Shrinkage stress kinetics of Bulk Fill resin-based composites at tooth temperature and long time

Highlights

  • We modified a tensometer with increased data collection rate and accuracy.

  • Five Bulk Fill resin-based composites (RBCs) were investigated at T = 33 °C.

  • At times ≥6 h, the shrinkage stress mean value for four RBCs were very similar.

  • Correlations were observed among several shrinkage stress kinetics parameters.

Abstract

Objective

To determine the shrinkage stress kinetics at up to 12 h after light exposure and at tooth temperature during placement of selected Bulk Fill resin-based composites (RBCs).

Methods

Five representative Bulk Fill RBCs from four companies were chosen with a wide range of viscosity and filler volume content. The shrinkage stress kinetics at T = 33 °C was measured continuously over a period of 12 h using a modified tensometer with the ability to measure the cantilever beam deflection to better than 40 nm accuracy at a sampling rate of up to 200 samples/s, and thermally stable resulting in a measurement accuracy better than 0.05 MPa at 12 h. The tensometer compliance was 0.105 μm/N. A custom made heater was used to control the RBC sample temperature at T = 33 °C with a temperature gradient across the sample of less than 1 °C. The samples were irradiated for 20 s with irradiance of 1.1 W/cm 2 and total energy density of 22 J/cm 2 . Three samples ( n = 3) were used for each RBCs.

Results

The shrinkage stress at 12 h for the five Bulk Fill RBCs ranged from 2.21 to 3.05 MPa, maximum stress rate ((d S /d t ) M ) varied from 0.18 to 0.41 MPa/s, time at which the maximum stress rate occurred ( t Max ) were between 1.42 to 3.24 s and effective gel time ( t gel ) varied from 50 to 770 ms. Correlations were observed between (d S /d t ) M and t Max ( r = −0.946), t Max and filler volume fraction ( r = −0.999), and between the shrinkage stress at 12 h and t gel ( r = 0.994). However, no correlation was observed between the stress at 12 h and filler volume fraction.

Significance

The shrinkage stress for four of the five Bulk Fill RBCs were not significantly different ( p < 0.05) at 6 h and beyond after photo-curing and that fully developed stress induced by photo-cured RBCs may only be reached at times longer than 12 h.

Introduction

Since the introduction of resin-based composites (RBCs) by Bowen in 1958 as an alternative to dental amalgam toward tooth restorations, tremendous research and development activities have followed . The primary motivation for such activities is the esthetic restoration of the tooth to its original appearance with a long clinical life. Recently, developments in novel monomers, monomer blends and filler nanotechnology , photoinitiators , and improved RBC transparency to blue light allow the placement and photocuring of up to 4 mm thick increment of “Bulk Fill” RBCs in the restoration . Despite such improvements over a 50 year period, the fundamental property of shrinkage strain of the RBC upon photocuring and its links on the viability of the restoration remains . The shrinkage strain and resulting development of built in stress on the tooth structure may result in cuspal deflection, marginal gap formation and leakage, enamel cracking and may eventually lead to restoration fracture, secondary caries, and ultimate clinical failure of the restoration .

The issue of shrinkage stress on the tooth walls was identified and two general approaches were introduced to quantify its impact . The ratio of the bonded to unbonded surface area of the RBC cylindrical shaped sample, the C-factor, was observed to play an important role toward the determination of the final stress value . Its value was found to decrease with decreasing C-factor under otherwise the same experimental conditions. These results were interpreted in terms of RBC flow and strain relaxation during the pre-gel phase of the curing RBC sample. These results pointed out the role of the gel time and RBC viscosity during the development of the shrinkage induced stress where low viscosity flowable RBCs would favor strain relaxation via flow at the beginning of photo-polymerization before the gel time than highly viscous RBCs. In recent studies , the shrinkage stress kinetics was measured on conventional RBCs at both room and clinical mouth temperature and over a period of up to 1 h after light exposure . It was observed that for all six RBCs tested the mean shrinkage stress was significantly higher ( p ≤ 0.007) at 60 min compared to that at 2 min after light exposure at 23 °C while the mean shrinkage stress was not significantly different ( p ≥ 0.112) at both times at 37 °C. Shrinkage stress studies were performed on “Bulk Fill” RBCs at room temperature with detailed analysis and conclusions drawn on samples with data collected over a period of time as short as 3 min. The longest duration experiments were carried out over 30 min.

The objective of this work was to investigate the shrinkage stress kinetics for up to 12 h at T = 33 °C for selected commercial Bulk Fill RBCs. The null hypotheses were: (1) the shrinkage stress is the same at 2 min, 1 h and 12 h for each RBC at a RBC temperature of 33 °C, (2) there is no correlation between the maximum shrinkage stress rate and time at which it occurs t Max , (3) there is no correlation between t Max and the filler volume fraction, (4) there is no correlation between the shrinkage stress at 12 h and the filler volume fraction, and (5) there is no correlation between the shrinkage stress at 12 h and effective gel time.

Materials and methods

Materials

Five commercial Bulk Fill RBCs were selected in this study. The RBC compositions as described by the manufacturers are given in Table 1 . Two low-viscosity (3M ESPE Filtek Flowable Restorative A2 shade and DENTSPLY Surefil SDR flow Posterior A3 shade), two medium viscosity (3M ESPE Filtek Posterior A2 shade and Voco GmbH X-tra fil) and one high viscosity (Ivoclar Vivadent Inc., Tetric Evoceram Posterior Bulk Fill IVA shade) were used to study the shrinkage stress with time, viscosity, gel time, and filler volume fraction.

Table 1
Composition of the Bulk Fill resin-based composites as provided by the manufacturers.
Product Manufacturer Resin matrix Filler Filler (wt.%/vol.%)
(lot number)
Filtek Bulk Fill Flowable Restorative A2 shade 3M ESPE bisGMA, UDMA, bisEMA(6) and Procrylat monomers 10 nm to 3.5 μm zirconia/silica filler, 100 nm to 5.0 μm ytterbium trifluoride filler 64.5%/42.5%
(N402919)
Filtek Bulk Fill Posterior Restorative A2 shade 3M ESPE proprietary AUDMA and AFM, DDDMA and UDMA 20 nm silica filler, 4 to 11 nm zirconia filler, and 100 nm ytterbium trifluoride filler 76.5%/58.4%
(N613079)
Tetric EvoCeram IVA Posterior Bulk Fill Ivoclar Vivadent bisGMA, bisEMA, and UDMA 100 nm to 1 μm Ba Al silicate glass filler, prepolymer particles, 200 nm ytterbium fluoride filler, 160 nm mixed spherical oxide filler 80%/61%
(U12196)
SureFil SDR flow Posterior Bulk Fill A3 Shade DENTSPLY SDR patented UDMA, DMA, di-functional diluent Ba B F Sr Al silicate glass filler 68%/45%
(120127)
Voco X-tra fil Voco GmbH bisGMA, UDMA, TEGDMA 2–3 μm Ba B Al Si glass filler 86%/70.1%
(1215125)
BisGMA: bisphenol A glycidyl methacrylate, BisEMA: ethoxylatedbis-phenol A dimethacrylate, BisEMA(6): (2,2-bis[4-methacryloxypolyethoxyphenyl)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).

Shrinkage stress and gel time measurements

A schematic diagram of the modified tensometer (American Dental Association Foundation, Paffenbarger Research Center, Gaithersburg, MD, USA) used to measure the shrinkage stress kinetics and gel time of RBCs at 33 °C is shown in Fig. 1 . The cantilever beam deflection was measured simultaneously with the built-in tensometer LVDT (Linear Variable Differential Transformer) position sensor and with an optical technique. The optical technique consists of using an interferometer where the “moving” mirror is mounted on the cantilever beam and the HeNe laser, beam splitter, fixed reference mirror, and Silicon photodiodes were mounted rigidly on the tensometer base. As the beam deflects caused by the photocuring of the RBC, constructive and destructive interferences of the recombined laser beam occur at the photodiodes. For instance, the cantilever beam deflecting from 0 to λ /2 at the “moving” mirror where λ is the HeNe emission wavelength (632.8 nm) results in a full cycle (fringe) of constructive to destructive to constructive laser light interference at the photodiodes. The interferometric technique is highly sensitive to beam deflection where a deflection of 1/8th of a cycle (fringe), corresponding to a beam deflection of 40 nm at the moving mirror, can be easily measured. In contrast, the detection limit of the LVDT position sensor provided with the apparatus is 100 nm. This measurement approach toward monitoring an oscillatory instead of a slowly varying signal is not sensitive to electronic drift in the detection system.

Fig. 1
Modified ADAF tensometer to measure accurately the shrinkage stress at tooth temperature and long time. A Michelson interferometer was set up to measure the cantilever beam deflection.

A simultaneous calibration done at room temperature using the LVDT position sensor and interferometric photodiode signal was performed by applying a calibrated force at the sample position and measuring the beam deflection at the position sensor and photodiodes. An effective cantilever spring constant of k eff = 4.283 N/μm was obtained using the position sensor. Using the overall beam length L = 29.3 cm and the distance between the beginning of the beam and sample position a = 12.8 cm (using the same parameters as in Ref. ), the cantilever spring constant experienced by the sample is k sample = 12.6 N/μm. Using a quartz rod length of 25.85 mm protruding outside the clamp for the upper and lower rod, a rod diameter of 6 mm and a Quartz Young’s modulus of 72 GPa a tensometer compliance of 0.105 μm/N was calculated.

A data acquisition hardware (National Instrument, USB-6281) and custom made LabVIEW program (National Instrument) were used to carry out data collection over 12 h at a sampling rate of 200 samples/s during the first hour and then at 1 sample/s where each sample was the average of 720 data points collected over 0.1 s. The recorded signals by the data acquisition hardware were the amplified signals from the two photodiodes monitoring the light output from the interferometer, the amplified photodiode signal measuring the LCU light transmitted through the RBC sample, output voltage from the cold junction compensator connected to a type T thermocouple (Omega, CJ-T), and voltage output from the control electronics monitoring the LVDT position sensor. Twenty seconds after the start of data collection, the custom program turned on and then off the LCU for a selected duration.

As shown in Fig. 1 , a copper block heater was clamped on the top and bottom quartz rod. Due to the quartz low thermal conductivity, only 1 W of electrical power was required to heat the RBC sample to 33 °C and the heater temperature was 34 °C indicating that the temperature gradient within the sample was less than 1 °C. Approximately 50 min was required to reach a steady sample temperature of 33 °C after the turning on of the heaters.

For mechanical and thermal stability, the tensometer was bolted on a 38 cm wide by 86.5 cm long by 2.5 cm thick aluminum plate and inserted into a thermally isolated enclosure. The LVDT position sensor electronic control was left outside the enclosure. The aluminum plate also acted as a large thermal reservoir to help maintain a constant temperature for the tensometer at room temperature. Fig. 2 shows a comparison between the shrinkage stresses measured simultaneously using the position sensor and the optical technique. To avoid daytime disturbance, the data were collected from 6 p.m. to 6 a.m. Fig. 2 (a) shows the case where the experimental procedure was followed except that no RBC was inserted between the two quartz rods. The tensometer was at room temperature while the block heater temperature was set at 33 °C. The heater temperature drifted by 0.30 °C over 12 h because no electronic feedback circuitry was used toward its control. The data collected using the position sensor drifted continuously with time reaching a value of −0.04 MPa at 1 h and −0.15 MPa at 12 h. The drift in the data is tentatively attributed in part to the drop in the laboratory temperature overnight resulting in a corresponding electronic drift in the LVDT position sensor electronic control. The shrinkage stress derived using the optical technique also drifted with time with a value of +0.01 MPa at 1 h and +0.035 MPa at 12 h. Since our optical set up is not designed to monitor the direction of deflection of the cantilever beam, the derived shrinkage stress will always be positive.

Fig. 2
Measurements of the tensometer stability with time using two simultaneous cantilever beam deflection monitoring systems. In (a) the procedure for the data collection was followed except no RBC sample was placed between the ends of the two quartz rods. The block heater temperature was set at T = 33 °C and drifted by less than 0.30 °C over 12 h while the tensometer was at room temperature. Typical shrinkage stress data (b) obtained using two 3M ESPE Bulk Fill RBCs at T = 33 °C.

The drift in shrinkage stress of 0.035 MPa measured using the optical technique is due to the fact that the two interferometer arms have different lengths and the two mirrors and beam splitter are mounted on metallic supports bolted to the tensometer. When the tensometer temperature varies the difference between the two arm lengths does not remain constant but also varies with temperature due to the thermal expansion of the metallic supports. As a result, the optical path length varies with temperature which can be interpreted as a deflection of the cantilever arm. A tensometer temperature drift of only 0.24 °C near room temperature is sufficient to account for the measured stress drift of 0.035 MPa. The analysis was confirmed by repeating the above measurements and monitoring the tensometer temperature over a 12 h period. The tensometer temperature drift was estimated to be less than 0.5 °C for the results presented in this study.

Fig. 2 (b) shows typical results obtained using the LVDT position sensor and optical technique for the shrinkage stress as a function of time of two Bulk Fill RBCs, 3M ESPE Flowable Restorative A2 shade and 3M ESPE Posterior A2 shade at a RBC temperature of 33 °C. Note that the shrinkage stress using both techniques parallel each other with a discrepancy of 0.03 MPa and 0.05 MPa at 1 h and, 0.14 MPa and 0.30 MPa at 12 h for the Posterior and Flowable RBC, respectively. Note that the observed discrepancy at 12 h for both RBCs are typical for the data presented in this study.

The gel time, time after the start of photopolymerization when the RBC sample converts from a liquid to a gel state across its whole thickness, was estimated by monitoring the start of deflection of the cantilever beam relative to the start of the LCU light transmission through the RBC sample. At this “effective” gel time, the RBC has developed a load bearing structure which enables the RBC to exert a force on the cantilever beam. This approach is consistent with data on the simultaneous measurements of a RBC shrinkage stress and degree of conversion (DC) as a function of time where the start of deflection of the cantilever beam was associated with a DC of 3–5% within the RBC, the latter value is typical for the DC at the gel point in methacrylates . It should be noted that this approach is not a direct measurement of the gel point as that obtained using a rheometer.

Sample preparation and photocuring

Both ends of each 6 mm diameter quartz rods were polished using 600 and then 1200 grit SiC sand paper with water as a lubricant to minimize reflection losses to blue light. The two ends facing the RBC sample were salinized using a ceramic primer (RelyX, 3M ESPE). The separation between the two quartz rods of 1.05 ± 0.04 mm which defines the sample thickness, d , was set using a filler gage. The RBC was inserted in the gap between the two rods and care was taken to ensure that the sample took the form of a cylinder with no excess of RBC along the rods. The heaters were turned on for 60 min before photocuring to reach a steady state temperature of 33 ± 1 °C. Over the course of the experiments the block heater temperature drifted by less than 0.5 °C.

The RBC samples were photocured for 20 s with a 455 nm single emission wavelength LED-based light curing unit (Elipar S10 LED unit, 3M ESPE, St Paul, MN, USA) with a radiant power transmitted through a 6 mm diameter quartz rod of 310 mW and irradiance of 1.1 W/cm 2 . The radiant power was measured using a calibrated thermopile and meter (PM-10 detector and FieldMax meter, Coherent Inc., Santa Clara, CA, USA) .

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 are any statistical differences in the data collected under different experimental conditions ( α = 0.05). Correlation analysis was performed on data exhibiting linear relationships between the variables.

Materials and methods

Materials

Five commercial Bulk Fill RBCs were selected in this study. The RBC compositions as described by the manufacturers are given in Table 1 . Two low-viscosity (3M ESPE Filtek Flowable Restorative A2 shade and DENTSPLY Surefil SDR flow Posterior A3 shade), two medium viscosity (3M ESPE Filtek Posterior A2 shade and Voco GmbH X-tra fil) and one high viscosity (Ivoclar Vivadent Inc., Tetric Evoceram Posterior Bulk Fill IVA shade) were used to study the shrinkage stress with time, viscosity, gel time, and filler volume fraction.

Table 1
Composition of the Bulk Fill resin-based composites as provided by the manufacturers.
Product Manufacturer Resin matrix Filler Filler (wt.%/vol.%)
(lot number)
Filtek Bulk Fill Flowable Restorative A2 shade 3M ESPE bisGMA, UDMA, bisEMA(6) and Procrylat monomers 10 nm to 3.5 μm zirconia/silica filler, 100 nm to 5.0 μm ytterbium trifluoride filler 64.5%/42.5%
(N402919)
Filtek Bulk Fill Posterior Restorative A2 shade 3M ESPE proprietary AUDMA and AFM, DDDMA and UDMA 20 nm silica filler, 4 to 11 nm zirconia filler, and 100 nm ytterbium trifluoride filler 76.5%/58.4%
(N613079)
Tetric EvoCeram IVA Posterior Bulk Fill Ivoclar Vivadent bisGMA, bisEMA, and UDMA 100 nm to 1 μm Ba Al silicate glass filler, prepolymer particles, 200 nm ytterbium fluoride filler, 160 nm mixed spherical oxide filler 80%/61%
(U12196)
SureFil SDR flow Posterior Bulk Fill A3 Shade DENTSPLY SDR patented UDMA, DMA, di-functional diluent Ba B F Sr Al silicate glass filler 68%/45%
(120127)
Voco X-tra fil Voco GmbH bisGMA, UDMA, TEGDMA 2–3 μm Ba B Al Si glass filler 86%/70.1%
(1215125)
BisGMA: bisphenol A glycidyl methacrylate, BisEMA: ethoxylatedbis-phenol A dimethacrylate, BisEMA(6): (2,2-bis[4-methacryloxypolyethoxyphenyl)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).

Shrinkage stress and gel time measurements

A schematic diagram of the modified tensometer (American Dental Association Foundation, Paffenbarger Research Center, Gaithersburg, MD, USA) used to measure the shrinkage stress kinetics and gel time of RBCs at 33 °C is shown in Fig. 1 . The cantilever beam deflection was measured simultaneously with the built-in tensometer LVDT (Linear Variable Differential Transformer) position sensor and with an optical technique. The optical technique consists of using an interferometer where the “moving” mirror is mounted on the cantilever beam and the HeNe laser, beam splitter, fixed reference mirror, and Silicon photodiodes were mounted rigidly on the tensometer base. As the beam deflects caused by the photocuring of the RBC, constructive and destructive interferences of the recombined laser beam occur at the photodiodes. For instance, the cantilever beam deflecting from 0 to λ /2 at the “moving” mirror where λ is the HeNe emission wavelength (632.8 nm) results in a full cycle (fringe) of constructive to destructive to constructive laser light interference at the photodiodes. The interferometric technique is highly sensitive to beam deflection where a deflection of 1/8th of a cycle (fringe), corresponding to a beam deflection of 40 nm at the moving mirror, can be easily measured. In contrast, the detection limit of the LVDT position sensor provided with the apparatus is 100 nm. This measurement approach toward monitoring an oscillatory instead of a slowly varying signal is not sensitive to electronic drift in the detection system.

Nov 23, 2017 | Posted by in Dental Materials | Comments Off on Shrinkage stress kinetics of Bulk Fill resin-based composites at tooth temperature and long time

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos