Laser Welding of Nitinol Shape Memory Alloys


P. Sathiya, Associate Professor,
Department of Production Engineering,
National Institute of Technology, Trichy,

Deepan Bharathi Kannan,
Assistant Professor, Department of Mechanical Engineering,
SRM Institute of Science & Technology,
Kattankulathur, Tamilnadu
In this article, an attempt is made to discuss the various aspects related to the laser welding of NiTinol shape memory alloys. Unique properties of NiTinol and its applications related to automobile, robotics, biomedical and fashion industries are discussed in detail. The properties variation of the NiTinol shape memory alloys after welding are discussed with respect to metallurgical, phase transformation temperature, mechanical and corrosion properties. The details given in this paper will help the researchers in understanding the basic concepts related to laser joining of NiTinol.
Keywords: NiTinol; Laser welding; Metallurgical characterization; Mechanical properties; Corrosion Properties; Post weld heat treatment
New materials are being continuously developed to sophisticate the human life. One such recently developed material is NiTinol. NiTinol stands for its constituents Nickel, Titanium, and Naval ordnance laboratory. NiTinol was discovered in the year 1962 by William Buehler and Frederick wang. NiTinol exists in two phases, viz. Austenite and Martensite. Austenite is a high-temperature stable phase, and martensite is a low-temperature stable phase. There are four temperatures associated with NiTinol shape memory alloys,and they are austenite start temperature (As), austenite finish temperature (Af), martensite start temperature (Ms) and martensite finish temperature (Mf). NiTinol has two unique properties, viz. shape memory effect (SME), super elasticity or pseudo elasticity. The ability of the NiTinol to return to its original shape on the application of heat is called shape memory effect. The pictorial representation of shape memory effect is given in Figure 1.

Fig 1: Pictorial representation of shape memory effect.
 From Figure 1, it can be seen that, below martensite finish temperature, NiTinol exists in the twinned martensite phase and on the application of load, NiTinol deforms by converting into detwinned martensite. On the application of heat (T >Af), NiTinol returns to its original shape by converting into austenite phase. If NiTinol is cooled below martensite finish temperature, then again it gets transformed into martensite phase. This transformation between austenite and martensite makes the feasibility of shape memory effect. The phenomenon of formation of stress-induced martensite is called super elasticity. Super elasticity is possible only at a temperature above austenite finish temperature. The temperature above which martensite can longer be stress induced is called martensite deformation temperature limit (Md). Properties such as shape memory effect, super elasticity makes NiTinol as one of the ideal materials for a vast number of industrial fields such aerospace, automobile, robotics, biomedical and fashion industries.
In automobile applications, NiTinol is preferred mainly, as it not only reduces the overall size of the vehicle but also helps in increasing the performance of the vehicles. With the automobile industries moving towards safer, comfortable vehicles with better performance, the need for some sensors and actuators became inevitable Butera et al. (2007), Stoeckel (1990) and Butera (2008). NiTinol is mainly used as sensors, actuators and they are entirely replacing conventional electromagnetic actuators in the automobile industries. The work density of the NiTinol actuator is 10 J/Cm3 which is 25 times greater than the electric motors, and they can lift weights 100 times more than its weight. Some of the examples, where NiTinol is used as actuators are lock/latch controls, climate control flaps, rear view mirror folding, etc. Stoeckel (1991). Now, attentions are given to expand NiTinol usage in aerodynamic and aesthetic applications. General motors are continuously involved in exploring NiTinol for automobile applications from the mid-1990’s,and they successfully implemented sensors made of NiTinol on their 2013 year car models.
