The influence of silane and silane–PMMA coatings on the in vitro biodegradation behavior of AE42 magnesium alloy for cardiovascular stent applications

Abstract

The squeeze-cast alkali pretreated (NaOH, 3 M, 10 days) AE42 Mg alloy is coated with silane as well as silane–polymethylmethacrylate (PMMA) by a sol–gel technique. The detailed microstructural characteristics, adhesion strength and in vitro corrosion behavior using hydrogen evolution, potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) of the coatings in simulated body fluid (SBF) under both static and dynamic conditions for 7 days has been systematically investigated. In addition, contact angle, protein adsorption and hemocompatibility of all the specimens have also been studied to establish their significance for cardiovascular stent application. For comparison, the same has also been carried out on the uncoated AE42 alloy. The coated specimens showed improvement in hydrophobicity leading to their corrosion resistance. The results of EIS, pH variation and Mg²⁺ ion release confirm that the silane–PMMA coating has the lowest in vitro biodegradation under both static immersion and dynamic flow conditions. Further, the protein adsorption and hemocompatibility results depict enhanced biocompatibility of the silane–PMMA-coated specimen making it a favorable candidate for biodegradable stent applications.

---

1. Introduction

In recent years, there has been a great increase in the use of biodegradable materials for advanced implants and tissue scaffolds. These biodegradable materials should be able to stimulate the healing responses of injured tissues in the physiological environment. For example, the stents made of conventional non-degradable metals and polymers, might negatively interact with the surrounding tissues and results in long-term endothelial dysfunction, delayed re-endothelialization, late thrombosis, permanent physical irritation, toxic metal ion release, local chronic inflammation and so on.¹ Thus, the materials exhibiting biodegradability are preferred over the stable and inert ones as in most cases body requires only the temporary presence of an implant. The biodegradable implanted stents are degraded or absorbed after they fulfill the desired objectives and eliminates the requirement of follow-up surgery to remove them.

In the past decade, pure magnesium (Mg) and its alloys received considerable attention over polymers as promising candidates for new generation stent materials owing to their biodegradability, appropriate mechanical properties, and favorable compatibility in the physiological environment.² Accordingly, Mg alloys were investigated extensively as biodegradable implants for the application of bone fixation devices, cardiovascular stents and tissue engineering scaffolds. Mg-Rare earth (RE) based alloys are reported to exhibit the highest strength and ductility, the best corrosion resistance and great biosafety in the form of both stents and screws.³ Mg is an electrochemically active metal and very prone to corrosion owing to its position in the electromotive force (EMF) series. Therefore, the major challenge of using Mg and its alloys as temporary implants for biomedical application is to control their high corrosion rate. Corrosion is too rapid even for a biodegradable material and it is not homogeneous, due to the strong tendency for localized corrosion exhibited by Mg alloys. In addition, there is a propensity for formation of hydrogen bubbles (also known as balloon effect) by corrosion reaction owing to the rapid evolution of hydrogen that body cannot absorb.⁴ It also hinders cell bonding to the implant surface. Moreover, a shift in pH towards alkaline in the vicinity of the corroded surface is also a concern for medical applications.

Several attempts were made to increase the corrosion resistance of Mg alloys by modifying them.^4,5 However, the modification by alloying additions is limited owing to the low solubility of elements in Mg and therefore, the application of coatings is regarded as one of the most promising methods to improve corrosion resistance. For biomedical applications, besides corrosion protection, the coatings should also possess other functions, such as enhancement of biocompatibility, bioactivity, antibiotic ability or local drug delivery ability. Moreover, the specific coatings should enable biodegradation at a desired rate, and hence, should offer only a limited barrier function. Such coatings should be nontoxic and their breakdown products should be biodegradable as well. Surface modifications based on silane coating were very attractive in enhancing corrosion resistance of metallic substrates. These coatings are environment friendly and their application processes are not complicated. Accordingly, they became attractive to improve the corrosion resistance of a wide range of metals and alloys. Silane coated surfaces have also been observed to be biocompatible and widely used for biological interactions, protein absorption, cellular and bacterial adhesion.^6–9 However, to the best of our knowledge, no studies have been reported earlier on monitoring the performance of AE42 Mg alloy in the biological environment with silane and PMMA as coatings. This maiden endeavour categorically focuses to reveal the effect of silane and silane–PMMA coatings on biodegradable AE42 Mg alloy up on exposure to in vitro bio-environment. In the present investigation, the squeeze-cast alkali pretreated AE42 Mg alloy was coated with silane (3-aminopropyltriethoxysilane (APTES)) as well as silane–polymethylmethacrylate (PMMA) by the sol–gel process. The detailed microstructural characteristics and in vitro corrosion behavior of the coatings in simulated body fluid (SBF) were systematically investigated.

