Casting Platinum Jewellery Alloys

Many platinum jewellery alloys are available,each being marketed according to applications forwhich it is suitable. For instance, platinum-5 wt.%iridium has high work hardening and is suitable formaking clasps, wire and the like, while softer plat-inum-5% palladium is used for fine casting anddelicate settings (1). Two commercially available,general purpose alloys, platinum-5% copper (Pt-5%Cu) and platinum-5% ruthenium (Pt-5%Ru),are in common use by manufacturing platinumjewellers. Jewellers who work with these two alloysreport some significant differences between them.

This paper will examine the differences in thesealloys and reasons for their preferential use. Wewill look for factors affecting their different char-

acteristics, and how working conditions can bemanipulated to optimise the performance of eachalloy in jewellery manufacture. Differencesbetween these alloys, experienced during hand-working and casting, are summarised in Table I.

Physical Properties of Pt-5%Cu and Pt-5%Ru

The relevant physical properties of pure plat-inum, Pt-5%Cu and Pt-5%Ru are summarised inTable II. The alloys are formulated in terms ofweight per cent (wt.%). Copper atoms are signifi-cantly lighter than ruthenium atoms. This meansthat the copper alloy contains a higher atomic percent (number of atoms) than the ruthenium alloy.

Platinum Metals Rev., 2005, 49, (3), 110–117 110

DOI: 10.1595/147106705X58268

Casting Platinum Jewellery AlloysTHE EFFECTS OF COMPOSITION ON MICROSTRUCTURE

By Duncan Miller, Tauriq Keraan, Penny Park-Ross, Victoria Husemeyer and Candy Lang*Centre for Materials Engineering, University of Cape Town, Rondebosch 7701, South Africa; *E-mail: [email protected]

There are numerous platinum jewellery alloys available today. Two commercially availablegeneral purpose alloys that are in common use by manufacturing platinum jewellers areplatinum-5% copper and platinum-5% ruthenium. In South Africa, the copper alloy isnormally preferred for casting because of its lower melting point and greater fluidity, whilethe ruthenium alloy is preferred for hand-working and machining, although some jewellersuse either alloy for all applications. In order to provide a scientific basis to the differencesin finish and workability, as experienced by jewellers, we set out to compare castingcharacteristics of platinum-copper and platinum-ruthenium jewellery alloys and look for anysubstantial differences between them. We also examine factors affecting their differentcharacteristics, and how working conditions can be manipulated to optimise the performanceof each alloy in jewellery manufacture.

Table I

Comparison of Some Working Attributes of Pt-5%Cu and Pt-5%Ru

Attribute Pt-5%Cu Pt-5%Ru

Hardness Slightly softer Slightly harderColour Slightly brownish-grey Shiny whiteMelting temperature Lower HigherCastability More fluid Less fluidSurface finish Susceptible to scratching Easier to polishCrucible corrosion More corrosive Less corrosiveFuming when molten Brown fumes Minimal fuming

There are 1.5 times more copper atoms thanruthenium atoms in an equivalent mass of 5 wt.%alloy. Both copper and ruthenium dissolve readilyin platinum, and so do not decrease the density ofthe actual metal substantially.

In both the cast and annealed state Pt-5%Ru isslightly harder than annealed Pt-5%Cu. The addi-tion of either of these alloying elementssignificantly increases the hardness of the plat-inum. The alloying atoms, acting like bollards on abusy pedestrian pavement, obstruct the movementof dislocations that allow planes of atoms in themetal crystals to slip past each other. This impedi-ment makes the metal more resistant todeformation and increases the hardness. The bulkhardness is also affected by the microstructure, aswill be seen later.

Pure platinum melts at 1769ºC. Alloys can meltover a range of temperatures, depending upontheir composition. As copper has a lower melting

point (1083ºC) than platinum, it lowers the melt-ing range of the alloy (to 1725–1745ºC). The muchhigher melting point of ruthenium (2314ºC) raisesthe alloy melting range (to 1780–1795ºC). Figure 1clearly shows the narrower range of solidificationat higher temperatures for Pt-Ru, compared to thebroader range of solidification at lower tempera-tures for Pt-Cu. The different compositions ofthese alloys lead to substantial differences in theirbehaviour and melting temperature ranges, andthese are described and explained here.

