This experiment investigated the principal factors involved in an identified offset in nitinol (NiTi alloy) peak transition temperatures logged when differential scanning calorimetry was used to measure the location of the transition temperatures.  The principal contributor for the observed differences in temperatures is the quench method used, and also appears to be more pronounced for the ternary alloy tested. However, the effect of the shift drops as the base transition temperatures rise, yielding essentially no difference once A_peak temperatures reached around 70 °C.  The experiment also investigated some alternative methods for assessing the completeness of anneal.  The best method is not clear, however, due to the potential for precipitation, migration, and other known changes in NiTi at elevated temperatures.  Follow-up testing to evaluate the possible root causes has shown mixed results, in that the aging of the sample for brief periods (4 – 6 seconds) can cause a downward shift in the observed temperature, and that neither set of quench conditions apparently yields a ground state anneal.  Microstructure issues still remain to be addressed.

 Aging:

The experiment was performed to evaluate a possible factor causing the downward shift observed in transition temperatures due to the use of air cooling as opposed to water quench for the post-anneal reduction in temperature.  In the air quench, the time spent in the aging region is on the order of seconds while cooling.  Evaluation of a series of samples using binary alloys demonstrated a clear correlation between the short-term aging and lowering of the transition temperatures.  The ternary alloy also showed a shift, but within-group variation prevented a finding of statistical significance.

 Purge Gas Selection:

Selection of a particular purge gas created a modification in the response of the Differential Scanning Calorimeter system.  A single indium standard was used to track responses on five parameters, with particular attention to the determination of the cell constant.  It is clear from the data that the helium causes an apparent loss of signal to the DSC system, due to its higher thermal conductivity relative to nitrogen.  This diversion of energy to purge results in much higher cell constants (e.g. the gain) needed to align the theoretical standard response to actual observations.  Therefore, it is recommended that the sole use of helium purge gas implemented in ASTM F2004-03 be dropped in favor of reporting what purge gas is used, due to the amplified noise.