NiTinol usage in the robotics field has tremendously increased since the mid-1980’s and mostly used as micro-actuators or artificial muscles. In robotic applications, NiTinol is mainly preferred due to its flexibility in shape and size. Actuators size and shape have a significant impact on the overall size, shape degrees of freedom of the robotic device. Chee Siong et al. (2005) in their work successfully developed prosthetic hand using NiTinol actuators. Instead of utilizing conventional push-pull type and biased spring, the authors used two NiTinol actuators to actuate the robotic finger which almost replicated the human finger actions namely, flexion and extension. Some robots developed using NiTinol had the capabilities to solve problems that are challenging for humans, by giving appropriate details form land, air, water, and space. Ali et al. (2010) in their work developed a microgripper which has the capability to work without the internal power source and can be activated wirelessly. An artificial rat whisker was developed by the researchers in the northwestern university of Illinois based on the super elasticity behavior of NiTinol which can be used to improve the robotic sensing capability and can be used to recognize micro surface features, navigate into small and tight interior locations Tuna et al. (2012). Some flying robots such as Bat robot, BATMAV were also developed with the help of NiTinol shape memory alloys Furst et al. (2012), Bunget et al. (2009) and Colorado et al. (2012). An artificial dragonfly having 13 degrees of freedom was recently developed using NiTinol shape memory alloys Festo (2013).
Even though NiTinol is expensive than the stainless steels, unique properties of NiTinol such as nonmagnetic, SME, unique physical characteristics similar to that of human bones and tissues have made NiTinol as an ideal candidate for biomedical applications. Some of the examples where NiTinol are used for biomedical applications are stents, guide wires, eyeglass frames, aneurysm treatments, medical tweezers, sutures, anchors for attaching the tendon to bone, etc. Pfeifer et al. (2013), Stirling et al. (2011), Pelton et al. (2008) and Miyazaki et al. (2006). Recently, research has been successfully done to study the feasibility of using NiTinol for making penis implants to treat penile erectile dysfunction. Penis will get erected once the implants reach a particular temperature. The problem associated with this is the erection irrespective of the time and location once the required temperature reached, and researchers are working to tackle this problem. One-third of men with erectile dysfunction between the ages of 40 and 70 don’t respond to the drugs, so this heat-activated penile implant could be a better option for the affected males and their partners.
NiTinol usage in fashion industries is picking up the pace as it is possible to change the shape of the shirts with respect to temperatures. In European countries, to tackle the high-temperature summer days, people are using shirts made of NiTinol. This shirt will reduce the size and change the shape in such a way that it enables the free flow of air into the body on reaching a high temperature (around 35 ℃) or vary depending on the geometrical locations. Women’s innerwear such as bra are also being manufactured using NiTinol SMA Wires as they have better durability.
Any materials success not only depends on its properties but also depends on the processing capabilities. NiTinol’s poor formability and sensitiveness to the higher temperature makes it as one of the unsuitable material for high-temperature processes. With the need to develop more complicated and more significant structures, development of suitable welding process for joining NiTinol becomes inevitable. Of the various welding processes, the laser welding process is found to be amore suitable process for joining NiTinol shape memory alloys. The main reasons for preferring laser welding for joining NiTinol are

