Effect of Annealing Temperature on Electrochemical Properties of Zr56Cu19Ni11Al9Nb5 in PBS Solution

The electrochemical properties of as-cast Zr56Cu19Ni11Al9Nb5 metallic glass and samples annealed at different temperatures were investigated using potentiodynamic polarization tests and electrochemical impedance spectroscopy (EIS) in phosphate buffer saline (PBS) solution. It was shown that passivation occurred for the as-cast sample and the samples annealed at 623–823 K, indicating good corrosion resistance. At higher annealing temperature, the corrosion resistance first increased, and then decreased. The sample annealed at 823 K exhibited the best corrosion resistance, with high spontaneous corrosion potential Ecorr at −0.045 VSCE, small corrosion current density icorr at 1.549 × 10−5 A·cm−2, high pitting potential Epit at 0.165 VSCE, the largest arc radius, and the largest sum of Rf and Rct at 5909 Ω·cm2. For the sample annealed at 923 K, passivation did not occur, with low Ecorr at −0.075 VSCE, large icorr at 1.879 × 10−5 A·cm−2, the smallest arc radius, and the smallest sum of Rf and Rct at 2173 Ω·cm2, which suggested the worst corrosion resistance. Proper annealing temperature led to improved corrosion resistance due to structural relaxation and better stability of the passivation film, however, if the annealing temperature was too high, the corrosion resistance deteriorated due to the chemical inhomogeneity between the crystals and the amorphous matrix. Optical microscopy and scanning electron microscopy (SEM) examinations indicated that localized corrosion occurred. Results of energy dispersive X-ray spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS) illustrated that the main corrosion products were ZrO2, CuO, Cu2O, Ni(OH)2, Al2O3, and Nb2O5.


Introduction
Zr-based bulk metallic glasses (BMGs) have attracted intense concern due to their high strength, high hardness, large elastic limit, good corrosion resistance, good glassforming ability (GFA), large critical size, and good biocompatibility [1][2][3][4][5][6]. Corrosion resistance was affected by the chemical composition, microstructure, heat treatment conditions, the types and concentrations of corrosive media, temperature, and so on. With higher Nb content, the corrosion resistance of Zr 46 Cu 30.14−x Nb x Ag 8.36 Al 8 Be 7.5 (x = 0, 2, 5, 10) in 0.1 mol/L HCl, 0.5 mol/L NaCl, and 0.5 mol/L H 2 SO 4 solutions was improved due to the promoted oxidation of Zr and thicker oxide film [7]. The electrochemical properties of Zr 56.2 Ti 13.8 Nb 5.0 Cu 6.9 Ni 5.6 Be 12.5 composite with crystalline dendrites in an amorphous matrix in 1 mol/L NaCl solution showed that corrosion in the amorphous matrix was faster [8]. In 1 mol/L HCl solution, the corrosion resistance is as follows: Zr 55 Ti 4 Y 1 Al 10 Cu 20 Ni 7 Co 2 Fe 1 > Zr 50 Ti 4 Y 1 Al 10 Cu 25 Ni 7 Co 2 Fe 1 > Zr 55 Ti 4 Y 1 Al 12 Cu 18 Ni 7 Co 2 Fe 1 > Zr 60 Al 12 Cu 28 > Zr 60 Al 10 Cu 30 > Zr 55 Al 10 Cu 35 [9]. Zr 55 Ti 4 Y 1 Al 10 Cu 20 Ni 7 Co 2 Fe 1 exhibited the highest corrosion potential E corr at −0.420 V SCE , and the