production of nanocrystalline aluminum alloy powders ... · production by cryogenie milling and...

Production of nanocrystalline aluminum alloy powders

through cryogenic milling and consolidation by dynamic

magnetic compaction

by

Umugaba Seminari

A Thesis Submitted to the Faculty of Graduate Studies and Research in

Partial Fulfillment of the Requirements for the Degree of Master of

Engineering

Department of Mining and Materials Engineering

McGili University

Montreal, Canada

August 2007

© Copyright S. Umugaba

1+1 Libraryand Archives Canada

Bibliothèque et Archives Canada

Published Heritage Branch

Direction du Patrimoine de l'édition

395 Wellington Street Ottawa ON K1A ON4 Canada

395, rue Wellington Ottawa ON K1A ON4 Canada

NOTICE: The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats.

The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

ln compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis.

While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis.

• •• Canada

AVIS:

Your file Votre référence ISBN: 978-0-494-38494-7 Our file Notre référence ISBN: 978-0-494-38494-7

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par télécommunication ou par l'Internet, prêter, distribuer et vendre des thèses partout dans le monde, à des fins commerciales ou autres, sur support microforme, papier, électronique et/ou autres formats.

L'auteur conserve la propriété du droit d'auteur et des droits moraux qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

Conformément à la loi canadienne sur la protection de la vie privée, quelques formulaires secondaires ont été enlevés de cette thèse.

Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant.

Don't measure yourself by what you have

accomplished, but by what you should have

accomplished with your ability. John Wooden

We are what we repeatedly do. Excellence,

therefore, is not an act, but a habit. Arisfofle

To my family and friends

ABSTRACT

Nanopowders and bulk nanostructred materials have gained large interest in recent

years. Bulk nanostructured materials exhibit properties that are far superior in

comparison to conventional micron grained alloys. The fabrication of large scale

nano-grained materials has been achieved in a two step process: (1) the production of

nanostructured aluminium alloy powders and (2) the consolidation of the powder using

a electromagnetic shockwave process.

The first part consists of cryo-milling; the milling of powder in an attritor filled with liquid

nitrogen. This causes successive welding and fracturing events as the powder is

milled, thereby creating the nano-structure. The low temperature prevents the

possibility of recrystallization and grain growth. The alloy used for this work was AI

5356 (AI-5%Mg). Two different types of raw source materials were investigated: pre

alloyed powders and a mixture of aluminum with pure magnesium or an A1 12Mg17

intermetallic. Experiments have been conducted in order to determine the optimum

milling parameters that will simultaneously give a grain size sm aller than 100 nm;

equiaxed milled particles and mechanically alloyed powder (in the case of the

mixture). The optimum milling parameters were established at 15 hours of milling time

with a rotational speed of 300 RPM and bail to powder weight ratio of 24: 1 in the case

of the pre-alloyed powders. For the mixture of pure aluminum with pure magnesium

the parameters were 15 hours, 300RPM and 32:1. The parameters for the mixture

with the intermetallic were 18 hours, 300RPM and 32: 1.

The dynamic magnetic compaction technique was done with a peak pressure of 1.1

GPa. This ultra-high strain rate process minimizes the exposure of the powders to

high temperature and therefore reduces the possibility of recrystallization and grain

growth. Relative densities of compacted pieces obtained ranged from 86.39% to

97.97%. However consolidation characterized by parti cie to particle bonding with a

melted layer was not accomplished.

ii

RÉSUMÉ

Les poudres nanocristallines et les pièces massives nanostructurées ont gagné en

intérêt ces dernières années. Ces pièces massives nanostruturées ont démontré des

propriétés mécaniques supérieures aux pièces conventionnelles dont la taille de grain

est de l'ordre du micron. L'obtention de larges pièces nanostructurées a été élaborée

en deux étapes: (1) la production de poudres d'alliage d'aluminium nanostructurées

et (2) la consolidation dynamique des poudres ainsi produites en utilisant une

compaction électromagnétique par onde de choc.

La première partie consiste au broyage cryogénique ; les poudres sont broyées dans

un attriteur rempli d'azote liquide. Étant donné la très basse température, il s'ensuit

une succession de soudage et rupture des poudres pendant qu'elles sont broyées et

créant ainsi la structure nanométrique. La basse température permet d'éviter la

recristallisation et de ce fait la croissance de la taille des grains. L'alliage choisi est AI-

5356 (AI-5%Mg). Deux types de source différente de poudres ont été utilisés: une

poudre pré-alliée et un mélange de poudres d'aluminium et magnésium pur ou d'un

intermétallique AI12Mg17. Une série d'expériences ont été conduites afin de

déterminer les paramètres qui influencent le plus les résultats escomptés à savoir la

taille des grains en deçà de 100 nm, des particules dont la forme est équiaxe et la

confirmation de la mécanosynthèse (dans le cas du mélange de poudres). Les

paramètres optimaux de broyage ont été établis à 15 heures de broyage avec une

vitesse de rotation de 300 RPM et un ratio massique de balles sur poudres de 24 : 1

pour la poudre préalliée. Dans le cas du mélange avec le magnésium pur, les

paramètres optimaux ont été fixés à 15 heures, 300 RPM et 32 : 1. Le mélange avec

l'intermétallique a aboutit avec 18 heures, 300 RPM et 32 :.

La consolidation des poudres à l'aide de compaction par ondes électromagnétiques a

été effectuée avec une pression de 1.1 GPa. Cette technique implique un très haut

taux de déformation et minimise l'exposition des poudres à une haute température et

iii

de ce fait réduit la possibilité de recristallisation des poudres. Les densités relatives

obtenues ont été de 86.39 % à 97.97%. La consolidation des poudres, caractérisée

par une couche fondue à l'interface des particules qui assure leur liaison, n'a pas été

réalisée.

iv

ACKNOWLEDGEMENTS

1 would like to extend my sincere gratitude to my thesis supervisor, Prof. Mathieu

Brochu for giving me the opportunity to work toward this project on Nanocrystalline

Production by Cryogenie Milling and Dynamic Consolidation. 1 would also like to thank

him for his unconditional support and technical expertise throughout the course of this

project. In addition, 1 would like to express my gratitude to Alcan, CQRDA, NSERC,

and REGAL for extending financial support to this project.

1 am grateful to Dr Sylvain Pelletier and Patrick England of NRC-IMI, Boucherville, for

providing the access to the dynamic magnetic compaction apparatus and offering

guidance during the consolidation of the pieces. 1 would like to thank Ms. Helen

Campbell, Dr. Kelly Sears, Dr Hojatollah Vali, and Dr Florence Paray for their help

with microscopie characterization. 1 would also like to use this opportunity to thank

Ms. Barbara Hanley, Mr. Raymond Langlois, Mr. Pierre Vermette, Dr. Farzad Jalilian

and Mr. Alain Gagnon for their help in logistic and materials. 1 would express my

sincere gratitude to Mr. Bernie Desjardins for proof reading my thesis work.

And many thanks to my fellow graduate students, especially Sean, Graeme, Ramona,

Andreas, Salim, Phuong, Ehab, Cecile, Yaneth, Paula, Ana Maria, Lydia, Vera, Xin,

Nasser, Dominique, Stella, Abdulaziz, Vikram, Geoff, Geremi, Jinbo, Jessica, Melissa,

Abdel, Curtis, Shawn, Eric and George for their precious help and support during my

master. Last but not the least, 1 would like to extend my sincere gratitude and

appreciation to my family without whose help, support and encouragement 1 would not

have been able to pursue advanced degree here at McGiII University.

v

TABLE OF CONTENTS

Chapter 1 ...................................................................................................................... 1

INTRODUCTION .......................................................................................................... 1

Chapter 2 ...................................................................................................................... 6

LITERATURE REVIEW ................................................................................................ 6

2.1. Nanostructured Materials ............................................................................. 6

2.2. Mechanical Milling ........................................................................................ 8

2.3. Mechanical Alloying .................................................................................... 13

2.3.1. The Process of Mechanical Alloying ................................................. 13

2.3.2. Mechanism of Mechanical Alloying ................................................... 19

2.4. Cryogenie Milling ........................................................................................ 22

2.4.1. Microstructure Evolution and Characterization ................................. 23

2.5. Dynamic Powder Consolidation Techniques ............................................. 31

2.5.1. Shockwave Compaction and Consolidation ..................................... 32

2.6. Mechanical Performance of Consolidated Nanostructured Powders ........ 36

Chapter 3 .................................................................................................................... 39

INSTRUMENTATION AND EXPERIMENTAL SET-UP ............................................ 39

3.1. Flow Chart .................................................................................................. 40

3.2. Materials ..................................................................................................... 41

3.3. Cryomill Setup and Design of Experiment ................................................. 41

3.3.1. Cryomill Setup ................................................................................... 41

3.3.2. Design of Experiments ...................................................................... 42

3.4. X-ray Diffraction .......................................................................................... 45

3.5. Dynamic Magnetic Compaction ................................................................. 49

vi

3.6. Microhardness ............................................................................................ 50

3.7. Microscopie Characterization ..................................................................... 51

3.8. Differentiai Scanning Calorimetry Analyses ............................................... 53

Chapter 4 .................................................................................................................... 54

RESULTS AND DiSCUSSiON ................................................................................... 54

4.1. Microstructural Characterization and Evolution of the Powders ................ 54

4.1.1. Starting Initial Powders ..................................................................... 54

A. SEM and X-Ray Diffraction Analyses ............................................... 55

B. Grain size and Lattice Strain Determination by X-Ray Diffraction of the

Pre-alloyed Powders ........................................................................................ 59

A. SEM and X-Ray Diffraction Analyses .............................................. 61

4.1.3. Characterization of the Cryomilled Powders With the Optimum Milling

Parameters After DOE ..................................................................................... 65

A. SEM and X-Ray Diffraction Analyses ............................................... 66

B. Lattice Constant Determination by X-Ray Diffraction ....................... 69

C. DSC Measurements .......................................................................... 72

D. Grain Size Calculation Based on XRD Compared to TEM

Measurements for the Selected Prealloyed Powders ..................................... 74

4.2. Dynamic Magnetic Consolidation (DMC) ................................................... 78

4.2.1. Optical Microscope Imaging .............................................................. 78

4.2.2. Relative Densities Measurements .................................................... 81

4.2.3 Microhardness and Indent Traces .................................................... 82

4.2.4 Considerations for Full Density Attainment.. ..................................... 86

Chapter 5 .................................................................................................................... 87

CONCLUSiONS ......................................................................................................... 87

vii

REFERENCES ........................................................................................................... 90

viii

Chapter 1

INTRODUCTION

LIST OF FIGURES

Figure 1-1: a) Cryogenie rocket turbopumps with impellers in Ti-5AI-2.3 Sn [10]. b)

Assault amphibious vehicle (AAVA71) using AI-5083 in its hull [11] ................. 3

Chapter 2

LlTERATURE REVIEW

Figure 2-1: Calculated volume fraction of atoms at grain boundaries of a

nanocrystalline sol id as function of grain diameter, assuming that the average

grain boundary thickness ranges from 0.5 to 1.0 nm [17] .................................. 6

Figure 2-2: Schematic skethes of the process of mechanical attrition [6] ..................... 9

Figure 2-3: Arrangement of rotating arms on a shaft in the attrition bail mill. .............. 10

Figure 2-4: Schematic illustration of dislocations evolution and nanometric grain size

formation [20] .................................................................................................... 11

Figure 2-5: Schematic evolution of powder during milling[22] ..................................... 11

Figure 2-6: SEM images of aluminum powders after 2, 4, 6, 8, 10 hours of milling[22]

.......................................................................................................................... 12

Figure 2-7: Mechanical milling parameters[1] .............................................................. 14

Figure 2-8: Change with milling time of crystallite size in mechanically alloyed Cu-Mo

alloys [28] .......................................................................................................... 16

Figure 2-9: Schematic illustration of oxides dispersed in a ductile matrix. The ductile

partiel es form lamella and the oxides are located at the boundaries [1] .......... 21

Figure 2-10 : TEM image showing brittle Er203 particles dispersed in a-titanium

aluminide alloy matrix [1] .................................................................................. 21

Figure 2.11: SEM image showing the embedment of Si in a Ge matrix [1] ................. 22

ix

Figure 2-12: Variation of the time exponent n for isothermal grain growth with the

normalized annealing temperature for nanocrystalline Ni powders[46] ........... 25

Figure 2-13:The DSC scan of the AI-Mg cryogenically milled [48] .............................. 27

Figure 2-14: Grain size and lattice strain versus annealing temperature for two

annealing time [48]. ........................................................................................... 28

Figure 2-15: Lattice constant versus annealing temperature for two annealing time .. 29

Figure 2-16: Nanostructured of a as-milled AI-Mg alloy from a dark filed TEM imaging

with its corresponding diffracted pattern [48] ................................................... 30

Figure 2-17: Distributions of grain size estimated from TEM for an annealed sample of

AI-Mg alloy [48] ................................................................................................. 31

Figure 2-18: Schematic circuit of dynamic magnetic compaction [53] ........................ 32

Figure 2-19: Illustration of the compression mechanism in DMC [52] ......................... 33

Figure 2-20: Schematic representation of two states of condensed matter ahead and

behind a propagating shockwave pulse [54] .................................................... 34

Figure 2-21: Examples of Hugoniot curves for a range of solid metals and alloys [54]

.......................................................................................................................... 36

Figure 2-22: Left: Melted layer surrounding Inconel powders; Right: Inefficient bonding

on Ti3AI particles [56] ....................................................................................... 37

Chapfer3

INSTRUMENTATION AND EXPERIMENTAL SET-UP

Figure 3-1: Cryomill Set-Up ......................................................................................... 42

Figure 3-2: X-Ray Diffraction Apparatus ...................................................................... 46

Figure 3-3: Microhardness Apparatus ......................................................................... 50

Figure 3-4: Optical Microscope Set-Up ........................................................................ 51

Figure 3-5: SEM Apparatus, JEOL 840-A. ................................................................... 52

