Carbon in Platinum and Palladium SOLUBILITY DETERMINATIONS AND DIFFUSION AT HIGH TEMPERATURES
BY G. L. Selman, B.SC., P. J. Ellison, and A. s. Darling, Ph.D., M.1.Mech.E. Research Laboratories, Johnson Matthey & Co Limited
These workers did not examine the micro- Carbon difluses rapidly through heated structure of their test specimens and assumed membranes of platinum and palladium that all the carbon found was in the dis- even though its solubility in the former solved form. The use of silica heat-treatment metal is less than 0.02 per cent by capsules also introduced a considerable factor weight at 1700°C. Palladium takes up of uncertainty into their determinations. more than 0.4 per cent by weight of The carbon-platinum and carbon-pal- carbon at 1400°C and this interstitial ladium alloys used in the present investigation solubility hardens it considerably and were carefully studied by metallographic and distends the lattice. The efects of X-ray diffraction techniques, and all treat- carbon on the microstructure and ments were carried out in platinum capsules! mechanical properties of platinum and to avoid any possible siliceous contamination. palladium are described in this article, The platinum sponge and sheet used during which dispels some illusions and surveys this investigation for the manufacture of the practical consequences of the con- alloys by melting or solid state diffusion stitutional relationships observed. contained the following impurities : Pd 0.003 per cent, Au 0.0001 per cent, Ca 0.0001 per cent, Fe 0.0005 per cent, Si
Platinum Metals Rev., 1970, 14, (l), 14-20 14 Fig. 1 Duplex carbon-platinum alloy Fig. 2 Duplex carbon-palladium alloy produced by melting pure platinum in a produced by melting palladium on graphite. graphite crucible. As cast. x 75 As cast. ~75 furnace in graphite crucibles. The charge The duplex platinum-carbon alloys pre- was held molten for five minutes, following pared by this method were not readily which the crucible, still containing the plati- amenable to cold work and cracked quite num metal, was quenched into water as severely under the forging hammer. The rapidly as possible. Small slices of the palladium alloys although considerably har- quenched ingots were then reduced in thick- der, were, by comparison, relatively ductile. ness approximately 50 per cent by cold Some typical microstructures are illustrated forging and solution treated at various in Figs I and 2. The differing geometry of temperatures for various times before being the primary graphite in the two metals quenched into water. These heat treatments provides a clear explanation for their differing were carried out under a nitrogen atmosphere response to cold work. to avoid oxidation. Lattice parameters and The lattice parameter of the platinum micro-hardness determinations were made on alloys changed little with quenching tem- carefully polished and etched microsections. perature whereas considerable expansion was
Platinum Metals Rev., 1970, 14, (1) 15 Fig. 4 Grain boundary porosity in the Fig. 5 Porosity throughout the sheath of a sheath of a platinum capsule containing palladium capsuZe containing graphite, graphite, heated in air for 75 hours at heuted in air for 18 hours at 1200°C. X 20 1200°C. X75. observed when the duplex alloys of palladium and sealed up in pure platinum capsules were quenched from high temperature. under a pressure less than IO-~Torr. These These results, plotted in Fig. 3, indicate that capsules consisted of solid drawn platinum palladium dissolves measurable quantities tubing having an internal diameter of 0.22 of carbon, and platinum very little. inch and a wall thickness of 0.020 inch. The slight changes in lattice parameter For the purposes of the equilibrium observed when carbon was equilibriated studies, these capsules were normally heat- with platinum in this way could well be treated under an atmosphere of nitrogen. caused by the introduction of small quantities Some of the initial specimens were annealed of dissolved impurity - usually of the order in air and in these cases some unusual effects of 30 p.p.m. - during melting and casting. were observed. After IOO hours at 1400°C considerable surface distortion occurred and Diffusion Experiments close examination revealed areas having a To provide specimens containing smaller surface structure quite unlike that normally quantities of carbon than those which were displayed by air annealed platinum. Similar readily obtained by melting and casting, effects were produced by burning carbon on sheets of platinum and palladium 0.020 inch the surfaces of a platinum sheet in air at thick were packed in high purity graphite 14oocCand it was concluded that the surface
Fig. 6 Layer of graphite on the outer Fig. 7 Graphite spheroids precipitated surface of a grphite-pocked palladium cap- towards the inner surface of the palladium sule, heated in vacuo for 100 hours at 1200°C sheath illustrated in Fig. 6. x 50 and slowly cooled. x 200
Platinum Metals Rev., 1970, 14, (1) 16 behaviour of the platinum capsules was caused by the combustion of carbon which must have diffused rapidly through the platinum mem- brane. Micro-sections taken through these plati- num capsules showed considerable grain boundary attack as shown in Fig. 4. Here the grains have opened up more than half way through the section of the tube. No attack of this sort was observed when empty platinum capsules were annealed in air, and the experimental observations are consistent with Fig. 8 Porosity in carbon saturated palladium the view that carbon, although sparingly sheet quenchedfrorn 1400°C. x 75 soluble in platinum, diffuses through it DUCUO. Fig. 7 shows that during the furnace very rapidly at high temperatures when cooling period following the annealing treat- surface combustion produces the concentra- ment graphite spherulites had precipitated tion gradients required. This localised out only within the inner half of the sheath intercrystalline attack was characteristic of the section. It seems likely therefore, that the internal oxidation behaviour of platinum. outer layer of graphite which formed initially Palladium capsules, when subjected to the as a consequence of surface volatilisation same treatment showed uniformly distributed acted as a nucleation site for graphite attempt- fine porosity throughout their cross-sections, ing to separate from solid solution within as indicated in Fig. 5. the outer layers of the capsule during the slow Both effects indicate a high rate of carbon cool. These carbon surface deposits were diffusion. Oxygen is of course fairly soluble not obscrvcd on palladium capsules annealed in the palladium lattice and this could account for the uniform porosity observed with the palladium capsules. The lower solubility of oxygen in pure platinum probably confines the effect of the carbonjoxygen reaction to the grain boun- daries. In order to avoid the complicating effects of palladium oxidation some experiments were made on palladium tube capsules which, after packing with carbon, evacuating and sealing, were finally annealed in vucm at 1400°C. These sealed capsules rapidly developed a black surface coating which was identified as pure carbon. The sections through such a palladium capsule shown in Figs 6 and 7 illustrate the thickness of this carbon sheath and also the spherulites of graphite precipitated out within the tube wall. The volume of carbon present on the surface of the capsules was too great to be accounted for entirely in terms of the vola- tilisation of carbon saturated palladium in
Platinum Metals Rev., 1970, 14, (1) 17 in argon and rapidly quenched from the geneity of the analytical specimens a series annealing temperature. of hardness impressions were made across At this stage attempts were made to deter- sections through the sheets which had been mine the rate of diffusion of carbon through saturated with carbon by diffusion. platinum and palladium by heating sealed Palladium and platinum specimens satu- carbon containing capsules in air and measur- rated with carbon and quenched from 140oCC ing the change in weight. By carrying out were porous as shown in Fig. 8. No porosity blank runs on empty capsules the losses was detected in platinum or palladium caused by carbon combustion could, it was quenched from IZOOOCor below. No free or hoped, be scparated from those caused by undissolved carbon or graphite was detected oxidation and volatilisation of the platinum in any of the quenchcd specimens. The curve metal. These tests were, however, unsuccessful of lattice parameter versus carbon content in as capsule failure occurred long before the palladium is given in Fig. 9. Superimposed establishment of steady rates of weight loss. on this curve are the lattice parameters of the quenched duplex alloys, the solubility curve Solubility Limits of carbon in palladium, shown in Fig. 10 Curves relating the lattice parameters being derived from this data. and carbon contents of single phase alloys Alloys quenched from temperatures above were required for accurate solubility deter- this line were simple solid solutions as shown minations. The alloys needed were prepared by Fig. 11. Subsequent heat treatment at by heating thin sheets of platinum and temperatures below the solubility line pre- palladium in contact with carbon in sealed cipitated out carbon which at 800°C or below capsules as described above. The capsules appeared as somewhat angular spheroids were quenched in water after heat treatments such as those shown in Fig. 12. These metal- ranging from 160 hours at 800°C to 7 hours lographic studies showed the true solubility at 14ooOC. Half the sheet samples were of carbon in palladium to be much lower analysed for carbon, the remainder being than the reported values of Siller, et al., whose sectioned, polished and etched for micro- results are plotted on Fig. 10for purposes of scopic and back reflection X-ray diffraction comparison. studies. The absence of significant lattice parameter In the case of palladium, microhardness variations made it necessary to determine the measurements provided a sensitive index of solubility of carbon in platinum by micro- its carbon content, and to confirm the homo- scopic methods. A number of synthetic
Platinum Metals Rev., 1970, 14, (1) 18 Fig. 11 Palladium containing 0.25 % of Fig. 12 Graphite spheroids precipitated carbon in solid solution quenched from .from 0.25% carbon-palladium alloy by 1200°C:. A 75 ageing at 800°C for 1 hour. )i 7.5 alloys was produced by melting in the argon although it was difficult to ascertain, even at arc furnace on a water cooled copper hearth. the highest optical magnification, whether Thesc alloys, containing from 1.5 to 0.004 these were particles of graphite, very fine per cent by weight of carbon, were carefully porosity, or areas where graphite spheroids analysed after preparation. Sections from had been removed by polishing. These the ingots were heat treated in vucuo in a solubility values are well below those reported tantalum furnace and quenched after 7 hours by Siller, ef al., which range up to 0.13 per at 1400OC or 5 hours at 1700'C. The carbon cent by weight at 1245°C. solubility appeared to be less than 0.02 per cent by weight at 1700~6. Small dark Hardness inclusions were observed in the 0.01 per cent Carbon increased considerably the hardness carbon-platinum alloy quenched from 14oo0C, of palladium although it had little effect on platinum. The quenched hardness values for the palladium alloys are plotted on Fig. 13. The supersaturated solid solutions of carbon in palladium might well respond to age hardening treatments and considerable scope for further work exists in this direction.
Discussion and Possible Applications These experimental results show that dissolved carbon has a considerable influence on the behaviour of both platinum and pal- ladium. Although the solubility of carbon in platinum is very low, the rate of diffusion at high temperatures is exceedingly high. In some applications this must be regarded as a disappointment because the use of high modulus graphite fibres for platinum rein- forcement has hitherto been considered an attractive possibility. The impracticability of this idea is unfortunately well illustrated
Platinum Metals Rev., 1970, 14, (1) 19 to the development of self-lubricating elec- trical slidewire resistances. It is not yet known whether the presence of dissolved carbon will accelerate or inhibit the diffusion of hydrogen through palladium. Since the solubility appears to be interstitial, further exploration in the low temperature regions of the solu- bility diagram might possibly reveal the existence of a miscibility gap similar to that observed in the hydrogen-palladium system. The greatly extended lattice parameter of Fig. 14 As cast carbon-pZatinu?n ingot after these alloys might also influence the behaviour heating in air for 90 hours at 1400'C. x 75 of palladium catalysts. Accurate determinations of the rate of by Fig. 14which shows the cross-section of a diffusion of carbon through platinum and duplex carbon-platinum alloy which has been palladium would indicate whether either of annealed in air at 1400OC for 90 hours. these metals could be seriously considered as a Before annealing, this specimen exhibited a semi-permeable membrane for carbon. microstructure comparable to that illustrated Acknowledgements in Fig. I. No graphite was observed in the The authors are greatly indebted to Dr J. C. section following the anneal in air, and the Chasten who provided much of the encourage- voids marking the outlines of the original ment and stimulus needed for this work. primary graphite flakes were seen to break References up and spheroidise as the heat treatment I L. T. Collier, T. H. Harrison, and G. W. A. became morc prolonged. Taylor, Trans. Faraday Soc., 1934,30, 581 2 M. R. Nadler and C. P. Kempter, J. Phys. The prospects for carbon-palladium alloys 1960, 64, 1468 are rather more attractive. The strengthening 3 E. Raub and G. Falkenburg, 2. Metallkunde, effects are considerable and it is possible that W64, 557 4 S. K. Rhee, Diss. Abs. B.,1966, 27, (6), 1963 the precipitation Of graphite from super- 5 R. H. Siller, W. A. Oates, and R, B. McLellan, saturated palladium-based alloys could lead J. Less-Common Metals, 1968, 16, 71
Spot Welding Platinum Foil on Titanium Anodes Titanium anodes coated with platinum are successfully over several years. The method used in many cathodic protection applica- is reliable and no selective corrosion of the tions. Electrodeposited platinum is usually foil at the spot welds has occurred. only a few micro-metres thick but for Baggerud has developed specifications for certain processes there is a platinum loss 30 and sopm foils and tolerance limits for the which calls for a layer of greater thickness. welding current have been determined. Hence, for example, platinum foil is used on Melting of the platinum foils must be the anodes that protect the drying cylinders minimised during welding lest magneto- of paper-making machines against the cor- hydrodynamic effects cause the formation of rosive effect of the cooling sea water (I). brittle intermetallic compounds. Oxides on For anodes used in this way, foil is cheaper the titanium surface have little effect on per unit weight than electrodeposited plat- weldability and the Peltier effect is no real inum and the platinum loss is much less. hindrance. A. Baggerud of the Technical University of References Norway reports (2) that spot welding of I A. Almar-Neiss and J. Drugli, Platinum Metals platinum foil to titanium has produced Rev., 1966, ID, (z), 48-51 anodes which have protected drying cylinders 2 A. Baggerud, Metal Construction, 1969, I, 412
Platinum Metals Rev., 1970, 14, (1) 20