PLATINUM METALS REVIEWPlatinum Metals Rev., 2005, 49, (2), 62Œ69 62 DOI: 10.1595/147106705X45604 New Stirrer Technology for the Glass Industry LONG-TERM BENEFITS FROM THE ‘DIFFUSION

E-ISSN 1471–0676

PLATINUM METALS REVIEWA Quarterly Survey of Research on the Platinum Metals and

of Developments in their Application in Industrywww.platinummetalsreview.com

VOL. 49 APRIL 2005 NO. 2

Contents

New Stirrer Technology for the Glass Industry 62By Duncan R. Coupland and Paul Williams

Iridium/Carbon Films Prepared by MOCVD 70By Changyi Hu, Jigao Wan and Jiaoyan Dai

Modern Palladium Catalysis 77A book review by Mark Hooper

Potential Applications of Fission Platinoids in Industry 79By Zdenek Kolarik and Edouard V. Renard

Ruthenium Catalyst for Treatment of Water Containing 91Concentrated Organic Waste

By YuanJin Lei, ShuDong Zhang, JingChuan He, JiangChun Wu and Yun Yang

Patents and Copyright for Scientists 98By Ian Wishart

Abstracts 102

New Patents 106

Final Analysis: Thermocouples Compensating Circuits 108By Roger Wilkinson

Communications should be addressed to: The Editor, Susan V. Ashton, Platinum Metals Review, [email protected] Matthey Public Limited Company, Hatton Garden, London EC1N 8EE

Good quality glass has to be homogeneous. Toachieve this, glass melting furnaces have beendeveloped to give a high degree of mixing andcapability to deliver uniform glass into the fore-hearth. However, the necessity to continuouslycondition (heat, cool, or de-gas, etc.) the glassflowing towards the working end can negate someof this design and can cause thermal and composi-tional inhomogeneities in the flowing glass. Thiscould compromise the quality of the finished prod-uct. To produce homogeneous glass it is thereforenecessary to stir the glass in the forehearth, andthis is widely employed. However, the introductionof stirrers has repercussions; the function of stir-ring is of value, but the physical presence of thestirrer is a drawback.

The choice of material for the stirrer helps todetermine the optimum benefit, as the cost, effec-tiveness and durability need to be balanced. Eachof these aspects depends on the final application ofthe glass and the nature of the molten glass, specif-ically, its viscosity, temperature, corrosivity, qualityand value. Stirrers with their glass contact surfacesmade in platinum or platinum alloys provide thebest solution to this issue, but for many installationsites, such as container glass forehearths, where the

value of the product has traditionally been low, thecost of fabricated platinum stirrers has historicallybeen too high.

The introduction in 1994 of ACTTM platinum-coated ceramics changed this (1, 2). ACTTM

technology provided great improvements in theresistance of ceramics to molten glass at relativelymoderate cost by providing enhanced durabilityand longevity compared with prior used unpro-tected ceramic (3).

For glasses of very high value, specifically opti-cal glasses where quality and clarity are paramount,stirrers fabricated from platinum alloys havealways been used, although they have limited dura-bility in high viscosity glass, especially when theglass is used for large components.

Stirrer cores made from molybdenum havebeen found (at least 15 years ago) to provide thestrength that is required by platinum for parts usedin high value glass making, such as in gobbing andhigh energy stirrers. Separation of the platinumand the molybdenum by an oxide diffusion barrierand evacuation of the resulting space are necessaryto avoid the cores from volatilising at temperaturesabove ~ 400ºC.

Other materials evaluated over the years for

Platinum Metals Rev., 2005, 49, (2), 6269 62

DOI: 10.1595/147106705X45604

New Stirrer Technology for the Glass IndustryLONG-TERM BENEFITS FROM THE ‘DIFFUSION CHOKE’

By Duncan R. Coupland* and Paul WilliamsJohnson Matthey Noble Metals, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.; *E-mail: [email protected]

The function of stirring in glass making is to create uniform, homogeneous glass. Stirringequipment operates at high temperatures and under high mechanical stresses, so stirringdevices have to be robust and often involve large amounts of platinum or platinum alloys. Thestirrers, stirrer bars, blenders, homogenisers, screw plungers and plunging stirrers currentlyused are generally effective in operation, reliable and with predictable lifetimes. Thus therehas been no incentive to improve the technology, and stirrer designs have changed little in thelast twenty or thirty years. However, the current economic climate in the glass industry demandslower costs, improved operational efficiency, and reduced platinum inventories glass makinguses large quantities of platinum, with stirring devices taking a large part of it. To help reducethese amounts work has been undertaken on stirrer technology. and recent developments haveresulted in lower platinum requirements (in some cases by over 90 per cent) without jeopardisingstirring effectiveness or stirrer longevity. Different types of glass stirrers are examined hereand a new concept in stirrer design, a diffusion choke, is described.

stirrer cores and similar applications in the glassindustry have included platinum alloy coated, highstrength oxide dispersion strengthened (ODS)superalloys, but without significant success.However, the recent development of diffusionchoke technology has overturned these failuresand enabled the use of these superalloys. Thispaper looks at the background to the diffusionchoke development and the improvements thatare now possible.

Recent Stirrer HistoryACTTM Platinum Coated Stirrers

ACTTM platinum coating technology has beenused in molten glass furnaces for more than tenyears (1, 2). Some of the earliest applications werecoated ceramic stirrers for application in the sever-est conditions, such as in opal and borosilicate

glass and colouring forehearths (where colour isadded to glass). The objective was to prevent theceramic from being eroded and to allow continu-ous efficient stirring. The effectiveness of theACTTM technology can be seen by the uncorrodedstirrer on the left in Figure 1; all three stirrers hadexperienced six months of continuous service.ACTTM technology is now being used to re-definethe nature of stirrers and to give more advantagesover conventional fabrications, see, for exampleFigure 2.

In a recent application, co-planar ceramic stir-rers with ACTTM platinum alloy coatings, seeFigure 3, replaced helical stirrers fully fabricatedfrom platinum alloy sheet. The improved perfor-mance they achieved in stirring molten TV panelglass has dramatically assisted in this economicallydifficult area. In one case, ACTTM-coated ceramic

Platinum Metals Rev., 2005, 49, (2) 63

Fig. 1 Three stirrers that were usedtogether at the same time for the sametime (approximately six months) in acolouring forehearth. The glass immersionline can be seen.

The stirrers were identical except the oneon the left has an ACTTM platinum coating.

The stirrers rotate in the glass. They are ~1 m in length and made from an alumino-silicate ceramic

Fig. 2 A conventional fabricated helicalscrew glass stirrer made of 10%Rh-Pt. Thisstirrer typically has a life of about 5 years.Such stirrers are used for all glass makingthat is inherently expensive due to the largeamount of precious metal required. Weldedjoins on this stirrer are visible

In a forehearth there may be from 2 to 16such stirrers operating in banks of up to 4.They are mechanically operated in optimumstirring patterns, with the other end of thestirrer being fixed to a geared drive

stirrers replaced traditional fabricated platinumones, and thus reduced the platinum that was beingused from a total of 84 kg to only 8 kg. This reduc-tion was partially accomplished by the superiordesign of the stirrer so that fewer were needed(from 10 to 4), and by the reduced thickness of thecoating as compared to the prior fabricated stirrer.The design of an ACTTM coated stirrer is dictatedby the ceramic and the requirements of the appli-cation. Many different configurations have beendefined and utilised.

Platinum-Clad Base Metal StirrersFor many years molybdenum has been used as

the material of choice for structural applicationswithin the glass furnace as it performs well inmolten glass. However, although it is used exten-sively as electrodes in electrically heated furnaces,if free oxygen impinges on its hot surface, it burnsrapidly. This is a major limitation. To be effectivethe molybdenum must be protected if it is to func-tion at any temperature > ~ 400ºC. Therefore inits major application of resistance-heating elec-trodes, it must be water-cooled to ensure that thezone not protected by immersion in molten glass iskept below this critical temperature.

Platinum does not have this limitation and canbe successfully used in such applications withoutwater-cooling. It is assumed that using platinumwould make an electrode too expensive, but this isnot always the case and the introduction of ACTTM

platinum coating technology has allowed electrodedesigns to be generated that have all the advan-tages of platinum without the disadvantages of

molybdenum. Indeed, in some applications wherea solid electrode is required, iridium, which hasunmatched stability in glasses that are especiallyaggressive when molten, and high environmentalresistance, is now being considered as a viablealternative to molybdenum.

In glass stirrer technology it is desirable to makeuse of the strength of molybdenum for applica-tions where the shear strength requirements arevery high and where unexpected failure would beexpensive. Protecting the molybdenum is critical inachieving this. Platinum cladding has convention-ally been utilised in a simple symbiosis: a platinumalloy cladding protects the strength-donatingmolybdenum. As in many symbiotic relationshipsthere is a parasitic component, and the two mate-rials can, under some circumstances, interact andform potentially detrimental intermetallic phases(4). The effect of this can be seen in Figure 4 whichshows the weight losses observed for a series ofmolybdenum samples protected by platinum coat-ings of high thickness. The simple platinum layercan protect the substrate until interdiffusion andinteraction promote failure of the platinum layer.Once this happens rapid oxidation of the molyb-denum occurs with dramatic loss of weight.

The addition of a ceramic barrier layer to keepthe two metals apart was a natural progression. Aceramic barrier can control interdiffusion and oxy-gen removal from the inner space (the volumebetween the cladding and the molybdenum sub-strate) (5). This situation must be maintained forthe duration of the service life of the component.

These stirrer designs have been used to great


Fig. 3 An ACTTM coated co-planar stirrer as used forhomogenising and disturbing laminar flows of glass.

These vanes will be fully immersed in the glass, withthe glass surface being a few centimetres above theupper vane. The vanes operate in simple rotation buteach pair of stirrers will be contra-rotated

success for several decades, and with care can havelives of more than five years. However, when thecladding fails either by mechanical damage, physi-cal change or chemical attack, the introduction ofoxygen onto the molybdenum can cause a dramat-ic and rapid failure. This failure can be anticipatedand avoided, but if it is unexpected the damage tothe forehearth and the resulting down-time can beextreme.

Alternative core materials have been tested, andsuperalloys are most likely to be suited to thisarduous task. These materials were developedspecifically for the gas turbine industry and weredesigned to have excellent strength, and very goodoxidation resistance up to temperatures of ~1100ºC. Oxide dispersion strengthened (ODS)alloys can, of course, be used to provide strengthat temperatures up to 1300ºC. Some of these ODSalloys have considerable resistance to the harshenvironment above molten glass and also whensubmerged in glass, but they tend to erode rela-tively rapidly at the glass line. This causes both

structural weakening and potential glass coloura-tion problems. It would seem feasible to use aplatinum cladding to negate this weakness, butwork done a few years ago (6) showed that the ten-dency of the nickel in the ODS alloy and platinumto interdiffuse was too great for long term success.

