it has the characteristics of both of buck and

boost converters and can operate in step up and down modes with extra flexibility.

The main advantages of PBB are as following 137

:


48

Decrease the number of components compared to cascade

combination of two complete buck and boost converters

Capability to use in three other DC converters by only one set of

control 138

Higher level of flexibility for the inductor current control as a current

source by using the extra degree of freedom during the buck converter

Control of output voltage in the boost converter by charging the

capacitor

Control of the output voltage in the distinct border in the case of any

load or input voltage changing (then it can be used as a voltage source

as well)

Figure 3.4 shows the general configuration of a positive buck-boost converter which

is comprised of two cascade current and voltage source.

Figure ‎3.4 Circuit diagram of positive buck-boost converter

The inductor (L) is charged by the input voltage through switches S1 and S2, creating

the current source. The appropriate duty cycle of S1 controls the level of inductor current


49

during this period. When the initial switch (S1) is switched off, the freewheel diode (D1)

conducts the current and keeps it constant in the desired level. The switch S2 that is

connected to the capacitor through the diode (D2) composes the voltage sources which

provides the adjusted high voltage level. When S2 is turned off, the inductor current flows

to the capacitor and stores in the form of voltage.

3.2.2 Switching Modes

This topology operates in two major modes including current and voltage sources,

which are presented in three switching states.

First state: charging inductor (S1: on, S2: on)

As demonstrated in Figure 3.5, in this switching state, both S1 and S2 are turned on

and the input voltage appears across the inductor. Hence, the inductor current is increased

to reach the desired level.

Figure ‎3.5 First switching state, charging the inductor

Second state: circulating the inductor current (S1: off, S2: on)

At this state which is shown in Figure 3.6, when the inductor current reaches to

defined level, the controller turns S1 (the current source switch) off and disconnect the

input voltage source. The freewheel diode (D1), conducts and causes the inductor current to

circulate through S2 while D2 is reversed biased and separate the rest of the circuit.


50

Although, the voltage drop across the diode and switch discharges the inductor moderately,

because it is not significant, the circulating current is considered to remain constant (a

current source). In this switching state, the load is disconnected from the input voltage

during power delivery period (because S1 is turned off). Thus, although, this state may

cause some conduction losses, it is necessary avoiding any stability concerns and prevent

from wasting a large amount of energy through the input source while an arc occur

suddenly at the load side.

Figure ‎3.6 Second switching state, circulating the inductor current

Third state: charging the capacitor and load supplying (S1: off, S2: off)

During this state, S2 is switched off and the inductor current is pumped into the

capacitor and charges it. The capacitor energy is kept constant at a certain level while S2 is

turned on and stay until it is turned off again. This state will repeat until the capacitor

voltage reaches to the desired level, providing the voltage source for delivery to resistive-

capacitive load (bone specimen wrapped in saline soaked cloth). Figure 3.7 illustrates the

switching state during charging the capacitor.


51

Figure ‎3.7 Third switching state, charging the capacitor

After charging the capacitor to desired voltage level, when the load switch (S3) is

turned on the specified voltage pulse is delivered to load at a relatively short time (several

microseconds). Obviously, the switching frequency of S3, which determines the output

frequency, should be less than of two other switches to avoid any disturb in charging of

inductor and capacitor and controlling pulse voltage magnitude in output. When the

required energy was delivered to load from the voltage and the current sources, the

topology will switch from the supplying mode to the charging inductor (first state) and

repeat to provide the appropriate high-voltage pulses with desired frequency. Figure 3.8

shows the load supplying state while S3 is turned on.

Figure ‎3.8 Power delivery through the load switch


52

3.3 The pulse power generators applied in this study

As mentioned earlier, in this study pulse power generator was a positive buck boost

(PBB) converter, which has been built and tested, in the power electronic group, Faculty of

Science and Engineering, QUT earlier. The PBB circuit was installed in a box with lead for

safety and simplicity of displacement. The output pulses parameters (magnitude, frequency

and duty cycle) have been controlled using a programmed microcontroller. Two circuit

boards that were fitted with different controls were applied for pulse generating.

The first generator (Pulse Generator A; PGA shown in Figure 3.9) was controlled

utilizing an NEC 32-bit 64MHz V850/IG3 micro-controller and delivered pulses maximum

to 180V with 100 Hz frequency. Two voltage pulse levels (80V and 180V) were available

manually in the output via a key installed on the board. PGA worked accumulating the

energy in five, 10nF- capacitors until the voltage over the capacitors reaches the desired

level (maximum 180 V). Then the voltage pulse is released over the load in a short time

(several µs), closing a fast switch in the load side. The magnitude and frequency of the

voltage pulses and the pulse duration are the parameters which can be adjusted in the

output pulses through the programming of the microcontroller.


53

Figure ‎3.9 Pulse power generator A (PGA) with NEC microcontroller

A TMS320F28335 Digital Signal Controller (Texas Instruments) achieved the

control of the output pulses in the second pulse power generator (Pulse Generator B; PGB

presented in Figure 3.9). The output from this generator was pulses up to 500V at a

frequency of 10 kHz which could be adjusted manually with four potentiometers. The

potentiometers controlled the duty cycle, frequency, maximum current level and maximum

voltage level manually. As demonstrated in Figure 3.10, several parallel inductors and

capacitors acted as current source and voltage source respectively. This configuration

provides more flexibility for inductor current and capacitor voltage levels by entering or

emitting the appropriate number of inductors and capacitors from the topology. For

example, if two parallel inductors are emitted from the topology, the average inductor will

increase which results in reduction of the required switching cycles and therefore, will


54

decrease the switching loss. On the other hand, the desired voltage level will not be

accessible if more inductors are send out from the current source part.

In addition to previous switches, in the latter topology, an extra fast switch (S4) was

installed in parallel to the load. This extra switch could provide more power delivery to

load (bone specimen) by increasing the

and therefore, causes more current pass

through the load capacitor (bone sample). Furthermore, it helped to produce more complete

high-voltage pulses rather than sawtooth waveforms that were generated because of

discharging the relatively large load capacitor.

S2 required to switch on and off at very high frequency in order to provide the

assigned current and voltage level at appropriate period (based on the output frequency).

To prevent this switch from heating and damaging, a heat sink was therefore installed on it

and a fan was also cooling that.

Figure ‎3.10 Pulse power generator B (PGB) with Digital Signal Controller (DSP)


55

Figure 3.11 presents a sample high-voltage pulses that were generated by PGB and

applied on the cortical bone samples through capacitive coupling method.

Figure ‎3.11 High-voltage high frequency pulses with 500V at 10 kHz generated by GPB

3.4 Load modeling

Similar to most of other pulse power applications, the bone tissue which is the target

of pulse power stimulation in this study, can be modelled as a capacitive-resistive load.

The pulsed power signals were delivered through two wire leads attached to two parallel

aluminium plates for bigger size samples and two series of metal screws to increase the

electric field intensity applied on smaller size specimens.

Initially, the electrodes were connected to bone samples directly. However, because

the sample needs to keep moist with physiological saline solution all through the

experiments (in order to avoid bone from dehydration), the direct connection of screws

and cables with bone provides very low impedance, a significant current can pass through

the bone and making it dry and causing it to burn. Therefore, in the rest of the experiments

the aluminium plates and screws were covered by electrical isolation tape, changing the

characteristics of the bone samples from a resistive load to a capacitive load. The pulsed


56

electric field was then applied to the bone samples through capacitive coupling method and

in this case the thermal effect was reduced while the electric field intesity on bone structure

was increased.

Figure ‎3.12 Topology of pulse generator B with load modeling

As demonstrated in Figure 3.12, the bone sample wrapped in saline soaked gauze

which is placed between two-isolated electrodes, can be modelled as three serial units

composed of parallel capacitors and resistors. The capacitors placed in two sides that are

corresponding to dielectric isolated tape are smaller compared to the middle capacitor

which simulated the bone dielectric. As mentioned previously, as the load capacitors are

relatively large, in order to discharge them thoroughly and have a complete pulse applied

on the bone samples, an extra switch (S4) was utilized parallel to the load. A time interval

requires to consider between the time that S3 is turned off and when S4 is turned on, in

order to prevent the switch from heating and damage due to applying the whole capacitor

voltage across it.

Chapter 4: Physical Characterisation of Bone

Exposed to Pulse Power in Bending

Chapter 4: Physical Characterisation of Bone Exposed to Pulse Power in Bending

58

4.1 Introduction

The study of biomechanical properties of bone is important for determination of bone

quality in order to perform its vital duties in the body. As stated earlier, stiffness and

strength are two main primary characterising properties of bone, which are normally

determined with mechanical testings. Between different methods, the bending test is a

common method for determining the load bearing properties of long bones. It can be

performed on both whole or a strip of bone prepared from cortical dyphysis with a constant

cross-sectional area in spherical, rectangular or square shapes. Because this method is not

as accurate as the tensile test, it is not generally used as the standard materials testing

method. Nevertheless, because the preparation of samples and test performance is less

complicated in a bending test compared with tensile test, it has often been used in different

studies 139

. In addition, the flexure test is a useful simulation of many bone fractures

resulting from bending stresses 140

. This test also provides a combination of compressive

and tensile stresses applied simultaneously along two opposite sides of bone. Because bone

has an asymmetry structure, the compressive and tensile stresses may not be equal. Bone is

weaker in tension compared to compression, so in a bending test, failure typically spreads

from the tensile side to the compressive side 139-142

.

In this work, to investigate the possible effect of high voltage, high frequency pulses

on flexural stiffness of long bone, a non-destructive three-point bending test (in linear

elastic region) was applied before and after exposure. The process was carried out on both

whole bone and cortical bone beams. To perform the test non-destructively, preliminary

trials were conducted to determine the maximum load and deformation the bone can

withstand without sustaining plastic deformation. It was almost at 30% of the failure load.


59

Furthermore, in this study, the suitable method for applying pulse power and its

appropriate parameters were established. Hence, the magnitude and frequency of voltage

pulses were increased stepwisely to explore the effect of changing the pulse power

parameters on bone elasticity. This chapter presents the results of three-point flexure

testing of cortical bone in two control and PP-exposed groups through four series of

experiments.

4.2 Factors influencing experimental measurement

There are two types of bending tests: three-point and four-point bending (Figure 4.1).

Three-point bending is simpler to implement, but it has the disadvantage of creating high

shear stresses at the centre of the load point. Four-point bending creates pure constant

bending with zero shear stresses between two upper load points. However, four-point

bending requires exerting equal force at each loading point. This requirement can be

obtained easily in regular shaped specimens but is difficult to achieve in whole bone,

which has a non-uniform cross section, a small length to diameter ratio and inconsistent

intramedullary components. Therefore a three point bending test is often used to determine

the mechanical properties of bone in bending 140, 142

.


60

Figure ‎4.1 Two types of bending tests and the compression-tension relationship of forces along the surfaces

of the loaded specimens[3]

Two important keys for performance of a successful three point bending test are the

distance between the lower supports which influence the length to diameter ratio and the

radius of curvature of the upper loader affecting the bone deformation beneath the loader

142. According to the standard testing method, adopted by the American Society of

Agricultural Engineers (ASAE), a three point bending test of animal bone should be

performed on straight bone with a symmetrical cross section and length to diameter ratio

greater than 10143

. However, the ideal length to diameter ratio (L:D) for reduction of bone

displacement and shear stress is 20. Since L:D ratio for whole long bone is typically less

than 10, shear stress causes substantial displacement and results in an overestimation of

strain and an underestimation of Young’s modulus. The best way to reduce the error is

strain measurement using a resistance strain gauge bonded directly to the middle of the


61

bending specimen140, 142

. The other way to obtain more accurate results in a three-point

bending test is to test a strip of bone machined from cortical diaphysis with a length to

thickness ratio of greater than 10 instead of whole bone142

. Hence, in this study, after some

preliminary testing on whole bone, bone strips obtained from cortical femoral dyaphysis

were applied as the test specimens. This reduction of the size of the samples can also

increase the possible effect of pulsed electric field stimulation on bone samples.

Biomechanical properties of bone are influenced by numerous factors such as the

anatomical site from which the bone sample is obtained, the activity level, hormones,

general health, sex and age of the donor. In addition, the preparation and storage condition

of the test samples like bone hydration and temperature can affect the mechanical

properties of the tissue (described in chapter 2). For example, the Young’s modulus and

strength of bone typically increase with the drying of bone, while the toughness decreases.

To obtain more precise results from testing, it is necessary to prevent bone samples from

drying. Therefore, the specimen should be bathed in physiological saline or wrapped with

gauze or paper tissue soaked in saline all through the tests. It is also necessary to obtain the

samples from similar sources and test them in similar environmental conditions (similar

temperature, humidity and air conditioning). Because the tissue degradation starts within

hours of removing bone from the body, it is better to take the samples as close to death of

the animal as possible.

The behaviour of bone under a bending test, similar to other mechanical testings, is

monitored by plotting the standard load-displacement curve obtained from data recorded

by the testing machine and its normalisation to stress-strain curve. Due to bone

viscoelasticity the strain rate during the mechanical tests also can affect the output results.

Depending on the nature of the investigation, the rate of loading could be in the


62

physiological range when studying normal bone under in vivo, non-impact condition and

much higher for trauma fracture studies 141, 142

.

4.3 Materials and MethodsSample preparation

Test specimens can be obtained from different animal sources like pig, rabbit, dog,

rat, cow (bovine) and sheep (ovine). Nevertheless, because ovine bone is structurally and

hormonally similar to human bone, it has been widely used in orthopaedic and trauma

research 144, 145

. Hence, in this work sheep long bones were applied for preparing test

samples.

Both right and left legs of two female merino sheep, that were freshly amputated at

the Queensland University of Technology’s “Medical Engineering Research Facility”

(MERF) located at the Prince Charles hospital, Brisbane, were obtained for preparation and

testing. Each leg provided three long bones including femur, tibia and metatarsus. To

minimize the changes of the bone in vitro properties, the hind legs were removed quickly

from the bodies and the femoral, tibial and metatarsal bones were amputated from them,

separated from the soft tissue (taking care not to put notches on the bone since this will

weaken it), and kept in 0.15M physiologic saline. Fresh samples were kept at 4 °C for

immediate experimentation. The rest of the samples were wrapped in gauze soaked with

saline and frozen at -20 °C. They were thawed at room temperature and equilibrated with

the room environment (20 to 22 °C temperature and about 60% humidity) before testing.

When the experiments continued for more than one day, bone samples can be refrigerated

for several days between tests without considerable alteration in their properties 140

.

The size and the shape of the test samples influence the outcomes of mechanical

testing as well, so preparation and processing of the specimens in the appropriate shape

and the desired structural level are important 140, 142

. In this study, which involves


63

performing non-destructive tests, to reduce the effect of size and shape of the test sample

on determination of bone elastic property, the same sample was subjected to the same

loading condition in elastic region before and after pulse power excitation.

The first experiment was conducted using whole metatarsus from the left and right

legs of one sheep to compare the possible effect of pulse power signals on whole long

bone.

For other experiments, 14 bone strip samples were obtained from the femoral

cortical dyphysis. They were cut with a handsaw and were polished afterwards using small,

precise files and fine sandpapers. Then, the physical dimensions of the bone samples were

measured. All samples had a length to thickness ratio greater than 10. To prevent the bone

samples from dehydration, the specimens were wrapped in saline-soaked gauze all through

the experiments (even when exposed to PP).

4.3.2 Three-point bending test

Following the preparation and processing of the bone specimens, non-destructive

three-point bending tests were conducted on bone samples using an Instron testing

machine (model 5944, 2KN load cell) before and after pulse power excitation. The span

length between the lower supports are adjustable to provide more precise bending tests

condition as required. The load up to about 30% of failure load (which was determined in

preliminary trials) at 10 mm/min for the whole bone samples and 0.1 mm/min for the bone

strip specimens was applied and load-displacement (LD) data were recorded. The extrinsic

stiffness (the slope of the load-displacement curve in linear region) was calculated in all

experiments.

To avoid the slippage of bone specimens on the supports and to hold them in a stable

orientation, particularly for the whole bone which is very slippery on the hard, round


64

stainless steel surface of the rig’s bending frame, approximately 50 N preloading for the

whole bone and 1-3 N for the bone beam samples were applied146

. Figure 4.2 presents the

three-cycle loading up to 30 N on the bone strip samples. From the figure it can be seen

that although the stress level was in the elastic region, the loading and unloading curves did

not completely coincide. This is the natural characterestics of the viscoeleastic material

such as bone in which some of the strain energy is stored as potential energy and released

when the stress removed and some is dissipated as heat. This energy is represented by the

area enclosed by loading and unloading curves (the hystersis loop)147

.

Figure ‎4.2 Three-cycle bending load in linear elastic region on the bone strip sample

If a test is repeated on the same sample (below elastic limit) within a short period, the

resulting load-displacement curves do not match exactly but appear to converge toward a

stable curve. It was advised to apply the load to the samples for several cycles before

performing the actual test146

. In this process which is called preconditioning, some settling

happens between the specimen and the mounts148

. Some other researchers have disagreed

with this opinion and argued that what is finally being measured after the preconditioning


65

process is not the natural property of the specimen but a modified one by a series of

cyclically applied load and so they apply just a single-cycle loading to a specimen141, 149

.

In this study, in spite of applying a preload on the bone specimens some slippage on

the stands was still possible during the experiments. Therefore, three cycles of loading on

the same sample provided a consistent load-displacement curve and if the specimens were

going to move they did so over the first cycle and reduced or eliminated errors which

ensure greater confidence in the accuracy of the results. For example, as illustrated in

Figure 4.3, a small drop occured in the first cycle that was omitted in the two other cycles

and might be caused by the slipage of the sample on the stands.

Figure ‎4.3 A small drop on the first cycle of bending test in elastic region that was removed on the further

cycles

4.3.3 Data collection and calculation

As demonstrated in previous section, the extrinsic stiffness of bone samples was

derived from load-displacement curves in the third cycle of loading which the Instron

testing machine recorded. The three common properties of bone which are often

determined for three-point bending are: the area moment of inertia, Young’s modulus

(modulus of elasticity) and toughness modulus. These parameters can be calculated using


66

beam-bending theory with the generalized assumption that bone is an isotropic,

homogeneous and linearly elastic material139

. For three-point bending test the relevant

equations are 142

:

(4.1)

(4.2)

(4.3)

(4.4)

(4.5)

Figure ‎4.4 Three point bending test142

Where M is bending moment; σ is applied stress, ε is strain; E is Young’s modulus; S

is stiffness; u is modulus of toughness; U is the strain energy (area under the stress-strain

curve); c is the distance from the farthest point in the cross-section to the neutral axis i.e.

a/2 (elliptical shape) or t/2(rectangular shape), F is the applied force, d is displacement, L

is the distance between two supports (presented in Figure 4.4) and I is the area moment of

inertia. The area moment of inertia for the whole bone with an assumed elliptical cross-

section (Figure 4.5) can be determined from equation 4.6 as:

(4.6)


67

Figure ‎4.5 Assumed elliptical cross-section for whole bone

The area moment of inertia can also calculated using numerical modeling, e.g.

simulation in a finite element software like ANSYS 150

following the accurate

measurement of the cross sectional area of the bone sample usually with a vernier caliper.

