Thermoplastic Forming of Bulk Metallic Glass PDF

Preview

Summary

Thermoplastic Forming of Bulk Metallic Glass andStainless SteelKaran GoldenwallaZienkiewicz Research Centre College of Engineering Swansea University Swansea UK [email protected]— Production of new metallic glasses in bulk form is increasing in demand due to their superior strength, elastici...

Description

Thermoplastic Forming of Bulk Metallic Glass and Stainless Steel Karan Goldenwalla Zienkiewicz Research Centre College of Engineering Swansea University Swansea UK [email protected]

Abstract— Production of new metallic glasses in bulk form is

increasing in demand due to their superior strength, elasticity and magnetic properties. As these metallic glasses can now be produced in bulk conditions, they can be used for many structural applications. This paper describes the experimental investigations undertaken to understand the force and stresses occurring at the overlap between a sample of bulk metallic glass and stainless steel in an experiment procedure where samples were placed in a furnace where thermoplastic forming would take place. This was done under two conditions where one of them included a weight being placed on the overlap between the plates, and these were further examined with a clamp securing the overlap tightened with a torque wrench. The effects of torque tightening on the stress in a single lap joint have been investigated numerically. To determine the clamping force resulting from the different torques applied with the torque wrench, an experiment method was designed and theoretical calculations were performed. Further FEM was also analyzed with both specimens and the obtained results revealed that the resultant stressed reduced as the tightening torque was increased due to the compressive stresses. Keywords—Bulk metallic glass, Stainless steel, Torque, Force, Stress, Thermoplastic Forming

I.

exception glass forming ability in rare-earth rich alloys [3]. These new amorphous alloys exhibit remarkable properties being structural materials. Roughly they have a yield strength of around 2 GPa, and a fracture roughness between 20 to 140 MPa m1/2, great corrosion resistance and a high specific strength. [4]. During the past several years, there have been many advances in this field as a result of the discovery and development of several families of alloys with improved glass forming ability greatly. By the 1990’s, new alloys were developed that form glasses at cooling rates as low as one Kelvin per second. Previously metallic glasses were usually formed at cooling rates of 105-106 Kelvin per second which shows a huge advance in the field. These bulk amorphous alloys can be cast from the molten state into glassy objects with dimensions of up to several centimetres as compared with maximum thickness or diameter. The ‘best’ forming glass forming alloys are based on Zirconium and Palladium, but alloys based on Magnesium, Copper, Iron and others are also known. One broadly studied example of this glass family is Vitreloy-1 which has a critical cooling rate of 1K/s [5]. With these superior properties that BMG’s have, our experiment would prove very useful in terms of applications and the future of BMG’s and SS.

INTRODUCTION

1.1 History and Background Metallic glasses were first discovered over forty years ago when rapid quenching methods were applied to the Au-Si system [1]. The first reported metallic glass was an alloy Au75Si25 produced at Caltech by W. Klement (Jr.), Willens and Duwez in 1960 [2], but these glass specimen dimensions were limited to micro meters due to their very quick cooling rates of around 106 K/s needed in order to avoid crystallization. In recent times, multicomponent alloys have been developed with an exceptional glass forming ability which allow the processing of bulk specimens called bulk metallic glasses (BMG). These were first found during the late 1980’s, when Akihisa Inoue and his co-workers in Sendai, Japan, found

Figure 1: Ashby map of fracture toughness against density (Source, Douglas C. Hofmann, Bulk Metallic Glasses and Their Composites, 2013)

