Electrochemical Properties of Suprastructures Galvanically Coupled to a Titanium Implant

Transcription

1 Electrochemical Properties of Suprastructures Galvanically Coupled to a Titanium Implant Keun-Taek Oh, Kyoung-Nam Kim Department of Dental Biomaterials and Bioengineering and Research Center for Orofacial Hard Tissue Regeneration, College of Dentistry, Yonsei University, 134 Shinchon-Dong, Seodaemun-Gu, Seoul, , South Korea Received 23 October 2003; revised 23 December 2003; accepted 23 December 2003 Published online 6 April 2004 in Wiley InterScience ( DOI: /jbm.b Abstract: In recent years, dental implants have been widely used for the aesthetic and functional restoration of edentulous patients. Dental implants and restorative alloys are required with high corrosion resistance. Suprastructures and implants of different compositions in electrical contact may develop galvanic or coupled corrosion problems. In addition to galvanic corrosion, crevice and pitting corrosion may occur in the marginal gap between dental implant assemblies. In this study, gold, silver palladium, cobalt chromium, and nickel chromium suprastructures were used to investigate their galvanic and crevice corrosion characteristics in combination with titanium (Ti) implants. Potentiodynamic and potentiostatic testing were performed in artificial saliva at 37 C. Potentiodynamic testing was carried out at the potential scan rate of 1 mv/s in the range of mv (SCE). Potentiostatic testing was performed with an open-circuit potential and current densities at 250, 0, and 250 mv (SCE) in artificial saliva. After electrochemical testing, surface morphologies and cross-sections were examined using micrographs of the samples. Potentiodynamic test results indicated that suprastructure/ti implant couples produced passive current densities in the range of A/cm 2 ; Ti abutment/ti implant and gold/ti implant couples exhibited relatively low passive current densities; Co Cr/Ti implant couples the highest. Co Cr and Ni Cr/Ti implant couples showed breakdown potentials of 700 and 570 mv (SCE), respectively. The open-circuit potentials of silver, Ti abutment, gold, Ni Cr, and Co Cr/Ti implant couples were , , , , and mv (SCE), respectively, and did not change with immersion time. The couples exhibited cathodic current densities at 250 mv (SCE); in particular, gold and silver alloys showed high cathodic current densities of 3.18 and 6.63 A/cm 2, respectively. At 250 mv (SCE), Ti abutment/ti implant couples exhibited a minimum current density of A/cm 2, but gold, Ni Cr, Co Cr, and silver/ti implant couples exhibited 0.313, 1.27, 5.60, and 8.06 A/cm 2, respectively. All couples exhibited relatively low current densities at 0 mv (SCE). Photomicrographs after electrochemical testing showed crevice or pitting corrosion in the marginal gap and at the suprastructure surface. Although of the tested samples Co Cr/Ti implant couples showed the possibility of galvanic corrosion, its degree was not significant. However, it should be borne in mind that galvanic corrosion can accelerate localized corrosion, such as crevice or pitting corrosion Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 70B: , 2004 Keywords: corrosion; dental/endosteal implant; implant interface; metal (alloys) INTRODUCTION When saliva penetrates into prosthetic components in contact with differential metals, the metals generate currents, owing to a potential difference created by the formation of a galvanic cell. In the case of dental implants, reports indicate that Correspondence to: K.-N. Kim, Department/Research Institute of Dental Biomaterials and Bioengineering, College of Dentistry, Yonsei University, 134 Shinchon- Dong, Seodaemun-Gu, Seoul, , South Korea ( kimkn@yumc. yonsei.ac.kr) Contract grant sponsor: Brain Korea 21 Project for Medical Science, Yonsei University Contract grant sponsor: the Medical Science and Engineering Research Program of the Korea Science & Engineering Foundation; contract grant number: R Wiley Periodicals, Inc. complicated electrochemical processes related to implant and suprastructure are linked to galvanic corrosion and pitting. 1 3 Various types of suprastructure materials are used including gold alloy, silver palladium alloy, nickel chrome alloy, cobalt chrome alloy, and titanium (Ti). These metals have high corrosion resistances, but when they are in electrical contact inside the oral cavity, they carry the risk of galvanic corrosion. Due to an array of corrosion factors such as the presence of chloride ion, difference in oxygen concentration, dental plaque, microorganisms, and mechanical stress, the possibility of corrosion occurring is increased still further. Dental implant prosthetics is composed of a titanium fixture and a suprastructure assembly composed of many parts. Therefore, microgaps exist in the interface between such 318

