SOME ASPECTS OF THE ANALYSIS OF HOUSEHOLD DETERGENTS 353 10 ml of the detergent solution, about 0.004 N, is put into a stoppered vessel with 25 ml of an acidified solution of methylene blue. 15 ml of chloroform is added and the mixture is shaken. All the indicator migrates to the chloroform phase. The quaternary salt, about 0.004 N, is added in small portions, with thorough agitation after each addition. Near the end point the colour begins to migrate back into the aqueous phase. The end point is taken quite arbitrarily when the two phases are of equal colour. The mechanism of the process is as follows: At first, the indicator reacts with a little of the anionic detergent to form a simple salt, which is completely extracted by the chloroform. As the quaternary is added, it too forms a salt which is extracted by the chloroform. When all the anionic detergent has reacted and been extracted, further increments of quaternary ions compete with the methylene blue for the detergent combined with it. Methylene blue is displaced and returns to the water layer. The equality of colour in the two layers at the end point is a convenient criterion but has no fundamental significance. This method has the advantage of not suffering from interference by any soap which may be present. But the end point is not easy to judge because the two phases are of different tints. In addition, the volume ratio of the two phases is critical, and if the titration differs appreciably from 10 m! the result is not accurate. Further, for both theoretical and practical reasons, the titration does not appear to be stoichiometric. But in spite of these drawbacks, the Epton method has been, and still is, very widely and successfully used. 2. The Barr Method This is one of two described in a paper by Barr et al ø, the indicator being bromophenol blue: Br Br ,-. .or S03Ne This is an anionic dyestuff, i.e. the coloured ion is negatively charged. The fundamental distinction between the Barr and Epton methods is that the former uses an indicator of the same charge as the detergent ions being determined, whereas the latter uses an indicator of opposite charge.

354 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS In the Barr method, the solution to be titrated is made neutral or slightly alkaline. Otherwise the mechanics of the process are very similar to those of the Epton method. In contrast, however, all the indicator remains in the water layer for practically the entire titration, during which the detergent is being converted into its quaternary salt and extracted by the chloroform, as before. Eventually, when almost all the detergent has reacted, the bromophenol blue begins to react, and the salt formed is extracted by the chloroform. The first trace of blue in the chloroform layer indicates the end point. $. The Cullum Method In a later study of the Barr method if was found 7 that the first appear- ance of blue in the chloroform did not mark the true end point, some detergent remaining unreacted in the water layer. On the addition of further increments of the quaternary, these last traces and the indicator react simultaneously, and both quaternary salts are extracted. At the point where the last trace of colour is removed from the water layer, the tendency for the two layers to emulsify reaches a minimum, and at the same time the emulsion type changes from water-in-chloroform to chloroform-in- water. This is easily observed while the two layers are separating after being shaken. Just before the end point the separation consists of the collapse of a network of hollow bubbles of chloroform through the water layer. Just after the end point, the separation consists of the settling out of solid drops of blue chloroform. The change-over is very sharp, one drop (ca.0.02 ml) of the quaternary solution being sufficient to bring it about. Frequently the water layer becomes quite clear at the change- over point. The actual procedure is as follows: Prepare a solution approximately 0.004 N with respect to the surface agent to be determined. Pipette 25 ml into a 200 ml stoppered bottle. Add, by pipette, 1 ml of 0.1% bromophenol blue. Make just acid with 0.1 N hydrochloric acid and restore the blue colour by the addition of a small quantity, about 0.5 g, of sodium bicarbonate. Add 50 ml of chloroform and titrate with 0.004 N cetyltrimethylammonium halide, or some other suitable long-chain quaternary ammonium salt, which has been standard- ised against pure sodium dodecyl sulphate. Add the quaternary in small portions and shake well after each addition. When the colour changes from purplish blue to clear blue, continue titrating in 0.1 ml increments or single drops, until the water layer is colourless and clear or almost clear. For the greatest precision, use the colour transfer only as a guide and take as end point the change-over from bubbles to drops of chloroform as it