Surface treatments for controlling corrosion rate of biodegradable Mg and Mg-based alloy implants

Drynda et al [44] developed a magnesium fluoride (MgF₂) coated Mg-Ca alloy, in which MgF₂ was deposited onto Mg–Ca alloy, and tested the corrosion performance in a saline (e.g. NaCl) solution. The degradation rate of the alloy was evaluated by measuring hydrogen generation. It was reported that the MgF₂ coated alloy degraded slowly as compared to the uncoated alloy (Mg–Ca alloy) and no hydrogen gas was detected from 8 to 40 h. Thomann et al [49] applied the same coating on a Mg–Ca0.8 alloy and found that the fluoride coating on the Mg–Ca implants slowed down the corrosion rate and retained higher mechanical properties for a long period of time (even after 6 months) than its uncoated counterpart. Similar effects of fluoride coatings on Mg-alloys in terms of improving corrosion and mechanical integrity performance were examined and reported by other researchers [50]. Figure 6 shows surface morphologies and electrochemical impedance spectroscopy (EIS) spectra results for uncoated and fluoride-coated AZ31 samples in a simulated blood plasma [51]. It is clear that magnesium fluoride coatings have been successful in modifying magnesium-based implants by both in vitro and in vivo assessments, and become a potential candidate for biomedical applications. However, it must be noted that caution must be taken as hydrofluoric acid typically used for MgF₂ coating generation may be harmful to operators as well as ensuing environments. Fluoride salt can be a potential alternative in this case, which requires further comprehensive investigations to confirm its use and benefits.

Zoom In Zoom Out Reset image size

In an aqueous environment like human bodily fluid, the Mg alloy forms Mg(OH)₂ which acts mainly as a protective layer and prevents corrosion. Alkaline treatments of Mg-alloys are thus used to create more of this layer on the implant surface. Gu et al [52], Li et al [53], and Zhu et al [54] studied the effect of alkaline heat treatments on Mg–Ca alloys. For instance, Gu et al [52] reported that after soaking the Mg–Ca alloy in three alkaline solutions (Na₂HPO₄, Na₂CO₃, and NaHCO₃) for 24 h and heat treating at 500 °C for 12 h, the alloy underwent a slower degradation rate (figure 7(a)). This is due to the thicker magnesium oxide films generated on the surface by alkaline treatments, in which the thicknesses of oxide layers on the Mg–Ca alloy substrate formed by Na₂HPO₄, Na₂CO₃, and NaHCO₃ treatments, are 13 μm, 9 μm and 26 μm, respectively. Furthermore, cytotoxicity tests revealed that none of the alkaline heat-treated Mg–Ca alloys induced toxicity in cells. In another study by Zhu et al [54], the AZ31 alloy was treated with 5.66 wt.% NaOH solution at 160 °C to generate a protective Mg(OH)₂ film. Tests performed in Hank's solution for up to 4 h showed that the corrosion rate of the treated alloy was inhibited effectively (figure 7(b)). They also found no effect on the cell growth or changes in cell morphology due to the additional oxide films.

Zoom In Zoom Out Reset image size

As an organic coating, CaP-based coating is of the highest interest and investigated to enhance the corrosion resistance of magnesium and magnesium alloys. There are many methods used to produce CaP coatings on the surface of a Mg-alloy. For instance, Cui et al [13] applied CaP coatings on a Mg–Ca1.0 alloy using electrochemical deposition. After performing electrochemical tests in Hank's solutions, CaP coated Mg–Ca alloy exhibited an increase in corrosion resistance as can be seen in (figure 8(a)). Further, the coated sample revealed higher hydrogen evolution rate than the uncoated sample. However, the corrosion rate increased as the pits were generated in the coatings. CaP coating was also shown to be beneficial for biocompatibility. However, the CaP layer formed on Mg–Ca alloy surface was an amorphous mixed (Mg, Ca)-phosphate which can be carbonated and hydrated [45].

