Impact of polyphenol-rich beverages on color stability, translucency and elemental surface composition of composite resins: correlation analysis after staining

Abstract

Background

The aim of this study was to conduct a multi-parametric assessment of optical degradation and surface chemical changes in composite resins following exposure to polyphenol-rich beverages, by integrating spectrophotometric measurements with SEM–EDS analysis to evaluate potential associations between elemental alterations and esthetic outcomes.

Methods

128 disc-shaped (2 mm × 8 mm) composite resin specimens divided into two main groups according to the composite resin used: Supra nano hybrid composite resin (Estelite Palfique Paste) and Microhybrid composite resin (Estelite Posterior). After four-step polishing with Sof-Lex discs, the specimens were weighed, their colors were recorded using a spectrophotometer according to the CIEDE2000 system. Specimens were then immersed for 14 days in black, rooibos, or peppermint teas for 2 h/day to simulate clinically relevant daily tea consumption habits. During the remaining time, specimens were stored in distilled water at 37 °C in an incubator. The tea solutions were prepared using commercial tea bags. Each tea bag (approximately 2 g) was immersed in 200 mL of freshly boiled distilled water and allowed to steep for 5 min. After removal of the tea bags, the solutions were cooled to a drinkable temperature (60 ± 2 °C) and used for the immersion procedures. Total polyphenol content (TPC) of the teas was determined using the Folin–Ciocalteu method. Color change (ΔE₀₀), translucency parameter (TP₀₀), and weight loss were measured at baseline and after post-staining (n = 10). SEM–EDS analysis (n = 3) was performed to assess surface morphology and elemental composition. Normality was assessed using Shapiro–Wilk, and parametric or non-parametric tests were applied accordingly. Both parametric (ANOVA) and non-parametric (Kruskal–Wallis) tests were used according to data distribution. Correlations were performed using Spearman’s rank correlation coefficient. The level of statistical significance was set at p < 0.05.

Results

TPC differed significantly among teas (p < 0.05). Black tea caused the greatest ΔE₀₀ and ΔP₀₀ change in both composite resins (p < 0.05). SEM–EDS analysis revealed tea-associated changes in surface elemental signals, including increased carbon and reduced silicon levels. ΔE₀₀ was positively correlated with total polyphenol content (ρ = 0.614) and surface carbon percentage (ρ = 0.660), and negatively correlated with silicon percentage (ρ = − 0.503) (all p < 0.001). ΔP₀₀ was positively correlated with silicon and negatively correlated with carbon percentage (p < 0.001).

Conclusion

Composite resins showed different susceptibility to extrinsic staining depending on tea type, with black tea causing the greatest changes in color and translucency properties. Optical degradation was accompanied by alterations in surface elemental composition, particularly involving carbon and silicon.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12903-026-08954-8.

Introduction

Composite resins are widely used in restorative dentistry due to their favorable optical properties, adhesive capability, and minimally invasive application techniques. Contemporary nanohybrid and supranano-filled composites are designed to mimic the natural appearance of tooth structure while providing adequate mechanical performance. However, the long-term clinical success of these materials is closely related to the stability of their optical properties, particularly color [1]. Despite advances in resin technology, discoloration remains a common clinical problem.

Extrinsic discoloration of composite resins is strongly influenced by dietary chromogens. Polyphenol-rich beverages such as tea and coffee contain tannins and phenolic compounds with high chromogenic potential that can interact with the organic matrix of resin-based materials. These compounds may adsorb onto the surface or diffuse into the resin matrix through absorption mechanisms, leading to progressive discoloration after repeated exposure [2]. Commonly consumed beverages such as black tea, rooibos tea, and peppermint tea contain pigments capable of interacting with composite resin surfaces and may therefore contribute to clinically perceptible esthetic deterioration [3]. Black tea is among the most widely consumed beverages globally and contains high concentrations of condensed tannins, theaflavins, and thearubigins with well-documented chromogenic potential [4]. Rooibos tea, increasingly popular for its antioxidant properties and caffeine-free composition, contains aspalathin and nothofagin as its principal polyphenolic compounds [5]. Peppermint tea, widely preferred as a herbal alternative, is characterized by rosmarinic acid, luteolin, and apigenin derivatives with comparatively lower but clinically relevant polyphenol content [6]. These three beverages represent a broad spectrum of polyphenol profiles and were selected based on their widespread daily consumption patterns and distinct chromogenic potential.

The interaction between composite resins and staining beverages is also associated with water sorption and solubility phenomena. Water uptake facilitates the diffusion of chromogenic molecules into the polymer network and may promote hydrolytic degradation of the resin matrix and filler–matrix interface [7]. As a result, changes in material mass may occur, and weight measurements are commonly used as indirect indicators of solubility and chemical stability in aqueous environments [7, 8].

Optical changes in composite resins are commonly evaluated using the CIEDE2000 (ΔE₀₀) color difference formula, which provides improved agreement with visual perception compared with the conventional CIELAB system [9, 10]. In addition to color, translucency is a key optical parameter contributing to the natural appearance of restorations by regulating light transmission and internal scattering within the material [11, 12]. Changes in translucency may occur due to alterations in resin matrix composition, filler characteristics, or degradation at the filler–matrix interface [13, 14].

Although numerous studies have investigated staining of composite resins, most have focused primarily on optical outcomes without evaluating possible surface chemical alterations [2, 9–12]. Surface analytical techniques such as scanning electron microscopy combined with energy-dispersive X-ray spectroscopy (SEM–EDS) enable evaluation of elemental changes involving carbon (C), silicon (Si), oxygen (O), and zirconium (Zr), which may reflect adsorption of organic chromogens or alterations at the filler–matrix interface [15–17]. While variations in C, O, and Si have been reported previously [15–17], zirconium is typically considered an inert filler component and its role in staining-related surface changes remains unclear.

Despite the growing number of studies on composite resin discoloration, this multi-parametric approach allows direct correlation between esthetic deterioration and surface elemental alterations, providing a more comprehensive understanding of material degradation than optical assessments alone. Furthermore, quantifying total phenolic content of the test beverages enables evidence-based interpretation of staining potential in relation to polyphenol composition.

Therefore, the aim of this study was to perform a multi-parametric evaluation of optical degradation and surface chemical changes in composite resins after exposure to polyphenol-rich beverages. By integrating spectrophotometric measurements with SEM–EDS surface analysis, the study examined possible associations between elemental alterations and esthetic deterioration. The study hypotheses were as follows:

• H₀: Tea type has no significant effect on color change (ΔE₀₀ < 0.8), translucency parameter (ΔP₀₀), or weight loss of composite resins.

• H₀₁: Composite resin type has no significant effect on color change (ΔE₀₀), translucency parameter (ΔP₀₀), or weight loss.

• H₀₂: Tea exposure does not significantly alter the surface elemental composition (carbon, silicon, oxygen, and zirconium weight percentages) of composite resins.

• H₀₃: There is no significant correlation (Spearman's ρ = 0) between total phenolic content, surface elemental composition, and optical properties (ΔE₀₀ and ΔP₀₀).

Material and method

Two A2 shade composite resin materials were investigated: a supra nanohybrid composite resin (Estelite Palfique Paste) and a microhybrid composite resin (Estelite Posterior). The properties, lot numbers and main composition of the composite resins and the phenolic characteristics of three polyphenol-rich teas are presented in Table 1. All the procedures are illustrated in Fig. 1. The sample size of n = 10 per group was determined based on previously published similar in vitro composite resin staining studies [18–20]. An a priori power analysis was performed using G*Power software (version 3.1, Heinrich Heine University, Düsseldorf, Germany). Based on a medium effect size (f = 0.25), an α error probability of 0.05, a power of 0.95, and 8 groups with 3 measurements, the required total sample size was calculated as 80 specimens, corresponding to n = 10 specimens per group. The current sample size meets this requirement. For SEM–EDS analysis, three specimens per group were selected, which is consistent with previously reported protocols for surface morphology and elemental composition assessment in dental materials research [15–17].

Table 1.

