國立中山大學材料與光電科學學系 博士論文
Department of Materials and Optoelectronic Science National Sun Yat-sen University
Doctorate Dissertation
結晶或非晶之鋯基與鈦基金屬玻璃在模擬體液下之 電化學與生物相容性質分析
Electrochemical and biocompatibility response of amorphous or partially crystallized Zr/Ti-based metallic glasses
in simulated body fluid
研究生：黃朝先 撰
Andy Chao-Hsien Huang
指導教授﹕黃志青 博士
Dr. Jacob Chih-Ching Huang
中華民國 103年7月
July 2014
立中山大學 博士論文 結晶或非晶之鋯基與鈦基金屬玻璃在模擬體液下之電化學與生物相容性質分析 研究生：黃朝先
撰 學年度 102
國立中山大學材料與光電科學學系 博士論文
Department of Materials and Optoelectronic Science National Sun Yat-sen University
Doctorate Dissertation
結晶或非晶之鋯基與鈦基金屬玻璃在模擬體液下之 電化學與生物相容性質分析
Electrochemical and biocompatibility response of amorphous or partially crystallized Zr/Ti-based metallic glasses
in simulated body fluid
研究生：黃朝先 撰
Andy Chao-Hsien Huang
指導教授﹕黃志青 博士
Dr. Jacob Chih-Ching Huang
中華民國 103年7月
July 2014
謝誌
遙想四年前，從生命科學領域跨行到材料工程領域，心中充滿不安與期待，到現在 順利拿到博士學位，心中除了充滿歡喜也有無盡的感激。俗話說隔行如隔山，跨領域的 道路並不是平坦好走的，還好在這求學的過程中有許多人的幫助，若不是有這些貴人，
我想這求學的道路一定會更加困難崎嶇。因此在完成這本論文的同時，我想要對這些支 持過、幫助過以及愛我的所有人們表達我最誠摯的謝意：
首先，我要感謝我的家人們。爸爸總是忙碌認真的對待家人和工作，也常常在我迷 失方向或是墮落時，給予建議跟拉我一把。小時後在我眼中爸爸的背影是高大強壯的，
現在我的努力有了結果，在這未來我不只會好好照顧媽媽與弟妹也會好好為社會貢獻自 己的所學，當個讓你感到驕傲的寶貝兒子。媽媽從我出生開始就一直照顧我擔心我這個 抵抗力不好的兒子，小時後調皮也搞出了不少事情讓你擔憂，但是很感謝媽媽無論發生 任何事情都願意站在我身邊陪我面對，我最大願望就是帶你跟爸爸一起出國遊玩並讓你 們過著充實美滿快樂的退休生活。還有給我的二弟朝雋，謝謝你這麼早就出社會工作幫 忙家裡的開銷，若不是你，我沒辦法繼續走完這段求學路。給我的么妹菩淨，謝謝妳總 是以我這個不常回家而且忙碌的哥哥為榮。也感恩上天給我這麼好的家庭，讓我可以在 心煩、遇到困難時得到庇護，家真的是最好的避風港。
再來要感謝實驗室的大家長：我的指導教授黃志青老師，謝謝你願意給予我這個完 全對材料領域外行的學生指導，您甚至為我接了國家型奈米計畫，也讓實驗室得以擴展 生物醫學材料的領域。在跨領域的研究路上，遇到了許多的風風雨雨，但是老師您總是 不厭其煩的給予我建議以及方向，在 paper 的發表上也常常幫我做修飾以及更改使得發
表上更為順利。在升博士二年級時，家裡出了一些狀況，真的很感謝老師在那時候給予 家裡的幫忙與協助。老師在我心目中不只是一位傑出優秀的學者，更像是我的第二個爸 爸，謝謝您！另外，中山大學機械工程學系的林哲信老師，謝謝您一直以來提供的實驗 資源以及在電化學領域上的指導，高雄醫學大學骨科學研究中心的陳崇桓醫師，謝謝您 在生物體內以及體外測試上給予了許多寶貴的知識傳授，美國田納西大學的聶台岡教授，
您在台上所展現的大師風範，讓我崇拜不已，也謝謝您在每次小組討論時給予我意想不
到的方向，更關心我的未來以及生活，Lawrence Livermore 國家實驗室熊烈銘教授，
感謝您給我在未來找博士後研究的路上給了很多有用資源與方法。也感謝儀器跟實驗室 管理員施淑媖小姐、王良珠小姐、李秀月小姐、以及許鎏先生，謝謝你們辛勞的維護貴 重儀器，我才能在這麼好的研究環境中成長茁壯，謝謝系辦朱惠敏小姐、陳秀玉小姐、
顏秀芳小姐，總是這麼辛苦的處理系上事務。
最後，感謝張育誠大神學長，感謝您在我博士班一入學就帶領並指導我這不成材的 學弟，李敬仁學長，感謝您在我研究難產時安慰我鼓勵我，實驗室同期的同窗好友維昭 以及盈翰，謝謝你們在我有困難的時候當我的垃圾桶聽我說話，與你們共築的回憶我這 一輩子不會忘記，另一個也是同期夥伴阿官，謝謝你在電子顯微鏡實驗上給我的幫助，
大衛學弟也謝謝你半年來願意收留我當室友，還有你的吐槽功力真的是一流的，傑瑞學 弟我想我再也找不到像你跟我電波這麼相近的人了吧(笑)，謝謝你總是帶給實驗室無比 的歡樂，跟我最親近的學弟 TY、Allen、Wayne，謝謝你們常常陪我日夜顛倒的待在實 驗室，子揚以及樹懶學弟，也謝謝你們常常幫忙我完成 XRD 的實驗，Nick 學弟也在我 感情受創時帶我去喝酒解愁，謝謝你!其他學弟妹小安迪、哲銘、Sunny、JR、Daco、Ricky、
涯中帶給我無限的歡樂。最後的最後，給我可愛的女朋友果汁，真的很感謝 iPhone 充 電器，更感謝上海 BMG 研討會把我們繫在一起，在我準備研討會英文演講以及博士論文 口試時，有妳陪在我身邊，我就有無比的勇氣，謝謝妳這樣愛著我也給我支持。
黃朝先 (Andy Chao-Hsien Huang) 于 國立中山大學 2014.07
中文摘要
本研究的主題主要分為四大部分，為了找出有潛力的生醫用金屬玻璃，一開始先以 電化學法研究鐵基、鎂基、鋯基及鈦基金屬玻璃。模擬體液漢克水溶液主要被用來當作 測量金屬玻璃腐蝕抗性的環境。簡便的循環伏安法被用來快速證明是否有電化學反應發 生。研究結果顯示鋯基與鈦基金屬玻璃在模擬體液中擁有較高的腐蝕抗性以及電化學穩
定性、良好的電化學穩定性以及相當低的細胞毒性，Ti65Si15Ta10Zr10 金屬玻璃在生物醫
學應用上有相當高的潛力。
第二，新穎的無毒元素鈦鋯基金屬玻璃，Ti42Zr40Si15Ta3與Ti40Zr40Si15Cu5的電化學 行為以及細胞毒性在本文中也被有系統性的探討。此兩金屬玻璃的電化學性質以及生物
相容性也被拿來與純鈦以及含銅量高的Ti45Cu35Zr20做比較。結果顯示含銅量較低之金
屬玻璃擁有較低的化學反應。在 MTT 分析中，純鈦、Ti42Zr40Si15Ta3、Ti40Zr40Si15Cu5
試片以及其定電位電化學反應後的水溶液皆無發現明顯的細胞毒性。然而 Ti45Cu35Zr20
卻展現出較差的細胞存活率。在一個月的紐西蘭大白兔活體植入實驗中可發現植入部位
其傷口復元狀況良好以及較低的 C-反應性蛋白指數，證明了鈦鋯基金屬玻璃擁有短期
上良好的生物相容性。從電化學測量、生物體外以及體內測試的結果共同確認了銅含量
低於5 %的鈦鋯基金屬玻璃相當適合生物醫學上的用途。
第三，不同銅含量的金屬玻璃如：不含銅之Ti45Zr40Si15、含銅之Ti40Zr40Si15-Cu5 以
及 Ti45Zr25-Cu30金屬玻璃之生物腐蝕反應之研究主要使用開路電位法、動電位極化法、
電化學阻抗圖譜以及MTT 分析來達成。銅元素對於鈦基金屬玻璃在於其當作生物植入
完全相反的兩種影響。銅元素會造成腐蝕電位往正極偏移，但是只是對於其氧化層是否 容易形成造成影響，並不是主要用來探討腐蝕抗性的主要參數。反之，銅元素的存在的
主要缺點是會使鈦基金屬玻璃產生局部點蝕並造成離子的釋出。因此不含銅Ti45Zr40Si15
以及含銅量較少的Ti40Zr40Si15-Cu5有較佳的表現。
最後，我們也鑑測了奈米晶效應對鋯基以及鈦鋯基金屬玻璃在人體體液下腐蝕行為 的影響。Zr53Cu30Ni9Al8 非晶片材以及 Ti42Zr40Si15Ta3非晶薄帶主要藉由 Tg點以上溫度
做不同時間的熱處理並順利在非晶基質中產生出不同結晶程度的 Zr2Cu 以及 β-Ti 奈米
晶相。由極化曲線測量法中可得知產生高反應性的Zr2Cu 以及 Zr2Ni 的奈米晶相之金屬
玻璃擁有較差的腐蝕抗性，這是因為奈米晶相產生嚴重的加凡尼腐蝕所致。比較上來說，
由於其優異的抗點蝕性能，擁有抗腐蝕β-Ti 奈米晶之金屬玻璃展現了較佳的腐蝕抗性。
關鍵字：金屬玻璃、模擬體液、生物相容性、奈米晶化、腐蝕抗性
Abstract
This research first presents the electrochemical investigations of Fe-, Mg-, Zr-, and Ti-based metallic glasses (MGs) for finding the potential MG-based bio-materials. The simulation body-fluid (SBF) Hanks solution is utilized for testing the corrosion resistance of MGs. In addition, a simple cyclic voltammetry method is used for rapid verification of the potential electrochemical responses. It is found that the Ti- and Zr-based MGs can sustain in the body-fluid, exhibiting the best corrosion resistance and electrochemical stability. The rapid screening process suggests that the Ti65Si15Ta10Zr10 metallic glass has high potential for biomedical applications due to its good electrochemical stability and very low cytotoxicity.
Secondly, the electrochemical behaviors and the cell toxicity of two newly developed TiZr-based MGs, Ti42Zr40Si15Ta3, Ti40Zr40Si15Cu5, with lower or without unfavorable elements are systematically investigated. The electrochemistry property and biocompatibility of these two MGs are also compared with the controlled sample of pure Ti and the MG with a higher Cu-content, Ti45Cu35Zr20. Results show that the MGs with a low Cu content exhibit low electrochemical response. Both the solid specimens and the mediums after the potential state test for pure Ti, Ti42Zr40Si15Ta3 and Ti40Zr40Si15Cu5 exhibit no significant cytotoxicity in the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) test, while the tested medium for Ti45Cu35Zr20 MG shows lower cell viability. The good healing condition and the low C-reactive protein (CRP) index for the implanted New Zealand rabbits in one-month in vivo test also show the satisfactory short-term biocompatibility of the TiZr-based MGs. The electrochemical measurements, in vitro and in vivo experiments confirmed that the developed TiZr-based MGs with lower Cu content (≦ 5%) are promising
Thirdly, the bio-corrosion response of the Cu-free Ti45Zr40Si15 and Cu-containing Ti40Zr40Si15-Cu5 and Ti45Zr25-Cu30 MGs are explored, in terms of open circuit potential, potentiodynamic polarization, electrochemical impedance, as well as cytotoxicity MTT testing. The role of Cu in the Ti-based MGs, tentatively applied for bio-implant, is established and modeled. The presence of nobler Cu will impose two opposite effects. Since the minor positive shift of Ecorr for forming oxide layers is not of a major issue, the negative effect on local pitting and ion release would cause major drawback. The Cu-free Ti45Zr40Si15 and minor-Cu Ti40Zr40Si15-Cu5 metallic glasses exhibits promising performance.
Finally, we examine the nanocrysalline effect on the corrosion behavior of the Zr- and TiZr-based MGs in SBF. The Zr53Cu30Ni9Al8 and Ti42Zr40Si15Ta3 metallic glasses were annealed at temperatures above the glass transition temperature, Tg, with different time periods under the protective argon atmosphere to result in MGs with different degrees of crystalline Zr2Cu and β-Ti nano-phases in the amorphous matrix. Because of the serious galvanic corrosion, the polarization measurements show lower corrosion resistance for the nanocrystallized MGs with reactive Zr2Cu phases. In comparison, the nanocrystallized MGs with corrosion resistant β-Ti phases exhibited more promising corrosion resistance, due to the superior pitting resistance.
Keywords: metallic glasses, simulated body fluid, biocompatibility, nanocrystallized, corrosion resistance
Table of Content
中文摘要……...………...v
Abstract………...…………..………..vii
Table of content ... ix
List of Tables ...xiii
List of Figures ... xv
Chapter 1 Introduction ... 1
1.1 Amorphous metallic alloys ... 1
1.2 Evolution of Zr/Ti-based amorphous metallic glasses ... 2
1.3 Motivation and aim of this work ... 3
Chapter 2 Background and Literature Review ... 7
2.1 The characterization of amorphous metallic alloys ... 7
2.1.1 Mechanical properties ... 7
2.1.2 Magnetic properties ... 8
2.1.3 Corrosion resistance ... 8
2.2 Empirical rules for synthesis of amorphous metallic alloys ... 9
2.3 Fabrication of amorphous metallic alloys ... 10
2.4 The parameters of glass forming ability (GFA) ... 12
2.5 Introduction of corrosion and biocompatibility ... 14
2.6 Traditional metallic alloys for load-bearing bio-implant application ... 16
2.7 Metallic glasses for bio-implant load-bearing applications ... 18
2.7.1 Protein adhesion and cell growth on the surface of bulk metallic glasses ... 19
2.9 Nanoindentation for mechanical properties ... 22
2.10 Electrochemical response ... 25
2.10.1 Cyclic voltammerty (CV) ... 25
2.10.2 Amperometry ... 27
2.10.3 Polarization measurement ... 27
2.10.4 Electrochemical impedance spectroscopy (EIS) ... 29
2.11 X-ray photoelectron spectroscopy (XPS) ... 30
Chapter 3 Experimental Procedures ... 32
3.1 The preparation of amorphous metallic alloy ribbons ... 33
3.1.1 Raw Materials ... 33
3.1.2 Melt spinning technique ... 34
3.1.3 Heat treatment of amorphous metallic ribbons ... 35
3.2 Microstructure and phase identification ... 35
3.2.1 X-ray diffraction (XRD) and Nanoindentation ... 35
3.2.2 Optical microscopy (OM) observations ... 36
3.2.3 Scanning electron microscopy (SEM) observations ... 36
3.2.4 Transmission electron microscopy (TEM) observations ... 36
3.3 Thermal analysis ... 37
3.4 Immersion test under SBF ... 38
3.5 Electrochemical analysis ... 38
3.6 Cell viability test ... 39
3.7 In vivo test ... 40
Chapter 4 Results and Discussion ... 42
4.1 Simulated body-fluid tests and electrochemical investigations on Fe-, Mg-, Zr-based metallic glasses ... 42
4.1.1 Structural characterization and mechanical properties ... 42
4.1.2 Short-term immersion test under SBF ... 43
4.1.3 Electrochemical activity evaluation ... 44
4.2 Rapid screening of various Zr- and Ti-based metallic glasses for biomedical application ... 45
4.2.1 Structural characterization and mechanical properties ... 45
4.2.2 Electrochemical activity... 46
4.2.3 Pitting reaction on the Cu-free and Cu-containing metallic glasses ... 49
4.2.4 Cell viability on the Cu-free and Cu-containing metallic glasses... 51
4.3 Electrochemical and biocompatibility response of Cu-free and low Cu-containing TiZr-based metallic glasses ... 52
4.3.1 Structural characterization and mechanical properties ... 53
4.3.2 Electrochemical activity... 54
4.3.3 Cell viability ... 57
4.3.4 In vivo test ... 58
4.4 Cu effects on electrochemical response of Ti-based metallic glasses under simulated body fluid ... 60
4.4.1 Structural characterization and mechanical properties ... 60
4.4.2 Electrochemical response... 61
4.4.3 Surface characterization and pit morphology ... 65
4.4.4 Cytotoxicity test ... 69
4.5 Simulated body fluid electrochemical response of Zr-based and TiZr-based metallic glasses with different degrees of crystallization ... 70
4.5.1 Structural characterization and thermal analysis ... 70
Chapter 5 Summary and Conclusions ... 80 Chapter 6 Prospective and future works………...……….. .. 85 References……….. ... 86 Tables………96-105 Figures……….106-176
List of Tables
Table 1.1 Fundamental characteristics and application fields of amorphous metallic alloys.