2 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 319 TABLE I. The Types and Compositions of Samples Used in This Study Type Composition (wt %) Specification Manufacturer cp Ti G3 C:0.010, Fe: 0.08, H: , Ti Grade 3, ASTM F 67 Dynamet Co., USA N: 0.004, O: 0.21, Ti bal. Gold alloy Pt: 1.5, Pd: 2, Ag: 12, Cu: 8, Au bal. Dental casting gold alloy Type II, ISO 1562 AP-76, Kujeong Co., Seoul, Korea Ag Pd alloy Au: 0.9, Pd: 26, Cu: 4, In: 0.1%, Zn 0.1 Ag bal. Silver palladium alloy Type II, ISO 8891 WG-2, Kujeong Co., Seoul, Korea Co Cr alloy Cr: 28.5, Mo: 4.5, Co bal. ASTM F 75 Biosil h, Degussa Co., Frankfurt, Germany Ni Cr alloy Cr: 22.0, Mo: 9.0, Si: 1.6, Fe: 0.5, Ce: 0.4, Ni bal. Porcelain fused to metal, ADA No. 14 Wirocer, Bego Co., Bremen, Germany components. In all screw-retained fixture abutment assemblies, a potential dead space exists at the abutment fixture interface, along the abutment screw threads, and in the cavity at the base of the screw. It has been postulated that this space is occupied by macromolecules, cervicular fluid, and/or saliva, which can leach out of surrounding tissue. Binon et al. 4 reported that there are interfacial discrepancies of a dimension ca. 49 m in fixture abutment connection areas. The diameters of oral microorganisms are less than 10 m, and the gap between the fixture and the abutment of different implant systems is in the range of m. Because the gap between the abutment and the prosthetic is larger than the dimensions of microorganisms, there is a possibility of bacteria penetration. 5 Traversy and Birek 6 reported bidirectional leakage of fluid and the penetration of Streptococcus sanguis along the Brånemark fixture abutment interface. Microleakage on each interface is thought to be the most likely cause of bacterial contamination inside an implant. This is a probable cause of the 1-mm alveolar ridge bone loss, which occurs during the year following implant surgery, and the source of clinically observed malodor, periimplant mucositis, and periimplantitis. To prevent such leakage, the use of a permanent seal (a permanent silicone ring) within the abutment was introduced. However, significant concentrations of microorganisms were observed on the apical part of abutment screws despite the presence of such a ring. 7 Wahl et al. 8 reported that a considerable leakage occurred along the components of implant system, even in the IMZ implants system. This report also found that the passive fit obtained using a POM (poly-oxy-methylene) intramobile element instead of the titanium abutment screw, reduces such leakage. Mairgüther and Nentwig 9 asserted that a conical abutment, which fits into the conical orifice of the implant perfectly, like the Astra implant system, could prevent such bacteria penetration. These gaps on the interface can lead to crevice corrosion of the metal. In general, the amalgam restoration often used clinically leaves a marginal gap of m. Amalgam restorations do not bond chemically to the tooth structure or completely seal prepared cavities. On exposure to the oral environment, a newly placed amalgam restoration is subject to corrosion, and the products of this corrosion are deposited in the gaps, resulting in reduced microleakage. 10 Minimizing the gap at the amalgam tooth interface can increase the self-sealing effect of corrosion byproducts. In this study we aimed to investigate the electrochemical characteristics of pure titanium used as an implant fixture and abutment, and gold alloy, silver palladium alloy, nickel chrome alloy, and cobalt chrome alloy used as suprastructures, when they are placed inside the oral cavity. Studies on galvanic and crevice corrosion were conducted upon titanium fixtures in artificial saliva and on assemblies of suprastructures manufactured from the above-mentioned alloys. The study was made to detect possibilities of marginal prevention and to select noble metals and base metals best suited as suprastructure materials. For this study, potentiodynamic polarization testing, open-circuit potential measurements, and potentiostatic testing were performed, and the corrosion behavior and the sealing effect of a marginal gap were observed using an optical microscope. MATERIALS AND METHODS Materials Table I shows the types and compositions of samples used in this study. For the suprastructure, the UCLA-type hexed plastic cylinder was used to manufacture a maxillary premolar tooth sample. The samples were cast in conventional ways using cast alloys and then polished. Gold alloy and silver palladium alloy were invested in calcium sulfate-bonded investment, and nickel chrome alloy and cobalt chrome alloy in a phosphate-bonded investment, as indicated by each manufacturer. Nickel chrome alloy was cast using a high-frequency casting machine (Nautilus, Bego Co., Bremen, Germany), and the remainder by using a gas oxygen torch. After ultrasonically cleaning the casts for 15 min and removing residual investments, the gold alloy cast was pickled for 5 min in the 50% hydrochloric acid solution. Silver palladium, nickel chrome, and cobalt chrome alloy casts were sandblasted to remove the oxide films and residual investments. Finishing and polishing were performed using light pressure and standard conditions, by using a high-speed polishing motor. After polishing with a green stone point, a green-

3 320 OH AND KIM Figure 1. Photographs of supraconstructions/ti implant couples before testing. course rubber wheel, a green clasp polisher, a green rubber point, a hard bristle brush with aluminum oxide, and a soft brush with a polishing paste, in order, at the bottom part of the cast abutment was refined with a milling device to ensure an excellent and consistent fit with the implant fixture. Each polishing phase was conducted for 90 s at 15,000 rpm. After polishing, the abutments were ultrasonically cleaned for 2 min to remove impurities and dried with warm air. Copper wires were soldered onto the finished cast with solder, and all sides, excluding the abutment surface, were covered with commercial resin. Copper wire was tied around the proximal-distal area of the standard-type regular-platform titanium fixture (cp Ti grade 3, IFA 313, Osstem Implant Co., Pusan, Korea.) of a length of 13 mm and a diameter of 3.75 mm and mounted to epoxy resin. The cast UCLA abutment was connected to the titanium fixture using a titanium screw (Ti grade 5, ASR 200, Osstem Implant co., Pusan, Korea.), and the closing torque recommended by the manufacturer. Locking and unlocking were performed using a torque controller with a force of 20 Ncm, and then it was finally locked on the third round (Figure 1). A titanium standard abutment (cp Ti grade 3, SAR 850, Osstem Implant Co., Pusan, Korea.) of 8.5 mm in height with a surface area similar to that of the manufactured UCLA abutment was connected to a similar type of a titanium fixture using a titanium screw (cp Ti grade 3, SARS 850, Osstem Implant Co., Pusan, Korea) (Figure 1). Exposed copper wire, the lower neck areas of the implant fixtures, and the holes of the occlusal planes of each assembly were covered with epoxy resin. The prepared specimens were dried after ultrasonically cleaning with acetone and ethyl alcohol. Method Electrochemical testing was performed according to ISO 10271:2001(E) Dental metallic materials-corrosion test methods, using a potentiostat (Model 263, EG&G, USA), two high-density carbon electrodes as auxiliary electrodes, and an SCE (saturated calomel electrode, SCE), as a reference electrode. Artificial saliva of 37 1 C was used as the test solution (Table II). A potentiodynamic polarization test and open-circuit potential measurements were performed, and three rounds of potentiostatic tests were conducted for each specimen at 250, 0, and 250 mv (SCE). To load the specimen and to eliminate the effects of oxide and other impurities on the surface, a cathodic potential of 600 mv (SCE) was applied. After a 10-min forced surface reduction, the open-circuit potential state was maintained for 10 min. In the potentiodynamic polarization test, the potential was increased to 1600 mv (SCE) at a rate of 1 mv/s, from 600 to 1600 mv (SCE). The open-circuit potential measurement and potentiostatic testing were performed for up to 5000 s. After the potentiodynamic polarization testing and potentiostatic testing had been performed, the surface of the suprastructure and the suprastructure fixture interface were observed under a microscope (Hiscope, Hirox/KH 1000 model, Micro Hiscope System, Japan). To assess the corrosion behavior and sealing effects at the marginal gap, the specimens tested electrochemically were mounted on epoxy resin. They were polished with SiC paper No in a step-by-step procedure and observed under an optical microscope (EPI- PHOT-TME, Nikon, Japan). RESULTS Potentiodynamic Polarization Behavior In terms of potentiodynamic polarization testing of the suprastructure, the titanium abutment showed a current density of 6.29 A/cm 2 at 250 mv (SCE), gold alloy recorded the lowest [0.15 A/cm 2 at 250 mv (SCE)], and silver palladium alloy the highest [9.24 A/cm 2 at 250 mv (SCE)]. Nickel chrome and cobalt chrome alloys showed passive current densities of 2.31 and 2.27 A/cm 2 at 250 mv (SCE), respectively, and generated a pitting at 580 and 600 mv (SCE), respectively. The zero-current potential of base metal alloys appeared to be lower than that of noble metal alloys (Figure 2). Potentiodynamic polarization testing of the suprastructure/ titanium couple in artificial saliva showed that the titanium abutment couple had the lowest passive current density, and TABLE II. Constituents of Artificial Saliva Constituent Concentration (g/l) NaCl 0.40 KCl 0.40 CaCl 2 2H 2 O NaH 2 PO 4 2H 2 O Na 2 S 9H 2 O CO(NH 2 ) 2 (Urea) 1.0 Distilled water Bal.

4 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 321 In artificial saliva, the open-circuit potential of each of the suprastructure/titanium implant couples was generally maintained at a constant level with time. Some titanium abutment couples and nickel chrome alloy couples showed rapid changes in the open-circuit potential. Depending on the specimen, the gold alloy couple exhibited open-circuit potentials in two ranges, and the silver palladium alloy couple showed variable open-circuit potentials (Figure 5). The silver palladium alloy, the titanium abutment, and the gold alloy couple had relatively high open-circuit potential, but among the alloy couples, there was no significant difference. Significant differences were found between the above-mentioned alloys (silver palladium alloy, the titanium abutment, and the gold alloy couples) and the nickel chrome and the cobalt chrome alloy couples. The nickel chrome alloy couple had a low open-circuit potential and the cobalt chrome alloy couple the lowest (Figure 6). Figure 2. Anodic polarization curves of supraconstructions. that the cobalt chrome alloy couple recorded the highest. As for the passive current density at 250 mv (SCE), the titanium abutment couple value was 0.5 A/cm 2, the gold alloy 1.7 A/cm 2, the silver palladium alloy couple 6.2 A/cm 2, the nickel chrome alloy couple 3.7 A/cm 2, and the cobalt chrome alloy couple 11.6 A/cm 2. Of these couples, the nickel chrome alloy couple and the cobalt chrome alloy couple generated pitting, and their pitting potentials were 700 and 570 mv (SCE), respectively. The gold alloy couple gets out of the immune area over 880 mv (SCE). In terms of the zero-current potential, the gold alloy couple and the silver palladium alloy couple showed the highest potentials, the nickel chrome alloy couple was the next highest in rank, and the titanium abutment couple and the cobalt chrome alloy couple low zero-current potentials (Figure 3). On comparing the dental alloy, suprastructure, and suprastructure/titanium implant couple potentiodynamic polarization curves, titanium did not exhibit a great difference in passive current density as an alloy, wrought abutment, or as a couple, and no pitting appeared. However, the zero-current potential showed great difference according to surface state. In the case of gold alloy, the suprastructure showed lowcurrent density, and in the case of cast alloy and of the suprastructure and implant in the couple, it exhibited similar current density. In the case of the silver palladium alloy, all showed similar potentiodynamic polarization behavior; however, there were differences in the zero-current potential. In the case of the cobalt chrome alloy, the precast alloy had the highest passive current density, the suprastructure the lowest, and the suprastructure/implant couple an intermediate density. All three cases generated pitting. In the case of the nickel chrome alloy, all three cases exhibited similar potentiodynamic polarization behavior and pitting (Figure 4). Potentiostatic Test Results The suprastructure and titanium implant couple s current density were measured at 250, 0, 250 mv (SCE) in the oral oxidation-reduction potential range. Overall, the results obtained showed that the current density was maintained at a relatively constant level. However, when there was a particularly low current density, current density fluctuations were observed (Figure 6). All couples showed a negative current density at 250 mv (SCE). However, the value of the current density was different for each couple. The titanium abutment couple and nickel chrome alloy couple current densities were negative, at and A/cm 2, respectively. Gold alloy and silver palladium alloy couples had high negative current densities of 3.18 and 6.63 A/cm 2, respectively. The cobalt chrome alloy couple showed a current density of A/cm 2. At 250 mv (SCE), the titanium abutment couple had the lowest current density ( A/cm 2 ). The gold alloy couple had a low current density of A/cm 2 ; next was the nickel chrome Open-Circuit Potentials Figure 3. Anodic polarizaiton curves of supraconstrutions/ti implant couples.

5 322 OH AND KIM Figure 4. Anodic polarization curves of dental alloys, supraconstructions, and spuraconstructions/ti implant couples. alloy couple with 1.27 A/cm 2. The cobalt chrome alloy couple exhibited a high current density of 5.60 A/cm 2, the silver palladium alloy couple was highest with 8.06 A/cm 2. Differences in the current density were detected for each alloy couple. At 0 mv (SCE), the titanium abutment couple and nickel chrome alloy couple exhibited low current densities of and A/cm 2, respectively, the silver palladium alloy couple A/ cm 2, the gold alloy and the cobalt chrome alloy couple and A/cm 2, respectively. Of these potentials, the lowest current density was exhibited at 0 mv (SCE), and the highest current density at 250 mv (SCE).