1.2 General applications of BMG’s Different types of BMG alloys are used in different ways. It has been previously reported that Zr-Based BMG’s were commercialized as sports goods, such as golf clubs, electromagnetic device casing, etc. In recent times, there has been a significant progress in applications for BMG’s in Zr-, Ti- and Fe-based alloy systems. Furthermore, SS and BMG composites have shown impressive properties which meant that there are many benefits from bonding the two materials together. Researchers in Germany [6] have successfully toughened a BMG with commercially available spring-shaped steel wire by centrifugal casting. This new composite showed more plasticity than when compared with the BMG without the steel wire. This creates a greater interest in composites of BMG’s. Currently, the most important application is due to the special magnetic properties of some ferromagnetic metallic glasses. Moreover, composite and BMG’s that are lightweight and inexpensive such as Al- based alloys are being developed for structural applications by multi-institution US Department of Defence programs. This included a team at the Centre for Science and Engineering of Materials Led by Caltech’s Johnson investigating the processing, microstructure and mechanical behaviour of Zr56.3Ti13.8Cu 6.9Ni5.6Nb5.0Be12.5 and other two-phase alloys. Johnson has pioneered the in situ transmission electron microscopy of shear band deformation [7]. By modelling the observed patterns using FEM, he has established the temperature dependence of the transition from homogeneous flow at low strain rates near or above Tg to inhomogeneous, shear band mediated plastic deformation at lower temperatures and high strain rates. BMG’s also have many medical applications; ‘Liquaalloy’ has a highly biocompatible, no allergenic form, which is ideal for corrosion and wear- resistant medical applications. For example, DePuy Orthopaedics, Inc. is using BMG as kneereplacement devices and other applications such as pacemaker casings. In 2002, Surgical Specialties began producing ophthalmic scalpel blades using this alloy. These were of higher quality and less expensive than diamond, sharper and longer lasting than steel. It was also more consistently manufacturable since they are produced from a single mold.

1.3 Thermoplastic Forming Thermoplastic forming (TPF) is the process in which a material is heated to a workable temperature which is then formed to a shape by a mould which is then cut to create a useful product. This process is especially well suited to replicate smaller features and thin sections with high aspect ratios, which makes this process suitable for microelectromechanical systems, micro & nano technology, optical applications and data storage etc. This is the process which is undertaken in the experiment in this project. This

process involves heating of amorphous BMG into the super cooled liquid region where it is formed under an applied pressure at a constant temperature. This processing window is done under isothermal conditions and is limited by the time taken to reach crystallization. An optimal alloy for this process should have a good glass forming ability meaning a low viscosity in the super cooled liquid region, a low processing temperature and a long processing time at the temperature before crystallization.

Figure 2: Schematic illustration of the TPF of BMG. (a) homogeneous bulk composites, (b) surface composites synthesized by this method [8]. (Source, 2007, Elsevier).

The formability of a BMG in the supercooled liquid region can be quantified using the Hagen–Poiseuille-equation: L

𝑝 = 16𝑣 ηd2

(1)

with p being the pressure required to move a liquid with a viscosity η, at a velocity, v, through a thickness of d, and length, L. The equation above assumes there is no velocity of liquid at the interface. The max time available for the forming is given by the crystallization time, tcryst. Substituting v = L/tcryst in Equation 1 gives the maximum length that can be filled. 𝐿=

𝑝𝑡𝑐𝑟𝑦𝑠𝑡 𝑑2 16 η

(2)

This filling length can be used to quantify the formability and determine the conditions which are most suitable for this process. It is dependent on the viscosity and filling time. It was found that in order to optimize the max formability the processing temperature should be chosen as high as possible, as long as crystallization is not occurring [9].

Furthermore, the flow law can be used to determine minimum feature size which can be replicated. When small sizes are being considered, there is a pressure contribution which has to be included in the calculation, as it results from the surface energy. This pressure increase is given by p = 4γ/d, where γ is the energy of the liquid to vacuum interface. This contributes to equation 1 and then gives 𝑝 = 16η𝑣

L 4𝑦 + 𝑑2 𝑑

(3)

Equation 3 can be used to calculate the minimum feature size BMG is the ideal material for smaller geometries as it is homogenous and isotropic. This is due to the no intrinsic limitations such as the grain size in crystalline materials is present. Stresses or porosity which results from solidification shrinkage are very smaller using this method compared with other ones. This solidification shrinkage in crystalline materials originates from thermal contraction and phase transition shrinkage. The absence for crystallization during the solidification of BMGs as well as the lower processing temperature reduces shrinkage. For example Zr44Ti11Cu10Ni10Be25, approximately had 0.4% solidification shrinkage during casting, whereas the same alloy during TPF has shrinkage of only 0.2% [10]. The crystallization of a BMG has to be avoided for many reasons; one being it degrades the properties of the superior glass. TPF is a very robust process. This is due from the fact in the crystallization in the super cooled liquid region is a welldetermined and controlled process [11,12]. This process eliminates the risk of unpredictable crystallization due to the nature of the crystallization process where TPF is performed whereas, in other processes such as die casting, crystallization can occur due to the nature of the method. Furthermore, for different BMG’s, the oxygen content can also affect the glass forming ability [13,14,15,16]. For the process of TPF, the oxygen content only has a minor influence on the crystallisation. On the other hand in die casting, the process has to be performed in a vacuum as the presence of oxygen reduces the processing time. This shows that TPF can be performed for most BMG’s in the air.

