Analysis of temperature change during polymerization according to resin thickness: an in vitro experimental study
Article information
Abstract
Objectives
This study aimed to analyze the temperature changes during the light curing of conventional flowable composite resin and bulk-fill composite resin of various thicknesses using an infrared thermographic camera.
Methods
Flowable composite resin (G-aenial Flo, GC Co.) and bulk-fill composite resin (SDR, Dentsply Caulk) were used. Specimens with thicknesses from 0.5 mm to 5.0 mm were prepared. The infrared thermographic camera measured the temperature changes at the maximum temperature rise point during light curing. The data were analyzed for maximum temperature, time to peak temperature, and temperature rise patterns.
Results
For G-aenial Flo, the maximum temperature tended to decrease with increasing thickness, whereas for SDR, the maximum temperature decreased up to 2.0 mm and then remained relatively consistent from 2.0 mm to 5.0 mm. At thicknesses of 1.5 mm or less, both resins showed a rapid temperature increase within the first 5 seconds, followed by a reduced rate of increase up to 80 seconds. At thicknesses of 2.0 mm or greater, the temperature peaked and then gradually decreased. Across all thicknesses, SDR was observed to reach peak temperature more rapidly than G-aenial Flo.
Conclusions
Observable differences in polymerization dynamics were identified between the two resin types, particularly at greater thicknesses. Although no statistical analysis was performed, these descriptive findings suggest that infrared thermographic cameras may be useful for indirectly assessing polymerization dynamics during resin polymerization.
INTRODUCTION
With the growing aesthetic demands of patients, composite resin usage has increased, leading to the development of various types tailored for different applications. For example, there are packable composite resins for posterior teeth with excellent physical properties such as strength, composite resins for anterior teeth requiring superior surface gloss and aesthetics, indirect restorative composite resins for extensive damage, and flowable composite resins designed for ease of use [1].
Flowable composite resins, introduced in the 1990s, aimed to enhance the convenience of application processes. However, due to the reduced physical properties caused by a lower content of inorganic filler compared to conventional composite resins, their use was limited [2,3]. To increase the flowability of composite resins, the content of the main viscous monomers, such as 2,2-bis[4-(2-hydroxy-3-methacryloxy-propyloxy)-phenyl] propane (Bis-GMA) or urethane dimethacrylate (UDMA) was reduced, and the amount of low-viscosity diluent monomers such as triethylene glycol dimethacrylate (TEGDMA) was increased. Additionally, the inorganic filler content was decreased, and larger particle fillers were used instead of finer particles. Consequently, early flowable composite resins were recommended for use in applications with minimal occlusal force or abrasion, such as Class V cavities or as liners [1,4].
Subsequent generations of flowable composite resins increased the inorganic filler content, making them suitable for posterior restorations [4]. However, the higher polymerization shrinkage compared to conventional composite resins necessitated careful application [5]. Recently, flowable bulk-fill composite resins, which simplify the filling process, have been commercialized. Bulk-fill composite resins can be polymerized in thicknesses over 4 mm in a single increment, and manufacturers claim they exhibit less polymerization shrinkage compared to traditional flowable composite resins [6,7].
Light-curable composite resins undergo polymerization through addition polymerization reactions involving carbon-carbon double bonds (C=C) in the dimethacrylate monomers. The relatively unstable and high-energy C=C bonds react readily with other molecules, generating a rise in temperature during the reaction [8]. The powerful visible light within the 400-500 nm wavelength range from light-curing units activates α-diketone in the light-curable composite resin, initiating free radical formation and polymerization [9]. In the past, quartz-tungsten-halogen lamps were primarily used as light sources, but due to issues such as high heat generation and short lifespan, light-emitting diode (LED) light sources are now more commonly used. LEDs emit blue light using junctions of differing properties within a gallium-nitride-based semiconductor. Previous studies have shown that LEDs are more efficient than halogens at converting energy to light and more closely match the absorption wavelength of camphorquinone, a common photoinitiator in light-curable composite resins [10]. Additionally, recent high-output LED light-curing units with outputs exceeding 1,000 mW/cm2 have been introduced, which cause significant temperature increase during polymerization due to their high power rather than the type of light source [11].
The temperature rise in composite resin during light curing is attributed to both exothermic polymerization and thermal energy from the light source [12–16]. Various studies have measured the temperature rise in composite resin and surrounding tissues using methods such as thermistors, thermocouples, differential scanning calorimetry, and differential thermal analysis [11,17–24]. However, these methods have limitations, such as measuring only a single point temperature and requiring direct contact with the surface [25]. In contrast, infrared thermography (IRT), using high-resolution infrared cameras, allows non-contact measurement of precise and sensitive temperature distributions by visualizing infrared radiation emitted from the sample. This method has been widely used in studies to visualize temperature rise during the polymerization process [16,25–28].
This study aimed to observe and compare the temperature dynamics of conventional flowable composite resins and flowable bulk-fill composite resins, which have reported differences in curing properties, using an IRT camera. Given that composite resin polymerization is an exothermic reaction, monitoring temperature changes through IRT provides indirect insight into polymerization dynamics and potentially the degree of conversion [16,29,30]. While this approach does not quantify conversion directly, it allows comparison of thermal characteristics among materials with differing composition and thickness. Because only a single specimen was tested per condition, this study did not involve statistical testing. Instead, observed trends in temperature change, time to peak temperature, and overall thermal dynamics were interpreted descriptively. The goal was not to test a formal hypothesis, but rather to identify material- and thickness-dependent trends that may inform future studies.
METHODS
Materials and equipment
In this study, the light-curing unit Dr’s Light (GoodDoctors Co., Seoul, Korea) was used. The Dr’s Light unit allows light output adjustments from 1% to 100% and curing time in 1-second increments up to 100 seconds. The LED curing unit utilized an 8 mm diameter guide tip, set to 80% output and an 80-second curing time. The 80% light output of the curing unit was measured using an LED-specific radiometer (GoodDoctors Co.), capable of measuring light output at 460 nm and 405 nm. To ensure consistency of light output, the irradiance from the LED unit was measured 10 times using the LED-specific radiometer, and the mean and standard deviation were calculated (Table 1). This repeated measurement was limited to verification of the light source and does not apply to the resin specimen experiments, which were conducted with a single specimen per condition. This corresponds to a high irradiance condition (approximately 1,786 mW/cm²), thus categorized as a high-intensity LED light source in this study. The composite resins used in this study were SDR (Dentsply Caulk, Milford, DE, USA), a flowable bulk-fill composite resin, and G-aenial Flo (GF; GC, Tokyo, Japan), a flowable composite resin in A2 shade (Table 2).
Light intensity of the LED light curing unit at 80% output
Materials, manufactures, and chemical composition of the matrix and filler
Infrared thermographic measurement equipment
The IRT camera used to measure surface temperature changes during polymerization in this study was the FLIR SC620 (FLIR Systems AB, Stockholm, Sweden) (Figure 1). IRT is a science that acquires and analyzes thermal information obtained from a non-contact thermal imaging device (camera). IRT measures the total emitted radiation energy, which includes energy absorbed by the object from external heat sources and radiated by the object itself, as well as energy reflected and transmitted by the object from external heat sources. This total emitted radiation energy is converted into thermal images based on infrared radiation wavelengths.
Measurement situation of infrared thermography.
Using the Stefan-Boltzmann Law, which relates temperature to wavelength, the infrared wavelengths detected by the infrared detector are expressed as a function of temperature, thereby visually representing temperature variations in the image. The total incident energy Wtotal received by the camera can be expressed as follows:
Wtotal = τϵWobj + τ (1 − ϵ) Wamb + (1 − τ) Watm
The first term, τϵWobj, represents the ambient radiative energy and the object’s reflected energy. The second term, τ (1 − ϵ) Wamb, represents the radiative energy of the object. The third term, (1 − τ) Watm, represents the atmospheric radiative energy. Here, τ denotes the transmittance, and ϵ denotes the emissivity (absorptivity). IRT inspection is not only used for detecting defects in equipment, devices, and components across all industrial sectors but also has a wide range of applications in the medical field and beyond [31]. The specifications of the infrared thermographic camera used in this study are provided in Table 3.
