Surface nanoengineering of titanium alloys for …...Surface Nanoengineering of Titanium Alloys for...

Surface Nanoengineering of

Titanium Alloys for Biomedical

Applications

A thesis submitted in total fulfilment of the requirements for the degree of

Doctor of Philosophy

By

Sepideh Minagar

Faculty of Science, Engineering and Technology

Swinburne University of Technology

Hawthorn, Melbourne

Australia

February 2015

ii

Abstract

Failure of a biomaterial implant occurs when it cannot be accepted by the body. In terms

of orthopaedic implants, this means that there is no bonding between the implant

material and bone cells. In this study, fabricated nanotubular layers on the surface of

Ti50Zr binary alloy was investigated for its bioactivity and osseointegration after

anodisation. This study investigates the effect of nanopatterning of the surface as metal

oxide nanotubes which convert the bio-inert metal surface to be bioactive. This

nanotubular layer was examined for its ability to induce bone-like calcium phosphate

(CaP), e.g. hydroxyapatite or other compositions such as octacalcium phosphate and

tricalcium phosphate through immersion in a modified simulated body fluid (m-SBF).

The biocompatibility of the nanotubular layer was also assessed by cell culture test

using human osteoblast-like cells (SaOS-2). As the bare TiZr metal exhibited potential

for metallic implant applications due to its excellent bioactivity and biocompatibility, it

is expected that the nanotubular layer plays a promising role in forming strong

adhesions with bone cells, allowing it to be used as an implant.

The first part of this study investigates the condition for fabrication of TiO2-ZrO2-

ZrTiO4 nanotubes. This includes the changing of the anodisation parameters such as the

applied potential, the concentration of F- ion in the electrolyte, the type of electrolyte

and the anodisation time. In this research project, a full characterisation of the TiO2-

ZrO2-ZrTiO4 nanotubes including morphological, thermal, topographical, chemical and

mechanical properties has been investigated using scanning electron microscopy

(SEM), thin film X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS)

analysis, nanoindentation and water contact angle measurements. Bioactivity and

biocompatibility assessment have been employed in order to demonstrate the

characteristics of the nanotubular layer to improve hydroxyapatite mineralisation and

bone cell integration.

It was found that in the aqueous electrolyte there was a distribution of inner diameter,

outer diameter and wall thickness of nanotubes in comparison to TiO2 nanotubes as well

as in non-aqueous electrolyte, due to the different oxidation rates of Ti and Zr. The

orientation of nanotubes followed the microstructure of the Ti50Zr alloy. This effect

was obvious in the aqueous electrolyte whilst in the non-aqueous electrolyte it was less

affected and the orientation can be observed when the water content of the electrolyte

increased from 5 to 10 wt %. The walls of nanotubes were not separated when they

iii

were fabricated in non-aqueous electrolyte. The roughness parameters of the

nanotubular layer of TiO2-ZrO2-ZrTiO4 were higher than the nanotubular layer of TiO2,

due to the different height of the grown nanotubes. It has been shown that there was a

direct link between roughness parameter and wettability of the nanotubular layer. The

wettability or hydrophilic properties of nanotubes increased with an increase in

roughness. The TiO2-ZrO2-ZrTiO4 nanotubes exhibited higher surface energy than TiO2

nanotubes. Hydroxyapatite mineralised on the TiO2-ZrO2-ZrTiO4 nanotubes with higher

Ca/P ratio and thickness than on the TiO2 nanotubes. When Sa of the as-formed TiO2-

ZrO2-ZrTiO4 nanotubes increased, the Ca/P ratio of HA increased in contrast to

annealed nanotube. Increasing the hydrophilic properties of either as-formed or

annealed nanotubes resulted in an increase of the Ca/P ratio of HA. After annealing, the

Ca/P ratio of HA increased. Results of the MTS assay indicated that the percentage of

cell adhesion on the nanotubes fabricated in aqueous and non-aqueous electrolyte was

critically affected by the nanoscale topographical parameters including the tube inner

diameter (Di), the tube wall thickness (Wt), the amplitude roughness (Sa) and the spacing

roughness (Sm) of the nanotubular surface. The highest percentage of cell attachment

was on the surface of TiO2-ZrO2-ZrTiO4 nanotubes with Di = 18 nm (optimum

nanospacing) and the lowest percentage of cell attachment was on the surface of TiO2-

ZrO2-ZrTiO4 nanotubes with Di = 59 nm (higher amplitude parameter of roughness, Sa)

when the nanotubes were fabricated in an aqueous electrolyte. It was found that a post-

treatment that reduces the contamination of nanotubular surface or mineralising

hydroxyapatite can increase cell attachment significantly from 41.0 % to 74.8 % and

59.7 %, respectively. When the nanotubes were fabricated in non-aqueous electrolyte,

bone cells behaved similarly. The highest percentage of cell attachment was on the

surface of TiO2-ZrO2-ZrTiO4 nanotubes with Di = 25 nm (optimum nanospacing) and

the lowest percentage of cell attachment were on the surface of TiO2-ZrO2-ZrTiO4

nanotubes with Di = 29 nm (higher spacing parameter of roughness, Sm).

iv

Acknowledgments

First and foremost I wish to express my deepest gratitude to my principal supervisor

Professor Cuie Wen for her continuous support, encouragement and unstinting

supervision in conducting research, and her optimistic manner and valuable guidance in

writing the research publications. I also appreciate her trust and friendship which means

a great deal to me. I would also like to express my sincere gratitude to my associate

supervisor Professor Christopher C. Berndt for his invaluable advice and guidance in

writing the research publications. I must thank my other associate supervisor Dr James

Wang for his continuous help and support in the metallographic and spectroscopy

laboratories. I am thankful to Professor Elena P. Ivanova for her supervision on the cell

biology aspect of this research. I also acknowledge the State Government of Victoria of

Australia and Swinburne University of Technology for supporting and funding this

research project through a Victorian International Research Scholarship (VIRS) and

Swinburne University Postgraduate Research Award (SUPRA).

I wish to thank Dr Yuncang Li for his guidance and advice in the cell culture study and

the nanoindentation research at the Institute for Frontier Materials, Deakin University,

Geelong. I also thank Dr Thomas Gengenbach for his help in XPS analysis at

CSIRO Materials Science and Engineering, Clayton.

The help of Dr Mehran Motamed Ektesabi, Dr Igor Sbarski, Dr Akbar Rhamdhani, Dr

Thomas Ameringer, Dr Mostafa Nikzad, Dr Yeannette Lizama and Dr Vi Khanh

Truong for permitting me to use their equipments and/or chemicals is highly

appreciated. My thanks are extended to Dr De Ming Zhu for training me to use the 3D-

Profilometer. The technical assistance of Messieurs Brian Dempster and Andrew Moore

and the technicians of the workshop of the Faculty of Science, Engineering and

Technology are gratefully acknowledged. I thank all my supervisor research group

members (PhD students and Research Fellows) for their scientific discussions and

advice. I wish to thank Ms Madeleine Bruwer, the liaison librarian of the Faculty of

Science, Engineering and Technology because she always answered my questions

patiently. My sincere thanks to Professor Syed H. Masood, Associate Professor Paul

Stoddart and Dr Ryan Cottam for their advice in my annual progress reviews.

v

I wish and must thank especially my close friend Ms Fatemeh Hamedi because of her

sister-like friendship and emotional support. She was, is and will always be willing to

help me no matter how hard the challenge. I would like to thank all my friends here and

in my home country for their friendship and support, which made the course of my PhD

studies an enjoyable experience. I would also like to sincerely thank Dr Gwyneth and

Prof Michael Asten, Ms Rosemarie and Mr John Worboys as they have not only helped

me improve my English but also made me feel right at home.

Last but not least, I should thank my mother Farazandeh and my father Mohammad

Hasan for their endless loves, prays, encouragements and support. I would also like to

express my love and gratitude to my sister Sima, my brother Sepehr and my niece Zahra

who encouraged me all the time.

vi

Declaration

This thesis contains no material which has been accepted for the award to the candidate

of any other degree or diploma, except where due reference is made in the text. I declare

that to the best of my knowledge this study contains no material previously published or

written by another person except where due reference is made in the text.

Sepideh Minagar

February 2015

vii

List of publications

Journal Papers:

Sepideh Minagar, Christopher C. Berndt, James Wang, Elena Ivanova, Cuie Wen. A

review of the application of anodization for the fabrication of nanotubes on metal

implant surfaces. Acta Biomaterialia, (2012), 8, 2875-2888.

Sepideh Minagar, James Wang, Christopher C. Berndt, Elena P. Ivanova, Cuie Wen.

Cell response of anodized nanotubes on titanium and titanium alloys - a review. Journal

of Biomedical Materials Research: Part A, (2013), 101 A (9), 2726-2739.

Sepideh Minagar, Christopher C. Berndt, Thomas Gengenbach, Cuie Wen. Fabrication

and characterization of TiO2-ZrO2-ZrTiO4 nanotubes on TiZr alloy manufactured via

anodization. Journal of Materials Chemistry B, (2014), 2 (1), 71-83.

Sepideh Minagar, Yuncang Li, Christopher C. Berndt, Cuie Wen. The influence of

titania-zirconia-zirconium titanate nanotube characteristics on osteoblast cell adhesion.

Acta Biomaterialia (2015), 12, 281–289.

Presentation at Research Conferences:

Poster presentation entitled “Simultaneous enhancing cell interaction and eliminating

bacterial infection in titania nanotubes”, Proceedings of the 4th International

Conference on Nanostructures (ICNS4), Kish Island, I.R. Iran, March 2012.

Oral presentation entitled “anodised nanotubes on titanium alloy”, 15th International

Conference on Advances in Materials & Processing Technologies Conference (AMPT),

Wollongong, Australia, September 2012.

viii

Table of contents

Abstract ............................................................................................................................. ii

Acknowledgments ............................................................................................................ iv

Declaration ....................................................................................................................... vi

List of publications .......................................................................................................... vii

Table of contents ............................................................................................................ viii

List of figures ................................................................................................................. xiii

List of tables .................................................................................................................. xxii

List of abbreviations ..................................................................................................... xxiv

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

1.1 Overview ................................................................................................................. 1

1.2 Thesis objective .................................................................................................. 4

1.3 Thesis structure ................................................................................................... 5

Chapter 2 Literature review .............................................................................................. 7

2.1 Introduction ............................................................................................................. 7

2.1.1 Bone: structure, composition and properties .................................................... 8

2.1.2 Bone implant materials .................................................................................. 11

2.1.3 Surface treatment for implant materials ......................................................... 15

2.2 Anodic oxidation as a metallic implant surface treatment ............................... 16

2.2.1 The influence of the type and concentration of aqueous electrolyte on TiO2

nanotubes ................................................................................................................ 20

2.2.2 The influence of the non-aqueous electrolyte on TiO2 nanotubes ................. 23

2.2.3 The effect of pH value on the formation of TiO2 nanotubes.......................... 24

2.3 Nanotube oxide layer on titanium alloys and titanium alloying metals ........... 26

2.3.1 Anodisation of biocompatible Ti-Nb-Ta-Zr alloy.......................................... 26

2.3.2 Anodisation of binary titanium alloys for implant applications..................... 28

2.3.3 Anodisation of tantalum as a β stabiliser ....................................................... 33

ix

2.3.4 Anodisation of niobium as a β stabiliser ........................................................ 35

2.3.5 Anodisation of zirconium as a neutral titanium alloying element (or an

effective β stabilizer in multi-elementary Ti alloys) ............................................... 35

2.4 Factors that influence bone cell adhesion ......................................................... 37

2.4.1 The influence of surface physicochemical, mechanical and electrical

properties on bone cell behaviour ........................................................................... 39

2.4.2 Effect of topography on the bone cell behaviour ........................................... 41

2.5 Effect of the characteristics of anodised TiO2 nanotubes on bone cell behaviour

43

2.5.1 Effect of nano-spacing of the surface of TiO2 nanotubes on cell behaviour .. 43

2.5.2 Effect of crystalline phase of TiO2 nanotubes on bone cell behaviour .......... 49

2.5.3 Effect of hydroxyapatite (HA) coating on the surface of TiO2 nanotubes on

bone cell behaviour ................................................................................................. 52

2.6 In vivo effect of micro/nanostructure of the surface of TiO2 nanotubes .......... 54

2.7 Effect of TiO2 nanotubes on bacteria attachment ............................................. 56

2.8 Summary .......................................................................................................... 56

Chapter 3 Materials and methods .................................................................................... 59

3.1 Introduction ........................................................................................................... 59

3.2 Sample preparation ........................................................................................... 59

3.3 Fabrication of nanoporous and nanotubular layers .......................................... 59

3.4 Surface characterisation.................................................................................... 61

3.4.1 Surface morphology and chemical composition characterisation .................. 61

3.4.2 Surface topography and water contact angle and surface energy measurement

................................................................................................................................. 61

3.5 Bioactivity assessment by SBF soaking ........................................................... 62

3.6 Assessing cell responses on nanotubes with different nanoscale dimensions and

surface topographies .................................................................................................... 64

3.7 Nanohardness and elasticity measurements .......................................................... 65

x

Chapter 4 Nanotubes formed in aqueous electrolyte ...................................................... 67

4.1 Introduction ...................................................................................................... 67

4.2 Materials and methods ...................................................................................... 70

4.3 Results and discussion ...................................................................................... 71

4.3.1 Formation of TiO2-ZrO2-ZrTiO4 nanotubes .................................................. 71

4.3.2 Surface roughness of the TiO2-ZrO2-ZrTiO4 nanotubular surface ................ 95

4.3.3 Hydrophilic properties of the nanotubular surfaces ..................................... 100

4.3.4 Mechanical properties of the TiO2-ZrO2-ZrTiO4 nanotubes ........................ 102

4.3.5 Effect of annealing on the TiO2-ZrO2-ZrTiO4 nanotubes ............................ 104

4.4 Conclusions .................................................................................................... 107

Chapter 5 Nanotubes formed in non-aqueous electrolyte ............................................. 110

5.1 Introduction .................................................................................................... 110

5.2 Materials and methods .................................................................................... 112

5.3 Results and discussion .................................................................................... 113

5.3.1 Formation and characterisation of TiO2-ZrO2-ZrTiO4 nanotubes ............... 113

5.3.2 Surface roughness and hydrophilic property of the TiO2-ZrO2-ZrTiO4

nanotubular surface ............................................................................................... 125

5.3.3 Mechanical properties .................................................................................. 131

5.4 Conclusions .................................................................................................... 133

Chapter 6 Bioactivity of nanotubes fabricated in aqueous and non-aqueous electrolytes

....................................................................................................................................... 135

6.1 Introduction .................................................................................................... 135



6.3.1 Dimensional, physical and chemical properties of TiO2-ZrO2-ZrTiO4

nanotubes fabricated in aqueous electrolyte ......................................................... 139

6.3.2 Bioactivity of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in aqueous

electrolyte .............................................................................................................. 146

xi


nanotubes fabricated in non-aqueous electrolyte .................................................. 155

6.3.4 Bioactivity of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in non-

aqueous electrolyte ................................................................................................ 158

6.4 Conclusions .................................................................................................... 165

Chapter 7 Cell response of nanotubes formed in both aqueous electrolyte and non-

aqueous electrolyte ........................................................................................................ 167

7.1 Introduction .................................................................................................... 168




nanotubes fabricated in aqueous electrolyte ......................................................... 174

7.3.2 Cell adhesion and spreading on TiO2-ZrO2-ZrTiO4 nanotubes fabricated in

aqueous electrolyte ................................................................................................ 180


nanotubes fabricated in non-aqueous electrolyte .................................................. 201


non-aqueous electrolyte ........................................................................................ 205

7.4 Conclusions .................................................................................................... 210

Chapter 8 Nanoporous and nanotubular metal oxide layers on biocompatible metals of

Ta, Nb and Zr and their potential applications .............................................................. 212

8.1 Introduction .................................................................................................... 213



8.3.1 Process conditions and formation mechanism of nanoporous/nanotubular

metal oxides .......................................................................................................... 217

8.3.2 The dynamics of the anodisation process for tantala (Ta2O5), niobia (Nb2O5)

and zirconia (ZrO2) ............................................................................................... 218

xii

8.3.3 Physical characteristics of nanoporous Ta2O5 and Nb2O5 and nanotubular

ZrO2 layers ............................................................................................................ 222

8.3.4 Bioactivity of nanoporous and nanotubular metal oxide layers ................... 230

8.4 Conclusions .................................................................................................... 235

Chapter 9 Conclusions .................................................................................................. 237

9.1 Introduction ......................................................................................................... 237

9.2 Major findings ................................................................................................ 237

9.3 Recommendation for future work .................................................................. 244

References ..................................................................................................................... 246

xiii

List of figures

Fig. 2.1 Hierarchical structure of cortical bone (adapted from [32]) ................................ 9

Fig. 2.2 Growth of regular TiO2 nanotubes, a) cathodic reaction, b) anodic reaction, c)

transition state of TiO2 layer, d) starting of nanotube formation and e) titania nanotubes

......................................................................................................................................... 18

Fig. 2.3 Schematic current-time curve (adapted from [58])............................................ 19

Fig. 2.4 SEM and TEM images of TiO2 nanotubes (top view, cross-section and bottom):

a) formed in 1 M H3PO4 + 0.3 wt % HF at 15 V [61], b) in 1 wt % HF at 20 V for 15

and 30 min) [66], c) in 1 M (NH4)2SO4 + 0.5wt % NH4F at 20V [67], d) in ethylene

glycol + 0.3 wt % NH4F + 2 vol % H2O [68] ................................................................. 21

Fig. 2.5 SEM and TEM images (top view, cross-section and bottom) of anodic oxide

nanotube and nanoporous layer formed on a) β phase titanium alloy after anodisation at

20 V for 4000 s [99], d) Zr substrate using 1M (NH4)2SO4 + 0.5 wt % NH4F

electrolyte [102], c) niobium substrate as a function of anodisation temperature at 15 °C

[103], b) tantalum substrate using 1 M H2SO4 + 2 wt % HF electrolyte for 2 h with a

sweep rate 100 mVs-1 [104] ............................................................................................ 28

Fig. 2.6 Representation of the cell proteins involved in cell adhesion on biomaterial: (a)

immediately after implantation; (b) adsorbing proteins from body fluid; and (c) attached

bone cell on an implant material surface in higher magnification (adapted from [15]) . 38

Fig. 2.7 SEM image of filopodia of the SaOs-2 cells on 200 nm deep round concentric

grooves and ridges in quartz [157] .................................................................................. 42

Fig. 2.8 Schematic illustration of a bone cell (osteoblast) attached on titania nanotubes

with a diameter less than 100 nm (adapted from [109] and [35]). .................................. 46

Fig. 2.9 SEM images of extended MC3T3-E1 preosteoblast cell filopodia on nanotube

layers with different diameters: (a) 20 nm, (b) 50 nm, (c) 70 nm, (d) 100 nm, and (e)

120 nm (×70,000), (f) 120 nm (×30,000) [36] ................................................................ 51

Fig. 4.1 SEM images of TiO2-ZrO2-ZrTiO4 nanotubes: (a-1) microstructure of etched

Ti50Zr, (a-2) top view of patterned nanotubes of different phases exhibited in the

microstructure of Ti50Zr alloy, (b) nanoporous patches of different phases, (c) cross

section of the nanotubes showing the nanotube length, (d) view of the nanotubes from

xiv

the bottom, (e) top view and cross section of separated nanotubes of and (f) top view of

damaged nanotubes ......................................................................................................... 72

Fig. 4.2 SEM images of TiO2-ZrO2-ZrTiO4 nanotubes: (a) top view of as-formed

nanotubes anodised in 0.4 wt % NH4F at 20 V, (b) the nanotubes annealed at 500 ºC for

3 h, (c) nanotubes as-formed in 0.5 wt % NH4F anodised at 20 V, (d) the nanotubes

annealed at 500 ºC for 3 h, (e) nanotubes as-formed in 0.5 wt % NH4F anodised at 15 V,

and (f) the nanotubes annealed at 500 ºC for 3 h ............................................................ 74

Fig. 4.3 (a) Illustration of an electrochemical cell that indicates the electrolyte ions

species, (b) SEM image of top view of nanoporous TiO2-ZrO2-ZrTiO4 fabricated on a

Ti50Zr alloy after anodisation for 15 min in 0.1 wt % NH4F, 5V, and c) SEM image of

top view of TiO2-ZrO2-ZrTiO4 nanotubes fabricated on Ti50Zr alloy after anodisation

for 2.75 h in 0.3 wt % NH4F, 5V .................................................................................... 76

Fig. 4.4 Histograms of as-formed and annealed TiO2-ZrO2-ZrTiO4 nanotube parameters

anodised at 20 V for 2 h for different concentrations of fluorine anion (F-): (a,d) inner

diameter (Di), (b,e) outer diameter (Do), and (c,f) wall thickness (Wt), Note: The

nanotube size distribution graphs were generated from100 nanotubes on different

positions for each of three samples (300 measurements)................................................ 79

Fig. 4.5 Histograms of TiO2-ZrO2-ZrTiO4 nanotube parameters anodised in 0.5 wt %

NH4F electrolyte for 2 h at different applied potentials: (a, b) inner diameter (Di) as-

formed and after annealing, (c, d) outer diameter (Do) as-formed and after annealing, (e,

f) wall thickness (Wt) as-formed and after annealing, respectively ................................ 83

Fig. 4.6 Mean nanotube size of Ti50Zr and CP-Ti as a function of CF-: (a) Di for As-

formed, (b) Di for Annealed (c) Do for as-formed, (d) Do for Annealed, (e) Wt for as-

formed and (f) Wt for Annealed ...................................................................................... 84

Fig. 4.7 Mean nanotube size of Ti50Zr and CP-Ti as a function of applied potential: (a)

Di for as-formed, (b) Di for Annealed (c) Do for as-formed, (d) Do for Annealed, (e) Wt

for as-formed and (f) Wt for Annealed ............................................................................ 85

Fig. 4.8 The effect on nanotube length by: a) anodisation time and b) applied potential

......................................................................................................................................... 86

Fig. 4.9 SEM images of nanotubular layer fabricated at a) 20V - as-formed, b) 20V -

annealed, c) 25V - as-formed, d) 25 V - annealed, e) 30 V - as-formed, f) 30V -

annealed, g) 35 V - as-formed and h) 35 V - annealed ................................................... 87

xv

Fig. 4.10 Histogram and distribution of Di, Do and Wt of as-formed and annealed TiO2-

ZrO2-ZrTiO4 nanotubes fabricated at a,b) 20 V, c,d) 25 V, e,f) 30 V and g,h) 35 V ..... 90

Fig. 4.11 Histogram and fitted normal and Weibull distribution of Di of as-formed

nanotubular TiO2-ZrO2-ZrTiO4 fabricated at: a) 20 V, b) 25 V, c) 30 V, d) 35 V and e)

normal distribution of all four conditions ....................................................................... 93

Fig. 4.12 Histogram and fitted normal and Weibull distribution of Di, Do and Wt of

annealed nanotubular TiO2-ZrO2-ZrTiO4 fabricated at a) 10 V, b) 15 V, c) 20 V .......... 94

Fig. 4.13 EDS analysis for the (a) top and (b) bottom of TiO2-ZrO2-ZrTiO4 nanotubes

formed in 0.5 wt % NH4F at 20 V after 2 h .................................................................... 95

Fig. 4.14 The mean roughness (Sa) and the mean water contact angle (W.C.A.) of the

nanotubular surfaces of Ti50Zr alloy as a function of: (a), (b) F- concentration and (c),

(d) applied potential, respectively ................................................................................... 96


nanotubular surfaces of CP-Ti as a function of: (a), (b) F- concentration and (c), (d)

applied potential, respectively ......................................................................................... 96

Fig. 4.16 Loading-unloading forces versus the nanoindentation depths of a) as-formed

TiO2-ZrO2-ZrTiO4 and b) annealed TiO2-ZrO2-ZrTiO4 fabricated at the applied potential

20 to 35 V ...................................................................................................................... 102

Fig. 4.17 The nano mechanical properties of: a) hardness, b) reduced elastic Modulus

and c) elastic Modulus of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at different applied

potential ......................................................................................................................... 103

Fig. 4.18 XRD patterns of the nanotube samples fabricated on Ti50Zr (α and β phases)

via anodisation. (a) as-formed amorphous TiO2 and ZrTiO4 and orthorhombic ZrO2; (b)

annealed at 500 °C for 3 h tetragonal anatase, srilankite (a mixture of orthorhombic

TiO2 and ZrO2) and orthorhombic ZrTiO4 .................................................................... 105

Fig. 4.19 XPS spectra for the as-formed and the annealed TiO2-ZrO2-ZrTiO4 nanotubes.

(a) O1s, (b) Ti 2p, and (c) Zr 3d .................................................................................... 106

Fig. 5.1 SEM images of top view of nanotubes fabricated at: a) 5 wt % H2O at 20 V, b)

10 wt % H2O at 20V, c) 5 wt % H2O at 30 V and d) 10 wt % H2O at 30V in ethylene

glycol ............................................................................................................................. 114

Fig. 5.2 Histograms of TiO2-ZrO2-ZrTiO4 nanotube parameters anodised in 0.5 wt %

NH4F and ethylene glycol consisting of 5 and 10 wt % H2O for 90 min at 20 and 30V:

xvi

(a, d) inner diameter (Di) of as-formed and annealed nanotubes, (b, e) outer diameter

(Do) of as-formed and annealed nanotubes, (c, f) wall thickness (Wt) of as-formed and

annealed, respectively ................................................................................................... 116

Fig. 5.3 Histogram and fitted normal and Weibull distribution of Di, Do and Wt of as-

formed and annealed nanotubular TiO2-ZrO2-ZrTiO4 fabricated at a,e) 5 wt % H2O, 20

V, b, f) 10 wt % H2O, 20 V c, g) 5 wt % H2O, 30 V, d,h) 10 wt % H2O, 30 V ............ 120

Fig. 5.4 Mean nanotube sizes of Di, Do and Wt: a,c) as-formed and b,d) annealed TiO2-

ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in two different water contents and two

different applied potentials in ethylene glycol .............................................................. 122

Fig. 5.5 EDS analysis for: (a) top and (b) bottom of TiO2-ZrO2-ZrTiO4 nanotubes

formed in 0.5 wt % NH4F, 5 wt % H2O in ethylene glycol at 20 V after 90 min ......... 123

Fig. 5. 6 The effect on TiO2-ZrO2-ZrTiO4 nanotubes length formed in organic

electrolyte by: a) anodisation time and b) applied potential ......................................... 124


via anodisation in non-aqueous electrolyte. (a) as-formed amorphous TiO2 and ZrTiO4

and orthorhombic ZrO2; (b) annealed at 500 °C for 2 h tetragonal anatase, rutile,

srilankite (a mixture of orthorhombic TiO2 and ZrO2) and orthorhombic ZrTiO4 ....... 125

Fig. 5.8 a), c) The mean roughness (Sa), (Sq) and b), d) the mean water contact angle

(W.C.A.) of the nanotubular surfaces of TiO2-ZrO2-ZrTiO4 and TiO2 respectively, Note:

Each data point is an average of five measurements..................................................... 126


TiO2-ZrO2-ZrTiO4 and b) annealed TiO2-ZrO2-ZrTiO4 fabricated at 5 and 10 wt % H2O

and the applied potential 20 and 30 V ........................................................................... 131

Fig. 5.10 The nano mechanical properties of of TiO2-ZrO2-ZrTiO4 nanotubes fabricated

at different water content and applied potential: a) hardness, b) reduced elastic Modulus

and c) elastic Modulus .................................................................................................. 132

Fig. 6.1 Mean nanotube size as a function of applied potential: (a) Ti50Zr - as-formed,

(b) Ti50Zr - annealed (c) CP-Ti - as-formed, (d) CP-Ti - annealed ............................. 140

Fig. 6.2 The mean roughness (Sa) of the nanotubular surfaces as a function of the

applied potential: (a) Ti50Zr alloy, (c) CP-Ti; and the mean water contact angle

(W.C.A.) of the nanotubular surfaces as a function of the applied potential: (b) Ti50Zr

alloy and (d) CP-Ti, Note: Each data point is an average of five measurements ......... 141

xvii

Fig. 6.3 Mineralisation of HA on the nanotubular surface of TiO2-ZrO2-ZrTiO4

fabricated at a), b) 20 V (Di = 37 nm)-as-formed, c), d) 20V (Di = 40 nm)- annealed, e),

f) 25 V (Di = 64 nm)-as-formed, g), h) 25V (Di = 59 nm)-annealed, i), j) 30 V (Di = 76

nm)-as-formed, k), l) 30V (Di = 64 nm)-annealed, m), n) 35 V (Di = 81 nm)-as-formed

and o), p) 35V (Di = 82 nm)-annealed .......................................................................... 147

Fig. 6.4 Mineralisation of HA on the nanotubular surface of TiO2 fabricated at a), b) 20

V (Di = 43 nm)-as-formed, c), d) 20V (Di = 51 nm)-annealed, e), f) 25 V (Di = 52 nm)-

as-formed, g), h) 25V (Di = 45 nm)-annealed, i), j) 30 V (Di = 65 nm)-as-formed, k), l)

30V (Di = 53 nm)-annealed, m), n) 35 V (Di = 68 nm)-as-formed and o), p) 35V (Di =

63 nm)-annealed ............................................................................................................ 148

Fig. 6.5 SEM images of cross section of mineralised HA layers on the surface of a)

TiO2-ZrO2-ZrTiO4 nanotubes and b) TiO2 nanotubes fabricated in aqueous electrolyte

....................................................................................................................................... 149

Fig. 6.6 XRD patterns of: a) TiO2-ZrO2-ZrTiO4 nanotubes anodised on Ti50Zr (α and β

phases) and annealed at 500°C for 3 h before and after immersion in m-SBF for 3

weeks, showing tetragonal anatase, srilankite (a mixture of orthorhombic TiO2 and

ZrO2) and orthorhombic ZrTiO4 with mineralised hydroxyapatite, b) TiO2 nanotubes

anodised on CP-Ti (α phase) and annealed at 500 °C for 3 h before and after immersion

in m-SBF for 3 weeks, showing tetragonal anatase with mineralised hydroxyapatite; and

c) hydroxyapatite ........................................................................................................... 150

Fig. 6.7 The percentage of atomic ratio of Ca/P of mineralised HA on four different

nanotubular surfaces of a) Ti50Zr and b) CP-Ti ........................................................... 151

Fig. 6.8 SEM images showing the transition of soluble salt crystal of present ions of m-

SBF into HA on the nanotubular surface of TiO2-ZrO2-ZrTiO4 with increasing

immersion time in m-SBF ............................................................................................. 153

Fig. 6.9 Schematic presentations of the process of bone-like apatite formation on

nanotubular surface in m-SBF ...................................................................................... 153

Fig. 6.10 The pH value of static m-SBF as a function of soaking time for nanotubular

TiO2-ZrO2 -ZrTiO4 fabricated at 20, 25, 30 and 35V .................................................... 154

Fig. 6.11 Loading-unloading forces versus the nanoindentation depths of HA on

nanotubular surface of TiO2-ZrO2-ZrTiO4 fabricated at the applied potential of 35 V 155

xviii

Fig. 6.12 Dimpled surface of TiO2-ZrO2-ZrTiO4 nanotubes after annealing and

removing the upper cracked layer a) 5 wt % H2O - 20 V, b) 10 wt % H2O - 20 V, c) 5 wt

% H2O - 30 V and d) 10 wt % H2O - 30 V ................................................................... 156

Fig. 6.13 a) The mean roughness (Sa), (Sq) and the mean water contact angle (W.C.A.)

of the nanotubular and dimpled surface of TiO2-ZrO2-ZrTiO4, respectively, Note: Each

data point is an average of five measurements.............................................................. 157

Fig. 6.14 SEM images of mineralisation of HA on the as-formed TiO2-ZrO2-ZrTiO4

nanotubes and dimpled surface fabricated in non-aqueous electrolyte: a, b) and c, d)

high and low magnification-5 wt % H2O-20V, e, f) and g, h) high and low

magnification-10 wt % H2O-20V, i, j) and k, l) high and low magnification-5 wt %

H2O-30V and m, n) and o, p) high and low magnification-10 wt % H2O-30V ............ 159

Fig. 6.15 SEM images of mineralisation of HA on the as-formed and annealed TiO2

nanotubes fabricated in non-aqueous electrolyte: a, b) and c, d) high and low

magnification-5 wt % H2O-20V, e, f) and g, h) high and low magnification-10 wt %

H2O-20V, i, j) and k, l) high and low magnification-5 wt % H2O-30V and m, n) and o,

p) high and low magnification-10 wt % H2O-30V ....................................................... 160


TiO2-ZrO2-ZrTiO4 nanotubes and b) TiO2 nanotubes fabricated in non-aqueous

electrolyte ...................................................................................................................... 161


via anodisation in non-aqueous electrolyte, annealed dimpled surface (at 500 °C for 2 h)

of tetragonal anatase, rutile, srilankite .......................................................................... 162

Fig. 6.18 The percentage of atomic ratio of Ca/P of mineralised HA on a) as-formed and

annealed dimpled surface of TiO2-ZrO2-ZrTiO4 nanotubes and b) as-formed and

annealed TiO2 nanotubes fabricated in non-aqueous electrolyte .................................. 163


TiO2-ZrO2-ZrTiO4 in a non-aqueous electrolyte with 5 and 10 wt % H2O at 20 and 30 V

....................................................................................................................................... 164


nanotubular surface of TiO2-ZrO2-ZrTiO4 fabricated at 10 wt % H2O and the applied

potential of 30 V............................................................................................................ 165

xix

Fig. 7.1 SEM images of annealed TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 10 V, b)

15 V, c) 20 V, and d) 25 V, e) 30 V and f) 35 V........................................................... 174

Fig. 7.2 Mean nanotube size of Di, Do and Wt for annealed TiO2-ZrO2-ZrTiO4 nanotubes

anodised on Ti50Zr as a function of applied potential .................................................. 176

Fig. 7.3 Illustration of changing roughness parameters of Sa and Sq via changing Di of

TiO2-ZrO2-ZrTiO4 nanotubes anodised on Ti50Zr ....................................................... 177

Fig. 7.4 SEM images of SaOS2 cells cultured after 24 h on three different pore sizes of

TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 10 V- low magnification, b) 10 V- higher

magnification, c) 15 V- low magnification, d) 15 V- higher magnification, e) 20 V- low

magnification and f) 20 V higher magnification ........................................................... 182




magnification and f) 35 V higher magnification ........................................................... 183

Fig. 7.6 Schematic illustration of exploring tools of an osteoblast cell on nanotubular

surface ........................................................................................................................... 184

Fig. 7.7 Proposed illustration of the position of integrins on a) Di = 18 ± 6 nm best

nanospacing, b) rough surface of Di = 59 ± 17nm and c) Wt = 24 ± 7 nm most existing

of proteins ..................................................................................................................... 186

Fig. 7.8 Illustration of a) the length of filopodia, b) surface area and c) height of SaOS2

cells after 1 day and 7 days grown on different nanotube sizes (ρ < 0.05) ................... 189

Fig. 7.9 SEM images of SaOS2 cells cultured for 7 days on the four different pore sizes

of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 10 V- low magnification, b) 10 V-

higher magnification, c) 15 V- low magnification, d) 15 V- higher magnification, e) 20

V- low magnification and f) 20 V higher magnification ............................................... 191

Fig. 7.10 SEM images of SaOS2 cells cultured for 7 days on the four different pore

sizes of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 25 V- low magnification, b) 25 V-


V- low magnification and f) 35 V higher magnification ............................................... 192

Fig. 7.11 The percentage of cells attached to the surface of different pore sizes of TiO2-

ZrO2-ZrTiO4 nanotubes after 7 days of cell seeding using MTS assay (ρ < 0.05) ....... 194

xx

Fig. 7.12 Confocal images of stained SaOS2 cells presenting nuclei (blue) and actin

stress fibres (red) after 7 days on TiO2-ZrO2-ZrTiO4 nanotubes with a) Di = 18 nm, b)

Di =30 nm, c) Di =40 nm, d) Di = 59 nm, e) Di = 64 nm and f) Di = 82 nm ................. 195

Fig. 7.13 SEM images of SaOS2 cells cultured for 24 h on the TiO2-ZrO2-ZrTiO4

nanotubes (Di = 40 nm) a) with reduced contamination - low magnification, b) with

reduced contamination - higher magnification, c) with post-treated by HA - low

magnification and d) with post-treated by HA - higher magnification ......................... 196

Fig. 7.14 Illustration of a) the length of filopodia, b) surface area and c) height of

SaOS2 cells after 1 day and 7 days grown on different post-treated nanotubular surface

with the same Di ............................................................................................................ 198

Fig. 7.15 The percentage of cells attached to the different post-treated nanotubular

surface of TiO2-ZrO2-ZrTiO4 with the same Di after 7 days of cell seeding using MTS

assay .............................................................................................................................. 199

Fig. 7.16 SEM images of SaOS2 cells cultured for 7 days on the TiO2-ZrO2-ZrTiO4



magnification and d) with post-treated by HA - higher magnification ......................... 200


stress fibres (red) after 7 days on TiO2-ZrO2-ZrTiO4 nanotubes a) with Di = 40 nm

before post-treating b) reduced contamination, c) with post-treated by HA................. 201

Fig. 7.18 Mean nanotube size of Di, Do and Wt of dimpled surface after annealing and

removing the upper cracked layer fabricated in two different water contents and two

different applied potentials in ethylene glycol, Note: The nanotube size distribution

graphs were generated from 100 nanotubes on different positions for each of three

samples (300 measurements) ........................................................................................ 202


sizes of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 5 wt % H2O - 20 V, b) 10 wt %

H2O - 20 V, c) 5 wt % H2O - 30 V, d) 10 wt % H2O - 30 V ......................................... 205


stress fibres (red) after 7 days on dimpled surface with a) Di = 25 nm, b) Di =29 nm, c)

Di =45 nm and Di =49 nm ............................................................................................. 207

xxi

Fig. 8.1 SEM images of nanoporous Ta2O5: (a) top view (b) the bottom view and the

cross section of multilayer, (c) shorted life nanotubes (arrow), (d) the top view of

multilayer formed on a Ta after anodisation for 120 min in 1M H2SO4 + 3.3 wt %

NH4F, 20V, SEM image of nanoporous Nb2O5 (e) top view and (f) cross section formed

on niobium after anodisation for 16 min in 1M H2SO4 + 3.3 wt % NH4F, 20V and SEM

image of ZrO2 nanotubes (g) top view, (h) cross section and (i) bottom view formed on

a zirconium after anodisation for 95 min in 1M (NH4)2SO4 + 0.3 wt % NH4F, pH=5 and

20V ................................................................................................................................ 219

Fig. 8.2 Demonstration of (a,c,e) changes of roughness (Sa) and (b,d,f) water contact

angle of bare tantalum, niobium, zirconium and titanium, after fabrication of

nanotubular and nanoporous layer, and subsequently annealing, respectively, Note:


Fig. 8.3 XRD patterns of (a) the nanoporous Ta2O5 and bare tantalum foil, (b) the

nanoporous Nb2O5 and bare niobium foil, (c) the ZrO2 nanotubes and bare zirconium

foil and (d) the TiO2 nanotubes and bare titanium foil ................................................. 229

Fig. 8.4 Illustration of pore size of the nanoporous and nanotubular layers ................. 230

Fig. 8.5 (a) low and (b) high magnifications of HA on as-formed nanoporous Ta2O5; (c)

low and (d) high magnifications of HA on annealed nanoporous Ta2O5 ...................... 231

Fig. 8.6 SEM images of HA on a) as-formed and b) annealed nanoporous Nb2O5 ...... 232

Fig. 8.7 (a) low and (b) high magnifications of HA on as-formed nanotubular ZrO2; (c,)

low and (d) high magnifications of HA on annealed nanotubular ZrO2 ....................... 232


nanoporous Ta2O5, b) nanoporous Nb2O5, c) nanotubular ZrO2 and d) nanotubular TiO2

....................................................................................................................................... 233

Fig. 8.9 Bioactivity of biocompatible nanoporous and nanotubular oxide metals after 3

weeks in m-SBF at 37 ˚C .............................................................................................. 234

Fig. 8.10 pH value of static m-SBF as a function of soaking time for: a) nanoporous

Ta2O5 and Nb2O5 and b) nanotubular TiO2 and ZrO2 ................................................... 235

xxii

List of tables

Table 2.1 Scale of sizes of the hierarchical bone components, in comparison to the scale

sizes of the implant structures ......................................................................................... 10

Table 2.2 Microstructures, mechanical properties and biological characteristics of

metallic implant materials ............................................................................................... 13

Table 2.3 Anodisation conditions, characteristics and properties of TiO2 nanotubes on

titanium ........................................................................................................................... 30

Table 2.4 Cell responses to different TiO2 nanotubes on titanium and the anodisation

conditions ........................................................................................................................ 47

Table 3.1 Compositions for the aqueous electrolyte used for fabricating nanoporous and

nanotubular layers ........................................................................................................... 60

Table 3.2 The reagents and their amounts for preparing 1000 mL m-SBF .................... 64

Table 4.1 The distribution of Di, Do and Wt according to normal and Weibull

distribution methods ........................................................................................................ 92

Table 4.2 Surface area index SI and roughness amplitude parameters of Sskw and Sku of

the nanotubular surfaces of Ti50Zr and CP-Ti anodised under various conditions, Note:

Each data point is an average of five measurements....................................................... 99

Table 4.3 Calculated surface energy of TiO2-ZrO2-ZrTiO4 nanotubular surfaces

fabricated on Ti50Zr and TiO2 nanotubular surfaces on CP-Ti, Note: Each data point is

an average of five measurements .................................................................................. 101

Table 5.1 The distribution of Di, Do and Wt according to normal and Weibull

distribution methods ...................................................................................................... 118




xxiii

Table 5.3 Calculated surface energy of as-formed and annealed dimpled surface TiO2-

ZrO2-ZrTiO4 nanotubular surfaces fabricated on Ti50Zr, Note: Each data point is an

average of five measurements ....................................................................................... 130

Table 6.1 Surface area index SI and roughness amplitude parameters of Sq, Sskw and Sku

of the nanotubular surfaces of Ti50Zr and CP-Ti anodised under various conditions,

Note: Each data point is an average of five measurements ........................................... 143

Table 6.2 Calculated surface energies of TiO2-ZrO2-ZrTiO4 nanotubular surfaces

fabricated on Ti50Zr and TiO2 nanotubular surfaces on CP-Ti, Note: Each data point is

an average of five measurements .................................................................................. 145

Table 7.1 Surface index and spacing parameters of TiO2-ZrO2-ZrTiO4 nanotubes

fabricated in different applied potential ........................................................................ 179

Table 7.2 Surface index, spacing parameters and calculated surface energy of TiO2-

ZrO2-ZrTiO4 nanotubes with dimpled surface fabricated in different water contents and

applied potentials, Note: Each data point is an average of five measurements ............ 204

Table 7.3 The percentage of cells attached to the surface of different pore sizes and

roughness parameters of TiO2-ZrO2-ZrTiO4 nanotubes fabricated in organic electrolyte

after 7 days of cell seeding using MTS assay (ρ < 0.05) .............................................. 208

Table 8.1 Surface area and volume index and roughness amplitude parameters of

nanoporous Ta2O5, Nb2O5 and nanotube ZrO2, Note: Each data point is an average of

five measurements ......................................................................................................... 224

Table 8.2 Water contact angle and surface energy of as-formed and annealed

nanoporous Ta2O5, Nb2O5 and nanotube ZrO2 in comparison to its bare metal, Note:


xxiv

List of abbreviations

ALP Alkaline phosphatase

ARB Accumulative roll bonding

at % Atomic percentage

BMD Bone mineral density

BSA Bovine serum albumin

Ca/P Atomic percentage of calcium to phosphate ratio

Ca5(PO4, CO3)3(OH) Carbonated apatite

CaP Calcium phosphate

CF- Concentration of fluorine anion

CP-Ti Commercial pure titanium

Di Inner diameter

Do Outer diameter

EB-PVD Electron-beam physical vapour deposition

ECM Extracellular matrix

EDM Electrical discharging machining

EDS Electron dispersive X-ray spectroscopy

FN Fibronectin

HA Hydroxyapatite

HEPES 2-(4-(2-hydroxyethyl)-1-piperazinyl) ethane sulfonic acid

HMDS Hexamethyldisilazane

ISQ Implant stability quotient values

xxv

L/W Length to width ratio

MC3T3-E1 Mouse osteoblast cells

MG63 Human osteosarcoma cells

mMSC Mouse bone marrow MSCs

m-SBF Modified simulated body fluid

MTS assay (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-

sulfophenyl)-2H-tetrazolium) assay

N2 Nitrogen gas

Nb Niobium

Nb2O5 Niobium pentoxide (niobia)

OCP Octacalcium phosphate

PBS Phosphate-buffered saline

PTES 1H, 1H, 2H, 2H-perfluorooctyl-triethoxysilane

RBA Rapid breakdown anodisation

RBF Rotating bending fatigue

real-time PCR Real-time polymerase chain reaction technique

Sa Mean roughness

SaOS-2 Human osteoblast-like cells

SAS.F Surface area of the actin stress fibre

SBF Simulated body fluid

Sc Core void volume (space roughness parameter)

SEM Scanning electron microscope

xxvi

SI Surface index

Sku Kurtosis (amplitude roughness parameter)

Sm Surface material volume(space roughness parameter)

Sp Largest peak height value from the reference surface within the

defined sampling area (amplitude roughness parameter)

Sq Root mean square (RMS) Roughness

Sskw Skewness (amplitude roughness parameter)

Sv Surface void volume (space roughness parameter)

Ta Tantalum

Ta2O5 Tantalum pentoxide (tantala)

TCP Tricalcium phosphate

TEM Transmission electron microscope

Ti Titanium

Ti50Zr Titanium zirconium alloy with 35 atomic percentage or 50 weight

percentage of zirconium

TiO2 Titanium dioxide (titania)

TiO2-ZrO2-ZrTiO4 Titanium-zirconium-zirconium titanate

TLM Ti-5Zr-3Sn-5Mo-15Nb

TNTZ Ti-Nb-Ta-Zr alloy

VI Volume index

Vol % Volume percentage

W.C.A or θW Water contact angle

Wt Wall thickness

xxvii

Wt % Weight percentage

XPS X-ray photoelectron spectroscope

XRD X-ray diffraction

Zr Zirconium

ZrO2 Zirconium dioxide (zirconia)

1

Chapter 1 Introduction

1.1 Overview

The fabrication, characterisation and application of biomaterials is an interesting field

which has been the focus of many studies in recent years. Although there are debates on

the exact kind of substance that can be classified as a biomaterial, in simple terms they

can be defined as implantable medical devices such as metals, ceramics and synthetic

polymers, biopolymers, self-assembled systems, nanoparticles, carbon nanotubes and

quantum dot, drug and gene delivery systems, tissue engineering and cell therapies,

organ printing and cell patterning, nanotechnology based imaging and diagnostic

systems and microelectronic devices, engineered tissues, cells, organs and even viruses

[1]. One of the applications of biomaterials is the repair of bone tissue, known as

orthopaedic biomaterials or implants [2]. These kind of biomaterials need to have

suitable mechanical and biological properties. Among different biomaterials used for

implants, such as ceramics, polymers, composites and natural products; metal implants

are preferred for load-bearing applications. Metal implants also exhibit excellent

mechanical properties such as Young’s modulus, tensile strength, ductility, fatigue life,

and wear resistance [3]. In addition, they have excellent chemical properties such as

corrosion resistance, biocompatibility and the ability to fuse and harmonise with other

implant materials [4]. These properties make them more suitable for long-term use in

hard tissue applications such as hip and knee joints [4, 5]. Stainless steels and cobalt-

chromium-molybdenum alloys are some examples alongside the more biocompatible

titanium and titanium alloys [2]. However, as bone is alive, an implant has to be

accepted by the bone cells in order to bond with it and to prevent loosening and

inflammation when it is inserted close to the bone. Surface modifications or treatments

can change the surface of the metal implant making it more stable and agreeable for

bone cells.

Surface modifications of metal implants play a major role in biomedical research. One

of the operations of surface modification techniques, either mechanical, physical or

chemical is the addition of substance to the surface such as plasma spray, physical

vapour deposition (PVD) and surface immobilisation of functional molecules. The

interface of the implant and body fluid can also be converted using ion implantation and

electrochemical oxidation. Another way to modify a surface is by removing undesirable

2

substances from the surface by surface machining, grit blasting and acid etching [6].

Based on the functionality and application of the metal implant, one or more surface

modification can be used. Considering the ability of titanium and titanium alloys to

form a protective oxide layer on their surfaces, electrochemical surface treatment is

utilised as a preferable technique.

Titanium, one of the best known transition metals, is of great interest in biomedical

research. It has an incompletely filled d shell in its electronic structure, which allows the

forming of solid solutions with other alloying elements. One of these elements is

zirconium which produces non-toxic β alloys with titanium. This β type alloy offers the

unique characteristic of mechanical biocompatibility such as low elastic modulus,

superior corrosion resistance, and good osseointegration [7, 8]. Titanium shape memory

alloys and porous titanium alloy are also of great interest because of their shape memory

function, or super-elastic function and bone tissue ingrowth into the porous network,

respectively[4, 9]. These metal implants are applied as plates, nails, screws and endo-

prostheses [8]. The protective oxide layer which forms naturally on the surface of

titanium and titanium alloy makes them bioinert. Therefore, electrochemical surface

modification such as anodisation is one of the suitable techniques available to make

them bioactive.

The significance of a surface treatment such as anodic oxidation, [10] which provides

the possibility of controlled nano-scale fabrication with suitable physico-chemical

properties on the metal surface, has been reported [11]. Highly ordered nanoporous

arrays of titanium dioxide that form on the titanium surface by anodic oxidation are

receiving increasing research interest due to their effectiveness in promoting

osseointegration. Uniform and controllable nano-patterned surfaces of titanium can be

fabricated using electrochemical anodic oxidation, which is an economical , simple and

versatile technique [12]. By applying a constant potential to titanium as a working

electrode in an electrochemical cell with a platinum counter electrode and a standard

reference electrode, the titania nanotubes can be obtained. Different TiO2 nanotube

diameters, wall thicknesses and lengths can be achieved by controlling the condition of

the anodising process, such as the applied potential, current density, anodisation time,

concentration of fluorine anion and type (aqueous and non-aqueous), pH value and the

3

viscosity of the electrolyte [13]. Such nano-scale surfaces exhibit similar properties to

those of physiological bone.

The important factors for cell response can be better controlled by understanding the

mechanism of cell attachment to the biomaterial surface. Various populations of cells,

cytokines, growth factors and the extracellular matrix affect the process of tissue

healing around implants. For example, chemotactic factors are essential for cell

recruitment and cell adhesion, and growth factors and cytokines are effective for the

proliferation and differentiation of cells. Protein adsorption at the implant surface

initiates interactions between cells and implants through a complex series of adsorption

and displacement steps. The interaction of cell and implant is followed by the

attachment phase, which is governed mainly by van der Waals forces. The next step,

termed the adhesion phase, is surface anchoring through fibronectin and vitronectin for

the formation of focal points at cell-membrane integrins. Filopodia and finger-like

protrusions are then produced to enable sensing of the optimum anchorage and

spreading toward the surface [14, 15].

In the specific instance of biomaterial applications, the highly ordered nanostructured

TiO2 promotes osseointegration because the size of integrins and focal contacts of cell-

implant interactions is also in the nano scale (~ 10 nm) [11, 16]. The effects of different

inner and outer diameters of nanotubes and the spaces between them on bone cells

during cell culture have been studied extensively [17-20]. The crystalline form of TiO2

nanotubes exhibited an encouraging effect on cell behaviour due to their enhanced

hydrophilic properties [21]. Furthermore, growing hydroxyapatite on the surface of

these so-formed nanotubes would improve cell behavior such as adhesion, proliferation,

morphology, migration, survival and differentiation due to beneficial changes in the

topographical and chemistry properties [22-25]. Cell biological responses of osteoblasts

on anodised titania zirconia nanotubes [26], the effect of surface energy for its

functionality [27], as well as the antibacterial properties of different sizes of titania

zirconia nanotubes [28] have also been investigated.

Several in vitro studies have examined the optimum nanotopography for cell response

and suggested that the optimum diameter of TiO2 nanotubes is less than 100 nm. The

15 nm diameter nanotubes revealed the best cell response; although there is controversy

[19, 20] regarding the optimum diameter required. The behaviour of the bone cells on

4

the nanotubular of oxide layer on the surface of Ti50Zr needs to be clarified.

Specifically, further studies are required to determine the optimum size, morphology,

physical and chemical properties of nanotubes on the biocompatible titanium-zirconium

(Ti50Zr) binary alloy for the adhesion of bone cell sensing elements. The current work

aims to improve the physicochemical and other properties of the nanotubular layer by

fabricating it on the surface of β alloy of Ti50Zr.

1.2 Thesis objective

This thesis was aimed at imparting new functionalities to the biocompatible Ti50Zr

alloy through surface modification. A nanotubular layer was anodised on the alloy with

enhanced bioactivity and biocompatibility because the nanoscale characteristics of the

nanotubes is able to interlock with bone cells, therefore improving cell adhesion,

proliferation and growth on the nanotubular surface. Extensive research has been

undertaken to fabricate and characterise the nanotubular layer on the surface of CP-Ti

and assess its bioactivity and biocompatibility. Although some research has been

undertaken to fabricate the nanotubular layer on the surface of Ti50Zr, few attempt have

been made to characterise the physical, chemical and mechanical properties of

nanotubular layer of TiO2-ZrO2-ZrTiO4 in respect of their influences on their bioactivity

and bone cell integration.

To accomplish the aforementioned objectives the following tasks were completed:

- Two different types of electrolytes were used to fabricate nanotubes with

different morphologies and physicochemical and mechanical properties.

- The effects of the applied potential and the concentration of fluoride ions in the

electrolyte on the morphology and other characteristics of the nanotubes were

systematically studied.

- The ability to induce hydroxyapatite formation was evaluated by immersing the

substrates with a nanotubular surface in the modified simulated body fluid.

- The cell adhesion was assessed by seeding osteoblast cells on the nanotubular

surfaces of Ti50Zr and CP-Ti.

Major contributions of the thesis include:

5

The TiO2-ZrO2-ZrTiO4 nanotubes fabricated on the surface of Ti50Zr alloy not only

revealed a distribution of nanotube sizes following the needle-like microstructure of the

alloy, but also exhibited a variety of nanotube height. These difference in the height of

nanotubes resulted in an interesting surface topography with a range of amplitude and

spacing roughness parameter. The link between roughness parameter and hydrophilic

properties of the nanotubular layer were evaluated. Hereby the surface of Ti50Zr alloy

was nanoengineered by fabricating the TiO2-ZrO2-ZrTiO4 nanotubes while the factors

which influence cell attachment were taken into consideration. This demonstrates that

the nanotubular layer not only patterned the surface in nano-scale but also created a

special topography in micro-scale. The surface chemistry and mechanical properties of

the substrate are also changed by the nanotubular layer. The TiO2-ZrO2-ZrTiO4

nanotubes were more bioactive than TiO2 nanotubes. The bone cell attachment

significantly increased on two different nanotubular layer of TiO2-ZrO2-ZrTiO4

fabricated in aqueous and non-aqueous electrolyte.

1.3 Thesis structure

Chapter 2 reviews the literature on anodisation of titanium, titanium alloy and transition

metal alloying elements to fabricate nanoporous and nanotubular layer after an

introduction of bone structure, metal implants and their surface modification. It presents

parameters which affect the morphology, topography, physical and chemical properties

of nanotubes. Then it describes in simple terms the mechanism of cell adhesion and the

factors that influence the adhesion of the bone cells on implant surface. Finally, it

highlights the characters of these nanotubular layers which influence bone cell

behaviour. Chapter 3 provides a methodological description of the materials and

methods for fabrication, characterisation and performance of the nanotubular layer. The

materials and methods of individual chapters are described as required in the relevant

chapter to avoid over-complication.

Chapter 4 presents the effect of anodisation parameters on the characteristics of the

TiO2-ZrO2-ZrTiO4 nanotubes fabricated in aqueous electrolyte such as inner, outer

diameter, wall thickness, roughness parameters, mechanical and hydrophilic properties

and surface energy. Chapter 5 presents the effect of anodisation parameters on the

characteristics of the TiO2-ZrO2-ZrTiO4 nanotubes fabricated in non-aqueous electrolyte

such as inner, outer diameter, wall thickness, roughness parameters, mechanical and

6

hydrophilic properties and surface energy. These characteristics are compared with TiO2

nanotubes fabricated on CP-Ti in the same condition to outline the significance of TiO2-

ZrO2-ZrTiO4 nanotubes.

Chapter 6 studies the effect of the characteristics of the nanotubular layer fabricated in

aqueous and non-aqueous electrolyte on their ability to induce hydroxyapatite. These

characteristics are compared with TiO2 nanotubes fabricated on CP-Ti in the same

condition. Chapter 7 investigates the cell adhesion on the surface of these nanotubular

layers and the influence of the morphology of the layer as well as the roughness

parameter, hydrophilic properties and the role of post-treatment on the percentage of

cell attachments is studied.

Chapter 8 presents the nanoporous and nanotubular layer fabricated on the surface of

some biocompatible transition metal such as tantalum, niobium and zirconium in

comparison to titanium. The characteristics of these nanoporous and nanotubular layer

such as roughness parameter, hydrophilic properties and surface energy is studied. The

ability of such a layer to induce hydroxyapatite is also assessed.

7

Chapter 2 Literature review

2.1 Introduction

Metal implants are the best choice for the long-term replacement of hard tissue, such as

hip and knee joints, because of their excellent mechanical properties. Titanium and its

alloys, due to their self-organised oxide layer that protects the surface from corrosion

and prevents ion release, are widely accepted as biocompatible metal implants. They

must also be accepted by bone cells, bonding with and growing on them to prevent

loosening. Surface modification is essential for the promotion of the osseointegration of

these biomaterials. Nanotubes fabricated on the surface of metal implants by

anodisation are currently receiving increasing attention for surface modification. The

response of bone cells to implant materials depends on the topography,

physicochemistry, mechanics and electronics of the implant surface and this influences

cell behavior, such as adhesion, proliferation, shape, migration, survival and

differentiation; for example the existing anions on the surface of a titanium implant

make it negative and this affects the interaction with negative fibronectin (FN).

Although optimal nanosize of reproducible titania nanotubes have not been found

significantly according to different protocols in the studies cell response has been

shown that were more sensitive on titania nanotubes with nanometer diameter and inter

space. By annealing amorphous TiO2 nanotubes change to a crystalline form and

become more hydrophilic, resulting in an encouraging effect on cell behaviour. The

crystalline size and thickness of the bone-like apatite that forms on the titania nanotubes

after implantation are also affected by the diameter and shape. This chapter provides an

overview of the employment of the anodisation for nanotubes fabricated on the surface

of titanium, titanium alloys and titanium alloying metals such as niobium, tantalum and

zirconium metal implants. It explains anodic oxidation and the manner by which

nanotubes form on the surface of the metals. It then assesses this topical research to

indicate how changes in anodising conditions influence nanotube characteristics such as

tube diameters and nanotube-layer thickness. Finally it describes how changes in

nanotube morphologies, such as the tube diameter, the thickness of the nanotube layer,

and the crystalline structure, influence the response of cells.

8

2.1.1 Bone: structure, composition and properties

Bone, as with other parts of our body, becomes damaged or weakened because of age,

accidents or disease. This damage, which includes bone fractures, low back pain,

osteoporosis, scoliosis and other musculoskeletal problems, usually occurs in elderly

people but not exclusively. Biomaterials such as implants are used for repairing injured

bones, cartilage or ligaments and tendons [2].

Weiner et al. [29] have described the basic bone composition that consists mostly of

fibrous protein collagen, carbonated apatite (Ca5(PO4, CO3)3(OH)) and water. The

crystal size and the proportion of these components change over time. As a result,

younger bones replace older bones. Uddin et al. [30] have emphasised that bones are

tissues that are alive and growing. Bone forming cells such as osteoblasts are

responsible for generating the synthesis and deposition of calcium phosphate crystals

that are required to confer hardness and strength in biomineralisation. A schematic

illustration of the hierarchical structure of cortical bone is shown in Fig. 2.1. It can be

seen that cortical bone contains many different structures that exist on several levels of

scale starting from sub-nanostructures. Mour et al. [31] described bone as a solid

material that is highly porous on a micrometre scale. Bone is a viscoelastic material due

to these pores, which are filled with fluid and cells; osteoblasts, osteoclasts, osteocytes

and bone lining cells that are regenerative. Bone, moreover, has values of compression

strength several times higher than, for example, concrete, but its low density is in the

range of aluminium. Table 2.1 lists the size scales of the hierarchical bone components,

in comparison to the size scales of the implant structures.

9

Fig. 2.1 Hierarchical structure of cortical bone (adapted from [32])

10

Table 2.1 Scale of sizes of the hierarchical bone components, in comparison to the scale

sizes of the implant structures

Structures Size

Ref. Micrometer Nanometer

Tissue

component

Osteon 10-500 - [32]

Osteoclast (lacunae) 100 or more - [33]

Osteoblast 20-30 - [33]

Lamella 3-7 - [32]

Collagen fibril 0.5 - [32]

Cell membrane - 10-100 [34]

Plate-like apatite crystals - 50×25×3 [32]

Integrins - 8-12 [35]

Proteins - 1-10 [34]

Collagen molecule - 1 [32]

Amino acids - 0.1-1 [34]

Water molecule - 0.1 [34]

Metal

Implant

Roughness of the surface >100 - [34]

titanium grain sizes 10-20 - [36]

Metal oxide - 5-15 [34]

Atom - 0.1 [34]

11

2.1.2 Bone implant materials

There are different types of implant materials, such as polymers, ceramics, metals,

composites and natural products. Metal implants can be load bearing up to several

decades and consequently are used as endoprostheses because of their mechanical

attributes. However, they need to be inserted into the human architecture and become

fixed and stable in the surrounding natural tissue. Furthermore, these implants need to

be bioinert within the highly corrosive and demanding environment of the human body.

Polymers, in comparison to the metals, give rise to inflammation because of monomers

intrinsic to their structure or that become available through infection; thereby creating

likelihood of degradation. However, polymers demonstrate excellent primary fixation

[2].

Ceramics and bioglass exhibit lower fracture toughness and higher elastic modulus than

bone, and also demonstrate property variations with respect to their formulation [37].

Their mechanical and biological properties depend on many factors during synthesising

such as maximum temperature, duration of the thermal steps, purity of the powder, size

and distribution of the grains and porosity; many of which cannot be accurately

controlled [2].

Biomaterials have progressed through three generations from the first, whose

particularities were matching its physical properties as a tissue replacement with the

least toxicity or biological inertness to the second generation that demonstrate bioactive

behaviour [38]. Surface treatments have been used to improve the bioactive nature of

these biomaterials, especially metals such as titanium and titanium alloys that change

the physico-chemical, mechanical and electrical properties of their surfaces [7]. The

third generation biomaterials are intended to promote specific cellular responses at the

molecular level [38]. Most of the metal implants used are stainless steels, cobalt

chromium alloy, and titanium and its alloys. SUS 316L stainless steel, with a Young’s

modulus of approximately 160 GPa, is the only stainless steel that is biocompatible;

however, its wear resistance is poor in comparison with other metal implants. When a

stainless steel alloy consists of nickel, it may possibly be allergenic [4] since the release

of metal ions can cause an adverse reaction due to corrosion and the soluble corrosion

products. Similar to stainless steel, cobalt alloys consist of nickel and chromium, and

consequently fabrication of nickel free stainless steels and cobalt chromium alloys is

12

under investigation since its corrosion resistance and mechanical properties are superior,

with the exception of the elastic modulus that is the same range as stainless steel [4].

When a metal implant material is able to form a stable and compact oxide layer that can

rebuild itself, then it can be described as bioinert since this surface can prevent

transmission of undesirable ions [7]. Other properties of a metal, which endow its

mechanical biocompatibility, are Young’s modulus, tensile strength, ductility, fatigue

life, fretting fatigue life and wear resistance [3]. As a result of the contact between the

inserted implant and bone, cyclic loading wear loss and fretting fatigue may occur and

the implant may become loose. Some oxide layers on metals increase the wear

resistance. A metal implant that has a lower Young’s modulus can homogeneously

transfer stress between itself and bone. However, as the modulus approaches that of the

bone, the strength of the implant will be lower than bone, and the possibility of failure

under large shear deformation will increase.

In vivo data show a smooth fracture callus for metals with a lower Young’s modulus

than that of bone. An in vivo fatigue test of implants is difficult due to the need to gather

data from tests that identically replicate in vivo conditions [3]. A standardised in vitro

fatigue test is therefore used. This includes tension/compression, bending, torsion, and

rotating bending fatigue (RBF) testing of the metal implant, which is then compared

with Ti6Al4V as a standard material [5]. Titanium and titanium alloys are more

biocompatible than other implant metals for the abovementioned reasons and these

metals have demonstrated ability to integrate with bone. The microstructures,

mechanical properties and biological characteristics of commonly used metallic implant

materials are summarised in Table 2.2.

13

Table 2.2 Microstructures, mechanical properties and biological characteristics of metallic implant materials

Alloy designation Microstructure

Nominal composition (wt%)

Ultimate

strength (MPa)

Elastic

modulus

E (GPa)

Yield

strength

(MPa)

Ref. Log polarisation resistance- Distinct form

Al/Cr V/Ni Mn/Ta Zr/Nb Others

Stainless steel 316L Austenite Cr=16-18

Ni=10-

14 Mn=2 - Mo=2-3

465-950 200 170-750 [5, 39]

7-8/ sequestration

Co-25Cr-6Mo Austenite(fcc)+hcp Cr=27-30/ Al=0.1 - - - Mo=5-7

600-1795 200-230 275-1585 [5, 39]

Titanium grade 2

(ASTM) Unalloyed

- - - - - 345 102.7 275 [39, 40]

- - - - -

Ti-6Al-4V α/β

Al=6 4 - - Fe=0.3

960-970 110 850-900 [5, 39] 4-5/ sequestration

5-6/

toxicity - -

4-5/

sequestration

Ti-6Al-7Nb α/β Al=6 - - Nb=7 -

900-1050 114 880-950 [39, 40] 4-5/ sequestration - - 6-7/

14

Inertness

Ti-Zr Cast ά/β

- - - Zr=50 -

900 - - [5, 39] - - -

7-8/

Inertness -

Ti-30Ta α″

- - Ta=29.6 - -

587 69 - [39, 41] - -

6-8

inertness - -

Ti-30Nb

- - - Nb=30 -

> 600 More than

120 530 [39, 42]

- - - 6-7/

Inertness -

Ti-29Nb-13Ta-4.6Zr Metastable β

- - Ta=13 Zr=4.6/N

b=29 -

911 80 864 [39, 40]

- - 6-8

inertness

7-8 and

6-8

Inertness

-

Bone Viscoelastic

composite 90-140 10-40 - [7, 39]

15

Researchers have paid more attention to titanium and its alloys for load bearing and

long term implants because of their good fatigue strength, relatively low elastic

modulus, low density, excellent corrosion resistance and biocompatibility; the last

property being reflected in low release of ions [39, 43]. Pure titanium and Ti-6Al-4V

ELI (Extra Low level of Interstitial content) are the prime metals used for biomedical

applications. However, aluminium is classified as belonging to a group of materials that

forms a capsule, or scar tissue, and vanadium is cytotoxic, and both can have

mutagenicity, cytotoxicity and allergic reactions. Research is active in developing

biocompatible Ti-Nb-Ta-Zr alloys. Recent β type titanium alloys have a lower elastic

modulus (Young’s modulus approximately 50 GPa) than commercial pure titanium and

Ti-6Al-4V (Young’s modulus approximately 105 and 110 GPa, respectively), which are

α and α+β types. The stability of β phase in titanium alloys depends on the alloying

element ratio and the β-transus temperature of the alloy. It has been established that

niobium and tantalum act as β-stabilisers and β phase exhibits a lower elastic modulus

in titanium alloys. In binary titanium alloys, it has been shown when β stabiliser

elements increase, the stability of β phase and also elastic modulus increase [4, 44-46].

2.1.3 Surface treatment for implant materials

The surface needs modification to optimise the properties of an implant and maximise

bioactivity when interfacing with natural tissue. The biological characteristics of

implants can be enhanced by adding material with desired properties, changing the

composition or removing unwanted material from the implant surface. These methods

are known as surface treatment or modification and can be classified into four

categories: mechanical, physical, chemical and biochemical surface modifications.

Machining, grinding, polishing and blasting are mechanical surface modifications.

Physical surface treatments include thermal spray, physical vapour deposition, ion

implantation and deposition, and glow discharge plasma. Chemical surface

modifications consist chemical treatment (acid, hydrogen peroxide, alkaline), anodic

oxidation, sol-gel and chemical vapour deposition. Some techniques have been

introduced for titanium and titanium alloys such as silanised titania, photochemistry,

self-assembled monolayers, protein resistance and protein immobilisation [6, 7].

A thin 2 - 5 nm surface layer of titania (TiO2) forms when titanium is exposed to air and

protects the surface from corrosion [47]. The layer that is in contact with the substrate

16

consists of TiO; the intermediate layer is Ti2O3 in composition and the outer layer is

TiO2. The titania outer layer is bioinert and cannot bond to bone cells easily since it is

encapsulated by fibrous tissue that isolates it from the bone cells. Such implants,

therefore, tend to become loose and are not recommended for lengthy periods [7]. On

the other hand, ceramics, glass and several other implant materials can bind and form

bone directly through the layer of apatite that forms on their surfaces [48]. Thus, surface

treatments are needed for titanium alloys to achieve better and more rapid bonding to

bone; i.e. osteoinduction. The optimum surface treatment would create a stable passive

layer on the surface of titanium implants with essential properties such as roughness,

wettability and electrostatic charges.

2.2 Anodic oxidation as a metallic implant surface treatment

Electrochemical anodic oxidation has been used to grow a thick and uniform oxide layer

on metals for a decade or so. It has been shown to play an excellent role in the

biocompatibility of metal implants [47]. A controllable and desired thickness is formed

by putting the metals that demonstrate a spontaneous protective oxide layer as the anode

of an electrochemical cell. An ordered oxide layer can be tailored when the type of

electrolyte, applied current density, electrolyte concentration, electrolyte temperature,

agitation speed, and cathode to anode surface area ratios are controlled [49]. It was

initially demonstrated that a nanoporous oxide layer could be formed on aluminium [50]

in such kind of conditions; thus seeding the fabrication of the same layer on titanium

[10], niobium [51], tantalum [52] and zirconium [53] has been undertaken. These are

termed as the “valve metals” that are able to form dense, stable, tightly adhering,

electrically insulating oxides on their surface.

A nanotube titanium oxide layer with a controlled and uniform diameter can be obtained

by anodising oxidation, which is a cost effective, versatile and simple technique. This

nanotube titania layer plays an important role in the enhancement of osseointegration

through improving the adhesion of the hydroxyapatite (HAP) coating that is deposited

onto TiO2. Oh et al. [54] indicated that due to the mechanical interlocking between the

HAP coating and the nanotube titanium oxide layer the cell adhesion could be improved

by up to 400 %.

17

Sibert [55] reported on the formation of a thick, adherent film of TiO2 on titanium

surface using anodic oxidation when the nature and concentration of electrolyte,

forming voltage, current density, and temperature were properly adjusted [55].

Mechanisms for apatite formation on TiO2 on titanium surfaces were indicated by

Kukubo [56]. Zwilling et al. [10] demonstrated the growth process of TiO2 nanotubes

on pure titanium and Ti6Al4V using anodisation in chromic acid (CA) electrolytes, with

or without the addition of fluorine species. The diameter and thickness of the nanotubes

were determined by the applied potential and time. Anodising can form TiO2 nanotubes

that are open at the top and closed at the bottom and of different sizes by changing the

concentration and type of electrolyte, the time and the voltage.

A standard two or three electrode setup can be employed for anodising the sample,

mostly commercial pure titanium. There are combinations of electrolytes: (i) an aqueous

acidic solution containing F-, (ii) an aqueous buffered solution, or (iii) a non-aqueous

electrolyte containing F- ion with or without a trace amount of H2O. Initially, compact

titanium dioxide will be created [57], as illustrated in Fig. 2.2.

18

Fig. 2.2 Growth of regular TiO2 nanotubes, a) cathodic reaction, b) anodic reaction, c) transition state of TiO2 layer, d) starting of nanotube

formation and e) titania nanotubes

19

Nanotubes then form as soluble fluoride complexes through the chemical dissolution of

the oxide in the presence of the fluoride ion and there will be an increase on current. A

stage is then reached where there is equilibrium between oxidation and dissolution; at

which point a nanostructured surface forms and the current flow is constant [57, 58]. A

schematic current-time curve is shown in Fig. 2.3.

Fig. 2.3 Schematic current-time curve (adapted from [58])

All chemical reactions respond to the in situ physical and chemical conditions. The

anodic oxidation mechanism is governed by the type and concentration of the

electrolyte, the applied potential or the applied current density, the temperature, and the

time of the electrochemical reaction. By changing one of these conditions when the

others remain constant, different morphologies of TiO2 nanotubes can be obtained on

commercial pure titanium. It has also been reported that the facing position of the

tubular shape working electrode (sample) towards the counter electrode influenced the

nanotube inner diameter and length. The nanotubes exhibited different diameters and

lengths at both inner and outer surface of the electrode, whether they were on the side

20

where faced the counter electrode or they were on the opposite side. The nanotubes

exhibited different diameters and lengths where they grew at the bottom or at the top of

the abovementioned tubular shaped working electrode. These observations were due to

the potential drop in the electrolyte, the field screening effect inside the tubular and

diffusion process [59]. Two different TiO2 morphologies; (i) the classic ordered

nanotubes and (ii) a nanoscale sponge layer were obtained when the rotated working

electrode was used [60].

2.2.1 The influence of the type and concentration of aqueous electrolyte on TiO2

nanotubes

When the anodisation of titanium takes place in an acidic solution, most of the

composition of hydrofluoric acid is in HF form. The diameter of the TiO2 nanotube that

grows in a HF-based electrolyte lies between 15 to 140 nm under conditions where

H3PO4 and H2SO4 were used and the applied potential changed from 1 to 25V [57, 61].

The optimum voltage was normally 20 V and the time of the anodisation, which had an

influence on the thickness of the oxide layer, was 12 h; a maximum length of ~ 500 nm

of TiO2 nanotube has been obtained [62]. The amorphous structure of nanotube arrays

was changed to anatase at 450 C inside the tube wall and it was entirely changed to

rutile at 600 C by annealing [63]. These phases exhibited different influences on the

apatite formation and/or cell fate [48, 64]. Titanium-based layers are more hydrophobic

than other layers formed in a buffered electrolyte. The hydrophilic value at the interface

of the implant surfaces with tissue is a crucial factor in cell adhesion and the best

hydrophilic value was reported to be 16 [65]. Fig. 2.4.a shows SEM images of TiO2

nanotubes (top view and cross-section) formed in 1 M H3PO4 + 0.3 wt % HF at 15 V

[61].

21

Fig. 2.4 SEM and TEM images of TiO2 nanotubes (top view, cross-section and bottom):

a) formed in 1 M H3PO4 + 0.3 wt % HF at 15 V [61], b) in 1 wt % HF at 20 V for 15

and 30 min) [66], c) in 1 M (NH4)2SO4 + 0.5wt % NH4F at 20V [67], d) in ethylene

glycol + 0.3 wt % NH4F + 2 vol % H2O [68]

Hydrofluoric acid selectively etches the surface of some metals. Therefore, the presence

of fluoride ions within the electrolyte during growth of titanium oxide allows the

formation of a nanotube oxide layer. A TiO2 nanotube of 70 - 100 nm in diameter forms

in less than 2 h when anodisation occurs in an electrolyte consisting of only HF at an

applied potential of 20 V [65, 66]. There is a certain concentration range of F- in which

regular uniform nanotubes can be formed; and beyond which nanotubes cannot be

uniformly fabricated. The nanotube diameter increases with an increase in voltage. It

has been reported that an oxide layer of TiO2 nanotube with a thickness of ~ 250 nm has

been formed in HF electrolyte [64, 69]. A combination of an amorphous and crystalline

22

structure for the TiO2 was observed; and the contact angle of the TiO2 nanotube that

grows in this type of electrolyte is still higher than the best proposed hydrophilic value

Fig. 2.4.b SEM and TEM images of TiO2 nanotubes (top view, cross-section and

bottom) formed in 1 wt % HF at 20 V for 15 and 30 min [66].

The bioactivity of the titanium oxide can be enhanced by immersion in a NaOH solution

so that sodium titanate nanostructures will be formed on the top edge of the nanotube

wall. Another process is to heat the TiO2 nanotube so that anatase will form; onto which

a nanostructured hydroxyapatite (HAP) phase can be created through soaking into a

simulated body fluid. The presence of the nanostructured TiO2 accelerates the formation

of HAP [64]. The adhesion of MC3T3-E1 mouse osteoblast cells was significantly

promoted with the filopodia of growing cells actually going into the nanotube pores,

producing an interlocked cell structure [54].

One proposal for the prevention of bacterial infection following orthopaedic

implantation is local antibiotic therapy, such as using bone cement or a polymer coating

that is loaded with a therapeutic agent [70]. It is possible to decrease the susceptibility

of the implant to bacterial infection due to limited chemical stability or controlled

kinetic release of the therapeutic agent. It is proposed that TiO2 nanotubes, following

phase transformation to anatase by annealing, can be loaded with antibiotics; for

instance 200, 400 and 600 µg of gentamicin, to minimise the initial bacterial adhesion

[70].

Schmuki et al. [67] have observed the electrolyte dissolubility and hydrolysis ability of

a titanium oxide layer and have indicated that smooth and longer nanotubes can be

formed in buffered electrolyte [((NH4)2SO4+NH4F) or (Na2SO4+NaF)]. The pH profile

within the tube was constructed to evaluate the effect of the electrochemical condition

on the geometry of TiO2 nanotubes using anodic oxidation in buffered electrolyte. The

dissolution current was adjusted and, as the electrolyte was neutral, acid was created at

the pore bottom. The pore thickness, regularity roughness, diameter and the crystalline

shape change with the variations in the concentration of F- in electrolyte, the sweep

potential rate and the applied potential. Under these conditions TiO2 nanotubes grew up

to ~ 2.5 µm, even 4 µm, in thickness and ~ 100 nm in diameter, based on a 20 V applied

potential [67, 71, 72]. The tube diameter is affected by the voltage but it is not affected

by time. However, the time of anodisation influences the thickness of the nanotube

23

layer. Fig. 2.4.c SEM images of TiO2 nanotubes (top view, cross-section and bottom)

formed in 1 M (NH4)2SO4 + 0.5wt % NH4F at 20V [67].

Apatite formation was evaluated on an annealed nanotube layer with a diameter of ~

100 nm and a thickness of 2 µm [48]. In comparison to a flat surface, such a nanotube

surface encourages the formation of apatite. The rate of apatite formation depends on

the thickness of the layer and the nanotube structure. Apatite forms more dominantly on

thick and crystalline nanotubes [48]. In cases where the oxide layers were not uniform,

precipitation was observed in buffered electrolytes consisting of K+ and Cs+ whereas no

precipitation was observed in Na+ and NH4+ electrolytes; thus, the electrolyte salts have

to be chosen carefully regardless of the importance of the final pH [73].

A TiO2 nanotube layer with a maximum layer thickness of 1 µm was formed in acidic

electrolyte consisting of fluoride salt at an applied potential of 20 V [74]. When the

length of the 35 - 70 nm tubes was increased, the Young’s modulus of TiO2 which was

approximately 36 - 43 GPa decreased because substrate as well as the functionality of

the coating influence measured Young’s modulus and hardness [75]. The modulus of

elasticity of nanotube oxide layer is more close to that of the substrate if it has lower

thickness or if it is closer to the substrate. An apatite coating can be obtained on the

TiO2 nanotube layer after immersing such an annealed layer in a solution with a high

calcium and phosphate concentration. A carbonated apatite crystal with a column-like

shape of nanometer size was formed, and the bond strength between the HA coating and

the nanotube layer was measured to be higher than 15.3 ± 2.5 MPa [74, 75].

2.2.2 The influence of the non-aqueous electrolyte on TiO2 nanotubes

An organic electrolyte has a low amount of oxygen in comparison to an aqueous

solution; therefore, the oxide chemical dissolution in an organic electrolyte depends on

the water concentration [68]. As a result, a thicker nanotube layer will form over a

longer time in an organic electrolyte and its viscosity, whether stirring [76] and the H2O

content influence the nanotube formation. Nanotubes with a diameter of 20 to 160 nm

were obtained at an applied potential ranging from 30 to 120 V using an organic

electrolyte [68, 76-79]. TiO2 nanotubes with a maximum height of 45 µm [68] or higher

(>150 µm) [77] in electrolytes such as glycerol [76, 80], ethylene glycol [68, 78] and

acetic acid [79] were formed after 1 to 15 h anodisation [68]. However, at short

24

anodisation time the tubes were connected by a ring around their outside walls, and with

a longer anodisation time they were smooth and not connected because of dissolution of

the rings [79]. A smooth layer of TiO2 nanotubes that were not connected by outside

walls was formed when NH4F in glycerol was used as the electrolyte [80]. Such an

electrolyte exhibits a lower diffusion constant, which has an influence on the pH value

of the pore tip and consequently on the current transient and dissolution rate. The

viscosity of the electrolyte is important too because in highly viscose electrolyte

diffusion constant decreases. The diameter of these nanotubes was 40 nm with a 7 µm

length and aspect ratio of 233 - 117 [80].

The shape of the nanotubes is influenced by the applied potential and the anodisation

time in a water-free electrolyte consisting of acetic acid and NH4F. Under low potential

conditions (i) the nanotube diameter is about 20 nm, (ii) the maximum length will be

100 nm and (iii) rings outside the walls connect the tubes [79]. Nanotubes with a coral

reef shape form and the rings dissolve when the applied potential and anodisation time

are increased. If the nonaqueous electrolyte consisting of glycerol and NH4F, and it is

stirred, the adhesion strength of the TiO2 nanotube (1.2 - 1.9 µm in thickness) to the

substrate decrease, and the contact angle with water ranges from 58 - 84 [76]. Fig.

2.4.d SEM images of TiO2 nanotubes (top view, cross-section and bottom) formed in

ethylene glycol + 0.3 wt % NH4F + 2 vol % H2O [68].

One interesting application of amphiphilic TiO2 nanotubes is the fabrication of a drug

release system. A hydrophobic cap using organic monolayer grafting was fabricated on

a TiO2 nanotube, which was treated using the anodising method in the presence of the

fluoride ion [81]. Another anodising process was then undertaken to fabricate a

hydrophilic TiO2 nanotube underneath the initial layer; and the hydrophilic segment was

loaded with hydrophilic drugs. Detachment of the cap would be performed using the

UV induced photocatalytic nature of TiO2; thereby permitting controlled drug release

[81].

2.2.3 The effect of pH value on the formation of TiO2 nanotubes

The pH value influences the dissolubility of the electrolyte and the hydrolysis ability of

titanium oxide layer. This behaviour has been surveyed for titanium foils that were

anodised in three electrolytes of different pH values and the mass loss of the anode

25

during the anodic oxidation was based on the current-time curve [82]. The time of

nanotube formation increases with increasing the pH; consequently the nanotubes have

a longer length compared to nanotubes formed in electrolytes with a lower pH [82]. A

length of 4.4 µm was achieved when the pH value was high but still acidic (pH=4.5)

[83]. As mentioned above, different tube sizes can be expected when the conditions of

the electrolyte and its pH, the applied potential and the anodisation time change.

Fabrication of TiO2 nanotubes can be classified as four generations based on the

anodising electrolyte: acidic, buffered, polar organic electrolytes and nonfluoride base

electrolyte [84]. An acidic electrolyte provides a suitable condition for an anodic

oxidation process. The dissolution rate based on nanometer per second can be

determined by measuring the XPS depth profiles when an anodised oxide layer etched

in different pH values. The highest dissolution rate matched the lower pH value. The

changes in pH values were observed during electrochemical anodisation [72].

To combine the in-growth of bone into micro/macro-sized (~ 1 to 20 µm) and

enhancing biological interaction by nanostructure (nanotubes with ~ 50 nm diameter)

Crawford et al. [85-87], recently fabricated a dual-porous hierarchical structure on

titanium using a simple anodic oxidation process. The anodisation time and pH

influence the nanotube layer length and random distribution of microscopic pores size

and density [86]. The fact that TiO2 nanotube growth can be affected by the orientation

of the titanium substrate have been reported by this group [87, 88]. They also have

investigated the mechanical properties and deformation behaviour of TiO2 nanotubes to

great detail and have measured the modulus of the nanotube coating [87, 89]. In another

study micro- and nano-patterned (MNP) surface was prepared via anodisation of

blasting titanium with large grit and acid etching in HF. The nanotubular layer with

mean inner diameter of 30 and 70 nm resulted in higher level of ALP activity,

mineralisation and osseointegration [90].

In the process of the fabrication of a TiO2 nanotube, it has been shown that when the pH

value is slightly acidic then smooth and long tubes can be formed [83]. An irregular

porous layer will be obtained if an inappropriate applied potential is used. The thickness

of the nanotube oxide layer can be increased by increasing the anodisation time;

however there is a limit beyond which no further changes will be observed or the oxide

26

layer might possibly dissolve. The time limit for a nonaqueous electrolyte is longer than

for an aqueous electrolyte.

There are few study on the effect of the nanotube length and wall thickness on adhesion,

wettability, or surface energy and consequently on cell adhesion, proliferation and

differentiation. Although a longer nanotube has been obtained by improving the method

and the condition, the question of the optimum length required to promote cell response

has not been reported to date except the work of Bauer et al. [91] that have indicated the

tube length does not affect the cell behaviour.

2.3 Nanotube oxide layer on titanium alloys and titanium alloying metals

2.3.1 Anodisation of biocompatible Ti-Nb-Ta-Zr alloy

Some titanium alloying elements are beta (β) stabilisers; e.g., molybdenum, vanadium,

niobium, tantalum (isomorphous; completely miscible elements in the phase), iron,

tungsten, chromium, silicon, cobalt, manganese (eutectoid; elements with eutectoid

temperature as 335 ºC below the unalloyed titanium transformation temperature); and

zirconium, a neutral (a continuous solid solution) alloying element. Beta titanium alloys

exhibit lower elastic modulus compared to α, and α+β titanium alloys [7, 92]: with Ti-

Nb-Ta-Zr (TNTZ) being such an alloy that is the choice for orthopaedic endoprostheses

[93]. Since TNTZ alloy cannot form apatite on its surface by the usual chemical and

heat-treatment processes, then another surface modification method is required [4, 93].

Nakada et al. [94] suggested using Ti-Nb-Ta-Zr alloy for orthopaedic implants because

of their excellent mechanical properties, anticorrosion ability, cytocompatibility and

biocompatibility. An alloy of Ti15Nb4Ta4Zr was compared to conventional Ti6Al4V in

terms of new bone formation and bone mineral density (BMD) [95]. It was

demonstrated that bone mineral density around Ti15Nb4Ta4Zr was equivalent to, or

even higher [94], than that of Ti6Al4V.

Hard tissue replacements, especially knee and hip joints, are eroded over time due to

micro-movement. As a result, microscopic wear particles gather at the implant-tissue

interface and give rise to inflammation, osteolysis, infection and, potentially, implant

loosening. The wear resistance of Ti29Nb13Ta4.6Zr was measured after different heat

treatments and oxidation treatments [96]. Wear resistance was improved by increasing

27

the niobium content, especially by means of an oxidation treatment, since the Nb2O5

layer is hard and exhibits lubricity; but no improvement was observed by heat treatment

[96].

Anodic oxidation of TNTZ alloy exhibits different reaction rates because of the

different electrochemical oxidation rates of elements in this alloy [97]. Consequently,

dissolution was more selective and a homogeneous, self-organised pattern of anodic

tube layers grew to different sizes. Each tube with a large diameter was surrounded by

nearly eight tubes of smaller diameter. The process was initiated with two layers; an

outer nanoporous layer and a nanotube layer when the sweep potential of the

potentiometer was employed. The top nanoporous layer dissolved when the sample was

held in the electrolyte and additional time was also necessary to grow nanotubes when

the applied potential was low [97, 98].

Homogeneous oxide layers of 100 nm thickness were formed on the α phase of

Ti6Al4V due to electrochemical anodisation of the titanium alloy [99]. On the other

hand, due to selective dissolution of oxide layers during anodisation, the β phase of

Ti6Al4V exhibited oxide layers that were inhomogeneous in terms of morphology and

diameter. An inhomogeneous oxide layer that was nanoporous and a mixture of

nanopores and nanoscratches were observed for a complex α + β phases. The element

niobium was observed in nanotubes formed on the surface of alloys consisting of

niobium [99, 100]. Two-size-scale structures of anodised nanotube oxide layer were

observed on Ti-28Zr-8Nb alloy using buffered electrolyte [101]. In this chapter, two

distinctly different tube diameters of 175 nm large nanotubes surrounded by several 98

nm small nanotubes at the bottom were formed with a length of 8.5 µm. The thickness

of this oxide layer depended on both the anodisation time and the applied potential and

it reached 22 µm by extending the time to 8 h; and increased from 3.2 µm to 11.3 µm

by increasing the applied potential from 10 to 40 V for 1 h anodisation [101]. This study

concluded that the nanotube length increased almost linearly with the anodisation time

in the first 60 min, and the rate slowed down with a longer time; whilst the nanotube

length increased linearly with the applied potential up to 40 V. Further studies are

needed concerning the anodic oxidation conditions to tailor the optimum nanotubes on

the titanium alloys, and to evaluate the cell behaviour with regard to their orthopedic

applications on this new pathway of multi-scale nanotubular geometry. Fig. 2.5.a SEM

28

and TEM images (top view, cross-section and bottom) of anodic oxide nanotube layer

formed on a) β phase titanium alloy after anodisation at 20 V for 4000 s [99].

Fig. 2.5 SEM and TEM images (top view, cross-section and bottom) of anodic oxide

nanotube and nanoporous layer formed on a) β phase titanium alloy after anodisation at

20 V for 4000 s [99], d) Zr substrate using 1M (NH4)2SO4 + 0.5 wt % NH4F electrolyte

[102], c) niobium substrate as a function of anodisation temperature at 15 °C [103], b)

tantalum substrate using 1 M H2SO4 + 2 wt % HF electrolyte for 2 h with a sweep rate

100 mVs-1 [104]

2.3.2 Anodisation of binary titanium alloys for implant applications

Studies [105, 106] on binary titanium alloys such as Ti-xTa, Ti-yZr and Ti-Al have

investigated the effect of alloy composition on the morphology of anodic oxide layers.

Uniform nanotube arrays of oxide layers were formed on Ti-Zr alloy and with

increasing zirconium content, the diameters of the tubes decreased and the length

increased. Tube separation occurred when the titanium content of the Ti-Al alloy

increased [105]. In the case of Ti-50Ta, the morphology and length of the nanotubes

29

depended on the phases of the alloy. It was revealed that the nanoporous oxide layers

that were formed first were dissolved after the growth of nanotubes beneath the

substrate surface. This dissolution arose because a local electrochemical was established

due to the different compositions of the complex phases [106].

Titanium-zirconium alloys (Ti-20Zr, Ti-50Zr, Ti-80Zr), which had been fabricated by

accumulative roll bonding (ARB), were treated electrochemically [107]. The ARB

process did not influence the diameter, length or arrangement degree of the tubes during

anodisation. Tubular oxide layers were formed at all ranges of zirconium content

without any transition from porous to tubular forms. As well, the tube diameter

decreased, although the tube length increased, when the zirconium content was

increased. Two dissimilar diameter tubes were observed on Ti-50Zr and Ti-80Zr. These

have been observed previously as forming on Ti-29Nb-13Ta-4.6Zr and Ti-50Ta [107].

Table 2.3 summarised the anodisation conditions, characteristics and properties of TiO2

nanotubes on titanium investigated by different research groups, to date.

30

Table 2.3 Anodisation conditions, characteristics and properties of TiO2 nanotubes on titanium

Anodisation condition Characteristics of TiO2 nanotubes Properties Ref.

Electrolyte

Applied

potential

(V)

Time Others

Nanotube

diameter

(nm)

Wall

thickness

(nm)

Nanotube

layer

thickness

Phase

structure Other treatment

Hardness / bonding

strength

Wetting

angle

(water)

Roughness

H2SO4/HF 20 2 h 50-100 amorphous HTa to Anatase &

Rutile [63]

H3PO4/NaF 20 28 min 30C

Stirred 100 amorphous

HTa to Anatase-

Deposition; Ca to

P=1.67, pH=4.5

bonding strength

;7.41 MPa [108]

H2SO4 /HF 20 3 h ~100 500 nm amorphous HTa to Anatase &

Rutile

enhanced apatite

formation Rsm=25nm [48]

H3PO4/HF 20 2 h 400 77.45 Ra=14 nm [65]

H2SO4/NaF/ Citric

acid 20

15min

-4 h pH=4.5 35-70 10-18

230-670

nm

Young’s modulus=

36-43 GPa [75]

HF /NH4H2PO4 20 1 h 20C

Stirred 100 19 1 µm

HTa to Anatase- HA

deposition by

immersing

bonding strength;

15.3 ± 2.5 MPa [74]

HF 20 2 h 100 83.39 Ra=19nm [65]

HF/NH4F 20 2 h 60 55.45 Ra=16nm [65]

31

HF 20 20min 80 10 250 nm amorphous HTa to Anatase &

Rutile [69]

Na2SO4 /NaF 20 3 h ~100 2 µm amorphous HTa to Anatase &

Rutile

significantly

enhanced apatite

formation

Rsm=77nm [48]

(NH4)2SO4/NaF 20 2 h 120 16° Ra=23nm [65]

Glycerol/NH4F 30 30 min 30C

Stirred

50-65

ripple 1.20 µm amorphous 3.5 ± 0.2 MPa 58 ± 2

Ra= 0.32 ±

0.2 µm [76]

Glycerol/NH4F 30 60 min 30C

Stirred 60-75 ripple 1.45 µm amorphous 4.1 ± 0.2 MPa 73 ±2

Ra= 0.49 ±

0.3 µm [76]

Glycerol/NH4F 30 120

min

30C

Stirred 75-85 ripple 1.7 µm amorphous

3.7±0.3

MPa 78 ± 2

Ra= 0.17 ±

0.2µm [76]

Glycerol/NH4F 30 240

min

30C

Stirred 110-130 ripple 1.9 µm amorphous 3.6 ± 0.2 MPa 84 ± 1

Ra= 0.13 ±

0.2 µm [76]

NH4F and H2O in

Ethylene glycol 60 18 h 100 25 45µm amorphous

HTa to Anatase &

Rutile [68]

Ethylene glycol/HF 120 15 h 70 45 261 µm [78]

CH3COOH-0.2%

H2O/NH4F 30 1 h 10 200 nm [79]

32

KF/NaHSO4/citric

Acid 25 20 h pH=4.5 110-120 4.4 µm amorphous

HTa to Anatase &

Rutile [83]

H2SO4/NaF/ citric

acid 20 4 h pH=4.5 51.1 50.58 600nm amorphous

HTa to Anatase &

Rutile

1.5-1.7 GPa (36-43

GPa)

Water=7.5±

3

Cell media=

2±1º

(surface

energy of

332.6

mJ/m2)

Rms=

51.83±7.77 [109]

acetic

acid/HF 5-20 30-100 amorphous HTa to Anatase 11,9,7,4º

Ra=13.0,12.7

,13.5,13.2

nm

[20]

HTa = Heat Treatment

33

An equation was proposed to describe the linear relationship between the outer diameter

of zirconium-titanate nanotube and the anodisation potential from 1 to 100 V in a

buffered electrolyte [110]. This study also indicated that the concentration of the

fluoride ions influenced the wall thickness and the composition of the nanotube layer

because the zirconium titanate nanotubes grew as a result of a competition between an

electrochemical oxide formation and chemical dissolution of oxide by fluoride ions. The

wall thickness was smaller for high fluoride ion concentrations due to the accelerated

dissolution rate of the oxide wall. The nanotube length depended on the anodisation

charge until etching of nanotube top was significantly observed [110]. Multiple

nanotube layers that can be formed using two step anodising have potential application

such as ultrafiltration or drug release systems. The lower layer of nanotubes started

forming in the gaps between the upper layer of nanotubes formed in the previous

anodisation process [111]. Zirconium titanate nanotubes that were formed on Ti-50Zr

(wt %) alloy surface were amorphous and they were converted to crystalline form after

annealing although TiO2 nanotubes on pure titanium were amorphous and ZrO2

nanotubes on pure zirconium were crystalline [112]. The nanotube oxide layer that grew

on Ti-45Nb (wt %) alloy were thicker and more regular in comparison to the nanotube

oxide layer formed on pure titanium and niobium [113]. Under the same anodisation

conditions, the oxide layer of TiO2 on pure titanium exhibited a thickness of 1.1 µm and

tube diameter 100 nm; and the Nb2O5 nanotubes on pure niobium exhibited a thickness

of 0.6 µm with 15 nm irregular pore diameter; whilst 2.9 µm in length uniform

nanotube oxide layer with 60 nm tube diameter grew on Ti-45Nb alloy after 2 h

anodisation and increased to 6.3 µm at 20 h [113].

2.3.3 Anodisation of tantalum as a β stabiliser

Tantalum is more expensive and has much higher density (16.6 g/cm3) [114] than

titanium (4.5 g/cm3) [115], and therefore it is not appropriate for implants. However, its

chemical and physical properties, such as biocompatibility [116], radiopacity, dielectric

properties [117] and high corrosion resistance to body fluids, are attractive [118].

Tantalum has applications for optical devices, storage capacitors for very large-scale

integrated circuits, and for electronic and sensor devices. Tantalum oxide has received

much attention due to its corrosion resistance as a protective coating material. In

34

addition to aluminium, tantalum is one of several metals onto which a porous oxide

layer can be formed by anodising [52, 119].

A porous oxide layer (Ta2O5) can be formed on tantalum metal when it is

electrochemically treated in acid that contains HF as the electrolyte. Pore diameters are

from 2 to 35 nm and the layer thickness is 350 - 400 nm. Similar to titanium, the

structure and distribution of the pores and the thickness of the porous layer can be

altered by changing the concentration of HF, the time allowed for anodic oxidation and

the anodisation voltage. The equilibrium between porous oxide formation and its

dissolution, as well as the time scale, controls the film thickness for the development of

porosity and ordering effects [52, 104, 117]. Before making dimpled tantalum metal

surfaces in H2SO4 and HF electrolyte without organic additive, tantala nanotubes can be

fabricated if the time of anodisation adjusted between 5 to 120 s. The reason of

nanotube loss from the surface was shown to be the formation of a thin, fluoride rich

(likely TaF5) layer builds up at the Ta/ Ta2O5 interface [120].

Tantalum oxide nanoporous layer can be fabricated in an organic electrolyte consisting

of ethylene glycol and glycerol with a background salt such as (NH4)2SO4 plus suitable

amount of NH4F to a maximum thickness of 11 µm and the pore diameters increased

from 9 to 16 nm with increasing the voltage from 10 to 40 V [121]. A highly ordered

Ta2O5 nanoporous layer grew on tantalum foils by anodisation using glycerol (1,2,3-

propanetriol) and NH4F as electrolyte . It was found that the average pore diameter and

the thickness of the porous layers were affected by the NH4F concentration. The

optimum conditions for forming a homogenous nanoporous oxide layer with high aspect

ratio and a pore diameter of 17 nm and a thickness of 16 µm were a 0.2 M NH4F

electrolyte and an applied potential of 20 V [122]. Fig. 2.5.b SEM images (top view

and cross-section) of anodic oxide nanoporous layer formed on tantalum substrate using

1 M H2SO4 + 2 wt % HF electrolyte for 2 h with a sweep rate 100 mVs-1 [104].

Tantalum has been deposited on the titanium surface by electrodeposition [123, 124],

and a sol-gel process [125] so that the good bulk properties of titanium can be enhanced

by the good surface properties of tantalum oxide. The aim was to increase the corrosion

resistance as well as the bond strength between the bone tissue and titanium.

35

2.3.4 Anodisation of niobium as a β stabiliser

The functionality of niobium for its application in gas sensors, catalysts, optical,

electrochromic devices, and biocompatible prostheses, has been investigated with

regard to increasing the local surface area. Nanostructured niobium oxide has been

fabricated by reactive sputtering [126], sol-gel processes [127], templating techniques

[128] and anodic oxidation [129]. Sieber et al. [51] were the first group of researchers to

fabricate self-organised porous niobium oxide by anodising oxidation in an electrolyte

containing F-. Uniform layers of porous niobium oxide with 20 - 30 nm pore diameters

were obtained. The HF concentration in the electrolyte and the anodisation time

influenced the thickness and microstructure of the porous layer. A steady state porous

film formation and film dissolution limited the film thickness to about 500 nm [51].

The effects of mixed electrolytes, applied potential and anodisation time were

investigated to obtain a high aspect ratio of ordered porous niobium oxide layers. It was

revealed that less than 10 nm in diameter of ordered porous niobium oxides were

formed by anodisation of niobium foils at 2.5 V in a mixture of 1 wt % HF and 1 M

H3PO4 for 1 h. Nanoporous niobium oxides could not formed on niobium if a single HF

electrolyte was used since the pore wall would dissolve as they form [130].

Choi et al. [103] targeted a thickness of porous niobium oxide more than 500 nm by

employing double anodising and one step annealing. The annealing step was designed

to form a protective oxide layer that would delay the chemical dissolution as part of the

anodic oxidation in fluorinated electrolytes. The thickness of the outer layer formed in

the first step was around 90 - 130 nm; and the inner layer of around 300 - 400 nm was

grown underneath the first layer after annealing [103]. Fig. 2.5.c SEM images (top view

and cross-section) of anodic oxide nanoporous layer formed on niobium substrate as a

function of anodisation temperature at 15 °C [103].

2.3.5 Anodisation of zirconium as a neutral titanium alloying element (or an

effective β stabilizer in multi-elementary Ti alloys)

Zirconium is one of the valve metals reported to form porous oxide layers on their

surface. Zirconium oxides have noteworthy properties, such as chemical and thermal

stability, mechanical strength, wear resistance, good ion-exchange and excellent

biocompatibility. Zirconium is used as a catalyst where an increase in its surface area

36

improves its functionality. Other applications include magnetic recording media, optical

devices, functional electrodes, and waveguides [131].

Tsuchiya et al. [53] were among the first researchers to report on the fabrication of

sponge-like porous zirconium oxide layers manufactured by electrochemical

anodisation in acidic electrolyte containing F-. An applied potential between 20 and 40

V was needed and the pore diameter was several 10 nm with a thickness of several 10

µm.

The formation mechanism of a self-organised nanotube oxide layer has been described

in detail [132] with regard to (i) the effect of changing the concentration of fluoride ions

and pH, (ii) the composition of the electrolyte, and (iii) the potential sweep rate and

holding potential, on the morphology and geometry of the porous oxide structures. By

improving the method, smooth high aspect ratio zirconia nanotubes with a diameter of ~

50 nm and a length of ~ 17 µm were obtained using a buffering electrolyte, instead of

the wavy and irregular wall morphology of pore arrays [102, 132]. Organic electrolytes

were also used to grow significantly thicker, and continuous and smooth ZrO2 nano-

tubular layers up to 40 µm on zirconium substrates [133-135]. It has been shown that

water content of the organic electrolyte influenced the chemical etching behaviour of

the electrolyte and the porous to tubular morphology, which is important in different

applications [134].

Pretreatments such as dip etching [133], electropolishing [135] and two step anodising

[133] were used in the investigations of the effect of pre-treatment of zirconium surface

on the anodised nanotubes. Regular nanotubes that formed on dip etched surface

indicated that the mechanical deformation layer and/or zirconium surface impurities

were the reason of the irregular, patchy (coherent) oxides. The needle-like (nano-pillars)

structures on the top of the tube walls were also observed due to thinning and partially

etching-out [133]. Highly regular hexagonal ordered dimples of the right spacing were

formed on zirconium surface after removing the layer resulted from the first anodisation

step. The nanotubular oxide layer exhibited a thickness of 3 µm and a diameter of 60

nm with highly ordered hexagonally porous which can be removed [133]. The

roughness of zirconium surface decreased from ~ 18.6 nm of bare surface to ~ 1.1 nm of

electropolished surface and this reduction had influence on a homogeneous electric field

distribution over the entire metal surface during the anodisation step, thus nanotubes

37

grew regularly on the entire surface [135]. Fluorine can be removed from the ZrO2

nanotubes by heat treatment at 300 °C for 6 h and it has been shown that although 800

°C heating will lead to crystallisation; collapse of the nanotubes will be observed [136].

Fig. 2.5.d SEM and TEM images (top view, cross-section and bottom) of anodic oxide

nanotube layer formed on Zr substrate using 1M (NH4)2SO4 + 0.5 wt % NH4F electrolyte

[102].

2.4 Factors that influence bone cell adhesion

The extracellular matrix (ECM) secreted by cells plays a role as the interface of a

biomaterial surface. ECM is synthesised and degraded by cells as a dynamic

surrounding substance that consists of 90 % collagenic proteins (type I collagen and

type V collagen) and 10 % non-collagenic proteins (NCP, for example osteocalcin,

osteonectin, bone sialoproteins, proteoglycans, osteopontin, fibronectin, growth factors

and bone morphogenetic proteins etc.) [15]. Anselme et al. [137] indicated that the cells

are always in contact with a biomaterial surface that has previously adsorbed water and

proteins from biological fluids, rather than with a bare surface. Specific receptor

proteins such as integrins, as well as the junctions of adherents containing cadherins

mediating cell-cell adhesions, are important for cell-substrate adhesion. Focal contacts

or adhesion plaques are junction locations of about 10 - 15 nm where adherent cells and

material surfaces are joined. Integrins are on the external side of focal contacts and they

can translate the attachment of external ligands to internal information that induces

adhesion, spreading, or cell migration, growth and differentiation [15]. It has been

shown that the vinculin and puxillin can play an essential role in continuing adhesion

and upcoming differentiated cell functions [138]. The cell proteins involved in cell

adhesion on biomaterial are illustrated in Fig. 2.6.

38

Fig. 2.6 Representation of the cell proteins involved in cell adhesion on biomaterial: (a)

immediately after implantation; (b) adsorbing proteins from body fluid; and (c) attached

bone cell on an implant material surface in higher magnification (adapted from [15])

39

The adhesion force at the cell/implant interface determines phenomena that involve

short term and long term adhesion due to (i) physicochemical linkages such as ionic

forces and van der Walls forces and (ii) different biological molecules such as

extracellular matrix proteins, cell membrane proteins, and cytoskeleton proteins. The

adhesion can be measured using a variety of methods such as micropipettes aspiration,

centrifugation (the force necessary to separate cells from a substrate is provided by

centrifugal force) [139] , and fluid flow; enzymatic procedures; optical or magnetic

tweezers (measure local viscoelastic properties of the surface of adhering cell by a

magnetic bead microrheometer [140]); and microcantilevers (detaching adhered cell by

applying a lateral load using a microcantilever and measuring the detachment force

[141]) for the detachment of cell populations by micromanipulation [137]. By

micropipettes aspiration the strength of cell adhesion can be measured by vertically

oscillating micropipette while it makes contact with cell and the pressure within the

micropipette is reduced to a level that cell can attach to the pipette by suction [142].

Glass, silane-adherent cells [143] which deposited in glass and silane capillary tubes can

be detached by calibrated laminar shear flows with a highly viscous dextran solution.

This method is known as fluid flow [144].

There are different opinions concerning the preferred bone growth direction around

implants, for example, towards the implant, on the implant surface or both together

[137, 145]. In addition to the cell type, the surface chemistry, mechanical cues and

topography influence cell behaviours such as adhesion, proliferation, shape, migration,

survival and differentiation [137, 145, 146]. It has also been reported that there are

different mechanisms of adhesion for blood cells, fibroblasts or osteoblasts (connective

tissues), and endothelial vascular cells or keratinocytes (endothelia or epithelia), such as

adherence and proliferation rate [137].

2.4.1 The influence of surface physicochemical, mechanical and electrical

properties on bone cell behaviour

Surface chemistry plays an important role in the cell response at interfaces. The effect

of surface energy, water contact angle or wettability (hydrophilicity and

hydrophobicity), and the zeta potential of different surfaces on cell behaviour have been

reviewed [137]. Protein adsorption, cell adhesion, proliferation and osteoblastic

differentiation will be increased by a decrease in the water contact angle or an increase

40

in surface energy. Electrical charges on titanium surfaces that are measured as zeta

potential are important in the interaction with negatively charged proteins such as

fibronectin (FN) [14, 147]. A titanium surface, normally, is negatively charged due to

the adsorption of anions (OH-, F- etc. from the electrolyte). On the other hand, cell

membranes are also negatively charged, so positively charged proteins at the cell

surface interface can play an important role, which is proposed as a dynamic model for

osteoblast attachment [147]. The adhesion constant and binding efficiency of adsorbed

proteins on different functional groups (e.g. OH, COOH, NH2 and CH3) demonstrate

different adhesion strengths [14]. The cell adhesion, proliferation and gene expression

were improved after UV irradiation due to the introduction of Ti-OH groups on the

nanostructured surface and the further enhanced hydrophilicity of nanostructured

surface [148].

Changing the surface topography to nano scale does not change the surface chemically

with respect to its bulk, but it might change the surface chemistry and energy at the

interface. Rising hydrophobicity is one example of the influence of surface density on

surface chemistry. Decuzzi et al. [149] proposed a mathematical model and introduced

three regimes for surface energy including small, intermediate and large. They then

investigated the effects of surface roughness on cell adhesion in these three regimes, and

found that cell adhesion will deteriorate with an increase in roughness for small energy

surfaces, exhibit maximum adhesion for an optimal roughness yet will be scarcely

affected by roughness for intermediate surface energies [149]. In the line of the

abovementioned model, stable cell adhesion and proliferation for four different cell

lines have been reported to be increased on moderately rough Brownian substrates with

nearly similar surface energy but the generality of the results need additional studies to

verify [150].

Osteoprecursor cell line attachment and growth behaviour on TiO2 nanotubes with ~ 50

nm diameter were also studied [109]. The effects of the physicochemical properties of

the surface, for instance, roughness, contact angle and surface energy, on the cell

behaviour were investigated and it was found that materials with a high nano scale

roughness, low contact angle and high surface energy showed the same positive effect

on cell behaviour, such as early differentiation and significant bone cell proliferation,

which means an improvement in cell adhesion and spreading [109]. Kim et al. [151]

41

indicated that cell behaviour is affected by the incorporated anions in the oxide layer

rather than the morphology of surface. It was shown that the composition of the

electrolyte, e.g., HF, HF+H3PO4, and H3PO4 not only determines the oxide

morphologies (dot-like structures, granules, nanopowders and nanotubes), but it also

affects the incorporation levels of the ions of F- and PO43-, which cause different cell

behaviours: the PO43- enriched oxides enhance cell proliferation in 7 days and the F-

enriched oxides stimulate initial cell attachment [151].

Mechanical cues, for instance, rigidity, stiffness and resilience, have been shown to

influence cell behaviour [137]. Influence of microenvironment stiffness on stem cell

specifications has been observed when the cells commit to the lineage specified by

matrix elasticity after several weeks in culture in inverse to the initial week in culture

that could be reprogrammed these lineages by adding soluble induction factors [152].

Mechanical feedback regarding substrate rigidity is essential for the cell to shape, grow

and survive, but recognition by the cells of their microenvironment and elasticity and its

influence on the structure and function of the cells is still under investigation [137, 153-

155].

2.4.2 Effect of topography on the bone cell behaviour

Chen et al. [156] reported that a decrease in the surface roughness of Ti resulted in an

increase in its corrosion resistance and a decrease in ion release. The roughness of a

metallic implant surface and its uniformity in the horizontal or vertical direction

influence its favorable mechanical locking to tissues. Several hypotheses regarding the

mechanisms of cell responses to surface topography were proposed in a review by

Anselme et al. [157]. Extracellular matrix protein adsorption acts as an energy barrier

for cells to modify their orientation, adhesion and spread. Changes in gene expression

and in cellular, cytoskeletal, and focal adhesions were observed in a study on cell

behaviour on micro- and nano-anisotropic topographies, regardless of the long-term

turnover [157]. Although there is an absence of comparable studies, cell operation is

motivated when the size is ~ 10 nm in height or depth [158], and cell operation is not

noticeable when the scale increases to ~ 100 nm.

The principal aim of all research in this field is to improve the wound healing that

accompanied by the promotion of osseointegration. Large macrostructure human bones

42

consist of nanosized organic and mineral phases, such as ions, DNA, proteins and the

viruses in the body. There has been insufficient study concerning the retention of

proteins on nanoscale surfaces, because fabrication of nano features or nano surfaces

cannot be easily reproduced. Nevertheless, investigation has shown that the structure

and function of absorbed proteins on smaller nanoparticles can be readily retained in

comparison to larger nanoparticles, but the details of the adsorption mechanisms are not

clear [157, 159]. As cell-substrate adhesion is based on integrins that have nanoscale

features, it can be predicted that cells will respond better to nanoscale surfaces. Further

research needs to be undertaken to clarify this hypothesis [137, 160]. In general it has

been shown that cells respond to nanosurfaces because the pores, ridges, and fibres of

the basement membranes have nano-scale characteristics [161]. Fig. 2.7 shows filopodia

of SaOs-2 cells on 200 nm deep round concentric grooves and ridges in quartz [157].

Fig. 2.7 SEM image of filopodia of the SaOs-2 cells on 200 nm deep round concentric

grooves and ridges in quartz [157]

43

Nanotubular layer is well-suited to an investigation of the mechanism of a bone cell

response to the various morphologies of a titanium and titanium alloy surfaces [11, 162-

164]. In revealing the effect of nanotubular pattern of Ti-30Ta on cell response in

comparison to a surface of Ti-30Ta alloy, cellular adhesion, proliferation, viability,

cytoskeletal organisation and morphology of human dermal fibroblasts (HDF, neonatal)

were investigated. Nanotube architecture anodised at 35 V for 40 min showed

enhancement of improved cellular functionality [165]. Good bioactivity for

mesenchymal stem cell adhesion and spreading and fast formation of extracellular

matrix (ECM) materials on the Ti-Nb-O nanotubes formed on Ti-35Nb has been

observed [166]. There were also some studies on ternary and higher titanium alloys

[167-170]. Nearly the same cell adhesion rate, the spreading out of attached stem cells,

the flat shapes and numerous filopodia after longtime culture occurred for mouse bone

marrow MSCs (mMSC) on Ti-35Nb-5Zr in comparison to Ti-35Nb [168]. Human

osteoblast cells growth on anodised nanotubes on Ti-6Al-7Nb in two inorganic and

hybrid electrolyte was found to be higher than that of cells cultured on untreated Ti-6Al-

7Nb alloy [169]. The protein adsorption, initial cell adhesion, cell differentiation and

osteogenesis related gene expression have been enhanced on sparsely distributed

nanotubes (SNT) on the surface of a near β titanium alloy Ti-5Zr-3Sn-5Mo-15Nb

(TLM) compared to the polished same alloy [170].

2.5 Effect of the characteristics of anodised TiO2 nanotubes on bone cell

behaviour

The size and distance between nanofeatures on an implant, in addition to adsorbed

proteins, as well as the shape or organisation of these nanofeatures, influence the cell

response [157, 159]. Although there are methods such as lithography and

nanoimprinting to fabricate uniform and reproducible nanofeatures on polymeric or

other material surfaces, these are not convenient for metals [14]. Anodic oxidation is a

method to tailor nanoscale patterned surfaces for metals and alloys [10].

2.5.1 Effect of nano-spacing of the surface of TiO2 nanotubes on cell behaviour

Park et al. [17] evaluated the adhesion, spreading, growth, and differentiation of rat

mesenchymal stem cells on the surface of TiO2 nanotubes of 15, 20, 30, 50, 70, and 100

nm diameters. Cells were found to adhere and spread widely on 15 nm tubes but the

44

adhesion decreased with an increase in the nanotube diameter. Focal contact formation

on nanotubes with a diameter ≤ 30 nm was higher than on those with larger diameters.

The cell proliferation rate decreased as the diameter of the nanotubes increased. The

highest cell differentiation was also observed on 15 nm tubes. This result was compared

with polished TiO2 surfaces, which showed that the 15 nm tubes offered optimal

spacing for cell proliferation, migration, and differentiation resulting from integrin

clustering and focal contact formation. Peri-implant bone formation of commercial Ti

covered by 30 nm TiO2 nanotubes was investigated in vivo in pigs and it was found that

the nanostructures not only promoted osteoblast formation in the initial step but also

resisted implant insertion shear force [18]. Osteoblasts and osteoclasts, which are

responsible for the formation and resorption of bone cells, demonstrated the same

reaction to the one-dimension surface nanotopography of 15 nm TiO2 nanotubes [19],

even though it is well known that their activities are different due to their different

working mechanisms. It has been claimed that the essential nanosize of TiO2 nanotubes

is less than 100 nm and the best diameter is 15 nm for adhesion and differentiation of

various cells [19].

Although MC3T3-E1 mouse osteoblast cells promoted higher adhesion on ~ 30 nm

TiO2 nanotubes, increased elongation of the cells and enhanced alkaline phosphatase

activity were observed on nanotubes of greater diameter between 70 - 100 nm that are

known to have greater bone forming ability [20]. This conclusion is different from the

results reported by Park et al. [19], in which the optimal diameter of nanotubes is 15

nm. However, the increase in the bone forming ability of a nanotube of ~ 100 nm

diameter was caused by crystallisation of the nanotube during heat treatment [20, 171].

The adhesion was accelerated by ~ 300 - 400 % increase in the number of the adhered

cells due to its significantly increased surface area and the presence of fluid between

annealed nanotubes or those soaked in sodium hydroxide [54, 172]. The optimum size

of a TiO2 nanotube for osteoconductivity and osseointegration ability has been reported

as 70 nm after assessments using an in vivo real-time polymerase chain reaction (real-

time PCR) technique, fluorochrome labels and histological analysis [173], but this is not

completely consistent with the studies mentioned above. Also in a study of the

relationship between Zr content and nanotube diameter in the Ti-xZr (x=10,20,30,40)

and Ti-30Ta-xZr (x=3,7,15) alloy, it has been reported that when Zr content increased

nanotube diameter decreased from 200 nm to 150 nm for Ti-xZr alloy and it decreased

45

from 200 nm to 50 nm for the ternary alloys. Good MC3T3-E1 mouse osteoblast cell

proliferation, migration and differentiation have been observed on the 50 nm diameter

of nanotubes formed on the surface of Ti-30Ta-15Zr, better than on Ti alloy with low Zr

content [167]. Although titania nanotubes formed on α phase and β phase of the

titanium alloys due to the content of alloying elements [167] were self-organised,

irregular [169, 174] and showed different diameters [175] which affected the HA

formation [176] there is still few study on the cell behaviour to different phases.

In summary, cells respond to the topography, physicochemistry, and the electrical and

mechanical properties of the implant surface. They can recognise the topography from a

few nanometers to several hundred micrometers following different chemical

treatments, especially anodic oxidation where the nanotube layer is reproducible and

can be fabricated in a cost effective way. Table 2.4 lists the different cell responses to

various TiO2 nanotubes on titanium and the anodisation condition from different

studies, to date. The interaction between the cells and the TiO2 nanotubes provides the

possibility of controlling the cell culture by ordering the physicochemical properties of

the surface. Fig. 2.8 schematically illustrates a bone cell attached to titania nanotubes

showing a hypothetical model of adhesion.

46

Fig. 2.8 Schematic illustration of a bone cell (osteoblast) attached on titania nanotubes

with a diameter less than 100 nm (adapted from [109] and [35]).

47

Table 2.4 Cell responses to different TiO2 nanotubes on titanium and the anodisation conditions

Nanotube characteristic Anodisation conditions

Hydroxyapatite deposition / cell response Ref. Nanotube

diameter

(nm)`

Wall

thickness

(nm)

Nanotube

layer

thickness

(nm)

Phase

structure Other treatment Electrolyte

Applied

potential (V)

Time

(h) Others

51.1 50.58 600 Amorphous HT* to anatase

and rutile

H2SO4/NaF/

citric acid 20 4 pH=4.5

Osteoblastic precursor cell line

(OPC1) /Dense focal contacts

Proliferation 0.342

High activity

Early differentiation in day 5

[109]

>100 - Short

length - - H3PO4 /HF 20 1

Stirred

(≈180 rpm)

MC3T3-E1 cells/good initial

attachment of cells

not changed through culturing for 7 days

[151]

15-20-30-

50 - - - - H3PO4/HF 1-20 - -

Rat mesenchymal stem cells /adhesion and

spreading, proliferation and differentiation were

highest for 15 nm diameter [17]

70-100 - - - - H3PO4/HF 1-20 - - Rat mesenchymal stem cells /adhesion and spreading

was impaired without stable extension of filopodia

lowest impaired [17]

30-100 - - Amorphous HT* to anatase Acetic 5-20 - - MC3T3-E1/increasing nanotube diameters [20]

48

acid/HF led to increased elongation/stretching of cell bodies,

increased levels of alkaline phosphatase and greater

bone forming ability

30-100 - Amorphous HT* into

crystalline phase

Acetic

acid/HF 5-20 0.5 -

Human mesenchymal stem cells (hMSCs)/

promoting adhesion without noticeable

differentiation at ~30 nm, eliciting a dramatic

stem cell elongation at ~70-100 nm, induced

cytoskeletal stress and selective differentiation into

osteoblast-like cells

[21]

100 - - - - H3PO4/HF 1-20 1 - Human primary osteoblasts and osteoclast/15nm was

recognised at least by both [19]

70 15 250 Amorphous HT* to anatase HF 20 0.5 - MC3T3-E1/Adhesion increase ~300-400% [54]

80 - 400 - HT* to anatase-

load with

gentamicin HF 20 0.75 -

MC3T3-E1/40% increase in the number of cells after

1 day

65% more cells present after 7 days

50% increase in ALP levels after 3 weeks

[70]

~70 ~15 ~250 -

Immersed in

NaOH then HT*

to anatase

HF 20 0.5 -

Soaked in SBF/ MC3T3-E1/

the kinetics of hydroxyapatite growth=7

The number of

the adhered cells increases= 400%

[172]

100 19 1000 - HT* to anatase

and rutile

NH4H2PO4/

NH4F 20 1 Stirred

HA with thickness of 6µm/bonding

strength;15.3±2.5 MPa [74]

*HT: Heat Treatment

49

2.5.2 Effect of crystalline phase of TiO2 nanotubes on bone cell behaviour

TiO2 nanotubes that were tailored on the surface of titanium using anodic oxidation

could be easily removed by a moderate touch. Heat treatment has been used to increase

the adhesion of nanotubes to the titanium substrates [22]. Crystalline structures of

titania, such as anatase, rutile and a mixture of both, have been observed after this

process, which are dependent on the temperature used [23]. At higher temperatures (≥

600 oC), rutile is the remaining structure along with changes in the shapes of the

nanotubes. However at a lower temperature, for instance 450 ºC, the growth of anatase

crystallites occurs along the length and curvatures of the nanotubes, thus the

morphology remains stable [24]. Loss of fluorine can occur during heat treatment,

which in turn affects cell adhesion and proliferation [151, 177]. The stability and

hydrophilic properties of TiO2 nanotubes can also be changed through annealing. The

diameter of the titania nanotubes annealed at 500 ºC for 2 h decreased by 34.43 %, the

wall thickness of the nanotubes increased by 35.65 % for a selected nanotube diameter

of 25 - 40 nm, and the measured contact angle value changed from 73.15 ° to 17.61 °

after annealing [178]. Although heat treatment at different temperatures did not

influence the surface roughness of the titania nanotubes, the average roughness of the

annealed nanotubes was higher than the as-formed nanotubes [177]. Another interesting

study [179] indicated that the anatase phase of TiO2 nanotubes was fabricated by hybrid

thermal heating using a microwave oven after only 2 or 3 min.

In addition to annealing that changes the hydrophilicity of TiO2 nanotubes (contact

angle value decreased from 24.62 ± 5.23 ° to 10.76 ± 2.35 ° [180]), aging also

influences the hydrophobicity properties of TiO2 nanotubes. Due to the formation of

Ti(OH)4, instead of TiO2, after anodising titanium in an electrolyte based on ethylene

glycol, the hydrophobicity properties of both as-formed and annealed titania nanotubes

increased after 92 days of aging; the contact angle value increased from 24 ° to 42 ° and

from 10 ° to 27 ° for the as-formed and annealed nanotubes, respectively [180]. It can

be deduced that after transferring to a more stable state of TiO2 after aging, Ti(OH)4

becomes more hydrophobic [180]. Anti-aging TiO2 nanotubes were obtained through

optimising the anodising and annealing conditions. It was reported that TiO2 nanotubes

fabricated at the applied potential of 60 V followed by 600 °C annealing maintained

their hydrophilicity longer than those annealed at 300 °C [181]. The viability,

50

proliferation of MC3T3-E1 preosteoblasts, and mineralisation on a mixture of anatase

and rutile crystallite were higher and more regular than on anatase only and the structure

was amorphous [177].

As anatase TiO2 is more biocompatible than amorphous TiO2, the adhesion, ALP

activity and mineralisation of MC3T3-E1 preosteoblast were investigated on titania

nanotubes with a diameter ranging from 20 to 120 nm and a length of 100 to 400 nm.

SEM images of extended MC3T3-E1 preosteoblast cell filopodia on nanotube layers

with different diameters are shown in Fig. 2.9: (a) 20 nm, (b) 50 nm, (c) 70 nm, (d) 100

nm, and (e) 120 nm (×70,000), (f) 120 nm (×30,000) [36]. Despite the increase in cell

proliferation with increases in diameter, there was no other cell behaviour observed for

nanotubes with a diameter of 70 nm and greater [36]. It has been proposed that this

phenomenon is related to the effect of length and roughness of the nanotubes and to the

sensitivity of the filopodia of the cells to the anatase structure [36]. The crystalline form

of TiO2 nanotubes showed a lower tendency to corrosion in Hank’s solution in

comparison to the amorphous phase, due to the thicker layer of oxide at the interface of

the metal surface and the bottoms of the nanotubes [24] and it has been demonstrated

that annealing helps to stabilise the surface of nanotubes, leading to a higher corrosion

resistance [178].

51

Fig. 2.9 SEM images of extended MC3T3-E1 preosteoblast cell filopodia on nanotube

layers with different diameters: (a) 20 nm, (b) 50 nm, (c) 70 nm, (d) 100 nm, and (e)

120 nm (×70,000), (f) 120 nm (×30,000) [36]

52

2.5.3 Effect of hydroxyapatite (HA) coating on the surface of TiO2 nanotubes on

bone cell behaviour

A bone-like apatite layer forms on the modified surface of biomaterials after

implantation in the body. This has been identified as an indispensable stage for bone-

bonding. The nanotubular structure of TiO2 promotes the tailoring of a nano-sized

apatite structure, which has been shown to improve the bioactivity and osteoblast

functions. An increased surface area, and pathways for fluid in between the nanotubes,

is the most obvious reasons for such functionality of titania nanotubes [25]. There are

several methods for calcium phosphate or hydroxyapatite deposition using physical and

chemical techniques, such as plasma spray [182], sputtering [183, 184], electron-beam

physical vapour deposition (EB-PVD) [185], electrodeposition [186-188] and a simpler

method: immersion in a simulated body fluid (SBF) such as a Hank’s solution [22, 189].

A large diameter of 200 nm and a small diameter of 100 nm nanotubes composed of

TiO2, Nb2O5 and ZrO2 formed on the surface of Ti-29Nb-5Zr have been coated by TiN,

HA and HA/TiN using a RF magnetron sputtering system. TiN coating had a columnar

shape and was well spread onto the nanotubes, exhibiting a high film nucleation and

growth rate compared to the HA-coated surface with their coating particles slightly

covered on the tops of the nanotubes and well spread on the surface. Multi-layer coating

of HA/TiN for 1 h deposition of HA and 5 min deposition of TiN showed an entirely

spherical shape and the coatings did not block the nanotube tips which could be an

optimum condition for enhance high Ca/P absorption, in comparison to the coatings of

HA/TiN for 2 h deposition of HA and 10 min deposition of TiN [190]. The

hydroxyapatite (HA) coating made of tooth ash has been coated on 150 and 100 nm

nanotubes on the surface of Ti-xHf by EB-PVD to examine their corrosion behaviours.

It has been shown that the diameter of nanotubes decrease with increasing Hf content

and HA coating covered the tip of nanotubes in lower content of Hf (Ti-20Hf), in

reverse to Ti-40Hf. Good corrosion resistance has been obtained by HA coating on top

of nanotubes in comparison to nanotubes without coating [176]. High corrosion

resistant HA coating using EB-PVD on top of nanotubes has been reported on the

surface of Ti-30Ta-xZr and Ti-30Nb-xZr alloys; with nearly complete covering on Ti-

30Ta-15Zr/Ti-30Nb-15Zr alloys, in comparison to a non-uniformly covering on Ti-

30Ta-3Zr/Ti-30Nb-3Zr alloys [175]. The Ca/P ratio of EB-PVD coated HA on Ti-

53

35Nb-10Zr alloy after heat treatment at 500 °C was around 1.67 and the HA exhibited a

thickness of 150 nm [191].

Uniform flaky and hexagonal columnar calcium phosphate ceramics layer, depending

on the pH of the electrolyte (pH=4, 6 respectively) has been grown on anodised titanium

surface with a pre-alkaline treatment nanotubular oxide layer by electrodeposition.

Bond strength of the calcium phosphate crystals with 100 - 200 nm in diameter and

about 2 µm long which have been started to deposit at the bottom of the nanotubes of

the oxide was from 16 to 19 MPa, whereas calcium phosphate coating onto a polished

non-anodised titanium surface have been easily removed by washing due to its weak

bonding and nonuniformity [192]. In another work [193] the bond strength between HA

and alkaline treated titania nanotubes that has been deposited by pulse electrodeposition

method was from 16 to 44 MPa. The bond strength of the as-deposited HA on

nanotubes was low; it increased after alkali treatment and annealing at 450 and 600 °C.

The ring like structure of sodium titanate that has been formed in alkaline treatment at

the neck of nanotubes has played an important role to improve the bond strength of the

coating with the substrate alongside with the annealing of the HA. Calcium hydrogen

phosphate (CaHPO4·2H2O, DCPD) crystals in nanometer scale precipitated on anodised

titanium with and without annealing when electrochemically polarised in SBF solution

transformed to HA coating in alkaline solution. The bond strength between the HA

coating and the substrate was 7.41 MPa because of both the anchoring in and between

the tubes, in comparison to Ti without anodic oxidation which exhibited a bond strength

of 3.29 MPa [108].

Calcium phosphate coatings grew uniformly on TiO2 nanotubes that were 40 - 110 nm

in diameter and about 0.7 µm in length that were annealed at a temperature up to 600 ºC

and immersed in Hank’s solution, due to the negative charge nature of titania nanotubes

[22]. The Ca/P ratio of the coating layer after immersion in Hank’s solution was 1.16

for 2 days and 1.37 for 7 days, which indicated that the coating layer tended to

incorporate more calcium with immersion time. The adsorption of bovine serum

albumin (BSA) on a high surface area, such as with a calcium phosphate coating,

increased in comparison to the titanium surface covered with native oxide film, and this

enhanced cell adhesion [22]. An alternative immersion method, preloading the TiO2

nanotubes with synthetic hydroxyapatite, was used to efficiently deposit synthetic

54

hydroxyapatite, which is affected by pore size and the ion content in the oxide material

[189]. The Ca/P ratio obtained through this method, which consists of 20 cycles of

alternating immersion in a saturated calcium solution followed by 1 min holding time

and 1 min washing with ultrapure water, was about 1.58, which is close to 1.67, the

Ca/P ratio of synthetic apatite. The deposition of calcium phosphate on annealed TiO2

nanotubes increased as the diameter increased from 15 to 100 nm and the length

increased from 880 nm to 1.2 µm [189].

Apatite formation differed in terms of the nucleation and growth on the TiO2 nanotubes,

which were formed using two different conditions of stirring electrolyte: i) a bath stirred

using a magnetic pellet, and ii) an ultrasonated bath [25]. The diameter of titania

nanotubes was 75 - 110 nm with a thickness of 700 nm when formed in an ultrasonated

bath and 50 - 90 nm with a thickness of 900 nm when formed in a magnetically stirred

bath. The apatite coatings formed on the larger diameter TiO2 nanotubes (75 - 110 nm)

after soaking in SBF for 504 h exhibited a smaller crystallite size (15 nm) and a

complete covering of the surface; whilst the apatite that formed on smaller diameter

nanotubes (50 - 90 nm) exhibited a larger crystallite size of 25 nm and an islands-

covering of the surface. The protein activity of MG63 human osteosarcoma cells was

higher on the finer apatite coating [25]. The thickness of the nanotubes showed only an

insignificant influence on the cell activity. Meanwhile, hydroxyapatite growth as well as

adhesion/proliferation of osteoblasts occurred more effectively on a mixture of anatase

and rutile structures, due to annealing, than on anatase only [24], although it has been

reported that a 10 µm thick hydroxyapatite coating was formed after 4 days immersion

in SBF on the as-formed TiO2 nanotubes that were treated using several dip-and-dry

steps, by which the TiO2 nanotubes were filled and covered with calcium phosphate

nucleation sites [194].

2.6 In vivo effect of micro/nanostructure of the surface of TiO2 nanotubes

According to the highly organised hierarchical structures of bone natural tissues that

consist of nano-, micro-, and macroscale building blocks [32] a few attempts have been

undertaken to study these phenomena through implantation in the body of selected

animals. To translate the role of the micro and nanotopographies on cell functions when

primary osteoblasts were seeded on the HF acid treated/anodised nanotubes substrate,

the micro/nano-textured surface topographies showed more biologically friendly

55

rendering, with more balanced promotion in multiple cell functions than the

microtopography [195]. The HF acid treated/anodised nanotubes substrates with screw-

shape were placed in mandibles of ovariectomised sheep for 12 weeks [196]. The

implant stability quotient (ISQ) values, the maximum pull-out forces, and the bone-

implant contact (BIC) have been investigated. Significantly increased bone volume ratio

and the trabecular number with decreasing trabecular separation evidenced the fact that

the osseointegration of titanium implant could be improved by a hierarchical

micro/nanotextured surface [196]. In another work [197] screw-shaped and anodised

titanium implants have been implanted in the femur condyle close to the knee joint of

New Zealand white male rabbits. The mean pore size of titania nanotubes was ~ 108

nm, the pore size distribution (PSD) was ~ 58 - 150 nm, the porosity was 60 % , the

roughness (Sa) value was 0.65 (± 0.02) µm and the equivalent values for developed

surface area (Sdr) was 14.3 % (± 0.9). The results of the animal studies based on the

osseointegration strengths, new bone formation and bone/cell contact at the bone-

implant interface in comparison to the blasted, moderately rough implants demonstrated

potential applications of titania nanotubes for bone tissue engineering [197]. To

examine the effect of the nanotube diameter on cellular activity, five different nanotube

diameters (15, 30, 50, 70 and 100 nm) have been implanted in the frontal skull of an

adult domestic pig [198]. The BIC for the 50, 70 and 100 nm groups were greater than

control group (untreated) and the BMP-2 expression within the 50, 70 and 100 nm

groups was different. This study showed the important effect of nanotube diameters for

the controlled formation of peri-implant bone around medical implant devices.

Annealed titania nanotubes with pore size of approximately 80 nm and length of 400

nm have been used to investigate the short- and long-term performance of MSCs in

vitro and implanted in the scruff region of the neck male Lewis rats in vivo [199].

Higher cell adhesion, proliferation and viability up to 7 days of culture, higher ALP

activity and 50 % higher calcium and phosphorous concentrations have been observed

in comparison to pure titanium and tissue culture polystyrene as a control system with

no adverse immune response occurred under in vivo conditions [199]. The in vitro

outcome of accelerated osteoblast adhesion by ~ 300 - 400 % [54] of annealed TiO2

nanotubes has been followed by an investigation of in vivo bone bonding in earloop of

rabbits for 4 weeks. Bone bonding strength, bone-implant contact area, new bone

formation, and calcium and phosphorus levels on the nanotube surfaces were increased

56

in comparison to TiO2 grit-blasted surfaces [200]. Further studies have to be undertaken

to investigate the influence of nanoarchitecture and properties of titanium dioxide

nanotubes on bone cell behaviour after implantation in the body of animals to confirm

the in vitro results.

2.7 Effect of TiO2 nanotubes on bacteria attachment

Another important factor in implantation is bacterial infection. It has been reported that

bacterial adherence is lower on titania nanotubes with a smaller diameter (20 nm) rather

than a larger diameter (100 nm), although both Staphylococcus aureus and

Staphylococcus epidermidis exhibited different adhesions on both anodised and un-

anodised Ti6Al4V surfaces [201]. Yu et al. reported that bacteria adhesion on titania

nanotubes increase with an increase in nanotube diameter and reaches the highest value

on 120 nm nanotubes, which is 300 - 400 % increase compared to smooth Ti; although

the proliferation of canine bone marrow stromal cells also increased by as much as some

300 % in contrast to the smooth Ti [202]. The nanotubular layer fabricated on the

surface of Ti-Nb-Zr-Mo (β titanium alloy) with inner diameter of 30 - 50 nm revealed

good antibacterial activity to resist actinomyces viscosus [203]. The smallest nanotubes

fabricated by two-step anodisation on the surface of TiZr alloy with inner diameter of

20 nm exhibited the most efficient antibacterial behaviour against Escherichia coli

bacterium than those with inner diameter of 100 nm [204]. To limit bacteria attachment

whereas cell fate still promote, loading nanotubes with gentamicine [70], silver

nanoparticle [205] and hydrophobicity treatments (using 1H, 1H, 2H, 2H-

perfluorooctyl-triethoxysilane (PTES) [206] or heat treatment [207]) was explored.

2.8 Summary

If bonding between an implant material and bone cannot be formed initially, then

acceptance of the implant by the body, or more precisely the bone cell, fails. It is well

known that titanium and its alloys, in comparison with other metal implants, are more

biocompatible and of these, the Ti-Nb-Ta-Zr alloys have lower Young’s modulus and

higher biocompatibility. If there is fluoride ion in electrolyte, a controllable nanotube

TiO2 can be fabricated by anodising oxidation on titanium, titanium alloying metals

(niobium, tantalum and zirconium) and biocompatible titanium alloys, that is, Ti-Nb-

Ta-Zr alloys and binary Ti-Nb, Ti-Ta, and Ti-Zr alloys. The nanotubes on titanium

57

promote bonding to bone due to their high surface area and the ability of cell-

interlocking.

The tube diameter and the length of TiO2 nanotubes would depend on the conditions of

electrochemical oxidation such as the type, concentration and the pH of the electrolyte,

the applied potential and the anodisation time. The tube diameter can be increased with

increasing the applied potential if the pH value remains constant; and the tube diameter

is pH-independent if the applied potential remains constant. In comparison to using an

aqueous electrolyte, a higher applied potential is required for obtaining the same

diameter nanotubes if an organic electrolyte is used. The tube length or the thickness of

the nanotube layer increases with the time of anodisation but there is always a point at

which the thickness remains constant with further increasing time. It is worth noting

that in an acidic electrolyte with pH value less than 1, although nanotubes form, their

length will not increase by increasing the time of anodisation. In an organic electrolyte,

although the process needs longer time, considerable and longer nanotubes can be

tailored beyond a certain amount of water content.

Nanotubes fabricated on implant material surfaces provide great potential in promoting

cell adhesion, proliferation and differentiation. A high surface area, especially the

spacing between nanofeatures that is provided by the nanotubes, offers convenient

conditions for interlocking with bone cells and the penetration of body fluid. Cells

respond to nano scale structures and nanopatterns are also sensible to the chemical and

mechanical properties of the surface, such as surface energy, water contact angle and

zeta potential. A low contact angle influences cell differentiation and proliferation.

Titania nanotubes after annealing existed in a mixture of anatase and rutile, exhibited a

more stable morphology, and demonstrated a higher proliferation of MC3T3-E1. A

hydroxyapatite layer with a thickness of up to 10 µm on the surface of TiO2 nanotubes,

with different nanotube sizes and forms, can be obtained. In order to establish the

optimum nanotopography for cell response, some investigations have been undertaken.

The results of in vitro studies suggested that the optimum diameter of TiO2 nanotubes is

less than 100 nm and that 15 nm is the best although there are some controversy

discussions. If the mechanism of the cell response is clarified, the cell fate at the surface

boundary will be more controllable. Nanotubes also offer possibility in bacterial

infection control through loading the tubes with antibacterial agents. More studies are

58

needed in order to find the optimum size of nanotubes in length and diameter for the

sensing element of a bone cell to be recognised and adhered to and more in vivo

investigations to confirm the in vitro results, especially for the new biocompatible

titanium binary, ternary and quaternary alloys.

59

Chapter 3 Materials and methods

3.1 Introduction

This chapter describes the sample preparation, surface modification and characterisation

procedures for the investigation of nanotubular layer on the surface of biocompatible

metals. The materials and methods requirements for of the assessment of bioactivity i.e.

mineralisation of hydroxyapatite and biocompatibility i.e. cell adhesion of osteoblast

cell (SaOS2) are also elaborated in this chapter.

3.2 Sample preparation

The main starting material used in this study was Ti50Zr alloy (wt % hereafter). Alloy

disc samples with a diameter 8 mm and 2 mm thick were cut from ingot using electrical

discharging machining (EDM). The other samples were tantalum (Ta, 10×10×0.1 mm),

niobium (Nb, 10×10×0.05 mm), zirconium (Zr, 10×10×0.05 mm) and commercially

available pure titanium (CP-Ti, 10×10×0.05 mm) foils (Baoji Boxin Metal Materials

Co. Ltd., Shaanxi, China).

All the surfaces of the disc samples were ground using silicon-carbide papers to 2400

grit and one of the surfaces was polished down to a finish of 1 μm diamond paste. All

the samples were degreased by sonification in methanol, isopropanol, acetone and

ethanol for 15 min each progressively between grinding and polishing processes.

Finally, the samples were washed with deionised water and dried using a nitrogen (N2)

gas stream.

3.3 Fabrication of nanoporous and nanotubular layers

In order to set the electrochemical connections for anodisation, the Ti50Zr discs were

spot-welded to titanium foil and connected to the anode of an electrochemical cell; the

Ta, Nb, Zr and CP-Ti foils were connected to the anode directly as working electrode.

Two electrode configurations were set up using a DC power supply with a 1 cm2

platinum plate as a counter electrode, and placed 4 cm from the working electrode.

Electrochemical experiments were performed at room temperature with stirring. The

compositions for the aqueous electrolyte were listed in Table 3.1. All the chemicals

were of Sigma-Aldrich reagent grades.

60

Table 3.1 Compositions for the aqueous electrolyte used for fabricating nanoporous and

nanotubular layers

Type of sample Concentration of NH4F Concentration of

(NH4)2SO4

Others

Ti50Zr 0.1 - 0.5 wt % 1 M -

CP-Ti 0.1 - 0.5 wt % 1 M -

Ta 3.3 wt % - 1 M H2SO4

Nb 3.3 wt % - 1 M H2SO4

Zr 0.3 wt % 1 M addition of H2SO4

to attain pH 5

In terms of non-aqueous electrolyte, 0.5 wt % of NH4F was used and 5 and 10 wt % of

pure water were added to ethylene glycol. Different applied potentials, i.e., 5, 10, 15,

20, 25, 30 and 35 V, were used to create a variety of nanoscale sizes and

nanotopographies of nanoporous and nanotubular layers. After the electrochemical

treatment the samples were rinsed with deionised water for 5 min and dried with N2

stream. The samples which were treated in non-aqueous electrolyte were rinsed with

acetone. For cell culture studies samples were firstly pretreated by immersion in 0.5 M

NaOH at 50 ˚C for 2 min, then rinsed with deionised water to increase

biomineralisation. Some samples were soaked at 70 °C for 14 days in pure water with

periodic replacements (three times per day) to reduce any contamination of remaining

chemicals from the electrolyte. The nanotubular layer formed in non-aqueous

electrolyte were post-treated by immersion in 0.5 M HCl at room temperature for 1 min,

then rinsed with deionised water for 1 min to dissolve the remained upper nanoporous

layer.

Annealing of samples was undertaken by heating in air at 500 ˚C for 3 h at a heating

rate of 2 ˚C/min in a conventional muffle furnace (Nabertherm LT15/13/P330;

Nabertherm GmbH, Lilienthal, Germany) after any necessary treatments after

anodisation. The nanotubular layers formed in non-aqueous electrolyte were annealed

61

by heating in air at 500 ˚C for 2 h at the heating rate of 2 ˚C/min. The annealing of

nanoporous and nanotubular layer on the surface of Ta, Nb and Zr was performed in air

at 290 °C for 10 min.

3.4 Surface characterisation

3.4.1 Surface morphology and chemical composition characterisation

Metallographic characterisation of the samples after each surface modification was

carried out using a field-emission scanning electron microscope (FESEM, ZEISS

SUPRA 40 VP) equipped with electron dispersive X-ray spectroscopy (EDS). The SEM

observation was conducted directly on the surface of nanoporouse and nanotubular

layers and mineralised hydroxyapatite (HA) at 3 kV. EDS analysis was performed with

an accelerating voltage of 20 kV. Determination of the surface crystallographic structure

and phases were performed by means of X-ray Diffraction (XRD, Bruker D8; Bruker

Pte Ltd, Singapore) using Cu Kα incident radiation at 40 kV and 40 mA with a 2θ

scanning step size of 0.02 ° from 10 to 90 °. The surface chemistry was analysed using

X-ray photoelectron spectroscopy (XPS, AXIS Ultra DLD, Kratos Analytical Inc.,

Manchester, UK) with a monochromated Al Ka source at a power of 180 W (15 kV and

12 mA) and a hemispherical analyser operating in the fixed transmission mode with a

standard aperture (analysis area: nominally 0.3 mm × 0.7 mm). The nanotube size

distribution measurements were generated from 100 nanotubes on different positions for

each of three samples (300 measurements). ImageJ software (Wayne Rasband, National

Institutes of Health, America) was used for measurements. By plotting the histogram of

nanotube sizes their distributions were presented by normal and Weibull distribution

using origin lab software. The highest percentage of normal distribution of nanotube

sizes represented by mean ± variance while it presented by scale parameter (β) ± shape

parameter (α) for Weibull distribution [208, 209].

3.4.2 Surface topography and water contact angle and surface energy

measurement

Roughness parameters were measured using a 3D-Profilometer (Bruker, Contour GT-

K1; Bruker Pte Ltd, Singapore) and analysed using the SurfVision software (Veeco

Instruments Inc.; Plainview, NY, USA). Surface topography was characterised by

several roughness parameters. Mean Roughness (Sa) and Root Mean Square (RMS)

62

Roughness (Sq) are some of the amplitude parameters studied in this thesis. These

parameters measure the vertical characteristics of surface deviation over the 3D surface.

A comprehensive description of the surface textures such as peaks, valleys and the

spacing can be represented by these parameters [210]. The third central moment of

profile amplitude, skewness of a 3D surface texture (Sskw), represents the dominant

nature of topography. Thus, a value of Sskw > 0 indicates high peaks about the mean

plane and Sskw < 0 indicates deep valleys such as would be formed from scratches [210].

The sharpness of the height distribution is indicated by a fourth central moment of

profile amplitude; i.e., the kurtosis (Sku) of the 3D surface texture where Sku = 3.0 for a

normal distribution of heights. However, when a few high peaks are spread out over a

3D surface, then the distribution is defined as ‘platykurtoic’ and Sku < 3.0; whereas Sku

> 3.0 in the instance of the surface exhibiting a high proportion of high peaks and low

valleys and this distribution is described as being ‘leptokurtoic’ [210].

The water contact angle was measured by sessile drop technique at room temperature

using a goniometer (NRL C.A. Goniometer, Ramé-hart, Inc.; Succasunna, NJ-USA).

The drop size was 5 µL. Five droplets of each liquid (pure water and glycerol) were

used for each sample. Surface energy is calculated based on Owens-Wendt (OW)

method [211], as in the following equation:

√

√

(3-1)

Where L is the liquid surface tension and Ld, L

p are its dispersive and polar

components, respectively. Sd and S

p refer to the dispersive and polar components of the

solid surface tension, S, which is the sum of Sd and S

p. The surface tension, dispersive

component, and polar component for water are =72.8 mJm-2, d=21.8 mJm-2, and

p=51.0 mJm-2 and for glycerol, =63.4 mJm-2, d=37.0 mJm-2, and p=26.4 mJm-2,

respectively [212]. An average of five readings per sample was acquired for the

roughness parameters, water contact and surface energy measurements.

3.5 Bioactivity assessment by SBF soaking

The in vitro bioactivity assessment of metal samples were carried out by soaking the as-

formed, annealed metal oxides samples and their respective bare metals in simulated

body fluid (SBF) and incubating at 37 ˚C for 1 day, 1 week, 2 weeks and 3 weeks. An

63

SBF solution is known as a solution with ion concentrations approximately equal to

those of human blood plasma. A modified (m-SBF) [213] was prepared by dissolving

chemicals in Table 3.2 in pure water as previously reported [213]. The m-SBF was

buffered at pH 7.4 at 37˚C using the prepared solution according to the Table 3.2 of

HEPES and 1 M NaOH. There are some important notes which have to be considered

while preparing m-SBF which has been picked up from reference [214]. An appropriate

preparation method has to be undertaken to avoid precipitation of apatite during the

preparation because the solution is supersaturated with respect to apatite. A plastic

beaker without any scratch should be used in this respect. When the temperature of 700

ml pure water in a water bath reaches 36.5 ± 0.5 ºC under stirring the reagents of the 1st

to the 6th order were dissolved into the solution one by one in the order given in Table

3.2. It means that a reagent should be dissolved only after the preceding one (if any) is

completely dissolved. The hygroscopic reagents such as KCl, K2HPO4 - 3H2O, MgCl2 -

6H2O, CaCl2, Na2SO4 should be measured in as short a period as possible. Then the

electrode of the pH meter should be inserted into the solution before adding HEPES

solution little by little taking careful note of the pH change. The pH has to be kept

between 7.42 and 7.45 at 36.5 ± 0.5 ºC by adding NaOH 1M until all the 100 ml of

HEPES solution to be added because increase or decrease in local pH (out of 7.42 -

7.45) of the solution can lead to the precipitation of calcium phosphate. Then the

temperature of the solution should be adjusted to 36.5 ± 0.2 ºC and the pH of the

solution should be adjusted by dropping 1M NaOH little by little at a pH of 7.42 ± 0.01.

The final pH adjustment after reaching 1000 ml should be 7.40 exactly at 36.5 ºC. Then

the solution should be poured from the beaker into 1000 ml volumetric flask and to be

kept in water to cool down to 20 ºC when it can be filled with pure water to the marked

line. The m-SBF solution should be kept in plastic container at 5 - 10 ºC in a refrigerator

and shall be used within 30 days after preparation.

64

Table 3.2 The reagents and their amounts for preparing 1000 mL m-SBF

Reagents Amount (g) Supplier

NaCl 5.403 Merck

NaHCO3 0.504 Sigma-Aldrich

Na2CO3 0.426 Merck

KCl 0.225 Chem-Supply

K2HPO4- 3H2O 0.230 Merck

MgCl2 - 6H2O 0.311 Chem-Supply

2-(4-(2-hydroxyethyl)-1-

piperazinyl) ethane sulfonic acid

(HEPES)

17.892 g HEPES has dissolved in

100 ml 0.2 M NaOH

aMResCO

CaCl2 0.293 Sigma-Aldrich

Na2SO4 0.072 Sigma-Aldrich

NaOH 1 M Merck

The ability of the pre-treated samples to form apatite was evaluated in a static SBF

environment. The samples were removed after incubation for 1 day, 1 week, 2 weeks

and 3 weeks in m-SBF; then rinsed with deionised water and dried at room temperature

for 24 h. The change of pH value during incubation was measured using pH/mV &

Ion/pH meter Series (Oakton, Eutech instruments Pte Ltd, USA). The atomic percentage

of calcium to phosphate ratio of HA is calculated using EDS results.

3.6 Assessing cell responses on nanotubes with different nanoscale

dimensions and surface topographies

All the samples for the cell culture studies were sterilised in a muffle furnace at 180 ˚C

for 3 h. Subsequently they were placed in a well in the cell culture plate. Osteoblast-like

cell (SaOS2) (Barwon Biomedical Research, Geelong Hospital, Victoria, Australia)

were seeded on the surfaces of the nanotubular TiO2-ZrO2-ZrTiO4 samples with

65

different inner and outer diameter, wall thickness and different surface roughness

parameters. The density of the seeded cell was 5×103 cells per well (100 mm2 area). To

measure the in vitro proliferation of the SaOS2 cells a MTS assay was used. The cells

were dehydrated by immersion in a buffer solution in which the ethanol concentrations

were increased progressively every 10 min to 60, 70, 80, 90 and 100 %. This procedure

was followed by chemical drying using hexamethyldisilazane (HMDS, Sigma - Aldrich,

Australia) for 10 min. The morphology of dehydrated and dried cells was observed

using SEM after cell culturing and coating with a gold layer. The cell morphology was

also observed using a confocal microscopy (Leica SP5, Leica Microsystems, Germany).

The cell-seeded samples after cell culture were fixed in paraformaldehyde, and then

permeabilised with triton-X 100 in phosphate-buffered saline (PBS) (Sigma - Aldrich,

Australia) for 10 min each at room temperature for the confocal microscopy

observation. Then the samples were stained with 1 % phalloidin and 40-6-diamidino-2-

phenylindole for 40 min at room temperature. Between each of the steps three washes

by PBS were included. If storage of stained samples was required they were stored in

PBS at 4 °C and the observations were conducted within a week of staining. Significant

differences in the cell number were analysed using one-way ANOVA (ρ < 0.05).

3.7 Nanohardness and elasticity measurements

By recording the penetration depth of an indenter into a sample material while the

applied load is measured, the area contact and consequently the hardness of the material

can be calculated as well as many other mechanical properties. These mechanical

properties can be the strain hardening index, fracture toughness, yield strength and

residual stress [215].The nano-mechanical characteristics of the nanotubular layers such

as hardness and elastic modulus were measured by using a nanoindenter (IBIS, Fischer-

Cripps Laboratories Pty Ltd, Sydney, Australia) with a three-side pyramid Berkovich

diamond tip which was calibrated against fused silica standard. The face half-angle of

the most popular geometry for nanoindentation testing; the three-sided Berkovich

indenter is 65.27 ° - 65.3 ° [215]. The samples were put on the platform and then used

load-unload nanoindentation testing using a needle that was pressed into the sample

surface. The maximum effective load was 10 mN. Once the maximum prescribed force

was reached, loading stopped and the load was maintained constant for 10 second

before unloading. The data reflects the mean values from multiple indentations (a

66

matrix of 2×2) separated by 10 μm which were made in each sample. Oliver-Pharr

method [216] used to draw the curve of loading and unloading results to calculate the

hardness, reduced elastic modulus and elastic modulus. The following equations were

used for calculation:

hc = hmax – ε Pmax / S (3-2)

where ε is a constant of 0.75 that depends on the geometry of the Berkovich indenter. S

is the slope of the unloading curve at maximum point before start of unloading. The

contact area A is then

A = 24.5 hc2

(3-3)

the hardness is estimated as below when the contact area is determined as above

H = Pmax / A (3-4)

According to the relationship of the elastic modulus to contact area and the measured

unloading stiffness (S) it is calculated as follows:

√ √ ⁄ (3-5)

Young’s modulus E of nanotubular layer with Poisson’s ratio υ of 0.27 [217] is

calculated using Eq. (3-6):

1/Er = 1 - υ2 / E – 1 - υi2 / Ei (3-6)

Where Ei and υi are elastic constants and Poisson’s ratio of indenter; 1141 GPa and 0.07

respectively.

67

Chapter 4 Nanotubes formed in aqueous electrolyte

Abstract

Titanium and its alloys are able to grow a stable oxide layer on their surfaces and have

been used frequently as substrates for anodisation in an electrochemical surface

treatment. A nanotubular oxide layer is formed in the presence of fluorine anion (F-) via

anodisation due to the competition between oxide formation and solvatisation. In this

chapter, a highly ordered titania-zirconia-zirconium titanate (TiO2-ZrO2-ZrTiO4)

nanotubular layer was formed on the surface of Ti50Zr alloy via anodic oxidation in an

F- containing electrolyte. The sizes of the nanotubes (i.e., the inner and outer diameters,

and wall thicknesses), morphology, crystal structure, hydrophilic properties and

components of the TiO2-ZrO2-ZrTiO4 nanotubular layer before and after annealing were

examined by scanning electron microscopy (SEM), thin film X-ray diffraction, X-ray

photoelectron spectroscopy (XPS) analysis and water contact angle measurements.

The results indicated that the mean inner diameter, outer diameter and wall thickness of

the as-formed TiO2-ZrO2-ZrTiO4 nanotubes were distributed in the ranges of 13 -

81 nm, 32 - 131 nm and 10 - 22 nm, respectively, and depended on the F- concentration

of the electrolyte and the applied potential during anodisation. The number of smaller

nanotubes increased with increasing F- concentration and the mean nanotube inner and

outer diameters increased with increasing applied potential. The as-formed TiO2 and

ZrTiO4 nanotubes exhibited an amorphous structure and the as-formed ZrO2 nanotubes

displayed an orthorhombic structure. These phases transformed into anatase TiO2 and

orthorhombic ZrO2 and ZrTiO4 after annealing. The hydrophilic properties of the TiO2-

ZrO2-ZrTiO4 nanotubular layer were affected by the size distribution of the nanotubes.

The surface roughnesses and the nanotubular character transformed the nanotubes to

exhibit superhydrophilic properties after annealing. The TiO2-ZrO2-ZrTiO4 nanotubular

surface on Ti50Zr alloy exhibited higher surface energy than that of the TiO2

nanotubular surface on commercially pure (CP) titanium.

4.1 Introduction

Titanium (Ti) is the fourth most plentiful metal on earth, lighter (4.5 g/cm3), stronger

(480 MPa yield strength) and more expensive than steels (~ 7.8 g/cm3 and 175 -

750 MPa yield strength, respectively) [218]. Titanium alloys possess exceptional high

68

specific strength and excellent corrosion resistance because of their ability to form a

protective oxide layer on their surfaces [219]. Applications in aerospace, biomedical

and cryogenic industries are examples where titanium and its alloys are used other than

as an alloying element for other metals [218, 219]. Titania (TiO2) that forms on titanium

at room temperature is amorphous [11] but in crystalline form it exists in 3 phases as

brookite, anatase and rutile. Titania has a wide range of applications, such as paint

pigment, whitener for paper and rubber and as a finishing agent in porcelain enamels

[218]. The electrical and optical properties of TiO2 depend on their band gaps, that is

3.0 eV for rutile, 3.2 eV for anatase and 3.2 - 3.5 eV for the amorphous phase; all of

which are influenced by the relative Ti3+ content [11]. The TiO2 surface has been

employed in gas phase adsorption and catalysis functions; for example CO oxidation,

selective reduction of NOx and O2 and water decomposition. The reactive sites are the

surface defects such as OH groups, oxygen vacancies and the presence of Ti3+ [11].

Improved functionalities of TiO2 in the above mentioned applications can be obtained

by increasing its specific surface area.

An important example where surface modification can be implemented to beneficially

alter material characteristics concerns the fabrication of TiO2 nanotubes by the anodic

oxidation method. The nanotube arrays are able to impart the substrates with novel

functionalities in various applications such as photoelectrochemistry, photocatalysis,

dye sensitised solar cells, and electrochromic devices. Anodisation, an electrochemical

method that has been used for fabrication of nanotubes for over a decade [17, 19, 21,

172, 220, 221]; is simple, reproducible and versatile. Titanium oxide nanotubes with

desirable morphological characteristics can be obtained by controlling the

electrochemical conditions such as the applied potential, time of anodisation and type

and concentration of the electrolyte. The shape and the capillary force of the nanotubes,

along with a highly ordered morphology and large surface area, make them an excellent

choice for the before-mentioned applications. Post processing by annealing the TiO2

nanotubes on the titanium alloy surface transforms the phase structure from amorphous

to anatase or a mixture of anatase and rutile. Other applications of TiO2 nanotubes

include gas sensing (CO, H2, NOx), solution sensing (O2 gas), a substitute for carbon

based supports in methanol fuel cells, as nano test tubes and biosensors [11].

69

Zirconium is one of the sister metallic elements of titanium and hafnium, also forms a

very stable, cohesive and protective oxide on its surface [222]. Zirconium has excellent

mechanical, chemical and thermal properties that enable it to be used as catalyst and

corrosion resistant structure materials such as pressure vessels, heat exchangers, pumps

and valves as well as alloying element for β titanium type alloys [223]. Acid and NOx

reduction catalyst, optoelectronic and biomedical application have been reported for

zirconium oxide known as zirconia [224-226]. Zirconia nanotubes can grow on the

surface of zirconium by anodisation in aqueous and organic electrolyte containing F- ion

under different formation conditions [53, 132, 133].

There has been great interest in fabricating nanotubes onto the surface of binary

titanium alloys; for example, Ti-Zr [105, 110], Ti-Ta [105, 165], Ti-Nb [105, 113], Ti-

Al [105], Ti-Mo [227] and Ti-W [228]. Metal oxide nanotubes have been grown on the

surfaces of these alloys and the diameter, shape and distribution of nanotubes were

related to the alloy element concentrations [105-107, 110-113]. Only a few studies have

examined the fabrication of ZrTiO4 nanotubes for applications as microwave resonant

components, optical devices and refractory ceramics [110-112, 229]. The functionalities

of nanotubes fabricated on titanium alloys for biomedical applications have also been

encouraging; hence the focus of this current investigation.

It has been reported that Ti50Zr (wt %) alloy possesses superior mechanical properties

[230] and excellent biocompatibility [231, 232]. Nanotubular oxide layer has been

fabricated on its surface and the effect of Zr content on the morphology of anodic oxide

layers [105, 233] and their corrosion behaviours [162, 232], crystal structure and

optical properties [229] have been discussed previously. The formation of single layer

[112] or multilayers [111] of self-organised zirconium titanate nanotubes, and the

control of their morphologies and compositions [110] have also been reported.

This chapter investigated the fabrication of TiO2-ZrO2-ZrTiO4 nanotubes onto the

surface of the Ti50Zr alloy via anodisation and contrasted to TiO2 nanotubes fabricated

on the surface of a commercial pure titanium (CP-Ti). A new approach for the

characterisation of nanotubes, for the first time, in terms of two different methods of the

size distribution of the nanotubes was described and this was affected by the fluoride

ion (F-) concentration of the electrolyte and the applied potential. The hydrophilic

properties and the roughness changes of the nanotubular surfaces before and after

70

annealing were characterised using a new set of surface roughness parameters such as

Sskw, Sku and SI to describe the topography of nanotubular surface.

4.2 Materials and methods

The disc samples of Ti50Zr (8 2 mm) prepared as described in Chapter 3 were

polished down to a finish of 1 μm diamond paste and then degreased by sonification in

methanol, isopropanol, acetone and ethanol. Finally, the samples were washed with

deionised water and dried using a nitrogen (N2) gas stream. CP-Ti foils 10×10 mm were

also degreased and washed by means of the identical procedure.

The Ti50Zr discs were spot-welded to titanium foil and connected to the anode of an

electrochemical cell; the CP-Ti foils were connected to the anode directly. Two

electrode configurations were set up using a DC power supply with a 1 cm2 platinum

plate as a counter electrode, and placed 4 cm from the working electrode.

Electrochemical experiments were performed at room temperature with the electrolyte

composed of 1M (NH4)2SO4 with the addition of small amounts of NH4F (Sigma-

Aldrich reagent grades). Electrolytes with different concentrations of fluorine anions (F-

) ranging from 0.1 to 0.5 wt % were used to investigate its effect on the nanotube

morphologies. After the electrochemical treatment the samples were rinsed with

deionised water for 5 min and dried with N2 stream.

The metallographic characterisation of the samples was carried out using a field-

emission scanning electron microscope (FESEM, ZEISS SUPRA 40 VP). Annealing of

the samples was carried out in air at 500 °C for 3 h in a muffle furnace (Nabertherm

LT15/13/P330). The nanotube size distribution measurements were generated from 100

nanotubes on different positions for each of three samples (300 measurements). Phase

characterisation was performed by means of X-ray Diffraction (XRD, Bruker D8) using

Cu Kα incident radiation at 40 kV and 40 mA with a scanning step size of 0.02 ° from

10 to 90 ° (2θ). The surface chemistry was analysed using X-ray photoelectron

spectroscopy (XPS, AXIS Ultra DLD, Kratos Analytical Inc., Manchester, UK) with a

monochromated Al Ka source at a power of 180 W (15 kV and 12 mA) and a

hemispherical analyser operating in the fixed transmission mode with a standard

aperture (analysis area: nominally 0.3 mm × 0.7 mm). Roughness parameters were

measured using a 3D-Profilometer (Bruker, Contour GT-K1) and analysed using the

71

SurfVision software. Water contact angle was measured using a goniometer (NRL C.A.

Goniometer, Ramé-hart, Inc.). The surface energy was calculated using Eq. 3-1. An

average of five readings per sample was acquired for the roughness parameters, water

contact and surface energy measurements. A laboratory nanoindenter (IBIS, Fischer-

Cripps Laboratories Pty Ltd, Sydney, Australia) was applied to measure hardness and

elastic modulus of nanotubular layer. A maximum force of 10 mN was chosen for the

nanoindentation. The loading and unloading curves were drawn and the hardness,

reduced elastic modulus and elastic modulus were calculated using Oliver-Pharr method

[216] (Chapter 3). A matrix of 2 × 2 μm separated by 10 μm measurement was applied

for each sample.

4.3 Results and discussion

4.3.1 Formation of TiO2-ZrO2-ZrTiO4 nanotubes

The nanotube inner and outer diameters, layer thickness and the shape (wavy or smooth

walls) were dependent on the applied potential, time of anodisation and the type and

concentration of the electrolyte. Four categories of electrolytes for anodising titanium

dioxide nanotubes have been reported; i.e., (i) acidic, (ii) buffered, (iii) polar organic

electrolytes, and (iv) nonfluoride based electrolyte [84]. For all of these electrolytes, the

diameter of the anodised nanotubes was controlled by the applied potential and the

nanotube diameter increased with increasing voltage. The nanotube length (in other

words, the nanotube layer thickness) is controlled by the anodisation time and it

increased with increasing time until a steady state was reached at which stage there was

no growth in thickness [110].

Titania-zirconia-zirconium titanate (TiO2-ZrO2-ZrTiO4) nanotubes were fabricated on

the surface of Ti50Zr alloy via anodisation at an applied potential of 5 - 20 V and time

of 2 h in the electrolyte of 0.1 - 0.5 wt % of NH4F. Fig. 4.1 shows the SEM images of

the nanotubes on the Ti50Zr alloy. Fig. 4.1(a-1) shows the microstructure of Ti50Zr

alloy after etching which exhibited a needle-like structure. Fig. 4.1(a-2) shows two

images; the left shows the surface after peeling of the nanotubes, demonstrating that the

original grain structure of the alloy was reflected in the subsequent growth of the

nanotubes. A needlelike morphology composed of different orientations of the

nanotubes can be observed from the top view of the nanotubes on the right of Fig. 4.1(a-

72

2); and a high magnification top view image, as shown in Fig. 4.1(b), shows the

equivalent patch-like regions when the time or the applied potential was not sufficient

enough. The cross-section view of the nanotubes, Fig. 4.1(c), indicates an average

thickness (i.e., nanotube length) of 34.1 µm. The bottom view of the nanotubes reveals

different sizes; in particular, features located along the grain boundaries appeared to be

larger than those inside phases, Fig. 4.1(d).

Fig. 4.1 SEM images of TiO2-ZrO2-ZrTiO4 nanotubes: (a-1) microstructure of etched

Ti50Zr, (a-2) top view of patterned nanotubes of different phases exhibited in the

microstructure of Ti50Zr alloy, (b) nanoporous patches of different phases, (c) cross

section of the nanotubes showing the nanotube length, (d) view of the nanotubes from

the bottom, (e) top view and cross section of separated nanotubes of and (f) top view of

damaged nanotubes

73

The effect of the F- concentration in the electrolyte on the nanotube parameters such as

inner diameter (Di), outer diameter (Do), wall thickness (Wt), roughness (Sa, Sq, Sskw,

Sku), wettability and surface energy at constant applied potential and anodisation time,

as well as the effects of the applied potential on the nanotube parameters at constant F-

concentration in the electrolyte and the anodisation time were investigated. The SEM

images of the TiO2-ZrO2-ZrTiO4 nanotubes fabricated on Ti50Zr alloy under different

anodisation conditions and subsequently annealed at 500 oC for 3 h are shown in Fig.

4.2. The top view of the TiO2-ZrO2-ZrTiO4 nanotubes fabricated in 0.4 wt % NH4F at

20 V for 2 h is shown in Fig. 4.2(a). The nanotubes after annealing at 500 oC for 3 h

exhibited a slightly larger mean inner and outer diameter, Fig. 4.2(b). By increasing the

F- concentration to 0.5 wt % NH4F, the nanotubes fabricated at the same applied

potential of 20 V for 2 h revealed slightly smaller mean inner diameters and larger mean

outer diameters, Fig. 4.2(c). The cross-section view of the nanotubes fabricated at 0.5 wt

% NH4F at a lower applied potential of 15 V for 2 h is shown in Fig. 4.2(e) and the top

view of the nanotubes annealed at 500 oC for 3 h is exhibited as Fig. 4.2(f).

74

Fig. 4.2 SEM images of TiO2-ZrO2-ZrTiO4 nanotubes: (a) top view of as-formed

nanotubes anodised in 0.4 wt % NH4F at 20 V, (b) the nanotubes annealed at 500 ºC for

3 h, (c) nanotubes as-formed in 0.5 wt % NH4F anodised at 20 V, (d) the nanotubes

annealed at 500 ºC for 3 h, (e) nanotubes as-formed in 0.5 wt % NH4F anodised at 15 V,

and (f) the nanotubes annealed at 500 ºC for 3 h

The application of a suitable voltage to the surface of a metal (M) allows the formation

of Mn+ ions that can (i) dissolve, or (ii) react with existing O2- and form an insoluble

metal oxide layer on the surface of metal, or (iii) partially dissolve due to the

composition and conditions of the electrolyte [234].