Figure 5 shows an example of a stirrer madefrom an ODS alloy of this type. It was ACTTM

platinum coated and then laboratory tested inmolten TV glass for 300 hours at 1150ºC.Approximately the top quarter of this componenthad platinum deposited directly onto the basemetal substrate. Through-diffusion resulted in thedevelopment of surface oxide on top of the plat-inum. The lower three quarters of the sample hada ceramic interlayer which effectively blocked thediffusion, although there was slight colouration ofthe glass still attached to the sample surface. Thisindicates that iron, nickel or chromium migratedfrom the core alloy. Thus, while ACTTM coatingtechnology offers improvement, a further techno-logical advance is required to allow the effective


Fig. 4 Results for test pieces ofplatinum covered molybdenumthat have undergone air oxidationfor 160 hours at 960ºC followedby 724 hours at 1300ºC.

These molybdenum samples canbe seen to lose weight even whencoated with platinum

Fig. 5 A glass stirrer which has an ACTTMplatinum coating on top of an oxide dispersionstrengthened nickel-based superalloy which isvery similar to PM 2000 alloy.

Approximately the first quarter (on the right ofthis component) had platinum depositeddirectly onto the base metal substrate. On theremaining sample the platinum coating wasseparated from the nickel substrate by anoxide interlayer

-4

-3.5

-3

-2.5

-2

-1.5

-1

-0.5

0

0.5

0 50 100 150 50 100 150 200 250 300 350 400 450 500 550 600 650 700

TIME, hR

ELA

TIV

E W

EIG

HT

LO

SS

use of the ODS materials.Innovative technology has now been developed

and patented (7), and the design and performanceof a stirrer made using it is described below.

The Diffusion ChokeThe process of melting glass and forming it into

high quality shapes requires stirrers that can oper-ate reliably for long periods in the temperaturerange 1000 to 1300ºC. Technical solutions do exist,but all have limitations either in performance orcost. These include the inherent limitations indesign embodied in ACTTM-coated ceramic stir-rers, or the cost and inherent potential forcatastrophic failure of the extremely strong plat-inum-clad molybdenum stirrers. The diffusionchoke was designed as an alternative to the latter.Elimination of the risk of catastrophic failure can

allow for some potential reduction in the usualplatinum cladding thickness, and hence somemodest cost reduction.

The technology was tested by a stirrer with aPM 2000 stirrer shaft and a 20%Rh-Pt sheet metalfabrication or cladding. A mesh or gauze of finelyknitted 10%Rh-Pt alloy, the diffusion choke, wasplaced between the two materials, see Figure 6.Advantages of diffusion choke technology are:• The diffusion choke is designed to separatecladding from the substrate, and thus reduce thediffusion that arises from contact at high tempera-ture causing the problems seen in Figure 5.• The diffusion choke is designed to maintainan airway to the outside and ensure that adequateamounts of oxygen reach the surface of the core.This enables the inherent oxidation properties ofthe ODS alloy to develop and remain throughout


Fig. 6 The typical structure of a knitteddiffusion choke gauze or mesh made from10%Rh-Pt alloy. Here it is wrappedaround the core of a stirrer

Fig. 7 Schematic diagram of atypical helical bladded stirrer.The position of a diffusionchoke in the form of knittedgauze is indicated in blue. Thiswould then cover the stirrercore before being coated withplatinum

prolonged operation at elevated temperatures. • The diffusion choke restricts oxygen flow tothe alloy surface, and ensures that excessive oxidethickness cannot develop. Under certain condi-tions this might otherwise give rise to anaggressive form of rapid oxidation.• An interesting further advantage results fromthe platinum of the diffusion choke being in con-tact with the core alloy, which has a known effectof increasing oxide stability (8).

A typical stirrer design is shown in Figure 7.This was tested in glass for TVs. The initial test

period was extended from 6 to 20 months, andthough still performing well, the stirrer wasremoved for investigation. Initial visual examina-tion indicated no surface degradation at any pointalong its length. Figure 8 shows sections of thestirrer prior to further dismantling. The dis-colouration at the upper end of the shaft inSection A was crusty and mineralised, indicative ofdeposition from the furnace vapours. At the glassline the platinum alloy was slightly brighter possi-bly indicating some minor surface interaction. Amicrofocus XRF unit examined the composition


Section A

Section B

Section C

Section D

1

2

3

4

5

6

7

8

9

10 11

12

13

Fig. 8 Preliminary sectioning ofa stirrer core of PM 2000 with a20%Rh-Pt sheet metal claddingbetween which was a gauze offinely knitted 10%Rh-Pt alloy.This stirrer was used in a hostileenvironment for 20 months.

The numbers represent positionswhere analyses were performed.

The stirrer has been divided intofour sections:

Section A is the top section linkedup to the drive motor;Section B fits into Section A;Section C fits into Section B andto Section D;Section D is the stirrer blade end

Table I

Microfocus X-Ray Fluorescence Analysis of the Stirrer Surface

Sample Pt, % Rh, % Fe, % Cr, % K, % Ca, % Sb, % As, % Bi, % Ba, %

1 27.4 19.7 0.7 44.1 7.5 0.62 14.5 27.7 0.2 5.2 21.3 31.0 0.13 77.2 17.6 1.1 3.4 0.74 77.9 18.8 2.5 0.75 79.7 19.3 1.06 80.2 19.2 0.67 80.4 19.68 80.1 19.5 0.39 80.1 19.910 80.3 19.711 80.4 19.612 80.9 19.113 80.7 19.3

of the external surfaces, see Table I.Analysis of the upper region of the stirrer, posi-

tions 1 to 6 showed that the yellowish encrustation

was derived from the molten glass probably viacondensation from the gas phase. This is, ofcourse, normal and expected. The rhodium con-tents of the alloy are exactly as the original alloyspecification, within the error expected for theanalytical equipment. On the lower portion of thestirrer there were no foreign elements detected onthe alloy surface except for a trace of barium, aglass component, in the region of the glass line.This lack of any surface impurities after immersionin molten glass indicates that the component hadbeen quite well cleaned before being returned forinvestigation, and thus the lack of evidence ofthrough-diffusion from the substrate was stillunproven. Further disassembly of the stirrer wastherefore needed.

The cladding on Section A slid easily from thebase metal shaft. The diffusion choke was

retained within the 20%Rh-Pt tube, butfurther examination showed this was byvery slight adhesion and minimal ten-sion was required to remove the gauze.The same situation was found for thewhole length of shaft (except, ofcourse, where the fixing screws hadbeen securely positioned to transfertorque on the shaft to the cladding andhence to the stirrer blades themselves).Along most of the length of the shaftthe gauze remained shiny and metallic,but in one region there was some grey-ness and at the very top some gauzewas yellow/brown.

Analyses of the inside of the tubeand of gauze samples at points corre-sponding to the external analyses aregiven in Tables II and III, respectively.

Analysis of the inner surfaces of the20%Rh-Pt protective tubing showedthat no elements were present thatwould not have been present in theoriginal alloy, with the possible excep-tion of one sample below the glass line.Interestingly, the rhodium level in theinside surface of the tube showed anapproximate 3% reduction from theoriginal bulk alloy composition, as did


Table II

XRF Results for the Inner Rh-Pt Tube Surface

Sample Pt, % Rh, % Fe, % Cr, %

1i 79.8 20.22i 79.1 20.93i 83.7 16.34i 83.4 16.65i 83.4 16.66i 82.5 17.58i 82.7 17.39i 82.6 17.410i 82.4 17.611i 82.9 16.9 0.2

Table IV

Microfocus XRF Analysis of the Substrate Core Surface

Sample Pt, % Rh, % Fe, % Cr, % Y, % Ti, %

14 1.0 0.1 79.1 17.8 0.8 1.115 5.8 0.5 73.6 18.2 0.8 1.116a (light) 6.1 0.5 72.9 18.5 0.7 1.216b (dark) 2.5 0.3 75.1 18.5 0.7 2.917 22.6 1.7 56.3 17.7 0.7 1.018 7.4 0.7 72.7 17.1 1.1 0.9

Table IIIa

Analysis of the Outer Surface of the Diffusion Choke Gauze


1g 88.3 11.76g 87.8 12.28g 86.1 13.911g 85.8 14.1 0.1

Table IIIb

Analysis of the Inside Surface of the Diffusion Choke Gauze


1gi 89.6 10.46gi 87.7 11.2 0.3 0.1 0.78gi 85.7 13.7 0.2 0.411gi 85.9 13.0 0.2 0.5 0.3

the average outer surface composition. Analysis ofthe diffusion choke showed where the rhodiumhad gone. This showed a corresponding increasein rhodium content indicating that either a diffu-sion process or a vapour phase transfer processhad been operating. In addition to an increasedaverage rhodium content the diffusion chokegave measurable levels of iron, chromium and tita-nium on the side in contact with the base metalsubstrate, but almost none on the side in contactwith the Rh-Pt tube. The surface of the choke incontact with the core was slightly discoloured, andappeared to have physical contamination ratherthan a chemically-bonded contamination.

PM 2000, the core of the stirrer, is an iron-based ODS alloy with major additions ofchromium and aluminium plus various otherminor ones. The key to its high strength at elevat-ed temperature is the presence of yttria which, as adispersed stable oxide, provides grain boundarystrengthening. Table IV shows the results formicrofocus XRF of the surface of the core PM2000 alloy after service. The absence of values foraluminium is linked to the analytical technique,rather than being a mechanistic issue. Alternateanalytical techniques can be used to confirm thataluminium has been retained.

The presence of the occasional high values forplatinum on the surface of the base metal core wasdue to very small adhered flecks of platinum. Thenature of this tiny platinum-rich particulate has notbeen determined, so it is impossible to say whetherthey have been transported by a vapour phasemechanism or are simple physical artifacts. Visualobservation, however, clearly indicated that a thin,protective surface oxide had been formed. Thiswould be expected to change only slowly allowingprotection to the substrate for a very long time.

ConclusionsThe diffusion choke maintained an effective

barrier to degradation of the stirrer for 20 monthsservice. Indeed, the analyses suggests that thecomponent would have maintained integrity for amuch longer time perhaps comparable to the max-imum life of clad molybdenum of 5 to 10 years.

The in-service trial and subsequent destructive

analysis of the 20%Rh-Pt clad, ODS iron-basedalloy stirrer reported here demonstrates that thereis new technology to overcome many of the prob-lems associated with traditional clad-molybdenumstirrers. The technology offers a breakthrough instirrer design and thus additional help to the glass-maker when using stirrers for improving glassquality. In this trial the stirrer design was simple,with reliance on traditional fabrication skills in itsconstruction. Diffusion choke technology haspotential for use in a wider range of stirrer types,and perhaps additional applications, where highstrength, durability and longevity, without risk ofcatastrophic failure, are paramount.

