TECHNICAL ASPECTS OF AEROSOL PACKAGING 357 now also available in this country. The cost of these valves is considerable and, in the long run, there is very little saving. In the United States these valves have failed to make very much headway. FILLING OF AEROSOLS There are two methods for the filling of fingertip dispensers, i.e., low temperature or cold filling and normal temperature filling, also known as pressure or injection filling. Cold filling is the method most commonly used in the United States. It might be well to point out that the propellents commonly used in the aerosol industry were developed as refrigerants, and as such it has been a natural thing to apply refrigerating machinery to this industry. Filling by this method is speedy, but the equipment necessary is fairly expensive. Pressure-filling equipment is relatively cheap but the filling speed is slower because the propellent has to be forced through a small orifice (in the valve) whereas in cold filling the propellent is filled through an opening at least 1 in. in diameter. Even to-day cold filling predominates in the United States because American fillers are obviously loath to dispense with expensive installation. Speed alone does not, however, dictate the choice of either method, and while it is true to say that all packs can be pressure filled it is not always economical to do so. To begin with, not all valves are suitable for pressure filling. The Precision valve, illustrated in Fig. 3, is amongst those valves which can be filled in this manner, while the Continental valve (Fig. 1) is not suitable. Cold filling is normally carried out in the following way: The cooled product is filled into the can, the cooled propellents are added and the can is sealed. The boiling off of a small amount of propellents sweeps out most of the air prior to the sealing of the can. It is impossible to cool water-based products prior to filling and in such cases cans containing the products would have to be cooled, a process which is laborious and definitely uneconomic. A further complication arises in such cases when the cans are immersed in hot water for leakage testing immediately after sealing. Because the contents are solidly frozen this immersion is equal to heating the propellents to 130 ø F. If the propellent consisted solely of dichlorodifluoromethane, as is often the case with shampoo and shaving cream, an internal pressure capable of deforming or even bursting the can would be generated. In pressure filling, the product is filled into the can, which is then sealed. The most common method of removing the air trapped is to connect the container to a vacuum line prior to the injection of the propellent through the valve. A better, more efficient and yet simpler way of removing this air is to purge the can with heavy propellent vapour introduced at the bottom of the can. Any air left in the containers will increase the final pressure of the pack, but pressures in cans purged as described approximate those of the

358 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS same product cold filled, several pounds under the pressure obtained when the conventional evacuation method is used.•0 Pressure filling is not restricted to water-based products. This method is also used for products which are sensitive to low temperatures for instance, some emulsions might break down irreversibly on cooling. Also whenever the amount of the product exceeds the amount of propellent pressure filling is called for. Glass aerosols are nearly always filled by this method. Pressure filling is completely anhydrous, whereas with cold filling elaborate precautions are necessary if the introduction of traces of moisture is to be avoided. FORMULATIONS Perfuming of products is considered an important aspect of the work of cosmetic chemists, and more care than ever has to be taken in the perfum- ing of aerosol products. Errors made in this respect, even if they do not render the product altogether unserviceable, will make them sufficiently offensive to be unsaleable. Perfume compositions consist of several, some- times very many, components, generally fairly complicated chemical com- pounds with widely divergent properties, and each of these compounds is subject to the influence, over a protracted period, of the active ingredient, propellent, container and the interaction between these items. A complete change of character in a number of commonly used aromatic ingredients or their complete degradation may be brought about by traces of hydrochloric acid created, say, as the result of hydrolysis of the propellent. Each individual perfume component may well present difficul- ties in itself, and it will therefore be obvious that a combination of a large number of these components renders the whole picture considerably more complicated. Innumerable reactions may take place, such as the formation of acetals from aldehydes and alcohols, the interchange of other radicals between various esters, dehydration, condensation, cyclisation, etc., all of which are capable of modifying the aromatic character of the composition beyond all recognition. As an example I will quote the experience of a European manufacturer who had been purchasing eau-de-cologne concen- trate from one of the leading perfumery houses for some years. Being enter- prising, he decided to market an aerosol cologne, and without informing his suppliers he proceeded to pack his standard perfume concentrate into glass aerosols. The result was commercial disaster for this line because the particular blend just was not suitable for atomisation. A few months ago, a sample of perfume was sent to our laboratories with a request that it be packed in an aerosol container as a room deodorant. The result was a very irritating spray, and it was quite clear that the customer had no idea of the problems involved. The selection of an appropriate perfume composition, therefore, does not