Corrosion behavior and surface characterization of titanium in solution containing fluoride and albumin

Shinji Takemoto^, , Masayuki Hattori, Masao Yoshinari, Eiji Kawada and Yutaka Oda

Department of Dental Materials Science, Tokyo Dental College, 1-2-2 Masago, Mihama-ku, Chiba 261-8502, Japan

Received 22 February 2004;  accepted 18 March 2004.  Available online 19 May 2004.
Biomaterials
Article in Press, Corrected Proof - Note to users

1. Abstract

The objective of this study was to demonstrate the role of albumin on the corrosion behavior of titanium in a solution containing 2.0 g/l fluoride and either 0.1 or 1.0 g/l albumin. The corrosion behavior and surface characterization of passive films on titanium immersed in such a solution were examined. In addition, the change in pH and the concentration of dissolved titanium in the solution were examined. The results showed that the corrosion of titanium in a solution containing fluoride was distinct, and that adding albumin to the solution containing fluoride suppressed corrosion. Fluorine was detected on the titanium surface immersed in the solution containing fluoride, and dissolution of the titanium was confirmed. The titanium immersed in a solution containing both fluoride and albumin had an albumin film regardless of the albumin concentration. In addition, the amount of dissolved titanium from the titanium immersed in the solution was less than when the solution contained no albumin. It was suggested that the formation of adsorbed albumin films on or in the passive film acted to not only protect the titanium from attack by the fluoride but also suppressed dissolution of the titanium-fluoride compounds.

Author Keywords: Titanium; Albumin; Fluoride; Corrosion; X-ray photoelectron spectroscopy

2. Article Outline

1. Introduction

2. Experimental procedure

2.1. Materials

2.2. Test solution

2.3. Electrochemical measurement

2.4. Surface characterization and analysis of dissolved titanium

2.5. Statistical analyses

3. Results

3.1. Electrochemical measurement

3.2. Surface characterization

3.3. Concentration of dissolved titanium and pH

4. Discussion

4.1. Solution containing fluoride

4.2. Solution containing fluoride and albumin

4.3. Albumin concentration

5. Conclusions

Acknowledgements

References

3. 1. Introduction

Titanium and its alloys are widely used in dentistry as prosthetic appliances because of a high corrosion resistance and good biocompatibility. These valuable properties are caused by passive films that are rapidly formed in the body fluid environment [1, 2 and 3]. Recently, mouth-rinses, toothpastes, and prophylactic agents containing fluoride are utilizing to prevent the dental caries. However, decreasing the corrosion resistance of titanium in solutions containing fluoride has been reported [4, 5, 6, 7, 8, 9, 10, 11 and 12]. The corrosion behaviors are related to the concentration of fluoride and the surrounding environment (e.g., pH, concentration of dissolved oxygen, and temperature). Therefore, the surface reaction on titanium under the existence of fluoride is important to understand the corrosion and tarnish of titanium in oral environment.

The influence of protein on the corrosion resistance of the implant materials is important because saliva contains 200-500 mg/dl of salivary protein. Williams et al. [13] reported that the corrosion of titanium in serum was enhanced. Khan et al. [14] suggested that the corrosion behavior of titanium alloys in protein solutions was dependent on the kind of alloy. In addition, they indicated that the alloys immersed in the protein solutions reduced the hardness of the surface oxides. However, how proteins are related to surface oxides has not been examined.

Recently, Ide et al. [15 and 16] reported that the corrosion of titanium was suppressed in solutions containing fluoride and albumin. They suggested a possible mechanism for suppression of the corrosion of the titanium. The albumin adsorption on the titanium prevented fluoride attack, or the buffering effect of the albumin increased the pH. Furthermore, Huang [17] also indicated that Ti-6Al-4 V alloys in artificial saliva containing fluoride and albumin had higher corrosion resistance than they did in saliva without albumin. He suggested that albumin adsorption could prevent destruction of the passive films by the fluoride ion. However, the mechanism of suppressing corrosion of titanium and its alloys in the presence of albumin is not clear, that is, the corrosion products and the composition of the surface oxide caused by albumin adsorption have not been examined in detail.

The present study was to examine the surface characterization of titanium and the dissolved titanium concentration after the titanium was immersed in solution containing fluoride and albumin, and to determine the effects of the addition of albumin on the corrosion behavior of the titanium. These results will enhance the understanding of the surface reaction of titanium prosthetic appliance in oral environment.

