Materials Science and Engineering R 87 (2015) 1–57

Contents lists available at ScienceDirect

Materials Science and Engineering R

journal homepage: www.elsevier.com/locate/mser

Metallic implant biomaterials

Qizhi Chen *, George A. Thouas
Department of Materials Engineering, Monash University, Clayton, Victoria 3800, Australia

A R T I C L E I N F O A B S T R A C T

Article history: Human tissue is structured mainly of self-assembled polymers (proteins) and ceramics (bone minerals),
Available online with metals present as trace elements with molecular scale functions. However, metals and their alloys
have played a predominant role as structural biomaterials in reconstructive surgery, especially
Keywords: orthopedics, with more recent uses in non-osseous tissues, such as blood vessels. With the successful
Metallic biomaterials routine use of a large variety of metal implants clinically, issues associated with long-term maintenance
Biocompatibility of implant integrity have also emerged. This review focuses on metallic implant biomaterials, identifying
Corrosion and discussing critical issues in their clinical applications, including the systemic toxicity of released
Mechanical properties metal ions due to corrosion, fatigue failure of structural components due to repeated loading, and
Medical devices
wearing of joint replacements due to movement. This is followed by detailed reviews on specific metallic
biomaterials made from stainless steels, alloys of cobalt, titanium and magnesium, as well as shape
memory alloys of nickel–titanium, silver, tantalum and zirconium. For each, the properties that affect
biocompatibility and mechanical integrity (especially corrosion fatigue) are discussed in detail. Finally,
the most critical challenges for metallic implant biomaterials are summarized, with emphasis on the
most promising approaches and strategies.
ß 2014 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1. Historical development of metallic biomaterials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2. Definitions of biomedical materials, biomaterials and biological materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3. Definition of biocompatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4. Classification of medical devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2. Essential considerations in design of metallic biomaterials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1. Biocompatibility of elements and selection of alloying elements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2. Corrosion of implant materials in the body. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3. Mechanical working conditions in the human body . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4. Wear of joint implants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5. Osseo-integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.6. The objectives of the rest of the review article . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3. Stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1. Alloying chemistry: corrosion resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2. Biocompatibility of alloying elements used in stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.1. Iron. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.2. Chromium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.3. Nickel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.3. Biocompatibility of 316L stainless steel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.4. Mechanical properties of implant-grade stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.4.1. General mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

* Corresponding author.
E-mail address: qzchen202@gmail.com (Q. Chen).

http://dx.doi.org/10.1016/j.mser.2014.10.001
0927-796X/ß 2014 Elsevier B.V. All rights reserved.
2 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

3.4.2. Fatigue properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

3.5. Medical applications of stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.6. Current issues and challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4. Cobalt-based alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.1. Alloying chemistry: corrosion resistance . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2. Biocompatibility of alloying elements: Co, Mo and W . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.2.1. Cobalt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2.2. Molybdenum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2.3. Tungsten . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.3. Biocompatibility of cobalt alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.4. Mechanical properties of medical-grade cobalt alloys . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.4.1. General mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.4.2. Fatigue properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.5. Medical applications of cobalt alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.6. Current issues and challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5. Titanium alloys used as orthopedic implants . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5.1. Alloying chemistry: microstructure design . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5.2. Grading and classification of titanium and its alloys . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3. Biocompatibility of alloying elements . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3.1. Titanium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3.2. Vanadium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3.3. Aluminum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.3.4. Niobium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.4. Biocompatibility of titanium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.5. Microstructure and general mechanical properties . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.5.1. a titanium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.5.2. a–b titanium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.5.3. b titanium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.6. Fatigue properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.7. Wear resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.8. Bone bonding mechanisms of Ti alloys . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.9. Clinical applications of titanium alloys . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.10. Current issues with titanium alloys as implant biomaterials . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.11. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6. Stainless steels, cobalt and titanium alloys in total joint replacement . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.1. Historical perspective of the development of total joint replacements [180] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.2. Materials used in total joint replacements . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6.3. Current issues and challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7. Magnesium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
7.1. Three generations of biomaterials in terms of clinical outcomes . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
7.2. Rationale of developing Mg-alloys as medical implants . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
7.3. Corrosion of Mg alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
7.4. Metallurgical roles of alloying elements in magnesium alloys . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
7.5. Biocompatibility and toxicity of alloying elements in Mg alloys . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
7.5.1. Magnesium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
7.5.2. Calcium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
7.5.3. Manganese . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.5.4. Copper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.5.5. Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.6. Mechanical properties of Mg-alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.6.1. Mg–Zn-based alloys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.6.2. Mg–Ca-based alloys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.7. Potential applications and challenges of magnesium alloys . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.8. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
8. NiTi shape-memory alloys [102] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
8.1. Corrosion of NiTi alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
8.2. In vitro evaluation of biocompatibility of NiTi alloy . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
8.3. In vivo evaluation of biocompatibility of NiTi in animal models . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
8.4. In vivo trials of NiTi implants in human [409] . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
8.5. Biocompatibility of NiTi wires as stents (filters) . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
8.6. Current and potential applications of NiTi . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
8.6.1. Self-expandable stents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
8.6.2. Orthopedic and orthodontic applications . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.7. Mechanical properties of NiTi alloy [43] . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.7.1. General mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.7.2. Fatigue properties of NiTi alloys . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.8. Issues and challenges of NiTi implants . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.9. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
9. Tantalum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
9.1. Corrosion and biocompatibility of tantalum . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 3

9.2. Clinical applications of tantalum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

10. Zirconium alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
10.1. Corrosion of zirconium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
10.2. Biocompatibility of zirconium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
10.3. Clinical application of zirconium alloy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
11. Silver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
11.1. Biocompatibility of silver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
11.2. Medical application of Ag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
12. Metals used as medical electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
13. Metallic materials used in orthodontic implants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
14. Summary and remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

1. Introduction in the vast majority of orthopedic medical devices available

commercially. These materials can be categorized in the following
1.1. Historical development of metallic biomaterials four groups based on the major alloying element (Table 1):
stainless steels, cobalt-based alloys, titanium-based alloys and
The use of metallic materials for medical implants can be traced miscellaneous others (e.g. NiTi and alloys of Mg and Ta) [6]. A
back to the 19th century, leading up to the era when the metal variety of medical implants made of the metallic materials in the
industry began to expand during the Industrial Revolution [1]. The first three groups have been approved by the United States Food
development of metallic implants was primarily driven by the and Drug Administration (FDA) [7] and are routinely used in
demands for approaches to bone repair, typically internal fracture orthopedic practice [2]. Figs. 1 and 2 depict some of the typical
fixation of long bones. However, almost no attempts of implanting clinical application scenarios. The materials of the last group have
metallic devices, such as spinal wires and bone pins made from recently been developed because of their unique material
iron, gold or silver, were successful until Lister’s aseptic surgical properties (such as the shape memory of NiTi and degradability
technique was implemented in the 1860s [1]. Since then, metallic of Mg alloys) that could potentially meet more specialized tissue
materials have predominated in orthopedic surgery, playing a requirements (Fig. 2) [8]. However, some medical implants made
major role in most orthopedic devices, including temporary with these newly developed alloys are not FDA approved yet,
devices (e.g. bone plates, pins and screws) and permanent implants primarily due to the significant issues associated with biocompat-
(e.g. total joint replacements) [2]. Concurrently, metals also found ibility [8], the principle requirement for clinical application of any
applications in dental and orthodontic practice, including tooth biomedical implant. Clinical applications and current status of the
fillings and roots [3]. Recently, increasing research effort in four classes of metallic biomaterial are summarized in Table 1.
metallic biomaterials has been invested in application of the non-
conventional reconstructive surgery of hard tissues/organs, such as 1.2. Definitions of biomedical materials, biomaterials and biological
the application of NiTi shape memory alloys as vascular stents [4] materials
and the development of new magnesium-based alloys for bone
tissue engineering and regeneration [5]. The common feature of biomaterials is that they are used in
Despite the large number of metals and alloys able to be intimate contact with the living body. In materials science, a
produced in industry, only a few are biocompatible and capable of biomaterial is defined as ‘‘a substance that has been engineered to
long-term success as an implant material. These form components take a form which, alone or as part of a complex system, is used to

Table 1
Four categories of metallic biomaterials and their primary applications as implants.

Type Primary utilizationsa Status of applications Ref.

Stainless steels 1. Temporary devices (fracture plates, screws, hip nails, etc.) (Class II) Routinely applied [9]
2. Total hip replacements (Class II)

Co-based alloys 3. Total joint replacements (wrought alloys) (Class II) Routinely applied [9]
4. Dentistry castings (Class II)

Ti-based alloys 5. Stem and cup of total hip replacements with CoCrMo or ceramic Routinely applied [9]
femoral heads (Class II)
6. Other permanent devices (nails, pacemakers) (Class III)

Miscellaneous others
NiTi 1. Orthodontic dental archwires (Class I) FDA approved [10]
2. Vascular stents (Class III) FDA approved
3. Vena cava filter (Class II) FDA approved
4. Intracranial aneurysm clips (Class II) FDA approved
5. Contractile artificial muscles for an artificial heart (Class III) Research
6. Catheter guide wires (Class II) FDA approved
7. Orthopedic staples (Class I) FDA approved

Mg Biodegradable orthopedic implants (Class III) Animal trial [11,12]

Ta 8. Wire sutures for plastic surgery and neurosurgery (Class III) FDA approved [13]
9. A radiographic marker (Class II) FDA approved
a
Class I–III are defined in Table 2.
4 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Fig. 1. (a) The Harrington rod, a stainless steel surgical device. (b) The stem of a total hip replacement, usually made from either stainless steel, cobalt- or titanium-based
alloys.

direct, by control of interactions with components of living
systems, the course of any therapeutic or diagnostic procedure’’. In
another words, a biomaterial is any biocompatible material,
natural or man-made, which is used to replace or assist part of an
organ or tissue, while in intimate contact with it. It should be
mentioned that the prefix ‘‘bio’’ of biomaterials refers to
‘‘biocompatible’’, rather than ‘‘biological’’ or ‘‘biomedical’’ as is often
misunderstood.
However, the term biomaterial is used within different
definition boundaries in the scientific and legal communities. In
the legal field, medical devices are defined as, ‘‘any instrument,
apparatus, implement, machine, appliance, implant, in vitro
reagent or calibrator, software, material or other similar or related
articles, intended by the manufacture to be used, alone or in
combination, for human beings for one or more of the specific
purposes of diagnosis, prevention, monitoring, treatment, investi-
gation, supporting or sustaining life, control of conception, and
disinfection of medical devices’’; a biomaterial is defined as a
component of a medical device [14]. According to this definition,
the materials used for contact lenses and total hip replacement are
biomaterials, whereas those used for corrective eye-wear and
artificial leg prostheses are biomedical devices, typically not dealt
with in the field of biomaterials.
In this review, the term ‘‘biomedical material’’ is used to
represent a component of any biomedical device applied with or
without intimate contact with living tissue, and the definition of
‘‘biomaterials’’ prevalent in the scientific community is used to
describe biomaterials used only in intimate contact with living

Fig. 2. (a) A vascular stent and (b) an aneurysm clip, both made from NiTi alloy.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 3. The definition boundaries of biomedical materials, biomaterials and
biological materials.
tissues. Hence, biomaterials are implanted biomedical materials. A
biomaterial can be either synthetic or naturally occurring, such as
bone and cotton. In this article, we refer to naturally occurring
materials as ‘‘biological materials’’. Fig. 3 demonstrates the
definition boundaries of ‘‘biomedical materials’’, ‘‘biomaterials’’
and ‘‘biological materials’’ used in this review.
1.3. Definition of biocompatibility
Since a biomaterial is designed to be used in intimate contact
with living tissue, it is essential that the implanted material does
not cause any harmful effects. Williams suggests that biocompati-
bility covers all aspects of bio-device function, including the
interaction of cells and tissues with the implanted biomaterials
[15–17]. The requirements for this biocompatibility are complex
and strict, varying with specific medical applications. Rejection of
an orthopedic implant due to the toxic release of metallic ions, for
instance, will lead to ultimate failure. Hence, a metallic implant is
expected to be made of non-toxic elements and therefore cause no
measurable inflammatory or allergic reactions in the human body.
A material used satisfactorily in orthopedic surgery may however
be inappropriate for cardiovascular applications because of its
thrombogenic properties. In some studies [18,19], biocompatibili-
ty of biomaterials is further classified according to their ability to
induce cell or tissue death (cytotoxicity), cancer formation
(carcinogenicity), to genetic damage (mutagenicity), immune
responses (pyrogenicity and allergenicity) or blood clotting
(thrombogenicity). Taking all of these types of biocompatibility
problems into account, a universally prevalent definition of
biocompatibility has been provided by FDA [7,14], to the effect
that the material induces no measurable harm to the host. Hence,
any requirement of a medical device to perform a useful function is
Table 2
Some accepted classifications of medical devices [14].
Authority Class
TGAa I IIa, IIb
FDAa I II
ECHCa I IIa
Health Canada I II
General description Non-invasive and/or transient Minimally invasive, short term
use (e.g. dermal) (e.g. eyes, ear canal)
Restrictions General General and specific
Health risk Low Low/moderate
Examples Surgical instruments, mechanical Contact lenses, ultrasound
barriers probes
a
FDA, Food and Drug Administration (USA); ECHC, European Commission on Health and Consumers (European Union); TGA, Therapeutic Good Administration (Australia).
5
a further demand based upon biological safety, i.e. biocompatibili-
ty.
Therefore, the biocompatibility of a medical device covers both
the compatibility of the materials used and the design (e.g. geometry,
mechanics and electrical control) of the device. Indeed many clinical
failures of joint replacements, for example, are due to suboptimal
mechanics of the device, rather than problems of the materials
properties [20]. Non-material issues are beyond the scope of this
review article, as well as this journal. As regards the biocompatibility
of implant materials, it includes not only the chemical interactions of
the implanted material with the host physiological system (e.g.
corrosion of alloys and toxicities of metal ions) but also the physical
impacts of the implanted material on the surrounding tissues (the
mechanical properties of the material), although the former is the
more common and primary concern. In this article, we confine our
discussion on biocompatibility to the chemical interactions of
materials with the biological systems, and the physical impacts of
biomaterials are categorized as the mechanical properties of
materials. Within this more confined concept, the biocompatibility
of a metallic implant biomaterial is directly determined by its
corrosion resistance and the biological effects of released metal ions.
In this article, therefore, the corrosion resistance of each alloy and
biological roles of alloying elements are reviewed first, followed by
the biocompatibility of the alloy.
1.4. Classification of medical devices
Medical devices are classified by government regulatory authori-
ties, including the FDA, Medical Devices Bureau of Health Canada,
European Commission on Health and Consumers (ECHC) and the
Therapeutic Goods Administration (TGA) [21] (Table 2). Classifica-
tion of medical devices is based on the complexity of the device, the
level of control necessary to assure its safety and effectiveness, and in
Australia a higher classification is given to devices with increasing
degree of invasiveness, depending on which tissue the device is
applied to (e.g. class III corresponds to chronic implantable devices).
Non-implant (class I and II) medical devices (e.g. surgical and dental
instruments) are manufactured from commercial-grade materials.
These materials adequately meet clinical requirements where
contact with human tissue is transient. Biomaterials of interest to
the scientific community are generally used in medical devices that
fall within the higher classes in Table 2, which must be suitable for
close and prolonged contact with human tissue, and thus require
premarket approval. This review is devoted to those metallic
biomaterials used in FDA class II and III medical devices.
2. Essential considerations in design of metallic biomaterials
The design and selection of a biomaterial depends on its
specific medical application. In order to serve safely and
III
III
IIb III
III IV
Short to medium term contact Medium to long term contact,
with blood, oral/nasal mucosae chronic implants, control systems
General control and premarket
approvals
Moderate/high High
Orthopedic implants, dialysis Pacemakers, perfusion pumps,
machines vascular stents
6 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 3
Elements in the human body [22].

Element O C H N Ca P K S Na Cl Mg Trace element

Wt% 65.0 18.5 9.5 3.3 1.5 1.0 0.4 0.3 0.2 0.2 0.1 <0.01
At% 25.5 9.5 63.0 1.4 0.31 0.22 0.06 0.05 0.3 0.03 0.1 <0.01

appropriately for a long period of time without rejection, a
metallic implant should possess the following essential char-
acteristics, but not limited to:
(1) Excellent biocompatibility (non-toxic)
(2) High corrosion resistance
(3) Suitable mechanical properties
(4) High wear resistance
(5) Osseo-integration (in the case of bone prosthetics)
The above requirements need to be considered in the design of
metallic biomaterials, as discussed as follows.
2.1. Biocompatibility of elements and selection of alloying elements
Since no material is completely inert in the living body over a
reasonable period of time, an immediate thought in the selection of
alloying elements may be those elements that already exist
naturally in the body. Most of the human body is made up of water,
H2O, with cells containing 65–90 wt% water. Therefore, it is not
surprising that most of a human body’s mass is oxygen and carbon.
A list of elements commonly found in the human body is given in
Table 3. Approximately 96% of the weight of the body results from
the elements oxygen, carbon, hydrogen and nitrogen, which are
the building blocks of both water and proteins. The rest (4%) of
the mass of the body exists largely either in bone as minerals (Ca,
Mg, and P) or in blood and extracellular fluid as major electrolytes
(Na, K, and Cl) (Table 4).
In addition, there are a number of elements, which are needed
in extremely low quantities for the proper growth, development,
and physiology of the body (Table 5). These elements are referred
to as trace elements or micronutrients, a list that is continually
increasing (Table 6). It is important to bear in mind that these trace
elements are all toxic at high levels, as demonstrated when
addition of cobalt compounds to stabilize beer foam in Canada
contributed to cardiomyopathy, which came to be known as beer
drinker’s cardiomyopathy [24,25]. Table 7 lists the percentage of
population who suffer from metal allergies [26]. In brief, many
metal elements are needed in the human body as micronutrients,
but are toxic at levels higher than required. The biocompatibility of
the pure metals used in metallic biomaterials and these alloys are
compared in Fig. 4 [27].

Table 4
Macro elements and their roles in the human body [23].

Macro elements Roles

O, C, H, N in water and the molecular structures of proteins

Ca Structure of bone and teeth.
P Structure of bone and teeth. Required for ATP, the energy carrier in animals.
Mg Important in bone structure. Deficiency results in tetany (muscle spasms) and can lead to a calcium deficiency.
Na Major electrolyte of blood and extracellular fluid. Required for maintenance of pH and osmotic balance.
K Major electrolyte of blood and intracellular fluid. Required for maintenance of pH and osmotic balance.
Cl Major electrolyte of blood and extracellular and intracellular fluid. Required for maintenance of pH and osmotic balance.
S Element of the essential amino acids methionine and cysteine. Contained in the vitamins thiamin and biotin. As part of
glutathione it is required for detoxification. Poor growth due to reduced protein synthesis and lower glutathione levels
potentially increasing oxidative or xenobiotic damage are consequences of low sulfur and methionine and/or cysteine intake.

Table 5
Some trace elements (also called micronutrients) and their roles in the human body.

Trace elements Roles

Fe Contained in heme groups of hemoglobin and myoglobin which are required for oxygen transport in the body, as well as many other metabolic
enzymes and Fe–S proteins. Part of the cytochrome p450 family of enzymes. Anemia is the primary consequence of iron deficiency. Excess iron
levels can enlarge the liver, may provoke diabetes and cardiac failure. The genetic disease hemochromatosis results from excess iron
absorption. Similar symptoms can be produced through excessive transfusions required for the treatment of other diseases.
Cu Contained in enzymes of the ferroxidase system which regulates iron transport in the blood and facilitates release from storage. A structural
element in the enzymes tyrosinase, cytochrome c oxidase, ascorbic acid oxidase, amine oxidases, and the antioxidant enzyme copper zinc
superoxide dismutase, amongst others. A copper deficiency can result in anemia from reduced ferroxidase function. Excess copper levels cause
liver malfunction and are associated with genetic disorder Wilson’s Disease
Mn Major component of the mitochondrial antioxidant enzyme manganese superoxide dismutase. A manganese deficiency can lead to improper
bone formation and reproductive disorders. An excess of manganese can lead to poor iron absorption.
I Required for production of thyroxine which plays an important role in metabolic rate. Deficient or excessive iodine intake can cause goiter (an
enlarged thyroid gland).
Zn Important for reproductive function due to its role in FSH (follicle stimulating hormone) and LH (leutinizing hormone). Required for DNA
binding of zinc finger proteins which regulate a variety of activities. A component of the enzymes alcohol dehydrogenase, lactic dehydrogenase
carbonic anhydrase, ribonuclease, DNA Polymerase and the antioxidant copper zinc superoxide dismutase. An excess of zinc may cause anemia
or reduced bone formation.
Se Contained in the antioxidant enzyme glutathione peroxidase and heme oxidase. Deficiency results in oxidative membrane damage with
different effects in different species. Human deficiency causes cardiomyopathy and is known as Keshan’s disease.
Co Contained in vitamin B12. An excess may cause cardiac failure.
Mo Contained in the enzyme xanthine oxidase. Required for the excretion of nitrogen in uric acid in birds. An excess can cause diarrhea and growth
reduction.
Cr A cofactor in the regulation of sugar levels. Chromium deficiency may cause hyperglycemia (elevated blood sugar) and glucosuria (glucose in
the urine). Elevated levels of some forms of chromium, such as Cr(VI), can be carcinogenic.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 7

Table 6 Table 8
List of known trace elements in the human body.a The ionic concentrations (mM) of human blood plasma [32].

Barium Chromium Iron Selenium Ion Human tissue fluid Human blood plasma
+
Beryllium Cobalt Lithium Strontium Na 142.0 142.0
Boron Copper Molybdenum Tungsten HCO3 4.2 27.0
Cesium Iodine Nickel Zinc K+ 5.0 5.0
a HPO42 1.0 1.0
These elements are all toxic at high levels.
Mg2+ 1.5 1.5
Cl 147.8 103.0
Ca2+ 2.5 2.5
Table 7 SO42 0.5 0.5
Metal sensitivity.

Percent Metal Sensitive

General population 10
Patients with stable total joints 25
Patients with loose total joints 60
Ideally, non-toxic elements should be selected as alloying
elements in developing a biomedical alloy. In reality no metals are
completely inert or non-toxic. Construction of metallic implants
therefore requires the use of alloys with virtually inert elements, or
those that exist as trace elements in the body; or in the case of
reactive alloys, the use of highly corrosion resistant elements (e.g.
Ti). More detailed reviews on the biocompatibility of trace
elements will be provided in the context of specific alloy systems
in which they are used as alloying elements (see Sections 3–13).
2.2. Corrosion of implant materials in the body
The environment inside the human body is physically and
chemically different from ambient conditions. Consequently, a
metal that performs well (is inert or passive) in the air may suffer a
severe corrosion in the body. As a matter of fact, the most corrosion
resistant stainless steels typically cause chronic allergy and toxic
reactions in the host body, which are only diagnosed after a
sufficiently long post-implantation period [28–30]. While corro-
sion resistance determines the long-term success of a metallic
implants, different parts of the body have different pH values and
oxygen concentrations. An implant that performs well in one
region of the body may therefore suffer an unacceptable amount of
corrosion in another, due to acidic erosion and oxidation.
Corrosion is also accelerated by aqueous ions, as commonly
seen for metals in ambient air near coastal areas. Under normal
conditions, most human body fluids contain around 0.9% saline,
solutions of mostly Na+, Cl and other trace ions (Table 8), as well
as amino acids and a range of soluble proteins. There is also trace
debris and cellular material that can result in focal adhesions onto
implants. These fluids have a nearly neutral pH value (7.2–7.4 at
37 8C, and 1 atm of pressure). However, the pH value of body fluid
may fall to 3–4 when there is inflammation caused by surgery or
injury, due to inflammatory cell secretions [31]. Combined with
fluctuations in ionic strength in relation to high blood pressure, or
due to ion deposits, the human body presents an aggressive
environment for any implant. Furthermore, the internal partial
pressure of oxygen is about one quarter of atmospheric oxygen
pressure. While less reactive in terms of oxidation, lower oxygen
actually accelerates corrosion of metallic implants by slowing
down the formation of protective passive oxide films on the metal
surfaces once an implant is broken or removed [31]. Ideally,
corrosion resistance should be such that the release of metal ions
from a metallic implant will be minimized in the harshest
conditions of the body, and remain at a satisfactorily low level
over a long service period (more than 30 years) under normal
physiological conditions.
2.3. Mechanical working conditions in the human body
To replace bone, which is satisfactorily strong and tough,
biomaterials must be able to match this mechanical performance.
The mechanical properties of general importance to biomaterials
development include Young’s modulus, ultimate tensile strength
(UTS) and toughness. A summary of these mechanical properties
for three dominant metallic biomaterials – stainless steel, cobalt-
based alloys and titanium-based alloys – is given in Table 9. These
three metallic biomaterials remain popular primarily because of
their ability to bear significant loads and undergo plastic
deformation prior to failure, as indicated by their respective UTS
and fracture toughness.
It must be mentioned that stainless steel, cobalt-based alloys
and titanium-based alloys have much higher Young’s modulus

Fig. 4. (a) Cytotoxicity of some pure metals. (b) The relationship between polarization resistance and biocompatibility of pure metals, cobalt–chromium alloy and stainless
steels [27].
8 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 9 Imperfections can arise from inhomogeneity of microstructure

Mechanical properties of metallic implant materials and cortical bone [33].
(e.g. impurities, second phase particles and grain boundaries);
Materials Young’s Ultimate Fracture manufacturing defects of the metallic component (e.g. holes, filets,
modulus/GPa tensile toughness welds, notches, and pits), or surface imperfections from machining
pffiffiffiffiffi
strength/MPa (MPa m)
operations [34,35]. Stress concentrates locally to these sites when
CoCrMo alloys 240 900–1540 100 the material is subjected to external loading, which can escalate to
316L stainless steel 200 540–1000 100 permanent defects (e.g. dislocations and microcracks) that cannot
Ti alloys 105–125 900 80
be removed by unloading. Instead, they either accumulate in
Mg alloys 40–45 100–250 15–40
NiTi alloy 30–50 1355 30–60 number or grow in size at these weak sites during the process of
Cortical bone 10–30 130–150 2–12 cyclic loading. Eventually a progressively growing crack develops
under normal service conditions until the final catastrophic failure.
In short, the fluctuating stresses inside a material may lead to the
(more than 100 GPa) than that of bone which is only 10–30 GPa initiation and growth of a crack, which upon reaching a critical size,
[33]. The higher modulus of elasticity of the implant may result in leads to complete fracture. Detailed fatigue mechanisms have been
its bearing nearly all the load, however bone that bears less provided by excellent review articles published elsewhere [34,35].
mechanical load can undergo biological responses such as atrophy, The mechanical working conditions of orthopedics implants are
particularly around the implant site; further revision surgery is frequently complicated by concurrent cyclic stress and friction, a
thus required. This is called the ‘‘stress shielding effect’’. Hence, it is scenario called fretting fatigue. This is contributed by cyclic friction
desirable to have an implant with similar Young’s modulus to that stress superimposed onto plain fatigue stress. When fretting
of bones. fatigue occurs, a foreign body is statically pressed to the surface of
The mechanical working conditions within the human body are the specimen to which cyclic stress is applied. Fretting occurs as a
complex. Human beings normally walk several thousand steps a result of relative movements of low amplitude, which may occur
day at a rate of 1 Hz. As such, skeletal bone implants such as between contacting surfaces of components. This results in
artificial hip joints, knee joints, spinal fixations, plates and wires production of oxide debris and fresh metal surfaces. Visible
suffer from fatigue due to cyclic loading. In the case of artificial hip damage is found even when the amplitude of slip is as little as
joints, the loading stress level is several times higher than that of 104 mm [36]. A crack can initiate at these contact sites, soon
the patient body weight [31]. This is because when hip joint is propagating before it finally breaks. Artificial hip joints, bone
located out of perpendicular alignment, such as during a step, most plates, and wires are likely to suffer from fretting fatigue [31].
of the body weight pivots on only one leg at a time. Loading stress Fatigue can also be complicated by concurrent corrosion. Inside
on a single leg is estimated to be 50 MPa on average, if a the living body, the surface wear of metallic materials leads to
hypothetical load of five times of the body weight is applied to the successive release of metal ions, metallic compounds and debris.
cross-section of the stem of a total hip prosthesis [31]. Assuming The release of these products into the tissues surrounding an
that a person walks 2  103 steps, the total number of steps over implant may provoke toxicity and foreign body reactions in local
20 years is estimated to be 2000  365 day  20 years  1  107 tissues or organs [37]. For instance, the black-coloring of the host
cycles (Table 10). This kind of cyclic stress also occurs in dental tissue surrounding an implant, a phenomenon called metallosis in
implants during chewing motion, and non-osseous tissue implants clinical orthopedics [38,39] is caused by the release of large
such as pacemaker electrodes in response to myocardial activity. In amount of debris, which eventually leads to aseptic loosening
any case, cyclic loading promotes material fatigue more efficiently (Section 2.4). When fatigue occurs along with corrosion, the
than static fixed loading. phenomenon is known as corrosion fatigue. Fretting fatigue with
Fatigue strength (fatigue limit or endurance limit) refers to the corrosion is thus known as ‘‘fretting corrosion fatigue’’. Metallic
maximal amplitude (or range) of cyclic stress that can be applied to implant biomaterials tend to be damaged by corrosion fatigue or
a material without causing fatigue failure. Fatigue strength fretting corrosion fatigue in human bodies [40,41]. In short, the
sensitively varies with the microstructure of materials, surface inside of the living body is a chemically and mechanically harsh
quality of products, and service conditions (e.g. load vectors, cyclic environment for metallic materials. Moreover, flaws on exposed
frequency, wearing, and corrosion environment). A material implant surfaces, which can be caused by an accidental scratch of
subject to a cyclic loading can fracture far below its UTS and even the surgical knife during operation, also decrease fatigue strength.
below the yield strength of the material. Fatigue fractures are The procedures of fatigue testing of metallic biomaterials can be
dangerous because they occur under normal service conditions found in ASTM standards: F1801-97 (compression fatigue), F1717-
with no warning prior to rupture. Indeed, medical devices 96 (Spinal device), F1612-95 (Stem of total hip replacement),
manufactured from any material that are expected to survive F1539-95 (Staple), F1440-92 and E602 (fatigue testing of metallic
millions of cyclic deformations over their lifetime require scrutiny implants). For fretting corrosion testing of metallic biomaterials,
of the fatigue and fracture resistance, with fatigue fracture being one can refer to F897-84. According to F1801-97, the standard
the major cause of premature failure in biomedical implants. conditions for axial tension tests with cylindrical specimens are
Fatigue usually initiates at a location that acts as a stress recommended to be 37 8C, 0.9% saline solution and air, under a
concentration. In the real world, no materials are perfect. loading frequency of 1 Hz. There is no standard for fretting fatigue
testing of metallic biomaterials, although a reliable test procedure
Table 10 for fretting fatigue is described by Nakazawa and co-workers [42].
Typical fatigue mechanical working conditions of some implants [31]. Procedures for predicting the fatigue life of materials are
Implants Loading strength Loading Expected total number conducted under either stress-controlled or strain-controlled
frequency of loading over the life conditions. The correct choice between these two conditions
(Hz) time of a 65y old patient requires an understanding of the cyclic conditions of service, such
(i.e. 20 years)
as the cyclic loading due to the human gait or movements unique
Joints Compression 50 MPa 1 107 to various anatomical locations. A stent placed in a popliteal artery
Bending 200 MPa located under the knee joint, for example, may experience cyclic
Pacemaker Not available 1 109
bending deformations of 908 in response to knee flexion, and thus
Tooth fillings Not available 1 107
requires a strain-based method. For the stem of total hip
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 9

Fig. 5. (a) A total hip prosthesis by Depuy, and (b) terminology of components.

replacements, the choice of method is more complex. A hip
replacement is typically subjected to cyclic loads proportional to
the weight of the patient in response to each footstep and hence
requires a stress-based approach [43]. However, for biomedical
applications, the notch fatigue properties of metallic biomaterials
are considered to be relative to the performance of hip implants,
because the stem of total hip replacements is often structured with
wedges (Fig. 5). Hence, strain-controlled notch fatigue perfor-
mance is considered to be more representative of in vivo conditions
by some researchers [44].
2.4. Wear of joint implants
Wear is an inevitable problem in any joint replacements no
matter what materials are used. The choice of material types for
joints is primarily determined by the type of joint. The skeletal
system contains several types of mobile joints between long bones
(e.g. hip, knee, ankle, shoulder, and elbow) and more static joints
(e.g. skull, wrist, and tooth). Mobile joints can be categorized as
congruent and incongruent, depending on how closely the opposed
bones fit together. In congruent joints, such as the hip and shoulder
joints, a ball-shaped head fits closely to a cup-like socket, so the
stress is distributed evenly (Fig. 6). Such mechanical load can be
sustained by any strong materials, including brittle ceramic
materials. The ranking of articulating surface for ball-and-socket
joint systems in terms of friction (wearing) resistance is shown in
Table 11. For incongruent joints, such as the knee and ankle joints,
contact of two incongruent hard surfaces creates highly concen-
trated (heterogeneous) stresses (Fig. 7), which are compensated by
the presence of thick cartilage layers and synovial fluid. Brittle
ceramic materials cannot sustain such stress impacts, so metallic
and hard polymeric materials are preferred in these cases. At
present, clinically applied joint replacements are made of Ceramic-
on-UHMPFE, Metallic-on-UHMPFE (categorized as Hard-on-Soft)
Ceramic-on-Ceramic, CoCrMo-on-CoCrMo, or Al2O3-on-CoCrMo
(categorized as Hard-on-Hard).
Improvements in prosthetics, implant design and methods of
surgical fixation have advanced the development of load-bearing
materials for joint replacement systems. The low wear resistance
or high friction coefficient of a joint system results in implant

Fig. 6. Two congruent joints in human bodies. (a) hip http://www.hughblackley.co.nz/total-hip-joint-replacement.htm and (b) shoulder http://www.emedicinehealth.com/
shoulder_dislocation/article_em.htm.
10 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Table 11
Ranking of articulating surfaces for joint prosthetics in terms of wear-resistance.
Ball and socket Wearing resistance
Ceramic-on-Ceramic (Al2O3 or ZrO2) Superior
CoCrMo-on-CoCrMo Excellent
Al2O3-on-CoCrMo Excellent
Al2O3 on UHMWPEa Excellent
CoCrMo on UHMWPE Good
Ti6Al4V on UHMWPE Good
Metal on metal (stainless steels or titanium alloys) Poor
a represents a currently unmet goal of tissue engineering, toward a
UHMWPE, ultra-high molecular weight polyethylene.
loosening. Although not fully understood, a frequent incident
caused by wearing damage is aseptic loosening, which can occur in
any material when micro-particles are generated at the bearing
couples of a joint replacement [45–47]. While joint replacements
initially perform their intended biomechanical function very well,
gradually increased levels of particles are produced due to long-
term wear, which then attracts macrophages of the immune
system that recognize and engulf the particles as foreign bodies,
similar to bacteria or pyrogens. This is a natural defence
mechanism, however foreign synthetic particles tend to kill
macrophages after ingestion [48]. As a result, dying macrophages
releases enzymes and metabolites which cause severe acidification
in the surrounding microenvironment. It is these enzymes, acidic
chemicals, ions and debris that contribute to implant and bone
erosion during aseptic loosening. To address the problems
associated with low wear resistance in the currently dominant
Hard-on-UHMWPE joint systems, Metal-on-Metal bearing systems
were re-introduced in joint replacements in the 1980s (Section
4.5), after having been almost completely replaced by the former in
the 1970s. However, there have been increasing concerns over the
adverse reactions to both local and systemic distribution of metal
wear particles and ions in Metal-on-Metal systems.
In short, wear debris causes severe adverse responses, whereby
a revision surgery is likely to be required. Therefore, high fatigue
strength, excellent corrosion and wear resistances are the key
properties that determine the longevity of joint implants in the
human body.
2.5. Osseo-integration
Osseo-integration is a fundamental requirement in orthopedics,
a term that describes the process of new bone formation and bone
healing [49]. The inability of an implant surface to bond to adjacent
bone and other tissues due to micromotions will cause the
formation of fibrous tissue around the implant, which promotes
loosening of the prosthesis [49]. Therefore, it is essential for an
implant to have an appropriate surface to integrate well with
Fig. 7. Two artificial incongruent joints (a) knee and (b) ankle.
surrounding bone. Surface chemistry, surface roughness and
surface topography are all factors that need to be considered for
good osseo-integration [50–53].
2.6. The objectives of the rest of the review article
An ideal metallic biomaterial should have a similar modulus to
that of bone, excellent resistance to fatigue, corrosion and wear,
and a good bone bonding ability. To develop such an alloy
synthetic entity that would substitute damaged bone at load-
bearing sites with a 100% success rate over more than 30 years.
The alloying chemistry, metallurgical physics and manufactur-
ing processes for metallic biomaterials are not necessarily unique
to the biomedical device industry. There is a range of comprehen-
sive references on material science (i.e. the structure–property
relationship) and material engineering (i.e. processing) of stainless
steels [54], Co-based alloys [55], Ti-based alloys [56], NiTi
shape memory alloys [57] and Mg-based alloys [58]. However,
the control of materials problems including biocompatibility
and resistance to corrosion is the foremost aspect of the safe
application of metallic implants [44], with mechanical perfor-
mance under biological working conditions of critical importance
to orthopedics at load-bearing sites. In the following sections of
this review, therefore, the control mechanisms of undesired factors
related to the materials properties of each alloying system are
discussed first, followed by detailed review on the biocompat-
ibilities of the alloy and its components. Discussion will then turn
to mechanical features, including general mechanical properties,
fatigue properties, wear-resistance and the clinical application
outcomes of the alloy system. Finally, both areas will be
summarized, with identification of the most challenging issues
inherent to current metallic biomaterials used as medical implants,
and the most promising approaches and strategies to address these
challenges.
3. Stainless steels
Since the Iron Age around 1200 BC, when iron metallurgy was
first mastered, containers made of iron were used in preparing and
keeping food in kitchens. While the benign nature of elemental
iron to the human body has been witnessed throughout history,
metal ion toxicity has more recently studied [59], classified in
decreasing order of toxicity as follows: cobalt > vanadium
> nickel > chromium > titanium > iron.
Stainless steel is the generic name for a number of iron-based
alloys that contain a high percentage (11–30 wt%) of chromium
[44] and varying amounts of nickel. Stainless steels can be
categorized into two groups: the chromium and chromium–nickel
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Table 12
Four categories of stainless steels and typical medical applications [44].
Material type Application grade
Martensitic Dental and surgical instruments
Ferritic Very limited surgical instruments
Austenitic A large number of non-implantable
medical devices
Many short-term implants
Total hip replacements
Duplex Not yet applied in the biomedical field
types, according to chemical composition. Alternatively, they can
also be grouped into four families based on the characteristic
microstructure of the alloys: martensitic, ferritic, austenitic, or
duplex (austenitic plus ferritic) [44], as listed in Table 12. Except
for the duplex type, each of the other three groups of stainless
steels finds applications in medical devices. The hardness of
martensitic stainless steels (up to 97 HRB) makes them ideally
suited for dental and surgical instruments. Ferrite stainless steels
find few applications in medical devices as well. Austenitic
stainless steels are applied in various non-implantable medical
devices where good corrosion resistance and moderate strength
are required. These applications often require a material that is
easily formed into complex shapes. However, only austenitic
stainless steels are used for implants [44]. Hence, the rest of this
section is devoted to these implant-grade stainless steels,
including type 316L and its variants.
3.1. Alloying chemistry: corrosion resistance
The minimum percentage of chromium in stainless steel is
11 wt%, the amount needed to prevent the formation of rust in
unpolluted atmosphere [44]. Table 13 lists the compositions of
316L (ASMT F138) stainless steel and its variants. These have
widely been used in temporary devices following bone trauma,
such as fracture plates, screws and hip nails, and in permanent
implants such as total hip replacements.
The chromium in stainless steel has a great affinity for oxygen,
which allows formation of an invisible chromium-rich oxide film
(2 nm thick). This surface layer is adhesive, promoting self-
healing in the presence of oxygen [60,61]. Further improvement in
corrosion resistance, as well as a wide range of mechanical
properties, is achieved by the addition of nickel. Nickel is the main
alloying element that stabilizes the formation of austenite in iron
and contributes to increased corrosion resistance by the formation
of protective oxide films on the surface of the alloys [44]. It must be
mentioned that nickel is highly toxic to the human body [62,63].
The addition of other alloying elements may enhance resistance
to specific corrosion mechanisms, or develop desired mechanical
and physical properties. For example, in addition to enhancing
corrosion resistance with a resultant strengthening passive film,
the use of molybdenum further increases resistance to pitting
corrosion caused by chromium carbide formation. Pitting corro-
sion is a form of highly localized corrosion that leads to the
Table 13
Compositions (wt%) of 316L stainless steel (ASTM F138) and variants [44].
ASTM code/UNS No of stainless steels Cr Ni
F138/S31673 17.00–19.00 13.00–15.00
F1314/S20910 20.50–23.50 11.50–13.50
F1586/S31675 (Orthinox) 19.50–22.00 9.00–11.00
F2229/S29108 19.00–23.00 0.10
11
Examples
Bone curettes, chisels and gouges, dental burs, dental chisels, curettes, explorers,
root elevators and scalers, forceps, hemostats, retractors, orthodontic pliers, and scalpels.
Solid handles for instruments, guide pins, and fasteners.
Canulae, dental impression trays, guide pins, hollowware, hypodermic needles,
steam sterilizers, storage cabinets and work surfaces, and thoracic retractors.
See Table 16
formation of small holes in metals. The thermodynamic reason for
pitting corrosion is that small de-passivated areas become anodic
while the remaining vast area becomes cathodic. When the
material is at a potential in the metastable pitting range, re-
passivation of mechanically de-passivated slip steps may not
commence, and so localized galvanic corrosion occurs at these
active sites. This mechanism is known as the slip-step dissolution
model [64].
In chromium steels, chromium has a strong affinity with
carbon, leading to the formation of carbides in carbon-rich regions,
typically at grain boundaries. The precipitation of chromium
carbide (Cr3C2, Cr7C3, and Cr23C6) results in the depletion of
chromium and thus the depassivation of small regions around
chromium carbides. Consequently, the corrosion resistance of the
steel surrounding the chromium carbides is reduced. The addition
of molybdenum, which has a strong tendency to form carbides as
well, can effectively trap carbon by forming molybdenum carbides,
and thus reduce the level of chromium carbide formation. The
chemistry of 316L stainless steel is designed to maximize the
pitting corrosion resistance and provide a ferrite-free microstruc-
ture. The pitting resistance equivalent (PRE) meets the following
requirement: %Cr + 3.3  %Mo  26. It should be noted that it is
necessary to decrease the carbon content to improve corrosion
resistance, though carbon contributes to the increased strength of
steels. Carbon is normally present in amounts from less than 0.03%
in austenitic steels (Table 13).
Crevice corrosion refers to corrosion occurring in a confined
space, to which the access of the surrounding fluid is limited. The
mechanism of crevice corrosion in alloys that can be passivated
(e.g. stainless steels), following exposure to chloride-rich body
fluids, is gradual acidification of the environment inside the
crevice, leading to the appearance of highly aggressive local
conditions that gradually destroy the passivity. Crevice corrosion is
encountered frequently between the underside of a bone screw
head and the countersunk area of the bone plate. It starts with the
depletion of oxygen in the crevice. While the anodic reaction
continues in the crevice, the oxygen concentration is not readily
replenished by the fluids outside the crevice. Meanwhile, smaller
chlorine ions flow into the crevice, drawn there by the metal ions
being released by the anodic reaction. The pH in the crevice region
thus decreases, causing accelerated metal oxidation [31].
Nitrogen is soluble in a relatively high content in austenitic
stainless steels and stabilizes the austenitic structure of iron. It can
Mo Mn Si Cu N C P S
2.25–3.00 2.00 0.75 0.50 0.10 0.030 0.025 0.010
4.00–6.00 2.00–3.00 0.75 0.50 0.20–0.40 0.030 0.025 0.010
2.00–4.25 2.00–3.00 0.75 0.25 0.25–0.50 0.08 0.025 0.010
21.00–24.00 0.50–1.50 0.75 0.25 >0.90 0.08 0.03 0.010
12 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
be used as a substitute for nickel in those variants of 316L
(Table 13) to increase mechanical strength as well as to enhance
resistance to pitting and crevice corrosion. In 1998, ASTM 1586
(i.e. Orthinox) came into use as a stem material in total hip
replacement [65]. In addition, nickel-free austenitic steels with
high nitrogen concentration (ASTM F2229) have also been
developed for medical use.
In addition to the above alloying chemistry, special production
routes such as vacuum melting [66], vacuum arc re-melting (VAR)
or electro-slag refining [67] are required to produce implant-grade
stainless steels. The vacuum melting step improves the cleanliness
of steels, such that the quantity and size of non-metallic inclusions
is minimized and the pitting and crevice corrosion resistance is
maximized in the products. Type 316L (18Cr–14Ni–2.5Mo) is a
vacuum-melted variant of the standard type 316 composition.
Finally, passivity of stainless steel implants is enhanced by nitric
acid passivation before the implant is sterilized and packaged for
delivery to a medical facility.
While resistance against pitting and crevice corrosion of
stainless steels has been improved by the above alloying chemistry
and metallurgical processing, the form of corrosion known as
stress corrosion cracking (SCC) cannot be prevented by these
methods. SCC occurs in stainless steels that work in a chloride-rich
medium, and is induced by the combined influence of tensile stress
and a corrosive environment. It can lead to unexpected sudden
failure of normally ductile metals subjected to a tensile stress.
Cracking from stress corrosion has been shown to occur under a
stresses much lower than the UTS, such as which might occur even
in an implant with residual stresses.
In summary, although the success of stainless steels makes
these alloys widely accepted as such, they are not completely
‘‘stain-proof’’. This is especially true in human body fluid, which is
highly corrosive as described in Section 2.2. As a matter of fact,
316L austenitic stainless steels are not sufficiently corrosion
resistant for long-term use as an implant material, and their
applications are now confined to temporary devices, such as
internal fixation or traction devices. In these applications, the
devices are removed after healing has taken place [68]. Currently,
the high-nitrogen stainless steel, Orthinox, is a predominant stem
material in the medical market of hip prosthesis [69].
3.2. Biocompatibility of alloying elements used in stainless steels
As listed in Table 13, the major alloying elements in 316L
stainless steel (ASTM F138) and variants are chromium, nickel,
molybdenum and manganese, in addition to the matrix element,
iron. In this section, we focus on the biological performance of the
three dominant elements: iron, chromium and nickel, with
bioactivity related to the release of elements as either soluble
ions or insoluble particles [70].
3.2.1. Iron
3.2.1.1. Iron in the body as a trace element. Iron is a necessary trace
element found in almost all living organisms, ranging from primitive
bacteria to humans. Iron is present in all cells in the human body, and
has several vital functions. Many cellular enzymes vital to life
contain iron, such as catalase and lipoxygenases [71], as well as
cytochromes, which oxidize food nutrients to produce energy
[71]. Iron is also an essential component of hemoglobin, an iron-
containing protein in blood that binds and carries oxygen to the
tissues from the lungs. Iron is absorbed into the body via oral intake,
and its deficiency (sideropenia or hypoferremia) is one of the most
common nutritional deficiencies. Too little iron can interfere with
these vital functions and lead to morbidity and death. The direct
consequence of iron deficiency is iron deficiency anemia [72].
3.2.1.2. Toxicity of iron. Large amounts of iron released from
metallic implants can cause excessive levels of iron in the blood.
High blood levels of free ferrous iron react with peroxides to
produce free radicals, which are highly reactive and can damage
DNA, proteins, lipids, and other cellular components. Iron typically
damages cells in the heart and liver, which can cause significant
adverse effects, including coma, metabolic acidosis, shock, liver
failure, coagulopathy, adult respiratory distress syndrome, long-
term organ damage, and even death if left untreated [73–75]. MRI
scanning shows that iron accumulates in the hippocampus of the
brains of those with Alzheimer’s disease and in the substantia nigra
of those with Parkinson’s disease [76]. Humans experience iron
toxicity above 20 mg/kg of body mass, and 60 mg/kg is considered
a lethal dose [77]. To the best knowledge of the present authors,
there have been no reports on any adverse effects caused by the
release of iron from stainless steel implants.
3.2.2. Chromium
3.2.2.1. Chromium in the body as a trace element. Chromium is a
member of the transition metals in group 6, having an electronic
configuration of 4s13d5. As a trace element, chromium is a cofactor
in the regulation of sugar levels in blood. Chromium deficiency
may cause hyperglycemia (elevated blood sugar) and glucosuria
(glucose in the urine). Hence, chromium concentrations in blood,
plasma, serum or urine may be measured to monitor for safety in
exposed individuals [78–80].
3.2.2.2. Toxicity of chromium. Among a range of possible oxidation
states of chromium, the Cr3+ and Cr6+ states are the most common
forms, with the toxicity depending on the oxidation state of the
metal. Water insoluble trivalent chromium (III) compounds and
chromium metal are not classified to be a health hazard, while the
toxicity and carcinogenic properties of hexavalent chromium (VI)
have been well documented [81–83]. The carcinogenicity of
chromate dust was known as early as 1890, when the first
publication described the elevated cancer risk of workers in a
chromate dye company [84–86]. Chromium salts (chromates),
which are often used to manufacture leather products, paints,
cement, mortar and anti-corrosives, are also the cause of allergic
reactions in some workers. Contact with products containing
chromates can lead to allergic contact dermatitis and irritant
dermatitis, resulting in ulceration of the skin (chrome ulcers)
[87,88]. This skin disease is often found in workers that have been
exposed to strong chromate solutions in electroplating, tanning and
chrome-producing industries [89–91]. An actual litigation con-
ducted in 2009 on hexavalent chromium release into drinking water
was used as the plot of 2010 biographical film Erin Brockovich.
Chromium, as well as nickel, is received into the body via the
lungs [92], oral intake [93], skin contact [83] or via implants
[78]. Once in the body, chromium (VI) is reduced to chromium (III)
in the blood before it enters the cells. The chromium (III) is
excreted from the body via urine. In vitro studies have indicated
that high concentrations of chromium (III) in cells can lead to DNA
damage [94–97]. The acute toxicity of chromium (VI) is due to its
strong oxidative properties. After it reaches the blood stream, it
damages the kidneys, liver and blood cells through oxidation
reactions, resulting in hemolysis, which causes renal and liver
failure [81,82]. The lethal dose for chromium (VI) ranges between
50 and 150 mg/kg [98]. World Health Organization recommended
maximum chromium (VI) is 0.05 mg/l [99–101].
3.2.3. Nickel
3.2.3.1. Nickel in the body as a trace element [102]. The biological
role of nickel as an essential trace element was not recognized until
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
the 1970s [103]. Nickel exists in urease, an enzyme that assists in
the hydrolysis of urea. In blood, nickel is mainly bound to the
albumin fraction, but also to some other proteins of serum
[104]. Most of the nickel is eliminated into urine (90%). The
concentrations of nickel in human tissues are estimated to be (mg/
kg of dry weight): 173 in lung, 62 in kidney, 54 in heart, 50 in liver,
44 in brain, 37 in spleen and 34 in pancreas [105].
Nickel deficiency has been found to have a number of
deleterious effects and result in pathological consequences in
animal models (e.g. goats, rats and chicks), including reduced
growth, weight loss increased perinatal mortality [106], skin
changes (pigmentation and parakeratosis) and uneven hair
development [107]. Animals with nickel deficiency have been
found to have impaired metabolism of iron, fats, glucose, and
glycogen [108,109]. Nickel deficiency interferes with the incorpo-
ration of calcium into the skeleton and decreases the length/width
ratios of long bones of the leg. It also suppresses the activity of
enzymes in the heart, liver and kidneys leading to the degeneration
of cardiac and skeletal muscle [107,108,110–113].
3.2.3.2. Toxicity and carcinogenicity of nickel [102]. Similar to
chromium, the toxicology of nickel was initially revealed by
contact-allergy related dermatitis, causing itchy and red skin due
to its use in ear-piercing [114,115]. The amount of nickel allowed
in products that come into contact with human skin is regulated by
the European Union, and nickel has been previously recognized as
one of the most important allergens by the American Contact
Dermatitis Society [116]. In the US, the minimal risk level of nickel
and its compounds is set to 0.2 mg/m3 for inhalation during
15–364 days [117,118]. Respiratory diseases caused by the toxicity
of nickel include acute pneumonitis from inhalation of nickel
carbonyl [7], chronic rhinitis and sinusitis from inhalation of nickel
aerosols and cancers of nasal cavities and lungs as observed in
nickel workers [119]. Following cutaneous contact, dermatitis and
hypersensitivity reactions to nickel alloys have also been reported
[7,120].
The toxic and carcinogenic properties of nickel are related to
certain nickel containing compounds, rather than pure Ni2+ ions.
Nickel carbonyl, Ni(CO)4, is an extremely toxic gas, while fumes
and dust of nickel sulphide, Ni3S2, are known to be carcinogenic
[121]. In one study, inhaled Ni3S2 was found to cause lung cancers
in rats, while inhaled nickel oxide did not [122]. In vivo, Ni2+ ions
may cross cell membranes using the Mg2+ ion transport system,
and then bind to cytoplasmic ligands [123,124], although soluble
Ni2+ is rapidly cleared. There is no specific mechanism for delivery
of Ni2+ to target sites in the cell nucleus that may result in cancer
causing genetic mutations [122].
A mechanism by which a nickel compound may be harmful,
especially in particle form, is due to active uptake by cells across
their membranes (endocytosis). Some characteristics of nickel
compounds that increase their ability to be endocytosed include
their crystalline nature, negative surface charge, 2–4 mm range
particle size and low solubility [125]. Ni3S2, which shows low
solubility in vivo, may act by this mechanism [126]. When nickel
compound particles are endocytosed by target cells, the endocytic
vesicles are acidified by fusion with lysosomes and Ni2+ is released.
Deleterious changes, such as the formation of oxygen radicals and
subsequent DNA damage can occur, a known mechanism for
initiating tumorigenesis [127,128].
Nickel has been shown to be harmful to cultured bone cells, but
to a lesser level than cobalt or vanadium, which are used in Co- and
Ti-alloys [59]. In vitro tests have also shown that cobalt, nickel and
chromium have a potential for carcinogenicity [129]. Pure nickel
implanted intramuscularly or inside bone has been found to cause
severe local tissue irritation and necrosis [130,131]. Malignant
fibrous histiocytomas or fibrosarcomas are also associated with an
13
ability for retention of nickel [132]. Released metal ions due to the
corrosion of the implants can also migrate into distant organs, and
are more readily able to cross cell membranes. Systemic toxicity
may be caused by the accumulation, processing, and subsequent
reaction of the host to corrosion products [133–135]. When high-
dose nickel salts were injected into mice, accumulation of nickel
was observed in the liver, kidney and spleen, causing local
deleterious effects [136]. Increased nickel concentrations have also
been found in tissues adjacent to stainless steel implant materials
(116 and 1200 mg/L) as well as in some distant organs [133,
137]. Other factors, such as infection and mechanical damage
due to wearing and friction may further encourage nickel
release, raising the concentration retained local to the implant
site [138–140].
3.3. Biocompatibility of 316L stainless steel
The use of stainless steel in medical implants began in the 1930s
[141], when cytotoxicity evaluation standards were yet to be
established. Hence there are almost no in vitro or in vivo reports
focusing on the biocompatibility of stainless steels both from this
period. The reasonably good biocompatibility of this alloy was
suggested by the early clinical success of total hip replacements in
1960s and 1970s. Further evidence of biocompatibility of 316L
stainless steel has been reported in many studies since then, when
used as a control material [142–144]. More detailed review will be
given in the context referring to NiTi alloys (Section 8). In general,
316L stainless steel shows relatively good biocompatibility, but to
a less satisfactory level than CrCrMo and titanium alloys, due to its
greater corrosion rates [44].
316L < CoCrMo < Ti6Al4V
!
Poor Biocompatibility Good
3.4. Mechanical properties of implant-grade stainless steels
3.4.1. General mechanical properties
Most mechanical properties (e.g. yield strength, fatigue
strength, UTS, and elongation) vary with both alloy type (i.e. alloy
chemistry) and processing (i.e. microstructure). The elastic
(Young’s or shear) modulus is an exception, which is determined
more by the alloy type than its microstructure (Table 9). Stainless
steels for implants are wrought alloys (i.e. they are fabricated by
forging and machining). A summary of microstructure-sensitive
mechanical properties of type 316L stainless steel and variants is
provided in Table 14.
Although 316L stainless steels are much stronger than bone, in
terms of elastic modulus, UTS and fracture toughness (Table 9),
their clinical application history has revealed that they are not safe
to be used for a permanent device in load-bearing sites. This is
because the mechanical working conditions inside the body are far
more complicated than simple compressive loading. As discussed
in Section 2.3, metallic biomaterials implanted into human bodies
are typically subject to fatigue damage [145], either corrosion
fatigue [146] or fretting corrosion fatigue [147].
3.4.2. Fatigue properties
The number of reports on the fatigue behavior of stainless steel
implants is rather low (Table 15), compared with those on cobalt
and especially titanium-based alloys [148]. This is because the use
of austenitic stainless steels is confined to temporary devices, due
to their poor corrosion resistance in body fluids as longer-term
implants, from which harmful ions can be released to adjacent
tissues [2,149–153]. The data listed in Table 15 and Fig. 8
collectively indicate that 316L stainless steels have lower fatigue
14 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 14
Mechanical requirements for 316L (ASTM F138) and variants in bar or wire [44].

Steels Conditions UTS, min/MPa Yield strength (0.2%), min/MPa Elongation in 4D, min/%

F138 Annealed 490 190 40

Cold worked 860 690 12
Extra hard 1350

F1314 Annealed 690 380 35

Cold worked 1035 862 12

F1586 Annealed 740 430 35

Cold worked 1000 700 20
Hard 1100 1000 10

F2229 Annealed 931 586 52

Cold worked 10% 1062 786 37
Cold worked 20% 1262 952 25
Cold worked 30% 1496 1227 19
Cold worked 40% 1731 1551 12

Table 15
Fatigue properties of 316L stainless steels (and F1586/S31675) (run-out at 106 or 107 cycles) [160].

Alloys Fatigue strength in air/frequency Fatigue strength in solution/frequency (Hz) R Refs.

Cast/Forged 316L 220 (smooth un-notched)/20 Hz 200 (smooth un-notched, PBS)/2 Hz 1 [159]
Cold-worked 316LVM – 300 (Notched, Ringer’s solution)/1 Hz 0.053 [161]
550 (smooth un-notched, Ringer’s solution)/1 Hz
Cold-worked 316LVM <600 (smooth un-notched)/10 Hz <600 (smooth un-notched, 0.9 wt% NaCl solution)/10 Hz 0.0 [162]
Cold-worked 316LVM 340 (notched)/120 Hz <340 (Notched, Ringer’s solution)/120 Hz 0.1 [163]
Annealed 316L 290 (notched)/120 Hz <290 (Notched, Ringer’s solution)/120 Hz 0.1 [163]
Annealed 316L – <400 (smooth un-notched, SBF)/1 Hz 1 [164]
Annealed 216L – <460 (smooth un-notched, SBF)/1 Hz 1 [164]
ISO5832-9 <550 <500 (smooth un-notched, 0.9 wt% NaCl solution)/10 Hz 0.01 [158]

PBS, phosphate buffered solution; SBF, simulated body fluid.

strength and shorter fatigue life in saline solutions than in ambient
air, and that the fatigue strength of 316L stainless steel is typically
between 200–300 MPa in a biological aqueous solution. It is
worthy of mention that, according to finite element analysis, the
maximal tensile stress in the femoral stem in the body is calculated
to be around 200 MPa [154]. Hence, cast and forged 316L stainless
steels are moderately reliable in terms of fatigue properties.
Furthermore, fatigue crack initiation has been correlated to the
presence of pits [155–157], which act as stress raisers and
preferential sites for fatigue crack initiation. Thus, as molybdenum
and nitrogen improve the pitting corrosion resistance of austenitic
stainless steels, adding such elements would be an effective way of
improving corrosion fatigue resistance of stainless steel implants,
as demonstrated in Fig. 8b [158].
3.5. Medical applications of stainless steels
The use of 316L stainless steels in long-term implant devices
owes its success to the work of Sir John Charnley, who drove the
evolution of the modern orthopedic procedure of Total Hip

Fig. 8. (a) Effect of SBF on the S–N relationship of cast 316L stainless steel [159]. (b) Effect of 0.9 wt% NaCl solution on fatigue properties of low nickel, high nitrogen stainless
steel [158].
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 9. (a) Corrosion scale on a Charnley stainless steel stem, and (b) pitting and corrosion of a Muller stainless steel stem after implant removal [166].
Arthroplasty (THA) [165]. Charnley’s design consisted of three parts:
a stainless steel one-piece femoral stem and head; a polyethylene
acetabular component, and a PMMA bone cement [165]. This design
resulted in a large number of THA operations by the 1970s. However,
the incidence of femoral stem fracture was higher with 316L
stainless steel than with Co-alloys [145]. Failure analysis revealed
that all fractures of stems were of the fatigue type, which was
initiated at defects in the stem material [145]. The initial clinical
success of stainless steels hip implants had also resulted in the
implantation of such devices into a younger population with a longer
life expectancy. However, reports of failure of those devices due to
aseptic loosening emerged in the 1990s [145]. Failure analysis of the
prosthesis after 9–21 years of service revealed that corrosion
occurred on the surface layer of either the stem (Fig. 9) [166] or head
[167,168] of the implant, which correlated to periprosthetic
metallosis and rapid failure [168]. Further clinical application of
THA from 1970 to 1990 has consistently demonstrated that 316L
stainless steels are not sufficiently corrosion resistant for long-term
use as hip joint implant materials.
In addition to poor resistance to fatigue and corrosion, 316L
stainless steels also have a number of other long-term issues,
including SCC, poor wear resistance, as well as the toxicity and
carcinogenicity of released nickel and chromium. 316L Stainless
steels have therefore been replaced by more corrosion and fatigue
resistant alloys in permanent implants, including high-nitrogen
variant of stainless steels (e.g. Orthinox), Co-alloys and Ti alloys.
Nowadays, 316L stainless steels are rarely used in any permanent
implant devices. Nevertheless, the low cost of stainless steels has
maintained their application in a large number of temporary
devices, such as internal fracture fixation and traction of the spine,
bone screws, bone plates, intramedullary nails and rods [44,68]. It
is also because of the low cost of stainless steel that high-nitrogen
variants of 316L stainless steel have been developed. According to
the 8th Annual Report 2011 of National Joint Registration (UK), the
top two hip stem brands, Exeter V40 and Charnley, both made from
Orthinox, occupy approximately 60 and 10% of the market in UK,
respectively [69]. Table 16 lists various implantable device
applications for stainless steels.
15
3.6. Current issues and challenges
Being used as a short-term device (from several months to
several years) does not guarantee the absence of a failure in
stainless steel, as premature failures of 316L stainless steels have
been reported in orthopedics implants (Fig. 10) [169–171]. The
time of failure can occur over a range from several months to
several years after implantation. Failure analysis on seven 316L
implants later revealed that the fractures appeared to be due to
fatigue failure, with implants showing very poor surface finishing
and fatigue cracking initiated at sites of machining imperfections
and crevice corrosion. Hence, the fatigue failure of 316L stainless
steel implants can be prevented by a quality surface finishing or
surface treatment, such as nitriding, which is primarily a quality-
control issue rather than a challenge in materials science and
engineering.
More importantly, there have also been premature fractures
of high-nitrogen Orthinox reported in the stems of hip joint
implants (Fig. 11), with a 10% incidence [172,173]. The fracture
positions are always around the middle of the stem, which is the
most highly stressed region of the distal part of the stem, as
indicated by finite element analysis (Fig. 12) [154]. The examina-
tions of fracture surfaces have revealed that the cause of failure is
indeed fatigue. Hence, it is reasonable to conclude that bending
fatigue is the major cause of stem fracture of Orthinox hip
replacements.
The adverse effects of nickel ions to the human body have
prompted the development of high-nitrogen, nickel-free austenitic
stainless steels for medical applications. Nitrogen replaces nickel
for austenitic structure stability, as well as improving the overall
steel structural properties [174,175]. By combining the benefits of
stable austenitic structure, high strength, good plasticity, better
resistance to corrosion and wear, and superior biocompatibility
compared to the currently used 316L stainless steel, the high-
nitrogen, nickel-free stainless steels (e.g. F2229) [176] could be a
better substitute. However, toughening of these alloys remains to
be explored, as high-nitrogen content can makes the alloys more
brittle.
16
Table 16
Medical applications of stainless steels [44].
Materials
Type 316L stainless steel
Type 316L stainless steel
Type 316L stainless steel
Type 316L stainless steel
Orthinox stainless steel
Type 316L stainless steel
Type 316L stainless steel
Types 316 and 316L stainless steel
Types 304, 316, and 316L stainless
steel
Types 316 and 316L stainless steel
17-7PH, 17-7PH (Nb), PH-15-7Mo, and
types 301, 304, 316, 316L, 420, and
431 stainless steel
Stainless steel wire formed in a zigzag
configuration of 5–10 bends
Stainless steel
Type 316 stainless steel
Type 304 stainless steel
Type 316 stainless steel, with spring
steel diaphragm
Types 304, 316, and 316L stainless
steel
Stainless steel
Type 316L stainless steel
Types 316 and 316L stainless steel
Types 304 and 316 stainless steel
Types 304 and 316 stainless steel
Type 304 stainless steel
Types 302, 303, 304, and
305 stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Stainless steel
Steel and stainless steel
Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Devices
Bone screws and pins
Onlay bone plates
Blade and nail bone plates
Intramedullar bone nails
Total joint prostheses
Wires
Harrington spine
instrumentation
Mandibular wire mesh
prostheses
Sutures
Stapedial prostheses for
middle ear repair
Neurosurgical aneurysm
and microvascular clips
Self-expanding stent
Balloon-expandable stent
Hydrocephalus
drainage valve
Trachea tube
Electronic laryngeal
prosthesis system
Electrodes and lead wires
Arzbaecher pill electrode
Cardiac pacemaker housing
Wire mesh
Retention pins for dental
amalgam
Preformed endodontic post
and core
Preformed dental crowns
Fixed orthodontic appliances
Percutaneous pin bone fixation
Variable capacitance transducer
Variable resistance transducer
Intrauterine device (IUD)
Intrauterine pressure-sensor
case
Osmotic mini-pump
Radiographic marker
Butterfly cannula
Cannula
Acupuncture needle
Limb prostheses, orthoses,
and adaptive devices
Applications
Internal fixation of diaphyseal fractures of cortical bone, metaphyseal and
epiphyseal fractures of cancellous bone; screws comprised of hexagonal or Phillips
recess driving head, threaded shaft, and self-tapping or non-self-tapping tip
Internal fixation of shaft and mandibular fractures: thin, narrow plate with slots or
holes for retaining screws
Internal fixation of fracture near the ends of weight-bearing bones: plate and nail,
either single unit or multicomponent
Internal fixation of long bones: tube or solid nail;
Replacement of total joints with metal and plastic components (shoulder, hip,
knee, ankle, and great toe): humeral, femoral (hip and knee), talus, and metatarsal
components
Internal tension band wiring of bone fragments or circumferential cerclage for
comminuted or unstable shaft fractures
Treatment of scoliosis by application of correction forces and stabilization of
treated segments: rod and hooks.
Primary reconstruction of partially resected mandible;
Wound closure, repair of cleft lip and palate, securing of wire mesh in cranioplasty,
mandibular and hernia repair and realignment, tendon and nerve repair
Replacement of nonfunctioning stapes: various types comprised of wire and piston
or wire and cup piston (synthetic fluorine-containing resin/stainless steel piston,
platinum and stainless steel cup piston, and all stainless steel prostheses)
Temporary or permanent occlusion of intracranial blood vessels; tension clips of
various configurations, approximately 2 cm (0.8 in.) or less in length and
constructed of one piece or jaw, pivot, and spring components (similar and
dissimilar compositions)
Treatment of tracheobronchial stenosis, tracheomalacia, and air collapse following
tracheal reconstruction: 0.457 mm (0.018 in.)
Dilation and postdilation support of complicated vascular stenosis (experimental):
Control of intercranial pressure: one-way valve
Breathing tube following tracheotomy and laryngectomy: tube-within-a-tube
construction
Electromagnetic voicing source following total laryngectomy: implanted unit
comprised of subdermal transformer, rectifier pack, and transducer encased in type
316 stainless steel, with spring steel diaphragm
Anodic, cathodic, and sensing electrodes and lead wires: intramuscular
stimulation, bone growth stimulation, cardiac pacemaker (cathode),
electromyography (EMG), electroencephalogram (EEG), and lead wires in a large
number of devices
Atrial electrocardiograms: swallowed sensing electrode of short metal tubing
segments forced over plastic tubing
Hermetic packaging of electronics and power source: welded capsule
Inguinal hernia repair, cranioplasty (with acrylic), orthopedic bone cement
restrictor;
Retention of large dental amalgam restorations: cemented, friction lock, and self-
threading pins, placed approximately 2 mm (0.08 in.) within dentin with
approximately 2 mm (0.08 in.) exposed
Restoration of endodontically treated teeth: post fixed within root canal
preparation, with exposed core providing a crown foundation
Restoration for extensive loss of tooth structure in primary and young permanent
teeth: preformed shell
Correction of malocclusion by movement of teeth: components include bands,
brackets, archwires, and springs
External clamp fixation for fusion of joints and open fractures of infected
nonunions: external frame supporting transfixing pins;
Measurement of pressure on sound: metal diaphragm, mounted in tension
Measurement of respiratory flow: metal arms supporting wire strain gage
Contraception: stainless steel (Majzlin spring, M-316, M device), stainless steel and
silicone rubber (Comet, M-213, Ypsilon device), stainless steel and natural rubber
(K S Wing IUD), stainless steel and polyether urethane (Web device)
Protective shroud for transducer
Continuous delivery of biologically active agents: implanted unit comprising
elastomeric reservoir, osmotic agent, rate-controlling membrane, and stainless
steel flow moderator and filling tube
Facilitation of postoperative angiography of bypass graft: open circle configuration
of 25 gage suture wires
Intravenous infusion
Coronary perfusion: silicone rubber reinforced with an internal wire spiral
Acupuncture: 0.26 mm (0.01 in.) diameter  5–10 cm (2–4 in.) length needles
Substitution, correction, support, or aided function of movable parts of the body,
and technical aids not worn by the patient: components such as braces, struts,
joints, and bearings of many items
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 10. Radiographic image of a failed stainless steel implant inside a patient [169].
3.7. Summary
Austenitic stainless steels remain popular for implant applica-
tions because of their availability, lower cost, excellent fabrication
properties, accepted biocompatibility and toughness. Their me-
chanical properties can be controlled over a wide range, for optimal
strength and ductility. However, 316L stainless steel implants are
often degraded due to pitting, crevices, corrosion fatigue, fretting
Fig. 11. (a) Stem fracture after 8.6 years. Endosteal bone lysis surrounds the proximal half of the stem. The stem has subsided 1.6 cm, partly within the cement, partly together
with the distal cement mantle. (b) Fractured Exeter stem after 3 years. Graft shows incorporation distal to the lesser trochanter and resorption proximal to it [172,173].
17
corrosion, stress corrosion cracking and galvanic corrosion in the
body. The wear resistance of 316L austenite stainless steel is also
relatively poor, and wear debris leading to allergic reaction in
surrounding tissue is another reason to restrict their application in
permanent implants. Today, 316L stainless steels are widely used
in a variety of surgical instruments and short-term implant devices
due to cost savings. This status is likely to continue for an indefinite
period. Nevertheless, the leading role of stainless steel as stem
materials in permanent hip prostheses has been maintained by the
development of high-nitrogen, nickel-free stainless steels. Cur-
rently, stems made of Orthinox stainless steel occupy 70% of the
market for total hip replacements in the United Kingdom.
4. Cobalt-based alloys
The cobalt–molybdenum superalloy was first developed by
Haynes for use in aircraft engines, and called stellite
[31,177,178]. It exhibited higher strength at elevated temperatures
and better corrosion resistance when compared to other super-
alloys [179]. Cobalt-based alloys were first used in medical
implants in the 1930s [180]. The CoCrMo alloy vitallium was
used as a cast dental alloy and then adapted to orthopedics
applications in the 1940s [44]. The corrosion resistance of CoCr
alloys is more than an order of magnitude greater than that of
stainless steels, and they possess excellent mechanical properties
[44]. By modifying vitallium, these alloys have been developed into
a number of compositions, as listed in Table 17.
4.1. Alloying chemistry: corrosion resistance
Cobalt–chromium based alloys are superior to stainless steels in
corrosion resistance, demonstrating excellent performance in a
chloride-rich environment, which is related to its chemical
composition. The high chromium content leads to spontaneous
formation of a passive oxide (Cr2O3) layer within the human body
environment [60,61,182–184]. The roles of Cr and other alloying
elements are summarized in Table 18. In brief, Cr, Mo and Ni are
responsible for corrosion resistance, very much like their roles in
18 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Fig. 12. The principal stress contours, in MPa, in a hip prosthesis. The maximal tensile stress in the stem is around 200 MPa, and approximately 350 MPa in the neck [154].

Table 17
CoCr alloys used in surgical implants [44,181].

ASTM standard Nominal compositions Cast/wrought status Medical application

F75-98 Co–28Cr–6Mo Cast Permanent implant

F90-97 Co–20Cr–15W–10Ni Wrought Short-term implant
F562-95 Co–35Ni–20Cr–10Mo Wrought Permanent implant
F563-95 Co–Ni–Cr–Mo–W–Fe Wrought Short-term implant
F799-99 Co–28Cr–6Mo Forged Permanent implant
F961-96 Co–35Ni–20Cr–10Mo Forged Permanent implant
F1058-97 Co–Cr–Ni–Mo–Fe Wrought Permanent implant
F1537-94 Co–28Cr–6Mo Wrought Permanent implant

Table 18
The roles of alloying elements.

