Effects of Cold Rolling or Precipitation Hardening Treatment on the Microstructure, Mechanical Properties, and Corrosion Resistance of Ti-Rich Metastable Medium-Entropy Alloys

Titanium-rich metastable medium-entropy alloys, designed for low elastic moduli, sacrifice strength. However, enhancing their mechanical strength is crucial for bio-implant applications. This study aims to enhance the mechanical properties and corrosion resistance of a metastable Ti80–Nb10–Mo5–Sn5 medium-entropy alloy using various treatments, including cold rolling (at 50% and 75% reduction) and precipitation hardening (at room temperature, 150 °C, 350 °C, 550 °C, and 750 °C). The results showed that the alloy underwent a stress-induced martensitic transformation during the rolling process. Notably, the α phase was precipitated in the β grain boundaries after 30 days of precipitation hardening at room temperature. The yield strengths of the alloy increased by 51% and 281.9% after room-temperature precipitation and 75% cold rolling, respectively. In potentiodynamic corrosion tests conducted in phosphate-buffered saline solution, the pitting potentials of the alloy treated using various conditions were higher than 1.8 V, and no pitting holes were observed on the surface of the alloys. The surface oxide layer of the alloy was primarily composed of TiO2, Nb2O5, MoO3, and SnO2, contributing to the alloy’s exceptional corrosion and pitting resistance. The 75% rolled Ti80–Nb10–Mo5–Sn5 demonstrates exceptional mechanical properties and high corrosion resistance, positioning it as a promising bio-implant candidate.


Introduction
In recent years, Ti-rich metastable β high-entropy alloys (HEAs) and medium-entropy alloys (MEAs) have garnered significant attention in the field of biomedical materials [1][2][3][4][5][6][7][8][9].These alloys are known for their high mechanical strength and low elastic moduli.A key research goal is to improve the strengths of MEAs while maintaining their low elastic moduli.Severe plastic deformation can achieve grain refinement in conventional Ti alloys, which not only greatly increases their strength but also retains their original low elastic modulus [10].Many studies have adopted cold rolling to produce grain refinement to improve the mechanical properties of HEAs/MEAs, such as Hf-Nb-Ta-Ti-Zr [11], Al 0.25 -Co-Cr-Fe-Ni [12], Cr-Co-Ni [13], Al 4 -Nb 8 -Ti 50 -Mo 4 -Zr 34 [14], Ti 65 -(Al-Cr-Nb) 35 [15], and Zr 35 -Ti 30 -Nb 20 -Al 10 -Ta 5 [16].The presence of an ultrafine grain structure not only provides the alloy with a high mechanical strength but also elevates its corrosion resistance [17,18].Furthermore, when a metastable β HEA/MEA is subjected to external stress, a stress-induced martensitic (SIM) transformation may occur, resulting in a partial transformation of the β phase into the α phase [19,20].After SIM transformation, a large number of dislocations and nano-sized α precipitates are formed, resulting in a notable Materials 2023, 16, 7561 2 of 18 enhancement in the alloy's strength [21].However, an excessive amount of the α phase in an alloy can lead to a decrease in strength [22,23].
Precipitation hardening is a common method to enhance the mechanical properties of Ti alloys, HEAs, and MEAs [24][25][26].Via control of the precipitation treatment, it is possible to regulate the size and distribution of the precipitated phases, allowing for adjustments in various alloy properties.Precipitation hardening treatment can be divided into singlestep (direct heat treatment) and two-step (solution treatment + heat treatment) processes.Two-step precipitation hardening treatment can produce a finer and more uniformly distributed α phase in β-Ti alloys than the single-step process, which can significantly increase the yield strength, ductility, and corrosion resistance of Ti alloys [24].Moreover, the fine and uniformly distributed α phase does not significantly increase the elastic moduli of Ti alloys [27].Furthermore, the precipitated phases can fill surface defects, improving the corrosion resistance of Ti alloys [28,29].A significant amount of research has been conducted on the precipitation hardening of high-entropy alloys [25,26,[30][31][32].However, limited literature exists concerning the precipitation hardening of biomedical HEAs or MEAs, particularly in terms of its impact on alloys' structure, mechanical properties, and corrosion resistance.
The innovative aspect of this study lies in the comprehensive investigation into the impact of various processing methods, specifically cold rolling and heat treatment, on Ti 80 -Nb 10 -Mo 5 -Sn 5 .This study delves into the intricate relationship between its microstructure, mechanical strength, and corrosion resistance.By exploring the alterations in the microstructure caused by these processing methods, this study provides novel insights into optimizing the alloy's performance.Additionally, the study aims to identify the most effective processing conditions to enhance both the mechanical strength and corrosion resistance of Ti 80 -Nb 10 -Mo 5 -Sn 5 , thereby advancing its potential applications in biomedical materials.

