Can household ultrasonic cleaners enhance the stain removal efficacy of chemical agents on orthodontic aligners?

Abstract

Objectives

This study aimed to evaluate the effectiveness of household ultrasonic cleaning with/without peroxide-based and peroxide-free cleaning methods for stain removal on polyurethane aligner material.

Methods

Seventy-two polyurethane aligner specimens (Invisalign, Align Technology Inc.) were stained for seven days in a standardized coffee solution and randomly assigned to six cleaning protocols (n = 12/group): peroxide-free cleaner (Invisalign Cleaning Crystals [ICC], Align Technology Inc., San Jose, CA, USA) or peroxide-based cleaner (Corega Proguard [CP], Stafford-Miller Limited, Waterford, Ireland), each applied without, 5, or 10 min of household ultrasonic assistance (Shantou Huan Te Tong Intelligent Technology Co., Guangdong, China; Serial No: 20240001) at 50 kHz. Color was measured at baseline (t₀), after staining (t₁), and after ultrasonuc cleaning (t₂) using a spectrophotometer (VITA Easyshade Compact; VITA Zahnfabrik, Bad Säckingen, Germany). The CIEDE2000 formula was used to calculate color changes (ΔE₀₀), specifically evaluating staining (ΔE_t0-t1), net color change ((ΔE_t0-t2), and cleaning efficacy (ΔE_t1-t2). Data were analyzed using Repeated Measures and two-way ANOVA, followed by Bonferroni adjustment (significant at p < 0.017).

Results

All specimens showed comparable discoloration after staining (p > 0.05). Cleaning efficacy (ΔE_t1–t2) did not differ between cleaning agents, ultrasonic durations, or their interaction (p > 0.05). Net color change (ΔE_t0–t2) differed between agents (p = 0.018), with ICC exhibiting lower residual discoloration than CP. Although a significant agent-ultrasonic interaction was detected (p = 0.043), Bonferroni-adjusted post-hoc comparisons did not reveal any statistically significant pairwise differences between individual cleaning protocols.

Conclusions

Household ultrasonic cleaning did not improve the stain-removal efficacy of either cleaning agent. Color recovery was primarily influenced by the chemical formulation of the cleaner rather than ultrasonic assistance.

Introduction

Clear aligner therapy has become an increasingly common orthodontic treatment modality due to its aesthetic appeal and patient comfort [1]. These appliances are removable, transparent trays manufactured from multilayer thermoplastic polyurethane/copolyester materials that apply controlled forces to teeth to achieve orthodontic movement [2].

Since clear aligners are worn for extended periods (typically 20–22 h per day), preservation of optical transparency is essential for maintaining their aesthetic advantage. However, studies have shown that many patients do not strictly adhere to the recommendation to remove them before consuming staining foods or beverages such as coffee, tea, or red wine [3]. This noncompliance leads to discoloration, which not only compromises their aesthetic invisibility but also raises concerns about oral hygiene [4, 5]. In addition to staining, bacterial biofilms frequently accumulate on aligner surfaces, especially in recessed areas like attachment dimples and cusp tips [6], further affecting their aesthetic invisibility that constitutes one of their primary advantages.

Brushing is a widely recommended and commonly practiced method for cleaning removable dental appliances [7]. However, mechanical cleansing alone is often insufficient for effectively removing bacterial biofilms and may increase surface roughness, which in turn makes the appliances more susceptible to pigment accumulation and microbial colonization [8, 9], ultimately affecting the color stability and transparency of clear aligners. To address this issue, patients frequently turn to over the counter (OTC) cleaning solutions, which are predominantly available in the form of effervescent tablets or gels. These products are marketed to remove stains and biofilms through a combination of chemical action (e.g., surfactants, oxidizing agents like hydrogen peroxide, and enzymes) and effervescence.

Concurrently, household ultrasonic cleaning devices have been marketed directly to consumers as an adjunctive or standalone method for washing dental appliances. These devices generate high-frequency sound waves (typically ranging from 20 to 120 kHz) that create cavitation bubbles in a liquid medium [10]. The implosion of these microbubbles produces microscopic scrubbing action, theoretically dislodging debris from hard-to-reach areas of the aligner’s complex geometry. Although predominantly promoted as hygiene-enhancing tools aimed at biofilm disruption, the cavitation-induced mechanical forces may also theoretically contribute to pigment detachment from the polymer surface, thereby influencing optical color recovery.