Elements On corrosion resistance On microstructure On mechanical properties

Cr Cr2O3 to corrosion resistance Form Cr23C6  Enhance wear resistance

Mo Increase corrosion resistance Refine grain size  Enhance solid-solution strengthening

Ni Increase corrosion resistance  Enhance solid-solution strengthening

 Increase castability

C Form Cr23C6  Enhance wear resistance

 Increase castability

W Decrease corrosion resistance Reduce shrinkage cavity, gas blow  Enhance solid-solution strengthening
hole and grain boundary segregation  Decrease corrosion fatigue strength

stainless steels. Tungsten, which is added to increase solid-solution
strengthening, and to control the distribution and size of carbides,
can however impair the corrosion resistance and corrosion fatigue
strength of Co-based alloys. As such, Co–20Cr–15W–10Ni and Co–
Ni–Cr–Mo–W–Fe alloys are restricted to shorter-term implants, such
as bone plates and wire, due to both unsatisfactory corrosion
resistance and a high amount of toxic nickel release when used for
permanent implantation.
Cobalt alloys were initially used as cast components, with
wrought alloys coming into use later. The casting, wroughting and
forging conditions of the alloys have a significant influence on the
corrosion resistance and mechanical properties of the alloys. In
general, casting gives rise to coarse grains, grain boundary
segregations, gas blow holes, and shrinkage cavities in CoCrMo
alloys. Although the cast alloys are superior to non-cast alloys in
wear resistance, pitting resistance and crevice corrosion resistance,
they are inferior to wrought or forged alloys in terms of fatigue
strength and fracture toughness.
4.2. Biocompatibility of alloying elements: Co, Mo and W
The major alloying elements in cobalt alloys include
chromium, nickel, molybdenum and tungsten, in addition to
the matrix element cobalt. In this section, we review the
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
biocompatibility of three dominant elements – cobalt, molyb-
denum and tungsten.
4.2.1. Cobalt
4.2.1.1. Cobalt in the body as a trace element. Cobalt is an essential
trace element found principally in the maturation of human red
blood cells as a constituent of vitamin B12 (cyanocobalamin). Vitamin
B12 exists as two types of alkyl ligand: methyl and adenosyl.
Methylcobalamin promotes methionine synthesis. Methionine sup-
ply ultimately influences DNA synthesis [185]. Deoxyadenosylcoba-
lamin performs a key role in the energy metabolism of ruminants by
facilitating the metabolism of propionate, which is an important
precursor of glucose in ruminants [186]. Changes in lipid [187] and
amino acid metabolism [188] during cobalt deficiency have been
reported. In short, vitamin B12 plays a critical role in the extraction of
energy from proteins and fats. Cobalt (vitamin B12) deficiency cause
metabolic deficiencies and poor conception rates.
4.2.1.2. Toxicity of cobalt. Being an essential trace element, cobalt
can also cause serious adverse health effects at high exposure
levels [189,190], evidenced by the beer drinker’s cardiomyopathy
[25]. Like chromium and nickel, cobalt is received into the body via
the lungs, oral intake, by skin contact and from implant release
products. The median lethal dose (LD50) value for soluble cobalt
salts has been estimated to be between 150 and 500 mg/kg. Hence,
for a 100 kg person the LD50 would be about 20 g [191,192]. After
nickel and chromium, cobalt is a major cause of contact dermatitis
[193]. The toxicity of cobalt-containing dust also caused interstitial
lung disease in workers of the metallurgical industry [194].
With hip replacements made from cobalt alloys, there is in
general a local release of metallic particles, and in some individuals
there is a hypersensitivity reaction that causes more severe damage
to the tissues in the immediate vicinity of the prosthesis [195]. Cobalt
toxicity has been reported to contribute to the pathology of systemic
and neurological symptoms in some patients with metal-on-metal
hip prostheses after 4–5 years of implantation [196]. The symptoms
include painful muscle fatigue and cramping, dyspnea, inability to
perform simple motor tasks, decline in cognitive function, memory
difficulties, severe headaches and anorexia [196]. It has been
recommended that health care providers other than orthopedic
specialists be aware that patients with a cobalt implant are at risk for
cobalt poisoning, and might present with cardiac or some of the
listed neurologic symptoms [189,190,197].
4.2.2. Molybdenum
4.2.2.1. Molybdenum in the body as a trace element. Molybdenum is
an essential trace element for a number of enzymes important to
cellular metabolism. The most important enzymes that require
molybdenum are sulphite oxidase, xanthine oxidase, and aldehyde
oxidase. Sulphite oxidases catalyze the oxidation of sulphite to
sulphate, which is involved in the metabolism of sulphur-based
amino acids. Sulphite oxidase deficiency or absence leads to
neurological symptoms and early death. Xanthine oxidase cataly-
ses oxidative hydroxylation of purines and pyridines, including
conversion of hypoxanthine to uric acid. Aldehyde oxidase is
responsible for oxidizing purines, pyrimidines and pteridines, and
is also involved in nicotinic acid metabolism. Low dietary
molybdenum leads to low urinary and serum uric acid concentra-
tions and excessive xanthine excretion [198].
The human body contains about 0.07 mg/kg of molybdenum
[199], with higher concentrations occurring in the liver and
kidneys, and lower concentrations in the vertebrae. Molybdenum
is also present within human tooth enamel and may help prevent
its decay [200]. Dietary molybdenum deficiency has been
19
associated with increased rates of esophageal cancer in a
geographical band from China to Iran [201], possibly due to low
soil levels that end up in crops. Compared to the United States,
which has a greater supply of molybdenum in the soil, people
living in deficient areas have about 16 times greater risk for
esophageal squamous cell carcinoma [201].
4.2.2.2. Toxicity of molybdenum. Molybdenum is much less toxic
than many other metals (e.g. Co, Cr and Ni) of industrial
importance. Molybdenum does not constitute a hazard to human
beings either in trace concentrations occurring in environmental
pollution, or from exposure to higher concentrations encountered
in industrial processes and applications [198,202,203]. Neverthe-
less, molybdenum dust and fumes generated by mining or metal
working can be toxic with chronic exposure. Low levels of
prolonged exposure can cause irritation to the eyes and skin.
Direct inhalation or ingestion of molybdenum and its oxides
should be avoided. Hence, precautions are recommended to avoid
repeated exposure of humans to molybdenum compounds,
especially particles of molybdenum metals and molybdenum
trioxide powders [202,204,205].
Acute toxicity has not been reported in humans, such as accidental
deaths due to molybdenum poisoning in industry. However, studies
on rats show a median lethal dose (LD50) as low as 180 mg/kg for
some molybdenum compounds. There is virtually no chronic toxicity
data for molybdenum in humans, but animal studies have shown that
chronic ingestion of more than 10 mg/day can cause diarrhea, growth
retardation, infertility, low birth weight and gout, with further effects
on the lungs, kidneys and liver [202,206]. Chronic exposure to
molybdenum and its compounds is blamed for some symptoms
including fatigue, headaches and joint pains [202,205,207].
Studies on the concentrations of chromium, cobalt and
molybdenum in patients with metal-on-metal total hip replace-
ment and hip resurfacing arthroplasty have shown that the level of
molybdenum in serum is generally low, compared with Cr and Ni
[208,209]. So far, no data are reported on systemic toxicity of
molybdenum, in regard of metallic implants.
4.2.3. Tungsten
Tungsten (atomic number 74) is chemically very similar
molybdenum (atomic number 42). However, the biological role
of tungsten has only recently been discovered in prokaryotes, but
not as yet in eukaryotes [210,211]. Tungsten is insoluble in water
and only very slightly soluble in nitric and sulphuric acids. The
toxicity of tungsten is very low; however, implantable materials
made from this metal degrade very rapidly in the body, and remain
in the serum, probably as tungsten particles [212]. Despite this, no
ill-effects were observed in patients given 25–80 g powdered
tungsten metal by mouth as a substitute for barium in radiological
examinations [213]. Moreover, the fact that patients can rapidly
recover both from seizures and renal failure whilst high levels of
tungsten persisted for several weeks in the patient’s serum and
urine suggests that there was no causal relation between tungsten
and toxicity. The excretion of tungsten is very rapid with a
biological half-life of a few hours for soft tissues [213]. Directly
implanted tungsten also showed very low toxicity, even though
significant levels could be detected after implantation [214]. While
the potential health effects of tungsten remain to be fully defined,
some researchers recommend caution [215].
4.3. Biocompatibility of cobalt alloys
The first application of cobalt alloy in hip implants was in 1939
[216], slightly later than stainless steels. In vitro evaluation showed
that the CoCrMo alloy is much less toxic than pure cobalt or nickel
due to its excellent corrosion resistance [217]. The huge clinical
20 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 19
Some physical properties of Fe, Co and Ni elements.

Elements Density (g/cm3) Structure Young’s modulus (GPa) Shear modulus (GPa) Melting point (8C)

Tensile Compressive

Fe 7.9 BCC [220] below 912 8C 211 170 82 1538

FCC [220] 912–1394 8C
BCC [220] above 1394 8C

Co 8.8 HCP below 417 8C 210 180 75 1493

FCC above 417 8C

Ni 8.9 FCC 200 180 76 1455

Ti 4.5 HCP [220] below 885 8C 116 110 44 1668

BCC [220] above 885 8C

success of CoCrMo stems in total hip replacements in1960s has
demonstrated that CoCrMo alloys are well tolerated in the body.
Recently CoCrMo hard-on-hard bearings were re-developed as a
result of improved mechanical properties of CoCrMo alloy and
enhanced surface polishing techniques. However, patients with
CoCrMo metal-on-metal bearing systems are still exposed to
wearing debris over a long period. The debris liberates cobalt and
chromium into blood via the synovial fluid, before being excreted
in the urine [218]. Recently, there have been increasing concerns
over elevated serum metal ion levels, potential teratogenic effects
and potent adverse local tissue reactions [196,197,199,218,219],
however the specific pathogenesis of these incidents remains
unclear.
4.4. Mechanical properties of medical-grade cobalt alloys
4.4.1. General mechanical properties
While the excellent corrosion resistance of Co–Cr-based alloy is
primarily imparted by alloying chromium, their superior mechan-
ical properties over stainless steels are due to the crystallographic
nature of the base element cobalt. With an atomic number of 27,
cobalt falls between iron and nickel on the periodic table. The
physical properties of cobalt are very similar to those of iron and
nickel (Table 19). The elastic modulus of pure cobalt is about
210 GPa in tension and about 180 GPa in compression, which are
similar to those of iron and nickel. The superior wear and fatigue
resistances of cobalt alloys over iron alloys primarily arise from its
two closely packed structures: hexagonal closely packed (HCP) and
face-center cubic (FCC) forms. The strengthening mechanism is the
solid-state phase transformation of part of the matrix from a FCC
crystal structure to a HCP structure in response to stress during
cold working. The presence of two equally close-packed but
distinct crystal structures poses a barrier to the motion of
dislocations and leads to pronounced strengthening. In addition,
the solid-solution-strengthening effects of chromium, tungsten,
and molybdenum, and the formation of metal carbides, all
contribute to the superb fatigue resistance of this alloy system.
For the purpose of comparison, it should be mentioned that iron
has two distinct crystallographic structures as well. However, the
body center cubic (BCC) structure is relatively loosely packed.
While the presence of the second BBC structured phase hardens
iron alloys, it makes the alloy susceptible to SCC, and lowers its
resistance to fatigue. For this reason, ferrite stainless steels find
limited applications in medical devices, and duplex stainless steels
have yet to make an impact in the biomedical field [44], as
summarized in Table 12.
As mentioned above, the cast/wrought/forged condition of the
alloys has a significant influence on the mechanical properties of
the alloys. In general, wrought/forged Co–Cr-based alloys have
higher strength values than cast counterparts. Table 20 shows that
forged, cold-worked, and hot isostatic pressed [221] wrought
cobalt alloys have substantial mechanical advantages over cast
alloys for orthopedic applications. The strength of human leg and
arm bones is in the range of 100–200 MPa and that of the vertebral
bodies is 1–10 MPa [222]. Finite element analysis shows that the
maximal tensile stress in the stem of hip prostheses in the body is
around 200 MPa [154]. The data summarized in Table 20 show that
the majority of cobalt alloys have yield strengths in excess of
500 MPa, well in excess of the above biological values, and thus
regarded to be safe for implant applications in the leg and arm.
However, cobalt–chromium alloys are difficult to machine. Hence,
the preferred method of producing cobalt-base alloy implants will
be a function of the trade-off between cost and properties
[44]. Where the properties of casting are sufficient, this method
will have priority. When maximum strength is required, hot
pressing and/or forging will be the preferred method.

Table 20
Mechanical properties of cast and wrought cobalt-based alloys [44].

Alloys Young’s UTS (MPa) 0.2% yield Elongation Smooth un-notched

modulus strength (%) fatigue strength/MPa
(GPa) (MPa) (106 cycles R = 1)

F75/Cast, annealed 210 650–890 450–520 15 200–310

F75/P/M HIP 250 1280 840 725–950
F799/Forged 210 1400–1590 900–1030 28 600–900
F90/Annealed 210 950–1220 450–650 NA
F90/44% cold worked 210 1900 1610 590
F562/Forged 230 1210 960–1000 500
F562/cold worked, aged 230 1800 1500 8 690–790
F563/annealed 230 600 280 50
F563/cold worked 230 1000–1310 830–1170 12–18
F563/cold worked, aged 230 1590 1310
F1058 wire 230 1860–2280 1240–1450
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
ASTM F75 is a cast Co–Cr–Mo alloy, and has a long history in the
aerospace and biomedical implant industries. In addition to its
excellent corrosion resistance, the main attribute of this alloy is
wear resistance. Distribution of Cr-rich carbides M23C6 and the
work-hardening ability of this alloy greatly enhance wear
resistance. Total artificial hip joints made from this alloy are even
more viable, due to its excellent tribological properties against
plastic sockets, i.e. in a metal-on-plastic joint. A further feature of
F75 is the extremely large grain size observed in the original
casting. A finer grain size in general results in superior mechanical
properties, hence wrought Co-alloy came into use [44].
ASTM F799 is a wrought version of F75 alloy, mechanically
processed by hot forging rough billets to make the final shape. To
have a good ability to be forged, the carbon content of a wrought
alloy must be sufficiently low. Although the low carbon content
compromises wear resistance of wrought alloys, they are superior
in fatigue strength imparted by the wrought microstructure. As
mentioned previously, the primary strengthening mechanism in
wrought cobalt alloys is the solid-state phase transformation of
part of the matrix from a FCC to a HCP crystal structure by cold
working. The presence of two distinct crystal structures poses a
barrier to the motion of dislocations and leads to pronounced
strengthening. The strength (fatigue strength, yield strength and
UTS) of F799 alloy is approximately twice those of as-cast F75
(Table 20). Modern Metal-on-Metal joint prostheses are almost
always made from the F799 alloy (Fig. 13).
ASTM F90 is a wrought Co–Cr–W–Ni alloy. Tungsten and nickel
are added to improve machinability and fabrication characteristics
[223]. In the annealed state, the mechanical properties of F990 are
similar to those of cast F75, but when cold worked to 44%, the
properties can be improved to be more than double (Table 16). This
difference in mechanical properties between F90 in the annealed and
work-hardened conditions means that caution must be taken to
ensure uniform deformation of the component [224], because
variations in mechanical properties may result in unexpected failure.
ASTM F562 is a wrought Co–Ni–Cr–Mo alloy [225]. The
microstructure of worked F562 is the HCP solid-state phase formed
from the FCC matrix by cold working. Subsequent age hardening in
the 425–650 8C range acts to further strengthen these two phases
through the precipitation of carbides, primarily Cr23C6 [225]. Cold
worked and aged F562 can exceed tensile strength levels in excess of
Fig. 13. Femoral bearing head and cups made from the F799 CoCrMo alloy. http://
www.zimmer.co.uk/z/ctl/op/global/action/1/id/9226/template/MP.
21
Fig. 14. Fatigue strength values (in air) of implant stainless steels and cobalt alloys
[148].
1795 MPa, which is the highest strength of any of the surgical
implant alloys (Table 20). It should be mentioned that even at high
yield strength levels of 1600 MPa, F562 alloy is still capable of
exhibiting elongations in excess of 8% [44].
The above four cobalt–chromium and cobalt–nickel–chromium
alloys are the most widely used cobalt alloys for implant applications.
Meanwhile other alloys have been rigorously tested and standardized
for medical devices, including F563, F1058, UNS R30004 (known as
Havar). These alloys have shown no adverse effects in terms of
cytotoxicity, systemic toxicity, intracutaneous or intramuscular
irritation, skin sensitization, blood hemolysis or pyrogenicity.
4.4.2. Fatigue properties
In general, the fatigue strength values of cobalt alloys are higher
than those of stainless steels in similar processing states (cast or
forged) (Fig. 14), and the cobalt alloys have a much lower notch
sensitivity compared to the both stainless steels and titanium alloys
[226]. It is also important to note that forged alloys, especially forged
cobalt–chromium, have significantly higher fatigue strengths than
the cast counterparts. Hot isostatic pressing (HIP), which introduces
fine microstructures, can further improve fatigue strength [227]. It
has also been indicated in Table 20 that the forged cobalt alloys have
fatigue strengths in excess of 500 MPa (in air) and thus could likely
be safe for orthopedic implant applications, such as for legs and
arms, the maximal strength of which is 200 MPa [222]. However,
the ultimate safety of these alloys has to be tested under
physiological conditions. The S–N data (Fig. 15) of cast and forged
cobalt alloys consistently shows that the fatigue strength and fatigue
life of CoCrMo alloys decreases in simulated body fluid (SBF),
therefore corrosion fatigue is very likely to occur when these alloys
are implanted. Considering the typical lifespan of joint replacements
of 107 cycles (Table 10), the corrosion fatigue strength value of cast
and wrought CoCrMo alloys in body fluid is below 100 MPa and
above 200 MPa, respectively, according to Fig. 15. Hence, cast cobalt
alloys are no less safe than stainless steels after 20 years service, and
wrought CoCrMo alloys are marginally safe, considering that the
maximal tensile stress in the stem of prosthesis in the body is around
200 MPa [154]. This can explain the relatively low success rate after
20 years implantation of total hip joint replacements (Section 6.3).
4.5. Medical applications of cobalt alloys
The superior fatigue resistance of CoCrMo alloys make them an
ideal choice of materials for total joint replacements. Indeed, the
22 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 15. S–N curves of (a) cast and (b) forged Co–Cr alloys in the air and in physiological solutions [228].
medical implant application of CoCrMo alloys started as a stem
material in total hip replacement in 1950s [180]. Currently 20% of
total hip replacements have stems and/or the hard-on-hard
bearing system made out of wrought CoCrMo alloys. As for total
knee and ankle replacements, the prostheses are almost exclu-
sively made out of CoCrMo alloys with UHMWPE as a lining.
The use of wrought, cobalt-based alloy for fracture-fixation
purposes is not as yet commonplace, probably as a result of its
increased cost compared with that of stainless steel.
The potential of cobalt alloys as coronary stents is currently
under investigation [229,230].
4.6. Current issues and challenges
Wrought CoCrMo alloys are expensive, which has limited their
percentage of the medical market, compared with stainless steels
[69]. Other issues associated with cobalt-based alloys include
stress shielding effects and metal toxicity. Cobalt alloys have a high
Young’s modulus (220–230 GPa), which is much higher than that
of cortical bone (20–30 GPa) [60,182], resulting in uneven energy
distribution from propagating shock waves during walking and
load bearing. The interface between a stress-shielded bone and an
implant deteriorates as the bone is weakened, with potential
loosening and fracture of the bone, the interface, or the implant.
The combination of stress shielding, wear debris, and motion at an
interface is especially damaging and often accelerates implant
failure. There also are concerns about the elements released from
cobalt-based alloys, as Ni, Cr and Co, are all found to be toxic. They
may cause systemic allergic reactions in the host body, which can
increase inflammation.
4.7. Summary
Cobalt based alloys are in general superior to stainless steels in
terms of resistance to corrosion, fatigue and wear. The range of
properties available for cobalt alloys make them suitable for a wide
range of orthopedic applications, including any of the metallic
components of all joint replacements, as well as fracture fixation
devices. Wrought CoCrMo alloys are currently the choice of
structural materials in permanent implants at loading-bearing
sites (i.e. the joint system rather than the stem), providing a more
than 20 year service longevity. Long-term clinical use has proved
that these alloys have good biocompatibility in bulk form.
However, a number of issues remain with these alloy systems,
including failure due to fretting fatigue and corrosion fatigue,
aseptic loosening due to wearing, stress shielding effects and
biological toxicity due to Co, Cr, and Ni ion or particle release.
Although imperfect, wrought CoCrMo alloys remain the most
popular metallic implant materials of joint bearing systems,
improving the life quality of thousands of people worldwide.
5. Titanium alloys used as orthopedic implants
Titanium is a low density element (approximately 60% of the
density of iron and nearly half of the density of cobalt) that can be
strengthened considerably by alloying and deformation proces-
sing. Pure titanium undergoes an allotropic transformation at
approximately 885 8C, changing from a HCP crystal structure (a
phase) to a BCC crystal structure (b phase). Based on their
microstructure after processing, titanium alloys are categorized
into four classes: a alloys, near-a alloys, a–b alloys, and b alloys.
The modulus of elasticity of titanium and its alloys is about half of
stainless steels and cobalt–molybdenum alloys (Table 19). Com-
pared with stainless steels and cobalt–chromium alloys, titanium
is superior in specific strength (strength/density) but inferior in
tribological properties. The use of titanium alloys as biomaterials
has been increasing, due to their lower modulus, superior
biocompatibility and enhanced corrosion resistance compared to
stainless steels and cobalt based alloys. These attractive properties
drove the early introduction of commercial pure titanium (CP-Ti)
and a + b Ti–6Al–4V alloys, as well as the more recent develop-
ment of metastable b alloys [44].
5.1. Alloying chemistry: microstructure design
Unlike stainless steels and cobalt based alloys for which the
main purpose of alloying is to enhance corrosion resistance, the
primarily concern in the design of titanium alloys is mechanical
properties, because the matrix element titanium already has an
excellent corrosion resistance by itself. The transformation
temperature between a-HCP and b-BCC phases is sensitively
influenced by the alloying properties of titanium, which are thus
classified into a or b stabilizing additions, depending on whether
they increase or decrease the a/b transformation temperature.
The interstitial elements oxygen, nitrogen, and carbon are also
strong a stabilizers, which raise the transformation temperature
with increasing solute content; and hydrogen is a strong b
stabilizer, which lowers the transformation temperature with
increasing solute content (Table 21). The high solubility of oxygen
and nitrogen also makes titanium unique among metals and also
creates problems not of concern to most other metals. For instance,
heating titanium in air at high temperatures results not only in
oxidation, but in solid-solution hardening of the surface, due to the
inward diffusion of oxygen (and nitrogen). This surface-hardened
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 23

Table 21 (lower interstitial content) are lower in strength, hardness, and

a and b stabilizers of titanium alloys [220].
Role Interstitial Substitutional element
element
a stabilizers O, N, C, B Al, Ga, Ge, rare earth elements
b stabilizers H V, Mo, Nb, Cr, Fe, Si, Ta (often
used)
Cu, Ni, Pd, Co, Mn, W (limited
usage)
Primarily neutral Zr, Hf, Sn
with light a or
b stabilization
effect
zone of a-case (also referred to as the air contamination layer)
impairs the fatigue strength and ductility, such that the layer has to
be removed by machining, chemical milling, or other mechanical
means prior to implantation usage [44].
The substitutional alloying elements also play an important role
in controlling the microstructure and properties of titanium alloys
(Table 21). Aluminum is a strong a stabilizers. Other a stabilizers
include B, Ga, Ge and the rare earth elements, but their solid
solubilities are much lower, compared to aluminum or oxygen, and
none of these elements is in common use as an alloying element at
present [220].
There are many more elements solubilized in b titanium than in
a titanium. b stabilizer elements are preferred as additions because
they do not form intermetallic compounds, and titanium does not
form intermetallic compounds with them. Eutectoid systems are
formed with vanadium, molybdenum, niobium, chromium, iron,
copper, nickel, palladium, cobalt, manganese, and certain other
transition metals. They are usually added to alloys in combination
with one or more of the b isomorphous elements to stabilize the b
phase and prevent or minimize formation of intermetallic com-
pounds, which could occur at elevated temperatures during thermo-
mechanical processing, heat treatment or service [44]. The most
frequently used b stabilizing elements are V, Mo, Nb, Cr, Fe and Si.
Iron, chromium, and other compound formers are sometimes used
in b-rich a–b alloys or in b alloys, because they are strong b
stabilizers, can improve hardenability and respond to heat
treatment. Nickel, molybdenum, palladium, and ruthenium improve
corrosion resistance of unalloyed titanium in certain media [44,231].
In addition, some elements (Zr, Hf and Sn) are essentially neutral
in their behavior. These elements have significant solubility in both a
and b phases, but can lower the a/b transformation temperature
very slightly, and then increase it again at higher concentrations
[220]. Alumina, tin and zirconium are commonly used together in a
and near-a alloys. In a–b alloys, these elements are distributed
approximately equally between the a and b phases. Almost all
commercial titanium alloys contain one or more of these three
elements because they are soluble in both a and b phases, and
particularly because they improve creep strength in the a phase [44].
transformation temperature than those higher in interstitial
content, and have greater formability. Titanium alloys are further
subdivided into four groups: a, near-a, a–b, and b alloys. The a–b
alloy Ti–6Al–4V is the most widely used titanium alloy, accounting
for approximately 45% of total titanium production. Unalloyed
grades comprise approximately 30% of production, and all other
alloys combined form the remaining 25%. In terms of biomedical
applications, Ti–6Al–4V and its Extra-Low Interstitial (ELI) variant
(Ti–6Al–4V ELI) and CP-Ti grades are also the most widely used,
although the past decade has seen increased use of b-titanium
alloys for surgical implant applications. Table 23 lists applicable
ASTM standards for titanium and its alloys used for medical
implants [44].
5.3. Biocompatibility of alloying elements
Among the long list of alloying elements of Ti alloys, V, Al, Nb,
Zr, Mo, Fe and Ta are the most important for medical implants, as
shown in Tables 22 and 23. In this section, we will briefly discuss
the toxicity of Ti, V, Al and Nb. Biological functions and
biocompatibility of Zr and Ta will be discussed in Section 9.
5.3.1. Titanium
Titanium is not found in the human body, and does not play any
known biological role human body [232], and is non-toxic even in
large doses. When quantities of up to 0.8 mg of titanium were
ingested by humans on a daily basis, most titanium was found to be
excreted without being digested or absorbed [233]. Titanium
implants are not rejected by the body, and generally make good
physical connections with the host bone. In vitro, titanium can
however inhibit osteogenic differentiation of mesenchymal stem
cells [234] and may cause genetic alterations in connective tissue
[235]. Titanium particles also have size-specific biological effects
on white blood cells in vivo [236].
5.3.2. Vanadium
Vanadium plays a less defined biological role in the human
body [237,238], and can have both negative and positive cellular
responses [236], with toxicity mainly from its compounds such as
oxides [239]. Animal trials show that oral or inhalation exposures
to vanadium and vanadium compounds result in carcinogenicity
and various adverse effects on the respiratory system, blood
parameters, liver, neurological system, and other organs
[240,241]. There is lack of reports on the toxicity of vanadium
as alloying elements in Ti alloy implants, although a recent case
Table 23
ASTM/UN standards for titanium and titanium alloys used for medical implants.
Category ASTM UNs no. Materials

a microstructure F67 R50250 CP-Ti grade 1

5.2. Grading and classification of titanium and its alloys R50400 CP-Ti grade 2
R50550 CP-Ti grade 3
R50700 CP-Ti grade 4
Unalloyed titanium grading is primarily based on its content
of oxygen and iron (Table 22). Grades of higher purity a–b microstructure F136 R56401 Ti–6Al–4V ELI (currently
standardized)
F1472 R56400 Ti–6Al–4V (currently
standardized)
Table 22
F1295 R56700 Ti–6Al–7Nb (not
Impurity limits [63] for commercial pure titanium (wt%).
currently standardized)
Materials N C H Fe O F2146 R56320 Ti–3Al–2.5V (not
currently standardized)
ASTM grade 1 0.03 0.08 0.015 0.20 0.18
ASTM grade 2 0.03 0.08 0.015 0.30 0.25 b microstructure F1713 Ti–13Nb–13Zr
ASTM grade 3 0.05 0.08 0.015 0.30 0.35 F1813 R58120 Ti–12Mo–6Zr–2Fe
ASTM grade 4 0.05 0.08 0.015 0.50 0.40 F2066 R58150 Ti–15Mo
24 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

study has suggested a possible links between vanadium release Table 24

Basic mechanical properties of titanium and titanium alloys developed for
and implant failure [242].
orthopedic implants [44].

5.3.3. Aluminum Materials Young’s 0.2% yield Ultimate Elongation

modulus strength tensile (%)
Aluminum is a naturally abundant element, but has little
(GPa) (MPa) strength
known function in the human body. Despite its an acute toxicity (MPa)
only in very high doses, public awareness of chronic aluminum
a microstructure
toxicity, especially its links to neurological problems [243], has
ASTM grade 1 115 170 240 24
increased in recent years, due to its use in domestic cookware. ASTM grade 2 115 280 340 20
Aluminum is a major candidate in several diseases. By competing ASTM grade 3 115 380 450 18
with calcium for absorption, increased amounts of dietary alumi- ASTM grade 4 115 480 550 15
num may contribute to the reduced skeletal mineralization a–b microstructure
(osteopenia) observed in infants. Aluminum is also associated with Ti–6Al–4V 110 860 930 10–15
altered function of the blood–brain barrier [244] and brain Ti–6Al–7Nb 105 795 860 10
Ti–5Al–2.5Fe 110 820 900 6
neurotoxicity [245]. Like other metals, aluminum toxicity is also a
Ti–3Al–2.5V 100 585 690 15
major problem in people with kidney disease [246]. Aluminum
intolerance also causes contact dermatitis and digestive disorders. b microstructure
Ti–13Nb–13Zr 79–84 840–910 970–1040 10–16
Furthermore, newer evidence suggests that aluminum can bind to
Ti–12Mo–6Zr–2Fe 74–85 1000–1060 1060–1100 18–22
estrogen receptors, causing increased gene expression in human (TMZF)
breast cancer cells in vitro [247]. Hence, there are concerns that Ti–15Mo 78 655 800 22
excessive exposure to aluminum may increase the risk of breast Ti–15Mo–5Zr–3Al 75–88 870–970 880–980 17–20
cancer and other neurological conditions, such as Alzheimer’s Ti–15Mo–2.8Nb– 83 950–990 980–1000 16–18
0.2Si–0.26O (21SRx)
disease [248,249]. However, the use of aluminum cookware has not Ti–16Nb–10Hf 81 730–740 850 10
been shown to lead to aluminum toxicity in general and there is no Ti–(10–80)Nb 65–93 760–930 900–1030
scientific consensus regarding whether or not aluminum exposure Ti–35.5Nb–7.3Zr–5.7Ta 55–66 800 830 20
could directly increase disease risk. (TNZT)
Ti–(70–80)Ta 80–100 350–600 600–650 10–25
Ti–Ta–Nb/Nb/Sn 40–100 400–900 700–1000 17–26
5.3.4. Niobium Ti–Zr–Nb–Ta 46–58 – 650–1000 5–15
While niobium has little known biological roles in humans,
Stainless steels and Co-alloys
some niobium-containing compounds are toxic, including
316L 200 200–700 500–1350 10–40
niobates and niobium chloride, two chemicals that are water Co-alloys 240 500–1500 900–1800 10–50
soluble [250]. One recent study actually found niobium to be one
of the more toxic metal ions, along with cobalt, tested for their
ability to induce DNA damage and cause immune cell death
[251]. As more information becomes available, this element
should be treated with care, especially when several alloying
elements are used.
Table 25
Smooth (unnotched) fatigue strength of orthopedic implant alloys [44].
5.4. Biocompatibility of titanium alloys
Alloy Test conditiona Fatigue Fatigue
Compared with stainless steel and cobalt alloys, titanium limit at 107 limit/yield
cycles/MPa strength
alloys have proven to be superior in terms of biocompatibility
because of their excellent corrosion resistance [143,252]. Muta- a microstructure
CP-Ti (Grade 1) RBF (R = 1, 100 Hz) 88 0.5
genicity is not significant, as determined by in vitro mutation
CP-Ti (Grade 2) RBF (R = 1, 100 Hz) 215 0.8
assays, indicating that titanium alloys are safe for humans and CP-Ti (Grade 3 and 4) RBF (R = 1, 100 Hz) 430 0.6
animals [252]. However, the first generation titanium alloys,
a–b microstructure
represented by Ti–6Al–4V (Ti64), have been reported to cause
Ti–6Al–4V Axial (R = 1, 292 Hz) 500 0.6
allergic reactions in the human body [253]. Second generation Axial (R = 1, 292 Hz) 330 0.4
titanium alloys (b-titanium alloys) have been developed and RBF (R = 1, 60 Hz) 610 0.7
investigated with great interest. Some b stabilizing elements, Ti–6Al–7Nb RBF (R = 1) 500–600 0.7
such as Mo, Ta, and Zr, are used as alloying elements. They are Ti–5Al–2.5Fe RBF (R = 1) 580 0.8
Ti–15Mo–5Zr–3Al RBF (R = 1, 100 Hz) 560–640 0.5
considered to be relatively safe, compared with vanadium and (aged a + b
aluminum [254]. So far, there is a lack of long-term clinical microstructure)
application data and follow up reports on the biocompatibility of
b microstructure
b-titanium alloys. Ti–13Nb–13Zr RBF (R = 0.1, 60 Hz) 500 0.6
Ti–12Mo–6Zr–2Fe RBF (R = 1, 67 Hz) 525 0.5
5.5. Microstructure and general mechanical properties (TMZF)
Ti–15Mo–3Nb–0.3O RBF (R = 1, 60 Hz) 490 0.5
(21SRx)
This section focuses on basic mechanical properties of titanium
TNZT RBF (R = 1, 60 Hz) 265 0.5
alloys, including Young’s modulus, UTS, yield strength and TNZT-0.4O RBF (R = 1, 60 Hz) 450 0.5
elongation at break. Mechanical properties associated with special
Stainless steels and Co-alloys
working conditions, such as fatigue and friction/wearing, will be
316L Axial (R = 1, 120 Hz) 300 0.5
discussed in more detail in later sections. Nevertheless, both CoCrMo (cast) Axial (R = 1,100 Hz) 200–300 0.5–0.6
fatigue and friction properties are always related with those basic CoCrMo (wrought) Axial (R = 1, 20– 400–500 0.5–0.6
properties. To a coherent review, fatigue properties and wearing 100 Hz)
resistance are briefly mentioned as necessary. a
RBF, rotating bending fatigue.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 25

Table 26 (within grade 1 oxygen level), the fatigue limit (107 cycles) is
Fatigue strength of orthopedic titanium alloys [44].
approximately 88 MPa, while at 0.27% O (close to grade 2 oxygen
Alloy Smooth Notch fatigue Kfa KTi–6Al–4Vb level), the fatigue limit increases to approximately 215 MPa.
fatigue limit at 107 Alpha-alloys that contain small additions of b stabilizers have
limit at 107 cycles/MPa
further been classed as super-a or near-a alloys. Although they
cycles/MPa
contain some b phase, these alloys behave more like CP-Ti alloys
a–b microstructure than a–b alloys. To date, a and near-a alloys have not found
Ti–6Al–4V 500 290 0.6 ...
applications in medical implants. Their utility for medical devices
Ti–5Al–2.5Fe 580 300 0.5 ...
Ti–15Mo–5Zr–3Al 560–640 190 0.3 1.0 has been limited by their low strength at ambient conditions when
(aged a+b compared to a–b or b alloys (Table 27). For non-load bearing,
microstructure) corrosion resistant applications, CP-Ti grades are preferred.
b microstructure Therefore, the rest of Section 5 is devoted to a–b and b titanium
Ti–13Nb–13Zr 500 335 0.7 1.0 alloys.
215 0.4 1.3
Ti–12Mo–6Zr–2Fe 525 410 0.8 1.4
5.5.2. a–b titanium alloys
(TMZF)
In contrast to a titanium alloys (including CP-Ti, super or near a
316LVM 550 300 0.5 1.0 alloys), a–b alloys can be strengthened by solution treating and
Wrought cobalt alloys 725–950 Non- 1.4–1.8 aging. The microstructure in final products is mainly determined
sensitive by alloy composition, solution-treating temperature (b-phase
to notch composition at the solution temperature), cooling rate and section
a
K f ¼ ðNotch fatigue limit=smooth fatigue limitÞ (under the same test condi- size. Microstructure can also be significantly varied by the
tions), is a fatigued strength factor. subsequent aging treatment conditions, normally at 480–650 8C,
b
K Ti6Al4V ¼ ðFatigue limit of alloy=fatigue limit of Ti  F6Al  4VÞ (under the
to precipitate a and produce a fine mixture of a and b
same test conditions), is a fatigue strength factor relative to Ti–6Al–4V.
microstructure. Transformation dynamics, kinetics and micro-
structure in a–b alloys are generally complex. Excellent analysis of
the subject has been provided by a number of topic review articles
5.5.1. a titanium alloys and books [56].
Mechanical properties of titanium alloys, except for Young’s Solution treatment and aging can enhance the strength of a–b
modulus, are sensitively affected by their alloying compositions, as alloys by 30–50%, while their Young’s moduli remain at a similar
shown in Table 24. In CP titanium (98.9–99.6% Ti) which is level (Table 24). The fatigue strengths of a–b alloys (Tables 25 and
essentially all a titanium with relatively low strength and high 26) are also significantly increased by solution treatment and aging
ductility, UTS and yield strength vary from 240 to 550 MPa, and compared with those of a alloys, being higher than those of 316L
from 170 to 480 MPa, respectively, as a result of variations in the stainless steels (Table 15) and comparable with those of cobalt
interstitial and impurity levels. Oxygen and iron are the primary alloys (Table 20). Despite the intensive investigation into the
variants in the CP-Ti grades, with strength increasing as their precipitation of hardened a–b alloys [56], there are few products
content increases. Similarly, fatigue strengths are also increased made from solution-treated and precipitated-(age)-hardened
with higher levels of oxygen (Tables 25 and 26). At 0.085% O titanium alloys. As shown in Table 23, there are four ASTM

Table 27
Comparison of a, near a, a–b and b Ti alloys [56].