Materials and Methods
The metastable β structure of the Ti 80 -Nb 10 -Mo 5 -Sn 5 (at%) MEA is achieved by adjusting the molybdenum equivalent ([Mo] eq ) to 15 and the valence electron concentration (VEC) to 4.2, which results in the low elastic modulus of the alloy.The thermodynamic parameters detailed in Table 1 were determined using calculations based on the formulas referenced in [9].These ingots were prepared using a commercial arc-melting vacuum pressure casting system (A-028, DAWNSHINE, Taichung, Taiwan) with pure Ti (99.7 wt%), Nb (99.95 wt%), Mo (99.95 wt%), and Sn (99.9 wt%).All ingots underwent re-melting and were flipped at least six times to ensure the homogeneity of the constituent elements.The initial dimensions of the Ti 80 -Nb 10 -Mo 5 -Sn 5 ingots were 15 mm × 15 mm × 15 mm, and after casting, the samples measured 40 mm × 5 mm × 5 mm.The solution treatment of each sample was conducted in a tubular furnace (MTF 12/38/250, Carbolite Gero, Hope Valley, UK) at 900 • C for 15 min in an argon atmosphere, followed by rapid cooling in ice water.Subsequently, Ti 80 -Nb 10 -Mo 5 -Sn 5 samples underwent cold rolling using a rolling machine (5TSB, Jong Yih Co., Kaohsiung, Taiwan) with a roll diameter of 15 cm.Reductions of 50% and 75% in the sample thickness were achieved using incremental passes of cold rolling, with each pass incrementally reducing the thickness by approximately 10%.Seven passes were required to achieve the 50% reduction, while a 75% reduction demanded thirteen passes of cold rolling.The precipitation hardening treatments were performed using a tubular furnace under an argon atmosphere at temperatures of 150 • C, 350 • C, 550 • C, or 700 • C for 15 min each.The natural precipitation hardening treatment was maintained at 25 • C for 30 days.The codes for the Ti 80 -Nb 10 -Mo 5 -Sn 5 alloy under various conditions are summarized in Table 2. X-ray diffraction (XRD) employing Cu-Kα radiation was carried out by using an XRD system (D8 Advance, Bruker, Karlsruhe, Germany) to perform phase analysis at 40 kV and 40 mA, with a scanning range of 2θ = 30-80 • , a scanning speed of 4 • /min, and a step size of 0.02 • /step.Electron backscatter diffraction (EBSD) was used to analyze the grain orientation and phase structure (SU5000, Hitachi, Tokyo, Japan).In order to evaluate the mechanical properties of the samples under the flexural stresses that occur during human body movement, three-point bending tests were performed using a desktop mechanical tester (HT-2102AP, Hung Ta Instrument, Taichung, Taiwan).The procedures of the threepoint bending experiments were in accordance with those described in previous works [33].The yield strength was determined using a 0.2% strain offset from the stress-strain curve obtained during the bending test.
To evaluate the corrosion behavior of the Ti 80 -Nb 10 -Mo 5 -Sn 5 sample, potentiodynamic polarization tests were conducted using a Potentiostat (PGSTAT12, Autolab, Utrecht, The Netherlands) in phosphate-buffered saline (PBS) at 37 • C and a pH of 7.4.The working electrode, reference electrode, and auxiliary electrode used were the Ti 80 -Nb 10 -Mo 5 -Sn 5 sample, a silver chloride electrode (Ag/AgCl), and a platinum plate, respectively.Each specimen was tested using three individual samples to investigate their corrosion resistance.The exposed surface area of the working electrode was 10 mm × 10 mm × 1 mm.To ensure accuracy, the electrolyte solution was purged with nitrogen gas for 30 min prior to testing.The potentiodynamic polarization tests were initiated after the open circuit potential (OCP) had stabilized for an hour in the PBS solution.The scan rate and scan ranges were set at 0.001 V/s and −0.3 to 1.8 V, respectively.The PBS solution contained NaCl (8.0 g/L), KCl (0.2 g/L), Na 2 HPO 4 (1.44 g/L), and KH 2 PO 4 (0.24 g/L) [34].The corrosion potentials and corrosion current densities were calculated using the Tafel extrapolation method.Following the potentiodynamic polarization tests, scanning electron microscopy (SEM; 6330, JEOL, Tokyo, Japan) was used to examine the microstructure of the alloys.Additionally, electrochemical impedance spectroscopy (EIS) measurements were conducted using an applied alternating current signal with a voltage amplitude of 10 mV in a frequency range from 10 5 to 10 −2 Hz, with measurements taken at the OCP after immersion in the electrolyte for 1 h.The EIS Spectrum Analyzer software was utilized to fit and analyze the experimental data.Moreover, X-ray photoelectron spectroscopy (XPS; JAMP-9500F, JEOL, Tokyo, Japan) was utilized to analyze the surface chemical compositions of the alloy after the potentiodynamic polarization tests.