Compositional characteristics and lot numbers of the composite resin materials used in the study

Material	Main compositional characteristics
Estelite Palfique Paste (Tokuyama Dental Corp., Tokyo, Japan). (Lot no: E7101)	Supra nano hybrid composite resin, Bis-GMA, TEGDMA, submicron spherical zirconia fillers, silica, particle size: 0.1–0.3 μm, 82 wt%, 71 vol%
Estelite Posterior (Tokuyama Dental Corp., Tokyo, Japan) (Lot No: W1732)	Microhybrid composite resin, Bis-GMA, TEGDMA, Bis-MPEPP, spherical silica–zirconia fillers, particle size: 0.1—10 µm, 84 wt%, 70 vol%
Black Tea (Lipton Yellow Label Tea, Unilever, Istanbul, Türkiye)	Tannins, theaflavins, thearubigins [4]
Rooibos Tea (Twinings Pure Rooibos Tea, Twinings & Co., Andover, Hampshire, United Kingdom)	Aspalathin, nothofagin, tannin [5]
Peppermint Tea (Twinings Pure Peppermint Tea, Twinings & Co., Andover, Hampshire, United Kingdom)	Rosmarinic acid, luteolin, apigenin derivatives [6]

Fig. 1.

Schematic illustration of the methodology of specimen preparation

Specimen preparation

A total of 128 disc-shaped specimens (8 mm diameter, 2 mm thickness) were fabricated using plastic molds according to the manufacturers’ instructions. Mylar strips and glass slides were positioned on the upper and lower surfaces of the molds during specimen preparation. Light polymerization was performed with a LED curing unit (Valo®, Ultradent Products Inc., USA) at 1000 mW/cm² for 20 s. The light intensity output was verified using a radiometer after every ten specimens.

After polymerization, all specimens were finished and polished using a four-step Sof-Lex™ disc system (3 M ESPE, St. Paul, MN, USA) under standardized conditions. Polishing was performed under dry conditions with a low-speed handpiece operating at 10,000 rpm, applying each disc in a unidirectional motion for 15 s. Each disc was used only once for all specimens to ensure consistency. After each polishing step, the specimens were rinsed with a water spray for 10 s and air-dried for 5 s. To minimize operator variability, all polishing procedures were performed by the same operator (HDK).

Specimens were stored at 37 °C under 100% humidity in an incubator for 24 h and subsequently in distilled water at 37 °C for 6 days to allow post-polymerization stabilization [21]. For the color evaluation, the specimens were randomly distributed as n = 10 per group for each composite resin across three beverages and a control group stored in distilled water. For SEM–EDS analysis, 3 specimens/group for each composite resin at two time points: baseline and after staining session were randomly prepared. A separate set of three specimens per group was assigned for baseline and final-stage SEM–EDS analysis.

Beverage preparation and immersion protocol

All samples were coded and randomly assigned to experimental groups using a computer-generated randomization sequence (Excel, Microsoft Corp., Redmond, WA, USA). Equal allocation was ensured for each subgroup: distilled water (control), black tea, rooibos tea, and peppermint tea. Specimens were immersed daily for 2 h over a 14-day period in freshly prepared tea solutions [22]. The immerison protocol was selected to simulate the cumulative daily oral exposure time associated with habitual tea consumption. Assuming an average tea consumption of three to four cups per day, each lasting approximately 15–20 min, the 2-h daily immersion approximates realistic intraoral contact time. The 14-day immersion period was selected to represent an accelerated aging model simulating several months of clinical use [22]. During the non-immersion periods, all specimens were stored in distilled water at 37 °C in an incubator.

Beverage pH was measured daily using a digital pH meter (Mettler Toledo digital pH meter, Greifensee, Switzerland) at room temperature, and immersion was performed at 60 ± 2 °C [23]. After staining protocols, the specimens were repolished using the same four-step Sof-Lex™ polishing protocol under standardized conditions to remove superficial extrinsic stains. Following the 14-day immersion protocol, weight loss measurements were performed. Subsequently, specimens were repolished using the same four-step Sof-Lex™ polishing protocol to remove extrinsic surface staining, simulating clinical polishing procedures. Final color measurements (T1) were then performed after repolishing to assess intrinsic color changes that could not be removed by polishing.

Total phenolic content analysis

Total polyphenol content (TPC) of the teas was determined using the Folin–Ciocalteu method with Ultraviolet–Visible (UV–Vis) spectrophotometry (765 nm) [24]. Fourteen independent brews were prepared and three technical replicates were measured from each brew (42 readings per tea). Gallic acid served as the calibration standard (≥ 6 points; 0–500 mg/L), and TPC values were expressed as mg gallic acid equivalents per liter (mg GAE/L).

Weight loss evaluation

For weight loss evaluation, mass measurements were performed using an analytical balance (Ohaus® Pioneer® analytical balance, Model PX124M, Ohaus Corporation, Parsippany, NJ, USA) with a precision of 0.0001 g at two time points: before (T0, baseline) and after 14 days of immersion (T1).

Color and translucency measurements

Color measurements were performed using a spectrophotometer (VITA Easyshade V, VITA Zahnfabrik, Bad Säckingen, Germany) at two time points: T0: Baseline (after preparation) and T1: after 14 days of staining and repolishing process. CIE L*, C*, and h° values were recorded at baseline and after the staining protocol using the spectrophotometer, and color differences were calculated using the CIEDE2000 formula (ΔE₀₀). Color differences were calculated using the CIEDE2000 formula (Eq. 1):

1

Color change measurements were performed in triplicate against a gray background under D65 standard illumination. Perceptibility and acceptability thresholds were set at ΔE₀₀ = 0.8 and 1.8, respectively [7, 25].

Translucency parameter (P₀₀) values were calculated based on CIEDE2000 color differences obtained against standardized black and white backgrounds. Each specimen was measured three times and mean values were used for statistical analysis.

SEM–EDS analysis

Surface morphology (at × 500 and × 1000 magnifications) and elemental composition were analyzed using scanning electron microscopy combined with energy-dispersive X-ray spectroscopy (SEM–EDS) (SEM, Apreo, Thermo Fisher Scientific, USA). The SEM was operated in secondary electron (SE) mode with an Everhart–Thornley detector (ETD) at an accelerating voltage of 20 kV, working distance of 9.5 mm, and spot size of 10.0.

Specimens were mounted on aluminum stubs using conductive carbon tape and sputter-coated with a thin palladium (Pd) layer (~ 10 nm) prior to observation to prevent charging. Representative micrographs were obtained to evaluate surface morphology, and elemental composition was quantified for carbon (C), oxygen (O), silicon (Si), and zirconium (Zr). EDS spectra were recorded at 20 kV accelerating voltage and 10 mm working distance. Quantitative data were obtained in weight percentage (wt%) for major elements using standardless ZAF correction. For each group, EDS analyses were performed at three randomly selected surface regions, and mean ± standard deviation values were calculated. For each group, three specimens per group were analyzed at baseline and after the staining procedure (n = 3).

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics software (version 22.0, IBM Corp., Armonk, NY, USA). Normality of data distribution was evaluated using the Shapiro–Wilk test. Depending on data normality, parametric (ANOVA) and non-parametric (Kruskal–Wallis) tests were applied with appropriate post-hoc comparisons. Since normality assumptions were violated in at least one group for each outcome variable, non-parametric tests were applied for all comparisons to ensure consistency. Phenolic content was analyzed using one-way ANOVA followed by the Games–Howell post-hoc test, as data were normally distributed. Weight loss, color change (ΔE₀₀), and translucency change (ΔP₀₀) were analyzed using the Kruskal–Wallis test, with pairwise comparisons performed using Bonferroni-adjusted Mann–Whitney U tests (significance threshold: p < 0.0083 for beverage comparisons; p < 0.0125 for composite comparisons). Elemental composition data (C, O, Si, and Zr) were analyzed using Kruskal–Wallis tests with pairwise post-hoc comparisons. Correlations between phenolic content, weight loss, optical parameters (ΔE₀₀, ΔP₀₀), and elemental composition were assessed using Spearman’s rank correlation coefficient. The significance level was set at p < 0.05.

Results

Phenolic content and pH

The mean pH values differed among the tea beverages, with black tea showing the lowest pH (5.3), followed by rooibos (6.2) and peppermint tea (6.5). Total phenolic content (TPC) also differed significantly among teas (F(2,57) = 2103.436, p < 0.001). Black tea exhibited the highest phenolic concentration (137.50 ± 3.10 mg GAE/L), followed by rooibos (99.89 ± 1.63 mg GAE/L) and peppermint tea (89.31 ± 2.45 mg GAE/L). Pairwise comparisons confirmed significant differences among all beverages (p < 0.001).

Weight loss

Weight loss values after 14 days of immersion are presented in Table 2. No significant differences among tea groups were observed for Estelite Palfique Paste (p = 0.303). In contrast, Estelite Posterior exhibited significantly higher weight loss in the rooibos tea group compared with peppermint tea (p = 0.006). Between materials, Estelite Posterior showed greater weight loss than Estelite Palfique Paste in distilled water (p = 0.012), whereas no differences were detected in the tea groups.

Table 2.

Weight loss in milligrams (mg) of composite resin materials after 14 days of immersion in different teas

Composite Resin	Distilled Water	Black tea	Rooibos tea	Peppermint tea
Estelite Palfique Paste	−0.10 (−0.10 – 0.20)A,a	0.10 (−0.80 – 0.70)A,a	0.00 (−0.60 – 0.30)A,a	0.00(−0.50 – 0.50)A,a
Estelite Posterior	0.20 (−0.30 – 0.50)AB,b	0.10 (−0.80 – 0.80)AB,a	0.15(0.00–0.30)B,a	−0.10(−0.50 – 0.20)A,a

^*Values are presented as Median (Min–Max). Different uppercase letters (A,B) indicate statistically significant difference between staining conditions for each composite resin (horizontal comparison: Bonferroni−adjusted Mann–Whitney test, p < 0.0083). Different lowercase letters (a,b) indicate statistically significant difference between composite resins within each aging condition (vertical comparison: Bonferroni−adjusted Mann–Whitney test, p < 0.0125)

Color change (ΔE₀₀)

Color change values are summarized in Table 3. Significant differences were observed among tea groups for both composite resins (Estelite Palfique Paste: p = 0.001; Estelite Posterior: p < 0.001). For Estelite Palfique Paste, black and rooibos teas produced significantly higher ΔE₀₀ values than distilled water (p = 0.003 and p = 0.01, respectively) and peppermint tea (p = 0.002 and p = 0.012, respectively), whereas no significant difference was observed between black and rooibos teas (p = 0.50). Estelite Posterior exhibited the highest color change after immersion in black tea. When comparing materials, Estelite Posterior showed significantly greater discoloration than Estelite Palfique Paste in the black tea group (p = 0.01). Based on the acceptability threshold (ΔE₀₀ = 1.8), all staining conditions exceeded the threshold for Estelite Palfique Paste. For Estelite Posterior, all tea groups exceeded the threshold, whereas the distilled-water control remained below this limit.

Table 3.

ΔE₀₀ values of composite resin materials after 14 days of immersion in different teas

Composite Resin	Distilled Water	Black tea	Rooibos tea	Peppermint tea
Estelite Palfique Paste	1.85 (0.99—2.48)A,b	2.90 (2.06—3.82)B,a	2.68 (1.36–4.19)B,a	1.98 (0.82–2.93)A,a
Estelite Posterior	1.39 (1.15—1.65)A,a	3.78 (2.89—4.22)C,b	2.90 (1.82—3.44)B,a	2.50 (1.76—2.95)B,a

^*Values are presented as Median (Min–Max). Different uppercase letters (A,B) indicate statistically significant difference between staining conditions for each composite resin (horizontal comparison: Bonferroni−adjusted Mann–Whitney test, p < 0.0083). Different lowercase letters (a,b) indicate statistically significant difference between composite resins within each aging condition (vertical comparison: Bonferroni−adjusted Mann–Whitney test, p < 0.0125)

Translucency parameter (ΔTP₀₀)

Changes in translucency are shown in Table 4. No significant difference in ΔP₀₀ values was found between the two composite resins (p = 0.053). However, tea type significantly affected translucency changes (p < 0.001). The greatest translucency reduction occurred in the black tea group, followed by rooibos and peppermint teas, whereas the distilled-water group showed the lowest change. Pairwise comparisons confirmed significant differences among most tea groups.

Table 4.

ΔTP₀₀ values of the composite resins after 14 days of immersion in different teas

Composite Resin	Distilled water	Black tea	Rooibos tea	Peppermint tea
Estelite Palfique Paste	−0.44 (−0.58—−0.17)A,b	−4.08 (−5.13—−2.39)D,a	−2.15 (−2.87—−1.69)C,a	−0.69 (−1.00—−0.61)B,b
Estelite Posterior	−0.13 (−0.26—−0.03)A,a	−2.92 (−3.93—−2.63)C,a	−2.00 (−2.15—−1.23)B,a	−0.40 (−0.65—−0.06)A,a

^*Values are presented as Median (Min–Max). Different uppercase letters within the same row indicate significant differences among tea groups for each composite resin (p <0.0083, Bonferroni−adjusted Mann–Whitney test). Different lowercase letters within the same column indicate significant differences between composite resins within the same tea group (p <0.0125, Bonferroni−adjusted Mann–Whitney test)

SEM images and Elemental compositions (EDS analysis)

Representative SEM images of the composite resin surfaces after immersion are presented in Fig. 2. Elemental distributions obtained by EDS analysis are illustrated in Figs. 3 and 4 for Estelite Palfique Paste and Estelite Posterior, respectively.

Fig. 2.

SEM images of the specimens at different magnifications: a × 500 and b × 1000

Fig. 3.

The overall distribution of elemental compositions (C, O, Si and Zr) for Estelite Palfique Paste after immersion in different tea solutions

Fig. 4.

The overall distribution of elemental compositions (C, O, Si and Zr) for Estelite Posterior after immersion in different tea solutions

Elemental compositions differed between the two composite resins, and tea type influenced the distribution of several elements.

Carbon (C)

Tea immersion significantly affected surface carbon levels (p < 0.001). Black tea produced the highest carbon percentages in both composite resins, followed by rooibos and peppermint teas.

Oxygen (O)

Oxygen content differed significantly between materials (p < 0.001), with Estelite Posterior showing higher values. Tea immersion increased oxygen levels compared with the distilled-water control.

Silicon (Si)

For Estelite Palfique Paste, no significant differences among tea groups were detected. In contrast, Estelite Posterior showed significantly lower silicon levels after black tea exposure compared with the other groups (p < 0.001).

Zirconium (Zr)

Zirconium content differed between composite resins (p < 0.001) but was not significantly influenced by tea type.

Correlation analysis

Spearman correlation analysis revealed significant associations among phenolic content, optical parameters, and surface elemental composition.

Phenolic content showed a positive correlation with weight loss (ρ = 0.359, p = 0.005) and color change (ρ = 0.614, p < 0.001), and a negative correlation with translucency change (ρ = − 0.894, p < 0.001). In addition, phenolic concentration was positively correlated with surface carbon percentage (ρ = 0.660, p < 0.001).

Carbon content was positively correlated with color change (ρ = 0.507, p < 0.001), whereas silicon content showed a negative correlation with color change (ρ = − 0.503, p < 0.001) and a positive correlation with translucency parameter (ρ = 0.715, p < 0.001). No significant correlations were observed between zirconium content and optical parameters.

Discussion

The present study investigated the effects of polyphenol-rich tea beverages on the color stability, translucency, and surface elemental characteristics of two contemporary composite resins. By combining optical measurements with surface-sensitive SEM–EDS analysis, this study aimed to explore potential associations between esthetic degradation and surface chemical alterations under controlled staining conditions.

The immersion protocol was designed to simulate realistic beverage exposure using standardized brewing conditions. Because elevated temperatures may independently influence optical properties of composite resins [14], the immersion temperature was maintained close to beverage-serving conditions to approximate daily consumption patterns while minimizing additional thermal aging effects. Nevertheless, the potential influence of temperature cannot be completely excluded.

Differences in discoloration among tea beverages may be associated with variations in their phenolic composition and concentration [22, 23, 25]. Black tea contains high-molecular-weight polyphenols such as tannins, theaflavins, and thearubigins with strong chromogenic potential. Rooibos and peppermint teas contain flavonoids and phenolic acids with comparatively lower staining capacity [4–6]. In the present study, black tea produced the greatest color change in both composite resins. This pronounced discoloration may be attributed to the high tannin content of black tea, which facilitates adsorption and penetration of chromogenic molecules into the resin matrix through water sorption pathways [26, 27]. Herbal teas contain distinct bioactive compounds such as flavonoids and hydrolysable tannins that may interact differently with resin matrices compared with beverages rich in condensed tannins and metal-polyphenol complexes [27].

In addition to phenolic composition, beverage acidity may also contribute to staining behavior. In this study, black tea presented a mean pH of 5.3, while rooibos and peppermint teas exhibited higher pH values. Acidic conditions may facilitate adsorption and absorption of pigments into the resin matrix, potentially accelerating discoloration and contributing to surface degradation processes [15, 27, 28].

Color differences were interpreted according to the CIEDE2000 system, where ΔE₀₀ values of 0.8 and 1.8 represent perceptibility and acceptability thresholds, respectively [10, 25]. In the present study, polyphenol-rich beverages produced color changes that frequently exceeded the perceptibility threshold and in several cases surpassed the clinical acceptability threshold. Clinically, such discoloration may compromise the esthetic appearance of restorations and may require polishing, repair, or replacement, particularly in visible regions [25, 29, 30].

The different discoloration behaviors observed between the composite resin materials may be related to variations in their structural composition. Differences in resin matrix chemistry, filler particle size distribution, and filler–matrix interactions can influence the susceptibility of composite resins to pigment adsorption and retention [19, 20]. Heterogeneous filler distributions may promote surface irregularities, facilitating chromogen retention. [31, 32]. Therefore, the variability observed between the tested materials likely reflects not only the staining potential of the beverages but also material-dependent structural characteristics.

Correlation analyses demonstrated a significant positive association between total phenolic content and color change (ΔE₀₀). Polyphenols, particularly tannins, possess strong chromogenic properties and can interact with the organic resin matrix through adsorption and absorption mechanisms, promoting pigment retention on the composite surface and within superficial layers of the material [33, 34].

In contrast, weight loss values showed only weak associations with phenolic content and were not significantly correlated with color change. This finding suggests that discoloration may predominantly arise from surface phenomena rather than bulk material degradation. Pigment adsorption and superficial penetration may therefore occur independently of measurable mass loss [19, 31]. Clinically, this implies that restorations may exhibit unacceptable esthetic changes despite relatively preserved structural integrity, and surface-oriented interventions such as repolishing may be considered before replacement.

Changes in translucency followed a pattern similar to color alterations, with black tea causing the greatest reduction in translucency for both materials. Because translucency is governed by light transmission and scattering within the composite structure, these changes may reflect surface modifications affecting refractive index matching at the filler–matrix interface [12, 20]. Distilled water produced the smallest translucency changes, indicating greater optical stability in the absence of chromogenic agents. Accordingly, the first null hypothesis (H₀), which stated that tea type would not affect optical properties, and the second null hypothesis (H₀₁), which proposed that composite type would not influence optical outcomes, were rejected.

EDS analysis revealed notable variations in surface elemental composition after immersion in the tea solutions, particularly regarding carbon and silicon levels. Increased carbon signals were observed following exposure to tea beverages, especially black and rooibos teas, which may indicate adsorption of organic compounds from the staining media onto the resin surface [35]. In contrast, reductions in silicon-related signals, particularly in Estelite Posterior after black tea immersion, may reflect surface alterations at the filler–matrix interface. Silicon is associated with silanized filler phases, hydrolytic degradation of siloxane bonds may weaken filler–matrix bonding under aqueous conditions [36]. Consequently, decreased silicon signals may indicate partial disruption of this interface and superficial material deterioration. Zirconium signals remained stable across all conditions, suggesting that zirconium-containing filler phases were less affected by the staining environment. Therefore, the third null hypothesis (H₀₂), stating that tea exposure would not influence surface elemental composition, was rejected.

SEM observations supported these findings, revealing surface irregularities following immersion in tea solutions. Previous studies have reported that beverages such as black tea can increase surface roughness and promote morphological changes including microcracks, matrix abrasion, and filler particle loss [15, 37, 38]. These changes may result from hydrolytic degradation processes in which water sorption leads to swelling and weakening of the resin matrix and silane interface [19, 39, 40].

Correlation analyses between optical parameters and surface elemental signals were conducted to identify potential associations, and findings should be evaluated within the descriptive nature of SEM–EDS data [15–17]. Nevertheless, consistent associations between silicon-related signals and optical parameters suggest that surface chemistry may contribute to the esthetic behavior of composite resins exposed to staining agents [15]. The negative relationship observed between silicon content and color change may indicate that disruption of the filler–matrix interface increases susceptibility to staining. In addition, increased carbon levels may reflect adsorption of organic chromogens, potentially altering light scattering behavior and contributing to translucency reduction. The observed associations between phenolic content, carbon enrichment, silicon-related signals, and optical parameters suggest that surface chemical alterations may contribute to the staining behavior of composite resins. Therefore, the fourth null hypothesis (H₀₃), which proposed no correlation between phenolic content, elemental composition, and optical parameters, was rejected.

