TiNi shape memory alloy obtained by ultrafine laminates

 

·      Recentdevelopments in the research of shape memory foils obtained by ultrafinelaminates.

·      Some experimental results of theshape memory foils (SEM, TEM, DSC).

·      3D “flower” like shape actuator in action.

·       Conclusions

 

 

 

Recent developments in the research of shapememory foils obtained by ultrafine laminates.

 

Shape memory alloys (SMAs) have beensuccessfully introduced in a variety of technological areas over the pastyears. One of the promising field for applications ofSMAs is the actuator technology, since with a shape memory element, apre-determined response can be obtained very easily by a thermal stimulus.

 

TiNi alloy exhibits superior shape memoryand superelastic properties and has been used in diverse areas of applications.A significant disadvantage of this alloy is the low plastic workability due toa high work hardening rate. Production of TiNi alloy thin sheets or foils byrolling requires complex annealing procedures, and this leads to a high cost ofthe materials.

 

Other possibility to produce TiNi thin sheets is to use multilayerednanostructures composed of pure Ti and Ni-layers. Production of suchmultilayered structures can be accomplished by cold rolling a stack ofelemental sheets of metals.  A subsequentheat treatment of the multilayers leads to the formation of intermetalliccompounds or amorphous phase by the solid state interdiffusion. Thus thismethod is advantageous for theproduction of thin plates or foils of intermetallics. Production of Ti/Nimultilayers by rollinghas been done starting from the foils of pure metals.

 

Sasaki et al. proposed a methodto produce materials with layered structures using a conventional rolling mill,referred to as ultrafine laminates method. In our research the Ti/Ni ultrafinelaminates were used to produce the TiNi shape memory alloy foils with thicknessdown to 50 μm.

 

The ultrafine laminates method isschematically illustrated in Fig. 1.

 

Fig. 1. Ultrafine laminates method.

 

In this method a thin sheets of pure Ti(0.2 mm thick) and pure Ni (0.12 mm thick) are stackedalternatively for 179 layers. The thickness and number of the pure metal sheetscan be adjusted to obtain the desired composition. The Ti/Ni stack was then putinto a steel container and sealed under vacuum. The sealed stack was hot andcold rolled to the final thickness of 50 μm. The size of the as-rolled foils is about 60 mm width and several meterslong. The designed composition of the foils was Ti-50.7at%Ni.

 

 

Some experimental results of the shape memory foils(SEM, DSC, TEM).

 

Fig. 2 shows a SEM micrograph of alongitudinal section of the Ti/Ni ultrafine laminates. Intermetallic compounds,possibly Ti2Ni, TiNi and TiNi3 were occasionally observedat the interface between the Ti and Ni layers.

   Ahomogeneous TiNi phase was obtained after the diffusion treatment at 1073 k for36 ks. In Fig. 3 is shown the SEM micrograph of sample after this diffusiontreatment. The Kirkendall voids are occasionally seen.

 

Fig. 2 SEM micrograph of longitudinal section of the Ti/Ni ultrafinelaminate.

 

Fig.3 SEMmicrograph of Ni-rich TiNi foils diffusion treated at 1073 K for 36 ks.

 

  The DSC curves of the TiNi foils with and without aging for differenttimes are presented in Fig. 4. There are three characteristic features thatchange with aging: (1) the type of transformation changes from three steps(after 3.6, 18 and 36 ks) to two steps (after 72 and 144 ks); (2) there is anadditional transformation, marked in Fig. 4 with triangles, at about 345 Kwhich is MS temperature inTi-rich alloys; (3) there are shifts in the transformation temperature.

 

Fig.4 DSCcurves of Ni-rich TiNi foils aged for 773 K for different times.

 

The reasons of this evolution could beexplained as follows: (1) the composition inhomogeneity that evolves duringaging as Ti3Ni4 precipitates grow.(2) The difference between nucleation barriers for R-phase(small) and B19’ (large).

 

TEM observation of specimen without agingrevealed that the B2 grain size is about 1μm. In Fig. 5(a) it can be seenthat the lenticular Ti3Ni4 precipitates are uniformlydistributed and aligned along two directions. The average size of theseprecipitates is less than 100 nm. Fig. 5 (b) is a typical diffraction patternof Ti3Ni4+TiNi (B2). There are 1/7 superlatticediffraction spots observed along <321>B2 and 1/3 superlatticediffraction spots <110>B2 .

 

 

           

       Fig. 5(a) Bright field TEM image of foil aged at 773 K for 3.6 ks         Fig. 5 (b) Typicaldiffraction pattern of Ti3Ni4+TiNi (B2)

 

 

3D “flower” like shape actuator in action.

 

 

In order to prove the possibility offabrication of three dimensional (3D) actuator, asample was constrained in the shape of flower and aged. The results of thisexperiment are shown in Fig. 6. (Click on picture to download a 1.1 Mb mpegfile).

 

 

 

 

 

 

 


Conclusions

 

 

·       Using ultrafine laminates method 50 μm foils ofTi-50.7at%Ni were produced. After diffusion treatment at 1073 K for 36 ks TiNiB2 phase was obtained. The diffusion treatment was followed by aging treatmentat 773 K in order to produce Ti3Ni4 precipitates.

 

·       In this study it was revealed that these TiNi foilshave a shape memory response comparable to that in TiNi bulk materials. Two-wayshape memory strain of 4x10-3 was realized. This type of processingmethod can produce a large quantity of TiNi alloy foils at muchlower cost than that for the conventional method of rolling bulk ingots.

 

·       Designing an actuator with large displacement and 3Dresponse is now possible using these TiNi foils.