LASER SURFACE ALLOYING AND IN-SITU FORMATION OF ALUMINIUM METAL COMPOSITES REINFORCED WITH CERAMICS AND INTERMETALLICS Luyolo Andrew Baxolise Mabhali A thesis submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, in fulfilment of the requirements for the degree of Doctor of Philosophy. Johannesburg, 2011 DECLARATION i DECLARATION I declare that this thesis is my own, unaided work. It is being submitted for the Degree of Doctor of Philosophy in Engineering to the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination in any other University. _______________________________________________ (Signature of candidate) _______________ day of ____________________________ (year) _________ ABSTRACT ii ABSTRACT This thesis describes the novel laser surface alloying of aluminium AA1200 with various combinations of Ni, Ti and SiC powders, using a 4.4kW Rofin Sinar Nd:YAG laser in order to improve its mechanical and tribological properties. The laser alloying parameters were optimized on the breakdown systems of the complex Al-Ni-Ti-SiC system. Various analytical techniques were used to study the microstructures produced. Wear testing was conducted under sliding and abrasion conditions while the fracture mechanisms were investigated using impact tests. Aluminium surfaces reinforced with metal matrix composites and intermetallic phases were achieved. The phases present depended on the composition of the alloying powder mixture. Al reacted with Ni to form Al3Ni and Al3Ni2 intermetallic phases while Ti reacted with Al to form an Al3Ti intermetallic phase. Some of the SiC particles dissociated and reacted with either Al or Ti to form Al4C3, Al4SiC4, TiC or Ti3SiC2 phases. Si reacted with Ti to form a Ti5Si3 phase. An increase in surface hardness was achieved, up to a maximum of 13 times that of aluminium when alloying with 80wt%Ni + 15wt%Ti + 5wt%SiC. The increase in hardness was attributed to the intermetallic phases especially the Al3Ni2 phase. Alloying led to a 4-38% improvement in the wear resistance of the pure aluminium under sliding wear conditions and a 19-82% improvement under three body abrasion wear conditions. The predominant wear mechanisms for both wear types were groove formation by ploughing and cutting action of the abrasive particles, smearing, material pile-up, extensive cracking of the intermetallic phases and fracturing of the embedded SiC particles in the MMCs. Alloying led to a 31-50% decrease in the impact resistance of the pure aluminium. Brittle fracture of the SiC particles and transgranular cracking of the intermetallic phases were observed for the laser alloyed surfaces while ductile fracture was observed for the bulk aluminium. DEDICATIONS iii DEDICATIONS To My children: Khanyisa Mabhali, Luyanda Mabhali and Nobuhle Gama My wife: Nonhlanhla Jacqueline Gama My parents: Margaret Xoliswa Plaatjie and Linda Oscar Mabhali My grandmother: Nondindi Julia Plaatjie The Plaatjie, Mabhali and Nzube families For Their love, support and patience ACKNOWLEDGEMENTS iv ACKNOWLEDGEMENTS I would firstly like to thank my supervisors Dr Natasha Sacks (Wits) and Prof. Sisa Pityana (CSIR) for giving me the opportunity to do this project and their patience. I?m eternally grateful to you for your assistance. The Department of Science and Technology (DST), the National Research Foundation (NRF) and Council for Scientific and Industrial Research (CSIR) are acknowledged for their financial support. I am also grateful to the University of the Witwatersrand for giving me an opportunity to conduct my studies with them. I would also like to thank my colleagues at the CSIR National Laser Centre, Herman Burger, Christopher Meacock, Lipson Rampedi, Oscar Sono and everyone in Laser Materials Processing, good luck in your future endeavours. I am eternally grateful to Mr Clive Oliphant (National Metrology Institute of South Africa) for his assistance with the SEM. My sisters, Zintle Plaatjie, Zimkhitha Plaatjie and Cebisa Mabhali, thank you for all the support and for believing in me. To all my friends, especially Sandile Xotyeni, Freddie Makgere and Neo Mokgatlhe, thank you for always being there, the drinking sessions and reminding me to have some fun. To Amaqhawe, thank you for always being there for me and reminding me where we come from and how far we?ve come. TABLE OF CONTENTS v TABLE OF CONTENTS DECLARATION .................................................................................. i ABSTRACT?. .................................................................................... ii DEDICATIONS .................................................................................. iii ACKNOWLEDGEMENTS ............................................................... iv TABLE OF CONTENTS .................................................................... v LIST OF FIGURES ........................................................................... ix LIST OF TABLES ........................................................................... xxv 1. INTRODUCTION ........................................................................... 1 2. LITERATURE REVIEW ............................................................... 4 2.1 Aluminium .................................................................................................................. 4 2.2 Laser Processing ......................................................................................................... 6 2.2.1 Nd:YAG (Neodymium yttrium aluminium garnet) lasers ............................... 8 2.2.2 Laser alloying .................................................................................................. 9 2.2.3 Advantages of laser alloying ......................................................................... 13 2.2.4 Disadvantages of laser alloying .................................................................... 13 2.3 Laser alloying with Al, Ni, Ti and SiC ...................................................................... 14 2.3.1 Laser alloying with Ni or Ni containing compounds .................................... 14 2.3.2 Laser alloying with Ti or Ti containing compounds ..................................... 18 2.3.3 Laser alloying with Ni and Ti ....................................................................... 19 2.3.4 Laser alloying with SiC or SiC containing compounds ................................ 22 2.3.5 Laser alloying with TiC ................................................................................. 26 2.3.6 Laser alloying with Ni and SiC or TiC.......................................................... 27 2.3.7 Laser alloying with Ti and SiC ..................................................................... 30 2.4 Wear .......................................................................................................................... 32 2.4.1 Factors affecting abrasive wear rates ............................................................ 33 2.4.2 Mechanisms of abrasive wear ....................................................................... 38 2.4.3 Wear of aluminium alloys ............................................................................. 42 TABLE OF CONTENTS vi 2.5 Fracture ..................................................................................................................... 45 2.5.1 Ductile fracture ............................................................................................. 45 2.5.2 Brittle fracture ............................................................................................... 46 3. EXPERIMENTAL PROCEDURE .............................................. 49 3.1 Aluminium AA1200 characterization........................................................................ 49 3.1.1 Specimen preparation .................................................................................... 49 3.1.2 Vickers Hardness ........................................................................................... 50 3.1.3 Density .......................................................................................................... 50 3.2 Powder Characterization ........................................................................................... 50 3.2.1 Powder particle morphology ......................................................................... 51 3.2.2 Hardness of the powder particles .................................................................. 51 3.3 Synthesis of the alloys ............................................................................................... 51 3.3.1 Optimization tests ......................................................................................... 52 3.3.2 Breakdown alloy systems .............................................................................. 55 3.4 Hardness of the alloys ............................................................................................... 55 3.5 Wear testing ............................................................................................................... 55 3.5.1 Two body abrasion wear tests ....................................................................... 56 3.5.2 Three body dry abrasion wear tests ............................................................... 58 3.6 Impact testing ............................................................................................................ 60 3.7 Microscopy analysis .................................................................................................. 62 3.7.1 Optical microscopy ....................................................................................... 62 3.7.2 Stereo microscopy ......................................................................................... 62 3.7.3 Scanning electron microscopy ...................................................................... 63 3.7.4 X-ray diffraction ........................................................................................... 63 4. RESULTS: SYNTHESIS OF THE ALLOYS ............................. 64 4.1 Aluminium AA1200 characterization........................................................................ 64 4.2 Powder characterization ............................................................................................ 66 4.3 Optimization of the laser surface alloying process ................................................... 71 4.3.1 Microstructure of the alloyed layers ............................................................. 72 4.3.2 Hardness of the alloyed layers ...................................................................... 75 4.4 Laser alloying of the breakdown alloy systems ........................................................ 78 4.4.1 Laser alloying Al with Ni .............................................................................. 79 4.4.2 Laser alloying Al with Ti .............................................................................. 80 4.4.3 Laser alloying Al with Si .............................................................................. 82 TABLE OF CONTENTS vii 4.4.4 Laser alloying Al with TiC ............................................................................ 83 4.4.5 Laser alloying Al with SiC ............................................................................ 85 4.4.6 Laser alloying Al with Ni and Ti ................................................................... 87 4.4.7 Laser alloying Al with Ni and Si ................................................................... 89 4.4.8 Laser alloying Al with Ti and Si ................................................................... 90 4.4.9 Laser alloying Al with Ni and SiC ................................................................ 91 4.4.10 Laser alloying Al with Ti and SiC ............................................................... 94 4.5 Laser alloying Al with Ni, Ti and SiC ....................................................................... 96 5. RESULTS: MECHANICAL AND WEAR TEST RESULTS OF THE ALLOYS ................................................................................ 102 5.1 Hardness .................................................................................................................. 102 5.2 Two body abrasion wear tests ................................................................................. 105 5.3 Three body dry abrasion wear tests ......................................................................... 111 5.4 Impact tests ............................................................................................................. 119 6. DISCUSSION .............................................................................. 124 6.1 Optimization of the laser surface alloying process ................................................. 124 6.2 Laser alloying of the breakdown alloy systems ...................................................... 126 6.3 Laser alloying Al with Ni, Ti and SiC simultaneously ............................................ 130 6.4 Effect of laser alloying on wear resistance .............................................................. 134 6.5 Effect of laser alloying on impact resistance........................................................... 138 6.6 Summary of the main results ................................................................................... 139 6.7 Industrial application of laser surface alloyed Al AA1200 ..................................... 141 7. CONCLUSIONS ......................................................................... 142 8. RECOMMENDATIONS ............................................................ 144 9. REFERENCES ............................................................................ 145 APPENDIX A: SYNTHESIS OF THE ALLOYS ........................ 158 TABLE OF CONTENTS viii APPENDIX B: WEAR TEST RESULTS ...................................... 200 APPENDIX C: IMPACT TEST RESULTS ................................. 260 APPENDIX D: PUBLICATIONS .................................................. 268 LIST OF FIGURES ix LIST OF FIGURES Figure Page 2.1 AA list for aluminium alloys [11]?..????????????...?..5 2.2 Schematic diagram of an Nd:YAG laser [15]???????????.8 2.3 Schematic diagram of a laser alloying process. [17]?????.???.10 2.4 Al-Ni phase diagram [30]??????????????????...15 2.5 Composition plot of aluminium concentration versus laser scan Speed at two different feed rates [33]???????????.??...16 2.6 Ti-Al phase diagram [42]??????????????????...19 2.7 Ni-Ti phase diagram [52]??????????????...................20 2.8 The Al-Ni-Ti partial isothermal section at 1050?C [57]?????........21 2.9 Al-Si-C phase diagram [64]?????????????????...23 2.10 MMC layer showing Al4SiC4, Al4C3, Si, and SiC phases [10]????..25 2.11 Micrograph showing TiC dendrites and Ti + Ti5Si3 eutectic phase [88]???????????????????????...?30 2.12 Schematic diagram illustrating two-body and three-body abrasive wear [102]???????????.?????????????.33 2.13 Graph of volumetric metal removal rate versus applied load [109]?...?35 2.14 Graph of removal rate versus grit diameter where D0 is the minimum grit size which will remove a significant amount of material [109]?????????????????????...?..36 2.15 (a) Round silica abrasives and (b) angular silica abrasives [98]??....?36 2.16 Diagram showing median and lateral crack formation in a brittle material due to indentation by a sharp indenter [98]. D is the damaged layer, M is the median crack and L is a lateral crack. The load is increased from (a) to (c) and progressively decreased from (d) to (f)????????????????????.......?..40 2.17 Schematic illustration of material removal in a brittle material by the growth of lateral cracks below a plastic groove [98]. b is the depth of the damaged layer where the lateral cracks occurred LIST OF FIGURES x and c is the length of the lateral cracks????????????.?.41 3.1 Experimental setup using A Nd:YAG laser??????????.?..52 3.2 Two body abrasive wear apparatus??..????????????..56 3.3 The traced sliding path of the specimens during two body wear Testing?????????????????????????....57 3.4 Three body dry abrasion wear instrument????.????????.59 3.5 The testing chamber of the three body abrasive wear apparatus showing a rubber wheel, a sample holder and a sand feeder?????.60 3.6 The impact testing machine?????????????.????..61 3.7 Specimen used for impact tests???????????.?????.61 4.1 SEM micrograph of aluminium AA1200 showing traces of FeAl3 in the ?-Al matrix?????????????????????.?.64 4.2 XRD pattern of aluminium AA1200 alloy???????????.?65 4.3 (a) SEM image of the Ni powder particles showing spherical, irregular and agglomerate particles. (b) Hardness indentations on a Ni particle????????????????????????67 4.4 XRD diffractograph of Ni powder showing Ni peaks??????.?..67 4.5 Ni powder particle size distribution curve??.?????????....68 4.6: (a) SEM image of the Ti powder particles showing irregular agglomerates. (b) Hardness indentations on a Ti particle?????..?68 4.7 XRD diffractograph of Ti powder showing Ti peaks??.?????...69 4.8 Ti powder particle size distribution curve??.??????????.69 4.9 (a) SEM image of the SiC powder particles showing irregular particles. (b) Hardness indentations on a SiC particle???????...70 4.10 XRD diffractograph of SiC powder showing SiC peaks??????...70 4.11 SiC powder particle size distribution curve????????...???71 4.12 Stereo micrographs showing the alloyed layers (AL) and thermo-affected layers (TL) for samples alloyed with powders containing (a) 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC, (b) 80wt%Ni + 15wt%Ti and 5wt%SiC, (c) 50wt%Ni + LIST OF FIGURES xi 20wt%Ti + 30wt%SiC, (d) 5wt%Ni + 80wt%Ti + 15wt%SiC, (e) 30wt%Ni + 50wt%Ti + 20wt%SiC , (f) 15wt%Ni + 5wt%Ti + 80wt%SiC and (g) 20wt%Ni + 30wt%Ti + 50wt%SiC at 10 mm/s laser scanning speed. The thickness of each layer is reported??????..???????????.??..74 4.13 Stereo micrographs of samples alloyed with (a) 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC, (b) 80wt%Ni + 15wt%Ti + 5wt%SiC, (c) 50wt%Ni + 20wt%Ti + 30wt%SiC, (d) 20wt%Ni + 30wt%Ti + 50wt%SiC at 20 mm/s laser scanning speed?...????????....75 4.14 Hardness profile of the laser treated aluminium AA1200.?????....76 4.15 Hardness profile of surface alloyed with powder containing 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC at various laser scanning speeds????????????.????????...?..77 4.16 SEM micrographs of an Al AA1200 laser alloyed with Ni powder showing an in-situ formed Al3Ni2 phase (grey) within an ?-Al (black) matrix??????????????.????????....79 4.17 XRD pattern of Al laser alloyed with Ni showing Al, Ni and Al3Ni2 phases??????????????????..????...80 4.18 SEM micrograph of an Al laser alloyed with Ti powder showing ?-Al (black) and Al3Ti (grey) phases?????????????.....81 4.19 XRD pattern of an Al laser alloyed with Ti showing Al, Ti and Al3Ti phases...............................................................................................81 4.20 SEM micrograph of an Al laser alloyed with Si showing ?-Al (black) and Al-Si eutectic (white) phases???...???????????....82 4.21 XRD of an Al laser alloyed with Si powder showing Al and Si phases????????????????????????.....83 4.22 SEM micrograph of an Al laser alloyed with a TiC powder showing TiC particles, ?-Al (black) and an Al3Ti (white) phase????...??..84 4.23 XRD of Al laser alloyed with TiC showing Al, Al3Ti and TiC phases.....84 4.24 Figure 4.24: SEM micrograph of an Al laser alloyed with SiC powder showing SiC particle (black particle), Al4SiC4 intermetallic phase (dark grey platelets), Si phase (white), ?-Al (grey) and LIST OF FIGURES xii Al-Si eutectic phase (white dots in the grey phase)????????...86 4.25 XRD of an Al laser alloyed with SiC showing Al, SiC, Si and Al4SiC4 phases??????????????????????...86 4.26 SEM micrograph of an Al laser alloyed with (a) 30wt%Ni + 70wt%Ti, (b) 50wt%Ni + 50wt%Ti and (c) 70wt%Ni + 30wt%Ti?.......88 4.27 XRD of an Al laser alloyed with 70wt%Ni + 30wt%Ti??...??.?....88 4.28 SEM micrographs of Al laser alloyed with 50wt%Ni + 50wt%Si showing (a) Al-Ni intermetallic phases and (b) Al, Al3Ni and NiSi2 phases???????????????????????...89 4.29 XRD of Al laser alloyed with 50wt%Ni + 50wt%Si???...?..............90 4.30 SEM micrographs of Al laser alloyed with 50wt%Ti + 50wt%Si showing (a) Al3Ti phase and (b) Al-Si eutectic?????????.....91 4.31 XRD of Al laser alloyed with 50wt%Ti + 50wt%Si????...???...91 4.32 SEM micrograph of an Al laser alloying with (a & b) 30wt%Ni + 70wt%SiC, (c) 50wt%Ni + 50wt%SiC and (d) 70wt%Ni + 30wt%SiC. In (a) AL is the alloyed layer and TL is the thermo-affected layer?????????????.??.....93 4.33 XRD of an Al surface laser alloyed with 70wt%Ni + 30wt%SiC?.........93 4.34 Micrographs of an Al laser alloyed with (a & b) 30wt%Ti + 70wt%SiC, (c) 50wt%Ti + 50wt%SiC and (d) 70wt%Ti + 30wt%SiC????.????????.................................................95 4.35 XRD of an Al surface laser alloyed with 50wt%Ni + 50wt%SiC?.........96 4.36 Effect of SiC content on alloys (a) 80wt%Ni + 15wt%Ti + 5wt%SiC and (b) 20wt%Ni + 30wt%Ti + 50wt%SiC. A good MMC was only formed in alloy (b)??????????????????.....98 4.37 Alloy 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC showing the different phases observed???????????????..??...100 4.38 Alloy 70wt%Ni + 20wt%Ti + 10wt%SiC showing the dendritic Al3Ni and the needle-like Al3Ti intermetallic phases?...?????...101 5.1 Hardness profile through cross-section of aluminium alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC. (A) is the alloyed, LIST OF FIGURES xiii is the interface between the alloyed and the aluminium substrate and (C) is the Al substrate?????????????????....103 5.2 Surface hardness of untreated aluminium and the laser alloyed surfaces?????????????????????....105 5.3 Wear response of untreated aluminium and laser alloyed surfaces?.....106 5.4 Wear rate of untreated and laser alloyed aluminium surfaces?..??...106 5.5 Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-sections?????????????????.....108 5.6 SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all alloyed layers. Images taken at low magnifications?????????????????.109 5.7 SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all alloyed layers. Images taken at high magnifications?????????...???????.110 5.8 SEM images of cross-section of the worn surface of the laser alloyed surfaces??????????????????????111 5.9 Size distribution curve for the silica sand???????????....112 5.10 Wear response of untreated aluminium and laser alloyed surfaces?.....113 5.11 Wear rate of untreated and laser alloyed aluminium surfaces?.??....114 5.12 Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-sections?????????????????.?115 5.13 SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at low magnifications????.....................................................116 5.14 SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at high magnifications????....................................................117 5.15 SEM images of cross-sections of the worn surfaces of the laser alloyed surfaces????........................................................................118 5.16 Untreated aluminium AA1200 after impact test?.???????.?120 5.17 A typical fracture surface for the laser alloyed samples?.????.?121 5.18 Fractured surface of the aluminium surface showing cuplets.???.?121 LIST OF FIGURES xiv 5.19 SEM fractographs of the laser alloyed samples. (a-c) are alloys with high SiC content (? 20wt%) while (d-f) are alloys with low SiC content (? 20wt%).????????????????????..123 A.1 Optical micrograph of a cross-section of a surface laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC showing SiC particles (black), distributed within the alloyed layer?..????????.....158 A.2 SEM cross-sectional micrograph of a surface laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC showing a near surface region with SiC particles, TiC dendrites, Al3Ni, Al3Ti and Al3Ni2 phases?????????????????????.?..160 A.3 SEM cross-sectional micrograph of a layer alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt% SiC showing a region in the middle of the alloyed layer with SiC particle, TiC, Al3Ti and Al3Ni2 phases??????????????????????...161 A.4 XRD of a surface laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC????????????....????...161 A.5 Optical cross-section micrograph of a surface laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC showing SiC particles (black) randomly distributed in the laser alloyed layer??????..???..162 A.6 SEM cross-sectional micrograph of a layer alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC showing a region near the surface of the alloyed layer????????????????????...163 A.7 XRD of a surface laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC???????????????????...163 A.8 Optical micrographs of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC showing SiC particles (black) in the matrix?...164 A.9 SEM cross-sectional micrograph of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC showing the near surface region?..165 A.10 XRD of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC???????????????????????....166 A.11 Optical cross-section micrograph of a surface layer alloyed with LIST OF FIGURES xv 20wt%Ni + 30wt%Ti + 50wt%SiC showing SiC particles (black) distributed within the alloyed layer??????????.?167 A.12 SEM cross-sectional micrograph of a layer alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC showing a region near the surface of the layer...168 A.13 SEM cross-sectional micrograph of a layer alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC showing a region in the middle of the alloyed layer???????????????????????168 A.14 XRD of a surface laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC????..???????????????????..169 A.15 Optical micrographs of a cross-section of an Al laser surface alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC showing SiC particles (black) within the alloyed layer??????????????.?...170 A.16 SEM cross-sectional micrograph of an Al layer laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC???????????...??...170 A.17 SEM cross-sectional micrograph of a layer alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC???????..???????????...171 A.18 XRD of a surface laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC?????......?????????????????..171 A.19 Optical micrographs of a cross-section of an Al surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC showing SiC particles (black) distributed within the alloyed layer???????...??......172 A.20 SEM cross sectional micrograph of a surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC showing in situ synthesised Al3Ni and Al3Ti intermetallic phases?????????????...173 A.21 SEM cross sectional micrograph of a surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC showing Al3Ni2, Al3Ni and Al3Ti intermetallic phases????????????????.173 A.22 XRD of an Al surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC?????..??????????????????..174 A.23 Optical micrograph of a cross-section of an Al surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC showing SiC particles (black) distributed within the alloyed layer??????...??...?...175 LIST OF FIGURES xvi A.24 SEM cross sectional micrograph of a surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC?????????...????...176 A.25 SEM cross sectional micrograph of a surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%Si??????.????????...176 A.26 XRD of an Al surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC???..????????????????????..177 A.27 Optical micrograph of a cross section of an Al surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC showing SiC particles (black) in the layer?????????...????????............178 A.28 SEM cross sectional micrograph of a surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC showing an Al4C3 phase????..179 A.29 SEM cross sectional micrograph of a surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC?????????????..?179 A.30 XRD of an Al surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC??????????????????????..?..180 A.31 Optical micrograph of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing SiC particles (black) distributed within the matrix????????????????....181 A.32 SEM cross sectional micrograph of an Al surface alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing SiC, TiC, Al3Ni, Al3Ni2 and Al3Ti phases?????????????????????..182 A.33 SEM cross sectional micrograph of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing Al3Ni, Al3Ti, Ti5Si3, SiC and TiC phases?????????.????????..182 A.34 SEM cross sectional micrograph of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing Al3Ti, Al3Ni, Al3Ni2, Ti5Si3 and Ti3SiC2 phases?????????????..?..183 A.35 XRD of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC??????????????????????...?.183 A.36 Optical micrographs of an Al surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing SiC particles (black) within the alloyed layer?????.??????????...184 LIST OF FIGURES xvii A.37 SEM cross sectional micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing SiC, TiC, Al3Ti and Al3Ni phases???????????????????????.185 A.38 SEM cross sectional micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing Al3Ti, Al3Ni, Al3Ni2, Ti5Si3, Ti3SiC2 phases?????????????????..??185 A.39 SEM cross sectional micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing Al3Ti, Ti5Si3 and Al3Ni phases???????????????????????.186 A.40 XRD of an Al surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC??????...?????????????...???..186 A.41 Optical micrograph of an Al surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC showing SiC particles (black) in the alloyed layer????????????????.???????187 A.42 SEM cross sectional micrograph of a surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC??????????????..188 A.43 SEM cross sectional micrograph of a surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC??????????????..188 A.44 XRD of an Al surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC????????????????????????189 A.45 Optical micrograph of a cross section of an Al surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing SiC particles (black) distributed within the matrix??????????????190 A.46 SEM cross sectional micrograph of a surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing SiC, TiC, Al3Ni2 and Ti5Si3 phases??????????????????????.?191 A.47 SEM cross sectional micrograph of a surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing Al3Ni, Al3Ni2 and Ti3SiC2 phases???????????????????...??...191 A.48 XRD of an Al surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC??????????????????????..?..192 A.49 Optical micrographs of an Al surface laser alloyed with LIST OF FIGURES xviii 40wt%Ni, 40wt%Ti and 20wt%SiC showing SiC particles (black) distributed within the matrix??????????????193 A.50 SEM cross sectional micrograph of a surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC at showing Al3Ni2 and Al3Ti intermetallic phases???????????????...??.194 A.51 SEM cross sectional micrograph of a surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC showing SiC, Al3Ti, Ti5Si3 and Ti3SiC2 phases??????????????????????..194 A.52 XRD of an Al surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC????????????????????????195 A.53 An optical micrograph of a cross section of an Al surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC (black) particles distributed within the matrix??????????.???.196 A.54 SEM cross sectional micrograph of a surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC, Al3Ni and Al3Ni2 phases??????????????????????...197 A.55 SEM cross sectional micrograph of a surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC, Al3Ti, TiC, and Ti5Si3 phases??????????????????????.?197 A.56 XRD of an Al surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC, Al3Ti, TiC, and Ti5Si3 phases??????..198 B.1 Graph of wear rate versus time for untreated aluminium surfaces??..200 B.2 Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-sections?????????????...????..201 B.3 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC??.????????????????????.202 B.4 Worn surfaces of Al samples laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..203 B.5 Graph of wear rate versus time for an untreated aluminium and LIST OF FIGURES xix aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC???204 B.6 Worn surfaces of Al samples laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..205 B.7 Graph of wear rate versus time for the an untreated aluminium and aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC?......206 B.8 Worn surfaces of Al samples laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..207 B.9 Graph of wear rate versus time for an untreated Al and specimen laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC???????208 B.10 Worn surfaces of Al samples laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..209 B.11 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC..??210 B.12 Worn surfaces of Al samples laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..211 B.13 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC?.?.212 B.14 Worn surfaces of Al samples laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..213 B.15 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC..?....214 B.16 Worn surfaces of Al samples laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..215 B.17 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC?.?.216 B.18 Worn surfaces of Al samples laser alloyed with 70wt%Ni + LIST OF FIGURES xx 10wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..217 B.19 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC?......218 B.20 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..219 B.21 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC??..220 B.22 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..221 B.23 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC??..222 B.24 Worn surfaces of Al samples laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..223 B.25 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC..?....224 B.26 Worn surfaces of Al samples laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..225 B.27 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC..?....226 B.28 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..227 B.29 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC..?....228 B.30 Worn surfaces of Al samples laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..229 LIST OF FIGURES xxi B.31 Graph of wear rate versus time for an untreated aluminium alloy?...?230 B.32 Worn surfaces of the untreated aluminium samples in (a) and (b)?..?231 B.33 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC??.????????????????????.232 B.34 Worn surfaces of Al samples laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..233 B.35 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC???234 B.36 Worn surfaces of Al samples laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..235 B.37 Graph of wear rate versus time for the an untreated aluminium and aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC?......236 B.38 Worn surfaces of Al samples laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..237 B.39 Graph of wear rate versus time for an untreated Al and specimen laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC???????238 B.40 Worn surfaces of Al samples laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..239 B.41 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC..??240 B.42 Worn surfaces of Al samples laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..241 B.43 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC?.?.242 B.44 Worn surfaces of Al samples laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC in (a) and (b) plan view; (c) and LIST OF FIGURES xxii (d) cross-sections?????????????????????..243 B.45 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC..?....244 B.46 Worn surfaces of Al samples laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..245 B.47 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC?..?246 B.48 Worn surfaces of Al samples laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..247 B.49 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC?......248 B.50 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..249 B.51 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC??..250 B.52 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..251 B.53 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC??..252 B.54 Worn surfaces of Al samples laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..253 B.55 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC..?....254 B.56 Worn surfaces of Al samples laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..255 B.57 Graph of wear rate versus time for an untreated aluminium and LIST OF FIGURES xxiii aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC..?....256 B.58 Worn surfaces of Al samples laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..257 B.59 Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC..?....258 B.60 Worn surfaces of Al samples laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections?????????????????????..259 C.1 Fractured surfaces of samples laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC?????????????????...260 C.2 Fractured surfaces of samples laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC???????????????????...261 C.3 Fractured surfaces of samples laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC?????????????????.........261 C.4 Fractured surfaces of samples laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC?????????????????.........262 C.5 Fractured surfaces of samples laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC?????????????????.........262 C.6 Fractured surfaces of samples laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC?????????????????.........263 C.7 Fractured surfaces of samples laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC?????????????????.........263 C.8 Fractured surfaces of samples laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC???????????????????.264 C.9 Fractured surfaces of samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC?????????????????.........264 C.10 Fractured surfaces of samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC?????????????????.........265 C.11 Fractured surfaces of samples laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC?????????????????.........265 LIST OF FIGURES xxiv C.12 Fractured surfaces of samples laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC?????????????????.........266 C.13 Fractured surfaces of samples laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC?????????????????.........266 C.14 Fractured surfaces of samples laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC?????????????????.........267 INTRODUCTION xxv LIST OF TABLES Table Page 3.1 Laser parameters used during the laser alloying experiments??...??53 