75

Titania nanotubes can be formed on the surface of titanium when it is used as working

electrode in an electrochemical cell with platinum as counter electrode and applying a

constant voltage between 1 - 30 V in an aqueous electrolyte, or between 5 - 150 V in an

organic electrolyte [234, 235]. The oxidation and reduction reactions at anode and

cathode of the cell forms titanium dioxide in the early minutes, according to the

reactions described by Eqs. (4-1) - (4-4). The reaction in aqueous electrolyte for water

can be given:

2H2O 2O2- + 4H+ (4-1)

The oxidation reaction at the interface of the titanium surface and electrolyte at a

constant applied potential is:

Ti Ti4+ + 4 e- (4-2)

Then,

Ti4+ + 2O2- TiO2 (4-3)

The reduction reaction at the surface of counter electrode, normally platinum, under

these conditions is:

4H+ + 4e- 2H2 (4-4)

In terms of formation of TiO2-ZrO2-ZrTiO4 or TiZrO4 the following reactions are

possible:

Zr Zr4+ + 4 e- (4-5)

Zr4+ + 2O2- ZrO2 (4-6)

2Ti4+ + 2Zr4+ + 8O2- TiO2 + ZrO2 + ZrTiO4 or TiZrO4 (4-7)

The presence of F- allows a complex of titanium and fluoride that is soluble in the

electrolyte to be created according to the mechanism proposed in Eqs. (4-8) and (4-9)

and a porous layer forms. Nanotubes grow when there is equilibrium between the

oxidation of titanium and dissolution of the hexafluorotitanate anion [11].

Ti4+ + 6F- [TiF6]2- (4-8)

Zr4+ + 6F- [ZrF6]2- (4-9)

76

If fluorine anions are not present in the electrolyte, then the thickness of the oxide layer

is limited and the current exponentially decays due to decreasing availability of Ti4+,

Zr4+ and O2- at the interface between the titanium alloy surface and the titanium-

zirconium oxide layer. The Ti4+ and Zr4+ at the interface of the oxide layer and

electrolyte will be precipitated as a hydroxide layer if a soluble condition is not

provided. In the presence of fluorine anions, a soluble complex of Ti4+ and Zr4+ forms

as presented by Eqs. (4-8) and (4-9).

The ionic radius of F- is smaller in comparison to that of O2-. Thus, more fluoride ions

will be available at the interface of the titanium surface and the oxide layer since this

will be a diffusion controlled process relative to the larger size of O2- [58].

Figure 4.3 shows an illustration of the electrochemical cell employed for fabricating

TiO2 nanotubes layer.

Fig. 4.3 (a) Illustration of an electrochemical cell that indicates the electrolyte ions

species, (b) SEM image of top view of nanoporous TiO2-ZrO2-ZrTiO4 fabricated on a

Ti50Zr alloy after anodisation for 15 min in 0.1 wt % NH4F, 5V, and c) SEM image of

top view of TiO2-ZrO2-ZrTiO4 nanotubes fabricated on Ti50Zr alloy after anodisation

for 2.75 h in 0.3 wt % NH4F, 5V

77

The F- concentration in the electrolyte, nanotube inner diameter, outer diameter and wall

thickness are denoted as CF-, Di, Do, and Wt here after respectively. The Di, Do and Wt of

the nanotubes anodised in the electrolyte with CF- increasing from 0.3 to 0.5 wt % were

measured and are quantitatively presented in Fig. 4.4.

79

Fig. 4.4 Histograms of as-formed and annealed TiO2-ZrO2-ZrTiO4 nanotube parameters anodised at 20 V for 2 h for different concentrations of fluorine anion

(F-): (a,d) inner diameter (Di), (b,e) outer diameter (Do), and (c,f) wall thickness (Wt), Note: The nanotube size distribution graphs were generated from100

nanotubes on different positions for each of three samples (300 measurements)

80

During anodisation, a nanoporous oxide layer was observed growing on the surface of

the Ti50Zr alloy after 15 min in an electrolyte with F- concentrations ranging from 0.1 -

0.5 wt % at an applied potential of 20 V. Although nanotubes grew underneath this

nanoporous layer after 2 h as shown in Fig. 4.2(b), the nanoporous layer existed when

the F- concentration was low; i.e., about 0.1 - 0.2 wt %. The mean Di of the nanotubes

was 40.3 ± 12.6 nm, the mean Do was 66.6 ± 14.6 nm, and the mean Wt was 10.0 ±

2.4 nm when the Ti50Zr alloy was anodised in an electrolyte with a CF- = 0.1 wt %.

When the CF- increased to 0.2 wt %, the mean Di increased to 44.2 ± 13.0 nm, the mean

Do increased to 72.7 ± 13.0 nm and the mean Wt decreased to 9.2 ± 1.8 nm.

As shown in Fig. 4.4, when the CF- increased from 0.3 to 0.5 wt %, the percentage of

nanotubes with both Di < 40 nm and 40 < Di < 80 nm increased; similarly, the

percentage of nanotubes with both Do < 60 nm and 60 < Do < 120 nm increased; and

also the number of nanotubes with Wt < 10 nm increased. The percentage of nanotubes

with Wt ranging from 10 to 20 nm did not show an obvious change. The mean Di, Do

and Wt achieved for CP-Ti in the same condition showed similar trends. It can be seen

that the distribution of both Di and Do increased with an increase in CF-, while the

distribution of Wt decreased slightly when CF- increased. These nanotubes were open at

the top and closed at the bottom. After 2.75 h anodisation, almost all the nanotubes were

separated (Fig. 4.1(e)). When the time of anodisation was extended to 6.25 h, the top

features of the nanotubes were damaged with randomly dispersed 2 µm observable

holes (Fig. 4.1(f)).

The different distribution of Di, Do and Wt of the TiO2-ZrO2-ZrTiO4 nanotubes is related

to the different chemical and electrochemical properties of titanium and zirconium. The

standard electrode potential of Zr→Zr4+ is 1.553 V, less than that of Ti→Ti4+, 2.132 V

[236]. Thus zirconium competes with titanium during anodisation, and will oxidised

first under the same electrochemical conditions. Also the standard enthalpies of

formation of ZrO2 and TiO2 are different, that is, -264.199 and -228.360 gram calories

per mole, respectively [237]. Hence nanotubes of two different heights formed on the

surface of the Ti50Zr alloy. Such a pattern caused different roughness parameters in

comparison to the single phase of titania nanotubes formed on pure titanium. The

nanotubes formed on the α phase of Ti50Zr exhibit smaller Di, Do and Wt. This

morphological difference arises because the α phase contains more zirconium compared

81

to the β phase of the alloy, and the solubility equilibrium constant of (NH4)2ZrF6 is

higher than that of (NH4)2TiF6. In terms of CP-Ti, although there was still a nanoporous

layer on the surface of nanotubes with large pore size that caused some difficulties for

measuring the actual distribution of the size of the nanotubes, the inner and outer

diameter increased specifically when CF- increased.

Figure 4.5 shows the Di, Do and Wt of TiO2-ZrO2-ZrTiO4 nanotubes anodised in an

electrolyte with a constant CF- of 0.5 wt % for 2 h when the applied potential varied

from 5 to 20 V. On anodisation of Ti50Zr with the applied potential increased from 5 to

20 V at a constant CF- of 0.5 wt % for 2 h, the percentage of nanotubes with Di < 20 nm

decreased; but the number of nanotubes with 20 < Di < 40 and 40 < Di < 60 nm

increased; similarly, the number of nanotubes with Do <40 nm decreased but the

number of nanotubes with both with 40 < Do < 80 and 80 < Do < 120 nm increased;

however, the number of nanotubes with Wt both < 5 nm and 5 < Wt < 15 nm increased.

83

Fig. 4.5 Histograms of TiO2-ZrO2-ZrTiO4 nanotube parameters anodised in 0.5 wt % NH4F electrolyte for 2 h at different applied potentials: (a, b) inner

diameter (Di) as-formed and after annealing, (c, d) outer diameter (Do) as-formed and after annealing, (e, f) wall thickness (Wt) as-formed and after annealing,

respectively

84

There was a nanoporous layer on the top of the nanotubular surface when the titanium

alloy was anodised at the applied potential of 5 V. It is assumed that increasing the

applied potential provides a larger anodic current for nanotube formation and leads to an

increase in inner and outer nanotube diameter [58, 110]. It can be concluded that

increasing applied potential did not affect the Wt of nanotubes. The mean sizes of as-

formed TiO2-ZrO2-ZrTiO4 nanotubes on Ti50Zr were smaller than those of as-formed

single phase TiO2 nanotubes on pure titanium. Figs. 4.6 and 4.7 demonstrate the mean

nanotube size of Ti50Zr and CP-Ti with changing CF- and applied potential. As

abovementioned, it was impossible to measure Di, Do and Wt of fabricated TiO2

nanotubes on CP-Ti anodised at 5 V because a nanoporous layer covered the entire

nanotubular surface.

Fig. 4.6 Mean nanotube size of Ti50Zr and CP-Ti as a function of CF-: (a) Di for As-

formed, (b) Di for Annealed (c) Do for as-formed, (d) Do for Annealed, (e) Wt for as-

formed and (f) Wt for Annealed

15

25

35

45

55

65

0.2 0.3 0.4 0.5 0.6

Mea

n D

i(n

m)

CF- (wt%)

Ti50Zr-As FormedCP-Ti-As Formed

30405060708090

100110

0.2 0.3 0.4 0.5 0.6

Mea

n D

o (n

m)

CF- (wt%)


c)

5

7

9

11

13

15

17

19

0.2 0.3 0.4 0.5 0.6

Mea

n W

t (n

m)

CF- (wt%)


e)

15

25

35

45

55

65

0.2 0.3 0.4 0.5 0.6

Mea

n D

i(n

m)

CF- (wt%)

Ti50Zr-AnnealedCP-Ti-Annealed

b)

30405060708090

100110

0.2 0.3 0.4 0.5 0.6

Mea

n D

o (n

m)

CF- (wt%)


5

7

9

11

13

15

17

19

0.2 0.3 0.4 0.5 0.6

Mea

n W

t (n

m)

CF- (wt%)


f)

a)

d)

85

Fig. 4.7 Mean nanotube size of Ti50Zr and CP-Ti as a function of applied potential: (a)

Di for as-formed, (b) Di for Annealed (c) Do for as-formed, (d) Do for Annealed, (e) Wt

for as-formed and (f) Wt for Annealed

The effects of applied potential and anodisation time on nanotube length were evaluated

as shown in Fig. 4.8. The nanotubes length increased with an increase of time of

anodisation. The length also increased with an increase of applied potential. The effect

of anodisation time was more significant than applied potential on nanotube growth.

0

10

20

30

40

50

60

70

0 5 10 15 20 25

Mea

n D

i (nm

)

Applied Potential (V)


a)

0

20

40

60

80

100

0 5 10 15 20 25

Mea

n D

o(n

m)



c)

02468

1012141618

0 5 10 15 20 25

Mea

n W

t(n

m)


Ti50Zr-As FormedCP-Ti_As Formed

e)

0

10

20

30

40

50

60

70

0 5 10 15 20 25

Mea

n D

i(nm

)



b)

0

20

40

60

80

100

120

0 5 10 15 20 25

Mea

n D

o(n

m)



d)

02468

1012141618

0 5 10 15 20 25

Mea

n W

t(n

m)



f)

86

Fig. 4.8 The effect on nanotube length by: a) anodisation time and b) applied potential

Another group of TiO2-ZrO2-ZrTiO4 nanotubes were fabricated when the applied

potential increased from 20 to 35 V in an electrolyte with a constant CF- of 0.5 wt % for

1 h. These applied potential was chosen to investigate the trend of nanotube sizes

distribution in higher applied potential. Fig. 4.9 shows SEM images of nanotubular

layer which fabricated at applied potential from 20 to 35V. Fig. 4.9 (a), (c), (e) and (g)

are before annealing and Fig. 4.9 (b), (d), (f) and (h) shows TiO2-ZrO2-ZrTiO4

nanotubes after annealing at 500 °C for 3 h. TiO2-ZrO2-ZrTiO4 nanotubes were the

result of anodisation and dissolution of oxide layer in presence of fluorine anions.

87

Fig. 4.9 SEM images of nanotubular layer fabricated at a) 20V - as-formed, b) 20V -

annealed, c) 25V - as-formed, d) 25 V - annealed, e) 30 V - as-formed, f) 30V -

annealed, g) 35 V - as-formed and h) 35 V - annealed

88

The growth rate of TiO2 was reported 2.5 nm/V [238] and it was 2.9 nm//V [239] for

ZrO2. Therefore an adherent layer of TiO2-ZrO2-ZrTiO4 grew on the surface which had

higher thickness where ZrO2 was exist because of its higher oxidation rate. These layer

consisted of two parts; substrate and oxide layer which is very compact and protective

and oxide layer and electrolyte which is porous. These pores are suitable places for

starting of tube growth because of trapping of F- and formation of dissoluble complex of

[TiF6]2- and [ZrF6]2- with different reaction rate. The difference between oxidation and

dissolution rate of titanium and zirconium caused to fabrication of different inner and

outer diameter and wall thickness of nanotubes as well as different nanotube length

which caused different roughness. Fig. 4.10 show distribution of Di, Do and Wt which

were as a function of applied potential.

90

Fig. 4.10 Histogram and distribution of Di, Do and Wt of as-formed and annealed TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a,b) 20 V, c,d) 25 V, e,f) 30 V and

g,h) 35 V

91

The highest percentage in Di at 20 V was between 28 < Di < 52 nm. The greatest

frequency increased with respect to potentials of 25, 30 and 35 V to corresponding

values of 42 < Di < 76 nm, 41 < Di < 87 and 56 < Di < 108 nm. A similar increase in the

highest percentage of Do nanotubes at 20 V was from 60 < Do < 88 nm to 85 < Do < 119

nm, 91 < Do < 143 nm and 111 < Do < 161 nm, respectively when the applied potential

increased to 25, 30 and 35 V. The difference in Wt was between 2 to 6 nm by increasing

the applied potential starting from 9 < Wt < 17 nm at 20 V and reaching 17 < Wt < 31

nm at 35 V.

The distribution of Di, Do and Wt for the TiO2-ZrO2-ZrTiO4 nanotubes fabricated on

Ti50Zr alloy resulted from the different oxidation and soluble complex formation of Ti

and Zr with F- ions [240]. The frequency of smaller sized Di and Do decreased on

increasing the applied potential; thus enabling the formation of larger inner and outer

diameters due to the higher anodic current. The broad distribution of Di and Do at 30

and 35 V shows a better condition for oxidation of Ti and Zr because of high current

flow while the dissolution rate of (NH4)2[TiF6] and (NH4)2[ZrF6] remained constant.

In order to find a better statistical distribution of nanotube sizes, Weibull distribution

was studied. There was not obvious difference between normal distribution and Weibull

distribution as detailed in Table 4.1 and Fig. 4.11 for Di of as-formed and Fig. 4.12 for

Di, Do and Wt of annealed TiO2-ZrO2-ZrTiO4 as an example.

92

Table 4.1 The distribution of Di, Do and Wt according to normal and Weibull distribution methods

Applied

Potential (V)

As-fromed Annealed

Di (nm) Do (nm) Wt (nm) Di (nm) Do (nm) Wt (nm)

Normal Weibull Normal Weibull Normal Weibull Normal Weibull Normal Weibull Normal Weibull

20 24<Di<50 Di<41±3 53<Do<89 Do<78±4 9<Wt<15 Wt<13±4 28<Di<52 Di<41±3 60<Do<89 Do<80±5 1<Wt<14 Wt<14±3




93

Fig. 4.11 Histogram and fitted normal and Weibull distribution of Di of as-formed

nanotubular TiO2-ZrO2-ZrTiO4 fabricated at: a) 20 V, b) 25 V, c) 30 V, d) 35 V and e)

normal distribution of all four conditions

94

Fig. 4.12 Histogram and fitted normal and Weibull distribution of Di, Do and Wt of

annealed nanotubular TiO2-ZrO2-ZrTiO4 fabricated at a) 10 V, b) 15 V, c) 20 V

These distributions of nanotubes sizes affected the physical and chemical properties of

nanotubular surface.

The EDS results of the as-formed TiO2-ZrO2-ZrTiO4 nanotubes showed that there was a

layer of TiO2 and ZrO2. Fig. 4.13(b) shows the EDS result measured at the bottom of

the nanotubular layer. The oxygen and fluorine concentrations were higher than those

at the top of the nanotubular layer as shown in Fig. 4.13(a).

95

Fig. 4.13 EDS analysis for the (a) top and (b) bottom of TiO2-ZrO2-ZrTiO4 nanotubes

formed in 0.5 wt % NH4F at 20 V after 2 h

4.3.2 Surface roughness of the TiO2-ZrO2-ZrTiO4 nanotubular surface

As described in Chapter 3 there are three types of parameters for characterising surface

topography: (i) the amplitude parameters, (ii) the spacing parameters, and (iii) hybrid

parameters. These parameters can be determined with a 3D-Profilometer that is coupled

with SurfVision software. Amplitude parameters such as Mean Roughness (Sa) and Root

Mean Square (RMS) Roughness (Sq) that measure the vertical characteristics of surface

deviations have been used in this chapter and evaluated over the complete 3D surface.

These parameters represent an overall description of the surface textures that enable the

differentiation of peaks, valleys, and the spacing of surface textures [210]. The

roughness amplitude parameters of Sa and Sq increased with increasing both the CF-

from 0.3 to 0.5 wt % and the applied potential from 5 to 20 V (Figs. 4.14(b) and

4.14(d)). As it can be seen in Fig. 4.14(b), the difference between Sa of nanotubular

layer fabricated in CF- = 0.4 and CF

- = 0.5 wt % was higher than the difference between

Sa of nanotubular layer fabricated in CF- = 0.3 and CF

- = 0.4 wt % according to the

different chemical and electrochemical reaction rates of titanium and zirconium in the

alloy. The roughness parameters of nanotubular layer formed at the applied potential 20

to 35 V were correspondingly presented and discussed in Chapters 6 and 7.

96


nanotubular surfaces of Ti50Zr alloy as a function of: (a), (b) F- concentration and (c),

(d) applied potential, respectively


nanotubular surfaces of CP-Ti as a function of: (a), (b) F- concentration and (c), (d)

applied potential, respectively

-2

3

8

13

18

23

0.2 0.3 0.4 0.5 0.6

Mea

n W

.C.A

(θ°)

CF- (wt%)

Ti50Zr

W.C.A-As formed

W.C.A-Annealed

a)

0.2

0.7

1.2

1.7

2.2

2.7

3.2

3.7

0.2 0.3 0.4 0.5 0.6

Mea

n S a

(µm

)

CF- (wt%)

Ti50Zr

Sa-As formedSa-Annealed

b)

-5

0

5

10

15

20

25

0 5 10 15 20 25

Mea

n W

.C.A

(θ°)

Applied Potencial (V)

Ti50Zr

W.C.A-As formed

W.C.A-Annealed

c)

0.2

0.7

1.2

1.7

2.2

2.7

3.2

3.7

0 5 10 15 20 25

Mea

n S a

(µm

)


Ti50Zr

Sa-As formed

Sa-Annealed

d)

5

10

15

20

25

30

35

40

0.2 0.3 0.4 0.5 0.6

Mea

n W

.C.A

(θ°)

CF- (wt%)

CP-Ti

W.C.A-As FormedW.C.A-Annealed

a)

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.2 0.3 0.4 0.5 0.6

Mea

n S a

(µm

)

CF- (wt%)

CP-Ti

Sa-As FormedSa-Annealed

b)

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25

Mea

n W

.C.A

(θ°)

Applied Potencial (V)

CP-Ti

W.C.A-As FormedW.C.A-Annealed

c)

0

0.2

0.4

0.6

0.8

1

1.2

0 5 10 15 20 25

Mea

n S a

(µm

)


CP-Ti

Sa-As FormedSa-Annealed

d)

97

The roughness parameter of Sa of TiO2 nanotubes on the surface of CP-Ti increased

with an increase of CF- but revealed lower value than nanotubular layer of TiO2-ZrO2-

ZrTiO4.The Sa of TiO2 nanotubes also increased when the applied potential increased

from 10 to 20V with lower value than TiO2-ZrO2-ZrTiO4 nanotubes.

The nanotubular layers of TiO2-ZrO2-ZrTiO4 grown on the surface of Ti50Zr were

nanoscale in the horizontal dimension. However, the layers exhibited different heights

depending on the microstructure (i.e., different phases of α and β) of the surface due to

the different chemical and electrochemical reaction rates of titanium and zirconium in

the alloy. The surface roughness parameters were on the micrometer scale and displayed

higher values than on CP-Ti before anodisation. The degree of symmetry of the surface

heights about the mean plane is represented by Sskw, the Skewness of a 3D surface

texture, and it is the third central moment of the profile amplitude probability density

function [210]. The sign of Sskw indicates the dominant nature of topography. For

example, Sskw > 0 implies high peaks; whereas Sskw < 0 indicates valley structures such

as deep scratches [210]. Almost the entire as-formed TiO2-ZrO2-ZrTiO4 nanotubular

surface exhibited negative Sskw or a lesser value in comparison to that of the as-formed

TiO2 nanotubes (Table 4.2).

The nature of the height distribution is indicated by Sku, i.e., the kurtosis of the 3D

surface texture that is the fourth central moment of the profile amplitude probability

density function and describes the sharpness of distribution [210]. Sku is 3.0 when the

surface heights are normally distributed. For Sku < 3.0, the distribution is platykurtoic,

indicating a few high peaks and low valleys. For Sku > 3.0, the distribution is

leptokurtoic and indicates many high peaks and low valleys. Almost all of the as-

formed TiO2-ZrO2-ZrTiO4 nanotubular surfaces were platykurtoic and all of the as-

formed TiO2 nanotubular surfaces were leptokurtoic (Table 4.2); thereby reinforcing the

changing natures of the topography dependent of the forming mechanism.

The surface index (SI) is calculated using Surfvision software by the following equation:

SI = Sp/SL (4-10)

where SP is projected surface area that is the total exposed three-dimensional surface

area being analysed, including peaks and valleys and SL is the lateral surface area that is

measured in the lateral direction. The surface index SI increased significantly due to the

98

increase in the nanotube size and the surface roughness. As-formed TiO2-ZrO2-ZrTiO4

nanotubes displayed a higher SI than that of the as-formed TiO2 nanotubes. The surface

area and volume index, roughness amplitude parameters of Sskw and Sku of the

nanotubular surfaces of Ti50Zr anodised under various conditions are listed in Table

4.2.

99

Table 4.2 Surface area index SI and roughness amplitude parameters of Sskw and Sku of the nanotubular surfaces of Ti50Zr and CP-Ti anodised under

various conditions, Note: Each data point is an average of five measurements

Anodisation conditions

As-formed Annealed

Surface area index

SI Sskw Sku Surface area index Sskw Sku

0.3 - 20V Ti50Zr 4.96 ± 0.73 -1.79 ± 0.43 6.64 ± 2.18 4.34 ± 0.22 -0.23 ± 0.08 21.05 ± 2.81

CP-Ti 1.57 ± 0.07 1.34 ± 0.59 8.11 ± 6.08 1.47 ± 0.05 -0.85 ± 0.41 3.84 ± 1.35

0.4 - 20V Ti50Zr 8.05 ± 1.59 -2.98 ± 0.66 16.85 ± 6.36 5.58 ± 1.13 -2.52 ± 1.13 39.44 ± 3.85

CP-Ti 1.80 ± 0.06 0.67 ± 0.22 3.40 ± 0.27 2.12 ± 0.06 -0.18 ± 0.18 2.49 ± 0.19

0.5 - 20V Ti50Zr 13.18 ± 0.38 0.61 ± 0.02 2.11 ± 0.03 3.12 ± 0.06 -1.43 ± 0.12 16.60 ± 1.6

CP-Ti 1.71 ± 0.07 -0.67 ± 0.12 3.30 ± 0.12 1.49 ± 0.01 -0.80 ± 0.03 2.97 ± 0.07

0.5 - 15V Ti50Zr 10.99 ± 2.04 0.25 ± 0.37 2.08 ± 0.28 4.156 ± 0.3 -0.93 ± 0.09 17.03 ± 3.49

CP-Ti 2.20 ± 0.13 0.83 ± 0.25 3.38 ± 0.69 2.61 ± 0.13 0.27 ± 0.29 2.37 ± 0.33

0.5 - 10V Ti50Zr 10.69 ± 6.56 0.06 ± 0.34 2.03 ± 0.06 2.66 ± 0.30 -2.02 ± 0.18 12.41 ± 2.5

CP-Ti 1.31 ± 0.03 0.33 ± 0.19 3.68 ± 0.09 1.29 ± 0.02 0.01 ± 0.07 2.75 ± 0.03

0.5 - 5V Ti50Zr 5.29 ± 2.53 0.85 ± 0.45 3.86 ± 2.26 2.53 ± 0.07 -1.15 ± 0.15 10.30 ± 2.33

CP-Ti* - - - - - -

*All the surface was covered by a nanoporous layer

100

4.3.3 Hydrophilic properties of the nanotubular surfaces

Hydrophobicity and hydrophilicity are the terms that are defined by the wetting

properties of a surface by water [241]. Accepted definitions generally consider water

contact angle as follows: (i) superhydrophobic: static water contact angle > 150 º, (ii)

hydrophobic: water contact angle between 90 º and 150 º, (iii) hydrophilic: water

contact angle between 10 º and 90 º, and (iv) superhydrophilic: water contact angle less

than 10 º [242]. Parameters such as type of the material, surface roughness,

heterogeneities of the surface, chemical composition and presence of contamination; as

well as parameters related to the ambient conditions, can influence the wettability of a

surface [241]. The nanotubular surfaces of TiO2-ZrO2-ZrTiO4 nanotubes anodised in all

conditions were hydrophilic. The hydrophilicity of the anodised nanotubular surface did

not change obviously on increasing CF-, but it increased with an increase in the applied

potential. Figures 4.14 and 4.15 show the mean roughness (Sa) and mean water contact

angle (W.C.A.) of the as-formed TiO2-ZrO2-ZrTiO4 nanotubular surfaces, annealed

TiO2-ZrO2-ZrTiO4 nanotubular surfaces, and the as-formed TiO2 and annealed TiO2

naotubular surfaces as a function of the CF- and the applied potential.

The wettability of the nanotubular surface increased when roughness parameters

increased. It is noticeable that the wettability of nanotubular surface on Ti50Zr was

higher than that of the nanotubular surface on CP-Ti. The wettability properties of TiO2-

ZrO2-ZrTiO4 nanotubes formed at the applied potential of 20 to 35 V for 1 h were

presented and discussed in Chapters 6 and 7 relatively.

The surface energy of the nanotubular surface was calculated according to the Owens-

Wendt (Equ. 3-1). Almost all of the as-formed TiO2-ZrO2-ZrTiO4 nanotubular surfaces

exhibited a higher surface energy than those of the as-formed TiO2 nanotubular

surfaces. The surface energies of the as-formed TiO2-ZrO2-ZrTiO4 and annealed TiO2-

ZrO2-ZrTiO4 nanotubular surfaces; as well as those of the as-formed TiO2 and annealed

TiO2 nanotubular surfaces were measured and are detailed in Table 4.3.

101

Table 4.3 Calculated surface energy of TiO2-ZrO2-ZrTiO4 nanotubular surfaces fabricated on Ti50Zr and TiO2 nanotubular surfaces on CP-Ti, Note:

Each data point is an average of five measurements


As-formed Annealed

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

0.3 - 20V Ti50Zr 20.50 ± 0.40 15.98 52.67 68.65 2.70 ± 0.17 14.93 58.71 73.64

CP-Ti 28.68 ± 10.08 22.45 41.87 64.32 22.50 ± 4.82 12.95 55.56 68.51

0.4 - 20V Ti50Zr 18.40 ± 1.04 17.88 51.39 69.27 2.00 ± 1.73 14.15 59.77 73.92

CP-Ti 22.97 ± 8.06 0.58 86.36 86.94 15.30 ± 3.75 13.57 57.84 71.41

0.5 - 20V Ti50Zr 16.03 ± 0.45 16.93 53.37 70.30 1.67 ± 1.44 15.44 58.10 73.54

CP-Ti 24.47 ± 0.45 14.39 52.65 67.04 14.97 ± 5.25 3.33 77.92 81.25

0.5 - 15V Ti50Zr 17.40 ± 0.53 17.89 51.77 69.66 2.40 ± 3.17 14.70 59.03 73.73

CP-Ti 22.00 ± 1.70 11.00 58.58 69.58 19.87 ± 2.70 7.52 65.58 73.10

0.5 - 10V Ti50Zr 19.33 ± 0.61 15.62 53.62 69.24 3.33 ± 0.29 14.42 59.33 73.75

CP-Ti 35.9 ± 11.46 17.44 42.19 59.63 26.67 ± 0.49 12.27 54.19 66.46

0.5 - 5V* Ti50Zr 20.11 ± 1.76 15.99 52.83 68.82 3.67 ± 0.58 14.98 58.57 73.55

CP-Ti* - - - - - - - -

*All the surface was covered by a nanoporous layer

102

4.3.4 Mechanical properties of the TiO2-ZrO2-ZrTiO4 nanotubes

The nanomechanical characteristics of the various as-formed and annealed TiO2-ZrO2-

ZrTiO4 nanotubes were measured using a nanoindentation to obtain the nano hardness

and Young’s modulus. Fig. 4.16 shows the curves of the load-unload forces versus the

nanoindentation depths of the as-formed and annealed TiO2-ZrO2-ZrTiO4 nanotubes

fabricated at different applied potentials.


TiO2-ZrO2-ZrTiO4 and b) annealed TiO2-ZrO2-ZrTiO4 fabricated at the applied potential

20 to 35 V

It is worth noting that the nano Young’s modulus and hardness can be determined

instantaneously as a function of depth. Using the method described in Chapter 3, the

hardness, reduced elastic modulus and elastic modulus were calculated and presented in

Fig. 4.17.

0

0.002

0.004

0.006

0.008

0.01

0.012

-5E-10 0 5E-10 1E-09 1.5E-09 2E-09 2.5E-09

Load

(N)

Displacement (m)

As Formed nanotubes

20V

25V

35V

(a)

0

0.002

0.004

0.006

0.008

0.01

0.012

-5E-10 0 5E-10 1E-09 1.5E-09 2E-09

Load

(N)

Displacement (m)

Annealed Nanotubes

20V

25V

35V

(b)

103

Fig. 4.17 The nano mechanical properties of: a) hardness, b) reduced elastic Modulus

and c) elastic Modulus of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at different applied

potential

The hardness of nanotubular layer of TiO2-ZrO2-ZrTiO4 was maximum 211.8 and 336.2

MPa before and after annealing which were lower than its respective bare metal which

104

revealed 4.38 GPa. The elastic modulus of the nanotubular layer of TiO2-ZrO2-ZrTiO4

was maximum 149.9 and 248.7 GPa before and after annealing which were higher than

its respective bare metal which revealed 102.9 GPa. It can be concluded that the

porosity of the nanotubes influenced the hardness and the ceramic properties of

nanotubular layer influenced elasticity of the bare metal. By increasing the applied

potential and consequently the nanotube size the hardness and elasticity of nanotubes

decreased.

4.3.5 Effect of annealing on the TiO2-ZrO2-ZrTiO4 nanotubes

TiO2 and ZrTiO4 nanotubes with an amorphous structure and ZrO2 with an

orthorhombic structure were fabricated via anodisation under the conditions of 0.5 wt %

NH4F, 20 V for 2 h. The amorphous TiO2 and ZrTiO4 nanotubes transformed into (i)

tetragonal anatase TiO2, (ii) a mixture of orthorhombic TiO2 and ZrO2 similar to

srilankite, and (iii) orthorhombic ZrTiO4 after annealing at 500 ºC for 3 h; as indicated

by the XRD patterns in Fig. 4.18. Varghese et al. [23] reported that amorphous TiO2

nanotubes transformed to anatase and rutile preferably along the length and the

curvatures of the nanotubes rather than wall thickness by annealing. A crystalline cubic

phase of ZrO2 accompanied by the phase of TiZrO4 or ZrTiO4 formed on the surface of

Ti50Zr alloy via anodisation [112, 132].

105


via anodisation. (a) as-formed amorphous TiO2 and ZrTiO4 and orthorhombic ZrO2; (b)

annealed at 500 °C for 3 h tetragonal anatase, srilankite (a mixture of orthorhombic

TiO2 and ZrO2) and orthorhombic ZrTiO4

After annealing at 500 °C for 3 h some changes in inner and outer diameter and wall

thickness of the nanotubes formed on Ti50Zr via anodisation were observed which can

explain the formation of crystalline phases of TiO2-ZrO2-ZrTiO4. Annealing also caused

significant diversity in the roughness parameters and hydrophilic properties of the

nanotubular surfaces fabricated under different CF- and applied potential conditions.

After annealing the values of Sa and Sq decreased for the nanotubular surface fabricated

at each applied potential of 5 to 20 V. The surface parameters increased when the

applied potential increased from 5 to 10 V, but decreased when the applied potential

increased from 15 to 20 V. Almost all the annealed TiO2-ZrO2-ZrTiO4 nanotubular

surfaces exhibited more negative Sskw values compared to the annealed TiO2

nanotubular surfaces, although both exhibited cracks after annealing that would have

106

influenced the results. After annealing, the value of Sku of the TiO2-ZrO2-ZrTiO4

nanotubular surfaces increased but those for TiO2 nanotubular surfaces decreased.

Annealed TiO2-ZrO2-ZrTiO4 nanotubular surfaces exhibited a higher surface index than

the annealed TiO2 nanotubular surfaces.

The hydrophilicity of the TiO2-ZrO2-ZrTiO4 nanotubular surfaces on Ti50Zr increased

significantly after annealing for both anodising conditions; i.e., for increasing CF- and

applied potential. That is, the hydrophilicity increased with increases in both the Di and

Do. After annealing, almost all the TiO2-ZrO2-ZrTiO4 nanotubular surfaces anodised

under all the conditions exhibited superhydrophilic properties. In contrast, the TiO2

nanotubular surfaces fabricated on CP-Ti showed hydrophilic properties. In general, the

surface energy of the annealed TiO2-ZrO2-ZrTiO4 nanotubular surfaces was higher than

those of annealed TiO2 nanotubular surfaces and the as-formed TiO2-ZrO2-ZrTiO4

nanotubular surfaces. This result was reflected in lower water contact angles.

The hardness and elasticity of nanotubular layer increased by annealing although

increasing the porosity resulted to a decrease of hardness and elasticity.

X-ray photoelectron spectroscopy (XPS) was used to evaluate the chemistry of the

nanotubes. The XPS spectra of as-formed and annealed TiO2-ZrO2-ZrTiO4 are shown in

Fig. 4.19.

Fig. 4.19 XPS spectra for the as-formed and the annealed TiO2-ZrO2-ZrTiO4 nanotubes.

(a) O1s, (b) Ti 2p, and (c) Zr 3d

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

522524526528530532534536538540

Inte

nsity

( a.

u.)

Binding energy (eV)

a)

Annealed

As Formed

O 1s

0

5000

10000

15000

20000

25000

30000

450452454456458460462464466468470472

Inte

nsity

(a.u

.)

Binding energy (eV)

b)

Annealed

As Formed

Ti 2p

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

174176178180182184186188190192194

Inte

nsity

(a.u

.)

Binding energy (eV)

Annealed

As Formed

c)Zr 3d

107

The nanotubular surface consisted of 48.55 at % O, 13.43 at % Ti, 8.64 at % Zr for the

as-formed nanotubes, and 51.23 at % O, 14.57 at % Ti and 8.35 at % Zr after annealing.

The remaining atomic percentages arose from carbon contamination and remnant

fluorine. The F residue after annealing was 0.15 at % in comparison to 3.23 at % for the

as-formed nanotubes. The Ti 2p3/2 and Zr 3d5/2 peaks have binding energies of 458.7 -

458.8 and 182.5 -182.7 eV, respectively, which represent the fully oxidised titanium ion

in its Ti4+ and zirconium ion in its Zr4+ state. A small splitting of the O 1s spectrum at

532.27 (as-formed) and 532.21 eV (annealed) showed the existence of ZrTiO4. This is

because of the different coordinations of oxide ions (Zr-O-Zr, Ti-O-Ti and Zr-O-Ti)

[243] as well as shifting to lower binding energy (Zr 3d5/2) due to phase transformation

during annealing.

4.4 Conclusions

Nanotubes formed on Ti50Zr alloy via anodisation have potential biomedical

applications such as promoting cell interaction; biomedical coatings; and drug delivery

and release of other payloads. The different phase components of Ti50Zr alloy led to the

formation of TiO2-ZrO2-ZrTiO4 nanotubes that are mostly microscopically needlelike

and separated from each other by an interspace. A distribution of nanotubes with

different sizes including inner diameter Di, outer diameter Do and wall thickness Wt,

resulting from the different zirconium contents in the α and β phases of the alloy, was

observed. The effects of the F- concentration (CF-) in the electrolyte, the applied

potential on the nanotube characteristics were revealed. The conclusions are as follows:

1. Because of the different chemical and electrochemical reaction rates of titanium

and zirconium with O2- and F- in the electrolytes, an increase in the CF- led to an

increase in the number of nanotubes with smaller Di, Do and Wt.

2. Increasing the applied potential has the consequence of increasing the anodic

current. The number of nanotubes with smaller Di and Do decreased; the number

of nanotubes with larger Di and Do increased; and the Wt of the nanotubes

fabricated at 5, 10, 15 and 20 V increased slightly. After annealing the mean of

Di and Do increased at higher voltage (e.g. 15 and 20 V) and Wt did not change

much.

108

3. The mean Di, Do and Wt of nanotubes fabricated on the surface of Ti50Zr

increased with an increase of applied potential from 20 to 35 V. Annealing

resulted in decreasing Di and Do; and Wt did not change.

4. The mean Di and Do of nanotubes fabricated on the surface of CP-Ti also

increased with an increase of applied potential from 20 to 35 V. But the Wt did

not show obvious change. Annealing resulted in decreasing Di and Do; and Wt

did not change.

5. According to little difference between normal distribution and Weibull

distribution it can be said that the normal distribution was an appropriate method

to represent the distribution of nanotube sizes of TiO2-ZrO2-ZrTiO4.

6. The Mean Roughness (Sa) and the Root Mean Square (RMS) Roughness (Sq) of

the nanotubular surfaces increased on increasing both the CF- and the applied

potential.

7. The hydrophilic properties of the anodised TiO2-ZrO2-ZrTiO4 nanotubular

surface of Ti50Zr increased with an increase in the CF- and the applied potential,

which is related to changes in the distribution of the nanotube size. The

hydrophilic properties of these surfaces increased when the roughness

parameters of the surface increased. The surface energy followed the same

increasing trend of hydrophilic properties. The TiO2-ZrO2-ZrTiO4 nanotubular

surfaces of Ti50Zr exhibited higher surface energy than those of TiO2

nanotubular surfaces on CP-Ti.

8. The surface area index SI increased with increasing CF- and applied potential,

which is linked to changes in the nanotube size distribution.

9. The hardness of bare metal decreased by the porosity of fabricated nanotubular

layer via anodisation while the elasticity increased by ceramic properties of the

oxide layer. Porosity affected the trend of hardness and elasticity of TiO2-ZrO2-

ZrTiO4 nanotubes.

10. Annealing changed the phase structure, surface roughness, hydrophilic property

and surface energy; i.e., (i) the amorphous nanotubes changed to a mixture of

anatase, orthorhombic TiO2 and ZrO2, and orthorhombic ZrTiO4, (ii) Sa and Sq

decreased, (iii) the hydrophilicity and surface energy increased, and (iv)

hardness and elasticity of nanotubular layer increased.

109

11. The mixed nanotubes of TiO2-ZrO2-ZrTiO4 on Ti50Zr alloy with a variety of

nanotube size distribution and other related properties such as hydrophilicity and

roughness are more promising for biomedical applications than the 100 % TiO2

nanotubes on CP-Ti.

110

Chapter 5 Nanotubes formed in non-aqueous electrolyte

Abstract

A high surface area of nanotubular layer of metal oxide fabricated via anodisation on

the surface of biocompatible titanium and titanium alloys offers high potential for

effective conditions for implant applications. In this chapter, the morphology, and the

physical and chemical characteristics of four groups of TiO2-ZrO2-ZrTiO4 nanotubes

that were fabricated via anodisation in a non-aqueous electrolyte were investigated.

Their nanoscale dimensional characteristics (i.e., inner diameter Di, outer diameter Do

and wall thicknesses Wt) were evaluated. The microstructure, morphology, hydrophilic

and mechanical properties of the nanotubes were characterised using scanning electron

microscopy (SEM), 3D-Profilometry, goniometer and nanoindentation. The applied

potential during anodisation and the water content of the non-aqueous electrolyte

influenced the shape and the profile of the titanium and zirconium oxide layer, resulting

in different nanoscale characteristics for the nanotubes. The oxidation and dissolution

rates were competing, resulting in different surface roughness parameters although these

rates are low in a non-aqueous electrolyte.

5.1 Introduction

Compared to other surface treatments such as hydrothermal [244], seeded growth [245]

and template-assisted deposition [246] for fabricating TiO2 nanotubes, anodisation is

simpler and reproducible. These nanotubes are more uniform and their characteristics

and morphology can be more easily controlled. It has been almost a decade when first

nanotubular layer of titanium dioxide fabricated via anodisation [10]. Several

generations of TiO2 nanotubes were formed by changing anodisation conditions [57, 61,

65, 66]. Apart from the applied potential which mostly influences the inner and outer

diameter and wall thickness of nanotubes [61] and the anodisation time which mostly

influences the length of nanotubes [62], the type of the electrolyte influences the

morphology of the nanotubular layer [73, 80, 247, 248]. In order to obtain more uniform

nanotubes with high aspect ratio several attempts has been undertaken through changing

the conditions of the electrolyte [67, 72, 79, 110, 132].

One attempt was to tailor the electrochemical conditions to enhance the aspect ratio of

the self-organised TiO2 nanotubes [67]. In this study, the pH profile within the tube was

111

constructed to modified the geometry of the TiO2 nanotubes using anodic oxidation in

buffered electrolyte (NH4F and (NH4)2SO4 or NaF and Na2SO4). Although F- is one of

the essential ions to fabricate nanotubes it reduces the length of the nanotubes up to

several hundred nanometer in an acidic pH with the same reason (dissolution of oxide

layer). The thickness of nanotubular layer was 2.5 µm with nm inner diameter in

buffered electrolyte [67]. Another attempt was tailoring the anodisation condition by

using non-aqueous electrolyte and smoother and longer nanotubes were fabricated in

this condition [79].

In a polar electrolyte such as ethylene glycol the oxide chemical dissolution is limited to

oxygen amount of its water (H2O) content [68]. As a result, fabricating a nanotubular

layer with the same thickness as a nanotubular layer formed in an aqueous electrolyte

takes a longer time. Furthermore, the viscosity, whether stirring [76] and the H2O

content of the electrolyte influence the formation of the nanotubes. It has been reported

that nanotubes with a diameter of 20 to 160 nm and a maximum height of 45 µm [68] or

higher (>150 µm) [77] were obtained at an applied potential ranging from 30 to 120 V

using an organic electrolyte such as glycerol [76, 80], ethylene glycol [68, 78] and

acetic acid [79]. When NH4F in glycerol was used as the electrolyte the TiO2 nanotubes

were smooth and not connected by outside walls [80]. The diffusion constant which is

the key factor of anodisation is relatively low in such an electrolyte. This has an

influence on the pH value of the pore tip and consequently on the current transient and

dissolution rate of ions.

This chapter describes how changes in the water content (H2O) of a non-aqueous

electrolyte and applied potential influenced the characteristics including the tube

diameter, roughness and hydrophilic and mechanical properties of a nanotubular TiO2-

ZrO2-ZrTiO4 layer. The nanotubular layers of TiO2-ZrO2-ZrTiO4 were fabricated with

mean Di (i) 29 ± 5, (ii) 35 ± 5, (iii) 34 ± 8 nm and (iv) 42 ± 10; mean Do (i) 51 ± 5 nm,

(ii) 61 ± 7, (iii) 59 ± 11 and (iv) 78 ± 10 nm; and mean Wt (i) 12 ± 1, (ii) 16 ± 1, (iii) 11

± 1 and (iv) 18 ± 1 nm, respectively. The roughness and water contact angle of different

nanoscale morphologies were measured and their surface energies were calculated.

112


The substrates for fabrication nanotubular layer was Ti50Zr alloy (wt % hereafter) discs

with 8 mm diameter and 2 mm thickness cut from a cast ingot by electrical discharge

machining (EDM) and titanium foils (10×10×0.05 mm). Prior to anodisation the disc

samples were first polished to a 1 µm finish with diamond paste. All the samples were

cleaned by sonification successively in methanol, isopropanol, acetone and ethanol for

15 min each, afterwards rinsed with deionised water and dried in a stream of nitrogen

gas.

The Ti50Zr discs were exposed to an electrolyte through spot-welded to titanium foil

and contacted to the anode of an electrochemical cell. The plate shape platinum counter

electrode with a 1 cm2 surface area was placed 4 cm from the samples in a two-

electrode configuration using a DC power supply. The electrolyte was composed of

0.5 wt % of NH4F with the addition of 5 and 10 wt % of water (H2O) in ethylene glycol

(Sigma-Aldrich reagent grades). After the electrochemical treatment the sample surfaces

were rinsed with acetone for 5 min and followed by drying in air.

Two applied potentials, i.e., 20 and 30 V were used for each electrolyte composition to

create different sizes and morphologies of nanotubes. The anodised TiO2-ZrO2-ZrTiO4

nanotubes were post-treated by immersion in 0.5 M HCl at room temperature for 1 min,

and then rinsed with deionised water for 1 min to dissolve the remained upper

nanoporous layer. The samples were then heat-treated at 500 ˚C for 2 h at the heating

rate of 2 ˚C/min in a conventional muffle furnace (Nabertherm LT15/13/P330;

Nabertherm GmbH, Lilienthal, Germany) to crystallise the amorphous nanotubular

layer. The inner diameter, outer diameter and wall thickness of the anodised nanotubes

were denoted Di, Do and Wt, respectively. Each data point is an average of 100

measurements. on different positions for each of three samples (300 measurements).

A field-emission scanning electron microscope (FESEM, ZEISS SUPRA 40 VP)

equipped with energy dispersive X-ray spectroscopy (EDS) was used for

microstructural characterisation. A 3D-Profilometer (Bruker, Contour GT-K1; Bruker

Pte Ltd, Singapore) that was coupled with SurfVision software (Veeco Instruments Inc.;

Plainview, NY, USA) was used to evaluate the amplitude parameters of the anodised

TiO2-ZrO2-ZrTiO4 nanotubes. The vertical characteristics of surface deviations,

113

amplitude parameters over the complete 3D surface [210], were measured and presented

by Mean Roughness (Sa) and Root Mean Square (RMS) Roughness (Sq). The symmetry

and nature of the surface heights represented by Sskw, the skewness of a 3D surface

texture, and Sku, the kurtosis of the 3D surface texture, are the third and fourth central

moment of the profile amplitude probability, respectively [210]. The water contact

angle was measured using a goniometer (NRL C.A. Goniometer, Ramé-hart, Inc.;

Succasunna, NJ-USA) and the surface energy was calculated based on the Owens-

Wendt (OW) method [211], given by:

(1 + cos θ )γL=2(√γLd γS

d+√γLp γS

p) (3 − 1)

where L indicates the liquid surface tension (water = 72.8 mJm-2, glycerol = 63.4 mJm-

2); Ld and S

d are the liquid and solid dispersive component (water = 21.8 mJm-2,

glycerol = 37.0 mJm-2). The variables Lp and S

p are indicative of the liquid and solid

polar components (water = 51.0 mJm-2, glycerol = 26.4 mJm-2) and S is the sum of Sd

and Sp [212].


5.3.1 Formation and characterisation of TiO2-ZrO2-ZrTiO4 nanotubes

Titania-zirconia-zirconium titanate (TiO2-ZrO2-ZrTiO4) nanotubes were fabricated on

the surface of Ti50Zr alloy via anodisation in a non-aqueous electrolyte (ethylene

glycol) which was consisted of 5 and 10 wt % of water (H2O) and 0.5 wt % NH4F. The

potential of 20 and 30 V were applied to samples for each composition of electrolyte for

90 min. The effect of the water content in the electrolyte on nanotubes size,

morphology, hydrophilic and mechanical properties were investigated. Fig. 5.1 shows

the SEM images of the top view of the fabricated nanotubes on the Ti50Zr alloy in non-

aqueous electrolyte.

114

Fig. 5.1 SEM images of top view of nanotubes fabricated at: a) 5 wt % H2O at 20 V, b)

10 wt % H2O at 20V, c) 5 wt % H2O at 30 V and d) 10 wt % H2O at 30V in ethylene

glycol

There were a distribution of size of Di, Do and Wt of nanotubes (Fig. 5.2).

116

Fig. 5.2 Histograms of TiO2-ZrO2-ZrTiO4 nanotube parameters anodised in 0.5 wt % NH4F and ethylene glycol consisting of 5 and 10 wt % H2O for 90 min at

20 and 30V: (a, d) inner diameter (Di) of as-formed and annealed nanotubes, (b, e) outer diameter (Do) of as-formed and annealed nanotubes, (c, f) wall

thickness (Wt) of as-formed and annealed, respectively

117

According to normal distribution the highest percentage of nanotube were evaluated and

presented in Table 5.1. To find better statistical distribution of nanotube sizes, Weibull

distribution was also studied. Fig. 5.3 shows that the difference between normal

distribution and Weibull distribution. It can be seen that the Weibull distribution can be

negligible and normal distribution can be an acceptable method to present.

118

Table 5.1 The distribution of Di, Do and Wt according to normal and Weibull distribution methods

Water Content

(wt%) and

Applied Potential

(V)

As-fromed Annealed

Di (nm) Do (nm) Wt (nm) Di (nm) Do (nm) Wt (nm)

Normal Weibull Normal Weibull Normal Weibull Normal Weibull Normal Weibull Normal Weibull

5- 20 24<Di<34 Di<31±6 46<Do<56 Do<54±11 11<Wt<12 Wt<12±24 16<Di<27 Di<24±4 42<Do<55 Do<51±8 12<Wt<15 Wt<14±12

10 - 20 29<Di<40 Di<37±6 53<Do<68 Do<64±8 15<Wt<17 Wt<16±31 25<Di<37 Di<33±6 50<Do<62 Do<59±10 16<Wt<20 Wt<19±12



120

Fig. 5.3 Histogram and fitted normal and Weibull distribution of Di, Do and Wt of as-formed and annealed nanotubular TiO2-ZrO2-ZrTiO4 fabricated at a,e) 5 wt

% H2O, 20 V, b, f) 10 wt % H2O, 20 V c, g) 5 wt % H2O, 30 V, d,h) 10 wt % H2O, 30 V

121

The walls of neighbouring nanotubes were not separated, although increasing water

content helped showing the needle-like microstructure of the underneath alloy.

Figure 5.4 presents the mean size of Di, Do and Wt of the fabricated nanotubues before

(Fig. 5.4(a)) and after annealing (Fig. 5.4(b)). The Di, Do and Wt of fabricated TiO2-

ZrO2-ZrTiO4 nanotubues increased when the water content of the organic electrolyte

increased from 5 to 10 wt % at the same applied potential. Increasing the applied

potential from 20 V to 30 V resulted in an increase in Di and Do but Wt did not increase

for each composition of electrolyte. After annealing Di of nanotubes decreased while the

change of Do was negligible. The thickness of the tube walls increased after annealing.

The Di, Do and Wt of the as-formed TiO2 nanotubes shown in Fig. 5.4(c) increased when

the water content of the organic electrolyte increased from 5 to 10 wt % at the same

applied potential. It can be observed that increasing the applied potential from 20 to 30

V resulted in an increase in Di, Do and Wt of the as-formed TiO2 for each composition of

electrolyte. The mean sized of nanotubes fabricated on the surface of CP-Ti were larger

than nanotubes fabricated on the surface of Ti50Zr. After annealing (Fig. 5.4(b) and (d))

Di of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes decreased, Do remained almost constant

and Wt increased. The crystallisation of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes occurred

at walls and grew towards the centre of nanotubes. The nanotubular surface fabricated

under non-aqueous electrolyte conditions showed a brittle layer after annealing which

was not suitable for further investigation.

122

Fig. 5.4 Mean nanotube sizes of Di, Do and Wt: a,c) as-formed and b,d) annealed TiO2-

ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in two different water contents and two

different applied potentials in ethylene glycol

Once a constant potential, i.e. 20 V was applied to the Ti50Zr disk during anodisation, it

was oxidised accordingly and a layer of TiO2-ZrO2-ZrTiO4 formed at the interface of

the substrate and electrolyte. It was described before [249] that the surface of this layer

randomly broke due to the formation of [TiF6]2- and [ZrF6]2- which are soluble

complexes in water. These breakdown sites were the places where the nanotubes and the

interspace between nanotubes started to grow as non-uniform nanoporous layer. The

Ti4+ and Zr4+ which were the production of oxidation attended at interface of substrate

and oxide layer at the bottom of these pores in every step of the process. The F- ions

which have higher diffusion rate than O2- reached at the bottom of the pores and formed

the soluble complex and took their places to the new F- ions and left the bottom of the

pores to the top because of diffusion flux. These oxidation and dissolution which were

equilibrated continued up to the end of anodisation time. In all these processes there was

always competition between Ti and Zr to be oxidised and making soluble complex. It

0

20

40

60

80

100

120

5%-20 10%-20 5%-30 10%-30

Nan

otub

e si

ze (n

m)

Water content (wt%)and applied potential (V)

Ti50Zr

Wt-As Formed

Di-As Formed

Do-As Formed

a)

0

20

40

60

80

100

120

5%-20 10%-20 5%-30 10%-30

Nan

otub

e si

ze (n

m)


Ti50Zr

Wt-Annealed

Di-Annealed

Do-Annealed

b)

0

20

40

60

80

100

120

5%-20 10%-20 5%-30 10%-30

Nan

otub

e si

ze (n

m)


CP-Ti

Wt-As Formed

Di-As Formed

Do-As Formed

c)

0

20

40

60

80

100

120

5%-20 10%-20 5%-30 10%-30N

anot

ube

size

(nm

)Water content (wt%)and applied potential (V)

CP-Ti

Wt-Annealed

Di-Annealed

Do-Annealed

d)

123

was Zr that was oxidised and formed the complex first because the standard electrode

potential of Zr→Zr4+ and Ti→Ti4+ are 1.553 V and 2.132 V [236] and the standard

enthalpies of formation of ZrO2 and TiO2 are -264.199 and -228.360 gram calories per

mole, respectively [237]. In aqueous electrolyte this completion resulted in obvious

different nanotube size and height on the α and β phases of the Ti50Zr alloy. But in non-

aqueous electrolyte the rate of these reactions are low and depend on the water content

of the electrolyte. Although ethylene glycol is a polar solvent, its polarity (79) is less

than water (100) and also has lower dielectric constant (37.7) than water (79.7).

Therefore, both oxidation rate and dissolution rate are limited and Ti and Zr have the

same chance to be oxidised to Ti4+ and Zr4+ and dissolve as [TiF6]2- and [ZrF6]2-. In this

case although there was difference in nanotube size and height but they were not largely

distributed and the nanotubues were not separated.

A layer of TiO2 and ZrO2 was observed at the bottom of the as-formed nanotubular

layer of TiO2-ZrO2-ZrTiO4 which has shown in Fig. 5.5 according to the EDS results.

The oxygen and fluorine weight percent of this layer were lower than those at the top of

the nanotubular layer.

Fig. 5.5 EDS analysis for: (a) top and (b) bottom of TiO2-ZrO2-ZrTiO4 nanotubes

formed in 0.5 wt % NH4F, 5 wt % H2O in ethylene glycol at 20 V after 90 min

Figure 5.6 shows the effect of applied potential and anodisation time on nanotube

length. The nanotubes length increased with an increase of anodisation time up to 20 h.

The length also increased when the applied potential increased from 20 to 35 V in the

same electrolyte. The time of anodisation exhibited more significant effect on nanotube

124

growth than applied potential. The length of nanotubes also increased when the water

content of the electrolyte increased from 5 to 10 wt % at the same applied potential and

anodisation time.

Fig. 5. 6 The effect on TiO2-ZrO2-ZrTiO4 nanotubes length formed in organic

electrolyte by: a) anodisation time and b) applied potential

Fabricated TiO2 and ZrTiO4 nanotubes revealed amorphous structure and ZrO2 revealed

orthorhombic structure after anodising in non-aqueous electrolyte. After annealing at

500 °C for 2 h, the amorphous TiO2 and ZrTiO4 nanotubes transformed into a mixture

of tetragonal anatase and rutile TiO2, a mixture of orthorhombic TiO2 and ZrO2 similar

to srilankite and orthorhombic ZrTiO4. Fig. 5.7 presents the XRD patterns of as-formed

and annealed nanotubular layer fabricated in non-aqueous electrolyte.