References1 D. R. Coupland, Platinum Metals Rev., 1993, 37, (2),

622 D. R. Coupland, R. B. McGrath, J. M. Evens and J.

P. Hartley, Platinum Metals Rev., 1995, 39, (3), 983 D. R. Coupland, J. M. Evens and M. L. Doyle, Glass

Technol., 1996, 37, (4), 1084 G. L. Selman, Platinum Metals Rev., 1967, 11, (4), 1325 A. S. Darling and G. L. Selman, Platinum Metals Rev.,

1968, 12, (3), 926 Johnson Matthey Noble Metals, internal communi-

cation, 19917 Johnson Matthey PLC, World Appl. 03/059,826;

20038 C. W. Corti, D. R. Coupland and G. L. Selman,

Platinum Metals Rev., 1980, 24, (1), 2

The Authors

Duncan Coupland manages theTechnology Team of Johnson MattheyNoble Metals in Royston. He isresponsible for all technology aspectswithin the business unit. He was theoriginator of ACTTM technology, nowextensively used in the glass industry.He is interested in all aspects of themetallurgical use of the platinum groupmetals and their utilisation in industrialand scientific applications.

Dr Paul Williams has worked forJohnson Matthey Noble Metals sinceJanuary 1997, as a developmentscientist then as a product specialist forACT™ platinum coatings and fabricatedproducts for the glass industry. He isnow Johnson Matthey’s EuropeanProduct Manager for medical products.He is interested in Nitinol shape memory

alloys and platinum alloys for medical applications, includingimplantable medical devices.


Noble metals are widely used as electrodes ingas sensors because of their unique physical andchemical properties, such as their inertness, goodoxidation resistance, electrical conductivity andcatalytic performance. However, due to sluggishcharge transfer reactions at the sensing electrodeinterface at low temperature (less than 500ºC) (1),a gas sensor constructed with traditional Pt elec-trodes and ZrO2 electrolyte needs to be heated to ahigher temperature to obtain sufficient voltage out-put and a shorter response time. In order toimprove the properties of these sensors, Ir clusterfilms have been prepared by MOCVD (metal-organic chemical vapour deposition) andinvestigated (28). This paper reports on the com-position, structure and electrochemical propertiesof some Ir/C films.

Experimental ProcedureA schematic diagram of a horizontal hot-wall

MOCVD apparatus is shown in Figure 1. The pre-cursor for the Ir/C films was 500 mg of iridiumtris-acetylacetonate, (CH3COCHCOCH3)3Ir,Ir(acac)3. Oxygen and argon were used as the reac-tant and transmission gases, respectively. Thesubstrates were quartz (10 mm × 10 mm × 1 mmthick) and YSZ (yttria stabilised zirconia): 6 mol %Y2O3, (10 mm Φ × 2 mm thick). The temperatureof the precursor (Tsor) was kept at 190ºC. The totalgas pressure in the chamber was fixed at 500 Pa,with argon flow maintained at 50 ml min1. Theprecursor was placed in a small quartz boat in theMOCVD apparatus. The deposition temperature(Tdep) was varied from 450 to 650ºC, for a deposi-tion time of 60 minutes. The flow of oxygen (FO2)


DOI: 10.1595/147106705X45631

Iridium/Carbon Films Prepared by MOCVDOBSERVATIONS AND ELECTROCHEMICAL PROPERTIES RELATING TO OXYGEN ADDITIONS

By Changyi Hu* and Jigao WanKunming Institute of Precious Metals, Kunming, Yunnan 650221, China; *E-mail: [email protected]

and Jiaoyan DaiInstitute of Materials and Engineering, Central South University, Changsha, Hunan 410083, China

Iridium/carbon (Ir/C) films were prepared by MOCVD using iridium acetylacetonate as theprecursor and some electrochemical properties were studied, in particular the effects of oxygenon the carbon content of the Ir/C films. Small additions of oxygen (4 ml min1) to the sourcegas drastically decrease the carbon content of the films. Ir grains are formed, up to ~ 3 nmin diameter, in the amorphous carbon. It was found that Ir/C films with higher carbon contenthave better catalytic performance for measuring the oxygen concentration than Ir/C filmswith lower carbon content. The Ir/C films were used as electrodes in an oxygen concentrationcell, and the sensitivity of the cell to oxygen was recorded. The Nernstian electromotiveforce of the cell is almost the same as that of a similar type of commercial oxygen concentrationsensor from Bosch, but the response time is faster.

Furnace FurnaceManometer

Vacuum pump

PrecursorArgon

OxygenSubstrate

Fig. 1 Schematic diagram of chemical vapour deposition equipment used for the preparation of Ir/C films

was varied from 0 to 10 ml min1.The composition of the deposits was analysed

by X-ray photoelectron spectroscopy (XPS). Theexciting source of the XPS is Al (Kα), the sensi-tivity factors are 4.4, 0.25 and 0.66 for Ir, C and O,respectively. The film structures were also investi-gated by XRD and SEM.

Figure 2 shows a schematic diagram of themeasurement of the Nernstian electromotive force(e.m.f.) of the oxygen concentration cell having anIr/C electrode attached to both sides of a YSZsolid electrolyte. Values of the e.m.f. were mea-sured by changing the partial pressure of theoxygen at temperatures from 300 to 600ºC.

A dynamic test apparatus (9) was used to assessthe performance of the oxygen sensor. The air and

fuel (natural gas) were adjusted to obtain thedesired λ values (normalised air/fuel ratios). Theexhaust gas was usually maintained in a rich con-dition, at λ = 0.95. A solenoid valve allowedadditional air to the burner to switch the exhaustcomposition quickly to lean, when λ = 1.05, thencutting off the additional air and switching back torich.

The sensor voltage output was measured by avoltmeter having an input impedance of 107 Ω.The voltage switching response was determinedusing an oscilloscope, also with an input imped-ance of 107 Ω connected in parallel to thevoltmeter. The response time was defined as thetime taken for the output voltage, recorded on theoscilloscope, to sweep between 600 and 200 mV.


Fig. 2 Schematic diagramof apparatus to measuree.m.f. values of an oxygenconcentration cell. The cellhas YSZ solid electrolyteand an Ir/C electrode

Fig. 3 XPS spectra before argon sputtering (lower curves) and after argon sputtering (upper curves) for Ir/C filmsprepared: (a) without oxygen addition and (b) with oxygen addition. B.E. is the binding energy

(b)(a)

600 500 400 300 200 100 0

B.E., eV600 500 400 300 200 100 0

B.E., eV

1#

1#k

6#k

6#

Ir4d

O1s C1s

Ir4f

Ir4p3

Ir4fIr4d

O1sIr4p3

280000

240000

200000

160000

120000

80000

40000

0

120000

90000

60000

30000

0

INT

EN

SIT

Y, c

ps

Furnace

Volt-ohm-milliammeter

Thermocouple

Oxygenflowmeter

Oxygengas

Argon flowmeter

Argon gas Oxygen gas

Oxygenflowmeter

Sample

C1s

ResultsComposition of Ir/C Films

Figure 3 shows the XPS spectrum before andafter argon sputtering (5 kV, 2 min) for Ir/C filmsprepared on quartz substrates with and withoutoxygen addition. Before argon sputtering, carbonwas observed at the surface of the two samples.After sputtering, no observable signals from car-bon were detected for the Ir/C films prepared withoxygen addition (trace 6#k), but signals of carbonwere observed from films prepared without oxy-gen addition after sputtering (trace 1#k).

The effects of oxygen and temperature on thecontents inside Ir/C films prepared on quartz areshown in Table I. The addition of oxygen (from 0

to 4 ml min1) is seen to decrease the carbon con-tent and thus increase the iridium content. There isan increasing trend of carbon content inside thefilms prepared without oxygen addition withincreasing deposition temperature.

Films obtained with the addition of oxygenwere smooth with silver-coloured surfaces: due tothe oxygen reacting with carbon. The reactionproducts (carbon dioxide or carbon monoxide) areexhausted from the deposition chamber.

Structure of Ir/C FilmsFigure 4 shows the surface appearances and ele-

mental maps of Ir/C films prepared without andwith oxygen addition. Film prepared without oxy-


Fig. 4 Elemental maps of carbon for Ir/C films. Top map: Film prepared at 650ºC without oxygen addition.Bottom map: Film prepared at 600ºC with 4 ml min1 oxygen addition

gen addition is seen to have a higher carbon con-tent than film with 4 ml min1 oxygen addition.The carbon is dispersed in the grain boundaries ofthe iridium.

Figure 5 shows characteristic X-ray patterns ofIr wire and Ir/C film prepared under differentoxygen flows. The height of the peak increaseswith increasing oxygen flow, indicating that thecarbon content of the films is decreasing. The X-ray peaks of Ir/C films are displaced in the samedirection, comparable to the X-ray peak of the Irwire. This indicates that the states of carbondeposited in these films are the same.

The peaks in Figure 5 starting at the highestrepresent: standard Ir wire; Ir/C films prepared

with oxygen additions of 10, 8, 4, 0 ml min1,respectively.

Figure 6 shows characteristic XRD patterns ofthe Ir/C films prepared under different depositionconditions. The Ir/C film with higher carbon con-tent has lower broader XRD peaks. Based oncalculations from the half-width of the X-raypeaks, the Ir grains are ~ 3 nm in size, Fig. 6(a),consistent with direct observations by TEM (6).


Table I

Composition of Ir/C Cluster Films Prepared under Different Deposition Conditions

Tdep 500ºC 550ºC 600ºC 650ºC

Flow O2, ml min–1 0 4 0 4 0 4 0 4

Ir, wt.% 89.5 98.6 82.5 94.9 83.6 97.4 66.9 98.8C, wt.% 9.8 0 17.2 4.7 15.1 2.1 32.2 0.6O, wt.% 0.7 1.4 0.3 0.4 1.3 0.5 0.9 0.6

Fig. 5 Characteristic X-ray patterns of Ir wire and Ir/Cfilms prepared in different oxygen flows.Peaks are: top: Ir wire, followed by Ir/C films preparedwith oxygen additions of 10, 8, 4, 0 ml min1

Fig. 6 Characteristics of the XRD patterns of Ir/C filmsprepared under different deposition conditions:(a) Film prepared at Tdep = 650ºC, FO2 = 0 ml min

1

(carbon content 32.2 wt.%)(b) Film prepared at Tdep = 550ºC, FO2 = 9 ml min

1

(carbon content 9.9 wt.%)

(a)

(b)

6.0200 6.0400 6.0600WAVELENGTH, Å

10.00 50.00 100.002θ

10.00 50.00 100.002θ

8000

7000

6000

5000

4000

3000

2000

1000

CO

UN

TS

CP

SC

PS

Fig. 9 WDS of the white granules on the surface of theIr/C film after argon sputtering

An SEM surface observation of Ir/C filmdeposited without addition of oxygen is shown inFigure 7. The granules on the surface of the film,

etched by argon sputtering, were identified bywavelength dispersive X-ray spectroscopy (WDS),see Figure 8 and 9. These figures show that theblack and white granules (in Figure 7) representcarbon and iridium, respectively. The carbon existsas an amorphous structure determined from theWDS.

