ANALYST, FEBRUARY 1986, VOL. 111 129 Determination of Hydroquinone in Skin-toning Creams Using High-performance Liquid Chromatography Jane Firth and Ian Rix Consumer Hazards Group, Laboratory of the Government Chemist, Corn wall House, Waterloo Road, London SE18XY, UK A simple, rapid reversed-phase HPLC method for the determination of hydroquinone in skin-toning creams is described that is suitable for routine use. The sample is dissolved in methanol or methanol - light petroleum and directly injected without further purification. The method successfully passed a ruggedness test and was applied to a range of 35 creams. Keywords: Hydroquinone determination; high-performance liquid chromatography; skin-toning creams; cosmetic product The major user of skin-toning creams in the UK is the West Indian population.The creams are applied to even out the skin colour on the facial areas. They are believed to function by decolourising the melanin in the skin and preventing new melanin being formed. The most favoured material for use in skin-toning preparations is hydroquinone, which was reported by Spencer1 to be effective at 1.5-2% in a vanishing cream producing a temporary lightening of skin colour. Spencer found that a concentration of 5% was liable to cause redness and burning. Cases of patchy de-pigmentation have arisen following the use of some skin-toning creams available on the retail market.2 Council Directive 76/768/EEC of the European Communi- ties makes the general point that cosmetic products must not be harmful under normal or foreseeable conditions of use and specifically allows hydroquinone to be used in cosmetic products at a level of 2% m/m subject to certain conditions of use and warnings that must be printed on the label.The field of application is not specified in the basic Directive 76/768/ EEC.3 The second amendment , Council Directive 82/368/ EEC,4 specified the field of application as oxidising colouring agents for hair dyeing and excluded hydroquinone for use as a skin lightener from the scope of the Directive. The fifth Commission Directive 84/415/EEC5 permitted the use of hydroquinone as a localised skin-lightening agent subject to a maximum concentration of 2% m/m in the finished cosmetic product and a warning on the label containing the information "contains hydroquinone, avoid contact with the eyes, apply to small areas, if irritation develops discontinue use, do not use on children under the age of 12." Member States were asked to bring into force the laws, regulations and administrative provisions necessary to comply with the Directive by not later than 31st December 1985.The method described in this paper will be submitted to the EEC for consideration as the adopted method of analysis. Hydro- quinone is permitted at a concentration of up to 2% by mass under the Cosmetic Products (Safety) Regulations; Statutory Instrument 1984: No. 1260, which had to be complied with by 1st January 1986. However, recent studies have shown that some available products contain more than the permitted level of hydroquinone .6 In the light of these facts it was decided to carry out an extensive survey of skin-toning creams to determine their hydroquinone content.There is a dearth of literature on the determination of hydroquinone in skin-toning creams. The most recent pub- lished work of Popov and Yanishlieva7 involved extraction of Crown Copyright. hydroquinone with acetic acid and conversion of the hydro- quinone into p-benzoquinone, with subsequent spectrophoto- metric determination. The determination of hydroquinone in various sample matrices has been described although none included skin-toning creams.Gl1 The method described here is of wide application and is rapid, allowing ten samples and associated standards to be analysed in less than 3 h. Experiment a1 Apparatus Reversed-phase HPLC was performed at ambient tempera- tures using a Spectrophysics SP800 solvent delivery system and a Shimadzu SPD-MIA diode-array UV - visible spectro- photometric detector.Reagents All reagents were of analytical-reagent grade. The methanol was of solvent for liquid chromatography grade. Chromatographic Conditions The analytical column used was of stainless steel (250 mm X 4.6 mm i.d.) packed with Spherisorb S O D S of 5 pm. The sample injection volume was 10 p1. The mobile phase was methanol - water (10 + 90 V/V) pumped at a flow-rate of 1.5 ml