125


via anodisation in non-aqueous electrolyte. (a) as-formed amorphous TiO2 and ZrTiO4

and orthorhombic ZrO2; (b) annealed at 500 °C for 2 h tetragonal anatase, rutile,

srilankite (a mixture of orthorhombic TiO2 and ZrO2) and orthorhombic ZrTiO4

5.3.2 Surface roughness and hydrophilic property of the TiO2-ZrO2-ZrTiO4

nanotubular surface

Figure 5.8 shows roughness parameters of Sa and Sq measured for the as-formed

nanotubular surfaces of TiO2-ZrO2-ZrTiO4 (Fig. 5.8(a)) and TiO2 (Fig. 5.8(c)). These

amplitude parameters (Sa and Sq) of TiO2-ZrO2-ZrTiO4 nanotubes increased with an

increase of water content for each applied potential. Increasing the applied potential for

each composition of electrolyte resulted in increasing of Sa and Sq. The Sa and Sq did

126

not change for TiO2 nanotubes when the water content of the electrolyte increased. The

amplitude parameters of Sa and Sq increased when the applied potential increased for

each composition of electrolyte. The nanotubular layer of TiO2 revealed lower

roughness than the nanotubular layer of TiO2-ZrO2-ZrTiO4.

Fig. 5.8 a), c) The mean roughness (Sa), (Sq) and b), d) the mean water contact angle

(W.C.A.) of the nanotubular surfaces of TiO2-ZrO2-ZrTiO4 and TiO2 respectively, Note:


The TiO2-ZrO2-ZrTiO4 nanotubes fabricated in the electrolyte with 5 wt % H2O at 20 V

with mean Di = 29 nm exhibited Sskw ≤ 0 and Sku ≥ 3, indicating uniform symmetry

normally distributed of surface heights, whilst the other 3 nanotubular surfaces

exhibited Sskw < 0 and Sku > 3, indicating leptokurtoic distribution with many high peaks

and valleys. The TiO2 nanotubes fabricated in the electrolyte with 10 wt % H2O at 30 V

with mean Di = 52 nm exhibited Sskw ≈ 0 and Sku ≈ 3, indicating uniform symmetry

normally distributed of surface heights, while the other 3 nanotubular surfaces exhibited

Sskw > 0 and Sku > 3, indicating leptokurtoic distribution with many high peaks and

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

29 (5 wt%H2O-20V) 35 (10 wt% H2O-20V) 34 (5 wt%H2O-30V) 42 (10 wt%H2O-30V)

Mea

n Sa

(µm

)

Mean Di (nm)

Roughness-Ti50Zr Sa-As Formed

Sq-As Formeda)

0

2

4

6

8

10

12

14

16


Mea

n W

.C.A

. (θ°

)Mean Di (nm)

Wettability-Ti50Zr W.C.A-As Formedb)

0

0.1

0.2

0.3

0.4

0.5

0.6


Mea

n Sa

(µm

)

Mean Di (nm)

Roughness-CP-Ti Sa-As Formed

Sq-As Formedc)

02468

1012141618


Mea

n W

.C.A

. (θ°

)

Mean Di (nm)

Wettability-CP-Ti W.C.A-As Formedd)

127

valleys. Upper layer of nanotubes were cracked after annealing at 500 °C for 2 h and it

was impossible to measure the roughness parameter and water contact angle.

The surface index (SI) is calculated using Surfvision software using the following

equation:

SI = Sp/SL (5-1)

The total exposed three-dimensional surface area being analysed (SP) is the projected

surface area, including peaks and valleys and SL is the lateral surface area that is

measured in the lateral direction.

Table 5.2 listed the surface index with other amplitude parameters of the TiO2-ZrO2-

ZrTiO4 and TiO2 nanotubes fabricated using four different water contents and applied

potentials. The nanotubular layer of TiO2-ZrO2-ZrTiO4 revealed higher surface area

index (SI) than nanotubular layer of TiO2 because of different height of TiO2 and ZrO2

nanotubes resulted of their different oxidation and dissolution rates.

128




Water Content (wt%) and Applied

Potential (V)

As-formed

Surface area index SI Sskw Sku

5 - 20 Ti50Zr 2.42 ± 0.95 -0.74 ± -0.57 4.56 ± 2.46

CP-Ti 1.14 ± 0.01 0.44 ± 0.10 5.60 ± 0.28

10 - 20 Ti50Zr 4.32 ± 0.11 -4.69 ± 0.18 31.94 ± 1.88

CP-Ti 1.37 ± 0.04 1.36 ± 0.16 7.36 ± 1.11

5 - 30 Ti50Zr 4.12 ± 0.96 -5.71 ± 2.15 49.38 ± 35.30

CP-Ti 1.82 ± 0.09 0.80 ± 0.30 3.58 ± 0.67

10 - 30 Ti50Zr 4.52 ± 0.92 -4.67 ± 0.53 31.75 ± 7.34

CP-Ti 1.76 ± 0.03 0.16 ± 0.09 2.95 ± 0.19

129

According to the generally accepted definitions for wetting properties [242], both TiO2-

ZrO2-ZrTiO4 and TiO2 nanotubular layer were hydrophilic (10° < θW < 90°).

Hydrophilic property of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes increased with an

increase in the nanotube sizes (Fig. 5.8 (b) and (d)). Hydrophilic property of TiO2-

ZrO2-ZrTiO4 and TiO2 nanotubes increased with an increase of roughness parameters

which were the result of increasing of water content and applied potential. The

difference between water contact angle on the nanotubular surfaces of TiO2-ZrO2-

ZrTiO4 and TiO2 was not vary although nanotubular layer of TiO2 exhibited lower

roughness parameters than TiO2-ZrO2-ZrTiO4 due to their larger nanotube sizes.

The surface energies of the as-formed TiO2-ZrO2-ZrTiO4 and TiO2 nanotubular surfaces

calculated using the Owens-Wendt (OW) method [211] are listed in Table 5.3. All of

the TiO2-ZrO2-ZrTiO4 nanotubular surfaces exhibited a higher surface energy than those

of the as-formed TiO2 nanotubular surfaces. The different heights of TiO2-ZrO2-ZrTiO4

nanotubular surface depending on the microstructure (i.e., different phases of α and β)

of Ti50Zr alloy resulted in the roughness parameters of 500 > Sa and Sq > 1500 nm

which was due to the different chemical and electrochemical reaction rates of titanium

and zirconium in the alloy although their rates were lower than what was observed for

nanotubular layer fabricated in aqueous electrolyte [240].

130

Table 5.3 Calculated surface energy of as-formed and annealed dimpled surface TiO2-

ZrO2-ZrTiO4 nanotubular surfaces fabricated on Ti50Zr, Note: Each data point is an

average of five measurements


As-formed

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

5 wt%H2O-20V Ti50Zr 14.3 ± 0.6 13.27 58.59 71.86

CP-Ti 16.5 ± 0.4 12.93 58.27 71.20

10 wt%H2O-20V Ti50Zr 10.0 ± 0.4 12.90 60.34 73.24

CP-Ti 11.0 ± 0.7 12.40 60.81 73.21

5 wt% H2O-30V Ti50Zr 11.0 ± 0.2 10.30 64.04 74.34

CP-Ti 12.5 ± 0.3 15.04 56.81 71.86

10 wt%H2O-30V Ti50Zr 9.0 ± 0.1 13.37 59.91 73.28

CP-Ti 10.1 ± 0.3 13.15 59.97 73.12

131

5.3.3 Mechanical properties

To evaluate the nanomechanical characteristics of the various as-formed and annealed

TiO2-ZrO2-ZrTiO4 nanotubes a nanoindentation was used to measure the nano hardness

and Young’s modulus. Fig. 5.9 shows the curves of the load-unload forces versus the

nanoindentation depths of the as-formed and annealed TiO2-ZrO2-ZrTiO4 nanotubes

fabricated at different water content in an organic electrolyte (ethylene glycol) and

applied potentials.


TiO2-ZrO2-ZrTiO4 and b) annealed TiO2-ZrO2-ZrTiO4 fabricated at 5 and 10 wt % H2O

and the applied potential 20 and 30 V

The nano Young’s modulus and hardness can be determined instantaneously as a

function of depth. The hardness, reduced elastic modulus and elastic modulus were

calculated using the method described in Chapter 3 and presented in Fig. 5.10.

0

0.002

0.004

0.006

0.008

0.01

0.012

0 5E-10 1E-09 1.5E-09 2E-09

Load

(N)

Displacement (m)

As Formed nanotubes

5%-20V

5%-30V

10%-30V

(a

0

0.002

0.004

0.006

0.008

0.01

0.012

-5E-10 0 5E-10 1E-09 1.5E-09

Load

(N)

Displacement (m)

Annealed Nanotubes

5%-20V

5%-30V

10%-30V

(b)

132

Fig. 5.10 The nano mechanical properties of of TiO2-ZrO2-ZrTiO4 nanotubes fabricated

at different water content and applied potential: a) hardness, b) reduced elastic Modulus

and c) elastic Modulus

133

The hardness of nanotubular layer of TiO2-ZrO2-ZrTiO4 was maximum 330.2 and

4019.0 MPa before and after annealing which was lower than its respective bare metal

which revealed 4.38 GPa. The elastic modulus of the nanotubular layer of TiO2-ZrO2-

ZrTiO4 was maximum 185.9 GPa before annealing which was higher than its respective

bare metal which revealed 102.86 GPa; and 84.4 GPa after annealing which was lower

than its respective bare metal. It can be concluded that the porosity of the nanotubes

influenced the hardness and the ceramic properties of nanotubular layer influenced the

elasticity of the bare metal. By increasing the applied potential and consequently the

nanotube size the hardness and elasticity of nanotubes decreased. Increasing the water

content of the organic electrolyte (ethylene glycol) resulted in increasing of hardness

and elasticity of the nanotubular layer.

5.4 Conclusions

The effects of water content of the non-aqueous electrolyte and the applied potential on

the nanospacing, morphology and physical, chemical and mechanical properties of as-

formed and annealed TiO2-ZrO2-ZrTiO4 nanotubular surfaces were investigated. The

key conclusions are as follows:

1. The mean inner diameter (Di) and outer diameter (Do) of TiO2-ZrO2-ZrTiO4

nanotubes fabricated in the electrolytes of ethylene glycol containing 5 and 10

wt % H2O increased with an increase in the applied potential of 20 to 30 V from

29 ± 5 and 51 ± 5 nm, and 35 ± 5 and 61 ± 7 nm and to 34 ± 8 and 59 ± 11 nm,

and 42 ± 10 and 78 ± 10 nm; but the Wt did not show obvious change. Increasing

water content resulted in an increase in Di, Do and Wt.

2. The Di and Do of TiO2 nanotubes fabricated in the electrolytes of ethylene glycol

containing 5 and 10 wt % H2O increased with an increase in the applied

potential of 20 to 30 V but the Wt did not show obvious change. Increasing water

content resulted in an increase in Di, Do and Wt. TiO2 nanotubes revealed higher

Di, Do and Wt than TiO2-ZrO2-ZrTiO4 nanotubes.

3. After annealing at 500 °C for 2 h the mean Di of both TiO2-ZrO2-ZrTiO4 and

TiO2 decreased, the mean Wt increased but the mean Do did not changed. It was

found that the top layer of nanotubes was brittle and cracked, and it was

removed for further investigation.

134

4. There was a little difference between normal distribution and Weibull

distribution of nanotube sizes. The normal distribution proves an appropriate

method in evaluating the distribution of nanotube sizes.

5. Increasing water content and applied potential resulted in an increase in Mean

Roughness (Sa), Root Mean Square (RMS) Roughness (Sq) and hydrophilic

properties and a decrease in the surface energy of TiO2-ZrO2-ZrTiO4 nanotubes.

TiO2 nanotubes exhibited lower Sa, Sq and almost the same hydrophilic

properties and surface energy in comparison to TiO2-ZrO2-ZrTiO4 nanotubes.

6. The surface area index SI of TiO2-ZrO2-ZrTiO4 nanotubes increased with

increasing water content and applied potential, which is linked to changes in the

nanotube size distribution and surface roughness. TiO2 nanotubes revealed lower

surface area index SI than that of TiO2-ZrO2-ZrTiO4 nanotubes.

7. Annealing changed the phase structure, Di, Do and Wt of nanotubes. The

amorphous nanotubes changed to a mixture of tetragonal anatase and rutile,

orthorhombic TiO2 and ZrO2, and orthorhombic ZrTiO4. Hardness and elasticity

of nanotubular layer increased after annealing.

8. The hardness of the nanotubular layer is lower than that of the bare metal due to

the porosity of the fabricated nanotubues via anodisation; and the elasticity of

the nanotubes is lower than that of the bare metal due to the ceramic properties

of the hard and brittle oxide layer. The hardness and elasticity of TiO2-ZrO2-

ZrTiO4 nanotubes were also influenced by the porosity of the nanotubular layer.

9. The mixed nanotubes of TiO2-ZrO2-ZrTiO4 on Ti50Zr alloy fabricated in the

non-aqueous electrolyte with a variety of nanotube size distribution and other

related properties such as hydrophilicity and roughness are more promising for

biomedical applications than the 100 % TiO2 nanotubes on CP-Ti.

135

Chapter 6 Bioactivity of nanotubes fabricated in aqueous and non-

aqueous electrolytes

Abstract

The chemical behaviour and topography of implant surfaces can be enhanced by

designing the biomaterial surface properties with regard to their influences on the

bioactivity and cell behaviour. Fabrication of metal oxide nanotubes on metallic

biomaterials, especially titanium alloys such as Ti50Zr via anodisation, could enhance

the growth of hydroxyapatite (HA), which is a necessary step to provide a bioactive

surface for the attachment of bone cells to attach. In this chapter, eight groups of TiO2-

ZrO2-ZrTiO4 and TiO2 nanotubes with a diversity of nanoscale dimensional

characteristics (i.e., inner diameter Di, outer diameter Do and wall thicknesses Wt) were

fabricated via anodisation in aqueous and non-aqueous electrolytes. The as-formed and

annealed nanotubes were characterised using scanning electron microscopy (SEM), X-

ray diffraction (XRD) and 3D-Profilometry. The different nanoscale characteristics of

the nanotubes and the different roughness parameters of the nanotubular surfaces were

obtained by adjusting the applied potential during anodisation, which influenced the

oxidation rates of titanium and zirconium. The bioactivity of a nanotubular layers of

TiO2-ZrO2-ZrTiO4 and TiO2 were assessed by immersion in a modified simulated body

fluid (m-SBF). The results revealed that the nanotubes were more effective in inducing

hydroxyapatite (HA) growth in contrast to the just anodised surface and the bare metal.

The Di, roughness, hydrophilic property and crystalline phase of the nanotubes

demonstrated an effective influence on inducing HA.

6.1 Introduction

The mineralised collagen fibril is the basic building block which is common in a family

of materials with different structural motif known as bone [29]. The plate shaped

carbonated apatite or dahllite (Ca5∙(PO4∙CO3)3(OH)) stands in a framework of these

collagen fibrils [29, 250]. Although dahllite displays a hexagonal crystal symmetry, the

crystals of bone are plate-shaped; probably because plate shaped octacalcium phosphate

is a transition phase. The average length and width of this crystal are 50 × 25 nm with a

uniform thickness of 1.5 - 4 nm and it has a highly ordered surface [29]. The organised

structure of bone which is hierarchically complex has two major functions: a

136

mechanical function and a metabolic function. The metabolic function of bone is a

reservoir for calcium and phosphate [250]. For example remodelling process in hard

tissue, which is guided by bone cells and proteins can also occur on surfaces of

bioactive implants by the biomineralisation of apatite [251]. Bioactive implant materials

can be mineralised into apatite by means of body fluid (in vivo) or simulated body fluid

(in vitro) when there is an interface to integrate spontaneously with living tissue [251].

Mineralisation of crystalline apatite occurs due to the electrostatic interaction of the

functional groups of the bioactive surface with the calcium and phosphate ions in the

fluid through a transition phase of amorphous calcium phosphate [251].

Recent research has focused on binary and tertiary alloys of titanium with

biocompatible alloying metals such as tantalum, niobium and zirconium for application

as implants [27, 46, 93-95, 230, 252, 253]. Ti50Zr, with a fine martensitic structure is

one of these new biocompatible alloys And, in comparison to CP-Ti, it exhibits 2.5

times higher hardness and ~ 2.3 - 3 times higher tensile strength [230]. Anodisation is a

surface modification method that can improve the biocompatibility and bioactivity of

metal implants. A layer of metal oxide nanotubes forms when the electrochemical

oxidation is undertaken in an electrolyte (aqueous or non-aqueous) consisting of

fluorine anions [11]. Titanium dioxide, fabricated both in aqueous and non-aqueous

electrolyte, exhibited bioactivity [11, 16, 234]. Hydroxyapatite can be deposited or

mineralised on the surface of metal implant using physical and chemical techniques,

such as plasma spray [182], sputtering [183, 184], electron-beam physical vapour

deposition (EB-PVD) [185] and electrodeposition [186]. However, a method that is

technically simpler; i.e., immersion in a simulated body fluid (SBF) such as Hank’s

solution [22, 189], is also feasible. Some attempts have been undertaken to achieve the

Ca/P ratio of stoichiometric synthetic hydroxyapatite (Ca10(PO4)6(OH)2) formed on a

nanotubular layer of TiO2 using the immersion method [189, 254] is 1.67. Although the

fabrication of a nanotubular layer on Ti50Zr alloy has been reported, its bioactivity in

inducing hydroxyapatite needs to be investigated.

In this chapter, eight different nanotubular layers of TiO2-ZrO2-ZrTiO4 and TiO2 were

fabricated with different inner diameter (Di), outer diameter (Do) and wall thickness

(Wt) dimensions and physical and chemical properties. Hydroxyapatite was mineralised

on the surface of (i) TiO2-ZrO2-ZrTiO4 nanotubes, (ii) the anodised surface (flat

137

surface), and (iii) bare Ti50Zr alloy and compared to (i) TiO2 nanotubes, (ii) the

anodised surface (flat surface), and (iii) bare CP-Ti to evaluate their degree of

bioactivity. The effect of nanotube size, the roughness and surface chemistry of the

nanotubular layer on inducing HA mineralisation was elucidated.


Ti50Zr alloy was prepared by casting and then discs 8 mm diameter and 2 mm thickness

were fabricated by electric discharge machining. After polishing to a 1 μm diamond

finish, the samples were ultrasonically cleaned for 15 minutes each in methanol,

isopropanol, acetone and ethanol prior to rinsing with pure water and drying in a

nitrogen stream. CP-Ti (Baoji Boxin Metal Materials Co., Ltd.) foils (10×10 mm) were

degreased and washed in the same manner. The cleaned discs and foils were divided

into four groups. TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes were fabricated on the surfaces

of the first group of samples using a DC power supply with two-electrode configuration

in 1 M (NH4)2SO4 and 0.5 wt % NH4F (Sigma-Aldrich reagent grades) at different

applied potentials. The electrolyte for second group of samples was ethylene glycol

contained 0.5 wt % NH4F and 5 and 10 wt % pure water (H2O). Compact TiO2-ZrO2-

ZrTiO4 and TiO2 films without nanotubular structure were produced on the surfaces of

the third group of samples by anodisation in 1 M (NH4)2SO4 at 20 V using the same

power supply. After anodisation the samples were washed with deionised water for 5

minutes and dried with nitrogen stream. The samples which were treated in non-

aqueous electrolyte were rinsed with acetone then were post-treated by immersion in 0.5

M HCl at room temperature for 1 min and finally rinsed with deionised water for 1 min

to dissolve the remained upper nanoporous layer. The forth group of samples was the

bare metals without any surface modification. Half of the electrochemically produced

surface layers were annealed at 500 ˚C for 3 h in air in a muffle furnace (Nabertherm

LT15/13/P330) at a heating rate of 2 ˚C min−1.

The sample surfaces, after anodisation and soaking in m-SBF [213], were analysed with

a field-emission scanning electron microscope (FESEM, ZEISS SUPRA 40 VP)

equipped with an energy-dispersive X-ray spectrometry (EDX). The crystalline phases

were determined by X-ray diffraction using Cu Kα incident radiation at 40 kV and

40 mA with a scanning step size of 0.02 ° from 2θ equal of 10 to 90 °.

138

The peaks, valleys, and the spacings of the surface textures were characterised by using

the roughness parameters including Mean Roughness (Sa) and Root Mean Square

(RMS) Roughness (Sq), the Skewness (Sskw) and the kurtosis (Sku) of a 3D surface

texture. These surface roughness parameters were measured using a 3D-Profilometer

(Bruker, Contour GT-K1; Bruker Pte Ltd, Singapore) coupled with SurfVision software

(Veeco Instruments Inc.; Plainview, NY, USA) [210]. The Sskw represents the degree of

symmetry of the surface heights about the mean plane and Sku indicates the nature of the

height distribution. It is notable that a Sskw > 0 implies high peaks; whereas Sskw < 0

indicates valley structures such as deep scratches. When Sku is 3.0, the surface heights

are normally distributed, Sku < 3.0, indicating a few high peaks and low valleys

(platykurtoic) and Sku > 3.0, indicating many high peaks and low valleys (leptokurtoic)

[210].

The water contact angle measurements were undertaken using a goniometer (NRL C.A.

Goniometer, Ramé-hart, Inc.; Succasunna, NJ-USA). The surface energy was calculated

using the Owens-Wendt (OW) method [211], as follows:


d+√γLp γS

p) (3 − 1)

where L indicates the liquid surface tension (water=72.8 mJm-2, glycerol=63.4 mJm-2);

Ld and S

d are the liquid and solid dispersive component (water=21.8 mJm-2,

glycerol=37.0 mJm-2). The variables Lp and S

p indicate the liquid and solid polar

components (water=51.0 mJm-2, glycerol=26.4 mJm-2) and S is the sum of Sd and S

p

[212].

The samples were soaked in modified simulated body fluid (m-SBF) to evaluate the in

vitro bioactivity and assess their ability for inducing apatite formation. All three groups

of Ti50Zr and CP-Ti samples were soaked in m-SBF for 3 weeks. The m-SBF solutions

were prepared according to reference [213] with the composition and the preparation

method presented in Ref. [214]. The m-SBF was buffered with 2-(4-(2-hydroxyethyl)-1-

piperazinyl) ethane sulfonic acid (HEPES) (17.892 g HEPES dissolved in 100 ml 0.2 M

NaOH) and 1 M NaOH at pH 7.4 at 37˚C. m-SBF with ionic concentrations equal to

human blood plasma was prepared by dissolving reagent grade chemicals with the

composition 5.403 g NaCl, 0.504 g NaHCO3, 0.426 g Na2CO3, 0.225 g KCl, 0.230 g

139

K2HPO4 - 3H2O, 0.311 g MgCl2 - 6H2O, 0.293 g CaCl2, and 0.072 g Na2SO4in 1 litre

deionised water. After soaking, the samples were rinsed with pure water and dried at

room temperature. The change of pH value during incubation was measured using

pH/mV & Ion/pH meter Series (Oakton, Eutech instruments Pte Ltd, USA). The atomic

percentage of calcium to phosphate ratio of HA is calculated using EDS results.



nanotubes fabricated in aqueous electrolyte

The nanotubular layer of TiO2-ZrO2-ZrTiO4 were fabricated in 0.5 wt % NH4F, 1 M

(NH4)2SO4 in pure water for 1 h at an applied potential of 20, 25, 30 and 35 V. Figure

4.9 in Chapter 4 shows the top view of as-formed and annealed nanotubes. The mean Di

of as-formed nanotubular layer of TiO2-ZrO2-ZrTiO4 increased from 37 ± 13 nm to 64 ±

18, 76 ± 20 and 81 ± 20 nm with an increase in applied potential from 20 to 25, 30 and

35 V. The mean Di of nanotubular layer of TiO2 also increased from 43 ± 10 nm to 52 ±

6, 65 ± 9 and 68 ± 6 nm with an increase of applied potential from 20 to 25, 30 and 35

V. Fig. 6.1 shows the mean size of nanotubular layer of TiO2-ZrO2-ZrTiO4 and TiO2.

140

Fig. 6.1 Mean nanotube size as a function of applied potential: (a) Ti50Zr - as-formed,

(b) Ti50Zr - annealed (c) CP-Ti - as-formed, (d) CP-Ti - annealed

The mean roughness (Sa) and the water contact angle (W.C.A.) of the nanotubular

surfaces as a function of the applied potential of the as-formed and annealed TiO2-ZrO2-

ZrTiO4 and TiO2 nanotubes anodised on Ti50Zr and CP-Ti are presented in Fig. 6.2. It

can be seen that the mean Sa of as-formed TiO2-ZrO2-ZrTiO4 nanotubes decreased with

an increase in Di of nanotubes and the Di increased with increasing applied potential.

Although the Sa of the annealed nanotubular surfaces slightly increased with the

increasing applied potential, they were smoother than those of as-formed TiO2-ZrO2-

ZrTiO4 nanotubular surfaces. The same tendency was observed for TiO2 nanotubes

although they revealed lower roughness amplitude parameters. After annealing the Sa of

the TiO2 nanotubes slightly increased or remained constant. The Sskw and Sku results for

the TiO2-ZrO2-ZrTiO4 nanotubes under all conditions were leptokutoic, which implies

0

20

40

60

80

100

120

140

160

180

20 25 30 35

Mea

n si

ze (n

m)


Ti50Zr-As Formed

Wt

Di

Do

a)

0

20

40

60

80

100

120

140

160

180

20 25 30 35

Mea

n si

ze (n

m)


Ti50Zr-Annealed

Wt

Di

Do

b)

0

20

40

60

80

100

120

140

160

180

20 25 30 35

Mea

n si

ze (n

m)


CP-Ti-As Formed

Wt

Di

Do

c)

0

20

40

60

80

100

120

140

160

180

20 25 30 35

Mea

n si

ze (n

m)


CP-Ti-Annealed

Wt

Di

Do

d)

141

that the surface exhibited mostly valleys as would be reflected by the needle-like

microstructure of the alloy. The same tendency was revealed after annealing except with

slightly lower values. The result of Sskw and Sku of the surface of TiO2 nanotubes

indicated that they displayed a normal distribution of peaks and valleys that were

similar to a flat surface and the bare metals of Ti50Zr and CP-Ti. After annealing the

Sskw and Sku of the nanotubular surfaces showed the same tendency with slightly lower

values.

Fig. 6.2 The mean roughness (Sa) of the nanotubular surfaces as a function of the

applied potential: (a) Ti50Zr alloy, (c) CP-Ti; and the mean water contact angle

(W.C.A.) of the nanotubular surfaces as a function of the applied potential: (b) Ti50Zr

alloy and (d) CP-Ti, Note: Each data point is an average of five measurements

The surface index of TiO2-ZrO2-ZrTiO4 nanotubes, which was calculated by dividing

the surface area of the nanotubular layer to the lateral surface area, decreased with an

increase in applied potential and Di, and the surface area index of TiO2-ZrO2-ZrTiO4

nanotubes was (i) a maximum 10.4 and a minimum of 3.5 times higher than TiO2

0

0.5

1

1.5

2

2.5

3

3.5

20 25 30 35

Mea

n Sa

(µm

)


Roughness of Ti50ZrAs Formed

Annealed

05

1015202530354045

20 25 30 35

Mea

n W

.C.A

(θ°)


Wettability of Ti50Zr

As Formed

Annealed

b)

0

0.5

1

1.5

2

2.5

3

3.5

20 25 30 35

Mea

n Sa

(µm

)


Roughness of CP-Ti As Formed

Annealed

c)

0

5

10

15

20

25

30

35

40

45

20 25 30 35

Mea

n W

.C.A

. (θ°

)


Wettability of CP-Ti As Formed

Annealed

d)

a)

142

nanotubes and (ii) a maximum of 17.6 and a minimum of 5.7 times higher than those of

the flat surface and bare Ti50Zr (Table 6.1).

143

Table 6.1 Surface area index SI and roughness amplitude parameters of Sq, Sskw and Sku of the nanotubular surfaces of Ti50Zr and CP-Ti anodised

under various conditions, Note: Each data point is an average of five measurements


As-formed Annealed

Surface area index

SI Sq Sskw Sku

Surface area index

SI Sq Sskw Sku

20V Ti50Zr 17.80 ± 2.25 3.94 ± 0.19 -1.6 ± 0.08 5.26 ± 0.34 3.88 ± 1.02 0.87 ± 0.12 -1.91 ± 0.51 20.22 ± 8.27

CP-Ti 1.71 ± 0.07 0.72 ± 0.03 -0.67 ± 0.12 3.30 ± 0.12 1.49 ± 0.01 0.74 ± 0.06 -0.80 ± 0.03 2.97 ± 0.07

25V Ti50Zr 8.56 ± 1.75 2.29 ± 0.46 -2.28 ± 1.42 12.88 ± 9.26 2.91 ± 0.59 0.85 ± 0.17 -1.74 ± 1.51 17.53 ± 13.66

CP-Ti 1.35 ± 0.07 0.33 ± 0.04 -0.31 ± 0.17 3.43 ± 0.95 1.29 ± 0.01 0.34 ± 0.07 0.01 ± 0.09 2.95 ± 0.38

30V Ti50Zr 8.62 ± 1.30 1.75 ± 0.41 -2.62 ± 1.20 33.24 ± 34.43 4.18 ± 0.60 0.95 ± 0.10 -0.81 ± 0.63 9.51 ± 5.6

CP-Ti 1.35 ± 0.02 0.31 ± 0.01 0.51 ± 0.13 3.88 ± 0.13 1.30 ± 0.01 0.30 ± 0.00 -0.11 ± 0.14 3.09 ± 0.10

35V Ti50Zr 5.82 ± 1.27 1.45 ± 0.29 -3.29 ± 1.27 30.1 ± 17.35 3.97 ± 0.47 0.94 ± 0.06 -0.87 ±0.36 16.34 ± 16.25

CP-Ti 1.67 ± 0.02 0.47 ± 0.01 -0.45 ± 0.06 2.48 ± 0.08 1.66 ± 0.04 0.58 ± 0.03 0.56 ± 0.07 2.71 ± 0.07

Anodised in (NH4)2SO4 Ti50Zr 1.01 ± 0.00 0.06 ± 0.01 0.12 ± 0.35 9.34 ± 4.72 1.04 ± 0.02 0.18 ± 0.08 0.41 ± 1.2 5.53 ± 5.18

CP-Ti 1.39 ± 0.02 0.62 ± 0.17 0.19 ± 0.05 2.31 ± 0.23 1.45 ± 0.03 0.29 ± 0.01 -0.26 ± 0.15 3.86 ± 0.21

Bare Metal Ti50Zr 1.00 ± 0.00 0.05 ± 0.01 0.20 ± 0.58 9.64 ± 3.48 - - - -

CP-Ti 0.18 ± 0.00 -0.23 ± 0.00 4.9 ± 0.00 - - - - -

144

The water contact angle of TiO2-ZrO2-ZrTiO4 nanotubes decreased with an increase of

Di for both the as-formed and the annealed nanotubes. The values of water contact angle

on the surface of TiO2-ZrO2-ZrTiO4 after annealing were lower than those of the as-

formed nanotubes. The nanotubular layer of TiO2-ZrO2-ZrTiO4 exhibited super-

hydrophilic properties under almost all conditions except for the as-formed TiO2-ZrO2-

ZrTiO4 nanotubes fabricated at an applied potential of 20 V with Di = 37 nm which

exhibited hydrophilic properties. The water contact angle of TiO2 nanotubes decreased

with an increase of Di for both the as-formed and the annealed nanotubes and they

revealed lower values after annealing. The TiO2 nanotubular layer exhibited hydrophilic

properties in all conditions. The flat surface and bare metal surface of Ti50Zr and CP-Ti

revealed hydrophilic properties with water contact angles that were at least 2 times

higher than their nanotubular counterpart layers. It can be seen that for both the as-

formed TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes the water contact angle increased with

an increase of Sa while it decreased on the annealed nanotubular layer. It can be

concluded that increasing Di and the crystalline phase of the nanotubular layer

influenced the hydrophilic properties more than the surface roughness. The surface

energy of TiO2-ZrO2-ZrTiO4, TiO2 nanotubes, flat surface and bare Ti50Zr alloy and

CP-Ti are listed in Table 6.2.

145

Table 6.2 Calculated surface energies of TiO2-ZrO2-ZrTiO4 nanotubular surfaces fabricated on Ti50Zr and TiO2 nanotubular surfaces on CP-Ti, Note:



As-formed Annealed

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

20V Ti50Zr 20.50 ± 0.4 15.98 52.67 68.65 2.7 ± 0.2 14.93 58.71 73.64

CP-Ti 28.7 ± 10.1 22.45 41.87 64.32 22.5 ± 4.8 12.95 55.56 68.51

25V Ti50Zr 9.5 ± 1.3 15.01 57.63 72.64 6.1 ± 0.3 14.69 58.66 73.35

CP-Ti 30.0 ± 1.0 18.14 45.24 63.38 13.5 ± 0.9 9.25 65.04 74.29

30V Ti50Zr 9.2 ± 0.3 15.45 57.13 72.58 5.9 ± 1.7 15.22 58.00 73.22

CP-Ti 25.3 ± 5.1 13.98 52.69 66.68 13.3 ± 2.0 13.33 58.83 72.16

35V Ti50Zr 6.5 ± 2.3 14.05 59.45 73.50 4.9 ± 0.7 14.20 59.47 73.67

CP-Ti 18.0 ± 3.0 13.62 56.74 70.36 13.3 ± 1.8 10.16 63.57 73.73

Anodised in (NH4)2SO4 Ti50Zr 47.7 ± 1.5 0.19 67.89 68.09 20.7 ± 1.5 3.59 74.42 78.01

CP-Ti 64.8 ± 3.8 7.29 30.25 37.54 47.2 ± 1.2 26.36 27.1 53.46

Bare Metal Ti50Zr 42.2 ± 1.8 14.56 40.68 55.24 - - - -

CP-Ti 54.5 ± 4.7 15.37 30.17 45.55 - - - -

146

6.3.2 Bioactivity of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in aqueous

electrolyte

A dense and uniform layer of the bonelike apatite was formed on the surface, as shown

in Figs. 6.3 and 6.4, when the TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes were soaked in m-

SBF. It has been reported that for various TiO2 layers, an anatase or a rutile structure of

TiO2 nanotubes were more favourable for apatite formation than an amorphous structure

[48, 254]. Therefore, some samples were annealed at 500 °C for 3 h to obtain crystalline

phases. The surface of the as-formed and annealed TiO2-ZrO2-ZrTiO4, TiO2 nanotubes

and flat surface were covered by nanobristle spheres (Fig. 6.3(f)) and a layer of

nanoflakes (Figs. 6.3(h) and 6.4(h)) after 3 weeks immersion in m-SB, respectively.

However, no HA precipitation was observed on bare metal samples after 3 weeks of

immersion and on flat surface within 3 weeks immersion in m-SBF.

147

Fig. 6.3 Mineralisation of HA on the nanotubular surface of TiO2-ZrO2-ZrTiO4

fabricated at a), b) 20 V (Di = 37 nm)-as-formed, c), d) 20V (Di = 40 nm)- annealed, e),

f) 25 V (Di = 64 nm)-as-formed, g), h) 25V (Di = 59 nm)-annealed, i), j) 30 V (Di = 76

nm)-as-formed, k), l) 30V (Di = 64 nm)-annealed, m), n) 35 V (Di = 81 nm)-as-formed

and o), p) 35V (Di = 82 nm)-annealed

148

Fig. 6.4 Mineralisation of HA on the nanotubular surface of TiO2 fabricated at a), b) 20

V (Di = 43 nm)-as-formed, c), d) 20V (Di = 51 nm)-annealed, e), f) 25 V (Di = 52 nm)-

as-formed, g), h) 25V (Di = 45 nm)-annealed, i), j) 30 V (Di = 65 nm)-as-formed, k), l)

30V (Di = 53 nm)-annealed, m), n) 35 V (Di = 68 nm)-as-formed and o), p) 35V (Di =

63 nm)-annealed

149

The maximum measured thickness of the HA layers mineralised on the surface of TiO2-

ZrO2-ZrTiO4 and TiO2 were 2.7 and 5.3 µm, respectively as shown in Fig. 6.5.


TiO2-ZrO2-ZrTiO4 nanotubes and b) TiO2 nanotubes fabricated in aqueous electrolyte

The EDS spectra of the above mentioned samples detected Ca, P and O peaks, which

are the essential ingredients of hydroxyapatite, after immersion in m-SBF for 3 weeks.

The XRD patterns of the TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes before and after

immersion in m-SBF are shown in Fig. 6.6. The annealed specimen revealed the anatase

peak of titania and the orthorhombic phase for a mixture of titania, zirconia and

zirconium titanate prior immersion in m-SBF. After immersion in m-SBF for 3 weeks,

some hydroxyapatite (HA) peaks appear in the XRD patterns.

150

Fig. 6.6 XRD patterns of: a) TiO2-ZrO2-ZrTiO4 nanotubes anodised on Ti50Zr (α and β phases) and annealed at 500°C for 3 h before and after

immersion in m-SBF for 3 weeks, showing tetragonal anatase, srilankite (a mixture of orthorhombic TiO2 and ZrO2) and orthorhombic ZrTiO4

with mineralised hydroxyapatite, b) TiO2 nanotubes anodised on CP-Ti (α phase) and annealed at 500 °C for 3 h before and after immersion in

m-SBF for 3 weeks, showing tetragonal anatase with mineralised hydroxyapatite; and c) hydroxyapatite

151

Figure 6.7 shows the atomic percentage of the calcium to phosphate ratio (Ca/P) of HA

that was mineralised on the surfaces of the TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes. The

Ca/P ratio of HA precipitated on the nanotubular layer of TiO2-ZrO2-ZrTiO4 increased

with an increase in Di for both the as-formed and annealed nanotubes. It can be

observed that the Ca/P ratio of HA increased with an increase of Sa for the as-formed

TiO2-ZrO2-ZrTiO4 nanotubes but it decreased with an increase of Sa for the annealed

samples. On the other hand, the Ca/P ratio increased with a decrease of water contact

angle for both the as-formed and annealed nanotubular layers. The Ca/P ratio of HA

increased after annealing for each nanotubular layer fabricated under the same

conditions regardless of the difference in the inner diameters (Di) of the nanotubes.

Fig. 6.7 The percentage of atomic ratio of Ca/P of mineralised HA on four different

nanotubular surfaces of a) Ti50Zr and b) CP-Ti

The Ca/P ratio of HA precipitated on the nanotubular layer of TiO2 decreased with an

increase of Di for both the as-formed and annealed nanotubes. Increasing Sa of both as-

formed and annealed TiO2 nanotubes resulted in an increase of the Ca/P ratio of HA;

although it decreased with a decrease of water contact angle. The Ca/P ratio of HA

remained constant after annealing for each nanotubular layer fabricated under the same

conditions regardless of different inner diameters. A HA layer formed on the flat surface

of the oxide layer on Ti50Zr and CP-Ti. The measured Ca/P ratios of HA were (i) 1.27

before and after annealing for Ti50Zr, and (ii) 1.29 before annealing and 1.22 after

1.21.251.3

1.351.4

1.451.5

1.551.6

1.651.7

20 25 30 35

Ca/

P (a

t %)


Ti50Zr As Formed

Annealed

1.21.25

1.31.35

1.41.45

1.51.55

1.61.65

1.7

20 25 30 35

Ca/

P (a

t %)


CP-Ti As Formed

Annealedb)a)

152

annealing for CP-Ti; the values of which were lower than their corresponding

nanotubular layers.

The reactions on the nanotubular layer subjected to the alkaline aqueous solution of m-

SBF are as follows:

TiO2.nH2O + OH- → HTiO3-.nH2O (6-1)

ZrO2.nH2O + OH- → HZrO3-.nH2O (6-2)

ZrTiO4. nH2O + OH- → HZrTiO5-. nH2O (6-3)

The isoelectric point of TiO2 is reported at a pH of 3.9 and the corresponding value for

ZrO2 is 5.5 [255, 256]. Both of these values are lower than the pH of m-SBF (7.4) and

indicate the formation of a negative layer of Ti-OH and Zr-OH. The negatively charged

surface absorbs the cations in the solution such as Na+, K+, Mg2+ and Ca2+ by Coulomb

force. According to our observation a patch like of salt crystals of the abovementioned

cations and existing anions such as Cl-, HCO3-, CO3

2-, HPO42- SO4

2- formed as a

transition state for the formation of the composition of calcium phosphate (Fig. 6.8).

These crystals were in equilibrium with their ions in the solution as they are not

sparingly soluble salts. A composition of calcium phosphate forms on the positively

charged surface (Ca rich layer) due to electrostatic interaction with the phosphate ion in

the m-SBF. Then the layer that is negatively charged absorbs more Ca2+ to form

amorphous apatite. With increasing the immersion time, the amorphous apatite

eventually transforms to a crystalline structure that has lower solubility in neutral and

basic solutions in comparison to other structural form of calcium phosphate. Sodium,

magnesium, chloride and carbonate ions may become incorporated into an apatite

similar to the compositional and structural features of bone mineral since the m-SBF

mimics the blood plasma not only in composition but also in concentration [251].

Octacalcium phosphate phase (Ca8H2(PO4)6·5H2O; OCP) is the first formed Ca-P

crystal nuclei in the simulated body fluid [257]. Thermodynamically OCP tends to

convert to a calcium deficient HA although it remains in the apatite lattice with a plate

like morphology. Fig. 6.9 schematically illustrates the process of bone-like apatite

formation on nanotubular surface in m-SBF.

153

Fig. 6.8 SEM images showing the transition of soluble salt crystal of present ions of m-

SBF into HA on the nanotubular surface of TiO2-ZrO2-ZrTiO4 with increasing

immersion time in m-SBF

Fig. 6.9 Schematic presentations of the process of bone-like apatite formation on

nanotubular surface in m-SBF

154

The local pH change up to 3 weeks of mineralisation of HA is shown in Fig. 6.10. The

pH value increased while negative hydrolysed nanotubular surface formed. During the

absorption of Ca2+ ions the changes of pH value were negligible. Then the pH value

increased as result of the exchange of OH- with PO43-. While the OH- consumed to form

hydroxyapatite [Ca10(PO4)6(OH)2] the pH value decreased and remaining constant after

the HA formation was completed.


TiO2-ZrO2 -ZrTiO4 fabricated at 20, 25, 30 and 35V

Considering the observation for changing the Ca/P ratio of HA mineralised on the

surface of TiO2-ZrO2-ZrTiO4 nanotubes it can be deduced that there was a direct

relation between Di, hydrophilic properties and crystalline phase of nanotubes rather

than the roughness parameters for inducing the formation and growth of HA on

nanotubular surface. In terms of TiO2 nanotubes, the hydrophilic properties and

roughness parameter revealed a more significant influence on the formation and growth

155

of HA than the Di and the crystalline phase of the nanotubular layer. The role of

morphological features that give rise to high roughness lies in the creation of the initial

reaction sites for apatite nucleation and formation.

The nanomechanical characteristics of the mineralised HA on the surface of TiO2-ZrO2-

ZrTiO4 nanotubes were measured using a nanoindentation. The curve of the load-

unload forces versus the nanoindentation depths of the HA is shown in Fig. 6.11. The

nano-Young’s modulus was 23.9 GPa and the nano hardness was 87.0 MPa for the

nanotubular layer of TiO2-ZrO2-ZrTiO4. The average elastic modulus for the osteons

and the interstitial lamellae in the longitude direction was reported as 24.7 GPa and 30.1

GPa. The hardness was reported to be varied among the microstructure components in

the range of 0.41 - 0.89 GPa [258].


nanotubular surface of TiO2-ZrO2-ZrTiO4 fabricated at the applied potential of 35 V


nanotubes fabricated in non-aqueous electrolyte

The nanotubular layer of TiO2-ZrO2-ZrTiO4 were fabricated in 0.5 wt % NH4F, 5 and

10 wt % pure water (H2O) in ethylene glycol at 20 and 30 V. Figure 5.1 in Chapter 5

0

0.002

0.004

0.006

0.008

0.01

0.012

0 5E-10 1E-09 1.5E-09 2E-09 2.5E-09

Load

(N)

Displacement (m)

156

shows the top view of the as-formed and Fig. 6.12 shows the dimpled surface of

annealed nanotubes. The mean Di of as-formed nanotubular layer of TiO2-ZrO2-ZrTiO4

increased from 29 ± 5 nm to 35 ± 5 with an increase of water content from 5 to 10 wt %

at 20 V and increased from 34 ± 8 to 42 ± 10 nm at 30 V. The mean Di of nanotubular

layer of TiO2 also increased from 32 ± 5 nm to 37 ± 5 with an increase of water content

from 5 to 10 wt % at 20 V and increased from 42 ± 5 to 52 ± 5 nm at 30 V. These

nanotubular layer exhibited a brittle layer after annealing therefore, the cracked upper

layer was removed by ultrasound and remaining dimpled surface used for bioactivity

investigation.

Fig. 6.12 Dimpled surface of TiO2-ZrO2-ZrTiO4 nanotubes after annealing and

removing the upper cracked layer a) 5 wt % H2O - 20 V, b) 10 wt % H2O - 20 V, c) 5 wt

% H2O - 30 V and d) 10 wt % H2O - 30 V

Figure 5.8 in Chapter 5 shows the roughness parameters of Sa and Sq measured for the

as-formed nanotubular surfaces and Fig. 6.13 shows the roughness parameters of Sa and

157

Sq measured for the annealed nanotubular layer with a dimpled surface. These

amplitude parameters (Sa and Sq) as fully explained in Chapters 5 and 7 increased with

an increase of water content for each applied potential and with an increase of the

applied potential for each composition of electrolyte. The hydrophilic property of

nanotubes increased with an increase of the nanotube sizes and roughness parameters

that were the result of increasing the water content and applied potential. The nanotubes

fabricated in the electrolyte with 5 wt % H2O at 20 V with mean Di = 29 nm exhibited

Sskw ≈ 0 and Sku ≈ 3 which indicats uniform symmetry and a normal distribution of

surface heights. The other 3 nanotubular surfaces exhibited a more distinct leptokurtoic

distribution with many high peaks and valleys. The majority of the 4 dimpled surfaces

exhibited a uniform symmetry with normal distribution in surface heights. Table 5.2 in

Chapter 5 presents these amplitude parameters plus the surface index (SI). The

hydrophilic property of nanotubes increased with an increase of the nanotube sizes and

roughness parameters that were the result of increasing the water content and applied

potential. The surface energies of the as-formed TiO2-ZrO2-ZrTiO4 and annealed

dimpled surface TiO2-ZrO2-ZrTiO4 nanotubular surfaces calculated using the Owens-

Wendt (OW) method [211] are listed in Table 5.3 (Chapter 5).

Fig. 6.13 a) The mean roughness (Sa), (Sq) and the mean water contact angle (W.C.A.)

of the nanotubular and dimpled surface of TiO2-ZrO2-ZrTiO4, respectively, Note: Each

data point is an average of five measurements

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8


Mea

n Sa

(µm

)

Mean Di (nm)

Roughness Sa-Annealed-Dimple

Sq-Annealed-Dimple

a)

0123456789

10


Mea

n W

.C.A

. (θ°

)

Mean Di (nm)

WettabilityW.C.A-Annealed-Dimpleb)

158

6.3.4 Bioactivity of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in non-

aqueous electrolyte

The dimpled surface after removing the upper cracked layer of annealed TiO2-ZrO2-

ZrTiO4 nanotubes with 4 different nanotube sizes was investigated for their ability to

induce hydroxyapatite after immersion in m-SBF for 3 weeks. SEM images of

mineralised hydroxyapatite on the as-formed TiO2-ZrO2-ZrTiO4 nanotubes and dimpled

surface are shown in Fig. 6.14 and SEM images of mineralised hydroxyapatite on the

as-formed and annealed TiO2 nanotubes are shown in Fig. 6.15.

159

Fig. 6.14 SEM images of mineralisation of HA on the as-formed TiO2-ZrO2-ZrTiO4

nanotubes and dimpled surface fabricated in non-aqueous electrolyte: a, b) and c, d)

high and low magnification-5 wt % H2O-20V, e, f) and g, h) high and low

magnification-10 wt % H2O-20V, i, j) and k, l) high and low magnification-5 wt %

H2O-30V and m, n) and o, p) high and low magnification-10 wt % H2O-30V

160

Fig. 6.15 SEM images of mineralisation of HA on the as-formed and annealed TiO2

nanotubes fabricated in non-aqueous electrolyte: a, b) and c, d) high and low

magnification-5 wt % H2O-20V, e, f) and g, h) high and low magnification-10 wt %

H2O-20V, i, j) and k, l) high and low magnification-5 wt % H2O-30V and m, n) and o,

p) high and low magnification-10 wt % H2O-30V

161

The maximum thickness of the HA layers which mineralised on the surface of

nanotubular TiO2-ZrO2-ZrTiO4 was 9.6 µm. The maximum thickness of the HA layers

which mineralised on the surface of TiO2 nanotubes was 7.1 µm (Fig. 6.16).


TiO2-ZrO2-ZrTiO4 nanotubes and b) TiO2 nanotubes fabricated in non-aqueous

electrolyte

The peaks of Ca, P and O peaks, the essential ingredient of hydroxyapatite, were

detected by the EDS spectra of all above mentioned samples. The XRD patterns of the

TiO2-ZrO2-ZrTiO4 nanotubes fabricated in non-aqueous electrolyte before and after

immersion in m-SBF are shown in Fig. 6.17. The anatase peak of titania and

orthorhombic phase of the mixture of titania, zirconia and zirconium titanate of the

annealed specimen before immersion in m-SBF are visible. Some hydroxyapatite (HA)

peaks appeared after immersion in m-SBF for 3 weeks, in the XRD patterns.

162


via anodisation in non-aqueous electrolyte, annealed dimpled surface (at 500 °C for 2 h)

of tetragonal anatase, rutile, srilankite

The atomic ratio of calcium to phosphate calculated using EDS results is shown in Fig.

6.18. If a biomaterial implant after implantation in vivo is encapsulated by fibrous

tissues it would be isolated from the surrounding bone. Formation of a carbonated

apatite layer on the surface of a biomaterial implant similar to hydroxyapatite (HA) of

bone composition is an ideal characteristic for healing mechanism. The Ti-OH and Zr-

OH groups located on the surface of TiO2-ZrO2-ZrTiO4 nanotubes are favoured sites for

apatite nucleation [56, 251, 254, 257]. The Ca2+ ions firstly are absorbed onto the

hydrolysed nanotubular oxide surface by Coulomb attraction forces. The surface would

be positively charged therefore the existing phosphate groups inside the SBF are

adsorbed to the surface and calcium phosphate forms. The stoichiometric Ca/P (at %) of

163

hydroxyapatite [Ca10(PO4)6(OH)2] is 1.67. According to the obtained Ca/P ratio of

eight different nanotubular surfaces, as-formed nanotubular surface with Di = 29 and 35

nm had higher value of Ca/P ratio than the as-formed nanotubular surface with Di = 34

and 42 nm. This may indicate the effect of nanospacing and roughness parameter and

hydrophilic properties although their differences are not considerable. The dimpled

surface of annealed TiO2-ZrO2-ZrTiO4 nanotubes with Di = 29 and 49 nm had higher

value of Ca/P ratio than Di = 25 and 45 nm because of their larger diameter and lower

Sa.

Fig. 6.18 The percentage of atomic ratio of Ca/P of mineralised HA on a) as-formed and

annealed dimpled surface of TiO2-ZrO2-ZrTiO4 nanotubes and b) as-formed and

annealed TiO2 nanotubes fabricated in non-aqueous electrolyte

Fig. 6.19 shows local pH changes up to 3 weeks of mineralisation of HA. Firstly the pH

value increased due to the formation of negative hydrolysed nanotubular surface. The

changes of pH value were negligible when the absorption of Ca2+ ions was undertaken.

The absorption of PO43- as a result of exchanging with OH- increased the pH value. The

pH value decreased when the OH- consumed to form hydroxyapatite [Ca10(PO4)6(OH)2]

and it remained constant after the HA formation completed.

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

5%-20 10%-20 5%-30 10%-30

Ca/

P (a

t %)


Ti50Zr As FormedAnnealed

a)

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

5%-20 10%-20 5%-30 10%-30

Ca/

P (a

t %)


CP-Ti As FormedAnnealed

b)

164


TiO2-ZrO2-ZrTiO4 in a non-aqueous electrolyte with 5 and 10 wt % H2O at 20 and 30 V

The nanomechanical characteristics of the mineralised HA on the surface of TiO2-ZrO2-

ZrTiO4 nanotubes were measured using a nanoindentation. Fig. 6.20 shows the curve of

the load-unload forces versus the nanoindentation depths of the HA. The Young’s

modulus was 49.9 GPa and the nano hardness was 599.7 MPa. It has been reported that

the average elastic modulus for the osteons and the interstitial lamellae in the longitude

direction was 24.7 GPa and 30.1 GPa. As mentioned before the hardness was reported

to be varied among the microstructure components in the range of 0.41 - 0.89 GPa

[258].

165


nanotubular surface of TiO2-ZrO2-ZrTiO4 fabricated at 10 wt % H2O and the applied

potential of 30 V

6.4 Conclusions

Nanotubular surfaces of TiO2-ZrO2-ZrTiO4 and TiO2 were fabricated via anodisation on

Ti50Zr and CP-Ti. The formation of hydroxyapatite on the TiO2-ZrO2-ZrTiO4 and TiO2

nanotubular surfaces in comparison to flat anodised surface were evaluated. The main

conclusions are as follows.

1. Increasing the applied potential resulted in an increase in Di, leading to (i)

decreasing roughness parameters of Sa and Sq, (ii) increasing hydrophilic

properties and (iii) increasing the surface energy of TiO2-ZrO2-ZrTiO4

nanotubes. The mean Sa and Sq of nanotubular surface decreased,

hydrophilic properties and surface energy increased after annealing. The

same tendency was observed for TiO2 nanotubes on the surface of CP-Ti,

although their hydrophilic properties and surface energy were lower than

those for TiO2-ZrO2-ZrTiO4 nanotubes.

2. All the nanotubular surfaces of TiO2-ZrO2-ZrTiO4 fabricated in aqueous

electrolyte were bioactive and the size, shape, hydrophilic properties and

crystalline phase of the surface layers affected the Ca/P atomic ratio of the

0

0.002

0.004

0.006

0.008

0.01

0.012

0 2E-10 4E-10 6E-10 8E-10 1E-09

Load

(N)

Displacement (m)

166

deposited HA during SBF immersion. All the nanotubular surfaces of TiO2

fabricated in aqueous electrolyte were bioactive and the growth of HA was

influenced by the hydrophilic properties and roughness parameter rather than

the Di and crystalline phase of the nanotubular layer.

3. The nanoscale topography and chemistry of the nanotubular surfaces

affected the mineralisation of hydroxyapatite while immersion in SBF. The

bioactivity of the nanotubular surfaces was clearly enhanced compared with

the flat anodised surfaces and the bare metals of Ti50Zr and pure Ti.

4. Increasing of water content and applied potential resulted in an increase in

mean Di, Sa and Sq of TiO2-ZrO2-ZrTiO4 fabricated in non-aqueous

electrolyte. The hydrophilic properties of the nanotubular surface increased

with an increase in the nanotube sizes and roughness parameters.

5. All the nano size surfaces of the as-formed and annealed dimpled surface of

TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes were bioactive and the size, shape,

roughness parameters and hydrophilic properties of the nanotubular layers

affected the atomic ratio of Ca/P after immersion in the m-SBF for 3 weeks.

