<%BANNER%>

Investigating Feature Formation by Thermoplastic Forming of Zirconium Based Bulk Metallic Glass on Stainless Steel Molds

Permanent Link: http://ufdc.ufl.edu/UFE0042468/00001

Material Information

Title: Investigating Feature Formation by Thermoplastic Forming of Zirconium Based Bulk Metallic Glass on Stainless Steel Molds
Physical Description: 1 online resource (43 p.)
Language: english
Creator: Mcintyre, Daniel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A relatively new material, bulk metallic glass is an amorphous metal that has mechanical properties that surpass aluminum and steel alloys. These enhanced properties allow this material to be considered in many applications ranging from microelectromechanical systems (MEMS) to structural uses, to electronics. Because of the ability to be thermoplastically formed, metallic glasses can be molded into complex shapes that conventional metals cannot achieve. The limiting factors for production for metallic glasses at this time include their cost and the ease of forming large shapes all while keeping a completely amorphous structure. This thesis focuses on the formation of features on the order of 200?m that were machined into a stainless steel mold. A total of three zirconium based metallic glass alloys were chosen and their ability to form small intricate features on stainless steel molds during molding was evaluated. A load frame fitted with heaters was utilized to enable the thermoplastic forming of metallic glass alloys in a laboratory setting. The reactivity between the metallic glass and the mold as well as the atmosphere is observed and evaluated. Along with this, the alloys require a relatively high cooling rate to achieve their vitreous nature. To determine crystallinity, an X-ray diffractometer was used and a scanning electron microscope (SEM) was utilized to examine the formed features. The molds were examined with an SEM to analyze the wear associated with the molding of the alloys.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Mcintyre.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Bourne, Gerald R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042468:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042468/00001

Material Information

Title: Investigating Feature Formation by Thermoplastic Forming of Zirconium Based Bulk Metallic Glass on Stainless Steel Molds
Physical Description: 1 online resource (43 p.)
Language: english
Creator: Mcintyre, Daniel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A relatively new material, bulk metallic glass is an amorphous metal that has mechanical properties that surpass aluminum and steel alloys. These enhanced properties allow this material to be considered in many applications ranging from microelectromechanical systems (MEMS) to structural uses, to electronics. Because of the ability to be thermoplastically formed, metallic glasses can be molded into complex shapes that conventional metals cannot achieve. The limiting factors for production for metallic glasses at this time include their cost and the ease of forming large shapes all while keeping a completely amorphous structure. This thesis focuses on the formation of features on the order of 200?m that were machined into a stainless steel mold. A total of three zirconium based metallic glass alloys were chosen and their ability to form small intricate features on stainless steel molds during molding was evaluated. A load frame fitted with heaters was utilized to enable the thermoplastic forming of metallic glass alloys in a laboratory setting. The reactivity between the metallic glass and the mold as well as the atmosphere is observed and evaluated. Along with this, the alloys require a relatively high cooling rate to achieve their vitreous nature. To determine crystallinity, an X-ray diffractometer was used and a scanning electron microscope (SEM) was utilized to examine the formed features. The molds were examined with an SEM to analyze the wear associated with the molding of the alloys.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Mcintyre.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Bourne, Gerald R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042468:00001


This item has the following downloads:


Full Text

PAGE 1

1 INVESTIGATING FEATURE FORMATION BY THERMOPLASTIC FORMING OF ZIRCONIUM BASED BULK METALLIC GLASS ON STAINLESS STEEL MOLDS By DANIEL MCINTYRE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FUL FILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Daniel McIntyre

PAGE 3

3 ACKNOWLEDGMENTS I would like to thank my parents for constant support and dedication to me. A special thanks goes to Dr. Geral d Bourne, my advisor for advice and guidance during my graduate education. I would like to also thank Dr. Gregory Sawyer for his support on this project as well as Dr. Tony Schmitz for his work.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF FIGURES ................................ ................................ ................................ .......... 5 LIST OF ABBREVIATIONS ................................ ................................ ............................. 7 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 Background on Metallic Glass ................................ ................................ ................ 10 History ................................ ................................ ................................ .............. 10 Values of BMG in Production ................................ ................................ ........... 11 Challenges in Production ................................ ................................ .................. 13 Problems Associated with Crystallinity ................................ ............................. 14 Motivation ................................ ................................ ................................ ......... 16 2 PROCEDURES ................................ ................................ ................................ ...... 17 Alloys Used ................................ ................................ ................................ ............. 17 Preparation of the Alloy ................................ ................................ ........................... 17 Arc Melting and Creating the Alloy ................................ ................................ ... 17 Creating the Proper Dimensions ................................ ................................ ...... 18 Steel Mold and the Load Frame ................................ ................................ .............. 19 Steel Mold ................................ ................................ ................................ ........ 19 Load Frame ................................ ................................ ................................ ...... 20 Measuring Mold Wear ................................ ................................ ............................. 24 SEM Analysis ................................ ................................ ................................ ... 24 Determi ning Crystallinity ................................ ................................ ................... 27 3 RESULTS AND DISCUSSION ................................ ................................ ............... 28 The Formation of Small Features ................................ ................................ ........... 28 Corrosion of Metallic Glass During Molding ................................ ............................ 30 Wear on the Molds Caused by Thermoplastic Molding ................................ ........... 32 4 CONCLUSION ................................ ................................ ................................ ........ 38 LIST OF REFERENCES ................................ ................................ ............................... 41 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 43

PAGE 5

5 LIST OF FIGURES Figure page 2 1 Copper drop mold containing a ca sted piece of metalli c glass ........................... 18 2 2 Schematic of the molding features to be machined into a mold ......................... 19 2 3 SEM image of the mold design that was machine d into a stainless steel ........... 20 2 4 Load frame setup showing plat ens and water cooling system ............................ 21 2 5 Schematic of the parts used wit h the load frame for molding ............................. 22 2 6 Cross section al view of the molding setup ................................ .......................... 22 2 7 Water cooling system temperature profile du ring a test ................................ ...... 23 2 8 Water quenching temperature profile during a test ................................ ............. 24 2 9 SEM secondary electron image illustrating the measur ement taken for circle diameter ................................ ................................ ................................ ............. 25 2 10 SEM secondary electron image illustrating the measurement taken for finger radius ................................ ................................ ................................ .................. 25 2 1 1 SEM secondary electron image illustrating the measurement taken for channel width ................................ ................................ ................................ ...... 26 2 12 SEM secondary electron image illustrating the measurement taken for finger width ................................ ................................ ................................ ................... 26 2 13 XRD scan of Vit 1 showing a highly amorphous piece of bulk metallic glass ..... 27 2 14 XRD scan of Vit 106a showing a crystalline peak ................................ ............... 27 3 1 SEM secondary electron image showing machine marks replicated on a molded piece of metallic glass ................................ ................................ ............ 28 3 2 An SEM secondary electron image of a successful m olding of bulk metallic glass ................................ ................................ ................................ ................... 29 3 3 An XRD scan of the Vit DH alloy after molding illustrating the retained amorphous structure. ................................ ................................ .......................... 30 3 4 Optical microscope image of a molded Vit 106a piece with surface corrosion ... 31 3 5 Optical microscope image demonstrating the good surface finish of the Vit DH alloy ................................ ................................ ................................ .............. 31

PAGE 6

6 3 6 Circle diameter dimensional changes with respect to molding attempts of Vit DH ................................ ................................ ................................ ...................... 32 3 7 Finger width dimensional chan ges with respect to molding attempts of Vit DH .. 32 3 8 Finger radius dimensional changes with respect to molding attempts of Vit DH ................................ ................................ ................................ ...................... 33 3 9 Channel width dimensional changes with respect to molding attempts of Vit DH ................................ ................................ ................................ ...................... 33 3 10 SEM secondary electron image of a failed demolding attempt ........................... 34 3 11 SEM secondary electron image showing an angled view of a mold finger demonstrating the amount of deformation after a failed molding attempt ........... 35 3 12 SEM second ary electron image showing the rough surface after a BN aerosol was used as a lubricant ................................ ................................ ......... 36 3 13 SEM secondary electron image of the rough surface after the use of a BN aerosol lubricant ................................ ................................ ................................ 37

PAGE 7

7 LIST OF ABBREVIATION S BMG Bulk metallic glass SEM Scanning electron microscope XRD X Ray diffractometer

PAGE 8

8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfill ment of the Requirements for the Degree of Master of Science INVESTIGATING FEATURE FORMATION BY THERMOPLASTIC FORMING OF ZIRCONIUM BASED BULK METALLIC GLASS ON STAINLESS STEEL MOLDS By Daniel McIntyre December 2010 Chair: Gerald Bourne Major: Materials Science and Engineering A relatively new material, bulk metallic glass is an amorphous metal that has mechanical properties that surpass aluminum and steel alloys. These enhanced properties allow this material to be considered in many applications rangin g from microelectromechanical systems (MEMS) to structural uses to electronics. Because of the ability to be thermoplastically formed, metallic glasses can be molded into complex shapes that conventional metals cannot achieve. The limiting factors for p roduction for metallic glasses at this time include their cost and the ease of forming large shapes all while keeping a completely amorphous structure. This thesis focuses on the formation of features on the order of 200 m that were machined into a stain less steel mold. A total of three zirconium based metallic glass alloys were chosen and their ability to form small intricate features on stainless steel molds during molding was evaluated. A load frame fitted with heaters was utilized to enable the therm oplastic forming of metallic glass alloys in a laboratory setting. The reactivity between the metallic glass and the mold as well as the atmosphere is observed and evaluated. Along with this, the alloys require a relatively high cooling rate to achieve t heir vitreous nature. To determine crystallinity, an X ray diffractometer was

PAGE 9

9 used and a scanning electron microscope (SEM) was utilized to examine the formed features. The molds were examined with a n SEM to analyze the wear associated with the molding of the alloys.