Figure 3: Schematic of a time temperature transformation (TTT) diagram indicating that metallic glasses can be processed via thermoplastic forming (2) [17]. (Source, Schroers Lab) 1.4 Glass Forming Ability Glass forming ability (GFA) is related to ease of devitrification and this is vital in understanding the basics of GFA. An early methodology for alloy design was created by Inoue [18] who states that in order to achieve a high GFA, the three following empirical rules must be satisfied; glass forming alloys must firstly have a negative mixing enthalpy amongst its constituent elements, this is because this increases the energy barrier at the liquid solid interface and decreases atomic diffusivity which increases the equilibrium melt viscosity to three orders of magnitude greater than binary alloys. This retards the local atomic rearrangements and the crystal nucleation rate, extending the supercooled liquid temperature range. Next, there must be an atomic radius mismatch between elements, greater than 12% leads to a greater packing density and smaller free volume in the liquid state compared with metallic melts, and requires a greater volume increase for crystallization. Lastly, the multicomponent alloy must be of three or more elements which increases complexity and size of the crystal unit cell and reduces the energetic advantage of forming an ordered structure of longer range periodicity than the atomic interactions. The rules for GFA were expanded by Egami [19] who added that an attractive force between large and smaller elements coupled with repulsion between the smaller elements increase GFA. There are many factors which affect the GFA of a metallic glass. Firstly, the alloy composition is an important element. It is known that deep eutectics are thermodynamically favourable for formation of glass. The free energy change between the glass state and crystalline phase is minimal therefore meaning there is only a small driving force for crystallisation. Next, large electronegativity differences for an alloy system are related to strongly negative values for enthalpies of mixing. This lowers eutectics temperatures meaning an increase in Trg which suggests a larger GFA. Glass formation is favourable when the atomic radii difference between elements in a given alloy is more than 12% [20]. The large difference in atomic radius deters the crystallisation kinetics by reducing the diffusion of elements in the liquid alloys. Lastly, it is found that GFA can be extended by the addition of other elements forming a multicomponent amorphous alloy. The choice of these additional components must satisfy the above factors. This was described as the ‘Confusion Principle’ as proposed by Greer [21].

1.5 Description of experiment

1.6 Bonding using clamp force

In this research project, forces are applied to the overlap of two plates of the two materials (BMG and SS) which are placed in the furnace, and the effects of this were then examined in terms of stress and the bond formed between the BMG and SS. The temperature at which the experiment is performed will be a factor which will further affect the stresses involved. However, this is very complex to take into account of in this study and was not gone into further detail on. In the next part of this project, a similar experiment was undertaken, but instead of a force being applied at the overlap between the two materials, a clamp was used to tighten the overlap. This clamp had a bolt where the handle was which was where an applied torque was used with a torque wrench to tighten the two pieces together. In addition, the bonding in which the BMG and SS are joined has also been examined. In the supercooled liquid state of a BMG, the atomic configuration easily rearranges to accommodate plastic flow, along with the fluidity of a supercooled BMG forming liquid to make micron/nano-sized pattern replication possible. It is reasonable to expect that a supercooled BMG forming liquid can wet and bond to the steel material during this rearrangement. This bonding is also dependent on the pressure applied on the joint. There is a pressure being applied to the overlap as there is a force acting on it. This depends on the area that the force is actually acting on, and this pressure will spread out and wet on the SS.

