Introduction: Thermally treating various calcium phosphates at 500°C for 24 hours reduce their chemical reactivity[1] and modify their osteoclastic response[2]. So far, these effects have been attributed to the removal of surface defects1. The aim of this study was to revisit this topic by performing a thorough XPS analysis of dense β-tricalcium phosphate (β-TCP, β-Ca3(PO4)2) samples.
Materials and Methods: 99% dense β-TCP samples were produced by a combination of slip-casting and sintering (1100 °C for 3 h). Crystalline purity was assessed by applying a Rietveld refinement analysis of X-ray powder diffraction patterns (Philips PW1800 θ/2θ diffractometer, Panalytical, Almelo, The Netherlands). XPS measurements were performed using an Axis Nova instrument (Kratos Analytical Ltd, UK). The system was calibrated according to ISO 15472 and the accuracy was better than ± 0.05 eV. The spectra were analyzed using CasaXPS software (V2.3.14, Casa Software Ltd., UK). Peak shifting was corrected by referencing aliphatic carbon to 285 eV. The areas of the peaks were determined after subtraction of an iterated Shirley background, corrected by the sensitivity factors given by Kratos.
Results and Discussion: The XPS survey spectra show prominent signals of O, Ca, P and C. After calcination, smaller signals of Mo and K as well as traces of Si and Na close to the detection limit were found. Before calcination, the Ca/P molar ratio was 1.53±0.02, which is close to the theoretical ratio of 1.5. After calcination, the ratio increased to 1.78±0.11 (Student’s t-test: p < 0.01). The various compounds detected after calcination (Si, Na, K, Mo) are most likely caused by a contamination during calcination. The increase of the Ca/P molar ratio of the platelets during calcination was attributed to the evaporation of phosphor pentoxide[3],[4]. Since neither more binding oxygen (P-O-P) was present nor the binding energy of phosphorus did change, the increase of the O/P molar ratio from 4.1 ± 0.1 to 5.2 ± 0.2 (Student’s t-test: p < 0.001) must be due to the formation of a P-depleted, Ca-rich phase such as hydroxyapatite (HA; Ca5(PO4)3OH), or tetracalcium phosphate (Ca4(PO4)2O; TetCP). In fact, as the O/P molar ratio can only increase from 4 to 4.3 or 4.5 by replacing β-TCP with HA or TetCP, it appears likely that a P-free, Ca-rich phase was formed, such as Ca oxide or Ca hydroxide. Angle-resolved XPS measurements confirmed the assumption that the Ca-rich phase was built on top of the calcium phosphate. Assuming the formation of an approximately 1 nm thick Ca oxide or Ca hydroxide layer during calcination is compatible with a 20% increase of Ca/P and O/P ratio as measured by XPS since XPS retrieves compositional information from the top 10 nm of the surface.
Conclusions: This work presents data suggesting the formation of a P-depleted Ca-rich surface layer during the calcination of β-TCP at 500 °C, possibly Ca oxide or hydroxide.
The financial support of Mathys Ltd (Bettlach, Switzerland) is kindly acknowledged
References:
[1] M. Bohner et al (2009) Acta Biomater, 5:3524-3535
[2] R. Egli et al (2011) Adv Eng Mater, 13:B102-B107
[3] J.H. Welch, W. Gutt (1961) J Chem Soc, 874:44424444
[4] B.C. Wang et al (1993) Surf Coat Tech, 58, 107-117