2021 Volume 62 Issue 12 Pages 1806-1809
This study aimed to investigate the effects of alloying elements on the fatigue properties of magnesium in simulated body fluids. The fatigue life of the Mg–Zn alloy in air was longer than that of Mg–Ca. This result is in accordance with the tensile yield stress. Crack propagation occurred along the grain boundaries in Mg–Ca and among the grains in Mg–Zn because zinc addition significantly effects grain boundary strengthening. In contrast, the fatigue life in simulated body fluid was longer for the Mg–Ca alloy than for the Mg–Zn alloy at a lower stress amplitude. These results suggest that the use of both calcium and zinc as additives contributes to the further improvement of the fatigue life of magnesium in simulated body fluids as the immersion time is prolonged.
Magnesium has attracted attention in recent years as a lightweight and biodegradable material.1) In particular, the favorable influence of magnesium on bone formation when implanted into bone tissue has led to its investigation as a potential material for use in bone fixation devices.2) However, major problems remain in terms of the mechanical and biodegradability of such devices. A variety of magnesium alloys developed for industrial applications have been found to exhibit excellent mechanical properties and biodegradable behavior.3) In these cases, although rare-earth elements have been applied to drastically improve the strength, ductility, and biodegradable behavior, the cytotoxicity of these elements towards cells has been reported.4) Moreover, they have also been found to accumulate in organs following the long-term implantation of magnesium alloys containing these elements.5)
To overcome these issues, the alloying of magnesium with essential elements such as calcium or zinc has been investigated owing to the expected high biocompatibility of the resulting alloys. The influence of alloying on biodegradation behavior has also been thoroughly investigated. The degradation rate of magnesium is decreased with the addition of appropriate amounts of Zn or Ca.6) Moreover, excellent in vivo biocompatibility has been demonstrated.7) As a result of these promising properties, these alloys have potential for application in plate systems for bone fixation devices. However, the bone tissues supported by magnesium plate systems must be able to support the repeated loads encountered during daily life. The plates and nails in the bone tissue are simultaneously subjected to degradation reactions with body fluids and repeated loads. Consequently, magnesium plate systems must exhibit excellent in vivo fatigue properties. Therefore it is important to determine the influence of the alloying elements on the fatigue properties in simulated body fluids (SBF). Corrosion fatigue properties have been evaluated in both humid environments and aqueous solutions, and the influences of alloying elements or the microstructure have been investigated.8) However, it was previously reported that the SBF used in the corrosion fatigue test strongly affects the fatigue life.9) In other words, the fatigue life in SBF of magnesium alloys may differ from the corrosion fatigue results obtained under other conditions. Although a limited number of studies have attempted to evaluate the fatigue properties in SBF of magnesium alloys containing calcium and zinc, these properties may be affected by the precipitates present in such alloy systems.10) Therefore, clarification of the effects of alloying elements on the fatigue properties in SBFs is still required. In this study, we examine the influences of the additives, calcium and zinc on the fatigue properties of Mg–Zn alloys and Mg–Ca11) solid solution alloys in an SBF.
Cast ingots of Mg–0.3 at% Zn and Mg–0.3 at% Ca were used in this study. Both alloys were subjected to two extrusion steps. The extrusion ratio during the second extrusion was 25:1. The extrusion temperatures of the Mg–Zn and Mg–Ca alloys were 503 and 623 K, respectively. To clarify the influence of the alloying element, the amount of additive was fixed at 0.3 at%, which is smaller than the solution limit at the extrusion temperature. The alloys are referred to herein as Mg–Zn and Mg–Ca. The microstructures of the extruded alloys were observed via scanning electron microscopy (SEM; JSM6510, JEOL, Japan)-electron backscatter diffraction (EBSD; OIM™, EDAX, USA).
Plate specimens with a gage length of 7.5 mm, width of 1.5 mm, and thickness of 1.0 mm were machined from the extruded bars. The plate specimens described above were used in the tensile and fatigue tests. The tensile test was conducted at an initial strain rate of 1.0 × 10−3 s−1. Fatigue tests were performed in either air or SBF. The tests were conducted by applying axial sinusoidal loading using a computer-controlled electromagnetic testing machine (MMT-500NV-10, Shimadzu, Japan). The presence of serum or proteins has been demonstrated to influence the fatigue life of magnesium in SBF.12) Therefore, in this study, Eagle’s minimum essential medium supplemented with 10% fetal bovine serum was used as the SBF. The custom-built chamber installed in the fatigue testing machine was filled with the SBF. The stress ratio was fixed at R = 0.1 for all fatigue tests, and the test frequency was 10 Hz. The fatigue tests were considered to have reached completion either when the specimen fractured completely or after 1 × 106 test cycles as the number of walks per year was 2 × 106 and the duration of fracture treatment was 6 months. In this study, the fatigue limit was defined as the stress amplitude at which the specimen did not fracture during the fatigue tests. After the fatigue tests, the fracture surfaces were observed via SEM.
