Effects of different intensities of the light-curing unit on the color change of bleach shade resin composites after immersion in coffee Authors: Zahra Ari Tabarestani, Ariana Rezaei Majd, Ghazaleh Ahmadi, Faraneh Mokhtarpour, Behnaz Esmaeili Published: 2026 ABSTRACT AbstractBackgroundThe popularity of white teeth among patients has led to the introduction of various restorative materials with bleach shades. However, color change in these materials remains a drawback to the long-term success of such restorations. The aim of this study was to compare the effects of different light-curing intensities on the color change of three Resin-Based Composite types in two shades after immersion in coffee.MethodsIn this in vitro study, 120 samples of three types of Resin-Based Composite were prepared, including Vit-l-escence in two shades (PN and OS), Gradia in two shades (A2 and XBW), and Filtek Z350 XT in two shades (A2 and XW). They were polymerized at intensities of 1000 mW/cm² and 1400 mW/cm² using a Valo light-curing device for 20 s. All specimens were incubated in distilled water at 37 °C for 24 h. After measuring the initial color, all specimens were immersed in a coffee solution at 50 °C for 7 days. The final color of the specimens was re-evaluated, and ΔE00 was calculated. Data were analyzed using One and Three-way ANOVA, and Post hoc Tukey HSD test in SPSS 26. P < 0.05 was considered statistically significant.ResultsResin-Based Composite type (P < 0.001), Resin-Based Composite shade (P < 0.001), and light-curing intensity (P < 0.001) had a significant effect on color change. The greatest color change was observed in Vit-l-escence (bleach shade, light-curing intensity of 1000 mW/cm2) and the lowest color change was in Gradia Direct (normal shade, light-curing intensity of 1400 mW/cm2). The higher light-curing intensity (1400 mW/cm2) significantly reduced color change in Gradia for both bleach and normal shades, as well as in Vit-l-escence for both bleach and normal shades.ConclusionsGradia Direct and Vit-l-escence showed less significant color change at a light-curing intensity of 1400 mW/cm² compared to 1000 mW/cm²; however, Filtek Z350 did not exhibit this result. The color change of bleach shade Resin-Based Composites was also greater than that of the normal shades. Thus, light-curing intensity may influence the color stability of Resin-Based Composites, and this effect likely depends on multiple factors, such as Resin-Based Composite shade and type.Keywords: Composite resins; Coffee; Curing lights, dental; Pigmentation Background Nowadays, having a beautiful smile has become an essential need among patients [1]. Additionally, most patients prefer white and shiny teeth with a natural appearance [2]. This preference has increased the popularity of restorative materials with bleach shades. These materials require multiple in vitro and clinical investigations to evaluate characteristics like color stability, translucency and strength [3]. RBCs (Resin-Based Composites) are extensively used in restorative dentistry. Their components include an organic polymer matrix, inorganic filler particles, a coupling agent, and a photoinitiating system. RBCs offer several advantages such as bonding to dentin and enamel, color similarity to tooth structures, and desirable physical properties. They are also less expensive than indirect treatments [4]. Although the quality of RBCs has improved in recent years, color change remains a barrier to the long-term success of these restorations, often making replacement necessary [5–7]. A previous study represented comparable performance between single-color RBCs and multi-color RBCs in terms of color stability over 12 months [8]. The color change of RBCs depends on several factors of internal or external origin. The internal factors are related to the structure of the restorative material, such as the physical and chemical properties of the matrix, the amount, size, and distribution of filler particles, the rate of water absorption, and the type of photoinitiator [9, 10]. Camphorquinone is the most widely used photoinitiator in RBCs. It is a dense mixture of yellow diketone. Therefore, large amounts of camphorquinone in the resin formulation result in unwanted yellowing of RBCs. To reduce this yellowish color in bleach RBCs, other photoinitiators such as diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide (TPO), bis-acryl phosphine oxide (BAPO), and 1-phenyl-2,1-propanedione (PPD) are used [11, 12]. On the other hand, external factors are associated with the absorption of pigments from external sources, which are related to oral hygiene, smoking, and nutrition [9, 10]. External color changes depend on several factors, including pigment type, surface roughness, duration of contact or immersion in the staining medium, and the type of RBC used. Previous studies on the color stability of RBCs have shown that beverages such as coffee, tea, soft drinks, and mouthwashes can change the color of RBCs to varying degrees [11, 13]. A study using values obtained from RBCs and color guide samples in the CIE L*a*b* color scale indicated a CIEDE2000 color difference between polymerized RBCs and color guides [14]. To reduce color change, the polymerization of RBC should also be improved. Several factors affect the degree of polymerization of RBCs, including those related to the light-curing unit, such as emission spectrum, intensity, distance, and direction of exposure [15, 16]. The light-curing intensity of the curing unit decreases from the surface to the depth of the restoration due to absorption and dispersion. There is no consensus on the optimum intensity of the light-curing unit for adequate curing of different RBCs [12, 17]. Teimourian et al. [18] examined the effect of light-curing intensity on the color stability of RBC with shade A2. They concluded that increasing light-curing intensity had no effect on the color stability of the RBC. Moreover, Farah et al. [2] investigated the color change of bleached RBCs in coffee and tea. They found that the color change in coffee was significantly greater than in tea. However, they did not investigate the effect of different light-curing intensities on the color change of the RBC. Since there is insufficient information on the color stability of bleach shade RBCs, the present study was conducted to compare the effects of different light-curing intensities on the color change of three RBC types in two shades after immersion in coffee. Null hypotheses were: RBC type does not affect color change, RBC shade does not affect color change and light-curing intensity does not influence color stability. Methods The study design and experimental procedure are shown in Fig. 1. In this in vitro study, color changes of three types of RBCs, including bleach-shade and normal shade, polymerized with 1000 mW/cm2 and 1400 mW/cm2 light-curing intensity, were evaluated after storage in a coffee solution. Diagram of the study design. This study investigated three types of RBCs, with two shades for each RBC, and two light-curing intensities, with 10 specimens per subgroup (n = 10). The mold was placed on a flat surface, and the mold hole was filled with each RBC type and shade in six groups. Each RBC group was randomly assigned to two subgroups for curing at two different light-curing intensities: 1000 mW/cm2 and 1400 mW/cm2. The tip of the light-curing device was in contact with a 1-mm-thick glass slide during polymerization. After exposure, specimens with a diameter of 7 mm and a thickness of 2 mm were removed from the mold. The upper surface of each specimen was polished with silicon carbide papers for 10 s, then incubated in distilled water at 37 °C for 24 h to ensure complete polymerization. The initial color was assessed. The coffee solution was prepared by mixing coffee powder in boiling water and filtring it. All specimens were immersed in the coffee solution and incubated at 50 °C for 7 days. They were then rinsed, dried, and the final color was assessed. The color change was calculated using the CIEDE2000 formula Sample size calculation The required sample size was determined using G*Power software (version 3.1.9.4; Heinrich Heine University, Düsseldorf, Germany), based on an effect size of 0.40, a significance level (α) of 0.05, and a statistical power (1–β) of 90%. This effect size corresponded to a moderate level of difference and was selected in accordance with values reported in a previous study [19]. This experimental study investigated three RBC types, two shades for each RBC, and two light-curing intensities, with 10 specimens per subgroup (n = 10). Vit-l-escence RBC (Ultradent, USA) was used in two enamel colors, PN and OS; Gradia RBC (Direct