Recently biodegradable metallic implants have notable consideration in biomaterial field. Their degradation after a certain time, leads to reduce the call for extra surgery to remove the short-term implant. Mg alloy has a great interest as a biodegradable implant because its biocompatibility as it found naturally in bone; it can directly influence bone-resorbing cells and consider as a vital factor for bone metabolism. Many clinical in vivo and in vitro studies have been recommended Mg-base implants due to their good biocompatibility and their mechanical properties comparable with those of natural bone   . The Mg density is (1.7 - 2.0 g/cm3) which is a slightly lower than that of natural bone (1.8 - 2.1 g/cm3), and its elastic modulus (41 - 45 GPa) is closer to those of the human bone (10 - 40 GPa)  . All these properties reduce the probability of stress shielding effect observed in the case of higher modulus materials such as titanium alloys  . Unfortunately, the application of magnesium as an implant has some restrictions such as its high degradation rate, especially in an aggressive chloride medium like a human body. This implant may degrade and reduce its mechanical properties before new tissues are completely healed  . Hence element alloying and surface modification are effective methods to improve rapid degradation of Mg and also to enhance their mechanical properties besides aiding in decrease corrosion.
Many elements such as Ca, Mn, Al, Zn, Si, and Sr are used as alloying elements with magnesium. In the present research, zinc (Zn) and antimony (Sb) were chosen as alloying elements. Zn is a commonly used as alloying element with magnesium and its strengthening effect is only following aluminum. It furthermore assists to eliminate the harmful effect of impurities such as iron and nickel, thus improving the corrosion resistance of magnesium alloys  . Also, 4wt% Zn which was chosen in this work was reported to be the optimum Zn content for tensile and mechanical strengthening over the range of alloying  . Moreover, zinc is vital for many biological functions, plays a fundamental role in the metabolic activities of the body’s enzymes, essential for cell division and the creation of protein and DNA. Whereas excessive exposure toxicity of Zn is rare, Zn shortage has a harmful effect on growth, neuronal development, and immunity  . On the other hand, it is reported that the addition of 0.4 wt% Sb gives high mechanical properties, however, at the higher Sb level of 0.7 wt% the strength and hardness decrease  . Therefore, in present work, 0.5 wt% Sb is added to Mg-4wt% Zn alloy to study its effect on mechanical properties and degradation behavior. On the other hand the biological effect of Sb still needs more experimental research.
To improve corrosion resistance and biocompatibility of prepared alloys, the surface will be coated with Ca-P layer. Ca and P are the major elements in bone minerals. Thus, Ca-P coatings, especially osteoconductive minerals such as hydroxyapatite (HA) and tricalcium phosphate (TCP), should be useful in constructing new bone and promoting osteointegration around Mg implants  . Niu, Jialin, et al.  reported that the Ca-P coat improves the corrosion resistance of Mg-Nd-Zn-Zr alloy and reduced the hemolysis rate from 48% to 0.68%. Song, Yang, et al.  showed that both the HA and FHA (fluoridated hydroxyapatite) coating could promote the nucleation of osteoconductive minerals (bone-like apatite or β-TCP), while the precipitates on the uncoated Mg-Zn alloy in modified simulated biological fluid had low Ca/P molar ratios, which delayed bone-like apatite formation.
Several coating techniques have been investigated, but due to the special properties of Mg as high reactivity and relatively low melting point (650˚C), not all coating procedures can be suitable for Mg  . Seeing that the biomimetic coating would be helpful in coating biodegradable substrates such as Mg-alloys, it reduces the incubation times in SBF during the coating procedure  . As reported this resulted in the formation of non-uniform and porous Ca-P coating on Mg substrates  , in addition to hydrogen bubbles formed on the Mg surface during immersion. When using a Sol gel technique for Mg substrates, the temperatures should be less than the melting point of pure Mg (650˚C)  to avoid affecting the surface integrity of the Mg substrate. Accordingly, in the present work autocatalytic technique which reported in previous studies    is used. This technique is simple, economical, does not require high temperature and setup in short time which minimize the chance of degradation.
So as there is little information on Mg-Zn-Sb, as biodegradable implant, the present work aims to study the effect of Sb addition on microstructure, mechanical and degradation behavior of Mg-4wt% Zn alloy. Also, we intend to study the effect of the autocatalytic technique on degradation to get an appropriate alloy for load bearing medical applications.