2. Experimental procedure

2.1. Materials and coating procedure

The AE42 Mg alloy was used in the present investigation and its chemical composition is shown in Table 1. The direct squeeze-casting technique was employed to fabricate the alloy. The alloy was first melted at 770 °C and it was then poured manually in a preheated die. The upper ram of a vertical hydraulic press squeezed the melt with a speed of 10 mm s⁻¹. The part solidified in 20 s under an applied pressure of 100 MPa. The AE42 alloy was coated with silane (S-coated) as well as silane combined with PMMA (SP-coated). For coating, specimens of 6 mm diameter × 3 mm length were cut from the castings and its surfaces were polished using SiC papers (up to 1500 grit) and ultrasonicated with deionized water and acetone. The ultrasonicated specimens were immersed in a 3 M NaOH solution of pH 12 maintained at 30 °C using a water bath in order to obtain a uniform hydroxide layer on the specimen surfaces before silane adhesion. This was done to improve the silane–substrate bonding. The alkali treated specimens were rinsed with deionized water and acetone, and then dried in hot air. For hydrolysis of ethoxy groups in APTES (Aldrich, 99% pure), 2.21 g of APTES and 18 g of H₂O were mixed and stirred at ambient temperature (∼25 °C) for 24 h. Concentrated H₂SO₄ of 0.98 g was added drop wise into the mixture and stirred for 24 h. The obtained solution is denoted as Si-solution.¹⁰ The specimens were immersed into the silane solution for 15 min and dried in air. This coating was cured by heating at 120 °C for 1 h in an oven. The PMMA solution was prepared by dissolving 2 wt% of PMMA (HiMedia, India) in acetone. The silanized alloy specimens were held vertically and immersed in PMMA solution for 1 h. The dipped specimens were dried in air for 20 min and then cured in an oven at 100 °C for 1 h.

Table 1 Chemical composition of the AE42 alloy

Element	Al	Zn	Mn	Si	Ce	La	Nd	Pr	Th	Be	Mg
Wt%	3.9	0.01	0.3	0.01	1.2	0.6	0.4	0.1	0.6	0.001	Balance

---

2.2. Characterization

X-ray diffraction (XRD) (Rigaku Ultima-IV) analysis was carried out to determine the phases present in the uncoated specimen using CuK_α radiation (λ = 1.541 Å). The scan range (2θ) was from 20 to 80°. The scan speed was 5° min⁻¹ and a step size of 0.05 was maintained. Both the silane and PMMA coatings used in the present investigation are amorphous (non-crystalline) in nature. Accordingly, these polymers based coatings never exhibit additional sharp diffraction peaks in the XRD patterns and therefore, XRD analysis were not carried out on the coated AE42 alloy specimens. The uncoated and coated specimens were characterized using scanning electron microscopy (FESEM, FEI Nova NanoSEM 450) to study the surface morphology and the cross-section of the coatings. Prior to SEM observation, all the specimens were gold sputter coated. The adhesion of the coatings was determined by tape test in accordance with ASTM D 3359-09 standard.¹¹ The coated specimen surfaces were cut using a scalpel with a cutting angle in between 15 and 30° and a pressure sensitive tape was applied and removed rapidly. The pencil hardness of the coatings was determined in accordance with ASTM D 3363-00 standard. Wood pencils of different hardness were procured and scratched on the coated surfaces at an angle of ∼45° at ambient temperature. Fourier transform infrared (FTIR) spectroscopy (Perkin Elmer, RX-I FTIR) was employed for all the coated specimens in order to study molecular interactions at the coatings. FTIR spectra were obtained over the range of 500–4000 cm⁻¹ with a spectral resolution of 4 cm⁻¹. The average surface roughness (R_a) of the uncoated and coated specimens (triplicates) were measured using a surface profilometer (Zygo NewView 7100, Zygo). The surface wettability of the coated and uncoated specimens was measured using a drop size analyzer (DSA 25, KRÜSS GmbH) using double distilled water at ambient temperature (∼25 °C) and 50% relative humidity. The contact angle measurements were carried out using sessile drop technique. The liquid of 10 μl droplets was allowed to fall on the surface of the specimen, and immediately image of the droplet was captured after 5 s of stabilization. The profile of the droplet was then automatically fitted and the contact angle was measured using DSA 4 software. Six specimens were tested in each condition.

2.3. In vitro degradation studies

In vitro biodegradation studies were performed following three different methods. The first method consisting of potentiodynamic polarization scan and electrochemical impedance spectroscopy (EIS) were performed in SBF solution prepared as per Kokubo's protocol using a potentiostat (Gill AC, ACM Instruments) consisting of a three electrode system.¹² The specimen (exposed area of 0.5 cm²) acted as working electrode, a platinum mesh acted as the counter electrode and a saturated calomel electrode (SCE) (4.5 mol l⁻¹ KCl) was the reference electrode. 200 ml of SBF incubated in a water bath at 37 ± 0.5 °C served as electrolyte. Open circuit potential was monitored for 1800 s. Potentiodynamic polarization tests were carried out −250 mV relative to the OCP at a scan rate of 0.2 mV s⁻¹. For EIS tests, a sinusoidal potential signal of 5 mV amplitude was used. The values of E_corr and I_corr were determined by Tafel extrapolation method from the potentiodynamic polarization plots. The value of corrosion potential (E_corr) was determined at the intersection of the Tafel slopes of both the cathodic as well as anodic branches of the potentiodynamic polarization curves. The value of corrosion current (I_corr) was determined at the intersection of the Tafel slope of the cathodic branch of the polarization curve with the line through the corrosion potential (E_corr). By using the exposed specimen surface area (0.5 cm²) and the Tafel equation corrosion rate was determined from the cathodic branch of the curve.⁵

The impedance response was measured over the frequencies ranging from 100 kHz and 10 mHz. The impedance spectra obtained from EIS were simulated using electrical equivalent circuit (EEC). The morphology of the corroded specimens was observed using FESEM. All the corroded specimens were gold sputter coated before observation under FESEM. The XRD analysis from the specimens after completion of EIS tests were also carried out in the 2θ range from 10 to 80° to confirm the presence of phases in the corrosion products of the coated and uncoated specimens. The scan speed was 5° min⁻¹ and a step size of 0.05 was maintained.

The second method was measurement of hydrogen evolution during corrosion. A schematic of the hydrogen evolution test set-up is shown in Fig. 1(a). The specimens of the uncoated, S-coated and SP-coated AE42 alloy having dimension of 10 × 10 mm² were placed at the bottom of a 1000 ml beaker containing 500 ml SBF solution approximately at 37 ± 0.5 °C for 168 h. The evolved hydrogen was collected in a measuring cylinder kept above the corroding specimens. The rate of hydrogen evolution was monitored continuously as a function of immersion time. Experiments were conducted in triplicate in order to examine the reproducibility of hydrogen evolution.

Fig. 1 Schematic diagram of (a) hydrogen evolution measurement setup and (b) dynamic test platform for performing in vitro biodegradation.

The third method was the in vitro biodegradation behavior of the coated and uncoated specimens evaluated at static immersion (SI) and dynamic flow (DF) conditions. The degradation (SI and DF) was assessed in SBF solution. For SI test, the specimens (triplicates) were immersed up to 168 h in SBF maintained at 37 ± 0.5 °C using a constant temperature water bath. The concentration of Mg²⁺ ion in the SBF solution was measured by atomic absorption spectroscopy (Perkin Elmer, AAnalyst 200) at the end of 24, 72, 120 and 168 h. For DF test, a test platform consisting of an electronic controller to monitor the flow rate was designed (Fig. 1(b)) to simulate the environment experienced by the stents in coronary arteries. A polyurethane tube with an inner diameter of 6 mm was used as channel for circulating the solution. The specimens (triplicates) having geometry of a circular tube were tightly locked by the pressure exerted on the outer wall by the enlarged tube. The pressure was enough to prevent penetration of solution in between the inner wall of the tube and the outer wall of the specimen. A constant flow rate of ∼250 ml min⁻¹ was maintained inside the specimen during the test period of 168 h.¹³ The pH of the solution was measured after 1, 2, 4, 6, 8, 10, 12, 24, 48, 72, 96, 120, 144 and 168 h for both SI and DF tests.

2.4. Protein adsorption studies

For protein adsorption test, bovine serum albumin (BSA) (HiMedia, India) was used as model protein. A calibration curve was drawn for standard BSA solution ranging from 200–1000 μg ml⁻¹. 25 ml of protein solution (1000 μg ml⁻¹ protein/saline solution) was pipette onto the uncoated and coated specimens (triplicates). The setup was allowed to remain for a period of 1, 3, 6 and 24 h separately and 1 ml of protein solution was removed and centrifuged. After that 100 μl of supernatant was taken and mixed with 1 ml of Bradford reagent (HiMedia, India), 2 ml of deionized water and kept in the dark for 10 min. The absorbance was recorded at 595 nm using UV-Vis spectrometer (Perkin Elmer, Lambda 35). The protein concentration was determined using the calibration curve.

2.5. In vitro hemocompatibility studies

2.5.1. Hemolysis studies. The specimens in triplicates (10 × 10 mm²) were taken in tubes containing 10 ml physiological saline. Human blood was collected by a doctor at Community Welfare Society Hospital Rourkela (Odisha, India) following Indian Council of Medical Research (ICMR) guidelines with the consent of the donor. Permission for the use of the blood was obtained from the hospital. Fresh blood (20 ml) with potassium citrate solution (1 ml, 1 mg ml⁻¹) was diluted up to 4/5 (v/v) with physiological saline, added (0.5 ml) into each test tube and kept at 37 °C for 1 h. For positive and negative controls 0.5 ml of blood was added in 10 ml of distilled water and physiological saline respectively. After 1 h specimens were centrifuged at 3500 rpm for 5 min and its optical density (OD) was taken at 545 nm. Percentage hemolysis ratio was calculated based on the optical density (OD) using the following formula:¹⁴where OD_test, OD_−ve and OD_+ve, are optical density values of test, negative control and positive control specimens respectively.

2.5.2. Platelet adhesion studies. Platelet adhesion test was conducted in order to evaluate thrombogenicity of the specimens and examined the interaction between blood and the specimens in vitro. For this, platelet-rich plasma (PRP) was prepared by centrifuging blood containing 3.8 wt% sodium citrate at 1500 rpm for 15 min. Then 60 μl of PRP was placed individually on top of each specimen, and incubated at 37 °C for 2 h. The specimens were gently rinsed with phosphate buffer solution (PBS) and 60 μl of 2.5% glutaraldehyde solution was added on to the specimens and kept for 30 min. The specimens were then dehydrated sequentially in 50%, 75%, 90% and 100% ethanol solution. The specimens were gold sputter-coated (JEOL JFC 1600 Autofine Coater), and observed using SEM (JEOL, JSM-6084LV).

2.6. Statistical analysis

Data were expressed as mean ± standard deviation and analyzed using software OriginPro 8 (Originlab Corporation, USA) at both significant (*p ≤ 0.05) and highly significant (**p ≤ 0.01) level. One-way analysis of variance (ANOVA) was accomplished to compare between groups and within groups for measuring the level of significance by using Tukey's test.

3. Results and discussion

3.1. Characterization

The XRD pattern of the squeeze-cast AE42 alloy is shown in Fig. 2(a). It is evident that the as-cast AE42 alloy consists of primary Mg (α-Mg) peaks along with the peaks corresponding to Al₄RE phase. The phases Al₁₁RE₃ and Al₄RE are considered to be the same and are conventionally denoted as Al₄RE phase in the literature.^15,16 There was no peak observed corresponding to the β-Mg₁₇Al₁₂ phase, which is generally present in Mg–Al alloys.

Fig. 2 XRD pattern obtained from the (a) as-cast uncoated AE42 alloy specimen; and corroded surfaces of the (b) uncoated (c) S-coated and (d) SP-coated specimens after completion of EIS tests.

The surface morphology of the coated and uncoated specimens observed under FESEM is shown in Fig. 3(a–c). The micrograph of the uncoated specimen shown in Fig. 3(a) reveals nearly polygonal grains of α-Mg with an average grain size of 25 μm. The detailed description of the microstructure is reported elsewhere.¹⁶ It is observed from Fig. 3(b) that the S-coated surface had a wavy appearance. A similar morphology of silane coating was also observed by Kelly et al.¹⁷ The reason behind this uneven distribution of the coating deposition is not known. In contrast, a reasonably smooth and uniform coating morphology was observed for the SP-coated specimen, as shown in Fig. 3(c). There was no crack observed on the coated surfaces of the specimens even after curing. The SEM micrograph of cross-section of the SP-coated specimen with the results of line scan (EDS) carried out across it is shown in Fig. 3(d). It displayed the dominant presence of C along with Si and O at the coated region confirming the presence of the silane–PMMA coating. The presence of Mg as a major element of the substrate was also seen.

Fig. 3 SEM micrographs of surfaces of the (a) uncoated, (b) S-coated, (c) SP-coated AE42 alloy specimens and (d) results of EDS analysis carried out across the cross-section of the SP-coated specimen.

The results of adhesion and pencil hardness tests of the coated surfaces of the specimens are shown in Table 2. The ‘B’ scale represents the rate of the adhesion whereas ‘H’ scale denotes the hardness of coating. The higher ‘B’ (i.e. 5B) scale of pencil means least coating area removed indicating a strong adhesion between the coating and specimen. Similarly, a high ‘H’ scale indicates a harder coating on the specimen. For both the coatings, the adhesion tests revealed strong adhesion with a rating of 5B, which is the best as per the ASTM D 3359-09 standard.¹¹ The values of pencil hardness of the coatings were 4H and 5H for the S-coated and SP-coated specimens, respectively. Gaur et al. too observed an adhesion of 5B and pencil hardness of 4H after investigating mechanical characteristics of the silane coating on Mg–6Zn–Ca alloy.⁶ A schematic representation of the overall reactions governing the coating processes in the S-coated and SP-coated specimens in sol–gel method is depicted in Fig. 4(a and b). The reason behind strong adhesion is the aminolysis reaction caused the –NH group of silane to couple with the ester group of PMMA to form an amide bond. This covalent coupling attributed to the adhesion of a firm, smooth and stable PMMA layer on the S-coated alloy specimen, as shown in Fig. 4(b).

Table 2 Results of pencil hardness and adhesion tests of the coatings

Specimen	Adhesion	Hardness
S-Coating	5B	4H
SP-Coating	5B	5H

---

Fig. 4 Schematic diagram of the sol–gel coating process for (a) silane coating and (b) silane–PMMA coating on the AE42 alloy; and (c) the FTIR spectra obtained from both the coated specimens.

The FTIR spectra obtained from the S-coated as well as SP-coated specimens are shown in Fig. 4(c). The alkali pre-treatment caused strong bonding between the –OH group of the Mg(OH)₂ layer and Si of the APTES. The bands related to the formation of OH–Mg–OH group was observed at 619 cm⁻¹.¹⁸ The bands that appeared at ∼1010 and 1100 cm⁻¹ correspond to the asymmetric stretching of Si–O–Si linkage. It confirmed the formation of a cross-linked siloxane network in the film.^19,20 The bands corresponding to Si–O–Et (asymmetric) and Si–OH were observed at 1389 and ∼850 cm⁻¹, respectively. The FTIR results in the present investigation match pretty well with the reported literature.⁶ The band at 1620 cm⁻¹ was attributed to the N–H bending region of free –NH₂ group owing to the presence of the amine group in the S-coated specimen.²¹ The characteristic band of the PMMA corresponding to the stretching of C=O bond was observed at 1732 cm⁻¹.²² The broad peak at ∼3400 cm⁻¹ was assigned to the vibration of –OH group. This helped to create a Si–O–Si network (Fig. 4(a)) whose presence was already confirmed by the results of FTIR.

The surface roughness of the specimens gradually descended from the uncoated specimen followed by the S-coated and SP-coated specimens. The uncoated, S-coated and SP-coated specimens exhibited R_a values of 3.09 ± 0.10 μm, 2.91 ± 0.15 μm and 0.75 ± 0.02 μm, respectively. This variation in surface roughness is expected to influence the contact angle, protein adsorption and other properties of the specimens.

3.2. Surface wettability

Surface wettability plays a vital role in the corrosion behavior of the specimens. Table 3 shows the average values of the contact angles. The contact angle for the uncoated AE42 specimen was found to be ∼65°. Majumdar et al. too observed a contact angle of 63° for the AZ91 alloy.²³ The S-coated specimen had a contact angle of ∼83°. The increase in contact angle with the silane coating indicates the increased hydrophobicity and it was attributed to the presence of OH group on the surface, as shown in Fig. 4(a). On the other hand, PMMA is a monopolar solid whose surface tension resulted only from the Lifshitz–van der Waals intermolecular interactions. It has free polar groups such as –COO and –OCH₃ in which the oxygen may act as an electron donor in contact with water and form a hydrogen bond.²⁴ Yeh et al. too showed that the incorporation of silica with PMMA increased the water contact angle and made the surface hydrophobic.²⁵ A contact angle of ∼114° for the SP-coated specimen was obtained. The wettability of the SP-coated surface decreased because the –COO group of the PMMA was already coupled to –NH group of the silane (Fig. 4(b)). This molecular level interaction between silane and PMMA helped to make the SP-coated surface inert and stable, which contributed to the hydrophobic behavior of the coating.

