Electrochemical corrosion of titanium and titanium-based alloys

Electrochemical corrosion of titanium and titanium-based alloys Chotiros Kuphasuk, DDS, MSD,a Yoshiki Oshida, PhD,b Carl J. Andres, DDS, MSD,c Suteera T. Hovijitra, DDS, MSD,d Martin T. Barco, DDS, MSD,e and David T. Brown, DDS, MSf School of Dentistry, Indiana University, Indianapolis, Ind. Statement of problem. Two varieties of unalloyed titanium, Ti-6Al-4V and NiTi, commonly are used in medical and dental fields. Several other types of alloys for potential use in these fields have been developed, including Ti-4.5Al-3V-2Mo-2Fe and vanadium-free alloys (Ti-5Al-2.5Fe and Ti-5Al-3Mo-4Zr). The corrosion of these alloys under simulated physiologic conditions is not known. Purpose. This study compared the corrosion behaviors of 6 titanium materials through electrochemical polarization tests in 37°C Ringer’s solution. Material and methods. The applied voltage was potentiostatically scanned from –0.6 to 1.0 V. From polarization curves, the corrosion rate (averaged over 3 samples) for each alloy was calculated and compared with that of other alloys. Analysis of variance (ANOVA) and the Student-Newman-Keuls multiple range test were performed at a 95% overall confidence level to identify statistically significance differences in corrosion rates. Surface oxide films were identified by electron diffraction, and the electrolyte medium was analyzed by atomic absorption spectrophotometry after each alloy was tested. Results. Commercially pure titanium and Ti-5Al-2.5Fe were the most resistant to corrosion; Ti-5Al-3Mo-4Zr, Ti-6Al-4V, and NiTi were the least resistant to corrosion. NiTi exhibited pitting corrosion along with transpassivation. Conclusion. Electron diffraction patterns indicated that all titanium alloys were covered mainly with rutile-type oxide (TiO2) after corrosion tests. The oxides that formed on Ti-5Al-2.5Fe were identified as a mixture of TiO2 and Ti9O17, and those that formed on NiTi were identified as a mixture of TiO2 and Ni2Ti4O. (J Prosthet Dent 2001;85:195-202.)

CLINICAL IMPLICATIONS In this study, NiTi corroded in Ringer’s solution at a higher potential than 500 mV. Surface modification of this alloy (such as selective dissolution of nickel from the surface layer or controlled passivation) may be needed before its intraoral use.

T

itanium and titanium alloys remain the first choice for implants in both medical and dental applications. Titanium materials are inert and resistant to corrosion because of the stable passivity of the surface oxide film. Selected elements, including iron and niobium,1 have been incorporated into titanium alloys to replace toxic elements such as vanadium. Some of these newer alloys are not commercially available, and their biocompati-

This research project was supported in part by a 1998 Stanley D. Tylman Research Award. aFormer Graduate Student, Prosthodontics, Department of Restorative Dentistry; Assistant Professor, Prosthodontics Department, Mahidol University, Bangkok, Thailand. bProfessor, Division of Dental Materials, Department of Restorative Dentistry. cProfessor, Department of Prosthodontics. dAssociate Professor, Department of Prosthodontics. eAssociate Professor, Department of Prosthodontics. fAssociate Professor, Department of Prosthodontics. FEBRUARY 2001

bility with human tissue has not been documented.2 There is no universally accepted definition of the term passivity. In conventional dental materials science, if an implant metal is oxidized and the oxide does not break down under physiologic conditions, the metal is said to be passive or passivated. The passive film (naturally, chemically, or electrochemically formed) on titanium implants is composed mainly of titanium oxide (TiO2), a rutile-type tetragonal structure.3,4 When an implant is introduced into the body, complex reactions begin to take place at the bioenvironment-oxide interface; when the former supplies enough oxygen, oxide growth occurs. Although the rate of formation and composition of this film is important, in the passive state, the rate of dissolution of elemental titanium is extremely low.5 Strietzel et al6 studied the in vitro corrosion of cast, cold-formed, and spark-eroded pure titanium in solutions containing chloride, fluoride, lactate, citrate, THE JOURNAL OF PROSTHETIC DENTISTRY 195

