Nano femtosecond laser processing in developing nanocrystalline functional materials for multi core orthogonal fluxgate sensor

Laser processing parameters such as laser fluence, laser irradiation time and number of pulses were studied.. Hysteresis loops, obtained from samples ablated using different laser fluenc

NANO/FEMTOSECOND LASER PROCESSING IN DEVELOPING NANOCRYSTALLINE FUNCTIONAL MATERIALS FOR MULTI-CORE ORTHOGONAL

FLUXGATE SENSOR

TAN LI SIRH

(B.Eng (Hons.),NUS)

A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ENGINEERING NANOSCIENCE & NANOTECHNOLOGY INITIATIVE

NATIONAL UNIVERSITY OF SINGAPORE

2009

I would like to express my utmost sincere and greatest appreciation to my supervisors, Professor Li Xiaoping and Associate Professor Hong Minghui, for their guidance throughout my Master project Their valuable opinions have been utmost importance to me

I am also grateful to Dr Seet Hang Li for all his advices and help throughout my project In addition, I would like to thank all members in Neurosensors Lab and Laser Microprocessing Lab for sharing their research experience and helping me in one way

or another I would also like to express my heartfelt gratitude to Miss Jasmin Lee from NUSNNI programme, Mr Lim Boon Chow from A-STAR Data Storage Institute (DSI), Miss Li Xue from A-STAR Institute of Materials and Research Engineering (IMRE) and staff at the Advanced Manufacturing Lab (AML) and Materials Science Lab (MSL) for all their kind assistance

Lastly, I wish to thank all those who have supported, helped and accompanied me throughout my two years of the Masters course

TABLE OF CONTENTS

ACKNOWLEDGEMENTS i

TABLE OF CONTENTS ii

SUMMARY v

LIST OF TABLES viii

LIST OF FIGURES ix

LIST OF SYMBOLS xv

LIST OF PUBLICATIONS xvi

Chapter 1 Introduction 1

1.1 Motivation 1

1.2 Objectives of Present Study 5

1.3 Organization of Thesis 6

Chapter 2 Literature Review 7

2.1 Magnetic Materials and Relevant Theories 7

2.1.1 Basic Classification of Magnetic Materials 7

2.1.2 Ferromagnetic Materials and Their Properties 9

2.1.3 Curie Temperature Tc 9

2.1.4 Hysteresis 9

2.1.5 Factors Affecting Magnetic Quality 10

2.1.6 Domain Wall Theories 11

2.1.7 Magnetization Rotation 11

2.1.7.1 Crystalline Anisotropy 12

2.1.7.2 Stress Anisotropy 12

2.1.7.3 Shape Anisotropy 12

2.2 Types of Lasers 13

2.2.1 Nanosecond Laser 13

2.2.2 Femtosecond Laser 14

2.2.3 Effects of Nanosecond and Femtosecond Lasers on Magnetic Materials 18

2.3 Fabrication Methods 21

2.3.1 Template Fabrication 21

2.3.1.1 Conventional Methods 21

2.3.1.2 Laser Micromachining Method 23

2.3.2 Magnetic Material Deposition Methods 24

2.3.2.1 Electrodeposition 24

2.3.2.2 Electron-beam evaporation 25

Chapter 3 Research Approach and Experimental Setup 28

3.1 Research Approach 28

3.2 Materials Development and Fabrication Processes 28

3.2.1 Laser Systems 28

3.2.1.1 355nm Nd:YAG nanosecond laser 28

3.2.1.2 800nm Ti:Sapphire femtosecond laser 29

3.2.2 Deposition Set-up 30

3.2.2.1 Electron-beam Evaporation 30

3.2.2.2 Electrodeposition 31

3.3 Materials Properties Characterization Setup 34

3.3.1 Profilometer 34

3.3.2 Optical Microscope 34

3.3.3 Scanning Electron Microscopy (SEM) and Field Emission Scanning Electron Microscopy (FESEM) 35

3.3.4 Energy Dispersive X-ray (EDX) 36

3.3.5 X-Ray Diffraction (XRD) 37

3.3.6 Vibrating Specimen Magnetometer (VSM) Setup 39

Chapter 4 Nanosecond Pulsed Laser-assisted Micro-drilling of Templates and Electroplating Deposition of Nanocrystalline Permalloy 40

4.1 Research Approach 40

4.2 Template Fabrication 41

4.2.1 Selection of Machined-drilled or Laser-drilled Templates 41

4.2.2 Methods of Adhesion 43

4.2.3 Fabrication of Single and Stacked Templates 44

4.3 Laser Parameters 45

4.3.1 Laser Fluence Effect 46

4.3.2 Laser Irradiation Time Effect 48

4.3.3 Pulse Number Effect 50

4.3.4 Laser Focal Length Effect 51

4.3.5 Optimization of Laser Parameters 54

4.4 Study of Electrodeposition Parameters 55

4.4.1 XRD Analysis 55

4.4.2 Plating Current Efficiency 57

4.4.3 Plating Current Density 58

4.4.3.1 Composition (Ni%) 58

4.4.3.2 Diameter of Electroplated Wires 60

4.4.3.3 Number of Electroplated Wires 62

4.4.4 Plating Time 64

4.4.4.1 Composition 64

4.4.4.2 Material Structure 67

4.4.5 Template Electrodeposition 68

4.5 Electrodeposited Structures 70

4.5.1 Single Template 70

4.5.2 Stacked Templates 74

4.5.3 Single Template with Varying Array Size 75

Chapter 5 Femtosecond Laser Machining of Magnetic Materials 79

5.1 Research Approach 79

5.2 Commercial Vitrovac 6025X 80

5.2.1 Material Removal Rate 81

5.2.2 Morphology of Laser-Machined Channels 82

5.2.3 Study of Laser Fluence Effect on Composition at Ablated and Damaged Areas 83

5.2.4 XRD Analysis 85

5.2.5 Study of Laser Fluence on Magnetic Properties 86

5.2.5.1 Hysteresis 86

5.2.5.2 Coercivity 87 5.2.6 Influence of Different Numbers of Ablated Channels on Magnetic Properties 89

5.2.6.1 Hysteresis 89

5.2.6.2 Coercivity 90

5.3 Electron-beam Deposited Permalloy 91

5.3.1 Surface Morphology of Laser-Machined Channels 91

5.3.1.1 Laser Fluence Effect 91

5.3.1.2 Number of Pulses 95

5.3.2 Study of Laser Fluence Effect on Composition at Ablated and Damaged Areas 97

5.3.3 XRD Analysis 99

5.3.4 Laser Fluence Effect 102

5.3.4.1 Hysteresis 102

5.3.4.2Coercivity 103

Chapter 6 Conclusions and Recommendations 108

6.1 Conclusions 108

6.2 Future Works 113

6.2.1 Methods to improve laser-drilling of templates and electroplating of nanocrystalline Permalloy 113

6.2.2 Method to improve femtosecond laser machining on magnetic materials 115

Reference 117

SUMMARY

In the development of magnetic sensing elements, the sensitivity of magnetic sensors have been found to increase exponentially when multi-core structures are used as sensing elements These multi-core structures can be achieved by template assisted deposition/electrodeposition or direct processing In this project, multi-core structured functional magnetic materials of extremely high permeability have been successfully developed with the assistance of either nanosecond or femtosecond lasers Methods such as nanosecond laser drilling of polymeric templates for electrodeposition of nanocrystalline nickel-iron wires and femtosecond laser machining of Vitrovac 6025X and nickel-iron thin films are investigated Both methods are found to be suitable to obtain multi-core structures

Nanosecond laser-drilled templates selected as smaller hole diameter can be achieved using this approach This would yield high aspect ratio holes for electroplating In addition, there is higher wire density per template, with the template dimensions held constant Laser processing parameters such as laser fluence, laser irradiation time and number of pulses were studied The diameter of the laser-drilled hole and the ablation depth are found to be dependent on the laser fluence An increase in the applied laser fluence will increase both parameters Higher laser irradiation time amplifies the heat affected zones and a threshold number of pulses are required to obtain a clean well-drilled hole Laser-drilled holes are also known to have tapered profile In order to minimize tapering, the focal point is set at the top surface of the template With the achieved templates, electrodeposition was then carried out to fill the drilled holes with Ni-Fe Using XRD, the electroplated Ni-Fe is found to be of FCC structure, with the

lattice constant calculated to be 0.354 nm and having an average crystallite/grain size

of 34.5 nm In order to gain an understanding and control of the electrodeposition process, the effect of varying plating current density and plating time on the resulting composition was investigated Experiments were also conducted to investigate the value of the plating current density in order to achieve Permalloy composition for templates with different hole diameters and array sizes (wire number) The results show that the materials composition is independent of both the hole diameters and the array sizes Since 1 mm thick polymeric templates are used, vacuum (for partial removal of air bubbles) and ultrasonic agitation during electroplating are introduced

to minimize “blockage” of the laser drilled holes This has been found to be effective

in achieving high aspect ratio (1:25) wires If the laser-drilled templates are stacked, the best aspect ratio achieved for deposited nanocrystalline Ni-Fe wires has been found to be 1:50 (pillars of ~ 40 m diameter and 2mm in length), with a composition

of 83% nickel and 17% iron When multiple holes of different array sizes on the same template are placed in the electrolyte, composition of nickel-iron varies due to the ions distribution in the cell varied along the electrolyte

Femtosecond laser (direct) processing of magnetic materials were successfully carried out on Vitrovac 6025X foils and electron-beam evaporated Ni-Fe thin films The results showed that femtosecond laser processing is a feasible way to fabricate multi-core sensing elements For multi-core Vitrovac 6025X, no composition distribution of the ablated magnetic foils were observed across the entire surface profile, suggesting

no diffusion process took place during ablation Hysteresis loops, obtained from samples ablated using different laser fluences as well as specimens with different numbers of ablated channels, show that the overall anisotropy of the magnetic foils

before and after femtosecond laser processing did not change significantly and that the magnetic property of the specimens (i.e coercivity) did not deteriorate significantly, even when a significant amount of the magnetic materials was ablated For multi-core Ni-Fe thin films, it is observed that there is a 5% increment in terms of coercivity for fully ablated channels However, the results show that femtosecond laser irradiation can be manipulated to modify the properties of Ni-Fe Surface modification was observed Ripples formation, tilted to an angle to the direction of the beam, was observed when different laser fluences were applied to the nanocrystalline films When the polarization state of the laser beam was changed to 90°, shorter ripples (parallel to beam’s scanning direction) were formed As the number of ablative pulses increased, formation and collapse of ripples occurred earlier than that

of the threshold laser fluence at single pulse XRD analysis suggests that quenching effect changes the surface layer from nanocrystalline to amorphous and that the average crystallite/grain size decreases upon ripples formation After laser ablation, regardless of the value of the laser fluences, anisotropy has shifted towards the easy axes of the films The appearance of ripples marks the point when coercivity

rapid-of the magnetic material begins to decrease It appears that there is a laser fluence threshold for both average crystallite/grain size and coercivity to decrease to a minimum, beyond which coercivity will increase at a gradual rate

LIST OF TABLES

Table 2.1: Summary of classification of magnetic properties 7 

Table 2.2: Types of anisotropy 12 

Table 3.1: List of fabrication equipment used in this project 28 

Table 3.2: List of characterization equipment used in this project 28 

Table 3.3: Electrolyte solution content 32 

Table 4.1: Calculations from XRD results 56 

Table 4.2: Summary of grain size results 57 

Table 4.3: Summary of current efficiency results 58 

Table 4.4: Summary of hole diameter, plating area, plating current in relation to a constant current density 62 

Table 5.1: Composition of Ni-Fe (%) after laser ablation at non-ablated and ablated areas 99 

