Effect of Nano-TiC Addition on Microstructure and Corrosion Resistance Behavior of NiTi Alloy Fabricated by Selective Laser Melting

NiTi shape memory alloys (SMAs) have excellent shape memory,¹  superelasticity,²  good corrosion resistance,³  and biocompatibility,⁴  making them promising for use in the medical field as biomedical implant materials.⁵  Although NiTi alloys exhibit excellent comprehensive performance, the high ductility and strong work hardening of the material make them very difficult to fabricate into aortic stents using conventional processing techniques, ultimately resulting in poor-quality workpieces.⁶  To address this issue, emerging additive manufacturing (AM) technologies are typically considered ideal solutions for fabricating NiTi alloy products with complex structures.⁷  Additionally, recent studies have indicated that AM can fully leverage the performance of NiTi alloys while ensuring high precision.⁸  In general, multiple laser AM processes can be applied for manufacturing NiTi alloys, including selective laser melting (SLM),⁹  electron beam melting (EBM),¹⁰  laser engineered net shaping (LENS),¹¹  and laser powder bed fusion (LPBF).¹²  Because of its advantages of simple processing technology, material saving, high precision, and strong productive efficiency, SLM is the most popular process for AM of various alloys.¹³  Moreover, previous SLM studies in many aspects, such as processing,¹⁴  microstructure,¹⁵  and mechanical properties,¹⁶  have also confirmed that SLM is an effective and rapid approach for preparing NiTi alloys.

The functional properties of NiTi alloys have been developed to fabricate precision structures for applications in the medical field,¹⁷  e.g., vascular stents,¹⁸  orthodontic braces,¹⁹  and other applications.²⁰  However, it is common to observe that vascular stents fabricated using NiTi alloys may undergo metal corrosion in the complex physiological environment of the human body during usage, thereby increasing the risk of thrombus formation and affecting patient recovery.²¹  Therefore, adequate safety and corrosion resistance are the basis for NiTi alloy parts for use in the interventional medical field. The corrosion resistance of NiTi alloys is strongly influenced by their microstructure. However, during SLM, the strength and corrosion resistance of NiTi alloys may be severely weakened because of the presence of intolerable microstructures with large columnar grains and an irregular second phase, which may have resulted from complex melting and solidification dynamics during the melting–solidification cycles.^22-23  Consequently, it is critical for the study to identify suitable ways for controlling the microstructure and improving the properties of SLM-fabricated NiTi alloys.

Generally, the microstructure of NiTi alloys can be modified by adjusting their compositions. However, alloying may modify the strength, storage modulus, and damping properties of NiTi alloys.²⁴  The addition of Sc, Hf, and Zr can significantly reduce the plasticity of these alloys.^25-27  Additionally, particle-reinforced metal matrix composites incorporate the metallic characteristics of the matrix and the properties of the reinforcement, offering excellent performance, such as high strength and good dimensional stability.²⁸  Thus, drawing on this design concept, the addition of second-phase particles to the NiTi matrix is an effective approach to improve the microstructure and properties of NiTi alloys. Lu, et al.,²⁹  proved that the grain and microstructure uniformity of AM-fabricated NiTi alloy were significantly refined and improved, respectively, after La₂O₃ addition, which could be mainly attributed to the synergistic action of La₂O₃ as heterogeneous nuclei. Therefore, for improved corrosion resistance, it is essential to adopt a suitable reinforcing phase for NiTi alloys.

Among various ceramics, including SiC,³⁰  Al₂O₃,³¹  TiB₂,³²  ZrO₂,³³  and TiC,³⁴  TiC is a suitable reinforcement material because of the absence of interface reactions and minimal impact on the shape memory effect of NiTi alloys. In the past decades, TiC at different contents has been widely added into NiTi alloys as second-phase particles using the powder metallurgy method to refine the grain and improve the mechanical properties.³⁵  In recent years, Zhang, et al.,³⁶  fabricated TiC–NiTi composites using laser-directed energy deposition. The results indicated that TiC addition refined the Ni₄Ti₃ precipitates and enhanced the strength and pseudoelasticity of NiTi. Meanwhile, Chen’s group³⁷  proved that NiTi SMAs with TiC addition prepared using LENS effectively improved the yield strength. Naturally, TiC nanoparticles can also be used to promote the corrosion resistance of AM-fabricated NiTi alloys. Although some scholars^38-40  have attempted to introduce TiC as a reinforcement phase in the process of preparing NiTi alloys using LSM technology, the research has primarily focused on improving the strength of NiTi alloys and enhancing superelasticity. However, there is limited research on using TiC as a second phase to promote the corrosion resistance of NiTi alloys.