6 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 323 Figure 5. Open-circuit potentials of supraconstruction/ti implant couples. Surface Morphologies and Cross-Sectional Photographs In the surface photographs taken prior to the titanium abutment/titanium implant couple test, an extremely small marginal gap appeared. It appeared much smaller than the gap in the cast suprastructure/implant couple. In the surface photograph taken after the potentiodynamic polarization test and the potentiostatic test, a black spot was observed on the surface of the titanium fixture, as well as pitting in parts of the marginal gap (Figure 7). The surface of the gold alloy couple showed a clean surface, and the marginal gap appeared to be greater than in the titanium abutment couple. Crevice corrosion was also observed. In the potentiostatic test, a large pit was observed on the surface of the titanium fixture at 0 mv (SCE). Potentiodynamic and potentiostatic tests both produced black spots on the surface of the titanium fixture, visible by the naked eye (Figure 8). Before the silver palladium couple test, the surface appeared clean. However, after the completion of the potentiodynamic testing, corrosion was apparent, the surface turned black. In the case of potentiostatic testing at 250 and 0 mv (SCE), corrosion was observed around the marginal gap. At 250 mv (SCE), the marginal gap was sealed. In all of the photographs taken after tests, black spots were observed on the surface of the titanium fixture (Figure 9). In the surface photograph of the cobalt chrome alloy couple before the test, the titanium fixture had a clean surface. In both potentiodynamic and potentiostatic tests, corrosion of suprastructure were observed, while at 250 mv (SCE), the marginal gap was sealed. Fewer spots appeared on the surface of the titanium fixture than on the titanium abutment, gold alloy, or silver palladium alloy couples (Figure 10). In the surface photograph of the nickel chrome alloy couple, tarnishing and pitting appeared on the suprastructure after potentiodynamic testing. Spots formed on the surface of titanium fixture during the potentiodynamic and potentiostatic tests, but they were very small in size. However, in the case of potentiostatic test at 0 mv (SCE), large pits appeared on the surface of the suprastructure and on the titanium fixture (Figure 11). In the cross-sectional photograph taken to observe corrosion within the implant and the shape of the marginal gap, most of the suprastructure/titanium fixture gaps were observed to be in close contact, unlike what was suggested by the appearance of the surface photograph. Before testing the titanium couple, the gaps were observed to be in close contact, but after the potentiodynamic testing, light corrosion was observed in the marginal gap. Potentiostatic testing showed no corrosion (Figure 7). In the cross-sectional photograph taken before the test on the gold alloy couple, irregularities that were formed during the casting process were observed within the suprastructure and the marginal gaps were in close contact. After the potentiodynamic testing, corrosion had taken place on the internal wall of the suprastructure and corrosion product was observed. Less severe corrosion was caused by potentiostatic testing than by potentiodynamic testing. However, suprastructures were damaged by corrosion in all cases. Particularly at 250 mv (SCE), corrosion of the titanium fixture was also observed (Figure 8). In the crosssectional photograph before the silver palladium alloy couple test, the internal marginal gap appears relatively large and irregularities were observed on the cast. Potentiodynamic testing, and potentiostatic testing at 250 mv (SCE), produced corrosion on suprastructures (Figure 9). In the cross-sectional photograph of cobalt chrome alloy couple, corrosion was observed within the implant (Figure 10). The nickel chrome alloy couple and the cobalt chrome couple showed no corrosion inside the implant (Figure 11). DISCUSSION Galvanic current causes a local reduction in oxygen and increases metal ions in solution. The current changes according to various factors, such as the electrode potential, the degree of polarization, the area ratio of anode and cathode, the distance between the electrodes, the surface condition of electrode, conductivity, diffusion, stirring, path of galvanic current in the electrolyte, deaeration, temperature, ph, composition of electrolyte, etc. In addition to these factors, different corrosion mechanism, namely crevice corrosion, which occurs owing to the geometric form of the assembly, should be considered. Owing to the local exhaustion of dissolved oxygen and the increased concentration of metal ions in the solution, metal hydrates are formed, the ph decreases locally and Cl ions diffuse from the external saliva. This phenomenon is a necessary for the initiation and propagation of crevice corrosion. According to time, if the surrounding acidity increases, the passive film of the alloy is dissolved and corrosion of the film is accelerated. Because a galvanic couple has an influence on anodic constituents, E couple corr should be lower than the breakdown potential (E bd )ofthe anodic constituent. Because the oxygen concentration decreases inside the crevice, if the film of the anodic constituent is destroyed, repassivation of the film is difficult. In the worst cases, the ph may be less than 3.0. The ph in the oral cavity is diverse, varying from neutral to strongly acid, and the ph of the body fluid is , which

7 324 OH AND KIM Figure 6. Results of potentiostatic testing of supraconstructions/ti implant couples. means that a ph of 7.4 to 3.0 exists within the gap. According to the Nernst equation, a potential difference of 262 mv translates into ph 4.4. Therefore, the breakdown potential of the anodic constituent has to be considerably larger than E couple corr ( 262 mv). In the case of the specimens tested in this study, the gold alloy couple, silver palladium alloy couple, and titanium couple films were not destroyed, whereas the cobalt chrome and nickel chrome alloy couple films were destroyed at around 600 mv (SCE). It can thus be considered that there is no risk of galvanic corrosion in any of the alloy couples examined in this study. Nevertheless, nickel chrome alloy

8 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 325 Figure 7. Photographs of Ti abutment/ti implant couples. and cobalt chrome alloy couples can generate localized corrosion (pitting, and crevice corrosion) due to the existence of cyclic motion and implant gaps. Pure titanium and the titanium abutment exhibited passive current densities of 7.1 and 6.29 A/cm 2, respectively. The titanium abutment seemed to have a lower current density than pure titanium, because its surface had been polished. Even when the galvanic couple was formed, it appeared to be similar to the potentiodynamic polarization behavior of the pure titanium specimen and titanium abutment, and galvanic corrosion was not visible. Gold alloy and the suprastructure exhibited current densities of 2.54 and 0.15 A/cm 2, respectively. The current density of cast gold alloy reduced after casting. This is considered to be because of a change in composition of the gold alloy when it was cast. In other words, when cast by the centrifugal casting method, more gold ended up in the suprastructure than the sprue, because of the specific gravity difference. Also, in the case of the cobalt chrome alloy, the corrosion resistance was higher for the suprastructure than for the cast alloy, and when a galvanic couple was generated, the current density seemed to increase by a greater extent than the suprastructure. The current den-

