In vitro biomedical corrosion and enzyme activity inhibition on modified Cu-Zn-Al bioalloy

2.1 Materials and chemicals

The shape memory alloy with a composition of Cu-25.38Zn-3.3Al (wt.%) was manufactured by melting high-purity copper, Cu-Zn, and Cu-Al pre-alloys in a graphite crucible in the resistance-heated furnace. A charcoal cover was utilized to prevent copper’s oxygen absorption and melt loss. The molten alloy was poured at 700 °C into preheated (120 °C) graphite mold of dimensions 14 mm × 65 mm × 100 mm. After machining, the cast ingot was cut into cylindrical samples with a 35 mm diameter which were subjected to the homogenization treatment at 850 °C for 2 h, followed by quenching into a water bath at room temperature. The homogenized specimens were hot forged at 830 °C after heating for 1 h, resulting in plates with a thickness of 4.0 mm. The multi-pass cold rolling was accompanied by intermediate annealing treatments at 500 °C for 20 min, followed by air cooling to produce the test specimens with a final thickness of 0.3 mm. Rolled sheets were quenched in a water bath at 17 °C ± 2 °C, after annealing at 890 °C for 10 min.

For designing the biosensor and its testing, the following were used: Enzyme catalase (CAT) (c100-50 MG; Sigma-Aldrich, Buchs, Switzerland); KH₂PO₄ and Na₂HPO₄ (Fisher Chemical, Wien, Austria); hydrogen peroxide (H₂O₂) p.a. 30% (Sigma-Aldrich, Buchs, Switzerland); l-cysteine (Sigma Aldrich, EC 200-158-2); Nafion 5% (Sigma-Aldrich, Buchs, Switzerland). Halogenated boroxine dipotassium-trioxohydroxytetrafluorotriborate K₂[B₃O₃F₄OH] as an inorganic boron derivative has been synthesized by the reaction of potassium hydrofluoride (KHF₂) with boric acid in a molar ratio of 2:3 at room temperature as described in the literature (Ryss and Slutskaya 1951).

2.2 Characterization techniques

The microstructures of the as-cast and homogenized specimens, as well as rolled and quenched plates, were analyzed with a Zeiss optical microscope. The samples were cut and cold-mounted in epoxy resin, followed by grinding using progressively finer silicon carbide (SiC) grinding papers. Samples were polished and etched using a solution of 5 g FeCl₃ + 30 mL HCl + 100 mL H₂O, which was used to reveal the microstructure of the cast and homogenized samples. The rolled and quenched specimens were electropolished (voltage: from 20 V to 70 V, solution: 300 mL HNO₃ and 600 mL CH₃OH, time: 10 s–60 s; cathode of stainless steel) and then etched with a solution of 5 g FeCl₃ + 30 mL HCl + 100 mL H₂O.

An X-ray diffractometer using Cu-Kα radiation examined the structural properties of quenched samples. The values of the angle φ between lines linking the closest neighbors in the 18R martensite’s basal plane were determined as follows:

where a and b are the lattice parameters of the monoclinic M18R martensite.

Using the four-point probe resistance measurement technique, a continuous, direct current of 1.5 A was supplied to the specimen (130 mm × 4 mm × 0.3 mm), and the resulting voltage drop was measured. The martensitic and reverse transformation temperatures were determined based on the change in electrical resistivity with temperature during cooling and heating cycles. Continuous acquisition and data processing were carried out using an amplifier, A/D converter, and relevant software executed on a personal computer. The data processing was accomplished using the original software application. Both applications leveraged the programming and numerical computing platform MATLAB.

At room temperature, mechanical testing of quenched sheets was conducted using an FPZ HECKERT RAUENSTEIN 100/1 standard testing machine until fracture. Five repeat tensile tests were performed to guarantee the reliability of data obtained for the yield stress, tensile strength, and elongation.

Tensile testing was used to investigate shape memory recovery until the tensile strain reached 2.9%, at which point the load was removed. Following unloading, the specimens were heated to 50 °C above A_f temperature. Before and after heating the samples, the shape recovery ratio was determined by measuring the distance between two bars within the gauge range.

2.3 Catalase immobilization

An amperometric biosensor for determining the H₂O₂ was formed by immobilizing catalase on the sheet surface produced from Cu-25.38Zn-3.3Al alloy using a Nafion matrix (Figure 1) as described in the literature (Herenda et al. 2018; Ostojic et al. 2017).

Figure 1:

Schematic representation of an amperometric biosensor for the detection of H₂O₂.