3.2 Starting powder mixtures???.....??????...???????..53 3.3 Compositions of powder mixtures???????...???????..54 4.1 Chemical composition of aluminium AA1200????...?????...65 4.2 Vickers hardness of Aluminium AA1200 alloy????.??????65 4.3 Density of aluminium AA1200 specimen????????????..66 4.4 Summary of the powder properties????????????.??...71 4.5 Hardness of the alloyed layers at different scanning speeds?..??..?..78 4.6 Phases observed and hardness for Al samples laser alloyed with Ni+Ti????????????????????????..89 4.7 Phases and hardness for Al surfaces laser alloyed with Ni + SiC?????????????????????????..94 4.8 Phases and hardness for Al samples laser alloyed with Ti and SiC????????????????????????..96 4.9 Phases found in all the Al surfaces laser alloyed with Ni, Ti and SiC?...97 4.10 Phases found in selected Al surfaces laser alloyed with Ni, Ti and SiC....97 5.1 Hardness of the alloyed layers????????????????.104 5.2 Absorbed energies during fracture?????...?????????119 6.1 Summary of the results from the breakdown systems?.??????127 6.2 Values used to calculate the surface temperature?????????131 6.3 Ranking of samples according to hardness, wear and impact???..?140 A.1 Phases observed in each of the laser alloyed layers????????.199 INTRODUCTION 1 1. INTRODUCTION Aluminium is extensively used in industry due to its low density, high strength to weight ratio, high thermal conductivity and good formability [1]. Despite these attractive attributes, its application range is limited by its poor surface properties such as hardness and wear resistance. These limitations can be overcome if the surface properties can be enhanced while retaining the bulk properties. There are several metallurgical processes which may be used to enhance surface properties such as flame spraying, plasma spraying, electroplating, physical vapour deposition and chemical vapour deposition. However these methods are not widely used as they do not offer good metallurgical bonding to the base material [2]. In this project, laser alloying was selected to modify the surface properties of aluminium since previous studies by various authors have shown positive results [1-7]. In this process the alloying materials are deposited as powders into a melt pool generated on the surface of a component by a focused laser beam [3]. The beam is scanned over the component?s surface and the deposited material resolidifies resulting in a good bonding between the substrate and the alloyed layer. This modifies the surface by changing the composition and microstructure without affecting the bulk properties of the material. Process parameters such as laser power, laser beam spot size, laser scanning speed and powder feed rate have to be controlled to achieve the desired metallurgical bonding and alloyed surface properties. Metals and ceramics can be used as alloying materials. Laser alloying with metals results in the formation of intermetallic phases in the alloyed layer and alloying with ceramics results in the formation of metal matrix composites. Intermetallic phases and metal matrix composites that are formed on aluminium surfaces during laser alloying result in improved hardness, high specific strength, high specific stiffness, high electrical and thermal conductivities, low coefficients of thermal expansion and high wear resistance [3-6]. These materials are typically used in the INTRODUCTION 2 automobile, mining and mineral, aerospace and defence sectors. In the automotive sector, typical uses are components such as brake drums, cylinder liners, cylinder blocks, drive shafts etc. In the aerospace sector, structural applications such as helicopter parts (body components, support of the rotor plates, drive shafts, etc), rotor vanes in compressors and aero-engines are the norm. Research on the laser alloying of aluminium alloys with either ceramic or metallic materials has been very active [1-4, 7-9], but limited work has been published on using both materials simultaneously. The published work has been limited to higher aluminium alloys (i.e. 2xxx to 7xxx series) [1-10] while to the author?s knowledge none exists on aluminium AA1200 (commercially pure aluminium). In this work, aluminium AA1200 was laser alloyed with Ni, Ti and SiC of different mixture compositions (refer to Table 3.3 in Chapter 3). The aim was to obtain an aluminium metal matrix composite reinforced with SiC particles and intermetallic phases. The Ni was chosen because Al-Ni intermetallic phases are known to improve the surface hardness of aluminium alloys during laser processing [7]. The SiC and Ti were chosen to form metal matrix composites with aluminium via in- situ reactions. The exposure of SiC to molten Al results in the formation of Al4SiC4 and/or Al4C3 phases [10]. The aluminium carbide phase (Al4C3) is extremely brittle and reacts with water to form aluminium hydroxide. Therefore Ti was added to compete with Al for C resulting in the formation of TiC instead of Al4C3 [2]. The Al-Ti intermetallic phases have good hardness and wear properties [3]. In this project the tribological and mechanical properties of the alloyed materials were studied and compared to those of unalloyed aluminium AA1200. Aluminium AA1200 is approximately 25% cheaper than higher aluminium alloys. If the properties of aluminium AA1200 laser alloyed with Ni, Ti and SiC powders are comparable or superior to those of the aluminium 2xxx and/or 7xxx series then this alloy could be used in industries where higher aluminium alloys are used. The study of this material is therefore of industrial and scientific interest. The novel aspects in this project are: INTRODUCTION 3 head2right Laser alloying aluminium AA1200 with Ni, Ti and SiC simultaneously; head2right Optimizing the laser parameters for the formation of homogeneous and crack-free alloyed surfaces; head2right Studying the in-situ formation of the metal matrix composites and intermetallic compounds; head2right Characterizing the mechanical and tribological properties of the alloyed material. This thesis is structured in the following manner: Chapter 1 explains the significance of the research. Chapter 2 introduces the aluminium alloys and reviews laser alloying as well as published research on mechanical and tribological studies of the materials. Chapter 3 describes the experimental techniques. The results are presented in Chapters 4 and 5 and discussed in Chapter 6. Chapters 7 and 8 provide the conclusions and recommendations, respectively. LITERATURE SURVEY 4 2. LITERATURE REVIEW This review starts by briefly introducing aluminium, then giving a description of laser processing and laser alloying. This is followed by a review of research involving Al, Ni, Ti and SiC. The chapter ends with a review on the fundamentals of wear and fracture. 2.1 Aluminium Aluminium has a face-centred cubic crystal lattice structure [11]. The melting point of pure aluminium is 660?C, a lower value than other commonly used engineering materials such as nickel (1453?C), titanium (1668?C), silicon carbide (2730?C) and steel (1535?C) [11]. The density of aluminium is 2.7g/cm3, which is lower than that of Ni (8.90g/cm3), Ti (4.51g/cm3) and SiC (3.50g/cm3) [12]. Aluminium is classified using a number of different systems. In this thesis, the American Aluminium Association (AA) standard is used. The International Organization of Standardization (ISO) standard whose designations specify the weight percentage of the major alloying elements and the new European Norms (EN) system which is based on the AA and ISO standards mentioned above are referred to in the literature review. The AA list for aluminium is shown in Figure 2.1 [11]. The first digit in the AA designation classifies the alloys according to the major alloying element(s). A distinction is made between wrought and cast alloys. Different products can be made from aluminium alloys by hot- or cold-rolling, homogenization, extrusion, casting (sand casting, gravity die-casting and pressure die-casting) and machining. Figure 2.1: AA list for aluminium alloys [11]. When atoms from another metal replace aluminium atoms in a crystal structure a solid solution is formed. Due to the difference in size between the aluminium atoms and the atoms of the increases the strength of aluminium as dislocation movement is impeded. When the alloying elements in solution are oversaturated, they precipitate as intermetallic compounds. All commercial aluminium all The most important alloying elements used to influence the properties of aluminium are silicon (Si), magnesium (Mg), manganese (Mn), copper (Cu) and Zinc (Zn). Other alloying elements such as bismuth (Bi), b (Cr), lead (Pb), nickel (Ni), titanium (Ti) and zirconium (Zr) can be added in small amounts (<0.1wt% although B, Pb and Cr may comprise up to 0.5wt%) to tailor alloys for special applications by improving their properties such as castability, machinability, heat [11]. LITERATURE SURVEY new metal, the aluminium lattice is strained. This oys contain approximately 0.1-0.4wt% iron (Fe). oron (B), chromium -resistance, corrosion resistance and tensile strength 5 LITERATURE SURVEY 6 Certain properties of aluminium such as the modulus of elasticity (E = 70GPa), density (? = 2.7g/cm3) and thermal expansion coefficient (24x10-6K) are dependent of the alloying elements and the processing chain. Certain properties are very sensitive to the microstructure and composition of the material. These properties are [11]: ? Strength, ductility and formability (volume properties); ? Fatigue resistance and fracture toughness (local properties or crack front properties); ? High temperature resistance and creep resistance (thermo-mechanical properties); ? Corrosion resistance, wear resistance and surface conditioning (surface properties). The alloy composition, shaping process and heat-treatment determines the microstructure and the microstructure determines the above mentioned properties. This work aims to improve the surface properties of aluminium AA1200 by laser alloying with Ni, Ti and SiC powder particles. 2.2 Laser Processing This section is an introduction to the fundamental concepts of laser materials processing. Research on the laser alloying of aluminium with various materials is also reviewed. The word ?laser? is an acronym for light amplification by stimulated emission of radiation [13]. When a photon strikes an atom in the excited state, it stimulates an electron to drop to a lower energy level emitting another photon. The energy of the incoming photon must correspond to the energy difference between the excited level and the lower energy level of the atom. The new photon has the same frequency and phase as the first photon and moves with it in the same direction. When the new photons (the incoming and the produced photons) strike LITERATURE SURVEY 7 other excited state atoms, they stimulate the release of other photons and the light ray is further amplified. Laser light is monochromatic (all photons have the same wavelength), coherent (all photons are in phase with each other) and directional (photons have the same direction and move nearly parallel to each other). The first laser was built by Theodore H. Maiman in 1960 [14]. It consisted of a synthetic ruby crystal rod that was placed inside a helical flash lamp. At each end of the rod were mirrors. As the flash lamp energized the rod, the chromium atoms of the ruby began to glow a deep red colour. The first emitted photons stimulated the emission of further photons and the laser beam was formed. The major components of a laser are the active medium, the resonator and the pumping and cooling mediums [13,14]. The active medium (also known as the gain medium) is the material that emits the laser light and is also the medium in which the laser light is amplified. The active medium can be a gas, solid or liquid. The critical factor is that the materials must emit electromagnetic radiation of a certain wavelength when it is stimulated and drops from the excited state to a lower energy state. The resonator consists of at least two mirrors that reflect the light repeatedly back into the active medium. The resonator determines the direction and propagation of the laser light and ensures sufficient amplification of the laser beam by stimulated emission. Laser pumping feeds energy to the active medium in order to excite it. Every beam source requires a pump source (or energizer), the gain medium with optics and electrical or chemical energy. The greater part of the excited energy is not converted into laser light but heat. The cooling of the active medium ensures that the heat is dissipated to ensure that the active medium and the resonator do not heat up excessively. Other components required are the elements required to supply the beam source with energy and the additional materials such as gas and cooling water. Various types of lasers, which each utilize different active mediums, are used in laser materials processing. In this work an Nd:YAG laser was used due to its characteristic wavelength which has high absorbivity when incident on metallic materials. LITERATURE SURVEY 8 2.2.1 Nd:YAG (Neodymium yttrium aluminium garnet) lasers The Nd:YAG laser is the most prevalent high power, solid state laser used in manufacturing applications today [14]. Figure 2.2 shows a schematic diagram of an Nd:YAG laser [15]. The gain medium has an Nd:YAG crystal and two mirrors (total reflector and output coupler). The output coupler is a partially transmitting mirror through which the output laser beam is emitted. The lasers are optically pumped using a flash lamp or a diode (using light from a diode laser). A flash lamp is an electric glow lamp designed to produce extremely intense, incoherent, full-spectrum white light for very short durations. A laser diode is more commonly used as a pumping source for solid state lasers because of its high pumping efficiency [13] and was used as the pumping source for the current work. The lasing action of an Nd:YAG laser is developed in the neodymium ions (Nd3+). It is based on a four-level system of electron energy changes within the ion. The neodymium doping level in the crystal of an Nd:YAG laser is between 0.5 and 1at%. The yttrium aluminium garnet (YAG) is chosen as a host for the Nd3+ ions because of its thermal, optical and mechanical properties. YAG is a very strong crystal even when distorted by the addition of the slightly larger Nd. It can withstand very high internal stresses produced from the optical or diode pumping and cooling at the outer diameter. Figure 2.2: Schematic diagram of an Nd:YAG laser [15]. LITERATURE SURVEY 9 YAG is optically transparent to both the pump wavelength and the laser wavelength, and it can take a very high-grade optical finish [13,14]. The neodymium ions turn the transparent YAG into a reddish crystal. Light from the arc lamps or diode laser is used to promote the electrons of the neodymium ions to an excited state where the electrons move to a higher laser level, releasing energy into the crystal host in the form of heat. This is followed by a transition back to the lower laser level, where light with the wavelength of 1.064 micrometers is produced. Once in the lower laser state, the electrons quickly decay to the ground state and give off heat. In a neodymium ion, the lower laser level is considerably higher than the ground state. As a result, the lower laser level is not populated when the crystal is in the ground state, making it easy to create a population inversion. Population inversion is the condition in which more atoms or molecules of the gain medium are in the upper laser level than in the lower laser level. In this situation, a photon is more likely to strike an excited-state atom or molecule and cause it to emit laser light than to be absorbed by the atom or molecule. The laser beam is then amplified. If more atoms or molecules were in the lower laser level than the upper laser level, the photons would strike an atom or molecule that is not excited and be absorbed, weakening the laser beam. An Nd:YAG laser uses fiber optic beam delivery. Bulk transmission efficiencies greater than 99% are possible with silica fibers at 1.064nm (the Nd:YAG wavelength). The major advantage of fiber optics over conventional beam delivery components (like mirrors) is the ability to transmit laser beams over long distances (up to 50m) and around curves, because of its ability to flex [14]. Fibers can be easily moved so that the beam can be manipulated about a fixed work- piece. 2.2.2 Laser alloying The laser alloying process involves melting a thin layer of the substrate surface and simultaneously adding the alloying material into the melt pool [3,16]. The LITERATURE SURVEY 10 alloying material can be in the form of powder, wire or paste. The surface re- solidifies rapidly and the composition of the surface layer is thereby modified. The motivation for laser surface alloying is to improve the mechanical, tribological and corrosion properties of a relatively inexpensive soft substrate. In a typical powder feeding system the powder particles are stored in a hopper. The powder is fed by an inert carrier gas from a powder feeder to the substrate via the transporting tube and a nozzle. A focused laser beam is used to generate a melt pool into which the alloying mixture is injected with a flow of air or gas. The alloying powder mixes with the molten base metal by convection and diffusion [14]. Figure 2.3 shows a schematic diagram of the laser alloying process [17]. The powder stream can lead or trail the laser beam. The laser is delivered to the substrate surface through an optical system. The powder to be used for alloying is delivered to the laser-substrate interaction zone via a powder delivery device. Figure 2.3: Schematic diagram of a laser alloying process [17]. A laser beam with a given power intensity distribution (power per unit area of beam cross-section) irradiates the surface of the base material (e.g. aluminium). Majority of the incident radiation is reflected by the base material and is dispersed away from the interaction zone [18]. The energy absorbed by the base material develops a molten pool in that area. Liquid transport occurs within the melt pool which is driven by a combination of Marangoni forces and buoyancy. Marangoni Water cooling Laser beam Aluminium substrate ?d Alloy Lens LITERATURE SURVEY 11 convection is induced by a surface tension gradient along the pool-heating surface, on which the temperature distribution is inhomogeneous, while the buoyancy convection is induced by the temperature gradient inside the melt pool [18,19]. Marangoni convection tends to dominate fluid flow patterns and the solid/liquid interface in the developing and solidifying processes of the melt pool. For a laser beam with a Gaussian energy distribution, the absorption of laser energy occurs mostly at the middle of the melt pool. A temperature gradient is therefore produced between the middle and the edge of the melt pool. The majority of metals display a negative surface tension coefficient and temperature gradients at the free surface induce thermocapillary-driven flow (Marangoni flow) directed from hot to cold [19,20]. This results in convection flow being directed from the middle to the edge of the melt pool increasing the width of the alloyed layer and reducing its depth [21]. Metals and ceramics can be used as alloying materials. Laser alloying with metals results in the formation of intermetallic phases and injecting ceramics result in the formation of metal matrix composites. Intermetallic phases and metal matrix composites formed during laser alloying have resulted in improved hardness and wear resistance of aluminium alloys [3-5,22]. The critical laser processing parameters include laser power, laser beam spot size, laser scanning speed, type and flow rate of shielding and carrier gases, powder particle size and powder feed rate. These parameters must be controlled carefully so that the required thickness, fine grain size and intermetallic compound can be achieved. Some of these parameters are discussed below. Effect of laser power during laser alloying As the melt pool is created on the surface of the substrate, increasing the power provides more energy for melting the surface and the alloying powders. Applying insufficient power results in insufficient melting of the substrate and undissolved powders. An increase in power also increases the depth of penetration [21,23]. When the same amount of alloying powder is used (assuming no loss due to LITERATURE SURVEY 12 evaporation or otherwise during alloying), a lower depth of penetration will result in a higher concentration of the alloying material compared to a higher depth [24]. Effect of laser beam spot size (or beam diameter) during laser alloying Increasing the laser beam diameter decreases the depth of alloying [21,24], because large beam spot diameters have lower intensities compared to smaller ones when the same power is used. When the beam spot size is very large, the powders might not dissolve in the melt pool due to low beam intensity. There is a correlation between the beam spot size and the powder feed rate. If too much powder is fed into a small melt pool, the powder efficiency will be compromised [21]. Powder efficiency is the ratio of the powder injected into the melt pool to the powder leaving the hopper. Effect of laser scanning speed during laser alloying At constant power and powder feed rates, increasing the laser scanning speed decreases the absorption of radiation on the substrate [14,21]. This may result in insufficient heat to melt the substrate and to dissolve the powder particles. A high laser scanning speed leads to a high cooling rate which could result in a refined microstructure. A high laser scanning speed is preferred for material processing as it increases the processing time. Effect of shielding gas during laser alloying An inert shielding gas (e.g. argon) is used to prevent oxidation of the powders and the substrate during the alloying process [3,7]. Effect of powder particle size during laser alloying Powder particle size plays an important role during laser alloying. Small particles coagulate within the powder delivery nozzle, due to their large surface area, LITERATURE SURVEY 13 resulting in nozzle blockage [25]. The carrier gas, which is applied for powder delivery, builds up pressure and forces the powder out of the nozzle. The blocking and unblocking action of the nozzle results in pulsing of the powder. This causes poor powder efficiency and an inhomogeneous alloying layer. Large powder particles produce poor powder focus as the powder spot size (area the powder is focused to) is greater than the laser spot size. This also leads to poor deposition efficiency. Powder particle size also affects the melting temperature of the particles which have an effect on the final microstructure e.g. larger particles of metals may not form intermetallics, and small ceramics may dissolve in the melt pool due. Effect of powder feed rate during laser alloying The proportion of the incident radiation absorbed by the powder stream increases with increasing powder flow rate [26]. If too much powder is fed some powder will not dissolve in the melt pool. When the powder feed rate is too low, there will not be sufficient powder particles to cover the alloyed surface leading to the formation of inhomogeneous layers. 2.2.3 Advantages of laser alloying Advantages of laser alloying include high deposition rates, low thermal distortion, limited metallurgical degradation of the base materials, refined microstructures due to high re-solidification rates and a variety of materials can be alloyed onto different substrates [1,26-28]. 2.2.4 Disadvantages of laser alloying Disadvantages of laser alloying include the high cost of the laser source and components, low efficiency of laser sources and difficulties in selectively producing one intermetallic phase since equilibrium conditions are not reached [1,27-29]. LITERATURE SURVEY 14 2.3 Laser alloying with Al, Ni, Ti and SiC Laser alloying with metallic materials may result in the formation of intermetallic phases. An intermetallic compound is a solid phase consisting of two or more metallic elements in definite proportions. These compounds ideally have superior properties (hardness and wear resistance) compared to the original metallic elements but have poor ductility. 2.3.1 Laser alloying with Ni or Ni containing compounds The Al-Ni phase diagram is shown in Figure 2.4 [30-32]. The phase diagram contains five intermetallic compounds (Al3Ni, Al3Ni2, Al3Ni5, AlNi and AlNi3). The first phase formed on the Al-rich side of the phase diagram is the Al3Ni (DO11) intermetallic phase which has an orthorhombic crystal structure [33-36]. This phase forms a eutectic with aluminium. The Al3Ni phase is a peritectic product of a reaction between liquid Al and the Al3Ni2 phase under equilibrium conditions. Jain and Gupta [37] showed that the Al3Ni phase also crystallizes from the liquid during cooling and appears as diamond-shaped particles enveloping the Al3Ni2 phase. The ?-Al3Ni2 (D513) phase is hexagonal and forms as a result of a peritectic reaction between AlNi and the liquid Al. The ?-AlNi phase has an ordered B2 structure and remains ordered up to the melting point of 1680?C. The Ni-rich side of the phase diagram has ?-Ni having an FCC structure containing aluminium as a solute in the Ni solid solution. The Ni3Al phase forms by a peritectic reaction between ?-Ni and liquid Al under equilibrium conditions. The Ni3Al phase has L12 ordering of Al and Ni atoms in the FCC structure. At low temperature a peritectoid reaction occurs between AlNi and Ni3Al leading to the formation of the orthorhombic Al3Ni5 phase. LITERATURE SURVEY 15 Figure 2.4: Al-Ni phase diagram [30]. Nickel aluminides are known for exhibiting high strength and corrosion/oxidation resistance at elevated temperatures [33]. Bysakh et al. [33] laser alloyed nickel with aluminium powder. The laser beam created a molten pool on the nickel substrate in which the aluminium powder was added as an alloying element. Argon was used as a shielding gas to prevent oxidation during laser alloying. The laser parameters used were 2.7kW of laser power and powder feed rates of 3.2 and 6.4g/min. The laser scanning speed was varied between 4.2 and 5.1mm/s to determine its effect on the microstructures formed. The authors observed that the depth of alloying decreased with increasing scanning speed. This is due to the high temperatures achieved at low scanning speeds. The alloyed layer became increasingly aluminium-rich as the scanning speed decreased due to the higher melting rates of the aluminium powder at lower scanning speeds. To achieve the same aluminium composition at a higher feed rate, a higher laser scanning speed would be required as shown in Figure 2.5. The phases observed in the microstructures were Al, Ni and Al3Ni at low scan rates. The Al3Ni intermetallic phase had a dendritic microstructure and was formed near the top of the alloyed layer. The interdendritic region consisted of ?-Al and ?-Al/Al3Ni eutectic phases. LITERATURE SURVEY 16 At higher scanning speeds, AlNi, AlNi2 and AlNi3 intermetallic phases were formed. Figure 2.5: Composition plot of aluminium concentration versus laser scan speed at two different feed rates [33]. Laser alloying of aluminium AA1100 with electrodeposited nickel was performed by Selvan et al. [38] using a CO2 laser. Alloying was performed with a beam diameter of 1mm for various scanning speeds and laser powers. The aim of the study was to improve the surface hardness of aluminium by forming Al-Ni intermetallic phases in the alloyed layer. The intermetallic phases formed were Al3Ni and Al3Ni2. These phases were also observed by other authors [30,39]. The hardness of the alloyed layers was in the range of 600-950 HV0.1 compared to 35HV for unalloyed aluminium AA 1100. Man et al [40] laser alloyed aluminium AA6061 with a NiCrSiB powder paste. NiCrSiB was chosen because it has high hardness, good wear resistance and good cavitation resistance. The laser processing was performed with a continuous wave Nd:YAG laser and the conditions were 1.5kW of laser power , a beam diameter of 2.5mm and a scanning speed of 5mm/s. Argon was used as the shielding gas at a LITERATURE SURVEY 17 flow rate of 20L/min. The laser tracks were overlapped by 50% track-width (i.e. a pass overlap of 50%). The results showed that Al3Ni and Al3Ni2 intermetallic phases were formed at the surface of the alloyed layer, Al3Ni2 and AlNi were formed in the middle of the alloyed layer while Al3Ni and ?-Al were formed in the region close to the substrate. The Ni content decreased from the region in the middle of the alloyed layer towards the substrate. There was a high Al content in the region near the surface of the alloyed layer due to the difference in densities of Al (? = 2.7g/cm3) and Ni (? = 8.91g/cm3). A hardness of approximately 900HV was achieved in the alloyed layer. The electrochemical corrosion and cavitation erosion resistance were significantly improved by laser processing. The improved corrosion resistance was attributed to the presence of Ni, Al-Ni intermetallic phases and the Ni- and Cr-rich aluminium solid solution in the alloyed layer. Das et al. [41] laser alloyed Al with Ni and studied the creep behaviour of the laser alloyed material. Creep is the tendency of a solid material to deform permanently under the influence of stress and occurs as a result of long term exposure to high levels of stress that are below the yield strength of the material. The phases observed in the alloyed layer were Al3Ni dendrites and a eutectic of ?- Al and Al3Ni. The creep resistance of aluminium was improved by laser alloying due to the formation of the Al3Ni phase which is non-deformable by creep. Ravi et al. [7] laser alloyed an Al-12wt%Si alloy with preplaced Ni and Cr powders and studied the microstructures and hardness of the alloyed layers. Chromium was added to improve the corrosion resistance of the alloy. The alloying was performed with a 3kW continuous wave CO2 laser. The processing parameters were a scanning speed of 0.5m/min, a beam diameter of 1mm and the laser power was varied between 1 and 2 kW. Argon was used as shielding gas. The authors observed that the hardness of the laser alloyed surface increased with decreasing Ni and a maximum of 490HV was achieved with the composition of 20wt%Ni-80wt%Cr. The microstructure showed needle-like structures for alloys with a high Ni composition which disappeared as the Cr content increased. The microstructure underwent a phase transformation when the Cr concentration was LITERATURE SURVEY 18 increased. The intermetallic phases formed in the microstructures were AlNi, Al3Ni2, AlNi3, AlCr2, Cr9Al17 and Cr3Ni2. 2.3.2 Laser alloying with Ti or Ti containing compounds The Ti-Al phase diagram is shown in Figure 2.6 [42,43]. The binary phase diagram consists of intermetallic phases and terminal solid solutions. The intermetallic phases are a hexagonal ?2-Ti3Al (DO19), tetragonal ?-TiAl (L10), tetragonal TiAl2, ? and tetragonal TiAl3 (DO22) [42-45]. The terminal solid solutions are an ?-Ti (A3), ?-Ti (A2) and fcc ?-Al (A1) [42,46,47]. Ohnuma et al. [44] reported the tetragonal Al5Ti and the bcc ?2-TiAl (B2) phases. The tetragonal Al5Ti and Al11Ti5 phases were also reported by Illekov? et al. [43] on the Al-rich side of the Al-Ti system near the Al3Ti phase but were later ruled out by Raghavan [36]. On the Al-rich side of the phase diagram, the Al3Ti phase forms when the Ti content exceeds 2at%. Based on thermodynamic assumptions, the Al3Ti phase forms preferentially during the reaction of aluminium and titanium [48]. Titanium aluminides have high oxidation and wear resistance [49]. The Al3Ti phase has a low density (3.3g/cm3) and a high melting point (1340?C) but is also brittle [50]. LITERATURE SURVEY 19 Figure 2.6: Ti-Al phase diagram [42]. Wendt et al. [51] laser alloyed aluminium with a titanium wire using CO2 and Nd:YAG lasers. The authors reported that the depth of alloying (or laser penetration depth) was greater when alloying with a Nd:YAG laser compared to a CO2 laser. The microstructure of the alloyed layer consisted of a Ti-supersaturated Al matrix and TiAl3 intermetallic phase. When the substrate was an Al-Si alloy, the intermetallic phase formed was Ti(Al,Si)3. 2.3.3 Laser alloying with Ni and Ti The Ni-Ti phase diagram is shown in Figure 2.7 [52]. The phase diagram shows the presence of the hexagonal ?-Ni3Ti (DO24), cubic NiTi (B2) and cubic NiTi2 phases [36,53,54]. LITERATURE SURVEY 20 Figure 2.7: Ni-Ti phase diagram [52]. The ternary phases observed in the Al-Ni-Ti phase diagram (Figure 2.8) are the cubic ?1-Al13Ni2Ti5, ?2-Al2NiTi and ?4-AlNi2Ti phases, and the hexagonal ?3- Al3NiTi phase [36,54,55]. A decagonal ?5 was also observed on the Al-rich side at the composition Al65Ni20Ti15 by various authors [36,53,56,57]. The ?3 and ?4 melt congruently at 1289 and 1500?C respectively, while ?1, ?2 and ?5 melt incongruently at 1347, 1255 and 1107?C respectively. The Al-Ni-Ti partial isothermal section at 1050?C is shown in Figure 2.8 [57]. LITERATURE SURVEY 21 Figure 2.8: The Al-Ni-Ti partial isothermal section at 1050?C [57]. Man et al. [3] used a continuous wave Nd:YAG laser to alloy aluminium AA6061 with preplaced NiTi (54 wt% Ni & 46 wt% Ti) powder to improve its hardness and wear resistance. The processing parameters were a laser power of 1.5kW, a beam diameter of 2.5mm and a scanning speed of 10mm/s. Argon was used as the shielding gas at a flow rate of 20L/min. A laser alloyed surface, free of cracks and pores was achieved and SEM micrographs showed a dendritic microstructure. XRD patterns confirmed the intermetallics formed as TiAl3 and Ni3Al. The interdendritic film was composed of ?-Al. The wear tests were performed on a pin-on-disc setup. A hardness increase of 200HV and wear resistance of about 5.5 times that of the virgin substrate was achieved for the modified layer. The increase in hardness and wear resistance was attributed to the formation of the TiAl3, Ni3Al and ?-Al phases. The intermetallic phases (TiAl3 and Ni3Al) resisted abrasion while the softer ?-Al phase suppressed crack growth. Ternary intermetallic phases were not reported by the authors. Mabhali et al [58] observed Al3Ni, Al3Ni2, Al3Ti and NiTi intermetallic phases when laser alloying aluminium LITERATURE SURVEY 22 AA1200 with Ni and Ti powders of different compositions. A hardness increase of up to 808HV was achieved after alloying with 80wt%Ni and 20wt%Ti. The authors reported that the hardness and wear resistance increased as the Ni content in the alloying powder increased. Gunn?s et al [50] investigated the microstructure development in laser-processed Al-Ti-Ni alloys with 25at.%Ti. Laser processing was performed with a CO2 laser and the power was varied between 5 and 6kW. The authors observed Al-rich dendrites with Ni-rich interdendritic phases. The dendritic phase of the alloy with 5at.%Ni contained the Al3Ti phase while alloys with higher Ni contents (8- 15at.%Ni) had Al67Ni8Ti25. Al3Ti has a low density (3.3g/cm3), high melting point (1340?C) and good corrosion resistance, but is a very brittle phase. The interdendritic phases consisted mainly of Al3Ni for the alloy with 5at.%Ni while the higher Ni containing alloys consisted of Al2NiTi and AlNi or Al3NiTi2. Biswas and Varin [59,60] also observed the Al2NiTi phase during induction melting of Al, Ni and Ti metals followed by homogenization at 1000?C. 2.3.4 Laser alloying with SiC or SiC containing compounds The previous section explored the literature regarding laser alloying with metallic materials which result in improvement of surface properties due to the formation of intermetallic phases. Surface properties can also be improved by incorporating (or injecting) ceramic particles (such as SiC, Al3O2, TiC and WC) into the aluminium melt pool [2,8-10]. An aluminium matrix composite is formed with aluminium as the matrix and the ceramic as reinforcement. The matrix could also be products from the dissociation of the ceramics and reactions of those products with aluminium. Since SiC was chosen as a ceramic for this project, this section will look at work done on processing aluminium with SiC particles. Some literature work on TiC will also be considered in Section 2.3.5. In the Al-Si phase diagram a eutectic ?-Al + Si is formed below 577?C and there are no intermetallic phases formed between Al and Si [61,62]. In the Si-C binary LITERATURE SURVEY 23 system, two phases are observed. These phases are the stable cubic ?-SiC and the metastable hexagonal ?-SiC [63]. The only intermetallic phase observed in the Al- C system is a rhombohydral Al4C3 phase [64,65]. The Al-Si-C phase diagram is shown in Figure 2.9 [64]. The ternary phases observed are Al4SiC4 (Al4C3.SiC), and the more Al-rich hexagonal Al8SiC7 (2Al4C3.SiC) which forms peritectically between 2273 and 2373K [64-68]. The hexagonal Al4Si2C5 (Al4C3.2SiC) phase which is stable between 2173 and 2243K, was not observed by Raghavan [64]. These phases together with a hexagonal Al4Si3C6 (Al4C3.3SiC) were however observed by Hu and Baker [10]. Figure 2.9: Al-Si-C phase diagram [64]. The main problem in the Al-Si-C system is the formation of the needle-like Al4C3 phase. This phase is extremely brittle and lowers fatigue resistance and thermal LITERATURE SURVEY 24 stability of the Al-Si-C system [69]. It also causes corrosion sensitivity due to its high water reactivity according to the following reaction [69,70]: Al4C3 + 12H2O ? 4Al (OH)3 + 3CH4 (Reaction 2.1) Various researchers have studied the phases formed when aluminium was laser alloyed with SiC [10,69,71-75]. The common observation was that the phases are temperature dependent. When the alloying temperatures are between 940 ? 1620K, the brittle Al4C3 phase is formed (reaction 2.2). At temperatures above or equal to 1670K, the Al4SiC4 phase is formed (reaction 2.3). 