Casting ExperimentsA large number of casting tests were carried out

in a Hot Platinum integrated induction meltingand casting machine. This machine melts and castsup to 250 g of platinum alloy into a shallow cylin-drical flask which is rotated around a vertical axis.The invested trees consist of a central sprue withthe test pieces arranged tangentially. The test

Platinum Metals Rev., 2005, 49, (3) 111

Table II

Comparison of Relevant Physical Properties of Pure Platinum, Platinum-5 wt.% Copper andPlatinum- 5 wt.% Ruthenium (1)

Property Platinum Pt-5%Cu Pt-5%Ru

Density, g cm–3 21.4 20.0 20.7Hardness, Hv (annealed) 50 125 130Melting range, ºC 1769 1725–1745 1780–1795

Fig. 1 Diagram showing the melting ranges for Pt-Cu and Pt-Ru alloys containing up to 10wt.% of solute metal. The solidification ranges for thealloys lie between the two sets oflines. There is a narrower rangeof solidification for Pt-Ru, compared to Pt-Cu. The range forPt-Ru is at higher temperaturesthan the solidification range forPt-Cu

Pt-5%Ru

Pt-5%Cu

pieces on each tree were two simple rings and two10 × 10 grids, see Figure 2.

The testing was aimed primarily at assessing thefill and porosity that can be obtained when castingPt-5%Cu and Pt-5%Ru in a number of differentinvestment materials at different flask tempera-tures and centrifugal speeds. The investmentmaterials used were Bego’s Wirovest, Ransom andRandolph’s Astro-Vest and Hoben’s Platincast.Duplicate tests were conducted mainly in pairs,under nominally identical conditions. Table IIIshows the test schedule. Results of a comparativemetallographic study of rings from twenty five ofthese tests are reported here. An assessment of filland porosity will be reported in a second paper.

Most of the twenty-five rings were sampledtwice, by sawing through the line joining the feed-er sprue to the ring shank and also perpendicularlythrough the shank at the point furthest from thesprue. This produced two samples for each ring.These were mounted in resin, ground flat, dia-mond polished, and then electrolytically etched forexamination by microscope in reflected light.

ResultsVisual comparison of the etched metallograph-

ic sections revealed the following:[1] Grain size varied considerably between sam-ples, ranging from about 0.01 mm to over 1 mm,

with the Pt-5%Cu tending to have larger grain sizethan the Pt-5%Ru under similar casting conditions.This gave rise to a measured 10% difference in theas-cast hardness.[2] Within sample pairs for both alloys, the spruesections always had coarser grains than the shanksections.[3] All the etched Pt-5%Cu samples had a markeddendritic structure, while most of the etched Pt-5%Ru samples had an equiaxed grain structurewithout obvious dendrites, see Figure 3.

Chemical analysis of selected etched andunetched samples by energy dispersive spec-troscopy (EDS) in a scanning electron microscoperevealed the following:[a] Analysis of unetched samples showed that thedendritic cores of the Pt-5%Cu were enriched inplatinum, while the interdendritic areas wereenriched in copper by about 1 wt.% (that is, about20% variation relative to the bulk composition).This is evidence of strong chemical segregation.[b] The aggressive electrolytic etching tended todissolve the copper-rich interdendritic areas com-pletely, producing strong visual contrast with theplatinum-rich dendritic cores. The etching alsoleached copper from the platinum-rich cores. Ifthe etched samples had been analysed alone, thesegregation would have been visible but notdetectable chemically.


Fig. 2 Photograph of awax test model tree withthe central feeder sprue.After burn-out, when theinvestment is cast, itrotates about a verticalaxis, throwing out themolten alloy into the tworings and the two 10 × 10grids. The model is 110mm across.Samples were taken fromthe cross-section where thefeeder sprue meets the ringshank and at the distalpoint through the ringshank. The grids gave ameasure of fill of the casting, while inspectionof the cross-sectionsassessed the porosity ofthe rings

[c] Neither the unetched nor etched Pt-5%Rualloy showed visible dendritic growth, but therewas some compositional inhomogeneity withinindividual grains. Chemical analysis of an unetchedsample revealed a maximum difference of about0.5 wt.% Ru between platinum-rich and rutheni-um-rich areas within individual grains.

DiscussionThese results can be explained best with refer-

ence to the compositional or phase diagrams forthese two alloys. A phase diagram is a plot of alloy

composition against temperature and is useful, butnot infallible, in predicting the state of an alloyunder given heating or cooling conditions.

First let us consider how dendrites with strongcompositional segregation form in the Pt-5%Cualloy. Figure 4 shows part of the phase diagram fora platinum-copper alloy including our 5 wt.%composition.

