HEMP-SEED AND OLIVE OILS 229 lower temperatures and increases oxidation temperatures, suggesting an increased pro­ tective effect upon heating (7). The physicochemical characteristics (8) of hemp-seed oil are examined through a number of analytical determinations: quality indices (free fatty-acid content, peroxide values, spectrophotometric characteristics in the UV region) parameters of oxidation processes (MDA and MONO analysis) and chlorophyll pigment content and its spectrophoto­ metric characteristics in the UV region. Hemp-seed oil was compared with a commercial virgin olive oil, since olive oil is considered to be one of the most valuable edible fats, thanks to its nutritional characteristics. The rheological characteristics of different for­ mulations of olive and hemp-seed oils were also investigated in order to determine the most suitable one for topical administration. MATERIALS Hamarnelis virginiana distillate, cetearyl alcohol, hydrolyzed milk protein, hydroxy ethyl cellulose, Polysorbate 20 (Tween 20®), Polysorbate 40 (Tween 40®), Peg-40- hydrogenated castor oil (Cremophor RH®40), and dimethicone (silicon oil 350 mPa s) were purchased from Acef. Acrylates/Cl0-30 alkyl acrylate crosspolymer (Carbopol 1382® and Pemulen TR-1 ® ) were gifts from Biochim. C 14-22 alkylalcohol/C 12-20 alkylglucoside (Montanov L®), cetearyl alcohol/cetearyl glucoside (Montanov 68EC®), polyacrylamide, C 13 _ 14 isoparaffin, laureth 7 (Sepigel 305®), and Na lauroyl oat amino acids (Proteol OAT®) were gifts from Seppic. Methyl glucose sesquistearate (20) OE (Glucamate SS E20®), and methyl glucose dioleate (Glucate DO®) were gifts from Amerchol. Imydazolidynylurea (Gram I®), methylisothiazolinone/methylchlo­ roisothiazolinone (Kathon CG®), magnesium chloride, and magnesium citrate were gifts from Sinerga S.r.l. Potassium hexadecyl hydrogen phosphate (Amphisol K®) was a gift from Roche. Poliglyceryl-3-methylglucose distearate (Tego Care 450®) was a gift from Goldschimdt Italia. Sucrose tristearate or palmitate (sucroesters DUB SE® 3S and 15 P) and sucrose mono (or di)stearate (or palmitate) (sucroesters SP® 30 and 50) were gifts from Sistema BV. C 12 _15 alkyl benzoate (Finsolv TN®) was a gift from Prodotti Gianni. Glyceryl monostearate (Cutina GMS®) was a gift from Henkel. Olive leaf extract (Eurol BT®) and Olivero 700® were gifts from B&T S.r.l. Cannabis sativa seed oil was a gift from Verdesativa, and olive oil was purchased from Carli S.r.l. LRI® solubilizer was a gift from Waker. Acetic acid, hydrochloric acid, thiobarbituric acid, trichloracetic acid, butanol, ethanol, methanol, n-hexane, diethyl ether, phenolphthalein, potassium hy­ droxide, potassium iodide, sodium chloride, sodium hydroxide, and starch were from Merck. Chlorophyll 3A (1522), Chlorophyll (5022), Chlorophyll Al 0, and Chlorophyll B 10 were from Biochim. INSTRUMENTS Viscosity of the oils was measured with a Schott-Gerate capillary viscometer (K = 0.01053, 0.4-6.0 cSt). An Abbie refractometer was used to characterize the oils, and a UV-VIS Lambda 2 spectrophotometer (Perkin-Elmer) was used to analyze the samples. Kathon CG® was quantified by high-performance liquid chromatography (HPLC) us-

230 JOURNAL OF COSMETIC SCIENCE ing an LC-6A Shimadzu HPLC, equipped with a C-R3A Chromatopac integrator, an SPD-2A UV-VIS spectrophotometric detector, and an RP-C18 column (particle diam­ eter 5 µm). Shear deformation of emulsions and oil viscosity were determined using a rotational viscometer (model DV-II, Brookfield) with a small adapter chamber, SC 21-29. An Ultra Turrax® T25 basic (Janke and Hunkel-IKA-Labortechnik), a Silverson SL 2 homogenizer (Silverson Machines Ltd), and a DLS stirrer (Velp) were used to prepare the emulsions. An Eppendorf 541 7 centrifuge was used for the stability tests on the emulsions. pH measurements were performed with an HI 9321 microprocessor pH meter (Hanna Instruments). Microscopic analysis used a Labovert Leitz optical micro­ scope equipped with a Wild MPS 46 Fotoautomat camera. Irradiation tests were per­ formed in Pyrex glass cells (5-ml solutions) under solarboxes equipped with a UVB TL 40/12 RST40Tl2 lamp (Philips®), a UVA TL K0540 W lamp (Philips®), and an Osram Ultravitalux lamp (solar spectrum). METHODS In order to make a qualitative evaluation of hemp-seed oil and test the validity of the analytical methods adopted, comparative studies were carried out on hemp-seed, olive, and extra-virgin olive oils simultaneously (4,8-11). Two different hemp-seed oil samples, obtained from the same batch of seeds, picked in 1998 and crushed in 1998 and in 1999, were also analyzed. The two hemp-seed oil samples had been produced by two different oil mills, adopting two different seed-crushing techniques. The hemp-seed oil produced in 1998 was obtained by mechanical extraction employing a screw press provided at its extremity with a filter to remove suspended impurities from the oil. The hemp-seed oil produced in 1999 was obtained by mechanical extraction with a method known as "Baglioni": hemp seeds are placed under a millstone with hazelnut shells, which, increasing the contact surface, reduce the crushing time required. DETERMINATION OF DENSITY Oil density was determined at 20°C by weighing an exact volume (5 ml) of oil the determination was carried out three times for each oil sample. The density of oil (p 20 , expressed in g/cm3) is the ratio of weight to volume at 20°C, while the relative density (d20 4 ) is the ratio of the density of oil to the density of water at 4°C. Results were expressed as relative density d20 4. DETERMINATION OF VISCOSITY Determination with rotational viscometer. The apparent viscosity of oil samples was deter­ mined with a Brookfield® rotational viscometer employing a small adapter chamber, SC 21, at 2 5 °C. The determinations were performed at 18. 6 s - 1 on 8 g of oil and were repeated three times per sample. Determination with capillary viscometer. Kinematic viscosity was determined with an Ostwald capillary viscometer by measuring oil-flow time through a capillary. The kinematic viscosity (v, expressed in cm2s- 1 ) was calculated from equation 1, whereas the