Figure 3-6: TEM Apparatus JEOL JEM 2010 .............................................................. 52

Figure 3-7: DSC Machine ............................................................................................ 53

x

Chapfer4

RESUL TS AND DISCUSSION

Figure 4-1: SEM images of starting powders ............................................................... 54

Figure 4-2: SEM images and schematic representations of milling parameters for AI-

5356 powders ................................................................................................... 56

Figure 4-3: SEM images and schematic representations of milling parameters for AI-

5356 powders ................................................................................................... 57

Figure 4-4: Diffraction patterns showing braadening of AI 5356 peaks for different

milling set-ups ................................................................................................... 58

Figure 4-5: SEM images and schematic representations of milling parameters of AI

and Mg cryomilled powders .............................................................................. 62

Figure 4-6: SEM images and schematic representations of milling parameters of AI

and Mg cryomilled powders .............................................................................. 63

Figure 4-7: Diffraction patterns showing the evolution of Mg peaks and braadening of

AI peaks for different milling set-ups ................................................................. 64

Figure 4-8: SEM illustrations of morphology of cryomilled powders from the three main

starting sources ................................................................................................. 67

Figure 4-9: Diffraction patterns of different starting powders ....................................... 68

Figure 4-10: Evolution of lattice parameter with increasing atomic percentage of

magnesium[51] ................................................................................................. 71

Figure 4-11: DSC curves obtained fram heating ......................................................... 73

Figure 4-12: TEM images of nanostructured AI 5356: a) Bright field; b) Dark field with

the corresponding diffracted plans .................................................................... 75

Figure 4-13: Grain size distribution based on TEM imaging measures ...................... 76

Figure 4-14: Schematic illustration of samples preparation after DMC ....................... 78

Figure 4-15: Compacted samples ranging fram 100 to 50% of milled powder ........... 79

xi

Figure 4-16: Compacted samples ranging from 50 to 0% of milled powder, plus a

sam pie from flaky powders ............................................................................... 80

Figure 4-17: Evolution of relative density with increasing percentage of milled powders

.......................................................................................................................... 82

Figure 4-18: Indent traces on samples ranging fram 100 to 50% of milled powder .... 84

Figure 4-19: Indent traces on samples ranging fram 50 to 0% of milled powder, plus a

sam pie fram flaky powders ............................................................................... 85

xii

Chapter 1

INTRODUCTION

LIST OF TABLES

Table 1-1: Comparative mechanical properties showing potential of nanostructured

materials .............................................................................................................. 2

Chapter 2

LITERA TURE REVIEW

Table 2-1: Weight capacities of the different types of mills [26] .................................. 15

Chapter3


Table 3-1: Materials Information: Different sources and characteristic of initial

powders .......................................................................................................... 41

Table 3-2: Detailed conditions of DOE with different levels seL ................................. 44

Table 3-3: Determination of the fourth parameter as a combination of the first three in

the DOE ............................................................................................................ 45

Table 3-4: Siopes and intercepts for grain size and micrastrain calculations[49] ....... 49

Chapter4


Table 4-1: Grain size calculated values for the pre-alloyed powders .......................... 59

Table 4-2: Values of grain size and lattice strain fram different milling pracess for

comparison ........................................................................................................ 60

Table 4-3: Lattice parameters values of AI and AI alloys under different states .......... 69

Table 4-4: Grain size comparision between XRD and TEM methods ......................... 76

xiii

Table 4-5: Densities and Hardness of Compacted Samples ....................................... 81

xiv

Chapter 1

INTRODUCTION

Research in nanocrystalline materials has increased dramatically in recent years. The

potential of improving the mechanical properties of bulk materials by reducing their

grain to the nanometric scale has been demonstrated by several researchers.

Numerous routes were developed to obtain nanomaterials. Among them, mechanical

milling and alloying (MA) [1], rapid solidification [2], plasma processing [3], vapor

deposition [4], severe plastic deformation processes (SDP) constitute the principal

ones. Severe plastic deformation processes include equal-channel angular pressing

(ECAP), high pressure torsion (HPT), accumulative rolled bonding (ARS), multi

directional forging (MDF), cyclic extrusion and compression (CEC), repetitive

corrugation and straightening (RCS) and twist extrusion (TE). For the present work

the powder metallurgy route has been chosen by means of mechanical milling and

alloying. This route is presented and applied for the production of nanostructured

powders.

Mechanical milling is a solid state process that involves plastic deformation under

shear stress conditions and higher strain rates (101 - 104

S-1) [5]. Lattice defects,

which will yield to high density of dislocations, are the precursors of the nanostructure

formation. They are created during the deformation of the micron scale crystals

composing the powder particles. The dislocations rearrange to form a subgrain

region. The subgrains will be transformed in new grains with a nanometric size.

Mechanical alloying involves a deformation similar to that of mechanical milling. The

difference resides in the formation of an alloy during milling from a mixture of different

metallic powders. Mechanical alloying is a non-equilibrium solid state process that

1

Chapter 1 INTRODUCTION

allows the production of super-saturated sol id solution of metals. The level of

solubility reached is higher compared to the level obtainable from conventional routes

[6].

The alloy chosen in this work is an AI-5% Mg, from the 5XXX series. The

technological importance of this series is related to particular properties such as

corrosion resistance. In addition, Mg (typically <5 wt.%) in AI has the benefit of

increasing strength without compromising ductility [7]. Most important, the alloys

possess a very high strength to weight ratio [8]. shows a comparison of mechanical

properties for a nanostructured AI-5083, a conventional AI-5083 and Ti-5AI-2.5Sn EU.

The n-AI 5083 has a- better specifie ultimate tensile strength (UTS) than the two other

materials. The conventional AI-5083 (AI-4.4Mg-0.7Mn-0.15Cr) is currently used in the

hull of an assault amphibious vehicle (MVA71, see Figure 1-1.) Replacing this

material with an-AI 5083 could save up to 30% of the weight in use. The n-AI 5083

could also replace the Ti-5AI-2.5Sn EU as the material constituting the impellers of the

cryogenie rocket engine turbopumps (see Figure 1-1). This also could save more

weig,ht in the pump due to the lower density of 2.66 for n-AI 5083 compared to 4.48 for

the existing alloy.

Table 1-1: Comparative mechanical properties showing potential of nanostructured materials

UTS Density Spec. UTS Elong. Alloy Temper

MPa g/cc kN.m 1 kg %

Ti-5AI-2.5Sn Annealed 720 4.48 160.7 15

EL! [9]

AI 5083 [9]. H32 320 2.66 120.3 10

AI n-5083 [10] Cryomilled-HIP 462 2.66 173.7 8.4

2


Figure 1-1: a) Cryogenie rocket turbopumps with impellers in Ti-5AI-2.3 Sn [11]. b) Assault amphibious vehicle (AAVA71) using AI-5083 in its hull [12] .

These two applications are examples of the potential of large scale bulk

nanomaterials.

Mechanical milling will be performed on an AI-5356 alloy (AI-5% Mg). This last alloy

already contains magnesium in solid solution. The formation of an AI-Mg alloy with

the same proportion will be achieved through the mechanical alloying process. During

milling, the tendency of ductile powder particles to adhere to container walls, to

agglomerate, and to sinter to form large millimeter-sized particles has been observed.

This is especially true of the ductile aluminum alloys [13]. Milling in a cryogenie liquid

has been introduced as a solution to eliminate these difficulties. Cryogenie milling, or

cryomilling, is a mechanical attrition process in which powders are milled in slurry

formed with milling balls and a cryogenie liquid. Liquid nitrogen is the cryogenie liquid

most often used.

As the process to fabricate nanostructured alloys is a powder metallurgy process, it

requires a consolidation step. The consolidation of the nanostructure powders can be

3


achieved following different options. Pressing and sintering, hot isostatic pressing

(HIP) are part of the static process. Shockwave compaction and consolidation is a

viable option because it allows one to maintain the grain size at a nanometer scale

and to achieve high densities in bulk nanostructure materials [14]. The amount of

pressure (over 1GPa) reached during a shockwave can allow consolidation of the

powders and attainment of full density bulk nanomaterials. The loading time, being

very short (-1I..1s), allows the non-equilibrium nanostructured states to be kept [15]. It

is important, at this point, to distinguish between compaction and consolidation. Both

of them can reach high density values, however a compacted sam pie will not have

particle to particle - bonding. A consolidated sample, on the other hand, is

characterized by maximum particle to particle bonding. Therefore, shockwave

compaction and consolidation must be conducted with a careful choice of parameters

in order to realize the bonding between particles. Different methods are used to

produce a shockwave front that will compact the powders. Explosives compaction,

gun compaction and dynamic magnetic compaction are the most common among

them. The process chosen for this work is dynamic magnetic compaction. In this

technique, magnetic fields are used to compact the powder placed in the core of a

coil. High currents are pulsed in the coil to produce magnetic fields in the bore.

Magnetic fields generated induce current in the armature. The armature will interact

with the magnetic fields to produce an inwardly magnetic force on the tube that will

compact the powder. The advantage of this technique relies on the precision of the

forces applied during compaction. In fact, the current produced can be tailored to

generate an exact force on the powders.

The report will present the study in the following order. Chapter 2 presents a brief

literature review on the production of nanostructured materials followed by a

description of dynamic consolidation technique. Topics included are nanostructured

materials definition, process of mechanical milling, mechanism of mechanical alloying

4


and presentation of cryogenie milling. Efforts are also taken to introduce magnetic

compaction and the evaluation of pieces made from this consolidation method.

Chapter 3 iIIustrates and describes the instrumentation and experimental setup

utilized for producing the nanostructured AI-Mg alloys powders and consolidating them

into bulk pieces. The setup for characterization of microstructures and mechanical

properties are also included.

Chapter 4 presents the experimental results of the work combined with the discussion.

The evolution of the powder with the variation of the milling parameters set by the

design of experiments is presented and explained in this chapter. Microstructural

characterization and selection of the powders for the dynamic compaction are also

presented and finally, mechanical evaluation and optical analyses of the compacted

pieces are executed.

Chapter 5 closes the work with summary and conclusions.

5

Chapter 2

LlTERATURE REVIEW

2.1. Nanostructured Materials

Nanostructured materials are characterized by a m icrostructu rai length in at least one

dimension being up to 100 nm [16]. The main characteristic of a nanostructured

material is the high ratio of atoms in the grain boundary region compared to atoms in

the core of the grain.

100

en

I~ 1.0 nm 1 'i 80 "C \

0.5 nm c: ::::> 0

\ III 60 c: ci! c!i "-.5 40

" cn

"-E

~ 20 ......... <f!-

0 1 10 100

d (nm)

Figure 2-1: Calculated volume fraction of atoms at grain boundaries of a nanocrystalline solid as function of grain diameter, assuming that the average grain boundary thickness ranges from 0.5 to 1.0 nm [16]

This significant difference when compared to conventional materials explains why

nanostructured materials exhibit different physical, mechanical and chemical

properties.

6

Chapter 2 LlTERATURE REVIEW

One should not confuse nanocrystalline materials with nanoparticles, as both

constitute nanomaterials. The former are, as defined in the previous paragraph, a

polycrystalline bulk materials possessing grain sizes sm aller than 100 nm in at least

one dimension, while the latter refers to ultrafine dispersoid particles with diameters

below 100 nm [16].

Structural materials are among the potential applications of bulk nanomaterials since

measured strength are several times higher than for conventional grain size materials.

The Hall-Petch relationship was initially developed for micron-scale grain size

materials and dictates the dependence of yield strength to the grain size, in metals

[17]. As the grain size decreases, an increase in the yield strength occurs, as

expressed in equation 1

Equation 1

Where ay is the yield stress of the material, ao is the friction stress, k is a constant and

d the grain diameter.

Experimental results have shown that the Hall-Petch relationship can be extended for

materials with grain size down to about 20nm. At this critical grain size, nearly 50

vol.% atoms are present at grains boundaries or triple junctions (see Figure 2-1). This

inflection point is called the Hall-Petch breakdown. The breakdown has been

aUributed to different deformation mechanisms that become dominant once the grain

size is reduced down below a critical value [18]. Dislocations pile up is not the major

deformation mechanism due to the fact that grains under the critical value (- 20nm)

cannot sustain a single dislocation within the grain. Therefore, bulk nanocrystalline

materials with grain size ranging between 20 and 100nm will have higher mechanical

properties according to the Hall-Petch relationship. The aUainment of the nanometer-

7


scale grains in bulk pieces can be achieved by various approaches which are rapid

solidification [2], plasma processing [3], vapor deposition [4], torsion straining under

high pressure [19] and equal channel angular pressing. The powder metallurgy route

consists of production of nanocrystalline powders and further consolidation into bulk

pieces. This route is more attractive over the other ones due to its capability of

producing powders in large scale and of maintaining the nanometer level after

consolidation. The starting powders can either consist of particles with an average

particle size distribution smaller than 100nm or be nanostructured, meaning they have

a micron scale particle size distribution but possess average grain sizes ranging

between 10-100nm -[20]. Nanostructured powders can be obtained through severe

plastic deformation, such as mechanical milling. The maintaining of the nanometer

grain size at the end of the fabrication process remains the most important challenge

in the production of nanostructured powder.

2.2. Mechanical Milling

Mechanical milling of a metallic powder is a solid state process that involves plastic

deformation under shear stress conditions and high strain rates (101 - 104 s-1)[5].

Lattice defects are created during the deformation of the micron scale crystals of

which the powder particles are composed. Plastic deformation of the powders

originates from the collision with the milling media where some kinetic energy is

transferred, a~ seen in Figure 2-2. Impact energy is related to the momentum of the

milling media. Collision speed of the balls during milling determines the type of milling

which can be separated into two groups, high energy and low energy bail mills.

8

Chapter 2 L1TERATURE REVIEW

lard Steel or "WC" Balls

l Figure 2-2: Schematic sketches of the process of mechanical attrition [5]

Shaker mills can provide high energy milling due to their high frequency and low

amplitudes of vibration. These mills are suitable for small powder quantities (10cm3)

and research purposes [21]. A low energy milling set up can also be used and it

includes attrition mills and tumbler mills. The main advantage of these mills is their

larger milling capacity.