  • Higher power density
  • Lesser distortion
  • Lesser heat affected zone width
  • Higher welding speed.

The major problems associated with joining of NiTinol are deterioration in the properties such as;

  • Microhardness
  • Ultimate Tensile Strength
  • Corrosion resistance

The shift in phase transformation temperature which in turn affects the super elasticity and shape memory effect.
The significant parameters that control the weld properties in laser welding are laser power, welding speed, beam diameter, shielding gases used. The total heat input to the weld which in turn controls the properties is mainly controlled by laser power and welding speed.
When the heat input is sufficient to get full penetration and minimum, then it is possible to obtain improved mechanical properties and corrosion resistance in the weld. The lower heat input values are generally achieved either by using smaller laser power or higher welding speed. The other primary concern with respect to the laser welding is the reflection by the material which in turn results in the incomplete penetration. One of the possible options to overcome this problem is to use lower wavelength lasers such Yb: YAG as it results in better absorption. Another solution is to bring the surface roughness to an optimized value as it helps in better absorption.
In most of the works related to laser welding, importance is given to the shielding gas flow rate, but no importance is given to shielding gas blown distance and shielding gas blown angle. All the shielding gas related parameters have a significant influence on the stability of the keyhole, which affects the porosity formation in the weld. It was understood that the selection of optimized blown distance and blown angle reduced the porosity formation which in turn lead to the increased mechanical and corrosion properties.
The microstructure of the laser-welded NiTinol shown a change of solidification mode from planar to cellular to dendrites on moving from the weld interface to the weld center. The NiTinol weld microstructure was controlled by two factors, viz. temperature gradient and growth rate. When the G/R ratio was low, it resulted in the equiaxed dendritic structure, and when G/R ratio was high, the planar structure was seen in the weld. Hence, it is recommended to select laser welding parameters such that it results in a low value of G/R which in turn lead to better weld mechanical and corrosion properties.
Phase transformation temperatures of the NiTinol shape memory alloys are hugely affected by the welding process parameters. The Phase transformation temperature plays a vital role with respect to shape memory effect. If the phase transformation temperatures are altered after welding, then the material will fail in its intended application. Phase transformation temperature is measured with the help differential scanning calorimeter (DSC) by changing the temperature at a rate of 10 ℃/minute. Based on the exothermic and endothermic reactions the austenite start, finish and martensite start, finish temperatures are identified. In most of the works related to laser welding, the phase temperatures of the weld have shown a significant variation due to the formation of intermetallic phase such as Ni3Ti,etc. These intermetallic phases, in turn, affect the nickel content present in the sample. It is understood that even 0.1 % variation in Nickel content would lead to shifting in phase transformation temperature by 10 ℃. Hence importance should be given to the laser welding parameters such as laser power, welding speed and beam diameter which significantly controls the heat input. When heat input to the weld is higher, the cooling rate becomes lower which in turn results in the formation of more brittle intermetallic phases. These intermetallic phases affect the phase transformation temperatures. Hence, welding parameters have to be chosen such that; the cooling rate is faster which in turn prevents the formation of brittle intermetallic phases. It is also recommended to subject the samples to annealing process before welding, as it avoids the formation of R-phase. R- Phase formation in the weld hinders the nucleation of martensite phase hence altering the martensite start and finish temperatures. R-phase formation occurs in both ways of transformation, i.e., from austenite to martensite or from martensite to austenite transformation but it is mostly observed during austenite to martensite transformation. But in most of the laser-welded NiTinol samples, R-phase was found to be forming during austenite to martensite transformation.
In most of the works related to laser welding, mechanical properties such as tensile strength and microhardness are given importance. Human bone has atensilestrength in the range of 150 Mpa and microhardness in the range of 150 HV. Hence, laser welding has to be done in such a way that the mechanical properties are meeting the requirement of biomedical applications. In most of the works, related to laser welding, the tensile strength of the weld was found to be lower than the base metal. The tensile strength of the weld is affected by the grain shape, size, and presence of intermetallic phases and defects. Porosity formation in the weld should be avoided as it reduces the load-bearing capacity. Heat input should be optimum such that the full penetration is achieved and better mechanical properties are obtained. When the heat input to the weld is higher, solidification cracks were seen in the weld center. In almost all the works related to laser welding, tested tensile samples broke in the weld zone and most of the samples failed in a mixed mode of fracture, i.e., both ductile and brittle mode, and it was confirmed by the presence of dimples and river marking in the fractography. Microhardness value of the laser-welded samples was found to maximum when the heat input to the weld is higher and vice versa. A trend of decrease in microhardness value was observed on moving from weld center to interface. The reason for this microhardness variation is due to the grain size variation in the different zone of the weld.
Corrosion resistance of the laser-welded sample has to be given utmost importance for laser welded NiTinol samples if they are used in biomedical applications. 0.9 % NaCl solution is the equivalent to human body fluids, hence corrosion resistance of the laser-welded specimens are analyzed in this solution. Potentio dynamic polarization measurement method is generally used for analyzing the corrosion resistance. The potential is varied from -1 to +1 V at a scan rate of 0.8 mV/seconds. From the potentio dynamic polarization curve, three different parameters, Viz. corrosion current density, corrosion potential, and polarization resistance are interpreted. Based on these values, the corrosion resistance of the weld and base metal are analyzed. Corrosion resistance will be better if the corrosion current density values are lower, and vice versa. Similarly, corrosion resistance will be better if the corrosion potential, polarization resistance is higher. Corrosion resistance of the weld is mainly controlled by two factors, Viz. surface finish and Ti/Ni ratio. If the surface finish is poor, then the corrosion resistance will be reduced, and if Ti/Ni ratio is higher, corrosion resistance will be higher. Hence, care must be taken in laser welding, such that it results in a higher value of Ti/Ni. A higher amount of Ti/Ni results in the formation of the stable passive layer made of Tio2 which protects the surface from corrosion.  It is understood that higher amount of Ti/Ni is possible when the cooling rate of the weld is higher. The cooling rate of the weld will be higher when the laser power is lower, or welding speed is higher. Presence of defects such as porosity, cracks also hinders the corrosion resistance and hence importance should be given to shielding gas parameters such as the type of shielding gas used, shielding gas flow rate, shielding gas blown angle and blown distance.
Heat treatment is one of the ideal options to vary the properties of laser welded NiTinol shape memory alloys. The problems such as lesser tensile strength, a shift in phase transformation temperature, reduced corrosion strength could be overcome by selecting correct post weld heat treatment parameters such as heating temperature, soaking time and cooling methods. The recrystallization temperature of the NiTinol is around 700 0C. Hence, it is recommended to set the post weld heat treatment temperature below 700 0C in order to avoid the formation of coarser grains. Depending on the temperature and aging time, three different intermetallic compounds, viz. Ni3Ti, Ti3Ni4, Ti2Ni3 gets formed in the NiTinol shape memory alloys. Post weld heat treatment in the temperature range of 200 0C – 400 0C results in improved mechanical and corrosion properties. Post weld heat treatment helps in reliving the residual stress and also helps in breaking the tangled network of dislocation.
A short review is done with respect to the laser welding of NiTinol shape memory alloys, and it was understood that laser welding is suitable for joining NiTinol shape memory alloys. Importance should be given to laser power, welding speed and shielding gas parameters as they mainly control the mechanical and corrosion properties. The only drawback with respect to laser welding is the economic aspects, and in future, an attempt can be made to overcome this by joining NiTinol sheets using pulsed TIG welding process as it is one of the low heat input and high precision processes. The initial investment is also very low on comparing with the laser welding setup. Post weld heat treatment is also recommended as it improved the corrosion properties. Post weld heat treatment through laser power source can be explored in future.
Ali. M. M and K. Takahata (2010) Frequency-controlled wireless shape-memory-alloy microactuators integrated using an electroplating bonding process. Sensors and Actuators A: Physical163(1), 363-372.
Bunget, Gheorghe, and Stefan Seelecke (2009) Actuator placement for a bio-inspired bone-joint system based on SMA. In Proc. SPIE, 7288, 1-15.
Butera. F (2008) Shape memory actuators for automotive applications. Advanced Materials Processes66, 37-44.
Butera. F, A. Coda, G. Vergani Shape memory actuators for automotive applications, In:Nanotec IT newsletter, Roma: AIRI/nanotec IT, 2007
Chee Siong. L, H. Yokoi and T Arai Improving heat sinking in ambient environment for the shape memory alloy (SMA), IEEE/RSJ international conference on intelligent robots and systems, 2005.
Colorado. J., A. Barrientos, C. Rossi and K. S. Breuer (2012) Biomechanics of smart wings in a bat robot: morphing wings using SMA actuators. Bioinspiration & Biomimetics7(3), 036006.
Festo BionicOpter – Inspired by dragonfly flight, Festo, 2013.
Furst. S. J., G. Bunget and S. Seelecke (2012) Design and fabrication of a bat-inspired flapping-flight platform using shape memory alloy muscles and joints. Smart Materials and Structures22(1), 014011.
Stoeckel. D (1990) Shape memory actuators for automotive applications. Materials & Design11(6), 302-307.
Stoeckel. D. F Tinschert Temperature compensation with thermovariable rate springs in automatic transmissions, SAE technical paper series: SAE, 1991.
Tuna C, J. H. Solomon, D. L. Jones and M. J. Z Hartmann (2012) Object shape recognition with artificial whiskers using tomographic reconstruction. In: 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); p. 2537–403.
Pfeifer. R., C. W. Müller, C. Hurschler, S. Kaierle, V. Wesling and H. Haferkamp (2013) Adaptable orthopedic shape memory implants. Procedia CIRP5, 253-258.
Stirling. L., C. H. Yu, J. Miller, E. Hawkes, R. Wood, E. Goldfield and R. Nagpal (2011). Applicability of shape memory alloy wire for an active, soft orthotic. Journal of materials engineering and performance20(4-5), 658-662.
Pelton. A. R., V. Schroeder, M. R. Mitchell, X. Y. Gong, M. Barney and S. W. Robertson (2008) Fatigue and durability of Nitinol stents. Journal of the mechanical behavior of biomedical materials1(2), 153-164.
Miyazaki. S., H. Y. Kim and H. Hosoda (2006) Development and characterization of Ni-free Ti-base shape memory and superelastic alloys. Materials Science and Engineering: A438, 18-24