•We performed fatigue and tensile tests of hydroxyapatite(HAp)-coated AZ31 Mg alloy.•The coated AZ31 showed high adhesiveness under 5% static and 3% cyclic elongations.•Fatigue limit of the AZ31 slightly decreased with the HAp coating.•Small pits with a depth of ca. 10 μm were formed on AZ31 during the HAp coating.•The small pits caused the decrease of fatigue strength.A hydroxyapatite (HAp) coating was directly formed on an extruded AZ31 magnesium alloy by a single-step chemical solution deposition. The HAp coating consists of an outer porous HAp layer, an inner continuous HAp layer, and a thin intermediate MgO layer, and the inner HAp and MgO layers are composed of nanocrystals. Tensile and fatigue tests were performed on the HAp-coated AZ31 in air. The HAp coating microscopically showed neither crack nor detachment at 5% static elongation (1.5% residual strain). With further elongation under tensile stress, cracks were formed perpendicularly to the tensile direction, and fragments of the coating detached with a fracture inside the inner continuous HAp layer. The fatigue strengths at 107 cycles (fatigue limit) of HAp-coated and mechanically polished AZ31 were ca. 80 MPa and ca. 90 MPa, respectively. The slight decrease in the fatigue limit with the HAp coating is attributed to small pits with a depth of ca. 10 μm formed on the substrate during the HAp-coating treatment. The HAp coating remained on the specimen without cracks after 107 cycles at the fatigue limit, which provides ca. 3% cyclic elongation.

Anticorrosion coatings are crucial for practical applications of magnesium alloys, which are used to reduce the weight of vehicles, aircraft, electronics enclosures etc. Hydroxyapatite (HAp) potentially offers high corrosion resistance and no environmental toxicity because its thermodynamic structural stability is high and it is a basic component of bone. However, direct synthesis of HAp on magnesium in aqueous solutions has been a scientific challenge because Mg ions prevent HAp crystallization. A new method of direct synthesis of HAp on magnesium was developed using a Ca chelate compound, which can maintain a sufficiently high concentration of Ca ions on the magnesium surface to overcome prevention of HAp crystallization with Mg ions. Highly crystallized HAp coatings were successfully formed on pure magnesium and AZ series alloys. Corrosion behavior of HAp-coated pure magnesium was examined by cyclic dry and wet tests with 1 g m−2 NaCl on the surface and polarization tests in a 3.5 wt% NaCl solution. A HAp-coated pure magnesium showed no noticeable corrosion pits after the dry and wet test. HAp-coated specimens showed 103–104 times lower anodic current density than as-polished specimen in the polarization test. The results demonstrate the remarkable anticorrosion performance of HAp coatings on magnesium for the first time.

Calcium phosphate precipitated on pure magnesium from artificial plasma (modified Hanks’ solution) was varied by anodization and autoclaving, aiming the control of corrosion rate of bioabsorbable magnesium. Rough and smooth anodized film was formed depending on anodizing voltage in 1 N NaOH. The amount of calcium phosphate precipitated on the porous film was 2–3 times larger than that on the smooth film. The Ca/P ratio on the porous film was slightly higher than that on the smooth film. The autoclaving did not significantly influence the morphology of anodized film; however, the precipitation of calcium phosphate was restricted. No significant local corrosion occurred after the immersion in modified Hanks’ solution. It is demonstrated that the precipitation of calcium phosphate on magnesium can be controlled by anodization and autoclaving.

Corrosion behaviour and microstructure of developed low-Ni Co–29Cr–(6, 8)Mo (mass%) alloys and a conventional Co–29Cr–6Mo–1Ni alloy (ASTM F75-92) were investigated in saline solution (saline), Hanks’ solution (Hanks), and cell culture medium (E-MEM+FBS). The forging ratios of the Co–29Cr–6Mo alloy were 50% and 88% and that of the Co–29Cr–8Mo alloy was 88%. Ni content in the air-formed surface oxide film of the low-Ni alloys was under the detection limit of XPS. The passive current densities of the low-Ni alloys were of the same order of magnitude as that of the ASTM alloy in all the solutions. The passive current densities of all the alloys did not significantly change with the inorganic ions and the biomolecules. The anodic current densities in the secondary passive region of the low-Ni alloys were lower than that of the ASTM alloy in the E-MEM+FBS. Consequently, the low-Ni alloys are expected to show as high corrosion resistance as the ASTM alloy. On the other hand, the passive current density of the Co–29Cr–6Mo alloy with a forging ratio of 50% was slightly lower than that with a forging ratio of 88% in the saline. The refining of grains by further forging causes the increase in the passive current density of the low-Ni alloy.

Changes in the composition of surface oxide film on titanium specimens in the presence of amino acids, serum proteins, and cells were characterized using X-ray photoelectron spectroscopy. The surface oxide film on titanium formed in the air is so protective that the further oxidation of titanium is prevented in various circumstances. During immersion of the specimen in Hanks’ solution, Eagle’s minimum essential medium (MEM), and MEM with the addition of fetal bovine serum (MEM+FBS), calcium phosphate precipitated, causing the increase in thickness of the surface oxide film. Calcium phosphate was also precipitated with culturing murine fibroblast L929, but the amount of the calcium phosphate was smaller than those in Hanks’ solution, MEM, and MEM+FBS. The relative concentration ratio of calcium to phosphorous, [Ca]/[P], increased with proteins charging negatively, while the ratio decreased with the cells whose extracellular matrix charging positively. In addition, sulfur precipitated as S0 and/or S2− only with culturing the cells. Sulfate ions in the MEM+FBS are reduced at the interface between titanium and the solution with the existence of cells.

Various electrochemical measurements were performed on titanium with and without culturing murine fibroblasts L929 to characterize the effects of cells on interface electrochemical properties between titanium and cells. Open-circuit potential of titanium decreased with L929 cells, which was caused by the shift of equilibrium potential between cathodic and anodic reactions indicated by the decrease in cathodic current density with L929 cells. In cathodic potential step test, the decrease in current density following to the peak current density was delayed with L929 cells, indicating that diffusivity of molecules and ions decreased with the cells. In addition, alternating current impedance measurement and data approximation to the electrical equivalent circuit model revealed that the circuit element for diffusion resistance of biomolecule adsorption layer increased with L929 cells. Consequently, the effect of cells on the interface property is the retardation of diffusion through the biomolecule adsorption layer due to the increase in biomolecule density with extracellular matrix consisting with proteins and glycosaminoglycans generated by the cells.