The only temperature coefficient hitherto measured for the atomic layer of oxygen on tungsten is that which demonstrates the 150,000 cal. /mol. expended in its desorption. If, however, some energy is required before the difference between this heat of adsorption and the heat of dissociation can be liberated, a further temperature coefficient will exist, giving variability to the “condensation coefficient” or fraction of impinging molecules which become enabled to contribute to this layer. Any such variability extends the notion of “activated adsorption” from the slow processes with which it has become associated to the almost instantaneous reaction of “chemisorption”. Since filling of the atomic layer is generally supposed to precede oxidation, and oxidation itself has already been studied through the subsequent volatilization of oxide (Langmuir 1915; Rideal and Wans-brough-Jones 1929), an experimental determination of whether the initial deposition also needs activating would be highly desirable. This would, however, be inaccessible to most technique because of the high speed of recording needed. The oscillographic study of reaction velocity, developed by one of us in collaboration with Dr F. A. Vick, and hitherto applied only to evaporation (Johnson and Vick 1935), offers the opportunity by suitable modifications. It fulfills the following requirements. (1) Reaction time may be kept below 1 sec., avoiding the usual slow contaminations due to gases which strike the walls. (2) The surface may be proved to be clean, within a second of gas being admitted, by observation of its thermionic emission, thus escaping the common ascription of activations to impurity. (3) The gas-covered fraction may be kept small enough to avoid interaction between surface particles. (4) No other substance beside oxygen and tungsten is introduced as reagent or as indicator.
Screening attenuation of coaxial cables determined in GTEM-cells
