HYGIENE IN MANUFACTURING PLANT AND EFFECT ON EMULSIONS 401 emulsion manufacturing plant, definite recommendations about usage cannot be made. TREATMENT AFTER PLANT STERILIZATION Residues of disinfectant solution must be removed from the plant after sterilization without recontaminating the machinery, and the best way of doing this is to irrigate with sterile water. The risk of recontaminating sterile machinery is high if cold water from a distilled or demineralized water storage tank is used, and boiled or autoclaved water is preferable, particularly if the plant is to be left idle before the next batch of product is made. Ideally, equipment should be washed, sterilized and rinsed immediately before use, but if it is necessary to stop operations at some point during this 3-stage process, the safest point to break off is after chemical steriliza- tion, and before rinsing. Leaving the plant idle after detergent washing or after final rinsing is potentially dangerous from the microbiological point of view unless the machinery is thoroughly dried inside and out. (Received ß 77 th September 7964) REFERENCES (1) Wedderburn, D. L. Advances in pharmaceutical sciences I 195 (1964) (Academic Press, London and New York). (2) Baker, J. H. J. Soc. Cosmetic Chemists 10 133 (1959). (3) Elsman, P. C., Kull, F. C. and Mayer, R. L. J. Am. Pharm. Assoc. Sci. Ed. •t8 88 (1949). (4) Cruickshank, G. A. and Braithwaite, D. G. Ind. Eng. Chem. 41 427 (1949). (5) Sokolski, W. T., Chidester, C. G. and Honeywell, G. E. Developments in industrial microbiology 3 179 (1962) (Plenum Press, New York). (6) Davis, J. G. J. Pharm. Pharmacol. Suppl. 12 29T (1960). Introduction by the lecturer In the cosmetic industry in particular many types of formulations are made in the same plant, and the intervals between making one formula and the next can vary from days to weeks, and the time during which the plant is idle can be very dangerous indeed from a microbiological point of view. Although most of the people concerned with the formulation of emulsions are fully aware of these dangers, plant operatives seldom are and a lot more ought to be done in their training. They are very seldom aware that an adequately preserved emulsion, under clean conditions, will behave very differently when there is massive and accidental contamination, either through faulty plant cleaning or through the use of a certain raw material that might, by chance, have been very heavily contaminated. Factory hands also cannot possibly be expected to know how quickly bacteria can multiply and change raw materials completely. Nor can they be expected to know which of the many raw materials handled are likely to be very heavily

402 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS contaminated, e.g. the gums acacia, karaya and tragacanth, and chalk, kaolin, rice starch and various pigments, often carry a very rich flora themselves. Dilute solutions of anionic and nonionic detergents, left in a plant between clean- and sterilizing can lead to very heavy contamination as micro-organisms, particularly the gram-negatives like Pseudomonads will grow quite readily in these solutions. I did mention the importance of having plant that is easy to dismantle, and can therefore easily be cleaned. Dismantling is therefore important, as is cleaning. The plant should be washed thoroughly using mechanical means if necessary, particularly with w/o emulsions, which are not easily rinsed and removed from the plant. Follow this by sterilization, and then by rinsing. These four procedures can very seldom be carried out in quick succession without a break. In my opinion it is important not to have a break after the cleaning process, because of the danger of having product residues diluted with anionic detergents that can easily and very quickly be metabolized by micro-organisms. If a break is necessary, it should be either after dismantling but before cleaning, and preferably after sterilization. In that event the sterilizing fluid is left in the plant during the idle time. One should take care that rinsing is not carried out with water from contaminated demineralized storage units. Either boiled, or very hot water which is unlikely to carry rich flora, should be used for this purpose. DISCUSSION MR. N.J. Vx• ABBg: I believe that the question of plant hygiene as discussed by you, is of great importance whether or not a finished product includes a pre- servative. For this reason, I should like to make one or two suggestions. In the first place, mains, distilled or demineralized water is a principal constituent of many products, but few companies, I believe, observe the same quality control over it that they apply to other raw materials. Bearing in mind the labyrinth of pipework through which water usually passes on its way to the production plant, and which may well result in contamination, I would strongly advocate the setting-up of internal bacteriological standards for water, and routine checks on this. Secondly, I feel that connections of plastic hose to metal pipework nozzles, as shown in Fig. $, offer a distinct contamination risk. A film of stagnant product is liable to remain between the plastic and the metal and this could easily serve as a reservoir of contamination for the bulk passing through. If joints of this type must be used, it is always prudent to fit a Jubilee clip around the plastic as close to the nozzle outlet as possible. TI•E LECTURER: I agree that daily bacteriological tests ought certainly to be carried out if one is suspicious of the water supply. Concerning the plastic pipe, I would not really think that the addition of a Jubilee clip would solve the problem. I would hope that the whole assembly would be taken apart and cleaned thoroughly, and not cleaned in situ. It is one of the advantages of using plastic pipe that one can take it off and discard it quite frequently, because it is not very expensive, and is very easy to replace. MR. G. SYKES: In terms of water supplies, I think that deionized water is the most dangerous product that has been introduced into the pharmaceutical industry in recent years, because if one runs a column continuously it appears to be all right. If, however, one stops the system overnight, and certainly over the weekend, an