Impact of NiTi Shape Memory Alloy Substrate Phase Transitions Induced by Extreme Temperature Variations on the Tribological Properties of TiN Thin Films

Abstract

NiTi alloys and thin film/NiTi composites are extensively utilized in frictional environments, particularly those experiencing extreme temperature fluctuations. Current studies mainly focus on preparing wear-resistant films on NiTi alloy surfaces but neglect the potential impact of temperature-induced phase transitions in the NiTi substrate on thin films’ performance. This study examines the effect of NiTi alloy phase transitions, induced by extreme temperature variations, on the tribological properties of TiN thin films on NiTi substrates. TiN films (1 μm thick) were deposited on NiTi alloy surfaces using magnetron sputtering technology. The transition of the main phase in the NiTi substrate between the R phase and the B19′ phase was achieved via liquid nitrogen cooling (−196 °C) and water bath heating (90 °C). XRD, EDS, SEM, and tribological tests analyzed the phase structure, elemental composition, micromorphology, and tribological behavior. Fatigue wear was identified as the predominant wear mechanism for the TiN films, with minor contributions from oxidative and abrasive wear. Phase transition from the R phase to the B19′ phase in the NiTi substrate induced by temperature change couls reduce the wear rate of the TiN film by up to 41.97% and decrease the friction coefficient from about 0.45 to about 0.25. Furthermore, the shape memory effect of the NiTi alloy substrate, caused by B19′ → B2 phase transition, resulted in the recovery of the TiN thin film wear track depth from 920 nm to 550 nm, manifesting a “self-healing” phenomenon. The results in this study are important and necessary for the provision of thin film/NiTi composites in frictional environments.

1. Introduction

With the rapid advancement of modern industrial technology, the wear resistance of materials has become a core consideration in mechanical design and manufacturing. The demand for wear-resistant materials is particularly pronounced in sectors such as biomedical engineering, aerospace, energy chemistry, and mechatronics. NiTi shape memory alloys, characterized by their unique phase transition between the B2 phase (austenite, body-centered cubic structure) and the B19′ phase (martensite, monoclinic structure), exhibit exceptional shape memory effects and superelasticity. These properties make NiTi alloys highly promising intelligent materials, driving their widespread adoption in these fields [1].

In biomedical applications, orthodontic arch wires [2,3] and stents [4,5] made from NiTi alloys are subjected to friction and wear when in contact with human tissues. Excessive release of toxic Ni ions can result from excessive wear. Similarly, in engineering, bearings [6,7], reinforcing bars [8], springs [9], and other components [10,11] require NiTi alloys with high wear resistance to ensure reliable service in the long term.

In comparison to traditional approaches to improve the preparation process of NiTi alloys [12,13], surface modification techniques such as depositing highly wear-resistant films on NiTi alloy surfaces to form film/NiTi alloy substrate composites have proven to be effective for boosting abrasion resistance and prolonging the lifespan of these materials. Wang, S. et al. and Wang, Q. et al. successfully deposited TiN films on NiTi alloy substrates [14,15]. Huang, S.Y. et al. investigated the deposition of diamond-like carbon films on NiTi arch wires [16]. Dai, D. et al. explored the application of graphene nanocoatings on NiTi alloy surfaces [17]. Extensive research has indicated that these thin films and coatings can notably enhance the biocompatibility and wear resistance of NiTi alloy-based implants. By effectively reducing the wear against human tissues, they significantly improve the practicality and durability of these implants in a variety of biomedical engineering applications. Furthermore, Choudhury, D. et al. and Zhou, Y. et al. deposited PDA/PTFE + graphite particle coatings [18] and hydrogenated diamond-like carbon coatings [19] on the surfaces of 60NiTi engineering components, respectively. The wear resistance of NiTi alloy devices can thus be significantly enhanced, providing an effective solution for their application under high-load and high-wear conditions. This approach ensures long-term stability and reliability in complex engineering and industrial environments.