Ti-alloys Advantages Disadvantages Medical applications

CP-Ti 1. Excellent corrosion resistance
2. Excellent biocompatibility
3. Good weldability
1. Cannot be significantly strengthened
by heat treatment
2. Poor forgeability especially below
b transus due to HCP structure
3. Having a narrow forging temperature range
4. Low strength at ambient temperature
For non-load-bearing, corrosion resistant
applications: e.g.
1. Pace-maker case
2. Housings for ventricular-assist devices
3. Implantable fusion drug pump
4. Dental implants
5. Maxillofacial and craniofacial implants
6. Screws and staple for spinal surgery

a or near a As above As above Not yet

microstructure

a–b microstructure 1. Can be strengthened by Ti–6Al–4V and Ti–6Al–4V ELI

heat treatment 1. Total joint replacement arthroplasty
(hips and knees).
Ti–6Al–7Nb
2. Femoral hip stems
3. Fracture fixation plates
4. Spinal components
5. Fasteners, nails, rods, screws, and wires
Ti–3Al–2.5V
Tubing and intramedullary nails

b microstructure 1. High hardenability 1. High density

2. Good ductility and toughness, 2.
excellent forgeability and good 3.
cold rolling capability (formability) 4.
at the solution-treated condition
3. Good fractural toughness
Low creep strength
Low tensile ductility in the aged state
low resistance to wearing
26 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
standardized a–b alloys used in medical devices. Ti–6Al–4V and
Ti–6Al–4V ELI are currently standardized and the most widely
applied titanium alloys. Ti–6Al–7Nb and Ti–5Al–2.5Fe are
metallurgically similar to Ti6Al–4V, except for the absence of
vanadium, which has been reported to be toxic and to show
adverse tissue effects [44] as discussed earlier.
5.5.3. b titanium alloys
b titanium alloys are often described as second generation
titanium biomaterials. One of the rationales behind their
development is to address the stress shielding associated with
high Young’s moduli of implant. b titanium alloys generally have a
modulus closer to that of bone (Table 24) [255–259].
In b alloys, the b phase can be completely retained at room
temperature upon a sufficiently fast cooling, following solution
treatment. In the solution-treated condition, b alloys have
excellent forgeability and good cold formability [56]. b alloys
also have high hardenability. After aging at 450–650 8C, fine a
phase particles can precipitate and distribute throughout the b
matrix, which is a desired microstructure in terms of precipitate-
strengthening [56]. Compared with a–b alloys, aged b alloys
generally have enhanced strength (including UTS, yield strength
and fatigue strength) and fracture toughness, but higher density,
compromised creep strength and tensile ductility [44]. b alloys are
metastable, not only because b phase can partially transform to a
and/or other ordered phases upon heating to an elevated
temperature, but because cold work at ambient temperature can
induce a martensitic transformation [56]. Nevertheless, b alloys
exhibit almost no work hardening [44].
The rapid development of b alloys for orthopedic implant
applications took place in the 1990s [27,260,254]. The principle
advantages of b alloys include lower elastic moduli (Table 24),
good corrosion resistance and enhanced biocompatibility when
compared to Ti–6Al–4V and other a–b alloys. These alloys are
vanadium-free, with the principal alloying elements being
niobium, zirconium, molybdenum, tantalum, and iron, all of which
exhibit suitable biocompatibility (Fig. 4). The Ti–Nb–Zr–Ta system
(TNZT alloys) possesses the lowest elastic moduli of any metallic
implant alloy developed to date [27,260,254].
It is useful to compare a–b and b alloys with stainless steels
and cobalt alloys. The UTS values of a–b and b titanium alloys are
comparable with those of 316L stainless steels, but lower than
those of cobalt alloys. The yield strength values of a–b and b
titanium alloys are also comparable with those of 316L stainless
steels, but closer to the lower side of the range for cobalt alloys
(Table 24). A comparison of fatigue strength of these alloys
(Tables 25 and 26) indicates that a–b titanium alloys collectively
have higher fatigue strength values than 316L stainless steels, and
quite similar values to cobalt alloys. However, fatigue strengths of
b alloys are lower than cobalt alloys, which forms one of the
drawbacks of b titanium alloys. Another drawback of b alloys is
their poor resistance to wear. These challenges will be discussed in
Sections 5.6 and 5.7.
5.6. Fatigue properties
Fatigue properties are not only influenced by microstructure
(such as grain size, ratio of a/b phases, second phase particle size
and shape) inside of a material, but are also extremely sensitive to
the surface conditions of the end product in service. Because of the
latter factor, fatigue testing specimens are either smooth or
notched. The fatigue strength values of a–b and b titanium alloys
are typically reduced by 40% with notched specimens (Table 26).
For biomedical applications, the notch fatigue properties of
metallic biomaterials are considered to be indicative of the
performance of hip implants, because the stem of total hip
replacements rarely have a smooth surface, often structured with
wedges and porous coatings (Fig. 5) that introduce stress-
concentration sites [44]. Porous coated hip stems show a large
reduction in the fatigue limit as compared to the same alloy in
smooth condition [261].
Among CP-Ti, a–b and b alloys, b-titanium alloys generally
have a lower smooth fatigue resistance than a–b alloys with
similar yield strength values (Table 25). As discussed above,
however, the stress-controlled smooth fatigue behavior may not
be an appropriate characterization for orthopedic applications for
these alloys, so strain-controlled notch fatigue performance is
considered more representative of in vivo conditions [44]. Fatigue
strain is the ratio between fatigue stress and elastic modulus, s/E,
and in this respect the strain-controlled fatigue strength of TMZF
and TNZT b alloys is comparable to that of a–b Ti–6Al–4V alloy
(Table 26).
One drawback of b titanium alloys is their unsatisfactory
fatigue strength, compared with a–b alloys. There are two
approaches to enhance fatigue strength of these alloys, while
retaining low Young’s modulus. One method is to add Y2O3 [262],
SiO2/ZrO2 [263], SrO [264], etc. particles, which cause dispersion
strengthening mechanisms that prevent the sliding of dislocations.
In addition to refinement of the microstructure and improvement
of mechanical properties, the addition of Y2O3 also enhances the
corrosion resistance and thus the biocompatibility of b alloys
[262]. Another method is to precipitate a small amount of v phase.
This can be achieved by aging alloys at low temperature for a short
time. However, caution must be taken, as too much v phase will
cause alloy embrittlement, as well as increase in Young’s modulus
of the alloys [265].
5.7. Wear resistance
Hip joint simulation testing has shown that the wear rates of
UHMWPE against Ti–6Al–4V are 35% greater than that against Co–
Cr–Mo alloy. The high UHMWPE wear rates against Ti alloys have
been related to the mechanical instability of the metal oxide layer
[266]. It has been proposed that the surface passivative oxide layer
of Ti alloys can be broken down by externally applied stresses. The
damaged surface layer may not be able to heal immediately,
leading to further loss of alloy material locally. The uneven
consumption of metal material increases the surface roughness of
the metal surface, which further increases UHMWPE wear rates.
Moreover, the breakdown of the oxide layer produces hard oxide
debris, which acts as a third abrasive body, resulting in further
surface damage. Hence, titanium is not used to articulate against
any materials, but is used for components of modular constraints,
whereby a titanium femoral stem is used with a cobalt–chromium
or a ceramic ball to articulate against a UHMWPE line.
Wear of joint prosthesis materials is a key factor affecting the
longevity of total joint replacements. The incidence of nonspecific
pain and prosthesis loosening has been attributed to the
accumulation of UHMWPE and, to a lesser extent, metal or
ceramic wear debris [266]. Nonspecific pain is a result of adverse
tissue reactions, such as chronic joint inflammation. Prosthesis
loosening is a result of adverse (e.g. aseptic) reaction to wear debris
of the implant materials, as discussed in Section 2.4, accounting for
approximately 80% for surgical revisions [267–269]. The reason for
the failure of the implants is due to the release of wear debris into
the surrounding tissue, which results in bone resorption, where the
implant becomes detached from fixation sites. The revision surgery
is not only expensive, but its success rate is usually lower
compared to the first implantation [9].
Some surface modifications are applied to increase wear
resistance of titanium alloys, such as nitriding or plasma
treatment. Methods such as physical vapor deposition coating
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 27

(TiN, TiC), ion implantation (N+), thermal treatments (nitriding and
oxygen diffusion hardening), and laser alloying with TiC have been
examined for improving wear [266]. Of these, ion implantation has
been the most common treatment employed so far.
5.8. Bone bonding mechanisms of Ti alloys
The family of titanium-based biomaterials is favored by most
clinicians, surgeons, materials scientists and medical device
designers, because they are more inert and more biocompatible
in implant applications than stainless steels and cobalt alloys.
Titanium and its alloys designed for use in implant devices have
little or no reaction with surrounding tissues [181]. Titanium
derives its corrosion resistance from the stable oxide film that
forms on its surface, which can reform at body temperatures and in
physiological fluids if damaged. Among all metallic biomaterials,
titanium alloys are the only material system with the ability to
bond with bone. In general, once a virtually bioinert device (such as
those made of cobalt alloys) [270] is implanted into the bone, the
body naturally forms a capsule around that material in recognition
of it as a foreign body (Fig. 16a). Although this is a perfectly natural
defense mechanism, the formation of capsule tissue also contrib-
utes to the loosening of permanent implants. To the surprise of
surgeons, titanium implants demonstrate intimate integration
with host bone tissue (Fig. 16b).
The bone bonding mechanisms of synthetic biomaterials
including bioceramics (e.g. hydroxyapatite, related calcium phos-
components provide compression strength to bone. Roughly 1/3 of
the weight of bone is from collagen fibers. Collagen fibers are tough
and flexible, and thus tolerate stretching, twisting, and bending. In
natural bone, hydroxyapatite is always carbonated, with a
significant amount of PO43 being substituted by CO32.
Historically, two fundamentally distinct mechanisms were
proposed to explain bone bonding of implants. One is chemical
bonding, the other is biological bonding. Hench et al. first
conducted systematic studies on the bonding between bone and
implants in Bioglass1 [271–273]. They revealed that the formation
of carbonated hydroxyapatite on the surface of an implant was a
critical step prior to bone bonding [271–273]. They proposed
various weak chemical bonds between bone and bioceramics,
including ionic, hydrogen and van de Waals bonds [271]. On the
other hand, researchers from the field of medical science proposed
a biological bonding theory, in which collagen fibers became
inserted into the surface layer of the bioactive material, and thus
strengthened the interface [21].
A comparison of strength of all possible chemical bonds
(Table 28) indicates that only covalent bonding could provide
the strength revealed in these experiments. Hydrogen and Van
de Walls bonds are far too weak, and ionic bonds, which are
compressively strong in dry conditions, become as weak as
Table 28
Strength of chemical bonds [275].

phates, bioactive glasses) and Ti alloys had been intensively Bond type Length (nm) Typical chemical bonding
studied since the 1970s [270]. The bonding strength between the strength (kcal/mol)
host bone and the described implant materials is such that fracture In vacuum In water
of bone/implant samples occurs in either bone or implant material,
Covalent 0.15 90 90
rather than at the interfaces, when examined after surgical
retrieval [271–273]. This unexpected bonding strength fascinated Non-covalent
Ionic 0.25 80 1–3
researchers for nearly half a century until 2004, when TEM
Hydrogen 0.30 4 1
observations revealed a biological mechanism for this bonding Van de Waals 0.35 0.1 0.1
[274]. To understand this mechanism, it is necessary to learn the
chemistry of bone matrix first. Hydroxyapatite Ca10(PO4)6(OH)2
(HA) accounts for nearly 2/3 of the weight of bone. These inorganic

Fig. 16. (a) The capsule tissue (marked with arrows) formed at the interface between cobalt alloy implant and bone. (b) The infused interface between a Ti implant and host
bone.
28 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 17. (a) Covalent bonding between implant and tissues; and (b) biological bonding.
hydrogen bonds once in an aqueous environment. However, it is
hard to envisage how hydroxyapatite and collagen proteins in the
host bone could covalently bond with a bioceramic or Ti alloy
(Fig. 17a). On the other hand, the biological bonding via the
insertion of collagen into biomaterials (Fig. 17b) can well account
for the range of bond strength observed experimentally. One may
argue that the biological bonding is actually covalent bonding, as
the covalent bonds in collagen molecular chains are involved, in
the strengthening mechanism, and are just a function of fiber
surface area. However, the mechanism of chemical bonding and
biological bonding are different, as illustrated in Fig. 17, with
biological bonding observed using TEM (Fig. 18).
A critical question is whether a pure chemical or biological bond
alone can provide a strong and tough mechanical performance in
bone-ceramic systems. The chemical bonding between an apatite
layer (e.g. by epitaxial growth) and implanted HA crystal should in
Fig. 18. In biological bonding, collagen fibers insert into the surface layer of the
hydroxyapatite implant 3 month post implantation, delineated by the arrows [274].
theory be ionic in nature, like the chemical bonding of HA crystal,
and thus should be as brittle as HA crystal itself. Hence, chemical
bonding alone cannot toughen the interface. On the other hand,
flexible collagen fibers alone cannot harden the interface without
the resistance of strong HA crystallites. It is very likely that the
biological interdigitation of collagen fibers with HA and chemical
bonding of an apatite layer on the surface of HA implant are both
necessary in strengthening and toughening an HA/bone interface.
In addition, osteoblast processes might invade into the surface of
the implant, laying down collagen and HA, which eventually
ossifies to create a fibrous crystalline lattice.
Although the chemical bonding mechanisms proposed in the
studies of Hench and co-workers on Bioglass1 were less likely to
occur in reality, their work revealed a number of important
discoveries, among which the foremost is the formation of
carbonated hydroxyapatite on the surface of an implant as a
critical step prior to bone bonding [271–273,276]. Biological
bonding involves more complexity than the chemical mechanisms
proposed by Hench and other investigators [271–273]. First,
naturally occurring collagen proteins and carbonated HA have a
strong tendency to mingle with each other. Second, it is apparent
that once a carbonated HA is formed on the surface of an implant,
particularly a porous one, the collagen fibers of the host tissue are
likely to infiltrate, thus forming a biological bond between the
implant and host tissue, depending on the orientation and
thickness of the bone-material interface. Also, natural movement
between the two surfaces further induces collagen, because
healing osteoblasts are responsive to directional stress and strain.
In brief, the bone bonding mechanism involves two major steps:
 First Step – formation of a carbonated hydroxyapatite on surface
via ion exchange
 Second Step – collagen fibers of host bone insert into the
carbonated apatite layer [277]
In the case of Ti alloy, the key question is how a carbonated HA
could form on these metallic materials, and what is the nature of
the interaction. Although the above process is still not fully
understood, a widely accepted mechanism has been proposed by
Kokubo and co-workers [278,279]. Titanate (e.g. Na2TiO3) forms on
the surface of Ti alloy via ion exchange, which progresses further to
result in the formation of carbonated apatite.
While the ability to bond with host bone is generally desired
from the fixation point of view, this is a problem for temporary
implants or those that need to be retrieved due to rupture. As such,
titanium and its alloys are primarily used in permanent implants or
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
reasonably short-term temporary devices, which will be retrieved
before bone bonding is formed [280,281].
5.9. Clinical applications of titanium alloys
Commercially pure titanium (CP-Ti) and Ti–6Al–4V ELI are the
most commonly used titanium materials for implant applications.
Examples of biomedical applications for CP-Ti grades include
pacemaker cases, housings for ventricular-assist devices and
implantable infusion drug pumps, dental implants, maxillofacial
and craniofacial implants, and screws and staples for spinal surgery.
As shown in Table 23, there are four ASTM standardized a–b
alloys currently used for medical devices. Ti–6Al–4V and Ti–6Al–
4V ELI are the most commonly employed alloys. The applications of
Ti–6Al–4V include dental implants and parts for orthodontic
surgery; replacement parts for hip, knee, shoulder, spine, elbow
and wrist joints; bone fixation devices such as nails, screws and
nuts; housing parts for pacemakers and artificial heart valves;
surgical instruments and components in high-speed blood
centrifuges [282–284]. However, both Al and V released from
Ti–6Al–4V alloy are found to be linked to long-term health
problems, such as Alzheimer’s disease, other neuropathies and
osteomalacia [285]. Vanadium-free Ti–6Al–7Nb and Ti–5Al–2.5Fe
(not currently standardized by ASTM) have therefore been
developed. Both are metallurgically quite similar to Ti–6Al–4V.
Ti–6Al–7Nb has been used for femoral prosthesis stems, fracture
fixation plates, spinal components, fasteners, nails, rods, screws
and wires. Ti–5Al–2.5V is known for its excellent cold formability,
and its tensile properties are 20–50% higher than CP-Ti grades; it is
used mainly for tubing and intramedullary nails.
5.10. Current issues with titanium alloys as implant biomaterials
In general, titanium alloys produces adverse tissue reactions
from fretting of the prostheses, and these alloys show relatively
poor wear resistance in an articulating situation, compared with
cobalt alloys [31]. While a–b Ti alloys are superior in UTS and
fatigue strength, they have relatively poor bending ductility, due to
the HCP structure of the a phase. Although BCC structured b-Ti
alloys have enhanced shearing deformation ability, their UTS and
fatigue strength are compromised, compared with a–b alloys. A
recent clinical failure case revealed the unreliability of current
titanium hip joint implants (Fig. 19), with premature fracture of
the neck (Fig. 8) occurring more often with stems made of Ti64
Fig. 20. Illustrations of joints suggested by Gluck [289,292].
29
Fig. 19. A broken stem of titanium total hip replacement in a patient.
alloys than stainless steels and cobalt alloys. Finite element
analysis (Fig. 12) shows that the maximal bending stress at the
neck of hip prostheses is approximately 350 MPa [154], hence,
the poor bending performance of a–b alloys is primarily
responsible for the premature fracture at this site in total hip
prostheses.
To address the poor bending properties of a–b titanium alloys,
one of b titanium alloys, Ti–Mo–Zr-Fe (TMZF), was introduced by
Stryker as a stem material in the total hip replacement, branded as
Accolade, in 2000s [286]. However, Accolade TMZF Plus Hip Stems
were recalled in 2011 due to grit blast media, which was observed
in the drive hole [7]. ‘‘Grit blast media’’, a term referring to any fine
powder used as abrasive, was actually the wearing debris caused
by compromised wearing resistance of this soft BCC structured b
titanium alloy. A satisfactory titanium alloy that has excellent
bending fatigue strength and wearing resistance has yet to be
developed. The wear resistance of Ti alloys could be improved by
the incorporation of refractory hard metal elements (e.g. Ta, Zr, W,
and Nb), and/or by surface modification of the final product.
Among a number of surface modification techniques, ion
implantation has been the most widely employed and successful
technique for these alloys, as mentioned earlier.
5.11. Summary
The use of titanium alloys as implant biomaterials has increased
over the past two decades because of their lower modulus, higher
30 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
corrosion resistance, and better biocompatibility than both stainless
steels and cobalt-based alloys. Comparing with CP-Ti and a–b (Ti–
6Al–4V/Nb) alloys, the second generation b-Ti alloys have reduced
elastic modulus, while their strain-controlled and notch fatigue
resistances are enhanced. However, the poor shear strength of a–b
alloys and compromised wear resistance of b titanium alloys have
hampered their applications as implant materials in joint replace-
ment. A mechanically reliable titanium implant alloy has yet to
become a reality.
6. Stainless steels, cobalt and titanium alloys in total joint
replacement
6.1. Historical perspective of the development of total joint
replacements [180]
Developments in metallic implant materials have focused on
repairing long bones and joints. It is interesting to consider the
evolution of total joint replacement [287], which is one of the most
Fig. 21. (a) Philip Wiles’ metal-on-metal prosthesis made out of stainless steel. (b) Austin Moors prosthesis made out of Co–Cr alloys. (c) McKee–Farrar prosthesis. (c) Ring
prosthesis. (d) Chanley’s prosthesis.
successful operations of modern times, a notable development of
the twentieth century [288,289]. The innovative German surgeon
Themistocles Glück (1853–1942) led the way in the development
of joint implant fixation. In May 1890, Glück produced an ivory ball
and socket joint, which he inserted into the knee of a 17-year-old
girl and fixed the materials to bone with nickel-plated screws
(Fig. 20) [290,291]. Later versions introduced in the second half of
the 20th Century are not dissimilar to this original design of Glück,
who reported performing 14 arthroplasties in that year, including a
hip, but only provided details on five cases: three knee, a wrist and
an elbow [289,292]. The procedures appeared successful over the
short term; however, all of the five patients in the report suffered
from tuberculosis and developed complications of chronic infec-
tion. Three of the five prostheses were removed, except for the
wrist and one of the knee prostheses, which were left in situ. He
later realized that prior joint infection was a contraindication to
joint arthroplasty [289].
In 1893, Jules Pean, one of the great French surgeons of the 19th
century, attempted the total joint arthroplasty with a prosthesis
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 31

made of a platinum tube and a rubber ball. It had to be removed

two years later due to infection. It was not until 1919 that French
surgeon Pierve Delbet used a rubber femoral prosthesis to replace
half of a hip joint [293]. A subsequent trial with a more advanced
prosthesis was performed in 1925 by Smith-Petersen from Boston.
However, the glass and Bakelite design that he used could not
withstand the mechanical demands and eventually failed
[165]. Smith-Petersen modified it over a period of time with
celluloid, Bakelite, and finally Pyrex [290,294], which had been
introduced to the dentistry market.
Since the late 1930s, several passionate and well-known
surgeons had sought to improve the clinical performance of
artificial hip joints with unique design changes, revolutionary
surgical procedures and innovative usage of new and existing
materials [180]. In 1938, Philip Wiles at the Middlesex Hospital,
London, undertook six operations with the first metal-on-metal
THA made of stainless steel, which were fixed to the bone with
screws and bolts, but all failed due to loosening (Fig. 21a) [141]. In
1939, Frederick. R. Thompson of New York and Austin T. Moors, an
American surgeon of South Carolina, separately developed the Fig. 22. Worn cup of a total hip replacement.
entire ball of the hip joint. This hemi-arthroplasty addressed the
diseased femoral head only, but the acetabulum (hip socket) was
not replaced. In the same year, Austin Moors collaborated with Briton, Peter Ring of Redhill Surrey, started his clinical experience
Bohlman to design the long-stemmed prosthesis (Fig. 21b). The with cementless components as part of a metal-on-metal
original prosthesis they designed was a proximal femoral articulation in 1964 (Fig. 21d) [302,303]. However, McKee’s
replacement, with a large fixed head made of vitallium, a Co–Cr metal-on-metal prosthetics were challenged by Professor Sir John
alloy [180,216,295]. Moore performed the metallic hip replace- Charnley of Wrightington Hospital, UK. Charnley’s work on
ment surgery. This prosthesis functioned well and influenced the tribology resulted in a design (Fig. 21e) that almost completely
development of subsequent long stem femoral head prostheses replaced the others by the 1970s, resulting in perhaps the most
[296]. successful hip joint system yet devised. By the late 1970s, usage of
In 1948, a lot of attention was generated by the Judet brothers metal-on-metal devices was replaced by metal-on-polyethylene,
from Paris, who developed an acrylic prosthesis. The Judet however McKee–Farrar’s metal-on-metal implants continued to
prostheses turned out to be exceptionally susceptible to wear function well in many patients [219,304,305]. During the 1980s,
and failed [293]. Edward Haboush of the Hospital for Special the hard-on-hard model was re-evaluated because of the wearing
Surgery and Kenneth McKee of Norwich, England, also developed of the ultrahigh molecular weight polyethylene (UHMWPE) in
acrylic prostheses in the late 1940s experimenting with acrylic metal-on-polymer implants (Fig. 22). Today another wave of
cement for fixation. Their high incidence of failure also resulted metal-on-metal bearings is on the rise [293]. It is likely that the use
from loosening of the components [297]. of hard-on-hard bearings will remain or even continue to increase,
In the 1950s, McKee designed the first metal-on-metal joint, especially in young and active patients [219]. The chronology of
which had been modified throughout 1950s and 1960s. Later total hip replacement is summarily provided in Table 29.
versions of McKee’s design, referred to as the McKee–Farrar
prosthesis (Fig. 21c) were widely accepted and successful applied 6.2. Materials used in total joint replacements
clinically in the 1960s [298–300]. His early THA’s provided
surprisingly good results, with up to 97% of implants surviving As reviewed above, artificial hip joints have gone through many
at 17 years of follow up [301]. During the same period, another early stages of failure which later improved tremendously with the

Table 29
Historical evolution of total joint replacement [180].

Year(s) Inventor(s) Joints Materials Clinical outcomes

1890 Themistocles Glück/Berlin, Germany Knee Ivory ball and socket joint Failed due to chronic infection
Hip
Wrist
Elbow
1893 Jules Pean/Paris, France Shoulder Platinum tube and a rubber ball Removed 2 year later due to infection
1919 Pierre Delbet/La Ferte Gaucher, France Hip Rubber femoral Failed due to wearing
1925–39 Smith-Petersen/Boston, USA Hip Glass, celluloid, pyrex, bakelite and Failed due to mechanical fracture
vitallium
1938 Philip Wiles/London, UK Hip Metal-on-Metal stainless steel Poor clinical results
1948 Robert and Jean Judet brothers/Paris, Hip Acrylic resin Removed due to loosening
France
1950 Austin Moore/South Carolina, USA Hip Metal-on-Metal Removed due to loosening
Co–Cr alloy (vitallium)
1950–1970 Kenneth McKee/Norwich, UK Hip Metal-on-Metal stainless steels or Successful with surviving for up to 17 years
Co alloys
1960–1970 Peter Ring/Redhill, UK Hip Metal-on-Metal stainless steel, Successful with surviving for up to 17 years
Co alloy)
1962–pres John Charnley/UK Hip Metal-on-UHMWPE (stainless steel, Successful with surviving for about 25 years
Co alloy, or Ti-alloys)
32 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Fig. 23. The structure and components of a modern total hip replacement.

use of appropriate materials. Modern total hip replacements
comprise primarily of three components: stem, head and socket,
replacing the femoral stem and head, and pelvic acetabulum
respectively (Fig. 23).
Table 30 summarizes the materials used in hip replacement
prostheses. Stainless steels were first used in total hip replace-
ments by Philip Wiles in 1938, and a Co–Cr alloy was wisely chosen
by Austin Moore in the 1950s. The issues associated with the poor
corrosion, fatigue and wear resistances of stainless steels and the
consequent issue of heavy metal toxicity have since barred them
from applications in permanent implants, and stainless steels are
rarely used in permanent implant devices anymore. Today, the stem
portions of most hip implants are made of Orthinox, cobalt/
chromium- or titanium-based alloys. Cobalt–chromium-based
alloys or ceramic materials (aluminum oxide or zirconium oxide)
are used in making the ball portions, which are polished smooth to
allow easy rotation within the acetabular socket. This can be made of
metal, UHMWPE, or a combination of polyethylene backed by metal.
Table 31 lists current major manufacturers of hip replacement
prostheses and the materials used. Recent market data (Fig. 24)
shows that Charnley’s model Metal-on-UHMWPE bearing surfaces
still accounts for the largest proportion (approximately 60%) of all

Table 30
Biomaterials used in modern total joint replacements [219].

Products Stem Head Line Cup Year(s)

Metal-on-Metal before 1960s

Philip Wiles Stainless steels Stainless steels No line Stainless steels 1938
Austin Moore CoCr alloy (vitallium) CoCr alloy (vitallium) No line No cup 1950s
McKee–Farrar Stainless steels Stainless steels No line Stainless steels 1950s
CoCrMo alloys CoCrMo alloys Co alloys
Peter Ring Stainless steels Stainless steels No line Stainless steels 1950s
CoCrMo alloys CoCrMo alloys CoCrMo alloys

Metal-on-UHMEPE
Charnley Stainless steelsa Stainless steelsa UHMWPE Stainless steelsa 1960s
CoCrMo alloys CoCrMo alloys CoCrMo alloys
Ti alloys Ceramics (Al2O3 or ZrO2) Ti alloys

Metal-on-Metal since 1960s

CoCrMo alloys CoCrMo alloys No line CoCrMo alloys 1960s
Ti alloys

Ceramic-on-Ceramic
CoCrMo alloys Ceramics (Al2O3 or ZrO2) No line Ceramics (Al2O3 or ZrO2) 1980s
Ti alloys

Ceramic-on-Metal
CoCrMo alloys Ceramics (Al2O3 or ZrO2) No line CoCrMo alloys 2000s
Ti alloys
a
Today, stainless steels are rarely used in the joint components of total hip replacements, though some previously implanted stainless steel prosthesis are still in service.
Stainless steel 316L/CoCr couplings should be avoided.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 33

Table 31
Major orthopedic implant manufacturers and materials of some hip replacements.

Company Products Materials

1 1
Zimmer Holdings (ZMH) Zimmer ML Taper Hip Prosthesis with Kinectiv Technology Metal-on-UHMWPE wrought Co-based alloy
Metasul1 Metal-on-Metal articulation Forged Co-based alloy

Depuy (Johnson and Johnson) (JNJ) AML1 Total Hip System CoCrMo alloy
Austin Moore Hemi-Arthroplasty CoCrMo alloy
Prodigy1 Total Hip System CoCrMo alloy

Stryker Corp. Accolade TMZF TMZF alloy

Accolade C Femoral Component Forged cobalt–chrome alloy

Biomet (BMET) EchoTM Hip System Forged cobalt–chrome alloy

Taperloc1 Complete Hip Stem Titanium alloy Ti–6Al–4V

Sminth & Nephew (SNN) SMF Short Modular Femoral Hip Cobalt–chrome alloy
1
Wright Medical Perfecta RS Stem Titanium alloy
Perfacta1 Plasma Spray Stems Titanium alloy

Exactech AcuMatch1 L-Series (Cemented Stems) Forged cobalt–chrome alloy

AcuMatch1 L-Series (Press-Fit Stems) Forged titanium alloy

total hip arthroplasty devices. Ceramic-on-UHMWPE joints account
for 20%, with the remainder metal-on-metal, which has recently
risen to 20% [306].
With the huge success of application in total hip replacements,
cobalt alloys were soon used in other total joint replacements for
load-bearing sites, including knee and ankle joints [287]. Figs.
25 and 26 demonstrate the structures of knee and ankle
prostheses. As discussed in Section 2.4, the knee and ankle are
incongruent joints that involve stress concentration at their
dynamic working conditions. Hence, brittle ceramic is forbidden.
Cobalt–chromium alloys are suitably tough and represent one of
the two most widely used metals in knee implants, along with
titanium, although there is no consensus as to which material is
more suitable. Cobalt-based alloys are used to make the femoral
and tibial components in total knee replacement, and tibial and
talar components in total ankle replacements, between which
UHMWPE is used. Ta and Zr have also been explored for use in knee
prostheses. More details on this will be discussed in Section 9.
6.3. Current issues and challenges
The cumulative survival rate of joint replacements has been
improving over the past half century (Fig. 27), with the revision
rate at 15 years survival decreasing from 50% in the middle of
1990s [307] to around 10% in the late 2010s [305,306,308,309]. De-
spite the huge success, failure of the femoral stem and loosening
caused by wearing of acetabular components, remain major and
serious post-implantation complications of all total joint replace-
ment [31,145]. The failure incidence is especially high after
20 years implantation, being 30–40% [308]. Failure analysis has
consistently revealed that fretting corrosion fatigue is the major
cause of stem fracture, with wearing between bearing surfaces
responsible for aseptic loosening [31,35,145]. As discussed in
Section 4.4, the corrosion fatigue strength of wrought CoCrMo
alloys in simulated body fluid is around 200 MPa after 10-years
service, a similar level to the working stress of 200 MPa for a stem
of hip prosthesis under normal working conditions. Hence, fatigue
or corrosion fatigue is likely to cause the fracture of the stem.
It is also important to bear in mind that the difference between
estimates of material parameters under in vitro versus in vivo
conditions is such that the corrosion fatigue and fretting corrosion
fatigue strengths of metallic material implants may be over-
estimated by in vitro testing. The in vivo working conditions are
also mechanically and biologically complicated. Normal physio-
logical loading provokes a combination of various deformation

Fig. 24. Proportion of total hip arthroplasty by bearing surface across time [306]. Fig. 25. Total knee replacement.
34 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Fig. 26. Total ankle replacement.
strains in bone, including compression, tension, shear, torsion, and
vibration. Complex biological conditions include changes in pH,
low-level infection, inflammation, tissue remodeling and the
presence of macrophages and debris. Hence, the in vivo working
environment may be far more aggressive than we estimate, so
further research is needed to bridge this knowledge gap.
Although the three reviewed metals are not perfect implant
materials, they predominate in current orthopedics. Each alloy has
advantages and disadvantages, as summarized in Table 32. A
common issue is stress-shielding, associated with their high
modulus, as discussed earlier. In a physiological solution, the
fatigue life of CoCrMo constructs tends to be greater than Ti
constructs [310]. Titanium alloys are desirable for their low density
and superb biocompatibility, without metal-related toxicities.
However, one of their main disadvantages is low resistance to wear
and friction. Fretting fatigue associated with friction remains the
main reason for failure in orthopedic implants, especially in stems
of hip joint prostheses.
The ideal alloy should have the modulus of bone, the strength of
cobalt–chromium alloys, the corrosion resistance and biocompati-
bility of titanium, and the fabrication cost of stainless steels
(Table 32) [44]. To understand the challenge of achieving such an
ideal material, it is useful to compare the mechanical properties of
metallic materials with bone (Table 9 and Fig. 28).
The data in Table 9 and Fig. 28 indicate that the three orthopedic
metallic implant materials are much stronger and tougher than
bone, in terms of their general mechanical properties and fatigue
strength. Why, then, do these metallic materials not have the
Fig. 27. Cumulative improvements in survival rate reported in (a) 1996 [307] and (b) 2007 [308].
longevity of healthy bone? It is certain that corrosion and wearing
associated with the aggressive environment and harsh mechanical
working conditions in joints contribute in many clinical failures.
However, these conditions could also theoretically cause subclini-
cal micro-damage to bone as well. The difference lies in the critical
feature of bone, unlike inert artificial materials, in that bone
responds to mechanical impacts and heals while being subjected to
them. Cancellous bone has a complex microstructure, comprising
of directional fibers, multiple layers and porosity, which collec-
tively dissipate the energy from impacts through the bone itself,
while remaining light and strong, compared to synthetic materials.
Healing bone also undergoes gradual increases in modulus as it
heals, from cartilaginous to osseous. Without this biological
structure and self-healing ability, the ideal metallic biomaterial
that imposes no stress-shielding effect on host bone, while having
a satisfactory working longevity longer than 30 years is unlikely to
be realistic. Researchers have to face this reality and seek for
alternative strategies, such as tissue engineering and regeneration
using degradable materials (Section 7), as well as smarter
materials with more complex micro- and nano-structure.
7. Magnesium alloys
7.1. Three generations of biomaterials in terms of clinical outcomes
The development of biomaterials has experienced three distinct
generations, which meet three levels of clinical requirements. In
the early stages, biomaterials were expected to meet the essential
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 35

Table 32
Advantages and disadvantages of stainless steels, Co-alloys and titanium alloys.

Alloys Advantages Disadvantages Typical application

Fe-based (316L) Good corrosion and fatigue resistance Tend to be corroded in long-term applications Instruments
in short-term applications High modulus (stress shielding effect) Temporary implants
Low cost Ni and Cr allergy
Easy to be machined

Co-based Long-term corrosion resistance Difficult to machine and thus expensive Permanent implants
Super fatigue and wear resistance to process
Biocompatibility High modulus (stress shielding effect)
Ni and Cr allergy

Ti-based Light Lower shear strength Stem of hip prostheses

Greatest corrosion resistance Low wear resistance Dental screws (permanent implant)
Excellent biocompatibility Expensive
Relatively low Young’s modulus High modulus (stress shielding effect)

clinical requirement that the tissue being repaired is not harmed in
any way. To meet this criterion, virtually inert biomaterials were
designed and selected as implant materials. Examples included Co-
alloys, Al2O3, and stable polyurethane, which are typical first
generation biomaterials. Loosening of implants made from these
biomaterials drove researchers to seek more complex surface
bioactive materials that could bond to the host tissue. These
surface bioactive materials formed the so-called second generation
of biomaterials, represented by titanium alloys, hydroxyapatite
and related calcium phosphates. However, the limited longevity of
both first and second generation biomaterials, as well as the low
probability of overcoming this limitation, has driven scientists to
adopt a new strategy of tissue engineering and regeneration.
Biomaterials are now sought to act as a temporary structure, which
enables them to degrade and allow native tissue to integrate with
the implant and eventually replace it. Degradable biomaterials
therefore comprise the so-called third generation of biomaterials.
Table 33 summarizes the three generations of biomaterials.
7.2. Rationale of developing Mg-alloys as medical implants
Magnesium alloys fall into the third category of implant
biomaterials. The potential application of this alloy system for
orthopedic implant materials is based on the following potential
benefits of these alloys. First, magnesium is a benign macro-element
in the body (Table 3). As early as the 1930s, feasibility studies
showed good resorbability and high biocompatibility of magnesium
bone-fixation implants [311]. Second, magnesium has a similar
density and Young’s modulus similar to that of bone (E = 10–30 GPa)
(Table 9). Third, magnesium alloys have controllable corrosion rates

Fig. 28. Fatigue strength at 107 cycles of biomedical stainless steel, cobalt alloys, titanium and its alloys, and bone. Data without designation of rotating bending are those
obtained from uniaxial fatigue tests in air [228].
36 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 33
Three generations of biomaterials.

Generation Bioactivity Clinical goals Examples

First Biologically inert No harm to tissues Co-alloys

Al2O3
Polyurethane

Second Surface erosion Tissue-bonding Titanium alloys

Hydroxylapatite
Calcium phosphate
Surface bioactive glasses

Third Biodegradation Tissue regeneration Magnesium alloys

Degradable Bioglass1
Degradable polymers (PLA, PGA, etc.)

(and thus resorbability) in physiological media. Magnesium alloys
are possible candidates for use in both cardiovascular intervention
and bone repair. Table 34 provides a list of the compositions of major
magnesium alloys.
7.3. Corrosion of Mg alloys
The basic electrochemical characteristic of magnesium, with
standard electrode potential of 2.372 V, contributes to a low
corrosion resistance. The surface of magnesium implants passi-
vates and builds up a thin gray layer of magnesium oxide, when
exposed to air, which prevents further chemical reactions.
However, magnesium is severely attacked in saline environments
of the human body environment, which actually enables Mg alloys,
particularly Mg–Ca varieties, to be used as absorbable implant
materials [11]. Like biodegradable polymers, magnesium can be
entirely absorbed in the human body, but in contrast are
mechanically much stronger.
Magnesium alloys corrode/degrade in aqueous materials via
several different oxidation reduction reactions, which are influ-
enced by the alloying elements. Generally, the corrosion of
magnesium in water will yield magnesium-hydroxide and
hydrogen gas evolution. The net reaction is given below:
Mg þ H2 O ! MgðOHÞ2 þ H2 :
Typically, zinc is used as an alloying element, as it also
possesses the ability to displace hydrogen ions from solution. If
zinc is used as in magnesium–zinc alloys, then the following
reactions would also occur [312]:
Zn þ H2 O ! ZnðOHÞ2 þ H2 :
Magnesium metal can also displace zinc ions from solution:
Mg þ Zn2þ ! Zn þ Mg2þ : (3)
The hydrogen evolution in filigree implants, such as cardiovas-
cular stents, seems to be of minor importance [5]. Preclinical and
clinical studies indicate good biocompatibility, with minimal
inflammatory changes and complete absorption within two
months [12,313–316]. In bone biosynthesis applications, where
vascularization and nutrient transport is minimal, hydrogen
evolution however remains problematic [5]. As a result of the
increased Mg mass and poor transport mechanisms, gas pockets
occur around these implants. In animal studies, subcutaneous gas
bubbles have had to be removed by means of puncture procedures
[11,317]. The corrosion of magnesium is strongly influenced by
alloying elements in solid solution [5]. Experimentation has shown
that the addition of Zn has the ability to significantly decrease the
amount of hydrogen gas evolved when measured by electrical
corrosion testing [4].
There have been a number of alloying elements used in
magnesium based materials in an attempt to control their
corrosion properties and feasibility as implants. Elements like
Mn, Cu, Al, Ca, Zr, Gd, and Zn have all been explored (Table 35)
(1) [39]. Progress in the development of new compositions and high-
purity alloys has produced improved corrosion resistance and thus
a decrease in hydrogen evolution rates [318]. However, the
solubility of alloying elements in crystalline Mg is limited, and thus
corrosion rates can only be altered within a limited range.
Hydrogen evolution will thus remain a problem during the
degradation of crystalline Mg alloys.
(2)

Table 34
Composition (wt%) for magnesium alloy.

