Effect of Annealing Time on Corrosion Behaviours of Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ in Hank Solution

Abstract

The microstructures of the as-cast and annealed Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ were investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM), their microhardness values were tested, and their corrosion behaviours in Hank solution were studied. XRD results and SEM analysis showed that the as-cast sample was amorphous, and crystallisation occurred in the samples annealed at 923 K for 5–30 min with crystals of Zr₂Cu and Zr₂Ni. Microhardness gradually increased and then levelled off, due to higher crystallisation degree with longer annealing time. Passivation occurred for all the samples in Hank solution. Prolonged annealing time leads to the initial rise and then a drop in corrosion resistance. Annealing for 5 min resulted in the highest corrosion resistance, with high corrosion potential E_corr at −0.007 V_SCE, versus saturated calomel electrode (SCE), i.e., 0.234 V_SHE, versus standard hydrogen electrode (SHE), the smallest corrosion current density i_corr at 2.20 × 10⁻⁷ A·cm⁻², the highest pitting potential E_pit at 0.415 V_SCE (i.e., 0.656 V_SHE), the largest passivation region E_pit–E_corr at 0.421 V_SHE, the largest arc radius, and the largest sum of charge transfer resistance and film resistance R_ct + R_f at 15489 Ω·cm². Annealing for 30 min led to the lowest corrosion resistance, with low E_corr at −0.069 V_SCE (i.e., 0.172 V_SHE), large i_corr at 1.32 × 10⁻⁶ A·cm⁻², low E_pit at −0.001 V_SCE (i.e., 0.240 V_SHE), small E_pit − E_corr at 0.068 V_SHE, the smallest arc radius, and the smallest R_ct + R_f at 4070 Ω·cm². When the annealing time was appropriate, the homogeneous microstructure of nanocrystals in an amorphous matrix resulted in improved passivation film, leading to the rise of corrosion resistance. However, if the annealing time was prolonged, the inhomogeneous microstructure of larger crystals in an amorphous matrix resulted in a drop in corrosion resistance. Localised corrosion was observed, with corrosion products of ZrO₂, Cu₂O, CuO, Ni(OH)₂, Al₂O₃, and Nb₂O₅.

1. Introduction

Bulk metallic glasses (BMGs) and bulk metallic glass composites (BMGCs) have been widely studied [1,2,3]. Zr-based BMGs showed promising biomedical applications due to high strength and hardness, good corrosion resistance, good glass forming ability, good biocompatibility, good wear resistance, and low elastic modulus [4,5]. Zr₆₁Cu_17.5Ni₁₀Al_7.5Si₄ exhibited good corrosion resistance in Hank solution, and good biocompatibility, indicating good potential for biomedical applications [6]. Corrosion resistance is affected by chemical composition. Proper addition of Nb and Ti led to better corrosion resistance [7,8]. Zr₅₅Al₂₀Co_25−xNb_x (x = 0, 2.5, 5) displayed good corrosion resistance in Hank solution and phosphate-buffered saline (PBS) solution, and the increase in Nb content led to better corrosion resistance and reduced water contact angle [7]. Zr_65−xTi_xCu₂₀Al₁₀Fe₅ (x = 4, 6) showed better corrosion resistance than Zr_65−xTi_xCu₂₀Al₁₀Fe₅ (x = 0, 2) in PBS solution [8]. Zr_60+xTi_2.5Al₁₀Fe_12.5−xCu₁₀Ag₅ (x = 0, 2.5, 5) exhibited good corrosion resistance in PBS solution, and higher Zr content resulted in better corrosion resistance due to higher Zr/Al ratio in the passive film [9]. Zr₄₆(Cu_4.5/5.5Ag_1/5.5)₄₆Al₈ showed better biocompatibility and better corrosion resistance than Zr_51.9Cu_23.3Ni_10.5Al_14.3 and Zr₅₁Cu₂₅Ni₁₀Al₉Ti₅ in Hank solution due to the amorphous structure and the formation of Al₂O_3-enriched passive film [10]. Zr₅₃Al₁₆Co₂₆Pt₅ showed the best corrosion resistance, followed by Zr₅₃Al₁₆Co₂₆Pd₅, and Zr₅₃Al₁₆Co₂₆Au₅ showed the worst corrosion resistance in PBS solution [11]. Zr_55.8Al_19.4(Co_1−xCu_x)_24.8 (x = 0–0.8) were prepared and Zr_55.8Al_19.4(Co_1−xCu_x)_24.8 (x = 0.3) illustrated good corrosion resistance in PBS solution, the largest 12 mm casting diameter, and good mechanical properties [12]. Zr_60.5Hf₃Al₉Fe_4.5Cu₂₃ and Zr_58.6Al_15.4Co_18.2Cu_7.8 illustrated good corrosion resistance in PBS solution, and good antibacterial properties [13,14]. Zr₄₀Ti₃₇Co₁₂Ni₁₁, Zr₅₀Ti₃₂Cu₁₃Ag₅, Zr₄₆Ti₄₀Ag₁₄ and Zr₄₆Ti₄₃Al₁₁ displayed better corrosion resistance than Ti and 316L steel in PBS solution, and Zr₄₆Ti₄₀Ag₁₄ showed better antibacterial properties than Zr₄₆Ti₄₃Al₁₁ [15,16]. Zr₄₅Ti₃₆Fe₁₁Al₈ exhibited better corrosion resistance than Ti in PBS solution due to the amorphous structure and the stable ZrO₂ and TiO₂ film [17]. Compared with Zr₆₅Cu₁₈Al₁₀Ni₇, Zr₅₅Cu₃₀Al₁₀Ni₅ displayed higher microhardness, higher E_corr and smaller E_pit − E_corr in PBS solution, and better wear resistance both in air and in PBS solution [18]. Zr₆₂Cu₂₂Al₁₀Fe₅Dy₁ showed amorphous structure, good corrosion resistance in PBS solution, artificial saliva solution (ASS), Hank solution, and artificial blood plasma (ABP) solution, as well as good biocompatibility [19]. Zr₃₇Co₃₄Cu₂₀Ti₉ displayed good corrosion resistance in PBS, ASS, ABP, and Hank solutions, and good biocompatibility [20].

