J. Soc. Cosmet. Chern. 22 3-14 (1071) ¸ 1971 Society of Cosmetic Chemists of Great Britain The analysis of odoriferous vapours, including head space analysis B. DUDLEY SULLY* Presented at the Symposium on "Perfumery", organised by the British Society of Perfumers and the Society of Cosmetic Chemists of Great Britain, at Eastbourne, Sussex, on 21st April 1970. Synopsis--HEAD SPACE ANALYSIS is normally concerned with the analysis of VAPOURS, but the 'perfumer would prefer to know the composition of a liquid that •vill reproduce a given vapour composition. The problem is illustrated by glc experiments with DIBUTYL PHTHALATE containing LINALOL and IONONE (1:4 by volume) which produces a vapour containing linalol and ionone in a weight ratio of more than 15: 1. It is shown that the vapour can be brought into equilibrium with a suspended micro-drop of dibutyl phthalate to reproduce the original liquid composition. EXPERIMENTS show that the slow step in the attainment of EQUILIBRIU• is the diffusion of the odoriferous yapours through the air space and that the rate is much increased by working under a partial vacuum. The procedure is illustrated by the equilibration of a micro-drop of dibutyl phthalate with the odoriferous yapours from a living ROSE. A similar experiment was made with culinary grade THYME. A complete analysis of an odoriferous vapour such as that produced by coffee, strawberries, or flowers such as roses would involve the identifica- tion and quantitative determination of hundreds and perhaps thousands of different organic compounds. The ultimate aim of the chemist is to reproduce an odour by chemical and physical techniques of analysis and without the use of the sense of smell, but at the present time this is extreme- ly laborious if not quite impossible. The identification and determination of perhaps 99•/o or more of the constituents of an odour is possible with the latest techniques, but this is far from sufficient and many have found that a synthetic mixture prepared according to an almost complete analysis of an essential oil may be no more than reminiscent of the original odour. *Bush Boake Allen Ltd., London, E.15.

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The nose is incredibly sensitive to certain odours and for a satisfactory analysis it is necessary to identify and include in the formulation constitu- ents that may be present in quantities less than one part per million and there are claims that much lower concentrations, even parts per billion, are of vital importance. For certain compounds, the nose is more sensitive than the flame ionization detector and in such cases it may be necessary to use spectrometer techniques. Interest in the gas chromatography of yapours began with the introduc- tion of instruments having ionization type detectors, for example those using strontium 90, or tritium. These detectors were far more sensitive than the earlier catharometers and there must be many who repeated the - at the time - fascinating experiment in which the head space yapours from a freshly punctured tin of ground roasted coffee were chromatographed. The interpretation of the chromatogram so obtained is a much more difficult problem. Retention times are often employed, but must be used with care particularly with packed columns and low retention times, and in all cases they should be measured on at least two stationary phases of different polarity. An unambiguous identification of each peak usually involves preparative chromatography whereby the compound producing each peak is separated and identified by physical means, such as a com- bination of infra-red, U.V. and nuclear magnetic resonance spectroscopy. Chemical reactions on a micro-scale are also used to aid complete identifica- tion. An important step forward was made when gas-chromatography was coupled with high resolution mass spectroscopy. With this technique it is possible to determine the molecular weight and probable chemical formula for each peak in the chromatogram and in certain cases it is possible to deduce the chemical structure. Thus, high resolution capillary chromato- graphy columns have been coupled to a fast-scan mass spectrometer to give unambiguous identification of the components of fruit volatiles (1). The work is still tedious because of the number of peaks involved, but it has been made possible by the use of computer techniques which will carry out the necessary calculations and tabulate the molecular weights and chemical formulae. This techn. ique is expensive and probably beyond the resources of most perfumers, quite apart from the fact that it is not the final answer. Many perfumers use gas chromatography to assist in the identification of the components of an essential oil or perfume. The usual procedure is to put the sample to be examined on to a chromatographic column and to divide the separated components at the exit into two streams. One stream