With the advancement of individual medicine, the additive manufacturing of NiTi alloys, which are used as biomedical implantation materials, is becoming increasingly popular. However, the insufficient corrosion resistance of these alloys still requires improvement. In this study, we fabricated various NiTi alloys with and without TiC addition using selective laser melting (SLM). The effects of TiC addition and content on the microstructure, phase transition, and corrosion resistance of SLM-fabricated NiTi alloys were thoroughly investigated. The results indicated that TiC addition promoted grain and microstructure refinement and induced a transition from columnar to equiaxed grains. In addition, the formation of martensite in NiTi was suppressed, which could be attributed to the synergistic action of grain refinement and second-phase dispersion toughening of TiC grains. Furthermore, because of the formation of a thicker protective passive layer with TiC addition, the corrosion resistance of the SLM-fabricated NiTi alloys was obviously improved. This study illustrated TiC addition as an effective method for enhancing the performance of SLM-fabricated NiTi alloys.

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 scholars38-40  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.

This study used NiTi alloys as raw materials, with an atomic ratio of 50% nickel and titanium each and a size of 30 μm to 100 μm (Avimetal AM Tech Co., Ltd.). TiC nanoparticles uniformly decorated the surface of the NiTi alloys using the electrostatic self-assembly method, which was proposed in 2017 by the research group of Martin, et al.41  During the process, the addition of surfactants adjusted the surface zeta potential values of the NiTi alloy and nano TiC, resulting in them becoming positively and negatively charged, respectively. Then, with the aid of ultrasound, mechanical stirring, and other methods, the two components were enabled to achieve electrostatic self-assembly. The TiC reinforcement phases were added at 0.5 wt% and 1 wt%. The two resulting composites are denoted as NiTi-TiC-0.5 and NiTi-TiC-1. The morphologies of the NiTi powders with and without TiC addition are shown in Figure 1.
FIGURE 1.

The morphologies of (a) NiTi powders, (b) NiTi-TiC-0.5 powders, and (c) NiTi-TiC-1 powders.

FIGURE 1.

The morphologies of (a) NiTi powders, (b) NiTi-TiC-0.5 powders, and (c) NiTi-TiC-1 powders.

Close modal
Bulk specimens corresponding to NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 with the same size of 10 mm × 10 mm × 5 mm were fabricated using an SLM device (EOS M290) under Ar atmosphere on Ti-6Al-4V rolled plates. A schematic of the corresponding SLM process is shown in Figure 2(a). The key technique parameters were as follows: layer thickness of 0.03 mm, laser power of 150 W, laser beam diameter of 0.1 mm, scan speed of 1,000 mm/s, and scan line space of 0.1 mm. The SLM device and a schematic of the SLM process are shown in Figures 2(a) through (c). A schematic of the island scanning strategy is shown in Figure 2(d). The vectors of adjacent island areas, measuring 5 mm × 5 mm, were perpendicular to each other, and the vector of the subsequent layer maintained an orthogonal relationship with the preceding layer. As shown in Figure 2(d), we defined a unified coordinate system for the samples with building direction (BD), normal direction (ND), and transverse direction (TD), which would be used for electron backscatter diffraction (EBSD) and x-ray diffraction (XRD) analyses.
FIGURE 2.

(a) Photograph of the SLM equipment used in this work, (b) and (c) schematic illustration of SLM process, and (d) schematic illustration of specimens.

FIGURE 2.

(a) Photograph of the SLM equipment used in this work, (b) and (c) schematic illustration of SLM process, and (d) schematic illustration of specimens.

Close modal

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 (TTs) 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.

3.1.1 |  Microstructure Analysis

To investigate the connection between the additional amount of TiC and the alloy microstructure, optical micrographs of the top layer (ND-TD plane) of various SLM-fabricated NiTi specimens with and without TiC addition are shown in Figure 3. The NiTi sample presents large grains with a dendritic structure, which displays a morphology similar to the top view of columnar grains observed in previous research. According to numerous literature reports,42  columnar grains growing along the BD possess average widths exceeding 50 μm during SLM, a trend that is consistent with the grain sizes observed in Figure 3(a). This suggests that all large grains can be attributed to columnar grains. The keyhole on the top side of the NiTi sample is predominantly located in the boundary of different molten pools. Thus, it can be deduced that these defects arise due to intense thermal convection induced by steep thermal gradients between the overlapping regions of the melt pools and their centers.43-44  In contrast, the addition of 0.5% and 1% TiC during SLM results in microstructure refinement and the disappearance of columnar grains and defects. As shown in Figures 3(b) and (c), the grain size gradually decreases with increasing TiC content, indicating mass formation of equiaxed grains. These results reveal that the introduction of TiC significantly refines the grain size and facilitates the transition from columnar to equiaxed grains. Moreover, based on the previous report,45  the TiC nanoparticles uniformly dispersed within the NiTi matrix are not directly exposed to laser irradiation. Instead, they dissolved into the molten pool through the sufficient wetting of the melted NiTi. During the re-precipitation process, TiC nanoparticles can act as a heterogeneous nucleation agent, thereby altering the molten pool dynamics and restraining the formation of keyhole pores.
FIGURE 3.

The optical photography of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1.

FIGURE 3.

The optical photography of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1.

Close modal
To obtain more microstructural information and finer details regarding the grain size, EBSD analyses were subsequently performed on the upper surface (ND-TD plane) of the three samples. The EBSD maps for NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 are shown in Figures 4(a) through (c), respectively. The relevant results were indexed with lattice information about the B2, B19’, and TiC phases. All drawings show an island pattern on the ND-TD plane, which is consistent with the scanning strategy results. As shown in Figure 4(a), the alloy tissues of the NiTi sample should comprise columnar grains. However, with the addition of TiC nanoparticles, the grain size is reduced, suggesting that the microstructure of the NiTi alloys has changed from columnar to equiaxed grains on the top layer.46  Furthermore, with increasing TiC content, the grains gradually become finer and smaller. For NiTi alloys, many columnar crystals may lead to defects, which may affect their corrosion resistance.47-48  The generation of numerous equiaxed grains after TiC addition can avoid various problems, including segregation, cracks, or internal pores.38,49-50 
FIGURE 4.

EBSD mapping of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1; Grain size information of (d) NiTi, (e) NiTi-TiC-0.5, and (f) NiTi-TiC-1.

FIGURE 4.

EBSD mapping of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1; Grain size information of (d) NiTi, (e) NiTi-TiC-0.5, and (f) NiTi-TiC-1.

Close modal

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.

3.1.2 |  Phase Characterization

Figure 5 shows the XRD patterns of the SLM-fabricated NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 alloys. XRD characterization was performed on a surface perpendicular to the BD of each specimen. From the image, the samples with or without TiC addition are mainly composed of an austenite phase (B2), as demonstrated in PDF 65-5537. According to PDF 35-1281, the main diffraction peak of (1-1 1)B19’ also appears near 42°, leading to its overlap with the peak of (110)B2. However, the low-intensity peak at 45° of the SLM-fabricated NiTi alloy is considered a characteristic peak belonging to the (012)B19’ plane. This result indicates a minor amount of martensite in NiTi, which can be attributed to the transformation temperature being too close to room temperature. However, because of the limited penetration depth of x-rays, the intensity of the martensite peak may not be representative of its bulk amount in the sample. The disappearance of the martensitic peak after TiC nanoparticle addition indicates that TiC, as a reinforcing phase, can effectively suppress the generation and expansion of martensite in the top region.51  XRD analysis can also reveal microstructure variables with TiC addition.36  Because the position of each B2 peak exhibits no significant change, the variations in the peak intensity of the (110)B2 and (200)B2 planes can be supposed as evidence of columnar-to-equiaxed grain transition (CET). The most intense peaks of NiTi-TiC-0.5 and NiTi-TiC-1 can be attributed to the (110)B2 peak of the B2 phase at 42.4°, and the characteristic peak corresponding to the (200) main crystal plane of TiC is also near 42°. To reveal the exact 2θ position of the diffraction peaks for TiC, XRD identification in the range of 40° to 46° (region A in Figure 4[a]) is shown in Figure 4(b). The (200) diffraction peaks belonging to NiTi-TiC-0.5 and NiTi-TiC-1 are observed at the same location (2θ = 41.7°).
FIGURE 5.

(a) XRD patterns and of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1; (b) XRD image of region A.

FIGURE 5.

(a) XRD patterns and of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1; (b) XRD image of region A.

Close modal

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 TiCx 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 TiCx 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.

To verify the presence of the TiCx phase and other nanoprecipitation, SEM and EDS were performed. Figures 6(a) through (c) exhibit the SEM micrographs of the NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples, and the insets show the corresponding high-magnification SEM micrographs. The EDS point analysis results for each sample are presented in Table 1. Figure 6(a) indicates that only a few nanoparticles are dispersed within the matrix of the NiTi sample. According to the previous analysis, in the matrix, the primary phase is B2, whereas the B19' phase is primarily located at the boundaries of the melt pool.22  Based on the EDS results, the nanoprecipitation is likely to be Ni4Ti3.56  As shown in the inset, the white precipitates marked as spot 1 are tightly linked to the matrix (marked as spot 2) with a grain size of about 1 μm. The proportion of Ni and Ti atoms in the two areas is inconsistent. The enrichment of nickel is observed at spot 1, while spot 2 has a Ni-Ti ratio of approximately 1:1. The EDS results illustrate that the matrix phases (spot 2) and precipitates (spot 1) are B2 and Ni4Ti3 phases, respectively.38 
FIGURE 6.

SEM images of different SLM fabricated NiTi alloys (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1.

FIGURE 6.

SEM images of different SLM fabricated NiTi alloys (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1.

Close modal
Table 1.

EDS Analysis at Different Areas Marked in Figure 6 (at%)

EDS Analysis at Different Areas Marked in Figure 6 (at%)
EDS Analysis at Different Areas Marked in Figure 6 (at%)

In addition, as SLM is processed through a full melting/solidification manner, therefore, during the process, the TiC or TiCx 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 TiCx.57  In contrast to the sporadically distributed TiC particles, a substantial number of flower-like TiCx 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 TiCx 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.

To more clearly identify these nanoprecipitates, the TEM characterization was conducted on the NiTi-TiC-1 sample (Figures 7[a] through [d]). Additionally, various high-resolution TEM (HRTEM) images and the corresponding selected area diffraction pattern (SADP) are present in Figures 7(e) through (m). As shown in Figure 7(a), the matrix of the sample mainly consists of the B2 phase (region A) and flower-like TiCx phase (region B). The HRTEM image of region A, exhibited in Figure 7(f), demonstrates that the matrix belongs to the B2 phase. Its SADP result is also consistent with the face-centered cubic (fcc) structure.58  Moreover, to view the flower-like TiCx more clearly, we have zoomed into region B and displayed it in Figure 7(b). The image reveals that the TiCx dendrites exhibit four petal-like structures, each with a particle size ranging from 20 nm to 30 nm. The EDS results are also in accordance with the previous, confirming the formation of carbon-rich TiCx nanocrystals through the re-precipitation process.35  The HRTEM image in Figure 7(f) indicates the (220) and (110) crystal faces, which belong to the TiCx phase and B2 phase, respectively. The corresponding SADP images, displayed in Figures 7(j) and (k), both conform to their respective lattice structures. Figure 7(c) shows a TiC particle with a large TiC particle, which can be attributed to the fusion and growth of nano-TiC grains during the SLM process. Its results of HRTEM and SADP essentially align with the TiCx. However, due to the variation in carbon content, they exhibit different spacing for the same crystal plane. From Figure 7(d), it is found a different nanoprecipitation phase, can be ascribed to the Ni4Ti3 according to its EDS analysis. The HRTEM shown in Figure 7(h) depicts a (112) crystal face, further confirming the inference regarding the formation of Ni4Ti3. Certainly, the corresponding SADP image (Figure 7[m]) also supports this conclusion.38 
FIGURE 7.

TEM bright-field images of (a) matrix, (b) region B, (c) TiC, and (d) Ni4Ti3. HRTEM images of (e) region A, (f) region B, (g) TiC and (h) Ni4Ti3. Corresponding SADP images of (i) region A, (j) through (k) region B, (l) TiC, and (m) Ni4Ti3.

FIGURE 7.

TEM bright-field images of (a) matrix, (b) region B, (c) TiC, and (d) Ni4Ti3. HRTEM images of (e) region A, (f) region B, (g) TiC and (h) Ni4Ti3. Corresponding SADP images of (i) region A, (j) through (k) region B, (l) TiC, and (m) Ni4Ti3.

Close modal

3.1.3 |  Crystallographic Texture

As shown in the XRD spectra in Figure 5, the crystallographic texture of the NiTi alloys changed with the addition of TiC. To discuss the variation in crystallographic orientation with TiC content, the inverse pole figure (IPF) maps and pole figure (PF) maps of the NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 specimens on the top side are present in Figures 7(a) through (f), respectively. The colors on the PF and IPF maps reveal that the B2 phase for the NiTi sample exhibits a more pronounced orientation along the <100> direction.59  From Figures 8(b) and (c), the weakened texture with growing TiC content indicates that TiC addition can result in a more discrete distribution of the crystallographic orientation. As shown in Figures 8(d) and (e), the NiTi sample has the highest microtexture distribution with a maximum index of 6.84, whereas the maximum polar densities of the NiTi-TiC-0.5 and NiTi-TiC-1 samples significantly fall to 2.15 and 1.45, respectively. This finding also determines that trace TiC can result in remarkable grain refinement in NiTi alloys.
FIGURE 8.

The inverse pole figure (IPF) maps of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1; and pole figure (PF) maps of (d) NiTi, (e) NiTi-TiC-0.5, and (f) NiTi-TiC-1.

FIGURE 8.

The inverse pole figure (IPF) maps of (a) NiTi, (b) NiTi-TiC-0.5, and (c) NiTi-TiC-1; and pole figure (PF) maps of (d) NiTi, (e) NiTi-TiC-0.5, and (f) NiTi-TiC-1.

Close modal
Figure 9 displays the DSC thermograms and phase transformation temperatures (TTs) of the SLM-fabricated NiTi alloys with and without TiC addition (NiTi, NiTi-TiC-0.5, and NiTi-TiC-1). As illustrated in Figure 9(a), because only one peak appears in the DSC curves during the heating or cooling process, the NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples exhibit a stable martensitic transformation (B2↔B19′). The findings from studies16,60  on the phase transformation behavior of SLM-fabricated NiTi alloys suggest that reducing heat input typically decreases nickel burn-off, thereby lowering the transformation temperature. However, a recent study35  has demonstrated that after adding TiC, reducing heat input increases the TTs. Moreover, previous research indicates that TiC particles, acting as a secondary phase within the NiTi matrix, can induce the movement and accumulation of dislocations, thus generating internal stress.61  Therefore, we can infer that the presence of secondary phase particles, including nano-TiC particles and TiCx dendrites, primarily alters the internal stress state of the NiTi matrix, which consequently reduces the TTs of the composite material. Figure 9(b) shows that TTs dramatically decreases with increasing TiC addition, which agrees well with previous reports. Specifically, the Ms temperature of NiTi is 16.4°C, whereas the introduction of 0.5 wt% and 1 wt% TiC nanoparticles results in significant decreases of 13.4°C and 20.4°C in Ms, respectively. However, it is noteworthy that the Ms temperatures of the three samples are lower than room temperature (25°C) irrespective of the incorporation of TiC. All SLM-fabricated samples undergo an extremely rapid cooling and solidification process. Because the Ms temperature coincides with the nucleation temperature of the NiTi (B19') phase, the amount of B19' generated is very limited once this temperature becomes lower than room temperature.61  Given that the TTs of all three samples fail to reach room temperature, they should predominantly form the austenitic B2 phase. A small quantity of the martensite phase can be developed in the NiTi sample, owing to the close approximation of its Ms temperature to room temperature. These results also agree with the XRD, SEM, and EBSD analysis.
FIGURE 9.

(a) DSC thermograms of SLM fabricated NiTi alloy with and without TiC addition and (b) TTs extracted from the DSC results of NiTi, NiTi-TiC-0.5 and NiTi-TiC-1.

FIGURE 9.

(a) DSC thermograms of SLM fabricated NiTi alloy with and without TiC addition and (b) TTs extracted from the DSC results of NiTi, NiTi-TiC-0.5 and NiTi-TiC-1.

Close modal
FIGURE 10.

OCP curve of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in 0.9 wt% NaCl.

FIGURE 10.

OCP curve of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in 0.9 wt% NaCl.

Close modal

In addition, the heat enthalpies of phase transition ΔHA→M (from the austenite B2 phase to the martensite B19′ phase) and ΔHM→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 ΔHA→M and ΔHM→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 ΔHA→M and ΔHM→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 ΔHA→M and ΔHM→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 

3.3.1 |  Open-Circuit Potential

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.

3.3.2 |  Potentiodynamic Measurements

Figure 11 exhibits the comparative analysis of the potentiodynamic polarization curves for the NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in a 0.9 wt% NaCl solution at 25°C. In the figure, the intersection of the two polarization potential curves represents the corrosion potential (Ecorr vs. SCE) for the corresponding sample. Ecorr represents the potential at which the net current density of the anode and cathode reactions is zero on the polarization curve.65  In general, Ecorr indicates the stable corrosion state of the potential without an applied current, which is also an important thermodynamic parameter for the corrosion tendency. However, Ecorr only reflects the corrosion trend, whereas the corrosion current density (icorr) determines the corrosion rate.66  Besides, the pitting potential (Epit vs. SCE), which is the lowest electrode potential value causing pitting on a passive layer, can be obtained at the inflection point.67 
FIGURE 11.

The potentiodynamic polarization curves of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in 0.9 wt% NaCl.

FIGURE 11.

The potentiodynamic polarization curves of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1 samples in 0.9 wt% NaCl.

Close modal

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 icorr 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 Ecorr values among the three samples. Moreover, all icorr values are less than 0.2 μA/cm2, which suggests good formation of the protective passive layer. In addition, the Epit 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.

Table 2.

Corrosion Parameters of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1

Corrosion Parameters of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1
Corrosion Parameters of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1

3.3.3 |  Electrochemical Impedance Spectroscopy Study

3.3.3.1 |  Electrochemical Impedance Spectroscopy
To further identify the variation in the corrosion resistance of the NiTi, NiTi–TiC-0.5, and NiTi–TiC-1 samples with TiC content, electrochemical impedance spectroscopy (EIS) was performed. Figure 12(a) depicts the Nyquist and Bode plots of the test samples in a 0.9 wt% NaCl solution at 25°C. As shown in Figure 12(a), the Nyquist plots of all samples display capacitive arc characteristics, which are consistent with the typical behavior of NiTi alloys. According to previous research, the diameter of the semicircle can be used to measure the corrosion resistance and stability of alloys.68  With the addition of TiC, the arc of the capacitance loop for the samples increases, indicating the best corrosion resistance of the NiTi-TiC-1 sample. This result agrees with the analysis of the polarization study.
FIGURE 12.

EIS results of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1: (a) Nyquist plots and (b) Bode plots.

FIGURE 12.