ANSYS has a useful potential for calculating the cross sectional geometrical properties of

materials which was applied here to determine the area moment of inertia of the whole

bone samples. Figure 4.6 presents the cross-sectional area of sheep whole metatarsus in

ANSYS.

Figure ‎4.6 Cross-sectional area of whole long bone in ANSYS for determination the area moment of inertia


68

Table 4.1 compares the area moment of inertia of an ovine metatarsus using ANSYS

and calculating by equation (4.6). The results of two methods show the differences less

than 1%. However, here to calculate Young’s modulus from equation 4.4, the area moment

of inertia obtained from ANSYS, was applied.

Area moment of inertia

(mm4)

From ANSYS From

PP-exposed sample 1217.9 1223.75

Control sample 1480.2 1487.82

Table ‎4.1 Comparison of the area moment of inertia of the whole bone samples obtained from ANSYS and

calculation

For a machined beam specimen with rectangular cross-section (Figure 4.7), the area

moment of inertia is calculated as:

(4.7)

Figure ‎4.7 Bone strip obtained from the cortical diaphysis

4.3.4 Pulse Power excitation

As described in chapter 2, two pulse power generators based on the topology of

buck-boost converters were applied in this study with adjustable voltage and frequency

capability. The first one (Pulse Generator A; PGA) which was controlled with

programmable microcontroller V850E/IF3 (NEC), delivered pulses up to 180V magnitude

and 100 Hz frequency. Two voltage pulse levels (80V and 180V) were available in the

output via a manual key installed on the board. A TMS320F28335 Digital Signal

Controller (Texas Instruments) achieved the control of the output pulses in the second


69

pulse power generator (Pulse Generator B; PGB). The output of this generator was

adjustable pulses up to 500V at 10 kHz by four potentiometers. The topologies and the

instructions of pulse power generators were presented in chapter 3. The methods of

applying output pulses over the bone samples varied for different experiments and are

explained in the next section.

4.4 Experimental procedure and Results

The experiments were conducted in two main rounds: I) applying voltage pulses up

to 180 V at 100 Hz using PGA on two types of the specimens (whole bone and bone

strips); II) Pulse Power exposure up to 500V and frequency of 10 kHz via PGB on cortical

bone beams. In all experiments, to determine possible changes in the elastic properties of

bone in response to pulse power stimulation, non-destructive three-point bending tests

were performed and the elastic stiffness was determined from the slope of load-

deformation curve in the third cycle of bending test. Then, the Young’s modulus of the

bone samples were calculated and compared before and after pulse power excitation using

stiffness and the area moment of inertia in equation 4.4.

4.4.1 Pulse power excitation with voltage up to 180V and 100 Hz frequency

To determine the assurance limit for pulse power signals and establish a suitable

experimental set-up, a preliminary experiment was conducted applying voltage pulses with

magnitude of 80 V at 100 Hz frequency on whole sheep metatarsus via two wire leads

attached directly to two ends of the long bone. To control the environmental condition and

bone hydration, the test sample was wrapped in saline-soaked cloth and was placed in a

covered box during the experiment. The shape of output pulses showed that due to high

impedance of the bone samples, the capacitors could not completely discharge and deliver


70

enough energy to whole bone sample. To resolve this issue, the pulse duration of the

output signals were prolonged. Therefore, the capacitors had more time to discharge and

deliver more energy to the whole bone sample. After establishing a stable and safe set up

and due to no significant changes in Young’s modulus of the bone sample after pulse

power excitation compared with that of before excitation, the main experiments were

performed with pulses of 180V at 100 Hz.

After wards, two series of experiments were carried out on whole bone and bone

strips using the first pulse generator.

Whole bone stimulation with Pulses of 180 V at 100 Hz

In the first experiment with PGA, the whole metatarsus were taken from the right and

left legs of a sheep as the control and PP-exposed samples. The left metatarsus was

exposed to pulse power signals (180 V, 100 Hz and 560 µs pulse duration) directly via two

cables for an average of 4 hours per day over 5 days. A three-cycle bending test was

performed non-destructively before and after every hour of pulse power excitation. The

test was carried out by applying load up to 200 N at a displacement rate of 10 mm/min.

The right metatarsus was used as a control sample in the same environmental condition but

without applying pulse power. Both control and PP-exposed samples were kept moist

during the test. Similar bending tests were applied on the right metatarsus at the same time

for consistency and comparison.

Figure 4.8 compares the changes in the Young’s moduli of the sample exposed to

pulse power for 5 days with those of the control sample in the same period. From the

graphs, it can be seen that the elastic properties of both the control and the PP-exposed

samples have similar fluctuation of less than 10%, which may not be a consequence of

exposure to pulse power rather it could be related to environmental condition or usual


71

experimental errors. It suggests that high voltage, high frequency pulses with selected

parameters, through direct connection of the electrodes to test bone sample, did not affect

the flexural elastic property of a long bone sample over 5 days excitation.

Day

1

Day

2

Day

3

Day

4

Day

5

0

5

10

15Control sample (unexposed to PP)

Yo

un

g's

mo

du

lus (

GP

a)

Day

1

Day

2

Day

3

Day

4

Day

5

0

5

10

15

PP-exposed sample

Yo

un

g's

mo

du

lus(M

Pa)

Figure ‎4.8 Variation of Young’s modulus of the ovine metatarsus exposed to 180V and 100 Hz pulses over 5

days (PP-exposed sample) compared to that of the control sample

Bone strips stimulation with pulses of 180 V at 100 Hz

The next experiment was performed on bone strips taken from a sheep femur

cortical diaphysis instead of whole bone. In this case, using more uniform and smaller size

samples can decrease the experiment errors causing by bending test set up while increase

the possible effect of pulsed electric field on bone tissue. In addition, to increase the

contact surface between electrodes and test samples, two aluminium plates

(98.8mm×10.44mm×0.7mm) were placed parallel on both sides of the test bone. To avoid

any electrical contact between aluminium plates, they were kept separate, using two plastic

screws through holes at each end. The bone specimen was wrapped in saline-soaked gauze

all through the experiment, for its prevention from dehydration. Figure 4.9 presents a

sketch of the experimental set up for pulse power stimulation of the bone samples via two


72

aluminium plates. The voltage pulses at 180 V, 100 Hz and 10 µs pulse width were applied

through the plates for 9 days with approximately 6 hours excitation per day. Three-point

bending tests for three cycles up to 30 N (in linear elastic region) with extension rate of 1

mm/min were conducted before and after the pulse power exposures in each day.

Figure ‎4.9 Sketch of experimental set-up for pulse power stimulation of the cortical bone strip sample

To explore the effect of the duration of pulse power excitation on bone elasticity, the

values of Young’s modulus of sheep femoral cortical beams exposed to 180 V with 100 Hz

high voltage pulses over 9 days were compared with that of the same samples without

pulse power excitation but in the similar environmental condition (Figure 4.10). In this

experiment, the duration of pulse power excitation has been increased as a variable which

might affect the results. The average values of Young’s modulus of the test samples after

each day under mentioned parameters of excitation were in the same range as those in the


73

day without pulse power exposure (with less than 10% variation which is probably

because of the experimental errors). Therefore, these results can imply that high voltage,

high frequency pulses up to 180 V at 100 Hz has not apparently affected the flexural

elasticity of cortical bone tissue during 9 days.

D1

D2

D3

D4

D5

D6

D7

D8

D9

No p

ulse

power

0

10

20

30

40 Y

oung

's m

odul

us(G

Pa)

Figure ‎4.10 Variation of the Young's modulus of femoral cortical strips exposed to 180V at 100 Hz pulse

power over 9 days compared with that of the same samples without pulse power excitation

4.4.2 Pulse power excitation with pulses up to 450 V magnitude at 10 kHz frequency

Following the observation of no significant changes in the elastic property of bone

samples due to the voltage pulses up to maximum 180V at 100 Hz frequency, pulse

generator B (PGB) with similar base circuit but different controller part was applied to

deliver adjustable voltage pulses up to 500V at 10 kHz. In this stage, two series of

experiments were conducted with new pulse generator at two different series of parameters

as presented below:

Pulses up to 450 V at 340 Hz

In the first experiment with PGB, the output pulses parameters were increased to

450V at 340 Hz (allowing the capacitors to discharge completely and deliver more energy

to the bone samples). Similar to previous experiments, the cortical bone samples were


74

excited via two parallel aluminum plates placed on two sides of the bone samples for three

days. To prevent the bone samples from dehydration, they were wrapped in saline soaked

gauze during the excitation. Because saline solution is a good electric connective liquid

and provides very low impedance therefore, significant current can pass through the gauze

soaked with saline and thereby small part will pass through the test sample placed in

between and hence, the pulse power effect especially on the bone sample will reduce. To

resolve this issue, the aluminum plates were therefore, covered with electrical isolation

tape in order to change the characteristics of the bone from resistive load to a capacitive

load. In this case, the pulsed electric field was applied to the bone samples using capacitive

coupling method reducing the thermal effect while increasing the electric field effect on

the bone samples.

Again, to determine the bending elastic responses of cortical bone samples, similar

non-destructive three-point bending test was conducted before and after pulse power

excitation. Three cycles of loading up to 30N after a preload of 1-5 N was applied. The

experiment was conducted over three days with approximately 6 hours excitation per day.

The applied energy per hour with mentioned parameters of pulsed electric field was nearly

9 times more than that of in previous experiment. Table 4.2 presents the Young’s modulus

of the cortical femoral samples after pulse power excitation compared with the values

before pulse power exposure. The comparison of Young’s modulus of the cortical bone

samples exposed to pulse power, before and after excitation in each day showed similar

changes less than 5% which may not be a consequence of exposure to pulse power signals.


75

Young’s modulus(GPa) PP-exposed group

before excitation

PP-exposed

group after

excitation

Day1 23.67±1.478 22.94± 0.9590

Day2 23.41 ± 0.9717 23.51 ± 1.105

Day3 23.53 ± 1.483 23.73 ± 0.8826

Table ‎4.2 Mean value± standard deviation for Young’s modulus of cortical bone before and after pulse

power excitation (450V at 340Hz) in three days

The non-parametric tests are usually more conservative compared to the parametric

ones (i.e. ANOVA and t-test) especially when the sample size is small and the normality

cannot be tested. Therefore, due to small size sample, the non-parametric test (Kruskal-

Wallis test) was performed. The result of the non-parametric analysis showed no

significant differences between the Young’s modules of the control samples (20.93±4.384

GPa) and those values of the samples exposed to pulse power in each day (P =

0.729>0.05). These results can imply that the pulse power excitation with nominated

parameters during three days had no considerable effect on the flexural elastic property of

sheep cortical bone samples.

Pulses up to 450 V at 10 kHz

In this trial, the frequency of high voltage pulses was increased to 10 kHz with 28µs

duty cycle (about 30 times more than previous experiment) but with similar magnitude.

Figure 4.11 shows the waveform of high power pulses applied with new parameters on

cortical bone beams through capacitive coupling method. The experiment was conducted

in the same manner as the previous test but with new pulse power parameters over 5 days,

for approximately 5 hours per day.


76

Figure ‎4.11 The pulse power waveform with 450V magnitude and 10 kHz frequency applied on cortical

bone samples

The bar graphs showing the Young’s modulus of the PP-exposed and the control

cortical bone samples are summarised in Figure 4.12.

Pulse Power with 450V and 10 KHz

Bef

ore P

P e

xcita

ion

Afte

r PP

exc

itatio

n

Contr

ol sam

ples

0

10

20

30

Yo

un

g's

mo

du

lus(G

Pa)

Figure ‎4.12 Elastic properties of the cortical bone samples exposed to pulse power (450 V at 10 kHz) before

and after excitation compared with those values of the control samples


77

The evaluation of the flexural stiffness of cortical bone sticks exposed to high

voltage, high frequency pulse signals (450V at 10kHz) reveals less than 5% differences

before and after 5-day excitation, that was almost in the same range of the Young’s

modulus of the samples which were not exposed to pulse power. These results generally

revealed that the pulse power signals up to 450 V and 10 kHz over 5 days did not affect the

stiffness of the cortical bone material.

4.5 Discussion

This chapter covered establishing a suitable set up for pulse power application on the

cortical bone samples under four experiments using two different pulse power generators.

The parameters of the pulse power signals changed from 80V to 450V and from 100 Hz to

10 KHz in frequency. The outcomes confirmed the safe and controlled application of pulse

power with parameters up to 450 V and 10 kHz using capacitive coupling method with no

thermal or destructive effect on the bone structure. The results of three-point bending tests

under four different conditions of pulse power stimulation demonstrated that this electrical

“loading” approach resulted in no significant changes to the elastic bending characteristics

of cortical bone samples. The experiment was started using whole bone as the test samples

and then, the size of the samples reduced to cortical bone strips increasing the possible

effect of pulsed electric field on the samples.

To diminish the effect of inhomogeneity and anisotropy of bone structure attributed

to different sites and anatomical locations and from one sample to another, on the results,

the same samples were applied all through the experiments in each round. Thus, bending

tests were performed non-destructively below the elastic limit by loading the bone up to

20% of the failure load for three cycles. This also helped to reduce the influence of

mechanical testing on the outcomes. In addition to the main experiments, some control


78

tests in similar environmental conditions were performed on similar bone specimens but

without exposure to pulse power. These control tests helped to reflect any possible effects

of mechanical test and environmental condition on the results.

The outcomes of the tests showed similar fluctuations up to 10% in Young’s

modulus for both the control and PP-exposed samples. However, performing the

experiments in a more isolated environment like an incubator can reduce these

fluctuations. The effect of the experimental errors including displacement of the samples

on the stands or environmental conditions like airconditioning were such that when the

samples were not removed from the stands during the experiments, the variation of

Young’s modulus, measured for both the control and PP-exposed specimens were reduced

to less than 5% .

This work applied and compared two methods of pulse power application including:

i) direct contact of electrodes with bone samples (in the first two experiments) and ii)

capacitive coupling method (in two other experiments) .In agreement with the previous

published studies 16, 89

, the direct contact method showed some destructive effect causing

the bone dehydrating and burning in the region of contact between the bone samples and

the electrodes. On the other hand, to prevent bone from dehydration (which influences

bone mechanical properties) and mimic the real bone tissue in the body saturated in body

fluid, it was wrapped in saline soaked cloth. This issue provides less impedance and

resulted in passing significant current through saline rather than excite the bone samples.

Hence, in general the application of capacitive coupling method through indirect contact of

the bone samples with the two parallel plate electrodes (changing bone characteristics from

resistive load to capacitive load, reducing the thermal effect, and increasing the pulsed


79

electric field effect) is more desirable and therefore, this method was applied in the rest of

experiments.

Although non-destructive testing could diminish the effect of interference of

mechanical testing on the experiment outcomes, it could just provide comparison of the

elastic properties of the bone samples. To consider the influence of pulse power

stimulation on other functional properties of bone (such as bone strength and toughness)

and in general fracture behaviour of bone in response to pulse power excitation, it is

required to conduct the mechanical tests until failure point. In addition, application of

samples with smaller geometries could increase the possibility of pulse power influence on

cortical bone material. Along this way, the other chapters cover investigating the

mechanical behaviour of smaller sized bone samples until failure through tensile and

compression testings and ultrasonic technique.

Chapter 5: Effect of Pulse Power Exposure on

Functional Behaviour of Cortical

Bone in Tension

Chapter 5: Effect of Pulse power exposure on Functional behaviour of Cortical Bone in

Tension

81

5.1 Introduction

The study of the mechanical behaviour of bone up to failure provides important

information on bone quality, particularly for investigating the effect of external stimuli or

disease on bone fracture risk. A plain fracture is usually generated in a material body (for

example bone) due to an applied stress that can be constant (static) or slowly changing

with time (quasi-static). Although, imposed stress may be tensile, compressive, bending or

torsion, this chapter is restricted to uniaxial quasi-static tensile load, because firstly, bone is

usually weaker in tension and secondly, tensile test is more severe compared with other

mechanical testings.

Ductility and brittleness of bone tissue are the bone characteristics that can show the

bone quality and might be influenced by exogenous stimulations. In general, there are two

possible fracture modes for materials (e.g. bone): ductile and brittle. This categorization is

attributed to the capability of a material to sustain plastic deformation. Brittle materials

normally show little or no plastic deformation with low energy absorption before fracture

while, for ductile material, there is approximate extensive plastic deformation with high

energy absorption before failure. Furthermore, crack initiation and propagation, which are

two essential steps in every fracture, are different in ductile and brittle fractures. For

ductile fracture, the crack propagates relatively slowly accompanied by considerable gross

deformation in the fracture surface (stable crack). Conversely, in a brittle fracture, the

cracks spread rapidly and continue once they are initiated without increase in applied stress

magnitude (unstable cracks) 151

.

The area under the stress-strain curve during loading represents the strain energy per

unit volume which an object (e.g. bone) absorbed. Toughness (fracture energy) gives the


Tension

82

strain energy required to completely break the material and is the area under loading curve

until the failure point 152

. Conversely, the area under the unloading curve gives the energy

released by the material. For elastic materials before yield point, because loading and

unloading curves coincide, the areas under them are equal and therefore the loss of energy

dissipated as heat would be zero. For viscoelastic material (like bone), apart from whether

stress or strain are small or large, some of the strain energy is stored in the body and some

dissipates as heat. Therefore, loading and unloading graphs do not coincide and there is

always an area (hysteresis loop) between two curves which is dependent on the strain rate

and reveals the amount of energy dissipated as heat during the recovery path 147

. Also,

when an elastic-plastic material is loaded into the plastic region, the loading and unloading

curves do not match. So that, the absorbed energy is more than energy released and their

difference shows the energy loss (dissipated as heat) by the material 153

. The area enclosed

by the hysteresis loop (loading-unloading cycle) represents energy loss (hysteresis energy).

To occur and progress a fracture, the strain (elastic) energy dissipated through the

crack propagation is required to be equal or greater than the essential energy to form the

surface material. Therefore, the more energy dissipated through fracture procedure ( the

more toughness), shows more resistance of the material to fracture and the more difficult is

to break it 154

.

Along this way, the determination of the fracture energy (toughness) and the bone

strength (ultimate tensile stress) can be used to evaluate the resistance of the tissue to

deformation and ultimately failure and indicates the bone material ductility or brittleness

and their variations due to pulse power excitation. The deliberation of the hysteresis energy

for the bone samples before and after pulse power exposure also provides more

information about bone behaviour due to this kind of stimulation. In addition, the


Tension

83

evaluation of the bone fracture surface (fractographic study) using scanning electron

microscopy (SEM) can offer supplementary information on bone quality in response to

high power, pulsed electric field excitation.

This chapter compares and contrasts the toughness, the strength and the hysteresis

energy between the control cortical bone samples (unexposed to pulse power) and the PP-

exposed samples in order to investigate how pulse power excitation influences the bone

quality in fracture mode. In addition, fractographic examination via SEM protocol provides

more detailed information on the microstructure and patterns of the fracture surfaces of

both normal and treated samples.

