The effect of in-office bleaching agents on the Vickers hardness and surface topography of polished and unpolished CAD/CAM composite materials

In-office bleaching, using hydrogen peroxide, is effective to remove dental enamel stains. However, bleaching agents can deteriorate surface properties of CAD-CAM materials. This in vitro study aimed to investigate the effect of in-office bleaching agents on Vickers hardness and surface topography of polished and unpolished dental CAD-CAM composite materials (Grandio blocs, Lava Ultimate, BRILLIANT Crios, Cerasmart), and a polymer-infiltrated ceramic network block (Vita Enamic). The specimens were randomly divided into two groups: unpolished or polished. The micro-hardness and surface topography of each group were measured before bleaching, after a 60 min bleaching period, and 24-h and one-month post-bleaching. In-office bleaching significantly influenced the Vickers hardness of both the polished and unpolished CAD/CAM composite blocks, with Vita Enamic exhibiting the least hardness stability among all groups. Furthermore, in-office bleaching significantly influenced the surface roughness of unpolished CAD/CAM composite blocks. There was a significant difference in hardness reduction between the polished and unpolished specimens for most of the investigated materials at different time points. The bleaching did not influence the surface roughness of the investigated polished group, except for Vita Enamic and Lava Ultimate. However, it did influence the surface roughness of the investigated materials in the unpolished group.


Materials and methods
Sample preparation.Five CAD/CAM blocks were investigated: four resin-composite blocks (RCB), Grandio blocs (Gr), Lava Ultimate (Lu), BRILLIANT Crios (Bc) and Cerasmart (Cs); and one polymer-infiltrated ceramic network (PICN) block Enamic (En), as shown in Table 1.They were selected due to their wider applications as representative CAD/CAM blocks for aesthetic restorations.
The CAD/CAM blocks were sectioned into rectangular bar shaped specimens (14 mm × 12 mm × 2 mm) using a cutting machine (MECATCH234, PRESI, France) under constant water cooling (Fig. 1).The specimens were wet-ground and polished using a polishing machine (BETA-VECTOR, Buehler, IL, USA) with silicon carbide Table 1.The manufacturers' compositional information and experimentally determined filler weight percentage 63 of the materials investigated, and bleaching agent used.

Materials (code) Manufacturer
Composition Bleaching procedure.Thirty samples of each material (15 polished and 15 unpolished) were randomly distributed into three study groups: four samples were used as control (unbleached), four for Vickers hardness, four samples for surface roughness, and three for SEM analysis.All samples were stored in distilled water at 37 °C for 3 months 40 (ISO 10993-13, 2010) before the bleaching procedure was performed using Opalescence Boost (in-office bleaching) for three intervals of 20 min (total 60 min).The bleaching procedure was performed by one investigator at room temperature, with the bleaching agent covering the whole top surface of each specimen.At www.nature.com/scientificreports/ the end of the respective bleaching procedure, the specimens were rinsed with a high-pressure water flow, dried with blotting paper and airflow, and placed in fresh distilled water for subsequent measurements.
Micro-hardness measurement.Micro-hardness of each specimen (n = 4) was measured using a Vickers micro-hardness instrument (HXD-1000 TMC, Shanghai Taiming Optical instrument, Shanghai, China).A fixed load of 300 gf was applied for 20 s.Ten indentations at a 0.5 mm interval were made randomly on each sample.The machine then automatically calculated the corresponding hardness value and presented the Vickers hardness number (VHN).For in-office bleaching, surface micro-hardness was measured before bleaching, after 60 min of bleaching and at 24-h and 1-month post-bleaching.VHN was also measured for the corresponding control samples from each group.The samples were cleaned and dried before the micro-hardness measurement.
The hardness reduction (HR as a percentage) after bleaching storage was calculated using Eq.(1).
where VHN( before ) and VHN( after ) indicate the Vickers hardness numbers before and after bleaching respectively for each time point.

Surface roughness.
The surface roughness of each specimen (n = 4) was measured using a stylus contact type profilometer (Mahr Perthometer M1 [L t = 5.6 mm, λ c = 0.800 mm]).R a , the arithmetic mean height of the roughness profile, was recorded during the measurement.Three vertical and three horizontal lines were measured on the specimen surface to record six R a values for each specimen, and the mean R a result was subsequently calculated.Surface roughness was also measured before bleaching and at the same timeframes after bleaching and post-bleaching similar to hardness measurement.
Scanning electron microscope imaging.Each specimen was attached to aluminium stubs (12 mm thickness) with double-sided adhesive tabs.The specimens were sputter-coated with gold and examined with a scanning electron microscope (FESEM, S-4800, Hitachi, Tokyo, Japan) (n = 3).SEM micrographs were produced at × 40k magnifications of representative areas of the samples.

Statistical analyses.
The data were analysed using a statistical software package (GraphPad Prism, version 9.1.2(226) and found to exhibit normal distribution according to the Shapiro-Wilk test.Two-way ANOVA was performed to investigate the material's effect, storage time effect, and their interaction.One-way ANOVA was followed by the Tukey's post hoc analysis for multiple comparisons between different materials and different time points.The Pearson correlation was determined to identify the correlation between the hardness reduction and filler weight percentages, in both the polished and unpolished groups.An independent t-test was performed to measure differences between the polished and unpolished groups (without bleaching, after 60 min of bleaching, 24 h and one-month post-bleaching) and for differences between the bleached and unbleached groups at each time point for each material.All tests were carried out at α = 0.05.

Results
The reduction of hardness percentage was calculated 60 min after bleaching, after 24 h and after one-month and for the control unbleached samples at the same time points.There was a significant difference between the bleached and unbleached (control) specimens for each material, as confirmed by the independent sample t-test (Table 2 and Fig. 2).After 60 min bleaching of the polished samples, the highest reduction of hardness was found for En, followed by Gr, Cs, Bc and Lu, with significant differences between all materials except Cs and Bc.The same was observed for the unpolished samples in the following order: En, Lu, Gr, Bc and Cs.There were significant differences between all the materials.Notably, there was an insignificant difference in the reduction of hardness between the polished and unpolished Gr and Cs specimens (Table 2).
After 24 h bleaching of the polished samples, the reduction of hardness was the highest for En followed by Gr, Lu, Cs and Bc, with significant differences between all the materials.The same was observed for the unpolished samples in the following order: En, Gr, Lu, Bc, and Cs.A significant difference was observed between all the materials.Whereas, remarkably, there was a significant difference in hardness reduction between the polished and unpolished samples for all investigated materials.
One-month post-bleaching, the reduction of hardness of the polished samples was the highest for En followed by Lu, Gr, Cs and Bc, with a significant difference between all the materials except Cs and Bc.The same was observed for the unpolished specimens in the following order: En, Lu, Gr, Bc and Cs.A significant difference was observed for all the materials except Cs and Bc.Notably, there was a significant difference in the reduction of hardness between the polished and unpolished samples for all the investigated materials except En.
Two-way ANOVA showed a significant material effect and time effect, including a significant interaction for hardness reduction (p < 0.0001) with more influence of the material effect.The greatest influence in the polished group was for the CAD-CAM material (partial eta squared ηp 2 = 0.798) followed by the interaction effect (ηp 2 = 0.099) while time had the lowest effect (ηp 2 = 0.063).The greatest influence in the unpolished group was for the CAD-CAM material (partial eta squared ηp 2 = 0.84) followed by the time effect (ηp 2 = 0.11) while interaction had the lowest effect (ηp 2 = 0.05).
The Pearson correlation of the reduction of hardness and filler weight percentage (data obtained from a previous study 32 ) showed an insignificant positive correlation between filler weight and the reduction of hardness www.nature.com/scientificreports/ in the polished group R 2 = 0.7, p = 0.076 and filler weight and the reduction of hardness in the unpolished group R 2 = 0.62, p = 0.115.There was also an insignificant positive correlation in the reduction of hardness between polished and unpolished groups R 2 = 0.52, p = 0.166 (Fig. 3).R a values recorded before bleaching, directly after bleaching, and after one month are presented in Table 3 and Fig. 4. The bleaching treatment did not influence surface roughness for all the investigated materials in the polished group except En and Lu.En showed significantly increased R a values after bleaching from both before bleaching and at one-month post-bleaching.Lu from the polished group exhibited significantly increased R a values after a month.Bleaching influenced the surface roughness of all investigated materials in the unpolished group.En showed significantly reduced R a values after one-month post-bleaching from both before and after bleaching.Also Gr showed significantly reduced R a values after 60 min of bleaching and one month post bleaching from that of before bleaching.Lu showed significantly increased R a values after one-month post-bleaching and after 60 min of bleaching from that before bleaching.Cs showed significantly increased R a values after bleaching from both before bleaching and one-month post-bleaching.Finally, Bc showed significantly increased R a values after one month of bleaching compared to both before and after 60 min of bleaching.
SEM images at a × 40k magnification showed less demarcated particle margins and more rounded particles directly after bleaching and after one-month post-bleaching when compared with both the polished and unpolished samples (Figs. 5 and 6).Small cracks were found in the matrix phase of both polished and unpolished specimens, especially one-month post-bleaching.For the polished specimens, En and Gr showed deeper grooves, whereas Bc and Cs exhibited no remarkable changes after one month.Nevertheless, Lu presented more dark spaces (the polymer matrix) and more prominent clustering of filler particles.More remarkable surface deterioration and margin fractures were observed for the unpolished specimens than for the specimens in the polished group.Unpolished Cs and En did not exhibit remarkable surface morphology variations.