Hydrogen peroxide-based cleaners are widely used because they release reactive oxygen species that break down organic deposits and disrupt bacterial biofilms, whereas peroxide-free agents rely primarily on detergent, surfactant, and chelating mechanisms [11]. While both hydrogen peroxide-containing and hydrogen peroxide-free OTC solutions and ultrasonic devices that enhance mechanical plaque and stain removal through cavitation-induced shear forces are widely promoted for cleaning dental appliances [12–15], a gap exists in the literature regarding their potential synergistic effect. Moreover, household ultrasonic cleaners are primarily promoted as hygiene-enhancing devices rather than pigment-targeted systems. Nevertheless, patients may assume that improved hygiene translates into enhanced aesthetic recovery. Consequently, this discrepancy underscores the need to specifically evaluate their impact on colorimetric outcomes.

To achieve standardized and reproducible discoloration prior to cleaning evaluation, a high-chromogen staining model is required. Coffee has been widely used in dental materials research as a reproducible extrinsic staining medium capable of inducing measurable chromatic changes under controlled laboratory conditions. Although clinical aligner discoloration is multifactorial, employing a single, standardized staining agent allows for controlled intergroup comparisons while minimizing confounding variability.

Accordingly, the present study aimed to evaluate the stain-removal efficacy of household ultrasonic cleaners used in combination with OTC chemical agents, and to assess the influence of different ultrasonic exposure durations on colorimetric recovery quantified by the CIEDE2000 (ΔE₀₀) metric, rather than microbiological decontamination, in order to directly address clinically relevant expectations regarding aligner transparency. The null hypothesis was that cleanser type, ultrasonic activation, and exposure duration would have no significant effect on ΔE₀₀ values in stained clear aligners.

Materials and methods

An a priori power analysis was conducted using G*Power (version 3.1.9.7; Franz Faul, Universität Kiel, Germany) to determine the required sample size. Assuming a significance level of 5%, a statistical power of 95%, and an effect size of f = 0.25, the analysis indicated that a total of 72 clear aligners would be required. An effect size of f = 0.25 was selected, corresponding to a medium effect according to Cohen’s conventions for ANOVA-based designs, as previous studies investigating color stability of thermoplastic aligners have reported moderate intergroup differences in ΔE₀₀ values [16, 17].

Orthodontic polyester-urethane clear aligners (Invisalign; Align Technology Inc., San Jose, CA, USA) were trimmed mesially to first molars and distally to second molars to span across two molars. Each specimen was assigned a numeric code prior to allocation, and randomization into six groups was performed using a computer-generated random sequence.

Baseline color measurements (t₀) were recorded using a spectrophotometer (VITA Easyshade Compact; VITA Zahnfabrik, Bad Säckingen, Germany). Before each measurement, the spectrophotometer was positioned centrally on the occlusal surface of the specimen and calibrated using the device’s dedicated white calibration standard after every 10 measurements. All assessments were performed under standardized D65 daylight conditions, with the specimens placed against a uniform white background. For each specimen, three consecutive readings were recorded and averaged. The investigator performing spectrophotometric measurements was blinded to group allocation.

Color evaluation was recorded using the Commission Internationale de l’Éclairage (CIE) Lab* color system, a three-coordinate model in which L* represents lightness (ranging from 0 for black to 100 for white), a* denotes the red-green axis (positive values indicate red, negative values indicate green), and b* reflects the yellow-blue axis (positive values indicate yellow, negative values indicate blue). Each color measurement was performed three times and averaged to obtain a single value for subsequent statistical analysis. The total color change (ΔE₀₀) between different time points was calculated using the CIEDE 2000 formula [16]:

Following baseline measurements (t₀), all aligners were stained by immersion for seven days in a coffee solution prepared by dissolving 15 g of coffee powder (Nescafé^® Classic, Nestlé SA, Vevey, Switzerland) in 500 mL of boiling distilled water. The mixture was stirred for 10 min, filtered through paper, and freshly prepared daily. Continuous immersion was used as an accelerated staining model to generate standardized chromatic changes under controlled laboratory conditions. After the staining period, initial color measurements (t₁) were recorded, and the aligners were stored individually in 15-mL Falcon tubes.

Two cleaning agents were used in this study: a peroxide-free effervescent tablet (Invisalign Cleaning Crystals [ICC], Align Technology Inc., San Jose, CA, USA) and a peroxide-based effervescent tablet (Corega Proguard [CP], Stafford-Miller Limited, Waterford, Ireland) (Table 1).

Table 1.