Table 5.2: Calculations from XRD results 100 

LIST OF FIGURES

Figure 2.1: A typical hysteresis curve 10 Figure 2.2: Process of magnetization 11 Figure 2.3: Nanosecond laser processing 13 Figure 2.4: SEM image of an ablated metal channel created by a nanosecond laser 14 Figure 2.5: SEM image showing minimum amount of re-deposited material after femtosecond laser ablation 15

Figure 2.6: Nanosecond laser processing 16

Figure 2.7: Easy axis M-H loops (a) before and (b) after fabrication of grooves

parallel to the as-deposited hard axis in FeN monolayer 19

Figure 2.8: Hysteresis loops of Finemet: (a) sample non-treated by laser, (b) laser treated sample with small density dotted lines, and (c) laser treated sample with high density of dotted lines 20

Figure 2.9: XRD analysis of amorphous alloy ribbon before and after laser micromachining 20

Figure 2.10: Photograph showing patterned template on developed negative photoresist layer 21

Figure 2.11: FESEM image of developed bi-layered photoresist after single exposure 22

Figure 3.1: Schematics of the laser drilling setup 29 Figure 3.2: Schematic representation of the laser set-up for cutting magnetic foils 30 Figure 3.3: Schematics of the electrodeposition setup 31 Figure 3.4: Diamond tip-wafer surface interaction 34 Figure 3.5: Set of atomic planes in a crystal, at an angle θ from the incident beam 38

Figure 4.1: Workflow for nanosecond laser-drilled templates for electrodeposition of nanocrystalline Ni-Fe 41

Figure 4.2: SEM photos showing (a) circular hole profile by machining; and (b) circular hole profile by laser drilling 42

non-Figure 4.3: SEM images of (a) tapered indentations on the Cu substrate by mechanical machining; (b) flat indentations on the Cu substrate by low fluence laser drilling; (c) round/concave indentations on the Cu substrate by high fluence laser drilling; Schematics showing (d) deposited hollow micro wires from mechanical machined templates; (e) solid micro wires resulting from the formation of flat indentations from low fluence laser drilling; (f) hollow micro wires resulting from the formation of round/concave indentations 43

Figure 4.4: PVC template piece was glued onto copper substrate before laser drilling

As a result, the copper surface is also ablated 44

Figure 4.5: Array is drilled into PVC before adhesion onto copper substrate The copper substrate is unaffected by the laser 44

Figure 4.6: The characteristic of this method is to drill an array of holes onto thin pieces of template material, and subsequently stacking and aligning them to form a resultant array of high aspect ratio 45

Figure 4.7: Relationship of laser fluence versus entry hole diameter at a pulse repetition rate of 1 kHz, laser irradiation time of 500 ms and the number of loop of 1 47

Figure 4.8: Dependence of diameter of heat affected zone on laser irradiation time at

a laser fluence of 19.1J/cm2, pulse repetition rate of 1 kHz, and number of loop of 1 The inset image shows a top view of a laser drilled hole on PVC surface 49

Figure 4.9: Schematics showing the effects of laser irradiation time with the HAZ size 50

Figure 4.10: Hole diameter versus number of pulses at a laser fluence of 19 J/cm2, pulse repetition rate of 1kHz, and laser irradiation of 500 ms 51

Figure 4.11: Surface profile of a laser drilled hole with profile 52

Figure 4.12: Tapering ratio versus distance from surface of template ( m) 53 Figure 4.13: Tapering angle (˚) versus distance from surface of template ( m) 54 Figure 4.14: (a) Front view of optimized laser micro-drilled 10×10 array; (b) back view of the array, showing that the holes were fully penetrated, and the small differences between front and back hole diameters (difference < 5 m) 55

Figure 4.15: Graph of counts per second against 2 angle 56 Figure 4.16: Graph of number of micro wires (with that particular iron content) against iron content (%) for the current density of 35 A/dm2 59

Figure 4.17: Graph of number of micro wires (with that particular iron content) against iron content (%) for the current density of 45 A/dm2 59

Figure 4.18: Graph of hole diameter (in µm) against composition (in %) for a current density of 45 A/dm2 and a plating time of 3 hours 61

Figure 4.19: Graph of nickel content (%) against number of micro wires for a hole diameter of 200 µm using a current density of 45 A/dm2 with a plating time of 3 hours 63

Figure 4.20: Graph of nickel content (%) against number of micro wires for a hole diameter of 300 µm using a current density of 45 A/dm2 with a plating time of 3 hours 63

Figure 4.21: Graph of number of micro wires (with that particular nickel content) against nickel content (%) for a plating period of 3 hours at a current density of 45 A/dm2 64

Figure 4.22: Graph of number of micro wires (with that particular nickel content) against nickel content (%) for a plating period of 6 hours at a current density of 45 A/dm2 65

Figure 4.23: Graph of number of micro wires (with that particular nickel content) against nickel content (%) for a plating period of 9 hours at a current density of 45 A/dm2 65

Figure 4.24: Graph of average nickel content (%) against plating time (in hours) for a current density of 45 A/dm2 66

Figure 4.25: Variation of deposited structures and deposition rate with plating time of (a) 3, (b) 6 hours and (c) 9 hours 67

Figure 4.26: Trapped air in template holes preventing electrolyte from reaching substrate and electrodeposition cannot take place 68

Figure 4.27: Photograph of template showing excess Ni-Fe on the template edges, as well as deposited Ni-Fe emerging from the template holes 71

Figure 4.28: SEM picture showing the overgrowth layers of deposited Ni-Fe 71 Figure 4.29: (a) SEM picture showing the overview of the whole array The number

of holes filled with Ni-Fe deposits come up to be 87 out of 100 (b) SEM picture showing a close-up view of the array 72

Figure 4.30: SEM pictures showing (a) top view; (b) 3-D view: of ablated copper surface after the covering PVC layer is removed During laser microdrilling the PVC was pasted onto the copper using epoxy adhesive and was removed thereafter so that the surface of the copper can be examined 73