Zoom In Zoom Out Reset image size

CaP-based hydroxyapatite coatings are widely used in biomedical implants due to their chemical and structural similarities with bone materials. For instance, Song et al [55] developed hydroxyapatite (HA, Ca₁₀(PO₄)₆(OH)₂) and fluoridated hydroxyapatite (FHA, Ca₅(PO₄)₃(OH)_1−xF_x) coatings on a Mg–Zn alloy and assessed their corrosion performance in simulated body fluid (SBF). HA and FHA coatings were found to delay the corrosion rate and promote the nucleation of osteo-conductive minerals for one month (figure 8(b)). Notably, as can be seen in figure 8(b), compared to HA coating, FHA showed more stable and improved corrosion resistance. Similar improvements in terms of cell growth and osteointegration were reported in a previous study by Li et al [56]. Furthermore, Kannan and Orr [57] studied the effect of potentiostatically alkaline treated hydroxyapatite coating on a AZ91 alloy and the results suggested that the corrosion resistance and biomechanical integrity can be improved by 20% using low cathodic voltage hydroxyapatite coatings, as compared to the uncoated samples.

A coating made of nanoparticles of TiO₂ was applied to a Mg–Ca alloy to increase the corrosion resistance. Li et al [58] and White et al [59] studied the effect of TiO₂ coating on the corrosion performance. For example, Li et al [58] reported that a TiO₂ coated Mg–Ca alloy showed delayed corrosion as compared to the uncoated alloy surface. Further, the coated alloy exhibited three times lower corrosion current density than the uncoated counterpart. Similarly, White et al [59] prepared and applied a TiO₂ thin layer on AZ31 Mg alloy and tested the sample in NaCl solutions. XRD and SEM evaluation of test samples indicated that the oxide films of MgO and Mg₂TiO₄ generated due to TiO₂ coating were uniform, porous and compact, which significantly improved the corrosion resistance as compared to the fresh AZ31 sample (figure 9(a)). However, the application of TiO₂ onto a Mg-alloy needs to be further investigated to increase the catalytic activity and to ensure the coated alloy is biocompatible and environmentally safe for its dedicated use in biomedical implants.

Zoom In Zoom Out Reset image size

Razavi et al [60] prepared and applied a CaMgSi₂O₆ coating on the AZ91 Mg alloy. In this study, using micro-arc oxidation (MAO) they first created a rough and porous structure on the Mg substrate, which provided satisfactory sites for depositing nanostructural CaMgSi₂O₆ diopside powders and increasing interfacial bonding strength. In the next step, they deposited CaMgSi₂O₆ diopside powders using an electrophoretic deposition (EPD). Electrochemical evaluation in SBF environment showed that the diopside-coated magnesium alloy improved the corrosion resistance and in vitro bioactivity (figure 9(b)). For instance, after immersion in SBF for 72 h, the corrosion rate of pure AZ91 and diopside-coated samples was 0.57 and 0.14 mg cm⁻² h⁻¹, respectively (see figure 9(b)). In particular, the initial degradation rate for AZ91 in first 72 h was 0.08 mg cm⁻² h⁻¹, which was pretty much similar to the one investigated by Song et al [18]; furthermore, this indicated that the results of immersion tests were consistent with those of electrochemical tests. Razavi et al [63] also studied akermanite (Ca₂MgSi₂O₇) coating on Mg alloys and found a similar improvement by the coating in terms of reducing the ecorrosion rate and preserving  biocompatibility.

As a bio-inert ceramic material, zirconia-based thin layers are widely applied on the surface of implants. It facilitates fixing the implant with the surrounding tissues without producing any chemical or biological bonds. It is reported that a bone-like apatite is produced on the surface of zirconium layer, which is responsible for the improved interfacial strength between the bone and the implant. Previously many researchers studied the effect of Zr and ZrO₂ based coating on Mg alloy to adjust the corrosion rate [18, 61, 64–66]. For instance, using cathodic arc deposition process, Xin et al [61] developed a 1.5 μm thick Zr–ZrO₂ dual layer on the AZ91 Mg alloy and tested its mechanical and corrosion behaviour in SBF. Results showed that the Vickers hardness of the coated sample was increased by about 88% as compared to the uncoated alloy substrate. Immersion and electrochemical tests (for up to 18 h) show that the corrosion resistance of the coated sample increased significantly (figure 9(c)) while electrolyte penetration was found to damage the protective coating layer after long exposure in SBF. Song et al [18] applied a compact and uniform Ni–P–ZrO₂ composite coating on AZ91D magnesium alloys, in which at first a Ni–P layer was deposited on the Mg alloy substrate and then another layer made of ZrO₂ nanoparticles was made on the Ni–P layer. Electrochemical evaluation indicated that the Ni–P–ZrO₂ coating exhibited better corrosion resistance than that of Ni–P coating due to additional ZrO2 together with Ni–P. In addition, the Ni–P–ZrO₂ coating enhanced hardness and wear resistance as compared to the uncoated sample. In fact, the initial Ni–P layer provided solution and precipitation strength while ZrO₂ facilitated dispersion hardening into Ni–P, making the overall composite coating harder than the Ni–P layer alone.

