JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS six months or a year after manufac- .ture, the failure can usually be attributed to an easily detected chemical or physical change. But this is not always the case, because occasionally a product will be re- turned because it does not 'perform properly, although it did the job when it was manufactured it looks and smells the same, and routine ... chemical analysis discloses no chemi- cal change in the product. •A chemi- cal change may in fact have occurred, but it is.not always a simple matter to determine precisely what slight and subtle change has taken place. Examples of functional instability are: the aqueous wave set which is still bright, viscous, stringy and pleasant smelling, but flakes badly when dry on the hair--which it did not do when it was freshly made. Another is the oxidation'type of hair colouring that looks and tests .the same as it always did, but it will no longer colour the hair--or it does colour the hair, but not in the original shade. TIME AS A DETERMINING FAcToR The point to make here is that even if a product passes all the usual accelerated stability tests it should be tested frequently by actual use, to be certain that it does the job it was designed to do. The actual sample that was used for the stability .tests should be used for the per- formance tests and compared with a sample of the product that did not receive the accelerated stability. tests.. In formulating a new cosmetic 182 product adequate shelf-testing is a vital necessity and, a/though acceler- ated stability tests have great value, there is still no substitute for time. Accelerated stability tests are not infallible and, if the maximum benefit is to be derived from them, an understanding of their limitations is essential. I have had the exasper- ating experience of seeing a product pass every accelerated stability test, but samples from the same batch that were not put through the tests went bad. It is not unusual to have a product pass all stability tests, only to have it develop some defect many months later. I can recall several such examples, but one will suffice. A company manufactured a cosmetic preparation in the form of a thick, pearly, liquid emulsion-- a beautiful product 'which had passed all the stability and shelf tests in the laboratory. Several months after distribution' of the product began, it' started to come back be- cause the emulsion was speckled i throughout with large. white Microscopic and other tests identified these specks as .crystalline clusters-i: • of cetyl alcohol. However, :. laboratory was never able to bring':::. about this condition using goodl samples from the same batch that. :•i went bad. ß What cosmetic chemist has not:i, had the experience of formulating new product in the laboratory, only •( to find that the first .production:•i?i batch or even a pilot batch, did have exactly the .same properties?:): It is for this reason that a laboratory'? prepared sample of a product is

STABILITY TESTING OF :. for use 'in any stability programme. Two or more or production batches, prefer- made from different lots of materials, should be used for the •final tests if they are to have any value. With these facts in mind, and with understanding of the limitations accelerated stability tests, I can •i%•i'• on-to discuss such tests for specific :{)!!.•i?ihair preparations. 'Raw material 2':•i•i': Control will not be included, although •'• •/•:?i'its importance' should not be mini- ½•!:i!:mised. The stability or instability ? ?of a cosmetic product very often !'?:i'depends upon the quality and control :)71( iof the raw materials used. :i i •!: 5 •' I shall limit my remarks to three ) !::)!ilVery popular t?pes of hair prepara-. ! ?):?}!'tions, namely cold wave solutions, :J?: shampoos, and hair-setting cosmetics. (iT!:/:'.:Dr. William Colburn has .already •.'•::':::::' ably discussed the stability testing :,:i!!i!.1øf emulsions (this Journal, Vol. 2, 3), and so I shall not attempt .. to discuss emulsified hair creams and lotions. TESTING COLD WAVE :. PREPARATIONS !i:•'•:" Most cold wave preparations 'on the market to-day consist essentially of an aqueous solution of thiogly- collic acid (usually as the ammonium !•i:.: salt) and an excess of alkali to bring :!',( the' pH to about 9'0-9'6. They' : may also contain a wetting agent to facilitate penetration into the hair, a clouding or opacifying agent to ": : impart a milky white appearance to the product, perfume, and colouring matter. The solution is usually HAIR PREPARATIONS packaged in 3 or 4 ounce bottles that are either sealed with a paraf-' fined cork plus a metal cap or a' plastic cap that has a vinylite liner. There are a number of difficulties that can occur in a packaged cold wave solution: 1. The thioglycollate concentra- tion can drop. The active content of thioglycollate salt can be deter- mined by iodimetr/c titrationi This drop can be due to inferior raw matedhals or to a defective seal. Thioglycollic •cid is easily oxidised to dithioglycollic acid, especially in alkaline solution in the presence of traces of metals, such as coppro or manganese, which catalyse the oxida- tion. 2.'The pH can drop..The pH of cold wave solutions must always be kept within a rather narrow range in order that the solution shall function properly when applied to the hair. Reaction of the alkali with the wetting agent or with the emulsified oils may cause the pH to drop. 3. Discoloration of the liquid ma.y occur." A pink colour is caused by. iron, which may have been present in one of the raw materials, in dirt or dust in the bottle, or may have got into the solution by corrosion of the metal cap. Only one part of iron in four million is needed to' turn the solution pink, so the impor- tance of keeping cold wave solutions free from iron is obvious. A yellow discoloration may appear, especially where a poorly paraffined cork has been used to seal the bottle. The tannin becomes extracted from the cork, resulting in the off-colour. ß 183