Mechanical Properties of 7075 Aluminium Matrix Composites Reinforced by Nanometric Silicon Carbide Particulates

By Zheng Ren

A thesis submitted for the Degree of Master of Engineering

School of Materials Science and Engineering Faculty of Science University of New South Wales

September 2007

PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Ren

First name: Zheng Other name/s:

Abbreviation for degree as given in the University calendar: ME

School: Materials Science and Engineering Faculty: Faculty of Science

Title: Mechanical Properties of 7075 Aluminium Matrix Composites Reinforced by Nanometric Silicon Carbide Particulates

Abstract 350 words maximum: (PLEASE TYPE)

Aluminium composites reinforced by particles have received considerable attention because of their superior mechanical properties over monolithic aluminum matrix. Over the last ten years, nanocomposites with nano-sized reinforcements have become a revolutionary progress for composites because they have different strengthening mechanisms as compared to that in composites with micro-sized reinforcements. Consequently novel properties can be expected from the nanometric particulate reinforced composites. The aim of this project was to fabricate SiC (50nm)/7075 aluminium composites via a modified powder metallurgy and extrusion route. Ageing treatment was used to increase the strength of the composites and mechanical tests, including tensile test and abrasive wear test, were performed. The effects of nanometric silicon carbide particulates to the ageing behaviours and mechanical properties of the composites have been studied by optical metallography, scanning electron microscopy and transmission electron microscopy. It was found that the dispersion of nanometric silicon carbide was not homogeneous, but tended to disperse along grain boundaries. Clustering of these nano-reinforcements was also found within the grains. This was particular true when the amount of nano-reinforcement increased to 5%. Compared with the monolithic 7075 alloy, the 1 vol.% SiC (50nm)/7075 aluminium had a higher strength because of effective dislocation pinnings by the reinforcements, while 5% SiC (50nm)/7075 had a much lower strength and ductility because of severe aggregation of nanometric particulates. Nanometric silicon carbide was not as effective as the micro ones in improving abrasive wear resistance of aluminium, this was because of micro-cracking in the aggregation and relatively large abrasive grit. In summary, the addition of a small amount of SiC nanoreinforcements has a high potential to further strengthen 7xxx aluminium alloy. However, the clustering of reinforcements in the matrix will detrimentally affect the strength and ductility of the alloy. The wear resistance of nanometric particulate reinforced composites was inferior to those with micrometric reinforcements. It is suggested that by improving the dispersion of nanometric reinforcements, as well as putting in reinforcememts with different sizes, the mechanical properties and wear resistance can both be increased.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

ORIGINALITY STATEMENT

„I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.‟

Signed …..……………………………………………......

Date……..……………………………………………......

i COPYRIGHT STATEMENT

„I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.' Signed ……………………………………………...... Date ……………………………………………......

AUTHENTICITY STATEMENT

„I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.‟ Signed ……………………………………………...... Date ……………………………………………......

ii ACKNOWLDGE

My deepest gratitude and thanks go to Dr Sammy Chan for his patient supervision and invaluable advice during the project. To all members of Nanomaterils for Hydrogen

Storage Group at University of New South Wales, Vincent Supratman, Thanh Lu ,Fred

Yuan, Ben Tsai, Vanessa Li, Dr Q. S. Song and Dr Z. M. Wang.

Thank you to Dr Guangqing Zhang and Mr John Budden for their helpful assistance in heat treatment and metallographic preparation. Also my thanks go to Ms Veira

Piegerova and Mr Barry Searle for their expert instruction on scanning electron microscopy. The technical support provided by the staff in the School of Materials

Science and Engineering, UNSW, are also acknowledged.

Most specially, my great thanks must go to my parents, Mr Sihai Ren and Ms Guanglan

Zhu, for their endless support and amazing encouragement. I love you.

iii ABSTRACT

Consequently novel properties can be expected from the nanometric particulate reinforced composites.

The aim of this project was to fabricate SiC (50nm)/7075 aluminium composites via a modified powder metallurgy and extrusion route. Ageing treatment was used to increase the strength of the composites and mechanical tests, including tensile test and abrasive wear test, were performed. The effects of nanometric silicon carbide particulates to the ageing behaviours and mechanical properties of the composites have been studied by optical metallography, scanning electron microscopy and transmission electron microscopy.

It was found that the dispersion of nanometric silicon carbide was not homogeneous, but tended to disperse along grain boundaries. Clustering of these nano-reinforcements was also found within the grains. This was particular true when the amount of nano-reinforcement increased to 5%. Compared with the monolithic 7075 alloy, the 1 vol.% SiC (50nm)/7075 aluminium had a higher strength because of effective dislocation pinnings by the reinforcements, while 5% SiC (50nm)/7075 had a much lower strength and ductility because of severe aggregation of nanometric particulates.

Nanometric silicon carbide was not as effective as the micro ones in improving abrasive wear resistance of aluminium, this was because of micro-cracking in the aggregation

iv and relatively large abrasive grit.

In summary, the addition of a small amount of SiC nanoreinforcements has a high potential to further strengthen 7xxx aluminium alloy. However, the clustering of reinforcements in the matrix will detrimentally affect the strength and ductility of the alloy. The wear resistance of nanometric particulate reinforced composites was inferior to those with micrometric reinforcements. It is suggested that by improving the dispersion of nanometric reinforcements, as well as putting in reinforcememts with different sizes, the mechanical properties and wear resistance can both be increased.

v TABLE OF CONTENTS

ORIGINALITY STATEMENT ...... i

AUTHENTICITY STATEMENT ...... ii

ACKNOWLDGE ...... iii

ABSTRACT ...... iv

TABLE OF CONTENTS ...... vi

LIST OF FIGURES ...... viii

LIST OF TABLES...... xi

LIST OF ABBREVIATIONS ...... xii

Chapter 1 INTRODUCTION ...... 1

Chapter 2 LITERATURE REVIEW ...... 3

2.1 Fabrication of Aluminum Matrix Composites ...... 3

2.1.1 Powder Metallurgy ...... 3

2.1.2 Deformation Processing of MMC ...... 5

2.2 Ageing Behaviour of MMC ...... 7

2.2.1 Ageing Characteristics of Al Alloys ...... 7

2.2.2 Accelerated Ageing of MMC ...... 13

2.3 Mechanical Properties of AMC ...... 15

2.3.1 Tensile Properties ...... 15

2.3.2 Creep of MMC ...... 19

2.4 Strengthening Mechanisms ...... 22

2.4.1 Grain Strengthening ...... 22

2.4.2 Sub-structure Strengthening ...... 22

2.4.3 Quench Strengthening ...... 22

2.4.4 Orowan Strengthening ...... 23

2.4.5 Work Hardening ...... 23

vi Chapter 3 EXPERIMENTAL ...... 25

3.1 Introduction ...... 25

3.2 Manufacturing ...... 25

3.2.1 Starting Materials ...... 25

3.2.2 Powder Metallurgy process ...... 27

3.3 Heat Treatment and Hardness Test ...... 27

3.4 Tensile Testing ...... 28

3.5 Wear Test ...... 29

3.6 Optical and scanning electron Microscopies ...... 30

3.7 TEM ...... 30

3.7.1 Sample Preparation ...... 30

3.7.2 Transmission Electron Microscopy ...... 31

Chapter 4 RESULTS AND DISCUSSION ...... 32

4.1 Microstructures of 7075 Al and 7075 Al Composites ...... 32

4.1.1 Optical Metallography of As-extruded Materials ...... 32

4.1.2 Distribution and Segregation of Nanoparticles ...... 36

4.1.3 Technology for Homogeneous Dispersion of Nanoreinforcements in Metals ...... 43

4.2 Aging Behaviours ...... 44

4.3 Room-temperature Tensile Properties ...... 49

4.4 Abrasive Wear Properties ...... 63

CHAPTER 5 CONCLUTIONS AND FUTURE WORK ...... 73

5.1 CONCLUTIONS ...... 73

5.2 FUTURE WORK ...... 74

Reference ...... 75

vii LIST OF FIGURES

NO Page

2.1 conventional P/M process [4] 4

2.2 Schematic diagram showing the principle of reciprocating extrusion 7

2.3 The distribution of particles in two composites 7

2.4 Al-Cu phase diagram showing the metastable GP zone, θ” and θ‟ Solvuses 9

2.5 Representation of the variation in GP zone size distribution with aging 10

time (t1

2.6 Schematic showing the variation of Strength (or hardness) with aging time 10

(at a fixed aging temperature) or precipitate size in an Al-Cu alloy [12]

2.7 TEM of the Al–Zn–Mg–Cu alloy [15] 13

2.8 Representation of profile of vacancy concentration adjacent to 13

grain boundary in quenched alloys[16]

2.9 Variation of matrix microhardness as a function of aging time at 177°C 13

for powder metallurgy processed 2124 aluminum alloy with and without

13.2vol. % SiC [16].

