Filler leachability during water storage of six composite materials
KARL-JOHAN M, SOOERHOLM Department of Denial Biomalerials, C&llege of Dentistry, University of Florida, Gainesville, Florida, USA
Soderholm K-JM: Filler leachability during water storage of six composite materials, Scand J Dent Res 1990; 98: 82-8, Abstract — Sis diflferent dental composite n:iaterials were investigated regarding leaching of filler elements such as silicon, barium,, and zinc. The leaching was conducted by storing the samples in distilled water at 60^C for half a year. The results could not prove that the leaching behavior of filler elements in most of tbe investigated cases decreases with time. The practical implications. of this study could be important. If other elements follow the same leaching pattern, therapeutic elements such as fluoride could be incorporated in filler particles. The tendency of composites to leach filler elements almost linearly wilh time, couJd be used lo generate a cotistani release rate of such therapeutic elements over time. Such an application could have a major impact on controlling caries adjacent to composite restorations and sealants. Key words: degradation; dental materials; hydrolysis; resins. Department of Dental Biomaterials, College of Dentistry, Box J-446, Gainesville, Florida 32610, USA, Accepted for publication 20 June 1989,
Previous studies have shown that water improved the wear properties of cotnposite leaches filler elements from dental composite materials (7—9), Furthermore, these glasses tnaterials (1, 2), This leaching process is ae- possess some degree of X-ray opacity which centuated when the silica network has been is a desirable property for a composite, modified with metal oxides such as barium Since filler particles consisting of glasses and strontium oxide to form different glasses possess important advantages, it is important (2, 3), Although these glasses are susceptible to determine how significant the hydrolytic to hydrolytic degradation, they also possess degradation of these less stable filler particles some advantages (4—6), Because they are is for material properties. Broken down filler softer than quartz, finer filler particles can be surfaces will increase the risk for filler-matrix made during filler processing and a smoother debonding, decrease strength, and reduce surface finish can be produced on composite the wear resistance (10—13), Therefore, quesrestorations. This reduction in particle size tions on the leaching characteristics of difand improvement of stirface texture may ex- ferent composites should be addressed since plain why the use of these glasses may have such properties could correlate with fracture
83
FILLER LEACHABILITY DURING WATER STORAGE Table I Jnuestigalsd materials
Material
Symbol
Batch No.
A
4GI
B C D E
240284 0426841 40442 UP 4142E GP 4iO7E 4PR1P
Silux Heliosit Prisma-Fil Command Ultrafine Prestige Plus
E
Manufacturer 3M, St Paul, MN, USA Vivadent, Liechtenstein The L. D. Caulk Company, Milford, DE, USA Kerr, Romulus, MI, USA Lee Pharmaceuticals, El Monte, CA, USA 3M, St Paul, MN, USA
and wear hehavior of composites when used clinically (14). The ohjective of this study was to determine how the leaching pattern changes over time when composite materials are stored in an aqueous environment. The working hypothesis was that the leaching rate in this type of environment would level out and reach a constant value within 6 months. Material and methods Six composite materials were investigated (Table 1). Five specimens of each material were made by inserting the composite material in a split ring 20.0 mm in diameter and 0.75 mnri w-:de. This ring was encircled by an aluminum plate serving as a support for the split ring. This mold was placed on a Mylar sheet backed by a glass slab. After insertion of the composite material, another glass slab faced with a thin Mylar sheet was used to press the composite into the ring. The mold, which was positioned between two glass slabs, was
loaded with a 1-kg weight for 1 min. The visiblelight cured composites were cured by initially exposing the center of the sample and then subsequently exposing eight evenly spaced locations close to tbe periphery' of the sample. Each location was exposed for 20 s with light from a Translux light curing unit (Kultzer, Inc., Irvine, CA). After one side was cured, the mold was turned upside down and the composite was then cured in the same way on the otber side. The chemically-cured composite (Material E) was mixed according to the manufacturer's instructions, placed in tlie mold as outlined for tbe visible-light cured materials, and allowed to set for 10 min under the 1-kg weight. After polymerization was completed, the samples were removed from the molds and transferred lo an oven where they were stored at 37°C for 24 h. Following this storage period, a piece of dental doss, approximately 15 cm long, was attached to each sample with a minimal amount of the same brand of composite material. Each sample was then transferred to a polypropylene bottle containing 60 ml of distilled water in which the sample
Table 2 Analytical results regarding concentrations* of Si, Ba, Z") •'I'J ""'' Q ' " ike filler particles
Material A B C D E F
Ba
46.7 46.7 28.4 32.8 40.7 37.5
*In percentages by weight.
