The Release of Mercury from Dental Amalgam: The Mechanism and in vitro Testing M. MAREK School of Materials Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0245
Dissolution of mercury from the Ag-Hg matrix phase of dental amalgam in distilled water and synthetic saliva, and the mercury evaporation from the solutions, were studied in vitro. The main objectives of the study were to evaluate the factors that affect the results of the mercury release tests, and to consider the possible mechanisms of the release in vivo. Specimens were exposed to the liquids in open or closed bottles, and the changes in the mercury concentration were determined by coldvapor Atomic Absorption Spectrophotometry. Concentration vs. time tests showed the dissolution rate decreasing with time. Tests involving sequences of short and long exposures with solution changes showed higher average rates for short-term dissolution into the fresh solution than for the longer preceding exposures. The differences were attributed to a stifling effect of the concentration of elemental mercury on the dissolution. It is believed that mercury dissolved mainly in the elemental form and that a continuous increase in the concentration was made possible by oxidation in the solution. In open cells, some of the mercury was lost by evaporation. The analysis showed that the results of mercury dissolution tests depend on many test variables, such as time, solution volume, oxidation and evaporation conditions, etc. Evaporation, dissolution, and evaporation/dissolution mechanisms of the mercury release in vivo were considered. It was concluded that the dissolution/evaporation model best described the mercury release from dental amalgam restorations. J Dent Res 69(5):1167-1174, May, 1990
saliva has been measured in vitro (Frykholm, 1957; Mayer and Diehl, 1976; Brune, 1985; Brune and Evje, 1985) and also detected in vivo (Ott et al., 1984). Numerous laboratory studies of mercury dissolution into solutions simulating the oral liquids have been reported (Babendererde et al., 1970; Oyama, 1974; Kawahara et al., 1976, 1979, 1981; Takahashi et al., 1981; Kozono et al., 1982; Brune et al., 1983; Habu et al., 1983; Hero et al., 1983; Ohta, 1983; Brune and Evje, 1984; Brune, 1985; Brune and Evje, 1985; Ferracane et al., 1986; Ahmad and Stannard, 1987; Chung et al., 1987; Marek, 1987; Okabe et al., 1987; Ferracane et al., 1988a; Marek, 1988). The summarized results of some of these studies (Okabe et al., 1987) show an extremely wide range of the reported rates of the release. This data dispersion indicates that the techniques and conditions strongly affect the outcome of the tests. The main objectives of this study were to evaluate the factors that affect the results of the mercury release rate measurements in vitro, to consider the possible mechanisms of the release in vivo, and to make conclusions concerning the knowledge that is needed to understand the mercury release from dental amalgam restorations. The experimental work was limited to tests designed to demonstrate the effects rather than to obtain quantitative data. We performed the tests using a tinfree silver-mercury y1 phase to focus attention on the basic characteristics of the release process rather than on the differences between various dental amalgams, and to avoid, for the time being, the effects of the film formation caused by the presence of tin.
Introduction.
Materials and methods.