To investigate the influence of TiC addition on the microstructure, corrosion resistance, and properties of SLM-fabricated NiTi alloys, including NiTi–TiC composites with different mass TiC ratios, were fabricated. Furthermore, the effect of TiC reinforcement on the microstructure and corrosion resistance of the alloys was investigated in detail.

The microstructure, phase, and texture of TiC were characterized using scanning electron microscopy (SEM, HITACHI-S-4800^†), XRD (Bruker D8^† advance), and a metallurgical microscope (Leica DM2700M^†). XRD was performed for phase constitution variations of different samples at room temperature. Texture analysis based on XRD was performed with Cu Kα radiation in the 2θ angle range of 30° to 80° at a scanning rate of 2°/min (λ = 0.15418 nm). Moreover, to explore the phase constituents using SEM and a metallurgical microscope, various samples were treated with Kroll’s reagent (80% H₂O + 15% HNO₃ + 5% HF). In addition, the microstructure of each sample was further characterized using energy-dispersive x-ray spectroscopy (EDS) and an EBSD (Oxford^† C-nano), along with SEM. The required samples were prepared by electrolytic polishing in a solution consisting of 6% HClO₄ and 94% CH₃OH by volume. Variations in phase transformation temperatures (TT_s) due to TiC addition were determined using differential scanning calorimetry (DSC^†, German Naichi DSC214) with a heating/cooling rate of 5°C/min in nitrogen atmosphere from −100°C to 50°C.

For the corrosion resistance test, the different bulk samples were measured using an electrochemical workstation (Princeton Versa STAT 3^†) with a three-electrode system in a 0.9 wt% NaCl solution. The NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples were each used as the working electrodes, Pt was used as the auxiliary electrode, and a saturated calomel electrode (SCE) was used as the reference electrode. The dynamic potential polarization curves and electrochemical impedance spectra of the three specimens were measured through the corresponding electrochemical tests.

The grain size within the scanning area of each sample was also measured and plotted in bar charts, as shown in Figures 4(d) through (f). Based on the grain size calculation, the average grain size of the NiTi sample is approximately 50.6 μm. The grain sizes of the NiTi-TiC-0.5 and NiTi-TiC-1 samples are fined to approximately 5 µm, which is a 95% reduction. By comparison, although the average grain size remains the same, an increase in TiC addition can reduce the number of large grains, resulting in a significant decline in the grain size distribution width. Thereby, a fine and equiaxed microstructure can be obtained by increasing the TiC content (1 wt%). The disappearance of columnar grains and the formation of numerous equiaxed grains reveal that nanocrystalline TiC can serve as the nucleus for heterogeneous nucleation during SLM, thereby achieving grain refinement.

In this study, sodium dodecyl benzene sulfonate (SDBS), constituting 5% of the TiC mass, was introduced into the electrostatic self-assembly process, leading to partial SDSB residue on the surface of the composite powders of NiTi-TiC-0.5 and NiTi-TiC-1. By performing SLM under the argon atmosphere, the residual organic groups can be converted into carbon.⁵²  Based on previous research, deviation from the ideal stoichiometric ratio results in the formation of carbon-rich TiC_x grains when the carbon content becomes excessive.^53-54  Because excess carbon is formed as an amorphous matrix embedding nanocrystalline TiC particles, the characteristic peaks for TiC_x are consistent with those for TiC of rocksalt structure.⁵⁵ 

In general, NiTi alloys produced through the SLM process are anticipated to contain specific nano-precipitate phases.⁴⁰  However, because of their potentially low concentrations, distinct characteristic peaks may not be easily observable. Consequently, it requires further characterization to enable an in-depth investigation.