smallest corrosion current density i corr at 1.0 × 10 −7 A·cm −2 , suggesting the best corrosion resistance. Zr 55 Al 10 Cu 35 illustrated the lowest E corr at −1.165 V SCE , and the largest i corr at 3.0 × 10 −7 A·cm −2 , suggesting the worst corrosion resistance. The corrosion resistance is better with higher E corr and smaller i corr . Zr 55 Ti 4 Y 1 Al 10 Cu 20 Ni 7 Co 2 Fe 1 , Zr 50 Ti 4 Y 1 Al 10 Cu 25 Ni 7 Co 2 Fe 1 and Zr 55 Ti 4 Y 1 Al 12 Cu 18 Ni 7 Co 2 Fe 1 showed higher content of Ti (4 at.%), Ni (7 at.%), and Co (2 at.%) and lower content of Cu (18-25 at.%) than Zr 60 Al 12 Cu 28 , Zr 60 Al 10 Cu 30 , and Zr 55 Al 10 Cu 35 , with better corrosion resistance. Cu decreased the corrosion resistance due to the deterioration of the protection of the film, and Ti, Ni, and Co increased the corrosion resistance because of the enhancement of the protection of the film. The electrochemical properties of Zr 52 Al 10 Ni 6 Cu 32 in 0.05-0.5 mol/L NaCl and 0.05-0.5 mol/L NaF solutions indicated that corrosion resistance decreased with higher concentrations of Cl − and F − [10]. Zr 60 Fe 10 Cu 20 Al 10 samples were prepared using selective laser melting with a laser power of 200 W and an exposure time of 40-70 µs, and their corrosion resistance in 3.5 wt.% NaCl solution decreased with the increase in exposure time [11].
Zr-based BMGs showed promising biomedical applications, such as orthopedic and dental device materials [12]. Zr 55.8 Al 19.4 Co 17.36 Cu 7.44 exhibited good corrosion resistance in phosphate-buffered saline (PBS) solution, combined with good glass-forming ability, a large critical diameter of 12 mm, a high yield strength of 2 GPa, and a high fracture toughness of 120 MPa·m 1/2 [13]. Zr 60.5 Hf 3 Al 9 Fe 4.5 Cu 23 showed good corrosion resistance in PBS solution, a large critical diameter of 10 mm, a high yield strength of 1.64 GPa, a large plastic strain of 4.0%, and good biocompatibility and wear resistance [14]. Zr 58.6 Al 15.4 Co 18.2 Cu 7.8 illustrated good corrosion resistance in PBS solution, a large critical diameter of 10 mm, a high yield strength of 1.95 GPa, a large plastic strain of 2.0%, and good antibacterial properties [15]. Zr 65−x Ti x Cu 20 Al 10 Fe 5 (x = 0-8) exhibited better corrosion resistance with higher Ti content, and the mechanical properties were the best with 2% Ti [16]. The corrosion resistance of Zr 53 Al 16 Co 26 M 5 (M = Pd, Au and Pt) in PBS solution was the best for Pt due to the increased Zr content and decreased Al content in the passive film, and the worst for Au [17]. Zr 45 Ti 36 Fe 11 Al 8 showed better corrosion resistance than commercially pure Ti in PBS solution [18]. Zr 55 Cu 30 Ni 5 Al 10 exhibited poorer corrosion resistance than the medical grade ASTM F 75 cast CoCrMo alloy and AISI 316LVM low carbon vacuum re-melted stainless steel alloy in PBS solution at a body temperature of 310 K [19]. Zr 40 Ti 37 Co 12 Ni 11 , Zr 50 Ti 32 Cu 13 Ag 5 , Zr 46 Ti 40 Ag 14 , and Zr 46 Ti 43 Al 11 indicated better corrosion resistance than commercially pure Ti and 316L stainless steel in PBS solution [20,21].