GC, Japan) in two colors, A2 and XBW; and Filtek Z350 XT RBC (3 M, ESPE, USA) in two colors, A2 and XW. The RBCs are presented in Table 1. These were cured with light-curing intensities of 1000 mW/cm2 and 1400 mW/cm2. A total of 120 samples were prepared: Table 1The characteristics of the used materialsMaterialCompositionTypeFiller Content (%w/w)ManufacturerLot NumberGradiaDirectUDMA,Dimethacrylate,Camphorquinone,Fluoroaluminosilicate,Glass silica powderMicrofillRBC73GC Corp,Tokyo JapanXBW: 1,909,091A2:1,807,071Filtek Z350Bis-GMA, Bis-EMA,UDMA, TEGDMA, CQ (468 nm) Silica (20 nm) Zirconia andsilica (0.6–1.4 μm)NanofillRBC78.53 M ESPE, StPaul, MN, USAXW: NA09459A2:NA14171Vit-l-escenceBis-GMA, Silica particles(0.7 μm)MicrofillRBC75Ultradent, USAPN: BX8HZOS:BM6Y5 The characteristics of the used materials Gradia Direct UDMA, Dimethacrylate, Camphorquinone, Fluoroaluminosilicate, Glass silica powder Microfill RBC GC Corp, Tokyo Japan XBW: 1,909,091 A2: 1,807,071 Bis-GMA, Bis-EMA, UDMA, TEGDMA, CQ (468 nm) Silica (20 nm) Zirconia and silica (0.6–1.4 μm) Nanofill RBC 3 M ESPE, St Paul, MN, USA XW: NA09459 A2: NA14171 Bis-GMA, Silica particles (0.7 μm) Microfill RBC PN: BX8HZ OS: BM6Y5 Preparation of specimens The mold was placed on a flat surface, and the mold hole was filled with each RBC type and shade in six groups (3 RBC types × 2 shades). The top surface was covered with a Mylar strip and pressed with a 1-mm-thick glass slide to obtain a smooth surface. Each RBC group that filled the mold was then randomly assigned to two subgroups for curing by a light-curing unit (Valo, Ultradent, USA): The first subgroups were polymerized with a standard intensity of 1000 mW/cm2 for 20 s (20 J/cm2); the second subgroups were polymerized with a high intensity of 1400 mW/cm2 for 20 s (28 J/cm2). The tip of the light-curing device was in contact with a 1-mm-thick glass slide during polymerization. The light-curing intensity was controlled using a radiometer (Kerr Corporation, Middleton, USA) at the beginning and after every 10 exposures. After completing the exposure, the specimens with a diameter of 7 mm and a thickness of 2 mm were removed from the mold, and the lower surface of the specimens was marked. Then, the upper surface of each specimen was polished using 600, 800, 1200, 2000 silicon carbide papers (Matador, Germany) for 10 s under running water to make the surface uniform and remove resin-rich surface. The silicon carbide papers were replaced after 5 specimen polishing. All specimens were polished by the same operator to avoid variability. All specimens were incubated (LTE Scientific LTD, Oldham, Great Britain) in distilled water at 37 °C for 24 h ensuring complete polymerization. Staining process Ground coffee (Starbucks Blonde Espresso whole bean dark roast) was used in this stage, 3.6 g of the powder was measured and mixed in 300 ml of boiling water. Then, the solution was filtered with filtering paper [7, 20]. The specimens were immersed in the coffee solution at 50 °C in an incubator (LTE Scientific LTD, Oldham, Great Britain) that this protocol represents an accelerated staining model rather than a direct simulation of clinical conditions [2]. Moreover, storage immersion was for 7 days and the solution was renewed daily. According to Guler et al. [21], the mean time to consume 1 cup of coffee is 15–20 min and among coffee drinkers, the mean coffee consumption is 3.2 cups per day; Therefore, in the present study, 7 days of immersion in coffee is equivalent to 6 months of coffee consumption. After 7 days of storage, the specimens were rinsed with distilled water and dried before color measurement. Color measurement and ΔE00 calculation The color measurement was performed using the Vita Easyshade (Vita Easyshade Advance, Germany) in the CIE L*a*b system on a white background, and the color difference (ΔE00) was calculated for specimens using the CIEDE2000 formula. Vita Easyshade is a contact-type spectrophotometer with D65 illumination and geometric 2°/0°. According to the CIE L*a*b system, the “L” index indicates the lightness and “a” and “b” indexes indicate the chroma for all the specimens. The spectrophotometer was calibrated before each measurement. Color measurements were done by the same operator. The probe tip of the Vita Easyshade was held at 90° over the RBC specimen. The initial and final color of