-
Correlation between hardness and degree of conversion during the setting reaction of unfilled dental restorative resins
Jack L. Ferracane Dental Materials Science Program, Baylor College of Dentistry, Dallas, Texas, USA.
Ferracane JL. Correlation between hardness and degree of conversion during the setting reaction of unfilled dental restorative resins. Dental Materials 1985; 1: 11-14.
Abstract. - Knoop hardness was correlated to degree of conversion (DC) of carbon double bonds, determined by IR, during the setting reaction of three unfilled dental resins. For a specific resin, increase in hardness correlates well with increases in DC during setting. However, an absolute hardness number cannot be used to predict a DC when comparing different resins. The two techniques for determining extent of cure in dental resins cannot always be used interchangeably, since each is sensitive to different variables.
Key words: hardness, degree of conversion, resins, composites, extent of cure.
Dr. Jack L. Ferracane, Dental Materials Science Program, Baylor College of Dentistry, 3302 Gaston Avenue, Dallas, TX 75246, USA.
Accepted for publication 21 August 1984.
It is desirable for a dental restorative resin to convert all of its monomer to polymer during the polymerization re- action. However, with Bis-GMA-based resins there is always a significant con- centration of unreacted carbon double bonds remaining in the resin when it is cured at or near the oral temperature. This is believed to be due mainly to limitations on the mobility of reactive species imposed by the rapid formation of a cross-linked polymeric network. Since the extent of cure may exert an effect on nearly every physical property of the resin system (including: me- chanical properties, solubility, dimen- sional stability, color change and bio- compatibility), degree of conversion may play an important role in deter- mining the ultimate success of the re- storative.
The degree of conversion in Bis- GMA-based restorative resins has been analyzed with infrared spectroscopy by several researchers and has been shown to be in the range of 50 to 80% (14) . However, it is common practice to measure mechanical properties as a means of indirectly evaluating degree of cure.
In 1982, Tirtha et al. (5) evaluated the Barcol hardness and transverse
strength of a series of light-cured dental composite resins to determine depth of cure. Barcol hardness was also used by Leung et al. (6) to determine depth of cure in visible light-cured composites as a function of curing and illumination time. Although there are several more reports in the literature in which degree of cure has been evaluated indirectly by measuring mechanical properties, there are only a few studies which have com- pared these indirect modes to IR tech- niques.
Asmussen (7) reported a good corre- lation between Wallace hardness, di- ametral tensile strength and degree of conversion by IR for experimental composite formulations in 1982. More recently, Ferracane (8) measured the degree of conversion in unfilled resins by IR and correlated the results with those obtained from tests of several mechanical properties. He found a good correlation between the degree of conversion and the storage modulus and glass transition temperature results obtained by performing dynamic me- chanical testing over a broad temper- ature range. The correlation between degree of conversion and ambient tem- perature properties was not as strong, however.
Objectives
The objective of this study was to de- termine the nature of the correlation between the Knoop hardness and the degree of conversion of carbon double bonds, as determined by IR analysis, for unfilled dental restorative resins. An attempt was also made to compare the time frames at which maximum hardness and degree of conversion are acquired in these quick-setting resins.
Methods
Three commercially available, chemi- cally-cured Bis-GMA-based resins were used in this study. They were: Delton Pit and Fissure Sealant (Johnson & Johnson, East Windsor, N J), Profile bonding agent (S. S. White Dental Products International, Phila- delphia, PA) and Concise Enamel Bond (Dental Products/3M, St. Paul, MN). The Profile bonding agent (Profile BA) and the Concise Enamel Bond (Concise EB) were supplied with their respective composite resin resto- ratives. These unfilled resins were chosen on the basis of differences in monomer compositions (9) and viscos- ities, which were expected to produce
-
12 Ferracane
resins with differences in degree of con- version (8).
The degree of conversion (DC) was analyzed with a Fourier Transform In- rared (FTIR) Spectrometer (FX6250- Analect Instruments, Irvine, CA) in transmission mode. The conversion during the first 30 rain of the reaction was followed by polymerizing the resin between AgC1 windows in the path of the IR. Spectra were acquired at time intervals of 1 min, 2 min, 3 min, 4 min, 5 min, 10 min, 20 min and 30 min under a nitrogen gas purge to insure a clean, stable background*. During this time period, the temperature in the specimen chamber rose from 34~ to 40~ Three runs were made for each resin system.
The DC values for the curing periods of 60 rain, 120 min, 6 h and 24 h were determined by polymerizing a drop of mixed resin between glass slides at 37~ for the appropriate time period. A thin resin film (2040 p,m) was produced in this manner. These films were peeled from the slides and ana- lyzed on the FTIR in transmission, ac- cording to a procedure outlined in detail previously (4). Three films were tested for each resin at each time period.
The DC was calculated by comparing the absorbance ratio of the C = C peak at 1640 cm 1 to the unchanging aro- matic ring peak at 1610 cm -a for the pre- and post-polymerized resins, using standard baseline procedures (2, 4).
Specimens were made for Knoop hardness (KHN) testing by curing the resins in a 6 mm x 3 mm steel disk mold for the appropriate time periods (5 min, 10 min, 20 min, 30 min, 60 min, 120 rain, 6 h and 24 h) at 37~ The
* Collection of the spectra took approx- imately 30 sec at each time period.
100
DC (%) vs log t
Q
90
80
70
60
50
40
30
20
10
0
. . . . . :=~r -=.~r-= - - . - - ' '~ ' '~ '~ '~
~:f2 .&- / t~r~ ~ ~ ~ - -
c i " .9" ~ . .~ !
/ ~/" ~ J
// " /
i / / 9
r / / / / /
o- . . . . . . -z De l ton
e - . - - . - - . Conc ise (EB)
9 - - - -A P ro f i le (BA)
I I I 10 100 1 ,000
t(min)
10,000
Fig. 1. Degree of Conversion (DC) vs. log time for three unfilled resins. Each point repre- sents an average of three values.
specimens were sanded fiat with 320 emery paper before the initial test to insure that any air-inhibited surface layer of resin had been removed. At least 10 specimens were made for each resin. An individual specimen was tested at alternate time periods to insure that it had been allowed to cure at 37~ for the entire specified time before testing. Each hardness reading was made at 23~ under a 100 g toad on a Tukon hardness tester (Page-Wilson Corp., Bridgeport, Conn.) using a Knoop diamond indenter. At least five values were recorded at each time in- terval from five different specimens for each resin.
The DC and hardness results were analyzed by analysis of variance and Scheffes test for making multiple com- parisons between unconfounded means at the p -< 0.05 level (10). Regression
analysis was performed to determine the correlation between DC and KHN. The correlation coefficient was ana- lyzed for significance at the p --< 0.05 level using a t-test for each resin (11).
Resu l ts
The three self-curing unfilled resins hardened very rapidly, with initial setting times of approximately 1~ min for Delton and Concise (EB) and 2 min for Profile (BA). Analysis of the DC with respect to time (Fig. 1) revealed that initial setting occurred when fewer than 40% of the available carbon double bonds had reacted in these sys- tems. After the first 5 to 10 min of cu- ring, the increase in DC was approx- imately linear with time through 24 h. No increase in DC was observed for any resin at 48 h. Therefore, DC at 24
Table 1. Hardness (KHN) vs DC during setting
Delton Profile (BA) Concise (EB) Time DC (%)* KHN (kg/mm2) + DC KHN DC KHN
5 min 62.4_+2.0 10 min 65.8_+0.7 20 min 68.2_+0.7 30 min 69.4_+0.7 60 min 70.5_+0.6
120 min 71.2_+0.7 6 h 74.7_+0.5
24 h 77.6+1.8
9.8_+0.2[ 43.4_+2.8 10.0+0.9 47.5+1.6 10.9-+0.9 50.6+1.9 12.0_+0.7 51.8_+1.5 13.3_+0.8 55.7+-0.5 14.9_+0.9 57.2+0.9 19.0_+I.0 58.6_+0.2 18.6_+1.5 62.0+_0.8
8.1-+0.8 58.5+1.0 9.4+0.9 I 62.7+1.3
10.6-+0.5 66.0-+1.4 11.2-+0.4 66.9+_2.0 13.1+0.5 67.7-+1.1 15.3_+0.6 70.1+0.3 18.2+ 1.4 72.2-+0.7 18.9-+0.8 76.9-+0.6
6.4+0.3 7.6+0.3 9.1_+0.7 9.7_+0.2
11.2_+0.4 12.7+_0.5 17.5-+0.3 17.6-+0.7
Values are mean +- standard deviation. Means connected by bars were not significantly different (p _< 0.05). * b=2. + b_>5.
-
Correlation between hardness and conversion in dental resins 13
h will be referred to as maximum DC. The DC values at 24 h were equiv-
alent for Delton and Concise (EB) and were approximately 15% greater than that of Profile (BA) (Table 1). Despite the differences in DC between the re- sins, there were no significant diffe- rences between the hardness values at 24 h. In fact, the hardness values for Delton and Profile (BA) were statisti- cally equivalent at all time periods, even though the DC was always 15 to 20% greater for Delton. The hardness of Concise (EB) was significantly lower than that of Delton at 5 and 10 rain and at 30 through 120 rain, but the DC values for the two resins were statis- tically equivalent at all time periods.
It is apparent from the results in Table 1 that both the hardness and DC continue to increase with time after setting for these resins. However, the acquisition of maximum (24 h) hardness and DC are not synchronized. Fig. 2 reveals that at 5 min, DC for Delton has reached 70-80% of its maximum value, while less than 55% of the maximum hardness has been at- tained. Both properties showed rather linear increases with time from 10 min to 6 h, but the slope for hardness was much greater. The curves for Profile (BA) and Concise (EB) were similar to that for Delton. There was no signif- icant increase in hardness beyond 6 h, and the only significant increase in DC after 6 h occurred with Concise (EB).
Regression analysis revealed a strong (r -> 0.937) and significant (p -< 0.05) correlation between DC and hardness with time for all three resins (Fig. 3). However, the correlation between hardness and DC was very poor (r = 0.498) when the regression analysis was repeated to include all three resins at all time periods (Fig. 3).
Discuss ion
For a dental restorative resin, there ap- pears to be a good correlation between increasing hardness and increasing degree of conversion during the setting reaction. However, the acquisition of hardness chronologically lags behind the conversion of carbon double bonds. The mechanical properties of these resins are influenced by the cross- linking and network formation taking place during setting. This network for- mation occurs after an initial stage of polymer chain propagation (12). A much greater percentage of carbon double bonds are reacted to form
DC fit KHN % o f Max vs log t
100
90
80
,~ 60
5O
3O
2O
10
-- -- DC
: : KHN
1 I I 10 100
t (min)
1,000 10 ,000
Fig. 2. Comparison of the percent of maximum DC and maximum hardness (KHN) vs. log time for Delton. Each point represents an average of three values for DC and at least 5 values for KHN.
polymer chains than are reacted to crosslink existing chains. Therefore, the greatest increase in hardness occurs during a period in which very subtle changes in DC take place, i.e. after 85% of the total conversion has been achieved.
Despite the correlation between hardness and DC during setting, a
specific hardness value cannot be cor- related to a specific DC when com- paring different resin formulations. For example, Delton achieves a DC of 62.4% at 5 min with a hardness value of 9.8 KHN. Concise (EB) achieves 62.7% conversion at l0 rain, with a hardness value of 7.6 KHN. The maximum DC for Profile (BA) i s
KHN vs DC.
9 Z [] /~t 9 / H
18.0 Z /~/e' I 9 ,;I"
/ " , / ) / 16.0 I / / / all points
r=~498 / 9 0 / / " r=.498 / / "
14.0 / /
- / /2 ~
C / 9 / ,,;P% / * / ,,;:/ - E 1oo / [] .,,:?.
~ ,,::/ 9
Z 8.0
"r" 0.o / / " ,/~/ I , / I / i
/ /~,. 4.0 /~/"
,S / o- . . . . . . -~ De l ton r = .937 9 o - . - - . - -e Conc ise (EB) r = .944
/7 2.0 / ,y * - - - - - -* P ro f i le (BA) r = .961 "7
~/- , , , 9 / / _ _ 50 60 70 0 100
DC (%)
Fig. 3. Correlation between KHN and DC for three individual resins and for all resins com- bined (solid line). Each point represents an average value for the different time periods.
-
14 Ferracane
62.0%. At this DC, Profile (BA) has a hardness of 18.9, approximately twice that of Delton and Concise (EB). Al- though the higher concentration of BIS-GMA in Profile (BA) contributes to its higher viscosity and to a lower DC (8), the hardness at ambient tem- perature is not similarly reduced. The discrepancy lies in the nature of the two experiments.
Infrared spectroscopic analysis of the degree of conversion of monomer to polymer in dental resins is a very ac- curate and reproducible technique. However, the conversion of a carbon double bond to a single bond during the setting of dimethacrylate resins may occur via several mechanisms, in- cluding: chain elongation, cross- linking, reaction with initiator radicals.
As stated previously, the mechanical properties of a resin system are very much dependent upon the cross-linking density and the quality of the network which forms during polymerization. Of less importance is the overall con- version of reactive species. The two
9 modes for evaluating degree of cure are sensitive to different variables. There- fore, it is possible that two resins with significantly different degrees of con- version may have identical properties when tested at ambient temperature, as has been shown previously (8) and in this study. This restricts the use of in- direct modes for testing degree of con- version. It would not be correct to indi- rectly determine the degree of cure in two different resin systems by evaluat- ing their hardness, when one is interes- ted in the effect of residual reactive species on biocompatibility or dimen- sional change.
Conclusions
The acquisition of hardness during the setting of unfillled dental restorative resins correlates well with the increase in degree of conversion for a specific resin. However, an absolute hardness value cannot be used to predict an ab- solute value for degree of conversion in all resins. This is due to the fact that the mechanical properties of the resin are very much dependent upon network formation, which is not equivalent to conversion in these materials. There- fore, indirect methods for measuring the degree of cure for comparison among resins are limited. It does appear to be valid, however, to use these indirect tests to predict the re- lative degree of conversion for a
specific resin at different time periods or under variable conditions.
Acknowledgements - The author would like to thank Analect Instruments, Irvine, CA for the use of the FX6250 FTIR spectropho- tometer used in this study.
References
1. RUYTER IE, GYOROSI PP. An infrared spectroscopic study of sealants. Scand J Dent Res 1976: 84: 396--400.
2. RUYTER ID, SVENDSEN SA. Remaining methacrylate groups in composite re- storative materials. Acta Odontol Scand 1978: 36: 75-82.
3. ASMUSSEN E. Factors affecting the quantity of remaining double bonds in re- storative resin polymers. Scand J Dent Res 1982: 90: 490-6.
4. FERRACANE JL, GREENER EH. Fourier transform infrared analysis of degree of polymerization in unfilled resins - methods comparison. J Dent Res (in press).
5. TIRTHA R, FAN PL, DENNISON JB, POWERS JM. In vitro depth of cure of photo- activated composites. J Dent Res 1982: 61: 1184-7.
6. LEUNG RL, FAN PL, JOHNSTON WM. Post-irradiation polymerization of visible light-activated composite resins. J Dent Res 1983: 62: 363-5.
7. ASMUSSEN E. Restorative resins: hardness and strength vs. quantity of re- maining double bonds. Scand J Dent Res 1982: 90: 484-9.
8. FERRACANE J. The correlation between the physical properties and degree of conversion in unfilled Bis-GMA-based dental resins. Evanston, Illinois: Northwestern University. 1983. 314 pp. Dissertation.
9. RUYTER IE, SJOVIK IJ. Composition of dental resins and composite materials. Acta Odontol Scand 1981: 39: 133-46.
10. CICCHETTI DV. Extension of multiple-range tests to interaction tables in the analysis of variance: a rapid approximate solution. Psych Bull 1972: 77: 405-8.
11. SPIEGEL MR. Probability and Statistics. Schaum Outline Series. New York: McGraw Hill~ 1975.
12. HORIE K, OTAGAWA A, MURAOKA, M, MITA I. Calorimetric investigation of polymerization reactions. V. crosslinked copolymerization of methylmetha- crylate with ethylene dimethacrylate. J Polym Sci Polym Chem Ed 1975: 13: 445-54.