Table 3 Results of the contact angle measurement

Specimen	Contact angle (degree)
Uncoated	64.6 ± 1.9
S-Coated	83.1 ± 5.0
SP-Coated	113.7 ± 1.8

---

3.3. In vitro degradation studies

3.3.1. Electrochemical corrosion response. Fig. 5(a) shows the open circuit potential (OCP) of all the three specimens recorded for 1800 s. It is obvious that the OCP for all the specimens increased initially and then stabilized after about 300 s. The OCP of both the coated specimens increased continuously with increase in time i.e., it shifted towards more noble potential, which indicates that the passive films formed on the coated specimens were relatively more protective than that on the uncoated AE42 alloy. The potentiodynamic polarization curves of the uncoated, S-coated and SP-coated AE42 alloy specimens are shown in Fig. 5(b). The corrosion current, I_corr, the corrosion potential, E_corr and the corrosion rate (mm per year) were determined by polarization test and the values were listed in Table 4. The I_corr value of the SP-coated specimen is three order of magnitude lower than that of the uncoated AE42 alloy, and one order lower than that of the S-coated specimen. The E_corr value of the SP-coated was lower than that of the uncoated and S-coated specimens by 83 and 221 mV, respectively. It is evident that the corrosion rate of the SP-coated specimen is two order and one order of magnitude lower than that of the uncoated and S-coated AE42 specimen, respectively.

Fig. 5 (a) Variation of open circuit potential for all the specimens (b) potentiodynamic polarization plots obtained from all the specimens, (c) EIS spectra obtained in SBF for all the specimens; electrical equivalent circuits (EECs) corresponding to the EIS data obtained from the (d) uncoated as well as (e) S-coated and SP-coated AE42 alloy specimens.

Table 4 Summary of the various parameters obtained from the potentiodynamic polarization tests

Specimen	Ecorr (mV vs. SCE)	Icorr (mA cm^−2)	Corrosion rate (mm per year)
Uncoated	−1755.5	0.13572	3.100
S-Coated	−1617.5	0.02082	0.325
SP-Coated	−1838.6	0.00097	0.0222

---

EIS was also carried out to investigate the long-term corrosion behavior of all the specimens and the plots are shown in Fig. 5(c). It is obvious that |Z|_SP-coated ≫ |Z|_S-coated > |Z|_uncoated where Z is the impedance offered. In order to develop a greater insight into the corrosion mechanism and the corrosion resistance provided by all the specimens, electrical equivalent circuit (EECs) were simulated based on the EIS data as shown in Fig. 5(d and e). The calculated data from the EECs simulation were fitted to the data obtained experimentally. The resistance offered by the SBF is denoted as R_s. R_ct is the resistance due to the charge transfer. A double layer capacitive zone (Q_dl) formed as a result of interaction between the free electrons on the metal surface and the negative ions present in the SBF. Instead of pure capacitance, a constant phase element (CPE) was considered. This kind of consideration was reported earlier as well.²⁶ The CPE impedance is provided as follows:²⁷

Z_CPE = Q⁻¹(jω)⁻ⁿ (2)

where Z_CPE is the CPE impedance (Ω cm²), Q is a constant (Ω⁻¹ cm⁻² s⁻ⁿ), n is a dimensionless constant in the range −1 ≤ n ≤ 1, j is the imaginary number where j = √(−1) and ω is the angular frequency (ω = 2πf, f is the linear frequency). There could be two reasons behind the observed inductive behavior of the uncoated alloy shown in Fig. 5(c). First, the adsorption of Mg²⁺ species through the Mg(OH)₂ corrosion layer and second, the presence of magnetic properties of the rare earth elements.²⁸ The inductive behavior is denoted by L and the resistance provided due to the adsorption of Mg²⁺ species through the intermediate porous corrosion layer is given by R_ad.

The dual semicircles observed in the Nyquist plots of both the coated specimens were attributed to the coating barrier. As a result, separate EEC was modelled and fitted to those curves. The capacitive loop created due to the coatings, consists of a CPE denoted by Q_coat and resistance offered by the coatings denoted by R_coat, as shown in Fig. 5(e). Various parameters employed to fit EEC are shown in Table 5.

Table 5 Summary of the parameters obtained from the EEC fitting

Specimen	Rct (Ω cm^2)	Qdl (Ω^−1 cm^−2 s^−n)	L (henry cm^2)	Rad (Ω cm^2)	Rcoat (Ω cm^2)	Qcoat (Ω^−1 cm^−2 s^−n)
Uncoated	258	4.022 × 10^−6	87.7	232	—	—
S-Coated	538.6	3.56 × 10^−6	—	—	266.4	175.9 × 10^−5
SP-Coated	2151	445.6 × 10^−6	—	—	2851	1.73 × 10^−5

---

It is worth noting that R_ct was the lowest indicating lowest long term corrosion resistance for the uncoated specimen and it increased by two and eight folds for the S-coated and SP-coated specimen. Further, R_coat for the SP-coated specimen was higher by an order as compared to the S-coated specimen, which indicated the protection provided by the silane–PMMA coating was much higher relative to that provided by the silane coating alone.

Fig. 6(a–c) shows the SEM micrographs of the uncoated, S-coated and SP-coated AE42 alloy specimens after completion of EIS corrosion tests. It is clearly visible that the surface of the uncoated specimen was degraded deeply at some areas showing substantial signs of corrosion (Fig. 6(a)). The surface of the S-coated specimen revealed a network of cracks exposing the base alloy at some areas (Fig. 6(b)), whereas, the SP-coated specimen exhibited very few shallow micro cracks indicating minimal damage to the surface (Fig. 6(c)).

Fig. 6 SEM micrographs of corroded surfaces of the (a) uncoated, (b) S-coated and (c) SP-coated AE42 alloy specimens. The areas marked with ‘A’ in (c) represent the inherited regions of the SP-coating.

The XRD patterns obtained from the corroded surfaces of all the specimens are presented in Fig. 2(b–d). The presence of Mg(OH)₂ was identified as the major corrosion product in both the uncoated (Fig. 2(b)) and S-coated (Fig. 2(c)) specimens. The formation of Mg(OH)₂ following the corrosion of Mg takes place by the following overall reaction.⁵

Mg(s) + 2H₂O → Mg(OH)₂ + H₂(g) (3)

Mg(OH)₂ peaks were dominantly observed on the XRD pattern taken from the uncoated alloy due to exposure of α-Mg to corrosion. The presence of Mg(OH)₂ on the S-coated specimen was very minimum owing to the presence of shallow microcracks on its surface. No prominent Mg(OH)₂ peak was observed for the SP-coated specimen (Fig. 2(d)) indicating the least corrosion of the surface.

3.3.2. Hydrogen evolution studies. One of the major shortcomings of using Mg as an implant material is the formation of H₂ gas when it corrodes in the body fluid by the following reaction:²⁹

Mg(s) + 2H₂O = Mg²⁺ + 2OH⁻ + H₂(g) (4)

The evolved H₂ accumulates and forms gas pockets that might lead to necrosis of the neighbouring tissues and delay the healing process.⁶ Therefore, the H₂ evolution from all the specimens was studied with increase in exposure time and the plots are shown in Fig. 7. It is observed that the H₂ evolution was the highest for the uncoated alloy and the lowest for the SP-coated alloy. The hydrogen evolved from the S-coated and SP-coated specimens exhibited a trend similar to that of their pH change with increase in immersion time. The curves show that the H₂ evolved was almost negligible up to 24 and 72 h for the S-coated and SP-coated specimens, respectively. Following 72 h of exposure the curve for the S-coated specimen rose moderately owing to the formation of cracks on the S-coated surface.

Fig. 7 Variation of hydrogen evolution with immersion time for all the specimens.

3.3.3. Degradation studies in SBF. The variation of Mg²⁺ ion release as a function of time is shown in Fig. 8(a). It is quite evident from results that the testing condition decided the degradation behavior of all the specimens. The rate of degradation of all the specimens in the SI tests was significantly lower as compared to that in the DF tests. The release of Mg²⁺ ion as well as its increase with increase in time was the lowest for SP-coated specimens employed in both SI and DF tests. In contrast, these were the highest for the uncoated specimen and intermediate for the S-coated specimen. The corrosion rate of all the specimens decreased following the formation of a passive layer in SI tests. However, the layer was washed away under the constant circulation of SBF in DF tests, which accelerated the degradation rate of the specimens. Lévesque et al. too observed a similar phenomenon in AM60B alloy under dynamic flow condition.³⁰ Unlike passive layer, the coatings acted as a barrier and provided a relatively higher protection to Mg substrate from dissolution. The coatings under SI test conditions were able to withstand the corrosion process for a longer time as compared to the DF test conditions where additional erosion took place under dynamic flow condition by the constant circulation of SBF. The circulation of SBF over the specimen surface inhibited local pH increase, and hence, a pH study was carried out to evaluate the alkaline shift in the SBF under different testing conditions.

Fig. 8 Variation of (a) Mg²⁺ ion released and (b) pH with time under both static immersion (SI) and dynamic flow (DF) conditions for all the specimens.

Local alkalization unfavorably affects the pH-dependent physiological reaction in the vicinity of the Mg implant. This might even lead to an alkaline poisoning effect as the local pH exceeds 7.8.³¹ Therefore, the change in pH of the SBF solution in contact with all the specimens in SI and DF tests was recorded as a function of time and the plots are shown in Fig. 8(b). The pH value of the SBF solution in contact with the uncoated specimen increases almost steadily with increase in time for the first 24 h. Then the rate slowed down, which might be due to the formation of passivation layer on the surface of the uncoated specimen. It was clearly visible that there was a significant difference in the pH values under SI and DF tests. The pH value of the SP-coated alloy after 168 h of immersion in SBF remained far below 7.8, in both SI and DF tests. On the other hand, it surpassed the acceptable limit within 24 h for the uncoated specimen. The S-coated specimen also exhibited a pH value within the acceptable limit in the SI, however, it reached slightly above 7.8 in DF test. The presence of OH⁻ ion due to corrosion in the SBF solution for the SP-coated specimen was the lowest and thus, promoting the acceptability of this coating.

3.4. Protein adsorption studies

Protein adsorption is an essential criterion that estimates the adhesion of cells on the specimen surface. The tailoring of specimen surfaces highly influences the protein adsorption. BSA was selected for this study since albumin is the most abundant protein present in the human blood plasma.³² The variation of protein adsorbed on the surfaces of the specimens was measured as a function of time and is shown in Fig. 9(a). From the figure, it is observed that initially the protein adsorption increases exponentially with increase in time and then stabilized after 6 h of incubation. Yang et al. too reported an exponential trendline of protein adsorption kinetics tested with albumin and fibronectin on Ti surface.³³ After 3 h of incubation, the protein adsorbed on the uncoated alloy specimen was higher than that of the S-coated and SP-coated specimens by 6.4 and 22.8%, respectively. Protein adsorption after attainment of steady state following 24 h of incubation on the uncoated specimen surface was observed to increase by 16.72 and 45.23%, respectively for the S-coated and SP-coated specimens. Therefore, the adsorption of protein on the specimen surfaces decreased with the corresponding decrease of their wettability. This also indicated a lower affinity of the BSA to adsorb on hydrophobic surfaces as compared to hydrophilic surfaces. This opinion is in agreement with the observation of Roach et al.³² In addition, the corrosion resistance of the specimens improved as the hydrophobicity of the surface improved.²⁵ In the context of a cardiovascular stent, the protein adsorption and wettability of the specimen surfaces might also provide an assessment about the platelet adhesion on the surfaces of the specimens.

Fig. 9 (a) Protein adsorbed as a function of time and SEM micrographs after platelet adhesion test on the surfaces of (b) uncoated, (c) S-coated and (d) SP-coated AE42 alloy.

3.5. In vitro hemocompatibility studies

3.5.1. Hemolysis studies. The hemolytic ratio (HR) obtained from the hemolytic activity of the specimens are shown in Table 6. The HR values of the S-coated and SP-coated specimens were 0.220 and 0.013% exhibiting no hemolytic activity. The uncoated specimen had an HR value of 4.260% that lied within the acceptable limit for clinical applications i.e., below 5%.³⁴ The HR values correlate with the corrosion behavior of the specimens. The concentration of Mg²⁺ ions in the blood plasma released due to corrosion had an adverse hemolytic effect. As the Mg²⁺ concentration was increased, the osmotic pressure of the solution was also enhanced. This caused swelling of the erythrocytes followed by the rupture of their cell membrane, which finally lead to the hemolysis of red blood cells.³⁵ In addition, the degradation of Mg produced OH⁻ ion in the blood plasma and increased the pH value of the blood, which enhances hemolytic activity. The Mg²⁺ ion release as well as the variation of pH value was the highest for the uncoated specimen and the lowest for the SP-coated specimen with the S-coated specimen showing intermediate values. The synchronization of Mg²⁺ ion release and the variation of pH value were directly related to the corrosion behavior of the specimens, which decided their hemolytic activity. In other words, higher is the resistance to corrosion of the specimens lower is the Mg²⁺ ion release that causes a minor increase in pH value and therefore, the hemolytic activity is controlled.

Table 6 Results of hemolysis tests on the specimens

Specimen	Hemolysis ratio (%)
Uncoated	4.260 ± 0.270
S-Coated	0.220 ± 0.110
SP-Coated	0.013 ± 0.004

---

3.5.2. Platelet adhesion studies. As per Goodman's observations, a platelet spreading can be divided into five categories for analysis: round, dendritic, spread dendritic, spread and fully spread.³⁶ A platelet with dendritic spreading having pseudopodial projections was observed (Fig. 9(b)), which showed signs of platelet activation on the uncoated alloy surface.

The S-coated specimen showed few platelets adhered on its surface (Fig. 9(c)). On the other hand, very few platelets were detected on the SP-coated specimen surface and their morphology was observed to be in round state, i.e., there was no sign of activation (Fig. 9(d)). It is evident that there were a larger number of adherent platelets on the uncoated alloy surface as compared to that on the surfaces of the coated alloys. The AE42 alloy has a weak propensity of platelet activation, which was attributed to the presence of rare earth elements. Vogler et al. indicated that the plasma coagulation induced by a hydrophobic surface was much less as compared to a hydrophilic surface.³⁷ The adhesion of platelets for the specimens behaved correspondingly in the order of their surface wettability. As the surface wettability of the specimens was decreased, a decrease in the number of platelets adhered was observed. Additionally, the protein adsorption kinetics also played a crucial role in the platelet adhesion process. A rapid adsorption of proteins might cause a higher number of platelet adhesion, which can trigger thrombus formation by platelet activation and ultimately results in blood coagulation. Hence, it can be concluded that the thrombogenicity of the specimens gradually decreased from the uncoated to S-coated and then to SP-coated specimens and it is in well agreement with the results obtained from their surface wettability and the protein adsorption kinetics.

4. Conclusion

The squeeze-cast AE42 Mg alloy was coated with silane as well as silane–PMMA. The detailed microstructural characteristics and in vitro corrosion behavior of the coatings in simulated body fluid were systematically investigated. The S-coated surface had a wavy appearance, whereas, a reasonably smooth and uniform coating with no cracks was observed for the SP-coated specimen. Adhesion tests revealed strong adhesion with a rating of 5B for both the coatings. The values of pencil hardness of the coatings were 4H and 5H for the S-coated and SP-coated specimens, respectively. The in vitro biodegradation test suggested that the SP-coated specimen exhibited superior corrosion resistance than the S-coated and uncoated specimens by three order and one order of magnitudes, respectively. The corrosion rate of the SP-coated specimen was two and one order of magnitudes lower than that of the uncoated and S-coated specimens. The rate of degradation of all the specimens in the SI tests were significantly lower as compared to that in the DF tests and a significant difference in the pH values were observed in both the tests. Further, an improvement in hydrophobicity of the SP-coated specimen resulted in significant reduction of protein adsorption and hemolysis ratio. From the overall results, the SP-coated AE42 alloy specimen exhibited high potential for cardiovascular stent application owing to its reasonably high resistance to corrosion along with the hemocompatible nature.

Acknowledgements

Authors acknowledge the facilities provided by Community Welfare Society Hospital Rourkela (Odisha, India) for performing hemocompatibility studies.

References

1. J. Wang, Y. He, M. F. Maitz, B. Collins, K. Xiong, L. Guo, Y. Yun, G. Wan and N. Huang, Acta Biomater., 2013, 9, 8678–8689.
2. G. Mani, M. D. Feldman, D. Patel and C. M. Agrawal, Biomaterials, 2007, 28, 1689–1710.
3. Y. Chen, Z. Xu, C. Smith and J. Sankar, Acta Biomater., 2014, 10, 4561–4573.
4. H. Hornberger, S. Virtanen and A. Boccaccini, Acta Biomater., 2012, 8, 2442–2455.
5. G. L. Song and A. Atrens, Adv. Eng. Mater., 1999, 1, 11–33.
6. S. Gaur, R. S. Raman and A. Khanna, Mater. Sci. Eng., C, 2014, 42, 91–101.
7. F. Zucchi, V. Grassi, A. Frignani, C. Monticelli and G. Trabanelli, Surf. Coat. Technol., 2006, 200, 4136–4143.
8. C. R. Jenney, K. M. DeFife, E. Colton and J. M. Anderson, J. Biomed. Mater. Res., 1998, 41, 171–184.
9. W. Jin, Q. Hao, X. Peng and P. K. Chu, Mater. Lett., 2016, 173, 178–181.
10. T. Tezuka, K. Tadanaga, A. Hayashi and M. Tatsumisago, J. Am. Chem. Soc., 2006, 128, 16470–16471.
11. J. Degner, F. Singer, L. Cordero, A. R. Boccaccini and S. Virtanen, Appl. Surf. Sci., 2013, 282, 264–270.
12. T. Kokubo and H. Takadama, Biomaterials, 2006, 27, 2907–2915.
13. T. Ramanathan and H. Skinner, Continuing Education in Anaesthesia, Critical Care & Pain, 2005, 5, 61–64.
14. T. Kaur and A. Thirugnanam, RSC Adv., 2016, 6, 39982–39992.
15. A. Mondal and S. Kumar, Compos. Sci. Technol., 2008, 68, 3251–3258.
16. A. Mondal, C. Blawert and S. Kumar, Mater. Corros., 2015, 66, 1150–1158.
17. D. N. Kelly, R. H. Wakabayashi and A. M. Stacy, ACS Appl. Mater. Interfaces, 2014, 6, 20122–20129.
18. G. Parthasarathy and U. Chandra, presented in part at 46^th Lunar and Planetary Science Conference, Houston, USA, March 2015.
19. X. Lu, Y. Zuo, X. Zhao and Y. Tang, Corros. Sci., 2012, 60, 165–172.
20. J. Hu, Q. Li, X. Zhong, L. Li and L. Zhang, Thin Solid Films, 2010, 519, 1361–1366.
21. D. Maity, P. Chandrasekharan, S.-S. Feng and D. Jun, presented in part at International Conference on Nanoscience and Nanotechnology, Sydney Australia, February 2010.
22. S. Sathish and B. C. Shekar, Indian J. Pure Appl. Phys., 2014, 52, 64–67.
23. J. D. Majumdar, U. Bhattacharyya, A. Biswas and I. Manna, Surf. Coat. Technol., 2008, 202, 3638–3642.
24. J. Krawczyk, K. Szymczyk, A. Zdziennicka and B. Jańczuk, Int. J. Adhes. Adhes., 2013, 45, 106–111.
25. J.-M. Yeh, C.-J. Weng, W.-J. Liao and Y.-W. Mau, Surf. Coat. Technol., 2006, 201, 1788–1795.
26. J.-B. Jorcin, M. E. Orazem, N. Pébère and B. Tribollet, Electrochim. Acta, 2006, 51, 1473–1479.
27. P. Zoltowski, J. Electroanal. Chem., 1998, 443, 149–154.
28. M. Ascencio, M. Pekguleryuz and S. Omanovic, Corros. Sci., 2014, 87, 489–503.
29. T. Hanawa, J. Artif. Organs, 2009, 12, 73–79.
30. J. Lévesque, H. Hermawan, D. Dubé and D. Mantovani, Acta Biomater., 2008, 4, 284–295.
31. G. Song, Corros. Sci., 2007, 49, 1696–1701.
32. P. Roach, D. Farrar and C. C. Perry, J. Am. Chem. Soc., 2006, 128, 3939–3945.
33. Y. Yang, R. Cavin and J. L. Ong, J. Biomed. Mater. Res., Part A, 2003, 67, 344–349.
34. T. Kaur and A. Thirugnanam, J. Mater. Sci. Technol., 2016 DOI:10.1016/j.jmst. 2016.06.020.
35. Z. Zhen, X. Liu, T. Huang, T. Xi and Y. Zheng, Mater. Sci. Eng., C, 2015, 46, 202–206.
36. S. L. Goodman, J. Biomed. Mater. Res., 1999, 45, 240–250.
37. E. A. Vogler and C. A. Siedlecki, Biomaterials, 2009, 30, 1857–1869.

---