2.1

Preparation of samples

For sample preparation the method of our earlier study has been improved: the bleaching treatment of the samples has been performed excluding daylight, because light sources additionally can activate peroxides of bleaching gels. Thus the bleaching process could be accelerated .

The sample preparation is described in detail: about 100 mg of each unpolymerized composite ( Table 1 ) was inserted in one increment in a Teflon ring (6 mm diameter, thickness 1.9 mm = surface 92.36 mm ³ ) placed on a plastic matrix strip (Frasaco, Tettnang, Germany). After covering samples with another plastic matrix strip, samples were polymerized according to instruction of manufacturer ( Table 1 ) by using a dental manual light-curing unit (Elipar S10, 3 M ESPE, Seefeld, Germany) which was directly placed on the matrix strip. The light intensity of the LED lamp (1200 mW/cm ² ) was checked using Demetron ^® Radiometer (Kerr, USA).

For bleaching procedure two concentrations of bleaching gel were used exactly according to the instructions of the manufacturer ( Table 2 ): Opalescence ^® PF 15% Tooth Whitening Systems (PF 15%) (home bleaching procedure) and Opalescence ^® PF 35% Tooth Whitening Systems (PF 35%) (in-office-bleaching procedure). For each composite ( Table 1 ) 3 groups with 4 samples each ( n = 4) were prepared: (1) samples bleached with PF 15% for 5 h; (2) samples bleached with PF 35% for 30 min and (3) unbleached samples: control group. During bleaching process the samples were stored excluding daylight.

After bleaching process the bleaching gel was removed: (1) roughly remove by spatula; (2) carefully remove by absolute dry cotton sticks.

Subsequently samples were incubated in brown glass vails (Macherey-Nagel, Düren, Germany) with 1 ml of methanol (GC Ultra Grade, RATISOLV ^® ≥99.9%, Roth, Karlsruhe, Germany) at 37 °C and analyzed after 1 d and 7 d by GC/MS . As internal standard caffeine (CF) solution (0.01 mg/ml) (HPLC ≥ 99.0%, Sigma–Aldrich, St. Louis, United States) was added.

Water (LC–MS-Grade, ROTISOLV ^® , Roth, Karlsruhe, Germany) elutions were carried out as outlined above. For GC/MS analysis water samples were extracted with ethyl acetate (1:1, v/v) (LC–MS-Grade, ROTISOLV ^® ≥99.9%, Roth, Karlsruhe, Germany).

2.2

Analytical procedure

The analysis of the eluates was performed on a Finnigan Trace GC ultra gas chromatograph connected to a DSQ mass spectrometer (Thermo Electron, Dreieich, Germany). A Factor Four ^® capillary column (length 25 m, inner diameter 0.25 mm; coating 0.25 μm; Varian, Darmstadt, Germany) was used as the capillary column for GC. The GC oven was heated from 50 °C (2 min isotherm) to 280 °C (5 min isotherm) with a rate of 25 °C/min, and 1 μl of the solution was injected with a split ratio of 1:30. Helium 5.0 was used as carrier gas at a constant flow rate of 1 ml/min. The temperature of the split–splitless injector, as well as of the direct coupling to the mass spectrometer, was 250 °C. MS was operated in the electron ionization mode (EI, 70 eV), ion source was operated at 240 °C; only positive ions were scanned. The performed scan ran over the range m/z 50–600 at a scan rate of 4 scan/s for scans operated in full scan mode in order to qualify the analytes.

The results were referred to an internal caffeine standard (0.1 mg/caffeine = 100%), which allows to determine the relative quantities of substances released from various resin-based materials. All eluates were analyzed five times. The integration of the chromatograms was carried out over the base peak or other characteristic mass peaks of the compounds, and the results were normalized by means of the internal CF standard. Identification of the various substances was achieved by comparing their mass spectra with those of reference compounds, the NIST/EPA library, literature data, and/or by a chemical analysis of their fragmentation pattern .

Bis-GMA and Bis-EMA were analyzed by LC/MS using the method described in previous study . Detected values for Bis-GMA and Bis-EMA of investigated dental composites were under detection limit.

2.3

Calculations and statistics

The results are presented as means ± standard deviation (SD). The statistical significance ( p < 0.05) of the differences between the experimental groups was analyzed by one-way ANOVA and the post hoc Test (Tukey’s HSD test) .