4. 2. Experimental procedure

2.1. Materials

Titanium specimens 14×14×1 mm in size or 4 mm in diameter and 1 mm in thickness were cut from a sheet of commercially available pure titanium (Ti grade 2, Kobe Steel, Japan). The former were prepared as specimens for the corrosion tests and surface characterization; the latter were prepared for X-ray photoelectron spectroscopy (XPS) measurement. The specimens were polished with silicon carbide papers No. 120, 240, 400, and 600 grits in series. They were ultrasonically washed in acetone and ethanol for 10 min.

2.2. Test solution

The solution was composed of sodium chloride (NaCl, Wako Chem., Osaka, Japan), sodium fluoride (NaF, Wako Chem., Osaka, Japan), bovine serum albumin (BSA, Wako Chem., Osaka, Japan), and distilled water. The starting composition in 1000 ml of each solution and the specimen codes are listed in Table 1. All solutions containing fluoride (NAF, BSA.01, and BSA.1) were adjusted to pH 5.0 using lactic acid (Wako Chem., Osaka, Japan) at 37°C, while SAL solution was not adjusted.

Table 1. Code, composition and pH of 1000 ml preparation solutions at 37°C

2.3. Electrochemical measurement

The apparatus for electrochemical measurement consisted of a potentiostat/galvanostat (Model 273A, EG&G, USA) controlled by a computer using research corrosion software M352C (EG&G, USA), a saturated calomel electrode (SCE) as reference electrode, a platinum plate as counter electrode, and a sample holder (the exposed area of the specimen was 1.0 cm²) as working electrode. A volume of 500 ml of test solution was deaerated with pure nitrogen gas for 30 min before measurement. The open circuit potential (E_open) was measured after the specimen was immersed in the solution for 30 min. The polarization resistance (R_p) of the specimen in the test solution was measured in a scanning range of −20 to 20 mV from the open circuit potential at a scanning rate of 0.1 mV/s. The potentiodynamic polarization behavior of the specimens was recorded in a scanning range of −1.2 to +2.5 V (vs. SCE) at a scanning rate of 0.33 mV/s. The passive current density at 0.5 V (I_0.5) was obtained from the potentiodynamic polarization curve. Each measurement was maintained at 37°C. Five specimens were tested for each condition.

2.4. Surface characterization and analysis of dissolved titanium

Some of the specimens were immersed in 10 ml per unit area (cm²) of the test solutions at 37°C. The specimens were picked up from the solutions after immersion for 3 days and gently washed with distilled water. Their surface structure was examined by a thin-film X-ray diffractometer (TF-XRD), an electron probe microanalyzer (EPMA), and XPS. The TF-XRD profiles were taken with a Rint 2000 (Rigaku, Japan) equipped with a Cu K
radiation (40 kV and 200 mA) source and with a thin film attachment; the incidence angle was 1°. EPMA measurements were performed with an X-3010 (Hitachi, Japan). The profiles of each element were obtained with voltage 15 kV and current 100 nA. All XPS spectra were measured with an ESCA 750 (Shimadzu, Japan) and obtained with a Mg K
: energy 1253.6 eV, voltage 8 kV, and current 30 mA. The reflectance angle for photoelectron detection was 0°. The binding energy was normalized to the C 1 s peak (285.0 eV) of hydrocarbon on the titanium specimen. The surface texture of the specimens was observed under a scanning electron microscope (SEM; JSM-6340F, JEOL, Japan). The titanium ion concentration and pH in the test solution after immersion of specimen were monitored with an inductively coupled plasma emission spectrometer (ICP; Vista-MPX, SII, Japan) and pH meter (BASIC, Denver Instrument, USA), respectively. Three specimens were prepared for each condition to examine the analysis of dissolved titanium.

2.5. Statistical analyses

The corrosion parameters, E_open, R_p, and I_0.5, the concentration of dissolved titanium, and pH were statistically analyzed using one-way analysis of variance (ANOVA) at a significance level of 5%, then compared among specimens by Scheffe's test at a significance level of 5%.