2.10 Tensile properties of the aluminium and composites [1] 17

2.11 TEM micrographs of (a) 1vol% Al2O3/Al (b) 5vol% Al2O3/Al [1] 17

2.12 Creep behavior of different materials at the same applied stress 20

and temperature [32]

2.13 TEM micrographs showing the particulates and 20

dislocations of 7775 Al

3.1 Morphology of nanometric silicon carbide particulates 26

3.2 P/M fabrication process for aluminium composites 28

viii 3.3 Schematic diagram of the pin-on-disk wear test apparatus 29

4.1 OM of as-extruded 7075 Al composites 33

4.2 OM of as-extruded 1vol. % SiC (50nm)/7075 Al composites 34

4.3 OM of as-extruded 5vol. % SiC (50nm)/7075 Al composites 35

4.4 SEM micrograph of as-extruded 5vol. % SiC (50nm)/7075 Al composites 37

4.5 EDS of agglomerations region and matrix region in Fig 4.4 (b) 38

4.6 SEM micrograph of an agglomeration region of 38

5vol. % SiC (50nm)/7075 aluminium composites at high magnification

4.7 TEM micrograph of peak-aged 7075 AMC 39

4.8 EDS of different regions in 5vol. % SiC (50nm)/7075 Aluminium 40

composites in Fig 4.7 (c)

4.9 TEM micrographs of AA7075 and AA7075 MMC in 41

as-received condition [48]

4.10 Schematic sketch of the milled composite after extrusion [49] 42

4.11 Bright field TEM image of the hot rolled Al-5083/SiC composite 42

4.12 As quenched hardness of 7075 aluminium and its composites 45

4.13 Hardness-aging curve of 7075 aluminium and its composites 46

4.14 Hardness variations of Al 7075 and Al 7075/TiCp s 47

amples with time at 120oC [55]

4.15 Comparison of the heat flow of precipitation reactions in 7775 Al 47

alloy and its composites with different amount of nano-SiC particles [62].

4.16 TEM graphs of the peak aged composites [63] 48

4.17 Stress-strain curve of 7075 aluminium and its composites 50

4.18 OM of fracture surface 51

4.19 OM of longitudinal cross section 52

ix 4.20 OM and SEM of longitudinal cross section of 53

5vol % SiC (50nm)/7075 AMC

4.21 SEM micrograph of fracture surface of 7075 aluminium 54

4.22 SEM micrograph of fracture surface of 7075 aluminium at large magnification 55

4.23 SEM micrograph of fracture surface of 1vol. % SiC (50nm)/7075 AMC 56

4.24 SEM micrograph of large crack in 1 vol. % SiC (50nm)/7075 AMC 57

4.25 SEM micrograph of a dimple containing clusters in 57

1vol. % SiC (50nm)/7075 AMC

4.26 SEM micrograph of fracture surface of 5 vol. % SiC (50nm)/7075 AMC 58

4.27 SEM micrograph shown nanometric particles and clusters 59 on the fracture surfaces of 5vol. % SiC(50nm)/7075AMC

4.28 Bimodal distributions of dimples in 5 vol. % SiC (50nm)/7075 AMC 60

4.29 TEM micrograph showing dislocations after tensile test 61

Fig4.30 Schematic sketch of composites microstructure reinforced 62

by nano-sized particles

4.31 Peak-aged hardness of 7075 aluminium and its composites 63

4.32 Abrasive wear rate of peak aged 7075 aluminium and its composites 63

4.33 Worn debris 65

4.34 SEM micrograph of worn debris in large magnification 67

4.35 Mode of materials removal form the groove by a process of cutting [69] 69

4.36 SEM micrograph of worn surfaces of 7075 aluminium alloy 70

4.37 SEM micrograph of worn surfaces of 71

1vol. % SiC(50nm)/7075 aluminium composites

4.38 SEM micrograph of worn surfaces of 5 vol. % SiC(50nm)/7075 72

aluminium composites

x LIST OF TABLES

NO PAGE

2.1 Probable precipitation processes in commercial aluminum alloys [13] 12

2.2 Summary of work on aging kinetics of particle-reinforced MMCs [16] 14

2.3 Published room temperature tensile properties of aluminum-based MMC 16

2.4 Tensile properties of nanocomposites [23] 18

3.1 Chemical composition of matrix alloy (wt. %) 25

4.1 Tensile properties of 7075 aluminium and its composites 50

xi LIST OF ABBREVIATIONS

MMC: metal matrix composite

AMC: aluminium matrix composite

P/M: powder metallurgy

CNTs: carbon nanotubes

SiCP: silicon carbide particulate

PFZs: precipitate-free zones

DSC: differential scanning calorimetry

OM: optical microscope

SEM: scanning electron microscope

TEM: transmission electron microscope

EDS: energy dispersive X-ray spectroscopy

CIP: cold isostatic pressing

UTS: ultimate tensile strength

YS: yield strength

xii Chapter 1 INTRODUCTION

Aluminium-based discontinuously reinforced metal matrix composites (MMC) have received considerable attention because of their improved strength, high modulus and increased wear resistance over conventional aluminium alloys. The size of reinforcements in commercial MMC generally ranges from a few micrometers to several hundred micrometers. Because of fabrication difficulties only recently reinforcements with nanometric size have been used. It was reported that the yield strength of pure aluminium was doubled by adding only 4vol. % Al2O3 particles [1], and the tensile strength and Young‟s modulus of 2024 Al were increased by 35.7% and

41.3% respectively by adding only 1.0wt.% carbon nanotubes (CNTs) in the matrix [2].

However, the use of advanced composites has not been realized because of the processing difficulties and a lack of understanding of the role of nano-particulates on the resulting mechanical properties.

Previous research indicated that nano-sized alumina particles tended to reduce alloy strength by attracting certain elements. Al-SiC system was mostly studied and it is a very stable system. 7xxx aluminium alloy is the strongest heat-treatable aluminium series; hence in this study 7xxx aluminium alloy and nano-sized silicon carbide particles were used to in order to gain maximum strength. The composites were fabricated via

Powder Metallurgy (P/M) method and the mechanical properties have been examined.

The role of the nanometric particles on dislocation motion and ageing behaviour has been determined.

This thesis is divided into five chapters. Chapter 1 gives an introduction and objectives of this work. It also outlines the structure of the thesis. Chapter 2 reviews previous

1 researches related to this study including manufacturing technology, mechanical properties and strengthening mechanisms. Chapter 3 describes details of all the experiments involved in this work. Chapter 4 reports the results and discussions of microstructures, ageing behaviours, tensile properties and abrasive wear properties.

Conclusions are given in Chapter 5 as well as recommendations of future work.

2 Chapter 2 LITERATURE REVIEW

2.1 Fabrication of Aluminum Matrix Composites

A metal matrix composites (MMC) is defined as a material that consists of at least two constituent parts, the matrix being metal, the other may be a different metal or ceramics.

They are bonded together along the interface in the composites.

Processing of MMC can be broadly divided into two categories of fabrication technique: solid state and liquid state. The liquid state processing is generally less expensive and easier to handle, and the composites can be produced in variable shapes, using techniques already developed in the casting industry for monolithic metals. However the technical difficulties related to liquid state processing include reinforcement segregation and clustering, detrimental interfacial chemical reaction, high localized residual porosity and poor interfacial bonding which degrade the properties. Meanwhile powder metallurgy processing can avoid strong interfacial reaction and minimize the undesired reaction between the matrix and the reinforcement because generally a lower manufacturing temperature is used in powder metallurgy. The content and distribution of reinforcement, as well as the microstructure of the matrix can be controlled relatively easy. Hence the P/M products normally have superior properties over that of their cast counterparts.

2.1.1 Powder Metallurgy

Powder Metallurgy is the most common method for fabricating particulate metal-ceramic composites. It usually involves mixing of powders of the matrix alloy with the reinforcing particles, followed by compacting and solid state sintering (Fig 2.1). 3 It is very important that all particles are homogeneously distributed in order to obtain a uniform microstructure. When whiskers are used as reinforcement, small particles for the matrix alloys are required for the improvement of the packing effect and to obtain a good dispersion of the fibers in the matrix [3].

Although most powder consolidation and processing work is carried out below the matrix solidus temperature, sometimes it becomes necessary to maintain the consolidation temperature slightly above the solidus to minimize deformation stresses and to avoid whisker damage. In liquid-phase sintering, powder consolidation may be achieved without the use of any external pressure because a low melting phase pulls solid particles together via the force of surface tension. The higher melting temperature phase in this process should be slightly soluble in this liquid [4].

Fig 2.1 conventional P/M process [4]

4 2.1.2 Deformation Processing of MMC

Secondary processing of the discontinuously reinforced composites leads to break up of particle (or whisker) agglomerates, reduction or elimination of porosity, and improved particle to particle bonding, all of which tend to improve the mechanical properties of these materials which involve high temperatures and large stain deformation [4,5]. a. Extrusion

The most common secondary processing methods is extrusion, which is performed at a temperature that enables strain-rate sensitivity (m = dlogσ/log ) to reach a relatively high value. Typically, the process is carried out at a high strain rate (1-100s-1) and primarily involves dislocation creep deformation of the matrix. Thus, the value of m does not exceed 0.3 during such an operation and can be as low as 0.1 for the particulate composites. Nevertheless, the highest possible values of m are sought in order to obtain the most uniform flow and minimize tendencies to cracking. Apart from improving the homogeneity of the product, extrusion can produce net-shape product forms in large lengths [5].

