NiTi shape memory alloys, known as Nitinol, are highly valuable in medical fields for their unique attributes, including superelasticity, wear resistance, and biocompatibility. Laser treatment provides precise control over surface characteristics, enhancing biocompatibility. This study focuses on the effects of laser irradiation on NiTi alloy surfaces, particularly considering the number of laser scans and their impact on surface features. Even at low laser power, multiple high-frequency scans significantly alter surface roughness and induce phase transformation. After 16 repeated laser irradiations, amorphous Ti oxide transforms into crystalline anatase. Remarkably, anatase can further transform into rutile due to the influence of Ni nearby and TiO, due to insufficient oxygen content. The most notable outcome is the formation of a thick Ti oxide layer, causing unbound Ni to emerge on the surface, resulting in a Ni oxide layer. These findings highlight the importance of precisely adjusting laser parameters to achieve tailored surface properties for medical applications, addressing challenges and enhancing biocompatibility.
Graphical abstract
×
Notes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
NiTi-based shape memory alloys, commonly referred to as Nitinol, have found extensive utilization within the medical field, spanning disciplines such as orthopedic, orthodontics, cardiovascular, and clinical instruments[1, 2]. The unique properties, including superelasticity[3], wear resistance[4, 5], and biocompatibility [6, 7], have established Nitinol as a highly sought-after material for medical applications[1, 8‐10]. However, given that post-implantation within the human body can elicit varied cellular responses[11, 12], which are influenced by the physicochemical properties of the material’s surface[13, 14], extensive research endeavors are focused on regulating these cellular responses and improving biocompatibility. Biocompatibility primarily hinges on the absence of toxicity, with equal emphasis on mitigating adhesion and infection risks arising from cell adhesion to the material’s surface.
NiTi alloys contain a significant amount of nickel (Ni) elements, approximately 50 to 51% by atomic percent, which can potentially induce allergic contact dermatitis and toxic reactions when exposed to the human body as Ni ions [15, 16]. Therefore, one approach to prevent the exposure of Ni elements on the surface and enhance corrosion resistance is by creating a layer of titanium (Ti) oxides on the surface. Additionally, increasing surface roughness can enhance cell adhesion properties. For instance, Ti oxide on the surface of titanium implant not only prevents the infiltration of harmful metal compounds into the body but also, when surface roughness is adjusted, improves cell adhesion and accelerates bone integration [17]. Consequently, various surface modification techniques, including surface coatings [18], anodic oxidation [19‐21], and laser treatment [22‐28], are under investigation for their application in biomedical materials to achieve the desired surface topology [29].
Advertisement
Among these methods, laser treatment stands out as an effective approach to enhance biocompatibility by forming a Ti oxide layer on the surface of NiTi alloys, thus preventing the harmful exposure of Ni elements while providing the desired surface topology [30]. Laser treatment demonstrates superior performance compared to other surface treatment methods in terms of precise control and uniform surface topology [31]. Furthermore, its short processing time mitigates material property degradation and contamination commonly associated with direct-contact surface treatment methods [31‐33]. Therefore, the only significant consideration is the naturally formed surface oxide layer during repetitive surface melting and solidification processes.
The type and thickness of the surface oxide depend on the parameters of the laser treatment and the operating environment. Typically, thermally induced surface oxide on NiTi alloys in atmospheric conditions consist of Ti oxides due to the higher oxygen affinity of Ti compared to Ni [34]. However, since most studies on oxidation behavior were conducted under long-term exposure condition at specific temperatures [35‐37], it becomes challenging to extrapolate their findings to laser-induced oxidation process occurring during picosecond or nanosecond melting and re-solidification events. To the best of the author’s knowledge, there is no report available regarding the evolution of nanoscale surface structures on NiTi alloy after extremely short laser-induced heat treatment.
In this study, we examined the surface oxidation and phase transformation behavior of the NiTi alloy in relation to the number of laser irradiations. Specifically, nanoscale structural analysis utilizing transmission electron microscope (TEM) has elucidated the evolution process of oxide layers on the order of several tens of nanometers, as well as the occurrence of martensite transformation induced by excessive Ni content.