Figure 4.31: Close up pictures of some of the deposited Ni-Fe pillars at (a) spot A; and (b) spot B; which have gaps or holes in them, which leads to the investigation of whether the pillars are hollow 73

Figure 4.32: Schematics of template on copper substrate The template was laser drilled before being pasted onto the copper 74

Figure 4.33: SEM images showing (a) top view; and (b) close-up 3-D view; of top surface of the array after filing 75

Figure 4.34: Schematics of the arrangement of arrays on the template 75 Figure 4.35: SEM images of the various arrays after filing: (a) 4×4 array; (b) 6×6 array; (c) 8×8 array; (d) 10×10 array 76

Figure 4.36: The template arrays and their compositions 77 Figure 5.1: Workflow for femtosecond laser machining of magnetic materials 79 Figure 5.2: Schematic diagrams showing dimensions of Vitrovac 6025X specimens with (a) 1 channel ablated and (b) 7 channels ablated 80

Figure 5.3: Influence of laser fluence on number of loop in order to ensure the ablation of a fully penetrated channel on the magnetic specimens with pulse repetition rate of 1 kHz and stage speed of 100 mm/min 81

Figure 5.4: SEM images of the channels fabricated at different laser fluences of (a) 13 J/cm2 and (b) 38 J/cm2 83

Figure 5.5: Charts showing the composition of Vitrovac 6025X at various locations when ablated by 3 J/cm2 84

Figure 5.6: Charts showing the composition of Vitrovac 6025X at various locations when ablated by 38 J/cm2 85

Figure 5.7: XRD results showing the intensity peaks of Vitrovac 6025X without laser ablation and at different laser fluences 86

Figure 5.8: Hysteresis loops obtained for applying an external field: (a) parallel; and (b) perpendicular to the magnetic foil fabricated at different laser fluences 87

Figure 5.9: Coercivities obtained for applying an external field: (a) parallel; and (b) perpendicular to the magnetic foil fabricated at different laser fluences 88

Figure 5.10: Hysteresis loops obtained for applying an external field: (a) parallel; and (b) perpendicular to magnetic foil with 1-7 laser fabricated channels 89

Figure 5.11: Coercivities obtained for applying an external field: (a) parallel; and (b) perpendicular to magnetic foil with 1-7 laser fabricated channels 90

Figure 5.12: FESEM images showing the effects of femtosecond laser irradiation on Permalloy at different laser fluences a) 14 mJ/cm2,b) 27 mJ/cm2,c) 34 mJ/cm2 and d)

51 mJ/cm2 The arrow indicates the scanning direction of the laser beam 93

Figure 5.13: FESEM images showing the effects of femtosecond laser irradiation on Permalloy different laser fluences a) 14 mJ/cm2,b) 27 mJ/cm2,c) 34 mJ/cm2 and d) 51 mJ/cm2 for 90° polarization The arrow indicates the scanning direction of the laser beam 94

Figure 5.14: FESEM images showing the effects of applying a) 2, b) 3, c) 4 and d) 5 laser pulses on Permalloy at laser fluence 14 mJ/cm2 96

Figure 5.15: FESEM images showing the effects of applying a) 2, b) 3, c) 4 and d) 5 laser pulses on Permalloy at laser fluence 27 mJ/cm2 97

Figure 5.16: Chart showing the average percentage of nickel and iron content for 8 different specimens before laser ablation 98

Figure 5.17: XRD results showing the intensity peaks of Permalloy at different laser fluences 100

Figure 5.18: Graph shows the variation in average crystallite/grain size due to different laser fluences 101

Figure 5.19: Hysteresis loops measured along the magnetic field after laser ablation 103

Figure 5.20: Percentage difference coercivity before and after laser ablation 104 Figure 6.1: SEM images of the drilled holes and their replicas 113 Figure 6.2: Exit of a hole produced in 1 mm thick steel plate using diffractive optical element without any change in the polarization state 114

Figure 6.3: Exit of a hole produced in 1 mm thick steel plate using diffractive optical element and polarization trepanning technique 114

I(t) Laser intensity

LIST OF PUBLICATIONS

Journal paper:

1 L.S Tan, H.L Seet, M.H Hong and X.P Li, “Effects of femtosecond laser

ablation on Vitrovac 6025X”, Journal of Materials Processing (In Press)

2 H.L Seet, L.S Tan, M.H Hong, K.S.Lee, H.H Teo, C.H Lui and X.P Li,

“Laser-drilled PVC template for electrodeposition of multi-core orthogonal

fluxgate sensing element”, Journal of Materials Processing (In Press)

3 L Jiang, L.S Tan, J.Z Ruan, W.Z Yuan, X.P Li and Z.J Zhao, “Intermittent

deposition and interface formation on the microstructure and magnetic

properties of NiFe/Cu composite wires”, Physica B 403 (2008) 3054-3058

Conference paper:

1 L.S Tan, H.L Seet, M.H Hong and X.P Li, “Effects of femtosecond laser

ablation on Vitrovac 6025X”, 8th Asia-Pacific Conference on Materials Processing, 15th – 20th June 2008, Guilin & Guangzhou, China

2 H.L Seet, L.S Tan, M.H Hong, K.S Lee, H.H Teo, C.H Lui and X.P Li,

“Laser-drilled PVC template for electrodeposition of multi-core orthogonal fluxgate sensing element”, 8th Asia-Pacific Conference on Materials Processing, 15th – 20th June 2008, Guilin & Guangzhou, China

Chapter 1 Introduction

1.1 Motivation

Magnetic sensors are widely used in areas such as high-density magnetic recording, navigation, military and security, target detection and tracking, anti-theft systems, non-destructive testing, magnetic marking and labeling, geomagnetic measurements, space research, measurements of magnetic fields onboard spacecraft and biomagnetic measurements in the body [1] The types of magnetic sensors available in the market include induction sensors, fluxgate sensors, Hall Effect magnetic sensors, magneto-optical sensors, giant magnetoresistive (GMR) sensors, resonance magnetometers, and superconducting quantum interference device (SQUID) gradiometers

