Platinum Metals Review

PLATINUM METALS REVIEW

A quurterly survey of reseurch on the platinum metuls urrd of dwelopments in their applications in industry

VOL. 8 JANUARY 1964 NO. 1

Contents

Duofunctional Platinum Catalysts in the Petroleum Industry 2

Oxygen Injection Engine for Space Research 8

The Magnetic Properties of Platinum Metals and Alloys 9

Platinum-lined Furnace for the Fluorination of Uranium Compounds I2

Determination of Thermal Conductivity I3

Platinum Metals in Electrochemistry I4

A High Temperature Waveguide Termination 1.5

Organometallic Compounds of the Platinum Metals 16

Precision Glaze Resistors 22

The Early History of the Thermocouple 23

Abstracts 29

New Patents 37

Communications should be addressed to The Editor, Platinum Metals Review Johnson, Mutthey & Co., Limited, Hatton Garden, London, E.C.1 Tuofunctional Platinum Catalysts in the Petroleum Industry

By Vladimir Haensel and Herman S. Bloch Universal Oil Products Company, Des Plaines, Illinois

The transition of petroleum refining from undergoing a number of reactions under the relatively simple thermal operations to more conditions of high temperature and pressure selective and efficient catalytic processes has prevalent in the reactor. The platinum and been one of the major advances of industrial acid functions in co-action with each other chemistry in modern times. This transition and with the support are balanced with the has been made possible by the development operating conditions to achieve a proper of highly selective catalysts designed to meet control of the several possible reactions of the exacting requirements of modern petrol- each of these hydrocarbon types, so that the eum conversion processes. Of such catalysts, course of the reaction is directed toward the those containing platinum as an active production of a maximum yield of desired component have dominated the petroleum product with a minimum amount of un- refining art during the past decade. desirable by-products. The ability to “tailor-make” catalysts for Catalytic reforming involves four primary specific purposes is well illustrated by the reactions. These are dehydrogenation of case of four duofunctional catalysts developed naphthenes containing either five- or six- for the petroleum refining industry by membered rings to aromatics, the de- Universal Oil Products Company and widely hydrocyclisation of paraffins to aromatics, used in processes licensed by it throughout the hydrocracking of higher to lower boiling the Western hemisphere. These catalysts- paraffins, and the isomerisation of paraffins to Platforming, Penex, Butamer and Hydrar*- more highly branched structures. all contain platinum as one of the functional The acid function incorporated in the agents. The other function is provided by Platforming catalyst accelerates the isomerisa- several types and degrees of acidity built into tion, cracking and cyclisation of hydrocarbons. the specially prepared catalyst support in such The platinum function accelerates hydro- a way as to provide the necessary balance genation and dehydrogenation, such as the and co-action between the platinum, the acid conversion of naphthenes into aromatic function and the support for the specific hydrocarbons, and the dehydrogenation of feedstocks and operating conditions charac- paraffins to olefins as one of the initiating teristic of the process involved. steps in paraffin cracking and isomerisation. These two functions are combined in the Platforming Process Platforming catalyst on a rugged, specially Platforming is a method of catalytically prepared alumina support through which the reforming petroleum naphthas to improve active components are uniformly dispersed so their anti-knock properties. Chemically as to provide ready access to the catalyst speaking, the naphtha feed contains three sites for the enormous hydrocarbon trafiic types of hydrocarbons (parafiins, naphthenes required for an economical process. In and aromatics), each of which is capable of addition to its primary function in the main

*Registered Trade marks of Universal Oil Products Company. Platforming reactions, platinum also plays

Platinum Metals Rev., 1964, 8, (l), 2-8 2 one other essential role in this process: it present in the feedstock are little affected for maintains a clean, continuously active catalyst this reason, except for some shortening of surface by hydrogenating intermediate un- the side-chain and rearrangement to a saturated by-products which might otherwise polymethylated structure. condense to high molecular weight carbon- It will be noted from the equations that the rich deposits that produce catalyst deactivation. platinum function and the acid function must The principal reactions involved in act not only simultaneously, but in some cases Platforming a naphtha are shown in the consecutively and in close co-operation, if representative equations given here. The the reactions involved are to be catalysed reactions are carried out at temperatures of cleanly. The objective of catalytic reforming about 475 to sso"C, and under hydrogen is the production of a reforrnate enriched pressures of approximately 250 to 500 p.s.i. in aromatic hydrocarbons and liquid iso- The naphthene-aromatic equilibrium under paraffins and with the concomitant production these conditions lies almost entirely on of a minimum of gaseous hydrocarbon by- the aromatic side. Aromatic hydrocarbons products. It is therefore apparent that some

Platinum Metals Rev., 1964, 8, (1) 3 Intensive research over a long period of years by 1;TniversaE Oil Products Company has provided the petroleum industry with a number of highly selectiwe platinum catalysts of widely different properties. Here catalyst performance is being investigated on a pilot plant scale of the reactions involved (for example concentration in their respective fractions. It hydrocracking to yield gaseous isobutane as is apparent that the Platforming catalyst, shown in equation (4)) must be minimised under Platforming conditions, does not and the other reactions maximised, if isomerise the isoparaflins first formed as a theoretical yields are to be approached. Yet, result of hydrocracking sufficiently to achieve the apparent paradox must be faced that equilibrium; nor is it desirable to isomerise the same catalyst function required for the the light isoparafhs. An increase in effective- desirable reactions likewise induces those ness of the acid function would not only which must be minimised. It is clear, cause such unwanted isomerisation, but the therefore, that in the design and manufacture catalyst would be unbalanced for the other of a catalyst, as well as in the engineering reactions necessary for successful Platforming. of the process in which the catalyst is For this reason, it was necessary to develop employed, there must be a very fine balance a catalyst for paraffin isomerisation that among the number of catalyst sites, their incorporated a different balance of acid ZI. location with respect to each other, and the platinum functions than is present in operating conditions employed. As refining Platforming catalyst. The resulting Penex needs have changed, it has been possible catalyst was developed specifically for the to vary this balance to maintain a high level hydroisomerisation of pentanes and hexanes of performance. to highly branched isomers desirable as components in high octane fuels. Its increased Penex Process acidity, as compared with that of the It is characteristic in the Platforming Platforming catalyst, permits operation at process just discussed that the isoparaffins temperatures of approximately 250 to 375'C- formed (for example isobutane and isopentane) sufficiently lower than the operating range are present in higher than equilibrium in Platforming to take advantage of the

Platinum Metals Rev., 1964, 8, (1) 4 higher equilibrium concentrations of iso- lowest-boiling product and all other isomers compounds which are possible at the lower recycled to obtain nearly quantitative ultimate temperatures. yields of this isomer. Like Platforming, the Penex process operates in a hydrogen atmosphere to ensure Butamer Process long catalyst life through continuous cleans- The need for a catalyst even more effective ing via the platinum component, thereby than the Penex catalyst for butane isomerisa- eliminating the need for frequent catalyst tion-one that would operate at several regeneration and providing process economy. hundred degrees below the effective tem- The Penex process can be employed for perature for Penex operation-led to the isomerisation of pentane alone, or of hexane development of the Butamer catalyst. Such a alone, or for isomerisation of a light naphtha catalyst could take advantage of the marked comprising a mixture of the two. So selective increase in the isobutane equilibrium yield is the catalyst and the process which employs as a function of temperature as shown in the it that volumetric yields of isopentane of over graph on page 6. 99 per cent have been obtained commercially The Butamer catalyst is a platinurn- in the isomerisation of pentane. containing, duofunctional, solid hydroiso- The isomerisation of hexane may be merisation catalyst of such enhanced activity conducted either to obtain a higher octane- and selectivity that it permits isomerisation number hexane product, or selectively to of butane (or pentanes and hexanes) sub- produce individual isomeric products. Thus, stantially to equilibrium at temperatures in z,a-dimethylbutane may be isolated as the the range of IOO to z50°C. This catalyst

The Penex process employs a platinum catalyst specijkally designed for the hydroisomerisation of pentanes and hexanes to highly branched isomers for use in high ortane Juels. This Pmex unit forms part of a refinery in Louisiana

Platinum Metals Rev., 1964, 8, (1) 5 represents a significant extension of platinum capable of uninterrupted use for many catalysis to a lower temperature region of months or even years. hydrocarbon conversions than hitherto was It will be noted that the mechanism of considered feasible. Like the Platforming isomerisation illustrated in equation (5) and Penex catalysts, the Butamer catalyst involves dehydrogenation of the paraffin by brings about hydroisomerisation in a con- the platinum function to an olefin as an tinuously self-cleansing manner, so that it is initiating step, followed by conversion of the

Platinum Metals Rev., 1964, 8, (1) 6 A Hydrar process unit at an Oklahoma rejinery. A platinum catalyst is used for the conversion of benzene to high purity cyclohexane, mainly for subsequent oxidation to adipic acid in the preparation of Nylon-type polyamides olefin to a carbonium ion by the acid function, catalyst as shown in equation (7). The isomerisation of the carbonium ion, genera- platinum serves to maintain catalyst cleanli- tion of an iso-olefin from the isomerised ion ness and high activity through continuous and, finally, rehydrogenation of the iso-olefin hydrogenation of polymeric carbonaceous to an isoparaffin. At the lower temperatures by-products, thus permitting their con- of operation, desired because of favourable tinuous removal. equilibrium conditions for butane isomerisation, the dehydrogenation step is so slow as Hydrar Process to offer a serious obstacle to feasible reaction Thus far we have considered three pro- rates if catalysts of the Penex or Platforming cesses employing acidic platinum-containing type are used. catalysts in which the acidity has been As a result of the enhanced acidity of the modified to conform with the requirements Butamer catalyst, it is believed that the of each process. In the Hydrar process, a initial paraffin activating step differs from platinum-containing catalyst is used for the that of the higher temperature operation convcrsion of benzene to high purity qclo- employed with Penex or Platforming catalysts. hexane. The Butamer initiating reaction is thought to Such cyclohexane is mainly used for oxida- be one of hydride abstraction via the acid tion to adipic acid in the preparation of function, as shown in equation (6). Once the Nylon-type polyamides. So exacting are the reaction has been initiated, it is propagated purity requirements of cyclohexane for this by a chain mechanism under the influence of use that no more than trace amounts of other the exceedingly acidic sites of the Butamer hydrocarbons may be tolerated.

Platinum Metals Rev., 1964, 8, (1) 7 Since the Hydrar reaction is conducted at a the feedstock, essentially quantitative yields temperature at which from about 2 to 6 per of cyclohexane are produced during several cent of methylcyclopentane might be formed years of continuous operation without con- if an isomerisation equilibrium with the tamination or deactivation of the catalyst. cyclohexane product could be established, it was necessary to employ a catalyst that did Summary not induce isomerisation. In addition, to The Platforming catalyst for reforming assure virtually quantitative yields of cyclo- naphthas, the Penex catalyst for hydro- hexane, it was necessary to suppress all isomerisation of pentanes and hexanes, the cracking to open-chain compounds. Butamer catalyst for hydroisomerisation of It will be noted from equations (2) and (4) butane, and the Hydrar catalyst for conver- that the acid function of duofunctional sion of benzene to cyclohexane, provide catalysts plays a key role in both the isomerisa- examples of four supported platinum catalysts tion and cracking of naphthenes. The Hydrar of widely differing properties obtained by catalyst which was developed for benzene regulating the acidity of thc catalyst com- hydrogenation is, therefore, one in which posite. Through proper balance of the the platinum is disposed on a specially platinum hydrogenation-dehydrogenation prepared support in which the acidity has function with the acid function, in co-action been virtually completely suppressed. As a with the support, such duofunctional catalysts result, benzene is hydrogenated cleanly to can provide a high degree of selectivity for cyclohexane of over 99.7 per cent purity in the hydrocarbon conversion processes operating Hydrar process. If a pure benzene is used as under widely different conditions.

Oxygen Injection Engine for Space Research PLATINUM ALLOY VALVE SEATS AND FACES Formidable design problems are encountered the plating surfaces and with good ability to in the development of a reciprocating power compensate for minor misalignment. unit for use in space research programmes. One of the most severe of such problems met with by the Vickers Inc. Division of Sperry Rand Corporation in developing a light- weight hydrogen-oxygen internal combustion engine (shown on the right) concerned the selection of materials for the oxygen injector valve. As this valve, operating at very high speed, handles gaseous oxygen at high temperatures, it must be made of a material that will resist oxidation and also maintain ade- quate strength and impact characteristics. Tests were carried out on austenitic stainless steel, a nickel-chomium-molybdenum alloy and 10per cent rhodium-platinum alloy for the poppet valve face, and on stainless steel, 10 per cent rhodium-platinum and the latter alloy flame-plated with alumina for the valve seat. The best combination was found to be a rhodium-platinum poppet face against a flame-plated rhodium-platinum seat. This combination has endured several hours of operation without leakage or deterioration of

Platinum Metals Rev., 1964, 8, (1) 8 The Magnetic Properties of Platinum Metals and Alloys

A SURVEY OF RECENT RESEARCH

By B. K. Coles, D.Phi1. Department of Physics, Imperial College, London

Well-marked magnetic properties are series (Co+Rh+Ir), and strong magnetic associated always with electrons in incom- interactions are absent in osmium, iridium plete inner electron shells which maintain a and platinum because the 5d electrons on comparatively localised and atomic character neighbouring atoms are behaving more like in the solid state - unlike the delocalised conduction electrons than like localised ones. conduction electrons which become a pro- Palladium, at the end of the second transition perty of the solid as a whole. There is not, series, is very nearly a magnetically ordered however, an absolutely clear-cut division material, and small amounts of iron in solid between localised and non-localised situa- solution produce a ferromagnetic moment tions. The best example of the well- which is not all provided by the iron atoms. localised situation is provided by the Even stronger effects are produced when incomplete f-electron shells of the rare small amounts of iron are added to those earth metals. In gadolinium the 7f-electrons palladium-rhodium alloys (-5 per cent per atom play no part in bonding and rhodium) where the work of F. E. Hoare f-electrons on neighbouring atoms in the and others at the University of Leeds has metal are only “aware of one another” shown that the conditions required for because of indirect coupling through the ferromagnetism are very nearly satisfied. three conduction electrons per atom which A large amount of work has been carried can, as it were, tell one atom of the orienta- out recently on dilute alloys of transition tion of the magnetic moment provided by metals of the 3d series dissolved in palladium the f-electrons on another atom. or platinum, much of it at the Bell Tele- The 3d electron shells of the first transition phone Laboratories. The results have been group metals (especially manganese, iron, compared with the magnetic behaviour of cobalt and nickel) are not as well localised solutions of iron in the earlier members of as the rare earth f-electrons, and in the pure the 4d series, where iron does not always metals strong direct interactions between carry a magnetic moment. In niobium d-electrons on neighbouring atoms are where the electronic density of states is certainly present. (We still do not know high dissolved iron atoms have no moment ; whether these direct interactions or indirect but the high density of states in platinum ones through the conduction electrons are and palladium does not prevent the appear- dominant in providing the ferromagnetism ance of a moment on the iron atoms, of iron, cobalt and nickel.) The d-electrons presumably because of the more localised become somewhat more localised on passing character of the d-electrons of the solvent. along a transition series (Ti-fNi) but If manganese, iron or cobalt is dissolved appreciably less localised on going from the in a non-transition metal (for example gold) first to the second to the third transition the d-electrons have no neighbouring d-elec-