The possibility that mercury in dental amalgams may be a health hazard to patients has been debated since the introduction of this material to dentistry. In the most recent re-evaluation (Stanford et al., 1984), the general safety of dental amalgam has been asserted, but the need for further studies of both the mercury release and the physiological consequences has also been recognized. The most concrete evidence that mercury is being released from dental amalgam restorations has been obtained by analysis of the oral atmosphere (Gay et al., 1979; Svare et al., 1981; Abraham et al., 1984; Ott et al., 1984; Patterson et al., 1985; Vimy and Lorscheider, 1985a,b; Berglund et al., 1988). Evaporation of mercury from dry dental amalgam surfaces has been observed and studied in laboratories (Chan and Svare, 1972; Svare et al., 1973; Reynolds, 1974; Boyer et al., 1980; Boyer and Svare, 1986; Dhuru et al., 1986; Mitchell and Hay, 1987; Boyer, 1988; Ferracane et al., 1988b; Powell et al., 1988). Dissolution of mercury from dental amalgam into human
Amalgam specimens. -Most of the tests involving dissolution were performed with two specimens of the mercury-silver -y phase containing 36% Ag, 64% Hg by weight (Type A specimens). They were prepared by trituration of an ultrafine (2-3.5 Atm) silver powder of 99.9% purity with mercury of 99.998% purity (Cat. Nos. 00786 and 00520, respectively, Morton Thiokol, Inc./Alfa Products, Danvers, MA), coldpressing in a steel die at 40 MPa, and annealing for one month at 60'C. The specimens, which were in the form of disks 1.13 cm in diameter and 0.3 to 0.4 cm thick, were then maintained at 370C. For dissolution tests, the specimens were placed in recesses in PTFE stoppers, and the porosities and crevices were filled with epoxy (Part No. 811-161, Leco Corp., St. Joseph, MI). The exposed flat surfaces of the disks (1.0 cm2 geometric area) were wet-ground on silicon carbide papers through 600 grit, and the specimens were dry-air-annealed at 370C for 16 to 20 h before a test was started. For one set of experiments, the Type A specimens were no longer usable, and new Yi specimens were prepared with the same materials as above but through a modified procedure (Type B specimens). Trituration was performed after the capsule was heated to 600C, and the amalgamated mass was hotpressed at 60'C for 24 h. Annealing and storing procedures were the same as above. Type B specimens contained 68% Hg and 32% Ag by weight, and were 0.95 cm in diameter. They were cast in epoxy, which was machined to the shape of the
Received for publication July 27, 1989 Accepted for publication January 5, 1990 This investigation was supported in full by USPHS Research Grant DE-07754 from the National Institute of Dental Research, National Institutes of Health, Bethesda, MD 20892.
Downloaded from jdr.sagepub.com at Monash University on June 17, 2015 For personal use only. No other uses without permission.
1167
1168
MAREK
J Dent Res
PTFE stoppers used in the previous tests. These specimens wet-ground on silicon carbide papers through 2400 grit. Two identically prepared specimens were used in the tests. Test solution. -The tests were performed with use of distilled water and synthetic saliva. The composition of the synthetic saliva was as follows: KCl, 1.5 g/L; NaHCO3, 1.5 g/L; NaH2PO4'H20, 0.5 g/L; KSCN, 0.5 g/L; and lactic acid, 0.9 g/L. The composition (Marek and Topfl, 1986) is a modification of a formula proposed by Tani and Zucchi (1967). The pH of the solution was 6.7 to 6.8. The temperature of the liquids and specimens in all tests was 370C. In two sets of experiments, the test solution contained dissolved mercury. For the dissolution tests, the solution was obtained by exposure of liquid mercury to de-aerated distilled water in a 150-mL bottle for various periods of time under closed-cell conditions. At the end of the exposure, about 100 mL of the solution was withdrawn and briefly stirred; 24 mL was used for the test and the rest analyzed for the initial mercury concentration. For the evaporation tests, mercury was dissolved from a specimen of the yi phase (Type B) in either distilled water or synthetic saliva, by use of a similar procedure. Dissolution tests. -The dissolution vessels were glass weighing bottles with ground stopper joints. The set-ups are shown schematically in Fig. 1. For the closed-cell tests (Fig. la), the bottles (Cat. No. 03-415-5C, Fisher Scientific, Pittsburgh, PA) were first completely filled with the solution and placed in an incubator for temperature stabilization at 370C. The stoppers with the specimens (Type A for most tests), warmed to the same temperature, were then inserted, displacing some of the solution; no air bubble was left between the specimen and the solution. The volume of the solution was 11 to 12 mL and was accurately determined for each set-up. The vessels were turned upside down and kept in the incubator for the period of the test. For examination of the possible effect of evaporation of mercury from the liquid, tests were also performed with the
were
b
a
c
