Contradictions and new aspects of the bubble model of doped tungsten wires

The interpretation of a series of small angle neutron scattering studies on lamp grade tungsten rods and wires is in contradiction with an extended body of the results that were obtained by fractographic methods and thin foils studies by means of scanning microscopy and transmission electron microscopy. In order to facilitate the re-interpretation of the scattering studies, the state of agglomeration of potassium in the dope phase has to be clarified, due to the peculiar scattering properties of the two-phase inclusions. These efforts yielded to the following results: (1) There are independent pieces of experimental evidence that the dope inclusion are filled with liquid potassium in heavily drawn lamp grade tungsten wires and rods in various stages of the processing. (2) A slight volume compression of liquid inclusion upon wire drawing is compatible with the scaling laws of the Taylor model in a good approximation, because the expected deviations are less than the accuracy of the required measurements. (3) The size increase of the inclusion cannot be explained as the expansion of individual bubbles that do not interact with each other. One has to take into account also various sorts Ostwald type ripening processes.