4Al + 3SiC ? Al4C3 + 3Si (Reaction 2.2) 4Al + 4SiC ? Al4SiC4 + 3Si (Reaction 2.3) Work has been conducted to suppress the formation of the Al4C3 phase during laser processing. Su and Lei [71] laser cladded Al-12wt%Si with a powder containing SiC and Al-12wt%Si in a 3:1 volume ratio. A CO2 laser was used with 2-4kW laser power, 2-10mm/s laser scanning speed and a 3mm diameter laser beam. The aim of the study was to form a surface MMC layer on the aluminium matrix. It was reported that the laser melting of SiC particles onto an aluminium substrate produces aluminium carbides. As stated above, the presence of the Al4C3 phase in MMC is not desirable as it is brittle and hydroscopic. The addition of Al- 12wt%Si was found to suppress or eliminate the aluminium carbides in the MMC layer. A good distribution of injected SiC particles was achieved near the surface. The microhardness of the coating was between 220 and 280 HV. Hu and Baker [10] formed metal matrix composite layers on aluminium AA6061 alloy surfaces by incorporating SiC particles using a CO2 laser at different energy densities. Energy density is given as: E = q/(vd) (Equation 2.1) LITERATURE SURVEY 25 Where q is the laser power, v is the laser scanning velocity and d is the diameter of the laser beam. Large surface areas of MMC layers were produced by overlapping single laser tracks by Hu and Baker [10]. The thickness of the MMC layer was dependent on the laser energy density. The microstructure and phases in the layers were also strongly dependent on the energy density. Al4C3 needles, Al4SiC4 platelets and free Si phases were formed at nominal energy densities of 100-200MJ/m2. These phases are shown in Figure 2.10. The formation of Al4C3 needles in the MMC layer could be reduced by applying high energy densities of ~560 MJ/m2. The laser energy density is directly related to the surface temperature generated by the laser beam. Figure 2.10: MMC layer showing Al4SiC4, Al4C3, Si, and SiC phases [10]. Anandkumar et al [72] conducted a detailed study of the microstructure of the aluminium MMC layers produced by a 2 kW Nd:YAG laser. The main processing parameter found to have a significant influence on the microstructure and properties of the coating was energy density. At low energy densities (~26 MJ/m2) the SiC particles were retained and dispersed in the coating. At high specific LITERATURE SURVEY 26 energies (~ 58 MJ/m2) the SiC dissolved and reacted with the molten aluminium. Analysis of the microstructure revealed small amounts of SiC dispersed near the surface along with phases such as Al4SiC and eutectic Al-Si. The undesirable Al4C3 phase was completely suppressed. Zhang and Chen [76] studied the influence of SiC particulates on the grain structure development of an aluminium AA7075 alloy during laser rapid solidification. A CO2 laser was used and the processing conditions were 600W of power, a 30mm/s scanning speed and a beam diameter of 3mm. Epitaxial columnar grains were observed for solidified laser melted alloys without SiC particles. The epitaxial grain growth of columnar grains was also observed by Abboud and West [46] while alloying titanium with aluminium powder. The addition of SiC particles resulted in a random growth of the primary Al, i.e. the arrangement in crystal growth changed from regular to random. The random growth was caused by the distortion of the subgrain structure. The SiC particles act as a barrier to the solute and heat transfer and the disturbed solute and temperature fields may give rise to unsteady solidification. Once the branched dendritic structure is formed, irregular growth may occur in any given columnar grain resulting in non-epitaxial grain development. This effect was observed with high volume fraction of particles (10vol%SiC) of 10?m in size. When the volume fraction of SiC particles was low (2vol%) and the particle size fine (4?m), the loss of the epitaxial relationship with unmelted substrate is ascribed to a particle restricted mechanism. The SiC particle ahead of the solidification interface creates a diffusion barrier to crystal growth. This restricted growth allows sufficient time for new grains to develop. These findings show that the particle volume and size play a role in the growth and development of microstructure. 2.3.5 Laser alloying with TiC TiC has a cubic crystal structure [77,78] and exhibits a very high melting point (1668?C), thermal stability, hardness and wear resistance [2]. Katipelli et al. [2] laser deposited TiC on an Al AA6061 alloy using a 2kW continuous wave Rofin LITERATURE SURVEY 27 Sinar Nd:YAG laser. 10wt%Si was added to TiC to increase wettability and fluidity of the Al due to the formation of Ti-Si intermetallic phases. Wettability can be defined as the ability of a liquid to spread on a solid surface, and represents the extent of intimate contact between a liquid and a solid [79]. The processing parameters were a laser power of 1.8kW, a scanning speed of 20mm/s and a beam diameter of 3.5mm. The phases observed after laser processing were Al, Si, TiC, SiC, TiSi2 and Al3Ti. The matrix consisted of aluminium, titanium and some silicon. The Ti in the matrix was due to the dissociation of TiC particles at high temperatures. The C released, reacted with Si to form the SiC observed. The Ti (together with the Si) in the matrix enhanced the bonding between TiC particles and the matrix. An improved hardness of approximately 7 times that of the substrate was achieved after laser processing. The wear rate was also significantly improved. Wu [80] laser cladded a 5CrMnMo steel with a coating consisting of Ni, Ti and graphite powder using a continuous wave CO2 laser. The phases produced in situ during laser cladding were ?-Ni, TiC, Cr23(C,B)6, Ni5Si2 and Cr2B. The Ni-Ti intermetallic phases were not observed. The volume fraction of TiC particles was high near the surface of the cladded layer and gradually decreased towards the cladded layer/substrate interface. The difference in microstructure and composition in different regions of the cladded layer (from the surface towards the substrate) results in the formation of a functionally graded coating (FGC). The hardness of the cladded layer was 1250HV near the surface and gradually decreased towards the base. The wear properties of the coating were also significantly enhanced. 2.3.6 Laser alloying with Ni and SiC or TiC The previous sections reviewed laser alloying with metallic materials and with ceramic materials in isolation. Due to the advantages achieved with both techniques, incorporating metals and ceramic simultaneously would further LITERATURE SURVEY 28 improve the surface properties. Work has been done to look at this scenario and the following sections review such work. Le?n and Drew [81] investigated the influence of nickel on the wettability of silicon carbide by aluminium. The problem encountered in fabrication of composites is the rejection of the ceramic phase by the liquid metal due to their poor wettability [79,81]. To improve wettability of SiC by liquid aluminium, a nickel coating was applied on the SiC which increased the overall surface energy of the solid, promoting wetting by the liquid aluminium [81-83]. Aluminium is highly reactive with silicon carbide and forms aluminium carbide (Al4C3) that results in reinforcement degradation and a reduction in the strength of the composite. The added nickel improved the wettability of SiC by liquid aluminium and the Al/SiC adhesion which led to the suppression of aluminium carbide formation. The Al3Ni and Al3Ni2 intermetallic phases were formed from the reaction of aluminium with nickel. The exothermic nature of the Ni-Al interaction together with the precipitation of the Al3Ni and Al3Ni2 intermetallic phases were the reported factors leading to the improvement of the wettability of SiC by aluminium. Selvan et al. [84] laser alloyed titanium with SiC and 50wt%Ni + 50wt%SiC and studied the microstructure and hardness of the alloyed layers. Laser alloying was performed with a CO2 laser and the processing parameters were laser powers of 1 and 5kW, a beam diameter of 1mm and scanning speeds of 0.5-1.5m/min. The shielding gas used was argon. The results showed the formation of dendrites when titanium was laser alloyed with SiC and the phases observed were TiC, TiSi, Ti5Si4 and Ti5Si3. Uniformly distributed Ni precipitates were observed when alloying with 50wt%Ni + 50wt%SiC. At low scan speeds and high laser power, the surface temperature was very high (approx. 2800?C). Due to these high temperatures, Ni dissolved completely and reacted with SiC and the molten titanium to form complete dendritic structures. The surface temperature was calculated from the following equation [14,84,85]: LITERATURE SURVEY 29 T(D,l) = (2F0/k) ? (? t/?) (Equation 2.2) Where F0 = P?/A, k ? thermal conductivity, ? ? thermal diffusivity, t ? interaction time, P ? input laser power, ? ? absorption coefficient and A ? area of laser irradiation. The phases formed after laser alloying with 50wt%Ni + 50wt%SiC were TiSi, Ti5Si3, TiNiSi and NiTi2. The NiTi2 phase improves ductility and fatigue resistance while TiSi, Ti5Si3 and TiNiSi are potential strengtheners which can improve oxidation, corrosion and wear resistance [84]. The surface hardness achieved was 900-1200HV when alloying with SiC and 600-1000HV when alloying with 50wt%Ni + 50wt%SiC. Fujimura and Tanaka [86] studied the in situ reaction at the Ni/SiC interface with a high temperature x-ray diffractometer. The authors reported that ?-Ni2Si and ?- Ni2Si was formed at the Ni/SiC interface. ?-Ni2Si is a high temperature (1400K) phase of ?-Ni2Si. The reaction for the formation of the Ni2Si phase is as follows: 2Ni + SiC ? ?-Ni2Si + C (Reaction 2.4) At high temperature, ?-Ni2Si ? ?-Ni2Si (Reaction 2.5) Viswanathan et al. [87] laser alloyed a Al-Si alloy with Ni and TiC powders of different composition using a 5kW continuous wave CO2 laser. A good MMC reinforced with TiC particles was formed when alloying with 75wt%TiC and 25wt%Ni. The phases formed in the alloyed layer were TiC, Al-Si and Al-Ni intermetallics. The Al-Ni intermetallic phases formed during rapid resolidification strengthened the interface between the TiC and the matrix. Al-Si also has good wetting with TiC which further strengthened the adherence between the metal/ceramic interface as well as the layer and the substrate. TiC did not dissociate to form Ti and C due to its high melting point (3160?C) and that prevented the formation of the Al4C3 phase as there was no free C. The hardness of the alloyed layer was 750HV which is approximately 9 times the hardness of the substrate (80HV). LITERATURE SURVEY 30 2.3.7 Laser alloying with Ti and SiC Kloosterman et al. [88] laser injected SiC particles into a Ti-6Al-4V alloy. The laser processing was performed with a Nd:YAG laser and the parameters used were a laser power of 1kW, a beam diameter of 2-3mm and scanning speeds of 8- 17mm/sec. Argon was used as the shielding gas. At high temperatures, SiC decomposed and the C reacted with Ti to form TiC dendrites shown in Figure 2.11. Figure 2.11: Micrograph showing TiC dendrites and Ti + Ti5Si3 eutectic phase [88]. These TiC dendrites were randomly orientated and distributed over the alloyed track. TiC was also observed around the SiC particles as either a cellular reaction layer or an irregular reaction layer. These reaction layers were also observed by Pei et al. [89]. The cellular layer was relatively thin and regularly shaped around the SiC particles. Spherical TiC grains were formed in the irregular layer with a ternary Ti3SiC2 phase found as small plates around the randomly orientated spherical TiC grains. This Ti3SiC2 phase was also confirmed by other authors TiC Ti + Ti5Si3 LITERATURE SURVEY 31 [65,68,90,91].The microstructure of the matrix consisted of ?-Ti and Ti5Si3 eutectic phases. Ti5Si3 is the more thermodynamically stable phase in the Si-Ti phase diagram [92,93]. Other titanium silicides (Ti3Si and Ti5Si4) are unstable especially in the presence of carbon and play a minor role in the Al-C-Si-Ti system [68]. The hardness of the alloyed layer was between 650 and 1100HV. Li et al. [94] reported the formation of TiC and Ti5Si3 when Ti reacts with SiC and when the temperature is between 1173 and 1373K, while Ti3SiC2 phase is formed above 1473K. Pleshakov et al. [95] also laser injected SiC particles into a Ti-6Al-4V melt pool. Phases observed in the modified layer were ?-Ti, TiC, SiC, Ti5Si3, TiSi2. The Ti3SiC2 phase was not identified since its reflections coincided with the TiC and TiSi2 in the investigated 2 theta range of 25-85?. Laser modification improved the wear resistance of the Ti-6Al-4V alloy with the specific mass loss improved by almost 180 times. Selvan et al. [96] laser alloyed a Ti6Al4V alloy with SiC particles using a CO2 laser. The phases observed in the alloyed layer were ?-Ti, SiC, TiC, TiSi and Ti5Si3. The TiSi2 and Ti3SiC2 phases were not observed. The dendrites of TiC were found to be embedded in the ?-Ti matrix. The dendrite-like population was high at the top surface and decreased towards the base. This resulted in a hardness gradient with a high hardness (650-800HV) at the top and a lower hardness (469-350HV) towards the substrate. The substrate hardness was 300HV. Man et al. [97] synthesized TiC in situ on an AA6061 aluminium surface by alloying with SiC and Ti powders. An Nd:YAG laser was used during the alloying at 1 to 1.5kW laser power. The processing speed was between 5 and 25mm/s and the track overlap was 50%. The optimum powder composition for a high quality surface metal matrix composite was achieved with 40wt% SiC and 60wt%Ti. The average size of the TiC particles that formed after alloying was less than 3?m and these were uniformly dispersed in the matrix. XRD analysis of the alloyed layer revealed TiC, TiAl, Ti3Al, SiC, Al and Si phases. The hardness increased from LITERATURE SURVEY 32 75HV to 650HV due to the formation of the TiC particles and TiAl and Ti3Al intermetallics. 2.4 Wear Wear can be classified into two groups, mechanical wear (e.g. mechanical interactions between two surfaces) and chemical wear (e.g. tribochemical reaction between counterfaces) [98,99]. Mechanical wear is mainly subdivided into abrasion, adhesion, surface fatigue, erosion and sliding wear [98]. This work deals with abrasive wear of composite materials. Abrasive wear refers to the removal or displacement of material from a surface by hard particles, or hard protuberances on a countersurface, forced against and sliding along a surface [98-100]. The amount of material removed depends on the microstructure and properties of the material, the abrasives and the operating conditions of the system [101]. There are two types of abrasive wear, namely two-body and three-body abrasive wear. Two-body abrasive wear refers to the wear which is caused by two bodies sliding against each other or by protuberances on the countersurface while three- body abrasive wear refers to the wear which is caused by hard particles that are free to roll and slide between two, perhaps dissimilar, sliding surfaces [98]. Figure 2.12 shows a schematic diagram of these types of wear. LITERATURE SURVEY 33 Figure 2.12: Schematic diagram illustrating two-body and three-body abrasive wear [102]. 2.4.1 Factors affecting abrasive wear rates Wear is influenced by the properties of the material (type of material, volume fraction of reinforcements, size and distribution of the reinforcements, interfacial bonding between the matrix and the reinforcements, hardness, etc), properties of the abrasives (type of abrasives, hardness, size and shape), system properties (applied load, sliding distance, etc) and the environmental properties (temperature, humidity, etc) [98-105]. The interaction between the material and the abrasives is also crucial especially in unlubricated wear as the temperature will rise due to friction [100]. In this work, the focus will be on the effect of material and abrasive properties. MMC material properties Increasing the volume fraction of reinforcement particles improves the wear resistance and hardness of the MMC [100,103-108]. The bond between the matrix and the reinforcements is also important. This bond is formed by a reaction between the ceramic particles and the metal matrix at the metal/ceramic interface. High wear rates are achieved when a strong bond is formed. Large reinforcement LITERATURE SURVEY 34 particles result in a lower volume fraction of embedded particles in the matrix thus reducing the wear resistance [103]. Ax?n and Zum Gahr [103] studied the abrasive wear resistance of steel substrates laser cladded with TiC particles. They concluded that abrasive wear resistance was influenced by the size of the reinforcement particles, matrix hardness and the abrasive. The mean free path (defined as the average spacing between reinforcement particles) also plays an important role during wear. Matrix protection is improved by a reduced mean free path between the reinforcing particles. Large reinforcement particles result in an increased mean free path which ultimately results in reduced wear resistance [100,103]. The effect of the size of the reinforcements depends on the matrix structure. Small reinforcement particles are of greater advantage within a hard matrix than large particles within a soft matrix [103]. Applied load The abrasive wear rate is generally proportional to the applied load [104,109] as shown in Figure 2.13. Larsen-Basse et al [109] found that the depth of penetration of the abrasive grits into the specimen increases as the applied load increased due to an increase in the average load per abrasive grain. This results in increased wear rates and should occur when the abrasive hardness is greater than that of the material and the abrasives do not fracture. Figure 2.13: Graph of volumetric metal removal rate versus applied load [109]. Abrasive properties Properties of the abrasives such as type of abrasive, particle size and shape play an important role during wear. Wear rate increase with increasing abrasive particle size up to a critical value and then it remains constant [103,109,110]. This relations The critical value is influenced by the hardness difference between the abrasive particles and the wearing material [110]. LITERATURE SURVEY hip is shown in Figure 2.14. 35 LITERATURE SURVEY 36 Figure 2.14: Graph of removal rate versus grit diameter where D0 is the minimum grit size which will remove a significant amount of material [109]. The wear rate is influenced by the shape of the abrasive particles due to its effect on the depth and cross-section of grooves formed and therefore the amount of material removed [98,101]. Angular or sharp abrasive particles give greater wear rates compared to rounded particles [98-101]. If the angle of the front face of the abrasive is sharp then the material it moves across is chipped, but for blunt abrasives the material will be pushed to the sides of the groove [101]. Figure 2.15 shows round and angular silica abrasives [98]. Figure 2.15: (a) Round and (b) angular silica abrasives [98]. (a) (b) LITERATURE SURVEY 37 The hardness of the abrasive material is important in wear as the ratio of the hardness of abrasive relative to the hardness of the material being worn (Ha/Hm) is related to the resultant wear mechanisms [98-100]. Abrasive wear can be classified as high or low wear depending on this ratio. High abrasive wear occurs when the ratio of the hardness of the abrasive particles to the hardness of the wearing material is greater than 1.2 (i.e. Ha/Hm > 1.2) [98]. The penetration depth of the abrasive increases and the material is removed or fractured during wear [100]. This results in deep groove formation and cracking of the material. Conditions that favour high abrasive wear are high applied loads and large abrasive particle sizes. Low abrasive wear occurs when the ratio of the hardness of the abrasive particles to the hardness of the wearing material is less than 1.2 (i.e. Ha/Hm < 1.2). The wear mechanism is characterized by plastic smearing of the material [100]. Conditions that favour low abrasive wear are low applied loads and small abrasive particle sizes. As wear progresses, abrasives are also worn and fractured thereby reducing their cutting abilities. Sin et al. [110] investigated the deterioration of abrasive grits during abrasive wear using a pin-on disk set-up with SiC abrasive papers of different grit sizes and applied loads of 4.9N and 39.2N. The specimens followed a spiral track and always passed over fresh abrasives. The materials tested were polymethylmethacrylate (PMMA), commercially pure nickel, AISI 1095 steel and OFHC copper. The authors found that the reduced cutting ability of the abrasives as wear progressed resulted in a reduction in the wear rates and that the wear debris clogged the interstices between abrasive grits causing a further reduction in the wear rate. This was also observed by Mabhali [111]. Depending on the ratio of the groove width to the radius of the spherical tip of the SiC grit particle, abrasive particles would either plastically deform or cut the surface of the material. As the abrasive particles become duller or the particle size decreases, the wear mechanisms exhibit a transition from cutting to delamination wear. LITERATURE SURVEY 38 Larsen-Basse and Premaratne [112] showed that craters form on the surface of a material when the grit was trapped (but could still rotate) between the counterface and the specimen. Trapped grit particles that could not rotate resulted in micro- grooving of the material?s surface. 2.4.2 Mechanisms of abrasive wear There are two predominant mechanisms of abrasive wear i.e. plastic deformation and brittle fracture [98]. This section reviews these mechanisms. Abrasive wear by plastic deformation This mechanism involves the removal of material by plastic deformation. As wear progresses, the material becomes strain hardened by plastic flow which increases the surface hardness of the material compared to the bulk. Shear strain due to wear decreases with depth into the bulk of the material. The depth of penetration of the abrasive material is proportional to both the depth of deformation and strain at a given depth [98]. Therefore, hardness of the worn surface is more important in this type of wear than the bulk hardness of the material and improving the surface hardness by laser alloying should result in improved wear resistance. There are three modes of plastic deformation caused by abrasive wear, namely cutting, ploughing and wedge formation. These modes of deformation depend on the abrasive attack angle which is the angle between the leading face of the abrasive particle and the wearing surface [98,100]. Cutting mode - This mode of deformation occurs at high attack angles [98,102]. The deformed material is deflected through a shear zone and forms a chip up the front face of the abrasive particle [98,99]. All the material displaced by the abrasive particle is removed in the chip. LITERATURE SURVEY 39 Ploughing mode - In this mode of deformation, a ridge of deformed material is pushed down ahead of the abrasive particle and to the sides of the groove [98,99,102]. There is little or no material removal. This mode occurs at low attack angles. Wedge formation mode - This mode represents the intermediate of the previous two modes. It involves limited slip or complete adhesion between the front face of the abrasive particle and a raised prow of material leading to continuous growth and detachment of the material [98,99]. Not all material removed during cutting and wedge formation forms wear debris. Some is displaced without being removed. The proportion of material which forms debris decreases with the ratio E/H (Young?s modulus/surface hardness) [98]. Abrasive wear by brittle fracture In this mechanism, material removal occurs by brittle fracture. Fracture occurs when the load on the specimen exceeds the fracture strength of the specimen. Zum Gahr [102] described the microcracking mode which occurs when highly concentrated stresses are imposed by abrasive particles. Large wear debris are removed from the surface of the material due to crack formation and propagation. In brittle materials, microcracking may cause wear volumes greater than the groove volume produced. Figure 2.16 shows the formation of median and lateral cracks by a point indenter. In abrasive wear the stress on the material is initially very high because the abrasives are fresh and sharp. To relieve these stresses, the tip of the abrasive subjects the material to local plastic flow or densification (Figure 2.16(a)). When the load reaches a critical value, a median crack is formed by tensile stresses across the vertical plan (Figure 2.16(b)) and grows in depth as the load is LITERATURE SURVEY 40 increased (Figure 2.16 (c)). During unloading, the median vent cracks close (Figure 2.16 (d)) and lateral vent cracks eventually form (Figure 2.16 (e)) and propagate towards the surface of the material (Figure 2.16 (f). Lateral cracks lead directly to wear and are formed by residual elastic stresses due to the relaxation of the deformed material around the region of contact [98,102]. Deuis et al. [100] also identified radial cracks which formed on pre-existing surface flaws like micro-cracks. Figure 2.16: Diagram showing median and lateral crack formation in a brittle material due to indentation by a sharp indenter [98]. D is the damaged layer, M is the median crack and L is a lateral crack. The load is increased from (a) to (c) and progressively decreased from (d) to (f). LITERATURE SURVEY 41 Lateral cracks form only when the applied load on the abrasive is greater than a critical value (w*) which depends on the fracture toughness (KIc) and the hardness (H) of the material as described by Equation 2.3 [98,100]. IcK 3 H IcK * ?? ??? ? ?w (Equation 2.3) Residual stresses in the sample may increase the crack length and reduce the critical load for microcracking [102]. The ratio H/KIC is the brittleness index and decreases with an increase in the critical load (w*). Reloading closes the lateral cracks and re-opens the median cracks. During abrasive wear, plastic grooves form and lateral cracks grow upwards causing the deformed layer to chip off (see Figure 2.17). High loads and large abrasive particle sizes favour brittle fracture and often result in high wear rates [98,100]. Figure 2.17: Schematic illustration of material removal in a brittle material by the growth of lateral cracks below a plastic groove [98]. b is the depth of the damaged layer where the lateral cracks occurred and c is the length of the lateral cracks. LITERATURE SURVEY 42 2.4.3 Wear of aluminium alloys Research has been conducted on wear of aluminium, MMCs and laser alloyed surfaces by various authors [107,113-116]. Elleuch et al. [115] studied the abrasive wear of aluminium alloys rubbed against sand. The author reported that the wear rate increased by 3 times as the incident angle of sand was increased from 0 to 45?. Abrasive grooves and wear debris were observed on the worn surfaces. ?ahin [116] studied the abrasive wear of aluminium AA2014 reinforced with SiC particles of different sizes (9?m, 14?m and 33?m). The wear tests were performed on a pin-on-disc machine and the counter surface was SiC abrasive paper. The authors observed that the wear rate was affected by hardness of the metal matrix composite, the particle size of the embedded particles, the applied load, sliding distance and abrasive particle size of the counter surface in the following manner: ? Wear rate decreased with increasing hardness of the MMC, ? Wear rate increased with increasing abrasive particle size of the counter surface, ? Wear rate increased with increasing applied load, ? Wear rate increased with sliding distance up to a maximum value, then either decreased or remained constant. Almeida et al. [117] studied the dry sliding wear mechanisms of Al-Mo deposited on an aluminium substrate by laser surface alloying. The wear mechanisms were predominantly adhesion followed by material detachment and transfer, oxidation and some abrasion, mainly by hard intermetallic compound particles on the steel counterbody. Staia et al. [113] alloyed aluminium A356 with a powder consisting of 96wt%WC, 2wt%Ti and 2wt%Mg. The alloying was performed with a 2kW Nd:YAG laser at different scanning speeds (16.7 to 66.7mm/s). Smaller and LITERATURE SURVEY 43 uniformly distributed WC grains were formed after alloying with lower scanning speeds (66.7mm/s). This was attributed to a higher beam-material interaction time which was achieved at lower scanning speeds. Different carbides (WC, W2C, W6C2.54 and Al4C3) were formed after alloying. Oxidation leading to the formation of Al2O3 and SiO2 occurred in the molten pool because an inert shielding gas was not used. The wear behaviour of the alloyed material against AISI52100 steel balls under a load of 5N was investigated. For high laser scanning speeds, large WC particles were formed which served as the load carrying particles and severely abraded the steel. For low laser scanning speeds, the WC particle size decreased and the wear mechanisms changed as the aluminium matrix participated in the transference process. The wear mechanisms of the aluminium A356 was adhesive with high quantities of aluminium transferred to the steel counterface. Miyajima and Iwai [107] reported that SiC and Al2O3 reinforcement caused severe wear of a steel counter surface during sliding wear of the aluminium matrix composite on a pin-on-disk wear tester. Metal matrix composites with low volume fractions of embedded ceramics produce severe wear characterized by plastic deformation and large grooves. MMCs with high volume fractions of embedded ceramics did not wear severely and the surfaces were smooth and flat without large grooves. The embedded ceramic acted as inhibitors against plastic flow and adhesion of the matrix metal to the steel counter surface. Majumdar et al. [114] laser alloyed Ti with Si, Ti with Al, and Ti with Si + Al using a 6kW continuous wave CO2 laser and studied the wear behaviour on a ball- on-disc wear testing machine. The microstructure of the surfaces alloyed with Si and Al + Si consisted of Ti5Si3 and ?-Ti phases and those alloyed with Al only consisted of ?-Ti phase. A hardness of 780HV was obtained when alloying with Si compared to 700HV when alloying with Si + Al and 450HV when alloying with Al. The high hardness for surfaces alloyed with Si was attributed to the formation of the Ti5Si3 phase. The hardness of the Ti substrate was 200HV. The increase in hardness for surfaces alloyed with Al was due to solid solution hardening. The wear resistance of the alloyed surfaces followed a similar trend as LITERATURE SURVEY 44 the hardness. The wear of pure Ti was characterized by extensive amounts of adhesive wear and localized deformation. This resulted in formation and propagation of micro-cracks. Surfaces alloyed with Al suffered extensive amount of abrasive and adhesive wear. The micro-cracks observed in these surfaces were smaller compared to pure Ti, which resulted in lower material loss. Surfaces alloyed with Si + Al suffered abrasive wear while no significant amount of adhesive or abrasive wear was observed in samples alloyed with Si. This decrease in the wear rates was associated with the formation of the Ti5Si3 phase. Surfaces alloyed with Si produced a higher volume fraction of the Ti5Si3 phase compared to surfaces alloyed with Si + Al. Shipway et al. [118] studied the sliding wear mechanisms of aluminium reinforced with TiC particles. The wear tests were conducted on a pin-on-disc testing machine. The counter surface was a disc of carbon-manganese steel (BS 080A15). The hardness of the metal matrix composite layer increased as the volume fraction of the TiC particles increased. The wear rates increased as the applied load increased. Das [6] also reported an increase in wear rate as the load increased for aluminium reinforced with SiC particles. Delamination cracks were observed in the MMC layer. Addition of TiC particles resulted in counter surface wear as hard particles act as abrasives in the sliding process. The wear rates of the steel counter surfaces increased as the volume fraction of the TiC particles increased. These TiC particles resulted in ploughing and cutting of the steel counter surfaces in the sliding direction. Ax?n and Zum Gahr [103] studied the abrasive wear of a tool steel 90MnCrV8 laser alloyed with TiC particles. Laser treatment was performed with a CO2 laser using powder densities of 8 and 10kWcm-2 and a scanning speed of 100mm/min. The wear tests were performed on an abrasive wheel machine with SiC abrasive papers of different grit sizes (20, 75, 135, and 200?m) and a load of 10N. The hardness of the MMC increased with decreasing TiC particles (from 30 to 3?m) due to the increased volume fraction of embedded TiC particles and the reduced mean free path between individual particles. A strong interfacial bond between the LITERATURE SURVEY 45 TiC reinforcements and the matrix is also important during wear. If the interfacial bond is weak, the reinforcements are easily pulled out during wear and act as additional abrasives increasing the wear rate. The wear resistance increased with decreasing size of the SiC abrasive grits. Laser injecting TiC particles on tool steel increased the wear resistance by a factor of approximately four against 200?m SiC paper and approximately six against 75?m SiC paper. Coarse abrasive grits (200?m) were able to crack and spall off large TiC particles (30?m) embedded in the soft matrix while fine abrasive grits (20?m) caused mild surface damage. Small TiC particles (3?m) were easily dug out of the soft matrix by coarse SiC abrasive grits (200?m) owing to greater deformation of a softer matrix at a given load. Loose TiC particles acted as abrasives in addition to SiC grits increasing the wear rate. Fine abrasive grits (20?m) were not able to pull out 3?m TiC particles. 2.5 Fracture Fracture is the end result of plastic-deformation processes [119]. There are two main types of fracture, brittle and ductile fracture. Fatigue fracture also occurs due to cyclic loading (or alternating stresses) but will not be reviewed in this work. 2.5.1 Ductile fracture During tensile tests, small cavities (pores) may occur in the metal near the centre of the cross-section [119]. The density of these pores increases with increasing deformation. Under high stresses, these pores grow and coalesce leading to the formation of a ductile crack. A ductile crack spreads as a result of intense localized plastic deformation of the metal at the tip of the crack. The crack propagates by a void-sheet mechanism. The stress concentrations at the end of the crack localizes the plastic deformation in these regions into shear bands that makes angles of 30? to 40? with the stress axis. As the deformation inside the band is very intense, the bands become filled with voids. The bands filled with voids are also known as void sheets. As the voids grow, they eventual impinge on each other splitting the void sheet into two and a crack advances. LITERATURE SURVEY 46 Failure of most ductile materials occurs with a cup-and-cone fracture. This type of fracture is closely associated with the formation of a neck in a tensile specimen. Completion of the fracture occurs at an angle of 45? with the tensile-stress axis. This leaves the surface of one half of the specimen with an appearance of a shallow cup and the other half with a cone flattened at the top. At low magnification, the fractured surface has a rough and spongy texture. At high magnification, cuplets are observed. 2.5.2 Brittle fracture A brittle fracture occurs when the movement of the crack involves little plastic deformation of the metal adjacent to the crack [119]. Fractured surfaces are sharp and smooth. Cleavage occurs when a crystal splits into two pieces along planes of low indices (or cleavage planes). A certain amount of plastic deformation almost always occurs when metals fail by brittle cleavage. This plastic deformation occurs because metals are never completely brittle even at very low temperature. Thus, cleavage cracks nucleate by plastic deformation processes. Plastic deformation associated with a moving crack is more liable to occur just ahead of the crack. This region (ahead of the crack) is in a state of high stress and the stress is normal to the plane of the crack. A tensile stress of this type is equivalent to a set of shear stresses on planes at 45? to the tensile axis. Since these shear stresses are large, it is possible to nucleate dislocations ahead of the crack on slip planes that are favourably oriented with respect to the shear stress. River patterns are tell- tail marks formed when cleavage cracks are formed. The origin of a cleavage crack can be traced by following the river pattern back to its origin. Brittle fracture can also occur along grain boundaries due to a grain-boundary film of a hard/brittle second phase or due to a concentration of solute in the metal close to the crystal boundary. LITERATURE SURVEY 47 Research has been conducted to investigate the failure mechanisms of laser alloyed metal matrix composites [120]. Vreeling et al. [120] studied failure mechanisms in a Al/SiC metal matrix composite produced by laser embedding with a 2kW Nd:YAG laser. Large, isolated and randomly distributed Al4C3 plates were formed near the surface where the temperature of the melt pool was the highest during laser injection. Short Al4C3 plates (? 20?m) were also observed at the Al/SiC interface. Tensile tests were performed in the longitudinal direction (direction of laser injection). The fracture mechanisms observed were: ? decohesion of the incoherent bond between the randomly distributed Al4C3 plates and Al matrix; ? brittle fracture of the embedded SiC particles (characterized by crack formation); ? decohesion of SiC particles from the Al matrix and ? ductile fracture of the Al characterized by a typical morphology of dimples. The Al4C3 particles which formed around the SiC particles assisted in anchoring the SiC particles to the Al matrix preventing decohesion of the SiC particles. However, when the matrix is deformed, the Al4C3 plates transfer the stress to the particle surface, which may lead to failure of the SiC particle. The brittle fracture of the SiC particle initiates and propagates through the particle by cleavage mechanisms. Suppressing the formation of the Al4C3 phase would postpone the failure initiation process and failure would only occur by particle cracking and Al/SiC interface debonding. The size of the powder particles affects their fracture process. Small powder particles are less prone to have internal defects and are thus more difficult to be fractured [121]. The stress concentration level on each particle is lower because there are more particles bearing the applied load which lowers the probability of fracturing. The high surface to volume ratio of small particles promotes LITERATURE SURVEY 48 agglomeration and clustering. Clusters have less ability to transfer shear and tensile stresses resulting in poor mechanical properties. EXPERIMENTAL PROCEDURE 49 3. EXPERIMENTAL PROCEDURE This chapter describes the experimental procedure followed. Laser surface alloying of Aluminium AA1200 with various combinations of Ti, Ni and SiC powders was done. The resulting alloy microstructures were characterized with optimized alloys being subjected to mechanical and wear testing. This chapter is arranged as follows: head2right Aluminium AA1200 characterization head2right Powder characterization head2right Synthesis of the alloys head2right Characterization of the alloyed surfaces head2right Hardness testing head2right Wear testing head2right Impact testing head2right Microscopy analysis 3.1 Aluminium AA1200 characterization 3.1.1 Specimen preparation Aluminium AA1200 specimen obtained from Non-Ferrous Metal Works (SA) (Pty) Ltd were sectioned to 10x10x6mm3 dimensions using a Struers Discotom-2 cutting machine with a Corundum L205 cut-off wheel. A Struers lubricant, which also acts as a coolant, was used during cutting. After sectioning, the specimens were mounted in a thermosetting bakelite resin using a Struers Prestopress-2 heat and pressure mounting press. The specimen were then ground and polished to a 0.04?m (OP-S suspension) surface finish using a Struers TegraForce-5 auto polisher. After polishing, specimens were rinsed in water and subsequently in alcohol, before being cleaned ultrasonically in EXPERIMENTAL PROCEDURE 50 alcohol. The polished samples were etched with Keller?s reagent (3mlHCl + 2mlHF + 5mlHNO3 + 190mlH2O) to reveal the microstructure. Microstructural investigations were conducted using different instruments (refer to section 3.7). 