Dual layer inorganic composite coatings were studied recently by researchers to adjust the corrosion rate of Mg-based alloys [62, 67]. Using an aqueous phosphating bath method, Singh et al [62] deposited homogeneous films of CaP and silicate on two magnesium alloy systems, AZ31 and Mg-4Y. In vitro and bioactivity tests, in which the samples were immersed in polymyxin B1 (PB1), polymyxin B2 (PB2), and polymyxin B3 (PB3) (polymyxin—a mixture of several closely related polypeptides) for up to 3 days of cell culture, showed that the silicate-based CaP coatings decreased the degradation of AZ31 while they increased the degradation of Mg-4Y (figure 9(d)). In addition, silicate–CaP coatings were observed to increase cell attachment on the AZ31 alloy. Further studies reported that silicon-substituted CaP had the ability to retain osteoclasts over a short period of time and increased the strength of the bond between the bone and implant as compared to pure CaP coated samples [68].

The chemical conversion approach is generally applied to form substrate-involving coatings. In this case, the thin layers are developed in situ via the chemical reaction between the Mg or Mg alloy substrate and the treatment agent. A detailed review of chemical conversion treatments is presented by Chen et al [70], and it is suggested that proper pre-treatment to functionalise the surface is crucial to determine the overall coating performance. Mg(OH)₂ and fluoride based coatings are two of the most widely used coatings developed by chemical conversion treatments, and are shown to increase the corrosion resistance along with no signs of induced toxicity of surrounding cells.

For instance, Gu et al [52] studied alkaline-based coatings, in which the magnesium substrate was soaked in three alkaline solutions (NaHCO_3, Na₂HPO₄ and Na₂CO_3,) and heat treated at a certain temperature. Electrochemical and cytotoxicity tests showed that all the coated samples decreased the corrosion rate and revealed no signs of toxicity. Ultrapure water, hydrofluoric acid, phytic acid and phosphor were also employed as the conversion treatment agents to form coatings which eventually reduced the corrosion rate and facilitated the cell growth surrounding the implant [49, 69, 71]. Further, rare earth elements such as Ce and Nd were used to form the conversion layers on Mg-based alloys. It was reported that a small amount of rare earth material was not detrimental to human health. Cui et al [72] and Levy and Aghion [73] developed Ce and Nd-based diffusion coatings on magnesium alloy substrate (e.g. AZ31) and found that both coatings enhanced the corrosion resistance significantly. However, caution must be taken about whether these rare earth materials may pose any toxicity when employed over a longer period of time. Chemical conversion coating is still considered a cost-effective technique, and therefore it is important to make use of the method to wider applications within the biomedical field.