Platinum Metals Rev., 1964, 8, (l), 9-1 1 9 trons to interact with and become more The striking change with temperature in localised -that is, more like the f-electrons magnetic properties in iron-rhodium alloys or rare earth metals - and there are interest- containing equal atomic percentages of each ing resemblances between lanthanum- element has been known for some time, and a gadolinium* alloys and gold-iron or short review of the data available up to copper-manganese alloys, since in all three about a year ago was given in this journal the only magnetic interactions are via the in January last year (2). This ordered CsCl conduction electrons. From this point of type alloy changes from antiferromagnetic view platinum, iridium and rhodium can to ferromagnetic at -360°K in zero external perhaps be regarded as simple (‘non- field, but at -220’K in a field of 120,000 magnetic” solvents for iron, but palladium gauss. certainly gives a different and more compli- Now Flippen and Darnel1 of du Pont cated situation. Recent work by the present (3) have discussed the thermodynamics of author has revealed striking effects in the this field dependence; while Shirane, Chen low temperature electrical resistance of and Flinn of Westinghouse, together with dilute solutions of iron in palladium, Nathans of Brookhaven (4) have published platinum, rhodium and iridium. These the results of Mossbauer studies of the seem to be related to, but much stronger magnetic field seen by the nucleus of an iron than, the well-known anomalies in electrical atom and the differences for iron atoms in resistance shown by the above mentioned (‘right” positions and “wrong” positions. gold-iron and copper-manganese alloys. Kouvel and his co-workers at the In more concentrated alloys of the iron Schenectady laboratory of General Electric group metals neighbouring atom interactions who carried out transformation studies on become important, and striking effects can the iron-rhodium alloy, have now moved sometimes be produced by transitions from on to investigate the possibility of an disordered structures to ordered structures. analogous behaviour being found in its FeAl and Au,Mn provide examples of this, isomorph MnRh (5). but for metallurgical reasons, such as There is a martensitic transformation size factors, most extensive solid solutions (with hysteresis) from a high temperature where ordering is possible are in alloys with ordered CsCl structure to a low temperature platinum metals. A striking example is the CuAu structure. The high temperature Pt,Fe alloy, which if perfectly ordered (no form is paramagnetic (it would probably Fe-Fe nearest neighbours) is antiferromag- become antiferromagnetic at low tempera- netic; but a few extra iron atoms that tures) and the low temperature form is occupy platinum sites convert much of the strongly antiferromagnetic. The change neighbouring material to a ferromagnetic therefore produces a change in density and state (I). resistance, but these are not sharp since there seems to be a rather extensive range Iron-Rhodium Alloys between the start and end of the martensite A number of papers presented to the transformation. In this connection a survey Eighth Conference on Magnetism and of the behaviour of the related phases Magnetic Materials held recently in FeRh, MnIr, MnNi and the effects of Pittsburgh dealt with various aspects of the substitutions into these of other transition magnetic properties of the platinum metals metals would probably be of interest. and their alloys which can be discussed in relation to the points made above. Platinum Alloys Alloys of platinum with small amounts *This notation indicates a dilute solid solution of gadolinium in lanthanum. of manganese, iron, cobalt and nickel were

Platinum Metals Rev., 1964, 8, (1) 10 described by Bozorth, Davis and Wernick atoms next to an iron atom have a moment of Bell Laboratories to the International and that it is -0.5 units. Conference on Magnetism in 1961 and were Gadolinium-palladium alloys with small compared with similar palladium-based amounts of gadolinium have been studied alloys (6). extensively by workers at the Bell Labora- For the platinum alloys effects are much tories, using the techniques of electron weaker (the highest Curie temperature for a paramagnetic resonance. The magnetic I per cent alloy (Co or Fe) is -17°K as field for this resonance is significantly compared with 80°K for Pd-Co I per cent different from that for gadolinium in alloys and the size of the magnetic moment with non-transition metals, so that although produced on platinum atoms is very much the magnetic properties are dominated less than that produced on palladium atoms. by the 7 localised 4f-electrons of gadolinium It is clear that even if the d-electrons of iron the nearly magnetic character of palladium can be regarded as localised when it is does modify matters slightly, although less dissolved in platinum (like the f-electrons strongly than in palladium-iron. This has of gadolinium when dissolved in lanthanum) now been shown (9) by adding hydrogen to the strength of the interaction between them palladium containing gadolinium and (the indirect interaction via the conduction thereby filling up with extra electrons the electrons) is very much stronger than in d-electron states of the palladium. The non-transition solvents because the conduc- effect is to make the gadolinium behave as it tion electrons in platinum have a lot of does in non-metallic situations or in solution &character themselves. in simple metals. The same effect is In another paper to the Pittsburgh Con- produced, incidentally, by adding silver to ference Pickart and Nathans (7) have palladium containing gadolinium. (With reported neutron diffraction studies on the metals osmium and ruthenium gado- manganese-platinum and chromium-plati- linium forms weakly ferromagnetic inter- num alloys. Both have the Cu,Au structure metallic compounds GdOs, and GdRu,, but and there seems to be some magnetic the similarity in properties to GdA1, moment on the platinum atoms, but in suggests that the transition metal character Pt,Cr this is in the opposite direction to the of the osmium and ruthenium is less tenfold larger moment on the chromium important than that of palladium in the atoms. above alloys).

Iron-Palladium and References Gadolinium-Palladium Alloys I G. E. Bacon and J. Crangle, Proc. Ray. SOC., Iron-palladium alloys with small amounts 1962,272, 387 2 A. S. Darling, Platinum Metals Rev., 1963, 7, of iron have also been studied by neutron 29 diffraction by Cable and his colleagues at 3 R. B. Flippen and F. J. Darnel1,J. Applied Oak Ridge (8). The magnetic moment on the Physics, 1963,34, (4), 1094 4 G. Shirane, C. W. Chen, P. A. Flinn and iron atoms has been found to be equivalent R. Nathans, ibid, p. 1044 to 3.0 electron spins (as compared with 2.22 5 J. K. Kouvel, C. C. Hartelius and L. M. in pure iron). There is a moment of 0.15 Osika, ibid. p. 1095 units on the palladium atoms in the 3 per 6 R. M. Bozorth, D. D. Davis and J. H. Wernick, J. Phys. SOC.Japan, 1962, 17, cent iron alloy. This has been calculated Suppl. B.I., 112 assuming that all palladium atoms have 7 S. J. Pickart and R. Nathans, J, Applied similar moments. There is other evidence Pb'SzCS, 196.3, 34, (41, 1203 8 J, W.Cable, E. 0. Wollan and \XI. C. Koehler, to suggest, however, that for very dilute ibid, p. 1189 alloys down to I per cent iron only palladium 9 D. Shaltiel, ibid, p. 1190

Platinum Metals Rev., 1964, 8, (1) 11 Platinum-lined Furnace for the Fluorination of Uranium Compounds

FEATURES OF DESIGN AND CONSTRUCTION

The high temperature stability of platinum The furnace liner requirements were and its immunity from attack by hydrogen found to be ideally fulfilled by the use of fluoride gas even at IOOO~Chave led to its platinum in the form of 0.010 inch thick being specified to form the liner material for sheet, and the furnace chamber, the purge a fmace recently designed and constructed outlet and inlct tubes and the chamber lid are for use in the fluorination of uranium sheathed with the metal. compounds at the Harwell research establish- AS pure platinum surfaces readily weld to ment of the United Kingdom Atomic Energy each other at high temperatures, a $-inch Authority. The design, testing, and opera- gap was left between the lid and the sides of tion of the furnace are described in report the furnace chamber after lining. A water AERE-R~z~~by D. N. Fletcher and L. J. cooled joint ring between the lid and the White of the Scientific Services Department furnace chamber was not permissible and of the Engineering Division at Harwell. after many tests a gasket of >&-inchthick The high level of radio-activity exhibited white asbestos millboard was used to prevent by uranium compounds, and eventually by the platinum surfaces from coming into the equipment, necessitated that the furnace contact and to provide an effective seal. To should be so designed that all operations avoid the platinum trays from sticking to the could be conducted inside a glove box, yet chamber lining a separator grid of a rhodium- easy access to the furnace chamber and platinum alloy was placed on the floor of the element housings had still to be retained. furnace. The basic design specification called for a Temperature control is effected by the use furnace capable of producing a chamber of three platinum : rhodium-platinum thermo- temperature of 10oo~Cand able to operate at couples, one sealed into the purge inlet tube a working pressure of I p.s.i. (gauge). A and protruding into the furnace chamber, the furnace liner was demanded that would be second fitted into a pocket at the centre of the unaffected by hydrogen fluoride gas at lid with its hot junction in contact with a flat- IOOO~C,would protect the charge from headed platinum rivet from the platinum- contamination by preventing the formation lining, and the third connected to one of the of corrosion products inside the furnace four silicon carbide heating elements. chamber and would itself be unaffected by In the furnace chamber, on the projecting contact with uranium compounds. The box section of the lid and in the purge inlet furnace chamber was required to be approxi- and outlet tubes, the thin platinum liner is mately 8 x 8 x z inches, to accommodate supported on Q-inch thick Inconel, which existing platinum trays which were to bear was selected bccause of its good high temper- the charges, and there was to be no water ature properties. At the development stage cooling of the joint between the lid of the much investigation was carried out to enable furnace and the chamber in order to avoid precautions to be taken to prevent contamina- any hazard that might arise if water were tion of the platinum liner by its close contact to come into contact with the charges. with Inconel. When platinum is in contact

Platinum Metals Rev., 1964, 8, (l), 12-13 12 The furnace chamber and the bor- sprtion lid lined with platinum

at high temperatures with certain base metals, in this case principally chromium, contamination of the platinum can result from the diffusion of these elements into its surface, leading to embrittlement and subsequent failure. To avoid this risk, a barrier was placed in the path of the diffusing elements between the platinum liner and the Inconel support, this taking the form of a thin layer of a proprie- tary aluminium silicate fibre. This material, which can withstand temperatures in the region of IOOO~C,has proved highly successful in protecting the platinum liner and also serves as a soft buffer layer was completed. The furnace construction between the metal surfaces. and assembly were carried out by R. M. Once this contamination problem and the Catterson-Smith Limited, the platinum lining difficulty of finding a suitable gasket material being manufactured and fitted to the furnace were overcome the construction of the furnace by Johnson Matthey & Co Limited. J. A. S. Determination of Thermal Conductivity PLATINUM AS A REFERENCE STANDARD

The growing demand for materials to Laboratory over the range oo to 950°C has operate at high temperatures has brought a now been reported by R. W. Powell and R. P. need for accurate knowledge of the manner Tye (Brit.J.Appl. Phys., 1963, 14,662), who in which they conduct heat. Reliable figures have for the first time used substantial bars for thermal conductivity are, however, notor- of platinum for the measurements. Two sets iously difficult to determine, and standard of observations were made on bars having reference materials would be invaluable for diameters of 2 and 4 inch respectively; with checking the reliability of testing equipment both samples the conductivity was found to and as standards in comparative methods. remain constant within 0.5 per cent of Platinum has many advantages as a standard 0.73 w cm-l deg C-l over the whole range. for use at high temperatures, but the values This result is as much as 20 per cent lower previously determined by various authorities than four out of the five previous deter- for thermal conductivity up to about IOOO~Cminations, but yields values of Lorenz have shown significant differences. A careful function in much closer agreement with the redetermination at the National Physical theoretical. J. C. C.

Platinum Metals Rev., 1964, 8, (1) 13 Platinum Metals in Electrochemistry J PAPERS AT THE MOSCOW CONFERENCE

By T. P. Hoar, s~.D. Department of Metallurgy, University of Cambridge

At the fourteenth meeting of the Inter- several technically important reactions. Thus national Committee for Electrochemical V. S. Bagotzky pointed out that both Thermodynamics and Kinetics (CITCE) oxygen reduction at a platinum cathode and in Moscow in August interest in the platinum organic oxidations at a platinum anode arc metals as “inert electrodes” continued to be inhibited by oxidation of the electrode sur- manifest. Electrochemists have long used face. E. V. Kasatkin and A. A. Rakov showed platinum as an “inert” basis for hydrogen, that persulphate production at a platinum oxygen and halogen gas electrodes, as well as anode in cooled sulphuric acid solution is for many redox electrodes such as the ferro- inhibited by oxidation of the anode; such cyanide/ferricyanide couple. It proved to be oxidation is accompanied by an increase of very difficult to set up a reversible oxygen/ the a.c. resistance and a decrease of the ax. hydroxyl-ion electrode on platinum; as long capacitance of the electrode, typical behaviour ago as 1933 I summarised the then scanty when very thin oxide films are formed. Also, knowledge of this system, and gave experi- the current-density changes at the platinum mental evidence (I) that the platinum acquires anode when its potential is fairly rapidly an oxide film that renders it a poor catalyst varied (the chronopotentiometric technique) for the 0,+zH20+4ee40H- change, as depend on its pre-history. Likewise, compared with the rather good catalytic Gonzalez, Capel-Boute and Decroly reported action of unoxidised platinum for the that the state of the platinum surface is hydrogen-electrode reaction, zHf fzeZH,. important for sulphur dioxide anodic oxida- It may be worth recalling that this evidence tion. M. W. Breiter, in one of a long series was, to say the least, somewhat sceptically of papers, noted that the anodic oxidation of received in “pure” electrochemical circles ;to methanol becomes less easy on a series of the pure all was pure-certainly the surface platinum metals in the order Pt>Pd>Rh>Ir of platinum. and that it is much reduced on oxidised sur- Today, with the greatly increased technical faces. The anodic oxidation of methanol on interest of “inert” platinum metal electrodes oxidised and “bare” platinum was examined in such fields as fuel cells, chlorine and per- in detail by J. E. Oxley, G. K. Johnson and salt production, and cathodic protection, a B. T. Buzalski of Leesona Moos Laboratories, very intensive study of the behaviour of the New York, who found little or no reaction platinum metals is being made. It is now unless the platinum was “bare”. Finally, generally agreed (2, 3, 4) that a platinum S. Toschev and B. Mutaftschiew of the surface is far from being “inert”, but can be Bulgarian Academy of Science reported that oxidised under several kinds of conditions; mercury will electrodeposit around the < I I I > only the form of the oxidation is in some cases poles on platinum single crystals, but not uncertain. around the , unless the initial oxide At the Moscow meeting, much attention film is removed cathodically or by drastic was paid to the influence of the surface state pickling: then, nucleation at the ( 100) poles of platinum on its effectiveness as a basis for is even faster than at the {III). Evidently,

Platinum Metals Rev., 1964, 8, (l), 14-15 14 the surface of platinum may be vitally affected Cl-

References I T.P.Hoar ...... Proc. Roy. SOC.,1933 A142,628 z T.P.Hoar ,...... Proc. Eighth Meeting CITCE, Madrid 1956, p. 439. Butterworths, London (1958) 3 J. O’M. Bockris and A. M. K. S. Huq . . Proc. Roy. Soe., 1956 A237,227 4 A. Damjanovic and J. O’M. Bockris . . Private communication, 1963 The CITCE papers will be published shortly in Electrochimica Acra

A High Temperature Waveguide Termination PLATINUM AS BOTH GUIDE AND HEATING ELEMENT In measuring the output from various of the Royal Radar Establishment, Great noise sources by means of a radiometer, a Malvern, describes (J. Sci. Instr., 1963, 40, standard noise source is required to calibrate (11), 524) a hot load that has been developed the instrument. This generally takes the form for operation at 1250°C. This load comprises of a hot load, i.e. a termination of known a one inch long, conically tapered rod of temperature connected to the radiometer by a pyrophyllite operating at this temperature in a suitably matched waveguide. For most heated circular waveguide. microwave receivers, a load immersed in In the construction of this waveguide boiling water can be used. assembly platinum sheet 0.005 inch thick is Because of the high noise level of microwave formed around a mandrel to make the round receivers in the short millimetric band, a waveguide 0.143 inch in diameter, the ends of relatively high output is also necessary from the sheet being extended parallel to each other the standard noise source if accurate calibra- to copper blocks. The platinum is resistance tion of the radiometer and hence the receiver heated by a current of 500 amp passed through is to be achieved. This makes it necessary to the copper blocks. The whole assembly is operate the hot load at an extremely high enclosed in laminated asbestos to reduce heat temperature. In a recent paper, Q. V. Davis losses. J. G. W.

