ANTIOXIDANT POWER OF DERMOCOSMETIC CREAMS 81 reference electrode is triangular. The voltage varies linearly with time and the slope is referred to as the scan rate. In a few seconds, a current density-potential curve can be obtained, giving information about the energy level necessary to perform the electrode reactions and the rate of these reactions. If the heterogeneous electron transfer rate is high compared to the diffusion rate, this curve presents a peak, whose current is pro­ portional to the concentration of the electroactive species present in solution and to the square root of the scan rate. Another important criterion to characterize the electrode reaction is the value of the peak potential, which is independent of the scan rate if the electron transfer is fast in relation to diffusion. If the product of an electrochemical reduction reaction may be reoxidized during the reverse scan rate, the reduction peak is accompanied by an oxidation peak and the difference between the peak potentials is an indication of the reversibility of the heterogeneous electron exchange reaction. Cyclic voltammetry is a very useful method to elucidate the mechanism of electrode reactions in the case where the heterogeneous electron transfer is accompanied by chemical reac­ tions occurring before or after the electrode reaction. In the present work, cyclic voltam­ metry is used to compare creams on the basis of the determination of the peak potential, the peak current intensity, and the charge involved in the electro-oxidation process. MATERIALS AND METHODS H 2 SO4 , NaOH, and thiolactic acid were purchased from Acros K2SO4 and H2O2 from Sigma and KH2PO4 and K2HPO4 from Merck. Unless otherwise indicated, all solu­ tions were prepared in potassium sulphate solution (0.4 mol/1 - 1, pH = 11.0). Creams were either commercially available or made in Pierre Fabre's laboratory. In the later case the samples were gels containing PEG 600, carbopol 980, paraffin, cremophor RH40, sorbic acid, nipagin, sodium hydroxide, and water. Three gels were made: one with pH = 4.86, one with pH = 6.88 and one containing BHT, with pH = 6.95. A depilatory cream from Klorane was chosen as an example. Others creams were studied: a restructuring cream from Nivea Vital a corrective dermatological cream for wrinkles, Active C, from Laroche-Posay a depigmenting emulsion, Trio D, from Laboratoires d'Evolution Dermatologique a whitening day cream from Decleor an emulsion, Ystheal +, from Laboratoires Dermatologigues Avene (Pierre Fabre) an epithelial cream, A Derma, from Laboratoires Dermatologiques Ducray, and an after-sun repair balm, Uriage, from Laboratoires Dermatologiques Uriage. All the electrochemical experiments were carried out in a single compartment cell at room temperature. An airtight cell was used to study the influence of oxidative stress. An air or nitrogen flux was introduced when necessary, and the flow was controlled with a flow meter: Brooks tube (R-2-15-AAA P-072 float: sapphire scale: 0-5 1/h). The electrochemical manipulations were performed with an Autolab Metrohm potentiostat interfaced to an HP omni-book XE 4500 microcomputer and using the GPES software. The working electrodes were platinum (0.03 cm2), gold (0.07 cm2), or vitreous carbon (0.07 cm2) rotating-disc electrodes. A large-surface-area platinum grid was used as counter electrode. All potentials were measured and expressed in reference to a saturated mercurous sulphate reference electrode Hg/Hg2SO4/K2SO4sar (MSE E = 0.656 V/SHE) connected to the cell by a Luggin capillary. Before each experiment, the working electrode surface was polished with abrasive paper (262X imperial lapping film sheets)

82 JOURNAL OF COSMETIC SCIENCE and rinsed with distilled water. For vitreous carbon, anodic polarization was then im­ posed at 1 V during one minute in sulphuric acid (H2S04 , 0.1 mol/1-1)_ For the platinum (respectively gold) electrode, cyclic voltammograms were performed in H2SO4 , 0.5 molW 1 (resp. 0.1 molW 1 ) at 50 mV s- 1 between -0.65 V and 0.75 V (resp. between -0.25 V and 1.1 V) until reproducible current density-potential curves (21) were obtained. In all the electrochemical experiments, the potential range was chosen according to the limits of the electroactivity domain of the solvent, in order to avoid the oxidation/reduction of water. RES UL TS AND DISCUSSION CYCLIC VOLTAMMOGRAM IN THE CREAMS Figure 1 shows the experimental device used. Experiments were performed by simply introducing the working, counter, and reference electrodes directly into the creams. A few grams (about 5 to 7 grams depending on the cream density) were necessary, in order for the electrodes to be in contact with the cream. Figure 2 shows cyclic voltammograms obtained with a platinum electrode introduced directly into two samples, e.g., a de­ pilatory cream from Klorane (pH = 11) containing BHA and thiolactic acid as antioxi­ dants (solid line) and an antioxidant-free base made in Pierre Fabre's laboratory (dashed line). In both cases, the curve presented a conventional shape with essentially no resistive or capacitive current. In the former case, a significant anodic current was recorded with a peak potential close to 0.31 V. Comparatively, only a little amperometric response was obtained with the other cream. The difference can therefore be attributed to the presence HP omnibook XE 4500 I Potentiostat µAutolab II Metrohm I I microcomputer l 1r,. - - Luggin capillary .... ,,,.- reference electrode -- (Hg/Hg2SOJK2SO4sat) phosphate buffer ... 30 0mm ,, pH=7,00 --- 50mm Figure 1. Experimental device used. The three-electrode system was introduced directly into the cream.