PAGE 10

10 CHAPTER 1 INTRODUCTION Background on Metallic Glass History The concept behind met allic glasses began in the 1950 s when an idea was devised through thermodynamics and kinetics that if a molten metal or metallic alloy is cooled quickly from a liquid, the amorphous structure that was present in the liquid would remain since the alloy would not have sufficient time to crystallize [22]. It was suggested that an alloy, consisting of metals with different crystal structures, would aide in slowing down the kinetics to stay amorphous upon rapid quenching from a liquid. It was theorized that since there would be no crystalline structure, there would be no slip planes or dislocations. This would increase the strength of the alloy and would warrant it to be used in structural applications. In 1960, Duwez created the first metallic glass by quenching a Au Si liquid alloy to form an amorphous solid [1]. Following this discovery, lead based ternary alloys were produced to allow for the liquid alloy to b e cooled slower but still achieve an amorphous structure [1] By adding metals with different crystal structures, melting temperatures and atomic sizes, the kinetics of the liquid are slowed down enough to allow for the production of a metallic glass when cooled at a fast enough rate. These alloying additions create a lower temperature eutectic point for the system, suppressing the melting temperature [1]. During this research, the discovery of a glass transition temperature was made in these alloys [2]. Further research was performed in the following decades that resulted with easy to form metallic glasses and the realization that they could have far reaching enginee ring implications. In the 1990 s, Peker and Johnson created a quinary alloy, Vitreloy 1, which

PAGE 11

11 consists of Zr Ti Cu Ni Be, which could be cast several centimeters in thickness, which lead to the term bulk metallic glass [3]. The dimensions of bulk metallic glass can be increased by either increasing the cooling rate or by slowing the kinetic s, since both resisting crystallization. Continuing research has led to alloys with increasing critical casting thicknesses because of a decreased critical cooling rate to lock in the amorphous structure, furthering their applications [4]. The first tec hniques to create metallic glass produced samples that had minimal use for structural parts. Atomization was one of these processes. Atomization involved melting an alloy and forcing the melt through a nozzle that separated the liquid stream into small d roplets, almost like a mist [21]. These droplets would spray into a chilled atmosphere and would cool quickly enough to stay amorphous. What is left is a powder of fine metallic glass particle. Another technique is splat forming [21]. This involved imp the plate and cool quickly, leading to a metallic glass flake [21]. One other way was called melt spinning. A chilled copper wheel was spun around and liquid metal wa s poured onto the wheel [21]. The liquid would cool quickly and exit off of the wheel as a thin sheet [21]. Values of BMG in Production Metallic glasses can be thermoplastically formed just like a polymer. If heated to a certain temperature, they will become viscous and will flow under a load, allowing them to be pressed into a mold with a desired shape. The viscosity of the liquid is temperature dependant and can vary by orders of magnitude when heated [7]. Essentially, metallic glasses are supercool ed liquids, leading to this property. Thermoplastic molding may be able to cut down on start up costs for a company looking

PAGE 12

12 to use metallic glasses for their part [7]. This process is inhibited by the cooling rates necessary to resist crystallization and a proper cooling system will be necessary to reach the critical cooling rates necessary. The cast parts can also contain defects caused by the high velocity flow of the material into the mold cavities [5]. Another benefit is the ability to obtain net sha pe forming and good surface finishes on these alloys [6]. The lack of a phase transformation from a liquid to a solid results in no change in volume associated with the transformation [22]. The amorphous qualities of the metallic glass lead to thermal ex pansion coefficient of less than 0.5% [7]. This means that when molding a piece, it will retain its dimensions after the cooling process. Little to no secondary machining would be needed to achieve proper dimensional tolerances on the molded part [7]. M etallic glasses are known to replicate a molding surface very well. They are able to replicate micrometer and nanometer features. If the molding surface is highly polished then the molded piece will be as well. Even the smallest machine marks can be rep roduced so care has to be involved to make the molds properly. This quality of metallic glass also allows less secondary processing to improve the surface finish after molding since the mold dictates the surface finish. Many mechanical properties of met allic glasses are superior to that of common commercial metals such as steel and nickel. Metallic glasses exhibit an increased tensile strength, ultimate tensile strength, hardness and a higher elastic limit [8]. Many of these improved properties arise f rom the fact that in the amorphous structure, dislocations do not form in addition to this, there is no slip plane so shear of the material is inhibited, leading to increased mechanical properties [9]. These enhanced properties make metallic glass a good choice for a part that has a load applied to it. These alloys