The vertical axis represents temperature in ºCand the horizontal axis represents alloy composi-tion from 100% Pt on the left and increasing Cutowards the right. The Pt-5%Cu composition is


Table IIISchedule of Casting Tests with Pt-5%Cu and Pt-5%Ru, in Duplicate Pairs, Except Where Indicated

Investment Alloy Melt, ºC Flask, ºC Centrifuge, rpm No. of tests

Wirovest Pt-5%Cu 2050 940 600 2xPt-5%Cu 2050 600 600 2xPt-5%Cu 2050 300 600 2xPt-5%Cu 2050 950 1200 2xPt-5%Ru 1950 950 600 2x

Astro-Vest Pt-5%Cu 1850 870 800 1xPt-5%Cu 1850 870 900 2xPt-5%Cu 1850 600 900 2xPt-5%Cu 1850 300 900 2xPt-5%Cu 1850 870 1000 2xPt-5%Cu 1850 870 1200 2xPt-5%Ru 2050 870 800 1xPt-5%Ru 2050 870 900 2xPt-5%Ru 2050 600 900 2xPt-5%Ru 2050 300 900 2xPt-5%Ru 2050 870 1000 2xPt-5%Ru 2050 870 1200 2x

Platincast Pt-5%Cu 1850 900 600 2xPt-5%Cu 1850 600 600 2xPt-5%Cu 1850 300 600 2xPt-5%Cu 1850 900 800 6xPt-5%Cu 1850 600 800 2xPt-5%Cu 1850 300 800 2xPt-5%Cu 1850 900 1000 2xPt-5%Ru 2050 900 600 2xPt-5%Ru 2050 600 600 2xPt-5%Ru 2050 300 600 2xPt-5%Ru 2050 900 800 6xPt-5%Ru 2050 600 800 2xPt-5%Ru 2050 300 800 2xPt-5%Ru 2050 900 1000 13xPt-5%Ru 1950 600 600 2xPt-5%Ru 2000 900 1000 4x

marked with a vertical dashed line. The upper linesloping down to the right is the liquidus, abovewhich the metal should be fully molten. The lowerline sloping down to the right is the solidus.Between these two lines we have a range of solidi-fication, with solid metal coexisting with liquid.Below the solidus the material should be fullysolid.

Let us consider what happens when we slowlycool a fully molten alloy with composition Pt-5%Cu through its solidification range. As thecooling liquid reaches the liquidus temperature, a,it does not freeze instantaneously, but solid beginsto form. This solid has a composition correspond-ing to the solidus at this temperature, b, which isenriched in platinum. The liquid is thus enriched incopper. As the material cools further, more solidforms with a composition that tracks down thesolidus, bd. The residual liquid tracks down the liq-uidus, ac, in composition. Under equilibriumconditions, that is, if the material cools slowly

enough, there is a continuous reaction between thesolid forming and the liquid, so that by the time thelast vestige of liquid freezes all the solid has a uni-form composition.

However, in reality cooling is usually too rapidfor equilibrium to occur, and as solid forms it pro-gressively removes the element with the highermelting point, in this case platinum, enriching themelt in copper. Because of the relatively widesolidification range of the Pt-5%Cu (from1745–1725ºC) and the substantial difference inmelting points of platinum and copper, there isopportunity during cooling for chemical segrega-tion to occur. This accounts for the formation ofdendritic crystals with platinum-rich cores andcopper-rich margins.

Now, let us consider the cooling of a Pt-5%Rualloy, with reference to the relevant part of thephase diagram, see Figure 5.

Ruthenium has a higher melting point than plat-inum, so on cooling a liquid Pt-5%Ru alloy the first


Fig. 3 Comparison of the etched microstructures of distal cross-sections of 8 mm wide shanks of cast rings (25 mm indiameter). Samples of Pt-5%Ru (above) and Pt-5%Cu (below) show the coarse dendritic segregation in Pt-5%Cu, andthe finer equiaxed grain growth of Pt-5%Ru. Both casts were in Platincast investment, with a flask temperature of900ºC and rotational speed of 800 rpm. The Pt-5%Ru cast was at 2050ºC and the Pt-5%Cu at 1850ºC, in a HotPlatinum induction melting and casting machine

crystals to form at 1795ºC should be ruthenium-rich, b, and the liquid will be enriched in platinum,a. On further cooling the solid compositionshould track down the solidus towards a moreplatinum-rich composition, bd. This explains theexistence of ruthenium-rich cores to the crystals.The solidification range of Pt-5%Ru is narrower(1795–1780ºC) in comparison with Pt-5%Cu, seeFigure 1, and the cooling rate faster because of thegreater difference between the temperature of themolten metal and the much cooler investmentmaterial. This means there is little opportunity forsubstantial segregation and cored dendritic growthto take place. The rapid solidification also explainswhy the equiaxed crystals in the Pt-5%Ru rings

were much smaller than the large dendritic grainsin the Pt-5%Cu.

In both cases, as the liquid freezes the first crys-tals to form are enriched in the higher meltingpoint element. This enriches the liquid in the lowermelting temperature element. Thus, for a shortdistance away from the freezing interface, thefreezing point of the liquid is lowered below theactual temperature of the melt because of thiscompositional segregation. This constitutionalundercooling is the mechanism that allows den-drites to grow from a freezing surface into a hotterliquid.

One way to suppress the dendritic growth, andproduce a finer grained and more uniform


Fig. 4 Part of the platinum-copperphase diagram (after Savitskii et al.(2)).Hot alloy cools from the high casting temperature, through itssolidification range. At the liquidustemperature, a, solid begins to formof composition, b, enriched in Pt,leaving the liquid copper rich. Asmaterial cools further, more solidforms, tracking down, bd. At thesame time residual liquid tracksdown, ac

Fig. 5 Part of the phase diagram for platinum-ruthenium (afterHutchinson (3))On cooling hot alloy the first crystalsto form will be Ru rich, at b, leavingthe liquid Pt rich. Then, on furthercooling, more solid forms, trackingdown the solidus, bd. The Pt-richliquid tracks down ac. As the solidification range is narrow (compared to Pt-5%Cu) and thecooling rate is faster, due to the larger temperature differencebetween alloy and investment, thereis less opportunity for segregation or dendritic growth to occur

microstructure is to increase the thermal gradientor temperature difference between the freezingsurface and the still molten metal (4). This can beachieved by selecting an alloy with a narrow solid-ification range, or by casting into a mould which iscool relative to the alloy’s melting temperature.

The fact that crystal size is influenced by thecooling rate, or thermal gradient, also explains whyin all the paired ring samples studied the distalshank sections had finer grain size than the proxi-mal sprue sections. The distal shank is surroundedby relatively cool investment material and freezesmore rapidly than the sprue connection, which isin contact with the hot metal ‘reservoir’ of thesprue itself. Even if a lower metal casting temper-ature is used, there will be a difference in coolingrate of thicker and thinner sections, which isreflected in the eventual grain size.

ConclusionsThe use of relevant phase diagrams coupled

with metallurgical analysis of the microstructurecan provide useful insight into the physical charac-teristics of jewellery alloys. However, using phasediagrams alone has limitations. They allow one topredict the compositions of phases present at equi-librium for different bulk alloy compositions andtemperatures. To some extent they also enable theprediction of disequilibrium phase compositions,suggesting the possibilities for significant segrega-tion on cooling, as observed for the Pt-5%Cualloy, for instance. However, they do not allowprediction of the grain size, or the geometric rela-tionships of grains to each other or to the marginsof a casting. In order to evaluate the effects ofgrain size or grain orientation, microstructuralanalysis must be used, usually coupled with directhardness measurements.

In practical terms, the finer and more uniformas-cast grain structure of Pt-5%Ru gives rise toabout 10% greater indentation hardness. This inturn should contribute to easier polishing andenhanced scratch resistance, compared to the Pt-5%Cu alloy. We noticed no discernable differencein as-cast surface texture between the two alloys. Ithas been reported that Pt-Ru has a rougher as-castfinish (5, 6), but this may be dependent on the

investment used. The main ingredient of mostinvestments is silica, which melts at 1723ºC. Thehigher melting temperatures required for Pt-Rualloys could result in more reaction and pitting ofthe investment surface.

The wide solidification range of the Pt-5%Cualloy gives rise to significant compositional segre-gation, with a 20% enrichment of copper to theinterdendritic areas. This not only contributes toundesirable shrinkage porosity, but also imparts abrownish colour to the cast alloy because of thecopper-rich areas.

From a microstructural point of view, it appearsthat if the higher melting temperature of the Pt-5%Ru alloy can be met, by induction melting forinstance, then it is a superior alloy for casting. Thegreater uniformity in crystal size, shape and com-position enhances both the colour and thehardness. The advantage of the greater fluidity andlower melting point of Pt-5%Cu is compromisedby chemical segregation, leading to large dendriticcrystals with compositional inhomogeneity, poorcolour, and reduced hardness.

AcknowledgementsWe thank Miranda Waldron of the Electron

Microscope Unit at the University of Cape Townfor assistance with the chemical analyses. TheCentre for Materials Engineering at the Universityof Cape Town provided laboratory and office facil-ities. This research was funded by a grant from theInnovation Fund, administered by the SouthAfrican National Research Foundation.

References1 Johnson Matthey, Platinum Jewellery Alloys,

http://www.noble.matthey.com/product/2 E. Savitskii, V. Polyakova, N. Gorina, and N.

Roshan, “Physical Metallurgy of Platinum Metals”,Pergamon Press, Oxford, 1978

3 J. M. Hutchinson, Platinum Metals Rev., 1972, 16, (3),88

4 R. E. Smallman, “Modern Physical Metallurgy”, 4thEdn., Butterworth, London, 1985

5 G. Normandeau and D. Ueno, ‘Platinum AlloyDesign for the Investment Casting Process’,Platinum Guild International, U.S.A., 2001http://www.pgi-platinum-tech.com/pdf/V8N7.pdf

6 G. Ainsley, A. A. Bourne and R. W. E. Rushforth,Platinum Metals Rev., 1978, 22, (3), 78



The AuthorsDuncan Miller is aResearcher in the Centrefor Materials Engineering,University of Cape Town(UCT) and a member ofthe Hot Platinum researchteam, responsible for test-ing the newly developed induction melt-ing and casting machine.

Tauriq Keraan is an M.Sc.graduate in the Departmentof Mechanical Engineering(UCT). His research topicwas the investigation ofplatinum superalloys.

Penny Park-Ross is aResearch Assistant in theCentre for MaterialsEngineering (UCT) and amember of the Hot Platinumresearch team, responsible forlaboratory analysis of alloys.

Victoria Husemeyer is a graduatemechanical engineer (UCT) and a memberof the Hot Platinum research team,involved in testing casting.

Candy Lang is AssociateProfessor in the Department ofMechanical Engineering at theUniversity of Cape Town (UCT),and leader of the Hot Platinumresearch project, developing novelplatinum technology for the jew-ellery industry.

Aggregating materials of low molecular weightby stimuli, electrochemically or with light, has beenstudied in gels, micelles, etc., to find ways to controlfluidity, optical transmission, elasticity, and so on.Sound transmission increases molecular movementin liquids, but has not been considered suitable formolecular switching as it usually only breaks weaknoncovalent interactions between molecules. Now,scientists in Japan (1) have developed the first mol-ecule that is assembled by brief irradiation withultrasound (sonication).

A dinuclear Pd complex, 1, stabilised by an intra-molecular π-stacking, but inert to formingassociations, can instantly become jelly-like in vari-ous organic solvents upon exposure to sound. Aclear, homogeneous solution of 1 was prepared bythe reaction of Pd(OAc)2 with N,N'-bis(salicyli-dene)-1,5-alkanediamines in boiling benzene (2).When placed in various solvents and irradiated withsound (0.45 W cm–2, 40 kHz) for a few seconds, thestable sol state was completely converted to gel. Forinstance, a 1.2 × 10–2 M solution of 1 in acetone irra-diated for 3 seconds at 293 K gave a totally opaque

gel. Other organic solvents, such as CCl4, 1,4-diox-ane and ethyl acetate with 1 also gelled completelyand instantly upon presonication for 10 seconds.However, without sonication these solutionsremained stable at ambient temperature.

The resulting gels are readily converted back tothe original solution by heating at above Tgel, fol-lowed by cooling to room temperature. Thecontrollable switching can be repeated indefinitelyas the gel transition is due only to a simple confor-mation change of the complex. The aggregation ratecan be controlled between ‘no gelation’ and ‘instantgelation’ by tuning the sonication time.

Conventional self-assembly depends on staticreaction parameters, such as temperature, concen-tration, solvents and additives, but sonication givesdynamic control to the aggregation rate, so is a use-ful addition to chemistry – and possibly industry.

References1 T. Naota and H. Koori, J. Am. Chem. Soc., 2005, 127,

(26), 93242 Japan Science & Technology Corp, Japanese Appl. 2003-

261,859; 2003

Reversible Gelation of Palladium-Based Fluid by Sound

DOI: 10.1595/147106705X58961

Documents

Casting Platinum Jewellery Alloys