Depth-Resolved Strain Investigation of Plasma Sprayed Hydroxyapatite Coatings Exposed to Simulated Body Fluid

The influence of exposure to simulated body fluid (SBF) on plasma sprayed hydroxyapatite (HAp) coatings on medical grade Ti6Al4V samples has been investigated. Through-thickness residual strain investigations of HAp coatings deposited on flat substrate surfaces incubated for 7, 28 and 56 days were performed using high-energy synchrotron diffraction techniques. In the as-sprayed condition, the results show the top half of the HAp coating to be under compression with the maximum around the near-surface region, relaxing with depth below the surface reaching a strain-free point around the coating thickness midpoint. On the contrary, the remainder of the coating is under tension increasing with further depth; the maximum tension is observed near the coating-substrate interface region. Upon immersion in SBF, both the slope of the normal strain components ε11 and ε33 relax, with the former experiencing a change in slope before saturating after 7 days; the highest change was observed within the first week of incubation. Introduction The second-generation biomaterial hydroxyapatite (HAp) has been extensively studied as a candidate material in biomedical applications due to its similarity to the mineral component of bone. These include filling of bone cavities [1] and medical implant coatings for improved biological fixation [2] amongst others. The poor mechanical properties of the material, however, limit its bulk utilisation in load-bearing applications. To overcome this limitation, the material is applied as a coating on metallic substrates such as Ti, Ti alloys and CoCrMo, combining the excellent mechanical properties of the metal with the osseoconductive ability of the coating [3]. With the plethora of coating techniques available for deposition [4], thermal spraying is still the method of choice. Although successfully utilised at an industrial scale (see, for example [5]), the high plasma temperature together with the cold substrate surface that the droplet impinges on, generally results in thermal decomposition of the HAp powder and rapid cooling on the substrate respectively. This leads to the introduction of undesirable thermal decomposition products [6], such as tricalcium phosphate (TCP), tetracalcium phosphate (TTCP), and sometimes calcium oxide as well as a reduced crystallinity [7]. These products are known to be susceptible to dissolution in simulated body fluids [8] and thus together with the strains and stresses generated MECA SENS 2017 Materials Research Forum LLC Materials Research Proceedings 4 (2018) 123-128 doi: http://dx.doi.org/10.21741/9781945291678-19 124 as a result of differential thermal mismatch (CTE) and quenching of the droplet, may compromise the mechanical stability and integrity of the coating. Although extensive investigation of the effect of incubation of HAp coatings in simulated body fluid have been carried out by many research groups around the world [6], the bulk of their work focused on the near-surface region coating, i.e. the region in immediate contact with living tissue. The present study is an extension of the author‘s previous work [9] on through-thickness investigation of HAp coating in the as-sprayed condition. Materials and methods Sample preparation: Hydroxyapatite powder (CAPTAL 90, batch P215, Plasma Biotal Limited, Tideswell, Derbyshire, UK) with size distribution of 120 ±20 μm was plasma-sprayed onto flat discs of 20 mm diameter medical grade Ti6Al4V alloy substrate supplied by Biomaterials Limited, North Yorkshire, UK. Details of deposition and spray parameters have been reported elsewhere [9]. Subsequent to spraying, the samples were incubated in simulated body fluid for 7, 28 and 56 days to mimic the physiological environment. Sample incubation was carried out in a revised simulated body fluid (rSBF) based on Kokubo’s formulation [10]. The solution had an ionic concentration similar to the human blood plasma but without proteins and enzymes. The temperature and pH of the solution during incubation experiment were kept at 36°C and 7.4, respectively. Subsequent to immersion, slices of approximately 5 mm thick were cut for investigation. Factors considered in determining the optimum slice length are reported elsewhere [9]. Through-thickness characterisation of HAp coating: Angular dispersive diffraction measurements utilising the high–energy synchrotron radiation, 100 keV (wavelength λ = 0.12331 Å), at the Advanced Photon Source’s X-ray Operation and Research 6-ID-D beamline at Argonne National Laboratory , USA was used for the experiments. Experimental details and measurement procedure have been reported elsewhere [9]. Measurements were done in transmission geometry using a 35(V) x 400 (H) μm beam and for one azimuth orientation hence the full strain tensor was not measured. The analysis of the data for phase composition and strain was done using TOPAS [11] and the traditional one-dimensional method [12] respectively. The error calculation in the latter was based on the standard deviation of the fit assuming a Gaussian distribution. Cross-section microstructure examination of the as-sprayed and sample subjected to immersion in simulated body fluid was done using scanning electron microscope. For better quality micrographs, the samples were metallographically prepared and images obtained in secondary electron (SE) mode. Results and discussion Phase analysis: Fig.1 show the superposed diffraction patterns of the as-sprayed and the sample immersed the longest (56 days) as well as the corresponding volume fractions of the starting HAp phase and main thermal product (TTCP) collected at different depths below the coating surface; the bottom patterns in the former represent the shallow depths probed in this geometry. The last top diffraction pattern(s) in the figures corresponds to the Ti alloy indicating that probing extended beyond the coating-substrate interface. The high temperature induced thermal products TTCP, TCP, and CaO can be seen through-out the as-sprayed coating, see Fig. 1a. Upon immersion these phases dissolve, with the latter being first to disappear. After 56 days of immersion, CaO has almost completely disappeared while TTCP and TCP only start appearing deeper in the coating see Fig. 1b. MECA SENS 2017 Materials Research Forum LLC Materials Research Proceedings 4 (2018) 123-128 doi: http://dx.doi.org/10.21741/9781945291678-19 125 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 210 μm 105 μm 70 μm 140 μm 175 μm Z-scan 35 μm Interface region (a) As-sprayed