- Email: [email protected]
base ratio on curing of resin luting agents: Polymerization exotherm analysis
Dent Mater 10:314-318, September, 1994
Sensitivity of catalyst/base ratio on curing of resin luting agents: Polymerization exotherm analysis Jason A. Griggs, Chiayi Shen, Kenneth J. Anusavice Department of Dental Biomaterials, College of Dentistry, University of Florida, GainesviUe, Florida, USA
ABSTRACT Objectives. Currently, the proposed test of the International Standardization Organization (ISO) for measuring working and setting times of resin luting agents is based on measurement of times to reach specified stages on the polymerization exotherm. The objective of this study was to use this test to investigate the influence of variations in the mass ratios of catalyst paste to base paste on the working and setting times of three dual-cured dental resin luting agents, Methods. The materials used were Dicor Light Activated Cement (Dentsply International Inc.), Palfique Inlay Resin Cement (Tokuyama Soda Co.), and Vivadent Dual Cement (Vivadent). Fifteen specimens of each material were tested for working time by spatulating mass ratios from 0.7 to 1.3 for 30s at 23°C and recording the time from beginning of spatulation to the time at which a temperature increase occurs. Ten specimens of each material were tested for setting time by spatulating in a similar manner at 37°C and recording the time at which the temperature reaches a maximum value, Results. The data were fitted to the relation, In t = In A + Bm, where t is the time in seconds, m is the mass ratio, and A and B are regression coefficients. The results suggest that working and setting times of the specimens were independent of variations in mass ratio. A comparison among the materials was made by using a multiple linear regression with the relation, In t = In C + Dm + E7 + Fm~,, where ~, is a dummy variable to help distinguish between materials, and C, D, E, and F are regression coefficients. The results suggest that differences in materials influence the working time but not the setting time. Significance. These results infer that variations in mass ratio (+ 20%) often observed in the clinical setting should not have a significant influence on the working and setting times of resin luting agents, INTRODUCTION The increased popularity of all-ceramic restorations has led to the recent introduction of several two-paste dual-cured resin luting agents. Dual-cured resin luting agents consist of a base and a catalyst and can be cured through light activation and chemical-curedmechanisms. In clinical practice, light activa314 Griggs et aL/Catalyst/base ratio vs. curing of resin luting agents
tion is performed after the clinician has finished seating the ceramic prosthesis. The chemical-cured component of the dual-cured resins begins polymerization after the two pastes are mixed. The length of time from the start of mixing to the time of insertion of the prosthesis is the working time. The time from the start of mixing until the restoration no longer requires external support from a matrix is the setting time. For a chemical-cured resin, the mixture should remain moldable within the first 1.5 to 2 min from the start of mixing to faci]itate p]acement ofthe mixture in the cavity preparation, and then the material should set within 5 min (Jacobsen and VonFraunhofer, 1974). These times also apply to dual-cured resins, especia]]y the working time. Resin ]uting agents, such as chemical-cured composites, can fail to exhibit the desired working time. The effects ofhydroquinone concentration on inhibition ofpremature initiation, differentfi]lersurfaceareas on viscosity, TEC-DMA to Bis-GMA ratios on viscosity, and different initiator concentrations on free-radical production have been shown to be factors that influence curing times greatly (Ruyter, 1981; Ruyter, 1985). Normally, the clinician or the assistant will dispense equal volumes of base and catalyst pastes from separate containers. The ratio ofthe mass ofcatalyst paste to the mass ofbase paste wi]] hereafter be referred to as the "mass ratio". It is not ]ike]y that the ratio of the two pastes will be precise each time. Ideally, the working and the setting times should be insensitive to the mass ratio. A measurement of this sensitivity necessitates a standardized test for measuring the working and setting times of resin products. Wo]cott et al. (1951) analyzed the exotherm profiles of resinous filling materials during curing. They defined the working and setting times in termsoftimelapsesbetweenpeaktemperaturesintheexotherm profile. Wilson(1964) adapted a curemeter fromthe rubber industry to study curing times in impression materials. The meter consists of a stylus under cyclical loading, which traverses through the material. The rheology of the material is observed as a decrease in the amplitude of the trace as the material cures. Several investigations (Houston and Miller, 1968; Bovis et al., 1971; Plant et al., 1972) have used this equipment to define the working and setting times of resin luting agents according to rheological measurements.
Bovis et al. (1971) followed American Dental Association Specifications No. 8 and No. 9 (ADA, 1970) to test composite filling materials. These procedures describe the use of a Gillmore needle as a penetrometer to determine setting times in zinc phosphate cements and silicate cements, respectively. The setting time is defined as the time from the beginning of mixing to the point at which the Gillmore needle no longer penetrates the material under a specified load. They found the results to be highly sensitive to operator variability. A current test proposal (Btichel, 1993) involves measuring the exotherm of a chemical-cured resin as it cures over time. During the induction period, the activation of the initiator requires a certain amount of energy. As this energy is absorbed, the resin temperature decreases. The temperature decrease is soon offset by the exothermic reaction, and the temperature increases as the reaction accelerates. The temperature then reaches a maximum and begins to decrease as the reaction nears completion. A curve recording this thermal history will contain a minimum point and a maximum point, which are used in defining the working and setting times. The objective of this study was to investigate the influence of variations in the mass ratios of catalyst paste to base paste on the working and setting times of three dual-cured dental resins using the polymerization exotherm profiles.