Properties of Ir/C Film ElectrodesThe relationship between e.m.f. values at differ-

ent temperatures and the ratio of the oxygenpartial pressures (P1/P2) in the oxygen concentra-tion cell is shown in Figure 10. The Ir/C filmelectrodes were deposited under various condi-tions. P1 is fixed at 0.1 MPa. The theoretical valuesare calculated from the Nernstian equation (10):

e.m.f. = RT/4F ln P1/P2

where R is the gas constant, F is the Faraday con-stant and T is the absolute temperature.


Fig. 7 SEM of Ir/C film deposited without addition ofoxygen (prepared at Tdep = 550ºC, FO2 = 0 ml min

1).The upper arrow indicates a black granule and the lowera white granule formed in the Ir/C films

WAVELENGTH, Å

WAVELENGTH, Å

carbon

iridium

Fig. 8 WDS of the black granules on the surface of theIr/C film after argon sputtering

WAVELENGTH, Å

iridium

WAVELENGTH, Å

carbon100

80

60

40

20

0

100

80

60

40

20

0

100

80

60

40

20

0

10000

5000

0

6.240 6.250 6.260 6.2706.255 6.260 6.265 6.270

43.000 44.000 45.000 46.000 43.000 44.000 45.000 46.000

CO

UN

TS

CO

UN

TS

7µmSEI

The difference between the experimental andthe theoretical values may be caused by electricalleakage (11). The e.m.f. values of Ir/C films pre-pared without oxygen addition were found to behigher than those prepared with 4 ml min1 oxygenaddition. This means the catalytic response, tooxygen, of film with more carbon content (pre-pared without oxygen addition) is higher.

Lastly, the response curves of the commercialsensor (BOSCH LSH6) and the oxygen concen-tration cell constructed with Ir/C film and YSZare shown in Figure 11. The voltage outputs arealmost identical, but the response time of the cellis shorter than that of the sensor.

ConclusionsIr/C films were prepared by MOCVD using

iridium acetylacetonate as the precursor. Smalladditions of oxygen to the source gas greatly

decrease the carbon content of the films. Ir grainsare formed up to ~ 3 nm in diameter by the amor-phous carbon. Ir/C films with higher carboncontent have better catalytic performance thanIr/C film of lower carbon content. The electro-chemical properties of the oxygen concentrationcell using Ir/C films as the electrodes is almost thesame as that for a commercial sensor, but theresponse time is shorter.

A research programme is currently beingundertaken to use the Ir/C films as electrodes forcommercial sensors.

AcknowledgementsThis project was supported by National Natural Science

Foundation of China, Grant No. 50171031, and YunnanScientific Project (Program No. 2003 PY10). The authors wouldlike to thank Mr Y. Wang and Senior J. M. Yang for their helpwith sample preparation and SEM observation, respectively.

References1 T. Goto, R. Vargas and T. Hirai, Mater. Sci. Eng.,

1996, A217218, 2232 T. Goto, R. Vargas and T. Hirai, J. Phys. IV, 1993, 3,

2973 R. Vargas, T. Goto, W. Zhang and T. Hirai, Appl.

Phys. Lett., 1994, 65, (9), 10944 B. S. Kwak, P. N. First, A. Erbil, B. J. Wilkens, J. D.

Budai, M. F. Chisholm and L. A. Boatner, J. Appl.Phys., 1992, 72, (8), 3735

5 Y. M. Sun, J. P. Endle, K. Smith, S. Whaly, R.Mahaffy, J. G. Ekerdt, J. M. White and R. L. Hance,Thin Solid Films, 1999, 346, 100


Fig. 10 Relationship between e.m.f. values and the oxygen partial pressure ratio of the oxygen concentration cell

Fig. 11 Voltage-time response curves operating at 300ºC

5

10

15

20

25

30

1 2

¡ ø

¡ ñ

¡ ö

5

10

15

20

25

30

35

1 2

¡ ñ

¡ ø

¡ ö

e.

m.f.

, mV

ln P1/P2

Measuringtemperature 500ºC 600ºC

theoretical value

0 ml min 1 550ºC 650ºC

4 ml min 1 550ºC 650ºC

(a) (b) Graph (a) (b)

TIME, ms

e.m

.f., m

V

BoschCVD Ir/C

1000 2000 3000 4000 5000

1000800600400200

0

!

!"

6 T. Goto, T. Hirai and T. Ono, Trans. Mater. Res. Soc.Jpn., 2000, 25, (1), 225

7 T. Goto, T. Ono and T. Hirai, Inorg. Mater., 1997, 33,(10), 1021

8 T. Goto, R. Vargas and T. Hirai, Mater. Trans., JIM,1999, 40, (3), 209

9 C. T. Young and J. D. Bode, Characteristics of ZrO2-type oxygen sensors for automotive applications,SAE Tech. Paper 790143, Int. Automotive Eng.

Congr. and Exposition, Detroit, Michigan, Feb., 197910 E. C. Sabbarao and H. S. Maiti, in Science and

Technology of Zirconia III, Advances in Ceramics,Vol. 24B, eds. S. Somiya, N. Yamamoto and H.Yanagida, American. Ceramic Society, Westerville,OH, 1989, pp. 731747

11 R. N. Blumenthal and M. A. Seitz, in ElectricalConductivity in Ceramics and Glass, Part A, ed. N.M. Tallan, Marcel Dekker, N.Y., 1974, pp. 35178


The Authors

Professor Changyi Hu is Professor ofMaterials Science at the Research andDevelopment Center, KunmingInstitute of Precious Metals, China. Hismajor work is the preparations of filmsand coatings of precious metals andwork pieces of refractory metals byMOCVD and CVD.

Jigao Wan is a Senior Researcher in theFunctional Materials Division, KunmingInstitute of Precious Metals, China. Hiscurrent research is on oxygen gas sensorsand other sensors.

Dr Jiaoyan Dai is an engineer at theInstitute of Materials and Engineering,Central South University, China. Herinterests include CVD of precious metalfilms, catalysis of precious metals andelectronic materials.


This book is intended as an update to the orig-inal title Palladium Reagents and Catalysts Innovations in Organic Synthesis written by thesame author and published by Wiley in 1995 (1). Itis to be used in conjunction with the originalreview to cover the whole of organopalladiumchemistry, from the past to the present (mid-2003). The book gives a detailed overview of themain recent advances in organopalladium chem-istry from a synthetic organic chemists view point.

The book is organised by types of organic reac-tions that are catalysed or effected byorganopalladium reagents. The first chapter com-prises a very concise and useful summary of thebasic chemistry of organopalladium catalysis. Thisis followed by separate chapters on each type ofsynthetic reaction.

The first types of reaction to be considered areoxidative reactions with Pd(II) compounds. As theauthor states in his introduction, oxidative nor-mally refers to a reaction of Pd where the oxidationstate of the metal is increased. This chapter, how-ever, refers to oxidation in the classical organicsense, for example, the conversion of an alkene toan aldehyde catalysed by a Pd(II) compound. Thenarrative begins with the first major example ofthis reaction, the Wacker process, and proceeds tomore specific and recent examples. This chapter isdetailed and includes some important chemistrycontributed by the author himself. This is obvious-ly an area close to his heart!

Pd(0)-Catalysed Reactions of Halides and Pseudohalides

The third chapter considers Pd(0)-catalysedreactions of sp2 organic halides and pseudohalides.This is the main body of the book, comprisingroughly half of the content (325 pages). It is a good

reflection of both the weight of academic researchinto this area of chemistry, and the increasing levelof industrial interest and application.

This field is often referred to as cross-couplingreactions. The introduction to the chapter tries tomake some sense of the many variations in thistype of reaction, and each subsection describes adifferent type of coupling reaction. The chapter isorganised in a systematic chemical manner basedon the type of substrate reacted with the arylhalide. A useful addition, however, is the inclusionof the generic names for each type of reaction inthe titles and contents. This makes it easy for a syn-thetic chemist to find details on each namedreaction, for example, Heck, Sonogashira, Suzuki,Stille, Negishi and Hiyama, which is often the waycoupling reactions are referred to in practice. Oneimportant area that is included, but not named assuch in this chapter, is the area often referred to asHartwig-Buchwald amination. This reaction is list-ed as arylation of nitrogen nucleophiles andincluded in the general group of C, N, O, S and Pnucleophiles.

This chapter reviews each type of couplingreaction well, with some mention of the historicaldevelopment of the methodology and good detailsof the most recent, important contributions andmethodologies. While it does not aim to providedetails on the synthetic methods, the subject cov-erage is very thorough and the references provideample leads for practical application of the chem-istry. I believe I am reasonably well informed insome areas of coupling chemistry, and I waspleased to see all of the major recent contributionsin the specific areas in the text. Based on thisobservation, it is clear that the author has provid-ed a well-researched and comprehensive overviewof this vast chosen field of palladium chemistry.

Modern Palladium CatalysisPALLADIUM REAGENTS AND CATALYSTS: NEW PERSPECTIVES FOR THE 21ST CENTURYBY J. TSUJI, John Wiley & Sons, Ltd., Chichester, 2004, 670 pagesISBN (hardcover) 0-470-85032-9, £175.00, €262.50; ISBN (paperback) 0-470-85033-7, £60.00, €90.00

Reviewed by Mark HooperJohnson Matthey Catalysts, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.; E-mail: [email protected]

DOI: 10.1595/147106705X46487

78

Pd(0)-Catalysed Reactions of Allylic Compounds

The next major area reviewed is that of thereactions of allylic compounds. This is a well-established area of chemistry and has both achiraland chiral synthetic utility. There is a useful gener-al introduction to the various types of this reaction.This chapter covers chemistry from 1965 to thepresent, so there is a much to cover. It is wellordered in a logical, chemistry/reagent based sys-tem. Almost every reaction possible with an allylicsubstrate catalysed by Pd is mentioned, and thereferences provide a useful follow-up.

Other Pd-Catalysed ReactionsThe next three chapters cover reactions of 1,2-

and 1,3-dienes and methylenecyclopropanes;propargyl compounds; and alkynes and benzynes.These short chapters (of 20 to 30 pages) provide agood flavour of these less common areas of Pdchemistry and again most of the main issues andrecent advances are covered.