2. Material and Methods
2.1. Material Preparation
The base composition of the studied alloy was Mg-4wt% Zn. 0.5 wt% Sb was added to the base alloy in order to investigate its effect on microstructure, mechanical properties and the degradation rate. The alloy was prepared from commercial pure Mg (99.8 wt% purity), Zn (99.98 wt% purity), and Sb (99.99 wt% purity). Before melting process a charge calculations are performed and with respect to experience, the 90% efficiency of melting process and 10% losses during melting can be estimated and be in consideration. After that melting the charge of about 1.5 kg of each alloy in a graphite crucible placed in an electric resistance furnace. The melting and pouring processes were carried out under gas mix consists of tetraﬂuoroethane (CF3CH2F, HFC-134a, 1 vol.%) and carbon dioxide (CO2, Bal.). Firstly, commercial Mg ingots were melted to above 650˚C, and then Zn with melting point 419˚C was added into Mg melt. After that, the melt were heated to about 720˚C, and then Sb with melting point 630˚C was added into Mg melt. The melt was manually stirred for 3 min using a stainless steel rod, and then was held for additional 15 min in order to get full homogenization. Before pouring the melt, test sample of the melt is required to make a chemical analysis correction using X-ray ﬂuorescence (XRF) analyzer. Depending on the test result a correction process is required either by diluting (adding pure Mg) or concentrating (adding alloying element as Sb or Zn or both) to get the required alloy composition. After that, the slag was removed, and then the melt was poured at about 720˚C into a preheated cast iron mold up to 100˚C with dimensions of outer diameter (100 mm), inner diameter (50 mm) and length (250 mm). The chemical compositions of the prepared alloy were determined by using X-ray fluorescence (model Axios advanced-PANALYTCAL, The Netherlands).
2.2. Microstructure Characterization
All metallographic specimens were cut at the same position of 10 mm from the bottom of castings. Grinding of the specimens were carried out on wet SiC papers, down to grit size 1200 then followed by fine polishing with diamond, particle size from 1 to 6 µm, on nylon cloth, rinsing with water and ethyl alcohol. For best results, the polished surface must be mirror like, i.e., free from any residual layer. After that, the specimens are etched by solution with 10 ml nitric acid, 30 ml acetic acid, 40 ml water, and 120 ml ethanol for 2 - 3 min  .
The microstructures of the specimens were analyzed by optical microscope (OM) (Zeis model Axiotech, Germany). Six OM micrographs were taken for each sample from the observed area at a magniﬁcation of 100×. The average grain size of α-Mg was measured by using an image analyzer. All α-Mg grains existed in one picture taken from the observed area were measured.
Phase constituents of samples were analyzed by X-ray diffraction (XRD) (model X’PERT PRO, The Netherlands) using Cu-Kα (K = 1.54056
2.3. Mechanical Tests
Tensile and compression tests were carried out using LFM-L-20KN Computer controlled and digital display electromechanical universal testing machine (model LFM-L-20KN (bench top), Switzerland). The dimensions of a cylindrical tensile specimen are shown in Figure 1 according to ASTM E8M  . Three specimens were tested with a strain rate of 0.1 mm/min for each alloy. Cylindrical samples with a diameter of 10 mm and a height of 8 mm were used for the compression test.
2.4. Sample Coating
Before coating, samples were cut into discs (10 mm diameter and 2 mm thickness). Each sample was cleaned with acetone, 70% ethanol, and finally distilled water for 15 min each, in an ultrasonic cleaner. Then samples were etched in Kroll’s reagents for 10 min at 20˚C next washed with distilled water to remove the native oxide surface. Afterward, samples were immersed in 10 M NaOH bath at 60˚C for 24 h followed by heating and holding at 100˚C for 1 h    .
2.4.2. Coating by Autocatalytic Bath
After the pretreatment step, samples were immersed in an acidic autocatalytic bath under magnetic stirrer for 2 h at 60˚C ± 2˚C at pH 5 to get different phases Ca-P coatings on samples. The composition of acidic bath was reported in previous work    .
2.4.3. Characterization of Coating
Surface morphology and the chemical compositions of the deposited layer were analyzed by field emission scanning electron microscope (Quanta 250) with EDS. Structure of coated layer was analyzed by X-ray diffraction (XRD) with similar condition used for microstructure composition.
Figure 1. Dimensions of the tensile specimen (mm)  .