The microstructure of the as-extruded materials is generally heterogeneous. The local volume fraction of SiC particulates in the clusters sometimes is as high as 50 to 60% and there are also areas that contain no particles [6]. Stain-rate effects during the deformation process have a strong effect on the break-up of particle clusters. Because the particle clusters act as harder regions than particle-free regions, strain partitioning takes place during flow. Thus at low strain rates, the imposed deformation is accommodated primarily by the flow of the particle-free regions. Because the stain-rate sensitivity of the matrix alloy exceeds that of the clustered region, a high-applied strain rate raises matrix flow stress preferentially thereby forcing deformation to take place in the particle clustered regions and leads to shearing of clustered regions. Thus, at high 5 strain rates, a reasonable amount of strain within the cluster can be expected.

Conversely, the stresses and strains in the cluster are so small at low strain rates that its break-up is insufficient even when the overall strain on the composites is 100 time that of the cluster [6]. b. Reciprocating Extrusion

The clustering of reinforcement can be eliminated by reciprocating extrusion. A reciprocating extrusion machine consists of two heated cylindrical containers, two rams and a die between them [7] as shown in Fig 2.2. Two billets of the same type are first placed into one container and heated to the extrusion temperature. Then it is extruded by the ram A into the pre-heated container B with a certain extrusion ratio. At the same time, ram B exerts a back pressure so that the work piece can recover its shape in container B. After the pass, the procedure is repeated but with the billet being extruded from B back to A. The billet can be extruded repeatedly for a predetermined number of times. After the final cycle the work piece will be directed extruded through the die with the opposite ram removed.

6061Al-0.3μm Al2O3 composites were successfully produced by reciprocating extrusion.

Fig 2.3 shows the SEM micrographs of two composites in the as-extruded condition and the fine Al2O3 particles were distributed uniformly in the matrix. This uniform dispersion of such submicron-sized particles could not be achieved by conventional casting or powder method, since those processes lack the homogenization resulting from kneading [7].

Fig 2.2 Schematic diagram showing the principle of reciprocating extrusion [32]

Fig 2.3 The distribution of particles in two composites:

(a) 10% Al2O3 composites (b) 20% Al2O3 composites [7]

2.2 Ageing Behaviour of MMC

2.2.1 Ageing Characteristics of Al Alloys

Ageing of a heat-treatable aluminium alloy is generally performed through a heat treatment cycle that involves: a) a high temperature solid solution treatment aimed at dissolving the alloying elements and producing a solid solution; b) quenching to obtain a supersaturated solid solution (SSSS) ; c) a subsequent heat treatment that allows controlled decomposition of the SSSS and the 7 formation of uniformly distributed strengthening precipitates [8]. a. Quench-in vacancies

The equilibrium concentration of vacancies increases exponentially with temperature.

Thus the equilibrium vacancy concentration will be relatively high at the solution treatment temperature and much lower at the ageing temperature. However when alloy is rapidly quenched from the high temperature there will be no time for the new equilibrium concentration to be established and the high vacancy concentration becomes quenched-in [9]. Excess vacancies are able to provide heterogeneous nucleation sites.

The quenched-in vacancies are to greatly increase the rate at which precipitating atoms can diffuse at the ageing temperature, which in turn speeds up the process of nucleation and growth of the precipitates. b. Precipitation in Aluminum-Copper Alloys

Although the alloy used in this study is an Al-Zn-Mg alloy, the ageing process can be illustrated by the precipitation in Al-Cu alloys, which is the most studied aluminium alloy for aging. The formation of hardening precipitates can be described from the view of classical thermodynamics, i.e. as a process of decomposition of a single phase into one or more phases, which goes through the stages of nucleation, growth of nuclei and coarsening.

GP Zones

Fig 2.4 shows the Al-rich end of the Al-Cu phase diagram. By quenching the specimen rapidly from the α field there is no time for any transformation to occur. And the solid solution is supersaturated with Cu and there is a driving force for the precipitation of the equilibrium θ phase, CuAl2.

The first precipitate to nucleate is not θ but GP zones. The reason for this can be understood on the basis of the relative activation energy barriers for nucleation. GP

8 zones are fully coherent with the matrix and therefore have a very low interfacial energy, whereas the θ phase has a complex tetragonal crystal structure which can only form with high-energy incoherent interface. In addition the zones minimize their strain energy by choosing a disc-shape perpendicular to the elastically soft <111>directions in the fcc matrix. The zones are about 2 atomic layers thick and 10 nm in diameter .The coherency misfit strain distorts the lattice causing local variations in the intensity of electron diffraction, which in turn shows up as variations in the image intensity [10].

An important concept is that of the GP zones solvus which shown as a metastable line in the equilibrium diagram (Fig 2.4). It defines the upper temperature limit of stability of the GP zones for different composition although its precise location can vary depending upon the concentration of excess vacancies. Solvus lines can also be determined for other metastable precipitates. The distribution of GP zones sizes with ageing time is shown schematically in Fig 2.5. There is strong experimental support for the model proposed by Lorimer and Nicholson whereby GP zones formed below the GP zones solvus temperature can act as nuclei for the next stage in the aging process, usually the intermediate precipitate, providing they have reached a critical size [11].

Fig 2.4 Al-Cu phase diagram showing the metastable GP zone, θ” and θ‟ Solvuses [10]

Fig 2.5 Representation of the variation in GP zone size distribution with aging time

(t1

Fig 2.6 Schematic showing the variation of strength (or hardness) with aging time (at a fixed aging temperature) or precipitate size in an Al-Cu alloy [12]

Transition Phases

The formation of GP zones is usually followed by the precipitation of transition phases.

In the case of Al-Cu alloys the equilibrium θ phase is preceded by θ” and θ‟. The total precipitation process can be written:

αo α1+GP zones α2 +θ‟‟ α3 +θ where αo is the original supersaturated solid solution, α1 is the composition of the matrix in equilibrium with GP zones, α2 the composition in equilibrium with θ‟‟ etc [10]. Table

2.1 shows the more details of precipitation processes in Al-Cu, Al-Mg-Si and Al-Zn-Mg

10 alloy systems.

The shape of the ageing curve in Figure 2.6 can be rationalized as follows. Primarily solid solution hardening is active immediately following quenching from the solutionizing temperature. As Guinier-Preston (GP) zones form during the initial stages of ageing, the strength or hardness of the alloy increase because additional stresses are necessary for dislocations to cut through (shear) the coherent zones during plastic deformation. The hardness increases as the size of the GP zones increases with time, making it even more difficult for the dislocations to shear them. The peak hardness or strength is associated with a critical dispersion of coherent or semicoherent (θ‟) precipitates. With still further increase in aging time, stable equilibrium (θ) precipitates with incoherent interfaces with the matrix begin to form and the dislocation become increasingly more capable of bypassing or looping around the alloy during plastic deformation. This circumventing of particles by dislocations results in the formation of so-called Orowan dislocation loops around particles. As the particles coarsen with time, they loss the effectiveness of pinning the dislocations and a progressively lower strength results as the bowing of dislocation around particles becomes easier with increasing aging time. [9] c. Precipitate-free zones (PFZs)

Most alloys in which precipitation occurs have zones adjacent to grain boundaries which are depleted of precipitate and this is called precipitate-free zones (PFZs). Fig 2.7 shows PFZs of an Al–Zn–Mg–Cu alloy.

These precipitate-free zones formed for two reasons:

 local solute depletion due to precipitate formation in the boundary (typically～

50nm), this causes solute to be drained from the surrounding matrix

 local depletion of vacancies which are important in assisting with nucleation of

11 precipitates

The width of the PFZs is determined by the vacancy concentration as shown in Fig 2.8.

At low temperatures, where the driving force for precipitation is high, the critical vacancy supersaturation is lower and narrower PFZs are formed. High quench rates will also produce narrow PFZs by reducing the width of the vacancy concentration profile.

Similar PFZs can also form at inclusion and dislocations [14].

Table 2.1 Probable precipitation processes in commercial aluminum alloys [13]

Fig 2.7 TEM of the Al–Zn–Mg–Cu alloy [15].

Fig 2.9 Variation of matrix microhardness as a function of aging time at 177°C for powder metallurgy processed 2124 aluminum alloy with and without 13.2vol. % SiC

[16].

2.2.2 Accelerated Ageing of MMC

In a metallic matrix containing brittle fibers, whiskers or particles, the difference in the thermal expansion coefficient, △α, between the matrix and the reinforcement can be so large, that even a small change in temperature will generate thermal residual stresses in the matrix. The matrix can undergo plastic yielding as the magnitude of the residual stresses locally exceeds the yield strength. The consequent development of dislocations

Table 2.2 Summary of work on aging kinetics of particle-reinforced MMCs [16]

in the matrix gives rise to a greater dislocation density in the matrix of the composite than in the unreinforced alloy. Theses dislocations can serve as heterogeneous sites for the nucleation of strengthening precipitates and can provide short circuit diffusion paths for solute atoms. As a result, both the nucleation and growth of precipitates in the matrix can be drastically altered by the presence of the reinforcements [16]. Experimental results show that in a wide variety of precipitation-hardenable aluminum alloys, the alloy with the brittle reinforcements exhibits a significantly shorter ageing time to 14 achieve peak strength than does the matrix alloy at the same ageing temperature. An example of accelerated ageing phenomenon can be show in Fig 2.9 which shows direct comparison of hardness with respect to ageing time for powder metallurgy processed

2124 aluminum alloy and its‟ composites with 13.2% SiC whisker.