167

Chapter 7 Cell response of nanotubes formed in both aqueous

electrolyte and non-aqueous electrolyte

Abstract

Studies of biomaterial surfaces and their influence on cell behaviour provide insights

concerning the design of surface physicochemical and topography properties of implant

materials. Fabrication of biocompatible metal oxide nanotubes on metallic biomaterials,

especially titanium alloys such as Ti50Zr via anodisation, alters the surface chemistry as

well as surface topography of the alloy. In this chapter, different groups of TiO2-ZrO2-

ZrTiO4 nanotubes that exhibit diverse nanoscale dimensional characteristics (i.e., inner

diameter Di, outer diameter Do and wall thicknesses Wt) and surface chemistry were

fabricated via anodisation in both aqueous and non-aqueous electrolytes. Two groups of

TiO2-ZrO2-ZrTiO4 nanotubes were post treated to assess their biocompatibility. The

nanotubes were annealed and characterised using scanning electron microscopy (SEM)

and 3D-Profilemetry. The potential applied during anodisation influenced the oxidation

rate of titanium and zirconium; thereby resulting in different nanoscale characteristics

for the nanotubes. Both the different oxidation and dissolution rates led to changes in

the surface roughness parameters. The in vitro cell response to the nanotubes with

different nanoscale dimensional characteristics was assessed using osteoblast cells

(SaOS2). Results of the MTS assay indicated that the nanotubes fabricated in aqueous

electrolyte with inner diameter (Di) ~ 18 nm exhibited the highest percentage of cell

adhesion of 65.2 %. This result can be compared to (i) 64.4 % cell adhesion at Di ~

30 nm, (ii) 41.0 % cell adhesion at Di ~ 40 nm, (iii) 25.9 % cell adhesion at Di ~ 59 nm,

(iv) 33.1 % at Di ~ 64 nm, and (v) 33.5 % at Di ~ 82 nm. The nanotubes with Di ~

59 nm exhibited the greatest roughness parameter of Sa (Mean Roughness), leading to

the lowest ability to interlock with SaOS2 cells. Post-treating on nanotubular surface by

making it cleaner and cover it with hydroxyapatite (HA) resulted in an increase in cell

adhesion notably to 74.8 % and 59.7 %, respectively. Results of the MTS assay

indicated that the percentage of cell adhesion on the nanotubes fabricated in non-

aqueous electrolyte was critically affected by the nanoscale topographical parameters

including the tube inner diameter (Di), the tube wall thickness (Wt), the amplitude

roughness (Sa) and the spacing roughness (Sm) of the nanotubular surface. The cell

adhesion was essentially promoted to 84.9 % on nanotubes with small inner diameter of

168

25 nm, or 80.3 % on nanotubes with large wall thickness of 34 nm due to the

accelerated integrin clustering and focal contacts formation. A nanotubular surface with

low spacing roughness of 33 nm3/nm2 led to a low cell adhesion of 61.0 %; similarly a

nanotubular surface with high amplitude roughness of 1.03 µm showed a cell adhesion

of 61.5 % even the inner diameters (29 nm) and the wall thicknesses (24 nm) of the

nanotubes were within the critical borderlines for cells to survive and thrive.

7.1 Introduction

The application of orthopeadic or dental implants is effectual when they form a solid

connective boundary between the bone tissue and the surface of the material in the

absence of a fibrous tissue interface [15]. This attribute is related to the molecules

involved in the adhesion of bone cells as well as the morphology and physico-chemical

properties of the implant surface. The morphology and capacity for proliferation and

differentiation of bone cells are affected by the quality of this adhesion [15]. The

biocompatible materials that are also known as ‘hybrid materials’ or ‘osteoconductive

structural materials’ can adsorb biomolecules and water from body fluid or culture

medium in vivo or in vitro [15] because cells are not equipped to interact directly with

biomaterials [34]. There are several stages regarding cell adhesion, which are known as

short-term and long-term adhesion. These phases of cell adhesion to an implant surface

involve different forces that are documented [15, 34, 137]. Surface chemistry and

surface energy influence short term adhesion; whereas topography affects long-term

adhesion as well as the cell morphology, proliferation, differentiation and migration [15,

259].

Chemical surface treatment is a surface modification method that increases

biocompatibility and bioactivity of biomaterial surfaces, especially that of metals.

Anodic oxidation, which is an electrochemical surface treatment, can be used to

fabricate a nanotubular or nanoporous layer of metal oxides on the surface of titanium

(Ti), zirconium (Zr), tantalum (Ta), niobium (Nb) and hafnium (Hf). Titanium and

binary, tertiary and higher titanium alloys especially alloyed with Zr, Ta and Nb

elements have been used extensively for orthopeadic implants due to their favourable

mechanical properties and biocompatibility, which is related to their ability to form a

protective oxide surface in air. A range of nanoscale dimensional characteristics (i.e.,

inner diameter, outer diameter and wall thickness) have been fabricated by altering

169

electrochemical conditions such the type and concentration of electrolyte, applied

potential and anodisation time [11, 234, 260]. Several studies [11, 165, 167, 177, 180]

have shown that the bioactivity of nanotubes is a function of their fabrication method

and that the bioactivity may increase by post treatment such as annealing.

The effect of macro, micro and nano sized topography on cell adhesion has been

reported [261-263]. Contact guidance of cells defines the alignment of cells that is

affected by the substrate surface morphology. The importance of the roughness

amplitude (height) parameter, space and hybrid parameters on cell adhesion has been

investigated [137, 264]. Surface chemistry plays an important role in cell adhesion

because it influences short term adhesion, which is a step that involves adsorption and

rearrangement of proteins [137]. It has been reported that cells adhere better on

hydrophilic surfaces, on surfaces with higher surface energy, and surfaces with

functional groups such as OH- and NH2- [137]. TiO2 nanotubes, as fabricated by an

electrochemical surface treatment, have exhibited an improved surface modification due

to (i) changes in the surface topography, such as the roughness and organisation of the

surface, and (ii) changes in surface chemistry that effect an increase in hydrophilic

properties and surface energy [180, 240, 265].

The investigation of different sizes of nanotubes indicated that cellular activity

increased when the length scale was less than 30 nm with a maximum activity at 15 nm.

A tube diameter larger than 50 nm decreased cellular activity [17]. The surface occupied

by the head of integrins (which are trans-membrane receptors that bind to specific

amino acids of proteins to form focal contact) has been reported to be around 10 nm

[137]. Consideration of these dimensions indicates that 15 nm can be an optimum

spacing for intergrin clustering and focal contact formation [17]. This larger length scale

was suitable for both bone forming cells and bone resorbing cells [266]. In another

study of human mesenchymal stem cells (hMSC), these cells adhered and grew with

minimum differentiation on small nanotubes of <50 nm. On the other hand, on larger

diameter nanotubes, i.e., > 50 nm, the hMSC cells became elongated and were guided to

differentiate into osteoblast cells. This result agreed with the general belief that cells

incline to differentiate into a specific lineage when they are forced to adapt under stress

[21]. There is also a report on an up-regulated level of ALP (alkaline phosphatase)

activity and osteoblast (MC3T3-E1 mouse) elongation on annealed TiO2 nanotubes;

170

which demonstrated a higher cell adhesion on 100 nm nanotubes than on 30 - 70 nm

nanotubes [20].

It has been shown that titanium zirconium alloy (TiZr) is a biocompatible titanium alloy

[252]. Therefore, studies have investigated the biocompatibility of TiZr alloy by

growing nanotubes on its surface via anodisation and rapid breakdown anodisation

(RBA) [151, 162, 267]. Various nanoarchitectures were fabricated using two-step

anodisation in an non-aqueous electrolyte to obtain nanopattern scaffold for cell growth

[268]. The influence of surface energy of titanium-zirconium alloy on osteoblast cell

functions was reported [27], therefore it would be a worthwhile study to investigate the

effect of the size, the roughness and wettability of nanotubes on the growth of bone

cells. The effect of size and the roughness of nanotubes have been discussed in Chapters

4 and 5. In this chapter TiO2-ZrO2-ZrTiO4 nanotubular layers were fabricated in

aqueous electrolyte with mean Di (i) 18 ± 6, (ii) 30 ± 10, (iii) 40 ± 12, (iv) 59 ± 17, (v)

64 ± 23 and (vi) 82 ± 26 nm; Do equal to (i) 38 ± 7, (ii) 63 ± 10, (iii) 74 ± 14, (iv) 102 ±

17, (v) 117 ± 26 and (vi) 136 ± 25; and Wt equal to (i) 7 ± 3, (ii) 11 ± 4, (iii) 13 ± 4, (iv)

17 ± 5, (v) 21 ± 7 and (vi) 24 ± 7 nm, respectively. The effect of some post surface

modification for improving the cell adhesion was assessed. Another group of TiO2-

ZrO2-ZrTiO4 nanotubes with dimpled surface were fabricated in non-aqueous

electrolyte by removing the upper layer of annealed nanotubes. The roughness and

water contact angle were measured and their surface energies calculated. SaOS2 cells

were seeded onto the TiO2-ZrO2-ZrTiO4 nanotubular surfaces and the effects of the

distribution of nanotube size, roughness parameters and surface energy of the

nanotubular surface on the cellular activity were comprehensively investigated.


Ti50Zr alloy (wt % hereafter) discs with 8 mm diameter and 2 mm thickness were

electrical discharge machined (EDM) from a cast ingot. The disc samples were polished

to a 1 µm finish with diamond paste. Subsequently the discs were degreased by

sonification in, successively, methanol, isopropanol, acetone and ethanol for 15 min

each; i.e., a total wash time of 1 hour. The final preparation involved a wash with

deionised water and drying in a stream of nitrogen gas.

171

The Ti50Zr discs were spot-welded to titanium foil to enable the construction of the

anode for an electrochemical cell. The samples were placed 4 cm from the platinum

counter electrode with a 1 cm2 surface area in a two-electrode configuration using a DC

power supply. One group of TiO2-ZrO2-ZrTiO4 nanotubes were fabricated via

anodisation at room temperature with the electrolyte composed of 1 M (NH4)2SO4 with

the addition of 0.5 wt % of NH4F (Sigma-Aldrich reagent grades). After the

electrochemical treatment the samples were rinsed in deionised water for 5 min and

dried with a nitrogen gas stream.

Six applied potentials, i.e., 10, 15, 20, 25, 30 and 35 V, were used to create different

sizes and morphologies of nanotubes. The TiO2-ZrO2-ZrTiO4 nanotubes were pre-

treated by immersion in 0.5 M NaOH at 50 ˚C for 2 min, and then rinsed with deionised

water to increase biomineralisation. The samples were then annealed at 500 ˚C for 3 h at

the heating rate of 2 ˚C/min in a conventional muffle furnace (Nabertherm

LT15/13/P330; Nabertherm GmbH, Lilienthal, Germany). To reduce any contamination

of remaining chemicals from the electrolyte some samples were soaked at 70 °C for 14

days in pure water with periodic replacements (three times per day). To increase

osseointegration one series of samples were soaked in modified simulated body fluid

(m-SBFand incubated at 37 ˚C for 3 weeks. The ion compositions of the m-SBF [213],

which have an ion composition nearly equal to that of blood plasma, were adjusted by

dissolving 5.403 g NaCl, 0.504 g NaHCO3, 0.426 g Na2CO3, 0.225 g KCl, 0.230 g

K2HPO4 - 3H2O, 0.311 g MgCl2 - 6H2O, 0.293 g CaCl2, and 0.072 g Na2SO4 in

deionised water as previously reported [213]. The m-SBF was buffered at pH 7.4 at

37 ˚C using 2-(4-(2-hydroxyethyl)-1-piperazinyl) ethane sulfonic acid (aka HEPES) and

1 M NaOH. An amount of 17.892 g HEPES has dissolved in 100 ml 0.2 M NaOH to

adjust the pH. The ability of the samples to form apatite was evaluated in a static SBF

environment. After removing the samples from m-SBF they were rinsed with deionised

water and dried at room temperature for 24 h.

Another group of TiO2-ZrO2-ZrTiO4 nanotubues were fabricated in non-aqueous

electrolyte. The electrolyte was composed of 0.5 wt % of NH4F with the addition of 5

and 10 wt % of H2O in ethylene glycol (Sigma-Aldrich reagent grades). After the

electrochemical treatment the sample surfaces were rinsed with acetone for 5 min and

followed by drying in air. Two applied potentials, i.e., 20 and 30 V were used for each

172

electrolyte composition to create different sizes and morphologies of nanotubes. The

TiO2-ZrO2-ZrTiO4 nanotubes were post-treated by immersion in 0.5 M HCl at room

temperature for 1 min, then rinsed with deionised water for 1 min to dissolve the

remained upper nanoporous layer. The samples were then heat-treated at 500 ˚C for 2 h

at the heating rate of 2 ˚C/min in a conventional muffle furnace (Nabertherm

LT15/13/P330; Nabertherm GmbH, Lilienthal, Germany) to crystallise the amorphous

nanotubular layer. The brittle and cracked nanotubular layer after annealing were

removed by sonification in pure water for 15 min and dried in a stream of nitrogen gas.

To enhance the bioactivity the nanotubular layer was chemically treated by immersing

in 0.5 M NaOH solution at 50 °C for up to 2 min then rinsed with pure water for 2 min

and dried in a stream of nitrogen gas. The inner diameter, outer diameter and wall

thickness of the anodised nanotubes were denoted Di, Do and Wt, respectively. Each data

point is an average of 100 measurements on different positions for each of three samples

(300 measurements).

A field-emission scanning electron microscope (FESEM, ZEISS SUPRA 40 VP) was

used for metallographic characterisation. The amplitude parameters and space

parameters was determined with a 3D-Profilometer (Bruker, Contour GT-K1; Bruker

Pte Ltd, Singapore) that was coupled with SurfVision software (Veeco Instruments Inc.;

Plainview, NY, USA). Mean Roughness (Sa) and Root Mean Square (RMS) Roughness

(Sq), amplitude parameters, which measure the vertical characteristics of surface

deviations, were evaluated over the complete 3D surface [210]. Sskw, the skewness of a

3D surface texture, represents the degree of symmetry of the surface heights about the

mean plane, and it is the third central moment of the profile amplitude probability

density function [210]. The kurtosis of the 3D surface texture, Sku, indicates the nature

of the height distribution that is the fourth central moment of the profile amplitude

probability density function and describes the sharpness of the distribution [210]. The

spacing parameters such as Sm (Surface Material Volume), Sc (Core Void Volume) and

Sv (Surface Void Volume) that represent the volume of material or space provided by

the surface relative to the cross sectional area of the measurement were evaluated [210].

A goniometer (NRL C.A. Goniometer, Ramé-hart, Inc.; Succasunna, NJ-USA) was

used to measure the water contact angle. The surface energy was calculated based on

the Owens-Wendt (OW) method [211], given by:

173


d+√γLp γS

p) (3 − 1)

where L indicates the liquid surface tension (water=72.8 mJm-2, glycerol=63.4 mJm-2);

Ld and S

d are the liquid and solid dispersive component (water=21.8 mJm-2,

glycerol=37.0 mJm-2). The variables Lp and S

p are indicative of the liquid and solid

polar components (water=51.0 mJm-2, glycerol=26.4 mJm-2) and S is the sum of Sd

and Sp [212].

All samples for the cell culture studies were sterilised in a muffle furnace at 180 ˚C for

3 h. The samples were placed in a well in the cell culture plate. SaOS2 cells (Barwon

Biomedical Research, Geelong Hospital, Victoria, Australia) were seeded on the surface

of the TiO2-ZrO2-ZrTiO4 samples with different inner, outer diameter and wall

thickness. The seeded cell density was 5×103 cells per well (100 mm2 area). A MTS

assay was used to measure the in vitro proliferation of the SaOS2 cells. The cell

morphology was observed by means of SEM after cell culturing. The cells were

dehydrated for the SEM observation by immersion in a buffer solution in which the

ethanol concentrations were increased progressively every 10 min to 60, 70, 80, 90 and

100 %. This procedure was followed by chemical drying using hexamethyldisilazane

(HMDS, Sigma - Aldrich, Australia) for 10 min. A gold layer was deposited on the

samples prior to SEM observation.

The cell morphology also was observed using a confocal microscopy (Leica SP5, Leica

Microsystems, Germany). The cell-seeded samples after cell culture were fixed in

paraformaldehyde, and then permeabilised with triton-X 100 in phosphate-buffered

saline (PBS) (Sigma - Aldrich, Australia) for 10 min each at room temperature for the

confocal microscopy observation. Then the samples were stained with 1 % phalloidin

and 40-6-diamidino-2-phenylindole for 40 min at room temperature. Between each of

the steps three washes by PBS were included. If storage of stained samples was required

they were stored in PBS at 4 °C and the observations were conducted within a week of

staining. Significant differences in the cell number were analysed using one-way

ANOVA (ρ < 0.05).

174



nanotubes fabricated in aqueous electrolyte

The SEM images of the TiO2-ZrO2-ZrTiO4 nanotubes fabricated on Ti50Zr alloy under

different applied potentials and then annealed at 500 oC for 3 h are shown in Fig. 7.1.

The top view of the TiO2-ZrO2-ZrTiO4 nanotubes fabricated in 0.5 wt % NH4F at 10 V

for 1 h is shown in Fig. 7.1(a). In this study, the F- concentration in the electrolyte,

nanotube inner diameter, outer diameter and wall thickness are denoted as CF-, Di, Do,

and Wt, respectively.

Fig. 7.1 SEM images of annealed TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 10 V, b)

15 V, c) 20 V, and d) 25 V, e) 30 V and f) 35 V

175

The mechanism of growing nanotubes on the surface of titanium, which is the result of

competition between oxidation of the metal to TiO2 and dissolution of TiO2 to form

[TiF6]2-, is elucidated in prior chapters [58, 249]. The formation of TiO2-ZrO2-ZrTiO4

nanotubes was presented previously [240]. The principal oxidation and reduction

reactions occur as follows:

Ti Ti4+ + 4 e- (7-1)

Zr Zr4+ + 4 e- (7-2)

4H+ + 4e- 2H2 (7-3)

The formation of a soluble complex of Ti4+ and Zr4+ with the ligand of F- takes place as:

Ti4+ + Zr4+ + 12 F-

[TiF6]2- + [ZrF6]2- (7-4)

Using anodisation in the electrolyte of 0.5 wt % of NH4F, titania-zirconia-zirconium

titanate (TiO2-ZrO2-ZrTiO4) nanotubes were fabricated on the surface of Ti50Zr alloy at

an applied potential in the range of 10 - 35 V for 1 h. The original grain structure of the

alloy was reflected in the subsequent growth of the nanotubes with an orientation based

on the needle-like morphology of the underlying alloy [240].

Figure 7.2 shows the mean Di, Do and Wt of annealed TiO2-ZrO2-ZrTiO4 nanotubes

anodised in an electrolyte with a constant CF- of 0.5 wt % and where the applied

potential increased from 10 to 35 V for 1 h. The mean Di of nanotubes increased with

respect to potentials of 10, 15, 20, 25, 30 and 35 V. A similar increase in the mean of Do

nanotubes at 10 V was from nm to 53 < Do < 73 and increased upto 111 < Do < 161 nm,

respectively when the applied potential increased to 15, 20, 25, 30 and 35 V. The

difference in Wt was between 12 to 14 nm by increasing the applied potential starting

10 V and reaching to 35 V.

176

Fig. 7.2 Mean nanotube size of Di, Do and Wt for annealed TiO2-ZrO2-ZrTiO4 nanotubes

anodised on Ti50Zr as a function of applied potential

The distribution of Di, Do and Wt for Ti50Zr alloy resulted from the different oxidation

and soluble complex formation of Ti and Zr with F- ions [240]. The frequency of

smaller sized Di and Do decreased on increasing the applied potential; thus enabling the

formation of larger inner and outer diameters due to the higher anodic current. The

broad distribution of Di and Do at 30 and 35 V shows a better condition for oxidation of

Ti and Zr because of high current flow while the dissolution rate of (NH4)2[TiF6] and

(NH4)2[ZrF6] remained constant.

Figure 7.3 shows the effect of changing nanotube size distribution on surface roughness

amplitude parameters. The values of Sa and Sq increased with an increase in applied

potential up to 25 V (Di = 59 nm). They revealed lower values at 30 and 35 V although

the mean Di increased. The texture of a surface, which is defined as surface roughness,

affects the character of the surface when it is contrasted to another surface of different

texture. There are several classical parameters used to evaluate surface roughnesses.

Amplitude parameters that are used to measure the vertical surface deviation

0

20

40

60

80

100

120

140

160

180

10 15 20 25 30 35

Mea

n na

notu

be si

zes (

nm)


WtDiDo

177

characteristics are one of the three roughness parameters [210]. The value of Sa

increased with increasing applied potential due to the availability of a larger anodic

current and its effect on oxidation reaction and consequently on dissolution.

Fig. 7.3 Illustration of changing roughness parameters of Sa and Sq via changing Di of

TiO2-ZrO2-ZrTiO4 nanotubes anodised on Ti50Zr

Ti50Zr alloy consists of two phases; α and β that contain a different amount of Zr. The β

phase consists of higher amount of Zr than α phase. The standard enthalpies of

formation of ZrO2 and TiO2 are different, and also the solubility equilibrium constant of

(NH4)2[ZrF6] is higher than that of (NH4)2[TiF6]. Therefore, two different heights of

TiO2 nanotubes and ZrO2 nanotubes with interspace formed [237]. The reduction of Sa

for nanotubular layers fabricated at 30 and 35 V can be explained by examining the

equilibrium state between the oxidation and dissolution reactions that was achieved at a

higher current density. Thus, the kinetics of oxidation in comparison to dissolution was

higher in the same period of time, which resulted in larger Di and Do and lower Sa and

Sq. The inference is that there was more physio-chemical activity inside the tubes than

along the length dimension.

0

0.5

1

1.5

2

2.5

18 30 40 59 64 82

Rou

ghne

ss a

mpi

tude

par

amet

ers -

(µm

)

Mean of Di (nm)

Sa

Sq

178

Consider a line, or ruler, that is located parallel to and intersecting a roughened surface.

This line will have intervals of intersection with the 2-D surface profile. The ratio of the

intersection lengths with the original measurement is defined as the bearing area. As the

ruler is moved more deep into the surface profile then the bearing area will increase and

the bearing area curve can be computed. The amount of material contained in the

surface peaks accounts for 0 to 10 % of the bearing area ratio and is represented by the

parameter Sm. In an alike manner, the volume that the surface would support from 10 -

80 % of the bearing ratio is Sc, and the relevant parameter from 80 - 100 % of the

bearing ratio is Sv. The dimensions of Sm, Sc, and Sv are nm3/nm2 [269]. Table 7.1

details the surface area index (SI) that is calculated by dividing the projected surface

area, which is the total exposed three-dimensional surface area being analysed, to the

surface area measured in the lateral direction, as well as the spacing parameters of Sm,

Sc and Sv. The nanotubular layer with Di = 64 nm revealed lowest Sm and the nanotubes

layer with Di = 59 nm revealed highest Sm.

179

Table 7.1 Surface index and spacing parameters of TiO2-ZrO2-ZrTiO4 nanotubes

fabricated in different applied potential

Anodisation Condition

(Mean size) SI Sm

(nm3/nm2)

Sc

(nm3/nm2)

Sv

(nm3/nm2)

10V (Di = 18 nm) 2.2 ± 0.3 34 ± 10 556 ± 106 153 ± 35

15V (Di = 30 nm) 2.5 ± 0.5 35 ± 10 820 ± 200 95 ± 33

20V (Di = 40 nm) 3.0 ± 0.6 35 ± 9 874 ± 204 124 ± 36

25V (Di = 59 nm) 5.3 ± 0.8 43 ± 26 1435 ± 223 231 ± 73

30V (Di = 64 nm) 2.9 ± 0.5 28 ± 4 711 ± 126 75 ± 13

35V (Di = 82 nm) 3.9 ± 0.4 30 ± 2 909 ± 95 100 ± 9

180

The dominant nature of topography is indicated by Sskw. High peaks are implied when

Sskw > 0; whereas Sskw < 0 is indicative of valley-like features such as deep scratches

[210]. The negative Sskw values show that the samples with the nanotubular layer

fabricated under different applied potentials exhibited low valleys rather than high

peaks reflected of their interspace. The nanotubular layer fabricated at 25 V revealed a

higher negative skewness than those nanotubes fabricated at other applied potential.

The sharpness distribution of the sample surface is described by Sku. For normally

distributed surfaces Sku is 3.0. When Sku < 3.0, the distribution is platykurtoic,

indicating a few high peaks and low valleys. When Sku > 3.0, the distribution is

leptokurtoic and indicates many high peaks and low valleys [210]. All the TiO2-ZrO2-

ZrTiO4 nanotubes fabricated at applied potentials from 15 to 35 V exhibited

leptokurtoic structures due to possible influence of their interspace although they were

near mesokurtic distributions when they formed at 10, 30 and 35 V. The nanotubular

layer fabricated at 25 V revealed a higher Sku value, indicating many peaks and valleys

that agree with other roughness parameters such as Sa and Sq in this study. The reason

for these roughness distribution responses relates to a higher physio-chemical activity

inside pores rather than at the end of nanotubes. The increased pore activity arises due

to the availability of anodic current at the higher applied potential while the dissolution

rate of (NH4)2[TiF6] and (NH4)2[ZrF6] remained constant.


aqueous electrolyte

In this part the cell adhesion on the surface of highly ordered nanotubular layer

fabricated at applied potentials from 10 to 35 V with a distribution of different Di, Do

and Wt was investigated. These distributions and interspace, which were related to the

different phase distributions of α and β phases and the needle like microstructure in

Ti50Zr alloy, played a key role in affecting the cellular response. SEM observations of

SaOS2 cells cultured on samples were carried out after 24 h of incubation. The cell

shape and spreading behaviour with formation of cytoplasmic extensions and filopodia

of the SaOS2 cells adhering on top of the various sizes of TiO2-ZrO2-ZrTiO4 nanotubes

were different. Fig. 7.4(e) shows the extension of filopodia, the exploring tools of the

cell, towards neighbouring spaces to find a more suitable place for their further

development. It was found that the cells were more flattened without any particular

181

orientation on the top of nanotubes with 12 < Di < 24 nm, 20 < Di < 40 nm, 28 < Di <

52 nm and 56 < Di < 108 nm than on top of nanotubes with 42 < Di < 76 nm and 41 < Di

< 87 nm, on which they appeared quasi round and non-spreading. The morphology of

the underneath nanotubular layer was visible through the extension part of cells at

higher magnification as shown in Fig. 7.4(f) and 7.5(f). Cell filopodia travelled between

nanotubes and the filopodia ends served as anchor points that allowed probing of the

local environment for proteins so that more stable focal contacts could be established.

Fig. 7.6 illustrates the possible movement of filopodia on nanotubular surface to make

the stable adhesion sites.

182




magnification and f) 20 V higher magnification

183




magnification and f) 35 V higher magnification

184

Fig. 7.6 Schematic illustration of exploring tools of an osteoblast cell on nanotubular

surface

The osteoblast cells which form new bone, and osteoclast cells which degrade bone

similar to other cells (except blood cells), adhere to extra cellular matrix, ECM [264]. In

vitro, within a few seconds, biomaterial surfaces adsorb water, ions and biomolecules

such as proteins from the culture medium. The cell then senses the foreign surface

through ECM and this adsorbed layer. The cell adheres via several phases [137]. The

biomaterial links to cell in the early phase, or attachment phase, by physicochemical

interactions such as ionic forces, van der Waals forces, etc. The later phase, or adhesion

phase, involves different biomolecules such as ECM proteins, cell membrane proteins,

cytoskeleton proteins, cell surface receptors and integrins [15, 137]. Integrins that act as

an interface between the intracellular and extracellular component are heterodimer,

which have different sub-units that bind to specific amino acids. Through this

interaction, they transmit a signal from the ECM to the nucleus to regulate many

cellular functions such as cell adhesion, motility, shape, growth and differentiation [137,

185

264]. Focal contacts are adhesion sites between tissue cultured cells and the substrate

surface that form when integrins are clustered because of interaction with the ECM-

adsorbed biomolecules [15, 264]. An interface of cell/matrix/substrate forms via ECM

protein synthesised by cells [137] when the cytoskeleton is recognised to allow the cell

to spread on the substrate [264]. Fig. 7.7 shows a schematic illustration of cells and

focal contacts which included integrins on the surface of nanotubular layer.

186

Fig. 7.7 Proposed illustration of the position of integrins on a) Di = 18 ± 6 nm best

nanospacing, b) rough surface of Di = 59 ± 17nm and c) Wt = 24 ± 7 nm most existing

of proteins

Regardless of other properties of the surface and the population of proteins, since the

surface occupancy by the head of an integrin is approximately 8 - 12 nm in size, they

187

were better clustered and flattened on nanospacing of 12 < Di < 24 nm and formed

stable focal contacts. Fig. 7.7 (a) and (c) shows how cells behave when the Di of

nanotube increased to adhere themselves on the nanotubular surface. This is in

accordance to the proposed optimum nanospacing (< 30 nm) reported by other

researchers [17, 19]. Although nanotubular layer with 56 < Di < 108 nm was not in the

range of optimum nanospacing, the range of Wt of 17 < Wt < 31 nm where the

population of proteins congregated onto the wall tops of nanotubes due to the large

empty space of big nanotubes on the surface results in the occurrence of the flat

morphology of cells. A higher aggregation and agglomeration of proteins on the surface

of TiO2 nanotubes in comparison to flat Ti (with a TiO2 native oxide layer) has been

reported previously [21]. The cell behavior on the surface of the nanotubular layer of

TiO2-ZrO2-ZrTiO4 with 42 < Di < 76 nm and 41 < Di < 87 nm was also affected by

other properties of the nanotubular surface such as roughness parameters.

For a better understanding of the quantity of spreading of cells on different substrates,

ImageJ software was used to measure the length of filopodia and the surface area of

spreading cells and the results are presented in Fig. 7.8(a) and 7.8(b). The longest

filopodia grew on the nanotube surfaces of mean Di = 40 nm after 1 and 7 days. The

shortest filopodia grew on nanotubes with mean Di = 18 nm after 1 day and mean Di =

30 nm after 7 days. The highest surface area was occupied by cells on the nanotubes

with mean Di = 40 nm after 1 day and mean Di = 59 nm after 7 days. The lowest surface

area was occupied by cells on the nanotubes with mean Di = 64 nm after 1 day and

mean Di = 30 nm after 7 days. The quasi round and non-spreading cells exhibit higher

peaks in profilometry measurement than flattened and spread cell. Therefore, using the

Sp, which is the largest peak height value from the reference surface within the defined

sampling area, the heights of selected cells were measured, as shown in Fig. 7.8(c). The

highest cells grew on the nanotubes with mean Di = 59 nm after 1 day and mean Di =

64 nm after 7 days. The lowest cells grew on the nanotubes with mean Di = 64 nm after

1 day and mean Di = 40 nm after 7 days. These graphs show results in accordance to the

above mentioned behaviour of cells on nanotubular surfaces. The cells exhibited shorter

filopodia, lower surface area and lower cell height on nanotubular surface with desirable

nanosize under 30 nm such as Di where the head of integrins were fitted great and Wt

where sufficient adsorbed protein existed for performing the requiring contacts than

188

nanotubular surface with nanosize upper 30 nm. The exceptions exist when other

surface parameter influenced the cell behaviour.

189

Fig. 7.8 Illustration of a) the length of filopodia, b) surface area and c) height of SaOS2

cells after 1 day and 7 days grown on different nanotube sizes (ρ < 0.05)

0

1

2

3

4

5

6

7

8

9

10

18 30 40 59 64 82

Filo

podi

a le

ngth

(µm

)

Mean of Di (nm)

Length of Filopodia

1 Day

7 Days

a)

-2000

0

2000

4000

6000

8000

10000

12000

18 30 40 59 64 82

Surf

ace

Are

a (µ

m2 )

Mean of Di (nm)

Surface area of elangated cells

1 Day

7 Days

b)

-10

0

10

20

30

40

50

60

18 30 40 59 64 81

Sp (µ

m)

Mean of Di (nm)

Heigth of Cells

1 Day

7 Days

c)

190

SaOS2 cells on nanotubular layer with 28 < Di < 52 nm and 56 < Di < 108 nm were

spread wider with lamellipods than on nanotubular layer with 12 < Di < 24 nm and 20 <

Di < 40 nm; thereby demonstrating a high level of cell migration. SaOS2 cells on

nanotubular layers of 42 < Di < 76 nm and 41 < Di < 87 nm were less spread than the

above mentioned nanotube size and exhibited dorsal ruffles; thereby demonstrating a

high cellular activity reflecting in their height (Sp) but they were relatively sparse on the

surface after 7 days. The SEM images of the cell growth after 7 days as shown in Fig.

7.9 and 7.10 indicate the effect of nanospacing and surface chemistry on cell behaviour.

TiO2-ZrO2-ZrTiO4 nanotubes with 41 < Di < 87 nm were lower in surface energy (Table

6.2) in comparison to other nanotubular layer with Di > 30 nm, which might influence

protein adsorption and their rearrangement on the nanotubular layer [15]. They also

demonstrated lower spacing roughness parameters than other nanotubular layers in this

study, which would potentially disrupt the formation of focal adhesion.

191

Fig. 7.9 SEM images of SaOS2 cells cultured for 7 days on the four different pore sizes

of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 10 V- low magnification, b) 10 V-


V- low magnification and f) 20 V higher magnification

192


sizes of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 25 V- low magnification, b) 25 V-


V- low magnification and f) 35 V higher magnification

The percentage of attached cells on six surface morphologies and the percentage of total

cells were obtained using MTS assay and are presented in Fig. 7.11. Among the six

nanotubular layers, the highest percentage of attached cells was 65.2 ± 9.4 % on the

nanotubes with mean Di of 18 nm. The percentage of cells attached to nanotubes with

193

higher mean Di of 30 and 40 nm decreased to 64.4 ± 5.7 % and 41.0 ± 2.6 %.

Considering previous work [17] it was expected that the percentage of attached cells

decreases with an increase in the distribution of Di of the nanotubes. In terms of

nanotubular layers with 42 < Di < 76 nm, which revealed lower cell attachment than

other, the roughness of the surface might also play an important role. It has been

reported that when the height and spacing between peaks of a surface increased, the

cells were not able to adhere as they adhere on the top of peaks of smoother surface

[137, 264]. This can be the reason for the lowest cell attachment on the nanotubular

surface with 42 < Di < 76 nm due to its highest Sa. The percentage of cells attached to

nanotubes with higher mean Di of 64 and 82 nm increased to 33.1 ± 2.2 % and 33.5 ±

4.1 % but they were still less than that on nanotubes with smaller mean Di. This result

agrees with the hypothesis that the human osteoblasts spread more intimately on lower

roughness amplitude surfaces than on rough ones when the topography is on the

nanoscale [262]. Two reasons may describe the lower cell attachment on nanotubular

surface with 41 < Di < 87 nm than 56 < Di < 108 nm: its lower Sm and surface energy

although the Sa did not reveal a large difference. In the contrary of expectation, the

nanotubular layer with 56 < Di < 108 nm did not reveal the lowest cell attachment

because of the large space of Di which prevents anchoring of the head of integrins. The

Wt of this nanotubular layer was 17 < Wt < 31 nm which was in the range of proposed

optimum nanospacing where the proteins can be adsorbed by the fluid. The mean

differences obtained by a one-way ANOVA test were significant at the level of 0.05

between Di of 40 ± 12 and 59 ± 17 nm. This result indicates the synergetic effects of a

variety of properties of the nanotubular surface as well as the important morphological

metric of the optimal nanospacing.

194

Fig. 7.11 The percentage of cells attached to the surface of different pore sizes of TiO2-

ZrO2-ZrTiO4 nanotubes after 7 days of cell seeding using MTS assay (ρ < 0.05)

The ratio of length to width (L/W) of cell nucleus and actin stress fibre on nanotubular

surface was measured by ImageJ software using confocal microscopy images after 7

days (Fig. 7.12). When the Di of TiO2-ZrO2-ZrTiO4 nanotubes increased from 18 to 30

and 40 nm, the L/W ratio of nuclei (~ 1.5) did not change considerably. It increased on

the nanotubular surface with Di = 59 nm then decreased to ~ 1.1 on Di = 64 and 82 nm.

The trend of the L/W ratios for the actin stress fibres was similar (~ 2.0) with the

maximum of 2.9 on Di = 59 nm. This result again confirms the before mentioned cell

behaviour against changing Di and roughness parameters of nanotubular layer.

0

20

40

60

80

100

120

140

18 30 40 59 64 82

Perc

enta

ge o

f cel

l adh

esio

n (%

)

Mean Di (nm)

Cell Adhesion

Total Cell

195


stress fibres (red) after 7 days on TiO2-ZrO2-ZrTiO4 nanotubes with a) Di = 18 nm, b)

Di =30 nm, c) Di =40 nm, d) Di = 59 nm, e) Di = 64 nm and f) Di = 82 nm

Figures 7.13 (a) and (b) show the SEM images of SaOS2 cells adhered on the surface of

annealed TiO2-ZrO2-ZrTiO4 nanotubes (Di = 40 nm) which were cleaned for two weeks

in 70 °C pure water while the pure water replaced three times per day. This cleaning

process was undertaken to minimise the remaining contamination of the electrolyte

during anodisation. The cells were flattened on the surface and they did not exhibit any

particular orientation. Fig. 7.14 shows the length of filopodia and the surface area of

spreading cells which were measured using SEM images by ImageJ software. On the

cleaner surface the filopodia did not elongated as much as those on the nanotubular

surface before cleaning after 1 day and 7 days. The surface area which occupied by cells

on the cleaner nanotubular surface was less than those on the nanotubular surface before

cleaning after 1 day and 7 days. The height of grown cells on cleaner surface was less

than the height of grown cells on the nanotubular surface before cleaning after 1 day

196

and 7 days. These result shows that the cleaner surface offered an acceptable condition

for cells to adhere and grow without much attempt to migrate to find better surface.

Fig. 7.13 SEM images of SaOS2 cells cultured for 24 h on the TiO2-ZrO2-ZrTiO4



magnification and d) with post-treated by HA - higher magnification

The nanotubular surface which was covered by HA after soaking in m-SBF for three

weeks revealed a convenient condition for SaOS2 cells to adhere. The length of

filopodia on HA surface was shorter than those on nanotubular surface without HA. The

cells did not expand on HA surface as much as they expand on nanotubular surface

before soaking in m-SBF. The grown cells on HA surface showed higher height than the

grown cells on the nanotubular surface without HA after 1 day and 7 days because of

the nature and height of HA. It can be concluded that changing the chemistry of the

197

surface served the cell to form better focal contact and decreasing the intention to

migrate.

198

Fig. 7.14 Illustration of a) the length of filopodia, b) surface area and c) height of

SaOS2 cells after 1 day and 7 days grown on different post-treated nanotubular surface

with the same Di

0

1

2

3

4

5

6

7

8

9

10

40 Cleaner Surface HA

Filo

podi

a le

ngth

(µm

)

Mean of Di (nm)

Length of Filopodia

1 Day

7 Days

a)

0

200

400

600

800

1000

1200


Surf

ace

Are

a (µ

m2 )

Mean of Di (nm)

Surface area of elangated cells

1 Day

7 Days

b)

0

1

2

3

4

5

6

7

8

9


Sp (µ

m)

Mean of Di (nm)

Heigth of Cells

1 Day

7 Days

c)

199

Fig. 7.15 illustrates the percentage of attached cells on nanotubular surface which was

cleaned in 70 °C pure water for two weeks to reduce the surface contamination and

nanotubular surface which was coated with HA by soaking in m-SBF in comparison to

nanotubular layer without these post treatment with same Di and the percentage of total

cells were obtained using MTS assay. The percentage of cells attached to nanotubes

with cleaner surface was 78.4 ± 7.7 % which was (almost two times) higher than the

nanotubular surface before cleaning (41.0 ± 2.6 %). The percentage of cells attached to

nanotubes covered with HA increased to 59.7 ± 13.6 %.

Fig. 7.15 The percentage of cells attached to the different post-treated nanotubular

surface of TiO2-ZrO2-ZrTiO4 with the same Di after 7 days of cell seeding using MTS

assay

It can be seen that the result obtained here was similar to the previous studies which

indicated that osteoblast proliferation was significantly greater on nanophase HA than

on conventional formulation of the same ceramic [270]. It has also been reported that

the growth of MG63 osteoblastic cells on electrodeposited CaP surface was more

significant than untreated titanium surface in a direct comparison [271]. Denser and

smaller crystalline size of HA on TiO2 nanotubes increased adhesion of MG63

0

20

40

60

80

100

120

140

40 Clean Surface HA

Perc

enta

ge o

f cel

l adh

esio

n (%

)

Mean Di (nm)

Cell Adhesion

Total Cell

200

osteoblastic cells due to increase in protein activity of such a layer [25]. Figures 7.16

and 7.17 show the SEM and confocal images of cells which were grown on the surface

of post-treated TiO2-ZrO2-ZrTiO4 nanotubes after 7 days, respectively.

Fig. 7.16 SEM images of SaOS2 cells cultured for 7 days on the TiO2-ZrO2-ZrTiO4



magnification and d) with post-treated by HA - higher magnification

201


stress fibres (red) after 7 days on TiO2-ZrO2-ZrTiO4 nanotubes a) with Di = 40 nm

before post-treating b) reduced contamination, c) with post-treated by HA

The ratio of length to width (L/W) of cell nucleus and actin stress fibre on nanotubular

surface was measured by ImageJ software using confocal microscopy images (Fig.

7.17) after 7 days. On cleaner surface of TiO2-ZrO2-ZrTiO4 nanotubes and covered with

HA, the L/W ratio of nuclei (~ 1.5) did not change considerably, however the L/W

ratios of the actin stress fibres increased to ~ 3.2 and ~ 2.2 on the cleaner nanotubular

surface and covered with HA, respectively. These results again interpret the before

mentioned behaviour of cells against changing of chemistry of nanotubular layer.


nanotubes fabricated in non-aqueous electrolyte

Figure 6.12 in Chapter 6 shows SEM images of top view of annealed TiO2-ZrO2-ZrTiO4

nanotubues fabricated in ethylene glycol with 5 to 10 wt % water (H2O) content and the

applied potential from 20 V to 30 V. As the nanotubular surface fabricated under non-

aqueous electrolyte conditions exhibited a brittle layer after annealing and it was not

suitable for further investigation, the cracked upper layer was removed by ultrasound

and the remaining dimpled surface used for cell culture. The mean sizes of Di, Do and

Wt of the dimpled surface of nanotubes is presented in Fig. 7.18. The Di, Do and Wt of

the surface with a dimpled surface increased when the water content of the organic

electrolyte increased from 5 to 10 wt %. Increasing the applied potential from 20 V to

202

30 V resulted in an increase in Di, Do and Wt for each electrolyte. The dimpled surface

of the nanotubular TiO2-ZrO2-ZrTiO4 layer was of larger nanoscale, revealing the bottle

shape of nanotubes that were larger at the bottom with a slightly thicker Wt.

Fig. 7.18 Mean nanotube size of Di, Do and Wt of dimpled surface after annealing and

removing the upper cracked layer fabricated in two different water contents and two

different applied potentials in ethylene glycol, Note: The nanotube size distribution

graphs were generated from 100 nanotubes on different positions for each of three

samples (300 measurements)

Figure 6.13 in Chapter 6 shows roughness parameters of Sa and Sq measured for the

annealed nanotubular layer with a dimpled surface. These amplitude parameters (Sa and

Sq) decreased with an increase of water content for each applied potential. Increasing

the applied potential for each composition of electrolyte resulted in increasing Sa and Sq.

It can be observed that the majority of the 4 dimpled surfaces, after removal of the

upper cracked layer of nanotubes, exhibited a uniform symmetry with normal

distribution in surface heights since Sskw ≈ 0 and Sku ≈ 3. Table 7.2 details the surface

area index (SI), the spacing parameters of Sm, Sc and Sv and surface energy of dimpled

surface of TiO2-ZrO2-ZrTiO4 nanotubes. The hydrophilic property of nanotubes also

0

20

40

60

80

100

120

140

5%-20 10%-20 5%-30 10%-30

Nan

otub

e si

ze (n

m)


Wt-Annealed-Dimple

Di-Annealed-Dimple

Do-Annealed-Dimple

203

increased with an increase of the nanotube sizes and roughness parameters that were the

result of increasing the water content and applied potential.

204

Table 7.2 Surface index, spacing parameters and calculated surface energy of TiO2-ZrO2-ZrTiO4 nanotubes with dimpled surface fabricated in

different water contents and applied potentials, Note: Each data point is an average of five measurements

Anodisation Condition (Mean size) SI Sm

(nm3/nm2)

Sc

(nm3/nm2)

Sv

(nm3/nm2)

Contact angle

()

sd

(mJm-2)

sp

(mJm-2)

s = sd + s

p

(mJm-2)

5 wt% H2O-20V (Di = 25 nm) 1.5 ± 0.1 62 ± 45 1074 ± 170 81 ± 16 9.0 ± 0.2 13.37 59.91 73.28

10 wt% H2O-20V (Di = 29 nm) 1.7 ± 0.0 33 ± 7 867 ± 62 67 ± 2 7.8 ± 0.4 15.16 57.77 72.93

5 wt% H2O-30V (Di = 45 nm) 1.7 ± 0.1 57 ± 28 1581 ± 682 110 ± 40 7.0 ± 0.7 13.85 59.64 73.49

10 wt% H2O-30V (Di = 49 nm) 1.7 ± 0.0 44 ± 6 1055 ± 95 111 ± 35 8.0 ± 0.2 15.20 57.69 72.89

205


non-aqueous electrolyte

The dimpled surfaces that were revealed after removing the upper cracked layer of

annealed TiO2-ZrO2-ZrTiO4 nanotubes were investigated for their induction of adherent

SaOS2 cells. Fig. 7.19 shows the formation of SaOS2 cell cytoplasmic extensions that

were widely spread with the extension of sheet-like protrusions, known as lamellipodia,

on all of the four dimpled nanotubular surfaces. The SaOS2 cells did not fabricate

finger-like protrusions of the plasma membrane, known as filopodia, to explore

neighboring spaces that might be more suitable locations for attachment and spreading.


sizes of TiO2-ZrO2-ZrTiO4 nanotubes fabricated at a) 5 wt % H2O - 20 V, b) 10 wt %

H2O - 20 V, c) 5 wt % H2O - 30 V, d) 10 wt % H2O - 30 V

206

The quantity and growth of cell spreading on the dimpled surfaces was quantitatively

assessed by application of ImageJ software. The ratio of length to width (L/W) of cell

nuclei and actin stress fibre (cytoplasmic protein) which is one of three types of filament

that physically joins the cell membrane and nuclear membrane and controls the

architecture and movement of cells [264, 272], was determined by using confocal

microscopy as well as the surface area of the actin stress fibre (SAS.F) after culture for 7

days. When the Di of the dimpled surface increased from 25 to 29 and 45 nm, the L/W

ratio of nuclei (~ 1.7) did not change considerably. The trend of the L/W ratios for the

actin stress fibres was similar (~ 3.1). The highest percentage of the surface area

measurements for the actin stress fibres were 646 < SAS.F. < 1230 µm2 on a dimpled

surface of Di = 25 nm. These values decreased to 562 < SAS.F. < 862 µm2 for Di =

29 nm. At Di = 45 nm for the dimpled surface, the highest percentage of the surface area

measurements of the actin stress fibre was 634 < SAS.F. < 948 µm2. This decreased to

330 < SAS.F. < 891 µm2 for the nanotubular surface with Di = 49 nm.

207


stress fibres (red) after 7 days on dimpled surface with a) Di = 25 nm, b) Di =29 nm, c)

Di =45 nm and Di =49 nm

Table 7.3 presents the percentage of attached cells and total cell numbers, as measured

by the MTS assay, on the four surface morphologies of different Di. The highest

percentage of attached cells was 84.9 ± 7.4 % on the nanotubes with mean Di of 25 nm

and the lowest one was 61.0 ± 7.1 % on the nanotubes with mean Di of 29 nm. The

percentage of cells attached to nanotubes with higher mean Di of 45 and 49 nm

increased to 61.5 ± 8.1 % and 80.3 ± 4.4 %, respectively; but these were still less than

that on the nanotubes with the lesser mean Di of 25 nm.

208

Table 7.3 The percentage of cells attached to the surface of different pore sizes and

roughness parameters of TiO2-ZrO2-ZrTiO4 nanotubes fabricated in organic electrolyte

after 7 days of cell seeding using MTS assay (ρ < 0.05)

Anodisation

conditions Cell adhesion (%) Di (nm) Wt (nm) Sa (µm) Sm (nm3/nm2)

5 wt%H2O-20V 84.9 ± 7.4 25 ± 5 27 ± 2 0.72 ± 0.12 62 ± 45

10 wt%H2O-20V 61.0 ± 7.1 29 ± 4 30 ± 2 0.55 ± 0.01 33 ± 7

5 wt% H2O-30V 61.5 ± 8.1 45 ± 6 24 ± 2 1.03 ± 0.28 57 ± 28

10 wt%H2O-30V 80.3 ± 4.4 49 ± 8 34 ± 3 0.73 ± 0.04 44 ± 6

209

The biological fluid functionalises the biomaterial implants in vivo, or the proteins

available in the culture media from the serum functionalise the biomaterial implants in

vitro, by providing protein and water that can be adsorbed onto the substrate. Therefore,

cells are now located in an environment that facilitates attachment since this is not

possible for the as-received, “naked” surface of biomaterial implant [15]. Extracellular

matrix (ECM), which is an intricate arrangement of glycoproteins, collagens,

proteoglycans and growth factors, has different roles. ECM acts as a physical scaffold

for cell attachment and the organisation of cellular structures. The ECM not only

provides a ligand for integrins expressed on the cell surface but also dictates the type of

integrin found within focal adhesions to which cells adhere [262, 273].

The most common biological molecules that can act as cell attachment molecules are

the integrins. These heterodimeric proteins are constituted of two subunits, α and β

[157]. There would be several phases for attachment when a cell studded with microvilli

perches on a surface. It has been reported that adhesion was not sufficient for survival;

that is, cells also need to spread on a surface to survive [274]. In a fraction of a second

the cell flattens due to non-specific electrostatic forces, such as van der Waals, and there

is the passive formation of ligand-receptor bonds. Then the cell aligns and the contact

area is widened [17, 259, 274]. Later, after several hours, the spreading phase proceeds

depending on the cell metabolism and the occurrence of clustering of integrins at the

anchoring sites. Focal adhesions are adhesion sites within 10 and 15 nm distance from

the substrate where they are formed by clustered receptors (integrins) using cytoplasmic

proteins [157, 264]. The head of integrins is approximately 8 - 12 nm. Thus, their

clustering has been proposed and proved to be enhanced when the optimum

nanospacing is less than 30 nm [19]. Therefore, they were initially clustered aptly on the

dimpled surface of 20 < Di < 30 nm; then on the dimpled surface of 31 < Wt < 37 nm,

due to an optimum nanospacing and similar roughnesses. However, the condition for

clustering of integrins on the dimpled surface of 25 < Di < 33 nm was not provided, as

might be expected due to optimal nanospacing, because of its low roughness spacing

parameter that could potentially disrupt the formation of focal adhesion. In the case of

the dimpled surface of 22 < Wt < 26 nm, the high roughness parameter of Sa = 1.03 µm

prevents formation of clusters.

210

These results agree with the fact that long-term adhesion is correlated to topography.

The hybrid roughness parameters, as measured in the current work, are especially

important since they affect the synthesis of ECM proteins by the cell themselves and the

cell-cell contacts. These phenomenological factors represent the strength of the cell-

material interface [264]. The mean differences obtained by a one-way ANOVA test

were significant at the 0.05 level between Di of 25 ± 5, 49 ± 8 nm and Di of 29 ± 4, 45 ±

6 nm. This result indicates the synergic effects of other properties of the dimpled

surface as well as the important morphological metric of the optimal nanospacing.

7.4 Conclusions

The fabrication of nanotubular layers on the surface of Ti50Zr under four different

electrochemical conditions was investigated. The influence of the characteristics of the

nanotubular surfaces on the cell behaviour of osteoblasts was evaluated. Conclusions

are as follows:

1. TiO2-ZrO2-ZrTiO4 nanotubes with six different distributions of inner

diameter (Di), outer diameter (Do) and wall thickness (Wt) were fabricated

via anodisation under different applied potentials in the aqueous electrolyte.

The annealed nanotubes exhibited a mean inner diameter (Di) of 18 ± 6, 30 ±

10, 40 ± 12, 59 ± 17, 64 ± 23 and 82 ± 26 nm when anodised at applied

potentials of 10, 15, 20, 25, 30 and 35 V, respectively.

2. The TiO2-ZrO2-ZrTiO4 nanotubes with Di = 59 ± 17 nm showed a roughness

with higher amplitude parameters of Sa and Sq, in comparison with

nanotubular surfaces that were fabricated at 10,15, 20, 35 and 35 V. These

parameters decreased when the applied potential increased to 30 and 35 V

because of the changes in the oxidation and dissolution rates.

3. The lowest density of SaOS2 cells occurred on the nanotubular surface with

59 ± 17 nm because of its high roughness amplitude parameters. Then next

lowest density evolved on the nanotubular surface with 64 ± 23 nm because

of its lowest surface energy and lowest roughness spacing parameters.

4. The highest density of SaOS2 cells was on the nanotubular surface with 18 ±

6 nm and the second highest density appeared on the nanotubular surface

with 30 ± 10 nm because of the optimum nanospacing.

211

5. SaOS2 cells exhibited the longest filopodia on the nanotubular surface with

Di = 40 ± 12 nm after culture for 1 and 7 days and the lowest height after

culture for 7 days. SaOS2 cells revealed the shortest filopodia and lowest

surface area and lowest height on the nanotubular surface with Di = 64 ±

23 nm after culture for 1 day and highest height of cell growth after culture

for 7 days.

6. The percentage of cells attached to nanotubes with cleaner nanotubular

surface and covered with HA increased to 78.4 ± 7.7 % and 59.7 ± 13.6 %

which was promising in comparison to the nanotubular surface before these

post-treatment (41.0 ± 2.6 %).

7. The upper nanotubular layer fabricated in ethylene glycol with 5 and 10 wt

% of water at 20 and 30 V after annealing was brittle and cracked. Therefore

it was removed to reveal a dimpled surface for bone cell adhesion study. The

nanotubes on the dimpled surface exhibited larger sizes of Di, Do and Wt than

the upper layer. The mean Di of nanotubes under the condition of 5 and 10

wt % of water anodised at 20 and 30 V were 25 ± 5, 29 ± 4, 45 ± 6 and 49 ±

8 nm, respectively.

8. The mean Sa and Sq of the dimpled surface decreased with an increase in

water content and increased with an increase in applied potential during

anodisation.

9. It was found that the optimal nanospacing was either Di = 25 ± 5 nm or Wt =

34 ± 3 nm for a maximum adhesion of bone cells (84.9 % and 80.3 %

respectively).

10. A lower percentage of cell attachment was observed on the dimpled surface

with the higher roughness of Sa = 1.03 µm (~ 61.5 %) and lower space

roughness of Sm = 33 nm3/nm2 (~ 61.0 %).

11. The surface area of actin stress fibres of SaOS2 cells was higher on optimal

nanospacing surfaces for the nanotubular surfaces with Di = 25 ± 5 nm.

212

Chapter 8 Nanoporous and nanotubular metal oxide layers on

biocompatible metals of Ta, Nb and Zr and their potential

applications

Abstract

Biocompatible metals such as titanium (Ti), zirconium (Zr), niobium (Nb) and tantalum

(Ta) that confer a stable oxide layer on their surfaces are commonly used as implant

materials or alloying elements for titanium-based implants due to their exceptional high

corrosion resistance and excellent biocompatibility. These metal oxide layers are highly

protective and stable and their surface properties can be tuned through creating a

nanoporous or a nanotubular layer with controlled topographical characteristics at the

nanometer scale using electrochemical surface modification. In this chapter, four kinds

of metal oxide nanostructures of tantala (Ta2O5), zirconia (ZrO2), titania (TiO2) and

niobia (Nb2O5) were fabricated via anodisation. The characteristics of these metal

oxides at nanoscale were characterised and their potential applications were discussed.

The tantala (Ta2O5) and the niobia (Nb2O5) exhibited a nanoporous structure and

zirconia (ZrO2) and titania (TiO2) a nanotubular structure. The nanosize distribution,

topography and both the physical and chemical properties of the nanolayers and their

bioactivities, as measured by the formation of hydroxyapatite (HA), were investigated

using scanning electron microscope combined with energy dispersive spectroscopy

(SEM-EDS), X-ray diffraction (XRD) analysis and 3D-Profilometry. The surface

energy of the nanoporous and nanotubular surfaces and its influence on the

hydroxyapatite formation in simulated body fluid (SBF) were evaluated. Observations

indicated that a Ta2O5 nanotube fabricated on the surface of tantalum were present for

several seconds prior to transforming to a nanoporous layer. The nanoporous Ta2O5

exhibited an irregular porous structure, high roughness and high surface energy as

compared to bare tantalum metal; and the highest bioactivity after annealing among the

four kinds of nanoporous structures. The amorphous nanoporous Nb2O5 showed a

uniform porous structure and low roughness, but no bioactivity before annealing.

Hydroxyapatite formation occurred only on the annealed nanoporous Nb2O5 with a

base-centred monoclinic structure. Fabrication of a uniform layer of ZrO2 nanotubes

required a higher applied potential as compared to that for TiO2 nanotubes with the

same inner diameters. Overall, the nanoporous and nanotubular layers of Ta2O5, Nb2O5,

213

ZrO2 and TiO2 imparted new functionality to the bare metal surfaces in the form of

enhanced bioactivity which is promising for their application as metal implants.