3.1.2 Vickers Hardness The Vickers hardness of polished specimens was determined using a Matsuzawa Seiki micro-hardness tester. Indenting loads of 100g and 1kg were applied for a dwell time of 15 seconds. Different loads were used in order to provide a direct comparison with the hardness results of the laser alloyed surfaces. 3.1.3 Density The density was determined using the Archimedes principle. The specimens were first weighed in air, then again while suspended in water. The density was calculated using equation 3.1 [108]. ?? ? ? ?= wa w ws ?? (Equation 3.1) Where: ?s = density of aluminium AA1200 ?w = density of water Ma = mass of specimen in air Mw = mass of specimen in water 3.2 Powder Characterization The Ni, Ti and SiC powders obtained from WEARTECH (Pty) Ltd were characterized using different instruments to determine the powder properties. EXPERIMENTAL PROCEDURE 51 3.2.1 Powder particle morphology The Ni, Ti and SiC powder particles were examined with a Jeol JSM840 scanning electron microscope (SEM) to determine the particle shape. Plan view and cross- sectional micrographs of the powder particles were taken. Cross-sections were made by mounting the powders in a thermosetting resin then grinding and polishing the particles with a Struers TegraForce-5 auto polisher to a 0.04?m finish (OP-S colloidal silica suspension). Energy dispersive spectroscopy (EDS) was used to identify the elements present in each powder. A PANaytical X?Pert Pro powder diffractometer (XRD) was used for phase analysis. The powder particle size and distribution were determined using a Malvern Mastersizer 2000. The powder particle size and distributions determined from the experiments were compared to the values obtained from the supplier (40-100?m for each of the Ni, Ti and SiC powders). 3.2.2 Hardness of the powder particles Vickers hardness measurements of the Ni, Ti and SiC powders were performed on cross-sections of mounted powder particles using a Matsuzawa Seiki micro hardness tester. The loads used were 10g (for Ni and Ti powder particles) and 100g (for SiC powder particles). A higher load was used for SiC particles because the indentations obtained with 10g load were too small to measure. The dwell time was 15 seconds for each hardness indent. 3.3 Synthesis of the alloys Aluminium AA1200 100x100x6mm3 plates were sand blasted and cleaned with acetone to enhance the absorption of laser energy by the aluminium substrate. The plates were laser alloyed using a laser injection system shown in Figure 3.1. The laser light was delivered to the target material through fibre optics and focused using a ZnSe lens. A KUKA articulated arm robot was used to control the movement of the alloying head and an off-axis nozzle of 2.5mm in diameter was EXPERIMENTAL PROCEDURE 52 used for powder delivery. The nozzle was mounted on the laser head and fixed at a distance of 12mm from the substrate. This arrangement assured that the powder stream coincided with the laser beam at the interaction zone. The alloying powders constituted different compositions of Ni, Ti and SiC powder mixtures. A commercial powder feed instrument equipped with a flow balance was used for controlling the powder feed rate; this was set to 2g/min. The powder feed rate varied between 2-3g/min depending on the powder composition. Argon was used as the carrier and shielding gas to prevent oxidation during the alloying process. Figure 3.1: Experimental setup using a Nd:YAG laser. 3.3.1 Optimization tests Optimization tests were performed on single tracked alloys in order to determine the optimum processing parameters. The selection criteria during optimization KUKA articulated arm robot Off axis nozzle Cooling water Aluminium sample Powder feeding tube Laser head Nd:YAG fibre optics EXPERIMENTAL PROCEDURE 53 tests were that surfaces must have a homogeneous layer free of porosity and have an alloyed thickness (depth of alloying) less than 3mm (i.e. 50% of the base aluminium thickness). The scanning speed was varied (10, 12, 15 and 20mm/sec) in order to determine the optimum processing parameters (i.e. parameters for homogeneous crack free surfaces). The laser power and laser beam diameter used were 4kW and 4mm. These laser parameters are listed in Table 3.1. Table 3.1: Laser parameters used during the laser alloying experiments. Laser power 4kW Laser beam diameter 4mm Laser scanning speeds 10, 12, 15 and 20mm/s Powder feed rate 2-3g/min Shielding gas Argon Shielding gas flow rate 4L/min Table 3.2 shows the powder compositions that were chosen for the optimization tests. These compositions were chosen through trial and error. A high laser power was used in order to melt the Ni and Ti powder particles, but partially melt the SiC particles. Some of the SiC particles were retained in order to form metal matrix composites. Table 3.2: Starting powder mixtures. Sample numbers Composition (wt %) 1-4 33.3 Ni + 33.3 Ti + 33.3 SiC 5-8 80 Ni + 15 Ti + 5 SiC 9-12 50 Ni + 20 Ti + 30 SiC 13-16 5 Ni + 80 Ti + 15 SiC 17-20 30 Ni + 50 Ti + 20 SiC 21-24 15 Ni + 5 Ti + 80 SiC 25-28 20 Ni + 30 Ti + 50 SiC After the optimization tests, laser alloying was performed with the powder compositions listed in Table 3.3. Single track and multi-track alloys were made. For multi-track alloying, ten overlapping tracks with a pass overlap of 50% were EXPERIMENTAL PROCEDURE 54 made (i.e. the width of one track was overlapped by 50% when forming the following track). After alloying, cross-sections were prepared using the same procedures described in Section 3.1.1. Table 3.3: Compositions of powder mixtures. Composition No. Ni (wt%) Ti (wt%) SiC (wt%) 1 33 33 33 2 85 5 5 3 80 15 5 4 70 10 20 5 70 20 10 6 60 30 10 7 60 20 20 8 50 30 20 9 50 20 30 10 50 10 40 11 50 40 10 12 40 20 40 13 40 30 30 14 5 85 5 15 5 80 15 16 10 70 20 17 20 70 10 18 10 60 30 19 20 60 20 20 30 50 20 21 20 50 30 22 10 50 40 23 40 50 10 24 20 40 40 25 30 40 30 26 5 5 85 27 15 5 80 28 20 10 70 29 10 20 70 30 30 10 60 21 20 20 60 32 20 30 50 33 30 20 50 34 10 40 50 35 40 10 50 36 40 40 20 37 30 30 40 EXPERIMENTAL PROCEDURE 55 3.3.2 Breakdown alloy systems Due to the complexity of the Al-Ti-Ni-SiC alloy system investigated in this project, laser surface alloying was also conducted on the breakdown systems of the alloy system. These breakdown systems were the Al-Ni, Al-Ti, Al-Si, Al-TiC, Al-SiC, Al-Ni-Ti, Al-Ni-Si, Al-Ti-Si, Al-Ni-SiC and Al-Ti-SiC systems. The purpose of making these systems was to understand the microstructures and phases formed in the complex Al-Ni-Ti-SiC system. The laser parameters used to form these breakdown systems were those optimized in Section 3.3.1 above. 3.4 Hardness of the alloys The Vickers hardness of polished alloyed surfaces was determined using the Matsuzawa Seiki hardness tester. An indenting load of 1kg and a 15 second dwell time was used for each hardness indent. The average of eleven indentations is reported for each alloy. Polished cross-sections were used to determine through-thickness hardness profiles (indentations from the surface of the alloyed layer through to the aluminium base) using a 100g load with 100?m spacing between subsequent indentations. 3.5 Wear testing Wear tests were performed on alloyed surfaces using two body and three body dry abrasion wear rigs. The size of the sample holders of these two machines determined the size of the specimen that could be used with each machine. Two body abrasion wear tests were performed on polished single track alloyed surfaces while the three body abrasion wear tests were performed on polished multi-track alloyed surfaces. EXPERIMENTAL PROCEDURE 56 3.5.1 Two body abrasion wear tests The two body abrasion wear apparatus is shown in Figure 3.2. The diameter of the sample holder of the two body abrasion wear apparatus is 10mm. In the laser alloying process the diameter of the beam used was 4mm resulting in a track width of approximately 4mm. Therefore, single track alloyed surfaces were used in this sample holder. The size of the samples used was 9mm x 4mm with a height of 6mm. The samples were polished to a 0.04?m finish prior to testing. The laser alloyed aluminium (body 1) was attached to the loading arm which moved the alloy across the SiC abrasive paper (body 2). 80-grit SiC grinding paper with a diameter of 200mm was used as the abrasive material. Sample Holder Start/Stop switches Figure 3.2: Two body abrasive wear apparatus. The loading arm moves horizontally in a straight line up to the centre of the rotating disc. The stop/start switches were used to control the movement of the loading arm. The rotating disc had a fixed speed of 262.7rpm which was Loading arm Rotating disc with SiC abrasive paper EXPERIMENTAL PROCEDURE 57 measured using a tachometer. The speed of the loading arm was set such that specimens constantly abraded on fresh grits. This speed was determined using a marking pen (body 1) and a sheet of paper (body 2) to simulate the wear experiment. The marking pen traced a spiral wear path (see Figure 3.3) with a spiral length of 2.53m. Figure 3.3: The traced sliding path of the specimens during two body wear testing. The samples traced the spiral path twice over the SiC paper before being weighed and the SiC paper changed. This resulted in a sliding distance of 5.06m in a time of 4.8s. The total duration of each wear test was 24 seconds but at intervals of 4.8 seconds the test was interrupted and the mass of the sample recorded. Three tests per alloy were conducted. The load applied onto the wear sample was based on the weight of the sample holder that pressed down onto the specimen during the test. The load on the specimen was therefore calculated as follows: F = mg Where F is the applied load, m is the mass of sample holder and g is the gravitational force. Therefore, F = (220.362/1000)kg x 9.8ms-2 = 2.16N EXPERIMENTAL PROCEDURE 58 3.5.2 Three body dry abrasion wear tests Specimens that were alloyed with multiple tracks were subjected to three body dry abrasion wear tests in accordance with the ASTM G65-04 dry sand, rubber wheel standard [122] using the machine shown in Figure 3.4. The three bodies were the sand, rubber and samples. Silica sand obtained from Rolfes (Pty) Ltd. with a particle size range of 0.3-0.65mm was used as the abrasive. Prior to wear testing the sand was sieved with a MACSALAB electronic sieve shaker to determine the particle size distribution. A sample of 2kg sand was taken from the top, middle and bottom of the bag for the sieve tests. Three samples were taken at the different depths in the bag (top, middle and bottom) to determine the consistency of the sand particles. The electronic sieve shaker was used in continuous vibration mode for 10 minutes. The silica sand is poured into the hopper and is introduced to the testing chamber by a vibratory feeder. The sand feed rate was 2.3g/s. This feed rate ensured that sufficient sand was introduced into the rubber wheel/specimen contact area in the testing chamber (see Figure 3.5). EXPERIMENTAL PROCEDURE 59 Figure 3.4: Three body dry abrasion wear instrument. The size of the multi-track samples used was 70x20x6mm. These samples were polished to a 0.04?m finish prior to wear testing and placed in the sample holder shown in Figure 3.5. A 1kg weight was placed onto the loading hook to give an applied load of 9.8N. The duration of each wear test was 60 minutes with the mass of the samples recorded at 10 minute intervals. Three tests were conducted to determine the average for each alloy. Sand hopper Vibratory feeder Sand fed to the testing chamber Testing chamber Sand collected after test Loading hook with weights EXPERIMENTAL PROCEDURE 60 Figure 3.5: The testing chamber of the three body abrasive wear apparatus showing a rubber wheel, a sample holder and a sand feeder. 3.6 Impact testing Aluminium AA1200 and multiple-track laser alloyed specimen were subjected to impact testing (Charpy V-notch) according to the ASTM G23-05 standard [123]. The tests were performed on a Tinius Olsen impact machine shown in Figure 3.6. The impact test is a standardized strain-rate test which determines the amount of energy absorbed by a material during fracture. The impact tester consists of a sample holder, a swinging pendulum axe and the energy indicator. Specimens were machined to 55x10x6mm dimensions (see Figure 3.7) and a 2mm deep notch machined at mid-length (27.5mm) at 45 degrees to the surface. Rubber wheel Sand feeder Sample holder Silica sand EXPERIMENTAL PROCEDURE 61 Figure 3.6: The impact testing machine. Figure 3.7: Specimen used for impact tests. Pendulum axe Pendulum lock Safety guards Sample holder Energy indicator 55mm Notch 6mm 10mm EXPERIMENTAL PROCEDURE 62 Specimens were placed in the sample holder using self-centering tongs. The pendulum axe was raised to a latched position (as shown in Figure 3.6) and locked before each test. The energy was set to the maximum position and the pendulum released. The pendulum axe strikes the specimen directly opposite the notch. The energy reading after fracturing the specimen was recorded as the energy absorbed by the specimen. 3.7 Microscopy analysis Surfaces and cross-sections of aluminium AA1200 and laser alloyed specimens were viewed with different types of microscopes at various magnifications in order to characterize the microstructures. The phases formed during laser alloying were determined using an X-Ray diffractometer. Worn and fractured surfaces were viewed with various microscopes to determine the wear and failure mechanisms. This section describes the microscopes used. 3.7.1 Optical microscopy An OLYMPUS BX51M optical microscope was used at various magnifications (5X to 100X). An ALTRA 20 Soft Imaging system digital camera is connected to the microscope and images of the specimen are observed and captured with analySIS? FIVE software. 3.7.2 Stereo microscopy An OLYMPUS SZ-CTV stereo microscope with a magnification range of 0.67X to 4X and a BAXALL CDX digital camera was used. The microscope is connected to a computer and images were observed and captured with analySIS? FIVE software. The depth of the alloyed layers was measured with the analySIS? FIVE computer software. EXPERIMENTAL PROCEDURE 63 3.7.3 Scanning electron microscopy A JEOL JSM-840 scanning electron microscope (SEM) was used for the following: o Determination of the powder morphology; o Microstructural characterization o EDS analysis o Examining the wear scars and fractured surfaces. 3.7.4 X-ray diffraction A PANalytical X? Pert Pro diffractometer (XRD) was used to determine the phases present in the alloyed surfaces. The radiation source used was Cu K? (1.54? wavelength). The XRD was operating at 40kV and 20mA. The data was acquired in the diffraction angle range of 10? ? 2? ? 130?, with a step size of 0.02?. The phases were identified using X?Pert Software with ICDD PDF reference cards. RESULTS: SYNTHESIS OF THE ALLOYS 64 4. RESULTS: SYNTHESIS OF THE ALLOYS This chapter presents the results of the synthesis of the aluminium alloys. It begins with characterization of the Aluminium AA1200 material which was subjected to surface modification. This is followed by the characterization of the Ti, Ni and SiC powders used to alloy the Aluminium AA1200. Next the optimization process for laser surface alloying is described. Due to the complexity of the Al-Ti-Ni-SiC alloy system investigated in this project, laser surface alloying was first conducted on the breakdown systems of the alloy system and then on the selected Ni-Ti-SiC compositions. 4.1 Aluminium AA1200 characterization The microstructure of Aluminium AA1200 is shown in Figure 4.1 and the chemical composition in Table 4.1. The SEM image shows the ?-Al phase with traces of FeAl3. Figure 4.1: SEM micrograph of aluminium AA1200 showing traces of FeAl3 in the ?-Al matrix. FeAl3 RESULTS: SYNTHESIS OF THE ALLOYS 65 Table 4.1: Chemical composition of aluminium AA1200. Element Cu Si Fe Al Composition (wt %) 0.12 0.13 0.59 Balance Figure 4.2 is the XRD pattern of aluminium AA1200 showing only Al peaks despite the chemical composition listed in Table 4.1; an indication of the high purity level of the aluminium. The Miller indices confirm that the crystal structure of aluminium is face-centered cubic (fcc). Figure 4.2: XRD pattern of aluminium AA1200. The average Vickers hardness for the Al AA1200 is 24.0?0.4HV0.1. This was calculated from the average of the ten hardness indentations listed in Table 4.2. Table 4.2: Vickers hardness of Aluminium AA1200. Test 1 2 3 4 5 6 7 8 9 10 HV 24.1 23.7 24.0 23.8 23.8 23.7 24.9 23.7 24.6 23.9 RESULTS: SYNTHESIS OF THE ALLOYS 66 The density of five aluminium AA1200 specimen is given in Table 4.3 and the average of 2.70?0.01g/cm3 is taken as the density of the alloy. Table 4.3: Density of aluminium AA1200 specimen. Sample number 1 2 3 4 5 Density (g/cm3) 2.70 2.70 2.71 2.69 2.70 4.2 Powder characterization The Ni, Ti and SiC powders were characterized in terms of particle shape and size distribution as well as hardness and phase analysis. The shape of the powder affects the manner in which it flows during laser alloying with round particles flowing more easily than irregular particles. The size of the particles affects the powder efficiency during laser alloying. Small particles coagulate within the powder nozzle, due to their large surface area, blocking the nozzle while large particles produce poor powder focus. The hardness of the powder particles were used for comparison with the hardness of the phases formed during laser alloying. Hardness was also useful in identifying un-dissolved powders after laser alloying. Ni powder Figure 4.3(a) is an SEM micrograph of the Ni powder particles showing that the majority of the particles are spherical while some are irregular. Large particles were formed by the agglomeration of small particles. Figure 4.3(b) shows the Vickers hardness indent made on one Ni powder particle using an applied load of 10g. The average hardness for the Ni powder is 154.8 ? 50.1HV0.01. RESULTS: SYNTHESIS OF THE ALLOYS 67 Figure 4.3: (a) SEM image of the Ni powder particles showing spherical, irregular and agglomerate particles. (b) Hardness indentation on a Ni particle. The XRD pattern of the Ni powder is shown in Figure 4.4; only Ni peaks were detected. The Miller indices show that Ni has a face-centered (fcc) crystal structure. Figure 4.4: XRD diffractograph of Ni powder showing Ni peaks. The Ni powder particle size distribution (7-200?m) is shown in Figure 4.5 with the average Ni powder particle size at 55.6?m. (a) (b) RESULTS: SYNTHESIS OF THE ALLOYS 68 Figure 4.5: Ni powder particle size distribution curve. Ti powder Figure 4.6(a) is an SEM micrograph of the Ti powder particles showing irregular agglomerates. Figure 4.6(b) shows the Vickers hardness indent made on one Ti powder particle using an applied load of 10g. The average hardness for the Ti powder is 83.0 ? 25.2HV0.01. Figure 4.6: (a) SEM image of the Ti powder particles showing irregular agglomerates. (b) Hardness indentation on a Ti particle. The XRD pattern of the Ti powder is shown in Figure 4.7 with only Ti peaks detected. The Miller indices show that Ti has a hexagonal close-packed (hcp) crystal structure. Particle Size Distribution 0.01 0.1 1 10 100 1000 3000 Particle Size (?m) 0 1 2 3 4 5 6 7 8 9 Vo lu m e (% ) Nickel - Average, Friday, November 14, 2008 10:21:08 AM (a) (b) RESULTS: SYNTHESIS OF THE ALLOYS 69 Figure 4.7: XRD diffractograph of Ti powder showing Ti peaks. The Ti powder particle size distribution (5-158?m) is shown in Figure 4.8 with the average Ti powder particle size at 69.2?m. Figure 4.8: Ti powder particle size distribution curve. SiC powder Figure 4.9(a) is an SEM micrograph of the SiC powder particles showing that the particles were irregular in shape. Figure 4.9(b) shows the Vickers hardness indent made on one SiC powder particle using an applied load of 100g. A higher load Particle Size Distribution 0.01 0.1 1 10 100 1000 3000 Particle Size (?m) 0 2 4 6 8 10 12 14 Vo lu m e (% ) Titanium - Average, Friday, November 14, 2008 10:54:48 AM RESULTS: SYNTHESIS OF THE ALLOYS 70 (compared to 10g used for Ni and Ti powder particles) was used since the indentations made with a 10g load were too small for measurement. The average hardness for the SiC powder is 2630.9 ? 760.5HV0.1. Figure 4.9: (a) SEM image of the SiC powder particles showing irregular particles. (b) Hardness indentation on a SiC particle. The XRD pattern of the SiC powder is shown in Figure 4.10 and only SiC peaks were detected. The Miller indices show that SiC (3C, ?-SiC) has a zinc blende cubic crystal structure similar to that of diamond [63]. Figure 4.10: XRD diffractograph of SiC powder showing SiC peaks. (a) (b) RESULTS: SYNTHESIS OF THE ALLOYS 71 The SiC powder particle size distribution (14-800?m) is shown in Figure 4.11 with the average particle size at 63.6?m. Figure 4.11: SiC powder particle size distribution curve. Summary of powder characterization results Table 4.4 summarizes the results obtained from characterization of the Ni, Ti and SiC powders. The SiC powder has a wide particle size distribution compared to the Ni and Ti powders. SiC also has the highest hardness with Ti showing the lowest. Ni powders were spherical while Ti and SiC powders were irregular. Table 4.4: Summary of the powder properties. Powder Particle size range (?m) Average particle size (?m) Particle Shape Vickers Hardness (HV) Ni 7 - 200 55.6 Spherical with agglomerates 154.8 ? 50.1 Ti 5 - 158 69.2 Irregular agglomerates 83.0 ? 25.2 SiC 14 - 800 63.6 Irregular 2630.9 ? 760.5 4.3 Optimization of the laser surface alloying process After laser alloying the aluminium AA1200 specimens with selected Ti-Ni-SiC compositions according to the procedure described in Chapter 3.3.1, alloyed Particle Size Distribution 0.01 0.1 1 10 100 1000 3000 Particle Size (?m) 0 1 2 3 4 5 6 7 8 9 Vo lu m e (% ) Silicon carbide - Average, Friday, November 14, 2008 11:10:18 AM RESULTS: SYNTHESIS OF THE ALLOYS 72 cross-sections were cut and metallurgically prepared for microscopic analysis and hardness measurements. The criterion for optimization includes production of a homogeneous, crack-free surface layer. Therefore alloyed surfaces which revealed a heterogeneous microstructure were excluded from further testing. Since the aim of the project was to modify only the surface of the aluminium, 3mm was selected as the maximum depth for the laser alloyed surface layer. The thickness of the base aluminium was 6mm. Therefore, a 3mm thickness for the alloyed layer meant that 50% of the base aluminium was laser alloyed. Thus alloyed layers with a thickness greater than 3mm were excluded from further testing. Phase identification is described in Chapter 4.5. 4.3.1 Microstructure of the alloyed layers Figure 4.12 shows the single-track alloyed layers for compositions alloyed using a laser scanning speed of 10 mm/s. These layers were crack-free. In the figure the alloyed layer (AL) represents a layer filled with alloyed microstructure while the thermo-affected layer (TL) is a heat treated layer with no (or little) alloyed microstructure. The shape of the alloyed layers is due to convective flow of the melt pool and the Gaussian shape of the laser beam. There is a temperature gradient from the centre towards the edges of the alloyed track. The centre is subjected to higher temperatures while the edges are subjected to lower temperatures. During alloying the molten metal pool flows from the region of high to the region of low temperature by convection. Good quality alloyed layers are those which have a homogeneous alloyed microstructure throughout the layer as shown in Figures 4.12 (a-c and g). The layers in Figures 4.12 (d-f) have a thermo-affected layer in the heat affected zone and were therefore excluded from further characterization. The thermo-affected layers were observed on surfaces laser alloyed with high Ti (? 50wt%Ti) and high SiC (? 80wt%SiC) contents in the alloying powder mixtures. The melting points of Ni, Ti and SiC are 1453, 1668 and 2730?C respectively. The surface temperature achieved during laser processing is 2362?C (see calculations in RESULTS: SYNTHESIS OF THE ALLOYS 73 Chapter 6, Section 6.3). Due to the high melting points of Ti and SiC, some material did not dissolve during laser processing, especially in layers which contained high proportions thereof (refer to Figures 4.12 (d-f)). Layers formed with high Ni contents (Figures 4.12 (b-c) did not form thermo-affected layers since all the Ni powder dissolved in the melt pool due to the lower melting temperature of Ni. The width of the alloyed layer track was approximately equal to the laser beam spot size (or laser beam diameter) of 4mm. RESULTS: SYNTHESIS OF THE ALLOYS 74 Figure 4.12: Stereo micrographs showing the alloyed layers (AL) and thermo- affected layers (TL) for samples alloyed with powders containing (a) 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC, (b) 80wt%Ni + 15wt%Ti and 5wt%SiC, (c) 50wt%Ni + 20wt%Ti + 30wt%SiC, (d) 5wt%Ni + 80wt%Ti + 15wt%SiC, (e) 30wt%Ni + 50wt%Ti + 20wt%SiC , (f) 15wt%Ni + 5wt%Ti + 80wt%SiC and (g) 20wt%Ni + 30wt%Ti + 50wt%SiC at 10 mm/s laser scanning speed. The thickness of each layer is reported. (a) (b) (d) (c) (e) (f) (g) AL =2.16mm AL = 2.37mm AL = 2.40mm AL = 0.91mm TL = 0.77mm AL = 1.90mm TL = 1.07mm AL = 1.67mm AL = 0.65mm TL = 1.83mm Length: 2.16mm Length: 1.90mm Length: 2.37mm Length: 1.07mm Length: 1.68mm Length: 0.65mm Length: 2.40mm RESULTS: SYNTHESIS OF THE ALLOYS 75 Figure 4.13 shows the single-track alloyed layers for samples alloyed using a 20 mm/s laser scanning speed. High laser scanning speeds result in low heat input and limited reaction times. This results in partial mixing of the powders in the melt pool as observed in Figure 4.13(b). The vestiges of convective flow lines induced in the melt pool by the Marangoni flow are observed by the coiled patterns in all the microstructures. Figure 4.13: Stereo micrographs of samples alloyed with (a) 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC, (b) 80wt%Ni + 15wt%Ti + 5wt%SiC, (c) 50wt%Ni + 20wt%Ti + 30wt%SiC, (d) 20wt%Ni + 30wt%Ti + 50wt%SiC at 20 mm/s laser scanning speed. 4.3.2 Hardness of the alloyed layers As stated in Chapter 4.1.3, the average hardness of aluminium AA1200 is 24.0? 0.4HV0.1. Aluminium AA1200 was also laser treated without any powder input to determine the effect of laser treatment on the hardness of the metal. Figure 4.14 shows a through-thickness hardness profile of the laser treated aluminium as (a) (d) (c) (b) Coiled patterns induced by convective flow lines Thickness = 1.74mm Thickness = 1.77mm Thickness = 2.25mm Thickness = 2.70mm RESULTS: SYNTHESIS OF THE ALLOYS 76 measured from the surface to the untreated base material. The average hardness increased to 30.9?2.3HV0.1 due to grain refinement. Grain refinement occurs due to the rapid heating and cooling rates (103 ? 1010 K/s) associated with laser alloying [16,21,27,124]. Figure 4.14: Hardness profile of the laser treated aluminium AA1200. Through-thickness hardness profiles were also constructed for the alloyed materials. Figure 4.15 shows hardness profiles for the surfaces alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC powder at various laser scanning speeds. The average hardness decreased as the laser scanning speed increased. From 195.1?3.9HV0.1 for a 10mm/s scanning speed to 171.2?8.3HV0.1 for a 12mm/s scanning speed to 161.7?9.7HV0.1 for a 15mm/s scanning speed and finally down to 138.0?6.1HV0.1 for a 20mm/s scanning speed. The overall increase in hardness compared to that of untreated aluminium was due to the formation of the intermetallic phases and the metal matrix composites (refer to Chapter 4.5.1). The hardness decreased as the scanning speed increased because low scanning speeds (e.g. 10mm/s) give sufficient time for the powders to melt and react to form the intermetallic phases and the metal matrix composites. High scanning speeds (e.g. 20mm/s) leads to incomplete mixing and un-dissolved alloying particles. The depth of alloying also decreased as the scanning speed increased because less energy is absorbed at high scanning speeds. Through-thickness hardness profiles for aluminium surfaces alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC, 50wt%Ni 0 10 20 30 40 50 0 1000 2000 3000 4000V ic ke rs H ar dn es s (H V 0 . 1) Position (?m) Laser treated Base RESULTS: SYNTHESIS OF THE ALLOYS 77 + 20wt%Ti + 30wt%SiC and 20wt%Ni + 30wt%Ti + 50wt%SiC followed similar trends. Different phases were formed in different regions within the alloyed layer due to the differences in density and melting points of Ni, Ti and SiC. These differences result in fluctuations of the measured hardness profile values. This will be discussed in further detail in Chapter 4.5. Figure 4.15: Hardness profile of surface alloyed with powder containing 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC at various laser scanning speeds. Table 4.5 summarizes the hardness results for the aluminium surfaces alloyed using the various laser scanning speeds (10mm/s, 12mm/s, 15mm/s and 20mm/s). For the same powder composition, hardness increased with decreasing laser scanning speed. Since the highest hardness and a homogenous microstructure were achieved when alloying with a 10mm/sec laser scanning speed, this speed was chosen as the optimum processing speed for further laser alloying tests. RESULTS: SYNTHESIS OF THE ALLOYS 78 Table 4.5: Hardness of the alloyed layers at different scanning speeds. Powder Composition Hardness (HV0.1) Laser scanning speed 10mm/s 12mm/s 15mm/s 20mm/s 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC 195.1 ?3.9 171.2?8.3 161.7?9.7 138.0?6.1 80wt%Ni + 15wt%Ti + 5wt%SiC 312.8?12.9 298.6?16.8 286 ? 13.7 254 ? 14.2 50wt%Ni + 20wt%Ti + 30wt%SiC 244.4?16.0 212?11.3 196?16.8 184?12.2 20wt%Ni + 30wt%Ti + 50wt%SiC 152.1?21.4 138.7?11.7 127.5?15.7 117.1?12.4 4.4 Laser alloying of the breakdown alloy systems Due to the complexity of the Al-Ti-Ni-SiC alloy system investigated in this project, laser surface alloying was first conducted on the breakdown systems of the alloy system. Studies on these systems would assist in understanding the microstructures and phases formed in the Al-Ni-Ti-SiC system. The laser parameters used to produce these breakdown systems were those optimized in Chapter 4.3 above. Ten breakdown systems were studied namely: ? Al-Ni ? Al-Ti ? Al-Si ? Al-TiC ? Al-SiC ? Al-Ni-Ti ? Al-Ni-Si ? Al-Ti-Si ? Al-Ni-SiC ? Al-Ti-SiC RESULTS: SYNTHESIS OF THE ALLOYS 79 The results for each system are presented in separate sections. The layers observed for the samples were free of cracks. 4.4.1 Laser alloying Al with Ni Aluminium AA1200 was laser alloyed with the Ni powder shown in Figure 4.3. During laser processing, alloyed layers with a thickness of 1.3mm were achieved. Figure 4.16 is the microstructure of the alloyed layer showing ?-Al (black) and Al-Ni (grey) eutectic phases. EDS analysis revealed that the grey ?flower-like? structures consisted of 58at%Al and 42at%Ni. From the Al-Ni phase diagram in Figure 2.4, this phase appears to be the hexagonal Al3Ni2 which was confirmed with XRD analysis (Figure 4.17). This Al3Ni2 intermetallic phase was formed in- situ by the reaction of Al and Ni. The Al3Ni2 phase was also reported by various authors [30,39] but the Al3Ni phase reported in [38-41] was not observed under these processing conditions. An improvement in hardness from 24.0?0.4HV0.1 (Al AA1200) to 766.9?38.5HV for the alloyed layer was achieved. The increase in hardness was attributed to grain refinement and the Al3Ni2 phase formed during laser processing. Figure 4.16: SEM micrograph of Al AA1200 laser alloyed with Ni powder showing an in-situ formed Al3Ni2 phase (grey) within an ?-Al (black) matrix. RESULTS: SYNTHESIS OF THE ALLOYS 80 Figure 4.17: XRD pattern of Al laser alloyed with Ni showing Al, Ni and Al3Ni2 phases. 4.4.2 Laser alloying Al with Ti Aluminium AA1200 was laser alloyed with the Ti powders shown in Figure 4.6. The thickness of the alloyed layer was 2.5mm. Figure 4.18 is the microstructure of the alloyed layer revealing in-situ formed dendritic structures. EDS analysis showed that the dendrites consisted of 78at%Al and 22at%Ti. According to the Al-Ti phase diagram in Figure 2.6, the phase formed in this region is the tetragonal Al3Ti intermetallic phase. This was confirmed by XRD analysis of the alloyed layer (Figure 4.19). The Al3Ti intermetallic phase was also observed by Wendt et al. [51]. The microstructure in Figure 4.18 is therefore a ?-Al matrix with a Al-Ti eutectic phase. The hardness of the alloyed layer was approximately 159.2?17.5HV0.1. This hardness is less than the 766.9?38.5HV0.1 achieved with the Al3Ni2 intermetallic phase when alloying Al with Ni. RESULTS: SYNTHESIS OF THE ALLOYS 81 Figure 4.18: SEM micrograph of an Al laser alloyed with Ti powder showing ?-Al (black) and Al3Ti (grey) phases. Figure 4.19: XRD pattern of an Al laser alloyed with Ti showing Al, Ti and Al3Ti phases. RESULTS: SYNTHESIS OF THE ALLOYS 82 4.4.3 Laser alloying Al with Si The alloying of Aluminium AA1200 with Si powder resulted in an alloy thickness of 0.8mm. The microstructure of the alloyed layer shown in Figure 4.20 reveals a ?-Al (black) matrix and an Al-Si eutectic (white) phase. XRD analysis (Figure 4.21) confirmed the Al-Si phase. The hardness of the alloyed layer was 39.9?3.4HV0.1. This is slightly harder than the aluminium AA1200 (24.0?0.4HV0.1) and the laser treated Al layer (30.9?2.3HV0.1). The increase in hardness is attributed to grain refinement and the formation of the Al-Si eutectic phase. Figure 4.20: SEM micrograph of an Al laser alloyed with Si showing ?-Al (black) and Al-Si eutectic (white) phases. RESULTS: SYNTHESIS OF THE ALLOYS 83 Figure 4.21: XRD of an Al laser alloyed with Si powder showing Al and Si phases. 4.4.4 Laser alloying Al with TiC Aluminium AA1200 was laser alloyed with TiC powder. The thickness of the alloyed layer was 0.9mm. A metal matrix composite (?-Al and Al3Ti) layer reinforced with TiC was formed in the alloyed layer as shown in Figure 4.22. EDS showed that the retained particles were 49at%Ti and 51at%C. XRD analysis, Figure 4.23, confirmed the presence of all three phases. The hardness of the matrix in the alloyed layer is 58.0 ? 9.0HV0.1. RESULTS: SYNTHESIS OF THE ALLOYS 84 Figure 4.22: SEM micrograph of an Al laser alloyed with a TiC powder showing TiC particles, ?-Al (black) and an Al3Ti (white) phase. Figure 4.23: XRD of Al laser alloyed with TiC showing Al, Al3Ti and TiC phases. TiC RESULTS: SYNTHESIS OF THE ALLOYS 85 4.4.5 Laser alloying Al with SiC Laser alloying aluminium AA1200 was performed with the SiC powders shown in Figure 4.9. The thickness of the alloyed layer was 1.9mm. Figure 4.24 is a SEM micrograph of the alloyed layer showing that a metal matrix composite reinforced with SiC particles was formed. Due to the high temperatures (2362?C) generated during laser alloying, some of the SiC particles dissociated to form Si and C. Al reacted with Si and C to form the Al4SiC4 phase. EDS analysis showed that the Al4SiC4 phase consisted of 43at%Al + 11at%Si + 46at%C. It has been reported [71-75] that the temperature required for this reaction to occur is above or equal to 1670K. The in-situ reaction of Al and SiC is shown below: 4Al + 4SiC ? Al4SiC4 + 3Si (Reaction 2.3) The free Si produced during this reaction is observed in Figure 4.24. The Al4C3 phase which forms at temperatures between 940 ? 1620K was not observed. This indicates that the temperature of the melt pool was above 1620K. Figure 4.25 shows the XRD analysis of the alloyed layer confirming that the phases observed are Al, Si, Al4SiC4 and SiC. The matrix consisted of ?-Al and Al-Si eutectic phases. The hardness of the alloyed layer is 238.3?33.7HV. RESULTS: SYNTHESIS OF THE ALLOYS 86 Figure 4.24: SEM micrograph of an Al laser alloyed with SiC powder showing SiC particle (black particle), Al4SiC4 intermetallic phase (dark grey platelets), Si phase (white), ?-Al (grey) and Al-Si eutectic phase (white dots in the grey phase). Figure 4.25: XRD of an Al laser alloyed with SiC showing Al, SiC, Si and Al4SiC4 phases. RESULTS: SYNTHESIS OF THE ALLOYS 87 4.4.6 Laser alloying Al with Ni and Ti Laser alloying aluminium AA1200 was performed with powders containing different weight ratios of Ni and Ti (30wt%Ni +70wt%Ti, 50wt%Ni + 50wt%Ti and 70wt%Ni + 30wt%Ti). The microstructures of the alloyed layers are shown in Figure 4.26. Layers alloyed with 30wt%Ni + 70wt%Ti resulted in the formation of Al3Ti and Al3Ni phases as shown in Figure 4.26(a). The Al3Ti phase was more dominant in the microstructure. Alloying with 50wt%Ni + 50wt%Ti resulted in the formation of Al3Ni and Al3Ti phases as shown in Figure 4.26(b). The Al3Ni and Al3Ti phases were equally present in the microstructure. Laser alloying with 70wt%Ni + 30wt%Ti resulted in the formation of Al3Ni, Al3Ni2 and Al3Ti phases as shown in Figure 4.26(c). The Al-Ni intermetallic phases (i.e. Al3Ni and Al3Ni2) were more dominant in the microstructure. All the phases were identified using EDS, XRD and phase diagrams. The XRD pattern for a surface alloyed with 70wt%Ni + 30wt%Ti is shown in Figure 4.27. The thickness of the alloyed layer was 2.81mm when alloying with 30wt%Ni + 70wt%Ti, 2.52mm when alloying with 50wt%Ni + 50wt%Ti and 2.61mm when alloying with 70wt%Ni + 30wt%Ti. The hardness of the alloyed surface was 139?21.5HV0.1 when alloying with 30wt%Ni + 70wt%Ti, 220?14.7HV0.1 when alloying with 50wt%Ni + 50wt%Ti and 400?23.4HV0.1 when alloying with 70wt%Ni + 30wt%Ti. Hardness increased with increasing Ni content in the Ni-Ti powder mixture. This shows that alloyed surfaces with Al-Ni intermetallic phases (Al3Ni and Al3Ni2) are harder than those with Al3Ti phases. The results for the microstructure and hardness of AA1200 laser alloyed with Ni and Ti are summarized in Table 4.6. RESULTS: SYNTHESIS OF THE ALLOYS 88 Figure 4.26: SEM micrograph of an Al laser alloyed with (a) 30wt%Ni + 70wt%Ti, (b) 50wt%Ni + 50wt%Ti and (c) 70wt%Ni + 30wt%Ti. Figure 4.27: XRD of an Al laser alloyed with 70wt%Ni + 30wt%Ti. (a) (c) (b) Al3Ti Al3Ti Al3Ni Al3Ni Al3Ni2 Al3Ni Al3Ti RESULTS: SYNTHESIS OF THE ALLOYS 89 Table 4.6: Phases observed and hardness for Al samples laser alloyed with Ni+Ti. Powder composition Phases observed Hardness (HV0.1) 30wt%Ni + 70wt%Ti ?-Al, Al3Ti and Al3Ni 139?21.5 50wt%Ni + 50wt%Ti ?-Al, Al3Ni and Al3Ti 220?14.7 70wt%Ni + 30wt%Ti ?