An approach to a more realistic cotton detergency test

As fa r back as 1933 however the Detergency Committee of the American Oil Chemists ' Society (1) issued a critical repor t on this test, emphasizing its poor reproducibi l i ty and q u e s t i o n a b l e significance. Ten years later this was followed by an even more adverse A.S.T.M. repor t (2), concluding that the procedure was of little value as a method for ra t ing detergents. Nevertheless the great major i ty of workers in this field have continued to employ a test of this type with only minor variations. In a recent pape r (3) the authors analyzed the shortcomings of this c o n v e n t i o n a l method in some detail, emphasizing in par t icu lar the need for differentiat ing clearly between precision ( reproducibi l i ty) and a c c u r a c y (correlation with field per formance) . I t was concluded that even though precision could be improved by closer control of variables and the use of larger numbers of swatches, the accuracy was inherent ly poor since the test fai led to d u p l i c a t e household conditions in several impor tan t respects, among which were mentioned the following: a) The Launder -Ometer itself departs great ly in design and mechanical action f rom the usual household ag i t a to r type washing machines. Par t icu la r ly undesirable is the relatively feeble washing action compared to the vigorous surge of an oscillating agitator. Since this lack of mechanical action may affect some detergents more than others, it would be preferable to use a degree of agitat ion more closely simulat ing that encountered in household washers. b) The type of soil used in the conventional test is also fa r f rom representative. Although the heavy deposits of lampblack and fa t usually employed might occur on mechanics ' overalls, they are hardly typical of soils normal ly encountered on shirts, sheets, towels, tablecloths, etc., which are usually complex mixtures of inorganic and organic particles conta ining very little free carbon. e) Fur the rmore this soil is applied so heavily that it i s pract ical ly impossible to wash out in a single operation and should real ly be classified as a stain ra ther than a normal household soil. The objective of practical launder ing is complete cleanliness, necessi tat ing removal of the last traces of ingrained soil; the conventional test method however, which washes only pa r t way, removes merely the superficially held carbon, leaving the more tenaciously bound particles untouched. Such a test may therefore be measuring only dispersion of the more loosely held carbon black ra ther than true detergency. In view of these considerations it was concluded in the above paper that a definite need existed for a more realistic cotton detergency test exhibit ing not only good precision but also a h i g h e r deg ree of accuracy. Practical Wash Tests One of the greatest obstacles to evaluating any proposed test method is the lack of definite information on the actual efiieiencies of different detergents in household use. Obviously there is no way of deciding whether the labora tory evaluation of a detergent correlates with practice unless the performance of that detergent in practice is known. Although manufacturers of household l aundry detergents possess some information of this type, based on consumer surveys, such data are extremely difficult to present quantita t ively and are subject to all the errors of personal judgment that inevitably arise in opinion polls. Fur thermore the results are not generally publicized. In order to set up a scale or " y a r d s t i c k " on which a number of detergents could be rated according to their effectiveness in household use, the following procedure was employed to obtain pract ical laundering data : Cotton rol ler towels (12" Huck towelling, 75 x 37 count, 5.7 oz./sq, yd., 75 ft. long) were placed in washrooms about the laboratory for repeated soiling in actual use. Fou r towels were employed in this series of tests, each being labelled with the number of the detergent to be used on it. For example, Towel I was washed with Detergent I a f t e r each soiling, Towel I I with Detergent I I , etc. Since the type of soil in the mechanics ' washroom was heavier than in the office washroom, for example, the locations were r o t a t e d af ter each soiling so that each towel received all types of soil. In order to increase the amount of soil wiped onto the towels, all soap and cleaners were removed f rom the washrooms for the durat ion of the tests and ordinary borax was subst i tuted instead. Af te r each soiling the towels were washed in a May-tag machine, using 0.25% of the appropr ia te detergent at 140~ in filtered Easton t apwate r (70 p.p.m. hardness) . Af te r washing, all towels were dried in a large cabinet to eliminate any bleaching action of sunlight. I n all, eight soil-wash cycles were carried out. In order to follow the progress of the washes the reflectance values of the towels were measured af ter each wash, a square foot at a time, and the results