Type Group (example) Al Mn Si Zn Ag Cu Zr Nd Ni Y

Cast AM series (AM20) 2 0.5 – – – – – – – –

AS series (AS21) 2 0.4 1 – – – – – – –
AZ series (AZ61) 6 0.2 – 0.7 – – – – – –
EQ series (EQ21) – – – – 1.5–2 0.075 0.7 2.25 – –
EZ series (EZ33) – – – 2–3.1 – 0–0.1 0.5–1 3 0–0.01 –
QE series – – – – 2–2.5 – 0.6 2 – –
WE series (WE43) – – – – – – 0.5 3.25 – 4
ZC series (ZC63) – 0.5 – 6 – 2.7 – – – –
ZE series (ZE41) – 0–0.15 – 3.5–5 – 0–0.1 0.4–1 1.3 0–0.01 –

Wrought AZ series (AZ31) 3 0.3 – 1 – – – – – –

EA series (EA55RS) 5 – – 5 – – – 4.9 – –
WE series (WE54) – – – – – – 0.5 3.15 – 5.1
Z series (Z6) – – – 6 – – – – – –
ZC series (ZC71) – 0.75 – 6.5 – 1.25 – – – –
ZK series (ZK60A-F) – – – 6 – – 0.45 – – –
ZM series (ZM21) – 1 – 2 – – – – – –
ZW series (ZW3) – – – 3.25 – – 0.6 – –
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Table 35
The in vitro and in vivo corrosion rates of magnesium alloys [321].
Alloys In vitro electrochemical corrosion rate (mA/cm2)
0.9 wt% Hank’s SBF
NaCl solution
Pure Mg (99.85%) 15.98 80.06
AZ31 34 31.60
AZ91 10
WE43 22.56 30.60 16.03
ZE41 27
LAE442 30
AZ91Ca
AZ61Ca
Cast Mg–Mn–Zn 1.45–1.60
Extruded Mg–Mn–Zn 79.17
Extruded Mg–Zn–Y 1.88–4.47
Cast Mg–1Ca 546.09
Extruded Mg–1Ca 75.65
In contrast, the potentially much greater range of alloying
elements in an amorphous, single-phase structure allows the
production of particular metallic glasses with significantly
improved corrosion characteristics [5]. Research by Zberg and
co-worker’s [5] shows that hydrogen evolution during degradation
can be significantly reduced or even prevented completely using
glassy MgZnCa alloys. Moreover, because there is no dislocation-
based plastic deformation mechanism in amorphous alloys, bulk
metallic glasses show greatly improved strength and elasticity
[319,320]. Although MgZnCa glasses possess lower ductility, these
materials show strength nearly three times, and an elastic limit
nearly four times higher than those of crystalline Mg alloys,
although their Young’s modulus is similarly low despite the greater
amounts of alloying elements. Therefore, the combination of
increased mechanical strength, greater corrosion resistance and
the potential absence of hydrogen evolution during degradation
makes certain glassy Mg alloy compositions very interesting
materials for new biodegradable implant applications [5].
7.4. Metallurgical roles of alloying elements in magnesium alloys
A minor addition of calcium to Mg-based alloys in general
enhances their corrosion resistance significantly [311]. Mg–Ca
alloys containing up to 0.8 wt% Ca show a homogenous texture and
a uniformly distributed corrosion. However, higher percentages of
calcium lead to irregular and more widespread corrosion
[311]. The addition of Mn to magnesium alloys is primarily aimed
to improve their corrosion resistance. Corrosion due to Fe impurity
is suppressed by Mn atoms acting as local cathodes [322,323]. The
addition of Cu also increases the strength of magnesium casts, but
at the same time can accelerate corrosion rates in NaCl solutions
[324]. Magnesium alloys containing Al generally possess a good
combination of mechanical properties, corrosion resistance, and
die-castability [325]. Although aluminum atoms tend to contribute
to creep deformation at elevated temperatures, 125 8C or above
[326,327], substantial creep deformation is unlikely to occur at
body temperature of 37 8C. Al is also a passivative element, and
thus enhances corrosion resistance of alloys.
Zn is the most commonly used alloying element in magnesium
alloys, effectively improving their yield strength [327,328]. One of
desirable features of magnesium alloys for orthopedic implants is
its Young’s modulus of 40 GPa [329,330], which is close to that of
bone (10–30 GPa) (Table 9). Another important benefit of adding
Zn to magnesium is that Zn-rich alloys form less hydrogen gas
[5,331]. Two mechanisms are proposed for this effect. The first is
that the Zn2+ ions in solution immediately around a bulk material
37
In vitro immersion corrosion rate (mg/cm2/h) In vivo corrosion
rate (mg/cm2/yr)
m-SBF 0.9 wt% Hank’s SBF m-SBF
NaCl solution
0.011 0.038 1.17
0.0065 1.38
65.70 0.0028 1.56
0.085
0.0626 0.39
17.80
36.50
0.003–0.010 0.92
0.05
0.015–0.04 1.28
0.136
0.040
tend to be removed by the reaction of Eq. (3). As Zn2+ ions compete
with the Mg2+ ions in solution for binding with free OH anions,
Zn(OH)2 is formed, which ultimately decreases the amount of H2
gas.
7.5. Biocompatibility and toxicity of alloying elements in Mg alloys
As described earlier, the chemical toxicity of these metals inside
the body depends not only on the toxicity of the elements and their
compounds, but the concentration of released ions or wear
particles. Even a poisonous substance may be less toxic in
sufficiently low concentrations, while micronutrients can cause
adverse responses when present in excessive amounts. Hence, to
design biologically safe, degradable magnesium alloys, it is
important to know the biological safe limit and release kinetics
of the relevant element used in the implanted material. It is also
important to understand that accumulation of elemental metals is
tissue specific; brain tissue, for example, has a high affinity for
metal ions.
7.5.1. Magnesium
Magnesium is needed for more than 300 biochemical reactions
in the body. At the biochemical level, magnesium is involved in
energy metabolism and protein synthesis, maintains normal
muscle and nerve function, supports a healthy immune system
and keeps bones strong. Magnesium also helps regulate blood
sugar levels and promotes normal blood pressure, playing
important roles in preventing and managing disorders such as
hypertension, cardiovascular disease and diabetes [332,333]. Die-
tary magnesium is absorbed in the small intestines and is excreted
through the kidneys [332,333]. The kidneys are efficient at
excreting excess magnesium and it is unlikely that the mineral
will accumulate to toxic levels, although there is a risk of renal
dysfunction with an overdose of magnesium, due to precipitation
of magnesium salts. A high intake of magnesium can compete with
calcium, and lead to impairment of its absorption [334]. Symptoms
of magnesium overload include diarrhea, difficulty breathing and
depression of the central nervous system, causing muscle
weakness, lethargy, sleepiness or even hyperexcitability [335].
7.5.2. Calcium
Calcium is the most abundant mineral and mainly stored in
bones and teeth. Other diverse biological roles in the human body
including blood clotting and co-activation and stabilization of
enzymes [336–338]. Compared with other metals, the calcium ion
and its compounds have very low toxicity. This is not surprising
38 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

given the high natural abundance of calcium compounds in the
body. Calcium poses few serious clinical problems, with kidney
stones the most common side-effect, and local tissue calcification
sometimes occurring around dead or dying tissue. Acute calcium
poisoning is rare, although excessive consumption of calcium
carbonate supplements can potentially cause renal failure
[339,340]. Consumption of more than 10 g/day of CaCO3 (=4 g
Ca) is at risk of developing milk-alkali syndrome [341].
7.5.3. Manganese
Manganese is an essential trace mineral, which plays a number
of roles in cellular systems, as cofactors for metalloenzymes,
including oxidases and dehydrogenases, DNA and RNA poly-
merases, kinases, decarboxylases and sugar transferases
[342,343]. In humans, excessive exposure to Mn has been reported
to induce ‘‘Manganism’’ which is a neurological disorder similar to
Parkinson’s disease [344,345].
7.5.4. Copper
Cu is another essential trace element present in human and
animal tissue [343,346]. The human body contains copper at a level
of about 1–3 mg/kg [347,348]. Cu is absorbed in the gut, then
transported to the liver [349,350]. After processing in the liver, Cu
is distributed to other tissues by binding to ceruloplasmin in the
blood stream. Normally about 1 mg per day of dietary Cu is
absorbed in the body, and excreted from the body, with some
ability to excrete excess Cu via emulsification with bile secretions
[351]. Excessive Cu amounts in the body have been linked to
neurodegenerative diseases including Alzheimer’s, Menkes, and
Wilson disease [352]. Some research suggests that serum levels
with either high Cu and low Mg alone, or in combination with low
Zn, can increase the mortality risk for middle-aged men [353].
7.5.5. Zinc
Zn, as a trace mineral in the human body, is also essential for
hundreds of biological enzymes and transcription factors, which
are often coordinated with amino acids [354]. Zn is less
detrimental than Mn, Al (see Section 5.3), and Cu, because it is
readily absorbable by biological functions within the cell
[354]. Like Ca, excessive amounts of Zn have the potential to be
corrosive in nature if ingested [355]. In a biological system, Zn2+
ions can form ZnCl2, which has been shown to damage parietal
cells lining the stomach. The pathology and toxicology of Mg and
the above listed alloying elements are summarized in Table 36.
7.6. Mechanical properties of Mg-alloys
Magnesium alloys have UTS and elongation in the range of
90–280 MPa and 3–20%, respectively (Table 37). The rapid
reduction in strength due to degradation is in principle not an
issue to tissue engineering applications, as the degradable implant
is expected to provide temporary support rather than a permanent
substitute for bone. The mechanical properties of magnesium alloy
can be tuned by the alloying and processing history. The addition of
Al, Ag, In, Si, Sn, Zn, and Zr can improve both strength and
elongation [358], while hot rolling, hot extruding, and ECAP can
further improve strength. However caution should be taken, as
these processes can deteriorate the ductility of alloys [358,
359]. The large number of variables in alloying and processing
provide a range of opportunities to improve the strength and
flexibility of magnesium alloys [321].
7.6.1. Mg–Zn-based alloys
In general, the addition of Zn up to 3 wt% reduces the grain size
and enhances the mechanical properties of the alloy matrix, with
Young’s modulus, UTS and yield strength all increasing (41–
45 GPa, 90–230, and 20–130 MPa, respectively) (Table 37). Beyond
3 wt% Zn, grain size and strength remain the same, while
elongation decreases significantly (Table 37). The fracture behavior
of Mg alloys also changes from nearly complete cleavage fracture
to quasi-cleavage fracture with the addition of Zn. When Zn
content exceeded 3 wt%, (Mg, Zn)-rich particles act as the crack
initiation sites [360].
7.6.2. Mg–Ca-based alloys
The influence of the amount of Ca on the mechanical properties
of Mg–Ca-based binary alloys has been investigated by a number of
research groups [311,361–365]. The addition of Ca up to 1.0 wt%

Table 36
The pathology and toxicology of Mg and its alloying elements [321,356,357].

Elements Blood serum Biological role Toxicology Daily allowance

level

Mg 0.9 mmol/L Activator of many enzymes; coregulator of Almost no evidence indicates toxicity of 0.7 g
protein synthesis and muscle contraction; magnesium
stabilizer of DNA and RNA
Ca 1.3 mmol/L Most abundant mineral and mainly stored in Calcium metabolism disorder; kidney 0.8 g
bone and teeth; participation blood clotting; stones
activator or stabilizer of enzymes
Cu 1.1–1.5 mg/ml Copper is mainly carried by ceruloplasmin in Neurodegenerative diseases including 2–3 mg
blood Alzheimer’s, Menkes, and Wilson
disease
Zn 46 mmol/L Essential trace element; appears in all enzyme Neurotoxic and hinder bone 15 mg
classes development at higher concentration
Mn 1 mmol/L Essential trace element; activator of enzyme; Mn Excessive Mn results in neurotoxicity 4 mg
deficiency is related to osteoporosis, diabetes
mellitus, atherosclerosis
Si Cross linking agent of connective tissue basement Excessive SiO2 causes lung diseases
membrane structures; necessary for growth an
bone calcification
Li 2–4 ng/g Used in the treatment of manic-depressive Reduced kidney function and central 0.2–0.6 mg
psychoses nervous system disorders
Al 2.1–4.8 mg Neurotoxicity and accumulation in Total amount in
bone human <300 mg
Zr Total, <250 mg High concentration in liver and gall 3.5 mg
bladder
Rare earth <47 mg Compound of drugs for treatment of cancer Accumulation in bone and liver
elements
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 39

Table 37 7.7. Potential applications and challenges of magnesium alloys

Mechanical properties of Mg–Zn- and Mg–Ca-based alloys [311,321].

Alloys Young’s UTS 0.2% yield Elongation As third generation metallic biomaterials, the development of
modulus (MPa) strength (%) magnesium alloys as degradable biomaterials has been primarily
(GPa) (MPa)
driven by applications in tissue engineering. A number of issues
Mg-cast 41 87 21 13 must be addressed prior to any clinical applications, among which
AZ91D-die cast 45 230 150 3 hydrogen generation is the most challenging one. Caution is
AZ31-extruded 45 235 125 7
advised also for the toxicity of any alloying elements used in new
LAE442 247 148 18
AE43-extruded T5 44 280 195 10 alloy compositions.
AM60B-die-cast 45 220 6–8
Mg–Zn–Mn extruded 180 246 22 7.8. Summary
Mg–1Ca-extruded 240 135 11

Magnesium alloys have attracted much attention from the field

leads to an increase in tensile strength up to approximately
220 MPa, and an increase of 0.2% results in an increased yield
strength of 140 MPa. This is an indication of relatively high
plasticity. No significant further increase in the tensile strength can
be achieved above 2.0 wt% calcium (Fig. 29) although that calcium
content dramatically affects the elongation at rupture (Fig. 29),
which reaches a maximum value of 14% at 1 wt% Ca.
of orthopedic implant materials and tissue engineering, because
the major alloying elements (Mg and Ca) can be tolerated by the
body at relatively high levels. The alloys are also mechanically
compatibility to bone and are suitably biodegradable. However,
several issues must be addressed prior to clinical applications,
including hydrogen bubble generation and infection. Magnesium
glasses also seem to offer a great alternative to crystalline alloys as
an opportunity to address the issue associated with the hydrogen
bubble effect.

Fig. 29. Tensile strength, 0.2% elastic limit, elongation at rupture, and elongation at tensile strength of several Mg–Ca alloys [361,365].
40 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Fig. 30. The use of NiTi in vascular stents is based in its superelasticity and shape memory effect.

8. NiTi shape-memory alloys [102]
The shape memory effect refers to the ability of materials to
return to their pre-defined dimensions upon heating, induced by
deformation training. Subjected to a proper entrainment, shape
memory alloys can undergo reversible deformation via a phase
transformation mechanism, also referred to as pseudoelasticity.
The shape memory effect exists in a number of alloy systems: Au–
Cd alloys discovered by Ölander in 1932 [366], Cu–Zn alloys by
Greninger and Mooradian [102] and NiTi alloys by Buehler and his
co-workers in the 1960s. In addition to Au-, Cu- and NiTi based
alloys, the shape memory effect has also been discovered in InTl,
NiAl, FePt, FePd, MnCu, and FeMnSi [367]. Among these alloys, NiTi
is the most attractive one, characterized by large plateau (super-
elasticity), relatively stable cyclic performance, good workability
and good resistance to corrosion and fatigue. Hence today, most
shape memory devices are produced from NiTi [367].
In the medical field, research to exploit the potential of NiTi as
an implant material was triggered in 1968 by the finding that NiTi
showed excellent corrosion resistance in sea water [368,369]. The
use of NiTi for medical applications was first reported in 1973
[370], leading to clinical trials in orthopedics in the early 1980s
[371]. It was only in the mid-1990s, however, that the first
widespread commercial intra-vascular stent applications made
their medical breakthrough (Fig. 30) [102].
8.1. Corrosion of NiTi alloy
As discussed in Section 2.2, the human body provides a relatively
aggressive environment for long term implants. However, most of

Table 38
The corrosion behaviors of NiTi alloys [102].

Testing conditions Major results Refs.

37 8C, Hank’s solution NiTi has better corrosion resistance than Co–Cr–Mo or 316L stainless steel [372]
37 8C, 0.9 wt% NaCl solution NiTi is more sensitive to corrosion than titanium. Pitting of the NiTi surface was observed [373]
37 8C, Hank’s solution NiTi has a better resistance to the chemical breakdown of passivity, compared to 316L [374]
37 8C, 0.9 wt% NaCl solution When stainless steel (316L) was coupled with NiTi, 316L was found to suffer from crevice corrosion [375]
Retrieved implants NiTi wires are no more subject to corrosion than stainless steel [376]
Tested in artificial saliva The release rates of nickel from stainless steel and nickel–titanium arch wires were not significantly [377]
different
Evaluated in physiological The Ni ion release was three times higher for NiTi than for austenitic stainless steels [378]
simulating fluids The characteristics of the passive film formed on NiTi are not so good as those on Ti6–Al–4V, but are
comparable or inferior to those on austenitic stainless steels
Ringer’s solution Annealed NiTi to be more corrosion-resistant than cold-worked material. Thus, the heat treatment [379]
and mechanical working had a significant influence on corrosion behavior. The same study also
indicated that straining of NiTi led to significant improvements in corrosion resistance. This may be
due to the development of a single martensite variant during deformation
In vivo: No generalized or localized corrosion on NiTi. Neutron activation analysis of distant organs in the [380]
NiTi plates, 17 months after same study showed no accumulation of trace metals from NiTi
implantation in dogs
Implanted 44 NiTi intraluminal 1. Only minimal corrosion was seen at 6 months [381]
stents in the iliac arteries 2. Pitting was the predominant type of corrosion. The pit penetration rate was estimated to be
of 22 sheep approximately 0.0046 cm per year
3. Corrosion product analysis around the pit sites indicated that the main product of pitting was a
titanium-bearing compound, probably an oxide
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 41

the knowledge on the corrosion behavior of NiTi is from in vivo
studies of dental arch wires and other models assessed under in vitro
conditions. Knowledge of the corrosion behavior of NiTi inside the
body, when used in implantable devices, is limited. Table 38 lists the
major results on the corrosion behaviors of NiTi alloys. In general,
NiTi has a better corrosion resistance than CoCrMo or 316L stainless
steel, but is more sensitive to corrosion than titanium.
8.2. In vitro evaluation of biocompatibility of NiTi alloy
NiTi is one of the most innovative concepts introduced in the
field of metallic biomaterials, but its biocompatibility remains
controversial [102]. Although we do not know the exact
concentrations of metallic compounds released from implanted
material due to the complicated local conditions (pH, fretting, etc.),
the high nickel content of NiTi (50 at.%) is of concern, as deleterious
amounts of nickel may be released [102], as discussed in Section
3.2.3. In this and next three sections, in vitro and in vivo evaluation
of NiTi biocompatibility will be reviewed.
There are only a small number of in vitro studies on the
cytocompatibility of NiTi (Table 39). In general, except for one
study [382], the cytocompatibility of NiTi is comparable with
CoCrMo alloys, 316 stainless steel and Ti-alloys. NiTi alloy can
therefore be regarded as a biologically safe implant material. More
recently, cytocompatibility of NiTi after surface modifications
including plasma treatment [383–385] and oxidation [386] were
assessed. In general, surface treatment of NiTi alloy encourages cell
attachment and proliferation, compared with untreated surfaces.
It must be mentioned that the releasing kinetics of metal ions
from alloys designed to be anticorrosive, such as NiTi, is slow. Past
clinical applications have demonstrated that the toxicity of
released metal ions from such materials could only be diagnosed
after years of implantation. Hence, the short-term in vitro
evaluation data on the toxicity associated with metal ion release
from bulk specimens should not be regarded as a reliable indicator
of long-term in vivo effects of the implants.
8.3. In vivo evaluation of biocompatibility of NiTi in animal models
Comprehensive evaluation of NiTi biocompatibility was first
made in 1976 by Castleman et al. [380] using a canine model. In
their work, neutron activation analyses were carried out on a

Table 39
In vitro studies on biocompatibility of NiTi alloys.

Cells Control materials Major results Refs.

Human fetal lung fibroblasts
Human fibroblasts
L-929 fibroblasts
Human gingival fibroblast spreading
Rat splenocytes
Human peripheral blood lymphocytes
Human peripheral blood lymphocytes
Fibroblasts
Murine fibroblasts and osteoblasts
Human osteoblast-like osteosarcoma
cells (SAOS-2, MG-63), primary
human osteoblasts [126], and
murine fibroblasts (3T3)
Rat osteosarcoma cell line ROS-17
Osteoblasts
MG63 cells
Osteoblasts
316L stainless steel 316L stainless steel and Co–Cr alloy did not differ in cell growth from the [382]
and Co–Cr alloy negative control cultures, but NiTi and titanium significantly reduced cell
growth. The morphological changes of cells with NiTi and titanium were
also more pronounced
No Nickel induces a significant inhibition of mitosis in human fibroblasts, [387]
whereas no significant effects of this kind were found for titanium or
NiTi. NiTi was considered biocompatible and comparable to titanium
316L stainless steel All metals induced a mild biological reaction. The cytotoxicity of NiTi was [388]
and Co–Cr alloy found to be approximately equal to that of Co–Cr–Mo, both being more
than that of pure titanium, Ti–6A1–4V or 316L stainless steel
No Human plasma fibronectin (pFN), an adhesive protein, can be covalently [389,390]
immobilized onto NiTi substrate and significantly improved human
gingival fibroblast spreading, suggesting that this chemical modification
enables the controlling of metal/cell interactions
No Cells exposed to NiTi are critically affected by the surface preparation. [391]
The hydrogen peroxide surface treatment of NiTi caused a toxic effect
comparable to that of pure nickel. However, the situation changed
tremendously when NiTi was treated by autoclaving in water or steam.
The reaction with these NiTi specimens was clearly non-toxic
316L stainless steel The NiTi alloy showed no cytotoxic, allergic or genotoxic activity. The [392]
findings were similar to those on 316L stainless steel. Conclusion: The
NiTi alloy can be regarded as a biologically safe implant material
CP-titanium and 316L These three alloys induced similar DNA strand breaks of interphase [393]
stainless steel chromatin, but stainless steel induction on metaphase chromatin was
more intense than with NiTi or pure titanium. Conclusion: NiTi
genocompatibility is promising in view of its biocompatibility approval
316L stainless steel, NiTi, stainless steel and b-titanium alloy wires had no effect on the rate [394]
Co–Cr alloy, b-titanium of cell proliferation. The most severe growth inhibition was induced by
alloy wires the CoCrNi alloy
No No cytotoxicity was detected in the direct-contact evolution testing [395]
No The results indicate a good biocompatibility for a nickel content up to [396]
about 50%
Stainless steel, pure In the NiTi and Ti groups, the number of dead cells was significantly [397]
titanium and pure nickel lower than in Ni group. Conclusion: NiTi is well tolerated by the
osteoblastic type ROS-17 cells
Stainless steel 1. The plasma-treated surfaces are cytologically compatible allowing the [383–385]
attachment and proliferation of osteoblasts
2. The sample with surface titanium nitride exhibits the largest degree of
cell proliferation whereas stainless steel fares the worst
No Oxidized NiTi surfaces enhance differentiation of osteoblast-like cells [386]
No The adhesion, spreading, and proliferation of osteoblasts on the [398]
implanted NiTi surface are assessed by cell culture tests. Our results
indicate that the nano-scale surface morphology that is altered by the
implantation frequencies impacts the surface free energy and wettability
of the NiTi surfaces, and in turn affects the osteoblast adhesion behavior
42
Table 40
In vivo studies on biocompatibility of NiTi alloys with animals.
Animal model/maximal duration
of implantation
45 rats subcutaneously/9 weeks
5 dogs, 12 beagles/17 months
N.A./3 months
Femoral shafts of 15 dogs/12 weeks
Frontal bone of 7 rabbits/12 weeks
Long crus of the incus and the incus
of 24 ears of 12 cats/355 days
Paravertebral implantation in
4 rabbits/4 weeks
6 rabbit tibias/12 weeks
Medullary canal of 15 rat tibiae/168
days
12 adult white rabbits/2 years
6 rabbits/3 weeks
Femur/tibia of New Zealand rabbits/15
weeks
Rabbits/No
Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Control materials Major results Refs.
316 stainless steel The tissue reaction was minimal. A dense, relative avascular fibrous [370]
connective tissue capsule formed around implants. Conclusion: NiTi
compares favorably with stainless steel and could be used in deep tissues
316L stainless steel The muscle tissue in dogs exposed to NiTi implants for 17 months [380]
and Co–Cr alloy showed some variability
There was no evidence of either localized or general corrosion on the
surfaces of the bone plates and screws. No signs of adverse tissue
reactions (e.g. bone resorption) resulting from the NiTi implants were
seen. In the NiTi group, some high nickel concentrations were observed
in bone due to contamination
Overall, the gross clinical, radiological, and morphological observations
of tissue at the implantation sites at autopsy revealed no signs of adverse
tissue reactions resulting from the NiTi implants. The study warranted
the conclusion that NiTi had no clearly toxic effects in vivo. The authors
concluded that NiTi alloy is sufficiently compatible with dog tissue to
warrant further investigation of its potential as a biomaterial
No After three months’ implantation, no corrosion was observed on the plate [399]
surfaces. Conclusion: Ti50Ni50xCux (x = 2, 6, 8) shape memory alloys also
have good biocompatibility
316L stainless steel Since the elastic modulus of the NiTi shape memory alloy is lower, the [400]
stress-shielding effect in the bone underneath the NiTi device is less than
316 stainless steel. The axial compression stress of the fracture line is
kept greater and the contact of that NiTi device with the bone was not so
close. This might be beneficial for the recovery of blood supply and bone
remodeling
Hydroxyapatite Porous NiTi [401]
No adjacent macrophage cells were seen for either implant type. Both
materials made bone contact with the surrounding cranial hard tissue,
and the percentage of ingrowth increased with the surgical recovery
time. The bone in contact with the implants was similar in quality to the
surrounding cranial bone. Porous NiTi implants appear to allow for
significant cranial bone ingrowth after as few as 12 weeks. Conclusion:
Porous NiTi appears to be suitable for craniofacial applications
No With the exception of pressure-induced bone erosions, there was no [402]
progressive bone resorption which was prosthesis-induced. The authors
concluded that the biocompatibility of the nickel-titanium alloy stapes
prosthesis with the long crus of the incus was hereby proven
The blood Ni concentration after implantation reached a level twice the [371]
normal in 6–9 h (28  11 vs. 13  5 ppb). After 4 weeks, the Ni
concentration was fourfold in the kidneys (140  43 ppb), twofold in the
liver (40  18 ppb), and 10-fold in urine (90  35 ppb). Concluded that Ni
elution from NiTi alloy should be limited by, for example, using some
coatings
Vitallium, CP-Ti, Duplex The biocompatibility results of the NiTi screws compared with the other [403]
austenitic-ferritic stainless screws showed a slower osteogenesis process characterized by no close
steel (SAF), and 316 stainless contacts between the implant and bone, disorganized migration of
steel osteoblasts around the implant, and a lower activity of osteonectin
synthesis
Pure titanium, anodic While NiTi and the other materials were progressively encapsulated with [404]
oxidized titanium (AO-Ti), bone tissues, Ni was encapsulated with connective tissues and showed
Ti–6Al–4V alloy and pure no bone contact through the 168-day experimental period. No significant
nickel differences between the tissue reactions to Ti, AO-Ti and Ti–6Al–4V, but
NiTi implants showed a significantly lower percentage of bone contact
and bone contact area than any of the other titanium or titanium alloy
materials. In terms of bone contact thickness, there were no significant
differences between NiTi and the other three materials (Ti, AO-Ti and
Ti–6Al–4V)
No The bioactivity and biocompatibility of NiTi alloy are significantly [405]
improved by coating the alloy with HA through chemical treatment
However, the untreated NiTi shows good biocompatibility after long time
implantation
No All six rabbits successfully completed the distraction. A continuity in the [406]
newly formed bone with similar transversal and horizontal dimensions
than the original bone can be observed. Conclusion: The application of a
constant force on distraction osteogenesis, using NiTi, may be a
successful alternative to the conventional gradual distraction
No New bone tissues adhere and grow well on the external surfaces as well [407]
as exposed areas on the inner pores of the NiTi scaffold
No The cell adherence and bone tissue inducing capability are respectively [408]
enhanced over 1.1–1.2 and 9–10 times by sputtering a uniform TiO2 film
on the surfaces of porous NiTi, compared with untreated NiTi
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 43

number of tissues including liver, spleen, brain, and kidney,
showing no metallic contamination of these organs due to the
release from the implants. The authors concluded that NiTi alloy is
sufficiently compatible to warrant further investigation of its
potential as a biomaterial for human use. The majority of the data
summarized in Table 40 suggest that NiTi is quite well accepted
into bone, however there are three conflicting studies
[371,403,404], in which NiTi has been found to have inferior
properties compared to the stainless steel, Co- and Ti-alloys. It is
interesting to note that although hydroxyapatite coatings improve
the bioactivity and biocompatibility of NiTi alloy at the early stage
of implantation, the untreated NiTi shows good biocompatibility
after long term implantation [405].
8.4. In vivo trials of NiTi implants in human [409]
NiTi has also been used as an implant material in humans for
both hard and soft tissue. The overall inflammatory response and
capsule thickness around Nitinol are similar to those observed for
stainless steels and Ti alloys (Tables 40–42). So far, there have been
no reports of tissue necrosis, granulomas, signs of tissue dystrophy
or calcification in vivo. The general immune response to Nitinol has
been found to remain low also during long term implantation. For
muscular tissue, NiTi implants are nearly inert, with porous NiTi
found to exhibit a thin, tightly adherent fibrous capsule with fibers
penetrating into implant pores. In neural and perineural tissue,
NiTi implants are also reported to be non-toxic and non-irritant. In
relation to bone and related connective tissues, Nitinol wire might
be a very promising new tendon suture material, because of high
mechanical strength comparing the conventional materials. The
biocompatibility of NiTi in tendon tissue is excellent. NiTi has no
negative effect on new bone formation and bone contact to NiTi has
been found to be very close, indicating good tissue tolerance.
Osteotomy healing with porous NiTi has also been found to be
good, with normal osteoclastic and osteoblastic activity. After
long term implantation no difference in nickel concentration
between the NiTi and stainless steel groups has been found in any
distant organs. The surface corrosion changes of retrieved NiTi
implants were minimal (refer to Tables 40–42 for references for
this paragraph).
Based on the published studies, NiTi appears to have good
potential for clinical use as its biocompatibility in vivo is very
good. When NiTi is intended to be used in long-term implants,
for example vascular stents, optimal surface treatment must
be considered [409]. However, worldwide medical applications
have been hindered for a long time because of the lack of
knowledge of the biocompatibility of NiTi [102]. A commercially
produced bone anchor (Mitec G21) includes a small piece of NiTi
wire, and this device has recently been approved by the FDA,
although this and other NiTi long-term implant devices remain to
have been proven in long term follow-ups. There are reports that
NiTi material has been successfully used in bone-related human
applications in Russia and China in a large number of patients
[391,400,409–412]. While the studies summarized in Table 41
apparently indicate that the NiTi material, in itself, has no
deleterious effects in human use, the number and quality of
clinical trial studies are such that no clear conclusion can be
made so far.

Table 41
In vivo trials of NiTi implants in human.

Applications/period Number of patients/implant Major outcomes Refs.

duration

Clamps for the fixation of mandibular fractures
Maxillo-facial fractures
Ventral intercorporeal lumbar spondylodesis
Staples were used to lock a tri-cortical iliac bone
graft in cervical anterior fusion
The diseased cervical and lumbal spine were
treated with anterior fusion and porous NiTi
implant grafts
Internal fixation of compression staple for hallux
valgus
Fixation of small bone fragments with NiTi
clamps
77 patients, 93 fractures
using 124 clamps/6 weeks
51 patients/9 months
50 patients/7 weeks
84 patients
36 patients
64 patients
In 72 patients the treatment progressed satisfactorily, while in five [413]
cases infections occurred. After removal of the clamps from
58 patients, there were no pathologic or atypical tissue reactions
or signs of disturbed cell maturation. Conclusion: The application
of NiTi for the surgical treatment of mandibular fractures
facilitates treatment while ensuring stable fixation of the bone
fragments
The surgical treatment of these fractures by NiTi devices ensured a [414,415]
good stability of the fracture surfaces, reduced the time needed for
operative procedures and rehabilitation, and allowed rapid bone
healing
In view of the easier operative technique, the earlier mobilization [416]
of the patients and the good fusion rate, the NiTi spondylodesis
seems to have important advantages over the transplantation of
bone chips
Good and very good clinical results were reported in 80% of the [417]
cases and the average bone fusion rate was fast
Porous NiTi implants can be successfully used, probably because [418]
their mechanical properties are similar to those of the vertebral
bodies, and the material itself shows a high degree of
biocompatibility
The recovery period preceding return to light work averaged [419]
19 days, and normal work and normal walking were resumed an
average of 41 days postoperatively. All the osteotomies united,
and the average angle of hallux valgus and the intermetatarsal
angle improved. The distal fragment during the healing of the
osteotomy was stable. No external fixation by plaster splintage
was needed. The benefits of this internal fixator were that the
period of bone healing was shortened and the patients were
allowed to bear weight earlier than usual
Non-union occurred in 4 patients treated with only one fixative. [420]
Two clamps implanted in non-parallel planes seem to be advisable
to exclude the need for longer immobilization. Neither toxic
manifestation nor episodes of allergic reaction occurred. The study
suggests that by using NiTi clamps in an appropriate way,
satisfactory outcomes could be achieved with respect to both
biological functionality and biocompatibility
44
Table 42
Biocompatibility of bare NiTi wires as stents (filters) and closures (without coating).
Number of patients/implant duration
Intravascular stents
66 endovascular NiTi prostheses implanted
in 36 dogs/14 months
12 intravascular NiTi stents implanted in
the iliac and femoral arteries of 6 normal
dogs/2 years
44 NiTi intraluminal stents implanted in
the iliac arteries of 22 sheep/6 months
14 vascular stents in 14 dogs/9 months
Iliac artery of 6 pigs
NiTi stents implanted in low flow velocity,
vertebral arteries in 6 dogs/9 months
Eleven NiTi and eleven stainless steel stents
implanted in human patients/28 days
22 self-expandable NiTi coil stents implanted
in 16 dogs/1–2 weeks, 1–12 months
27 NiTi wire devices implanted into the venae
cavae of 16 dogs and one sheep/1 week to
4 years
Urethral stents
18 Urethral NiTi stents implanted in 18 dogs/1
week and 1, 3, 6, 12 and 18 months
Urethral stents implanted in 39 patients
(human)/26 months
Closures
Atrial septal defect closure device in 20 adult
dogs/8 weeks
50 patients operated on with NiTi
8.5. Biocompatibility of NiTi wires as stents (filters)
The most exciting clinical application of NiTi is as an alloy for
cardiovascular stents, which provides minimally invasive treatment
Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Major outcomes Refs.
Good and prolonged permeability of the NiTi prostheses [421]
The endovascular prosthesis was wrapped by a thin layer of connective tissue, while inside it
was lined with a layer of endothelial cells
No migration, erosion, inflammation, surface thrombus, or stenosis of the side branches was [422]
seen. Nor were any histopathologic effects detected
Conclusion: The good biocompatibility manifested as a completely endothelialized, thin and
stable neointima, satisfactory delivery and long-term patency at 2 years
All but one stent remained widely patent during the follow-up period. Minimal corrosion was [381,423]
seen at 6 months, and the stent appeared to be biocompatible
Conclusion: A stent can be reliably and safely deployed in the vascular system
No incompletely occluded aneurysms were visible after the implantation of NiTi stents. After [424]
nine months, significantly more abundant intimal fibrocellular tissue growth surrounded the
tantalum filaments than the NiTi filaments, which were covered with a smooth, thin
neointimal layer
Conclusion: NiTi stents may become the treatment of choice for broad-based and fusiform
aneurysms of the internal carotid artery
The early proliferative reaction of smooth muscle cells in the media of the iliac artery [426,427]
following percutaneous transluminal angioplasty [425] was compared with the reaction on
the insertion of NiTi stents. The cell reaction appeared to be more pronounced after PTA than
after the insertion of a self-expanding stent
Five arteries remained patent without significant narrowing. The total mean thickness of the [428]
intima covering the stents showed no significant differences over time. The histologic
findings on the stented vessels showed atrophic compression of the media, but intact
endothelial cell linings without necrosis or perforation were observed
Conclusion: no significant risk of thromboembolic events exists after the implantation of NiTi
stents in the vertebral arteries in dogs
At 3 days, the stainless steel stents had more inflammatory cells adjacent to the stent wires [429–432]
than their NiTi counterparts. After 28 days, the vessel response was similar for the NiTi and
stainless steel designs. The mean neointimal area and the percentage of stenosis were
significantly lower in the NiTi than in the stainless steel group
Conclusion: A NiTi stent exerts a more favorable effect on vascular remodeling with less
neointimal formation, than a balloon-expandable design. Progressive intrinsic stent
expansion after the implantation does not appear to stimulate neointimal formation and may
therefore prevent in-stent restenosis
Angiographic artery dimensions measured immediately after stent implantation did not [433]
differ from those noted at follow-up. Outward stent pressure compressing the internal elastic
membrane and the media in most cases. Intimal hyperplasia started at 2 weeks and was most
apparent at 3 and 6 months
Conclusion: The NiTi self-expandable stent provokes a moderate cellular proliferative
response that reaches its maximum in 3–6 months without further progression
All cleaned NiTi wire filters remained patent, but some showed venographic filling defects [434]
caused by adherent organized thrombi. The filters in larger veins tended to have less
thrombus formation. Surface polishing and filter shape had no observable effect on
thrombogenicity. Patchy chronic inflammation on the surface of uncleaned filters, but only a
benign fibrous tissue reaction on cleaned filters. Neointimal tissue overgrowth was observed
in the contact area of the vena cava. Platelet adhesion and plasma coagulation effects of NiTi
wire were tested in vitro in human blood and found to be similar to those of stainless steel
Conclusion: NiTi may be a promising material for human intravascular prosthetic applications
Conclusion: Despite the excellent biocompatibility of the material with no evidence of foreign [435]
body reactions or corrosion, there were no complete incorporations of the stent by
epithelialization. Clinical application therefore appears to be problematic
39 patients with benign prostatic hyperplasia had NiTi urethral stents implanted with a [436,437]
clinical success rate of 89%. Follow-up for 26 months showed no incrustation or migration of
the spiral
Percutaneous transcatheter closures were attempted using the new device. The closures [438]
were successful in 19 studies and unsuccessful in one. At 8 weeks in 3 dogs showed the
devices to be covered by smooth endocardium enmeshed in mature collagen tissue, with
minimal mononuclear cell infiltration
Conclusion: This new device permits effective and safe atrial septal defect closure in a canine
model
No problems related to early migration and expulsion were observed, and no anastomotic [439]
leakage and bleeding occurred
Conclusions: Intestinal anastomosis with the NiTi was safe and feasible without anastomotic
leakage and reoperation compared with the stapling technique
instead of major surgery. Since the first experiments by Cragg and
co-workers [423], several studies have provided further information
on the biocompatibility of NiTi as vascular stent materials. Based on
the studies summarized in Table 42, the histopathological changes
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 45

Table 43
Biocompatibility of coated versus bare NiTi stents.

Number of patients/implant duration Major outcomes Refs.