8.1 Introduction

Metals and alloys that are used as biomaterials also have a range of applications in

industry and medicine due to their excellent mechanical, physical and chemical

properties. Extensive studies have been carried out on titanium that has been coated by

its natural oxide layer formed in air. Other biocompatible metals and their respective

oxide layers are also of interest. For example, the protective oxide layer that forms

naturally on a tantalum surface or that is fabricated on the surface of other metals by

surface treatments such as chemical vapour deposition [275], electro deposition [123,

124] and sol gel [276] methods have other applications. Tantalum pentoxide, Ta2O5, is

used as a protective coating for chemical equipment due to its excellent corrosion

resistance [275]. A thin film of Ta2O5, formed by radio frequency (RF) sputtering of

SiO2-Ta2O5 films, has been used in optical devices such as transmission media for

integrated optical circuits [277]. Ta2O5 has demonstrated modest dielectric properties

with an ability to form a high quality thin film [278] for the miniaturisation of the

capacitive component of memory devices. The biocompatibility of the tantalum

pentoxide layer suggests tantalum as a good candidate for bioengineering implant

applications [116]. However, tantalum metal is more dense (ρTa = 16.6 g.cm-3) [114]

and more expensive than titanium (ρTi = 4.5 g.cm-3) [115]; therefore, researchers have

taken an alternative processing route to form a thin layer of tantalum onto the surface of

titanium and other metal implants. By anodisation a nanoporous Ta2O5 can be formed

on the metal in an acidic electrolyte containing HF [104, 117, 279]. The organic

electrolytes were also used for anodising a nanoporous oxide layer at an applied

potential of 10 to 40 V [121, 122].

The application of niobium pentoxide (Nb2O5), which can be produced via different

methods such as reactive sputtering [126], sol-gel processes [127], templating

techniques [128] and anodic oxidation [51, 129], has been reported in applications for

gas sensors [280], catalysts [281], both optical and electrochromic devices [126] and

biocompatible prostheses [127]. Niobium pentoxide (Nb2O5) is also a N type

semiconducting oxide with an electrical conductance in the presence of oxidising or

reducing gases that is reversible; thus, it has been investigated as an oxygen and

214

ammonia sensor material [280]. The niobium pentoxide is distinguished with

characteristics of displaying promoter and supporter effects in Fischer-Tropsch

synthesis, acidic and redox properties, and photosensitivity. The layered structure of

Nb2O5 can act as a catalyst under different chemical reactions due to its surface

properties and structure [281]. As well, the optical transmission of Nb2O5 changes

rapidly and reversibly from the near UV to near IR when H+ and Li+ ions are placed in

its layer lattice and, therefore, enable Nb2O5 to be used in solid state electrochemical

devices [282]. There are studies on the formation of a nanoporous layer of Nb2O5 by

controlling the effects of mixed electrolytes, applied potential and anodisation time.

These nanoporous layers displayed different range of pore sizes and thickness formed

mostly in an acidic electrolyte [51, 130, 283].

An important application for zirconium dioxide (ZrO2) is as an industrial catalyst,

especially with its use as an acid catalyst and for NOx reduction because it is more

stable under hydrothermal conditions in comparison to zeolites. This metal oxide also

exhibits large pores and a flexible mix in metal oxide composition [224, 225]. The

electrochemical deposition of a zirconium thin film on the surface of stainless steel has

been reported for these applications [284]. Zirconia, ZrO2, can have optoelectronic and

biomedical applications due to its high mechanical, chemical, and thermal stability

[226]. The investigation of optoelectronic applications of ZrO2 nanotubes has proceeded

via photoluminescence and cathodoluminescence measurements and the as-forme ZrO2

nanotubes were crystalline and were observed to demonstrate bright luminescence

[226]. The formation of self-organised sponge-like porous ZrO2 [53] and nanotube

oxide layer with different nanoscale sizes and thicknesses [132] has been described in

detail along with the effect of changing the condition of electrolyte, applied potential

and time of anodisation [102, 132-135].

The composition of the substantial mineral form of hard tissue such as bone and dentin

is apatite, which is secreted by bone cells on the implant surface during the process of

attachment. The bioactive materials which contained Si-OH, Ti-OH, Zr-OH, Nb-OH

and Ta-OH have been reported to exhibit the ability to induce apatite formation [37].

The effect of Ta-OH groups on the surface of alkaline treated bioactive tantalum metal

to form bone-like apatite was investigated and the Ca/P ratio of 1.59 was reported for

crystalline apatite after immersing in SBF [285]. The nanoporous niobium oxides which

215

prepared by sol-gel method and coated on 316LSS enhanced hydroxyapatite formation

[286]. Nanosheets and nanofibre-like surface formed on niobium surface by

hydrothermal alkaline treatment significantly influenced its apatite inducing ability

[287]. Amorphous nanoporous bioactive sodium niobate hydrogel layer formed on the

surface of niobium by alkaline treatment induced the deposition of a CaP layer during

soaking in SBF [288]. As-formed ZrO2 nanotubes have been reported to induce apatite

formation [289] and annealing improved its bioactivity [290]. Using pre-treatment such

as effective dipping treatment improved bioactivity of as-formed ZrO2 nanotube with

diameter 35 to 80 nm [291]. Although a few studies have investigated the bioactivity of

ZrO2 nanotubes, there is still a lack of research in this aspect for nanoporous Nb2O5 and

Ta2O5 which is the object of this chapter.

In this chapter the size of the pores of the as-formed and annealed Ta2O5, Nb2O5 and

ZrO2 nanoporous and nanotubular layer fabricated on the surfaces of tantalum, niobium

and zirconium and different electrolyte compositions of other studies is presented. Their

hydrophilic properties and surface energies are investigated and linked to changes of

their roughness and bioactivities. These properties are important for the biomedical

applications of these biocompatible metals. However there are insufficient studies in

this aspect, which is a prime motivation for the current work.


Tantalum (10×10×0.1 mm), niobium (10×10×0.05 mm), zirconium (10×10×0.05 mm)

and titanium (10×10×0.05 mm) foils (Baoji Boxin Metal Materials Co. Ltd., Shaanxi,

China) were degreased by sonification in methanol, isopropanol, acetone and ethanol

for 15 min each progressively. The substrates were then washed with deionised water

and dried using a nitrogen gas stream. A two-electrode configuration was employed

using a DC power supply with a 1 cm2 platinum plate acting as the counter electrode

and placed 4 cm from the working electrode. Electrochemical experiments were

performed at room temperature with the electrolyte composed of 1 M H2SO4 +

3.3 wt % NH4F for tantalum and niobium at 20 V for 120 min and 16 min, respectively;

1 M (NH4)2SO4 + 0.3 wt % NH4F (with the addition of H2SO4 to attain pH 5) for

zirconium at 30 V for 95 min and 1M H2SO4 + 0.5 wt % NH4F for titanium at 10, 15

and 20V for 120 min. The electrolytes were prepared from reagent grade chemicals

(Sigma Aldridge) and pure water. After the electrochemical treatment the samples were

216

rinsed with deionised water for 5 min and dried with a nitrogen stream. Annealing of

the samples was performed in air at 290 °C for 10 min in a conventional muffle furnace

(Nabertherm LT15/13/P330; Nabertherm GmbH, Lilienthal, Germany).

Microstructural characterisation of the samples was carried out using a field-emission

scanning electron microscope (FESEM, ZEISS SUPRA 40 VP). Phase characterisation

was performed by means of X-ray Diffraction (XRD, Bruker D8; Bruker Pte Ltd,

Singapore) using Cu Kα incident radiation at 40 kV and 40 mA with a 2θ scanning step

size of 0.02 ° from 10 to 90 °.

Roughness parameters were measured using a 3D-Profilometer (Bruker, Contour GT-

K1; Bruker Pte Ltd, Singapore) and analysed using the SurfVision software (Veeco

Instruments Inc.; Plainview, NY, USA). Mean Roughness (Sa) and Root Mean Square

(RMS) Roughness (Sq) were studied along with some other amplitude parameters

(skewness (Sskw) and kurtosis (Sku)). These parameters were described in Chapter 3.

The water contact angle was measured using a goniometer (NRL C.A. Goniometer,

Ramé-hart, Inc.; Succasunna, NJ-USA). Surface energy was calculated based on

Owens-Wendt (OW) method [211], as described in Chapter 3.

Bioactivity assessments were carried out by soaking the as-formed and annealed metal

oxides samples in simulated body fluid (SBF) and incubating at 37 ˚C for up to 3 weeks.

A modified (m-SBF) [213] with an ion composition nearly equal to blood plasma was

prepared by dissolving 5.403 g NaCl, 0.504 g NaHCO3, 0.426 g Na2CO3, 0.225 g KCl,

0.230 g K2HPO4 - 3H2O, 0.311 g MgCl2 - 6H2O, 0.293 g CaCl2, and 0.072 g Na2SO4 in

pure water as previously reported [213]. The m-SBF was buffered at pH 7.4 at 37 ˚C

using 2-(4-(2-hydroxyethyl)-1-piperazinyl) ethane sulfonic acid (aka HEPES) and 1 M

NaOH. An amount of 17.892 g HEPES has dissolved in 100 ml 0.2 M NaOH. The

ability of the pretreated samples to form apatite was evaluated in a static SBF

environment. The samples were removed after incubation for 3 weeks in m-SBF; then

rinsed with deionised water and dried at room temperature for 24 h. The atomic

percentage of calcium to phosphate ratio is calculated using EDS results.

The nano size distribution measurements were generated from 100 nano-pores and

nanotubes at different positions for each sample. An average of five readings per sample

217

was acquired for the roughness parameters, water contact and surface energy

measurements.


8.3.1 Process conditions and formation mechanism of nanoporous/nanotubular

metal oxides

The application of a suitable voltage to a metal surface produces a Mn+ ion that can

dissolve in an electrolyte or form an insoluble metal oxide layer by reacting with

existing O2- and forming an oxide layer on the metal surface. However, partial

dissolution is also possible due to the composition and condition of the electrolyte [11]

and this processing conditions needs to be controlled since this influences the formation

of the nanolayers of metal oxides.

The oxidation reaction at the interface of the metal surface and electrolyte at a constant

applied potential is given by:

M Mn+ + ne- (M = Ta, Nb, Zr and Ti) (8-1)

As well:

xH2O xO2- + 2xH+ (8-2)

Then,

2Mn+ + nO2- M2On (8-3)

The reduction reaction at the surface of counter electrode (normally platinum) in this

condition is given by:

2nH+ + 2ne- nH2 (8-4)

In the presence of a fluorine anion, chemical dissolution occurs, i.e.:

Ta5+ + 7F-[TaF7]2- (8-5)

Nb5+ + 7F-[NbF7]2- (8-6)

Zr4+ + 6F-[ZrF6]2- (8-7)

Ti4+ + 6F- [TiF6]2- (8-8)

218

Nanoporous Ta2O5 can be formed by anodisation [119, 279] and under specific

conditions, such as high viscosity of the concentrated acid, nanotubes can be grown in 5

- 20 s [120]. The diameter of the pores of Ta2O5 were 2 to 35 nm in cases where the

oxide layer was 350 - 400 nm thick [104, 117, 279], although the height of the

nanoporous layer has been reported to be up to 16 µm [121, 122]. A porous Nb2O5 layer

with pore diameter of 20 - 30 nm and layer thickness of ~ 500 nm [51] can be fabricated

on the surface of niobium by anodisation in the presence of F- ions. This layer system

has been investigated for use as a capacitor [292]. Nanotubular ZrO2 can be obtained by

anodic oxidation of zirconium in an electrolyte consisting of the fluorine anion by

means of a process that is similar to other valve metals [53]. The anodising conditions

can be improved by employing a buffered electrolyte to obtain ZrO2 nanotubes of ~

50 nm diameter and layer thickness of ~ 17 µm [102, 132].

8.3.2 The dynamics of the anodisation process for tantala (Ta2O5), niobia (Nb2O5)

and zirconia (ZrO2)

Figure 8.1(a) shows a top view of a nanoporous Ta2O5 that exhibits irregularly

distributed pores with a diameter range of 35 - 65 nm. The thickness of this layer was

1.24 µm, taking note that the sub-structure revealed multiple layers (Fig. 8.1(b)), as

reported previously [122], where every layer exhibited a closed bottom. Fig. 8.1(c)

shows the cross section of the multiple layers and the arrow indicates Ta2O5 nanotubes

with a length of nearly 500 nm that formed initially after several seconds during

anodisation. These long nanotubes transformed into nanoporosity after several seconds

during anodisation and exhibited similar morphological features to other layers. A top

view of the multiple layers can be observed in Fig. 8.1(d).

219

Fig. 8.1 SEM images of nanoporous Ta2O5: (a) top view (b) the bottom view and the

cross section of multilayer, (c) shorted life nanotubes (arrow), (d) the top view of

multilayer formed on a Ta after anodisation for 120 min in 1M H2SO4 + 3.3 wt %

NH4F, 20V, SEM image of nanoporous Nb2O5 (e) top view and (f) cross section formed

on niobium after anodisation for 16 min in 1M H2SO4 + 3.3 wt % NH4F, 20V and SEM

image of ZrO2 nanotubes (g) top view, (h) cross section and (i) bottom view formed on

a zirconium after anodisation for 95 min in 1M (NH4)2SO4 + 0.3 wt % NH4F, pH=5 and

20V

When anodisation starts, a compact oxide layer forms on the surface of tantalum for a

finite thickness according to the decay characterisitics of the current. Tantalum ions,

Ta5+, arrive at the interface of the oxide layer and electrolyte and become soluble in the

electrolyte by forming [TaF7]2-. Thus, pores formed at the interface between the oxide

220

layer and electrolyte. These pores continued to develop during anodisation because they

preferred sites that trap the necessary ions. Then, the growth rate of the metal oxide

nanoporous and nanotubes may be determined by the diffusion rate of F- ions and the

soluble metal complex [58].

Although the formation mechanism for the nanoporosity of tantalum and other valve

metals is uncertain, the nanoporosity could be related to the molecular dimension of

[TaF7]2- and the formation kinetics of Ta2O5. The standard electrode potential of

Ta→Ta5+ is 0.75 V, which is less than that of Ti→Ti4+, 2.132 V [236]. Thus Ta5+

oxidises quickly. In addition, the standard enthalpies of formation for Ta2O5 and TiO2

are -492.790 and -228.360 gram calories per mole, respectively [237]. It can be

concluded that the oxidisation process and the formation of a Ta2O5 layer on tantalum

are faster than the processes on the titanium surface.

The mechanism and kinetics for the oxide layer formation can be calculated on the basis

of the ionic radii for Ta5+ and F- on a first principles basis. In this instance, the physical

and chemical environment around the [TaF7]2- at the pore sites, which affect the

molecular size of [TaF7]2-, are ignored. The structure of the complex has been reported

to be a two pentagonal pyramid. The volume of the circumscribed sphere of this

structure can be calculated as 26.094×10-30 m3, which is smaller than the octahedral

volume of [TiF6]2- = 29.182×10-30 m3. It can be expected that the diffusion of [TaF7]2- is

faster than [TiF6]2- and that the pore growth does not only occur solely at the bottom,

but also at the walls, thereby resulting in pores that merge and agglomerate. Over a

period of several seconds the pore was bottle-shaped similar to that of nanotubes with a

closed bottom where the diameter of tubes near the bottom was greater than that at the

tube necks. Gaps are also created at the bottom of the layer when the pores merged,

which permits the electrolyte to become available underneath the nanotubes. Thus, a

new layer of nanotubes can form under the same mechanism, thereby resulting in the

formation of multilayers of nanoporous layer over time. The nanotubes dissolve

according to the dissolution rate of the oxide layer in the presence of the fluorine anions

(F-).

The nanoporous Nb2O5 (as shown in Fig. 8.1(e)) exhibited nearly uniform pore

diameters of 22 - 42 nm. The thickness of this layer was ~ 235 nm (Fig. 8.1(f)). The

same formation mechanism can be hypothesised for nanoporous Nb2O5 layer formation

221

as has been described for the formation of the nanoporous Ta2O5 layer. Niobium (Nb)

and Zr are in the same period and adjacent columns in periodic table of the elements.

The standard electrode potential of Nb→Nb5+ is 0.644 V, which is less than that of

Zr→Zr4+, 1.553 V [236] and Nb is, therefore, oxidised rapidly. In addition, the standard

enthalpy of formation for Nb2O5 is -458.640 gram calories per mole; whereas the

enthalpy of formation is -264.199 gram calories per mole for ZrO2 [237].

The processes of oxidisation and formation of the oxide layer on the niobium surface

are rapider than those on the zirconium surface. Consider, for example, that the physical

and chemical environment around the [NbF7]2- ion within the pore structure are ignored.

The volume of the ion complexes can then be calculated on the basis of the ionic radii

for Nb5+ and F- ions and assuming a lattice structure of a two pentagonal pyramid. The

volume of the circumscribed sphere of the [NbF7]2- ion is calculated as 32.02×10-30 m3,

which is smaller than the volume of the octahedral [ZrF6]2- ion which is 36.08×10-30 m3.

It can be expected, therefore, that the diffusion of a [NbF7]2- ion within a pore is rapider

than for a [ZrF6]2- ion. The growth of the pore does not arise only at the bottom but also

at the walls, which will result in the merging of pores.

Figure 8.1(g-i) show the fabricated ZrO2 nanotubes that exhibit (i) uniform inner pore

diameters of 21 - 35 nm, (ii) an outer diameter of 54 - 68 nm, and (iii) wall thickness of

10 - 16 nm. The thickness of this layer was about 24.4 µm. As previously explained

[240] the standard electrode potential of Zr→Zr4+ is 1.553 V is lower than that of

Ti→Ti4+, 2.132 V [236] which results in rapid oxidation. In addition, the standard

enthalpy of formation for ZrO2 is -264.199 gram calories per mole whereas that for

TiO2 is -228.360 gram calories per mole [237]. The process of oxidation and the

formation of the oxide layer on the surface of zirconium are faster than those for the

titanium surface. However, the formation of the [ZrF6]2- ion takes more time than for the

[TiF6]2- ion because the solubility equilibrium constant of (NH4)2ZrF6 is higher than that

of (NH4)2TiF6. Based on the hypothesis described previously, the volume of the

circumscribed sphere of the octahedral structure of [ZrF6]2- can be calculated as

36.08×10-30 m3 and 29.18×10-30 m3 for [TiF6]2- which causes a lower diffusion rate

inside the pore. It has been reported that nanotubes grow longer with a smaller diameter

when the dissolution rate is slower compared to the oxidation process [249].

222

8.3.3 Physical characteristics of nanoporous Ta2O5 and Nb2O5 and nanotubular

ZrO2 layers

Figure 8.2(a) demonstrates the influence of the formation of nanoporous Ta2O5 and the

TiO2 nanotubes on the Sa with respect to the bare metals. The nanoporous layer of

Ta2O5 revealed a lower surface roughness than the TiO2 nanotubes with almost the same

distribution of inner diameter (Di) when the concentration of fluoride ion (CF- hereafter)

was 0.5 wt % at 20 V. The roughness similarity arose because the TiO2 nanotubes were

separated from each other whereas the nanoporous Ta2O5 formed a continuous layer.

The surface area index (SI) may be calculated by dividing the projected surface area,

viz. the total exposed three-dimensional surface area being analysed, including peaks

and valleys, to the surface area measured in the lateral direction. The volume index (VI)

may be calculated by dividing the natural volume, viz. the amount of liquid that it

would take to submerge the dataset to its highest point, to the normal volume that is

measured in the lateral of nanoporous layer. The SI and VI values of nanoporous Ta2O5

are shown in Table 8.1 with the corresponding roughness amplitude parameters and

verify that the Ta2O5 nanoporous layer did not change significantly the surface area and

roughness of the bare metal. Although the distribution of nanoporous Ta2O5 was not

uniform, there were no high peaks and low valleys according to the measured Sskw

which is close to zero. The peaks and valleys exhibited a platykurtoic distribution that

was almost uniform due to the Sku being near to 3.

223

Fig. 8.2 Demonstration of (a,c,e) changes of roughness (Sa) and (b,d,f) water contact

angle of bare tantalum, niobium, zirconium and titanium, after fabrication of

nanotubular and nanoporous layer, and subsequently annealing, respectively, Note:


224

Table 8.1 Surface area and volume index and roughness amplitude parameters of nanoporous Ta2O5, Nb2O5 and nanotube ZrO2, Note: Each

data point is an average of five measurements

Surface area Index

(SI)

Volume Index

(VI) Sa () Sq () Sskw Sku

Nanoporous Ta2O5 -as-formed 1.34 ± 0.01 18393 ± 86 0.27 ± 0.01 0.35 ± 0.01 0.60 ± 0.23 4.82 ± 0.52

Nanoporous Ta2O5 -Annealed (10 min at 290°C) 1.27 ± 0.01 1838 ± 132 0.25 ± 0.01 0.32 ± 0.01 0.72 ± 0.03 3.90 ± 0.17

Bare tantalum foil (0.1 mm) 1.06 ± 0.00 18229 ± 0 0.23 ± 0.00 0.29 ± 0.00 0.06 ± 0.00 2.79 ± 0.00

Nanoporous Nb2O5 -as-formed 1.50 ± 0.00 18398 ± 134 0.26 ± 0.00 0.34 ± 0.00 0.64 ± 0.03 3.78 ± 0.21

Nanoporous Nb2O5 -annealed (10 min at 290°C) 1.38 ± 0.00 18357 ± 144 0.26 ± 0.01 0.33 ± 0.01 0.39 ± 0.02 3.28 ± 0.05

Bare niobium foil (0.05 mm) 1.04 ± 0.00 18327 ± 0 0.31 ± 0.00 0.38 ± 0.00 0.22 ± 0.00 2.68 ± 0.00

ZrO2 Nanotube--as-formed 2.44 ± 0.30 18422 ± 12 0.56 ± 0.05 0.71 ± 0.05 -0.86 ± 0.018 3.75 ± 0.49

ZrO2 Nanotube-annealed (10 min at 290°C) 2.94 ± 0.15 18428 ± 79 0.61 ± 0.01 0.76 ± 0.00 -0.57 ± 0.20 3.30 ± 0.35

Bare zirconium foil (0.05 mm) 1.03 ± 0.00 18531 ± 0.00 0.12 ± 0.00 0.20 ± 0.00 -4.27 ± 0.00 39.31 ± 0.00

225

Figure 8.2(b) shows the change of hydrophilic properties of bare tantalum and titanium

after fabrication of the nanoporous and nanotubular layers on their surfaces. The water

contact angle measurement of a surface represents the wetting properties of the surface.

In the literature, a superhydrophobic surface is defined when the water contact angle is

more than 150˚ and hydrophobic when it is 90˚ < θw < 150˚ [241]. A surface has

superhydrophilic properties when θw < 10˚ and has hydrophilic properties with a water

contact angle 10˚ < θw < 90˚ [241]. The as-formed nanoporous Ta2O5 exhibited

hydrophilic properties that were similar to the TiO2 nanotubes with the same

distribution of inner diameters, Di (under the conditions of CF- = 0.5 wt %, 20 V). Fig.

8.2(a) and (b) also suggest a direct link between roughness and water contact angle.

When the roughness increased there was a decrease in the water contact angle. The

calculated surface energy for nanoporous Ta2O5 increased after anodisation and also

after annealing due to the crystallisation. The water contact angle and surface energy

have been detailed in Table 8.2 for nanoporous Ta2O5.

226

Table 8.2 Water contact angle and surface energy of as-formed and annealed nanoporous Ta2O5, Nb2O5 and nanotube ZrO2 in comparison to its

bare metal, Note: Each data point is an average of five measurements

Contact Angle

(˚)

sd

mJm-2

sp

mJm-2

s = sd + s

p

mJm-2

Nanoporous Ta2O5 -as-formed 26.0 ± 5.0 11.8 55.2 67.0

Nanoporous Ta2O5 -annealed (10 min at 290°C) 12.1 ± 1.6 15.2 56.7 71.9

Bare tantalum foil (0.1 mm) 47.9 ± 1.2 5.1 49.5 54.6

Nanoporous Nb2O5 -as-formed 25.0 ± 7.0 11.02 56.94 67.96

Nanoporous Nb2O5 -annealed (10 min at 290°C) 15.5 ± 1.4 18.11 52.20 70.31

Bare niobium foil (0.05 mm) 61.0 ± 0.0 6.44 34.94 41.38

ZrO2 Nanotube-as-formed 17.0 ± 1.0 11.56 60.04 71.60

ZrO2 Nanotube-annealed (10 min at 290°C) 8.5 ± 1.0 12.84 60.77 73.61

Bare zirconium foil (0.05 mm) 58.7 ± 1.3 5.82 38.042 43.87

227

Figure 8.2(c) indicates the increments in roughness for the nanoporous Nb2O5 layer and

nanotubes while exhibiting almost the same distribution of Di (under the conditions of

CF- = 0.5 wt %, 15 V) with respect to the corresponding bare metals. The roughness of

the titanium increased upon the creation and separation of TiO2 nanotubes that give rise

to gaps in the surface architecture, whereas the roughness of the niobium decreased

upon the growth of nanoporous Nb2O5. The Sskw of the Nb2O5 nanoporous surface is

higher than zero, indicating the presence of high peaks and low valleys with a

leptokurtoic distribution. The surface area and volume indices of nanoporous Nb2O5 are

shown in Table 8.1, as well as the corresponding roughness amplitude parameters. Fig.

8.2(d) shows the changes of hydrophilic properties of the bare niobium and titanium

after fabrication of the nanoporous and nanotubular layers on their surfaces. The

nanoporous surface revealed a water contact angle in the range of almost the same

distribution of Di (under the conditions of CF- = 0.5 wt %, 15 V) for TiO2 nanotubes that

showed hydrophilic properties. This data also implies that there is a link between

roughness and the water contact angle. The surface energy of this layer was higher than

bare niobium. The water contact angle and surface energy are listed in Table 8.2 for the

nanoporous Nb2O5.

Figure 8.2(e) shows the effect of ZrO2 and TiO2 nanotube formation on the roughness,

Sa, of the bare metals. The presence of the nanotubular ZrO2 has increased the

roughness of the bare metal and also revealed a higher roughness than the nanotubular

layer on the TiO2 nanotubes surface with almost the same distribution of Di (under the

conditions of CF- = 0.5 wt %, 10 V). The Sskw of the ZrO2 nanotubular layer is lower

than zero, which signifies deep valleys such as scratches with a leptokurtoic

distribution. The surface area and volume indices of ZrO2 nanotube is shown in Table

8.1 along with their roughness amplitude parameters.

228

The change of the hydrophilic properties of bare zirconium and titanium after

fabrication of naotubular layer on their surfaces is shown in Fig. 8.2(f). Although both

nanotubular layers fabricated on the zirconium and titanium surfaces are hydrophilic,

the ZrO2 nanotubes exhibited a lower water contact angle in comparison to TiO2 with

nearly the same distribution of Di (CF- = 0.5 wt %, 10 V) which led to a high surface

energy. The water contact angle and surface energy is detailed in Table 8.2 for ZrO2

nanotubes. A direct link between roughness and the water contact angle can be observed

in this figure.

Nanoporous Ta2O5 with a mixture of amorphous and hexagonal phase was fabricated on

tantalum via anodisation using an electrolyte of 1 M H2SO4 + 3.3 wt % NH4F at an

applied potential of 20 V for 120 min. The amorphous nanoporous Ta2O5 transformed

into hexagonal Ta2O5 after annealing at 290 ºC for 10 min, as indicated by the XRD

patterns in Fig. 8.3(a). After annealing the pore size decreased to the range of 23 -

49 nm as a result of crystallisation. Annealing increased the hydrophilic properties of

the nanoporous Ta2O5 layer as much as TiO2 nanotubes because of their different

porosity due to their structures.

229

Fig. 8.3 XRD patterns of (a) the nanoporous Ta2O5 and bare tantalum foil, (b) the

nanoporous Nb2O5 and bare niobium foil, (c) the ZrO2 nanotubes and bare zirconium

foil and (d) the TiO2 nanotubes and bare titanium foil

Nanoporous Nb2O5 with a mixture of amorphous and monoclinic phase was fabricated

on niobium via anodisation using an electrolyte of 1 M H2SO4 + 3.3 wt % NH4F at an

applied potential of 20 V for 16 min. The amorphous nanoporous Nb2O5 transformed

into base-centred monoclinic Nb2O5 after annealing at 290 ºC for 10 min, as indicated

by the XRD patterns in Fig. 8.3(b). The pore size, after annealing, was irregular and laid

in the range of 26 - 60 nm. The pore size was affected by the crystallisation process

during the heat treatment. Annealing did not influence the roughness of the nanoporous

Nb2O5 but decreased the water contact angle of the layer.

Cubic ZrO2 nanotubes were fabricated on zirconium via anodisation using an electrolyte

of 1 M (NH4)2SO4 + 0.3 wt % NH4F (with the addition of H2SO4 to attain a pH=5) at an

230

applied potential of 30 V for 95 min, as indicated by the XRD patterns in Fig. 8.3(c).

After annealing, the inner diameter of the nanotubes laid in the range of 20 - 36 nm,

whilst the nanotube outer diameter (Do) was 48 - 68 nm and the wall thickness (Wt) was

in the range of 9 - 11 nm. The roughness (Sa) and hydrophilic properties of ZrO2

nanotubes increased after annealing. Fig. 8.4(a) shows the pore sizes of nanoporous

Ta2O5 and Nb2O5, compared to TiO2 nanotube with nearly the same Di. Fig. 8.4(b), (c)

and (d) show the Di, Do and Wt of ZrO2 nanotubes compared to TiO2 nanotube with

nearly the same pore sizes as a visual illustration.

Fig. 8.4 Illustration of pore size of the nanoporous and nanotubular layers

8.3.4 Bioactivity of nanoporous and nanotubular metal oxide layers

Bioinert implant materials upon implantation in vivo are encapsulated by fibrous tissues

that isolate them from the surrounding bone. This is not the ideal healing mechanism,

whereas a bioactive material is preferred which bonds to living bone by forming a

carbonated apatite layer on their surfaces similar to hydroxyapatite (HA) of bone

231

composition [56]. The bioactivity of the biocompatible metals with the as-formed and

annealed nanoporous and nanotubular metal oxides was assessed by immersion in the

m-SBF for a period up to 3 weeks. The response of the oxide layers to the SBF

immersion were observed after 1 day, 1 week, 2 weeks and 3 weeks. After 1 day

immersion, no growth of HA was perceived on the surfaces of the as-formed and

annealed nanoporous Ta2O5 and Nb2O5 and the nanotubular TiO2 and ZrO2. However

after 3 weeks immersion as shown in Fig. 8.5, 8.6 and 8.7 HA were deposited onto the

surfaces of the nanoporous and nanotubular layers.

Fig. 8.5 (a) low and (b) high magnifications of HA on as-formed nanoporous Ta2O5; (c)

low and (d) high magnifications of HA on annealed nanoporous Ta2O5

232

Fig. 8.6 SEM images of HA on a) as-formed and b) annealed nanoporous Nb2O5

Fig. 8.7 (a) low and (b) high magnifications of HA on as-formed nanotubular ZrO2; (c,)

low and (d) high magnifications of HA on annealed nanotubular ZrO2

233

The thickness of the HA layers mineralised on the surface of nanoporous Ta2O5,

nanoporous Nb2O5, nanotubular ZrO2 and TiO2 were 6.9, 3.7, 3.3 and 3.8 µm

respectively as shown in Fig. 8.8.


nanoporous Ta2O5, b) nanoporous Nb2O5, c) nanotubular ZrO2 and d) nanotubular TiO2

The atomic ratio of calcium to phosphate calculated using EDS results after 3 weeks is

indicated in Fig. 8.9.

234

Fig. 8.9 Bioactivity of biocompatible nanoporous and nanotubular oxide metals after 3

weeks in m-SBF at 37 ˚C

The M-OH groups located on the surface of biocompatible metal oxides are favoured

sites for apatite nucleation [56, 251, 254, 257]. First, Ca2+ ions are absorbed onto the

hydrolysed nanoporous and nanotubular oxide surface by Coulomb attraction forces.

Then, existing phosphate groups inside the SBF are adsorbed to the positively charged

surface, resulting in the formation of calcium phosphate. The stoichiometric Ca/P (at %)

of octacalcium phosphate [Ca8H2(PO4)6×5H2O], tricalcium phosphate [Ca3(PO4)2] and

hydroxyapatite [Ca10(PO4)6(OH)2] are 1.33, 1.5 and 1.67, respectively. Fig. 8.10 shows

local pH changes up to 3 weeks of mineralisation of HA. When negative hydrolysed

nanotubular surface formed the pH value increased. When the absorption of Ca2+ ions

was accomplished the changes of pH value were negligible. The exchange of OH- with

PO43- as a result of their absorption increased the pH value. By consuming the OH- to

form hydroxyapatite [Ca10(PO4)6(OH)2] the pH value decreased and remaining constant

after the HA formation completed.

235

Fig. 8.10 pH value of static m-SBF as a function of soaking time for: a) nanoporous

Ta2O5 and Nb2O5 and b) nanotubular TiO2 and ZrO2

According to the obtained Ca/P ratio for the metal oxides, annealed nanoporous Ta2O5

had a high value of Ca/P ratio which may indicate the presence of a mixture of calcium

and phosphate abovementioned compositions. Other nanoporous and nanotubular metal

oxide layers induce calcium phosphate and need more time or pre-treatment to induce

crystalline hydroxyapatite. Ta2O5 reaches an isoelectric point at a pH of 2.7 - 3.0, which

can be compared to Nb2O5 that attains an isoelectric point at a pH of 4.0 [255, 256].

Thus, the Ta2O5 surface, compared to the Nb2O5 surface, becomes more negatively

charged in SBF at a pH of 7.4. Therefore there is an appropriate chemical environment

for a higher Ca/P atomic ratio to be attained for Ta2O5 in comparison to Nb2O5. In

addition, the isoelectric point of TiO2 has been reported at a pH of 3.9 and the

corresponding value for ZrO2 is 5.5 [255, 256], which would be expected to lead to a

higher Ca/P ratio for TiO2.

8.4 Conclusions

Nanoporous oxide layers on the surfaces of tantalum and niobium and nanotubular

layers on the surfaces of zirconium and titanium were fabricated via anodisation. The

bioactivity of the nanoporous and nanotubular layers were evaluated. The prime

conclusions are as follows.

1. The pore size distributions of the nanoporous tantala (Ta2O5) were not uniform

in comparison to TiO2 nanotubes. The pore size distribution of the nanoporous

niobia (Nb2O5) was almost uniform in comparison to the TiO2 nanotubes. Higher

236

applied potential was needed in fabricating ZrO2 nanotubes with the same

distribution of inner diameter than TiO2 nanotubes during anodisation.

2. The nanoporous Ta2O5 layer and nanotubular ZrO2 and TiO2 layer exhibited a

higher roughness than their respective bare metals but the nanoporous Nb2O5

layer exhibited a lower roughness than its bare metal.

3. The nanoporous layers of Ta2O5 and Nb2O5 and the nanotubular layers of ZrO2

and TiO2 revealed an increase in hydrophilic property and surface energy

compared to their respective bare metals.

4. The hydrophilic property and the surface energy of the nanoporous layers of

Ta2O5 and Nb2O5 and the nanotubular layers of ZrO2 and TiO2 were increased

after annealing.

5. After annealing the pore size of nanoporous Ta2O5 decreased whereas the pore

size of nanoporous Nb2O5 increased. This is due to their different crystalline

phases before and after annealing which possess different lattice parameters. The

inner diameter of ZrO2 nanotubes after annealing did not show any obvious

change because the crystalline phase did not changed but annealing led to an

increase in the inner diameter of anatase TiO2.

6. As-formed Ta2O5 and ZrO2 exhibited a good bioactivity similar to TiO2.

Annealed Ta2O5 showed a high level of bioactivity which is promising for

biomedical applications. The as-formed amorphous nanoporous Nb2O5 did not

show bioactivity; but the annealed nanoporous Nb2O5 with a base-centred

monoclinic structure was bioactive.

237

Chapter 9 Conclusions

9.1 Introduction

The motivation and principal objectives of this thesis were the investigation of the

bioactivity and biocompatibility of Ti50Zr alloy with a nanotubular surface fabricated

via anodisation. The nanotubular layer which is the result of electrochemical surface

modification played an important role in endowing new surface functionalities of metal

implants. The nanotubular layer fabricated in this study revealed an interesting

morphology. This morphology which was a reflection of the needle-like microstructure

of the alloy exhibited a distribution of inner diameter (Di), outer diameter (Do) and wall

thickness (Wt) of nanotubes and a range of surface roughness parameters due to

different heights of nanotubes. This attractive structure needed to be considered while

their bioactivity and biocompatibility were investigated.

9.2 Major findings

Initially two different morphologies of TiO2-ZrO2-ZrTiO4 nanotubes were fabricated in

two different types of electrolyte; aqueous and non-aqueous electrolyte. In Chapter 4,

firstly the effect of fluorine (F-) anion concentration (CF-) in the aqueous electrolyte on

the nanotube parameters at constant applied potential and anodisation time was

investigated. An increase in the CF- from 0.3 to 0.5 wt % led to an increase in the

number of nanotubes with smaller Di, Do and Wt. Because the chemical and

electrochemical reaction rate of Zr was higher than Ti as a result Zr reacted faster than

Ti. In a same condition the nanotube size of ZrO2 was smaller than TiO2 and the

nanotube length was longer. The mean Di and Do of TiO2 nanotubes on the surface of

CP-Ti increased specifically when CF- increased. Then the effects of the applied

potential on the nanotube parameters at constant F- ion concentration in electrolyte and

the anodisation time were investigated. Conclusions are as follows:

1. Increasing the applied potential from 5 to 35 V resulted in boarder Di and Do. It

means that the number of nanotubes with smaller Di and Do decreased or the

number of nanotubes with larger Di and Do increased due to the increase in the

anodic current. The Wt of nanotubes increased slightly with an increase in the

applied potential from 5 to 35 V.

238

2. Increasing applied potential from 5 to 35 V resulted in an increase in mean Di

and Do of TiO2 nanotubes on the surface of CP-Ti. The extent of increasing the

mean Wt was not as much as the mean Di and Do.

3. At lower applied potential (5 to 20 V) the mean Di and Do of TiO2-ZrO2-ZrTiO4

nanotubes were lower than those of TiO2 nanotubes in contrast to higher applied

potential.

The effect of water content of 5 and 10 wt % in a non-aqueous electrolyte (ethylene

glycol) which contained 0.5 wt % NH4F were investigated on fabrication of TiO2-ZrO2-

ZrTiO4 nanotubes in Chapter 5. The effect of microstructure of Ti50Zr alloy on the

orientation of nanotubes was more observable when the water content increased. The

walls of nanotubes were not separated when they were fabricated in non-aqueous

electrolyte. The key conclusions are as follows:

1. Increasing the water content from 5 to 10 wt % in electrolyte resulted in an

increase in the number of nanotubes with larger Di, Do and Wt or a decrease in

the number of nanotubes with smaller Di, Do and Wt due to the increased rate of

chemical and electrochemical reactions.

2. Increasing the applied potential from 20 to 30 V during anodisation with the

same water content in the electrolyte resulted in boarder distribution of Di and

Do. It means that the number of nanotubes with smaller Di and Do decreased or

the number of nanotubes with larger Di and Do increased due to the increased

anodic current. The extent of increasing of the mean Wt was not as much as the

mean Di and Do.

3. Increasing the applied potential from 20 to 30 V resulted in an increase in the

mean Di and Do of TiO2 nanotubes on the surface of CP-Ti but the mean Wt did

not change obviously. Increasing the water content from 5 to 10 wt % in the

electrolyte at the same applied potential resulted in an increase in the mean Di,

Do and Wt.

4. The mean Di, Do and Wt of TiO2 nanotubes were higher than those of TiO2-

ZrO2-ZrTiO4 nanotubes.

5. The mean Di and Do of nanotubes fabricated in non-aqueous electrolyte were

smaller than the mean Di and Do of nanotubes fabricated in aqueous electrolyte.

239

The distribution of nanotube sizes fabricated in both aqueous and non-aqueous

electrolytes were analysed using normal and Weibull distribution methods. The results

indicated that the normal distribution was an appropriate method to represent the

distribution of nanotube sizes of TiO2-ZrO2-ZrTiO4.

The Mean Roughness (Sa) and the Root Mean Square (RMS) Roughness (Sq) of the

nanotubular layer of TiO2-ZrO2-ZrTiO4 were evaluated using 3D-Profilometry

(Chapters 4, 5, 6 and 7). As it mentioned before because of different oxidation and

dissolution rates of the elements of the alloy; Ti and Zr, the nanotubes not only

exhibited different nanotube sizes but also exhibited different lengths, resulting in

different amplitude and space roughness parameters. The key conclusions are as

follows:

1. Increasing the CF- in the aqueous electrolyte resulted in an increase in the Sa

and Sq.

2. Increasing the applied potential in the aqueous electrolyte from 5 to 20V

resulted in an increase in the Sa and Sq but these roughness parameters

decreased when the applied potential increased from 20 to 35 V due to the

increased anodic current.

3. Increasing the water content from 5 to 10 wt % in the non-aqueous

electrolyte at the same applied potential resulted in an increase in Sa and Sq.

4. Increasing the applied potential from 20 to 30 V in the non-aqueous

electrolyte with the same water content resulted in an increase in Sa and Sq.

5. The Sa and Sq of the nanotubular layer of TiO2 on the surface of CP-Ti

increased with an increase in the CF- in the aqueous electrolyte and it

increased when the applied potential increased from 10 to 20 V then

decreased when the applied potential increased from 20 to 30V.

6. The Sa and Sq of the nanotubular layer of TiO2 on the surface of CP-Ti

increased with an increase in water content in the non-aqueous electrolyte

and applied potential.

7. The value of Sa and Sq of the nanotubular layer of TiO2-ZrO2-ZrTiO4 were

higher than those of the nanotubular layer of TiO2 anodised under the same

condition.

240

8. The values of Sa and Sq of the nanotubular layer of TiO2-ZrO2-ZrTiO4

fabricated in aqueous electrolyte were higher than those of the nanotubular

layer of TiO2-ZrO2-ZrTiO4 anodised in the non-aqueous electrolyte.

In Chapters 4 and 5, it was found that there was a direct link between roughness

parameter and wettability or hydrophilic properties of nanotubular layers of TiO2-ZrO2-

ZrTiO4 fabricated in both aqueous and non-aqueous electrolyte. The key conclusions are

as follows:

1. Increasing the CF- in the aqueous electrolyte resulted in an increase in the

hydrophilic properties of TiO2-ZrO2-ZrTiO4 nanotubes.

2. Increasing the applied potential in the aqueous electrolyte from 5 to 30 V

resulted in an increase in the hydrophilic properties of TiO2-ZrO2-ZrTiO4

nanotubes.

3. The hydrophilic properties of TiO2 nanotubes on the surface of CP-Ti increased

with an increase of CF – in aqueous electrolyte from 0.3 to 0.5 wt % and an

increase of the applied potential from 5 to 30 V, and did not change when the

applied potential increased from 30V to 35V.

4. The hydrophilic properties of TiO2-ZrO2-ZrTiO4 nanotubes fabricated in

aqueous electrolyte were higher than those of the TiO2 nanotubes anodised under

the same applied potential and it increased with an increase in mean Di and

roughness parameters.

5. Increasing the water content from 5 to 10 wt % in the non-aqueous electrolyte at

the same applied potential resulted in an increase in the hydrophilic properties of

TiO2-ZrO2-ZrTiO4 nanotubes.

6. Increasing the applied potential from 20 to 30 V in the non-aqueous electrolyte

with the same water content resulted in an increase in the hydrophilic properties

of TiO2-ZrO2-ZrTiO4 nanotubes.

7. The hydrophilic properties of TiO2 nanotubes on the surface of CP-Ti increased

with an increase of water content in non-aqueous electrolyte from 5 to 10 wt %

and an increase of the applied potential from 20 to 30 V.

8. The hydrophilic properties of TiO2-ZrO2-ZrTiO4 nanotubes fabricated in non-

aqueous electrolyte were almost the same in comparison to the hydrophilic

properties of TiO2 nanotubes anodised under the same condition.

241

9. The hydrophilic properties of TiO2-ZrO2-ZrTiO4 nanotubes fabricated in

aqueous electrolyte were higher than those of the TiO2-ZrO2-ZrTiO4 nanotubes

fabricated in non-aqueous electrolyte probably because of their different mean

mean Di and roughness parameters of Sa.

There was a direct link between the hydrophilic properties and the calculated surface

energy of nanotubular layer of TiO2-ZrO2-ZrTiO4 fabricated in aqueous electrolyte. The

surface energy of the nanotubular layer increased when the hydrophilic properties

increased. Such a link was not observed for the nanotubular layer of TiO2-ZrO2-ZrTiO4

fabricated in non-aqueous electrolyte due to the possibility of remaining of the organic

electrolyte inside the nanotubes. The surface energies of the nanotubular layers of TiO2-

ZrO2-ZrTiO4 were higher than those of the nanotubular layers of TiO2 fabricated in

aqueous electrolyte under the same applied potential and anodisation time. The surface

energies of the nanotubular layers of TiO2-ZrO2-ZrTiO4 fabricated in aqueous

electrolyte were lower than those of the nanotubular layers of TiO2-ZrO2-ZrTiO4

fabricated in non-aqueous electrolyte under the same applied potential and anodisation

time.

The hardness and elastic modulus of both nanotubular layers of TiO2-ZrO2-ZrTiO4

fabricated in aqueous and non-aqueous electrolyte were measured using nanoindenter,

Chapters 4 and 5. The porosity of both nanotubular layers of TiO2-ZrO2-ZrTiO4

fabricated in aqueous and non-aqueous electrolyte caused a decrease in the hardness of

the Ti50Zr alloy surface. However the ceramic properties of the metal oxide layer

fabricated in aqueous electrolyte caused an increase in the elasticity of the alloy surface.

The hard and brittle nanotubular layer fabricated in non-aqueous electrolyte exhibited

lower elastic modulus than the bare alloy. The nano-hardness and elasticity were

influenced by the values of the porosity (nanotube sizes) for both nanotubular layers.

In Chapters 4 and 5, the nanotubular layer of TiO2-ZrO2-ZrTiO4 and TiO2 on the surface

of Ti50Zr and CP-Ti fabricated in aqueous and non-aqueous electrolyte were

amorphous. After annealing crystalline phases formed and their physical, chemical and

mechanical properties changed. Conclusions are as follows:

1. The amorphous TiO2-ZrO2-ZrTiO4 nanotubes fabricated in aqueous electrolyte

changed to a mixture of anatase, orthorhombic TiO2 and ZrO2, and orthorhombic

242

ZrTiO4. The amorphous TiO2 nanotubes changed to tetragonal anatase after

annealing.

2. The mean Di and Do and Wt of TiO2-ZrO2-ZrTiO4 nanotubes anodised at lower

applied potential decreased after annealing; but increased when they were

anodised at higher applied potential. The mean Wt did not show an obvious

change. The mean Di and Do of TiO2 nanotubes anodised at higher applied

potential increased after annealing and the mean Wt of TiO2 nanotubes did not

exhibit a remarkable change.

3. The mean Sa and Sq of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes decreased and

their hydrophilic properties, surface energy increased after annealing; whilst the

hardness and elasticity of the nanotubular layer of TiO2-ZrO2-ZrTiO4 increased.

4. The amorphous TiO2-ZrO2-ZrTiO4 nanotubes fabricated in non-aqueous

electrolyte changed to a mixture of tetragonal anatase and rutile, orthorhombic

TiO2 and ZrO2, and orthorhombic ZrTiO4 after annealing.

5. The mean Di and Do and Wt of TiO2-ZrO2-ZrTiO4 nanotubes changed after

annealing. The mean Di of both TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes

decreased, the mean Wt increased and the mean Do did not change.

6. The top layer of nanotubes fabricated in non-aqueous electrolyte was brittle and

cracked after annealing. Therefore it was removed and the underneath dimpled

surface of nanotubes was used for further investigation.

The bioactivity of TiO2-ZrO2-ZrTiO4 and TiO2 nanotubes fabricated in aqueous and

non-aqueous electrolyte was evaluated by immersing them in m-SBF and the Ca/P ratio

(at %) was measured using EDS. The nanotubular layer of TiO2-ZrO2-ZrTiO4 and TiO2

fabricated in both aqueous and non-aqueous electrolytes with a variety range of

nanotube sizes were bioactive and the size, the shape, hydrophilic properties and

crystalline phase of the nanotubes affected the Ca/P ratio of HA.

The cell behaviour of osteoblast (SaOS2) on the surface of TiO2-ZrO2-ZrTiO4

nanotubes fabricated in aqueous and non-aqueous electrolyte with a variety of nanotube

sizes and shapes with different physical and chemical properties was evaluated. It was

found that besides the Di of nanotubes the topography of the nanotubular surface and its

chemical properties affect the density of cell attachment. The highest percentage of cell

attachment of 65.2 % was on the surface of nanotubular layer with Di = 18 ± 6 nm.

243

Although it decreased when the Di increased to 59 ± 17 nm as expected (the lowest

percentage of cell attachment of 25.9 %), the percentage of cell attachment increased on

the surface of TiO2-ZrO2-ZrTiO4 nanotubes with Di = 64 ± 23 nm and Wt = 21 ± 7 and

Di = 82 ± 26 nm and Wt = 24 ± 7 nm, respectively. It can be concluded that the high cell

attachment was on the surface of nanotubes with either Di or Wt in the range of optimum

nanospacing and increasing the roughness parameter of the surface distracted the

process of cell attachment. Two post treatments; cleaning and coating with HA which

were undertaken on the nanotubular layer with Di = 40 ± 12 nm resulted in an increase

of ~ 1.5 to 2 times of the percentage of cell attachment. Almost the same result was

observed when the osteoblast cells seeded on the surface of TiO2-ZrO2-ZrTiO4

nanotubes fabricated in non-aqueous electrolyte. Although the morphology and

topography of nanotubular layer fabricated in aqueous electrolyte are different from

nanotubular layer fabricated in non-aqueous electrolyte two surfaces with almost the

same Sa and Sm were compared. The percentage of attached cells on dimpled

nanotubular surface fabricated in non-aqueous electrolyte was higher than that of the

nanotubular layer fabricated in aqueous electrolyte with smaller Di and higher

hydrophilic properties and surface energy. It can be concluded that the non-separated

wall of the nanotubes may receive more protein adsorption and serves as a more suitable

nano-pattern surface for bone cell attachment.

It can be concluded from the study presented here that the mixed nanotubes of TiO2-

ZrO2-ZrTiO4 on Ti50Zr alloy with a variety of nanotube size distribution and

morphology and other related properties such as roughness, hydrophilic and bioactivity

along with other excellent properties of the substrate (bare alloy) are more promising

and have potential for biomedical application than 100 % TiO2 nanotubes on CP-Ti.

The nanoporous and nanotubular layer of metal oxides fabricated on the surface of the

most commonly used titanium alloying element of Ta, Nb and Zr were processed and

their physical, chemical and bioactive properties in comparison to TiO2 nanotubes with

almost the same Di were evaluated. The nanoporous Ta2O5 with non-uniform pore size

exhibited a decrease of the roughness and nanoporous Nb2O5 with uniform pore size

exhibited an increase of the roughness in comparison to their respective bare metal. The

nanotubular layer of ZrO2 with almost the same Di as nanotubular layer of TiO2 was

fabricated at a higher applied potential and exhibited higher roughness than its

244

respective bare metal. The hydrophilic properties and surface energy of these

nanoporous and nanotubular layer was higher than their bare metal and increased after

annealing and their crystalline phase changed. Similar to TiO2 nanotube, as-formed

nanoporous Ta2O5 and ZrO2 nanotubes revealed a good bioactivity but as-formed

nanoporous Nb2O5 were not bioactive. All these nanoporous and nanotubular layers

were bioactive after annealing and annealed nanoporous Ta2O5 exhibited a high level of

bioactivity. It can be concluded that these alloying elements are promising for

biomaterial application alone or accompany with other biomaterials.

9.3 Recommendation for future work

Some area of this study remained still to be identified and future research can be

explored as follows:

1. There was a direct link between the roughness parameters of the surface and its

wettability. If the water contact angle of nanotubular surface with same

distribution of Di and different roughness parameters, and same roughness

parameter and different distribution of Di were measured, a modelling study

could be undertaken to clarify this relationship. With this model an optimum

nanotubular layer for enhanced bone cell attachment could be fabricated by

controlling the anodisation parameters.

2. Although this study attempted to interpret the hydrophilic property, bioactivity

and bone cell attachment by studying the roughness amplitude and spacing

parameters and surface area index (SI), evaluating the nanoporosity of the

nanotubular layer with different Di could provide a more reliable explanation.

3. More fundamental study is needed to modify the mechanical properties of

nanotubular layer fabricated in non-aqueous electrolyte after annealing.

4. There is a debate about optimum nanospacing for cell adhesion. There were

conclusions that Di ≤ 30 nm of as-formed nanotubes or Di ≤ 100 nm of annealed

nanotubes was the optimum nanospacing. Therefore cell attachment study with

the same distribution of Di and roughness parameter could be undertaken on as-

formed and annealed nanotubes to clarify this debate.

5. The effect of applying post treatment on the nanotubes anodised in non-aqueous

electrolyte such as removing contaminations of the nanotubular layer or

245

deposition of an HA layer via immersion in SBF on cell adhesion could be

undertaken.

6. More fundamental knowledge is needed to explain the different relationship

between water contact angle and surface energy of the nanotubular layer

fabricated in aqueous and non-aqueous electrolyte and to modify it.

7. The nanotubular layer fabricated in both electrolytes can be coated by HA

electrodeposition and the effect of this layer on improving bone cell attachment

could be studied and compared with the mineralised HA by soaking in m-SBF.

The physical, chemical and mechanical properties of these final layers can be

measured and compared.

246

References

[1] Williams DF. On the nature of biomaterials. Biomaterials 2009;30:5897-909. [2] Navarro M, Michiardi A, Castaño O, Planell JA. Biomaterials in orthopaedics. Journal

of The Royal Society Interface 2008;5:1137-58. [3] Niinomi M. Mechanical biocompatibilities of titanium alloys for biomedical

applications. Journal of the Mechanical Behavior of Biomedical Materials 2008;1:30-42.

[4] Niinomi M. Metallic biomaterials. Journal of Artificial Organs 2008;11:105-10. [5] Long M, Rack HJ. Titanium alloys in total joint replacement - A materials science

perspective. Biomaterials 1998;19:1621-39. [6] Duan K, Wang R. Surface modifications of bone implants through wet chemistry.

Journal of Materials Chemistry 2006;16:2309-21. [7] Liu X, Chu P, Ding C. Surface modification of titanium, titanium alloys, and related

materials for biomedical applications. Materials Science and Engineering: R: Reports 2004;47:49-121.

[8] Hsu HC, Wu SC, Sung YC, Ho WF. The structure and mechanical properties of as-cast Zr-Ti alloys. Journal of Alloys and Compounds 2009;488:279-83.

[9] Wang X, Li Y, Xiong J, Hodgson PD, Wen Ce. Porous TiNbZr alloy scaffolds for biomedical applications. Acta Biomaterialia 2009;5:3616-24.

[10] Zwilling V, Aucouturier M, Darque-Ceretti E. Anodic oxidation of titanium and TA6V alloy in chromic media. An electrochemical approach. Electrochimica Acta 1999;45:921–9.

[11] Roy P, Berger S, Schmuki P. TiO2 Nanotubes: Synthesis and Applications. Angewandte Chemie International Edition 2011;50:2904-39.

[12] Liu X, Chu PK, Ding C. Surface nano-functionalization of biomaterials. Materials Science and Engineering: R: Reports 2010;70:275-302.

[13] Gulati K, Simovic S, Losic D. Highly ordered titania (TiO2) nanotube arrays fabricated by electrochemical self-ordering process toward development of implantable drug delivery devices with triggered drug release. Canberra, ACT2010. p. 159-60.

[14] Lavenus S, Ricquier JC, Louarn G, Layrolle P. Cell interaction with nanopatterned surface of implants. Nanomedicine 2010;5:937-47.

[15] Anselme K. Osteoblast adhesion on biomaterials. Biomaterials 2000;21:667-81. [16] Minagar S, Wang J, Berndt CC, Ivanova EP, Wen C. Cell response of anodized

nanotubes on titanium and titanium alloys. Journal of Biomedical Materials Research - Part A 2013;101 A:2726-39.

[17] Park J, Bauer S, Von Der Mark K, Schmuki P. Nanosize and vitality: TiO2 nanotube diameter directs cell fate. Nano Letters 2007;7:1686-91.

[18] Von Wilmowsky C, Bauer S, Lutz R, Meisel M, Neukam FW, Toyoshima T, Schmuki P, Nkenke E, Schlegel KA. In Vivo Evaluation of Anodic TiO2 Nanotubes; An Experimental Study in the Pig. Journal of Biomedical Materials Research - Part B Applied Biomaterials 2009;89:165-71.

[19] Park J, Bauer S, Schlegel KA, Neukam FW, von der Mark K, Schmuki P. TiO2 Nanotube Surfaces: 15 nm-An Optimal Length Scale of Surface Topography for Cell Adhesion and Differentiation. Small 2009;5:666-71.

247

[20] Brammer KS, Oh S, Cobb CJ, Bjursten LM, Heyde Hvd, Jin S. Improved bone-forming functionality on diameter-controlled TiO2 nanotube surface. Acta Biomaterialia 2009;5:3215-23.

[21] Oh S, Brammer KS, Li YSJ, Teng D, Engler AJ, Chien S, Jin S. Stem cell fate dictated solely by altered nanotube dimension. Proceedings of the National Academy of Sciences 2009;106:2130-5.

