Analyst, July, 1973, Vol. 98, @. 493-501 493 The Determination of Di-n-alkyl Phthalates in Cosmetic Preparations by Gas - Liquid Chromatography BY E. W. GODLY AND A. E. MORTLOCK (Department of Trade and Industry, Laboratory of the Government Chemist, Cornwall House, Stainford Street, London, SE 1 9NQ) An improved gss-chromatographic method for the direct determination of C,-C, di-n-alkyl phthalates in toiletry samples that contain ethanol is des- cribed and a range of perfume essential oils and perfume synthetic chemicals is examined for possible interference. MANUFACTURERS of preparations that contain ethanol are required by H.M. Customs and Excise to include denaturants in order to render the preparations unpotable. The suitability of newly proposed denaturants and their effective concentration are determined according to the recommendations of the Laboratory of the Government Chemist, which examines also the finished products to ensure that the required level of nauseousness is maintained.The list of accepted denaturants is large and increasing and, as the samples examined are drawn from the entire cosmetic range and indeed extend to any retailed preparations (other than beverages) that contain spirit, special analytical problems arise. Diethyl phthalate is commonly used as a denaturant of toiletries because it is usually compatible with other ingredients and has properties that may be convenient, e.g., as a perfume fixative or plasticiser. Its inclusion at a minimum level of 1 per cent. V/V in per- fumes made with Q-grade industrial methylated spirit to the Statutory Formula I1 is manda- tory.1 In a previous paper2 from this laboratory, three methods for the determination of diethyl phthalate were described as follows: by (i) direct gas - liquid chromatography with two alternative column systems; (ii) isolation of the diethyl phthalate by column cliromatography followed by measurement of optical density at 227 nm ; and (iii) gravimetric determination as phthalanil, which involves alkaline hydrolysis, liberation of the free phthalic acid by acidification and extraction into diethyl ether followed by treatment with aniline to give a precipitate that can be weighed.The acidification stage provides the possibility of determining the plithalic acid by back-titration. Formerly, the only comparable published gas - liquid chromatographic method for the determination of diethyl phthalate3 involved the use of 30 per cent.sodium dodecylbenzene- sulphonate as the stationary phase and a thermal conductivity detector. Although this system yielded symmetrical peaks, the time required for a single analysis was excessive. By making use of low loadings of polar phases that were stable and non-fugitive at high temperatures, Hancock, Rose and Singer2 were able to shorten the analysis time and thus provided a useful routine method. Some tailing of peaks was experienced but this was not considered unacceptable. The stationary phases were fluorosilicone oil 1;s 1265 and silicone gum rubber SE-30, which were used in each instance a t a loading of only 1.5 per cent.m/m on HMDS-treated Chromosorb W. These authors compared results obtained from a range of toiletry samples by the gravimetric method with those obtained by ultraviolet absorption measurements, and those for gas - liquid chromatographic determinations, including also the results calculated from the phthalic acid values obtained by back-titration. Agreement was generally good in both instances but some discrepancies were