May 5, 1960
COMMUNICATIONS TO THE EDITOR
a t 253 and 294 mp were consistent with the presence of the expected 2-acetylthiazolium salt. The band near 300 mk, which is characteristic of the 2acetyl substituted thiazole chromophore, was absent when the spectra were run in water or methanol. Furthermore, solution of the mixture for ten minutes in methanol a t room temperature, then ether precipitation, led to a 50% decrease in the intensity of the carbonyl band a t 5.85 p , accompanied by the appearance of a new band a t 10.7 p ; repeated methanol treatment led to further decrease in the carbonyl band and increase in the 10.7 k absorption. Since this latter is the C-H bending vibration charactelistic of thiazolium salts unsubstituted a t C-22, it is apparent that the process being observed is a deacylation a t C-2. The leaving group in the deacylation reaction is the same thiazolium zwitterion which has been detected previously,2 but the lability of the 2acetylthiazoliurn salts is much greater than would be expected solely from the stability of this thiazoliurn zwitterion, and it seems likely that dipoledipole repulsion in the ketone is also an important factor. At any rate, these results furnish additional support for the mechanism of thiamine action already proposed.
TABLE I m2e/l.
0.000225 .00076 .00102 .00500 .OlOl .0252 .0286 .0495 ,0575 .097
Sir:
D(obs.) cm.l/sec.
1.335 1.332 1.327 1.319 1.312 1.297 1.298 1.294 1,289 1.280
D(ca1cd.) crn.z/sec.
1.330 1.326
1.326 1.315 1.30; 1,291 1,288 1.274
1.269 1,250
of f O.lS%, together with values calculated from the limiting law. Extrapolation of the last four points gives a D o of 1.337 compared with the Nernst liniiting value of 1.334. The results also are illustrated in Fig. 1. It will be observed that for concentrations