solution open to the atmosphere. The dissolution cells were weighing bottles of the same diameter as above, but longer (Cat. No. 03-415-5D, Fisher Scientific, Pittsburgh, PA), with the bottoms cut off (Fig. lb). In these tests, the stoppers with the specimens were inserted in the bottles, the bottles were placed in the incubator with the open end up, pre-warmed, and filled with 12 mL of the warm solution. The surface area of the liquid was 4.52 cm2, and the column of the liquid above the specimen was 2.65 cm high. In each test, the solution was analyzed for mercury at the end of the test period. In some tests, several exposures were performed consecutively; after one exposure, the specimens were quickly washed with distilled water, re-inserted into bottles filled with a fresh solution, and another exposure was started. The sequence of exposure times was as follows: one h, 24 h, one h, 72 h, and one h. Evaporation tests. -Evaporation from a mercury-containing solution, in the absence of a specimen, was determined with use of weighing bottles (Cat. No. 03-415-SC, Fisher Scientific, Pittsburgh, PA) without specimens (Fig. ic). Two pre-warmed bottles were filled with 12 mL of the mercury-containing solution, which was then analyzed after various periods of evaporation in an incubator at 370C. Analysis for mercury. -The mercury concentration following periods of dissolution or evaporation was determined by Atomic Absorption Spectrophotometry (AAS) by the cold-vapor technique. Samples were stabilized immediately following the exposure with use of a solution of 0.01% K2Cr2O7 in 5% HNO3. The dissolution bottles were washed with the same stabilizer, and the washings were included in the analyzed sample. The equipment consisted of an AA Spectrophotometer and a Cold-Vapor Generation Accessory (Models 1200 and VGA-76, respectively, Varian Techtron Pty. Limited, Mulgrave, Australia). The detection limit was 0.1 ppb.
Results. Dissolution tests. -The results of the analyses for mercury dissolved from Type A specimens in synthetic saliva following various test periods are shown in Fig. 2a. A curve was fitted to the data by nonlinear regression, by the following empirical
function:
Glass bottles
k.tn where c is the mercury concentration (ppb), t is the exposure time (h), and k and n are regression parameters. The curve was then numerically differentiated with respect to time after multiplication of each concentration with the average volume of the solution per test; the result is the curve in Fig. 2b showing the dissolution rate as a function of time. The results of the sequential exposure tests (Type A specimens), showing the average dissolution rates for each exposure (as well as the maximum dissolved mercury concentration), are presented in the Table, and the average dissolution rates are shown graphically in Figs. 3 and 4 for distilled water and synthetic saliva, respectively. The differences between the means of the average dissolution rates for each pair of conditions were evaluated by the Student's t test; the level of significance was p
Dissolution of mercury from the Ag-Hg matrix phase of dental amalgam in distilled water and synthetic saliva, and the mercury evaporation from the solutions, were studied in vitro. The main objectives of the study were to evaluate the factors that affect the results of the mercury release tests, and to consider the possible mechanisms of the release in vivo. Specimens were exposed to the liquids in open or closed bottles, and the changes in the mercury concentration were determined by coldvapor Atomic Absorption Spectrophotometry. Concentration vs. time tests showed the dissolution rate decreasing with time. Tests involving sequences of short and long exposures with solution changes showed higher average rates for short-term dissolution into the fresh solution than for the longer preceding exposures. The differences were attributed to a stifling effect of the concentration of elemental mercury on the dissolution. It is believed that mercury dissolved mainly in the elemental form and that a continuous increase in the concentration was made possible by oxidation in the solution. In open cells, some of the mercury was lost by evaporation. The analysis showed that the results of mercury dissolution tests depend on many test variables, such as time, solution volume, oxidation and evaporation conditions, etc. Evaporation, dissolution, and evaporation/dissolution mechanisms of the mercury release in vivo were considered. It was concluded that the dissolution/evaporation model best described the mercury release from dental amalgam restorations. J Dent Res 69(5):1167-1174, May, 1990
saliva has been measured in vitro (Frykholm, 1957; Mayer and Diehl, 1976; Brune, 1985; Brune and Evje, 1985) and also detected in vivo (Ott et al., 1984). Numerous laboratory studies of mercury dissolution into solutions simulating the oral liquids have been reported (Babendererde et al., 1970; Oyama, 1974; Kawahara et al., 1976, 1979, 1981; Takahashi et al., 1981; Kozono et al., 1982; Brune et al., 1983; Habu et al., 1983; Hero et al., 1983; Ohta, 1983; Brune and Evje, 1984; Brune, 1985; Brune and Evje, 1985; Ferracane et al., 1986; Ahmad and Stannard, 1987; Chung et al., 1987; Marek, 1987; Okabe et al., 1987; Ferracane et al., 1988a; Marek, 1988). The summarized results of some of these studies (Okabe et al., 1987) show an extremely wide range of the reported rates of the release. This data dispersion indicates that the techniques and conditions strongly affect the outcome of the tests. The main objectives of this study were to evaluate the factors that affect the results of the mercury release rate measurements in vitro, to consider the possible mechanisms of the release in vivo, and to make conclusions concerning the knowledge that is needed to understand the mercury release from dental amalgam restorations. The experimental work was limited to tests designed to demonstrate the effects rather than to obtain quantitative data. We performed the tests using a tinfree silver-mercury y1 phase to focus attention on the basic characteristics of the release process rather than on the differences between various dental amalgams, and to avoid, for the time being, the effects of the film formation caused by the presence of tin.