The final three chapters deal with alkenes, andmiscellaneous reactions, and mention palladium-catalysed reactions that the author sees as impor-tant but which do not fit in with the systematicsubject order of the main chapters. This is usefuland interesting to see glimpses of possible future

areas of important chemistry. There is also a usefulset of tables detailing a long list of the ligands men-tioned in the book. On perusal it appears that all ofthe major advances in ligand technology are there.

In conclusion, this monograph is well writtenand a very well researched review of recent years inpalladium chemistry. It provides the reader with areliable starting point for learning about and evenperforming palladium-catalysed reactions. It is def-initely worth the investment.

The author has succeeded in completing his aimto cover the whole of organopalladium chemistry,in a systematic and logical manner. The book canact as a valuable learning tool and reference pointto release the potential of the wealth of palladiumchemistry that is now available.

The only criticism, from my point of view,could be that the book does not try to comparevarious contributions to the fields of chemistry interms of their actual usefulness to the practical orindustrial chemist. However, the author is to becongratulated on taking on such a massive task andin his success in making some sense of the vastexplosion in palladium-catalysed chemistry overrecent years.

Reference1 M. V. Twigg, Platinum Metals Rev., 1996, 40, (3), 126

The Reviewer

Mark Hooper is a Senior Development Chemist in the Catalyst Development Department, at Johnson Mattheyin Royston, U.K. He holds a B.A. (Hons), chemistry, and a D.Phil., in organometallic chemistry from OxfordUniversity. From 2000–2002 he held a post-doctoral position with Professor John Hartwig at Yale University,working on palladium catalysed amination. He joined Johnson Matthey in 2002. He is interested in novelhomogeneous catalysts, especially Pd catalysts for coupling chemistry and anchored homogeneous catalystsand has worked with Smopex for the recovery/ separation of precious metals.

Platinum Metals Rev., 2005, 49, (2)


The potential utilisation of fission-producedplatinum metals (fission platinoids, FPs) as valu-able products has attracted attention in the last fewdecades, as large amounts of spent nuclear fuelhave accumulated worldwide. One metric ton ofspent fuel, at a burn up of 33 GWd/t (gigawattdays per metric ton) contains > 1 kg palladium(Pd), > 400 g rhodium (Rh) and > 2 kg ruthenium(Ru) (1). Indeed, by 2030 spent nuclear fuel couldsupply up to 1000 t Pd and 340 t Rh. This wouldbe a considerable addition to the yield from natur-al sources.

FPs can be isolated from radioactive wastes thatoriginate in the reprocessing of spent fuel by thePurex process. The FPs are contained mainly inthe solid residue left after the dissolution of thefuel at an early processing stage, and in the aque-ous waste stream emerging from the first processcycle (high-level liquid waste). Processes for therecovery of FPs from both fractions have beendeveloped worldwide (2).

The purification of FPs during recovery can beso effective that the radioactivity of the fissionproducts left is compatible with safety regulations.However, there remains the intrinsic radioactivity

of the isolated FPs. Fission Pd contains 17 wt.% ofthe radioactive isotope 107Pd (half-life t½ = 7 × 106

years). Besides, fission Pd only contains stable iso-topes with atomic masses 104 (17 wt.%), 105 (29wt.%), 106 (21 wt.%), 108 (12 wt.%) and 110 (4wt.%). 107Pd is a soft beta emitter (maximum ener-gy, Emax = 0.035 MeV), but the radiation intensityat the surface of a foil of fission Pd metal (0.2 mgcm2) is 520 Bq cm2 (3) and this is higher than per-mitted by safety regulations in most countries. Thespecific beta radioactivity was compiled as 1.7 ×106 Bq g1 (1), while 2.6 × 106 Bq g1 was foundexperimentally (3).

The intrinsic radioactivity of fission Rh and Rumay be a more serious problem. Fission Rh con-sists almost exclusively of the stable isotope 103Rhand trace mass fractions of the isotopes 102Rh (t½ =2.9 years) and 102mRh (t½ = 207 days). Electron cap-ture is the exclusive decay mode of 102Rh and it isthe main decay mode of 102mRh, which also is abeta and positron emitter and undergoes an inter-nal transition. The gamma radiation of the isotopesis rather energetic (0.47 to 1.1 MeV). Radioactivedecay can reduce the radioactivity to an acceptablelevel after a suitable, indeed long storage time (≥ 30

DOI: 10.1595/147106705X35263

Potential Applications of FissionPlatinoids in IndustryBy Zdenek Kolarikretired from Forschungszentrum Karlsruhe, POB 3640, 76021 Karlsruhe, Germany

Present address: Kolberger Str. 9, 76139 Karlsruhe, Germany; E-mail: [email protected]

and Edouard V. RenardAll-Russian Institute of Inorganic Materials, 123060 Moscow, Russia

Amounts of fission-generated platinoids, as recovered from high-level liquid radioactive wastes,could considerably supplement amounts of metals claimed from natural sources. Of particularinterest are fission palladium and rhodium, which can be decontaminated from other fissionproducts to a non-hazardous level. What remains is intrinsic radioactivity which is weak infission palladium, and which in fission rhodium decays to an acceptable level after 30 years.The intrinsic radioactivity should not play a negative role when fission platinoids are appliedto nuclear technology. Some non-nuclear applications of fission platinoids may be possible,if irradiation and contamination of personnel as well as uncontrolled release of the platinoids,are avoided.

years). The specific radioactivity of isolated Rhafter a 5 year storage is ~ 107 Bq g1 (1).

Fission Ru exhibits higher intrinsic radioactivitythan Rh, caused by the isotopes 103Ru (0.0036wt.%, t½ = 39 days) and 106Ru (3.8 wt.%, t½ = 1.02years). 103Ru emits beta particles with Emax = 0.76MeV and little gamma radiation (0.050.61 MeV),and decays to stable 103Rh. 106Ru is a soft beta emit-ter (Emax = 0.039 MeV), which is in equilibriumwith 106Rh (t½ = 30 seconds), a hard beta emitter(Emax = 3.54 MeV), also releasing some gammaradiation (0.510.62 MeV). The stable isotopes are99Ru (2.4 × 104 wt.%), 100Ru (4.2 wt.%), 101Ru and102Ru (both 34 wt.%), and 104Ru (24 wt.%). Thespecific radioactivity of isolated Ru after 5 year and20 year storage has been compiled as 3 × 1011 and1 × 107 Bq g1, respectively (1).

It is clear that intrinsic radioactivity restricts theapplicability of isolated FPs. It has been suggestedthat the radioactive isotopes should be removedeither by current methods of isotope separation orby special methods. Atomic vapour laser (4) andplasma (5) separation processes are applicable toall three FPs, laser separation to remove 107Pd fromfission Pd (6) and electromagnetic separation toremove radioactive isotopes from fission Ru (7).However, all these operations would inevitablyincrease the price of isolated FPs which might notbe acceptable by the market.

In another approach (8), only stable isotopes ofPd and Rh would be obtained as final products iffission Ru was the exclusively separated platinoid,that is: beta decay of 106Ru via 106Rh would give sta-ble 106Pd, while stable 103Rh would be formed from103Ru. However, this would, of course, essentiallyreduce the yield of FPs; large amounts of Pd andRh would be left unexploited in the radioactivewaste.

Of the three FPs, Pd and Rh are most applica-ble. Fission Ru is too radioactive, due to the high106Ru content, while Ru obtained from naturalsources has a lower commercial value than eitherPd or Rh.

The intrinsic radioactivity of FPs does notrestrict their applications in fields in which it is notin conflict with safety regulations, for example, innuclear engineering. In other fields, two require-

ments must preferably be fulfilled: Irradiation and contamination of personnelmust be avoided and, uncontrolled release of the FPs radioactivitymust be suppressed to well below the legally per-mitted level.

The first requirement is fulfilled without specialprecautions in using fission Pd; the major part ofits soft beta radiation is self-absorbed in Pd itselfor in its solid support. The range of the radiationin air is 0.2 cm, and is < 0.002 cm in tissue whichis considerably shorter than the thickness of thehorny layer of human skin. Fulfilling the secondrequirement differs from application to applica-tion.

One precaution is inevitable both in nuclearand non-nuclear applications. Substances contain-ing FPs would have to be treated as radioactivematerials in common operations, such as fabrica-tion, refabrication, regeneration and disposal. Suchoperations would have to be made in correspond-ingly licensed and equipped facilities and respectsafety regulations. However, the impact of this ontotal productions costs would not necessarily be ofgreat importance.

This review outlines the potential for industrialand small scale applications of FPs. It shows thatin some applications the intrinsic radioactivitywould play no role, or a subordinate role.Elsewhere the use of FPs could be made compati-ble with safety regulations, but would not bealways practicable. Due to the critical attitude ofthe public toward nuclear technology and applica-tions, FPs could not be used in the production ofconsumer goods, even if their role in the produc-tion process was indirect and contamination of thefinal product excluded. On the other hand, the FPsmay well be used in the fabrication of products forindustrial use. Applications that are not acceptableare medical uses such as the production of bacteri-cidal and antitumour pharmaceuticals, surgicalimplants, medical equipment and jewellery.

Nuclear Technology In any applications in this field the intrinsic

radioactivity of FPs would play only a minor role.Possible applications are shown below, excluding

Platinum Metals Rev., 2005, 49, (2), bbmm 80

cold nuclear fusion which, although havingpromised to be a revolutionary source of energy,turned out to be a misinterpretation of experimen-tal results.

Structural and Special MaterialsAreas where platinoid additives improve the

properties of structural materials: In the Canadian deuterium-uranium reactor(CANDU), pressure tubes made from a Zr-Nballoy are in contact with heavy water coolant (580K, 11.1 MPa) on the inner side and with CO2coolant on the outer side. Elemental deuteriumformed in corrosion reactions diffuses towards theouter side of the tubes and weakens them due tohydrogen embrittlement. To inhibit this, the con-centration of deuterium is suppressed by oxidationbelow the dissociation pressure of Zr hydride. Pdcoating catalyses the oxidation, as shown in modelexperiments with pure Zr and hydrogen (9).Associated problems, such as oxygen corrosion ofZircaloy, catalyst deactivation, neutron absorptionby Pd and radioactive waste production wereshown to be manageable (10). A Pd layer on nickel or cobalt-based alloys andstainless steels catalyses the reaction of hydrogenwith oxygen or hydrogen peroxide in water at >150ºC. This lowers the corrosion potential ofthese materials in pressurised water nuclear reac-tors (11). A platinoid catalyses the recombinationof hydrogen and oxygen in a gas stream and,simultaneously, the decomposition of hydrogenperoxide in a water stream, when the streams arein counter-current contact (12). Embrittlement of Zircaloy cladding of oxidefuel rods by fission product cadmium is preventedby Pd (0.252.0 g kg1). The Pd can be blendedwith the bulk of the fuel, dispersed as coating onthe oxide particles before or after their pressing topellets, or applied as a coating on the inner side ofthe cladding tubes (13). The oxidation resistance of the Zircaloy-4cladding of fuel elements is increased by alloyingits surface with Pd. For example, a Pd layer (2 µm)is electroplated onto the Zircaloy-4 surface andannealed at 950ºC and < 104 Pa (14). 60Co-embedded oxide scales are formed on the

surface of stainless steel in boiling water reactors.The formation is reduced by a thin surface film ofPd, deposited either by vacuum evaporation orelectrolysis (15). Pd can be a component of shape memoryalloys, that is, materials acquiring a prescribedform when heated to transformation temperatureand restored to their initial shape on cooling. Suchalloys may be TiNiPd, sputter-deposited as a thinamorphous film and crystallised at 700750ºC (16)and Ti50Pd50xNix, especially when improved bythermomechanical treatment (17). Shape memoryalloys can be used in passive safety systems, ther-mocouplings for pipes and electric drives,equipment for repair and assembly of units, ther-momechanical drivers, dampers, flow rateregulators, thermodetectors, self-operating emer-gency systems, units and elements in electricaltransmission lines and electric contact devices(18). The Ti-0.2Pd alloy is a prospective material forthe construction of containers for solid high-levelradioactive wastes, which are to be disposed in arock salt depository. The passive layers of thematerial are adequately resistant to gamma radia-tion when it is in contact with salt brine (19).

