platinum metals · pdf file · 2016-01-28platinum metals review e-issn...

PlatinumMetalsReview

www.platinummetalsreview.comE-ISSN 1471–0676

VOLUME 51 NUMBER 3 JULY 2007

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. 51 JULY 2007 NO. 3

Contents

Building a Thermodynamic Database for 104Platinum-Based Superalloys: Part I

By L. A. Cornish, R. Süss, A. Watson and S. N. Prins

Enhancement of Industrial Hydroformylation Processes 116by the Adoption of Rhodium-Based Catalyst: Part I

By Richard Tudor and Michael Ashley

The 4th Cape Organometallic Symposium: 127Organometallics and Their Applications

A conference review by David J. Robinson

Vapour Pressure Equations for the Platinum Group Elements 130By J. W. Arblaster

“Nonporous Inorganic Membranes: for Chemical Processing” 136A book review by Hugh Hamilton

Platinum-Copper on Carbon Catalyst Synthesised 138by Reduction with Hydride Anion

By Hany M. AbdelDayem

2007 Fuels and Emissions Conference 145A conference review by Andrew P. E. York

Sir Geoffrey Wilkinson: New Commemorative Plaque 150By W. P. Griffith

New Autocatalyst Plant for South Korea 154By Carlos Silva

“Platinum 2007” 155Abstracts 156

New Patents 159Final Analysis: Why Use Platinum in Catalytic Converters? 162

By S. E. Golunski

Communications should be addressed to: The Editor, Barry W. Copping, Platinum Metals Review, [email protected]; Johnson Matthey Public Limited Company, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.

Platinum Metals Rev., 2007, 51, (3), 104–115 104

Work has been ongoing in building a thermody-namic database for the prediction of phaseequilibria in Pt-based superalloys (1–4). The alloysare being developed for high-temperature applica-tions in aggressive environments. The database will

aid the design of alloys by enabling the calculationof the composition and proportions of phases pre-sent in alloys of different compositions. Currently,the database contains the elements platinum, alu-minium, chromium and ruthenium. This paper is a

Building a Thermodynamic Database forPlatinum-Based Superalloys: Part IINTRODUCTION, AND INITIAL RESULTS FROM THE COMPOUND ENERGY FORMALISM MODEL

By L. A. CornishAdvanced Materials Division, Mintek, Private Bag X3015, Randburg 2125, South Africa,

DST/NRF Centre of Excellence in Strong Materials, Johannesburg 2050, South Africa,

and School of Chemical and Materials Engineering, University of the Witwatersrand, Private Bag 3, Johannesburg 2050, South

Africa

R. Süss*Advanced Materials Division, Mintek, Private Bag X3015, Randburg 2125, South Africa,

DST/NRF Centre of Excellence in Strong Materials, Johannesburg 2050, South Africa,

and School of Chemical and Materials Engineering, University of the Witwatersrand, Private Bag 3, Johannesburg 2050, South

Africa; *E-mail: [email protected]

A. WatsonInstitute for Materials Research, University of Leeds, Leeds LS2 9JT, U.K.

and S. N. PrinsNational Metrology Institute of South Africa, Private Bag X34, Lynwood Ridge 0040, South Africa,

and Phases Research Lab, Department of Materials Science and Engineering, Pennsylvania State University, University Park, PA

16802, U.S.A.

Work is being done at Mintek, the University of Leeds and the University of Bayreuth tobuild up a platinum-aluminium-chromium-ruthenium (Pt-Al-Cr-Ru) database for the predictionof phase diagrams for further alloy development by obtaining good thermodynamic descriptionsof all of the possible phases in the system. The available databases do not cover all of thephases, and these had to be gleaned from literature, or modelled using experimental data.Similarly, not all of the experimental data were known, and where there were gaps orinconsistencies, experiments had to be undertaken. A preliminary version of the database wasconstructed from assessed thermodynamic data-sets for the binary systems only. The binarydescriptions were combined, allowing extrapolation into the ternary systems, and experimentalphase equilibrium data were compared with calculated results. Very good agreement wasobtained for the Pt-Al-Ru and Pt-Cr-Ru systems, which was encouraging and confirmedthat the higher-order systems could be calculated from the binary systems with confidence.Since some of the phase models in earlier databases were different, these phases had to beremodelled. However, more work is ongoing for information concerning the ternary phasespresent in the Al-Cr-Ru, Pt-Al-Ru (two ternary compounds in each) and Pt-Al-Cr (possiblymore than three ternary compounds) systems. Later in the work, problems with thethermodynamic descriptions of the Cr-Ru and Pt-Cr binary systems were found, and aprogramme of experimental work to overcome these has been devised, and is being undertaken.

DOI: 10.1595/147106707X212967

Platinum Metals Rev., 2007, 51, (3) 105

revised account of work presented at the confer-ence: Southern African Institute of Mining andMetallurgy ‘Platinum Surges Ahead’ at Sun City,South Africa, from 8th to 12th October 2006 (4).

Ever since the possibility of basing a new seriesof alloys on platinum was seen (1), work has beenongoing at Mintek, Fachhochschule Jena andBayreuth University, Germany, with input fromthe National Institute for Materials Science(NIMS), Japan. Experimental work in this field istime-consuming and very expensive in terms ofequipment, materials and expertise. A number ofimportant commercial alloy systems, such as steels,nickel-based alloys, and aluminium alloys nowhave thermodynamic databases which have beenderived from copious experimental results pub-lished by experts. These databases can be usedwith appropriate software to calculate phase dia-grams, phase proportion diagrams, and Pourbaixdiagrams. They can be used instead of experimen-tation, saving both time and money. A similardatabase is being derived within this programme,so that it will facilitate further alloy development,and also be a tool to help designers to select alloycompositions and conditions in the future.However, steels, nickel-based and aluminiumalloys have been used extensively, and there aremore data and accepted phase diagrams for thesesystems than for Pt alloys. As inputs to the Pt-based database, there are fewer commercial alloys,experimental data, and very few accepted ternarysystems. There are problems even with some ofthe binary systems. Thus, part of this work includ-ed the study of phase diagrams to address the lackof data, and to use these data to compile a thermo-dynamic database. Since the basis of the alloys isthe Pt84:Al11:Cr3:Ru2 alloy, the thermodynamicdatabase will be built on the Pt-Al-Cr-Ru system.The Scientific Group Thermodata Europe(SGTE) database (5) includes all the elements andsome of the most commonly used phases, i.e.those that are industrially important, but containsfew of the required Pt phases. Additionally, theruthenium data have been updated to captureAnsara’s modification, to obtain a better estimateof the calculated melting temperature for (hypo-thetical) b.c.c.-Ru (6) than hitherto. Although there

is a database for precious metals (7) it is not suffi-ciently complete for the purposes of thisinvestigation, and does not contain all the elementsof interest to this study, or all the phases.Additionally, not all the phase descriptions neces-sary are present in Spencer’s database. The Al-Crsystem has also been independently assessed (8),although some of the phases might ultimately bemodelled a different way by this group.

The phase diagram programs (e.g. Thermo-CalcTM, MTDATA and Pandat) comprise thesoftware itself, accessed in modules through amain interface, and a series of databases where thestructural and thermodynamic data are stored. Inthese databases, each phase is described by a seriesof parameters. The SGTE database covers thephases of only the most common and well knownsystems, and all the stable elements (5). The inter-metallic phases in the Al-Ru and Pt-Al systems arenot included in the SGTE database. Providing thatthe elements are available in a database (and all ofthe stable elements are in the SGTE database, orother available databases), a phase diagram can becalculated and drawn. However, if there is nodescription for a particular phase, then the calcu-lated phase diagram cannot include it. A givendatabase can be modified to include new phases, orrun in conjunction with another database. The aimof this programme is to develop a database specif-ically for the Pt-rich alloys in this investigation.Prior to building a database, it must be knownwhich phases need descriptions. Information onthe elements, and on any phase that is alreadyincluded in the SGTE database (5), can beaccessed from that database. For phases that arenot represented by the SGTE database, a numberof factors must be taken into consideration. Firstly,the structure of the phase has to be decided,including the number of lattice sites for the atoms,and which particular atoms fit on the sites. Eachphase is modelled with sublattices, and each sublat-tice usually corresponds with a type of atomposition. Elements allowed in a particular sublat-tice are those actually found in those positions bycrystallographic measurement. This information isusually derived from X-ray diffraction (XRD)structural information and composition ranges,


and is usually made to be as simple as possible.Next, some values have to be obtained for theinteraction parameters. Initial values may beguessed, or the parameters set to zero, and the usercan decide which parameter can be varied duringoptimisation. In preparing for optimisation, exper-imental data are compared against thethermodynamic description, which is adjusted tobest fit the experimental data. A ‘pop’ file is creat-ed which contains the experimental data (these caninclude phase compositions in equilibrium witheach other at known temperatures, reaction infor-mation, enthalpies, etc.). Then optimisation can beconducted. Thermo-CalcTM uses the informationin the pop file and, through iteration, calculates theparameters required (those that were set to be var-ied) to best fit the data in the pop file. The resultof this process is the incorporation of new phases,which now have parameters that can be used tocalculate a phase diagram that agrees with theinput data.

Optimisation is the iterative process in whichselected expressions of the thermodynamicdescriptions are allowed to vary so that agreementwith the experimental results is improved. Theoptimisation was carried out with the PARROTmodule (for the assessment of experimental data,and establishment of thermodynamic coefficients)(9) of the Thermo-CalcTM software (10). With thismodule, the Gibbs energy functions can bederived by a least-squares fit to experimental data.Different types of experimental data can be used,with weightings assigned on the basis of the uncer-tainties associated with the original data. Oncecalculated phase diagrams that agree with theexperimental data are obtained, and the thermody-namic descriptions have been rationalised, thebase systems will be complete. Selected importantbinaries were optimised first, for example, Al-Ru(11) and Al-Pt (12). More work has to be carriedout on the Al-Pt system because there is nodescription for the two major Pt3Al phases. Sincethese phases are crucial to the programme, theyhave to be modelled satisfactorily, before incorpo-ration into the main database. Once each binarysystem has been modelled satisfactorily, it can beadded into the ternary systems, after which each

ternary system must be optimised individually.This is done using the experimental data, eitherdrawn from the literature, or, as was mostly thecase, derived experimentally within the pro-gramme at the University of the Witwatersrandand Mintek (for Al-Cr-Ru (13, 14), Pt-Cr-Ru (15,16), Pt-Al-Cr (17, 18) and Pt-Al-Ru (19)) or theCSIR and Mintek (for Pt-Al-Ru (20)). Only oncethe ternaries have been finalised can they be com-bined for the Pt-Al-Cr-Ru quaternary. TheThermo-CalcTM database will then be optimisedagainst some quaternary alloys that have alreadybeen made for the alloy development work (2–4).Once this stage is complete, then the other smalladditions, to improve the properties (as in nickel-based superalloys), can be included in theoptimisation. It is envisaged that the very finalstage will be focused on the optimisation of onlythe important phases: at least the cubic and tetrag-onal structures of ~ Pt3Al, (Pt), ~ Pt2Al and (Ru).Here, (Pt) and (Ru) denote combinations of fouratoms of the elements in the four-compound sub-lattice formalism (4CSF); arithmetically, Pt4 andRu4, respectively.

The Pt-Al-Cr-Ru system is optimised by study-ing the four-component ternary systems. Thereason for undertaking an optimisation of wholeternaries rather than portions of them is that thereare very few data available for the system, and anythermodynamic model needs to be valid over thecomplete range of compositions in the base sys-tem before the minor components can be added.If only a small region is to be optimised (e.g. theregion between the (Pt) and Pt3Al phases only),then it is likely that although the model would besufficiently good locally, the fit would either bevery erratic or the calculations would not be capa-ble of converging when new elements were added,or other elements added beyond their originalcompositions. (This phenomenon is well knownfor Thermo-CalcTM and has been experienced atMintek for copper additions in duplex stainlesssteels (21).) Thus, the ternary systems for the Pt-Al-Cr-Ru quaternary will be studied in full toprovide a sound basis for the computer database.The Pt-Al-Cr-Ru system is shown schematically inFigure 1.


Using the Compound EnergyFormalism Model

At the beginning of the programme, it wasassumed that the six binary phase diagrams report-ed by Massalski (22) were correct, but it wasrealised after subsequent ternary work that thisassumption was wrong. For the ternary systems,experimental work has already been completed forAl-Cr-Ru (13, 14), Pt-Cr-Ru (15, 16), Pt-Al-Ru (19,20), and is nearly complete for Pt-Al-Cr (17, 18).Some quaternary alloys have already beenaddressed (3, 4), but any new alloys chosen willprobably be located only within the Pt-rich corner.The aim is to input results from the phase diagramwork, together with enthalpies from the single-phase or near single-phase compositions from theUniversity of Leeds (23), for optimisation. Therewill also be inputs from ab initio work from theUniversity of Limpopo, South Africa, onenthalpies of formation for the Pt3Al (24) and Cr-Ru (25) phases. Additionally, the transmissionelectron microscopy (TEM) results will be utilisedin changes to modelling, especially of the ~ Pt3Alphase (26–28). The Pt-Al-Cr-Ru system needed tobe thoroughly researched through experimentalwork, so that the phases could be realisticallydescribed (to be as true to their crystallographicform as possible, so that any additional elementscould be correctly incorporated) and then opti-mised using the software. Only then can the otherelements be added to the database descriptions.These will be the additional elements, added insmaller proportions to ‘tweak’ the metallurgicalproperties of the systems. These will include at

least cobalt and nickel. Experimental work hasincluded the phase investigations alluded to above.Studies of as-cast alloys were done to determinetheir solidification reactions (13–15, 20). The solid-ification reactions and the temperatures at whichthey occur (found by differential thermal analysis(DTA)) are important inputs to the phase diagramprograms. The as-cast alloy samples were alsoheat-treated at 600ºC and 1000ºC (16), thenanalysed so that the phase compositions at knowntemperatures could be input.

Ruthenium-AluminiumInitially, a simplified version of the four-com-

pound sublattice formalism (4CSF), a version ofthe compound energy formalism model (29),which models different combinations of theatoms, was used for the RuAl phase (11). Thisworked very well for the Al-Ru system, as is shownin Figures 2(a)–(c), where the calculated diagram iscompared both with that of Boniface and Cornish(30, 31) and with a phase diagram by Mücklich andIlic (32) which was published subsequently to thecalculated work. The RuAl (B2) phase was actuallydescribed by two different models: the compoundenergy formalism (CEF), which is a simplifiedform of the 4CSF model, and is designated SL (forsublattice model) in Figure 2, as well as the modi-fied sublattice formalism (MSL), which describesthe order-disorder transformation with one Gibbsenergy function. The MSL model allowed a widerRuAl phase field (by giving more flexibility in atompositions), which is closer to experimental find-ings. The RuAl6 phase was described as astoichiometric phase (i.e. ‘line compound’, with nocomposition range), and the other intermetallicphases (Ru4Al13, RuAl2 and Ru2Al3) were modelledwith the sublattice model. The solubility of Ru in(Al) was considered negligible. The coefficientswere also within a range comparable to those ofother phases in other systems.

It will be noticed that the two experimentalphase diagrams are very similar, except for the sta-bility of the Ru2Al3 phase, and the appearance ofthe Ru2Al5 phase. Boniface had observed a similarphase, but attributed it to being a ternary phasebecause it was found only with zirconium and

Fig. 1 Schematic diagram of the Pt-Al-Cr-Ru system,showing four ternary systems and six binary systems

Pt

Cr

Al

Ru


silicon impurities (33). Differences in the experi-mental phase diagrams are due to the use ofdifferent techniques. Boniface and Cornish (30,31) studied both as-cast and annealed samples, andthe as-cast specimens showed that Ru3Al2 solidi-fied at higher temperatures than RuAl2. Annealedsamples (32) are less likely to show this feature.Since data from the experimental diagram are usedto optimise the calculated phase diagram, the lattershould agree with the former. Where there are dif-ferences, this is usually due to the mathematicalmodel not allowing flexibility, or sometimes toomuch flexibility for complex models with limiteddata. In some cases, simpler models have to be

used because there are insufficient data for all theparameters required by a more complex (butpotentially more accurate) model.

Platinum-AluminiumAt the outset of the programme, there were

two conflicting phase diagrams: those of McAlisterand Kahan (34) and Oya et al. (35). The major dif-ference, which was very important to thedevelopment of the Pt-based alloys using the ~ Pt3Al precipitates in a (Pt) solid solution, was thephase transformations in the ~ Pt3Al (γ′) phase,and the number of types of the ~ Pt3Al (γ′) phase.McAlister and Kahan (34) reported one transfor-

Fig. 2 Comparison of the Al-Ru phasediagrams: (a) calculated (11) (Courtesy ofCALPHAD; MSL: modified sublattice model;SL: sublattice model)

2700

2400

2100

1800

1500

1200

900

600

3000 0.2 0.4 0.6 0.8 1.0

Ru, mole fraction

Tem

pera

ture

, K

- - - MSL— SL

RuA

l 6

RuA

l 2

Ru 4

Al 13

Ru 2

Al 3

RuA

l

(a)

0 10 20 30 40 50 60 70 80 90 100Ru, at.%

2400

1900

1400

900

400

Tem

pera

ture

, ºC Al 5R

u

Al 13

Ru 4

Al 2R

u Al 3R

u 2

AlR

u

Fig. 2(b) Comparison of the Al-Ruphase diagrams: Experimental fromBoniface and Cornish (31) (Courtesyof Elsevier Science)

(b)


mation of the high-temperature Pt3Al phase fromL12 (γ′) to a tetragonal low-temperature variant(designated D0′c) (γ′1) at ~ 1280ºC. However, Oyaet al. (35) observed the highest transformationγ′ → γ′1 at ~ 340ºC, and gave an additional trans-formation γ′1 → γ′2 at 127ºC. Previous attempts toresolve this conundrum by scanning electronmicroscopy (SEM), XRD and DTA had beenunsuccessful, although Biggs found peaks at 311 to337ºC and 132ºC for different composition sam-ples using DTA (36). Recent work (4) using in situheating in a TEM showed that the diagram of Oyaet al. (35) was more correct, although there is a pos-sibility that very minor impurities are responsible,since the different workers sourced their raw mate-rial differently. The Pt-Al system had beencalculated by Wu and Jin (37) using the CALPHAD method, but there was a need forreassessment (12) because they considered onlyone Pt3Al phase, i.e. they did not reflect the order-ing in the Pt3Al phases. This needs to be doneusing a model that allows ordering to be calculated(described below). Wu and Jin did not include thePtAl2 or β phases, owing to a lack of experimentaldata. A study of Pt-Al-X ternaries (where X = Ru,Ti, Cr, Ni) confirmed the presence of the Pt2Alphase (19, 36). Experimental work on the Pt-Al-Ru

ternary confirmed the presence of the β phase inthe Al-Pt binary (20).

Initially, the four-compound sublattice formal-ism (4CSF) was applied. This models differentcombinations of four atoms of two different ele-ments, for example: (A) (arithmetically A4), A3B,AB (arithmetically A2B2), AB3, and (B) (arithmeti-cally B4) where at least two of these appear in asystem. This method was used for the (Pt) andPt3Al phases (12), because this model had beenused in the development of the nickel-based super-alloy database (38, 39).

However, when the 4CSF model was applied tothe Pt-Al system (12), the results were less success-ful, mainly because there were very few data, andthe system was more complex. The intermetalliccompounds Pt5Al21, Pt8Al21, PtAl2, Pt2Al3, PtAl,Pt5Al3 and Pt2Al were treated as stoichiometriccompounds. The β phase was assumed to be stoi-chiometric, since very little experimentalinformation was available, and was treated asPt52Al48. The phase diagram shown in Figures3(a)–(c) appears to agree with that of Massalski(22), which is actually from McAlister and Kahan(34), but the 4CSF model did not produce the dif-ferent Pt3Al phases. The calculated compositionsand temperatures for the invariant reactions of the

Fig. 2(c) Comparisonof the Al-Ru phasediagrams: Experimentalby Mücklich and Ilic(32) (Courtesy ofElsevier Science; EDS:energy dispersivespectroscopy; WDS:wavelength dispersivespectroscopy; referencenumbers are as cited inReference (32))

Tem

pera

ture

, ºC

0 10 20 30 40 50 60 70 80 90 100Ru, at.%

Al(R

u)

Al 6R

u

Al 13

Ru 4

Al 2R

u

Al 3R

u 2

RuA

l

Ru(

Al)

Al5Ru2

657ºC734ºC

1340ºC1420ºC

1492ºC

1675ºC

1805ºC

2334ºC

1920 ± 20ºC

[64] Annealed, EDS[64] DTA heating, EDS[64] DTA cooling, EDS[78] Quenched, WDS[78] Annealed, WDS[36] Annealed, WDS[63] As-cast, EDS[58] Annealed, EDS

~ 54 74 ~ 86

Liquid2400

2200

2000

1800

1600

1400

1200

1000

800

600

400

200

0

660

(c)


intermetallic phases are generally in good agree-ment with those reported from experiment.However, there are some areas in less goodagreement, in most cases for reasons inherent inthe models being used.

Calculated results for the congruent forma-tion of the Pt3Al phase and L → Pt3Al + (Pt)eutectic reactions are not in very good agree-

ment with the experimental diagram, as bothreactions are shifted to lower platinum compo-sitions in the calculated system. The 4CSFmodel is such that the formation composition ofPt3Al is fixed at 75 at.%, whereas the phase hasbeen reported in the literature to form congru-ently at 73.2 at.%. This non-stoichiometricformation cannot be described with the model,

Fig. 3 Comparison of Al-Pt phasediagrams: (a) calculated (12)

Fig. 3(b) Comparisonof Al-Pt phasediagrams: Experimentalfrom Massalski (22)(Courtesy of ASMInternational; LT: lowtemperature structure)(McAlister and Kahan(34).)