EIS results of NiTi, NiTi-TiC-0.5, and NiTi-TiC-1: (a) Nyquist plots and (b) Bode plots.

Close modal

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 

A good fit to the EIS data for these samples was obtained using an electrical equivalent circuit with two time constants, as shown in Figure 13. The corresponding fit curves are plotted in Figure 12, and the fitting parameters are summarized in Table 3. This model presents a NiTi|oxide layer|solution system, which has been used to study the corrosion behavior of NiTi alloys with defects in previous studies. Based on Chembath’s report,3  the first time constant, observed at high frequencies, was associated with an inner oxide layer (Rb, CPEb), whereas the second time constant, detected at low frequencies, was attributed to an outer layer (Rp, CPEp) with keyhole pores on the surface. Here, Rs represents the solution resistance, Rp and Rb denote the additional resistance of the solution inside the pores and the passive layer resistance, respectively, CPEf stands for the capacitance of the pore wall, and CPEb corresponds to the capacitance of the passive layer.71 
FIGURE 13.

Equivalent circuit used to fit EIS data.

FIGURE 13.

Equivalent circuit used to fit EIS data.

Close modal
Table 3.

Fitted Values of the Parameters of Different NiTi Samples from the EIS Spectra

Fitted Values of the Parameters of Different NiTi Samples from the EIS Spectra
Fitted Values of the Parameters of Different NiTi Samples from the EIS Spectra
Owing to the microscopic roughness of the surfaces of all specimens, the capacitance in the proposed circuit is not ideal. Therefore, the constant phase element (CPE), which is related to the angular frequency of the excitation signal, was used to represent the double-layer capacitance of the NiTi/electrolyte interface. Herein, it can be expressed as follows72 
where Q0 is the CPE constant, j denotes the symbol of an imaginary number, ω is the angular frequency, and n represents the CPE component. In general, n ranges from 0.5 to 1, and j2 = –1. When n = 1, the CPE is equivalent to the pure capacitance.73  However, resulting from the dispersion effect, the n value typically reduces, thus deviating from the pure capacitance.

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

The reciprocal of the passive layer capacitance (Cb) and the thickness of the passive film are in the direct ratio.68  To further investigate the impact of TiC addition on passive layer formation, the Cb values for the SLM-fabricated samples are listed in Table 3, which are calculated as follows65 

Hence, the lower Cb 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.

Trade name.

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).