PAGE 13

13 also possess increased resistance to corrosion and are electrically conductive [7,10]. Many of the alloys contain elements, including beryllium that are oxygen getters and this leads to a resi stance to oxidize at elevated temperatures [11]. Beryllium is toxic when in a powder form so it must be decided if alloys containing this element can be used. Challenges in P roduction During molding, the alloy is subjected to elevated temperatures, which can cause corrosion. The titanium in many alloys is insufficient to prevent corrosion by itself and it needs to be with another oxygen getter to provide enough protection [11]. To prevent oxidation in non beryllium containing alloys, a controlled atmosph ere must be involved This can be achieved by placing a chamber around the molding apparatus and backfilling this space with an inert gas such as argon. This would eliminate the problem of oxidation in the non beryllium containing alloys. In a productio n process, the use of a controlled atmosphere can reduce the efficiency of the process. Each molding attempt would require a backfilling of argon and a subsequent removal of argon to extract the part from the mold and set up a new molding attempt. This w ould slow the system down and will increase the cost to produce the parts. Another reason to use a controlled atmosphere is because as the concentration of dissolved oxygen increases in the alloy, the amount of time to process before crystallization occur s decreases because of the increased propensity for nucleation, leading to crystallization. [12]. With a controlled atmosphere, more time can be taken to mold and cool the part than if the process was performed in regular atmosphere conditions. Metallic g lass alloys need to be cooled quickly from their molding temperature in order to achieve an amorphous structure. Each alloy has a critical cooling necessary to keep its amorphous nature [7]. Cooling systems in the molding system can be

PAGE 14

14 implemented to red uce the heat right after molding to ensure that the critical cooling rate is achieved. Mold compatibility with the alloy poses a problem for processing as well. Reactions between the mold and the alloy can lead to problems that will make the molded part useless. Crystallization can occur, usually on the mold surfaces first, in the alloy when it is held too long at the molding temperature Making all surfaces of the mold smooth and polished can minimize this problem. At molding temperatures, diffusion i n the alloy and mold can lead to diffusion bonding between the parts, ultimately fusing them together. Diffusion of elements from the mold into the metallic glass can also lead to compositional changes in the metallic glass. This would ultimately change the properties of the alloy in the diffusion layer and could also cause crystallization. This crystallization can lead to permanent mold deformation and a rejection of the molded part. To prevent this, tests must be performed to confirm that in a product ion process, there will be no reaction between the mold and the metallic glass that would compromise either the part or the mold. Problems Associated with Crystallinity Crystallinity in this alloy is unwanted and adversely affects the properties that are a ssociated with metallic glasses. A significant degree of crystallinity prior to molding will result in a failed molding attempt since the alloy will not become viscous when subjected to an elevated temperature. When unable to flow, damage to the mold wil l occur when a load is applied. This introduces a problem for the surface finish of a molded piece. Metallic glass can replicate a surface very well when molded and the presence of crystallization on the surface inhibits this property and causes a dull f inish on the molded piece. The different crystallographic grains on the surface reflecting light

PAGE 15

15 differently cause this dull finish. When the crystalline nature is produced after or during molding, the mechanical, electrical and corrosion properties that are desired in the alloy no longer exist. The formation of a crystalline structure forms in the alloy when three things happen either individually or together. The first is when the alloy is heated to above its crystallization temperature. This temperat ure is determined using a DSC and can be avoided during molding [14,15]. Above the crystallization temperature, the kinetics of the alloy are fast enough to enable nucleation and growth or crystals upon cooling [16]. The second way is to cool the alloy a t a slower rate than the critical cooling rate. A slow rate will allow atomic movements and will result in nucleation of crystals. A slow cooling rate is an issue during the drop molding process of the alloy. The drop mold and hearth are two separate pi eces and do not form a good contact to one another. This results in a poor heat transfer between the two pieces. When the molten alloy drops into the mold, the cooling rate is thus reduced and can be slow enough to cause crystallization in parts of the a lloy, usually towards the top of the drop mold where the most heat is contained. Each alloy of metallic glass has different crystallization temperatures and critical cooling rates. To avoid crystallization of the alloy, this temperature must not be excee ded. The third way to cause crystallinity in these alloys is to allow oxygen into the composition. This can occur if the metal is at an elevated temperature in ambient atmosphere, which occurs during molding. The higher the concentration of oxygen, the less time available to work and mold the metal before crystallization occurs. The presence of oxygen at high temperatures can lead to oxidation of the piece. Oxidation ruins the surface finish and thus makes the use of metallic glass less desirable.

PAGE 16

16 The choice of alloy is important since oxygen can be absorbed into the alloy at high temperatures. Some metallic glass alloys contain beryllium, which helps to reduce this problem. Beryllium is an oxygen getter so when the alloy is subjected to elevated temp erature, the oxygen will preferentially be attracted to the beryllium and will prevent a shortened processing time as well as surface corrosion. Unfortunately, beryllium is a toxic material so it is undesirable to choose a beryllium containing alloy for us e in a product that would be used by people. Most beryllium containing alloys have a lower critical cooling rate than an alloy deficient of beryllium. Because of this, keeping an alloy amorphous after molding is more difficult and requires the use of a q uench tank rather than a platen cooling system that is attached to the load frame. In a mass production based operation, this issue would need to be addressed with a better system to cool the alloy post molding. Motivation This research will focus on for ming features approximately 200 m in size in two beryllium containing alloys and one non beryllium containing alloy and investigating whether molding and demolding can be achieved with a stainless steel mold for production processes. The features machined in the mold will be measured after each molding attempt to record deviations in dimensions. Reactions of the alloys, during molding, with the mold and the atmosphere will be also evaluated to determine if they affect the molding process.

PAGE 17

17 CHAPTER 2 PR OCEDURES Alloys Used Three zirconium based alloys were chosen for the molding of bulk features. The alloys were chosen based on their mechanical properties as well as on the critical cooling rates and crystallization temperature. The alloys were Vit 1 (Z r 41.2 Be 22.5 Ti 13.8 Cu 12.5 Ni 10 ), Vit DH (Zr 35 Ti 30 Be 27.5 Cu 7.5 ) and Vit 106a (Zr 58.5 Cu 15.6 Ni 12.8 Al 10.3 Nb 2.8 ) [7,23]. In order to keep an amorphous structure in the these alloys the part must be rapidly cooled from the liquid at a rate from 1 C to 10 C per second deemed the critical cooling rate The critical cooling rates of these alloys are low enough so that an amorphous structure can be obtained simply by quenching in water directly after molding. Preparation of the Alloy Arc Melting and Creating the Alloy The alloys were obtained and produced from outside sources. Each alloy was created from high purity powders to reduce impurities. A bell jar arc furnace was used to melt the powders. The chamber of the arc melter was backfilled with argon to minim ize the amount of oxygen present prior to melting the alloy. A titanium getter was struck with an arc and melted prior to melting the alloy to remove any residual o xygen in the chamber that would compromise the alloy. A copper water chilled hearth was us ed to cool the molten alloy as quickly as possible to prevent crystallization. The final alloy was in the shape of a small button that is a result of the hemispheres machined into the hearth to allow the pooling and consolidation of the melt.

PAGE 18

18 Creating th e Proper Dimensions A low speed diamond saw was used to section the arc melted button of metallic glass into 3.5 to 5.5 g pieces. A copper drop mold, Figure 2 1, has a channel for casting to fit that amount of material in the mold. Figure 2 1. Copper drop mold containing a casted piece of metallic glass [7] Two drop molds were used to melt the alloy into the desired shape. The first drop mold had the melted alloy sitting above the water chilled hearth, separated by the drop molds copper lining about 5 mm thick as seen in Figure 2 1 above The second drop mold had a through cut that allowed the melted alloy to directly touch the hearth during cooling. The second drop mold was created for Vit 106a, which required a faster cooling rate to ensure an amo rphous structure. Similar to before, the chamber of the arc melter was backfilled with argon 3 times and a titanium getter was used to ensure the oxygen concentration in the chamber was minimal. To ensure a high rate of cooling, the copper water chilled hearth was used.

PAGE 19

19 Steel Mold and the Load Frame Steel Mold The material chosen for the mold was a 316 stainless steel. Stainless steel was chosen for its relatively cheap cost and corrosion resistance. The mold was a cylinder 25mm in diameter and 6.3 mm thick made from stainless steel. On one face of the mold, a design was machined into it as seen in Figure 2 2 and 2 3. The design was simply four fingers, each with a channel and hemisphere machined into them. Figure 2 2. Schematic of the molding feat ures to be machined into a mold 5 mm SQ

PAGE 20

20 Figure 2 3. SEM image of the mold design that was machined into a stainless steel Load Frame The loa d frame as seen in Figure 2 4 was capable of applying a maximum load of 5000 N. The load frame was first build by Jeff Bardt for his research to develop a similar process for silicon molds [7] The top and bottom platens contain cartridge heaters inside of them and the whole setup is demonstrated in Figure 2 5. The heaters provide adequate heating to the platens, which t hen heat metallic glass in to a viscous enough state to be pressed into a mold. The time necessary to bring the platens and

PAGE 21

21 molding pieces to a proper operating temperature is approximately 60 seconds This relatively quick time to heat up reduced the opp ortunity for both corrosion and any reactivity that can exist between any of the molding materials and the metallic glass A load of 2000 N was applied to the mold at the designated temperature of around 350 C depending on the alloys glass transition te mperature The amount of pressure was load controlled so when the metallic glass alloy starts to flow, a constant pressure was applied, which allows a uniform flow of the alloy into the mold [7]. There were 4 parts accompanying the mold that were used to ensure proper alignment and pressure from the load frame on the metallic glass and the mold as seen in a simplified Figure 2 6. Essentially, the mold and metallic glass were placed onto a platen and a ram directed the force down onto these pieces for mold ing. Figure 2 4. Load frame setup showing platens and water cooling system [7]

PAGE 22

22 Figure 2 5. Schematic of the parts used with the load frame for molding [7] Figure 2 6. Cross sectional view of the molding s etup [7] F(t) Mold pocket Ram BMG Mold Heating/cooling base for T(t)

PAGE 23

23 A water cooling system was implemen ted into the top and bottom platen to provide sufficient cooling to keep the alloy amorphous. The cooling system starts when the load is removed after the molding has finished. If a faster cooling rate t han can be achieved by the system is needed, the mo ld and accessories can be quickly dropped into a water filled quench tank. The cooling rate for the quench tank is much greater than the cooling system, which is seen in comparing Figure 2 7 and 2 8. The cooling system results in a maximum cooling rate o f approximately 0 .87 C/sec. The water quench rate is essentially instantaneous and thus cannot be quantified. Figure 2 7. Water cooling system temperature profile during a test