Figure 1 shows the inverse pole figure (IPF) maps and pole figures for both alloys. The observed plane was parallel to the extrusion direction. Both alloys exhibited fully recrystallized equiaxed grains. The average grain diameters for Mg–Zn and Mg–Ca were 5.7 and 3.2 µm, respectively. The pole figures of the basal plane shown in Fig. 1 demonstrate the characteristic distributions of the extruded square bars in the three alloys. The intensity of the basal texture was found to be higher for Mg–Zn than for Mg–Ca.
Initial microstructures of the extruded alloys and the corresponding pole figures. (a)–(b) IPF maps for the (a) Mg–Ca and (b) Mg–Zn alloys. (c)–(d) Pole figures for the (c) Mg–Ca and (d) Mg–Zn alloy.
The tensile deformation behaviors of both alloys are shown in Fig. 2. The tensile yield stresses of Mg–Zn and Mg–Ca were 228 and 217 MPa, respectively. The Mg–Zn alloy exhibited a higher tensile yield stress, although the grain size of Mg–Zn was slightly larger than that of Mg–Ca. This may be caused by the greater basal texture in Mg–Zn.
Nominal stress–nominal strain curves obtained under quasi-static tension for the extruded Mg–Zn and alloys.
Figure 3 shows the S-N curves for Mg–Zn and Mg–Ca11) in air and SBF. A comparison of the fatigue properties in air revealed that the fatigue life of Mg–Zn was higher at all stress amplitudes. The fatigue limits in the air of Mg–Zn were slightly higher than those of Mg–Ca.
S-N curves for the Mg–Zn and Mg–Ca alloys determined in air and the SBF at room temperature. The fatigue tests were conducted until failure or 1 × 106 cycles were completed.
The fatigue life in the SBF of Mg–Zn was lower than that in air, as well as in the Mg–Ca alloy. In addition, the relationship between the stress amplitude and fatigue life was dependent on the alloying element in the SBF. The Mg–Zn alloy exhibited a shorter fatigue life than the Mg–Ca alloy at stress amplitudes lower than 30 MPa. The ratio of the fatigue life in SBF to that in air was smaller for Mg–Zn than for the Mg–Ca alloy at all stress amplitudes. This indicates that alloying with calcium is more effective in improving the fatigue properties of these alloys in SBF.
Figure 4 shows representative SEM images of the fracture surfaces of the Mg–Zn and Mg–Ca alloys. Figures 4(a)–(b) and 4(c)–(d) show the surfaces of the alloys fractured in air and SBF, respectively. Because fractures close to the surface were covered by degradation products after the fatigue tests in SBF, the crack propagation area of the fracture surfaces are shown in these images. The fracture surfaces of the alloys were independent of the stress amplitude. Steps of a size similar to the grain size were observed in the fatigued Mg–Ca sample. In contrast, no such steps were found in the Mg–Zn alloy, although cleavage steps were observed. The alloys tested in SBF showed the same fracture surfaces as those tested in air, demonstrating that the SBF did not substantially affect the fracture surface. The results indicate that the Mg–Ca alloy underwent brittle fracture at the grain boundary and the Mg–Zn underwent relatively ductile fracture, which was independent of the SBF and stress amplitude.
SEM images of the fracture surfaces after the fatigue tests: (a) Mg–Ca in air, (b) Mg–Zn in air, (c) Mg–Ca in the SBF, and (d) Mg–Zn in the SBF.
As demonstrated above, the crack propagation pathway varied depending on the alloying element. Examination of the surfaces of the materials containing cracks is useful for elucidating the crack propagation pathway. The extruded bars after heat treatment were utilized to precisely observe the crack propagation pathway. The increased average grain size was approximately 30 µm for both alloys. Figure 5 shows IPF+IQ maps near the crack propagation pathway. For the Mg–Ca alloy, the crack propagated along the grain boundaries. Crack propagation along grain boundary is confirmed in pure magnesium (data not shown). In contrast, in the Mg–Zn alloy, intergranular fracture was observed. Therefore, the additive elements affected the crack propagation pathways during fatigue deformation. It is generally known that cracks propagate faster at grain boundaries. Therefore, it can be said that the onset of intergranular fracture due to zinc addition resulted in the superior fatigue life of Mg–Zn compared to that of Mg–Ca at all stress amplitudes.
IPF+IQ maps near the crack propagation pathways for the (a) Mg–Ca and (b) Mg–Zn alloys.
It has been demonstrated that enhanced grain boundary strength suppresses intergranular fractures.13) The grain boundary strength can be estimated using the grain boundary cohesive energy determined via first-principles calculations, as demonstrated by Yamaguchi.13) The value of this parameter was lowest for pure Mg, followed sequentially by Mg–Ca and Mg–Zn.14) This result indicates that both of the essential elements contributed to grain boundary strengthening. However, only a small difference in the grain boundary cohesive energy was observed between the Mg–Ca alloy and pure Mg. Therefore, in the case of the Mg–Ca alloy, the strengthening effect of calcium is insufficient to prevent crack propagation along the grain boundaries, thereby resulting in lower elongation and fatigue lives in the air of Mg–Ca compared to those of Mg–Zn.
For both alloys investigated in this study, the fatigue life was found to decrease when the fatigue tests were performed in SBF rather than in air, as shown in Fig. 3. The crack propagation pathways were found to be unaffected by changes in the surrounding environment because the fracture surfaces of the alloys were unchanged. These observations indicate that the decrease in the fatigue life may be caused by degradation at the specimen surface. In SBF, a layer of degradation products initially covered the specimen surface. This layer functions as a protective layer against further degradation and moderates the degradation rate. However, a portion of the protective layer was somewhat dissolved at the ion equilibrium. In addition, because the specimens were also exposed to cyclic loads, the protective layer may have been disrupted during the fatigue tests in the SBF. These holes in the protective layer enable contact between the magnesium alloy and the SBF, thereby promoting local degradation of the alloy during static immersion.15) These local exposures to SBF cause local degradation, which causes crack initiation. The surfaces of both alloys after the fatigue tests in the SBF are shown in Fig. 6. The specimens were subjected to fatigue tests in the SBF for similar durations. Although several cracks are clearly visible in the Mg–Ca alloy specimen, these cracks have a short length, and connections between the cracks cannot be seen. In contrast, many long and wide cracks can be observed in the Mg–Zn alloy specimen. Therefore, the addition of calcium contributed to the improved toughness of the protective layer compared to the addition of zinc. As the properties of the protective layer are influenced by the composition of the mother alloy, it can be stated that the fatigue life may be affected by the additive elements. Thus, this change in the surface of the alloy may contribute to the longer fatigue life of the Mg–Ca alloy in SBF. These results suggest that increased calcium concentrations may contribute to this change at the surface. To confirm this effect, further investigations into the toughness or strength of the degradation product layer are required.
Surfaces of the specimens after the fatigue tests for the (a) Mg–Ca and (b) Mg–Zn alloys.
The fatigue life of human bone is on the order of 5 × 105 at a stress amplitude of 23 to 30 MPa.16) In this study, both Mg–Ca and Mg–Zn fractured before the number of cycles reached 5 × 105 at a stress amplitude of approximately 30 MPa, indicating that the singular addition of Ca or Zn did not provide sufficient fatigue properties for bone fixation devices. However, we found that Zn contributes to the improvement of the fatigue properties by suppressing the crack growth rate, while Ca contributes to the suppression of crack initiation in the SBF. Therefore, ternary alloys containing both elements are expected to exhibit an excellent fatigue life with a combination of both effects. Moreover, the co-segregation of zinc and calcium further improved the grain boundary strength compared to the Mg–Ca alloy.17) In addition, the Mg–Zn–Ca alloy may exhibit a slower degradation rate than the Mg–Ca alloy because of the inhibition of precipitation by the addition of zinc.18) In addition to alloying, microstructure control, such as grain refinement, which can contribute to the improvement of strength and corrosion resistance, is considered to be effective. In future studies, it will be necessary to clarify the effects of ternary alloying and microstructural factors on the fatigue properties of these materials, as well as to develop magnesium alloys that exhibit a fatigue life superior to that of bone.
The effects of the alloying elements calcium and zinc on the fatigue properties of magnesium in air and an SBF were investigated, and the following conclusions were obtained:
This work was supported by JSPS KAKENHI Grant Numbers JP18K14015 and 20K05132. The author (N.I.) also appreciates the financial support from the Light Metals Education Foundation, Japan.
