Stabilization of the shape memory effect in NiTi: an experimental investigation

Abstract

Over the last 20 years or so, shape memory alloys have become well known for their ability to undergo large shape changes that traditional metals could never do. One such alloy, NiTi, has been of particular interest due to it’s high efficiency (when compared to other SMA’s) to convert thermal energy into mechanical energy. While NiTi is currently used in industry, its use is hindered by an extreme sensitivity to processing techniques. Certain parameters such as alloy composition, annealing procedures and various loading conditions can improve the shape memory effect or completely make it disappear. Many studies have focused on these variables. Of particular interest to this study is the one-way shape memory effect (OWSME) after the NiTi has been cycled (deformation followed by recovery) many times. It is very desirable to find a processing technique that would stabilize the OWSME and make it repeatable after many cycles. This paper explores these parameters in hopes of finding an optimal process to stabilize the shape memory effect in NiTi. Many studies have been conducted examining the cyclic behavior of NiTi. Most common are studies done on pseudoelastic NiTi (1,2,3) and on the two way shape memory effect (TWSME) (4,5,6,7). However, very few have explored these cyclic effects on the OWSME. One study that was conducted on the OWSME used wire samples and carried tests out to 100 cycles (8). It was reported that the recovery strain was reduced to 93.4% and 81% of the total strain when total strains of 6% and 8% were used, respectively over the 100 cycles. (Note, that these values will differ from ours since they used “hard” cycles with a fixed maximum strain where as we used “soft” cycles with a fixed maximum stress.) This displays the degradation of the shape memory effect that hinders NiTi’s use. After this work, another study was performed that investigated a conditioning technique to try to stabilize the shape memory effect (9). Their research recommended that after SME (strain-heat-recover-cool-repeat) cycling the NiTi 30 times at 6% strain, a stable strain range was obtained for strains less than 6%. This was only shown over a very few cycles and in wire samples. It is our hypothesis that some form of “conditioning” may be performed on NiTi after annealing to improve its cyclic shape memory properties. The result will be a protocol that can be applied to NiTi before its service use. In doing so, it is important to define a few performance criteria that are considered indicative of “good behavior” and clarify what will hopefully be stabilized: