dental materials Dental Materials 19 (2003) 232±239 www.elsevier.com/locate/dental Evaluation of the long-term corrosion behavior of dental amalgams: in¯uence of palladium addition and particle morphology Pierre Colon a,b,*, Nelly Pradelle-Plasse a, Jacques Galland b a Laboratoire de BiomateÂriaux Dentaires, Department of Restorative Dentistry and Endodontics, Universite Denis Diderot, Service d'Odontologie GarancieÁre, HoÃtel Dieu, Paris, France b Laboratoire CFH, Ecole Centrale Paris, Paris, France Received 17 April 2001; revised 14 November 2001; accepted 11 December 2001 Abstract Objectives: The purpose of this investigation was to evaluate the long-term corrosion behavior of experimental amalgams as a function of particle morphology and palladium content. Methods: Samples of four experimental high copper amalgams were prepared according to ADA speci®cations. Two of them had the same chemical composition but one had lathe cut particles (LCP) and the other had spherical particles (SP). The two others had spherical powders with an addition of 0.5 wt% of palladium (SP 0.5) and 1 wt% of palladium(SP 1) for the other. Corrosion resistance was evaluated by electrochemical techniques in Ringer's solution in a thermostated cell at 37 8C for samples aged 5, 8, 12, 16 months and 10 years. Potentiokinetic curves were drawn and the potential and the current density corresponding to the ®rst anodic peak were registered. Results: For all the amalgam samples the corrosion behavior improves over the 10-year period. SP samples exhibit a better behavior than LCP. Palladium addition improves corrosion behavior as compared to samples without palladium. No real difference is found regarding the amount of palladium between 0.5 and 1%. The potentials progress from a range between 0 and 20 mV/SCE to a range of 60±80 after 10 years. The stabilization of the potential begins after only 16 months. Except for the LCP, all the values converge to the same level of 80 mV/ SCE. Signi®cance: The addition of no more than 0.5 wt% Palladium in a high copper amalgam powder improves the corrosion behavior of the amalgam up to a period of 10 years. The potential of the ®rst anodic peak increases for each amalgam, probably in relation to the evolution of the structure of the material. Clinically, it is of interest to consider the good electrochemical behavior of older restorations when contemplating the repair or replacement of such ®llings. At the same time, galvanic current can occur when a new amalgam restoration is placed in contact with an old one even if the same amalgam is used. In this situation, the new ®lling will be anodic and its degradation will be accelerated. The evaluation of the corrosion behavior of dental amalgams has to take into account the age of the samples. q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. Keywords: Dental materials; Amalgam alloy; Corrosion; Long-term behavior; In vitro; Palladium 1. Introduction Dental amalgams are discussed according to their mercury content and release [1±3]. However, amalgam is a ®lling material which is still widely used because of its cost, durability and ease of manipulation [4,5]. In the US, approximately, 96 million amalgam restorations were placed in 1991 [6]. Corrosion is the primary degradation process dental * Corresponding author. Address: 5, rue Garanciere, 75 006 Paris, France. Tel.: 133-1-53-10-50-10; fax: 133-1-53-10-50-11. E-mail address: colon@ccr.jussieu.fr (P. Colon). amalgam undergoes [7,8]. Since the longevity of the restoration is put forth as an argument in its defense [5,9,10], a consideration of its corrosion behavior as a function of time could be of interest. The setting reaction of high copper amalgam leads to the formation of two new phases: [11] the silver±mercury gamma 1 phase, and the copper±tin eta-prime phase, the most corrodable phase in high copper amalgam. Some transformation from the gamma 1 phase to the more thermodynamically stable beta 1 phase [12] can occur. The composition of tin in the different phases is modi®ed as a function of time [13]. Evidence exists that low rates of mercury are released 0109-5641/03/$30.00 + 0.00 