, 'I ~ '
4
, 1 , ~' "P~I I , ,
~" (~ 4 ~_ //'\\/".7 ) ' / / / / / / / / / / / /
r : -" 2 .-
~
| __ _~'_. 1 - -~ -
1
/ / // /
MATERIALS AND METHODS Three dual-cured resins were chosen for study: Dicor Light Activated Cement, batches 513315 and 513386 (Dentsply International Inc., York, PA, USA), Palfique Inlay Resin Cement, batches RCA153 and RCB653 (Tokuyama Soda Co., LTD., Tsukuba-City, Japan), and Dual Cement, batches 560049 and 560051 (Vivadent, Schaan, Liechtenstein). All specimens of each material were prepared from the same batches, and all materials were mixed according to the manufacturers' instructions. The materials were all dual-cured resins; however, only the chemical-cured mechanism was activated so that the working and setting times recorded would represent the amount of time available to a clinician for manipulation of the material. Upon qualitative examination, the lack of light activation did not appear to detract from the degree of cure. A volume of catalyst paste was placed on a mixing pad, and the mass was recorded. An approximately equal volume by appearance of base paste was placed on the mixing pad, and the mass was recorded. To simulate clinical variation, no attempt was made to keep the paste masses constant by weight. A total of 15 specimens was prepared for measuring the working time of each material. After dispensing the two pastes by visual estimation of a ratio of 1.0, the first 10 specimens of each group varied in mass ratio from 0.8 to 1.2. For each material, an additional five specimens with mass ratios ranging from 0.7 to 1.3 were prepared to assess possible effects of excessive operator errors. A total of 10 specimens was prepared for measuring the setting time of each material. The same mass ratio variation, 0.8 to 1.2, also resulted in this group of specimens. No additional specimens were prepared to include greater mass ratio variations, A timer was started at the moment mixing began. At 30 s after the start of mixing, the material was placed in a mold, and the temperature was recorded at 1.4 s intervals by a
/ // / // / // // // /
~
J~
/
Fig 1. Apparatus for determination of working and setting time, polyethylene tubing(A), nylon block (B), stainless steel tube (C), copper/constantan (type T) thermocouple(D). All measurementsin ram,
thermocouple connected to an analog-to-digital converter (pMega Intelligent Scanning System, Omega Engineering, Inc., Stamford, CT, USA) to an accuracy of 0. I°C. The mold, shown in Fig. 1, consists of polyethylene tubing on a nylon block which has a stainless steel tube inserted through it containing a copper/constantan (type T) thermocouple. The working time was determined while maintaining the specimen mold at room temperature (23 _+1°C), and the setting time was determined with the mold in a constant temperature oven maintained at 37 _+l°C. The temperature fell slightly from T 1to T O(Fig. 2) after the material was inserted into the mold. The time at which the temperature began to increase denotes the start of the curing reaction. The working time (t w) was determined at this minimum. The time at which the temperature began to fall again Dental Materials~September 1994 315
103
~
,
,
T2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dicor
o~ Start of mixing Insertion
~c~ I--
30 s
tw
102
I
I
I
103
I
I
I
I
I Palfique
E
ts
Time
~
o
o
Fig 2. T y p i c a l recording trace showing temperature changes with time for determination of working and setting times.
102
~
i
~
103
i
i
i
t i
Vivadent 0
0
~
¢K)n
0
from T 2 denotes the end of the curing reaction. The setting time (t s) was determined at this maximum. The recorded times and the measured mass ratio were used to determine the effect of mass ratio on the working and setting time by a linear regression analysis for each material. The effect of material on the working and setting time was evaluated by a multiple linear regression model using both material and mass ratio as the independent variables.
1o2 0.4
, 0.6
O
Q
0
, , 0.8 1.0 1.2 Cataryst/Base Ratio
1.4
Fig 3. Working time vs. mass ratio.
RESULTS The natural logarithms of the times were used in the analysis because of the distribution of the recorded values of working and setting times. To determine the influence of the mass ratio on the times measured for each group, several mathematical models were examined, and the following simple linear regression model demonstrated the best fit of data: In t = In A + Bm
lo3 Dicor
o ~ ~
(1) 10 2
where t is the working or setting time in seconds, A is a constant in seconds, B is a constant that dictates the effect of mass ratio on the working or setting time, and m is the mass ratio. The null hypothesis to be tested was Ho: B=0. The equivalent form of equation (1) is: t = Ae B=
lO3 o ~- v - o,,~
"~ e ~= ==
O
Palfique
o
7'
(2)
For the multiple linear regression, the following mode] was
102
i
lO 3
,
used:
Vivadent
lnt=lnC+Dm+ET+Fmy
where t is the working or setting time in seconds, C is a constant in seconds, m is the catalyst/base ratio, ?is a dummy variable t h a t assigns a numerical code to each of the three materials, and D, E, and F are the coefficients for re, T, and my terms, respectively. The values of A and B were determined by a linear regression of In t against m. The intercept of the straight line at m=0 is the value of In A, and the slope of the same line is the 316 Griggs et al./Catalyst/base rafio vs. curing of resin luting agents
o¢
(3)
~ : ~ o %
10 2
o.s
0.8 Catalyst/Base
Fig 4.
Setting time vs. mass ratio.
I 1.0 Ratio
o
1.2
TABLE 1: COEFFICIENTS OF CURVES FIT TO CURING TIMES VS. MASS RATIO DATA USING THE RELATION, In t = In A + Bm. Working Time Group
occur routinely.
Statistical
analysis has confirmed that both working and setting times are not sensitive to the
Setting Time
Mean +_S.D.
of 20% in the mass ratio can
Mean +_S.D.
In A
B
In A
B
degree of mass ratio varia-
Dicor
5.53 + 0.12
-0.117 + 0.143 (p = 0.4304)
5.64 + 0.13
-0.099 + 0.139 (p = 0.4950)
tion t h a t m a y be occurring during routine clinical use.
Palfique
5.69 + 0.53
-0.534 + 0.484 (p = 0.2905)
6.67 + 0.46
0.366 + 0.493 (p = 0.4794)
Vivadent
5.99 + 0.11
-0.113 + 0.161 (p = 0.4963)
6.19 + 0.24
0.133 + 0.241 (p = 0.5957)
Asmussen (1981) investigated the effect of various combinations of activator, initiator, and inhibitor on the curing time. He varied the mass percent of the activator between 0.5% and 3.0%, the
initiator from 0.25% to 2.0%, and the inhibitor from 0% to TABLE 2: COEFFICIENTS ASSOCIATED WITH MULTIPLE LINEAR REGRESSION ANALYSIS OF THE EQUATION, In t = In C + Dm + ET + Fmy. Working Time
Setting Time
1.0%. He concluded t h a t the curing t i m e is proportional
In C
D
E
F
to the mass percent of the
4.71 + 0.53
0.505 + 0.236
0.900 _+ 0.578
-0.306 + 0 . 2 4 9
inhibitorbutisinverselypro-
(p = 0.1270)
(p = 0.0379)
(p = 0.2260)
portional to the product of the mass percents of the acti-
0.523 _+ 0.535
0.389 + 1.222
-0.371 + 0.557
(p = 0.7525)
(p = 0.3370)
(p = 0.5117)
5.34 + 1.16
value of B. Table 1 shows the results of the regression by a general linear model of SAS (1985). The statistical analysis shows that the p-values for B range from 0.2905 to 0.5957. Thus, the null hypothesis was accepted for both working and setting times. This means that the range of mass ratio variation in this study had no influence on either the working or setting times of the resin luting agents tested. Therefore, the values of A shown in Table 1 are the means of the experimental data. The mean working times were 280 _+20 s, 499 _+83 s, and 448 + 40 s for Dicor, Palfique, and Vivadent groups, respectively. The mean setting times were 310 _+12 s, 568 _+90 s, and 430 t 29 s for Dicor, Palfique, and Vivadent groups, respectively. The experimental data along with the best fit curves of the three materials tested are shown in Figs. 3 and 4 for working and setting times, respectively, Table 2 shows the results of the multiple linear regression, In this analysis, the interest was two-fold, i.e., whether the working or setting times differed among the materials tested and whether there was an interaction between the material and the mass ratio. The results show that the working time was significantly influenced by the materials (p = 0.0379), while the setting time was independent of the material (p = 0.3770). Furthermore, there was no interaction between the material and the mass ratio for either working (p = 0.2260) or setting time (p = 0.5117).
DISCUSSION Information obtained from the manufacturers of resin-based luting agents indicated that similar quantities of activator and initiator are incorporated in the base and catalyst pastes, respectively. A certain amount of inhibitor is also added to both pastes. Ideally, the working or setting time for an individualmaterialshouldnotvarygreatlyasa functionofthe ratio of pastes. The present study has shown that a variation
vator and initiator. Theprod-
uct of the mass percent of the activator and the initiator used in that study ranged from 0.25 to 4.0. In the present study, increasing the catalyst paste mass means proportionally decreasing the base paste mass and vice versa. That means the actual changes in the product of the mass percent of the two components may not be asgreataschangesinthemassratio. Assumingthat, themass percent of either ingredient in the respective paste is 1.0%. The calculations show that as the mass ratio ranges from 0.7 to 1.3, the product of mass percent values increases from 0.97 up to 1.00 at a mass ratio of 1.0 and then decreases to 0.98. Compared with Asmussen's study (1981), this study used a smaller range of variation. Assuming that Asmussen's model is applicable to the materials used in the present study and the content of the inhibitor is 0.033%, the model predicts that the setting time will increase by approximately 13 s when the product is reduced from 1.0 to 0.97. This predicted difference was not possible to detect, as it is much lower than the variance of data observed in the present study. For specimens in the Vivadent group, the mean working time recorded exceeds the mean setting time recorded. At first these results may seem paradoxical, since setting occurs after working. However, one must realize that the setting times obtained in this study are not directly clinically relevant, because the specimens were immediately placed in a 37°C environment to provide a common initial condition. In clinical use, the materials remain at room temperature until working is finished, leading to longer setting times. Ahigh degree of variability in the data was caused by a few outliers. Factors that may be responsible for these outliers include temperature of the apparatus, oxygen inhibition, and ambient light exposure. The temperature varied only slightly from specimen to specimen (_+ I°C), so variation in temperature should not have significantly contributed to data variation. A Mylar strip was placed over the specimen surface in several past studies to prevent an oxygen-inhibited layer from Dental Materials~September 1994 317
forming (Ruyter, 1981; Watts et al., 1984; Yearn, 1985). The inhibited layer can be as thick as 80 I~m (Ruyter, 1981). The thickness between the exposed surface and the tip of the thermocouple is about 3 mm (Fig. 1). It is not likely that an oxygen-inhibited surface layer had a significant effect on the exotherm profile; however, oxygen incorporated into the bulk of the material may have affected the exotherm profile produced. Variation in exposure to ambient light is the most likely explanation for the outliers. Although dual-cured resins are designed to be activated by a wavelength of 470 nm, significant effects have been observed from radiation sources that produce other frequencies (Ruyter and Sj~vik, 1984; Watts et al., 1984; Yearn, 1985). The blue-white fluorescent lights that were present during mixing and curing emit radiation in the 380-720 nm range with a maximum output at 480 nm (Amick, 1947). Plastic molds are ineffective at screening blue to ultraviolet radiation (Watts et al., 1984; Yearn, 1985). If this testing procedure is used as an ISO standard, future specifications should include lighting conditions that lack the blue-violet range. ACKNOWLEDGMENTS The authors would like to thank Ben Lee for his services as a dental technician and for constructing the testing apparatus, This study was supported by NIH-NIDR Grants DE09307 and DE06672. Received July 5, 1994 / Accepted August 29, 1994 Address correspondence and reprint requests to: Jason A. Griggs
P.O. Box 100446, HSC University of Florida Gainesville, FL 32610-0446,USA
REFERENCES American Dental Association (1970). Guide to Dental Materials and Devices. 5th ed. Chicago: American Dental Association, 150-160.
318 Griggs et aL/Catalyst/base ratio vs. curing of resin luting agents
AmickCL(1947). Fluorescent Lighting Manual. 2nded. New York: McGraw Hill Book Company, Inc., 16. Asmussen E (1981). Setting time of composite restorative resins vs. content of amine, peroxide, and inhibitor. Acta Odontol Scand 39:291-294. Bovis SC, Harrington E, Wilson HJ (1971). Setting characteristics of composite filling materials. Brit Dent J 131:352356. B~ichelT (1993). Method proposed for testing dental resin based luting materials at Pforzheim, Germany, October 1993 by ISO TC106/SC1/WG12. Houston WJB, Miller MW (1968). Cements for orthodontic use. Dent Pract Dent Rec 19:104-109. Jacobsen PH, Von Fraunhofer JA (1974). Setting characteristics of anterior restorative materials. J Dent Res 53:461467. Plant CG, Jones IH, Wilson HJ (1972). Setting characteristics of lining and cementing materials. Brit Dent J 133:21-24. Ruyter IE (1981). Unpolymerized surface layers on sealants. Acta Odontol Scand 39:27-32. Ruyter IE (1985). Monomer systems and polymerization. In: Vanherle G, Smith DC, editors. Posterior Composite Resin Dental Restorative Materials. Proceedings of the International Symposium on Posterior Composite Resin Dental Restorative Materials: 1985 Jan, St. Maarten, MN. Netherlands: Peter Szulc Publishing Co., 109-137. Ruyter IE, Sj~vik IJ (1984). Light-activated restorative resin materials: Sensitivity to ambient light. JDentRes 63:579, Abstr. No. 66. SAS Institute (1985). SAS User's Guide: Statistics. 5th ed. Cary, NC: SAS Institute Inc., 433-506. Watts DC, Amer O, Combe EC (1984). Characteristics of visible-light-activated composite systems. Br Dent J 156:209-215.
Wilson HJ (1964). Amethodofassessingthesettingcharacteristics of impression materials. Brit Dent J 117:526-540. Wolcott RB, Paffenbarger GC, Schoonover IC (1951). Direct resinous filling materials: Temperature rise during polymerization. J Am Dent Assoc 42:253-263. Yearn JA (1985). Factors affecting cure of visible light activated composites. Inter Dent J 35:218-225.
Sensitivity of catalyst/base ratio on curing of resin luting agents: Polymerization exotherm analysis Jason A. Griggs, Chiayi Shen, Kenneth J. Anusavice Department of Dental Biomaterials, College of Dentistry, University of Florida, GainesviUe, Florida, USA
ABSTRACT Objectives. Currently, the proposed test of the International Standardization Organization (ISO) for measuring working and setting times of resin luting agents is based on measurement of times to reach specified stages on the polymerization exotherm. The objective of this study was to use this test to investigate the influence of variations in the mass ratios of catalyst paste to base paste on the working and setting times of three dual-cured dental resin luting agents, Methods. The materials used were Dicor Light Activated Cement (Dentsply International Inc.), Palfique Inlay Resin Cement (Tokuyama Soda Co.), and Vivadent Dual Cement (Vivadent). Fifteen specimens of each material were tested for working time by spatulating mass ratios from 0.7 to 1.3 for 30s at 23°C and recording the time from beginning of spatulation to the time at which a temperature increase occurs. Ten specimens of each material were tested for setting time by spatulating in a similar manner at 37°C and recording the time at which the temperature reaches a maximum value, Results. The data were fitted to the relation, In t = In A + Bm, where t is the time in seconds, m is the mass ratio, and A and B are regression coefficients. The results suggest that working and setting times of the specimens were independent of variations in mass ratio. A comparison among the materials was made by using a multiple linear regression with the relation, In t = In C + Dm + E7 + Fm~,, where ~, is a dummy variable to help distinguish between materials, and C, D, E, and F are regression coefficients. The results suggest that differences in materials influence the working time but not the setting time. Significance. These results infer that variations in mass ratio (+ 20%) often observed in the clinical setting should not have a significant influence on the working and setting times of resin luting agents, INTRODUCTION The increased popularity of all-ceramic restorations has led to the recent introduction of several two-paste dual-cured resin luting agents. Dual-cured resin luting agents consist of a base and a catalyst and can be cured through light activation and chemical-curedmechanisms. In clinical practice, light activa314 Griggs et aL/Catalyst/base ratio vs. curing of resin luting agents
tion is performed after the clinician has finished seating the ceramic prosthesis. The chemical-cured component of the dual-cured resins begins polymerization after the two pastes are mixed. The length of time from the start of mixing to the time of insertion of the prosthesis is the working time. The time from the start of mixing until the restoration no longer requires external support from a matrix is the setting time. For a chemical-cured resin, the mixture should remain moldable within the first 1.5 to 2 min from the start of mixing to faci]itate p]acement ofthe mixture in the cavity preparation, and then the material should set within 5 min (Jacobsen and VonFraunhofer, 1974). These times also apply to dual-cured resins, especia]]y the working time. Resin ]uting agents, such as chemical-cured composites, can fail to exhibit the desired working time. The effects ofhydroquinone concentration on inhibition ofpremature initiation, differentfi]lersurfaceareas on viscosity, TEC-DMA to Bis-GMA ratios on viscosity, and different initiator concentrations on free-radical production have been shown to be factors that influence curing times greatly (Ruyter, 1981; Ruyter, 1985). Normally, the clinician or the assistant will dispense equal volumes of base and catalyst pastes from separate containers. The ratio ofthe mass ofcatalyst paste to the mass ofbase paste wi]] hereafter be referred to as the "mass ratio". It is not ]ike]y that the ratio of the two pastes will be precise each time. Ideally, the working and the setting times should be insensitive to the mass ratio. A measurement of this sensitivity necessitates a standardized test for measuring the working and setting times of resin products. Wo]cott et al. (1951) analyzed the exotherm profiles of resinous filling materials during curing. They defined the working and setting times in termsoftimelapsesbetweenpeaktemperaturesintheexotherm profile. Wilson(1964) adapted a curemeter fromthe rubber industry to study curing times in impression materials. The meter consists of a stylus under cyclical loading, which traverses through the material. The rheology of the material is observed as a decrease in the amplitude of the trace as the material cures. Several investigations (Houston and Miller, 1968; Bovis et al., 1971; Plant et al., 1972) have used this equipment to define the working and setting times of resin luting agents according to rheological measurements.
Bovis et al. (1971) followed American Dental Association Specifications No. 8 and No. 9 (ADA, 1970) to test composite filling materials. These procedures describe the use of a Gillmore needle as a penetrometer to determine setting times in zinc phosphate cements and silicate cements, respectively. The setting time is defined as the time from the beginning of mixing to the point at which the Gillmore needle no longer penetrates the material under a specified load. They found the results to be highly sensitive to operator variability. A current test proposal (Btichel, 1993) involves measuring the exotherm of a chemical-cured resin as it cures over time. During the induction period, the activation of the initiator requires a certain amount of energy. As this energy is absorbed, the resin temperature decreases. The temperature decrease is soon offset by the exothermic reaction, and the temperature increases as the reaction accelerates. The temperature then reaches a maximum and begins to decrease as the reaction nears completion. A curve recording this thermal history will contain a minimum point and a maximum point, which are used in defining the working and setting times. The objective of this study was to investigate the influence of variations in the mass ratios of catalyst paste to base paste on the working and setting times of three dual-cured dental resins using the polymerization exotherm profiles.
, 'I ~ '
4
, 1 , ~' "P~I I , ,
~" (~ 4 ~_ //'\\/".7 ) ' / / / / / / / / / / / /
r : -" 2 .-
~
| __ _~'_. 1 - -~ -
1
/ / // /
MATERIALS AND METHODS Three dual-cured resins were chosen for study: Dicor Light Activated Cement, batches 513315 and 513386 (Dentsply International Inc., York, PA, USA), Palfique Inlay Resin Cement, batches RCA153 and RCB653 (Tokuyama Soda Co., LTD., Tsukuba-City, Japan), and Dual Cement, batches 560049 and 560051 (Vivadent, Schaan, Liechtenstein). All specimens of each material were prepared from the same batches, and all materials were mixed according to the manufacturers' instructions. The materials were all dual-cured resins; however, only the chemical-cured mechanism was activated so that the working and setting times recorded would represent the amount of time available to a clinician for manipulation of the material. Upon qualitative examination, the lack of light activation did not appear to detract from the degree of cure. A volume of catalyst paste was placed on a mixing pad, and the mass was recorded. An approximately equal volume by appearance of base paste was placed on the mixing pad, and the mass was recorded. To simulate clinical variation, no attempt was made to keep the paste masses constant by weight. A total of 15 specimens was prepared for measuring the working time of each material. After dispensing the two pastes by visual estimation of a ratio of 1.0, the first 10 specimens of each group varied in mass ratio from 0.8 to 1.2. For each material, an additional five specimens with mass ratios ranging from 0.7 to 1.3 were prepared to assess possible effects of excessive operator errors. A total of 10 specimens was prepared for measuring the setting time of each material. The same mass ratio variation, 0.8 to 1.2, also resulted in this group of specimens. No additional specimens were prepared to include greater mass ratio variations, A timer was started at the moment mixing began. At 30 s after the start of mixing, the material was placed in a mold, and the temperature was recorded at 1.4 s intervals by a
/ // / // / // // // /
~
J~
/
Fig 1. Apparatus for determination of working and setting time, polyethylene tubing(A), nylon block (B), stainless steel tube (C), copper/constantan (type T) thermocouple(D). All measurementsin ram,
thermocouple connected to an analog-to-digital converter (pMega Intelligent Scanning System, Omega Engineering, Inc., Stamford, CT, USA) to an accuracy of 0. I°C. The mold, shown in Fig. 1, consists of polyethylene tubing on a nylon block which has a stainless steel tube inserted through it containing a copper/constantan (type T) thermocouple. The working time was determined while maintaining the specimen mold at room temperature (23 _+1°C), and the setting time was determined with the mold in a constant temperature oven maintained at 37 _+l°C. The temperature fell slightly from T 1to T O(Fig. 2) after the material was inserted into the mold. The time at which the temperature began to increase denotes the start of the curing reaction. The working time (t w) was determined at this minimum. The time at which the temperature began to fall again Dental Materials~September 1994 315
103
~
,
,
T2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dicor
o~ Start of mixing Insertion
~c~ I--
30 s
tw
102
I
I
I
103
I
I
I
I
I Palfique
E
ts
Time
~
o
o
Fig 2. T y p i c a l recording trace showing temperature changes with time for determination of working and setting times.
102
~
i
~
103
i
i
i
t i
Vivadent 0
0
~
¢K)n
0
from T 2 denotes the end of the curing reaction. The setting time (t s) was determined at this maximum. The recorded times and the measured mass ratio were used to determine the effect of mass ratio on the working and setting time by a linear regression analysis for each material. The effect of material on the working and setting time was evaluated by a multiple linear regression model using both material and mass ratio as the independent variables.
1o2 0.4
, 0.6
O
Q
0
, , 0.8 1.0 1.2 Cataryst/Base Ratio
1.4
Fig 3. Working time vs. mass ratio.
RESULTS The natural logarithms of the times were used in the analysis because of the distribution of the recorded values of working and setting times. To determine the influence of the mass ratio on the times measured for each group, several mathematical models were examined, and the following simple linear regression model demonstrated the best fit of data: In t = In A + Bm
lo3 Dicor
o ~ ~
(1) 10 2
where t is the working or setting time in seconds, A is a constant in seconds, B is a constant that dictates the effect of mass ratio on the working or setting time, and m is the mass ratio. The null hypothesis to be tested was Ho: B=0. The equivalent form of equation (1) is: t = Ae B=
lO3 o ~- v - o,,~
"~ e ~= ==
O
Palfique
o
7'
(2)
For the multiple linear regression, the following mode] was
102
i
lO 3
,
used:
Vivadent
lnt=lnC+Dm+ET+Fmy
where t is the working or setting time in seconds, C is a constant in seconds, m is the catalyst/base ratio, ?is a dummy variable t h a t assigns a numerical code to each of the three materials, and D, E, and F are the coefficients for re, T, and my terms, respectively. The values of A and B were determined by a linear regression of In t against m. The intercept of the straight line at m=0 is the value of In A, and the slope of the same line is the 316 Griggs et al./Catalyst/base rafio vs. curing of resin luting agents
o¢
(3)
~ : ~ o %
10 2
o.s
0.8 Catalyst/Base
Fig 4.
Setting time vs. mass ratio.
I 1.0 Ratio
o
1.2
TABLE 1: COEFFICIENTS OF CURVES FIT TO CURING TIMES VS. MASS RATIO DATA USING THE RELATION, In t = In A + Bm. Working Time Group
occur routinely.
Statistical
analysis has confirmed that both working and setting times are not sensitive to the
Setting Time
Mean +_S.D.
of 20% in the mass ratio can
Mean +_S.D.
In A
B
In A
B
degree of mass ratio varia-
Dicor
5.53 + 0.12
-0.117 + 0.143 (p = 0.4304)
5.64 + 0.13
-0.099 + 0.139 (p = 0.4950)
tion t h a t m a y be occurring during routine clinical use.
Palfique
5.69 + 0.53
-0.534 + 0.484 (p = 0.2905)
6.67 + 0.46
0.366 + 0.493 (p = 0.4794)
Vivadent
5.99 + 0.11
-0.113 + 0.161 (p = 0.4963)
6.19 + 0.24
0.133 + 0.241 (p = 0.5957)
Asmussen (1981) investigated the effect of various combinations of activator, initiator, and inhibitor on the curing time. He varied the mass percent of the activator between 0.5% and 3.0%, the
initiator from 0.25% to 2.0%, and the inhibitor from 0% to TABLE 2: COEFFICIENTS ASSOCIATED WITH MULTIPLE LINEAR REGRESSION ANALYSIS OF THE EQUATION, In t = In C + Dm + ET + Fmy. Working Time
Setting Time
1.0%. He concluded t h a t the curing t i m e is proportional
In C
D
E
F
to the mass percent of the
4.71 + 0.53
0.505 + 0.236
0.900 _+ 0.578
-0.306 + 0 . 2 4 9
inhibitorbutisinverselypro-
(p = 0.1270)
(p = 0.0379)
(p = 0.2260)
portional to the product of the mass percents of the acti-
0.523 _+ 0.535
0.389 + 1.222
-0.371 + 0.557
(p = 0.7525)
(p = 0.3370)
(p = 0.5117)
5.34 + 1.16
value of B. Table 1 shows the results of the regression by a general linear model of SAS (1985). The statistical analysis shows that the p-values for B range from 0.2905 to 0.5957. Thus, the null hypothesis was accepted for both working and setting times. This means that the range of mass ratio variation in this study had no influence on either the working or setting times of the resin luting agents tested. Therefore, the values of A shown in Table 1 are the means of the experimental data. The mean working times were 280 _+20 s, 499 _+83 s, and 448 + 40 s for Dicor, Palfique, and Vivadent groups, respectively. The mean setting times were 310 _+12 s, 568 _+90 s, and 430 t 29 s for Dicor, Palfique, and Vivadent groups, respectively. The experimental data along with the best fit curves of the three materials tested are shown in Figs. 3 and 4 for working and setting times, respectively, Table 2 shows the results of the multiple linear regression, In this analysis, the interest was two-fold, i.e., whether the working or setting times differed among the materials tested and whether there was an interaction between the material and the mass ratio. The results show that the working time was significantly influenced by the materials (p = 0.0379), while the setting time was independent of the material (p = 0.3770). Furthermore, there was no interaction between the material and the mass ratio for either working (p = 0.2260) or setting time (p = 0.5117).
DISCUSSION Information obtained from the manufacturers of resin-based luting agents indicated that similar quantities of activator and initiator are incorporated in the base and catalyst pastes, respectively. A certain amount of inhibitor is also added to both pastes. Ideally, the working or setting time for an individualmaterialshouldnotvarygreatlyasa functionofthe ratio of pastes. The present study has shown that a variation
vator and initiator. Theprod-
uct of the mass percent of the activator and the initiator used in that study ranged from 0.25 to 4.0. In the present study, increasing the catalyst paste mass means proportionally decreasing the base paste mass and vice versa. That means the actual changes in the product of the mass percent of the two components may not be asgreataschangesinthemassratio. Assumingthat, themass percent of either ingredient in the respective paste is 1.0%. The calculations show that as the mass ratio ranges from 0.7 to 1.3, the product of mass percent values increases from 0.97 up to 1.00 at a mass ratio of 1.0 and then decreases to 0.98. Compared with Asmussen's study (1981), this study used a smaller range of variation. Assuming that Asmussen's model is applicable to the materials used in the present study and the content of the inhibitor is 0.033%, the model predicts that the setting time will increase by approximately 13 s when the product is reduced from 1.0 to 0.97. This predicted difference was not possible to detect, as it is much lower than the variance of data observed in the present study. For specimens in the Vivadent group, the mean working time recorded exceeds the mean setting time recorded. At first these results may seem paradoxical, since setting occurs after working. However, one must realize that the setting times obtained in this study are not directly clinically relevant, because the specimens were immediately placed in a 37°C environment to provide a common initial condition. In clinical use, the materials remain at room temperature until working is finished, leading to longer setting times. Ahigh degree of variability in the data was caused by a few outliers. Factors that may be responsible for these outliers include temperature of the apparatus, oxygen inhibition, and ambient light exposure. The temperature varied only slightly from specimen to specimen (_+ I°C), so variation in temperature should not have significantly contributed to data variation. A Mylar strip was placed over the specimen surface in several past studies to prevent an oxygen-inhibited layer from Dental Materials~September 1994 317
forming (Ruyter, 1981; Watts et al., 1984; Yearn, 1985). The inhibited layer can be as thick as 80 I~m (Ruyter, 1981). The thickness between the exposed surface and the tip of the thermocouple is about 3 mm (Fig. 1). It is not likely that an oxygen-inhibited surface layer had a significant effect on the exotherm profile; however, oxygen incorporated into the bulk of the material may have affected the exotherm profile produced. Variation in exposure to ambient light is the most likely explanation for the outliers. Although dual-cured resins are designed to be activated by a wavelength of 470 nm, significant effects have been observed from radiation sources that produce other frequencies (Ruyter and Sj~vik, 1984; Watts et al., 1984; Yearn, 1985). The blue-white fluorescent lights that were present during mixing and curing emit radiation in the 380-720 nm range with a maximum output at 480 nm (Amick, 1947). Plastic molds are ineffective at screening blue to ultraviolet radiation (Watts et al., 1984; Yearn, 1985). If this testing procedure is used as an ISO standard, future specifications should include lighting conditions that lack the blue-violet range. ACKNOWLEDGMENTS The authors would like to thank Ben Lee for his services as a dental technician and for constructing the testing apparatus, This study was supported by NIH-NIDR Grants DE09307 and DE06672. Received July 5, 1994 / Accepted August 29, 1994 Address correspondence and reprint requests to: Jason A. Griggs
P.O. Box 100446, HSC University of Florida Gainesville, FL 32610-0446,USA
REFERENCES American Dental Association (1970). Guide to Dental Materials and Devices. 5th ed. Chicago: American Dental Association, 150-160.
318 Griggs et aL/Catalyst/base ratio vs. curing of resin luting agents
AmickCL(1947). Fluorescent Lighting Manual. 2nded. New York: McGraw Hill Book Company, Inc., 16. Asmussen E (1981). Setting time of composite restorative resins vs. content of amine, peroxide, and inhibitor. Acta Odontol Scand 39:291-294. Bovis SC, Harrington E, Wilson HJ (1971). Setting characteristics of composite filling materials. Brit Dent J 131:352356. B~ichelT (1993). Method proposed for testing dental resin based luting materials at Pforzheim, Germany, October 1993 by ISO TC106/SC1/WG12. Houston WJB, Miller MW (1968). Cements for orthodontic use. Dent Pract Dent Rec 19:104-109. Jacobsen PH, Von Fraunhofer JA (1974). Setting characteristics of anterior restorative materials. J Dent Res 53:461467. Plant CG, Jones IH, Wilson HJ (1972). Setting characteristics of lining and cementing materials. Brit Dent J 133:21-24. Ruyter IE (1981). Unpolymerized surface layers on sealants. Acta Odontol Scand 39:27-32. Ruyter IE (1985). Monomer systems and polymerization. In: Vanherle G, Smith DC, editors. Posterior Composite Resin Dental Restorative Materials. Proceedings of the International Symposium on Posterior Composite Resin Dental Restorative Materials: 1985 Jan, St. Maarten, MN. Netherlands: Peter Szulc Publishing Co., 109-137. Ruyter IE, Sj~vik IJ (1984). Light-activated restorative resin materials: Sensitivity to ambient light. JDentRes 63:579, Abstr. No. 66. SAS Institute (1985). SAS User's Guide: Statistics. 5th ed. Cary, NC: SAS Institute Inc., 433-506. Watts DC, Amer O, Combe EC (1984). Characteristics of visible-light-activated composite systems. Br Dent J 156:209-215.
Wilson HJ (1964). Amethodofassessingthesettingcharacteristics of impression materials. Brit Dent J 117:526-540. Wolcott RB, Paffenbarger GC, Schoonover IC (1951). Direct resinous filling materials: Temperature rise during polymerization. J Am Dent Assoc 42:253-263. Yearn JA (1985). Factors affecting cure of visible light activated composites. Inter Dent J 35:218-225.