Abstract
Surface coatings are widely used for improving the biocompatibility or corrosion resistance of Mg alloy implant materials. Polytetrafluoroethylene (PTFE) coating with about 100 nm thickness was prepared on AZ31 alloy substrate by radio frequency magnetron sputtering. The coating was characterized via X-ray photoelectron spectroscopy, fourier transform infrared spectrometer, X-ray diffractometer, scanning electron microscope, contact angle goniometer, atomic force microscope and white light interferometer. The corrosion behavior of PTFE coating were analyzed by electrochemical and immersion measurements in SBF. The results showed that PTFE coating with dense and flat surface can be formed on AZ31 alloy substrate by sputtering a PTFE target directly at room temperature. The corrosion current density declined from 1.92 × 10−4 A/cm2 (bare AZ31 alloy) to 1.07 × 10−8 A/cm2 (sample with PTFE coating), and the 7 days immersion tests revealed a complete surface morphology and very slight variation in pH value of the SBF, suggesting a significant increase in corrosion resistance. High viability rate of endothelial cells of the PTFE coated AZ31 proved the good biocompatibility. These results manifest that the PTFE is a protective coating on magnesium alloys for implants application.
Introduction
Magnesium (Mg) is a beneficial and essential element for the human body. In recent years, magnesium alloys have attracted attention from researchers because of the potential application in medical implants, such as orthopedics and cardiovascular implants, etc.1–3 Compared to the traditional permanent metallic biomaterials, mainly including stainless steel, and titanium (Ti) alloys,4,5 Mg alloys have unique advantages because they can be absorbed in the physiological environment of the human body. The Mg alloys implants therefore don’t have to be taken out of the body via a second surgery once the tissue has been completely cured. Also, Mg alloys have antibacterial performance thus could reduce the risk of bacterial infections during the treatment.6,7 Moreover, the mechanical properties of Mg alloys are suitable for acting as bone pins, plates or endovascular stents, due to the low density close to bone and high specific strength.
Nevertheless, the major challenge is how to prevent the rapid corrosion of magnesium alloys in the physiological environment, and therefore avoid the implantation failure caused by the lost of mechanical integrity.8,9 To solve this problem, anti-corrosive coatings have been used as a important strategy. It should be noted that the coatings which are adopted to improve the corrosion resistance of magnesium alloys, must be non-toxical and biocompatible.10–14 For example, hydroxyapatite (HA) has the same composition and structure as human bone crystals, with excellent biocompatibility, biological activity and biodegradability. It can directly bind with bone. These advantages above make it an ideal hard tissue substitute material. It has been used as a filling material for bone defects, providing a scaffold for the new bone formation. Up to now, a variety of coating techniques to coat HA on Mg alloys are available for medical implant application.15,16 Moreover, other inorganic materials such as Ti-O, 17 TiN, 18 parylene C, 19 and microarc oxidation (MAO)-based coating 20 with good corrosion resistant and biocompatibility, has been investigated as coatings for biomedical Mg alloys.
Other than inorganic coatings, polymeric materials have been widely used to make coatings on Mg alloys for many biomedical applications. The most widely used synthetic biomedical polymer are polylactic acid (PLA), polyglycolic acid (PGA) or poly (lactic-co-glycolic acid) (PLGA), with good biodegradability and biocompatibility, which could gradually degrade after a certain period of use and ultimately become harmless water and carbon dioxide to the human body.21,22 For Mg alloys implants, they could be applied as drug delivery carriers or other functions.23,24
Polytetrafluoroethylene (PTFE) is chemically stable, and it's corrosion resistance is excellent. In the field of medical treatment, many people have conducted in-depth research on it as a biomaterial. In recent years, PTFE membranes have been developed and used as artificial organs, such as blood vessels and heart valves.25,26 In clinical practice, it has been used as a scaffold material for artificial blood vessels and nasal plastic surgery, with good results.27,28 Thus it can be seen that the biocompatibility of PTFE is good enough.
Based on the current available research, there are still no studies on the PTFE coated biomedical magnesium alloys. Thus, in this study, the researchers attempted to coat the AZ31 magnesium alloy substrate with a combination of radio frequency (RF) magnetron sputtering and PTFE coatings. The corrosion resistance and cytocompatibility of the PTFE coating were discussed.
Materials and methods
Preparation of the coating
In this work, PTFE coating was prepared on the surface of AZ31 magnesium alloy (composition: 3.12 wt.% Al, 0.93 wt.% Zn, 0.30 wt.% Mn, 0.018 wt.% Si, 0.0029 wt.% Fe, 0.001 wt.% Ni, 0.0009 wt.% Cu, balance Mg). AZ31 contains a small amount of Al, Zn and other elements beneficial to human body, and has good mechanical properties, corrosion resistance and good processability, so it has been studied and applied in many fields, including biomaterials. First, AZ31 alloy pieces with size of ∅10 × 3 (mm) were cut from an extruded bar. The pieces were polished using sandpaper, then washed ultrasonically in ethanol for 3 min. Surface of the pieces must be dried completely before they are used as substrates. PTFE coating was deposited on the polished AZ31 pieces by sputtering a PTFE target for 3 h, under Ar flow of 20 sccm, working pressure of 0.6 Pa and radio frequency power of 100 W.
Coating characterizations
X-ray photoelectron spectroscopy (XPS), fourier transform infrared spectrometer (FT-IR), X-ray diffractometer (XRD), scanning electron microscope (SEM), contact angle meter, atomic force microscope (AFM), white light interferometer were used for the coating characteristics. The morphology and roughness were determined using SEM and AFM, respectively. The element composition and distribution of the coating were determined by XPS. The chemical bond information of PTFE was characterized by FT-IR spectrum. The phase composition of the coating was analyzed by XRD. The thickness of the coating was measured by white light interferometer. The surface wettability of PTFE coating was evaluated by testing the water contact angle (CA) via a contact angle meter.
Corrosion behavior tests in vitro
For corrosion tests, simulated body fluid (SBF) was used as a medium according to literature, 29 with composition of 8.035 g NaCl, 0.355 g NaHCO3, 0.225 g KCl, 0.231 g K2HPO4·3H2O, 0.311 g MgCl2·6H2O, 40 mL of 1.0M-HCl, 0.292 g CaCl2, 0.072 g Na2SO4, and 6.118 g Na2HPO4, in 1 L of deionized water. Bare and PTFE coated AZ31 samples (n = 3) were immersed in SBF at 37 °C for up to 7 days. The solution volume-to-specimen area ratio was 40 mL/cm2, based on ASTM G31-21. After a decent interval in 7 days, the pH value were monitored and samples were removed from solution. After samples being dried, the corrosion appearance were examined under a microscope.
In addition, electrochemical corrosion behavior of samples were tested by employing an electrochemical workstation (CorrTest CS2350, China). 30 A platinum electrode and a saturated calomel electrode were acted as the counter electrode and reference electrode. Working electrode is the sample. The potentiodynamic polarization test was performed at a scanning rate of 0.5 mV/s, after the open circuit potential (OCP) of tested sample was monitored for 1800 s. Also, the electrochemical impedance spectroscopy (EIS) of samples was carried out between 100 kHz to 0.01 Hz at the OCP. Additionally, the EIS of samples was monitored at the end of 1 d, 3 d, 5 d and 7 d of the immersion, to find out the change of corrosion resistance with time.
Biocompatibility testing
In vitro cell viability and proliferation assays were conducted to assess the biocompatibility of the bare and coated AZ31, using human umbilical vein endothelial cell lines (HUVEC, Ea.hy926). First, samples were immersed in complete culture medium (Roswell Park Memorial Institute [RPMI] 1640) for 72 h with a ratio of 6 mL/cm2 to obtain the extracts, according to ISO 10993-12:2021 standard. Then, cells were cultured in 24-well plates (6000 cells per well) for 4 h, thereafter the complete culture medium was replaced by equal extracts. After 24 and 72 h of culture, the cell viability rates were measured by Cell Counting Kit-8 (CCK-8). Meanwhile, fluorescent images of cells were taken to present the living/dead staining results.
Results and discussion
Coating characteristics
Figure 1(a) is the XPS spectrum of PTFE coating, which shows that the coating surface mainly contains C and F elements. It should be noted that a small amount of O element and a part of C element may be related to the adsorption when the sample exposed to air. Large number of F elements were found, but elements such as magnesium, aluminum and zinc in AZ31 alloy were not detected. We speculate that chemical bond between F and metal elements may not be formed.

XPS spectrum (a), FT-IR spectrum (b) and XRD patterns (c) of PTFE coating.
Figure 1(b) shows the fourier transform infrared spectrum of PTFE coating. The absorption peak appearing at wave number 3439 cm−1 represents the stretching vibration peak of —OH. The absorption peak with a wave number of 2869—2960 cm−1 is the stretching vibration peak of the C—H group. The absorption peak at wave number of 2381 cm−1 is the asymmetric stretching vibration peak of the cumulative double bond of —C = C—. There exists the following chemical reaction equation between tetrafluoroethylene and polytetrafluoroethylene: n[F2C = CF2] ↔ —(CF2—CF2)n—. The presence of —C = C— and the weak corresponding peaks indicate the presence of a small quantity of tetrafluoroethylene in the PTFE coating. The absorption peaks with wave numbers 1215 cm−1 and 1153 cm−1 are the stretching vibration peak of C—F and the stretching vibration peak of C—C skeleton in polytetrafluoroethylene. The absorption peak with a wave number of 640 cm−1 is the bending vibration peak of the C—H group. It indicates that a good PTFE coating has been formed on AZ31 surface by RF magnetron sputtering.
Figure 1(c) shows the XRD patterns of PTFE coated AZ31 and the PTFE target alone. It is found that the PTFE target is a crystalline polymer, and there are several main diffraction peaks at 18.1°, 31.5°, 37.1° and 41.3°, which correspond to the standard card PDF#47-2217. However, no diffraction peaks of PTFE were detected in PTFE coated AZ31 sample, but all the diffraction peaks came from AZ31 matrix, which may be due to the failure to reach crystallization conditions of PTFE under sputtering conditions at room temperature. And on the other hand, it may be related to the too small coating thickness.
Figure 2 presents the surface microstructure of PTFE coating prepared on AZ31 substrate after 3 h sputtering. It can been seen that the as-prepared PTFE coating was very dense and flat. AFM results revealed the three-dimensional island growth of the PTFE coating, as shown in Figure 3(a) and (b). The size of deposited particle was below 100 nm and the coating showed a smooth surface with roughness (Ra) of 7.8 nm. It is believed that the coating is tightly bonded to the AZ31 surface, which determined by the nature of sputtering. When charged Ar+ particles with high energy bombard the solid surface of target material (PTFE target, as cathode) under the action of the electric field, surface atom collisions occur, causing the target material atoms or molecules to escape from the surface and deposit on the substrate material (Mg substrate, as anode). Ultimately, the deposited atoms or molecules condensed and a coating formed on surface of the substrate.

SEM morphologies of PTFE coating (a) low magnification, (b) high magnification.

The AFM micrograph of the PTFE coating (a,b) and contact angles of AZ31 substrate (c), PTFE coating (d). AFM, atomic force microscope.
As is well known, PTFE is a hydrophobic material, and the microscopic surface morphological feature can also caused a change of wettability. As the results shown in Figure 3(c) and (d), the water contact angle of original AZ31 surface increased from 53.2° to 114.2° after being coated with PTFE, revealing a increased hydrophobicity. Because the surface of PTFE is covered with carbon-fluorine bonds (C-F, with a low surface energy), it has remarkable hydrophobicity. In the similar study of reference, 20 PTFE was sprayed on the surface of micro-arc oxidation layer to form a MAO-based composite coating on AZ31 magnesium alloy, and the composite coating exhibited a hydrophobic surface with a contact angle of 101°, close to the result in this paper. Besides, many research findings revealed that by proper treatment (such as laser etching, etc.), the specific micro-nano composite structure can be formed on the surface, which can improve the hydrophobicity of PTFE.
Figure 4 is the white light interference test results of PTFE coating. It clearly revealed that the thickness of the coating is near 100 nm.

White light interference test results of PTFE coating.
Corrosion behavior in vitro
Figure 5 shows the potentiodynamic polarization curves of the PTFE coated AZ31 and the bare AZ31 samples. Normally, the corrosion resistance can be reflected in Ecorr (corrosion potential) and Icorr (corrosion current density), which can be obtained from polarization curves. A larger Ecorr suggests a smaller corrosion tendency, and a lower Icorr represents a lower corrosion rate and better corrosion resistance. As shown in Figure 5, the corrosion potentials of PTFE coated AZ31 and bare AZ31 were −1.58 V and −1.60 V respectively. The PTFE coated samples showed a more positive Ecorr than the bare AZ31 samples, which improved by 20 mV. Thus, it can be determined that to some extent the PTFE coating could retard the corrosion progress of AZ31 substrate in the same corrosive environment.

Potentiodynamic polarization curves of AZ31 coated with PTFE and bare AZ31 substrate in SBF.
Icorr is a key index to judge the corrosion behavior of materials. Figure 5 shows that the Icorr of PTFE coated sample reduced from 1.92 × 10−4 A/cm2 to 1.07 × 10−8 A/cm2, compared to the bare AZ31 sample. The data clearly suggested that the corrosion resistance of AZ31 was dramatically enhanced by the thin PTFE coating. During the potentiodynamic test of the PTFE coating, there is a sharp inflection point on the anodic region of the polarization curve (marked as P). It might be caused by pitting corrosion of the PTFE coating.
Corrosion of magnesium will increase the OH– concentration and the pH value of the SBF solution. Figure 6 shows the change of pH value with immersion time. It's found that the PTFE coated AZ31 group showed lower pH values throughout the immersion time. At the end of 7 d immersion, the pH was about 8.4 with a small increase. However, the bare AZ31 group exhibited constantly aggravated corrosion with the immersion time, and the severe degradation induced higher pH values of the SBF. After 7 d immersion in SBF, the bare AZ31 group showed a high pH at 9.5. The PTFE coated group has the lower pH and better corrosion resistance, in marked contrast to the uncoated group. This is consistent with the conclusion of the potentiodynamic polarization tests.

pH changes of SBF solution in which samples immersed over time up to 7 days.
The corrosion morphologies of two groups of experimental samples during 7 days are shown in Figure 7 and 8 respectively. As the immersion time prolongs, surface of uncoated AZ31 was destroyed and seriously corroded finally. Compare with the original polished surface, there were many obvious cracks and abundant corrosion products deposited. As known, the corrosion of magnesium alloy in solution environment is more serious than that in air. when magnesium is immersed in SBF, insoluble magnesium hydroxide formed and corrosion pits will soon appear on the surface. 31 Magnesium hydroxide on the surface of magnesium alloy is unstable, so the internal magnesium will also be corroded. Affected by the film density, the magnesium hydroxide film on the surface of magnesium will still be partially dissolved in the solution containing strong corrosive ions (such as Cl− ions). At the same time, magnesium hydroxide generated when magnesium ions in the solution meet hydroxide ions may return to the surface of the substrate, and the hydrogen generated during the corrosion process will affect the quality of the newly formed surface film. Therefore, the film deposited in this way is loose and does not play any protective role.

SEM images of bare AZ31 samples after 1 d, 3 d, 5 d and 7 d immersion in SBF. SEM, scanning electron microscope.

SEM images of PTFE coated AZ31 samples after 1 d, 3 d, 5 d, and 7 d immersion in SBF. SEM, scanning electron microscope.
By contrast, the corrosion of the PTFE coated AZ31 was milder. Figure 8 shows that the corrosion appearance of PTFE coated AZ31 was uniform and intact on a large scale range. After 7 d immersion in SBF, the PTFE coated surface was basically flat. Exiguous corrosion products can be seen on the surface. Also, the PTFE coating was not cracked or detached from the substrate, even if corrosion products were produced. The comparisons above show that the PTFE coating could effectively prevent the corrosion of magnesium alloy in SBF solution. But, it should be noted that the tracks of coating delamination were observed after samples immersing in SBF for 3 days, which means that hydrogen evolution increases with the exposure of the samples to the media.
In addition to producing magnesium hydroxide, hydrogen will be produced simultaneously in the corrosion process of magnesium alloy. For the bare AZ31 samples, as soon as they were put into SBF, hydrogen bubbles was observed immediately. However, for the PTFE coated samples, no obvious hydrogen evolution was observed, especially in the early stage of immersion.
The EIS can be adopted to analyze the corrosion of materials. 32 The Nyquist plots is an important type of EIS data. Figure 9 is the Nyquist plots of PTFE coated AZ31 after different immersion time in SBF solution. According to literature, 14 the corrosion behavior of the tested material have a relationship with the capacitive loops in the Nyquist plots. In Figure 9, the Nyquist plots of PTFE coated AZ31 only have a single capacitive loop within 1 day, which indicates the substrate had not been corroded by electrolyte. But after 3 day immersion, there are two capacitive loops in the Nyquist plots of PTFE coated AZ31. In this situation, the electrolyte (SBF) maybe permeate through the PTFE coating and substrate has been corroded in the very early phase. Throughout the immersion tests, the capacitive loops was regularly narrowed, revealing a decline of corrosion resistance of PTFE coating. The results are consistent with the surface appearance, as represented in Figure 8.

Nyquist plots of PTFE coated AZ31 samples after immersion in SBF for different time.
We compared the above corrosion test results with those of the plasma electrolytic oxidation (PEO) coating reported in literature, 33 in which the authors employed the same AZ31 matrix and corrosion test method. By comparing the thickness of the coating, it is found that the thickness of PTFE coating is much less than 1 micron, while the PEO coating has about 60 microns thickness. However, corrosion test results show that the Icorr of PTFE coated AZ31 is 1.07 × 10−8 A/cm2, while that of PEO coating is 1.21 × 10−6 A/cm2. Icorr of the two coatings showed a significant difference. Moreover, after the same immersion time of 7 days, PTFE coating remained intact, while the PEO coating showed obvious cracking. Based on the above analysis, we think that although the thickness of PTFE coating is smaller, it shows better corrosion resistance. The main reason may be that PTFE is more hydrophobic and the PTFE coating can still remain intact after immersion in the SBF.
Based on the results and discussion as mentioned above, the better corrosion resistance of the PTFE coating is mainly conditional on the chemically stability, strong combination with substrate, also the hydrophobicity, making the coating more resistant to the penetration of corrosive media.
Cytocompatibility tests
A promising material for biomedical implant application should has enhanced biocompatibility. Compared with corrosion resistance, the biocompatibility is equally important for magnesium alloy implants.33,34 Here, in-vitro cell viability of PTFE coated AZ31 was evaluated.
Figure 10 shows the viability of endothelial cells during incubation with various extracts after 1 and 3 days. After 1 and 3 d culture, the cell viability of the bare AZ31 was 78.9% and 76.6%, respectively (vs. 98.7% and 96.3% of the PTFE coated sample). According to ISO10993-5 standard, the PTFE coated AZ31 has grade 1 cytotoxicity, thus it can be considered that PTFE coated AZ31 was biocompatible. It is thought that bare Mg alloy was corroded rapidly, with a high pH of the extract, which was unfavorable to cell proliferation. The results are consistent with that of Esmaili. 33

Cell viability of Ea.hy926 cultured for 1 d and 3 d in different extracts of bare AZ31 and PTFE coated AZ31.
Figure 11 shows the fluorescent images of Ea.hy926 cells after culturing for 1 d and 3 d in extracts of the bare and PTFE coated AZ31. At each time point, the number of cells in PTFE coated AZ31 extract was significantly higher than that of bare AZ31. After 3 d culture, the cells of PTFE coated AZ31 almost overspread the culture plates, with a fusiform shape. The results show that the PTFE coated AZ31 showed good cytocompatibility.

Fluorescent images of Ea.hy926 cells after culturing for 1 d and 3 d in extracts of the bare and PTFE coated AZ31.
In this study, the good corrosion resistance and biocompatibility in the field of biomedical magnesium alloys have been proved. It's believed that there is great potential of the PTFE coating for Mg alloy medical device applications.
Conclusions
In this work, PTFE coating was obtained on AZ31 alloy substrate via RF magnetron sputtering. The corrosion behavior and cytocompatibility of the PTFE coated AZ31 have been investigated. The main conclusions can be drawn as follows.
At room temperature after 3 h sputtering, 100 nm thick PTFE coating was prepared on AZ31 surface. The PTFE coating was dense and had a flat surface with obvious hydrophobicity. At the microscopic scale, the PTFE coating was composed of nanoparticles. After being coated by PTFE, the AZ31 sample showed improved corrosion resistance. Corrosion tests indicated that PTFE coated AZ31 has a very low corrosion current density. The surface of PTFE coated AZ31 was less corroded after 7 day immersion in SBF. After 1 and 3 d culture, the PTFE coated AZ31 showed better cell morphology and higher cell proliferation rate (98.7% and 96.3%, respectively).
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by Key Scientific Research Project of College and University of Henan Province, China (Grant No. 22A430035).