MAO, often also called plasma electrolytic oxidation (PEO), is another widely used technique to form porous and robust thin coatings with high adhesion strength onto magnesium and its alloys. The typical chromate coating is no longer available due to a ban imposed by EU regulatory authorities. MAO has been regarded as the most commercially viable route for developing corrosion protective layers on Mg/Mg-based alloys. MAO consists of an electrolytic bath, a working electrode consisting of an electrically connected Mg alloy and a stainless steel counter-electrode. The quality of the produced layer heavily relies on the bath, alloy properties, and processing parameters, e.g. reaction time, applied voltage [48]. CaP and ceramic based coating such as silicate and ZrO₂ are typically developed by using MAO. It is found that MAO fabricated coating works as an intermediate layer for depositing nano-sized HA coating because it can produce pinning force when HA is deposited in the pores. Per Gao et al [74], the MAO coating was fabricated under a pulse voltage mode, and the cell potential was increased gradually up to 165–175 V and then the working electrode was oxidised for 30 min. The electrolyte was prepared by using a range of solution mixtures. For instance, Gu et al [75] prepared the electrolyte from the solution of 10 g L⁻¹ sodium silicate with 3.5 g L⁻¹ sodium. In another study by Shi et al [76], MAO was performed in 500 mL aqueous solution containing 50 g L⁻¹ NaOH, 40 g L⁻¹ Na₂SiO₃, 20 g L⁻¹ Na₂B₄O₇ and 40 g L⁻¹ Na₃C₆H₅O₇ at a constant potential of 90 V for about 140 min. However, the disadvantages with MAO process are that surface pores and cracks formed during MAO accelerate the corrosion rate in a chloride-rich physiological environment. Further, the main phase of coating is MgO which is often not suitable for  effective cell growth.

ED, a non-substrate-involving coating technique, is one of the most recognised and widely used coating formation methods for biodegradable magnesium and its alloys. Though the process is mainly concerned with the deposition of inorganic phases, the coating produced can be achieved via both purely deposition and, to some extent, the conversion at the interface. It has the capability of forming uniform coatings, controllability of maintaining the required thickness and chemical composition of the coating, and most importantly, low deposition temperature [45]. Cathodic ED, a traditional process, is commonly used to deposit CaP based on coatings. However, HA coatings made by cathodic ED are sometimes fragile and less stable, hence degrading its long term corrosion resistance. Therefore, a modified pulse electrodeposition was recently employed to fabricate CaP coatings. Using this technique, Wang et al [77] developed Ca-deficient HA (Ca-def HA) coating on a Mg–Zn–Ca alloy and found that both corrosion resistance and mechanical properties such as ultimate tensile strength (and time to fracture) were improved by the Ca-def HA coatings. It was also shown that, compared to traditional cathodic electrodeposition, the pulse electrodeposition process assisted more effectively in precipitating magnesium, calcium and oxide ions, which created a strong, uniform and compact protective layer, thus preventing the magnesium alloy substrate from further corrosion. Further, Wen et al [78] and Kannan and Orr [57] studied multi-step ED processes to produce HA-based coatings on AZ31 Mg alloys, in which alkaline treatments using NaOH solution produce a fine microstructure. Electrodeposition is then used to modify or transform the microstructure from platelets to nanowiskers or rod-shaped particles. In vitro evaluation indicates that the coating produced by the modified ED process enhances mechanical strength significantly.

Anodisation is another coating technique which uses the concept of classical electrochemical conversion, in which a film having a thickness of ranging from 5–200 μm is developed on the substrate material. During anodisation, the metal (of coating material) as anode is transformed into oxide film with controlled corrosion protective, decorative, and functional properties. As reported by many researchers, the anodising performance of Mg and its alloys generally depends on a range of key parameters of the process. They are: electrolytic composition (e.g. aqueous or non-aqueous solutions), applied constant voltage or current, substrate types, quality of Mg and alloying element concentrations [45]. In order to further improve the anodisation performance, self-ordering electrochemical processes have been studied. The benefit of the process is that it can produce nano-structured oxide films from aqueous and non-aqueous solutions, and even under certain optimized conditions, a nano-tubular porous structure is created on the substrate material. However, it was reported that Mg alloys facilitated in forming nano-structured oxide films in non-aqueous solutions rather than aqueous solutions [79]. The degree of quality of the corrosion resistant oxide layers depends on the anodisation parameters. For example, appropriate adjustment of the process parameters such as electrolytic concentration, current density, and anodisation time, affects the quality of coating structure, including the density of porosity and crystallinity [48]. While anodisation has a significant potential in creating corrosion resistive oxide layers, the process may need to be further investigated to explore and determine the optimized parameters which will able to develop a well-defined and structured coating, thus enhancing corrosion resistance, cell growth (i.e. interaction between cells and films) and biocompatibility.