Tem

pera

ture

, K2100

1800

1500

1200

900

600

3000 0.2 0.4 0.6 0.8 1.0

Pt, mole fraction

Pt 5A

l 21

Pt 8A

l 21P

tAl 2

Pt 2A

l 3

PtA

l

Pt 5A

l 3

Pt 2A

l

β Pt 3A

l (Pt)

(Al)

Liquid

Pt, wt.%0 20 40 50 60 70 80 85 90 95 100

0 10 20 30 40 50 60 70 80 90 100Pt, at.%

Tem

perta

ure,

ºC

2000

1800

1600

1400

1200

1000

800

600

400

Pt 5A

l 21

Pt 8A

l 21P

tAl 2

Pt 2A

l 3

PtA

l

Pt 2A

l

Pt2Al(LT)

Pt3Alβ (Pt)

(Al)

Pt3Al(LT)

Liquid

660.452

657ºC

806ºC

1127ºC

1406

1527ºC 1554ºC 1556ºC

1468ºC13971260

1456ºC1507ºC

1769ºC

Pt5Al3

(a)

(b)


and had consequences for the temperature aswell as the enthalpy of formation for the Pt3Alphase. The symmetry and fixed compositionsassumed in the 4CSF model also made it diffi-cult to fix the eutectic reaction to lower Ptcontents in the calculation. Furthermore, thephase area of the (Pt) solid solution is too nar-row, especially at lower temperatures, althoughthe phase area for the Pt3Al phase is acceptable.However, the Pt3Al phase is ordered throughoutits phase area. The unstable PtAl3 (L12) andPt2Al2 (L10) phases, which are introducedthrough the 4CSF model, are not stable at anycomposition or temperature in the phase dia-gram, which is correct. Further work on thissystem was postponed until more data todescribe the (Pt) and Pt3Al phases had beenobtained. Currently, the Pt-Al binary is beinginvestigated at Mintek using the NovaNanoSEM, and good results are being obtained(40). The data from these alloys will be used tooptimise the Pt3Al phase in the Pt-Al binary.

Platinum-Aluminium-RutheniumThe resulting database files from the Ru-Al,

Pt-Al and Pt-Ru systems were added and the phasediagram was plotted, as an extrapolation from thebinary systems, without any ternary interactionparameters or optimisation with ternary data (41).There were problems in the calculation of isother-mal sections; these arose because the currentmodels were not sufficiently robust to allow forextension into the ternary. However, the liquidusprojection showed very good agreement with theexperimental projection (Figures 4(a)–(b)).Obviously, the two ~ Pt18Al18Ru64 and ~ Pt12Al15Ru73 ternary phases (20) were not calcu-lated, because data for these were not input. Thestability of the Pt2Al phase was calculated as toohigh in the ternary, because it solidified from themelt as a primary phase, which rendered the liq-uidus inaccurate for that region. This was probablybecause of the inadequacies of modelling the phases, which resulted in Pt2Al being shown astoo stable.

1800

1400

1000

600

200

Tem

pera

ture

, K

0 10 20 30 40Al, at.%

γ

γ ′

γ1′

γ2′

Pt 3A

l

Pt 2A

l

Pt 5A

l 3

613 K

400 K

1780 K

1829 K

1738 K1670 K

1533 K

Fig. 3(c) Comparison of Al-Pt phasediagrams: Experimental from Oya et al. (35)(Courtesy of Carl Hanser Verlag)

(c)

Chromium-PlatinumAn assessment by Oikawa et al. (42) showed

that the values for the two eutectic temperatures inthe Cr-Pt binary should be reversed when com-pared with those results (22), as shown in Figure 5.Oikawa et al.’s conclusion was initially thought to

be wrong, even considering that the original tem-perature data were very close (within 30 ± 10ºC).Thus, when work began on the Cr-Pt system, thework of Oikawa et al. (42) was ignored and the4CSF model was used on the (Pt), Pt3Cr and PtCrordered phases (43). However, more data were


X, (L

iq., R

u)

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

00 0.40.2 0.6 0.8 1.0

X, (Liq., Pt)

Fig. 4 Comparison of Pt-Al-Ruliquidus surface projections:(a) calculated by Prins et al.(41) (Courtesy of CALPHAD;X = mole fraction)

Fig. 4(b) Experimental by Prinset al. (20) (Courtesy of ElsevierScience; X: ~ Ru12Pt15Al73;T: ~ Ru18Pt18Al64)

Ru

PtAl

Al, a

t.%

Ru, at.%

(Ru)

RuAl(Pt)

Pt8Al21

Pt5Al21

RuAl3

(Al)

Ru4Al13 RuAl2

Ru2Al3

Ru4Al13

PtAl2 Pt2Al3 PtAl

Pt3Al

Pt5Al21

β

Pt5Al21 Pt8Al21 PtAl2 Pt2Al3 PtAl Pt5Al3 Pt2Al Pt3Al

Pt5Al3Pt, at.%

X

X T

(a)

(b)


required for the 4CSF model than were available,and consequently the fit was very poor (22).Subsequently, results from experimental work onthe Cr-Pt-Ru, Al-Cr-Pt and Cr-Ni-Pt ternary sys-tems also agreed with the findings of Oikawa et al.(42), and those parameters are being used untilsubsequent experimental work indicates that arevision is necessary.

Chromium-RutheniumFor the Cr-Ru system, there was no previous

assessment, and the first calculation once againused the 4CSF model. Reproducibility was poorbecause the model was too complex for the data

available. Further work and an extrapolation ofthe Pt-Cr, Cr-Ru and Pt-Ru binaries (the latterfrom Spencer’s database (7)) demonstrated ratherthat the calculations were poor (44) although thefit with the ternary Pt-Cr-Ru liquidus was reason-able. It was evident that another form ofmodelling was required. Subsequently, it wasrevealed (45) that there were problems withMassalski’s (22) phase diagram, and two alloyswere manufactured to study the sequence of reac-tions in the Cr-Ru binary. The Cr-Ru system isvery difficult to study experimentally because thediffusion rates are very low (large atoms and highmelting points), and Cr oxidised easily on

0 20 40 60 80 100Pt, at.%

2000

1800

1600

1400

1200

1000

800Te

mpe

ratu

re, º

CCr3Pt

Müller [51]Waterstrat [52]Waterstrat [52]

f.c.c.

b.c.c.

Liquid

Fig. 5 Comparison of Cr-Pt phasediagrams: (a) calculated by Oikawaet al. (42) (Courtesy of ElsevierScience; reference numbers are ascited in Reference (42))

Pt, wt.%0 20 40 50 60 70 80 90 100

2000

1500

1000

500

Tem

pera

ture

, ºC

0 10 20 30 40 50 60 70 80 90 100Pt, at.%

(Pt)

Liquid

Cr 3P

t

CrP

t

Tc

CrPt3

~ 1130ºC

1785 ± 5ºC~ 80

1500ºC1530ºC

970ºC~ 23 ~ 34

(Cr)

~ 24~ 29

~ 20

~ 13~ 10

~ 1600ºC

1769ºC

1863ºC

~ 85

Fig 5(b) Experimental (22)(Courtesy of ASMInternational; Tc is thecritical temperature)

(b)

(a)


protracted annealing, despite precautions.However, this work is ongoing.

ConclusionAlthough good results were obtained for Ru-Al

using the 4CSF representation for RuAl, there wereinsufficient data to obtain good results for theother systems since more phases were representedin each system. A different approach was neededwith simpler representation to allow for sparsedata. The use of simpler thermodynamic models totreat binary and ternary alloys will be reported inPart II of this series of papers, to be published in afuture issue of Platinum Metals Review.

1 I. M. Wolff and P. J. Hill, Platinum Metals Rev., 2000,44, (4), 158

2 L. A. Cornish, J. Hohls, P. J. Hill, S. Prins, R. Süssand D. N. Compton, J. Min. Metall. Sect. B: Metall.,2002, 38, (3–4), 197

3 L. A. Cornish, R. Süss, L. H. Chown, S. Taylor, L.Glaner, A. Douglas and S. N. Prins, ‘Platinum-BasedAlloys for High Temperature and SpecialApplications’, in “International Platinum Conference‘Platinum Adding Value’”, Sun City, South Africa,3rd–7th October, 2004, Symposium Series S38, TheSouth African Institute of Mining and Metallurgy,Johannesburg, South Africa, 2004, pp. 329–336

4 L. A. Cornish, R. Süss, A. Watson and S. N. Prins,‘Building a Database for the Prediction of Phases inPt-based Superalloys’, in “Second InternationalPlatinum Conference ‘Platinum Surges Ahead’”, SunCity, South Africa, 8th–12th October, 2006,Symposium Series S45, The Southern AfricanInstitute of Mining and Metallurgy, Johannesburg,South Africa, 2006, pp. 91–102; http://www.plat-inum.org.za/Pt2006/index.htm

5 A. T. Dinsdale, CALPHAD, 1991, 15, (4), 3176 A. T. Dinsdale (National Physical Laboratory,

Teddington, U.K.), private communication to I.Ansara (LTPCM-INPG, Grenoble), 2003

7 P. J. Spencer, The Noble Metal Alloy (NOBL)Database, The Spencer Group, Trumansburg,U.S.A., 1996

8 I. Ansara, in “Cost 507 – Definition ofThermochemical and Thermophysical Properties toProvide a Database for the Development of NewLight Alloys: Thermochemical Database for LightMetal Alloys”, eds. I. Ansara, A. T. Dinsdale and M.H. Rand, European Cooperation in the field ofScientific and Technical Research, Brussels, 1998,Volume 2, pp. 23–27

9 B. Jansson, “Evaluation of Parameters inThermodynamic Models Using Different Types ofExperimental Data Simultaneously”, TRITA-MAC-0234, Royal Institute of Technology, Stockholm,

Sweden, 198410 B. Sundman, B. Jansson and J.-O. Andersson,

CALPHAD, 1985, 9, (2), 15311 S. N. Prins, L. A. Cornish, W. E. Stumpf and B.

Sundman, CALPHAD, 2003, 27, (1), 7912 S. Prins, L. A. Cornish, W. Stumpf and B. Sundman,

‘A Reassessment of the Al-Pt System’, in “CAL-PHAD XXXI: Abstracts”, Stockholm, Sweden,5th–11th May, 2002; see B. Sundman, CALPHAD,2004, 28, (3), 221

13 D. Compton, “The Constitution of Al-Rich Alloys ofthe Al-Cr-Ni System”, Ph.D. Thesis, University ofthe Witwatersrand, Johannesburg, South Africa, 2002

14 D. Compton, L. A. Cornish and M. J. Witcomb, J.Alloys Compd., 2001, 317–318, 372

15 R. Süss, L. A. Cornish and M. J. Witcomb, J. AlloysCompd., 2006, 416, (1–2), 80

16 R. Süss, L. A. Cornish and M. J. Witcomb,‘Investigation of Isothermal Sections at 1000ºC and600ºC in the Pt-Cr-Ru System’, accepted by J. AlloysCompd.

17 R. Süss, L. A. Cornish and B. Joja, ‘An Investigationof As-Cast Pt-Al-Cr Alloys’, in “Second InternationalConference of the African Materials ResearchSociety”, Johannesburg, South Africa, 8th–11thDecember, 2003, pp. 138–139

18 R. Süss and L. A. Cornish, ‘MicrostructuralEvolution in the Pt-Al-Cr System’, in Proc. Microsc.Soc. south. Afr., Vol. 36, Port Elizabeth, 28thNovember–1st December, 2006, p. 14

19 T. Biggs, P. J. Hill, L. A. Cornish and M. J. Witcomb,J. Phase Equilib., 2001, 22, (3), 214

20 S. N. Prins, L. A. Cornish and P. S. Boucher, J. AlloysCompd., 2005, 403, (1–2), 245

21 J. Kerr, S. M. Wessman and L. A. Cornish, J. Min.Metall. Sect. B: Metall., 2001, 37, (1–2), 1

22 “Binary Alloy Phase Diagrams”, 2nd Edn., eds. T. B.Massalski, H. Okamoto, P. R. Subramanian and L.Kacprzak, in 3 volumes, ASM International, Ohio,U.S.A., 1990

References

AcknowledgementsFinancial assistance from the South African

Department of Science and Technology (DST); thePlatinum Development Initiative (PDI: AngloPlatinum, Impala Platinum and Lonmin); DST/NRFCentre of Excellence in Strong Materials andEngineering; and Physical Sciences Research Council(EPSRC) Platform Grant GR/R95798 is gratefullyacknowledged. The authors would like to thankCompuTherm LLC, Wisconsin, U.S.A., and theNational Physical Laboratory (NPL), Teddington,U.K., for the provision of the WinPhaD, Pandat andMTDATA software. This paper is published with thepermission of Mintek and the Southern AfricanInstitute of Mining and Metallurgy.


23 L. Glaner, A. Watson, L. A. Cornish and R. Süss,‘Calorimetric Measurements for a Pt-BasedThermodynamic Database’, in “CALPHADXXXIV: Program and Abstracts”, Maastricht, TheNetherlands, 22nd–27th May, 2005, p. 139; see H. A.J. Oonk, CALPHAD, 2006, 30, (2), 97

24 H. R. Chauke, D. Nguyen-Manh, D. G. Pettifor andP. E. Ngoepe, ‘First Principles Phase-Stability Studyof Pt/Pt3Al Alloys’, in “CALPHAD XXXII:Program and Abstracts”, Quebec, Canada,25th–30th May, 2003, p. 38

25 M. M. Mokagane, P. Ngoepe, B. Montanari and N.Harrison, ‘Computational Studies of Binary MetalAlloys’, 10th Annual Materials Modelling Meeting,Sovenga, South Africa, 5th–6th April, 2006, p. 6

26 A. Douglas, J. H. Neethling, R. Santamarta, D.Schryvers and L. A. Cornish, ‘TEM Investigation ofthe Microstructure of Pt3Al Precipitates in a Pt-AlAlloy’, in Proc. Microsc. Soc. south. Afr., Vol. 32,Cape Town, 3rd–5th December, 2003, p. 14

27 R. Santamarta, J. H. Neethling, D. Schryvers and A.Douglas, ‘HRTEM Investigation of the LowTemperature Phase of Pt3Al Precipitates in Pt’, inProc. Microsc. Soc. south. Afr., Vol. 32, Cape Town,3rd–5th December, 2003, p. 15

28 A. Douglas, J. H. Neethling, R. Santamarta, D.Schryvers and L. A. Cornish, J. Alloys Compd., 2007,432, (1–2), 96

29 B. Sundman, S. G. Fries and W. A. Oates, CALPHAD, 1998, 22, (3), 335

30 T. D. Boniface and L. A. Cornish, J. Alloys Compd.,1996, 233, (1–2), 241

31 T. D. Boniface and L. A. Cornish, J. Alloys Compd.,1996, 234, (2), 275

32 F. Mücklich and N. Ilic, Intermetallics, 2005, 13, (1), 5and references therein

33 T. D. Boniface, “The Ruthenium-AluminiumSystem”, M.Sc. Dissertation, University of theWitwatersrand, Johannesburg, 1994

34 A. J. McAlister and D. J. Kahan, Bull. Alloy PhaseDiag., 1986, 7, 45

35 Y. Oya, U. Mishima and T. Suzuki, Z. Metallkd., 1987,

78, (7), 48536 T. Biggs, “An Investigation into Displacive Phase

Transformations in Platinum Alloys”, Ph.D. Thesis,University of the Witwatersrand, Johannesburg, 2001

37 K. Wu and Z. Jin, J. Phase Equilib., 2000, 21, (3), 22138 (a) I. Ansara, N. Dupin, H. L. Lukas and B.

Sundman, J. Alloys Compd., 1997, 247, (1–2), 20; (b)Applications of Thermodynamics in the Synthesisand Processing of Materials, Rosemont, IL, U.S.A.,2nd–6th October, 1994, pp. 273–283

39 N. Dupin, I. Ansara and B. Sundman, CALPHAD,2001, 25, (2), 279

40 W. Tshawe, A. Douglas, B. Joja and L. A. Cornish, ‘AStudy of the Pt3Al, Pt2Al and β Phases in the Pt-AlSystem’, in Proc. Microsc. Soc. south. Afr., Vol. 35,Port Elizabeth, 28th November–1st December,2006, p. 15

41 S. N. Prins, B. Sundman, L. A. Cornish and W. E.Stumpf, ‘Thermodynamic Assessment of the Al-Pt-Ru System’, in “CALPHAD XXXII: Program andAbstracts”, Quebec, Canada, 25th–30th May, 2003.p. 35; see C. K. Pollard, CALPHAD, 2004, 28, (3),241

42 K. Oikawa, G. W. Qin, T. Ikeshoji, O. Kitakami, Y.Shimada, K. Ishida, and K. Fukamichi, J. Magn. Magn.Mater., 2001, 236, (1–2), 220 and references therein

43 U. Glatzel and S. N. Prins, ‘ThermodynamicAssessments of the Pt-Cr and Cr-Ru Systems with anExtrapolation into the Pt-Cr-Ru System’, in “CAL-PHAD XXXII: Program and Abstracts”, Quebec,Canada, 25th–30th May, 2003, p. 118; see C. K.Pollard, CALPHAD, 2004, 28, (3), 241

44 R. Süss, L. A. Cornish and U. Glatzel, ‘Comparisonof Experimentally Determined and CALPHAD-Determined Results of the Pt-Cr-Ru System’, in“CALPHAD XXXIII: Program and Abstracts”,Krakow, Poland, 30th May–4th June, 2004; see C. K.Pollard, CALPHAD, 2004, 28, (4), 383

45 X. Gu (NIMS, Japan), private communication to L.A. Cornish, June 2005

Lesley Cornish is an Honorary Professor atthe University of the Witwatersrand, SouthAfrica, and is associated with the DST/NRFCentre of Excellence for Strong Materialsand Mintek. Her research interests includephase diagrams, platinum alloys andintermetallic compounds.

Dr Andy Watson is a Senior Research Fellowin the Institute for Materials Research at theUniversity of Leeds. He has worked inexperimental and computationalthermodynamics for many years and hasinterests in lead-free solders andintermetallic phases as well as pgm alloys.

Rainer Süss is Section Head of theAdvanced Metals Group in the AdvancedMetals Division at Mintek, as well as the co-ordinator of the Strong Metallic AlloysFocus Area in the DST/NRF Centre ofExcellence for Strong Materials. Hisresearch interests include phase diagrams,platinum alloys and jewellery alloys.

Sara Prins is a research metrologist at theNational Metrology Institute of South Africa.She is undertaking Ph.D. research at thePennsylvania State University, U.S.A., whereshe is working on phase diagram and first-principles calculations.

The Authors

116

2-Ethylhexanol (2EH) is the most widely used(‘workhorse’) plasticiser alcohol, and butanols –the normal and iso isomers – are used as solventsor chemical intermediates. Both 2EH andbutanols are derivatives of butyraldehyde madefrom propylene by hydroformylation. From theearly 1940s until the early 1980s, the world’s majorproducers of 2EH and butanols operated propy-lene hydroformylation (often termed ‘oxo’)processes for producing the required butyr-aldehyde using a cobalt catalyst system. This deliv-ered poor conversion and low selectivity of theprincipal feedstock, propylene, to the desiredproducts, in complex and cumbersome plantsoperating at high pressure.

The ‘Low Pressure Oxo’ process (LP OxoSM

Process) was developed and then licensed to theoxo industry through a tripartite collaboration

beginning in 1971. The principals were JohnsonMatthey & Co. Ltd. (now Johnson Matthey PLC),The Power-Gas Corporation Ltd. (a former nameof Davy Process Technology Ltd., now a sub-sidiary of Johnson Matthey PLC) and UnionCarbide Corporation (now a subsidiary of TheDow Chemical Company). Using rhodium-basedcatalysis, the LP OxoSM Process offered such greateconomic advantages over the established cobalt-catalysed processes, as well as technical elegance,that many cobalt systems were replaced by brandnew plants. In the thirty years or so since the LPOxoSM Process was first introduced, it has main-tained its position as the world’s foremost oxoprocess, having undergone much improvementand refinement. About two thirds of the world’sbutyraldehyde is now produced in LP OxoSM

plants. Most LP OxoSM systems are licensed

Enhancement of IndustrialHydroformylation Processes by theAdoption of Rhodium-Based Catalyst:Part IDEVELOPMENT OF THE LP OXOSM PROCESS TO THE COMMERCIAL STAGE

By Richard Tudor* and Michael AshleyDavy Process Technology Ltd., 20 Eastbourne Terrace, London W2 6LE, U.K.; *E-mail: [email protected]

The adoption of a low-pressure rhodium-based catalyst system in place of high-pressure cobaltfor the hydroformylation of propylene by reaction with carbon monoxide and hydrogen toproduce butyraldehydes (an ‘oxo’reaction) has brought large cost benefits to oxo producers.The benefits derive from improved feedstock efficiency, lower energy usage and simpler andcheaper plant configurations. The technical and commercial merits of the ‘LP OxoSM Process’for producing butyraldehydes have made it one of the best known applications of industrial-scale chemistry using a platinum group metal (pgm). Today, practically all butyraldehyde ismade by rhodium catalysis, and this should provide convincing encouragement to researcherswho are keen to exploit pgms as catalyst research materials, but are apprehensive as to theimplications of their very high intrinsic value. It should also encourage developers and designersresponsible for turning pgm chemistry into commercial processes, who may be daunted byproblems such as containment and catalyst life. This article (Part I) reviews the backgroundto the LP OxoSM Process, and its development to the point of first commercialisation.Part II, covering some of the key improvements made to the process and its use in non-propyleneapplications, will appear in a future issue of Platinum Metals Review.

Platinum Metals Rev., 2007, 51, (3), 116–126

DOI: 10.1595/147106707X216855

plants, nearly all of which have been built underlicences granted by Davy Process Technology (1)working in cooperation with The Dow ChemicalCompany (2); the remainder are plants owned andoperated by Dow’s Union Carbide subsidiary (3).

This article (Part I) reviews the background tothe LP OxoSM Process, addressing some of thechallenges that faced its developers and designersin planning the first commercial plant, and duringthe period immediately following commercialis-ation. Insights are given on the chemical functionof the homogeneous liquid-phase catalyst system.In Part II, some examples of advancements of thetechnology in the years following the first roundof licensing will be outlined.

The Beginnings of the DevelopmentCommercialisation of the LP OxoSM Process

was the culmination of an intensive joint effort inchemistry and engineering by the three compa-nies, dating back to 1964. Early exploratory workby the chemicals producer Union Carbide in theU.S.A. demonstrated promise for rhodium coordi-nation complexes in solution as hydroformylationcatalysts at low pressure, yielding a high propor-tion of the straight-chain aldehyde product andwith high enough catalyst productivity to justifyexamining the commercial potential of rhodium.The company obtained a basic patent for thiswork in 1970 (4). In operating a number of high-pressure oxo plants, Union Carbide had becomevery familiar with cobalt systems and their short-comings, and viewed the potential for rhodiumwith guarded excitement. At that time, all industri-al oxo production used the classic high-pressurecobalt process described below, or a modifi-cation of it.

Meanwhile, independent research by the lateProfessor Sir Geoffrey Wilkinson (5–7) (later towin a Nobel Prize for Chemistry) at ImperialCollege, London, supported by the precious metalrefiner and processor Johnson Matthey, producedresults using a suitable coordination complex ofrhodium (e.g. (5)) which basically reproduced orcomplemented the Union Carbide findings.Johnson Matthey in turn approached The Power-Gas Corporation (now Davy Process Technology).