Platinum Metals Rev., 1964, 8, (1) 15 Organomet allic Compounds of the Platinum Metals

A SURVEY OF THE TYPES OF COMPOUNDS, THEIR STRUCTURES AND REACTIONS

By G. Wilkinson, Ph.D., A.R.C.S. Professor of Inorganic Chemistry, Imperial College, London

into the ways in which carbon compounds or Although platinum provided the jrst radicals can be attached to transition metal known exurnpie of an organorndlic atoms-especially regarding the unusual types complex of a transition metal, intensive of chemical bonds not found in the chemistry studies of cornpounds of the plutinurn of non-transition metals and non-metals. metals with metal to carbon bonds have However, the chemical utility of these come mainly during th past decade. complexes, as chemical compounds, is virtually The main chsses of srlable organo- zero at the present time. Despite this fact, complexes such as those having oleJilas, such organometallic species are of considerably acetylenes, alkyl or cyclopentadienyd more than purely academic interest since groups as 1iganJs are discussed. The their structures and the nature of the bonding study of such stable complexes should in the compounds may throw light on the assist in the elucidation of the mcch- mechanisms involved in the use of platinum onism of carious reactions which are metal catalysts, especially in homogeneous catalysed by platiniirn metal salts or media, and may allow the design of new complexes. useful processes. Some of the necessary information may already be at hand and we Although a platinum ethylene complex, need only consider that the facts of the Zeise’s salt, K[C,H,PtCl,], was the first hydrolysis of [C,H,PtCl,]- to acetaldehyde platinum metal-indeed, the first transition and of the reduction of palladous salts by metal-organometallic compound to be pre- ethylene were known for many years (2) pared (I), and although the methyl platinum before the development of the Wacker olefin compounds such as (CH,PtI), were among oxidation process. Other reactions which the first metal alkyls to be prepared, it is doubtless involve labile organometallic inter- generally true to say that only during the mediates are the stereospecific polymerisation past twelve or so years when the organo- of butadiene using emulsions containing metallic chemistry of the transition elements rhodium salts, polymerisations of acetylenes, has been intensively studied have the organo olefin isomerisations and hydrogenation and compounds of the platinum metals been carbon monoxide insertion reactions of examined in detail as part of this general olefins using ruthenium, rhodium or platinum research effort. species. It is not the purpose of this brief The organometallic compounds of the review to discuss such matters but only to platinum metals are of particular interest consider the isolable organo compounds of since they have provided considerable insight the platinum metals.

Platinum Metals Rev., 1964, 8, (l), 16-22 16 Some Representative Platinum Metal Organometallic Compounds

Co m p 1ex Form Structure type Comment

Colourless oil z-cyclopentadien yl Volatile. Stable in air m.p. -5" alkyl Petroleum soluble Yellow crystals Alkyl Formed by CO insertion into m.p. 50" PdC1CHs(PEt3) Pale yellow Olefin Water-soluble crystals Yellow crystals Olefin Soluble organic solvents. m.p. -zqo"(dec.) (nor bornadiene) Air-stable. Halogen bridged White crystals Acetylene Stable. Soluble in organic m.p. 215" solvents Orange crystals Acetylene-derived Extremely chemically inert map.150" cyclopentadienone White crystals Sandwich Undergoes aromatic substi- tution reactions. Oxidisable (z-CjHg) ,Ru+ C,H,OPdCl(C,H,N) Yellow needles Allylic Has mesityl oxide bound as ally1 group with free ketone group I

As part of the main d-block transition to recent review articles. The table gives the elements, the six platinum metals have certain properties of a few representative compounds. similarities in common with the rest of the group, and many of the classes of compounds Alkyls discussed below are found also for other These are compounds of the type usual metals. For a specific metal, not all types of among non-transition and non-metallic compound may be known-or indeed be elements but less common among transition capable of existence. In some cases, metals, in which aliphatic or aromatic explanations of special behaviour may be radicals are bound to a metal atom (3). possible on an ad hoc basis. However, the There are two classes: (a) normal alkyls, reasons for stability or criteria for bonding in where the M-CE bond is of the normal certain types of complex are not well known 0-bond type, (6) those compounds where the and quantitative data such as bond energies hydrocarbon radical, usually a methyl group, and other thermodynamic data, bond dis- acts as a bridge, tances, etc., are often lacking. Nevertheless HHH there is already an appallingly extensive \I/ ..C literature, which is growing rapidly, covering M,' ,~ '.M especially preparations of compounds and their reactions and of recent years details of and where the bonding must be regarded as infra-red and high-resolution nuclear magnetic of the multicentre type as in the polymeric resonance spectra. The latter technique has alkyls of Li, Be or Al; this type of bonding is been of especial importance in the study and not found for transition metals. characterisation of organometallic compounds. Although no simple binary alkyls, i.e. In the following sections we survey, with M(CR,),, are known for the platinum group examples chosen from the platinum metal of metals, many complex alkyls where other complexes, all of the major types of organ0 ligands are also bound to the metal atom are compounds known at present. References are well characterised. The criteria for obtaining

Platinum Metals Rev., 1964, 8, (1) 17 alkyls stable enough to isolate under normal conditions and various factors, whose effects I , are not well understood, are complicated. Some of these factors are (i) the oxidation state of the metal, (ii) the nature of the attached ligands, (iii) the electronegativity of the metal-ligand system-this involves both (i) and (ii), (iv) the electronegativity of the attached -CR, group, (v) steric factors such I as large bulky groups around the metal which can inhibit the approach of attacking molecules, so stabilising the M-C bonds. Concerning the ligands, it appears that x-bonding ligands, which, in effect, remove electron density from the metal, tend to stabilise M-CR, bonds ; however, though carbon monoxide is doubtless the best J;-bonding ligand, platinum metal carbonyl alkyls have not been isolated. For compounds of analogous structure, the greater the carbonylation reactions to give acyl deriva- electronegativity of the -CR, group, generally ties, e.g. the more stable the M-C bond and fluorinated (Et3P),PdC1CH3+ CO =(Et3P)2PdC1(COCH3) derivatives are especially stable. In the absence of any data on M-C bond energies, Such carbon monoxide insertion reactions however, explanations of relative stabilities are believed to be involved in hydroformy- of complex systems are not necessarily of lation reactions of olefins. much significance. Among the alkyls, there are a few points Olefin Complexes of special interest. Platinum, especially PtIV, Compounds containing >C=Ci groups of all the metals appears most readily to can generally be bound to transition metals form stable Pt-C bonds. This is illustrated (4a,b,c). In addition to mono-olefins, cyclic by the fact that whereas p-diketones form conjugated and non-conjugated olefins can chelate complexes normally, with PtIV the be bound and in this class can also be ligand forms an M-C bond as in (I) and this included such unsaturated compounds as is so stable that in reactions, e.g., with quinones or cyclopentadienones although the 2,~'-dipyridylit is the Pt-0 bond and not the latter are usually obtained in metal complexes Pt-C bond which breaks, to give (11). indirectly through the interactions of acety- The square planar complexes of Pd", lenes with carbon monoxide containing metal Pt" and Ir', chiefly with phosphine ligands, complexes (see below). Olefin complexes were can undergo two unusual reactions. They the first and are the most numerous of can add alkyl halides, halogens or hydrogen organ0 compounds of platinum metals. They halides to give octahedral species of formally are commonly obtained by direct interaction oxidation state two higher, e.g. of platinum metal salts or complexes with olefins, e.g. trans-(Et3P),PtICH3+CH31= aqueous (Et3P)Zpt12(CH3)2 KzPtCl,+CzH, -+ K[C2H4PtC13]-+ KC1 so1 u t io n and they can, like complexes of metal carbonyls such as CH,Mn(CO),, undergo

Platinum Metals Rev., 1964, 8, (1) 18 Apart from the salts, the complexes are stronger donors are used, to displacement of usually air-stable crystalline solids, soluble the olefin. Where not all of the olefinic groups in organic solvents. in a polyolefin are bound to the metal, the The nature of the binding of an olefin to “loose” double bonds may be protonated by a metal is fairly well understood. X-ray acids to give cationic species. These are not diffraction on several compounds has shown usually stable unless it is possible for the that the orientation is generally as in (111) hydro-carbon entity to rearrange to give a partially delocalised allylic type system (see below) in which case the resulting cations may be quite stable. For example in cyclo- octatetraene complexes, x-C&RhC8H8, where the olefin is bound in the form shown in (IV), gives an unstable species on pro- tonation since a delocalised system cannot with the olefin group “side-on” to the metal. be formed, whereas C,H,Fe(CO), (V) can The C=C bond length is not precisely give such a system. known but it appears to be longer than the The inter-relationships of olefin complexes unco-ordinated bond. There is also infra-red and allylic species are still in the process of cvidence which indicates C-C bond weakening development. It is noteworthy that in some on co-ordination. It is not essential that the instances the metal appears to be the dominant metal atom lies on a line perpendicular to the factor. Thus palladium has given allylic C-C axis-indeed in many cyclic olefin complexes whereas, with the same ligands, complexes it is not possible for this to be so. platinum forms the olefin complex. Thus Regarding the bonding, it is generally accepted with mesityl oxide we have the species (VI) that the bond is rather similar to that in the and (VII). metal-carbon monoxide bond in metal carbonyls with donation of electron density Acetylenic and Acetylene-derived from the %-electronsof the C=C bond into Complexes acceptor orbitals on the metal together with a Acetylene and acetylenic compounds can type of %-bondresulting from “back-donation” interact with platinum metal species in a of electron density from filled metal orbitals number of ways depending on the nature of to the empty anti-bonding orbitals on the the reacting species and on the reaction carbon atoms. Estimates of the relative conditions. In addition to forming complexes, contributions of these two components are difficult to make but nuclear magnetic evidence suggests that the amount of “double bond” character in the metal-olefin bond (unstable cation) is rather small-insufficient in fact to prevent free rotation of ethylene about the bond axis in simple ethylene complexes. There has been relatively little work on the reactions of the hydrocarbon when bound to metals. (stable cation) Attack by many reagents leads to oxidation of the metal and dis- ruption of the complex or, where V

Platinum Metals Rev., 1964, 8, (1) 19 0 A

VI XI VIII

compounds some which may be allylic in cH3 of / -3 nature in addition to other reactions. Thus o=c n / - in alcohol, hexaphenylbenzene is formed together with (IX); this complex with hydrochloric acid gives (X). The formation of hexaphenylbenzene is an example of the /c\ polymerisations that can occur; similarly CH, CH, hexafluorobut-2-yne can give CJCF,),, VII C,(CF,),CO or (C4F&,, the latter being an infusible white powder. the acetylenes may be polymerised catalyti- Another class of complex is that which cally. There are two main classes of compound : appears to have the acetylene bound by (4 a, b) (a) those in which the acetylenic o-bonds as in (XI) which can be made by group is still recognisable-and sometimes treating PtCl,(PPh,), with hydrazine in recoverable-as such; (b) those in which the presence of the acetylene. Acetylenes can identity of the acetylene has been lost by displace each other in such complexes, the polymerisation and complexation of the one with most electronegative groups being polymer; in presence of carbon monoxide or bound most strongly. metal carbonyl compounds, the polymerisation can occur with incorporation of CO to Sandwich Compounds give a cyclopentadienone. Although extensive This class of compound, whose name studies have already been made in this area, derives from the structure of (x-C5HJ2Fe, mostly on Pd, Pt and Rh, we can quote only ferrocene, is that in which a completely a few illustrative examples. symmetrically delocalised carboxylic system Hexafluorobut-2-yne reacts with with six x-electrons is bound to a metal (5 a, b). x-CjH,FUi(CO), to give a cyclopentadienone The most extensive class is that of the complex (VIII). Diphenylacetylene reacts 7c-cyclopentadienyls, but arene and cyclohep- with palladium halides to give series of tatrienyl complexes are known, although the

Pd Pd C1, Ph OEt

Ph Ph Ph Ph IX X Ph3P CS6 Hqa (Octaphenylcubane)

Platinum Metals Rev., 1964, 8, (1) 20 \/ \/ C C \/ /7 H-C-M T\ C /\ n R XI I XI11 (anri) latter not for the platinum metals. In some Arene complexes are of little importance cases, two rings can be bound to the metal for platinum metals although a few benzene but in others one ring with other ligands and substituted benzene species of Ru, 0s occupying the remaining co-ordination posi- and Rh are known; they are not especially tions. A x-C,H, ring can crudely be stable. regarded as occupying three co-ordination sites on a metal. Allylic and Related Complexes After the recognition of the sandwich These complexes have been noted in structure, the ruthenium analogue was the passing above. They are in a sense inter- first other complex of this type to be made; mediate between olefin and sandwich com- it is one of the few organo complexes of pounds in having a partially delocalised platinum metals which are commercially “open-ended”, x-electron system bound to available. The osmium analogue also exists a metal (4a,b). The allyls can be written and cationic species, e.g. (x-C,H,),Rh’, for as involving canonical variants of (XII) but Ru, Os, Rh and Ir. The ability to form nuclear magnetic resonance spectra have sandwich compounds appears to be related shown that a more realistic and correct to the stability of a particular oxidation state formulation is (XIII) or (XIV). Two isomers, of the metal and to its ability to give the anti and syra with respect to the hydrogen pseudo-octahedral bonding orbitals necessary atom on the “central” carbon atom not only to provide overlap with the rings. There are exist theoretically but have been separated no stable neutral (x-C,H,),M compounds of in some cases. Rh, Ir, Pd and Pt since these metals have The first allyls-which are of quite either an unstable I1 state or give normally recent recognition-were obtained from the square planar complexes. Attempts to reduce (x-C,H,),Rh+ lead to a cyclopentadiene olefin complex of Rh‘ NaBH, (T-C5H,),Rh+ +H- :+ --C5H,RhC5H6 in tetra- xv hydrofuran All cyclopentadienyl complexes are made by essentially standard procedure, namely H H the interaction of a metal halide or complex \/ C CI halide with the sodium salt C,H5-Na in tetrahydrofuran solution. Mono x-cyclopentadienyls result when a complex halide with other suitable ligands are used, e.g. for ally1 palladium chloride [C,H,PdCl] +zC,H,Na -+zz-C,H,PdC,H, XVI