... 96 Table 4.1 The glass transition (Tg), crystallization temperatures (Tx), and supercooled
region (ΔTx) of Mg-based, Fe-based, and Zr-based amorphous metallic alloys.
... 97 Table 4.2 Representative thermal properties, in termed of glass transition temperature (Tg), crystallization temperature (Tx), and supercooled region (ΔTx) of Zr-based metallic glasses, as well as the mechanical properties, in terms of elastic modulus E and nano-hardness H. The deviations of all the data are less than 10%.. ... 98 Table 4.3 Representative thermal properties, in termed of glass transition temperature (Tg),
crystallization temperature (Tx), and supercooled region (ΔTx) of pure Ti and Ti-based metallic glasses, as well as the mechanical properties, in terms of elastic modulus E and nano-hardness H. The deviations of all the data are less than 10%. ... 99 Table 4.4 EDS measured atomic compositions for the MGs, the oxygen contents in the
corroded region, and the weight loss after the potential state test. ... 100 Table 4.5 The corrosion properties of the metal speciments in the Hank’s solution. Noted that three inidvidual repeating tests were done for this test. Within our measured regime, pure Ti do not show pitting, thereby no Epit can be measured. The Ipass reading for Ti45Cu35Zr20 is rather low, about 0.4 A/cm2, but this value is meaningless since the just formed oxide layer would immediately be subject to
Table 4.6 Measured concentration variation (ppm) for the target ions inthe mediums after the potential state tests under Hank’s solution for Ti-based and TiZr-based metallic glasses. ... 102 Table 4.7 Bio-corrosion properties of the Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and Ti45Zr25-Cu30
metallic glasses, in comparison with commercial purity (cp) Ti, in the Hank's solution. ... 103 Table 4.8 Bio-corrosion properties of the as-cast and annealed Zr53Cu30Ni9Al8 metallic
glasses in Hank's solution. All specimens are tested two to three times to ensure reproducibility. ... 104 Table 4.9 Bio-corrosion properties of the as-cast and annealed Ti42Zr40Si15Ta3 metallic
glasses in Hank's solution. All specimens are tested two to three times to ensure reproducibility. ... 105
List of Figures
Figure 1.1 A scheme of (a) long range ordered structure crystalline metals and (b) short
range ordered structure of amorphous metallic alloys. ... 106
Figure 1.2 (Left) The short range ordering of the local atom clusters in different forms. (Right) The packing of local clusters into an amorphous material. ... 107
Figure 2.1 X-ray diffraction pattern of amorphous and crystalline materials. ... 108
Figure 2.2 The phenomenon of amorphous metallic alloy under an applied stress. ... 109
Figure 2.3 Young’s modulus vs. yield strength data for amorphous metals [x] and ductile-phas reinforced amorphous metals [+], shown together with data for stainless steels (green), Co-Cr-based (purple), and Ti-based alloys (blue).. .... 109
Figure 2.4 Vickers hardness vs. yield strength data for amorphous metals [x], shown together with data for stainless steels (green), Co-Cr-based (purple), and Ti-based alloys (green).. ... 110
Figure 2.5 Schematic diagram of (a) sputtering and (b) vacuum evaporation. ... 111
Figure 2.6 Schematic diagram of splat quench method.. ... 112
Figure 2.7 Schematic diagram of two roller quench method.. ... 112
Figure 2.8 Schematic diagram of chill block melting spinning.. ... 113
Figure 2.9 Schematic diagram of planar flow casting.. ... 113
Figure 2.10 Relationship between the critical cooling rate (Rc), maximum sample thickness (tmax) and reduced glass transition temperature (Tg/Tm) for bulk amorphous alloy system. ... 114
Figure 2.11 Relationship between the critical cooling rate (Rc), maximum sample thickness and supercooled liquid range ΔT (= T – T ) for bulk amorphous alloys.. ... 115
Figure 2.13 Various types of corrosion for metallic alloys.. ... 117 Figure 2.14 The application fields of metal and ceramic biomaterials. ... 118 Figure 2.15 BAE cell attachment and growth on metallic glass. (a) Shiny side plus FCS, at 6 h; (b) shiny side plus FCS, at 4 days; (c) shiny side minus FCS, at 6 h; (d) dull side plus FCS, at 4 days; (e) dull side after crystalline conversion, plus FCS at 6h. ... 119 Figure 2.16 Fibroblast cell attachment on Zr-Ti-Co-Be and high-density polyethylene
(HDPE) discs. (a) Micrograph of the amorphous metal surface after 7 days, (arrows point to the cell-layer buildup at the interface.) (b) Cell proliferation on the amorphous metal and HDPE discs. ... 120 Figure 2.17 Micrographs of the tissue surrounding a Pd-Ag-P-Si rod (circular region in the center of the image) implanted intramuscularly in rat for 28 day: (a) 40x magnification, (b) 100x magnifications... 121 Figure 2.18 Anodic and cathodic polarization curves of the Ti40Zr10Cu36Pd14 bulk metallic glass and its crystalline alloys at 310 K in Hanks’ solution.. ... 122 Figure 2.19 Median pitting and repassivation potentials for Al90Fe5Gd5, Al87Ni7Gd6 and pure polycrystalline and single-crystal Al in deaerated 0.6 M NaCl. Error bars represent the 25th and 75th percentile of the data when expressed as the cumulative probability of obtaining the given pitting potential. Scan rate is 0.1 mV/s... 123 Figure 2.20 Potentiodynamic polarization curves of the (a) as-spun and heat-treated samples
of Al88Ni6La6 and (b) as-spun and heat-treated Al86Ni9La5 in 0.01 M NaCl alkaline solution.. ... 124 Figure 2.21 Mechanisms of the electron transfer reaction. (A) An oxidation process of
species A (A → A+ + e−). (B) An reduction process of species B (B + e− →
B−). ... 125
Figure 2.22 (a) Cyclic voltammerty waveform and (b) standard cyclic voltammetry of oxidation-reduction reaction... ... 126
Figure 2.23 Standard cyclic anodic polarization curves.. ... 127
Figure 2.24 A simple electrified interface. The oxidants (red) with a positive charge diffuse toward the negatively charged electrode, accept electrons from the electrode at the interface, become the reductants (blue), and diffuse to the bulk of the solution. Furthermore, IHP and OHP are the inner and outer Helmholtz planes, respectively. ... 128
Figure 2.25 (a) presents a standard equivalent circuit model of a double layer formed by applying a negative potential on the surface of the electrode. (b) shows classical Nyquist plot, the start point of the high frequency region is the Rs and the end point of the low frequency region is the Rs + Rp. ... 129
Figure 2.26 PES as a three-step process: (1) photoexcitation of electrons; (2) travel to the surface with concomitant production of secondaries (shaded); (3) penetration through the surface (barrier) and escape into the vacuum... ... 130
Figure 3.1 The experimental flow chart. ... 131
Figure 3.2 The illustration of a single-roller melt spinning process. ... 132
Figure 3.3 The picture of Perkin Diamond DSC. ... 133
Figure 3.4 The schematic diagram of power compensation DSC... 133
Figure 3.5 The standard Nano Indenter XP is a complete, turnkey system consisting of the major components illustrated... 134 Figure 3.6 Preparation of a cross-section specimen by the liftout technique, (a) deposition of a platinum strap over the region of interest, (b) cutting of the staircase cuts, (c)
a sample by 45° and cutting of its base, (e) further thinning of a cross-section specimen until it is about 70 nm thick, and (f) cutting of the edges of a cross-section specimen to free it from the substrate... ... 135 Figure 3.7 The picture of CHI 614D electrochemical work station. ... 136 Figure 3.8 Schematic diagram of electrochemical workstation in a three electrodes cell...