Experimental work by a number of researchers has provided clear evidence for occurrence of accelerated ageing and for the precipitation characteristics of a wide range of metal-matrix composites. A summary of work on the ageing characteristics of reinforced metals is provided in Table 2.2. These studies have employed diverse experimental and analytic tools that include transmission electron microscopy (TEM), differential scanning calorimetry (DSC), microhardness measurements, macrohardness and strength measurements of composites, and electrical conductivity measurements

[16].

2.3 Mechanical Properties of AMC

2.3.1 Tensile Properties a. Tensile Properties of Micrometric Particulate-reinforced Aluminium Composites

The most common and commercial particulate composite system is aluminium reinforced with silicon carbide (SiC). The widely used commercial SiC particulates generally have a size ranging from few micrometers to several hundred micrometers.

The composites can be improved in mechanical properties such as yield stress and tensile strength, and also increases in high-temperature properties. Examples of published room-temperature tensile properties are shown in Table 2.3. It is found that yield strength and tensile strength tend to increase, and toughness and ductility decrease with increasing volume fraction of particulates. Tensile properties such as yield strength, tensile strength and ductility generally tend to be improved with decreasing particle size 15 for a given volume fraction of reinforcement. However, it can also be seen that the reinforcement has little effect on high strength 7XXX series alloys. In some cases, the tensile strength actually decreases. b. Tensile Properties of Nanometric Particulate Reinforced Aluminium Matrix

Composites

Table 2.3 Published room temperature tensile properties of aluminum-based MMC [17].

Fig 2.10 Tensile properties of the aluminium and composites [1]

(a) (b)

Fig. 2.11 TEM micrographs of (a) 1vol% Al2O3/Al (b) 5vol% Al2O3/Al [1]

Table 2.4 Tensile properties of nanocomposites [23]

Ma et al. [22] showed that the tensile strength of the 1vol%Si3N4 (15 nm)/Al composite is comparable to that of the 15 vol. % SiCp (3.5 μm)/Al composite, and that yield strength of the former is much higher than that of the latter. Amorphous Si-N-C particulate

(25-30nm) was also used for aluminium composites and all the tensile properties were compared as shown in Table 2.4 [23]. Kang and Chan [17] also reported the same trend for Al reinforced with nanometric Al2O3: a 1vol. % of Al2O3 added to commercial pure aluminium will have the same strengthening effect as that of a 10%SiCp (10μm) composite (Fig 2.10). However, the strengthening effect flattened as the nanometric

Al2O3 added exceeded 4%, and this may be considered as the optimum fraction of nanometric Al2O3 that can be added to the alloy using the conventional P/M route. As the amount of nanometric Al2O3 addition increased, the degree of agglomeration of the particulates increased. This is evident from Fig 2.11 that, nanometric Al2O3 particulates are distributed in the matrix relatively evenly for 1vol% addition, but the degree of Al2O3 segregated along the boundaries rapidly increased for 5% of nano- Al2O3 in Al [17].

18 2.3.2 Creep of MMC a. Creep

The creep behaviours of AMC have been the subject of investigations recently. Klocey al.

[24] showed that the creep apparent stress exponent and activation energy of a 2024-Al alloy produced by P/M process was higher than that of the alloy produced by casting.

Mohamed and his group [25, 26] found that the creep apparent stress exponents (n) in their experiments were too high to be explained by traditional creep theory. An effective stress (e= -0, where  is applied stress, 0 is the threshold stress for dislocations to move) proposed by Nardone and Strife [37] has been used to elucidate this phenomenon.

It was found [24, 26] that 0 originated from the small particles and dislocations in the material. These small particles may be the precipitates formed during aging [24], or oxides introduced into the P/M matrix during processing [26]. Artz and his co-workers

[28-30] set up theoretical models for the effect of prediction of dislocation climb over different types of particles. For climb without attractive particle-dislocation interaction

(i.e. coherent particles), they obtained a natural power-law dependence of the dislocation velocity on the applied stress where n = 3 or 4. The 0 was very low in this case and the theory can be applied to lightly aged attractive particle-dislocation interaction [30]. This mechanism can be used to explain the high 0 found in the oxide-containing aluminium and sub-micron SiC reinforced 6061 AMC. However, Artz‟s works was based on a low volume fraction dispersion of cube-shaped particles. Obviously for the composites which have a much higher volume fraction of reinforcement, different sizes, shapes and morphologies, and a combination of coherent precipitates and hard dispersoids coexisting in the matrix, the situation is more complicated.

Fig 2.12 Creep behavior of different materials at the same applied stress and temperature

[32].

Fig 2.13 TEM micrographs showing the particulates and dislocations of

7775 Al (a) and 1 vol. % SiC p (50nm)/7775 Al composites (b) at 723K under 27.4MPa applied stress, 7775 Al (c) and 1 vol. % SiC p (50nm)/7775

Al composites (d) at 773K under 22MPa applied stress [32].

20 b. Creep of nano-particulate reinforced AMC

Ma et al. [23, 31] showed that the creep resistance of the 1vol. % Si-N-C (25-30nm)/Al and the 1vol%Si3N4 (15 nm)/Al composites is about two orders of magnitude higher than that of the15vol. % SiCp (3.5 μm)/Al composites. Kang [32] also studied high temperature creep behaviour of 7775 composites with 1vol.% SiCp(50nm) and found that there was a big improvement for the nanocomposites compared with monolithic alloy and composites with micrometric reinforcement at 732K and 775K as shown in Fig 2.12.

Fig 2.13 shows the microstructures of the different materials after creep test. At such a high temperature of 775K (~0.8Tmel) all the precipitates dissolved in the matrix and the nanometric particle could be found along grain boundary and within grains (as seen in

Fig 2.13d). Pinning grain boundary and dislocations are two possible explanations. There is no recognizable theory so far.

According to his results, it is highly possible that the composite can be used above 300oC to 450oC. This will be a significant improvement on the creep resistance of aluminium alloys or composites, and may compete with Ti alloys at these temperatures. c. Wear

The wear resistance of AMC was found to be improved by the addition of a hard phase

[33, 34, and 35]. However, with larger abrasive SiC particles (60μm) added to the composite, the wear resistance of a composite actually decreased after a critical amount of SiC added [36, 37]. With the advent of nano-particle production and beneficial effects on yield strength and hardness [17] when incorporation into a metal matrix, it may offer a new method for increasing the wear resistance of ductile metals. Influence of nano-particles (<50nm), its properties, size, shape and volume fraction on the wear behaviour of ductile metals will be studied.

21 2.4 Strengthening Mechanisms

Several strengthening models originally proposed for monolithic alloys have been modified for MMC, and have suggested that the strengthening due to the presence of the particulate is dependent on a variety of mechanisms [38].

2.4.1 Grain Strengthening

During thermomechanical processing particulate reinforced MMC may recrystallize. For particles of diameter greater than about 1 m, recrystallization may occur by particle stimulated nucleation [1]. The resultant grain size is often finer than that found in an unreinforced alloy and may contribute to the yield strength of the MMC according to the

Hall-Petch equation [39, 40]:

-1/2 gb = 0 + KD where gb is the yield stress of polycrystalline metals with an average grain size (D), and

0 and K are constants.

2.4.2 Sub-structure Strengthening

It is show that above a critical ratio of volume fraction to particle size, the material will retain a fine worked structure on solution treatment [41]. In this case, the sub-structure will contribute to the MMC strength via the Hall-Petch equation.

2.4.3 Quench Strengthening

The large difference in thermal expansion between aluminum and SiC results in the generation of dislocations on quenching from the recrystallization or solution treatment temperature [42, 43]. The dislocation density generated on quenching () is a function of

22 reinforcement size (d) and volume fraction (V) and the product of the thermal mismatch

(C) and the temperature change (T). The strength (q) can be estimated by an expression such as:

1/2 q=Gb

=12TCV/bd where G is the material's shear modulus, b is the Burger's vector and  is a constant between 0.5 and 1 [43].

The analysis above assumes that the dislocations are uniformly distributed and that all the dislocations generated can contribute to strength. This may not be true for a solution hardened matrix [38].

2.4.4 Orowan Strengthening

The Orowan bypassing of particles by dislocations can increase a material‟s strength. If the particles are assumed to be equiaxed then the strengthening (or) is estimated to be [44, 22]:

or = mor = 0.84mGb/(-d)

 = (6V/)-1/3d where or is the resolved shear stress, m is the orientation factor, G is the material‟s shear modulus, b is the Burger‟s vector,  is the interparticle spacing, d is the mean particulate diameter and V is the volume fraction of particulates.

