six polar hydroxy sites. Sorbitol, in contrast, appearsAcknowledgmerti. The financial assistance of the to be a struchre breaker, as is indicated by the mono-National Science Foundation is gratefully aeknowlronic negatiw erilhalpies over the entire concentration edged. J. K. S. wishes to thank Professor Y, Marcus range investigated, This behavior of ART, is similar at> the Hebrew University for his hospitality. for XaCE in acjil~aoixs hydrogen peroxideJ5 ureaJ6 and %,hi& differ d r o~i d e s .~ in position of one of their six hy-bility of Inorganic and Organic Compounds,'' Supplement to 3rd Publication costs assisted by the Muage Royal de E'Afrigue CentraleThe proton motions in the methanol-silica gel (Xerogel) systems have been deduced from the measurement of the spin-lattice (TI) and spin-spin (T,) proton relaxation times in the systems C€LOD-XOD, CDIOH-XOK, and CH,OH-XOH, where XOH and XOD stand for the Xerogel-OH or Xerogel-OD surfaces, respectively. In CH30D-XOD, only one TI or T2 is observed. The experimental data may be accounted for by a laandational motion of CH,OD, the methyl. group rotating very rapidly around the triad axis. In CD30H-XOH, one TI aiid two Tt are measured. TI is the superimposition of two functions. The rate of exchange of nuclei between the two populations is fast with respect to T1-l but slow with respect to T2-I. The long TZ represents the diffusion motion observed in the first system, while the short T2 is assumed to be due to a proton exchange process between the hydroxylic protons of C&OH and XOH or between CDIOI~ molecules. Moreover, it has been possible to compute the observed Ti for the CHIOH-XOH system from the data obtained for the two simpler cases. At 25", the surface diffusion coefficient of adsorbed methanol is reduced by two orders of magnitude with respect to that in the liquid for a degree of coverage e ^I 0.6. For 1 < e < 2, the lowering is by a factor of 10. For e 2 0.7, the surface diffusion coefficient at 25' is the same as that in the liquid state at -85". More than three layers should be probably required to find a diffusion coefficient in the adsorbed state close to that in the liquid state. In the assumed exchange process a proton is transferred from the surface into a cluster of adsorbed species where it jumps from a molecule to another before coming back on the surface: [-OH + CB301-I 3 1 0-+ CH10H2+; CH30H2+ + CH80J1 -+ CH30H + C€InOIh+; CH3OH2" + -0-[ -+ ] -OH + CHaOH. The ratio of the exchange to the molecular diffusion jump frequency ( v e / u d ) is 6 x 10-3 for e = 1.7 and 2.5 x 10-2 for e = 0.8. This means that the number of times a methanol molecule is protonated during its stay on an adsorption site increases at decreasing degree of coverage. For 0 = 0.8, u./ud is very close to the value observed in the liquid protonated methanol al, 25" The activation energies of both the diffusion and the proton exchange processes are dose to each other (-6 kcal at 8 = l), and there are approximately equal to half the isosteric heat of adsorption. Both these motio...