5.1.1 Fractographic study

The fracture surfaces, which are created during fracture process, can be analysed at

both macroscopic and microscopic levels. Visual examination of the macroscopic fracture

features can provide a strong clue of the breaking process. In addition, microscopic

examination of the failure surface, usually via scanning electron microscopy (SEM), can

present more complete information about the fracture mechanism. This kind of studies is

called fractographic151

.

Ductile and brittle fracture surfaces show their own distinctive characteristics on both

macroscopic and microscopic levels. Figure 5.1 indicates schematic illustrations for typical

macroscopic tensile test fracture profiles. In brittle materials, the crack propagates rapidly

almost perpendicular to the direction of applied stress and therefore creates a nearly

smooth, plateau fracture surface (Figure 5.1C). The fracture surface may have a bright

granular appearance 155

.


Tension

84

Figure ‎5.1Typical macroscopic tensile test fracture (A) ductile shear fracture (B) moderately ductile fracture

(C) brittle fracture 155

Figure 5.1 A and B demonstrate that macroscopically, ductile fractures have

generally uneven and rough surfaces with almost a fibrous appearance. On microscopic

level, the ductile surfaces also appear to be very rough and irregular, consisting of dimples

and micro voids.

5.2 Materials and Methods

5.2.1 Practical consideration for tensile testing

Tensile testing can be one of the most accurate methods used to evaluate bone

fracture behaviour and measure its biomechanical properties like: Young’s modulus, the

ultimate tensile strength and the fracture and hysteresis energy. Nevertheless, no specified

standards and clear guidelines on the specimen shape exist for the tensile testing of cortical

bone. Although, several standards including the American Society for Testing and

Materials (ASTM) standards, were established for the tension testing of engineering

materials, these standards cannot always be applied to the bone samples due to restriction

imposed by the size and the geometry of the specimens, the difficulties in preparing and

gripping the test samples and/or comparatively low loads that can be applied on the bone


Tension

85

samples 148, 156

. However, this work attempted to employ ASTM standards developed for

testing engineering materials whenever it was possible.

The main source of errors in tensile testing method is slippage of the test sample in

the grips. To reduce this error, both ends of the bone specimens were roughened using

sandpaper and files.

Rectangular and dumbbell are the most commonly used specimen shapes for tensile

testing. Although, the strip samples are easier to manufacture, to diminish the tensile

strength measurement errors, it is recommended to use dumbbell shape samples 156

(Figure5.2). Furthermore, the reduced cross-section area in the middle portion of the

dumbbell shaped specimen (gage section) compared to that of the two ends (grips portion)

of the specimen, causes the majority of strain to occur in the central part and therefore,

decreases the chance of fracture in the grip parts 142, 156

.

Figure ‎5.2Dumbbell shape specimen with round junction (GL, GW and GT are gage length, gage width and

gage thickness respectively)

5.2.2 Sample preparation

A fresh tibia, obtained from a slain ovine within 24 hours of slaughter, was used for

tensile test specimen preparation. The surrounding soft tissue was removed from the bone

and the whole tibia was wrapped in 0.15M physiologic saline soaked cloth, placed in


Tension

86

sealed plastic bags and kept at -20 until required for testing. Prior to specimen

treatment, the whole bone sample was thawed in saline at room temperature.

Dumbbell-shaped specimens were prepared from the cortical diaphysis of ovine tibia

in the following way. The epiphysis parts of ovine tibia were cut off from two ends and the

mid section (shaft) was separated to equal four segments (25% of total diaphysis length)

named TP (Tibia Proximal), TMP (Tibia Midshaft Proximal), TMD (Tibia Midshaft

Distal) and TD (Tibia Distal) according to their positions156

(Figure 5.3).

Afterward, each segment was cut into four to six pieces and polished into strips of

specific thickness values using the small, precise files and fine sandpapers. These strips

were then filed into dumbbell shapes with the longest dimension corresponding to the

longitudinal axis of bone. To compare the effect of specimen geometry on the tensile

properties of bone samples, both types of dumbbell shape were created. The strip samples

provided from the TD segment were converted to the dumbbell- shaped with sharp

junction specimens (Figure 5.4), while the beams obtained from three other segments were

changed to dumbbell shape samples with a round junction (Figure 5.2). Throughout the

preparation process, the bone specimens were kept moist with 0.15 M physiological saline

solution. The total 20 test samples were prepared for the experiments in this part.


Tension

87

Figure ‎5.3 Sketch of partitioned tibia used for tensile test specimen preparation

Figure ‎5.4 Dumbbell shape specimen with sharp junction (GL, GW and GT are gage length, gage width and

gage thickness respectively)

The specimens were labelled according to their sites and segregated randomly into

two groups of PP-exposed samples which were exposed to high power, high frequency

pulses (pulse power) and the control specimens that were kept in the same environmental

conditions (e.g. similar room temperature, humidity) as the PP-exposed samples, but

without pulse power excitation. All samples were then refrigerated at 4 , wrapped in

0.15M saline soaked gauze for immediate experimentation and stored at -20 for later

testing.


Tension

88

5.2.3 Pulse Power excitation

The second pulse power generator (PGB) was used to stimulate cortical bone

samples with high voltage pulses up to 500 V at 10 kHz frequency (maximum available

voltage and frequency that could be obtained from the pulse generator). The pulsed power

signals (presented in figure 3.11) were delivered trough two aluminium strips covered with

electrical isolation tape (similar to last experiment setup in chapter 4) using capacitive

coupling method. In this case, the thermal effect and thus, burning and damage probability,

were reduced while the effect of the electric field would be increased on bone structure. To

prevent the bone samples from dehydrating during the excitation process, both the control

and the treated samples were kept moist wrapping in 0.15M physiological saline soaked

gauzes during the experiment.

The bone specimens were exposed to a high power pulsed electromagnetic field in

the middle (gage portion). Figure 5.5 shows a sketch of the experiment set up including the

tensile test samples wrapped in saline soaked cloths which were placed in their gage

section between two parallel isolated aluminium plates.


Tension

89

Figure ‎5.5 Top view of a sketch of experimental set up for Pulse Power excitation of the bone tensile test

specimens between two isolated aluminium strips

The duration of pulse power stimulation were 28, 35 and 145 hours on three groups

of bone specimens under similar environmental condition. This provided the possibility to

investigate the effect of timing in the results of the experiment, the period of excitation was

considered as a variable. In addition, to consider any possible effect of setting and

atmospheric conditions on the results, the cortical bone samples control group were placed

in similar environmental circumstances as the PP-exposed groups all through the

experiment.


Tension

90

5.2.4 Uniaxial quasi-static tensile test

For tensile testing, the prepaired cortical bone specimens were placed in the grips

attached to an Instron testing machine (model 5944, 2kN load cell). The tensile tests were

conducted in two protocols. Firstly, for hystersis energy measurement, one cycle of loading

up to 150N (after yield point and before failure) and unloading down to 0 N, at 0.1mm/min

strain rate was conducted on both the control and the PP-exposed samples before and after

the period of 145-hour pulse power excitation. Secondly, in oreder to evaluate the effect of

high voltage, high frequancy pulses on the toughness and strength of the cortical bone

samples, both control and PP-exposed samples were loaded at an extention rate of 0.1

mm/min until compelete fracture (90% load drop) occured. The load-deformation results

were recorded and converted into stress-strain data for analysis. During testing the gage

portion of the specimens were wrapped in saline soaked gauze to prevent the bone from

dehydrating (Figure 5.6).

Figure ‎5.6 Tensile testing of the cortical bone specimen

Tensile grips

Bone sample wrapped in saline

soaked gauze


Tension

91

5.2.5 Scanning electron fractograph

To characterize any alterations in the cortical bone specimen’s microstructure due to

pulse power excitation, after tensile testing to failure, the fracture surfaces of both control

and PP-exposed samples were examined microscopically using a scanning electron

microscope (FEI QUANTA 200) at 10 kV and 16.2 mm working distance.

Sample preparation for SEM procedure

The fracture surface of the bone samples were processed for SEM protocol as

follows. Firstly, the bone pieces were chemically fixed in 3% gluteraldehyde . They were

washed in a series of three 10 min buffer washes (0.1 M sodium cacodylate). These were

post-fixed in a mixture of 1% osmium cacodylate in sodium cacodylate for one hour. The

samples were washed in Distilled water (2 changes of 10 minutes each). Then, the samples

were dehydrated through a series of ethanol solutions (50%, 70%, 90% and 100%) for 2

changes of 10 minutes each. The dried samples were labelled and mounted on the SEM

stubs (with the fracture surface up) and placed in a desiccator until gold coating time. The

samples were coated with gold using a sputter coater (BioRad SC500) (Figure 5.7).

Figure ‎5.7 Cortical bone samples mounted on the SEM stubs, place for gold coating


Tension

92

5.3 Experimental procedure and Results

5.3.1 Dumbbell shape tensile test samples with round junction versus those with

sharp junction

The size and geometry of the tensile test samples can affect the measurement of the

properties of bone specimens like the ultimate tensile strength156

. As mentioned earlier,

dumbbell shape specimens are better choice for tensile testing compared to the strip

specimens. However, they can themselves be in two different shapes: with either a round

junction or sharp junction between the grip and gage sections (Figure 5.2 & 5.4). To

compare the effect of these two shapes on fracture energy and ultimate tensile stress

measurement, preliminary tensile tests until fracture were conducted on two groups of

specimens prepared in the dumbbell shape with arc junction and sharp transition. The

force-displacement curves obtained from experiments were normalised to stress-strain

curves and the toughness and the strength of the samples were determined (Figure 5.8).

From the graph, it can be seen that the dumbbell shaped samples with a sharp junction

(DS) require less energy to fracture compared with dumbbell-shaped samples with a round

junction (DR). It could be related to stress concentration occurred in these samples.


Tension

93

Figure ‎5.8 Tensile Stress-Strain responses until failure of dumbbell shape samples with round junction (

) versus those of dumbbell shape samples with sharp junction ( )

Figure 5.9 compares the mean strength and toughness (fracture energy) between two

different-shaped samples.

strength

(MPa)

fractu

re e

nergy(N

.m)

0

50

100

150samples with sharp junction

samples with round junction

Figure ‎5.9 Comparison of the strength and toughness of dumbbell shaped samples with round junction and

those of samples with sharp junction

Although the mean toughness of dumbbell shaped samples with sharp junction is

considerably less than that of dumbbell-shaped samples with round junction, their mean

-20

0

20

40

60

80

100

120

140

160

-0.50 0.00 0.50 1.00 1.50

Stre

ss (

MP

a)

Strain

Dumbbell shape samples with round junction

Dumbbell shape samples with sharp junction


Tension

94

strength do not show significantly differences. This outcome in general, also agrees with

the results of the other researchers which, have recommended the utilization of the

dumbbell shaped samples with round junction for tensile testing due to their lower stress

concentration156

. Hence, the dumbbell shape samples with round junction were applied in

the rest of experiments.

5.3.2 Hysteresis energy absorption for PP-exposed samples versus the control samples

Hysteresis energy is a measure of energy absorption or dissipation as heat by bone

specimen during a tensile loading-unloading cycle. To study the variation in hysteresis

energy caused by 145 hours of stimulation of bone specimens by 500V and 10 KHz high

voltage pulses, a tensile loading-unloading(L-U) cycle until 40 N was conducted on PP-

exposed samples and the area enclosed in hysteresis loop were determined before and after

excitation.

Figure 5.10 and 5.11 compare the hysteresis loop of the representative PP-exposed

and the control samples respectively before and after 145 hours excitation.

Figure ‎5.10 Hysteresis loops in tensile loading-unloading cycle for a bone specimen exposed to pulse power

before and after 145 hours excitation

0

2

4

6

8

10

12

14

16

18

0.0% 10.0% 20.0%

Stre

ss (

MP

a)

Strain (%)

PP-exposed sample

After 145 h PP excitation

Before 145 h PP excitation


Tension

95

From Figure 5.10, it can be seen that the samples exposed to pulse power for 145

hours at 40 N loading sustain higher strain relative to before excitation. This results in a

greater area under the hysteresis loop in the L-U cycle for the sample which was exposed

to pulse power. It can show that the hysteresis energy after 145 hours excitation

significantly increased compared with that of before pulse power stimulation (more

than20% differences).

Figure ‎5.11 Hysteresis loops in the tensile loading-unloading cycle for a control bone sample before and

after 145 hours being in similar environmental condition as PP-exposed samples

In contrast for the control specimen (Figure 5.11), the strain at 40 N loading after 145

hours, was the same as initial measurement and the hysteresis energy also before and after

that period (145 hours being in similar situation as PP-exposed sample but unexposed to

pulse power), was not considerably different (less than 5% variation).

The graph bars in figure 5.12 compares the average hysteresis energy dissipated by

samples exposed to pulse power for 145 hours before and after excitation with those of the

control samples in the same period. It demonstrates that the samples exposed to pulse

power dissipated more energy after 145 hours excitation compared with before exposure,

while the mean hysteresis energy for control samples appeared to remain quite unchanged

after that period.

0

5

10

15

20

25

30

0.0% 2.0% 4.0% 6.0%

Stre

ss (

MP

a)

Strain (%)

control sample

after 145 hours

before 145 hours


Tension

96

PP-exp

osed s

ample

s

Contr

ol sam

ples

0

10

20

30Before 145h excitation

After 145h excitation

Hyste

rsis

en

erg

y (

N.m

)

Figure ‎5.12 Mean hysteresis energy of the control samples versus the samples exposed to pulse power before

and after 145 hours excitation

5.3.3 Tensile toughness and strength measurement

In this part, to consider the effect of timing and environmental condition in pulse

power stimulation of bone samples, they were divided into four groups of Control and PP-

exposed with 28 hours, 35 hours and 145hours pulse power excitation. The samples in the

latter group were exposed to pulse power continuously but for two other groups, the

excitation periods were not continuous (it was over 5 days with approximately 6 to 7 hours

excitation per day). Tensile testing up to failure was conducted on cortical bone specimens

for all groups. Figure 5.14 shows representative samples of tensile stress-strain curves from

four groups. Regardless of differences in general pattern of the graphs in different groups,

all stress-strain curves were divided into i) a very small linear elastic part in the beginning

(up to a strain of about 0.05), ii) relatively extensive nonlinear plastic region before

fracture and iii) catastrophic failure part. This behaviour exhibits the characteristics of


Tension

97

ductile material. Furthermore, the fracture processes in both the control and the PP-

exposed bone samples occurred relatively slowly, which confirmed the ductile fracture

characteristic. The comparison of the stress-strain graphs in all group samples shows that

pulse power stimulation did not affect the overall ductile behaviour of the bone samples.

From the graphs, it can also be observed that the general trend of stress-strain curves for

samples of each group appears to be identical. It will be discussed more in section 5.3.4.

Figure ‎5.13 Tensile stress-strain graphs of the cortical bone samples in four groups up to failure

Area under the stress-strain graphs until fracture point (fracture energy) and ultimate

tensile stress (strength) were determined for all samples. All data was expressed as means

± standard deviation.

The non-parametric tests are usually more conservative compared to the parametric

ones (i.e. compared to ANOVA and t-test) especially when the sample size is small and the

normality cannot be tested. Due to very small size samples in some groups, the normality

of the samples could not be checked and therefore the data was analysed by non-parametric

test (Kruskal-Wallis). There was a trend in the toughness and strength of the bone samples

-20

0

20

40

60

80

100

120

140

160

0 0.5 1 1.5 2

Stre

ss(M

Pa)

Strain

control samples (unexposed to pulse power)

after 35h PP excitation

after145h PP excitation

after 28 hours PP excitation


Tension

98

which shows no statistically significant differences between the control and different

timing PP-exposed specimens (P>0.05). Table 5.1 presents the values of the ultimate

tensile stress and the toughness for four-group samples.

Parameter

Control

samples

(unexposed to

pulse power)

28 hours pulse

power excitation

35 hours pulse

power excitation

145 hours pulse

power

excitation

P value

Strength

(MPa) 83.51±3.946 58.998±7.262 93.363±21.435 114.3±6.710 0.112(>0.05)

Fracture

energy(N.m) 77.99±6.626 40.916±28.725 40.502±7.4119 73.04±22.4 0.244(>0.05)

Table ‎5.1 Mean value ±standard deviation for the toughness and strength of the

tensile bone samples in four treated groups

5.3.4 Fractographic examination using SEM

The fracture surfaces of the cortical bone samples from both the control and the PP-

exposed groups, tested in tensile testing were categorised and analysed using fractographic

examination. Figure 5.14, 15 &16 show scanning electron micrographs from the top and

side views of the fracture surfaces of the control and PP-exposed specimens with their

appropriate tensile stress-strain graphs up to failure.

The overall rough and irregular appearance of the fracture surfaces, with steep

gradients in some cases, exhibits the ductile fracture behaviour in all samples. For

example, steep gradient surfaces were observed for both the control and the PP-exposed

samples. These results are generally, consistent with the stress-strain graphs obtained from

tensile testing that showed an extensive plastic region with high strain energy absorption

until failure. However, a great deal more specimens might help to find a significant trend

in differently treated groups.


Tension

99

As the samples were harvested from different sites of tibia, they appeared differently

in the porosity and density. The micrographs (e.g. Figure 5.14A1&2) show that, in the

porous portion, the calcified fibres are pulled out, creating a relatively brushy appearance

on the surface. Several depressions also appear in this portion that may be produced by the

fibre pullout process. On the contrary, the dense (fibreless) portion have shorter pull-out

calcified fibres with more little depressions, caused smoother fracture surface compared to

the porous portion.

Some samples (e.g. Figures 5.14A & D, 5.15B and 5.16A) had larger fibre pull-out

length at one edge compared to the opposite edge, yielded a steep gradient on the surface

which is characteristic of ductile shear fractures. In more rigid samples, the fibre pull-out

length appeared longer near the edge of the surface and became smaller in the interior

region (e.g. Figure 5.14C1&2). The comparison of the scanning electron micrographs of

the samples with their appropriate tensile strain-stress graphs reveals the equivalent trends

in the fracture process.

Figure 5.14A & B show two distinctive areas (one is rougher and more porous while

the other is smoother and more rigid) on the fracture surfaces which are matched with their

equivalent stress-strain graphs (the below of the same figure) demonstrating two separate

portions in the plastic regions before failure. The stress-strain graphs for sample A and B

have similar patterns with a moderate slope in the plastic region up to failure point

following a very small elastic portion. They show a small drop in the plastic area before

failure. From figure 5.14B & D, it can be seen that the more porous samples in addition to

having a steeper gradient on their fracture surface, require less fracture strain energy until


Tension

100

failure. This is very noticeable for sample D which has a small fracture toughness with a

very sharp slope fracture surface.

Figure ‎5.14 SEM micrographs from the top and side views of the control samples (unexposed to pulse

power) with their corresponding stress-strain graphs

0

20

40

60

80

100

-20% 0% 20% 40% 60% 80% 100% 120% 140%

Stre

ss (

MP

a)

Strain(%)

Control samples

A

B

C

D


Tension

101

Figure ‎5.15 SEM micrographs from top and side views of cortical bone samples exposed to 500Vand 10 kHz

pulse power for 145 hours with their corresponding stress-strain graphs

Figure 5.15 shows the samples that were exposed to pulse power for 145 hours with

their appropriate stress-strain graphs. Both graphs (with almost similar patterns) show that

after a very small, initial elastic linear region, there is a sharp rise, until the graphs reach a

maximum stress before catastrophic failure. The scanning electron micrographs show

fibreless and denser material with little depression on the surfaces. This provides a

relatively smoother appearance on the surface compared to the control samples.