Removal of Hydrogen Isotopes from Gasesand Liquids

Platinoid catalysed reactions can be of impor-tance in gaseous, liquid and solid phases: Tritium, free or bound in tritiated hydrocar-bons, is removed from the off-gas stream of afission reactor by conversion to tritiated water orits mixture with CO2. Catalysed by Pd, Rh or theirmixture deposited on alumina or silica, the reac-tion proceeds at 90500ºC and atmosphericpressure (20). Tritium is removed from the aqueous effluentsof a nuclear plant and directed into an aqueousconcentrate in combining the electrolysis of tritiat-ed water with the catalysed isotopic exchangereaction:

HT(gas) + H2O(liquid) = H2(gas) + HTO(liquid)

Deposited on a styrene-divinylbenzene copoly-mer, Pd catalyses the exchange less efficiently than


Pt, but it can be used in mixture with Pt (21, 22).The catalyst is prepared by agglomeration, crush-ing the agglomerate, pressing it into a cake andcutting catalyst particles from the cake (22). Hydrogen in the primary cooling circuit of agas-cooled nuclear power reactor is separated fromradioactive impurities if it is passed through a Pdalloy film. The cooling gas must be pressurised andfree from moisture and oxygen (23, 24). To prepare a catalyst for hydrogen/oxygenrecombination in a nuclear power plant, a platinoidis deposited on a porous metal, and heated at400850ºC, when it diffuses into the carrier sur-face where it forms an alloy layer (25).Alternatively, a ceramic granular material can becoated with Pd and used in a passive catalytic mod-ule, which is incorporated in a nuclear reactor forhydrogen mitigation during a core-melt accident(26, 27). Alumina beads carrying 0.5% Pd catalyse theH2/O2 recombination to 99% at 25ºC in compact-ed solid radioactive waste stored in sealedpackages. Hydrogen is formed if the waste ishumid and swells as a result of the corrosion ofaluminium, steel and zinc (28).

Hydrogen Isotope Diffusion, Trapping andCleanup

The utilisation of platinoids in the above oper-ations finds broad application in non-nuclearindustry (see later section on HydrogenProduction). Pure Pd metal, but not Pd alloys,exhibits considerable adsorption and permeationcapacity for hydrogen; the capacity decreases in theorder: Pd > Pd95Co5 > Pd90Co10 > Pd95U5 > Pd3U(300600 K, ≤ 50 bar) (29).

Hydrogen isotopes are separated from othercomponents of the gas output of a fusion reactorby permeation through Pd-Ag (75/25 wt./wt.)membranes at 350450ºC. The isotopic effects areH2/D2 = 1.72 and H2/DT = 2.06. The Pd-Ag alloyis poisoned by tritiated methane, but is regenerat-ed by heating in air (30). Efficient devices havebeen constructed and tested in the U.S.A.(Savannah River Site) (31), Russia (32) and Japan(Japan Atomic Energy Research Institute) (33, 34).The dimension and operating conditions of a per-

meator can be calculated by mathematical model-ling (35).

Other materials used in permeation membranesare Pd alloys containing 1040 wt.% Ag, 525wt.% Au, 1020 wt.% Pt or 510 wt.% Rh. Verypromising materials are Pd-Ag or Pd-Au alloyswith additions of Pt, Rh, Ru or Ir. Pd alloyed with25 wt.% Ag, Au and Ru exhibits excellent hydro-gen permeability and mechanical properties, and isalso resistant to hydrogen embrittlement andswelling and fractures caused by helium bubbleformation (36). Other applicable Pd alloys, devel-oped for non-nuclear industry, contain 1030%Ag, 0.55% Au, ≤ 2% Y, 0.22% Ru, ≤ 1% Pt and0.010.5% Al (37).

The methane poisoning is avoided in a double-function membrane reactor which incorporates aPd-Ag tube packed in a Ni catalyst bed. After thebulk of HT in the inlet gas is oxidised to HTOover a Pt catalyst, He is added, and the gas is con-tacted with the Ni catalyst which converts theHTO and CH2T2 into HT, CO and CO2 at310600ºC. A HT product and a He + CO + CO2waste stream are obtained (3840). A mathematicalmodel accounts for coupled effects of transport-limited permeation of H isotopes and variouschemical reactions (41).

Tritium can be separated from liquid Li in athermonuclear power plant by transfer through aniobium window into a helium stream at 980ºC.The Nb window is protected from oxygen attackon the He side by an electrolytically deposited Pdlayer (0.001 cm). Diffusion of Pd into Nb is pre-vented by an intermediate layer (250 nm) ofyttrium which does not form solid solutions withNb (42). Diffusion through a double-layer Zr-Pdwindow also separates tritium from liquid Li, andalso from Li alloys (for example, Li17Pb83). Li or itsalloy flows on the Zr side of the window, and apurge stream of argon and oxygen flows on theside of the Pd coating. At 450ºC tritium diffusesrapidly through the window and is recovered asT2O. Problems like reaction at the Pd surface andcorrosion deserve attention (43).

Hydrogen isotopes can be removed from fis-sion and fusion liquid coolants by pumpingagainst a partial-pressure drop. A gas containing H


isotopes as impurities is kept in a vacuum (108 Pa)or a reducing atmosphere on the lower concentra-tion side, and H isotopes are permeated through aPd or Pd-Ag (75/25) diaphragm (44) or a Pd-coat-ed Zr membrane (45) into an oxidisingatmosphere on the higher concentration side.There they are oxidised to water (600700 K,upstream pressure 0.00070.03 Pa). The mecha-nism is discussed in (46).

The behaviour of tritium on a Pd-Ag (75/25)cathodic membrane with and without a Pd blackdeposit, that is, the amount of diffused andtrapped tritium, the retrodiffusion, diffusion coef-ficient, tritium concentrations in the alloy sublayerand the diffusion layer thickness, all depend uponthe applied cathodic potential, temperature, Pd-Agmembrane thickness, presence of Pd blackdeposits and time. Without a Pd black deposit, thedouble layer capacitance is 40 µF cm2 and theapparent diffusion coefficient is 3 × 107 cm2 s1 at~ 20ºC. A Pd black deposit increases the diffusioncoefficient to 3 × 103 cm2 s1 (47).

Separation of Hydrogen IsotopesThis separation is based on isotope exchange

reactions in Pd, such as:H2 + Pd-T → HT + Pd-H

andD2 + Pd-T → DT + Pd-D.

The separation factors are:ln αHT = 284/T + 0.03

andln αDT = 114/T + 0.002,

that is, at 296 K, αHT = 2.69 and αDT = 1.47 (48),while the ranges αHT = 2.684.16 and αDT =1.471.54 are given elsewhere (49). Pelletised Pdblack can be used at 0ºC, and the separation fac-tors depend on the starting H2/D2, H2/T2, andD2/T2 ratios (50) and on the temperature (80 to100ºC) (50, 51) (see also (52)).

Chromatography is currently used to separateH isotopes. The adsorbent can be Pd deposited onalumina (51, 53) and on carbon and other supportsin frontal and displacement chromatography (51),and in twin-bed periodic counter-current flow(54). Alternatively, spongy Pd black is used in dis-placement chromatography (55) and Pd on

kieselguhr has been used in a pilot plant and a pro-duction facility constructed in the U.S.A.(Savannah River Plant) (56, 57). Pd can also be car-ried by a sulfonic acid cation exchanger (58).Problems arising from volume changes of Pdadsorbers are avoided if they are in the form ofmoulded granules containing a binder. Theabsorption rate is determined by surface reactionsat 78ºC but mainly by hydrogen diffusion inpores of the adsorbent at 0ºC. Counter-currentcontact of the gas phase with the adsorber ispreferably achieved by intermittent opening ofindependent column sections for the gas flow (see(48) and references therein).

Adsorbers consisting of Pd or Rh on alumina,kieselguhr or other suitable oxide can be coveredby a lipophilic layer (silicon resin, teflon, etc.),which is permeable to hydrogen gas and watervapour but not to liquid water. Then the isotopeexchange reactions are (59):

HD(gas) + H2O(vapour) = H2(gas) + HDO(vapour)

HDO(vapour) + H22O(liquid) = HDO(liquid) + H2O(vapour)

Other Nuclear ApplicationsDissolution of pulverised UO2 pellets in 8 M

HNO3 at 80ºC (a modified head-end operation inthe Purex process) is accelerated if they contain0.11.0 wt.% Ru, Rh or Pd. The platinoid is addedas RuO2, Rh2O3 or PdO to UO2 powder in the fab-rication of the pellets, which are sintered at 1750ºCin hydrogen (60).

Hydrogen ProductionPlatinoid-containing membranes are utilised in

non-nuclear industry for the separation of hydro-gen from other gases. Negligible amounts of FPsmight be released to gaseous products from com-pact metallic or glassy materials. The stability ofFPs dispersed as coatings on ceramic or oxidematerials would have to be checked in each case.The mechanical properties and the plasticity ofPd-containing membranes can be improved whenthey are repeatedly loaded with hydrogen, andthen unloaded (isobarically or isothermally) (61).

Membranes made from Pd alloys containing1030 % Ag, 0.55% Au, ≤ 2% Y, 0.22% Ru, ≤


1% Pt and 0.010.5% Al (B-X alloys) have beenapplied on industrial scale in the production ofhydrogen from ammonia purge gas (37). Pd alloymembranes containing 68 wt.% In and 0.51.0wt.% Ru can be used for purification of hydrogenat 400900ºC and 510 atm (62).