each specimen was measured three times and the average of them was recorded. Initial color was measured 24 h after polymerization (L1, a1, b1) and final color was measured after 7 days of coffee immersion (L2, a2, b2). Color difference (ΔE00) was calculated for all specimens using the following CIEDE2000 formula [22]: ΔL’, ΔC’, ΔH’ in the formula refer to the lightness, chroma and hue of the specimens. The KL, KC, KH parametric factors were used as computational correction terms for experimental monitoring conditions. RT is a rotation function and deciphers the interaction between saturation and tone differences in the blue region. SL, SC, SH are weighting functions and make the total color difference adjustment for changes in the location of color difference pairs in the L*, a*, b* coordinates. In the current, the parametric factors of the CIEDE2000 formula were set to 1. According to the CIEDE2000 formula, ∆E00 * < 0.8 were considered as ‘undetectable’ and 0.8 < ∆E00 * < 1.8 were evaluated as clinically ‘acceptable’. Values above 1.8 were considered clinically unacceptable [22]. Statistical analysis The distribution of data was normal (Kolmogorov-Smirnova ). One and Three-way ANOVA and Post hoc Tukey HSD test were used in SPSS 26 (IBM, New York, USA) Software. The P < 0.05 was considered statistically significant. Results Table 2 presents ΔE00 of the different RBCs according to type, color and light-curing intensity. The RBC type (P < 0.001), the RBC shade (P < 0.001), and the intensity of the light-curing (P < 0.001) significantly affected the color change of the RBCs. At equal light-curing intensity, the RBC shade and type significantly affected the color change (P = 0.002). The ΔE00 of different RBCs at light-curing intensities of 1000 mW/cm2 and 1400 mW/cm2, according to type and shade, are presented in Figs. 2 and 3. Table 2ΔE00 (Mean ± SD) of RBCs according to type, shade and light-curing intensityRBCShadeLight-curing intensity (mW/cm2)P-value*10001400GradiaBleach (XBW)12.51 ± 0.8611.33 ± 0.930.047Normal (A2)8.46 ± 0.866.40 ± 0.720.000Vit-l-escenceBleach (OS)12.89 ± 0.5210.11 ± 0.960.000Normal (PN)8.93 ± 0.857.24 ± 0.930.000Z350 XT FiltekBleach (XW)10.85 ± 0.6210.28 ± 0.410.892Normal (A2)7.77 ± 0.896.72 ± 0.610.115*Post hoc Tukey HSD test ΔE00 (Mean ± SD) of RBCs according to type, shade and light-curing intensity *Post hoc Tukey HSD test ΔE00 of RBCs at light-curing intensity of 1000 mW/cm2 ΔE00 of RBCs at light-curing intensity of 1400 mW/cm2 Effect of RBC type on ΔE00 Regardless of the RBC shade and light-curing intensity, different RBC types suggested a significant ΔE00 (P < 0.001). The most ΔE00 was in Vit-l-escence (bleach shade, intensity of 1000 mW/cm2) and the lowest ΔE00 was in Gradia Direct (normal shade, intensity of 1400 mW/cm2). The ΔE00 was clinically unacceptable in all three types. The color change for all RBC types was above the clinically acceptable range (ΔE00> 1.8). Effect of RBC shade on ΔE00 At the same light-curing intensity and RBC type, the color change of different RBC shades was significant, and the bleach shade than the normal shade exhibited a higher ΔE00 (P < 0.05). Effect of light-curing intensity on ΔE00 All specimens were affected by light-curing intensity, except for RBC Z350 bleach shade (XW) and normal shade (A2). Higher light-curing intensity (1400 mW/cm2) effectively reduced color change in Gradia and Vit-l-escence. Three-way ANOVA results for ΔE00 are presented in Table 3. Table 3Three-way ANOVA results of ΔE00SourceType III Sum of SquaresdfMean SquareFP-value*Corrected Model535.4831148.68079.5810.000Intercept10732.560110732.56017545.3150.000RBC type18.55629.27815.1680.000RBC shade419.9271419.927686.4860.000Light-curing intensity72.852172.852119.0970.000RBC type * RBC shade8.42724.2146.8880.002RBC type * Light-curing intensity10.18425.0928.3240.000RBC shade * Light-curing intensity0.07110.0710.1160.734RBC type * RBC shade * Light-curing intensity5.46522.7324.4670.014Error66.0641080.612Total11334.107120Corrected total601.547119*Three-way ANOVA Three-way ANOVA results of ΔE00 *Three-way ANOVA Discussion The present study compared the effects of different light-curing intensities on the color change of three RBC types in two shades after immersion in coffee. All null hypotheses were rejected: RBC type, RBC shade, and light-curing intensity significantly affected