min-1. The detector was operated at 226 nm with a sensitivity of 0.50 A full scale and a chart speed of 5 mm min-1. Skin-toning Cream Samples Samples of cream were bought from a number of retail outlets in the South London area.They were stored at room temperature throughout the investigation. Procedure Extraction of hydroquinone Transfer 0.05 g of cream containing 0.2-4.0% of hydro- quinone into a 10-ml calibrated flask and add 8.0 ml of methanol. Heat to 40 "C in a water-bath and shake occasion- ally until dissolved. Allow to cool and make up to the mark with methanol. If the cream fails to dissolve under these conditions repeat the procedure with 4.0 ml of light petroleum (60-80 "C boiling range) and make up to the mark with methanol.130 ANALYST, FEBRUARY 1986, VOL. 111 Table 1. Multi-factorial experiments for the ruggedness test Experiment Factor Ala . . . . Blb . . . . CIC . . . . Dld . . . . E/e . . . . F l f . . . . . . Glg . . . . Results . . . . Hydroquinone, YO 1 .. . . A . . . . B . . . . c . . . . D . . . . E . . , . F . . . . G . . . . 2.062 . . * . s Wavelength of measurementlnm . . Mass of cream taken for analysislg . . Detector type . . . . . . . . Columnlengthlmm . . . . . . Internaldiameter/mm . . . . . . Packing material . . . . . . . . Constructionalmaterial . . . . Mobile phase flow-ratelm1 min-1 . . Final volume of extracting solution/ml 2 A B D e g t 2.045 C f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mobile phase composition (methanol - water) 3 4 5 6 A A a a b b B B C C C C d d d d E e e E F F f G 8 G g f U V W X 2.127 2.020 2.086 2.070 Factor 7 a b C D e G Y 2.100 f 8 a b D E F g 2.056 C 2 ~~~~~~~ ~~ A B C SPD-MIA D . . 250 . . 4.6 . . Spherisorb SS-ODS . . Stainless steel E .. 1.5 F . . 10 G . . 10 + 90 . . 226 . . 0.05 . . Shimadzu ~ ~~ a 230 b 0.08 Pye Unicam PU4020 d 100 3 Cp-t,-Spher C18 Glass e 2.0 25 g 12.5 + 87.5 C f t v) 0 5? U 0 5 10 0 5 10 Ti meimin Fig. 1. Typical chromatograms for ( a ) a standard solution containing 0.152 g 1-1 of hydroquinone and (b) a skin-toning cream containing 1.6494% m/m of hydroquinone Preparation of the stock standard solution Prepare a stock solution of standard hydroquinone by dissolving the solid hydroquinone in methanol at a concentra- tion of 10 g 1-1. Preparation of the calibration graph Prepare a range of solutions by diluting aliquots of the stock hydroquinone standard with methanol to 100 ml in calibrated flasks. Inject the standard solutions and measure the peak- height absorbance. A straight-line calibration graph of absor- bance versus concentration was obtained which passed through the origin.Table 2. Effect of each factor in the ruggedness test D, = 1/4 (S + t + u + v - w - X - y - Z) = 0.0145 D,= 1 / 4 ( ~ - t+ u - v + w - x + y - Z) =0.0460 D, = 1 / 4 ( ~ - t + u - v - w + X - y + Z) = 0.0160 Db = 114(~ + t - u - v + w + x - y - 2) = 0.0100 Dd= 1/4(s + t - u - v - w - x + y + 2) = 0.0100 Df= 114(~ - t - u + v + w - x - y + 2) = 0.0295 Dg = 1/4(s - t - u + v - w + x + y - 2 ) = 0.0155 Results and Discussion Under the experimental conditions used, hydroquinone had a retention time of 5.0 min. Fig. l ( a ) and ( b ) depict typical chromatograms of a standard solution and skin-toning cream. It can be seen that there is no overlap from other compounds present in the cream as these are not eluted by the relatively weak mobile phase composition employed in the method.Performance Characteristics The limit of detection, which was based on a solution containing 0.001 g 1-1 of hydroquinone and defined as 5 times the standard deviation of this standard, was found to be 0.13% mlm with 4 degrees of freedom. A skin-toning cream containing 2% mlm of hydroquinone produces a solution that gives an absorbance of approximately 0.17. A proprietary skin-toning cream containing 1.694% rnlm of hydroquinone and spiked with 2.000% rnlm of hydroquinone gave a recovery of 3.658 k 0.232% rnlm (95% confidence limits, 4 degrees of freedom). A proprietary moisturising cream base spiked with 2.000% mlm of hydroquinone gave a recovery of 2.045 k 0.0186% mlm (95% confidence limits, 4 degrees of freedom). The full analytical procedure was followed through for fiveANALYST, FEBRUARY 1986, VOL.111 131 Table 3. Results for a range of skin-toning creams Sample No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Test No. 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Hydroquinone, Yo 5.40 5.69 8.30 8.07 2.25 2.20 1.94 1.86 2.09 2.08 5.99 6.28 5.55 5.41 1.90 2.01 1.75 1.60 1.74 1.61 1.88 1.87 1.96 1.90 5.20 5.08 2.06 1.93 2.37 2.24 2.11 2.20 5.10 4.86 1.52 1.56 Sample No. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Test No. 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Hydroquinone, % 2.07 1.89 1.62 1.76 2.08 1.99 1.76 1.74 1.94 1.95 1.62 1.64 2.37 2.49 6.44 6.63 1.72 1.75 1.64 1.71 1.66 1.64 1.98 2.07 1.79 1.76 1.80 1.76 2.18 2.18 2.03 2.11 5.36 5.19 samples of the same tube of skin-toning cream; a mean of 1.694% mlm and standard deviation of 0.017% mlm of hydroquinone were found.Replicate injections of standard solutions of hydroquinone of concentrations of 0.010 and 0.200 g 1-1 gave standard deviations of 0.000299 and 0.00217, respectively (4 degrees of freedom), which is the same order of magnitude to that obtained with skin-toning creams, indi- cating that the extraction procedure does not contribute sig- nificantly to the over-all error of the procedure. Ruggedness Test The results presented so far indicate that the method is precise and without significant bias. However, these results were obtained by a single analyst using a rigidly defined set of operating conditions in one laboratory.It was thought desirable to simulate use of the method in other laboratories by altering slightly the various analytical parameters and determining the effect on the result. The Youden and Steinerl2 model was employed for this test. The ruggedness test is carried out by deliberately varying the factors that are likely to have an effect on the result, from value A to value a. Youden and Steiner give a set of multi-factorial experiments for varying up to seven factors simultaneously (Table 1). From these eight results the effect of each factor can be calculated for a proprietary skin-toning cream (Table 2). The standard deviation (a) of the eight individual results found on this occasion was 0.0334.Any valueof Dlcan be considered significant (P<0.05) if 1 D 1 > v 2 a As no value of D exceeds d- 20 the method can be considered to be rugged for the parameters chosen. v 2 a = 0.0472). Application to Samples The method was applied to a range of skin-toning creams. Four products were waxy in nature and required preliminary dissolution in light petroleum before addition of methanol. Each sample was analysed in duplicate and the results are given in Table 3. Eight products contained hydroquinone well in excess of the 2% mlm permitted by the Cosmetic Products (Safety) Regulations, 1984. In order to obtain some indication of the relative errors of the procedure the difference between the duplicate results was plotted against the mean value for each sample.This plot revealed two distinct clusters at about 2% mlm and 5.5% rnlm of hydroquinone. The ratio of the mean of the differences to the mean of the sample means was taken for each cluster and was found to be very similar at the two hydroquinone levels (0.0368 at 2% hydroquinone, 0.0371 at 5.5% hydroquinone) indicating that the relative errors of the procedure are not dependent on hydroquinone concentration in the sample. The method presented here allows the determination of hydroquinone in a complex sample matrix with relative ease, accuracy and precision and has been shown to be insensitive to changes in many instrumental parameters, permitting its application in a wide range of laboratories. References 1. 2. 3. Spencer, M. C., Arch. Dermatol., 1961, 84, 131. Ridley, C. M., Br. Med. J., 1984, 287, 1537. Council Directive 76/768/EEC, Off. J . Eur. Commun., 1976, L262.132 ANALYST, FEBRUARY 1986, VOL. 111 4. Second Amendment Council Directive 82/368/EEC, Of$ J . Eur. Commun., 1982, L167. 5. Fifth Amendment Commission Directive 84/415/EEC, Off. J. Eur. Commun., 1984, L228. 6. Bush, S. J . , personal communication, 9 November 1984. 7. Popov, A., and Yanishlieva, N., Fresenius 2. Anal. Chem., 1970, 249, 191. 8. Miller, R. L., Chromatogr. Newsl., 1981, 9, 10. 9. Greenlee, W. F., Chisn, J. P., and Richert, D. E., Anal. Biochern., 1981, 112, 367. 10. 11. 12. Sticher, O., Soldati, F., and Lehmann, D., Plant. Med., 1979, 35, 253. Raghavan, N. V., J. Chrornatogr., 1979, 168, 523. Youden, W. J., and Steiner, E. H., “Statistical Manual of the Association of Official Analytical Chemists,” AOAC, Wash- ington, DC, 1975. Paper A51283 Received August 2nd, 1985 Accepted August 22nd, 1985