... 137 Figure 4.1 The (a) XRD patterns and (b) DSC scans of Mg65Cu25Gd10, Mg67Cu25Y8, Zr61Cu17.5Ni10Al7.5Si4, and Fe70B20Si10 amorphous metallic alloys. ... 138 Figure 4.2 The bio-corrosion response for the four MGs immersed in Hank’s solution for 24 h under two pH levels of 2.0 and 6.5: (a) Mg65Cu25Gd10, (b) Mg67Cu25Y8, (c) Zr61Cu17.5Ni10Al7.5Si4, and (d) Fe70B20Si10. ... 139 Figure 4.3 Variations of the pH values as a function of immersion time for the Hank’s
solution itself, as well as the Hank’s solution immersed with the four MGs. Only the variation for initial pH=6.5 is shown. ... 140 Figure 4.4 The bio-corrosion response for the four MGs immersed in Hank’s solution for 24 h under two pH levels of 2.0 and 6.5: (a) Mg65Cu25Gd10, (b) Mg67Cu25Y8, (c) Zr61Cu17.5Ni10Al7.5Si4, and (d) Fe70B20Si10. ... 141 Figure 4.5 (a) XRD scans and (b) DSC scans for the MGs under study. The XRD peak in (a) for the Ti65Si15Ta10Zr10 MG is referred to the {110} planes of the minor crystalline Ta phase embedded in the amorphous matrix. The Ta57Zr23Cu12Ti8
thin film metallic glass is too thin for DSC measurement. But from the smooth XRD hump in (a) it can be ensured that this MG is fully amorphous. ... 142 Figure 4.6 The comparison of cyclic voltammogram responses for pure Ti, Ta57Zr23Cu12Ti8, Ti65Si15Ta10Zr10, Ti40Cu36Pd14Zr10, Ti45Cu35Zr20, Zr61Cu17.5Ni10Al7.5Si4, Zr53Cu30Ni9Al8 and Zr53Cu30Al8Pd4.5Nb4.5 MGs.. ... 143
Figure 4.7 The comparison of cyclic voltammograms for Ta57Zr23Cu12Ti8 and Ti65Si15Ta10Zr10 MGs and pure Ti in Hank's solution. ... 144 Figure 4.8 (a) The measured i–t curves for various MG and the control group of Ti, with the applied voltage of 80 mV. (b) The close-up i–t curves for the three samples with low current response. ... 145 Figure 4.9 The SEM images for the MGs after the potential state test. ... 146 Figure 4.10 The EDS results for inspecting the composition labeled in red square in Fig. 6.
(a) Ti40Cu36Pd14Zr10, (b) Ti45Cu35Zr20, (c) Zr61Cu17.5Ni10Al7.5Si4, (d) Zr53Cu30Ni9Al8 (e) Zr53Cu30Al8Pd4.5Nb4.5 and (f) Ti65Si15Ta10Zr10. (1:
composition before potential state test, 2: composition after test.). ... 147 Figure 4.11 The results of cell viability tests for pure Ti and various MGs cultured for 72 h, and the medium after the potential state test (24 h culture). The viability of the control pure Ti is set to be 100% as a reference. (a) pure Ti, (b) Ti65Si15Ta10Zr10, (c) Ti40Cu36Pd14Zr10, (d) Ti45Cu35Zr20, (e) Zr61Cu17.5Ni10Al7.5Si4, (f) Zr53Cu30Ni9Al8, (g) Zr53Cu30Al8Pd4.5Nb4.5. ... 148 Figure 4.12 (a) XRD patterns and (b) DSC scans of TiZr-based and Ti-based metallic
glasses. ... 149 Figure 4.13 The comparison of the cyclic voltammogram responses for Ti45Cu35Zr20 and
Ti42Zr40Si15Ta3, Ti40Zr40Si15Cu5, and pure Ti (inset) in Hank’s solution. Note that the current response for Ti45Cu35Zr20 is around 1000-fold greater than the other samples.. ... 150 Figure 4.14 The comparison of thei-t curves for TiZr-based metallic glasses and pure Ti
with the applied low-voltage (80 mV) in the Hank’s solution.. ... 151 Figure 4.15 The potential polarization curves of TiZr-based and Ti-based metallic glasses
Figure 4.16 The comparison of MTT tests for (a) pure Ti, (b) Ti42Zr40Si15Ta3, (c) Ti40Zr40Si15Cu5, and (d) Ti45Cu35Zr20 with culturing for 72 hours and the medium with culturing for 24 hours after the potential state test. The control pure Ti is defined to be 100 % viability for reference.. ... 153 Figure 4.17 (a) and (b) Photo images of thesurgical opertion for the implantation (left) and the corresponding X-ray image (right) after the operttions. Note that the MGs were placeat theepiphyseal growth plate of the right tibia. The implantation sites for the corresponding MG are marked.. ... 154 Figure 4.18 The 2D (a-c) and 3D (a’-c’) micro-CT images of the three MGs at theepiphyseal growth platesafter one month of implantation.(a)(a’) Ti42Zr40Si15Ta3, (b)(b’) Ti40Zr40Si15-Cu5, and (c)(c’) Ti45Cu35Zr20. The dotted circles indicate the corresponding implantation sites of each TiZr-based and Ti-based metallic glass.