Microstructure
The XRD patterns of Ti 80 -Nb 10 -Mo 5 -Sn 5 under different conditions are presented in Figure 1.In the ST state, Ti 80 -Nb 10 -Mo 5 -Sn 5 exhibited a single β phase, while after 50% and 75% cold rolling, a β + α phase was observed due to the SIM transformation.After the precipitation hardening treatment, the P150, P550, and P700 samples displayed a single β phase, while P350 and PRT showed a β + α phase.The alloy still presented a single β phase after low-temperature (150 • C) heat treatment, attributed to an insufficient heating time for α phase precipitation.Similarly, the high-temperature heat treatment (>550 • C) resulted in a single β phase due to the heating temperatures exceeding the β-phase transus temperature.In contrast, at room temperature, the precipitation hardening behavior of the Ti 80 -Nb 10 -Mo 5 -Sn 5 was primarily attributed to the abundant vacancies accelerating atomic diffusion and α phase nucleation and growth [35].The rapid cooling process during casting caused vacancies to remain in the alloy, while high-temperature solution treatment increased the equilibrium concentration of the vacancies [36].
ing casting caused vacancies to remain in the alloy, while high-temperature solution treatment increased the equilibrium concentration of the vacancies [36].
EBSD images of the Ti80-Nb10-Mo5-Sn5 samples with a single β phase under various conditions are shown in Figure 2. All β-phase samples (ST, P150, P550, and P700) exhibited equiaxed grains, which is characteristic of the β phase and confirms the XRD results.The average grain size of the P150 (35.5 ± 5.3 µm) was similar to that of the ST (34 ± 5.0 µm) due to the lower heating temperature of the P150.In contrast, P550 and P700 under hightemperature treatment exhibited larger grain sizes (49.5 ± 9.9 and 54.5 ± 2.5 µm, respectively).The EBSD images of the Ti80-Nb10-Mo5-Sn5 samples with a non-single β phase in this study are shown in Figure 3.The grain size of the Ti80-Nb10-Mo5-Sn5 decreased significantly due to fragmentation after cold rolling.The average grain sizes of the β phase were <10 and <1 µm for CR50 and CR75, respectively.Furthermore, the CR75 had a larger grain size for the α″ phase (~5 µm) than the CR50 due to the induction of a greater α″ phase due to cold rolling.In P350 (Figure 3h), a small amount of the α phase precipitated, with the grain sizes of the β and α phases measuring 38.4 ± 6.1 and <1 µm, respectively.Additionally, a significant amount of the α phase was observed in the PRT sample (Figure 3k), confirming the precipitation of the α phase in the Ti80-Nb10-Mo5-Sn5 after ST and 30 days of precipitation hardening treatment at room temperature.Moreover, P350 and PRT had ultrafine grain sizes for the α phase (~1 µm), attributed to the short heating time (15 min) and low heating temperature (room temperature), respectively.This phenomenon can be attributed to the sluggish diffusion effect in high-entropy alloys, which impedes atomic diffusion and contributes to the formation of ultrafine α grains in the alloy [37].EBSD images of the Ti 80 -Nb 10 -Mo 5 -Sn 5 samples with a single β phase under various conditions are shown in Figure 2. All β-phase samples (ST, P150, P550, and P700) exhibited equiaxed grains, which is characteristic of the β phase and confirms the XRD results.The average grain size of the P150 (35.5 ± 5.3 µm) was similar to that of the ST (34 ± 5.0 µm) due to the lower heating temperature of the P150.In contrast, P550 and P700 under high-temperature treatment exhibited larger grain sizes (49.5 ± 9.9 and 54.5 ± 2.5 µm, respectively).The EBSD images of the Ti 80 -Nb 10 -Mo 5 -Sn 5 samples with a non-single β phase in this study are shown in Figure 3.The grain size of the Ti 80 -Nb 10 -Mo 5 -Sn 5 decreased significantly due to fragmentation after cold rolling.The average grain sizes of the β phase were <10 and <1 µm for CR50 and CR75, respectively.Furthermore, the CR75 had a larger grain size for the α phase (~5 µm) than the CR50 due to the induction of a greater α phase due to cold rolling.In P350 (Figure 3h), a small amount of the α phase precipitated, with the grain sizes of the β and α phases measuring 38.4 ± 6.1 and <1 µm, respectively.Additionally, a significant amount of the α phase was observed in the PRT sample (Figure 3k), confirming the precipitation of the α phase in the Ti 80 -Nb 10 -Mo 5 -Sn 5 after ST and 30 days of precipitation hardening treatment at room temperature.Moreover, P350 and PRT had ultrafine grain sizes for the α phase (~1 µm), attributed to the short heating time (15 min) and low heating temperature (room temperature), respectively.This phenomenon can be attributed to the sluggish diffusion effect in high-entropy alloys, which impedes atomic diffusion and contributes to the formation of ultrafine α grains in the alloy [37].