A Pd-Ag alloy can be spread on a thin film of γ-alumina, which is supported by a porous ceramichollow fibre (63), or the alloy or Pd can be deposit-ed electrochemically on a fine metal fabric (2080µm thread and 520 µm mesh) (64). A Pd/porous-glass membrane (65) and a γ-alumina membraneimpregnated with Pd in its bulk volume (66) sepa-rate hydrogen from nitrogen and carbon monoxide.

CatalysisIn this much used application the platinoids are

mostly contained in closed systems and, if FPs areused, the risk of personnel irradiation and contam-ination can be minimised. Release of the FPs to theproduct of the catalysed reaction can be minimisedby the catalyst preparation, and minimum losses ofthe platinoid components are, in any case, strivedfor to achieve long catalyst lifetimes. Thus, in somesystems the application of FPs could be quiteacceptable. Incidentally, the use of an intrinsicallyradioactive element (technetium) has already beensuggested: it strongly increases the catalytic activityof Pd (67).

However, using FPs in automobile catalystswould not be acceptable. The release to the envi-ronment, even if minimised, would be worthy ofconsideration, due to the broad utilisation of suchcatalysts. Data on Pt concentrations in dust, soiland sediments, biological material and naturalwaters has been published (68).

The following examples give a value to the pos-sible extent of platinoid release from a catalyst, andalso illustrate the extent of handling the weaklyradioactive FPs if they are used as catalyst compo-nents.

The lowest release of platinoids can be expect-ed from compact metal bodies or layers. Forexample, Pd-Ru alloys (8090/520 w/w) are usedas foils covered on one or both sides by a porouscopper layer (69). A Pd/Pt/Rh alloy is shaped to acomposite wire, containing in its body a fibre of a

Rh/Y alloy; this is braided into a net (70). Anotherexample is a package in which nets from two dif-ferent alloys (Pd/Rh/Ru/Pt and Pd/Pt/Au) arealternately layered (71). A film of Pd can be sup-ported by a layer of a siloxane polymer on a porouscopper membrane (72). Catalysts of the typePdZnTe0.2, PdZn, or PdZn2 are prepared by reduc-tion with formaldehyde or by metal displacement(73).

A not-so-low release of platinoids might occurfrom catalysts in which platinoids are deposited onporous pretreated oxides: most frequently γ-Al2O3,less frequently TiO2, ZrO2, SiO2, V2O5 or MoO3.They are used either without mechanical support,or on a ceramic support such as cordierite. Thepretreatment of γ-Al2O3 consists of calcining in airat 200ºC (74) or 550ºC (75, 76), after eventual ballmilling in 0.5 M nitric acid (75), or contacting withan (NH4)2SeO3 solution and drying at 50ºC (76).Zr(IV) hydroxide can be converted to a superacidby treatment with an (NH4)2MoO4 solution, dryingat 110ºC and calcining at 600ºC (77).

The pretreated oxides (mentioned in the previ-ous paragraph) are contacted with an aqueoussolution of H2PdCl4, Pd(NO3)2 or RhCl3 at roomtemperature, dried at 50120ºC and heated in air to400ºC (74, 75, 77) or 540ºC (76). The Pd(II) is thenreduced to metal by hydrogen at atmospheric pres-sure and 200ºC (77) or 400ºC (74), and the Rh(III)is reduced to metal at 0.1 MPa and 500ºC (75).

In a wet sol-gel process, Al2O3 sol, for example,is formed by reacting Al isopropoxide with hexy-lene glycol at 120ºC. A Rh(III) solution hydrolysesthe sol at 85ºC to a gel, which is then aged at 80ºC,dried at reduced pressure and heated to 600ºC inan atmosphere of air or nitrogen (78).

Again calcination is the usual final step of thecatalyst preparation if zeolites of various types andionic forms (79, 80), mordenite (81) and siliconnitride (82) are used as carriers for platinoids. Anon-calcined catalyst with encapsulated dicarbonylrhodium(I) is prepared by introducing Rh(III) intozeolite Y by ion exchange with Na(I) and heatingin a CO atmosphere at 120ºC and 1 MPa (83).

On sulfide carriers, Pd or Rh in a valency state> 0 is bound to S atoms. For example, the reactionof bis(2-ethoxyethylxanthato)palladium(II) or


tris(2-ethoxyethylxanthato)rhodium(III) withmolybdenyl dithiocarbamate at 430ºC and 6.9 kPain a hydrogen atmosphere results in the take up ofPd or Rh into highly dispersed molybdenum sul-fide (84).

Charcoal granules can be loaded with Rh(III)from an aqueous chloride solution, and the Rh(III)is converted to an oxide at 220ºC and reduced to ametal crystallite by humidified hydrogen at 325ºC(85). An iron/graphite carrier is loaded with Pd(II)from a solution of (π-C3H5PdCl)2 or π-C3H5PdP(C6H5)3 in benzene and subsequentlycalcined in air at 750850ºC (86).

An organic carrier, the styrene-divinylbenzenecopolymer HP20, is loaded with Pd from a Pd(II)acetate solution. The Pd(II) is reduced by hydro-gen at 100ºC and the obtained catalyst is soakedwith trichlorobenzene or trimethylbenzene (87).

To prepare a silicon carrier, silica gel is treatedwith dimethylethoxysilane or triethoxy(2-ethyl-3-pyridyl)silane and propylamine, or by(dimethylethoxysilylmethyl)diphenylphosphine. Asilicon polymer is formed on the surface which, ifcontacted with an aqueous solution of PdCl2,incorporates Pd(II) bound to a nitrogen or a phos-phorus donor atoms (88).

A polymeric support can also be based onsilane, silicone or carbon fluoride (89), and aporous material carrying a platinoid can be coatedby a layer of a carbon fluoride polymer which ispermeable only to gases (90).

Electrochemical TechnologyIn this area the release of FPs and risks for per-

sonnel are reduced if the FPs are contained incompact and corrosion-resistant parts of theequipment involved. These FPs can be applied tothe production of dimensionally stable anodes,cathodes for hydrogen evolution, platinised titani-um electrodes as diffusion electrodes,three-dimensional electrodes (91), electrodes forfuel cells, monocrystalline electrodes and micro-electrodes in microsensors for organics.

Examples of electrode manufacturing show theextent of manipulation with FPs, and examples ofthe corrosion rate and the lifetimes give a figurefor the FPs expected to be released.

Anodes for electrolytic reactions, made fromamorphous alloys Rh-B (75/25), Rh-B-P(70/20/10) and Rh-B-Ti (60/20/20) are represen-tative of compact electrodes. Their corrosion ratein chlorine evolution from aqueous chloride solu-tions is as low as 0.040.07 µm/year at 200 mAcm2, 1.2 V vs. SCE and 6080ºC (92). Electrodesmade from amorphous alloys, such asPd76xPtxSi18Cu6, need no activation treatment (93).

Platinoid Layer/Ti Support ElectrodesElectrodes consisting of a compact support

carrying a platinoid-containing layer are more typ-ical. In these cases Ti is often chosen as thesupport material. It is treated as follows: It is electrolytically coated with the FeSn2 alloy,immersed in a nitrate solution of Pd, Fe and Cd,and heated to 600ºC. An active layer 0.04 cm thickis formed, containing 3336% Pd. The lifetime ofsuch an anode, used to electrodeposit Zn from a 1M H2SO4 + 1 M ZnSO4 solution, is 255287 daysat 50 mA cm2, 1.6 V and 35 ºC (94). A pre-etched Ti coupon is repeatedly wettedwith a solution of RuCl3, PdCl2, Ti(C4H9O)4 andHCl in butanol, dried at 120ºC and heated to500ºC. This forms an active layer in which Pdoxide is finely dispersed in a solid solution of Ruand Ti oxides, containing 2255 mol% Ru, 0.222mol% Pd and 4478 mol% Ti. The active layer canbe top-coated with a porous layer of Ta2O5,formed by applying a solution of TaCl5 in pentanoland heating to 525ºC. The lifetime of the electrodeis 140 hours in 1.5 M H2SO4 at 50ºC and the anodecurrent density is 7.5 kA m2. In a hypochloritegenerator the electrode operated 24 days in dilutedbrine at a chloride current efficiency of 8085%(95). Ti is mechanically polished and etched by 0.2 Moxalic acid. Then it is repeatedly wetted with asolution of RuCl3, SnCl2 and HCl, then dried at50ºC and heated to 350ºC (450ºC after the lastcycle). A RuO2/SnO2 layer is formed, the compo-sition of which is controlled by adjusting theconcentration of the components in the appliedsolution. At a Ru content of 30 mass%, its maxi-mum lifetime as an anode in 0.5 M H2SO4 at 500mA cm2 and 30ºC is ~12 h (96).


Fuel Cell Catalyst ElectrodesA catalyst electrode for a fuel cell is fabricated

by forming a monoatomic layer of Pd or Rh ongold crystallites (510 nm in diameter) carried bycarbon particles. The metals are underpotentialdeposited from 110 M NaOH or KOH contain-ing 104105 M Pd or Rh (97).

Another catalytic electrode is prepared bydepositing Pd onto a sputter-etched silicon sur-face. A 13.5 MHz radio frequency voltage can beused in an argon atmosphere (0.018 torr) both forthe sputtering (500 W r.f. power input into theresulting Ar gas discharge, 30 s) and the Pd depo-sition (50 W, 5 s) (98).

Electrical Technology andElectronics

The potential acceptability of FPs in this field issimilar to that in the area of electrochemical tech-nology. Examples of applications are: Superconductivity is exhibited at < 2 K by Cr-Ru alloys containing > 17 at.% Ru (99), and hasbeen predicted to be a property of the compoundLiPdHx (100). Ru, Pd and Rh not only enhance thesuperconducting transition temperature of hightemperature superconductors but, for example,also shorten the synthesis of YBa2Cu3O7δ from 60to 10 hours, at a temperature of 880ºC instead ofat 920950ºC (101). The compound Al2Ru is a semiconductor atlow temperature, exhibiting rather anomalousdirect current conductivities of ~ 10 and 0.21 Ω1

cm1 at 300 and 0.46 K, respectively (102). The Pd-Ag alloy (70/30 w/w) does not reactwith the YBa2Cu3O7δ superconductor at 980 and1100ºC. A foil of the alloy can thus serve as a con-ductive barrier between the superconductor and asubstrate (103). A non-porous, ductile and shinycoating of a Pd-Ag alloy can be deposited elec-trolytically from a solution containing PdCl2(NH3)2or Pd(NO3)2(NH3)2, AgNO3, ammonium acetateor ammonium phosphate plus boric acid and mer-captosuccinic acid or mercaptopropionic acid plussuccinic acid monoamide (104). A Pd coating with increased microhardness can bedeposited from a solution containing PdCl2(NH3)4,ammonium sulfate and a complex of ZnCl2 with

1,3,6,8-tetraazatricyclo(4,4,1,13,8)dodecane (105),or from a solution containing the salt (RH)2PdCl4(R = tetramethylenediethylenetetramine) andammonium sulfate (106). Polypyridine complexes of Ru(NCS)2 and RuCl2are used as sensitisers in solar energy conversioncells based on TiO2 mesoporous electrodes (107).