Figure 5.16 presents the micrographs for two other groups of bone samples exposed

to pulse power over 28 and 35 hours with their equivalent stress-strain graphs.

0

20

40

60

80

100

120

140

0% 50% 100% 150%

Stre

ss (

MP

a)

Strain(%)

Samples exposed to pulse power for 145 hours

A

B


Tension

102

Figure ‎5.16 SEM micrographs from top and side views of the cortical bone samples exposed to pulse power,

A and B for28 hours, C and D for 35 hours with their equivalent stress-strain graphs

Again, some similarities were observed in the stress-strain graphs of the samples in

each group. For example, the graphs of the samples exposed to pulse power for 35 hours

(C & D) after the initial elastic part, rise moderately to failure point before catastrophic

fracture. In contrast, for two other samples which were under 28 hours pulse power

0

20

40

60

80

100

120

0% 50% 100% 150% 200%

Stre

ss(M

Pa)

Strain(%)

A 28h exposed to PP

B 28h exposed to PP

C 35h exposed to PP

D 35h exposed to PP


Tension

103

exposure, the graphs show two steps fracture in the last portion which is different from the

fracture processes of the other samples. Sample B showed a considerably different graph.

In this case, after the elastic region, the graph went up slightly and then extended

moderately down to less than 10% of ultimate stress. Its micrograph plane, from top view,

shows relatively dense feature with less fibber collagen pull-out on the surface. Sample A

absorbed smaller strain energy before failure with a steeper gradient on the fracture surface

compared to three other samples.

Figure 5.17 presents higher magnification micrograph of the fracture surface of a

representative specimen that was exposed to pulse power. These photos highlight the

dimples and microvoides created in the fibrous fracture surface of the sample through

failure process. Figure 5.17D shows crack bridging by the collagen fibrils. This crack

bridging has been illustrated to toughen the bone by reducing the stress magnitude imposed

on the tip of cracks in longitudinal directions. Such bridges sustain the load that could

spread the cracks 157, 158

. Figure 5.17B presents micro-cracks and uncracked ligament

bridges on fracture surface of the treated sample. These mechanisms which generated

through fracture procedure and acted as proposed mechanisms involved in the bone

toughness and its resistance to crack growth, were presented in both the control and PP-

exposed samples 158

.


Tension

104

Figure ‎5.17 Details of scanning electron micrographs of fracture surface in higher magnification (A)

Dimpled, irregular appearance of fracture surface (B) Microcrack diffusion (C) Microvoids (D) Crack

bridging by collagen fibrils

Figure 5.18 compares the higher magnification micrographs of the fracture surfaces

of the representative samples from three different treated groups with the normal sample.

From the photos, it can be seen that there is not significant variation in microstructure of

the samples exposed to pulse power at different periods (including their porosity and

diffusion of microcracks on their surfaces) due to excitation compared with those

morphological characteristics of the control sample.

Microcracks

Microvoids

Crack bridging

Uncracked

ligament

bridges


Tension

105

Figure ‎5.18 Higher magnification of scanning micrographs of the fracture surfaces of the representative

samples from each group (A) Control sample (B) Samples exposed to pulse power for 28 hours (C) Sample

exposed to pulse power for 35 hours (D) Sample exposed to pulse power for 145 hours

5.4 Discussion

This chapter investigates the effect of applying high voltage, high frequency pulsed

electric field (with capacitive coupling method) on fracture behaviour of cortical bone

through quantitative and qualitative analysis using tensile test, which is a usual standard

method in fracture studies. The quantitative analysis involved comparing the cortical bone

strength and toughness (two crucial properties for functional behaviour of bone particularly

for determination of the fracture risk) with and without pulse power excitation using


Tension

106

tension test until failure. Additionally, the amount of energy dissipated by bone specimens

through a tensile loading–unloading cycle was evaluated before and after pulse power

exposure. The result presented here, demonstrates that the bone toughness and strength

appeared to remain unchanged after applying pulse power signals. Nevertheless, the

amount of hysteresis energy during tensile loading-unloading cycle shows quite increase

after pulse power excitation compared to the control samples.

Both toughness and strength of bone tissue, which are two intrinsic mechanisms to

limit microstructural damage through fracture process 154

, are primarily attributed to the

bone material’s inherent resistance to microstructural fracture and therefore are related to

relative amount and properties of minerals (hydroxyapatite) and the collagen matrix inside

the bone 43

. Hence, the results can suggest that pulse power excitation with particular

parameters (500v and 10 kHz) did not influence the bone material properties in tension.

Beyond the amount and the properties of the bone material, their arrangement in the

space comprising the structural and microstructural properties of cortical bone are

important factors in mechanical competence of bone such as its rigidity, strength and

stiffness. Although bulk structural properties of cortical bone including thickness of the

cortex, cortical cross-sectional area and area moment of inertia are most commonly applied

features for determination of its mechanical competence, microstructural properties such as

cortical porosity, crystallinity or the presence of microcracks also certainly involves in

bone’s mechanical competence. For example, microcracks are mentioned as a particular,

effective mechanism for energy dissipation which also weaken the cortical bone tissue30

.

Therefore, investigation of microstructure of bone tissue after pulse power stimulation and

compared with that of the samples unexposed to pulse power can provide detailed

information about effect of pulse power on bone’s structure and its resistance to fracture.


Tension

107

In the qualitative analysis, the fracture surface of broken bone samples were analysed

using scanning electron microscopy. The irregular, fibrous appearance of the failure

surfaces of both normal and treated samples, illustrates the ductile behaviour of bone

samples in tensile loading. Nevertheless, in general, the apparent feature of the samples

exposed to pulse power for 145 hours showed little pores and fibreless surface compared

with other samples.

The micrographs were also matched with their corresponding stress-strain graphs up

to failure. A relatively extensive plastic region for most of the samples confirmed their

ductile behaviours. Juxtaposing these curves and their micrographs reveal similar pattern in

fracture process for samples of the same treated groups. In addition, although, relatively

identical patterns were observed between stress-strain graphs of cortical bone samples

from the same group, a great deal more specimens would be required to find a significant

trend in differently treated samples. It can confirm that pulse power stimulation of bone

with nominated parameters appeared to be safe with no destructive effect on bone

structure.

Higher magnification micrographs of the fracture surfaces showed crack bridging (by

unbroken collagen fibrils) and defusing of microcracks (around the large cracks) in both

control and PP-exposed samples. They are mentioned as the main extrinsic mechanisms

involved in bone toughening against crack propagation through fracture process157, 158

.

However, comparison of the porosity and microcracks distribution showed no significant

differences between both normal and treated samples which can suggest that pulse power

did not influence microstructure of cortical bone tissue.

Comparison of micrographs of fracture surfaces with the corresponding stress-strain

curves generally showed that samples which required less fracture energy before failure,


Tension

108

have a steeper gradient fracture surface with more holes and dimples on the surface (e.g.

Figure 5.16A & 5.14D). In contrast, the rigid samples (for example samples exposed to

pulse power for 145 hours), showed smoother face with little dimples on the top surface

accompanied with a relatively steeper stress-strain curve in their plastic region prior to

catastrophic failure (Figure 5.15A & B). Nevertheless, because both control and exposed

samples were including both types of graphs, it cannot be concluded that this difference is

definitely because of the effect of pulse power stimulation on bone microstructure rather it

could be due to differences between location and anatomical sites where specimens

obtained from.

Few factors involved in experimental setup that could be the sources of errors and

influence the results of tensile testing as a detective method such as: slippage of the sample

between grips, placement of the specimen on the testing machine (so that it was completely

straight to be imposed to pure tensile loading without shear stresses) and the effect of the

testing machine compliance. Also, the geometry of tensile test sample can influence the

measurement of the sample properties (for example cause underestimate of toughness).

Hence, it was recommended to use dumbbell shape samples with round junction to reduce

stress concentration and testing error.

In microscopic level analysis, application of another method such as Transmission

Electron Microscopy (TEM) which provide higher resolution photos from interior structure

of the materials could also be more appropriate method to show possible changes in

microstructure of cortical bone samples (such as its porosity, microcracks diffusion and

crystallinity).

Furthermore, here pulse power was applied in radial and tensile loading in

longitudinal directions. Consideration of the other directions for applying pulse power and


Tension

109

tensile loading can provide more complete assessment from the effect of high voltage, high

frequency pulses on bone tissue. Application of samples with smaller dimension can

increase the possibility of the effect of pulse power on bone tissue. Next chapter, therefore,

will consider the variation in the fracture behaviour of smaller size specimens due to pulse

power stimulation using compression testing until failure.

Chapter 6: Effect of Pulse Power Excitation on

Basic Mechanical Properties of

Cortical Bone in Compression

Chapter 6: Effect of Pulse power on Basic Mechanical Properties of Cortical Bone in

Compression

111

6.1 Introduction

In addition to tensile and bending, compression is another normal loading mode that

imposes on bones during daily life for example in locomotion. It in particular occurs in

some regions of skeleton like the vertebrae. Although compressive fracture appears to be

less important in life, it can occur in the vertebrate and diaphysial regions of long bone in

particular as a result of fatigue 41

. On the other hand, bone shows different mechanical

behaviour (e.g. stiffness and strength) in response to divergent types of loading. Hence, the

supplementary study of the bone behaviour in compression especially after applying

exogenous stimuli e.g. pulse power exposure can provide useful information regarding to

possible effect of this kind of stimulation on the bone quality and its assurance application

on bone tissue.

Achievement of accurate results using compressive test is noticeably more difficult

compared with tensile tests due to friction and end effect imposed on the samples through

the tests. However, the compressive test has significant advantages compared with tensile

tests. Firstly, it allows the use of relatively smaller samples (with dimension in some

millimetres). This advantage is particularly desired to increase the concentration effect of

pulsed electromagnetic field on bone construction. Secondly, the preparation of the

compressive samples is not as complicated as tensile test specimens. Nevertheless, the

interface surfaces of the bone specimen with platen need to be relatively flat and parallel.

Otherwise, stress concentration may occur in high spots, which cause some measurement

errors like underestimation of the compressive strength 140, 148, 159

. Despite to the usual

measurement errors in compressive testing, it is very precise particularly for assessing the

effect of a kind of treatment or an external stimulus like pulse power stimulation which,


Compression

112

requires comparison of data from the experimental and the control groups (assuming the

measurement errors did not change as a result of treatment) 148

.

Following the previous chapter that investigated the effect of pulse power excitation

on the fracture behaviour of cortical bone using tensile loading, this chapter compares

compressive fracture toughness and strength between samples exposed to pulse power and

the control specimens in order to investigate whether or not pulse power excitation can

influence the bone tissue quality. On the other hand, to explore how pulse power exposure

to small-size bone samples for a particular period will influence the bone material quality.

Hence, compressive test until fracture occurred, were performed on the control specimens

and the samples exposed to pulse power for 66 hours and their compressive strength and

toughness were compared.



Fresh sheep tibia was obtained from a slain ovine within 24 hours of slaughter. The

surrounding soft tissue was removed from the bone and tibia was wrapped in a 0.15M

physiologic saline soaked cloth and stored at -20 until required for testing. Prior to

sample treatment, the tibia was thawed for one hour. Parallel-side cubic specimens were

prepared from cortical dyaphysis of the ovine tibia (Figure 6.1). As mentioned previously,

creating flat, parallel surfaces at two end sides of the samples, is crucial for accurate

compressive testing. Hence, the cubic specimens were cut using a linear high precision saw

(Isomet 5000) while the bone was kept moist and their physical dimensions were then

measured using digital caliper. The total 10 samples were applied in this experiment for

both the control and PP-exposed samples.


Compression

113

Figure ‎6.1 Position and directions of the rectangular specimen obtained from the tibial cortical dyaphysis

The specimens were divided randomly and labelled as PP-exposed samples, which

were exposed to pulse power and the control specimens which were kept in the same

environmental conditions (room temperature and humidity) as PP-exposed samples, but

without pulse power excitation. All samples were then refrigerated at 4 in 0.15M

physiological saline solution for immediate experimentation and stored at -20 for later

testing.

6.2.2 Experimental Procedure

The general process applied to investigate the effect of pulse power excitation on

compressive fracture toughness and strength of cortical bone was including high voltage,

high frequency pulses exposure to PP-exposed samples and determination and comparison

of the fracture toughness and strength of the cortical bone specimens in both the PP-

exposed and the control groups using compressive testing until failure occurred.

Bone samples stimulation with pulse power signals

The pulse power signals (voltage pulses up to 500 V at 10 kHz), generated by pulse

generator B (PGB), were delivered through two wire leads attached to two series of metal


Compression

114

screws. Application of millimetre-sized samples and screws with small contact cross-

section will increase the electric field intensity applied on bone specimens. As described

earlier in chapter 3, since the direct connection of screws and cables with bone provides

very low impedance, significant current can pass through the bone and makes it dry and

causes it to burn. Thus, the screws have been covered by electrical isolation tape in order to

change the characteristics of the bone from a resistive load to a capacitive load. Therefore,

the pulsed electric field has been applied to the bone samples through capacitive coupling

method reducing the thermal effect while increasing the influence of the electric field on

the bone structure.

The small cortical bone specimens in the PP-exposed group were placed in radial direction

between isolated screws for stimulation. Figure 6.2 presents a sketch of the experimental

set-up for pulse power stimulation of the small-sized bone samples. The samples were

exposed to the high power, high frequency pulsed electric field for 66 hours continuously.

The applied pulse power waveform was shown in figure 3.11. All through this period, the

control specimens were placed in similar environmental condition as the PP-exposed

samples with no pulse power exposure. Both control and treated samples were kept moist

with 0.15M physiological saline during the experiment.


Compression

115

Figure ‎6.2 Sketch of experimental set-up for pulse power stimulation of millimetre-sized cortical bone

samples

Compressive testing

After 66 hours pulse power excitation, both control and PP-exposed samples were

imposed to compressive loading until fracture. Hence, the millimetre-sized samples were

placed on a flat platen attached to an Instron testing machine (model 5944, 2kN load cell).

The compressive tests were performed in displacement control at the extension rate of 0.1

mm/min until complete failure occurred and the load was measured from the load cell. The

results were recorded as load and displacement data and converted to stress and strain data

for farther analysis. For load cell safety, the upper actuator was stopped before touching

the downer platen (at 1 mm distance). Figure 6.3 shows the compressive testing set up

while the bone specimen was placed on the flat platen.


Compression

116

Figure ‎6.3 Compressive testing of cortical bone specimen

6.3 Toughness and strength measurement (results)

Figure 6.4 presents the stress-strain graphs obtained from compression testing on

both control specimens and the samples exposed to high power pulsed electromagnetic

field for 66 hours. The area under the stress-strain curves and the ultimate stress present the

total fracture work and the compressive strength respectively. The elastic modulus of the

specimens were also determined from the slope of elastic linear portion of stress-strain

graphs.


Compression

117

Figure ‎6.4 Compressive stress-strain responses for the control specimens ( ) verse those for the

samples exposed to pulse power ( )

As illustrated in figure 6.4, although the pattern of stress-strain curves of the samples

in both groups were similar, the samples loaded in compression showed different stress-

strain curves compared with the graphs obtained from tension test in previous chapter.

Regardless of one exceptional graph observed in each group with different pattern, the

stress-strain curves of the compression tests showed a linear region until the sample

completely breaks without any distinctive plastic deformation, which was previously

observed in tension graphs (Figure 5.13). This difference can confirm the different

behaviour of cortical bone in response to different loading mode.

From area underneath the stress-strain curves shown in figure 6.3, it can be seen that,

the samples that were exposed to pulse power for 66 hours, absorbed larger amount of

strain energy until the complete fracture occurred. Additionally, the PP-exposed samples

appeared to show higher strength (larger ultimate stress) compared with the control

samples.

0

50

100

150

200

250

0.00% 2.00% 4.00% 6.00% 8.00%

Stre

ss (

MP

a)

Strain(%)

control sample

sample exposed to pulse power


Compression

118

The total fracture work and the strength and the stiffness were determined and

compared between the control and PP-exposed samples. Figure 6.5 and 6.6 illustrate the

average total fracture energy and strength of the control specimens and those of the

samples exposed to high-voltage pulsed electromagnetic field respectively.

Control s

maple

s

PP-exposed s

ample

s

0

1

2

3

4

5

To

tal f

ract

ure

en

erg

y (N

.m)

Figure ‎6.5 The total strain fracture energy of the samples exposed to 500V, 10 KHz electromagnetic field

compared to that of the control samples

Contro

l sam

ples

PP-exposed s

ample

s0

50

100

150

200

Com

pres

sive

Str

engt

h (M

Pa)

Figure ‎6.6 The strength of the samples exposed to 500V, 10 KHz electromagnetic field compared to that of

the control samples


Compression

119

The graph bars demonstrates that the PP-exposed samples appeared to require larger

amount of energy in order to fracture and showed also more compressive strength

compared with the control samples.

Figure 6.6 compared the average Young’s modulus of the samples exposed to pulse

power with that of control samples. From the graph bars, it can be seen that the mean

stiffness of PP-exposed specimens after 66 hours exposure remained unchanged compared

with that of the control samples.

PP-exposed

sam

ples

Contr

ol sam

ples

0

10

20

30

Yo

un

g's

mo

du

lus(G

Pa)

Figure ‎6.7 Comparison of the stiffness of the samples exposed to pulse power with that of the control

samples

As stated in chapter 2, the cortical bone toughness predominantly pertained to the

integrity of the collagen matrix while its stiffness and rigidity is strongly associated with

mineral content 27, 40

. Hence, the results presented in this chapter can suggest that high

voltage, high frequency pulsed electromagnetic field exposure may have altered the


Compression

120

orientation or the quality of collagen fibrils whereas the mineral constitutes and mineral

density were not affected due to this stimulation.

6.4 Discussion

Following two different loading patterns presented in previous chapters, to provide

more comprehensive investigation regarding to the assessment of the basic functional

properties of cortical bone samples in response to pulse power excitation, this chapter

evaluates the variation in compressive strength and toughness of cortical bone samples due

to pulse power exposure. This test has the advantage of applying small-sized samples

which can result in the increase in the probable effect of electromagnetic field intensity

over the samples. Furthermore, preparation of the compressive samples is less complicated

compared to the tensile specimens.

The results presented in this chapter, demonstrate the positive effect of pulse power

excitation via two series of isolated screws (through capacitive coupling method) on the

compressive strength and the total failure strain energy absorption of the cortical bone

samples. They show that the ultimate compressive stress and the total fracture energy of

the cortical samples increased after 66 hours pulse power stimulation. As stated previously,

the total strain failure energy that is measured to be the area under the stress-strain curve

until complete fracture is an indication of sample toughness. Therefore, the findings

suggest that the cortical bone tissue became tougher and stronger due to high voltage, high

frequency pulsed electromagnetic field exposure.