-Al, Al3Ni, Al3Ni2 and Al3Ti 400?23.4 4.4.7 Laser alloying Al with Ni and Si Aluminium AA1200 was laser alloyed with a powder containing 50wt%Ni and 50wt%Si. The thickness of the alloyed layer was 1.1mm. Figure 4.28(a) shows the microstructure revealing Al3Ni and Al3Ni2 phases. The reaction between Ni and Si resulted in the formation of a NiSi2 intermetallic phase shown in Figure 4.28(b). This is thermodynamically the most stable phase above 750?C in the Ni-Si phase diagram [125]. Due to the high heating rates associated with laser alloying, Ni2Si and NiSi which form at approximately 200?C and 350?C respectively did not form in the alloyed layer. The XRD pattern of the alloyed layer is shown in Figure 4.29 confirming the phases were Al, Si, Al3Ni, Al3Ni2 and NiSi2. The matrix consisted of ?-Al and an Al-Si eutectic. The hardness of the alloyed layer is 216.6?22.8HV. Figure 4.28: SEM micrographs of Al laser alloyed with 50wt%Ni + 50wt%Si showing (a) Al-Ni intermetallic phases and (b) Al, Al3Ni and NiSi2 phases. ?-Al Al3Ni2 Al3Ni NiSi2 (a) (b) Al3Ni RESULTS: SYNTHESIS OF THE ALLOYS 90 Figure 4.29: XRD of Al laser alloyed with 50wt%Ni + 50wt%Si. 4.4.8 Laser alloying Al with Ti and Si Aluminium AA1200 was laser alloyed with a powder containing 50wt%Ti and 50wt%Si. The thickness of the alloyed layer was 1.1mm and Figure 4.30 shows the microstructure. The reaction of Al and Ti resulted in an in-situ formation of an Al3Ti intermetallic phase. The matrix consisted of ?-Al and an Al-Si eutectic. The reaction of Ti with Si resulted in the formation of a Ti5Si3 phase. This phase is thermodynamically the most stable in the Ti-Si phase diagram [126]. These phases were confirmed with XRD (Figure 4.31). The hardness of the alloyed layer is 120.9?19.3HV0.1. RESULTS: SYNTHESIS OF THE ALLOYS 91 Figure 4.30: SEM micrographs of Al laser alloyed with 50wt%Ti + 50wt%Si showing (a) Al3Ti phase and (b) Al-Si eutectic. Figure 4.31: XRD of Al laser alloyed with 50wt%Ti + 50wt%Si. 4.4.9 Laser alloying Al with Ni and SiC Laser alloying of aluminium AA1200 with a powder containing Ni and SiC of different ratios (30wt%Ni +70wt%SiC, 50wt%Ni + 50wt%SiC and 70wt%Ni + 30wt%SiC) resulted in microstructures of the alloyed layers as shown in Figure 4.32. Figure 4.32(a) is a typical optical micrograph of the alloyed surface showing an alloyed layer (AL) and a thermo-affected layer (TL). As previously stated, an (a) (b) Al-Si eutectic Ti5Si3 Al3Ti Al3Ti RESULTS: SYNTHESIS OF THE ALLOYS 92 alloyed layer (AL) represents a layer filled with alloyed microstructure and a thermo-affected layer (TL) is a heat treated layer with no (or little) alloyed microstructure. The alloyed layer formed at the surface of the alloyed material and consisted of a homogeneous microstructure. The thickness of the heat treated layer (alloyed layer + thermo-affected layer) was approximately 4.9mm for all these alloyed surfaces. Figure 4.32(b) is a laser alloyed 30wt%Ni + 70wt%SiC layer showing a SiC particle, Al4C3 phase and an Al-Si eutectic. Due to the high temperatures reached in the alloyed layer, some of the SiC particles dissociated and reacted with Al to form a needle-like Al4C3 phase. An Al3Ni phase was formed from the reaction of Al with Ni. The Al4C3 phase was more dominant in the alloyed layer compared to the Al3Ni dendritic phase. Figure 4.32(c) is the 50wt%Ni + 50wt%SiC alloyed layer showing a SiC particle, Al3Ni2 phase and an Al-Si eutectic. Al3Ni formed from a peritectic reaction between Al and Al3Ni2. The dissociation of SiC particles resulted in the formation of an Al4C3 phase. Figure 4.32(d) is a 70wt%Ni + 30wt%SiC layer showing Al4C3, Al-Si eutectic and Al3Ni2 phases. The Al-Ni intermetallic phases (Al3Ni2 and Al3Ni) increased as the Ni content in the Ni-SiC powder mix increased, while the Al4C3 phase was dominant when the SiC content was high. The phases in the microstructures were identified using XRD (Figure 4.33). The hardness of the alloyed surface was 104?11.5HV0.1 when alloying with 30wt%Ni + 70wt%SiC, 110?10.4HV0.1 when alloying with 50wt%Ni + 50wt%SiC and 138?17.2HV0.1 when alloying with 70wt%Ni + 30wt%SiC. This represents a 5 times increase in hardness compared to the base aluminium AA1200. Increasing the Ni content and decreasing the SiC content in the alloying powder mixture did not have a significant effect on the hardness. The results for the phases and the hardness of Al A1200 laser alloyed with Ni and SiC are summarized in Table 4.7. RESULTS: SYNTHESIS OF THE ALLOYS 93 Figure 4.32: SEM micrograph of an Al laser alloying with (a & b) 30wt%Ni + 70wt%SiC, (c) 50wt%Ni + 50wt%SiC and (d) 70wt%Ni + 30wt%SiC. In (a) AL is the alloyed layer and TL is the thermo-affected layer. Figure 4.33: XRD of an Al surface laser alloyed with 70wt%Ni + 30wt%SiC. (a) (d) (c) (b) AL TL Base Al4C3 SiC Al-Si eutectic Al3Ni2 Al-Si eutectic SiC Al4C3 Al3Ni2 Al-Si eutectic Thickness = 4.6mm RESULTS: SYNTHESIS OF THE ALLOYS 94 Table 4.7: Phases and hardness for Al surfaces laser alloyed with Ni + SiC. Powder composition Phases observed Hardness (HV0.1) 30wt%Ni + 70wt%SiC ?-Al, Al3Ni, SiC, Al4C3 & Al-Si eutectic 104?11.5 50wt%Ni + 50wt%SiC ?-Al, Al3Ni, Al3Ni2, SiC, Al4C3 & Al-Si eutectic 110?10.4 70wt%Ni + 30wt%SiC ?-Al, Al3Ni, Al3Ni2, SiC, Al4C3 & Al-Si eutectic 138?17.2 4.4.10 Laser alloying Al with Ti and SiC Laser alloying aluminium AA1200 with powders containing Ti and SiC of different ratios (30wt%Ti +70wt%SiC, 50wt%Ti + 50wt%SiC and 70wt%Ti + 30wt%SiC) led to the microstructures shown in Figure 4.34. A homogeneous layer was formed as shown in Figure 4.34(a). All the phases in the microstructures were identified using XRD, EDS and phase diagrams. Figure 4.34(b) is a 30wt%Ti + 70wt%SiC layer showing a SiC particle, Al4SiC4, Ti5Si3, Al3Ti and Al phases. Again due to high laser alloying temperatures, some of the SiC particles dissociated and reacted with Al and Ti to form different intermetallic phases. The reaction of Al with Si and C (from SiC) resulted in the formation of the Al4SiC4 phase and free Si, as shown in reaction 2.3. The Al4C3 phase which forms at lower temperature (between 940 ? 1620K) compared to Al4SiC4 (above 1670K) was not formed. This was also observed when Al AA1200 was alloyed with only SiC particles (see Section 4.4.5). The free Si reacted with Ti to form the Ti5Si3 phase. This is the most stable phase in the Ti-Si system as it has the lowest energy of formation [126]. Al reacted with Ti to from the Al3Ti intermetallic phase. Figure 4.34(c) is a 50wt%Ti + 50wt%SiC layer. The phases identified by XRD (Figure 4.35) are Al, SiC, TiC, Ti5Si3 and Al3Ti. The dissociation of SiC particles resulted in the formation of TiC and Ti5Si3 phases. The interfacial TiC phase was RESULTS: SYNTHESIS OF THE ALLOYS 95 formed around the SiC particles due to adsorption of Ti on the SiC particle surface. As the Ti content in the alloying powder was further increased to 70wt%, an increase in the Al3Ti phase was observed in the microstructure as shown in Figure 4.34(d). The Ti5Si3 phase was also observed from the dissociation of SiC and the reaction of Si with Ti. The hardness of the alloyed surface was 81.8?8.5HV0.1 when alloying with 30wt%Ti + 70wt%SiC, 111.3?20.3HV0.1 when alloying with 50wt%Ti + 50wt%SiC and 149.0?21.3HV0.1 when alloying with 70wt%Ti + 30wt%SiC. These results show that the hardness increases as the Ti content increased because of the increase of the hard and brittle Al3Ti phase in the microstructure. The results for samples laser alloyed with Ti and SiC of different weight ratios are summarized in Table 4.8. Figure 4.34: Micrographs of an Al laser alloyed with (a & b) 30wt%Ti + 70wt%SiC, (c) 50wt%Ti + 50wt%SiC and (d) 70wt%Ti + 30wt%SiC. (a) (b) (c) (d) SiC Al4SiC4 Ti5Si3 Al3Ti ?-Al SiC Ti5Si3 Al3Ti TiC ?-Al Ti5Si3 SiC Al3Ti RESULTS: SYNTHESIS OF THE ALLOYS 96 Figure 4.35: XRD of an Al surface laser alloyed with 50wt%Ni + 50wt%SiC. Table 4.8: Phases and hardness for Al samples laser alloyed with Ti and SiC. Powder composition Phases observed Hardness (HV0.1) 30wt%Ti + 70wt%SiC ?-Al, SiC, Al4SiC4, Ti5Si3 & Al3Ti 81.8?8.5 50wt%Ti + 50wt%SiC ?-Al, SiC, TiC, Ti5Si3 & Al3Ti 111.3?20.3 70wt%Ti + 30wt%SiC ?-Al, SiC, TiC, Ti5Si3 & Al3Ti 149.0?21.3 4.5 Laser alloying Al with Ni, Ti and SiC Aluminium AA1200 surfaces were laser alloyed with powders containing Ni, Ti and SiC in different ratios as listed in Table 3.3. The processing conditions were the same as those listed in Table 3.2 using a laser scanning speed of 10mm/s as optimized in Chapter 4.3. The width of the single track alloyed layers was approximately equal to the beam diameter of 4mm. The thickness of the layers produced varied between 1.8 to 2.8mm. The microstructure of alloyed layers that did not form a thermo-affected layer was further characterized to determine the phases present. Of the 37 alloys listed in Table 3.3 only 14 alloys met this RESULTS: SYNTHESIS OF THE ALLOYS 97 criterion. Detailed microstructural analysis of the 14 alloys is given in Appendix A. In this section a summary of the results of all 14 alloys is presented. Table 4.9 lists the phases that were detected in all 14 alloys using XRD, while Table 4.10 lists the phases that were only found in selected alloys, also using XRD. The XRD patterns for all the alloys as well as microstructure images reflecting the phases are given in Appendix A. Table 4.9: Phases found in all the Al surfaces laser alloyed with Ni, Ti and SiC. ?-Al SiC Al3Ni Al3Ti Table 4.10: Phases found in selected Al surfaces laser alloyed with Ni, Ti and SiC. Phases Results for Alloys TiC, Ti5Si3 Found in all alloys except 80wt%Ni + 15wt%Ti + 5wt%SiC and 10wt%Ni + 70wt%Ti + 20wt%SiC Al3Ni2 Found in all alloys except 10wt%Ni + 70wt%Ti + 20wt%SiC and 20wt%Ni + 40wt%Ti + 40wt%SiC Al4C3 Only found in alloys 50wt%Ni + 20wt%Ti + 30wt%SiC and 70wt%Ni + 10wt%Ti + 20wt%SiC Ti3SiC2 Only found in alloys 40wt%Ni + 20wt%Ti + 40wt%SiC, 40wt%Ni + 30wt%Ti + 30wt%SiC, 20wt%Ni + 40wt%Ti + 40wt%SiC, 30wt%Ni + 40wt%Ti + 30wt%SiC, 40wt%Ni + 40wt%Ti + 20wt%SiC and 30wt%Ni + 30wt%Ti + 40wt%SiC Al4SiC4 Only found in alloys 20wt%Ni + 30wt%Ti + 50wt%SiC, 40wt%Ni + 20wt%Ti + 40wt%SiC, 20wt%Ni + 40wt%Ti + 40wt%SiC, 30wt%Ni + 40wt%Ti + 30wt%SiC and 30wt%Ni + 30wt%Ti + 40wt%SiC Si Only found in alloys 20wt%Ni + 40wt%Ti + 40wt%SiC, 30wt%Ni + 40wt%Ti + 30wt%SiC and 30wt%Ni + 30wt%Ti + 40wt%SiC Al3TiC2 Only found in alloy 20wt%Ni + 30wt%Ti + 50wt%SiC RESULTS: SYNTHESIS OF THE ALLOYS 98 Due to different densities, Ni (? = 8.91 g/cm3), Ti (? = 4.51 g/cm3) and SiC (? = 3.21 g/cm3) reacted with the Al at different positions within the alloyed layer. Generally the SiC reacted with molten Al close to the surface while Ni reacted with Al near the middle of the alloyed layer. The needle-like Al3Ti phase was observed throughout the microstructure. The formation of an aluminium SiC-embedded metal matrix composite (MMC) depended on the amount of un-dissolved SiC particles in the alloyed layers which in turn depended on the amount of SiC added to each alloy as well as the size of the SiC particles. All 14 alloys formed a MMC but the alloy to which 5wt%SiC was added did not form a good MMC as few SiC particles were retained in the alloyed layer. Figure 4.36 shows the effect of SiC content on the microstructure of two alloys. It also appeared as though the addition of Ti influenced the formation of MMCs. When Ti was added together with low amounts of SiC (10 ? 20wt%), the amount of retained SiC particles decreased. Figure 4.36: Effect of SiC content on alloys (a) 80wt%Ni + 15wt%Ti + 5wt%SiC and (b) 20wt%Ni + 30wt%Ti + 50wt%SiC. A good MMC is reflected in alloy (b). Due to the high surface temperatures achieved during laser alloying, some of the SiC powder particles dissolved in the melt pool. The dissolved SiC particles dissociated to form Si and C. The C reacts with either Ti or Al to form TiC or the brittle Al4C3 phase. The Gibbs free energy is more negative for the formation of TiC and thus there was a higher tendency for the formation of TiC than for Al4C3. Two types of TiC phases were observed, namely dendritic and interfacial TiC as (a) (b) RESULTS: SYNTHESIS OF THE ALLOYS 99 indicated in Figure 4.37. The interfacial TiC phases were formed around SiC particles due to adsorption of Ti on the SiC surface. Due to the high cooling rates associated with laser treatment, these TiC phases did not grow or aggregate. The dendritic TiC phases were formed inside the melt pool from the reaction of the dissolved SiC particles and Ti. TiC formed in all except two alloys. The brittle Al4C3 phase was only detected in two alloys which both had low Ti and SiC contents. The formation of the Al4C3 phase can be controlled by (1) selecting a suitable laser energy density to reduce the melting temperature and restrain the melting and dissolution of SiC particles, (2) using a large laser scanning speed to shorten the melt pool time and weaken the reaction of SiC and Al, and (3) increasing appropriately the amount of SiC to restrain the nucleation and growth of the Al4C3 phase [73]. RESULTS: SYNTHESIS OF THE ALLOYS Figure 4.37: Alloy 33.3 phases observed. The dissociated Si and reaction with Ti generally formed the thermodynamically favourable Ti5Si3 phase which has the lowest Gibbs free energy compared to other Ti-Si phases such as TiSi, TiSi increased the formation of Ti few alloys traces of Ti Al3Ni2 Al3Ni Al3Ni2 SiC Al3Ti (a) (b) wt%Ni + 33.3wt%Ti + 33.3wt% SiC showing 2 and Ti5Si4. Increasing the level of Ti 5Si3. In three alloys free Si was detected while in a 3SiC2 and Al4SiC4 were found. 100 the different added TiC SiC TiC ?-Al Al3Ti Al3Ti ?-Al Edge of the surface Ti5Si3 RESULTS: SYNTHESIS OF THE ALLOYS Due to the high Al content in the base material compared to the Ti and Ni content, the Ni-rich and Ti-rich Al Al with Ni generally the reaction of Al with Ti Al3Ti (refer to Figure 4.38). When high Ti contents were added the microstructure was dominated by the formation of the Al contents were added the microstructure was dominated by the formation of usually the Al3Ni phase and occasionally the Al formed in the Al rich side of the Al forms from a eutectic r was seen to envelope the Al between the liquid and the Al Figure 4.38: Alloy 70wt%Ni + 20wt%Ti + 10wt%SiC and the needle-like Al Al3Ti -intermetallic phases were not formed. resulted in the formation of the dendritic Al generally resulted in the formation of 3Ti intermetallic phase. 3Ni2 phase. These phases are -Ni phase diagram. The dendritic Al eaction between Ni and liquid Al. The Al3 3Ni2 phase was formed by a peritectic reaction 3Ni2 phase. showing the dendritic Al 3Ti intermetallic phases. 101 The reaction of 3Ni phase while the needle-like When high Ni 3Ni phase Ni phase which 3Ni Al3Ni RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 102 5. RESULTS: MECHANICAL AND WEAR TEST RESULTS OF THE ALLOYS The 14 alloys described in Chapter 4, Section 4.5 were subjected to abrasion and impact testing. The results of these tests are presented in this chapter. The surface hardness, as well as hardness profiles through cross-sections of each alloy, was also determined and the results described here. 5.1 Hardness Through-thickness hardness profiles (indentations from the surface of the alloyed layer through to the base) were constructed using a 100g indenting load on polished cross-sections of the alloyed surfaces. The spacing between consecutive indentations was 100?m. Figure 5.1 shows a typical hardness profile for an aluminium surface alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC. The average hardness of the alloyed layer (region A) is 195.1 ? 3.9HV0.1. The hardness is highest in the alloyed layer (region A) showing a sharp decrease in the interface region (region B) between the alloyed layer and the pure aluminium and is lowest in the pure aluminium (region C). This confirms that laser alloying improved the surface hardness. The improvement in hardness is related to the formation of hard intermetallic phases and metal matrix composites as well as grain refinement due to the rapid heating and cooling rates associated with laser alloying. During alloying a temperature gradient is formed in the material due the heat source (the laser beam) and the heat sink, namely the substrate, which acts to extract heat from the surface. This temperature gradient together with the density variations between Al (2.7g/cm3), Ni (8.91g/cm3), Ti (4.51g/cm3) and SiC (3.21g/cm3) produces a strong convective flow in the melt pool which in turn results in the formation of different phases (see Chapter 4, Section 4.5). This in turn leads to the small fluctuations in hardness observed in region A. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 103 Figure 5.1: Hardness profile through cross-section of aluminium alloyed with 33.3wt%Ni, 33.3wt%Ti and 33.3wt%SiC. (A) is the alloyed layer, (B) is the interface between the alloyed layer and the aluminium substrate and (C) is the Al substrate. The hardness results for all the alloyed cross sections (average hardness of region A in Figure 5.1) are listed in Table 5.1. The surface hardness of these alloyed samples using a 1kg load are also listed in Table 5.1 and illustrated in Figure 5.2. The two hardness values for each alloy are similar within experimental error. The results show that laser alloying improved the surface hardness for all the compositions used. Grain refinement, due to the rapid heating and cooling rates associated with laser alloying plays a role in increasing the hardness of the laser alloyed surfaces [124]. The highest hardness was obtained when alloying with high Ni contents. This is likely due to the high hardness of the Al3Ni (732HV) and Al3Ni2 (1013HV) phases present in this alloy [127]. The density of the equiaxed dendritic Al3Ni2 grains was high resulting in small Al mean free paths between the grains. This limits the contribution of the pure aluminium to the hardness of the high Ni alloys. The Si containing intermetallic phases namely, Ti5Si3, Ti3SiC2 and Al4SiC4 did not increase the hardness significantly due to the low volume fraction of these phases. A B C RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 104 Table 5.1: Hardness of the alloyed layers. Powder Composition Hardness (HV0.1) from Region A Surface Hardness (HV1) Untreated aluminium AA1200 24.0 ? 0.4 24.0 ? 0.3 Laser treated aluminium 30.9 ? 2.3 30.7 ? 4.2 10wt%Ni + 70wt%Ti + 20wt%SiC 136.5 ? 12.5 143.7 ? 9.3 20wt%Ni + 30wt%Ti + 50wt%SiC 152.1 ? 21.4 149.5 ? 10.1 20wt%Ni + 40wt%Ti + 40wt%SiC 168.3 ? 28.6 172.4 ? 13.6 30wt%Ni + 30wt%Ti + 40wt%SiC 183.0 ? 17.4 184.5 ? 6.9 30wt%Ni + 40wt%Ti + 30wt%SiC 189.0 ? 11.4 192.2 ? 8.2 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC 195.1 ? 3.9 197.4 ? 7.5 40wt%Ni + 20wt%Ti + 40wt%SiC 195.4 ? 28.6 198.2 ? 13.5 40wt%Ni + 30wt%Ti + 30wt%SiC 229.7 ? 17.6 218.9 ? 6.4 40wt%Ni + 40wt%Ti + 20wt%SiC 235.7 ? 21.3 239.7 ? 10.3 50wt%Ni + 20wt%Ti + 30wt%SiC 244.4 ? 16.0 250.1 ? 7.8 60wt%Ni + 30wt%Ti + 10wt%SiC 265.3 ? 22.7 264.3 ? 9.4 70wt%Ni + 10wt%Ti + 20wt%SiC 289.6 ? 10.6 290.4 ? 11.9 70wt%Ni + 20wt%Ti + 10wt%SiC 292.5 ? 21.9 295.8 ? 11.2 80wt%Ni + 15wt%Ti + 5wt%SiC 312.8 ? 12.9 318.6 ? 9.1 RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 105 Figure 5.2: Surface hardness of untreated aluminium and the laser alloyed surfaces. 5.2 Two body abrasion wear tests Two body abrasion wear tests were performed on polished surfaces of the alloys according to the procedure outlined in Chapter 3, Section 3.5.1. The pure aluminium and all the alloyed layers displayed the same wear rate response in terms of an initial high wear rate during the first 5s of testing followed by a levelling off of the wear rate for the remainder of the test. Figure 5.3 illustrates this behaviour for several alloys while Figure 5.4 provides a comparison of the wear rates for all the alloys after 24 seconds of wear. Graphs, similar to Figure 5.3, reflecting the wear rates for each alloy are provided in Appendix B1. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 106 Figure 5.3: Wear response of untreated aluminium and laser alloyed surfaces. Figure 5.4: Wear rate of untreated and laser alloyed aluminium surfaces. The results in Figure 5.4 show a reduced average wear rate after laser alloying which is due to the formation of the metal matrix composites and intermetallic phases. The common linear trend of increasing hardness leading to increasing wear resistance was not observed. The 80wt%Ni + 15wt%Ti + 5wt%SiC alloy RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 107 which has the highest hardness did not show the best wear resistance. This alloy has the ninth highest wear rate of the fifteen materials. The best wear resistance was shown by the 20wt%Ni + 30wt%Ti + 50wt%SiC alloy which only had the thirteenth highest hardness. SEM images of the worn surfaces were taken to identify the wear mechanisms. Similar features were found on the wear scars of all the alloys. The degree to which the different types of mechanisms occurred depended on the alloying composition. SEM images of the untreated aluminium AA1200 alloy worn surfaces are shown in Figure 5.5. The wear features observed were grooves formed by the motion of the SiC abrasive grits (Figures 5.5(a) and (b)) and material pile-up due to groove formation (Figure 5.5(b)). Plastic deformation (Figure 5.5(c) showing differences in grey levels) and cracking in the deformed layer (Figure 5.5(d)) were observed in the cross-sections. Cracks which initiated in the deformed layer, and at the interface between the deformed layer and the bulk material, facilitated material removal during the cutting action of the abrasive grits leading to high mass losses. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS Figure 5.5: Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-section Figures 5.6 to 5.8 illustrate the typical wear scars observed on the alloy surfaces. Images of the wear scars all the alloys can be found in Appe Groove formation, material pile were generally observed alloyed surfaces were shallow compared to those formed on the untreated aluminium. This is attributed to intermetallic compounds (a) (c) s. -up and cracking, as shown in Figures 5.6 and 5.7 on all the worn surfaces. The wear grooves in the laser the presence of the hard ceramic particles and which hinder groove formation. (d) (b) 100?m 200?m 108 ed ndix B1. 10?m 50?m RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 109 Figure 5.6: SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at low magnifications. (a) (c) (b) (d) RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 110 Figure 5.7: SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at high magnifications. Figure 5.8 shows SEM images of the cross-sections of several alloys. No sub- surface plastic deformation was observed in the laser alloyed materials due to the presence of the hard ceramic particles and intermetallic compounds. In Figures 5.8(a) and (b), the SiC particles appear to hinder crack growth. Fracturing of the SiC particles (Figure 5.8(c)) and cracking of the intermetallic matrix (Figure 5.8(f)) was also observed. Lateral cracking seen in some of the cross-sectional (a) (d) (c) (b) (e) (f) RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS views was found to be surface layer as seen in Figures 5.8 Figure 5.8: SEM images of alloyed surfaces. 5.3 Three body Three body dry abrasion wear tests were performed on polished surfaces of the alloys according to the procedure outlined in Chapter 3, Section 3.5.2. Silica (SiO2) sand was used as the abrasive material. The sand was sieved to determine (a) (e) (c) transgranular and contributed to chipping of the worn (d) and (e). cross-sections of the worn surfaces of the laser dry abrasion wear tests 10?m 10?m 10?m (f) (d) (b) 111 10?m 10?m 10?m RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 112 the particle size and distribution. Figure 5.9 shows the size distribution curve for the silica sand used. Majority of the sand particles were in the range of 500- 600?m with a D50 of 525?m. The silica particles were generally angular in shape. Figure 5.9: Size distribution curve for the silica sand. The wear response of all the alloys was similar with an initial high wear rate during the first 10 minutes of testing followed by a levelling off of the wear rate for the remainder of the test. Figure 5.10 illustrates this behaviour for several alloys while Figure 5.11 provides a comparison of the wear rates for all the alloys after 60 minutes of wear. Graphs, similar to Figure 5.10, reflecting the wear rates for each alloy are provided in Appendix B2. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 113 Figure 5.10: Wear response of untreated aluminium and laser alloyed surfaces. The results in Figure 5.11 show a reduced average wear rate after laser alloying which is due to the formation of the metal matrix composites and intermetallic phases. Again no correlation was observed between wear response and hardness. The 80wt%Ni + 15wt%Ti + 5wt%SiC alloy which had the highest hardness did not show the best wear resistance. This alloy has the twelfth highest wear rate of the fifteen materials. The best wear resistance was shown by the 40wt%Ni + 20wt%Ti + 40wt%SiC alloy which only had the eighth highest hardness. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 114 Figure 5.11: Wear rate of untreated and laser alloyed aluminium surfaces. SEM images of the worn surfaces were taken to identify the wear mechanisms. Similar features were found on the wear scars of all the alloys. The degree to which the different types of mechanisms occurred depended on the alloying composition. Figures 5.12 to 5.15 illustrate the typical wear scars observed on the metals. Images of the wear scars of all the alloys can be found in Appendix B2. Figure 5.12 shows the wear scars on the pure Al. Due to high ductility of aluminium, some of the SiO2 particles were embedded in the deformed microstructure. Figures 5.12(a) and (b) show cracking and plastic deformation of the surface. During testing the rubber wheel pushes the sand against the specimen leading to point indentation on the aluminium surface by the applied load and the SiO2 particles. The extent of deformation and cracking on the surface is evident on the cross-sections of the aluminium shown in Figures 5.12(c) and (d). Cracks which initiated in the deformed layer facilitated material removal during wear. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS Figure 5.12: Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-section Similar wear features of cracking were observed in the laser surface alloyed materials as shown in Figures 5.13 and 5.14. The SiC particles seem to hinder crack growth and groove formation as seen in Figure 5.14 (a) and (b). shorter compared to those formed during two body abrasion wear degree of plastic deformatio This is attributed to silicon carbide abrasive grits (2500HV) for the two tests (higher in the 3 body (c) (a) s. ploughing, smearing, material pile-up and The wear grooves were shallow and n was greater under three body abrasion conditions. the lower hardness of the silica sand (800HV) compared to as well as the different applied loads used wear tests). (d) (b) 10?m 115 extensive . However the 10?m RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS Figure 5.13: SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at magnifications. (a) (c) (d) (b) 116 low RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 117 Figure 5.14: SEM plan view images of wear scars of the laser alloyed surfaces showing typical features observed in all the alloyed layers. Images taken at high magnifications. Sub-surface plastic deformation occurred in the laser alloyed materials. This was evident in the cross-sections of the worn surfaces where distortions in the microstructures were observed, for example uniform changes in grain orientation in the direction of motion of the abrasive. This deformation is caused by the cyclic action of the abrasive over the surface under the influence of the applied load. Lateral cracking seen in the cross-sectional views was transgranular and (a) (f) (e) (d) (c) (b) RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS contributed to chipping of the worn surface layer as seen in Figures 5.15(a) and (b). Fracturing of the SiC particles (Figures 5.15(c of the intermetallic matrix (Figure 5.15(f)) was observed. Figure 5.15: SEM images of alloyed surfaces. (a) (e) (c) -e)) and transgranular cracking cross-sections of the worn surfaces of the laser 10?m 10?m 10?m (f) (d) (b) 118 10?m 20?m 10?m RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 119 5.4 Impact tests Charpy impact tests were performed on untreated aluminium and laser alloyed samples. The tests were conducted according to the procedure outlined in Chapter 3, Section 3.6. The energy absorbed by aluminium and the laser alloyed samples during fracture was recorded and is listed in Table 5.2. The reported energy is the average of 4 measurements. The data shows that high energy was absorbed during fracture of the ductile aluminium AA1200 metal. The laser alloyed samples absorbed lower energies due to the increased hardness of the surface layers. Table 5.2: Absorbed energy during fracture. Sample Average Energy (J) Aluminium 14.8?0.2 10wt%Ni + 70wt%Ti + 20wt%SiC 10.2?0.2 20wt%Ni + 40wt%Ti + 40wt%SiC 10.0?0.0 40wt%Ni + 40wt%Ti + 20wt%SiC 10.0?0.0 30wt%Ni + 30wt%Ti + 40wt%SiC 10.0?0.0 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC 9.8?0.2 30wt%Ni + 40wt%Ti + 30wt%SiC 9.5?0.4 80wt%Ni + 15wt%Ti + 5wt%SiC 9.5?0.2 20wt%Ni + 30wt%Ti + 50wt%SiC 9.2?0.3 70wt%Ni + 20wt%Ti + 10wt%SiC 9.0?0.0 70wt%Ni + 10wt%Ti + 20wt%SiC 9.0?0.0 40wt%Ni + 30wt%Ti + 30wt%SiC 9.0?0.0 40wt%Ni + 20wt%Ti + 40wt%SiC 8.7?0.2 50wt%Ni + 20wt%Ti + 30wt%SiC 8.5?0.4 60wt%Ni + 30wt%Ti + 10wt%SiC 7.3?0.2 After impact, the pure aluminium samples did not completely fracture but were plastically deformed and simply bent, as shown in Figure 5.16. Therefore, plastic deformation occurred in these surfaces, not fracture, and the absorbed energy RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 120 listed in Table 5.2 for aluminium is due to this plastic deformation and not fracture. Figure 5.16: Untreated aluminium AA1200 after impact test. A typical fractured surface for the laser alloyed samples is shown in Figure 5.17. Three different regions are observed; the laser alloyed layer followed by the aluminium surface and then the notched aluminium. The fractured surface of the aluminium surface region is shown in Figure 5.18. Cuplets characteristic of ductile fracture were observed in this region. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS Figure 5.17: A typical fracture surface for the laser alloyed samples. Figure 5.18: Fracture surface of the aluminium surface showing cuplets. 121 Alloyed layer Al surface Notched Al RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS 122 The typical fracture surfaces of the alloyed layers are shown in Figure 5.19. Images of the fractured surfaces of all the alloys can be found in Appendix C. In alloys with a high SiC content (Figure 5.19(a-c)), crack propagation was promoted by transgranular fracture of these hard SiC particles. Fracture of the SiC particles occurred by cleavage which is characteristic of brittle fracture while the matrix was dull and fibrous which is characteristic of ductile fracture. Cleavage is known to occur due to transgranular fracture [119]. Few cracks were observed in alloys with a low SiC content, but decohesion of the intermetallic phases was observed as shown in Figures 5.19(d-f). This was also observed by Vreeling et al. [120] while studying the failure mechanisms in an Al/SiC metal matrix composite. Cracking of the intermetallic phases was also observed in the fractographs. In general the overall fracture was ductile. RESULTS: WEAR AND MECHANICAL TEST RESULTS OF THE ALLOYS Figure 5.19: SEM fractographs of the high SiC content (? 20wt%). (a) (c) (e) laser alloyed samples. (a-c) are alloys with 20wt%) while (d-f) are alloys with low SiC content ( (b) (d) (f) 123 ? SiC Matrix with intermetallic phases Decohesion of the intermetallic phases DISCUSSION 124 6. DISCUSSION The originality of this research was the laser alloying of aluminium AA1200 with Ni, Ti and SiC powders simultaneously. Optimization of the laser parameters for the formation of homogeneous and crack-free alloyed surfaces; studying the in situ formation of the metal matrix composites and intermetallic compounds; and characterizing the mechanical and tribological properties of the alloyed material were also novel aspects with regard to this specific alloy system. This chapter discusses the results obtained with emphasis on these novel components. 6.1 Optimization of the laser surface alloying process The initial challenge in this work was to produce a good quality alloyed layer on the aluminium surface. A good quality alloyed layer has a homogeneous alloyed microstructure throughout the layer. This layer was achieved by carefully (through trial and error) determining the optimum processing parameters. Only the laser scanning speed was varied while determining the optimum processing parameters. Samples that formed a thermo-affected layer (layer with no alloyed microstructure) due to heterogeneous mixing in the melt pool were not characterized further. As the aim of the research was to modify only the surface of the aluminium AA1200 while retaining the bulk properties, samples with an alloyed thickness (laser alloying depth) greater that 3mm were not characterized. The thickness of the base aluminium was 6mm and therefore a 3mm alloying depth represents 50% of the aluminium base plate. Prior to laser processing the untreated aluminium AA1200 metal was characterised. The major phase observed in the microstructure was ?-Al with traces of FeAl3. The crystal structure of aluminium is face-centered cubic and the density is 2.70?0.01g/cm3. The Ni, Ti and SiC powders were also characterised to determine the particle size and shape. The size of the particles affects the powder efficiency during laser alloying while the shape affects the manner in which they DISCUSSION 125 flow. The SiC powder had a wide particle size distribution compared to the Ni and Ti powders. SiC also had the highest hardness with Ti showing the lowest. Ni powders were spherical while Ti and SiC powders were irregular. Laser alloying Al with Ni, Ti and SiC simultaneously was successfully performed with an Nd:YAG laser. The processing conditions used to determine the optimum conditions are given in Table 3.2. In determining the optimum processing parameters, only the laser scanning speed was varied. The off-axis nozzle and powder feed rate used ensured that sufficient powder was introduced to the laser/material interaction zone. A laser power of 4kW and a laser beam diameter of 4mm were sufficient for the dissolution of the powders. The laser scanning speed affects the heat input during laser alloying. Increasing the laser scanning speed decreases the absorption of radiation on the substrate which results in insufficient heat to melt the substrate and to dissolve the powder particles. The optimum laser scanning speed was determined to be 10mm/s based on the following reasons: head2right At high laser scanning speeds (e.g. 20mm/s), vestiges of convective flow lines induced in the melt pool by the Marangoni flow were observed. These vestiges of convective flow lines show that homogeneous mixing did not occur. head2right Due to the fast cooling rates (103?1010 K/s) associated with laser alloying, high scanning speeds did not give sufficient time for the powders to dissolve in the melt pool. head2right The surface hardness increased as the alloying speed decreased because the powders dissolved and reacted with Al to form hard intermetallic phases. Some of the powders did not dissolve at the high scanning speeds (e.g. 20mm/s). head2right Low scanning speeds (e.g. 10mm/s) resulted in a greater depth of penetration due to the high heat input. This produced alloyed layers with the required thickness. DISCUSSION 126 The study of the in-situ formation of the metal matrix composites and intermetallic compounds was then conducted on the optimized layers. Due to complexity of the Al-Ni-Ti-Si-C system investigated in this project, the breakdown systems were first studied. 6.2 Laser alloying of the breakdown alloy systems A summary of the results obtained from the breakdown systems are shown in Table 6.1. The increase in hardness after laser alloying was attributed to grain refinement and the in-situ formation of phases (intermetallic phases and/or metal matrix composites) within the ?-Al matrix. The highest hardness of 766.9?38.54HV0.1 was achieved when laser alloying with Ni powder. A dense network of equiaxed dendritic Al3Ni2 grains was observed in the microstructure. The ?-Al mean free path (spacing between two consecutive Al3Ni2 intermetallic phases seen in Figure 4.16) was low compared to surfaces alloyed with the Ti or SiC powders. Therefore, the high volume fraction of the Al3Ni2 phase and the low ?-Al mean free path were major contributors to the high hardness of the laser alloyed surface. The dendritic Al3Ti phase formed when laser alloying Al with Ti resulted in a lower hardness (159.2?17.5HV0.1) compared to the effect of the Al3Ni2 phase. The ?-Al mean free path was higher due to the dendritic nature of the Al3Ti phases which resulted in a low hardness due to the increased volume fraction of the ?-Al in the alloyed layer. DISCUSSION 127 Table 6.1: Summary of the results from the breakdown systems. Alloying powder composition Phases observed Hardness (HV0.1) Ni ?-Al & Al3Ni2 766.9?38.5 Ti ?-Al & Al3Ti 159.2?17.5 Si ?-Al & Al-Si eutectic 39.9?3.4 TiC ?-Al, TiC & Al3Ti 58.0 ? 9.0 SiC ?-Al, SiC, Si, Al4SiC4 & Al-Si eutectic 238.3?33.7 30wt%Ni + 70wt%Ti ?-Al, Al3Ti and Al3Ni 139?21.5 30wt%Ni + 50wt%Ti ?-Al, Al3Ni and Al3Ti 220?14.7 70wt%Ni + 30wt%Ti ?-Al, Al3Ni, Al3Ni2 and Al3Ti 400?23.4 50wt%Ni +50wt%Si ?- Al, Si, Al3Ni, Al3Ni2 and NiSi2 216.6?22.8 50wt%Ti + 50wt%Si ?-Al, Al3Ti and Ti5Si3 & Al-Si eutectic 120.9?19.3 30wt%Ni +70wt%SiC ?-Al, Al3Ni, SiC, Al4C3 & Al-Si eutectic 104?11.5 50wt%Ni + 50wt%SiC ?-Al, Al3Ni, Al3Ni2, SiC, Al4C3 & Al-Si eutectic 110?10.4 70wt%Ni + 30wt%SiC ?-Al, Al3Ni, Al3Ni2, SiC, Al4C3 & Al-Si eutectic 138?17.2 30wt%Ti +70wt%SiC ?-Al, SiC, Al4SiC4, Ti5Si3 & Al3Ti 81.8?8.5 50wt%Ti + 50wt%SiC ?-Al, SiC, TiC, Ti5Si3 & Al3Ti 111.3?20.3 70wt%Ti + 30wt%SiC ?