The corrosion behaviours of Zr-based BMGs were influenced by annealing temperature and time. The as-cast Zr₅₆Co₂₈Al₁₆ sample and the samples annealed at 660 K (<T_g), 800 K and 973 K (>T_x) for 5 h displayed poorer corrosion resistance in Ringer’s solution and poorer biocompatibility with higher annealing temperature [21]. Zr₅₈Nb₃Cu₁₆Ni₁₃Al₁₀ samples were annealed at 523 K and 673 K (<T_g), as well as at 773 K and 873 K (>T_x), for 6 h, and with higher annealing temperature, the microhardness gradually increased, and the corrosion resistance in 1 mol/L H₂SO₄ solution at 333 K gradually decreased [22]. Zr₆₀Cu₂₀Ni₈Al₇Hf₃Ti₂ samples were annealed at 500 K and 600 K (<T_g) and remained amorphous, and with higher annealing temperature, the microhardness rose, and the corrosion resistance in 0.01 mol/L H₂SO₄ solution slightly dropped [23]. Zr_50.7Ni₂₈Cu₉Al_12.3 samples were annealed at 719 K (T_g~T_x), 768 K (>T_x), and 810 K for 30 min, and the structures were amorphous, nanocrystals of ZrO₂ in an amorphous matrix, and crystals of ZrO₂, Al₂Zr, Cu₁₀Zr₇, and CuZr₂ in an amorphous matrix, respectively, and the sample annealed at 768 K exhibited the highest corrosion resistance in 0.5 mol/L H₂SO₄, 1 mol/L NaCl and 1 mol/L HCl solutions [24]. Zr_41.2Cu_12.5Ni₁₀Ti_13.8Be_22.5 and Zr₅₇Cu_15.4Ni_12.6Al₁₀Nb₅ samples annealed at 0.9T_g for 4 h showed better corrosion resistance in NaCl solution than the as-cast sample, due to the decreased free volume [25]. Zr₆₈Al₈Ni₈Cu₁₆ samples annealed at 673 K and 713 K displayed crystals of Zr₂Ni and Zr₂Cu in the amorphous matrix with a crystallinity of 10% and 77%, and the sample annealed at 713 K exhibited the maximum microhardness, the lowest strength, the smallest plasticity, and the worst corrosion resistance in 1 mol/L HCl solution [26]. The Zr₆₅Cu₁₅Ni_12.5Al_7.5 samples were annealed at 643 K, 663 K and 683 K (T_g~T_x) for 10 min, and with higher annealing temperature, the plasticity gradually decreased, and the corrosion resistance in 3.5% NaCl solution gradually decreased, and the cast sample showed higher corrosion resistance than the annealed samples [27]. The fully crystallised Zr₄₈Cu_46.5Al₄Nb_1.5 sample obtained by flash-annealing showed better corrosion resistance than the as-cast Zr₄₈Cu_46.5Al₄Nb_1.5 sample and Zr_47.5Cu_47.5Al₅ sample in 3.5% NaCl solution and 0.05 mol/L H₂SO₄ solution [28]. Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ as-cast sample and samples annealed at 623 K (<T_g), 723 K (T_g~T_x), 823 K and 923 K (>T_x) for 30 min displayed good corrosion resistance in PBS solution, and at higher annealing temperature, the corrosion resistance firstly rose and then dropped, with the best corrosion resistance for the sample annealed at 823 K and the worst corrosion resistance for the sample annealed at 923 K [29].