In addition, as SLM is processed through a full melting/solidification manner, therefore, during the process, the TiC or TiC_x can be both formed through a dissolution/re-precipitation mechanism via the heterogeneous nucleation of TiC nuclei and subsequent grain growth.³⁵  Therefore, as shown in Figures 6(b) and (c), the numerous nanoscale fine grains homogeneously disperse on the B2 phase after TiC addition. The insets show two different phases with diverse microstructural morphologies. According to the EDS analysis, it is supposed that the phase exhibiting a large pellet shape is laser-sintered TiC, present in low content, whereas the phase with a flower-like morphology is carbon-rich TiC_x.⁵⁷  In contrast to the sporadically distributed TiC particles, a substantial number of flower-like TiC_x dendrites with approximately 100 nm, dispersed along the grain boundaries of the B'19 phase, not only refine the grains but also can help to form a passivation film, thereby enhancing the corrosion resistance of the NiTi alloy matrix. Moreover, increased TiC content leads to larger partial TiC particles due to crystallite fusion and growth, and also results in the coarsening of TiC_x dendrites. In general, these large reinforcement phase particles may facilitate the formation of passive films and improve the corrosion resistance of the NiTi-TiC-1 sample.

In addition, the heat enthalpies of phase transition ΔH_A→M (from the austenite B2 phase to the martensite B19′ phase) and ΔH_M→A (from the martensite B19′ phase to the austenite B2 phase) were measured by integrating the corresponding DSC peaks. The corresponding calculation results are shown in Figure 9(a). When TiC is not added, the ΔH_A→M and ΔH_M→A of the NiTi sample are 17.40 J/g and 17.94 J/g, respectively. Because a decrease in TTs can reduce the corresponding enthalpy of phase transitions, the ΔH_A→M and ΔH_M→A of the NiTi–TiC-0.5 and NiTi–TiC-1 samples with TiC addition decrease to 5.99 J/g and 5.42 J/g and 9.41 J/g and 8.95 J/g, respectively. In comparison, the introduction of TiC significantly decreases ΔH_A→M and ΔH_M→A, whereas the enthalpy loss due to a further increase in TiC content is minor. This can be illustrated as follows: During the transformation process, mismatches resulting from the incompatibility of transformation strains between the matrix and particles induce internal stress in the matrix, thereby altering the kinetics of the transformation, and the enthalpy of the phase transition.⁶² 

The open-circuit potential (OCP) represents the voltage variance between the working and reference electrodes in the absence of a load. Consequently, its value is primarily influenced by the material and the corrosive solution. In general, a variation in the OCP represents the progressive evolution of the electrode from an initial unstable condition to a final stable state within the corrosive solution.⁶³  Thus, the voltage sweep range of the polarization curve for the specimens can be determined by identifying the OCP values.

Figure 10 shows the OCPs of the NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in a 0.9 wt% NaCl solution. Notably, the different OCP values stabilize rapidly after a short decline, and these values are approximately −0.3 V. The stable OCP of the NiTi sample is approximately −0.32 V, with certain fluctuations due to the corrosion dissolution of the passive layer on the surface.⁶⁴  With the addition and elevated content of TiC, the OCPs of the NiTi–TiC-0.5 and NiTi–TiC-1 samples are increased to −0.29 and −0.27, respectively, which are significantly higher than that of the sample without TiC addition. In addition, the fluctuations in the OCP curves of the NiTi-TiC-0.5 and NiTi-TiC-1 samples are significantly less. TiC addition can improve the potential and reduce the corrosion tendency on the surface of NiTi specimens. In conclusion, the NiTi sample without TiC addition demonstrates a marginally lower steady-state OCP with wider fluctuations than the NiTi-TiC-0.5 and NiTi-TiC-1 samples under identical preparation conditions.

The corrosion parameters extracted from the polarization curves are summarized in Table 2. From the results, we conclude that the corrosion potentials of the different NiTi samples exhibit little variation with increasing TiC content, whereas the corrosion current density gradually decreases. These results indicate that the addition of TiC significantly reduces the corrosion rate of NiTi alloys even though it is difficult to weaken the corrosion trend of NiTi alloys. The order of i_corr from low to high is as follows: NiTi-TiC-1 < NiTi-TiC-0.5 < NiTi. Thus, the NiTi-TiC-1 sample exhibits the optimum corrosion resistance because of the similar E_corr values among the three samples. Moreover, all i_corr values are less than 0.2 μA/cm², which suggests good formation of the protective passive layer. In addition, the E_pit of these samples increases with TiC content, which further confirms that TiC addition can effectively enhance the corrosion resistance of NiTi alloys. In conclusion, TiC with a heterogeneous nucleation effect can significantly improve the corrosion resistance of NiTi alloys.