-Al, SiC, TiC, Ti5Si3 & Al3Ti 149.0?21.3 Laser alloying Al with a powder consisting of mixed Ni + Ti in different compositions resulted in the formation of Al3Ni, Al3Ni2 and Al3Ti intermetallic phases. No Ni-Ti phases were present. The hardness of the laser alloyed surfaces increased with increasing Ni content in the powder mixture due to the high DISCUSSION 128 hardness of the Al3Ni (1013HV [127] ) and Al3Ni2(732HV [127] ) phases. The highest hardness of 400?23.4HV0.1 was observed when laser alloying with 70wt%Ni + 30wt%Ti. Si did not form intermetallic phases with Al but was in solid solution in the Al-Si eutectic. This was also observed by other authors [61-62]. The hardness of the laser alloyed surface was 39.9?3.4 HV0.1. The increase in hardness from 24.4?0.4 HV0.1 for the untreated aluminium was attributed to grain refinement and the Al- Si eutectic formed. When Al was laser alloyed with 50wt%Ni + 50wt%Si, Al3Ni, Al3Ni2 and NiSi2 intermetallic phases were formed in the Al-Si eutectic matrix. The Ni and Si in the melt pool resulted in the formation of the thermodynamically stable (above 750?C) NiSi2 intermetallic phase. The Ni2Si and Ni-Si phases which form at lower temperatures did not form due to the high heating and cooling rates associated with laser alloying. A surface hardness of 216.6?22.8HV0.1 was achieved after laser alloying. Al3Ti and Ti5Si3 intermetallic phases were formed when laser alloying aluminium with 50wt%Ti + 50wt%Si. The Ti5Si3 phase is the most thermodynamically stable phase in the Ti-Si phase diagram compared to the TiSi, TiSi2 and Ti5Si4 phases [126]. The surface hardness was 120.9?19.3HV0.1, which is lower than 216.6?22.8 HV0.1 observed when laser alloying with Ni + Si. This highlights the efficiency of the Al-Ni (Al3Ni and Al3Ni2) intermetallic phases to increase the surface hardness compared to the Al3Ti and Ti5Si3 intermetallic phases. Laser alloying aluminium with ceramic materials (TiC and SiC) resulted in the formation of metal matrix composites. A metal matrix composite (MMC) is a composite material consisting of at least two materials with one material being a metal. In this work, ?-Al was the metal matrix and the ceramic (TiC or SiC) was the reinforcement. Due to the high temperatures achieved during laser alloying and the wide size distribution of the powder particles, some of the ceramic particles dissociated and reacted with aluminium to form in-situ products. An aluminium MMC reinforced with TiC particles was formed when laser alloying Al with TiC powder. Some of the TiC particles dissociated and reacted with Al to DISCUSSION 129 form the Al3Ti intermetallic phase. A surface hardness 58.0 ? 9.0 HV0.1 was achieved after laser alloying. This is the hardness of the matrix which consisted of ?-Al and Al3Ti phases with a high volume fraction of the ?-Al phase. A metal matrix composite was also formed when Al was laser surface alloyed with SiC. Some of the SiC particles dissociated in the melt pool due to the high temperatures achieved during laser irradiation and reacted with Al to form Al4SiC4 and free Si. The Al4SiC4 phase is reported to form at temperatures above 1640K [10] while the brittle Al4C3, which forms at lower temperatures (940- 1620K) was avoided due to the high temperatures in the melt pool. The hardness of Al4SiC4 is 1200HV while that of Al4C3 is 300HV [129]. The surface hardness of the matrix (consisting of ?-Al, Si and Al4SiC4 phases) was 238.3?33.7HV0.1. This increased hardness was due to the high hardness of the Al4SiC4 phase [129]. To determine the combined effect of intermetallic phases and metal matrix composites, aluminium was laser alloyed with mixed Ni + SiC powders and mixed Ti + SiC powders of different compositions. Two regions, the alloyed layer and the thermo-affected layer, were observed in the microstructures of surfaces laser alloyed with mixed Ni + SiC powders. The alloyed layer was formed in the region near the surface. The thermo-affected layer was a heat treated layer with no (or little) alloyed microstructure. The phases observed in this layer were ?-Al and ?-Al-Si eutectic. The hardness of the thermo-affected layer was 48.1 ? 8.2HV0.1. This is twice the hardness of aluminium AA1200. The increase in hardness was attributed to the ?-Al-Si eutectic formed and ?-Al grain refinement. The microstructures of the alloyed layers consisted of Al-Ni (Al3Ni + Al3Ni2) phases and the Al4C3 phase. The volume of the Al-Ni intermetallic phases in the alloyed layer increased as the Ni content in the powder mixture increased, while the Al4C3 phase was dominant when the SiC content was high. The needle-like Al4C3 phase formed from the dissociation of SiC (to form Si + C) and reaction of Al with C. An increase in hardness with increasing Ni content in the Ni + SiC powder mixture was observed. The hardness increase was attributed to the Al3Ni + Al3Ni2 phases formed which have a higher hardness (732HV and 1013HV respectively) compared to the needle shaped Al4C3 phase (300HV) [127,129]. The highest DISCUSSION 130 hardness of 138?17.2HV0.1 was achieved after laser alloying with 70wt%Ni + 30wt%SiC. This represents a five times hardness increase compared to that of a aluminium AA1200 metal (24.0?0.4HV0.1). When laser alloying Al with mixed Ti + SiC powders, some of the SiC particles dissociated due to the high temperatures achieved during laser alloying to form Si and C. The C reacted with either Al to form Al4C3, or with Ti to form TiC. The Gibbs free energy is higher for the formation of Al4C3 compared to TiC. Therefore, the formation of TiC was a more thermodynamically favourable product. The free Si reacted with Ti to form a thermodynamically stable Ti5Si3 phase while reactions between Al and Ti resulted in the formation of the Al3Ti intermetallic phase. When the SiC content in the alloying powder mixture was high (70wt%SiC + 30wt%Ti), the Al4SiC4, formed from the reaction of Al with Si and C. The surface hardness increased with increasing Ti content in the alloying powder mixture and the highest hardness of 149.0?21.3HV0.1 was achieved when laser alloying with 70wt%Ti + 30wt%SiC. 6.3 Laser alloying Al with Ni, Ti and SiC simultaneously Following the study of the breakdown systems, the complex Al-Ni-Ti-SiC system was investigated in detail. Studies of the in situ formation of the metal matrix composites and intermetallic compounds during laser alloying of Al with various combinations of Ni, Ti and SiC simultaneously was conducted and the phases observed reported in Chapter 4, Section 4.5. During laser alloying, heat from the laser beam created a melt pool on the aluminium surface and the Ni, Ti and SiC powder mixtures were injected into the melt pool. The molten powders react with aluminium and each other to form various in-situ products the composition of which depended on the percentage of each powder added and the temperature of the melt pool. The melt pool temperature depends on the absorbed heat which in turn depends on laser power, laser scanning speed, laser beam spot size, thermal conductivity and absorption coefficient of the base material. The laser energy density was maintained at 100MJ/m2 after the scanning speed of 10mm/s was DISCUSSION 131 determined during optimization. The laser energy density is calculated from: E (MJ/m2) = q/(vd) ,where q is the laser power (4kW), v is the laser scanning speed (10mm/s) and d is the diameter of the laser beam (4mm). The surface temperature of aluminium during laser alloying was calculated using the following equation [14]: Ts = {(I/k) [? (4??/?)]} C Where Ts is the surface temperature, I is the laser beam intensity, k is the thermal conductivity of Al, ? is the thermal diffusivity, ? is the laser-material interaction time and C is the absorption coefficient. Intensity is calculated from I = P/A (where, I = intensity, P = power, A = spot area = ?r2). Substituting the values from Table 6.2 gave Ts = 2635 K. Table 6.2: Values used to calculate the surface temperature Thermal Conductivity (k) 237 Wm-1K-1 Thermal diffusivity (?) 8.418x10-5 m2/s Laser-interaction time (?) 0.3 s Intensity (I) 318x106 W/m2 Absorption Coefficient (C) 0.2 The phases in the alloyed microstructures were identified using EDS elemental analysis, phases diagrams and XRD. The phase diagrams were used as guidelines during phase identification. The reaction between Al and Ni is exothermic and the heat provided by the laser beam initiates this reaction. The Al-Ni binary phase diagram in Figure 2.4 shows that two solid solutions and five stable intermetallic phases (Al3Ni, Al3Ni2, NiAl, Ni5Al3 and Ni3Al) exist. Two of these phases were detected in this study. The molten Al reacted with Ni to form Al3Ni and Al3Ni2 intermetallic phases in-situ. Two types of Al3Ni phases were observed. The dendritic Al3Ni phase was produced from a eutectic reaction between Ni and liquid Al, while the Al3Ni DISCUSSION 132 phase was observed around the Al3Ni2 phase was produced as a peritectic product of a reaction between liquid Al and the Al3Ni2 phase. These Al-Ni intermetallic phases were also observed by other authors [30,38,39] when laser alloying Al with Ni. The morphology of the Al3Ni2 phase was equiaxed dendritic. Ke et al. [130] also reported equiaxed sub-micron sized Al3Ni particles dispersed in the Al matrix while columnar Al3Ni particles surrounded the Al3Ni2 grains. The reaction of Al with Ti resulted in the formation of Al3Ti. This is the most thermodynamically stable phase in the Al-Ti system and was also observed by Wendt et al. [51] when laser alloying Al with Ti. Due to the high Al content (from the Al base) available during laser alloying compared to the Ti and Ni contents, only the Ni-rich (Al3Ni and Al3Ni2) and Ti-rich (Al3Ti) Al-intermetallic phases were formed. Due to the high surface temperatures achieved during laser alloying and the wide distribution of the SiC particles, some SiC particles dissolved in the melt pool and dissociated into C and Si. The C reacts with either Ti or Al to form TiC or Al4C3, while Si reacted with Ti to form the thermodynamically favourable Ti5Si3 phase. The Gibbs free energies for the formation of other Ti-Si phases (TiSi, TiSi2 and Ti5Si4) are greater than that of Ti5Si3 [126]. The Gibbs energy for the formation of TiC is lower than that of Al4C3, therefore the reaction of C with Ti was more favourable. Two types of TiC phases were observed, namely dendritic and interfacial. The interfacial TiC phases were formed around the SiC particles due to adsorption of Ti on the SiC surface. Due to the high cooling rates associated with laser treatment, these TiC phases did not grow or aggregate. The dendritic TiC phases were formed inside the melt pool from the reaction of the dissolved SiC particles and Ti. When the SiC content was greater than or equal to the Ti content and there was sufficient SiC (? 20wt%) in the powder mixture, Al reacted with the excess SiC to form either the Al4C3 or Al4SiC4 phase. The in-situ products from the reaction of DISCUSSION 133 Al with SiC have also been reported by other authors [71-75]. The Al4C3 phase was formed when the SiC content was low (? 20wt%) while Al4SiC4 formed with high SiC contents (? 40wt%). Ti also reacted with SiC to form a Ti3SiC2 ternary phase according to the reaction below [90,91]: 3Ti + SiC + C ? Ti3SiC2 Surface hardness values of the laser alloyed samples were investigated and compared to the pure aluminium AA1200 metal in Table 5.1. The effect of the laser beam on the hardness of the pure Al, without the addition of the alloying powders, was also tested. This resulted in a 25% increase in the hardness due to refinement of the ?-Al grains. Laser surface alloying with Ni, Ti and SiC simultaneously resulted in a further improvement of the surface hardness due to the formation of the intermetallic phases and the metal matrix composites as well as the refinement of the ?-Al grains. A minimum hardness increase of 6 times that of aluminium was achieved when alloying with 10wt%Ni + 70wt%Ti + 20wt%SiC and a maximum of 13 times was achieved when alloying with 80wt%Ni + 15wt%Ti + 5wt%SiC. The hardness increased with increasing Ni content. This trend occurred primarily because the in-situ formed Al-Ni (Al3Ni and Al3Ni2) intermetallic phases have high hardness values. An additional reason is the high density of the equiaxed dendritic Al3Ni2 grains which resulted in low Al mean free paths. This limited the contribution of the ?-Al to the overall surface hardness. The Al3Ti phase was dendritic which led to high volume fractions of ?-Al with large mean free paths. This resulted in a lower hardness with increasing Ti (and reducing Ni and SiC) in the alloying powder mixture. Increasing the SiC content in the alloying powder mixture led to the formation of needle-like (e.g. Al4C3) and platelet (e.g. Al4SiC4) phases which promoted the formation of large ?-Al mean free paths. This resulted in an increased contribution from the ?-Al phase to the overall alloy hardness. DISCUSSION 134 6.4 Effect of laser alloying on wear resistance A reduction in wear rate (or improved wear resistance) was observed for the laser alloyed surfaces compared to the aluminium metal under sliding (two body abrasion) and three body abrasion wear conditions. A maximum reduction in wear rate of 38% (compared to pure Al) was observed when laser alloying with 20wt%Ni + 30wt%Ti + 50wt%SiC during sliding wear and of 82% when laser alloying with 40wt%Ni + 20wt%Ti + 40wt%SiC during three body abrasion. This enhanced wear performance was attributed to the metal matrix composites and the intermetallic phases formed during laser alloying. The abrasive particles could easily cut into and plough through the ductile aluminium while the presence of the hard metal matrix composites and intermetallic compounds in the laser alloyed surfaces limited deep cutting and ploughing. The wear rates measured under two body abrasion conditions were at least four times higher than those determined under three body abrasion and the associated wear mechanisms were found to be more severe under two body abrasion. There are several factors which contributed towards this response including the Ha/Hm ratios (ratio of the hardness of the abrasive particles to the hardness of the material surfaces), the applied stress and the degree to which the abrasive is able to facilitate material removal. The hardness values of the SiC and SiO2 abrasives are 2500HV and 800HV respectively [98]. The hardness of the laser alloyed surfaces are in the range of 144-319HV1. Therefore the ratio of the hardness of the abrasive particles to the hardness of the alloyed surfaces (Ha/Hm) is in the range of 7.8-17.4 for two body abrasion wear and in the range of 2.5-5.6 for three body abrasion wear. The Ha/Hm ratios for two body abrasion wear are three times higher than those under three body wear leading to more severe material damage. The applied stress was calculated to be 0.06MPa during two body abrasion and 0.02MPa during three body abrasion, despite a higher load being applied during three body wear testing (9.8N versus 2.16N). This is due to the smaller contact area between abrasive and material during two body wear testing, namely 36mm2 compared to 420mm2. The degree to which the abrasive facilitates DISCUSSION 135 material removal is influenced by the depth to which an abrasive indents the material [98]. During three body abrasion the particles are free to roll between the surfaces compared to the fixed particles on the counterface in two body wear. The free rolling particles generally cause more indenting and ploughing on the surface while the fixed particles tend to indent and cut into the surface resulting in higher material removal rates. The depth of indentation of the abrasives increases with increasing hardness of the abrasive particles. Hence the SiC is expected to indent the surface to a greater depth than the SiO2. The indentation depth is also proportional to the depth of deformation and the induced strain at a given depth. Hutchings [98] reported that abrasion wear introduces very high strains into the surface of the material which decreases with depth into the bulk. The higher stresses exerted during the two body wear tests are therefore expected to have caused higher induced strains compared to the three body abrasion conditions. The predominant wear mechanisms for the alloyed layers were groove formation by ploughing and cutting action of the abrasive particles, smearing, material pile- up, extensive cracking of the intermetallic phases and fracturing of the embedded SiC particles in the MMCs. Groove formation of aluminium MMCs during abrasive wear was also reported by Elleuch [115]. The main wear mechanisms of the pure aluminium metal were extensive plastic deformation and deep groove formation. This severe wear led to surface cracking (also at a sub-surface level) and material pile-up. This type of severe wear scarring of both the pure aluminium and the alloyed layers can be predicted from the Ha/Hm ratios discussed above. The calculated ratios are greater than 1.2 which signifies the boundary condition between mild and severe wear as defined by Hutchings [98]. Abrasive wear occurs by plastic mechanisms or fracture mechanisms or both [98]. In this work, both mechanisms were observed as ploughing and cutting (plastic mechanisms), and fracturing and cracking (fracture mechanisms) occurred. The wear rates predicted by brittle fracture are generally higher than those due to plastic mechanisms [98]. The action of the abrasive particle on a region of the hard matrix reinforcement can lead to plastic flow or to fracture DISCUSSION 136 depending on the load carried by each abrasive particle and on its geometry, as well as on the size and the mechanical properties of the abrasive and the reinforcement particles and of the matrix. Alloys with low Al mean free paths had high wear rates as brittle fracture was promoted by the formation of high densities of the hard and brittle intermetallic phases and MMCs in which cracking of the intermetallic phases and fracturing of the SiC particles were observed. The Ti containing and the Al-SiC intermetallic phases resulted in large Al mean free paths. This promoted wear by plastic ploughing and cutting. The hard phases, e.g. the equiaxed dendritic Al3Ni2 grains and the SiC particles, were also seen to interrupt groove formation (plastic mechanism). Man et al. [3] observed abrasive wear by microcracking and flaking during wear of aluminium AA6061 laser surface alloyed with NiTi. The authors reported that the presence of the hard and fine intermetallic phases (Al3Ni and Al3Ti) together with the soft ?-Al in the interdendritic region, constituted a favourable combination with hard phases resisting abrasion and the soft phase suppressing crack growth. Guo et al. [49] also reported that hardness and brittleness of the intermetallic phases play a role in determining which phase would have more wear resistance. Phases that have a high hardness and are less brittle will have better wear resistance but when ductility is low (for brittle phases) fracture occurs leading to high wear rates. Hutchings [98] also reported that fracture toughness of the materials sometimes plays a more important role than the hardness during wear by fracture mechanisms and that wear rates decrease with increasing fracture toughness. This direct correlation was not observed in the current work and no trends of increasing fracture toughness or hardness leading to high wear resistance could be formulated. The alloy with the highest hardness namely, the 80wt%Ni + 15wt%Ti + 5wt%SiC alloy, had a moderate toughness based on the impact results and had the ninth highest wear rate of the fifteen materials tested. The alloy with the best toughness and hardness combination, namely the 70wt%Ni + 20wt%Ti + 10wt%SiC alloy had one of the highest wear rates. DISCUSSION 137 There is a direct correlation between the microstructural properties, bonding and volume fraction of the reinforcement particles in the MMCs and intermetallic phases and their relationship to abrasive wear [98,116]. These factors determine if the alloyed aluminium surfaces would respond homogenously or heterogeneously under wear test conditions. When the size of the abrasive particles is greater than the size of the reinforcements, the material behaves in a homogeneous manner during wear. If the size of the reinforcements is greater than or equal to the size of the abrasive particles, then the material will respond heterogeneously [98]. In this work, the size distribution of the SiC particles in the alloyed surfaces was wide (14-800?m), the average size of the SiO2 particles used in the three body abrasion tests was 530?m while the size of the SiC abrasive grit particles used for two body abrasion was 251?m. Due to the wide particle size distribution in the alloyed surfaces the materials responded either homogenously or heterogeneously depending on the volume fraction of the reinforcements. It is well known that a high volume fraction of small particles will result in low mean free paths which would generally lead to a high wear resistance [98]. However this correlation could not be established in the current work due to the complex nature of the alloyed surfaces. The hard intermetallic phases cracked due to repeated loading and the high stresses applied during wear and these cracks facilitated material removal as wear progressed. Fracturing of the embedded SiC particles also occurred due to cyclic loading but the strong bonding between the particles and the matrix ensured that decohesion (interfacial debonding and pull-out) of the SiC particles was prevented. A strong interfacial bond between the reinforcements and the matrix is important. When the bond is weak the reinforcements are easily pulled out during wear and can act as additional abrasives thereby increasing the wear rate, as was observed by Ax?n and Zum Gahr [103]. Due to the strong interfacial bonding and prior to fracture, the embedded SiC particles in the MMCs also hindered groove formation by the abrasives. This was also observed by Anandkumar et al. [72] who observed discontinuous grooves confined to the matrix on the worn surfaces DISCUSSION 138 of Al-7wt%Si laser alloyed with SiC. The authors reported that the hard SiC reinforcement particles interrupted the abrasive grooving action. 6.5 Effect of laser alloying on impact resistance Laser surface alloying the Al AA1200 metal led to a 31-50% decrease in the impact resistance of the pure metal. This is predominantly due to the increased hardness achieved by the laser alloying process. Ozden et al. [131] also observed an increase in impact resistance for untreated aluminium alloys (AA2124, AA5083 and AA6063) compared to the aluminium MMCs. The high impact resistance of the pure aluminium was noted by the high degree of plastic deformation experienced by the samples during impact testing. The pure aluminium samples did not fracture; they simply bent. All the laser alloyed samples fractured and displayed similar fracture characteristics. Cleavage occurred in the SiC particles and transgranular cracking occurred in the intermetallic phases. Alloyed surfaces with large ?-Al mean free paths resulted in high absorbed energies being recorded during fracture. This was observed in microstructures which had high volumes of dendritic, needle-like or platelet shaped phases. Cuplets characteristic of ductile fracture were observed in the bulk aluminium. The intermetallic phases assist in anchoring the SiC particles to the Al matrix by preventing decohesion of the SiC particles. Vreeling et al. [120] reported that when the matrix is deformed, the intermetallic phases transfer the stress to the SiC particle surface which leads to fracture of the SiC particles. The brittle fracture of the SiC particles is initiated by cleavage mechanisms. Vreeling et al. [120] concluded that suppressing the formation of intermetallic phases would postpone the fracture initiation process. For alloys with a low SiC content, decohesion and cracking of the intermetallic phases resulted in fracture of the alloys. Ozden et al. [131] reported that when the distribution of SiC particles in the matrix is heterogeneous, formation of clusters occur which decreases the matrix- reinforcement bonding and reduces the impact strength of the composite as the DISCUSSION 139 clustered particles are easily separated during impact loading. Interfacial debonding occurs and voids nucleate around the SiC particles. In regions of clustering and decohesion of the interface, a low Al mean free path facilitates linkage between neighbouring voids and cracks as a direct result of a decreased propagation distance between cracked SiC particles. This behaviour was observed in several alloyed surface layers in the current project. The impact strength has been shown to increase with SiC particle size [131] but in this work the SiC particle size distribution was wide (14-800?m) and this effect was not observed. 6.6 Summary of the main results The results from this research project are summarized in Table 6.3. The highest ranking (1) is given to a sample that performed the best i.e. highest hardness or lowest wear rate or highest absorbed energy during impact, while a ranking of 14 represents a sample that performed the worst. The sample numbering used in the table represents the following compositions: head2right Sample 1 = untreated aluminium AA1200 head2right Sample 2 = 10wt%Ni + 70wt%Ti + 20wt%SiC head2right Sample 3 = 20wt%Ni + 30wt%Ti + 50wt%SiC head2right Sample 4 = 20wt%Ni + 40wt%Ti + 40wt%SiC head2right Sample 5 = 30wt%Ni + 30wt%Ti + 40wt%SiC head2right Sample 6 = 30wt%Ni + 40wt%Ti + 30wt%SiC head2right Sample 7 = 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC head2right Sample 8 = 40wt%Ni + 20wt%Ti + 40wt%SiC head2right Sample 9 = 40wt%Ni + 30wt%Ti + 30wt%SiC head2right Sample 10 = 40wt%Ni + 40wt%Ti + 20wt%SiC head2right Sample 11 = 50wt%Ni + 20wt%Ti + 30wt%SiC head2right Sample 12 = 60wt%Ni + 30wt%Ti + 10wt%SiC head2right Sample 13 = 70wt%Ni + 10wt%Ti + 20wt%SiC head2right Sample 14 = 70wt%Ni + 20wt%Ti + 10wt%SiC head2right Sample 15 = 80wt%Ni + 15wt%Ti + 5wt%SiC DISCUSSION 140 From the table, it can be observed that samples with a high Ni content had the best hardness, samples with a high SiC content had the best wear resistance and samples with a high Ti content had the best impact resistance. The hard, dense Al3Ni2 phase coupled with the low Al mean free paths were responsible for the increase in hardness. A combination of high hardness and homogeneously distributed SiC particles were required for good wear resistance. The large Al mean free paths between the Al3Ti dendrites and the Ti5Si3 and Ti3SiC2 phases led to increased absorbed energies and better resistance to impact. Table 6.3: Ranking of samples according to hardness, wear and impact. Ranking Hardness Sliding wear Abrasive wear Impact 1 Sample 15 Sample 3 Sample 8 Sample 1* 2 Sample 14 Sample 7 Sample 4 Sample 14 3 Sample 13 Sample 11 Sample 5 Sample 2 4 Sample 12 Sample 8 Sample 9 Sample 4 5 Sample 11 Sample 5 Sample 10 Sample 5 6 Sample 10 Sample 9 Sample 6 Sample 10 7 Sample 9 Sample 4 Sample 7 Sample 7 8 Sample 8 Sample 12 Sample 3 Sample 6 9 Sample 7 Sample 15 Sample 2 Sample 15 10 Sample 6 Sample 14 Sample 11 Sample 3 11 Sample 5 Sample 10 Sample 12 Sample 9 12 Sample 4 Sample 6 Sample 15 Sample 13 13 Sample 3 Sample 13 Sample 13 Sample 8 14 Sample 2 Sample 2 Sample 14 Sample 11 15 Sample 1 Sample 1 Sample 1 Sample 12 * Samples did not fracture but simply bent DISCUSSION 141 6.7 Industrial application of laser surface alloyed Al AA1200 The hardness of the laser surface alloyed aluminium AA1200 samples was in the range of 140-320HV1 which is higher than the hardness of aluminium AA6061 (100HV), AA7075 (186HV) and AA2024 (130HV) alloys. The aluminium AA2024 and AA7075 alloys are used in aircraft structures especially in wings and fuse structures under tension, while the aluminium AA6061 alloy is used in less demanding structural applications. Based on hardness alone, the laser surface alloyed aluminium AA1200 can replace the high aluminium alloys (2xxx, 6xxx and 7xxx series) in application. With respect to wear resistance, by extrapolating the wear rate-time curves it would take 8.3 minutes under sliding wear conditions and 22 hours under three body dry abrasion conditions for the entire 3mm alloyed layer to be removed. The SiC particles used during sliding wear are very hard and led to aggressive wear conditions while the silica abrasive used for the three body abrasion resulted in mild wear conditions. These results indicate that these samples are not ideal for applications in which severe wear will be experienced. However they may be suitable for mild wear conditions where short component lifetimes (less than 20 hours) are required. It must be emphasized that in order to assess the true suitability of these materials in industrial applications, field tests should be conducted. CONCLUSIONS 142 7. CONCLUSIONS The novel components of this research were laser alloying aluminium AA1200 with Ni, Ti and SiC simultaneously, optimizing the laser parameters for the formation of homogeneous and crack-free alloyed surfaces, studying the in situ formation of the metal matrix composites and intermetallic compounds, and characterizing the mechanical and tribological properties of the alloyed materials. Aluminium AA1200 was successfully laser alloyed with Ni, Ti and SiC powders simultaneously using an Nd:YAG laser. The optimum laser parameters were experimentally determined to be: ? 4kW laser power; ? 4mm laser beam spot size; ? 10mm/s laser scanning speed; ? 2-3g/min powder feed rate; ? 4 l/min Argon (used as a shielding and carrier gas) These parameters ensured that sufficient heat was supplied to create a melt pool and melt the powders during laser alloying. The optimum powder feed rate ensured that sufficient powder was supplied to the laser/material interaction zone (on the substrate) and that sufficient powder was injected into the melt pool. The Argon was used as a shielding gas to prevent oxidation and as a carrier gas to transport the powders from the hopper to the melt pool created on the aluminium substrate. Metal matrix composites reinforced with SiC particles and intermetallic phases were formed in the alloyed layers. The predominant intermetallic phases were Al3Ni, Al3Ni2, Al3Ti, TiC and Ti5Si3. The Al3Ni, Al3Ni2 and Al3Ti phases were formed due to the reaction of Al with Ni and Ti. The dissociation of SiC particles and reactions with Ti resulted in the formation of TiC and Ti5Si3 phases. In alloys with a high SiC content, Al4C3, Al4SiC4, Ti3SiC2 and Al3TiC2 intermetallic phases were also formed from reaction of the dissociated SiC with either Al or Ti. CONCLUSIONS 143 An improvement in surface hardness was achieved after laser alloying the aluminium AA1200 metal. Hardness was found to increase with increasing Ni content. A minimum hardness increase of 6 times that of pure aluminium was achieved when alloying with 10wt%Ni + 70wt%Ti + 20wt%SiC while a maximum hardness of 13 times that of pure aluminium was achieved when alloying with 80wt%Ni + 15wt%Ti + 5wt%SiC. The grain morphology of the intermetallic and metal matrix composite phases as well as the aluminium mean free path influenced the hardness. Dense, equiaxed dendritic phases and small mean free paths led to high hardness values while the needle-like and platelet phases with large mean free paths produced low hardness values. In general laser alloying improved the wear resistance of the aluminium AA1200 surface. In 2 body abrasion (sliding) wear conditions an improvement of 4-38% was achieved, with alloy 20wt%Ni + 30wt%Ti + 50wt%SiC showing the best resistance to wear. The predominant wear mechanisms under these conditions were groove formation by ploughing and cutting action of the abrasive particles, smearing, material pile-up, extensive cracking of the intermetallic phases and fracturing of the embedded SiC particles in the MMCs. In 3 body abrasion wear conditions an improvement of 19-82% was achieved, with alloy 40wt%Ni + 20wt%Ti + 40wt%SiC showing the best resistance to wear. The wear mechanisms under these conditions were similar to those observed for two body wear but less severe. The aluminium mean free path was low for the Al-Ni intermetallic phases and this promoted cracking and fracturing of the SiC particles. The Ti-containing and the Al-SiC intermetallic phases resulted in a large aluminium mean free path. This promoted wear by plastic ploughing and cutting. From the impact studies, brittle fracture of the SiC particles and transgranular cracking of the intermetallic phases was observed for the laser alloyed samples, while ductile fracture was observed for the bulk aluminium. Alloyed layers with a high Ti content had high absorbed energies which represent a reduction in brittleness, while alloyed layers with a high Ni content had low absorbed energies which indicate a preference for brittle fracture. RECOMMENDATIONS 144 8. RECOMMENDATIONS A continuation of this investigation would increase the knowledge base on the general understanding on laser alloying of aluminium with mixed Ni, Ti and SiC powders. Further studies could include: head2right Quantifying the volume fraction of each phase in the laser alloyed layers which would aid in quantifying their effect on the mechanical properties; head2right Corrosion studies on the laser alloyed samples and aluminium AA1200; head2right Tensile tests to determine the strength of the interface between the alloyed layer and the base aluminium; head2right Determination of the high temperature properties of the alloyed materials. REFERENCES 145 9. REFERENCES [1] Tomida, S.; Nakata, K. (2003); Fe-Al composite layers on aluminium alloy formed by laser surface alloying with iron powder; Surface and Coatings Technology; vol. 174 -175; pp. 559-563. [2] Katipelli, L. R.; Agarwal, A.; Dahotre, N. B. (2000); Laser surface engineered TiC coating on 6061 Al alloy: microstructure and wear; Applied Surface Science; vol. 153; pp. 65-78. [3] Man, H. C.; Zhang, S.; Cheng, F. T. (2007); Improving the wear resistance of AA 6061 by laser alloying with NiTi; Materials Letter; vol. 61; pp. 4058- 4061. [4] Almeida, A.; Anjos, M.; Vilar, R.; Li, R.; Ferreira, M. G. S.; Steen, W. M.; Watkins, K. G. (1995); Laser alloying of aluminium alloys with chromium; Surface and Coatings Technology; vol. 70; pp. 221-229. [5] Tomida, T; Nakata, K.; Saji, S.; Kubo, T. (2001); Formation of metal matrix composite layer on aluminium alloy with TiC-Cu powder by laser surface alloying process; Surface and Coatings Technology; vol. 142-144; pp. 585- 589. [6] Das, S. (2004); Development of aluminium alloy composites for engineering applications; Transactions of the Indian Institute of Metals; vol. 57; no. 4; pp. 325-334. [7] Ravi, N.; Sastikumar, D.; Subramanian, N.; Nath, A. K.; Masilamani, V. (2000); Microhardness and microstructure studies on laser surface alloyed aluminium alloy with Ni-Cr; Materials and Manufacturing Processes; vol. 15; no. 3; pp. 395-404. [8] Staia, M. H.; Cruz, M.; Dahotre, N. B. (2001); Wear resistance of a laser alloyed A-356 aluminium/WC composite; Wear; vol. 251; pp. 1459-1468. [9] Kok, M. (2005); Production and mechanical properties of Al2O3 particle- reinforced 2021 aluminium alloy composites; Journal of Materials Processing Technology; vol. 161; pp. 381-387. [10] Hu, C.; Baker, T. N. (1997); A new aluminium silicon carbide formed in laser processing; Journal of Materials Science; vol. 32; pp. 5047-5051. REFERENCES 146 [11] St?ren, S.; Skanaluminium (1994); Understanding aluminium as a material; Training in Aluminium Application Technology (TALAT) Lecture 2101.01: Aluminium in Production Development; Norway; pp. 1-23. [12] Thornton, P. A.; Colangelo, V. J. (1985); Fundamentals of Engineering Materials; Prentice-Hall Inc., New Jersey, USA. [13] Buchfink, G. (2007); The Laser as a Tool ? A light beam conquers industrial production; 1st edition; Vogel Buchverlag; W?rzburg; Deutschland. [14] Ready, J. F., Farson, D. F. (2001), LIA Handbook of Laser Materials Processing, Laser Institute of America, Magnolia Publishing Inc.; USA. [15] Chryssolouris G. (1991); Laser Machining: Theory and Practice; Springer- Verlag; New York; USA. [16] Riabkina-Fishman, M.; Zahavi, J. (1996); Laser alloying and cladding for improving surface properties; Applied Surface Science; vol. 106; pp. 263- 267. [17] Pityana, S. L.; Rossouw, R. (2008); Laser alloyed Al-Ni-Fe coatings; Proceedings of the 27th International Congress on Applications of Laser and Electro Optics; Temecula, USA; pp. 55-60. [18] Lei, Y. P.; Marakawa, H.; Shi, Y. W.; Li, X. Y. (2001); Numerical analysis of the competitive influence of Marangoni flow and evaporation on heat surface temperature and molten pool shape in laser surface remelting; Computation Materials Science; vol. 21; pp. 276-290. [19] Velde, O.; Gritzki, R.; Grundmann, R. (2001); Numerical investigations of Lorentz force influenced Marangoni convection relevant to aluminium surface alloying; International Journal of Heat and Mass Transfer; vol. 44; pp. 2751-2762. [20] Sarkar, S.; Mohan, P.; Chakraborty, S.; Phanikumar, G. (2003); Transport phenomena in laser surface alloying; Journal of Materials Science; vol. 38; pp. 155-164. [21] Vaziri, S. A.; Shahverdi, H. R.; Torkamany, M. J.; Shabestari, S. G (2009); Effect of laser parameters on properties of surface-alloyed Al substrate with Ni; Optics and Lasers in Engineering; vol. 47; pp. 971-975. REFERENCES 147 [22] Yang, S.; Chen, N.; Liu, W.; Zhong, M. (2003); In situ formation of MoSi2/SiC composite coating on pure Al by laser cladding; Materials Letters; vol. 57; pp. 3412-3416. [23] Yue, T. M; Xu, J. H.; Man, H. C. (1997); Pulsed Nd:YAG laser welding of a SiC particulate reinforced aluminium alloy composite; Applied Composite Materials; vol. 4; pp. 53-64. [24] Das, D. K. (1994); Surface roughness created by laser surface alloying of aluminium with nickel; Surface and Coatings Technology; vol. 64; pp. 11- 15. [25] Kong, C. Y.; Caroll, P. A.; Brown, P.; Scudamore, R. J. (2007); The effect of average powder particle size on deposition efficiency, deposit height and surface roughness in the direct metal laser deposition process; Proceedings of the International Conference on Joining of Materials; Helsing?r; Denmark. [26] Villar, R. (1999); Laser Alloying and Laser Cladding; Materials Science Forum; vol. 301; pp. 229-252. [27] Majumdar, J. D.; Manna, I. (2003); Laser processing of materials; Sadhana; vol. 28; parts 14; pp. 495-562. [28] Toyserkani, E.; Khajepour, A.; Corbin, S. (2005); Laser Cladding; CRC Press, USA. [29] Capello, E., Previtali, B. (2006); The influence of operator skills, process parameters and materials on clad shape in repair using laser cladding by wire; Journal of Materials Processing Technology; vol. 174; pp. 223-232. [30] Bao, C. M.; Dahlborg, U.; Adkins, N.; Calvo-Dahlborg, M. (2009); Structural characterization of Al-Ni powders produced by gas atomization; Journal of Alloys and Compounds; vol. 481; pp. 199-206. [31] Krivoroutchko, K.; Kulik, T.; Matyja, H.; Portnoy, V. K.; Fadeeva, V. I. (2000); Solid state reactions in Ni-Al-Ti-C system by mechanical alloying; Journal of Alloys and Compounds; vol. 308; pp. 230-236. [32] Morsi, K. (2001); Review: reaction synthesis processing of Ni-Al intermetallic materials; Materials Science and Engineering; vol. A299; pp. 1-15. REFERENCES 148 [33] Bysakh, S.; Mitra, S. K.; Phanikumar, G.; Mazumder, J.; Dutta, P.; Chattopadhyay, K. (2003); Characterization of microstructure in laser- surface-alloyed layers of aluminium on nickel; Metallurgical and Materials Transactions A; vol. 34A; pp. 2621-2631. [34] L?pez, G.A.; Sommadossi, S.; Gust, W.; Mittemeijer, E. J. (2002); Phase characterization of diffusion soldered Ni/Al/Ni interconnections; Interface Science; vol. 10; pp. 13-19. [35] Ansara, I.; Dupin, N.; Luka, H. L.; Sundman, B. (1997); Thermodynamic assessment of the Al-Ni system; Journal of Alloys and Compounds; vol. 247; pp. 20-30. [36] Raghavan, V. (2009); Al-Ni-Ti (Aluminium-Nickel-Titanium); Journal of Phase Equilibria and Diffusion; vol. 30; no. 1; pp. 77-78. [37] Jain, M.; Gupta, S. P. (2003); Formation of intermetallic compounds in the Ni-Al-Si ternary system; Materials Characterization; vol. 51; pp. 243-257. [38] Selvan J. S., Soundararajan G, Subramanian K (2000); Laser alloying of aluminium with electrodeposited nickel: optimization of plating thickness and processing parameters; Surface and Coatings Technology; vol. 124; pp. 117-127. [39] Watkins, K. G.; MacMahon, M. A.; Steen, W. M. (1997); Microstructure and corrosion properties of laser surface processed aluminium alloys: a review; Materials Science and Engineering; vol. A231, pp. 55-61. [40] Man, H. C.; Zhang, S.; Yue, T .M.; Cheng, F. T. (2001); Laser surface alloying of NiCrSiB on Al6061 aluminium alloy; Surface and Coatings Technology; vol. 148; pp. 136-142. [41] Das, D. K.; Paradkar, A. G.; Mishra, R. S. (1992); Microstructure and creep behaviour of laser surface alloyed aluminium; Scripta Metallurgica et Materialia; vol. 26; no. 8; pp. 1211-1214. [42] Froes, F. H.; Suryanarayana, C.; Eliezer, D. (1991); Production, characteristics and commercialization of titanium aluminides; The Iron and Steel Institute of Japan (ISIJ) International, vol. 31; no. 10; pp. 1235-1248. REFERENCES 149 [43] Illekov?, E.; ?vec, P.; Jani?kovi?, D. (2009); Influence of the processing on the ordering process in the Al-Ti binary system with composition close to Al3Ti; Journal of Physics: Conference Series; vol. 144; no. 1; pp. 1-6. [44] Ohnuma, I.; Fihita, Y.; Mitsui, H.; Ishikawa, K.; Kainuma, R.; Ishida, K. (2000); Phase equilibria in the Ti-Al binary system; Acta Materialia; vol. 48; pp. 3113-3123. [45] Banerjee, R.; Zhang, X. D.; Dregia, S. A.; Fraser, H. L. (1999); Phase stability in the Al/Ti multilayers; Acta Materialia; vol. 47; no. 4; pp. 1153- 1161. [46] Abboud, J. H.; West, D. R. F. (1990); Laser surface alloying of titanium with aluminium; Journal of Materials Science Letters; vol. 9; pp. 308-310. [47] Wang, G.; Wu, A.; Zou, G.; Zhao, Y.; Chen, Q.; Ren, J. (2009); Bending properties and fracture behavior of Ti-23Al-17Nb alloy laser beam welding joints; Tsinghua Science and Technology; vol. 14; no. 3; pp. 293-299. [48] Kreimeyer, M.; Wagner, F.; Vollertsen, F. (2005); Laser processing of aluminium-titanium tailored blanks; Optics and Lasers in Engineering, vol. 43; pp. 1021-1035. [49] Guo, B.; Zhou, J.; Zhang, S.; Zhou, H.; Pu, Y.; Chen, J. (2008); Tribological properties of titanium aluminides coatings produced on pure Ti by laser surface alloying; Surface and Coatings Technology; vol. 202; pp. 4121- 4129. [50] Gunn?s, A.E.; Olsen, A.; Taft?, J. (1998); Microstructure development in laser-processed Al-Ti-Ni alloys with 25at.