Titanium dioxide nanotubes incorporated into bleaching agents: physicochemical characterization and enamel color change

Abstract Titanium dioxide nanotubes are nanostructures that can accelerate the oxidation reaction of bleaching procedures and promote a more effective whitening effect. Objective This study evaluated physicochemical properties of bleaching agents incorporated with titanium dioxide (TiO2) nanotubes, and the effects on tooth color change at different periods. Methodology 40 premolars were treated according to the following groups (n=10): CP - 10% carbamide peroxide (1 hour daily/21 days); CPN - CP incorporated into TiO2; HP - 40% hydrogen peroxide (three 40-minute sessions/7 days apart); HPN - HP incorporated into TiO2. Color shade was evaluated at five different periods (baseline, after 7, 14 and 21 days of bleaching, and 7 days after end of treatment) according to Vita Classical, CIELab and CIEDE2000 scales. Mean particle size (P), polydispersity (PO) and zeta potential (ZP) were evaluated using dynamic light scattering. Data on the different variables were analyzed by mixed model tests for measures repeated in time (ZP e L*), generalized linear models for measures repeated in time (P, PO, Vita Classical and b*), and Friedman and Mann-Whitney tests (a* and color change/ΔE and ΔE00). Results CP and CPN presented higher P, higher PO and lower ZP than HP and HPN (p≤0.05). All groups showed a significant decrease in Vita Classical color scores after 7 days of bleaching (p<0.05), and HPN presented a greater significant reduction than the other groups. L* increased in TiO2 presence, in all groups, without any differences (p>0.05) in bleaching time. A significant reduction occurred in the a* and b* values for all the groups, and HPN presented lower a* and b* values (p<0.05) than CPN. ΔE was clinically noticeable after 7 days, in all groups, and all groups resulted in a perceptible color change according to ΔE00. Conclusion TiO2 did not influence physicochemical properties of the bleaching agents. HPN presented more effective tooth bleaching than CPN.


Introduction
Dental sensitivity is one of the adverse effects directly related to whitening treatment [1][2][3][4][5][6] and to concentration of the whitening agent, as well as time frame, application technique and composition of the whitening agent. [1][2][3][4][5][6] High concentrations of hydrogen peroxide seem to increase the penetration and diffusion of peroxides and their by-products, leading to more intense inflammatory response and/or greater sensitivity. [1][2][3][4][5][6] Some recommended methods to minimize sensitivity include using lower concentrations of bleaching agents, 6 shortening the application time 7 and introducing catalysts or accelerators of the oxidation reaction, such as sodium, potassium, iron, and manganese compounds, as well as titanium and graphene-based nanoparticles. [8][9][10][11] The use of catalysts or accelerators could provide a more pronounced bleaching effect in a shorter time, thus, achieving more efficient bleaching results, while requiring a lower hydrogen peroxide concentration.
Among the components used as catalysts for the oxidation reaction, titanium dioxide (TiO 2 ) is biocompatible and it has antimicrobial properties, 12 in addition to its known use as a pigment in cosmetics and food additives. 13,14 When associated to a light source, it excites electrons and generates oxygen ions, producing superoxide. 15 The catalytic action of TiO 2 has strong oxidative power and high chemical stability, and being influenced by reactions with molecules on its surface. This process can be influenced by several factors, such as catalyst concentration, particle size and shape, pH, and surface area of the substrate. 15,16 Because of the catalytic action of TiO 2 , some studies 17,18 have found that its incorporation in inoffice bleaching agents with lower concentrations of hydrogen peroxide (3.5%) and irradiation by light can promote a greater release of free radicals, and, consequently, more effective and clinically noticeable bleaching effect than non-incorporation of TiO 2 .
However, it may be suggested that the incorporation of TiO 2 in higher hydrogen peroxide concentrations than those used for the in-office bleaching technique could decrease the application time, and prevent against the common effect of dental sensitivity related to this procedure.
In most studies, [19][20] TiO 2 particles incorporated into bleaching agents have a spherical nanostructure, which tends to form clusters 12 that may compromise the effectiveness of the agent. However, TiO 2 nanotubes, characterized by a hollow structure and a high relationship between surface and particle volume, 15,21 could offer more benefits regarding the catalytic reaction promoted by the nanoparticle, even in the absence of light. 16 This shape of the nanotube thus provides a larger surface area, and produces a greater number of radicals resulting from the increase in amount of active reaction sites for radical production.
This suggests that the long one-dimensional structure of nanotubes is a better option. 15 By adding a new component to the bleaching gel, like TiO 2 nanotubes, the resulting physicochemical characterization may be altered. This could lead to sedimentation or aggregation of particles, which, in turn, could influence the expected effectiveness.
Thus, this study aims to evaluate physicochemical characteristics of whitening agents containing 10% carbamide peroxide and 40% hydrogen peroxide when incorporated in TiO 2 nanotubes, and the effects on the color change of the dental structure at different periods. The null hypotheses evaluated were that the addition of TiO 2 nanotubes to whitening agents containing 10% carbamide peroxide or 40% hydrogen peroxide did not influence: 1) pH properties, particle size, polydispersibility or zeta potential of the whitening agents; 2) color alteration of the dental structure at different whitening periods.

Methodology
Selection of teeth, specimen preparation and distribution among groups J Appl Oral Sci. 2020;28:e20190771 3/11 to the amelocemental junction with self-curing acrylic resin (VipiFlash, Vipi, Pirassununga, SP, Brazil), and the long axis of the tooth was kept perpendicular to the horizontal plane. PVC molds were removed after resin polymerization.
A spectrophotometer (VITA Easyshade® Advance, Vita, Germany) was used to perform initial color evaluation applying the Vita Classical scale as criterion, after which the very light or very dark teeth were excluded. A total of 40 teeth were selected with colors that included A2, A3, B3, and A3.5, and they were distributed homogeneously among the four groups investigated (n=10): CP-10% carbamide peroxide;

CPN-10% carbamide peroxide incorporated with
TiO 2 nanotubes; HP-40% hydrogen peroxide; and HPN-40% hydrogen peroxide incorporated with TiO 2 nanotubes. The teeth were maintained at 37°C for 7 days in a bacteriological incubator before the start of the experiment, and then individually immersed in an artificial saliva solution (20 mL), composed of 1.5 mMol/L Ca; 50 mMol/L KCl; 0.9mMol/L PO 4 ; 20 mMol/L Tris buffer (pH=7). 22 Specification of bleaching agents, obtaining of TiO 2 nanotubes, and incorporation The bleaching agents used in this study, their compositions, time of daily use, and total treatment time, are described in Figure 1.
TiO 2 nanotubes were formed from a single rolled sheet (~10 nm diameter and ~200 nm long), and they were produced in spiral structures; they were synthesized using the alkaline method. 23 Nanotubes were prepared by mixing 12 g of TiO 2 anatase phase, 99% purity, with 200 mL of 10 M NaOH. This mixture was maintained at 120°C for 24 h in an open Teflon container, placed in a glycerin bath, and heated with a mantle heater. The syntheses were carried out at ambient pressure, where only precursor reagents were subjected to alkaline treatment. After treatment, the mixture was washed repeatedly with 0.1 M hydrochloric acid and deionized water to remove the sodium ions. Next, the pH of the solution was adjusted to 7. 23 TiO 2 nanotubes were weighed on a 0.1 mg precision scale (BEL Engineering, Monza, Milan, Italy) and added manually to 9.9 mg of bleaching agents (CP or HP), at a concentration of 1%. 18,20 The nanotubes were incorporated into the bleaching agents by spatulation (plastic spatula) on waterproof paper for 1 minute.
Then, the resulting content was deposited into a sterile, disposable syringe. The TiO 2 was incorporated into the bleaching agents right before each application.
Physicochemical characterization of bleaching agents: pH analysis, average particle size, polydispersibility and zeta potential The pH values were obtained in triplicate for the bleaching agents, according to respective measurement periods. The pH of the carbamide peroxide-based agents was measured initially after 30 minutes (half of the application time), and lastly after 1 hour (total application time). The pH of the hydrogen peroxide-based agents was measured initially after 20 minutes, and lastly after 40 minutes (total application time). The pH of both agents was measured by a or without TiO 2 were necessary for these analyses.
The zeta potential was evaluated to measure colloidal stability, using Helmholtz- Smoluchowski model,25 which measures the electrophoretic mobility of dispersed particles in the applied electric field.
The zeta potential analysis was performed by laser electrophoresis, with 30 runs per measurement at 25°C. The zeta potentials were estimated automatically using electrophoretic mobility, with the Smoluchowski approach: UE=2*ε*z*f(ka)/3*η→z≈UE*η/ε, where UE is the electrophoretic mobility, ε the dielectric constant, z is the zeta potential, f (ka) is the function of Henry, and η is the viscosity. 25 The analyses (particle size, polydispersibility, and zeta potential) were performed in triplicate at 25°C and three times after the preparation of the bleaching agents with or without the addition of TiO 2 nanotubes at 1%, namely: immediately after, 14 days, and 28 days.

Bleaching protocol
The teeth were removed from the artificial saliva solution before each application of bleaching agent Each tooth from HP and HPN groups received a 0.03 mL layer of bleaching agent on the vestibular surface. The agent remained on the dental surface for 40 minutes without replenishing the gel. 3,4 At the end of the process, the gel was removed from the dental surface using a surgical aspiration tip, cleaned with gauze and rinsed with water for 10 seconds.
The bleaching treatment was performed twice again, at 7-day intervals between sessions (total of three sessions of bleaching application). After completing the bleaching treatment, all teeth were stored in artificial saliva solution for 7 days, changing the solution every two days.

Tooth Color Shade Evaluation
A spectrophotometer (VITA Easyshade ® Advance, Vita, Germany) was used to measure color at the middle third of the labial surface of the teeth. Tooth color was verified using the Vita Classical shade guide, and parameters L*, a*, and b* from the CIEL*a*b* system at different times, namely: T1-before bleaching procedure (baseline); T2-after 7 days of bleaching procedures; T3-after 14 days of bleaching; T4-after 21 days of bleaching; T5-7 days after the end of bleaching. The specimens were positioned inside a box with a white background for standardization of illumination, to perform the color measurements, conducted by the same rater at all times.
These measurements were duplicated at each period to ensure accuracy. When the two readings were the same for the Vita Classical scale, the value and other parameters (CIEL*a*b*) were recorded after the second reading. If the two readings did not match, a new measurement was carried out until agreement was achieved between readings.

Statistical Analysis
The sample (n=3) for particle size, polydispersity and zeta potential data provided a power of 0.80 to test a 5% level of significance and an effect size higher than 1.0. 28 The sample size for color evaluation (n=10) also provided a power of 0.80 to test for a mean size effect higher than 0.47 for color parameters (Vita Classical, L*, a* and b*) and 1.4 for color change Titanium dioxide nanotubes incorporated into bleaching agents: physicochemical characterization and enamel color change J Appl Oral Sci. 2020;28:e20190771 5/11 (ΔE and ΔE 00  Data on zeta potential were analyzed by mixed models for measures repeated in time, and they were presented as means and standard deviations.
Generalized linear models for measures repeated in time were used, since particle size and polydispersity data did not meet the assumptions of parametric analysis, which were presented as medians, and minimum and maximum values. All the analyses were performed with SAS 30 and R 31 software tools, at 5% significance level.

Results
Bleaching agents presented pH value close to neutral, which remained similar and stable over the time of application, regardless of TiO 2 incorporation ( Table 1).
The mean particle size showed stability in CPN group during the evaluation time (Table 2). HPN showed a decrease in mean particle size over a 14day period, and a subsequent increase at 28 days (p=0.0037), without a significant difference from the initial time. The mean particle size was significantly larger in the CP group (p=0.0007), regardless of period and presence or absence of TiO 2 . As for polydispersibility ( The Vita Classical scale color score decreased significantly for all groups after 7 days, but there were no differences between 7 and 14 days (Table 3). After 21 days, a significant decrease occurred in the score related to the previous period (14 days) only for CP.
At 7 days, after the end of whitening, only CP and HP showed a significant decrease in the color score, in relation to the 7-day whitening period. At the period of 21 days of whitening and 7 days after whitening, CPN presented a higher color score than CP. At 7 days of whitening, HPN presented a lower color score than HP. At the periods of 7, 14, 21 days and after 7 days of whitening, HPN presented a score significantly lower than CPN. There was an increase in L* (Table 3) in all groups, especially after 7 days of evaluation, and a significant increase especially in CP and HP after 14 days. However, no significant differences occurred among the groups at any evaluation period.
Over time, all groups presented a significant decrease in mean/median values for a* and b*. There were no significant differences among groups at the evaluation periods, except CPN, that presented a significantly higher b* value than CP, in 21 days of whitening. HPN had a significantly lower b* value than CPN at 14, 21 days and after 7 days of whitening. The CPN group had a b* value similar to the baseline at 21 days of whitening, whereas HPN group had a b* value significantly lower than that of the baseline at 14-day period, and similar to that obtained 7 days after whitening.

Bleaching agent Time
As for ΔE and ΔE 00 (Table 4), they were significantly higher (p<0.05) in HPN than CPN at 7 days after whitening in relation to the baseline time, but ΔE was not significantly different among the groups at the other periods. Meanwhile, ΔE 00 was significantly higher in HP than HPN only at 14 days of bleaching treatment (p<0.05).

Discussion
When TiO 2 nanotubes were incorporated into the bleaching agents, the pH was stable during the application time, and the gel pH remained close to neutral. Such as gel composition and temperature influence hydrogen peroxide decomposition, 32,33 pH value can also affect the bleaching efficacy. 30,32 It is well-known that the bleaching efficacy of hydrogen peroxide is directly proportional to pH, and that pH increases as the speed of the reaction to hydrogen peroxide decomposition increases, triggering the release of free radicals. 34 The higher the pH, the greater the dissociation of hydrogen peroxide, leading to greater formation of reactive free radicals that improve bleaching efficacy. 35 On the other hand, a more acidic pH promotes the stability of hydrogen peroxide, keeping it from decomposing, and favoring bleaching agent longevity during storage. 34 Although it has been observed that the lower the pH of the bleaching agent, the greater the diffusion of peroxides in the dental structure, there is also a greater risk and intensity of dental sensitivity, compared with a whitening gel of more alkaline pH. 35,36 The incorporation of TiO 2 at a concentration of 1% also produced a gel with clinical handling characteristics, regarding viscosity, similar to those of the original commercial product. The properties of polydispersity and colloidal stability (zeta potential) were similar between gels. This indicates that there was no tendency of particles agglomeration in the whitening gel-leading to greater material stability- Groups with medians or means followed by distinct letters (uppercase horizontally and lowercase vertically, comparing groups with and without nanotubes for each bleaching agent) differ from each other (p≤0.05). * Values differ from CP, in reference to nanotubes and timepoints (p≤0.05).   Table 4-Median (minimum value; maximum value) color variation (CIEL*a*b* system and CIEDE2000) as a function of group and time period 2020;28:e20190771 8/11 carbamide peroxide-based agents had larger particle sizes and polydispersibility, and lower zeta potential than the hydrogen peroxide-based agents. However, the addition of TiO 2 to carbamide peroxide led to an agent with smaller mean particle size, although it did not influence the product's polydispersibility characteristic.
The polydispersibility index is a parameter that defines the distribution of particles in the material, where greater uniformity of distribution is evidenced by lower values of polydispersion. 37,38 The polydispersibility values for HP and HPN agents were significantly lower at all evaluation periods than those for CP and CPN agents, thus contributing to the lower probability of formation of agglomerates in the bleaching gel. In this sense, values of zeta potential above (+/-) 30 mV indicate stable molecules in suspension, since the surface load prevents the aggregation of particles. 39 In this study, the zeta potential values showed good colloidal stability for all agents, with higher values for agents containing CP, considering that it is desirable for a system to present high surface load by generating repulsive forces that tend to prevent the aggregation of particles. 38 Note that mean particle size, polydispersibility and TiO 2 in its crystalline form can be found in three different phases, namely: anatase, rutile, and brookite. The anatase phase can be transformed into nanotubes, which is relevant as a catalytic agent and photocatalyst. 15,23,41 The anatase phase of TiO 2 nanoparticles is stable; 23 this may also have contributed to pH properties, mean particle size, polydispersibility and zeta potential. Although no light source was used for photocatalysis of the agent, TiO 2 particles at nanoscale may be reactive, even in the absence of activation sources, 16 especially when associated to 40% hydrogen peroxide. 20 The catalytic action of TiO 2 has strong oxidative power and high chemical stability, and it is influenced by reactions with molecules on its surface. 15 However, TiO 2 nanotubes shape provides larger surface area, increasing the amount of superoxide radicals produced, such as O 2and hydroxyl radicals. 15 This is expected to improve color change when associated to hydrogen peroxide agents. This finding was also identified in the study by Sakai, et al. 17 (2007), who showed that gel with hydrogen peroxide at 3.5% and TiO 2 (less than 1%), even without diode laser activation (405 nm), generated hydroxyl radicals, albeit in significantly smaller amounts than the irradiated group.
On the other hand, association of TiO 2 to carbamide peroxide did not seem to show any potentiation of the bleaching effect when evaluating Vita Classical scale, since the reduction in the color score after 7 days was only 2 shades, after which the color remained stable throughout the treatment; moreover, this association proved less effectiveness than 10% carbamide peroxide at the end of the bleaching treatment. Lee, et al. 40 (2005)  hydrogen peroxide concentration, showing that the improvement in bleaching efficacy may be associated to the hydrogen peroxide concentration and the amount of radicals generated with the incorporation of TiO 2 nanotubes. Nevertheless, the addition of TiO 2 at 0.1% in the same study 20 was able to improve the performance of the bleaching gel, regardless of the hydrogen peroxide concentration.
Regarding the evaluation of color change according to CIEL*a*b* system, an increase in luminosity (L*) was observed in all groups, especially after 7 days. Cardoso, et al. 7 (2010) showed that patients with a bleaching treatment using 10% carbamide peroxide with daily applications of one-hour were as satisfied as those who used it for eight hours a day. The color change (ΔE) values obtained by all the techniques were higher than 2.7 (minimum value of ΔE that is clinically perceptible), compared to the baseline, denoting the effectiveness of all bleaching procedures. 3,7 However, the color change obtained by the CIEDE2000 formula (ΔE 00 ) is better correlated with visual perception than the CIELAB. 26 In this study, ΔE 00 values between baseline and 7 days after end of the bleaching treatment (above 4.07) were higher than the 50:50% perceptibility threshold, 26 showing their effectiveness in dental bleaching. By incorporating TiO 2 nanotubes (10 mg) into the hydrogen peroxide hydrogen peroxide with another containing 6% hydrogen peroxide + TiO 2 nanotubes doped with nitrogen (irradiated with hybrid LED/Laser light), they observed a similarity between the groups in the subjective analysis. Regarding the objective analysis, a remarkable difference of more than 2 ΔE units was observed, with higher values for the gel of higher concentration. In our study, the presence of TiO 2 in the hydrogen peroxide agent promoted more significant color change than that of the CPN group, thus leading to the rejection of the second null hypothesis, since the addition of TiO 2 nanotubes to the whitening agents influenced the color change of the dental structure at different whitening time-points. This effect may depend on the concentration of nanotubes, and whitening agent. The performance of TiO 2 nanotubes may have been enhanced by the higher hydrogen peroxide concentration (40%).
The results of this study suggest that the inclusion of TiO 2 nanotubes seems to be promising (especially