Platinum Metals Rev., 1964, 8, (1) 21 interaction of palladous chloride with ally1 more experimental work will be required to alcohol. They can be obtained in a variety ascertain the reaction mechanisms in detail. of ways. The recognition of allylic binding also led to the reformulation of a number References of known complexes, previously held to I H. Zeise, Pogg. Annalen, 1827, 9, 632 be olefin complexes. Typical examples are 2 See N. V. Sidgwick, The Chemical Elements the I :3-cyclohexadiene palladium complex and their Compounds, Vol. 11, pp. 1560, 1588. Oxford which is unequivocally now known to be 3 Far comprehensive reviews see F. A. Cotton, (XV) and the butadiene palladium halide Chem. Rev., 1955, 55, 551; see also for complex now shown to be (XVI). It is now discussion J. Chatt, Proc. Chem. SOL.,1962, 318 recognised that allylic species may be most 4 For reviews see: (a) M. A. Bennett, Chem. important intermediates in a variety of Rev., 1962, 62, 11; (b) R. G. Guy and B. &. Shaw, Advances in Inorganac and Radio- reactions involving olefins and platinum chemistry; 1962, 4, 77; (c) E. 0. Fischer and metal salts. It seems probable, for example, H. Werner, Angew. Chem. International, 1963, that the polymerisation of butadiene by 2,80 5 For reviews see: (a) G. Wilkinson and F. A. rhodium salts is of this type and doubtless Cotton, Progress in Inorganic Chemistry, other olefin reactions will be interpreted in 1939, I, I; (b) E. 0. Fischer and H. P. Fritz, Advances in Inorganic and Radio- this way in future, although considerably chemistry, 1958, I, 56

P rc~isionGlaze K esistors PALLADIUM IN NEW SCREEN PRINTING COMPOSITIONS

For use in printed circuits there is a demand (0.1 to 0.5~)palladium and silver as con- for resistors with a wide range of values and ductors. The oxidation-reduction behaviour good electrical properties and capable of of palladium heated in air is thought to being produced by the simple method of promote sintering of the palladium and silver screen printing and firing the resistor film on particles into chain-like aggregates. These to a ceramic substrate. In a paper given to the are claimed to give continuous rather than Electronics Division of the American Ceramic particulate conduction, thereby reducing the Society in October 1962, and now published dependence of resistance on metal concen- in the A.C.S. Bulletin (1963, 42, (9), 490), tration in the glaze, promoting heat dissipation L. C. Hoffman, of the Electrochemicals and reducing current noise. Department of du Pont, describes the develop- The resistor glazes are made by ball-milling ment of resistors of this type in which the the glassy component or frit to an average elements consist of conducting glazes con- particle size of 5p, and mixing this with the taining palladium and silver. metal powders. Screen printing preparations The disadvantages are briefly discussed of are made by milling two-thirds of the inor- resistors produced by screen printing carbon- ganic powder with one-third of organic resin dispersions, and of those made by vehicle and controlling the viscosity of the deposition and attenuation of metal films. pastes between 170 and 230 poises. The The failure of early attempts to produce glazes are fired on ceramic substrates at glaze resistors containing particulate oxide 760°C. and oxide-metal mixtures is attributed mainly A range of resistance values can be obtained to poor resistance-temperature relationships by varying the ratio of palladium to silver and critical dependence of resistance values and the concentration of metal powder in the on the concentration of powder conductors glaze, and by attenuating the current path by in the glaze. selection of the screen printed design. The These difficulties, and the current noise author gives data on the resistivity, tempera- which is characteristic of conduction between ture coefficients and current noise of various particles, are claimed to have been overcome palladium-silver compositions. by using glazes containing finely divided P. E. K.

Platinum Metals Rev., 1964, 8, (1) 22 The Early History of the Thermocouple

By L. B. Hunt, M.SC., Ph.D. Johnson Matthey & Co Limited

The name oJ Seebeck is indissolubly associated with the discovery of thermoelectricity and the thermocouple. But how did Seebeck’s researches jit into the general background of contemporary scientijic work, and how did they lead on to the development of the modern thermocouple pyrometer? This article outlines the early history of the couple up to the time when it became accepted as an accurate and reliable means of measu.ring high temperatures.

At this distance from the events, and Among the small band of physicists who looking back from our present understanding were active in this way was Thomas Johann of the nature and applications of electricity, it Seebeck. Born at Reval in Estonia on April is almost impossible to appreciate the excite- gth, 1770, the son of a wealthy merchant, he ment and enthusiasm that prevailed in the left his native town at the age of 17 and took little world of physics around the year 1820. up the study of medicine in Berlin. His strong Until Volta announced the discovery of his inclination for natural science, together with pile in 1800 physicists had had no means of his financial independence, caused him to studying the effects of a steady current. This, change his plan, however, and he embarked on and the voltaic cell which rapidly succeeded it, a career of private research, first in Bayreuth provided for the first time a simple means of and later in Jena. Here he worked on optics maintaining a continuous current, and led to a and on the nature of colour, but in 1810 he burst of activity in the study of the chemical left Jena for Nuremberg, and it was here that effects of electricity. Oersted spent some time as Seebeck‘s guest. In 1818he accepted a position with the Berlin Oersted and Ampbre Academy of Sciences, and moved to that city. As this first wave of enthusiasm was perhaps dying away, Oersted discovered that a current Seebeck’s Discovery of electricity flowing in a wire lying parallel Here it was, of course, that he learned of with, and close to, a magnetic needle had the Oersted’s discovery, and at once applied power of deflecting the needle. This discovery himself to the study of electromagnetism; in was announced in 1820, and it immediately December of the same year, 1820, he read a set off a new wave of interest all over Europe, paper to the Academy dealing with the this time in the mechanical effects of a current. magnetic influence of a current. Only a few Arago, in the same year, produced the first months later, in August 1821, he announced electromagnet, while Ampere, within a week to the Academy (I) his discovcry that two of hearing of Oersted’s experiment, had different metals forming a closed circle, in the shown that one electric current had a magnetic absence of moisture, showed magnetic pro- influence upon another. No doubt was left perties when subjected to a difference of that magnetism was essentially an electrical temperature at the point of contact. He had phenomenon. experimented with a number of combinations

Platinum Metals Rev., 1964, 8, (l), 23-28 23 of metals (finding an antimony-bismuth At this time Ohm was working on the combination the most effective), and had propagation of electricity through a conductor ascertained the effects of both heating and and on the concept of resistance, but he was cooling one of the junctions, He established having trouble in his experiments owing to the that the deflection of the magnetic needle variations in the current obtained from his arose from the difference in temperature of batteries. A suggestion was made to him by the metallic junctions, that the effects varied Poggendorf that he should replace the for different metals, and were greater for batteries by a thermo-electric circuit, and this greater differences of temperature. In his idea Ohm adopted in his classical research in results he reported the movement of the 1826. He used a circuit of bismuth and needle in terms of an easterly or a westerly copper, one junction being immersed in a deflection, and he described the phenomenon steam jacket and the other in ice. as ‘ctherm~-magneti~m”,taking objection in later years to the expression “thermo- The First Measurement of electricity”. High Temperature Second in importance at the time only to The first recorded suggestion to make use of Oersted’s experiment, Seebeck’s discovery Seebeck’s discovery as a means of measuring also spread rapidly among European physi- high temperatures came from A. C.Becquerel cists, and it was repeated in every centre of (2)in a paper read to the Academie Royal des research. Sciences in Paris on March 13th, 1826. His investigations included observations of the Faraday and Ohm needle deflection obtained with a number of Among others, Faraday carried out the combinations of metal wires when one junc- experiment, and he records it in his diary as tion was heated in a spirit lamp, and he follows: deduced that, for certain of these com- Octr. 21, 1822 binations, the intensity of current developed Dr. Seebeck’s Expt. An. Phil. 2V.S. VoE. iv. was proportional to the rise in temperature. p. 318 The most suitable combination, he decided, Bar of Antimony and brass wire; the bur was a circuit consisting of platinum and being heated at one end the north pole of a needle palladium wires. would go round it as represented in the lower Jigure-the effect on the needle very decided, Becquerel further showed that the charac- powerful even and constant. teristics were independent of the diameter of The dotted lines represent the state of the the wire, and also that an impure platinum loire as ascertained from former experiments. wire would give rise to a current if coupled with a pure platinum wire; he pointed out, in fact, the necessity for cleaning the platinum in nitric acid to avoid spurious effects due to contamination. The Magnetic Pyrometer In 1836 Professor C. S. M. Pouillet (3), of Paris, also before the Academie Royal des Sciences, put forward his “magnetic pyrometer” and detailed its construction. This instrument, almost incredible by today’s standards, comprised a platinum wire sealed into the breech of a gun, the wire passing up the barrel but prevented from touching the

Platinum Metals Rev., 1964, 8, (1) 24 sides by a filling of magnesia or asbestos. The breech of the gun was then to be inserted into the hot zone. In the course of his long and classic researches on heat Henri Regnault (4) made use of Pouillet’s iron-platinum couple, but he found such irregularities that he emphatically condemned the whole idea of the thermo- electric method. Regnault’s unhappy experiences were due partly to his use of iron as one element, and also to his failure to employ a high-resistance galvanometer. Later, in 1862,Edmond Becquerel(5) took up the study of his father’s platinum-palladium thermocouple and used it as an intermediary with an air thermometer in determining the melting points of a number of substances. As a result of his researches he succeeded to some extent in rehabilitating the reputation of the thermocouple, and he derived an expression-much too complex-for the relationship between temperature and electromotive force. Henry Le Cliatelier The father of the modern thermocouple, Le Chatelier E.m.f. - Temperature Relationship was the $rst to employ a rhodium-platinum alloy Avenarius (6) also investigated this rela- against platinum and to recommend calibration in terms of the fired points of melting or boiling tionship, working with the rather curious of pure substances combinations of steel and nickel-silver, and copper and zinc. He arrived at a parabolic first reference to the use of iridium-platinum formula of the type: alloys in thermocouples. E =a +bt +cte The Work of Le Chatelier but a few years later apparently realised his And so to 1885 and Henry Le Chatelier (8), error and withdrew his arguments. whose name, together with that of Seebeck, In the meantime Professor P. G.Tait (7) of will always be associated with the thermo- Edinburgh University had conducted a series electric pyrometer and with the use of a of experiments in an attempt to construct rhodium-platinum alloy. Born in Paris in “thermo-electric diagrams”, and concluded 1850, Le Chatelier studied chemistry under that the electromotive force is in general a Sainte-Claire Deville, but in 1870 was called parabolic function of the absolute tempera- into the army and took part in the siege of ture, He also reported that a very small Paris. Subsequently he practised as a mining amount of impurity, or even of permanent engineer, but in 1877 joined the French strain, is capable of considerably altering the School of Mines to teach chemistry, becoming line of a metal in the diagram. Professor of Industrial Chemistry some nine Professor Tait used “platinum-iridium years later. Many years afterwards Le alloys containing respectively 5, 10 and 15 Chatelier recalled the origins of his work on per cent of the latter metal. These were thermocouples in the following words: prepared for me from pure metals by Messrs. “In 1885, when I attacked the problem of Johnson and Matthey”. This constitutes the the measurement of high temperatures, it is

Platinum Metals Rev., 1964, 8, (1) 25 fair to say there existed nothing definite organisation of the United States Geological available on this important question; we Survey with the objective of studying the possessed only qualitative observations for temperatures above 500°C. Engaged at that physical constants of rocks. Barus was put in time in industrial studies relative to the charge of this laboratory, which was located manufacture of cement, I sought a method first in New Haven, Connecticut, but moved which above all would be rapid and simple, and decided on the use of thermo-electric couples, to Washington two years later. He realised intending to determine the order of magnitude that few important steps in the study of this of the sources of error noticed by Regnault. branch of geology could be made until The readings of even a crude galvanometer might be very useful in technical work, methods for the accurate measurement of high provided the limitations of its accuracy were temperatures and pressures had not only been appreciated. I soon recognised that the errors perfected but rendered easily available, and he attributed to this method could easily be eliminated by discarding in the construction therefore undertook a most comprehensive of the couples certain metals, such as iron, study of temperature measurement; this was nickel, and palladium, which give rise to singular anomalies. Among the different metals published as a memoir of some 300 pages in and alloys studied, pure platinum and the alloy 1889. Numerous alloys of platinum were of platinum and rhodium which are still used investigated as thermocouple elements, most today, gave the most satisfactory results. . . . I recommended also the calibration of the of them having additions of 2, 5 and 10 per couples, not against the air thermometer cent of other elements, but he came to rest directly, as Becquerel had tried to do, but in for general use on platinum against 20 per terms of the fixed points of boiling or fusion of certain pure substances, in such a way that, cent iridium-platinum. Barus was aware of when these temperatures should be known the importance of purity and homogeneity in more exactly, as is the case since my earlier researches, the results could be corrected with certainty.” Le Chatelier devoted considerable time and effort to the development of the thermocouple pyrometer, and arranged for the instrument to be manufactured by Carpentier, the suc- cessor of the famous Ruhmkorff, at 20 Rue Delambre, Paris. The reputation of these instruments spread rapidly and widely. In 1890, for instance, the great American metallurgist, Professor H. M. Howe wrote (9): “Thanks to the labors of M. Le Chatelier, we have at last a pyrorneter capable of measuring easily, accurately and rapidly extremely high temperatures, indeed, those approaching the melting point of platinum. And this is not an apparatus which each must construct for himself; it is for sale ready made. Indeed, it is SO far simplified that it has actually entered into practical use for the control of high temperatures in steel works, glass works and gas works.” American Investigations Sir William Roberts-Austen Contemporary with Le Chatelier but quite Chemist to the Royal Mint, Professor of Metallurgy independently of him, Dr. Carl Barus (10) at the Royal School of Mines and Jirst investigator was engaging himself actively in the measure- to the Alloys Research Committee of the Institution ment of high temperatures. In 1882 a new of Mechanical Engineers, Roberts-Austen was quick to appreciate the usefulness of Le Chatelier’s physical laboratory had been set up within the thermocouple pyrometer

Platinum Metals Rev., 1964, 8, (1) 26 his thermocouple materials, and he drew upon the resources of the Bishop and Co. Platinum Works at Malvern, Pa., for the preparation of his materials. The couples were then cali- brated at the boiling points of mercury, zinc and certain organic substances.

Roberts-Austen’s Contribution At this time the Institution of Mechanical Engineers had established its Alloys Research Committee, with the initial objective of studying the effects of alloying elements on the properties of metals, and had appointed W. C. Roberts-Austen, who combined the posts of Chemist to the Royal Mint and Professor of Metallurgy at the Royal School of Mines, as investigator. In his first report (11) in 1890 to the Institution, Professor (later Sir William) Roberts-Austen said: “In the present investigation it is necessary to measure much higher temperatures; and fortunately an accurate method is at hand. Edward Matthey Early in 1889 I had occasion to employ the pyrometer devised by M. H. Le Chatelier, and Investigated the homogeneity of rhodium-platinum was satisfied as to its being extremely trust- alloys and conjirmed the $findings of Le ChateEier worthy and convenient up to temperatures and Roberts-Austen as to their suitability for over 10oo”C. or 18oo”Fahr. The instrument thermometric use in fact enabled me to confirm the fundamental observations of M. Osmond respecting the critid points of iron and steel, and to dernon- “It is asserted that even long wires of the strate the results in a lecture delivered before platinum-rhodium alloy are homogeneous, the members of the British Association in and therefore do not give rise to subsidiary September 1889.” currents which would disturb the effect of the main current produccd by hcating the junction; Since 1875 Roberts-Austen had interested but very careful experiments to determine himself in the problems of liquation or whether this is the case have yet to be made.” segregation of the constituents of alloys, and This uncertainty aroused the interest of had been most painstaking in his measurement Edward Matthey, who carried out a lengthy of temperatures using the laborious calori- investigation (12) on the liquation of alloys of metric methods then available. He therefore the platinum metals. On the rhodium- welcomed most readily the new type of platinum alloys he had the following comment instrument and proceeded to adapt it for the to make: production of autographic records of the cooling and solidification of molten metals “Much attention has lately been drawn to an and alloys. alloy of pure platinum, with 10 per cent of rhodium, which has become important from the excellent service it has rendered in the Problems of Homogeneity determination of high temperatures. The alloy of platinum with ID per cent of rhodium is Some doubt still remained, however, con- used with pure platinum as a thermocouple, cerning the absolute reliability of the rhodium- and it is, therefore, interesting to be able to set at rest any doubt which might arise as to platinum alloy, and Roberts-Austen referred this alloy being uniform in composition when to this: melted and drawn into wire.”

