sign in sign up
<<
Home , Copper, Platinum

Copper- Alloys

THE EQUIATOMIC Cu-Pt R. S. Irani and R. W. Cahn Laboratory, University of Sussex, Brighton, England

We have been trying to elucidate the reactions. Moreover, the structural mechanisms of ordering of the lattice structure change involved (a rhombohedral lattice in various systems, among which replacing a cubic one) means that substantial the copper-platinum system is especially internal stresses are created which sub- interesting because of its unique features. sequently strengthen the material during rhe The random f.c.c. solution in the course of transformation, as shown in Fig. 2. equiatomic copper-platinum system shown in Such hardening behaviour, coupled with the Fig. Ia is replaced below the critical temper- and oxidation resistance of the ature of 812°C by a long-range ordered alloy, can be usefully employed when the rhombohedral lattice. The latter superlattice configuration consists of alternate layers of pure copper and pure platinum arranged parallel to the (111) planes, there being the same number of nearest neighbours in both the ordered and disordered states, as shown in Fig. Ib. CuPt is the only known superlattice with this structure but in fact surprisingly little work has been performed on the equiatomic alloy. The fact that there is no change in the number of nearest neighbours upon ordering has meant that most existing theories fail to account for the ordering in CuPt. However, Clapp and Moss (I) have attempted a formu- lation in terms of first and second nearest neighbours, the approach being referred to as the central pair-wise (CPW) model. The analysis predicts the long-range order be- haviour from short-range order data, obtained by Walker (2) from diffuse X-ray scattering experiments on powdered samples, but there is no mention of why CuPt orders at all. The theory could be more rigorously tested using diffuse X-ray scattering from single-crystal samples of CuPt. Our recent experimental study (3) has shown that Curt conforms to the earlier established thermodynamic criteria (4) that order-disorder transformations are first-order

Platinum Rev., 1972, 16, (2), 48-49 48 alloy has been annealed for intermediate Fig. 3 Thermally polished sample of equiatomic times below the critical temperature (81zT). copper-platinum ulloy observed under polarised A thermal polishing technique (5) for pro- light after for 8% at 550'C, showing ordered domains growing from surface imperfec- ducing strain-free surfaces without recourse tions, e.p. grain boundaries and pores to electrolytic polishing, which is impractic- able, has meant that it is now possible to References perform polarised light metallography on bulk I P. C. Clapp and S. C. Moss, Phys. Rev., 1966, samples of CuPt (see Fig. 3) and thus interpret 142,418; 1968, 171,754; 1968, I7Ii 764 2 C. Walker,J. Appl. Phys., 1952, 23, 1x8 the mechanical characteristics in terms of the 3 R. S. Irani and R. W. Cahn, Nature, 1970, ordered morphologies. 226, I045 Further work is in progress and a forth- 4 F. N. Rhines and J. B. Newkirk, Trans. A.S.M., 19.53, 45, 1029 coming publication will provide full details of 5 R. S. Irani and R. W. Cahn, Metallography, our studies. 1971, 4, 91 - Alloys The high affinities of both palladium and tectically at 822'C. The first eutectic is platinum for silicon are well known, for dis- formed between pure palladium and PdjSi astrous consequences result when these metals and melts at 825OC. A second eutectic melting are heated under reducing conditions in at 798°C then forms between Pd,Si and Pd,Si. contact with siliceous materials. Although The diagram advanced by Roschel and low phases are formed, the Raub, although considerably more compli- alloying processes which result in this eutectic cated than that previously accepted, is soundly formation have been incompletely under- based on the thermal analysis and metallo- stood and since 1957 (I) it has been known graphic evidence, both of which fix the that the palladium-silicon diagram originally eutectic compositions very precisely. Al- published in Hansen was inaccurate. Rao though the structures of Pd,Si and Pd,Si have and Winterhagen confirmed the existence of not been completely elucidated, their the compound Pd,Si but concluded incor- diffraction lines are tabulated in considerable rectly that it reacted directly with pure pal- detail and the only remaining uncertainty in ladium to form a eutectic melting at 760°C. this system is the extent of the primary solid Dr E. Roschel and Dr C. J. Raub of the solution of silicon in palladium As previously Forschungsinstitut fur Edelmetalle at showr? (3) this interstitially dissolved silicon Schwabisch Gmund (2) have now re-examined causes discontinuous yielding effects. the system, using silicon of semiconductor A. S. D. quality, and have revealed a fairly complex References situation. As silicon is taken up by palladium, I N. K. Rao and H. Winterhagen, Trans. the compounds Pd,Si and PdaSi are suc- Indian Inst. Metals, 1956-57, 10, 139-148 E. Roschel and C. J. Raub, Z. MetaZlkunde, cessively formed. Pd,Si and Pd$i melt 2 1971,62, (11), 840-842 congruently at 835" and 1070OC and the 3 R. G. Hollister and A. S. Darling, Platinum intermediate compound Pd,Si forms peri- Metals Rev., 1967~IT, (3), 94-99

Platinum Metals Rev., 1972, 16, (2) 49