1.
Kaya
M.
,
Çakmak
Ö.
,
Gülenç
B.
,
Atli
K.C.
,
Mater. Res. Bull.
95
(
2017
):
p
.
243
247
.
2.
Nargatti
K.
,
Ahankari
S.
,
J. Intell. Mater. Syst. Struct.
33
(
2022
):
p
.
503
531
.
3.
Chembath
M.
,
Balaraju
J.N.
,
Sujata
M.
,
Mater. Sci. Eng. C
56
(
2015
):
p
.
417
425
.
4.
Stergioudi
F.
,
Vogiatzis
C.A.
,
Pavlidou
E.
,
Skolianos
S.
,
Michailidis
N.
,
Smart Mater. Struct.
25
(
2106
):
p
.
article id 095024
.
5.
Elahinia
M.H.
,
Hashemi
M.
,
Tabesh
M.
,
Bhaduri
S.B.
,
Prog. Mater. Sci.
57
(
2012
):
p
.
911
946
.
6.
Zhao
Y.Z.
,
Guo
K.
,
Li
J.F.
,
Sun
J.
,
Int. J. Adv. Manufac. Technol.
116
(
2021
):
p
.
1913
1923
.
7.
Safavi
M.S.
,
Bordbar-Khiabani
A.
,
Khalil-Allafi
J.
,
Mozafari
M.
,
Visai
L.
,
J. Manufac. Mater. Process.
6, 3
(
2022
):
p.
65
.
8.
Alagha
A.N.
,
Hussain
S.
,
Zaki
W.
,
Mater. Des.
204
(
2021
):
article id 109654
.
9.
Liu
T.
,
Wu
Z.G.
,
Zhou
W.
,
Zhong
M.T.
,
Lin
J.
,
Yang
Y.
,
J. Alloys Compd.
933
(
2023
):
article id 167694
.
10.
Zhang
L.
,
Ren
D.
,
Ji
H.
,
Ma
A.
,
Daniel
E.F.
,
Li
S.
,
Jin
W.
,
Zheng
Y.
,
npj Mater. Degrad.
6
(
2022
):
article id 79
.
11.
Chen
F.
,
Lu
J.Q.
,
Liu
Y.
,
Zhang
H.F.
,
Zhang
C.
,
Shen
Q.
,
Adv. Eng. Mater.
25
(
2023
):
article id 2200504
.
12.
Xue
L.
,
Atli
K.C.
,
Zhang
C.
,
Hite
N.
,
Srivastava
A.
,
Leff
A.C.
,
Wilson
A.A.
,
Sharar
D.J.
,
Elwany
A.
,
Arroyave
R.
,
Acta Mater.
229
(
2022
):
article id 117781
.
13.
Kubásová
K.
,
Drátovská
V.
,
Losertová
M.
,
Salvetr
P.
,
Kopelent
M.
,
Korínek
F.
,
Havlas
V.
,
Dzugan
J.
,
Daniel
M.A.
,
Materials
17, 6
(
2024
):
p.
1248
.
14.
Khoo
Z.X.
,
Liu
Y.
,
An
J.
,
Chua
C.K.
,
Shen
Y.F.
,
Kuo
C.N.
,
Materials
11, 4
(
2018
):
p.
519
.
15.
Lu
H.Z.
,
Liu
L.H.
,
Yang
C.
,
Luo
X.
,
Song
C.H.
,
Wang
Z.
,
Wang
J.
,
Su
Y.D.
,
Ding
Y.F.
,
Zhang
L.C.
,
Li
Y.Y.
,
J. Mater. Sci. Technol.
101
(
2022
):
p
.
205
216
.
16.
Wang
X.B.
,
Yu
J.Y.
,
Liu
J.W.
,
Chen
L.G.
,
Yang
Q.
,
Wei
H.L.
,
Sun
J.
,
Wang
Z.C.
,
Zhang
Z.H.
,
Zhao
G.Q.
,
Addit. Manufac.
36
(
2020
):
article id 101545
.
17.
Zhang
Y.T.
,
Attarilar
S.
,
Wang
L.Q.
,
Lu
W.J.
,
Yang
J.L.
,
Fu
Y.F.
,
Int. J. Bioprinting
7
(
2021
):
p
.
15
42
.
18.
Li
X.
,
Hao
S.J.
,
Du
B.P.
,
Feng
B.
,
Li
H.H.
,
Qiu
P.
,
Huang
B.M.
,
Cui
L.S.
,
Yang
Y.
,
Met. Mater. Int.
29
(
2023
):
p
.
1510
1521
.
19.
Qiu
C.X.
,
Zhang
Y.C.
,
Qi
J.
,
Li
H.
,
Smart Struct. Syst.
21
(
2018
):
p
.
479
491
.
20.
Qiu
C.X.
,
Jiang
T.Y.
,
Liu
J.W.
,
Du
X.L.
,
J. Construct. Steel Res.
197
(
2022
):
article id 107480
.
21.
Srivastava
A.K.
,
Snapper
D.M.
,
Zheng
J.W.
,
Yildrim
B.S.
,
Srivastava
S.
,
Wood
S.C.
,
J. Immunotoxicol.
19
(
2022
):
p
.
61
73
.
22.
Yang
C.
,
Lu
H.Z.
,
Ma
H.W.
,
Cai
W.S.
,
Acta Metall. Sin.
59
(
2023
):
p
.
55
74
.
23.
Ou
S.F.
,
Peng
B.Y.
,
Chen
Y.C.
,
Tsai
M.H.
,
Metals
8, 5
(
2018
):
p.
342
.
24.
Qin
Q.H.
,
Peng
H.B.
,
Fan
Q.C.
,
Zhang
L.H.
,
Wen
Y.H.
,
J. Alloys Compd.
739
(
2018
):
p
.
873
881
.
25.
Li
B.
,
Zheng
L.
,
Chen
X.
,
Zhang
H.
,
Mater. Chem. Phys.
295
(
2023
):
article id 127114
.
26.
Lia
B.
,
Zheng
L.
,
Zhanga
H.
,
Mater. Charact.
208
(
2024
):
article id 113649
.
27.
Li
B.
,
Zhang
F.
,
Zheng
L.
,
Zhang
H.
,
Mater. Today Commun.
31
(
2022
):
article id 103556
.
28.
Pattar
N.
,
Patil
S.F.
,
Adv. Compos. Hybrid Mater.
2
(
2019
):
p
.
571
585
.
29.
Lu
B.W.
,
Cui
X.F.
,
Jin
G.
,
Dong
M.L.
,
Fang
Y.C.
,
Wen
X.
,
Ma
W.Y.
,
Opt. Laser Technol.
129
(
2020
):
article id 106290
.
30.
Jiang
H.J.
,
Cao
S.
,
Ke
C.B.
,
Ma
X.
,
Zhang
X.P.
,
Mater. Lett.
100
(
2013
):
p
.
74
77
.
31.
Farvizi
M.
,
Bahamirian
M.
,
Faraji
A.
,
Kim
H.S.
,
Met. Mater. Int.
30
(
2024
):
p.
843
856
.
32.
Promakhov
V.
,
Matveev
A.
,
Schulz
N.
,
Dronov
P.
,
Zhukov
A.
,
Vorozhtsov
A.
,
Materials
15, 15
(
2022
):
p.
5327
.
33.
Farvizi
M.
,
Ghanbariha
M.
,
Faraji
A.
,
Met. Mater. Int.
27
(
2021
):
p
.
5407
5424
.
34.
Sharma
A.K.
,
Singh
R.
,
Tiwari
A.K.
,
Sharma
A.K.
,
J. Mater. Eng. Perform.
31
(
2022
):
p
.
2934
2945
.
35.
Ma
C.L.
,
Ge
Q.
,
Yuan
L.H.
,
Gu
D.D.
,
Dai
D.H.
,
Setchi
R.
,
Wu
M.P.
,
Liu
Y.
,
Li
D.Y.
,
Ma
S.
,
Peng
X.
,
Fang
Z.
,
Compos. Part B
250
(
2023
):
article id 110457
.
36.
Zhang
D.Z.
,
Li
Y.Z.
,
Wang
H.
,
Cong
W.L.
,
Mater. Sci. Eng. A
809
(
2021
):
article id 140976
.
37.
Chen
F.
,
Guo
W.
,
Li
J.
,
Liu
Y.
,
Zhang
C.
,
Lu
J.Q.
,
Huang
Z.F.
,
Shen
Q.
,
J. Mater. Eng. Perform.
33
(
2024
):
p.
3899
3911
.
38.
Gu
D.D.
,
Ma
C.L.
,
Appl. Surf. Sci.
441
(
2018
):
p
.
862
870
.
39.
Krooss
P.
,
Lauhoff
C.
,
Gustmann
T.
,
Gemming
T.
,
Sobrero
C.
,
Ewald
F.
,
Brenne
F.
,
Arold
T.
,
Nematolahi
M.
,
Elahinia
M.
,
Thielsch
J.
,
Hufenbach
J.
,
Niendorf
T.
,
Shape Memory Superelasticity
8
(
2022
):
p
.
452
462
.
40.
Ma
C.
,
Gu
D.
,
Dai
D.
,
Xia
M.
,
Chen
H.
,
Mater. Charact.
143
(
2018
):
p
.
191
196
.
41.
Martin
J.H.
,
Yahata
B.D.
,
Hundley
J.M.
,
Mayer
J.A.
,
Schaedler
T.A.
,
Pollock
T.M.
,
Nature
549
(
2017
):
p
.
365
.
42.
Lu
B.W.
,
Cui
X.F.
,
Ma
W.Y.
,
Dong
M.L.
,
Fang
Y.C.
,
Wen
X.
,
Jin
G.
,
Zeng
D.C.
,
Addit. Manufac.
33
(
2020
):
article id 101150
.
43.
Elahinia
M.
,
Moghaddam
N.S.
,
Andani
M.T.
,
Amerinatanzi
A.
,
Bimber
B.A.
,
Hamilton
R.F.
,
Prog. Mater. Sci.
83
(
2016
):
p
.
630
663
.
44.
Zhang
C.
,
Xue
L.
,
Atli
K.C.
,
Arróyave
R.
,
Karaman
I.
,
Elwany
A.
,
J. Manufac. Sci. Eng. Trans. ASME
144, 9
(
2022
):
article id 091011
.
45.
Gu
D.D.
,
Wang
H.Q.
,
Zhang
G.Q.
,
Metall. Mater. Trans. A
45
(
2015
):
p
.
464
476
.
46.
Bermingham
M.J.
,
StJohn
D.H.
,
Krynen
J.
,
Tedman-Jones
S.
,
Dargusch
M.S.
,
Acta Mater.
168
(
2019
):
p
.
261
274
.
47.
Fu
J.
,
Hu
Z.H.
,
Song
X.
,
Zhai
W.
,
Long
Y.
,
Li
H.
,
Fu
M.W.
,
Opt. Laser Technol.
131
(
2020
):
article id 106374
.
48.
Fu
J.
,
Li
H.
,
Song
X.
,
Fu
M.W.
,
J. Mater. Sci. Technol.
122
(
2022
):
p
.
165
199
.
49.
Zhao
C.
,
Parab
N.D.
,
Li
X.X.
,
Fezzaa
K.
,
Tan
W.D.
,
Rollett
A.D.
,
Sun
T.
,
Science
370
(
2020
):
p
.
1080
.
50.
Zhu
J.N.
,
Ding
Z.Y.
,
Borisov
E.
,
Yao
X.Y.
,
Brouwer
J.C.
,
Popovich
A.
,
Hermans
M.
,
Popovich
V.
,
Virtual Phys. Prototyping
18
(
2023
):
article id e2246437
.
51.
Feng
X.
,
Sui
J.H.
,
Cai
W.
,
Liu
A.L.
,
Scr. Mater.
64
(
2011
):
p
.
824
827
.
52.
Wang
X.J.
,
Liu
J.Y.
,
Yang
L.J.
,
He
Y.
,
Wang
Y.
,
Appl. Surf. Sci.
448
(
2018
):
p
.
133
137
.
53.
Zhang
D.D.
,
Liu
H.L.
,
Sun
L.P.
,
Bai
F.
,
Wang
Y.
,
Wang
J.G.
,
Crystals
7, 7
(
2017
):
p.
205
.
54.
Liepack
H.
,
Bartsch
K.
,
Arnold
B.
,
Bauer
H.D.
,
Liu
X.
,
Knupfer
M.
,
Leonhardt
A.
,
Diamond Relat. Mater.
13
(
2004
):
p
.
106
110
.
55.
Ma
C.L.
,
Gu
D.D.
,
Dai
D.H.
,
Yu
G.Q.
,
Xia
M.J.
,
Chen
H.Y.
,
Crystengcomm
19
(
2017
):
p
.
1089
1099
.
56.
Wang
C.
,
Tan
X.P.
,
Du
Z.
,
Chandra
S.
,
Sun
Z.
,
Lim
C.W.J.
,
Tor
S.B.
,
Lim
C.S.
,
Wong
C.H.
,
J. Mater. Process. Technol.
271
(
2019
):
p
.
152
161
.
57.
Ye
D.
,
Li
S.F.
,
Misra
R.D.K.
,
Zheng
R.
,
Yang
Y.F.
,
Addit. Manufac.
47
(
2021
):
article id 102344
.
58.
Yang
Y.
,
Zhan
J.
,
Sun
Z.
,
Wang
H.
,
Lin
J.
,
Liu
Y.
,
Zhang
L.
,
J. Alloys Compd.
804
(
2019
):
p
.
220
229
.
59.
Dai
K.J.
,
He
X.
,
Zhang
W.
,
Kong
D.C.
,
Guo
R.
,
Hu
M.L.
,
He
K.T.
,
Dong
C.F.
,
Mater. Des.
235
(
2023
):
article id 112386
.
60.
Yang
Y.
,
Zhan
J.B.
,
Sun
Z.Z.
,
Wang
H.L.
,
Lin
J.X.
,
Liu
Y.J.
,
Zhang
L.C.
,
J. Alloys Compd.
804
(
2019
):
p
.
220
229
.
61.
Sam
J.
,
Franco
B.
,
Ma
J.
,
Karaman
I.
,
Elwany
A.
,
Mabe
J.H.
,
Scr. Mater.
146
(
2018
):
p
.
164
168
.
62.
Mari
D.
,
Dunand
D.C.
,
Metall. Mater. Trans. A
26
(
1995
):
p
.
2833
2847
.
63.
Kassab
E.
,
Frotscher
M.
,
Eggeler
G.
,
Gome
J.
,
Mater. Today Commun.
33
(
2022
):
article id 104401
.
64.
Liu
M.
,
Zhu
J.N.
,
Popovich
V.A.
,
Borisov
E.
,
Mol
J.M.C.
,
J. Mater. Res. Technol.
23
(
2023
):
p
.
2991
3006
.
65.
Barros
C.D.D.
,
Gomes
J.
,
J. Mech. Behav. Biomed. Mater.
112
(
2020
):
article id 104015
.
66.
Chmielewska
A.
,
Dobkowska
A.
,
Kijenska-Gawronska
E.
,
Jakubczak
M.
,
Krawczynska
A.
,
Choinska
E.
,
Jastrzebska
A.
,
Dean
D.
,
Wysocki
B.
,
Swieszkowski
W.
,
Int. J. Mol. Sci.
22, 24
(
2021
):
p.
13209
.
67.
Safavi
M.S.
,
Khalil-Allafi
J.
,
Ahadzadeh
I.
,
Walsh
F.C.
,
Visai
L.
,
Surf. Coat. Technol.
470
(
2023
):
article id 129822
.
68.
Qiu
P.
,
Gao
P.P.
,
Wang
S.Y.
,
Li
Z.H.
,
Yang
Y.
,
Zhang
Q.Q.
,
Xiong
Z.W.
,
Hao
S.J.
,
Corros. Sci.
175
(
2020
):
article id 108891
.
69.
Sun
X.T.
,
Kang
Z.X.
,
Zhang
X.L.
,
Jiang
H.J.
,
Guan
R.F.
,
Zhang
X.P.
,
Electrochim. Acta
56
(
2011
):
p
.
6389
6396
.
70.
Liu
M.
,
Cheng
X.Q.
,
Li
X.G.
,
Lu
T.J.
,
J. Electroanal. Chem.
803
(
2017
):
p
.
40
50
.
71.
Bolat
G.
,
Mareci
D.
,
Iacoban
S.
,
Cimpoesu
N.
,
Munteanu
C.
,
J. Spectrosc.
2013
(
2013
).
72.
Shanaghi
A.
,
Chu
P.K.
,
J. Alloys Compd.
790
(
2019
):
p
.
1067
1075
.
73.
Wu
J.L.
,
Deng
Y.L.
,
Guo
X.B.
,
Zhou
S.Y.
,
Mater. Today Commun.
39
(
2024
):
article id 108636
.

Supplementary data