J. Phys. Chem.of Figure 12 corresponds to gzz, g, , , and gYy, respectively, in the above equations. (Note that in circumstances that the 0, ground state becomes a12b22a12a~b12; 2Bl, then one has to reverse the x and y axis designation.) For the case of the 2A2 ground state one calculates 6 = 1.23 eV. The larger crystal field parameter 6 = 1.23 eV for [ C~~C r ] + o~-/ N a~~-y compared to 6 = 0.424 eV for 02-/NaM-Y is n o t e~o r t h y .~~,~' ,~~ For matrix-isolated alkali-metal superoxide M 0 2 species, 6 is relatively insensitive to the nature of the cation for Na, K, Rb, and Cs, as well as the matrix material for N2, Ar, and Kr.39 The enhanced 6 value for the superoxide anion in [Cp2Cr]+0y/Na56-Y appears to be due to the presence of the Cp2Cr+ cation. The effect could be steric and/or electrostatic in origin, although the mechanism of the enhancement a t present remains obscure. The coexistence of Cp2Cr+ in the proximity of the Na+.-02-ion pair might induce a change in the electronic ground state from the normal 2A2 to 2Bl as alluded to earlier. One should also consider the possibility that the oxygens of the 02-are inequivalent, with the 0-0 bond making an angle to the interior zeolite surface36 rather than being oriented parallel to it (see "0-labeling studiesM). Further work will be required to clarify this interesting point.From an examination of the data of Table IV, one learns that 0,-dynamical effects (EPR line broadening) are only observed for crystal field interactions having 6 I 0.50 eV.36 One deduces from this that the 6 = 1.23 eV value for superoxide in [Cp2Cr]+02-/NaS6-Y is consistent with the lack of any obvious temperature-dependent 0,-EPR line-shape effects and supports the rigid-limit, intrazeolite tight ion-pair description, Na+-02-depicted in Figure 16. This paper describes the use of a microscopic imaging ellipsometer to determine both the macroscopic spreading parameters and the profile of the microscopic precursor of spreading drops of siloxane oil on glass, fused silica, and freshly cleaved mica surfaces. The glass surfaces were subjected to a series of acid etches, which altered their spreading dynamics. Many of the features of the precursor follow de Gennes' predictions, but particularly the profile shows consistently different features. Experimental values for the van der Waals interaction parameter agree with theoretical estimates confirming the importance of this interaction in determining the profile shape, and lower bounds for the positive spreading coefficient S lying between 1 and 2 erg/cm2 for the three surfaces have been determined.de Gennes' has recently revived interest in the microscopic shapes of drops spreading on solids. If the spreading parameter S = usv -(usL + uLv) is negative, a drop will spread to a finite limiting size and contact angle. If however S I 0, the drop spreads