Specification of the infrared camera
Specimen preparation
Acrylic plates with thicknesses of 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, and 5.0 mm were cut into 40 × 40 mm squares. A 5.5 mm diameter hole was drilled in the center of each plate using a benchtop drill. To create a cavity structure with one closed side, one face of each acrylic plate was attached to a 40 × 40 mm acrylic plate with a thickness of 0.5 mm using adhesive. The cavities were then filled with flowable composite resin and high-flow bulk-fill resin. To achieve a uniform resin layer, the cavity was covered with a thin, transparent vinyl plate, and excess resin was removed by applying pressure with a glass plate. The acrylic cavity structure was used as a physical reference to control specimen thickness, and the combination of a transparent vinyl sheet and glass plate ensured a flat and uniform upper surface. Although the thickness was not directly measured after curing, the cavity design was intended to maintain consistent thickness across specimens. To improve the emissivity of the surface for accurate IRT measurements, the vinyl plate was coated with Krylon 4290 Ultra Flat Black paint (Krylon Industrial, Cleveland, OH, USA), which has an emissivity of 0.97, and allowed to dry thoroughly.
For temperature measurements, the prepared specimen was placed on a thermal insulation plate, which was made of 150 × 150 mm compressed styrofoam with a thickness of 5 mm. A 9.0 mm diameter hole in the center of the insulation plate allowed the resin-filled cavity to be exposed to the curing light source. To ensure proper alignment and reproducibility during the curing process, an “L”-shaped acrylic guide plate was attached to the edges of the specimen to secure its position. The LED light-curing unit was fixed to a rod stand and positioned perpendicular to the surface of the resin at a distance of 10 mm. To minimize external light interference and reflections, the thermal insulation plate and surrounding setup were coated with the same emissive paint. Although multiple specimens were fabricated for each resin and thickness, only one specimen per condition was used for measurement (n = 1).
Measurement of polymerization heat
The prepared specimen was placed at the designated position on the thermal insulation plate, and the device was fixed such that the tip of the curing unit was positioned 10 mm away from the bottom of the specimen. The IRT camera was mounted 100 mm above the specimen, and the measurement area was adjusted and verified using the camera’s imaging mode (Figure 2). The light-curing unit was set to 80% output and 80 seconds of curing time. Polymerization was initiated by turning on the light-curing unit, and simultaneously, the IRT camera began recording. The camera was set to capture 30 frames per second to collect data. All experiments were conducted in a light-blocked darkroom at room temperature. The temperatures measured by the IRT camera for each frame were converted using FLIR Tools (FLIR Systems AB). The time each frame was captured was converted to temperature per second for analysis.
Schematic diagram set-up for the infrared thermography camera measurement of the composite surface temperature. The infrared thermographic camera is located above the composite in a mold.
Data analysis
All temperatures within a 5.0 mm diameter circle centered at the highest temperature rise point in the center were measured. The maximum temperature at the highest temperature rise point over time was extracted using FLIR Tools. Subsequently, the maximum temperature, the time to reach maximum temperature, and the temperature rise pattern for each thickness of the two resins were analyzed. Since each condition was tested with only a single specimen (n = 1), statistical analysis was not performed. Observed differences were interpreted descriptively based on thermal trends.
RESULTS
The temperature data within the 5 mm diameter circle at the highest temperature rise point were recorded from the start to the end of polymerization (Table 4, Figure 3). The maximum temperature rise for GF decreased as the thickness increased, whereas for SDR, the temperature decreased up to 2.0 mm but remained relatively constant between 2.0 mm and 5.0 mm. For thicknesses of 1.5 mm or less, neither resin showed a peak temperature. At these thicknesses, both resins exhibited a rapid temperature increase within the first 5 seconds of polymerization, followed by a reduced rate of increase, and continued to rise steadily up to 80 seconds. For thicknesses of 2.0 mm or greater, the temperature rose rapidly after the start of polymerization, reaching a maximum peak, and then gradually decreased. At a thickness of 2.0 mm, the temperature initially increased, then decreased slightly after reaching an early peak, and increased again to reach the maximum temperature at 80 seconds. For thicknesses of 2.5 mm or greater, the peak temperature coincided with the maximum temperature. In GF, the peak temperature tended to decrease as the resin thickness increased, while SDR showed relatively stable peak temperature values across thicknesses. The time required to reach peak temperature increased as the resin thickness increased, in both materials, especially beyond 2.0 mm. The time to reach the peak temperature was 8.93 seconds for SDR and 16.58 seconds for GF at 2.0 mm thickness, and 12.93 seconds for SDR and 29.67 seconds for GF at 5.0 mm thickness. Across all tested thicknesses, SDR was observed to reach peak temperature more rapidly than GF.
Descriptive comparison of time to peak temperature, peak temperature, and temperature at 80 seconds for SDR and G-aenial Flo at each thickness
Temperature rise during light cure with a high intensity LED curing unit (Dr’s Light, 1,786 mW/cm² at 80% output) for different resin thickness. (A) SDR (bulk-fill composite resin), 0.5–5.0 mm. (B) SDR, 0.5–2.0 mm. (C) SDR, 2.5–5.0 mm. (D) G-aenial Flo (GF; flowable composite resi), 0.5–5.0 mm. (E) GF, 0.5–2.0 mm. (F) GF, 2.5–5.0 mm. Each plot represents a single recorded measurement for each condition (n = 1). LED, light-emitting diode. Dr’s Light: GoodDoctors Co., Seoul, Korea; SDR: Dentsply Caulk, Milford, DE, USA; G-aenial Flo: GC Co., Tokyo, Japan.
DISCUSSION
The introduction of flowable composite resins has improved direct restorative techniques by enabling better adaptation to cavity walls, especially in deep or irregular areas, compared to conventional packable resins [2,3]. The GF resin used in this study contains 68 wt% (44 vol%) of inorganic filler, making it suitable for larger posterior restorations, though its higher polymerization shrinkage warrants caution [4,5]. Bulk-fill composite resins are commercially available in both high-flow and conventional packable forms. Their primary characteristic is a deeper polymerization depth compared to conventional composite resins, allowing for depths of 4 mm or more in a single increment. However, the method of filling large quantities of resin at once must overcome issues that can arise from polymerization shrinkage. According to the manufacturer, the SDR used in this study includes a high molecular weight polymerization modulator in its matrix to reduce polymerization stress. This structure delays gelation, allowing stress relief during bulk application [6]. Although SDR reached peak temperature faster than GF, this does not contradict its delayed gel point, as the two reflect distinct aspects of polymerization governed by different formulation factors [5,6]. Jang et al. [5] further reported that SDR exhibits less polymerization shrinkage and lower polymerization shrinkage stress compared to conventional flowable composite resins, along with a deeper polymerization depth.
Both GF and SDR used in this study employ UDMA, which is less viscous than Bis-GMA, as their primary matrix. TEGDMA is also used. Therefore, it was expected that the temperature rise between the two materials using similar matrices would be similar, and the actual measured results showed similar maximum temperature values between the two materials (Table 4). In this study, the peak temperature rise was about 20°C. However, other experiments measuring the temperature rise of flowable composite resin using IRT cameras observed a temperature rise of about 40°C [14,28]. Notably, Chang et al. [28] used the same materials as this study. This discrepancy is likely due to uncontrolled environmental factors, especially surrounding light sources. In this study, surrounding light sources were completely blocked using blackout curtains, and emissive paint with an emissivity of 0.97 was applied, considering reflections from the material surface. Preliminary tests showed that emissive paint improved measurement accuracy, suggesting that the lower peak temperatures were due to the experimental setup, not the paint itself.
Based on these observations, the goal of this study was to determine whether IRT cameras could be used to understand the curing characteristics, such as the degree of conversion and polymerization depth of composite resins. The findings suggest that IRT has meaningful potential for evaluating resin polymerization behavior. Previous studies have demonstrated that differential thermal analysis can estimate the degree of conversion, with trends comparable to Fourier transform infrared spectroscopy (FTIR), especially in matrices with higher TEGDMA content [22,24,29]. Both resins have high TEGDMA content for flowability, and although their exact UDMA/TEGDMA ratios are unknown, intra-material comparisons by thickness using thermal data are still feasible. It should be noted that the thermal parameters observed in this study are not direct measures of the degree of conversion. However, the temperature rise patterns, particularly maximum temperature and time to peak, may serve as indirect indicators of polymerization kinetics and material behavior. These findings align with previous studies that have demonstrated the feasibility of using thermal analysis for estimating conversion trends [29].
In addition to compositional differences, the optical properties of the materials may also have contributed to the observed temperature dynamics. The higher translucency of SDR (universal shade) enhances light transmission and polymerization depth, making it advantageous for bulk-fill applications compared to more opaque materials like GF (A2 shade). The difference in translucency may have influenced the time to peak temperature and the rate of temperature change observed between the two materials in this study. These findings are consistent with previous studies indicating that the optimized translucency of SDR enhances light penetration and depth of cure [6]. Therefore, shade and translucency should be considered important variables in future investigations evaluating polymerization dynamics using thermal analysis.
Based on the temperature rise patterns observed in this study, it can be inferred that GF may exhibit more limited polymerization dynamics beyond 2.5 mm thickness, whereas SDR maintained more stable thermal characteristics up to 5.0 mm. These trends, including differences in time to peak temperature and thermal response, suggest that polymerization dynamics differ between the two materials, even without direct measurement of conversion. A common method for measuring polymerization depth involves comparing the microhardness at the surface and at a certain depth. Jang et al. [5] reported that SDR has a deeper polymerization depth than GF using this method. Analyzing the temperature rise patterns from this study shows similar trends. Although further studies are needed, temperature measurements may serve as a useful indirect tool for evaluating polymerization dynamics, particularly when comparing materials with different thickness-dependent thermal dynamics.
In interpreting the thermal data obtained in this study, it is important to acknowledge that temperature-based parameters such as peak temperature, time to peak, and post-peak slope serve as indirect indicators of polymerization dynamics rather than direct measures of the degree of conversion. The peak temperature primarily reflects the total exothermic heat released during the polymerization process. Under certain conditions, this may correlate with the extent of polymerization; however, it does not provide quantitative information about the actual monomer-to-polymer conversion ratio. The time to peak temperature is influenced by the rate of radical generation and propagation, and was used in this study to infer relative differences in polymerization kinetics between materials. In contrast, the slope of the temperature curve after the peak reflects the rate of heat dissipation and is influenced by thermal conductivity and environmental conditions rather than chemical reactivity. Furthermore, while a rapid rise in temperature and a short time to peak temperature may suggest a faster onset of polymerization, this does not necessarily imply that the overall conversion process has been completed. In bulk-fill composites like SDR, polymerization may continue beyond the temperature peak due to factors like light attenuation and delayed gelation. This concept is consistent with previous studies that demonstrated ongoing polymer network formation after the cessation of significant heat release [30]. Therefore, in the context of IRT evaluation, thermal parameters should be interpreted as supporting data for understanding polymerization trends, rather than definitive indicators of the degree of conversion. For a more accurate assessment of conversion, direct analytical methods such as FTIR or Raman spectroscopy should be employed in conjunction with thermal analysis [29].
The high-output LED curing unit (1,700 mW/cm²) likely contributed to temperature increases, especially in thinner specimens [11]. For specimens with a thickness of 1.5 mm or less, no distinct peak temperature was observed; instead, the temperature continued to increase steadily throughout the 80-second curing period. In thin specimens, the continuous temperature increase and early thermal spike observed during curing likely reflect superficial heating from the light source, rather than exothermic heat from polymerization. In contrast, specimens with a thickness of 2.0 mm or greater exhibited a typical polymerization thermal profile, characterized by a rapid temperature increase, a clear peak, and a subsequent gradual decline. Moreover, in both materials, a sharp temperature rise was observed within the first 5 seconds of light exposure, especially in thinner specimens. Therefore, when interpreting thermal curves, particularly in thin samples (≤1.5 mm) and during the early phase of irradiation (first 5 seconds), it is important to consider that the measured temperature may reflect light-induced heating rather than polymerization dynamics. From a clinical standpoint, temperature rise during light curing is a concern because of its potential to affect pulpal health. While our study did not directly measure intrapulpal temperatures, previous studies have indicated that increases above 5.5°C may lead to irreversible pulp damage [32]. In light of this, the temperature rises we observed, particularly at lower resin thicknesses, highlight the importance of using curing protocols that minimize heat exposure to the pulp. Therefore, further research on light output and temperature rise by thickness is needed, which could establish application standards for IRT cameras as useful research tools for understanding the curing characteristics of light-curing composite resins.
One limitation of this study is that the specimen thickness was not remeasured after polymerization. Although the acrylic cavity structure served to standardize the thickness during preparation, minor dimensional changes due to polymerization shrinkage may have occurred and were not quantified. To improve the accuracy of polymerization-related thermal analysis in future studies, experimental designs should account for potential dimensional changes caused by polymerization shrinkage.
Many previous studies on temperature rise measurements reported results from specific points. However, the studies by Chang et al. [28] and this study indicate that the same temperature is not measured across all illuminated areas. Thus, traditional temperature measurement methods might have yielded different results depending on the measurement point. IRT cameras can provide more accurate results by identifying the highest temperature in a specific area and confirming the overall temperature distribution. Additionally, advancements in technology allow for temperature comparisons at very short intervals.
SDR consistently showed a shorter time to reach peak temperature compared to GF, indicating a faster thermal response during polymerization. However, this rapid temperature increase did not necessarily correspond to higher overall exothermic output or greater polymerization shrinkage stress [5,30]. These findings suggest that polymerization kinetics and stress development depend on distinct material-specific properties [5,6,30].
CONCLUSIONS
This study demonstrated that conventional flowable and bulk-fill composite resins exhibit distinct thermal behaviors during light curing, particularly as resin thickness increases. Bulk-fill resins maintained a more consistent temperature profile and showed a faster thermal response at greater depths, supporting their use in posterior restorations that require deeper curing.
These results also suggest that IRT is a promising non-contact tool for indirectly evaluating polymerization dynamics. While the degree of conversion was not directly measured, the observed thermal patterns offer meaningful insight into material behavior. Further research involving repeated measurements and statistical validation is warranted to confirm these findings and to define clinically relevant thresholds.
Notes
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING/SUPPORT
This study was not supported by any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors declare no financial interest in the companies whose materials are included in this article.
AUTHOR CONTRIBUTIONS
Conceptualization: Hwang IN, Cho YT; Formal analysis: Bae KB, Noh EY, Cho YT; Investigation: Noh EY, Cho YT; Methodology: Hwang IN, Oh WM, Hwang YC; Visualization: Bae KB, Noh EY; Supervision: Lee BN, Chang HS, Hwang IN; Writing - original draft: Bae KB, Noh EY, Cho YT, Hwang IN; Writing - review & editing: Bae KB, Lee BN, Chang HS, Oh WM, Hwang YC, Hwang IN.
DATA SHARING STATEMENT
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request. Requests for access to the data should include a clear description of the intended use and the conditions under which reuse is permitted.