In general, these results show that the capacitive coupling pulse power exposure with

nominated parameters accompanying with continuous hydration of the bone samples

appeared to be safe and controlled with no athermal or destructive effect.


Compression

121

They also confirm the differences in fundamental characteristics of the cortical bone

in different loading mode in response to pulse power stimulation. Additionally, these

findings show that the increase in the pulsed electromagnetic field intensity using

electrodes with small cross section (applying screws) and millimetre-sized samples can

enhance the possible effect of pulse power excitation on cortical bone tissue resulting in

augmentation of bone strength and toughness due to pulse power excitation.

The toughness and the strength of the cortical bone tissue are directly associated to

the quality and integrity of the collagen matrix while its stiffness is primarily related to

bone mineral content 40, 47, 50

. On the other hand, it was illustrated that the electrical field

can align the collagen fibrils46

. Hence, although the mechanism by which pulse power

stimulation has increased the strength and toughness of the cortical bone samples is not

fully clear, it is proposed that exposure of millimetre-sized cortical bone samples to high

voltage, high frequency pulsed electromagnetic field may have positively altered the

orientation and the quality of collagen fibrils in extracellular matrix. Nevertheless, it did

not affect the bone mineral phase.

Chapter 7: Evaluation of Cortical Bone

Elasticity in Response to Pulse

Power Excitation Using Ultrasonic

Technique

Chapter 7:Evaluation of Cortical Bone Elasticity in Response to Pulse Power Excitation

Using Ultrasonic Technique

123

7.1 Introduction

Bone needs to be strong and stiff enough to play its important roles as a supportive

and protective frame for other organs and tissues in the body. Therefore, there is always a

crucial interest in obtaining information about bone strength and stiffness particularly in

detection of bone diseases and investigation of the effect of an external stimulus. The

anisotropic and inhomogeneous structure of bone cause some problems in determining its

functional properties using conventional mechanical testing. Viscoelasticity in bone (strain

rate dependency) and environmental conditions (like temperature and bone hydration) are

other factors that can influence the outcomes especially when mechanical tests are required

to continue over a long period. An alternative, non-destructive method is ultrasonic bone

measurement that can present direct information about the elastic properties of bone and

can predict whole bone strength160

. By preparing small parallel-sided specimens, ultrasonic

technique provides several anisotropic property measurements of a single bone specimen.

Additionally, it can use smaller, less complicated bone samples compared to conventional

mechanical testing methods161

. The other important advantages of ultrasonic measurement

are the possibility of its application several times with no, destructive effect on bone

structure and its ability to produce more accurate results. Hence, it can be applied even for

in vivo studies. Table 7.1 reviews some of the advantages of ultrasonic technique over

conventional mechanical testing methods161, 162

.

In this work, an ultrasonic velocity measurement was conducted to determine the

possible changes in elastic properties of the cortical bone due to pulse power excitation.

Running the procedure in water, prevents the bone from dehydrating during the test and

can control the effects of the environmental conditions on bone properties. In addition, this



124

method used small samples increasing the possible influence of the pulsed electric field on

the bone material structure.

Parameter Bending Compressive Tensile Ultrasonic technique

Specimen shape Rectangular

parallelepiped

Length-cross section

ratio is critical.

Right cylinders or

cubes. Parallel faces

are critical.

Difficult to machine

specialized shapes

for mounting.

Cylinders or

parallelepiped. Parallel

faces not necessarily

critical. Less complicated

shape.

Anisotropic

elastic properties

Three orthogonal

specimens for three

moduli.

Determination of

Poisson’s ratio in pure

bending tests. This

method requires

specimens with

relatively large cross

sections.

Three orthogonal

moduli from cube.

May be possible to

measure Poisson’s

ratio with

extensometer , but

as yet no reports of

Poisson’s ratio

measured in this

way.

Three orthogonal

specimens for three

moduli. Shear

moduli possible if

cross section is

round. Poisson’s

ratio possible with

bi-axial

extensometer.

Three moduli, three shear

moduli, six Poisson’s

ratios possible from a

cube as small as several

millimetres dimension.

Notes For determination of

elastic constants,

several series of

specimens with

different h/l ratio are

necessary.

Inaccuracies occur

due to specimen

misalignment, friction

at the load points,

imprecise strain

measurement,

inadequate h/l ratio,

and elastic-plastic

deformation. The

limitations imposed

by theoretical

considerations must

be taken into account.

This technique can

be accurate if faces

are parallel and if

strain is measured

with an

extensometer

instead of platen

motion.

Compressive testing

is less common for

engineering

materials, but

ASTM standards

have been written

and are used for

rigid plastics. The

ASTM suggested

specimen size is 12

mm by 12mm by

50mm, not cubic

specimens.

If induced bending

is accounted for,

and if strain is

measured with an

extensometer this

technique can be

accurate. Tensile

testing is most

common method of

measuring elasticity

of engineering

materials.

Actual path length is

unknown unless specimen

shape is simple. Path

length is determined by

averaging the actual

lengths. The velocity of

propagation of an

ultrasonic wave can be

dependent on the

frequency of oscillation.

Pure longitudinal and

shear waves propagates

only in directions parallel

to axis of material

symmetry.

Table ‎7.1 Comparison between the conventional mechanical tastings and the ultrasonic technique161, 162

The porosity in cortical bone is low and the pore size is normally smaller than the

ultrasound wavelength. Therefore, the ultrasonic technique provides a straightforward

relationship to deduce the elastic properties of cortical bone. To measure ultrasound

velocity in the cortical bone, relatively high frequency ultrasound waves ranging from 2-10

MHz have been utilized in different studies. These ranges of frequencies provide a

comparatively accurate elastic property measurement in very small samples162

.



125

These all advantages of the ultrasonic technique compared with mechanical testing

make it a more practical option to evaluate the effect of an external stimulus like pulse

power on bone elasticity. This chapter demonstrated the results of comparing Young’s

modulus of cortical bone samples in the control and PP-exposed groups before and after

pulse power excitation using 5-MHz ultrasound waves.

7.2 The theoretical consideration

According to the theory of small amplitude elastic wave propagation in anisotropic

solids 163, 164

, the rate at which the shear or longitudinal waves travel through the solid

matters is dependent upon its elastic properties and density. Figure 7.1 shows two kinds of

ultrasound wave propagation in bone specimen. A longitudinal wave is generated when the

transmitter vibrates in the same direction as wave propagation. If the transmitter vibrates in

a perpendicular direction to the wave propagation, shear waves are produced.

Figure ‎7.1 Ultrasound wave propagation in a bone specimen142



126

Both longitudinal and shear waves can propagate in two modes inside the bone based

on specimen geometry and wavelength of the waves (velocity/frequency). If the cross-

sectional dimension of the specimen is greater than the ultrasound wavelength, the wave

does not reach the sample boundaries. It is referred to as bulk wave propagation. The

second case, where the characteristic specimen dimensions are smaller than the

wavelength, is called bar wave propagation. In this case, the ultrasound wave propagates as

a complex bar wave, consisting of both shear and longitudinal waves and the entire

specimen cross section is excited by the passing wave142, 162

.

For bulk wave propagation, velocity is given by165

:

(7.1)

Where K is bulk modulus and G is shear modulus which for isotropic material are

defined by Young’s modulus (E) and Poisson’s ratio (ν) as:

(7.2)

(7.3)

For bar wave propagation, the velocity can be defined directly by the Young’s

modulus and the density given as162, 165

:

(7.4)

Where v is velocity, E is young’s modulus and ρ is density.

Therefore, if the density of bone samples and the ultrasound velocity are specified,

the young’s modulus is determined as:

(7.5)



127

For the ultrasound wave velocity determination, the time in which the wave pass

through the specimen is measured by the substitution method. In this method, the

difference in ultrasound transit time with and without a sample in the position gives the

time delay.



After initial processing on the sheep fresh tibia, parallel-side cubic specimens were

prepared from cortical dyaphysis of the ovine tibia (Similar to the test samples for

compressive testing in previous chapter, Figure 6.1). Producing parallel surfaces, is crucial

for accurate determination of the ultrasound velocity and the bone elasticity. Therefore,

cutting was conducted with a linear high precision saw (Isomet 5000) while the bone was

kept moist.

The physical dimensions of bone samples with consideration of their orientation in

respect to the bone axis were measured. Table 7.2 provides the average size of 10

specimens prepared for this work.

Table ‎7.2 Mean values ± standard deviation for the specimens’ dimensions

The cubic specimens were labelled according to their site and segregated randomly

into two groups of PP-exposed samples which were exposed to high power, high frequency

pulses (pulse power) and the control specimens which were kept in the same

Direction Mean (mm) Standard

Deviation(S.D)

Longitudinal(L) 10.3 0.05

Tangential(T) 3.21 0.16

Radial(R) 1.91 0.96



128

environmental conditions (room temperature and humidity) as PP-exposed samples, but

without pulse power excitation. All samples were then refrigerated at 4 in PBS solution

for immediate experimentation and stored at -20 for later testing.

7.3.2 Density measurement

There is a direct positive correlation between bone density and its strength and

stiffness166

. To calculate Young’s modulus from ultrasonic technique, it is necessary to

measure cortical bone specimen density. For cortical bone, the material density can be

measured by the wet weight divided by the specimen volume, which is the function of both

porosity and mineral content of the bone. Because there is no marrow space in cortical

bone, its apparent density is the same as its material density167

.

There are also some non-invasive methods for determination of bone material density

such as quantitative computed tomography(QCT), dual-energy X-ray, micro-CT, Magnetic

resonance imaging (MRI)166, 167

. For example, true volumetric density of cortical bone

samples can be derived via micro-CT utilizing Scanco μCT40 scanner. The calibration

phantom was performed to convert Hounsfield numbers into volumetric density. It has

been suggested (particularly for in vivo studies) that the BMD (Bone mineral density)

obtained from micro-CT data can be substituted into equation 7.5 and combined with

ultrasound velocity to find bone stiffness165

.

(7.6)

In this study, the cortical bone density was measured using: 1) The conventional

method (wet weight/specimen volume) and 2) Micro-CT before and after pulse power

excitation. Micro-CT data provides bone mineral density (BMD) of the specimens. For

first method, tissue mass was obtained using a precise scale and the volume was calculated

from the physical dimensions of the specimens. No significant variation (using two-tail



129

paired t-test) was found in cortical bone density measurement from both methods due to

pulse power stimulation. Table 7.3 presents the mean value (MV) ± standard deviations

(SD) obtained from the two methods. The density obtained from micro-CT was used in all

calculations.

Density measurement

method Before PP excitation

After PP

excitation P value

Micro-CT (g/Cm3) 1.148±0.049

1.165±0.06

0.33(>0.05)

Weight (g/Cm3) 2.030±0.019 2.036±0.03 0.93(>0.05)

Table ‎7.3 Mean density ± standard deviation for cortical bone specimens before and after pulse power

excitation

7.3.3 Experimental Procedure

The general process applied in this study to investigate the effect of pulse power excitation

on the elastic property of cortical bone consisted of two main procedures: firstly,

determination of Young’s modulus of cortical bone specimens using ultrasound velocity

measurement pre and post pulse power exposure for both PP-exposed and the control

samples and secondly, applying high voltage, high frequency pulses on the bone samples

in PP-exposed group.

Ultrasound velocity measurement

High precision measurement of ultrasound velocity were conducted using a high-frequency

pulser-receiver (Panametrics PR5800), water tank containing two matched 5MHz, 12.5

mm diameter ultrasound transducers; one acting as transmitter and the other as receiver.

They were highly damped to provide short pulses (Figure 7.2) and a 100 MHz PC-housed

digitisation card (NI PCI5122)



130

Figure ‎7.2 Ultrasound velocity measurement set up inside water tank

The water tank was filled with warm water to above the face of the upper transmitting

transducer and the water temperature was measured and recorded. Existence of any air

bubbles on the faces of both transducers was checked regularly and if present, wiped away.

The cables were connected between computer and pulser-receiver in their appropriate

locations and the initial setting on the pulser-reciever was carried out. Ultrasound waves

produced by the transducers were monitored and recorded in “Labview Signal Express”

software. Before main test on the bone samples, to check and calibrate the ultrasound

signals, an initial testing on a recognized specimen (Perspex) was performed and the

required setting was applied.

The “substitution” method was applied to calculate the ultrasound velocity. In this method,

the difference in ultrasound transit time with and without, a sample in position was

measured and recorded. For this purpose, one of the cursors (solid or dashed cursor) was

fixed on the initial peak of the ultrasound wave running before placing the sample in the

water. The cortical bone specimen whose density and dimensions were measured

previously, was then placed on top of the downer transducer. The second cursor was placed

on the initial peak of the new ultrasound wave while it passed through the sample. The

difference between the two cursors gave the difference transit time of the ultrasound wave



131

through the sample (Figure 7.3). The water temperature (T ), measured transition time

(dt) and sample thickness(D) in each direction were applied to determine the ultrasound

velocity in water (Vo) and through the sample (Vs) as 168

:

Vo = 1405.03 + 4.624T – 0.0383T2 (7.7)

Vs=

(7.8)

Ultrasound velocity was measured 5 times in longitudinal, tangential and radial directions

of samples before and after pulse power excitation. The average of the measurements was

used for calculation. To find the elastic properties from ultrasound velocity, as the lateral

dimension of the cortical bone samples in this study were small (compared with ultrasound

wavelength), this study assumed that the bar wave was propagated through the sample and

therefore the straightforward equation 7.5 was used to calculate Young’s modulus of the

bone samples.

Figure ‎7.3 Ultrasound wave propagation trough the sample and time delay measurement on Lab view Signal

Express



132

Pulse Power excitation

Similar to compressive testing (previous chapter), voltage pulses at 500V and 10 kHz

generating by the second pulse power generator (described in chapter 3) were exposed to

millimetre-sized bone samples via two series of parallel metal screws. As mentioned

earlier, to prevent any thermal or electrochemical effect due to direct contact of electrodes

and bone samples, the screws were covered with electrical isolation tape. Therefore, the

pulsed electric field has been applied to the bone samples through capacitive coupling

method and in this case, the thermal effect was reduced while the electric field effect on

bone structure was increased. Figure 3.11 showed the waveform of high voltage pulses

applied on PP-exposed samples.

After the first stage of the ultrasound velocity measurement, small cortical bone specimens

in the PP-exposed group were placed in a radial direction between isolated screws for

stimulation (similar to compressive samples). They were exposed to a high power, high

frequency pulsed electric field for 144 hours continuously in a set up similar to previous

chapter (Figure 6.1). To consider any possible effect of the environmental condition on the

results, the specimens from the control group were placed in a similar environmental

situation as the PP-exposed group but they were not exposed to the pulse power signals.

The test specimens were kept moist during the experiment, with 0.15M physiological

saline preventing them from dehydration. Then, the bone specimen density and the

ultrasound velocity were measured again for both the control and PP-exposed samples and

their elastic properties were calculated using equation (7.5).



133

7.4 Results

The ultrasound velocity was measured in three main orthogonal directions of cortical

bone cubic samples namely 1) longitudinal, 2) radial and 3) tangential. Using bone mineral

density obtained from the microCT data in the equation 7.5, Young’s modulus of cortical

bone specimens for both the PP-exposed and the control samples were calculated. All data

was expressed as means value ± standard deviation and were analysed by non-parametric

test to be more conservative due to small size samples in some groups and inability to

check the normality of the samples distribution. The two-tail paired Wilcoxon signed rank

test (a non-parametric paired test) compared the ultrasound velocity and Young’s modulus

variation in each group (control and PP-exposed) before and after pulse power excitation.

All differences were considered significant at the value P<0.05(95% confidence). Table 7.4

and 7.5 present the values of ultrasound velocities and Young’s modulus of the cortical

bone samples in PP-exposed and control groups before and after PP excitation (control

group was not exposed to PP) respectively. The mean ultrasound velocity passing through

the samples did not change significantly in both the control and the PP-exposed group after

pulse power excitation compared with the initial measurement (P>0.05). In addition, no

significant variation in elastic properties of the cortical bone specimens of both groups was

found, after application of high power pulses compared to those before stimulation.



134

Table ‎7.4 Mean value± standard deviation for ultrasound velocity and Young’s modulus of PP-exposed

samples before and after pulse power excitation in longitudinal, radial and tangential directions respectively

However, Young’s modulus of cortical bone in longitudinal direction was

significantly greater than that of two other crosswise directions (P<0.05). This is in

consistent with other appropriate reported researches.

Parameter Control group before

excitation

Control group after

excitation

P value

V1 (m/s) 4158±29.88 4191±152.1 0.3271

V2 (m/s) 3491±322.9 3713±24.66 0.4346

V3 (m/s) 3757±145.8 3917±103.4 0.2492

E1 (GPa) 19.96±1.383 21.10±0.1579 0.3664

E2 (GPa) 14.05±1.825 15.79±1.443 0.4011

E3 (GPa) 16.91±1.536 18.52±1.137 0.3557

Table ‎7.5 Mean value± standard deviation for ultrasound velocity and Young’s modulus of control samples

before and after pulse power excitation period in longitudinal, radial and tangential directions respectively

Young’s modulus of the cortical bone samples and the ultrasound velocities in the

control group compared with those of the PP-exposed group were summarized in table 7.6.

Parameter PP-exposed samples

before excitation

PP-exposed samples

after excitation

P value

V1 (m/s) 4032± 355.6 4562± 137.1 0.1772

V2 (m/s) 3774± 391.3 3810± 111.7 0.8644

V3 (m/s) 3937± 153.9 4191± 303.7 0.7545

E1 (GPa) 18.63±2.656 22.17± 1.476 0.1127

E2 (GPa) 16.36± 2.873 17.06± 1.079 0.6938

E3 (GPa) 17.72± 1.519 20.75± 3.431 0.5041



135

Table ‎7.6 Mean value ± standard deviation of ultrasound velocity and Young's modulus in PP-exposed

groups after pulse power excitation compared with those of the control group in the same time

The P values greater than 0.05 can show that the ultrasound velocity and Young’s modulus

of the cortical samples which were exposed to high voltage, high frequency pulses were

not statistically different from that of the control samples.

7.5 Discussion

This chapter investigated the effect of a high power, high frequency pulsed

electromagnetic field with 500V at 10 KHz frequency on the cortical bone material

elasticity using an ultrasonic technique. Performing the experiments in two parallel groups,

with and without pulse power application, but in a similarly controlled environmental

condition, is likely to omit the possible influence of the other issues (e.g. environmental

condition) on the bone material elasticity. There appeared to be no statistically significant

changes in ultrasound velocity passing through the samples and bone density in both

groups before and after pulse power excitation. The comparison of the elastic properties of

millimetre-size cortical bone samples in control and PP-exposed groups also confirmed

that application of high-voltage pulses with specified parameters in the period of 144 hours

did not affect significantly the elastic property of cortical bone tissue. This result can

Parameter PP-exposed group Control group P value

V1 (m/s) 4562± 137.1 4423± 23.90 0.2498

V2 (m/s) 3810± 111.7 3713± 24.66 0.3147

V3 (m/s) 4191± 303.7 3917± 103.4 0.3033

E1 (GPa) 22.17± 1.476 21.10± 0.158 0.3883

E2 (GPa) 17.06± 1.079 15.79± 1.443 0.279

E3 (GPa) 20.75± 3.431 18.52± 1.137 0.4426



136

suggest that high voltage, high frequency pulsed electromagnetic field exposure did not

affect the mineral phase structure in bone tissue which are the predominant factor affecting

bone stiffness.