2 Experimental Methods
2.1 Sample Preparation by Laser Irradiation
A commercial equiatomic NiTi alloy was utilized, and it was mechanically polished with 100, 600, 1000, and 2000 grit SiC paper to remove contaminants and create a smooth surface for laser treatment. The alloy's surface was irradiated using a nanosecond ytterbium fiber laser (IPG Photonics, USA), capable of providing a peak power of 15 kW, a pulse duration ranging from 4 to 200 ns, and repetition rates from 2 to 1000 kHz. The NiTi alloy surfaces were exposed to laser light with a wavelength of 1064 nm, an output power of 4 W, a pulse duration of 4 ns, a scanning speed of 0.5 m/s, and a repetition rate of 800 kHz, as illustrated in Fig. 1. The polished samples underwent irradiation in a single direction, with the laser beam being precisely controlled in terms of interval gap and overlap frequency. The laser irradiation width was 32 μm, and the interval gap between irradiating beams was 12.8 μm, resulting in a 60% overlap. Subsequently, depending on the surface roughness, we will refer to the sample after a single irradiation (n = 1) as 'nanopattern' and the one after 16 irradiations (n = 16) as 'micropattern' based on the surface roughness.
×
Advertisement
2.2 Morphology and Structure Analysis
The irradiation-induced surface roughness of the samples was evaluated by capturing scanning electron microscopy (SEM) images at a 52° tilt angle and by obtaining cross-sectional images after ion milling using a dual-beam focused ion beam (FIB) instrument (Helios NanoLab 600, FEI). The roughness of the patterned surface was quantified and characterized by atomic force microscopy (AFM) (XE-100, Park System). The non-contact mode was used to inspect the surface using a non-contact cantilever (PPP-NCHR, Park System). To analyze the multi-layered cross-sectional structure generated through laser beam irradiation, cross-sectioned TEM lamella samples were prepared using FIB. For the surface protection, Pt was deposited on the samples. The phase evolution beneath the surface was characterized in terms of crystal structure using selected area diffraction patterns (SADPs) and in terms of chemical composition through energy dispersive spectroscopy (EDS) analysis in a TEM (Talos F200X, FEI) equipped with EDS having four SDD detectors.
3 Results and Discussion
3.1 Surface Roughness as a Function of the Number of Laser Irradiations
The pristine NiTi alloy exhibits no distinctive features aside from surface polishing scratches, as seen in Fig. 2a. The arithmetical mean roughness, \({R}_{a}\) measured by AFM is 9.8 nm. Following a single laser irradiation (n = 1), a faint wavy pattern, referred to as ‘nanopattern’ emerges on the surface, with the \({R}_{a}\) of 27.2 nm, as more clearly depicted in the tilted image in Fig. 2b. After 16 rounds of laser irradiation, the surface roughness intensifies, yielding an \({R}_{a}\) of 732.5 nm, termed the ‘micropattern’ (Fig. 2c). The scale of the surface roughness, contingent on laser irradiation parameters, can be classified as macro-, micro-, nanometer-sized textures, each exhibiting varying effects on implantation. It is well-known that the micrometer-scale roughness enhances implant fixation to bones, while the nanometer-scale roughness relates to protein adsorption and the adhesion of osteoblastic cells [38]. Cross-sectional SE images are used to roughly measure oxide thickness and the heat-affected zone (HAZ). In the nanopattern sample, the HAZ measured as 218 nm, including the approximately 59 nm oxide film on the surface(Fig. 2b). Conversely, the HAZ depth significantly increases to 4.3 μm with 656 nm oxide film in the micropattern sample (Fig. 2c).
×
3.2 Surface Layer Structure in ‘Pristine’
To assess the oxide structures, TEM analysis was conducted on three samples: the pristine sample, the nanopattern sample, and the micropattern sample. The pristine sample was divided into four layers with contrasting features along the depth axis, as depicted in the bright-field (BF) TEM image in Fig. 3a. The predominant phase within the matrix is B19′ phase [39‐41], which remains thermodynamically stable at room temperature due to its high martensite start temperature (\({T}_{Ms}\)) inherent in equatomic NiTi alloy (as shown in Fig. 3a) [42, 43]. Above the matrix, a 24 nm-thick Ni-rich amorphous layer (Ni: ~ 59 at%, Ti: ~ 41 at%) was observed.