Introduction.
Materials and methods.
The possibility that mercury in dental amalgams may be a health hazard to patients has been debated since the introduction of this material to dentistry. In the most recent re-evaluation (Stanford et al., 1984), the general safety of dental amalgam has been asserted, but the need for further studies of both the mercury release and the physiological consequences has also been recognized. The most concrete evidence that mercury is being released from dental amalgam restorations has been obtained by analysis of the oral atmosphere (Gay et al., 1979; Svare et al., 1981; Abraham et al., 1984; Ott et al., 1984; Patterson et al., 1985; Vimy and Lorscheider, 1985a,b; Berglund et al., 1988). Evaporation of mercury from dry dental amalgam surfaces has been observed and studied in laboratories (Chan and Svare, 1972; Svare et al., 1973; Reynolds, 1974; Boyer et al., 1980; Boyer and Svare, 1986; Dhuru et al., 1986; Mitchell and Hay, 1987; Boyer, 1988; Ferracane et al., 1988b; Powell et al., 1988). Dissolution of mercury from dental amalgam into human
Amalgam specimens. -Most of the tests involving dissolution were performed with two specimens of the mercury-silver -y phase containing 36% Ag, 64% Hg by weight (Type A specimens). They were prepared by trituration of an ultrafine (2-3.5 Atm) silver powder of 99.9% purity with mercury of 99.998% purity (Cat. Nos. 00786 and 00520, respectively, Morton Thiokol, Inc./Alfa Products, Danvers, MA), coldpressing in a steel die at 40 MPa, and annealing for one month at 60'C. The specimens, which were in the form of disks 1.13 cm in diameter and 0.3 to 0.4 cm thick, were then maintained at 370C. For dissolution tests, the specimens were placed in recesses in PTFE stoppers, and the porosities and crevices were filled with epoxy (Part No. 811-161, Leco Corp., St. Joseph, MI). The exposed flat surfaces of the disks (1.0 cm2 geometric area) were wet-ground on silicon carbide papers through 600 grit, and the specimens were dry-air-annealed at 370C for 16 to 20 h before a test was started. For one set of experiments, the Type A specimens were no longer usable, and new Yi specimens were prepared with the same materials as above but through a modified procedure (Type B specimens). Trituration was performed after the capsule was heated to 600C, and the amalgamated mass was hotpressed at 60'C for 24 h. Annealing and storing procedures were the same as above. Type B specimens contained 68% Hg and 32% Ag by weight, and were 0.95 cm in diameter. They were cast in epoxy, which was machined to the shape of the
Received for publication July 27, 1989 Accepted for publication January 5, 1990 This investigation was supported in full by USPHS Research Grant DE-07754 from the National Institute of Dental Research, National Institutes of Health, Bethesda, MD 20892.
Downloaded from jdr.sagepub.com at Monash University on June 17, 2015 For personal use only. No other uses without permission.