THE JOURNAL OF PROSTHETIC DENTISTRY

oxalate, acetate, and thiocyanate ions. It was concluded that the titanium ion release increased in the presence of fluoride ions; hence, it was recommended that the presence of fluoride ions be avoided even though no corrosion effects were observed clinically. Chen et al7 studied corrosion resistance and ion release in Hanks solution for commercially pure titanium (ASTM grade 2) with different surface morphologies and found that, when surface roughness increased, corrosion resistance decreased and the ion release rate increased. It was observed that the oxide film formed by oxidation in air at 400°C enhanced the corrosion resistance of titanium. Khan et al8 investigated the corrosion behavior of Ti-6Al-4V, Ti-6Al-7Nb, and Ti-13Nb-13Zr as a function of pH in phosphate-buffered saline (PBS), bovine albumin solution in PBS, and 10% fetal calf serum in PBS solution. It was reported that an increase in pH greatly affected the corrosion behavior of all 3 alloys, and the addition of protein to the PBS solution reduced the influence of pH on the corrosion of all alloys. When exposed to conditions present in the oral cavity, metallic restorations and prostheses are subject to the effects of corrosion processes. Interaction between the metallic material and the environment gives rise to electrochemical reactions. One of the problems associated with corrosion studies of dental alloys is that the knowledge of the electrochemical environment of the oral cavity is limited. This makes preparation of the artificial testing media and the interpretation of in vitro data somewhat hypothetical. To evaluate the stability of passive film, it is convenient to refer to the Pourbaix (E-pH) diagram9 if the clinically relevant potential range (E) in the intraoral environment and the pH value range of saliva are known. The oxidation-reduction potential associated with the mouth has received attention since first reported by Eisenbrandt.10,11 He found that the oxidation potential of unstimulated saliva ranged from –17 to +152.5 mV relative to a standard saturated calomel electrode (SCE), with an average of +59 mV (SCE), whereas the pH of saliva for 5 subjects ranged from 6.55 to 6.8. Ewers and Greener12 measured the oxidation potential (Eh) and pH in vivo in 9 patients who did and did not have periodontal therapy. They found that the Eh ranged from –58 to +212 mV (SCE) and that the pH ranged from 6.1 to 7.9. Variations in the age of metallic restorations of the same type can result in potential differences in the oral cavity. Nilner and Holland,13 using 28 patients (20 men and 8 women with a mean age of 25.4 ± 3.2 years), measured the in vivo potential of amalgam restorations against an Ag/AgCl reference electrode. The readings varied from –23 to –595 mV. During the study, new amalgam restorations were placed, and 196

KUPHASUK ET AL

their potentials varied from –180 to –565 mV; this range was significantly lower than that of the older restorations, and it was speculated that the differences in potential were due to the attack on the most corrosion-prone γ2 phase.13 Although it is difficult to find a common range of intraoral potentials for patients without metallic restorations, based on the information presented previously, it is speculated that the range overlapping all reported data is a narrow potential zone of –20 to +150 mV. In this study, the electrochemical behavior of commercially pure titanium (cp Ti), Ti-6Al-4V, NiTi, and 3 other titanium alloys were evaluated in 37°C Ringer’s solution. Electron diffraction and atomic absorption spectrophotometry were performed to identify the crystalline structure(s) of the oxide film and detect dissolved elements to understand the electrochemical reaction of these titanium materials in Ringer’s solution.

MATERIAL AND METHODS Unalloyed titanium and 5 titanium alloys (Daido Steel Co Ltd, Nagoya, Japan) that are used or that represent experimental material for medical and dental applications were tested in this study (Table I). Unalloyed titanium is available in 4 different grades, which are classified by their levels of impurities (primarily, dissolved oxygen) and the resultant effect of these impurities on strength and ductility. ASTM grade 1 has the highest purity, lowest strength, and best room-temperature ductility and formability. Unalloyed titanium belongs to the α-type structure, a hexagonal closed, packed (HCP) crystal structure. Grade 1 cp Ti is an ideal material to withstand a wide variety of corrosive environments, particularly those containing the chlorine ion.2 Ti-6Al-4V presently is the most widely used titanium alloy, accounting for more than 50% of all titanium tonnage in the world. The largest application of Ti-6Al-4V, after the aerospace industry, is medical prostheses, which account for 3% of the market. Ti-6Al-4V has a high specific strength (ratio of strength and density), stability at temperatures up to 400°C, and good corrosion resistance. This alloy is an α-β phase alloy, with the β phase being a body-centered cubic (BCC) structure. Wrought Ti-6Al-4V is a useful material for surgical implants because of its low elastic modulus, good tensile and fatigue strengths, and biologic compatibility. It is used for bone screws and for partial and total replacement of the hip, knee, elbow, jaw, finger, and shoulder joints.2 Materials used for permanent implants in the human body must exhibit corrosion resistance, biocompatibility, amenability to osseointegration, and biofunctionality. These requirements are met by vanaVOLUME 85 NUMBER 2

KUPHASUK ET AL

dium-free Ti-5Al-2.5Fe. This alloy belongs to an α-β duplex phase type. Potential applications for Ti-5Al2.5Fe include total hip endoprostheses, knee joints, spinal implants, dental implants, and all types of joint prostheses as well as bone nails, screws, and plates.1 A new titanium alloy for surgical implant application, vanadium-free Ti-5Al-3Mo-4Zr (Ti-5/3/4), has been developed. This new alloy is advantageous over Ti-6Al-4V because the biohazardous alloying elements (vanadium, for example) have been eliminated; moreover, it is superior to austenitic stainless steel in mechanical properties and corrosion abrasive wear resistance.1 This alloy has a β-rich α-β duplex phase and has been used to make a newly designed artificial hip joint.1 Ti-4.5Al-3V-2Mo-2Fe (SP-700) is also a β-rich α-β duplex-phase titanium alloy designed to offer super plastic formability properties that are superior to those of Ti-6Al-4V.14 The former alloy’s low-flow stress and fine-grain microstructure result in good super plastic formability at 700°C (hence the name SP-700). The microstructure also results in an attractive combination of mechanical properties and corrosion resistance. The alloy exhibits excellent heat-treatability, coldformability, and hot-forgeability, which make it suitable for the fabrication of a full denture base.1 Nickel-titanium (NiTi) alloys with 49.0% to 50.7% titanium are common commercially. Because of their uniqueness (including super elasticity and shapememory, good corrosion resistance, and high damping capacity), these alloys are highly versatile; they are used for fabricating blade-type dental implants, orthopedic implants, orthodontic appliances, and endodontic files.2

Electrochemical corrosion test Three specimens of each of the 6 materials were prepared. Each specimen was wet polished with a series of SiC abrasive papers with grit size up to 2000. A mixture of epoxy glue and silver paste was used to attach the back surface of each specimen to a copper wire, which provided electrical contact to the electrochemical corrosion testing apparatus. Ringer’s was the electrolyte solution chosen for testing; it is composed of 8.6 g NaCl, 0.3 g KCl, and 0.33 g CaCl2, with the remainder being distilled water (these figures represent a 1000-mL solution with a 7.0 pH). Electrochemical corrosion studies were conducted at 37°C ± 1°C, and the system was deaerated with nitrogen gas to simulate the intraoral environment and reduce fluctuation in the oxygen concentration of the electrolyte.15,16 The potential was measured against an SCE as a standard electrode. A polarization curve was recorded with a potentiostat (Model 362, EG&G, Instruments, Princeton Applied Research, Princeton, N.J.) at scanning rate of FEBRUARY 2001

THE JOURNAL OF PROSTHETIC DENTISTRY

0.5 mV/sec. A scanning range of –600 to +1000 mV was recorded. The polarization tests were repeated at least 3 times for each specimen until all repeated runs exhibited similar polarization curves. The specimens and electrolyte solution also were used for electron diffraction and atomic absorption tests, respectively. The corrosion rate was estimated with the following equation: Corrosion rate (mpy) = 0.13 × ICORR × EW/d where mpy stands for mils per year, EW is an equivalent weight of the corroding species in grams (g), d is the density of the corroding species (g/cm3), and ICORR is a corrosion current density (µA/cm2). The corrosion current density, ICORR, was obtained graphically by finding the intersecting point of the catholic Tafel slope and the anodic Tafel slope.17 For this study, we used d = 4.51 (g/cm3), and EW = 11.975 (= 47.90/4) (g), respectively.