The compositions of the cleaning agents

Cleaners	Manufacturer	Ingredients	Lot No
Invisalign Cleaning Crystals	Align Technology Inc., San Jose, CA, USA	Sodium Sulfate, Sodium Carbonate, Sodium Tripolyphosphate, Sodium Dichloroisocyanurate, Sodium Lauryl Sulfate	20363B
Corega Proguard	Stafford-Miller Limited, Waterford, Ireland	Sodium Bicarbonate, Citric Acid, Potassium Monopersulfate, Sodium Carbonate, Sodium Carbonate Peroxide, TAED, Sodium Benzoate, PEG-180, Sodium Lauryl Sulfate, VP/VA Copolymer, Aroma, Subtilisin, Cellulose Gum, Sodium Nitrite, CI 4209 CI 73,015	SM8P

Cleaning was performed with or without adjunctive ultrasonic activation using a portable household ultrasonic device (Model S920; Shantou Huan Te Tong Intelligent Technology Co., Guangdong, China). The device operated at a nominal frequency of 50 kHz with a manufacturer-reported rated power of 0.6 W and a tank volume of 200 mL (≈ 3 W/L power density). The unit was battery-operated (3.7 V; 500 mAh), functioned in continuous mode, and did not include active heating. All procedures were conducted at controlled room temperature (22–24 °C) and no external heating was applied during ultrasonic activation.

Specimens were randomly assigned to six independent groups (n = 12 per group). The cleaning methods used were as follows:

• Group 1: ICC without ultrasonic assistance,

• Group 2: ICC with 5-minutes ultrasonic assistance,

• Group 3: ICC with 10-minutes ultrasonic assistance,

• Group 4: CP without ultrasonic assistance,

• Group 5: CP with 5-minutes ultrasonic assistance,

• Group 6: CP with 10-minutes ultrasonic assistance.

For Groups 1 to 3, ICCs were prepared by dissolving one sachet in 70 mL of distilled water with gentle agitation for 20 s. For Groups 4 to 6, CP tablets were dissolved in 70 mL of distilled water according to the manufacturer’s instructions. In each group, aligners were immersed in 10 mL of the prepared solution. Non-ultrasonic groups were soaked for 15 min at room temperature. Ultrasonic groups underwent activation for either 5–10 min.

Following the cleaning protocols, all aligners were thoroughly rinsed under running water for 30 s to remove any residual cleaning agents, and final color measurements (t₂) were performed (Fig. 1). ΔE₀₀ values were calculated according to time intervals as ΔE_t0−t1 (baseline - after 1 week of discoloration), ΔE_t0−t2 (baseline - after the cleaning procedure) and ΔE_t1−t2 (after 1 week of discoloration - after the cleaning procedure). The ΔE threshold was set at 0.8 for perceptibility and 1.8 for acceptability [17].

Fig. 1.

Flow chart of study

Statistical analysis

Normality was assessed using the Shapiro-Wilk test. Dexcriptive statistics were summarized as mean and standard deviation (SD). Changes in color coordinates (L*, a* and b*) across time points (t₀, t₁ and t₂) were analyzed using Repeated Measures ANOVA. The assumption of sphericity was evaluated using Mauchly’s test and, when violated, a Greenhouse-Geisser correction was applied. The color change parameters (ΔE₀₀) were evaluated using a two-way ANOVA to determine the main and interaction effects of the cleaning agent and ultrasonic duration. The complete ANOVA models were tabulated to report the degrees of freedom (df), Sum of Squares (SS), Mean Square (MS), F-statistics, and partial eta-squared (η_p²). For post-hoc evaluations, pairwise comparisons were performed using a Bonferroni correction, with the significance threshold applied at p < 0.017.

All statistical analyses were conducted using IBM SPSS Statistics software (version 26.0; IBM Corp., Armonk, NY). The statistical significance was set at p < 0.05.

Results

Colorimetric analysis of L*, a* and b*

Descriptive statistics for the L*, a*, and b* values at baseline (t₀), after staining (t₁), and after cleaning (t₂) are presented in Table 2. Statistical analysis revealed that the L* values did not change significantly after staining (p > 0.05). In contrast, the chromatic parameters showed highly significant changes. Pairwise comparisons confirmed a significant decrease in a* values (p < 0.001) and a significant increase in b* values (p < 0.001) following immersion in the coffee solution. Notably, the subsequent cleaning process significantly reduced the yellowness (b*, p < 0.001 ), but did not produce a significant shift in the red-green axis (a*, p > 0.05).

Table 2.