Despite repolishing after the staining procedure, discoloration persisted in several specimens, suggesting that staining was not restricted to superficial deposits. The observed associations between phenolic content, carbon enrichment, and color change further support the role of matrix–chromogen interactions in staining susceptibility, particularly under highly chromogenic conditions such as black tea exposure.

The present findings indicate that discoloration of composite resins involves not only esthetic alterations but also material-dependent surface processes. Integrating optical measurements with elemental analysis may therefore provide additional insight into how dietary chromogens interact with restorative materials and influence their long-term esthetic performance.

From a clinical perspective, the ΔE₀₀ values observed in this study may have implications for restorative decision-making. When ΔE₀₀ exceeds the perceptibility threshold of 0.8, the color change may become detectable to the trained eye, warranting closer monitoring of the restoration. When ΔE₀₀ surpasses the acceptability threshold of 1.8, the discoloration may no longer be clinically acceptable, and intervention could be considered. In the present study, both composite resins exceeded the acceptability threshold following immersion in black tea, suggesting that restorations in patients with habitual black tea consumption may require earlier maintenance intervals. Repolishing may represent a conservative first-line approach; however, the persistence of color changes even after repolishing in the present study suggests that staining may extend beyond superficial deposits into the resin matrix. In such cases, restoration replacement might ultimately be necessary, particularly in esthetically sensitive regions. These findings suggest that clinicians may benefit from considering patients' dietary habits, including the type and frequency of polyphenol-rich beverage consumption, when selecting restorative materials and planning maintenance protocols. Nevertheless, clinical validation through in vivo studies is needed to confirm these observations.

This in vitro study has several limitations that should be considered when interpreting the findings. First, the use of only two composite resin materials limits the generalizability of the results to other commercially available materials with different matrix compositions and filler characteristics. Second, the in vitro nature of the study does not fully replicate the complex oral environment, including the effects of saliva, oral biofilm, masticatory forces, and temperature fluctuations on composite resin degradation. Third, the immersion protocol, while designed to simulate daily beverage consumption, may not accurately reflect the variable exposure patterns observed in clinical practice. Fourth, the specific phenolic compounds responsible for the observed optical and surface changes could not be identified, as total phenolic content was measured rather than individual compound fractions. Fifth, the study examined only short-term effects over a 14-day period; longer observation periods may reveal different degradation patterns. Finally, color and translucency measurements were performed using an intraoral spectrophotometer, which may have inherent limitations compared to laboratory-grade benchtop devices.

Conclusion

Exposure to polyphenol-rich beverages produced measurable changes in color and translucency of composite resins, accompanied by alterations in surface elemental signals. Among the tested materials, Estelite Palfique Paste demonstrated better color stability under black tea exposure, whereas Estelite Posterior exhibited relatively better translucency stability. Carbon enrichment and silicon reduction were associated with optical degradation, suggesting possible alterations at the filler–matrix interface. Zirconium content remained unaffected.

Neither material demonstrated overall superiority across all optical parameters. Clinicians may consider these material-specific optical behaviors when selecting composite resins for patients with habitual polyphenol-rich beverage consumption, particularly in esthetically demanding regions.

Correlation analyses revealed significant associations between total phenolic content and color change, as well as between surface elemental signals and optical parameters, supporting the hypothesis that polyphenol-induced surface chemical alterations contribute to esthetic degradation of composite resins. These findings suggest a relationship between surface chemical characteristics and esthetic behavior of composite resins, although further studies are required to confirm underlying mechanisms.

Authors’ contributions

HDK performed material preparation, data collection and statistical analysis, wrote the first draft. LST contributed to writing, review, and editing of the manuscript. All authors contributed to the study's conception and design. All authors commented on previous versions. All authors read and approved the final manuscript before submission and agreed to be accountable for all aspects of the study.

Funding

Not applicable.

Data availability

The corresponding author can be contacted to provide the data of this study on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.