Intravascular stents
14 patients Perivascular inflammation in 79% of patients with polyester-covered NiTi stents. Clinical [440]
symptoms were seen in 57% of these patients. No reaction was evident among the controls
with uncovered NiTi stents and the subjects who underwent peripheral percutaneous
transluminal angioplasty. Conclusion: The polyester-covered NiTi stent may induce
systemic and severe local reactions
10 PTFE (Dacron)-coated NiTi stents implanted Clinical signs of acute inflammation manifested as fever and local tenderness. No [441]
in10 patients/1 month inflammation was found in the control groups with bare NiTi stents
Heparin-coated Dacron-covered NiTi stent-grafts Severe inflammatory perigraft responses, marked vascular wall thickening and adhesions [442]
in 4 sheep/6 months around the Dacron fabric, and a pronounced inflammatory foreign-body response. There
was almost no response to bare NiTi stents. Conclusion: The use of noncovered stents
should thus be preferred to the use of Dacron-covered stent-grafts
6 polyurethane-coated and 6 bare NiTi stents At 4 weeks, all stent struts were endothelialized. Mild proliferative responses with some [443]
implanted in rabbit carotid arteries/4 weeks neovascularization around both stent types were seen. No differences in the degree of
neointimal proliferation between the stents were found, but the polyurethane coating
was associated with an inflammatory tissue response consisting of lymphocytic
infiltration and foreign-body reaction and the appearance of multinucleated giant cells.
Conclusion: A low biocompatibility of polyurethane, which may thus not be an ideal
material for coating intravascular devices
PTFE, polytetrafluoroethylene.

caused by bare NiTi wires are associated with a mild inflammatory
response, some atrophy of vessel media, acceptable fibrocellular
tissue growth and endothelialization. The biocompatibility of NiTi
stents seems to be equal or better compared to stainless steel
stents.
Urethral stents have also been used for the treatment of
urethral strictures, and outcomes are controversial. While one
study has shown NiTi urethral stents to be problematic in terms of
compatibility [435], another reported that in a sample of
39 patients, a clinical success rate of 89% was achieved and
follow-up for 26 months showed no incrustation or migration of
the spiral [436,437].
To minimize ion release from NiTi implants, especially in stents
that have high surface area and are in a dynamic flow, polymer
coatings have been applied. However, severe inflammation
associated with polyester- or polyurethane-coating has been
consistently reported (Table 43). Nonspecific inflammatory reac-
tions characterized by local discomfort and flu-like symptoms have
been seen in patients implanted with PTFE (Dacron)- or polyure-
thane-coated NiTi stents.
8.6. Current and potential applications of NiTi
Since the early 1970s, NiTi wires have found a variety of
applications [444–446], including orthodontic archwires [447],
vascular stents [380,435], and orthopedic devices for closure [438]
and fixation [448]. Among these applications, the most successful
ones have been self-expandable stents in gastroenterology and
cardiovascular applications (Fig. 30). The idea of using NiTi stents
was first reported in 1983 independently by the research teams of
Cragg and co-workers [423], and Dotter and co-worker [449]. Using
stents, major surgical operations can be avoided. For critically ill
patients, a stent may be the only choice [102].
8.6.1. Self-expandable stents
pelvis or the lower limbs. SNF was approved by the FDA in
1990. Self-expandable NiTi-based stents have since become the
general trend in stent production, whereby a thin stent is placed in
the narrowed artery to then expand and dilate it. Today, there are a
number of self-expandable stents/filters approved by USFDA,
commercially available in medical markets and applied widely in
clinics (Figs. 31 and 32).
8.6.1.2. Gastrointestinal applications. Self-expanding stents have
been applied for esophageal strictures and the palliation of
malignomas by a number of surgeons (Fig. 33) [426,451,452]. Esoph-
ageal NiTi stents are routinely implantable, providing effective
palliation of malignant esophageal obstructions, and have a low risk
of severe complications. The only disadvantage has been that
incomplete initial stent expansion as well as tumor ingrowth/
overgrowth occurred in nearly one third of the patients, in several
studies [426,451,452]. Biliary stents (Fig. 31g) are also effective in
achieving long-term palliation in patients with malignant obstruc-
tive jaundice, and in 1999, the FDA released a stent for this purpose.
For benign biliary strictures, metallic stent placement is associated
with a low long-term patency rate [453–457].
8.6.1.3. Urological applications. The use of NiTi stents has also been
expanded to treat prostatic obstruction by Lopatkin and co-
workers [458,459]. For patients with a high operative risk the
insertion of a permanent metal stent system offers a useful
alternative to treat subvesical obstruction caused by prostatic
carcinoma [460–462], and benign prostatic hyperplasia [461,
463]. The use of urethral stents was found to considerably
decrease the number of repeated dilatations and urethrotomies in
recurrent urethral strictures [464,465]. Despite the good biocom-
patibility of NiTi in a long-term canine study, there were no
complete incorporations of the stent by epithelialization, and so
there is a concern that clinical application might be problematic
[435].

8.6.1.1. Cardiovascular applications. The first vascular NiTi device
was the Simon Nitinol Filter (SNF) (Fig. 31a) used to treat
pulmonary embolism [450]. The filter is inserted as a straight thin
wire via the small bore catheter used for angiographic diagnosis.
Upon reaching the lumen of the inferior vena cava and sensing
body temperature, it reverts to its preset complex filter shape and
locks into place permanently, trapping any further emboli from the
8.6.1.4. Other applications. The use of self-expanding NiTi stents
was first reported by Rauber and co-workers to prevent major
airway occlusion [466]. In those early tests, stenting was shown to
be very effective in inoperable tracheal or bronchial stenoses, due
to intraluminal tumor invasion [467,468]. A NiTi-based mesh-
expanding prosthesis for laparoscopic hernioplasty significantly
shortened the operating time in a study by Himpens [469]. The
46 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Fig. 31. Various Nitinol (NiTi) stents/filters. (a) Simon Nitinol Filter (SNF), FDA approved in 1990. (b) G2 (Bard Peripheral Vascular), FDA approved for permanent use in
2005 and for retrievable use in 2008. (c) OptEase, FDA approved for permanent use in 2002 and retrievable use in 2004. (d) SafeFlo, FDA approved for permanent use only in
2009. (e) Option. (f) Vena Tech LP, FDA approved in 2001. (g) Biliary stent, FDA approved in 1999. http://www.whichmedicaldevice.com/editorial/article/104/a-brief-history-
of-inferior-vena-cava-filters-and-analysis-of-current-devices.

good holding and atraumatic characteristics of detachable clamps
for gastrointestinal tract surgery have been demonstrated using
laparoscopic and thoracoscopic follow up [470]. Another type of
device is the NiTi stapes prosthesis, used in ossicular fixation after
stapedectomy, introduced in the later 1990s during otoplasty
[402,471].
8.6.2. Orthopedic and orthodontic applications
Since the 1960s, studies have been carried out into the
application of NiTi for orthopedic procedures, including correction

Fig. 32. Radiograph of the inferior vena cava implanted with Vena Tech LP (yellow
arrow) and OptEase (blue arrow). http://www.ceessentials.net/article12.html. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of the article.) Fig. 33. Esophageal stent. http://www.rush.edu/rumc/page-1175113035472.html.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 47

Table 44
Orthopedic and orthodontic applications of NiTi.

Animal models Major outcomes Refs.

Correction rods of scoliosis

The functional principle of the NiTi memory
wire was demonstrated in an experiment
carried out on a plastic model
26 patients
1 scoliosis monkey/4 weeks
6 goats with experimental scoliosis were
implanted with 6-mm nitinol rods/
6 pigs/3 and 6 months
38 scoliosis patients (ranging from 508 to
1208; 22 cases over 708) who underwent
NiTi-assisted correction/4 months
NiTi wire prestretched by 7% is led through eyelets on the convex side and fixed at the ends. On [482]
being heated, the wire shortens, righting the model so that it assumes a straight shape
Correction was reported to be good and there were no complications [473]
The blood Ni concentration after implantation of the NiTi alloy rod reached a level twice normal [371]
after 6–9 h. After 4 weeks, the Ni concentration increased fourfold in the kidneys, twofold in the
liver, and 10-fold in the urine. Scoliosis with a Cobb angle of 438 was completely corrected
The rods were transformed, and the scoliosis corrected, in the awakened goats by 450-kHz radio [472]
frequency induction heating. The curves averaged 418 before instrumentation, 338 after
instrumentation, and 118 after rod transformation. The animals tolerated the heating without
discomfort, neurologic injury, or evidence of thermal injury to the tissues or the spinal cord. The use
of shape memory alloys allows continuous true rotational correction by rod torsion
The induced curve of about 40degrees Cobb angle remained constant during the follow-up [483]
The postoperative serum nickel measurements were around the detection limit, and were not
significantly higher compared to the preoperative nickel concentration
The device was almost overgrown with newly formed bone. Corrosion and fretting processes were
not observed. No evidence of a foreign body response in lung, liver, spleen and kidney
The major Cobb angle improved from an average 78.48 preoperatively to 24.38 postoperatively (total [474]
percent correction 71.4%). In 16 patients with a major curve <708 and flexibility of 52.7%, the
deformity improved from 58.48 preoperatively to 12.38 postoperatively (percent correction, 78.9%).
In 22 patients with a major curve >708 and flexibility of 25.6%, the deformity improved from 94.18
preoperatively to 30.18 postoperatively (percent correction, 68.1%). Only one case had a deep
infection. There were no neurologic, vascular or correction-related complications such as screw
pullout or metal fracture. The study shows that the intraoperative use of a NiTi rod is a safe and
effective method to correct scoliosis

Compressive staples, clamps or clips

133 NiTi clips were applied in No procedural complication or adverse reaction to the clip was noted. There was no movement at [425,448]
119 patients/2–8 years the operated level in dynamic lateral view X-ray of cervical spine at the 1st postoperative day as
well as on follow-up. Graft extrusion was seen in one patient on the 2nd day after surgery and was
re-operated. Bony fusion occurred in all patients after 9–12 months of surgery. There was no
incidence of breakage or dislodgement of the clip from the site where it was inserted
Conclusion: NiTi clips are a simple alternative for cervical spine stabilization after discoidectomy.
Their insertion is simple, minimally invasive, does not require any special set of instruments and
they are much more economical than other established methods of treatment. These clips are
accepted well by human tissue and do not interfere with MRI

rods for scoliosis and fixation staples for long bone (Table 44). Early
trials indicated that the scoliosis-correction system based on NiTi
shape memory had quite complicated biomechanical problems
related to compression and distraction control. NiTi may not
provide any improvements compared to the traditional implant
systems [371,472]. However, a follow-up study from China in 1986
[473] reported satisfactory performance of NiTi rods in 26 scoliosis
patients. Successful applications of NiTi in correction of scoliosis in
38 patients were reported in another more recent follow up report
in 2011 [474].
NiTi compression staples were first introduced in China.
According to Dai [448] a shape memory staple was first used
inside the human body in 1981. After that, NiTi staples and clamps
were used in comminute fractures of short tubular bone [400], for
fixation of mandibular fractures [413], metatarsal osteotomies
[419], anterior cervical decompression and fusion [417,418,475],
fixation of small bone fragments [420], and for several other
cursory applications [412,476,477]. The only NiTi-containing
orthopedic implant widely used in western countries is the Mitek
G2 suture anchor (Fig. 34) [478]. It has superelastic NiTi wings

Fig. 34. Non-screw anchors Mitek G2, Mitek Rotator Cuff Anchor, and Linvatec Ultrafix RC from left to right, all containing NiTi [478].
48 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57

Table 45
Mechanical properties of NiTi alloys in comparison with the 316L stainless steel and cortical bone [44,444].

Alloys Young’s 0.2 yield Maximal elastic UTS Elongation Fracture toughness
modulus (GPa) stress (MPa) strain (%) at rupture (MPa m1/2)

316L 200 200–700 0.20 500–850 10–40 100

NiTi (austenite) 50–80 200–700 10 900–1355 14.3 30–60
NiTi (martensite) 30–40 70–140 10 NA NA NA
Cortical bone 7–30 NA NA 50–150 NA 2–12

which prevent the anchor from pulling out of the bone after
insertion and securing of tendons or ligaments back onto the bone
[479,480]. Another potential application of NiTi is a hook used to
restore the dislocated acromio-clavicular joint of the shoulder
[481].
Most published studies in the orthopedic field have not satisfied
the basic quality criteria of scientific study. The data are not
convincing enough to say that a certain NiTi implant can be used
without harm. To be considered fully successful, it must be proved
to be better than the existing competitors. At present, there are no
comparative clinical studies and the sample groups have generally
been small. Randomized prospective studies are needed to apply
new NiTi implant devices for long-term clinical use in humans
[102].
8.7. Mechanical properties of NiTi alloy [43]
8.7.1. General mechanical properties
The mechanical properties of NiTi alloy are unique. Young’s
modulus of NiTi alloy is in the range of 30–50 GPa [444], the closest
to that of cortical bone of human, even closer than b Ti alloys and
Mg–Zn alloys (Table 9). The relatively low level of Young’s modulus
in NiTi alloys does not affect their yield strength and UTS, which are
comparable with those of stainless steels (Table 45). The most
attractive mechanical property of NiTi is its large elastic strain of
10%, which is one order higher than any traditional alloys.
8.7.2. Fatigue properties of NiTi alloys
The mechanical properties, especially fatigue properties, of NiTi
alloy sensitively depend on the deformation temperatures,
whether in austenite or martensite phase. The fatigue strength
of austenitic NiTi alloys is in general higher than that of martensitic
alloys (Fig. 35a), and the fatigue strength of austenitic NiTi alloys is
higher at elevated temperatures (Fig. 35b). At both room and body
temperature, the fatigue strength of austenitic NiTi alloys is
typically around 400 MPa (Fig. 35).
The majority of Nitinol fatigue investigations in the literature
have been conducted under constant deformation (strain) condi-
tions (Fig. 36). Even though a fatigue life of 1000 cycles is obtained
for a cyclic strain amplitude of 10%, conventional engineering
materials sustain a strain amplitude of only 1% or below before
fracture, under which a fatigue life is typically in the range of
105–107. The data of Fig. 36 indicate that larger total strains for
given lifetime are achieved when the deformation temperature is
between As and Af, compared with deformation of thermally
formed martensite for these fully annealed materials (i.e. thermal
martensite).
A comparison of fatigue properties of NiTi alloys with those of
stainless steel, Co- and Ti-based alloys (Table 46) reveals that
Nitinol alloys are not advantageous over other alloys in terms of
fatigue properties, which explains why these alloys are not the
choice of joint replacements.
8.8. Issues and challenges of NiTi implants
Despite the most exciting and successful applications of NiTi as
stents in the treatment of the occlusion of various vessels and ducts
in the body, there are concerns over the systemic toxicity associated
with the release of pure Ni over a long-term implantation period,
especially as a bone implant. New nickel-free shape memory alloys
are under development [486,487].
8.9. Summary
The shape memory property of Nitinol has put it in a special
place for minimally invasive surgical applications. Today, self-
expandable stents, clamps and clips made out of NiTi alloys have
been applied in various types of surgery. Despite extensive

Fig. 35. S–N curves of NiTi alloys. (a) Fully reversed (R = 1) stress based fatigue curves to 107 cycles. The fatigue limit for the austenite (Af = 10 8C) is significantly higher
than the martensite (Af = 110 8C) phase [484]. (b) The tension–tension stress based fatigue curves up to 106 cycles, with an Af of 27 8C. The 106 cycles fatigue limit at 30 8C is
lower than that obtained at 60 or 160 8C [485]. Note: Af is the finish temperature of phase transformation in the alloy.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 49

Fig. 36. (a) ‘Plastic’ strain amplitude per cycle for four Nitinol compositions tested under constant strain tension–compression conditions at room temperature. In general,
these data show that fatigue life tends to increase with increasing Af temperature for annealed Nitinol. (b) Effect of cyclic strain amplitude on the rotary bending fatigue life for
NiTi wires at three test temperatures for an Af of 40 8C. The low cycle fatigue behavior is greatly affected by the test temperature with an order of magnitude improvement in
fatigue at 20 8C compared with that at 80 8C. However, with this limited dataset, there does not appear to be a difference at the high cycle regime. (c) The thermal martensite
data have greater low cycle and high cycle lives compared with the three sets of data for superelastic conditions [43].

investigation on the biocompatibility of NiTi alloys, reports have
been controversial, especially those on orthopedic implant trials.
So far, many researchers believe that NiTi alloys have excellent
corrosion resistance and biocompatibility, while on the other hand
there are serious concerns over the long-term systemic toxicity of
nickel ion release. The development of new nickel-free shape
memory alloys may offer new opportunity to obviate these
concerns.
9. Tantalum
Tantalum is one of the refractory metals, among others
including niobium, molybdenum, tungsten, and rhenium. Except
for two of the platinum-group metals (osmium and iridium), the
refractory metals have the highest melting temperatures
(>2000 8C) and the lowest vapor pressures of all. The use of
niobium, molybdenum, and tungsten for biomedical applications
is confined to alloying in stainless steels, cobalt alloys, and
titanium alloys. Radioactive rhenium is occasionally used in stents
to prevent re-stenosis [488,489]. Tantalum has found a number of
clinical applications, due to its excellent biocompatibility, flexibil-
ity and corrosion resistance.
9.1. Corrosion and biocompatibility of tantalum
Pure tantalum has excellent resistance to corrosion in a large
number of acids, most aqueous solutions of salts, organic
chemicals and in various combinations and mixtures of these
agents. The corrosion resistance of tantalum is approximately
the same as that of glass. Tantalum has no known biological role
[490], and is non-toxic. Compounds containing tantalum are
rarely encountered in the natural environment. Tantalum is
among the most biocompatible of metals used for implantable
devices [44]. There is however some evidence linking tantalum to
local sarcomas [491] and toxicity of its oxide to alveolar cells
[492].

Table 46
Fatigue strength of NiTi alloys in comparison with stainless steel, cobalt and titanium based alloys.

Alloys Fatigue strength [160] in the air at 107 cycles Fatigue strength [160] in Refs.
solution at 107 cycles

NiTi alloys 100–400 (strain control) NA

Forged 316L 300–350 (unnotched stress control) 100–200 (stress control) [228]
Forged CoCrMo alloys 600–900 (unnotched stress control) 200–300 (stress control) [228]
Ti alloys 500–600 (unnotched stress control) 400–600 (stress control) [228]
50 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
Table 47
Compositions, maximum weight percent allowed for unalloyed tantalum in accordance with ASTM F560.
Processing condition Hardness (HV) Young’s modulus (GPa)
a
Annealed 80–110 186–191 (27–28)
Cold-workedb 120–300 186–191 (27–28)
a
Electron beam or vacuum arc cast tantalum.
b
Sintered tantalum.
Table 48
Mechanical properties for unalloyed tantalum (ASTM F560).
Processing condition Hardness (HV) Young’s modulus (GPa)
Annealed 80–110 186–191 (27–28)
Cold-worked 120–300 186–191 (27–28)
9.2. Clinical applications of tantalum
The chemical requirements for medical-grade tantalum are
given in Table 47. Requirements on the mechanical properties of
annealed and cold-worked unalloyed tantalum are listed in
Table 48. Since the 1950s, tantalum has been used in surgery as
suture wires for skin closure, tendon, and nerve repair [493–495];
foils and sheets for nerve anastomoses [496]; clips for the ligation
of vessels [497]; staples for abdominal surgery [498,499]; and as
pliable sheets and plates for cranioplasty and reconstructive
surgery [500,501]. Sintered tantalum capacitor electrodes are also
used in electrical stimulation devices [502].
Tantalum has also been used to coat other metals, such as
titanium implants [503], and carbon foam skeletons used as a
biocompatible replacement for vertebral bodies of the spinal
column [504]. Tantalum coatings, which are 70–80% porous,
have a macroporous structure similar to that of cancellous bone
[503–505]. Besides spinal implants, carbon–tantalum cellular
materials have potential applications for hip and knee construction
and bone scaffold void filling applications [504]. Porous scaffolds
have also been made from tantalum [506,507], including Trabecular
Metal [508,509], which contains pores, the size of which makes this
material very good for bone in-growth [508–511]. It is believed that
Trabecular Metal has an elastic nature which aids bone remodeling
[506–509].
10. Zirconium alloys
10.1. Corrosion of zirconium
Zirconium is mainly used as an alloying element for its strong
resistance to corrosion. Zirconium is similar to titanium in that it is
refractory and has a high affinity for oxygen. When zirconium is
exposed to an oxygen-containing environment, a protective oxide
film spontaneously forms on its surface, in both dry and wet
conditions. Moreover, this film is self-healing and protects the base
metal from chemical attack at temperatures up to 300 8C. As a
result, zirconium is very resistant to corrosive attack in most
mineral and organic acids, strong alkalis and saline solutions.
10.2. Biocompatibility of zirconium
Zirconium, which exists in the body only 1 mg on average, does
not play natural biological role in humans. The daily intake of
zirconium is approximately 50 mg. Short-term exposure to
zirconium powder can cause irritation, and inhalation of zirconium
compounds can cause skin and lung granulomas. Persistent
Yield strength (MPa) UTS (MPa) Elongation (%)
140  20 205  30 20–30
345  50 480  70 1–25
Yield strength (MPa) UTS (MPa) Elongation (%)
140  20 205  30 20–30
345  50 480  70 1–25
exposure to zirconium tetrachloride results in increased mortality
in rats and guinea pigs and a decrease of blood hemoglobin and red
blood cell in dogs [512–514]. Nonetheless, zirconium metal
exhibits the highest biocompatibility of all metals in the body
(Fig. 4), and zirconium compounds are of low toxicity.
10.3. Clinical application of zirconium alloy
A zirconium alloy, Zr–2.5Nb, is used in a new ceramic knee
implant (Fig. 37) [515–518]. The Zr–2.5Nb alloy has a relatively
low modulus of 100 GPa. The zirconium alloy is combined with an
all-plastic tibial component, replacing the metal tray and plastic
insert used in other knee replacements. The zirconium alloy is
strengthened with small additions of oxygen and coated with a
hard ceramic surface. The ceramic coating is developed through
heating at 500 8C, at which the zirconium reacts with oxygen to
produce zirconium oxide, or zirconia. The ceramic zone extends
approximately 5 mm deep. Underneath the ceramic, the material
gradually transitions from the ceramic into zirconium metal
through several more micrometers. It is believed that this new
knee could last for 20–25 years [515,516], substantially more than
the 15– 20 years over which cobalt chromium alloy and
polyethylene implants last [519]. The new combination can also
be lubricated, which results in a smoother and easier articulation
through the plastic. Another important characteristic of this
material is that it is systemically more biocompatible, meaning
that sufferers of nickel allergies who may not tolerate knee
implants made of cobalt chromium alloy, may well tolerate a
zirconium alloy alternative [520].
Fig. 37. OXINIUM oxidized zirconium has been introduced to reduce the wear rate
over the CoCrMo alloy total knee implants. http://global.smith-nephew.com/us/
patients/OXINIUM.htm.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57 51

11. Silver Table 49

Metallic materials used in dentistry [543].