Platinum Metals Rev., 1964, 8, (1) 27 Matthey prepared a melt of one and a half alloys of platinum and rhodium, lately published in the ‘Philosophical Transactions’, kilograms of 10 per cent rhodium-platinum, settled the question in favour of the rhodium- which he cast into a sphere of two inches platinum thermo-junction, for I was satisfied diameter. The sphere was then sectioned, and that the alloy of platinum with 10 per cent of rhodium is as homogeneous as any known samples were taken for analysis from a number alloy could well be, and is therefore admirably of locations between the surface and the adapted for use as a thermo-junction, pure centre. The maximum difference between the platinum being the opposing metal.” centre and the outside was found to be 0.6 The instrument employed was obtained per cent of platinum and 0.04 per cent of from Paris and was used in the full scale rhodium. He concluded: operation of the oxidation process in the “This result proves that the alloy is not works of Johnson Matthey, being one of the subject to liquation, and fully justifies the first such pyrometers to be used in industry. high opinion that H. Le Chatelier and Roberts- Austen have formed as to its suitability for The only earlier record of a Le Chatelier thermometric measurements.” pyrometer being used in this country is that At much the same time, 1892, Edward given by Sir Robert Hadfield (rq), who Matthey was concerning himself with the purchased one from Carpentier in April 1890. extraction and refining of bismuth and he This gave excellent service in the steel works contributed a series of papers on this subject of Hadfields Ltd. in Sheffield until British- to the Royal Society. An extract from one of made pyrometers became available just after these papers (13)~dealing with the tempera- the turn of the century. ture at which arsenic can be oxidised off from The first mention of thermocouples being bismuth, reads as follows: available for temperature measurement in “The work of Roberts-Austen has shown this country appears in a Cambridge that a thermo-junction is practically the only Instrument Company catalogue of 1898, but form of pyrometer that can be used for delicate the manufacture them did not begin until thermal investigations of rhis kind, but the of question arose which particular thermo- 1902, when special stocks of platinum and junction should be adopted. Was it well to rhodium-platinum were procured from use the platinum-iridium one as advocated by Barus, or the platinum-rhodium one suggested Johnson Matthey and marketed in suitable by H. Le Chatelier? My previous work on the porcelain tubes (15).

References I T. J. Seebeck . . *. Abhandlurgen der physikalische Klasse dcr Komglichen Akudemie dm Wissenschafrer ZU Berlin, 1822-23, pp. 265-373 2 A. C. Becquerel . . __ Ann. chim. phys., 1827, 31, pp. 371-392 3 c. s. AM. l’ouillet , . . . Compt. rend., 1836, 3, pp. 782-790 4 H. V. Regnault . . .. Relation des Experiences, Paris, 1847, I, p. 246 S E. Becquerel . . .. Compt. rend., 1862, 55, p. 826 6 Avenarius ...... Pogg. Ann, 1863, 119, p. 406 1864, 122, p. 193 7 P. G. Tait ...... Trans. Roy. SOC.Edin., 1872-73, 27, p. 125 8 1-1. Le Chatelier . . .. Compt. rend., 1886, 102, p. 819 J. de Phys., 1887, 6, p. 23 9 13. M. Howe . . .. ling. and Min. J., 189, 50, p. 426 I0 C. Barus ...... 13~11.U.S. Geol. Surwey, 1889, h’o. 54 Phil. Mug., 1892, 34, p. 376 I1 W. C. Roberts-Austen . . 1st Rep. Alloys Hes. Comm., Inst. Mech. Big., 1891 I2 E. Matthey .. .. I’hiZ. Trans., 1892, 183, p. 629 I3 B. Matthey .. .. I’roc. Roy. Soc., 1893, 52, p. 467 14 Sir Robert Hadfield , . Trans. Fareday SOL.,1917-18, 13, p, 208 IS Anon ...... Eyineeriqq, 1945, 159,iMay 11th and zgth, pp. 361 and 41

Platinum Metals Rev., 1964, 8, (1) 28 ABSTRACTS of current literature on the platinum metals and their alloys

PROPERTIES 605fS'C. The effects of variations in the treatment of the alloys were also studied. Fe-Pd alloys On the Thermal Expansion Coefficient and behaved similarly to Ni-Fe alloys after quenching, the Temperature Coefficient of Young's particularly by showing an Invar effect. Modulus of the Alloys of Iron and Platinum H. MASUMOTO and T. KOBAYASHI, J. Japan Inst. The Iron-Palladium Phase Diagram below Met., 1963, 27, (91, 459-460 950°C 53.5 to 57.0'6 Pt alloys with Fe have negative E. RAUB, H. BEESKOW and 0. LOEBICH, Z. Metall- coefficients of thermal expansion with a maximum kfmde, 1963, 54, (101, 549-552 value of - 23.68 x IO-~for 56Oh Pt alloy at o to The results of room- and high-temperature X-ray 40°C. 51 to 69:/, Pt alloys have positive tempera- analysis of the Fe-Pd system are tabulated. No ture coefficients of Young's modulus with a y/u eutectoid was observed but a miscibility gap maximum value of +161.5xro-~ for 5596 Pt extended from 7 to 23 at.X Fe, with a critical alloy in the same temperature range. A vertical temperature of about 900°C. A eutectoid between dilatometer and a vibrator-controlled oscillator u-Fe and the yl phase was confirmed at about with an electrostatic transducer were used for the 62ooC, 43 at.(:; Fe. The two ordering phases y1 respective measurements. and yz were separated by a narrow two-phase region below a eutectoid area. yz has a higher Theory of Hydrogen Adsorption on Platinum critical temperature than yl. (8zofzo"C and T. TOYA, 3. Res. Inst. Catalysis Hokkaido Univ., 790 &2Ooc). 1962, 10, (3), 236-260 There are two types of H adsorption on Pt. On the Structural Form of Copper-Palladium r-type adsorption increases the work function and Solid Solutions near the Composition Cu,Pd electrical resistance of clean Pt while s-type A. A. PRESNYAKOV, L. I. DAUTOVA and E. A. adsorption decreases them. The total effect on DZHANBUSINOV, Fiz. Met. Metalloved., 1963, 16, Pt is complex as the difference between the two (I), 61-64 heats of adsorption is only 0.4 kcal/mol whereas 28.8 at.:6 Pd-Cu alloy samples were tempered at for Ni, which behaves similarly, it is ro kcal/mol. 750"C, annealed and studied by X-ray methods. The positions and entropies of the two types of The crystal lattice structure is discussed. At high adsorption are discussed. The broad, intense temperatures a structure was detected which infra-red band at 4.8611 in the spectrum of H on Pt indicated the existence of a new phase but its is caused by s-type adsorption and the sharp, lattice parameters were not determined. weak band at 4.74~by r-type adsorption. Properties of Palladium-Rhenium Alloys Ultra-pure Hydrogen by Diffusion through M. A. TYLKINA and I. A. TSYGANOVA, Zh. hreorg. Palladium Alloys Khim., 1963, 8, (IO), 2346-2350. J. B. HUNTER, Abs. Papers, 145th Meeting, Am. Additions of up to 11.9 wt.76 Re increased the Chem. SOC.,1963, 12s-130 hardness and strength and the specific electrical The development of a stable, highly permeable resistance of Pd but decreased the temperature alloy of Pd has led to its industrial application coefficient of resistance. Re also raised the in a variety of diffusion units. H, transfer rate initial temperature of reaystallisation of Pd; by data are presented for wide ranges of pressure and up to 600°C in the case of 8.74 wt.?,, Re-Pd. temperature. Annealing studies on Re-Pd showed that higher temperatures reduced the hardness and strength Magnetic and Dilatometric Measurements on but improved the elastic properties of the alloys. the Transformation Kinetics of the Iron- Electrical properties were unaffected by the Palladium Alloys annealing treatments. A. KUSSMAN and K. JESSEN, z.Metallkunde, 1963, 541 (911 504-510 The Temperature Dependence of the Mag- Studies on y-a transformations and yl-yz ordering netic Susceptibility of Some Palladium Alloys confirmed the existence of a miscibility gap w. KOSTER, D. HAGMANN and R. LUECK, Ann. between 10 and 25 at.% Pd-Fe alloys at high Phy~dk,1963, 11, (1-6), 52-58 temperatures. The critical temperature curve for Temperature-reciprocal paramagnetic suscepti- yI-y2 ordering showed a maximum at 8oo"C, bility curves for Pd alloys with up to 4 at.:/, Ru 65 at. :k Pd. A eutectoid occurred at 46 at. :d Pd, or 10at. a& Cd, In or Rh between 120 and I 100°K

Platinum Metals Rev., 1964, 8, (l), 29-37 29 agreed with the transport aspects of Stoner's tion temperature between 01- and P-Bi,Rh theory of paramagnetism. Various explanations occurred at 430°C. Peritectic temperatures were of the electron specific heat-susceptibllity dis- 456°C and 775°C. crepancy in Pd showed that no one definition of Stoner's parameters OF, q, O1 was possible. An On the Superconductivity of Ti- and Zr-Rh approximation using the reciprocal temperature Alloys coefficient of the reciprocal susceptibility ex- c. J. RAUB and c. A. ANDERSEN, Z. Physik, 1963,175, plained vacancy variations in the alloys but (11, 10.5-114 this quantity was not the same as the Curie The connection between superconductivity and constant unless T+OO,. The rigid band model valence electron concentration was studied in Ti- gave a partial interpretation of the results. and Zr-alloys containing up to IZ at.% Rh. In both the h.c.p. or-phases and b.c.c. 8-phases the Hysteresis in the Palladiuni-Hydrogen Sys- transformation temperature increased with valence tem electron concentration. By extrapolation to 0% N. A. SCHOLTUS and w. K. HALL, J. Chern. Phys., Rh it was shown that the transformation tem- 1963, 391 (412 868-870 peratures for Ti and Zr p-phases lie below those of the a-phases. When H, dissolves in Pd, the GL phase forms first. The phase nucleates and grows in this causing plastic deformation of the M phase beyond its The Structure of the Stable Tantalnm- elastic limit because of the large M+P volume Ruthenium Alloys change. Absorption is accompanied by ,!3 phase E. RAUB, H. BEESKOW and w. FRITZSCHE, Z. growth under a phase compression but desorption Metallkunde, 1963, 54, (81, 451-454 takes place from the /3 phase of the plastically The Ru-Ta system is similar to the Ru-V and deformed solid. This is not reversible and Ru-Nb systems. X-ray and microscopic analysis hysteresis occurs. Earlier theories are discussed. of alloys with increasing Ru content showed the The absorption loop of the isotherm is calculated occurrence of transitions from the b.c.c. solid from the desorption loop and the yield strength of solution to an ordered Bz structure, followed by Pd. tetragonal distortion, an ordered orthorhombic form and an ordered f.c. tetragonal form at high Role of Hydrogen Atoms in Palladium Ru content. T. TSUCHIDA, J. Phys. Soc. Japan, 1963, 18, (7), 1016-1019 Melting Points of LaRu,, CeRu, and PrRu, Ag-Pd alloys are similar in electronic structure to R. D. REISWIG and K. A. GSCHNEIDNER, J. Less- Pd but do not have the narrow regions of H, Common Metals, 1963, 5, (51, 432-433 content stability of pure Pd. Measurements of The observed melting points of the Laves phases the Hall coefficient and magnetic susceptibility were: LaRu, at 1431*3o'C, CeRu, at 1539& of hydrides of 10 to 30 at.% Ag-Pd alloys were 30T, PrRu, at 1681+15OC. Melting was made as functions of Hzcontent. As Ha was probably incongruent. occluded the paramagnetic susceptibility decreased and the Hall coefficicnt decreased con- Vapour Pressures of Platinum Metals. 111. siderably in the high H, contcnt p-phase alloys. Iridium and Ruthenium The H atom was considered to act like a proton in the Pt lattice by transferring a valence electron R. c. PAULE and J. L. MARGRAVE, J. Phys. Chem., to the Pd 4d band. 1963967, (9XI896-1897 Vapour pressures, heats of sublimation and The Phase Diagram Rh-Sb boiling points were obtained for solid Ir and Ru. R. N. KUZ'MIN and N. N. ZHURAVLEV, Vestnik Mosk. Ir has log P& (solid)=10.1zo-~3,680/T for Univ., Ser. ZII, Fiz. Astron., 1963, 18, (z), 9-14 2140°K < T < ~477°K~AH,&= 160.9 32.8 kcall X-ray and micrographic studies of the Rh-Sb mole and estimated normal b.p. -4850&10o"K. system indicated the existence of a y-phase at Ru has log Pzz (solid)= 11.200-33,600/T for about the Rh,Sba composition and a eutectic 201IoK

Platinum Metals Rev., 1964, 8, (1) 30 Laves phase ZrRu,, which decomposes eutectoid- Superconductivity of Some New Pt-Metal ally at 1300t20"C. Ru has low solubility in Compounds a-Ti and cc-Zr but more in /%Ti and P-Zr. Ru c. J. RAUB, w. H. ZACHARIASEN, T. H. GEBALLE and stabilises P-Ti and can depress the fl fcc trans- B. T. MATTHIAS, J. Phys. Chem. Solids., 1963, 24, formation below room temperature for 16-20 at. :; (9)i 1093-1 100 Ru. Considerable precipitation hardening occurs Tests on compounds of the Pt metals with P, As, in Ti-rich alloys. Study of the P+cc transforma- Sb, Bi and S revealed several previously unknown tion in Zr-Ru alloys is made difficult by the superconductors in the systems Rh-As, Pd-P and oxidation of Zr, causing the alloys to appear Pd-As. Transition temperatures and crystallo- ternary. graphic data are listed for all known superconductors of the Pt metals with elements of groups A Study of the Adsorption and Decomposi- IVA, VA, and VIA. tion of Hydrocarbons on Clean Iridium Surfaces The Problem of Producing Superconducting R. W. ROBERTS, J. PhYS. CheWZ., 1963, 67, (IO), Materials 8 2035-203 E. M. SAVITSKII and V. V. BARON, Metallurgiyu i CH,, CzHs and C,H, were admitted to Pyrex Gmnoe Delo, 1963, (9,3-12 flasks on the inner surfaces of which Ir films had Among a large number of superconducting been prepared under ultra-high vacuum con- substances reviewed there are more than thirty ditions. They were adsorbed on the clean films compounds of the Pt metals, for which the at 27 and CH, did not decompose on Ir 100°C. transition temperatures and Curie points are but C,H, decomposed to CH, and H, at 27% and tabulated. Their formation, properties and uses to CH, at 100°C. 0, preadsorbed on Ir inhibited are discussed. (24 references.) C,H, decomposition. CaHl was self-hydrogenated to C,H6 and CHI. The roughness factor of the Ir films was about 7. High Temperature Properties of Refractory CHEMICAL COMPOUNDS Alloys Structure and Reactivity of the Oxyanions of E. J. RAPPERPORT and A. L. GEARY, U.S.A.E.C. Repmt N.M.I.-I~~~,67 pp. Transition Metals A. CARRINGTON and M. C. R. SYMONS, Chem. Rev., Ten alloy systems of the refractory metals Mo, W, Nb and Ta with additions of Hf, Re, Os, Ru, 19633 63, (9,443-460 Rh, Ir and Zr were studied by hot hardness tests Known oxyanions of the Pt metals are RuO,, up to 1200°C and oxidation resistance tests at RuO,, RuOZ- and OsO, and Rho% is also IOOO"Cto determine their suitability for use in believed to exist. The electronic structure and nuclear reactors. Ru, 0s and Re additions reactions of these tetrahedral oxyanions, together increased the hardness by up to 7 times at room with those of V, Nb, Ta, Cr, Mo, W, Mn, Ti, Re temperature and up to 17 times compared to the and Fe are critically discussed. (115 references,) unalloyed refractory metals. These alloys were then tested for tensile strength up to 180orC. Unusual Oxidation States of the Noble 2-10 at.% Ru-W,2-1oat.% Ru-Taand2-1oat.O; Elements Re-Nb showed greater strength worth further N. BARTLETT, Chem. in CUnadU, 1963,I5, (8),33-40 study. The structure and properties of the fluorides, Oxidation of the Platinum-Group MetaIs oxides and oxyfluorides of the Pt metals are c. A. KRIER and R. I. JAFFEE, J. Less-Common discusscd. Particular note is made of OsOF, and PtO,F,. Work on the [PtF,]- ion led to the Metals, 19633 53 411-431 (5)~ discovery of the noble gas fluoride compounds, Samples of the six Pt-group metals were heated the bonding of which is also reviewed. (38 in a slowly moving stream of dry air at IOOO- references.) 14oo0C and all showed linear weight loss rates in the range 1200-1400@C by the formation of volatile oxides. Rh gained weight at ro0o"C by Aroniaticity of Five-Membered Rings Con- the formation of an oxide film and Pd initially taining Platinum (11) absorbed 0,. At 14oo0C in a slow air stream at P. HAAKE and P. A. CRONIN, Imrg. Chem., 1963, 2, I atm. the rates of weight loss were 6.8~10-~(4)J 879-880 mg/cm2/hrfor Rh, 9.6 x IO-~ for Pt, 3.1 for Ir, Conducmmetric studies on the kinetics of 1.2 x 102 for Ru and 1.2 x 10-3 for 0s. The rates displacement of chloride by dithiooxamide in depended on the gas flow rate, the partial pressure methanol indicated that a strong trans effect of Oeand the total gas pressure. The kinetics of resulting from aromaticity in the Pt- cc-diimine the reactions also depended on the temperature. chelate ring caused the Pt(bipy)Cl, complex to The mechanism of reaction of the 0, at the react marly one hundred times faster than the surface is discussed and specimens are illustrated. cis-?t(py),Cl, or Pt(en)Cl, complexes.

Platinum Metals Rev., 1964, 8, (1) 31 Aromatic Complexes of Metals. LXXV. ELECTROCHEMISTRY C yclopentadienyl-palladium-~kosyl E. 0. FISCHER and A. vOGLER, 2. Naturf., 1963,186, Relation between the Activation Energy of (9)Y 771-772 Some Electrochemical and Catalytic Pro- C,H,PdNO was prepared by the reaction of cesses and the Properties of Metals PdNOCl with NaC,H, in pentane with air v. v. DEMCHENKO, Zh. Fiz. Khim., 1963, 37, (8), excluded and its absorption spectrum was studied. 1718-1725 Complexes between Palladium and Electrochemical and catalytic processes such as (II) hydrogenation, isotope exchange and Ha for- Pyridine-2-aldoxime mation at the electrode, where metal-H, bonding c. F. LIU and C. H. LIU, Znorg. Chem., 1963, 2, occurs as an intermediate step, are reviewed. (419706-707 Activation energy is proportional to the increase The tetrachloropalladate (11), chloride and nitrate in surface tension of the cathode or catalyst of monohydrogen bis-(pyridine-2-a1doxime)-pal- surface. Graphs illustrate this effect for Pt, Pd, ladium (11) ion were prepared and compared to Rh, Ir, 0s and a number of base metals. (16 similar Pt(I1) and Cu(I1) compounds. The inner references). complex between pyridine-2-aldoxime and Pd(I1) was also prepared. In contrast to the Pt(I1) inner On the Thermodynamics of Platinum Oxide complex only the tram isomer of Pd(I1) was Electrodes obtained. B. NOVAK and T. MARKOVIC, Monat. Chenz., 1963, Preparation and Properties of Anhydrous 94, (31, 607-420 Rhodium (11) Acetate and Some Adducts The energy of formation of each oxide, hydroxide and oxide hydrate of Pt in aqueous solution was Thereof determined. Data are fully tabulated and S. A. JOHNSON, H. R. HUNT and H. M. NEUMANN, extensively discussed. Inorg. Chem., 1963, 2, (5), 960-962 Rh(I1) acetate was prepared by the reaction of Electrochemical Oxygen- and Hydrogen- Rh(OH),. H,O with glacial acetic acid. Stable Chemisorption on Smooth and Rough Plati- adducts with 2 ligands to each dimer were formed with H,O, tetrahydrofuran, acetonitrile, dimethyl num sulphoxide, dimethyl sulphide, trimethylamine, H. DIETZ and H. GOHR, Z. phys. Chem., Leapzig, NH,, NO2 and ethylenediamine. The properties 1963, 223, (I12), 113-131 and spectra of Rh(I1) acetate and these adducts The chemisorbed H, or 0, films which can be are discussed. formed on Pt electrodes are produced continuously only if the electrodes are continually Ruthenium Tetrafluoride cleaned. Film formation then becomes a steady J. H. HOLLOWAY and R. D. PEACOCK,J. Chem. SOL, process whereas if the films are allowed to build 1963, !July), 3892-3893 up the potential of the circuit alters and the RuF, was formed by the reaction of I, with RuF, current is reduced. dissolved in IF,. RuF, was highly reactive and reacts violently with H,O to form RuO,. The The Mechanism of Electrochemical Oxida- magnetic moments of RuFI were measured in the tion of Carbon Monoxide and Methanol on range 90 to 300°K. X-ray diffraction analysis Platinum. I. Carbon Monoxide Adsorption suggested that it has a simple structure. and Desorption and SimuItaneous Oxidation of the Platinum Surface at Constant Poten- Anhydrous Ruthenium Chlorides tial J. M. FLETCHER, W. E. GARDNER, E. W. HOOFER, K. R. s. GILMAN, J. Phys. Chem., 1963, 67, (9), 1898- HYDE, F. H. MOORE and J. L. WOODHEAD, Nature, 190.5 1963,199, (4898), 1089-1090 The oxidation of the surface of a Pt electrode The preparation and magnetic properties of covered with adsorbed CO in a saturated solution r-RuCl,, 6-RuCI, and Ru,OCl, are described. in I NHClO, of a IOO?;, or I”: CO gas mixture is a-RuC1, formed by heating Ru and C1, in siliceous rate-controlled by the removal of adsorbed CO, vessels at 600°C contains some Ru,OCI, which is as shown by current-time traces after raising the diamagnetic and reduces the susceptibility. electrode potential from 0.4 to >0.8V. a-RuC1, has now been made by heating fi-RuCI, above the &transition temperature of 450°C. F-RuCl, is formed by reacting Ru in CO and C1, 11. Interpretation of the CO and “Oxygen” at 360-39ooC, with some u-RuC1, formation. Adsorption Data Obtained at Constant Ru,OCI, occurs as a volatile oxide chloride in the Potential mixture of aquochloro complexes known as S. GILMAN, Abs. Papers, 145th Meeting, Am. Chem. commercial RuCl,. u-RuCI, is antiferromagnetic SOC.,1963, 22B-49 and /?-RuCI, has low magnetic susceptibility. The oxidation of both the adsorbed CO and the

Platinum Metals Rev., 1964, 8, (1) 32 Pt surface are interpreted by a reactant-pair and medium temperature heating stages incor- mechanism which assumes that the actual porate solid Pt in the grid holders and caps because reactants are adjacent CO and H,O adsorption of its high melting point, low electrical resistance molecules. and resistance to oxidation. The Pt carries the current to heat the grid in the high temperature Mechanism of Anodic Oxidation of Organic stage. Pt wire forms a resistance furnace in the Compounds on Platinum other case. v. s. BAGOTZKY, Abs. Papers, 145th Meeting, Am. Chem SOC.,1963, 2rK-58 The Durability of Pt-Rh Alloys in Analytical The adsorption of reacting species on the Pt Apparatus surface affects the rate of oxidation. Adsorption G. REINACHER, Werkszoffe u. Korrosion, 1963, 14, of 0%on the surface reduces the reaction rate (7), 574-579 exponentially. Five typical analytical laboratory operations tested Rh-Pt alloys to show that they can well be The Normal Oxygen Potential on Bright used in corrosive conditions up to IOOOT.A Platinum tarnish developed during annealing in air and J. P. HOARE,?. Electrochem. SOL.,1963, 110, (9), during soda-potash fusions but weight constancy 10I 9-102 I was of the same order as ordinary apparatus Pt. Experiments showed that when a Pt surface was Resistance to acids was as good as with Pt. completely covered by a chemisorbed monolayer Weight losses during potassium bisulphate fusions of 0, the surface was passivated with respect to increased with temperature but, by lining with Pt, its reaction with oxygen. The layer was produced the crucibles combined resistance to attack with by passivating Pt beads in HNO,. A rest potential resistance to deformation. zo?/o Rh-Pt has been of 1225310 mV was observed for up to 24 hours used for ferrophosphorus and ferrosilicon extrac- or more. tions and roo4 Rh-Pt for HF and H,S04 evapora- tions and various fusions. The dark Rh oxide Electrochemical Corrosion of Iridium in tarnish could be avoided by Pt linings. Lined and Hydrochloric Acid Solutions unlined 1o,20 and 30 wt.;;:) Rh-Pt alloy apparatus J. LLOPIS and L. JORGE, J.Electrochem. soc., 1963, is therefore profitable for many high temperature analyses. 11% (9X 947-951 Electrolysis of Ir in HCI showed that in all respects Ir is more corrosion resistant than k. BRAZING Ir was attacked more strongly as the HCl concentration and temperature increased. An a.c. New Era Brazing Turns to Filler Metals with superimposed on the d.c. increased the effect, Palladium especially at low frequencies, but the corrosion A.s. CROSS and J. B. ADAMEC, J., 1963, 42, was negligible until a threshold value of ia.c.Sid.c. Welding was reached. Dissolution and passivation mechan- (8), 645-649 isms were studied. Pd-containing brazing alloys are ductile with good wetting and flowing properties and do not erode the base metals with which they are used. LABORATORY APPARATUS Sixteen alloys are described of which five Ag-Pd and Ag-Cu-Pd alloys cover the melting range AND TECHNIQUE 1490°C to 2250T which is so important for electronic equipment. The alloys possess Purity of Hydrogen Permeating through Pd, considerable strength at high temperature. They Pil-25?/, Ag, and Ni also fill gaps satisfactorily and reduce stress J. R. YOUNG, Rev. Scient. Instrum., 1963, 34, (8), cracking. Their low vapour pressure makes them 891-892 suitable for vacuum applications. (13 references). A technique for determining the purity ofdiffused H, is described. Impurities in H, permeating Ceramic-to-Metal Seals for High Tempera- through Pd or 25:< Ag-Pd alloy tubes were a ture Operation few parts in 10~". The lowest impurity content E. L. BRUNDIGE and G. s. HANKS, U.S.A.E.C. Report from Ni was found to be I part in ro6. CO was LAMS 2~17,1963, 36 pp. the main impurity in all cases. A nuclear fuel element design required Nb components to be separated by Al,O,-Y,O,, a ceramic Specimen Heating with Temperature capable of withstanding Cs vapour at 1500°C. Measurement from -1150°C to 2200°C inside Bonding the ceramic to Nb was by metallking the EM6 Electron Microscope with W and then brazing with Pd applied by J. A. HEADLEY and J. MCGEAGH, J. Scient. Instrim., plating the W surface or as Pd foil before joining 1963, 40, (I0)i 484-486 and sealing. Seals for use to rooo°C can be made Equipment is described to enable the temperature from Co-Pd braze alloy added to the joint as foil range -150 to zzoo"C to be studied. The high or by plating with alternate layers of Co and Pd.

Platinum Metals Rev., 1964, 8, (1) 33 CATALYSIS 7moC further increased the activation energy for tests in the ~oG-~oo'Crange, An interpretation of the results relates the condition of the active part Active Centres of Platinum Adsorption of the Pt surface to the thermodynamic equilibrium Catalyst in the Oxidation of Ammonia of the system. v. I. SHEKHOBALOVA and N. I. KOBOZEV, Zh. Fiz. Khim., 1963, 37, (9), 2131-2132 Stereoisomeric 3/3,17/3-Dihydroxyandrostan- Analysis of the reaction products of oxidation of 16-ylacetic Acids NH, over Pt/SiO, showed that the Pt exists in P. KURATH, W. COLE, J. TADANIER, M. FREIFELDER, monoatomic form on the catalyst support. G. R. STONE and E. v. SCHUBER, J. org. Chem., Theoretical relations were established such that 1963~28,(9), 2189-2194 for SiO, with 300 m*/g and temperatures of 300 and 35o"C, the apparent activation energy could The conditions of hydrogenation of 38, 178- be calculated as 10,700 cal/mol. diacetoxy-5-androsten-16-ylidenacetic acid over PtO, were adjusted to improve the yield of 38, 17~-diacetoxy-5-androsten-16/3-ylaceticacid and The Relation of the Catalytic Properties of the stercochemistry of the four possible tetra- Platinised Silica Gel to the Conditions of its hydro products was derived. Preparation M. D. ADAMENKOVA and 0. M. POLTORAK, vat. The Hydrogenation of the Stereoisomers of Mosk. Univ., Ser. II, Khim., 1963, (5), 12-16 Piperylene on Metallic Catalysts The catalytic activity of Pt/SiO, prepared by the L. KH. FREIDLIN and E. F. LITYIN, Neftekhimiya, adsorption of ammines of Pt on SiO, depends on 19631 3j (319 326-329 the SiO, grain size, the drying temperature and The hydrogenation of the cis and trans isomers of the conditions of ammine reduction by H,. piperylene (I, 3-pentadiene) in methanol using SO, should be homogeneous but too much grind- Pt and Pd blacks and Raney Ni as catalysts yields ing reduces the activity. Pt ammine adsorption is all the three possible pentenes in amounts which uniform after 3 days but not after 5 minutes. differ substantially from one system to another. Drying at 100°C gives higher activity than drying Very little pentane is formed in the early stages of at 60°C. 300°C is the optimum reduction the reaction over Pd and Ni, but pentane con- temperature for the Pt ammines. Tests on three stitutes about 35% of the initial products formed series of differently prepared samples showed that over Pt. No isomerisation between the piperylene the level of activity for 0.5 to 3.15:6 Pt contents isomers was detected. is higher than for 3.5 to 7% Pt. Selectivity and Stereospecificity in the Hy- The Hydrogenation of Diene Hydrocarbons drogentdtion of Acetylene Hydrocarbons on over a Platinum Catalyst in Liquid Phase Metal Catalysts L. KH. FREIDLIN and E. F. LITVIN, Izv. Akad. Nauk L. m. FRIEDLIN and YU. w. KAUP, Doklady Akad. S.S.S.R., Ser. Khim., 1963, (7), 1307-1312 Nauk S.S.S.R., 1963, 152, (6), 1383-1386 Saturated hydrocarbons were formed by the Tests on pentyne-2 and pentyne-r, using Raney hydrogenation of dienes over Pt black catalyst at Ni, Raney Co, Pd black, Pt black, and Rh black conversions of up to 70 to 80%, mainly by two catalysts established that the order of selectivity H, molecules becoming attached to each diene in the first stage of hydrogenation is Pd>R> molecule without desorption of intermediate Rh>Ni>Co and that the order of stereospecifi- products from the catalyst surface. Tests with city is Pd> Ni=Co>Pt>Rh. Isomerisation of isoprene, tram-piperylene and 2, 3-dimethyl- the products was also studied. butadiene-I, 3 and binary mixtures of a-olefins with isoprene showed that, at high degrees of Hydrogenation of 2-Methyl-5-acetylfurans diene conversion, some saturated hydrocarbon in Liquid Phase was derived from the a-olefins, simultaneous hydrogenation having taken place. There was N. I. SHUIKIN, I. F. BEL'SKII, G. K. VASILEVSKAYA high diene selectivity with alkaline solution but and Y. M. sHOSTAKOVSKII, Izv. Akad. Nauk poor selectivity with acid solutions. S.S.S.R., Ser. Khim., 1963, (8), 1475-1478 When 2-methyl-5-acetylfuran was hydrogenated Kinetics of Dehydration of Cyclohexa- over Pt/C and PdjC in liquid phase at 100°Cwith IOO atm. initial pressure of H,, the furan ring was diene-1, 3 on Platinum Films over a Wide not reduced but hydrogenolysis occurred at the Range of Temperature C-0 bond adjacent to the CO group. Heptane- V. D. YAGODOVSKII, V. M. GRYAZNOV and E. A. dione-2, 6 was produced which was further SAVEL'EVA, Kinetika i Kataliz, 1963,4, (5), 746752 converted by cyclisation to cyclohexanone and The activation energy of the dehydrogenation of cyclohexanol (20-300;,) and by reduction to cyclohexadiene-I, 3 on thin Pt films was shown to heptanediol-a, 6, which was partly cyclised to 2, increase sharply at higher temperatures in the 6-dimethyltetrahydropyrane (about 50-6ooi;). 20c-500"C range. Treatment of the Pt films at Over Pd/Al,Os at 180°C the CO group was

Platinum Metals Rev., 1964, 8, (1) 34 reduced to CH2followed by hydrogenation of the the numbers of H atoms were equal, isomerisation furan ring to produce 2-methyl-5-ethyltetra- occurred more readily with the compound more hydrofuran (about 607,). difficult to hydrogenate.

The Oxidation of Olefines by Palladium Isomerisation of Hexenes in the Presence of Chloride Catalyst Palladium on Charcoal J. SMIDT, Bull. Assoc. Franc. Tech. PCtrole, 1963, B. A. KAZANSKII, N. B. DOBROSERDOVA, G. S. (160), 467-487 BAKHMET’EVA and I. V. GOSTUNSKAYA, Neftek- The oxidation of olefines to carbonyl compounds himiya, 1963,3, (41,503-5c6 with the same number of C atoms by 0, in The transfer of double bonds in hexenes was aqueous solutions of PdCle and CuC1, is described studied at 80°C in the presence of Pd/C and a and reaction mechanisms are discussed. Higher stream of H2. The relation bctween the initial olefines and other unsaturated compounds are hexenes and the rates of transfer was obtained and attacked by O2 at points within the molecule this agreed with the associative mechanism for the depending on the groups attached to the double- reaction. bond C atoms and on the Pd ligands. Ketones and aldehydes are formed from a-olefines in fixed Inhibitors in the Palladium-Catalysed Hy- proportions. The transesterification of vinyl drogenation of Aryl Nitro Groups esters is the most important non-oxidation H. GREENFIELD, J. OTg. Chem., f963. 28, (9), reaction of unsaturated compound in which Pd 2434-2435 salts act as catalysts. Potential inhibitors of the hydrogenation of p-nitrotoluene to p-toluidine over 5”/D Pd/C were Kinetic Study on the Oxidation of Propylene added to the reaction mixture at the rate of 5 in the Presence of Palladium Chloride mole‘)” with respect to p-nitrotoluene, i.e. more T. DOZONO and T. SHIBA, Bull. Japan Petroleum than trace amounts. Sodium nitrate, acetate, Inst., 1963,5, (March), 8-12 sulphate, carbonate, phosphate, hydroxide, fluor- The rate of the catalytic oxidation of propylene to ide, bromide and chloride and ferrous chloride, form carbonyl compounds by Smidt’s process, n-octyl chloride, chlorobenzenc, bromobenzene, using acidic PdCl,;CuCl, solution, was shown to aniline, piperidine, sodium methoxide, phenol [PdClJ p-nitrophenol caused no inhibition. Various be r=k pprop, where [PdCl,] and and [Cl-] 2.21 [H+] amounts of poisoning were caused by potassium TC1-1 are the total concentrations of PdCl, and nitrite and sodium nitrate, sulphite, iodide, kl’ ion respectively, where [Hf] is the initial H+ cyanide, sulphide and bisulphite and also by nickel ion concentration and where pprop is the partial (11), ferric, cobalt (11), chromium (111), copper pressure of propylene; k is a constant. (11), silver, aluminium, zinc and lead nitrates, ferric chloride, copper chloride and by certain Oxidation of Alcohols by Palladium Salts in nitroso compounds. Phenyl disulphide caused Aqueous Solution inhibition at very low concentrations. A. v. NIKIFOROVA, I. I. MOISEEV and YA. K. SYRKIN, Zh. Obshch. Khim., 1963,33, (lo), 32393242 Comparison of the Rates of Catalytic Aliphatic alcohols were oxidised to aldehydes and Reduction and Electroreduction of Some aliphatic-aromatic alcohols to ketones by PdCl, in Organic Substances on Rhodium. 111 aqueous solution. The reactions were slower than T. M. GRISHINA, L. P. KHOMCIIENKO and G. D. for the corresponding olefines. Alcohols tested VOVCHENKO, Vest. Mosk. Univ., Ser. 11, Khim., were ethyl-, isopropyl, N-butyl, N-arnyl- and 1963, (41, 55-58 benzoyl-alcohol. The catalytic reduction rate on Rh was several times as great as its electroreduction rate for Isomerisation of Hexenes in Liqnid Phase crotonaldehyde, acetone and amylaldehyde. Both Hydrogenation Conditions in the Presence of rates decreased in the same order of substances. Palladium Black The rates decreased with changes in the electro- I. V. GOSTUNSKAYA, A. I. LEONOVA, N. B. lyte in the order H2S04,HCl, KOH and HBr. DOBROSERDOVA and B. A. KAZANSKII, Nefte- Br- ions particularly poisoned the reductions. Rh has lower reduction rates than Pt but higher khimiya, 1963, 3, (4,498-502 than Ru. Double bond transfer from position f to position 2, but no further along the C chain, was observed of during partial hydrogenations of hexene-I, Study Ruthenium Catalyst. I. Hydrogen- 2-methylpentene-I, 3-methylpentene-I, 4- ation of Organic Compounds by Ruthenium methylpentene-1 and 2, 3-dimethylbutene- f in Catalyst the presence of Pd black at atm. pressure and Y. TAGAKI, sci. Papers Inst. Phys. Chem. Res. 20°C in C2H,0H medium. Compounds with two (Japan), 1963,57, (2), 105-109 H atoms in the ally1 positions were isomerised Ru and RuIC catalysts were prepared and used more readily than those with only one. Where in liquid phase hydrogenations at ordinary and

Platinum Metals Rev., 1964, 8, (1) 3s high pressures. Carbonyl groups were more and PdjC showed the relative amounts of isomer- easily hydrogenated at ordinary pressure over Ru isation and hydrogenolysis of cis- and trans-I, than over Pd. HzO and alkaline solvents were 2-dimethylcyclopentane. The amount of hydro- more effective than alcohol for these reactions. genolysis and the proportion of cis- to trans- The hydrogenation of double bonds was difficult. isomer increase with temperature. PdjC and Nitro groups could be hydrogenated but nitriles RhjC appeared to be most active in these respects. could not. At high pressure the rate of hydro- The reaction mechanism is discussed and the genation of ally1 alcohol was rapid. Citronella1 effects are compared with PtjC catalyst action. was used for studies on the hydrogenation of unsaturated aldehydes and the selectivity of such processes. FUEL CELLS The Production of Polyhydroxy Alcohols A High Performance Saturated Hydrocarbon from Wood-Polysaccharides Fuel Cell v. I. SCHARKOW, Chem. -Ing. -Tech., 1963, 35, (7), W. T. GRUBB and L. W. NIEDRACH, J. Electrochem. 494-497 Soc., 1963,110, (IO), 10861087 Optimum conditions for the simultaneous Propane has been oxidised rapidly in a new hydrolysis and hydrogenation of polysaccharides propane-oxygen fuel cell operating at only 150°C in the presence of RuiBaSO, or of RujC are with a new porous eleLTrode structure to preserve 0.794 H,PO, solution, 16c-165T, 60-80 atm. the catalytic properties of the Pt incorporatedin pressure of H, and a reaction time of 50-60 min. it. The electrolyte was HJ'O,. A voltage-current Other processes developed in the U.S.S.R. are density curve when 14.6 M H,PO, was used based on the hydrolysis of the polysaccharides to showed no limiting current in the range examined monosaccharides which are subsequently hydro- thanks to the thin structure of the electrodes. genated, on hydrogenolysis of monosaccharides and on the pyrolysis of cellulose. A Direct Hydrocarbon/Air Fuel Cell H. G. OSWIN, A. J. HARTNER and F. MALASPINA, An Investigation of Catalysts Produced by Nature, 1963,200, (49031,256257 Reactions between Sodium Borohydride and The combustion of propane in a fuel cell using Salts of Heavy Metals Pt black electrodes and H,P04 electrolyte at A. M. TABER, B. D. POLKOVNIKOV, N. N. MAL'TSEVA, temperatures up to 220'C produced currents of V. I. MIKHEEVA and A. A. BALANDIN, Doklady Akad. 100-zoo mA/cms at 20-350/0 thermal efficiency. Nauk S.S.S.R., 1963, 152,(1), 119-121 The process appeared to consist of catalytic 0.25 g samples of PdCl,, RhC1, and H,PtCl, were cracking followed by combustion and again reacted with I g samples of NaBH, and the demonstrated the feasibility of direct hydro- products were analysed chemically, thermally and carbon/air fuel cells. by spectrography. The infra-red spectra of Pd-B alloys and of borided Pd catalyst were very similar. The Development of a High-efficiency H, in the catalysts acted similarly to adsorbed H,. Hydrogen-diffusing Palladium-Alloy Anode for Use in Fuel Cells Charging Curves of the Bode Catalysts of H. G. OSWIN, s. M. CHODOSH and N. I. PALMER, Pt-Group Metals Abs. Papers, 145th Meeting, Am. Chem. SOC., A. M. TABER, A. A. BALANDIN, D. v. SOKOL'SKII and 1963, 14K-39 B. D. POLKOVNIKOV, Doklady Akad. Nauk S.S.S.R., Thin Pd alloy membranes for use as H,-diffusing 1963,152, (z), 379-381 anodes were studied electrochemically and the Results were plotted for Pt, Pd and Rh boride effect of various operating parameters on anode catalysts in 0.1N solutions of NaOH, HzSO, and polarisation were determined. Variables discussed HC1 and for the anodic polarisation of phenyl- include temperature, electrolyte composition, gas acetylene and diphenylacetylene. Charging was pressure and composition, membrane thickness most rapid for the NaOH electrolyte and the and surface preparation. acetylenes were most readily polarised over Pt. Isotherms for the solution of H, in the catalysts Thin Fuel Cell Electrodes were also determined. R. G. HALDEMAN, w. P. COLMAN, s. H. LANGER and w. A. BARBER, Abs. Papers, 145th Meeting, Am. Hydrogenolysis and Reciprocal Transitions Chem. SOC.,1963, I5K-42 of cis- and trans-1, 2-Dimethylcyclopen- Thin electrodes consisting of Pt and Pt-C tanes in the Presence of Rhodium, Osmium, supported on screens were found to be capable of Iridium and Palladium Catalysts very high performance. The Pt used was from I 0. V. BRAGIN, A. L. LIBERIMAN, G. K. GURIANOVA to 9 mg/cma. Electrodes were tested in acidic and and B. A. KAZANSKII, Doklady Akad. Nauk basic media and were studied for initial polarisa- S.S.S.R., 1963, 152, (4), 865-868 tion, length of service, effect of temperature and Studies at 150 to 280°C over Rh/C, Os/C, IrIC operation in air.

Platinum Metals Rev., 1964, 8, (1) 36 ANODIC PROTECTION ELECTRICAL ENGINEERING

Anodic Protection of Carbon Steel in Sul- Effect of Organic Vapours on Contact phuric Acid Materials in Communication Engineering w. P. BANKS and J. D. SUDBURY, Corrosion, 1963, J. TLAMSA and J. NUSZBERGER, Nachrichtentechnik, 19, (91, 300t-307t 1963, 13~(7), 272-277 Pt cathodes were used in tests on 102.0-type mild Ag and Pd contacts were tested for the effects of steel samples. The tests showed that the effective- organic vapours arising from the atmosphere ness of anodic protection and the current density around them or from wire sheathing materials. for maintaining protection depended on acid The vapours can cause sparking, deposits, concentration and temperature. H,S04 varied erosion and changes of resistance. Results are from 45 to IO~"!~concentration and temperature tabulated and cover most of the more common from 80 to 535°F. Graphs illustrate the large organic materials likely to be present. The increase in current density required with higher activities of contacts in various surroundings are temperatures. also listed.