The engineering contracting company drew on itsstrong background in process engineering toinvestigate the commercial potential for a low-pressure route to butyraldehyde. With thepublication of patents by Union Carbide (e.g. (4))and Johnson Matthey (e.g. (5)), the three partiesrealised that they had a common interest, so in1971 they launched a joint development pro-gramme to convert the laboratory rhodium-oxochemistry into a commercial process with a viewto exploiting its technical merit.

The focus of the collaboration was a processfor the hydroformylation of propylene using a mix-ture of carbon monoxide and hydrogen (in theform of synthesis gas) to produce normalbutyraldehyde and iso-butyraldehyde according toReaction (i):

2CH3CH=CH2 + 2CO + 2H2

→ CH3CH2CH2CHO + (CH3)2C(H)CHO (i)

normal butyraldehyde iso-butyraldehyde

The normal butyraldehyde isomer is usuallymore highly valued than iso-butyraldehyde. Amuch improved normal-to-iso yield ratio observedin the laboratory with rhodium catalysis, as com-pared with the then-current commercial cobaltsystems (i.e. about ten as opposed to typicallybetween three and four) was unquestionably a keydriver for collaborative development. The highselectivity of conversion of propylene to normalbutyraldehyde has since become a hallmark of theLP OxoSM Process.

The collaborators’ success in exploiting theirdevelopment efforts (8) would eventually result inthe LP OxoSM Process becoming the technologyof choice for many of the world’s oxo producers,with whom Davy Process Technology negotiatedlicences on behalf of the collaborators. The highreputation which the process would acquirebecause of its operating excellence and lowproduction costs, and a sustained growth in themarkets for the end products, drove investment incontinuing research and process developmentprogrammes aimed at improving the technologyto ensure the long-term sustainability ofthe process.



The Uses and Market forButyraldehyde

The major use of butyraldehydes was, and stillis, for the production of 2EH and butanols; seeFigure 1. Normal butyraldehyde has always beenthe more valuable of the two aldehyde isomers,because unlike iso-butyraldehyde, it can be used toproduce 2EH, by a sequence involving an aldolcondensation reaction followed by hydrogenationof the aldol product. Furthermore, normalbutanol, produced by the direct hydrogenation ofnormal butyraldehyde, usually offers solvent andderivative value superior to that of iso-butanol.The world production levels of 2EH and normaland iso-butanols combined are presently about 2.5 million and 4.5 million tonnes respectively (9).

In today’s marketplace, butanol and its deriva-tives have gained prominence from the long-termgrowth potential of water-based coatings (such asindoor paints), driven by environmental consider-ations, with demand for butyl acrylate andmethacrylate esters particularly strengthened bythis trend. Meanwhile, most of the world’s 2EH isesterified with phthalic anhydride to produce di(2-ethylhexyl) phthalate (DEHP), often referredto as dioctylphthalate or DOP, a plasticiser in wideuse for the production of flexible PVC. DOP hasbeen around for a long time, and its market issomewhat mature. Increasing amounts of 2EHare, however, being esterified with acrylic acid toproduce 2-ethylhexyl acrylate, used for adhesives,resins for latex, paper coatings and textile finish-ing. 2EH is also used to produce 2-ethylhexylnitrate, a diesel fuel additive, and also lubricantadditives. Propylene hydroformylation is increas-ingly being used as the first step in the productionof 2-ethylhexanoic acid, the wide applications of

which include alkyd resins and adhesives for lami-nated glass. Therefore, the range of productapplications linked to propylene hydroformylationis increasing, and the growth in global demand forbutyraldehyde is between about 2% and 3% peryear.

The Classic Cobalt ‘Oxo’ RouteIn 1938, the German chemist Otto Roelen,

working in the laboratories of Ruhrchemie AG,discovered that it was possible to react a mixtureof carbon monoxide and hydrogen with an olefinto form products containing oxygen. Roelen’s ini-tial work identified aldehydes and ketones in theproduct, and the reaction was named the ‘oxo’reaction. Later work established that using olefinsother than ethylene, the product is principally analdehyde, with very little ketone formation, andthe reaction was renamed ‘hydroformylation’.Both names are in common use, but ‘oxo’ hasbecome the more convenient and more interna-tionally recognisable name.

The process was commercialised in Germanyduring the early 1940s, and was then widely usedthroughout the world from the late 1940sonwards. The classic oxo process uses a cobalt cat-alyst in solution, operating at very high pressure inthe range 200 to 450 bar and at temperatures in therange 140 to 180ºC. The active compound iscobalt hydridocarbonyl HCo(CO)4. A very highCO pressure is needed to ensure catalyst stabilityduring hydroformylation. The catalyst has to bedecomposed before the reaction product can berecovered; therefore the process involves a cum-bersome and costly catalyst recovery cycle. Usingpropylene as feedstock, the ratio of normal to isoproducts is typically between about three and four,

PropyleneSyngas LP OxoSM

n-Butyraldehyde

n- + iso-Butyraldehyde

Aldolisation Hydrogenation

Hydrogenation

Hydrogen

Hydrogen

Product refining

Product refining

2-Ethylhexanol

n-Butanoliso-Butanol

Fig. 1 Schematic showing the production of oxo alcohols from propylene by the LP OxoSM Process

and the severe operating conditions mean thatthere is a high level of byproduct formation.Derivatives of butanol present in the reactionproduct could adversely affect the environmentalimpact of the process. The two or three high-pressure cobalt plants remaining in operation forproducing 2EH and butanols are very inefficient.They require considerable operator attention, arecostly to maintain, and leave a poor impression onthe environment.

A modification of the classic cobalt processwas commercialised in the 1960s, using as the catalyst cobalt hydridocarbonyl trialkylphosphine,HCo(CO)3PR3. The process operates at a lowerpressure than the ‘classic’ process (around 50 bar),although a higher temperature is needed. Withpropylene, the product shows much-improved lin-earity, the normal-to-iso ratio being around seven.The better selectivity to normal butyraldehyde is,however, partly offset by an increase in reactionbyproducts and an unavoidable production ofalcohols during oxo synthesis.

The Appeal of Rhodium-CatalysedHydroformylation

The first commercial plant to employ theLP OxoSM Process to produce butyraldehydes suc-cessfully started up in 1976. It was built by UnionCarbide at its petrochemical complex at Ponce,Puerto Rico, with a capacity of 136,000 tonnes perannum. As a result, the collaborators saw muchinterest in the technology from both existing andnew oxo producers. By the end of 1982, DavyProcess Technology had licensed and designed tenLP OxoSM plants that were built around the world.

Several advantages of the LP OxoSM Processappealed at that time. The high activity and goodstability of the rhodium catalyst meant that it wasnot necessary to use the very high pressures need-ed with cobalt to retain catalyst integrity. TheLP OxoSM Process operated at less than 20 bar,and a lower reaction temperature of between 90and 100ºC resulted in less byproduct formation.The lower temperature also brought other advan-tages over cobalt catalysis. Overall, the productmix from the reaction was much ‘cleaner’ and freeof many of the components formed using cobalt

chemistry. For example, the absence of butanol inthe product meant that esters and acetals were notformed – unlike with the cobalt process, for whichspecial measures were often needed to reduce theirenvironmental impact. With LP OxoSM, theproduct could be worked up using a much simplersystem and, very significantly, the selectivity ofconversion of propylene to the preferred normalbutyraldehyde was much better than with cobalt,the normal to iso ratio being improved about three-fold. These characteristics meant that propylenecould be converted to normal butyraldehyde moreeffectively and efficiently than had hitherto beenpossible. The lower operating pressure comparedwith cobalt eliminated or reduced the need forcompression of the incoming synthesis gas, andwith a simpler distillation system needed to workup the product butyraldehyde, overall energydemand was reduced.

In the thirty years since rhodium was first usedcommercially in hydroformylation, rhodium chem-istry of one form or another has been adopted tomeet at least 95% of world butyraldehyde demand.First- and subsequent-generation LP OxoSM plantsaccount for more than 60% of this; see for exam-ple Figure 2. (It is believed that the only remainingcobalt-based butyraldehyde production plants arein Russia, all other cobalt plants having been shutdown, with many of them being replaced by LP OxoSM plants.) Rhodium catalysis has also madeinroads into non-propylene hydroformylationapplications, and the possibilities here may wellincrease with time. Some of these applications willbe discussed in Part II.

How the LP OxoSM Process wasDeveloped

The active catalyst used in the LP OxoSM

Process is a hydridocarbonyl coordination com-plex of rhodium, modified with triphenyl-phosphine (TPP) ligand. The catalyst is formed,under process conditions, from rhodiumacetylacetonato carbonyl triphenylphosphine(Rh(acac)(CO)PPh3 or ‘ROPAC’), or a suitablealternative catalyst precursor. From the outset, theprocess concept involved a homogeneous liquid-phase catalyst, in which the active catalyst species


are dissolved in the reaction mixture along withreactants and reaction products, that is, normal andiso-butyraldehyde and high-boiling aldol condens-ation byproducts. The beauty of this route is thatno extraneous solvent is necessary. A characteristicof fully mixed homogeneous catalyst systems isthat short molecular diffusion ranges encouragehigh reaction rates. These were achieved at labora-tory scale, suggesting that rhodium concentrationsin the low hundreds of parts per million (ppm)would be suitable. This in turn implied affordablerhodium inventories for commercial-scale plants,provided that the rhodium could be sufficientlyprotected from poisoning and that excessivedeactivation could be avoided.

In the early stages of development, UnionCarbide needed to relate the rates of propylenehydroformylation and of the main byproduct-forming reaction, i.e. the hydrogenation ofpropylene to propane, to the main process vari-ables. A statistical approach was used to design aset of laboratory experiments to develop kineticmodels to determine these relationships. Modelswere also developed for the rate of formation ofheavy byproducts resulting from aldehyde conden-sation reactions. Drawing upon thesemathematical models, Davy Process Technology

was able to optimise relationships among equip-ment size and cost, reactant concentrations,feedstock consumption, and rhodium inventory,seeking the lowest possible overall produc-tion cost.

The Design of the First CommercialProcess

A key challenge to the developers and designersof a first commercial LP OxoSM Process, resultingfrom the intrinsic characteristics of the homoge-neous catalyst system, was how best to separate thebutyraldehyde product from the reaction mixture.The solution adopted needed to address key fac-tors such as losses of unrecoverable reactants andproduct, energy usage and capital cost. There werehowever two very important additional considera-tions that were directly linked to the use of a pgmof high intrinsic value: firstly, rhodium contain-ment, and secondly, the impact of process designon catalyst stability and catalyst life. For the for-mer, the physical loss of even relatively smallamounts of rhodium had to be avoided. As to thelatter, much care had to be applied to the design ofthe complete catalyst system, including the facili-ties needed for preparing, handling, treating andprocessing the raw materials and the various


Fig. 2 2-Ethylhexanol plant built by Sinopec Qilu Petrochemical Co. Ltd., China,employing the LP OxoSM Process

rhodium-containing streams. The object was toavoid design measures that might unduly harm thecatalyst, thus shortening its useful life.

At the outset of commercialisation, there wasconsiderable uncertainty as to the likely lifetime ofa rhodium catalyst charge in a commercial plant.Moving up from the laboratory to industrial scalewas not seen in itself as significantly influencingcatalyst life; the more salient issue was that duringlaboratory testing, it had not been possible toreplicate completely the operating regime to whichthe catalyst system would be subjected in a com-mercial plant, due to various limitations, and thisintroduced its own uncertainties. Only a certainamount could be learned in the laboratory aboutthe tendency for the catalyst to lose activity.

It was recognised, for example, that the catalyststability observed in small-scale rigs using high-purity feedstocks could not reflect the effects oncatalyst life of impurities present in commercialfeedstocks. Nor, with the limitations of rig designand scope, would it be possible to simulate thelong-term effects on the catalyst of operating con-ditions that could well occur in the plant butcannot be reproduced in the laboratory. Such con-ditions might negatively impact catalyst life. Thepredictive models for deactivation rates based onlaboratory studies therefore had their limitations,and considerable further effort would be neededhere as the technology developed. Despite theuncertainties, the conceptual process design forthe first commercial application of the LP OxoSM

Process built in as much protection for the rhodi-um as was thought desirable. The degree ofprotection was based on the known science, or,where there were large gaps in knowledge, on whatwas considered intuitively correct, in either casebearing in mind capital cost constraints.

The fact that the rhodium catalyst used insmall-scale rigs was not seeing representative com-mercial feedstocks, and the concerns this raisedwith respect to catalyst life, had to be addressedbefore the flowsheet for a commercial plant couldbe outlined. Early poisoning studies in the labora-tory by Union Carbide had concluded that thepropylene and synthesis gas mixtures produced inindustrial-scale plants were likely to contain impu-

rities that could either poison the rhodium orinhibit its performance. To put that problem intoperspective, it is useful to look at some data for thescale of butyraldehyde production that was thenbeing contemplated: based on predictive modelsgenerated from laboratory results, a commercialplant designed to produce 100,000 tonnes perannum of normal butyraldehyde would need acharge equivalent to about 50 kg of fresh rhodium.Given the rhodium metal price at the time, thereplacement value of this rhodium was aboutU.S.$1 million (allowing for the processingcharge). During one year of operation, each kilo-gram of rhodium, if it could last that long inservice, would be exposed to more than 2,500,000times its own mass of commercial feedstocks. Thequestion was whether there could be present inthat huge quantity of raw materials enough harm-ful contaminants, albeit at low concentrations, tothreaten serious damage to the catalyst, evendestroying its activity, within an unacceptablyshort period of plant operation. The answer was aresounding ‘yes’.

The poisoning studies carried out by UnionCarbide had shown that certain likely contami-nants such as hydrogen sulfide and carbonylsulfide (often found in commercial propylene andsynthesis gas streams), and organic chlorides oftenseen in propylene, were definite catalyst poisons.Other impurities, in particular dienes present inpropylene, had shown strong inhibiting effects onthe rhodium catalyst. Impurities that might catal-yse the aldol condensation reaction had also beenconsidered. If this reaction were allowed to occurto excess, it would produce too many high-boilingbyproducts in the reactor. Having identified targetimpurities, and quantified the problem in terms ofthe permissible concentrations of those impuritiesin raw material streams to be fed to commercialplants, new analytical techniques were required.Their sensitivity and repeatability had to be suffi-cient to measure the target impurities present inreal feed streams down to sub-ppm levels. Armedwith such analytical methods, Davy ProcessTechnology built laboratory rigs to develop andcharacterise processing schemes, employing het-erogeneous catalysts and adsorbents for removing


(to desired residual levels) the potentially trouble-some impurities likely to be found in commercialpropylene and synthesis gas streams.

The impurity guard beds and other purificationplant that Davy Process Technology developed forcommercial feedstocks ultimately featured in thedesign of commercial LP OxoSM plants, and wereto contribute to ensuring that catalyst deactivationrates in the operating plants were within permissi-ble limits.

Using the Gas Recycle PrincipleTo address the key challenge of how best to

separate the products and byproducts of the oxoreaction from the catalyst, several distillationcolumns were proposed in an early LP OxoSM

flowsheet. However, it was felt that this schemewould only exacerbate concerns regarding catalystdeactivation. Thermodynamic modelling, in con-junction with the kinetic models, revealed that itshould be possible to remove from the catalystsolution the reaction products, including high-boil-ing aldol condensation byproducts, by means ofgas stripping. This emerged as the makings of the‘gas recycle flowsheet’ adopted for the firstcommercial LP OxoSM plant, and several sub-sequent plants.

The flowscheme of an early LP OxoSM plantemploying the gas recycle principle is shown inFigure 3. A stirred, back-mixed reactor is config-

ured in a loop, also containing a gas recycle com-pressor, product condenser and liquid-vapourseparator. The catalyst solution, containing ligand-ed rhodium and excess triphenylphosphine (TPP)dissolved in the products and byproducts ofhydroformylation, is retained in the stirred reactor.The incoming fresh raw materials, after pretreat-ment to remove impurities, merge with recycledgas containing the chemical components of thesynthesis gas and vaporised organics from thereactor, to enter the base of the reactor throughdistributor spargers. The gaseous reactants pass asbubbles of small size (and hence large interfacialarea) into the liquid phase, where reaction takesplace at a closely controlled temperature, typicallyselected between 90 and 100ºC. While oxo synthe-sis takes place in the reactor, the reaction productsare stripped from the catalyst solution by anupward gas flow. Heat of reaction is taken outpartly via the latent heat of vaporisation of aldehydes into the gas, and partly by circulating acoolant through coils inside the reactor. The prod-ucts are condensed from the gas/vapour effluentleaving the top of the reactor, and the resulting liq-uid products are separated from the recycle gas.The gas/uncondensed vapour is then recom-pressed for recycling to the reactor. Operatingconditions, in particular the gas recycle rate, are setso that all liquid products leave the system at thesame rate at which they have been formed, so that


Key1 Pretreatment2 Reactor3 Catalyst preparation4 Condenser5 Separator6 Stabiliser7 Cycle compressor8 Overhead compressor

PropyleneSynthesis gas 1

4

8

62 5

3

7Purge gas

Mixedaldehydes Fig. 3 Gas recycle

flowsheet of an earlyLP OxoSM plant

the reactor inventory remains constant. Passivecomponents in the synthesis gas, such as nitro-gen, methane and carbon dioxide, along withpropane present in the propylene or formed byhydrogenation, are purged in a blow-off to a fuelheader, to prevent them from accumulating in thesystem. Unreacted propylene, propane, CO andhydrogen dissolved in the condensed productleaving the separator are removed from the prod-uct in a stabiliser column, and recompressedbefore being recycled to the reactor.

The basic flowscheme of the LP OxoSM

Process emerged as both simple and elegant. Theprinciple of using in situ gas stripping to separateproduct from the catalyst appeared sound,because the high molecular weight of the rhodi-um catalyst complex should mean that the loss byvaporisation of rhodium in the product would bepractically zero. The rhodium catalyst was safelycontained in the reactor, and provided sufficientenergy could be imparted through the mixerimpeller, the catalyst would be exposed to operat-ing conditions more or less replicating those usedin the laboratory. There was no reason for anysignificant amount of rhodium to leave the reac-tor during day-to-day operation, provided leakagewas avoided and the physical entrainment of cat-alyst solution in the reactor overhead gas streamwas minimised. Neither of these containmentrequirements were expected to pose undue diffi-culties. Catalyst leakage could be virtuallyeliminated by good engineering practice, includ-ing the careful selection of construction materialsand mechanical seals for moving parts; entrain-ment could be dealt with by using proprietary, butinexpensive, entrainment filters on the overheadline from the reactor. The use of in situ strippingobviated the need to remove catalyst solutionfrom the reactor to separate product using exter-nal distillation equipment, thereby eliminating anypotential for increased catalyst deactivation dueto concentrating the catalyst, and exposing it tohigher temperatures than those used in the reac-tor.

The adoption of the gas recycle principle notonly led to a simple and affordable process flow-sheet, it also appeared to provide the best overall

working regime for the catalyst, in terms of bothloss prevention and deactivation, based on the‘state of the art’ at the time.

Success from the First LP OxoSM PlantHaving decided to build a commercial plant at

Ponce, Union Carbide erected a 200 tonnes perannum gas recycle pilot plant at the same site totest the process on the feedstocks available there,and to provide scale-up data. While the pilot plantwas being built and commissioned, Davy ProcessTechnology started the process and basic engi-neering design of the 136,000 tonnes per annumfull-scale unit. This was to be built almost along-side the pilot plant. The process design wasrefined and further developed once operatingdata were available from the pilot plant, whichcontinued to operate for a short time after thecommercial unit first started up in January 1976.

The initial start-up of the full-scale Ponceplant was easier than anticipated. Excluding out-side interruptions, the plant was online for all butone hour in its first month of operation. Duringits first year, its on-stream operational availabilitywas greater than 99%. This contrasted with a typ-ical availability of about 90% for a conventionalcobalt-based oxo plant, based on Union Carbide’sown experience. The operation continued to bemarked by what was until then unusual ease,stability and smoothness. Design targets for pro-ductivity, selectivity, feedstock usage efficiencyand product quality were all met. The ratio of normal to iso-butyraldehyde was usually con-trolled at around 10:1, but higher ratios up to 16:1were achieved. Significantly, the costs attributableto catalyst were lower than expected, and the lifeof the first catalyst charge exceeded one year.

The reaction temperature was kept as low aspossible, and in the range of about 90 to 100ºC,consistent with being able to achieve sufficientcatalyst productivity from the volume of catalystsolution available to meet the productiondemands, and being able to control the liquid lev-els in the reactors. (Product stripping was easier athigher temperatures because of the higher vapourpressures of the products.) It was known thathigher reaction temperatures would lead to an


increased production of reaction byproducts andan increased rate of catalyst deactivation; effec-tive temperature control was therefore important.The reaction temperature could be regulated veryclosely – to within ± 0.5ºC. The operating pres-sure of the reactors was also well controlled at about 18 bar.

The process characteristics and controlsystems used meant that the unit needed littleday-to-day operator attention. Again, thiscontrasted with experience on high-pressurecobalt plants. The rhodium unit could quickly berestarted from a full shutdown, and it waspossible to restore production following outagesmuch more rapidly than had been the casewith cobalt.

How the Catalyst WorksThe active rhodium species for the LP OxoSM

Process is formed under hydroformylation reac-tion conditions, and there is no need for complexcatalyst synthesis and handling steps. The proba-ble sequence of the reaction with propylene toform normal butyraldehyde is shown in Figure 4.

Rhodium is introduced to the oxo reactor inthe form of a solution of ROPAC (a stable crys-talline compound) in butyraldehyde. Complex Ain Figure 4 is formed from the fresh rhodium in

the presence of carbon monoxide and TPP. Inthis coordination complex the rhodium atom car-ries five labile-bonded ligands: two TPP, twocarbon monoxide and one hydrogen. In the firstreaction step, a propylene ligand is added to formcomplex B, which rearranges to the alkyl com-plex C. This undergoes carbon monoxideinsertion to form the acyl complex D. Oxidativeaddition of hydrogen gives the dihydroacyl com-plex E. Finally, hydrogen transfers to the acylgroup, and normal butyraldehyde is formedtogether with complex F. Coordination of F withcarbon monoxide regenerates complex A.

Some iso-butyraldehyde is produced alongwith the normal butyraldehyde, but a high selec-tivity to the latter is ensured by exploiting a sterichindrance effect as follows. The reaction is car-ried out in the presence of a large excess of TPP.Under the low-pressure conditions of the reac-tion, the high TPP concentration suppresses thedissociation of complex A into one containingonly a single phosphine ligand. If largely undisso-ciated complex A is present, with its two bulkyTPP ligands incontact with the propylene, then ahigh proportion of primary alkyl is favoured – iffewer such ligands were present, then morepropylene would form secondary alkyl groups,leading to more iso-butyraldehyde.