... 155 Figure 4.19 (a) XRD and (b) DSC patterns of the Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and
Ti45Zr25-Cu30 metallic glasses.. ... 156 Figure 4.20 Open circuit potentials (OCP) curves of the Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and
Ti45Zr25-Cu30 metallic glasses in the Hank's solution. ... 157 Figure 4.21 Potential polarization Tafel curves of the Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and
Ti45Zr25-Cu30 metallic glasses in the Hank's solution.. ... 158 Figure 4.22 The Nyquist plot showing the EIS spectra of the Ti45Zr40Si15, Ti40Zr40Si15-Cu5
and Ti45Zr25-Cu30 metallic glasses in Hank's solution.. ... 159 Figure 4.23 Representative EDS element mapping near the pitted region in the Ti-based
metallic glasses immersed in Hank's solution. (a) SEM secondary electronic image showing a pitted region of Ti45Zr40Si15 on the right side, and the EDS mappings for Ti, Zr, Cl, and Si. (b) SEM secondary electronic image showing a
pitted region of Ti40Zr40Si15-Cu5, and the EDS mappings for Ti, Zr, Cl, Si, and Cu. Note that both Ti and Zr were depleted and Cl was enriched in the pitted region... ... 160 Figure 4.24 XPS spectrums for the surface analysis of Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and
Ti45Zr25-Cu30 after 7-day immersion in Hank’s solution: the (a) O 1s, (b) Ti 2p, (c) Zr 3d, (d) Si 2p, and (e) Cu 2p peak. ... 161 Figure 4.25 Schematic illustration of the pitting corrosion mechanisms of three Ti-based
metallic glasses in the Hank’s solution. The mechanism in (a) to (b) is for Cu-free Ti45Zr40Si15, and that in (a) to (e) is for Cu-containing Ti40Zr40Si15-Cu5
and Ti45Zr25-Cu30, respectively. (a) continuous passive film formation in the Hank’s solution, (b) local passive layer breakdown due to the galvanic corrosion, (c) the further dissolution of the Cu, (d) the precipitated CuCl formation on the surface of the Cu-containing Ti-based metallic glasses, and (e) pitting propagation.. ... 162 Figure 4.26 Comparison of the MTT tests for the Ti45Zr40Si15, Ti40Zr40Si15-Cu5 and
Ti45Zr25-Cu30 metallic glasses. The control pure DMEM is defined to be 100 % viability for reference. ... 163 Figure 4.27 DSC curve of (a) Zr53Cu30Ni9Al8 and (b) Ti42Zr40Si15Ta3 metallic glasses.. ... 164 Figure 4.28 (a) DSC curves and (b) XRD patterns of as-cast Zr53Cu30Ni9Al8 and its partial
crystalline alloys. ... 165 Figure 4.29 (a) DSC patterns of as-cast Ti42Zr40Si15Ta3 and its partial crystalline alloys. (b) XRD curves of as-cast Ti42Zr40Si15Ta3 and its partial crystalline alloys. ... .166 Figure 4.30 (a) Bright field TEM micrograph, (b) Dark field TEM micrograph, and (c)
corresponding diffraction pattern of Ti42Zr40Si15Ta3 after 10 min annealing. (d)
Ti42Zr40Si15Ta3 glassy matrix. ... 167 Figure 4.31 Open circuit potentials as a function of immersion time of the as-cast and
annealed Zr53Cu30Ni9Al8 metallic glasses in Hank's solution. ... 168 Figure 4.32 Open circuit potentials as a function of immersion time of the as-cast and
annealed Ti42Zr40Si15Ta3 metallic glasses in Hank's solution. ... 169 Figure 4.33 Potential polarization curves of the as-cast and annealed Zr53Cu30Ni9Al8
metallic glasses in Hank's solution.. ... 170 Figure 4.34 Potential polarization curves of the as-cast and annealed Ti42Zr40Si15Ta3 metallic glasses in Hank's solution.. ... 171 Figure 4.35 SEM images of the pitting morphology for the as-cast Zr53Cu30Ni9Al8 metallic glasses after polarization measurement in Hank’s solution... 172 Figure 4.36 The equivalent circuit model of the as-cast and annealed Zr53Cu30Ni9Al8 and
Ti42Zr40Si15Ta3 metallic glasses for fitting the curve of the Nyquist plot in Hank's solution. The Rs, Rp, and CPE are the solution resistance, polarization resistance, and constant phase element, respectively. ... 173 Figure 4.37 The Nyquist plot showing the EIS spectra of the as-cast and annealed
Zr53Cu30Ni9Al8 metallic glasses in Hank's solution.. ... 174 Figure 4.38 The Nyquist plot showing the EIS spectra of the as-cast and annealed
Ti42Zr40Si15Ta3 metallic glasses in Hank's solution. ... 175 Figure 4.39 The structure of passive layers form on the surface of the nanocrystallized
Zr53Cu30Ni9Al8 and Ti42Zr40Si15Ta3 metallic glasses via passive process after polarization measurements. ... 176
Chapter 1 Introduction
1.1 Amorphous metallic alloys
Amorphous metallic alloys are metallic materials that own disordered atomic structure and lack the periodicity of crystal. The short-range ordered structure of amorphous metallic alloys is different from the long-range ordered structure of crystalline metals, as shown in Figure 1.1 [1]. The short range ordering for the local atom clusters usually measures from 1-2 nm, consisting of 10-20 atoms (Figure 1.2). The tight-bonding local clusters make them one of the smallest nano materials. The relatively loose bonding between the short range clusters would lead to presence of excess free volume for special shear transition zone (STZ) and shear band (SB) formation at room temperature, and homogeneous or viscous flow at elevated temperatures. The near net shape viscous superplastic forming and the adhesion of tissue onto the metallic glass surface are both interesting and worth of extensive studies. Due to the absence of symmetry short-range ordered structure, which is similar to the glass, the amorphous metallic alloys are also called as liquid metals, non-crystalline metals, glassy metals or metallic glasses.
In contrast to crystalline materials with same composition, amorphous metallic alloys exist several unique physical and chemical properties, such as higher tensile strength, lower Young’s modulus, elastic elongation about 2% [2-6], larger elastic energy, higher corrosion resistance [7-13], outstanding electromagnetic properties [14-22]. Hence, there are various application fields of amorphous metallic alloys, which are summarized in Table 1.1 [23].
1.2 Evolution of Zr/Ti-based amorphous metallic glasses
Zr-based amorphous metallic alloys not only show good corrosion resistance [24] and above mechanical properties, but also higher bending flexural strength and bending fatigue strength [25]. In addition, formation of a large number Zr2(Cu,Pd) or (Zr,Ti)2Al nanocrystalline phases in Zr-Al-Cu-Pd and Zr-Al-Cu-Ni-Ti amorphous metallic alloys would improve the ductility and tensile strength [26-27]. Furthermore, the higher crystallization temperature (Tx) and supercooled liquid region (above 100 K) mean that the Zr-based amorphous alloy exist higher thermal stability and glass forming ability (GFA) [28].
In 1993, Inoue et al. developed a family of Zr-Al-Ni-Cu amorphous metallic alloys with a high GFA and thermal stability, and casted the 15 mm thickness Zr65Al7.5Ni10Cu17.5
amorphous metallic alloy [29]. At the same time, the Zr-Ti-Cu-Ni-Be family with excellent GFA was developed by Peker and Johnson. The cooling rate of Zr41.2Ti13.8Cu12.5Ni10Be22.5 is 1 K/s and casting diameter is several centimeters [30]. Unfortunately, the existence of toxic beryllium (Be) makes this system hard to use in the biomedical application. Hence, Lin et al.
investigated another family: Zr-Ti(Nb)-Cu-Ni-Al, including Zr52.5Ti5Cu17.9Ni14.6Al10 and Zr57Cu15.4Ni12.6Al10Nb5 [31]. These amorphous metallic alloys do not contain toxic beryllium and still exhibit good mechanical properties. Besides, the Zr-Al-Cu-based system, containing nanocrystalline phases, exhibits high tensile strength and good ductility [32].
