Role of Calcium Hydroxyapatite in Bioglasses: ab-initio characterization of its Surface Properties.

Hydroxyapatite [HA, Ca10(PO4)6(OH)2] is the main constituent of the mineral phase in mammalian bones and teeth enamel. Moreover, its fundamental role in consenting the integration of bioactive glasses implants with damaged bone tissues is well known. In fact, according to Hench’s mechanism, a thin layer of crystalline carbonate hydroxyapatite (HCA) is forming after immersion of bioglasses in SBF solutions (Simulated Body Fluid). Just because of this thin layer, bone reparation can be promoted and the bioactive glass implant result strongly attached. Chemical structure of HCA is very similar to HA’s one, therefore, studying hydroxyapatite microstructure allows to better understand the crucial step of bioglass bonding to living tissues. Within this huge research area, computer simulations have been recently exploited to provide information about materials’ structure and properties, extremely useful when compared with experimental results. So, it is possible to merge both aspects, and obtain a more complete insight of the studied system. In the present work, ab-initio electronic structure methods have been applied to describe both hydroxyapatite bulk and its surfaces. All calculations have been performed using the periodic ab-initio CRYSTAL code, which adopts a Gaussian basis set with the B3LYP functional. Hydroxyapatite is a ionic compound with a hexagonal crystal structure (space group P63/m) and 44 atoms in the unit cell. X-rays data are affected by proton disorder inside the unit cell, due to the presence of OH groups aligned in the c direction. Therefore, it was necessary to lower the symmetry to the P63 space group to define a proton ordered structure. HA structure has been optimised (lattice parameters and internal coordinates) and its properties (band structure and density of states) have been fully characterised. The vibrational spectrum of HA has been evaluated in the harmonic approximation. Agreement with experiments is good, showing the reliability of the B3LYP functional. In order to study the interfacial reactions between HA and bioglasses, different models of the (001) surface have been analysed. The (001) cut brings about a neutral face which has already been reported in literature as the most probable surface. A slab containing only one atomic layer (Figure 1, formation energy ~ 1.2 J/m2) with a thickness of ~ 5 Å has been optimised, showing remarkable changes in its geometry, due to large relaxation process. Figure 1: the relaxed (001) face of HA The second model designed is a double layered slab (thickness ~ 14 Å, formation energy ~ 1.0 J/m2) which undergoes to a far less relaxation showing an optimised geometry close to that of the bulk. Both these surfaces have been characterised by means of various properties (band structure, density of states…), and in particular their electrostatic potential maps at the surface. The latter shows large positive values associated to calcium ions, which are good candidates as adsorption sites on HA surfaces. Both slabs have been loaded with increasing amounts of water molecules, which are ubiquitous in the biological environment. The water binding energy for both slab models is ~ 20/30 kcal/mol, in agreement with microcalorimetry measurements data. The adsorption process is shown to be molecular, without any bond dissociation. These results confirm a good affinity of the HA surfaces to water, as expected. The vibrational frequencies of adsorbed water have also been calculated and shows a comparison with those of free water, a blue shift of the bending mode (average value ~ 80 cm-1), whereas a red shift (~ 100-1200 cm-1) has been calculated for the OH stretching. This behaviour is a consequence of the hydrogen bond between water and oxygens of superficial phosphates groups (hydrogen bond length varies from 1.5 to 2 Å). These results are in remarkable agreement with the experimental IR data. In conclusion, the (001) hydroxyapatite surface has been analysed by theoretical methods with the purpose of understanding their structure and reactivity, in particular towards water adsorption. So far, results have shown good agreement with measured data, allowing for the further step, which will involve the interaction of larger molecular probes. On one hand, we will aim at modelling amino-acids adsorptions on hydroxyapatite surfaces. On the other hand, it would be of great interest simulating the interaction between HA and silica based materials, thus approaching to Hench’s reaction steps.