×
At the top surface, there exists a continuous 4 nm-thick layer displaying a distinct bright contrast. Chemical composition analysis via EDS revealed a composition close to TiO2 (Ti: ~ 28 at%, O: ~ 71 at%). This continuous amorphous Ti oxide layer, often referred as the native oxide, is a characteristic feature found on NiTi alloys and Ti-based alloys at room temperature, primarily due to the selective oxidation of Ti, which has a strong affinity [34] for oxygen. In the literature [44], it has been reported that the thickness of native oxide on Ti metal at room temperature typically falls within the range of 2–7 nm, consistent with the findings of this study. The selective oxidation of Ti on the surface induces the inward diffusion of Ni toward the interface between the oxide and the matrix, resulting in the formatin of a Ti-depleted Ni–Ti amorphous layer (Ni: ~ 69 at%, Ti: ~ 31 at%) with a thickness of approximately 10 nm beneath the Ti oxide layer. Below this, there exists a Ni3Ti2 amorphous layer (Ni: ~ 59 at%, Ti: ~ 41 at%) with a thickness of approximately 24 nm. All three of these layers present on the surface possess amorphous structures which is confirmed through the halo ring pattern observed in the fast Fourier transform (FFT) in Fig. 3a. High-angle annular dark-field (HAADF) STEM imaging and corresponding EDS mapping, as shown in Fig. 3b, c, clearly visualizes the Ti oxide layer along with the underlying Ni-enriched layers on the surface.
3.3 Surface Layer Structure Transformation in ‘Nanopattern’
Following a single laser irradiation, the surface structure undergoes a significant transformation, as illustrated in Fig. 4, resulting in the division of the HAZ into three distinct layers with a depth of 340 nm. At the uppermost surface, a Ti oxide layer exhibits a bright contrast, extending to a thickness of approximately 66 nm, a substrantial increase compared to the 4 nm native oxide observed in the pristine sample. Notably, spherical Ni particles ranging in size from 2 to 17 nm, displaying a gray contrast, are embedded within the Ti oxide film. The FFT pattern obtained from the oxide layer, including the embedded Ni particles, indicates the crystalline nature of the Ni particles, while the Ti oxide maintains its amorphous structure. Under atmospheric thermal oxidation conditions, Ni cations do not typically form Ni particles within the Ti oxide due to the selective oxidation of the Ti element; instead, they accumulate beneath the Ti oxide film[35, 45], as observed in the pristine sample (Fig. 3). However, the rapid solidification induced by nanosecond laser irradiation may facilitate the formation of some embedded Ni particles, as it does not provide sufficient time for Ni to diffuse below the oxide film. The HAADF STEM image and the corresponding EDS elemental map in Fig. 4b clearly depict the presence of trapped Ni particles within the Ti oxide layer.
×
Below the Ti oxide layer, a featureless dark contrast layer with a thickness of approximately 163 nm has been identified as amorphous NiTi, exhibiting a chemical composition close to Ni3Ti2. The relatively high Ni content (62 at%) comparison to the Ni50Ti50 matrix can be attributed to partial inner diffusion of Ni due to Ti oxide layer growing during laser irradiation. Furthermore, Ni3Ti2 alloy demonstrates sufficient glass-forming ability [46] to maintain its amorphous structure even after undergoing melting and re-solidification processes through rapid quenching induced by laser irradiation.
The third layer, not distinctly discernible from the B19’ phase of the matrix due to similar chemical composition and crystalline characteristics in the SEM image (Fig. 2c), has been identified as the B2 NiTi phase through SADP analysis. This phase transformation from the B19’ to the B2 phase can be attributed to the dependency of the martensitic transformation (MT) temperature on chemical composition. The EDS line profile analysis in Fig. 4a reveals a continuous variation in Ni and Ti content within the B2 NiTi layer. As a consequence of the outward diffusion of Ti, driven by selective oxidation towards the interface of amorphous Ni3Ti2 and B2 NiTi during surface melting induced by laser irradiation, the Ni concentration in the B2 NiTi layer becomes slightly elevated compared to the B19’ matrix. The MT temperature of binary NiTi alloys is highly sensitive to Ni content, as illustrated in Fig. 4c, where a 1 at% increase in Ni content results in a reduction of the austenite finish temperature (\({T}_{Af}\)) and martensite start temperature (\({T}_{Ms}\)) by more than 100 K [47‐49]. Consequently, when the Ni content surpasses 50.7 at%, the B2 austenite phase becomes more stable compared to the B19’ phase at room temperature due to the lowered \({T}_{Af}\) [42]. In other words, the decrease in \({T}_{Af}\) within the Ni-rich region of the matrix triggers the formation of the B2 layer above the B19’ matrix. As a result, the surface of the nanopatterned sample comprises four layers: crystalline B19’ and B2 NiTi phases, along with amorphous Ni3Ti2 and Ti oxide embedded with Ni nanoparticles.