Mechanical Properties
The stress-deflection curves of the Ti 80 -Nb 10 -Mo 5 -Sn 5 samples in various states under three-point bending tests are shown in Figure 4a.The comparison of the bending strengths, yield strengths, and elastic moduli of the Ti 80 -Nb 10 -Mo 5 -Sn 5 samples after various treatments is illustrated in Figure 4b.Due to the contribution of work hardening and grain refinement, both CR50 and CR75 exhibited ultra-high yield strengths (1007 and 1351 MPa, respectively), which were far greater than ST (354 MPa), by 184% and 282%, respectively.Despite this, CR50 and CR75 still maintained a high bending deformation capability (deflection > 8 mm), a combination of the high strength and ductility often found in HEAs [38].The high yield strengths of CR50 and CR75 are attributed to the work hardening induced by dislocation activity [12].Additionally, cold rolling suppresses the dynamic recovery in the alloy, increasing the dislocation density as the grain size decreases, thereby enhancing the yield strength [12].Among all heating conditions, P350 exhibited the highest yield strength (772 MPa) due to the strengthening of the ultrafine α precipitates.The yield strength of the Ti 80 -Nb 10 -Mo 5 -Sn 5 after natural precipitation hardening treatment (536 MPa) was 51% higher than that of ST (354 MPa).However, the improvement in yield strength after natural precipitation hardening treatment is not as significant as that after cold rolling and certain annealing conditions.The experimental results demonstrate the possibility of enhancing mechanical properties using natural precipitation hardening treatment.
The elastic moduli of the Ti 80 -Nb 10 -Mo 5 -Sn 5 increased by 40% and 24% after cold rolling of 50% (CR50) and 75% (CR75), respectively, due to the SIM transformation.It is well documented that the elastic modulus of the metastable β phase is lower than that of the α phase in Ti-Nb-based alloys [23,39,40].Furthermore, the elastic anisotropy caused by cold rolling could have led to the increase in the elastic modulus of CR50 and CR75 [41].The lower elastic modulus of CR75 compared to CR50 is attributed to an increase in the amount of cold rolling, which induced a greater α phase and increased dislocation density [42][43][44][45].Overall, the bending strength and yield strength of the Ti 80 -Nb 10 -Mo 5 -Sn 5 were improved significantly at the cost of a slightly increased elastic modulus after 75% cold rolling.Additionally, the elastic modulus of the Ti 80 -Nb 10 -Mo 5 -Sn 5 slightly increased after the precipitation hardening treatment at 150 • C, which could be related to the increase in crystal defects caused by low-temperature heat treatment.When the heating temperature was further raised to 350 • C, the elastic modulus of the Ti 80 -Nb 10 -Mo 5 -Sn 5 significantly increased to 59 GPa due to the fine precipitation of the α phase.Although the EBSD results (Figure 3) show that the size of the α phase in P350 was very fine, the α-phase precipitation did increase its elastic modulus, indicating that the α phase may not obtain a nanoscale size.When the heating temperature was further increased to 550 • C and 700 • C, both the elastic moduli and yield strengths of the alloys decreased.Since P550 and P700 had a single β phase, the β-phase transus temperature of the Ti 80 -Nb 10 -Mo 5 -Sn 5 may have been below 550 • C. Therefore, the annealing of the Ti 80 -Nb 10 -Mo 5 -Sn 5 at 550 • C and 700 • C is similar to the effect of solution treatment, resulting in mechanical properties similar to those of the ST.
The potential of an alloy as a biomedical implant can be evaluated by simultaneously comparing its mechanical strength and elastic modulus (E).An excellent biomedical implant requires a high yield strength (σ y ) to prevent fracture within the body, coupled with a low elastic modulus (E) to avoid stress-shielding effects.Therefore, for the alloy's elastic modulus (<80 GPa) to remain significantly lower than cortical bone, a higher σ y /E value indicates its greater potential as a biomedical implant.In Figure 5, the yield-strength-to-elastic-modulus ratio (σ y /E, ×1000) is presented for several materials, including conventional Ti alloys (Ti-6Al-4V ELI and Ti-29Nb-13Ta-4.6Zr, in wt%) [46], a biomedical HEA (Ti 45 -Zr 37 -Nb 16 -Fe 1 -Mo 1 , in at%) [20], a biomedical MEA (Ti 33.33 -Zr 33.33 -Nb 33.33 in at%) [47], and the samples from this study.The CR75 had the highest σ y /E ratio (20.7), primarily attributed to its high σ y value (1351 MPa), resulting from the combined effects of lattice distortion and grain refinement.Furthermore, P350 exhibited the highest σ y /E ratio (13) among all alloys with an elastic modulus below 60 GPa.The following section of this study focuses on the corrosion resistance of ST, CR75, and P350, elucidating their potential as biomedical implants.
among all alloys with an elastic modulus below 60 GPa.The following section of this study focuses on the corrosion resistance of ST, CR75, and P350, elucidating their potential as biomedical implants.[20], a biomedical MEA [47], and conventional Ti alloys [46].

Potentiodynamic Polarization Test
To understand the corrosion behavior of the Ti80-Nb10-Mo5-Sn5 in the human body after different treatments, potentiodynamic polarization tests were conducted on three kinds of alloys with different structures, including ST (β), CR75 (β + α″), and P350 (β + α).The polarization curves of ST, CR75, and P350 in PBS at 37 °C are shown in Figure 6a.In addition, the electrochemical parameters obtained from the potentiodynamic polarization tests, such as the corrosion potential (Ecorr), corrosion current density (icorr), passivation potential (Epass), passive current density (ipass), the slope of the anodic curve (βa), the slope of the cathode curve (βc), and polarization resistance (Rp), are shown in Table 3.The trend charts of Ecorr, icorr, and Epass for the three kinds of samples are shown in Figure 6b.
From the polarization curves of ST, CR75, and P350, it can be seen that the corrosion resistance of Ti80-Nb10-Mo5-Sn5 significantly increased after cold rolling or precipitation hardening treatment, and the pitting potentials of the three kinds of samples were all higher than 1.8 V.Among them, CR75 had the highest Ecorr value (0.2 V), significantly higher than ST (-0.06 V) and P350 (0.02 V).Generally, alloys will generate a large number of defects on the surface and inside after cold rolling, and these high-energy defects will lead to a decrease in the corrosion resistance of the alloy [48].In addition, the phase boundaries of multi-phase alloys are also high-energy areas that are prone to preferential corrosion.Furthermore, several studies have shown that the Ecorr value of Ti alloys decreased significantly after cold working [48][49][50][51].However, this study found that the Ecorr value of the Ti80-Nb10-Mo5-Sn5 significantly increased after cold rolling.Some of the literature has noted that the increase in the Ecorr value of the alloy after deformation is related to its ultrafine grain structure [17,[52][53][54][55].According to Martin et al. [54], the corrosion behavior of the alloy is concurrently influenced by both its microstructure and grain size.A microstructure with uniformly distributed fine grains exhibits superior corrosion resistance compared to needle-like or lamellar microstructures, attributed to the minimization of tendencies for microgalvanic cell formation.Hence, CR75 demonstrates the highest Ecorr value, signifying its optimal corrosion resistance.On the other hand, the icorr value of  [20], a biomedical MEA [47], and conventional Ti alloys [46].

Corrosion Properties 3.3.1. Potentiodynamic Polarization Test
To understand the corrosion behavior of the Ti 80 -Nb 10 -Mo 5 -Sn 5 in the human body after different treatments, potentiodynamic polarization tests were conducted on three kinds of alloys with different structures, including ST (β), CR75 (β + α ), and P350 (β + α).The polarization curves of ST, CR75, and P350 in PBS at 37 • C are shown in Figure 6a.In addition, the electrochemical parameters obtained from the potentiodynamic polarization tests, such as the corrosion potential (E corr ), corrosion current density (i corr ), passivation potential (E pass ), passive current density (i pass ), the slope of the anodic curve (β a ), the slope of the cathode curve (β c ), and polarization resistance (R p ), are shown in Table 3.The trend charts of E corr , i corr , and E pass for the three kinds of samples are shown in Figure 6b.Table 3. Corrosion potential (E corr ), corrosion current density (i corr ), passivation potential (E pass ), passive current density (i pass ), the slope of the anodic curve (β a ), the slope of the cathode curve (β c ), and polarization resistance (R p ) of Ti 80 -Nb 10 -Mo 5 -Sn 5 under different conditions after potentiodynamic polarization tests.From the polarization curves of ST, CR75, and P350, it can be seen that the corrosion resistance of Ti 80 -Nb 10 -Mo 5 -Sn 5 significantly increased after cold rolling or precipitation hardening treatment, and the pitting potentials of the three kinds of samples were all higher than 1.8 V.Among them, CR75 had the highest E corr value (0.2 V), significantly higher than ST (-0.06 V) and P350 (0.02 V).Generally, alloys will generate a large number of defects on the surface and inside after cold rolling, and these high-energy defects will lead to a decrease in the corrosion resistance of the alloy [48].In addition, the phase boundaries of multi-phase alloys are also high-energy areas that are prone to preferential corrosion.Furthermore, several studies have shown that the E corr value of Ti alloys decreased significantly after cold working [48][49][50][51].However, this study found that the E corr value of the Ti 80 -Nb 10 -Mo 5 -Sn 5 significantly increased after cold rolling.Some of the literature has noted that the increase in the E corr value of the alloy after deformation is related to its ultrafine grain structure [17,[52][53][54][55].According to Martin et al. [54], the corrosion behavior of the alloy is concurrently influenced by both its microstructure and grain size.A microstructure with uniformly distributed fine grains exhibits superior corrosion resistance compared to needle-like or lamellar microstructures, attributed to the minimization of tendencies for microgalvanic cell formation.Hence, CR75 demonstrates the highest E corr value, signifying its optimal corrosion resistance.On the other hand, the i corr value of P350 (2.54 nA/cm 2 ) was significantly lower than those of ST (9.02 nA/cm 2 ) and CR75 (131.26 nA/cm 2 ), indicating that P350 had the lowest corrosion rate.The low i corr value of P350 is attributed to the precipitation of the α phase, which reduces the conductivity [56].Additionally, P350 exhibited the highest R P value (2409.44 kΩ cm 2 ), attributed to its low i corr value.Yet, the polarization resistances acquired using potentiodynamic polarization tests do not account for the effects of the electrolyte and capacitance.Further elucidation of the alloy's corrosion resistance is necessary, using EIS testing.
Materials 2023, 16, x FOR PEER REVIEW 10 of 19 P350 (2.54 nA/cm 2 ) was significantly lower than those of ST (9.02 nA/cm 2 ) and CR75 (131.26 nA/cm 2 ), indicating that P350 had the lowest corrosion rate.The low icorr value of P350 is attributed to the precipitation of the α phase, which reduces the conductivity [56].Additionally, P350 exhibited the highest RP value (2409.44 kΩ cm 2 ), attributed to its low icorr value.Yet, the polarization resistances acquired using potentiodynamic polarization tests do not account for the effects of the electrolyte and capacitance.Further elucidation of the alloy's corrosion resistance is necessary, using EIS testing.
The difference between the Ecorr and Epass values acts as an indicator for assessing an alloy's passivation capability.Table 3 shows that the CR75 had the lowest Ecorr-Epass value (0.27 V), indicating that the alloy can quickly undergo passivation after corrosion.The high passivation ability of CR75 is attributed to its ultrafine grain size.After grain refinement treatment, the increase in the grain boundary density and dislocation density can reduce the surface work function and promote the formation of thicker and less defective passivation films [17,55].In addition, more grain boundaries can provide more nucleation sites for oxidation in the passivation layer [57,58].The difference between the E corr and E pass values acts as an indicator for assessing an alloy's passivation capability.Table 3 shows that the CR75 had the lowest E corr -E pass value (0.27 V), indicating that the alloy can quickly undergo passivation after corrosion.The high passivation ability of CR75 is attributed to its ultrafine grain size.After grain refinement treatment, the increase in the grain boundary density and dislocation density can reduce the surface work function and promote the formation of thicker and less defective passivation films [17,55].In addition, more grain boundaries can provide more nucleation sites for oxidation in the passivation layer [57,58].