Electron diffraction The oxide film was removed from the metal substrates by immersing the sample in a mixed solution of HF, NHO3, and H2O (1:4:75 by volume).17,18 After washing the oxide film with distilled water and methyl alcohol several times, an electron diffraction pattern of the oxide film was obtained with a transmission electron microscope (Model CM-10, Phillips, Amsterdam, The Netherlands) using an accelerating voltage of 100 kV. In the Bragg diffraction equation, 2dsinθ = nλ (where λ is a wavelength, and n is an integral that indicates the order of diffraction, normally taken as unity), λ (wavelength for incident electrons) can be approximated by λ = (150/V) 1/2. When the accelerating voltage (V) is 100 kV, then λ ≈ 0.038 Å. Accordingly, the diffraction angle (2θ) can be 5 × 10–2 radian ≈ 3 degrees for a crystalline structure having an interplanar spacing of d = 0.5 ∆. Because 2θ is very small, the Bragg equation can be simplified as: 2dθ = λ (if n = 1). If λ is kept constant, the radius (r) of the electron diffraction ring is given by the following relationship: 2rd = λ = constant. If a standard sample with known d-spacings is used, the d-spacings of an unknown sample can be readily obtained. In this study, a thin gold foil (Au) was used as a standard sample, so that 2rAudAu = constant = 2rsampledsample.17 The constant (2rAudAu) was calculated and averaged as 8.854 over all diffracted rings from the Au foil.

Atomic absorption spectroscopy Chemical analysis was performed on the collected electrolyte using a graphite furnace (Model HGA 400, Perkin Elmer, Norwalk, Conn.) with an atomic absorption spectrophotometer (Model 306, Perkin Elmer). The furnace and spectrophotometer were programmed according to the manufacturer’s recom197

THE JOURNAL OF PROSTHETIC DENTISTRY

KUPHASUK ET AL

Fig. 1. Polarization curves of cp Ti in Ringer’s solution.

mended parameters for Ti, Al, V, Fe, Mo, Zr, and Ni. The detectable limit with this technique for each element is as follows: 0.05 ppm for Ti, 0.03 ppm for Al, 0.04 ppm for V, 0.003 ppm for Fe, 0.03 ppm for Mo, 0.4 ppm for Zr, and 0.004 ppm for Ni.

Statistical analysis The means and standard deviations of the corrosion rates were obtained from 3 runs for each alloy. Analysis of variance (ANOVA) was performed at a significance level of α=.05. If the ANOVA indicated that there were significant differences among the tested materials, the Student-Newman-Keuls multiple range test was conducted.

RESULTS Polarization curves Figures 1, 2, and 3 are typical polarization curves obtained from cp Ti, Ti-4.5Al-3V-2Mo-2Fe, and NiTi, respectively. In Figure 1, the current density (approximately 3 µA/cm2) and potential (approximately –200 mV) when the passivation started are 198

marked. The passive region is also indicated. Because cp Ti did not exhibit a breakdown of the passive film (transpassivation) at up to 1000 mV in this study, there is no mark for a higher potential end of the passive region. All titanium materials tested in this study showed passivation current density ranging from 6 to 30 µA/cm2. On the other hand, NiTi exhibited transpassivation at approximately 500 mV for the N3 run (or at 750 mV for N12 and N1 runs) (Fig. 3). Transpassivation takes place with the generation of oxygen gas accompanied by the anodic dissolution of a metallic ion, which was a nickel ion in this study. Many corrosion pits were observed on tested NiTi surfaces.

Corrosion rate Table I summarizes individual ICORR and CR (corrosion rates in mpy) for 3 readings for each tested titanium material. With the use of those calculated CR values, an ANOVA was conducted at a significance level of α=.05. The ANOVA indicated that there were some significant differences among the materials testVOLUME 85 NUMBER 2

KUPHASUK ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 2. Polarization curves of Ti-4.5Al-3V-2Nb-2Fe in Ringer’s solution.

ed. Accordingly, the Student-Newman-Keuls multiple range test was conducted. The result indicated that there was no significant difference in corrosion resistance among Ti-6Al-4V, Ti-5Al-3Mo-4Zr, Ti-4.5Al-3V-2Mo-2Fe, and NiTi, and there was no significant difference among cp Ti, Ti-5Al-2.5Fe, and Ti-4.5Al-3V-2Mo-2Fe. However, there was a significant difference between Ti-6Al-4V and NiTi and between cp Ti and Ti-5Al-2.5Fe at a 95% overall confidence level.

