Tiziana Drago FEM Research Institute for Precious Metals & Metals Chemistry Schwaebisch Gmuend, Germany

Transcription

1 Tiziana Drago FEM Research Institute for Precious Metals & Metals Chemistry Schwaebisch Gmuend, Germany B. Sc. Tiziana Drago has a Bachelor in Materials Engineering with the specialization in surfaces treatments (Politecnico di Milano, Italy) and a Master in Materials and diagnostic techniques in heritage manufactures (University of Pisa, Italy). She works in the Department of Physical Metallurgy at the Research Institute for Precious Metals & Metals Chemistry (fem) in Schwaebisch Gmuend, Germany. Ulrich E. Klotz FEM Research Institute for Precious Metals & Metals Chemistry Schwaebisch Gmuend, Germany Dr. Ulrich E. Klotz is Diploma Engineer in Physical Metallurgy (University of Stuttgart, Germany) and has a PhD in Materials Science from ETH Zurich, Switzerland. He is Head of the Department of Physical Metallurgy at the Research Institute for Precious Metals & Metals Chemistry (fem) in Schwaebisch Gmuend, Germany. Platinum is a challenging material for casters because of its physical properties, which result in possible crucible and flask reactions during melting and casting, high shrinkage porosity and difficulties in filling of filigree items. This paper describes a collaborative research effort of several industrial partners and fem on the influence of casting process parameters, initiated and financed by PGI. Based on an analysis of casting defects from industrial practice experiments at fem were performed in order to identify optimum casting parameters. Two common Pt alloys (Pt-5Co and Pt-5Ru) and four different investment materials were used for casting experiments with variation of atmosphere, casting and flask temperature, tree design and centrifugal machine parameters. Detailed sample investigation found shrinkage porosity and surface defects as main problems. Optimized process parameters for heavy and filigree items were identified. Future research on platinum casting should focus on casting simulation in order to reduce experimental efforts and costs.

2 The role of process parameters in Platinum casting Introduction In recent years several articles on casting properties of platinum have been published [1, 2]. Different aspects such as suitable alloys for casting [3-6], tree design [7-10] and investment reactions [11, 12] have been treated. Articles from South African authors are describing the effect of centrifugal casting parameters for different alloys and investments [12-14]. Pt-5Co was identified as very versatile casting alloy showing excellent form filling of filigree parts even for flask temperatures as low as 100 C. Pt-5Ru on the other hand showed poor form filling of filigree parts for flask temperatures below 800 C [12]. Besides casting properties functional alloy properties such as colour, hardness, ductility and magnetic properties have to be taken into account for jewellery purposes. In this regard Pt-5Ru is more versatile compared to Pt-5Co or Pt-5Cu and can be used for all jewellery purposes. In the present project the focus has been on Pt-5Ru and Pt- 5Co as the most common alloys for jewellery purposes. The results on platinum investment casting described in this paper are result of a research project commissioned by the Platinum Guild International, USA (PGI) in cooperation with several industrial partners. This paper summarises the results of a presentation given at the Santa Fe Symposium 2010 [15] and was accepted for publication in Platinum Metal Review [19]. Experimental Casting experiments were made using a TopCast TCE10 casting machine with induction heating and a power of 10kW. Metal temperature during heating and melting was registered by a quotient pyrometer. Most trees used a Diabolo type setup. A typical tree containing some standard items (ball rings and grid) together with jewellery pieces are shown in Figure 1. Figure 1: Example of typical tree design with Diabolo setup. Four different investments from different suppliers were tested during the project. Table 1 gives an overview on the properties of the investment and briefly describes the experience in working with them. Table 1: Investment properties Manufacturer No. 1 Type two-part Base and liner paper base and liner Mixing time [min] Working time [min] Burnout time [h] Burnout temp. [ C] Remarks curing does not start at room temperature, extended working time possible No. 2 2-part paper base and liner curing does not start at room temperature, extended working time possible No. 3 1-part / fibre rubber base cold water required to reach upper working time limit; high viscosity No. 4 1-part / fibre rubber base cold water required to reach upper working time limit; high viscosity, risk of bending of filigree parts, especially plastic parts 2

