Discussion on the Peak Shift of Α-Ti Phase in Tio 2 Nanostructured Coatings on Ti-6 Al-4 V Alloy

In this study, TiO2 nanostructured coatings on Ti-6A-4V alloys were fabricated by two methods: H2O2 oxidation and RF sputtering. In the annealing temperature range of 25 C 500 C, there were the peaks at 35, 37, 40 and 52 corresponding to {100}, {002}, {101} and {102} crystal planes of hcp structure of α-Ti. At the annealing temperature of 600 C, there was the presence of peaks corresponding to crystal planes of anatase and rutile TiO2. The relative intensities of anatase and rutile phases of the sample fabricated by RF sputtering were 3.62 and 10.25 %, respectively; while those of the sample fabricated by H2O2 oxidation were 21.27 and 3.20 %, respectively (The relative intensity of α-Ti phase was 100 %). The results investigated the peak shift of α-Ti phase in TiO2/Ti-6Al-4V nanostructured coatings fabricated by the two methods which was reasonably explained from the difference in the thermal expansion coefficients of Ti alloy and TiO2 components, as well as the difference in the ratio of anatase to rutile phases.

Dumbleton et al. [11] have proven that HAp coating should be in crystalline form which has an ability to provide a better substrate for the development of cells, to prevent the formation of adverse fibrous tissue.To increase the adhesion strength of HAp coating on the substrate, it should exhibit sub-layers between HAp coating and Ti-6Al-4V surface [12].The addition of the sub-layers is expected to reduce a thermal expansion mismatch of the layers, as well as to achieve an abundance of surface hydroxyl and superoxide radicals groups, consequently, to achieve a surface free of cracks and a high adhesion of the modified surface to the substrate [13,14].Kim H. W. et al. [15] investigated that the insertion of a TiO2 buffer layer was to improve the bonding strength between the HAp layer and Ti substrate, as well as to prevent the corrosion of the Ti substrate.The results reported in the publication [15] showed that the bonding strength of the HAp/TiO2 double layer coating on Ti markedly increased, with the highest strength of the double layer coating at 55 MPa after annealing at 500 C.Annealing the coating at the high temperature to keep the surface of hydroxyapatite stable, which is important for the interaction between hydroxyapatite substrate and bone tissue cells.Moreover, annealing the coating is to obtain nanostructured coatings with pre-determined structure, chemical and phase composition [16].Other publications have also discussed the proper temperature to improve the strength and quality of the titanium-hydroxyapatite interface in the annealing temperature range of 500 -750 C [17][18][19][20].However, at the high annealing temperature, the values of thermal expansion coefficients of Ti alloy and TiO2, as well as those of HAp and TiO2 should be concerned carefully.In this research, at the beginning, we focused on the difference in thermal expansion coefficients of Ti alloy and TiO2 components.According to the literature [21], the SI units of the coefficient of thermal expansion is K −1 and typical values for alloys are in the range of 10×10 −6 to 30×10 −6 K −1 , and this value of Ti alloy is 8.70×10 -6 /K.The computed linear thermal expansion coefficient (TEC) of TiO2, which is similar with the experimental data, is 6.55×10 −6 K −1 [22].Some original results determined the effect of the difference in the TEC values, as well as the preparation methods on the structure of TiO2 sub-layer on Ti-6A-4V alloy.
The main aim of study is to investigate the peak shift of α-Ti phase in TiO2/Ti-6Al-4V nanostructured coatings fabricated by a controlled oxidation in hydrogen peroxide via RF magnetron sputtering technique.

Material and methods
The specimen used for this study was Ti-6Al-4V alloy (Gr 5, ASSTM 136, BAOJI TI-LEADER METAL PROCESSING CO., LTD) with a circular shape of 10.0 mm diameter and 1.0 mm thickness.The specimens were polished using the abrasive silicon carbide (SiC) paper up to 1200 grade.Final polishing was done using 0.02 -0.25 µm corundum grits (Struers AP-Paste SQ) and micro-cloth (40-72222, Buehler, Vibromet 2) to obtain a surface without scratch, followed by rinsing with distilled water and acetone in ultrasonic.
Preparation of TiO2 coating on Ti-6A-4V alloy via RF magnetron sputtering method: Using a TiO2 target (High Purity Chemicals Lab.Corp., Grade: 99.99%) as the source material and Ar gas (99.99%) as the sputtering gas.An RF generator was used at a frequency of 13.56 MHz and a power of 100 W. Ar gas was then introduced into the vacuum chamber by a mass flow controller at 50 sccm and kept at 5.0 × 10 −3 mbar as the total pressure.Before the deposition of the samples, the chamber was kept in the pre-sputtering regime for 10 min (shutter closed) to remove contaminations on the target surface in order to stabilize the deposition parameters.Ti-6Al-4V surfaces were fixed onto the substrate holder which was centrally positioned in parallel just above the source material with a target-to-substrate distance of 60 mm.The deposition time was 2 hours.
The TiO2 nanostructured coatings on Ti-6A-4V alloy were annealed in a vacuum furnace at 25 Torr and the rate of temperature increase of 10 C /min, kept at the highest temperature (400, 500 and 600 C) for 2 hours and cooled freely down to the room temperature.
The structure and phase composition of the surfaces were investigated via X-ray powder diffraction (XRD, Rigaku Ultima JV, Japan) with Cu Kα radiation (λ = 1.540Å), the acquisition time of 4° per minute, and the step size of 0.01.Remarkably, the peak shift phenomenon of all four diffraction peaks of α-Ti phase was observed in XRD pattern of TiO2/Ti-6A-4V alloy annealed at 600 C, while these peaks did not shift in all the cases of TiO2/Ti-6A-4V alloy annealed in the range of 25 C to 500 C, as well as those of Ti-6A-4V alloy annealed at 600 C.The central positions of these peaks shifted to lower 2-theta side, consequently, the corresponding d spacing values of the reflections increased (Table 1).Changes in lattice spacing of titanium crystalline were presented in some publication as following: R. Montanaria et al. [28] investigated the effects of nitrogen and oxygen absorption

