國立中山大學材料與光電科學學系 碩士論文

國立中山大學材料與光電科學學系 碩士論文

Department of Materials and Optoelectronic Science National Sun Yat-sen University

Master Thesis

金屬玻璃鍍膜對鎂合金表面之硬度改質 金屬玻璃鍍膜對鎂合金表面之硬度改質 金屬玻璃鍍膜對鎂合金表面之硬度改質 金屬玻璃鍍膜對鎂合金表面之硬度改質

Surface Hardness Improvement in Magnesium Alloy by Metallic-Glass Sputtered Film

研究生：陳柏佑

Bo-You Chen 指導教授：黃志青 博士

Dr. Jacob C. Huang

中華民國 100 年 7 月

July 2011

Table of Content

Table of Content………...………..i

List of Tables……….………..…….iv

List of Figures………...……v

致謝………..………ix

中文摘要………...………x

Abstract………...…………...………xi

Chapter 1 Introduction……….…………1

1-1 Characteristics of Mg alloys………...……….1

1-2 Classifications of magnesium alloys………...…………3

1-3 Amorphous alloys………4

1-4 Status of bulk metallic glasses and thin film metallic glasses…...…………..6

1-5 Motivation………...………8

Chapter 2 Background and literature review……….………..…….11

2-1 Applications of Mg alloys………...…………..……11

2-2 Mg alloy systems………...…………...…….11

2-3 Fabrication of amorphous alloys………...………13

2-3-1 Cooling from gaseous state to the solid state………...……...14

2-3-2 Cooling from liquid state to the solid state………...………...15

2-3-3 Transformation from solid state to solid state……….16

2-4 Characteristics of amorphous alloys………...…………...17

2-4-1 Glass forming ability (GFA)………...……….17

2-4-2 Recently developed GFA criterions…………...………..18

2-5 Empirical rules for forming amorphous alloys………...………...19

2-6 Physical vapor deposition………...………...21

2-6-1 Introduction of sputtering………...……….21

2-6-2 DC and RF sputtering………...………...23

2-6-3 Nucleation and growth of thin films………24

2-6-4 Growth of amorphous films………...26

2-7 Properties of TFMGs………...………..26

2-7-1 Thermal properties………...27

2-7-2 Mechanical properties………..28

2-7-3 Chemical properties………...………..31

Chapter 3 Experimental Procedures………...………..32

3-1 Materials………...……….32

3-2 Sample preparation………...……….33

3-2-1 Substrate preparation………...………33

3-2-2 Film preparation………...………...33

3-3 Property measurements and analyses………...……….34

3-4 Microhardness tests………...………35

3-5 Nanoindentation tests………...……….35

3-6 Observation of indentation marks………..36

Chapter 4 Results and Discussions………...37

4-1 Amorphous nature………...………..37

4-2 Microhardness and Nanoindentation tests of Pd77Cu₆Si₁₇ TFMGs on AZ31………...………...38

4-3 Hardness marks of Pd77Cu₆Si₁₇ coatings on AZ31………41

4-4 Hardness calculation and comparison……….…………...……...45

Chapter 5 Conclusions………...………..50

References………...………52

Tables………...………58 Figures………...………..65

List of Tables

Table 1-1 The standard four-part ASTM designation system of alloy and temper for the magnesium alloy………..………..58 Table 1-2 The effect of separate solute addition on the mechanical properties……...59 Table 1-3 Fundamental properties and application fields of bulk amorphous and

nanocrystalline alloys………...………...60 Table 2-1 Binary amorphous systems and mixing enthalpy values calculated based on

Miedema’s macroscopic model………...………61 Table 3-1 Chemical composition of the AZ31 Mg alloy (in wt%)………...…...62 Table 4-1 Composition difference between the alloy targets and the as-deposited thin

films. The compositions of the film are analyzed by SEM/EDS………….63 Table 4-2 The fitting parameters of the PCS TFMGs deposited on the AZ31 substrate

obtained from the nanoindentation data……….…………...…….……….64

List of Figures

Figure 1-1 The atomic arrangment of long-range-order structure…….………….65 Figure 1-2 The atomic arrangment of short-range-order structure………65 Figure 1-3 The shiny and smooth apperance of metallic glasses……...…………...66 Figure 1-4 The frame for the upscale models of the Vertu mobile phone is made of

liquid-metal alloy due to its high strength, hardness, and scratch resistance………...………..66 Figure 1-5 (a) Conical spring of microactuator, and (b) a fundamental structure of

micro-switch made of thin film metallic glasses………...………..67 Figure 2-1 Schematic drawing of (a) sputtering and (b) vacuum evaporation……..68 Figure 2-2 Schematic drawing of binary phase diagram……….………..69 Figure 2-3 Mechanisms for the stabilization of supercooled liquid and the high

glass-forming ability………70 Figure 2-4 Events that occur on a surface being bombarded with energetic

atomic-sized particles………..71 Figure 2-5 Schematic illustrations of three basic growth modes for thin film……..71 Figure 2-6 DSC thermography curve of the Pd-TFMG………...……….72 Figure 2-7 DSC thermography curve of the Zr-TFMG……….72 Figure 2-8 TTT diagram for the onset of crystallization in the Zr-TFMG…………73 Figure 2-9 TTT diagram for the onset of crystallization in the Pd-TFMG…...……73 Figure 2-10 (a) Nanoindentation hardness measurement results of the as-deposited

and the annealed films as a function of the concentration of Zr-Cu-Al TFMGs; (b) Nanoindentation Young’s modulus measurement results of the as-deposited and the annealed films as a function of the concentration of Zr-based TFMGs……….74

Figure 2-11 Relationship between Vickers hardness (Hv) and Young’s modulus (GPa) for various BMGs………..………..75 Figure 2-12 The arrangement of atoms in (a) crystalline and (b) amorphous states...76 Figure 2-13 The illustration of the shear transformation zones (STZs) (a) before shear

deformation and (b) after shear deformation in two-dimensional space.76 Figure 2-14 Schematic drawing of the fluid zones of amorphous alloy…………...77 Figure 3-1 The flow chart of the experimental procedures in this study……...…...78 Figure 3-2 The standard Nano Indenter^® XP is a complete, turnkey system

consisting of the major components illustrated………...79 Figure 4-1 XRD pattern of (a) the AZ31 substrate, (b) the Pd77Cu6Si17 thin film

deposited on silicon substrate………..………80 Figure 4-2 XRD pattern of the Pd77Cu6Si17 thin film deposited on the AZ31

substrate……….………..…81 Figure 4-3 (a) AFM image of the AZ31 substrate topography, and the roughness Ra

is ~70 nm after diamond paste polishing, (b) OM image of the AZ31 substrate morphology………..………..…..82 Figure 4-4 (a) AFM image of the AZ31 substrate topography, and the roughness Ra

is ~10 nm after SiO2 polishing, (b) OM image of the AZ31 substrate morphology………..83 Figure 4-5 The microstructure of the as-received AZ31 billet after etching……...84 Figure 4-6 The hardness-β curves of the PCS-1000 and PCS-2000, measured from

the microhardness tests………..………..85 Figure 4-7 The hardness-β curves of the PCS-200, PCS-500, PS-1000, and