1167
1168
MAREK
J Dent Res
PTFE stoppers used in the previous tests. These specimens wet-ground on silicon carbide papers through 2400 grit. Two identically prepared specimens were used in the tests. Test solution. -The tests were performed with use of distilled water and synthetic saliva. The composition of the synthetic saliva was as follows: KCl, 1.5 g/L; NaHCO3, 1.5 g/L; NaH2PO4'H20, 0.5 g/L; KSCN, 0.5 g/L; and lactic acid, 0.9 g/L. The composition (Marek and Topfl, 1986) is a modification of a formula proposed by Tani and Zucchi (1967). The pH of the solution was 6.7 to 6.8. The temperature of the liquids and specimens in all tests was 370C. In two sets of experiments, the test solution contained dissolved mercury. For the dissolution tests, the solution was obtained by exposure of liquid mercury to de-aerated distilled water in a 150-mL bottle for various periods of time under closed-cell conditions. At the end of the exposure, about 100 mL of the solution was withdrawn and briefly stirred; 24 mL was used for the test and the rest analyzed for the initial mercury concentration. For the evaporation tests, mercury was dissolved from a specimen of the yi phase (Type B) in either distilled water or synthetic saliva, by use of a similar procedure. Dissolution tests. -The dissolution vessels were glass weighing bottles with ground stopper joints. The set-ups are shown schematically in Fig. 1. For the closed-cell tests (Fig. la), the bottles (Cat. No. 03-415-5C, Fisher Scientific, Pittsburgh, PA) were first completely filled with the solution and placed in an incubator for temperature stabilization at 370C. The stoppers with the specimens (Type A for most tests), warmed to the same temperature, were then inserted, displacing some of the solution; no air bubble was left between the specimen and the solution. The volume of the solution was 11 to 12 mL and was accurately determined for each set-up. The vessels were turned upside down and kept in the incubator for the period of the test. For examination of the possible effect of evaporation of mercury from the liquid, tests were also performed with the
were
b
a
c
solution open to the atmosphere. The dissolution cells were weighing bottles of the same diameter as above, but longer (Cat. No. 03-415-5D, Fisher Scientific, Pittsburgh, PA), with the bottoms cut off (Fig. lb). In these tests, the stoppers with the specimens were inserted in the bottles, the bottles were placed in the incubator with the open end up, pre-warmed, and filled with 12 mL of the warm solution. The surface area of the liquid was 4.52 cm2, and the column of the liquid above the specimen was 2.65 cm high. In each test, the solution was analyzed for mercury at the end of the test period. In some tests, several exposures were performed consecutively; after one exposure, the specimens were quickly washed with distilled water, re-inserted into bottles filled with a fresh solution, and another exposure was started. The sequence of exposure times was as follows: one h, 24 h, one h, 72 h, and one h. Evaporation tests. -Evaporation from a mercury-containing solution, in the absence of a specimen, was determined with use of weighing bottles (Cat. No. 03-415-SC, Fisher Scientific, Pittsburgh, PA) without specimens (Fig. ic). Two pre-warmed bottles were filled with 12 mL of the mercury-containing solution, which was then analyzed after various periods of evaporation in an incubator at 370C. Analysis for mercury. -The mercury concentration following periods of dissolution or evaporation was determined by Atomic Absorption Spectrophotometry (AAS) by the cold-vapor technique. Samples were stabilized immediately following the exposure with use of a solution of 0.01% K2Cr2O7 in 5% HNO3. The dissolution bottles were washed with the same stabilizer, and the washings were included in the analyzed sample. The equipment consisted of an AA Spectrophotometer and a Cold-Vapor Generation Accessory (Models 1200 and VGA-76, respectively, Varian Techtron Pty. Limited, Mulgrave, Australia). The detection limit was 0.1 ppb.
Results. Dissolution tests. -The results of the analyses for mercury dissolved from Type A specimens in synthetic saliva following various test periods are shown in Fig. 2a. A curve was fitted to the data by nonlinear regression, by the following empirical
function:
Glass bottles
k.tn where c is the mercury concentration (ppb), t is the exposure time (h), and k and n are regression parameters. The curve was then numerically differentiated with respect to time after multiplication of each concentration with the average volume of the solution per test; the result is the curve in Fig. 2b showing the dissolution rate as a function of time. The results of the sequential exposure tests (Type A specimens), showing the average dissolution rates for each exposure (as well as the maximum dissolved mercury concentration), are presented in the Table, and the average dissolution rates are shown graphically in Figs. 3 and 4 for distilled water and synthetic saliva, respectively. The differences between the means of the average dissolution rates for each pair of conditions were evaluated by the Student's t test; the level of significance was p