11.1. Biocompatibility of silver Alloys Compositions (wt%)

Cu–Zn alloys 65–88Cu, 12–35Zn

Silver has no known biological roles, and possible health effects (no longer used)
of silver are a disputed subject [521–523]. Although silver itself is Nickel silver 47–65Cu, 10–25Ni, 15–42Zn, 0–1Pb
non-toxic, most silver salts are, and some may be carcinogenic (no longer used)
HgAgSnCu amalgam 50Hg, 22–32Ag, 14Sn, 8Cu
[524]. Silver ions can bind to sulphur groups in inter-molecular
Au-alloys 58–75Au, 7–18Cu, 10–26Ag, 1–10Pd,
bonds in some biomolecules [525]. 5–25Pt, 0–19lr, 1–2Ni
Stainless steels 45–84Fe, 8–30Cr, 8–25Ni, 0.1–0.2C
11.2. Medical application of Ag CoCrNiMo alloys 40Co, 20Cr, 16Fe, 15Ni, 7Mo, 2Mn, 0.15C, 0.04Be
CP-Ti 99Ti, <0.10C, <0.50Fe, <0.06H, <0.40N, <0.40O
(a + b)-Ti alloys 88–91Ti, 5–7Al, 3–5V
Silver ions and silver compounds show a toxic effect on some b-Ti alloys 76–82Ti, 10–12Mo, 5–7Zr, 3–5Sn
bacteria, viruses, algae and fungi, but without causing the high NiTi-martensitic 44–52Ni, 45–51Ti, 5–6Cu, 0.2–0.5Cr, 0–3Co
toxicity to humans. Its germicidal effects can kill many microbial NiTi-austenitic 44–52Ni, 45–51Ti, 5–6Cu, 0.2–0.5Cr, 0–3Co
organisms [526,527]. Because of this, silver salts and the metal Ti–Nb 50–60Ti, 40–50Nb
Ti–Ta–Nb/Zr/Sn 50–60Ti, 25–35Ta, 5–25Nb
itself have played an important part in the development of
Ti–Zr Ti–10Zr–5Nb–5Ta
medicine. Interest in the use of silver is being rediscovered in a Ni–Ta 50–70Ni, 30–50Ta
wide variety of biomedical applications, where it can prevent the
growth of microorganisms responsible for disease [528,529]. The
antimicrobial properties of silver are used in the form of silver salts
and nano-scale complexes that break down to release silver ions applied as corrective archwires, and HgAgSn amalgam is used as a
(Ag+). restorative material. HgAgSnCu (50% Hg, 22–32% Ag, 14% Sn, and
Metallic silver is also used in a number of surgical applications, 8% Cu) amalgam has been used in dentistry as a restorative
both for structural devices (e.g. cranial support plates, suture wire, material for a number of reasons. First, its melting point is close to
aneurysm clips, and tracheostomy tubes) and for prostheses (e.g. body temperature, and thus it is relatively easy to use and
silicon-silver penile implants) [530,531]. manipulate during placement. It remains soft for a short time, so it
can be packed to fill any irregular volume, and then forms a hard
12. Metals used as medical electrodes compound. Second, it has satisfactory longevity. Third, it is
inexpensive. Nonetheless, HgAgSnCu has been increasingly
An important and challenging medical use of implanted replaced by white tooth fillings and composite resins, primarily
electrodes is in prosthetic devices for neural [532–534] or muscle for cosmetic reasons, as well as the toxicity of Hg ions.
stimulation. These devices employ metal electrodes to transmit the The use of Ti–Ta, Ti–Nb, Ti–Zr and Ni–Ta alloys in dentistry is
current required for electrical stimulation of appropriate areas of primarily driven by their high wear resistance. Moreover, for the
the nervous system. Neural prostheses for direct control of diffusion bonding of Ti alloys to alumina, materials with a similar
peripheral organs include the cardiac pacemaker, the phrenic thermal expansion coefficient to alumina are required. Materials are
stimulator for respiratory control and spinal cord stimulators for developed by alloying the biocompatible elements Ta, Nb and Zr to
bladder control [535–539]. More complex neural control devices titanium, including Ti–(10–80)%Ta, Ti–(40–50)%Nb, Ti–(10–50)%Zr.
include auditory prostheses for deafness, experimental visual There is no significant difference in the corrosion behavior of these
prostheses for blindness and neuromuscular prostheses for alloys in 0.9 wt% saline solution. But the Ti–Ta alloys are superior in
muscular entrainment to restore limb function in paralyzed their mechanical properties to Ti–Nb alloys [544,545]. Ti–Zr alloy
individuals. has better biocompatibility than Ti alloy [546,547]. The general
The most frequently considered metals for electrical stimula- mechanical properties of these metallic materials are listed in
tion are the so-called noble or precious metals: platinum, iridium, Table 50 for the purpose of comparison.
rhodium, gold, and palladium. This is because of their resistance to
chemical and electrochemical corrosion. However, all of these
metals show corrosion effects during both in vitro and in vivo
electrical stimulation. Corrosion effects include weight loss, Table 50
formation of unstable surface films that tend to spall from the Mechanical properties of dental metallic implant materials [543].
surface, and dissolution of metal [540]. Of the noble metals, Alloys Young’s Yield UTS (MPa)
platinum and platinum-iridium alloys containing 10–30% Ir are the modulus strength
most widely used for electrical stimulation. Metal oxides such as (MPa) (MPa)
iridium oxide have also shown promise. Some non-noble metals Cu–Zn alloys 100–120 70–460 260–900
are candidates for electrode applications requiring high mechani- Nickel silver 120 140–540 390–640
cal strength and fatigue resistance such as demanded by HgAgSnCu Varying Varying 150–160
with time with time (compressive)
intramuscular electrodes. These include vacuum melted type
Au-alloys 85–110 170–570 320–1120
316L stainless steel, Cobalt alloys Elgiloy and MP35N, pure forms of Stainless steel 180–220 790–2450 930–2860
zirconium, tungsten, tantalum, and titanium, and tungsten CoCrNiMo alloys 180–230 960–2140 1210–2540
bronzes made by powder metallurgy processing [540]. CP-Ti 100–110 170–1000 240–1100
(a + b)-Ti 100–120 740–1130 860–1220
NiTi-martensite 28–44 70–1240 900–1930
13. Metallic materials used in orthodontic implants
NiTi-austenite 80–110 180–690 800–1670
b-Ti 65–70 520–1380 690–1500
Metallic biomaterials have been applied in dentistry and dental Ti–Nb 65–93 760–930 900–1030
surgery (Table 49) for dental restoration, endodontic implantations Ti–Ta 65–110 300–600 500–700
Ti–Zr 45–60 NA 650–1000
and orthodontics [541–543]. Cobalt-based alloys and Ti–6Al–4V
Ni–Ta 173–220 65–500 300–850
are used to replace tooth roots, stainless steels and NiTi alloys are
52 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
14. Summary and remarks
Although imperfect, stainless steels, cobalt- and titanium-based
implant materials remain dominant in orthopedic surgery.
Although these alloys are much stronger and tougher than bone
in the laboratory, their service longevity is limited to 20–25 years,
much shorter than the lifespan of humans (except elderly
patients), and the incidence of failure after 15 years service is
unsatisfactorily high. Until now, major attempts to reduce stress-
shielding effects have involved reducing the Young’s modulus of
bulk materials (e.g. the development of b Ti alloys), which
inevitably compromises the fatigue resistance. While it is possible
to achieve metallic materials that would have Young’s modulus of
bone while maintaining a fatigue resistance of longer than
20 years, this remains unlikely. This is because biological tissues,
such as bone, have the ability to self-heal, remodel and recover
their mechanical properties in vivo while subjected to cyclic
wearing, whereas synthetic materials do not. Hence, any bulk
material that has a Young’s modulus similar to that of bone would
be at a high risk of premature failure. Hence we believe that the
most promising strategy to achieve a synthetic permanent
substitute that has a Young’s modulus of bone would be porous
metallic implant. Such porous networks could be tailored to
mechanically match the host bone, and the porous structure can
also encourage host bone to grow into it.
Another hope that may enable long-term success of metallic
implants is from the development of degradable metallic
materials, such as magnesium alloys and metal ceramics. Using
this strategy, researchers must take great caution in the choice of
metal to prevent metal-related toxicity, because biological safety
levels of most metal elements are lower than 0.01%. The corrosion
rate of a degradable implant must also be lower than the tolerable
level of each alloying element in the body. In this regard, careful
design must be conducted based on the biological profile of each
alloying element (their safe level in the body, excreting mechanism
and rate, etc.) and the corrosion profile of the implant in an
appropriately simulated physiological environment, combined
with appropriate non-human implant models.
Acknowledgement
The authors would like to thank Ms. Xuenyuan Yang for
providing Fig. 19 and her assistance in preparing Table 34.
References
[1] J.B. Park, R.S. Lakes, Biomaterials: An Introduction, Springer, New York, 2007.
[2] A.D. Mazzocca, J.P. DeAngelis, A.E. Caputo, B.D. Browner, J.W. Mast, M.W.
Mendes, Principles of internal fixation, in: B.D. Browner (Ed.), Skeletal Trauma,
W.B. Saunders Company, Philadelphia, 2008.
[3] F. Rupp, J. GeisGerstorfer, K.E. Geckeler, Adv. Mater. 8 (1996) 254.
[4] V. Biehl, T. Wack, S. Winter, U.T. Seyfert, J. Breme, Biomol. Eng. 19 (2002) 97–101.
[5] B. Zberg, P.J. Uggowitzer, J.F. Loeffler, Nat. Mater. 8 (2009) 887–891.
[6] D.J. Blackwood, Corros. Rev. 21 (2003) 97–124.
[7] h.w.a.f.g.s.c.c.c.r.c.i. PM], in: vol 2012, Food and Drug Administration, 2011,
http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/Devi-
ceApprovalsandClearances/default.htm (accessed on 29 Oct 2014).
[8] Y. Okazaki, E. Goth, Corros. Sci. 50 (2008) 3429–3438.
[9] M. Geetha, A.K. Singh, R. Asokamani, A.K. Gogia, Prog. Mater. Sci. 54 (2009)
397–425.
[10] T.W. Duerig, K.N. Melton, D. Stockel, C.M. Wayman, Engineering Aspects of Shape
Memory Alloy, Butterworth-Heinemann, London, 1990.
[11] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Biomaterials 27 (2006) 1728–1734.
[12] B. Heublein, R. Rohde, V. Kaese, M. Niemeyer, W. Hartung, A. Haverich, Heart 89
(2003) 651–656.
[13] J.Y. Wong, J.D. Bronzino, Biomaterials, CRC Press, New York, 2007.
[14] Handbook of Biomaterials Evaluation: Scientific, Technical and Clinical Testing
of Implant Materials, Taylor & Francis, London, UK, 1999p. 719.
[15] D.F. Williams, Definitions in Biomaterials, Elsevier, Amsterdam, 1987, pp.
49–59.
[16] D.F. Williams, General concepts of biocompatibilitty, in: J. Black, G.W. Hastings
(Eds.), Handbook of Biomaterials Properties. Part III, Chapman and Hall, New
York, 1998, pp. 482–489.
[17] D. Williams, Med. Device Technol. 10 (2013) 10–13.
[18] R. Yoda, J. Biomater. Sci. Polym. Ed. 9 (1998) 561–626.
[19] Y. Marois, R. Guidoin, Biocompatibility of polyurethanes, in: P. Vermette, H.J.
Griesser, G. Laroche, R. Guidoin (Eds.), Biomedcial Applications of Polyurethanes,
Landies Bioscience, Georgetown, TX, 2001, pp. 77–96.
[20] F. D’Angelo, L. Murena, E. Vulcano, G. Zatti, P. Cherubino, Hip Int. 20 (2010)
81–86.
[21] J. Ganeles, M.A. Listgarten, C.I. Evian, J. Periodontol. 57 (1986) 133–140.
[22] R.R. Seeley, T.D. Stephens, P. Rate, Anatomy and Physiology, McGrew Hill, New
York, 2006.
[23] Whitney, Rolfes, Understanding Nutrition.
[24] JAMA 202 (1967) 1145.
[25] D.G. Barceloux, J. Toxicol. Clin. Toxicol. 37 (1999) 201–216.
[26] N. Hallab, J. Clin. Rheumatol. 7 (2001) 215–218.
[27] M. Niinomi, J. Miner. Met. Mater. Soc. 51 (1999) 32–34.
[28] J.J. Jacobs, A.K. Skipor, L.M. Patteson, N.J. Hallab, W.G. Paprosky, J. Black, J.O.
Galante, J. Bone Joint Surg. Am. 80A (1998) 1447–1458.
[29] J.J. Jacobs, C. Silverton, N.J. Hallab, A.K. Skipor, L. Patterson, J. Black, J.O. Galante,
Clin. Orthop. 358 (1999) 173–180.
[30] M. Brayda-Bruno, M. Fini, G. Pierini, G. Giavaresi, M. Rocca, R. Giardino, Int. J.
Artif. Organs 24 (2001) 41–49.
[31] M. Sumita, T. Hanawa, Failure processes in biometallic materials, Bioengineer-
ing, vol. 9, Elsevier Science Ltd., London, 2003, pp. 131–167.
[32] T. Kokubo, K. Hata, T. Nakamura, T. Yamamura, Apatite formation on ceramics,
metals, and polymers induced by a CaO–SiO2-based glass in simulated body
fluid, in: W. Bonfield, G.W. Hastings, K.E. Tanner (Eds.), Bioceramics 4, Butter-
worth-Heinemainn, Guildford, London, 1991, pp. 113–120.
[33] H.J.R.M. Long, Biomaterials 19 (1998) 1621–1639.
[34] S. Suresh, Fatigue of Materials, Cambridge University Press, Cambridge, UK,
2003.
[35] R.A. Antunes, M.C. Lopes de Oliveira, Acta Biomater. 8 (2012) 937–962.
[36] J.J. O’Connor, The role of elastic stress analysis inthe interpretation of fretting
fatigue failures, in: R.B. Waterhouse (Ed.), Fretting Fatigue, Applied Science
Publishers, London, 1981, pp. 23–66.
[37] A. Yamamoto, T. Kobayashi, N. Maruyama, K. Nakazawa, M. Sumita, J. Jpn. Inst.
Met. 59 (1995) 463–470.
[38] H.F. Hildebrand, J.C. Hornez, Biological response and biocompatibility, in: J.A.
Helsen, H.J. Breme (Eds.), Metals as Biomaterials, Wiley & Sons, Chichester, UK,
1998, pp. 265–290.
[39] R.S. Petrie, A.D. Hanssen, D.R. Osmon, D. Ilstrup, Am. J. Orthop. 27 (1998)
172–176.
[40] R.B. Waterhouse, Wear 58 (1980) 391–392.
[41] R.B. Waterhouse, M. Lamb, Wear 60 (1980) 357–368.
[42] K. Nakazawa, M. Sumita, N. Maruyama, Effect of contact pressure on fretting
fatigue of high strength steel and titanium alloy, in: M. Helmi Attia, R.B. Water-
house (Eds.), Standardization of Fretting Fatigue Test Methods and Equipment,
ASTM STP 1159, American Society for Testing and Materials, Philadelphia, 1992,
pp. 115–125.
[43] S.W. Robertson, A.R. Pelton, R.O. Ritchie, Int. Mater. Rev. 57 (2012) 1–36.
[44] Mettalic materials, J.R. Davies (Ed.), Handbook of Materials for Medical Devices,
ASM International, Materials Park, Ohio, 2003, pp. 21–50.
[45] D.W. Lennox, B.H. Schofield, D.F. McDonald, L.H. Riley, Clin. Orthop. (1987)
171–191.
[46] A.V. Lombardi, T.H. Mallory, B.K. Vaughn, P. Drouillard, J. Bone Joint Surg. Am.
71A (1989) 1337–1342.
[47] A. Schweizer, U. Riede, T.B. Maurer, P.E. Ochsner, Arch. Orthop. Trauma Surg. 123
(2003) 353–356.
[48] I. Papageorgiou, V. Shadrick, S. Davis, L. Hails, R. Schins, R. Newson, J. Fisher, E.
Ingham, C.P. Case, Mutat. Res. 643 (2008) 11–19.
[49] M.B. Nasab, Trends Biomater. Artif. Organs 24 (2010) 69.
[50] B. Subramanian, R. Ananthakumar, A. Kobayashi, M. Jayachandran, J. Mater. Sci.
Mater. Med. 23 (2012) 329–338.
[51] A. Parsapour, S.N. Khorasani, M.H. Fathi, J. Mater. Sci. Technol. 28 (2012)
125–131.
[52] R.L. Sammons, Modifying biomaterial surfaces to optimise interactions with
bone, in: R. Williams (Ed.), Surface Modification of Biomaterials; Methods,
Analysis and Applications, Woodhead Publishing, Oxford, 2011, pp. 365–391.
[53] G. Mani, M.D. Feldman, S. Oh, C.M. Agrawal, Appl. Surf. Sci. 255 (2009)
5961–5970.
[54] Handbook of Stainless Steels, McGraw-Hill, New York, 1977.
[55] J.R. Davis, Nickel, Cobalt, and Their Alloys, ASM International Handbook Com-
mittee, Materials Park, Ohio, 2001.
[56] G. Lütjering, J.C. Williams, Titanium, Springer, Berlin, 2007.
[57] H.R. Chen, Shape Memory Alloys: Manufacture, Properties, and Applications,
Nova Science, New York, 2010.
[58] M.M. Avedesian, H. Baker, Magnesium and Magnesium Alloys, ASM Interna-
tional, Materials Park, OH, 1999.
[59] A. Yamamoto, R. Honma, M. Sumita, J. Biomed. Mater. Res. 39 (1998) 331–340.
[60] J. Alvarado, Biomechanics of Hip and Knee Prostheses, in: Applications of
Engineering Mechanics in Medicine, GED – University of Puerto Rico Mayaguez,
2003.
[61] M. Navarro, A. Michiardi, O. Castaño, J.A. Planell, J. R. Soc. Int. 5 (2008) 1137–1158.
[62] H.J. Raithel, K.H. Schaller, Zentralbl. Bakteriol. 173 (1981) 63–91.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
[63] A. Munoz, M. Costa, Toxicol. Appl. Pharmacol. 260 (2012) 1–16.
[64] R. Sonnleitner, G. Mori, M. Panzenboeck, R. Fluch, Mater. Corros. 61 (2010)
97–104.
[65] M. Rokkum, K. Bye, K.R. Hetland, A. Reigstad, Acta Orthop. Scand. 66 (1995)
435–439.
[66] T.C. Diamantino, L. Guilhermino, E. Almeida, A. Soares, Ecotoxicol. Environ. Saf.
45 (2000) 253–259.
[67] F. Rask, A. Mihic, L. Reis, S.M. Dallabrida, N.S. Ismail, K. Sider, C.A. Simmons, M.A.
Rupnick, R.D. Weisel, R.-K. Li, M. Radisic, Soft Matter. 6 (2010) 5089–5099.
[68] A.D. Mazzocca, J.P. DeAngelis, A.E. Caputo, B.D. Browner, J.W. Mast, M.W.
Mendes, Principles of internal fixation, in: B.D. Browner (Ed.), Skeletal Trauma,
vol. I, Elsevier, W.B. Saunders Company, Philadelphia, 2008.
[69] A.F. Kamath, P.B. Voleti, T.W.B. Kim, J.P. Garino, G.-C. Lee, J. Arthroplast. 26 (2011)
1520–1526.
[70] A. Doran, F.C. Law, M.J. Allen, N. Rushton, Biomaterials 19 (1998) 751–759.
[71] S.J. Lippard, J.M. Berg, Principles of Bioinorganic Chemistry, University Science
Books, Mill Valley, 1994.
[72] J. Umbreit, Am. J. Hematol. 78 (2005) 225–231.
[73] K. Cheney, C. Gumbiner, B. Benson, M. Tenenbein, J. Toxicol. Clin. Toxicol. 33
(1995) 61–66.
[74] A.B. Abdelmageed, F. Oehme, Vet. Hum. Toxicol. 32 (1990) 324–328.
[75] J.O. Hoppe, G.M.A. Marcelli, M.L. Tainter, Am. J. Med. Sci. 230 (1955) 558–571.
[76] S. Brar, D. Henderson, J. Schenck, E.A. Zimmerman, Arch. Neurol. 66 (2009)
371–374.
[77] C. Esteve, E. Alcaide, R. Urena, Aquat. Toxicol. 109 (2012) 176–184.
[78] T. Tsuchiya, Y. Ikarashi, T. Uchima, H. Doi, A. Nakamura, Y. Ohshima, M. Fujimaki,
K. Toyoda, E. Kobayashi, T. Yoneyama, H. Hamanaka, Mater. Trans. 43 (2002)
3058–3064.
[79] S. Denizoglu, Z.Y. Duymus, Dent. Mater. J. 25 (2006) 365–370.
[80] S.L. Meacham, J. Cizdziel, A. Sadik, G. McFadden, A. Buffalo, FASEB J. 17 (2003),
A706–A706.
[81] A.D. Dayan, A.J. Paine, Hum. Exp. Toxicol. 20 (2001) 439–451.
[82] E. Rudolf, Acta Med. (Hradec Kralove) 41 (1998) 55–65.
[83] D.G. Barceloux, J. Toxicol. Clin. Toxicol. 37 (1999) 173–194.
[84] S. Langard, Am. J. Ind. Med. 17 (1990) 189–215.
[85] D.M. Stearns, K.E. Wetterhahn, Abstracts of Papers of the American Chemical
Society, vol. 212, American Chemical Society, New York, 1996, 208-INOR.
[86] N. Becker, J. Occup. Environ. Med. 41 (1999) 294–303.
[87] J. Blair, JAMA 90 (1928) 1927–1928.
[88] D. Colomb, H. Perrot, A.J. Beyvin, Bull. Soc. Fr. Dermatol. Syphiligr. 76 (1969)
130–131.
[89] A. Mahto, B. De Silva, Br. J. Dermatol. 161 (2009) 126–127.
[90] M.B. Hansen, S. Rydin, T. Menne, J. Duus Johansen, Contact Dermat. 47 (2002)
127–134.
[91] M.H. Kim, Y.W. Choi, H.Y. Choi, K.B. Myung, Contact Dermat. 44 (2001) 289–292.
[92] K.D. Rosenman, M. Stanbury, Am. J. Ind. Med. 29 (1996) 491–500.
[93] C.S. Dubey, B.K. Sahoo, N.R. Nayak, Bull. Environ. Contam. Toxicol. 67 (2001)
541–548.
[94] S.C.B. Oliveira, A.M. Oliveira-Brett, Anal. Bioanal. Chem. 398 (2010) 1633–1641.
[95] K.P. Nickens, S.R. Patierno, S. Ceryak, Chem. Biol. Interact. 188 (2010) 276–288.
[96] S. Burkhardt, R.J. Reiter, D.X. Tan, R. Hardeland, J. Cabrera, M. Karbownik, Int. J.
Biochem. Cell Biol. 33 (2001) 775–783.
[97] E. Rudolf, M. Cervinka, Toxicol. Lett. 188 (2009) 236–242.
[98] W.J. Berry, W.S. Boothman, J.R. Serbst, P.A. Edwards, Environ. Toxicol. Chem. 23
(2004) 2981–2992.
[99] I.A. AlSaleh, Sci. Total Environ. 181 (1996) 215–221.
[100] I. Al-Saleh, I. Al-Doush, Sci. Total Environ. 216 (1998) 181–192.
[101] S.I. Segura-Munoz, T.M.B. Trevilato, A.M.M. Takayanagui, S.E. Hering, P. Cupo,
Arch. Latinoam. Nutr. 53 (2003) 59–64.
[102] J. Ryhanen, Biocompatibilty Evaluation of Nickel–Titanium Shape Memory
Metal Alloy, Oulu University Linrary, Oulu, 1999.
[103] H. Sigel, A. Sigel, R.K.O. Sigel, Nickel and Its Surprising Impact in Nature: Metal
Ions in Life Sciences, John Wiley & Sons, New York, 2007.
[104] J.L. Nielsen, O.M. Poulsen, A. Abildtrup, Electrophoresis 15 (1994) 666–671.
[105] W.N. Rezuke, J.A. Knight, F.W. Sunderman, Am. J. Ind. Med. 11 (1987) 419–426.
[106] M. Anke, B. Groppel, H. Kronemann, M. Grun, IARC Sci. Publ. (1984) 339–365.
[107] M. Szilagyi, M. Anke, I. Balogh, Acta Vet. Hung. 39 (1991) 231.
[108] F.H. Nielsen, T.R. Shuler, T.G. McLeod, T.J. Zimmerman, J. Nutr. 114 (1984)
1280–1288.
[109] G.I. Stangl, M. Kirchgessner, Trace Elem. Electrolytes 13 (1996) 117–122.
[110] F.H. Nielsen, D.R. Myron, S.H. Givand, T.J. Zimmerman, D.A. Ollerich, J. Nutr. 105
(1975) 1620–1630.
[111] F.H. Nielsen, D.R. Myron, S.H. Givand, D.A. Ollerich, J. Nutr. 105 (1975) 1607–1619.
[112] G.I. Stangl, M. Kirchgessner, J. Anim. Physiol. Anim. Nutr. 75 (1996) 164–174.
[113] G.I. Stangl, M. Kirchgessner, J. Nutr. 126 (1996) 2466–2473.
[114] B. Santucci, P.V. Ferrari, A. Cristaudo, C. Cannistraci, M. Picardo, Contact Dermat.
21 (1989) 245–248.
[115] L.E. Gaul, Arch. Dermatol. 78 (1958) 475–478.
[116] http://en.wikipedia.org/wiki/Allergen_of_the_Year.
[117] T. Kawamoto, P. Thi-Thu-Phuong, T. Matsuda, T. Oyama, M. Tanaka, H.-S. Yu, I.
Uchiyama, Int. J. Hyg. Environ. Health 214 (2011) 296–304.
[118] K. Hughes, M.E. Meek, P.K.L. Chan, J. Shedden, S. Bartlett, L.J. Seed, J. Environ. Sci.
Health C 12 (1994) 417–433.
[119] R.B. Ashman, J.G. Birch, L.B. Bone, J.D. Corin, J.A. Herring, C.E. Johnston 2nd, J.F.
Ritterbush, J.W. Roach, Clin. Orthop. 227 (1988) 113–125.
[120] F.W. Sunderman, Ann. Clin. Lab. Sci. 7 (1977) 377–398.
53
[121] J. Zhao, X. Shi, V. Castranova, M. Ding, J. Environ. Pathol. Toxicol. Oncol. 28 (2009)
177–208.
[122] A.R. Oller, M. Costa, G. Oberdorster, Toxicol. Appl. Pharmacol. 143 (1997)
152–166.
[123] M.P. Abbracchio, J. Simmonshansen, M. Costa, J. Toxicol. Environ. Health 9 (1982)
663–676.
[124] M.P. Abbracchio, R.M. Evans, J.D. Heck, O. Cantoni, M. Costa, Biol. Trace Elem. Res.
4 (1982) 289–301.
[125] F.W. Sunderman Jr., J. UOEH 9 (Suppl.) (1987) 84–94.
[126] J.K. Dunnick, M.R. Elwell, A.E. Radovsky, J.M. Benson, F.F. Hahn, K.J. Nikula, E.B.
Barr, C.H. Hobbs, Cancer Res. 55 (1995) 5251–5256.
[127] C.B. Klein, K. Frenkel, M. Costa, Chem. Res. Toxicol. 4 (1991) 592–604.
[128] C.B. Klein, K. Conway, X.W. Wang, R.K. Bhamra, X.H. Lin, M.D. Cohen, L. Annab,
J.C. Barrett, M. Costa, Science 251 (1991) 796–799.
[129] R.B. Hayes, Cancer Causes Control 8 (1997) 371–385.
[130] C.D. Martz, J. Wickstro, V.H. Frankel, P.G. Laing, Can. Med. Assoc. J. 96 (1967) 568.
[131] P.G. Laing, A.B. Ferguson, E.S. Hodge, J. Bone Joint Surg. Am. 47 (1965) 1102.
[132] K. Takamura, K. Hayashi, N. Ishinishi, T. Yamada, Y. Sugioka, J. Biomed. Mater.
Res. 28 (1994) 583–589.
[133] M. Bergman, B. Bergman, R. Soremark, J. Oral Rehabil. 7 (1980) 325–330.
[134] S.J. Lugowski, D.C. Smith, A.D. McHugh, J.C. Vanloon, J. Biomed. Mater. Res. 25
(1991) 1443–1458.
[135] S. Ishimatsu, T. Kawamoto, K. Matsuno, Y. Kodama, Biol. Trace Elem. Res. 49
(1995) 43–52.
[136] M.C. Pereira, M.L. Pereira, J.P. Sousa, J. Biomed. Mater. Res. 40 (1998) 40–47.
[137] R. Michel, J. Zilkens, Z. Orthop. Ihre Grenzgeb. 116 (1978) 666–674.
[138] S. Hierholzer, G. Hierholzer, K.H. Sauer, R.S. Paterson, Arch. Orthop. Trauma Surg.
102 (1984) 198–200.
[139] R.M. De Souza, L.M. De Menezes, Angle Orthod. 78 (2008) 345–350.
[140] I. Bruce, C.E. Swanson, Spine 32 (2007) 1963–1968.
[141] P. Wiles, Br. J. Surg. 45 (1958) 488–497.
[142] F. Escalas, J. Galante, W. Rostoker, P. Coogan, J. Biomed. Mater. Res. 10 (1976)
175–195.
[143] J.E. Lemons, K.M.W. Niemann, A.B. Weiss, J. Biomed. Mater. Res. 10 (1976)
549–553.
[144] B.C. Syrett, E.E. Davis, J. Biomed. Mater. Res. 13 (1979) 543–556.
[145] W. Rostoker, E.Y.S. Chao, J.O. Galante, J. Biomed. Mater. Res. 12 (1978) 635–651.
[146] J.R. Cahoon, R.N. Holte, J. Biomed. Mater. Res. 15 (1981) 137–145.
[147] D.W. Hoeppner, V. Chandrasekaran, Wear 173 (1994) 189–197.
[148] S.H. Teoh, Int. J. Fatigue 22 (2000) 825–837.
[149] R. Muller, J. Abke, E. Schnell, F. Macionczyk, U. Gbureck, R. Mehrl, Z. Ruszczak, R.
Kujat, C. Englert, M. Nerlich, P. Angele, Biomaterials 26 (2005) 6962–6972.
[150] M. Fini, N.N. Aldini, P. Torricelli, G. Glavaresi, V. Borsari, H. Lenger, J. Bernauer, R.
Glardino, R. Chiesa, A. Cigada, Biomaterials 24 (2003) 4929–4939.
[151] Y. Okazaki, E. Gotoh, Biomaterials 26 (2005) 11–21.
[152] D.-J. Lin, C.-C. Chuang, J.-H.C. Lin, J.-W. Lee, C.-P. Ju, H.-S. Yin, Biomaterials 28
(2007) 2582–2589.
[153] R. Singh, N.B. Dahotre, J. Mater. Sci. Mater. Med. 18 (2007) 725–751.
[154] E.F.S. Hussam, Finite element simulation of Hip joint replacement under static
and dynamic loading, in: School of Mechanical and Manufacturing Engineering,
vol Doctor of Philosophy, Dublin City University, 2002p. 346.
[155] T.S. Sudarshan, T.S. Srivatsan, D.P. Harvey, Eng. Fract. Mech. 36 (1990) 827–852.
[156] R. Ebara, Corrosion fatigue crack initiation behavior of stainless steels, in: P.
Lukas (Ed.), Fatigue 2010 Procedia Engineering, vol. 2, Elsevier, London, UK,
2010, pp. 1297–1306.
[157] Handbook of biomaterials properties, Part II, Chapman and Hall, New York,
1998.
[158] E.J. Giordani, V.A. Guimaraes, T.B. Pinto, I. Ferreira, Int. J. Fatigue 26 (2004)
1129–1136.
[159] N. Maruyama, D. Mori, S. Hiromoto, K. Kanazawa, M. Nakamura, Corros. Sci. 53
(2011) 2222–2227.
[160] B.W. Minto, C.V.S. Brandao, G.J.C. Pereira, D. Campagnol, M.J. Mamprim, C.R.
Padovani, J.J.T. Ranzani, Acta Vet. Scand. 53 (2011), 6 pages.
[161] M.D. Roach, R.S. Williamson, L.D. Zardiackas, Comparison of the corrosion
fatigue characteristics of 23Mn–21Cr–1Mo low nickel, 22Cr–13Ni–5Mn, and
18
Cr–14Ni–2.5Mo stainless steels, in: K.L. Jerina (Ed.), Fatigue and Fracture of
Medical Metallic Materials and Devices, vol. 1481, American Society for Testing
and Materials, West Conshohocken, PA, 2007, pp. 56–66.
[162] M. Taira, E.P. Lautenschlager, J. Biomed. Mater. Res. 26 (1992) 1131–1139.
[163] L.M. Weldon, P.E. McHugh, W. Carroll, E. Costello, C. O’Bradaigh, J. Mater. Sci.
Mater. Med. 16 (2005) 107–117.
[164] S.G. Rao, V. Singh, L.K. Singhal, Mater. Sci. Eng. A 538 (2012) 224–230.
[165] W. Pospula, Kumait Med. J. 36 (2004) 250–255.
[166] J. Walczak, F. Shahgaldi, F. Heatley, Biomaterials 19 (1998) 229–237.
[167] J. Allain, F. Roudot-Thoraval, J. Delecrin, P. Anract, H. Migaud, D. Goutallier, J.
Bone Joint Surg. Am. 85A (2003) 825–830.
[168] J. Allain, D. Goutallier, M.C. Voisin, S. Lemouel, J. Bone Joint Surg. Am. 80A (1998)
1355–1360.
[169] S.S.M. Tavares, F.B. Mainier, F. Zimmerman, R. Freitas, C.M.I. Ajus, Eng. Fail. Anal.
17 (2010) 1246–1253.
[170] U.K. Mudali, T.M. Sridhar, B. Raj, SADHANA 28 (2003) 601–637.
[171] U.K. Mudali, T.M. Sridhar, N. Eliaz, B. Raj, Corros. Rev. 21 (2003) 231–267.
[172] O. Reigstad, P. Siewers, M. Rokkum, B. Espehaug, Acta Orthop. 79 (2008)
194–202.
[173] W.J. van Doorn, F.C. van Biezen, P.J. Prendergast, J.A.N. Verhaar, Acta Orthop.
Scand. 73 (2002) 111–113.
54 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
[174] K. Yang, Y. Ren, P. Wan, Sci. China Technol. Sci. 55 (2012) 329–340.
[175] K. Yang, Y. Ren, Sci. Technol. Adv. Mater. 11 (2010) 014105–014118.
[176] R.F. Vera Villamil Jaimes, M.L. Chaves de Andrade Afonso, S.O. Rogero, S.M. Leite
Agostinho, C.A. Barbosa, Mater. Lett. 64 (2010) 1476–1479.
[177] K. Nakazawa, M. Sumita, N. Maruyama, J. Jpn. Inst. Met. 63 (1999) 1600–1608.
[178] K. Nakazawa, M. Sumita, N. Maruyama, Fatigue and fretting fatigue of austenitic
and ferritic stainless steels in pseudo-body fluid, in: Fatigue’99: Proceedings of
the 7th International Fatigue Congress, Beijing (P.R. China), 8–12 June, 1999, X.R.
Wu (Ed.), EMAS Publishing, vol. 2, pp. 1359–1364.
[179] Q.Z. Chen, D.M. Knowles, Mater. Sci. Eng. A 356 (2003) 352–367.
[180] S. Pramanik, A.K. Agarwal, K.N. Rai, Trends Biomater. Artif. Organs 19 (2005)
15–26.
[181] IARC, On the Evalution of Carcinogenic Risks to Humans: Surgical Implants and
Other Foreign Bodies, IARC, Lyon, 1999.
[182] J.J. Ramsden, D.M. Allen, D.J. Stephenson, J.R. Alcock, G.N. Peggs, G. Fuller, G.
Goch, CIRP Ann. Manuf. Technol. 56 (2007) 687–711.
[183] O. Öztürk, U. Türkan, A.E. Eroğlu, Surf. Coat. Technol. 200 (2006) 5687–5697.
[184] C.V. Vidal, A.I. Muñoz, Electrochim. Acta 54 (2009) 1798–1809.
[185] D.G. Kennedy, P.B. Young, S. Kennedy, J.M. Scott, A.M. Molloy, D.G. Weir, J. Price,
Int. J. Vitam. Nutr. Res. 65 (1995) 241–247.
[186] D.G. Kennedy, P.B. Young, W.J. McCaughey, S. Kennedy, W.J. Blanchflower, J.
Nutr. 121 (1991) 1236–1242.
[187] G.I. Stangl, F.J. Schwarz, M. Kirchgessner, Nutr. Res. 19 (1999) 415–427.
[188] G.I. Stangl, F.J. Schwarz, M. Kirchgessner, J. Anim. Physiol. Anim. Nutr. 80 (1998)
40–48.
[189] S.S. Tower, Br. Med. J. 344 (2012) e430 (abstract).
[190] S.S. Tower, J. Bone Joint Surg. Am. 92A (2010) 2847–2851.
[191] M. Kobayashi, S. Shimizu, Eur. J. Biochem. 261 (1999) 1–9.
[192] J.D. Donaldson, D. Beyersmann, Cobalt and Cobalt Compounds, in: V.A. Welch,
K.J. Fallon, H.P. Gelbke (Eds.), Ullmann’s Encyclopedia of Industrial Chemistry
2005, Wiley-VCH, Weinheim, 2005.
[193] D.A. Basketter, G. Angelini, A. Ingber, P.S. Kern, T. Menne, Contact Dermatitis 49
(2003) 1–7.
[194] D. Lison, Crit. Rev. Toxicol. 26 (1996) 585–616.
[195] D.A. Jones, H.K. Lucas, M. Odriscoll, C.H.G. Price, B. Wibberley, J. Bone Joint Surg.
Am. 57 (1975) 289–296.
[196] X. Mao, A.A. Wong, R.W. Crawford, Med. J. Aust. 194 (2011) 649–651.
[197] S.S. Tower, Bulletin of Devision of Public Health, State of Alaska Epidermiology,
2010, http://www.epi.alaska.gov/.
[198] J.R. Turnlund, Molybdenum metabolism and requirements in humans, in: A.
Sigel, H. Sigel (Eds.), Molybdenum and Tungst their Roles in Biological Processes.
Handbook on the Toxicology of Metals, Elsevier, London, UK, 2002.
[199] U.E. Pazzaglia, P. Apostoli, T. Congiu, S. Catalani, M. Marchese, G. Zarattini, Arch.
Orthop. Trauma Surg. 131 (2011) 1299–1308.
[200] M.E.J. Curzon, J. Kubota, B.G. Bibby, J. Dent. Res. 50 (1971) 74.
[201] P.R. Taylor, B. Li, S.M. Dawsey, J.Y. Li, C.S. Yang, W.D. Guo, W.J. Blot, W. Weng, B.Q.
Liu, S.F. Zheng, Q. Yang, Y. Yu, Y. Sun, G.Y. Li, S.F. Liu, X.N. Zhou, Y.P. Wu, Z. Wang,
S.X. Lu, Y.H. Zhang, K. Yang, G. Yang, Z. Chen, Z.Y. Wang, D.H. Zheng, G.T. Lian, J.A.
Tangrea, A.G. Ershow, S.D. Mark, M. Gail, B.J. Stone, J.F. Fraumeni, P. Greenwald,
K. Lewin, R. Nieberg, M. Weiner, W. Weinstein, P. Engstrom, P. Correa, S. Lagakos,
Cancer Res. 54 (1994) S2029–S2031.
[202] D.G. Barceloux, J. Toxicol. Clin. Toxicol. 37 (1999) 231–237.
[203] M.P. Coughlan, J. Inherit. Metab. Dis. 6 (1983) 70–77.
[204] Nutr. Rev. 18 (1960) 54–56.
[205] K. Liber, L.E. Doig, S.L. White-Sobey, Ecotoxicol. Environ. Saf. 74 (2011) 1171–1179.
[206] M. Anke, M. Seifert, W. Arnhold, S. Anke, U. Schaefer, Acta Aliment. 39 (2010)
12–26.
[207] A. Vyskocil, C. Viau, J. Appl. Toxicol. 19 (1999) 185–192.
[208] W.-C. Witzleb, J. Ziegler, F. Krummenauer, V. Neumeister, K.-P. Guenther, Acta
Orthop. 77 (2006) 697–705.
[209] J. Antoniou, D.J. Zukor, F. Mwale, W. Minarik, A. Petit, O.L. Huk, J. Bone Joint Surg.
Am. 90A (2008) 142–148.
[210] A. Kletzin, M.W.W. Adams, FEMS Microbiol. Rev. 18 (1996) 5–63.
[211] J.P. Smart, M.J. Cliff, D.J. Kelly, Mol. Microbiol. 74 (2009) 742–757.
[212] M. Peuster, C. Fink, C. von Schnakenburg, G. Hausdorf, Cardiol. Young 12 (2002)
229–235.
[213] D. Lison, J.P. Buchet, P. Hoet, Lancet 349 (1997) 58.
[214] M. Peuster, C. Fink, C. von Schnakenburg, Biomaterials 24 (2003) 4057–4061.
[215] M.L. Witten, P.R. Sheppard, B.L. Witten, Chem. Biol. Interact. 196 (2012) 87–88.
[216] A.T. Moore, H.R. Bohlman, J. Bone Joint Surg. Am. 25 (1943) 688–692.
[217] E.J. Evans, I.T. Thomas, Biomaterials 7 (1986) 25–29.
[218] C. Delaunay, I. Petit, I.D. Learmonth, P. Oger, P.A. Vendittoli, Orthop. Traumatol.
Surg. Res. 96 (2010) 894–904.
[219] M.G. Zywiel, S.A. Sayeed, A.J. Johnson, T.P. Schmalzried, M.A. Mont, Expert Rev.
Med. Dev. 8 (2011) 187–207.
[220] G.L.t. (Gerd), J.C.W.J. Case), Titanium, Springer, New York, 2003.
[221] A.E. Goodship, T.J. Lawes, C.T. Rubin, J. Orthop. Res. 27 (2009) 922–930.
[222] F.G. Evans, H. Yamada, Strength of Biological Materials, Williams & Wilkins,
Baltimore, 1970.
[223] J.B. Brunski, Metals, in: B.D. Ratner, A.S. Hoffman, F.J. Schoen, J.E. Lemons (Eds.),
Biomaterials Science: An introduction to Materials in Medicine, Academic Press,
London, 1996, pp. 37–50.
[224] A.M. Weinstein, A.J.T. Clemow, Cobalt-based alloys, in: D.F. Williams (Ed.),
Concise Encyclopaedia of Medical & Dental Materials, Pergamon Press and
the MIT Press, 1990, pp. 106–112.
[225] Cobalt-based Alloys, ASM International, Materials Park, Ohio, 2000.
[226] T.M. Grupp, T. Weik, W. Bloemer, H.-P. Knaebel, BMC Musculoskelet. Disord. 11
(2010) 3 (abstract).
[227] H.A. Luckey, L.J. Barnard, Improved Properties of Co–Cr–Mo Alloy by Hot
Isostatic Pressing of Powder, John Wiley and Sons, New York, 1980.
[228] M. Niinomi, Int. J. Fatigue 29 (2007) 992–1000.
[229] S. Garg, P.W. Serruys, J. Am. Coll. Cardiol. 56 (2010) S1–S42.
[230] D.J. Kereiakes, D.A. Cox, J.B. Hermiller, M.G. Midei, W.B. Bachinsky, E.D. Nukta,
M.B. Leon, S. Fink, L. Marin, A.J. Lansky, R. Guidant MultiLink Vision Stent, Am. J.
Cardiol. 92 (2003) 463–466.
[231] A. Biesiekierski, D.H. Ping, Y. Yamabe-Mitarai, C. Wen, Mater. Des. 59 (2014)
303–309.
[232] I. Pais, M. Feher, E. Farkas, Z. Szabo, I. Cornides, Commun. Soil Sci. Plant Anal. 8
(1977) 407–410.
[233] S. Yaghoubi, C.W. Schwietert, J.P. McCue, Biol. Trace Elem. Res. 78 (2000)
205–217.
[234] M.L. Wang, R. Tuli, P.A. Manner, P.F. Sharkey, D.J. Hall, R.S. Tuan, J. Orthop. Res. 21
(2003) 697–707.
[235] N. Coen, M.A. Kadhim, E.G. Wright, C.P. Case, C.E. Mothersill, Br. J. Cancer 88
(2003) 548–552.