L

CO CO

CO

+ COCO

Rh

Rh

Rh

Rh

Rh

Rh

Rh

CO

CO CO CO CO

CO

CO

CO

L

L

L

L

L

L

L

+ L

L

L

L

L

L

L

L

H+ H2

H HCH2=CHCH3 CH2CH2CH3+ CH2=CHCH3

CH2CH2CH3

CH2CH2CH3

– CH3CH2CH2CHO

HH

H

A B C

DEF

Fig. 4 Probable reaction cycle for formation of normal butyraldehyde from propylene (L = triphenylphosphine (TPP);A: product of reaction of ROPAC with carbon monoxide and TPP; B: addition product of A and propylene; C: alkylcomplex resulting from rearrangement of B; D: acyl complex resulting from carbon monoxide insertion to C;E: dihydroacyl complex resulting from oxidative addition of hydrogen to D; F: product of elimination of butyraldehydefrom E)

Measures to Deal with CatalystDeactivation

The TPP-modified catalyst has a tendency todeactivate over time due to the formation from themonomeric rhodium species of rhodium clusters.This type of deactivation is termed ‘intrinsic’, todistinguish it from deactivation caused by an exter-nal source such as catalyst poisons present in thefeedstocks. Catalyst management models weredeveloped to help operators of the LP OxoSM

Process to optimise the economic return fromtheir catalyst charges, in recognition that intrinsicdeactivation had to be tolerated to some extent.For example, operating temperatures could not belowered to reduce catalyst deactivation if this alsoreduced catalyst productivity to uneconomic orunmanageable levels. Rhodium catalyst manage-ment guidelines from Union Carbide and DavyProcess Technology recommended operatingadjustments to compensate for deactivation, inresponse to accumulated operating data whichindicated the time evolution of catalyst activity.The guidelines were couched so as to optimise thebalance between reaction rate, selectivity to nor-mal butyraldehyde and catalyst stability. Whileoperators felt some obligation to comply with thelicensor’s recommendations, at least until perfor-mance warranties had been met, it was interestingto observe how the long-term catalyst operatingstrategies adopted by plant owners varied sowidely between plants, depending on specific cir-cumstances and preferences.

Plant operators observed rates of catalyst deac-tivation that meant that a rhodium catalyst chargewould typically last for about 18 to 24 monthsbefore its activity had declined to the point whenit would have to be discharged from the reactorand replaced by a fresh catalyst charge.

The earliest LP OxoSM plants contained verysimple equipment to concentrate the dischargedspent catalyst solution. The idea was that concen-trated catalyst, containing say 2000 ppm ofrhodium, would be shipped to Johnson Matthey inthe U.K., who would then recover the rhodium ina form suitable for reprocessing to ROPAC. Butthe logistics of actually reprocessing around 20 tonnes of concentrate for a typical plant were

somewhat daunting. There were concerns abouthandling and transporting such material in suchlarge quantities. With rhodium metal prices rising,the logistics might put the security of, say, U.S.$2million worth of rhodium at undue risk. Therewere also uncertainties about what other sub-stances might be present in the rhodiumconcentrate that could cause Johnson Mattheyprocessing problems. Although metals like ironand nickel that are usually found in commercialfeedstocks could be anticipated, would metal con-tamination compromise rhodium recovery? Therequirement for off-site rhodium recovery frombulk catalyst solution detracted from the eleganceof the LP OxoSM Process. Fortunately, by the timethe first licensed plants actually started operation,Union Carbide had proven a catalyst reactivationtechnique that would virtually obviate off-siterecovery.

Catalyst ReactivationBy the early 1980s, before any need had arisen

to resort to off-site rhodium recovery, UnionCarbide had developed a means to deal with theintrinsic deactivation – effectively by reversing it.This involved concentrating the spent catalyst andthen treating the rhodium present in the resultingresidue to convert it into a form capable of reacti-vation. The concentration process was carried outusing specialised equipment (a proprietary evapo-rator) under very precise conditions, includinghigh vacuum, designed to prevent catalyst damage.The overall process could conveniently be per-formed at the plant site, and required no chemicalreagents. It resulted in a ‘declustering’ of rhodiumto enable the restoration of activity once thetreated residue had been returned to a hydro-formylation environment. Eventually, all operatorseither added reactivation equipment to theirplants, or arranged to share facilities. Catalyst reac-tivation was incorporated into the standard designof all new plants, and a measure of lost elegancewas restored to the LP OxoSM Process!

The catalyst reactivation technique was used tocarry out repeated reactivations of what was essen-tially a single catalyst charge. This drasticallyreduced the need for off-site recovery, which was


normally deployed only on rhodium that could nolonger be reactivated economically. In that case,the recovery could be performed on residues typi-cally containing about 8000 ppm of rhodium, fourtimes the concentration initially envisaged, thusimproving the logistics and reducing the cost ofoff-site processing.

ConclusionThis article (Part I) has sought to demonstrate

the initial promise of the LP OxoSM Process,employing rhodium-based catalysis, in terms ofhigh availability, selectivity and productivity, lowenvironmental impact and low maintenance. PartII, to be published in a future issue of PlatinumMetals Review, will address subsequent key improve-ments to the process, and its use in non-propyleneapplications.

LP OxoSM is a service mark of The Dow Chemical Company.

References1 Davy Process Technology Ltd.:

http://www.davyprotech.com/

2 The Dow Chemical Company: http://www.dow.com/3 Union Carbide Corporation:

http://www.unioncarbide.com/4 R. L. Pruett and J. A. Smith, Union Carbide Corporation,

‘Hydroformylation Process’, U.S. Patent 3,527,809; 19705 G. Wilkinson, Johnson Matthey, ‘Improvements in

Catalytic Hydrogenation or Hydroformylation’, BritishPatent 1,219,763; 1971

6 M. L. H. Green and W. P. Griffith, Platinum Metals Rev.,1998, 42, (4), 168

7 W. P. Griffith, Platinum Metals Rev., 2007, 51, (3), 1508 F. J. Smith, Platinum Metals Rev., 1975, 19, (3), 939 Production estimates provided by RXN Petrochemical

Consulting Inc.: http://rxnpetrochem.com/page4.html

Further Reading‘Low-pressure oxo process yields a better product mix’,Chem. Eng. (New York), 5th December, 1977, 84, (26), 110;marking the award of the 1977 Kirkpatrick ChemicalEngineering Achievement Award to the winners: UnionCarbide Corporation, Davy Powergas Ltd., and JohnsonMatthey & Co. Ltd.

J. L. Stewart, ‘LP OxoSM process – a success story’,Indications, Winter 1982/83; the international journal ofDavy McKee


The AuthorsRichard Tudor is a chartered chemicalengineer. He has played a leading part inDavy Process Technology’s oxo licensingactivities for over thirty years, firstly asProcess Manager, and then as BusinessManager after a period as LicensingManager. As a Vice President of sales andmarketing, he now has overall responsibilityfor the oxo business.

Mike Ashley spent many years with JohnBrown, involved with process technology andbusiness development, before joining DavyProcess Technology. He is now concernedwith business analysis, technologyacquisition, marketing, website developmentand all aspects of public relations.

The annual Cape Organometallic Symposia(COS) series was started in 2003 through the com-bined efforts of researchers at the Universities ofCape Town, Stellenbosch and the Western Cape inSouth Africa, and the COS 2006 chairman wasProfessor John R. Moss, Department ofChemistry, University of Cape Town ([email protected]). These symposia weredesigned as informal one-day sessions, bringingtogether researchers with similar interests from thethree institutions, and successfully exposing gradu-ate students to a mixture of invited lectures andstudent presentations. In particular, all studentswould have opportunities to network with theirpeer group and to present their work in oral orposter format for discussion.

In 2006 this event was timed for 10th and 11thof August, to run immediately before the 37thInternational Conference on CoordinationChemistry (ICCC) (1), affording visitors an oppor-tunity to attend both events, and exposing localgraduate students to some of the best researchersin organometallic chemistry from around theworld. The format therefore changed to a two-dayevent, and a greater proportion of presentationsthan usual were given by these leading invitedinternational researchers. However, the event waslimited to around 100 delegates to ensure that theatmosphere remained fairly informal, and theevent more intimate than most international gath-erings. The theme for the COS 2006 event was“Organometallics and their Applications” (OATA)(2), with a view to emphasising the links betweenthe multiplicity of academic research directionsbeing followed and the potential for practicalapplications of organometallic chemistry.

The presentations were fairly evenly divided

between the organometallic chemistry of early, lateand first row transition metals. However, it is thepurpose of this review only to consider those pre-sentations involving the development of theorganometallic chemistry of the platinum groupmetals (pgms), and its potential applications.

Nanotechnology and Catalysis The systematic development of the field of nan-

otechnology and catalysis was well illustrated byProfessor Brian F. G. Johnson’s (University ofCambridge, U.K.) presentation on supportednanoparticle preparations and nanocatalyst activity,entitled: ‘Small and Beautiful: Nano-Catalysts byDesign or Strategically Designed Single-SiteHeterogeneous Catalysts for Clean Technology,Green Chemistry and Sustainable Development’.Starting with a myriad of well characterised mixed-metal clusters, this group has designed andprepared a range of very active catalysts with vary-ing metals, in particular these include platinum andruthenium in well controlled and varied ratios.When these were supported inside mesoporousmaterials, an added dimension to their reactivitywas described. In the second part of this presenta-tion, the tethering of active metal centres to a‘non-passive’ support with a chiral ligand wasdescribed as a further advantageous manner inwhich novel catalysts could be prepared.

Similarly starting with the useful precursorsRuHCl(CO)L3 and trans-IrCl(CO)L2 (L = PPh3),Professor Anthony Hill’s group (AustralianNational University), have previously prepared andstudied the reactivity of a number of mono-, bi-and trimetallic carbene and carbyne complexes. Inthis talk: ‘Tricarbido Complexes:LnM≡C–C≡C–M″Ln’ a fairly general route to

127

The 4th Cape Organometallic Symposium:Organometallics and Their ApplicationsTHE ORGANOMETALLIC CHEMISTRY OF PLATINUM GROUP METALS

Reviewed by David J. RobinsonCSIRO Minerals, Parker Cooperative Centre for Integrated Hydrometallurgy Solutions, PO Box 7229, Karawara, WA 6152,

Australia; E-mail: [email protected]


DOI: 10.1595/147106707X216927

mono-, di- and tricarbido complexes was present-ed, involving the isolation and characterisation of arange of interesting new compounds containingone or more of the pgms, gold, mercury, molybde-num, tungsten and other metals.

Professor Anna M. Trzeciak (University ofWroclaw, Poland) presented interesting resultsfrom her group’s study of the methoxycarbonyla-tion of iodobenzene and the Sonogashira couplingof an aryl acetylene and aryl halide both using a pal-ladium catalyst in an ionic liquid, in particularaddressing the inhibiting effect of imidazoliumhalides: ‘Palladium Catalysed Methoxycarbonyl-ation in Ionic Liquids, Inhibiting Effect ofImidazolium Halides’. With the methoxycarbonyla-tion reaction, a simple palladium precursor (such as[PdCl(COD)], COD = cycloocta-1,5-diene) and abase were sufficient to achieve good yields in cer-tain ionic liquids. Similarly, good yields could beobtained in the Sonogashira reaction in the pres-ence of a base and a [PdCl2{P(OPh)3}2] catalyst.Systematic variation in the components of the ionicliquid, and the effects of such changes on the reac-tion yields and on isolation of a number ofinteresting adducts and intermediates, have helpedto throw light on these respective mechanisms.

Mechanistic Studies Recent developments in our understanding of

the mechanism of the asymmetric hydrogenation ofolefins using a number of derivatives of the clustersM3(CO)12 (M = Ru or Os), H2Os3(CO)10 orH4Ru4(CO)12 were presented by Professor EbbeNordlander (Lund University, Sweden): ‘Cluster-based Catalytic Systems for Asymmetric Reactions’.In particular, a range of catalysts[H4Ru(CO)10(P-P*)] and [H4Ru(CO)8(P-P*)2] (P-P* = BINAP, MOBIPH, DIOP) were prepared.Their activity and enantioselectivity were studiedfor the hydrogenation of 2-methyl-2-butenoic acidat elevated temperature and pressure. Most inter-esting were the enhanced reactivity andenantioselectivity obtained when ferrocene-con-taining diphosphine ligands were used, and thefurther enhancement in enantioselectivity whichwas observed when the reaction was performed inthe presence of trace amounts of mercury.

As part of a broader presentation on catalyticchemistry, and particularly ethylene oligomeris-ation/polymerisation, a number of new palladiumcomplexes ligated to one or more derivatisedferrocene units were prepared and used as catalystsin Suzuki coupling reactions. This work, assessingthe potential for ‘modular design’ of efficient cata-lysts and the isolation of several palladium(0) andpalladium(II) intermediates in the proposed mech-anism, was presented by Professor T. S. Andy Hor(National University of Singapore): ‘Design ofSmart Catalysts by the Combinative &Complementary Uses of Hemilability and MetalUnsaturation’.

Synthesis of Metallacycles andClusters

Two presentations from the group of ProfessorJohn Moss extended the range of metallacycles thathave been made, typically via one of two routesinvolving either a di-Grignard reagent or the appli-cation of the Grubbs catalyst and a ring closingmetathesis reaction upon a di-alkenyl complex.Akella Sivaramakrishna presented a talk entitled:‘Synthesis & Structure of Metal-alkenyl Complexes– Novel Precursors to Fascinating Chemistry!’.Starting with [PtCl2(COD)] or Cp*IrClL2, a rangeof di-alkenyl compounds of PtL2R2, PtL'R2 orCp*IrLR2, respectively, (R = alkenyl group; L =PPh3 or PtBu3, L' = dppe or dppp), of varying sta-bility were prepared by reaction with appropriateGrignard reagents (and additional phosphine L orL' for platinum). While the thermal decompositionof several of these new compounds was studied, ofgreater interest was the use of a ring closingmetathesis reaction catalysed by first-generationGrubbs catalyst to form platina- and iridacycleswith up to 21 ring atoms. Several side reactions andother interesting transformations including dimeri-sation, internal isomerisation, allyl formation andtransmetallation reactions were reported. The inser-tion of metal carbonyls and small molecules intometal–alkenyl ligand bond and the potential for for-mation of hetero-bimetallic clusters and othernovel complexes were described. Emma Hager pre-sented ‘The Synthesis of Novel Rhodacycloalkanes:“Old” and New Methods’. The analogous use of


Cp*RhLCl2 (L = PPh3, PPh2Me or PPhMe2) andappropriate di-Grignard reagents also resulted in aseries of compounds of the type:Cp*RhL{(CH2)n}. These metallacycles are consid-ered to be model compounds for several chainforming catalytic reactions, including ethylenetrimerisation. As such, their thermal decomposi-tion is also of interest and was reported. Di-alkenylrhodium complexes were similarly prepared from[Cp*RhCl2]2 and excess appropriate Grignardreagent with longer carbon chain groups. Thesame reaction with short chain Grignard reagentsresulted in interesting allylic rhodium complexes.

A student presentation from Cathrin Welker(University of Cape Town, South Africa): ‘Fischer-Tropsch Synthesis on Organometallic Ru-ModelCatalysts’, addressed the systematic preparation ofmodel ruthenium clusters of varying sizes and thestudy of their thermal decomposition and Fischer-Tropsch activity.

Concluding Remarks Over 30 excellent posters were also presented

during the symposia. Student participation, asdesired, was evident throughout the proceedings.The format for 2007 is again to be slightly differ-ent to allow the symposium to be incorporatedinto the South African Chemical Institute,Inorganic Chemistry Conference (3), again chairedby Professor John Moss. It is clear that the seriesof Cape Organometallic Symposia, and indeedinorganic and organometallic chemistry, are very

much alive and thriving in the Western Cape andmore generally in South Africa. The timing of theCOS 2006 and the 37th ICCC (1) in consecutiveweeks afforded an almost unique opportunity formany local students and academics to meet andinteract with experts from around the world, whilethe large amount of pgm chemistry presented atboth conferences appropriately highlighted theimportance of these metals and their applicationsin today’s world.

References1 D. J. Robinson and M. Arendse, Platinum Metals Rev.,

2007, 51, (1), 362 4th Cape Organometallic Symposium:

Organometallics and their Applications;http://www.wildmice.co.za/chemistry/oata2006.html

3 INORG 007: Inorganic Chemistry Conference2007, Club Mykonos, Western Cape, South Africa,8th–12th July, 2007;http://www.wildmice.co.za/chemistry/inorg007.htm


The AuthorDavid John Robinson was active in bothfundamental and applied pgm chemistryresearch, and in particular, the development ofimproved separation technologies over a 15year career with Anglo Platinum. He wasinvolved directly in their production at themodern precious metals refinery nearRustenburg, South Africa. Since writing this

review he has moved to CSIRO Minerals’ Parker CooperativeResearch Centre for Integrated Hydrometallurgy Solutions,Karawara, Western Australia. His interests include developingimproved industry-academia research collaborations and theapplication of improved fundamental knowledge to the solution ofreal refining problems. In his new position he will be active in thearea of base metal hydrometallurgy and developing advantages forindustry through better scientific research, optimisinghydrometallurgical processing and efficient technology transfer.


Calculation of Vapour PressureValues

Vapour pressure values are calculated fromthermodynamic data following the proceduredescribed in Appendix A. Since free energy func-tions are given for one bar standard-state pressure,then the vapour pressure will also be given in bars.Values for the free energy functions and the select-ed heats of sublimation are given in the reviews onplatinum (1) and the other five pgms: palladium(2), ruthenium (3), osmium (4), rhodium (5) andiridium (6). Since the behaviour of the specific heatcapacities of the solids and their effect on the freeenergy functions is more complicated than for theliquids, then vapour pressure values were evaluat-ed for the solids at 25 K intervals and the meltingpoint, and for the liquids at 50 K intervals and themelting point, so as to give approximately equalnumbers of data points. In the review on platinum(1), a minimum lower temperature of 1200 K wasconsidered. This gives a vapour pressure of about10–17 bar, which is considerably below practicalmeasurements. For the other metals, the use ofrounded temperatures gives a vapour pressure

between 10–16 and 10–17 bar, which was used as thelower bound.

Selection of a Vapour PressureEquation

The derivation of the selected vapour pressureequation (Equation (i)) is described in Appendix B.

ln(p, bar) = A + Bln(T ) + C/T + DT + ET 2 (i)

where A, B, C, D and E are constant coefficients.Equation (i) was used by Honig and Kramer (7) torepresent the vapour pressures of the elementsover a wide range of temperatures and pressures.

Recent Data and Their Effect on theSelected Values

Since publication of the reviews on platinumand the other metals (1–6) newer data havebecome available for consideration:(a) The standard value of the atomic weight of

platinum has been changed from 195.078 ±0.002 to 195.084 ± 0.009 (8). In the review onplatinum (1) the thermodynamic properties of

Vapour Pressure Equations for thePlatinum Group ElementsIMPROVED CALCULATIONS OF VAPOUR PRESSURE AND TEMPERATURE

By J. W. ArblasterColeshill Laboratories, Gorsey Lane, Coleshill, West Midlands B46 1JU, U.K.; E-mail: [email protected]

While a knowledge of the vapour pressure curve of any material is of theoretical significancein understanding its basic physical properties, it can also be of practical importance in, forexample, the use of the material in high-temperature vacuum applications. Therefore readilyusable equations which accurately predict values of vapour pressure over a wide range oftemperatures and pressures can have an important practical use. For the platinum groupmetals (pgms) the vapour pressures can be immediately assessed from about 10–16 bar to justabove the boiling point by the use of Equation (i), fitted for the solid and liquid metals separately:ln(p, bar) = A + B ln(T) + C/T + DT + ET 2, where p is the vapour pressure, T is the temperaturein kelvin and A, B, C, D and E are constants. Although containing five coefficients, this equationcan easily be evaluated by computers and scientific calculators. Although it gives values ofvapour pressure at fixed temperatures, by a simple and rapid use of iteration values oftemperature at fixed vapour pressure, temperature-dependent values of vapour pressure canalso be obtained.

DOI: 10.1595/147106707X213830


the condensed phases were based on an atom-ic weight of 195.08, and the gas phaseproperties had been corrected using the atomicweight of 195.078. Since the change of 0.006to the new atomic weight was from one singledetermination to another, it is likely that adefinitive value for the atomic weight has notyet been obtained. Therefore, for consistencywith the treatment of the condensed phases,the values for the gas phase have been recalcu-lated using an atomic weight of 195.08. Theeffect on the tabulated values is negligible atthe level of accuracy given.

(b)Two further determinations of the vapourpressure of palladium by Zaitsev, Priselkovand Nesmeyanov (9) (1267 to 1598 K) andFerguson, Gardner and Nuth (10) (1473 to1973 K) led to heats of sublimation of 377.3 ±0.1 kJ mol–1 and 377.7 ± 0.2 kJ mol–1, respec-tively, in excellent agreement with the selectedvalue of 377 ± 5 kJ mol–1.

(c) A number of amendments to the calculationof the thermodynamic properties of thegaseous elements from energy levels have beenfound to have a negligible effect on the freeenergy functions at the level of accuracy givenin the tables in References (1–6):

(i) For the metals other than platinum thethermodynamic properties were calculatedusing mainly the 1986 CODATAFundamental Constants (11) but all valueshave now been recalculated using the 2006CODATA Fundamental Constants (12).

(ii)Joueizadeh and Johansson (13) haverevised seventeen of the energy levels ofruthenium.

(iii)Engleman et al. (14) have completelyrevised the energy levels of palladium andextended their number.

Application of the Selected VapourPressure Equation to the PGMs

Coefficients corresponding to the selectedvapour pressure equation for both the solid andliquid pgms are given in Table I. One criterion tobe met is that the calculated vapour pressuresshould be equal at the melting point, or more correctly the triple point which is the solid-liquid-vapour equilibrium temperature. However,because of the high resistance of the pgms tocompression, then at the level of accuracy obtain-able the melting points are considered to be equalto the triple points as indicated in Appendix C.