Electron diffraction Figures 4 and 5 are typical electron diffraction patterns of the oxide films formed on Ti-5Al-2.5Fe and NiTi, respectively. Oxide films formed on all 6 titanium materials were identified as rutile-type tetragonal structure TiO2, whereas the oxide film formed on Ti-5Al-2.5Fe also had a trace amount of Ti9O17. In addition to TiO 2, Ti 4Ni 2O was identified in the oxide film formed on NiTi, indicating that nickel was incorporated into the surface oxide film. The FEBRUARY 2001

diffraction patterns obtained from oxide films formed on the NiTi surface showed spotty patterns, indicating that the oxide crystals had preferred orientation, although it was not possible to specify the orientation from this pattern.

Atomic absorption spectroscopy analysis The concentrations of Ti, Al, V, Fe, Mo, and Zr were below the limits of detection. However, 0.6 ppm Ni ion concentration was detected in the Ringer’s solution used for testing NiTi.

DISCUSSION Analysis of the crystalline surface oxide films removed from the tested titanium materials indicated that the oxide was composed mainly of rutile TiO2, although the oxides on the Ti-5Al-2.5Fe and NiTi alloys contained an additional phase. As mentioned previously, the potential range corresponding to intraoral condition lies in a narrow zone from –20 to +150 mV. At the same time, a stable passive film (mainly of TiO2) 199

THE JOURNAL OF PROSTHETIC DENTISTRY

KUPHASUK ET AL

Fig. 3. Polarization curves of NiTi in Ringer’s solution.

Table I. Results of electrochemical corrosion tests Titanium materials

Run

ICORR (µm/cm2)

Corrosion rate (MPY)

Average corrosion rate (MPY)

Significant difference

cp Ti

C32 C33 C34 F1 F2 F3 S1 S2 S3 T23 T24 T34 V13 V22 V23 N1 N12 N3

0.039 0.036 0.033 0.064 0.068 0.044 0.043 0.102 0.113 0.110 0.111 0.090 0.260 0.100 0.240 0.260 0.119 0.240

0.013 0.012 0.011 0.022 0.023 0.015 0.015 0.035 0.039 0.038 0.038 0.031 0.089 0.034 0.082 0.89 0.041 0.082

0.012 ± 0.014

a

0.0200 ± 0.025

a

0.0297 ± 0.038

a,b

0.0357 ± 0.044

b

0.0683 ± 0.088

b

0.0707 ± 0.090

b

Ti-5Al-2.5Fe

Ti-4.5Al-3V-2Mo-2Fe (SP-700) Ti-5Al-3Mo-4Zr (Ti-5/3/4) Ti-6Al-4V

NiTi

MPY = mils (1/1000 in) per year; cp Ti = commercially pure titanium.

200

VOLUME 85 NUMBER 2

KUPHASUK ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 4. Transmission electron diffraction pattern of oxide film (identified as rutile-type TiO2 with a trace of Ti9O17) formed on Ti-5Al-2.5Fe.

Fig. 5. Transmission electron diffraction pattern of oxide film (identified as mixture of rutile-type TiO2 and Ni2Ti4O) formed on NiTi.

formed on all tested titanium materials can provide excellent corrosion (oxidation) resistance, even if the applied potential rises to 500 to 750 mV for NiTi and to 1000 mV or higher for the rest of the tested titanium materials. The initial stability of titanium oxide film is assumed to be related inversely to corrosion rate. From a physical metallurgy point of view, each alloying element (such as Al, Fe, V, Mo, Zr, and Ni for tested titanium alloys in this study) has different alloying effects on various properties. Elements such as Al and Zr, plus the interstitial elements (C, O, and N), are strong solid-solution strengtheners that produce little change in the transformation temperature (β-transus) from the HCP (α) to BCC (β) structure of pure titanium. Hence, these elements, known as α-stabilizers, produce a good high-temperature performance. Conversely, alloying additions that decrease the β-transus temperature are referred to as β-stabilizers. Generally, they are transition metals, such as V, Mo, Nb, and Ta, that provide a friability. 19 Although an aluminum neurotoxicity might cause Alzheimer’s disease, 20-22 Al is still an indispensable alloying element to titanium. As indicated by the atomic absorption spectroscopy data, no Al ion was detected in the Ringer’s solution used for testing Al-containing titanium alloys even after being polarized up to 1000 mV. Several studies on vanadium toxicity exist in the literature 23,24 and