3 A number of casting experiments was carried out to analyze the influence of casting parameters (melt temperature, flask temperature, casting atmosphere, casting machine type), alloy and investment material by using standard sample geometries such as ball rings and grids. The grid represents filigree items and will proof form filling ability under certain process parameters. The ball ring represents heavy section pieces. With the large ball acting as hotspot it will provoke investment reactions and it is prone to shrinkage porosity. The as-cast samples were evaluated in terms of surface quality, shrinkage porosity and investment reactions by optical microscopy, metallography and SEM. Results and Discussion Alloy properties Alloy properties differ in dendrite morphology, segregation, and melting temperature, which is about 150 C higher for Pt-5Ru [16]. The phase diagrams give the same melting range of 18 C for both alloys, but due to segregation the melting range of Pt-5Co is about twice that of Pt-5Ru, which is probably the reason for the better form filling ability of Pt-5Co. Co segregates to the melt promoting oxidation of Co and investment reactions, even for vacuum casting. Form filling Form filling is a critical issue for filigree items and was determined on standard grids (Figure 2). Figure 2: Grid filling as function of flask temperature for air casting (except value for 550 C). Averaged values of all grids per tree. Pt-5Co clearly has superior form filling ability over Pt-5Ru and shows excellent results for a flask temperature of 850 C. For Pt-5Ru form filling considerably increases with increasing centrifugal speed and flask temperature, which should be 950 C for filigree items. Vacuum casting and in case of Pt-5Ru also overheating of the melt allowed complete filling of filigree items with three part investments, however, promoting investment reactions at the same time. Surface of cast parts The surface of cast parts is controlled by investment reactions, inclusions of investment particles and rough or matt appearing surfaces. Table 2 provides a comparison of both alloys and the different investment materials. Table 2: Relative comparison of investment performance in terms of surface quality for air casting: +++ (best) ++ (medium) + (worst). No. 3 was only used for casting trials with Indutherm MC15 tilting machine. Investment Devesting Surface quality of T flask Surface quality of T flask 850 C 950 C 1050 C 850 C 950 C 1050 C No n.a n.a. No (+) ++(+) + No n.a. + (cracks) + (cracks) n.a. No n.a. + + n.a. 3

4 Further it shows the influence of flask temperature. Best surface quality was observed for Pt-5Ru, because this alloy did not show any investment reactions despite its considerably high casting temperature. Investment reactions were observed for Pt-5Co independent of casting atmosphere and resulted in a blue layer of Co silicate (Figure 3). a) b) c) Figure 3: Investment reactions of Pt-5Co and EDX analysis of reaction products on position 2 (Co oxide) and 3 (Co-Mg silicate). 4

5 Investment inclusions (Figure 4) on the surface were observed especially in case of the one-part investment No. 4. With increasing flask temperature the surface became rougher and the inclusions more frequent. For the two-part investments No. 1 and 2 inclusions were seldom observed, probably due to the lower porosity and higher stability of these investments. Therefore, the surface quality of these investments was usually superior over investment No. 4. Figure 4: Inclusions of investment material on the surface of apt-5ru alloy. The cast parts often showed matt and glossy areas. The matt areas show a dendritic like surface and were observed on the heavy sections (Figure 5) of a part. The matt surface is caused by shrinkage of isolated melt volumes during solidification. Figure 5: Matt surface of a Pt-5Ru casting with dendritic surface. Shrinkage porosity Shrinkage porosity was the main issue for bulk, heavy pattern as well for filigree, light weight pattern. It was therefore the most significant problem in the casting trails performed. The effect of casting parameters and the position on tree was found to be relatively low. Table 3 and Table 4 provide a relative comparison of shrinkage porosity. Table 3: Relative comparison of porosity in metallographic sections of Pt-5Ru casting trials depending on investment: +++ (low = best) ++ (high = medium) + (very high = worst). Pt-5Ru No. 1 No. 2 No C / air n.a C / air C / vacuum + n.a. n.a. 950 C / vacuum ++ n.a. n.a. 5

6 Table 4: Relative comparison of porosity in metallographic sections of Pt-5Co casting trials depending on investment: +++ (low = best) ++ (high = medium) + (very high = worst). Pt-5Co No. 1 No. 2 No C / air C / air C / vacuum + n.a. n.a. 950 C / vacuum + n.a. n.a. In general shrinkage porosity is less pronounced for Pt-5Ru. Both alloys showed significant difference in shrinkage pore morphology. Pt- 5Co shows few but large pores while Pt-5Ru often shows scattered pores built by intersecting dendrites (Figure 6). a) b) Figure 6: Shrinkage porosity in the sphere of a ball ring. a) Pt-5Co, b) Pt-5Ru. Flask temperature and casting atmosphere showed little influence. Investment material was found to influence shrinkage porosity and lowest levels of shrinkage porosity were achieved for investment No. 4. This is probably an effect of the lower thermal conductivity of this investment, resulting in slower and more homogeneous cooling of the melt. 6