PCS-2000, measured by nanoindentaiton tests.………...…86 Figure 4-8 The combined hardness-β plots for all the experimental data measured

from the microhardness and nanoindentaiton tests………..…87

Figure 4-9 The SEI image of the AZ31 substrate at an applied load of 10 g…...….88 Figure 4-10 The SEI image of the Pd77Cu6Si17 TFMG 2000 nm in thickness (a)

without obvious shear bands, and (b) with shear bands at an applied load of 10 g………..89 Figure 4-11 The SEI image of the Pd77Cu₆Si₁₇ TFMG 2000 nm in thickness at an

applied load of 25 g………..…………..….90 Figure 4-12 The SEI image of the Pd77Cu₆Si₁₇ TFMG 2000 nm in thickness at an

applied load of 50 g………..………..…….90 Figure 4-13 The SEI image of the Pd77Cu₆Si₁₇ TFMG 2000 nm in thickness at an

applied load of 100 g………...91 Figure 4-14 (a) The enlarged SEI image taken from Figure 4-13, (b) the enlarged SEI

image taken from Figure 4-14 (a)………92 Figure 4-15 The SEI image of the Pd77Cu₆Si₁₇ TFMG 1000 nm in thickness at an

applied load of 10 g………...……….……….93 Figure 4-16 The SEI image of the Pd77Cu₆Si₁₇ TFMG 1000 nm in thickness at an

applied load of 25 g……….93 Figure 4-17 The SEI image of the Pd77Cu₆Si₁₇ TFMG 1000 nm in thickness at an

applied load of 50 g……….94 Figure 4-18 The SEI image of the Pd77Cu₆Si₁₇ TFMG 1000 nm in thickness (a) with

obvious cracks, and (b) without any crack at an applied load of 100 g...95 Figure 4-19 The SEI image of the Pd77Cu₆Si₁₇ TFMG 2000 nm in thickness under nanoindentation testing.………..……….………96 Figure 4-20 (a) The enlarged SEI image (b) the enlarged and rotated SEI image of the

Pd77Cu6Si17 TFMG 2000 nm in thickness under nanoindentation testing.………..…97 Figure 4-21 (a) The SEI image of the Pd77Cu6Si17 TFMG 1000 nm in thickness under

nanoindentation testing. (b) the enlarged SEI image taken from Figure 4-23 (a).……….………..……98 Figure 4-22 Comparison of the experimental data (in various symbols) and the best

fit predictions based on equation 4-3 (in various lines) for the PCS 200, 500, 1000 and 2000 samples under nanoindentaiton. Note that the horizontal axis is presented in log scale.………...……….….99 Figure 4-23 The presicted and postualated trend of the relative minimal substrate hardness Hs as a function of the surface hard coating thickness. The optimum hard coating thickness might be around 200 nm………100

致謝 致謝致謝 致謝

即將結束兩年的碩士班生涯，回顧這些學習做研究的日子，從大學部的專題 實驗到碩士論文。首先，我要感謝恩師 黃志青教授。老師除了在課業以及實驗 的的指導，更花很多時間教導我們做人做事的道理。每當我們遇到困難，不論是 研究上或是生活上的，老師總是會關心我們，並陪伴我們渡過低潮。除了研究，

老師也很重視英文能力，幾年來利用英文的小測驗、競賽，甚至是英文的研究討 論，來幫助我們增強英文能力。對於剛踏入研究領域懵懂無知的我來說，老師的 訓練與照顧都是全方位的。

此外，在這充滿酸甜苦辣的兩年碩士生涯，非常感謝實驗室各位學長、學弟 妹們的幫助以及支持，特別感謝剛進實驗室的時候，帶我入門的鴻昇學長；一起 陪伴我做研究、學英文的良師益友─宇庭學長；可以陪我聊心事，以及指導我實 驗的名哲學長、浩然、大豪學長；常常給予我們實驗啟發以及批判的敬仁、育誠、

海明、炎暉等諸位學長。也謝謝從大學時代就一直陪伴我的好友昭憲，還有球隊 裡常常帶給我歡笑的球伴們，讓我可以適時地調節生活緊湊的步調。

最後，我要感謝一路上無怨無悔為我付出的家人們，沒有他們默默的付出與

支持，我就沒有辦法順利的完成我的學業。『天下沒有不散的筵席』。即使畢業之

後，我都會常常想起這些良師益友們帶來的啟發，我會懷念這些年來在西子灣生 活的點點滴滴，相信總有一天，我們一定會在社會上闖出自己的一片天，並且重 逢分享這些喜悅。

摘要 摘要 摘要 摘要

鈀銅矽非晶質薄膜的玻璃形成能力和硬度佳鈀銅矽非晶質薄膜的玻璃形成能力和硬度佳鈀銅矽非晶質薄膜的玻璃形成能力和硬度佳鈀銅矽非晶質薄膜的玻璃形成能力和硬度佳，，，，因此被選為用來改善因此被選為用來改善因此被選為用來改善 AZ31 鎂合因此被選為用來改善 鎂合鎂合鎂合 金表面硬度的

金表面硬度的 金表面硬度的

金表面硬度的鍍層鍍層鍍層鍍層。。。。實驗中實驗中實驗中實驗中，，，，從三十到兩百奈米從三十到兩百奈米從三十到兩百奈米從三十到兩百奈米，，，，不同厚度的鈀銅矽薄膜試片都不同厚度的鈀銅矽薄膜試片都不同厚度的鈀銅矽薄膜試片都不同厚度的鈀銅矽薄膜試片都 會經過微硬度還有奈米壓痕的測試

會經過微硬度還有奈米壓痕的測試 會經過微硬度還有奈米壓痕的測試

會經過微硬度還有奈米壓痕的測試。。。。硬度和相對應的壓印深度可由一種量化模型硬度和相對應的壓印深度可由一種量化模型硬度和相對應的壓印深度可由一種量化模型硬度和相對應的壓印深度可由一種量化模型 推算而出

推算而出 推算而出

推算而出。。。。相關的作用參數和硬度值可以經過反覆的計算得到相關的作用參數和硬度值可以經過反覆的計算得到相關的作用參數和硬度值可以經過反覆的計算得到相關的作用參數和硬度值可以經過反覆的計算得到。。。根據實驗結果顯。根據實驗結果顯根據實驗結果顯根據實驗結果顯 示

示 示

示，，，，鎂合金的淺層區表面硬度可以藉由鈀銅矽非晶質薄膜鎂合金的淺層區表面硬度可以藉由鈀銅矽非晶質薄膜鎂合金的淺層區表面硬度可以藉由鈀銅矽非晶質薄膜鎂合金的淺層區表面硬度可以藉由鈀銅矽非晶質薄膜，，，隨著壓印深度的減少，隨著壓印深度的減少隨著壓印深度的減少隨著壓印深度的減少 而大幅地提升