One of the most important factors that will affect the stresses occurring at the overlap of the second specimen is the amount of clamping force which results from the tightening torque applied to the bolt on the handle on the vice. This clamping force is achieved by using the torque wrench, which applies a torque to the head of the bolt. The applied torque correlates with induced clamping force [32,33,34,35,36]. This clamping force Fcl can be calculated using the following equation.

This differs to usual studies, as one of these materials happens to be bulk metallic glass, and the other is stainless steel. These materials are dissimilar and this area should be further investigated in detail as the properties and features of the stainless steel would change in this investigation. Details of the features cannot be investigated very deeply in this project as the experiment wasn’t performed. Homogenous glassy BMG’s usually are not thick enough for structural applications and extensive research efforts have been made to develop welding technologies. There are studies in which BMG’s have been welded to SS, which can be a very difficult procedure as it fails due to the tendency to crystallize into the weld material but methods such as electron beam welding, laser beam welding etc. have been successful. Weldability is an important index that makes BMGs join with different materials and extend their applications in engineering [22 ]. To date, several welding methods, including friction welding [23], explosion welding [24], electron beam welding [25], ultrasonic welding [26], and laser welding [27] have been used to connect the BMGs materials. For example Kawamura et al. [28,29,30] have reported that the Zr41Be23Ti14Cu12Ni10 BMG can be successfully welded to crystalline Zr, Ti and Ni metal without welding defects or crystallization in the weld. This has proved to increase the strength of the composite adequate for industrial applications along with higher yield strength being achieved [31]. The experiment described here is not welding but it is a similar process in which features will be transferred.

𝑇 = 𝐾𝐹𝑐𝑙 𝑑

(4)

Where T, Fcl and d are the applied tightening torque to the bolt, the clamping force and the nominal thread diameter, respectively, and K represents the torque coefficient or nut factor depending on the variety of parameters. This torque coefficient used for steel is generally 0.2 which is what was used in the calculations shown. In this study, the effect of this torque tightening on the stress occurring and clamping force present has been investigated numerically. The bond formed between the SS and BMG also has importance. This is as if a huge torque or weight was applied, it would most likely cause the experiment to fail and the BMG would melt completely and there would be no bond between the two materials. II. EXPERIMENTAL PROCEDURE The specimens employed in this investigated were constructed using a 2mm thick 302 stainless steel alloy. This is the most familiar and frequently used alloy in the stainless steel family and can be used in a wide variety of applications due to a large number of its properties such as resistance to corrosion, ease of fabrication etc. Table 1 displays the mechanical properties of this stainless steel alloy, along with the chemical compositions illustrated in Table 2. Stainless steel is used more widely than pure metal, such as Zr, Ti and Ni, in a range of applications in the automotive and marine industries. This is why the experiment could prove very useful in terms of applications of the formed product. Table 1 Mechanical Properties of 302 Stainless Steel Alloy Tensile Yield Youngs Poisson’s % strength Strength Modulus Elongation Ratio MPa MPa GPa in 2” Minimum 620

275

193

0.27-0.30

50%

Figure 5: Isometric view of specimen setup

Table 2 Chemical Composition of 302 Stainless Steel Alloy (%) Cr Ni Mn Si C S P 17-19

8-10

2

1.00

0.15

0.03

0.045

2.1 Specimen Preparation The two specimens were prepared based on the EN1465 standards guidelines. Each specimen would be a cuboid of dimensions 20mm×10mm×1.6mm of stainless steel and bulk metallic glass as illustrated in Figure 4. To eliminate any possible surface scratched of both these surfaces, beforehand the surfaces of both plates were polished mechanically by rubbing with grinding papers. The experiment will be performed and tested in a standard laboratory atmospheres specified in ISO 291. The two samples would then be placed together with an overlap of 5mm and will be held together with a suitable pair of self-aligning grips as well as a weight being placed on the overlap for the first specimen. With the second specimen, a vice with an M6 bolt placed on the handle will be used to secure the overlap tightly and a torque wrench was then used to fasten this clamp at a certain torque.

Figure 4: Side view of specimen setup

Bulk metallic glass

Steel

Figure 6: Specimen setup with weight placed on overlap

2.2 Test Procedure The tests were performed using a furnace which was set at a certain temperature. During the test, the first specimens were held together with self aligning grips and then placed into the furnace where a weight was pla...