2.4.5 Work Hardening

The work hardening of a MMCs will be influenced by the dislocation structure formed during quenching. However, there are other factors that

23 will increase the work hardening rate of the composite compared with that of the matrix alloy. Firstly, in the early stages of deformation there will be load transfer between the matrix and the reinforcing phase by means of

Orowan loops or some equivalent mechanism [38]. The expected contribution to strength (wh1)is simply dependen on V (volume fraction) according to the following equation [45]:

wh1 = 4.5GVε where ε is the true strain

As deformation proceeds the stress induced at the particle/matrix interface leads to the relaxation of the Orowan loops at strains as low as 10 -4. At higher strains, the particles will contribute to work hardening due to the creation of geometrically necessary dislocations [ 46]. This contribution to strength (wh2 ) has been estimated by Ashby [10] to be:

1/2 1/2 wh2= 5G(2Vb/d) ε

It should be noted that the above contributions to the MMC strength will be superimposed on any alloying effects such as solution or precipitation hardening, i.e. the various mechanisms will interact with each other. The models require modifications to consider interactions between the various mechanisms. All theories here describe the strengthening of the matrix by particulates larger than 1 m. These mechanisms seem to require modifications for nano-scale reinforcements.

24 Chapter 3 EXPERIMENTAL

3.1 Introduction

As discussed in literature survey, most of the nanocomposites research was based on pure metals while in this project, a heat-treatable 7075 alloy was used as matrix for the composites. Rod shape monolithic alloy and composites with 1% and 5% nanometric silicon carbide particulates were fabricated via modified powder metallurgy (P/M) and extrusion. Ageing behaviours of the alloy and composites were studied and mechanical tests, including room temperature tensile test and abrasive wear test, were performed on peak-aged samples. Optical microscopy (OM) and scanning electron microscopy (SEM) were used on fracture surfaces and wear surfaces for morphology study. The shape and dispersion of nanometric ceramic particles and precipitates were observed by

Transmission Electron Microscopy (TEM). Details of the experiments are described in this chapter.

3.2 Manufacturing

3.2.1 Starting Materials

Matrix alloy Mg Si Zn Cu Fe Ni Al

7075 Al 2.23 0.06 5.71 0.98 0.10 0.01 Bal.

Table 3.1 Chemical composition of matrix alloy (wt %)

Fig 3.1 Morphology of nanometric silicon carbide particulates

(a) SEM of as-received powders (b) TEM micrograph of ultrasonic-dispersed powders

26 The starting materials used were micro-sized 7075 aluminium alloy powders (Ampal,

U.S.A.) and nanometric silicon carbide powder which had average sizes of 78 µm and

50nm respectively. The chemical composition of the aluminium powder is shown in

Table 3.1. Fig 3.1(a) is an SEM picture of as-received silicon carbide powder which shown a very large SiC with small particles on its surface and Fig 3.1(b) is a TEM picture of ground and ultrasonic-dispersed SiC powder. It can be seen that the SiC particulates are now less than 100nm after the ultrasonic dispersion.

3.2.2 Powder Metallurgy process

The P/M process employed in this work included wet mixing, cold isostatic pressing

(CIP) and sintering. First the aluminum powder were mixed with different volume fractions (1 and 5 vol. %) of silicon carbide powder in pure ethanol slurry followed by ultrasonic dispersion and ball milling, then the mixture was dried at 150oC and compacted by CIP. The compacted billets were sintered in nitrogen at 595oC for two hours. The last step was hot-extrusion of the sintered billets with a reduction ratio of

36:1, to produce bars of 13 mm diameter. Fig. 3.2 is the flow chart of the composite fabrication.

3.3 Heat Treatment and Hardness Test

All specimens were subjected to a T6 artificial aging treatment. Firstly, 7075 aluminium alloy and the composites were solutionized at 480oC for 2 hours and water quenched.

Subsequently, 7075 Al alloy and the composites were aged at 120oC for different period of time in a constant temperature oil bath.

Before the hardness test, all the samples were mounted and ground successively using

320#, 800#, 1200# SiC paper, and finally polished with 3µm and 1µm diamond paste

27 using a polishing machine. A LECO M-400-H1 hardness-tester was used to determine the changes in hardness during artificial ageing. A loading of 500g and a loading time of

15 seconds were used for the Vickers hardness test and 20 measurements were taken for each sample.

Fig 3.2 P/M fabrication process for aluminium composites

3.4 Tensile Testing

Tensile tests were performed at room temperature on each material in peak-aging condition, using a Instron 1185 Screw testing machine at nominal applied strain rate of

1.67 x 10-4/S. Laser extensometer were used for the measurement of the displacement of the samples. The dog-bone specimens were machined to have a gauge length of 30 mm 28 and a diameter of 6mm and the surfaces of the tensile specimens were polished before tensile test. Three specimens were tested for each material.

3.5 Wear Test

Fig 3.3 Schematic diagram of the pin-on-disk wear test apparatus.

The abrasive wear tests, which were of the pin-on-disc type, were performed using a polishing machine [33], as shown in Fig 3.3. Experiments were carried out under dry condition by sliding the samples under an applied load of 10 N on a 120# abrasive paper stuck to the grinding disk which rotated at 300 rev /min. A fixed track diameter of

150mm was used and the duration time was 60 second.

Each specimen was ground up to 600# abrasive paper to make the wear surface contact well with the surface of the abrasive paper. Each test was conducted on a fresh abrasive paper and three runs were conducted for each test condition. Wear losses were obtained by weighing the samples before and after wear test and the wear rate was calculated by converting the mass loss to volume loss using respective densities:

3.6 Optical and scanning electron Microscopies

Optical microscopy was used for the metallographic study of the peak-aged samples and the cross section of the samples after tensile testing. All the samples were mounted and ground successively using 320#, 800#, 1200# SiC paper, and finally polished with 3µm and 1µm diamond paste using a polishing machine. Keller‟s enchant, consisting of 1%

HF (48%), 1.5% HCl, 2.5% HNO3, and 95% H2O, was use for the peak-aged samples to observe the grain structures.

A Hitachi S4500 Field Emission Scanning Electron Microscope (FESEM) was used on tensile test samples and wear test samples to investigate fractography of failed samples.

The fracture surfaces of the tensile test samples and the worn surfaces and debris of the wear test samples were preserved carefully to avoid contamination.

3.7 TEM

3.7.1 Sample Preparation

Firstly, slices with thickness of about 500µm were cut slowly by a low speed diamond saw, followed by grinding successively using 800#, 1200# SiC papers to a thickness of

200µm. Then 3mm discs were punched from the thin slices by electro-discharge machine (EDM), using copper discharge electrode while submerged in LECO cutting oil bath.

Finally, the discs were electropolished by a Cryo-Ultramicrotome Struers Tenupol 30 Electropolisher operated at 10V and 0.1A. The electropolishing solution was a mixture of 30% nitric acid and 70% methanol and the temperature of the solution was maintained at about -20oC.

3.7.2 Transmission Electron Microscopy

Philips CM200 field emission gun transmission electron microscope operating at 200kv was used to examine the thin foils and energy dispersive x-ray spectroscopy (EDS) was used to determine the composition of various regions.

31 Chapter 4 RESULTS AND DISCUSSION

4.1 Microstructures of 7075 Al and 7075 Al Composites

4.1.1 Optical Metallography of As-extruded Materials

Optical micrograph of 7075 aluminium alloy and its composites both in transverse and longitudinal directions are shown in Figs 4.1 to 4.3. Secondary phases were seen to have aligned in the longitudinal direction through hot extrusion process. The grains were elongated along the extrusion direction because sever deformation during extrusion.

Compared with composites, the monolithic alloy was clean and had less second phases due to the absence of reinforcements. The inclusions in 7075 alloy were most probably intermetallic phases combined with oxide inclusion which formed during powder metallurgy processing. The intermetallic particles could form from elements such as Cr,

Fe, Mn and Zr [47]. Figs 4.2 and 4.3 reveal that the composites had a tremendous increase of secondary phases with a size of several micros, dispersed mainly along the grain boundaries and the amount of clusters increases with increasing volume fraction of reinforcements. These second phase were considered to be nanometric silicon carbide which formed large clusters. Due to the large difference between the aluminium powder size (78µm) and silicon carbide powders (50µm), the nanometric reinforcements tended to fill in the gaps between matrix particles during mixing. Even after extrusion there was still certain amount of clusters.

It is also noticed that 7075 composites have a smaller grain size than the monolithic alloy. This kind of grain refinement effect was much notable for pure aluminium nanocomposites which decreased the grain size from 4.6 µm to 1.1µm by adding only

(a)

(b)

Fig 4.1 OM of as-extruded 7075 Al composites

(a) in transverse direction, and (b) in longitudinal direction 33

(a)

(b)

Fig 4.2 OM of as-extruded 1vol. % SiC (50nm)/7075 Al composites

(a) in transverse direction, and (b) in longitudinal direction

(a)

(b)

Fig 4.3 OM of as-extruded 5vol. % SiC (50nm)/7075 Al composites

(a) in transverse direction, and (b) in longitudinal direction

35 4vol. % nano-sized alumina particles [1]. It was pointed out that the nanometric particles could pin grain boundaries hence gave rise to grain refinement. Since 7075 aluminium got oxide particles and intermetallic phases which could also help pinning grains, the grain refinement effect by nanoreinforcements for alloy matrix was not as significant as that for pure aluminium.