have led to the development of a vanadium-free titanium material.1 Preliminary studies on the effects of alloying elements on corrosion behavior suggested that Fe had no influence; Al exhibited a slight effect; V, Zr, and Ni showed adverse effects; and Mo exhibited a beneficial alloying effect. 25 These results explain why Ti-5Al-2.5Fe showed better corrosion resistance than other titanium materials, except cp Ti. As described previously, a 0.6 ppm Ni ion concentration was detected in the Ringer’s solution used for testing NiTi over its transpassive potential. Hypersensitivity to nickel has been well documented.26-28 Although NiTi has wide potential applications in both medical and dental fields,2 the potential risk of nickel ion release from the NiTi surface layer should be prevented. The use of chemical treatment to selectively dissolve only nickel from the surface layer of NiTi, so that a titanium-rich surface layer is created, is 1 potential solution.4,29,30 Another alternative would be administering an oxidation-passivation treatment to the NiTi alloy under a controlled oxygen partial pressure before use; this would prevent nickel from being incorporated into the surface oxide film.31

FEBRUARY 2001

CONCLUSIONS This study compared the corrosion behavior of 6 titanium materials through electrochemical polariza201

THE JOURNAL OF PROSTHETIC DENTISTRY

tion studies in 37°C Ringer’s solution. Under the experimental conditions of this study, the following conclusions were drawn: 1. All samples showed good resistance to electrochemical corrosion over the potential of relevance for intraoral conditions. 2. Commercially pure Ti and Ti-5Al-2.5Fe showed significantly different corrosion properties than Ti-6Al-4V and NiTi. The former pair exhibited the lowest corrosion rate; the latter pair exhibited the highest corrosion rate. 3. NiTi alloys revealed transpassive behavior at the potentials between 0.5 and 0.75 V (vs SCE) accompanied by pitting corrosion. 4. All samples tested in 37°C Ringer solution were covered mainly with rutile-type TiO2. 5. There was no detectable level of ion dissolution from 5 of the tested alloys; however, 0.6 ppm nickel ions were detected in the Ringer’s solution used for testing NiTi. REFERENCES 1. Higo Y, Ouchi C, Tomita T, Murata K, Sugiyama H. Properties evaluation and its application to artificial hip joint of newly developed titanium alloy for materials. Proc 3rd Jpn Int SAMPE Symp 1993:1621-6. 2. Moore BK, Oshida Y. Materials science and technology in dentistry. In: Wise DL, editor. Encyclopedic handbook of biomaterials and bioengineering. New York: Marcel Dekker Co; 1995. p. 1325-1430. 3. Oshida Y, Sachdeva R, Miyazaki S. Changes in contact angles as a function of time on some pre-oxidized biomaterials. J Mater Sci:Mater Med 1992;3:306-12. 4. Oshida Y, Sachdeva RC, Miyazaki S. Microanalytical characterization and surface modification of TiNi orthodontic archwires. Biomed Mater Eng 1992;2:51-69. 5. Solar RJ. Corrosion resistance of titanium surgical implant alloys: a review. In: Syrett BD, editor. Corrosion and degradation of implant materials. ASTM STP 684. Philadelphia, PA: American Society for Testing and Materials; 1979. p. 259-73. 6. Strietzel R, Hösch A, Kalbfleisch H, Buch D. In vitro corrosion of titanium. Biomaterials 1998;19:1495-9. 7. Chen G, Wen X, Zhang N. Corrosion resistance and ion dissolution of titanium with different surface microroughness. Biomed Mater Eng 1998;8:61-74. 8. Khan MA, Williams RL, Williams DF. The corrosion behaviour of Ti-6Al4V, Ti-6Al-7Nb and Ti-13Nb-13Zr in protein solutions. Biomaterials 1999;20:631-7. 9. Pourbaix M. Atlas of electrochemical equilibria in aqueous solutions. Oxford: Pergamon Press; 1966. p. 213-22. 10. Eisenbrandt LL. Initial oxidation reduction potential of saliva. J Dent Res 1943;22:293. 11. Eisenbrandt LL. Studies on the oxidation reduction potentials of saliva. J Dent Res 1945;24:247. 12. Ewers GJ, Greener EH. The electrochemical activity of the oral cavity—a new approach. J Oral Rehabil 1985;12:469-76. 13. Nilner K, Holland RI. Electrochemical potentials of amalgam restorations in vivo. Scan J Dent Res 1985;93:357-9. 14. Ouchi C, Minakawa K, Takahashi K, Ogawa A, Ishikawa M.