%Ti; Journal of Materials Science; vol. 33; pp 4961 ? 4969. [51] Wendt, U.; Settegast, S.; Grodrian, I. U. (2003); Laser alloying of aluminium with titanium wire; Journal of Materials Science Letters; vol. 22; pp. 1319-1322. [52] Chen, K. C. (2003); NiTi ? magic or phase transformation?; Proceedings of the American Society of Engineering Education Annual Conference & Exposition; Nashville; Tennessee, USA. REFERENCES 150 [53] Le, D. H.; Colinet, C.; Hicter, P.; Pasturel, A. (1991); Theoretical description of phase equilibrium in the nickel-titanium system; Journal of Physics: Condensed Matter; vol. 3; pp. 9965-9974. [54] Zeng, K.; Schmid-Fetzer, R.; Huneau, B.; Rogl, P.; Bauer, J. (1999); The ternary system Al-Ni-Ti Part II: Thermodynamic assessment and experimental investigation of polythermal phase equilibria; Intermetallics; vol. 7; pp. 1347-1359. [55] Huneau, B.; Rogl, P.; Zeng, K.; Schmid-Fetzer, R.; Bohn, M.; Bauer, J. (1999); The ternary system Al-Ni-Ti Part I: Isothermal section at 900?C: Experimental investigation and thermodynamic calculation; Intermetallics; vol. 7; pp. 1337-1345. [56] Schuster, J. C.; Pan, Z.; Liu, S.; Weitzer, F.; Du, Y. (2007); On the constitution of the ternary Al-Ni-Ti; Intermetallics; vol. 15; pp. 1257-1267. [57] Raghavan, V. (2010); Al-Ni-Ti (Aluminium-Nickel-Titanium); Journal of Phase Equilibria and Diffusion; vol. 31; no. 1; pp. 55-56. [58] Mabhali, L.; Pityana, S.; Rampedi, L. (2008); Laser alloying of Al with mixed Ti and Ni powders to improve surface properties; Proceedings of the Microscopy Society of Southern Africa, Gaborone, Botswana; vol. 38; pp.19. [59] Biswas, S.; Varin, R. A. (1996); Evolution of microstructures in the nickel modified titanium trialuminides near the L12 phase field; Metallurgical and Materials Transactions A: Physical Metallurgy and Materials Science; vol. 27A; pp. 5-17. [60] Biswas, S., Varin, R. A. (1996); Mechanical behaviour of the in situ composite alloys in the Al-Ni-Ti system near the L12 phase field; Metallurgical and Materials Transactions A: Physical Metallurgy and Materials Science; vol. 27A; pp. 71-79. [61] ASM International (1992); Alloy phase diagrams; ASM Handbook; vol. 3; USA. [62] Martin, J. (1996); Materials for Engineering; The Institute of Materials, Minerals and Mining; England. REFERENCES 151 [63] Syv?j?rvi, M. (1999); High growth rate of SiC: growth processes and structural quality; PHD Thesis; Department of Physics and Measurement Technology; Link?ping University; Link?ping; Sweden. [64] Raghavan, V. (2008); Al-C-Si (Aluminum-Carbon-Silicon); Journal of Phase Equilibria and Diffusion; vol. 29; no. 4; pp. 365-366. [65] Gu, W. L.; Yan, C. K.; Zhou, Y. C. (2003); Reactions between Al and Ti3SiC2 in temperature range of 600-650?C; Scripta Materialia; vol. 49; 1075-1082. [66] Gr?bner, J.; Lukas, H. L.; Aldinger, F. (1996); Thermodynamic calculation of the ternary Al-Si-C; Computer Coupling of Phase Diagrams and Thermochemistry (Calphad); vol. 20; no. 2; pp. 247-254. [67] Oden, L. L.; McCune, R. A. (1987); Phase equilibria in the Al-Si-C system; Metallurgical Transactions A; vol. 18A; pp. 2005-2014. [68] Viala, J. C.; Peillon, N.; Bosset, F.; Bouix, J. (1997); Phase equilibria at 1000?C in the Al-C-Si-Ti quaternary system: an experimental approach; Materials Science and Engineering; vol. A229; pp. 95-113. [69] Walter, D.; Karyasa, I. W. (2005); Solid state reaction in the Al-Si-C system; Journal of the Chinese Chemical Society; vol. 52; pp. 873-876. [70] Yue, T. M.; Wu, Y. X.; Man, H. C. (1999); Improvement in the corrosion resistance of aluminum 2009/SiCw composite by Nd:YAG laser surface treatment; Journal of Materials Science Letters; vol. 18; pp. 173-175. [71] Su, R.; Lei, Y (2008); Microstructure and hardness of laser clad SiCp-Al composite coatings on Al alloys; Materials Letters; vol. 62; pp. 3272 - 3275. [72] Anandkumar, R.; Almeida, A.; Cola?o, R.; Vilar, R.; Ocel?k, V.; De Hosson, J. Th. M. (2007); Microstructure and wear studies of laser clad Al-Si/SiC(p) composite coatings; Surface and Coating Technology; vol. 201; pp. 9497- 9505. [73] Xu, X.; Han, J.; Li, Y.; Liu, L. (2006); SiC particulate reinforced aluminium matrix composite coatings prepared by laser powder deposition; Journal of Ceramic Processing Research; vol. 7; no. 2; pp. 167-171. REFERENCES 152 [74] Hu, C.; Xin, H.; Baker; T.N. (1996), Formation of continuous surface Al- SiCp metal matrix composite by overlapping laser tracks on AA6061 alloy, Materials Science and Technology, vol. 12, pp. 227 - 232. [75] Lee, J.; Byun, J.; Park, S.; Lee, H. (1998); Prediction of Si contents to suppress the formation of Al4C3 in the SiCp/Al composite; Acta Materialia; vol. 46; no. 5; pp. 1771-1780. [76] Zhang K., Chen, G. N. (2000); Influence of SiC particulates on grain structure development of an aluminium 7075 alloy during rapid laser solidification; Journal of Materials Science Letters; vol. 19; pp. 795-796. [77] Birol, Y. (2006); Grain refining efficiency of Al-Ti-C alloys; Journal of Alloys and Compounds; vol. 422; pp. 128-131. [78] Chen, Y.; Wang, H. M. (2006); Microstructure and wear resistance of laser- melted TiC reinforced nickel aluminide dual-phase matrix in situ composite; Intermetallics; vol. 14; pp. 325-331. [79] Hashim, J.; Looney, L.; Hashmi, M. S. J. (2001); The wettability of SiC particles by molten aluminium alloy; Journal of Materials Processes Technology; vol. 119; pp. 324-328. [80] Wu, X. (1999); In situ formation by laser cladding of a TiC composite coating with a gradient distribution; Surface and Coatings Technology; vol. 115; pp. 111-115. [81] Le?n, C. A.; Drew, R. A. L. (2002); The influence of nickel coating on the wettability of aluminium on ceramics; Composites: Part A; vol. 33; pp. 1429-1432. [82] Chung, W.; Li, S. (1996); Ni-coated SiCp reinforced aluminum composites processed by vacuum infiltration; Materials Research Bulletin; vol. 31; no. 12; pp. 1437-1447. [83] Leon-Pati?o, C.; Drew, R. A. L. (2005); Role of metal interlayers in the infiltration of metal-ceramic composites; Current Opinion in Solid State and Materials Science; vol. 9; pp. 211-218. [84] Selvan, J. S.; Subramanian, K.; Nath, A. K. (1999); Laser surface alloying of (pre-placed) SiC and Ni-SiC coating on commercial titanium; Materials and Manufacturing Processes; vol. 14; no. 2; pp. 285-296. REFERENCES 153 [85] Lou, B.; Chen, Z.; Bai, W.; Dong, G. (2006); Structure and erosion of Ni60A/SiC coating by laser cladding; Transactions of Nonferrous Metals Society of China; vol. 16; pp. 643-646. [86] Fujimura, T.; Tanaka, S. I. (1999); In-situ high temperature X-ray diffraction study of Ni/SiC interface reactions; Journal of Materials Science; vol. 34; pp. 235-239. [87] Viswanathan, A.; Sastikumar, D.; Kamachimudali, U.; Kumar, H.; Nath, A. K. (2007); TiC reinforced composite layer formation on Al-Si alloy by laser processing; Surface Engineering; vol. 23; no. 2; pp. 123-128. [88] Kloosterman, A. B.; Kooi, B. J.; De Hosson, J. Th. M. (1998); Electron microscopy of reaction layers between SiC and Ti-6Al-4V after laser embedding; Acta Materialia; vol. 46; no. 17; pp. 6205-6217. [89] Pei, Y. T.; Ocelik, V.; De Hosson, J. Th. M. (2002); SiCp/Ti6Al4V functionally graded materials produced by laser melt injection; Acta Materialia; vol. 50; pp. 2035-2051. [90] Kooi, B. J.; Kabel, M.; Kloosterman, A. B.; De Hosson, J. Th. M. (1999); Reaction layers around SiC particles in Ti: An electron microscopy study; Acta Materialia; vol. 47; no. 10; pp. 3105-3116. [91] Konishi, R.; Yasukochi, R.; Nakatsuka, O.; Koide, Y.; Moriyama, M.; Murakami, M. (2003); Development of Ni/Al and Ni/Ti/Al ohmic contact materials for p-type 4H-SiC; Materials Science and Engineering; vol. B98; pp. 286-293. [92] Qin, F.; Fan, T.; Zhang, D. (2008); Thermodynamic effects of alloying elements on solid-state interfacial reactions in a multicomponent system: Theoretical treatment and application in SiCf/Ti composite; Metallurgical and Materials Transactions A; vol. 40A; pp. 462-471. [93] Rao, K. P.; Zhou, J. B. (2004); Thermal stability and phase formation of mechanically alloyed Ti-Al-Si-C powders; Journal of Materials Science; vol. 39; 5471-5476. [94] Li, S.; Zhou, Y.; Duan, H. (2002); Wettability and interfacial reaction in SiC/Ni plus Ti system; Journal of Materials Science; vol. 37; pp. 2575- 2579. REFERENCES 154 [95] Pleshakov, E.; Senyavs?kyi, Ya.; Filip, R. (2002); Laser surface modification of Ti-6Al-4V alloy with silicon carbide; Materials Science; vol. 38; no. 5; pp. 646-652. [96] Selvan, J. S.; Subramanian, K.; Nath, A. K. (1999); Ti-SiC composite and TiC ceramic layer formation on Ti6Al4V surface by laser alloying of pre- placed SiC coating; Advanced Performance Materials; vol. 6; pp. 71-83. [97] Man, H. C.; Zhang, S.; Cheng, F. T.; Yue, T. M. (2002); In situ synthesis of TiC reinforced surface MMC on Al 6061 by laser surface alloying; Scripta Materialia; vol. 46; pp. 229-234. [98] Hutchings, I. M. (1992), Tribology: Friction and Wear of Engineering Materials, University of Cambridge, Edward Arnold, pp. 133-171. [99] Kivikyt?-Reponen, P. (2006); Correlation of material characteristics and wear of powder metallurgical metal matrix composites; Doctoral Theses; Helsinki University of Technology; Espoo; Finland. [100] Deuis, R. L.; Subramanian, C.; Yellup, J. M. (1996); Abrasive wear of aluminium composites ? a review; Wear; vol. 201; pp. 132-144. [101] Larsen-Basse, J. (1990), Role of microstructure and mechanical properties in abrasion, Scripta Metallurgica et Materialia, vol. 24, no. 5, pp. 821-826. [102] Zum Gahr, K. H. (1987), Microstructure and wear of materials, Tribology series 10, Elsevier. [103] Ax?n, N.; Zum Gahr, K. H. (1992); Abrasive wear of TiC-steel composite clad lasers on tool steel; Wear; vol. 157; pp. 189-201. [104] Rao, R. N.; Das, S. (2010); Effect of matrix alloy and influence of SiC particle on the sliding wear characteristics of aluminium alloy composites; Materials and Design; vol. 31; issue 3; pp. 1200-1207. [105] Thouless, M. D. (1992); Frictional sliding and pull-out of a fibre; Scripta Metallurgica et Materialia; vol. 27; no. 9; pp. 1211-1214. [106] Tomida, S.; Nakata, K.; Shibata, S.; Zenkouji, I.; Saji, S. (2003); Improvement in wear resistance of hyper-eutectic Al-Si cast alloy by laser surface remelting; Surface and Coatings Technology; vol. 169-170; pp. 468- 471. REFERENCES 155 [107] Miyajima, T.; Iwai, Y. (2003); Effect of reinforcements on sliding wear behavior of aluminium matrix composites; Wear; vol. 255; pp. 606-616. [108] O?Quigley, D. (1996); The properties and relationship between properties of a wide range of WC-Co hardmetals; MSc dissertation; University of Witwatersrand; Johannesburg; South Africa. [109] Larsen-Basse, J.; Shishido, C. M.; Salem, L. K. (1976); Abrasion of some cemented carbides by SiC paper; Advances in Hard Materials Tool Technology; R. Komanduri (ed.); Carnegie-Mellon University; Pittsburg; pp. 560-565. [110] Sin, H.; Saka, N.; Suh, N. P. (1979); Abrasive wear mechanisms and the grit size effect; Wear; vol. 55; issue 1; pp. 163-190. [111] Mabhali, L. A. B. (2007); Effect of SiC abrasive breakdown on the wear rate of WC-12wt%Co alloy; MSc dissertation, University of Witwatersrand, Johannesburg; South Africa. [112] Larsen-Basse, J.; Premaratne, B. (1983); Effect of relative hardness on transitions in abrasive wear mechanisms; International Conference on Wear of Materials; Virginia; USA; pp. 161-166. [113] Staia, M. H.; Cruz, M.; Dahotre, N. B. (2000); Microstructure and tribological characterization of an A-356 aluminium alloy superficially modified by laser alloying; Thin Solid Films; vol. 377-378; pp. 665-674. [114] Majumdar, J. D.; Mordike, B. L.; Manna, I. (2000); Friction and wear behavior of Ti following laser surface alloying with Si, Al and Si+Al; Wear; vol. 242; pp. 18-27. [115] Elleuch, K.; Mezlini, S.; Guermazi, N.; Kapsa, Ph. (2006); Abrasive wear of aluminium alloys rubbed against sand; Wear; vol. 261; pp. 1316-1321. [116] ?ahin, Y. (2010); Abrasive wear behaviour of SiC/2014 aluminium composite; Tribology International; vol. 43; pp. 939-943. [117] Almeida, A.; Carvalho, F.; Carvalho, P. A.; Vilar, R. (2006); Laser developed Al-Mo surface alloys: microstructure, mechanical and wear behaviour; Surface and Coatings Technology; vol. 200; issue 16-17; pp. 4782-4790. REFERENCES 156 [118] Shipway, P. H.; Kennedy, A. R.; Wilkes, A. J. (1998); Sliding wear behaviour of aluminium-based metal matrix composites produced by a novel liquid route; Wear; vol. 216; pp. 160-171. [119] Reed-Hill, R. E; Abbaschian, R. (1994); Physical Metallurgy Principles; 3rd edition; PWS Publishing company; Boston; USA. [120] Vreeling, J. A.; Ocel?k, V.; Hamstra, G. A.; Pei, Y. T.; De Hosson, J. Th. M. (2000); In situ microscopy investigation of failure mechanisms in Al/SiCp metal matrix composite produced by laser embedding; Scripta Materialia; vol. 42; pp. 589-595. [121] Kamrani, S.; Simchi, A.; Riedel, R.; Seyed Reihani, S. M. (2007); Effect of reinforcement volume fraction on mechanical alloying of Al-SiC nanocomposite powders; Powder Metallurgy; vol. 50; no. 3; pp. 276-282. [122] American Standard Testing Methods (2004), Standard Test Method for Measuring Abrasion Using the Dry Sand/Rubber Wheel Apparatus, ASTM G65-04, pp.1-12. [123] American Standard Testing Methods (2005), Standard Test Method for Norched Bar Impact Testing of Metallic Materials, ASTM E23-05, pp.1-27. [124] Brytan, Z.; Bonek, M.; Dobrza?ski, L. A. (2010); Microstructure and properties of laser surface alloyed PM austenitic stainless steel; Journal of Achievements in Materials and Manufacturing Engineering; vol. 40; no. 1; pp. 70-78. [125] Julies, B. A.; Knoesen, D.; Pretorius, R.; Adams, D. (1999); A study of the NiSi to NiSi2 transition in the Ni-Si binary system; Thin Solid Films; vol. 347; issue 1-2; pp. 201-207. [126] Guan, Q. L.; Wang, H. Y.; Li, S. L.; Wang, W. N.; L?, S. J.; Jiang, Q. C. (2009); Influence of Al addition on the products of the self-propagating high-temperature synthesis Al-Ti-Si system; Materials Chemistry and Physics; vol. 114; pp. 709-715. [127] Zamzam, M. A.; El-Sabbagh, A. S.; Milad, M. M. (2002); Study of thermal cycling effects of Al-Ni turbine blades on their lifetime; Materials and Design; vol. 23; pp. 161-138. REFERENCES 157 [128] Li, Y. X.; Hu, J. D.; Wang, H. Y.; Guo, Z.X.; Chumakov, A. N. (2007); Thermodynamic and lattice parameter calculation of TiCx produced from Al-Ti-C powders by laser igniting self-propagating high-temperature synthesis; Materials Science and Engineering A; vol. 458; pp. 235-239. [129] Ure?a, A; Rodrigo, P.; Gil, L.; Escalera, M. D.; Baldonedo, J. L. (2001); Interfacial reactions in an Al-Cu-Mg (2009)/SiCw composite during liquid processing; Journal of Materials Science; vol. 36; pp. 429-439. [130] Ke, L.; Huang, C.; Xing, L.; Huang, K. (2010); Al-Ni intermetallic composites produced in situ by friction stir processing; Journal of Alloys and Compounds; vol. 503; pp. 494-499. [131] Ozden, S; Ekici, R.; Nair, F. (2007); Investigation of impact behaviour of aluminium based SiC particle reinforced metal-matrix composites; Composites Part A: Applied Science and Manufacturing; vol. 38; issue 2; pp. 484-494. APPENDIX A 158 APPENDIX A: SYNTHESIS OF THE ALLOYS This Appendix provides detailed descriptions of laser surface alloying Aluminium AA1200 with 15 combinations of Ni, Ti and SiC powders. These results were summarized in Chapter 4, Section 4.5. Alloy 1: 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC Laser alloying of aluminium AA1200 surface with equal amounts of Ni, Ti and SiC powders (i.e. 33.3wt%Ni, 33.3wt%Ti and 33.3wt%SiC) was performed and the microstructure and phases observed in the alloyed layer characterized. Figure A.1 is a cross-section optical micrograph of the alloyed layer showing SiC particles distributed within the layer. The difference in size of the SiC particles observed in the layer occurred to a wide particle size distribution of the SiC powder used (see Table 4.4). The thickness of the alloyed layer was 2.16mm. Figure A.1: Optical micrograph of a cross-section of a surface laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC showing SiC particle (black) distributed within the alloyed layer. APPENDIX A 159 SEM cross-section micrographs of surface laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC are shown in Figures A.2 and A.3. The phases observed were SiC particles, TiC dendrites, ?-Al, Al3Ni, Al3Ni2, Al3Ti and Ti5Si3. These phases were identified with the aid of EDS elemental analysis, phase diagrams and the XRD. Due to high surface temperatures achieved during laser alloying (2362?C), some powder particles dissolved in the melt pool. Some of the SiC particles were retained and formed a metal matrix composite as observed in Figures A.1. The dissolved SiC particles dissociated to form Si and C. The C reacts with either Ti or Al to form TiC or Al4C3. The changes in Gibbs free energy, ?G, for the reactions are given by [97]: 4/3 Al + C ? 1/3 Al4C3 ?G (kJ/mol) = 1/3 ?GAl4C3 = -89.611 + 0.0328T Ti + C ? TiC ?G (kJ/mol) = -205.351 + 0.0291T The Gibbs free energy is more negative for the formation of TiC and thus there was a higher tendency for the formation of TiC than for Al4C3. Two types of TiC phases are observed namely dendritic (Figure A.2) and interfacial TiC (Figure A.3). The interfacial TiC phases were formed around the SiC particles due to adsorption of Ti on the SiC surface. Due to the high cooling rates associated with laser treatment, these TiC phases did not grow or aggregate. The dendritic TiC phases were formed inside the melt pool from the reaction of the dissolved SiC particles and Ti. The reaction of Al with Ni resulted in the formation of Al3Ni and Al3Ni2 phases while the reaction of Al with Ti resulted in the formation of Al3Ti. Around Al3Ni2 phase, the Al3Ni phase (white phase around Al3Ni2 in Figure A.2) is formed by a peritectic reaction between the liquid and the Al3Ni2 phase. Due to the high Al content in the base material compared to the Ti and Ni content, the Ni-rich and Ti- rich Al-intermetallic phases were not formed. The formation of Al3Ni, Al3Ni2 and Al3Ti were shown to increase the hardness and abrasive wear resistance of aluminium alloys [54]. The authors reported a high hardness of approximately 800HV when laser alloying with 80wt% Ni and 20wt% Ti at 10mm/s laser scanning speed. The Si (from SiC) reacted with Ti to form a Ti in Figure A.2. The equ 3/8 Si + 5/8 Ti ? 1/8 Ti The Gibbs free energies for the formation of other Ti Ti5Si4) are greater than that of Ti in the alloyed surfaces and cross identified in the XRD are Figure A.2: SEM cross 33.3wt%Ni + 33.3wt% particles, TiC dendrites, Al3Ni2 Al3Ni Al3Ti APPENDIX A 5Si ation for the formation of Ti5Si3 is given below [126]: 5Si3 ?G (kJ/mol) = 1/8 ?G4Ti5Si3 ? -Si phases (TiSi, TiSi 5Si3 [126]. The XRD used to identify the phases -sections is shown in Figure A.4. Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti and Ti -section micrograph of a surface laser Ti + 33.3wt%SiC showing a near surface regi Al3Ni, Al3Ti and Al3Ni2 phases. 160 3 phase shown -72 + 0.0056T 2 and The phases 5Si3. alloyed with on with SiC TiC SiC Al3Ti ?-Al Edge of the surface Ti5Si3 Figure A.3: SEM cross 33.3wt%Ti + 33.3wt% SiC with SiC particle, TiC, Al Figure A.4: XRD of 33.3wt%SiC. Al3Ni2 SiC APPENDIX A -section micrograph of a layer alloyed with showing a region in the middle of the alloyed layer 3Ti and Al3Ni2 phases. a surface laser alloyed with 33.3wt%Ni + 33.3 161 33.3wt%Ni + wt%Ti + TiC ?-Al Al3Ti APPENDIX A 162 Alloy 2: 80wt%Ni + 15wt%Ti + 5wt%SiC Aluminium AA1200 was laser alloyed with a powder containing 80wt%Ni + 15wt%Ti + 5wt%SiC and the microstructure and phases formed studied. Due to low SiC content in the alloying powder, there were very few SiC particles retained in the alloyed layer as shown in Figure A.5. Therefore, metal matrix composite was not formed. The thickness of the alloyed layer was 1.90mm. Figure A.5: Optical cross-section micrograph of a surface laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC showing SiC particles (black) randomly distributed in the laser alloyed layer. SEM cross-section micrograph of the laser alloyed layer is shown in Figure A.6. The microstructure was dominated by the dendritic structures of Al3Ni and needle-structured Al3Ti phases. XRD of a surface laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC is shown in Figure A.7. The phases identified are Al, SiC, Al3Ni, Al3Ni2 and Al3Ti. Figure A.6: SEM cross 15wt%Ti + 5wt%SiC Figure A.7: XRD of 5wt%SiC. Al3Ti APPENDIX A -section micrograph of a layer alloyed with showing a region near the surface of the alloyed layer. a surface laser alloyed with 80wt%Ni + 15 163 80wt%Ni + wt%Ti + SiC Al3Ni Al3Ni2 APPENDIX A 164 Alloy 3: 50wt%Ni + 20wt%Ti + 30wt%SiC Figure A.8 shows an optical micrograph of a cross-section of aluminium AA1200 laser surface alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC. Due to high SiC content in the alloying powder mixture, a metal matrix composite (MMC) was formed as evident by the SiC particles distributed within the alloyed layer. The thickness of the alloyed layer was 2.37mm. Figure A.8: Optical micrographs of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC showing SiC particles (black) in the matrix. SEM cross-section micrograph the laser alloyed layer is shown in Figure A.9. Due to low Ti content in the alloying powder, the formation of the brittle Al4C3 phase was not avoided (by forming TiC only) as observed in the microstructure. This Al4C3 phase was formed according to the following equation. 4Al + 3SiC ? Al4C3 + 3Si The Si produced from the dissociation of SiC, reacted with Ti to form the thermodynamically favourable Ti5Si3 phase. Some of the C from this dissociation reacted with Ti to form TiC phases. These were observed mainly around the SiC particles. The formation of the Al suitable laser energy den melting and dissolution of SiC particles; (2) using a large laser scanning speed to shorten the melt pool time and weaken the reaction of SiC and Al; (3) increasing appropriately the amount of SiC to phase [73]. Other phases observed are Al was also observed near the surface of the alloyed layer. XRD of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC is show phases identified were Al, SiC, TiC, Al phase was also detected due to the reaction of Al with Ti. Figure A.9: SEM cross 50wt%Ni + 20wt%Ti Al4C3 Al3Ni2 Al3Ni APPENDIX A 4C3 phase can be controlled by (1) selecting a sity to reduce the melting temperature and restrain the restrain the nucleation and growth of the Al 3Ni, Al3Ni2 and Ti5Si3. Undissolved Ti n in Figure A.10. The 3Ni, Al3Ni2, Al4C3 and Ti -section micrograph of a surface laser alloyed with + 30wt%SiC showing the near surface region. 165 4C3 5Si3. The Al3Ti SiC Ti5Si3 Undissolved Ti powder TiC around SiC particle APPENDIX A 166 Figure A.10: XRD of a surface laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC. Alloy 4: 20wt%Ni + 30wt%Ti + 50wt%SiC An optical micrograph of a cross-section of an aluminium surface laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC is shown in Figure A.11. A metal matrix composite was formed and SiC particles are distributed throughout the alloyed layer. The thickness of the alloyed layer was 2.40mm. APPENDIX A 167 Figure A.11: Optical cross-section micrograph of a surface layer alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC showing SiC particles (black) distributed within the layer. SEM micrographs of the laser alloyed layers are shown in Figures A.12 and A.13. The phases in the microstructures were identified with the aid of EDS, XRD and phase diagrams. In situ formed Al3Ni, Al3Ni2 and Al3Ti intermetallic phases were observed from the reactions of Al with Ni and Ti. The in situ reaction of Al and SiC resulted in the formation of the Al4SiC4 and free Si. Some of the SiC particles dissociated to form Si and C phases. The C reacted with Ti to form a TiC phase. Due to the high Ti content, some of the Ti reacted with Si to form Ti5Si3 and Ti3SiC2 phases. The Ti3SiC2 phase is formed according to the reaction below [90,91]: 3Ti + SiC + C ? Ti3SiC2 Figure A.12: SEM cross 30wt%Ti + 50wt%SiC Figure A.13: SEM cross 30wt%Ti + 50wt%SiC XRD of the laser alloyed layer is shown in Figure A14. The phases identified were Al, SiC, TiC, Al Al4SiC4 Ti3SiC2 SiC Al3Ti Al APPENDIX A -section micrograph of a layer alloyed with 20 showing a region near the surface of the layer -section micrograph of a layer alloyed with 20 showing a region in the middle of the alloyed layer. 3Ni, Al3Ni2, Al3Ti, Ti5Si3, Al3TiC2 and Al4SiC 168 wt%Ni + . wt%Ni + 4. TiC Ti5Si3 TiC Al Al3Ni2 Al3Ni APPENDIX A 169 Figure A.14: XRD of a surface laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC. Alloy 5: 60wt%Ni + 30wt%Ti + 10wt%SiC The optical micrograph of an aluminium AA1200 surface laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC is shown in Figure A.15. An aluminium matrix composite reinforced with SiC particles was formed during laser alloying. The thickness of the alloyed layer was 2.21mm. SEM micrographs of the laser alloyed layers are shown in Figures A.16 and A.17. The phases observed are SiC, TiC, Al3Ni, Al3Ni2, Ti5Si3 and Al3Ti. These phases were identified with the assistance of the XRD in Figure A.18. APPENDIX A 170 Figure A.15: Optical micrographs of a cross-section of an Al laser surface alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC showing SiC particles (black) within the alloyed layer. Figure A.16: SEM cross-section micrograph of an Al layer laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC. SiC Al3Ni TiC Al3Ti APPENDIX A 171 Figure A.17: SEM cross-section micrograph of a layer alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC. Figure A.18: XRD of a surface laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC. Al3Ni Al3Ti Ti5Si3 Al3Ni2 APPENDIX A 172 Alloy 6: 70wt%Ni + 20wt%Ti + 10wt%SiC An optical micrograph of a cross-section of an aluminium AA1200 surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC is shown in Figure A.19. A metal matrix composite was formed and SiC particles are distributed throughout the alloyed layer. The thickness of the alloyed layer was 1.94mm. Figure A.19: Optical micrographs of a cross-section of an Al surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC showing SiC particles (black distributed within the alloyed layer). SEM micrographs of surface layers alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC are shown in Figures A.20 and A.21. The microstructure is dominated by the formation of Al3Ni and Al3Ti phases. Al3Ni2 phase was also observed within the Al3Ni phase. This observation was also reported by Jain and Gupta [37]. Al3Ni phase solidified from the liquid during cooling and enveloped the Al3Ni2 phase. Figure A.22 is the XRD of a laser alloyed specimen showing Al, SiC, Al3Ni, Al3Ni2 and Al3Ti phases. APPENDIX A 173 Figure A.20: SEM cross-section micrograph of a surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC. Figure A.21: SEM cross-section micrograph of a surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC showing Al3Ni2, Al3Ni and Al3Ti intermetallic phases. Al3Ti Al3Ni Al3Ni Al3Ti Al3Ni2 APPENDIX A 174 Figure A.22: XRD of an Al surface laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC. Alloy 7: 10wt%Ni + 70wt%Ti + 20wt%SiC A cross-section microstructure of an Al surface laser alloyed with a powder containing 10wt%Ni + 70wt%Ti + 20wt%SiC is shown in Figure A.23. A metal matrix composite was formed and SiC particles are distributed throughout the alloyed layer. The thickness of the alloyed layer was 2.03mm. APPENDIX A 175 Figure A.23: Optical micrograph of a cross-section of an Al surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC showing SiC particles (black) distributed within the alloyed layer. SEM micrographs of the laser alloyed layers are shown in Figures A.24 and A.25. The microstructure is dominated by the Al3Ti phase as shown in Figure A.24. Some of the SiC particles dissociated to form Si and C. Due to the abundance of Ti in the alloying powder, these products (Si + C) reacted with Ti to form Ti5Si3 and TiC respectively. Al also reacted with Ti to form Al3Ti. These phases are shown in Figure A.25. XRD of the laser alloyed sample is shown in Figure A.26 and the phases identified are Al, SiC, TiC, Al3Ti, Ti5Si3 and Al3Ni. The Al3Ni phase was formed from the reaction of Al with Ni. APPENDIX A 176 Figure A.24: SEM cross-section micrograph of a surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. Figure A.25: SEM cross-section micrograph of a surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. Al3Ti SiC TiC SiC Ti5Si3 AlTi3 Al APPENDIX A 177 Figure A.26: XRD of an Al surface laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. Alloy 8: 70wt%Ni + 10wt%Ti + 20wt%SiC An optical micrograph of a cross-section of an Al surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC is shown in Figure A.27. A metal matrix composite was formed and SiC particles are distributed throughout the alloyed layer. The thickness of the alloyed layer was 1.97mm. Figure A.27: Optical micrograph of with 70wt%Ni + 10wt% SEM micrographs of the laser alloyed layers are shown in Figures A.28 and A.29. The microstructure is dominated by the formation of Al to high Ni content in the alloying powder mixture. TiC and Al observed due to the dissociation of SiC and reaction of C with either Al or Ti. The Si (from SiC) reacted with Ti to form a Ti was formed in-situ from the reaction of Al with Ti. These phases were identified using the XRD in Figure A.30. APPENDIX A a cross-section of an Al surface laser alloyed Ti + 20wt% showing SiC particles (black) in the layer. 3Ni and Al 4 5Si3 phase. An Al3Ti intermetallic 178 3Ni2 phases due C3 phases were phase APPENDIX A 179 Figure A.28: SEM cross-section micrograph of a surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC showing an Al4C3 phase. Figure A.29: SEM cross-section micrograph of a surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC. Al4C3 Al3Ni2 SiC Al3Ni APPENDIX A 180 Figure A.30: XRD of an Al surface laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC. Alloy 9: 40wt%Ni + 20wt%Ti + 40wt%SiC An optical micrograph of a cross-section of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC is shown in Figure A.31. A metal matrix composite reinforced with SiC particles was formed during laser alloying. The thickness of the alloyed layer was 2.14mm. Figure A.31: Optical micrograph of 20wt%Ti + 40wt%SiC showing SiC (black) particles distributed within the matrix. SEM micrographs of surface layers alloyed with 40wt% 40wt%SiC are shown in Figures A.32 to A.34. The XRD of the laser alloyed surface is shown in Figure A.35. The phases identified are Al, Al Al3Ti, SiC, TiC, Ti5Si The dendritic Al3Ni phase shown in Figures A.32 and A.33 formed from a eutectic reaction between Ni and liquid Al. The Al formed as a peritectic product of a reaction between liquid Al and the Al phase. The Al3Ti intermetallic phase Dendritic structures of TiC (Figures A.33 and A.34) and interfacial phases (Figure A.32) were formed from the reactions of Ti with C (from SiC). The interfacial TiC phases formed around the SiC particles due to adso surface. The free Si (from SiC) reacted with Ti to form Ti Ti3SiC2 phase was formed due to the reaction of Ti, C and SiC. An Al ternary phase was also formed from the reaction of Al and SiC. APPENDIX A an Al surface laser alloyed with 40wt%Ni + Ni + 20wt%Ti + 3 and Ti3SiC2. Two types of Al3Ni phases were observed. 3Ni phase in Figure A.34 formed from the reaction of Al with Ti. rption of Ti on the SiC 5Si3 phase. A ternary 181 3Ni, Al3Ni2, 3Ni2 4SiC4 APPENDIX A 182 Figure A.32: SEM cross-section micrograph of an Al surface alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing SiC, TiC, Al3Ni, Al3Ni2 and Al3Ti phases. Figure A.33: SEM cross-section micrograph of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing Al3Ni, Al3Ti, Ti5Si3 SiC and TiC phases. Al3Ti SiC Al3Ni TiC Ti5Si3 SiC Al3Ni TiC TiC Al3Ti Al3Ni2 APPENDIX A 183 Figure A.34: SEM cross-section micrograph of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC showing Al3Ti, Al3Ni, Al3Ni2, Ti5Si3 and Ti3SiC2 phases. Figure A.35: XRD of an Al surface laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC. Al3Ti Ti5Si3 Ti3SiC2 Al3Ni Al3Ni2 APPENDIX A 184 Alloy 10: 40wt%Ni + 30wt%Ti + 30wt%SiC An optical micrograph of a cross-section of an Al surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC is shown in Figure A.36. A metal matrix composite was formed. The thickness of the alloyed layer was 1.81mm. Figure A.36: Optical micrographs of an Al surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing SiC particles (black) within the alloyed layer. SEM micrographs of the laser alloyed layers are shown in Figures A.37 to A.39. The reaction of Al with Ni and Ti resulted in the formation of Al3Ni, Al3Ni2 and Al3Ti intermetallic phases observed in the micrographs. The dissociation of SiC and reaction with Ti resulted in TiC, Ti5Si3 and Ti3SiC2 phases. Dendritic and interfacial TiC phases are observed in Figure A.37. XRD of a laser alloyed specimen is shown in Figure A.40 and the phases identified are Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3 and Ti3SiC2. APPENDIX A 185 Figure A.37: SEM cross-section micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing SiC, TiC, Al3Ti and Al3Ni phases. Figure A.38: SEM cross-section micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing Al3Ti, Al3Ni, Al3Ni2, Ti5Si3, Ti3SiC2 phases. SiC TiC Al3Ti Al3Ni TiC Ti5Si3 Ti3SiC2 Al3Ni2 Al3Ti Al3Ni APPENDIX A 186 Figure A.39: SEM cross-section micrograph of a surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC showing Al3Ti, Ti5Si3 and Al3Ni phases. Figure A.40: XRD of an Al surface laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC. Ti5Si3 Al3Ni Al3Ti Alloy 11: 20wt%Ni + 40wt%Ti + 40wt%SiC The optical micrographs of aluminium AA1200 alloy surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC is shown in Figures A.41. A metal matrix composite reinforced with SiC particles was layer was 2.29mm. Figure A.41: Optical micrograph of 40wt%Ti + 40wt%SiC showing SiC particles (black) in the alloyed layer. SEM micrographs of cross Figures A.42 and A.43. The reaction of Al with Ni and Ti resulted in the formation of Al3Ni and Al of SiC and reaction with Ti resulted in TiC, Ti phases were identified from the XRD in Figure A.44. The reaction of Al with SiC resulted in the formation of an Al dissociation of SiC is also observed in the XRD. APPENDIX A formed. The thickness of the alloyed an Al surface laser alloyed with 20wt%Ni + -sections of the laser alloyed layers are shown in 3Ti intermetallic phases respectively. The dissociation 5S3 and Ti3SiC 4SiC4 ternary phase. Free Si from the 187 2 phases. These Figure A.42: SEM cross 20wt%Ni + 40wt%Ti Figure A.43: SEM cross 20wt%Ni + 40wt%Ti Al3Ti Ti5Si3 Al3Ni Al APPENDIX A -section micrograph of a surface laser alloyed with + 40wt%SiC. -section micrograph of a surface laser alloyed with + 40wt%SiC. 188 SiC TiC Al3Ti Ti3SiC2 APPENDIX A 189 Figure A.44: XRD of an Al surface laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC. Alloy 12: 30wt%Ni + 40wt%Ti + 30wt%SiC The optical micrographs of a cross-section of an aluminium surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC is shown in Figure A.45. A metal matrix composite reinforced with SiC particles was formed. The thickness of the alloyed layer was 2.24mm. APPENDIX A 190 Figure A.45: Optical micrograph of a cross-section of an Al surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing SiC particles (black) distributed within the matrix. SEM micrographs of cross-sections of the laser alloyed layers are shown in Figures A.46 and A.47. Al reacted with Ni and Ti to form Al3Ni, Al3Ni2 and Al3Ti phases. The dissociation of SiC and reaction with Ti resulted in TiC, Ti5Si3 and Ti3SiC2 phases. These phases are identified from the XRD in Figure A.48. Free Si and Al4SiC4 phases were also identified in the XRD. The Al4SiC4 phase formed from the reaction of Al with SiC. APPENDIX A 191 Figure A.46: SEM cross-section micrograph of a surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing SiC, TiC, Al3Ni2 and Ti5Si3 phases. Figure A.47: SEM cross-section micrograph of a surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC showing Al3Ni, Al3Ni2 and Ti3SiC2 phases. SiC Ti5Si3 TiC Al3Ni2 Al3Ni Al3Ni2 Ti3SiC2 APPENDIX A 192 Figure A.48: XRD of an Al surface laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC. Alloy 13: 40wt%Ni + 40wt%Ti + 20wt%SiC An optical micrograph of a cross-section of an aluminium surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC is shown in Figure A.49. A metal matrix composite reinforced with SiC particles was formed. The thickness of the alloyed layer was 1.93mm. APPENDIX A 193 Figure A.49: Optical micrographs of an Al surface laser alloyed with 40wt%Ni, 40wt%Ti and 20wt%SiC showing SiC particles (black) distributed within the matrix. SEM micrographs of cross-sections of the laser alloyed layers are shown in Figures A.50 and A.51. The reaction of Al with Ni and Ti resulted in the in situ formation of Al3Ni, Al3Ni2 and Al3Ti intermetallic phases. The dissociation of SiC and reaction with Ti resulted in TiC, Ti5Si3 and Ti3SiC2 phases. These phases were identified using the XRD in Figure A.52. APPENDIX A 194 Figure A.50: SEM cross-section micrograph of a surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC at showing Al3Ni2 and Al3Ti intermetallic phases. Figure A.51: SEM cross-section micrograph of a surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC showing SiC, Al3Ti, Ti5Si3 and Ti3SiC2 phases. SiC Al3Ni2 Al3Ti Ti3SiC2 Ti5Si3 Al3Ti APPENDIX A 195 Figure A.52: XRD of an Al surface laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC. Alloy 14: 30wt%Ni + 30wt%Ti + 40wt%SiC An optical micrograph of a cross-section of an aluminium surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC is shown in Figure A.53. A metal matrix composite reinforced with SiC particles was formed. The thickness of the alloyed layer was 1.95mm. APPENDIX A 196 Figure A.53: An optical micrograph of a cross-section of an Al surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC particles (black) distributed within the matrix. SEM micrographs of cross-sections of surfaces laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC are shown in Figures A.54 and A.55. The reaction of Al with Ni and Ti resulted in the in situ formation of the Al3Ni, Al3Ni2 and Al3Ti intermetallic phase. The dissociation of SiC and reaction with Ti resulted in TiC and Ti5Si3. All the phases were identified using the XRD in Figure A.56. Ternary Ti3SiC2 phase observed in the XRD was formed from a reaction of SiC with Ti and C, while Al4SiC formed due to the reaction of Al with SiC. Figure A.54: SEM cross 30wt%Ni + 30wt%Ti Figure A.55: SEM cross 30wt%Ni + 30wt%Ti Ti5Si3 Al3Ti TiC Al3Ni APPENDIX A -section micrograph of a surface laser alloyed + 40wt%SiC showing SiC, Al3Ni and Al3Ni2 -section micrograph of a surface laser alloyed with + 40wt%SiC showing SiC, Al3Ti, TiC, and Ti 197 with phases. 