The Bode diagrams of all samples in Figure 12(b) show minimal changes as the TiC content increases. Their phase angles exhibit a broad peak in the low-middle frequency range, indicating two superimposed individual time constants: the double layer and the passive film. Figure 12(b) shows the high |Z|_f→0 values and phase angles close to −90° for the three samples, suggesting typical high corrosion resistance due to the formation of a stable passive film on the sample surfaces.⁶⁹  The NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples perform larger |Z|_0.01 values and higher phase angles than the NiTi sample, revealing that TiC addition can lead to more perfect capacitive behavior and thus significantly improve the corrosion resistance of NiTi alloys.⁷⁰ 

The fitting values show that the SLM-fabricated NiTi alloys with or without TiC addition exhibit R_s of approximately 20 Ω, which remains roughly constant. Furthermore, resistance R_s and R_b increase with TiC content, indicating an obvious improvement in the corrosion resistance of NiTi alloys. In general, polarization resistance (R_polar), the sum of R_s and R_b, is commonly used to assess corrosion resistance. The NiTi–TiC-1 sample has the largest R_polar value (1.49 × 10⁷ Ω·cm²), demonstrating that the protective ability of the passive film can be enhanced through TiC addition.

Hence, the lower C_b value for NiTi-TiC-1 suggests that a thicker passive film is formed on the surface. Due to the positive correlation between the thickness of the passive film and the corrosion resistance of the alloys, the NiTi-TiC-1 sample exhibits better performance, which is in line with previous studies.

Moreover, we performed SEM characterization of the NiTi, NiTi–TiC-0.5, and NiTi–TiC-1 samples following electrochemical corrosion, and the results are presented in Supplemental Figure S1. It is observed that with an increase in TiC content, the number of corrosion pits decreases while the surface integrity of the samples improves. The analysis reveals that the addition of nano-TiC can effectively improve the corrosion resistance of NiTi alloys.

In this work, we carefully studied the microstructure, phase transformation, and corrosion resistance of NiTi alloys fabricated by the SLM process, both with and without the addition of TiC. The main findings are summarized as follows:

• During the SLM process, the added TiC served as a nucleus for heterogeneous nucleation, promoting grain and microstructure refinement and inducing a transition from columnar to equiaxed grains. Moreover, nano-TiC particles adhering to the NiTi surface could increase the laser absorptivity of the powder and improve the heat transfer characteristics, thus altering the molten pool dynamics and restraining the formation of keyhole pores.

• With TiC dispersed throughout the B2 phase, this reinforcing phase effectively suppressed the generation and expansion of martensite in NiTi alloys and refined the grains, potentially leading to improved shape memory characteristics. Besides, crystallographic orientation was altered by the addition of TiC, resulting in a more discrete distribution of crystallographic orientation.

• The addition of TiC did not vary the martensitic transformation mechanism of NiTi alloys. However, because of an internal stress induced by the mismatch between the NiTi matrix and TiC particles, the growth of TiC led to a reduction in transformation temperatures and a decrease in the enthalpy of the phase transition, resulting in a higher amount of B2 phase at room temperature.

• Compared to the NiTi sample without TiC, the corrosion resistance of the sample with TiC addition was significantly improved, as reflected by corrosion potential and corrosion current density. This improvement could be attributable to the refinement of the microstructure and formation of large reinforcement phase particles. A thicker protective passive layer was formed in NiTi-TiC-1 sample on the EIS results, as confirmed by EIS results, indicating superior corrosion resistance.

Min Jin: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Writing: original draft.

Qing Zhou: Supervision, Writing: Review & Editing.

Shuchun Li: Methodology.

Haitao Zhang: Resources, Validation.

Pengcheng Lu: Validation, Formal analysis.

Dongjin Wang: Conceptualization, Methodology, Data curation, Writing: Review & Editing, Project administration.

Yi Cao: Conceptualization, Methodology, Supervision, Funding acquisition.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was financially supported by Nanjing Medical Science and the Technology Development Foundation (grant numbers YKK23091), National Natural Science Foundation of China (grant numbers 82241212, 82270346), Natural Science Foundation of Hubei Province (grant numbers 2023AFD028), and Open project of National Laboratory of Solid State Microstructures (grant numbers M36003).