In the review on platinum (1), only the first

Table I

Coefficients in Equation (i) for the PGMs in the Solid and Liquid States

Element Phase Temperature A B C D Erange, K

Ru Solid 1400–2606 23.31345 – 0.632925 – 78385.0 3.36362 × 10–4 – 8.85627 × 10–8

Liquid 2606–4600 54.00959 – 4.54744 – 80366.4 4.73549 × 10–4 – 1.54492 × 10–8

Rh Solid 1200–2236 45.43958 – 3.95580 – 68981.1 2.32882 × 10–3 – 2.96772 × 10–7

Liquid 2236–4200 38.32595 – 2.60178 – 67855.0 – 5.86242 × 10–5 4.32765 × 10–9

Pd Solid 850–1828 14.37701 0.270634 – 45327.0 – 1.30189 × 10–3 2.07872 × 10–7

Liquid 1828–3300 92.64931 – 10.78435 – 51456.6 3.94327 × 10–3 – 2.33111 × 10–7

Os Solid 1700–3400 26.80257 – 1.17147 – 95027.6 5.67800 × 10–4 – 6.25215 × 10–8

Liquid 3400–5600 44.97739 – 3.42327 – 93300.9 2.65730 × 10–4 – 5.86394 × 10–9

Ir Solid 1400–2719 27.23601 – 1.22965 – 81010.4 4.34895 × 10–4 – 5.80991 × 10–8

Liquid 2719–5000 51.13835 – 4.06675 – 83829.3 1.29129 × 10–4 – 3.77392 × 10–9

Pt Solid 1200–2041.3 20.55547 – 0.279512 – 68277.9 – 1.49389 × 10–4 – 3.60502 × 10–8

Liquid 2041.3–4200 34.89596 – 2.24178 – 68166.4 4.95301 × 10–5 8.91166 × 10–10


three terms of Equation (i) were used, which gavean adequate representation of the vapour pressure.However, when all five terms are used to ensureprecise reproducibility for all of the metals, thecorrelation for platinum is extraordinary. Theaverage deviation between the thermodynamicallyderived vapour pressure and the result from theequation is only ± 0.0004% for both the solid andliquid, and similar results were generally obtainedfor the other metals as given in Table II. Even inthe worst case, for solid rhodium, agreement is stillwithin five significant figures; this tolerance is wellbeyond that of any practical determination ofvapour pressure.

Whereas the equations for the liquids repro-duce the thermodynamic boiling points to within0.01 K, more realistic estimates of the uncertain-ties in the boiling points can be ascertained fromthe uncertainties assigned to the enthalpies of sub-limation, with boiling point uncertainties roundedto the nearest 10 K (see Table III).

The uncertainties given in Table III must beconsidered to be minimum values, because theuncertainties in the free energy functions are nottaken into account. For example, the enthalpy of

fusion and thermodynamic properties of liquidosmium are all estimated, and therefore a morerealistic estimate of the uncertainty in the boilingpoint would be ± 100 K.

Temperatures Corresponding toGiven Vapour Pressures

These values are given in Table IV. Since thelowest temperature used corresponds to a vapourpressure between 10–17 and 10–16 bar, then 10–16 baris used as the lower bound for fixed pressures.Table IV provides a check on values obtained byiteration of the equations. It is an amendment to aprevious summary (15), in which the lowest pres-sure included was 10–12 bar. Vapour pressurescorresponding to fixed temperatures are obtaineddirectly from the equations, and a check is provid-ed by Table V, which lists the melting pointvapour pressures.

Table II

Percentage Deviations betweenThermodynamically Derived Vapour PressureCurves and Results from Equation (i) for PGMsin the Solid and Liquid States

Element Phase Accuracy of fit, %

Ru Solid 0.0014Liquid 0.0001

Rh Solid 0.0059Liquid 0.0002

Pd Solid 0.0020Liquid 0.0014

Os Solid 0.0008Liquid 0.0000

Ir Solid 0.0002Liquid 0.0003

Pt Solid 0.0004Liquid 0.0004

Table III

Uncertainties in the Boiling Points of PGMs

Element Enthalpy of sublimation Boiling point± uncertainty ± uncertainty

ΔH°298.15, kJ mol–1 T, K

Ru 649 ± 4 4592 ± 30

Rh 558 ± 10 4114 ± 90

Pd 377 ± 5 3263 ± 50

Os 788 ± 4 5576 ± 30

Ir 670 ± 6 4898 ± 50

Pt 565 ± 2 4149 ± 20

References1 J. W. Arblaster, Platinum Metals Rev., 2005, 49, (3),

1412 J. W. Arblaster, CALPHAD, 1995, 19, (3), 3273 J. W. Arblaster, CALPHAD, 1995, 19, (3), 3394 J. W. Arblaster, CALPHAD, 1995, 19, (3), 3495 J. W. Arblaster, CALPHAD, 1995, 19, (3), 3576 J. W. Arblaster, CALPHAD, 1995, 19, (3), 3657 R. E. Honig and D. A. Kramer, RCA Rev., 1969, 30,

2858 M. E. Wieser, Pure Appl. Chem., 2006, 78, (11), 20519 A. I. Zaitsev, Yu. A. Priselkov and A. N.

Nesmeyanov, Teplofiz. Vys. Temp., 1982, 20, (3), 589;Chem. Abstr., 1982, 97, 98769


Table IV

Temperatures Corresponding to Fixed Vapour Pressures for the PGMs

Pressure Temperature, K

p, bar Ru Rh Pd Os Ir Pt

10–16 1403 1219 870 1706 1456 1238

10–15 1464 1273 911 1780 1520 1292

10–14 1530 1332 956 1861 1590 1352

10–13 1603 1396 1005 1950 1667 1417

10–12 1684 1468 1060 2048 1751 1489

10–11 1772 1546 1122 2156 1844 1569

10–10 1871 1634 1191 2277 1949 1659

10–9 1982 1733 1269 2411 2065 1759

10–8 2107 1845 1359 2563 2197 1872

10–7 2249 1972 1462 2736 2347 2002

10–6 2412 2119 1582 2934 2520 2156

10–5 2602 2293 1725 3163 2721 2339

10–4 2842 2508 1899 3435 2976 2558

10–3 3134 2772 2120 3793 3288 2824

10–2 3498 3102 2400 4239 3681 3155

10 1 3965 3530 2765 4810 4193 3580

1 4588 4110 3259 5571 4894 4146

NBP* 4592 4114 3263 5576 4898 4149

Table V

Vapour Pressures at the Melting Points for thePGMs

Element Melting point Vapour pressureT, K p, bar

Ru 2606 1.045 × 10–5

Rh 2236 5.050 × 10–6

Pd 1828.0 4.227 × 10–5

Os 3400 7.753 × 10–5

Ir 2719 9.837 × 10–6

Pt 2041.3 1.896 × 10–7

*NBP = normal boiling point at one atmosphere pressure (1.01325 bar)Values corresponding to the liquid region are given in italics

10 F. T. Ferguson, K. G. Gardner and J. A. Nuth, III,J. Chem. Eng. Data, 2006, 51, (5), 1509

11 E. R. Cohen and B. N. Taylor, ‘The 1986Adjustment of the Fundamental Physical Constants(Report of the CODATA Task Group onFundamental Constants)’, CODATA Bull., No. 63,November 1986

12 P. J. Mohr, B. N. Taylor and D. B. Newell, ‘The2006 CODATA Recommended Values of theFundamental Physical Constants’, (Web Version5.0), April 2007 (database developed by J. Baker, M.Douma and S. Kotochigova) Available:http://physics.nist.gov/constants, NationalInstitute of Standards and Technology,Gaithersburg, Maryland, U.S.A.

13 A. Joueizadeh and S. Johansson, InternationalColloquium on Atomic Spectra and OscillatorStrengths for Astrophysical and Laboratory Plasmas,National Institute of Standards and Technology,Gaithersburg, Maryland, U.S.A., 14th–17thSeptember, 1992, NIST Special Publication 850,April, 1993, p. 131


Appendix ACalculation of Vapour Pressure Values

Vapour pressure values may be calculated from thermodynamic data via an inversion of the thirdlaw of thermodynamics used to calculate the heat of sublimation (Equation (ii)):

RTln(p) = T[δ – (G°T – H°298.15)/T] – ΔH°298.15(III) (ii)

where R is the universal gas constant (8.314472 J mol–1 K–1 (12)); T is the temperature in kelvin; p isthe vapour pressure in bar; ΔH°298.15(III) is the third law enthalpy of sublimation at 298.15 K (Equation(iii));

δ – (G°T – H°298.15)/T = [– (G°T – H°298.15)/T(g)] – [– (G°T – H°298.15)/T(s, l)] (iii)

where (G°T – H°298.15) is the free energy of the phase relative to that at 298.15 K.

Appendix BDerivation of the Selected Vapour Pressure Equation

The variation of vapour pressure (p) with temperature (T) can be represented by the Clapeyronequation (Equation (iv)):

dp/dT = ΔH/TΔV (iv)

where ΔH is the latent heat of sublimation (below the melting point) or vaporisation (above themelting point); ΔV is the change in volume during the transition from condensed phase to gas. Sincethe volume change is very large, the value for the condensed phase can be neglected and the equa-tion becomes Equation (v) or (vi):

dp/dT = ΔHp/RT 2 (v)

d ln(p)/dT = ΔH/RT 2 (vi)

Equation (vi) is then integrated to give the Clausius-Clapeyron equation (Equation (vii)):

ln(p) = A/T + B (vii)

where B is an integration constant, and A = ΔHT/R where ΔHT is the average enthalpy centred onthe mid-range temperature; ΔHT can only be taken as constant over a relatively narrow temperaturerange. The Clausius-Clapeyron equation is used practically to fit experimental vapour pressure mea-surements, and also to calculate the ‘second law’ heat of sublimation through the relationship(Equation (viii)):

ΔH°298.15(II) = – δ(H°T – H°298.15) – RA (viii)

where δ(H°T – H°298.15) and (H°T – H°298.15) (see Equation (ix)) are enthalpy values relative to the

14 R. Engleman, U. Litzén, H. Lundberg and J.-F.

Wyart, Phys. Scr., 1998, 57, (3), 345

15 J. W. Arblaster, Platinum Metals Rev., 1996, 40, (2), 62

16 H. M. Strong and F. P. Bundy, Phys. Rev., 1959, 115,(2), 278

17 R. E. Bedford, G. Bonnier, H. Maas and F. Pavese,Metrologia, 1996, 33, (2), 133


Appendix CMelting Point and Triple Point Differences

Although the pressure at the standard state in thermodynamics is now set at one bar, meltingpoints are still quoted at one atmosphere pressure. This is an arbitrary choice, whereas the triplepoint is that at which the vapour pressures of the solid, liquid and gas are exactly equal; it is there-fore a universal point. The difference between the melting and triple points can be determined fromthe melting pressure curves, for example Strong and Bundy (16) determined the initial values forthese curves to be 138 atm deg–1 for platinum and 160 atm deg–1 for rhodium, equivalent to 0.007deg atm–1 and 0.006 deg atm–1, respectively. Assuming that the triple point pressures are equal tozero, then this indicates that for platinum, the triple point would be 0.007 K below the melting point,and for rhodium 0.006 K. However, these values are negligible when compared to melting pointuncertainties of ± 0.4 K for platinum and ± 3 K for rhodium at the secondary fixed points on theInternational Temperature Scale of 1990 (ITS-90) (17). On these grounds, triple point correctionsmay be regarded as meaningless. These examples for platinum and rhodium are considered as rep-resentative for all six of the pgms.

The AuthorJohn W. Arblaster is Chief Chemist,working in metallurgical analysis on a widerange of ferrous and non-ferrous alloys forstandards in chemical analysis, at ColeshillLaboratories in the West Midlands ofEngland. He is interested in the history ofscience and in the evaluation of thethermodynamic and crystallographicproperties of the elements.

enthalpy at 298.15 K;

δ(H°T – H°298.15) = (H°T – H°298.15)(g) – (H°T – H°298.15)(s, l) (ix)

ΔHT may be taken as constant only for a narrow temperature range. In order to cover a wider tem-perature range it may be expanded to Equation (x):

ΔHT = ΔH0 + a1T + a2T 2 + a3T 3 + … (x)

Substituting Equation (x) into the Clapeyron equation (Equation (iv)) gives Equation (xi):

d ln(p) = ΔH0dT/RT 2 + a1dT/RT + a2dT/R + a3TdT/R (xi)

Rearranging the coefficients and integrating gives Equation (xii):

ln(p) = A + a1ln(T )/R – ΔH0/RT + a2T/R + a3T 2/R (xii)

where A is the integration constant. Substituting B = a1/R, C = – ΔH0/R, D = a2/R, E = a3/R givesEquation (i):

ln(p, bar) = A + B ln(T ) + C/T + DT + ET 2 (i)

136

Amid the current global preoccupation with thelightest element as a means to alleviate the conse-quences of the dwindling supplies of fossil fuels,this book sets out to provide an overview of thearea of dense oxygen and hydrogen transportmembranes and the roles they could play withintechnologies being developed to improve on thepoor efficiencies associated with simple combus-tion processes. The book’s editors, themselveslong-time practitioners in the field of dense mem-brane research, have assembled a group of authorswhose names will be familiar to researchers in the area.

Membrane Applications of thePlatinum Group Metals

Application of the platinum group metals(PGMs) within the area of fully dense membranesfalls into three categories:– The entire membrane may be fabricated from

a PGM or PGM-based alloy, as in the case ofhydrogen transport membranes fabricatedfrom Pd and alloys thereof.

– The PGM phase can be added to a ceramicoxide phase to form a ‘cermet’ material.

– Thin PGM coatings may be deposited uponthe surface of fully dense materials to improvedissociation/association kinetics, act as protec-tive coatings or as electrode materials in drivenmembrane systems.Each of these areas is discussed, in varying

lengths, within the context of the various chapters.The first two chapters cover the area of hydro-

gen-permeable, dense, ceramic materials withmixed protonic-electronic conductivity that arebased almost exclusively upon oxides. Of specificinterest to the PGM industry are the sections deal-

ing with the cermet materials, where pairing ofoxides with metals (typically 10 to 40 vol.% of pal-ladium, platinum, palladium alloys or refractorymetals such as tantalum or niobium etc.) allowsresearchers to circumvent potential problems relat-ing to having both conduction types in a singlephase. The practical caveat with these materials istheir poor mechanical stability, primarily due tointernal interfacial stresses.

In a reference-laden Chapter 3, Stephen N.Paglieri (Los Alamos National Laboratory, U.S.A.)amply highlights the huge effort that has beendevoted to fabricating thin palladium-based hydro-gen diffusion membranes. ‘Thick’ membranes,capable of achieving ultra-high purity hydrogen,have been available for over thirty years, but theseare prohibitively expensive for applications thatrequire only moderate purity levels. The authorprovides a concise overview of the areas that mustbe addressed in order to produce a viable thinmembrane – the nature of the support, depositionand/or alloying techniques for the active layer, theeffects and mitigation of poisons, the difficultiessurrounding sealing of the membranes into hous-ings, and of course the final cost implications. It isa measure of the dedication of the membraneindustry and funding bodies that so much fundinghas been made available in the quest for whatremains an elusive product.

The metallic competition to Pd systems isdescribed in Chapter 4 by Michael V. Mundschau,Xiaobing Xie and Carl R. Evenson (EltronResearch Inc). The so-called ‘superpermeable’membranes, based upon niobium, tantalum andvanadium, were developed for the nuclear industryto separate hydrogen isotopes from helium in plas-mas and from molten metal cooling fluids. The

“Nonporous Inorganic Membranes: forChemical Processing”EDITED BY ANTHONY F. SAMMELLS AND MICHAEL V. MUNDSCHAU (Eltron Research Inc, U.S.A.), Wiley-VCH, Weinheim, Germany,

2006, 291 pages, ISBN 978-3-527-31342-6, £80.00, €120.00

Reviewed by Hugh HamiltonJohnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading RG4 9NH, U.K.; E-mail: [email protected]


DOI: 10.1595/147106707X216800

burgeoning interest in the hydrogen economy andthe high cost of Pd have reinvigorated research inthese alloys as an alternative for hydrogen purifica-tion in the chemical and energy industries.However, these metals are very prone to formationof surface oxides, carbides and nitrides that poisonthe hydrogen dissociation reaction. This leadsdevelopers to coat the surfaces with PGM layerswhich in part protect the surface from poisoningand facilitate hydrogen ingress into the membranebeneath. The importance of matching the expan-sion properties of the catalyst layer with themembrane and the criticality of the depositionmethod are highlighted, but the authors clearlybelieve that these membrane systems will providea commercially viable industrial membrane.

The down-to-earth comments by David J.Edlund at the start of Chapter 5 remind us that,without a suitable module, a membrane is merelya laboratory curiosity. Of the myriad groups devel-oping new membrane compositions only a few areconcurrently engineering new modules, which fallinto two designs – those that will accommodatefoils and those for tubular membranes. ‘Thick’palladium-silver alloy membrane systems, such asthose available from Johnson Matthey (1), tend tobe tubular in form, primarily because during the1960s when they were developed, sealing of foilswas problematic. Nevertheless, a huge amount ofwork on brazing alloys was necessary before com-mercial systems were suitably reliable. Subsequentdevelopment of palladium-copper alloy foil hasallowed modules based around stacks of foils tobe constructed successfully for commercial appli-cations. This excellent chapter provides theengineering viewpoint that is often neglected bymembrane developers.

The following three chapters cover the area ofoxygen transport membranes which, although fas-cinating materials with great possibilities, haveonly very limited opportunities for PGM usage.

Early researchers doped oxide materials with Pdor Pt to form ‘dual phase’ structures in which thedopant introduced electronic conductivity.However, cost and long-term mechanical instabil-ity resulted in cessation of this practice. ThePGMs have also been introduced onto the mem-brane surface to act as catalysts. For example,“significant improvements” in ethene yields andselectivities were noted when dispersed Pd wasintroduced onto the membrane for alkane dehy-drogenation reactions.

The book finishes with discussion of the eco-nomics of membrane reactors using the water gasshift reaction over a Pd membrane as a case study.In agreement with the comments by Edlund inChapter 5, the author notes that, although thepotential of membrane reactors has been widelyacclaimed, the technology has yet to proceedbeyond laboratory scale. There are several techni-cal barriers to be overcome, including sealing,thermal cyclability, high temperatures and pres-sures in aggressive environments, catalyst andmembrane poisoning. However, the reviewerwould like to offer the thought that, while one ofthe most widely cited barriers is the cost of Pdmetal, this is in effect an upfront cost. The Pd willultimately be easily recoverable and, if the metalprice has risen during the lifetime of the mem-brane, could be considered to be an investment.

Reference1 Johnson Matthey Gas Purification Technology:

http://www.jmgpt.com/html/all_data_sheets_link.html

The Reviewer

Dr Hugh Hamilton has worked at the JohnsonMatthey Technology Centre for nearly 19 years,during which time he has researched in a rangeof areas including autocatalysts, palladiummembranes, fuel cell MEA manufacture andhydrogen storage. His current role includessorbent development for Hg from syngas and Tipowder metal injection moulding processdevelopment.



Platinum-Copper on Carbon CatalystSynthesised by Reduction with Hydride AnionINITIAL FINDINGS ON REACTIVITY AND DISPERSION CHARACTERISTICS

By Hany M. AbdelDayemChemistry Department, Faculty of Science, Ain Shams University, 11566-Abassia, Cairo, Egypt; E-mail: [email protected]

With a view to improving the catalytic performance of supported bimetallic platinum-coppercatalysts in hydrogen-assisted dechlorination of halogenated alkanes, a range of catalystswere prepared by reduction of oxide precursors by the hydride anion H–, using both sodiumand calcium hydrides (NaH and CaH2). The catalytic performance of the resulting catalystsamples in the hydrodechlorination (HDCl) of 1,2-dichloroethane at 220ºC was investigated,to gain understanding of metal alloying phenomena governing the variation in ethene selectivitywith time on stream (TOS). Metal dispersion was also investigated by O2 chemisorption andtransmission electron microscopy (TEM). PtCuCaH(b) catalyst, synthesised by reduction withCaH2 at 450ºC, showed a high selectivity towards ethene in comparison with that of catalystssynthesised by reduction with either NaH or hydrogen. In view of the chemisorption and TEMresults, the significant high selectivity of this catalyst towards ethene was attributed to the factthat reduction by CaH2 enhanced alloying of Pt and Cu. On the other hand, the ethene selectivityof PtCuCaH(b) catalyst did not show any variation with TOS, but reached a steady state atearly TOS. This suggested that Pt and Cu alloying did not take place during the course ofthe reaction, but might have occurred during the reduction process.

DOI: 10.1595/147106707X214316

Bimetallic alloys are versatile catalysts for impor-tant industrial processes, including Fisher-Tropschsynthesis, fuel reforming, hydrogenation and dehy-drogenation of alkanes, decomposition ofmethanoic acid, and hydrodechlorination ofchloroorganic compounds (1). Selectivity maypotentially be greatly improved through therational design of bimetallic alloy catalysts from fun-damental principles (2). However, the preparationof supported bimetallic alloy catalysts presents someproblems as compared with model catalysts (3).

The production of alkenes from halogenatedalkanes by hydrogen-assisted dechlorination oversupported metal catalysts has received considerableattention in the past few years (4–7), especially thehydrodechlorination of 1,2-dichloroethane toethene (7–10). Supported monometallic noble metalcatalysts (from Group VIII) are very active forHDCl reactions (11, 12). Extensive research withmonometallic catalysts (platinum, palladium, silver,copper, etc.) has shown that the dechlorinated C2H4

is immediately converted into the fully hydrogenat-

ed product, C2H6, which is much less industriallyuseful (7, 13). However, several authors havedemonstrated that bimetallic catalysts, composed ofalloys such as Pt-Cu, Pd-Cu and Pd-Ag, possess ahigh ability to convert chlorinated alkanes selective-ly into non-chlorinated alkenes (10, 14–16). In thecase of supported Pt-Cu catalyst, earlier investiga-tion suggested that the increase in ethene selectivitywith TOS was due to an increase in the degree ofalloying of Pt and Cu under reaction conditions(17). On the other hand, recent results indicate thatalloying of Pt and Cu depends on the pretreatmentreduction conditions of the catalyst (18).

The hydride anion is one of the most powerfulreducing agents known; the reduction potential ofthe H–/H2 couple has been estimated at –2.25 V(19). LiAlH4 and KBH4 are used extensively asreducing reagents in solution chemistry (20).Recently, NaH and CaH2 have been used as power-ful reducing agents in solid-state topotacticreduction, at lower temperatures than would berequired for a H2 gas process (21–23).


The present work addresses the following ques-tions:– Can we discover a new route for enhancing the

alloying of Pt and Cu in supported catalysts,based on the reduction of their metal oxide pre-cursors in solid-state reaction using metalhydrides?

– Can this suggest a new way to master catalystselectivity in HDCl reactions?The bimetallic catalyst, Pt-Cu supported on car-

bon, was prepared by reduction of its oxideprecursor, using different metal hydrides (CaH2

and NaH) at different temperatures. Hydrogen wasalso used to prepare a reference sample. The reac-tivity of all catalysts studied was tested for thehydrodechlorination of 1,2-dichloroethane (DCE)as a model reaction, looking for a significant mod-ification in catalytic performance. Oxygenchemisorption and TEM were used to investigatemetal dispersion.

Experimental WorkCatalyst Preparation

0.5 wt.% Pt-0.5 wt.% Cu/C catalyst was pre-pared by co-impregnation of a carbon support(Aldrich, Darco® KB, SBET 1500 m2 g–1 and porevolume 2 ml g–1) with an aqueous solution contain-ing the appropriate quantity of H2PtCl6·6H2O(Aldrich, 99.9%) and CuCl2·2H2O (Aldrich, 99.9%)overnight. The material was allowed to dry at roomtemperature for 24 h, and then at 100ºC for 2 h invacuo. The produced solid was calcined at 200ºC for4 h to give an oxide precursor sample. This sampleof PtO and CuO was mixed and ground with atwofold stoichiometric excess of the metal hydride(CaH2, Aldrich, 99.9% or NaH, Aldrich, 95%) in aHe-filled glove box, and then sealed in an evacuat-ed Pyrex ampoule (p < 2 × 10–4 torr). The sealedreaction vessel was then heated for two periods of4 days at 300ºC with intermediate grinding. Thebyproduct (CaO or Na2O) and any unreactedmetal hydride from the reaction mixture wereremoved from the produced solid by washing witha solution of 1M NH4Cl in CH3OH, in a Schlenkfilter under a nitrogen atmosphere. The productwas then further washed with CH3OH beforebeing dried under vacuum (p < 1 × 10–1 torr). The

complete removal of Ca and Na from the catalystsamples was confirmed by energy dispersive X-ray(EDX) analysis.

Another catalyst sample was prepared by reduc-tion with CaH2 at 450ºC, following the sameprocedure as above. The catalyst prepared byreduction with NaH was denoted as PtCuNaH,and catalysts prepared by reduction with CaH2 at300 and 450ºC were denoted as PtCuCaH(a) andPtCuCaH(b), respectively. A reference sample(denoted as PtCuH2) was prepared by reductionwith 10% H2/He gas mixture at 300ºC for 4 h.Monometallic catalysts, 0.5 wt.% Pt/C and 0.5wt.% Cu/C, were also prepared by reduction withhydrogen.

Dispersion StudyMetal dispersion in the catalysts was deter-

mined from O2 chemisorption at 350ºC. Thecatalyst was charged into the adsorption cell in aHe-filled glove box to avoid any oxidation of thecatalyst. Before measurements were taken, the cat-alyst was heated in vacuo at room temperature for4 h at 1.33 × 10–3 Pa. The net adsorption(μmol g–1) on the supported metal catalyst was cal-culated by measuring adsorption on thenonmetallised support, and subtracting this valuefrom total adsorption on the catalyst. The netadsorption was simply converted into the apparentdegree of dispersion (Dapp = [O]/[Pt + Cu]), whichis regarded as a measure of the number of atomsof adsorbed oxygen per metal atom (24–26). Theadsorption stoichiometry was assumed to equal 2.

Transmission electron micrographs wereobtained using a JEOL 1200 EX II transmissionelectron microscope operated with an accelerationvoltage of 50 kV.

Catalytic TestsHydrodechlorination of 1,2-dichloroethane

(DCE) was carried out at 220ºC in a flow reactor.Catalyst was charged into a quartz reactor (12 mminternal diameter) in the He-filled glove box toavoid any oxidation of catalyst. Then the reactorwas connected directly to the catalytic system in aflow of H2 (4 ml min–1)/He (28 ml min–1) at 110ºC.After 1 h, the gas stream was switched to a mixed


flow of 41 ml min–1, consisting of DCE (7000ppm), H2 (35,000 ppm) and the remainder He. Thefeed and product streams were analysed using aShimadzu GC-17A gas chromatograph equippedwith a 27.5 mm Chrompack PoraPLOT capillarycolumn and a flame ionisation detector. Note thatthe HCl detected was not quantified in this study.The weight of bimetallic catalyst was adjusted tomaintain the conversion at approximately 2%.

Results and DiscussionSelectivity and Reactivity

Figures 1 to 4 represent the variation of ethaneand ethene selectivities with time on stream. In thecase of PtCuH2 catalyst, both ethane and etheneselectivity changed by only 5% during the courseof the reaction (Figure 1). However, the variationsin selectivity were more pronounced for thePtCuNaH catalyst: before a steady state wasachieved, the change observed was ~ 15% (Figure2). As shown in Figure 3, in the case of

PtCuCaH(a) there was an initial sharp rise in selec-tivity towards ethene, from 70 to 80% during thefirst 3 h TOS; after which it reached steady state(at 7.5 h), with selectivity to ethene ~ 82%.However, bimetallic PtCuCaH(b) (Figure 4) didnot show significant variation in the selectivitytowards products with TOS. On the other hand,none of the catalysts under investigation showedsignificant variation in the conversion ofdichloroethane with TOS.

The activity results for the hydrodechlorinationof 1,2-dichloroethane at steady state over the stud-ied catalysts are summarised in Table I. It is clearthat the PtCuCaH(b) catalyst exhibited the highestselectivity (100%) towards ethene, but with lowactivity. Ethene is also the major product detectedwith the PtCuCaH(a) catalyst, showing selectivity~ 82%, with a higher activity than for thePtCuCaH(b) catalyst. Both PtCuH2 and PtCuNaHcatalysts were selective for the dechlorination of1,2-dichloroethane to ethane; the major product

0102030405060708090

100

0 5 10 15 20 25TOS, h

Sele

ctiv

ity, %

EthyleneEthane

0 5 10 15 20 25Time on Stream, h

100908070605040302010

0

Sel

ectiv

ity, %

EtheneEthane

Fig. 1 Selectivities toward etheneand ethane vs. time on stream forhydrodechlorination of1,2-dichloroethane over PtCuH2

catalyst at 220ºC

0102030405060708090

100

0 5 10 15 20 25TOS, h

Sele

ctiv

ity, %

EthyleneEthane

Sel

ectiv

ity, %


100908070605040302010

0

EtheneEthane

Fig. 2 Selectivities toward etheneand ethane vs. time on stream forhydrodechlorination of1,2-dichloroethane over PtCuNaHcatalyst at 220ºC

was ethane, at ~ 81% and ~ 78 %, respectively.However, monometallic Pt/C is the most selectiveto ethane, at ~ 94%. Monometallic Cu/C catalysthas high ethene selectivity, ~ 93 %, but its activityis very low.

Dispersion ResultsThe chemisorption results are summarised in

Table II. The net adsorption (micromoles oxygenper gram of sample) on the metal was taken as thedifference between the total adsorption on the cat-alyst and the adsorption on the correspondingsupport. From this, it is evident that the values ofthe degree of dispersion of PtCuCaH catalysts arelower than for the other catalysts. A more detailedstudy is required before quantitative dispersion


0102030405060708090

100

0 5 10 15 20 25TOS, h

Sele

ctiv

ity, %

Ethylene

Ethane

Sel

ectiv

ity, %


100908070605040302010

0

0

20

40

60

80

100

120

0 5 10 15 20 25TOS, h

Sele

ctiv

ity, %

Ethylene Ethane

Sel

ectiv

ity, %


120

100

80

60

40

20

0

EtheneEthane

Ethene

Ethane

Fig. 3 Selectivities toward etheneand ethane vs. time on stream forhydrodechlorination of1,2-dichloroethane overPtCuCaH(a) catalyst at 220ºC

Fig. 4 Selectivities toward etheneand ethane vs. time on stream forhydrodechlorination of1,2-dichloroethane overPtCuCaH(b) catalyst at 220ºC

Table II

Chemisorption Data of Oxygen on the Surface ofDifferent Samples of Pt-Cu/C Catalysts

Catalyst Net adsorption, Degree of dispersion,

μmol g–1 Dapp = [O]/[Pt + Cu]

Pt/C 11.8 0.92

PtCuH2 14.2 0.74

PtCuNaH 15.4 0.80

PtCuCaH(a) 14.6 0.24

PtCuCaH(b) 2.0 0.10

Table I

Steady-State Activity Parameters of Pt-Cu/CCatalysts for the Hydrodechlorination of 1,2-Dichloroethane

Catalyst Time on Selectivity, mol% Activity,

stream, h Ethene Ethane μmol min–1 g–1

Pt/C 3 6 94 2.0PtCuH2 13 19 81 7.7PtCuNaH 17 22 78 5.2PtCuCaH(a) 8 82 18 6.7PtCuCaH(b) 0.5 100 0 1.4Cu/C 9 93 7 0.05

changes are discussed, but the present results couldgive a qualitative account of the relative dispersionobtained.

TEM micrographs of the PtCuH2, PtCuNaHand PtCuCaH(b) catalysts are shown in Figure 5.The micrographs of both PtCuH2 and PtCuNaH

(Figures 5(a) and 5(b)) show small Pt and Cu par-ticles in a highly dispersed state, without significantaggregations. However, aggregated large particlesof Pt and Cu are shown in the micrograph of thePtCuCaH(b) catalyst (Figure 5(c)).

Catalytic ActivityRecent studies on the HDCl of DCE over Pt-M

(M = Cu, Ag) catalysts proposed that alloying of Ptand M atoms played an effective role in deter-mining the selectivity of the catalyst towardsethene (4, 9). Recalling the present activity results,the observed difference in ethene selectivitybetween the catalysts studied may be due to thedifferent capacities of various reducing agents toenhance the alloying of Cu and Pt. In the case ofPtCuH2 catalyst, the microcrystallites of Pt and Cuions might be separated on the support surfaceafter drying due to chromatographic effects (1).These effects are expected to be especially pro-nounced due to the high surface area and narrowpores of the activated carbon support used. Thus,if Pt and Cu chlorides in a dry catalyst are not inclose proximity to one another, the standardreduction by hydrogen would not result in signifi-cant alloying (as shown in Figure 5(a)). Thissuggestion can also be verified from the highapparent degree of dispersion of Pt and Cuobserved in PtCuH2 (Table II).

We have shown that PtCuCaH catalysts havethe highest ethene selectivity among the catalystsstudied. According to the TEM and chemisorptionresults, reduction by CaH2 may promote surfacemigration of Pt and Cu, which results in alloying ofthe two metals. It is important to note that thechange in ethene selectivity of PtCuCaH(b) cata-lyst is independent of TOS; this result suggests thatPt and Cu alloying did not take place during thecourse of the reaction, but might have occurredduring the reduction process. This phenomenonmight be attributable to the different processes forthe reduction of Pt-Cu catalyst by metal hydridesand by H2. However, the observed low etheneselectivity of PtCuNaH, which was reduced fol-lowing the same procedure as for PtCuCaH(a),excludes this possibility. Furthermore, PtCuNaHcatalyst showed approximately the same ethene


10nm(c)

10nm

Pt

Cu

(b)

Fig. 5 TEM micrographs of the catalysts: (a) PtCuH2;(b) PtCuNaH and (c) PtCuCaH(b)

10nm

Pt

Cu

(a)

selectivity as PtCuH2 catalyst, independently of thedifferent reduction procedures used. It is wellknown in the literature that many of the transitionmetal oxychlorides migrate more easily over thesurface of a carbon support than do the pure metalatoms (27). Hayward et al. (22, 23) reported thatreduction of metal oxides by CaH2 (hydride anion)could produce metal oxide hydrides. Thus, we canenvisage that during the reduction with CaH2, Ptand/or Cu, oxy-hydride species might be formedas intermediates of high surface mobility. The for-mation of these intermediates could facilitate theaggregation of Pt and Cu, as shown from themicrographs and their enhanced alloying. To clar-ify this hypothesis further, specific characterisationstudies are required; these are beyond the scope ofthis work.

As already observed, PtCuCaH(a) catalyst pre-pared by reduction with CaH2 at 300ºC has ahigher ethene selectivity than does PtCuNaHreduced by NaH at the same temperature. Thismay be due to the difference in thermal stabilitybetween CaH2 and NaH. CaH2 has a high decom-position temperature (~ 885ºC), therefore mainlyH– may be present as reducing agent at the tem-peratures used (300 and 450ºC) (26). However,NaH thermally decomposes at a lower tempera-ture (~ 220ºC) (26); in this case, hydrogen ispresent in thermal equilibrium with H– duringreduction of the catalyst, and the probability offormation of Pt and Cu oxy-hydride intermediatesis decreased. It seems that 450ºC is an optimumtemperature to reduce the Pt-Cu catalyst for theselective HDCl of DCE to ethene, but furtherinvestigations are necessary to determine preciselythe most efficient temperature.

Finally, the catalytic performance of thePtCuCaH(b) catalyst might also be affected by thepresence of trace amounts of calcium from thehydride precursor. As mentioned above, the com-plete removal of Ca from all catalysts wasconfirmed by EDX analysis, but one cannotexclude the possibility that highly dispersed Caspecies may be present on a support of high sur-face area, such as the carbon used here. The Caissue is outside the scope of this article; however,some conclusions can be drawn on the basis of lit-

erature results. It has been reported that calciumcan improve the dispersion of both supported Cuand Pt (29, 30). This effect might lead to a changein the geometry of Pt sites responsible forhydrogenolysis of 1,2-dichloroethane to ethane.Nevertheless, at the same time, the greatest effectof Ca is to decrease the possibility of alloying Ptand Cu. The above interpretation suggests that Cadoes not play a decisive role in determining theselectivity of PtCuCaH catalysts towards ethene.

ConclusionsThe results discussed above demonstrate that

supported Pt-Cu catalysts synthesised by reduc-tion of oxide precursors with CaH2 exhibit highethene selectivity, because alloying of Pt and Cuwas enhanced during the reduction process. CaH2

apparently affords the hydride anion as a reducingspecies at the reduction temperatures used. Theformation of intermediate Pt and/or Cu oxidehydride species of high surface mobility was pro-posed. To clarify this hypothesis, further specificcharacterisation studies are still required; these arebeyond the scope of the present work. As areagent, CaH2 is easily to handle, readily available,and has a high decomposition temperature, allow-ing the reduction of various supported bimetalliccatalysts over a wide temperature range.

References1 V. Ponec and G. C. Bond, “Catalysis by Metals and

Alloys”, eds. B. Delmon and Y. T. Yates, Studies inSurface Science and Catalysis, Vol. 95, Elsevier,Amsterdam, 1995; for a review see: D. E. Webster,Platinum Metals Rev., 1996, 40, (2), 70

2 J. Greeley and M. Mavrikakis, Nature Mater., 2004, 3,(11), 810

3 E. Antolini, Mater. Chem. Phys., 2003, 78, (3), 5634 B. Heinrichs, F. Noville, J.-P. Schoebrechts and J.-

P. Pirard, J. Catal., 2003, 220, (1), 2155 L. S. Vadlamannati, V. I. Kovalchuck and J. L. d’Itri,

Catal. Lett., 1999, 58, (4), 1736 L. S. Vadlamannati, D. R. Luebke, V. I. Kovalchuck

and J. L. d’Itri, in: “12th International Congress onCatalysis”, eds. A. Corma, F. V. Melo, S. Mendiorozand J. L. G. Fierro, Proceedings of the 12th ICC,Granada, Spain, 9th–14th July, 2000, Studies inSurface Science and Catalysis, Vol. 130, Elsevier,Amsterdam, 2000, pp. 233–238

7 B. Heinrichs, J.-P. Schoebrechts and J.-P. Pirard, J.Catal., 2001, 200, (2), 309

8 B. Heinrichs, F. Noville, J.-P. Schoebrechts and J.-



P. Pirard, J. Catal., 2000, 192, (1), 1089 B. Heinrichs, F. P. Delhez, J.-P. Schoebrechts and J.-

P. Pirard, J. Catal., 1997, 172, (2), 32210 S. Lambert, F. Ferauche, A. Brasseur, J.-P. Pirard

and B. Heinrichs, Catal. Today, 2005, 100, (3–4), 28311 S. Lambert, J.-F. Polard, J.-P. Pirard and B.

Heinrichs, Appl. Catal. B: Environ., 2004, 50, (2), 12712 S. Lambert, C. Cellier, P. Grange, J.-P. Pirard and B.

Heinrichs, J. Catal., 2004, 221, (2), 33513 P. Delhez, B. Heinrichs, J.-P. Pirard and J.-P.

Schoebrechts, Solvay SA, U.S. Patent 6,072,096; 200014 S. Lambert, B. Heinrichs, A. Brasseur, A. Rulmont

and J.-P. Pirard, Appl. Catal. A: Gen., 2004, 270,(1–2), 201

15 W. Zhu, J. Zhang, F. Kapteijn, M. Makkee and J. A.Moulijn, in: “Catalyst Deactivation 2001”, eds. J. J.Spivey, G. W. Roberts and B. H. Davis, Proceedingsof the 9th International Symposium, Lexington, KY,U.S.A., October, 2001, Studies in Surface Scienceand Catalysis, Vol. 139, Elsevier, Amsterdam, 2001,pp. 21–28

16 D. Chakraborty, P. P. Kulkarni, V. I. Kovaluchukand J. L. d’Itri, Catal. Today, 2004, 88, (3–4), 169

17 “Metal-Support Interactions in Catalysis, Sintering,and Redispersion”, eds. S. A. Stevenson, J. A.Dumesic, R. T. K. Baker and E. Ruckenstein, VanNostrand Reinhold, New York, 1987

18 D. R. Luebke, L. S. Vadlamannati, V. I. Kovalchuckand J. L. d’Itri, Appl. Catal. B: Environ., 2002, 35, (3),211

19 “Advanced Inorganic Chemistry”, 4th Edn., eds. F.A. Cotton and G. Wilkinson, Wiley-Interscience,New York, 1980

20 C. Tsang, J. Kim and A. Manthiram, J. Solid StateChem., 1998, 137, (1), 28

21 K. Poeppelmeier, Science, 2002, 295, (5561), 184922 M. A. Hayward, M. A. Green, M. J. Rosseinsky and

J. Sloan, J. Am. Chem. Soc., 1999, 121, (38), 884323 M. A. Hayward, E. J. Cussen, J. B. Claridge, M.

Bieringer, M. J. Rosseinsky, C. J. Kiely, S. J. Blundell,I. M. Marshall and F. L. Pratt, Science, 2002, 295,(5561), 1882

24 H. M. AbdelDayem, Adsorpt. Sci. Technol., 2004, 22,(9), 755

25 S. A. Hassan, H. A. Hassan, K. M. Hashem and H.M. AbdelDayem, Appl. Catal. A: Gen., 2006, 300, (1),14

26 H. M. AbdelDayem, H. A. Hassan and S. A. Hassan,J. Dispersion Sci. Technol., 2007, 28, in press

27 J. R. Anderson, “Structure of Metallic Catalysts”,Academic Press, London, 1975

28 “Chemistry of the Elements”, 2nd Edn., eds. N. N.Greenwood and A. Earnshaw, Butterworth-Heinemann, Oxford, U.K., 1997

29 J. Wu, Y. Shen, C. Liu, H. Wang, C. Geng and Z.Zhang, Catal. Commun., 2005, 6, (9), 633

30 G. C. Cabilla, A. L. Bonivardi and M. A. Baltanás,Appl. Catal. A: Gen., 2003, 255, (2), 181

The AuthorHany AbdelDayem holds a B.Sc. degree in Chemistry and an M.Sc. incatalysis from Ain Shams University, Egypt, as well as a Ph.D. inChemistry for research on chemical kinetics and catalysis from theUniversité catholique de Louvain, Belgium. He subsequently held aFulbright postdoctoral award at the University of Pittsburgh, U.S.A. anda postdoctoral fellowship from the Agence universitaire de laFrancophonie (AUF) at the Université catholique de Louvain. In 2001 hewas appointed Assistant Professor in the Chemistry Department at AinShams University. He joined King Faisal University, Saudi Arabia, asAssistant Professor in 2005. His main focus is the development of

heterogeneous catalytic processes, including oxidations/oxidative dehydrogenations andhydrodechlorination. He has particular interests in catalyst synthesis, and in ex- and in situcharacterisation of catalysts and reaction kinetics.

The Society of Automotive Engineers (SAE)Fuels and Emissions Conference was held from23rd to 25th January 2007 in Cape Town, SouthAfrica. The main conference sponsor was Sasol Ltd.It is the first time the SAE have held a conferencein South Africa. It was pointed out that this is animportant event for Africa, because it brings togeth-er engineers, scientists and suppliers for a globaldiscussion on the latest evolving technologies infuels and lubricants, as well as future emissions controls.

The conference centred on a keynote lecture byan eminent speaker on each morning, followed bytwo or three parallel sessions of technical presenta-tions. These sessions covered a wide array of topics,such as alternative and potential new fuels, hydro-gen, engine technology, emissions control, andmeasurement and calibration techniques. Platinumgroup metals (pgms) feature prominently in theseareas, for applications from fuel reforming to emis-sions control catalysis. The keynote lectures aredescribed in some detail here, along with a briefsummary of some of the other work from the tech-nical sessions. Technical paper numbers are given inparentheses following the title of each paper, andare available through the SAE (1).

Synthetic FuelsJohannes Botha (Sasol Ltd, South Africa) got the

proceedings underway with a keynote address on‘Synthetic Fuels’. The presentation concentrated oncoal/gas/biomass-to-liquid technologies (CTL,GTL, BTL, respectively) from the standpoint ofSasol, covering drivers, history, technologicaladvances and future importance. The use of coal asa feedstock has been historically important to SouthAfrica for a number of reasons, not least because ithas large coal reserves; there is also the desire tobecome less dependent on imported oil. Today,30% of transportation fuels in South Africa

(equivalent to 205,000 barrels per day (bpd)) aresupplied by coal- or natural gas-fed facilities, such asSasol’s GTL operations. Coal may well also be thefuel of the future in other countries and regions,such as China, India, the U.S.A. and Russia, where alarge proportion of fossil fuel reserves are as coal.The environmental benefits of using synthetic fuelswere highlighted: synthetic fuels burn much morecleanly than do conventional fuels derived fromcrude oil, resulting in lower particulate and sulfuremissions. Greenhouse gas emissions are also lowerthan with crude oil for BTL technology, and com-parable to those with crude oil for GTL technology.The potential future importance of synthetic fuelsproduced by the CTL, GTL and BTL processes isevident from the number of plants currently underconstruction or in the planning stages in all theworld’s inhabited continents. However, even if allthe current probable and possible plants were built,the amount of synthetic fuel produced would still beonly a fraction of worldwide demand.

The exhaust emissions advantages of diesel fuelsderived from GTL technology were also highlight-ed by Monica Larsson and Ingemar Denbratt(Chalmers University of Technology, Sweden): ‘AnExperimental Investigation of Fischer-TropschFuels in a Light-Duty Diesel Engine’ (2007-01-0030). Two synthetic diesel fuels were tested, andtheir performance compared with that of conven-tional diesel fuel in a single-cylinder research engine,where the effect of injection timing and exhaust gasrecirculation could be studied. Lower carbonmonoxide (CO), total hydrocarbon (THC) and sootemissions were seen with the synthetic GTL diesel.The improved particulate matter (PM) and hydro-carbon emissions with synthetic diesel were alsoconfirmed by Taku Tsujimura and coworkers(National Institute of Advanced Industrial Scienceand Technology (AIST), Japan), in a joint study withthe Mitsubishi Corporation: ‘A Study of PM

145

2007 Fuels and Emissions ConferenceA SELECTIVE REPORT ON THE SAE INTERNATIONAL CONFERENCE

Reviewed by Andrew P. E. YorkJohnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading RG4 9NH, U.K.; E-mail: [email protected]


DOI: 10.1595/147106707X214703

Emission Characteristics of Diesel Vehicle Fueledwith GTL’ (2007-01-0028). Furthermore, PaulSchaberg and colleagues (Sasol Ltd), in a study withSasolChevron Consulting Ltd and DaimlerChryslerAG, found that very low NOx emissions wereachievable with GTL diesel fuel, when used in com-bination with some economically andtechnologically viable engine hardware changes:‘HSDI Diesel Engine Optimisation for GTL DieselFuel’ (2007-01-0027).

Future Fuels and TechnologiesDay two began with a keynote lecture by Hans-

Otto Herrmann (DaimlerChrysler AG, Germany).The lecture, entitled ‘The Role of Future Fuels forSustainable Mobility’, was concerned with address-ing the challenges for future mobility and identifyingthe major stakeholders in finding solutions to pollu-tion problems, before concentrating on the role ofthe automotive industry. Five chronological stepstoward “Energy for the future” were demonstratedfrom a DaimlerChrysler standpoint: (a) optimisation of combustion engines(b) improvement of conventional fuels(c) CO2-neutral biofuels(d) hybrid vehicles(e) fuel cell technology.

The immediate challenge is to improve internalcombustion engine performance, so that dieselemissions can be reduced to the very low levelsassociated with gasoline engines, while gasolineengine fuel consumption is improved to match thatof diesel engines. This is now becoming a reality inEurope due to such technical advances as exhaust

gas recirculation (EGR), improved fuel quality, on-board diagnostics (OBD), and pgm-containingemissions control devices (e.g. filter systems, NOxtraps and selective catalytic reduction (SCR) sys-tems). (While SCR systems are not usuallypgm-based, SCR systems contain pgms for the pur-pose of ammonia slip control.) However, theexample diesel vehicle shown in the lecture relied ona NOx trap for NOx control. It is therefore notavailable in the U.S.A., because ultra-low sulfurdiesel (ULSD) fuel is required. Herrmann’s lectureaddressed DaimlerChrysler’s approach to aftertreat-ment, and two systems were presented: Bluetec Iand Bluetec II (Figure 1). In Bluetec I, most of theNOx control is performed by the NOx trap compo-nent. However, when trap regeneration occurs,under rich engine conditions, some ammonia is pro-duced; this can be stored on the SCR catalyst andused for some additional NOx removal activity.This system relies on sulfur-free fuel and has nourea injection.

The Bluetec II system is more expensive thanBluetec I, and employs urea injection (Adblue) infront of the SCR catalyst to control the NOx emis-sions. Depending on which system is chosen,DaimlerChrysler can achieve 50 to 80% NOxremoval. The lecture also included some discussionon fuels, including detailing the importance ofimproving conventional fuel quality. It went on toconsider the diversification of fuel types, startingwith synthetic fuels such as those produced bySasol, then leading on to biofuels, and in the future,hydrogen produced by renewable energy. By thisroute, fuels become progressively cleaner, in terms


Fig. 1 Schematic ofDaimlerChrysler’s Bluetec I andBluetec II systems. DOC = dieseloxidation catalyst; LNT = leanNOx trap; DPF = dieselparticulate filter; SCR = selectivecatalytic reduction; Adblue = ureainjection

DOC

DOC

LNT

DPF

DPF SCR

SCR

Adblue

Bluetec I

Bluetec II

of both PM/NOx and CO2 emissions. Herrmannpointed out that GTL synthetic fuels can give someemissions advantages with no engine modifications,but by modifying and adapting an engine for GTLfuel it may be possible to achieve 70% NOx emis-sions reduction without a NOx removal device, ascompared with the baseline emissions for a Euro IVengine. Therefore, the use of GTL fuels is a possi-ble approach to meeting future stringent emissionslegislation, though the scenario described requiresdedicated GTL engines and GTL fuels which areboth readily available and economically viable.Biodiesel was briefly discussed, particularly jatrophabiodiesel in India (discussed in detail later).Biodiesel is currently two to three times moreexpensive than conventional diesel fuels, but itsimpact on CO2 emissions is potentially huge.Finally, the route to fuel cell vehicles was outlined,with hybrid technology as an interim step, and thisis expected to be especially important in Japan andthe U.S.A. However, Herrmann was concernedthat, despite the potential advantages of fuel cellvehicles, including zero emissions, low noise andhigh efficiency, there are still a number of chal-lenges to be overcome.

Other measures discussed by Herrmann asimportant for sustainable mobility included: contri-butions from policy makers, such as improved roadand traffic management and traffic light synchroni-sation; from consumers and drivers, for example,20% fuel consumption savings can be made byadaptive driving behaviour, efficient accelerationand optimised gear shifting; and the oil industry,including optimised fuels, renewable fuels (BTL),and a hydrogen infrastructure. Herrmann’s conclu-sion was that, for optimum results, interactionbetween the oil and auto industries is necessary.

Catalysts and Converter TechnologiesLouise Arnold and coworkers (Johnson Matthey

PLC, U.K.) presented: ‘Development andApplication of New Low Rhodium Three-WayCatalyst Technology’ (2007-01-0046). This isimportant because the rhodium price has increasedenormously recently, while 83% of world rhodiumusage is in three-way catalysts (TWCs). By employ-ing advanced washcoat formulations and coating

techniques, it has been possible to reduce rhodiumloadings to 1 g ft–3 (from 5 g ft–3), while still meet-ing Euro IV and Euro V (2) emissions limits, evenafter extended engine-bench ageing. GuillaumeBrecq et al. (Gaz de France), in a study with theUniversité Pierre et Marie Curie: ‘ComparativeStudy of Natural Gas Vehicles CommercialCatalysts in Monolithic Form’ (2007-01-0039),employed commercial TWC systems for emissionscontrol in natural gas vehicles (NGVs); all theTWCs tested were efficient for CO and NOx treat-ment. However, methane conversion proved to bea problem, with high light-off temperaturesobserved, typically in the range 315 to 400ºC. Alsofrom Johnson Matthey, Andrew York and cowork-ers presented: ‘Modeling of the CatalyzedContinuously Regenerating Diesel Particulate Filter(CCR-DPF) System: Model Development andPassive Regeneration Studies’ (2007-01-0043). Themodel showed that the CCR-DPF (Catalysed CRT®

or CCRT®) can operate successfully even over low-temperature drive cycles or with challengingNOx/PM ratios; the CCR-DPF is therefore applic-able to a wider range of more challenging dieselapplications than the CRT®. The model can be usedto design the system and choose which is most suit-able for a particular application.

Bernard Bouteiller and coworkers (Saint-Gobainand Université d’Orléans, France), in: ‘OneDimensional Backpressure Model forAsymmetrical Cells DPF’ (2007-01-0045), demon-strated that they had adapted a one-dimensionalDPF model to operate with new asymmetric cellgeometry DPF systems. Furthermore, AchimHeibel and Rajesh Bhargava (Corning Inc, U.S.A.),in: ‘Advanced Diesel Particulate Filter Design forLifetime Pressure Drop Solution in Light DutyApplications’ (2007-01-0042), showed that theseasymmetrical cell technology (ACT) DPFs show a65% higher ash capacity for the same pressure dropin engine dynamometer tests, when compared withstandard cell design substrates. Thus, using ACT itis possible to design filter systems with a lower life-time pressure drop or longer service intervals for aparticulate filter system of the same size. Frank-Walter Schütze and coworkers (Umicore AG andCo KG, Germany), in: ‘Challenges for the Future


Diesel Engines Exhaust Gas AftertreatmentSystem’ (2007-01-0040), briefly reviewed the impor-tance of thermally stable catalysts for future dieselengine aftertreatment systems. The benefits of earlyintegration of catalyst development and selection inthe engine development process for optimum emis-sions reduction performance was highlighted.Finally in this section, Kevin Lang and Wai Cheng(Massachusetts Institute of Technology (MIT),U.S.A.) addressed the problem of fast catalyst light-off in a spark ignition engine: ‘A Novel Strategy forFast Catalyst Light-Off Without the Use of an AirPump’ (2007-01-0044). This is achievable using anexhaust air pump; however, the novel approachreported by Lang was to operate the engine on start-up using only three of the four cylinders, runningunder rich conditions, while the unused fourthcylinder is used as a ‘pump’ to supply air to theexhaust manifold. This air can then oxidise theproducts of incomplete combustion emitted fromthe engine, and hence supply a large amount of heatto the catalyst to accelerate its warm-up.

BiodieselOn the final day, the keynote address was deliv-

ered by Manohar K. Chaudhari (AutomotiveResearch Association of India), to highlight the cur-rent status of biodiesel in India, as well as alternativefuels such as ethanol and hydrogen-compressed nat-ural gas (H2-CNG). The Indian automotive industryis expected to grow enormously over the next 10years. There is concern over fuel supply/demand inIndia. By 2016 India is expected to account foraround 5% of the world’s oil consumption, with70% of the oil consumed being imported. The useof biodiesel as an alternative fuel is a reality in India;its potential has been widely surveyed and a numberof feedstocks tested. India’s tropical climate, whichfavours the growth of crops, combined with hugeareas of waste land and cheap labour make India aprime candidate for biodiesel use. Jatropha is themost environmentally and economically feasiblebiofuel crop in India, and can theoretically provide10% of its diesel requirements; this is ideal for usein a diesel blend. This will have a significant impacton the rural economy, and concerns over the envi-ronmental impact of growing large amounts of

crops are currently being addressed. On the otherhand, biodiesel provides some emissions benefits,compared to standard diesel, with lower CO, THCand PM emissions. In addition to biodiesel blends,compressed natural gas (CNG) and liquefied petro-leum gas (LPG) are becoming more important inIndia. For example, in 1998 Delhi was listed as oneof the 10 most polluted cities in the world; 70% ofthe pollution was from vehicles and the PM levelwas 10 times the legal limit. The use of CNG inDelhi’s vehicles has resulted in much better air qual-ity, and a CNG infrastructure is growing quicklythere and in Mumbai, with plans to expand to a fur-ther 28 cities. This is leading to growing consumerconfidence in CNG, even though it is necessary totune engines for CNG use, otherwise increasedNOx emissions result. LPG use is also growing, butat a much slower rate than CNG. Other fuels men-tioned by Chaudhari as potential alternatives inIndia were 5% ethanol/gasoline blends and 5 to20% H2-CNG, though little detail on their applica-tion was given. Looking into the future, the use ofhydrogen and fuel cells was proposed, with one mil-lion vehicles to be in operation by 2020.

Rapeseed methylester (RME) biodiesel was test-ed on a heavy-duty diesel engine by Hu Li andcoworkers (University of Leeds, U.K. andUniversidad Nacional de Asuncion, Paraguay):‘Study of Emission and Combustion Characteristicsof RME B100 Biodiesel from a Heavy Duty DIDiesel Engine’ (2007-01-0074). The emissions werecompared with those from conventional diesel, anda significant reduction in PM, volatile organic frac-tion (VOF), CO and THC was observed. This wasconfirmed by Sathaporn Chuepeng and coworkers(University of Birmingham, U.K.), in collaborationwith Jaguar Cars Ltd, U.K.: ‘A Study of QuantitativeImpact on Emissions of High Proportion RME-Based Biodiesel Blends’ (2007-01-0072). Theyvaried the proportion of biodiesel contained indiesel blends, and found that the amount ofbiodiesel affected the emissions: increased NOx wasobserved, but the use of high RME blends resultedin significant reduction in PM. Finally, DelanieLamprecht (Sasol Ltd), in: ‘Elastomer Compatibilityof Blends of Biodiesel and Fischer-Tropsch Diesel’(2007-01-0029), investigated the effect of the use of


biodiesel on elastomers typically used in sealingapplications in engines. It is known that neatbiodiesel, and high percentage blends, can degradecertain types of elastomer over time. Using standardnitrile-butadiene rubber (NBR), it was found thatelastomer compatibility should not be a problemwith the 20% biodiesel/GTL synthetic diesel blendemployed in the study.

Other Oxygenated FuelsA number of presentations addressed other oxy-

genated fuels. For example, Mitsuharu Oguma andShinichi Goto (AIST, Japan) presented the success-ful operation of a medium duty truck on publicroads using dimethyether (DME) fuel: ‘Evaluationof Medium Duty DME Truck Performances –Field Test Results and PM Characteristics’ (2007-01-0032). Jamie Turner et al. (Lotus Engineering,U.K.), in: ‘Alcohol-Based Fuels in HighPerformance Engines’ (2007-01-0056), discussedthe operation of a high-speed sports car engine run-ning on an ethanol-based fuel (ethanol containing15% gasoline by volume: E85). Some engine modi-fications were required for optimised running, forexample, to injectors. The results of the study wereapplied to a Lotus Exige 265E sports car: betterCO2 emissions, compared with gasoline, wereobserved, as well as excellent performance. Also,Miloslaw Kozak and coworkers (Poznan Universityof Technology and BOSMAL Automotive R&DCentre, Poland) presented an investigation of theeffect of dosing Euro V diesel fuels with 5% of anoxygenated fuel additive on the exhaust emissionsfrom a Euro IV passenger car: ‘The Influence ofSynthetic Oxygenates on Euro IV Diesel PassengerCar Exhaust Emissions’ (2007-01-0069). The gen-eral conclusion was that the addition of someoxygenates, such as triethylene glycol dimethylether,can significantly reduce the emissions of CO, THCand PM, but with slightly higher NOx emissions.

Concluding RemarksSummarising, the general theme running

through the entire conference was the well-knownproblem of ensuring a secure and adequate fuel sup-ply, which is only likely to worsen in the future.Therefore, a wider range of fuels will become avail-

able, necessitating improved engine design andaftertreatment devices that operate with a widerange of fuels. In addition, reducing emissions willcontinue to be a challenge.

The obvious effect of this will be the future needfor more and improved pgm and other catalysts fora whole host of applications. For example, enor-mous amounts of catalysts for use in gas-to-liquids,coal-to-liquids and biomass-to-liquids plants will berequired to bring the extra synthetic fuels capacityto the levels desired, as discussed in the conference.Furthermore, biodiesel manufacture, for exampleby transesterification, will become more important.The knock-on effect of this wide range of new fuelson emissions control will need to be studied andunderstood: for example, what will be the effect ofusing biodiesel on the efficacy of a particulate filterin controlling emissions of particulate matter?Finally, in the long term, the use of hydrogen as analternative fuel will place demands on materialsmanufacturers. This will be due not only to therequirements of hydrogen production, using alter-natives to fossil fuels as feedstocks, together withnew catalyst technologies, but also because of theneed for efficient and high-capacity hydrogenstorage media.

References 1 2007 Fuels and Emissions Conference, South Africa,

Technical Papers, SAE, Warrendale, PA, U.S.A.;http://www.sae.org/servlets/PaperEvents?OBJECT_TYPE=PaperEventsInfo&PAGE=getPaperTopics&GEN_NUM=144556

2 For information on the proposed Euro V limits see:‘Clean cars: Commission proposes to reduce emis-sions’, Europa press releases, Brussels, 21/12/2005,Reference IP/05/1660;http://europa.eu/rapid/pressReleasesAction.do?ref-erence=IP/05/1660&format=HTML&aged=0&language=EN&guiLanguage=en


The Reviewer

Andy York is a Principal Scientist in the GasPhase Catalysis Department at the JohnsonMatthey Technology Centre, U.K. He isinterested in reaction kinetics and computersimulation of vehicle emissionsaftertreatment systems, with a particularemphasis on diesel systems.

150

On 3rd May 2007, two Royal Society ofChemistry National Chemical Landmark plaqueswere unveiled at Imperial College London (IC),by Sir Richard Sykes FRS, the Rector of the col-lege, and by Professor Jim Feast FRS, Presidentof the Royal Society of Chemistry. They com-memorate two past Nobel Laureates: SirGeoffrey Wilkinson FRS, 1921–1996, (seeFigure 1), and Sir Derek Barton FRS,1918–1998. Both had been students at IC; bothleft and returned some fifteen years later to takeup professorial chairs there. The unveiling eventalso formed part of the Centenary celebrations(1) for the college, which was founded in 1907by merging the Royal College of Science, theRoyal School of Mines and the City and GuildsCollege. Some 150 people attended; there was areception afterwards and a dinner for thirty ofthe guests.

Barton did innovative and distinguished work

on the conformational analysis of complexorganic molecules. Here we concentrate on SirGeoffrey Wilkinson (Figure 2), who devotedmuch of his research to platinum group metals(pgms) chemistry (indeed some 200 of his 557publications concern the pgms). Virtually all hispgm research used materials from the JohnsonMatthey Loan Scheme, so it was fitting that DrBarry Murrer, Director of the Johnson MattheyTechnology Centre, Sonning Common, U.K.,was present at the unveiling ceremony and thedinner. An exhibition of Wilkinson memorabiliaincluded a model of ferrocene presented to himby IC inorganic chemists on the occasion of his25th year at IC in 1981, and rhodium-plated byJohnson Matthey (Figure 3). This neatly symbol-ises both the work for which he won his Nobelprize (on ferrocene and other organometalliccompounds) and his subsequent rhodium andother pgm work.


Sir Geoffrey Wilkinson:New Commemorative PlaqueCONTRIBUTIONS TO PLATINUM GROUP METALS CHEMISTRY MARKED AT IMPERIAL COLLEGE LONDON

By W. P. GriffithDepartment of Chemistry, Imperial College, London SW7 2AZ, U.K.; E-mail: [email protected]

DOI: 10.1595/147106707X214398

Fig. 1 Plaque at Imperial College, London,commemorating Professor Sir Geoffrey Wilkinson FRS,1921–1996; photo courtesy of the Royal Society ofChemistry

Fig. 2 Sir Geoffrey Wilkinson FRS, Professor ofInorganic Chemistry, Imperial College, London,1956–1988 (Sir Edward Frankland chair from 1978);Nobel Prize for Chemistry, 1973


Career at Imperial CollegeWilkinson’s early life and scientific work are out-

lined in Reference (2). In 1955, at the age of only 34,he was appointed to the chair of InorganicChemistry at IC. Highlights of Wilkinson’s time atIC include his Fellowship of the Royal Society in1965, Nobel Prize in 1973 and knighthood in 1976.He became head of the Chemistry Department inthe same year. He formally retired in 1988, but untilthe day before he died in 1996 he continued doinginnovative research with a dedicated team.

Work on Platinum Group MetalsChemistry

His work on transition metal and organometal-lic chemistry was wide-ranging and led toapplications of coordination complexes to organiccatalysis. Here we concentrate on the significantand innovative work which he carried out with thepgms. The following account draws in part fromtwo earlier sources written by the present authorand Wilkinson’s first Ph.D. student, MalcolmGreen (2), and with other ex-students (3).

Without the support of Johnson Matthey, whoprovided him with pgms (2) under their LoanScheme, a generous contribution to science enduringto this day, Wilkinson could have done little researchin the area. He knew well four Johnson Matthey stal-

warts of that time – Leslie Hunt, Henry Connor,Frank Lever (who instigated the Loan Scheme in1955) and A. R. Powell, and co-published with thelatter two in the 1960s. In 1964 he wrote a prescientarticle for Platinum Metals Review on ‘OrganometallicCompounds of the Platinum Metals’ (4). A frequentvisitor to the Johnson Matthey Technology Centre,he would sometimes leave his and his colleagues’pgm residues there in a supermarket shopping bag.In 1988 Johnson Matthey provided him with a spa-cious ‘Johnson Matthey Research Laboratory’ at IC,and here he and a dedicated small group of talentedstudents continued innovative research until hisdeath in 1996 – he went to that laboratory, as was hisdaily practice, until the day before he died.

RhodiumWilkinson used to talk of the ‘three R’s’ – rhodi-

um, ruthenium and rhenium. He isolated ashort-lived 106Rh isotope, a fission product of 235U,while working at the University of California,Berkeley, and in 1953 made salts of the [Cp2Rh]+ and[Cp2Ir]+ cations. In 1961 he did some laboratorywork himself (5) (something which always terrifiedhis students), reacting cis- and trans-[RhCl2(en)2]+

with Na(BH4) in a test-tube over a Bunsen burner,giving the new [RhHCl(en)2]+ cation as a foamingbrown solution – this he brandished around, callingout “Who wants a Ph.D.?”. With that JohnsonMatthey wizard of practical chemistry, A. R. Powell,he isolated salts of the new hydrides [RhH(NH3)5]2+

and [RhH(H2O)(NH3)4]2+ in 1966 (6).In 1965 he found that fac-RhCl3(PPh3)3 converted

hex-1-ene to n-heptaldehyde with H2 and CO underpressure at 55ºC (7). The catalyst was difficult tomake, and it was during these preparations thatRhCl(PPh3)3 was isolated; this is now universallyknown as Wilkinson’s catalyst, 1, (8). This versatile

material effects the hydrogenation of alkenes, hydro-gen transfer reactions, hydrosilation, hydroacylation,decarbonylation, hydroformylation, hydroboration,oxidation, and bond cleavage of organic materials.

Fig. 3 The model of the ferrocene structure presented toGeoffrey Wilkinson on the occasion of his 25th year atImperial College (1981), and rhodium-plated by JohnsonMatthey

1


His full paper of 1966 describing its discovery andthe realisation that it would, with molecular hydro-gen, catalyse the hydrogenation of alkenes, is aclassic, despite its uncharacteristically overlong andcompletely unpunctuated title (9). This was a turn-ing point in his career, perhaps more significantthan his elucidation of the ferrocene structure.Although not mentioned in his Nobel citation, thiswork probably contributed to the case for the prize.

At around the same time Wilkinson realised thatthe hydroformylation reactions seemingly catalysedby RhCl(PPh3)3 were actually due to the knownRhH(CO)(PPh3)3, 2, (10, 11). Among many otherreactions, this species catalyses the hydroformyla-

tion of propylene to n-butyraldehyde (a PVCprecursor). This was one of his most importantcontributions to industry.

IridiumIn the mid-1960s to 1970s, research on iridium

was quite rare, and Wilkinson came to it late. Afterhis 1953 work on [Cp2Ir]+, it was not until 1991 thathe made Ir(mesityl)3 and Ir(mesityl)4 (12) – veryunusual chemistry – and in 1992 a squareplanar, paramagnetic iridium(II) complex,Ir(mesityl)2(SEt2)2 (13).