The mechanical properties and the corrosion and electrochemical properties of Zrbased BMGs were affected by annealing conditions, such as annealing temperature and annealing time. Proper annealing led to the formation of nanocrystals in the amorphous matrix, which acted as the initiation sites for shear bands and hindered the propagation of shear bands. Therefore, the shear band density was increased, resulting in improved plasticity [22]. If the annealing temperature was too high or the annealing time was too long, large size crystals were formed in the amorphous matrix, and strength and plasticity decreased due to the stress concentration and formation of microcracks at the interfaces. Zr 65 Cu 17.5 Fe 10 Al 7.5 samples were annealed at 573 K, below the glass transition temperature T g , for 0.5-4 h, and a large plasticity of 7.1%, high hardness of 487 HV, and improved pitting corrosion resistance in 3.5% NaCl solution was obtained for the sample annealed for 1 h, due to the formation of nanocrystals in the amorphous matrix, the reduced free volume, and the increased shear band density [22]. With the increase in annealing time, the plasticity, the hardness, the pit potential E pit , and the passivation region E pit -E corr first increased, and then decreased. Zr 58 Nb 3 Cu 16 Ni 13 Al 10 samples were annealed at 523 K, 673 K (lower than T g ), 773 K (higher than the crystallization temperature T x ), and 873 K for 6 h, and at the higher annealing temperature, the corrosion resistance in 1 mol/L H 2 SO 4 solution at 333 K deteriorated [23]. Zr 60 Cu 20 Ni 8 Al 7 Hf 3 Ti 2 samples were annealed below T g , and the good corrosion resistance in H 2 SO 4 solution was maintained [24]. Zr 50.7 Ni 28 Cu 9 Al 12.3 samples were annealed at 719 K (T g~Tx ), 768 K (>T x ), and 810 K for 30 min, and the electrochemical properties in 0.5 mol/L H 2 SO 4 , 1 mol/L NaCl, and 1 mol/L HCl solutions showed that the corrosion resistance was the best for the sample annealed at 768 K due to the proper quantity of ZrO 2 nanocrystals in the amorphous matrix [25]. Zr 41.2 Cu 12.5 Ni 10 Ti 13.8 Be 22.5 and Zr 57 Cu 15.4 Ni 12.6 Al 10 Nb 5 samples were annealed at 0.9 T g for 4 h, and the corrosion resistance in NaCl solution was improved due to the reduced free volume [26]. Zr 68 Al 8 Ni 8 Cu 16 samples were annealed at 673 K and 713 K with a crystallinity of 10% and 77%, and the corrosion resistance in 1 mol/L HCl solution decreased at the higher annealing temperature [27]. Zr 59 Ti 6 Cu 17.5 Fe 10 Al 7.5 samples were annealed at 573 K (<T g ) for 0.5-4 h, and the corrosion resistance remained good in PBS solution [28].
The effects of annealing temperature and time on the corrosion and electrochemical properties of Zr-based BMGs and the composites in simulated body fluids, such as PBS solution, were rarely reported, and systematic study was needed. In this work, the corrosion and electrochemical properties of the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 metallic glass and the samples annealed at different temperatures (<T g , T g -T x , >T x ) in PBS solution were investigated using potentiodynamic polarization tests, electrochemical impedance spectroscopy (EIS), optical microscopy, scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS), and X-ray photoelectron spectroscopy (XPS). Our findings make contributions to the research and development of Zr-based metallic glass and composites in biomedical applications.

Material Preparation
Zr 56 Cu 19 Ni 11 Al 9 Nb 5 metallic glass was prepared using copper-mold suction after the arc-melting of Zr, Cu, Ni, Al, and Nb with high purity under vacuum conditions and filled with Ar. Remelting was carried out 5 times to obtain chemical homogeneity. The samples were cut to a size around 5 mm × 4 mm × 1 mm using a SYJ-160 low speed diamond saw (Shenyang Kejing Automatic Equipment Limited Company, Shenyang, China). The samples were annealed at 623 K (below T g ), 723 K (between T g and T x ), 823 K (above T x ), and 923 K (far above T x ) for 30 min, and then cooled to room temperature. The sample surfaces were ground using 800, 1200, and 1500 grit sandpaper, polished using 2.5 and 0.5 µm diamond paste, and then cleaned in acetone.