5. 3. Results

3.1. Electrochemical measurement

Table 2 presents the values of E_open after a 30-min immersion, R_p calculated from linear polarization curve, and I_0.5 obtained from the potentiodynamic polarization profiles (Fig. 1). The E_open values in the SAL and the NAF were approximately −0.4 and −0.7 V, respectively, and the E_open value decreased in the solutions containing the fluoride. The E_open values in the BSA.01 and the BSA.1 increased slightly in comparison with that for the NAF, but the values were smaller than that for the SAL. No significant difference in the E_open values between the BSA.01 and the BSA.1 was apparent. The corrosion resistance, R_p, abruptly decreased from 300 to 12 k
 cm² in the solution containing fluoride. The values among the NAF, the BSA.01, and the BSA.1 were not significantly different. With the addition of sodium fluoride to the saline, the values of I_0.5 increased from 1.7×10⁻⁶ A/cm² for the SAL to 89.9×10⁻⁶ A/cm² for the NAF. The values for the BSA.01 and the BSA.1 were 46.8×10⁻⁶ A/cm² and 38.2×10⁻⁶ A/cm², respectively, and were significantly smaller than that for the NAF. The difference in the additive amount of BSA was not statistically significant.

Table 2. Open circuit potential (E_open) after 30 min immersion, polarization resistance (R_p) obtained from linear polarization test, and passive current density at +0.50 V (I_0.5) obtained from the potentiodynamic polarization curves shown in Fig. 1, for titanium specimens in the various solutions

Average ± standard deviation (n=5), Different letters (a, b, c) indicate statistical differences within each group based on ANOVA and Scheffe's test at p<0.05.

(10K)

Fig. 1. Typical potentiodynamic polarization profiles for titanium specimen in the various solutions.

3.2. Surface characterization

Fig. 2 shows the TF-XRD patterns for the titanium specimens immersed in the various solutions for 3 days. All specimens consisted only of
-Ti as the diffractions were only 35.1°, 38.4°, 40.2°, 53.0°, 62.9°, 70.7°, 76.2°, and 77.4° in 2
(JCPDS card: 44-1294).

(10K)

Fig. 2. TF-XRD patterns for the titanium specimens after immersion in the various solutions for 3 days at 37°C.

Table 3 lists the elements detected on the titanium specimens after immersion in various solutions for 3 days using the XPS and the EPMA analyses. XPS detected carbon, oxygen, nitrogen, and titanium on all specimens. Apart from C, O, N, and Ti, a few sodium and chloride were detected on the SAL and the NAF, indicating that sodium chloride was deposited on the specimens. EPMA detected fluorine on the NAF, the BSA.01, and the BSA.1 after immersion in the solution containing fluoride. XPS spectra of the binding energy regions of (a) O 1s, (b) Ti 2p, (c) N 1s (d) C 1s, and (e) F 1s on the titanium specimens immersed in various solutions for 3 days are shown in Fig. 3. Spectrum "Polished" for the titanium polished with silicon carbide papers up to 600 grids is indicated as the reference spectrum. In Fig. 3(a), the shoulders in the SAL and the NAF located on the higher energy side were smaller than for that of the Polished. In contrast, the core-level of the binding energy for the BSA.01 and the BSA.1 shifted to the higher energy side and the energy at 531.8 eV increased. The XPS spectrum of the O 1 s electron binding energy region consists of at least three peaks originating from the metal oxide, O²⁻, hydroxide or hydroxyl groups, OH⁻, and hydrate and/or adsorbed water, H₂O, and their peaks appear close to 530.5, 532.0, and 533.5 eV, respectively [18]. Proportions of these oxygen species and relative concentration ratio of OH⁻ to O²⁻, [OH⁻]/[O²⁻], are listed in Table 4. The [OH⁻]/[O²⁻] ratios of the SAL and the NAF were smaller than that of the Polished. The results in increased O²⁻ and decreased OH⁻ and H₂O suggested that the SAL and the NAF were oxidized by immersion in each solution after polishing. The [OH⁻]/[O²⁻] ratios of the BSA.01 and the BSA.1 were higher than that of the Polished, indicating that the BSA in solution adsorbed on the specimen.

Table 3. Elements detected on the titanium specimen immersed in various solutions by using XPS and EPMA

Tr: Trace element; polished indicates before immersion in the various solutions.

(34K)

Fig. 3. XPS spectra obtained from the titanium specimens immersed in the various solutions for 3 days at 37°C. "Polished" indicated before immersion in the various solutions. (a) O 1s region, (b) Ti 2p region, (c) N 1s region, (d) C 1s region, (e) F 1s region.

Table 4. The ratio of total amount of [OH⁻]/[O²⁻] of the titanium specimen immersed in various solutions

Shown in average (n=2).

In the XPS spectra of the Ti 2p electron binding energy region shown in Fig. 3(b), the spectra of the Polished, the SAL, and the NAF appeared in the spectra originating from both the metallic (Ti⁰) and oxide (Ti²⁺, Ti³⁺, and Ti⁴⁺) states [19]. The peak originating from the metallic state of the SAL and the NAF is smaller than that of the Polished, while the shoulder at 456-458 eV originating from the Ti²⁺ and Ti³⁺ oxidation states is similar. Then, the oxidation state of the SAL and the NAF mainly existed in Ti⁴⁺ The intensity of the N 1s XPS spectra for the BSA.01 and the BSA.1 shown in Fig. 3(c) was larger than that for the other specimens. In the C 1s XPS spectra for the BSA.01 and the BSA.1 shown in Fig. 3(d) new peaks appeared at close to 288.3 eV. These N 1s and C 1s peaks originated from peptide bonding (CO-NH), carboxyl groups (−COOH), and amino groups (−NH₂) of the BSA. Fig. 4 shows the ratio in the proportion of the intensity for carbon at 282-291 eV to that for titanium at 452-467 eV (I_C/I_Ti). The BSA.01 and the BSA.1 had larger I_C/I_Ti values than the SAL and the NAF. The I_C/I_Ti values for the BSA.1 were larger than that for the BSA.01. The F 1 s XPS spectra in Fig. 3(f) indicated that the NAF had larger amount of fluorine, while the little fluorine was detected on the BSA.01. The fluorine on the BSA.1, however, was not detected by XPS.

(7K)

Fig. 4. Ratio of the intensity for C 1s XPS spectra to that for Ti 2p XPS spectra, I_C/I_Ti, for the titanium specimens after immersion in the various solutions for 3 days at 37°C. "Polished" indicated before immersion in the various solutions.

Surface textures of the titanium specimens immersed in various solutions for 3 days were observed, as shown in Fig. 5. A rough surface, which included localized corrosion, was observed on the NAF, while only scratches made by polishing were observed on the SAL (Figs. 5(a) and (b)). The BSA.01 and BSA.1 had smoother surfaces than the NAF, but rougher surfaces than the SAL.

(62K)

Fig. 5. SEM images of the titanium specimens immersed in various solutions for 3 days at 37°C. A rough surface was observed on the NAF, while the BSA.01 and BSA.1 had smoother surfaces than the NAF.

3.3. Concentration of dissolved titanium and pH

The concentration of dissolved titanium and the pH value for the various solutions when the titanium specimens were immersed for 3 days are shown in Fig. 6 and Fig. 7. No dissolved titanium in the saline was detected, and the pH was almost the same as that for the saline without the specimens. In the solution containing NaF, the dissolution of titanium was detected. The dissolved amount of titanium in the solution containing both fluoride and BSA was clearly lower than that in the solution containing only fluoride. No significant difference in the amount of dissolved titanium between the BSA.01 and the BSA.1 was detected. The pH values increased as the amount of dissolved titanium increased, and the pH values of the solution containing fluoride increased by about 0.5 (pH).

(11K)

Fig. 6. The concentration of dissolved titanium in the various solutions after immersion for 3 days at 37°C. Different letters (a, b, c) indicate statistical differences between groups based on ANOVA and Scheffe's test at p<0.05.

(10K)

Fig. 7. The pH value in the various solutions after immersion for 3 days at 37°C. "Blank" indicate the solutions without specimen maintained. Different letters (a, b) indicate statistical differences between groups based on ANOVA and Scheffe's test at p<0.05.

6. 4. Discussion

Corrosion behaviors on titanium and its alloys in solutions containing fluoride or protein have been reported. From the perspective of electrochemical evaluation, fluoride corroded titanium [5, 6, 7, 8, 9, 10, 11, 12, 15, 16 and 17] and proteins influenced the corrosion of titanium [13, 14, 15, 16 and 17]. This study is useful for understanding the reactions on titanium surfaces in solutions containing fluoride with/without protein. The relationship between corrosion behavior and surface reaction in saline solutions containing fluoride and protein is discussed below.

4.1. Solution containing fluoride

The results of EPMA analyses detected fluorine on the surface of titanium immersed in the solution containing fluoride, while the detection of sodium by XPS was low. Therefore, fluoride-titanium or fluoride-hydrogen compounds were deposited on the titanium surface, rather than sodium fluoride. This result supports the suggestion of Wilhelmsen et al. [5] and Nakagawa et al. [9]: Hydrofluoric acid was attributed to destroy the passive film on titanium but not the fluoride ion. At first, NaF is decomposed into sodium ion and fluoride ion in the solution. The fluoride ion becomes hydrofluoric acid partially depending on pH of the solution. The hydrofluoric acid attacks the passive films on titanium surface. Judging from the increase in pH value after immersion of titanium in the solution containing fluoride, it might be that the hydrofluoric acid decomposes to form fluoride ions and protons or water molecules on the titanium surface by following formulas:

Ti2O3+6HF→2TiF3+3H2O,TiO2+4HF→TiF4+2H2O,TiO2+2HF→TiOF2+H2O.

Titanium-fluoride compounds are formed because the fluoride ions bound to the titanium or titanium oxide degrade in the solution. Protons are absorbed by the titanium substrate or dissolve into the solution as water molecules since it is reported that the titanium alloys absorb hydrogen [20].

In this study, the regeneration of the passive films in the solution containing fluoride was confirmed by the Ti 2p and O 1s XPS spectra, while the passive films on the titanium surface were dissolved by the above reactions. In addition, since TF-XRD detected no crystallite except for
-Ti on the titanium specimens, the passive films had amorphous, or low- and too thin-crystallite structures. Therefore, both dissolution and regeneration of the thin passive films on the titanium surface have occurred in the solution containing fluoride.

4.2. Solution containing fluoride and albumin

Because saliva contains 200-500 mg/dl of salivary protein, it is important to investigate the resistance of titanium to corrosion under circumstances in which the fluid environment contains protein. The presence of protein leads to the release of metal ions [21 and 22] and to the reduction of the mechanical properties of metals [20], indicating that protein adsorption influences resistance to corrosion. Therefore, investigation of the corrosion behavior and the surface reaction of titanium under circumstances in which the fluid environment contains fluoride and protein is required to understand the characteristics of prosthesis made of titanium.

Ide et al. [15 and 16] indicated that the passive current density at +0.30 V of titanium in solutions containing both fluoride and albumin was less than that in solutions containing only fluoride. That is, the addition of albumin suppressed the corrosion of the titanium caused by the fluoride. Huang [17] reported that the presence of albumin in acid media containing fluoride improved the corrosion resistance of Ti-6Al-4 V. One of the reasons for this was that the adsorption of albumin via the calcium ion binding prevented destruction of the passive films. Passive films containing calcium phosphate formed on titanium in artificial saliva, Hanks' solution [1, 2 and 3]. Therefore, the formation of passive films containing calcium and phosphorus should be recognized when considering the corrosion of titanium. Since the test solutions in this study did not contain calcium and phosphorus, their influence on passive films was not taken into consideration.

The corrosion behaviors of titanium in a solution containing fluoride and albumin via the electrochemical measurements agreed with Ides' and Huang's results. We suggest the following mechanism for the suppression of the corrosion of titanium in solutions containing fluoride and albumin. Albumin adsorbs on the titanium surface and forms thin film by itself. Hydrofluoric acid generated in the solutions containing the fluoride can attack the titanium surface before and after formation of the albumin film. Titanium-fluoride compounds then form on the titanium, and are dissolved. However, dissolution of the compounds could not occur by the adsorbed albumin film on the titanium easily. Evidence of this suggested mechanism could be seen in the passive current density, the detection of fluorine, the degradation of the titanium, and the pH value. First, the I_0.5 values for the BSA.01 and the BSA.1 were lower than for the NAF, indicating that the corrosion resistance of titanium in a solution containing fluoride and albumin is larger than that in a saline containing fluoride. Therefore, the albumin in a solution containing fluoride prevented the corrosion of titanium. Second, XPS analyses detected few fluorine on the BSA.01 and the BSA.1, while EPMA analyses detected a considerable amount of fluorine on the BSA.01 and the BSA.1 as well as on the NAF. Considering the limitation on detection depth for XPS and EPMA analyses, which are about 10 nm and 1 
m, respectively, the fluoride could not exist on a albumin film but could exist in or on a passive film (between the passive film and the albumin film). Finally, when the titanium specimens were immersed in the solution containing fluoride and albumin, the titanium dissolved and the pH values increased. In other words, the specimens were corroded even if the albumin existed in the solution. The presence of albumin, however, suppressed the dissolution of titanium and pitting on the titanium, as shown in Fig. 5, Fig. 6 and Fig. 7. This demonstrated that the albumin film adsorbed on the titanium surface suppresses not only the attack of the fluoride but also the degradation of the passive films.

4.3. Albumin concentration

The electrochemical parameters, such as E_open, R_p, and I_0.5, showed no significant differences between the BSA.01 and the BSA.1, indicating that the corrosion resistance of the titanium was independent on the concentration of BSA in the solution containing the fluoride. In addition, the I_C/I_Ti values for the BSA.01 and the BSA.1 did not increase 10 times even when the concentration of albumin added to the solution containing the fluoride was increased 10 times. Williams et al. [23 and 24] demonstrated that titanium had less protein adsorption among metals, and that the amount of adsorption was not dependent on time. Sundgren et al. [25] suggested that fibrinogen, as a plasma protein, maintained a native conformation on titanium. In this respect, albumin adsorption on titanium increases slightly as the amount of albumin added to a solution containing fluoride is increased, that is, the thickness of the albumin film on the titanium hardly increases. Therefore, the albumin films adsorbed on titanium could alleviate the corrosion of the titanium from the fluoride attack, even when the albumin concentration in the solution is 0.1 g/l.

7. 5. Conclusions

The reactions on titanium in a solution containing fluoride and albumin were demonstrated based on electrochemical properties, surface composition, dissolved titanium concentration, and pH. These results are summarized as following.

1. The corrosion resistance of titanium specimens in a solution containing fluoride decreased in comparison with that in a solution not containing fluoride. Adding albumin to the solution resulted in preventing the corrosion of the titanium from the fluoride.

2. EPMA analysis confirmed that fluorine was present on titanium immersed in the solutions containing fluoride. XPS analysis showed that titanium immersed in a solution containing fluoride and albumin had carbon, nitrogen, and oxygen originating from the albumin.

3. Dissolution of titanium in a solution containing fluoride after titanium immersion was confirmed. Titanium immersed in a solution containing fluoride and albumin had less titanium dissolution than did that in a solution not containing albumin. The amount of albumin added to the solution had little effect on the corrosion resistance of the titanium.

From these results, the effects of albumin are not only the protection of titanium from fluoride attack but also the suppression of dissolution of titanium ions via formation of the albumin films. These effects evidently suppress the corrosion of titanium by fluoride.

8. Acknowledgements

This work was supported in part by a Grant-in-Aid for Developmental Scientific Research (No. 13470420) from the Ministry of Education, Science, Culture and Sports, Japan. The authors would like to thank Mr. Katsumi Tadokoro, Ms. Eiko Watanabe and Mr. Yoshiaki Kitazawa of the Oral Health Science Center of Tokyo Dental College for their valuable support with the experiments.

9. References

1. T. Hanawa and M. Ota, Characterization of surface film formed on titanium in electrolyte using XPS. Appl Surf Sci 55 (1992), pp. 269-276. Abstract

2. J.L. Ong, L.C. Lucas, G.N. Raikar, R. Connatser and J.C. Gregory, Spectroscopic characterization of passivated titanium in a physiologic solution. J Mater Sci Mater Med 6 (1995), pp. 113-119. Abstract-EMBASE | Abstract-Compendex  

3. T. Hanawa, K. Asami and K. Asaoka, Repassivation of titanium and surface oxide film regenerated in simulated bioliquid. J Biomed Mater Res 40 (1998), pp. 530-538. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE   | Full Text via CrossRef

4. J. Lausmaa, B. Kasemo and S. Hansson, Accelerated oxide grown on titanium implants during autoclaving caused by fluorine contamination. Biomaterials 6 (1985), pp. 23-27. Abstract

5. W. Wilhelmsen and A.P. Grande, The influence of hydrofluoric acid and fluoride ion on the corrosion and passive behavior of titanium. Electrochim Acta 32 10 (1987), pp. 1469-1472.

6. G. Boere, Influence of fluoride on titanium in an acidic environment measured by polarization resistance technique. J Appl Biomater 6 (1995), pp. 283-288. Abstract-Compendex | Abstract-MEDLINE  

7. Y. Oda, E. Kawada, M. Yoshinari, K. Hasegawa and T. Okabe, The influence of fluoride concentration on the corrosion of titanium and titanium alloys. Jpn J Dent Mater 15 (1996), pp. 317-322.

8. L. Reclaru and J-.M. Meyer, Effects of fluorides on titanium and other dental alloys in dentistry. Biomaterials 19 (1998), pp. 85-92. Abstract | PDF (504 K)

9. M. Nakagawa, S. Matsuya, T. Shiraishi and M. Ohta, Effect of fluoride concentration and pH on corrosion behavior of titanium for dental use. J Dent Res 78 9 (1999), pp. 1568-1572. Abstract-MEDLINE  

10. M. Nakagawa, S. Matsuya and K. Udoh, Corrosion behavior of pure titanium and titanium alloys in fluoride-containing solutions. Dent Mater J 20 4 (2001), pp. 305-314. Abstract-MEDLINE  

11. M. Nakagawa, S. Matsuya and K. Udoh, Effect of fluoride and dissolved oxygen concentrations on the corrosion behavior of pure titanium and titanium alloys. Dent Mater J 21 2 (2002), pp. 83-92. Abstract-MEDLINE  

12. N. Schiff, B. Grosgogeat, M. Lissac and F. Dalard, Influence of fluoride content and pH on the corrosion resistance of titanium and its alloys. Biomaterials 23 (2002), pp. 1995-2002. SummaryPlus | Full Text + Links | PDF (505 K)

13. R.L. Williams, S.A. Brown and K. Merritt, Electrochemical studies on the influence of proteins on the corrosion of implant alloys. Biomaterials 9 (1988), pp. 181-186. Abstract

14. M.A. Khan, R.L. Williams and D.F. Williams, The corrosion behavior of Ti-6Al-4V, Ti-6Al-7Nb and Ti-13Nb-13Zr in protein solutions. Biomaterials 20 (1999), pp. 631-637. Abstract | PDF (445 K)

15. K. Ide, M. Hattori, K. Hasegawa, M. Yoshinari, E. Kawada and Y. Oda, The effect of fluoride and albumin on corrosion of titanium. J Dent Res 80 (2001), p. 663 (IADR abstract No. 1089) .

16. K. Ide, M. Hattori, M. Yoshinari, E. Kawada and Y. Oda, The influence of albumin on corrosion resistance of titanium in fluoride solution. Dent Mater J 22 3 (2003), pp. 359-370. Abstract-MEDLINE  

17. H.H. Huang, Effects of fluoride and albumin concentration on the corrosion behavior of Ti-6Al-4V alloy. Biomaterials 24 (2003), pp. 275-282. SummaryPlus | Full Text + Links | PDF (228 K)

18. K. Asami and K. Hashimoto, X-ray photoelectron spectra of several oxides of iron and chromium. Corros Sci 17 (1977), pp. 559-570. Abstract-INSPEC | Abstract-Compendex  

19. K. Asami, S.C. Chen, H. Habazaki and K. Hashimoto, The surface characterization of titanium and titanium-nickel alloys in sulfuric-acid. Corros Sci 35 (1993), pp. 43-49. Abstract

20. K. Kaneko, K. Yokoyama, K. Moriyama, K. Asaoka, J. Sakai and M. Nahumo, Delayed fracture of beta titanium orthodontic wire in fluoride aqueous solutions. Biomaterials 24 (2003), pp. 2113-2120. SummaryPlus | Full Text + Links | PDF (2464 K)

21. J.C. Wataha, S.K. Nelson and P.E. Lockwood, Elemental release from dental casting alloys into biological media with and without protein. Dent Mater 17 (2001), pp. 409-414. SummaryPlus | Full Text + Links | PDF (380 K)

22. N.J. Hallab, A. Skipor and J.J. Jacob, Interfacial kinetics of titanium- and cobalt-based implant alloys in human serum: metal release and biofilm formation. J Biomed Mater Res 65A (2003), pp. 311-318. Abstract-Compendex | Abstract-MEDLINE   | Full Text via CrossRef

23. D.F. Williams, I.N. Askill and R. Smith, Protein adsorption and desorption phenomena on clean metal surfaces. J Biomed Mater Res 19 (1985), pp. 313-320. Abstract-EMBASE | Abstract-MEDLINE  

24. R.L. Williams and D.F. Williams, Albumin adsorption on metal surfaces. Biomaterials 9 (1988), pp. 206-212. Abstract

25. J-.E. Sundgren, P. Bodö and I. Lundström, Auger electron spectroscopic studies of the interface between human tissue and implants of titanium and stainless steel. J Colloid Interface Sci 110 1 (1986), pp. 9-20. Abstract-Compendex  

Corresponding author. Tel.: [phone]; fax: [phone]

---