Zr₆₀Cu₂₀Al₁₀Fe₅Ti₅ as-cast sample and the samples annealed at 648 K (slightly < T_g) for 1 min and 5 min showed amorphous structure and the sample annealed at 873 K (>T_x) for 1 min displayed crystalline structure with crystals of Zr₂Cu and Al₂Zr₃. They exhibited gradually increased hardness, higher E_corr and better corrosion resistance in Hank solution, due to structural relaxation, reduced free volume and increased crystallisation [30]. The Zr₆₅Cu_17.5Fe₁₀Al_7.5 samples were annealed at 573 K (<T_g) for 0.5–4 h, and with prolonged annealing time, the microhardness firstly rose and then dropped, and the sample annealed at 573 K for 1 h showed the maximum microhardness of 487 HV, the highest plasticity of 7.1%, and the highest corrosion resistance in 3.5% NaCl solution due to the decreased free volume [31]. Zr₃₅Ti₃₀Be_26.75Cu_8.25 samples were annealed at 683 K (T_g~T_x) for 5 min and 12 min, and the corrosion resistance in 3.5% NaCl solution decreased, and the corrosion resistance increased after holding at 77 K for 30 min, and decreased after cryogenic cycling [32]. The as-cast Zr_50.7Cu₂₈Ni₉Al_12.3 sample and the samples annealed at 736 K (T_g~T_x) for 140, 164, and 193 min displayed amorphous structure and crystals in the amorphous matrix with the volume fraction of the crystalline phases of 14%, 40%, and 100%, and the sample annealed for 140 min exhibited the best corrosion resistance in the simulated groundwater, and with longer annealing time, the corrosion resistance gradually dropped [33]. The as-cast Zr₅₉Ti₆Cu_17.5Fe₁₀Al_7.5 sample and the samples annealed at 573 K (<T_g) for 0.5–2 h displayed amorphous structure, and the sample annealed at 573 K for 4 h exhibited crystals of Al₃Zr₂ in amorphous matrix, and with prolonged annealing time, the strength and plasticity firstly rose and then dropped, with the highest strength and the largest plasticity for the sample annealed at 573 K for 0.5 h, and the corrosion resistance of all the samples remained good in PBS solution [34].

The effects of microstructure, annealing temperature and time displayed complex effects on the corrosion resistance of Zr-based BMGs and BMGCs in simulated bio-environments, and further study is necessary. In this work, the microstructures of as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ metallic glass and the samples annealed at 923 K (>T_x) for 5–30 min were studied by X-ray diffraction (XRD) and scanning electron microscopy (SEM), their microhardness values were tested, and their corrosion behaviours in Hank solution were studied. Our work helps the further understanding of corrosion mechanisms of Zr-based metallic glass and composites and further research and development in biomedical applications.

2. Materials and Methods

2.1. Material Preparation

Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ samples were obtained through arc-melting of pure Zr, Cu, Ni, Al and Nb [29]. The samples were cut to 5 mm × 4 mm × 1 mm, as shown in Figure 1, and were annealed at 923 K (far above T_x) for 5, 8, 10, 15, and 30 min. The surfaces were ground using 800–1500 grit sandpaper, polished using 2.5 and 0.5 µm diamond paste, and then cleaned in acetone. The surface roughness of the sample was determined by Dimension IconIR atomic force microscopy (Bruker, MA, USA), as shown in Figure 2, with R_a around 0.007 μm and R_q around 0.014 μm.