[236] R. Kumazawa, F. Watari, N. Takashi, Y. Tanimura, M. Uo, Y. Totsuka, Biomaterials
23 (2002) 3757–3764.
[237] V.V. Koval’skii, L.T. Rezaeva, Usp. Sovrem. Biol. 60 (1965) 45–61.
[238] E.P. Daniel, E.M. Hewston, Am. J. Physiol. 136 (1942) 0772–0775.
[239] M.A. Altamirano-Lozano, M.E. Roldan-Reyes, E. Rojas, Vanadium in the Envi-
ronment, Pt. 2: Health Effects, vol. 31, 1998, pp. 159–179.
[240] L.S. Rhoads, W.T. Silkworth, M.L. Roppolo, M.S. Whittingham, Toxicol. In Vitro 24
(2010) 292–296.
[241] N.B. Ress, B.J. Chou, R.A. Renne, J.A. Dill, R.A. Miller, J.H. Roycroft, J.R. Hailey, J.K.
Haseman, J.R. Bucher, Toxicol. Sci. 74 (2003) 287–296.
[242] B. Moretti, V. Pesce, G. Maccagnano, G. Vicenti, P. Lovreglio, L. Soleo, P. Apostoli,
Lancet 379 (2012) 1676.
[243] S.V. Verstraeten, L. Aimo, P.I. Oteiza, Arch. Toxicol. 82 (2008) 789–802.
[244] W.A. Banks, A.J. Kastin, Neurosci. Biobehav. Rev. 13 (1989) 47–53.
[245] R.A. Yokel, Neurotoxicology 21 (2000) 813–828.
[246] D.N.S. Kerr, M.K. Ward, H.A. Ellis, W. Simpson, I.S. Parkinson, Ciba Found. Symp.
169 (1992) 123–141.
[247] P.D. Darbre, J. Appl. Toxicol. 26 (2006) 191–197.
[248] P.D. Darbre, Best Pract. Res. Clin. Endocrinolo. Metab. 20 (2006) 121–143.
[249] P.C. Ferreira, K.d.A. Piai, A.M. Magosso Takayanagui, S.I. Segura-Munoz, Rev. Lat.
Am. Enfermagem 16 (2008) 151–157.
[250] W.L. Downs, J.K. Scott, C.L. Yuile, F.S. Caruso, L.C.K. Wong, Am. Ind. Hyg. Assoc. J.
26 (1965) 337.
[251] M. Caicedo, J.J. Jacobs, A. Reddy, N.J. Hallab, J. Biomed. Mater. Res. A 86A (2008)
905–913.
[252] T. Kodama, Kokubyo Gakkai zasshi, J. Stomatol. Soc. Jpn. 56 (1989) 263–288.
[253] M. Niinomi, Mater. Trans. 49 (2008) 2170–2178.
[254] M. Niinomi, Metall. Mater. Trans. A 33 (2002) 477–486.
[255] V.D. Cojocaru, D. Raducanu, I. Cinca, I. Dan, S. Ivanescu, E. Jalba, Metal. Int. 16
(2011) 35–38.
[256] D. Raducanu, E. Vasilescu, V.D. Cojocaru, I. Cinca, P. Drob, C. Vasilescu, S.I. Drob, J.
Mech. Behav. Biomed. Mater. 4 (2011) 1421–1430.
[257] Y.X. Tong, B. Guo, Y.F. Zheng, C.Y. Chung, L.W. Ma, J. Mater. Eng. Perform. 20
(2011) 762–766.
[258] M. Niinomi, Mater. Sci. Eng. A 243 (1998) 231–236.
[259] M. Niinomi, J. Mech. Behav. Biomed. Mater. 1 (2008) 30–42.
[260] M. Niinomi, Recent Biocompatible Metallic Materials, in: N. Niinomi, T. Okabe,
E.M. Taleff, D.R. Lesure, H.E. Lippard (Eds.), Structural Biomaterials for the 21st
Century, 2001, the 2001 TMS Annual Meeting in New Orleans, LA, on February
11–12, p. 3–14.
[261] Coatings, D. J.R. (Ed.), Handbook of Materials for Medical Devices, ASM Interna-
tional, Materials Park, Ohio, 2003, pp. 179–194.
[262] X. Song, M. Niinomi, M. Nakai, H. Tsutsumi, L. Wang, Mater. Sci. Eng. C 32 (2012)
542–549.
[263] Y. Li, C. Han, X.K. Zhu, C. Wen, P. Hodgson, J. Mater. Sci. 47 (2012) 4410–4414.
[264] Y. Wang, C. Wong, C. Wen, P. Hodgson, Y. Li, J. Mater. Chem. B 2 (2014)
5854–5861.
[265] M. Nakai, M. Niinomi, T. Oneda, Metall. Mater. Trans. A: Phys. Metall. Mater. Sci.
43 (2012) 294–302.
[266] M. Long, H.J. Rack, Biomaterials 19 (1998) 1621–1639.
[267] J. Charnley, Z. Cupic, Clin. Orthop. 95 (1973) 9–25.
[268] M. Harty, Orthop. Clin. North Am. 13 (1982) 667–679.
[269] H. Malchau, P. Herberts, Scientific Exhibition Presented at the 65th Annual
Meeting of the American Academy of Orthopaedic Surgeons, 1998, 19–23.
[270] M. Jarcho, J.F. Kay, K.I. Gumaer, R.H. Doremus, H.P. Drobeck, J. Bioeng. 1 (1977)
79–92.
[271] L.L. Hench, R.J. Splinter, W.C. Allen, J. Biomed. Mater. Res. 2 (Pt. 1) (1971)
117–141.
[272] L.L. Hench, H.A. Paschall, J. Biomed. Mater. Res. 7 (1973) 25–42.
[273] L.L. Hench, H.F. Paschall, M. Paschall, J. McVey, Am. Ceram. Soc. Bull. 52 (1973)
432.
[274] Q.Z. Chen, C.T. Wong, W.W. Lu, K.M.C. Cheung, J.C.Y. Leong, K.D.K. Luk, Bioma-
terials 25 (2004) 4243–4254.
[275] W.D. Callister, Materials Science and Engineering: An Introduction, John Wiley
& Sons, Inc., New York, 2007.
[276] Q.Z. Chen, K. Rezwan, D. Armitage, S.N. Nazhat, A.R. Boccaccini, J. Mater. Sci.
Mater. Med. 17 (2006) 979–987.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
[277] F.X. Maquart, G. Bellon, B. Chaqour, J. Wegrowski, L.M. Patt, R.E. Trachy, J.C.
Monboisse, F. Chastang, P. Birembaut, P. Gillery, J.P. Borel, J. Clin. Invest. 92
(1993) 2368–2376.
[278] T. Kitsugi, T. Nakamura, M. Oka, W.Q. Yan, T. Goto, T. Shibuya, T. Kokubo, S.
Miyaji, J. Biomed. Mater. Res. 32 (1996) 149–156.
[279] H.M. Kim, F. Miyaji, T. Kokubo, T. Nakamura, J. Biomed. Mater. Res. 38 (1997)
121–127.
[280] X.L. Zhao, M. Niinomi, M. Nakai, T. Ishimoto, Mat. Sc. Eng. C 31 (2011)
1436–1444.
[281] E. Kobayashi, M. Ando, Y. Tsutsumi, H. Doi, T. Yoneyama, M. Kobayashi, T.
Hanawa, Mater. Trans. 48 (2007) 301–306.
[282] R.S. Kirby, S.R. Heard, P. Miller, I. Eardley, S. Holmes, J. Valve, J. Urol. 148 (1992)
1192–1195.
[283] K. Machara, K. Doi, T. Matsushita, Y. Susaki, Mater. Trans. 43 (2002) 2936–2942.
[284] C. Boehlert, M. Niinomi, M. Ikeda, Mater. Sci. Eng. C 25 (2005) 247.
[285] S. Nag, R. Banerjee, H.L. Fraser, Mater. Sci. Eng. C 25 (2005) 357–362.
[286] F.a.D. Administration (Ed.) K993768, 2000.
[287] G. Luetjring, J. Albrecht, C. Sauer, T. Krull, Mater. Sci. Eng. A 468 (2007) 201–209.
[288] N.J. Eynonlewis, D. Ferry, M.F. Pearse, Br. Med. J. 305 (1992) 1534–1536.
[289] R.A. Brand, M.A. Mont, M.M. Manring, Clin. Orthop. 469 (2011) 1525–1527.
[290] S. hughes, I. McCarthy, Science Basic to Orthopaedics, WB Saunders Company
Ltd., Philadelphia, PA, 1998.
[291] M. Rang, Anthology of Orthopaedics, Churchil Livingstone, London, 1966.
[292] T. Gluck, Arch. Klin. Chir. 41 (1891) 187–239.
[293] P.F. Gomez, J.A. Morcuende, Iowa Orthop. J. 25 (2005) 25–29.
[294] M.N. Smith-Peterspn, J. Bone Joint Surg. (Br) 21B (1939) 269–288.
[295] A.T. Moore, H.R. Bohlman, Clin. Orthop. (2006) 22–24.
[296] R.B. Welch, J. Charnley, Arthritis Rheum. 13 (1970) 358.
[297] G.K. McKee, Clin. Orthop. (1970) 85.
[298] G.K. McKee, J. Bone Joint Surg. Am. 40 (1958) 345.
[299] G.K. McKee, J. Watsonfarrar, J. Bone Joint Surg. Am. 45 (1963) 624.
[300] G.K. McKee, J. Bone Joint Surg. Am. 45 (1963) 616.
[301] P.A. Ring, Ann. R. Coll. Surg. Engl. 48 (1971) 344–355.
[302] P.A. Ring, Proc. R. Soc. Med. 60 (1967) 281.
[303] P.A. Ring, J. Bone Joint Surg. Br. 50 (1968) 720–731.
[304] C.C.P.M. Verheyen, J.A.N. Verhaar, Lancet 380 (2012) 106.
[305] A.J. Smith, P. Dieppe, K. Vernon, M. Porter, A.W. Blom, Lancet 379 (2012) 1199–1204.
[306] E.W. Paxton, C.F. Ake, M.C.S. Inacio, M. Khatod, D. Marinac-Dabic, A. Sedrakyan,
Pharmacoepidemiol. Drug Saf. 21 (2012) 53–59.
[307] M.E. Torchia, R.A. Klassen, A.J. Bianco, J. Bone Joint Surg. Am. 78A (1996)
995–1003.
[308] I.D. Learmonth, C. Young, C. Rorabeck, Lancet 370 (2007) 1508–1519.
[309] G. Labek, M. Thaler, W. Janda, M. Agreiter, B. Stoeckl, J. Bone Joint Surg. Am. 93B
(2011) 998.
[310] T.Q. Nguyen, J.M. Buckley, C. Ames, V. Deviren, Proc. Inst. Mech. Eng. H 225
(2011) 194–198.
[311] M. Salahshoor, Y.B. Guo, Mater. Des. 5 (2012) 135–155.
[312] D. Persaud-Sharma, A. McGoron, J. Biomim. Biomater. Tissue Eng. 12 (2012) 25–39.
[313] P. Erne, M. Schier, T.J. Resink, Cardiovasc. Intervent. Radiol. 29 (2006) 11–16.
[314] R. Erbel, C. Di Mario, J. Bartunek, J. Bonnier, B. de Bruyne, F.R. Eberli, P. Erne, M.
Haude, B. Heublein, M. Horrigan, C. Ilsley, D. Boese, J. Koolen, T.F. Luescher, N.
Weissman, R. Waksman, P.-A. Investigators, Lancet 369 (2007) 1869–1875.
[315] P. Peeters, M. Bosiers, J. Verbist, K. Deloose, B. Heublein, J. Endovasc. Ther. 12
(2005) 1–5.
[316] C. Di Mario, H. Griffiths, O. Goktekin, N. Peeters, J. Verbist, M. Bosiers, K. Deloose,
B. Heublein, R. Rohde, V. Kasese, C. Ilsley, R. Erbel, J. Intervent. Cardiol. 17 (2004)
391–395.
[317] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C.J. Wirth, H.
Windhagen, Biomaterials 26 (2005) 3557–3563.
[318] M. Salahshoor, Y.B. Guo, Materials 5 (2012) 135–155.
[319] Y. Li, S.J. Poon, G.J. Shiflet, J. Xu, D.H. Kim, J.F. Loeffler, MRS Bull. 32 (2007)
624–628.
[320] A.L. Greer, Mater. Today 12 (2009) 14–22.
[321] X.N. Gu, Y.F. Zheng, Mater. Sci. China 4 (2010) 111–115.
[322] G.L. Makar, J. Kruger, Int. Mater. Rev. 38 (1993) 138–153.
[323] R.W. Revie, H. Uhlig, Corrosion and Corrosion Control, Wiley-Interscience, 2008.
[324] Y.F. Jiang, H.S. Yang, Y.F. Bao, Y.Y. Zhang, Characterization of composite coatings
produced by plasma electrolytic oxidation plus cathodic electrophoretic depo-
sition on magnesium alloy, in: S. Jiao, Z.Y. Jiang, J.L. Bu (Eds.), Advances in
Superalloys, Pts. 1 and 2, vols. 146–147, Trans Tech Publications Ltd., Stafa-
Zurich, 2011, pp. 1126–1131.
[325] Y.C. Lee, A.K. Dahle, D.H. StJohn, Metallur. Mater. Trans. A 31 (2000) 2895–2906.
[326] A. Shibata, M. Suzuki, T. Tsukeda, K. Saito, K. Maruyama, J. Jpn. Inst. Met. 70
(2006) 968–974.
[327] S. Spigarelli, M. Regev, E. Evangelista, A. Rosen, Mater. Sci. Technol. 17 (2001)
627–638.
[328] E. Aghion, B. Bronfin, Magnesium alloys development towards the 21(st) centu-
ry, in: Y.A.T.K.S. Kojima (Ed.), Magnesium Alloys 2000, Materials Science Forum,
vols. 350–351, 2000, pp. 19–30.
[329] S. Virtanen, Mater. Sci. Eng. B 176 (2011) 1600–1608.
[330] M. Salahshoor, Y. Guo, Materials 5 (2012) 135–155.
[331] B. Zberg, E.R. Arata, P.J. Uggowitzer, J.F. Loeffler, Acta Mater. 57 (2009) 3223–3231.
[332] A.L. Pleshchitser, Clin. Chem. 4 (1958) 429–450.
[333] I.Y. Torshin, O.A. Gromova, The biological roles of magnesium, in: Magnesium
and Pyridoxine: Fundamental Studies and Clinical Practice, Nova Science Pub-
lishers, Inc, New York, 2009, pp. 1–17.
55
[334] R. Stahlmann, J. Antimicrob. Chemother. 47 (2001) 11–12.
[335] W.S. Kim, Magnesium toxicity and its use for prediction of calcium deficiency in
apple fruit before harvest, in: J.M.L.M.E. Val (Ed.), Mineral Nutrition and Fertil-
izer Use for Deciduous Fruit Crops, International Society for Horticultural
Science, Brussels, Belgium, 1997, pp. 358–359.
[336] R.J.P. Williams, Calcium: The Developing Role of Its Chemistry in Biological
Evolution, Oxford University Press, Oxford, UK, 1999, pp. 3–27.
[337] H. Laborit, R. Favre, R. Guittard, G. Laborit, F. Buchard, Anesth. Anal. 12 (1955)
593.
[338] L. Ambard, F. Schmid, C. R. Seances Soc. Biol. Fil. 98 (1928) 1220–1222.
[339] K.E. Hovda, C. Guo, R. Austin, K.E. McMartin, Toxicol. Lett. 192 (2010) 365–372.
[340] J.G. Zhang, W.E. Lindup, Toxicol. In Vitro 10 (1996) 205–209.
[341] C.S. Bailey, J.J. Weiner, M. Gibby, M.D. Penney, Ann. Clin. Biochem. 45 (2008)
527–529.
[342] V.V. Nasolodin, Vopr. Pitan. 4 (1985) 3–6.
[343] V.S. Artamonov, Pediatr. Akus. Ginekol. 5 (1968) 43–46.
[344] M.A. Lebda, M.S. El-Neweshy, Y.S. El-Sayed, Neurotoxicology 33 (2012) 98–104.
[345] Y. El-Sayed, M. Lebda, M. El-Neweshy, Toxicol. Lett. 211 (2012) S145.
[346] Z. Elbowicz-Waniewska, Pol. Tyg. Lek. (Warsaw, Poland: 1960) 23 (1968) 882–884.
[347] Harvard Women’s Health Watch 10 (2003) 6–7.
[348] R. Carrel, E.D. Mackowiak, A.J. Chialastri, W.H. Binns Jr., Pa. Dent. J. (Harrisb.) 45
(1978) 14–16.
[349] W.E.C. Wacker, S.J. Adelstein, D.D. Ulmer, B.L. Vallee, J. Clin. Invest. 35 (1956)
741–742.
[350] S.J. Adelstein, T.L. Coombs, B.L. Vallee, N. Engl. J. Med. 255 (1956) 105–109.
[351] I. Bremner, R.K. Mehra, J.N. Morrison, A.M. Wood, Biochem. J 235 (1986) 735–739.
[352] A.B. Abdelmageed, F.W. Oehme, Vet. Hum. Toxicol. 32 (1990) 230–234.
[353] N. Leone, D. Courbon, P. Ducimetiere, M. Zureik, Epidemiology 17 (2006)
308–314.
[354] A.B. Abdelmageed, F.W. Oehme, Vet. Hum. Toxicol. 32 (1990) 34–39.
[355] M.S. Salga, H.M. Ali, M.A. Abdulla, S.I. Abdelwahab, Int. J. Mol. Sci. 13 (2012)
1393–1404.
[356] H. Waizy, J.M. Seitz, J. Reifenrath, A. Weizbauer, F.W. Bach, A. Meyer-Lindenberg,
B. Denkena, H. Windhagen, J. Mater. Sci. 48 (2013) 39–50.
[357] Y.F. Ding, C. Wen, P. Hodgson, Y. Li, J. Mater. Chem. B 2 (2014) 1912–1933.
[358] X. Gu, Y. Zheng, Y. Cheng, S. Zhong, T. Xi, Biomaterials 30 (2009) 484–498.
[359] Z. Li, X. Gu, S. Lou, Y. Zheng, Biomaterials 29 (2008) 1329–1344.
[360] D. Yin, E. Zhang, S. Zeng, China Foundry 6 (2009) 43–47.
[361] A. Drynda, T. Hassel, R. Hoehn, A. Perz, F.-W. Bach, M. Peuster, J. Biomed. Mater.
Res. A 93A (2010) 763–775.
[362] Z. Trojanova, P. Lukac, P.P. PalVal, L.N. PalVal, Characterisation of Mg–Ca alloys
by mechanical spectroscopy, in: Proceedings of 5th European Conference on
Advanced Materials and Processes and Applications (EUROMAT 97), vol. 1,
Maastricht, Netherlands, April 21–23, Netherlands Soc Materials Science, Zwijn-
drecht, Netherlands, 1997, pp. 265–268.
[363] J.F. Nie, B.C. Muddle, Scr. Mater. 37 (1997) 1475–1481.
[364] Z. Zhang, Q. Le, H. Cui, Rare Metal Mater. Eng. 37 (2008) 436–439.
[365] T. Hassel, F.W. Bach, C. Krause, Influence of alloy composition on the mechanical
and electrochemical properties of binary Mg–Ca alloys and its corrosion behavior
in solutions at different chloride concentrations, in: K.U. Kainer (Ed.), Proceedings
of the 7th International Conference on Magnesium Alloys and Their Applications,
Wiley-VCH Verlag GmbH & Co, Hoboken, NJ, USA, 2007, pp. 789–795.
[366] K. Tsuchiya, K. Marukawa, J. Phys. IV 5 (1995) 901–905.
[367] M.H. Elahinia, M. Hashemi, M. Tabesh, S.B. Bhaduri, Prog. Mater. Sci. 57 (2012)
911–946.
[368] F.E. Wang, W.J. Buehler, S.J. Pickart, J. Appl. Phys. 36 (1965) 3232.
[369] S.M. Motzkin, V.L. Bonawit, W. Szablowski, L.S. Castleman, F.P. Alicandri, A.A.
Johnson, J. Dent. Res. 55 (1976) B291.
[370] D.E. Cutright, S.N. Bhaskar, B. Perez, R.M. Johnson, G.S.M. Cowan, Oral Surg. Oral
Med. Oral Pathol. Oral Radiol. Endod. 35 (1973) 578–584.
[371] K. Matsumoto, N. Tajima, S. Kuwahara, Nippon Seikeigeka Gakkai Zasshi 67
(1993) 267–274.
[372] K.M. Speck, A.C. Fraker, J. Dent. Res. 59 (1980) 1590–1595.
[373] N.K. Sarkar, W. Redmond, B. Schwaninger, A.J. Goldberg, J. Oral Rehabil. 10
(1983) 121–128.
[374] D.J. Wever, A.G. Veldhuizen, J. de Vries, H.J. Busscher, D.R.A. Uges, J.R. van Horn,
Biomaterials 19 (1998) 761–769.
[375] J.A. Platt, A. Guzman, A. Zuccari, D.W. Thornburg, B.F. Rhodes, Y. Oshida, B.K.
Moore, Am. J. Orthod. Dentofac. Orthop. 112 (1997) 69–79.
[376] J.W. Edie, G.F. Andreasen, M.P. Zaytoun, Angle Orthod. 51 (1981) 319–324.
[377] R.D. Barrett, S.E. Bishara, J.K. Quinn, Am. J. Orthod. Dentofac. Orthop. 103 (1993)
8–14.
[378] G. Rondelli, Biomaterials 17 (1996) 2003–2008.
[379] C. MonteroOcampo, H. Lopez, A.S. Rodriguez, J. Biomed. Mater. Res. 32 (1996)
583–591.
[380] L.S. Castleman, S.M. Motzkin, F.P. Alicandri, V.L. Bonawit, A.A. Johnson, J. Biomed.
Mater. Res. 10 (1976) 695–731.
[381] A.H. Cragg, S.C. Dejong, W.H. Barnhart, S.K. Landas, T.P. Smith, Radiology 189
(1993) 775–778.
[382] L.S. Castleman, S.M. Motzkin, The biocompatibility of Nitinol, in: D.F. Williams
(Ed.), Biocompatibility of Clinical Implant Materials, vol. 1, CRC Press, Inc., Boca
Raton, FL, 1981, pp. 129–154.
[383] K.W.K. Yeung, R.W.Y. Poon, X.Y. Liu, J.P.Y. Ho, C.Y. Chung, P.K. Chu, W.W. Lu, D.
Chan, K.M.C. Cheung, J. Biomed. Mater. Res. A 75A (2005) 256–267.
[384] K.W.K. Yeung, R.W.Y. Poon, X.Y. Liu, J.P.Y. Ho, C.Y. Chung, P.K. Chu, W.W. Lu, D.
Chan, K.M.C. Cheung, J. Biomed. Mater. Res. A 72A (2005) 238–245.
56 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
[385] K.W.K. Yeung, R.W.Y. Poon, X.M. Liu, P.K. Chu, C.Y. Chung, X.Y. Liu, S. Chan, W.W. Lu,
D. Chan, K.D.K. Luk, K.M.C. Cheung, Surf. Coat. Technol. 201 (2007) 5607–5612.
[386] A. Michiardi, E. Engel, C. Aparicio, J.A. Planell, F.J. Gil, J. Biomed. Mater. Res. A 85A
(2008) 108–114.
[387] J.L.M. Putters, D. Sukul, G.R. Dezeeuw, A. Bijma, P.A. Besselink, Eur. Surg. Res. 24
(1992) 378–382.
[388] M. Assad, S. Lombardi, S. Berneche, E.A. Desrosiers, L.H. Yahia, C.H. Rivard, Ann.
Chir. 48 (1994) 731–736.
[389] K. Endo, Dent. Mater. J. 14 (1995) 185–198.
[390] K. Endo, Dent. Mater. J. 14 (1995) 199–210.
[391] S.A. Shabalovskaya, Bio-Med. Mater. Eng. 6 (1996) 267–289.
[392] D.J. Wever, A.G. Veldhuizen, M.M. Sanders, J.M. Schakenraad, J.R. vanHorn,
Biomaterials 18 (1997) 1115–1120.
[393] M. Assad, L.H. Yahia, C.H. Rivard, N. Lemieux, J. Biomed. Mater. Res. 41 (1998)
154–161.
[394] E.C. Rose, I.E. Jonas, H.F. Kappert, J. Orofac. Orthop. 59 (1998) 253–264.
[395] F.L. Nie, Y.F. Zheng, Y. Cheng, S.C. Wei, R.Z. Valiev, Mater. Lett. 64 (2010)
983–986.
[396] D. Bogdanski, M. Koller, M. Bram, D. Stover, H.P. Buchkremer, J. Choi, M. Epple, G.
Muhr, Biomed. Tech. (Berl.) 47 (Suppl. 1, Pt. 1) (2002) 500–502.
[397] A. Kapanen, J. Ilvesaro, A. Danilov, J. Ryhanen, P. Lehenkari, J. Tuukkanen,
Biomaterials 23 (2002) 645–650.
[398] X.M. Liu, S.L. Wu, P.K. Chu, C.Y. Chung, C.L. Chu, Y.L. Chan, K.O. Lam, K.W.K.
Yeung, W.W. Lu, K.M.C. Cheung, K.D.K. Luk, J. Nanosci. Nanotechnol. 9 (2009)
3449–3454.
[399] X.J. Wen, N. Zhang, X.H. Li, Z.W. Cao, Bio-Med. Mater. Eng. 7 (1997) 1–11.
[400] P.J. Yang, J.C. Tao, M.Z. Ge, Q.M. Yang, H.B. Yang, Q. Sun, Chin. Med. J. (Engl.) 105
(1992) 312–315.
[401] S.J. Simske, R. Sachdeva, J. Biomed. Mater. Res. 29 (1995) 527–533.
[402] F. Kasano, T. Morimitsu, Auris Nasus Larynx 24 (1997) 137–142.
[403] M. BergerGorbet, B. Broxup, C. Rivard, L.H. Yahia, J. Biomed. Mater. Res. 32 (1996)
243–248.
[404] F. Takeshita, H. Takata, Y. Ayukawa, T. Suetsugu, Biomaterials 18 (1997) 21–25.
[405] C.Y. Li, X.J. Yang, L.Y. Zhang, M.F. Chen, Z.D. Cui, Mater. Sci. Eng. C 27 (2007)
122–126.
[406] S. Idelsohn, J. Pena, D. Lacroix, J.A. Planell, F.J. Gil, A. Arcas, J. Mater. Sci. Mater.
Med. 15 (2004) 541–546.
[407] X. Liu, S. Wu, K.W.K. Yeung, Y.L. Chan, T. Hu, Z. Xu, X. Liu, J.C.Y. Chung, K.M.C.
Cheung, P.K. Chu, Biomaterials 32 (2011) 330–338.
[408] B. Yuan, H. Li, Y. Gao, C. Chung, M. Zhu, Sci. China Technol. Sci. 55 (2012)
437–444.
[409] J. Ryhanen, S. Shabalovskaya, L.H. Yahia, Bioperformance of nitinol: in vivo
biocompatibility, in: Y.Y.Z.L.C. Chu (Ed.), Shape Memory Materials and Its
Applications, vols. 394–393, 2001, pp. 139–144.
[410] P.J. Yang, Y.F. Zhang, M.Z. Ge, T.D. Cai, J.C. Tao, H.P. Yang, Chin. Med. J. (Engl.) 100
(1987) 712–714.
[411] C.-L. Kuo, M.-C. Wang, A.-S. Shiao, J. Chin. Med. Assoc. 73 (2010) 241–247.
[412] K.R. Dai, Y.Y. Chu, Bio-Med. Mater. Eng. 6 (1996) 233–240.
[413] J. Drugacz, Z. Lekston, H. Morawiec, K. Januszewski, J. Oral Maxillofac. Surg. 53
(1995) 665–671.
[414] P.G. Sysolyatin, V.E. Gyunter, A.V. Starokha, I.A. Makarova, S.P. Sysolyatin, V.N.
Denisov, B.Q. Rahman, The use of Ni–Ti implants in maxillofacial surgery, in: A.R.
Pelton, D. Hodgson, T.W. Duerig (Eds.), Shape memory and superelastic tech-
nologies, proceedings of SMST-94, SMST, Pacific Grove, CA, 1994, pp. 470–475.
[415] A. Itro, V. Garau, G.P. Tartaro, G. Colella, Minerva Stomatol. 46 (1997) 381–389.
[416] G.F. Vonsalissoglio, Z. Orthop. Ihre Grenzgeb. 127 (1989) 191–196.
[417] O. Ricart, The use of memory shape staple in cervical anterior fusion, in: A.R.
Pelton, D. Hodgson, S.M. Russell, T.W. Duerig (Eds.), Shape Memory and Super-
elastic Technologies, Proceedings of SMST-97, SMST, Pacific Grove, CA, 1997, pp.
623–626.
[418] B. Silberstein, Subtotal and total vertebral body replacement and interbody
fusion with porous Ti–Ni implants, in: A.R. Pelton, D. Hodgson, S.M. Russell, T.W.
Duerig (Eds.), Shape Memory and Superelastic Technologies, Proceedings of
SMST-97, SMST, Pacific Grove, CA, 1997, pp. 617–621.
[419] R.G. Tang, K.R. Dai, Y.Q. Chen, Bio-Med. Mater. Eng. 6 (1996) 307–312.
[420] J. Musialek, P. Filip, J. Nieslanik, Arch. Orthop. Trauma Surg. 117 (1998) 341–344.
[421] D.I. Rabkin, S.M. Minkina, A.A. Kadnikov, B.P. Khasenov, Med. Radiol. (Mosk). 31
(1986) 55–63.
[422] C.S. Sutton, T. Oku, H. Harasaki, H.E. Kambic, Y. Sugita, S. Murabayashi, E. Shirey,
Y. Nose, Am. J. Roentgenol. 151 (1988) 597–601.
[423] A. Cragg, G. Lund, J. Rysavy, F. Castaneda, W. Castanedazuniga, K. Amplatz,
Radiology 147 (1983) 261–263.
[424] A.K. Wakhloo, F. Schellhammer, J. Devries, J. Haberstroh, M. Schumacher, Am. J.
Neuroradiol. 15 (1994) 493–502.
[425] D. Singh, S. Sinha, H. Singh, A. Jagetia, S. Gupta, P. Gangoo, M. Tandon, Minim.
Invasive Neurosurg. 54 (2011) 172–178.
[426] W. Cwikiel, R. Willen, H. Stridbeck, R. Lillogil, C.S. Vonholstein, Radiology 187
(1993) 667–671.
[427] W. Cwikiel, J. Harnek, E. Zoucas, U. Stenram, Acta Radiol. 38 (1997) 124–128.
[428] A.K. Wakhloo, F.O. Tio, B.B. Lieber, F. Schellhammer, M. Graf, L.N. Hopkins, Am. J.
Neuroradiol. 16 (1995) 1043–1051.
[429] A.J. Carter, T.A. Fischell, Int. J. Rad. Oncol. Biol. Phys. 41 (1998) 127–133.
[430] A.J. Carter, M. Foster, L.R. Bailey, J. DeVries, J. Piltz, K.T.T. Rao, E. McCarroll, R.
Jones, R. Virmani, Circulation 98 (1998) 189.
[431] A.J. Carter, D. Scott, J.R. Laird, L. Bailey, J.A. Kovach, T.G. Hoopes, K. Pierce, K. Heath,
K. Hess, A. Farb, R. Virmani, Cathet. Cardiovasc. Diagn. 44 (1998) 193–201.
[432] S. Sheth, F. Litvack, V. Dev, M.C. Fishbein, J.S. Forrester, N. Eigler, Circulation 94
(1996) 1733–1740.
[433] E. Grenadier, R. Shofti, M. Beyar, H. Lichtig, D. Mordechowitz, O. Globerman, W.
Markiewicz, R. Beyer, Am. Heart J. 128 (1994) 870–878.
[434] M.R. Prince, E.W. Salzman, F.J. Schoen, A.M. Palestrant, M. Simon, Invest. Radiol.
23 (1988) 294–300.
[435] D. Latal, J. Mraz, P. Zerhau, M. Susani, M. Marberger, Urol. Res. 22 (1994)
295–300.
[436] C.Y. Qiu, Zhonghua Wai Ke Za Zhi (Chin. J. Surg.), 31 (1993) 272–274.
[437] C.Y. Qiu, J.M. Wang, Z.X. Zhang, Z.X. Huang, Q.Y. Zheng, Z.Q. Du, M.J. Liu, C.L. Liu, J.
Endourol. 8 (1994) 65–67.
[438] G.S. Das, G. Voss, G. Jarvis, K. Wyche, R. Gunther, R.F. Wilson, Circulation 88
(1993) 1754–1764.
[439] H.R. Kim, W.Y. Lee, K.U. Jung, H.R. Yun, Y.B. Cho, S.H. Yun, H.C. Kim, H.K. Chun, J.
Laparoendosc. Adv. Surg. Tech. 22 (2012) 472–478.
[440] W. Kellner, G. Kuffer, T. Pfluger, F.T. Rosa, K. Hahn, Radiology 202 (1997)
327–331.
[441] D. Hayoz, D.D. Do, F. Mahler, J. Triller, F. Spertini, J. Endovasc. Surg. 4 (1997)
354–360.
[442] K. Schurmann, D. Vorwerk, A. Bucker, J. Neuerburg, B. Klosterhalfen, G. Muller, R.
Uppenkamp, R.W. Gunther, Radiology 204 (1997) 757–763.
[443] E. Rechavia, F. Litvack, M.C. Fishbien, M. Nakamura, N. Eigler, Cathet. Cardiovasc.
Diagn. 45 (1998) 202–207.
[444] L. Petrini, F. Migliavacca, J. Metall. 2011 (2011), Article ID 501483, 15 pages (e-
article).
[445] J.C. Stinville, P. Villechaise, C. Templier, J.P. Riviere, M. Drouet, Acta Mater. 58
(2010) 2814–2821.
[446] A. Biesiekierski, J. Wang, M.A.-H. Gepreel, C. Wen, Acta Biomater. 8 (2012)
1661–1669.
[447] Gf Andrease, T.B. Hilleman, J. Am. Dent. Assoc. 82 (1971) 1373.
[448] K.R. Dai, Zhonghua Wai Ke Za Zhi (Chin. J. Surg.]), 21 (1983) 343–345.
[449] C.T. Dotter, R.W. Buschmann, M.K. McKinney, J. Rosch, Radiology 147 (1983)
259–260.
[450] M. Simon, R. Kaplow, E. Salzman, D. Freiman, Radiology 125 (1977) 87–94.
[451] C. Ell, A. May, E.G. Hahn, Dtsch. Med. Wochenschr. 120 (1995) 1343–1348.
[452] B. Acunas, I. Rozanes, S. Akpinar, A. Tunaci, M. Tunaci, C. Acunas, Radiology 199
(1996) 648–652.
[453] M. Bezzi, F. Orsi, F.M. Salvatori, F. Maccioni, P. Rossi, J. Vasc. Interv. Radiol. 5
(1994) 287–293.
[454] P. Rossi, M. Bezzi, M. Rossi, A. Adam, N. Chetty, M.E. Roddie, V. Iacarl, W. Cwikiel,
C.L. Zolllkofer, F. Antonucci, L. Boguth, J. Vasc. Interv. Radiol. 5 (1994) 279–285.
[455] P. Rossi, F. Maccioni, F.M. Salvatori, M. Bezzi, R. Gandini, L. Broglia, D. Ingianna,
Radiol. Med. 87 (1994) 585–596.
[456] M. Smits, K. Huibregtse, G. Tytgat, Endoscopy 27 (1995) 505–508.
[457] J. Tack, A.M. Gevers, P. Rutgeerts, Gastrointest. Endosc. 48 (1998) 267–271.
[458] N.A. Lopatkin, A.V. Morozov, Urol. Nefrol. (1984) 3–7.
[459] N.A. Lopatkin, A.G. Martov, B.L. Gushchin, J. Endourol. 14 (2000) 721–726.
[460] H.W. Gottfried, H.P. Schlmers, J. Gschwend, E. Brandle, R.E. Hautmann, Arch. Esp.
Urol. 47 (1994) 933.
[461] H.W. Gottfried, H.P. Schimers, J. Gschwend, E. Brandle, R. Hautmann, Urol. Ausg.
A 34 (1995) 110–118.
[462] H.W. Gottfried, R. Gnann, E. Brandle, R. Bachor, J.E. Gschwend, K. Kleinschmidt,
Br. J. Urol. 80 (1997) 623–627.
[463] A. Gesenberg, R. Sintermann, J. Urol. 160 (1998) 72–76.
[464] D. Yachia, Ann. Urol. (Paris) 27 (1993) 245–252.
[465] D. Yachia, Z. Markovic, B. Markovic, V. Stojanovic, Acta Chir. Iugosl. 54 (2007)
105–114.
[466] K. Rauber, S. SyedAli, M. Hofmann, H.J. Terpe, G. Bachmann, W.S. Rau, Radiology
202 (1997) 281–283.
[467] K. Yanagihara, H. Mizuno, H. Wada, S. Hitomi, Ann. Thorac. Surg. 63 (1997)
1786–1789.
[468] R.W. Hauck, R.M. Lembeck, H.P. Emslander, A. Schomig, Chest 112 (1997)
134–144.
[469] J.M. Himpens, Surg. Endos. Ultrasound Interv. Tech. 7 (1993) 315–318.
[470] T. Frank, G.J. Willetts, F. Carter, A. Cuschieri, Proceedings of the Institution of
Mechanical Engineers. Part H, J. Eng. Med. 209 (1995) 111–115.
[471] F. Kasano, A. Naichu, T. Morimitsu, Utilization of Ni-Ti shape memory alloy for
ossicular prosthesis, Ear Res. Jpn. 20 (1) (1989) 117–118.
[472] J.O. Sanders, A.E. Sanders, R. More, R.B. Ashman, Spine 18 (1993) 1640–1646.
[473] S.B. Lu, J.F. Wang, J.F. Guo, Zhonghua Wai Ke Za Zhi (Chin. J. Surg.), 24 (1986)
129–187.
[474] Y. Wang, G. Zheng, X. Zhang, Y. Zhang, S. Xiao, Z. Wang, Eur. Spine J. 20 (2011)
118–122.
[475] F.R. Mei, X.J. Ren, W.D. Wang, Spine 22 (1997) 2083–2088.
[476] T. Iwabuchi, S. Suzuki, K. Ebina, T. Honma, J. Neurosurg. 42 (1975) 733–735.
[477] P.P. Kuo, P.J. Yang, Y.F. Zhang, H.B. Yang, Y.F. Yu, K.R. Dai, W.Q. Hong, M.Z. Ke, T.D.
Cai, J.C. Tao, Orthopedics 12 (1989) 111–116.
[478] P.K. Ritchie, E.C. McCarty, Oper. Tech. Sports Med. 12 (2004) 215–220.
[479] F.A. Barber, M.A. Herbert, J.N. Click, Arthroscopy 12 (1996) 32–38.
[480] F.A. Barber, O. Hapa, J.A. Bynum, Arthroscopy 26 (2010) S134–S141.
[481] J. Ryhanen, M. Kallioinen, J. Tuukkanen, J. Junila, E. Niemela, P. Sandvik, W. Serlo,
J. Biomed. Mater. Res. 41 (1998) 481–488.
[482] F. Baumgart, G. Bensmann, J. Haasters, A. Nolker, K.F. Schlegel, Arch. Orthop.
Trauma Surg. 91 (1978) 67–75.
[483] D.J. Wever, J.A. Elstrodt, A.G. Veldhuizen, J.R. van Horn, Eur. Spine J. 11 (2002)
100–106.
 Q. Chen, G.A. Thouas / Materials Science and Engineering R 87 (2015) 1–57
[484] K.N. Melton, O. Mercier, Acta Metall. 27 (1979) 137–144.
[485] S. Miyazaki, Y. Sugaya, K. Otsuka, Effects of various factors on fatigue life of Ti–
Ni–alloys, in: Proceedings MRS int. Meeting on ‘Advanced materials’, vol. 9, MRS,
Tokyo, Japan, 1988, pp. 251–256.
[486] J. Xiong, Y. Li, P.D. Hodgson, C.e. Wen, Design of a new biocompatible ti-based
shape memory alloy and its superelastic deformation behaviour, Mater. Sci.
Forum 654–656 (2010) 2087–2090.
[487] H. Kanetaka, H. Hosoda, Y. Shimizu, T.-a. Kudo, Y. Zhang, M. Kano, Y. Sano, S.
Miyazaki, Mater. Trans. 51 (2010) 1944–1950.
[488] U.O. Hafeli, M.C. Warburton, U. Landau, Biomaterials 19 (1998) 925–933.
[489] B. Nowak, J.M. Meyer, T. Goergen, D. Fluehs, S. Block, R.W. Guenther, H. Hoecker,
U. Buell, Cardiovasc. Radiat. Med. 2 (2001) 246–253.
[490] J. Black, Clin. Mater. 16 (1994) 167–173.
[491] B.S. Oppenheimer, E.T. Oppenheimer, I. Danishefsky, A.P. Stout, Cancer Res. 16
(1956) 439–441.
[492] R.A. Matthay, P.A. Balzer, C.E. Putman, J.B. Gee, G.J. Beck, R.H. Greenspan, Invest.
Radiol. 13 (1978) 514–518.
[493] S. Brumme, G. Lowicke, W. Knofler, Z. Exp. Chir. Transplant. Kunstliche Organe
22 (1989) 308–313.
[494] M. Khalid, Urol. Nefrol. 17 (1981) 49–52.
[495] T.V. Kalinina, O.R. Bogomolova, Vopr. Neirokhir. 23 (1959) 41–42.
[496] S. Bando, Kobe Ika Daigaku Kiyo 26 (1964) 98–112.
[497] T.V. Kalinina, Khirurgiia (Sofiia) 44 (1968) 97–100.
[498] M.F. Vyrzhikovskaia, Eksp. Khir. 4 (1959) 38–39.
[499] M.G. Anan’Ev, N.V. Antoshina, I.I. Gritsman, Eksp. Khir. 2 (1957) 28–35.
[500] A.H. Andrews Jr., M.J. Tamari, Ill. Med. J. 103 (1953) 175–178.
[501] R. Molicki, B. Klaptocz, M. Sypniewski, E. Wierzynski, Czas. Stomatol. 32 (1979)
595–599.
[502] I. Arcos, R. Davis, K. Fey, D. Mishler, D. Sanderson, C. Tanacs, M.J. Vogel, R. Wolf, Y.
Zilberman, J. Schulman, Artif. Organs 26 (2002) 228–231.
[503] V.K. Balla, S. Banerjee, S. Bose, A. Bandyopadhyay, Acta Biomater. 6 (2010)
2329–2334.
[504] S. Eriksen, B. Gillesberg, L.N. Langmaack, E. Christensen, H. Li, M. Lind, C. Bunger,
Tantalum coated carbon-carbon composite material for surgical implants, in:
Medical Device Materials II: Proceedings of the Materials & Processes for
Medical Devices Conference 2004, ASM International, MaterialsPark, OH,
2005, pp. 245–250.
[505] S. Dittrick, V.K. Balla, S. Bose, A. Bandyopadhyay, Mater. Sci. Eng. C 31 (2011)
1832–1835.
[506] P.S. Nebosky, S.R. Schmid, J. Manuf. Sci. Eng. Trans. ASME 133 (2011).
[507] P.S. Nebosky, S.R. Schmid, M.A. Selles, J. Manuf. Sci. Eng. 133 (6) (2011), 061022,
9 pages.
[508] D.A.J. Wilson, G. Richardson, A.W. Hennigar, M.J. Dunbar, Acta Orthop. 83 (2012)
36–40.
[509] A. Sternheim, D. Backstein, P.R.T. Kuzyk, G. Goshua, Y. Berkovich, O. Safir, A.E.
Gross, J. Bone Joint Surg. Am. 94B (2012) 158–162.
[510] J. Lu, X. Zheng, Z. Wang, Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 26 (2012)
244–247.
[511] N. Patil, K. Lee, S.B. Goodman, J. Biomed. Mater. Res. B Appl. Biomater. 89B (2009)
242–251.
[512] A.S.B. Martins, P.M.F.C. Fresco, M.J. Pereira, Fresenius Environ. Bull. 16 (2007)
869–874.
[513] J.L. Delongeas, D. Burnel, P. Netter, M. Grignon, J.M. Mur, R.J. Royer, G. Grignon, J.
Pharmacol. 14 (1983) 437–447.
[514] L.T. McClinton, J. Schubert, J. Pharmacol. Exp. Ther. 94 (1948) 1–6.
[515] P. Hernigou, G. Mathieu, A. Poignard, O. Manicom, P. Filippini, A. Demoura, Eur. J.
Orthop. Surg. Traumatol. 17 (2007) 243–246.
57
[516] R.B. Bourne, R. Barrack, C.H. Rorabeck, A. Salehi, V. Good, Clin. Orthop. 441 (2005)
159–167.
[517] R. Sonntag, J. Reinders, J.P. Kretzer, Acta Biomater. 8 (2012) 2434–2441.
[518] R. Bader, P. Bergschmidt, A. Fritsche, S. Ansorge, P. Thomas, W. Mittelmeier,
Orthopade 37 (2008) 136–142.
[519] B.M. Spector, M.D. Ries, R.B. Bourne, W.S. Sauer, M. Long, G. Hunter, J. Bone Joint
Surg. Am, 83-A (Suppl. 2, Pt. 2) (2001) 80–86.
[520] A. Dalal, V. Pawar, K. McAllister, C. Weaver, N.J. Hallab, J. Biomed. Mater. Res. A
100A (2012) 2147–2158.
[521] R.B. Raskin, New York State Dental J. 50 (1984) 587.
[522] J.W. Gorsuch, S.J. Klaine, Environ. Toxicol. Chem. 17 (1998) 537–538.
[523] D.J. Call, C.N. Polkinghorne, T.P. Markee, L.T. Brooke, D.L. Geiger, J.W. Gorsuch,
K.A. Robillard, Environ. Toxicol. Chem. 25 (2006) 1802–1808.
[524] S. Hemati, O. Asnaashari, M. Sarvizadeh, B.N. Motlagh, M. Akbari, M. Tajvidi, A.
Gookizadeh, Support. Care Cancer 20 (2012) 1613–1618.
[525] K.C. Chaw, M. Manimaran, F.E.H. Tay, Antimicrob. Agents Chemother. 49 (2005)
4853–4859.
[526] J.T. Seil, T.J. Webster, Int. J. Nanomed. 7 (2012) 2767–2781.
[527] A. Hambidge, Health Estate 55 (2001) 23–25.
[528] S. Cheruthazhekatt, M. Cernak, P. Slavicek, J. Havel, J. Appl. Biomed. 8 (2010) 55–66.
[529] Y. Qiu, N. Zhang, Y.H. An, X. Wen, Int. J. Artif. Organs 30 (2007) 828–841.
[530] Y. Okada, Y.J. Kuo, S. Hida, Y. Nishio, O. Yoshida, Y. Arai, Y. Kihara, T. Mishina,
Hinyokika Kiyo 33 (1987) 1640–1646.
[531] P.H. Rowe, M.G. Royle, J. R. Soc. Med. 76 (1983) 1019–1022.
[532] K.G. Watkins, W.M. Steen, I. Manna, D.F. Williams, S. Rhodes, P. Mazzoldi, S. Lo
Russo, M.G.S. Ferreira, J.T. Rito, T. Moura Silva, A.M.P. Simoes, A. Laser Inst, Enhanced
control of electrochemical response, in: Proceedings of The Laser Materials Proces-
sing Conference, Laser Inst America, Orlando, 1996, pp. A37–A46.
[533] S. Micera, P.N. Sergi, J. Carpaneto, L. Citi, S. Bossi, K.P. Koch, K.P. Hoffmann, A.
Menciassi, K. Yoshida, P. Dario, Experiments on the development and use of a
new generation of intra-neural electrodes to control robotic, Conf. Proc. IEEE
Eng. Med. Biol. Soc. 1 (2006) 1476–1479.
[534] S. Steltenkamp, K. Becher, T. Doerge, R. Ruff, K.-P. Hoffmann, Conf. Proc. IEEE Eng.
Med. Biol. Soc. 2009 (2009) 5478–5481.
[535] C. Ramasubbu, A. Flagg 2nd, K. Williams, Curr. Pain Headache Rep. 17 (2013)
315.
[536] P. Khong, A. Lazzaro, R. Mobbs, J. Clin. Neurosci. 17 (2010) 205–208.
[537] K.T. Ragnarsson, Spinal Cord 46 (2008) 255–274.
[538] L. Venkatraghavan, V. Chinnapa, P. Peng, R. Brull, Can. J. Anaest. J. Can. D Anesth.
56 (2009) 320–326.
[539] G.M. Yarkony, E.J. Roth, G.R. Cybulski, R.J. Jaeger, Arch. Phys. Med. Rehabil. 73
(1992) 195–200.
[540] I. Kostov, F. Filipov, Biotechnol. Biotechnol. Equip. 13 (1999) 81–90.
[541] R. Strang, C.J. Whitters, D. Brown, R.L. Clarke, R.V. Curtis, P.V. Hatton, A.J. Ireland,
C.H. Lloyd, J.F. McCabe, J.W. Nicholson, S.N. Scrimgeour, J.C. Setcos, M. Sherriff, R.
van Noort, D.C. Watts, D. Woods, J. Dent. 26 (1998) 273–291.
[542] R. Strang, C.J. Whitters, D. Brown, R.L. Clarke, R.V. Curtis, P.V. Hatton, A.J. Ireland,
C.H. Lloyd, J.F. McCabe, J.W. Nicholson, S.N. Scrimgeour, J.C. Setcos, M. Sherriff, R.
van Noort, D.C. Watts, D. Woods, J. Dent. 26 (1998) 191–207.
[543] R.P. Kusy, Angle Orthod. 72 (2002) 501–512.
[544] J. Breme, V. Wadewitz, Int. J. Oral Maxillofac. Implants 4 (1989) 113–118.
[545] M. Kaga, T. Aomori, K. Oikawa, M. Ota, S. Kondo, S. Ohkawa, Dent. Mater. J. 2
(1983) 161–168.
[546] Y. Ikarashi, K. Toyoda, E. Kobayashi, H. Doi, T. Yoneyama, H. Hamanaka, T.
Tsuchiya, Mater. Trans. 46 (2005) 2260–2267.
[547] Y. Ikarashi, K. Toyoda, E. Kobayashi, H. Doi, T. Yoneyama, H. Hamanaka, T.
T’Suchiya, J. Jpn. Inst. Metals 71 (2007) 395–401.