noted with the comment that the chroma tographic method served as a screening technique to overcome the non-specificity inherent in the other methods. It was frustrating that the FS 1265 system, although less prone to tailing, and therefore giving better chromatograms than those obtained with the SE-30 system, failed to resolve diethyl phthalate from isopropyl myristate, a fairly common toiletry ingredient, while, even with SE-30, with which this resolution was complete, it was found that commercial isopropyl myristate contained an impurity that eluted close to die thy1 phthalate .0 SAC; Crown Copyright Reserved.494 GODLY AND MORTLOCK: DETERMINATION OF PHTHALATES [Analyst, VOl. 98 In more recent work in which the determination of diethyl phthalate in small-arms propellents is de~cribed,~ the stationary phase used was 20 per cent. SE-30 on an acid-washed Chromosorb FV support at 200 "C, and the published chromatograms show that, even with such high loading of the stationary phase, the problem of tailing had not been entirely elimin- ated. The system for the determination of diethyl phthalate now proposed is, as previously ~laimed,~ free from the interference complained of and providess ymmetrical peaks at retention times compatible with routine analysis.I t is applicable to many other denaturants including dimethyl phthalate and, after slight modification, to di-n-butyl phthalate. APPARATUS- A Pye, Model 104, gas chromatograph with a flame-ionisation detector was used. CoEwumns-Usually glass, 1-52m x 3 mm i.d. (5 feet x Q inch o.d.), but stainless-steel columns have also been used without difficulty. Packiwg-This consists of 8 per cent. of nonylphenoxypoly(ethy1eneoxy)ethanol on 80 to 100-mesh acid-washed DMCS-treated Chromosorb W. The prepared column contains 5.6 to 5.7 g of packing and should have an efficiency of 1200 to 1800 plates per metre, as measured on the peak obtained from a 1-pl injection of 2 per cent.diphenylamine in ethanol. Columns are packed under gentle suction to within 11 cm of the injection head and both ends are plugged with glass string. The stationary phase, also known as Antarox CO-990 and Igepal CO-990, has been mentioned6 in connection with the determination of dimethyl phthalate in propellent plasticisers but under slightly different conditions. Its recommended temperature range is 50 to 225 "C. CoZwnn conditioni.Yzg-Conditioning overnight at 225 "C with nitrogen passing through the column a t the rate of 5 ml min-l has proved satisfactory. Carrier gas-Nitrogen at a flow-rate of 55 to 60 ml min-l. Column te.lnperntwre--For dimethyl phthalate and diethyl phthalate, 200 to 210 O C , and The temperature of both the injection block and detector oven was 250 "C, and the for di-n-butyl phthalate, 220 "C.attenuation was x 5000. The detector response is measured from a chart recorder as peak height. 1 4 I I 15 10 5 0 Time/m i nutes Fig. 1. Gas chromatogram of mix- ture A. Peaks: 1, ethanol; 2, methyl salicylate ; 3, 1-chloro-4-nitrobenzene ; 4, isopropyl myristate ; 6, dimethyl phthalate ; 6, dicthyl phthalate; 7, di-n-propyl phthalate; 8, diphenylamine; and 9, di-n-butyl phthalateJuly, 19731 IN COSMETIC PREPARATIONS BY GAS - LIQUID CHROMATOGRAPHY 495 Injections were made from a 1-pl syrinse fittcd wit11 an ll-cm needle. Uri1ms otherwise stated, p!urr:ss. grade cli~~micn!s wr.v used ; the ethanol uszd was absolute ethanol. Fig. 1 shows a c l i ~ o i i ~ ’ ohk+~cd i~r~d~ar t!jnsz conditions a t 205 “C from a 0.2-pl injcciion of ‘ ‘ I ~ ~ Y ~ Z I P C A,” a u n 01 riiiwd didkyl p:i tl,ilatcs and some otbcs compounds of interest in ethanol a t conrr-n% ions 1mgir-g i1-0m 0.1 to 2.0 g per 100 ml, ai2J in Table I are listed the retention t k c s o€ tile constitu:-xts o! this mistwe in the ordei- oT their elution.TABLE I ELUTION OF CONSTITUENTS OF “MIXTURE A” Compound Ethanol . . . . Methyl salicylate . . l-Chloro-4-nitrobenzene Jsopropyl myristate Dimethyl phthalate . . Diethyl pl-ithalate . . Di-n-propyl phthalate Diphenylamine . . Di-n-butyl phthalate Retention tirnc/miiiutes .. . . 0.4 . . .. 1.2 . . .. 1.8 . . . . 2.2 . . .. 4.0 . . .. 4.9 .. . . 7-65 ,. .. 9.0 .. .. 12.8 R* 0.08 0.24 0.37 0.45 0.82 1.53 1.84 2.61 (1.00) TEST OF LINEARITY OF DETECTOR RESPONSE- A series of standard solutions containing from 0-5 Lo 4.0 per cent.V/V of diethyl phthalate in ethanol was preparcd and each solution was then mixed with an equaI voilime of a solution containing 2 g of diphenylamine in 100 ml of ethanol. Each mixture was injected under the conditions described and the peak heights were measured for both components. In Fig. 2, the peak height ratio of diethyl phthalate to diphenylamine is plotted against the initial dietliyl phthalate concentration and is clearly linear within the limits of measurement Y m a 0 1.0 2.0 3.0 4.0 Diethyl phthalate/per cent. V/V Fig. 2 . Detector response over a range of cliethyl phthalate concentrations 0.6 0.7 0.8 0.9 Peak height ratio of dimethyl phthalate t o internal standard 13:.3. Calibralioii graph for de tel-in {nation o 1 -1 ime;liyl plitlia- late. liatio: A, of di1me“lyl phtha- late to clipheriylami~ie, :itid B, of djirlcihvl phthaLtte io I rhloro- 4-17 I I obenzene. x , ui (5 Eor first day; and and 0, rcsuitc for second day during continuous running496 GODLY AND MORTLOCK: DETERMINATION OF PHTHALATES [AndySt, VOl. 98 over the phthalate concentration range 0.2 to 2.0 per cent., which is twice the normal de- naturant level. ANALYTICAL METHODS DIMETHYL PHTHALATE- Prepare a standard solution containing 0.5 per cent. V/V of dimethyl phthalate and 1.0 g of diphenylamine per 100 ml of ethanol. Make successive injections of this solution over the volume range 0.1 to 1.0 p1 under the conditions described above.Measure the peak heights for dimethyl phthalate and diphenylamine and then plot the dimethyl phthalate peak height as ordinate against the peak height ratio of dimethyl phthalate to diphenylamine as abscissa (Fig. 3A). This peak height ratio is 0.6 to 0.7, which enables samples containing excess of dimethyl phthalate up to 50 per cent. above the legal denaturant minimum to be analysed at a single attenuation without dilution. To a known volume of each sample add an equal volume of a standard solution containing 2 g of diphenylamine per 100 ml of ethanol (for convenience, 2-ml aliquots were dispensed from pipettes into 10-ml McCartney bottles). Inject the mixture and measure the heights of the peaks due to dimethyl phthalate and diphenylamine. Determine the ratio of the former to the latter (x) and read off the ratio (y) corresponding to each dimethyl phth2late peak height from the standard graph. The concentration of dimethyl phthalate in the sample is then x/y per cent. V/V.If the standard graph is considered to be close enough to the vertical, plotting of ratios can be dispensed with and an average ratio applied to any volume injected; the calculation then becomes one of simple proportion. The same procedure can be followed with an alternative internal standard. If a solution containing 0.6 g of 1-chloro-4-nitrobenzene in 100 ml of ethanol is substituted throughout for the 2 per cent. diphenylamine solution, analytical results are obtained in exactly the same way. The only differences are that, as 1-chloro-4-nitrobenzene elutes before dimethyl phthalate, the duration of each analysis is reduced while the risk of interference from perfume constituents is increased. A typical standard graph obtained with this internal standard is shown in Fig.3B. DIETHYL PHTHALATE- The analytical procedure is exactly as described above for dimethyl phthalate and the same two alternative internal standards are used. Fig. 4 shows a standard graph with diphenylamine as internal standard, to which points were added over a period of 8 days' continuous running. DI-n-PRoPYL PHTHALATE- This ester has not yet found use as a denaturant but it has been included in order to complete the C,-C, series. The method is as described for dimethyl phthalate except that the temperature of the column oven is set at 215 "C.Diphenylamine is the appropriate internal standard and, under these conditions, the retention times are 4.9 minutes for di-n-propyl phthalate and 5.6 minutes for diphenylamine. DI-n-BUTYL PHTHALATE- The method is as described for dimethyl and diethyl phthalates but with the following modifications: the temperature of the column-oven is 220 "C; the nitrogen flow-rate is 80 ml min-l; and the two alternative internal standards are diphenylamine and methyl fi-hydroxybenzoate, each at a concentration of 2 g per 100 ml of ethanol. Under these conditions, the retention times are diphenylamine 5.4, di-n-biityl phthalate 7.3 and methyl 9-hydroxybenzoate 11.4 minutes. Fig. 5A shows a graph of di-n-butyl phthalate peak height against the peak height ratio of di-n-butyl phthalate to diphenylamine over a 2-day period, while Fig.5B shows the graph obtained over a similar period when the diphenylamine was replaced with methyl $-hydroxy- benzoate. Turbidity occasionally results when the sample is mixed with the solution of internal standard. The solution can sometimes be clarified for long enough to permit an injection of homogeneous liquid by gently warming the capped bottle, but if this method fails, dilutionJuly, 19731 IN COSMETIC PREPARATIONS BY GAS - LIQUID CHROMATOGRAPHY 497 of the mixture with a known excess of ethanol can be combined with appropriate reduction in attenuation. Alternatively, the internal standard can be replaced with the second choice. I I I 0.5 0.6 0.7 Peak height ratio of diethyl phthalate to diphenylamine Fig.4. Calibration graph for determination of diethyl phthalate. Continuous running for 8 days: dif- ferent symbols indicate results obtained on different days Peak height ratio of di-n-butyl phthalate to internal standard Fig. 5 . Calibration graph for determination of di-n-butyl phthalate. Ratio : A, of di-n-butyl phthalate to diphenylamine; and B, of di-n-butyl phthalate to methyl p-hydroxyben- zoate RESULTS RECOVERIES FROM SOLUTIONS OF PURE PHTHALATES I N ETHANOL- A range of standard solutions of each ester in ethanol was prepared; with di-n-propyl phthalate a single concentration was considered sufficient. Each ester was weighed, dissolved in and made up to volume with ethanol in calibrated flasks at 20 "C and the solution was then taken for analysis as an unknown. Three to ten determinations were made at each level by the methods described above as appropriate to each ester.Mean recoveries are given in Table 11, where the percentage of ester in the prepared solutions has been converted for comparison from grams per 100 ml into per cent. V/V by means of published density data at 20 "C. These results provided assurance that the system was functioning correctly and was capable of yielding significant results. RECOVERIES FROM TOILETRY SAMPLES- For these recovery experiments toiletry samples were chosen to represent various types of goods received for analysis. In each instance the content of the ester concerned had been found to be nil or negligible. Each sample was treated as follows: a 6-ml aliquot was transferred into each of four 14-ml McCartney bottles, 2.0, 1.5, 1.0 and 0-5 ml of ethanol were added to successive bottles followed by 0, 0.5, 1.0 and 1.5 ml, respectively, of the phthalate concerned in ethanolic solution at a concentration of 10 per cent. V / V , all additions being made by means of graduated pipettes.These mixtures were then analysed by the appropriate procedure as described above. The results are set out in Tables 111, IV and V.498 GODLY AND MORTLOCK: DETERMINATION OF PHTHALATES [A%&!@, VOl. 98 TABLE I1 RECOVERIES OF PURE ESTERS FROM ETHANOLIC SOLUTION Found by analysis, per cent. Ester dispensed, Ester per cent. V / V Dimethyl phthalate 0.18 0.45 0.75 0.99 1-38 1-77 Diethyl phthalate Mean recovery, per cent.0.22 0.47 0.80 1.25 1.51 1-38 Mean recovery, per cent. Di-n-propyl phthalate 1.10 Di-n-butyl phthalate 0.20 0.40 0-79 1.19 1.98 Mean recovery, per cent. Against l-chloro-4-nitrobenzene 0.18 & 0.01 0.47 & 0.01 0.76 f 0.02 0.99 f 0.01 1.38 f 0.02 1.74 f 0.02 102 0.21 f 0.01 0.47 f 0-00 0.80 f 0.00 1.25 f 0-02 1.52 & 0.02 2.02 0.02 100 Against methyl p-hydroxybenzoat 2 0.38 f 0.00 0.79 f 0.03 1-22 f 0.03 2-13 f 0.03 99 - 0-19 f 0.00 > Against d j phen ylamine 0.19 f 0.01 045 f 0.02 0.74 f 0.00 0.99 f 0.01 1-37 5 0.01 1.76 f 0.02 99 0.22 & 0.00 0-48 & 0.00 0.81 & 0.01 1.27 f 0.01 1.52 f 0-01 1.96 f 0-01 101 1-09 0.03 Against d jphen ylamine 0.19 f 0.00 0-38 i 0.00 0.79 f 0.00 1.21 f 0.01 2.14 & 0-03 100 TABLE I11 RECOVERY OF DIMETHYL PHTHALATE Dimethyl phthalate recovered, per cent.V / V , after addition of- 7- > Description 0.5 1.0 1.5 Internal standard of sample Nil per cent. per cent. per cent. Diphenylamine Cleansing lotion Negligible 0-44 1.03 1.48 After-shave lotion 0 0-52 0.98 1.51 Toilet water Negligible 0.51 0.96 1.52 Eau-de-cologne 0 0.49 1.08 1.50 Skin tonic 0 0.50 1.02 1.50 Mean recovery per cent. 98 101 100 1-Chloro-4-nitrobenzene Perfume 0 0.49 0.95 1.49 Friction lotion 0 0.50 1.02 1-52 After-shave lotion 0 0.5 1 1.10 1.50 Bay-rum* 0.02 0.51 (0.49) 0.99 (0.97) 1.57 (1.55) Mean recovery per cent. 100 100 101 * Figures in parentheses were used in calculation of mean recovery. EXAMINATION OF PERFUME ESSENTIAL OILS AND PERFUME SYNTHETIC CHEMICALS FOR POSSIBLE It is clear from the results that the method provides a quick and efficient analytical pro- cedure for the determination of these toiletry denaturants.However, it is appreciated that dependence on a single-column system fails to eliminate the possibility of undetected inter- ference. It was, therefore, a matter of immediate interest to examine a range of typical INTERFERENCE-July, 19731 IN COSMETIC PREPARATIONS BY GAS - LIQUID CHROMATOGRAPHY TABLE IV RECOVERY OF DIETHYL PHTHALATE Diethyl phthalate recovered, per cent. V / V , aftcr addition of- 499 Description Internal standard of sample Diphen ylamine Lavender water Hair lotion A Hair lotion B Spirit shampoo* Alter-shave lotion Vegetable hair tonic Setting lotion? Tonic cleanser Toilet water Skin tests Nil 0 0 0 0 Ncgl j gible 0 0.03 0 Negligible 0 Mean recovery per cent.l-Chloro-4-nitrobenzene Deodorant 0 Toilet water A 0 Toilet water B 0 Skin cleanser 0 After-shave lotion Negligible Friction lotion 0 Perfume Xegligible Skin tonic 0 Mean recovery per cent. 0-5 per cent. 0.5 1 0.50 0.50 0.41 0.50 0.50 0.50 (0.47) 0.48 0.50 0.51 99 0.49 0.50 0-51 0-5 1 0.50 0.49 0.50 0-49 100 1.0 per cent. 0.99 1.01 1.02 0-90 1.00 1.00 0.99 0.96 1.00 100 1 *oo 0.99 0.99 1.08 1.00 1-03 1.03 0.98 101 1.02 (0.99) 7 1-5 per cent. 1.51 1.50 1.50 1-45 1.50 1.51 1-51 (1.4s) 1.52 1-65 1.50 101 1.48 1.52 1-54 1.51 1.6 1 1.49 1.49 1.52 101 * Persistent sediment on mixing. t Figures in parentheses used in calculation of mean recovery. pErfume synthetic chemicals and some available perfume oils for this possibility. The perfume chemicals were injected as 10 per cent.V/V solutions in ethanol but the perfume oils were injected undiluted with increased attenuation in order to demonstrate trace constituents. In most instances, elution of all of the compounds was complete within 2 minutes of injection under the conditions of the method. In Table VI the perfume synthetic chemicals are listed alphabetically, giving the retention time of the principal constituent and its RDEP value (diethyl phthalate = l), together with the number of other constituents; figures in parentheses convey the number of separate peaks that can be ascribed to trace constituents. TABLE V Internal standard Diphen ylamine RECOVERY OF DI-n-BUTYL PHTHALATE Di-n-butyl phthalate recovered, per cent. V / V , after addition of- A r 7 Description 0.5 1.0 1.5 of sample Nil per cent.per cent. per cent. Toilet water 0 0.51 0.98 1-52 Hair lotion Negligible 0.47 1.02 1.48 After-shave lotion Negligible 0-53 1-02 1.55 Skin cleanser 0 0.50 1.00 1.52 Mean recovery per cent. 100 100 101 Methyl p-hydroxybenzoate Skin cleanser 0 0.49 1-04 1.56 Hair lotion Negligible 0.53 0.98 1-46 After-shave lotion 0 0.49 0.99 1-50 Perfume 0 0.53 1.04 1-53 Mean recovery per cent. 102 101 101500 GODLY AND MORTLOCK: DETERMINATION OF PHTHALATES [Analyst, VOl. 98 In some instances spectrographically pure chemicals might have been obtainable but the interest in this survey applied equally to impurities, isomers, etc., that are normally found in commercial-grade products. Comparable results for some perfume essential oils are listed in Table VII, except that no RDEP values are given because, in many instances, no single constituent predominated.Compound Pentyl acetate . . Benzyl benzoate . . Rornyl acetate . . Butyl acetate . . Cinnamyl acetate . . Citronellol . . . . Citronellyl acetate. . Citronellyl butyrate Ci tronellyl isobutyrate Citronellyl valerate Diethyl nialonate . . Diethyl phthalate . . Diethyl succinate . . Diethyl tartrate . . Diethyl sehacate . . Ethyl acetate . . Ethyl butyrate . . Ethyl octanoate . . Ethyl cinnamate . . Ethyl cyanoacetate Ethyl heptanoate . . Ethyl hexylacetate Ethyl lactate . . Ethyl palmitate . . Ethyl phenylacetate Geraniol . . . . Geranyl acetate . . Geranyl propionate Isopropyl palmitate D-Limonene .. Linalol . ... Linalyl butyrate . . Linalyl propionate Menthyl acetate . . Menthyl salicylate. . Musk (synthetic) . . Nerol . . .. Nerolidyl acetate . . Xerol idyl propionate a-Pinene . . .. Pulegol . . . . Sextol phthalate . . Terebene . . .. a-Terpineol .. Terpineol . . .. Terpinoline . . a-Terpinyl acetate. . Te trah y droger aniol .. .. .. .. . . . . . . . . . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . .. .. .. . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. TABLE V I PERFUME SYNTHETIC CHEMICALS Retention time of principal No. of constituent/minutes RDEP* impurities . . 0.3 0.05 1 .. 11.4 2.07 ,. 0.9 0.17 i + (2) . . 0.4 0.08 (2) .. 3- 1 0.56 1 + ( 8 ) .. 1.0 0.1s (11) .. 0.9 0.17 1 . . (14 constituents in 3.7 minutes) .. 1.0 .. 1.6 .. 0.6 .. 6.5 ..0.8 .. 4.4 .. 11.2 .. 0.2 .. 0.3 .. 0-6 .. 3-0 .. 0.9 .. 0.4 .. 0.5 .. 0 4 .. 4.6 .. 1-2 .. 1.1 .. 0.8; 1.0 .. 0.2 .. 4.6 .. 0.4 . . 0.6 .. 0.9 .. 0.S .. 0.7 .. 7.3 .. 9.7 .. 1.0 .. 0.6 .. 3.0 .. 0.3 .. 0.7 .. 14 .. 0.6 .. 0.9 .. 0.9 .. 0.4 .. 1-1 .. 0-7 * DEP denotes diethyl phthalate. 0.18 0.30 0.12 1.00 0.15 0.80 2.14 0.05 0.05 0.7 1 0.54 0.18 0-08 0.10 0.08 0.84 0.23 0.2 1 0.14; 0.18 0.04 0.83 0.09 0.S 1 0.17 0-15 0.12 1.32 1.77 0.18 0.11 0-64 0.06 0.14 1.54 0.10 0.17 0.17 0-07 0.2 1 0-14 It will be noted that, on this evidence, the risk of interference is small. The first con- stituent peak obtained from Indian sandalwood oil is the only one of all those examined which would affect the validity of the analysis. The ultraviolet absorption spectrum of a solution of diethyl phthalate in ethanol was compared with that of a solution of Indian sandalwood oil in ethanol at the same concentration.The sandalwood oil was shown to be free from diethyl phthalate and the ultraviolet method would therefore resolve any difficulty with formulations that contain both of these substances.July, 19731 the Government Chemist for permission to publish this paper. IN COSMETIC PREPARATIONS BY GAS - LIQUID CHROMATOGRAPHY 501 The authors thank Mr. A. J. Blake and Mr. G. A. Pask for technical assistance and TABLE VII PERFUME ESSENTIAL OILS Oil (with source, if known) Oil of fennel (U.S.S.R.) . . Geranium bourbon essence Juniper berry oil (U.K.) . . Lavender oil . . .. Lavender exotic . . .. Lavender . . .. .. Lavender abrialis (France) Lemon oil (Italy) . . . . Oil of limes (U.K.) .. Oil of marjoram . . . . Myrcene . . .. . . Oil of rosemary . . .. Litsea cabeba (China) . . Mace oil . . . . .. Oil of rosemary (Spain) . . Oil of rosemary (Peru) . . Sandalwood oil (India) . . Tangerine oil . . . . Oil of thyme, red . . . . Oil of thyme, white . . Oil of ylang ylang (U.K.) Oil of ylang ylang (France) Diethyl phthalate . . .. . . . . . . . . .. .. .. .. . . . . .. .. . . . . . . .. . . .. . . . . . . .. .. Retention times of principal constituents/ minutes 0.5; 1-3 0.7; 0.9; 1.1 0.4; 0.4 0.5; 0.7; 0.8 0.4; 0-6; 0.7 0.4; 0.6; 0.8 0.2; 0 . 5 ; 0.5 0.5 0.3; 0.4; 0.9; 1.0 0.4; 1.0 0.3; 0.4; 0.7; 4.1 0.5; 0-5; 0.7; 0.9 0.4; 4.1; 4.7 0.3; 0.4; 0.8 0.4; 0.5; 0.8 0.5; 0.9 5-4; 6.7 0-4; 0.5 0.6; 3.2 0.6; 3.2 0.5; 0.6; 1.0; 1.2; 1-3 0.8; 0.9; 1.1; 1.2 5.5 Number of subsidiary constituents 9 14 12 8 10 8 9 8 4 15 12 20 9 6 6 14 13 5 14 10 13 15 - Retention time of final constituent/ minutes 9.1 5.0 4.1 2.4 2.4 2.3 2.4 1.5 1.2 4.7 4.15 4.9 4.7 2.4 2.4 4.8 9.0 2.7 8.5 3.4 5.2 5.2 - REFERENCES 1. 2. 3. 4. 5. 6. “The Methylated Spirits Regulations,” S.I. 1952 No. 3320, H.M. Stationery Office, London. Hancock, W., Rose, B. A., and Singer, D. D., Analyst, 1966, 91, 449. Baines, C. B., and Proctor, K. A., J . Phavm. Pharmac., 1959, 11, 230T. Norwitz, G., and Apatoff, J. B., J , Chromat. ?ci., 1971, 9, 682. “Report of the Government Chemist, 1968, Tunstall, F. I. H., Analyt. Chem., 1970, 42, 542. H.M. Stationery Office, London, 1969, p. 12. Received December l3th, 1972 Accepted March 16th, 1973