L*, a*, and b* values of clear aligners at baseline (t₀), after staining (t₁), and after cleaning (t₂) for each experimental group

	Timepoint	L* (Mean ± SD)	a* (Mean ± SD)	b* (Mean ± SD)
Group 1 ICC + 0 min	Baseline (t0)	86.31 ± 8.09A	1.01 ± 0.55A	3.98 ± 1.86A
	After Staining (t1)	83.64 ± 2.79AB	-1.41 ± 1.53B	16.97 ± 4.41B
	After cleaning (t2)	85.77 ± 3.87B	-0.45 ± 1.03B	13.48 ± 2.56C
Group 2 ICC + 5 min	Baseline (t0)	81.23 ± 7.09A	1.22 ± 0.54A	3.37 ± 1.19A
	After Staining (t1)	83.83 ± 4.86AB	-0.81 ± 1.07B	16.56 ± 4.04B
	After cleaning (t2)	81.59 ± 5.95B	-1.31 ± 0.77B	14.90 ± 3.02C
Group 3 ICC + 10 min	Baseline (t0)	78.26 ± 5.58A	1.09 ± 0.76A	3.87 ± 1.11A
	After Staining (t1)	81.32 ± 4.47AB	-0.48 ± 1.28B	18.38 ± 3.23B
	After cleaning (t2)	83.63 ± 4.21B	-1.33 ± 1.06B	16.22 ± 2.34C
Group 4 CP + 0 min	Baseline (t0)	79.51 ± 6.06A	1.27 ± 0.42A	2.79 ± 1.65A
	After Staining (t1)	83.00 ± 3.15AB	-0.40 ± 1.34B	19.71 ± 3.77B
	After cleaning (t2)	85.07 ± 2.62B	-1.09 ± 1.03B	16.14 ± 3.09C
Group 5 CP + 5 min	Baseline (t0)	82.08 ± 9.42A	1.25 ± 1.01A	3.53 ± 1.21A
	After Staining (t1)	84.22 ± 4.81AB	-0.12 ± 1.55B	16.41 ± 4.10B
	After cleaning (t2)	83.06 ± 3.16B	-0.94 ± 1.09B	14.67 ± 3.64C
Group 6 CP + 10 min	Baseline (t0)	81.60 ± 6.13A	1.43 ± 0.53A	3.20 ± 1.36A
	After Staining (t1)	81.86 ± 5.56AB	0.07 ± 1.48B	17.82 ± 3.76B
	After cleaning (t2)	85.14 ± 3.41B	-0.61 ± 1.11B	14.28 ± 3.37C

SD Standard deviation, ICC Invisalign Cleaning Crystals, CP Corega Proguard, 0 min No ultrasonic cleaning, 5 min 5 min of ultrasonic cleaning, 10 min 10 min of ultrasonic cleaning

Different uppercase superscript letters within the same group column indicate statistically significant differences over time based on Bonferroni pairwise comparisons

Color change (ΔE₀₀)

The effects of the cleaning method and ultrasonic assistance on the staining effect (ΔE_t0−t1), net color change (ΔE_t0−t2), and cleaning efficacy (ΔE_t1−t2) are presented in Table 3, while the corresponding two-way ANOVA results and effect sizes are detailed in Table 4.

Table 3.

Effects of cleaning method and ultrasonic duration on color change parameters (ΔE₀₀) of clear aligners

	Group 1 ICC + 0 min	Group 2 ICC + 5 min	Group 3 ICC + 10 min	Group 4 CP + 0 min	Group 5 CP + 5 min	Group 6 CP + 10 min
ΔEt0−t1
Mean ± SD	26.95 ± 9.80	31.10 ± 10.76	26.25 ± 9.90	37.30 ± 14.19	29.35 ± 12.42	31.85 ± 10.81
95% CI	20.72; 33.17	24.22; 37.90	19.96; 32.54	28.28; 46.32	21.45; 37.24	24.99; 38.72
ΔEt0−t2
Mean ± SD	22.90 ± 9.45	31.10 ± 11.90	27.19 ± 8.87	38.27 ± 13.55	30.27 ± 12.33	31.72 ± 9.64
95% CI	16.90; 28.91	23.53; 38.62	21.56; 32.83	29.66; 46.88	22.44; 38.10	25.59; 37.84
ΔEt1−t2
Mean ± SD	8.92 ± 5.35	9.47 ± 3.93	8.91 ± 4.32	7.89 ± 3.20	10.30 ± 5.66	9.95 ± 4.77
95% CI	5.53; 12.32	6.97; 11.96	6.17; 11.65	5.85; 9.93	6.70; 13.89	6.92; 12.99

SD Standard deviation, CI Confidence intervals, ICC Invisalign Cleaning Crystals, CP Corega Proguard, 0 min No ultrasonic cleaning, 5 min 5 min of ultrasonic cleaning, 10 min 10 min of ultrasonic cleaning

Table 4.

Two-way ANOVA for color change (ΔE₀₀)

	df	SS	MS	F	P value	Partial η2
ΔEt0−t1 (Staining Effect)
Cleaning Agent	1	405.793	405.793	3.111	0.082	0.045
Ultrasonic Duration	2	115.476	57.738	0.443	0.644	0.013
Cleaning Agent × Ultrasonic Duration	2	443.578	221.789	1.701	0.190	0.049
Error	66	8607.710	130.420
ΔEt0−t2 (Net Color Change)
Cleaning Agent	1	727.902	727.902	5.922	0.018*	0.082
Ultrasonic Duration	2	22.158	11.079	0.090	0.914	0.003
Cleaning Agent × Ultrasonic Duration	2	814.552	407.276	3.314	0.043*	0.091
Error	66	8112.025	122.909
ΔEt1−t2 (Cleaning Effect)
Cleaning Agent	1	1.397	1.397	0.066	0.799	0.001
Ultrasonic Duration	2	27.457	13.728	0.645	0.528	0.019
Cleaning Agent × Ultrasonic Duration	2	15.705	7.853	0.369	0.693	0.011
Error	66	1405.652	21.298

df Degrees of freedom, SS Sum of Squares, MS Mean Square, F ANOVA test statistic, Partial η² Partial Eta Squared

*Significant at P < 0.05

Staining effect (ΔE_t0–t1)

All aligners exhibited visible discoloration after immersion in the staining solution for seven days. No differences found among the groups in terms of staining intensity (p > 0.05).

Net color change (ΔE_t0–t2)

The analysis of net color change revealed significant differences in the final color of the aligners depending on the cleaning protocol. While there was no significant main effect for the ultrasonic cleaning duration (p > 0.05), a significant main effect was found for the cleaning agent (p = 0.018). Overall, the ICC group exhibited significantly lower mean ΔE₀₀ values (27.06 ± 10.42) compared to the CP group (33.42 ± 12.13).

A significant interaction effect was observed between the cleaning agent and ultrasonic cleaning duration (p = 0.043) (Fig. 2). For the ICC group, the use of ultrasonic cleaning appeared to hinder efficacy, with mean ΔE₀₀ values increasing from 22.90 ± 9.45 (no ultrasonic cleaning) to 31.08 ± 11.90 (5 min of ultrasonic cleaning). Conversely, for the CP group, the use of ultrasonic cleaning appeared to improve efficacy, with mean ΔE₀₀ values decreasing from 38.27 ± 13.55 (no ultrasonic cleaning) to 30.27 ± 12.33 (5 min of ultrasonic cleaning). However, this interaction must be interpreted with caution, as Bonferroni-adjusted pairwise comparisons within each agent (0 vs. 5 min; 0 vs. 10 min; 5 vs. 10 min) did not reveal statistically significant differences (p > 0.017).

Fig. 2.

Mean ΔE_t0−t2 values of clear aligners across experimental groups. t₀: baseline color measurements; t₂: after cleaning; ICC: Invisalign Cleaning Crystals; CP: Corega Proguard

Cleaning effect (ΔE_t1–t2)

The cleaning effect was similar across all experimental protocols (p > 0.05). No significant main effect was found for the cleaning agent, ultrasonic cleaning duration, or their interaction effect (p > 0.05) (Fig. 3).

Fig. 3.

Mean ΔE_t1−t2 values of clear aligners across experimental groups. t₁: staining; t₂: after cleaning; ICC: Invisalign Cleaning Crystals; CP: Corega Proguard

Color change relative to the clinical acceptability threshold

The clinical acceptability threshold for color difference in this study was set at ΔE₀₀=1.8, with a perceptibility threshold of ΔE₀₀=0.8 [18]. Following the seven-day immersion in the coffee solution, the initial staining effect (ΔE_t0−t1) produced severe discoloration that vastly exceeded both thresholds. Mean staining values ranged from 26.25 ± 9.90 to 37.30 ± 14.19 across the groups. The subsequent cleaning protocols yielded a highly perceptible cleaning effect (ΔE_t1−t2). The mean color change during this phase ranged from 7.89 ± 3.20 to 10.30 ± 5.66, which confirms that the cleaning agents did induce a visually noticeable removal of stains across all experimental groups. However, despite this visually perceptible stain removal, the net color change from baseline to after the cleaning procedures (ΔE_t0−t2) demonstrated residual discoloration that remained substantially above the clinical acceptability threshold (ΔE₀₀ > 1.8). Specifically, the mean ΔE₀₀ values across all six protocols ranged from 22.90 ± 9.45 in the ICC group (0 min) to 38.27 ± 13.55 in the CP group (0 min). Consequently, while the cleaning methods provided a noticeable improvement, none of the tested combinations were capable of fully restoring the aligners to an optically acceptable baseline.