4.1.2 Distribution and Segregation of Nanoparticles

SEM micrograph and its related EDS for two different regions of 5vol. % SiC

(50nm)/7075 AMC are shown in Figs 4.4 and 4.5. Dark regions (5-10µm) and white second phase were found out to be large nanometric silicon carbide clusters and

Fe-containing intermetallic phase. Fig 4.6 give more details of a single cluster containing large amount of nano-sized particle which had a similar morphology with as-received nanometric silicon carbide powders, as shown in Fig 3.1 (a).

TEM micrographs of 1vol. % SiC (50nm)/7075 AMC and 5vol. % SiC (50nm)/7075

AMC are shown in Fig 4.7. Two different distributions of nanometric particles can be identified in these micrographs. In Figs 4.7 (a) and (c), all the nanoparticles dispersed along grain boundaries and some formed very small clusters, but there were almost no particles on sub-grain boundaries and within grains. In Fig 4.7 (b) and (d), much larger clusters with size of 1 to 2µm dispersed at junctions of three grains and within grains, and some of these clusters were close to each other which could form very fine grains.

Either way, the agglomeration increased with increasing fraction of the nanometric reinforcements.

EDS results in Fig 4.8 confirmed that the black regions were silicon carbide clusters.

TEM micrograph of AA7075 and its composites in Fig 4.9 [48] show similar dispersions of nano-particles.

(a)

(b)

Fig 4.4 SEM micrograph of as-extruded 5vol. % SiC (50nm)/7075 Al composites.

(a) at low magnification (b) at large magnification

Large agglomeration region (black) and Fe-containing intermetallic phase (white)

Fig 4.5 EDS of agglomerations region and matrix region in Fig 4.4 (b)

Fig 4.6 SEM micrograph of an agglomeration region of 5vol. % SiC (50nm)/7075

aluminium composites at high magnification

(a) (b)

Fig 4.7 TEM micrograph of peak-aged 7075 AMC

(a), (b) 1vol. % SiC (50nm)/7075 Al composites

(c), (d) 5vol. % SiC (50nm)/7075 Al composites

Fig 4.8 EDS of different regions (marked as +1 and +2 in Fig 4.7 (c)) of 5vol. % SiC

(50nm)/7075 Aluminium composites

Fig 4.9 TEM micrographs of AA7075 and AA7075 MMC in as-received condition [48]

AA7075 (a) transverse direction (b) longitudinal direction

AA7075 MMC (c) transverse direction (d) longitudinal directions

Fig 4.10 Schematic sketch of the milled composite after extrusion [49]

(a) nano-structured regions [51]

(b) coarse grains in the inter-particle region [52]

Fig 4.11 Bright field TEM image of the hot rolled Al-5083/SiC composite

42 4.1.3 Technology for Homogeneous Dispersion of

Nanoreinforcements in Metals

As discussed above dispersing nanoreinforcements in matrix becomes an issue of fabricating nanocomposites. There are several studies aiming to avoid segregation.

Magnesium composites reinforced by silicon carbide nanoparticles were fabricated via powder metallurgy and extrusion by Ferkel and Mordike. It was found that ball milling could improve the mixing effect and the dispersion of the nanoparticles, as shown in Fig

4.10. This kind of microstructure is described as “Nanoparticles decorated the grain boundaries of the metal matrix and form together with the matrix along the grain boundaries” [49] which is very similar as our finding. Razavi also found that 72 hours of ball milling could embed nanoparticles in metal matrix particles [50]. Cryomilling has also been used to disperse nanoparticles and produce nanograins [51, 52]. As shown in Fig 4.11 (a), ultra fine grains and homogeneous dispersion of nanoparticles can achieve, however, there were also regions free of nanoparticles which were shown in fig

4.11 (b). The reason could be loss of powder with the flow of evaporated liquid nitrogen during cryomilling. Both ball milling and cryomilling aim to embed nanoparticles within metal matrix, but nanoparticles free regions and clusters were common phenomena especially at large fraction of nanoparticles.

Yamagiwa [53] claimed that three-dimension shaking without balls could coat the metal powders with nanocarbon materials which could give a homogeneous dispersion of nanocomposites, but till now limited publication about this method could be found.

43 4.2 Aging Behaviours

As quenched hardness measurements in Fig 4.12 show that hardness tended to increase with presence and increasing fraction of silicon carbide particulates, this suggested that the nanometric SiC particles can contribute to the strengthening of composites. This was less notable than the reports of other researches, which were based on nanoreinforcements including alumina particulates, silicon nitride particulates and carbon nanotubes [1, 23, and 54]. This could be due to the fact that the grain refinement effect is not significant for 7075 aluminium composites.

Fig 4.13 show hardness-ageing curves during ageing at 120oC which indicate that both monolithic alloy and its composites reached peak ageing at about 48 hours (two days).

Similar ageing curve in Fig 4.14 shown that Al 7075/TiCp composites reached a little bit higher peak-ageing hardness than 7075 Al after same period of ageing.

Tremendous findings of accelerated aging were reported [55-58] based on dislocation generation because of large thermal mismatch between metal matrix and ceramic reinforcements. Theses dislocation could serve as heterogeneous sites for the nucleation of strengthening precipitates and provide short circuit diffusion paths for solute atoms.

As a result, both the nucleation and growth of precipitates in the matrix can be drastically altered by the presence of the reinforcements [16]. While for nanocomposites, limited volume change of reinforcements was resulted because of the small size of the reinforcements, which could generate much less dislocations and thus did not contribute to the ageing process of the matrix alloy.

High strength 7XXX alloys are heavily alloyed with Zn and Mg which form the major strengthening phases ‟ and  (MgZn2) [59-61]. The precipitation sequence is:

SSS  GP zone  ”   (MgZn2 ) 44 Kang [62] did DSC test on 7775 aluminium nanocomposites which was shown in Fig 4.15 and it indicated the peaks which was related to the volume fraction of precipitating occurred at approximately the same temperature. The nanoreinforcements did not change the precipitates sequence or accelerate the ageing precess which agreed with our result s.

TEM micrograph in Fig 4.16 [63] show microstructures of peak aged 7775 nanocomposites which contained very fine precipitates. Clusters of nanoparticles could be seen and there were PFZs on grain boundaries and interfaces between reinforcements and matrix. For 7775 composites reinforced by SiC no alloy elements segregation was found and the PFZs were associated with vacancies absorption which reduce the diffusion rate of atoms, and hence reduced amount of in situ precipitates. For 7775 composites reinforced by Al2O3, high concentration of magnesium was found on the surface of alumina particles and larger PFZs formed due to lack of solute atoms. It is suggested that nanometric alumina particles should not be use for magnesium containing matrix. For our composites, no PFZs was found at interfaces between nano-SiC particles and the matrix either, which proved that SiC is a better choice due to its‟ chemical stability.

Fig 4.12 As quenched hardness of 7075 aluminium and its composites

45 200 200

180 180

160 160

140 140

7075 Al alloy HV HV 120 1% 7075 AMC 120 100 5% 7075 AMC 100

80 80

60 60 0.1 1 10 100 1000 0.1 1 10 100 1000 Ageing time, hours Ageing time, hours 200 200

180 180

160 160

140 140 HV 120 HV 120

100 100

80 80

60 60 0.1 1 10 100 1000 0.1 1 10 100 1000 Ageing time, hours Ageing time, hours

Fig 4.13 Hardness-aging curve of 7075 aluminium and its composites

o Fig 4.14 Hardness variations of Al 7075 and Al 7075/TiCp samples with time at 120 C [55]

7775 Al 10%SiC(10um)/7775 Al 1%SiC(50nm)/7775 Al 2%SiC(50nm)/7775 Al 3%SiC(50nm)/7775 Al

5%SiC(50nm)/7775 Al



50 100 150 200 250 300 350 400 450 o Temperature ( C)

Fig. 4.15 Comparison of the heat flow of precipitation reactions in 7775 Al

alloy and its composites with different amount of nano-SiC

particles [62].

(a)

(b)

Fig.4.16 TEM graphs of the peak aged composites [63]

(a) 1vol.% SiC(50 nm)/7775 Al composites (b) 1vol.% Al2O3(50 nm)/7775

Al composites

48 4.3 Room-temperature Tensile Properties

Table 4.1 and Fig 4.17 shows the tensile properties of the 7075 alloy and its composites .The alloy used in this project had quite similar yield strength (YS) and ultimate tensile strength (UTS) compared with data from ASM handbook [1]. It was notice that the 1vol. % SiC (50nm)/7075 aluminium composites had the greatest YS and

UTS and the 5vol. % SiC (50nm)/7075 aluminium composites had the lowest strength.

The ductility decreased with increase of the volume fraction of reinforcements.

Fig 4.18 and Fig 4.19 shown macrostructures of cross section in transverse and longitudinal directions and it was shown that the fracture surface was getting rough with adding of nanometric reinforcements and the shear lips become much smaller for composites, while 5% composites even have no shear laps.

Enlarged grains after tensile test and broken grains at edge of fracture of 5vol. % SiC

(50nm)/7075 aluminium composites could be found in Fig 4.20 which indicated that for composites the bonding of grains become weaker, mainly because of segregation of nanometric particulates on grain boundaries.

