858 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The results obtained in these experiments indicate that: 1. The acid rinse lessened the cation uptake by 34%. In addition, the network-like ash characteristic of the salt-bromate neutralizer was not at all evident only a small clumplike residue was visible. 2. The waving performance on the mandrel, despite the lessened uptake, was somewhat improved. The general range of values was not much changed without added salt, the AW values re- mained in the fair range with salt, close to excellent. The im- provement was manifest in a greater degree of regularity. 3. Serigraphically, there is no loss of strength. 4. Although the in-between acid rinse improved the waving sequence when no salt was added to the neutralizer, this three-stage waving sequence cannot be said to approach the over-all performance of reductant followed by bromate-containing magnesium sulfate. Citric acid could be expected to include, among the possible actions, that of neutralization of the excess negative charges in the keratin. The fact that the acid displaced a significant portion of the cationic uptake permits the conclusions to be drawn that: 1. Part of the mechanism of action of the added magnesium sulfate is one of charge neutralization and 2. Uptake does strongly parallel improvement in waving per- formance, but it is not a sole operating mechanism. Dehydration Mechanisms and Itydrogen Bond Re-formation The question arises: What other mechanism(s) could be involved? The consistent improvement in waving properties throughout a wide range of increasing salt concentration suggested a possible dehydration mechanism. Two possible dehydration mechanisms were considered: one osmotic and the other a salt-protein interaction in which the water that is hydrating the protein is removed by a salt in sufficient concentration. In order to determine whether an osmotic mechanism is operating, the vapor pressures were determined of a series of neutralizers in which a number of salts in varying molalities were combincd with sodium bromate whose final concentration was 0.67m. Tresses were then waved utilizing these neutralizers. The effects of the added salt on the AW values are shown in Table V. In utilizing this approach, the assumption was made that if the effect of the salt were due only to the solution properties (as osmotic pressure)

PERMANENT WAVE NEUTRALIZER 85,0 and not to specific salt-protein interaction, then the determination of vapor pressure lowering should correlate with the effect on curl forma- tion. In Table V the results of the measurements of vapor pressure lowering indicated that this property could not be used to predict the effect of a salt on air stability there was no indication that a given vapor pressure range or a required vapor pressure lowering could be cited as required for optimum waving results. The experimental results to this point may be interpreted as follows: The magnesium sulfate in relatively low concentration neutralizes the ex- cess negative charges and brings the keratin closer to its isoelectric point here, the protein is most easily susceptible to dehydration which is then effected by the salt's relatively higher concentration (4). During de- hydration, the hydrogen bonds whose strength and tendency to form are minimal in water itself, have, in the relatively decreased water content (provided by the high concentration of the added salt) an increased like- lihood to re-form (5). Consecutive or Simultaneous Application of the Added Salt with the Oxidant The favorable results of adding a salt in sufficient concentration to the oxidant raises the question of how the effectiveness of the salt would be influenced by its being applied separately to the hair just prior to or fol- lowing the oxidant. To answer this, tresses were waved by first reducing them for six minutes with ammonium thioglycolate these were then divided into two groups which were treated by either of the two following procedures: 1. 30% aqueous solution of magnesium sulfate followed by a bromate neutralizer, each for 8 minutes or 2. A bromate neutralizer followed by a 30% aqueous solution of magnesium sulfate, each for 8 minutes. Rinsing was done after application of each of the above solutions. The experimental results are summarized in Table VI. From the tabulated results, it is evident that the tresses in the two groups somewhat resemble one another and that both are quite similar to tresses waved with the salt-free bromate neutralizer. This is especially noticeable in their rather poor AW values. These results clearly indicate that for the added salt to be of benefit in conferring air stability on the tress, it must be applied simultaneously with the oxidant. Apparently, the re-formation of the secondary bonds is facilitated only if done at the same time as those of the disulfide bridges.