Ti-based alloys are suitable for biomedical application due to their good biocompatibility and corrosion resistance. There are many Ti-based amorphous systems had been investigated, for example, Ti-Cu-Ni [33], Ti-Cu-Ni-Co [34], Ti-Cu-Ni-Zr [35] and Ti-Cu-Ni-Zr-Be [36]. Because the Ti- based amorphous metallic alloys with high GFA always
contain toxic Niobium (Nb) and Beryllium (Be), Inoue et al. investigated Ti-Zr-Cu-Pd and Ti-Zr-Ta-Si amorphous system without toxic elements in 2007 [37,38]. Both Ti-Zr-Cu-Pd and Ti-Zr-Ta-Si amorphous system exhibited higher strength, lower Young’s modulus and corrosion resistance than pure Ti and Ti-6Al-4V. Compare to the Ti-Zr-Cu-Pd, Ti-Zr-Ta-Si are more potential candidate for biomedical application due to lack the reactive copper (Cu).
Although nanocrystalline Zr/Ti-based amorphous metallic alloys show good mechanical properties, the study of electrochemical response and biocompatibility behavior in simulated body fluid (SBF) is still rare. Hence, it is necessary to pay more attention on the electrochemical response in metallic glasses under the conditions of fully amorphous state or partially amorphous state with a certain amount of nanocrystalline phases.
1.3 Motivation and aim of this work
Metallic materials have been used in a number of artificial parts of the human body such as the substitutes for hard tissue replacement, fracture healing aids and the fixation devices due to their excellent mechanical properties [39-42]. The implanted devices are suffering from corrosion since they are exposed to the surrounding body fluids that are typically with a high ionic strength. The concentrations of chloride, potassium and sodium ions are relatively high and may cause the simultaneous electrochemical reaction between the implanted metals and the surrounding fluids. Metal corrosion may release ions into the body fluids and induce allergy, inflammatory, diseases or cancer [43-45]. Over the metallic materials, Ti-based and Co-based alloys are the most promising metals for biomaterial applications due to their light weight, good corrosion resistance and the bio-inert property in the early stage [46,47].
weakening issues [48-50]. Nevertheless, the difference in the electrochemical potential of the multiple compositions in these alloys may cause more significant electrochemical corrosion [51,52].
Recently, another category in metal alloys for biomaterial applications, called metallic glass (MG), has attracted a number of researchers working in this field. MGs are amorphous since there is no crystal structural deficiency in the metallic glasses such as dislocation, twin, vacancy, or grain boundary. Metallic glasses have a homogenate composition, providing higher strength, hardness and elastic limit compared to typical alloys [53-58]. The lack of grain boundaries in this amorphous structure excludes the micro-structural defects in the materials so that the resistance against electrochemical corrosion is enhanced [59,60]. The risk for the necrosis and apoptosis of the tissues induced by severe corrosion-induced ion release can also be eliminated [61]. However, some MGs containing active elements such as Mg, Ca, and even Fe atoms may still suffer from electrochemical corrosion [57,58,62-64].
The MG compositions may influence the bio-stability and biocompatibility in biomedical applications.
In this study, simulation body-fluid immersion is used to test the corrosion resistance of different compositional systems of MGs (Mg-, Fe-, Zr-, Ti-, and TiZr-based). A novel electrochemical test by measuring the cyclic voltammetry response of the metallic glasses in the simulation body-fluid is developed. With this approach, analyzing the electrochemical response of the MGs in body can be achieved in a short time. The long-term electrochemical corrosion response of the MGs in human body-fluid can be predicted without long-term and continuous observation. Therefore, we can easily determine which compositional system of MGs is more suitable for biomedical uses.
Previous studies mentioned that some elements used to produce these MGs might have toxicity issues for implants. For example, an in vivo test reported by Laing et al. [65]
indicated that the Ni, Co, Cr, Fe, Mo, V, and Mn are considered as toxic elements due to the unfavorable tissue responses between the metallic implants and rabbit muscle. Elshahawy et al. [66] also evaluated various commercial biomedical alloys by the in vitro testing, the Cu2+, Ni2+, and Be2+ were identified as toxic ions in the fibroblast cell culture. Calin et al. [67] also summarized the potential harmful and non-toxic elements for biomedical implants.
This study develops the new Ti-Zr-Si amorphous systems with a lower Ta (Ti-Zr-Si-Ta) or a minor Cu content (Ti-Zr-Si-Cu), respectively. The newly developed MG systems will reduce the cost and the process temperature for producing the bio-implantable MGs. One positive control and one negative control, utilizing the pure Ti and another MG composed of higher Cu content, Ti45Cu35Zr20, are also prepared for the biocompatibility comparison with the currently developed TiZr-based MGs.
Most of metallic glasses, no matter Zr-, Mg- or other-based, small size Cu (with the atomic radius r~1.28 Å) is almost inevitably added to ensure the GFA. However, there have been numerous studies reporting that Cu would impose harmful impact to human body [68,69], causing cytotoxicity or killing cells. Although Cu has been viewed as a bio-unfriendly element to human body (not so serious as Ni or Be), the understanding of Cu effect on the bio-corrosion behavior of Ti-based metallic glasses is still very limited.
In this study, the electrochemical responses of Ti-based amorphous alloys with different
exploring the Cu effect. For the Cu-free Ti45Zr40Si15, another element also with a small atomic size, Si (r~1.1 Å), is substituted to maintain the sufficient GFA. The electrochemical behaviors and passive layer structure of these three Ti-based metallic glasses are studied in SBF for better systematic understanding Cu effect on the bio-corrosion behavior of Ti-based metallic glasses.
Recently, some studies started to point out the different corrosion behavior of partially amorphous and fully nanocrystalline alloys [70-76]. It appears that the formation of different quantities of nanocrystalline phases would affect the corrosion behavior of the amorphous matrix in various corrosive media. But it is still not certain whether the metallic glass alloys in fully amorphous or partially nanocrystalline state would exhibit better bio-corrosion behavior. This can be an important issue for MGs bio-implant application, since all MGs could accidentally induce various amounts of partially nanocrystalline phases within the amorphous matrix, either during fabrication or long-term service.
In this study, we examine how nanocrystals would affect the corrosion behavior of the Zr- and TiZr-based amorphous alloys under SBF. The Zr53Cu30Ni9Al8 plates and Ti42Zr40Si15Ta3 ribbons were annealed for the preparation of different degrees of partially crystalline Zr2Cu and β-Ti nano-phases in the amorphous matrix. In this work, the electrochemical responses of amorphous and nanocrystalline Zr-based and TiZr-based alloys with different crystallinities are investigated in the SBF for establishing the clear profile of their corrosion behavior.
Chapter 2 Background and Literature Review
2.1 The characterization of amorphous metallic alloys
Due to the short range disorder or order atomic arrangement of amorphous metallic alloys, the distance between atoms are not constant. Hence, the amorphous metallic alloys are homogeneous and isotropic. The X-ray diffraction pattern reveals boarding effects at 30° to 40°, which shows the different diffraction peak between amorphous and crystalline materials, as shown in Figure 2.1 [77]. Therefore, the special structure makes properties of amorphous materials different from the crystalline materials.