而大幅地提升 而大幅地提升

而大幅地提升。。。。除此之外除此之外除此之外除此之外，，，，使用比較薄的非晶質鍍層使用比較薄的非晶質鍍層使用比較薄的非晶質鍍層使用比較薄的非晶質鍍層((((例如兩百奈米例如兩百奈米例如兩百奈米)例如兩百奈米))，)，，基板和薄，基板和薄基板和薄基板和薄 膜之間的作用會比較強

膜之間的作用會比較強 膜之間的作用會比較強

膜之間的作用會比較強。。。。而使用比較厚的非晶質鍍層而使用比較厚的非晶質鍍層而使用比較厚的非晶質鍍層而使用比較厚的非晶質鍍層((((例如兩千奈米例如兩千奈米例如兩千奈米)例如兩千奈米))，)，，較容易造，較容易造較容易造較容易造 成薄膜的破裂

成薄膜的破裂 成薄膜的破裂

成薄膜的破裂。。。。本實驗估算本實驗估算本實驗估算本實驗估算，，，，適當的非晶質薄膜厚度大約在兩百奈米左右適當的非晶質薄膜厚度大約在兩百奈米左右適當的非晶質薄膜厚度大約在兩百奈米左右適當的非晶質薄膜厚度大約在兩百奈米左右。。。 。

關鍵字 關鍵字 關鍵字

關鍵字：：：：鎂合金鎂合金鎂合金鎂合金、、、、硬度硬度硬度硬度、、、、濺鍍濺鍍濺鍍濺鍍、、、、非非非非晶質薄膜晶質薄膜晶質薄膜晶質薄膜、、、、奈米壓痕奈米壓痕奈米壓痕奈米壓痕

Abstract

The Pd77Cu6Si17 (PCS) thin film metallic glasses (TFMGs) with high glass forming ability and hardness are selected as a hard coating for improving the surface hardness of the AZ31 magnesium alloy. Both micro- and nano-indentation tests are conducted on the specimens with various PCS film thicknesses from 30 to 2000 nm. The apparent hardness and the relative indentation depth (β) are integrated by a quantitative model.

The involved interaction parameters and relative hardness values are extracted from iterative calculations. According to the results, surface hardness can be enhanced greatly by PCS TFMGs in the shallow region, followed by gradual decrease with increasing β ratio. In addition, the specimens with thinner coating (for example, 200 nm) show greater substrate-film interaction and those with thick coating (for example, 2000 nm) become prone to film cracking. The optimum TFMG coating thickness in this study is estimated to be around 200 nm.

Keywords: Magnesium alloys, hardness, sputtering, thin film metallic glass, nanoindentation

Chapter 1 Introduction

1-1 Characteristics of Mg alloys

Among many engineering materials, Mg alloys have raised much attention in recent

years. For industrial need of developing high strength materials with light weight, Mg alloys

have attracted more and more interests due to their specific strength among all structural

materials. Because of the good recyclability, Mg alloys are regarded as a potential structural

material.

The density of pure Mg (1.74 g/cm³) is lower than that in pure Al (2.70 g/cm³) and Ti

(4.54 g/cm³), but Mg shows a high strength to weight ratio among all structural materials. For

thermal conductivity, compared with plastics, Mg alloys exhibit a high thermal conductivity

due to good thermal dissipation. For damping and crash resistance, the damping resistance of

Mg alloys is better than that of aluminum and plastics. The crash resistance of Mg alloys is

notable to be superior to that of plastics. In electromagnetic shielding capability, the Mg wall

can shield the electromagnetic wave effectively [1]. If electronic products are made of Mg

alloys, or Mg alloys are coated on the surface of electronic products, the electromagnetic

shielding capability may protect human bodies from electromagnetic wave damage.

Moreover, the good casting ability of Mg alloys makes it easier to be die-cast into

complicated shapes. Similar to other alloys, Mg alloys can also be reused and recycled

through remelting procedures. The recyclability makes Mg alloys possible to lower capital

cost and economize the natural resources. This characteristic benefits the extension of

commercial use of Mg alloys.

The above advantages make Mg alloys become irreplaceable in industry. For the

computer, consumer and communication (3C) of electronic products, Mg alloys are used for

the shell and radiator components. Moreover, Mg alloys show the good electromagnetic

interference (EMI) capability. Mg alloys was regarded as a new family of promising materials

in application of electronic products. Although Mg alloys exhibit a lot of advantages, there

are still some undesirable properties: poor corrosion resistance, poor wear resistance,

unsatisfactory creep resistance, and high chemical reactivity. Poor corrosion resistance

drastically deteriorates the properties of Mg alloys in severe environments. High chemical

reactivity will cause the formation of oxidization easily. In addition, Mg alloys have faced a

challenging problem in the intrinsic properties, such as low hardness, and low strength.

Compared with other industrial alloys, the strength of Mg alloys is too low to bear intense

stress. These disadvantages restrict the engineering applications of Mg alloys. As a result,

many efforts have been devoted to the improvement in these properties.

1-2 Classifications of magnesium alloys

For commercial Mg alloys, a variety of elements are added for the formation of solid

solution, such as Al, Ce, Li, Ag, Th, Zn, and Zr [2]. In general, the main components and

compositions of Mg alloys can be indicated by the capital letters and the followed numbers.

Table 1-1 [3] presents the standard four-digit American Society for Testing and Material

(ASTM) designation system for Mg alloys and their heat treatments in temper. The first two

letters indicate two main components with abbreviations of the elements. A is the

abbreviation of aluminum, C is for copper, E is for rare earths, F is for iron, H is for thorium,

K is for zirconium, L is for lithium, Q is for silver, and Z is for zinc. The first letter describes

the element which possesses a larger quantity than the other element in addition to Mg itself.

If the quantities are equal of two components, the letters will be listed alphabetically. The

numbers followed by the letters stand for the compositions of the above elements in weight

percent (wt%). Taking AZ31B-H24 for example, it means that the alloy is composed of

nominal 3 wt% aluminum, 1 wt% zinc, and is referred to B modification. B is used to

distinguish the same AZ31 form that contains different levels of impurity. The H24

designation indicates that the alloy is strain-hardened and partially annealed.

The solute elements are usually added to Mg alloys for the improvement in casting

capability, corrosion resistance, etc. For example, the addition of Al to Mg solvent is mainly

for solute hardening and precipitation hardening through the Mg₁₇Al₁₂β phase. The

addition of Zn will enhance casting capability of the Mg alloys. The effects of various solutes

on mechanical, corrosion, and casting behaviors are all listed in Table 1-2 [4]. Due to these

advantages above, Mg alloys, such as AZ, AM, ZK series, have become popular in

commercial markets.