2.2. Tests

XRD analysis was performed using a D8 Advance X-ray diffractor (Bruker, Karlsruhe, Germany) with a diffraction angle 2θ in the range of 20–90°, and an SEM image was obtained using Zeiss Ultra Plus field emission scanning electron microscopy (Zeiss, Jena, Germany) with the voltage of 5 kV, in order to investigate the microstructures of the as-cast sample and the samples annealed at 923 K for 5–30 min. The amorphous-to-crystalline ratio is equal to the area of the amorphous structure divided by the area of the crystalline structure. Vickers microhardness measurements were carried out using a MICRO-586 microhardness tester (Shandong Shancai Testing Instrument Co., Ltd., Yantai, Shandong, China) with a load of 2 N and a holding time of 10 s, and the measurements were repeated 10 times for each sample. Potentiodynamic polarisation (PP) tests and electrochemical impedance spectroscopy (EIS) tests of the as-cast and annealed samples in Hank solution were carried out through the CHI 660E electrochemical station (Shanghai CH Instruments, Shanghai, China) [29]. The composition of Hank solution is as follows, 0.14 g/L CaCl₂, 0.1 g/L MgCl₂·6H₂O, 0.1 g/L MgSO₄·7H₂O, 0.35 g/L NaHCO₃, 1 g/L D-glucose, 0.01 g/L phenol red. The saturated calomel electrode (SCE) was used as the reference electrode, and the graphite electrode was used as the counter electrode. The electrodes were stabilised in Hank solution at open circuit potential (OCP) for 60 min. In PP tests, the potential was scanned from −0.8 V_SCE (i.e., −0.559 V_SHE) to 0.5 V_SCE (i.e., 0.741 V_SHE) at 0.33 mV/s. In EIS tests, the complex impedance was obtained at OCP, with the frequency of 10⁻²–10⁵ Hz and the amplitude of 5 mV. Zsimpwin 3.6 software was used to fit the EIS data through the appropriate equivalent circuit (EC). After PP and EIS tests, the corrosion morphology and products were analysed by SH11/YF-III optical microscopy (OM, Shanghai Optical Instrument Factory, Shanghai, China), Zeiss Ultra Plus field emission SEM (Zeiss, Jena, Germany), X-Max 50 X energy dispersive X-ray spectroscopy (EDS, Oxford Instruments, Abingdon, UK) and Thermo Scientific ESCALAB 250Xi X-ray photoelectron spectroscopy (XPS, Thermofisher Scientific, Waltham, MA, USA) [29]. The measurement accuracy of the equipment is shown in

Table 1. List of equipment and the measurement accuracy.

Equipment	Model	Parameter	Accuracy
X-ray diffractor	D8 Advance	angle	10−2°
Electrochemical workstation	CHI 660E	potential	10−4 V
Electrochemical workstation	CHI 660E	current	10−8 A
Microhardness tester	MICRO-586	microhardness	10−1 MPa

.

3. Results and Discussion

3.1. Microstructure Analysis

The XRD patterns and SEM images of the as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample and the samples annealed at 923 K for 5–30 min are shown in Figure 3 and Figure 4. The as-cast sample shows an amorphous structure, and crystallisation occurred in the samples annealed at 923 K for 5–30 min with crystals of Zr₂Cu and Zr₂Ni. With longer annealing time, the intensity of the diffraction peaks gradually increased, and the crystallisation degree gradually increased, with crystallisation percentages of 0, 17%, 18%, 27%, 27% and 85% for the as-cast sample and the samples annealed at 923 K for 5 min, 8 min, 10 min, 15 min and 30 min, respectively. SEM images display the amorphous structure of the as-cast sample and microstructures of nanocrystals in the amorphous matrix of the annealed samples. With prolonged annealing time, the number of crystals firstly increased and then decreased, and the size of crystals gradually increased, due to the increase in crystallisation degree and the growth and coalescence of crystals. The amorphous-to-crystalline ratio was 4.9, 4.6, 2.7, 2.7 and 0.2 for the samples annealed at 923 K for 5 min, 8 min, 10 min, 15 min and 30 min, respectively.

