Surface analysis of titanium dental implants with different topographies

Surface Analysis of Titanium Dental Implants with Different Topographies

M.H. Prado da Silva^a, G.A. Soares^a*, C.N. Elias^b, J.H.C. Lima^c,

H. Schechtman^d, I.R. Gibson^e, S.M. Best^e

^aCOPPE/Universidade Federal do Rio de Janeiro, C.P. 68505,

21945-970 Rio de Janeiro - RJ, Brazil

^bEEIMVR/Universidade Federal Fluminense, Volta Redonda - RJ, Brazil

^cInstituto Brasileiro de Implantodontia - RJ, Brazil

^dFIOCRUZ, Rio de Janeiro - RJ, Brazil

^eIRC in Biomedical Materials, QMWC, University of London, U. K.

e-mail: [e-mail]

Received: December 17, 1999; July 31, 2000

Cylindrical dental implants made of commercially pure titanium were analysed in four different surface finishes: as-machined, Al₂O₃ blasted with Al₂O₃ particles, plasma-sprayed with titanium beads and electrolytically coated with hydroxyapatite. Scanning electron microscopy (SEM) with Energy Dispersive X-ray Analysis (EDX) revealed the topography of the surfaces and provided qualitative results of the chemical composition of the different implants. X-ray Photoelectron Spectroscopy (XPS) was used to perform chemical analysis on the surface of the implants while Laser Scanning Confocal Microscopy (LSM) produced topographic maps of the analysed surfaces. Optical Profilometry was used to quantitatively characterise the level of roughness of the surfaces. The implant that was plasma-sprayed and the hydroxyapatite coated implant showed the roughest surface, followed by the implant blasted with alumina and the as-machined implant. Some remnant contamination from the processes of blasting, coating and cleaning was detected by XPS.

Keywords: titanium, dental implants, surface characterisation

1. Introduction

When selecting materials to be used as implant devices, many aspects must be considered and, depending on the application, one or more properties will be a decisive factor on the choice of the material. In the study of dental implants, the physics and chemistry of the implant's surface must be studied closely as the presence of debris or contaminant particles can affect the tissue response. This emphasis is related to the nature of bone-metal interactions that take place at an atomic scale¹.

The surface of a metal may differ in properties from the bulk material and these differences may vary from surface free energy to differences in composition due to segregation or contamination. Techniques such as X-ray photoelectron spectroscopy (XPS), also called electron spectroscopy for chemical analysis (ESCA), can be used to quantify the chemical composition of surfaces. In this kind of analysis, a specimen is submitted to an X-ray beam and electrons from the valence layers are ejected giving rise to transitions of electrons from inner layers to occupy the generated vacancies. Peaks of binding energy corresponding to these transitions are detected and quantified, giving a “finger print” of the surface². Scanning electron microscopy (SEM) is very useful for the study of surfaces as this technique has good depth of focus and energy dispersive analysis (EDX) can be performed to chemically characterise features observed on the analysed surface. Laser scanning confocal microscopy (LSM) and profilometry can give qualitative and/or quantitative information of topography. Wennerberg et al.³ developed an optical system for the topographic description of surfaces and measured some roughness parameters to quantify the roughness of different surfaces. Some authors have also used Auger Electron Spectroscopy (AES) and Atomic Force Microscopy (AFM)⁴ for the surface characterisation of biomaterials.

The role of surface roughness on the biocompatibility of titanium implants has been discussed by various authors^5-7. There are two main mechanisms of bone-implant attachment: mechanical attachment and chemical attachment. Roughness mainly improves mechanical attachment as it permits bone to grow through pores or features on the surface of the implant. On the other hand, the contribution of roughness to the improvement of osseointegration is not only mechanical. Features like surface tension and thus surface energy, can change the hydrophobic character of a surface. The ability a surface has to adsorb organic molecules like proteins is directly related to biocompatibility. In this context, surface roughness has been found to positively influence cell response to titanium implants^8-10.Some authors have also investigated the influence of surface roughness on bone attachment and concluded that rougher surfaces result in improved tissue responses to titanium implants^6,11-13.

In this study, four different types of surfaces were produced on titanium implants and these were characterised by SEM, LSM, XPS and optical laser profilometry. The surface conditions were: as-machined; blasted with alumina particles; plasma sprayed with titanium beads and electrolytically coated with hydroxyapatite (HA). The HA coating procedure was developed by Silva and co-workers ¹⁴ from the studies of Asaoka¹⁵. This process offers the advantage of being a less expensive alternative to the plasma-spray process. Another advantage is the lower temperature of the process, in comparison to the high temperature of plasma-spraying, that gives rise to the decomposition of hydroxyapatite into a mixture of amorphous and transformed phases.

2. Materials and Methods

2.1. Materials preparation

The implants used in this study were made from commercially pure titanium 7 mm in length by 2.2 mm in diameter. Four different surface treatments were carried out which resulted in four different implant surfaces: the first condition consisted of an as-machined implant, which was the condition that the implant came from the turning process; the second condition was a surface blasted with Al₂O₃ particles of granulometry in the range 250-600 mm; the third surface was plasma-sprayed with titanium beads in the range 300-500 mm; the fourth condition was an implant electrolytically coated with hydroxyapatite.

The hydroxyapatite coating process was developed in the IRC in Biomedical Materials from the technique by Asaoka et al.¹⁵ and consisted of electrolytically coating the implants with monetite and the subsequent conversion of the monetite to hydroxyapatite. Details of this procedure is given in the study by Da Silva and co-workers¹⁴.

2.2. Surface characterisation techniques

SEM analysis was performed using a Jeol field emission SEM with an accelerating voltage of 10 kV. A laser confocal microscope LSM410 (Zeiss) was used to construct topographic maps. These maps were built from a series of virtual sections of each sample that were interpreted by the LSM 3.95 software. XPS analysis was carried out in a 1257 PHI X-ray photoelectron spectrometer operating at 13.0 kV, at an angle of 54 degrees and with the Mg-ka energy (1253.6eV). The XPS analysis was performed after a sputtering of about two atomic layers. For the profilometry measurements, an optical laser profilometer Mahr GmbH TERTHOMETER CONCEPT was used. The measured parameters were: R_a, the arithmetic average of the absolute values of all points of the profile; R_q, the root-mean-square of the values of all points and R_z, the average value of the absolute heights of the five highest peaks and the depths of the five deepest valleys. The roughness parameters were measured in five regions for each surface condition.

3. Results and Discussion

SEM analysis of the as-machined specimen showed a relatively clean surface, as shown in Fig. 1a. The specimen blasted with alumina was found to exhibit a rougher surface and the presence of some aluminium-rich particles resulting from the blasting process, Fig. 1b. These particles appeared to be evenly distributed on the implant surface.

Figure 1.
SEM picture showing the surface of the implant: (x 3.000). a) as-machined; b) Al₂O₃ blasted; c) plasma-sprayed with titanium; d) electrolitically coated with hydroxyapatite.

The titanium plasma-sprayed sample showed an uniform and very rough surface, Fig. 1c, and a melted aspect of some particles. The implant that was electrolytically coated with hydroxyapatite exhibited an homogeneous surface, as can be seen in Fig. 1d. The microstructure consisted of small hydroxyapatite crystallites, in the range of 1-5 mm and X-ray diffraction indicated no other calcium phosphate phase than hydroxyapatite. At higher magnification, as shown in Fig. 2, it could be observed that nanometric crystals grew from the tip of each individual hydroxyapatite crystallite. The rough surfaces observed in Figs. 1b to 1d suggest, as might be expected, that the Al₂O₃ blasted, the titanium plasma-sprayed and the hydroxyapatite samples would have a much higher surface area than the as-machined sample.

Figure 2.
Hydroxyapatite coated implant under high magnification. (x 10.000).

Figure 3 shows a cross section of the electrolytically produced hydroxyapatite coating from this study. The thickness of the electrolytically produced hydroxyapatite coating varied from 5 to 30 mm. Other processes result in thicker coatings as Lacefield¹⁶ found the range 40 to 60 mm for electrochemical processes and, according to Willmann¹⁷, the thickness of hydroxyapatite coatings produced by plasma-spray varies from 50 to 250 mm. The thinner layer can be a positive point for electrolytic deposition process as, usually, adhesive strength increases with decreasing coating thickness¹⁷. Willmann¹⁷ explains the compromise between coating thickness and adhesion in terms of the tendency thicker coatings have to peel off their substrates. The ideal value for coating thickness has not already been determined because there is a compromise between thickness and resorption and it is not well established how long a coating needs to be present after insertion in the human body¹⁸.

Figure 3.
Cross section the specimen electrolytically coated with hydroxyapatite (x 1.000).

The topographic maps, Figs. 4a to 4d, showed, qualitatively, the difference in roughness between the four surfaces. When comparing the maps of the different specimens, it should be noted that a different scale was used for the machined specimen. Laser scanning confocal microscopy can be an useful tool for roughness measurements but this quantification depends on pattern calibration, so this qualitative analysis was complemented by the measurement of the roughness parameters by profilometry. The roughness parameters of the analysed surfaces are shown in Table 1. The titanium plasma-sprayed sample and the electyrolytically coated specimen exhibit R_a equal to 5.0 mm, a higher value than the measured one for the other two surface conditions. The similarity in roughness between the plasma-sprayed and the electrolytically coated specimens was also observed for the other two roughness parameters (R_q and R_z). The alumina blasted specimen showed the highest values of R_q and R_z, indicating a very irregular surface. The arithmetic average roughness (R_a) for the electrolytically coated implant is in the same range of values reported in the literature for hydroxyapatite coated by a plasma-spraying process, according to Wong et al.¹⁹. Moreover, as mentioned before, the implant that was electrolytically coated with hydroxyapatite showed crystallites in the scale of nanometres. This feature suggests a large increase in surface area compared to the other specimens that had features in the scale of microns. As the profilometry measured roughness at a micrometric scale, results from Table 1 could not describe the topography of all samples in a nonometric scale.

Table 1.
Roughness parameters of the analysed implants.

XPS analysis revealed the characteristic adsorption of carbon on the four analysed surfaces. This carbon adsorption occurs on every surface that is exposed to the atmosphere and is detected by the XPS technique. As might be expected, the as-machined implant showed no traces of contaminants, Fig. 5. The Al₂O₃ blasted specimen spectrum (Fig. 6) showed contamination with aluminium, also identified by SEM, which was associated to the alumina particles from the blasting process. Analysis of the titanium plasma-sprayed specimen indicated some contamination with aluminium and zinc, Fig. 7b. This can be associated to the blasting process prior to the plasma-spraying, as this specimen was first blasted with alumina particles to create a rough surface. It is thought that the detergents used in the cleaning procedure were the most probable origin of the zinc contamination²⁰. The hydroxyapatite coated specimen showed traces of sodium, remaining from the coating process, as indicated in the spectrum shown in Fig. 8d. The measured Ca/P ratio varies from 1.58 to 1.73. Considering a possible variation of 10% when quantification is done using XPS spectra, this range could correspond to stoichiometric hydroxyapatite (Ca/P = 1.67) or calcium-deficient hydroxyapatite (Ca/P = 1.50). With infrared spectroscopy using the diffuse reflectance mode (DR-IR) small amounts of carbonate-apatite were identified²¹.

Figure 5.
XPS spectrum for the as-machined specimen.

Figure 6.
XPS spectra for Al₂O₃-blasted sample. a) general view; b) detail near aluminium peak.

Figure 7.
XPS spectra for sample plasma-sprayed with titanium. a) general view; b) details near Al peak.

Figure 8.
XPS spectra for sample electrolytically-coated with hydroxyapatite. a) general view; b) details near calcium peak; c) details near phosphorus peak; d)details near sodium peak.

4. Conclusions

The techniques utilised in this study were effective in the characterisation of the morphology, surface quality and chemical purity of the analysed surfaces. As might be expected, the as-machined implant presented the smoothest surface. The implant that was electrolytically coated with hydroxyapatite showed roughness similar to data obtained for titanium plasma-sprayed specimen and similar to the data reported in the literature for commercial implants coated with hydroxyapatite by the plasma-spraying process. Moreover the nanometric crystallites identified in the sample electrolytically coated with hydroxyapatite act to increase the surface area, although the correlation of this event with cell attachment remains unknown.

SEM/EDS is, as predicted, a good tool for topographic visualisation and/or particle identification while contamination from the different surface treatments was detected by XPS analysis. These findings suggest that careful consideration must be given while manufacturing, cleaning and sterilising the implants since the nature of the surface can influence bone-bonding events.

Acknowledgements

The authors would like to acknowledge the support of CAPES, CNPq, FAPERJ and FUJB for the Brazilian government grant and the EPSRC for funding of the core programme of the IRC in Biomedical Materials. They would also like to thank Conexão Sistemas de Prótese S.A., for supplying the implants. The XPS spectra were obtained in the NUCAT Lab. at PEQ/COPPE/UFRJ.

• 1. Kasemo, B.; Lausmaa, J. Tissue-Integrated Prostheses - Osseointegration in Clinical Dentistry. Quintessence Publishing Co., Inc., Chicago, p. 99, 1985.
• 2. Muster, D.; Demri, B.; Hage Ali, M. Enclyc. Handbook of Biomaterials and Bioengineering, part A, v. 2, p. 785-812, 1995.
• 3. Wennerberg, A.; Albrektsson, T.; Ulrich, H.; Krol, J.J. J. Biom. Eng., v.14, p. 412-418, 1992.
• 4. Siedlecki, C.A.; Marchant, R.E. Biomaterials, v. 19, p. 441-454, 1998.
• 5. Oshida, Y.; Sachdeva, R.; Miyazaki, S.; Daly, J. J. Mater. Sci.: Mater. in Medicine, v. 4, p. 443-447, 1993.
• 6. Wennerberg, A.; Albrektsson, T.; Andersson, B. J. Mater. Sci.: Mater. in Medicine, v. 6, p. 302-309, 1995.
• 7. Ishizawa, H.; Fujino, M.; Ogino, M. Bioceramics, v. 9, Pergamon/Elsevier, UK, p. 329-332, 1996.
• 8. Keller, J.C.; Draughon, R.A.; Wightman, J.P.; Dougherty, W.J.; Meletiou, S.D. J. Oral & Maxillofacial Impl, v. 5, p. 360-367, 1990.
• 9. Kieswetter, K.; Shwartz, Z.; Hummert, T.W.; Cochran, D.L.; Simpson, J.; Dean, D.D.; Boyan, B.D. J. of Biom. Mater. Res., v. 32, p. 55-63, 1996.
• 10. Boyan, B.D.; Hummert, T.W.; Dean, D.D.; Schwartz, Z. Biomaterials, v. 17, p. 137-146, 1996.
• 11. Suzuki, K.; Aoki, K.; Ohya, K. Bone, v. 21, p. 507-514, 1997.
• 12. Wennerberg, A.; Ektessabi, A.; Albrektsson, T.; Johansson, C.; Anderson, B. J. Oral & Maxillofacial Impl, v. 12, p. 486-494, 1997.
• 13. Wong, M.; Eulenberger, J.; Schenk, R.; Hunziker, E. J. of Biom. Mater. Res., v. 29, p. 1567-1575, 1995.
• 14. Da Silva, M.H.P.; Soares, G.A.; Elias, C.N.; Gibson, I.R.; Best, S.M.; Bonfield, W. Bioceramics, v. 11, p. 223-226, 1998.
• 15. Asaoka, N.; Best, S.; Bonfield, W. Bioceramics, v. 10, p. 447-450, 1998.
• 16. Lacefield, W.R. Implant Dentistry, v. 7, n. 4, p. 315-321, 1998.
• 17. Willmann, G. Bioceramics, v. 10, p. 353-356, 1997.
• 18. De Groot, K.; Wolke, J.C.K.; Jansen, J.A. Journal of Oral Implantology, p. 232-234, 1994.
• 19. Wong, M.; Eulenberger, J.; Schenk, R.; Hunziker, E. Journal of Biom. Mater. Res., v. 29, p. 1567-1575, 1995.
• 20. Prado da Silva, M.H. Ph.D. Dissertation, COPPE/UFRJ, 166 p., 1999.
• 21. Oliveira, J.F.; Sena, L.A.; Prado da Silva, M.H.; Soares, G.A.; Rossi, A.M. Bioceramics, to be published, 2000.

Publication Dates

• Publication in this collection
  17 Nov 2000
• Date of issue
  July 2000

History

• Received
  17 Dec 1999
• Accepted
  31 July 2000