Tests
Electrochemical tests of the as-cast samples and the annealed samples in PBS solution were performed using a CHI660E electrochemical station (Shanghai CH Instruments, Shanghai, China). Saturated calomel electrode (SCE) was the reference electrode, and the graphite electrode was the counter electrode. The sample was the working electrode. The PBS solution contained 8.0 g/L NaCl, 1.44 g/L Na 2 HPO 4 , 0.24 g/L KH 2 PO 4 , and 0.20 g/L KCl, and it was prepared using reagent-grade chemicals and deionized water. The electrochemical tests were performed in the open air at room temperature. In potentiodynamic polarization tests and EIS tests, the electrodes were first stabilized in PBS solution at open circuit potential (OCP) for 60 min. In potentiodynamic polarization tests, the potential was scanned from −0.8 V SCE to 0.8 V SCE at the rate of 0.33 mV/s, and auto-sensitivity was set. In EIS tests, alternative current impedance mode was chosen, and the initial potential was set at the OCP. The frequency was in the range of 10 −2 -10 5 Hz. The amplitude was 5 mV, and the stabilization period was 2 s. The complex impedance was measured. Zsimpwin software was used to fit the EIS data using the proper equivalent circuit. SH11/YF-III optical microscopy, Zeiss Ultra Plus field emission scanning electron microscopy, X-Max 50× energy dispersive X-ray spectroscopy, and Thermo Scientific ESCALAB 250Xi X-ray photoelectron spectroscopy were used to study the corrosion morphology and corrosion products after the potentiodynamic polarization tests. Figure 1 shows the potentiodynamic polarization curves for the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample and the samples annealed at 623-923 K in PBS solution, and Table 1 summarizes the obtained electrochemical parameters. Similar trends were observed for the as-cast sample and the samples annealed at 623-823 K, and passivation occurred. With the increase in annealing temperature, the spontaneous corrosion potential E corr gradually increased and then decreased, and the corrosion current density i corr gradually decreased and then increased. Corrosion resistance increased with higher E corr , higher pitting potential E pit , smaller i corr , and a larger passivation region E pit -E corr . For the as-cast sample, E corr was the lowest at −0.083 V SCE , the pitting potential E pit was the highest at 0.496 V SCE , and the width of passivation region E pit -E corr was the largest at 0.579 V. Passivation occurred for the as-cast sample and the samples annealed at 623 K (below T g ), 723 K (between T g and T x ), and 823 K (above T x ), indicating good corrosion resistance. For the sample annealed at 623 K, E corr was higher at −0.042 V SCE , i corr was smaller at 1.466 × 10 −5 A·cm −2 , E pit was lower at 0.157 V, and the passivation region E pit -E corr was smaller at 0.199 V, suggesting similar corrosion resistance. For the sample annealed at 723 K, E corr was the highest at −0.036 V SCE , and i corr was the smallest at 9.977 × 10 −6 A·cm −2 , with a wide passivation region E pit -E corr of 0.395 V, indicating excellent corrosion resistance. The sample annealed at 823 K exhibited high E corr at −0.045 V SCE , small i corr at 1.549 × 10 −5 A·cm −2 , high E pit at 0.165 V SCE , and a wide passivation region E pit -E corr of 0.210 V, indicating good corrosion resistance. For the sample annealed at 923 K, i corr was the largest at 1.879 × 10 −5 A·cm −2 , and passivation did not occur, indicating the worst corrosion resistance in PBS solution. With the increase in annealing temperature, the corrosion resistance gradually increased and then decreased.  