ANTIPOLLUTION COSMETIC EFFECTIVITY AGAINST AIR POLLUTANT ABSORPTION 39 The density of both cosmetic products is similar, presenting values of 1.01–1.03 g cm-3, which is a norm al density value for this type of cream cosmetic product. As cosmetic creams are applied directly on the skin, it is preferable that it possesses a pH similar to tha t of natural skin. Variable skin pH values are being reported in the literature, all in the acidic range but with a broad range from pH 4.0 to 7.0 (14). As it can be seen in Table 2, both cosmetic products have a similar slightly acidic pH value, with values between 6 and 7. The conductivity test is widely used to determine the emulsion type and to estimate emulsion stability. Hig h conductivity values (higher than 50 μS cm-1) indicate oil/water emulsions, whereas low conductivity values ( 1 μS cm-1) indicate water/oil emulsions (15). Conductivity values of cosmetic products are given in Table 2. As it can be seen, product B (1,888–2,832 μS cm-1) has a substantially higher conductivity value than cos- metic A (377–701 μS cm-1), both being oil/water emulsions. Regarding the refraction index, it can be seen that developed cosmetic products show comparable values in the 1.37–1.42 r ange. DESIGN AND EVALUATION OF THE SIMULATION CHAMBER CONDITIONS The air concentration of HAPs inside the chamber is controlled by the concentration and fl ow of the standard solution and the air fl ow. Acetone was selected as carrier solvent be- cause of the solubility of HAP compounds and its high vapor pressure. Moreover, acetone does not enhance the skin absorption of lipophilic compounds, such as HAPs (16). The simulation chamber was preconditioned overnight using the aforementioned working conditions, assuming a zero input–output balance of target analytes after the overnight cycle. Infi nite dose conditions were assured by a constant fl ow of air containing HAPs introduced inside the chamber. Homogeneity of the chamber was assessed during HAP permeability experiments using the aforementioned vertical cell design, at different loca- tions inside the chamber. The results showed a precision, established as relative standard deviation of fi ve measurements (each corner and the center of the chamber) lower than 16% for HAP permeability experiments. Preliminary experiments were carried out at 15 mg m-3 air working concentration ad- justing pure and dry air fl ow to 3.0 L min-1 and 0.9% (w/v) BTEX standard solution (in acetone) fl ow to 5.0 μL min-1. Active sampling was used to evaluate the real concentra- tion of target analytes inside the calibration chamber several times a day, during three consecutive days. Moreover, these concentrations were compared with those obtained af- ter application of Henry’s law taking into consideration the concentration of the analytes Table II Physicochemical Properties of the Developed Antipollution Cosmetic Products Antipollution cosmetic A Antipollution cosmetic B Viscosity (Pa s-1) 2.53–3.53 3.95–4.95 Density (g cm-3) 1.0158–1.0258 1.0136–1.0236 pH 6–7 6–7 Conductivity (μS cm-1) 377–701 1888–2,832 Refraction index (nD) 1.3748–1.3948 1.3999–1.4199

JOURNAL OF COSMETIC SCIENCE 40 in the water solution once the equilibrium was reached. As it can be seen in Table 3, all these values were in the same range as those theoretically expected, with HAP air concen- tration from 12 to 15 mg m-3 and from 8 to 14 mg m-3, except for nitrobenzene that provided 4 and 3 mg m-3, for the active sampling measurement and using Henry’s law, respectively. Thus, the effi ciency of the volatilization system to introduce HAPs inside the simulation chamber was positively assessed. EFFECTIVENESS OF ANTIPOLLUTION COSMETICS Evaluation of the effectiveness of developed antipollution cosmetics has been carried out using the simulation chamber, as previously described. Control experiments were carried out using Strat-M® membranes without any added cosmetic layer, whereas cosmetic sam- ples experiments were performed with Strat-M® membranes with a homogeneous layer of 2 mg product per cm2 exposed membrane. Vertical cell devices for control and cosmetics experiments were introduced in the exposition chamber for different exposition times, from 0.5 to 24 h. After each exposition time, control and cosmetic samples were removed, and receiving solutions were analyzed by GC-MS to evaluate the antipollution effective- ness over time of the evaluated cosmetic. It should be highlighted that for each exposition time, 10 experiments were carried out: one blank, three controls, three cosmetic A– treated experiments, and three cosmetic B–treated experiments. F igure 1 shows permeation curves obtained for the studied HAPs in the control and cosmetic samples A and B. As it can be seen, an initial linear uptake was observed for all compounds that reached the equilibrium at exposition times higher than 12 h, indi- cating that the analyte permeability followed Fick’s fi rst law. The equilibrium state in the case of cosmetic samples is reached at longer times than that in control experiments, Table III Concentration of HAPs in the Simulation Chamber Air Measured by Active Sampling and Using Henry’s Law and Diffusion Parameters HAPs [HAPS] [mg m-3 ± s] KOWa HCCb Permeability at equilibrium (μg cm-2 ± s) J (μg cm-2 h-1 ± s) Active sampling Henry’s law 1,2-dichloroethane 15 ± 3 12 ± 3 1.47 19.83 2.8 ± 0.3 0.566 ± 0.007 Benzene 12 ± 3 9 ± 1 2.13 4.21 0.68 ± 0.07 0.140 ± 0.006 Bromodichloromethane 14 ± 3 11 ± 1 2.10 9.92 1.9 ± 0.2 0.377 ± 0.006 Toluene 12 ± 2 10 ± 1 2.69 3.72 0.67 ± 0.07 0.138 ± 0.008 1,2-dibromoethane 13 ± 3 10 ± 3 1.60 44.62 5.5 ± 0.5 1.2 ± 0.2 Chlorobenzene 14 ± 3 12 ± 3 2.84 7.44 1.6 ± 0.1 0.24 ± 0.04 Ethylbenzene 13 ± 2 10 ± 1 3.13 3.22 0.59 ± 0.06 0.114 ± 0.005 m+p-xylene 25 ± 4 21 ± 2 3.20–3.18 3.47 1.2 ± 0.1 0.25 ± 0.04 Bromoform 13 ± 4 10 ± 4 2.35 42.14 7.4 ± 0.5 1.4 ± 0.3 o-xylene 14 ± 3 14 ± 2 3.13 4.71 1.2 ± 0.1 0.21 ± 0.02 Nitrobenzene 3 ± 1 4 ± 2 1.85 1,140.32 78 ± 13 5.1 ± 0.3 Naphthalene 12 ± 3 8 ± 3 3.36 57.01 7.8 ± 0.8 1.1 ± 0.1 a Kow values obtained from Ref. (22). b Hcc values obtained from Ref. (13).