COSMETIC KNOWLEDGE THROUGH INSTRUMENTAL TECHNIQUES 131 •3 I I • Stort 5 4 3 a 1 0 TIME (minutes) Figure 3. Gas chromatogram of an oilin toluene solution. Carrier gas, helium flow rate, 65 mi./min. temperature, 153øC. stationary phase, celite coated with silicone oil column length, 2 meters sample volume, 0.02 ml. 1 Start 6 5 4 3 2 1 0 TIME (minutes) Figure 4.--Gas chromatogram ooe "pure" toluene, Carrier gas, helium flow rate, 65 ml./min temperature, 153øC. stationary phase, celite coated with silicone oil column length, 2 meters sample volume 0.02 ml. when "pure" toluene was examined by gas chromatography, it was dis- covered, as can be seen in Fig. 4, that many of the small peaks resulted fiom the so-called "pure" toluene. This is probably true for many of the things which in the cosmetic industry today are called "pure." Many off- odor problems, lack of uniformity, and even allergenic reactions might be traced back to the starting materials. It is well known that "pure" com- pound A from supplier B often cannot be used in a formulation to replace "pure" compound A from supplier C. Here is a source of vexation that might speedily be removed if chromatographic specifications could be set for certain materials.

132 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Whenever a new method or technique is announced, there is, regrettably, a slavish inclination to follow in the footsteps of the pioneers and possibly to push on a few more paces into the unknown. It is important, however, to step back and examine this new development to see if, in some other way, it could better be turned to one's specific needs and problems. The advent of gas chromatography affords a good example. It has been reported (4) that good separations in gas chromatography can be achieved if the combination of inert support and high boiling liquid is replaced by the detergent "Tide" (The Procter and Gamble Company), and it has been found, indeed, that useful separations can be performed with it in the column. Other common household detergents have been ex- amined, and it has been found that they, too, can be used. The separations given by these detergents are not of the same degree. Ira standard mixture of alcohols and esters is separated by each of the detergent columns, dif- ferent patterns are obtained on the recorder chart. Thus, detergents might be identified in a novel way by packing the unknown detergent into a column and using it to separate the standard mixture. The pattern obtained un- der standard conditions might then be matched with one given by a known detergent. Similarly, surface active agents, such as the polyoxyethylene ethers and esters, when used as the inert liquid support in a chromato- lOO lOO z n• I I I I I I I I I I. I 3 4 5 6 7 8 9 10 11 12 1,3 14 15 MICRONS Figure 5.--Spectra of G-2162 and Tween 80. z/ = Tween 80 and B -- G-2162. Both solutions were 5% by weight in chloroform. The spectra were obtained on a double-beam spectrophotometer.