To further improve the corrosion resistance, Mg and its alloys have been ion implanted as reported by previous research [48]. Ion implantation involves a process in which ions are accelerated and impinged into the modified surface. As opposed to the surface coatings techniques described above, as an advanced technique, the ion implantation process introduces ions into the substrate without requiring any electrolytic solutions. Further, an ion-implanted layer does not have an abrupt interface and hence potential layer fracture or delamination is not likely to occur [69]. As the process only can produce a very thin layer of the modified surface, generally the corrosion resistance is not adequate in the long term [16, 80, 81]. In another study by Wu et al [82], ion implantation was used a pre-treatment process to improve the adhesion strength between a Ti coating and the AZ31 substrate. Depending on the implanted elements, ion implantation for Mg modification is categorised into three types—gas ion implantation, metal ion implantation, and dual ion implantation. Organic elements such as oxygen, nitrogen, and carbon dioxide gases are implanted into Mg and its alloy substrates and the surface modification is found to improve the corrosion resistance of the alloy (see figure 10(a)) [83, 84]. While gas ion implantation creates an oxygen or nitrogen-rich layer on the substrate, metal ion implantation introduces metal elements to form surface alloying into magnesium and its alloy substrate. Al, Zn, Zr, Ti and Ta are few metallic elements which are often ion implanted into magnesium alloys (e.g. AZ31, AZ91) [16, 85]. The modified layer with metal ion implantation enhances the corrosion resistance (see figure10(b)). The mechanism for improved corrosion resistance is ascribed to the formation of intermixed layer consisting of an outer oxide layer (of MgO or Mg(OH)₂), an intermediate metallic layer and a bottom layer rich in metallic elements [85]. While these elements are found to be individually biologically friendly to the human body, further investigations are needed to ensure their biocompatibility when the composite layer consisting of multiple metals are implanted. In order to develop an oxygen rich metal oxide layer, metal and oxygen dual ion implantation has been employed to achieve better corrosion resistance than either gas ion implantation or metal ion implantation used individually. For instance, Zhao et al [86] used Al and O ion implantation to create an Al₂O₃ protective layer and found that the modified layer delayed the degradation of the substrate (figure 10(c)). Similar improvements were observed for Ti–O, Zr–O, and Cr–O dual ion implanted surfaces studied by other researchers [87–89]. Furthermore, the detailed description of different types of ion implantation can be found in a comprehensive review performed by Tian and Liu [69]. Despite its unique benefits as outlined above, the ion implantation technique is generally costly and often may not be suitable for intricate geometries of components such as implants and porous scaffolds.

Zoom In Zoom Out Reset image size

While traditional coating techniques, including chemical conversions, electrochemical treatment and ion implantation as described above, are widely studied, the results in improving the corrosion resistance are variable and often lack the bioactive properties necessary for controlling cellular behaviour [90]. Biomimetic coating is a novel and promising way to significantly enhance corrosion resistance and biocompatibility. It is a biologically derived coating which provides cell signalling capabilities for better tissue generation. For instance, Cui et al [90] comprehensively developed biomimetic coatings based on highly acidic bone protein, known as dentin sialophosphorprotein (DSPP), on a Mg alloy and investigated their corrosion and biological performance. DSPP is a non-collagenous extracellular matrix (ECM) protein found in bone and teeth while phosphophoryn (PP) is a cleavage product of DSPP which provides dentine mineralization and cell signalling in bone and dentine. As technical isolation of PP protein is quite difficult because of rapid degradation, the same authors [13] followed the principles of bio-mimicry and developed a biomimetic coating based on a peptide motif using the amino acid sequence of PP. The peptide is composed of three repeats of DSS (Asp–Ser-rich) amino acids. The study is focused on testing the feasibility of coating biomimetic peptides onto a Mg alloy (AZ31B), examining the reduction in corrosion rate and assessing whether the peptide induces CaP precipitation onto the alloy substrate. SEM analysis showed that 3DSS peptide coating on AZ31B was smooth and compact while polishing marks were evident on non-3DSS coated surfaces (see figure 11(a)). Immersion tests in a SBF solution showed that hydrogen volume generation from the 3DSS-coated AZ31B alloy was found to be consistently lower than that from the non-coated alloy. The corrosion potential for the 3DSS-coated sample was 400 mV higher than the non-coated one, while the corrosion current density (of 3.16 × 10⁻⁴ μA) was significantly lower than the uncoated sample (3.161 μA) (figure 11(b)), all of which indicated that 3DSS biomimetic peptide coatings were stable and corrosion resistant. Further, it was reported that the 3DSS peptide-induced CaP coating on AZ31B alloys showed a larger and more organised crystal-like structure as compared to that without the facilitation of 3DSS biomimetic peptide.