9 326 OH AND KIM Figure 8. Photographs of gold alloy supraconstruction/ti implant couples. sity in this case seemed to increase because of the galvanic couple. Nickel chrome alloy exhibited similar potentiodynamic polarization behavior for the cast alloy, cast suprastructure, and galvanic couple. Nickel chrome alloy is an alloy, which has a relatively high corrosion resistance and seems to be little affected by galvanic corrosion. In the case of the Ag Pd alloy, the alloy before casting showed the lowest open-circuit potential, but the potentiodynamic polarization behavior of the cast suprastructure and galvanic couple appeared to be similar. This means that galvanic corrosion hardly occurred. In particular, the cast suprastructure showed the highest current density (9.24 A/cm 2 ). In the case of the Ag Pd alloy, such properties resulted because the melting temperature was in the narrow range of C and each of the specimens could not be cast uniformly; thus, cast defects formed during the casting process, which in turn, caused the corrosion resistance to decrease. The Co Cr and Ni Cr alloys exhibited pitting corrosion during the potentiodynamic polarization test. But because the potential in the oral cavity was lower than the potential at which pitting corrosion occurred, it was determined that the risk of the pitting corrosion was small.

10 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 327 Figure 9. Photographs of Ag Pd alloy supraconstruction/ti implant couples. In view of the galvanic corrosion, it can be considered that if a titanium abutment or gold alloy is coupled with a titanium fixture, galvanic corrosion may not take place. The Co Cr alloy couple exhibited a current density of 11.6 A/cm 2. This increased the possibility of galvanic corrosion, and galvanic corrosion also seemed to occur when an Ag Pd alloy crown, with a high corrosion resistance, forms a couple with the titanium implant, but is not given the due attention during the casting process or when crown has a defect in the cast. The nickel chrome alloy used in this study seems to have relatively stable corrosion resistance and casting attributes. The reason seems to be because the study used an excellent corrosion resistant alloy, and because it was cast using a high-frequency casting machine as recommended by the alloy manufacturer, unlike other alloys. Ravnholt and Jensen 11,12 reported that the galvanic current densities of titanium/precious alloy, titanium/cobalt chromium alloy, and titanium/stainless steel couples are zero under aerated conditions. The formation of a passive film is enhanced under aerated conditions, but CO 2 and NO 2 were found to destroy the stability of the passive film and cause corrosion. Therefore, the low galvanic current density mea-

11 328 OH AND KIM Figure 10. Photographs of Co Cr alloy supraconstruction/ti implant couples. sured in the present study was due to the fact that the test was performed under aerated conditions, and thus the galvanic corrosion was not accelerated. Moreover, generally, crevice corrosion is accelerated in the deaerated state, but in the present study, for the above stated reasons, it is considered that such crevice corrosion did not occur in almost all cases. Corrosion currents in the oral cavity may be perceived as metallic taste sensations. Föns 13 and Johansson et al. 14 reported that the threshold for eliciting such sensations is in the range of A/cm 2 (75 A/cm 2 mean), A/ cm 2 (190 A/cm 2 mean) in healthy test persons. The current densities of suprastructure and suprastructure/titanium implant couple were measured by potentiodynamic polarization and potentiostatic testing. Although the current density varied according to each alloy, they all demonstrated lower current densities than that required for eliciting such sensations in the mouth. Potentiodynamic polarization test of the suprastructure, showed a current density of A/cm 2 at 250 mv (SCE). In the case of the suprastructure/titanium implant couple, this was A/cm 2 at 250 mv (SCE), and the same couple in the potentiostatic test at 250 mv (SCE) produced a current density in the range of A/

12 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 329 Figure 11. Photographs of Ni Cr alloy supraconstruction/ti implant couples. cm 2. The maximum current density measured in this study was for a cobalt chromium alloy, which reached 11.6 A/ cm 2. This appears to be smaller than the reported minimum value of the threshold for eliciting such sensations. In other words, it is thought to fall short of an amount that could affect a sensitive person. However, this study was an in vitro study, which means it is a more simplified version than what may actually happen inside the mouth, and the possible clinical implications require more thought. In the case of the cobalt chromium alloy, the possibility of corrosion increases when the exposed area is not taken into full consideration during its exposure inside the electrolyte; this may entail not only ion release but also a reaction that can be sensed. Silver palladium alloy appears to undergo great changes in terms of electrochemical attributes, depending on the way it is cast. Therefore, prudence is required when using this alloy. In the case of cast alloys and galvanic couples, similar open-circuit potentials, whether high or low, were detected for each alloy. Gold alloy, the silver palladium alloy, and the titanium couple showed similar high open-circuit potentials; the nickel chromium alloy couple demonstrated low opencircuit potentials, and the cobalt chromium alloy couple re-