1-3 Amorphous alloys

The structure of crystalline materials shows long-range-order (LRO) with repeating unit

cell, as shown in Figure 1-1 [5]. Different from crystalline materials, the structure of

amorphous alloys exhibits a disordered arrangement that lacks crystalline periodicity. The

random atomic structures are displayed in amorphous alloys. Although amorphous alloys are

considered to be random atomic arrangement, the atomic arrangements are not completely

random but with a short-range-ordered (SRO) structure, as shown in Figure 1-2 [5]. Based on

the structure and thermodynamic characteristics, amorphous alloys are called non-crystalline

metals, liquid metals, glassy metals, or metallic glasses. Amorphous alloys show metal

lustrous appearance and smooth surface, as shown in Figure 1-3 [6]. They are in contrast to

the environmental transparent glasses.

To fabricate amorphous alloys in the form of ribbons, flakes, or powders, rapid

quenching is required. The nucleation of crystalline phase can be prevented at a high cooling

rate. In recent years, many methods to produce amorphous alloys have been reported. With

the third element added into the original binary system, the cooling rate for fabricating

amorphous alloys can be lowered from 10⁷ K/s to 10³ K/s. Because of the increasing demands

for light and strong materials which can resist the severe environment, bulk metallic glasses

(BMGs) have been developed.

In general, the vacant space among atoms in metallic glasses is called free volume. A

shear localization occurs at the sites with a high amount of free volume. Amorphization of

metallic material might result in excellent properties due to the atomic structure of

amorphous phase, such as lower Young’s modulus, better tensile strength, higher electric

resistance, and excellent gas absorption ability, which is different from the corresponding

crystalline alloys.

Recently, metallic glasses are regarded as potential materials, and can be used as a

coating, in an attempt to strength the substrate due to the unique physical and mechanical

properties. The fundamental properties and application field of metallic glasses are listed in

Table 1-3 [7]. Ternary and multi-component BMGs with the functional applications on

protective coatings were fabricated by Inoue’s group [8]. The shell of Vertu, the famous cell

phone of Nokia, was coated by liquid-metal alloys. It shows better mechanical properties,

good-looking shiny surface as shown in Figure 1-4 [9]. To expand the applications in the

electronic products and microscaled devices, metallic glasses with excellent mechanical

properties are good candidates for the thin film coating. So far, the Pd-based and Zr-based

thin film metallic glasses (TFMGs) have been used for nanopatterning and microactuator,

which can be seen in Figure 1-5 [10].

1-4 Status of bulk metallic glasses and thin film metallic glasses

Over the past decades, bulk metallic glasses (BMGs) have attracted extensive interests

because of their characteristics, such as high elastic energy, high yield strength, good wear

resistance, reasonable corrosion resistance, and good forming in the viscous state. With the

development of thin film technology, thin film metallic glasses (TFMGs) were successfully

fabricated by co-sputtering and alloy sputtering. The liquid quenching method operated at a

cooling rate of 10³-10⁸ K/s was widely employed to prepare various types of metallic glasses.

TFMGs could also be fabricated in the binary Al-Fe and Au-La systems [11, 12] at a

sufficient fast cooling rate (above 10⁸ K/s). To lower the critical cooling rate, multicomponent

systems consisting of more than three elements were adopted following the empirical rules

for the formation of metallic glasses. The ternary and quaternary systems, Zr-Cu-Al [13],

Zr-Cu-Al-Ni [14], and Pd-Cu-Si [15] TFMGs, were successfully deposited via alloy

sputtering process.

In the past decade, the composition for forming metallic glasses was a very

time-consuming process. With the sputtering process, the amorphous structure of the films

can be well controlled, and thus the sputtering process can eliminate the time-consuming

selection of glass forming composition. It can be expected that the amorphization of the films

will be beneficial to the physical and mechanical properties. Moreover, the smoother and

shinier surfaces can be obtained. Recently, TFMGs have been extensively applied to the

fields of semiconductor industry, and micro-electromechanical systems (MEMSs). The

metallic glasses which would become the viscous flow state in the supercooled liquid region

(SLR), and TFMGs will form any shape easily upon heating at the SLR temperature [16].

After forming, the cooled TFMGs can still maintain their excellent properties of amorphous

structure. The thermoplastic forming properties of TFMGs will be a promising material for

use of microforming and microreplication.

1-5

Motivation

Nowadays, Mg alloys are extensively used in industrial and commercial products due to

their specific strength and low cost. The light weight of Mg alloys makes it practicable that

cell phones can be carried conveniently. The electromagnetic shielding capability of Mg

alloys plays important role in cell phone industrial, which can protect human body from

electromagnetic wave damage. However, the lower strength to resist intense stress limits the

engineering applications of Mg alloys. If the problems of lower hardness and lower strength

can be solved, Mg alloys will be more promising in commercial uses.

From a viewpoint of low-weight alloy with the high hardness and strength, it is noted

that TFMGs coated on Mg alloy is a good candidate. In the past decade, many rigid ceramic

thin films with a crystalline structure were usually used to be as a modified coating for the

improvement of hardness or wear resistance [17]. TFMGs with an amorphous structure can

also be employed for the achievement of hardness enhancement. Compared with the ceramic

thin films, the amorphous structure of TFMGs without crystalline defects, like grain boundary,

could make sample surfaces smooth and shiny, resulting in valued aesthetic appearance.

Moreover, ceramic materials always exhibit a drawback of brittle nature. Once rigid ceramic

thin films are subjected to overloaded force, crack opening will be triggered, and thus the

coating tend to fragment into several pieces by catastrophic fracture without noticeable

ductility. TFMGs can possess ductility rather than the stubborn brittle nature of ceramic thin

films,. The above properties make TFMGs attractive for many applications, including

electronic, medical, and surface coating.

The AZ31 magnesium alloys were selected as the raw substrates. In a view of ductile

metallic glass with a high glass forming ability, we note that the ternary Pd-Cu-Si alloy

systems may be good candidates for preparing the TFMGs. It is uneasy to control the

composition for co-sputtering. In an attempt to obtain accurate composition of the Pd-Cu-Si

TFMGs, alloys target is selected in the sputtering process. In this study, the AZ31 Mg alloys

coated with the Pd-Cu-Si TFMGs are prepared and analyzed in terms of the mechanical

properties.

In sputtering process, the thickness of film is a function of the deposition time. In our

general senses, the thicker the film is, the more rigid in nature the film will be. However,

once the thickness of the deposited film is over a critical value, the films will readily be

peeled off form the substrates. On the contrary, the film with a very thin thickness would

contribute little to its underlying properties. Therefore, it is crucial issue that the suitable

thickness of the film can satisfy a good coherence to the substrate. The effects of various

thicknesses in the Pd-Cu-Si are investigated. We compare these types of TFMGs with

different indentation depth to various thicknesses, and discuss the difference in mechanical

behavior.

Chapter 2 Background and literature review

2-1 Applications of Mg alloys

Mg alloys have become one of the most popular materials in industry ever since Mg

element was discovered by Davy in 1808. Due to the tendency of light weight utilization and

environmental consciousness, Mg alloys were widely used for structural utilization,

especially in the fields of transportation vehicles. For the light weight goal, it is necessary to

adopt the light metals in the components of vehicles [18]. According to the advantages

mentioned in chapter 1, Mg alloys can be regarded as promising materials to replace Al alloys

or plastics on the applications of electronic products and important structural materials.

