each other. Under the best conditions the mass Acknowledgments of one kilogram can be determined in terms of The author is pleased to acknowledge the assistance given by Johnson Matthey Metals Limited and the a reference standard to rather better than I part National Physical Laboratory in the development of in 109, that is to somewhat less than I pg. The the procedures described in this article. Many helpful steam-cleaning procedure is repeated until discussions were held with J. Leroux of the BIPM weighings show that no further decreases in workshop who carries out the machining and G. Girard, who does the weighing. It is a pleasure also mass occur. Figure 6 shows one of the new to note that the work described in this article was car- kilogram prototypes alongside one made using ried out just one hundred years after George Matthey the traditional methods. collaborated with the then recently founded Bureau International des Poids et Mesures in the manufac- The first kilogram prototype made using ture of the alloy from which the international pro- these new procedures of polishing and adjust- totype of the kilogram and its copies were made. ment was given the number 64 in the BIPM References series. Since then, numbers 65, 66 and 67 have T. J. Quinn, “BIPM-a profde”, 3. Phys. E., successfully been completed. Also from the 1985,18, 368 same ingot two platinum-iridium standards J.-R. Benoit, “Rapport sur la construction, les comparaisons et autres opfrations ayant servi were made for NPL and were designated A and dCterminer les equations des nouveaux prototypes B. During the course of the development of the mttriques”, BIPM Travaux et Memoires, 1890,7, technique two standards were made whose I01 L. B. Hunt, “George Matthey and the Building of mass was just outside the tolerance of *I mg. the Platinum Industry”, Platinum Metals Rev., These were designated 650 and 651. A second 1979,23, (2), 68; D. McDonald and L. B. Hunt, “A History of Platinum and its AUied Metals”, ingot has been delivered by Johnson Matthey Johnson Matthey, London, 1982,pp. 271-299 and the machining of a further group of A. Bonhoure, “La construction des Prototypes de kilogram prototypes is now well advanced. Masse”, Mumtechnic, 1952, 6, (4), 204

The Platinum-Zirconium-Yttrium System For the metallurgist the availability of phase The isothermal section is characterised by equilibrium data is an essential requisite if he is regions of ternary solid solubility, based on the to relate the microstructure of alloys with their initial binary compounds, which spread into behaviour. While binary diagrams of the the ternary system along the corresponding platinum group metals are now widely pseudobinary section. The solubility of yttrium available, it is unfortunate that there is still in platinum-zirconium intermetallic com- relatively little published on ternary and higher pounds is less than 5 atomic per cent whereas order systems. It is, therefore, encouraging to zirconium has much higher solubility in find that new work on ternary systems of platinum-yttrium compounds, extending up to platinum has been undertaken. The phase 18 atomic per cent in the Pt,Y intermetallic equilibria of the platinum-zirconium-yttrium which has a AuCu, type structure. The Laves system at 10ooOC has been reported recently by phase Pt Y only extends into the ternary Yu. J. Konobas, M. V. Raevskaya and I. G. system to about z atomic per cent zirconium, Sokolova of Moscow State University u. Less- not interacting with other binary compounds. Common Met., 1986, 115, (2), L5-L6). Below the solidus line, the Pt ,Zr-Pt ,Y, PtZr- This ternary system is of some interest since PtY and the F’t,Zr,-Pt,Y, sections are both the platinum-zirconium and the platinum- pseudobinary, as might be anticipated. At the yttrium binary systems exhibit a number of in- platinum-rich comer, the platinum solid solu- termetallic compounds, whereas the zirconium- tion is in equilibrium with two three-phase yttrium system is a simple eutectic. regions, each containing two intermetallic com- The 10ooOC isothermal section published by pounds. This could have significance for new the authors was based on experiments on arc- platinum-alloy development, possibly as an melted alloys homogenised at 10ooOC for 1200 alternative approach to the high strength oxide hours prior to quenching in iced water. They dispersion-strengthened platinum materials found no new ternary phases, only those which are based, coincidentally, on the oxides reported previously in the binary systems. of zirconium and yttrium. C.W.C.

Platinum Metals Rev., 1986, 30, (2) 79