13 330 OH AND KIM corded the lowest open-circuit potential. The galvanic couple had lower open-circuit potential than the cast alloy. Perhaps this was due to the increased possibility of corrosion following the formation of a galvanic couple. In any case, it was difficult to provide accurate results because the surface area, surface polishing, and the geometric forms varied between the cast alloy and galvanic couple. The nickel chromium alloy and the cobalt chromium alloy couple maintained a relatively stable open-circuit potential. However, the gold alloy, silver palladium alloy, and the titanium couple showed a wide range of open-circuit potentials; the silver palladium alloy was the largest. The gold alloy and silver palladium alloy seemed to vary, depending on whether or not there were surface or internal defects in the cast. This seems to be the result of a deviation, which largely depends on the surface state and internal defects of the gold alloy and the silver palladium alloy casts. In the case of titanium, the deviation is also effected largely by the characteristics of the materials, such as the degree of surface finishing, the amount of impurities in the alloy, the size of the marginal gap, and the corrosion environment. In the case of the noble metal alloy, open-circuit potential generally has a positive value, remains within the immunity range, and reacts as a cathode, whereas titanium remains as an anode. When noble metal alloy and a titanium couple are integrated, titanium, which is the anode, dissolves according to the ratio of the i couple corr value calculated in accordance with the mixed potential theory. Base metal alloy has an open negative circuit potential, and if such a couple is integrated, the titanium will react as a cathode, and the base metal alloy will be dissolved according to the ratio of the i couple corr. In such a case, the titanium remains as a cathode. In general, the greater the difference between alloy and titanium s circuit potential, the greater the possibility of galvanic currents within the sensitive range of acute pain. During the potentiostatic testing of each suprastructure and titanium implant couple, all couples showed a negative current density at 250 mv (SCE), meaning that there was a reduction in the reaction at this potential. The absolute value of the current density is equivalent to the rate of the reduction reaction. A small absolute value of current density means that potential of the material is close to the redox potential. If the absolute value is high, it means the reduction reaction is accelerating at a great rate. In other words, when the cathodic current density is high, it becomes a noble metal. The titanium abutment couple and nickel chromium couple s current density showed a negative current density of as low as 10 3 A/cm 2, meaning that it produces a redox potential of around 250 mv (SCE). Such a galvanic couple is deemed less damaging in terms of galvanic corrosion. The gold alloy couple and silver palladium couple showed relatively high negative current densities of 3.18 and 6.63 A/cm 2, respectively, which shows that the two couples are noble metals. At 250 mv (SCE), the anodic activation or passive areas of the titanium abutment and gold alloy couples showed the lowest current density, and the nickel chromium alloy, cobalt chromium alloy, and silver palladium alloy couples exhibited high-current densities. The current densities that these alloys generated at 250 mv (SCE) were relatively low and stable, meaning that the risk of galvanic corrosion was very low. It is considered that the difference of the current densities of cast alloys, crown, and their galvanic couples results from their cast conditions, the marginal gap of the assembled implant, and so on. Corrosion resistance of the alloy itself and cast condition and the assembled condition of the implant determine the possibility of galvanic and crevice corrosion. For the applied potentials, the lowest current density was generated at 0 mv (SCE) and the highest at 250 mv (SCE). The redox potential in the oral cavity may change according to the measuring point, periodontal condition, and the quality and quantity of the implant. Therefore, it is desirable to maintain low redox potentials in the oral cavity. Before electrochemical testing, the titanium abutment was observed to have a clean surface, but the cast crown seemed to have an irregular surface. This seems to be the result of casting defects and the irregularity of the polishing during the manufacturing process of the crown of each alloy. In the surface figure, each of the gaps between the implant and the suprastructure were different. But the size of these gaps was within the range of those reported previously. The crosssectional figure shows that the gaps were in close contact. Close contact between the abutment and the suprastructure decreased the applied load of the abutment and the suprastructure. This phenomenon is important to obtain maximum efficiency, especially when restoring a single tooth. From the biological point of view, fitting stability is an important factor for successful tooth restoration. After electrochemical testing, specimens showed spots on the surface of the titanium abutment. Although the current density was low, surface corrosion occurred. It was considered that the localized corrosion occurred in the artificial saliva. The corroded part was observed in the photograph after the potentiodynamic polarization and potentiostatic testing at 250 mv (SCE). The edge around gaps was corroded, and the corrosion of the crown surface accelerated. In the cross-sectional figure, there was no trace of crevice corrosion. In the surface and cross-sectional photograph before and after the electrochemical test, there was low current generated by the electrochemical test. However, surface degradation and crevice corrosion were observed in several photographs, although it was considered that such corrosion products did not have the effect of sealing the marginal gap. In the case of dental amalgam, the corrosion product was accumulated at interfacial gap between the amalgam and the tooth, and had the effect of sealing the marginal gap. However, at the implant abutment interface, metal release occurred instead of an accumulation of corrosion product. CONCLUSIONS In the present study, we investigated if microbial penetration is prevented and the marginal fit of an implant fixture and suprastructure interface were sealed by galvanic corrosion, and by corrosion products in artificial saliva. We selected