[22] Roguska A, Pisarek M, Andrzejczuk M, Dolata M, Lewandowska M, Janik-Czachor M. Characterization of a calcium phosphate-TiO2 nanotube composite layer for biomedical applications. Materials Science and Engineering C 2011;31:906-14.

[23] Varghese OK, Gong D, Paulose M, Grimes CA, Dickey EC. Crystallization and high-temperature structural stability of titanium oxide nanotube arrays. Journal of Materials Research 2003;18:156-65.

[24] Bai Y, Park IS, Park HH, Lee MH, Bae TS, Duncan W, Swain M. The effect of annealing temperatures on surface properties, hydroxyapatite growth and cell behaviors of TiO2 nanotubes. Surface and Interface Analysis 2011;43:998-1005.

[25] Narayanan R, Lee HJ, Kwon TY, Kim KH. Anodic TiO2 nanotubes from stirred baths: Hydroxyapatite growth & osteoblast responses. Materials Chemistry and Physics 2011;125:510-7.

[26] Sista S, Nouri A, Li Y, Wen C, Hodgson PD, Pande G. Cell biological responses of osteoblasts on anodized nanotubular surface of a titanium-zirconium alloy. Journal of Biomedical Materials Research - Part A 2013;101:3416-30.

[27] Sista S, Wen Ce, Hodgson PD, Pande G. The influence of surface energy of titanium-zirconium alloy on osteoblast cell functions in vitro. Journal of Biomedical Materials Research Part A 2011;97A:27-36.

[28] Grigorescu S, Ungureanu C, Kirchgeorg R, Schmuki P, Demetrescu I. Various sized nanotubes on TiZr for antibacterial surfaces. Applied Surface Science 2013;270:190-6.

[29] Weiner S, Wagner HD. The material bone: Structure-mechanical function relations. Annual Review of Materials Science 1998;28:271-98.

[30] Uddin MH, Matsumoto T, Okazaki M, Nakahira A, T S. Biomimetic fabrication of apatite related biomaterials. In: M A, editor. Biomimetics, learning from nature. Vukovar, Croatia: In-Teh; 2010. p. 289-303.

[31] Mour M, Das D, Winkler T, Hoenig E, Mielke G, Morlock MM, Schilling AF. Advances in Porous Biomaterials for Dental and Orthopaedic Applications. Materials 2010;3:2947-74.

[32] Rho JY, Kuhn-Spearing L, Zioupos P. Mechanical properties and the hierarchical structure of bone. Medical Engineering and Physics 1998;20:92-102.

[33] Wheeless CR. Wheeless' textbook of orthopaedics. [S.l.]: C.R. Wheeless, M.D.; 1996. [34] Anselme K. Biomaterials and interface with bone. Osteoporosis International

2011;22:2037-42. [35] Mark K, Park J, Bauer S, Schmuki P. Nanoscale engineering of biomimetic surfaces:

cues from the extracellular matrix. Cell and Tissue Research 2009;339:131-53. [36] Yu WQ, Jiang XQ, Zhang FQ, Xu L. The effect of anatase TiO2 nanotube layers on

MC3T3-E1 preosteoblast adhesion, proliferation, and differentiation. Journal of Biomedical Materials Research - Part A 2010;94:1012-22.

[37] Kokubo T, Kim H-M, Kawashita M. Novel bioactive materials with different mechanical properties. Biomaterials 2003;24:2161-75.

248

[38] Hench LL, Polak JM. Third-generation biomedical materials. Science 2002;295:1014+6-7.

[39] Steinemann SG. Titanium - The material of choice? Periodontology 2000 1998;17:7-21.

[40] Niinomi M. Mechanical properties of biomedical titanium alloys. Materials Science and Engineering A 1998;243:231-6.

[41] Zhou YL, Niinomi M, Akahori T. Effects of Ta content on Young's modulus and tensile properties of binary Ti-Ta alloys for biomedical applications. Materials Science and Engineering A 2004;371:283-90.

[42] Kikuchi M, Takahashi M, Okuno O. Mechanical properties and grindability of dental cast Ti-Nb alloys. Dental Materials Journal 2003;22:328-42.

[43] Niespodziana K, Jurczyk K, Jurczyk M. The synthesis of titanium alloys for biomedical applications. Reviews on Advanced Materials Science 2008;18:236-40.

[44] Elias LM, Schneider SG, Schneider S, Silva HM, Malvisi F. Microstructural and mechanical characterization of biomedical Ti–Nb–Zr(–Ta) alloys. Materials Science and Engineering: A 2006;432:108-12.

[45] Okazaki Y. A new Ti-15Zr-4Nb-4Ta alloy for medical applications. Current Opinion in Solid State and Materials Science 2001;5:45-53.

[46] Okazaki Y, Rao S, Ito Y, Tateishi T. Corrosion resistance, mechanical properties: corrosion fatigue strength and cytocompatibility of new Ti alloys without A1 and V. Biomaterials 1998;19:1197-215.

[47] Lausmaa J. Surface spectroscopic characterization of titanium implant materials. Journal of Electron Spectroscopy and Related Phenomena 1996;81:343-61.

[48] Tsuchiya H, Macak JM, Müller L, Kunze J, Müller F, Greil P, Virtanen S, Schmuki P. Hydroxyapatite growth on anodic TiO2 nanotubes. Journal of Biomedical Materials Research Part A 2006;77A:534-41.

[49] Sul YT, Johansson CB, Jeong Y, Albrektsson T. The electrochemical oxide growth behaviour on titanium in acid and alkaline electrolytes. Medical Engineering and Physics 2001;23:329-46.

[50] Masuda H, Hasegwa F, Ono S. Self-ordering of cell arrangement of anodic porous alumina formed in sulfuric acid solution. Journal of the Electrochemical Society 1997;144:L127-L30.

[51] Sieber I, Hildebrand H, Friedrich A, Schmuki P. Formation of self-organized niobium porous oxide on niobium. Electrochemistry Communications 2005;7:97-100.

[52] Sieber I, Kannan B, Schmuki P. Self-Assembled Porous Tantalum Oxide Prepared in H2SO4/HF Electrolytes. Electrochemical and Solid-State Letters 2005;8:J10-J2.

[53] Tsuchiya H, Schmuki P. Thick self-organized porous zirconium oxide formed in H2SO4/NH4F electrolytes. Electrochemistry Communications 2004;6:1131-4.

[54] Oh S, Daraio C, Chen L-H, Pisanic TR, Fiñones RR, Jin S. Significantly accelerated osteoblast cell growth on aligned TiO2 nanotubes. Journal of Biomedical Materials Research Part A 2006;78A:97-103.

[55] Sibert ME. Electrochemical Oxidation of Titanium Surfaces. Journal of the Electrochemical Society 1963;110:65-72.

[56] Kokubo T. Apatite formation on surfaces of ceramics, metals and polymers in body environment. Acta Materialia 1998;46:2519-27.

249

[57] Fojt J, Moravec H, Joska L. Nanostructuring of titanium for medical applications. Nanocon 2010, 2nd International Conference. Olomouc, Czech Republic, EU: Tanger Ltd 2010.

[58] Macak JM, Tsuchiya H, Ghicov A, Yasuda K, Hahn R, Bauer S, Schmuki P. TiO2 nanotubes: Self-organized electrochemical formation, properties and applications. Current Opinion in Solid State and Materials Science 2007;11:3-18.

[59] Sun L, Wang X, Li M, Zhang S, Wang Q. Anodic titania nanotubes grown on titanium tubular electrodes. Langmuir 2014;30:2835-41.

[60] Sánchez-Tovar R, Lee K, García-Antón J, Schmuki P. Formation of anodic TiO2 nanotube or nanosponge morphology determined by the electrolyte hydrodynamic conditions. Electrochemistry Communications 2013;26:1-4.

[61] Bauer S, Kleber S, Schmuki P. TiO2 nanotubes: Tailoring the geometry in H3PO4/HF electrolytes. Electrochemistry Communications 2006;8:1321-5.

[62] Beranek R, Hildebrand H, Schmuki P. Self-organized porous titanium oxide prepared in H2SO4/HF electrolytes. Electrochemical and Solid-State Letters 2003;6:B12-B4.

[63] Neupane MP, Park IS, Lee MH, Bae TS, Watari F. Influence of heat treatment on morphological changes of nano-structured titanium oxide formed by anodic oxidation of titanium in acidic fluoride solution. Bio-Medical Materials and Engineering 2009;19:77-83.

[64] Oh S, Finones R, Daraio C, Chen L, Jin S. Growth of nano-scale hydroxyapatite using chemically treated titanium oxide nanotubes. Biomaterials 2005;26:4938-43.

[65] Demetrescu I, Ionita D, Pirvu C, Portan D. Present and Future Trends in TiO2 Nanotubes Elaboration, Characterization and Potential Applications. Molecular Crystals and Liquid Crystals 2010;521:195-203.

[66] Zhao J, Wang X, Chen R, Li L. Fabrication of titanium oxide nanotube arrays by anodic oxidation. Solid State Communications 2005;134:705-10.

[67] Macak JM, Tsuchiya H, Schmuki P. High-Aspect-Ratio TiO2 Nanotubes by Anodization of Titanium. Angewandte Chemie International Edition 2005;44:2100-2.

[68] Wan J, Yan X, Ding J, Wang M, Hu K. Self-organized highly ordered TiO2 nanotubes in organic aqueous system. Materials Characterization 2009;60:1534-40.

[69] Lee W-J, Alhoshan M, Smyrl WH. Titanium Dioxide Nanotube Arrays Fabricated by Anodizing Processes. Journal of the Electrochemical Society 2006;153:B499-B505.

[70] Popat KC, Eltgroth M, LaTempa TJ, Grimes CA, Desai TA. Decreased Staphylococcus epidermis adhesion and increased osteoblast functionality on antibiotic-loaded titania nanotubes. Biomaterials 2007;28:4880-8.

[71] Ghicov A, Tsuchiya H, Macak JM, Schmuki P. Titanium oxide nanotubes prepared in phosphate electrolytes. Electrochemistry Communications 2005;7:505-9.

[72] MacAk JM, Sirotna K, Schmuki P. Self-organized porous titanium oxide prepared in Na2SO4/NaF electrolytes. Electrochimica Acta 2005;50:3679-84.

[73] Macak JM, Taveira LV, Tsuchiya H, Sirotna K, Macak J, Schmuki P. Influence of different fluoride containing electrolytes on the formation of self-organized titania nanotubes by Ti anodization. Journal of Electroceramics 2006;16:29-34.

[74] Feng B, Chu X, Chen J, Wang J, Lu X, Weng J. Hydroxyapatite coating on titanium surface with titania nanotube layer and its bond strength to substrate. Journal of Porous Materials 2009;17:453-8.

250

[75] Crawford G, Chawla N, Das K, Bose S, Bandyopadhyay A. Microstructure and deformation behavior of biocompatible TiO2 nanotubes on titanium substrate. Acta Biomaterialia 2007;3:359-67.

[76] Narayanan R, Kwon T-Y, Kim K-H. TiO2 nanotubes from stirred glycerol/NH4F electrolyte: Roughness, wetting behavior and adhesion for implant applications. Materials Chemistry and Physics 2009;117:460-4.

[77] Vasilev K, Poh Z, Kant K, Chan J, Michelmore A, Losic D. Tailoring the surface functionalities of titania nanotube arrays. Biomaterials 2010;31:532-40.

[78] Albu SP, Ghicov A, Macak JM, Schmuki P. 250 µm long anodic TiO2 nanotubes with hexagonal self-ordering. physica status solidi (RRL) – Rapid Research Letters 2007;1:R65-R7.

[79] Tsuchiya H, Macak JM, Taveira L, Balaur E, Ghicov A, Sirotna K, Schmuki P. Self-organized TiO2 nanotubes prepared in ammonium fluoride containing acetic acid electrolytes. Electrochemistry Communications 2005;7:576-80.

[80] Macak JM, Tsuchiya H, Taveira L, Aldabergerova S, Schmuki P. Smooth Anodic TiO2 Nanotubes. Angewandte Chemie International Edition 2005;44:7463-5.

[81] Song YY, Schmidt-Stein F, Bauer S, Schmuki P. Amphiphilic TiO2 nanotube arrays: An actively controllable drug delivery system. Journal of the American Chemical Society 2009;131:4230-2.

[82] Cai Q, Yang L, Yu Y. Investigations on the self-organized growth of TiO2 nanotube arrays by anodic oxidization. Thin Solid Films 2006;515:1802-6.

[83] Cai Q, Paulose M, Varghese OK, Grimes CA. The Effect of Electrolyte Composition on the Fabrication of Self-Organized Titanium Oxide Nanotube Arrays by Anodic Oxidation. Journal of Materials Research 2011;20:230-6.

[84] Xin Y CP. Nanostructured Titania coatings for Biological Applications: Fabrication and Characterization. In: Z S, editor. Biological and Biomedical Coatings Handbook : Processing and Characterization. Hoboken: CRC Press; 2011. p. 203-36.

[85] Crawford GA, Chawla N, Ringnalda J. Processing and microstructure characterization of a novel porous hierarchical TiO2 structure. Journal of Materials Research 2009;24:1683-7.

[86] Crawford GA, Chawla N. Porous hierarchical TiO2 nanostructures: Processing and microstructure relationships. Acta Materialia 2009;57:854-67.

[87] Chawla GACaN. Biocompatible TiO2 nanotube coatings fabricated by anodic oxidation of Ti. In: RL Karlinsey, editor. Recent Developments in Advanced Medical and Dental Materials Using Electrochemical Methodologies. Kerala, India: Research Signpost; 2009. p. 177-99.

[88] Crawford GA, Chawla N. Tailoring TiO2 nanotube growth during anodic oxidation by crystallographic orientation of Ti. Scripta Materialia 2009;60:874-7.

[89] Crawford GA, Chawla N, Houston JE. Nanomechanics of biocompatible TiO2 nanotubes by Interfacial Force Microscopy (IFM). Journal of the Mechanical Behavior of Biomedical Materials 2009;2:580-7.

[90] Moon SK, Kwon JS, Uhm SH, Lee EJ, Gu HJ, Eom TG, Kim KN. Biological evaluation of micro-nano patterned implant formed by anodic oxidation. Current Applied Physics 2014;14:S183-S7.

[91] Bauer S, Park J, Faltenbacher J, Berger S, Von Der Mark K, Schmuki P. Size selective behavior of mesenchymal stem cells on ZrO2 and TiO2 nanotube arrays. Integrative Biology 2009;1:525-32.

251

[92] S. L. Wrought Titanium and Titanium Alloys, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials. ASM Handbook: ASM International; 1990. p. 592–633.

[93] Fukuda A, Takemoto M, Saito T, Fujibayashi S, Neo M, Yamaguchi S, Kizuki T, Matsushita T, Niinomi M, Kokubo T, Nakamura T. Bone bonding bioactivity of Ti metal and Ti-Zr-Nb-Ta alloys with Ca ions incorporated on their surfaces by simple chemical and heat treatments. Acta Biomaterialia 2011;7:1379-86.

[94] Nakada H, Numata Y, Sakae T, Okazaki Y, Tanimoto Y, Tamaki H, Katou T, Ookubo A, Kobayashi K, LeGeros RZ. Comparison of bone mineral density and area of newly formed bone around Ti-15%Zr-4%Nb-4%Ta alloy and Ti-6%A1-4%V alloy implants. Journal of Hard Tissue Biology 2008;17:99-108.

[95] Choe H-C, Saji VS, Ko Y-M. Mechanical properties and corrosion resistance of low rigidity quaternary titanium alloy for biomedical applications. Transactions of Nonferrous Metals Society of China 2009;19:862-5.

[96] Li SJ, Yang R, Li S, Hao YL, Cui YY, Niinomi M, Guo ZX. Wear characteristics of Ti-Nb-Ta-Zr and Ti-6Al-4V alloys for biomedical applications. Wear 2004;257:869-76.

[97] Tsuchiya H, Macak JM, Ghicov A, Schmuki P. Self-Organization of Anodic Nanotubes on Two Size Scales. Small 2006;2:888-91.

[98] Tsuchiya H, Macak JM, Ghicov A, Tang YC, Fujimoto S, Niinomi M, Noda T, Schmuki P. Nanotube oxide coating on Ti-29Nb-13Ta-4.6Zr alloy prepared by self-organizing anodization. Electrochimica Acta 2006;52:94-101.

[99] Tsuchiya H, Macak JM, Ghicov A, Schmuki P. Anodic oxide nanotubes on Ti alloys. ECS Transactions 2007;3:365-74.

[100] Macak JM, Tsuchiya H, Taveira L, Ghicov A, Schmuki P. Self-organized nanotubular oxide layers on Ti-6Al-7Nb and Ti-6Al-4V formed by anodization in NH4F solutions. Journal of Biomedical Materials Research Part A 2005;75A:928-33.

[101] Feng XJ, Macak JM, Albu SP, Schmuki P. Electrochemical formation of self-organized anodic nanotube coating on Ti-28Zr-8Nb biomedical alloy surface. Acta Biomaterialia 2008;4:318-23.

[102] Tsuchiya H, Macak JM, Taveira L, Schmuki P. Formation of self-organized zirconia nanostructure. 208th Meeting of the Electrochemical Society. Los Angeles, CA: ECS Transactions; 2005. p. 351-7.

[103] Choi J, Lim JH, Lee J, Kim KJ. Porous niobium oxide films prepared by anodization-annealing-anodization. Nanotechnology 2007;18:055603 (6pp).

[104] Sieber IV, Schmuki P. Porous Tantalum Oxide Prepared by Electrochemical Anodic Oxidation. Journal of the Electrochemical Society 2005;152:C639-C44.

[105] Tsuchiya H, Nakata J, Fujimoto S, Berger S, Schmuki P. Anodic porous and tubular oxide layers on Ti alloys. 3 ed. Honolulu, HI2008. p. 359-67.

[106] Tsuchiya H, Akaki T, Nakata J, Terada D, Tsuji N, Koizumi Y, Minamino Y, Schmuki P, Fujimoto S. Anodic oxide nanotube layers on Ti-Ta alloys: Substrate composition, microstructure and self-organization on two-size scales. Corrosion Science 2009;51:1528-33.

[107] Tsuchiya H, Akaki T, Nakata J, Terada D, Tsuji N, Koizumi Y, Minamino Y, Schmuki P, Fujimoto S. Metallurgical aspects on the formation of self-organized anodic oxide nanotube layers. Electrochimica Acta 2009;54:5155-62.

252

[108] Wang Y, Tao J, Wang L, He P, Wang T. HA coating on titanium with nanotubular anodized TiO2 intermediate layer via electrochemical deposition. Transactions of Nonferrous Metals Society of China 2008;18:631-5.

[109] Das K, Bose S, Bandyopadhyay A. TiO2 nanotubes on Ti: Influence of nanoscale morphology on bone cell-materials interaction. Journal of Biomedical Materials Research Part A 2009;90A:225-37.

[110] Yasuda K, Schmuki P. Control of morphology and composition of self-organized zirconium titanate nanotubes formed in (NH4)2SO4/NH4F electrolytes. Electrochimica Acta 2007;52:4053-61.

[111] Yasuda K, Schmuki P. Electrochemical formation of self-organized zirconium titanate nanotube multilayers. Electrochemistry Communications 2007;9:615-9.

[112] Yasuda K, Schmuki P. Formation of self-organized zirconium titanate nanotube layers by alloy anodization. Advanced Materials 2007;19:1757-60.

[113] Ghicov A, Aldabergenova S, Tsuchyia H, Schmuki P. TiO2-Nb2O5 nanotubes with electrochemically tunable morphologies. Angewandte Chemie - International Edition 2006;45:6993-6.

[114] M. S. Tantalum-Properties of Pure Metals, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials. ASM Handbook: ASM International; 1990. p. 1099–201.

[115] WS. L. Titanium-Properties of Pure Metals, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials. ASM Handbook: ASM International 1990. p. 1099–201.

[116] Levine BR, Sporer S, Poggie RA, Della Valle CJ, Jacobs JJ. Experimental and clinical performance of porous tantalum in orthopedic surgery. Biomaterials 2006;27:4671-81.

[117] Sieber I, Hildebrand H, Friedrich A, Schmuki P. Initiation of tantalum oxide pores grown on tantalum by potentiodynamic anodic oxidation. Journal of Electroceramics 2006;16:35-9.

[118] Arnould C, Delhalle J, Mekhalif Z. Electrodeposition from ionic liquid of 2D ordered Ta2O5 on titanium substrate through a polystyrene template. Journal of the Electrochemical Society 2009;156:K186-K90.

[119] Q. Lu a SM, Skeldon P, Thompson GE, Masheder D, Habazaki H, Shimizu K. Anodic film growth on tantalum in dilute phosphoric acid solution at 20 and 85 °C. Electrochimica Acta 2002;47:2761-7.

[120] El-Sayed HA, Birss VI. Controlled interconversion of nanoarray of Ta dimples and high aspect ratio Ta oxide nanotubes. Nano Letters 2009;9:1350-5.

[121] Wei W, MacAk JM, Shrestha NK, Schmuki P. Thick self-ordered nanoporous Ta2O5 films with long-range lateral order. Journal of the Electrochemical Society 2009;156:K104-K9.

[122] Wei W, Macak JM, Schmuki P. High aspect ratio ordered nanoporous Ta2O5 films by anodization of Ta. Electrochemistry Communications 2008;10:428-32.

[123] Arnould C, Delhalle J, Mekhalif Z. Multifunctional hybrid coating on titanium towards hydroxyapatite growth: Electrodeposition of tantalum and its molecular functionalization with organophosphonic acids films. Electrochimica Acta 2008;53:5632-8.

[124] Arnould C, Delhalle J, Mekhalif Z. Corrigendum to "Multifunctional hybrid coating on titanium towards hydroxyapatite growth: Electrodeposition of tantalum and its

253

molecular functionalization with organophosphonic acids films" [Electrochim. Acta 53 (2008) 5632-5638] (DOI:10.1016/j.electacta.2008.03.003). Electrochimica Acta 2009;54:2402.

[125] Arnould C, Volcke C, Lamarque C, Thiry PA, Delhalle J, Mekhalif Z. Titanium modified with layer-by-layer sol-gel tantalum oxide and an organodiphosphonic acid: A coating for hydroxyapatite growth. Journal of Colloid and Interface Science 2009;336:497-503.

[126] Aagard RL. Optical waveguide characteristics of reactive dc-sputtered niobium pentoxide films. Applied Physics Letters 1975;27:605-7.

[127] Velten D, Eisenbarth E, Schanne N, Breme J. Biocompatible Nb2O5 thin films prepared by means of the sol-gel process. Journal of Materials Science: Materials in Medicine 2004;15:457-61.

[128] Mozalev A, Sakairi M, Saeki I, Takahashi H. Nucleation and growth of the nanostructured anodic oxides on tantalum and niobium under the porous alumina film. Electrochimica Acta 2003;48:3155-70.

[129] Lu Q, Hashimoto T, Skeldon P, Thompson GE, Habazaki H, Shimizu K. Nanoporous Anodic Niobium Oxide Formed in Phosphate/Glycerol Electrolyte. Electrochemical and Solid-State Letters 2005;8:B17-B20.

[130] Choi J, Lim JH, Lee SC, Chang JH, Kim KJ, Cho MA. Porous niobium oxide films prepared by anodization in HF/H3PO4. Electrochimica Acta 2006;51:5502-7.

[131] Tsuchiya H, Macak JM, Sieber I, Schmuki P. Self-Organized High-Aspect-Ratio Nanoporous Zirconium Oxides Prepared by Electrochemical Anodization. Small 2005;1:722-5.

[132] Tsuchiya H, Macak JM, Taveira L, Schmuki P. Fabrication and characterization of smooth high aspect ratio zirconia nanotubes. Chemical Physics Letters 2005;410:188-91.

[133] Berger S, Faltenbacher J, Bauer S, Schmuki P. Enhanced self-ordering of anodic ZrO2 nanotubes in inorganic and organic electrolytes using two-step anodization. Physica Status Solidi - Rapid Research Letters 2008;2:102-4.

[134] Berger S, Jakubka F, Schmuki P. Formation of hexagonally ordered nanoporous anodic zirconia. Electrochemistry Communications 2008;10:1916-9.

[135] Shin Y, Lee S. A freestanding membrane of highly ordered anodic ZrO2 nanotube arrays. Nanotechnology 2009;20:105301 (5pp).

[136] Lee W-J, Smyrl WH. Zirconium Oxide Nanotubes Synthesized via Direct Electrochemical Anodization. Electrochemical and Solid-State Letters 2005;8:B7-B9.

[137] Anselme K, Ploux L, Ponche A. Cell/Material Interfaces: Influence of Surface Chemistry and Surface Topography on Cell Adhesion. Journal of Adhesion Science and Technology 2010;24:831-52.

[138] Finke B, Luethen F, Schroeder K, Mueller PD, Bergemann C, Frant M, Ohl A, Nebe BJ. The effect of positively charged plasma polymerization on initial osteoblastic focal adhesion on titanium surfaces. Biomaterials 2007;28:4521-34.

[139] Lotz MM, Burdsal CA, Erickson HP, McClay DR. Cell adhesion to fibronectin and tenascin: Quantitative measurements of initial binding and subsequent strengthening response. Journal of Cell Biology 1989;109:1795-805.

[140] Bausch AR, Ziemann F, Boulbitch AA, Jacobson K, Sackmann E. Local measurements of viscoelastic parameters of adherent cell surfaces by magnetic bead microrheometry. Biophysical Journal 1998;75:2038-49.

254

[141] Yamamoto A, Mishima S, Maruyama N, Sumita M. A new technique for direct measurement of the shear force necessary to detach a cell from a material. Biomaterials 1998;19:871-9.

[142] Bowers VM, Fischer LR, Francis GW, Williams KL. A micromechanical technique for monitoring cell-substrate adhesiveness: measurements of the strength of red blood cell adhesion to glass and polymer test surfaces. Journal of Biomedical Materials Research 1989;23:1453-73.

[143] Truskey GA, Proulx TL. Relationship between 3T3 cell spreading and the strength of adhesion on glass and silane surfaces. Biomaterials 1993;14:243-54.

[144] Mege JL, Capo C, Benoliel AM, Bongrand P. Determination of binding strength and kinetics of binding initiation. A model study made on the adhesive properties of P388D1 macrophage-like cells. Cell Biophysics 1986;8:141-60.

[145] Arvidsson A, Franke-Stenport V, Andersson M, Kjellin P, Sul Y-T, Wennerberg A. Formation of calcium phosphates on titanium implants with four different bioactive surface preparations. An in vitro study. Journal of Materials Science: Materials in Medicine 2007;18:1945-54.

[146] Bruckert F, Weidenhaupt M. Applications of Micro- and Nano-technology to Study Cell Adhesion to Material Surfaces. Journal of Adhesion Science and Technology 2010;24:2127-40.

[147] Kabaso D, Gongadze E, Perutková Š, Matschegewski C, Kralj-Iglič V, Beck U, van Rienen U, Iglič A. Mechanics and electrostatics of the interactions between osteoblasts and titanium surface. Computer Methods in Biomechanics and Biomedical Engineering 2011;14:469-82.

[148] Li B, Li Y, Li J, Fu X, Li C, Wang H, Liu S, Guo L, Xin S, Liang C, Li H. Improvement of biological properties of titanium by anodic oxidation and ultraviolet irradiation. Applied Surface Science 2014;307:202-8.

[149] Decuzzi P, Ferrari M. Modulating cellular adhesion through nanotopography. Biomaterials 2010;31:173-9.

[150] Gentile F, Tirinato L, Battista E, Causa F, Liberale C, di Fabrizio EM, Decuzzi P. Cells preferentially grow on rough substrates. Biomaterials 2010;31:7205-12.

[151] Kim S, Lim J, Lee S, Nam S, Kang H, Choi J. Anodically nanostructured titanium oxides for implant applications. Electrochimica Acta 2008;53:4846-51.

[152] Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix Elasticity Directs Stem Cell Lineage Specification. Cell 2006;126:677-89.

[153] Pelham Jr RJ, Wang YL. Cell locomotion and focal adhesions are regulated by substrate flexibility. Proceedings of the National Academy of Sciences of the United States of America 1997;94:13661-5.

[154] Kim TI, Jang JH, Kim HW, Knowles JC, Ku Y. Biomimetic approach to dental implants. Current Pharmaceutical Design 2008;14:2201-11.

[155] Wang HB, Dembo M, Wang YL. Substrate flexibility regulates growth and apoptosis of normal but not transformed cells. American Journal of Physiology - Cell Physiology 2000;279:C1345-C50.

[156] Chen G, Wen X, Zhang N. Corrosion resistance and ion dissolution of titanium with different surface microroughness. Bio-Medical Materials and Engineering 1998;8:61-74.

255

[157] Anselme K, Davidson P, Popa AM, Giazzon M, Liley M, Ploux L. The interaction of cells and bacteria with surfaces structured at the nanometre scale. Acta Biomaterialia 2010;6:3824-46.

[158] Jung YL, Donahue HJ. Cell sensing and response to micro- and nanostructured surfaces produced by chemical and topographic patterning. Tissue Engineering 2007;13:1879-91.

[159] Lord MS, Foss M, Besenbacher F. Influence of nanoscale surface topography on protein adsorption and cellular response. Nano Today 2010;5:66-78.

[160] Li JR, Shi L, Deng Z, Lo SH, Liu GY. Nanostructures of designed geometry and functionality enable regulation of cellular signaling processes. Biochemistry 2012;51:5876-93.

[161] Yim EKF, Leong KW. Significance of synthetic nanostructures in dictating cellular response. Nanomedicine: Nanotechnology, Biology and Medicine 2005;1:10-21.

[162] Kim WG, Choe HC. Nanostructure and corrosion behaviors of nanotube formed Ti-Zr alloy. Transactions of Nonferrous Metals Society of China (English Edition) 2009;19:1005-8.

[163] Jeong Y-H, Lee K, Choe H-C, Ko Y-M, Brantley WA. Nanotube formation and morphology change of Ti alloys containing Hf for dental materials use. Thin Solid Films 2009;517:5365-9.

[164] Jang SH, Choe HC, Ko YM, Brantley WA. Electrochemical characteristics of nanotubes formed on Ti-Nb alloys. Thin Solid Films 2009;517:5038-43.

[165] Capellato P, Smith BS, Popat KC, Claro APRA. Fibroblast functionality on novel Ti30Ta nanotube array. Materials Science and Engineering C 2012;32:2060-7.

[166] Ding D, Ning C, Huang L, Jin F, Hao Y, Bai S, Li Y, Li M, Mao D. Anodic fabrication and bioactivity of Nb-doped TiO2 nanotubes. Nanotechnology 2009;20:305103 (6pp).

[167] Choe HC. Nanotubular surface and morphology of Ti-binary and Ti-ternary alloys for biocompatibility. Thin Solid Films 2011;519:4652-7.

[168] Li Z, Ning C, Ding D, Liu H, Huang L. Biological properties of Ti-Nb-Zr-O nanostructures grown on Ti35Nb5Zr alloy. Journal of Nanomaterials 2012;2012:834042-9.

[169] Mîndroiu M, Pirvu C, Ion R, Demetrescu I. Comparing performance of nanoarchitectures fabricated by Ti6Al7Nb anodizing in two kinds of electrolytes. Electrochimica Acta 2010;56:193-202.

[170] Mei S, Zhao L, Wang W, Ma Q, Zhang Y. Biomimetic titanium alloy with sparsely distributed nanotubes could enhance osteoblast functions. Advanced Engineering Materials 2012;14:B166-B74.

[171] Oh S, Brammer KS, Li YSJ, Teng D, Engler AJ, Chien S, Jin S. Reply to von der Mark et al.: Looking further into the effects of nanotube dimension on stem cell fate. Proceedings of the National Academy of Sciences of the United States of America 2009;106.

[172] Oh S, Jin S. Titanium oxide nanotubes with controlled morphology for enhanced bone growth. Materials Science and Engineering: C 2006;26:1301-6.

[173] Wang N, Li H, Lü W, Li J, Wang J, Zhang Z, Liu Y. Effects of TiO2 nanotubes with different diameters on gene expression and osseointegration of implants in minipigs. Biomaterials 2011;32:6900-11.

256

[174] Li Y, Ding D, Ning C, Bai S, Huang L, Li M, Mao D. Thermal stability and in vitro bioactivity of Ti-Al-V-O nanostructures fabricated on Ti6Al4V alloy. Nanotechnology 2009;20:065708 (6pp).

[175] Choe HC, Kim WG, Jeong YH. Surface characteristics of HA coated Ti-30Ta-xZr and Ti-30Nb-xZr alloys after nanotube formation. Surface and Coatings Technology 2010;205:S305-S11.

[176] Jeong YH, Kim WG, Park GH, Choe HC, Ko YM. Surface characteristics of HA coated Ti-Hf binary alloys after nanotube formation. Transactions of Nonferrous Metals Society of China (English Edition) 2009;19:852-6.

[177] Yu WQ, Zhang YL, Jiang XQ, Zhang FQ. In vitro behavior of MC3T3-E1 preosteoblast with different annealing temperature titania nanotubes. Oral Diseases 2010;16:624-30.

[178] Mazǎre A, Voicu G, Truscǎ R, Ioniţǎ D. Heat treatment of TiO2 nanotubes, a way to significantly change their behaviour. UPB Scientific Bulletin, Series B: Chemistry and Materials Science 2011;73:97-108.

[179] Aquino JM, Rocha-Filho RC, Bocchi N, Biaggio SR. Microwave-assisted crystallization into anatase of amorphous TiO2 nanotubes electrochemically grown on a Ti substrate. Materials Letters 2014;126:52-4.

[180] Shin DH, Shokuhfar T, Choi CK, Lee SH, Friedrich C. Wettability changes of TiO2 nanotube surfaces. Nanotechnology 2011;22:315704 (7pp).

[181] Hamlekhan A, Butt A, Patel S, Royhman D, Takoudis C, Sukotjo C, Yuan J, Jursich G, Mathew MT, Hendrickson W, Virdi A, Shokuhfar T. Fabrication of anti-aging TiO2 nanotubes on biomedical Ti alloys. PLoS ONE 2014;9:e96213.

[182] Khor KA, Dong ZL, Quek CH, Cheang P. Microstructure investigation of plasma sprayed HA/Ti6Al4V composites by TEM. Materials Science and Engineering A 2000;281:221-8.

[183] Kim WG, Choe HC. Surface characteristics of hydroxyapatite/titanium composite layer on the Ti-35Ta-xZr surface by RF and DC sputtering. Thin Solid Films 2011;519:7045-9.

[184] Lee K, Choe HC, Kim BH, Ko YM. The biocompatibility of HA thin films deposition on anodized titanium alloys. Surface and Coatings Technology 2010;205:S267-S70.

[185] Choe HC. Photofunctionalization of EB-PVD HA-coated nano-pore surface of Ti-30Nb-xZr alloy for dental implants. Surface and Coatings Technology 2013;228:S470-S6.

[186] Ma J, Liang CH, Kong LB, Wang C. Colloidal characterization and electrophoretic deposition of hydroxyapatite on titanium substrate. Journal of Materials Science: Materials in Medicine 2003;14:797-801.

[187] Parcharoen Y, Kajitvichyanukul P, Sirivisoot S, Termsuksawad P. Hydroxyapatite electrodeposition on anodized titanium nanotubes for orthopedic applications. Applied Surface Science 2014;311:54-61.

[188] Chen J, Zhang Z, Ouyang J, Chen X, Xu Z, Sun X. Bioactivity and osteogenic cell response of TiO2 nanotubes coupled with nanoscale calcium phosphate via ultrasonification-assisted electrochemical deposition. Applied Surface Science 2014;305:24-32.

257

[189] Kodama A, Bauer S, Komatsu A, Asoh H, Ono S, Schmuki P. Bioactivation of titanium surfaces using coatings of TiO2 nanotubes rapidly pre-loaded with synthetic hydroxyapatite. Acta Biomaterialia 2009;5:2322-30.

[190] Kim EJ, Jeong YH, Choe HC, Brantley WA. Surface phenomena of HA/TiN coatings on the nanotubular-structured beta Ti-29Nb-5Zr alloy for biomaterials. Applied Surface Science 2012;258:2083-7.

[191] Jeong YH, Choe HC, Eun SW. Hydroxyapatite coating on the Ti-35Nb-xZr alloy by electron beam-physical vapor deposition. Thin Solid Films 2011;519:7050-6.

[192] Raja KS, Misra M, Paramguru K. Deposition of calcium phosphate coating on nanotubular anodized titanium. Materials Letters 2005;59:2137-41.

[193] Kar A, Raja KS, Misra M. Electrodeposition of hydroxyapatite onto nanotubular TiO2 for implant applications. Surface and Coatings Technology 2006;201:3723-31.

[194] Wang LN, Luo JL. Formation of hydroxyapatite coating on anodic titanium dioxide nanotubes via an efficient dipping treatment. Metallurgical and Materials Transactions A: Physical Metallurgy and Materials Science 2011;42:3255-64.

[195] Zhao L, Mei S, Chu PK, Zhang Y, Wu Z. The influence of hierarchical hybrid micro/nano-textured titanium surface with titania nanotubes on osteoblast functions. Biomaterials 2010;31:5072-82.

[196] Xiao J, Zhou H, Zhao L, Sun Y, Guan S, Liu B, Kong L. The effect of hierarchical micro/nanosurface titanium implant on osseointegration in ovariectomized sheep. Osteoporosis International 2011;22:1907-13.

[197] Sul YT. Electrochemical growth behavior, surface properties, and enhanced in vivo bone response of Tio2 nanotubes on microstructured surfaces of blasted, screw-shaped titanium implants. International Journal of Nanomedicine 2010;5:87-100.

[198] Von Wilmowsky C, Bauer S, Roedl S, Neukam FW, Schmuki P, Schlegel KA. The diameter of anodic TiO2 nanotubes affects bone formation and correlates with the bone morphogenetic protein-2 expression in vivo. Clinical Oral Implants Research 2012;23:359-66.

[199] Popat KC, Leoni L, Grimes CA, Desai TA. Influence of engineered titania nanotubular surfaces on bone cells. Biomaterials 2007;28:3188-97.

[200] Bjursten LM, Rasmusson L, Oh S, Smith GC, Brammer KS, Jin S. Titanium dioxide nanotubes enhance bone bonding in vivo. Journal of Biomedical Materials Research - Part A 2010;92:1218-24.

[201] Peremarch CPJ, Tanoira RP, Arenas MA, Matykina E, Conde A, De Damborenea JJ, Barrena EG, Esteban J. Bacterial adherence to anodized titanium alloy. Journal of Physics: Conference Series 2010;252.

[202] Yu WQ, Jiang XQ, Xu L, Zhao YF, Zhang FQ, Cao X. Osteogenic gene expression of canine bone marrow stromal cell and bacterial adhesion on titanium with different nanotubes. Journal of Biomedical Materials Research - Part B Applied Biomaterials 2011;99 B:207-16.

[203] Qi YM, Ma BH, Geng YJ, Deng JY, Cui CX. Nano-surface modifacation of biomedical β Titanium alloy for dental implant and its antibacterial property. Advanced Materials Research 2014;904: 142-5.

[204] Grigorescu S, Ungureanu C, Kirchgeorg R, Schmuki P, Demetrescu I. Various sized nanotubes on TiZr for antibacterial surfaces. Applied Surface Science 2013;270:190-6.

258

[205] Zhao L, Wang H, Huo K, Cui L, Zhang W, Ni H, Zhang Y, Wu Z, Chu PK. Antibacterial nano-structured titania coating incorporated with silver nanoparticles. Biomaterials 2011;32:5706-16.

[206] Lin C, Tang P, Zhang W, Wang Y, Zhang B, Wang H, Zhang L. Effect of superhydrophobic surface of titanium on staphylococcus aureus adhesion. Journal of Nanomaterials 2011;2011: 178921 (8 pp).

[207] Ercan B, Taylor E, Alpaslan E, Webster TJ. Diameter of titanium nanotubes influences anti-bacterial efficacy. Nanotechnology 2011;22:295102 (11pp).

[208] Mendenhall W, Sincich T. Continuous random variables. In: S Yagan, editor. Statistics for engineering and the sciences. 5th ed. ed. New Jersey, USA: Prentice Hall; 2007. p. 168-210.

[209] Rinne H. Definition and properties of the WEIBULL distribution. The Weibull Distribution: A Handbook. United Kingdom: Chapman and Hall/CRC; 2008. p. 27-97.

[210] Gadelmawla ES, Koura MM, Maksoud TMA, Elewa IM, Soliman HH. Roughness parameters. Journal of Materials Processing Technology 2002;123:133-45.

[211] Owens DK, Wendt RC. Estimation of the surface free energy of polymers. Journal of Applied Polymer Science 1969;13:1741-7.

[212] Dann JR. Forces involved in the adhesive process. I. Critical surface tensions of polymeric solids as determined with polar liquids. Journal of Colloid and Interface Science 1970;32:302-20.

[213] Oyane A, Onuma K, Ito A, Kim HM, Kokubo T, Nakamura T. Formation and growth of clusters in conventional and new kinds of simulated body fluids. Journal of Biomedical Materials Research - Part A 2003;64:339-48.

[214] Kokubo T, Takadama H. How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 2006;27:2907-15.

[215] Fischer-Cripps AC. Critical review of analysis and interpretation of nanoindentation test data. Surface and Coatings Technology 2006;200:4153-65.

[216] Oliver WC, Pharr GM. Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinements to methodology. Journal of Materials Research 2004;19:3-20.

[217] Li H, Khor KA, Cheang P. Impact formation and microstructure characterization of thermal sprayed hydroxyapatite/titania composite coatings. Biomaterials 2003;24:949-57.

[218] Halka M, Nordstrom B. Transition Metals. New York: Facts On File; 2010. [219] Campbell FC. Elements of Metallurgy and Engineering Alloys. Materials Park, OH,

USA: ASM International; 2008. [220] Park HH, Park IS, Kim KS, Jeon WY, Park BK, Kim HS, Bae TS, Lee MH.

Bioactive and electrochemical characterization of TiO2 nanotubes on titanium via anodic oxidation. Electrochimica Acta 2010;55:6109-14.

[221] Oh S, Brammer KS, Moon K-S, Bae J-M, Jin S. Influence of sterilization methods on cell behavior and functionality of osteoblasts cultured on TiO2 nanotubes. Materials Science and Engineering: C 2011;31:873-9.

[222] H. Nielsen R, H. Schlewitz J, Nielsen H, Updated by S. Zirconium and Zirconium Compounds. Kirk-Othmer Encyclopedia of Chemical Technology: John Wiley & Sons, Inc.; 2000.

259

[223] Webster RT, Albany TWC. Zirconium-Properties of Pure Metals, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials. ASM Handbook: ASM International 1990. p. 1099–201.

[224] Yamaguchi T. Application of ZrO2 as a catalyst and a catalyst support. Catalysis Today 1994;20:199-217.

[225] Bethke KA, Kung MC, Yang B, Shah M, Alt D, Li C, Kung HH. Metal oxide catalysts for lean NOx reduction. Catalysis Today 1995;26:169-83.

[226] Hahn R, Berger S, Schmuki P. Bright visible luminescence of self-organized ZrO2 nanotubes. Journal of Solid State Electrochemistry 2010;14:285-8.

[227] Agarwal P, Paramasivam I, Shrestha NK, Schmuki P. MoO3 in self-organized TiO2 nanotubes for enhanced photocatalytic activity. Chemistry - An Asian Journal 2010;5:66-9.

[228] Paramasivam I, Nah YC, Das C, Shrestha NK, Schmuki P. WO3/TiO2 nanotubes with strongly enhanced photocatalytic activity. Chemistry - A European Journal 2010;16:8993-7.

[229] Liu G, Lu H, Chen Z, Li F, Wang L, Watts J, Lu GQ, Cheng HM. Ti-Zr-O nanotube arrays with controlled morphology, crystal structure and optical properties. Journal of Nanoscience and Nanotechnology 2009;9:6501-10.

[230] Kobayashi E, Matsumoto S, Doi H, Yoneyama T, Hamanaka H. Mechanical properties of the binary titanium-zirconium alloys and their potential for biomedical materials. Journal of Biomedical Materials Research 1995;29:943-50.

[231] Li Y, Wong C, Xiong J, Hodgson P, Wen C. In vitro cytotoxicity of binary TI alloys for bone implants. Gold Coast, QLD2009. p. 295-8.

[232] Kim WG, Choe HC, Ko YM, Brantley WA. Electrochemical characteristics of nanotube formed Ti-Zr alloy. Quebec City, QC2008. p. 462-5.

[233] Kim WG, Choe HC, Ko YM, Brantley WA. Nanotube morphology changes for Ti-Zr alloys as Zr content increases. Thin Solid Films 2009;517:5033-7.

[234] Minagar S, Berndt CC, Wang J, Ivanova E, Wen C. A review of the application of anodization for the fabrication of nanotubes on metal implant surfaces. Acta Biomaterialia 2012;8:2875-88.

[235] Wang Y, Wen C, Hodgson P, Li Y. Biocompatibility of TiO2 nanotubes with different topographies. Journal of Biomedical Materials Research - Part A 2013.

[236] Chemical Properties of Materials. In: CRC Materials Science and Engineering Handbook, ed. J. F. S. a. W. Alexander, CRC Press 2000, Boca Raton, 2000, ch. 9, tbl. 309.

[237] Thermodynamic and Kinetic Data. In: CRC Materials Science and Engineering Handbook, ed. J. F. S. a. W. Alexander, CRC Press 2000, Boca Raton, 2000, ch. 4, tbl. 67.

[238] Taveira LV, Macák JM, Tsuchiya H, Dick LFP, Schmuki P. Initiation and growth of self-organized TiO2 nanotubes anodically formed in NH4F/(NH4)2SO4 electrolytes. Journal of the Electrochemical Society 2005;152:B405-B10.

[239] Madou MJ, Kinoshita K. Electrochemical measurements on metal oxide electrodes-I. Zirconium dioxide. Electrochimica Acta 1984;29:411-7.

[240] Minagar S, Berndt CC, Gengenbach T, Wen C. Fabrication and characterization of TiO2-ZrO2-ZrTiO4 nanotubes on TiZr alloy manufactured via anodization. Journal of Materials Chemistry B 2014;2:71-83.

260

[241] EL-010 SASNZC. Guidance on the measurement of wettability of insulator surfaces. GPO Box 5420, Sydney, NSW 2001, Australia: Standard Australia; 2005. p. 15.

[242] Webb HK, Hasan J, Truong VK, Crawford RJ, Ivanova EP. Nature inspired structured surfaces for biomedical applications. Current Medicinal Chemistry 2011;18:3367-75.

[243] Ikawa H, Yamada T, Kojima K, Matsumoto S. X-Ray Photoelectron-Spectroscopy Study of High-Temperature and Low-Temperature Forms of Zirconium Titanate. Journal of the American Ceramic Society 1991;74:1459-62.

[244] Tsai CC, Teng H. Regulation of the physical characteristics of titania nanotube aggregates synthesized from hydrothermal treatment. Chemistry of Materials 2004;16:4352-8.

[245] Tian ZR, Voigt JA, Liu J, McKenzie B, Xu H. Large oriented arrays and continuous films of TiO2-based nanotubes. Journal of the American Chemical Society 2003;125:12384-5.

[246] Sander MS, Côté MJ, Gu W, Kile BM, Tripp CP. Template-assisted fabrication of dense, aligned arrays of titania nanotubes with well-controlled dimensions on substrates. Advanced Materials 2004;16:2052-7.

[247] Macak J, Schmuki P. Anodic growth of self-organized anodic TiO2 nanotubes in viscous electrolytes. Electrochimica Acta 2006;52:1258-64.

[248] Kim D, Schmidt-Stein F, Hahn R, Schmuki P. Gravity assisted growth of self-organized anodic oxide nanotubes on titanium. Electrochemistry Communications 2008;10:1082-6.

[249] Yasuda K, Macak JM, Berger S, Ghicov A, Schmuki P. Mechanistic aspects of the self-organization process for oxide nanotube formation on valve metals. Journal of the Electrochemical Society 2007;154:C472-C8.

[250] Weiner S, Traub W, Wagner HD. Lamellar bone: Structure-function relations. Journal of Structural Biology 1999;126:241-55.

[251] Kim HM, Himeno T, Kawashita M, Kokubo T, Nakamura T. The mechanism of biomineralization of bone-like apatite on synthetic hydroxyapatite: an in vitro assessment. Journal of The Royal Society Interface 2004;1:17-22.

[252] Chen X, Nouri A, Li Y, Lin J, Hodgson PD, Wen Ce. Effect of surface roughness of Ti, Zr, and TiZr on apatite precipitation from simulated body fluid. Biotechnology and Bioengineering 2008;101:378-87.

[253] Michelle Grandin H, Berner S, Dard M. A review of Titanium Zirconium (TiZr) alloys for use in endosseous dental implants. Materials 2012;5:1348-60.

[254] Kunze J, Muller L, Macak J, Greil P, Schmuki P, Muller F. Time-dependent growth of biomimetic apatite on anodic TiO2 nanotubes. Electrochimica Acta 2008;53:6995-7003.

[255] Points of Zero Charge. Chemical Properties of Material Surfaces: CRC Press; 2001. p. 731-44.

[256] Surface Charging in Absence of Strongly Adsorbing Species. Chemical Properties of Material Surfaces: CRC Press; 2001. p. 65-309.

[257] Ma Q, Li M, Hu Z, Chen Q, Hu W. Enhancement of the bioactivity of titanium oxide nanotubes by precalcification. Materials Letters 2008;62:3035-8.

[258] Wang XJ, Chen XB, Hodgson PD, Wen CE. Elastic modulus and hardness of cortical and trabecular bovine bone measured by nanoindentation. Transactions of Nonferrous Metals Society of China (English Edition) 2006;16:s744-s8.

261

[259] Anselme K, Bigerelle M. Statistical demonstration of the relative effect of surface chemistry and roughness on human osteoblast short-term adhesion. Journal of Materials Science: Materials in Medicine 2006;17:471-9.

[260] Zhang W, Xi Z, Li G, Wang Q, Tang H, Liu Y, Zhao Y, Jiang L. Highly ordered coaxial bimodal nanotube arrays prepared by self-organizing anodization on Ti alloy. Small 2009;5:1742-6.

[261] Popat KC, Chalvanichkul KI, Barnes GL, Latempa Jr TJ, Grimes CA, Desai TA. Osteogenic differentiation of marrow stromal cells cultured on nanoporous alumina surfaces. Journal of Biomedical Materials Research - Part A 2007;80:955-64.

[262] Anselme K, Bigerelle M. Topography effects of pure titanium substrates on human osteoblast long-term adhesion. Acta Biomaterialia 2005;1:211-22.

[263] Anselme K, Linez P, Bigerelle M, Le Maguer D, Le Maguer A, Hardouin P, Hildebrand HF, Iost A, Leroy JM. The relative influence of the topography and chemistry of TiAl6V4 surfaces on osteoblastic cell behaviour. Biomaterials 2000;21:1567-77.

[264] Anselme K, Ponche A, Bigerelle M. Relative influence of surface topography and surface chemistry on cell response to bone implant materials. Part 2: Biological aspects. Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine 2010;224:1487-507.

[265] Balaur E, Macak JM, Tsuchiya H, Schmuki P. Wetting behaviour of layers of TiO2 nanotubes with different diameters. Journal of Materials Chemistry 2005;15:4488.

[266] Park J, Bauer S, Schlegel KA, Neukam FW, Von Mark KD, Schmuki P. TiO2 nanotube surfaces: 15 nm - an optimal length scale of surface topography for cell adhesion and differentiation. Small 2009;5:666-71.

[267] Jha H, Hahn R, Schmuki P. Ultrafast oxide nanotube formation on TiNb, TiZr and TiTa alloys by rapid breakdown anodization. Electrochimica Acta 2010;55:8883-7.

[268] Grigorescu S, Pruna V, Titorencu I, Jinga VV, Mazare A, Schmuki P, Demetrescu I. The two step nanotube formation on TiZr as scaffolds for cell growth. Bioelectrochemistry 2014;98:39-45.

[269] Inc. VI. Vision for profiler. 4.20 ed. 100 Sunnyside Blvd., Woodbury, NY 11797: Veeco Instruments Inc.; 2009.

[270] Webster TJ, Ergun C, Doremus RH, Siegel RW, Bizios R. Enhanced functions of osteoblasts on nanophase ceramics. Biomaterials 2000;21:1803-10.

[271] Peng P, Kumar S, Voelcker NH, Szili E, Smart RSC, Griesser HJ. Thin calcium phosphate coatings on titanium by electrochemical deposition in modified simulated body fluid. Journal of Biomedical Materials Research - Part A 2006;76:347-55.

[272] Anselme K, Bigerelle M. Role of materials surface topography on mammalian cell response. International Materials Reviews 2011;56:243-66.

[273] Burridge K, Chrzanowska-Wodnicka M. Focal adhesions, contractility, and signaling. Annual Review of Cell and Developmental Biology; 1996. p. 463-519.

[274] Pierres A, Benoliel AM, Bongrand P. Cell fitting to adhesive surfaces: A prerequisite to firm attachment andsubsequent events. European Cells and Materials 2002;3:31-45.

[275] Ishikawa M, Sugimoto K. Multiple layer protective coating composed of Ta2O5, Cr2O3 and Al2O3 produced by the MO-CVD technique. Corrosion Engineering 1989;38:619-30.

262

[276] Arnould C, Volcke C, Lamarque C, Thiry PA, Delhalle J, Mekhalif Z. Titanium modified with layer-by-layer sol–gel tantalum oxide and an organodiphosphonic acid: A coating for hydroxyapatite growth. Journal of Colloid and Interface Science 2009;336:497-503.

[277] Terui H, Kobayashi M. Refractive-index-adjustable SiO2-Ta2O5 films for integrated optical circuits. Applied Physics Letters 1978;32:666-8.

[278] Cava RF, Peck WF, Krajewski JJ. Enhancement of the dielectric-constant of Ta2O5 through substitution with TiO2. Nature 1995;377:215-7.

[279] Sieber I, Kannan B, Schmuki P. Self-assembled porous tantalum oxide prepared in H2SO4/HF electrolytes. Electrochemical and Solid State Letters 2005;8:J10-J2.

[280] Chambon L, Maleysson C, Pauly A, Germain JP, Demarne V, Grisel A. Investigation, for NH3 gas sensing applications, of the Nb2O5 semiconducting oxide in the presence of interferent species such as oxygen and humidity. Sensors and Actuators B: Chemical 1997;45:107-14.

[281] Tanabe K, Okazaki S. Various reactions catalyzed by niobium compounds and materials. Applied Catalysis A, General 1995;133:191-218.

[282] Pawlicka A, Atik M, Aegerter MA. Synthesis of multicolor Nb2O5 coatings for electrochromic devices. Thin Solid Films 1997;301:236-41.

[283] Choi J, Lim JH, Lee J, Kim KJ. Porous niobium oxide films prepared by anodization–annealing–anodization. Nanotechnology 2007;18:055603 (6pp).

[284] Stefanov P, Stoychev D, Stoycheva M, Ikonomov J, Marinova T. XPS and SEM characterization of zirconia thin films prepared by electrochemical deposition. Surface and Interface Analysis 2000;30:628-31.

[285] Kim HM, Kaneko H, Kokubo T, Miyazaki T, Nakamura T. Mechanism of apatite formation on bioactive tantalum metal in simulated body fluid. Key Engineering Materials 2003; 240-242:11-14.

[286] Nagarajan S, Raman V, Rajendran N. Synthesis and electrochemical characterization of porous niobium oxide coated 316L SS for orthopedic applications. Materials Chemistry and Physics 2010;119:363-6.

[287] Wang XJ, Xiong JY, Li YC, Hodgson PD, Wen CE. Apatite formation on nano-structured titanium and niobium surface. Materials Science Forum 2009;614:85-92.

[288] Godley R, Starosvetsky D, Gotman I. Bonelike apatite formation on niobium metal treated in aqueous NaOH. Journal of Materials Science: Materials in Medicine 2004;15:1073-7.

[289] Guo L, Zhao J, Wang X, Xu R, Lu Z, Li Y. Bioactivity of zirconia nanotube arrays fabricated by electrochemical anodization. Materials Science and Engineering C 2009;29:1174-7.

[290] Guo L, Zhao J, Wang X, Xu X, Liu H, Li Y. Structure and bioactivity of zirconia nanotube arrays fabricated by anodization. International Journal of Applied Ceramic Technology 2009;6:636-41.

[291] Wang LN, Adams A, Luo JL. Enhancement of the capability of hydroxyapatite formation on Zr with anodic ZrO2 nanotubular arrays via an effective dipping pretreatment. Journal of Biomedical Materials Research - Part B Applied Biomaterials 2011;99 B:291-301.

[292] Yoo JE, Choi J. Electrochemical surface enlargement of a niobium foil for electrolytic capacitor applications. Electrochemistry Communications 2011;13:298-301.

Surface nanoengineering of titanium alloys for …...Surface Nanoengineering of Titanium Alloys for...

Documents

Transcript of Surface nanoengineering of titanium alloys for …...Surface Nanoengineering of Titanium Alloys for...