2.1.1 Mechanical properties
There are many superior mechanical properties of amorphous materials. When applied a stress on amorphous metals, the random atomic structure will make the atoms collide with each other, which restrict the atomic movement, as shown in Figure 2.2 [78]. Hence, due to the lack of dislocation mechanism, the amorphous metals are not easy to be deformed and possess higher strength. The Liaw et al. [59] compared several mechanical properties (Young’s modulus, hardness, toughness and fatigue) between biomedical alloys with metallic glasses. Young’s modulus is a very important mechanical factor for hard-tissue implantation, material with lower modulus can transfer the stress to the bone easily for reducing the stress-shielding effect. Figure 2.3 shows that metallic glasses would own the lower Young’s modulus than biomedical CoCr-based and Ti-based alloys, indicating the promised potential
materials which is defined as H/E, where H is hardness and E is modulus. Therefore, the material with higher hardness possesses desired wear resistance. Figure 2.4 demonstrated that the metallic glasses exhibit larger hardness than the biomedical CoCr-based and Ti-based alloys.
2.1.2 Magnetic properties
Permeability and coercive force are basic magnetic properties of materials. Due to interaction between the random arrangement atoms and heterogeneous magnetism, the amorphous metals possess superior magnetic properties [78]. Permeability means that materials can be magnetized and be demagnetized when it is close to and far from the magnetic field, and this property is also called soft magnetism. Fe-based and Co-based amorphous metallic alloys are soft magnetic, including Fe-(Al,Ga)-(P,C,B,Si), Co-Cr-(Al,Ga)-(P, B,C), Fe-(Co,Ni)-(Zr, Nb,Ta)-B and Co-Fe-Nb-B systems.
2.1.3 Corrosion resistance
In general, the grain boundary of crystal materials is high chemical reactive region which is preferred to be corroded. In addition, there are many types of defects in crystal materials. Although some metals are easy to be oxidized, the presence of defects make them form inhomogeneous protective oxide layer. Hence, corrosion will occur through defects first.
Due to lack of crystallographic defects (grain boundaries, dislocations and secondary grains), the amorphous materials can form more homogeneous protective oxide layer than the crystal materials to avoid corrosion. Hiromoto et al. [79-81] investigated the corrosion behavior of
the Zr-based amorphous alloy in the phosphate buffered solution (PBS) and SBF. Morrison et al. [82] conducted major studies on the different corrosion behavior between the Zr-based amorphous alloy and various metals (316 stainless steel, CoCrMo, and Ti–6Al–4V) in PBS.
In our previous studies, the Zr-based amorphous alloy exhibited good corrosion resistance and electrochemical stability [83]. Besides, some reports mentioned that the addition of niobium, tantalum, titanium and cobalt to Zr-Al-Ni-Cu systems will improve the corrosion resistance [84-86]. These results revealed that Zr-based amorphous alloys are high potential materials for biomedical applications.
2.2 Empirical rules for synthesis of amorphous metallic alloys
To improve the GFA of amorphous metallic alloys, there have been some basic guild lines by selecting the alloy composition. The three simple empirical rules are described as below [3,16,87-90]:
(1) Multi-component systems consisting of more than three elements
For the alloys with more than three elements, the difference in atomic size makes atoms not easy to move and increases the random packing density of atoms when the melt is quenched to the solid state. Hence, atoms in solid state exhibit short-range ordered arrangement. In general, the more elements in the alloys will improve GFA.
(2) Above 12% difference in atomic size ratio among the elements
In to the Hume-Rothery rules [90], for the formation of solid-solution, the atomic size between solution and solute have to less than 12% ~ 15%. According to the concept, for the
obstruction of the movement of atoms and viscosity, making atom arrangement for crystallization not easy to occur and increasing the glass transition temperature (Tg).
Consequently, the atomic size ratio increase will improve the GFA of amorphous metallic alloys.
(3) Negative heat of mixing among the elements
Based upon the thermodynamic theory, the heat of mixing is an important criterion for determining the bonding type of atoms. The atoms tend to bond with different atoms and tend to separate from the same ones, when the heat of mixing is very high. Hence, the different atoms tend to bond with each other and arrange randomly to form amorphous metallic alloys.
2.3 Fabrication of amorphous metallic alloys
The three basic methods to fabricate amorphous metallic alloys can be discussed by the transformation between the three states of matter:
(1) The gaseous state to the solid state
This method are majorly used to fabricate thin film metallic glasses (TFMGs) with highest cooling rate (1010 ~ 1012 K/s) compared to the method of liquid state to solid state and the solid state to solid state. There are two methods to prepare TFMGs, including sputtering [91,92] and vacuum evaporation [91], as shown in Figure 2.5 [93]. Sputtering is carried out by applying a high voltage direct current on two electrodes to produce an electronic field that accelerate the electrons emitted from cathode, and then the accelerated electrons with enough kinetic energy can ionize the gas molecules (argon or nitrogen) to produce large number of ions and electrons. Then, these ions attack the alloy or metal target, and the alloy and metal
target will deposit on the substrate to form TFMGs. Besides, vacuum evaporation is to gasify the alloys or metals under high temperature environment, and the element vapors will deposit on the low temperature substrate to form TFMGs.
(2) The liquid state to the solid state
This method is widely used to produce amorphous metallic alloys with cooling rate (103~108 K/s) and is usually called liquid quenching method. There are many types of amorphous metallic alloys that can be fabricated by this method, such as powder, ribbon and bulk. Furthermore, liquid quenching method can be carried out by twin roll quenching, melting spinning, planar flow casting, metallic mold casting, high pressure die casting and spray forming [94-97]. Figures 2.6~2.9 show the schematic diagrams of the splat quench method, two roller quench method, chill block melting spinning and planar flow casting, respectively. The theory of this method is to heat the alloy to melt, and then quench the melt through the low temperature mold or substrate to form the amorphous metallic alloys.
(3) The solid state to the solid state
The solid state to the solid state process can be carried out by solid state reaction, particle bombardment and solid state inter-diffusion method. The theory of solid state reaction method is to refine the grain size of the materials to form amorphous metallic alloys by making the materials severe plastic deformation. Generally, there are many way to make sever plastic deformation, such as cyclic extrusion or compression, torsion straining under high pressure and equal channel angular pressing (ECAP), mechanically alloying (MA) and accumulative roll bonding (ARB) [98-103]. Besides, the particle bombardment method is to destroy the surface of the materials by high quantities of high energy electrons or heavy ions
implantation and ion beam mixing are the ways to accomplish the particle bombardment method. Moreover, solid state inter-diffusion is to fabricate amorphous metallic alloys by diffusion of atoms between two distinct metals through heat treatment.
2.4 The parameters of glass forming ability
GFA is important for understanding the origins of glass formation and for designing and developing bulk metallic glasses. In early-stage, the critical cooling rate (Rc) is believed an essential way to evaluate the GFA of a glassy system, which is defined as minimum cooling rate for maintaining the amorphous structure of metallic melts during the quenching process.
The higher cooling ratecompared to Rc can suppress nucleation of crystalline phases and to fabricate amorphous alloys. On the other words, a system with lower Rc possesses higher GFA. Unfortunately, the Rc can only be determined for a system with a well-known composition. Therefore, four popular GFA parameters and criteria have been investigated to evaluate the relative GFA for metallic glass systems on the basis of thermal analysis by differential scanning calorimetry (DSC) and differential thermal analysis (DTA):
(1) Reduced glass transition temperature, Trg (=Tg/Tl, where Tl is the liquidus temperature) [104],
(2) Supercooled liquid range, ΔTx (= Tx – Tg, where Tx and Tg are the onset crystallization temperature and the glass transition temperature, respectively) [105], (3) γ (=Tx/Tg+Tl)) [106],
(4) γm (=(2Tx-Tg)/Tl) [107].