Figure 1 shows the potentiodynamic polarization curves for the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K in PBS solution, and Table 1 summarizes the obtained electrochemical parameters. Similar trends were observed for the as-cast sample and the samples annealed at 623-823 K, and passivation occurred. With the increase in annealing temperature, the spontaneous corrosion potential Ecorr gradually increased and then decreased, and the corrosion current density icorr gradually decreased and then increased. Corrosion resistance increased with higher Ecorr, higher pitting potential Epit, smaller icorr, and a larger passivation region Epit-Ecorr. For the as-cast sample, Ecorr was the lowest at −0.083 VSCE, the pitting potential Epit was the highest at 0.496 VSCE, and the width of passivation region Epit-Ecorr was the largest at 0.579 V. Passivation occurred for the as-cast sample and the samples annealed at 623 K (below Tg), 723 K (between Tg and Tx), and 823 K (above Tx), indicating good corrosion resistance. For the sample annealed at 623 K, Ecorr was higher at −0.042 VSCE, icorr was smaller at 1.466 × 10 −5 A•cm −2 , Epit was lower at 0.157 V, and the passivation region Epit-Ecorr was smaller at 0.199 V, suggesting similar corrosion resistance. For the sample annealed at 723 K, Ecorr was the highest at −0.036 VSCE, and icorr was the smallest at 9.977 × 10 −6 A•cm −2 , with a wide passivation region Epit-Ecorr of 0.395 V, indicating excellent corrosion resistance. The sample annealed at 823 K exhibited high Ecorr at −0.045 VSCE, small icorr at 1.549 × 10 −5 A•cm −2 , high Epit at 0.165 VSCE, and a wide passivation region Epit-Ecorr of 0.210 V, indicating good corrosion resistance. For the sample annealed at 923 K, icorr was the largest at 1.879 × 10 −5 A•cm −2 , and passivation did not occur, indicating the worst corrosion resistance in PBS solution. With the increase in annealing temperature, the corrosion resistance gradually increased and then decreased.   Table 2 shows the fitting parameters. The Nyquist plots exhibit half-circles, which suggest that the control step is the electrochemical reaction accompanied by the transfer of electrons. The Bode plots illustrate two time-constants, with frequencies around 3 Hz and 100 Hz. The maximum phase is reached around 3 Hz. The equivalent circuit diagram as shown in Figure 2c was previously used for the fitting of EIS results of Ni70Cr21Si0.5B0.5P8 and Ni72.65Cr7.3Si6.7B2.15Fe8.2Mo3 glassy alloys in 1-12 mol/L HNO3 solution [29], and EIS results of Zr65Cu17.5Al7.5Ni10-xCox in 3.5% NaCl solution [30].   Figure 2 illustrates the Nyquist plots, Bode plots, and the equivalent circuit diagram for the EIS results of the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample and the samples annealed at 623-923 K in PBS solution, and Table 2 shows the fitting parameters. The Nyquist plots exhibit half-circles, which suggest that the control step is the electrochemical reaction accompanied by the transfer of electrons. The Bode plots illustrate two time-constants, with frequencies around 3 Hz and 100 Hz. The maximum phase is reached around 3 Hz. The equivalent circuit diagram as shown in Figure 2c was previously used for the fitting of EIS results of Ni 70 Cr 21 Si 0.5 B 0.5 P 8 and Ni 72.65 Cr 7.3 Si 6.7 B 2.15 Fe 8.2 Mo 3 glassy alloys in 1-12 mol/L HNO 3 solution [29], and EIS results of Zr 65 Cu 17.5 Al 7.5 Ni 10−x Co x in 3.5% NaCl solution [30]. The passivation film consists of two layers, i.e., the compact inner layer and the porous outer layer. R s is the solution resistance between the working electrode and the reference electrode. R f is the film resistance, and R ct is the charge transfer resistance at the interface between the solution and the film. The constant phase element (CPE) represents the capacitance, considering the surface roughness and inhomogeneity. CPE1 and CPE2 are the constant phase elements for the inner layer and outer layer of the passivation film.