The basic principle of mechanical attrition, as illustrated in Figure 2-3, consists of

powder particles put together with stainless steel or tungsten carbide or ceramic balls

in a container which is shaken or agitated by a rotary shaft. The density of the

grinding medium should be high enough so that the balls create enough impact force

on the powder [1].

9


Rota/log impeller

Figure 2-3: Arrangement of rotating arms on a shaft in the attrition bail mill.

As the powders are hit between two milling balls, they are subjected to the successive

steps of flattening, fracturing and cold welding. In the mean time, at a crystalline level,

a nanometer grain size is formed through a three-stage process, as illustrated in

Figure 2-4. It starts with the localization of deformation into shear bands containing a

high dislocation density. The second stage is a combination of annihilation and

recombination of dislocations, resulting in a nanometer-scale sub-grain. As the milling

progresses, the subgrain structure extends throughout the sam pie and the final stage

being the transformation of sub-grain boundary structure to randomly oriented high

angle boundaries [20].

10

Chapter 2

00 LlTERATURE REVIEW

~' .... .'"

$, • . . , .' , ~.. ,~

INCRJ!ASlNG l\-llLLING To.m

============================~~

Figure 2-4: Schematic illustration of dislocations evolution and nanometric grain size formation [20].

Figure 2-5shows the stages schematically and Figure 2-6 depicts Scanning Electron

Microscope (SEM) rnicrographs of the evolution in morphology of pure aluminum

powders cryomilled for different lengths of time [22]. The first stage consists of

flattening particles into flakes. After milling for 2 hours, during stage 2, flakes

formation continues, especially for finer particles and the flakes weld to each other. A

multilayer structure is developed by cold welding. After 4 hours of milling (during

stage 3), the multilayered elongated particles are transformed into a relatively

equiaxed shape. Multilayered flakes fold and reweld without any preferred orientation,

causing a convoluted structure. The layers become convoluted rather than linear. For

milling times of longer than 10 hours, the internai layered structure has disappeared

being replaced by a homogeneous structure and the powder is equiaxed and relatively

spherically shaped [22].

o Stage 0 Stage 1 Stage 2 Stage 3 Stage 4

Figure 2-5: Schematic evolution of powder during milling[22]

11

Chapter 2 LITERA TURE REVIEW

Figure 2-6: SEM images of aluminum powders after 2, 4, 6, 8, 10 hours of milling[22]

Powder characteristics are important in the production of high density compacts. In

the case of cold pressing, aUainment of high green density could be compromised due

to the hardness of the milled powders. An important characteristic is the morphology

of the particles. Hard particles will reach higher green densities, when cold pressed, if

the powders have a spherical shape. Spherical powders tend to have a good

flowability which makes the aUainment of high green density possible. Powders loose

their spherical morphology during milling. The equiaxed morphology is the closest

that powders can be to the spherical shape (see Figure 2-6).

12


The next two sections describe two applications of mechanical milling. The first

focuses on mechanical alloying which is a process that enables the production of new

alloys from elemental powders. The second one is cryogenie milling which is a

process in which milling takes place in cryogenie media.

2.3. Mechanical Alloying

2.3.1. The Process of Mechanical AI/oying

Mechanical alloying _ (MA) is a solid-state powder processing technique involving

repeated welding, fracturing, and rewelding of powder particles. This technique has

been used to synthesize equilibrium and metastable phases [1].

The interest and one of the greatest advantages of MA resides in the possibility of

formation of novel alloys that can not be prepared by liquid metallurgy, e.g, alloying of

normally immiscible elements, such as AI-Ta [23] and AI-Nb [24]. This is possible

since MA is fully a solid-state processing technique and therefore limitations imposed

by phase diagrams do not apply [25].

Severa 1 parameters play distinctive roles in MA, they are summarized in Figure 2-7. It

should be noted that the parameters described in Figure 2-7 are not totally

inde pendent.

13


Types of mill: High or Low energy

Materials of milling tools: Types of milling media:

Stainless Stee/s or Rods or Balls

Tungsten Carbides or Ceramics

Mechanically 1 Milling 1 Milled Powders Milling temperature

atmosphere 1 1

Milling media-to-powder

1 weight ratio -

l Milling speed J 1 Milling time 1 Process control agent

Figure 2-7: Mechanical milling parameters[1]

The design of a milling experiment has to be done after careful consideration of ail the

parameters.

1. Types of mills and milling media

Milling apparatus can be regrouped into three categories; the following table shows

their respective typical capacities.

14


Table 2-1: Weight capacities of the different types of mills [26]

Mill type Sam pie weight

Mixer mills Up to 2 X 20 9

Planetary mills Up to 4 X 250g

Attritors 0.1 -100 kg

2. Milling temperature

Temperature has a crucial role in the mechanical alloying mechanism due to the fact

that atomic solubility varies with temperature. It was found in the case of the

production of AI-Mg alloy, using a Retsch PM 400-MA planetary mill[27] , that

temperature of milling strongly affected the concentration of magnesium in the alloy.

The concentration of magnesium went from 2-3 at. % to almost 25 at% when the

milling temperature was decreased from 70-80 Oc to 20-30 oC. At higher milling

temperatures, the formation of intermetallic phases is favored [27]. This was also the

case in another study, conducted on a Cu-Ag alloy. It was found that the extent of

solid solubility was reported to decrease at higher milling temperatures during

planetary bail milling of a Cu ± 37 at% Ag powder mixture. It was noted that a mixture

of amorphous and crystalline (supersaturated solid solution) phases was obtained

during milling at room temperature while, only a Cu±8at%Ag solid solution was

obtained on milling the powder at 200°C [1].

3. Process control agent

A process control agent (PCA) is used in mechanical milling in order to mediate the

welding of particles. The powder particles indeed get cold-welded to each other,

15


especially if they are ductile, due to the heavy plastic deformation experienced during

milling. A balance has to be maintained between cold welding and fracturing of

particles in order to achieve alloying. The PCAs can be solids, liquids, or gases and

they are mostly, but not necessarily, organic compounds [1]. The surface of the

powder particles absorbs the PCA, thus minimizing cold welding between powder

particles. This inhibits agglomeration and promotes fracturing of the particles [1].

4. Milling time

Change in crystallite size, microdeformation, dislocation densities and shape of the

milled particles are affected by the milling time. As an example, in the case of Cu-Mo

alloys, the grain size and particle size decrease with increasing milling time and

approach a minimum value of 20-30 nm. Increasing the milling time beyond this value

results in an increase in the grain size up to 40 nm (Figure 2-8) [28]. This increase

can be due to adynamie recrystallization, grain rotation and growth process

associated with a combination of high local temperature generated during milling and

a large amount of powder with stored energy after severe plastic deformation [28].

100 .1% al. Mo -- e8% al. Mo E 80 e E-CD .e .t::! 60

U)

~ e • 40 }§ • e e e IJ) • ~ 20 • Ü • 0

0 20 40 60 80 100

Milling Ti me (h)

Figure 2-8: Change with milling time of crystallite size in mechanically alloyed Cu-Mo alloys [28]

16


5. Milling speed

The energy put into the powder during milling is increased in direct relation to the

milling speed. However, the maximum speed that could be employed encounters

limitations, depending on the design of the mill. In the case of an horizontal mill,

increasing the speed of rotation will also increase the rotational movement of the balls

and above a critical speed, the balls will be pinned to the inner walls of the vial and will

not fall down to impact the powders. Therefore, the maximum speed should be just

below this critical value so that the balls fall down to produce the collision energy [1].

That critical value can be determined experimentally. Calka et al. [29] reported that

when vanadium and carbon powders were milled together at different energy levels,

the final constitution of the powder was different for each level. For example, at a very

low milling speed, the powder consisted of nanometer-sized grains of vanadium and

amorphous carbon, which on annealing formed either V2C or a mixture of V + VC. At

intermediate speed level, the as-milled powder contained a nanostructure, which on

annealing transformed to VC. At the highest speed level, the VC formed directly on

milling [1].

6. Milling media-to-powder weight ratio

The milling media-to-powder weight ratio, also noted as ball-to-powder weight ratio

(BPR) has a significant effect on the time required to achieve a particular phase in the

powder being milled. The higher the BPR, the shorter the time required [1]. The study

of the effect of ball-to-powder weight ratio was elaborated during the milling of an

elemental Ni-AI powder mixture containing 25 at. % of aluminum [30]. It was found

that by decreasing the ball-to-powder weight ratio from 6.5 to 3.25, the start time of the

formation of NbAI phase was delayed from -20 to -40 hour. Similar conclusions were

found in a study on Ti ± 33 at.% AI powder mixture milled in a SPEX mill. The

17


formation of an amorphous phase was achieved in 7 hours at a BPR of 10: 1, in 2

hours at a BPR of 50:1 and in 1 hour at a BPR of 100:1 [31]. The number of collisions

per unit time increases at higher BPR because of an increase in the weight proportion

of the balls and consequently more energy is transferred to the powder particles and

so alloying takes place faster [1]. The desired quantity of the powder after milling and

the time doing milling will be the factors determining the optimum BPR.

7. Milling atmosphere

The atmosphere of milling is very important due to the contribution on the

contamination of the powder. The large surface area resulting from small powder

particles size and increasing during milling contributes to the powder's susceptibility to

contamination [25]. In order to avoid such contamination, the powders are milled in

containers that have been either evacuated or filled with an inert gas such as argon or

helium. High-purity argon is preferred in order to prevent oxidation of the powder [1,

32]. Glove boxes with a controlled atmosphere are also used to load and unload the

powder before and after milling. These glove boxes are evacuated and refilled with an

inert gas, usually argon [1]. The use of nitrogen to prevent oxidation has been limited

because of its reactivity with metals, except where the production of nitrides is desired

[1]. The section describing cryogenie milling in liquid nitrogen explains the role of

aluminum nitrides in contributing to an increase in the yield strength and thermal

stability

8. Materials composition of milling tools

They can be made of hardened steel, tool steel, hardened chromium steel, tempered

steel, stainless steels, tungsten carbides or ceramics. Depending of the type of

metals to be milled, the choosing of appropriate media must be given careful

consideration. Milling material is a source of contamination of the powder being

18


milled, therefore it is always desirable, whenever possible, to have the grinding vessel

and the grinding medium made of the same material as the powder being milled [1]. A

study on alloying elemental silver and tellurium powder using agate and zirconia as

milling media has resulted in formation of silver silicate (Ag2Si03) and zirconium

telluride (ZrTe2) simultaneously to the Ag2Te phase formation. However, using

tungsten carbide instead of the ceramics for the same powder mixture has not

resulted in the undesired phases [33].

2.3.2. Mechanism of Mechanical Alloying

The mechanism described earlier for mechanical milling also applies for mechanical

alloying. During the milling, the powder partieles are subjected to flattening, cold

welding, fracturing and rewelding, on repeated cycles. In addition here, two or three

types of starting powders are put together in the mill with a given weight percentage.

The powder partieles are plastically deformed by the force of impact of the balls and

the energy input leads to work hardening and fracture. The new surfaces thus created

weld together and form partiel es with an increase in size. The partieles formed have a

composite layered structure containing the starting constituents. Steady-state

equilibrium is reached, after milling for a given period, and it is characterized by a

balance between the rate of welding and fracturing. At this stage of homogenization,

each partiele represents the targeted proportion of ail the original constituents.

Simultaneously, the formation of the nanostructure takes place and the volume

fraction of grain boundary regions also increases, enhancing grain boundary diffusion

mechanism of solute atoms. The increase in temperature during milling favors the

diffusion mechanism and the process of alloying the constituents is established. Even

though mechanical alloying is performed at room temperature, it might be necessary

19


to anneal the milled powder to complete the alloying, specially when intermetallic

formation is desired [1].

The three types of material combination that can be created using MA are ductile

ductile, ductile-brittle and brittle-brittle as in the case of metal oxides and

intermetallics.

A. Ductile-Ductile Components

This is the ideal case for mechanical alloying. In order to have plastic deformation that

will result in cold wëlding, a combination of ductile materials is best suited. Ductile

materials also form a layer on the balls surfaces, preventing wear of the milling media

and contamination of the powder [34]. When the homogeneous structure of the

powder is obtained, the completion of the MA process is considered done [1]. A work

on Ni-Cr alloy has shown that it is possible, starting from elemental powders mixture,

to produee an al/oy having the same magnetie behavior as a homogeneous Ni-Cr

alloy produced bya melting procedure [35].

B. Ductile-Brittle Components

The researeh on oxides dispersion strengthening (ODS) falls into this category. The

steps described in the section on the proeess of mechanical milling apply to the ductile

eomponent as weil. While the ductile partiel es are flattened, fractured, and cold

welded, the brittle oxides or intermetallic partiel es get fragmented. The brittle

fragmented partiel es become trapped and oecluded between ductile deformed

particles, as shown in Figure 2-9 [1]. An illustration of the dispersion of Er203 in Q

titanium aluminide alloy created with a Transmission Electron Microscope is shown in

Figure 2-10 [1].

20

Chapter 2 L1TERATURE REVIEW

Figure 2-9: Schematic illustration of oxides dispersed in a ductile matrix. The ductile particles form lame lia and the oxides are located at the boundaries [1]

Figure 2-10 : TEM image showing brittle Er203 particles dispersed in a-titanium aluminide alloy matrix [1].

c. Brittle-Brittle Components

The fact that the two materials are not ductile, precluding cold welding, it could be

assumed that MA will not be achieved. However, mechanical alloying has been

reported for two brittle components, silicon and germanium where a solid solution from

21


brittle elemental starting powders (Si and Ge ) was produced after 8 hours of bail

milling with a calculated lattice parameter approximately equal to that of a similar melt

formed alloy [36]. In the case of two brittles components, the particles get fragmented

during milling and their size gets reduced continuously until they reach a limit where

further reduction is not possible, this is called limit of comminution [37]. Like the

embedment of a brittle component in a ductile one, two brittle materials will be.have

the same way with the harder component getting embedded in the less hard

component. Therefore, the harder Si particles are embedded in the less hard Ge

matrix, as shown in Figure 2.11 [1].