SEM micrograph of fracture surfaces for the monolithic alloy were shown in Fig 4.21 and Fig 4.22. The centre part of fracture surfaces were totally ductile failure which contained fine dimples, while the lips contained dimples and sheared faces. It is noticed that inclusions were found in some dimples which could be intermetallic particles, discussed in chapter 4.1.

For 1vol. % SiC/7075 AMC, Fig 4.23 shows both a ductile fracture surface and a brittle surface. Although large crack were also found in Fig 4.24, it was stopped by a series of very fine dimples. This explains why the 1vol. % SiC/7075 AMC maintained its ductility. Fig 4.25 shows a large dimple containing inclusions in 1vol. % SiC/7075

AMC and it is found the inclusions were nanometric particulates clusters. 49 UTS (MPa) YS (MPa) El (%)

7075 Al 574 533 9.6

1 Vol.% SiC/7075 AMC 634 599 8.9

5 Vol.% SiC/7075 AMC 547 527 3.1

7075 alloy [64] 572 503 11

Table 4.1 Tensile properties of 7075 aluminium and its composites

Fig 4.17 Stress-strain curve of 7075 aluminium and its composites

(a)

(b)

(a)

(b)

AMC (c) 5 vol % SiC (50nm)/7075 AMC

(a)

(b)

Fig 4.20 OM and SEM of longitudinal cross section of 5 vol % SiC (50nm)/7075 AMC

Notice enlarged grains after tensile test and break of grains

(a)

(b) Fig 4.21 SEM micrograph of fracture surface of 7075 aluminium

(a) centre (b) shear lips 54

(a)

(b) Fig 4.22 SEM micrograph of fracture surface of 7075 aluminium in large magnification

(a) centre (b) shear lips 55

(a)

(b) Fig 4.23 SEM micrograph of fracture surface of 1vol. % SiC (50nm)/7075 AMC

(a) ductile surface (b) brittle surface 56

Fig 4.24 SEM micrograph of large crack in 1 vol. % SiC (50nm)/7075 AMC

Fig 4.25 SEM micrograph of a dimple containing clusters in 1 vol. % SiC (50nm)/7075

AMC 57

Fig 4.26 SEM micrograph of fracture surface of 5 vol. % SiC (50nm)/7075 AMC

(a)

(b) Fig 4.27 SEM micrograph shown nanometric particles and clusters on the fracture surfaces of 5vol. % SiC(50nm)/7075AMC (a) brittle region (b) ductile region including dimples 59

Fig 4.28 Bimodal distributions of dimples in 5 vol. % SiC (50nm)/7075 AMC

Brittle fracture surfaces are shown in Fig 4.26. Large amount of particles and clusters were found in varies regions as shown in Fig 4.27. For brittle region the dispersion was random and for ductile regions they all located in dimples. A lot of internal crackings has been found in Fig. 4.26, which were initiated from the clustering of reinforcements.

These micro-cracks would act as stress raisers for large cracks to take place, hence the ductility decreased. This agrees the result in tensile testing, where the elongation of

5vol. % SiC/7075 AMC was only 3.1%.

A bimodal distribution of dimples in 5% AMC was found in Fig 4.28 which containing very fine dimples (about 1 µm) and larger dimples (about 5 µm.) This could be explained by different degree of clusters found in chapter 4.1, which generated different sized dimples of different sizes.

The dislocations/precipitates interactions in 7075 aluminium after tensile testing have been observed by TEM in Fig 4.29(a). Kang and Chan[65] studied studied 60 1vol. %SiC(50nm)/6061Al composite and the microstructure after tensile testing are shown in Fig 4.29(b). For composites additional dislocations were found around nanometric silicon carbide which could provide further strengthening by Orowan loops.

(a)

(b) Fig 4.29 TEM micrograph showing dislocations after tensile test

(a) 7075 aluminium alloy (b) 1vol.% SiC/6061 Al composites [65] 61

Fig4.30 Schematic sketch of composites microstructure reinforced by nano-sized particles (a) pure aluminium as matrix (b) peak-aged 7075 aluminuim composites

As discussed in section 2.44, the Orowan strengthening effect is determined by the interparticle spacing (), the mean particulate diameter (d) and the volume fraction of particulates (V). For peak-aged 7075 aluminuim and its composites, the sketch of microstructures was shown in Fig 4.30. There are two kinds of particles which both could strengthen the matrix. From literature study and microstructure observation, all the parameters (, d, V) of the precipitates particles are much higher than that of the added ceramic particles. This explains why the nano-sized ceramic particles could largely strengthen pure metals but slightly strengthen 7xxx peak-aged alloys. In another word, precipitates particles dominate strengthening effect in this composites based on high strength aluminium alloy matrix.

62 4.4 Abrasive Wear Properties

Peak-aged hardness

220

200

180 HV 160

140

120

100 7075 Al 1vol. % SiC/7075 AMC 5vol. % SiC/7075 AMC

Fig 4.31 Peak-aged hardness of 7075 aluminium and its composites

abrasive wear

0.8 /m)

3 0.6

0.4

0.2 wear rate(mm wear 0.0 7075 Al 1vol.%SiC/7075 5vol.%SiC/7075 AMC AMC

Fig 4.32 Abrasive wear rate of peak aged 7075 aluminium and its composites

63 Peak-aged hardness and abrasive wear rates of 7075 aluminium and its composites are shown in Figs. 4.31 and 4.32. The hardness of all materials were quite similar, while the wear rate of the composites decreased with increase of volume fraction of reinforcements and the wear rate of 1vol. % SiC(50nm)/7075 Al composites was about half of its matrix alloy. These results indicate that hard second phase could reduce wear rate of the matrix alloy, even though their hardness was similar. However, the effect of improvements is much less notable than composites with micro-size reinforcement [66].

For composites containing the same amount of SiC particles the wear rate decreased with increasing SiC size [66, 67], but these findings were only basis on micro-size reinforcements. For nanometric reinforcements the interfacial area was much larger than that reinforced with micro-sized particulates. This increased the chance of the nanometric reinforcements to escape from the matrix, while large reinforcements could embed longer before they were broken down [68].

SEM of wear debris of 7075 alloy and its composites are shown in Figs 4.33 and 4.34, respectively. The length of debris decreased with the increase of the reinforcement content. The debris of the monolithic alloy contained only long curved chips, while the

1% SiC/7075 Al composites got shorter and wider chips and for 5% SiC/7075 Al composites there were short fragments instead of chips. Fig 4.35 shows the mode of materials removal from the groove by a process of cutting which produced long chips

[69]. Fig 4.34 (a) and (b) show morphology of both sides of the short chips cut from

1vol. % SiC(50nm)/7075 aluminium composites, it is clear that the wear mechanism for monolithic alloy and 1vol. % SiC(50nm)/7075 aluminium composites was microcutting by the small abrasive particles. For 5vol. % SiC(50nm)/7075 aluminium composites, the debris consisting brittle fragments which could be produced by delaminating or cracking.

64 Figs. 4.36 to 4.38 showed the worn surfaces of 7075 alloy and its composites. It is noticed that for monolithic alloy the worn surfaces consist of long grooves which were caused by cutting of chips, while for composites the grooves become shallower and cracking occurred. The presence of hard reinforcements were responsible for shallower grooves, while the large agglomeration region in Fig 4.4 could be the origin of cracking by fatigue as found by Ma et al. [70]. By comparing the worn surfaces and worn debris of alloy and composites, there was a transition of wear mechanism which changed from micro-ploughing and micro-cutting to micro-cracking [71].

Fig 4.33 (a) Worn debris of 7075 Al

Fig 4.33 (b) Worn debris of 1vol. % SiC (50nm)/7075 Aluminium composites

Fig 4.33 (c) Worn debris of 5vol. % SiC (50nm)/7075 Aluminium composites 66

(a)

(b)

(c)

(d)

Fig 4.34 SEM micrograph of worn debris in large magnification (a), (b) short chips cut from 1vol. % SiC/7075 AMC (c), (d) fragments from 5 vol. % SiC/7075 AMC

Fig 4.35 Mode of materials removal form the groove by a process of cutting [69]

Fig 4.36 SEM micrograph of worn surfaces of 7075 aluminium alloy

Fig 4.37 SEM micrograph of worn surfaces of 1vol. % SiC (50nm)/7075 aluminium composites 71

(a) (b)

Fig 4.38 SEM micrograph of worn surfaces of 5 vol. % SiC (50nm)/7075 aluminium composites

72 CHAPTER 5 CONCLUTIONS AND

FUTURE WORK

5.1 CONCLUTIONS

Nanocomposites were fabricated by traditional powder metallurgy and extrusion, microstructures, ageing behaviour and mechanical properties were studied in this work.

The following conclusions can be made:

1. Different volume fractions of nanometric particulates could be introduced to 7xxx aluminum matrix by traditional powder metallurgy technique.

2. The dispersion of the nanometric particulates in aluminium matrix was not homogeneous, most reinforcements dispersed along grain boundaries and clustering were formed and aggregation tends to increase with increasing particulates content.

3. Nanometric silicon carbide particulates did not change the aging kinetic and precipitates phases of 7075, but formed narrow PFZs at the interface.