Owing to the hexagonal closed-packed (HCP) structure, Mg alloys show poor

workability. Thus, the shaping of Mg alloys needs adopting the die casting, thixomolding [19]

or thixocasting [20]. The improvement in the workability of Mg alloys will be important for

the purpose of promoting mass production of Mg alloys in the engineering applications.

2-2 Mg alloy systems

The main application of Mg metals and Mg alloys can be divided into structural

applications and non-structural applications. In structural applications, there are several

systems based on various alloy compositions for conventional cast.

Take some Mg alloys systems for examples. In the Mg-Al-X systems, Mg-Al-Zn,

Mg-Al-Mn, Mg-Al-Si, and Mg-Al-RE (rare earth elements) series have been widely

investigated. For the Mg-Al-Zn series alloys, Zn plays the role in strengthening. For example

AZ91D show a good mechanical performance, castability and corrosion resistance. The

ductility of Mg-Al-Mn series alloys can be improved by the addition of Mn. For the

Mg-Al-RE series alloys, AE42 show the improved creep resistance, ductility, and corrosion

resistance. Moreover, it is indicated that the addition of Ca in AZ series alloys can make the

grain smaller to improve yield strength because of the Al₂Ca and β-Mg17Al₁₂ phases formed

on the grain boundaries [21, 22].

In the Mg-Zn-X systems, the Mg-Zn-Zr and Mg-Zn-RE series alloys have been

commonly used. For the Mg-Zn-Zr series alloys, the ability of grain refinement in the Mg-Zn

alloys contributed to the ZK51 and ZK61 alloy products, which exhibit a high yield stress and

good casting ability. For the Mg-Zn-RE series alloys, ZE33 and ZE41 exhibit a good creep

resistance. In addition, they were extensively used for the casting in elevated temperature. In

the Mg-Th-X systems, the addition of Th can contribute a good creep resistance to Mg alloys.

For the Mg-Th-Zn series alloys, the addition of Zn can increase the creep resistance of Mg

alloys. Zh62A is a famous family of Mg alloys because it exhibits relatively high strength at

room temperature.

Finally, in the Mg-Ag-X systems, the Mg-Ag-RE series are mostly popular. In the

Mg-Ag-RE series alloys, QE22 is widely used for the aerospace applications. Besides the

addition introduced above, there are still some often used. For instance, yttrium (Y) and rare

earth element can enhance creep resistance to the QE series alloys after fully hardened by

means of the T6 thermal treatment [23]. It is found that the properties of various Mg alloy

systems depends on the addition of various alloys elements according to different demands

for applications.

2-3 Fabrication of amorphous alloys

In the past twenty years, many techniques for fabrication of amorphous alloys have been

developed. The process can be classified into three main groups according to the difference in

cooling rates. The first one is forming amorphous alloys from gaseous state to solid state.

This process can be carried out by sputtering and evaporation in a high vacuum environment.

The second one is forming amorphous alloys from the liquid state to solid state. By means of

the temperature difference between the target and substrate, sufficient cooling rate for

quenching can be achieved. The third one is forming amorphous alloys from solid state to

solid state by a variety of plastic deformation, such as ion beam mixing, mechanical alloying

(MA) and accumulative roll bonding (ARB) [24]. The three groups will be discussed in the

following chapters.

2-3-1 Cooling from gaseous state to the solid state

The amorphous alloys can be fabricated by depositing alloys or metal elements with

gaseous phase on the substrate with a relatively high cooling rate of 10¹⁰-10¹² K/s. The

processes of sputtering and vacuum evaporation belong to this method [25], as illustrated in

Figure 2-1 [26].

In the sputtering process, a high voltage is added to electrodes to create an electronic

field, and electrons are accelerated and emitted from the cathode. After gas molecules are

ionized, the metal or alloy vapors will be deposited on the substrate to form amorphous thin

film while the target is impacted by the gaseous ions. In order to achieve a sufficient cooling

rate, liquid nitrogen or helium is used in the cooling process.

In 1954, Buckel [27] synthesized amorphous films composed of pure Ga and Bi. The

thermal evaporation impact the substrate which was maintained at a liquid helium

temperature with a cooling rate above 10¹⁰ K/s. Subsequently, in 1986, Cotts et al. [28]

fabricated the Ni-Zr multilayer thin films with the magnetron sputter method. The

amorphization was successfully observed by using a differential scanning calorimetry (DSC).

With the development of these techniques, cooling from gaseous state to solid state has been

a common way to fabricate amorphous films. Moreover, co-sputtering deposition process [29]

and alloy sputtering deposition process [13] are also adopted to synthesize multi-component

monolayer TFMGs.

2-3-2 Cooling from liquid state to the solid state

Liquid quenching at a cooling rate of 10³-10⁸ K/s provides another way to fabricate

amorphous alloys. Amorphous alloys can be made into different forms by the liquid

quenching method, inclusive of splat quenching, melt spinning, two rollers quenching, planar

flow casting, metallic mold casting, spray forming, and high-pressure die casting [30-33]. In

the process of fabrication, after the alloys are atomized or melt, they will be quenched on

substrate with relatively high thermal conductivity or on a low-temperature mold with a

water-cooling system. By means of this method (gun quenching), the first amorphous metal

Au-Si was fabricated by Klement et al. [34] in 1960.

2-3-3 Transformation from solid state to solid state

The ways of transformation from the solid state to solid state include severe plastic

deformation, solid-state reaction, particle bombardment, and solid-state interdiffusion. The

cooling rate becomes no more important in the process. The severe plastic deformation leads

to the grain size refinement and the lattice rearrangement. As long as the grain size becomes

small enough, the crystalline structure will turn into amorphous structure. In order to carry

out such a large amount of plastic deformation, the ways of cyclic extrusion or cyclic

compression, torsion straining under high pressure, equal channel angular pressing (ECAP),

mechanical alloying (MA), and accumulative roll bonding (ARB) were commonly adopted

[35-40].

According to the principles of particle bombardment, as long as the alloy surface is

impacted by the heavy ions or electrons with high energy, the structure of alloy surface will

be rearranged to amorphous state. The processes of neutron and ion particle bombardment

irradiation, electron beam radiation, ion implantation, and ion beam mixing are all based on

the same concept.

In the reaction of solid-state interdiffusion, atoms in different metal layer can diffuse

through the interface to reach amorphous state under appropriate heat treatment. However,

when the temperature of heat treatment is not high enough, the intermetallic compound will

be formed.

2-4 Characteristics of amorphous alloys

2-4-1 Glass forming ability (GFA)

In order to make the amorphous alloys extensively applied for industry, it is essential to

understand their natures. Among the factors in developing the metallic glasses, the glass

forming ability (GFA) play a critical role in glass forming. GFA is the key to design and

develop new metallic glasses systems [41], and it is determined by tow parameters, the

critical cooling rate (Rc) and maximum attainable size (Dmax). The formation of glass will be

easier and more controllable when GFA is higher. Thus, larger Dmax or smaller Rc is preferred

to acquire higher GFA. Since Dmax significantly is dependent on process of fabrication, and

Rc is difficult to measure, several models and theories have been proposed for discussing the

reliable criteria for GFA in various metallic glasses systems. These established models are

based on the principles of atomic packing or thermodynamic parameters. Such models as

structural models [42], theory of nearly free electron [43], chemical factors [44], phase

diagrams features [45], model of solid solution [46], and atomic size criterion [47] make it

possible to estimate the GFA of MGs. The kinetics is not taken into consideration [48] in the

approaches above. Due to the difficulty in quantification, these approaches are limited for

applications. Therefore, several GFA criterions that can be easier to quantify have been

developed. Recently, the γ parameter has been viewed as reliable criteria.

2-4-2 Recently developed GFA criterions

In addition to γ parameter, researchers are dedicated to developing new indicators for

describing the GFA of metallic glasses. Fan et al. [49] demonstrate a dimensionless criterion,

γ, represented by Trg(∆Tx/T_g)^a. Chen et al. [50] propose another criterion, δ, expressed by

Tx/(Tl - Tg). Many researchers consider that the criterions should be updated with the

development of new metallic glasses systems because these criterions may be inappropriate

to the new systems. Hence, the GFA criterions must be modified with time. After the previous

argument stated by Lu and Liu [41, 51], a modified γ parameter, defined as γm = (2T_x - T_g)/T_l,

has been proposed by Du et al. [52] in recent years.

The parameter, γm reflects T_g, T_x and T_l based on the measurement of GFA. Since these

relevant factors can be correctly obtained during the formation of metallic glasses, the

parameter, γm displays the best correlation with the GFA, especially for BMGs. Among this

relationship, the statistical correction factor R² of γm can be consistent with various GFA

parameters. The parameter, γm is more reliable than the previous ones. Moreover, it is easier

to acquire exacter experimental data for the related factors by DSC and DTA. Thus, the

parameter, γm become a useful and friendly tools for expressing the GFA of metallic glasses.

2-5 Empirical rules for forming amorphous alloys

The formation of amorphous phase in a binary system has much to do with the specific

compositions around the deep eutectic points. A schematic drawing of binary phase diagram

is shown in Figure 2-2 [53]. For composition 1, the melting state passes liquidus line at a

higher temperature (point a). As the melt is cooled down, it will go through a larger

temperature range. The crystallization and growth may occur before reaching Tg. For

composition 2, the melting state passes liquidus line at a lower temperature (point b). Since

the temperature range between Tl and Tg is small, the thermal energy for crystallization will

become lower. In conclusion, the composition of binary system quenching at the temperature

around the deep eutectic points will lead to a lower Tl, and thus, the GFA of metallic glasses

can be enhanced.

The first amorphous alloys were successfully fabricated around 1960s [34].

Subsequently, various techniques were developed to synthesize BMGs empirically. Gradually,

researchers found that the amorphous alloys could be fabricated at critical cooling rates as

low as 1-100 K/s for specific elemental constituents. With the development of various

amorphous alloy systems in the past decade by Inoue’s group, three basic empirical rules for

BMGs designs with high GFA [7,54-56] have been proposed. These empirical rules are listed

as below: (1) multicomponent systems composed of at least three elements, (2) difference in

atomic size more than 12% among the main constituent elements, and (3) negative heats of

mixing among the main constituent elements.

The difference among atomic sizes for more than three elements system will hinder the

motion when the system is quenched from melt to a solid state. As long as the difference in

size among the main constituent elements is larger than 12%, the GFA of amorphous alloys

increase with increasing categories in multicomponent systems. After quenching, the

short-range-ordered structure will be formed in a high dense packing arrangement. In

Hume-Rothery criterion [53], the difference in atomic size to form solid solution is less than

15%.

On the basis of the famous thermodynamic equation ∆G = ∆H^mix – T∆S^mix, the heat of

mixing, ∆H^mix represents atomic bonding ability between two atoms. Larger negative ∆H^mix

stands for weaker the atomic bonding ability of the same atoms. When the melt is quenched

with a larger negative ∆H^mix, usually ranged from -20 to -30 kJ/mol, it will be easier for the

distinct atoms to be bound together to form amorphous alloys. According to the Miedma’s

macroscopic model, the mixing enthalpy values of various ternary amorphous systems and

three binary subsystems can be calculated, as listed in Table 2-1 [57].

The nucleation of crystalline phase and the rearrangement of atoms will be suppressed

by following the above-mentioned empirical to fabricate amorphous alloys due to a higher

liquid/solid interfacial energy. The lower atomic diffusivity and higher viscosity result in the

difficulty in the growth of crystalline phase. These mechanisms for forming BMGs shown in

Figure 2-3 [7] consist in three characteristics for a particular liquid structure.

2-6 Physical vapor deposition

2-6-1 Introduction of sputtering

Sputtering deposition is a physical process involved in ion bombardment and

momentum transfer to fabricate thin films. In the deposition process, the incident particles

with high energy impact the surface of the target, resulting in the bond breaking and atoms

dislodging. It is assumed that elastic collision occurs when the incident particles hit the target

due to almost no loss in momentum transfer.

Generally, sputtering deposition should be operated in: (1) a good vacuum environment

(<10^-5 torr); (2) a low pressure gas environment. When the sputtered particles emit from the

target to the substrate, no gas phase collision happens in a low pressure gas environment.

Once any particle starts hitting the substrate surface with sufficient energy to break bond and

dislodge atoms, the sputtering process occurs. The parameter of sputtering yield is the ratio of

atom sputtered to the number of incident particles with sufficient energy [58], and it is closely

related to the mass of the bombarding particle as well as its energy.

The illustration in Figure 2-4 [58] shows that when the bombarding particles collide the

surface, they penetrate in ten near-surface region. More than 95% transferred energy

dissipates in form of heat near the surface region. Similar to elastic collision, some

bombarding particles show are reflected as high energy neutrals, while some behaving as

complete non-elastic collision are implanted in the surface.

The diluted Ar gas is poured into the ultra-high vacuum chamber. The gas Ar molecules

will be ionized to Ar⁺ as long as bias between anode and cathode becomes large enough. The

incident particle, which is also called plasma, is produced by glow discharge. The electrons

produced at the beginning of glow discharge are called primary electrons. The primary

electrons are accelerated by the electric field. When discharge occurs, the primary electrons

collide with gas molecules, resulting in the generation of the positive ions. After that, the

positive ions bombard the cathode surface, and thus, the secondary electrons are generated.

With the secondary electrons, the efficiency of gas ionization will be increased, and the

self-sustained discharge [59] will be generated.

Sputtering deposition provides an easy way to control the film thickness and refine the

surface roughness. Moreover, the excellent uniformity of film over large areas can be

acquired by employing the sputtering techniques. Hence, sputtering deposition has been

extensively used for surface refinement in industry.

2-6-2 DC and RF sputtering

In DC (direct current) diode sputtering, the electrons are emitted and accelerated away

from the cathode. However, it is not an efficient way to sustain the discharge. By controlling

the magnitude and the arrangement of the magnetic field appropriately, the electrons can be

deflected to stay near the surface of the target. The high density plasma is created by the high

flux electrons. The ions are extracted from the plasma to the sputtered target producing a

magnetron sputtering configuration. In RF (radio frequency) sputtering, the power supply

works at high frequency. Different from DC sputtering, RF sputtering can handle insulating

materials even if the sputtering yield is low. In a word, RF sputtering can provide stable

power to dielectric targets. The major advantage both the DC and RF magnetron sputtering

configuration possess is no energy loss attributed by physical and charge-exchange collisions.

Therefore, DC and RF magnetron sputtering configuration exhibit a higher sputtering rate

with a lower potential on the target compared with the DC diode configuration.

2-6-3 Nucleation and growth of thin films

Many properties of film, such as grain size, surface morphology, and film density are

dominated by the nucleation and the growth of the film. In terms of film growth, several

essential aspect are listed below [58]: (1) substrate surface roughness; (2) surface temperature;

(3) adatom surface mobility; (4) geometrical showing effects (angle-of-incidence effects);

and (5) reaction and mass transport during deposition such as segregation effects and void

formation.

As thin film grows, the substrate surface morphology changes due to the faster growth

of some features or crystallographic planes. Hence, the surface roughness will be increased.

The adatom surface mobility increases with increasing surface temperature. As long as the

flux of deposition on the substrate exceeds that of leaving, the nucleation and growth of the

film will start. The energy absorption between the adatoms and substrate dominates the

nucleation process. In addition, the surface diffusion energy affects the binding and migration

of the deposited atoms [60]. Indeed, the migration of atoms is not randomly distributed. It is

related to crystallographic directions and surface topography of the substrate. Besides, atoms

mobility induced by temperature can also affect the migration. Basically, the three common

modes of film contain island (or Volmer-Weber) mode, layer (or Frank-Vander Merwe)

mode, and Stranski-Krastanov mode [61]. These modes are illustrated schematically in

Figure 2-5 [61].

The island mode results from the nucleation of the most instable cluster on the substrate.

The atoms and molecules are bound more tightly than the atoms and substrate. Moreover, the

growth mechanism leads to the three dimension island. Such metallic glasses on insulators as

alkali halide crystals, graphic, and mica substrates can be classified in this mode. For layer

mode, the film grows layer by layer in two dimensions when the binding between coating

atoms to substrate is stronger than that between coating atoms. In the beginning, the atoms

condense as a single monolayer and no energy barrier is there for nucleation. The growth

mode is common especially for heteroepitaxial thin film growth in

semiconductor-semiconductor systems and some metal-metal systems. The mechanism of

Stranski-Krastanov mode is the combination of the island mode and the layer mode.

2-6-4 Growth of amorphous films

Amorphous films with short-range order can be fabricated by several techniques [69]: (1)

deposition of a natural glassy material; (2) deposition of complex metal alloys; (3) deposition

at low temperatures that make the mobility insufficient to form crystalline structures; (4) ion

bombardment during deposition; (5) ion bombardment of films after deposition; and (6)

deposition of materials, whose bonds are partially saturated by hydrogen such as a-Si:H,

a-C:H and a-B:H [62, 63].

2-7 Properties of TFMGs

2-7-1 Thermal properties

As mentioned above, metallic glasses display special thermal properties such as broad

SCLR. At the SCLR, most metallic glasses show a viscous flow. The formability of MGs can

be improved due to the viscous flow. This thermal property makes TFMGs be used as the

three-dimensional MEMS. In order to measure the thermal properties exactly, the working

temperature should be confirmed. After that, heat treatment such as annealing process should

be carried out. Thus, the thermal properties and SCLR of TFMGs are usually measured by

differential scanning calorimetry (DSC).

From the thermography trace of DSC result (Figures 2-6 and 2-7) [15, 64], the glass

transition temperature (Tg), crystallization temperature (Tx) and SCLR (∆T^x = Tx - Tg) can be

determined according to the changes of the slops. For the Pd-TFMG, the Tg, Tx and ∆T^x are

determined to be 637 K, 669 K and 32 K, respectively. For Zr-TFMGs, its Tg, Tx and ∆T^x are

643 K, 713 K and 70 K, separately. Because Zr-TFMG exhibits an endothermic reaction, the

slop changes due to glass transition at about 643 K. Moreover, an exothermic reaction is

owing to crystallization at about 713 K, as shown in Figure 2-8 [64].

In the time-temperature-transformation (TTT) diagrams which also conducted by using

DSC, the thermal stability of the THMGs can be represented. From two typical TTT

diagrams of TFMGs, as shown in Figures 2-8 and 2-9 [64], the time till the onset of

crystallization and amorphization can be observed. Besides, the allowable interval of the

temperature and the heating time can also be ensured for the micro-forming process and

annealing heat treatment.

2-7-2 Mechanical properties

With the development of BMGs, their mechanical properties of have been extensively

studied for a long time. The superior mechanical properties of BMGs are the reasons why

they have a variety of applications. From the view of structure, BMGs display the dense

packing and randomly atomic arrangement configuration. Thus, the displacements of

amorphous alloys are limited when an external stress is applied. Without dislocation

mechanisms for plastic deformation, the amorphous alloys therefore exhibit some unique

properties such as high strength, hardness, Young’s modulus, and good fracture toughness.

However, these superior mechanical properties as mentioned above are not completely

performed in the TFMGs for some reasons. For instance, because the sample sizes of TFMGs

are different from those of the BMGs, the testing methods for BMGs are difficult to directly

performed on TFMGs. Therefore, the properties of TFMGs should be measured by the

sensitive instruments and careful methods. Furthermore, the testing of bulk material exhibits

the average properties over a large section. However, the testing of thin film material displays

the localized and specific properties in a small scale due to the limit of measurements. For the

reasons above, the microtester or sensitive instrument, such as nanoindenter, will be a safe

way to analyze the TFMGs.

For example, the Zr-based TFMGs with micro-scale thicknesses on silicon substrates are

analyzed by nanoindentation techniques. The nanoindentation-hardenss-measurement results

of the as-deposited and the annealed films as a function of composition on the ZrxCu(1-x)

TFMGs are shown in Figure 2-10 [65]. It is shown that the moderate Zr concentrations (45 –

65 at %) would result in relatively high hardness (~7 GPa) and Young’s modulus (~125 GPa)

for the annealed films. In addition, the relationship between Vickers hardness (Hv) and

Young’s modulus is also shown in Figure 2-11 [7]. In comparison with these two types of the

Zr-based metallic glasses, both of the the hardness and Young’s modulus of the Zr-based

TFMGs are slightly higher than those of bulk BMGs.

