ERRORS IN THE ANALYSING OF PERFUMERY RAW MATERIALS a7 Langenau, E., and Rogers jr., J.A. 21m. t•erfumer 21romat. 47 (1960) (3) a0 Nogax, S. P., and Safranski, L. W., in Mitchell, jr., S., I•ohltoff, I. M., Proskauer, E. S., and Weissberger, A. Organic 21nalysis 4 141 (1960) (Interscience Publications Ltd., London) a9 James, A. T., and Martin, A. J.P. Blochem. J. 50 •79 (1952) 40 Kirkland, Holmgren, in Coates, V. J., Noebels, H. J., and Fagerson, I.S. Gas Chromatography 38 (1958) (Academic Press, Inc., New York and London) 4• Lund, N.A. _Perfumery t•ssent. Oil Record 50 152 (1959) 4• I•napman, C. E. N., and Scott, C.G. Gas Chromatography 2t bstracts (Butterworths, London). Introduction by the lecturer Gas liquid chromatography (GLC) has become one of the most important tools in the analysis and quality control of perfumery raw materials. Its application has, however, brought in its wake numerous difficulties and inconveniences, simply due to the lack of knowledge of some of the principles that are involved, and due to failure to take certain basic precautions. Equipment and techniques which are known, which may be quite suitable for the analysis of other products, are quite useless with perfumery raw matehals because the latter comprise many constituents which differ widely in their polarities and chemical structures. One has also to take into account the added complication that certain constituents are liable to alteration in the injector, in the column, and in the detector. Because of the complexity of the perfumery materials, the diversity of the chemical characteristics of their constituents, and the even greater diversity in the relative proportions in which they are present in an oil, the identification of components and their quantitative determination impose conditions that are not always taken into account nor completely understood. Where GLC is intended to be used for technical purposes, we ought not to neglect economic considerations either. The analysis must be made reli- able, without the use of pa/nstaking labour on the part of the probably very expensive experts. GLC must be considered merely as an analytical process. The results obtained by its use should and must be confirmed, and must be amplified by data which are obtained by means of other analytical techniques. In its instrumentation it cannot be divorced from the instrumentation of other analytical methods. The application of GLC in everyday technical analysis must go hand in hand with continuous research into better experi- mental conditions, and with an accumulation of exper/ence that is subject to an equally continuous and stringent critical review. DISCUSSION MR. R. N. BEVITT: You have noted vahous changes, undergone by terpenes during gas chromatographic analysis, induced (i) thermally, by reaction in the gas or liquid phase, e.g., isomehsation of ocimene to allo-ocimene,

40 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (ii) catalytically, by adsorption at the support/liquid-phase interface, e.g., decomposition of linalol, isomerisation of •-pinene. Decompositions of type (ii) may be prevented by the use of a liquid phase which is more polar than the components of the sample and is therefore preferentially adsorbed at the solid/liquid interface. The use of a really inert support, perhaps powdered polytetrafluorethylene, is also a possible solution. However, a more general solution might be the use of capillary columns, which may be operated at lower temperatures and have the column material itself as the support. Do you consider that these, and other potential advantages, recommend the capillary column for general use in essential oil analysis, in spite of the reduced sensitivity brought about by the use of a necessarily small sample ? THE LECTURER: I agree that by the use of multi-stage columns, the decomposition is likely to be reduced. Apart from that, the isomerisation must be considered as an important factor because, for example, with allo- ocimenes a temperature of 140 ø is already sufficient to cause the change. There are other examples, such as the change of pulegone into isopulegone, and carvone into spicatone, which are similarly caused. It is, of course, possible to use completely inert supports, such as Haloport, which is made by the Scientific Corporation in the United States. Although Haloport seems to have a very high melting point it does start to agglomerise at temperatures of 180 ø. It must not be forgotten that with capillary columns there are advantages and disadvantages. It is, after all, the perfumer who is working with the product, and in view of the fact that the quantities injected into a capillary column are so minute, the detection, by odour, of the minor constituents often becomes impossible. Apart from that, the very minute- ness of the quantity obtained also makes it almost impossible to use physical methods such as spectrometry of various types. Economy must not be forgotten either. I do not know what a capillary column might cost in England, but in Switzerland it would cost between oe50 and oe100 per unit. In view of the tremendous variety of problems that have to be handled, between 15 and 20 columns might possibly be required, and that represents quite a large investment. DR. }V[. H. KLOUWEN : For the determination of retention values different authors have chosen different standard materials. In order to obtain an index system of highest practical value to science and industry it is desirable that we all agree, as much as possible, on one generally accepted system of standards. Dr. Kovats, of the Ziirich Technische Hochschule, developed a system of retention values determined in relation to the even numbered saturated