Electrochemical Tests
. Y 0 is the constant, ω is the angular frequency, and n is the parameter between 0 and 1. The sample annealed at 823 K illustrates the largest arc radius and the largest sum of R f and R ct , 5909 Ω·cm 2 , indicating the best corrosion resistance in PBS solution. The sample annealed at 923 K illustrates the smallest arc radius and the smallest sum of R f and R ct , 2173 Ω·cm 2 , indicating the worst corrosion resistance in PBS solution. With the increasing annealing temperature, the corrosion resistance first increases, and then decreases, which is in agreement with the potentiodynamic polarization results. The passivation film consists of two layers, i.e., the compact inner layer and the porous outer layer. Rs is the solution resistance between the working electrode and the reference electrode. Rf is the film resistance, and Rct is the charge transfer resistance at the interface between the solution and the film. The constant phase element (CPE) represents the capacitance, considering the surface roughness and inhomogeneity. CPE1 and CPE2 are the constant phase elements for the inner layer and outer layer of the passivation film. The CPE impedance is cos sin . Y0 is the constant, ω is the angular frequency, and n is the parameter between 0 and 1. The sample annealed at 823 K illustrates the largest arc radius and the largest sum of Rf and Rct, 5909 Ω•cm 2 , indicating the best corrosion resistance in PBS solution. The sample annealed at 923 K illustrates the smallest arc radius and the smallest sum of Rf and Rct, 2173 Ω•cm 2 , indicating the worst corrosion resistance in PBS solution. With the increasing annealing temperature, the corrosion resistance first increases, and then decreases, which is in agreement with the potentiodynamic polarization results.       Figure 3 shows the optical microscopy images of the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution. Obvious localized corrosion occurred for the as-cast sample and the samples annealed at 623 K and 723 K. Minor corrosion was observed in the sample annealed at 923 K. Corrosion was not obvious for the sample annealed at 823 K, suggesting good corrosion resistance.  Figure 3 shows the optical microscopy images of the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution. Obvious localized corrosion occurred for the as-cast sample and the samples annealed at 623 K and 723 K. Minor corrosion was observed in the sample annealed at 923 K. Corrosion was not obvious for the sample annealed at 823 K, suggesting good corrosion resistance.

SEM and EDS Analysis
SEM images and EDS analysis of the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution are shown in Figures 4 and 5. Localized corrosion occurred for all the samples. The weakest corrosion was observed in the sample annealed at 823 K, indicating the best corrosion resistance. Spots A, C, E, G, and J are in the smooth non-corroded area, in which the oxygen content was low, less than 20%, and the content of Zr, Cu, Ni, Al, and Nb was close to the original content. On the other hand, spots B, D, F, H, and K are in the corroded area, in which the oxygen content was higher, and the content of Zr, Cu, Ni, Al, and Nb was lower than the original content.
While annealing below Tg, structural relaxation led to reduced free volume, resulting in improved corrosion resistance [22,26]. When the annealing temperature was slightly above Tx, crystallization occurred, leading to the formation of nanocrystals in the amorphous matrix. The stability of the passivation film was increased, resulting in better corrosion resistance. When the annealing temperature was well above Tx, the size of the crystals increased. Corrosion was susceptible to occurring at the interface between the crystals and amorphous matrix due to the chemical inhomogeneity, leading to reduced corrosion resistance.

SEM and EDS Analysis
SEM images and EDS analysis of the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution are shown in Figures 4 and 5. Localized corrosion occurred for all the samples. The weakest corrosion was observed in the sample annealed at 823 K, indicating the best corrosion resistance. Spots A, C, E, G, and J are in the smooth non-corroded area, in which the oxygen content was low, less than 20%, and the content of Zr, Cu, Ni, Al, and Nb was close to the original content. On the other hand, spots B, D, F, H, and K are in the corroded area, in which the oxygen content was higher, and the content of Zr, Cu, Ni, Al, and Nb was lower than the original content. Figure 3 shows the optical microscopy images of the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution. Obvious localized corrosion occurred for the as-cast sample and the samples annealed at 623 K and 723 K. Minor corrosion was observed in the sample annealed at 923 K. Corrosion was not obvious for the sample annealed at 823 K, suggesting good corrosion resistance.

Optical Microscopy Observation
(a) as-cast (b) 623 K (c) 723 K (d) 823 K (e) 923 K Figure 3. The optical microscopy images of the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution.

SEM and EDS Analysis
SEM images and EDS analysis of the as-cast Zr56Cu19Ni11Al9Nb5 sample and the samples annealed at 623-923 K after potentiodynamic polarization tests in PBS solution are shown in Figures 4 and 5. Localized corrosion occurred for all the samples. The weakest corrosion was observed in the sample annealed at 823 K, indicating the best corrosion resistance. Spots A, C, E, G, and J are in the smooth non-corroded area, in which the oxygen content was low, less than 20%, and the content of Zr, Cu, Ni, Al, and Nb was close to the original content. On the other hand, spots B, D, F, H, and K are in the corroded area, in which the oxygen content was higher, and the content of Zr, Cu, Ni, Al, and Nb was lower than the original content.
While annealing below Tg, structural relaxation led to reduced free volume, resulting in improved corrosion resistance [22,26]. When the annealing temperature was slightly above Tx, crystallization occurred, leading to the formation of nanocrystals in the amorphous matrix. The stability of the passivation film was increased, resulting in better corrosion resistance. When the annealing temperature was well above Tx, the size of the crystals increased. Corrosion was susceptible to occurring at the interface between the crystals and amorphous matrix due to the chemical inhomogeneity, leading to reduced corrosion resistance.
(a) as-cast (b) 623 K (c) 723 K (d) 823 K (e) 923 K  While annealing below T g , structural relaxation led to reduced free volume, resulting in improved corrosion resistance [22,26]. When the annealing temperature was slightly above T x , crystallization occurred, leading to the formation of nanocrystals in the amorphous matrix. The stability of the passivation film was increased, resulting in better corrosion resistance. When the annealing temperature was well above T x , the size of the crystals increased. Corrosion was susceptible to occurring at the interface between the crystals and amorphous matrix due to the chemical inhomogeneity, leading to reduced corrosion resistance.    Figure 6 shows the XPS analysis of the Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample annealed at 723 K after the potentiodynamic polarization test in PBS solution. In the spectra, the peaks with binding energy of 182.58 eV and 184.88 eV represented Zr 4+ 3d 5/2 and Zr 4+ 3d 3 Nb 3d3/2. The peak at 530.95 eV suggested O . The corrosion products mainly consist of ZrO2, CuO, Cu2O, Ni(OH)2, Al2O3, and Nb2O5.

XPS Analysis
Relative to the potential of a standard hydrogen electrode (SHE), the electrode potential of Al/Al 3+ , Zr/Zr 4+ , Nb/Nb 5+ , Ni/Ni 2+ , Cu/Cu 2+ , and Cu/Cu + is −1.662 VSHE, −1.529 VSHE, −1.200 VSHE, −0.250 VSHE, 0.337 VVSH, and 0.521 VSHE. At lower potential, the metal element is more active and is easier to corrode. Al, Zr, and Nb are more active than Ni and Cu, and they are corroded first with the corrosion products of Al2O3, ZrO2, and Nb2O5. Ni and Cu are then corroded with the corrosion products of Ni(OH)2, Cu2O, and CuO. The corrosion products of the Zr56Cu19Ni11Al9Nb5 sample are mainly ZrO2, CuO, Cu2O, Ni(OH)2, Al2O3, and Nb2O5. Passivation occurred due to the formation of oxide film or the adsorption of oxygen atoms or oxygen ions. The Cl − ions in PBS solution, which contained 8.0 g/L NaCl, 0.20 g/L KCl, 1.44 g/L Na2HPO4 and 0.24 g/L KH2PO4, caused the localized corrosion of the as-cast and annealed Zr56Cu19Ni11Al9Nb5 samples. According to the oxidefilm theory, the Cl − ions caused the localized dissolution of the passivation film, leading to localized corrosion. According to the competitive adsorption theory, the localized preferential adsorption of Clions hindered the adsorption of oxygen atoms or oxygen ions, resulting in the localized corrosion.  