The effect of wire history on the coarsened substructure and secondary recrystallization of doped tungsten
- PDF / 2,475,482 Bytes
- 9 Pages / 597.28 x 785 pts Page_size
- 116 Downloads / 201 Views
I.
INTRODUCTION
THE tungsten wire that is used in lamp filaments has a small diameter that is produced by repeated wire drawing. The microstructure o f this as-drawn wire is fibrous and consists o f long, narrow grains. In this condition, the potassium, w h i c h is the key doping material in lampquality tungsten, is contained in ellipsoids that have very high aspect ratios. II-41 The precursors o f these ellipsoids are potassium-filled pores in the sintered ingot. After this wire is coiled to make a filament, it must be heated to the operating temperature o f the lamp. During this initial heatup, the potassium ellipsoids breakup into bubble rows, and secondary recrystallization occurs. During this latter process, a few grains grow and consume all others present in the material, and the grains that form often occupy the entire wire diameter. Because o f the interaction between the growing grains and the bubble r o w s , the grains that are formed have a wavelike shape, known as an interlocking structure, in w h i c h most o f the grain boundary is parallel to the wire axis and only small parts o f the boundary are perpendicular to it. This type o f grain structure allows very little grain boundary sliding at elevated temperatures ]5-1°~ and thus produces a long-life filament. Micrographs o f this grain morphology and the bubble rows are shown in References 1 through 3. Because the shape o f the secondary grains is so important in determining filament life, there have been a number o f studies examining the process by which the grains are formed, t2,1~-171 Most studies have concluded that the potassium bubble size and bubble row length and C.L. BRIANT, Visiting Research Scientist, O. HORACSEK, and K. HORACSEK are with the Research Institute for Technical Physics of the Hungarian Academy of Sciences, H-1325 Budapest, Hungary. C.L. BRIANT is permanently affiliatedwith General Electric Company, Research and Development Center, Schenectady, NY 12019. Manuscript submitted June 3 , 1992. METALLURGICAL TRANSACTIONS A
spacing are the most important factors in determining the shape o f the grains. However, there have been a few studies that have indicated that the entire process history o f the wire (powder processing, doping, thermomechanical processing, and annealing) can affect the secondary recrystallization temperature and the shape o f the secondary grains, tJT,ls,19] In this article, we report a study o f secondary recrystallization in two different tungsten wires. These wires had different process histories, but their bulk potassium concentrations were approximately the same. O u r results show that these two wires had different secondary recrystallization temperatures, different morphologies o f the secondary grains, and different grain structures a f t e r primary recrystallization. W e will propose that these wires differ because their thermomechanical processing history was different. The differences did not appear to result from differences in the potassium bubble distribution in the w i r e s . W e also foun
Data Loading...
