potentially lead to the formation of non-apatitic calcium phosphate phases or salts other than calcium phosphates.46 The Ca:F ratio of that method was calculated to be about 5:1, which is the appropriate stoichiometry for FAp.34 The incubation of different substrates inside the fluoride-rich supersaturated calcium phosphate solutions, are thought to drive the occurrence of heterogeneous nucleation on the surfaces, which is known to take place faster than the homogenous nucleation, as the substrates decrease the energy required to form a nucleus.47,48 The morphology, coverage, and crystal sizes of apatite coatings precipitated on the substrates, is known to be dependent on the substrate’s chemistry, topography, charge, grain size, lattice geometry, and architecture.47,49 Concave surfaces usually have high spatial charge, enhance the assembling of ions in three dimensions, thus they are considered as good nucleation sites. While convex surfaces dissipate the charge, which then decrease the nucleation.47 Moreover, planar surfaces were also known to act as two dimensions nucleation sites with equal distribution of charge depending on the substrates’ isoelectric point itself rather than its topographical properties.47
After mineralization, we have observed multiple organized structures made of apatite nanocrystals. For example, dandelion shaped nanocrystals (Chart 1c) were seen on all the substrates at 37◦C and 70◦C (Figs. 4a, b, d, e, g, h, j, and k). As well, bowknot-like structures made of assembled nanocrystals (Chart 1d), were observed on NE-Ti at 37◦C on the E-Ti disc at 70◦C (Figure 4a, e). Furthermore, smooth ball-like structures either with spikes (Chart 1a, Figs 4a, b, d, e, g, j, h, and k) or without spikes (Chart 1b, Figs. 4a, b, d, e, g, j, h, and k) were observed on all substrates, except on the NE-Ti, E-Ti, P-Sa at 90◦C. On the other hand, disordered nanocrystals (Chart 1e) were detected on all substrates at 90◦C except the U-Sa (Figure 4c, f, i, and l).
The difference in the morphology between the ball-like structures with and without the spiky features, is potentially due to the outward growth of nanocrystals from inside the ball-like structures. Kinetically, it is proposed that the ball-like structures with the spiky features, can transform at later stages into bowknots. This then leads to the morphological transformation into the dandelion structures. Therefore, those morphologies which are found mainly at 37ᵒC could be tested as potential biocompatible material at the bone-implant interface.
Activation energy for nucleation is known to be dependent on the nucleus-substrate interactions. Crystal growth is favoured at low energies, as it decreases the overall free energy of the critical nucleus formation, thus promoting the sequential phase transformations.50 In our scenario, spherical structures (Chart1a, b, f) were mostly seen at lower temperature as these structures are energetically favoured with less energy needed for their formation. With increasing the temperature to 70ᵒC, self-assembled organized nanocrystals start to grow out of the ball-like structures (Chart 1c,d), with further formation of ordered apatite features. However, at high temperatures (90ᵒC), the nanocrystals seem to lack the hierarchical organization and appeared to be arranged randomly.
Effect of Substrate and Temperature on the Coverage of the Apatite Coatings
Fujishiro et al.51 reported deposited HAp on various substrates; they found that the coverage, morphology, and adhesive strength of the coatings. For example, they found that the coverage changes considerably between different substrates, as titanium was entirely coated compared to sparsely coated alumina substrate. Furthermore, it is well known that at pH of 6, titanium substrate acts as a negatively charged surface,52 while sapphire substrate is either neutral or exhibits a slightly positive surface at values of the pH below the isoelectric point, sapphire has a positive charge.53 Zhu et al.54 observed that the nucleation of HAp is favoured on negatively charged substrates compared to the positively charged ones. Using image analyses, we quantified the extent of the coating coverage on the different substrates at 37C, 70ᵒC and 90C. This quantification was conducted, to assess the effect of the different surface properties including the surface chemistry and topography on the coverage. The highest apatite coverage was found on P-Sa 37C and U-Sa at 70C with a percentage of 45.5 ± 0.03% and 35±0.02% respectively. All other substrates at different temperatures showed moderate coverage between 12-29%. The enhanced apatite coverage, which is seen on the E-Ti surfaces, gives an indication that the apatite nucleation and growth is favoured due to increased grooves and pits on the surface, which in turn increase the nucleation sites.55 and overall coverage. Holbrough et al.56, who confirmed that the surface topography can control efficiently the nucleation sites and in turn the coverage, stated that some cavities on the substrate surface could have the same dimensions of a critical nucleus. Moreover, our results agree with Dennig and Stevenson, and Campbell et al.57,58who stated that the nucleation events prefer rougher surfaces such as scratches.