2.5. Corrosion Evaluation by Immersion in Simulated Body Fluid (SBF)
Corrosion resistance by in vitro degradation method was obtained by weight loss measurements which were performed by immersing the alloy samples in SBF. The chemical composition of c-SBF in comparison to human blood plasma was reported in several studies  . The pH value of the initial solution was adjusted to 7.4. The whole test cell was maintained at 37˚C ± 1˚C in a shaking water bath to prevent aggregation of the precipitates on sample surfaces  . In this test two categories of samples (non-coated (bare), coated) were studied. The samples were immersed for 14 days in SBF solution with refreshment after 3 days. The refreshment was done to simulate the metabolic environment of the human body and study the formed layer. The appearance and microstructure of the specimens were characterized by SEM with EDS. The pH value of the solution was measured by a pH meter with an accuracy of ±0.01.
The sample weight was measured every 24 hours of immersion after removal of corrosion products with a chromic acid. The degradation rates were calculated by the following equation according to the standard:
where: DR is the degradation rate (mm/hr), is the mass loss (g), A is the initial surface area exposed to the SBF (mm2), T is the immersion time (hr), and is the density of tested sample (g/mm3).
The concentrations of Ca and P ions in SBF after immersion of coated Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb for 14 days were measured using a UV-Vis spectrophotometer and the specific kits (Biodiagnostic, Diagnostic and research reagents, Egypt) at wavelength 640 nm, 585 nm for phosphorus and calcium respectively.
3.1. Chemical Composition
The chemical compositions of the prepared alloy were determined using X-ray ﬂuorescence (XRF) analyzer. Table 1 showed the obtained chemical composition of the investigated alloys.
3.2. Phases and Microstructure Characterization
Table 1. The chemical composition of the prepared alloys (wt%).
Figure 2. (a) XRD patterns of as-cast alloys Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb, optical micrographs of as-cast alloys (b) Mg-4wt% Zn and (c) Mg-4wt% Zn-0.5wt% Sb.
constituents of the Mg-4wt% Zn alloy. These two phases have also been reported to exist in the as-cast microstructure of an Mg-4wt% Zn alloy  . In the Sb containing materials, however, some new peaks corresponding to the Mg3Sb2 intermetallic phase appear in the patterns. Figure 2(b) & Figure 2(c) show that the coarse dendritic structure of the Mg-4wt% Zn alloy is generally reﬁned from about 100 µm (Figure 2(b)) to about 53 µm after the addition of 0.5 wt% Sb (Figure 2(c)).
For clear observations of the constitutive phases, SEM micrographs with EDS and elemental mapping of Mg, Zn and Sb of investigated alloys are shown in Figure 3 & Figure 4, respectively. Figure 3(a) shows the microstructure of the Mg-4wt% Zn alloy which consists of the primary α-Mg matrix (region A), while the second phase appears as white island-like (region B) which was identified by EDS to be composed of Mg and Zn elements. Combined with the XRD results shown in Figure 2(a), it is found that the island-like particles are most likely the MgZn particles and thus the lamellar eutectic structure consists of α-Mg +
Figure 3. SEM micrographs with EDS showing intermetallic phases in as-cast alloys (a) Mg-4wt% Zn and (b) Mg-4wt% Zn-0.5wt% Sb.
Figure 4. (a) SEM micrograph and EDS elemental mapping of (b) Mg and (c) Zn in as-cast alloy Mg-4wt% Zn and (d) SEM micrograph and EDS elemental mapping of (e) Mg, (f) Zn and (g) Sb in as-cast alloy Mg-4wt% Zn-0.5wt% Sb.
Figure 4 shows EDS elemental mapping of Mg, Zn and Sb in the investigated alloys. It can be seen that the non-uniform distribution of Zn (Figure 4(c)) in the interdendritic regions of α-Mg (Figure 4(b)). With addition of 0.5 wt% Sb, the distribution of Sb is mostly at the boundaries of the α-Mg dendrites (Figure 4(g)) and the uniform distribution of Zn at dendrites boundaries (Figure 4(f)).
3.3. Mechanical Properties
The ultimate tensile strength, elongation and Compressive strength of investigated alloys, are summarized in Table 2.
3.4. Characterization of Coating
It is observable that autocatalytic bath changes the morphologies of the based alloys as shown in Figure 5. Homogeneous coating was observed for both alloys. At higher magnification it was found that bulk coating was formed after immersion of Mg-4wt% Zn in the bath with distribution of white grain on its surface (Figure 5(b)). EDS analysis of this layer demonstrates the presence of Mg, O
Table 2. The ultimate tensile strength, elongation and Compressive strength of investigated alloys.