When utilized as mechanical components in complex and variable environments, NiTi alloy composites are often subjected to extreme temperature variations, such as those encountered in space, desert, or high-altitude regions. These fluctuations are significantly greater than those in the relatively stable temperature conditions within the human body, where implantable devices are typically used. The temperature responsiveness of phase transitions in NiTi alloys is extremely significant, transitioning to the B2 phase at elevated temperatures and transforming into the B19′ phase at lower temperatures, with some alloys also exhibiting an intermediate R phase [20,21]. Such extreme temperature changes, often spanning several hundred degrees Celsius, can induce phase transitions in NiTi alloys, thereby altering the properties of the NiTi substrate within the composite material. The wear-resistant coating bonded to the NiTi substrate forms an integrated structure. During the phase transition of the NiTi substrate, the rearrangement and nucleation of Ni and Ti atoms [22] inevitably induce changes in the microstructure at the film/substrate interface and within the film itself. Previous research conducted by our team demonstrated that phase transitions in the NiTi substrate substantially alter the TiN thin film stress state and the interfacial adhesion strength of the TiN thin film/NiTi substrate composite [23], thereby modifying the mechanical characteristics associated with the TiN film/NiTi alloy composite. Considering that the tribological performance of the film is a critical mechanical property, current research on the tribological performance of film/NiTi alloy composites primarily focuses on the preparation of films with high wear resistance on the surfaces of NiTi alloys to adapt to tribological environments [14,15,16,17,18,19]. However, these studies often overlook the potential impact of phase transition behavior in the NiTi substrate under extreme environmental conditions on the tribological behavior of the surface film.

In this research, TiN films, which are representative of high-wear-resistant hard ceramic films [24,25,26], were deposited on the NiTi alloy surface using unbalanced magnetron sputtering technology [27]. Going beyond the conventional focus on enhancing wear resistance under stable conditions, our research uniquely simulates the extreme temperature fluctuations encountered in demanding environments such as space, desert, or high-altitude regions. By inducing phase transitions in the NiTi alloy through liquid-nitrogen cryogenic cooling and water bath heating and conducting a series of friction and wear experiments, we delve into the unexplored territory of how these transitions affect the tribological performance of the TiN thin film/NiTi alloy composites. This pioneering approach not only addresses the critical gap in understanding the impact of phase transition behavior under extreme conditions but also provides a novel theoretical foundation and practical insights for the surface modification of NiTi alloys. Our findings are expected to have significant implications for the development of more durable and reliable NiTi alloy-based components, particularly in scenarios where materials are subjected to harsh and variable temperature conditions.

2. Materials and Methods

2.1. Material Preparation

The NiTi alloy (50.7 at.% Ni, 49.3 at.% Ti) that served as the primary focus of this study was provided by Suzhou Haichuan Rare Metal Products Co., Ltd. (Suzhou, China). As 304 stainless steel is often used as the metal substrate for wear-resistant coatings and is less prone to phase transition, 304 stainless steel provided by Hebei Qinbang New Material Technology Co., Ltd. (Shijiazhuang, China), was employed as the substrate material of the control sample to eliminate the influence of a single temperature factor on the tribological properties of the TiN films in this study. They were cut into square specimens measuring 20 × 20 mm², each 2 mm in thickness. Then, the specimens were subjected to fine polishing using a grinder polisher (Buehler AutoMet 250, Lake Bluff, IL, USA) with SiC sandpaper ranging from 240 to 5000 grit until a mirror-like surface finish was achieved. After polishing, the specimens were sequentially immersed in acetone, anhydrous ethanol, and deionized water for 5 min of ultrasonic cleaning to remove residual surface contaminants. The cleaned sheets were then air-dried and stored for subsequent use. The NiTi alloy used in this study was identical to that employed in our previous research. The phase transition temperatures, including the martensitic start (Ms), martensitic finish (Mf), austenitic start (As), austenitic finish (Af), R-phase start (Rs), and R-phase finish (Rf) temperatures, were determined using differential scanning calorimetry (DSC). According to the DSC curves in the literature [23],

Table 1. Phase transition temperatures of the NiTi alloy.

	Ms	Mf	As	Af	Rs	Rf
Temperature	25 °C	−3 °C	45 °C	68 °C	54 °C	38 °C

enumerates the specific values of the phase transition temperatures.