q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S01 09- 5641(02)0003 5-0 233 P. Colon et al. / Dental Materials 19 (2003) 232±239 Table 1 Tested amalgams Materials LCP SP SP 0.5 SP 1 Batch number 3282 3727 3715 3714 Particle morphology Lathe cut Spherical Spherical Spherical from aged dental amalgam restorations [14±20]. Powder morphology, especially the use of spherical particles, is well known as a way of reducing the mercury ratio. Efforts have been made to reduce this mercury emission by altering the alloy ratio [21,22], and by trituration of the alloy with binary mercury/indium [23,24]. Mahler et al. studied the in¯uence of tin associated with the gamma 1 phase on the vapor pressure of gamma 1 and on the release of mercury [25]. More recently, addition of palladium into the amalgam powder has been shown to be an effective way to reduce mercury release [26]. Mercury release from dental amalgam decreases as a function of amalgam age during the ®rst 30 days. [27]. This result could occur in relation to microstructural changes associated with the solid state transformation reported by Marshall and Marshall [28]. It would, therefore, be of interest to consider the preservation of some clinically acceptable older restorations, in view of the low mercury release rates in such restorations. The addition of palladium has been reported to have a positive in¯uence on the corrosion behavior of dental amalgams [29,30]. Corrosion behavior is often evaluated on samples of amalgams without information about their age. Studies concerning the improvement of corrosion resistance obtained by palladium addition are also short-term studies. The aim of this electrochemical study is to evaluate, in vitro, based on an experimental single formulation of a high copper amalgam powder: ² the in¯uence of particle morphology ² the in¯uence of addition of 0.5 and 1% palladium on the corrosion behavior ² the evolution of this corrosion behavior on a long-term clinically signi®cant period extending over 10 years. 2. Materials and methods Powder composition (wt%) Hg (mass%) Ag Sn Cu Pd 45.9 45.9 45.74 45.56 31.4 31.4 31.28 31.07 22.7 22.7 22.48 22.37 0 0 0.5 1 44.44 41.48 42.86 42.90 particles with differing ratios of palladium. Their chemical compositions are represented in Table 1. The amalgams were triturated in an amalgamator (Silamat, Ivoclar, Schann, Liechtenstein) for 8 s and mechanically condensed into cylindrical specimens (4.0 mm diameter X 5.0 mm height) in a steel mold under 14 MPa of pressure by following the procedure outlined in ADA Speci®cation No. 1. The mass% of residual Hg in each amalgam was determined by collecting and weighing the mercury that was expressed during condensation and subtracting this amount from the amount originally used in the mix. Samples were embedded in an epoxy resin (Epo®x, Struers, Copenhagen, Denmark). The interface amalgam/ resin was covered with an electrically insulating lacquer (Plastik 70, Kontakt Chemie, Rastatt, Germany) because it was the best way to prevent crevice corrosion at the interface between the sample and the resin. An internal contact was obtained with silver lacquer and an electrical cable before embedding. Samples were stored in air at room temperature for periods of 5, 8, 12, 16 months and 10 years. All the samples were prepared 10 years ago, and then tested after each time period. The test conditions were the same over these 10 years, particularly the new reference electrodes (SCE) were controlled are new Ringer's solutions were elaborated before each series of tests. The standard surface preparation consisted of a metallographic polish obtained with a 1200 Si±C grit paper, ultrasonic cleaning and washing with distilled water just prior to immersion in electrolyte solution. 2.2. Electrochemical test All tests were performed using a double-walled glass polarization cell (500 cm 3) with three electrodes, temperature controlled at 37 8C with a thermostat. The working electrode was placed in the same place for each experiment. 2.1. Amalgams Table 2 Ringer's solution characteristics The four experimental powders were prepared by a French manufacturer, Dentoria, in order to obtain the same chemical composition, but differing in spherical or lathe cut morphology, and to obtain the same spherical Composition (g/l) pH Ionic force (mol l 21) Ohmic drop (V) 7.4 17.04 £ 10 22 NaCl:9.15; KCl:0.42; CaCl2:0.24; NaHCO3:0.15 58 234 P. Colon et al. / Dental Materials 19 (2003) 232±239 between the different samples was not statistically signi®cant. 3. Results Fig. 1. Schematic electrical set-up. The reference electrode was a saturated calomel electrode (SCE) (XR 140, Tacussel Radiometer Analytical, Copenhagen, Denmark). The auxiliary electrode was a platinum electrode (surface 1 cm 2) (XR 110, Tacussel Radiometer Analytical, Copenhagen, Denmark). The electrolyte used was Ringer's solution (Table 2) at 37 8C de-aerated for 20 min with azote. Stability of pH and good conductibility were obtained and the ohmic fall phenomena was negligible in the range of potentials studied. The schematic electrical set-up used to measure the electrochemical parameters is shown in Fig. 1. The cell was connected to potentiostat (PGS201 T, Tacussel Radiometer Analytical, Copenhagen, Denmark), monitored by a microprocessor and speci®c software (Voltamaster, Tacussel Radiometer Analytical, Copenhagen, Denmark). An in-vitro potentiodynamic polarization test was employed to evaluate the electrochemical behavior of all the samples. The specimens were ®rst immersed for 45 min, and the open circuit potential was then registered. Anodic curves at a rate of 5 mV min 21 were scanned from the imposed value of the open circuit potential measurement to the potential of 1400 mV versus SCE. Two samples of each material are tested and for each sample, the measurements are twice conducted in order to insure the reproducibility of the method. However, the evaluation of the corrosion behavior focused on the values of the potential and of the current density corresponding to the ®rst anodic peak. Statistical analysis was not performed on the data because the difference in the values of potentials and intensities Table 3 Open circuit potentials after 45 min Amalgam 5 months (mV/SCE) 10 years (mV/SCE) LCP SP SP0.5 SP1 2440 2480 2350 2320 2300 2290 2280 2250 The open circuit potentials at 5 months and 10 years are shown on Table 3. For each specimen open circuit potential increases after 10 years. When comparing the values between the different amalgams, the data range decreases after 10 years. Potentiokinetic curves of different amalgams for specimens 5 months old are shown on Fig. 2. The current density increases for the LCP sample at a lower potential value than for all the other spherical amalgams. With regard to the current density of the ®rst anodic peak, two groups can be analyzed. The ®rst one is the SP and LCP group. The current density of the ®rst peak is similar but the current increases for the LCP sample at a lower value of potential than for the SP one. The second group is the SP0.5 and SP1 group. The current density of the ®rst peak is also quite similar but the values of currents are lower than for the LCP and SP groups. No visible modi®cation can be found for the second peak. The SP sample exhibits better behavior than LCP, and palladium addition improves corrosion behavior. Potentiokinetic curves of different amalgams, as a function of time, are shown in Figs. 3±6. For each amalgam, the different curves are drawn on the same graph in order to evaluate the evolution of the corrosion behavior. For all the amalgam samples, the corrosion behavior improves over the 10 years elapsed since the ®rst anodic peak moved in the direction of increasing potentials. Fig. 7 shows the evolution of potentials of the ®rst anodic peak versus time for the different amalgams while Fig. 8 presents the values of corresponding current density. The potentials progress from a range between 0 and 20 mV/SCE to a range of 60±80 mV/SCE after 10 years. The stabilization of the potential begins after 16 months. Except for the LCP, all the values converge to the same level of 80 mV/SCE. Regarding the current density values of the SP sample versus the LCP, results are similar for 5 months and for 10 years but lower values of the SP are noted for the other periods. Current density values between SP 0.5 and SP 1 are almost similar but about two times less than LCP and SP samples. 4. Discussion The current density at the ®rst anodic peak appears to be a good method for studying the corrosion behavior of dental amalgams. High rates of degradation occur for this potential value but the range of potential is beyond the open circuit P. Colon et al. / Dental Materials 19 (2003) 232±239 Fig. 2. Potentiokinetic curves at 5 months. Fig. 3. Evolution of the potentiokinetic curves as a function of time sample LCP. 235 236 P. Colon et al. / Dental Materials 19 (2003) 232±239 Fig. 4. Evolution of the potentiokinetic curves as a function of time sample SP. potential data. However, in oral situations, galvanic situation with other metallic restorations are able to reach this range of potential values (approximately 140 mV/SCE) as demonstrated by previous studies [31,32]. Otherwise, the range of potential examined for testing dental amalgams has been suggested from 0 to 100 mV (SCE) [33]. Analysis of the potentiokinetic curves versus time shows an evolution of the potential of this peak over a 10-year period. This evaluation could be a good approach in determining the long-term behavior of each amalgam. Fig. 5. Evolution of the potentiokinetic curves as a function of time sample SP0.5. P. Colon et al. / Dental Materials 19 (2003) 232±239 237 Fig. 6. Evolution of the potentiokinetic curves as a function of time sample SP1. The potential of the ®rst anodic peak probably increases in relation to the evolution of the metallurgical structure of the material. Previous studies [12,13] have shown the transformation in the composition of phases. These new phases exhibit better corrosion resistance and could explain the evolution of the electrochemical curves. At the same time, some porosities could be closed by the growth of crystals of new phases (Cu±Sn and Ag±Hg or Ag±Hg±Sn). The longevity of amalgam restorations could be partly explained by this improvement of corrosion resistance. Fig. 7. Potentials of the ®rst anodic peak versus time. 238 P. Colon et al. / Dental Materials 19 (2003) 232±239 Fig. 8. Current densities of the ®rst anodic peak versus time. Spherical particles are a good means of reducing, at the same time, mercury content [21,34] and the corrosion process. The addition of no more than 0.5 wt% Palladium in a high copper amalgam powder improves the corrosion behavior of the amalgam over a long-term study period from 5 months to 10 years. Since Palladium also reduces mercury release [26], the addition of this element is certainly of interest. Clinically it is also interesting to consider the good electrochemical behavior of older restorations when contemplating the repair or replacement of such ®llings. Since the open circuit potential registered on the samples 5 months old ranged from 2480 to 2320 mV/SCE and after 10 years from 2300 to 2250 mV/SCE, galvanic interaction between fresh and old amalgams could occur even if the same amalgam is used. Previous studies have demonstrated the existence of this galvanic current regardless the age of the amalgam [35]. In this situation, the new ®lling will be anodic and its degradation will be accelerated. The evaluation of the corrosion behavior of dental amalgams must take into account the age of the samples. This parameter may have the same in¯uence on the electrochemical behavior as the composition of different high copper amalgams, since an increase in potential of 60 mV over 10 years has been observed. Acknowledgements The authors are grateful to the manufacturer, Dentoria, for the elaboration of experimental amalgam powders. References [1] Abraham JE, Svare CW, Franck CW. The effect of dental amalgam restorations on blood mercury levels. J Dent Res 1984;63:71±3. [2] Ekstrand J, BjoÈrkman L, Edlund C, Sandborgh-Englung G. Toxicological aspects on the release and systemic uptake of mercury from dental amalgam. Eur J Oral Sci 1998;106:678±86. [3] Vimy MJ, Lorscheider EL. Intra-oral air mercury release from dental amalgam. J Dent Res 1985;64:1069±71. [4] Berry TG, Summitt JB, Chung AKH, Osborne JW. Amalgam at the new millenium. J Am Dent Assoc 1998;129:1517±56. [5] Leinfelder KF. Dental amalgam alloys. Curr Opinion Dent 1991;1:214±7. [6] Association Report. Dental amalgam: update on safety concerns. J Am Dent Assoc 1998;129:494±503. [7] Eley BM. The future of dental amalgam: a review of the literature. Part 1: dental amalgam structure and corrosion. Br Dent J 1997;182:247±9. [8] Milleding P, Wennberg A, Hasselgren G. Cytotoxicity of corroded and non corroded dental silver amalgams. Scand J Dent Res 1985;93:76±83. [9] Roulet JF. Bene®ts and disadvantages of tooth-coloured alternatives to amalgam. J Dent 1997;25:459±73. [10] Mjor IA. Problem and bene®ts associated with restorative material: side effects and long-term cost. Adv Dent Res 1992;6:7±16. [11] Marshall SJ, Marshall GW. Dental amalgam: the materials. Adv Dent Res 1992;6:94±9. [12] Marshall SJ, Marshall GW, Letzel JH. Gamma 1 to beta 1 phase transformation in retrieved clinical amalgam restoration. Dent Mater 1992;8:162±6. [13] Boyer DB, Edie JW. Composition of clinically aged amalgam restorations. Dent Mater 1990;6:146±50. [14] Ferracane JL, Adey JD, Nakajima H, Okabe T. Mercury vaporization from amalgams with varied alloy compositions. J Dent Res 1995;74:1414±7. P. Colon et al. / Dental Materials 19 (2003) 232±239 [15] Haikel Y, Gasser P, Salek P, Voegel JC. Exposure to mercury vapor during setting, removing and polishing amalgam restorations. J Biomed Mater Res 1990;24:1551±8. [16] Heintze U, Edwardsson S, Derand T, Birkhed D. Methylation of mercury from dental amalgam and mercuric chloride by oral streptococci in vitro. Scand J Dent Res 1983;91:150±2. [17] Ahmad R, Stannard JG. Mercury release from amalgam: a study in vitro and in vivo. Oper Dent 1990;15:207±18. [18] Berglund A. Estimation by a 24 hour study of the daily dose of intraoral mercury vapor inhaled after release from amalgam. J Dent Res 1990;69:1646±51. [19] Berglund A. An in vitro and in vivo study of the release of mercury vapor from different types of amalgam alloys. J Dent Res 1993;72:939±46. [20] Halbach S. Combined estimation of mercury species released from amalgam. J Dent Res 1995;74:1103±9. [21] Bracho-Troconis C, Colon P, Bartout JD, Bienvenu Y. In¯uence of thermal treatments on Ag Sn Cu powders in order to reduce mercury contents in dental amalgam. J Mater Sci: Mater Med 2000;11:1±9. [22] Eichmiller FC. Promising new dental materials on the horizon. Compendium 1997;18:254±69. [23] Washimi H. Optical measurement of mercury activity in an In±Hg alloy. Jpn J Appl Phys 1969;8:1268. [24] Okabe T, Yamashita T, Nakajima H, Berglund A, Zhao L, Guo I, Ferracane JL. Reduced mercury vapor release from dental amalgams prepared with binary Hg±In liquid alloys. J Dent Res 1994;73:1711±6. [25] Mahler DB, Adey JD, Fleming MA. Hg emission from dental amal- [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] 239 gam as related to the amount of Sn in the Ag±Hg (g1) phase. J Dent Res 1994;73:1663±8. Neme AL, Wagner WC, O'Brien WJ. Effects of palladium addition on emission of mercury vapor from dental amalgam. Dent Mater 1999;15:382±9. Berdouses E, Vaidyanathan TK, Dastane A, Weisel C, Houpt M, Shey Z. Mercury release from dental amalgams: an in vitro study under controlled chewing and brushing in an arti®cial mouth. J Dent Res 1995;74:1185±93. Marshall SJ, Marshall GW. Time dependant phase change in Cu rich amalgams. J Biomed Mater Res 1979;13:395±406. Mahler DB, Engle JH, Adey JD. Effect of Pd on the clinical performance of amalgam. J Dent Res 1990;69:1759±61. Chern Lin JH, Greener EH. Microstructures of Pd-containing dispersants for admixed dental amalgams. Dent Mater 1991;7:254±7. Momoi Y, Asanuma A, Kohno A, Yanagisawa K. A measurement of galvanic current and electrical potential in extracted human teeth. J Dent Res 1986;65:1441±4. Sutow EJ, Jones DW, Hall GC. Correlation of dental amalgam crevice corrosion with clinical ratings. J Dent Res 1989;68:82±8. Marek M. Corrosion test for dental amalgam. J Dent Res 1980;59:63±9. Colon P, Para-Morice B. In¯uence de la morphologie des particules sur la reÂsistance aÁ la corrosion d'un amalgame HCSC. J BiomateÂr Dent 1989;4:171±6. Ravnholt G, Holland RI. Corrosion current between fresh and old amalgam. Dent Mater 1988;4:251±4.