Figure 5. SEM micrographs with EDS showing the coat formed on the surface of Mg-4wt% Zn ((a)-(c)), Mg-4wt% Zn-0.5wt% Sb ((d), (e)).
with traces of phosphorus and calcium on the surface of the alloy (Figure 5(c)). While after addition of Sb, an agglomerate of fine grains covers the surface of the alloys (Figure 5(e)). EDS of this layer consist of higher amount of Ca, P in addition to Na ions. The addition of Sb allows the precipitation of Ca-P layer as a nucleation for hydroxyapatite (HA). The atomic calcium phosphate ratio n(Ca)/n(P) value suggest that formed layer may be dicalcium Phosphate dihydrate (DCPD), also known as brushite which act as a precursor phase.
The XRD patterns Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb alloys after coating (Figure 6) show existence of Mg(OH)2 peaks which confirmed by SEM results. It is noted that Mg-4wt% Zn displays high intensity of Mg(OH)2 peaks compared with Mg-4wt% Zn-0.5wt% Sb.
Other than brushite phase cannot be detected by XRD as it is very thin and non uniform layer. Improvement of the formed layer will be considered in future work by increasing the immersion time and pH of autocatalytic bath.
3.5. Corrosion Evaluation by Immersion in Simulated Medium
The average mass loss and the rate of degradation of Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb alloys before and after coating during 14 days of immersion in SBF are illustrated in Figure 7.
The change in mass is an indicator for the degradation of the alloy samples. Figure 7 illustrates that, the coated sample of Mg-4Zn showed the lowest rate of degradation in the first 24 hr of immersion. As the sample has a porous, and
Figure 6. XRD patterns of Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb samples after coating.
Figure 7. The degradation rate during immersion of coated and bare alloys of Mg-4Zn and Mg-4Zn-0.5Sb.
dense layer of Mg(OH)2, which can reduce the corrosion rate. However, more than 48 hr of immersion, the degradation rate of both coated samples increased. As the protective Mg(OH)2 layer converted into a more soluble MgCl2. This will cause the coated Mg-4Zn alloys to expose to the solution and stimulate the degradation. As well, the coated Mg-4Zn-0.5Sb sample displayed high rate of degradation due to dissolving of brushite coat. On the other hand, the degradation rate of the bare samples of Mg-4Zn and Mg-4Zn-0.5Sb decreased due to formation of Mg(OH)2 protective layer.
By refreshing the solution after 3 days of immersion, it is found that, the degradation rate increased again for all the samples. This caused by Cl− ions which accelerate the electrochemical reaction rate from magnesium and leads to more degradation of the Mg matrix. Seeing that the exposure period continues, the corrosion rate reached a steady state as the rate of formation of Mg(OH)2 is equal to destruction of it by MgCl2. At the end of immersion time it is observed that the Mg-4Zn-0.5Sb samples shows the lowest rate of degradation specially the coated one.
Figure 8 shows the micrograph of the surface of Mg-4Zn and Mg-4Zn-0.5Sb alloys after immersion in the SBF solution for 336 h. Figure 8(a), Figure 8(d), Figure 8(g), and Figure 8(j)) show the surface cracks which formed from dehydration of the surface layer in the air  . Existence of cracks accelerates corrosion of the substrate by making solution come into contact with the matrix  . The bare Mg-4Zn alloy and Mg-4Zn-0.5Sb (Figure 8(a), Figure 8(g)) showed deeper
Figure 8. SEM micrographs of samples surface after SBF immersion. ((a)-(c)) bare Mg-4Zn, ((d)-(f)) coated Mg-4Zn, ((g)-(i)) bare Mg-4Zn-0.5Sb, ((j), (k)) coated Mg-4Zn-0.5Sb.
and thicker cracks than coated alloys (Figure 8(d), Figure 8(j)). The EDS analysis showed the presence of Mg(OH)2 consistent with the atomic ratio of Mg to O (1:2) on the surface of bare Mg-4Zn (Figure 8(c)). The EDS results of the corroded layer on coated samples (Figure 8(f), Figure 8(k)) showed the existence of Mg and O as well as amounts of Ca and P, which represents formation of HA. As observed in EDS results that the bare samples have trace amounts of Ca and P in compare with the coated samples.
This consequence points that the coated samples are more successful to form nucleation of Ca-P which initiates the formation of HA. To confirm that, the concentrations of Ca and P ions in SBF after immersion of coated Mg-4Zn and Mg-4Zn-0.5Sb for 14 days are measured as presented in Figure 9.
It is illustrious that the difference in the electronegativity values between Mg of 1.31, Zn of 1.6 and Sb of 2.05, provides the Sb more tendencies to form a compound with Mg than with Zn   . Therefore, all of the Sb added to Mg-4wt%
Figure 9. Concentrations of Ca and P ions in SBF after immersion of coated alloy samples for 14 days.
Zn alloy, is consumed to form Mg3Sb2 particles  . The reﬁning effect of Sb in cast magnesium alloys has attracted some attention    . It has been reported that small additions of Sb had a great effect on the microstructural reﬁnement and mechanical properties of the cast Mg-5%Sn alloy. Adding Sb to ZA84  and AZ91  has also resulted in microstructural reﬁnement and improved room and high-temperature mechanical properties. Based on the SEM observations (Figure 4(d), Figure 4(g)), where no Mg3Sb2 compound is detected inside the Mg grains and all particles are found to be formed inside the interdendritic spaces. Therefore, the segregation of Mg3Sb2 particles at the liquid/ solid interface would restrict the growth of α-Mg dendrites during solidiﬁcation and thus reﬁne the structure. These results are also confirmed by the previous work reported by R. Alizadeh et al.  . Also the addition of 0.5 wt% Sb could improve the mechanical properties of Mg-4wt% Zn alloy. Owing to the relation of mechanical properties with the microstructure of the investigated alloys, the improved mechanical properties of the Sb-added alloy outcome from their microstructures. It is well known that the ﬁne microstructure is usually beneﬁcial to tensile properties of engineering alloys. According to Grifﬁth’s theory  , the fracture stress of a material increases with decreasing grain size, in inverse proportion to the square root of the grain size. As shown before, the addition of Sb into the base Mg-4wt% Zn alloy cause grain refinement. Therefore, that leads to improve mechanical properties of the as-cast Sb-added alloys which is biocompatible with mechanical properties of cortical bone  in comparison with the base alloy and pure Mg  . It’s clear that the addition of Sb leads to delay the loss of the mechanical properties of the binary Mg-4wt% Zn alloy during implantation
For a closer explanation, in the early stage of an immersion, Mg matrix of bare samples dissolved and transformed into the stable Mg2+ ion (Equations ((2), (3))). This dissolution of Mg is associated with release H2 bubbles observed after immersion and OH− ions which lead to the increase pH of SBF  . After that, a protective layer of Mg(OH)2 is formed (Equation (4)) and prevent solution from direct contact with the substrate. This leads to reduce the degradation rate for a while. However, the aggressive Cl− ions in the medium attack the thin film of Mg(OH)2 and convert it into a more soluble MgCl2 (Equation (5)). So the substrate of the bare samples becomes in contact with the solution again  .
Equations (2)-(5) 
The coated Mg-4Zn sample which has a dense protective layer of Mg(OH)2. This layer protects the sample from degradation at first, but after that the Cl- ions dissolve it into MgCl2 according to (Equation (5)). In case of coated Mg- 4Zn-0.5Sb, the brushite (DCPD) layer is unstable and soluble in SBF especially with increasing pH value. Also brushite convert into HA   and form layer against corrosion (Equation (6)).
As shown in Figure 9 the increase of Ca and P ion concentrations after immersion is due to partially dissolve of the coated layer within the corrosion. Fortunately the presence of the Ca and P ion, and also the increasing of pH during immersion can encourage HA nucleation. So finally it remarks that the formation of the hydroxyapatite in the SBF is self-generated after a certain period of immersion. The formation of HA not only decrease the degradation rate, but it is also reported that during implantation, the living bone will bond through this apatite layer formed on the surface of implant so the coating step is very helpful.
In this article, the as-cast Mg-4wt% Zn and Mg-4wt% Zn-0.5wt% Sb alloys were investigated as degradable metallic implants. The alloys are mainly composed of the matrix α-Mg and MgZn as secondary phase. The addition of Sb refines the grain size and improves mechanical properties due to the formation of the secondary phase of Mg3Sb2.
Immersion result showed that the Ca-P treatment provides a protective layer that improves the corrosion resistance. The degradation rate measured at the end of immersion time improved from 3.5 mm/yr, 2.5 mm/yr for bare and coated Mg-4Zn respectively to 1.4 mm/yr, 0.7 mm/yr for bare and coated Sb-added alloy respectively. So it is clear that the addition of Sb and coating procedure are very effective steps to enhance degradation behavior.