Currently, the market demand is shifting towards cheaper, smaller size and lower power consumption alternatives without any comprise on the performance level in terms of sensitivity However, there is a trade-off between sensitivity, size, power and cost By making a comparison between existing magnetic sensors in terms of the processing cost and power consumption, GMR sensor shows the lowest cost and power consumption [1] However, the field sensitivity of the GMR sensor is rather low (~1%/Oe) A recent discovery based on giant magnetoimpedance (GMI) has lead

to an improvement for high performance magnetic sensors The field sensitivity of a typical GMI sensor can reach a value as high as 500%/Oe [1]

GMI effects occur when a soft ferromagnetic conductor is subjected to a small alternating current (ac) in the presence of a direct current (dc) field and a large change

in the ac complex impedance of the conductor is achieved [1] The relative change of

the impedance (Z) with applied field (H), which is defined as the GMI effect, is

expressed by

)(

)()(

%100(%)/

max

max

H Z

H Z H Z Z

where H max is the external magnetic field sufficient to saturate the impedance

It has been demonstrated by Chirac et al [2] on glass-coated microwires array that magnetoimpedance (MI) response of the array is dependent on the number of active microwires and the current frequency The results show that the highest magneto-impedance response was found close to 6MHz for one microwire, 3MHz for two connected microwires, 2MHz for three, four and five connected microwires, and 1MHz for the next microwires, when using a current of 10 mA Recent research has also shown that a multi-core sensing element increases the sensitivity of orthogonal fluxgate sensor as a function of the number of ferromagnetic cores [3] Measured sensitivity increased exponentially with the number of ferromagnetic core wires increased, where the “linear” curve was calculated by multiplying the number and the sensitivity of a single-core sensor [3]

In order to enhance the performance of such magnetic sensors, one available option is through improved processing and manufacturing[4] For magnetic sensors used to detect weak magnetic fields, a highly sensitive sensing element, with high permeability, is needed Therefore, the focus on this project is towards improving the sensitivity of the magnetic sensors through its fabrication processes In this project, two methods are adopted to obtain regularly packed multi-core structures

The first method makes use of nanosecond 355nm Nd:YAG laser to produce polymeric template for electroplating nickel-iron (Ni-Fe) ferromagnetic cores Currently, wires array can be achieved through template electrodeposition and the templates used in the process are usually anodic aluminum oxide (AAO) [5], PMMA and co-polymer templates developed by e-beam lithography [6] However, the holes arrangement for these templates is random Alternative approach is to laser-drill the templates Advantages of using laser-drilled templates include fast patterning, repeatability and the desired template design can be programmed by software such as Mastercam In this research work, studies are made on the selection of machine-drilled or laser-drilled templates, methods of adhesion and fabrication of single and stacked templates Different laser processing parameters such as laser fluence, laser irradiation time, number of pulses and laser focal length are investigated to optimize laser drilling process on polymeric template

Soft magnetic material (nickel-iron alloy) is chosen as the study material due to its high initial permeability as well as near-zero magnetostriction These quantities are particularly significant in nickel-iron alloys between 75 and 80% nickel content [7] Therefore, Permalloy with the composition percentage ratio of 81:19 NiFe was selected as the ferromagnetic material Research has shown that to obtain the selected composition for the electroplated wires, the general approach is to increase the deposition rate of iron in three different methods: 1) the current density can be decreased; 2) the plating duration can be decreased and; 3) the amount of iron (II) sulfate heptahydrate (FeSO4.7H2O) in the plating solution can be increased[8, 9] Hence, template electrodeposition of nanocrystalline Ni-Fe is adopted as a mean of achieving arrays of magnetic wires which would function as multi-core sensing

elements for weak magnetic fields Templates with different array and hole sizes are fabricated using laser drilling techniques Thereafter, direct current electrodeposition was carried out under controlled environment to deposit nanocrystalline Ni-Fe The magnitude of the plating current and plating time were varied to gain further understanding of the electrochemistry involved Characterization of the obtained

structures is carried out in terms of materials properties

The second method involves femtosecond 800nm Ti:Sapphire laser machining on commercial Vitrovac 6025X ribbons and electron-beam evaporated (Ni-Fe) thin films

in ambient conditions to produce regular arrays of long magnetic stripes Unlike nanosecond laser machining, processing of magnetic materials using femtosecond laser does not create any domain pinning effects [10] In addition, A Semerok et al has shown that femtosecond laser has higher ablation efficiency than nanosecond and picosecond laser The reason being for a femtosecond regime, a laser pulse terminates before the energy is completely redistributed in the solid matter It is likely that the energy is deposited in the matter without laser plasma interaction resulting in better ablation efficiency than the one in the picosecond and nanosecond regimes [11] Hence, these justify the use of femtosecond laser as a suitable candidate to fabricate magnetic sensor Thin film and ribbons are chosen due to the even surface they have

as compared to the curved profile of wires This allows laser processing to be carried out with minimum interference from scattered reflected rays

Electron-beam evaporation is adopted as sputtering process is not available Literatures have shown that this method is not recommended for deposition of alloy material, as the composition is dependent on the vapour pressure of the elements

However, the vapor pressure of nickel and iron are quite close to each other The vapor pressure of nickel and iron are 1 Pa at 1783K [12] and 1 Pa at 1728 K [13] respectively Hence, electron-beam evaporation method is used Amorphous (non-crystalline) ferromagnetic alloy such as Vitrovac 6025X was also obtained from Sekels, Germany The amorphous state is achieved by rapid quenching of a liquid alloy In order to be termed as amorphous, the material must consist of about 80% transition metal (usually Fe, Co, or Ni) Metalloid component (B, C, Si, P, or Al) is also required to lower the melting point Advantages of amorphous materials are that their properties are nearly isotropic (not aligned along a crystal axis); this results in low coercivity, low hysteresis loss, high permeability, and high electrical resistivity [14]

