Biomaterials, Volume 24, Issue 2, January 2003, Pages 263-273

Corrosion behaviour of commercially pure titanium shot blasted with different materials and sizes of shot particles for dental implant applications

Conrado Aparicio^{, , a}, F. Javier Gil^a, Carlos Fonseca^b, Mario Barbosa^b and Josep Anton Planell<sup>a

a</sup> Research Centre in Biomedical Engineering (CREB), Department of Materials Science and Metallurgical Engineering, Universitat Politècnica de Catalunya, Av. Diagonal 647, 08028, Barcelona, Spain
^b Laboratorio de Biomateriais, Instituto de Engenharia Biomédica, Universidade do Porto, Rua do Campo Alegre 823, 4150-180, Porto, Portugal

Received 30 January 2002;  accepted 23 July 2002.  Available online 25 September 2002.

1. Abstract

It is well known that the osseointegration of the commercially pure titanium (c.p. Ti) dental implant is improved when the metal is shot blasted in order to increase its surface roughness. This roughness is colonised by bone, which improves implant fixation. However, shot blasting also changes the chemical composition of the implant surface because some shot particles remain adhered on the metal.

The c.p. Ti surfaces shot blasted with different materials and sizes of shot particles were tested in order to determine their topographical features (surface roughness, real surface area and the percentage of surface covered by the adhered shot particles) and electrochemical behaviour (open circuit potential, electrochemical impedance spectroscopy and cyclic polarisation).

The results demonstrate that the increased surface area of the material because of the increasing surface roughness is not the only cause for differences found in the electrochemical behaviour and corrosion resistance of the blasted c.p. Ti. Among other possible causes, those differences may be attributed to the compressive residual surface stresses induced by shot blasting.

All the materials tested have an adequate corrosion and electrochemical behaviour in terms of its possible use as dental implant material.

Author Keywords: Titanium; Dental implant; Shot blasting; Corrosion; Electrochemical impedance spectroscopy

2. Article Outline

1. Introduction

2. Materials and methods

2.1. Materials

2.2. Methods

2.2.1. Sample preparation

2.2.2. Surface roughness

2.2.3. Covered-surface Percentage

2.2.4. Real surface area and corrected real surface area

2.2.5. Corrosion tests

2.2.6. Open circuit tests

2.2.7. Electrochemical impedance spectroscopy tests

2.2.8. Cyclic polarisation tests

3. Results

4. Discussion

5. Conclusions

Acknowledgements

References

3. 1. Introduction

Commercially pure titanium (c.p. Ti) is widely used as dental implant material [1 and 2] because of its suitable mechanical properties and excellent biocompatibility [3, 4 and 5]. The latter is mainly due to its excellent corrosion behaviour in the physiological environment [6 and 7]. But clinical success is achieved not only because of implant material but also because of other properties as implant design, surgery technique, host bone quality, load bearing and surface quality [8]. Among them, one of the most important is surface quality, which refers to its mechanical, physicochemical and topographic properties [9 and 10]. In this sense, it is known that an increased implant-surface roughness increases the values of cell adhesion, proliferation and differentiation [11, 12, 13, 14, 15, 16, 17 and 18]. Moreover, a better long-term in vivo response is achieved since the percentage of implant in direct contact with bone increases as well as loads and torques for extracting implant from bone [19, 20, 21, 22, 23, 24, 25, 26, 27 and 28]. The improvement of implant fixation is mainly because the implant-bone interlocking increases when surface roughness is colonised by host bone [10 and 29].

Shot blasting is one of the most frequently used treatments for obtaining a rough surface of a dental implant [10 and 30]. The materials of the shot particles, which are bombarded on the dental implant surface, are chemically stable materials that will not stimulate negative responses of the biological behaviour of the implant [31]. This is mainly because shot blasting provokes the adhesion of the shot particles on the implant even after its ultrasonical cleaning, acidic passivation and sterilisation because of the high impact velocity between shot particles and implant surface [32, 33 and 34]. If some of these particles are loosened to the surrounding tissues, they can interfere with adequate bone mineralisation [29 and 35] or stimulate cell adhesion and differentiation [31 and 36]. Moreover, chemical heterogeneity of the implant surface may change the excellent corrosion resistance of the c.p. Ti in physiological environment [3 and 37].

The aim of this work was to determine the influence of size and chemical composition of shot particles used in shot blasting on the electrochemical and corrosion behaviour of the c.p. Ti for dental implant applications. First, the shot-blasted samples were physically characterised by laser profilometry, confocal laser scanning microscopy and scanning electron microscopy. The electrochemical behaviour was studied in Hank's Balanced Salt Solution (HBSS) by means of open-circuit potential vs. time (E vs. t), electrochemical impedance spectroscopy (EIS) and cyclic polarisation tests (CPT). Corrosion natural potentials, breakdown potentials and electrical resistances were determined.

4. 2. Materials and methods

2.1. Materials

The c.p. Ti. Grade II discs of 6 mm diameter were shot blasted with three different sizes and two different materials of shot-blasting particles resulting in seven different series of samples depending on their final surface quality:

Ctr: as machined.

Al2: shot blasted with Al₂O₃-particles of 200 
m mean size.

Al6: shot blasted with Al₂O₃-particles of 600 
m mean size.

Al9: shot blasted with Al₂O₃-particles of 900 
m mean size.

Si2: shot blasted with SiC-particles of 200 
m mean size.