Trg (=Tg/Tm, where Tm is the melting point) is one of GFA parameters, which is initially developed by Turnbull et al. [108]. Their previous research shows that the GFA is increasing
with the increase of Tg/Tm, as shown in Figure 2-10. Moreover, Lu et al. mentioned that Trg
given by Tg/Tl shows better correlation with GFA the given by Tg/Tm. According to the Trg
(=Tg/Tl), Tl is related to the composition of an alloy system. Therefore, a careful selection of concentration of an alloy system near the eutectic point can promote the probability of cooling process without crystallization. Another famous parameter for evaluating the GFA is the ΔTx, which can be used as an index of thermal stability. The larger ΔTx means that the amorphous structure of an alloy system can exist in a wide temperature range. Figure xx presents the relations between ΔTx and Rc and show that the GFA improved with the increasing ΔTx [109].
Although Trg and ΔTx are both useful parameters for determining the GFA, they are not very reliable in some cases. Waniuk et al. [110] shows that Trg is correlated with GFA of Zr-Ti-Cu-Ni-Be alloy very well whereas has no relationship with the ΔTx. The same phenomena are also revealed by Inoue et al. from the Cu-Zr-Ti and Cu-Hf-Ti alloy systems [111]. Furthermore, Trg is not very suitable for inferring the GFA of Pd40Ni40-xFexP20 (x = 0 to 20) [112] and Mg65Cu15M10Y10 (M = Ni, Al, Zn and Mn) [113] amorphous alloy systems.
Therefore, other parameters for more accurate evaluations of the GFA were investigated for the bulk metallic glasses.
In 2002, for more suitable prediction of the GFA, a new parameter γ based on Tx/Tg+Tl
is reported by Lu and Liu [114]. In comparison, γ is more ideal way to judge the coherence of the GFA than Trg, due to the combination of thermal stability and resistance of crystallization, as shown in Figure 2-11. In addition, γ has the relation with the critical cooling temperature (Rc) and critical specimen thickness (Zc), as described follows:
Rc = 5.1 x 1021exp(-117.19γ), (2.1) Zc = 2.80 x 10-7exp(41.70γ). (2.2)
In 2007, Du et al. [107] modified the parameter γ to γm which can be defined as (2Tx-Tg)/Tl, as shown in Figure 2.12. Compared to γ, γm reported by Du et al. majorly focus on the relationship between the liquidus stability and resistance of crystallization. Besides, the Lu et al. also use the statistical correlation factor (R2) to evaluate which GFA parameter is more convinced. The result presents that γm is most promising GFA parameter, due to the R2=0.931.
2.5 Introduction of corrosion and biocompatibility
Corrosion can be defined as the phenomenon of material degradation caused by chemical erosion which is reacted with external environment. Naturally the metallic minerals are stable compounds, and most of the stable compounds are of the form of oxidation state.
Therefore, as long as the conditions are sufficient in external environment, the metal will be transformed into the initial state (oxidation state), and this is the corrosion reaction.
Corrosion is an electrochemical reaction and it can be classified into eight categories [115] as shown in Figure 2.13.
(1) Uniform corrosion: Metallic surface generates a layer of uniform corrosion, meaning that the entire surface of the metal is doing the electrochemical reaction at the same time.
(2) Galvanic corrosion: This is also known as heterogeneous metal corrosion. When two
different metals contact, the metal whose activity is larger would accelerate the corrosion reaction.
(3) Crevice corrosion: This means that the local electrochemical corrosion occurred in the gap and the stagnant solution which has covered on the surface. Because the dissolved oxygen in the aperture is more rarefied than the external, this area acts as the formation of anode, and the external oxygen-rich area is the cathode. The different concentration of oxygen in internal and external aperture forms oxygen concentration cell, and it results in the corrosion.
(4) Pitting corrosion: This only occurs in a metal material which has the passive film, and it is a partial erosion of the metal surface. The pitting corrosion can easily occur in the place which has impurities on the surface, non-uniform structure, and the uneven composition.
(5) Intergranular corrosion: The grain boundary has a strong chemical activity, so it may be more prone to corrosion at the position. When the alloy composition or impurities in the grain boundary separates, the grain boundary would generate potential difference and proceed corrosion reaction in the metal contact phase which has more negative potential. This corrosion would take place along the grain interfaces, causing the crack-like corrosion reaction.
(6) Selective corrosion: A particular metal is within a solid metal alloy because of the priority remove in this corrosion process. Usually the ingredients with active chemical characteristics would be extracted first.
(7) Erosion leaching: The corrosive fluid and the metal surface relative motion can cause the accelerated rate of metal corrosion. The feature is that the metal surface will have a groove hole and the direction is the same as the corrosive fluid flow
(8) Stress corrosion: When the stress and corrosion exist at the same time, the metal will be gradually broken earlier before it reaches the breaking point. Stress corrosion cracking (SCC) of metals is the breakdown of the metal surface that combines tensile stress and corrosive environment effects. During the SCC reaction, the material usually shows minor erosion, but the local crack spreads very quickly along the cross-sectional area in the metal. And the last causes serious damage.
Biocompatibility is a term which describes the interaction between biomaterial and biological system. Some definitions of biocompatibility are mentioned as follows: (1) The ability of a material to perform with an appropriate host response in a specific application [116]. (2) The quality of not having toxic or injurious effects on biological systems [117]. (3) Biocompatibility is the capability of a prosthesis implanted in the body to exist in harmony with tissue without causing deleterious changes [118]. Hence, biomaterials with good biocompatibility are more suitable for biomedical applications. The application fields of biomaterials are summarized in Figure 2.14 [40].
2.6 Traditional metallic alloys for load-bearing bio-implant application
Ideal clinical biomaterials have to satisfy the following conditions: good biocompatibility, non-toxicity, non-allergic responses, noncarcinogenic effects and applicable mechanical properties. Metals, polymers, ceramics, and biopolymers are four well-known major types of biomaterials. Polymers and ceramics are widely used in many biomedical applications [119-122], but they are not favorable for load-bearing applications due to their poor strength or fatigue endurance limit [40, 41]. On the contrary, the sufficient yield strength and the resistance to cyclic loading make metals suitable for orthopedic load-bearing
applications, including dental implants, artificial joint replacement, and bone fixation [39-41].
Notwithstanding that Ti-based and Co-based alloys are popular for metallic implants in early stages, their unsatisfactory wear resistance gives rise to toxic debris after long-term uses [48,49,51,52].
Recently, some metals such as cobalt chromium (Co-Cr) alloys, tantalum (Ta), niobium (Nb) and titanium (Ti) have been used for implants, since they have excellent corrosion resistance [123]. Titanium (Ti) is the most popular metal for producing long-term implantable devices due to its excellent biocompatibility. However, the lower strength and low hardness of commercial pure titanium (CP Ti, typically ~300–500 MPa for tensile strength and ~1.5 GPa for hardness) are issues for some clinical applications. In this regard, a number of titanium alloys were developed for biomedical applications. However, some studies have reported that lower wear resistance of titanium alloys may produce toxic debris after long-term usages [51,52]. In Ti alloy (Ti–6Al–4V), it was reported that aluminum (Al) and vanadium (V) were dissolved [124]. Al is a growth inhibitor of bone and a possible cause of Alzheimer's disease [125] and V has strong cytotoxicity. Cytotoxicity is often dependent on the ionization tendency of the metals. Highly corrosive materials in the body may release cytotoxic ions and cause cell apoptosis and necrosis after long-term use [61].
It is known that implants with a high Young's modulus may cause more significant stress shielding effect. The Young's modulus of conventional implant materials of Ti and Co–Cr alloy are 121 GPa and 241 GPa, respectively, which are much higher than the human cortical bone of around 3–20 GPa [40, 58]. The modulus mismatch between the implants and the bond tissue may cause stress shielding effect and may result in the decrease of the bone