NEW PATENTS Production of Semi-Conductor Devices Isomerisation of Olefines THE INTERNATIONAL NICKEL CO. (MOND) LTD. BRITISH PETROLEUM GO. LTD. British Patent 932,748 British Patent 930,091 The isomerisation of a branched chain olefine is An electrically conducting joint is made between a carried out using as catalyst a complex of the wire and a metal layer adherent to a semi- olefine with a halide of a platinum group metal. conductor material with the use of a paste-like dispersion of a metal powder in a liquid vehicle Electrolytic Apparatus consisting of a thermally decomposable com- D. J. EVANS (RESEARCH) LTD. British Patent 932,945 pound. Pt metal and a decomposable Pt com- Provides an improved closed electrolytic cell for pound is used. the production of chlorine by the electrolysis of brine. Electrodes consist of thin sheets of Ti Catalytic Decomposition of Hydrazine coated with Pt. ENGELHARD INDUSTRIES INC. British Patent 930,499 Hydrazine is contacted with a catalyst composed Reforming Catalyst of one or more of Rh, Ir, Ru and Pd, with, if THE STANDARD OIL CO. British Patent 934,080 desired, Pt. A hydroforming catalyst comprises Pt/Al,O, in which at least 40% by wt. of the Pt is maintained Preparation of 20-Alkylamine Steroid Deri- in an HF-soluble form. vatives SMITH KLINE & FRENCH LABORATORIES. British Selective Hydrogenation Patent, 930,676 FARBENFABRIKEN BAYER A.G. British Patent 934,429 A Pt oxide hydrogenation catalyst is used in the Process for the selective hydrogenation of preparation of the above substances. acetylene/diolefine components of a liquid hydrocarbon mixture comprises trickling the Isomerisation of Olefinic Hydrocarbons hydrocarbon mixture over a Pt or Pd hydrogena- BRITISH PETROLEUM CO. LTD. British Patent tion catalyst carried on a macroporous support. 931,922 The isomerisation of an olefine is carried out Ignition Device using as catalyst a compound of the olefine with a ROLLS-ROYCE LTD. British Patent 934,499 halide of a Pt group metal. Ignition device for combustion eqpipment comprises a mass of refractory material and a foramin- Preparation of Cyanoalkyl Chlorosilanes ate catalytic element of Pt, or Rh, or a Pt alloy GENERAL ELECTRIC GO. British Patent 932,380 of less than 0.030 in. thick. Relates to a method of forming alpha-cyanoethyl methyl dichlorosilane which comprises reacting Fuel Cells methyl dichlorosilane with acrylonitrile in the LEESONA CORP. British Patent 935,430 presence of a catalyst composition comprising Provides a fuel cell constructed for high tem- palladous chloride, a trialkylamine and a poly- perature operation in which at least one electrode amine. is composed of a mixture of zinc oxide and

Platinum Metals Rev., 1964, 8, (l), 3740 37 metallic Ag, the gas interface of the electrode bcing known carbon electrode in an atmosphere of coated with A1 silicate or alumina activated with an carbon dioxide at 800"C, wet proofing the fired activating metal, e.g. Pd, Pt or Rh. electrode in the usual way and then impregnating it with chloroplatinic acid containing 90 mgiml Pt Removal of Methane so that the electrode contains 2 mg Pt per sq. m. of ENGELHARD INDUSTRIES INC. British Patent 935,951 electrode. Process for removing methane or oxygen from a gaseous mixture comprises adding oxygen or Thin Film Noble Metal Thermocouple methane respectively to the gaseous mixture to ENGELHARD INDUSTRIES INC. US.Patent 3,099,575 form a stoichiometric excess, and passing the New thermocouples consist of thin films of resultant mixture over a Pt group metal catalyst electrically dissimilar noble metals bonded to a to effect combustion. refractory support and connected together, one of these films comprising at least one of Pt, Pd, Rh Metal-Loaded Molecular Sieves and Ir, and the other film of at least two of Pd, UNION CARBIDE CORP. British Patent 937,748, Rh and Ir, the films being produced by applying the finely divided metals in an organic liquid 937,749, 937,750 carrier vehicle. In an example, one film consists All relate to metal-loaded molecular sieves suitahle for use as catalysts, scavengers, getters and the of Pt metal and the other of 90X Pt and IO"~Rh. like, the sieves being loaded with one or more of the Pt group metals. Hydrocracking Catalyst PHILLIPS PETROLEUM GO. U.S. Patent 3,099,618 Hydrogenation Cyclisation Catalyst High sulphur petroleum oils are cracked at lower MINNESOTA MINING & MANUFACTURING CO. GT.s. temperatures with less coke formation by using a Patent 3,09~,423 Pt-activated Co-molybdate catalyst containing Succinonitrile is catalytically hydrogenated and 1-3 wt.O/D Pt, 1-10 wt.9; Co and 1-20 wt.9, Mo cyclised simultaneously to 2-pyrrolidone by on a suitable support. heating in an aqueous medium at 20-2o0°C in the presence of a Ru or Pt oxide catalyst. Anode Assembly for Cathodic Protection CHEMIONICS ENGINEERING LABORATORIES INC. u.s. Fuel Cell Electrode Patent 3,101,311 AIR PRODUCTS & CHEMICALS INC. U.S. Patent In providing cathodic protection with an im- pressed current, the anode is an assembly 3,0973974 The electrode is produced by forming a suspen- consisting of a Pt disc of foil thickness in a suitable sion of active C powder impregnated with Pt or holder, the foil having a grid-like pattern on its Pd and depositing the suspension in the porcs of surface to facilitate the release of gaseous electro- an electrode matrix made from Ni, Ag, Fe, etc. lysis products. Catalytic Production of Hydrocyanic Acid Platinum Composite Hydrocracking Catalyst IMPERIAL CHEMICAL INDUSTRIES LTD. US. Patent SOCONY MOBIL OIL CO. U.S. Patent 3,098,030 3,102,001 A high boiling petroleum hydrocarbon or hydro- The reaction of the NH,, a hydrocarbon and carbon mixture is cracked in the presence of oxygen over a Pt metal or alloy is improved by hydrogen using a physical particle-form mixture adding 1-500 mg/m3 of S and controlling the of (I) a Mo oxide combined with an acidic oxygen content to give maximum conversion of component of two or more oxides of Groups IIA, the NHs. In No. 3,102,269 the catalyst is specifi- IIIB, IVA and VIB and (2) an inert support cally claimed and consists of a Pt metal or alloy in containing 0.05-10wt."& of a Pt metal. massive form, such as one or more layers of metal gauze. At least 50°(, Pt must be present, as in an Catalytic Removal of Oxygen from Mixtures alloy of 90°{,Pt and ID":, Rh. Containing Nitric Oxide ENGELHARD INDUSTRIES INC. U.S. Patent 3,098,712 Brine Purification Cathode A Pt or Rh catalyst, e.g. supported on alumina, IMPERIAL CHEMICAL INDUSTRIES LTD. US. Patent may be used to strip oxygen from mixturcs 3,102,085 containing oxygen and nitric oxide, and optionally Brine containing free and available chlorine is other gases, when hydrogen is added to the purified by passing it through an electrolytic mixture before passing over the catalyst. This diaphragm cell having a cathode not attacked by keeps down the catalyst operating temperature. Cl,, e.g. a Pt-coated Ti electrode.

Fuel Cell Electrode Platinum Cathode for Titanium Protection US. SECRETARY OF ARMY. U.S. Patent 3,098,772 IMPERIAL CHEMICAL INDUSTRIES LTD. U.S. Patent A platinised carbon electrode specifically for use 3,102,086 with acid electrolytes is produced by firing a When Ti and its alloys are exposed to corrosion by

Platinum Metals Rev., 1964, 8, (1) 38 strong acids, they are protected by immersing in methane and oxygen is catalysed on a high the acid a Pt/C electrode and impressing a temperature plate made of an alloy of 90 '7; Pt and current on the system. approximately 10% Rh. Sulphur-resistant Hydrogenation Catalyst Fritted Platinum Surface Coatings THE STANDARD OIL co. US.Putent 3,102,864 INTERNATIONAL NICKEL CO. (MOND) LTD. French Catalysts suitable for a variety of feedstocks where Patent 1,320,963 the fatty substance, olefine, etc., contains S consist Pt-coated metal and refractory surfaces, par- of an alumina support impregnated with 0.05-1y0 ticularly for use in contact with molten glass and of Pt and 0.05-6';" of thiocyanate ion and option- enamels, are produced by applying a dispersion ally also a small amount of halogen. of flake Pt metal in a liquid medium, capable of being volatilised or destroyed, and then heating Catalytic Hydrogenation of Nitrosamine the coated surface to remove the dispersion E. I. DU PONT DE NEMOURS & co. U.S. Patent medium and sintering or fritting the resulting layer of Pt metal flakes. Typically Pt coatings 3,102,887 Unsymmetrical disubstituted hydrazines are can be produced on alumina in this way. produced by the catalytic hydrogenation of the corresponding nitrosamine using a Pt metal Catalyst for the Reduction of Mononitro catalyst, as hitherto, but now hydrogenation, e.g. Aromatic Compounds with Pd/C, is carried out in a solution of ionic GENERAL ANILINE AND FILM CORP. French Patent strength of at least 0.6 to improve the yields. 1,3213689 Mononitro ethers are reduced to the correspond- Complexed Platinum Metal Hydrogenation ing amines with hydrogen in the liquid phase Catalyst using a noble metal catalyst having a surface of at SHELL OIL CO. U.S. Patent 3,102,899 least about 150 sq. m/g at a temperature of The Pt metals are amongst the transition metals 25-125"C with the reaction water held in disper- which may be complexed with trihydrocarbon sion. Suitable catalysts are PdfC or Pt/Al,O,. phosphites, arsenites and stibenites to provide new hydrogenation catalyst, etc. Activation of Platinum Metal-Alumina Catalysts Fuel Cell Reduction System TEXACO DEVELOPMENT CORPORATION. French Patent IONICS INC. U.S. Patent 3,103,473 1,322,457 Fuel cell producing the electrochemical reduction The activity of catalysts containing a Pt metaI of organic and inorganic compounds has a used for hydrocarbon isomerisation is increased 'permanent' electrode for the fuel and this by contacting them with an activating chloride contains a Pt metal as catalyst. In No. 3,103,474 (e.g. a chloroalkane or an organic chloro acid) the same invention is applied to the recovery of having a C1 : C atom ratio of at least z : I at a metals, e.g. Cu, Fe, Zn, Cr, Ni, Mn, Co, or Cd, by temperature between 149 and 345°C. Typical the reduction of solutions of their compounds. activating compounds are CCI,, CH,CI and trichloroacetic acid chloride. Reduction Catalyst for Organometallic Com- pounds Oxidation of a Palladium or Platinum Group Metal ETHYL CORP. U.S. Patent 3,103,525 SHELL INTERNATIONAL RESEARCH MAATSCHAPPIJ Zerovalent aryl cyclopentadienyl compounds of N.V. French Putent 1,325,696 the iron subgroup (Mn, Tc, Rh, Fe, Ru, 0s) are produced by the reduction of the corresponding Metallic Pd, Pt, Rh, Ru, 0s or Ir or a lower halide by hydrogen over Pt, Pd, etc. valency compound thereof is oxidised in a non- aqueous medium by a gas containing oxygen in the presence of one or more carboxylic acids, one Palladium and/or Platinum Reformation or more other acids and one or more metal Catalyst oxidation promoters. The carboxylic acid THE STANDARD OIL CO. Dutch Patent 234,680 preferably provides the oxidation medium. This A new Pt and/or Pd supported catalyst containing process is particularly intended for reoxidising Pd 0.1-10 wt.o/b Ge is produced by impregnating compounds produced when the higher Pd com- the support with one or more Pt or Pd compounds pounds are reduced in producing alkenyl esters of and at least one Ge compound and reducing the carboxylic acids. product. Duplex Chemical Plating Catalyst for Hydrogen Cyanide Production ENGELHARD INDUSTRIES INC. Ererzclz Patent IMPERIAL CHEMICAL INDUSTRIES LTD. French Patent 1,326,441 80,273. Addition to 1,248,895. New economic form of the noble metal-plated The reaction of NH,, a hydrocarbon(s) such as deposit required for printed circuit manufacture

Platinum Metals Rev., 1964, 8, (1) 39 consists of a layer of Pt, Rh, Pd or Ru covered Pt and/or Pd represents 5 to 95Y0 of the mixture with a thin layer of Au, instead of the wholly Au with Ge. The mixture is said to be a solid solution layer previously employed. Normal chemical and is 0.1to 10 wt.:& of the total weight of the plating methods are used. catalyst and support together. Fusion Production of Platinum Catalyst for Methane Production COMPTOIR LYON-ALEIMAND, LOUYOT & CIE. French ENGELHARD INDUSTRIES LTD. German Patent Patent 1,326,597 1,151,247 Pt, Rh and similar metals are produced by first Catalyst giving low ignition temperatures (e.g. pyrolysing a compound to yield a sponge or foam, of the order of 271-40o"C)for mixtures lean in as hitherto, which is then degassed by heating to a CHI and 0, consists of Rh, or Rh and another temperature just below its melting point, com- Pt metal, on a support, preferably activated pressed, melted in a vacuum and cast to give alumina. articles such as spinnerets which come into contact with molten glass. Gold-platinum Alloy for Tensioning Strips W. C. HERAEUS G.m.b.H. German Patent 1,151,944 Gold-Palladium Brazing Alloy An alloy consisting of I--soo/,, preferably 1-3074, AEROJET-GENERAL CORP. French Patent 1,326,619 of one or more Pt metals, particularly Pt and Rh Brazing alloys which require less critical con- and/or 1-50u/i,,preferably 5-400/;, of one or more ditions consist of 1-96% Au, 1-720/" Pd and of the metals Fe, Co and Ni, remainder at least 2-61?; of Ni, Co and/or Cry e.g. 45-35O4 Au, zoy$ Au, is used as material for making tensioning 30-200/~Pd, 40-30% Ni and 0-100/4 Cr. strips in measuring instruments with rotating measuring parts. Electroless Metallising of Powdered Metal Oxides and/or Sulphides Catalyst for Benzyl Chloride Amination SHERRITT GORDON MINES LTD. German Patent SHELL INTERNATIONALE RESEARCH MAATSCHAPPIJ 1,1439372 N.V. German Patent 1,152,098 In the electroless metallising of difficult-to-melt ~,6-Dichlorobenzonitrile is produced from 2,6- powdered metal oxides, and/or sulphides, by dicblorobenzyl chloride or 2,6-dichlorobenzal dispersion of the previously activated powder in a chloride by reaction at I50-50Ooc with a mixture metallic salt solution under reducing conditions, of ammonia and oxygen in the presence of a Pt the powder is treated in an ammoniacal metal salt metal catalyst on a C support, e.g. palladised solution of Os, Rh, Ru, Pt or Pd with reducing animal charcoal. gases under a partial pressure of over 4 atm at a temperature above 90°C. See also No. 1,143,373 Platinum Metal Alloy for Tensioning Strips for similar treatment of powdered carbides, W. C. HERAEUS G.m.b.H. German Patent 1,152,826 borides, silicides or nitrides. An alloy for making tensioning strips for precision instruments consists of r-50'Z/oYpreferably 5-40;,, Electrodeposition of Thick Stress Free of one or more of Fey Co, Ni, W, Mo, Cu or Ag Platinum Coating with the remainder Pt, Pd andlor Rh with a SEL-REX CORP. German Patent 1,144,074 minimum of 15% Fe group metal in the case of A plating bath for depositing thick stress-free Pt binary alloys. Optionally part of the three Pt coatings comprises an aqueous solution of platino- metals may be replaced by other Pt metals, diamino-nitrite, in amount corresponding to a Pt particularly Ir. content of at least 6 g /I in sulphamic acid. Catalytic Olefine Oxidation Process for Production of Highly Active FARBWERKEHOECHST A.G. German Patent 1,153,008 Metal Catalysts from Laminar Silicates Aldehydes and/or ketones are produced from VEB FARBENFABRIKEN WOLFEN. German Patent olefines by oxidation in the presence of a solution 1,1441696 containing a Pt metal salt in a new process in Barium phyllosilicate, BaSi,O, or sodium phyllo- which the amount of oxygen allowed to come silicate Na,Si,O, is treated with a salt solution of into contact with the catalyst is strictly limited. catalytically active metals at normal pressure or in an autoclave and the metal laminar silicate so Separation of Palladium from Other Plati- formed is, after separation, reduced. The treating num Metals solution may be a solution of bivalent salts of CANADIAN COPPER REFINERS LTD. German Patent e.g. Pd or Pt. 'J153>737 A solution of Pt metals in HCI, HNO, or H,SO, Reforming Catalyst is treated with SO, to give an SO2 content of THE STANDARD OIL CO. German Patent 1,151,082 0.3-4 gll and then a soluble iodide is added in A new reforming catalyst consists of Ge and Pt stoichiometric amount to precipitate Pd quan- and/or Pd deposited on a support. Preferably the titatively by itself as PdI,.

Platinum Metals Rev., 1964, 8, (1) 40