Discussion

The aesthetic appeal of clear aligners is often compromised by their tendency to stain when exposed to common food and drink pigments, making reliable daily cleaning essential for maintaining their transparency and ensuring oral hygiene. Although numerous OTC cleaning agents and consumer-grade ultrasonic devices are marketed for aligner maintenance, evidence regarding their combined effect on optical color recovery remains limited. This study therefore addresses a significant gap in the literature by systematically assessing the impact of ultrasonic assistance on the stain-removal performance of different cleaning agents. The primary outcome of this investigation revealed that the addition of ultrasonic cleaning, regardless of duration, did not confer a significant advantage in stain removal efficacy compared to the use of chemical cleaners alone. However, an interaction was observed, indicating that the ultrasonic cleaning influenced the two cleaning agents, improving the performance of the peroxide-based CP while hindering the efficacy of the non-peroxide ICC. Consequently, the null hypothesis of this study was partially rejected.

The results show that stain removal efficacy (ΔE_t1−t2) was similar across all groups. This finding indicates that neither the type of cleaning agent nor the use of ultrasonic assistance conferred a significant advantage in the capacity to remove stains. Sambhavaphol et al. [12]. also reported that none of the color parameters demonstrated significant differences in their in-vivo evaluation of four different cleaning methods used with ultrasonic assistance. Analysis of the net color change (ΔE_t0−t2), however, revealed that the choice of cleaning agent played a decisive role in the final appearance of the aligners. The ICC group exhibited significantly better color restoration compared to the CP group. However, the cleaning protocols responded differently to physical agitation. While CP performed poorly on its own, its efficacy improved with ultrasonic assistance. In contrast, ICC performed best when used as a soaking solution alone, with ultrasonic agitation appearing to diminish its effectiveness. The difference between these two groups may therefore confirm a performance disparity between peroxide-based and non-peroxide-based cleaners.

Changes in the mechanical properties of clear aligners substantially influence their functional performance and overall longevity [19]. Consequently, the choice of appropriate cleaning protocols and compliance with recommended maintenance practices are crucial for preserving appliance integrity. This consideration becomes even more critical during treatment phases that require prolonged wear of the same aligner, such as extended revision periods. Agarwal et al. [18]. demonstrated that the light transmittance of polyurethane retainers (Vivera) significantly decreased over a 6-month cleaning period, with toothbrushing producing the greatest reduction and ICCs, Listerine, and Polident causing the least optical alteration. Similarly, Wible et al.. reported that all tested cleaning approaches, including ICCs, Retainer Brite tablets, Polident, Listerine Cool Mint, 2.5% vinegar, 0.6% sodium hypochlorite, 3% hydrogen peroxide, and brushing with distilled water, led to long-term changes in the light transmittance of Essix C+ retainers. Consistent with these previous findings, this study showed no differences in color change among the six cleaning protocols.

Although overall stain removal efficacy was comparable across groups, the most perceptible visual improvements were observed following cleaning with ICCs. From a clinical perspective, the most important observation is that post-cleaning ΔE₀₀ values in all groups remained well above established perceptibility and acceptability thresholds [17]. Although visually perceptible improvement occurred, none of the protocols restored the aligners to an optically acceptable baseline. This finding underscores that stain reversal may be inherently limited once chromogens have interacted with the polyurethane matrix. However, the material used, sample morphology, and the specific ultrasonic device employed may influence the comparability of the results. The limited additive effect of ultrasonic activation may partly relate to the low rated power (0.6 W; approximately 3 W/L) of the consumer-grade device evaluated. Cavitation intensity depends on acoustic energy delivery, and lower power densities may generate insufficient shear forces to meaningfully disrupt pigment–polymer interactions. Moreover, ultrasonic devices are primarily marketed as hygiene-enhancing tools rather than pigment-targeted systems, and their effectiveness may therefore depend on material composition, stain characteristics, and device-specific parameters.

Perhaps the most clinically significant finding of this study is that none of the tested cleaning protocols were able to restore the stained aligners to an optically acceptable state. The post-cleaning ΔE₀₀ values in all groups remained substantially above the established 1.8 clinical acceptability threshold. This indicates that while daily cleaning is essential for hygiene and may partially improve aesthetics, it cannot fully reverse the discoloration caused by extended exposure to strong pigments like coffee. From a clinical perspective, this underscores the critical importance of patient education on avoiding staining substances during aligner wear, as prevention remains more effective than any available cleaning method for maintaining aligner transparency.

The superior net color stability of the peroxide-free ICC (ΔE_t0–t2), despite its comparable stain removal efficacy (ΔE_t1–t2) to CP, suggests that the optical outcome is not solely determined by the amount of pigment extracted. It is plausible that the hydrogen peroxide in CP, while effectively breaking down organic stains, may induce microstructural alterations on the polyurethane surface. Such oxidative damage can increase surface roughness or create sub-surface voids that scatter light, leading to a clinically perceptible loss of translucency and an overall darker or more opaque appearance. In contrast, the detergent-based action of ICC might be gentler on the polymer matrix, preserving its optical integrity even if some pigment remnants persist. We hypothesize that this chemical interaction is a key factor, though future studies combining colorimetric analysis with surface profilometry and scanning electron microscopy are necessary to confirm this mechanism and directly correlate chemical cleaning action with surface degradation.