In terms of the deformation mechanism, the amorphous alloys are quite different from

the crystalline alloys. In crystalline alloys, once the external shear stress exceeds the critical

resolved shear stress on a slip plane, the plastic deformation will occur. Meanwhile, the

dislocations are formed, and then start to glide along the slip direction on a slip plane. Thus,

in crystalline materials, the mechanical properties have much to do with the crystal and

electrical structures. On the other hand, no slip system exists in amorphous alloys due to the

lack of long-range-order structure. Their plastic deformation is operated by the mechanisms

of shear bands.

The amorphous alloys display short-range-order structures. The small spaces called free

volumes exist in the structure, as shown in Figure 2-12. Recently, several theories have been

proposed to describe the heterogeneous plasticity in MGs. The plastic flow in amorphous

alloy is viewed as a diffusion-like process. This is involved in the stress-induced

self-assembled of smaller unit of plasticity called flow defects or shear transformation zones

(STZs) [66, 67]. STZs are the fundamental unit of plasticity in a form of a small cluster of

random close-packed atoms. These flow defects or STZs are associated with about 10 to 50

atoms in open spaces or free volume sites, which are distributed over the amorphous structure.

The illustration of STZ deformation in two-dimensional space is shown in Figure 2-13. The

dominated mechanisms in glass structure and deformation are generally localized to this mere

fine scale. In other words, it is usually assumed that the plastic deformation in amorphous

alloys is dominated by the amount and distribution of the free volume.

Moreover, the moving of STZ is attributed by the releasing of adiabatic heat near the

STZ [68]. When more and more STZs start moving, the huge adiabatic accumulates, resulting

in the raise of the temperature around the shear plane. After the temperature reach the glass

transition temperature, a fluid region or fluid layer will be formed in the shear plane, as

shown in Figure 2-14. The fluid region will possess more free volume sites for STZs. Thus,

the amorphous alloys will be deformed more easily.

2-7-3 Chemical properties

In addition to good mechanical properties, it has been discovered that the homogeneous

structures of single phase make BMGs display high corrosion resistance. That is because the

amorphous structures are lack of grain boundaries, dislocations, and other crystal defects.

Moreover, the corrosion resistance can further improved by adding some corrosive solutes.

Take the Zr-Al-Ni-Cu amorphous alloys system for instance, when elements of corrosive

solutes such as Nb, Ta, Ti, and Cr [69] are added, the corrosion resistance can be enhanced.

Recently, some researches indicate that the Pd-based and Fe-based [70] amorphous alloys

also exhibit high corrosion resistance. Thus, they can be used for practical corrosion

resistance materials.

Chapter 3 Experimental Procedures

The Pd77Cu6Si17 (in atomic percent) alloy films were selected to be deposited on the

polished AZ31 substrates by magnetron sputtering, denoted as PCS. The structural

characteristics of thin films were examined by X-ray diffraction (XRD). The film

compositions were confirmed by scanning electron microscopy (SEM) with energy dispersive

X-ray spectrometer (EDS). The mechanical properties of PCS were evaluated by

microhardness and nanoindentation. The scratch test was operated to understand the films

deformation in different depths. In addition, the wear coefficient was calculated to compare

the differences among AZ31, and PCS. The flow chart of the experimental procedures is

shown in Figure 3-1.

3-1 Materials

The AZ31 billets were purchased from the CDN company, Deltabc, Canada. This

as-received alloy was fabricated through semi-continuous casting and showed the form of

billet with 178 mm in diameter and 300 mm in length. The as-received AZ31 billets were cut

into the small pieces measuring 8 mm in width, 8 mm in length, and 3 mm in thickness. The

detailed chemical composition of AZ31 Mg alloy is listed in Table 3-1. The Pd77.5Cu6Si16.5

alloy target with a purity of 99.99% was purchased from Gredmann Taiwan, corp.

3-2 Sample preparation

3-2-1 Substrate preparation

In this study, the AZ31 plate was selected as the substrate. To make the surface smoother,

the substrates were grinded with abrasive paper and polished with diamond paste. To avoid

impurities and greasy dirt contaminating the substrates, the following steps for cleaning the

surfaces were adopted. An ultrasonic cleaner was used to clean the substrates in alcohol for

10 minutes, followed by the acetone rinsing for 10 minutes. Finally, air was employed to dry

the substrates.

3-2-2 Film preparation

The Pd77Cu6Si17 alloy (PCS) coatings were deposited on the smooth AZ31B substrates

by a single-gun magnetron sputtering system using Pd77.5Cu6Si16.5 alloy target of 50.8 mm (2

inches) in diameter.

The main chamber was pumped down to a base pressure less than 1×10^-6torr by a

cryo-pump. The Pd77Cu6Si17 alloy target was located on the DC gun with a power of 200 watt,

and pure argon atmosphere as the working gas was maintained at the rate of 30 standard

cubic centimeters per minute (sccm). The substrate was rotated at an average speed of 10 rpm

for ensuring uniform distribution of film thickness. The working distance between the holder

and the DC gun is 120 mm. With appropriate time for sputtering, PCS thin films were

deposited on the AZ31 substrate with various thicknesses of 30, 50, 100, 200, 300, 500, 1000

and 2000 nm. These specimens are denoted as PCS-30, PCS-50, PCS-100, PCS-200,

PCS-300, PCS-500, PCS-1000, and PCS-2000, respectively.

3-3 Property measurements and analyses

The film structure was characterized by Bede D1 HR-XRD grazing incidence in-plane

X-ray diffraction (XRD), and the film compositions were confirmed by JEOL-6400 scanning

electron microscopy (SEM) linked with energy dispersive X-ray spectrometry (EDS). The

glassy nature of as-deposited Pd77Cu6Si17 thin films was characterized by Bede D1 HR-XRD

grazing incidence in-plane X-ray diffraction (XRD), and the film compositions were

confirmed by JEOL-6400 scanning electron microscopy (SEM) linked with energy dispersive

X-ray spectrometry (EDS). The films deposited on P-type (100) silicon wafers and deposited

on AZ31 Mg alloys were both examined by XRD.

3-4 Microhardness tests

The difference in hardness between the AZ31 Mg alloys and AZ31 Mg with the

Pd77Cu6Si17 amorphous coatings was examined by using microhardness test. The

microhardness tests of samples were conducted using a SHIMADZU HMV-2000 Vickers

Microhardness Tester with various loads from 10 to 100 g for a fixed duration time of 10

seconds. The hardness values of each sample were averaged from 30 datum points chosen

randomly. The microhardness data was analyzed to inspect the hardness change.

3-5 Nanoindentation tests

For comparison of hardness with different testing instruments, all the coating samples were

also measured by the MTS XP nanoindenter system via the continuous stiffness measurement

(CSM) mode. The nanoindentation testing was executed with a maximum load of 500 mN

(Figure 3-2). The Berkovich tip was a three sided pyramid with a geometrical symmetry.

There existed a flat profile with a centerline-to-face angle of 65.3 degrees. Nanoindenter is

very sensitive to the external environment and the flat condition of sample surface. It would

be necessary to keep clean and dry surface of thin films before nanoindentation test. In the