Quarterly progress reports on various projects in the Chemistry Division.

Analysis of the problem of accelerating ions in a Thomas cyclotron has been confined almost entirely to the integration of various equations of motion (for single particles) for which a complete field description is necessary. In contrast, the problem of principal interest to rf system engineers concerns the gross transfer of electrical energy from an rf generator into an accelerated beam.

Measurements were made on the full scale drift tube magnets to check values predicted from the model magnets measurements. These investigations were carried out by the magnetic measurements group.

The intuitive concept that a sector shaped centrifuge cell is free from convection is criticized. Not only is a form of convection present for a single sedimenting species, but a more insidious type occurs in a mixture having an appreciable Johnston-Ogston effect. Rather than striving for convection-free sedimentation, the proposal is to utilize if possible an apparently harmless type of convection occurring in a very thin annulus in order to avoid the convection extending between boundaries in a mixture. The requirement that the concentrations be independent of time meets this condition and yields a hyperbolic cell, which is approximated by a sector cell placed in the rotor backwards. Simultaneously, area measurements and calculations involving the Johnston-Ogston anomaly are simplified because of the time independence.