202

KUPHASUK ET AL

15. 16. 17. 18.

19. 20.

21.

22.

23.

24.

25.

26. 27.

28. 29.

30.

31.

Development of β-rich α-β titanium alloy: SP700. NKK (Nihon Kou Kan) Tech Rev 1992;65:61-7. Reclaru L, Meyer JM. Study of corrosion between a titanium implant and dental alloys. J Dent 1994;22:159-68. Venugopalan R, Lucas LC. Evaluation of restorative and implant alloys galvanically coupled to titanium. Dent Mater 1998;14:165-72. Oshida Y, Moore BK. Anodic polarization behavior and microstructure of a gallium-based alloy. Dent Mater 1993;9:234-41. Platt JA, Guzman A, Zuccari A, Thornburg DW, Rhodes BF, Oshida Y, et al. Corrosion behavior of 2205 duplex stainless steel. Am J Orthod Dentofacial Orthop 1997;112:69-79. Collings EW. The physical metallurgy of titanium alloys. Columbus, OH: The American Society for Metals; 1984. p. 3-5. Tsunoda M, Shama RP. Modulation of tumor necrosis factor alpha expression in mouse brain after exposure to aluminum in drinking water. Arch Toxicol 1999;73:419-26. Nordenran G, Ryd-Kjellen E, Johansson G, Nordstrom G, Winblad B. Alzheimer’s disease, oral function and nutritional status. Gerodontology 1996:13:9-16. Minami T, Ichii M, Tohno Y, Tohno S, Utsumi M, Yamada MO, et al. Agedependent aluminum accumulation in the human aorta and cerebral artery. Biol Trace Elem Res 1996;55:199-205. Grabowski GM, Paulauskis JD, Godleski JJ. Mediating phosphorylation events in the vanadium-induced respiratory burst of alveolar macrophages. Toxicol Appl Pharmacol 1999;156:170-8. Sanchez DJ, Colomina MT, Domingo JL, Corbella J. Prevention by sodium 4,5-dihydroxybenzene-1,3-disulfonate (Tiron) of vanadium-induced behavioral toxicity in rats. Biol Trace Elem Res 1999;69:249-59. Kuphasuk C. Electrochemical corrosion behaviors of titanium and titanium-based alloys. [MSD Thesis.] Indianapolis, IN: Indiana University School of Dentistry; 1999. Fleming CJ, Burden AD, Forsyth A. The genetics of allergic contact hypersensitivity to nickel. Contact Dermatitis 1999:41:251-3. Rapisarda E, Bonaccorso A, Tripi TR, Fragalk I, Condorelli GG. The effect of surface treatments of nickel-titanium files on wear and cutting efficiency. Oral Surg Oral Med Oral Path Oral Radiol Endod 2000;89:363-8. Barr ES, Kleier DJ, Barr NV. Use of nickel-titanium rotary files for root canal preparation in primary teeth. Pediatr Dent 2000;22:77-8. Oshida Y. Requirements for successful biofunctional implants. Proceedings of the 2nd Symposium International on Advanced Biomaterials, 2000, Montreal, Canada. p. 5. Kapanen A, Ryhanen J, Tuukkanen J. The effect of nickel titanium shape memory metal alloy on ectopic bone formation. Proceeding of the 2nd Symposium International on Advanced Biomaterials, 2000, Montreal, Canada. p. 13. Venugopalan R, Trepanier C, Pelton AR. Corrosion resistance of passivated NiTi. Proceedings of the 2nd Symposium International on Advanced Biomaterials, 2000, Montreal, Canada. p. 77.

Reprint requests to: DR YOSHIKI OSHIDA DEPARTMENT OF RESTORATIVE DENTISTRY DENTAL MATERIALS DIVISION INDIANA UNIVERSITY SCHOOL OF DENTISTRY 1121 W MICHIGAN ST INDIANAPOLIS, IN 46202-5186 FAX: (317)274-2419 E-MAIL: [email protected] Copyright © 2001 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/2001/$35.00 + 0. 10/1/113029 doi:10.1067/mpr.2001.113029

VOLUME 85 NUMBER 2