Inhomogeneity and anisotropy of bone tissue have always been challenging issues in

determination of mechanical properties of bone using conventional mechanical testing.

Bone hydration, viscoelasticity and preparation of the samples were other noticeable

concerns that affected the experiment results. For that reason, ranges of values for

biomechanical properties of bone have been reported in different studies. The application

of a non-destructive method which has less effect on bone structure and allows the same

sample to be tested before and after excitation is more reliable to determine the possible

effect of pulse power simulation on elastic properties of cortical bone.

In comparison with mechanical testing methods, ultrasonic techniques comprise

significant advantages in determination of the bone elastic property. It is a non-invasive,

non-destructive method which uses small samples with less complicated shape which

allows measurement of bone elastic property in multi directions reducing the errors caused

by unidirectional measurements techniques160

.

The present work showed the inequalities on the values of the cortical bone elastic

property in different directions, which is consistent with similar previous researches.

Although it was shown that cortical bone elasticity in three main orthogonal directions was

not affected by pulse power stimulation, analysis of the elastic properties of bone in other

directions and levels inside the sample could be useful to determine the full effect of this

kind of electrical stimulation on the bone structure.



137

Furthermore, this work applied pulse electric filed in one bone crosswise direction

(radial direction). For a more complete assessment, analysis of other directions of pulse

power excitation would be required.

Chapter 8: Effect of Pulse Power Stimulation

on Functional and Physical

Characteristics of Cortical Bone

(Discussion and Conclusion)

Chapter 8: Effect of Pulse Power Stimulation on Functional Characteristics of Cortical

Bone

139

8.1 Introduction

Low-power electromagnetic fields have been applied during the last forty years as a

stimulation for osteogenesis and as a useful treatment for some chronic musculoskeletal

disorders like non-union bone fractures 4, 25

. Nevertheless, the behaviour of bone in

response to high voltage and high frequency electromagnetic fields (pulse power) has been

poorly explored. Applying this type of electrical stimulation on live bone firstly requires

the identification and introduction of controlled parameters and a safe method for applying

pulse power to bone tissue which requires investigating its effect on the fundamental

physical properties of bone structure. This thesis provides a step in this direction.

The main aim of this research was to investigate how the functional properties of

bone are influenced by pulse power stimulation. In the other words, whether or not pulse

power excitation affect the basic mechanical properties of bone.

Cortical bone bears a considerable portion of the load applied to the body and its

characteristics plays significant role in the mechanical competence of bone. This study

therefore, focused on evaluating the mechanical behaviour of cortical bone in response to

high power pulsed electromagnetic field exposure in terms of its structure and

microstructure.

Bone primarily consists of cells (living component) ensconced in an extra cellular

matrix (ECM). ECM is the composite material portion of cortical bone and consists of

about 70% mineral (mostly hydroxyapatite), 22% organic matrix (more than 90% type I

collagen and less than 10 % non-collagenous proteins) and 8% water by weight. Chapter 2

detailed bone structure and its functional behaviour. ECM is the base substance of the

functional and mechanical competence of bone and is in particular the target of this study.


Bone

140

Research on bone cell behaviour, in response to pulse power exposure, is itself a vast

research and is beyond the scope of this study.

The quality and spatial arrangements of bone constituents determine its functional

characteristics and can be influenced by different factors such as mechanical environment,

diseases, aging and other internal or external stimuli.

The mineralized collagen fibrils form the main structure of extra cellular matrix and

determine the mechanical properties of bone in nanoscale. The structural quality of this

matrix pertains to both the quality and the orientation of its collagen fibrils30

. The

orientation of collagen fibrils varies in the adjacent lamellae to bear the loads applied in

different directions. The elastic collagen fibrils provide the bone its elasticity and the

capacity to dissipate energy under deformation. Additionally, the cross-links between

collagen fibrils play an important role in the bone toughness and as such is strongly related

to the quality of collagen network40, 169

.

The integrity and the quality of the collagen matrix have a direct impact on the

toughness and strength of cortical bone tissue while it has no considerable effect on bone

stiffness50

. If collagen composition is altered (in quality or orientation) or denatures (e.g.

by heating over 160˚C) cortical bone toughness and strength will be changed47

.

Contrary to collagen fibrils, the mineral phase has less ability to withstand tensile

stresses and it is directly associated to the bone stiffness27, 41

. Although, the bone strength

has a direct correlation with the increase of the mineralisation, the ultimate strength of the

cortical bone tissue does not have such a deep association with the mineral content as does

the Young’s modulus 41

. On the other hand, increasing mineralisation makes bone more

brittle which results in less required energy to fracture 30

.


Bone

141

In general, a combination of collagen and mineral phases provides the bone with its

required stiffness and strength in response to applied loads 27

.

8.2 Research procedure description and justification

Typically, bone is loaded in a combination of compression and tension, bending and

torsion. It shows different characteristics in response to different loading patterns. In

addition, due to the anisotropic and heterogeneous nature of bone tissue, a range of values

for its behavioural parameters has been measured and reported in different studies aimed at

characterising bone. This issue happened in this study as well. Stiffness, strength and

toughness are three basic functional properties of bone that together provide insights for

characterising the bone’s mechanical viability.

Therefore, to have a more comprehensive assessment about the possible effect of

pulse power stimulation on the functional behaviour of cortical bone tissue, this study

applied three main conventional loading patterns including three point bending, tension

and compression tests. This also help to ensure that any variation measured in mechanical

properties of the cortical bone after pulse power excitation is not related to a particular

loading. The fundamental properties of cortical bone in these three loading patterns were

determined and compared with and without pulse power exposure. The effect of size and

geometry of the test samples on the pulse power influence on bone material structure was

also considered through the experiments. A supplementary fractographic study was

conducted by scanning electron microscopy to analyse the fracture surfaces of the broken

tensile test samples and investigate the possible effect of pulse power stimulation on

microstructure of cortical bone. A non-invasive, non-destructive method using ultrasonic

technique was also applied to evaluate the effect of high-voltage, high-frequency pulsed


Bone

142

electromagnetic field on the elastic property of millimetre-size samples in three orthogonal

directions.

Therefore, in summery the research procedure has established through three different

standard biomechanical experiments, ultrasound velocity measurement and scanning

electron microscopy in four steps to achieve its aim as follow:

8.2.1 Introduction of a suitable pulse power application set up and evaluation of the

flexural elasticity of cortical bone through non-destructive 3-point bending test

Because cortical bone, especially in long bones, is mostly loaded in bending, the first

experiments investigated the effect of pulse power exposure on flexural elastic modulus of

cortical bone. The primary test was conducted on whole bone samples but other

experiments shifted the test samples to tissue level. The cortical strips obtained in

dimensions of centimetres from femoral and tibial diaphysis. Three experimental

procedures were performed to establish a suitable, controlled set up. For this purpose

different pulse power parameters in magnitude, frequency and pulse width up to maximum

available power from pulse power generators were applied which led to the choice of high-

voltage pulses with 500V at 10 kHz.

Two different methods were examined for application of high-voltage, high-

frequency pulses to cortical bone samples. Firstly, the pulse power signals were exposed

with direct connection of electrodes with bone samples. This invasive method resulted in

passing a significant current through the sample, causing thermal effect, drying the sample,

and finally their burning. This method was therefore changed to capacitive coupling,

covering the electrodes with electrical isolation tape and placing the bone sample, wrapped

in saline soaked cloth, between them. This action changed the characteristics of the bone

sample from a resistive load to a capacitive load reducing the thermal and electrochemical


Bone

143

effects while the pulsed electromagnetic field effect on bone structure increased. This

method isolated the electrodes from the tissue, so that the polarisation, electrolysis,

electroosmosis and infection effect, reported in using direct electric connect77, 170, 171

, were

considerably reduced. Therefore, the remainder of the experiments conducted using pulses

of 500V at 10 kHz through the capacitive coupling method.

To determine and compare the elastic modulus of cortical bone strips with and

without pulse power excitation, a non-destructive three-point bending test was performed

(in the elastic region) on bone samples before and after exposure. Although a non-

destructive method gives the advantage of testing the same sample before and after

excitation and spontaneously omit the typical errors due to testing different biological

samples, it cannot provide the other fundamental parameters of bone such as strength and

the total fracture energy. The other mechanical testings therefore were conducted until

failure point.

8.2.2 The effect of pulse power exposure on the tensile strength and total fracture

energy accompanying the microstructure analysis of the test bone fracture

surfaces

In the next step, a destructive tensile test was performed until failure to evaluate the

possible variation in the strength and toughness of the bone samples exposed to pulse

power, compared with those of the control samples. In parallel, a non-destructive tensile

loading and unloading cycle was conducted to compare hysteresis energy absorption of the

samples after pulse power excitation, compared with that before exposure.

In addition to intrinsic material properties, microstructural properties of cortical bone

such as porosity, crystallinity and the presence of microcrackc determine the bone

mechanical potential30

. Therefore, the fracture surfaces of both control and PP-exposed

broken specimens were further examined microscopically to analyse the microstructural


Bone

144

properties of the cortical bone samples (including diffusion new microcracks or variation

in porosity or crystallinity of the samples) after exposure to pulse power. The fracture

pattern were also inspected to investigate whether or not pulse power stimulation led to

changes in ductile or brittle behaviour of the cortical bone tissue.

8.2.3 The effect of the pulse power excitation on the compressive strength and

toughness of the small sized samples

The next step used compressive testing which allowed application of less

complicated, millimetre-sized samples. This experiment again, determined and compared

the strength and total failure energy absorption of the samples exposed to high-voltage

pulses and the control samples. The reduction in the size of the samples accompanied with

a decreased electrode contact surface (using two series of isolated screws instead of

parallel plates) would increase the intensity of pulsed electromagnetic field on the bone

tissue (i.e. more leakage current passing through a smaller area).

8.2.4 Application of ultrasonic technique to evaluate the effect of pulse power on bone

elasticity

Mechanical tests are the usual methods in biomechanical studies for investigating the

structural and functional behaviour of biological samples in particular bone. Though the

application of the control samples in addition to the experimental samples can reduce

errors in the results caused by the mechanical methods, loading of the samples themselves

to some extent can affect the results of the experiments. These effects in general, are not

desirable particularly when investigate the effect of an exogenous stimulus. Therefore, the

application of an alternative non-destructive method such as ultrasonic technique is

helpful. This method determines the elastic property of bone samples using ultrasound

waveform velocity through the sample and bone density measurement which has several

advantages compared with conventional mechanical testing. Firstly, this method can be


Bone

145

repeated several times without a destructive effect on the bone structure, and is able to

produce more accurate results. Secondly, it allows application of small size samples (in

millimetre size) and finally, enables a simple measure of the bone sample elasticity in

different directions which is not easily available in other mechanical testings.

The bone mineral density, required for elastic modulus calculation, was determined

and compared using microCT, before and after pulse power excitation.

8.3 The effect of pulse power stimulation on functional behaviour of cortical bone

tissue

8.3.1 Results Interpretation

The results obtained from the above mentioned four-step procedure provide the

capacity to assess the basic functional properties of cortical bone tissue by examining the

micro and macrostructural changes in response to high voltage, high frequency pulsed

electromagnetic field stimulation.

As stated in the flowchart at the beginning of the thesis, the overall research

procedure was divided into two categories: non-destructive and destructive methods.

The elastic modulus of the cortical bone samples obtained from the non-destructive

three-point bending test in elastic portion and ultrasonic technique (chapter 4 and 7)

indicated that stiffness of cortical bone samples remained in the same range before and

after pulse power excitation in both control and PP-exposed samples. This suggests that

high voltage, high frequency pulsed electromagnetic fields (maximum 500V at 10 kHz) did

not influence the elasticity of cortical bone tissue after up to 144 hours continuous

excitation.


Bone

146

The detailed results of ultrasound velocity through the sample and its equivalent

elastic modulus in three directions (presented in chapter 7) showed that although the elastic

property of cortical bone were obviously different in three directions (show the anisotropic

nature of bone), similar variation was found between the elasticity of the control and PP-

exposed samples. The results of corresponding statistics analysis also showed no

significant differences in the elastic property of the cortical bone tissue before and after

pulse power excitation even on small-sized specimens. Figure 8.1 compares the total

average elastic modulus of the control and the PP-exposed samples, before and after

excitation, obtained from the ultrasonic technique. The graph highlights that the elastic

modulus of both normal and treated samples appeared to remain unchanged before pulse

power exposure compared with that after excitation (less than 5% variation).

PP-exposed

sam

ples

Contr

ol sam

ples

0

5

10

15

20

25 Before excitaion

After excitation

Ela

sti

c m

od

ulu

s (

GP

a)

Figure ‎8.1 The elastic modulus of the normal specimens compared with the samples exposed to pulse power

for 144 hours obtained from ultrasonic technique


Bone

147

Similarly, a comparison of the bending test results (in the elastic region) for the

normal and PP-exposed cortical bone strips, demonstrated that pulse power excitation (up

to 450V at 10 kHz for 5 days with 6 hours per day) did not affect significantly the elasticity

of the cortical bone tissue. Figure 8.2 compares the elastic modulus of the cortical bone

strips exposed to 450 V with 10 KHz pulse power signals from last experiment results in

chapter 4. The other experiments with lower pulse power magnitude and frequency

(presented in chapter 4) showed similar results which were not repeated here.

PP-exposed

sam

ple

control s

ample

0

10

20

30Before excitation

After excitation

Ela

sti

c m

od

ulu

s (

GP

a)

Figure ‎8.2 Comparison of the flexural elastic modulus of the control and the PP-exposed samples before and

after pulse power stimulation

As stated earlier, the stiffness and rigidity of cortical bone tissue is predominantly

related to mineral crystals embedded in collagen matrix and there is a significant

correlation between the mineral content and Young’s modulus of the bone tissue 29, 40, 41

.

The comparison of the cortical bone mineral density (BMD) determined from

microCT measurement in chapter 7, also showed that the bone mineral content remained


Bone

148

unchanged in both the control and PP-exposed samples after 144 hours pulse power

excitation. This outcome also confirms the results of comparison of Young’s modulus and

stiffness of the control and PP-exposed samples. Figure 8.3 illustrates the average amount

of bone mineral density of the control samples compared with that of the samples exposed

to pulse power for 144 hours.

The combination of these outcomes, along with the results of the elasticity

measurement demonstrates that pulse power stimulation does not apparently affect the

mineral phase structure in cortical bone. As stated in chapter 2, they are the predominant

factor in bone stiffness and obviously because bone was dead, the mineralisation process of

the bone tissue (related to bone cells activities) cannot also be influenced by this

stimulation.

PP-e

xpose

d sam

ples

Contr

ol sam

ples

0

500

1000

1500Before excitation

After excitation

Bo

ne M

inera

l D

en

sit

y (

g/c

m3)

Figure ‎8.3 Comparison of the bone mineral density of the control and PP-exposed samples before and after

pulse power excitation

A non-destructive tensile loading–unloading test was applied to determine and

compare the hysteresis energy dissipated by cortical bone specimens exposed to pulse


Bone

149

power and that of the control samples. This energy is measured by the area enclosed

between tensile loading and unloading cycle. The result of this experiment revealed that,

the samples stimulated by pulsed electromagnetic field showed a greater strain at a

constant load (40 N) compared with before excitation (Figure 5.10 & 5.11). This effect

shows a larger amount of hysteresis energy dissipation during a loading-unloading cycle

compared with the control samples in a similar loading pattern. Figure 8.4 compares the

average amount of dissipated hysteresis energy by both control and PP-exposed samples

before and after excitation.

PP-exp

osed s

ample

s

Contr

ol sam

ples

0

10

20

30Before 145h excitation

After 145h excitation

Hyste

rsis

en

erg

y (

N.m

)

Figure ‎8.4 Comparison of the hysteresis energy dissipated by the control and the PP-exposed samples before

and after excitation

From destructive experiments including tensile and compressive tests, the strength

and the total failure strain energy absorption of the samples exposed to pulse power were

determined and compared with those of the control samples. The total failure strain energy

was determined by the area underneath the stress-strain graphs in destructive tests and

therefore showed a larger value for the tougher tissue.


Bone

150

Figure 8.5 (A) and (B) compare the mean tensile strength and the total fracture

energy absorption by samples exposed to pulse power relative to the control samples. The

tensile test results indicate that the average strength and the total failure strain energy

absorption (toughness) of the cortical bone samples exposed to pulse power for 145 hours

were relatively similar to those of the control samples. It can suggest that these functional

properties of the cortical bone were not apparently influenced by pulse power stimulation.

Figure ‎8.5 Comparison of the tensile strength and total failure strain energy of the samples exposed to pulse

power for 145 hours with those of the control samples

In contrast, the ultimate compressive stress (compressive strength) and the total

fracture energy (bone toughness) of small sized samples after 66 hours of pulse power

excitation considerably increased relative to those of the control samples. This suggests

that smaller size samples can be affected more by pulse power excitation through smaller

cross-section electrodes. Therefore, the excitation method and the size of the samples can

influence the amount of pulse power effect on bone characteristics. Additionally, it can

show the variation in the functional behaviour of the cortical bone due to pulse power

excitation in different loading patterns.


Bone

151

Figure 8.6A shows that the samples which were stimulated by a high voltage, high

frequency pulsed electromagnetic field can absorb much larger amounts of energy before

fracture. It suggests that these specimens become tougher compared with the samples not

exposed to pulse power. The PP-exposed samples showed also higher compressive strength

compared to the control samples (Figure 8.6B).

Figure ‎8.6 The strength and total fracture energy absorption of the samples exposed to pulse power for 66

hours compared with those parameters of the control samples

The comparison of the average Young’s modulus of the samples exposed to pulse

power with that of the control samples, also supports the results of the non-destructive

experiments that showed no significant effect on the cortical bone stiffness due to pulse

power stimulation. Figure 8.7 presents the comparison of average compressive elastic

modulus of the cortical bone samples with and without pulse power excitation obtained

from compression test.


Bone

152

PP-exposed s

ample

s

Control s

ample

s

0

10

20

30

Youn

g's

mod

ulus

(GP

a)

Figure ‎8.7 Comparison of the Young’s modulus of the samples exposed to pulse power with that of the

control samples obtained from compression tests

8.3.2 Final results

The total results from four series of experiments performed in this study implied that

pulse power stimulation did not change the elastic properties of cortical bone samples

while it appeared to increase the bone strength and toughness. These findings may address

the effect of pulse power on collagen network portion of bone material rather the mineral

phase.

8.4 Discussion and Conclusion

As stated in chapter 2, the collagen matrix provides the bone with its ductility and the

ability to absorb energy (e.g. bone toughness) and sustain the tensile strain while the

mineral phase provides the bone with its stiffness and effective resistance to compressive

loading 21, 30, 32, 40

. The mineral contents have more effect on Young’s modulus than

ultimate strength of cortical bone tissue41, 42

whereas the collagen matrix has a poor

relation with Young’s modulus of bone tissue and a direct strong effect on its toughness32

.