SEM Images
The SEM images of the ST, CR75, and P350 after potentiodynamic polarization testing are shown in Figure 7.The surfaces of all samples remained smooth after potentiodynamic polarization.No large corrosion products or pitting holes were observed.P350 (Figure 7c), with a high E corr value and the lowest i corr value, did not show any corrosion or products, attributed to the precipitation of the α phase.The ultrafine-grained α precipitates can fill the internal defects of grain boundaries and grains, which can inhibit intergranular corrosion and impede the permeation of electrolyte solutions, thereby enhancing the chemical stability of the alloy [28,29].Small white corrosion products were observed on the surface of the ST (Figure 7a), not found in the other two samples, due to its low E corr and high E corr -E pass value.Several light corrosion stripes parallel to the rolling direction were observed on the surface of CR75 (Figure 7b), inferred to be the preferential corrosion of the residual stress regions generated by cold rolling.As a result, the CR75 had the highest i corr value.Despite the limited number of stripes observed on the CR75 surface, these stripes could compromise the integrity of the passive film.Therefore, it is imperative to conduct further analysis of the chemical composition and characteristics of the passive film using EIS and XPS.The SEM images of the ST, CR75, and P350 after potentiodynamic polarization testing are shown in Figure 7.The surfaces of all samples remained smooth after potentiodynamic polarization.No large corrosion products or pitting holes were observed.P350 (Figure 7c), with a high Ecorr value and the lowest icorr value, did not show any corrosion or products, attributed to the precipitation of the α phase.The ultrafine-grained α precipitates can fill the internal defects of grain boundaries and grains, which can inhibit intergranular corrosion and impede the permeation of electrolyte solutions, thereby enhancing the chemical stability of the alloy [28,29].Small white corrosion products were observed on the surface of the ST (Figure 7a), not found in the other two samples, due to its low Ecorr and high Ecorr-Epass value.Several light corrosion stripes parallel to the rolling direction were observed on the surface of CR75 (Figure 7b), inferred to be the preferential corrosion of the residual stress regions generated by cold rolling.As a result, the CR75 had the highest icorr value.Despite the limited number of stripes observed on the CR75 surface, these stripes could compromise the integrity of the passive film.Therefore, it is imperative to conduct further analysis of the chemical composition and characteristics of the passive film using EIS and XPS.

EIS
Figure 8 presents the EIS results of ST, CR75, and P350 in PBS solution at 37 °C, with Nyquist diagrams and Bode plots at frequencies ranging from 10 5 to 10 -2 Hz shown in Figure 8a,b, respectively.From the Nyquist plot, it can be observed that the capacitance semicircles of all three samples exhibited an incomplete shape, which is related to the capacitive response of the passive film [59].Compared to CR75 and P350, ST has a capacitance semicircle with a larger diameter, indicating that the passive film of the ST has the best corrosion resistance.From the Bode modulus curves, it can be seen that the slopes of the curves of all the samples in the high-frequency range (10 4 to 10 5 Hz) were close to 0, indicating that all samples have a similar electrolyte resistance [59].Furthermore, all the Bode phase curves had a phase angle of approximately 0° at high frequencies, indicating no phase signal delay.In the low-frequency range (10 -1 to 10 1 Hz),   8a,b, respectively.From the Nyquist plot, it can be observed that the capacitance semicircles of all three samples exhibited an incomplete shape, which is related to the capacitive response of the passive film [59].Compared to CR75 and P350, ST has a capacitance semicircle with a larger diameter, indicating that the passive film of the ST has the best corrosion resistance.From the Bode modulus curves, it can be seen that the slopes of the curves of all the samples in the high-frequency range (10 4 to 10 5 Hz) were close to 0, indicating that all samples have a similar electrolyte resistance [59].Furthermore, all the Bode phase curves had a phase angle of approximately 0 • at high frequencies, indicating no phase signal delay.In the low-frequency range (10 −1 to 10 1 Hz), the Bode phase slopes of all samples were close to 0, demonstrating the formation of a passive film on the surface of the samples.The negative phase angles of the Bode phase curves of all three samples were greater than 80 • , indicating that the passive films generated in PBS exhibit a good capacitive behavior [60].indicating the best corrosion resistance of the passive film on the surface of ST.CR75 showed comparatively lower R 1 and C eff values owing to the compromised integrity of the passive layer resulting from the residual stress induced during cold rolling.The P350 exhibited a high E corr value and the lowest i corr value in the potentiodynamic test (Figure 6), but the R 1 value of P350 in the EIS test was lower than that of ST.The precipitated phase (α) in P350 may hinder the diffusion of oxygen ions and slow down the growth rate of the passive layer [61,62].Furthermore, the lower C eff value of P350 compared to ST also indicates a thinner passive layer on the surface of P350.Nevertheless, ST, CR75, and P350 all exhibit ultra-high R 1 values (>10 6 Ω cm 2 ) and low C eff values (<3 × 10 −5 F cm 2 ), which are sufficient for the corrosion resistance requirements of biomedical implants.

XPS
The XPS chemical characterization results of the CR75 surface after potentiodynamic polarization are presented in Figure 10.The full spectrum (Figure 10a) indicates the presence of Ti, Nb, Sn, Mo, C, and O elements in the oxide film/passivation layer on CR75.The presence of C and O elemental peaks in the full spectrum result from surface carbon contamination and the oxide layer.Narrow scans of the O 1s, Ti 2p, Nb 3d, Sn 3d, and Mo 3d elements of the CR75 sample are shown in Figure 10b-f, respectively.The O 1s peaks correspond to the OH − , O − , and O 2− oxidation states; the Ti 2p peaks belong to the Ti 4+ oxidation state; the Nb 3d peaks are related to the Nb 5+ oxidation state; the Sn 3d spectrums are ascribed to the Sn 0 and Sn 4+ oxidation states; and the Mo 3d peaks are composed of the Mo 0 , Mo 2+ , Mo 4+ , and Mo 6+ oxidation states.Therefore, the surface oxide film/passivation layer of the CR75 mainly comprises TiO 2 , Nb 2 O 5 , SnO 2 , MoO, MoO 2 , and MoO 3 .The presence of Nb 2 O 5 improves the resistance to Cl − ion erosion and enhances the structural integrity of the oxide film/passivation layer [48].In the passivation layer of Ti alloys, there may be O vacancies and Ti defects, which will lead to a decrease in the corrosion resistance of the passivation layer.When ions with a larger radius, such as Nb 5+ and Mo 6+ , are present in the alloy passivation layer, they can form a chemical bond with Ti atoms and eliminate Ti defects, thereby improving the corrosion resistance [63].

Conclusions
The microstructure, mechanical properties, and corrosion properties of the Ti80-Nb10-Mo5-Sn5 alloy were investigated under different conditions, including cold rolling and In contrast to the Ti, Nb, and Sn elements, Mo exhibits multiple oxidation states in the passivation layer of the CR75.The electron configuration and elemental properties of Mo affect the formation of its oxidation state.The electron configuration of Mo is [Kr]4d 5 5s 1 , in which 4d electrons have relatively high energy and easily participate in chemical reactions to form compounds, while the chemical properties of 5s electrons are relatively inert.Hence, in Ti-rich alloys with Mo as an alloying element, the electron configuration and elemental properties of Mo promote the formation of 2+, 3+, and 4+ oxidation states while impeding the formation of 6+ oxidation states.Moreover, other alloy elements present in the alloy (Ti, Nb, and Sn) compete with Mo elements and further limit the oxidation state of Mo.As a consequence, the existence of several oxidation states of Mo in the passivation layer of the CR75 is unrelated to its uneven surface.In summary, although the surface of the CR75 is not very uneven, the presence of Nb, Sn, and Mo oxides in the passivation layer still makes it have a high R 1 value (>10 6 Ω cm 2 ) and low C eff value (<3 × 10 −5 F cm 2 ).