2.4 Electrochemical measurements

The characterization of examined samples was based on the measurements performed on a PAR 263A potentiostat/galvanostat with a conventional three-electrode system. The reference electrode was a saturated Ag/AgCl electrode with a Pt counter, while a sample of Cu-25.38Zn-3.3Al bioalloy was used as a working electrode. The following electrochemical methods were carried out for the measurements: cyclic voltammetry, linear polarization (Tafel), chronoamperometry, and impedance spectroscopy. Cyclic voltammetry was used to examine the immobilization, the effect of different substrate concentrations on enzyme activity, and the thickness of the enzyme film on the electrode surface. All cyclic voltammetry tests were conducted in the phosphate buffer (pH 7) in the potential range of −1.2 V–0.7 V versus Ag/AgCl and at a scan rate of 50 mV/s. In addition, cyclic voltammetry was used to evaluate the effects of different scan rates from 10 mV/s to 90 mV/s. The success of enzyme immobilization on a Cu-25.38Zn-3.3Al bioalloy was examined by linear polarization and impedance spectroscopy. The chronoamperometric technique was implemented for the determination of kinetic parameters, the Michaelis-Menten constant (K_m) and the maximum current under saturated substrate conditions (I_max), which is equivalent to the maximum reaction rate (V_max) and resolve the type of inhibition (Adeyoju 1995). Chronoamperometric measurements were performed in the electrochemical cell containing 25 mL of the phosphate buffer solution at a constant potential of 0.9 V applied to the working electrode and at the constant stirring rate of 400 rpm. The reaction was monitored in different conditions, i.e., in the absence of l-cysteine and boroxine and in different concentrations of l-cysteine and boroxine.

3.1 Optical microscope observation and X-ray diffraction analysis

The microstructures of as-cast and homogenized specimens and rolled and quenched sheets of Cu-25.38Zn-3.3Al bioalloy are shown in Figure 2. The light needles of α-solid solution were observed in the β-phase matrix, while a small number of dendrites were observed in the as-cast samples (Figure 2a). In order to improve the technological plasticity and stability of the mechanical properties as well as reduce the directionality, the bioalloy was homogenized at 850 °C for 2 h and cooled into a water bath at room temperature. A microstructural study of homogenized samples indicated α-phase as the dominating phase and a small amount of martensite (Figure 2b). In terms of future plastic deformation, the homogeneity of the structure is superior to that of the as-cast samples. In light of the fact that one of the needed features of Cu-Zn-Al alloys is a fine grain structure that increases grain boundary strength (Tadaki 1998), cold working was selected as the most effective approach for refining the grain size. The samples were subjected to cold rolling and intermediate annealing treatment at 500 °C for 20 min, followed by air cooling. The intermediate annealing was required after every two or three passes to produce samples with a flat, shiny surface and uniform thickness below 1 mm and avoid crack occurrence and grain coarsening. After annealing, the specimens were cooled in the air to eliminate the internal stresses generated during cold deformation passes. The resulting sheets exhibited a dual-phase microstructure consisting of a minor quantity of β-phase combined with a workable α-phase (Figure 2c). Cold rolling resulted in the elongation of the grains.

Figure 2:

Optical micrograph of the (a) as-cast sample, (b) homogenized specimen, (c) final cold-rolled sheet, and (d) quenched sheet of bioalloy Cu-25.38Zn-3.3Al.

After solution treating at 890 °C for 10 min and quenching into a water bath at 17 °C ± 2 °C, blocks of parallel-side martensite plates were formed (Figure 2d). Besides 94.7% of martensite, the quenched sample exhibited approximately 5.3% of the β-phase with DO₃ structure (a = 0.5852 nm). The M18 R-type martensite lattice parameters are as follows: a = 0.4464 nm, b = 0.5273 nm, c = 3.7077 nm, and β = 88.35°, while φ = 61.13°. The 2H-type martensite has lattice parameters of a = 0.3166 nm and b = 0.5177 nm.

3.2 The phase transformation temperatures

By monitoring the change of electric resistance with temperature, transformation temperatures of quenched samples were found. The martensitic transition in the rolled and quenched samples began around 257 °C (M_s) and ended at 202 °C (M_f). The quench rate, solution treatment duration, and temperature all substantially impacted the martensitic transformation in the samples studied. The reverse transition of martensite was seen at temperatures ranging from 223 °C (A_s) to 316 °C (A_f temperature).

3.3 Mechanical behavior and shape memory recovery

Figure 3 shows the stress-strain curve of a quenched sample of the bioalloy Cu-25.38Zn-3.3Al. After the initial linear part caused by martensite phase elastic deformation, the stress plateau may be recognized, implying that the martensite reorientation process is at a strain of 0.7%–4.7%. Deformations that occur with subsequent stress application are related to the permanent deformation of the martensite.

Figure 3:

Stress-strain curve for Cu-25.38Zn-3.3Al bioalloy.

Tension along the rolling direction till fracture was evaluated on five quenched specimens. The average values of the five measurements were used to calculate the yield stress, tensile strength, and elongation. The yield stress was determined to be 222.0 MPa ± 9.5 MPa, and the ultimate tensile strength to be 481.0 MPa ± 7.8 MPa. The elongation was determined to be 7% ± 1%.

The shape recovery curve shown in Figure 3 indicates a recoverable strain of up to 2.9%, indicating that such an essential thermomechanical response of Cu-25.38Zn-3.3Al shape memory alloy is an unusual deformation associated with the plateau area. Deformation of about 2.9% recovered between 223 °C (A_s) and 316 °C (A_f) temperatures after applied stress has been removed, and samples were heated at 370 °C (temperature which was for 50 °C above A_f temperature). Since martensite plates are the only ones that can recover their original configuration by retracing the transformation path during the reverse transformation, as suggested by Ghosh et al. (1986), martensite plates in the microstructure of the quenched sample (Figure 2d) that are oriented favorably concerning the stress axis are primarily responsible for shape recovery.

3.4 Electrochemical characterization

The cyclic voltammograms of a Cu-25.38Zn-3.3Al bioalloy and catalase-modified Cu-25.38Zn-3.3Al bioalloy in phosphate buffer pH 7 at 50 mV/s are shown in Figure 4. The electrochemical response and redox potential formation on the catalase-modified bioalloy may be recognized (E_pc = −0.52 V and E_ac = 0.05 V). The redox activity depends on the pH and the electrolyte solution composition (Mažeikiene et al. 2003). Different film components causing the potential shifts may interact with the catalase or affect the electrode’s electric double-layer (Salimi et al. 2007).

Figure 4:

Cyclic voltammograms of bioalloy samples in phosphate buffer solution pH 7 at a scan rate of 50 mV/s: (a) without immobilized enzyme, (b) with immobilized enzyme.

The effect of the scan rate in the phosphate buffer solution in the absence of oxygen was investigated to obtain the catalase kinetic parameters of on the bioalloy film. Figure 5 shows cyclic voltammograms recorded at various scan rates. A linear relationship (R² = 0.9698) can be seen in the plot of peak currents versus scan rates (Figure 6), implying that catalase adsorbed on the surface undergoes quasi-reversible electron transfer from the surface of the bioalloy and saline solution. ΔE_P was increased with increasing scan rate. The peak-to-peak potential separation values were proportional to the square root of the scan rate. By using the Laviron method for a typical thin-layer electrochemical system (Laviron 1979a,b), the heterogeneous electron transfer rate constant (k_s) of the catalase immobilized on the electrode (bioalloy) was estimated. The transfer coefficient and heterogeneous electron transfer rate constant of catalase were about 0.45 and 3.7 ± 0.1 s⁻¹, respectively. Results indicate that the transfer of catalase electrons to the bioalloy is an easy and uninterrupted process.

Figure 5:

Cyclic voltammograms of bioalloy in the phosphate buffer at different scan rates.

Figure 6:

The plot of anodic peak currents versus square root of scan rate.

The cyclic voltammograms for the reduction of hydrogen peroxide on the catalase-modified bioalloy electrode at different concentration ranges are shown in Figure 7. The proportionality of the catalytic peak currents with the concentration of hydrogen peroxide was found with a linear range from 0.3 mM to 3 mM (Figure 8), for which the linear regression equation may be written as follows:

Figure 7:

Cyclic voltammograms of immobilized bioalloy in the presence of different substrate concentrations of H₂O₂ at a scan rate of 50 mV/s.

Figure 8:

The catalytic currents response versus hydrogen peroxide concentrations.

Cyclic voltammetry response for all, including higher hydrogen peroxide concentration, showed a leveling-off tendency of the typical Michaelis-Manten process of formation of the enzyme-substrate complex (Stryer 1988).