5Si3 phases. SiC SiC Al3Ni2 APPENDIX A 198 Figure A.56: XRD of an Al surface laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC showing SiC, Al3Ti, TiC, and Ti5Si3 phases. The results for the phases observed in each laser alloyed layer are summarized in Table A.1. APPENDIX A 199 Table A.1: Phases observed in each of the laser alloyed layers Powder composition Phases observed 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti & Ti5Si3 80wt%Ni + 15wt%Ti + 5wt%SiC ?-Al, SiC, Al3Ni, Al3Ni2 & Al3Ti 50wt%Ni + 20wt%Ti + 30wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3 & Al4C3 20wt%Ni + 30wt%Ti + 50wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3, Al3TiC2 & Al4SiC4 60wt%Ni + 30wt%Ti + 10wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti & Ti5Si3 70wt%Ni + 20wt%Ti + 10wt%SiC ?-Al, SiC, Al3Ni, Al3Ni2 & Al3Ti 10wt%Ni + 70wt%Ti + 20wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ti & Ti5Si3 70wt%Ni + 10wt%Ti + 20wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3 & Al4C3 20wt%Ni + 70wt%Ti + 10wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ti & Ti5Si3 40wt%Ni + 20wt%Ti + 40wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3, Ti3SiC2 & Al4SiC4 40wt%Ni + 30wt%Ti + 30wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3 & Ti3SiC2 20wt%Ni + 40wt%Ti + 40wt%SiC ?-Al, Si, SiC, TiC, Al3Ni, Al3Ti, Ti5Si3, Ti3SiC2 & Al4SiC4 30wt%Ni + 40wt%Ti + 30wt%SiC ?-Al, Si, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3, Ti3SiC2 & Al4SiC4 40wt%Ni + 40wt%Ti + 20wt%SiC ?-Al, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3 & Ti3SiC2 30wt%Ni + 30wt%Ti + 40wt%SiC ?-Al, Si, SiC, TiC, Al3Ni, Al3Ni2, Al3Ti, Ti5Si3, Ti3SiC2 & Al4SiC4 APPENDIX B 200 APPENDIX B: WEAR TEST RESULTS This Appendix provides detailed results of the abrasion wear tests conducted on the 15 alloys. These results were summarized in Chapter 5, Sections 5.3 and 5.4. Appendix B1 details the results of the two body abrasion wear tests and Appendix B2 details the results of the three body abrasion wear tests. B1: Two body abrasion wear tests Pure Aluminium AA1200 The graph of wear rate versus time for pure aluminium is shown in Figure B.1. The average wear rate is 4.48?0.21x10-4cm3/sec. Figure B.1: Graph of wear rate versus time for untreated aluminium surface. SEM images of the untreated aluminium AA1200 alloy worn surfaces are shown in Figure B.2. The wear features observed were grooves formed by the motion of the SiC abrasive grits (Figures B.2(a) and (b)) and material pile-up due to groove formation (Figure B.2(b)). Plastic deformation (Figure B.2(c) showing differences in grey levels) and cracking in the deformed layer (Figure B.2(d)) were observed in the cross-sections. Cracks which initiated in the deformed layer, and at the interface between the deformed layer and the bulk material, facilitated material removal during the cutting action Figure B.2: Worn surfaces of the untreated aluminium in (a) and (b) plan view; (c) and (d) cross-section Alloy 1: 33.3wt%Ni + The graphs of the wear rate versus time for the untreated aluminium and aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC are shown in Figure B.3. The average wear rate of the laser alloyed aluminium was 3.40?0.13x10-4cm3/sec. Therefore, laser alloying Al AA1200 with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC improved the wear resistance by approximately 25% under the tested conditions. The reduction in wear rate was attributed mainly to the formation of intermetallic matrix. (a) (c) APPENDIX B of the abrasive grits leading to high mass losses. s. 33.3wt%Ti + 33.3wt% SiC compounds and the retained SiC particles in the (d) (b) 100?m 200?m 201 10?m 50?m APPENDIX B 202 Figure B.3: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC are shown in Figure B.4. Wear grooves which were formed due to the cutting action of the SiC abrasive grits are observed in Figures B.4(a) and (b). These wear grooves were narrower and shallow compared to those formed for the untreated aluminium. Cracking inside and outside the wear grooves was observed in Figure B.4(b). The strong bonding between particles and the matrix ensured decohesion (interfacial debonding and pull-out) of SiC particles did not occur during wear. The randomly distributed SiC particles and the intermetallic phases act as load bearing constituents leading to improved wear resistance. The cross-sections in Figures B.4(c) and (d) show that plastic deformation, which occurred in the untreated Al, does not occur because of the presence of hard ceramic particles and intermetallic compounds. Fracturing of SiC particles is observed in Figure B.4(c). Transgranular cracking of the Al3Ni intermetallic phase is observed in Figure B.4(d) Figure B.4: worn surfaces 33.3wt%Ti + 33.3wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 2: 80wt%Ni + 15wt%Ti + The graphs of the wear rate versus time for the untreated aluminium and aluminium laser alloyed Figure B.5. Laser alloying resulted in a 12% reduction in the wear rate. The average wear rate of the las (a) (c) APPENDIX B of Al samples laser alloyed with 33.3wt%Ni + cross 5wt%SiC with 80wt%Ni + 15wt%Ti + 5wt%SiC are er alloyed aluminium was 4.03?0.07x10 (d) (b) 10?m 203 -sections. shown in -4cm3/sec. 10?m APPENDIX B 204 Figure B.5: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC are shown in Figure B.6. The wear features observed were groove formation (Figures B.6(a) and (b)), material pile-up (Figure B.6(b)) and cracking (Figures B.6(b) and (c)). The wear grooves were narrow and shallow compared to those observed for the untreated aluminium. Lateral cracking is observed in the cross-sections in Figures B.6(c) and (d). Figure B.6: worn surfaces + 5wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 3: 50wt%Ni + 20wt%Ti + 30wt%SiC The graphs of the wear rate versus time for aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC Figure B.7. Laser alloying improved the wear rate by approximately 19%. The average wear rate of the laser alloyed aluminium was 3. (a) (c) APPENDIX B of Al samples laser alloyed with 80wt%Ni + 1 cross-sections. an untreated aluminium and 70?0.12x10 (d) (b) 205 5wt%Ti are shown in -4cm3/sec. APPENDIX B 206 Figure B.7: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC are shown in Figure B.8. Narrow wear grooves are observed in Figure B.8(a) and (b). SiC particles hindered crack propagation as shown in Figure B.8(b). Lateral cracking is observed in the cross-sections in Figures B.6(c) and (d). Transgranular subsurface cracking which resulted in chipping and detachment of the worn layer were observed in the cross-sections in Figure B.8(c) and (d). Figure B.8: worn surfaces + 30wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 4: 20wt%Ni + 30wt%Ti + 50wt% SiC The graphs of the wear rate versus time for aluminium laser alloyed wit Figure B.9. There is a 38% reduction in the wear rate of aluminium due alloying. The average wear rate of the 4cm3/sec. The improvement in wear was greater compared to alloying with lower SiC content in the powder mixture. This shows the significance of SiC particles as load bearing constituents during two (a) (c) APPENDIX B of Al samples laser alloyed with 50wt%Ni + 20wt%Ti cross-sections. an untreated aluminium and h 20wt%Ni + 30wt%Ti + 50wt%SiC are laser alloyed aluminium was 2.83 body abrasion wear tests. (d) (b) 10?m 207 shown in to laser ?0.13x10- 10?m APPENDIX B 208 Figure B.9: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC are shown in Figure B.10. Narrow wear grooves are observed in Figure B.10(a) and (b). SiC particles hindered crack propagation as shown in Figure B.10(a) and (d). Lateral cracking is observed in the cross-sections in Figures B.10(c) and (d). A transgranular crack which resulted in chipping and detachment of the worn layer were observed in the cross-sections in Figure B.10(c). Figure B.10: worn surfaces + 50wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 5: 60wt%Ni + 30wt%Ti + 10wt% SiC The graphs of the wear rate versus time for aluminium laser alloyed with 60w B.11. The average wear rate of the 4cm3/sec. A 10% reduction in wear rate wa (a) (c) APPENDIX B of Al samples laser alloyed with 20wt%Ni + 30wt%Ti cross-sections. an untreated aluminium and t%Ni + 30wt%Ti + 10wt%SiC are laser alloyed aluminium was 4.06 s achieved after laser alloying. 10?m (d) (b) 209 shown in ?0.07x10- 10?m APPENDIX B 210 Figure B.11: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC are shown in Figure B.12. The wear grooves observed in Figures B.12(a) and (b). SiC particles hindered crack propagation as shown in Figure B.12(b). Cracking is observed in Figures B.12(b-d). Transgranular cracking of Al3Ti phase and lateral cracking which resulted in chipping and detachment of the worn layer were observed in the cross- sections in Figures B.12(c) and (d). Figure B.12: worn surfaces + 10wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 6: 70wt%Ni + 20wt%Ti + 10wt%SiC The graphs of the wear rate versus time for an untreated aluminium and aluminium laser alloyed wit Figure B.13. The average wear rate of the laser alloyed aluminium was 4.12?0.07x10-4cm3/sec. A reduction in wear rate of approximately 8% was achieved after laser alloying. (a) (c) APPENDIX B of Al samples laser alloyed with 60wt%Ni + 30wt%Ti cross-sections. h 70wt%Ni + 20wt%Ti + 10wt%SiC are (b) (d) 10?m 211 shown in 10?m APPENDIX B 212 Figure B.13: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC are shown in Figure B.14. Wear grooves were deeper compared to those observed for alloys with higher SiC content as shown in Figure B.14(a) and (b). Lateral cracking is observed in the cross-sections in Figures B.14(c) and (d). Figure B.14: worn surfaces + 10wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 7: 10wt%Ni + 70wt%Ti + 20wt%SiC The graphs of the wear rate versus time for aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC Figure B.15. The average wear rate of the 4.27?0.14x10-4cm3/sec. The wear rate was similar to that of untreated aluminium with only 4% reduction. (a) (c) APPENDIX B of Al samples laser alloyed with 70wt%Ni + 20wt%Ti cross-sections. an untreated aluminium and laser alloyed aluminium was (d) (b) 213 are shown in APPENDIX B 214 Figure B.15: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC are shown in Figure B.16. Wear grooves due to the cutting action of the SiC abrasive particles are observed in the worn micrographs as shown in Figure B.16(a) and (b). Lateral cracking is observed in the cross-section in Figure B.16(c) and the SiC particle in Figure B.16(d) hindered crack growth. Figure B.16: worn surfaces + 20wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 8: 70wt%Ni + 10wt%Ti + 20wt%SiC The graphs of the wear rate versus time for an aluminium laser alloyed Figure B.17. Laser alloying resulted in a 7% reduction in the wear rate. The average wear rate of the laser alloyed aluminium was 4. (a) (c) APPENDIX B of Al samples laser alloyed with 10wt%Ni + 70wt%Ti cross-sections. untreated aluminium and with 70wt%Ni + 10wt%Ti + 20wt%SiC 22?0.07x10 (d) (b) 10?m 215 are shown in -4cm3/sec. 10?m APPENDIX B 216 Figure B.17: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC are shown in Figure B.18. Wear grooves, cracks and material pile-up are observed in Figures B.18(a) and (b). Lateral cracking which results in chipping and detachment of the worn layer is observed in Figures B.18(c) and (d) Figure B.18: worn surfaces + 20wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 9: 40wt%Ni + 20wt%Ti + 40wt%SiC The graphs of the wear rate versus time for an untreated aluminium and aluminium laser alloyed with Figure B.19. Laser alloying resulted in a 17% reduction in the wear rate. The average wear rate of the laser alloyed aluminium was 3.7 (a) (c) APPENDIX B of Al samples laser alloyed with 70wt%Ni + 10wt%Ti cross-sections. 40wt%Ni + 20wt%Ti + 40wt%SiC are 6?0.07x10 (d) (b) 10?m 217 shown in -4cm3/sec. 10?m APPENDIX B 218 Figure B.19: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC are shown in Figure B.20. Shallow wear grooves are observed in Figures B.20(a) and (b). SiC particles hindered crack propagation as shown in Figure B.20(a).Cracks parallel to the wear groove are observed in figure B.20(b). Lateral cracking and fracturing of a SiC particle are observed in the cross-sections in Figures B.20(c) and (d). APPENDIX B 219 Figure B.20: worn surfaces of Al samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections. Alloy 10: 40wt%Ni + 30wt%Ti + 30wt%SiC The graphs of the wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC are shown in Figure B.21. The average wear rate of the laser alloyed aluminium was 4.01?0.13x10-4cm3/sec. An 11% reduction in the wear rate was achieved after laser alloying. (a) (d) (c) (b) APPENDIX B 220 Figure B.21: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC are shown in Figure B.22. Shallow wear grooves are observed in Figures B.22(a) and (b). Cracks parallel to the wear groove are observed in Figure B.22(b). Lateral cracking and fracturing of a SiC particle are observed in the cross-sections in Figures B.22(c) and (d). APPENDIX B 221 Figure B.22: worn surfaces of Al samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC in (a) and (b) plan view; (c) and (d) cross-sections. Alloy 11: 20wt%Ni + 40wt%Ti + 40wt%SiC The graphs of the wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC are shown in Figure B.23. The average wear rate of the laser alloyed aluminium was 4.06?0.13x10-4cm3/sec representing a 9% reduction in the wear rate compared to that of untreated aluminium. (a) (d) (c) (b) APPENDIX B 222 Figure B.23: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC are shown in Figure B.24. Shallow wear grooves are observed in Figures B.24(a) and (b). SiC particles hindered crack propagation as shown in Figure B.24(b). Lateral cracking is observed in the cross-sections in Figures B.24(c) and (d). Figure B.24: worn surfaces + 40wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 12: 30wt%Ni + 40wt%Ti + 30wt%SiC The graphs of the wear rate versus time for aluminium laser alloyed with 30w Figure B.25. A slight reduction in wear rate of approximately 7% reduction was observed for the alloyed The average wear rate of the (a) (c) APPENDIX B of Al samples laser alloyed with 20wt%Ni + 40wt%Ti cross-sections. an untreated aluminium and t%Ni + 40wt%Ti + 30wt%SiC are aluminium compare to untreated aluminium AA1200. laser alloyed aluminium was 4.21?0.0 (b) (d) 10?m 223 shown in 7x10-4cm3/sec. 10?m APPENDIX B 224 Figure B.25: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC are shown in Figure B.26. Shallow wear grooves are observed in Figures B.26(a) and (b). SiC particles hindered crack propagation as shown in Figure B.26(b). Lateral cracking which resulted in chipping of the worn layer is observed in the cross-sections in Figures B.26(c) and (d). Figure B.26: worn surfaces + 30wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 13: 40wt%Ni + 40wt%Ti + 20wt%SiC The graphs of the wear rate versus time for an aluminium laser alloyed wit Figure B.27. The average wear rate of the laser alloyed aluminium was 4.22?0.07x10-4cm3/sec. This wear rate was similar to that observed for an untreated aluminium with a slight improvement of about 5%. (a) (c) APPENDIX B of Al samples laser alloyed with 30wt%Ni + 40wt%Ti cross-sections. untreated aluminium and h 40wt%Ni + 40wt%Ti + 20wt%SiC are (d) (b) 225 shown in APPENDIX B 226 Figure B.27: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC are shown in Figure B.28. Shallow wear grooves are observed in Figures B.28(a) and (b). SiC particles hindered crack propagation as shown in Figure B.28(b). Lateral cracking and fracturing of a SiC particle are observed in the cross-sections in Figures B.28(c) and (d). Figure B.28: worn surfaces + 20wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 14: 30wt%Ni + 30wt%Ti + 40wt%SiC The graphs of the wear rate versus time for an untreated aluminium and aluminium laser alloyed wit Figure B.29. The average wear rate of the laser alloyed aluminium was 3.86?0.13x10-4cm3/sec. A reduction in wear rate of approximately 12% was achieved for the laser alloyed surface compared to the untreated aluminium alloy. (a) (c) APPENDIX B of Al samples laser alloyed with 40wt%Ni + 40wt%Ti cross-sections. h 30wt%Ni + 30wt%Ti + 40wt%SiC are (d) (b) 10?m 227 shown in 10?m APPENDIX B 228 Figure B.29: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC. SEM micrographs of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC are shown in Figure B.30. Shallow wear grooves are observed in Figures B.30(a) and (b). Cracks parallel to the wear groove are observed in figure B.30(b). Fracturing of SiC particles is observed in the cross-sections in Figures B.30(c) and (d). Figure B.30: worn surfaces + 40wt%SiC in (a) and (b) plan view; (c) and (d) (a) (c) APPENDIX B of Al samples laser alloyed with 30wt%Ni + 30wt%Ti cross-sections. (d) (b) 10?m 229 10?m APPENDIX B 230 B.2: Three body dry abrasion wear tests Pure Aluminium AA1200 The graph of wear rate versus time for untreated aluminium AA1200 is shown in Figure B.31. The average wear rate was 7.53?0.13x10-3cm3/min. Figure B.31: Graph of wear rate versus time for an untreated aluminium alloy. SEM micrographs of worn surfaces (plan view and cross-sections) of an untreated aluminium laser are shown in Figure B.32. Due to high ductility of aluminium, some SiO2 particles were embedded in the deformed microstructure. Grooves which were formed during 2body sliding wear tests were not observed as ploughing did not occur. Cracking, material removal and embedded SiO2 particles are observed in the plan views in Figures B.32(a) and (b). Plastic deformation occurred due to point indentation on the aluminium surface by the applied load and the SiO2 particles. Cracking in the deformed layer is observed in the cross- sections in Figures B.32(c) and (d). Figure B.32: worn surfaces plan view; (c) and (d) Alloy 1: 33.3wt%Ni + 33.3wt%Ti + 33.3wt% SiC The graphs of wear rate versus time for untreated aluminium and an aluminium laser surface alloyed with 33.3w Figure B.33. Approximately 44% alloying due to the intermetallic phases and metal matrix composites wear rate for the alloyed samples SiO2 SiO2 (a) (c) APPENDIX B of the untreated aluminium samples in (a) and (b) cross-sections Al. t%Ni + 33.3wt%Ti + 33.3wt%SiC are reduction in wear rate was achieve was 4.25?0.26x10-3 cm3/min. 10?m (d) (b) 231 shown in d after laser . The average SiO2 10?m APPENDIX B 232 Figure B.33: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC are shown in Figure B.34. Wear grooves which were formed by the repeated action of the SiO2 abrasives are observed in Figures B.34(a) and (b). These wear grooves were narrower and shallow compared to those formed during two body abrasion wear tests. Cracking inside and outside the wear grooves was observed in Figure B.34(b). SiC particles embedded in the alloyed layer during laser processing inhibited the growth of the grooves. The strong bonding between particles and the matrix ensured decohesion (interfacial debonding and pull-out) of SiC particles did not occur during wear. The randomly distributed SiC particles and the intermetallic phases act as load bearing constituents leading to improved wear resistance. The cross-sections in Figures B.34(c) and (d) show that plastic deformation, which occurred in the untreated Al, does not occur because of the presence of hard ceramic particles and intermetallic compounds. Lateral cracking and fracturing of SiC particles are observed in the cross-sections. SiO2 abrasive particles embedded in cracks are observed in Figure B.34(c). Figure B.34: worn surfaces 33.3wt%Ti + 33.3wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 2: 80wt%Ni + 15wt%Ti + 5wt%SiC The graphs of wear rate versus time laser surface alloyed with B.35. The average wear rate for the laser alloyed samples was 5.53?0.22x10 3cm3/min. Therefore, approximately 27% reduction in wear rate was achieved after laser alloying. (a) SiC particle (c) SiO2 APPENDIX B of Al samples laser alloyed with 33.3wt%Ni + cross for an untreated aluminium and 80wt%Ni + 15wt%Ti + 5wt%SiC are (d) (b) 10?m 233 -sections. aluminium shown in Figure - SiC particle 10?m APPENDIX B 234 Figure B.35: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC are shown in Figure B.36. Cracking and shallow grooves are observed in Figures B.36(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking and fracturing of the SiC particles are observed in the cross-sections in Figure B.36(c) and (d). Figure B.36: worn surfaces + 5wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 3: 50wt%Ni + 20wt%Ti + 30wt%SiC The graphs of wear rate versus time laser surface alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC B.37. A 31% reduction in wear rate was wear rate for the laser alloyed samples was 5.16?0.10x10 (a) (c) APPENDIX B of Al samples laser alloyed with 80wt%Ni + 15wt%Ti cross-sections. for an untreated aluminium and are achieved after laser alloying. -3cm3/min. (d) (c) 10?m 235 aluminium shown in Figure The average 10?m APPENDIX B 236 Figure B.37: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC are shown in Figure B.38. Cracking and shallow grooves are observed in Figures B.38(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking and fracturing of the SiC particles are observed in the cross-sections in Figure B.38(c) and (d). Figure B.38: worn surfaces + 30wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 4: 20wt%Ni + 30wt%Ti + 50wt% SiC The graphs of wear rate versus time laser surface alloyed wit B.39. The average wear rate for the 3cm3/min. Therefore, approximately 43% reduction in wear rate was achieved after laser alloying. (a) (c) APPENDIX B of Al samples laser alloyed with 50wt%Ni + 20wt%Ti cross-sections. for an untreated aluminium and h 20wt%Ni + 30wt%Ti + 50wt%SiC are shown in Figure laser alloyed samples was (d) (b) 10?m 237 aluminium 4.23?0.16x10- 10?m APPENDIX B 238 Figure B.39: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC are shown in Figure B.40. Cracking and shallow grooves are observed in Figures B.40(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.40(c) and (d). Figure B.40: worn surfaces + 50wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 5: 60wt%Ni + 30wt%Ti + 10wt% SiC The graphs of wear rate versus time laser surface alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC B.41. A 28% reduction in wear rate was observed for the laser alloyed s compared to the untreated aluminium. samples was 5.44?0.45x10 (a) (c) APPENDIX B of Al samples laser alloyed with 20wt%Ni + 30wt%Ti cross-sections. for an untreated aluminium and are shown in Figure The average wear rate for the laser alloyed -3cm3/min. (d) (b) 10?m 239 aluminium amples 10?m APPENDIX B 240 Figure B.41: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC are shown in Figure B.42. Cracking and shallow grooves are observed in Figures B.42(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral and transgranular cracking are observed in the cross-sections in Figure B.42(c) and (d). Figure B.42: worn surfaces + 10wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 6: 70wt%Ni + 20wt%Ti + 10wt%SiC The graphs of wear rate versus time laser surface alloyed with B.43. A 19% reduction in wear rate wear rate for the laser alloyed (a) (c) APPENDIX B of Al samples laser alloyed with 60wt%Ni + 30wt%Ti cross-sections. for an untreated aluminium and 70wt%Ni + 20wt%Ti + 10wt%SiC are shown in Figure was achieved after laser alloying. samples was 6.10?0.26x10-3cm3/min. (d) (b) 10?m 241 aluminium The average 10?m APPENDIX B 242 Figure B.43: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC are shown in Figure B.44. Cracking and shallow grooves are observed in Figures B.44(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.44(c) and (d). Figure B.44: worn surfaces + 10wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 7: 10wt%Ni + 70wt%Ti + 20wt%SiC The graphs of wear rate versus time for untreated aluminium and an aluminium laser surface alloyed with B.45. The average wear rate which is a 32% reduction in wear rate compared to the untreated aluminium. (a) (c) APPENDIX B of Al samples laser alloyed with 70wt%Ni + 20wt%Ti cross-sections. 10wt%Ni + 70wt%Ti + 20wt%SiC are for the alloyed surface was 5.13?0. (d) (b) 10?m 243 shown in Figure 21x10-3cm3/min 10?m APPENDIX B 244 Figure B.45: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC are shown in Figure B.46. Cracking is observed in Figures B.46(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross- sections in Figure B.46(c) and (d). Figure B.46: worn surfaces + 10wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 8: 70wt%Ni + 10wt%Ti + 20wt%SiC The graphs of wear rate versus time laser surface alloyed with B.47. A 20% reduction in wear rate was achieved after laser alloying and the average wear rate for the (a) (c) APPENDIX B of Al samples laser alloyed with 10wt%Ni + 70wt%Ti cross-sections. for an untreated aluminium and 70wt%Ni + 10wt%Ti + 20wt%SiC are shown in laser alloyed samples 6.02?0.22x10-3cm3 (b) (d) 10?m 245 aluminium Figure /min. 10?m APPENDIX B 246 Figure B.47: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC are shown in Figure B.48. Cracking and shallow grooves are observed in Figures B.48(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking, fracturing of SiC particles and transgranular cracking are observed in the cross-sections in Figure B.48(c) and (d). Figure B.48: worn surfaces + 20wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 9: 40wt%Ni + 20wt%Ti + 40wt%SiC The graphs of wear rate versus time laser surface alloyed with B.49. A reduction of approximately 82% in wear rate was achieved after laser alloying and the average wear rate for the laser alloyed samples was 1.30?0.26x10-3cm3/min. (a) (c) APPENDIX B of Al samples laser alloyed with 70wt%Ni + 10wt%Ti cross-sections. for an untreated aluminium and 40wt%Ni + 20wt%Ti + 40wt%SiC are shown in Figure (d) (b) 10?m 247 aluminium 10?m APPENDIX B 248 Figure B.49: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC are shown in Figure B.50. Cracking and shallow grooves are observed in Figures B.50(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.50(c) and (d). Figure B.50: worn surfaces + 40wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 10: 40wt%Ni + 30wt%Ti + 30wt%SiC The graphs of wear rate versus time laser surface alloyed with B.51. A 66% reduction in wear rate was achieved after laser alloying and the average wear rate for the laser alloyed samples was (a) (c) APPENDIX B of Al samples laser alloyed with 40wt%Ni + 20wt%Ti cross-sections. for an untreated aluminium and 40wt%Ni + 30wt%Ti + 30wt%SiC are shown in Figure 2.58?0.05x10- (d) (b) 10?m 249 aluminium 3cm3/min. 10?m APPENDIX B 250 Figure B.51: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC are shown in Figure B.52. Cracking and shallow grooves are observed in Figures B.52(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.52(c) and (d). Figure B.52: worn surfaces + 30wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 11: 20wt%Ni + 40wt%Ti + 40wt%SiC The graphs of wear rate versus time laser surface alloyed with B.53. A 82% reduction in wear rate was achieved after laser alloying and the average wear rate for the laser allo (a) (c) APPENDIX B of Al samples laser alloyed with 40wt%Ni + 30wt%Ti cross-sections. for an untreated aluminium and 20wt%Ni + 40wt%Ti + 40wt%SiC are shown in Figure yed samples was 1.36?0.21x10- (d) (b) 10?m 251 aluminium 3cm3/min. 10?m APPENDIX B 252 Figure B.53: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC are shown in Figure B.54. Cracking and shallow grooves are observed in Figures B.54(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.54(c) and (d). Figure B.54: worn surfaces + 40wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 12: 30wt%Ni + 40wt%Ti + 30wt%SiC The graphs of wear rate versus time laser surface alloyed with B.55. A 63% reduction in wear rate was achieved after laser alloying and the average wear rate for the laser alloyed samples was (a) (c) APPENDIX B of Al samples laser alloyed with 20wt%Ni + 40wt%Ti cross-sections. for an untreated aluminium and 30wt%Ni + 40wt%Ti + 30wt%SiC are shown in Figure 2.79?0.23x10- (d) (b) 10?m 253 aluminium 3cm3/min. 10?m APPENDIX B 254 Figure B.55: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 30wt%Ni + 40wt%Ti + 30wt%SiC are shown in Figure B.56. Cracking and shallow grooves are observed in Figures B.56(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.56(c) and (d). Figure B.56: worn surfaces + 30wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 13: 40wt%Ni + 40wt%Ti + 20wt%SiC The graphs of wear rate versus time laser surface alloyed with B.57. A 64% reduction in wear rate was achieved after laser alloying and average wear rate for the laser alloyed samples was (a) (c) APPENDIX B of Al samples laser alloyed with 30wt%Ni + 40wt%Ti cross-sections. for an untreated aluminium and 40wt%Ni + 40wt%Ti + 20wt%SiC are shown in Figure 2.71?0.33x10- (d) (b) 10?m 255 aluminium the 3cm3/min. 10?m APPENDIX B 256 Figure B.57: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 40wt%Ni + 40wt%Ti + 20wt%SiC are shown in Figure B.58. Cracking and shallow grooves are observed in Figures B.58(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking which resulted in chipping of the worn layer and fracturing of the SiC particle are observed in the cross-sections in Figure B.58(c) and (d). Figure B.58: worn surfaces + 20wt%SiC in (a) and (b) plan view; (c) and (d) Alloy 14: 30wt%Ni + 30wt%Ti + 40wt%SiC The graphs of wear rate versus time laser surface alloyed with B.59. A 78% reduction in wear rate was achieved after laser alloying and the average wear rate for the laser alloyed samples was (a) (c) APPENDIX B of Al samples laser alloyed with 40wt%Ni + 40wt%Ti cross-sections. for an untreated aluminium and 30wt%Ni + 30wt%Ti + 40wt%SiC are shown in Figure 1.63?0.09x10- 10?m (d) (b) 257 aluminium 3cm3/min. 10?m APPENDIX B 258 Figure B.59: Graph of wear rate versus time for an untreated aluminium and aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC. The microstructures of worn surfaces (plan view and cross-sections) of aluminium laser alloyed with 30wt%Ni + 30wt%Ti + 40wt%SiC are shown in Figure B.60. Cracking and shallow grooves are observed in Figures B.60(a) and (b) due to the repeated action of the SiO2 abrasive particles. The SiC particles embedded in the microstructure hindered the growth of the grooves. Lateral cracking is observed in the cross-sections in Figure B.60(c) and (d). Figure B.60: worn surfaces + 40wt%SiC in (a) and (b) plan view; (c) and (d) (a) (c) APPENDIX B of Al samples laser alloyed with 30wt%Ni + 30wt%Ti cross-sections. (d) (b) 10?m 259 10?m APPENDIX C: IMPACT This Appendix provides images of the fracture surfaces for impact tests which may be used for comparison purposes. The fracture was summarized in Chapter 5, Section 5.4. Alloy 1: 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC Fractured surfaces of samples laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC are shown in Figure C.1 Figure C.1: Fractured surfaces of samples laser alloyed with 33.3wt%Ni + 33.3wt%Ti + 33.3wt%SiC. Alloy 2: 80wt%Ni + Fractured surfaces of samples laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC are shown in Figure C.2 APPENDIX C TEST RESULTS the 14 alloys . 15wt%Ti + 5wt%SiC . 260 from the behaviour Figure C.2: Fractured surfaces of samples laser alloyed with 80wt%Ni + 15wt%Ti + 5wt%SiC Alloy 3: 50wt%Ni + 20wt%Ti Fractured surfaces of samples laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC are shown in Figure C.3. Figure C3: Fractured surfaces of samples laser alloyed with 50wt%Ni + 20wt%Ti + 30wt%SiC. Alloy 4: 20wt%Ni + 30wt%Ti + 50wt%SiC Fractured surfaces of samples laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC are shown in Figure C.4. APPENDIX C . + 30wt%SiC 261 Figure C.4: Fractured surfaces of samples laser alloyed with 20wt%Ni + 30wt%Ti + 50wt%SiC. Alloy 5: 60wt%Ni + 30wt%Ti + 1 Fractured surfaces of samples laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC are shown in Figure C.5. Figure C.5: Fractured surfaces of samples laser alloyed with 60wt%Ni + 30wt%Ti + 10wt%SiC Alloy 6: 70wt%Ni + 20wt%Ti + 10wt%SiC Fractured surfaces of samples las 10wt%SiC are shown in Figure C.6. APPENDIX C 0wt%SiC . er alloyed with 70wt%Ni + 20wt%Ti + 262 Figure C.6: Fractured surfaces of samples laser alloyed with 70wt%Ni + 20wt%Ti + 10wt%SiC. Alloy 7: 10wt%Ni + 70wt%Ti + 20wt%SiC Fractured surfaces of samples laser alloyed with 1 20wt%SiC are shown in Figure C.7. Figure C.7: Fractured surfaces of samples laser alloyed with 10wt%Ni + 70wt%Ti + 20wt%SiC. Alloy 8: 70wt%Ni + 10wt%Ti + 20wt%SiC Fractured surfaces of samples laser alloyed with 70wt%Ni + 10wt%Ti 20wt%SiC are shown in Figure C.8. APPENDIX C 0wt%Ni + 70wt%Ti + 263 + Figure C.8: Fractured surfaces of samples laser alloyed with 70wt%Ni + 10wt%Ti + 20wt%SiC. Alloy 9: 40wt%Ni + 20wt%Ti + 40wt%SiC Fractured surfaces of samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC are shown in Figure C.9. Figure C.9: Fractured surfaces of samples laser alloyed with 40wt%Ni + 20wt%Ti + 40wt%SiC. Alloy 10: 40wt%Ni + 30wt%Ti + 30wt%SiC Fractured surfaces of samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC are shown in Figure C. APPENDIX C 10. 264 Figure C.10: Fractured surfaces of samples laser alloyed with 40wt%Ni + 30wt%Ti + 30wt%SiC. Alloy 11: 20wt%Ni + 40wt%Ti + 40wt%SiC Fractured surfaces of samples laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC are shown in Figure C.11. Figure C11: Fractured surfaces of samples laser alloyed with 20wt%Ni + 40wt%Ti + 40wt%SiC. Alloy 12: 30wt%Ni + 40wt%Ti + 30wt%SiC Fractured surfaces of samples laser alloyed with 30wt%SiC are shown in Figure C. APPENDIX C 30wt%Ni + 12. 265 40wt%Ti + Figure C.12: Fractured 40wt%Ti + 30wt%SiC. Alloy 13: 40wt%Ni + 40wt%Ti + 20wt%SiC Fractured surfaces of samples laser alloyed with 20wt%SiC are shown in Figure C. Figure C.13: Fractured surfaces of sa 40wt%Ti + 20wt%SiC. Alloy 14: 30wt%Ni + 30wt%Ti + 40wt%SiC Fractured surfaces of samples laser alloyed with 40wt%SiC are shown in Figure C. APPENDIX C surfaces of samples laser alloyed with 30wt%Ni + 40wt%Ni + 13. mples laser alloyed with 40wt%Ni + 30wt%Ni + 14. 266 40wt%Ti + 30wt%Ti + Figure C.14: Fractured surfaces of samples laser 30wt%Ti + 40wt%SiC. APPENDIX C alloyed with 30wt%Ni + 267 APPENDIX E 268 APPENDIX D: PUBLICATIONS D1: Journal Papers 1. Mabhali, L.A.B.; Pityana, S. L.; Sacks, N. (2010); Laser surface alloying of aluminium (AA1200) with Ni and SiC powders; Materials and Manufacturing Processes; vol. 25; issue 12; pp. 1397-1403 (ISSN: 1042- 6914). 2. Mabhali, L.; Pityana, S.; Sacks, N. (2010); Laser alloying of Al with mixed Ni, Ti and SiC powders; Journal of Laser Applications; vol. 22; no. 4; pp. 121-126 (ISSN: 1042- 346X). 3. Mabhali, L.; Sacks, N.; Sliding wear of laser surface alloyed Aluminium AA1200 alloys; in preparation. 4. Mabhali, L.; Sacks, N.; Abrasive wear of laser surface alloyed Aluminium AA1200 alloys; in preparation. 5. Mabhali, L.; Sacks, N.; Impact toughness of laser surface alloyed Aluminium AA1200 alloys; in preparation. D2: Conference Papers 1. Mabhali, L.; Pityana, S.; Sacks, N. (2010); Laser alloying of Al with mixed Ni, Ti and SiC powders; Proceedings of the 4th Pacific International Conference on Applications of Lasers and Optics (PICALO); Wuhan, Peoples Republic of China; 23-25 March 2010, vol. 4; pp 6 (ISBN: 978-0- 912035-56-7). 2. Mabhali, L.; Pityana, S.; Sacks, N. (2010); Laser alloying of aluminium to improve surface properties; Proceedings of the 46th Conference of the Microscopy Society of Southern Africa (MSSA); Bela-Bela, Republic of South Africa; 21-25 July; vol. 40; pp 76 (ISBN: 0-620-35056-3).