14 SUPRASTRUCTURES GALVANICALLY COUPLED TO A Ti IMPLANT 331 noble and base metal alloys that could be used as implant suprastructures. 1. Titanium with alloy, abutment, and couple did not exhibit significant difference in passive current density and pitting. In the case of the gold alloy, the suprastructure showed the lowest current density, and in the cast alloy and couple, the current density appeared to be similar. Silver palladium alloy exhibited similar potentiodynamic polarization behavior. In the case of the cobalt chromium alloy, the alloy exhibited the highest passive current density, the suprastructure the lowest, and the couple medium: pitting occurred in all cases. The nickel chromium alloy showed similar potentiodynamic behaviors, and pitting for all cases. 2. Open-circuit potential of each of the suprastructure/titanium implant couples was generally maintained at a constant level with time. The silver palladium alloy, the titanium abutment, and the gold alloy couple had relatively high open-circuit potential, but among the alloy couples, there was no significant difference. Significant differences were found between the above-mentioned alloys (the silver palladium alloy, the titanium abutment, and the gold alloy couples) and the nickel chrome and the cobalt chrome alloy couples. The nickel chrome alloy couple had a low open-circuit potential and the cobalt chrome alloy couple the lowest. 3. All alloys exhibited a cathodic current density at 250 mv (SCE). The gold alloy and silver palladium alloy exhibited a high cathodic current density of 3.18 and 6.63 A/cm 2, respectively. At 250 mv (SCE), the titanium abutment exhibited the lowest current density ( A/cm 2 ), the gold alloy A/cm 2, the nickel chromium alloy 1.27 A/cm 2, the cobalt chromium alloy 5.60 A/cm 2, and the silver palladium alloy 8.06 A/cm 2. At 0 mv (SCE), the titanium abutment exhibited A/cm 2, the nickel chromium alloy A/cm 2, the silver palladium alloy A/cm 2, the gold alloy A/cm 2, and the cobalt chromium alloy A/cm After the electrochemical testing, black spots were observed on the surface of the titanium fixture, as well as pitting in parts of the marginal gap. Although the current density was low, surface corrosion occurred. It was considered that the localized corrosion occurred in the artificial saliva. The corroded part was observed mainly in the photograph after the potentiodynamic polarization and potentiostatic testing at 250 mv (SCE). It was considered that such corrosion products did not have the effect of sealing the marginal gap. The above results show that galvanic corrosion can occur in the Co Cr alloy, but this is determined to have low risk. In the other alloys, the possibility of corrosion and its risks appear lower. It can be confirmed that because of the structure of the implant, the risk of the crevice corrosion exists. Galvanic corrosion, in addition to crevice corrosion, increases the risk of corrosion, and thus various factors must be considered during alloy selection. The current density in the electrochemical test appeared lower than the current density in the oral cavity. But despite the low-current density, released ions may cause have side effects on the human body. Therefore, the selection of alloys is very important. REFERENCES 1. Geis-Gerstorfer J, Weber H, Sauer KH. In vitro substance loss due to galvanic corrosion in Ti implant/ni Cr supraconstruction systems. Int J Oral Maxillofac Implants 1989;4: Reclaru L, Meyer JM. Study of corrosion between a titanium implant and dental alloys. J Dent 1994;22: Venugopalan R, Lucas LC. Evaluation of restorative and implant alloys galvanically coupled to titanium. Dent Mater 1998; 14: Binon P, Weir D, Watanabe L, Walker L. Implant component compatibility. In: Laney WR, Tolman DE, editors. Tissue integration in oral orthopedic & maxillofacial reconstruction. Chicago: Quintessence Publishing; p Quirynen M, Bollen CML, Eyssen H, van Steenberghe D. Microbial penetration along the implant components of the Brånemark system. Clin Oral Implants Res 1994;5: Traversy MC, Birek P. Fluid and microbial leakage of implantabutment assembly in vitro. IADR ABCS 1992;71: Quirynen M, van Steenberghe D. Bacterial colonization of the internal part of two-stage implants. Clin Oral Implants Res 1993;4: Wahl G, Müller F, Schaal KP. Die mikrobielle Besiedlung von Implantatelementen aus Kunststoff und Titan. Schweiz Monaisschr Zahnmed 1992;102: Mairgüther R, Nentwig GH. Das Dichtigkeitsverhalten des Verbindungssystems beim zweiphasigen NM-Implantat. Z Zahnärzt Implantol 1992;8: Ben-Amar A, Cardash HS, Judes H. The sealing of the tooth/ amalgam interface by corrosion products. J Oral Rehabil 1995; 22: Ravnholt G. Corrosion current and ph rise around titanium coupled to dental alloys. Scand J Dent Res 1988;96: Ravnholt G, Jensen J. Corrosion investigation of two materials for implant supraconstructions coupled to a titanium implant. Scand J Dent Res 1991;99: Föns M. Elektrogustometri. Copenhagen: FADL s Forlag; p Johansson B, Stenman E, Bergman M. Clinical study of patients referred for investigation regarding so-called oral galvanism. Scand J Dent Res 1984;92: