Thermal properties of a methyl methacrylate-based orthodontic bonding adhesive Frederick A. Rueggeberg, DDS, MS," Felix T. Maher, BS, ~ and Michael T. Kelly, BS ~
Augusta, Ga. Methyl methacrylate-based (MMA-based) bonding resins have been used in orthodontics because they offer easy removal of both the bonded bracket and the residual adhesive at case completion. However, these materials are not cross-linked, and the brackets bonded with this type of product may undergo drifting when subjected to temperatures slightly higher than those in the mouth. This research investigated the influence of heat on the debonding characteristics of a MMA-based bonding resin compared with those of a BIS-GMA-based system. The temperature of initial bracket movement, as well as of final bracket displacement, was noted for a variety of applied loads (141, 226, 425, 934, and 1727 gm) with stainless steel brackets bonded to etched bovine enamel. The results showed that the MMA-based material underwent a glass transition near 47 ~ C in which the initiation of bracket drift resulted on the tooth surface. This temperature proved independent of the applied load. Further heating resulted in the release of heat from the resin as a result of further curing. The extent of bracket drift associated with this secondary heat release was dependent on the applied load. The debonding temperature of the BIS-GMA-based system was three to six times greater than that of the MMA-based product. Practitioners should be aware that brackets bonded with MMA-based resins have the potential for drifting.when subjected to temperatures within the normal range of hot fluids consumed by their patients. Bracket drift could result in deactivation of orthodontic force and could prolongthe treatment time. (AMJ ORTHOO~ENTOFAC ORTHOP 1992;101:342-9.)
Many agents used to attach orthodontic brackets to etched enamel are similar in formulation to composite restorative materials. This formulation is mostly a combination of BIS-GMA (2,2-Bis[4-(2hydroxy-3-methacroyloxypropoxy)phenyl]propane) and functional comonomers. Inorganic fillers are added to reduce polymerization shrinkage and water sorption, to increase strength, and also to impart color characterization to the material so that it matches the appearance of a natural tooth. On curing, these types of adhesive systems are highly cross-linked, which indicates that the individual polymer chains are chemically attached to one another at frequent intervals. This joining of individual chains results in a much more rigid and stronger material than in a material in which few or no cross-links occur. The clinical retention of brackets bonded with BISGMA-based materials has been highly successful. On the completion of treatment, however,, the brackets and bonded resins must be removed. The removal of stain-
From the School of Dentistry, Medical College of Georgia. "Assistant Professor, Dental Physical Sciences. q:qourth-year student dentist. "Second-year student dentist. 8 / 1 / 27284
342
less steel brackets is easily accomplished with crimping pliers. ~ However, a layer of bonding resin usually remains firmly attached to the etched enamel surface. 2z Removal of this resin necessitates a large expenditure of time and often results in the iatrogenic removal of enamel to obtain a clean, smooth tooth surface. 2'4 Ceramic brackets provide an unequaled esthetic result in bonded brackets, s Removal of single-crystal sapphire brackets frequently causes the fracture of the bracket because of the brittle nature of the material. This partial debonding requires time-consuming drilling of the remainder of the bracket from the tooth surface. 6 Also it is not uncommon to find that a large portion of enamel has been removed along with ceramic brackets. 6 To ease bracket and bonded resin removal, some manufacturers have used a methyl methacrylate-based (MMA-based) monomer system. 2.4 This system is a powder/liquid product that is quite similar to the autocured denture-base resins. Liquid methyl methacrylate is added to previously polymerized polymethyl methacrylate beads. The liquid wets the prepolymerized particles and undergoes a free radical polymerization at mouth temperature resulting in the attachment of a bracket to etched enamel. The cured polymer is not cross-linked, as are fhe BIS-GMA systems. As a result of this lack of cross-linking, the physical properties of
Volume 101
Methyl methaco'late bonding adhesions
Number 4
the M M A - b a s e d materials are inferior to those of the more highly cross-linked products, ttowever, the c o m : paratively poorer physical properties o f this type o f material aid in the removal of the bonded brackets at the case completion. 2 Because their strength of attachment to etched enamel is less than that of the B I S - G M A products, the M M A - b a s e d materials are easier to debond and tend to leave a clean tooth surface that requires only mild hand scaling and pumicing to complete the resin removal.2'4 Therefore these materials are marketed with the emphasis on the safety and ease of removal. During previous research on thermal debracketing of orthodontic resins; 7 it was noted that the M M A - b a s e d materials debonded at temperatures only slightly above moutli levels. It was also noted that brackets bonded with methyl methacrylate products showed evidence o f significant bracket drift before debonding at elevated temperatureS. Therefore the possibility exists that brackets bonded with this type of material may drift when subjected to higher temperatures, which may occur when a patient consumes hot liquids. ~ The purpose o f this research is to examine the effect o f temperature and load on the debonding characteristics of a commercial methyl methacrylate-based orthodontic bonding resin as compared with a B I S - G M A formulation. MATERIALS AND METHODS
Lower anterior bovine teeth were extracted, cleaned of periodontal attachment, stored in a 2% solution of sodium azide (an antimicrobial agent), and placed in a refrigerator until ready to use. Tooth preparation consisted of removal of the lingual surface and the pulp. A flat surface was prepared on the facial enamel with 600 grit silicone carbide. A No. 1/4 bur was used to drill a hole from the middle of the flattened facial surface into the pulp chamber. A K-type thermocouple (Models TFCY-010 and TFAL-101, Omega Engineering, Stamford, Conn.) was passed through this hole from the lingual side and secured so the tip was flush with the flat enamel surface. The thermocouple served to monitor the temperature of the resin/enamel interface during the thermal debonding procedure. A stainless steel mesh bracket (A5701/D2500, RMO, Inc., Denver, Colo.) was bonded to the flattened enamel surface according to the procedures and with the components prescribed in the directions of a MMAbased material, Quasar orthodontic debonding adhesive (Lot No. 81101, RMO, Inc., Denver, Colo.) and a BIS-GMAbased product, Concise orthodontic bonding resin (Lot No. P890207, 3M, St. Paul, Minn.). The brackets were positioned so that the center of the base was directly over the thermocouple, and the large bracket slot was horizontally positioned in a mesiodistal direction. The excess resin around the bracket base was removed, and the freshly bonded bracket was allowed to cure at room temperature for 10 minutes before being placed in water. The immersed, bonded teeth were stored at 37 ~ C until testing.
343
KNIFEBLADE SOLDERINGGUN TIP
"":'"':":":":"::""'::
ttOLDI NG
JIG - - I Fig. 1, Diagram of debonding apparatus.
After 24 hours of immersion, the teeth were removed and mounted in a brass ring with dental stone. A removable mounting fork was used to ensure that the bracket Was consistently positioned with respect to the long axis of the brass ring before the stone set. In this manner, shearing was applied on the bracket to minimize peeling forces. The brass ring with oriented tooth/bracket assembly was locked into a holder. A shear!ng force on the upper surface on the bracket Was applied ~vith a knife blade. Various loads wereapplied as shearing forces: 141 gm (5 oz), 226 gm (8 oz), 425 gm (15 oz), 934 gm (33 oz), and 1727 gm (61 oz). These loads were chosen to rePresent a wide range of clinically encountered activation forces. 9 The vertical position of the knife blade was monitored with a linear variable differential transformer (Schaevitz Engineering, Pennsauken, N.J.). The output of this transformer was amplified and sent to the Y axis of an X-Y recorder. A drop in the vertical displacement of the bracket was noted by a downward deflection of the recording pen along the Y axis. The thermocouple was attached to an electronic cold junction compensator (Model MCJ-K, Omega Engineering, Stamford, Conn.). The output of the compensator was applied to the X axis of the X-Y recorder. Calibration of the linear displacement of the X axis from inches into millivolts was accomplished with a millivolt potentiometer (Model 8690-2, Leeds and Northrup, Atlanta, Ga.). In this manner, the displacement of tile pen in the X direction could be changed from inches into millivolts. A reference book t~was consulted to convert the millivoltage output at debonding into a temperature value. lleat was applied to the bonded bracket with a soldering iron (Model 09540406, Sears Craftsman, Chicago, 111.) that was suspended so that it imparted no force on the bracket. The flat blade of the soldering iron t!p was inserted into the large horizontal slot of the orthodontic bracket, and current was applied to initiate heating. Fig. I diagrams the debonding apparatus. The downward deflection of the recording pen, which diverted from the straight baseline, represented the start of -th~'lSracketdrift. The temperature of this event was determined and recorded as the initial drop temperature (T~). Complete dcbonding of the bracket from the tooth was denoted by a
344
Rueggeberg, Maher, and Kelly
Am. J. Orthod. Dentofac. Orthop. April 1992
Table I. Mean initial and final debonding temperatures Temperature
Resin
Load (gin)
Quasar Quasar Quasar Quasar Quasar Concise Concise
141 226 425 934 1727 141 1727
Initial drop (~
44.8 46.0 48.4 50.7 40.6
(7) (7) (5) (9) (6)
Final drop (=C)
69.4 (I I) 69.4 (6) 79.2 (18) 70.3 (13) 49.0 (8) 237.2 (21) 173.2 (20)
*Mean of five replications. ( ) = Standard deviation.
37 ~ C for 24 hours. After this time, the specimen was placed in a DSC (Model 910, Du Pont Instruments, Wilmington, Del.). The standard against which the thermal events of the specimen were compared was an empty aluminum pan and lid of similar weight. A thermal scan of each of the cured materials was run at a ramp of 5 ~ C/rain from 20 ~ to 120~ C for the MMA-based product and 20 ~ to 260 ~ C for the BIS-GMA-based material. Room air was used as a purge gas at a rate of 50 ml/min. Previous to running the thermal scans of the commercial products, the temperature scale of the DSC was calibrated with benzophenone (melting point 48.5 ~ C) and indium (156.2 ~ C). The heated specimen was allowed to cool to room temperature, and a second thermal run was accomplished under similar conditions. From the resulting thermal plots, the exothermic, endothermic, and changes in the heat capacity of the specimens were observed. RESULTS
sharp: drop in the recording pen. This temperature was noted as the final drop temperature (Tf). Five replications of debonding were performed for each applied load with the MMA-based product. The BIS-GMA material was tested with only five replications each of the 141 and 1727 gm loads. Fresh teeth and brackets were used for each replication. The mean initial and final temperatures o f bracket drop were determined for each applied load of the MMA-based product. The t test was used to compare mean initial and final debonding temperatures for each load of the methyl methacrylate material. A regression analysis was applied to the relationship between the applied load and the debonding temperatures (initial and final) to determine the effect of load on the debonding event. The slopes of the regression lines representing the linear relationship between the applied load and the debonding temperatures for the initial and final events were compared to detect any significant difference. All statistical tests were performed at the 95% level of confidence. A one-way analysis of variance (ANOVA) was used to compare the mean debonding initial and final temperatures of the MMA-based material with the 141 gm load and the debonding temperature of the BIS-GMA product with a similar load. The Fisher's PLSD test was used to detect the presence of a significant difference between specific pairs of mean debonding temperatures among these conditions. A similar procedure was performed for the 1727 gm load debonding temperatures. Differential scanning calorimetry (DSC) was used to examine the thermal events occurring in both the MMA-based and BIS-GMA-based products. It is an experimental technique that measures the amount of heat flow between an empty specimen pan and a similar pan filled with a test specimen when both are heated at identical rates. Exothermic (heat releasing) and endothennic (heat absorbing) events are thus detected and measured. Approximately 5 mg of mixed material of each resin type was allowed to cure in separate aluminum DSC dishes. After curing, the upper lid was crimped to the lov,'er, and the specimens ,.,,'ere stored dry at
Table I displays the mean debonding temperatures observed as a function o f applied load for both the M M A - b a s e d product and the B I S - G M A material. The MMA-based material demonstrated both initial and final debonding temperatures, whereas the B I S - G M A material demonstrated only a single, final temperature, with no bracket movement before total debonding. The trends of these data are more easily seen in Figs. 2 and 3. Fig. 2 displays the debonding data for the M A A based product in a graph. The temperature of initial bracket movement (Ti) did not vary greatly with the applied load, whereas the temperature o f final bracket position before full debonding (Tf) appears to have an inverse relationship with the applied load. When comparing Ti with Tr for e a c h l o a d condition of Quasarbonded brackets, the t test revealed significant differences between all temperature means with the exception of the 1727 gm values (Table II). An ANOVA of the mean Ti values revealed that there was no significant difference among these values with the applied load. An ANOVA o f the mean Tr values indicated that the temperature associated with the 1727 gm load was significantly different from all other values. The results o f the regression analysis o f the Ti indicate that the slope of the linear model relating the applied load to the initial drop temperature does not significantly differ from zero (Table III). The slope o f the linear regression o f Tf values as a function of load does differ from zero (Table III) and has a negative value. The mean slope for each regression line does not lie within the 95% confidence limits for the corresponding slope value. This relationship indicates that the two slopes are significantly different from each other. The mean debonding values for the B I S - G M A -
Volume 101 Number 4
Methyl methacry/ate bonding adhesions
345
10o 90 80 o
tll F,< rr tll uJ F-
i0.014x
70
R = 0.551
60 50
(Ti)
40 30
y = 47.881 - 0 , 0 0 3 x
R = 0.212
20
10 2 0 0
i
i
i
500
1000
1500
i
2000
SHEAR LOAD (gin) Fig. 2. Initial and final d e b o n d i n g temperature of Quasar as a function of s h e a r load.
Table II. Significant differences in debonding
300
temperature with applied load
Quasar: Ti vs Tt
Concise
One'tail
~" o
:250.
Load (gin)
p-va/ue
200"
141 226 425 934 1727
0.0016' 0.00025* 0.0029* 0.0117" 0.051
W" -rr"
Augusta, Ga. Methyl methacrylate-based (MMA-based) bonding resins have been used in orthodontics because they offer easy removal of both the bonded bracket and the residual adhesive at case completion. However, these materials are not cross-linked, and the brackets bonded with this type of product may undergo drifting when subjected to temperatures slightly higher than those in the mouth. This research investigated the influence of heat on the debonding characteristics of a MMA-based bonding resin compared with those of a BIS-GMA-based system. The temperature of initial bracket movement, as well as of final bracket displacement, was noted for a variety of applied loads (141, 226, 425, 934, and 1727 gm) with stainless steel brackets bonded to etched bovine enamel. The results showed that the MMA-based material underwent a glass transition near 47 ~ C in which the initiation of bracket drift resulted on the tooth surface. This temperature proved independent of the applied load. Further heating resulted in the release of heat from the resin as a result of further curing. The extent of bracket drift associated with this secondary heat release was dependent on the applied load. The debonding temperature of the BIS-GMA-based system was three to six times greater than that of the MMA-based product. Practitioners should be aware that brackets bonded with MMA-based resins have the potential for drifting.when subjected to temperatures within the normal range of hot fluids consumed by their patients. Bracket drift could result in deactivation of orthodontic force and could prolongthe treatment time. (AMJ ORTHOO~ENTOFAC ORTHOP 1992;101:342-9.)
Many agents used to attach orthodontic brackets to etched enamel are similar in formulation to composite restorative materials. This formulation is mostly a combination of BIS-GMA (2,2-Bis[4-(2hydroxy-3-methacroyloxypropoxy)phenyl]propane) and functional comonomers. Inorganic fillers are added to reduce polymerization shrinkage and water sorption, to increase strength, and also to impart color characterization to the material so that it matches the appearance of a natural tooth. On curing, these types of adhesive systems are highly cross-linked, which indicates that the individual polymer chains are chemically attached to one another at frequent intervals. This joining of individual chains results in a much more rigid and stronger material than in a material in which few or no cross-links occur. The clinical retention of brackets bonded with BISGMA-based materials has been highly successful. On the completion of treatment, however,, the brackets and bonded resins must be removed. The removal of stain-
From the School of Dentistry, Medical College of Georgia. "Assistant Professor, Dental Physical Sciences. q:qourth-year student dentist. "Second-year student dentist. 8 / 1 / 27284
342
less steel brackets is easily accomplished with crimping pliers. ~ However, a layer of bonding resin usually remains firmly attached to the etched enamel surface. 2z Removal of this resin necessitates a large expenditure of time and often results in the iatrogenic removal of enamel to obtain a clean, smooth tooth surface. 2'4 Ceramic brackets provide an unequaled esthetic result in bonded brackets, s Removal of single-crystal sapphire brackets frequently causes the fracture of the bracket because of the brittle nature of the material. This partial debonding requires time-consuming drilling of the remainder of the bracket from the tooth surface. 6 Also it is not uncommon to find that a large portion of enamel has been removed along with ceramic brackets. 6 To ease bracket and bonded resin removal, some manufacturers have used a methyl methacrylate-based (MMA-based) monomer system. 2.4 This system is a powder/liquid product that is quite similar to the autocured denture-base resins. Liquid methyl methacrylate is added to previously polymerized polymethyl methacrylate beads. The liquid wets the prepolymerized particles and undergoes a free radical polymerization at mouth temperature resulting in the attachment of a bracket to etched enamel. The cured polymer is not cross-linked, as are fhe BIS-GMA systems. As a result of this lack of cross-linking, the physical properties of
Volume 101
Methyl methaco'late bonding adhesions
Number 4
the M M A - b a s e d materials are inferior to those of the more highly cross-linked products, ttowever, the c o m : paratively poorer physical properties o f this type o f material aid in the removal of the bonded brackets at the case completion. 2 Because their strength of attachment to etched enamel is less than that of the B I S - G M A products, the M M A - b a s e d materials are easier to debond and tend to leave a clean tooth surface that requires only mild hand scaling and pumicing to complete the resin removal.2'4 Therefore these materials are marketed with the emphasis on the safety and ease of removal. During previous research on thermal debracketing of orthodontic resins; 7 it was noted that the M M A - b a s e d materials debonded at temperatures only slightly above moutli levels. It was also noted that brackets bonded with methyl methacrylate products showed evidence o f significant bracket drift before debonding at elevated temperatureS. Therefore the possibility exists that brackets bonded with this type of material may drift when subjected to higher temperatures, which may occur when a patient consumes hot liquids. ~ The purpose o f this research is to examine the effect o f temperature and load on the debonding characteristics of a commercial methyl methacrylate-based orthodontic bonding resin as compared with a B I S - G M A formulation. MATERIALS AND METHODS
Lower anterior bovine teeth were extracted, cleaned of periodontal attachment, stored in a 2% solution of sodium azide (an antimicrobial agent), and placed in a refrigerator until ready to use. Tooth preparation consisted of removal of the lingual surface and the pulp. A flat surface was prepared on the facial enamel with 600 grit silicone carbide. A No. 1/4 bur was used to drill a hole from the middle of the flattened facial surface into the pulp chamber. A K-type thermocouple (Models TFCY-010 and TFAL-101, Omega Engineering, Stamford, Conn.) was passed through this hole from the lingual side and secured so the tip was flush with the flat enamel surface. The thermocouple served to monitor the temperature of the resin/enamel interface during the thermal debonding procedure. A stainless steel mesh bracket (A5701/D2500, RMO, Inc., Denver, Colo.) was bonded to the flattened enamel surface according to the procedures and with the components prescribed in the directions of a MMAbased material, Quasar orthodontic debonding adhesive (Lot No. 81101, RMO, Inc., Denver, Colo.) and a BIS-GMAbased product, Concise orthodontic bonding resin (Lot No. P890207, 3M, St. Paul, Minn.). The brackets were positioned so that the center of the base was directly over the thermocouple, and the large bracket slot was horizontally positioned in a mesiodistal direction. The excess resin around the bracket base was removed, and the freshly bonded bracket was allowed to cure at room temperature for 10 minutes before being placed in water. The immersed, bonded teeth were stored at 37 ~ C until testing.
343
KNIFEBLADE SOLDERINGGUN TIP
"":'"':":":":"::""'::
ttOLDI NG
JIG - - I Fig. 1, Diagram of debonding apparatus.
After 24 hours of immersion, the teeth were removed and mounted in a brass ring with dental stone. A removable mounting fork was used to ensure that the bracket Was consistently positioned with respect to the long axis of the brass ring before the stone set. In this manner, shearing was applied on the bracket to minimize peeling forces. The brass ring with oriented tooth/bracket assembly was locked into a holder. A shear!ng force on the upper surface on the bracket Was applied ~vith a knife blade. Various loads wereapplied as shearing forces: 141 gm (5 oz), 226 gm (8 oz), 425 gm (15 oz), 934 gm (33 oz), and 1727 gm (61 oz). These loads were chosen to rePresent a wide range of clinically encountered activation forces. 9 The vertical position of the knife blade was monitored with a linear variable differential transformer (Schaevitz Engineering, Pennsauken, N.J.). The output of this transformer was amplified and sent to the Y axis of an X-Y recorder. A drop in the vertical displacement of the bracket was noted by a downward deflection of the recording pen along the Y axis. The thermocouple was attached to an electronic cold junction compensator (Model MCJ-K, Omega Engineering, Stamford, Conn.). The output of the compensator was applied to the X axis of the X-Y recorder. Calibration of the linear displacement of the X axis from inches into millivolts was accomplished with a millivolt potentiometer (Model 8690-2, Leeds and Northrup, Atlanta, Ga.). In this manner, the displacement of tile pen in the X direction could be changed from inches into millivolts. A reference book t~was consulted to convert the millivoltage output at debonding into a temperature value. lleat was applied to the bonded bracket with a soldering iron (Model 09540406, Sears Craftsman, Chicago, 111.) that was suspended so that it imparted no force on the bracket. The flat blade of the soldering iron t!p was inserted into the large horizontal slot of the orthodontic bracket, and current was applied to initiate heating. Fig. I diagrams the debonding apparatus. The downward deflection of the recording pen, which diverted from the straight baseline, represented the start of -th~'lSracketdrift. The temperature of this event was determined and recorded as the initial drop temperature (T~). Complete dcbonding of the bracket from the tooth was denoted by a
344
Rueggeberg, Maher, and Kelly
Am. J. Orthod. Dentofac. Orthop. April 1992
Table I. Mean initial and final debonding temperatures Temperature
Resin
Load (gin)
Quasar Quasar Quasar Quasar Quasar Concise Concise
141 226 425 934 1727 141 1727
Initial drop (~
44.8 46.0 48.4 50.7 40.6
(7) (7) (5) (9) (6)
Final drop (=C)
69.4 (I I) 69.4 (6) 79.2 (18) 70.3 (13) 49.0 (8) 237.2 (21) 173.2 (20)
*Mean of five replications. ( ) = Standard deviation.
37 ~ C for 24 hours. After this time, the specimen was placed in a DSC (Model 910, Du Pont Instruments, Wilmington, Del.). The standard against which the thermal events of the specimen were compared was an empty aluminum pan and lid of similar weight. A thermal scan of each of the cured materials was run at a ramp of 5 ~ C/rain from 20 ~ to 120~ C for the MMA-based product and 20 ~ to 260 ~ C for the BIS-GMA-based material. Room air was used as a purge gas at a rate of 50 ml/min. Previous to running the thermal scans of the commercial products, the temperature scale of the DSC was calibrated with benzophenone (melting point 48.5 ~ C) and indium (156.2 ~ C). The heated specimen was allowed to cool to room temperature, and a second thermal run was accomplished under similar conditions. From the resulting thermal plots, the exothermic, endothermic, and changes in the heat capacity of the specimens were observed. RESULTS
sharp: drop in the recording pen. This temperature was noted as the final drop temperature (Tf). Five replications of debonding were performed for each applied load with the MMA-based product. The BIS-GMA material was tested with only five replications each of the 141 and 1727 gm loads. Fresh teeth and brackets were used for each replication. The mean initial and final temperatures o f bracket drop were determined for each applied load of the MMA-based product. The t test was used to compare mean initial and final debonding temperatures for each load of the methyl methacrylate material. A regression analysis was applied to the relationship between the applied load and the debonding temperatures (initial and final) to determine the effect of load on the debonding event. The slopes of the regression lines representing the linear relationship between the applied load and the debonding temperatures for the initial and final events were compared to detect any significant difference. All statistical tests were performed at the 95% level of confidence. A one-way analysis of variance (ANOVA) was used to compare the mean debonding initial and final temperatures of the MMA-based material with the 141 gm load and the debonding temperature of the BIS-GMA product with a similar load. The Fisher's PLSD test was used to detect the presence of a significant difference between specific pairs of mean debonding temperatures among these conditions. A similar procedure was performed for the 1727 gm load debonding temperatures. Differential scanning calorimetry (DSC) was used to examine the thermal events occurring in both the MMA-based and BIS-GMA-based products. It is an experimental technique that measures the amount of heat flow between an empty specimen pan and a similar pan filled with a test specimen when both are heated at identical rates. Exothermic (heat releasing) and endothennic (heat absorbing) events are thus detected and measured. Approximately 5 mg of mixed material of each resin type was allowed to cure in separate aluminum DSC dishes. After curing, the upper lid was crimped to the lov,'er, and the specimens ,.,,'ere stored dry at
Table I displays the mean debonding temperatures observed as a function o f applied load for both the M M A - b a s e d product and the B I S - G M A material. The MMA-based material demonstrated both initial and final debonding temperatures, whereas the B I S - G M A material demonstrated only a single, final temperature, with no bracket movement before total debonding. The trends of these data are more easily seen in Figs. 2 and 3. Fig. 2 displays the debonding data for the M A A based product in a graph. The temperature of initial bracket movement (Ti) did not vary greatly with the applied load, whereas the temperature o f final bracket position before full debonding (Tf) appears to have an inverse relationship with the applied load. When comparing Ti with Tr for e a c h l o a d condition of Quasarbonded brackets, the t test revealed significant differences between all temperature means with the exception of the 1727 gm values (Table II). An ANOVA of the mean Ti values revealed that there was no significant difference among these values with the applied load. An ANOVA o f the mean Tr values indicated that the temperature associated with the 1727 gm load was significantly different from all other values. The results o f the regression analysis o f the Ti indicate that the slope of the linear model relating the applied load to the initial drop temperature does not significantly differ from zero (Table III). The slope o f the linear regression o f Tf values as a function of load does differ from zero (Table III) and has a negative value. The mean slope for each regression line does not lie within the 95% confidence limits for the corresponding slope value. This relationship indicates that the two slopes are significantly different from each other. The mean debonding values for the B I S - G M A -
Volume 101 Number 4
Methyl methacry/ate bonding adhesions
345
10o 90 80 o
tll F,< rr tll uJ F-
i0.014x
70
R = 0.551
60 50
(Ti)
40 30
y = 47.881 - 0 , 0 0 3 x
R = 0.212
20
10 2 0 0
i
i
i
500
1000
1500
i
2000
SHEAR LOAD (gin) Fig. 2. Initial and final d e b o n d i n g temperature of Quasar as a function of s h e a r load.
Table II. Significant differences in debonding
300
temperature with applied load
Quasar: Ti vs Tt
Concise
One'tail
~" o
:250.
Load (gin)
p-va/ue
200"
141 226 425 934 1727
0.0016' 0.00025* 0.0029* 0.0117" 0.051
W" -rr"