4. The 1% vol. SiC/7075 composites exhibited the highest YS and UTS in all materials in this work. The nanometric particulates did strengthen 7075 matix based on Orowan strengthening mechanism.

5. Abrasive wear rate of SiC/7075 nanocomposites decreased with increasing reinforcements content. The improvement was not notable compared with composites reinforced by micro-sized particulates due to nanometric particulates aggregation and the large size of abrasive grit.

73 5.2 FUTURE WORK

1. Dispersion of nanometric reinforcements become a critical step for fabricating nanocomposites because of negative effect of segregation. To improve this work, different kinds of mixing technology such as attrition mill, three dimension coating and cryomilling may be applied.

2. To thoroughly understand the strengthening effect of the nanometric particulates, different strain rate with different load at different temperature need to be done.

3. To confirm the second phases and reinforcements, HRTEM and electron diffraction need to be applied.

4. Other kind of nanometric reinforcements (such as carbon nanotubes) with different size and shape should be used to get a fully understanding of nanocomposites.

74 Reference

[1] Y. C. Kang, S. L. I. Chan, Mater Chem and Phy. 85 (2004) 438.

[2] C.F. Deng, D.Z. Wang, X.Y. Zhang, A.B. Li, Mater. Sci. Eng. A 444 (2007) 138.

[3] J.M. Torralba, C.E. da Costa, F. Velasco. Journal of Materials Processing

Technology 133 (2003) 203

[4] S. Suresh, A. Mortensen, A. Needleman. Fundamentals of Metal Matrix Composites.

Butterworth-Heinemann, 1993, P24

[5] T.W. Clyne, P.J. Withers. An Introduction to Metal Matrix Composites, Cambridge

University Press, 1993, P350

[6] S. Suresh, A. Mortensen, A. Needleman. Fundamentals of Metal Matrix Composites.

Butterworth-Heinemann, 1993, P25

[7] H.S. Chu, K.S. Liu, J.W. Yeh. Script Mater (2001) 45

[8] Buha, Joka, Interrupted ageing of Al-Mg-Si-Cu alloys. PHD thesis, UNSW 2005

[9] David A. Porter and K.E. Easterling, Phase transformations in metals and alloys, 3rded

(1992), P303

[10] David A. Porter and K.E. Easterling, Phase transformations in metals and alloys,

3rded (1992), P293

[11] I. J. Polmear, Light alloys: metallurgy of the light metals, 3rded (1995), P29

[12] S. Suresh, A. Mortensen, A. Needleman, Fundamentals of Metal-Matrix

Composites, Butterworth-Heinemann, 1993, P120.

[13] I. J. Polmear, Light alloys: metallurgy of the light metals, third edition (1995), P38.

[14] Porter, David A, Phase transformations in metals and alloys, third edition (1992),

P305.

75 [15] X. Liu, J. Cui, Materials Science and Engineering: A402 (2005) 1.

[16] S. Suresh, A. Mortensen, A. Needleman, Fundamentals of Metal-Matrix composites,

Butterworth-Heinemann, 1993, P122.

[17] Yung-Chang Kang. Mechanical Properties of Nanometric Particulate Reinforced

Aluminium Composites, PhD thesis (2004). National Taiwan University. P13

[18] D. Brooksbank and K. W. Andrews, JISI 207 (1969) 474.

[19] D. L. McDanels, Metall. Trans. 16A (1985) 1105.

[20] C. P. You, A. W. Thompson and I. M. Bernstein, Scripta Metall. 21

(1987) 181.

[21] T. J. A. Doel and P. Bowen, Comp. 27A (1996) 655.

[22] Z.Y. Ma, Y.L. Li, Y. Liang, F. Zheng, J. Bi, S.C. Tjong, Mater Sci Eng A 219 (1996), 229.

[23] Z.Y. Ma, S.C. Tjong. Composite Science and Technology 59 (1999) 263

[24] L. Kloc, S Spigarelli., E. Cerri, E. Evangelista, T.G. Langdon Acta Mater (1997) 45

[25] F.A. Mohamed, K.T. Park, E.J. Lavernia, Mater Sci Eng A150 (1992) 21

[26] F.A. Mohamed, Script Mater 38 (1998) 457

[27] V.C. Nardone, J.R. Strife, Metall. Trans A 18 (1987) 109

[28] E. Artz , D.S. Wilkinson, Acta Metall 34(1986), 1893

[29] J. Rösler and E. Arzt, Acta Metall 36 (1988) 1043.

[30] E. Arzt and J. Rösler, Acta Metall 36(1988) 1053.

[31] Z.Y. Ma, S.C. Tjong, Y.L. Li, Y. Liang. Mater Sci Eng A225 (1997) 125

[32] Yung-Chang Kang. Mechanical Properties of Nanometric Particulate Reinforced

Aluminium Composites, PhD thesis (2004). National Tai Wan University. P162

[33] H.L. Lee, W.H. Lu, S.L.I. Chan, Wear 159 (1992), 223.

[34] S.V. Praasad, P.K. Rohatgi , J Met 39 (1987) 22

[35] H.L. Lee, W.H. Lu, S..LI. Chan, Chin J Mater Sci 24(1992) 40

76 [36] A.G. Wang, I.M. Hutching, Mater Sci Technol 5 (1989) 76

[37] A.K. Jha, S.V. Prasad, G.S. Upadhyaya Wear 133 (1989) 163

[38] W. S. Miller, F. J. Humphreys, Scr. Met. Mater. 25 (1991) 33.

[39] E. O. Hall, Proc.Phys.Soc. 64B (1951) 747.

[40] N. J. Petch, J.IronSteelInst. 173 (1953) 25.

[41] F. J. Humphreys 9th Riso Conference Denmark P25.

[42] R. J. Arsenault, N. Shi. Mater. Sci. Eng. 81 (1986) 175.

[43] R. J. Arsenault, Mater. Sci. Eng. 64 (1984) 171.

[44] U. F. Kocks, Phil. Mag. 13 (1966) 541.

[45] L. M. Brown, W.M. Stobbs Phil Mag 23 (1971) 1187.

[46] M. F. Ashby. Phil Mag 14 (1984) pi157.

[47] Sahin, A. and F. Sarioglu. Scripta Material, Vol 37, no8, pp1117-1121. 1997

[48] Hamidreza Mohseni. Microstructural Development and Thermal Stability of

Aluminium-based Composites Processed by Severe Plastic Deformation. PhD

Thesis. University of New South Wales, July 2006

[49] H, Ferkel, B.L. Mordike. Materials Science and Engineering A298 (2001) 193-199

[50] Z. Razavi Hesabi, H. R. Hafizpour, A. Simchi. Mater Sci Eng. A 454-455 (2007) 89

[51] F. Tang, B. Q. Han, M. Hagiwara. Advanced Engineering Materials. 2007, 9. No. 4

[52] F. Tang, M. Hagiwara, J. M. Schoenumget. Materials Science and Engineering

A407 (2005) 306–314

[53] United States Patent Application : 0060269435

[54] C. Deng, X. Zhang, D. Wang, Q. Lin, A. Li. Materials Letters 61 (2007) 1725–1728

[55] A. Sahin and F. Sarioglu, Scripta Materials, Vol. 37, No. 8, pp. 1117-1121, 1997

[56] L. Salvo, G. L'Espérance, M. Suéry, J. G. Legoux Mater. Sci. Engng A 177 (1994)

173

77 [57] M. P. Thomas, J. E. King, J. Mater. Sci 29 (1994) 5272

[58] C. Badini, F. Marino, E. Verne, Mater. Sci. Engng A191 (1995) 185

[59] J. M. Gómez de Salazar and M. I. Barrena, Scripta Mater. 44 (2001)

2489.

[60] G. Dlubek, P. Lademann, H. Krause, S. Krause and R. Unger, Scripta

Mater. 39 (1998) 893.

[61] J. M. Papazian, Metall. Trans. 13A (1982) 761.

[62] Yung-Chang Kang. Mechanical Properties of Nanometric Particulate Reinforced

Aluminium Composites, PhD thesis (2004). National Tai Wan University. P137

[63] Yung-Chang Kang. Mechanical Properties of Nanometric Particulate Reinforced

Aluminium Composites, PhD thesis (2004). National Tai Wan University. P138

[64] J.R Davis and Davis & Associates, ASM Specialty Handbook-Aluminums and

Aluminums Alloys, The Materials Information Society, 1994, P699

[65] Yung-Chang Kang. Mechanical Properties of Nanometric Particulate Reinforced

Aluminium Composites, PhD thesis (2004). National Tai Wan University. P120

[66] H. L. Lee, W. H. Lu and S. L. I. Chan. Materials Letters 15 (1992) 49-52

[67] R. L. Deuis, Wear 201 (1996) 132-144

[68] O. Yilmaz, S. Buytoz. Composites Sci and Tech 61 (2001) 2381-2392

[69] A. G. Atkins, J. H. Liu. Wear 262 (2007) 146-159

[70] Ma Zongyi, Bi Jing, Lu Yuxiong, Shen Hongwei and Gao Yinxuan, wear, 148

(1991) 287-293

[71] K-H Zum Gahr. Microstructure and Wear of Materials. New York, 1987. P234