.
Zn
Al
20.3.
8.8
9.2 6.3
10.4
1.3
12.4
3.1 9.7
O
%-fiiler
53.3 53.3 42.5 47.6 49.9 39.1
51.1 35.6 74.9 74.4 76.3 83.0
SODERHOLM was stored for 6 months at 60°C. By using the dentai floss and the screw cap, the samples could be hung; inside the bottles with both surfaces in direct contact with the water. Except for the bottles containing the specimens, five bottles containing 60 ml distilled water only served as a control group. Each month, 5 nal of the storage water was withdrawn from each bottle and replaced with an equal amount of distilled water. This' procedure was repeated until the samples had been stored for a total of 6 months in water. Additional samples were made and analyzed for filler composition. These analyses were made by use of electron dispersive X-ray analysis (EDXA) {ED.\X International Inc., Prairie View, IL, USA), using an acceleration voltage of 15 kV (Novascan 30, SEMCO Instruments Company, Ltd,, Ottawa, Canada), In addition to the chemical composition of the filler, the filler fraction by weight was also determined. This determination was done by measuring the weight change of one sample of each material before and after the matrix had been burned away for 2 h at 600°C, After the filler elements had been identified, the concentration of these elements in the storage media were determined by atomic absorption spectroscopy (Model 360 Atomic Absorption Spectrophotomeler, Perkin-Elmer, Norwalk, GT). This silica content was analyzed at a wavelength of 251,6 nm using a hollow-cathode lamp \\ath the slit setting at 0.2 nm and a reducing nitrous oxideacetylene flame. No dilution was required since
the working range for silica is linear up to concentrations of approximately 150 |ig/mi. Before the barium samples were analyzed, they were diluted by taking equal volumes of sample liquid and a 2000 ^g/ml KGl solution to reduce the risk for ionizadon interference, A hollow-cathode lamp was used as the light source. The wavelength was set to 553,6 nm and the slit setting to 0,2 nm, A reducing nitrous oxide-acetylene liame was used. Since the working range for Ba is linear up to concentrations of approximately 25 Hg/ml and the addition of the alkali salt resulted in a concentration below this value, no additional precautions were taken regarding the barium specimens. Zinc was analyzed using a zinc hollow-cathode lamp, a slit setting at 0,7 nm, a wavelength of 213,9 nm, and an oxidizing air-acetylene flame. Since the sensitivity for zinc is very high, preliminary analyses revealed that tlie zinc-containing specimens had to be diluted four times in order to give analytical results within the linear working range of concentrations up to approximately 1 ^lg/ml, All samples were analyzed at the same time after 6 months so that the same standards could be used. Since 5 ml of the 60-ml sample volume were drawn and replaced with 5 ml distilled water every time an analysis was conducted, 5/60 of leached filler elements were removed each time an analysis was conducted. To compensate for the diluting effect, the leaching value for each sample and month was adjusted according to the relationship:
Q.
& o Z m
UJ
o
o o o
o o
3
4
5
MONTH
Fig. I. Leaching data over time for Material A, Unfilled squares represent mean values of leached silica, and lines drawn in connection with each value represent standard deviation.
1
2
3
4
5
6
MONTH
Fig. 2. Leaching data over time for Material B, Unfilled squares represent mean values of leached silica, and lines drawn in connection with each value represent standard deviations.
FILLER LEAGHABILITY DURING WATER STORAGE
A(ni) + 5/60 x
2., A
(n)
Results
•n = 0
where Aj,jj(m) = adjusted amount of leached element al month m A (m) = analyzed amount of leached element at month m To test the working hypothesis that the concentration of leached filler elements levels out with time and reaches a constant value within 6 months at 60°C, the adjusted experimental data was analyzed by linear regression analysis. This analysis was conducted individually on the leached values of each sample. During this analysis the coefficient of correlation for each sample and element was. calculated and used to determine tbe strength of the linear relationship hetween time and concentration. Since the working hypothesis was that the amount of leached elements levels out and reaches a constant value over time, such a leaching pattern should be non-linear. Tbus, this hypothesis could he tested by formulating a null hypothesis that tested for linearity on the 95% significant level. The null hypothesis was rejected when the r-valuc was smaller than |rD2j|, and each rejection was taken as. a support for the formulated working hypothesis.
85
The results of the energ)' dispersive X-ray analyses (EDXA) and the filler fraction determinations are giveti in Table 2. The EDXA results shoold be regarded as qualitative rather than quantitative since they do not consider the existence of elements such as lithium, boron, and fluorine, which are essentially undetectable with this analytical method. It should be noted that the filler weight numbers being presented in Table 2 represent the weight of a particular element relative to the weight of the filler. Thus, a material with a lower element concentration could still contain more of that element if that material has a higher filler content. This can be seen by considering the values under "%-Cller" in Table 2. Accordingly,, the concentration of Si in the filler of tnaterial A is 46.7% and since material A contains 51.1% filler by weight, the total amount of Si in material A is 23.86% by weight. For material F, as atl example, the total amount of Si is 31.29% by weight. The adjusted results of the atomic absorption analyses are presented in Figs. 1 to 6. Regarding the control group consisting of distilled water without any composite mate-
& O < CC 1-
z m o z
UJ
O
MONTH
Fig. 3. Leaching data over time for Material C. Unliiled squares represent mean values of leached silica while filled and tilted squares represent that of barium. Lines drawn in connection with each value represent standard deviations.
MONTH
Fig. 4. Leaching data over time for Material D. Unfilled squares represent mean values ofleached silica and filled and tilted squares that of barium. Lines drawn in connection with each value represent standard deviations.
86
SODERHOLM
rial, no traces of Si, Ba or Zn could be detected at any time. By use of linear regression analysis, the leachability of the filler elements was described as linear functions of time within the investigated time interval (1-6 months). Of 30 investigated samples from which Si had leached, only one specimen (one of the Material A samples) had a significant (i'
KARL-JOHAN M, SOOERHOLM Department of Denial Biomalerials, C&llege of Dentistry, University of Florida, Gainesville, Florida, USA
Soderholm K-JM: Filler leachability during water storage of six composite materials, Scand J Dent Res 1990; 98: 82-8, Abstract — Sis diflferent dental composite n:iaterials were investigated regarding leaching of filler elements such as silicon, barium,, and zinc. The leaching was conducted by storing the samples in distilled water at 60^C for half a year. The results could not prove that the leaching behavior of filler elements in most of tbe investigated cases decreases with time. The practical implications. of this study could be important. If other elements follow the same leaching pattern, therapeutic elements such as fluoride could be incorporated in filler particles. The tendency of composites to leach filler elements almost linearly wilh time, couJd be used lo generate a cotistani release rate of such therapeutic elements over time. Such an application could have a major impact on controlling caries adjacent to composite restorations and sealants. Key words: degradation; dental materials; hydrolysis; resins. Department of Dental Biomaterials, College of Dentistry, Box J-446, Gainesville, Florida 32610, USA, Accepted for publication 20 June 1989,
Previous studies have shown that water improved the wear properties of cotnposite leaches filler elements from dental composite materials (7—9), Furthermore, these glasses tnaterials (1, 2), This leaching process is ae- possess some degree of X-ray opacity which centuated when the silica network has been is a desirable property for a composite, modified with metal oxides such as barium Since filler particles consisting of glasses and strontium oxide to form different glasses possess important advantages, it is important (2, 3), Although these glasses are susceptible to determine how significant the hydrolytic to hydrolytic degradation, they also possess degradation of these less stable filler particles some advantages (4—6), Because they are is for material properties. Broken down filler softer than quartz, finer filler particles can be surfaces will increase the risk for filler-matrix made during filler processing and a smoother debonding, decrease strength, and reduce surface finish can be produced on composite the wear resistance (10—13), Therefore, quesrestorations. This reduction in particle size tions on the leaching characteristics of difand improvement of stirface texture may ex- ferent composites should be addressed since plain why the use of these glasses may have such properties could correlate with fracture
83
FILLER LEACHABILITY DURING WATER STORAGE Table I Jnuestigalsd materials
Material
Symbol
Batch No.
A
4GI
B C D E
240284 0426841 40442 UP 4142E GP 4iO7E 4PR1P
Silux Heliosit Prisma-Fil Command Ultrafine Prestige Plus
E
Manufacturer 3M, St Paul, MN, USA Vivadent, Liechtenstein The L. D. Caulk Company, Milford, DE, USA Kerr, Romulus, MI, USA Lee Pharmaceuticals, El Monte, CA, USA 3M, St Paul, MN, USA
and wear hehavior of composites when used clinically (14). The ohjective of this study was to determine how the leaching pattern changes over time when composite materials are stored in an aqueous environment. The working hypothesis was that the leaching rate in this type of environment would level out and reach a constant value within 6 months. Material and methods Six composite materials were investigated (Table 1). Five specimens of each material were made by inserting the composite material in a split ring 20.0 mm in diameter and 0.75 mnri w-:de. This ring was encircled by an aluminum plate serving as a support for the split ring. This mold was placed on a Mylar sheet backed by a glass slab. After insertion of the composite material, another glass slab faced with a thin Mylar sheet was used to press the composite into the ring. The mold, which was positioned between two glass slabs, was
loaded with a 1-kg weight for 1 min. The visiblelight cured composites were cured by initially exposing the center of the sample and then subsequently exposing eight evenly spaced locations close to tbe periphery' of the sample. Each location was exposed for 20 s with light from a Translux light curing unit (Kultzer, Inc., Irvine, CA). After one side was cured, the mold was turned upside down and the composite was then cured in the same way on the otber side. The chemically-cured composite (Material E) was mixed according to the manufacturer's instructions, placed in tlie mold as outlined for tbe visible-light cured materials, and allowed to set for 10 min under the 1-kg weight. After polymerization was completed, the samples were removed from the molds and transferred lo an oven where they were stored at 37°C for 24 h. Following this storage period, a piece of dental doss, approximately 15 cm long, was attached to each sample with a minimal amount of the same brand of composite material. Each sample was then transferred to a polypropylene bottle containing 60 ml of distilled water in which the sample
Table 2 Analytical results regarding concentrations* of Si, Ba, Z") •'I'J ""'' Q ' " ike filler particles
Material A B C D E F
Ba
46.7 46.7 28.4 32.8 40.7 37.5
*In percentages by weight.
.
Zn
Al
20.3.
8.8
9.2 6.3
10.4
1.3
12.4
3.1 9.7
O
%-fiiler
53.3 53.3 42.5 47.6 49.9 39.1
51.1 35.6 74.9 74.4 76.3 83.0
SODERHOLM was stored for 6 months at 60°C. By using the dentai floss and the screw cap, the samples could be hung; inside the bottles with both surfaces in direct contact with the water. Except for the bottles containing the specimens, five bottles containing 60 ml distilled water only served as a control group. Each month, 5 nal of the storage water was withdrawn from each bottle and replaced with an equal amount of distilled water. This' procedure was repeated until the samples had been stored for a total of 6 months in water. Additional samples were made and analyzed for filler composition. These analyses were made by use of electron dispersive X-ray analysis (EDXA) {ED.\X International Inc., Prairie View, IL, USA), using an acceleration voltage of 15 kV (Novascan 30, SEMCO Instruments Company, Ltd,, Ottawa, Canada), In addition to the chemical composition of the filler, the filler fraction by weight was also determined. This determination was done by measuring the weight change of one sample of each material before and after the matrix had been burned away for 2 h at 600°C, After the filler elements had been identified, the concentration of these elements in the storage media were determined by atomic absorption spectroscopy (Model 360 Atomic Absorption Spectrophotomeler, Perkin-Elmer, Norwalk, GT). This silica content was analyzed at a wavelength of 251,6 nm using a hollow-cathode lamp \\ath the slit setting at 0.2 nm and a reducing nitrous oxideacetylene flame. No dilution was required since
the working range for silica is linear up to concentrations of approximately 150 |ig/mi. Before the barium samples were analyzed, they were diluted by taking equal volumes of sample liquid and a 2000 ^g/ml KGl solution to reduce the risk for ionizadon interference, A hollow-cathode lamp was used as the light source. The wavelength was set to 553,6 nm and the slit setting to 0,2 nm, A reducing nitrous oxide-acetylene liame was used. Since the working range for Ba is linear up to concentrations of approximately 25 Hg/ml and the addition of the alkali salt resulted in a concentration below this value, no additional precautions were taken regarding the barium specimens. Zinc was analyzed using a zinc hollow-cathode lamp, a slit setting at 0,7 nm, a wavelength of 213,9 nm, and an oxidizing air-acetylene flame. Since the sensitivity for zinc is very high, preliminary analyses revealed that tlie zinc-containing specimens had to be diluted four times in order to give analytical results within the linear working range of concentrations up to approximately 1 ^lg/ml, All samples were analyzed at the same time after 6 months so that the same standards could be used. Since 5 ml of the 60-ml sample volume were drawn and replaced with 5 ml distilled water every time an analysis was conducted, 5/60 of leached filler elements were removed each time an analysis was conducted. To compensate for the diluting effect, the leaching value for each sample and month was adjusted according to the relationship:
Q.
& o Z m
UJ
o
o o o
o o
3
4
5
MONTH
Fig. I. Leaching data over time for Material A, Unfilled squares represent mean values of leached silica, and lines drawn in connection with each value represent standard deviation.
1
2
3
4
5
6
MONTH
Fig. 2. Leaching data over time for Material B, Unfilled squares represent mean values of leached silica, and lines drawn in connection with each value represent standard deviations.
FILLER LEAGHABILITY DURING WATER STORAGE
A(ni) + 5/60 x
2., A
(n)
Results
•n = 0
where Aj,jj(m) = adjusted amount of leached element al month m A (m) = analyzed amount of leached element at month m To test the working hypothesis that the concentration of leached filler elements levels out with time and reaches a constant value within 6 months at 60°C, the adjusted experimental data was analyzed by linear regression analysis. This analysis was conducted individually on the leached values of each sample. During this analysis the coefficient of correlation for each sample and element was. calculated and used to determine tbe strength of the linear relationship hetween time and concentration. Since the working hypothesis was that the amount of leached elements levels out and reaches a constant value over time, such a leaching pattern should be non-linear. Tbus, this hypothesis could he tested by formulating a null hypothesis that tested for linearity on the 95% significant level. The null hypothesis was rejected when the r-valuc was smaller than |rD2j|, and each rejection was taken as. a support for the formulated working hypothesis.
85
The results of the energ)' dispersive X-ray analyses (EDXA) and the filler fraction determinations are giveti in Table 2. The EDXA results shoold be regarded as qualitative rather than quantitative since they do not consider the existence of elements such as lithium, boron, and fluorine, which are essentially undetectable with this analytical method. It should be noted that the filler weight numbers being presented in Table 2 represent the weight of a particular element relative to the weight of the filler. Thus, a material with a lower element concentration could still contain more of that element if that material has a higher filler content. This can be seen by considering the values under "%-Cller" in Table 2. Accordingly,, the concentration of Si in the filler of tnaterial A is 46.7% and since material A contains 51.1% filler by weight, the total amount of Si in material A is 23.86% by weight. For material F, as atl example, the total amount of Si is 31.29% by weight. The adjusted results of the atomic absorption analyses are presented in Figs. 1 to 6. Regarding the control group consisting of distilled water without any composite mate-
& O < CC 1-
z m o z
UJ
O
MONTH
Fig. 3. Leaching data over time for Material C. Unliiled squares represent mean values of leached silica while filled and tilted squares represent that of barium. Lines drawn in connection with each value represent standard deviations.
MONTH
Fig. 4. Leaching data over time for Material D. Unfilled squares represent mean values ofleached silica and filled and tilted squares that of barium. Lines drawn in connection with each value represent standard deviations.
86
SODERHOLM
rial, no traces of Si, Ba or Zn could be detected at any time. By use of linear regression analysis, the leachability of the filler elements was described as linear functions of time within the investigated time interval (1-6 months). Of 30 investigated samples from which Si had leached, only one specimen (one of the Material A samples) had a significant (i'