Discussion
In the present study there was a significant difference in hardness reduction percentages between the bleached and unbleached (control) specimens for each material, and thus, the first null hypothesis was rejected.Moreover, there were significant material and time effects, with a significant interaction for the reduction of hardness (p < 0.0001) with more influence from the material effect.Given that, the second null hypothesis was rejected as well.Markedly, there was a significant difference in the reduction of hardness between polished and unpolished specimens for all investigated materials, and thus, the third null hypothesis was rejected.The filler weight percentage showed an insignificant effect on the reduction of hardness, and thus, the fourth null hypothesis was accepted.The bleaching influenced the surface roughness of some of the investigated materials, and thus, the fifth null hypothesis was partially accepted.
En showed the highest hardness reduction of all investigated materials for both the polished and unpolished surfaces at all time points.This was an interesting finding as En is a PICN composed of an interpenetrating ceramic-polymer network that is strong and resistant to breakdown 41 .Previous studies have found that En has the least hardness reduction when compared to other RCBs upon storage in different media 42,43 .However, this contradicted the findings of some alternative studies, which, in contrary, showed an increase in the material microhardness 44,45 .This might be attributable to the fact that bleaching agents can penetrate and diffuse through the polymeric matrix of the material to a greater extent than the other storage media such as water, artificial saliva or 70% ethanol solution 43 due to oxidative cleavage of the polymer chains 46 or similar solubility factor to Table 2.The mean and standard deviation values of the reduction of hardness percentage (HR %) of polished and unpolished samples after 60 min in-office bleaching, 24 h and one-month post-bleaching.Vita™ Enamic (EN); Grandio™ Blocs (GR); Lava™-Ultimate (LU); BRILLIANT™ Crios (BC); Cerasmart™ (CS); Control (cont).Values with the same superscript letters represent a non-significant difference among different materials (the Tukey post hoc tests, α = 0.05), at each time point.Values with the same superscript asterisk represent a nonsignificant difference for polished/unpolished surfaces for each material at each time point; the independent sample t test.There was a significant difference for bleached/unbleached (control) samples for both polished/ unpolished surfaces for each material at each time point; the independent sample t test.www.nature.com/scientificreports/ that of the bleaching agent 47 .Furthermore, the samples in this study were stored in water for three months to simulate clinical conditions as such the materials would have been bleaching after servicing for a while in the oral cavity while the compared studies have measured bleaching effect without any previous storage in water.Nevertheless, RCBs were less affected by the in-office bleaching procedure.GR exhibited the highest reduction of hardness of all the resin-composite blocks.However, it has the highest filler percentage (86 wt.%).This, again, contradicted previous findings 42,48 and could be attributed to the aforementioned reason of different experimental conditions.Lu, Bc and Cs had similar filler weight percentages (75 wt.%, 70 wt.%and 66 wt.%) and the same level of the reduction of hardness.Lu contains more zirconium silicate in its filler composition; as such, it is more prone to hydrolysis of the silane-coupling agent as a consequence of inefficient salinization of high crystalline content in the zirconium silicate 49 or due to material inhomogeneity and presences of large filler particles 50 .Then again, silicate materials are more prone for silane-aided adhesion promotion than zirconia or zirconium silicate.However, silanes may degrade over time in resin composites 51 .Furthermore, nano-cluster filler particles in Lu typically have defects and voids, which could explain the higher levels of surface degradation 4 .Lu with zirconia nanoparticles has been found to exhibit higher softening resistance compared to En with a feldspathic ceramic structure.This is in line with studies reporting that zirconia-containing materials have higher hardness and strength values than feldspathic-containing materials 52,53 .Although bleaching agents can influence the surface microhardness of dental restorations, the mechanism for such changes is not fully clear.One possible explanation is that the peroxides in bleaching agents release very reactive free radicals at the resin-filler interphase.These free radicals may attack glass particles, silica, and alumina, causing a separation of fillers and filler-matrix debonding 54 .Alternative explanations are that the bleaching agents can cause an oxidative cleavage of the polymer chains 46 or that the resin matrix has a similar solubility factor to that of the bleaching agent 47 .Bleaching agents might have a greater effect on the surface microhardness at higher temperatures 28 .
All unpolished surfaces exhibited a significantly higher hardness reduction.When servicing in the oral environment and exposed to fluids, acids and moisture, the surface roughness of the materials tends to increase 9,[30][31][32] , rendering the restoration more prone to penetration by bleaching agents which affect their physical properties, such as microhardness 35,55 .Our finding might indicate that polishing dental restoration before bleaching procedures is advisable to reduce any potential softening of the restorative materials.
Two-way ANOVA showed significant material and time effects with a significant interaction for hardness reduction (p < 0.0001) with more influence from the material effect.Considering the different resin matrices  and filler compositions of CAD/CAM composite blocks, the bleaching agents can cause various changes in the material's microhardness and surface roughness.This indicates that bleaching effects are material dependent.The reduction of hardness was greatest after 60 min bleaching and increased slightly after 24 h and after one month.This finding is consistent with other studies, as most material changes occur in the first interaction with a solvent or contacting solution, here a bleaching agent, after which the material effect may reach an equilibrium [56][57][58][59] .Further, the bleaching agent might be diluted with a storage medium (deionized water) after performing the bleaching for the recommended time.In this study, the lowered hardness values were positively correlated with the amount of filler loading for all CAD/CAM composite materials, where En with the highest weight percentage of fillers (86%), exhibited least softening resistance.On the other hand, factors other than filler type and concentration, such as microstructure homogeneity, cross-linking between the fillers, and resin matrix, have a significant impact on the interaction between the bleaching agents and dental materials 54 .However, all the materials experienced hardness reduction over time due to plasticization of the polymer matrix caused by water sorption and consequently degradation and reduction in material stiffness 42,60 .
In this research, both profilometry and SEM were used to study the surface topography of the materials of interest.In-office bleaching did not influence the surface roughness of any of the investigated materials of the polished group except for En, which showed significantly increased R a values after 60 min bleaching when compared with the values before bleaching and one-month post-bleaching.Polished Lu specimens exhibited significantly increased R a values after one month.This finding was in line with similar studies 29,54 .This can be attributed to the Bis-GMA composition of Lu, as Bis-GMA-containing materials have higher surface roughness values 46 .Furthermore, for the unpolished group, the lower R a values of Lu might be attributed to the fact that it has smaller filler size than En or the highly cross-linked polymer matrix and zirconia fillers of Lu, which resist free radicals caused by bleaching agents 54 .
Furthermore, the bleaching agent and diverse dental materials may chemically interact in a way that changes the surface topography 54 .Surface roughness might increase due to the composition, exposure time, and concentration of bleaching agents 61 .The increased surface roughness of dental restorations can cause discolouration and secondary caries by enhancing and accelerating biofilm accumulation.The R a value should not exceed 0.2 μm to prevent biofilm accumulation 62 .This value had not been exceeded for all polished samples; however, it was exceeded for all the unpolished samples except CS.Bleaching influenced the surface roughness of for all the investigated materials in the unpolished group of specimens.
Compared to the unbleached condition, SEM images at ×40k for the polished specimens, En and Gr exhibited deeper flaws, whereas Bc and Cs showed no significant changes after one month.Lu contained more dark spaces (the polymer matrix) and a more prominent clustering of filler particles.That being said, more remarkable surface deterioration and margin fractures were observed on the unpolished specimens compared with the polished group.Unpolished Cs and En did not present significant surface morphology variations.In this study, all samples were stored in water at 37 °C as simulator of intraoral fluids 61 and temperature and being more reflective of the clinical conditions.According to ISO standards (ISO 10993-13), any devices intended to be used for more than 30 days and tested in simulated conditions should be tested at 1-month, 3-month, 6-month, and 1-year 40 , hence a short term storage for 3 months was applied.
Future work should involve bleaching at high temperatures and using 3D micro-CT or AFM to facilitate a deeper understanding of the changes bleaching causes inside the materials.