In order to confirm the success of the immobilization of catalase on the bioalloy, we have also used the impedance spectroscopy technique and linear polarization resistance measurements. The oxidation-reduction processes were best observed on the bioalloy at the potential of 0.9 V during our preliminary research. Therefore, electrochemical impedance diagrams were recorded at a constant voltage of 0.9 V. A frequency ranged from 10,000 Hz–0.01 Hz with a 5 mV amplitude sine wave. The Nyquist diagram in Figure 9 represents the obtained results of the electrochemical impedance spectroscopy of bioalloy in the absence and presence of the immobilized enzyme. The initial parts of the curves are not clearly visible and experimentally recorded, which indicates that they correspond to very high frequencies (above 10,000 Hz). Also, it can be noticed that the experimental curves do not form closed semicircles, which indicates that the missing parts of the curves are related to frequencies lower than 0.01 Hz. The obtained impedance spectrum revealed a successful enzyme immobilization onto the Cu-Zn-Al alloy sheet. In order to confirm the inhibitory activity of analyzed substances, the impedance spectra have been recorded under the same conditions. The measurement was performed in the absence of K₂[B₃O₃F₄OH] as well as its presence at a concentration of 0.028 mM for a fixed concentration of immobilized catalase of 7.01 × 10⁻³ g/mL, with the addition of a substrate at a concentration of 3.3 mM. The values obtained by the intersection of semicircles with the real axis indicate the electrolyte resistance R_el. In Figure 10, for measurement without K₂[B₃O₃F₄OH], the electrolyte resistance R_el value was found as 135 Ω, while 130 Ω was determined for measurement with K₂[B₃O₃F₄OH]. A difference between the initial measurements can be seen better in Figure 11, which presents the enlarged view of the initial part of the Nyquist diagram presented in Figure 10. The charge transfer resistance value through the electrolyte R_ct, for measurement in the absence of K₂[B₃O₃F₄OH], was recorded as ∼2200 Ω. After adding K₂[B₃O₃F₄OH], the charge transfer resistance value was determined as ∼2000 Ω. Obtained results indicate the inhibitory effect of K₂[B₃O₃F₄OH] on catalase activity. Since the corrosion rate of the material in saline can be compared to the polarization resistance, it may be seen that the corrosion rate of Cu-25.38Zn-3.3Al bioalloy is higher than that of catalase-modified Cu-25.38Zn-3.3Al bioalloy in the presence of inhibitors.

Figure 9:

Nyquist diagram of the impedance spectrum of bioalloy in the (a) absence and (b) presence of immobilised enzyme at a potential of 0.9 V.

Figure 10:

Nyquist diagram of the impedance spectrum of catalase activity without () and in the presence of K₂[B₃O₃F₄OH] 0.028 mM ().

Figure 11:

Enlarged view of the initial part of the Nyquist diagram of the impedance spectrum of catalase activity without () and in the presence of K₂[B₃O₃F₄OH] in the concentration of 0.028 mM ().

Obtained charge transfer resistance value through the electrolyte R_ct in this work is almost twice as lower as the values determined by the researchers who have investigated the corrosion and corrosion inhibition of Mg/Mn alloy in 3.5% NaCl solutions by 5-(3-aminophenyl)-tetrazole (APT) after different exposure intervals (Babouri et al. 2017; El-Sherif and Almajid 2011). The observed difference is attributed to the influence of chloride ions. The increase in resistance of Mg with APT concentration is a consequence of the adsorption of its molecules on the alloy surface due to the chloride ions’ attack. On the other hand, the semicircles at high frequencies are generally associated with the relaxation of electrical double-layer capacitors, while the diameters of the high-frequency semicircles can be considered as the charge transfer resistance (El-Sherif and Almajid 2011).

The success of enzyme immobilization on the bioalloy at a scan rate of 0.166 mV is illustrated in Figure 12. The value of the corrosion potential E_P was −0.03 V compared to the alloy E_P (0.07 V). The drop in potential indicates that the biosensor is cross-linked in the polymer, as seen by the curves in Figure 12. Electron mobility and diffusion are impeded, resulting in a decrease in potential. The cathodic polarization of Cu-25.38Zn-3.3Al bioalloy was better than for immobilized alloy from the polarization curves. The passivation of Al was visible. Since the surface ionization of Al was more substantial than that of Zn in Cu-25.38Zn-3.3Al bioalloy samples, it could be concluded that this compact Al layer inhibited cathodic oxygen depletion. The corrosion rate of catalase-modified bioalloy was lower than that of Cu-25.38Zn-3.3Al bioalloy due to the lower potential. The corrosion potentials of the material tested in the same medium could be a symbol of thermodynamic stability. In the physiological saline solution, the thermodynamic stability of the Cu-25.38Zn-3.3Al bioalloy was more significant than the catalase-modified bioalloy due to the possibility of uneven surface immobilization.

Figure 12:

Tafel polarisation curves of bioalloy in the presence of biosensor CAT (b) and absence of it (a).