Large GMI effect exists in magnetic materials having low resistivity, high magnetic permeability, high saturation magnetization and small ferromagnetic relaxation parameter [1] Though crystalline ferromagnetic materials such as Ni-Fe have lower resistivity, but amorphous ones have better soft magnetic behavior in terms of higher magnetic permeability and saturation magnetization because they lack magnetocrystalline anisotropy For both magnetic materials, the influence of different laser parameters on the magnetic materials will be investigated and the results are characterized in terms of both material and magnetic properties

1.2 Objectives of Present Study

The project primary objective is to develop multi-core structured functional magnetic materials of extremely high permeability, using either nanosecond or femtosecond lasers The scope of the project includes:

1) Utilizing nanosecond laser and optimizing laser processing parameters in approaches to achieve the desired structures, in particular, optimizing laser process parameters for laser assisted micro-drilling of templates used in electrodeposition to achieve Ni-Fe pillars arrays

2) Carrying out studies and optimization of femtosecond laser processing parameters on magnetic materials such as Vitrovac 6025X foils and Ni-Fe thin films

3) Characterizing the materials and magnetic properties of the achieved structures, in particular verifying the extent of laser processing on the resulting properties

4) Evaluating the feasibility of using laser processing approaches on magnetic materials fabrication

1.3 Organization of Thesis

This thesis is divided into 6 chapters Chapter 1 gives an overall view of the introduction, motivation, project objectives and organization of the thesis Chapter 2 consists of the literature reviews done on magnetic materials and their relevant theories and the effects of nanosecond and femtosecond lasers processing on magnetic materials Fabrication methods used to obtain multi-core structures were also discussed Chapter 3 covers the experimental setups and characterization equipment used Chapter 4 deals with the use of nanosecond (Nd:YAG) laser-assisted micro-drilling of templates and electroplating deposition Chapter 5 talks about the effects of femtosecond (Ti:Sapphire) laser machining on magnetic materials such as Vitrovac 6025X and nanocrystalline Permalloy Chapter 6 gives the final conclusions of the thesis and suggested recommendations to improve the work

Chapter 2 Literature Review

2.1 Magnetic Materials and Relevant Theories

2.1.1 Basic Classification of Magnetic Materials

Magnetic materials can be classified according to their magnetic behaviors into three classical categories, namely: diamagnets, paramagnets and ferromagnets Beside these three categories, there are ferrimagnets, antiferromagnets, helimagnets and superparamagnets which are closely related to ferromagnets [14] as they are magnetically ordered The classification is dependent on their bulk magnetic susceptibility Table 2.1 shows the summary of the various categories of magnetic materials

Table 2.1: Summary of classification of magnetic properties

Ferromagnetism

Parallel aligned magnetic moments

For diamagnetic materials, the electrons are paired with electrons of opposite spin Therefore, the orbital currents are zero As a result, the atoms do not have any net magnetic moment in the absence of an applied field The spinning electrons precede under the influence of an applied external field Magnetization occurs in the direction opposite to that of the applied field, as predicted by Lenz’s Law This law states that when a current is induced by a change in magnetic field, the magnetic field produced

by the induced current will oppose the change Diamagnetic effects are present in all materials Examples of diamagnetic materials include water, wood, copper, mercury and gold [15]

Paramagnetic materials have positive magnetic susceptibility These atoms have randomly oriented magnetic moments due to the thermal motion Unlike ferromagnets, paramagnets under the influence of an external magnetic field, the electrons would be weakly attracted to the magnetic poles Hence, the total magnetization will drop to zero when the external field is removed The amount of induced magnetization is rather small since only a small fraction of the spins will be oriented by the external field And this amount is proportional to the field strength which explains the linear dependency [16]

Atoms in ferromagnetic materials are regularly arranged in a lattice The magnetic moments are parallel aligned The attraction experienced by ferromagnets is non-linear and much stronger, so that it is easily observed Examples of ferromagnetic materials at and above room temperature are iron, cobalt and nickel When these ferromagnetic materials are heated, they become thermal agitated The degree of alignment of the atomic magnetic moments decreases and hence the saturation

magnetization also decreases Eventually the thermal agitation becomes so great that the material becomes paramagnetic The temperature of this transition is the Curie temperature, TC (Fe: TC =770°C, Co: TC =1115°C and Ni: TC =354°C) [14]

2.1.2 Ferromagnetic Materials and Their Properties

Ferromagnetic materials are characterized by having high magnetic flux when under

an applied magnetic force Magnetic flux is defined as the total number of magnetic lines of force passing through a specified area in a magnetic field [17] However, there is a limit to the increasing trend of magnetic flux This phenomenon is termed as the magnetic saturation [18] Generally, ferromagnetic materials can be divided into two main categories: hard or soft magnetic materials Hard magnetic materials are characterized by low permeability and are more difficult to be magnetized, while soft magnetic materials are the opposite They are defined by high permeability and low coercivity

2.1.3 Curie Temperature Tc

Curie temperature is the maximum temperature beyond which any ferromagnetic properties vanish An example is magnetic saturation It is temperature-dependent At Curie temperature, it practically vanishes The maximum value is reached at absolute zero [7]

2.1.4 Hysteresis

Hysteresis is the ability of ferromagnets to remain magnetized after being exposed to

an external magnetic field When the field is removed, the fraction of magnetic

saturation left is termed as remanence of the material [19] Depending on the anisotropy of the magnetic material, the shape of the hysteresis curve varies Virgin curve, indicated by the dotted line in Figure 2.1, exists if the material is heated beyond its Curie temperature

Figure 2.1: A typical hysteresis curve [19]

2.1.5 Factors Affecting Magnetic Quality

Permeability,μ, refers to the magnetic conductivity of magnetic materials in the presence of magnetic fields It is described by the following equation [17]