PAGE 24

24 Figure 2 8. Water quenching temperature profile during a test Measurin g Mold Wear SEM Analysis A scanning electron microscope (SEM) was used to obtain secondary elec tron images of the mold that were then analyzed and measured to determine any changes of dimension. The SEM was operated at 30kV at a pressure of less than 5e 5 mbar. The mold was analyzed prior to molding and repeatable measurements were recorded These measurements were taken after each success ive molding of the alloy. The amount of deformation caused by molding of Vit DH was analyzed for this process and a total of four measurements were taken to measure deformation and they are represented as a yellow line in Figures 2 9 to 2 12.

PAGE 25

25 Figure 2 9. SEM secondary electron image illustrating t he measurement taken for circle diameter Figure 2 10. SEM secondary electron image illustrating t he measurement taken for finger radius

PAGE 26

26 Figure 2 11. SEM secondary electron image illustrating t he measurement taken for channel width Figure 2 12. SEM secondary electron image illustrating t he measurement taken for finger width

PAGE 27

27 Determining Crystallinity To ensure the alloy used in th e molding process was compl etely amorphous, the samples were scanned in a powder X Ray diffractometer (XRD) before and after the molding process as shown in Figures 2 13. In circumstances that show post molding crystallinity like in Figure 2 14, the temperature used during molding can be altered to ensure an optimum molding temperature for that specific alloy. Figure 2 13. XRD scan of Vit 1 showing a highly amorphous piece of bulk metallic gl ass Figure 2 14. XRD scan of Vit 106a showing a crystalline peak

PAGE 28

28 CHAPTER 3 RESULTS AND DISCUSSI ON The Formation of Small Features The two beryllium containing metallic glasses replicate d micrometer sized features. An example of this can be seen in Figu re 3 1 showing micrometer sized machine marks replicated on a molded piece of Vit DH alloy A successful molding of the feature is seen in Figure 3 2 where the entire piece molded except for one hemisphere on one finger. This was the result of a lack of m aterial and not a failure to mold. These two alloys also remained amorphous through molding confirmed by a XRD scan shown in Figure 3 3 Vit 106a the non beryllium containing alloy, w as also able to form the features in the mold, but t he molding was u nsuccessful because the alloy crystallized during molding A XRD scan de monstrating crystallinity in Vit 106a after molding is shown in Figure 2 14. Altering the molding temperature to prevent crystallization in the alloy was unsuccessful. Figure 3 1. SEM secondary electron image showing m achine marks replicated on a molded piece of metallic glass

PAGE 29

29 F igure 3 2. An SEM secondary electron image of a successful molding of bulk metallic glass. The top finger not moldin g a hemisphere was a result of a lack of material and not a failed molding

PAGE 30

30 Figure 3 3. An XRD scan of the Vit DH alloy after molding illustrating the retained amorphous structure. Corrosion of Metallic Glass During Molding The a lloys that contain ed beryllium were not susceptible to oxidatio n during molding. Vit 106a formed a visible surface oxide during molding. An example of surface oxidation can be seen in the optical microscope image in Figure 3 4 which was a molded sample of Vit 106 a. As a comparison, Figure 3 5 is a molded sample of Vit DH which shows a polished metallic surface The molding was partially successful for

PAGE 31

31 Vit 106a in that the features were formed but the layer of oxide negatively affected the surface finish and compromised the part Figure 3 4 Optical microscop e im age of a molded Vit 106a piece with surface corrosion Figure 3 5 Optical microscope image demonstrating the good surface finish of the Vit DH alloy

PAGE 32

32 Wear on the Molds Caused by Thermoplastic Molding The stresses and strains associated with molding t he metallic glass were not sufficient to cause significant changes in the dimensions of the molds even after multiple molding attempts with VIT DH, as seen in Figure 3 6 to 3 9 Figure 3 6 Circle diameter dimensional changes with respect to molding a ttempts of Vit DH Figure 3 7 Finger width dimensional changes with respect to molding attempts of Vit DH

PAGE 33

33 Figure 3 8 Finger radius dimensional changes with respect to molding attempts of Vit DH Figure 3 9 Channel width dimensional changes with resp ect to molding attempts of Vit DH The change in dimensions of the mold during the molding of Vit DH is on the order of 1 3%. The amount of deformation is relatively low and demonstrates that multiple

