No abstract

Although melts in the lead fluoride-lead oxide system are commonly used as solvents for growing crystals of refractory oxides by the fluxed-melt method [e.g., White (1)], very few physical data have been published for these liquids. As part of an investigation into fluxed-melt growth processes, a fairly rapid survey of the viscosities of melts in the lead-fluoride-lead oxide system was made using a rotating crucible viscometer. This work was not designed to yield viscosity values of high absolute accuracy, but to determine quickly the relative magnitudes of viscosities of different compositions. ApparatusThe method is essentially that of Margules (2) and Lillie (3), in which a crucible containing the liquid is rotated at constant speed, and the viscous torque on a bob suspended in the liquid is measured.The crucible was of platinum to reduce chemical attack by the liquids and was 50 mm deep and 50. mm in diameter. It was supported inside a Kanthal electric furnace (maximum operating temperature 1250~ by an alumina tube which could be rotated about a vertical axis at several speeds between 2 and 20 rpm, selected by a gear-box.The liquids used evaporated fairly rapidly, changing the liquid depth appreciably in a few hours. The platinum bob was therefore in the form of a disk 2 mm thick and 40 mm in diameter, which was totally immersed and suspended by a platinum rod 2 mm in diameter. The viscous drag on the rod was relatively small so that small variations in liquid depth had a negligible effect on the total viscous torque.The platinum rod projected above the furnace and was suspended by a steel wire 7 ft in length. The crucible was rotated in each direction alternately, and the difference between the equilibrium positions of the bob was measured with an optical lever 2 meters long. At equilibrium the torque exerted on the bob by the rotating liquid was balanced by the elastic torque of the suspension, and the deflection of the optical lever from the rest position provided a measure of the viscosity.Vertical adjustment of the disk position to within about 0.05 mm was effected with a micrometer screw. This was used in setting up, and also to measure the liquid depth, the point at which the disk touched the crucible bottom or the liquid surface being determined visually with a mirror.A Pt-Pt/13%Rh thermocouple was dipped into the liquid to measure its temperature to within +_3~ between each viscosity measurement. Reproducible curves of viscosity against temperature were obtained by making measurements while the liquid temperature slowly changed through about 100~ thus avoiding the need for elaborate temperature control. The rotation of the crucible provided adequate stirring. About five minutes was required for each viscosity measurement, during which time the temperature change was negligible. Observations were made during periods of up to 30 hr in some cases. The depth of the liquid was measured and samples taken for chemical analysis at frequent intervals. ~' ! SL ~.o G o I 9 "~" 0.759 0.8 0-85 o.9 I000 --~ TEMPER...

No abstract