- Email: [email protected]
Post-curing in dental resin-based composites
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Post-curing in dental resin-based composites William Germscheid a , Louis Gosse de Gorre a , Braden Sullivan b , Catherine O’Neill b , Richard B. Price b , Daniel Labrie a,∗ a b
Department of Physics and Atmospheric Science, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada Department of Dental Clinical Sciences, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To determine the post-curing in six commercial contemporary resin-based com-
Received 3 April 2018
posites (RBCs) using axial shrinkage, the degree of conversion, and Vickers hardness.
Received in revised form 6 June 2018
Methods. Five Bulk Fill and one conventional RBCs from three companies were selected with
Accepted 7 June 2018
a wide range of filler volume content. The axial shrinkage of samples that were 1.00 mm thick by 9–10 mm diameter was measured using a modified bonded disk method over a time between 15 h and 19 h at temperatures of 26 ◦ C and 34 ◦ C (mouth temperature). The degree
Keywords:
of conversion (DC) was collected continuously for 10 min using mid-infrared spectroscopy
Resin composite
in the attenuated total reflectance geometry. Vickers hardness was measured at 1 h post-
Polymerization
irradiation using a load of 300 gf. For all three tests, the samples were irradiated at five
Shrinkage
exposure times, 20, 5, 3, 1.5 and 1 s with a light curing unit radiant exitance of 1.1 W/cm2 .
Degree of conversion
Three samples (n = 3) were used for each experimental condition.
Microhardness
Results. After light exposure, the axial shrinkage and degree of conversion exhibited a func-
Kinetics
tional time dependence that was proportional to the logarithm of time. This suggests an
Temperature
out-of-equilibrium polymer composite glass that is transitioning to thermal equilibrium. At a sufficiently long time and among the RBCs investigated, the shrinkage related physical aging rate was found to vary between 1.34 and 2.00 m/log(t). The rate was a function of the filler content. Furthermore, 15 h after light exposure, the post-curing shrinkage was estimated to be an additional 22.5% relative to the shrinkage at 100 s for one RBC at T = 34 ◦ C. The hardness in the photo-cured RBC was varied by using different light exposure times. The first two experimental techniques show that the higher the initial DC 10 min after light exposure, the smaller is the post-curing shrinkage related and DC related physical aging rates. A direct correlation was observed between the shrinkage related and the DC related physical aging rates. Significance. Post-curing shrinkage should be evaluated for longer than 1 h. The post-curing shrinkage 15 h after light exposure in dental RBCs can be appreciable. The long-term development of built-in stress within the tooth wall structure may shorten the restoration’s lifespan. © 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.
∗
Corresponding author. E-mail address: [email protected] (D. Labrie). https://doi.org/10.1016/j.dental.2018.06.021 0109-5641/© 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.
1368
1.
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Introduction
A paradigm shift in the field of dentistry occurred with the introduction of photo-cured resin-based composites (RBCs) as an alternative to amalgam in dental restorations [1–3]. The main driving force is that the restored teeth can look like a natural decay free tooth. Due to the fundamental nature of the resin used in RBCs, after photo-curing, polymerizationinduced shrinkage strain develops with time. This produces shrinkage-induced stress within the cavity walls of the teeth [4]. There is evidence that such stresses may result in enamel cracking and tooth post-operative pain [5], margin debonding [6], secondary caries [7], and premature failure of the restoration [8]. Dental manufacturers have taken different approaches to relieve the stress developed both during and after photo-curing by altering the monomers [9,10], or by incorporating pre-polymerized filler particles [11] in RBCs. After the light curing unit (LCU) is turned off, one important contribution to shrinkage has been attributed to the continued cross-linking as the unterminated radicals continue to react, albeit rather slowly. This process is sometimes referred as the densification of the cured RBC with time [12] because it leads to an increased RBC density and elastic modulus. Several studies have been carried out in cured RBCs to investigate how the degree of conversion [13–17], glass transition temperature [18], axial shrinkage [19], density and mechanical properties [20–22] vary with time for up to a month after light exposure. It has been reported that during the postcuring period, the elastic modulus [21], ion viscosity [21], and degree of conversion (DC) [23] were found to be proportional to the logarithm of time, but no explanation was given for this unique time dependence. Post-curing was investigated in experimental photo-cured resins using six experimental techniques [20]. An overall explanation of the results was offered using the free volume fraction model to explain the properties of out-of-equilibrium amorphous polymer glasses [24–27]. In this model, an out-of-equilibrium amorphous polymer glass can have an excess of free volume in the form of a large number of sub-nanometer-sized voids dispersed throughout the polymer glass. Due to the thermal diffusion of these excess voids out to the surface, the polymer continues to shrink with time. As indicated in Fig. 2 of Ref. [27], after a relatively short time, the decrease in sample thickness can be predicted to be proportional to the logarithm of time [log(t)]. The observation of the log(t) dependence for the volume shrinkage of an out-of-equilibrium RBC in the glass state provides a direct evidence for this model. In the above study [20], the sample axial shrinkage was measured continuously over a period of 10 h. The axial shrinkage varied in a non-monotonic manner with time and, as a result, was inconclusive toward testing the applicability of the free volume fraction model to photocured methacrylate-based resins. In another study [19], the axial shrinkage on RBCs was measured continuously over a period of 1 h; this time period was too short to observe the log(t) dependence. The free volume fraction model has been studied extensively in the field of polymer glasses and composites [27–31]. The process for the out-of-equilibrium polymer glass to reach thermal equilibrium is commonly referred to as physical
aging. In this work, the constant of proportionality between the glass parameter such as axial shrinkage (or DC) and the logarithm of time is called the axial shrinkage (or DC) related physical aging rate [28]. To optimize dental RBC blends and photo-curing conditions with the goal of minimizing post-curing shrinkage, it is vital to understand the underlying mechanisms governing post-curing. Furthermore, it is also important to quantify the post-curing shrinkage in RBCs at times longer than 1 h. In this study, the post-curing of six commercial contemporary RBCs was investigated using axial shrinkage and degree of conversion as a function of time for different light exposure times, and hardness measured one hour after light exposure for different light exposure times. The research hypotheses were: (1) at a sufficiently long time, the axial shrinkage is not proportional to the logarithm of time, (2) the axial shrinkage related physical aging rate, ˛S , does not vary with the light exposure time, (3) at a sufficiently long time, the degree of conversion is not proportional to the logarithm of time, (4) the degree of conversion related physical aging rate, ˛DC , does not vary with the light exposure time, (5) ˛S and ˛DC do not vary with the degree of conversion, (6) ˛DC is not correlated with ˛S , and (7) ˛S and ˛DC are not correlated with the hardness.
2.
Materials and methods
2.1.
Materials
Six commercial contemporary RBCs were used in this study. They consisted of five Bulk Fill and one conventional dental RBCs. The RBC compositions as described by the manufacturers are reported in Table 1. Two low (≤45 vol.%) filler content (3M Filtek Bulk Fill Flowable Restorative A2 shade and DENTSPLY Surefil SDR flow Posterior Bulk Fill A3 shade), and four high (>58 vol.%) filler content (3M Filtek Bulk Fill Posterior Restorative A2 shade, Filtek One Bulk Fill Restorative A2 shade, and Z100 Restorative A2 shade, and Voco GmbH X-tra fil Universal shade) were used to study the effect of time and filler volume fraction on the shrinkage and degree of conversion. Temperatures of 26 ◦ C (representing a warm room and the minimum after washing and drying a cavity) and 34 ◦ C (representing a maximum intraoral temperature inside a cavity) [32,33].
2.2.
Sample preparation and photo-curing conditions
For all three experimental techniques used, the sample geometry and sample dimensions, and photo-curing conditions were the same. The substrate was a 25.4 mm diameter by 3 mm thick quartz disk. One surface was lightly sandblasted and then silanized using a ceramic primer (RelyX Ceramic Primer, 3M, St Paul, MN, USA). A 1.00 mm thick by 9–10 mm diameter RBC sample was inserted at the center of the substrate. For the axial shrinkage measurements, a 100 m thick glass coverslip disk was placed on top of the sample. The coverslip was supported along its edge by a 1.00 mm thick brass
1369
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Table 1 – Composition of the resin-based composites as provided by the manufacturers. Product name
Manufacturer (lot number)
Filtek Bulk Fill Flowable Restorative A2 shade
3M ESPE (N632976)
BisGMA, UDMA, bisEMA(6) and Procrylat monomers
Filtek bulk fill posterior restorative A2 shade
3M ESPE (N894570, N793515, N723218, N772739, N761323, N833958) 3M ESPE (N889317)
Proprietary AUDMA and AFM, DDDMA and UDMA
3M ESPE (N232301) DENTSPLY (120127)
BisGMA, TEGDMA SDR patented UDMA, DMA, di-functional diluent BisGMA, UDMA, TEGDMA
Filtek One Bulk Fill Restorative A2 shade
Z100 Restorative A2 shade SureFil SDR flow posterior bulk fill A3 shade X-tra fil universal shade
Voco GmbH (1525575)
Resin matrix
Proprietary AUDMA and AFM, DDDMA and UDMA
Filler 10 nm–3.5 m zirconia/silica and 100 nm–5.0 m ytterbium trifluoride filler 20 nm silica, 4–11 nm zirconia, and 100 nm ytterbium trifluoride filler 100 nm ytterbium trifluoride, 20 nm silica, 4–11 nm zirconia, zirconia/silica cluster filler 600 nm silica/zirconia filler Ba B F Sr Al silicate glass filler 2–3 m Ba B Al Si glass filler
Filler (wt.%/vol.%) 64.5%/42.5%
76.5%/58.4%
76.5%/58.4%
–/66% 68%/45% 86%/70.1%
BisGMA: bisphenol A glycidyl methacrylate, bisEMA: ethoxylatedbis-phenol A dimethacrylate, bisEMA(6): (2,2-bis[4methacryloxypolyethoxyphenyl)propane], DMA: dimethacrylate, UDMA: urethane dimethacrylate, TEGDMA: triethylene glycol dimethacrylate, DDDMA: 1,12-dodecanediol dimethacrylate, proprietary AUDMA: high molecular weight aromatic dimethacrylate, proprietary AFM: addition-fragmentation monomers, +Procrylat (2,2-bis[4-(3 methacryloxypropoxy)phenyl]propane).
ring. For the degree of conversion measurements, the coverslip was replaced by the base plate of the attenuated total reflectance (ATR) attachment where the diamond prism was located at the center of the sample. For the hardness measurements, MylarTM sheets were inserted between the quartz disk and RBC sample and between the glass coverslip and sample. The RBC samples were photo-cured 20 s after starting the data collection using a single emission peak wavelength) LEDbased LCU (Paradigm Deep Cure LED unit, 3M) for 20, 5, 3, 1.5 and 1 s exposure times. This LCU delivered a radiant exitance of 1.1 W/cm2 , and the wavelength output was centered at 455 nm. The LCU radiant power was measured using a calibrated thermopile and meter (PM-10 detector and FieldMax meter, Coherent Inc., Santa Clara, CA, USA) [34,35].
2.3.
Axial shrinkage measurements
A variation of the bonded disk method was used to measure the axial shrinkage of RBC samples [36–41]. The technique is based on a Michelson interferometer where the sample replaces the moving mirror and the glass coverslip on top of the sample acts as a mirror to partially reflect the HeNe laser beam. As the sample shrinks with time, constructive and destructive interferences of the recombined laser beam at the output of the interferometer are monitored by a photodiode. A program was utilized to carry out data collection for between 15 h and 19 h at a sampling rate of 4000 samples/s during the first 30 min and then at 1 sample/s where each sample was the average of 720 data points collected over 0.1 s. Twenty seconds after the start of data collection, the program turned on and then off the LCU for the specified exposure time. Another program calculated the axial shrinkage with time from the photodiode signal. To minimize the effect of temperatureinduced drifts in the interferometric signal, the interferometer was kept inside an enclosure, and the interferometer temperature was kept constant using a temperature controller (Lake Shore Cryotonics Inc., Westervile, OH, USA). The sample temperature was measured to vary by less than 0.08 ◦ C, and the
drift in the interferometric signal was estimated to result in an error in deflection of less than 0.1 m over 19 h. Three repeats were performed for each experimental condition.
2.4.
Degree of conversion measurements
The degree of conversion was measured using a Fouriertransform infrared spectrometer (FTIR) (Tensor 27, Bruker, Billerica, MA, USA) equipped with a temperature-controlled ATR unit (Golden Gate, Specac, Orpington, Kent, UK) with a 2 × 2 mm diamond prism. Spectra were collected in dynamic mode for 10 min at a spectral resolution of 4 cm−1 and sample temperature of 26 ◦ C. The change in peak area of the aliphatic C C peak at 1635 cm−1 relative to the aromatic C C peak at 1607 cm−1 was monitored to determine the degree of conversion. A total of three repeats were collected for each experimental condition.
2.5.
Hardness measurements
Vickers hardness was measured (HM 123, Mitutoyo, Kawasaki, Kanagawa, Japan) on RBC samples one hour after light exposure; this time delay is sufficient for the hardness not to change during measurement and allowed the RBC to reach close to its maximum hardness value [22]. Three indentations were made 0.5 mm apart on both the top and bottom surface of each sample with a 300 g load applied for 8 s. Three samples were made for each photo-curing condition for a total of 9 data points to average per experimental condition.
2.6.
Statistical analysis
For each RBC, three repeats of photo-cured samples were used for statistical analysis. Two-way ANOVA followed by Tukey/Kramer post-hoc multiple comparison tests were used to determine if there were statistical differences in the data collected under different experimental conditions (˛ = 0.05). Linear least square fits (LLSF) where the errors in the y- and
1370
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Fig. 1 – Deflection as a function of time (a) and the logarithm of time (b) for five dental resin-based composites measured at T = 26 ◦ C with a sample thickness and diameter of 1.00 mm and 9–10 mm, respectively. The deflection for two RBCs was also collected at T = 34 ◦ C. The samples were photo-cured with a light curing unit with a radiant exitance of 1.1 W/cm2 and exposure time of 20 s. In the legend, BF stands for Bulk Fill.
x-coordinates were included as weights in the fits and calculation of the adjusted R-squared were performed on data exhibiting linear relationships between the variables.
3.
Results
Fig. 1 shows the deflection (D) for four Bulk Fill and one conventional RBCs as a function of time (a) and the logarithm of time (b) at a temperature of 26 ◦ C and sample thickness and diameter of 1.00 mm and 9–10 mm, respectively. In addition, the deflections are also illustrated for two RBCs at T = 34 ◦ C. The sharp variation in deflection observed in (b) at 40 s occurred when the LCU was turned off which resulted in a cooling down and thermal contraction of the samples. All the RBCs were measured for 15 h at which time the two flowable RBCs displayed the most substantial deflection compared to the other RBCs. Of note, the deflection for the two RBCs for which the
Fig. 2 – Deflection as a function of log(t) for three samples made with the SureFil SDR flow Posterior Bulk Fill A3 shade RBC, sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, and T = 26 ◦ C (a). Deflection rate as a function of time for the X-tra fil Universal shade RBC with a sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, and T = 26 ◦ C (b). In the legend, BF stands for Bulk Fill.
data were collected at 34 ◦ C was greater than those collected at 26 ◦ C. Fig. 1(b) depicts that, although the deflection with the logarithm of time is complex when the LCU is turned on and shortly after it is turned off, the deflection for all RBCs at both temperatures of 26 ◦ C and 34 ◦ C was found to vary linearly with log(t) for times longer than 500 s. Table 2 presents the mean deflection at 100 s over three repeats,, mean deflection at 15 h, , and the mean of the percent increase in deflection from 100 s to 15 h relative to that at 100 s, <D/D(100 s)>, for all the measured RBCs and at the two temperatures. Fig. 2(a)displays typical data for the deflection as a function of log(t) for the RBC SureFil SDR flow Posterior Bulk Fill A3 shade at a temperature of 26 ◦ C, and times greater than 6300 s. Linear least squares fits were performed on each data set with LLSF values for the slope, ˛S , equal to 2.008, 1.999, and 2.006 in units of m/log(t) for the three repeats.
0.1 0.3 0.5 0.1 0.09 0.05 0.07 0.4
0.88 0.88 1.02 0.92 1.00 0.76 0.88 0.39
0.05 0.04 0.05 0.01 0.02 0.03 0.04 0.02
1371
0.02 0.03 0.01 0.01 0.03 0.04 0.03 0.03 0.3 2 0.5 0.1 0.5 0.4 0.1 0.7 15.8 13 17.4 18.3 24.8 29.4 22.5 23.0 36.8 39.2 35.3 26.9 22.9 21.9 25.0 22.5 26 34 26 26 26 26 34 26 Filtek Bulk Fill Flowable Restorative A2 shade Filtek Bulk Fill Flowable Restorative A2 shade SureFil SDR Flow Posterior Bulk Fill A3 shade Z100 Restorative A2 shade Filtek One Bulk Fill Restorative A2 shade Filtek Bulk Fill Posterior Restorative A2 shade Filtek Bulk Fill Posterior Restorative A2 shade X-tra fil Universal shade
31.8 34.7 30.1 22.8 18.3 17.0 20.4 18.3
0.2 0.3 0.5 0.1 0.1 0.2 0.3 0.4
0.2 0.3 0.5 0.1 0.1 0.2 0.3 0.4
2.00 1.92 2.00 1.34 1.65 1.79 1.66 1.41
7.6 8.9 6.9 5.5 3.57 2.91 4.00 7.5
(s) SD (m/s) <(dD/dt)MAX > (m/s) SD (m/log(t)) <˛S > (m/log(t)) SD (%) <D/D(100 s)> (%) SD (m) (m) SD (m) (m) T (◦ C) RBC name
Table 2 – Mean ± standard deviation (SD) axial shrinkage kinetic parameters for six RBCs. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C and T = 34 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure time of 20 s.
SD (s)
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Fig. 3 – (a) Shrinkage related physical aging rate as a function of filler content. Fig. 3 (b) depicts the post-curing shrinkage as a function of the shrinkage rate. In (b), the anomalous point at (7.46, 23.04) has been excluded in the linear fit and calculation of R2 .
Fig. 2(b) depicts the deflection rate, dD/dt, as a function time for the X-tra fil Universal shade RBC at T = 26 ◦ C. The maximum deflection rate, (dD/dt) MAX , was 8.0 m/s at a time, tMAX , of 0.36 s after the turning on of the LCU. The oscillations in dD/dt at times after 23 s are due to the imperfect Hilbert transform performed on the raw oscillatory data from the interferometer photodiode output. The mean axial shrinkage kinetic parameters and mean ˛S for all RBCs investigated in this study are given in Table 2. Fig. 3(a) depicts the shrinkage related physical aging rate as a function of the filler content (vol.%) fitted using a LLSF resulting in an adjusted R2 of 0.853. Fig. 3(b) displays the post-curing shrinkage as a function of the maximum shrinkage rate with a LLSF and adjusted R2 of 0.929. Note that the data point located at (7.46, 23.04) was excluded from the fit as it is well outside the trend observed by the other data points. This anomalous data point may be attributed to the proprietary formulation of the X-tra fil Universal shade RBC. Fig. 4 depicts the deflection as a function of time (a) and the logarithm of time (b) for the RBC Filtek Bulk Fill Posterior
0.2 0.1 0.6 0.3 0.07 7.5 8.2 9.8 10.1 10.75 0.4 0.5 3 1 0.2 20.4 24.1 34 38 50.0 0.9 0.3 0.6 0.3 0.4 62.5 59.9 53.5 50.4 44.6 0.7 0.1 1 0.1 0.3 51.9 48.2 40 36.5 29.7 Two repeats for the DC data were used. a
1.82 2.0 2.19 2.58 2.73 20 5 3 1.5a 1a
0.06 0.1 0.03 0.06 0.03
<˛DC > (%/log(t)) SD (%) <DC/DC(25 s)> (%) SD (%) (%) SD (%) (%) SD (m/log(t)) <˛S > (m/log(t))
Restorative A2 shade at T = 26 ◦ C and photo-curing exposure times of 1, 1.5, 3, 5, and 20 s. The sharp variation in deflection observed in (b) at a time of 40 s and LCU exposure time of 20 s is due to the LCU turning off, cool down and thermal contraction of the sample. At a sufficiently long collection time where the deflection is proportional to log(t), ˛S increases with decreasing exposure time. The mean values of ˛S are given in Table 3. Fig. 5 shows the degree of conversion (DC) as a function of time (a) and the logarithm of time (b) for the RBC Filtek Bulk Fill Posterior Restorative A2 shade at T = 26 ◦ C and exposure times of 1, 1.5, 3, 5, and 20 s. The degree of conversion at 580 s increases as the LCU exposure time is increased. As shown in Fig. 5(b), the DC becomes linear as a function of log(t) shortly after the LCU is turned off up to the maximum data collection time of 10 min. Furthermore, the DC related physical aging rate, ˛DC , decreases with increasing exposure time. The mean values of ˛DC and DC at 25 s over three repeats,, mean DC at 580 s, , and the mean of the percent
Exposure time (s)
Fig. 4 – Deflection as of function of time (a) and the logarithm of time (b) for the Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure times of 1, 1.5, 3, 5, and 20 s. In the legend, BF stands for Bulk Fill.
SD (%/log(t))
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Table 3 – Mean ± standard deviation (SD) axial shrinkage and degree of conversion kinetic parameters for RBC samples made with Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and five exposure times.
1372
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
1373
Fig. 5 – Degree of conversion as a function of time (a) and the logarithm of time (b) for the Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure times of 1, 1.5, 3, 5, and 20 s. In the legend, BF stands for Bulk Fill.
increase in DC from 25 s to 580 s relative to that observed at 25 s, <DC/DC(25 s)> are reported in Table 3. The results shown in Table 3 together with data of the Vickers hardness collected 1 h after light exposure on RBC samples using the Filtek Bulk Fill Posterior Restorative A2 shade and prepared under the same experimental conditions are depicted in Figs. 6 and 7. Fig. 6(a) and (b) display the shrinkage and DC related physical aging rate as a function of the degree of conversion collected at t = 580 s. Linear relationships are observed with adjusted R2 values of 0.935 and 0.918, respectively. Fig. 6(c) depicts a plot of the DC related physical aging rate as a function of the shrinkage related physical aging rate with an adjusted R2 equal to 0.669. Fig. 7(a) displays the Vickers hardness measured on the top surface of the RBC sample made with the Filtek Bulk Fill Posterior Restorative A2 shade as a function of the degree of conversion collected at t = 580 s with an adjusted R2 of 0.950. Fig. 7(b) and (c) show the shrinkage related, and DC related
Fig. 6 – The shrinkage (a) and degree of conversion (b) related physical aging rate ˛S and ˛DC , respectively, as a function of the degree of conversion collected at t = 580 s for the Filtek Bulk Fill Posterior Restorative A2 shade. In Fig. 6 (c) ˛DC is displayed as a function of ˛S . In the legend, BF stands for Bulk Fill.
1374
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
physical aging rate as a function of Vickers hardness with adjusted R2 equal to 0.990 and 0.808, respectively.
4.
Fig. 7 – (a) Vickers hardness measured at 1 h after light exposure as a function of the degree of conversion collected at t = 580 s for the Filtek Bulk Fill Posterior Restorative A2 shade. Fig. 7 (b) and (c) depict ˛S and ˛DC as a function of the Vickers hardness measured at 1 h after light exposure, respectively. In the legend, BF stands for Bulk Fill.
Discussion
The objective of this study was to investigate the post-curing of six commercial contemporary RBCs at temperatures of 26 ◦ C (warm room) and 34 ◦ C (near tooth temperature). Five Bulk Fill and one conventional dental RBCs were selected for this work and all contain methacrylate-based monomers and have filler contents up to 70 vol.%. Three experimental techniques, namely, axial shrinkage and degree of conversion as a function of time for different LCU exposure times, and hardness measured 1 h after light exposure for different LCU exposure times were used to study the effects of post-curing. The axial shrinkage measurements were carried out for between 15 h and 19 h with an uncertainty in the observed deflection of less than 0.1 m over 19 h. Two low and four high filler content RBCs were selected for this investigation. According to the RBC manufacturer’s specifications listed in Table 1, the RBCs have different monomers with different reactivities and different filler volume fractions. SureFil SDR flow Posterior Bulk Fill A3 shade and X-tra fil Universal shade had different nominal shades than the remaining RBCs that were A2 shade used in this work. Although this difference in shade is small and that both of these Bulk Fill RBCs are highly transparent, the nominal shade difference (A2 vs A3 or Universal) should have minimal impact on the materials properties measured on 1 mm thick samples. For all six RBCs and temperatures of 26 ◦ C and 34 ◦ C, the first research hypothesis that the axial shrinkage is not proportional to log(t) at a sufficiently long time is rejected. The proportionality with the functional log(t) dependence is characteristic of the free volume fraction model developed to describe the change in volume with time of out-of-equilibrium polymer glass [27,28]. This model was studied extensively in various out-of-equilibrium polymer glasses and polymer/filler composites [29–31]. Evidence on the decrease in volume and hence densification of a photo-cured RBC sample at times of up to 2.8 days and its impact on the elastic modulus (E) was observed where E was found to be proportional to log(t) [21]. Moreover, the sample ion viscosity which is the sample electrical resistivity [42] was also shown to be proportional to log(t) at long times [21], although the interpretation of such data can be complex [43]. The deflection measured at T = 34 ◦ C for one low and one high filler content RBC is shown in Fig. 1. For each RBC, the deflection at T = 34 ◦ C is systematically larger than that collected at T = 26 ◦ C. The increase in deflection with temperature is attributed to the delayed gelation at the higher temperature. This results in prolonged monomer mobility and ultimately a higher degree of conversion [12]. This behavior is reflected, in part, for the high filler content RBC with a longer tMAX of 0.88 ± 0.04 s at T = 34 ◦ C compared to tMAX of 0.76 ± 0.03 s at T = 26 ◦ C (p < 0.05). However, tMAX is the same (p < 0.05) at both temperatures (tMAX = 0.88 ± 0.05 s) for the low filler content RBC. It is interesting to note that tMAX for the high filler content RBC at T = 34 ◦ C and having a lower viscosity than at T = 26 ◦ C is the same (p < 0.05) as that for the low filler content
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
RBC. A comparison of tMAX between the Filtek Bulk Fill Flowable Restorative A2 shade (tMAX = 0.88 ± 0.05 s) and SureFil SDR Flow Posterior Bulk Fill A3 shade (tMAX = 1.02 ± 0.05 s) indicate that the latter is significantly longer (p < 0.05) than the former. These two RBCs have very similar filler content. Other factors such as the reactivity of the monomer-photoinitiator system may also play important roles in determining tMAX . Although the ˛S listed in Table 2 differs from RBC to RBC, a strong correlation (R2 adj. = 0.853) is noted in Fig. 3(a) between ˛S and the filler volume content, namely, as the filler content increases ˛S decreases. This is expected because the filler content’s role is to minimize the overall shrinkage of the polymer matrix/filler content composite and hence of ˛S . An assessment of the post-curing shrinkage relative to the shrinkage at 100 s is given by <D/D(100 s)> for all the measured RBCs and at the two temperatures. The deflection data used in this evaluation were collected at t = 100 s where the contribution of the thermal contraction to the deflection caused by the LCU turning off at t = 40 s is minimized and at t = 15 h, which was the maximum data collection time for one RBC. Although <D/D(100 s)> was evaluated at a data collection time of only 15 h and not after the post-curing is complete which may be one to four weeks later [17,18], a lower limit to the post-curing shrinkage can be estimated. Table 2 indicates that the maximum value of <D/D(100 s)> is obtained for the Filtek Bulk Fill Posterior Restorative A2 shade with 29.4 ± 0.4% and 22.5 ± 0.1% at T = 26 ◦ C and T = 34 ◦ C, respectively. At T = 26 ◦ C, the post-curing shrinkage among all RBCs investigated varied between 15.8 and 29.4% with an average post-curing shrinkage of 21 ± 5%. It is important to note that <D/D(100 s)> is significantly smaller (p < 0.05) at near tooth temperature than at room temperature. Among the variables listed in Table 2 and excluding the data point collected with the X-tra fil Universal shade RBC, a correlation is shown in Fig. 3(b) between the post-cure shrinkage and the maximum shrinkage rate, namely, as (dD/dt)MAX increases <D/D(100 s)> decreases with a R2 adj. = 0.929 for a linear relationship. In other words, the faster the RBC goes from the liquid to the gel to the glass state, the smaller is the post-cure shrinkage. However, the total post-cure time when the RBC in the glass state stops shrinking and reaches thermal equilibrium is not included in the analysis. This total time may well be longer than for a smaller (dD/dt)MAX as more time may be required to reach thermal equilibrium for an RBC in a deep glass state. Note that the data point obtained using the X-tra fil Universal shade RBC is anomalous and may be related to the proprietary monomers used for this RBC. In Figs. 4 and 5 the degree of polymerization and hence of the glass state in the RBC Filtek Bulk Fill Posterior Restorative A2 shade at T = 26 ◦ C was varied by using five photo-curing exposure times. As depicted in Fig. 4(b), the second research hypothesis that ˛S does not vary with the light exposure time is rejected. Moreover, Fig. 5(b) indicates that the third and fourth research hypothesis, namely, that at a long time, the degree of conversion is not proportional to log(t) and that ˛DC does not vary with the light exposure time are both rejected. Considering that the samples used to carry out the DC measurements were made under the same experimental conditions as those used for the shrinkage measurements where the log(t) dependence was clearly observed, we attribute in
1375
part the log(t) dependence observed in the DC data to the densification of the out-of-equilibrium RBC polymer glass caused by the thermal diffusion of voids to the sample surface. The log(t) dependence for the degree of conversion has been previously reported [23], but the mechanism used to describe the time evolution is novel. A typical behavior observed in Figs. 4 and 5 is that the magnitude of the axial shrinkage and degree of conversion at the longest observation time increases at longer LCU exposure times while the opposite is true for ˛S and ˛DC . The ˛S listed in Table 2 differ from RBC to RBC due to the varying chemical nature, filler content, and reactivity of each RBC ((˛S )MAX /<˛S >) = 39%). Table 3 gives the values of ˛S ((˛S )MAX /<˛S >) = 40%) caused by different exposure time resulting in different glass state for a single RBC at T = 26 ◦ C. As shown in Table 3, the variation in ˛DC caused by the different exposure times is comparable ((˛DC )MAX /<˛DC >) = 36%). These results indicate that the longer the exposure time, the higher is the degree of conversion, the higher is the glass state of the photo-cured RBC, and hence the smaller are ˛S and ˛DC . This is shown in Fig. 6(a) and (b) where a high level of correlation is obtained between ˛S and ˛DC and the degree of conversion collected at t = 580 s. In turn, a positive correlation is also observed in Fig. 6(c) between ˛DC and ˛S . In a previous study [44], the void formation in photo-cured BisGMA/TEGDMA polymer glass network was investigated using positron annihilation lifetime spectroscopy. For the 40 mol% of TEGDMA in the co-monomer composition and for five exposure times, the average free volume of the voids in the polymer network decreased with increasing exposure time and with increasing DC. Although the relative void concentration with exposure time was not determined, the results suggest that the density of the polymer glass increases with increasing DC and with decreasing average free volume of the voids. In turn, the decreasing void free volume and increase in density of the polymer glass would lead to a smaller ˛S as illustrated in Fig. 4. Similarly, if the increase in DC in the post-cure region as depicted in Fig. 5 is mainly driven by the densification of the polymer network then ˛DC would decrease with increasing DC and density, and as the average free volume of the voids decrease. The degree of post-curing can also be evaluated by using <DC/DC(25 s)> with the DC data listed in Table 3 for the five exposure times. The DC related post-curing ranged from 20.4 ± 0.4% to as high as 50.0 ± 0.2% for exposure times from 20 s to 1 s, respectively. Using the recommended manufacturer’s exposure time of 20 s, the DC increases by as much as 20% relative to the DC at t = 25 s in the first 9 min after light exposure. The increase in DC due to post-curing is attributed in part to the post shrinkage of the out-of-equilibrium RBC that is in a glass state before it reaches thermal equilibrium. A measure of the crosslinking density within the polymer matrix is provided by the RBC hardness measured 1 h after light exposure. A high level of correlation is obtained between the hardness measured at the top surface of the sample and degree of conversion in Fig. 7(a). Moreover, the high correlation between ˛S and ˛DC and the hardness confirms that the shrinkage and DC related physical aging rate decreases where there is a higher degree of cross-linking and with the deeper
1376
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
glass state of the cured RBC. A similar analysis was performed using the hardness measured on the bottom surface of the RBC samples with the same conclusion.
5.
Clinical significance
Most previous studies have reported accurate shrinkage data at short times after light exposure. These values are misleadingly low as the shrinkage continues to increase with time. The results of this study indicate that the shrinkage occurring during post-curing can be quite appreciable; namely, 22.5% at 15 h relative to that at t = 100 s after light exposure at near tooth temperature for a contemporary Bulk Fill RBC. This post shrinkage results in a gradual built up of stress within the tooth wall structure, potentially shortening the restoration’s life. Different approaches have been taken by RBC manufacturers to minimize the built-in stress within the restored teeth. Recent shrinkage stress data [41] collected on five contemporary Bulk Fill RBCs at up to 12 h and T = 33 ◦ C showed that four RBCs had the same shrinkage stress values (p < 0.05) at times of at least 6 h. One of the four RBCs was the low filler content SureFil SDR flow Posterior Bulk Fill A3 shade. The shrinkage strain for this RBC is appreciably higher than those for all the high filler content RBCs used in this work providing substantial evidence of the effectiveness of the proprietary monomers used to release stress during and after photo-curing. Furthermore, the Filtek Bulk Fill Restorative A2 shade, which has an appreciable post-curing deflection at 15 h, has a low value of shrinkage stress at 12 h [41]. This illustrates the effectiveness of the novel monomers that are included in this RBC to relieve stress both during and after photo-curing.
6.
Conclusions
The effects of post-curing on the shrinkage of six commercial contemporary RBCs with filler contents up to 70 vol.% were evaluated. Axial shrinkage, the degree of conversion, and hardness at a temperature of 26 ◦ C and 34 ◦ C were measured. Within the limitations of this study, it can be concluded that:
1) Axial shrinkage in photo-cured RBCs at times of at least 15 h after light exposure displayed a functional dependence that was proportional to the logarithm of time. This time dependence is a signature of the free volume fraction model. 2) The shrinkage related physical aging rate at T = 26 ◦ C differed from RBC to RBC by up to 39%. A high correlation was obtained between shrinkage related physical aging rate and filler volume content. 3) In samples prepared using the same experimental conditions as those which displayed post-curing shrinkage proportional to log(t), the degree of conversion in these photo-cured RBCs samples was found to be proportional to the logarithm of time shortly after the photoexcitation light was turned off. This time dependence was attributed in part to the applicability of the free volume fraction model to the degree of conversion of the RBC samples.
4) The level of the out-of-equilibrium state in the RBCs in the glass state was varied by using different LCU exposure times. The shrinkage and degree of conversion related physical aging rate were found to be proportional to the degree of conversion collected at t = 580 s and Vickers hardness measured 1 h after light exposure. A direct correlation (R2 adj. = 0.669) was found between the shrinkage related physical aging rate and degree of conversion related physical aging rate. 5) The post-curing shrinkage in dental photo-cured RBCs can be appreciable (an additional 15.8–29.4% relative to the shrinkage at t = 100 s and T = 26 ◦ C). RBC blend and photo-curing conditions should be optimized to reduce the post-curing shrinkage.
Acknowledgments We would like to thank L. Kreplak and J. Stansbury for a fruitful discussion. We would like to express our sincere gratitude to C. Hudson 3M for the donation of the LCU, ceramic primer, and RBCs and, Voco GmbH, and DENTSPLY SIRONA for the donation of the RBCs used in this study. We thank G. Earle IObIO Inc. for the donation of the LabVIEW program used for the data collection. This work was supported in part by the Department of Physics and Atmospheric Science, and Department of Dental Clinical Sciences, Dalhousie University.
references
[1] Leprince JG, Palin WM, Hadis MA, Devaux J, Leloup G. Progress in dimethacrylate-based dental composite technology and curing efficiency. Dent Mater 2013;29:139–56. [2] Ferracane JL. Resin composite—state of the art. Dent Mater 2011;27:29–38. [3] Cramer NB, Stansbury JW, Bowman CN. Recent advances and developments in composite dental restorative materials. J Dent Res 2011;90:402–16. [4] Ferracane JL. Buonocore lecture placing dental composites—a stressful experience. Oper Dent 2008;33:247–57. [5] Oliveira LRS, Braga SSL, Bicalho AA, Ribeiro MTH, Price RB, Soares CJ. Molar cusp deformation evaluated by micro-CT and enamel crack formation to compare incremental and bulk-filling techniques. J Dent 2018;74:71–8. [6] Moorthy A, Hogg CH, Dowling AH, Grufferty BF, Benetti AR, Fleming GJ. Cuspal deflection and microleakage in premolar teeth restored with bulk-fill flowable resin-based composite base materials. J Dent 2012;40:500–5. [7] Alvanforoush N, Palamara J, Wong RH, Burrow MF. Comparison between published clinical success of direct resin composite restorations in vital posterior teeth in 1995–2005 and 2006–2016 periods. Aust Dent J 2017;62:132–45. [8] van Dijken JWV, Pallesen U. Bulk-filled posterior resin restorations based on stress-decreasing resin technology: a randomized, controlled 6-year evaluation. Eur J Oral Sci 2017;125:303–9. [9] Ilie N, Hickel R. Investigations on a methacrylate-based flowable composite based on the SDR technology. Dent Mater 2011;27:348–55. [10] Shah PK, Stansbury JW, Bowman CN. Application of an addition-fragmentation-chain transfer monomer in
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25] [26] [27]
di(meth)acrylate network formation to reduce polymerization shrinkage stress. Polym Chem 2017;8:4339–51. Blackham JT, Vandewalle KS, Lien W. Properties of hybrid resin composite systems containing prepolymerized filler particles. Oper Dent 2009;34:697–702. Stansbury JW. Dimethacrylate network formation and polymer property evolution as determined by the selection of monomers and curing conditions. Dent Mater 2012;28:13–22. Al-Ahdal K, Ilie N, Silikas N, Watts DC. Polymerization kinetics and impact of post polymerization on the Degree of Conversion of bulk-fill resin-composite at clinically relevant depth. Dent Mater 2015;31:1207–13. Par M, Gamulin O, Marovic D, Klaric E, Tarle Z. Raman spectroscopic assessment of degree of conversion of bulk-fill resin composites-changes at 24 hours post cure. Oper Dent 2015;40:E92–101. Par M, Gamulin O, Marovic D, Klaric E, Tarle Z. Effect of temperature on post-cure polymerization of bulk-fill composites. J Dent 2014;42:1255–60. Scotti N, Venturello A, Borga FA, Pasqualini D, Paolino DS, Geobaldo F, et al. Post-curing conversion kinetics as functions of the irradiation time and increment thickness. J Appl Oral Sci 2013;21:190–5. Schneider LF, Consani S, Ogliari F, Correr AB, Sobrinho LC, Sinhoreti MA. Effect of time and polymerization cycle on the degree of conversion of a resin composite. Oper Dent 2006;31:489–95. Sideridou ID, Vouvoudi EC, Adamidou EA. Dynamic mechanical thermal properties of the dental light-cured nanohybrid composite Kalore, GC: effect of various food/oral simulating liquids. Dent Mater 2015;31:154–61. Alnazzawi A, Watts DC. Simultaneous determination of polymerization shrinkage, exotherm and thermal expansion coefficient for dental resin-composites. Dent Mater 2012;28:1240–9. Truffier-Boutry D, Demoustier-Champagne S, Devaux J, Biebuyck JJ, Mestdagh M, Larbanois P, et al. A physico-chemical explanation of the post-polymerization shrinkage in dental resins. Dent Mater 2006;22:405–12. Steinhaus J, Frentzen M, Rosentritt M, Moginger B. Dielectric analysis of short-term and long-term curing of novel photo-curing dental filling materials. Mod Trends Polym Sci—Epf 09 2010;296:622–5. Leung RL, Fan PL, Johnston WM. Post-irradiation polymerization of visible light-activated composite resin. J Dent Res 1983;62:363–5. Selig D, Haenel T, Hausnerova B, Moeginger B, Labrie D, Sullivan B, et al. Examining exposure reciprocity in a resin based composite using high irradiance levels and real-time degree of conversion values. Dent Mater 2015;31:583–93. Angell CA. Formation of glasses from liquids and biopolymers. Science 1995;267:1924–35. Hutchinson JM. Physical aging of polymers. Prog Polym Sci 1995;20:703–60. Tant MR, Wilkes GL. An overview of the nonequilibrium behavior of polymer glasses. Polym Eng Sci 1981;21:874–95. Curro JG, Lagasse RR, Simha R. Diffusion model for volume recovery in glasses. Macromolecules 1982;15:1621–6.
1377
[28] Jin KL, Li LQ, Torkelson JM. Bulk physical aging behavior of cross-linked polystyrene compared to its linear precursor: effects of cross-linking and aging temperature. Polymer 2017;115:197–203. [29] Cangialosi D, Alegria A, Colmenero J. Effect of nanostructure on the thermal glass transition and physical aging in polymer materials. Prog Polym Sci 2016;54–55:128–47. [30] Cangialosi D, Boucher VM, Alegria A, Colmenero J. Volume recovery of polystyrene/silica nanocomposites. J Polym Sci B—Polym Phys 2013;51:847–53. [31] Cangialosi D, Boucher VM, Alegria A, Colmenero J. Free volume holes diffusion to describe physical aging in poly(mehtyl methacrylate)/silica nanocomposites. J Chem Phys 2011;135:014901. [32] Rueggeberg FA, Daronch M, Browning WD, DE Goes MF. In vivo temperature measurement: tooth preparation and restoration with preheated resin composite. J Esthet Restor Dent 2010;22:314–22. [33] Runnacles P, Arrais CA, Pochapski MT, dos Santos FA, Coelho U, Gomes JC, et al. Direct measurement of time-dependent anesthetized in vivo human pulp temperature. Dent Mater 2015;31:53–9. [34] Price RB, Rueggeberg FA, Labrie D, Felix CM. Irradiance uniformity and distribution from dental light curing units. J Esthet Restor Dent 2010;22:86–101. [35] Price RB, Labrie D, Rueggeberg FA, Felix CM. Irradiance differences in the violet (405 nm) and blue (460 nm) spectral ranges among dental light-curing units. J Esthet Restor Dent 2010;22:363–77. [36] Watts DC, Cash AJ. Determination of polymerization shrinkage kinetics in visible-light-cured materials: methods development. Dent Mater 1991;7:281–7. [37] Watts DC, Marouf AS. Optimal specimen geometry in bonded-disk shrinkage-strain measurements on light-cured biomaterials. Dent Mater 2000;16:447–51. [38] Lee IB, Cho BH, Son HH, Um CM, Lim BS. The effect of consistency, specimen geometry and adhesion on the axial polymerization shrinkage measurement of light cured composites. Dent Mater 2006;22:1071–9. [39] Kostylev I, Labrie D, Price RBT. Time-resolved 2D shrinkage field of dental resins using laser interferometry. Appl Opt 2015;54:1852–60. [40] Kostylev I, Winters T, Labrie D, Price RB. Role of the beam profile and sample geometry in the bonded-disk method on the shrinkage kinetics of two dental resins. Dent Mater 2016;32:1026–35. [41] Kalliecharan D, Germscheid W, Price RB, Stansbury J, Labrie D. Shrinkage stress kinetics of Bulk Fill resin-based composites at tooth temperature and long time. Dent Mater 2016;32:1322–31. [42] Zahouily K, Decker C, Kaisersberger E, Gruener M. Real-time UV cure monitoring. Eur Coat J 2003;49:14–8. [43] Steinhaus J, Hausnerova B, Haenel T, Grossgarten M, Moginger B. Curing kinetics of visible light curing dental resin composites investigated by dielectric analysis (DEA). Dent Mater 2014;30:372–80. [44] Pfeifer CS, Shelton ZR, Braga RR, Windmoller D, Machado JC, Stansbury JW. Characterization of dimethacrylate polymeric networks: a study of the crosslinked structure formed by monomers used in dental composites. Eur Polym J 2011;47:162–70.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Post-curing in dental resin-based composites William Germscheid a , Louis Gosse de Gorre a , Braden Sullivan b , Catherine O’Neill b , Richard B. Price b , Daniel Labrie a,∗ a b
Department of Physics and Atmospheric Science, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada Department of Dental Clinical Sciences, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To determine the post-curing in six commercial contemporary resin-based com-
Received 3 April 2018
posites (RBCs) using axial shrinkage, the degree of conversion, and Vickers hardness.
Received in revised form 6 June 2018
Methods. Five Bulk Fill and one conventional RBCs from three companies were selected with
Accepted 7 June 2018
a wide range of filler volume content. The axial shrinkage of samples that were 1.00 mm thick by 9–10 mm diameter was measured using a modified bonded disk method over a time between 15 h and 19 h at temperatures of 26 ◦ C and 34 ◦ C (mouth temperature). The degree
Keywords:
of conversion (DC) was collected continuously for 10 min using mid-infrared spectroscopy
Resin composite
in the attenuated total reflectance geometry. Vickers hardness was measured at 1 h post-
Polymerization
irradiation using a load of 300 gf. For all three tests, the samples were irradiated at five
Shrinkage
exposure times, 20, 5, 3, 1.5 and 1 s with a light curing unit radiant exitance of 1.1 W/cm2 .
Degree of conversion
Three samples (n = 3) were used for each experimental condition.
Microhardness
Results. After light exposure, the axial shrinkage and degree of conversion exhibited a func-
Kinetics
tional time dependence that was proportional to the logarithm of time. This suggests an
Temperature
out-of-equilibrium polymer composite glass that is transitioning to thermal equilibrium. At a sufficiently long time and among the RBCs investigated, the shrinkage related physical aging rate was found to vary between 1.34 and 2.00 m/log(t). The rate was a function of the filler content. Furthermore, 15 h after light exposure, the post-curing shrinkage was estimated to be an additional 22.5% relative to the shrinkage at 100 s for one RBC at T = 34 ◦ C. The hardness in the photo-cured RBC was varied by using different light exposure times. The first two experimental techniques show that the higher the initial DC 10 min after light exposure, the smaller is the post-curing shrinkage related and DC related physical aging rates. A direct correlation was observed between the shrinkage related and the DC related physical aging rates. Significance. Post-curing shrinkage should be evaluated for longer than 1 h. The post-curing shrinkage 15 h after light exposure in dental RBCs can be appreciable. The long-term development of built-in stress within the tooth wall structure may shorten the restoration’s lifespan. © 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.
∗
Corresponding author. E-mail address: [email protected] (D. Labrie). https://doi.org/10.1016/j.dental.2018.06.021 0109-5641/© 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.
1368
1.
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Introduction
A paradigm shift in the field of dentistry occurred with the introduction of photo-cured resin-based composites (RBCs) as an alternative to amalgam in dental restorations [1–3]. The main driving force is that the restored teeth can look like a natural decay free tooth. Due to the fundamental nature of the resin used in RBCs, after photo-curing, polymerizationinduced shrinkage strain develops with time. This produces shrinkage-induced stress within the cavity walls of the teeth [4]. There is evidence that such stresses may result in enamel cracking and tooth post-operative pain [5], margin debonding [6], secondary caries [7], and premature failure of the restoration [8]. Dental manufacturers have taken different approaches to relieve the stress developed both during and after photo-curing by altering the monomers [9,10], or by incorporating pre-polymerized filler particles [11] in RBCs. After the light curing unit (LCU) is turned off, one important contribution to shrinkage has been attributed to the continued cross-linking as the unterminated radicals continue to react, albeit rather slowly. This process is sometimes referred as the densification of the cured RBC with time [12] because it leads to an increased RBC density and elastic modulus. Several studies have been carried out in cured RBCs to investigate how the degree of conversion [13–17], glass transition temperature [18], axial shrinkage [19], density and mechanical properties [20–22] vary with time for up to a month after light exposure. It has been reported that during the postcuring period, the elastic modulus [21], ion viscosity [21], and degree of conversion (DC) [23] were found to be proportional to the logarithm of time, but no explanation was given for this unique time dependence. Post-curing was investigated in experimental photo-cured resins using six experimental techniques [20]. An overall explanation of the results was offered using the free volume fraction model to explain the properties of out-of-equilibrium amorphous polymer glasses [24–27]. In this model, an out-of-equilibrium amorphous polymer glass can have an excess of free volume in the form of a large number of sub-nanometer-sized voids dispersed throughout the polymer glass. Due to the thermal diffusion of these excess voids out to the surface, the polymer continues to shrink with time. As indicated in Fig. 2 of Ref. [27], after a relatively short time, the decrease in sample thickness can be predicted to be proportional to the logarithm of time [log(t)]. The observation of the log(t) dependence for the volume shrinkage of an out-of-equilibrium RBC in the glass state provides a direct evidence for this model. In the above study [20], the sample axial shrinkage was measured continuously over a period of 10 h. The axial shrinkage varied in a non-monotonic manner with time and, as a result, was inconclusive toward testing the applicability of the free volume fraction model to photocured methacrylate-based resins. In another study [19], the axial shrinkage on RBCs was measured continuously over a period of 1 h; this time period was too short to observe the log(t) dependence. The free volume fraction model has been studied extensively in the field of polymer glasses and composites [27–31]. The process for the out-of-equilibrium polymer glass to reach thermal equilibrium is commonly referred to as physical
aging. In this work, the constant of proportionality between the glass parameter such as axial shrinkage (or DC) and the logarithm of time is called the axial shrinkage (or DC) related physical aging rate [28]. To optimize dental RBC blends and photo-curing conditions with the goal of minimizing post-curing shrinkage, it is vital to understand the underlying mechanisms governing post-curing. Furthermore, it is also important to quantify the post-curing shrinkage in RBCs at times longer than 1 h. In this study, the post-curing of six commercial contemporary RBCs was investigated using axial shrinkage and degree of conversion as a function of time for different light exposure times, and hardness measured one hour after light exposure for different light exposure times. The research hypotheses were: (1) at a sufficiently long time, the axial shrinkage is not proportional to the logarithm of time, (2) the axial shrinkage related physical aging rate, ˛S , does not vary with the light exposure time, (3) at a sufficiently long time, the degree of conversion is not proportional to the logarithm of time, (4) the degree of conversion related physical aging rate, ˛DC , does not vary with the light exposure time, (5) ˛S and ˛DC do not vary with the degree of conversion, (6) ˛DC is not correlated with ˛S , and (7) ˛S and ˛DC are not correlated with the hardness.
2.
Materials and methods
2.1.
Materials
Six commercial contemporary RBCs were used in this study. They consisted of five Bulk Fill and one conventional dental RBCs. The RBC compositions as described by the manufacturers are reported in Table 1. Two low (≤45 vol.%) filler content (3M Filtek Bulk Fill Flowable Restorative A2 shade and DENTSPLY Surefil SDR flow Posterior Bulk Fill A3 shade), and four high (>58 vol.%) filler content (3M Filtek Bulk Fill Posterior Restorative A2 shade, Filtek One Bulk Fill Restorative A2 shade, and Z100 Restorative A2 shade, and Voco GmbH X-tra fil Universal shade) were used to study the effect of time and filler volume fraction on the shrinkage and degree of conversion. Temperatures of 26 ◦ C (representing a warm room and the minimum after washing and drying a cavity) and 34 ◦ C (representing a maximum intraoral temperature inside a cavity) [32,33].
2.2.
Sample preparation and photo-curing conditions
For all three experimental techniques used, the sample geometry and sample dimensions, and photo-curing conditions were the same. The substrate was a 25.4 mm diameter by 3 mm thick quartz disk. One surface was lightly sandblasted and then silanized using a ceramic primer (RelyX Ceramic Primer, 3M, St Paul, MN, USA). A 1.00 mm thick by 9–10 mm diameter RBC sample was inserted at the center of the substrate. For the axial shrinkage measurements, a 100 m thick glass coverslip disk was placed on top of the sample. The coverslip was supported along its edge by a 1.00 mm thick brass
1369
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Table 1 – Composition of the resin-based composites as provided by the manufacturers. Product name
Manufacturer (lot number)
Filtek Bulk Fill Flowable Restorative A2 shade
3M ESPE (N632976)
BisGMA, UDMA, bisEMA(6) and Procrylat monomers
Filtek bulk fill posterior restorative A2 shade
3M ESPE (N894570, N793515, N723218, N772739, N761323, N833958) 3M ESPE (N889317)
Proprietary AUDMA and AFM, DDDMA and UDMA
3M ESPE (N232301) DENTSPLY (120127)
BisGMA, TEGDMA SDR patented UDMA, DMA, di-functional diluent BisGMA, UDMA, TEGDMA
Filtek One Bulk Fill Restorative A2 shade
Z100 Restorative A2 shade SureFil SDR flow posterior bulk fill A3 shade X-tra fil universal shade
Voco GmbH (1525575)
Resin matrix
Proprietary AUDMA and AFM, DDDMA and UDMA
Filler 10 nm–3.5 m zirconia/silica and 100 nm–5.0 m ytterbium trifluoride filler 20 nm silica, 4–11 nm zirconia, and 100 nm ytterbium trifluoride filler 100 nm ytterbium trifluoride, 20 nm silica, 4–11 nm zirconia, zirconia/silica cluster filler 600 nm silica/zirconia filler Ba B F Sr Al silicate glass filler 2–3 m Ba B Al Si glass filler
Filler (wt.%/vol.%) 64.5%/42.5%
76.5%/58.4%
76.5%/58.4%
–/66% 68%/45% 86%/70.1%
BisGMA: bisphenol A glycidyl methacrylate, bisEMA: ethoxylatedbis-phenol A dimethacrylate, bisEMA(6): (2,2-bis[4methacryloxypolyethoxyphenyl)propane], DMA: dimethacrylate, UDMA: urethane dimethacrylate, TEGDMA: triethylene glycol dimethacrylate, DDDMA: 1,12-dodecanediol dimethacrylate, proprietary AUDMA: high molecular weight aromatic dimethacrylate, proprietary AFM: addition-fragmentation monomers, +Procrylat (2,2-bis[4-(3 methacryloxypropoxy)phenyl]propane).
ring. For the degree of conversion measurements, the coverslip was replaced by the base plate of the attenuated total reflectance (ATR) attachment where the diamond prism was located at the center of the sample. For the hardness measurements, MylarTM sheets were inserted between the quartz disk and RBC sample and between the glass coverslip and sample. The RBC samples were photo-cured 20 s after starting the data collection using a single emission peak wavelength) LEDbased LCU (Paradigm Deep Cure LED unit, 3M) for 20, 5, 3, 1.5 and 1 s exposure times. This LCU delivered a radiant exitance of 1.1 W/cm2 , and the wavelength output was centered at 455 nm. The LCU radiant power was measured using a calibrated thermopile and meter (PM-10 detector and FieldMax meter, Coherent Inc., Santa Clara, CA, USA) [34,35].
2.3.
Axial shrinkage measurements
A variation of the bonded disk method was used to measure the axial shrinkage of RBC samples [36–41]. The technique is based on a Michelson interferometer where the sample replaces the moving mirror and the glass coverslip on top of the sample acts as a mirror to partially reflect the HeNe laser beam. As the sample shrinks with time, constructive and destructive interferences of the recombined laser beam at the output of the interferometer are monitored by a photodiode. A program was utilized to carry out data collection for between 15 h and 19 h at a sampling rate of 4000 samples/s during the first 30 min and then at 1 sample/s where each sample was the average of 720 data points collected over 0.1 s. Twenty seconds after the start of data collection, the program turned on and then off the LCU for the specified exposure time. Another program calculated the axial shrinkage with time from the photodiode signal. To minimize the effect of temperatureinduced drifts in the interferometric signal, the interferometer was kept inside an enclosure, and the interferometer temperature was kept constant using a temperature controller (Lake Shore Cryotonics Inc., Westervile, OH, USA). The sample temperature was measured to vary by less than 0.08 ◦ C, and the
drift in the interferometric signal was estimated to result in an error in deflection of less than 0.1 m over 19 h. Three repeats were performed for each experimental condition.
2.4.
Degree of conversion measurements
The degree of conversion was measured using a Fouriertransform infrared spectrometer (FTIR) (Tensor 27, Bruker, Billerica, MA, USA) equipped with a temperature-controlled ATR unit (Golden Gate, Specac, Orpington, Kent, UK) with a 2 × 2 mm diamond prism. Spectra were collected in dynamic mode for 10 min at a spectral resolution of 4 cm−1 and sample temperature of 26 ◦ C. The change in peak area of the aliphatic C C peak at 1635 cm−1 relative to the aromatic C C peak at 1607 cm−1 was monitored to determine the degree of conversion. A total of three repeats were collected for each experimental condition.
2.5.
Hardness measurements
Vickers hardness was measured (HM 123, Mitutoyo, Kawasaki, Kanagawa, Japan) on RBC samples one hour after light exposure; this time delay is sufficient for the hardness not to change during measurement and allowed the RBC to reach close to its maximum hardness value [22]. Three indentations were made 0.5 mm apart on both the top and bottom surface of each sample with a 300 g load applied for 8 s. Three samples were made for each photo-curing condition for a total of 9 data points to average per experimental condition.
2.6.
Statistical analysis
For each RBC, three repeats of photo-cured samples were used for statistical analysis. Two-way ANOVA followed by Tukey/Kramer post-hoc multiple comparison tests were used to determine if there were statistical differences in the data collected under different experimental conditions (˛ = 0.05). Linear least square fits (LLSF) where the errors in the y- and
1370
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Fig. 1 – Deflection as a function of time (a) and the logarithm of time (b) for five dental resin-based composites measured at T = 26 ◦ C with a sample thickness and diameter of 1.00 mm and 9–10 mm, respectively. The deflection for two RBCs was also collected at T = 34 ◦ C. The samples were photo-cured with a light curing unit with a radiant exitance of 1.1 W/cm2 and exposure time of 20 s. In the legend, BF stands for Bulk Fill.
x-coordinates were included as weights in the fits and calculation of the adjusted R-squared were performed on data exhibiting linear relationships between the variables.
3.
Results
Fig. 1 shows the deflection (D) for four Bulk Fill and one conventional RBCs as a function of time (a) and the logarithm of time (b) at a temperature of 26 ◦ C and sample thickness and diameter of 1.00 mm and 9–10 mm, respectively. In addition, the deflections are also illustrated for two RBCs at T = 34 ◦ C. The sharp variation in deflection observed in (b) at 40 s occurred when the LCU was turned off which resulted in a cooling down and thermal contraction of the samples. All the RBCs were measured for 15 h at which time the two flowable RBCs displayed the most substantial deflection compared to the other RBCs. Of note, the deflection for the two RBCs for which the
Fig. 2 – Deflection as a function of log(t) for three samples made with the SureFil SDR flow Posterior Bulk Fill A3 shade RBC, sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, and T = 26 ◦ C (a). Deflection rate as a function of time for the X-tra fil Universal shade RBC with a sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, and T = 26 ◦ C (b). In the legend, BF stands for Bulk Fill.
data were collected at 34 ◦ C was greater than those collected at 26 ◦ C. Fig. 1(b) depicts that, although the deflection with the logarithm of time is complex when the LCU is turned on and shortly after it is turned off, the deflection for all RBCs at both temperatures of 26 ◦ C and 34 ◦ C was found to vary linearly with log(t) for times longer than 500 s. Table 2 presents the mean deflection at 100 s over three repeats,
0.1 0.3 0.5 0.1 0.09 0.05 0.07 0.4
0.88 0.88 1.02 0.92 1.00 0.76 0.88 0.39
0.05 0.04 0.05 0.01 0.02 0.03 0.04 0.02
1371
0.02 0.03 0.01 0.01 0.03 0.04 0.03 0.03 0.3 2 0.5 0.1 0.5 0.4 0.1 0.7 15.8 13 17.4 18.3 24.8 29.4 22.5 23.0 36.8 39.2 35.3 26.9 22.9 21.9 25.0 22.5 26 34 26 26 26 26 34 26 Filtek Bulk Fill Flowable Restorative A2 shade Filtek Bulk Fill Flowable Restorative A2 shade SureFil SDR Flow Posterior Bulk Fill A3 shade Z100 Restorative A2 shade Filtek One Bulk Fill Restorative A2 shade Filtek Bulk Fill Posterior Restorative A2 shade Filtek Bulk Fill Posterior Restorative A2 shade X-tra fil Universal shade
31.8 34.7 30.1 22.8 18.3 17.0 20.4 18.3
0.2 0.3 0.5 0.1 0.1 0.2 0.3 0.4
0.2 0.3 0.5 0.1 0.1 0.2 0.3 0.4
2.00 1.92 2.00 1.34 1.65 1.79 1.66 1.41
7.6 8.9 6.9 5.5 3.57 2.91 4.00 7.5
Table 2 – Mean ± standard deviation (SD) axial shrinkage kinetic parameters for six RBCs. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C and T = 34 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure time of 20 s.
SD (s)
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Fig. 3 – (a) Shrinkage related physical aging rate as a function of filler content. Fig. 3 (b) depicts the post-curing shrinkage as a function of the shrinkage rate. In (b), the anomalous point at (7.46, 23.04) has been excluded in the linear fit and calculation of R2 .
Fig. 2(b) depicts the deflection rate, dD/dt, as a function time for the X-tra fil Universal shade RBC at T = 26 ◦ C. The maximum deflection rate, (dD/dt) MAX , was 8.0 m/s at a time, tMAX , of 0.36 s after the turning on of the LCU. The oscillations in dD/dt at times after 23 s are due to the imperfect Hilbert transform performed on the raw oscillatory data from the interferometer photodiode output. The mean axial shrinkage kinetic parameters and mean ˛S for all RBCs investigated in this study are given in Table 2. Fig. 3(a) depicts the shrinkage related physical aging rate as a function of the filler content (vol.%) fitted using a LLSF resulting in an adjusted R2 of 0.853. Fig. 3(b) displays the post-curing shrinkage as a function of the maximum shrinkage rate with a LLSF and adjusted R2 of 0.929. Note that the data point located at (7.46, 23.04) was excluded from the fit as it is well outside the trend observed by the other data points. This anomalous data point may be attributed to the proprietary formulation of the X-tra fil Universal shade RBC. Fig. 4 depicts the deflection as a function of time (a) and the logarithm of time (b) for the RBC Filtek Bulk Fill Posterior
0.2 0.1 0.6 0.3 0.07 7.5 8.2 9.8 10.1 10.75 0.4 0.5 3 1 0.2 20.4 24.1 34 38 50.0 0.9 0.3 0.6 0.3 0.4 62.5 59.9 53.5 50.4 44.6 0.7 0.1 1 0.1 0.3 51.9 48.2 40 36.5 29.7 Two repeats for the DC data were used. a
1.82 2.0 2.19 2.58 2.73 20 5 3 1.5a 1a
0.06 0.1 0.03 0.06 0.03
<˛DC > (%/log(t)) SD (%) <DC/DC(25 s)> (%) SD (%)
Restorative A2 shade at T = 26 ◦ C and photo-curing exposure times of 1, 1.5, 3, 5, and 20 s. The sharp variation in deflection observed in (b) at a time of 40 s and LCU exposure time of 20 s is due to the LCU turning off, cool down and thermal contraction of the sample. At a sufficiently long collection time where the deflection is proportional to log(t), ˛S increases with decreasing exposure time. The mean values of ˛S are given in Table 3. Fig. 5 shows the degree of conversion (DC) as a function of time (a) and the logarithm of time (b) for the RBC Filtek Bulk Fill Posterior Restorative A2 shade at T = 26 ◦ C and exposure times of 1, 1.5, 3, 5, and 20 s. The degree of conversion at 580 s increases as the LCU exposure time is increased. As shown in Fig. 5(b), the DC becomes linear as a function of log(t) shortly after the LCU is turned off up to the maximum data collection time of 10 min. Furthermore, the DC related physical aging rate, ˛DC , decreases with increasing exposure time. The mean values of ˛DC and DC at 25 s over three repeats,
Exposure time (s)
Fig. 4 – Deflection as of function of time (a) and the logarithm of time (b) for the Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure times of 1, 1.5, 3, 5, and 20 s. In the legend, BF stands for Bulk Fill.
SD (%/log(t))
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
Table 3 – Mean ± standard deviation (SD) axial shrinkage and degree of conversion kinetic parameters for RBC samples made with Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and five exposure times.
1372
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
1373
Fig. 5 – Degree of conversion as a function of time (a) and the logarithm of time (b) for the Filtek Bulk Fill Posterior Restorative A2 shade. Sample thickness and diameter of 1.00 mm and 9–10 mm, respectively, T = 26 ◦ C, LCU radiant exitance of 1.1 W/cm2 and exposure times of 1, 1.5, 3, 5, and 20 s. In the legend, BF stands for Bulk Fill.
increase in DC from 25 s to 580 s relative to that observed at 25 s, <DC/DC(25 s)> are reported in Table 3. The results shown in Table 3 together with data of the Vickers hardness collected 1 h after light exposure on RBC samples using the Filtek Bulk Fill Posterior Restorative A2 shade and prepared under the same experimental conditions are depicted in Figs. 6 and 7. Fig. 6(a) and (b) display the shrinkage and DC related physical aging rate as a function of the degree of conversion collected at t = 580 s. Linear relationships are observed with adjusted R2 values of 0.935 and 0.918, respectively. Fig. 6(c) depicts a plot of the DC related physical aging rate as a function of the shrinkage related physical aging rate with an adjusted R2 equal to 0.669. Fig. 7(a) displays the Vickers hardness measured on the top surface of the RBC sample made with the Filtek Bulk Fill Posterior Restorative A2 shade as a function of the degree of conversion collected at t = 580 s with an adjusted R2 of 0.950. Fig. 7(b) and (c) show the shrinkage related, and DC related
Fig. 6 – The shrinkage (a) and degree of conversion (b) related physical aging rate ˛S and ˛DC , respectively, as a function of the degree of conversion collected at t = 580 s for the Filtek Bulk Fill Posterior Restorative A2 shade. In Fig. 6 (c) ˛DC is displayed as a function of ˛S . In the legend, BF stands for Bulk Fill.
1374
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
physical aging rate as a function of Vickers hardness with adjusted R2 equal to 0.990 and 0.808, respectively.
4.
Fig. 7 – (a) Vickers hardness measured at 1 h after light exposure as a function of the degree of conversion collected at t = 580 s for the Filtek Bulk Fill Posterior Restorative A2 shade. Fig. 7 (b) and (c) depict ˛S and ˛DC as a function of the Vickers hardness measured at 1 h after light exposure, respectively. In the legend, BF stands for Bulk Fill.
Discussion
The objective of this study was to investigate the post-curing of six commercial contemporary RBCs at temperatures of 26 ◦ C (warm room) and 34 ◦ C (near tooth temperature). Five Bulk Fill and one conventional dental RBCs were selected for this work and all contain methacrylate-based monomers and have filler contents up to 70 vol.%. Three experimental techniques, namely, axial shrinkage and degree of conversion as a function of time for different LCU exposure times, and hardness measured 1 h after light exposure for different LCU exposure times were used to study the effects of post-curing. The axial shrinkage measurements were carried out for between 15 h and 19 h with an uncertainty in the observed deflection of less than 0.1 m over 19 h. Two low and four high filler content RBCs were selected for this investigation. According to the RBC manufacturer’s specifications listed in Table 1, the RBCs have different monomers with different reactivities and different filler volume fractions. SureFil SDR flow Posterior Bulk Fill A3 shade and X-tra fil Universal shade had different nominal shades than the remaining RBCs that were A2 shade used in this work. Although this difference in shade is small and that both of these Bulk Fill RBCs are highly transparent, the nominal shade difference (A2 vs A3 or Universal) should have minimal impact on the materials properties measured on 1 mm thick samples. For all six RBCs and temperatures of 26 ◦ C and 34 ◦ C, the first research hypothesis that the axial shrinkage is not proportional to log(t) at a sufficiently long time is rejected. The proportionality with the functional log(t) dependence is characteristic of the free volume fraction model developed to describe the change in volume with time of out-of-equilibrium polymer glass [27,28]. This model was studied extensively in various out-of-equilibrium polymer glasses and polymer/filler composites [29–31]. Evidence on the decrease in volume and hence densification of a photo-cured RBC sample at times of up to 2.8 days and its impact on the elastic modulus (E) was observed where E was found to be proportional to log(t) [21]. Moreover, the sample ion viscosity which is the sample electrical resistivity [42] was also shown to be proportional to log(t) at long times [21], although the interpretation of such data can be complex [43]. The deflection measured at T = 34 ◦ C for one low and one high filler content RBC is shown in Fig. 1. For each RBC, the deflection at T = 34 ◦ C is systematically larger than that collected at T = 26 ◦ C. The increase in deflection with temperature is attributed to the delayed gelation at the higher temperature. This results in prolonged monomer mobility and ultimately a higher degree of conversion [12]. This behavior is reflected, in part, for the high filler content RBC with a longer tMAX of 0.88 ± 0.04 s at T = 34 ◦ C compared to tMAX of 0.76 ± 0.03 s at T = 26 ◦ C (p < 0.05). However, tMAX is the same (p < 0.05) at both temperatures (tMAX = 0.88 ± 0.05 s) for the low filler content RBC. It is interesting to note that tMAX for the high filler content RBC at T = 34 ◦ C and having a lower viscosity than at T = 26 ◦ C is the same (p < 0.05) as that for the low filler content
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
RBC. A comparison of tMAX between the Filtek Bulk Fill Flowable Restorative A2 shade (tMAX = 0.88 ± 0.05 s) and SureFil SDR Flow Posterior Bulk Fill A3 shade (tMAX = 1.02 ± 0.05 s) indicate that the latter is significantly longer (p < 0.05) than the former. These two RBCs have very similar filler content. Other factors such as the reactivity of the monomer-photoinitiator system may also play important roles in determining tMAX . Although the ˛S listed in Table 2 differs from RBC to RBC, a strong correlation (R2 adj. = 0.853) is noted in Fig. 3(a) between ˛S and the filler volume content, namely, as the filler content increases ˛S decreases. This is expected because the filler content’s role is to minimize the overall shrinkage of the polymer matrix/filler content composite and hence of ˛S . An assessment of the post-curing shrinkage relative to the shrinkage at 100 s is given by <D/D(100 s)> for all the measured RBCs and at the two temperatures. The deflection data used in this evaluation were collected at t = 100 s where the contribution of the thermal contraction to the deflection caused by the LCU turning off at t = 40 s is minimized and at t = 15 h, which was the maximum data collection time for one RBC. Although <D/D(100 s)> was evaluated at a data collection time of only 15 h and not after the post-curing is complete which may be one to four weeks later [17,18], a lower limit to the post-curing shrinkage can be estimated. Table 2 indicates that the maximum value of <D/D(100 s)> is obtained for the Filtek Bulk Fill Posterior Restorative A2 shade with 29.4 ± 0.4% and 22.5 ± 0.1% at T = 26 ◦ C and T = 34 ◦ C, respectively. At T = 26 ◦ C, the post-curing shrinkage among all RBCs investigated varied between 15.8 and 29.4% with an average post-curing shrinkage of 21 ± 5%. It is important to note that <D/D(100 s)> is significantly smaller (p < 0.05) at near tooth temperature than at room temperature. Among the variables listed in Table 2 and excluding the data point collected with the X-tra fil Universal shade RBC, a correlation is shown in Fig. 3(b) between the post-cure shrinkage and the maximum shrinkage rate, namely, as (dD/dt)MAX increases <D/D(100 s)> decreases with a R2 adj. = 0.929 for a linear relationship. In other words, the faster the RBC goes from the liquid to the gel to the glass state, the smaller is the post-cure shrinkage. However, the total post-cure time when the RBC in the glass state stops shrinking and reaches thermal equilibrium is not included in the analysis. This total time may well be longer than for a smaller (dD/dt)MAX as more time may be required to reach thermal equilibrium for an RBC in a deep glass state. Note that the data point obtained using the X-tra fil Universal shade RBC is anomalous and may be related to the proprietary monomers used for this RBC. In Figs. 4 and 5 the degree of polymerization and hence of the glass state in the RBC Filtek Bulk Fill Posterior Restorative A2 shade at T = 26 ◦ C was varied by using five photo-curing exposure times. As depicted in Fig. 4(b), the second research hypothesis that ˛S does not vary with the light exposure time is rejected. Moreover, Fig. 5(b) indicates that the third and fourth research hypothesis, namely, that at a long time, the degree of conversion is not proportional to log(t) and that ˛DC does not vary with the light exposure time are both rejected. Considering that the samples used to carry out the DC measurements were made under the same experimental conditions as those used for the shrinkage measurements where the log(t) dependence was clearly observed, we attribute in
1375
part the log(t) dependence observed in the DC data to the densification of the out-of-equilibrium RBC polymer glass caused by the thermal diffusion of voids to the sample surface. The log(t) dependence for the degree of conversion has been previously reported [23], but the mechanism used to describe the time evolution is novel. A typical behavior observed in Figs. 4 and 5 is that the magnitude of the axial shrinkage and degree of conversion at the longest observation time increases at longer LCU exposure times while the opposite is true for ˛S and ˛DC . The ˛S listed in Table 2 differ from RBC to RBC due to the varying chemical nature, filler content, and reactivity of each RBC ((˛S )MAX /<˛S >) = 39%). Table 3 gives the values of ˛S ((˛S )MAX /<˛S >) = 40%) caused by different exposure time resulting in different glass state for a single RBC at T = 26 ◦ C. As shown in Table 3, the variation in ˛DC caused by the different exposure times is comparable ((˛DC )MAX /<˛DC >) = 36%). These results indicate that the longer the exposure time, the higher is the degree of conversion, the higher is the glass state of the photo-cured RBC, and hence the smaller are ˛S and ˛DC . This is shown in Fig. 6(a) and (b) where a high level of correlation is obtained between ˛S and ˛DC and the degree of conversion collected at t = 580 s. In turn, a positive correlation is also observed in Fig. 6(c) between ˛DC and ˛S . In a previous study [44], the void formation in photo-cured BisGMA/TEGDMA polymer glass network was investigated using positron annihilation lifetime spectroscopy. For the 40 mol% of TEGDMA in the co-monomer composition and for five exposure times, the average free volume of the voids in the polymer network decreased with increasing exposure time and with increasing DC. Although the relative void concentration with exposure time was not determined, the results suggest that the density of the polymer glass increases with increasing DC and with decreasing average free volume of the voids. In turn, the decreasing void free volume and increase in density of the polymer glass would lead to a smaller ˛S as illustrated in Fig. 4. Similarly, if the increase in DC in the post-cure region as depicted in Fig. 5 is mainly driven by the densification of the polymer network then ˛DC would decrease with increasing DC and density, and as the average free volume of the voids decrease. The degree of post-curing can also be evaluated by using <DC/DC(25 s)> with the DC data listed in Table 3 for the five exposure times. The DC related post-curing ranged from 20.4 ± 0.4% to as high as 50.0 ± 0.2% for exposure times from 20 s to 1 s, respectively. Using the recommended manufacturer’s exposure time of 20 s, the DC increases by as much as 20% relative to the DC at t = 25 s in the first 9 min after light exposure. The increase in DC due to post-curing is attributed in part to the post shrinkage of the out-of-equilibrium RBC that is in a glass state before it reaches thermal equilibrium. A measure of the crosslinking density within the polymer matrix is provided by the RBC hardness measured 1 h after light exposure. A high level of correlation is obtained between the hardness measured at the top surface of the sample and degree of conversion in Fig. 7(a). Moreover, the high correlation between ˛S and ˛DC and the hardness confirms that the shrinkage and DC related physical aging rate decreases where there is a higher degree of cross-linking and with the deeper
1376
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
glass state of the cured RBC. A similar analysis was performed using the hardness measured on the bottom surface of the RBC samples with the same conclusion.
5.
Clinical significance
Most previous studies have reported accurate shrinkage data at short times after light exposure. These values are misleadingly low as the shrinkage continues to increase with time. The results of this study indicate that the shrinkage occurring during post-curing can be quite appreciable; namely, 22.5% at 15 h relative to that at t = 100 s after light exposure at near tooth temperature for a contemporary Bulk Fill RBC. This post shrinkage results in a gradual built up of stress within the tooth wall structure, potentially shortening the restoration’s life. Different approaches have been taken by RBC manufacturers to minimize the built-in stress within the restored teeth. Recent shrinkage stress data [41] collected on five contemporary Bulk Fill RBCs at up to 12 h and T = 33 ◦ C showed that four RBCs had the same shrinkage stress values (p < 0.05) at times of at least 6 h. One of the four RBCs was the low filler content SureFil SDR flow Posterior Bulk Fill A3 shade. The shrinkage strain for this RBC is appreciably higher than those for all the high filler content RBCs used in this work providing substantial evidence of the effectiveness of the proprietary monomers used to release stress during and after photo-curing. Furthermore, the Filtek Bulk Fill Restorative A2 shade, which has an appreciable post-curing deflection at 15 h, has a low value of shrinkage stress at 12 h [41]. This illustrates the effectiveness of the novel monomers that are included in this RBC to relieve stress both during and after photo-curing.
6.
Conclusions
The effects of post-curing on the shrinkage of six commercial contemporary RBCs with filler contents up to 70 vol.% were evaluated. Axial shrinkage, the degree of conversion, and hardness at a temperature of 26 ◦ C and 34 ◦ C were measured. Within the limitations of this study, it can be concluded that:
1) Axial shrinkage in photo-cured RBCs at times of at least 15 h after light exposure displayed a functional dependence that was proportional to the logarithm of time. This time dependence is a signature of the free volume fraction model. 2) The shrinkage related physical aging rate at T = 26 ◦ C differed from RBC to RBC by up to 39%. A high correlation was obtained between shrinkage related physical aging rate and filler volume content. 3) In samples prepared using the same experimental conditions as those which displayed post-curing shrinkage proportional to log(t), the degree of conversion in these photo-cured RBCs samples was found to be proportional to the logarithm of time shortly after the photoexcitation light was turned off. This time dependence was attributed in part to the applicability of the free volume fraction model to the degree of conversion of the RBC samples.
4) The level of the out-of-equilibrium state in the RBCs in the glass state was varied by using different LCU exposure times. The shrinkage and degree of conversion related physical aging rate were found to be proportional to the degree of conversion collected at t = 580 s and Vickers hardness measured 1 h after light exposure. A direct correlation (R2 adj. = 0.669) was found between the shrinkage related physical aging rate and degree of conversion related physical aging rate. 5) The post-curing shrinkage in dental photo-cured RBCs can be appreciable (an additional 15.8–29.4% relative to the shrinkage at t = 100 s and T = 26 ◦ C). RBC blend and photo-curing conditions should be optimized to reduce the post-curing shrinkage.
Acknowledgments We would like to thank L. Kreplak and J. Stansbury for a fruitful discussion. We would like to express our sincere gratitude to C. Hudson 3M for the donation of the LCU, ceramic primer, and RBCs and, Voco GmbH, and DENTSPLY SIRONA for the donation of the RBCs used in this study. We thank G. Earle IObIO Inc. for the donation of the LabVIEW program used for the data collection. This work was supported in part by the Department of Physics and Atmospheric Science, and Department of Dental Clinical Sciences, Dalhousie University.
references
[1] Leprince JG, Palin WM, Hadis MA, Devaux J, Leloup G. Progress in dimethacrylate-based dental composite technology and curing efficiency. Dent Mater 2013;29:139–56. [2] Ferracane JL. Resin composite—state of the art. Dent Mater 2011;27:29–38. [3] Cramer NB, Stansbury JW, Bowman CN. Recent advances and developments in composite dental restorative materials. J Dent Res 2011;90:402–16. [4] Ferracane JL. Buonocore lecture placing dental composites—a stressful experience. Oper Dent 2008;33:247–57. [5] Oliveira LRS, Braga SSL, Bicalho AA, Ribeiro MTH, Price RB, Soares CJ. Molar cusp deformation evaluated by micro-CT and enamel crack formation to compare incremental and bulk-filling techniques. J Dent 2018;74:71–8. [6] Moorthy A, Hogg CH, Dowling AH, Grufferty BF, Benetti AR, Fleming GJ. Cuspal deflection and microleakage in premolar teeth restored with bulk-fill flowable resin-based composite base materials. J Dent 2012;40:500–5. [7] Alvanforoush N, Palamara J, Wong RH, Burrow MF. Comparison between published clinical success of direct resin composite restorations in vital posterior teeth in 1995–2005 and 2006–2016 periods. Aust Dent J 2017;62:132–45. [8] van Dijken JWV, Pallesen U. Bulk-filled posterior resin restorations based on stress-decreasing resin technology: a randomized, controlled 6-year evaluation. Eur J Oral Sci 2017;125:303–9. [9] Ilie N, Hickel R. Investigations on a methacrylate-based flowable composite based on the SDR technology. Dent Mater 2011;27:348–55. [10] Shah PK, Stansbury JW, Bowman CN. Application of an addition-fragmentation-chain transfer monomer in
d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1367–1377
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25] [26] [27]
di(meth)acrylate network formation to reduce polymerization shrinkage stress. Polym Chem 2017;8:4339–51. Blackham JT, Vandewalle KS, Lien W. Properties of hybrid resin composite systems containing prepolymerized filler particles. Oper Dent 2009;34:697–702. Stansbury JW. Dimethacrylate network formation and polymer property evolution as determined by the selection of monomers and curing conditions. Dent Mater 2012;28:13–22. Al-Ahdal K, Ilie N, Silikas N, Watts DC. Polymerization kinetics and impact of post polymerization on the Degree of Conversion of bulk-fill resin-composite at clinically relevant depth. Dent Mater 2015;31:1207–13. Par M, Gamulin O, Marovic D, Klaric E, Tarle Z. Raman spectroscopic assessment of degree of conversion of bulk-fill resin composites-changes at 24 hours post cure. Oper Dent 2015;40:E92–101. Par M, Gamulin O, Marovic D, Klaric E, Tarle Z. Effect of temperature on post-cure polymerization of bulk-fill composites. J Dent 2014;42:1255–60. Scotti N, Venturello A, Borga FA, Pasqualini D, Paolino DS, Geobaldo F, et al. Post-curing conversion kinetics as functions of the irradiation time and increment thickness. J Appl Oral Sci 2013;21:190–5. Schneider LF, Consani S, Ogliari F, Correr AB, Sobrinho LC, Sinhoreti MA. Effect of time and polymerization cycle on the degree of conversion of a resin composite. Oper Dent 2006;31:489–95. Sideridou ID, Vouvoudi EC, Adamidou EA. Dynamic mechanical thermal properties of the dental light-cured nanohybrid composite Kalore, GC: effect of various food/oral simulating liquids. Dent Mater 2015;31:154–61. Alnazzawi A, Watts DC. Simultaneous determination of polymerization shrinkage, exotherm and thermal expansion coefficient for dental resin-composites. Dent Mater 2012;28:1240–9. Truffier-Boutry D, Demoustier-Champagne S, Devaux J, Biebuyck JJ, Mestdagh M, Larbanois P, et al. A physico-chemical explanation of the post-polymerization shrinkage in dental resins. Dent Mater 2006;22:405–12. Steinhaus J, Frentzen M, Rosentritt M, Moginger B. Dielectric analysis of short-term and long-term curing of novel photo-curing dental filling materials. Mod Trends Polym Sci—Epf 09 2010;296:622–5. Leung RL, Fan PL, Johnston WM. Post-irradiation polymerization of visible light-activated composite resin. J Dent Res 1983;62:363–5. Selig D, Haenel T, Hausnerova B, Moeginger B, Labrie D, Sullivan B, et al. Examining exposure reciprocity in a resin based composite using high irradiance levels and real-time degree of conversion values. Dent Mater 2015;31:583–93. Angell CA. Formation of glasses from liquids and biopolymers. Science 1995;267:1924–35. Hutchinson JM. Physical aging of polymers. Prog Polym Sci 1995;20:703–60. Tant MR, Wilkes GL. An overview of the nonequilibrium behavior of polymer glasses. Polym Eng Sci 1981;21:874–95. Curro JG, Lagasse RR, Simha R. Diffusion model for volume recovery in glasses. Macromolecules 1982;15:1621–6.
1377
[28] Jin KL, Li LQ, Torkelson JM. Bulk physical aging behavior of cross-linked polystyrene compared to its linear precursor: effects of cross-linking and aging temperature. Polymer 2017;115:197–203. [29] Cangialosi D, Alegria A, Colmenero J. Effect of nanostructure on the thermal glass transition and physical aging in polymer materials. Prog Polym Sci 2016;54–55:128–47. [30] Cangialosi D, Boucher VM, Alegria A, Colmenero J. Volume recovery of polystyrene/silica nanocomposites. J Polym Sci B—Polym Phys 2013;51:847–53. [31] Cangialosi D, Boucher VM, Alegria A, Colmenero J. Free volume holes diffusion to describe physical aging in poly(mehtyl methacrylate)/silica nanocomposites. J Chem Phys 2011;135:014901. [32] Rueggeberg FA, Daronch M, Browning WD, DE Goes MF. In vivo temperature measurement: tooth preparation and restoration with preheated resin composite. J Esthet Restor Dent 2010;22:314–22. [33] Runnacles P, Arrais CA, Pochapski MT, dos Santos FA, Coelho U, Gomes JC, et al. Direct measurement of time-dependent anesthetized in vivo human pulp temperature. Dent Mater 2015;31:53–9. [34] Price RB, Rueggeberg FA, Labrie D, Felix CM. Irradiance uniformity and distribution from dental light curing units. J Esthet Restor Dent 2010;22:86–101. [35] Price RB, Labrie D, Rueggeberg FA, Felix CM. Irradiance differences in the violet (405 nm) and blue (460 nm) spectral ranges among dental light-curing units. J Esthet Restor Dent 2010;22:363–77. [36] Watts DC, Cash AJ. Determination of polymerization shrinkage kinetics in visible-light-cured materials: methods development. Dent Mater 1991;7:281–7. [37] Watts DC, Marouf AS. Optimal specimen geometry in bonded-disk shrinkage-strain measurements on light-cured biomaterials. Dent Mater 2000;16:447–51. [38] Lee IB, Cho BH, Son HH, Um CM, Lim BS. The effect of consistency, specimen geometry and adhesion on the axial polymerization shrinkage measurement of light cured composites. Dent Mater 2006;22:1071–9. [39] Kostylev I, Labrie D, Price RBT. Time-resolved 2D shrinkage field of dental resins using laser interferometry. Appl Opt 2015;54:1852–60. [40] Kostylev I, Winters T, Labrie D, Price RB. Role of the beam profile and sample geometry in the bonded-disk method on the shrinkage kinetics of two dental resins. Dent Mater 2016;32:1026–35. [41] Kalliecharan D, Germscheid W, Price RB, Stansbury J, Labrie D. Shrinkage stress kinetics of Bulk Fill resin-based composites at tooth temperature and long time. Dent Mater 2016;32:1322–31. [42] Zahouily K, Decker C, Kaisersberger E, Gruener M. Real-time UV cure monitoring. Eur Coat J 2003;49:14–8. [43] Steinhaus J, Hausnerova B, Haenel T, Grossgarten M, Moginger B. Curing kinetics of visible light curing dental resin composites investigated by dielectric analysis (DEA). Dent Mater 2014;30:372–80. [44] Pfeifer CS, Shelton ZR, Braga RR, Windmoller D, Machado JC, Stansbury JW. Characterization of dimethacrylate polymeric networks: a study of the crosslinked structure formed by monomers used in dental composites. Eur Polym J 2011;47:162–70.