the color change. The results indicated that higher light-curing intensities significantly affected the color stability of Gradia Direct and Vit-l-escence; however, Filtek Z350 showed no such effect. When examining the color change of bleach shade RBCs at different light-curing intensities, it was found that the color changes of Gradia Direct and Vit-l-escence RBCs depended on the light-curing intensity, with color change at 1400 mW/cm2 being significantly lower than at 1000 mW/cm2. This may result from higher polymerization rates of these RBC types at increased light-curing intensity. However, no significant difference in color change was observed for Filtek Z350 at different intensities. This indicates that at a 20-second light-curing time, a light-curing intensity of 1000 mW/cm2 was sufficient for Z350 polymerization. The results of the current study were consistent with those of Badir et al. [23] regarding mono RBC shade discoloration. Their study revealed that power density (J/cm2) is more important than light-curing intensity (mW/cm2) for the polymerization of RBCs. It was recommended that the light energy transferred to the RBC should be about 21–24 J/cm2 for adequate polymerization [24]. In their study, a light energy of 1400 mW/cm2 intensity was higher than 1000 mW/cm2, which can lead to increased color stability. On the other hand, previous studies were inconsistent with our findings and reported that light-curing intensity is not an effective factor in color stability [18, 25]. It seems that power density should be considered an effective factor in color stability instead of light-curing intensity and light exposure time separately. In the present study, bleach shade RBCs showed more staining than normal shade RBCs, which is consistent with other studies [2, 26, 27] reporting that the color change of RBCs depends on their color and opacity, with lighter materials discoloring more than darker ones. Vichi et al. [28] examined the effect of color and opacity on the color change of three different RBCs. Their results revealed that brighter RBCs (with higher L values) showed more color change, which was consistent with the results of the present study. Sakaguchi et al. [29] suggested that increasing opacity reduces light-curing penetration in RBC. The white pigments used in bleach shade RBCs increase the opacity of the RBC. Moreover, the white color of these RBCs causes the dispersion of light. Thus, the greater color change of bleach shade RBCs may be due to reduced light penetration in these RBCs compared to normal shades [30]. Therefore, a longer exposure time is recommended for white and opaque RBCs [31]. The main difference between bleach-shaded RBC and other shades is the type of photoinitiator used. Unlike regular shades of RBCs that use type II photoinitiators (CQ), bleach-shaded RBCs use type I white and colorless photoinitiators such as lucirin TPO and phenylpropanedione, either alone or in combination with CQ, to avoid the photo-yellowing effect of the RBC [26, 32]. These photoinitiators absorb shorter wavelength violet light (approximately 350–420 nm) [31]. Effective polymerization of each photoinitiator system depends on using a compatible light-curing unit with an appropriate emission spectrum [31]. Single-emission-peak LED units (first and second generation) have usually been designed for CQ activation (460–480 nm). To address this challenge, multiple-emission-peak LED units (third generation) have been introduced, such as blue-phase G2 and Valo, which emit multiple peaks and a broad wavelength range (395–480 nm) to ensure simultaneous activation of both initiator types [31]. Due to lack of access to the details of the RBC formula, a third generation LED device with multiple emission peaks, Valo, was used in the current study. According to the results of the ongoing study, the effect of RBC type on color change was significant. The color change of Vit-l-escence RBC compared to Filtek Z350 XT RBC was significantly higher in the bleach shade at an exposure intensity of 1000 mW/cm2. However, at an exposure intensity of 1400 mW/cm2, the color change of Vit-l-escence and Filtek Z350 in the bleach shade was almost similar. This result is inconsistent with other studies that found nanoresin composite had much higher discoloration than microhybrid composite [11, 33]. This may be due to