Results and discussion
on lattice expansion of Ti-6Al-4V by high-temperature X-ray diffraction.After annealing this material at 600 °C, the oxygen or nitrogen atoms may occupy in the octahedral interstices of (hcp) structure of α-Ti, which results in the expansion of cell volume and the changes in the ratio of cell parameters.According to the literatures [29,30] mechanical and physical properties of titanium alloys strongly depend on interstitial elements at high temperature.The nature of the reinforcement/matrix interface plays a significant role in the stress transfer of the composites, thus influences the mechanical properties of composites.In this study, oxygen and nitrogen elements may not exist in the coatings due to annealing in a vacuum furnace.Thus, the TiO2 film, which was crystallized in anatase and rutile phases, formed onto the surface of Ti-6A-4V alloy and affected on the structure of the original alloy as the temperature increased up to 600 C.In fact, the thermal expansion coefficient (TEC) of Ti alloy was 8.70×10 -6 /K and that of TiO2 was 6.55×10 −6 K −1 .The TEC, which is one of the structural parameters, was given as a reason in the increase of d spacing values of the reflections of TiO2/Ti-6A-4V alloy annealed at the phase transition temperature of TiO2.The TEC of Ti alloy is larger than that of TiO2, so the rate of contraction of TiO2 layer should be smaller than that of Ti substrate during cooling, which resulted in the lattice expansion.
Particularly, Table 1 also presents that the degree of lattice expansion of the sample fabricated by the controlled oxidation in hydrogen peroxide was less than that of the sample synthesized by the RF magnetron sputtering method.This result can be explained by the difference in the thermal expansion coefficients of anatase and rutile phases, as well as the difference in the ratio of the two phases in the samples fabricated by the different processes.To determine the relative ratios of anatase phase to rutile phase coexisting in TiO2/Ti6Al4V samples, the strongest reflections for anatase and rutile phases were conveniently located at 25.2 and 27.4, after baseline correction (Fig. 2).
(a) H 2 O 2 oxidation (b) RF sputtering Fig. 2: The presence of the peaks in the 2-theta range of 20-30 of the samples annealed at 500 C and 600 C, where anatase phase is symbolized as ▲, and rutile phase is symbolized as Δ Fig. 2 investigated the difference in the temperature of formation of anatase, rutile phases in the two fabrications, as well as in the ratio of anatase phase to rutile phase.Firstly, the anatase phase appeared in the sample fabricated by the H2O2 oxidation after annealing at 500 C, while phase did not appear in the sample fabricated by RF sputtering at the same annealing temperature.Moreover, the sample fabricated by RF sputtering had an amount of rutile phase increasing fast after annealing from 500 to 600 C.According to James's explanation (J. Ovenstone et al. [27]), the presence of a small amount of brookite phase contamination in the anatase resulted in rapid conversion to the rutile.In case of the sample fabricated by the controlled oxidation in hydrogen peroxide, the amount of anatase phase was higher than that of rutile phase, while the sample fabricated by the RF magnetron sputtering method had an opposite ratio.This result means TiO2 coatings sputtered via RF sputtering may contain a small amount of brookite phase, however, that was not confirmed by XRD examination.Secondly, the difference in the ratio of anatase phase to rutile phase was identified due to the relative intensity (%) of the peaks (Table 2).▲ ▲ Δ Δ Table 2 also showed the crystallite sizes of the anatase and rutiles phases coexisting in TiO2/Ti6Al4V samples.In case of the sample fabricated by the oxidation, the widths of peaks at 25.2 and 27.4 for anatase and rutile phases, respectively, were practically the same, so anatase and rutile crystallites had the same size.While the sample fabricated by RF sputtering had the sizes of anatase and rutile crystallites were slightly higher.The difference in crystallite sizes calculated from XRD data was appropriate to that in particle sizes observed from SEM images (Fig. 3).Thus, the degree of lattice expansion of α-Ti phase in the sample fabricated by the H2O2 oxidation was less than that in the sample fabricated by the RF sputtering because of two reasons: First, the TEC of TiO2 anatase was smaller than that of TiO2 rutile in the same condition [31].Second, the amount of anatase phase was higher than that of rutile phase in the case of the sample fabricated via H2O2 oxidation process.

Fig. 3 :
Fig. 3: SEM images of the samples annealed at 600 C

Table 2 :
Data of the peaks at 25.2 and 27.4 of the samples annealed at 500 C and 600 C