PAGE 34

34 molding attempts can be performed before the mold is determined to be unusable because of dimensional considerations. The water cooling system on the load frame was tested with a sample of Vit DH to determine if the cooling rate was sufficient to prevent crystallization. The XRD scan after molding showed a definite crystalline peak, signifying that the water c ooling system does not cool the metallic glass faster than the critical cooling rate for that alloy. The de molding of the part also presents some difficulties and an example of a failed d emolding is pr esented in Figure 3 10 Many times during the molding of, because of the complex design of the mold, the metallic glass will be frictionally bound to the mold. A relatively large force must be applied to the metallic glass in order to separate it from th e mold. During this mechanical process forces are imparted onto the mold causing plastic deformation on the relatively soft stainless steel an example of this is shown in Figure 3 11 Along with frictional effects, diffusion bonding occurred with the mo ld and metallic glass Because of the high temperatures involved the kinetics are increased and diffusion between the viscous metallic glass and the stainless steel mold walls can occur. The result is the inability to remove the sample from the mold. Figure 3 10 SEM secondary electron image of a failed demolding attempt. Part of the metallic glass did not separate from the mold because of a lack of material

PAGE 35

35 Figure 3 11 SEM secondary electron image showing an angled view of a mold finger demons trating the amount of deformation after a failed molding attempt Lubrication was suggested to prevent difficulties in the de molding process. The lubrication used was a boron nitride aerosol. Characterization of the BN particles found that particles as large as 10 m in diameter existed The surface finish on the part was negatively influenc ed by the lubrication because it interfered with the ability of the metallic glass to replicate the surface of the mold, which is visible in Figures 3 1 2 and 3 1 3

PAGE 36

36 Figure 3 1 2 SEM secondary electron image showing the rough surface after a BN aerosol was used as a lubricant

PAGE 37

37 Figure 3 1 3 SEM secondary electron image of the rough surface after the use of a BN aerosol lubricant

PAGE 38

38 CHAPTER 4 CONCLUSION Metallic glasses provide enhanced performance over many of their metallic counterparts such as steel and aluminum. While metallic glass alloys are relatively expensive and more diffi cult to fabricate, they are able to be formed into smaller parts. C ompared to common load bearing metals such as steel and aluminum, metallic glasses generally have a higher yield strength, ultimate tensile strength, a lower metallic glasse s that provide improved mechanical properties over the mechanical fatigue limitations of the other materials used in the devices [17,18]. The corrosion properties of these alloys trump common metals as well. The oxygen getters common in the composition o f many metallic glasses protect against oxidation of the alloy. Metallic glasses also possess the ability to be thermoplastically formed using common procedures of vacuum die and suction casting [5,20]. At elevated temperatures, these alloys flow viscou sly and this property allows the alloy to be processed in a similar manner to polymers that have been made in this manner for many years. Because of the viscous flow of the material, the molding of complex parts can be achieved. Such parts would be too h ard or expensive to produce with c ommon processing techniques for crystalline metals. In addition, metallic glass alloys inherently have a relatively low coefficient of thermal expansion and do not change phase, so there is no change in volume during mold ing Because of this, net shape parts can be molded in one step and will not have to be machined down to proper dimensional tolerances This means a higher efficiency and lower cost for production. Furthermore, these alloy s are able to replicate feature s and surfaces very well While molding, if the

PAGE 39

39 design in the mold is highly polished then the metallic glass surface will be as well, reducing any secondary finishing. The molding of the three zirconium based metallic glass alloys in ambient atmosphere u sing 316 stainless steel molds was investigated to determine the feasibility in a production process. Mold part feature reproduction and mold wear was monitored and characterized for 8 molding attempts with Vit DH, a beryllium containing alloy. The featur es that were molded with Vit 1 and Vit DH showed ease of release, reproducibility and minimal mold wear. Molding attempts with Vit 106a, a beryllium free alloy, were unsuccessful due to crystallization and oxidation of the sample Molding of the two bery llium containing alloys was successfully demonstrated in an ambient atmosphere while molding attempts with the beryllium alloy was unsuccessful. The amount of wear from the molding process on 316 stainless steel molds was not a significant issue when moldi ng Vit DH. Stainless steel is a viable mold material for the thermoplastic forming of metallic glass. When the proper molding temperatures are used, the amount of wear on the mold after each molding attempt was between 1% and 3% This is suitable for a large production run where the most uses out of a single mold is needed for the cost efficiency. Another benefit to metallic glass processing is the fact that simple alloys such as stainless steel or nickel can be used as mold materials, making the proces s even more cost efficient. To reduce the overall wear on the mold and extend its lifetime, lubrication can be used to aid in the de molding process. The issues with using lubrication are surface finish and reactivity with the alloy. While molding on th e micron scale, the smallest particle of contamination from the lubrication can cause adverse affects on the surface finish of the part. Elements in the alloy can

PAGE 40

40 also react with certain elements in the lubrication and can create unwanted phases that can negatively affect the properties of the alloy. Future work on these alloys includes research into non beryllium containing alloys that do not oxidize under elevated temperatures and have a large processing window This would eliminate the need for a con trolled atmosphere around the molding apparatus and will cut costs to produce parts in a production run. The use of a controlled atmosphere will also be examined. Research into the causes of crystallization and oxidation in beryllium free alloys will be investigated to explore a solution. The load frame will be setup to allow for an argon containment unit around the platens an d mold to eliminate oxidation. A process to automate the molding and de molding, all while allowing the use of the controlled atmo sphere chamber will be evaluated. These processes will be further researched as to whether they influence the amount of mold wear caused by molding and de molding of the part.