Zoom In Zoom Out Reset image size

VD is one of the well-known and commercially employed coating techniques, in which, a protective layer is deposited on the substrate. Due to its capability in producing highly-dense nano-structured coatings, the technique is widely used to deposit a hard coating layer on the substrate material to improve wear and corrosion resistance. As compared to electro-chemical coating processes such as MAO, the PVD technique can produce thick and strongly adhering layers on the substrate. Studies have shown that the corrosion can be affected by the coating material selection and process parameters [91]. Magnetron sputtering also enables the formation of metal alloys that are not accessible via conventional processing techniques [92]. This so called vapour quenching process has the potential to generate a corrosion resistive coating on Mg alloys for which the process has not been previously studied. In the past, using the magnetron sputtering method, many studies explored the application of hard coatings on Mg/Mg-based alloys to improve corrosion resistance as well as wear resistance, in which the typical multi-layer coatings used are TiN, CrN, TiAlN, TiO₂, Al₂O₃ and DLC [93–95]. For instance, Altun and Sen [96] developed multi-layer AlN and AlN+TiN on a AZ91 magnesium alloy using PVD DC magnetron sputtering and reported that the multilayer coating increased the corrosion resistance significantly. Researchers studied the growth and characteristics of a dense nano-structure biocompatible CaP-based HA coating deposited using a PVD technique on orthopaedic implant surfaces [97, 98]. Most of the recent studies focused on depositing the sputtered HA coating on the surface of metallic implants including titanium/titanium alloys, NiTi, stainless steel, and silicon plates [98–100]. It was reported that the HA films deposited on such metallic surfaces using RF magnetron sputtering exhibit a highly dense and nano-composite structure, providing good biocompatibility and mechanical integrity [101]. Recently, Surmeneva et al [102] employed the RF magnetron PVD technique to deposit CaP based HA coating on bioabsorbable AZ31 magnesium alloys and reported an increase in the wear resistance and hardness of the alloy. They found that by applying a negative bias (−50 V) on the substrate during deposition process reduces the grain and crystallites size, hence further increasing the hardness of the coating.

In order to reduce the acceleration of galvanic corrosion between the substrate and protective coating produced by PVD, Hoche et al [103] studied DC magnetron sputtering (DCMS) and high power impulse magnetron sputtering (HiPIMS) to produce a coating of TiMgN with rare earth elements such as Y and Gd on a AZ31hp Mg alloy. It was shown that compared to the DCMS and TiMgYN coating, the TiMgGdN coating with HiPIMS technique exhibited excellent corrosion resistance. As compared to DCMS, HiPIMS was found to increase the grain refinement, hence reducing corrosion initiation sites such as pinholes, clusters, and droplets. On the other hand, the addition of Gd into TiMgN causes the formation of dense and stable oxide films, which prevent galvanic corrosion at the interface between the coating layer and the substrate. In a similar study, Xie et al [104] investigated the gradient duplex coating combining nitrogen ion implantation and AlN–MoS₂–phenolic resin produced by the RF magnetron sputtering method and reported that the AlN interlayer, together with ion implantation, significantly improves the tribo-corrosion resistance of the Mg alloy. While ion implantation creates more dense oxide films, AlN–MoS₂ retards the galvanic corrosion between the coating and the substrate. While grain refinement is shown to improve the corrosion resistance, other factors such as surface free energy and residual stress within the protective oxide films may impact the corrosion performance [105], which is still considered to be a topic of interest and needs to be further investigated. Plasma anodisation together with RF magnetron sputtering was applied to produce hard coatings such as CrN and Al₂O₃ [106], and the corrosion behaviour was investigated at macroscopic and microscopic levels. In this case, at first a 0.05 μm thick plasma anodised layer is deposited on the substrate, and the process is similar to the anodisation in an aqueous electrolyte by replacing the electrolyte with oxygen plasma. The final step is to apply a relatively thick (4–5 μm) layer of hard coating by conventional magnetron sputtering.

It is therefore clear that the performance of the coating developed by PVD (e.g. magnetron sputtering) depends not only on the choice and integration of appropriate coating materials but also on the coating process conditions including power pulse and bias. As a potential technique, PVD can thus be employed to produce a stable, strong coating which can delay the degradation of orthopaedic implants made of Mg alloys. A hybrid coating process combining RF/DC magnetron sputtering together with additional ion implantation or plasma anodisation would potentially further increase the corrosion resistance of the alloys. Along with proper choice of coating materials, the process parameters including applied pulse power and bias negative voltage, are critical factors impacting the coating performance. Further research should focus on how PVD can robustly be applied to produce CaP HA and biomimetic coatings, which are considered potentially new and accepted coating materials for the control of corrosion process in Mg alloy-based implants.

As a simple and inexpensive method, shot peening (SP) is widely used to improve the fatigue and the corrosion performance of structural metallic alloys [107–109]. SP is a cold surface work hardening process which introduces near surface compressive residual stress into the material to prevent fatigue crack initiation and growth. With an appropriate choice of SP process parameters, fatigue performance can be improved significantly. For instance, Liu et al [110], Wagner et al [111], and Fouad et al [112] have applied SP on Mg-based alloys, e.g. Mg–10Gd–3Y, AZ31 and AZ80, and ZK60 alloy, and reported that the surface modifications with SP generated effective compressive stresses which shifted the fatigue crack nucleation site from the surface to subsurface regions, thus improving the overall fatigue strength of the material. Recently, Mhaede et al [113] applied SP with ceramic shots (Z850, Ø850 μm) on the AZ31 magnesium alloy followed by the electro-deposition of a calcium phosphate coating, e.g. dicalcium phosphate dihydrate (DCPD). Microhardness and surface roughness were measured for mechanical assessment while electrochemical tests were performed to evaluate the corrosion performance. Results showed that the hardness of the SP sample increased from 55 up to 101 HV0.025 at a surface depth of 0.20 mm. DCPD coating on the SP surface showed higher surface roughness as compared to SP-only surface. It was reported that the DCPD coating formed on the SP sample was relatively dense and adherent to the surface, which potentially improved the mechanical strength of the coating (see figure 12(a)) and high surface roughness implied a positive result in terms of osseointegration. Polarization resistance of the SP substrate decreased 22–71 times as compared to the fresh substrate (i.e. only ground surface), meaning that SP caused an increased surface energy and induced more residual stresses, thus making the surface more active. However, polarization resistance of the SP + DCPD coating was higher than those without DCPD coating (see figure 12(a)). Furthermore, the SP + DCPD coated sample exhibited a higher corrosion resistance as compared to the ground + DCPD coated sample. This clearly indicates the impact of SP in combination with the coating in controlling the corrosion resistance of Mg-based alloys.

Zoom In Zoom Out Reset image size

Tool-based material removal, i.e. machining, is another mechanical surfacing process to improve corrosion resistance by generating a high finish on machined Mg-based implants. Von der Höh et al [114] employed three different surface treatments on Mg–0.8Ca alloy implants and investigated their effect on the in vivo degradation behaviour of the alloy for a 3–6 month period. The implant surfaces were mechanically processed with turning, cylinder threading, and sand blasting, in which medial roughness was 3.65 μm for the turned sample and 32.7 μm for the sand-blasted one. Different cell growth or bone formation around the implant and hydrogen gas generation were considered as tools to evaluate the degradation process. It was shown that sand-blasted samples degraded the fastest as compared to turned and threaded ones. On the other hand, the turned surface exhibited the lowest structural loss, showing hardly any hydrogen gas formation. In addition, after 3 months the turned implants showed the best bone–implant integration as compared to sand-blasted and thread surfaces. The highest amount of fibrotic tissue along with hyaline cartilaginous was found to grow around the turned sample, as compared to threaded cylinder and sand-blasted implants. It is quite apparent that a smooth surface with fine finishing (i.e. lower surface finish) by the turning process is favourable for achieving osteointegration between the bone and the Mg-based alloy implants.

