Can TPO as Photoinitiator Replace “Golden Mean” Camphorquinone and Tertiary Amines in Dental Composites? Testing Experimental Composites Containing Different Concentration of Diphenyl(2,4,6-trimethylbenzoyl)phosphine Oxide

The aim of this research was to compare the biomechanical properties of experimental composites containing a classic photoinitiating system (camphorquinone and 2-(dimethylami-no)ethyl methacrylate) or diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide (TPO) as a photoinitiator. The produced light-cured composites consisted of an organic matrix-Bis-GMA (60 wt.%), TEGDMA (40 wt.%) and silanized silica filler (45 wt.%). Composites contained 0.27; 0.5; 0.75 or 1 wt.% TPO. Vickers hardness, microhardness (in the nanoindentation test), diametral tensile strength, resistance to three-point bending and the CIE L* a* b* colorimetric analysis was performed with each composite produced. The highest average Vickers hardness values were obtained for the composite containing 1 wt.% TPO (43.18 ± 1.7HV). The diametral tensile strength remains on regardless of the type and amount of photoinitiator statistically the same level, except for the composite containing 0.5 wt.% TPO for which DTS = 22.70 ± 4.7 MPa and is the lowest recorded value. The highest average diametral tensile strength was obtained for the composite containing 0.75 wt.% TPO (29.73 ± 4.8 MPa). The highest modulus of elasticity characterized the composite containing 0.75 wt.% TPO (5383.33 ± 1067.1 MPa). Composite containing 0.75 wt.% TPO has optimal results in terms of Vickers hardness, diametral tensile strength, flexural strength and modulus of elasticity. Moreover, these results are better than the parameters characterizing the composite containing the CQ/DMAEMA system. In terms of an aesthetic composite containing 0.75 wt.%. TPO is less yellow in color than the composite containing CQ/DMAEMA. This conclusion was objectively confirmed by test CIE L* a* b*.


Introduction
In every branch of medicine doctors use golden means to treat their patients, but is it always the best solution for every situation? In operative dentistry, especially dental materials, dentists use most often composites to fill most of the dental cavities. Do dentists focus on ingredients of composite they use? Is it possible that 1% of a whole composite can change the aesthetic and quality of dental composite? Dentists accepted the advantages and disadvantages of golden mean which is camphorquinone. The technology and science before use, was silanized with γ-methacryloxypropyltrimethoxy silane (Unisil Sp. z o. o., Tarnów, Poland). The composites were manually mixed (with an agate mortar, the laboratory method of producing composites) till a smooth paste was achieved at room temperature without daylight and artificial light. All samples were cured with polywave Valo Lamp Ultradent Products Inc., South Jordan, UT, USA) with three irradiance outputs (1000 mW/cm 2 , 1450 mW/cm 2 and 3200 mW/cm 2 ) and a light range of 395-510 nm. After multiple tries, the optimal curing duration was 20 s per 2 mm of material high. After this, the samples had optimal features. The samples that were used to hardness and diametral tensile strength testing were cylindrical (3 mm high and 6 mm diameter). The material was put into silicon molds and then irradiated on both sides. The Vickers hardness (HV) test was performed as a first examination. Eleven samples were tested of every composite. It was determined with semiautomatic hardness tester (ZHV2-m Zwick/Röell, Ulm, Germany). The diamond shape as a square-based pyramid with apex angle 136 degree is used and the indenter was loaded with 9.81 N. The contact of pyramid and sample was 10 s.
The microhardness of the composites was tested with the NanoTest 600 (Micromaterials Ltd., Wrexham, UK) using a Berkovich indenter. The maximum force was 10 mN, the loading and unloading speed was dP/dt = 0.5 mN/s. The measurements were carried out in controlled conditions of temperature (T = 20 • C) and relative humidity (60 ± 5%). The samples that were tested was 2 mm high and 10 mm in diameter. The composite microhardness and reduced modulus were calculated on the basis of the unloading curve using the procedure proposed by Olivier and Pharr [25]. The distances between measurements were, successively, 0, 450, 900, 1350, and 1800 µm. In this article, the results of nanotesting the composite containing 0.5 wt.% TPO are not stated because when the samples were cured on one side, the bottom stayed uncured. The bottom of these samples remained a soft paste.
The diametral tensile strength test (DTS) assesses the tensile properties of the dental composite materials. Eleven cylindrical samples (diameter 6 mm and thickness 3 mm) were tested of every composite. DTS is the maximum resistance against loads tending to destroy a sample. The crosshead speed was 2 mm/min. The examination was determined with universal testing machine (Z020, Zwick/Röell, Ulm, Germany). The numerical value is calculated according to the following formula (1): where: • DTS-diametral tensile strength (MPa); • P-load applied (N); • D-diameter of sample (mm); • T-high of sample (mm).
Flexural strength was performed on a three-point bending test. Six samples were tested of every composite. The samples used in this test were rectangular (25 mm × 2 mm × 2 mm), and they were irradiated at 3 points twice for 20 s on both sides: In total, the sample was irradiated for 120 s. The samples were then placed on two supports 20 mm apart. A force was applied in the middle, downwards at a 90 • angle. The test was evaluated in a universal testing machine (Zwick Z020, Zwick/Röell, Ulm, Germany), the crosshead speed was 1 mm/min and the examination complied with ISO regulations [26]. The maximum force, which destroyed the sample, was measured for each specimen.
Flexural strength (MPa) was defined with following Equation (2): where • W-force, which caused the destruction of the sample (N); • l-distance between supports, 20 mm; The last examination which was performed was CIE L* a* b* color system on spectrophotometer KONICA MINOLTA CM-3600A (Germany). Every color can be described by three values: hue, brightness and saturation. This test was performed on cylindrical samples (2 mm height and 10 mm diameter) that were cured before testing. The three measurement were made for each composite. The spectrophotometer was calibrated in accordance with the manufacturers' recommendation. The control group was sample containing CQ and DMAEMA as a photoinitiator system. System CIE L* a* b* contains three axes: a* and b*, which are at right angles to each other and define the basic colors. The third axis L* means brightness and it is perpendicular to plane created by axes a* and b*. A scale of axis a* is from −120 (green color) to +120 (red color). A scale of axis b* is from −120 (blue color) to +120 (yellow color). The scale of axis L* is from 0 (black saturation) to 100 (white saturation). System CIE L* a* b* assumes considering the differences between colors on basis of the distance between points in the three-axes spatial layout.
For the statistical analysis of the results, Microsoft Excel from the Microsoft Office 2010 and Statistica v. 13 (Statsoft Polska Sp.z o. o., Krakow, Poland) were used. To evaluate the distributions of particular parameters, the Shapiro-Wilk test of normality was applied. In cases of nonnormal distribution, the Kruskall-Wallis test was used. In case of normal distribution of particular parameters, the equality of variances was assessed with the use of Levene's test. For equal variances, ANOVA with Scheffe's post hoc test was applied. The accepted level of significance was p = 0.05.

Results
All results are summarized in Table A1, but individual studies (i.e., HV, DTS, etc.) are discussed separately for a more accurate analysis of the relationships. The first examination that was performed was Vicker's hardness test. The control group containing CQ and DMAEMA had the lowest HV. The tendency was noticeable that the higher the concentration of TPO, the higher the HV. According to the ANOVA, a statistically significant difference was demonstrated in the HV (p-value = 0.00000) of manufactured composites. The post hoc Scheffe's test showed statistically significant differences ( Figure 1 The highest microhardness on the top of sample characterize the composite with 0.27 wt.% of TPO. The lowest microhardness on the top of sample characterize the composite with 1 wt.% of TPO. On the high 1800 µm the lowest microhardness and reduced modulus has composite with conventional photoinitiator-CQ. Inside the sample the highest values of microhardness has composite containing CQ ( Table 2). The second test was diametral tensile strength. The values for the control group were not the lowest; rather, they were in the optimal middle of the chart. However, there is no tendency concerning dependence values of DTS from concentration of TPO. According to the ANOVA, a statistically significant difference was demonstrated in the DTS (p-value = 0.00521). The post hoc Scheffe's test showed statistically significant differences ( Figure 2 Table 2). The second test was diametral tensile strength. The values for the control group were not the lowest; rather, they were in the optimal middle of the chart. However, there is no tendency concerning dependence values of DTS from concentration of TPO. According to the ANOVA, a statistically significant difference was demonstrated in the DTS (p-value = 0.00521). The post hoc Scheffe's test showed statistically significant differences (     The results of CIE L* a* b* color system are shown in Figures 5 and 6. There were three color measurements made for each composite. The values closest to the control group of CQ and DMAEMA on axis a* are connected with concentration of TPO. The higher the TPO concentration, the greater the distance between control group and research group. According to the ANOVA, a statistically significant difference was demonstrated in the a* (p-value = 0.00000). The post hoc Scheffe's test showed statistically significant differences between:   The results of CIE L* a* b* color system are shown in Figures 5 and 6. There were three color measurements made for each composite. The values closest to the control group of CQ and DMAEMA on axis a* are connected with concentration of TPO. The higher the TPO concentration, the greater the distance between control group and re-   The results of CIE L* a* b* color system are shown in Figures 5 and 6. There were three color measurements made for each composite. The values closest to the control group of CQ and DMAEMA on axis a* are connected with concentration of TPO. The higher the TPO concentration, the greater the distance between control group and re-  There was no specific tendency between distance in the control group and the concentrations in the research groups. According to the ANOVA, a statistically significant difference was demonstrated in b* (p-value = 0.00000). The post hoc Scheffe's test showed statistically significant differences ( Figure 6) between:

Discussion
This article compares the various features of dental composites containing CQ/DMAEMA and different concentration of TPO. This range of analyses highlights the most important properties of dental composites. The null hypothesis of our research was accepted that dental resin containing TPO performs no worse than composites with CQ/tertiary amines.
The first examination that was performed was Vicker's hardness. It was statistically confirmed that composites with TPO as a photoionitiator show higher values of hardness. Moreover, the higher concentration of TPO, the better the values of hardness. This tendency is also confirmed in other research. Salgado et al. [23] performed the Knoop analysis of hardness and proved that dental resin containing CQ (1.0 mol%) and ethyl 4-dimethylaminobenzoate (1.0 mol%) showed lower hardness values than dental composite including TPO (1.0 mol%). Randolph et al. [14] also confirmed that composites containing TPO (0.42 wt.%) has higher values of hardness than CQ with dimethylaminoethylmethacrylate There was no specific tendency between distance in the control group and the concentrations in the research groups. According to the ANOVA, a statistically significant difference was demonstrated in b* (p-value = 0.00000). The post hoc Scheffe's test showed statistically significant differences ( Figure 6

Discussion
This article compares the various features of dental composites containing CQ/DMAEMA and different concentration of TPO. This range of analyses highlights the most important properties of dental composites. The null hypothesis of our research was accepted that dental resin containing TPO performs no worse than composites with CQ/tertiary amines.
The first examination that was performed was Vicker's hardness. It was statistically confirmed that composites with TPO as a photoionitiator show higher values of hardness. Moreover, the higher concentration of TPO, the better the values of hardness. This tendency is also confirmed in other research. Salgado et al. [23] performed the Knoop analysis of hardness and proved that dental resin containing CQ (1.0 mol%) and ethyl 4-dimethylaminobenzoate (1.0 mol%) showed lower hardness values than dental composite including TPO (1.0 mol%). Randolph et al. [14] also confirmed that composites containing TPO (0.42 wt.%) has higher values of hardness than CQ with dimethylaminoethylmethacrylate (0.2/0.8 wt.% appropriately) regardless of the exposure time. Frequently, hardness is used as an indirect measurement of degree of conversion and as an indicator of crosslinking [27]. Bertolo et al. showed that composite containing TPO (0.5 mol%) had a higher degree of conversion than did material with CQ and EDMAB (0.5 mol%/1 mol%). For TPO, it was 63.0 ± 1.0% and for CQ/EDMAB, 56.05 ± 0.5% [27]. Our findings are in line with above mentioned results. Randolph et al. also proved that TPO (0.42 wt.%) has a higher degree of conversion than CQ and DMAEMA (0.2/0.8 wt.%). Moreover, TPO achieves better results of degree of conversion in a shorter exposure time [21].
The test of nanoindentation shows that composite containing CQ/DMAEMA has the best values of hardness inside the samples, and the deeper the penetration within the sample, the better hardness values. However, the dental resin containing 0.27 wt.% of TPO manifests higher values of hardness than CQ/DMAEMA at distance 450 µm and it retains a similar growth tendency as CQ/DMAEMA. Inside the samples, at 900 µm and 1350 µm, the highest hardness value other than that of CQ/DMAEMA was TPO 0.75 wt.%. According to Salgado et al., research composites including TPO has lower depth of cure values than dental resin with CQ and DMAEMA system [23]. Palin [29]. The same conclusion was proved also in our last research [8]. Filtek Ultimate containing golden mean-CQ and tertiary amine has higher values of microhardness inside the sample (1570 ± 160 MPa) than Tetric EvoCeram Powerfill (1260 ± 200 MPa) and Tetric EvoCeram Bleach (1250 ± 170 MPa) which include not only CQ, but also TPO and Ivocerin [8]. Ivocerin -dibenzoyl germanium is a patented photoinitiator of one manufacturer Vivadent and it can be found only in their composites. The absorption range of Ivocerin is 390-445 nm and absorbance maximum is 418 nm. Ivocerin forms at least two radicals, that means it is also more efficient than CQ [8,22,30]. Our research due to nanoindentation shows that composites with TPO have lower values of hardness inside samples than those with CQ. However, the highest concentration of TPO, the highest values of hardness inside the samples. But it is worth underlining again that composite containing 0.5 wt.% TPO as a photoinitiator was not cured on unexposed side, when the sample was 2 mm high. The depth of cure depends on the concentration of photoinitiator.
The nanoindentation test not only assesses the microhardness but also reduced modulus of elasticity. Inside sample containing CQ/DMAEMA reduced modulus is growing with the distance of measurement. This trend is the same as microhardness.
The  [31]. Considering the results of flexural strength of TPO 0.27 wt.% and 0.5 wt.% and comparing them with the results for CQ/DMAMEA, the results of golden mean are better than TPO in this low concentration.
The last examination was CIE L* a* b*. This method is indicated for determination the shade of dental materials, not only composites but also ceramic and let scientists compare these material shades with color of dentine or enamel of natural teeth. The axis a* refers to color green and red. There was also significant statistical difference in values of a*. The CQ/DMAEMA is a control group and reference point. The color of composite with CQ/DMAEMA is closer to red, so this hue is warmer than the hue of composites containing TPO in concentrations 0.5 wt.%, 0.75 wt.%, and 1 wt.%. However, the composite containing 0.27 wt.% TPO has hue values deviated to green color. The closest values to red had composite including 1 wt. % of TPO. Axis b* refers to yellow and blue. The composite containing CQ/DMAEMA has color values closer to yellow than composites containing TPO in concentration 0.27 wt.%, 0.5 wt.% and 0.75 wt.%. The hue of composite including 0.27 wt.% TPO were the most distant from yellow. In contrast, composite containing 1% TPO had the closest values to yellow, even closer than CQ/DMAEMA. The lower the concentration of TPO, the less yellow the composite. Bertolo et al. [24] also analyzed the color of the composites containing alternative photoinitiator systems. The concentration of TPO was 0.5 mol% and 0.5 mol%/1.0 mol% of CQ and ethyl-4-(dimethylamino)benzoate (EDMAB). The highest yellow values according to their studies has composites containing CQ/EDMAB, so this statement also confirms our results. Not only do Bertolo et al.'s results confirm our conclusions, Salgado et al. [23] also showed that composite containing TPO (1.0 mol%) was less yellow than composite with CQ/EDMAB (1.0 mol%/1.0 mol%). In clinical use, this less yellowish composite will match with ultra-white teeth after bleaching.

Conclusions
All the composites were tested to assess the most important mechanical properties, and the hues of the composites were also rated. The concentration of photoinitiator influenced the Vicker's hardness value. Additionally, samples containing TPO had better hardness values. There was also a noticeable influence of type and concentration of photoinitiator on the diametral tensile strength. To sum up: the best mechanical and aesthetic properties were found in the composite containing 0.75 wt.% TPO as a photoinitiator. This composite had higher microhardness, diametral tensile strength, flexural strengths and modulus of elasticity than the composite with the conventional/often used photoinitiator system, i.e., CQ with tertiary amine. Not only were the mechanical properties better, but this composite was less yellow than material with CQ/DMAEMA. However, to answer the question in the title, more tests on the composite with TPO need to be done, i.e., conversion degree measurements, shrinkage stress analysis and material cytotoxicity tests.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.