Additionally, the quality and the orientation of collagen fibrils play a key role in the

bone strength and toughness 32

so that reorganization of collagen fibrils causes the


Bone

153

maintenance of the mechanical properties of bone tissue including its strength, although

bone mineral density was decreased 44

. The concentration, pattern and specific structure of

collagen crosslinks was also reported to play key role in the bone strength and material

deformation. For example, the collagen crosslink density in old bones appeared to be

higher than young bones resulting in the ability of lower bone components to dissipate

energy before fracture172

.

On the other hand, the orientation of the collagen fibrils can be affected by several

factors such as an electromagnetic field exposure. This effect was used as a most common

method for collagen fibril alignment in the synthesis of scaffolds that mimic the aligned

collagen fibrils in very regular tissue like tendon and ligament or as an aligned sheets in

bone and corneal tissue45, 46

.

The denaturation and deterioration of the bone collagen network, by heating or

ionising radiation for example, without changing in mineral content was reported to

significantly decrease the toughness and strength of the bone tissue with no effect on its

Young’s modulus 32, 47

. These outcomes reinforce previous studies suggesting that the

collagen network plays a key role in the bone toughness and overall strength while having

minimal effect on the bone elasticity.

These overall discussion accompanying the obtained results from this study suggest

that the pulse power stimulation can influence the arrangement or the quality of collagen

fibrils (e.g. the crosslink density between the collagen molecules), increasing the total

compressive strength and toughness of cortical bone material. However, it apparently did

not affect the mineral phase in the cortical bone tissue that is the main recognized factor for

the bone stiffness.


Bone

154

The outcomes of this research also confirmed that the indirect application of high

power pulsed electromagnetic field with nominated parameters through capacitive

coupling method, was athermal and did not destroy the bone tissue construction. The

continuous hydration of bone samples during the experiments was also an effective factor

preventing variation in bone mechanical properties.

At a microscopic level, according to scanning electron micrographs, no significant

changes appeared in microstructure (including crystallinity, porosity and microcracks

distribution) of cortical bone samples exposed to pulse power compared to the

morphological characteristics of the samples unexposed to pulse power (chapter 5).

Additionally, comparison of the fracture patterns on the fracture surfaces of the PP-

exposed samples and the control samples showed no significant variation in the ductile

behaviour of the cortical bone tissue due to this kind of stimulation.

The combined results of the experiments suggest that, although pulse power

stimulation appeared to have no significant effect on the mineral content, the porosity of

the cortical bone tissue and the diffusion of the new micro-cracks in the microstructure

level, it may have positively contributed to the arrangement and integrity of collagen

network. Nevertheless, clear understanding of the equivalent mechanisms under the effect

of pulse power stimulation on the increase of cortical bone compressive strength and

toughness and probably its embedded collagen network needs further research.

To author’s knowledge, this study was the first research investigating the effect of

high voltage, high frequency pulsed electromagnetic field on basic mechanical and

physical characteristics of cortical bone tissue.


Bone

155

8.5 Research limitations

This research was primarily established as a pilot study investigating the feasibility

of the safe and controlled application of pulse power on bone tissue. Due to the difficulty

in preparing large numbers of specimens manually, diversity in the experiments and their

required appropriate samples and lack of time, due to the scope of this research, a small

number of samples was employed in the experiments. Therefore, the first significant

limitation of this study was that the experiments and analysis were performed using a small

sample size. Additionally, although sheep bone is reported to be structurally and

hormonally similar to the human bone and also is readily available as well as widely

applied in orthopaedic research, for future research, it is suggested pulse power stimulation

be tested on a larger number of human bones before in vivo and clinical studies, for more

confidence. Other factors that may affect the results which need to be considered in this

study include bone type, gender and age of the donor, more isolated controlable

environmental conditions and other pulse power parameters such as pulse width and

application of current pulse instead of voltage pulse.

Another limitation of this thesis was that, because the present research hypothesis

was the investigation of pulse power on functional/mechanical properties of bone tissue

(composite material), high voltage, high frequency electromagnetic field was applied on

dead bone. However, for an actual evaluation of structural and functional behaviour of

bone and its real response to pulse power stimulation, it is proposed to investigate the

application of pulse power on live bone (the cell bones or in vivo study). For example,

although, this study suggested the positive effect of pulse power on collage network due to

increase on the compressive strength and total fracture energy absorption of the samples,


Bone

156

application of this kind of stimulation on living tissue may give higher elastic modulus due

to physiological response of bone cell which may lead to increase of mineralization.

Furthermore, in this thesis, pulse electric filed was applied in one bone crosswise

direction (radial direction) and in mechanical testings, only in a longitudinal direction.

Nevertheless, due to the anisotropic and inhomogeneous nature of cortical bone tissue, it

responds differently to different direction loading pattern. Hence, consideration of the

other directions of pulse power excitation and mechanical loading would be required for a

more complete assessment.

8.6 Future work and recommendation

As this thesis was the first step towards the application of pulse power on bone

tissue, there are different aspects for further research. The future work required to

overcome the limitations of the current study and extend this project apply pulse power in

clinical study. Although this thesis as a first step highlighted some ambiguous points about

the application of pulse power on bone tissue, the analysis and results would have been

more reliable if a larger number of samples from human bone are applied. Additionally,

further microscopic investigation for example using TEM with higher resolution capability

, micro computed tomography (MicroCT) with ability to present the possible systematic

variation in bone tissue microstructure, polarized light microscopy with the ability to show

the possible reorganisation of collagen fibrils or a histological analysis with appropriate

collagen staining, can clarify the mechanism behind the specified functional characteristics

of cortical bone tissue in response to pulse power exposure.

In this research, just one direction for pulse power stimulation of bone samples

(radial) and mechanical loading (longitudinal) were considered. Ultrasonic techniques to


Bone

157

some extent resolve this issue by offering three-directional measurement of bone elasticity.

However, due to anisotropic and inhomogeneous nature of bone and to obtain a more

complete assessment about bone behaviour in response to pulse power stimulation, it

would be beneficial to consider other directions for pulse power excitation and mechanical

loading.

This study applied the capacitive coupling and direct connection of the electrodes

and the bone samples for pulse power exposure. It may also be useful to explore inductive

coupling method using Helmholtz coil apparatus in particular for future in vivo studies.

The main motivation for this research was the successful application of low power

electrical stimulation for the therapeutic purposes of some bone disease. When this aim can

be achieved that bone as a dynamic, self-adaptive tissue can respond to stimulus. This is

not possible unless bone cells which are the sensors in living organs, are alive and sense

the stimulation and respond appropriately to it (increasing bone formation and improving

bone mechanical properties). Therefore, in order to have a more realistic view of the effect

of pulse power stimulation on the functional/structural behaviour of bone and its growth

and osteogenesis, it is proposed to explore the application of high power, high frequency

pulsed electromagnetic fields on living tissue.

158

References

1. Burr, H. and F. Northrop, Evidence for the existence of an electro-dynamic field in

living organisms. Proceedings of the National Academy of Sciences of the United

States of America, 1939. 25(6): p. 284.

2. Isaacson, B.M. and R.D. Bloebaum, Bone bioelectricity: What have we learned in

the past 160 years? Journal of Biomedical Materials Research Part A, 2010.

95A(4): p. 1270-1279.

3. Bassett, C.A.L., Beneficial Effects of Electromagnetic Fields. J Cellular Biochem,

1993. 51: p. 387-393.

4. Shapak, N.M., Therapeutic uses of pulsed magnetic-field exposure, a review. The

radio science bultein, 2003(307).

5. Hartshorne, E., On the causes and treatment of pseudarthrosis, and especially of

that form of it sometimes called supernumerary joint. The American Journal of the

Medical Sciences, 1841. 1(1): p. 121.

6. Lente, R., Cases of un-united fracture treated by electricity. New York State J Med,

1850. 5: p. 317-19.

7. Behari, J., Biophysical Bone Behaviour and Application. 2009, I.Hoboken:John

Wiley and sons.

8. Singh, S. and S. Saha, Electrical properties of bone: a review. Clinical

orthopaedics and related research, 1984. 186: p. 249.

9. HUANG, L.Q., et al., Clinical update of pulsed electromagnetic fields on

osteoporosis. Chinese medical journal, 2008. 121(20): p. 2095-2099.

159

10. Trock, D.H., Electromagnetic Fields and Magnets: Investigational Treatment for

Musculoskeletal Disorders. Rheumatic Disease Clinics of North America, 2000.

26(1): p. 51-62.

11. TREHARNE, R.W., et al., An in Vitro Study of Electrical Osteogenesis Using

Direct and Pulsating Currents. Clinical Orthopaedics and Related Research, 1979.

145: p. 300-306.

12. Brighton, C., J. Cronkey, and A. Osterman, In vitro epiphyseal-plate growth in

various constant electrical fields. The Journal of Bone and Joint Surgery, 1976.

58(7): p. 971.

13. K.Li, Y.H., S. Ma, G. Ding,Y. Guo , J. Liu, Y. Li, G. Guo, Inhibition of bone

formation by high intensity pulsed electromagnetic field in MC3T3-E1 cells.

Progress in electromagnetics research 2011. 112: p. 139-153.

14. Mankowski, J. and M. Kristiansen, A Review of Short Pulse Generator Technology.

IEEE TRANSACTIONS ON PLASMA SCIENCE, 2000. 28( 1).

15. Schoenbach, K.H., et al., Bioelectrics-new applications for pulsed power

technology. Plasma Science, IEEE Transactions on, 2002. 30(1): p. 293-300.

16. Behari, J., V.P. Arya, and Z.C. Alex, bone fracture healing using a capasitively

coupled RF field. journal of bioelectricity, 1991. 10( 1&2): p. 231-239

17. Gürgül, S., et al., Deterioration of bone quality by long-term magnetic field with

extremely low frequency in rats. Bone, 2008. 42(1): p. 74-80.

18. Akdag, M.Z., et al., The Effect of Long-Term Extremely Low-Frequency Magnetic

Field on Geometric and Biomechanical Properties of Rats' Bone. Electromagnetic

Biology and Medicine, 2010. 29(1–2): p. 9-18.

19. Luo, E., et al., Effects of the PEMFs of different intensity on BMD and

biomechanical properties of rabbits' femur]. Sheng wu yi xue gong cheng xue za

160

zhi= Journal of biomedical engineering= Shengwu yixue gongchengxue zazhi,

2005. 22(6): p. 1168.

20. Behari, J., Biophysical Bone Behavior: principles and applications. 2009: Wiley

Online Library.

21. Ritchie, R.O., M.J. Buehler, and P. Hansma, Plasticity and toughness in bone.

Physics Today, 2009. 62(6): p. 41-47.

22. Ahn, A.C. and A.J. Grodzinsky, Relevance of collagen piezoelectricity to "Wolff's

Law": A critical review. Medical Engineering & Physics, 2009. 31(7): p. 733-741.

23. Rho, J., L. Kuhn-Spearing, and P. Zioupos, Mechanical properties and the

hierarchical structure of bone. Medical engineering & physics, 1998. 20(2): p. 92-

102.

24. Hoffler, C., et al., A hierarchical approach to exploring bone mechanical

properties. Mechanical testing of bone and the bone–implant interface. Boca Raton:

CRC Press. p, 2000: p. 133–149.

25. Gueldner, S.H., et al., eds. Osteoporosis : clinical guidelines for prevention,

diagnosis, and management 2008.

26. Bone. Available from: http://en.wikipedia.org/wiki/Bone.

27. Pearson, O.M. and D.E. Lieberman, The aging of Wolff's “law”: Ontogeny and

responses to mechanical loading in cortical bone. American Journal of Physical

Anthropology, 2004. 125(S39): p. 63-99.

28. Burr, D., Targeted and nontargeted remodeling. Bone, 2002. 30(1): p. 2.

29. Beniash, E., Biominerals—hierarchical nanocomposites: the example of bone.

Wiley Interdisciplinary Reviews: Nanomedicine and Nanobiotechnology, 2011.

3(1): p. 47-69.

http://en.wikipedia.org/wiki/Bone

161

30. Augat, P. and S. Schorlemmer, The role of cortical bone and its microstructure in

bone strength. Age and Ageing, 2006. 35(suppl 2): p. ii27-ii31.

31. Voet, D. and J. Voet, Biochemistry, vol. 1. 2004, New York, J. Wiley.

32. Viguet-Carrin, S., P. Garnero, and P. Delmas, The role of collagen in bone

strength. Osteoporosis International, 2006. 17(3): p. 319-336.

33. Landis, W., et al., Mineral and organic matrix interaction in normally calcifying

tendon visualized in three dimensions by high-voltage electron microscopic

tomography and graphic image reconstruction. Journal of structural biology, 1993.

110(1): p. 39-54.

34. Manjhi, J. and J. Behari. Effect of electrical stimulation in mineralization and

collagen enrichment of osteoporotic rat bones. in Recent Advances in Microwave

Theory and Applications, 2008. MICROWAVE 2008. International Conference on.

2008.

35. P.A.Timmins, J.C.W., Bone water. Calcified tissue research, 1977. 23(1): p. 1-5.

36. Zhang, D., S. Weinbaum, and S. Cowin, Estimates of the peak pressures in bone

pore water. Journal of biomechanical engineering, 1998. 120: p. 697.

37. Robinson, R. and M. Watson, Collagen‐crystal relationships in bone as seen in the

electron microscope. The anatomical record, 1952. 114(3): p. 383-409.

38. Nomura, S., et al., Interaction of water with native collagen. Biopolymers, 1977.

16(2): p. 231-246.

39. LeGeros, R.Z., G. Bonel, and R. Legros, Types of “H 2 O” in human enamel and in

precipitated apatites. Calcified tissue international, 1978. 26(1): p. 111-118.

40. Zioupos, P. and J. Currey, Changes in the stiffness, strength, and toughness of

human cortical bone with age. Bone, 1998. 22(1): p. 57-66.

41. Currey, J.D., Bones: structure and mechanics. 2002: Princeton University Press.

162

42. Weiner, S., et al., Rotated plywood structure of primary lamellar bone in the rat:

orientations of the collagen fibril arrays. Bone, 1997. 20(6): p. 509-514.

43. Boskey, A.L., T.M. Wright, and R.D. Blank, Collagen and Bone Strength. Journal

of Bone and Mineral Research, 1999. 14(3): p. 330-335.

44. Puustjärvi, K., et al., Do More Highly Organized Collagen Fibrils Increase Bone

Mechanical Strength in Loss of Mineral Density After One‐Year Running Training?

Journal of bone and mineral research, 1999. 14(3): p. 321-329.

45. Guo, C. and L.J. Kaufman, Flow and magnetic field induced collagen alignment.

Biomaterials, 2007. 28(6): p. 1105-1114.

46. Becker, R. and G. Selden, The body electric: Electromagnetism and the foundation

of life. New York: William Morrow & Company. 1985, Inc.

47. Zioupos, P., J. Currey, and A. Hamer, The role of collagen in the declining

mechanical properties of aging human cortical bone. Journal of Biomedical

Materials Research, 1999. 45(2): p. 108-116.

48. Currey, J.D., et al., Effects of ionizing radiation on the mechanical properties of

human bone. Journal of Orthopaedic Research, 1997. 15(1): p. 111-117.

49. Currey, J.D., Role of collagen and other organics in the mechanical properties of

bone. Osteoporosis International, 2003. 14: p. 29-36.

50. Wang, X., et al., The role of collagen in determining bone mechanical properties.

Journal of orthopaedic research, 2001. 19(6): p. 1021-1026.

51. Paschalis, E.P., et al., Bone Fragility and Collagen Cross‐Links. Journal of Bone

and Mineral Research, 2004. 19(12): p. 2000-2004.

52. Oxlund, H., L. Sekilde, and G. Ørtoft, Reduced concentration of collagen reducible

cross links in human trabecular bone with respect to age and osteoporosis. Bone,

1996. 19(5): p. 479-484.

163

53. Martin, B., Aging and strength of bone as a structural material. Calcified tissue

international, 1993. 53: p. 34-40.

54. Boskey, A., Bone mineral crystal size. Osteoporosis International, 2003. 14: p. 16-

21.

55. Nyman, J.S., et al., The influence of water removal on the strength and toughness of

cortical bone. Journal of Biomechanics, 2006. 39(5): p. 931-938.

56. Lees, S., A mixed packing model for bone collagen. Calcified tissue international,

1981. 33(1): p. 591-602.

57. Currey, J.D., K. Brear, and P. Zioupos, The effects of ageing and changes in

mineral content in degrading the toughness of human femora. Journal of

Biomechanics, 1996. 29(2): p. 257-260.

58. Dempster, W.T. and R.T. Liddicoat, Compact bone as a non‐isotropic material.

American Journal of Anatomy, 1952. 91(3): p. 331-362.

59. Gupta, H.S., et al., Cooperative deformation of mineral and collagen in bone at the

nanoscale. Proceedings of the National Academy of Sciences, 2006. 103(47): p.

17741.

60. Almer, J. and S. Stock, Micromechanical response of mineral and collagen phases

in bone. Journal of structural biology, 2007. 157(2): p. 365-370.

61. Lees, S., N.J. Tao, and S. Lindsay, Studies of compact hard tissues and collagen by

means of Brillouin light scattering. Connective Tissue Research, 1990. 24(3-4): p.

187-205.

62. Riches, P.E., et al., Microhardness anisotropy of lamellar bone. Journal of

Biomechanics, 1997. 30(10): p. 1059-1061.

63. Fratzl, P. and R. Weinkamer, Nature's hierarchical materials. Progress in Materials

Science, 2007. 52(8): p. 1263-1334.

164

64. Weiner, S. and H.D. Wagner, The material bone: Structure mechanical function

relations. Annual Review of Materials Science, 1998. 28: p. 271-298.

65. Peterlik, H., et al., From brittle to ductile fracture of bone. Nature materials, 2005.

5(1): p. 52-55.

66. Ebacher, V. and R. Wang, A Unique Microcracking Process Associated with the

Inelastic Deformation of Haversian Bone. Advanced Functional Materials, 2009.

19(1): p. 57-66.

67. Ahn, A. and A. Grodzinsky, Relevance of collagen piezoelectricity to" Wolff's

Law": A critical review. Medical engineering & physics, 2009.

68. Niigata, N., Bioelectrical Repair and Growth, S.I. E. Fukada , T. Sakou, H.

Takahashi, N. Tsuyama, Editor. 1985.

69. Fukada, E., Piezoelectric properties of organic polymers. Annals of the New York

Academy of Sciences, 1974. 238(1): p. 7-25.

70. Fukada, E. and I. Yasuda, On the piezoelectric effect of bone. J. Phys. Soc. Japan,

1957. 12: p. 1158-1162.

71. Mycielska, M.E. and M. Djamgoz, Cellular mechanisms of direct-current electric

field effects: galvanotaxis and metastatic disease. Journal of cell science, 2004.

117(9): p. 1631.

72. Jahn, T.L., A Possible Mechanism for the Effect of Electrical Potentials on Apatite

Formation in Bone*. Clinical Orthopaedics and Related Research, 1968. 56: p. 261.