Conclusion
The following can be concluded: the hardness of CAD/CAM composite blocks was affected by in-office bleaching, with PICN exhibiting the least hardness stability of all of the resin-composite blocks.In-office bleaching significantly influenced the surface roughness of unpolished CAD/CAM composite blocks.In summary, polishing the restorations in advance of a bleaching procedure is advisable.Figure 5. SEM images at ×40k of the polished surfaces of all investigated materials before bleaching, 60 min after bleach, after a month from that of unbleached condition.Note the less demarcated particles margins and more rounded particles after bleach, and after a month from that of unbleached condition.Lu showed more dark spaces (polymer); orange arrows showed more prominent clustering of filler particles (blue arrows).Figure 6.SEM images at ×40k of the unpolished surfaces of all investigated materials before bleach, 60 min after bleach, and after a month from that of unbleached condition.Note the less demarcated particles margins and more rounded particles after bleach and after a month from that of unbleached condition.

Figure 1 .
Figure 1.A diagram of overall experimental and characterization procedure.

Figure 2 .
Figure 2. A bar chart illustrating the mean values of hardness reduction percentage (HR %) (A) after 60 min, (B) 24 h post-bleaching, and (C) one-month post-bleaching.Error bars represent the standard deviation and ns indicates 'not significant' .

Figure 4 .
Figure 4.A bar chart illustrating the mean values of surface roughness (Ra) of polished (A) and unpolished (B) samples before, after 60 min of in-office bleaching and 1-month post-bleaching.Error bars represent the standard deviation.