Grain boundary migration in niobium

Grain boundary migration in niobium

JOURNALOF THE LESS-COMMON METALS Short 399 Communications Grain boundary migration in niobium Heretofore, virtually all evidences of grain boun...

514KB Sizes 0 Downloads 0 Views





Grain boundary


in niobium

Heretofore, virtually all evidences of grain boundary migration have been surface markings. This note reports on some observations of a distinctive internal structure believed to result from grain boundary migration in niobium. A 65 g button of niobium produced by non-consumable electrode arc melting was cold-rolled 50% and then annealed for 3 h at 17oo’C in a vacuum of less than to r/r0,u. This temperature, well above the recrystallization temperature, about IZOO’C, was chosen in order to produce large grains. The surface of the slab was ground, polished metallographically, and etched with the IO ml HaO, IO ml H&O*, IO ml HF, I ml HzOa reagent reported by MICHAELAND HUEGEL~to produce etch pits at the emergence points of dislocations. Figs. I and 2 show typical structures produced. The dark, rod-like particles in the grains where the etch pits are most clearly revealed are carbides of niobium which have been over-etched in the generation of the etch pits. The features of interest, however, are the regular lines of etch pits which appear related by their contours to certain nearby grain boundaries. Many of the arrays of etch pits were single lines of pits (Fig. I), while others were

Fig. I. Discreteetch pit lines J. Less-Common






more diffuse lines (Fig. z). In virtually all families of etch pit lines, the shape of the lines progressively became more similar to the shape of the associated grain boundary, the nearer they were to this grain boundary. This similarity to the many observations of thermal grooving occurring on the surface of a metal in which the grain boundaries are migrating has led to the conclusion that these lines of etch pits delineate periodic positions of migrating grain boundaries. There are two possible internal structures which would cause the positions of the migrating grain boundaries to be marked by etch pits. These etch pits may form either on dislocation emergence points or on precipitate sites, If the former is true, this means that as a high-angle grain boundary changes orientation during migration, its structure alters in such a way that it must periodically “precipitate” dislocations in a regular sub-boundary array. The second possibility is that the solubility of carbon in the neighbourhood of grain boundaries at 17oo’C is greater than in the interior of the grains, so that as a migrating grain boundary sweeps through a grain it dissolves the carbides in the grain being consumed and periodically precipitates them in the growing grain. Conditions for this are favourable since the analyzed carbon content of the sample, 0.030 wt.%, is very close to the solid solubility of carboh at 16oo-17oo~C~. Although an electron microscope study of the etch pits at high magnification did not reveal these precipitates, there is insufficient evidence to dismiss this mechanism. The author is inclined to favour the first mechanism, however, because this observation has been made in a number of other samples of both alloyed and unalloyed niobium; one would expect that the second mechanism would require rather special

Fig. 2. Diffuse etch pit lines. J. Less-Common Metals,


(1960) 399-401



The nearly regular periodicity of the rows composition-temperature conditions. suggests that in the case of either mechanism, the rate of motion of the boundaries is discontinuous. Occurrence of this phenomenon opens the possibility of studying in three dimensions the change in shape of an entire grain during grain boundary migration. Its shape could be revealed by successively removing layers of metal and etching to reveal the various positions of the grain boundaries. Union Carbide Metals Compalzy, Niagara


N. Y. (U.S.A.)


1 A. I3. MICHAEL AND F. J. HUEGEL, Acta Met., 5 (1957) 339. 2 M.L. POCHON,C.R.MCKINSEY, R. A. PERKINS AND W. D. FORGENG, RegionalConference on Reactive Metals, hn. Inst. Mech. Engrs., 1958. Received May 23rd, 1960 J. Less-Common

The ammonolysis of zirconium(W)

Metals. 2 (1960) 399-401


Tensimetric studies made by FOWLES AND POLLARD’ showed that one mole of ammonium chloride was formed in the reaction of zirconium(IV) chloride with ammonia at -63” and -36°C. However, although these experiments suggested the formation of the ammonobasic zirconium(IV) chloride ZrC13(NHz), xNH3, this compound was not isolated, and in comparable experiments with vanadium(IV) chloride213, where VC12(NH2)2, xNH3 was formed under tensimetric conditions, ammonolysis was found to be more extensive when ammonium chloride was removed from the reaction mixture by washing it with liquid ammonia; VCl(NH&, 2NH3 was isolated in this way. Removal of ammonium chloride from the products of the ammonolysis of zirconium(IV) chloride does not, however, effect a greater replacement of chlorine atoms, and the ammonobasic zirconium(IV) chloride predicted in the tensimetric experiments can be isolated in almost quantitative yield. The small amount of zirconium (7%) found in the soluble portion probably results from the formation of complex zirconium anions similar to those found in the titanium(IV) halide-ammonia systemsd. Such complex formation would be small and would occur only in concentrated solutions of ammonium chloride. More extensive ammonolysis may sometimes be effected by heating the ammonobasic metal chlorides in vacua and then re-treating with liquid ammonias, but this is not very effective with zirconium(IV) chloride because a second zirconium-chlorine bond is only partially ammonolysed even when the product is heated to 100’. After such a treatment, however, more zirconium is found in the soluble portion, and it may be that the heating helps to break down the polymeric solid so that it dissolves in the presence of ammonium chloride. The ammonia which is attached to the ammonobasic zirconium(IV) chloride is J. Less-Common Metals. 2 (1960) 401-403