73. Onibere, R. and A. Khanna, The Role Of Electrical Stimulation In Fracture

Healing. The Internet Journal of Orthopedic Surgery, 2009. 11(2).

74. Anderson, J. and C. Eriksson, Piezoelectric properties of dry and wet bone. Nature,

1970. 227(5257): p. 491.

165

75. Reinish, G.B. and A.S. Nowick, Piezoelectric properties of bone as functions of

moisture content. Nature, 1975. 253(5493): p. 626-627.

76. Bassett, C.A.L. and R.O. Becker, Generation of electric potentials by bone in

response to mechanical stress. science, 1962. 137(3535): p. 1063.

77. Bassett, C.A.L., R.J. Pawluk, and R.O. Becker, Effects of Electric Currents on

Bone In Vivo. Nature, 1964. 204(4959): p. 652-654.

78. Becker RO., B.C.A.L., Bioelectrical factors controlling bone structure. Bone

Biodynamic, 1964.

79. Brighton, C.T., et al., A multicenter study of the treatment of non-union with

constant direct current. The Journal of bone and joint surgery. American volume,

1981. 63(1): p. 2.

80. Paterson, D., G. Lewis, and C. Cass, Treatment of delayed union and nonunion with

an implanted direct current stimulator. Clinical Orthopaedics and Related

Research, 1980. 148: p. 117.

81. Mackenzie D.A., P.K., Demetriades D., Non-invasive electrical stimulation of the

growh plate. Trans. Orthop. Res. Soc., 1977. 1: p. 241.

82. Piekarski, K., D. Demetriades, and A. Mackenzie, Osteogenetic stimulation by

externally applied DC current. Acta Orthopaedica, 1978. 49(2): p. 113-120.

83. Bodamyali, T., et al., Effect of faradic products on direct current-stimulated

calvarial organ culture calcium levels. Biochemical and biophysical research

communications, 1999. 264(3): p. 657-661.

84. Bushinsky, D.A., Metabolic alkalosis decreases bone calcium efflux by suppressing

osteoclasts and stimulating osteoblasts. American Journal of Physiology-Renal

Physiology, 1996. 271(1): p. F216-F222.

166

85. Cho, M., T.K. Hunt, and M.Z. Hussain, Hydrogen peroxide stimulates macrophage

vascular endothelial growth factor release. American Journal of Physiology-Heart

and Circulatory Physiology, 2001. 280(5): p. H2357.

86. Keramaris, N., et al., Fracture vascularity and bone healing: a systematic review of

the role of VEGF. Injury, 2008. 39: p. S45-S57.

87. Spadaro, J.A., Bioelectric Stimulation of Bone Formation: Methods, Models, and

Mechanisms. Electromagnetic Biology and Medicine, 1982. 1(1): p. 99-128.

88. Aaron, R.K. and D.M. Ciombor, therapeutic effect of electromagnetic fields in the

stimulation of connective tissue repair. celullar biochemistry, 1993. 52: p. 42-46.

89. Becker, R.O., Cross Currents — The Perils of Electropollution —The Promise of

Electromedicine. 1990: JP Tracher.

90. Brighton, C. and S. Pollack, Treatment of recalcitrant non-union with a

capacitively coupled electrical field. A preliminary report. J Bone Joint Surg Am,

1985. 67(4): p. 577-585.

91. Traina, G.C., et al., use of electric and magnetic stimulation in orthopaedics and

traumatology. Italian journal of orthopedic and traumatology, 1998. 24(1): p. 1-31.

92. Bassett, C.A., Fundamental and practical aspects of therapeutic uses of pulsed

electromagnetic fields (PEMFs). Critical Review in Biomedical Engineering, 1989

17(5): p. 451-529.

93. Nelson, F.R.T., et al., Use of physical forces in bone healing. Journal of the

American Academy of Orthopaedic Surgeons, 2003. 11(5): p. 344.

94. Kuzyk, P.R. and E.H. Schemitsch, The science of electrical stimulation therapy for

fracture healing. Vol. 43. 2009. 127-31.

167

95. Bassett, C.A., R.J. Pawluk, and A.A. Pilla, Acceleration of fracture repair by

electromagnetic fields: a surgically non-invasive method. Annals of the newyork

academy of sciences, 1974. 238: p. 242-262.

96. Bassett, C.A., R.J. Pawluk, and A.A. Pilla, Augmentation of bone repair by

electromagnetic fields. science, 1974. 184(136): p. 575-577.

97. Bassett, C.A., R.J. Pawluk, and A.A. Pilla, A non-operative Salvage of surgically-

resistent pseudarthroses and non-union by pulsing electromagnetic field. Clinical

Orthopaedics & Related Research 1977. 124(8): p. 128-143.

98. Matsumoto, H., et al., Pualsed electromagnetic fiels promote bone formation

around dental implants inserted into the femur of rabbits. Clinical oral implants

research, 2000. 11: p. 354-360.

99. Tabrah, F., et al., Bone density changes in octeoprosis-prone Women Exposed to

pulsed electromagnetic fields(PEMFs). Journal of bone and mineral research 1990.

5(5): p. 437-442.

100. Hartig, M., U. Joos, and H.P. Wiesmann, Capacitively coupled electric fields

accelerate proliferation of osteoblast-like primary cells and increase bone

extracellular matrix formation in vitro. European Biophysics Journal, 2000. 29(7):

p. 499-506.

101. Manjhi, J., R. Mathur, and J. Behari, Effect of low level capacitive-coupled pulsed

electric field stimulation on mineral profile of weight-bearing bones in

ovariectomized rats. Journal of Biomedical Materials Research Part B: Applied

Biomaterials, 2010. 92B(1): p. 189-195.

102. Spadaro, J. and R. Becker, Function of implanted cathodes in electrode-induced

bone growth. Medical and Biological Engineering and Computing, 1979. 17(6): p.

769-775.

168

103. Hassler, C., et al., Studies of Enhanced Bone Healing via Electrical Stimuli

Comparative Effectiveness of Various Parameters. Clinical Orthopaedics and

Related Research, 1977. 124: p. 9.

104. Black J., F.Z.B., Brighton C.A., growth response of intact bone to direct current

and pulsed current. Proc. 27th annual conf. Engineering Med. Biol., 1974(16): p.

289.

105. Aaron R.K. , C.D.M., role of electrical stimulation in the treatment of a vascular

necrosis of the femoral head. current orthopaedics, 1997. 11: p. 187-192.

106. Supronowicz, P.R., The effects of biophysical stimuli on select bone cell functions

pertinent to osteogenesis. , in Biomedical Engineering. New York,. Rensselaer

Polytechnic Institute. 2002.

107. Bassett, C.A.L., Biophysical principles effecting bone structure, in biochemistry

and phyisiology of bone. Academic press New York 1971. III.

108. Bassett, C.A.L., pulsating electromagnetic fields: a new approach to surgical

problems, in metabolic surgery. Grune and stratton Inc. , 1978: p. 225.

109. Brighton, C.T., et al., In vitro bone-cell response to a capacitively coupled

electrical field The role of field strength, pulse pattern, and duty cycle. Clinical

orthopaedics and related research, 1992. 285: p. 255.

110. Spadaro, J.A., Electrical osteogenesis—Role of the electrode material. Electrical

Properties of Bone and Cartilage, CT Brighton, J. Black and SR Pollack, eds.,

Crune and Stratton, New York, 1979: p. 189-197.

111. Reddy, G.N. and S. Saha, Electrical and Dielectric Properties of Wet Bone as a

Function of Frequency. Biomedical Engineering, IEEE Transactions on, 1984.

BME-31(3): p. 296-303.

169

112. Kosterich, J.D., K.R. Foster, and S.R. Pollack, Dielectric Permittivity and

Electrical Conductivity of Fluid Saturated Bone. Biomedical Engineering, IEEE

Transactions on, 1983. BME-30(2): p. 81-86.

113. Singh, S. and J. Behari, Frequency dependence of electrical properties of human

bone. Journal of bioelectricity, 1984. 3(1&2): p. 347-356.

114. Hannay, G.G., Mechanical and electrical environments to stimulate bone cell

development. 2006, PhD thesis, Queensland University of Technology.

115. Bassett, C.A.L. and M. Schink-Ascani, Long-term pulsed electromagnetic field

(PEMF) results in congenital pseudarthrosis Calcified Tissue International, 1991.

49,( 3).

116. Darendeliler MA, D.A., Sinclair PM, Effects of static magnetic and pulsed

electromagnetic fields on bone healing. Int J Adult Orthodon Orthognath Surg,

1997. 12(1): p. 43-53.

117. Luben, R., Effects of low-energy electromagnetic fields (pulsed and DC) on

membrane signal transduction processes in biological systems. Health Physics,

1991. 61(1): p. 15.

118. Berg, H. and L. Zhang, Electrostimulation in Cell Biology by Low-Frequency

Electromagnetic Fields Electromagnetic Biology and Medicine, , 1993. 12, 1993(

2): p. 147 - 163

119. Spadaro, J.A. and W.H. Bergston, In vivo and in vitro effects of electromagnetic

field on net calcium flux in rat calvarial bone. calcified tissue international, 2001.

70: p. 496-502.

120. Katsuki, S., et al., Biological effects of narrow band pulsed electric fields.

Dielectrics and Electrical Insulation, IEEE Transactions on, 2007. 14(3): p. 663-

668.

170

121. Deng, J., et al., The effects of intense submicrosecond electrical pulses on cells.

Biophysical journal, 2003. 84(4): p. 2709-2714.

122. Weaver, J.C., Electroporation of cells and tissues. Plasma Science, IEEE

Transactions on, 2000. 28(1): p. 24-33.

123. Liboff, A., et al., Time-varying magnetic fields: effect on DNA synthesis. science,

1984. 223(4638): p. 818.

124. Diniz, P., et al., Effects of pulsed electromagnetic field (PEMF) stimulation on bone

tissue like formation are dependent on the maturation stages of the osteoblasts.

Bioelectromagnetics, 2002. 23(5): p. 398-405.

125. Icaro Cornaglia, A., et al., Stimulation of osteoblast growth by an electromagnetic

field in a model of bone-like construct. European journal histochemistry, 2006.

50(3): p. 199-204.

126. Landry, P.S., et al., Electromagnetic fields can affect osteogenesis by increasing the

rate of differentiation. Clinical Orthopaedics and Related Research, 1997. 338: p.

262.

127. Yamamoto, Y., et al., Effects of static magnetic fields on bone formation in rat

osteoblast cultures. Journal of Dental Research, 2003. 82(12): p. 962-966.

128. Aaron, R.K., D.M.K. Ciombor, and G. Jolly, Stimulation of experimental

endochondral ossification by low‐energy pulsing electromagnetic fields. Journal of

Bone and Mineral Research, 1989. 4(2): p. 227-233.

129. Brighton, C.T., et al., Fracture healing in the rabbit fibula when subjected to

various capacitively coupled electrical fields. Journal of Orthopaedic Research,

1985. 3(3): p. 331-340.

130. Collazo, K.J.M.a.L., Suppression of a Differentiation Response in MC-3T3-E1

Osteoblast-Like Cells by Sustained, Low-Level, 30 Hz Magnetic-Field Exposure.

171

Radiation Research, The Electric and Magnetic Fields Research and Public

Information Dissemination (EMF-RAPID) Program, 2000. 153( 5, Part 2): p. 706-

714

131. Lohmann, C., et al., Pulsed electromagnetic field stimulation of MG63

osteoblast‐like cells affects differentiation and local factor production. Journal of

Orthopaedic Research, 2000. 18(4): p. 637-646.

132. Chang, W.H.S., et al., Effect of pulse‐burst electromagnetic field stimulation on

osteoblast cell activities. Bioelectromagnetics, 2004. 25(6): p. 457-465.

133. Hannay, G., D. Leavesley, and M. Pearcy, Timing of pulsed electromagnetic field

stimulation does not affect the promotion of bone cell development.

Bioelectromagnetics, 2005. 26(8): p. 670-676.

134. Akiyama, H., et al., Industrial Applications of Pulsed Power Technology.

Dielectrics and Electrical Insulation, IEEE Transactions on, 2007. 14(5): p. 1051-

1064.

135. Zabihi, S., et al., A new pulsed power supply topology based on positive buck-boost

converters concept. Dielectrics and Electrical Insulation, IEEE Transactions on,

2010. 17(6): p. 1901-1911.

136. Chakraborty, A., A. Khaligh, and A. Emadi. Combination of Buck and Boost

Modes to Minimize Transients in the Output of a Positive Buck-Boost Converter. in

IEEE Industrial Electronics, IECON 2006 - 32nd Annual Conference on. 2006.

137. Boora, A.A., et al. A general approach to control a Positive Buck-Boost converter

to achieve robustness against input voltage fluctuations and load changes. in

Power Electronics Specialists Conference, 2008. PESC 2008. IEEE. 2008.

172

138. Weissbach, R.S. and K.M. Torres. A noninverting buck-boost converter with

reduced components using a microcontroller. in SoutheastCon 2001. Proceedings.

IEEE. 2001.

139. An, Y. and R. Draughn, Mechanical testing of bone and the bone-implant interface.

2000: CRC.

140. Horan, M.P. and Y.H. An, Bone, Mechanical Testing of. Wiley Encyclopedia of

Biomedical Engineering. 2006: John Wiley & Sons, Inc.

141. Mandi, L. and M. Mark, Bending Tests of Bone, in Mechanical Testing of Bone and

the Bone-Implant Interface. 1999, CRC Press. p. 207-217.

142. Cowin, S.C., Bone mechanics handbook (second edition):, hardback ISBN 0-8493-

9117-2, chapter 7 Experimental technique for bone mechanics. 2001: Boca Raton,

Fla. ; London : CRC, c2001. 981.

143. Saffar, K., N. JamilPour, and S. Rajaai, How Does The Bone Shaft Geometry Affect

its Bending Properties? American Journal of Applied Sciences, 2009. 6(3): p. 463-

470.

144. Grant, C.A., et al., Creation of a validated 3D finite-element model of an ovine

tibia. 2009.

145. Newman, E., A. Turner, and J. Wark, The potential of sheep for the study of

osteopenia: current status and comparison with other animal models. Bone, 1995.

16(4): p. S277-S284.

146. Grant, C.A., et al., Creation of a validated 3D finite-element model of an ovine

tibia, in 4th Asia Pacific Biomechanics Conference. 2009: University of

Canterbury, Christchurch.

147. Özkaya, N., M. Nordin, and D. Leger, Fundamentals of biomechanics: equilibrium,

motion, and deformation. 1991, New York: Van Nostrand Reinhold.

173

148. Tony, K. and L. Michael, Tensile and Compression Testing of Bone, in Mechanical

Testing of Bone and the Bone-Implant Interface. 1999, CRC Press. p. 175-205.

149. Aspden, R., Mechanical testing of bone ex vivo. Methods in molecular medicine,

2003. 80: p. 369-380.

150. Inc., A., ANSYS. 2010.

151. Callister, W.D. and D.G. Rethwisch, Materials science and engineering: an

introduction, 8th Edition. January 2011, ©2009, John Wiley & Sons New York. p.

992.

152. Yuehuei, A., B. William, and D. Robert, Basic Concepts of Mechanical Property

Measurement and Bone Biomechanics, in Mechanical Testing of Bone and the

Bone-Implant Interface. 1999, CRC Press. p. 23-40.

153. Roylance, D., Stress-strain curves. Department of Materials Science and

Engineering, Massachusetts Institute of Technology, 2001.

154. Ritchie, R.O., How does human bone resist fracture? Annals of the New York

Academy of Sciences, 2010. 1192(1): p. 72-80.

155. Ductile and brittle failure. 2007; Available from:

http://www.materialsengineer.com/CA-ductbrit.htm.

156. Feng, L. and I. Jasiuk, Effect of specimen geometry on tensile strength of cortical

bone. Journal of Biomedical Materials Research Part A, 2010. 95A(2): p. 580-587.

157. Ritchie, R. and R. Nalla, Fracture, Aging and Disease in Bone and Teeth. Fracture

of Nano and Engineering Materials and Structures, 2006: p. 23-24.

158. Nalla, R.K., J.J. Kruzic, and R.O. Ritchie, On the origin of the toughness of

mineralized tissue: microcracking or crack bridging? Bone, 2004. 34(5): p. 790-

798.

http://www.materialsengineer.com/CA-ductbrit.htm

174

159. Turner, C.H. and D.B. Burr, Experimental techiques for bone mechanics, in Bone

Mechanics Handbook, S.C. Cowin, Editor. 2001, CRC Press. p. 7-1-7-35.

160. Mehta, S.S., et al., Bone material ultrasound velocity is predictive of whole bone

strength. Ultrasound in Medicine & Biology, 2001. 27(6): p. 861-867.

161. Jae-Young, R., An ultrasonic method for measuring the elastic properties of human

tibial cortical and cancellous bone. Ultrasonics, 1996. 34(8): p. 777-783.

162. Jae-Young, R., Ultrasonic Methods for Evaluating Mechanical Properties of Bone,

in Mechanical Testing of Bone and the Bone-Implant Interface. 1999, CRC Press.

p. 357-369.

163. H, K., Stress waves in solids. Journal of Sound and Vibration, 1964. 1(1): p. 88-

110.

164. Hearmon, R.F.S., An introduction to applied anisotropic elasticity. 1961: Oxford

University Press.

165. Njeh, C., et al., Orthogonal relationships between ultrasonic velocity and material

properties of bovine cancellous bone. Medical engineering & physics, 1996. 18(5):

p. 373-381.

166. Yuehuei, A., Mechanical Properties of Bone, in Mechanical Testing of Bone and

the Bone-Implant Interface. 1999, CRC Press. p. 41-63.

167. Yuehuei, A., B. William, and K. Ivars, Methods of Evaluation for Bone

Dimensions, Densities, Contents, Morphology, and Structures, in Mechanical

Testing of Bone and the Bone-Implant Interface. 1999, CRC Press. p. 103-118.

168. Langton, C., High Precision Quantitative Ultrasound Measurements of velocity and

Attenuation in Small Samples. 2010, Queensland University of Technology (QUT).

175

169. Banse, X., T. Sims, and A. Bailey, Mechanical Properties of Adult Vertebral

Cancellous Bone: Correlation With Collagen Intermolecular Cross‐Links. Journal

of Bone and Mineral Research, 2002. 17(9): p. 1621-1628.

170. Brighton, C.T., et al., Cathodic oxygen consumption and electrically induced

osteogenesis. Clinical Orthopaedics and Related Research, 1975. 107: p. 277.

171. Steckel, R., et al., Electrical stimulation on skin wound healing in the horse:

preliminary studies. American journal of veterinary research, 1984. 45(4): p. 800.

172. Buehler, M.J., Molecular nanomechanics of nascent bone: fibrillar toughening by

mineralization. Nanotechnology, 2007. 18: p. 295102.

Documents

Application of High Voltage, High Frequency Pulsed ...eprints.qut.edu.au/53105/1/Hajarossadat_Asgarifar_Thesis.pdf · Application of High Voltage, High Frequency Pulsed Electromagnetic