Figure 2.11: SEM image showing the embedment of Si in a Ge matrix [1]

2.4. Cryogenie Milling

Milling at cryogenie temperature is done when powders are milled in a slurry formed

with milling balls and a cryogenie liquid such as liquid nitrogen [20]. Cryomilling was

first introduced in the literature for a composite, AI-AI20 3 [38]. [38]. Early research

focused on mechanical alloying of metallic and non-metallic constituents for the

development of dispersion-strengthened alloys for creep resistance improvement [34,

22


39, 40]. The organic process control agent (PCA) added to the milling mixture, along

with liquid nitrogen, leads to the formation of oxide (AI20 3), and nitride (AIN)

compounds [41, 42]. These compounds, which are contaminant products or by

products, are useful as dispersoids in the alloys and produce an increase in strength

by acting as dislocation pinning points. Thus, further consolidation of the cryomilled

powder at high temperature would not result in grain growth and the hardness

reached during cryomilling will not be lost. These characteristics allow the creation of

hard homogeneous powder amen able to powder metallurgy (PM) processing when

aluminum powders are used in the cryomilling process [22].

Metals that have a face centered cubic structure, like aluminum and copper base

alloys, stay ductile even at very low temperatures and exhibit plastic deformation [43].

The development of nanostructures is dominated by lattice strain [13], as described in

Figure 2-4. Milling in cryogenie media has been successful for nanostructure

formation of ductile materials, such as aluminum and its alloys. Ductile alloys are

indeed difficult to mill in a non-cryogenie bail mill because of the tendency of powder

particles to adhere to container walls, to agglomerate and to sinter to form large

millimeter-sized particles [13].

2.4.1. Microstructure Evolution and Characterization

A. Grain Growth Kinetics

Thermal stability of materials is related to grain growth mechanisms. In order to

maintain the higher mechanical strength gained by the nanocrystalline structure, the

microstructure evolution under elevated temperature should be understood and if

possible, controlled. Studies on isothermal grain growth kinetics have suggested that

normal grain growth can be expressed as [44, 45]

23

Chapfer 2 LlTERATURE REVIEW

D 2 -D 2 = kt o Equation 2

where 0 is the grain size at time f, Do is the initial grain size and k is the temperature

dependant rate constant. Equation 2 has been unsuccessful in predicting grain

grawth except for high purity metals at high homologous temperatures, which is a ratio

of the actual temperature to the melting point (TIT M) [46]. Therefore, a modification of

the empirical relation was done to obtain better fitting with the experimental results.

The modified equation is represented by Equation 3:

Equation 3

where n (;zV.5) is an empirical time constant determined experimentally and varies

with annealing temperature. Figure 2-12 shows the variation of the exponent n in the

case of nanocrystalline nickel cryomilled powders [46]. This figure demonstrates that

higher temperature is needed to achieve grain growth. Ideally, when the value of n is

0.5 the only driving force on the grain boundary results fram its curvature. However,

deviations fram this value are currents [47] and are attributed to pinning of the grain

boundaries caused by the oxides and nitrides dispersed in the milled powder. This is

the basis of the oxides and nitrides strengthening mechanism described earlier.

24


O.!}r------------------,

•

• •

• 0.1'---...1..--........ ---1.------...1..------'

0,4 0,5 0.6 0,7

Normalized Annealing Temperature (TIT M)

Figure 2-12: Variation of the time exponent n for isothermal grain growth with the normalized annealing temperature for nanocrystalline Ni powders [46]

Consequently, the pinning forces of the oxides and the nitrides exerting a drag on the

grain boundaries migration should also be taken into account in the prediction of grain

growth kinetics. As a result, equation 4 was developed to take into consideration the

positive effect of the dispersoids [45].

Equation 4

where Dm is the maximum grain size that results due to the pinning force.

The rate constant k can be expressed, in an Arrhenius form, as follow,

Equation 5

25


where Q is the activation energy for grain growth, ka is a constant that is inde pendent

of the absolute temperature T, and R is the molar gas constant [46].

Prior to grain growth, in the annealing process, are recovery and recrystallization [48].

Hence, characterization of grain growth should also imply analyses of the two

preceding mechanisms in order to predict the behavior the cryomilled powder when

subjected to high temperature. The objective is to maintain the grain size below

100nm in order to benefit from the higher mechanical properties related to a

nanometric grain size.

B. Annealing Characterization

Mechanical milling induces large plastic deformation within a material. Therefore, the

material milled contains a large amount of stored energy and when it is subjected to

annealing at elevated temperatures, it will revert to a lower energy state by structural

evolution during recovery and recrystallization [47]. During this annealing process, the

material goes through different phase transformations that can be tracked by enthalpy

changes with Differentiai Scanning Calorimetry (DSC). F. Zhou et al [48] have

analyzed the recovery and recrystallization of a nanocrystalline AI-Mg alloy prepared

by cryogenie milling. The DSC curve (Figure 2-13), performed with a heating rate of

16 Klmin, indicates two distinct exothermic peaks prior to the occurrence of the

melting process (around 530 OC). The first peak is at 164 oC and the second peak at

327 oC. The enthalpy release (llH) was obtained by calculating the area under each

exothermic peak and the values are 450 J/mol, and 410 J/mol, respectively [48].

26


(a) HR: 16 Klrnir'

""

• ~ ___ ~~ .... _~ ...... ___ .J--.~ .. _ .....

---..., .~ '. ::1 .. .ci !. 1 v7'C 1 :: 1 Hj4~::' ! 0

1 iI ~ • 1 " 'iiI 1

, <J.I

l-

" J :I: 1,1

a. . 'd ::l 1 g , ,/ 11X) 200 XC 400

/ w

a j'DO 200 ~C<J .. 00 seo SCi} 700

T «q

Figure 2-13:The DSe scan of the AI-Mg cryogenically milled [48]

The exothermic peaks iIIustrate the transformations that occur in the alloy during

annealing. The first peak corresponds to a recovery and the second, to a

recrystallization of the powders [48].

C. Grain Size, Lattice Constant and Lattice Strain Characterization with X-Ray

Diffraction

XRD is often used to calculate grain size in nanostructured materials (5-100 nm) [49].

XRD analyses are limited to the grain size under 100 nm, because the diffraction peak

line broadening that results from a grain ranging from 100 to 300 nm is considered to

be negligibly sm ail [50]. Methods used to estimate grain size with XRD are based on

the broadening of the diffraction peaks, calculated upon the full width half maximum

(FWHM) and the integral width (IntW). Complete descriptions of the methods used to

calculate the grain size are included in the experimental procedure section.

27


Along with grain size calculations, XRO is also used to calculate the lattice constant

and lattice strain of the milled powders. The calculated values allow correlating the

grain size, the lattice strain and the lattice constant to the evolution of the ose curve

in order to track phase transformation in the milled alloy. F. Zhou et al [48], in their

study of the microstructure evolution of AI-Mg cryomilled alloy, noted that the grain

size increases from 23 nm in the as-milled powder to a value of 32 nm at the

temperature of the first peak of ose (164°e) and finally to reach 43 nm at 3700 e after

the second peak (see Figure 2-14). This last value indicates the thermal stability of

the alloy. The lattice strain (the root mean square: rms or <e2> 1/2) went from a value of

0.21 ± 0.03% to a value of 0.08% (see Figure 2-14) after the first peak and remained

unchanged after the second peak [48].

w

• 2 min . 45 6 GOmin.

E + -S § 4Qf '" .~ 3-5 ~ '" 1 t :lOr il/ r ~\ < 25l . "T

2Q L.:-"-,-~ .... ~--_..J .. _._. __ l-_-,,"-~_ ••.• _-'----

025r • 2 mm. r 11 60 nnn.

o 2Q~ .'

~ f ~ a 15r ~ ~

i 01û~ 005 ~

l

.. o 00 ! __ .~J.._ .. __ J........-~----L ____ .. j _~ .. _~.l..--..L-_~~_..o.-.-....!_M-'O""~

() 50 100 150 200 250 300 350 400

Annealing tomperatUfe (oC)

Figure 2-14: Grain size and lattice strain versus annealing temperature for two annealing time [48].

28


The lattice constant is used to examine the changes in solubility of a solute in the

lattice, in this case of Mg in AI, during the annealing pracess. It was found [48] (see

Figure 2-15) that the lattice constant (ao) decreases fram -0.4078 to 0.4073 nm in the

region of the first peak (150 to 250 OC) corresponding to a decrease of Mg content in

the AI lattice, fram about 7.6 to 5.4 at%. In the region of the second peak (300 to

370°C), an increase of the lattice parameter to 0.4078 nm was found, indicating that

ail of the Mg atoms were again dissolved in the AI lattice. The association between

the lattice constant and the solubility percentage is estimated fram the relationship

between Mg concentration and ao for AI(Mg) solid solutions [51].

4.G8G

4.005

•

• • •

1

~ 1

• 2 min Il 60 min.

1

o 50 100 150 200 250 3{l(l 3-[">0 40{)

Annealing temp€lraturel ('C)

Figure 2-15: Lattice constant versus annealing temperature for two annealing time

29


D. Grain Size Characterization with TEM

TEM images analyses is a direct method for grain size measurement. Dark field

imaging shows grain areas of a selected diffraction direction. Despite the fact that

TEM allows one to see the actual image of a grain, it requires more sam pie

preparation and the analysis is done on a small quantity of the powder milled. This

later implies an uncertainty as to the correlation between the sam pie observed and the

total quantity of the powder [49]. Having a high number of measured grains (see

Figure 2-17) and pairing the TEM technique with XRD grain size analyses is

recommended. Figure 2-16 shows an example of TEM image analyses of an AI-Mg

cryomilled alloy [48]. The diffracted planes form ring patterns showing the crystallinity

and absence of a texture in the cryomilled powders. Each concentric circle represents

a diffracted plane of the powder.

Figure 2-16: Nanostructured of a as-milled AI-Mg alloy from a dark field TEM imaging with its corresponding diffracted pattern [48]

30


0.5

150 °C/60 min 0.4

--.. -:!< ~ D'" 35.7 nrn c:: 0.3 .Q tî 391 grains ~ "- 0.2 ..8 E ;;, z

0.1

0.0 0 100 200 300 400

Grain size (nm)

Figure 2-17: Distributions of grain size estirnated from TEM for an annealed sample of AI-Mg alloy [48]

2.5. Dynamic Powder Consolidation Techniques

Static routes for consolidation of metal powders into shapes usually consist of

pressing followed by sintering. The two mechanisms can be done in two steps, which

are cold pressing then sintering; or in one step, done by hot pressing or hot isostatic

pressing.

Dynamic consolidation is a viable approach for both maintaining the grain size at a

nanometer scale and achievement of high densities in bulk nanostructured materials

[14]. The choice of dynamic compaction over static is justified by the applied pressure

that can reach over 1 GPa. This high pressure allows achievement of near full density

due to the high energy input and the short duration time (-1I-1s). The short application

time of that high pressure permits maintenance of the non-equilibrium nanostructured

states [15].

31


2.5.1. Shockwave Compaction and Consolidation

This principle relies on the generation of a shockwave front that will pass through

green powder compacts in order to consolidate them. It is the energy of the

shockwave that consolidates the powders into dense materials. The densification

occurs at an extremely high strain rate (1 07_1 08S-1) due to pressure levels exceeding 1

GPa imposed in less than 11Js.

Shockwaves can be generated by means of explosives, gas gun or magnetic pulse.

ln this project, the fo~us will be on dynamic magnetic compaction.

A. Dynamic Magnetic Compaction

This technique, also called Electromagnetic Forming (EMF), is based on the

compression of powders within a conductive container which has been placed in the

bore of a coil, by repulsive forces generated by the opposite magnetic field in adjacent

conductors. The primary field is developed by the rapid discharge of a capacitor

through the forming coil and the opposing field results from the eddy current induced

in the armature (Figure 2-18and Figure 2-19). The powders are pressed to full density

via the transmiUed impact energy from the armature, with the entire compaction

occurring within a few microseconds of the compaction cycle [52, 53].

Figure 2-18: Schematic circuit of dynamic magnetic compaction [53]

32


Armature Powder

"'Current ... Magnetie flux c:::> Mag netie pressure

Figure 2-19: Illustration of the compression mechanism in DMC [52]

B. Conditions at a Shock Front and Equations of States

A schematic illustration of a propagating shock front is presented in Figure 2-20. In

front of the planar shock wave is the undeformed material and behind it is the

deformed material.

33


A ~

Pressure i·"AB

.1 Lltt\

: p 1

1 r

Time

C

Figure 2-20: Schematic representation of two states of condensed matter ahead and behind a propagating shockwave pulse [54]

Equations of states during shockwave consolidation appear as conservations of mass,

momentum and energy. The undisturbed region ahead of the shock front is

characterized by density Po, volume Va, internai energy Eo, pressure Po [54]. The

particle velocity refers to the velocity u that a given element in the material acquires,

as a result of the shock wave passing over the element. The shock velocity is the

velocity U with which the disturbance moves through the body. The undeformed

material to the right of the shock front can be envisioned to move toward the shock

front with a velocity of U-u. Conservation of mass is then written as

peU -u) = PoU Equation 6

34


Momentum conservation is written

Equation 7

The shock front applies on the system a pressure equal to P. The pre-existing

pressure is Po. The net force acting on the system is P-Po. The shock wave

accelerates a mass equal to po *U per unit time to the velocity u, which results in a

momentum transfer of po *U*u of the mass element.

Final/y, conservation of energy is written

(E _ E ) = (P + ~) )(Vn - V) o 2 Equation 8

Equations 6 to 8 are referred as the Rankine-Hugoniots equations [54]. By

rearranging these equations, the shock velocity and the particle velocity can be

expressed by:

u= (P-PJ

(Vo - V)

Equation 9

Equation 10

Equations 6 to 10 can lead to the construction of the Rankine-Hugoniot Curve which

represents a two-dimensional curve relating pressure and volume for a material after

the passage of a shock wave. The P-V Hugoniot is a locus of ail the possible end

states that can be achieved behind a single shock wave passing through a material at

a given initial state. The Hugoniot curve is not an equation of state of a material

35


because it's a curve containing only one independent variable instead of a surface

(two independent variables) that defines properly an equation of state in a

thermodynamic space.

eo~--~----~-----r----'-----'

50

-• A. <:J ... w CI: 30 j co co U.I a: CI. 20

10

oL---~----~----~----~--~ 0.75 0.80 0.86 O.tO 0.815 1.00

RELATIVE VOLUME (VIVo)

Figure 2-21: Examples of Hugoniot curves for a range of solid

metals and alloys [54]

The path followed by steady shock compression is the chord connecting the initial

state and final shocked state, the" Rayleigh line".

2.6. Mechanical Performance of Consolidated Nanostructured Powders

Mechanical performance of pieces from powder metallurgy consolidation routes relies

deeply on the degree of density attained after consolidation. Legros et al .[55] and

Tellkamp et al [10] experimented with warm compaction and hot isostatic pressing,

36


respectively for the attainment of high density and fulfill consolidation of the powders.

Dynamic consolidation techniques have to be used carefully order to achieve full

density and bonding. The process parameters, mostly the impact energy, must be

controlled to assure the bonding between particles. The bonding can be obtained with

a melted layer of the particles during compaction without compromising ail the

nanostructure formed in previous processes. The melted layer guarantees the higher

mechanical properties because the powders are consolidated. When the particles do

not bond with a melted layer, properties rely only on mechanical interlocking, the

powders only having been compacted. This phenomenon has been observed by

Meyers et al [56] where a surrounding layer melted on shockwave consolidated

Inconel 718 powders. Figure 2-22 illustrated examples of shockwave compacted

particles with a surrounding melted layer on Inconel 718 and an inefficient bonding on

T 3AI sample.

Figure 2-22: Left: Melted layer surrounding Ineonel powders; Right: Ineffieient bonding on Ti3AI partiel es [561

The strength of nanostructured metals is often increased at the expense of ductility

[57, 58]. One option to improve ductility is to incorporate larger grains particles to the

nanocrystalline powder at the consolidation stage as Legros et al. [55] and Tellkamp

et al. [10] did with copper and AI 5083 alloy respectively. The tensile tests performed

37


on the samples resulted in elongation at fracture of 2.1% and 8.4% respectively, both

of which are higher than with typical nanocrystallinity [59]. In these studies, the

formation of larger grains was achieved by recrystallization during warm compaction

[55] or by bimodal grain growth during hot isostatic pressing (HIP) consolidation and

extrusion [10].

8rochu et al [60] have attempted ta achieve full density and improved ductility by

controlling the shockwave process. Their work aimed to produce pieces where the

grains melted on the surface, producing an ultra-fine layer structure but the core of the

grains maintained the nanostructure characteristic. However an oxide layer present at

the surface of the particles prevented bonding and no liquid film was obtained.

Therefore, the material obtained was made with mechanically interlocked particles.

38

Chapter 3


ln this chapter, the materials, the procedure and the instruments used are presented

in the order they appear in the chronology of the whole process. The flow chart

summarizes these steps.

39

Chapter 3

3.1. Flow Chart

Startmg Matenals

Expenmental Setup

Execution

Mechanlcal Evaluation

1

1

1

INSTRUMENTAL AND EXPERIMENTAL SET UP

Design of Experiments

Grain Size Lattice Strain

Lattice Constant Mechanical Alloying

40

Grain Size

Chapter3 INSTRUMENTAL AND EXPERIMENTAL SET UP

3.2. Materials

The alloy studied is AI-5356, contains a nominal content of 5 wt % Mg. Two milling

procedures were used, the first one involving solely nanostructure formation (milling

pre-alloyed powders) and the second involving mechanical alloying during

nanostructure formation. Two sources of Mg were used, namely commercial purity Mg

and Ah2Mg17 intermetallic. Table 3-1 presents the characteristics of the starting

powders used. For the two cases where mechanical alloying was involved, the two

powders were carefully weighted to target 5 wt% Mg in the alloy.

Table 3-1: Material$ Information: Different sources and characteristic of initial powders

Powder Supplier Particle size Morphology Purity (Mesh) (%)

AI-5356 pre-VAL/MET -325 Spherical

AI: 94.40 alloyed Mg:5.05

ATLANTIC AI pure EQUIPMENT -325 Spherical 99.9

ENGINEERS Magnesium

ALFAAESAR -325 Spherical 99.8 pure

Ah2Mg17 ATLANTIC AI: 50.05 (Mg50%- EQUIPMENT -200 Angular Mg: 49.70 AI50%) ENGINEERS

3.3. Cryomill Setup and Design of Experiment

3.3.1. Cryomill Setup

The procedure is described as a mechanical attrition technique in which powders are

milled in slurry consisting of mi"ing ba"s and cryogenic liquid. The temperature was

maintained at cryogenic range by contro"ing the lever of liquid nitrogen, which was

monitored by a thermocouple.

41


A Szegvari 01 HD low-energy mechanical attritor (Union Process) with 440C

stainless steels balls (4,76 mm in diameter) was used. Stearic acid

(CH3(CH2)16C02H) was added as process control agent in a proportion of 0.3 wt%.

The container was placed in an enclosed box after milling until completion of liquid

nitrogen evaporation, th en returned to room temperature. The powder was

separated from the milling media by sieving.

Figure 3-1: Cryomill Set-Up

3.3.2. Design of Experiments

Cryomilling of the powder had been performed with the intention of producing

nanostructured pre-alloyed powder on one hand and to obtain mechanically alloyed

aluminum and magnesium nanostructured powder on the other hand. Both of the

routes have to be done with the optimal parameters. A 24-1 fractional plan has been

selected for the cryomill runs in order to determine which parameters most influence

the final powder. This plan totalizes eight different tests for four different parameters,

42

Chapfer3 INSTRUMENTAL AND EXPERIMENTAL SET UP

having each of them vary on two different modalities. After the eight tests, the

preliminary results inform parameter adjustment for further milling. The four milling

parameters that have been chosen are as follows:

• A: Type of powder: Pre-alloyed AI5356 and a mixture of aluminum and

magnesium with the same weight percentage as the alloy which is AI 95 wt%

and Mg 5wt%.

• B: Milling times: 4, 8 and possibly 12 and 16 hours

• C: Rotational speed: 180, 240 and a possibility to increase the speed to 300

RPM (Rotations Per Minute).

• D: Balls-to-powder weight ratio: 24:1 and 32:1

The choice of a fractional plan over complete one allows a reduction in the number of

tests from 16 to 8. The 4th factor becomes a resulting combination of the first three.

Table 3-2 summarizes the conditions of the Design of Experiences.

The expected results will have to conform with the following criteria:

• Reaching the nanometer grain size

• Morphology: Equiaxed particles

• Completion of mechanical alloying in the case of the mixture

43


Table 3-2: Oetailed conditions of OOE with different levels set

Modalities Identification Labels Ident.

Pre- AI95 wt%- Mg - a

alloyed 5wt%

Powder source

Elemental AI 5356 + A

Mixture

-

Milling Time Short 4 hours - b

Hours Long 8 hours + B

Slow 180 RPM - c

Rotational Speed

High 240 RPM + C

Balls-to-powder Low 24:1 - D

weight ratio High 32:1 + D

44


Table 3-3 shows the combination of the parameters in order to perform a fractional 24-

1 plan.

Table 3-3: Determination of the fourth parameter as a combination of the first three in the DOE

A B C 0 Identification

(A*B*C)

1 - - - - Abcd -

2 + - - + AbcO

3 - + - + aBcO

4 + + - - ABcd

5 - - + + abCO

6 + - + - AbCd

7 - + + - aBCd

8 + + + + ABCO

3.4. X-ray Diffraction

X-ray diffraction (XRO) was used to calculate the grain size, lattice strain and lattice

parameters, so it is possible that it could be used to confirm and quantify the degree of

45

Chapter 3 INSTRUMENTAL AND EXPERIMENTAL SET UP

alloying. The spectra were acquired using a PHILIPS PW1070 diffractometer

equipped with a monochromatic Cu Ka (A=1.54056 A) radiation and operated at

40keV and 20 mA. The angles scanned were from 35 to 120 degrees (29) at a speed

of 0.0100/s (100s/ 0), a dwell time of 0.5 s was used for each step and the step size

was 0.005°. An annealed 5356 powder was used as reference to correct for

instrumental broadening.

Figure 3-2: X-Ray Diffraction Apparatus

The grain size and lattice strain calculations were based on the Williamson-Hall

method [61]. The method, as mentioned earlier, is based on the full width half

maximum (FWHM) and the integral width (IntW) of the diffraction peaks for the milled

and the standard samples, noted f3m , f3a, om and oa respectively. The measured

profile is a convolution of the instrumental broadening (instrument contribution) and

the intrinsic broadening (small grain size and microstrain contribution). Deconvolution

of the measured profile is done on the assumption that the li ne shapes follow a

Gaussian (Equation 11) or Cauchy (Equation 12) approximation or a combination of

the two, Cauchy-Cauchy (CC), Gaussian-Gaussian (GG) and Cauchy-Gaussian (CG)

46


[50]. The best fit between the measured profile and the approximation will decide

which one to use for deconvolution.

Equation 11

Equation 12

ln the deconvolution process, the instrumental profile can be represented by the

measured profile of the annealed sam pie. The representation of the intrinsic and the

instrumental could therefore assume to follow the subsequent combinations [50]:

Equation 13: CC

Equation 14: GG

p2 52 j3 = j3 __ '_1 or 5 = 5 __ a_

m j3 m 5 m m

Equation 15: CG

The overall intrinsic broadening of the milled powders results from small grain size

and microstrain. The deconvolution of the integral intrinsic broadening profile will

consist of separating the integral grain size broadening 00 and the integral microstrain

broadening, os.

Based on the same assumptions as for the instrumental and the intrinsic profiles, the

following equations are applied for the deconvolution of the overall intrinsic profile

[62]:

47


Equation 16: CC

Equation 17: GG

Equation 18: CG

The following equations express the integral grain size and the microstrain broadening

in the milled materia~

JD

= kÂ Dcos8

Js = 4etan8

Equation 19

Equation 20

Where k ranges from 0.9 to 1; " is the wavelength and in the case of CuKa 1, "

=0.154056 nm; 8 is the diffraction angle; D is the grain size; e is an approximate upper

limit of the lattice distortion [50].

Grain size and microstrain can be obtained with the equations obtained when

substituting Equ 19-20 into Equ 16-18, as shown below:

J cos 8 = 4e sin 8 + kÂ D

"cos8 = [ (4esin8)2 + ( ~)2 r'

Equation 21

Equation 22

48


s: B 16 2 sin 2 B kÂ uCOS = e +-

8cosB D Equation 23

ln order to determine the grain size and the microstrain with the previous equations,

the value of 0 has to be calculated first, using Equations 16-18, depending on the

combination (CC, GG, CG) chosen. A plot is do ne with Equations 21-23 and the

slopes and the intercepts of the straight li ne obtained are summarized in the following

table.

Table 3-4: Stopes and intercepts for grain size and microstrain calculations [49]

Combination 1 Plot Siope Intercept

Cauchy-Cauchy 4e kNO

o cos e against sin e

Gaussian-Gaussian

16e2 (kNol ri cos2 e against sin2 e

Cauchy-Gaussian

16e2 kNO o cos e against sin2 elo cos e

3.5. Dynamic Magnetic Compaction

A magnetic press working under a tension of 15.8 kV and generating a total pressure

of 1.1 GPa was used to compact the powders sample. Powders were cold pressed in

49


copper containers to aUain a green density of 65%. A total of 12 samples were

prepared for compaction based on the prealloyed AI 5356. These samples consisted

of a mixture of cryomilled powders and un-milled ones. Starting with a sam pie of

totally cryomilled powders, the percentage of un-milled powders was increased by

10% per sample. An extra sam pie consisting of flaUened powders was also

compacted. Following compaction, samples were cut into two pieces, mounted, and

polished for optical microscopy analyses. The second piece was used to measure the

relative density of each sample. The density was obtained by dividing the measured

weight of the cut piece by the volume calculated with the measured values.

3.6. Microhardness

Microhardness measurements had been conducted on a Clark Microhardness Tester

(CM-100AT) connected to a computer using Clemex Image Analysis Software

(Clemex CMT) for the readings of the measures.

Figure 3-3: Microhardness Apparatus

50


3.7. Microscopie Characterization

Optical microscopies were done with a Nikon (EPI HOT 200) connected to an

automated Clemex stage (JS 2000) and to a computer using Clemex Vision

Professionnal Edition for the edition of the images.

Figure 3-4: Optical Microscope Set-Up

Scanning electron microscope (JEOL-840-A) was used to characterize the particles

morphology. Transmission electron microscopy observations were done using a

JEOL JEM-201 0 operating at 200 kV.

51


Figure 3-5: SEM Apparatus. JEOL 840-A

Figure 3-6: TEM Apparatus JEOL JEM 2010

The TEM samples were prepared by spreading the powders on a carbon film.

52


3.8. Differentiai Scanning Calorimetry Analyses

DSe was used to determine through enthalpy evolution, the charaderistics of the

phases present in the powders. The three main powder sources were the prealloyed,

the mixture of pure aluminum and pure magnesium and the mixture of pure aluminum

and the intermetallic. For each sample, 20 mg were used to run the scan. A heating

rate of 10K per minute for a temperature range of 20 to 5000 e was set for the three

samples run.

The apparatus consisted of a High Performance Modular DSe & DT A Thermal

Analyzers, from Setaram, Model Setsys Evolution.