3.4 Surface Layer Structure Transformation in ‘Micropattern’
Repeated laser irradiation, conducted 16 times (Micropattern), leads to the formation of a thicker oxide layer, approximately 695 nm in thickness, comprising multiple oxide layers (Fig. 5a). Notably, a continuous NiO layer with a thickness of around 10 nm is confirmed above the rutile phase thorugh EDS elemental mapping (as depicted in Fig. 5b). Building upon the presence of trapped Ni particles within the Ti oxide following a single laser shot, as observed in Fig. 4, we hypothesize that these trapped Ni particles act as a source for supplying the final thin NiO layer on the top surface after repeated laser irradiations.
×
Another intriguing observation in the samples subjected to repeated laser irradiation is the presence of not just a single Ti oxide layer, but three distinct oxide layers, namely rutile, anatase and fcc TiO layers. It is hypothesized that these oxide layers are formed through a three-step process: i) the intial formation of the anatase phase, ii) simultaneous out-diffusion of excess Ni elements from the anatase phase, leading to the formation of an amorphous Ni oxide layer, and iii) a phase transition from anatase to rutile.
Let us first consider the appearance of anatase. Fathi-Hafshejani et al. demonstrated that the surface oxide structure of Ti or Ti alloys is influenced by the heat generated during laser processing, including factors such as laser power, scanning speed, laser pulse duration, and frequency [50]. Under conditions of higher laser energy, frequence and slower irradiation speed, rutile TiO2 tend to preferentially form, whereas lower laser power and/or faster irradiation speed promote anatase formation. It is worth noting that the laser irradiation conditions employed in our study, specifically a laser power of 4 W and an irradiation speed of 0.5 m/s, can be categorized as relatively mild conditions, potentially falling within the unaffected zone of the heatmap presented in reference [50]. However, the higher frequency (800 kHz) and repeated laser scans, which provide sufficient power density, may contribute to the formation of the anatase phase.
Next, let’s examine the structure of the rutile phase. The formation of the rutile phase typically demands a higher amount of thermal energy compared to the anatase phase [50]. However, our experiments do not align with this scenario. Instead, it is anticipated that the excess Ni atoms in anatase TiO2, which remains even after the repeated laser scans, trigger a phase transformation from anatase to rutile [51‐53]. These excess Ni atoms substitute into the Ti sublattice of the anatase phase, creating oxygen vacancies due to the difference in valence between Ti4+ and Ni3+ ions. These oxygen vacancies induce relaxation within the large oxygen sublattice, leading to atomic rearrangement and the transformation from anatase to rutile [51]. The detailed mechanism for this phase transformation is depicted in Fig. 5c.
Lastly, let us consider how the TiO phase forms beneath anatase TiO2. TiO2 can undergo conversion into the TiO phase when there is insufficient oxygen supply. The presence of thick TiO2 phases, including rutile and anatase, creates challenges for oxygen penetration into the sample interior. Additionally, the large grain size of anatase reduces the number of grain boundaries, which serve as paths for ion conduction, thereby impeding the rapid movement of oxygen ions. The formation of the TiO phase beneath the rutile/anatase phase aligns with observations reported in other long-term thermal oxidation experiments [40, 41].
In the micropattern sample, crystalline Ni3Ti2 is observed just above NiTi matrix, not the amorphous Ni3Ti2 which is found in the nanopattern sample (Fig. 5a). The crystalline Ni3Ti2 is recognized as a metastable phase and its crystal structure is not well-defined. In this study, the diffraction pattern we obtained was similar to Ni3Ti2 with cubic structure shown in ref. [54]. Thus we identified it as Ni3Ti2 intermetallic phase based on EDS quantitative analysis results and diffraction pattern indexing results. Further, the excess oxygen in Ni3Ti2 layer results in the formation of nano-sized TiO2 particles in the grains as well as in the grain boundaries.
3.5 Laser-Induced Surface Microstructure Modification Steps in the NiTi alloy
The microstructural evolution of NiTi alloy upon the number of laser irradiation, encompassing changes in surface roughness, oxide layers and phase transformations is summarized in Fig. 6. With an increasing number of laser irradiations, the surface roughness also increases, leading to the designations "nanopattern" and "micropattern."
×
Upon a single laser irradiation, the native Ti oxide layer thickens, and the residual Ni forms crystalline Ni nanoparticles in Ti oxide. Additionally, the Ni3Ti2 amorphous layer also thickens to approximately 160 nm. Below this layer, a B2 structure is formed as the B19’ martensite undergoes a transformation due to excess Ni content. Repetitive laser irradiations lead to the development of TiO oxide layers, encompassing rutile, anatase, and TiO, with layered structures. On the top surface, Ni expelled from the TiO forms an exceedingly thin oxide layer. Despite the low laser power and scanning speed, the repetitive use of high-frequency laser scanning allows for the formation of anatase as a Ti oxide. At the interface between anatase and Ni oxide, Ni2+ probably acts as nucleation site of rutile and promote the phase transformation [55]. The lower region, however, does not experience sufficient oxygen diffusion from the surface, resulting in the formation of TiO. As a result, 4 oxide layers are evolved under repetitive laser irradiation. Furthermore, the Ni3Ti2 layer, formed at a considerable depth from the surface and no longer subjected to rapid cooling rates, no longer remains in an amorphous state.
In this study, we maintain a relatively low laser power while varying the number of laser irradiations under high repetition frequency conditions. We not only demonstrate the control of surface roughness but also propose that the type of surface oxide can change with the number of irradiations. Consequently, it is essential to consider how these surface structural alterations may impact different applications within the field of biomedical materials, depending on the intended site of use.
4 Conclusion
In this study, we employed high-resolution TEM to examine the cross-sectional nanoscale structure of NiTi alloy's surface layer resulting from extremely short laser-induced heat treatment. Our findings revealed that even with relatively low laser power, multiple laser scans at high repetition frequencies significantly impact both the surface roughness and the surface phase transformation process. Consequently, we identified changes in the type of surface oxide at the top layer. The following are key features observed at each stage:
1.
The pristine NiTi alloy consists of a B19' matrix with an extremely thin native Ti amorphous oxide layer, approximately 4 nm thick, on the surface.
2.
Following a single laser irradiation, involving rapid melting and re-solidification, the Ti oxide layer thickens to approximately 60–70 nm. It is noteworthy that during the oxidation process, residual Ni forms as nano-crystalline particles within the oxide layer. Another crucial point is the occurrence of selective Ti oxidation, leading to the formation of Ni-rich compounds beneath it. As a result, the upper part of the B19' phase undergoes martensitic transformation induced by excess Ni, resulting in the formation of the B2 layer.
3.
With 16 repeated laser irradiations, the amorphous Ti oxide becomes crystalline anatase. Notably, in this scenario, anatase can transform into rutile due to the influence of Ni or into TiO due to insufficient O content. The most significant finding here is the development of a thick Ti oxide layer, which causes the unbound Ni to emerge on the surface, forming a Ni oxide layer.
For future studies, careful adjustment of laser parameters is necessary to achieve the desired surface topology and oxide layer suitable for medical applications.
5 Author Contritutions
IJ, HJ, YCK and HJC conceived the idea and designed the experiments. IJ and HJ prepared patterned NiTi samples and AFM measurement. HC, MYN and HJC conducted microstructure analysis, interpreted the analytical results and wrote the manuscript. All authors discussed results and revised the manuscript. HJC supervised the project.
Acknowledgements
This work was supported by the Ministry of Trade, Industry, and Energy (MOTIE) of Korea through the project No. P0022331 supervised by the Korea Institute for Advancement of Technology (KIAT). I.J, H.J, Y.C.K was supported by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trad, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety)(NTIS Number:999007189)
Declarations
Conflict of interest
There are no conflicts to declare.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.