Si6: shot blasted with SiC-particles of 600 
m mean size.

Si9: shot blasted with SiC-particles of 900 
m mean size.

2.2. Methods

2.2.1. Sample preparation

The shot blasting was carried out with a laboratory shot-blasting machine at 2.5 MPa-pressure during the time required for roughness saturation of the samples. The particles used for each sample series were those mentioned above.

After shot blasting the samples were ultrasonically cleaned during 15 min in acetone and 15 min in distilled water. Then the discs were passivated with an 0.6 
HCl solution during 30s at room temperature, and immediately washed in distilled water.

The side of the disc used in electrical connections was polished with 600-SiC-abrasive paper. Then the samples were ultrasonically cleaned with the same steps mentioned before. The mechanical contact between the sample and the wire was guaranteed by means of a cyanoacrylate polymer, and the electrical contact by means of silver ink. The electrical contacts as well as the non-tested sample areas were isolated by means of a wax coverage consisting of a mixture of Beeswax (Fluka, Germany) and Colophony (Fluka, Germany).

2.2.2. Surface roughness

The surface roughness measurements were done with a laser profilometer (Perthometer S4P^©, Mahr Perten, Germany); and the profiles and R_a-values were obtained and calculated with a measuring-station instrument (Perthometer S4P^©, Mahr Perten, Germany). R_a is the arithmetical mean deviation of the profile and is caculated as the arithmetical mean of the absolute values of the profile deviations from the mean line.

The main assay conditions were:

• Sampling length: 0.8 mm.

• Number of sampling lengths: 3.

• Evaluation length: 2.4 mm (0.8 mm×3).

• Pre- and post-travel: 0.8 mm.

• Filter type: Gaussian standard.

• Cut-off length: 0.8 mm.

• Number of measurements per sample series: 9 (3 measurements×3 samples).

2.2.3. Covered-surface Percentage

After performing the shot blasting, passivation and cleaning treatments, the percentage of surface area (CS%) covered by the remaining shot particles on sample surfaces was determined by means of a two-step determination. Firstly, backscattered scanning electron microscopy (BS-SEM) images of the different surfaces were obtained. Secondly, BS-SEM images were treated with an image analyser (Paqi^©, CEMUP, Portugal) in order to recognise dark areas (corresponding to shot particles) and light areas (corresponding to non-covered titanium surfaces). Finally, the percentage of dark area was calculated with the image analyser.

Eighteen fields in each group of samples were analysed by means of inscribing squares of 3×3 mm² that were divided in 36 equal square-shaped portions (0.5×0.5 mm²). Alternate portions were selected in order to obtain the 200×-magnification BS-SEM images.

2.2.4. Real surface area and corrected real surface area

The real surface areas (RSA) were calculated from the ratio of real surface area/geometric surface area (RSA/A), which were measured after obtaining 3D-topographical images of the surfaces by means of confocal laser scanning microscopy (CSLM). The geometric surface area (A) was 28.3 mm² for all the discs. CLSM was performed with a Leica TCS 4D (Leica Laser Technik, Germany) adapted to an inverted microscope (DMRIB, Leitz, Germany). Confocal images were captured in the reflection mode using the 488 nm line of a krypton-argon laser as illumination source.

Images were taken using a 10× objective. Images size was 512×512 pixels. Maximum lateral resolution (660 nm) with minimum noise was obtained by adjusting the detector-pinhole size. Each line of the image was the average of four line-scans at the standard scan rate, which reduces noise and improves image quality. By alternating the output of the laser to 10 mW, the intensity of the specimen illumination was set up. The scanned geometric-surface area was 1000×1000 
m². The recording and processing of the 3D-image and calculation of the RSA/A ratios were done with a standard Leica software. A total of six discs per series were analysed.

An approximated value of the exposed titanium surface (CRSA) was calculated as the corresponding value of RSA minus the projected area that is covered by the remaining shot particles:

CRSA=RSA−(RSA*(CS%/100)).

2.2.5. Corrosion tests

The corrosion tests were performed in a glass cell (200 ml) containing HBSS (Sigma, UK) at 37°C. Chemical composition of HBSS is detailed in Table 1. A calomel reference electrode (ESC) and a platinum counterelectrode were used. At least two discs of each series were tested, except for Al2, in which only a sample was tested.

Table 1. Chemical composition of HBSS

2.2.6. Open circuit tests

The E vs. t tests were carried out according to ASTM G31-90 [38] standard. The natural corrosion potential (E_corr) was measured when a maximum 3 mV variation in half an hour was attained. The data were recorded at 4 s-intervals. Electrical-potential recording and data acquisition were carried out with a potentiostat (273, EG&G PAR, USA) and a computer with appropriate software (352/252 v 2.01, EG&G PAR, USA). A cleaned and polished sample without passivation treatment (Non-Pass) was also tested in order to study the influence of passivation in E_corr.

2.2.7. Electrochemical impedance spectroscopy tests

The electrochemical Impedance Spectra were acquired using a Solartron 1250^© frequency response analyser connected to a potentiostat (273, EG&G PAR, USA). The frequency domain analysed ranged from 2 mHz to 60 kHz, the amplitude of the imposed ac signal was 5 mV (rms) and the experiments were carried out at the E_corr achieved in the E vs. t test. Data acquisition was done with a computer with appropriate software (m388^© v2.90, EG&G PAR, USA). Data were recorded at 5-frequencies/decade. In order to obtain the equivalent circuits and their parameter-values, all data were analysed and fitted using the software Equivalent Circuit^© developed by Boukamp [39].