The TiN thin film was deposited using an unbalanced magnetron sputtering apparatus (UBMS450) provided by Sky Technology Development Co., Ltd. (Shenyang, China). High-purity titanium targets (with a purity exceeding 99.9%) served as the sputtering material. Meanwhile, prefabricated NiTi alloy sheets, 304 stainless steel sheets, and single-crystal silicon wafers were used as substrates. During the preparation of the TiN thin film, the pressure inside the chamber was lowered to 2 × 10⁻³ Pa, after which high-purity argon gas was introduced. After the power supply was connected, plasma pre-treatment was conducted on the target and substrate materials for 10 and 20 min, respectively, to eliminate the oxides and contaminants on the surface. Subsequently, the TiN thin film was deposited on the substrate materials under the parameters specified in

Table 2. Deposition parameters of TiN films.

	Ar:N2 Ratio	Power Supply	Substrate Bias/V	Target–Substrate Distance/mm	Working Pressure/Pa	Working Time/min
Ti transition layer	40:0	DC-2A	−50	90	0.53	2
TiN layer	40:10	DC-2A	−50	90	0.61	25

, with a Ti interlayer incorporated to boost the bond strength between the TiN film and the substrate materials [28]. Three parallel samples were prepared for each material. Single-crystal silicon wafer samples were utilized to characterize the morphological details of the deposited TiN thin films.

2.2. Characterization Methods

X-ray diffraction (XRD; Malvern Panalytical Empyrean, Almelo, The Netherlands) was utilized to characterize the phase composition of the film/substrate composites. The experiment utilized a Cu Kα1 radiation source. The samples were scanned from 20° to 80°, with the scanning rate set to 0.33°/s. The XRD shooting method was symmetrical, and the incident angle and reflection angle of the X-rays were equal. Energy-dispersive X-ray spectroscopy (EDS; OXFROD ULTIM Max65, Oxford, UK) was employed to analyze the elemental composition and proportions within the microregions of the samples. Scanning electron microscopy (SEM; Thermo Scientific Apreo 2C, Waltham, MA, USA) was utilized to observe the micromorphologies and structures of the samples. A tribometer (CSEM, Peseux, Switzerland) was used to characterize the tribological performance of the fabricated composites. A profilometer (Ambios XP-2, Bozeman, MT, USA) was utilized to determine the wear scar profile curves after abrasion. The wear rate of the materials was subsequently calculated based on these data.

2.3. Experimental Methods

Liquid nitrogen cryogenic treatment (−196 °C) and water bath heating (90 °C) were employed to simulate cryogenic–thermal cycling under extreme environmental conditions, enabling the evaluation of the tribological performance of thin film/substrate composite materials under such conditions. Specifically, the prepared thin film/substrate composite samples were first subjected to a 5 min liquid nitrogen cryogenic treatment, followed by a return to room temperature. This was followed by a 5 min water bath heating treatment, after which the samples were returned to room temperature. This complete cryogenic–thermal treatment process was defined as one cryogenic–thermal cycle. If a sample had undergone only the liquid-nitrogen cryogenic treatment without water bath heating, it was considered to have undergone 0.5 cryogenic–thermal cycles. In the experimental design of this study, the same sample was subjected to a uniform standard of tribological tests at eight distinct treatment stages: the initial state post fabrication, after one deep-cryogenic treatment, following one cryogenic–thermal cycle with an additional deep-cryogenic treatment, after ten cryogenic–thermal cycles, subsequent to ten cryogenic–thermal cycles with an additional deep-cryogenic treatment, following twenty cryogenic–thermal cycles, and after twenty cryogenic–thermal cycles with an additional deep-cryogenic treatment. The specific process is illustrated in Figure 1, the sample codes are mentioned in the subsequent text, and their corresponding processing information is listed in

Table 3. Sample coding and information.

Sample Coding	Sample Information
TiN/substrate-0	The sample underwent friction in its initial state.
TiN/substrate-0.5	The sample underwent friction after one deep-cryogenic treatment.
TiN/substrate-1	The sample underwent friction after one cryogenic–thermal cycle treatment.
TiN/substrate-1.5	The sample underwent friction after one cryogenic–thermal cycle treatment with an additional deep-cryogenic treatment.
TiN/substrate-10	The sample underwent friction after ten cryogenic–thermal cycle treatments.
TiN/substrate-10.5	The sample underwent friction after ten cryogenic–thermal cycle treatments with an additional deep-cryogenic treatment.
TiN/substrate-20	The sample underwent friction after twenty cryogenic–thermal cycle treatments.
TiN/substrate-20.5	The sample underwent friction after twenty cryogenic–thermal cycle treatments with an additional deep-cryogenic treatment.
TiN/substrate-0(0.5)	The sample underwent friction in its initial state and then underwent one deep-cryogenic treatment.
TiN/substrate-0(1)	The sample underwent friction in its initial state and then underwent one cryogenic–thermal cycle treatment.

.

In the tribological wear experiments, a 6 mm diameter Al₂O₃ ball served as the counterface to apply reciprocating friction on the TiN film surface for 1000 cycles. During the friction process, a constant normal load (F) of 2 N was applied with a wear scar length (L) of 6 mm, resulting in a total sliding distance (s) of 12 m for the ball. The ball’s sliding speed was maintained at 12 mm/s. The approximate ranges of these parameters for the tribological wear experiments were taken from Reference [14] and ultimately determined based on the experimental conditions and objectives of this study. After each friction and wear test, a profilometer was used to obtain five equidistant measurements of the wear scar and acquire detailed contour information about the scar. From these data, the wear scar’s cross-sectional area was calculated, and the average cross-sectional area ($\overline{S}$) was determined. This average value, when multiplied by the wear scar length, provided the wear volume of a single track, which was subsequently used for the quantitative analysis of the wear rate. The specific calculation method for the wear rate is expressed in Equation (1):

$$WearRate={\displaystyle \frac{\overline{S}\cdot L}{F\cdot s}}$$

(1)

3. Results and Discussion

3.1. Phase Transitions of NiTi Substrate

As shown in the phase transition temperatures in Table 1, plasma bombardment during the film preparation process elevated the temperature of the NiTi substrate [29,30], resulting in its transition to the austenitic phase. Upon cooling to room temperature, the NiTi substrate transformed into the R phase. In the TiN/NiTi alloy composites, martensitic phase transition occurred when the NiTi substrate was subjected to liquid-nitrogen cryogenic treatment. During the subsequent water bath heating treatment, the NiTi substrates underwent austenitic phase transition. Finally, when the samples were cooled to room temperature, the NiTi substrates transitioned from austenite back to the R phase. The XRD results presented in Section 3.3 further confirm this phase transition process. The NiTi alloy in this study demonstrated both the R phase and the martensitic phase at room temperature, thereby providing suitable conditions for exploring the differences in the frictional and wear behavior of TiN films on NiTi alloys under different phase states.

3.2. Thin Film Microstructure

Figure 2a shows the cross-sectional microstructure of a TiN thin film prepared on a silicon substrate using the deposition parameters outlined in Table 2. The thin film exhibits vertical growth relative to the substrate, forming columnar crystals with a thickness of 1 μm. Figure 2b further reveals that these columnar crystals are closely packed, resulting in a dense film surface. This densification is attributed to the substrate bias applied during film deposition, which significantly increased the film’s density [31,32]. Figure 2e includes the EDS line scanning (the subfigure ① in Figure 2e) corresponding to Figure 2a and the EDS surface scanning (the subfigure ② and ③ in Figure 2e) corresponding to Figure 2b. Both the EDS scanning results of the cross-section and the surface of TiN/Si show that the atomic ratios of Ti and N elements are almost equal, which is consistent with the characteristics of TiN thin films.

Although directly observing the cross-sectional morphology of films on metal substrates to determine their growth structure is challenging, inferences can be made through comparisons, as shown in Figure 2b–d. The surface morphology of the films on all three substrates exhibits a closely packed granular structure. The sample on the silicon substrate shows a higher degree of surface density compared to the other two metal substrates. This can be attributed to differences in the thermal expansion coefficients and initial surface conditions of the different substrates. However, overall, the surface morphologies of the films on the three substrates still show a high degree of similarity. It can thus be inferred that the TiN films also grow in the form of columnar crystals on the surfaces of the NiTi and 304 substrates.