3.2. Microhardness

Figure 5 shows the microhardness of the as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample and the samples annealed at 923 K for 5–30 min. With a longer annealing time, the microhardness gradually rose from 4.89 GPa (i.e., 499 Hv) to 6.37 GPa (i.e., 650 Hv) and then levelled off. The brittleness gradually increased, reaching the maximum, and then slightly dropped. It is due to the increase in crystallisation degree with prolonged annealing time, as shown by XRD analysis and SEM images. The interface between the crystals and the amorphous matrix and the grain boundaries hindered the deformation, leading to the increase in microhardness. A similar trend was reported for Zr₆₅Cu_17.5Fe₁₀Al_7.5 samples annealed at 573 K (<T_g) for 0.5–4 h, and the microhardness increased from 422 Hv for the as-cast sample to 487 Hv for the sample annealed at 573 K for 1 h and then dropped to 444 Hv for the sample annealed at 573 K for 4 h [31].

3.3. Corrosion Behaviours

Figure 6 shows the PP curves for the as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample and the samples annealed at 923 K for 5–30 min in Hank solution, and

Table 2. PP results of Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ as-cast sample and samples annealed at 923 K for 5–30 min in Hank solution.

Annealing Time (min)	Ecorr (VSHE)	icorr (A·cm−2)	Epit (VSHE)	ipit (A/cm2)	Epit − Ecorr (VSHE)
0	0.239	3.14 × 10−7	0.484	1.67 × 10−6	0.245
5	0.234	2.20 × 10−7	0.656	2.81 × 10−6	0.421
8	0.240	4.40 × 10−7	0.584	3.25 × 10−7	0.344
10	0.215	5.75 × 10−7	0.352	1.64 × 10−6	0.138
15	0.128	1.05 × 10−6	0.145	5.11 × 10−6	0.017
30	0.172	1.32 × 10−6	0.240	8.09 × 10−6	0.068

summarises the obtained electrochemical parameters. Figure 7 illustrates the change of corrosion potential E_corr, corrosion current density i_corr, pitting potential E_pit, pitting current density i_pit with annealing time. For the as-cast sample and the samples annealed at 923 K for 5–10 min, E_pit was determined from the kink where the current density increased sharply. For the samples annealed at 923 K for 15 and 30 min, E_pit was determined from the intersection point of the tangent lines where the current density increased dramatically. Passivation occurred for all the samples, indicating good corrosion resistance in Hank solution. With longer annealing time, E_corr and E_pit first rose and then dropped, i_corr and i_pit first dropped and then rose, and the corrosion resistance first rose and then dropped. The sample annealed at 923 K for 5 min exhibited the best corrosion resistance, with high E_corr at −0.007 V_SCE (i.e., 0.234 V_SHE), the smallest i_corr at 2.20 × 10⁻⁷ A·cm⁻², the highest E_pit at 0.415 V_SCE (i.e., 0.656 V_SHE), and the largest passivation region E_pit − E_corr at 0.421 V_SHE. The sample annealed at 923 K for 30 min exhibited the worst corrosion resistance, with low E_corr at −0.069 V_SCE (i.e., 0.172 V_SHE), large i_corr at 1.32 × 10⁻⁶ A·cm⁻², low E_pit at −0.001 V_SCE (i.e., 0.240 V_SHE) and small E_pit − E_corr at 0.068 V_SHE. Figure 8 illustrates the Nyquist plots, Bode plots and the equivalent circuit diagram for the EIS results of the as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample and the samples annealed at 923 K for 5–30 min in Hank solution, and

Table 3. EIS analysis results for as-cast and annealed samples in Hank solution.

Annealing Time (min)	EIS Fitting Results
	Rs (Ω·cm2)	Y01 (Ω−1·sn·/cm2)	n1	Rf (Ω·cm2)	Y02 (Ω−1·sn·/cm2)	n2	Rct (Ω·cm2)	Rct + Rf (Ω·cm2)
0	11	1.05 × 10−4	0.94	4444	2.68 × 10−4	0.64	46	4490
5	11	1.19 × 10−4	0.89	15,460	1.92 × 10−4	0.71	29	15,489
8	11	1.19 × 10−4	0.90	13,300	1.55 × 10−4	0.72	44	13,344
10	12	1.02 × 10−4	0.88	12,050	1.21 × 10−4	0.75	37	12,087
15	16	1.15 × 10−4	0.79	8053	2.60 × 10−5	0.89	23	8076
30	17	9.17 × 10−5	0.69	47	9.24 × 10−5	0.87	4023	4070

summarises the obtained electrochemical parameters.

The Nyquist plots displayed a half-circle, indicating that the control step was the electrochemical reaction accompanied by the transfer of electrons. The Bode plots showed two time constants, with frequencies around 3 Hz and 1000 Hz. The maximum phase was reached around 3 Hz. The equivalent circuit diagram as shown in Figure 8c was previously used for the fitting of EIS results of Zr₄₅Ti₃₆Fe₁₁Al₈ in PBS solution [16], Zr₅₈Nb₃Cu₁₆Ni₁₃Al₁₀ as-cast sample and samples annealed at 523 K, 673 K, 773 K and 873 K in 1 mol/L H₂SO₄ solution at 333 K [21], Zr₅₂Cu₃₂Al₁₀Ni₆ in 0.05–0.5 mol/L NaF solutions [35], Al₈₆Ni₉Y₅ as-spun sample and the samples annealed at 423 K for 5 min and 30 min, at 543 K for 5 min, and at 673 K for 5 min in 0.6 mol/L NaCl [36]. The passivation film contains defects. R_s represents the solution resistance between the working electrode and the reference electrode. R_f represents the film resistance, and R_ct represents the charge transfer resistance at the interface between the solution and the film. CPE₁ and CPE₂ represent the constant phase element (CPE) of the film and the capacitance of the double layer between the solution and the film. The impedance $\mathrm{Z}={Y_0}^{-1}{\omega }^{-n}\left[\cos \left({\displaystyle \frac{n\pi }{2}}\right)-j\sin \left({\displaystyle \frac{n\pi }{2}}\right)\right]$. Y₀ is the constant, ω is the angular frequency, and n is the parameter between 0 and 1. Figure 9 shows the change of the sum of charge transfer resistance and film resistance R_ct + R_f with annealing time. With the increase in annealing time, the arc radius gradually increased and then decreased, R_ct + R_f gradually decreased and then increased, indicating that the corrosion resistance in Hank solution first increased and then decreased. The sample annealed at 923 K for 5 min showed the largest arc radius and the largest R_ct + R_f, 15,489 Ω·cm², indicating the best corrosion resistance. The sample annealed at 923 K for 30 min displayed the smallest arc radius and the smallest R_ct + R_f, 4070 Ω·cm², indicating the worst corrosion resistance.

Figure 10 and Figure 11 exhibit OM images, SEM images and EDS analysis of the as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample and the samples annealed at 923 K for 5–30 min after PP tests in Hank solution. Pitting was observed in all the samples. Spots A, C, E, G, L and N are located in the non-corroded area, with a low oxygen content of 11–48 at.%, and concentrations of Zr, Cu, Ni, Al and Nb of 24–47 at.%, 10–15 at.%, 6–14 at.%, 7–9 at.%, 4–5 at.%, which are similar to the original 56 at.%, 19 at.%, 11 at.%, 9 at.%, and 5 at.%, respectively. Spots B, D, F, M and O are located in the corroded area, with a higher oxygen content of 25–66 at.% and concentrations of Zr, Cu, Ni, Al and Nb of 10–37 at.%, 9–25 at.%, 1–12 at.%, 2–10 at.%, 1–6 at.%, respectively, much lower than the original concentrations. The decrease in Zr content is the largest.

The as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample exhibited an amorphous structure, resulting in good corrosion resistance in Hank solution. The annealed sample at 923 K (>T_x) for 5 min displayed a homogeneous microstructure of nanocrystals of Zr₂Cu and Zr₂Ni in the amorphous matrix, as confirmed by XRD analysis and SEM image. The improved stability of the passivation film led to higher corrosion resistance. With a longer annealing time, the crystallisation content rose. The microstructure became inhomogeneous, with larger crystals of Zr₂Cu and Zr₂Ni in the amorphous matrix, leading to a drop in corrosion resistance.

Figure 12 shows the XPS analysis of the Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ as-cast sample and the samples annealed at 923 K for 5–30 min after EIS tests in Hank solution. The experimental data, the base line, the fitted peaks and the sum of fitted peaks were illustrated in different colours. The peaks at 182.2 eV and 184.6 eV exhibited Zr⁴⁺ 3d_5/2 and Zr⁴⁺ 3d_3/2 [3,29,37]. The peaks at 932.0 eV and 952.0 eV represented Cu⁺ 2p_3/2 and Cu⁺ 2p_1/2 [3,29,37]. The peaks at 75.6 eV and 77.6 eV indicated Cu²⁺ 2p_3/2 and Cu²⁺ 2p_1/2 [3,29,37]. The peaks at 851.9 eV and 856.0 eV indicated Ni⁰ 2p_3/2 and Ni²⁺ 2p_3/2, and the peaks at 869.5 eV and 873.5 eV indicated Ni⁰ 2p_1/2 and Ni²⁺ 2p_1/2 [3,29]. The peak at 74.0 eV represented Al³⁺ 2p [29,37]. The peaks at 207.0 eV and 209.8 eV indicated Nb⁵⁺ 3d_3/2 and Nb⁵⁺ 3d_1/2 [3,29]. The peaks at 529.8 eV and 531.3 eV represented O²⁻ and OH⁻ [3,29,37]. Therefore, the corrosion products consist of ZrO₂, CuO, Cu₂O, Ni(OH)₂, Al₂O_3, and Nb₂O₅.

The electrode potential of Al/Al³⁺, Zr/Zr⁴⁺, Nb/Nb⁵⁺, Ni/Ni²⁺, Cu/Cu²⁺, and Cu/Cu⁺ is as follows, −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, respectively. Al, Zr, and Nb are more active than Ni and Cu, due to lower potential, and the following corrosion reactions occur first. 2Al + 3H₂O → Al₂O₃ + 3H₂, Zr + 2H₂O → ZrO₂ + 2H₂, 2Nb + 5H₂O → Nb₂O₅ + 5H₂. The corrosion of Ni and Cu occurs next, with higher potential. Ni + 2H₂O → Ni(OH)₂ + H₂, Cu + H₂O → CuO + H₂, 2Cu + H₂O → Cu₂O + H₂.

Larsson et al. reported that for the Zr_59.3Cu_28.8Al_10.4Nb_1.5 sample manufactured by selective laser melting, the ion release under both simulated physiological and inflammatory conditions is as follows, Zr > Al > Cu > Nb due to the lower potential of Zr and Al, and the ion release under inflammatory condition is higher than that under simulated physiological condition [38]. Zr is a biocompatible element, and Al and Cu are trace elements present in the human body. Ni is a toxic element, which may cause an allergy response. Large ion release higher than the legal limits may cause health problems [39,40,41].

4. Conclusions

The as-cast Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample showed an amorphous structure, and the samples annealed at 923 K for 5–30 min showed crystals of Zr₂Cu and Zr₂Ni in the amorphous matrix. With prolonged annealing time, the crystallisation degree gradually increased, and the microhardness gradually increased and then levelled off. Passivation was observed for all the samples in Hank solution, indicating good corrosion resistance. With a longer annealing time, the corrosion resistance first rose and then dropped. The annealed sample at 923 K for 5 min displayed the highest corrosion resistance, with high E_corr at −0.007 V_SCE (i.e., 0.234 V_SHE), the smallest i_corr at 2.20 × 10⁻⁷ A·cm⁻², the highest E_pit at 0.415 V_SCE (i.e., 0.656 V_SHE), the largest E_pit − E_corr at 0.421 V_SHE, the largest arc radius, and the largest R_ct + R_f at 15,489 Ω·cm². The annealed sample at 923 K for 30 min exhibited the lowest corrosion resistance, with low E_corr at −0.069 V_SCE (i.e., 0.172 V_SHE), large i_corr at 1.32 × 10⁻⁶ A·cm⁻² (i.e., 500% increase), low E_pit at −0.001 V_SCE (i.e., 0.240 V_SHE), small E_pit − E_corr at 0.068 V_SHE (i.e., 84% decrease), the smallest arc radius, and the smallest R_ct + R_f at 4070 Ω·cm² (i.e., 74% decrease). When the annealing time was appropriate, the microstructure was homogeneous with nanocrystals in an amorphous matrix, resulting in improved passivation film and the rise of corrosion resistance. However, if the annealing time was prolonged, the microstructure became inhomogeneous, with larger crystals in an amorphous matrix, leading to a drop in corrosion resistance. Localised corrosion was observed, with corrosion products of ZrO₂, Cu₂O, CuO, Ni(OH)₂, Al₂O₃, and Nb₂O₅. Zr₅₆Cu₁₉Ni₁₁Al₉Nb₅ sample annealed at 923 K for 5 min exhibited potential candidate for biomedical applications due to high hardness and good corrosion resistance in Hank solution. In the future, in vivo tests can be performed to investigate the biocompatibility.