Conclusions
The microstructure, mechanical properties, and corrosion properties of the Ti 80 -Nb 10 -Mo 5 -Sn 5 alloy were investigated under different conditions, including cold rolling and precipitation hardening treatment at various temperatures.The XRD patterns revealed that the alloy exhibited a single β phase in the solution-treated state.In contrast, the SIM transformation induced a β + α phase after cold rolling (50% and 75%), while heating at 350 • C for 15 min resulted in a β + α phase.Cold rolling significantly reduced the grain size of the β phase, while high-temperature precipitation hardening treatment led to larger grain sizes.Additionally, α precipitates were found in the ST sample after precipitation hardening treatment at room temperature for 30 days.After undergoing 75% cold rolling, the Ti 80 -Nb 10 -Mo 5 -Sn 5 alloy (CR75) exhibited a significant increase in yield strength, albeit with a slight increase in elastic modulus.Moreover, the corrosion resistance of the Ti 80 -Nb 10 -Mo 5 -Sn 5 alloy significantly increased after both the cold rolling and precipitation hardening treatments.The CR75 exhibited the highest E corr value, which was attributed to its ultrafine grain size and high passivation ability.The EIS results indicated ultra-high R 1 values (>10 6 Ω cm 2 ) and low C eff values (<3 × 10 −5 F cm 2 ) for the ST, CR75, and P350 samples, affirming their good corrosion resistance.The XPS analysis indicated that the surface oxide film/passivation layer of the CR75 mainly consisted of TiO 2 , Nb 2 O 5 , SnO 2 , MoO, MoO 2 , and MoO 3 .To summarize, CR75 displayed the highest σ y /E ratio alongside excellent corrosion resistance, positioning it as a promising candidate for biomedical implants.

Figure 8
Figure8presents the EIS results of ST, CR75, and P350 in PBS solution at 37 • C, with Nyquist diagrams and Bode plots at frequencies ranging from 10 5 to 10 −2 Hz shown in Figure8a,b, respectively.From the Nyquist plot, it can be observed that the capacitance semicircles of all three samples exhibited an incomplete shape, which is related to the capacitive response of the passive film[59].Compared to CR75 and P350, ST has a capacitance semicircle with a larger diameter, indicating that the passive film of the ST has the best corrosion resistance.From the Bode modulus curves, it can be seen that the slopes of the curves of all the samples in the high-frequency range (10 4 to 10 5 Hz) were close to 0, indicating that all samples have a similar electrolyte resistance[59].Furthermore, all the Bode phase curves had a phase angle of approximately 0 • at high frequencies, indicating no phase signal delay.In the low-frequency range (10 −1 to 10 1 Hz), the Bode phase slopes of all samples were close to 0, demonstrating the formation of a passive film on the surface of the samples.The negative phase angles of the Bode phase curves of all three samples

Figure 8 . 1 n 1 R 1 1−n 1 n 1 .
Figure 8.(a) Nyquist diagrams and (b) Bode plots of Ti 80 -Nb 10 -Mo 5 -Sn 5 at various conditions in PBS solution at 37 • C.An equivalent electrical circuit (EEC) model of the three Ti 80 -Nb 10 -Mo 5 -Sn 5 samples is shown in Figure 9.The design of this EEC model is based on the results of the potentiodynamic polarization tests for ST, CR75, and P350, which did not experience pitting corrosion in the potential range of -0.3 to 1.8 V. Within this EEC model, R s represents the electrolyte resistance, CPE 1 denotes the constant phase element of the passive layer, and R 1 stands for the polarization resistance of the passive layer.The EIS data of each sample, including the electrolyte resistance (R s ), CPE 1 , R 1 , deviation parameter for CPE 1 (n 1 ), effective constant capacitance value (C eff ), and chi-square value (χ 2 ) fitted based on the EEC in Figure 9 using the EIS Spectrum Analyzer software, are listed in Table 4.The formula for calculating C eff is as follows [59]: C eff,1 = CPE 1

Table 4 .
The EIS data of Ti 80 -Nb 10 -Mo 5 -Sn 5 under different conditions in the PBS electrolyte, including the electrolyte resistance (R s ), constant phase element of the passive layer (CPE 1 ), polarization resistance of the passive layer (R 1 ), deviation parameter for CPE 1 (n 1 ), effective constant capacitance value (C eff ), and chi-square value (χ 2 ) fitted based on the EEC in Figure9.

Materials 2023 , 19 Figure 10 .
Figure 10.The chemical characterization of the surface of CR75 after potentiodynamic polarization test.(a) Full spectrum and (b-f) narrow scans.

Figure 10 .
Figure 10.The chemical characterization of the surface of CR75 after potentiodynamic polarization test.(a) Full spectrum and (b-f) narrow scans.