different light-curing protocols. In the previous study by Hashemikamangar et al. [33], staining of the bleach shade of Gradia Direct and Filtek Z350 using a light-curing intensity of 1000–1100 mW/cm2 was evaluated. Specimens of each RBC type were immersed separately in sour cherry juice and coffee. Filtek Z350 exhibited more discoloration than Gradia Direct, which disagrees with the present study. This discrepancy may result from different polymerization rates caused by varying light-curing times [34]. Several factors can influence the extent of color change in RBCs. These factors include the composition and structure of the RBC, the properties of the filler particles, the degree of polymerization achieved during curing, and the amount of water absorbed by the material [7]. RBCs with larger filler particles have more color change due to the effect on water absorption [28, 35], which is consistent with our findings. It occurred due to hydrolytic reactions at the interface between the filler and matrix. Dietschi et al. [36] also reported that the color change of composite is related to the amount of water absorption. Water carries pigments that penetrate the resin matrix. Increasing the amount of filler reduces polymerization shrinkage, thermal expansion, and water absorption. Therefore, a higher amount of resin matrix increases water absorption. Additionally, several studies have shown that composites with a lower weight% of filler exhibit more color changes [14, 37, 38]. The lower color stability of Gradia Direct RBC in their study may be due to its lower weight% of filler compared to the other RBCs studied. Moreover, ALShetili et al. [39] examined the effect of RBC polishing on the surface roughness of two RBCs, Filtek Z350 (nanofilled) and Filtek Z250 (microhybrid). They observed that the microhybrid RBC has higher surface roughness than the nanohybrid RBC. The surface roughness of RBCs can affect their color change [40]. In this study, the effects of surface roughness, surface finishing and polishing, different light-curing exposure times, multiple staining solutions, translucency, thermocycling protocols, and degree of conversion were not evaluated and could be considered in future research. Using hand pressure during specimen preparation and conducting an in vitro study were also limitations of this study. Furthermore, the effects of different staining beverages, the effects of immersion at different temperatures, additional surface finishing and polishing methods, and saliva’s buffering effect should be investigated further. Conclusions The color change of the bleach shade RBCs was greater than that of the normal shade RBCs. Gradia Direct in both bleach and normal shades, and Vit-l-escence in both bleach and normal shades, showed less significant color change at a light-curing intensity of 1400 mW/cm² compared to 1000 mW/cm²; however, this result was not observed for Filtek Z350 in either bleach or normal shade. Thus, light-curing intensity may influence the color stability of RBCs, and this effect likely depends on multiple factors, such as the light-curing intensity evaluated and the material tested. Acknowledgements The authors would like to thank Dental Materials Research Center and Research Council of Babol University of Medical Sciences. Abbreviations Color difference Resin based composite Analysis of variance Diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide Bis-acryl phosphine oxide 1-phenyl-2,1-propanedione Urethane dimethacrylate Bisphenol A-glycidyl methacrylate Bisphenol A-ethoxylated dimethacrylate Triethylene glycol dimethacrylate Camphorquinone Light-emitting diode Standard deviations 24 h 7 days Authors’ contributions Zahra Ari Tabarestani developed the original idea and protocol, summarized the study, and collected the data. Ghazaleh Ahmadi collaborated in data collection, and also the data were analyzed by Ghazaleh Ahmadi and Faraneh Mokhtarpour. Ariana Rezaei Majd and Behnaz Esmaeili drafted the manuscript and edited the article. The Study was supervised by Behnaz Esmaeili. Funding Babol University of Medical Sciences. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The current study was approved by the Research Ethics Committees of Babol University of Medical Sciences (Ethics approval code: IR.MUBABOL.REC.1400.119). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.