The trend of activation-passivation appeared in the anodic polarization of Cu-25.38Zn-3.3Al bioalloy, while the trend of passivation did not occur in the catalase-modified bioalloy. In Cu–Zn–Al shape memory alloys, elements Zn and Al have stronger metal activity than Cu and may be easier to ionize in solutions (Chen et al. 2005). Adding Zn and Al as alloying elements to copper to form Cu-25.38Zn-3.3Al alloy contributes to decreasing the tendency of anodic polarization and improving the performance of cathodic polarization, which was observed during examining pure copper and the 70Cu-26Zn-4Al shape memory alloy in the simulated uterine fluid by Chen et al. (2005). Immobilization success and biocorrosion parameters indicate that the direct electrochemistry of enzymes can provide a good model for mechanistic studies of their electron transfer activity in biological systems, as discussed by Mani et al. (2012). The CAT enzyme immobilized in the film on the surface of Cu-25.38Zn-3.3Al alloy shows well-defined and direct electrochemical reactions. It retains its primary structure, bioactivity, and good stability.

Cu-based alloys are prone to aging treatment which affects shape memory behavior. The properties of these alloys are improved by the presented method of immobilization and reaction of alloy ions and iron metal ions in the enzyme. Therefore, it is essential to emphasize that the analysis of the corrosion potential is critical before shape memory alloys are placed in biomedical or industrial applications. It was found that alloys with shape memory have higher corrosion resistance than traditional ones due to the elastic behavior of the polycrystalline structure (Al-Hassani et al. 2017; Raheem et al. 2010). Raheem et al. (2010) observed that austenite’s current density (633.62 μA/cm²) was greater than that of martensite (336.45 μA/cm²) in Cu-Al-Ni shape memory alloy.

As one of the most applied biosensor techniques (Amine and Mohammadi 2019; Zhang et al. 2008), hydrodynamic amperometry has been used to study the electrocatalytic reduction of hydrogen peroxide on the CAT film of the bioalloy. The constant potential of the modified electrode (mixing at 400 rpm) at 0.9 V versus Ag/AgCl was applied. The catalytic reduction current was monitored while the different hydrogen peroxide aliquots were added. The increase in catalytic reduction currents was caused by a gradual increase in the concentration of hydrogen peroxide from 6 mM to 32 mM in a physiological saline solution. From the Lineweaver-Burk diagram presented in Figures 13 and 14, the inhibition type has been defined and values of maximum current (I_max) and Michaelis-Menten constant (K_m) were calculated in the absence as well as the presence of l-cysteine and K₂[B₃O₃F₄OH] inhibitors.

Figure 13:

Lineweaver-Burk diagram for the determination of kinetic constants (I_max and K_m) in the absence and presence of different concentrations of l-cysteine.

Figure 14:

Lineweaver-Burk diagram for the determination of kinetic constants (I_max and K_m) in the absence and presence of different concentrations of K₂[B₃O₃F₄OH].

In the enzyme reactions carried out, the competitive type of inhibition is recognized from the Lineweaver-Burke diagram shown in Figure 13. The obtained I_max value is constant (0.8 μA) while the K_m values were increased with raising the inhibitor concentration (0.0184 mM; 0.0396 mM; 0.0935 mM), indicating that the inhibitor binds to the active site of the enzyme on the bioalloy. The ESI complex can be formed by binding l-cysteine to the previously formed enzyme-substrate complex. The reaction proceeds in the direction that the ESI complex yields the product but less than the reaction without an inhibitor. With the addition of K₂[B₃O₃F₄OH] to the reaction mixture, the inhibitor has bound competitively to catalase, as shown in Figure 14. The calculated values of the kinetic constants I_max and K_m are changed by adding different concentrations of inhibitors. While I_max values were 3.3 μA; 1.4 μA and 0.66 μA, the values of K_m were 0.016 mM; 0.0075 mM and 0.0043 mM. The initial formation of the ES binding and inhibition complexes is required for this modality of inhibitors (Copeland 2013). Therefore, after the initial non-covalent binding of the substrate to the enzyme’s active site, these inhibitors affect the catalysis steps. The binding of boroxine to the ES complex is occurred according to the mechanism (Marangoni 2003) as follows:

(3)

where E, enzyme; S, substrate; I, inhibitor; ES, enzyme–substrate complex; ESI, ES–inhibitor; P, product.

Based on the mechanism and the obtained K_m constant, our research proved that K₂[B₃O₃F₄OH] has higher catalytic stability due to lower K_m values, which is in good agreement with the theory. Todorovic et al. (2014) evaluated the biocompatibility of Cu-Al-Ni shape memory alloys in vitro and proved that those alloys almost entirely reduced the metabolic activity of peripheral blood mononuclear cells (PBC).