PAGE 41

41 LIST OF REFERENCES [1] R. DeHoff. Thermodynamics in Mat erials Science. Secon d Edition. Taylor & Francis Group, LLC. ( 2006 ) [2] W Klement, R H Willens P Duwez, Non crystalline structure in silicon alloys. Nature 869 8 70 187 (1960) [3] H S Chen, D. Turnbull, Formation, stability and structure of palladium silicon based alloy glasses. Acta Metall 1021 10 31 17 (1969) [4] A Peker, W L. Johnson, A highly processable metall ic glass Zr 41.2 Ti 13.8 Cu 12.5 Ni 10.0 Be 22.5 Appl Phys Lett 2342 234 4 63 (1993) [5] J Loffler Review Bulk metallic g lasses. Intermetallics. 529 540 11 (2003) [6] S. Kalpakjian, Manufacturing Processes For Engine ering Materials, Fifth Edition, Pearson Ed ucation, Inc. p. 125, 236, 671 (2008) [ 7 ] J. Bardt, G. Bourne, T. Schmitz, J. Ziegert, W.G. Sawyer, Micromolding three dimensional amorphous me tal structures. J. Mater. Res. Vol. 22, No. 2 ( 2007 ) [8 ] A. Wiest, J. Harmon, M. Demetriou, R. Conner, W. Joh nson, Injection molding metall ic glass. Scripta Materialia 160 163 60 (2009) [ 9 ] M. Macht, T. Zumkley, S. Suzuki, S. Mechler, Near net shape microcomponents obtained by superplastic forging of bulk metallic glass. Praktische Metallographie. 215 223 No 5, Vol. 43 ( 2006 ) [ 10 ] A. Wiest, G. Duan, M. Demetriou, L. Wiest, A. Peck, G. Kaltenboeck, B. Wiest, W. Johnson, Zr Ti based Be bearing glasses optimized for high thermal stability and thermoplastic formability. Acta Materia lia 2625 2630 56 (2008) [ 1 1 ] P. Zhang, H. Wei, X. Wei, Z. Long, X. Su, Evaluation of glass forming ablility of bulk metallic glasses based on characteristic temperature. Journ al of Non Crystalline Solids 2183 2189 355 (2009) [ 12 ] F. Spaepen, Homogeneous flow of metallic glasses: A free volume pers pective. Scripta Materialia 363 367 54 (2006) [ 13 ] J. Scully, A. Gebert, J. Payer, Corrosion and related mechanical properties of bulk me tallic glasses. J. Mater. Res. Vol. 22, No. 2 ( 2007 ) [ 14 ] Z. Lu, C. Liu, Role of minor alloying additions in formation of bulk metallic glasses: A Review. Journal of Materials Science 3965 3974 39 (2004)

PAGE 42

42 [ 15 ] R. Bhowmick, B. Majumdar, D. Misra, U. Ramamurty, K. Chattopadhyay, Synthesis of bulk metallic glass composites using high oxygen con tainin g Zr sponge. J Mater Sci. 9359 9365 42 (2007) [ 16 ] H. Hng, Y. Li, S. Ng, C. Ong, Critical cooling rates for glass formation in Zr Al Cu Ni alloys. Journal of Non Crystalline Solids 127 138 208 (1996) [ 17 ] G. Duan, A. Wiest, M. Lind, J. Li, W. Rhim, W Johnson, Bulk metallic glass with benchmark thermoplastic proc essability. Advanced Materials 4272 4275 19 (2007) [18 ] D. Xu, W. Johnson, Crystallization kinetics and glass forming ability of bulk metallic glasses Pd 40 Cu 30 Ni 10 P 20 and Zr 41.2 Ti 13.8 Cu 12.5 Ni 10 Be 22.5 from classical theory. Physical Review B 74 024207 (2006) [19] S. Mathaudhu, Fabrication of amorphous metal matrix composites by severe plastic deformation Office of graduate studies of Texas A&M University. p iii ( 2006 ) [ 20 ] J. Schroers Q. Pham, A. Desai, Thermoplastic forming of bulk metallic glass A technology for MEMS and microstructure fabrication. Journal of Microelectromechanica l Systems. Vol. 16, No. 2 ( 2007 ) [ 21 ] P. Sharma, N. Kaushik, H. Kimura, Y. Saotome, A. Inoue, Nano fa brication with metallic glass an exotic material for nano electromechanical systems. Nanotechnology, Vol. 18, 035302 (2007)

PAGE 43

43 BIOGRAPHICAL SKETCH Daniel McIntyre was born i n 1987 in Boynton Beach, Florida to Kenneth and Barbara McIntyre. He lived his w hole life in south Florida until his graduation from Atlantic High School when he then moved to Gainesville, FL to attend college at the University of Florida. Daniel graduated the University of Florida with a Bachelor of Science in materials science and engineering in May of 2005. He then joined Dr. Gerald Bourne for graduate school at the University of Florida where he is scheduled to complete his Master of Science degree in December of 2010.