Figure3-7: ose Machine

53

Chapfer4


4.1. Microstructural Characterization and Evolution of the Powders

4.1.1. Starting Initial Powders

The following scanning electron microscope (SEM) images show the morphology of

each powder prior t.9 milling. The intermetallic AI12Mg17 is the only non spherical

powder and possesses a different initial size, 200 mesh (74IJm) compare to 325 mesh

(44IJm) for ail the remaining powders.

(a) AI 5356 Pre-alloyed (b) Aluminum pure

(c) Magnesium Pure (d) Intermetallic AI12Mg17 (Mg50%- AI50%)

Figure 4-1: SEM images of starting powders

54

Chapfer4 RESUL TS AND DISCUSSION

4.1.2 Characterization of the Cryomilled Powders from the Design of

Experiment Scheme

The characterization of the milled powders, consisting of 8 different setups based on

the design of experiments (DOE), has been separated into two groups. The milling

procedures described earlier (section 3.2), was used to create the two groups (pre

alloyed powder and the elemental mixture). Microstructural characterization is done in

order to confirm the following conditions that the powders have to satisfy:

- Nanometer grain size

- Equiaxed particles morphology

- Completion of the mechanical alloying in the case of the elemental mixture.

4.1.2.1. Microstructural Characterization of Pre-alloyed Powders

A. SEM and X-Ray Diffraction Analyses

The SEM images are paired up with the XRD pattern for a better understanding of the

milling mechanism illustrated by the evolution of particles morphology and peak

broadening. In Figure 4-2 (a), the powder had low milling time (4hours) low rotation

speed (180 RPM) and high ball-to-powder weight ratio (BPR of 32: 1). This is traduced

by flattened, fractured and non-welded particles. Figure 4-2 (b), illustrates equiaxed

particles of powder milled for 8 hours at 180 RPM and BPR of 24: 1. The particles are

in stage 3 of the process of mechanical milling. Figure 4-3 (a) shows powder after 8H

of milling time, 240 RPM and 32:1 for BPR but the particles are not equiaxed. In

stage 2 of the mechanical milling the powders become flattened, fractured and start to

55

Chapter4 RESUL TS AND DISCUSSION

cold weld to each other. The image showed in Figure 4-3 (b) (4H, 240RPM and 24:1

BPR), indicates that the powders are flattened and start to cold weld. A difference on

the edges of the flattened particles of AbCd and the edges of the flattened and

fractured of AbcD has to be exposed.

From this point, one concludes that a higher BPR has the tendency to produce more

fracturing than cold welding. On the other hand, a higher RPM a"ows the powders to

be equiaxed.

(a) AbcD: 4H 180 RPM Ratio 32:1 (b) ABcd: 8H 180 RPM Ratio 24:1

Time (H)

15 12

Ratio

8

4

Time (H)

15 12

8 -/: 1 RPM 4 1 1

1 1

24:1,(1 32y Ratio

Figure 4-2: SEM images and schematic representations of milling parameters for AI-5356 powders

56

RPM

Chapter 4 RESUL TS AND DISCUSSION

(a) ASCD: 8H 240 RPM Ratio 32:1 (b) AbCd: 4H 240 RPM Ratio 24:1

Time (H)

15 12

Ratio

8 4

-----" RPM

Time (H)

15 12

8

Figure 4-3: SEM images and schematic representations of milling parameters for AI-5356 powders

RPM

The XRD patterns illustrated in Figure 4-4 show the broadening of the peaks after

milling powders for the same design of experiment modalities. In order to have an

accurate analysis of this broadening, grain size calculations have been done and are

presented in the next section. However, it can be noted that the lowest spectrum

(AbcD: short time, low rotational speed and high ratio) demonstrated the least

broadening of the peaks.

57


6000.---------------------------------------------------------------------------------,

5000 -

4000 -~ !: ::J

-ë ~ 3000 ~ III !: $ .5

2000

1000

-AbcD: 4H 180RPM Ratio 32:1 -AScd: 8H 180RPM Ratio 24:1

- ASCD: 8H 240RPM Ratio 32:1 - AbCd: 4H 240RPM Ratio 24:1

o ft!"" ..J r' ....... d' 'L.. J'=

35 40 45 50 55 60

29

65 70 75

Figure 4-4: Diffraction patterns showing broadening of AI 5356 peaks for different milling set-ups

58

80 85


S. Grain size and Lattice Strain Determination by X-Ray Diffraction of the

Pre-alloyed Powders

Determination of grain size is a direct way of confirming the nanostructured state of

the powder. The method used is based on X-Ray diffraction. This method has been

chosen over the TEM method at this stage in the experiments because it requires less

sam pie preparation and is therefore less time-consuming.

Added to grain size calculations is the lattice strain (microstrain) determination. This

characteristic is alsC? calculated with the X-Ray diffraction patterns using the Hall

Williamson relation. Table 4-1 shows the values of the calculated grain size and the

corresponding lattice strain of the prealloyed powders for different milling setups.

Table 4-1: Grain size calculated values for the pre-alloyed powders

Grain

size (nm) Lattice strain (%)

AbcD: AI-5356 4H 180RPM Ratio 32:1 80 ±29 0.13 ± 0.05

AScd: AI 5356 8H 180 RPM Ratio 24:1 30 ± 11 0.16 ± 0.06

AbCd: AI 5356 4H 240 RPM Ratio 32: 1 120 ± 4 0.31 ± 0.01

ASCO: AI 5356 8H 240 RPM Ratio 32: 1 40 ± 7 0.23 ± 0.04

The value obtained with the 4 hours sam pie (AbCd: 120 ± 4 nm) will not be

considered for the continuation of the work because the grain size obtain is over 100

59


nm. The grain sizes of the three other samples are ail under 100 nm. These values

satisfy the nanometer grain size criterion. However the morphology of the particles

favors the ASCO sam pie. In fact only this sam pie possesses powders with equiaxed

morphology. This second criteria is only satisfied by the ASCO sam pie.

Consequently, longer milling time and lower (24:1) ball-to-powder weight ratio is the

combination that would be retained at this point of analysis. Table 4-2 shows values

of grain size and lattice strain obtained for cryomilled pure aluminum and AI-7.5% Mg

alloy [13, 63]. The grain size values obtained in the present study are within the same

time range as those fram the studies presented.

Table 4-2: Values of grain size and lattice strain from different milling process for comparison

Milling Grain size LaUice

Sam pie (nm) strain (%) Reference time (h)

AI 8 26 ±2 0.16 ± 0.03 [63]

AI-7.5%Mg 8 26 ±2 0.23 ± 0.03 [13]

AI-5356 40 ±7 0.23 ± 0.04 8 This work

(5%Mg)

60


4.1.2.2. Microstructural Characterization of Mixture of Elemental

Powders


Figure 4-5 (a) shows powders in stage 2 of the mechanical milling process and its

corresponding schematic milling parameters representation. The particles were

flatlened, fractured and started to weld and form lamellas. This sam pie (abCD) was

subjected to 4H of milling time, 240 RPM and 32: 1 of ball-to-powder weight ratio

(BPR). Figure 4-5 (b) illustrates particles before the beginning of stage 2, they are

flattened but not fractured and have not started to weld. Thus, 180 RPM and 24:1

BPR requires more time to reach stage 2. The XRD patterns (see Figure 4-7) of these

samples corroborate the SEM pictures analyses. The pure magnesium peak is still

visible, confirming the early stage in mechanical milling process. In the course of the

milling, the magnesium particles will be embedded as the particles are cold welded

and acquire equiaxed morphology (stage 3) in the aluminum matrix. The

homogenization of the powderrs completes the AI-Mg alloy. Figure 4-6 a) shows

images of aBcD powders that have been milled for 8H at 180 RPM and with a BPR of

32: 1. Figure 4-6 b) illustrates of aBCd milled for 8H at 240 RRM and with a BPR of

24: 1. A BPR of 32: 1 introduces more energy and accelerates the alloying process [1]

but 240 RPM plays in favor of more equiaxed particles, as shown in the case of aBCd.

The comparison of the XRD patterns (see Figure 4-74-7) of these samples (aBcD and

aBCd) does not favor one setup over the other one because both of them still show

pure magnesium peaks.

61


(a) abCD: 4H 240 RPM Ratio 32:1 (b) abcd: 4H 180 RPM Ratio 24:1

Time (H) Time (H)

15 15 12 12

8 8 RPM

Ratio Ratio

Figure 4-5: SEM images and schematic representations of milling parameters of AI and Mg cryomilled powders.

62

RPM


(a) aBcD: 8H 180 RPM Ratio 32:1(b) aBCd: 8H 240 RPM Ratio 24:1

lime (H)

15 12

8 4

~

32:1,~'

Ratio ,1

-"/' 1 1

1 1 1 1

1

RPM

lime (H)

15 12

Ratio

8

4

-----7 1 1

1 1 / 1

1

Figure 4-6: SEM images and schematic representations of milling parameters of AI and Mg cryomilled powders.

RPM

At this point, the conclusion that in order to achieve the alloying of the elemental

powders and the equiaxed morphology, it is necessary to mill the elemental mixture

for a much longer time with high RPM and high BPR.

The grain size and the lattice strain of these samples were not calculated because

they did not satisfy mechanical alloying completion condition.

63

~ 'c :1

.ci ... ~ >-:t: t/I C

.Sl

.5

Chapter4

8000

7000

6000

5000 ~ 1 1

4000

3000

O·+-' --

30

• •

0

0

Jill 1\ •

40 50


- abCD: 4H 240RPM Ratio 32:1 - abcd: 4H 180RPM Ratio 24:1

- aBcD: 8H 180RPM Ratio 32:1 - aBCd: 8H 240RPM Ratio 24:1

• Mg o AI

o o

• • •

60 70 80

26

Figure 4-7: Diffraction patterns showing the evolution of Mg peaks and broadening of AI peaks for different milling set-ups

64


4.1.3. Characterization of the Cryomilled Powders With the Optimum

Milling Parameters After DOE

After the execution of the 8 tests according to the design of experiment, further tests

were performed with optimized parameters based on the initial conclusion from the

two different types of powder. Therefore, the following results will be presented based

on the following three types of starting powder:

• The pre-alloyed AI 5356

-

• The mixture of pure aluminum and pure magnesium

• The mixture of pure aluminum and the intermetallic AI12Mg17 (Mg50%- AI50%)

The analysis of the results of the DOE has informed new cryomilling parameter levels.

• For the pre-alloyed powder, the previous conclusion suggested long milling

time and lower ball-to-powder weight ratio. In addition, the SEM image

analysis of the two 8-hour samples of the elemental mixture (aBcD and aBCd)

showed that high RPM resulted in more equiaxed particles than lower RPM

(see Figure 4-6). Therefore, the rotation speed has been increased to a value

of 300 RPM, the BPR maintained at 24:1 and for more homogenization, the

milling time has also been increased to 15 hours.

• For the mixture of aluminum and magnesium, milling for a linger time at a high

RPM and BPR has been suggested. Thus, the RPM was increased to 300, the

BPR fixed at 32:1 and the milling time increased to 15 hours.

• The third type of starting powder, mixture of pure aluminum and the

intermetallic had a total milling time of 18 hours, a rotational speed of 300 RPM

65


and a ratio of 24: 1. The higher ratio of 32: 1 was not necessary because the

intermetallic powder already forms a phase with aluminum. Therefore the

accomplishment of the solid solution is halfway through. In addition the

intermetallic powder is a hard material; the milling of two materials of different

hardness implies that the harder material will act as a cutter of the softer one.

As a result, formation of finer powder is observed in this particular case.


Optimized paramet~rs produced equiaxed particles for the three types of starting

powders, as shown in Figure 4-8.

The SEM images of the three samples ail show equiaxed particles, suggesting that

the stage 3 in the milling process has been successfully reached. As for the

homogenization (stage 4) of the powders, SEM image analysis is not sufficient as a

tool to confirm the attainment of this stage. Therefore, further analyses are necessary.

The XRD patterns presented in Figure 4-9 are the first stage in the analyses. The

spectra imply that the mechanical alloying has been completed as no pure

magnesium peaks, nor intermetallic peaks are still visible.

66


(a) 535615 H 300RPM Ratio 24:1 (b) AI+Mg Pure 15H 300RPM Ratio 32:1

(cyAI + A1 12Mg17 (Mg50%- AI50%) 18H 300RPM Ratio 24:1

Figure 4-8: SEM illustrations of morphology of cryomilled powders from the three main starting sources.

67

(il

5000

4500

4000

3500

~ 3000 ::1

€ ~ 2500 ~ //1

ai 2000 -.E

1500

10001

500

o 35

Chapter4

40 45 50 55


- AI 5356 15H 300 RPM Ratio 24:1

-AI pure + Mg pure 15H 30Ç RPM Ratio 32:1

- AI pure + Intermetallic 18H 300 RPM Ratio 24:1

60

29

65 70 75 80

Figure 4-9: Diffraction patterns of different starting powders

68

l

85


B. Lattice Constant Determination by X-Ray Diffraction

ln order to confirm the mechanical alloying as stated by XRD patterns, lattice

parameters determination is a powerful tool. Calculations of the lattice parameters for

the three cryomilled powders will permit a clear comparison between the prealloyed

powders and the powders obtained fram a mixture of pure aluminum and pure

magnesium on one hand and pure aluminum and an intermetallic on the other hand.

The assertion is based upon a relation between the percentage attained of sol id

solution of magnesium in the aluminum matrix and the corresponding lattice

parameter of the aluminum alloy [51]. This assumes that the lattice expansion can be

attributed solely to the dissolved Mg atoms. For this purpose, the program X-LAT

[64](Least Squares Refinement of Cell Constants) has been used for the

determination of the different lattice parameters.

Table 4-3: LaUice parameters values of AI and AI alloys under different states

Lattice parameters

(Angstroms)

AI pure 4.0495 [51]

AI 5356-0 4.0678 [51]

AI 5356 15H 300RPM Ratio 24:1 4.0640 ± 0.0007

AI pure + Mg pure 15H 300RPM Ratio 32: 1 4.0508 ± 0.0007

AI pure + Intermetallic 18H 300RPM Ratio 24:1 4.0508 ± 0.0008

69


Despite the fact that the X-Ray spectrum show a complete disappearance of pure

magnesium and intermetallic peaks, the values of calculated lattice parameters

demonstrate that the alloying was incomplete. The two samples from a mixture of

pure aluminum and magnesium show the same lattice parameter value of 4.0508 A. Figure 3-10 depicts the evolution of lattice parameters with increasing atomic

percentage of solid solution of magnesium in aluminum. A linear tendency curve has

been added for the equivalent weight percentage. From this tendency curve, the

amount of Mg that has formed a solid solution with aluminum, for both sources of

magnesium, was calculated to be 0.6259 wt%. Note that the equilibrium solubility of -

Mg in AI is only 1.08 wt. % at room temperature [65].

70


• [51EII] • [50Dor] [51 Poo] • [62Hel] " [64Luo] • Weight% - Linear (Weight %) i

0.424 -,--..... _--... 1

1

0.422 i---"·' . --------------

.. ".L·_·······I 0.420 +-- -----.-" ... -.".-...... ".".----." .-.. 1

0.418 1'- .-------.-. ------ .. __ --'L.~. -1 I°.416 " ~ --- --------~------ --

m "

1

~ 0.414 1

----- -------------_. ----~ ID E "

1

~ If. 0.412 ---------------- .~"

ID • () 1 'e

1 j 0.410 -..-c' ._---"--,,

• 0.408j---· ... --Jl~ '--1 0.406 1

. .,- ,, ___ o.

1

-1

1 1

0.404 +- ..... ----.---." .. _-_.~

0.402 '

0 5 10 15 20 25 30 35 40 45

Atomic Percent Magnesium

Figure 3-10: Evolution of lattice parameter with increasing atomic percentage of magnesium[51]

71


C. DSC Measurements

One other way to determine the state of the powder milled is to measure the enthalpy

released through a Differentiai Scanning Calorimetry. This technique allows an

evaluation of a compound's change in enthalpy and links it with a modification in the

phases constituting the material. In this case, the liberation of magnesium from the

solid solution with aluminum is the phase transformation that has been aimed at [48].

For the prealloyed AI 5356, the phase transformation happened at 375°C and was

related to a change of enthalpy of 28.2506 J/g. The samples fabricated from

mechanical milling of pure aluminum and magnesium/ intermetallic did not have a

change in enthalpy, as observed in Figure 4-11.

72


25 .

- AI 5356 - AI pure + Mg pure - AI pure + Intermetallic

20

15

§' 10 .§.

~ 0 Li: -CIl 5 <II :I:

0 6, 100/ // 200 300 400

~ 500

-5

-10

Temperature (oC)

Figure 4-11: Dse curves obtained fram heating

73


D. Grain Size Calculation Based on XRD Compared to TEM

Measurements for the Selected Prealloyed Powders.

As per the results presented in the previous sections, the analyses were continued

only for the prealloyed powder (15H 300 RPM Ratio 24: 1) in this work. These

powders were chosen for the attainment of the nanometric scale, the formation of

equiaxed particles (better flowability) and the fact that they contain Mg in sol id

solution.

TEM images, shown_ in Figure 4-12, were taken fram cryomilled particles thin enough

to let the electron beam pass thraugh them. Most of the parti cl es used had size under

11..lm; the bright field image was (Figure 4-12 a) done with a particle -0.7I..lm in length.

These particles will have a smaller grain size than of the average of ail the cryomilled

powder grains. Similar results were found by G.Lucadamo et al.[66] where the

particles analyzed had -0.5I..lm in size and possessed grain sizes of -10 nm.

The dark field image (Figure 4-12b) was formed fram a selected plane of the

diffraction pattern rings. The rings in the diffraction pattern confirm the polycrystalline

nature of the cryomilled alloy. The limited number of diffracted planes resulted in a

series of dots in a circle pattern instead of full circles. These dots could suggest a

texture in the cryomilled powders.

74

Chapfer4

lilIIJ·'bI


a

Figure 4-12: TEM images of nanostructured AI 5356: a) Bright field; b) Dark field with the corresponding diffracted plans

75


For the TEM measures, a total number of 158 grains has been used to directly

measure on the image the crystallite size. A distribution of the measured grain size is

shown in Figure 4-13.

60 ,------------------------------------------------,

50

- 40 e c o ~ .ê 30 ... QI

J:I E :::1

Z 20

10

o.

0·-10 10--20 30--40 Grain size (nm)

D=19.6 nm 158 grains

30--40

Figure 4-13: Grain size distribution based on TEM imaging measures

40--50

The grain size obtained from the two methods is summerized in Table 4-4 ..

Table 4-4: Grain size comparision between XRD and TEM methods

Sample X-Ray Diffraction TEM measures

Grain size (nm) 48 ± 3 19.6 ± 6.0

76


The average value obtained from the TEM is sm aller than the one from the XRD

technique. This is explained by the fact that the particles used for TEM imaging had,

as mentioned earlier, to be small enough to let the electron beam cross them. This

implies that sm ail and thinner particles will also consist of grains with sm aller diameter

size. On the other hand, XRD is done on a much bigger sam pie size and the grain

size obtained is an average value from particles beUer representing the cryomilled

powder.

77


4.2. Dynamic Magnetic Consolidation (DMC)

Dynamic magnetic compaction (DMC) of the powders was done by mixing cryomilled

with an increasing quantity of non milled powder having larger grains. The selected

powders for DMC were the prealloyed AI 5356 (15 hours 300RPM, Ratio 24:1).

4.2.1. Optical Microscope Imaging

The compacted samples have been eut in the longitudinal direction and then mounted,

as iIIustrated in Figure 4-14.

~-1 1

1

1

Figure 4-14: Schematic illustration of samples preparation after DMC

The longitudinal faces of the compacted pieces are presented in Figure4-15 and

Figure 4-16. The increasing percentage of non milled powders, is illustrated by the

presence of bigger and clearer (white) particles. The cryomilled particles consist of

darker (grey) particles. The black spots in the indicate porosity, due to a lack of

complete compaction. The last sample was compacted with flaUened-fracture

powders (flaky powders). The curvature of the compacted particles is indicative of

deformation during compaction.

78


100% Milled powder 90% Milled Powder

80% Milled powder 70% Milled Powder

60% Milled Powder 50% Milled Powder

Figure 4-15: Compacted samples ranging from 100 to 50% of milled powder

79




0% Milled Powder Flaky Powder

Figure 4-16: Compacted samples ranging from 50 to 0% of milled powder, plus a sample from flaky powders

80


4.2.2. Relative Densifies Measuremenfs

Table 4-5 shows the values of the measured densities for the different combinations of

compacted samples. The measures had been done by weighting eut samples of

calculated volumes. Theoretical density fram which the relative densities are

compared is the density of AI 5356, which is 2.64g/cc.

Table 4-5: Densities and Hardness of Compacted Samples

D~creasing P prel percentage of milled

powders. (g/cc) (%)

100% 2.28 ± 0.03 86.39 ± 5.88

90% 2.31 ± 0.01 87.49 ± 2.42

80% 2.43 ± 0.01 91.89 ± 2.22

70% 2.41 ± 0.03 91.37 ± 6.33

60% 2.46 ± 0.01 92.93 ± 1.94

50% 2.46 ± 0.01 93.01 ± 1.09

40% 2.50 ± 0.01 94.73 ± 2.56

30% 2.48 ± 0.03 93.90 ± 5.49

20% 2.44 ± 0.01 92.48 ± 1.78

10% 2.59 ± 0.01 97.97 ± 2.03

81


The values of the relative density can be related to the pictures of the compacted

samples (see Figure4-15 and Figure 4-16). Figure 4-17 shows a tendency of a

decrease in relative density while the milled powder percentage increases. This

tendency is also visible in the pictures of the compacted samples where the porosity

area (black spots) decreases with decreasing percentage of milled powders

105,00·,-----------------------------,

100,00 .

l ~ ~

95,00·

1 ijj

1 l 1

c CIl C CIl .~ 'lii 90,00 (jj

! Il:::

85,00·

80,00 +--------,---,------,------,------,--------,---,------,----1

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 110%

Percentage of Milled Powders

Figure 4-17: Evolution of relative density with increasing percentage of milled powders

4.2.3 Microhardness and Indent Traces

An indication of how much the powders are consolidated is the indent load mark after

microhardness measurements. In fact consolidation is characterized by a bonding

layer between particles. Compaction, on the other hand, is characterized by a

82


mechanical interlocking of powders without. An apparent crack at the corner of the

load mark indicates an unconsolidated powder. The images shown in Figure 4-18

and Figure 4-19 indicate a lack of bonding between particles as it appears that ail of

them possess a crack, even if very small for some of the samples, at one of the

corners of the indent trace. Therefore, complete consolidation was not achieved

during the dynamic magnetic process.

83





Figure 4-18: Indent traces on samples ranging fram 100 to 50% of milled powder

84




0% Milled Powder Flaky Milled Powder

Figure 4-19: Indent traces on samples ranging fram 50 to 0% of milled powder, plus a sample fram flaky powders

85


4.2.4 Considerations for Full Density Attainment

The peak pressure reached during compaction was 1.1 GPa. Ali samples were

prepared prior to compaction with a green density of 65%. After dynamic compaction,

none of the samples reached maximum density (100%). Shockwave peak pressure

applied during compression generates energy that was dissipated into void collapse,

defect and microkinetic energies [56]. The void collapse energy is affected by partiele

geometry, partiele contact areas and mechanical properties of the particle and

adjacent partieles [56]. Partiele characteristics such as hardness, size, distribution

and aspect ratio will-have a direct effect on the void collapse energy. Defect energy

originated from energy deposited into the interiors of the partieles by generating high

dislocation densities [56]. Microkinetic energy consists of the plastic flow of the

materialleading to an interparticle impact [56]. Consolidation of powders occurs when

the surplus of shock energy dissipates through reaction bonding energy and melting

energy. This could be achieved by increasing the shockwave energy to a higher level,

allowing a formation of melted layer on the partieles that will enhance the bonding [56]

and simultaneously increasing the final density. Dynamic magnetic compaction

(DMC) with its ease of adjusting the pressure is a powerful tool to realize both

compaction and consolidation of powders.

86

Chapter5

CONCLUSIONS

The production of nanostructured AI-Mg alloy powders was processed via the

cryomilling procedure. Two options were available for the formation of AI-5%Mg

nanocrystalline alloy: Milling of a pre-alloyed AI-5356 powder and attempted

mechanical alloying of AI and Mg. Two potential sources of Mg were pure magnesium

and an intermetallic AI12Mg17 powder. Consolidation of the nanostructured powders

was performed with a dynamic magnetic compaction technique. The conclusions are

summarized below.

1. Design of experiment has been conducted and the optimum milling parameters

for the formation of nanostructure pre-alloyed powder were determined. It was

found that 300 RPM (rotational speed), a ball-to-powder weight ratio of 24:1

and milling time of 15 hours were the optimum milling parameters for the

attainment of a nanocrystalline structure and equiaxed powder morphology.

2. In the case of mixture of elemental AI and Mg. The milling conditions were

fixed at 300 RPM, a ratio of 32:1 and a milling time of 15 hours. The

confirmation of the alloying could not be assured despite the disappearance of

the pure Mg peaks in the X-ray diffraction (XRD) spectra. Differentiai scanning

calorimetry (DSe) and lattice constant calculations revealed the lack of alloying

formation during mechanical milling. No enthalpy variation was found during

heating and the lattice constant obtained has the same value as the pure AI,

4.050 A.

87

Chapter5 CONCLUSIONS

3. The second source of Mg, the intermetallic A1 12Mg17 powder, was milled with

pure aluminum at 300 RPM, having a ratio of 24:1 for a total time of 18 hours.

As for the pure magnesium powder, alloying was not confirmed, either by OSC,

or the lattice constant. The enthalpy variation with OSC measurement did not

change and the lattice constant found was also equal to 4.050 A.

4. Analyses were profoundly conducted for the pre-alloyed nanostructured AI-

5356 powder. The grain size obtained with the XRO method was 48.3 nm and

the measure obtained with TEM imaging analysis was 19.6 nm. The difference -

in the values resides in the particles selected for TEM. They were small and

thin, in order to let the electron beam cross them.

5. OSC measurement of the pre-alloyed AI-5356 powders reveals an enthalpy

variation of 28.2506 J/g at 375°C.

6. Oynamic magnetic compaction was done on the pre-alloyed nanostructured AI-

5356 powders. Samples were set by mixing the powder with an increased

percentage of un-milled AI-5356 powders, for better ductility. The resulting

formed pieces attained relative density ranging from 86.39 to 97.97%. The

lowest value came from the 100% milled powder sam pie and the highest from

the mixture of 90% unmilled and 10% milled powders. Oespite that relatively

high density, particle to particle consolidation was not successfully achieved.

7. General conclusions:

a. Cryogenie milling has successfully produced a nanometric grain size in

AI-5356 powder alloys having equiaxed particle morphology. In

addition, milling parameters were optimized for further purposes.

88

Chapter5 CONCLUSIONS

b. Dynamic magnetic compaction was found to be a precise technique for

powder compaction. However, in this work, pressure obtained (1.1

GPa) was not sufficient to ensure consolidation of the powders.

c. In order to have an evaluation of the mechanical properties of the

dynamically compacted samples, longer specimen will have to be

prepared in future work. These samples will have to be long enough so

that tensile testing could be carried out following standard size

specimens.

89

REFERENCES

REFERENCES

[1] C. Suryanarayana, "Mechanical alloying and milling," Progress in Materials

Science, vol. 46, pp. 1-184,2001.

[2] S. H. Kim, D. H. Kim, and N. J. Kim, Materials Science and Engineering A, vol.

226, 1997.

[3] F. Khonsari, J. Kurdi, M. Tatoulian, and J. Amouroux, Surf. Coat. Tech, vol.

142,2001.

[4] S. Keller, P.- Waltereit, P. Cantu, U. K. Mishra, J. S. Speck, and S. P.

DenBaars, Opt. Mater., vol. 23, 2003.

[5] H. J. Fecht, Nanostructured Materials and Composites Prepared by Solid State

Processing,

ed. Norwich, NY: Noyes Publication / William Andrew Publishing, 2002.

[6] S. M. Umbrajkar, M. Schoenitz, S. R. Jones, and E. L. Dreizin, "Effect of

temperature on synthesis and properties of aluminum-magnesium mechanical

alloys," Journal of AI/oys and Compounds, vol. 402, pp. 70-77, 2005.

[7] J. E. Hatch, Aluminum: Properties and Physical Metal/urgy. Metals Park, OH:

ASM,1984.

[8] G. Ribarik, J. Gubicza, and T. Ungar, Materials Science and Engineering A,

vol. 387, 2004.

[9] M. 1. D. Source, "www.matweb.com." 2007.

[10] V. L. Tellkamp, A. Melmed, and E. J. Lavernia, "Mechanical behavior and

microstructure of a thermally stable bulk nanostructured AI alloy," Metal/urgical

and Materials Transactions a-Physical Metallurgy and Materials Science, vol.

32, pp. 2335-2343, 2001.

[11] O. N. Senkov and F. H. Froes, International Journal of Hydrogen Energy, vol.

24, pp. 565-576, 1999.

90

REFERENCES

[12] A. P. Newbery, S. R. Nutt, and E. J. Lavernia, "Mu/ti-Sca/e A/ 5083 for Mi/itary

Vehicles with /mproved Performance," JOM, vol. 58, pp. 56-61,2006.

[13] F. Zhou, S. R. Nutt, C. C. Bampton, and E. J. Lavernia, "Nanostructure in an

A/-Mg-Sc alloy processed by /ow-energy bail milling at cryogenie temperature,"

Metallurgical and Materials Transactions a-Physical Metallurgy and Materials

Science, vol. 34A, pp. 1985-1992,2003.

[14] T. G. Nieh, P. Luo, W. Nellis, D. Lesuer, and D. Benson, "Dynamic compaction

of a/uminum nanocrysta/s," Acta Materialia, vol. 44, pp. 3781-3788, 1996.

[15] G. H. Lee, C. K. Rhee, M. K. Lee, W. W. Kim, and V. V. /vanov, -

"Nanostructures and mechanica/ properties of copper compacts prepared by

magnetic pu/sed compaction method," Materials Science and Engineering a

Structural Materials Properties Microstructure and Processing, vol. 375-77, pp.

604-608,2004.

[16] S. C. Tjong and H. Chen, "Nanocrystalline materia/s and coatings," Materials

Science & Engineering R-Reports, vol. 45, pp. 1-88, 2004.

[17] H. H. Fu, D. J. Benson, and M. A. Meyers, "Ana/ytica/ and computationa/

description of effect of grain size on yie/d stress of meta/s," Acta Materialia, vol.

49, pp. 2567-2582, 2001.

[18] M. A. Meyers, A. Mishra, and D. J. Benson, "Mechanica/ properties of

nanocrystalline materia/s," Progress in Materials Science, vol. 51, pp. 427-556,

2006.

[19] 1. V. A/exandrov, K. Zhang, A. R. Ki/mametov, K. Lu, and R. Z. Va/iev, Materials

Science and Engineering A, vol. 234, pp. 331, 1997.

[20] D. B. Witkin and E. J. Lavernia, "Synthesis and mechanica/ behavior of

nanostructured materia/s via cryomilling," Progress in Materials Science, vol.

51, pp. 1-60,2006.

[21] H. J. Fecht, "Nanostructure formation by mechanica/ attrition," Nanostructured

Materials, vol. 6, pp. 33-42, 1995.

91

REFERENCES

[22] J. A. Rodriguez, J. M. Gallardo, and E. J. Herrera, "Structure and properties of

attrition-milled aluminium powder," Journal of Materia/s Science, vol. 32, pp.

3535-3539, 1997.

[23] M. S. EI-Eskandarany, K. Aoki, and K. Suzuki, "Calorimetrie study of solid-state

amorphization reaction of rod-milled AlxTa1-x alloy powder," Journal of AI/oys

and Compounds, vol. 186, pp. 15-31, 1992.

[24] M. S. EI-Eskandarany, K. Aoki, and K. Suzuki, "Calorimetrie characterization of

the amorphization process for rod milled AI50Nb50 alloy powders," Scripta

Metal/urgica, vol. 25, pp. 1695-1700, 1991. ~

[25] C. Suryanarayana, E. Ivanov, and V. V. Boldyrev, "The science and technology

of mechanical alloying," Materia/s Science and Engineering a-Structural

Materia/s Properties Microstructure and Processing, vol. 304, pp. 151-158,

2001.

[26] O. Kobayashi, T. Aizawa, and J. Kihara, "High speed, bulk mechanical alloying

of Cu/Ag/Co systems," Materia/s Transactions Jim, vol. 37, pp. 1497-1504,

1996.

[27] S. M. Umbrajkar, M. Schoenitz, S. R. Jones, and E. L. Dreizin, "Effect of

temperature on synthesis and properties of aluminum-magnesium mechanical

alloys," Journal of AI/oys and Compounds, vol. 402, pp. 70-77, 2005.

[28] C. Aguilar, S. Ordonez, J. Marin, F. Castro, and V. Martinez, "Study and

methods of analysis of mechanically alloyed Cu-Mo powders," Materia/s

Science and Engineering: A, vol. 464, pp. 288-294, 2007.

[29] A. Calka, J. Nikolov, and B. Ninham, Mechanical al/oying for structural

applications. Materials Park, OH: ASM International, 1993.

[30] M. H. Enayati, Z. Sadeghian, M. Salehi, and A. Saidi, "The effect of milling

parameters on the synthesis of Ni3AI intermetallic compound by mechanical

alloying," Materia/s Science and Engineering A, vol. 375-377, pp. 809-811,

2004.

92

REFERENCES

[31] C. Suryanarayana, G. H. Chen, and F. H. S. Froes, "MILLING MAPS FOR

PHASE IDENTIFICATION DURING MECHANICAL ALLOYING," Scripta

Metallurgica Et Materialia, vol. 26, pp. 1727-1732, 1992.

[32] C. Cetinkaya, T. Findik, and S. Ozbilen, "An investigation into the effect of

experimental parameters on powder grain size of the mechanically milled 17-4

PH stainless steel powders," Materials & Design, vol. 28, pp. 773-782,2007.

[33] J. De, A. M. Umarji, and K. Chattopadhyay, "Origin of contamination and role of

mechanochemistry during mechanical alloying: the case of Ag-Te alloys,"

Materials Science and Engineering: A, vol. 449-451, pp. 1062-1066,2007. -

[34] J. S. Benjamin and T. E. Volin, Metallurgical Transactions vol. 5, pp. 1929,

1974.

[35] J. S. Benjamin, "Mechanical Alloying," Scientific American, vol. 234, pp. 40-49,

1976.

[36] R. M. Davis and C. C. Koch, "Mechanical alloying of brittle components: Silicon

and germanium," Scripta Metallurgica, vol. 21, pp. 305-310,1987.

[37] C. Harris, Trans Soc Min Engrs, vol. 238, 1967.

[38] M. J. Luton, C. S. Jayanth, M. M. Disko, S. Matras, and J. Vallone, presented

at Materials Research Society Proceedings, 1989.

[39] J. Benjamin, Metallurgical Transactions vol. 1, pp. 2943, 1970.

[40] P. S. Gilman and J. S. Benjamin, Ann Rev Mater Sci, vol. 13, pp. 279, 1983.

[41] W. C. Oliver, R. F. Singer, and W. D. Nix, "Formation and Identification of

Particles in Dispersion Strengthened Aluminum," Journal of Metals, vol. 31, pp.

109-109,1979.

[42] R. F. Singer, W. C. Oliver, and W. D. Nix, "Identification of Dispersoid Phases

Created in Aluminum during Mechanical Alloying," Metallurgical Transactions

a-Physical Metallurgy and Materials Science, vol. 11, pp. 1895-1901, 1980.

[43] J. William D. Callister, Materials Science And Engineering

An Introduction, Sixth Edition ed: John Wiley & Sons, Inc., 2003.

93

REFERENCES

[44] P. A. Beek, J. C. Kremer, L. J. Demer, and M. L. Holzworth, "Grain Growth in

High-Purity Aluminum and in an Aluminummagnesium Alloy," Transactions of

the American Institute of Mining and Metallurgical Engineers, vol. 175, pp. 372-

400,1948.

[45] J. E. Burke, "Some Factors Affecting the Rate of Grain Growth in Metals,"

Transactions of the American Institute of Mining and Metallurgical Engineers,

vol. 180, pp. 73-91, 1949.

[46] J. Lee, F. Zhou, K. H. Chung, N. J. Kim, and E. J. Laverina, "Grain growth of

nanocrystalline Ni powders prepared by cryomilling," Metallurgical and -

Materials Transactions a-Physical Metallurgy and Materials Science, vol. 32,

pp. 3109-3115, 2001.

[47] J. Humphreys and M. Hatherly, Recrystallization and related annealing

phenomena. Amsterdam: Pergamon, 1995.

[48] F. Zhou, X. Z. Liao, Y. T. Zhu, S. Dallek, and E. J. Lavernia, "M ierostructu rai

evolution during recovery and recrystallization of a nanocrystalline AI-Mg alloy

prepared by cryogenie bail milling," Acta Materialia, vol. 51, pp. 2777-2791,

2003.

[49] J. He, J. Ye, E. J. Lavernia, D. Matejczyk, C. Bampton, and J. M. Schoenung,

"Quantitative analysis of grain size in bimodal powders by X-ray diffraction and

transmission electron microscopy," Journal of Materials Science, vol. 39, pp.

6957-6964, 2004.

[50] H. P. Klug and L. E. Alexander, X-Ray Diffraction Procedures. New York: John

Wiley & Sons, 1974.

[51] J. Murray, Bulletin of Alloy Phase Diagrams, vol. 3, 1982.

[52] J. Barber and B. Chelluri, "Magnetic compaction process nears market," Metal

Powder Report, vol. 55, pp. 4,2000.

[53] A. G. Mamalis, D. E. Manolakos, A. G. Kladas, and A. K. Koumoutsos,

"Physical principles of electromagnetic forming process: a constitutive finite

94

REFERENCES

element model," Journal of Materials Processing Technology, vol. 161, pp.

294-299, 2005.

[54] L. E. Murr, Shock Waves For Industrial Applications. Park Ridge New Jersey

U.S.A: Noyes Publications, 1988.

[55] M. Legros, B. R. Elliot, M. N. Rittner, J. R. Weertman, and K. J. Hemker,

Philosophical Magazine a-Physics of Condensed Matter Structure Defects and

Mechanical Properties, vol. 80,2000.

[56] M. A. Meyers, D. J. Benson, and E. A. Olevsky, "Shock Consolidation:

Microstructurally-Based Analysis And Computational Modeling," Acta

Materialia, vol. 47, pp. 2089-2108, 1999.

[57] C. C. Koch, D. G. Morris, K. Lu, and A. Inoue, "Ductility of nanostructured

materials," Mrs Bulletin, vol. 24, pp. 54-58, 1999.

[58] R. W. Siegel and G. E. Fougere, Nanostructured Materials, vol. 6, 1995.

[59] D. Witkin, Z. Lee, R. Rodriguez, S. Nutt, and E. Lavernia, "AI-Mg alloy

engineered with bimodal grain size for high strength and increased ductility,"

Scripta Materialia, vol. 49, pp. 297-302, 2003.

[60] M. Brochu, T. Zimmerly, L. Ajdelsztajn, E. J. Lavernia, and G. Kim, "Dynamic

Consolidation of Nanostructured A-7.5%Mg Alloy Powders," Materials Science

and Engineering A, 2007.

[61] G. K. Williamson and W. H. Hall, Acta Metallurgica, vol. 1, pp. 22-31, 1953.

[62] H. H. Tian and M. Atzmon, "Comparison of X-ray analysis methods used to

determine the grain size and strain in nanocrystalline materials," Philosophical

Magazine a-Physics of Condensed Matter Structure Defects and Mechanical

Properties, vol. 79, pp. 1769-1786, 1999.

[63] F. Zhou, D. Witkin, S. R. Nutt, and E. J. Lavernia, "Formation of nanostructure

in AI produced by a low-energy bail milling at cryogenie temperature," Materials

Science and Engineering a-Structural Materials Properties Microstructure and

Pro cessing , vol. 375-77, pp. 917-921,2004.

95

REFERENCES

[64] B. Rupp, "Least squares refinement of cell constants," Scripta Metal/urgica, vol.

22, 1988.

[65] T. Massalki, Binary al/oy phase diagram. Metals Park, OH: ASM International,

1994.

[66] G. Lucadamo, N. Y. C. Yang, C. S. Marchi, and E. J. Lavernia, "Microstructure

characterization in cryomilled AI 5083," Materia/s Science and Engineering a

Structural Materia/s Properties Microstructure and Processing, vol. 430, pp.

230-241, 2006.

[67] R. Hill, The Mathematical Theory of Plasticity. Oxford: Oxford University Press,

1950.

[68] H. F. Fischmeister and E. Arzt, Powder Metal/., vol. 26, pp. 82, 1983.

[69] J. Davis, ASM Specialty Handbook-Aluminum and Aluminum AI/oys. Materials

Park, OH: ASM INTERNATIONAL, 1994.

[70] B. Q. Han, J. Y. Huang, Y. T. Zhu, and E. J. Lavernia, "Effect of strain rate on

the ductility of a nanostructured aluminum alloy," Scripta Materialia, vol. 54, pp.

1175-1180,2006.

96

production of nanocrystalline aluminum alloy powders ... · production by cryogenie milling and...

Documents