Further special applications are molecularsuperconductors based on platinoid complexeswith organic ligands, photoelectrochemical cells,microwave components, thin film resistors, ther-mocouples, multilayer structures and superthinwire for IC-chips, amorphous soft magneticrecording materials, magnetic and photorecordingmaterials, antiferromagnetic corrosion-resistantfilms, and sandwich cermet capacitors.

Production of Corrosion ResistantMaterials

In this field as well, the FPs would be incorpo-rated in a solid phase and control of their releaseshould thus be possible. Uses may include: Up to a few per cent of Pd improves or,depending upon the steel composition and type,causes deterioration to the corrosion resistance ofstainless steels in diluted sulfuric acid (108, 109)and in solutions of hydrochloric acid or ferric chlo-ride (110). Pd can suppress a particular form ofhydrogen embrittlement (flaking) of even lowalloy steels, and it can also improve mechanicalproperties (109). Platinoids enhance the corrosionresistance of alloys by modifying the cathodic reac-tion (cathodically modified alloys) (111). Addition of ≤ 5% Pd enhances the resistance ofchromium stainless steels to high-temperaturewater that contains hydrogen (112), to pressurisedsuperheated steam at 1200ºC (109), to ≥ 90% sul-furic acid at ≤ 220ºC (113) and to air oxidation at500ºC (114) and 900ºC (115). A content of ≤ 0.7wt.% Rh improves the stability of chromium stain-less steel toward sulfuric and nitric acids (116) and≤ 0.3 wt.% Ru enhances the resistance of the fer-ritic Fe-40Cr alloy to diluted sulfuric acid (117).Laser surface alloying enhances the resistance to0.5 M HCl by forming a surface layer of fine cellu-lar dendrites containing 52 wt.% Ru (118). Passivefilms have been characterised in 0.5 M HCl at


00.2 wt.% Pd (119), and in 0.5 M H2SO4 and 0.5M HCl at 0.10.2 wt.% Ru (120). Steel 316 containing 0.5 wt.% Pd is passivatedto 1 N H2SO4 by a single cycle of hot pressing andsintering (121). A cathodic alloying additive of Pd(≤ 0.5%) improves the resistance of Cr and Tialloys in non-oxidising acids or reducing mediaand also, depending on which components arepresent, improves the resistance of multicompo-nent stainless steels in aggressive environments(122). Addition of 0.15% Pd or coating withPdO/TiO2 enhances the resistance of Ti in boilingnon-acidic NaCl and MgCl2 solutions (123). Mo-Cr alloys are resistant to inorganic and organicacids if they contain ≤ 10 wt.% Pd or Ru (124).Promising corrosion resistance in air is exhibitedby alloys Al47Ru53, Al48Ru51Y, Al44.5Ru50.5Cr5 andAl44.3Ru50.2Cr5B0.5 at 1100ºC and by alloysAl46Ru52Sc2 and Al43Ru52Sc5 at 1350ºC (125). Timetal or Ti-based alloys are resistant to acid chlo-ride brines, if they contain 0.1% Ru (126).

Surface alloying of a Pd plated Ti alloy isachieved by bombarding with Xe ions which dis-perse Pd homogeneously in the surface layer. Thissuppresses corrosion in boiling 1 N H2SO4 (127).The mechanism of the beneficial effect of Pd onthe oxidation resistance of Mo-W-Cr alloys to airand oxygen at 10001250ºC is elucidated in (128).

Miscellaneous ApplicationsA metallic and a carbon-containing material can

be joined if a Pd/Si brazing material and an activemetal (Ti, Zr, etc.) or hydride are placed betweenthe surfaces and heated in vacuum (129). Pd canbe a component of high temperature strain gaugealloys, such as Au-Pd-Cr, Au-Pd-Cr-Ni, Au-Pd-Cr-Pt-Al or Au-Pd-Cr-Pt-Fe-Al-Y (130). Otherpotential applications are hydrogen getters in vac-uum cryogenics, cryogenic temperature sensitiveelements and crucible materials for growing crys-tals at superhigh temperatures.

Conclusions[1] Although fission platinoids (FPs) separatedfrom high-level radioactive wastes will have resid-ual radioactivity, this need not be an insuperable

barrier to their industrial use in particular cases.[2] A wide range of applications where the use ofFPs might be possible has been identified. Thisincludes applications in the nuclear industry,where the materials involved are themselvesradioactive or become radioactive during opera-tion, and other applications where the impact ofthe residual radioactivity could be satisfactorilycontrolled.[3] Any industrial utilisation of FPs must meet thefollowing general criteria: The cost of separating, processing and usingthe FPs should not exceed the costs of using nat-urally derived platinum group metals. Irradiation and contamination of personnel aswell as uncontrolled release of the FPs into theenvironment must be avoided. It must be ensured that recycling of platinumgroup metals, which frequently occurs in industry,does not result in the contamination of the gener-al stock of the metals by the weakly radioactiveFPs. Especially, any risk of introducing the FPsinto materials which later can be used in medicineor jewellery must be excluded. It has to be assessedwhether this could be guaranteed by commonsafety regulations for the treatment of radioactivematerials (which in many countries have becomevery strict in recent decades) or whether addition-al safeguards must be introduced by authoritiesand efficiently established by the industry.

References1 Feasibility of Separation and Utilization of

Ruthenium, Rhodium and Palladium from High-Level Wastes, IAEA Tech. Rep. Ser., No. 308, 1989

2 (a) Z. Kolarik and E. V. Renard, Platinum Metals Rev.,2003, 47, (2), 74; (b) Z. Kolarik and E. V. Renard,Platinum Metals Rev., 2003, 47, (3), 123

3 B. N. Zaitsev, V. A. Korolev, V. P. Popik, Yu. Z.Prokopchuk and M. N. Chubarov, Radiokhimiya,1988, 30, (3), 411; Sov. Radiochem., 1988, 30, (3), 387

4 J. W. Dubrin, Proc. Symp. Waste Management 83,27 Feb.3 March 1983, Tucson, Arizona, U.S.A.,Vol. II, p. 117

5 S. N. Suchard, Proc. Symp. Waste Management 83,27 Feb.3 March 1983, Tucson, ANS/ASME,Arizona, U.S.A., Vol. II, p. 113

6 H. Yamaguchi and N. Sasao, Proc. Int. Symp. Adv.Nucl. Energy Res. Near-Future Chemistry in Nucl.Energy Field, 1516 Feb. 1989, Oarai, Ibaraki,JAERI, 1990, p. 129


7 V. P. Bochin, B. E. Gavrilov and V. S. Zolotarev,Isotopenpraxis, 1971, 7, (6), 232

8 F. J. Smith and H. F. McDuffie, Sep. Sci. Technol.,1981, 16, (9), 1071

9 A. Sawatzky and G. A. Ledoux, Proc. 2nd Int. Congr.Hydrogen in Metals, 610 June 1977, Paris, France,Pergamon Press, New York, Vol. 1, Paper 1C8

10 C. L. Stokes and R. E. Buxbaum, Nucl. Technol., 1992,98, 207

11 General Electric Co., U.S. Patent 5,135,709; 199212 Power Reactor & Nuclear Fuel Dev. Corp. and

Nippon Engeruharudo K.K., Japanese Patent Appl.55-126,502; 1980

13 General Electric Co., U.S. Patent 4,097,402; 197814 G. A. Eloff, C. J. Greyling and P. E. Viljoen, J. Nucl.

Mater., 1993, 202, 23915 T. V. Rao, R. W. Vook, W. Meyer and C. Wittwer, J.

Vac. Sci. Technol., 1987, A5, (4), 270116 P. Schlossmacher, Mater. Lett., 1997, 31, 11917 D. Golberg, Ya Xu, Y. Murakami, K. Otsuka, T.

Ueki and H. Horikawa, Mater. Lett., 1995, 22, 24118 R. R. Ionaitis, V. V. Kotov and I. M. Shchukin, At.

Energiya, 1995, 79, (4), 304; Sov. At. Energy, 1995, 79,(4), 712

19 A. Michaelis, S. Kudelka and J. W. Schultze,Electrochim. Acta, 1998, 43, (12), 119

20 Engelhard Minerals & Chemicals Corp., U.S. Patent4,178,350; 1979

21 Yu. A. Sakharovskii, M. B. Rozenkevich, B. M.Andreev, É. P. Magomedbekov, Yu. S. Pak, D. M.Nikitin, I. Z. Khairullin and V. V. Uborskii, At. Energiya,1998, 85, (1), 35; Soviet At. Energy, 1998, 85, (1), 462

22 Studiecentrum voor Kernenergie, S.C.K., U.S. Patent4,376,066; 1983

23 Mitsubishi Genshiryoku K.K., Japanese Patent Appl.49-002,000; 1974

24 Mitsubishi Genshiryoku K.K., Japanese Patent 58-15,078 B4; 1983

25 Hitachi Seisakusho K.K., Japanese Patent Appl. 57-046,198; 1982

26 K. Fischer, Nucl. Technol., 1995, 112, 5827 T. Kanzleiter and M. Seidler, Atomwirtschaft, 1995,

40, (6), 39228 M. Kelm and R. Köster, Proc. 3rd Int. Conf. Nucl.

Fuel Reprocessing Waste Management (RECOD91), 1418 Aug. 1991, Sendai, Japan, Vol. II, p. 965

29 R. Feenstra, D. G. de Groot, R. Griessen, J. P.Burger and A. Menovski, J. Less-Common Met., 1987,130, 375

30 R.-D. Penzhorn, U. Berndt, E. Kirste and J. Chabot,Fusion Technol., 1997, 32, 232

31 E. A. Clark, D. A. Dauchess, L. K. Heung, R. L.Rabun and T. Motyka, Fusion Technol., 1995, 28, 566

32 V. A. Goltsov, V. V. Latyshev, A. F. Volkov, A. L.Zvyagintsev, A. P. Kuzin, V. K. Kapyshev and P. A.Fefelov, At. Energiya, 1982, 52, (2), 135; Sov. At.Energy, 1982, 52, (2), 145

33 M. Okamoto and K. Okuno, Fusion Technol., 1995,28, 460

34 S. Konishi, H. Yoshida, H. Ohno, Y. Naruse, D. O.C

Documents

PLATINUM METALS REVIEWPlatinum Metals Rev., 2005, 49, (2), 62Œ69 62 DOI: 10.1595/147106705X45604 New Stirrer Technology for the Glass Industry LONG-TERM BENEFITS FROM THE ‘DIFFUSION