The difference in efficacy, where ultrasonic agitation improved the peroxide-based cleaner but hindered the non-peroxide crystals, may be attributed to distinct physical-chemical interactions, though these remain speculative and require further investigation. Firstly, regarding the negative interaction with the ICCs, we hypothesize that an interference between chemical and physical processes may have occurred. The turbidity produced by the dissolution of the crystal salts could have dampened the ultrasonic waves, preventing effective cavitation [14]. Furthermore, the same turbulence might have caused the redeposition of dissolved pigment particles back onto the aligner surface before they could be rinsed away, explaining the increase in ΔE₀₀ values in this group. Secondly, regarding the improvement in the CP group, the specifications of the household ultrasonic cleaner could have been a limiting factor. Although 50 kHz represents the upper frequencies commonly avalaible in these devices, this frequency may not generate cavitation bubbles with sufficient energy to disrupt strongly bound organic coffee chromogens that form physicochemical interactions with the polyurethane matrix [20, 21]. Finally, the relatively smooth surface topography of the aligners might have limited the necessary turbulence for optimal cavitation bubble formation and implosion.

The complex internal surface topography of the aligners, which mimics tooth morphology, also may have played a role in limiting the efficacy of ultrasonic cleaning, potentially through several hypothesized mechanisms [22]. It is possible that a “cavitation trap” formed within the narrow and deep fissures, where cavitation bubbles likely collapsed away from the target surface, resulting in the dissipation of cleaning energy. An “acoustic shielding” effect may have also occurred, where the aligner’s geometry physically blocked ultrasonic waves from reaching recessed areas such as the lingual surfaces and deep grooves, creating zones where no cleaning occurred. Restricted fluid circulation may also have been a limiting factor, as the confined volume inside the aligner impeded the replenishment of fresh cleaning solution and the removal of debris, thereby diminishing the impact of ultrasonic turbulence. It appears that the hard-to-reach recessed areas, which are most prone to coffee stain accumulation, were the most affected by these limitations. However, because only a two-molar segment was evaluated rather than a full-arch aligner, the extent to which these mechanisms occurred across the entire appliance remains a limitation of the present study.

Torlak and Sert [21] demonstrated that the combination of benzalkonium chloride and ultrasound markedly reduced Listeria monocytogenes biofilms on plastic surfaces, whereas ultrasound or disinfectant alone produced weaker effects. Similarly, Lombardo et al. [14]. demonstrated that the most effective method for biofilm reduction was a combination of cationic detergent and high-frequency ultrasound at 42 kHz. This aligns with the concept that ultrasonic cavitation can enhance antimicrobial efficacy by disrupting biofilm architecture and facilitating chemical penetration. Yet, these advantages seem to relate primarily to microbiological hygiene and not optical improvement. Even in the study by Lombardo et al. [14]., improvements were observed in biofilm removal rather than in color restoration. In contrast to microbial biofilms, coffee-induced staining involves chromogenic molecules capable of diffusing into superficial polymer layers and the formation of stable physicochemical interactions, which may not be fully disrupted by cavitational forces alone. Accordingly, while ultrasound may enhance disinfectant activity, the present findings suggest limited benefit in overcoming pigment-polymer interactions, which explains the absence of significant differences in the cleaning efficacy among the protocols.

Bernard et al. [23]. reported substantial discoloration of polyurethane-based retainers, following coffee exposure and suggested that ultrasonic cleaning may contribute to superficial stain reduction. However, the effectiveness of pigment removal was more closely related to the cleaning chemistry than the ultrasonic action itself. Similarly, other studies [5, 24] emphasized that ultrasonic washing can remove loosely bound surface pigments but does not prevent deeper penetration or fully reverse discoloration. These observations align with the present findings, in which cleanser formulation was the primary determinant of final color stability. In clinical settings, aligner discoloration is not limited to liquid chromogens. Cigarette smoke and aerosolized pigments have also been associated with clinically relevant color alterations in dental polymers. Smoke exposure has been associated with clinically relevant color changes in dental polymers, and experimental protocols highlight substantial variability in smoke-staining simulation methods. Therefore, the present coffee-based model should not be extrapolated to smoke-related discoloration without further investigation. Future studies incorporating combined staining models (liquid chromogens and smoke/aerosol exposure) would provide a more comprehensive representation of real-world conditions.

