Materials Science and Engineering A 438–440 (2006) 679–682 The effects of vacuum induction melting and electron beam melting techniques on the purity of NiTi shape memory alloys J. Otubo a,b,∗ , O.D. Rigo b , C. Moura Neto a , P.R. Mei b a Instituto Tecnologico de Aeronautica (ITA), 12228-900 S.J. Campos, SP, Brazil b DEMA-FEM-UNICAMP, 13083-000 Campinas, SP, Brazil Received 21 April 2005; received in revised form 16 January 2006; accepted 7 February 2006 Abstract The usual process to produce NiTi shape memory alloys is by vacuum induction melting (VIM) using graphite crucible that contaminates the bath with carbon. The contamination by oxygen comes from residual oxygen inside the melting chamber. A new alternative process to produce NiTi alloys is by electron beam melting (EBM) using water-cooled copper crucible that eliminates the carbon contamination and the oxygen contamination would be minimized due to operation in high vacuum. This work compares the two processes and shows that the carbon contamination is four to ten times lower for EBM compared to VIM products and that the final oxygen content is much more dependent on the starting raw materials. The purity of the final product should be very important mainly in terms of biomedical applications and the contaminations by carbon and oxygen affect the direct and reverse martensitic transformation temperatures. © 2006 Elsevier B.V. All rights reserved. Keywords: NiTi; Shape memory alloys; Vacuum induction melting; Electron beam melting 1. Introduction Although the NiTi shape memory alloys (hereafter called NiTi SMAs) are known worldwide since the 1970s [1], the lit- erature relating its processing is limited owing to production difficulties (there are few producers in the world). Our group has been working on NiTi SMA processing since 1997 using two processes: electron beam melting (EBM) and vacuum induction melting (VIM) in order to obtain ingot in pilot scale [2–4]. The conventional process to produce NiTi alloy is by VIM that presents some advantages such as ease chemical composi- tion control due to magnetic agitation, relatively ease operational control and possibility of degassing. One of the disadvantages is the contamination of the bath by carbon which comes from the use of graphite crucible and graphite mold. The use of other types of crucible such as MgO and Al 2 O 3 could contaminate the bath with oxygen [1]. An alternative to VIM process is the EBM process to produce NiTi SMA. This process is known since the 1950s for refining refractory and reactive metals. More recently this process has been used to produce alloys such as Ti6Al4V [5–8] and very ∗ Corresponding author. Tel.: +55 12 39475905; fax: +55 12 39475906. E-mail address: jotubo@ita.br (J. Otubo). clean superalloys [9]. The use of EBM process to produce NiTi SMA is quite rare [10,11] and practically started with this group [2,3]. The ingot production by EBM completely eliminates the carbon contamination due to melting in a water-cooled copper crucible, and also the oxygen contamination is minimum due to operation in high vacuum (better than 10 -2 Pa). Therefore, the carbon and oxygen contents in the final product depend only on their levels in the initial raw material. One of the disadvantages of working in high vacuum during melting and remelting is the difficulty of controlling the nominal chemical composition due to some component evaporation mainly nickel which has higher vapor pressure than Ti. Our recent work producing small NiTi SMA ingots using EBM has proved that the composition control is perfectly possible as far as certain care is taken during the feeding material preparation [3,4]. This work compares some results of producing NiTi SMA by EBM and VIM processes mainly in terms of final ingot purity and how those impurities affect the martensitic transformation temperature. 2. Experimental procedures 2.1. VIM process By this process three experimental procedures were planned to check the influence of graphite crucible quality and its dimen- 0921-5093/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2006.02.171