7 Effect of casting machine A comparison of two types of casting machines, a centrifugal and a tilting machine, showed that form filling of filigree items was superior with centrifugal casting. Both machines provided comparable results for heavy items cast in Pt-5Co alloy. Defect free castings of the ball ring were obtained at 950 C/vacuum, which was not possible by centrifugal casting under comparable conditions. Pt-5Ru was difficult to cast in the tilting casting machine, because of the low heating rate in the specific model used, which resulted in hot tearing of the parts. Machines with higher power and sufficiently short melting time may enable successful casting of Pt-5Ru also. Conclusions In order to solve the main problem of shrinkage porosity, sprue design and tree setup are most important. Directional solidification has to be assured. Typical measures such as increase of flask temperature are limited by the thermal stability of the investment materials. However, optimization of casting behaviour solely by experimental means remains challenging. During the last years casting simulation proofed out being a valuable tool for gold and silver casting [17, 18]. Sophisticated software packages are available on the market to determine form filling and shrinkage porosity depending on alloy, tree setup, melt and flask temperature, allowing optimization by computer simulation. In case of platinum this would pay off even more, because of high material price and extreme casting and flask temperatures. Detailed knowledge of the thermophysical properties of alloys and investment are required, and, as such data are scarce they have to be determined in suitable experiments. Benchmark experiments with sophisticated thermal recording during the centrifugal casting process have to be performed to calibrate the casting simulation results. Investment materials were found playing an important role on form filling and shrinkage porosity. It is assumed that properties such gas permeability and thermal conductivity are responsible for that behavior. Therefore, the influence of water:powder ratio, burnout cycle, flask temperature and casting atmosphere requires further investigation to understand how the physical properties of the investments can be tailored. Acknowledgements The authors are grateful for financial support by (PGI) with special thanks to Jurgen Maerz. C. Hafner GmbH, Germany enabled the project by providing platinum alloys, which is kindly acknowledged. The companies Ransom&Randolph, Lane Industries and Specialist Refractory Services (SRS) are acknowledged for allocation of investment materials. Special thanks to Dieter Ott for fruitful discussions of the project results and to the staff members of the metallurgy department at FEM, especially to Franz Held and Ulrike Schindler for casting trials and metallography. 7

8 References [1] Swan, N., Improvements in platinum casting. Jewellery in Britain 19(Dec) (2004) [2] Swan, N., Casting Platinum Jewellery A Challenging Process. Platinum Metals Review 51(2) (2007) 102. [3] Ainsley, G., A.A. Bourne, and R.W.E. Rushforth, Platinum investment casting alloys. Platinum Metals Review 22(3) (1978) [4] Huckle, J. The development of platinum alloys to overcome production problems. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (1996) pp [5] Maerz, J. Platinum alloy applications for jewelry. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (1999) pp [6] Maerz, J. Platinum alloys: features and benefits. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2005) pp [7] Maerz, J. Casting tree design and investing technique for platinum induction casting. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2002) pp [8] Nooten-Boom II, A. Controlled flow = controlled solidification. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2004) pp [9] Nooten-Boom II, A. Dynamics of the restricted feed tree. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2006) pp [10] Maerz, J. Platinum casting tree design. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2007) pp [11] Frye, H., M. Yasrebit, and D.H. Sturgis. Basic ceramic considerations for the lost wax processing of high melting alloys. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2000) pp [12] Lester, P., S. Taylor, and R. Süss. The effect of different investment powders and flask temperatures on the casting of Pt alloys. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2002) pp [13] Miller, D., T. Keraan, P. Park-Ross, V. Husemeyer, and C. Lang, Casting Platinum Jewellery Alloys. The effects of composition on microstructure. Platinum Metals Review 49(3) (2005) [14] Miller, D., T. Keraan, P. Park-Ross, V. Husemeyer, and C. Lang, Casting Platinum Jewellery Alloys. Part II: The effects of casting variables on fill and porosity. Platinum Metals Review 49(3) (2005) [15] Klotz, U.E., T. Drago. The role of process parameters in platinum casting. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2010) pp [16] McCloskey, J.C. Microsegregation in Pt-Co and Pt-Ru jewelry alloys. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA. Met-Chem Research (2006) pp [17] Fischer-Bühner, J. Computer simulation of jewelry investment casting. in: The Santa Fe Symposium on Jewelry Manufacturing Technology. Albuquerque, NM, USA (2006) pp [18] Actis-Grande, M., L. Porta, and D. Tiberto, Computer simulation of the investment casting proess: widening of the filling step, in The Santa Fe Symposium on Jewelry Manufacturing Technology, E. Bell, Editor. 2007: Albuquerque, NM, USA. p [19] Klotz, U. and T. Drago, The role of process parameters in Platinum casting. Platinum Metals Review (2011). 8