A research on pacifiers, toothbrushes, and dentures has shown that certain combinations of disinfectants and mechanical energy can enhance antimicrobial activity [25]. Yet, these materials differ substantially from polyurethane aligners in terms of surface morphology and physicochemical behavior. Acrylic resin appliances may respond favorably to ultrasonic agitation due to their rougher surface morphology, whereas thermoplastic aligners exhibit different cavitation behavior and are more vulnerable to optical changes when exposed to oxidative agents. Accordingly, the mixed evidence from removable appliance literature cannot be directly extrapolated to thermoplastic aligners. Ultrasonic cleaners did not provide a significant advantage for pigment removal or optical restoration in the stained thermoplastic aligners in this study. Furthermore, the superior net color stability (ΔE_t0−t2) of the peroxide-free ICCs compared with the peroxide-containing CP, despite removing a comparable amount of stain, is consistent with reports that oxidative cleaners can induce light-scattering surface alterations in polyurethane materials.

Overall, while ultrasound may improve antibacterial cleaning when combined with cationic agents [14], it offered no meaningful benefit in reversing coffee-induced discoloration in the present experimental conditions. The findings suggest that optical outcomes are primarily governed by cleanser chemistry and pigment-polymer interactions rather than by cavitational agitation alone. From a practical standpoint, the selection of an appropriate cleaning agent appears more critical than adjunctive use of household ultrasonic devices for maintaining aligner transparency.

Several limitations must be acknowledged when interpreting these findings. First, this study was conducted using an in vitro model, which does not accurately represent the intricate oral environment, including salivary dynamics, temperature fluctuations, and mechanical wear. Second, discoloration was induced exclusively using a continuous seven-day coffee immersion model, representing an accelerated staining protocol rather than intermittent clinical exposure. Although coffee was selected as a standardized high-chromogen medium to ensure reproducibility, other staining agents may exhibit different adsorption and diffusion behaviors. Furthermore, this study evaluated only a single home-use ultrasonic device operating at 50 kHz. Given that ultrasonic efficacy depends on frequency, power density, transducer configuration, and bath characteristics, these findings should be considered device-specific. In addition, surface roughness and microstructural changes (e.g., profilometry or scanning electron microscopy) were not assessed. Therefore, potential cleaning-induced surface degradation and its contribution to optical changes could not be directly quantified. Finally, although the study was powered to detect medium effect sizes, smaller differences may have remained undetected and should be interpreted with caution. To confirm these results, further investigations incorporating surface characterization techniques, repeated cleaning cycles, combined staining models, and well-designed clinical trials are necessary to further elucidate the relationship between cleaning protocols and long-term optical performance of thermoplastic aligners.

Conclusions

Within the limitations of this in vitro study, neither the use of household ultrasonic cleaning nor the duration of ultrasonic exposure significantly enhanced the stain-removal performance of peroxide-free or peroxide-based cleaning agents. Importantly, none of the tested protocols restored stained aligners to a clinically acceptable optical condition, as post-cleaning ΔE₀₀ values remained well above established perceptibility and acceptability thresholds.

While the peroxide-free formulation exhibited superior net color stability compared with the peroxide-based cleaner, the addition of ultrasonic agitation did not significantly alter the efficacy of either agent. Overall, both cleaning approaches demonstrated comparable efficacy in reducing coffee-induced discoloration, and ultrasonic agitation did not provide an additional benefit.

Clinically, these findings reinforce the importance of advising patients to minimize exposure to staining agents, as prevention remains the most reliable strategy for maintaining aligner transparency. The present findings are specific to the tested consumer-grade 50 kHz device and should not be generalized to other ultrasonic systems. Further investigations, particularly clinical studies incorporating variable staining diets, longer cleaning cycles, and higher-power ultrasonic devices, are warranted to clarify the potential synergistic effects between chemical cleaners and ultrasonic energy on aligner color stability.

Authors’ contributions

Yağmur Lena Sezici, Aslı Aşık, Genta Agani Sabah, Sibel Acar and Enver Yetkiner contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yağmur Lena Sezici, Aslı Aşık and Genta Agani Sabah. The first draft of the manuscript was written by Yağmur Lena Sezici, Aslı Aşık and Genta Agani Sabah and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors certify that there are no financial organisations associated in relation to this study.

Data availability

The entirety of the data analyzed in this study has been included in the published article. Upon making a reasonable request, the corresponding author will provide access to the raw data.

Declarations

Ethics approval and consent to participate

This study did not involve any biological materials; instead, it conducted experiments solely on dental materials. Therefore, there was no need for ethics committee approval.

Consent for publication

The present study did not involve human participants.

Competing interests

The authors declare no competing interests.