Relative to the potential of a standard hydrogen electrode (SHE), the electrode potential of Al/Al 3+ , Zr/Zr 4+ , Nb/Nb 5+ , Ni/Ni 2+ , Cu/Cu 2+ , and Cu/Cu + is −1.662 V SHE , −1.529 V SHE , −1.200 V SHE , −0.250 V SHE , 0.337 V VSH , and 0.521 V SHE . At lower potential, the metal element is more active and is easier to corrode. Al, Zr, and Nb are more active than Ni and Cu, and they are corroded first with the corrosion products of Al 2 O 3 , ZrO 2 , and Nb 2 O 5 . Ni and Cu are then corroded with the corrosion products of Ni(OH) 2 , Cu 2 O, and CuO. The corrosion products of the Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample are mainly ZrO 2 , CuO, Cu 2 O, Ni(OH) 2 , Al 2 O 3 , and Nb 2 O 5 . Passivation occurred due to the formation of oxide film or the adsorption of oxygen atoms or oxygen ions. The Cl − ions in PBS solution, which contained 8.0 g/L NaCl, 0.20 g/L KCl, 1.44 g/L Na 2 HPO 4 and 0.24 g/L KH 2 PO 4 , caused the localized corrosion of the as-cast and annealed Zr 56 Cu 19 Ni 11 Al 9 Nb 5 samples. According to the oxide-film theory, the Cl − ions caused the localized dissolution of the passivation film, leading to localized corrosion. According to the competitive adsorption theory, the localized preferential adsorption of Cl − ions hindered the adsorption of oxygen atoms or oxygen ions, resulting in the localized corrosion.

Conclusions
Passivation occurred for the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 metallic glass and the samples annealed at 623 K (below T g ), 723 K (between T g and T x ), and 823 K (above T x ), indicating good corrosion resistance in PBS solution. Passivation did not occur for the sample annealed at 923 K (far above T x ). With the increase in annealing temperature, the corrosion resistance first increased, and then decreased. The sample annealed at 823 K exhibited high E corr at −0.045 V SCE , small i corr at 1.549 × 10 −5 A·cm −2 , high E pit at 0.165 V SCE , a wide passivation region E pit -E corr of 0.210 V, the largest arc radius, and the largest sum of R f and R ct , 5909 Ω·cm 2 , indicating the best corrosion resistance in PBS solution. For the sample annealed at 923 K, passivation did not occur, and the sample illustrated the highest i corr at 1.879 × 10 −5 A·cm −2 , the smallest arc radius, and the smallest sum of R f and R ct , 2173 Ω·cm 2 , indicating the worst corrosion resistance in PBS solution.
Optical microscopy, SEM, EDS, and XPS analysis showed that localized corrosion occurred for the as-cast Zr 56 Cu 19 Ni 11 Al 9 Nb 5 sample and the samples annealed at 623-923 K. In the non-corroded area, the content of Zr, Cu, Ni, Al, and Nb was close to the original content of the sample. In the corroded area, the content of Zr, Cu, Ni, Al, and Nb was lower than the original content. The main corrosion products are ZrO 2 , CuO, Cu 2 O, Ni(OH) 2 , Al 2 O 3 , and Nb 2 O 5 .
The proper annealing temperature led to improved corrosion resistance. When the annealing temperature was below T g , structural relaxation led to reduced free volume, resulting in improved corrosion resistance. When the annealing temperature was slightly above T x , crystallization started to occur, and the formation of nanocrystals in the amorphous matrix led to improved stability of the passivation film, resulting in better corrosion resistance. However, if the annealing temperature was well above T x , the size of the crystals increased, and the chemical inhomogeneity led to corrosion at the interface between the crystals and amorphous matrix, resulting in reduced corrosion resistance. Zr 56 Cu 19 Ni 11 Al 9 Nb 5 metallic glass and the samples annealed at the proper temperature are promising candidate materials for biomedical applications.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors. The data are not publicly available due to privacy reasons.