Using the turning process, Denkena and Lucas [117] attempted to control the degradation rate of a Mg–3.0Ca alloy via modifying the surface integrity including surface roughness and residual stress. They studied the effect of different process parameters on the generation of compressive residual stress into the sub-surface, which were likely to improve the corrosion resistance. It was reported that higher cutting forces due to low cutting speed induced higher and deeper compressive stress and hence decreased the corrosion rate, while a fine surface finish at the same condition was observed to have a negligible impact. Later on, Denkena et al [115] investigated the application of both turning and milling processes along with deep rolling to adjust the in vitro corrosion performance of Mg–0.8Ca and Mg–3.0Ca alloys. Turning followed by deep rolling at a higher force (of 200 N) can induce a significant amount of compressive stress at higher surface depth of 800 μm (figure 12(b)). Milling with a relatively larger cutting edge radius of insert is able to introduce larger compressive stress at higher surface depth as compared to the turning process. While the surface and subsurface properties (i.e. residual compressive stress) from mechanical processing are approximately similar for both Mg–3.0Ca and Mg–0.8Ca alloys, the corrosion behaviour of two differs significantly (figure 12(b)). As can be seen from figure 12(b), turned and deep-rolled Mg–0.8Ca generated hydrogen at the same rate, while for the Mg-3.0Ca alloy, the deep -rolled sample showed a significant reduction of hydrogen generation, thus delaying degradation of the alloy. This further indicates that deep rolling after the turning process reduces the corrosion rate for the Mg–3.0Ca alloy. This can be attributed to higher and deeper compressive stress into the subsurface, which closes the superficial micro-pores and stops the formation and propagation of cracks (see figure 12(b)).

Low plasticity burnishing, analogous to deep rolling, is a new mechanical treatment process used to improve the corrosion and the fatigue resistance of metallic alloys. It produces cold work hardening and introduces minimal plastic deformation generating compressive residual stress into the subsurface. Low cold working offers both thermal and mechanical stability of the beneficial compression, and the process has recently been used in aerospace and biomedical industries [109, 118, 119]. Similar to low plasticity burnishing, as already indicated, Denkena and Lucas [117] employed deep rolling to develop adaptable degradation profiles for the purpose of different medical applications of Mg–Ca alloys. Both processes can introduce higher and deeper compressive stress as high level of cold working associated with the process often degrades the mechanical stability of compressive layer generated. Interestingly low plasticity burnishing typically requires less cold working and the technique with appropriate processing parameters would, therefore, be a promising way to improve the biomechanical stability. Compared to shot peening, the benefit of using low plasticity burnishing or deep rolling is that the processed surface does not become contaminated with any external material and hence third body wear is unlikely to occur [120]. Salahshoor and Guo [15, 17] studied the process mechanics of ball burnishing and low plasticity burnishing and investigated the effect of different process parameters on improving the corrosion resistance and the mechanical surface integrity of Mg–Ca alloys.

Very recently, unlike low plasticity burnishing, severe plasticity burnishing integrated with cryogenic cooling has been employed to enhance the corrosion resistance of Mg–Al–Zn alloys [116] and compared the performance with traditional grinding process. Cryogenic burnishing produced a fine-grained microstructure at a relatively larger depth, inducing severe compressive stress as well as generating a fine surface finish (figure 12(c)). Hydrogen evolution tests in NaCl solutions suggested that cryogenic burnishing reduced the corrosion rate by almost 50% as compared to the ground-only sample (see figure 12(c)). The above results clearly imply that that burnishing or deep rolling as a simple, cost-effective mechanical process would be able to offer a unique opportunity to remarkably enhance the functional performance of metallic materials, such as the corrosion resistance of Mg-based alloys, by tailoring their microstructures and introducing residual stress into the subsurface.

Overall, it seems that burnishing or deep rolling would be an efficient way to mitigate corrosion-induced failure of the implants without changing either the material or design of the implants. While shot peening may negatively influence the corrosion resistance, burnishing could be potentially employed to control both bio-corrosion and mechanical integrity, and hence further research works are needed to achieve the maximum benefits. It has been reported that it is possible to generate the depth of compression into the surface, which may exceed the maximum corrosion pit depth in implants, and thus save the implants from cracking or pitting-induced failures.