2.2.8. Cyclic polarisation tests

The CPT measurements were performed according to the standard ASTM G61-86 method [40]. The samples were immersed for 15 min in the electrolyte prior to starting the polarisation scan at −400 mV. The scan was initiated in the more noble direction at a scan rate of 1 mV/s. When 3000 mV was reached, the scanning direction was reversed. The scan was terminated when the potential reverted, once again to −400 mV. The Potentiostat, ramp generator and software were the same as used in the E vs. t-tests. The data were recorded at 3 s-intervals.

5. 3. Results

The R_a-values are shown in Table 2. As expected, the greater the particle size, the higher the R_a, with statistically significant differences (Fisher test). In the same way, statistically significant differences between samples shot blasted with different particle material and the same particle size have been determined. The discs treated with SiC-particles show slightly lower R_a-values than those shot blasted with Al₂O₃-particles.

Table 2. R_a, CS%, RSA/A, RSA and mean CRSA-values for all sample series


SD: standard deviation; A=28.3 mm².

The results of the percentage of covered surface (CS%) (Table 2) show that there are no statistically significant differences between samples shot blasted with the same particle material but different particle size. However, the discs treated with Al₂O₃-particles have higher CS%-values than those treated with SiC-particles, whatever the size of the particles used (p-value<0.001, t-Student test). BS-SEM images obtained for Si6 and Al6-samples are shown in Fig. 1.

(60K)

Fig. 1. Comparing example of BS-SEM images for Si6 (left) and Al6 (right), where the dark areas correspond to attached particles.

The RSA and CRSA calculations are shown in Table 2. Results show that the CRSA strongly increases with shot blasting and the higher the size of shot particle, the higher the CRSA-values, irrespective of the material of the particles. However, the differences in CRSA observed are only statistically significant (Fisher's test) between the samples shot blasted with 200 
m particles and all other blasted series. Statistically significant differences (Fisher's test) between series shot blasted with the same size of the particle but different particle material have not been determined. The Ctr-samples have the lowest mean CRSA-value showing statistically significant differences (Fisher's test) with all the blasted series.

Fig. 2 shows a representative example of the E vs. t-results for each series. The E_corr is approximately between −150 and −275 mV for all the studied samples, except for Non-Pass series, which had significantly less-noble E_corr-value. Significant trends either due to shot-particle material or shot-particle sizes have not been found.

(12K)

Fig. 2. Example of E vs. t-curves for each series.

A simple equivalent circuit (Fig. 3) consisting of the electrolyte resistance R_s and a parallel CPE-R_f couple, corresponding to the titanium oxide film has been proposed in order to electrochemically describe the behaviour of all the different c.p. Ti surfaces, whatever the particle material or particle size used in shot-blasting treatment. The CPE element represents a shift from the ideal capacitive behaviour and in this case it takes account of both the surface roughness and any oxide heterogeneity. The parameters of the equivalent circuit and the
²-values, which give an estimation of the fit quality, are shown in Table 3. A representative graphic example of the excellent fitting accuracy is reported in Fig. 4. The R_f's in all cases are higher than 0.3 M
 cm² and the CPE's higher than 4.3×10^−5 
−1 cm⁻² sⁿ. Another interesting result is that the higher the size of the particles, the lower the R_f except for Al2-sample.

(4K)

Fig. 3. Model of equivalent circuit obtained from the EIS-results for all the series.

Table 3. Values for the elements of the equivalent circuit (R_s, R_f, CPE, n) shown in Fig. 3 and
²-values for all sample series

(12K)

Fig. 4. Bode plot for Al2 EIS-result with the experimental points of phase (
) and module (Z) and the fitted points obtained with the simulation of the model shown in Fig. 3.

Fig. 5 and Fig. 6 show the CPT-results. The higher the size of the shot particles, the higher the current passing through the samples. The Al₂O₃-blasted materials do not show remarkable differences compared to the potentiodynamic behaviour of Ctr-samples (Fig. 5). However, when the polarising potential is beyond 1000 mV, materials blasted with SiC-particles show increased currents more than an order of magnitude compared to 900 
m Al₂O₃-blasted series (Fig. 6).

(9K)

Fig. 5. Representative CPT-curves of Ctr-series and series shot blasted with Al₂O₃-particles.

(11K)

Fig. 6. Representative CPT-curves of Al9-series and series shot blasted with SiC-particles.

6. 4. Discussion

The R_a-values (Table 2) show that, as expected, differences in surface roughness due to different sizes of the shot-blasting particles are statistically significant. Furthermore, there are differences in roughness if the shot particles are Al₂O₃ or SiC (Table 2). This result may be explained taking into consideration different causes. Firstly, size distribution of the particles is a main influencing factor. The distribution of sizes for a standard mean size [41] can significantly change depending on the material or the manufacturer [10 and 42]. A previous non-published study done with the same particles used in this work demonstrated that most of the Al₂O₃-particles had particles with sizes in the upper value in the permitted standard range of sizes, whereas most of the SiC-particles had values in the middle of the range. This fact implies that despite the mean size of the particles being equal for both SiC and Al₂O₃, it is more likely that Al₂O₃-particles will result in larger marks, which in turn will increase the roughness of the blasted surface. Secondly, the mechanical and physical properties (hardness, fracture toughness and density) of the shot particles can influence surface roughness because the energy transfer when the particle impacts depends on its facility for breaking and its mass as well [43]. In this sense, Al₂O₃ particles have a higher density and a higher fracture toughness (3.96 g/cm³ and 4.2 MPa m^1/2, respectively) than SiC particles (3.22 g/cm³ and 3.3 MPa m^1/2). Moreover, the ball mill friability index (BMFI) is significantly higher for SiC-particles (54) than for Al₂O₃-particles (11.4). The BMFI is a measure of the ease of abrasive-particle breakage when submitted to impacts. All of these properties are related to the energy transfer to the surface during the impact of the particles. Moreover, the higher the impact energy, the higher the plastic deformation [43]. Consequently, the energy transfer is higher with Al₂O₃-particles than with SiC-particles. This is because the higher the mass, the higher the kinetic energy of the particle with the same impact velocity; and the less the particles breakage, the less the loss of energy during the impact, and the higher the energy transfer as well. As a result, the surface roughness of the surfaces blasted with Al₂O₃-particles increases compared to those blasted with SiC-particles. The higher energy transfer also provokes a higher compressed cold-worked surface layer with higher residual stresses, which in turn may influence the corrosion behaviour, as explained later on. Finally, shape of the particles, i.e., the number of edges in a particle and/or the value of their angles, may influence the qualitative roughness aspect of the surface, which in turn may change quantitative roughness parameters [44, 45 and 46]. However, both Al₂O₃- and SiC-particles are sharp-edged with no significant differences. Each of these results can explain the higher R_a-values for Al₂O₃-blasted series compared with SiC-blasted series. However, the total differences in surface roughness must be explained as a combination of all the causes referred before.

In a different way, non-statistically significant differences in RSA between the c.p. Ti blasted with 600 
m particles and 900 
m particles in spite of the higher mean RSA-values for Al9 and Si9 comparing with Al6 and Si6, were not obtained (Table 2). This is a surprising result which cannot easily be explained.

An especially interesting result is that differences in %CS between Al₂O₃ and SiC-blasted materials have been found (Table 2). The selection of SiC as blasting material could be the right choice if it is intended to obtain a surface with no other materials than the protective TiO₂-layer. Furthermore, this result can be improved by means of changing the passivation acid for HF/HNO₃, where HF can react with SiC particles (SiC+4HF+2H₂O→SiF₄↑+4H₂↑+CO₂↑).

The E vs. t-results reveal that corrosion potentials do not depend on the material or the size of the shot particles, i.e., roughness does not influence the E_corr (Fig. 2). E_corr-values for passivated series (−275-−150 mV, approximately) and for Non-Pass samples (lower than −400 mV) are in good agreement with the values previously reported by other authors [47 and 48] working with similar electrolytes. Moreover, these values demonstrate that passivation with HCl-solution increases the corrosion potential of c.p. Ti to significantly more noble values due to changes in TiO₂-layer thickness and/or structure [47]. Variability between samples of the same series was observed with changes in the E_corr lower than 75 mV in all of the different surfaces. These scattered values can be caused by: (i) the chemical instability of HBSS, which is not a buffered solution; (ii) differences in the surface state in terms of the degree of internal defects caused by shot blasting; and (iii) thickness of the passive layer [47].

The equivalent circuit describing the Ti-HBSS interface has been obtained taking into consideration its simplicity, physical meaning and very good fitting (Fig. 3 and Fig. 4). It must be highlighted that a dense solid has been chosen to model the protective TiO₂-layer [49]. This model, consisting of a parallel R_f-CPE circuit, fits with excellent
²-values (Table 3). A CPE instead of a pure capacitor has been selected taking into account the significantly improved fitting. Not in vain CPE is a very useful element in order to describe deviations of the pure C-R response. Those deviations can be induced by physical, chemical and topographical heterogeneity on the studied surface of the metal [50].

Very high R_f-values (Table 3) for all the series have been determined (higher than 0.3 M
 cm²), which denotes that neither increasing roughness nor shot particles adhered on the c.p. Ti prevents the TiO₂-film from protecting titanium. The R_f-values are higher than in other works [49 and 51] probably due to the acidic passivation treatment. A relationship between surface roughness, i.e., real surface area, and R_f has been proven. The higher the RSA, the lower the R_f. This is probably because R_f is the total resistance of the film per geometric area of specimen and, as Solar et al. proposed [47], if the surface real area increases, the active anodic areas in the surface increase and, as a result, the total resistance of TiO₂-layer will decrease. The results of Al2-series do not follow the general tendency of the other series, not only with respect to the EIS-results (Table 3) but also with respect to the CPT-results ( Fig. 7 and Fig. 8), which will be discussed below. However, since only one specimen of Al2-series was tested, it is difficult to draw general conclusions about the causes of this particular behaviour.

(6K)

Fig. 7. I_{E vs. ESC=500 mV} vs. R_a for all the series. I_{E vs. ESC=500 mV} is the total current passing through the electrode when E vs. ESC equals 500 mV in the cyclic polarisation tests in the anodic scanning direction.

(6K)

Fig. 8. Real current densities (i_{E vs. ESC=500 mV}=I_{E vs. ESC=500 mV}/CRSA) vs. CRSA for all the series. I_{E vs. ESC=500 mV} is the same as in Fig. 7.

The surface roughness of the c.p. Ti or material composition of the shot particles do not alter the qualitatively electrochemical response of c.p. Ti (Fig. 3). This is probably because the imbedded ceramic particles are just insulating parts of the Ti surface and do not take part in the electrochemical interfacial processes.

Cyclic potentiodynamic curves reveal that the higher the size of the shot particles, the higher the total current (I) for a given potential, comparing series treated with the same shot-particle material, both Al₂O₃ (Fig. 5) and SiC ( Fig. 6). Again, the reason of this behaviour can be attributed to the increasing surface roughness when shot-particle size increases. In fact, as said before, an increasing surface roughness of the electrode implies an increased real surface area, which in turn result in an increased total current passing through the specimens. In Fig. 7 the total current when the potential equals 500 mV in the anodic scanning direction (I_{E vs. ESC=500 mV}) vs. R_a-values are represented. The relationship between surface roughness and current is clearly and surprisingly linear. The 500 mV has been selected because it is in a range of potentials where all the series are in a passivation state (Fig. 5 and Fig. 6). If this surprising linear relationship is considered, a current density of 0.15 
A/cm² is expected when the surface is ideally a plane (R_a=0 
m). However, further studies are necessary in order to determine the causes and meanings of this relationship.

If the only reason for the current increase with shot blasting was the increasing surface area, as suggested by Cai et al. [52], then the real current density, i_{E vs. ESC=500 mV}=I_{E vs. ESC=500 mV}/CRSA, would be constant. But, this is not the case as Fig. 8 shows. Again, the higher the CRSA, the higher the i_{E vs. ESC=500 mV}, comparing series treated with shot particles of the same material. i_{E vs. ESC=500 mV} has been calculated dividing I_{E vs. ESC=500 mV} by CRSA instead of RSA because the adhered shot particles are non-conducting materials, so the areas where they are imbedded do not provide current paths. The reason for the increasing of the real current density with roughness may be found in the fact that shot blasting induces surface compressive residual stresses in the metal. This is because impingement of the particles provokes plastic deformation on the surface, which in turn provokes surface roughness increase and a compressed cold-worked surface layer [45 and 53]. The compressed surface has a changed reactivity compared with the non-compressed surface because of the changes in the internal parameters of the crystal lattice. Furthermore, it is known that the size of the shot particles and other variables of shot-blasting treatment influence the magnitude of the induced residual stresses [45 and 54]. Therefore, the electrochemical behaviour of the c.p. Ti may be influenced, which in this case is revealed by changes in i_{E vs. ESC=500 mV}. There is another important result emerging from Fig. 8. It can be seen (by means of the connecting lines) that the mean i_{E vs. ESC=500 mV}-values for materials blasted with Al₂O₃-particles are higher than that fitted with i_{E vs. ESC=500 mV}-values for materials blasted with SiC-particles. Again, compressive surface residual stresses may explain that difference because residual stresses not only depend on the mean size and size distribution of the particles but also on their mechanical properties (hardness and fracture toughness) [45]. Therefore, surface residual stresses of the samples must be determined in a future work in order to elucidate their actual influence on the corrosion resistance and electrochemical behaviour of blasted c.p. Ti.

Qualitative differences in the polarisation behaviour of samples blasted with Al₂O₃-particles (Al2, Al6, Al9) and samples without shot blasting (Ctr) have not been found (Fig. 5). As reported in previous works [47, 48, 51, 52, 55, 56, 57 and 58], a first passive region in the potentials between 300 and 1500 mV with very low passive current is reached. Then a moderate increase of current is produced at approximately 1700 mV. Finally, a new passive region with low passive currents also develops up to more than 2500 mV, where the current densities increase again because of the dissociation of water. These results demonstrate that Al₂O₃-particles behave as electrical isolators all over the range of scanned potentials and they do not interfere in the corrosion resistance of the shot-blasted c.p. Ti.

The SiC-blasted materials (Si2, Si6, Si9) behave similar to Ctr- and Al₂O₃-blasted materials up to 1000 mV (Fig. 6). Therefore, they also reach the first passive region with low passive currents. However, the current passing through, the samples significantly increases above 1000 mV with a peak located at approximately 1800 mV. The currents reached in the peak are more than an order of magnitude higher than those registered for Ctr and Al series. Finally, they also show a decrease in currents to a new passive region. Such a high current increase cannot be ascribed only to a geometric effect related to an increase of the real surface area exposed to the solution after shot blasting. Taking into consideration that the main difference between the Al- and the Si-series is the presence of different shot-particle materials adhered on the c.p. Ti, the cause of such significant differences may be the oxidation of the adhered SiC. Although this hypothesis had not been confirmed this should not be a crucial problem for our application since the highest potentials recorded for the oral cavity are only of the order of 200 mV vs. SCE [59]. Moreover, the equilibrium potential for the oxygen reduction reaction at a biological pH is 575 mV vs. ESC, so the shot-blasted c.p. Ti under non-polarised conditions can be indefinitely passive in electrolytes similar to biological fluids.

Values of the first passive current densities (Fig. 5, Fig. 6 and Fig. 7) are not higher than 4 
A/cm², whatever the series. Those values are similar to values previously found in polished c.p. Ti [47, 48, 51, 52 and 57]. Moreover, they demonstrate the excellent corrosion resistance of all the studied series, especially because, as said before, this passive region spans up to potentials as high as 1000 mV. All the materials show high repassivation kinetics, which is revealed by the continous high slopes of the curves when the potential scan is just reversed to the cathodic direction (Fig. 5 and Fig. 6).

7. 5. Conclusions

The increased surface area of the material because of the increasing surface roughness, when shot blasting is performed on the c.p. Ti, is not the only cause for the differences found in the electrochemical behaviour and corrosion resistance of the SiC- and Al₂O₃-blasted c.p. Ti. Among other possible causes, those differences may be attributed to the compressive residual surface stresses induced by shot blasting.

All the materials tested, i.e., the c.p. Ti blasted with Al₂O₃- or SiC-particles of 200, 600 or 900 
m in mean size, have an adequate corrosion and electrochemical behaviour in terms of its possible use as dental implant material. This is despite the significantly increased currents found in cyclic polarisation tests for the SiC-blasted c.p. Ti compared with the Al₂O₃-blasted c.p. Ti.

8. Acknowledgements

The authors would like to acknowledge Dr. Susanna Castel from the Scientific and Technical Services (Confocal Microscopy and Cellular Micromanipulation Facility) of the Universitat de Barcelona for her help in the confocal microscopy study; Dr. Carlos Sa from the Centro de Materiais da Universidade do Porto for his help in the scanning electron miscrocopy and image analysis study; Klockner Dental Implants, S.L., and Materias Primas Abrasivas, S.L. (MPA) for their technical help; and the Comisión Interministerial de Ciencia y Tecnología (Grant No 2000-1736) for its financial support.

9. References

1. P.I. Brĺnemark, B.O. Hansson, R. Adell, U. Breine, J. Lindström, O. Hallen and A. Öhman , Osseointegrated implants in the treatment of edentulous jaw experience from a 10-year period. Scand J Reconstr Surg 16 Suppl. (1977), pp. 1-132.

2. J.E. Lemons , Application of materials in medicine and dentistry. Dental implants. In: B.D. Ratner, A.S. Hoffman, F.J. Schoen and J.E. Lemons, Editors, Biomaterials science: an introduction to materials in medicine, Academic Press, San Diego (1996), pp. 308-318.

3. M. Pourbaix , Electrochemical corrosion of metallic biomaterials. Biomaterials 5 (1984), pp. 122-134. Abstract

4. D.I. Bardos , Titanium and titanium alloys. In: D. William, Editor, Encyclopaedia of medical and dental materials, Pergamon Press, Oxford (1990), pp. 360-365.

5. Boyer R, Welsh G, Colling EW, editors. Materials properties handbook: titanium alloys. Ohio: ASM International 1994. p. 181-219.

6. B. Kasemo and J. Lausmaa , Surface science aspects on inorganic biomaterials. CRC Crit Rev biocompatibilty 4 (1986), pp. 335-380. Abstract-EMBASE  

7. P. Tengvall and I. Lundtröm , Physico-chemical considerations of titanium as a biomaterial. Biomaterials 9 (1992), pp. 115-134. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

8. T. Albrektsson, P.I. Brĺnemark, B.O. Hansson and J. Lindström , Osseointegrated titanium implants. Acta Orthop Scand 52 (1981), pp. 155-170. Abstract-MEDLINE | Abstract-EMBASE  

9. K.A. Thomas and S. Cook , An evaluation of variables influencing implant fixation by direct bone apposition. J Biomed Mater Res 19 (1985), pp. 875-901. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

10. Wennerberg A. On surface roughness and implant incorporation. Ph.D. thesis, Göteborg: University of Göteborg, 1996.

11. K.T. Bowers, J.C. Keller, B.A. Randolph, D.G. Wick and C.M. Michaels , Optimization of surface micromorphology from enhanced osteoblast response in vitro. Int J Oral Maxillofac Implants 7 (1992), pp. 302-310. Abstract-MEDLINE  

12. B. Groessner-Schereider and R.S. Tuan , Enhanced extracellular matrix production and mineralization by osteoblasts cultured on titanium surfaces in vitro. J Cell Sci 101 (1992), pp. 209-217.

13. Z. Schwartz and B.D. Boyan , Underlaying mechanisms at the bone biomaterial interface. J Cell Biochem 56 (1994), pp. 340-347. Abstract-MEDLINE | Abstract-EMBASE  

14. B.D. Boyan, T.W. Hummert, K. Kieswetter, D. Schraub, D.D. Dean and Z. Schwartz , Effect of titanium surface characteristics on chondrocytes and osteoblasts in vitro. Scanning Electron Microsc (Cells and Materials) 5 (1995), pp. 323-335.

15. J.Y. Martin, Z. Schwartz, T.W. Hummert, D.M. Schraub, J. Simpson, J. Lankford, D.D. Dean, D.L. Cochran and B.D. Boyan , Effect of titanium surface roughness on proliferation, differentiation and protein synthesis of human osteoblast-like cells (MG63). J Biomed Mater Res 29 (1995), pp. 389-401. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

16. J.L. Ong, C.W. Prince, G.N. Raikar and L.C. Lucas , Effect of surface topography of titanium on surface chemistry and cellular response. Implant Dent 5 (1996), pp. 83-88. Abstract-MEDLINE  

17. D.D. Deligianni, N. Katsal, S. Ladas, D. Sotiropoulou, J. Amedee and Y.F. Missirlis , Effect of surface roughness of the titanium alloy Ti-6Al-4V on human bone marrow cell response and on protein adsorption. Biomaterials 22 (2001), pp. 1241-1251. SummaryPlus | Full Text + Links | PDF (427 K)

18. Z. Schwartz, C.H. Lohmann, M. Sisk, D.L. Cochran, V.L. Sylvia, J. Simpson, D.D. Dean and B.D. Boyan , Local factor production by MG63 ostoblast-like cells in response to surface roughness and 1,25-(OH)₂D₃ is mediated via protein kinase C- and protein kinase A-dependent pathways. Biomaterials 22 (2001), pp. 731-741. SummaryPlus | Full Text + Links | PDF (463 K)

19. P. Predecki, J.E. Stephan, B.A. Auslander, V.L. Mooney and K. Kirkland , Kinetics of bone growth into cylindrical channels in aluminium oxide and titanium. J Biomed Mater Res 6 (1972), pp. 375-400. Abstract-Compendex | Abstract-MEDLINE  

20. D. Buser, R.K. Schenk, S. Steinemann, J.P. Fiorelini, C.H. Fox and H. Stich , Influence of surface characteristics on bone integration of titanium implants. A histomorphometric study in miniature pigs. J Biomed Mater Res 25 (1991), pp. 889-902. Abstract-EMBASE | Abstract-MEDLINE  

21. I. Ericsson, C. Johansson, H. Bystedt and M.R. Norton , A histomorphometric evaluation of bone-to-implant contact on machined-prepared roughened titanium dental implants. Clin Oral Implant Res 5 (1994), pp. 202-206. Abstract-MEDLINE   | Full Text via CrossRef

22. J.E. Feighan, V.M. Goldberg, D. Davy, J.A. Parr and S. Stevenson , The influence of surface-blasting on the incorporation of titanium alloy implants in a rabit intramedullary model. J Bone Jt Surg Am 77 (1995), pp. 1380-1395. Abstract-MEDLINE | Abstract-EMBASE  

23. A. Wennerberg, T. Albrektsson and B. Andersson , An animal study of c.p. titanium screws with different surface topographies. J Mater Sci: Mater Med 6 (1995), pp. 302-309. Abstract-EMBASE  

24. M. Wong, J. Eulenberger, R. Schenk and E. Hunziker , Effect of surface topology on the osseointegration of implant materials in trabecular bone. J Biomed Mater Res 29 (1995), pp. 1567-1575. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE  

25. A. Wennerberg, T. Albrektsson, C. Johansson and B. Andersson , Experimental study of turned and grit-blasted screw-shaped implants with special emphasis on effects of blasting material and surface topography. Biomaterials 17 (1996), pp. 15-22. SummaryPlus | Full Text + Links | PDF (1326 K)

26. Cochran DL, Schenk RK, Lussi A, Higginbottom FL, Buser D. Bone response to unloaded and loaded titanium implants with a sandblasted and acid etched surface: a histometric study in the canine mandible. J Biomed Mater Res 1998;40:1-11.

27. A. Piatelli, L. Manzon, A. Scarano, M. Paolantonio and M. Piatelli , Histologic and histomorphometric analysis of the bone response to machined and sandblasted titanium implants: an experimental study in rabbits. Int J Oral Maxillofac Implants 13 (1998), pp. 805-810.

28. M. Svehla, P. Morberg, B. Zicat, W. Bruce, D. Sonnabend and W.R. Walsh , Morphometric and mechanical evaluation of titanium implant integration: comparison of five surface structures. J Biomed Mater Res 51 (2000), pp. 15-22. Abstract-MEDLINE | Abstract-Compendex   | Full Text via CrossRef

29. K. Gotfredsen, L. Nimb, E. Hjorting-Hansen, J.S. Jensen and A. Holmen , Histomorphometric and removal torque analysis for TiO₂-blasted titanium implants. Clin Oral Implant Res 5 (1994), pp. 202-206.

30. F. Jones , Teeth and bones: applications of surface science to dental materials and related biomaterials. Surf Sci Rep 42 (2001), pp. 75-205. SummaryPlus | Full Text + Links | PDF (1733 K)

31. A. Naji and M.F. Harnad , Cytocompatibility of two coating materials, amorphous alumina and silicon carbide, using human differentiated cell cultures. Biomaterials 12 (1991), pp. 690-694. Abstract

32. B.W. Darvell, N. Samman, W.K. Luk, K.F. Clark and H. Tideman , Contamination of titanium casting by aluminium oxide blasting. J Dent 23 (1995), pp. 319-322. SummaryPlus | Full Text + Links | PDF (1146 K)

33. H.E. Placko, S. Mishra, J.J. Weimer and L.C. Lucas , Surface Characterization of titanium-based implant materials. Int J Oral Maxillofac Implants 15 (2000), pp. 355-363. Abstract-MEDLINE  

34. G. Orsini, B. Assenza, A. Scarano, M. Piatelli and A. Piatelli , Surface analysis of machined versus sandblasted and acid-etched titanium implants. Int J Oral Maxillofac Implants 15 (2000), pp. 779-784. Abstract-MEDLINE  

35. U. Gross and V. Strunz , The interface of various glasses and glass ceramics with a bony implantation bed. J Biomed Mater Res 19 (1985), pp. 251-271. Abstract-EMBASE | Abstract-MEDLINE  

36. Aparicio C, Gil FJ, Peraire C, Padrós A, Planell JA. Effect of size and material of the shot peening particles on adhesion and differentiation of osteoblast-like cell cultures on commercially pure titanium. North Sea Biomaterials 1998 Proceedings, The Hague, 1998. p. 137.

37. M.G. Fontana and N.D. Green Corrosion engineering: Mc Graw Hill series in materials science and engineering, McGraw-Hill, USA (1978).

38. ASTM G31-90. Standard practice for laboratory immersion corrosion testing metals. In: 1992 Annual Book of ASTM Standards, vol. 03.02. Philadelphia: ASTM, 1992. p. 102-9.

39. B.A. Boukamp Equivalent circuit (EQUIVCRTPAS). User's manual, University of Twente, Enschede (1989).

40. ASTM G61-86. Standard test method for conducting cyclic potentiodynamic polarization measurements for localized corrosion susceptibility of iron-nickel, or cobalt-based alloys. In: 1992 Annual Book of ASTM Standards, vol. 03.02. Philadelphia: ASTM, 1992, p. 231-5.

41. FEPA 42-F-1984. Norme FEPA des grains de corindon electrofondu et de carbure de silicium pour abrasifs agglomeres. Paris: Fedération Européenne des Fabricants de Produits Abrasifs, 1984.

42. P. Bohn , Qualitaty control of shot media by sieve analysis. Shot Peener 13 (1999), pp. 6-9.

43. Mehelich CS, Van Kuiken L, Woelfel MM. Shot Peening. In: ASM metals handbook. vol. 5: surface engineering. USA: ASM International, 1995. p. 138-49.

44. G. Udupa, M. Singaperumal, R.S. Sirohi and M.P. Kothiyal , Characterization of surface topography by confocal microscopy: II. The micro and macro surface irregularities. Meas Sci Technol 11 (2000), pp. 315-329. Abstract-Compendex | Abstract-INSPEC   | Full Text via CrossRef

45. R.D. Arnell, P.B. Davie, J. Halling and T.L. Whomes Tribology: principles and design applications, Mac Millan, London (1991).

46. G. Udupa, M. Singaperumal, R.S. Sirohi and M.P. Kothiyal , Characterization of surface topography by confocal microscopy: I. Principles and the measurement system. Meas Sci Technol 11 (2000), pp. 305-314. Abstract-Compendex | Abstract-INSPEC   | Full Text via CrossRef

47. R.J. Solar, S.R. Pollack and E. Korostoff , In vitro corrosion testing of titanium surgical implant alloys: an approach to understanding titanium release from implants. J Biomed Mater Res 13 (1979), pp. 217-250. Abstract-EMBASE | Abstract-Compendex | Abstract-MEDLINE  

48. T.P. Hoar and D.C. Mears , Corrosion resistant alloys in chloride solutions, materials for surgical implants. Proc Royal Soc London Ser A 294 (1966), pp. 486-510.

49. J.E.G. Gonzalez and J.C. Mirza-Rosca , Study of the corrosion behaviour of titanium and some of its alloys for biomedical and dental implant applications. J Electroanal Chem 471 (1999), pp. 109-115. SummaryPlus | Full Text + Links | PDF (213 K)

50. J.R. McDonald Impedance spectroscopy, Wiley, New York (1987).

51. M. Aziz-Kerrzo, K.G. Conroy, A.M. Fenelson, S.T. Farrell and C.B. Breslin , Electrochemical studies on the stability and corrosion resistance of titanium-based implant materials. Biomaterials 22 (2001), pp. 1531-1539. SummaryPlus | Full Text + Links | PDF (240 K)

52. Z. Cai, H. Nakajima, M. Woldu, A. Berglund, M. Bergman and T. Okabe , In vitro corrosion resistance of titanium made using different fabrication methods. Biomaterials 20 (1999), pp. 183-190. SummaryPlus | Full Text + Links | PDF (153 K)

53. M. Kobayashi, T. Matsui and Y. Murakami , Mechanism of creation of compressive residual stresses by shot peening. Int J Fatigue 20 (1998), pp. 351-357. Abstract | PDF (540 K)

54. P.S. Prevey , X-ray diffraction characterization of residual stresses produced by shot peening. Shot Peener 15 (2001), pp. 4-8.

55. H.J. Mueller and E.H. Greener , Polarization studies of surgical materials in Ringer's solution. J Biomed Mater Res 4 (1970), pp. 29-41. Abstract-MEDLINE  

56. K.M. Speck and A.C. Fraker , Anodic polarization behavior of Ti-Ni and Ti-6Al-4V in simulated physiological solution. J Dent Res 59 (1980), pp. 1590-1595. Abstract-MEDLINE  

57. S.R. Sousa and M.A. Barbosa , Corrosion resistance of titanium cp in saline physiological solutions with calcium phosphate and proteins. Biomaterials 14 (1993), pp. 287-294. Abstract-EMBASE | Abstract-Compendex  

58. M.A. Khan, R.L. Williams and D.F. Williams , In-vitro corrosion and wear of titanium alloys in the biological environment. Biomaterials 17 (1996), pp. 2117-2126. SummaryPlus | Full Text + Links | PDF (1038 K)

59. G.J. Ewers and E.H. Greener , The electrochemical activity of the oral cavity—a new approach. J Oral Rehab 12 (1985), pp. 469-476. Abstract-EMBASE | Abstract-MEDLINE  

Corresponding author. Fax: [phone]; email: [e-mail]

---