3.3. Thin Film and Substrate Phase Composition

XRD was used to analyze the phase structural changes in TiN/NiTi alloy and TiN/304 stainless steel samples during a single cryogenic–thermal cycle, and the results are presented in Figure 3. In the initial state, the XRD pattern of the NiTi alloy substrate shows two distinct high-intensity R-phase diffraction peaks between 42° and 43° [33], indicating that the NiTi mainly exhibits the R phase at this time. Additionally, the diffraction peaks of the B19′ phase are located at 2θ = 39.3° and 2θ = 41.5°, while the diffraction peak of the B2 phase is located at 2θ = 77.8°. These findings indicate the presence of residual austenite and martensite phases in the NiTi alloy substrate, and the diffraction peak at 2θ = 43.3° confirms the existence of Ni₄Ti₃ precipitates. After cryogenic treatment, the diffraction peaks corresponding to the R and B2 phases in the XRD pattern of the NiTi sample disappeared, while the diffraction peak associated with the B19′ phase intensified significantly. The above observations indicate that the main phase of the NiTi substrate underwent a phase transition from the R phase with the P$\overline{3}$ space group to the B19′ phase with the P21/m space group. Subsequent water bath heating restored the diffraction pattern of the NiTi sample to its initial state, confirming the reverse phase transition process of the substrate from the B19′ phase to the B2 phase with the Pm3m space group and then back to the R phase.

Conversely, the XRD pattern of the 304 stainless steel substrate remained stable throughout the entire cryogenic–thermal cycle, indicating that no phase transition had occurred within the tested temperature range. In the XRD patterns of both samples, a low-intensity diffraction peak attributed to the Ti transition layer appeared at 2θ = 35.3°.

The TiN film exhibited a high-intensity diffraction peak at 2θ = 36.5°, indicating its preferred orientation along the (111) crystal plane. Additionally, a low-intensity diffraction peak corresponding to the (222) crystal plane of TiN was detected at 2θ = 77.8°. These diffraction peaks remained unchanged during the single cryogenic–thermal cycle, suggesting that the prepared TiN film maintained its phase stability under temperature variations. Furthermore, the data of full width at half maximum β and diffraction angle 2θ associated with the most intense (111) diffraction peak of the TiN film were extracted from the XRD spectra and are presented in

Table 4. Grain size of TiN thin films.

	β/°	2θ/°	Grain Size/nm
TiN/NiTi-0	0.46	36.50	18.09
TiN/NiTi-0.5	0.45	36.52	18.40
TiN/NiTi-1	0.45	36.50	18.50
TiN/304-0	0.27	36.59	31.14
TiN/304-0.5	0.26	36.49	31.44
TiN/304-1	0.26	36.46	31.24

. By substituting these values into the Scherrer formula, the grain sizes of the TiN films deposited on the NiTi and 304 substrates were calculated. The findings revealed that the grain size of the TiN film remained unaffected by temperature fluctuations and the phase transitions of the NiTi substrate. Additionally, it was observed that the TiN films prepared on the two distinct substrates using the identical preparation process exhibited disparate grain sizes.

3.4. Wear Mechanism of TiN Film

Under SEM at 1000× magnification, the initial wear scars generated by friction on the TiN/NiTi alloy and TiN/304 stainless steel substrates were observed, revealing distinct surface characteristics. As shown in Figure 4a,c, the wear scar edges on the TiN film/NiTi substrate appeared relatively blurred, and its wear scar width was significantly larger than that on the TiN film/304 substrate. Further analysis combining SEM images with the corresponding EDS images indicated that the surface distribution of Ti and N was remarkably homogeneous across both samples, with no significant difference in the elemental distribution observed in areas with and without wear scar. Additionally, the primary elements of the substrates (Ni or Fe) exhibited a uniform distribution and were detected in only a few areas. These results suggest that the TiN film did not penetrate during the friction and wear tests, with only this film experiencing direct abrasive interaction with the Al₂O₃ counterpart surface.

The distribution of O, depicted in Figure 4b,d, also appeared uniform and exhibited a sparse content. This observation indicates that no significant macroscopic oxidation wear was detected on the TiN film during the tests.

A TiN/NiTi alloy served as the primary focus of this study. We conducted a thorough microscopic analysis of the first wear scar on the surface, aiming to acquire a more profound insight into the wear mechanism of TiN films prepared using the process parameters outlined in Table 2. Figure 5a depicts a continuous fish scale-like flaking structure observed within the film’s wear scar, which is likely attributed to shallow microcracks formed under friction, eventually leading to flaking under repeated shear stress. Similarly, Figure 5a reveals a larger area of extensional cracks within the wear scar, resembling the cracks surrounding the flaky defects shown in Figure 5c. These cracks are initiated by cyclic contact stresses on the film under load [34]. As the stress cycles progressed, these cracks extended longitudinally, eventually leading to film delamination and the formation of flaky defects, as shown in Figure 5c (the subfigure ①, ② and ③). The EDS mapping results of defect ① are depicted in Figure 5d, which indicates the absence of Ti and N elements in the defect region while detecting the presence of Ni from the substrate and the enrichment of O elements. This finding suggests that the defect originated from delamination at the film–substrate interface. The initiation and propagation of the cracks and the film’s delamination from the substrate are characteristic indicators of fatigue damage in thin-film materials under friction and wear [35]. These observations indicate that the TiN film experienced significant fatigue wear during this study’s friction and wear processes.

Additionally, numerous strip-like particles were observed within the wear scar, which was likely wear debris formed due to the delamination and fracture of the thin film, resulting from fatigue microcracks under repeated shearing actions of the counterface. By employing EDS line scanning technology (Figure 5b), the elemental composition of these debris particles was analyzed, revealing a significantly higher oxygen content compared to other areas within the wear scar. Although no significant oxygen enrichment was observed within the wear scar in the previous macroscopic EDS results (Figure 4b), the EDS line scan results of the debris particles suggest that a certain degree of oxidation wear happened to the TiN film during friction.

According to the literature, the friction products of the TiN (111) crystal plane are primarily amorphous Ti oxides [36]. As the friction and wear experiments progressed, the previously generated debris particles inevitably became interposed between the film and counterface, undergoing relative sliding. In Figure 5a, slight wear marks along the sliding direction of the counterface were observed; however, no deep or distinct furrows, characteristic of abrasive wear, were formed. This outcome may be attributed to the lower hardness of the amorphous Ti oxides compared to that of the TiN film, leading to only minor abrasive wear of the TiN film during the tribological process.

3.5. Wear Rate of TiN Thin Film

Tribological tests were conducted using identical conditions on R-phase NiTi alloy substrates and 304 stainless steel substrates, both without film coating. The results in Figure 6 reveals that the NiTi substrate manifested superior wear resistance compared to the 304 substrate. This phenomenon is not only attributed to the difference in hardness between the two materials but also closely related to the unique ability of the NiTi alloy to undergo phase transition under external stress [37].

After the successful deposition of the TiN film, the wear resistance of these two film/substrate composite materials in their initial state was compared with that of bare metal substrates without film coverage. The experimental data in Figure 6 indicate that depositing the TiN film significantly reduced the materials’ wear rates, resulting in excellent wear resistance. Compared to the TiN/NiTi composite, the TiN/304 composite manifested a lower wear rate, in agreement with the differences between the wear scar widths shown in Figure 4. This finding highlights that, even under identical film preparation process parameters, the different characteristics of substrate materials can markedly impact the wear resistance of the film.

Figure 7a illustrates the trend in wear rate variation for three parallel samples of the TiN/NiTi composite during the cryogenic–thermal cycling process. Initially, the NiTi substrate was in the R phase, and the initial samples’ wear rates were determined based on the first set of friction and wear tests conducted at this stage. After the deep-cryogenic treatment, the NiTi substrate transitioned into the martensitic phase, resulting in a significant reduction in the wear rate observed during the second friction and wear test conducted on the TiN film surface.

Subsequently, after the water bath heat treatment, the NiTi substrate underwent a continuous phase transition from martensite to austenite and back to the R phase, a process which caused the TiN film wear rate in the third friction and wear test to increase again. In multiple subsequent cryogenic–thermal cycles, the TiN film wear rate continued to follow this pattern. In contrast, Figure 7b shows that the three parallel TiN/304 composite samples’ wear rates fluctuated within a specific range as the cryogenic–thermal cycling treatment progressed, without exhibiting a clear trend. This observation suggests that temperature variations within a certain range generally do not exert a notable influence on the wear performances of TiN films on substrates that do not undergo phase transition. However, for the NiTi substrate, the temperature-induced phase transitions significantly altered the tribological characteristics of the surface TiN film. For example, when the NiTi substrate transitioned from the R phase to the martensitic phase, the wear rate of the surface TiN film markedly reduced.

In the experiments conducted on the TiN/NiTi alloy and TiN/304 stainless steel, representative samples from each material were selected, and their wear scar profiles during the first cryogenic–thermal cycle were analyzed. As shown in Figure 8a, the initial sample produced the first wear scar with a depth of approximately 920 nm after the friction and wear test. After the deep-cryogenic treatment, the NiTi substrate underwent a martensitic phase transition. A notable decrease in the wear scar depth to approximately 550 nm was observed on the TiN/NiTi composite surface in the subsequent friction and wear test.

Subsequently, water bath heating restored the NiTi substrate to the austenite and then back to the R phase, producing a third wear scar during friction, similar in depth to the first one. In contrast, Figure 8b demonstrates that the wear scar profiles from the three friction and wear tests conducted on the TiN/304 stainless steel sample under identical treatment conditions showed minimal variation, with no significant differences in wear scar depth or width observed.

This result indicates that the phase transition induced by temperature changes in the NiTi alloy substrate significantly affects the frictional and wear characteristics of the surface TiN film. Specifically, the wear scar depth produced by the TiN film on the martensitic NiTi alloy surface during friction exhibited a substantial decrease relative to that of the NiTi alloy in the R phase, despite the absence of any notable alteration in scar width.

A significant cyclic pattern of variation was observed when evaluating the impact of phase transition phenomena experienced by NiTi substrates during the cryogenic thermal cycling treatment on the TiN/NiTi composites’ resistance to wear. This pattern was attributed to the alternating phase transitions of the NiTi substrate, which induced corresponding changes in the stress state of the surface TiN film. Specifically, the deep-cryogenic treatment caused the R-phase NiTi substrate to transition to martensite, shifting the stress state of the surface TiN film towards compressive stress. Conversely, during the water bath heating treatment, the martensitic NiTi substrate transitioned to the austenite and then reverted to the R phase, resulting in a shift in the stress state of the TiN film from compressive to tensile stress [23]. Studies have shown that the stress state of a film critically influences its tribological properties, with films under optimal compressive stress conditions typically exhibiting superior wear resistance [38,39]. The TiN film on the surface of the martensitic NiTi substrate was observed to exhibit compressive stress, which promoted closer packing of columnar crystals, thereby enhancing density and hardness [40]. Moreover, compressive stress inhibits the initiation and propagation of fatigue cracks in the TiN film during friction and wear processes, thereby reducing material wear.

3.6. Coefficient of Friction of Composites

During the cryogenic–thermal cycling treatment of composite materials, notable changes in the friction coefficients of the TiN/NiTi composite were monitored at different stages, while the friction coefficient of TiN/304 remained relatively stable throughout the process, according to Figure 9. Initially, the friction coefficient of the TiN/NiTi composite was stable at roughly 0.45. However, when the NiTi substrate underwent the deep-cryogenic treatment and transitioned into the martensitic phase, the friction coefficient decreased significantly to roughly 0.25. Furthermore, when the NiTi substrate was restored to the R phase through the water bath heating treatment, the friction coefficient returned to the initial level of 0.45. The TiN film on the martensitic NiTi alloy, which demonstrated a lower wear rate, also exhibited a reduced friction coefficient. This observation may be attributed to the effects of phase transition on the mechanical compatibility between the NiTi substrate and the surface TiN film, thereby affecting the friction coefficient of TiN/NiTi during friction and wear processes.

3.7. Evolution of Wear Scar Characteristics on TiN/NiTi Alloy

This section examines the evolutions of the first wear scar on the sample during the initial cryogenic–thermal cycles. As shown in Figure 10a, after the NiTi substrate underwent a martensitic phase transition from the R phase under temperature induction, the wear rate of the first wear scar on the TiN film exhibited a minor fluctuation upon remeasurement. However, the change was not significant compared to the previous measurement, which was likely due to errors in the measurement process. Furthermore, after the sample was subjected to water bath heating, the wear rate of the identical wear scar was remeasured and monitored to decrease significantly relative to the first two measurements. In contrast, Figure 10b illustrates that the wear rate of the first wear scar on the TiN/304 stainless steel sample remained relatively consistent across the three measurements, with only slight fluctuations observed. The analysis of this phenomenon suggests that, although the normal load applied during the friction and wear tests was relatively small and the TiN film exhibited high hardness, repeated stress loading inevitably caused some degree of deformation in the metal substrate. However, when the NiTi substrate was heated, it underwent an austenitic phase transition and exhibited a shape memory effect. This phase transition facilitated the deformation recovery at the wear scar site, resulting in a decrease in the worn volume and, thus, a lower wear rate.

In this section, we analyze the changes in the profiles of the first wear scar on the two composite materials during their initial cryogenic–thermal cycles with the corresponding wear scars and their variation trajectories illustrated in Figure 11, as referenced in Figure 8. As shown in Figure 11a, there was almost no change in the wear scar profile before and after the deep-cryogenic treatment, which could be attributed to the insignificant shape change in the NiTi alloy owing to the martensitic phase transition. However, following the water bath heat treatment, when the NiTi substrate underwent austenitic phase transition accompanied by the triggering of the shape memory effect, the wear scar was significantly reduced. Specifically, the depth of the wear scar decreased from an initial 920 nm to 450 nm, a change which demonstrates the recovery of the surface film wear scar on the NiTi alloy substrate under the action of the shape memory effect. As shown in Figure 11b, the wear scar profiles of the TiN/304 composite sample remained consistent across the three measurements, with only minor fluctuations observed. These fluctuations were likely due to measurement inaccuracies rather than substantial changes in the material properties.

The recovery process of wear scars is closely associated with the evolution of the surface morphology. As shown in Figure 12, SEM was used to document the morphological changes at the tail of the first wear scar on the two materials during the initial cryogenic–thermal cycle. The analysis of Figure 12a versus Figure 12b reveals that the martensitic phase transition induces negligible variations in the wear scar morphology. However, Figure 12c illustrates significant variations in the morphological features of the wear scar after the austenitic phase transition. The triggering of the shape memory effect not only renders the wear scar contour more distinct but also induces a certain degree of alteration in the morphology of the heavily worn areas at the edges of the scar. In contrast, the overall morphology of the wear scar tail on TiN/304 stainless steel, as shown in Figure 12d–f, exhibits no significant changes.

This study demonstrates the “self-healing” behavior due to phase transition in NiTi alloys and their composites after friction and wear, which aligns with observations reported in previous research [41,42]. This intriguing phenomenon represents a significant characteristic that differentiates shape memory alloy substrates from traditional substrate materials. NiTi alloys’ “self-healing” capability, as representatives of shape memory alloys and their composites, warrants further exploration and development for practical applications.

4. Conclusions

This study differs from studies that only focus on the wear resistance of films themselves and explores the effect of phase transitions in the NiTi alloy substrate on the tribological properties of the surface film under extreme temperature variations. The main research results indicate that the cyclic temperature changes cause the NiTi substrate to undergo a cyclic transition between the R phase and the B19′ phase, which in turn leads to periodic changes in the wear resistance of the TiN film prepared by magnetron sputtering. The transition from the R phase to the B19′ phase results in a maximum reduction of 41.97% in the wear rate of the TiN film and also causes the friction coefficient to decrease from about 0.45 to about 0.25. Moreover, the shape memory effect triggered by the B19′ to B2 phase transition in the NiTi alloy substrate results in a notable recovery of the TiN thin film wear track depth, reducing it from 920 nm to 550 nm, which demonstrates a self-healing phenomenon.

These quantitative results emphasize the necessity for focusing on the tribological properties of film/NiTi composite materials in applications with drastic temperature fluctuations. This study provides a valuable theoretical foundation and practical insights for the surface modification of NiTi alloys in the field of tribology. However, this study’s scope was limited to a specific temperature range and film type. Future research should expand the temperature range and vary the film type to further investigate the underlying mechanisms and optimize the performance of these composites.