2.1.6 Domain Wall Theories

Ferromagnetic materials exhibit a long-range ordering phenomenon at the atomic level, which causes the unpaired electron spins to line up parallel with each other in a region called a domain [19] When ferromagnetic material is subjected to an external magnetic field, two situations occur simultaneously Firstly, individual domain will align itself to the direction of the applied field Next, domains that are in the direction

of the field dominate over those that are not as seen in Figure 2.2 Greater field is required in order to ensure all domains shift their alignment

Figure 2.2: Process of magnetization[19]

This leads to the effect of magnetic saturation [17] The extent of domain wall rotations is limited by the thickness of the material Ignoring the influence of temperature on the magnetic moment, the amount of saturation depends on the magnitude of the magnetic moment of the atoms involved and on the number of such moments per unit volume [17] It is thus determined by composition and crystal lattice of the magnetic material [17]

2.1.7 Magnetization Rotation

Magnetic anisotropy is the preferred direction of a material to magnetize along a particular axis and to return to the same axis in the absence of an external magnetic field [20] This axis is defined as the easy axis Based on the factors affecting the

direction of the easy axis, there are several types of magnetic anisotropy as illustrated

in Table 2.2 The magnitude and the type of magnetic anisotropy affect the magnetization and hysteresis curves in magnetic materials

Table 2.2: Types of anisotropy

Type of anisotropy

Factors that influence the easy axis of

magnetization

2.1.7.1 Crystalline Anisotropy

A ferromagnetic single crystal has its preferred magnetizing directions In hexagonal

close packed cobalt, the path is parallel to the hexagonal axis [7] For a cubic crystal

(such as iron or nickel), the easy axes directions are along the cube edge or cube diagonal respectively [7] Magneto-crystalline anisotropy arises from the coupling between the electron spin and the orbital motion of the electron [21]

2.1.7.2 Stress Anisotropy

The effect of tensile stress on anisotropy is similar to applying an external magnetic

field to the ferromagnetic material except for the changes in dimensions Tensile stress pulls the atomic magnets in the direction of the force This causes the atomic magnets to be closer to each other vertically

2.1.7.3 Shape Anisotropy

Demagnetizing field is present in all magnetized bodies Its effect will depend strongly on the geometric shape of the bodies A long thin needle-shaped grain will

have a smaller demagnetizing factor if magnetization is along its longitudinal axis, as compared to its perpendicular axis

2.2 Types of Lasers

2.2.1 Nanosecond Laser

For nanosecond laser pulse (Figure 2.3), the pulsed laser beam causes the molten material to be ejected Due to the repercussion effects of the shock waves, surface ripples occur and adjacent structures are damaged Microcracks with significant heat affected zones can be observed Formation of surface debris and recast layer are found along the ablated region Figure 2.4 shows the SEM image of an ablated metal channel produced by a 532nm/30ns Nd:YAG laser

Figure 2.3: Nanosecond laser processing [22]

Figure 2.4: SEM image of an ablated metal channel created by a nanosecond laser

The interaction between the laser pulse and the solid material depends on physical properties of the material, the ambient conditions and the laser processing parameters such as wavelength, pulse duration, energy and spot size

2.2.2 Femtosecond Laser

The advantages of material processing with femtosecond laser pulses include defined and much lower ablation threshold as compared to nanosecond laser ablation; negligible heat affected zone; highly precise control of ablation geometry; high repeatability and process efficiency; fast, efficient and localized deposited energy into nearly all kinds of materials, even if they are transparent for the laser wavelength In addition, femtosecond laser processing provides flexibility processing for direct-write approach and the interaction volume is less than 1 μm3

[23-27] Figure 2.5 shows a typical SEM image of a channel produced by femtosecond (Ti:Sapphire) laser ablation

Figure 2.5: SEM image showing minimum amount of re-deposited material after femtosecond laser ablation

The exact mechanism and theoretical explanation behind femtosecond laser ablation

on metallic materials have been discussed by F Korte et al [26] The laser energy is initially absorbed inside the surface layer by bound and free electrons Due to inverse bremsstrahlung, excitation and ionization of the material and heating of free electrons occur The absorbed laser energy is deposited into the electronic subsystem The absorbed energy is transferred to the lattice for bond breaking and material expansion

As all these processes occur in picoseconds, thermal diffusion is negligible for laser fluences close to the ablation threshold Therefore, ablation occurs without the formation of heat- and shock-affected zones Figure 2.6 gives an illustration of the effect of femtosecond laser pulse drilling

Figure 2.6: Nanosecond laser processing [22]

The spatial and temporal evolution of the electron and lattice temperatures (T e and T i, respectively) in a thin surface layer with the subsequent material expansion is described by the following one-dimensional equations [26]:

x

u P S T T x

x Q

−

−

=γ( ) ( ) (2.3)

)(P c P e P i

Where x is the direction perpendicular to the target surface, d/dt = ∂/∂t + u∂/∂x, and C e

and C i are the heat capacities (per unit volume) of the electron and lattice subsystems The parameter γ characterizes the electron-lattice coupling, ρ and u the density and velocity of the evaporated material, P e and P i the thermal electron and ion pressures,

Pc the elastic ( or ‘cold’) pressure, Q(x) = -k e (T e )∂T e /∂x the heat flux, and S =

I(t)Aαexp(-αx) the laser heating source term Here I(t) is the laser intensity, A and α

are the surface absorptivity and the material absorption coefficient, and k e is the electron thermal conductivity The first two equations are the energy conservation equations for the electron and ion subsystems Equation 2.4 expresses Newton’s law and the last one is the continuity equation During a femtosecond laser pulse, material expansion can be neglected The last terms in (1) and (2) (containing ∂u/∂x) can be omitted

As mentioned previously, due to inverse bremsstrahlung, the laser energy is initially absorbed by free electrons This is followed by fast energy relaxation within the electron subsystem, thermal diffusion, and an energy transfer to the lattice due to electron-phonon coupling The thermal conductivity of the lattice can be neglected since the heat diffuses much faster through the electron subsystem

Since the heat capacity of the lattice is much higher than the electronic heat capacity, the free electrons of the material are heated to very high transient temperatures during the laser pulse Hence, the electron pressure term in Equation (2.4) becomes dominant This creates a tensile force and causes material expansion Ablation takes place when the expanding force exceeds that of binding force determined by the

gradient of the cold pressure The threshold electron pressure, P th, required to initiate

ablation can be estimated by P th ∼ ρo c o 2, where ρo is the solid density and c o is the

speed of sound in the solid material At P ≥ P th, atoms overcome the potential barrier and become completely free Assuming that the whole laser energy is deposited into the electron subsystem, the threshold laser fluence necessary for ablation by ultrashort laser pulses can be estimated by:

F th∼ P thδ/2 (2.6) Where δ = 1/α is absorption length For metals, considering ρo ~ 10 g/cm3, c o ~ 5 x

105 cm/s, and the skin depth δ ~ 10nm, the threshold pressure is estimated to be 2.5 ×

1011 Pascal and laser fluence threshold is 125 mJ/cm2

However, P.R Herman et al [27] had also mentioned that undesirable effects and poorly defined processing windows exist for femtosecond laser These include poor morphology, such as periodic surface structures, melt, debris and surface swelling; shock-induced microcracking; and ‘gentle’ and ‘strong’ ablation phases developing with increasing number of pulses

2.2.3 Effects of Nanosecond and Femtosecond Lasers on Magnetic Materials

Le Gall et al [28] had investigated the effects of nanosecond multi-wavelength ionized argon laser on sputtered nanocrystalline magnetic thin film, such as NiFe, FeN and Co/Cu Ablated tracks had been produced in magnetic films when submitted

in air or near the focalization point Figure 2.7 shows the hysteresis loops for the deposited film and after fabrication of grooves parallel to the as-deposited hard axis

as-by the laser After laser ablation, coercivity of the film was found to be decreased with anisotropy towards the hard axis of the film Domains pinning towards the hard axis may have occurred which explained the reduction in coercivity

Figure 2.7: Easy axis M-H loops (a) before and (b) after fabrication of grooves

parallel to the as-deposited hard axis in FeN monolayer [28]

To investigate the influence of the density of dotted lines on the domain structure and shape of the hysteresis loops, A Zele et al [10] performed multiple pulses of ultraviolet (λ = 308 nm) XeCl excimer laser radiation on the surface of Finemet ribbons These lines represent new pinning centres for domain wall movements, leading to the increase of the static coercivity as seen in Figure 2.8 It was observed that the hysteresis loops of laser-treated samples are steeper Coercivity increases with the density of circular holes because the domain structures consist of larger number of wider and movable domain walls These holes act as domain walls pinning centres

(a)

(b)

Figure 2.8: Hysteresis loops of Finemet: (a) sample non-treated by laser, (b) laser treated sample with small density dotted lines, and (c) laser treated sample with high density of dotted lines [10]

Jia et al [29] reported the detailed studies on the effects of femtosecond laser ablation

on surface characteristics and subsurface microstructures of amorphous Finemet The results show that three types of ripple structures existed on the material surface in the gentle ablated (damaged) zone In addition, as observed with x-ray diffraction analysis (Figure 2.9), amorphous form is kept in the damaged zone, and there is few crystallization form in ablation zone

Figure 2.9: XRD analysis of amorphous alloy ribbon before and after laser micromachining [29]

Figure 2.10: Photograph showing patterned template on developed negative photoresist layer

However, photolithography may not be an ideal case to fabricate templates for electrodeposition of magnetic material because the photoresist layer cannot be used as the actual template as the layer will not be able to withstand the plating conditions for example, long plating hours and acidic environment Further steps, such as, etching are required to transfer the patterned photoresist to other suitable template materials for example a hard mask

Another template fabrication method is laser interference lithography (LIL) Advantages of LIL are low cost and simple set-up, fast fabrication speed, non-contact

and maskless process and lastly, it is applicable to any type of substrate LIL positions two monochromatic waves to form periodic patterns The period of the patterns (Figure 2.11) depends on the wavelength (λ) and wave intersecting angle, (θ)

super-as shown in the standing wave period,

Figure 2.11: FESEM image of developed bi-layered photoresist after single exposure

Subsequent deposition of material through sputtering or electron-beam evaporation produces long magnetic nanowires The ability to ensure that every nanowire can be used as sensing elements without any damage proves to be a great challenge By giving the negative photoresist layer a second exposure after rotating the substrate at 90°, meshed-like photoresist layer can be obtained In this case, nano-dots or nano-pillars are formed However, the aspect ratio of such structures is not high enough to meet the requirements of the sensing applications

Positive photoresist

PMGI

2.3.1.2 Laser Micromachining Method

Laser micromachining is an effective method to fabricate large and regular array of micro-holes The ablation rate is dependent on a range of parameters such as the laser fluence, pulse repetition rate and laser irradiation duration, as well as thermal and chemical properties of the material Advantages of laser drilling include applicable to wide range of materials, high resolution and accuracy and non-contact machining process Important parameters of laser include laser power, wavelength of laser, spot size and depth of focus Material properties (thermal conductivity, thermal diffusivity, melting and boiling temperature, and latent heat of vaporization) can also affect the machining quality

The effects of using different types of lasers on laser-drilling for polymeric materials have been investigated by different researchers In particular, Lazare and Tokarev (2004) [30] reported some experimental and theoretical work on microdrilling of polymers with pulsed KrF laser radiation while Masmiati and Philip (2007) [31] presented experimental investigations on the drilling of polymers using a carbon dioxide (CO2) laser There have also been many studies of laser processing on polymers Serfetinides et al (2001) [32] reported the experimental results of the ablation rate per pulse as a function of the laser fluence and images of the surface morphology for several organic polymer materials Pham et al (2002) [33] investigated the relationship between ablation rates and various polymer thermal properties