1292
Langmuir 2009, 25, 1292-1295
Assessing Surface Area Evolution during Biomimetic Growth of Hydroxyapatite Coatings Albert Mihranyan,*,† Johan Forsgren,† Maria Strømme,† and Håkan Engqvist‡ DiVision for Nanotechnology and Functional Materials and DiVision for Materials Science, Department of Engineering Sciences, The Ångstro¨m Laboratory, Uppsala UniVersity, Box 534, 75121 Uppsala, Sweden ReceiVed October 22, 2008. ReVised Manuscript ReceiVed December 16, 2008 The surface area of biomimetically deposited hydroxyapatite (HA) coatings on metallic implants is important for the biological performance of the implant. Thus, a nondestructive method of assessing this quantity directly on the solid substrate would be highly valuable. The objective of this study was to develop such a method and for the first time assess the evolution of surface area of HA during biomimetic growth. The surface area of a TiO2-covered titanium substrate was measured prior to and following the biomimetic coating deposition using Ar gas adsorption at 77 K. The presence of HA on the surface was verified with scanning electron microscopy and X-ray diffraction. The specific surface area of the coating was found to increase linearly during 1 week of deposition at a rate of ∼100 cm2 day-1 (g substrate)-1. The presented method may be used as a tool for studying the evolution in surface area of coatings on solid substrates during biomimetic deposition or other growth processes.
Biomimetic growth of hydroxyapatite (HA; chemical formula Ca10(PO4)6(OH)2) on implants has been studied vastly during recent years, mainly because this growth can be used as an in vitro test of a surface’s possible bioactivity, and because biomimetically grown HA coatings can be used to obtain a better biological response.1-5 HA is an appealing implant coating material because of its similarity to the low-crystallinity calcium phosphate mineral that constitutes the hard phase of bone. It can be deposited on substrates in several ways, e.g., by plasma spraying,6 pulsed laser deposition,7 and electron-beam deposition.8 The industrial golden standard for depositing HA coatings on implants is plasma spraying, giving some tens- to hundreds-of-micrometers-thick coatings. However, plasma spraying is a line-of-sight method and cannot be used for deposition of coatings on complex geometries. HA can also be deposited on metallic substrates by biomimetic precipitation from a liquid solution after treatment of the metallic surface.9,10 The development of biomimetically grown HA coatings has been addressed by several researchers,9–23 because this type of coating is easily applied to complex * Corresponding author.
[email protected]. † Division for Nanotechnology and Functional Materials. ‡ Division for Materials Science. (1) Hyup Lee, J.; Ryu, H.-S.; Lee, D.-S.; Sun Hong, K.; Chang, B.-S.; Lee, C.-K. Biomaterials 2005, 26, 3249–3257. (2) Rocca, M.; Fini, M.; Giavaresi, G.; Nicoli Aldini, N.; Giardino, R. Biomaterials 2002, 23, 1017–1023. (3) Yildirim, O. S.; Aksakal, B.; Celik, H.; Vangolu, Y.; Okur, A. Med. Eng. Phys. 2005, 27, 221–228. (4) Jaffe, W. L.; Scott, D. F. J. Bone Jt. Surg. (Am.) 1996, 78A, 1918–1934. (5) Sun, L. M.; Berndt, C. C.; Gross, K. A.; Kucuk, A. J. Biomed. Mater. Res. 2001, 58, 570–592. (6) Ha, S. W.; Mayer, J.; Koch, B.; Wintermantel, E. J. Mater. Sci. Mater. Med. 1994, 5, 481–484. (7) Blind, O.; Klein, L. H.; Dailey, B.; Jordan, L. Dent. Mater. 2005, 21, 1017–1024. (8) Lee, S. H.; Kim, H. E.; Kim, H. W. J. Am. Ceram. Soc. 2007, 90, 50–56. (9) Wang, X. X.; Yan, W.; Hayakawa, S.; Tsuru, K.; Osaka, A. Biomaterials 2003, 24, 4631–4637. (10) Kim, H. M.; Miyaji, F.; Kokubo, T.; Nakamura, T. J. Biomed. Mater. Res. 1996, 32, 409–417. (11) Barrere, F.; Layrolle, P.; van Blitterswijk, C. A.; de Groot, K. J. Mater. Sci. Mater. Med. 2001, 12, 529–534. (12) Feng, Q. L.; Wang, H.; Cui, F. Z.; Kim, T. N. J. Cryst. Growth 1999, 200, 550–557. (13) Jonasova, L.; Muller, F. A.; Helebrant, A.; Strnad, J.; Greil, P. Biomaterials 2004, 25, 1187–1194.
geometries at low temperature and also because of the possibility to include drugs and growth factors in the coatings. The latter has been demonstrated by loading various antibiotics,18,24 bisphosphonates,25 and bone morphogenetic proteins22 into such coatings. The biomimetic growth process has been shown to result in homogeneous porous coatings with excellent biocompability and bioactivity.26 Whereas physicochemical properties of HA have been shown to determine surface reactivity in vivo, which is important for chemical and cell-mediated degradation of the material as well as for osteogenic differentiation of endogenous cells,27 structural properties, such as surface area, have also been proven to affect biological performance. Surface interaction with cells, proteins, and body fluids in vivo as well as the possibility to load HA coatings with active molecules before implantation are expected to be dependent on the HA surface area.28 Extensive porosity in the coating may be used to improve nutrient supply and to allow for a faster tissue ingrowth.29–32 Several measures may be taken to increase the surface area of an HA implant coating to improve the biological (14) Kikuchi, M.; Ikoma, T.; Itoh, S.; Matsumoto, H. N.; Koyama, Y.; Takakuda, K.; Shinomiya, K.; Tanaka, J. Compos. Sci. Technol. 2004, 64, 819–825. (15) Kim, H. M.; Miyaji, F.; Kokubo, T.; Nakamura, T. J. Mater. Sci. Mater. Med. 1997, 8, 341–347. (16) Lu, X.; Leng, Y. Biomaterials 2005, 26, 1097–1108. (17) Stigter, M.; Bezemer, J.; de Groot, K.; Layrolle, P. J. Controlled Release 2004, 99, 127–137. (18) Stigter, M.; de Groot, K.; Layrolle, P. Biomaterials 2002, 23, 4143–4153. (19) Liu, Y.; Huse, R. O.; de Groot, K.; Buser, D.; Hunziker, E. B. J. Dent. Res. 2007, 86, 84–89. (20) Liu, Y.; Li, J. P.; Hunziker, E. B.; de Groot, K. Philos. Trans. R. Soc. A-Math. Phys. Eng. Sci. 2006, 364, 233–248. (21) Aebli, N.; Stich, H.; Schawalder, P.; Theis, J. C.; Krebs, J. J. Biomed. Mater. Res. A 2005, 73A, 295–302. (22) Liu, Y. L.; Hunziker, E. B.; Layrolle, P.; de Bruijn, J. D.; de Groot, K. Tissue Eng. 2004, 10, 101–108. (23) Ginebra, M. P.; Traykova, T.; Planell, J. A. J. Controlled Release 2006, 113, 102–110. (24) Forsgren, J.; Brohede, U.; Mihranyan, A.; Engqvist, H.; Strømme, M. Key Eng. Mater. 2009, 396-398, 523–526. (25) Åberg, J.; Brohede, U.; Mihranyan, A.; Strømme, M.; Engqvist, H. Key Eng. Mater. 2009, 396-398, 523–526. (26) Ma, J.; Wong, H. F.; Kong, L. B.; Peng, K. W. Nanotechnology 2003, 14, 619–623. (27) LeGeros, R. Z. Clin. Mater. 1993, 14, 65–88. (28) Vanis, S.; Rheinbach, O.; Klawonn, A.; Prymak, O.; Epple, M. Materialwiss. Werkstofftech. 2006, 37, 469–473.
10.1021/la803520k CCC: $40.75 2009 American Chemical Society Published on Web 12/30/2008
Letters
performance, e.g., optimizing the coating deposition process and incorporation of rapidly resorbable phases. The surface area of HA powders is routinely obtained from Brunauer-Emmet-Teller (BET) analysis33 of N2 adsorption isotherms. The structural properties of a HA coating are dependent on the properties of the substrate on which it is deposited. If a coating is scraped off a substrate to obtain a powder, some of the structural properties may change. Hence, structural parameters obtained from measurements on such a powder may not be representative for the coating. A nondestructive method of assessing the parameters of biomimetic HA coatings directly on the surface of metal implants would therefore be highly valuable. However, to our knowledge, no such methods have been hitherto reported. In this study, we demonstrate a procedure to assess the evolution of the surface area of HA coatings biomimetically grown on TiO2-coated titanium substrates. The procedure is performed ex situ by BET analysis of Ar adsorption isotherms recorded on samples following successive biomimetic depositions. HA has been proven to precipitate and bond to crystalline TiO2 surfaces when soaked in noncellular body-like fluids.9,34,35 This was utilized in the present study where a layer of crystalline TiO2 was deposited to titanium substrates prior to soaking in phosphate buffered saline (PBS). The resulting coating when using this solution is a calcium-deficient layer of HA.36 A coating of Ti oxide was deposited on 20 mm × 20 mm × 1 mm plates of commercially pure grade 2 titanium by physical vapor deposition (PVD). To achieve thin films of crystalline titanium dioxide, a reactive magnetron sputtering process was used. A titanium target was mounted in a PVD chamber (Balzers 640R) with oxygen as reactive gas. The substrates were cleansed before deposition in an alkaline cleaning agent (UPON, pH 11.6) in a 60 °C ultrasonic water bath for 5 min and then rinsed in deionized water, followed by 5 min in an ultrasonic bath of ethanol. The substrates were carefully dried with compressed air and then mounted on the substrate holders. The coating process was performed using the following parameters: Arc current 150 A, coating time 20 min, substrate voltage -20 V, magnetron effect 1.5 kW, an oxygen partial pressure in the range of 10-3 mbar and a temperature during the coating process of 350 °C. The thickness of the titanium dioxide sputtered during 20 min on reference 100 silicon wafer was 74 nm as measured using cross-section SEM. The TiO2-covered titanium substrates were cut into strips to the dimensions 20 mm × 5 mm × 1 mm to fit into the measuring glass tube bulb of the below described gas adsorption equipment. Before soaking in PBS, the strips were ultrasonically cleaned in acetone, ethanol, and water consecutively for 5 min in each medium. Subsequently, the strips were placed in preheated Dulbecco’s PBS (Sigma-Aldrich) in an incubator at 60 °C. The strips were removed from the buffer for characterization, as described below, after 1 and 3 days of (29) Karageorgiou, V.; Kaplan, D. Biomaterials 2005, 26, 5474–5491. (30) Adachi, T.; Osako, Y.; Tanaka, M.; Hojo, M.; Hollister, S. J. Biomaterials 2006, 27, 3964–3972. (31) Okuda, T.; Ioku, K.; Yonezawa, I.; Minagi, H.; Kawachi, G.; Gonda, Y.; Murayama, H.; Shibata, Y.; Minami, S.; Kamihira, S.; Kurosawa, H.; Ikeda, T. Biomaterials 2007, 28, 2612–2621. (32) Habibovic, P.; Yuan, H. P.; Van den Doel, M.; Sees, T. M.; Van Blitterswiik, C. A.; De Groot, K. J. Orth. Res. 2006, 24, 867–876. (33) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309–319. (34) Forsgren, J.; Svahn, F.; Jarmar, T.; Engqvist, H. Acta Biomater. 2007, 3, 980–984. (35) Lindberg, F.; Heinrichs, J.; Ericson, F.; Thomsen, P.; Engqvist, H. Biomaterials 2008, 29, 3317–3323. (36) Forsgren, J.; Svahn, F.; Jarmar, T.; Engqvist, H. J. Appl. Biomater. Biomech. 2007, 5, 23–27.
Langmuir, Vol. 25, No. 3, 2009 1293
Figure 1. XRD spectra of a TiO2-covered titanium substrate and of a biomimetic HA coating deposited in PBS for 7 days.
soaking in PBS and thereafter reimmersed in the buffer to continue the biomimetic growth process unless otherwise stated. The deposition was stopped after 7 days, and the final coating was characterized. The sputter-deposited TiO2 coatings as well as the biomimetic coating deposited during 7 days in PBS were analyzed using grazing-incidence X-ray diffraction (XRD, Siemens D5000 diffractometer) operating with an incidence angle of 1° in parallel beam geometry using Cu KR radiation (wavelength of 1.540 598 Å) scanned between 24° and 30° 2θ. Images of all coatings were recorded using environmental scanning electron microscopy (ESEM, FEI/Phillips XL30, Netherlands) at a pressure of 3.5 × 10-6 mbar. The samples were mounted on aluminum stubs with adhesive tape. The samples analyzed by electron microscopy were not reimmersed in the PBS buffer after characterization. Hence, the samples analyzed by electron microscopy after 3 and 7 days in PBS had not been removed from the buffer for previous analysis. The specific surface area of the samples was measured with an ASAP 2020 (Micromeritics, USA) using Ar as the analysis gas at 77 K. The specific surface area was obtained by a multipoint (8-point) BET measurement at relative pressures p/p0 between 0.01 and 0.2. In order to achieve as accurate measurement results as possible, the number of strips used for each adsorption measurement was maximized resulting in a total weight of around 8 g (Mettler Toledo, Al 204). Figure 1 shows XRD spectra of the sputter-deposited TiO2 sample as well as of the coating formed when the sample had been stored in a PBS buffer for 7 days. The TiO2 coating was clearly crystalline with grains of both anatase and rutile phase with grain sizes on the order of 10 nm according to an analysis of the diffraction peaks using the Scherrer equation,37 and the spectrum of the biomimetically deposited coatings showed the presence of crystalline HA. The latter was to be expected since both anatase and rutile TiO2 have been shown to be bioactive and enhance spontaneous formation of HA on the surface when stored in a simulated body fluid buffer.38 XRD spectra recorded on samples that had been stored in PBS during 1 and 3 days, respectively, did not show distinct HA peaks, even if HA crystals were clearly present on the surface in the SEM images (Figure 2). This is in accordance with earlier findings concluding that XRD patterns from the early stage of HA growth on implant surfaces rarely show distinct HA peaks35 either because the
1294 Langmuir, Vol. 25, No. 3, 2009
Letters
Figure 2. ESEM images of TiO2-covered titanium substrate (a), soaked in PBS buffer for 1 (b), 3 (c), and 7 (d) days.
Figure 3. Specific surface area of TiO2-covered titanium substrate (0 days) during biomimetic deposition of HA (1-7 days). The surface area values are given per gram TiO2-covered titanium substrate. Error bars indicate the standard deviations.
coatings are too thin or because the crystallinity increases with aging time. Electron microscopy images of the different samples are shown in Figure 2. Evidently, the TiO2-covered titanium substrate surfaces appeared smooth and free from adsorbed minerals (Figure 2a). Already after 1 day of soaking in PBS, islands of HA crystals were clearly visible (Figure 2b), and after 3 days in the buffer, they covered the entire surface (Figure 2c). In Figure 3, the increase in specific surface area of the biomimetically deposited HA coatings as a function of soaking time in PBS is depicted. Interestingly, the surface area was found to increase linearly with soaking time, and the rate of increment
was ∼100 cm2 g-1 day-1. It should be emphasized that the obtained surface area values are given per gram of TiO2-covered titanium substrate. The weight of the HA coating is negligible compared to the weight of the substrate, and thus, the weight of the sample was considered constant during the measurements. N2 gas adsorption and desorption measurements are the most common methods employed for assessing the surface area and pore size distribution, respectively, of materials. However, for surface areas well below 1 m2 g-1, N2 gas adsorption is not suitable. For this purpose, we chose Ar gas adsorption at liquid nitrogen temperature (77 K, p0 ) 195 Torr), since Ar gas molecules have the same dimensions as N2 molecules but provide 4-fold gain in sensitivity of measurement due to the lower saturation pressure.39 Whereas the use of Ar at liquid N2 temperature is very practical to measure small specific surface areas, it also imposes some limitations. In particular, the pore size distribution of the HA coating cannot be obtained by considering condensation of Ar in pores according to the classical Kelvin equation at liquid N2 temperature. At 77 K, Ar is below its triple point and will therefore solidify at a saturation pressure p0 of 195 Torr. At these conditions, there is no surface tension, which is a necessity for employing the Kelvin equation. To summarize, a procedure to study the increase in surface area of HA coatings during biomimetic growth on crystalline TiO2covered titanium substrates was demonstrated. The method relied on ex situ Ar gas adsorption measurements following successive biomimetic depositions. The surface area of the coating was shown to increase linearly with biomimetic deposition time. The presented method may be used as a tool for studying the evolution in surface area of coatings on solid substrates during biomimetic deposition or other growth processes. The presented approach is the first step in introducing useful techniques for assessing the relationship
Letters
between the surface area, local porosity, coating density and roughness, and clinical properties of biomimetically grown HA on metal implants. (37) Guinier, A. X-Ray Diffraction: In Crystals, Imperfect Crystals, and Amorphous Bodies; Dover: New York, 1994. (38) Zhou, W.; Zhong, X. X.; Wu, X. C.; Yuan, L. Q.; Shu, Q. W.; Xia, Y. X.; Ostrikov, K. J. Biomed. Mater. Res. A 2007, 81A, 453–464.
Langmuir, Vol. 25, No. 3, 2009 1295
Acknowledgment. Mattias Lindgren is gratefully acknowledged for producing the TiO2 coatings used in the present project. The Swedish Funding Agency Vinnova is greatly acknowledged for the VinnVerification grant used to support the presented project. LA803520K (39) Webb, P. A.; Orr, C. Analytical Methods in Fine Particle Technology; Micromeritics Instrument Corporation; Norcross, USA, 1997.