RutheniumThis, Wilkinson said, was “an element for the

connoisseur”. He made ruthenocene, Cp2Ru andsalts of [Cp2Ru]+ in 1952, and with A. R. Powell ofJohnson Matthey in 1964–5 worked on mixed-valence ruthenium carboxylates (14, 15). He workedtoo on RuX2(LPh3)3 (X = Cl, Br; L = P, As, Sb),useful catalysts for a variety of reactions and alsoprecursors for many other ruthenium complexes(16). In later work he isolated tetrahedral homolep-tic RuR4 complexes, for example, thetetrakis(mesityl) Ru(C9H11)4 (17) and even salts of[Ru(CH3)6]3+ (18).

OsmiumWilkinson was chagrined by the fact that it was

his rival (though co-Nobel laureate) Ernst Fischerwho first made osmocene, Cp2Os, in 1958. As withiridium, he came to osmium chemistry late in hiscareer, but made up for this with his preparation ofthe astonishing tetrakis(tbutylimido) osmiumspecies, 3, a catalyst for amination reactions (19, 20).

Palladium and PlatinumThese received less attention from Wilkinson,

although he was particularly proud of his transmu-tation of platinum to gold, a feat which caught thepublic imagination with the headline: ‘ScientistDiscovers Goldmine in the Cyclotron’, in the SanFrancisco Chronicle, 1948. This may have beenbecause these metals are less versatile in their oxida-tion states than the other four, but he did someseminal work with reactions of M(PPh3)3 (M = Pd,Pt) (21) and with the nitrosyl (22) and carboxylatecomplexes (11, 23) of the two metals.

Concluding RemarksWilkinson made many contributions to chem-

istry, many of them concerned with the pgms, andwas a prime contributor to what was (in those headydays) called the renaissance of inorganic chemistry.He was fortunate to have been associated with theatomic bomb project, assimilating profound chem-ical knowledge from that experience, and to havelived at a time when one could still do research forits own sake. He used his knowledge well, with aremarkable instinct for the new experiment, thenew area of research, producing really original andcreative new science.

References1 Imperial College: Centenary Website:

http://www.imperial.ac.uk/centenary/2 M. L. H. Green and W. P. Griffith, Platinum Metals

Rev., 1998, 42, (4), 168

2

3


3 M. A. Bennett, A. A. Danopoulos, W. P. Griffithand M. L. H. Green, J. Chem. Soc., Dalton Trans.,1997, 3049

4 G. Wilkinson, Platinum Metals Rev., 1964, 8, (1), 165 G. Wilkinson, Proc. Chem. Soc., 1961, 726 J. A. Osborn, A. R. Powell and G. Wilkinson, Chem.

Commun. (London), 1966, 4617 J. A. Osborn, G. Wilkinson and J. F. Young, Chem.

Commun. (London), 1965, 178 J. F. Young, J. A. Osborn, F. H. Jardine and G.

Wilkinson, Chem. Commun. (London), 1965, 1319 J. A. Osborn, F. H. Jardine, J. F. Young and G.

Wilkinson, J. Chem. Soc. A, 1966, 171110 C. O’Connor, G. Yagupsky, D. Evans and G.

Wilkinson, Chem. Commun. (London), 1968, 42011 D. Evans, J. A. Osborn and G. Wilkinson, J. Chem.

Soc. A, 1968, 313312 R. S. Hay-Motherwell, G. Wilkinson, B. Hussain-

Bates and M. B. Hursthouse, J. Chem. Soc., DaltonTrans., 1992, 3477

13 A. A. Danopoulos, G. Wilkinson, B. Hussain-Batesand M. B. Hursthouse, J. Chem. Soc., Dalton Trans.,1992, 3165

14 S. M. Morehouse, A. R. Powell, J. P. Heffer, T. A.Stephenson and G. Wilkinson, Chem. Ind. (London),1964, 544

15 T. A. Stephenson and G. Wilkinson, J. Inorg. Nucl.Chem., 1966, 28, (10), 2285

16 T. A. Stephenson and G. Wilkinson, J. Inorg. Nucl.Chem., 1966, 28, (4), 945

17 P. D. Savage, G. Wilkinson, M. Motevalli and M. B.Hursthouse, J. Chem. Soc., Dalton Trans., 1988, 669

18 R. S. Hay-Motherwell, G. Wilkinson, B. Hussain-Bates and M. B. Hursthouse, Polyhedron, 1990, 9,(17), 2071

19 D. W. H. Rankin, H. E. Robertson, A. A.Danopoulos, P. D. Lyne, D. M. P. Mingos and G.Wilkinson, J. Chem. Soc., Dalton Trans., 1994, 1563

20 A. A. Danopoulos, G. Wilkinson, B. Hussain-Batesand M. B. Hursthouse, J. Chem. Soc., Dalton Trans.,1991, 1855

21 C. J. Nyman, C. E. Wymore and G. Wilkinson, J.Chem. Soc. A, 1968, 561

22 W. P. Griffith, J. Lewis and G. Wilkinson, J. Chem.Soc., 1961, 775

23 T. A. Stephenson, S. M. Morehouse, A. R. Powell, J.P. Heffer and G. Wilkinson, J. Chem. Soc., 1965, 3632

The AuthorBill Griffith is Professor of Inorganic Chemistry atImperial College, London. He has considerableexperience with the platinum group metals,particularly ruthenium and osmium. He haspublished over 270 research papers, manydescribing complexes of these metals as catalystsfor specific organic oxidations. He has writtenseven books on the platinum metals, and is theSecretary of the Historical Group of the RoyalSociety of Chemistry.

Platinum Metals Rev., 2007, 51, (3), 154 154

During 2005, the global car market generated rev-enues to the tune of U.S.$839.7 billion, and this isexpected to increase by 24.5 per cent by 2010. TheAsia-Pacific region is a promising market, with a shareof more than 26 per cent of global car sales in 2005;this is forecast to increase to 33 per cent by 2010. Interms of car production volumes, Asia-Pacificaccounts for 41 per cent of the global market, andSouth Korea produces 15 per cent of the region’s out-put. With the industry forecasting that globalproduction by South Korean manufacturers will growby 11 per cent from 2005 to 2010, South Korea’sstrong automotive position within Asia-Pacific is setto continue for many years to come. Moreover, SouthKorea experienced economic growth of 4.8 per centduring 2006 and is expected to see 4.2 per cent growthin 2007.

Korea introduced stringent emissions legislation inthe 1980s, ahead of the European Union. Initiallyadopting procedures similar to those in the UnitedStates, the legislation has since been adapted with theintention of harmonising with European as well asU.S. standards (1).

Against this optimistic background, in January2007 the latest of eleven Johnson Matthey autocatalystmanufacturing plants opened at Jangan-Myeon inSouth Korea. The plant was designed to satisfyincreasing market demands, while exploiting the latesttechnologies with a view to optimising performanceand economic efficiency. New vehicle technologieshave emerged in order to meet increasing environ-mental concerns worldwide. The Jangan-Myeon plantensures that Johnson Matthey’s autocatalysts not onlymeet but exceed customer specifications.

The first stage of the plant consists of the manu-facturing building, with an initial capacity of 2 millionautocatalysts per year, and designed with sufficientforesight and flexibility for future expansion. Robotsare a key contributor to the manufacturing process,accelerating it, making it more efficient and reliable,decreasing human errors, and improving productquality through enhanced accuracy. A computerisedcontrol system fully links and integrates all stages

of process operation. Machinery has been procuredworldwide as part of manufacturing process optimisation.

The second stage, to begin operating by the end of2007, is the Applications and Test Centre. Again usingthe latest technologies and state-of-the-art equipment,this facility will offer customers a reliable and efficientresponse to their specific catalyst development needs.A full range of research and development activitieswill include testing, analysis, benchmarking and simu-lation. One of the principal aims for the Applicationsand Test Centre is to ensure that washcoats (contain-ing platinum group metals) continue to meetcustomers’ ever more stringent specifications. Enginetest cells will enable compliance with the quality pro-gramme requirements of the Korean motor industry.

The Jangan-Myeon autocatalyst plant will enableJohnson Matthey to build on our already successfulrelationships with South Korean automotive compa-nies, providing them with industry-leading emissionscontrol technologies. The new plant further strength-ens Johnson Matthey’s position in Asia, underliningour commitment to this fast-growing region.

CARLOS SILVA

Reference1 Johnson Matthey Emission Control Technologies,

‘Korea’s Resurgent Market’, Global EmissionsManagement, Spring 2004, 2, (5), pp. 4–5:http://ect.jmcatalysts.com/pdf/2,5%20Koreas%20resurgent%20market.pdf

Carlos Silva is General Manager of the Johnson Matthey EmissionControl Technologies plant at Jangan-Myeon, South Korea.

New Autocatalyst Plant for South KoreaJOHNSON MATTHEY MAKES USE OF THE LATEST TECHNOLOGIES

DOI: 10.1595/147106707X216819

Serving a growing market, the new Johnson Mattheyplant at Jangan-Myeon, South Korea, has an initialcapacity to produce 2 million autocatalysts per year

Johnson Matthey has published the“Platinum 2007” edition of its annual survey ofthe supply and demand of the platinum groupmetals (pgms), addressing the calendar year2006.

Global demand for platinum rose by 80,000oz to 6.775 million oz in 2006. The autocatalystmarket grew by over 10 per cent, attributablemainly to emission controls fitted to diesel vehi-cles. Supplies of platinum also climbed in 2006,to a record 6.785 million oz. Supply and demandfor platinum effectively balanced, with a nomi-nal surplus of only 10,000 oz in 2006.

With the market share of light-duty dieselvehicles in Europe growing to over 50 per cent,and new emissions standards applied in early2006, platinum consumption for autocatalysts inEurope rose by 200,000 oz to 2.160 million oz.The new requirement for catalysts on medium-sized diesel vehicles in the U.S.A. and rapidgrowth in production of passenger vehicles inAsia also contributed to global auto demand of4.195 million oz, reports Johnson Matthey; thisis an increase of 400,000 oz over 2005.

The rising price of platinum led to a smalldecline in the volume of platinum jewellery pro-duced in 2006. However, the price had littleimpact on consumer purchasing in China.Manufacturers’ new metal purchases were fur-ther reduced by increased jewellery recycling inChina and Japan. Total world jewellery demandfor platinum fell by 360,000 oz in 2006 to 1.605million oz.

Platinum consumption for autocatalysts isexpected to increase again in 2007, with most ofthe rise attributable to diesel vehicles. Jewellerymanufacturers in Asia will continue to sourceplatinum partly from old stock, but JohnsonMatthey regards consumer demand for platinumjewellery as encouragingly resilient. Weaker thanplanned supply from South Africa earlier in2007 and a recent hiatus in Russian exports due

to regulatory problems are expected to havemeant a tight market in the first half of 2007,with liquidity set to increase in the second half asSouth African mine output rises.

After five years of growth, annual demandfor palladium fell by 720,000 oz in 2006 to 6.635million oz. Although demand for autocatalystswas buoyant, demand for new metal from jew-ellery manufacturers fell sharply. There was alsoless interest in palladium physical investmentproducts. Production of palladium from SouthAfrica increased, but sales from state stocks byRussia were significantly lower than in 2005.Supplies accordingly fell to 8.060 million oz, adecline of 345,000 oz. Overall, the palladiummarket showed another large surplus of 1.425million oz.

Johnson Matthey note autocatalyst demandup by 150,000 oz to 4.015 million oz for 2006,due to continuing substitution of platinum-based catalysts by palladium on gasoline vehicles.However, demand for palladium in the jewellerysector fell by 435,000 oz to 995,000 oz, with vir-tually all of the decline occurring in China.

The autocatalyst market is again predicted totake more palladium in 2007. The prospects forthe palladium jewellery market are less certain,according to Johnson Matthey. Primary produc-tion of palladium is expected to rise, and supplyis to be augmented by sales from the largeamount of Russian state stocks shipped toSwitzerland at the end of 2006.

Two Special Features are included in“Platinum 2007”: ‘Heavy Duty Diesel: AGrowing Source of PGM Demand’, and‘Memories Are Made of This’, a survey of pgmusage for personal computer hard disk drives.

“Platinum 2007”, Johnson Matthey PLC,Precious Metals Marketing, Orchard Road,Royston, Hertfordshire SG8 5HE, U.K.;E-mail: [email protected]; website:http://www.platinum.matthey.com/publications.


DOI: 10.1595/147106707X214460

Platinum 2007

PROPERTIESCyclic Oxidation of Ru-Containing Single CrystalSuperalloys at 1100ºCQ. FENG, B. TRYON, L. J. CARROLL and T. M. POLLOCK, Mater.Sci. Eng.: A, 2007, 458, (1–2), 184–194

The cyclic oxidation behaviour at 1100ºC of Ru(3.5–9 at.%)-containing Ni-base single crystal superal-loys (1) with Cr additions (0–8 at.%) has beeninvestigated. High levels of Cr additions (8 at.%) sig-nificantly improved oxidation resistance. Amultilayered scale formed on (1); this generally con-sisted of an external layer of NiO, an intermediatelayer of spinel and an α-Al2O3 inner layer (2). (2)improved oxidation performance. Ru-rich precipi-tates were observed in the spinel layer of (1) whichdisplayed poor oxidation resistance.

Electronic Properties of the Semiconductor RuIn3

D. BOGDANOV, K. WINZER, I. A. NEKRASOV and T. PRUSCHKE,J. Phys.: Condens. Matter, 2007, 19, (23), 232202

Single crystals of RuIn3 (1) were grown using theflux method with In as reactant and flux medium.Temperature-dependent measurements of the resis-tivity of (1) show a semiconducting behaviour, incontrast to previously published results. In the high-temperature range the semiconducting gap is 0.4–0.5eV. An anisotropy of the resistivity along [110] and[001] orientations of the tetragonal (1) was observed.

CHEMICAL COMPOUNDSOrganometallic Molecular Materials: Self-Assemblythrough Hydrogen Bonding of an OrganoplatinumNetwork Structure with Zeolite-Like TopologyF. ZHANG, M. C. JENNINGS and R. J. PUDDEPHATT, Chem.Commun., 2007, (15), 1496–1498

[Pt(OH)2Me2(dpa)] (1) (dpa = di-2-pyridylamine)was formed by oxidation of [PtMe2(dpa)] by H2.[Pt2(μ-OH)2Me4(dpa)2][B(OH)(C6F5)3]2 (2) wasobtained by abstraction of a hydroxo ligand from (1)by reaction with B(C6F5)3. (1) and (2) contain both H-bonding PtOH groups and NH groups. Self-assemblyof (1) to form a complex network structure gave thefirst “organometallic zeolite”.

Solvent-Free Synthesis of Metal ComplexesA. LAZUEN GARAY, A. PICHON and S. L. JAMES, Chem. Soc. Rev.,2007, 36, (6), 846–855

Polymeric PtCl2 will react with solid phosphinesafter grinding in a ball mill to give cis-[PtCl2(PPh3)2],which can subsequently react with solid K2CO3 togive [Pt(CO3)(PPh3)2]. Also, supramolecular hoststructures, such as a tetraplatinum square and a nano-scale bowl-shaped hexapalladium cage, self-assemblewith remarkable efficiency in the solid state. (39 Refs.)

ELECTROCHEMISTRYElectrochemical Performance of Nano-Pt-SupportedCarbon Anode for Lithium Ion BatteriesW.-S. KIM, H. S. KIM, I.-S. PARK, Y. KIM, K.-I. CHUNG, J. K. LEEand Y.-E. SUNG, Electrochim. Acta, 2007, 52, (13), 4566–4571

The title anode (1) was prepared by supporting Ptnanoparticles onto C powder. The Pt nanoparticleson the C surface helped to suppress the solventdecomposition reaction: a protective film was formedas soon as (1) had contact with the electrolyte. Inaddition, the Pt nanoparticles act as catalyst reactionsites to improve the Li discharge reaction.

Compositional and ElectrochemicalCharacterization of Noble Metal–DiamondlikeCarbon Nanocomposite Thin FilmsN. MENEGAZZO, C. JIN, R. J. NARAYAN and B. MIZAIKOFF,Langmuir, 2007, 23, (12), 6812–6818

Pt– and Au–diamondlike C (DLC) nanocompositefilms (1) were deposited onto Si substrates by modi-fied pulsed laser deposition. Cross-sectional TEMrevealed that metal was present as arrays of noblemetal islands embedded within the DLC matrix. (1)exhibited greater conductivity than their metal-freecounterparts. The electrochemical properties of (1)were studied using quasi-reversible redox couples.

PHOTOCONVERSIONPhotocatalytic Oxidation of NOx by Pt-ModifiedTiO2 Under Visible Light IrradiationY. ISHIBAI, J. SATO, S. AKITA, T. NISHIKAWA and S. MIYAGISHI,J. Photochem. Photobiol. A: Chem., 2007, 188, (1), 106–111

Photooxidation of NOx was carried out using TiO2

(Ishihara ST-01) treated with H2PtCl6. Pt-modifiedTiO2 was obtained with different ligand structuresaccording to the Pt treatment method. TiO2 photo-catalysts with certain Pt complexes producedsignificant photocatalytic activity under visible lightirradiation without decreasing photoactivity underUV light irradiation. The visible-light-induced photo-catalytic activity depended on the amount of Pt.

An Investigation on Palladium Sulphide (PdS)Thin Films as a Photovoltaic MaterialI. J. FERRER, P. DÍAZ-CHAO, A. PASCUAL and C. SÁNCHEZ, ThinSolid Films, 2007, 515, (15), 5783–5786

Polycrystalline PdS thin films (1) with tetragonalstructure were grown by direct sulfuration of Pd lay-ers. (1) exhibited a Seebeck coefficient, S = –250 ± 30μV K–1. Electrical resistivity of (1), measured by afour contact probe, was (6.0 ± 0.6) × 10–2 Ω cm. Halleffect measurements confirmed n-type conductivity.Optical absorption coefficient in the range of photonenergies hυ > 2.0 eV was higher than 105 cm–1.


ABSTRACTSof current literature on the platinum metals and their alloys

DOI: 10.1595/147106707X219113

Platinum Metals Rev., 2007, 51, (3)

Iridium(III) Complexes Bearing Quinoxaline Ligandswith Efficient Red Luminescence PropertiesK. TANI, H. FUJII, L. MAO, H. SAKURAI and T. HIRAO, Bull. Chem.Soc. Jpn., 2007, 80, (4), 783–788

Excellent quantum efficiencies (50–79%) for pho-toluminescence were attained for cyclometallatedIr(III) complexes (1) bearing 2,3-diphenylquinoxa-lines in CH2Cl2. Luminescence peak wavelengths of(1) were within the preferred range of 653–671 nm inthin films. The most vivid red electrophosphores-cence was achieved with an acetylacetonato Ircomplex bearing 2,3-diphenylquinoxaline.

ELECTRODEPOSITION AND SURFACECOATINGSIn Situ Raman Spectroscopy of AnnealedDiamondlike Carbon–Metal Composite FilmsC. JIN, H. ZHOU, S. GRAHAM and R. J. NARAYAN, Appl. Surf.Sci., 2007, 253, (15), 6487–6492

Films of diamondlike C, diamondlike C–Pt com-posite and diamondlike C–Au composite wereannealed to 523ºC. The Raman spectra for these filmswere fitted using a two-Gaussian function. The varia-tions of the G-peak position, the D-peak position,and the ID/IG ratio were examined as a function oftemperature. The diamondlike C film exhibitedgreater thermal stability than the diamondlikeC–noble metal composite films.

Synthesis of PVP Stabilized Cu/Pd Nanoparticleswith Citrate Complexing Agent and Its Applicationas an Activator for Electroless Copper DepositionS. H. Y. LO, Y.-Y. WANG and C.-C. WAN, J. Colloid Interface Sci.,2007, 310, (1), 190–195

Cu/Pd nanoparticles (1) were synthesised in aque-ous solution using trisodium citrate as additive. Theprotecting agent PVP was added after the citrate. (1)exist in a stable suspension. XPS of (1) revealed smallamounts of oxidised Pd on the surface and the exis-tence of zerovalent Cu and oxidised Cu. (1) exhibitedcatalytic activity comparable to that of conventionalPd/Sn activator. (1) have potential as an activator forelectroless Cu deposition in the PCB industry.

Electrodeposition of Magnetic CoPd Thin Films:Influence of Plating ConditionF. M. TAKATA and P. T. A. SUMODJO, Electrochim. Acta, 2007,52, (20), 6089–6096

The title thin films (1) were electrodeposited from achloride plating bath containing glycine as additive.The Co content in (1) could be varied from 6.4–94.0at.% by controlling the pH and [Co2+]/[Pd2+] ratio inthe bath. Current densities > 50 mA cm–2 gavedeposits with a typical ‘cauliflower’ morphology. Forcurrent densities < 25 mA cm–2 cracks were observed.XRD showed that (1) were amorphous. The magnet-ic properties for (1) revealed that the coercivity (Hc)values ranged from 84–555 Oe and the magnetic sat-uration (Ms) from 0–1.73 T.

APPARATUS AND TECHNIQUEPd Encapsulated and Nanopore Hollow FiberMembranes: Synthesis and Permeation StudiesB. K. R. NAIR and M. P. HAROLD, J. Membrane Sci., 2007, 290,(1–2), 182–195

“Pd encapsulated” (1) and “Pd nanopore” (2) mem-branes on α-Al2O3 hollow fibres were synthesised bysol slip casting, film coating, and electroless platingsteps. The unaged (1) exhibited good performancewith ideal H2/N2 separation factors of 3000–8000 andH2 flux ~ 0.4 mol m–2 s–1 at 370ºC. The unaged (2)had a lower initial flux and permselectivity, but exhib-ited superior performance after 200 h.

HETEROGENEOUS CATALYSISHCN Synthesis from Methane and Ammonia overPlatinumS. DELAGRANGE and Y. SCHUURMAN, Catal. Today, 2007, 121,(3–4), 204–209

TAP (temporal analysis of products) experimentswere conducted for the synthesis of HCN from NH3

and CH4 over Pt black. At 1173 K the HCN produc-tion rate depends on the order of introducing thereactants. HCN is formed rapidly on the CH4 pulsejust after introducing NH3. A slow formation of HCNis observed on the NH3 pulse that follows a CH4

pulse. The rate-determining step for the formation ofHCN is the NH3 decomposition rate.

Synthesis and Characterization of Pt/Mg(Al)OCatalysts Obtained From Layered DoubleHydroxides by Different RoutesO. LORRET, S. MORANDI, F. PRINETTO, G. GHIOTTI, D. TICHIT,R. DURAND and B. COQ, Microporous Mesoporous Mater., 2007,103, (1–3), 48–56

Pt-containing Mg/Al layered double hydroxides atdifferent Pt loadings (0–3.2 wt.%) were prepared bycoprecipitation, impregnation and sol-gel methods.After activation and reduction treatments, Pt nano-clusters interacting with Mg(Al)O supports wereobtained. The behaviour of Pt/Mg(Al)O systems asmultifunctional catalysts was investigated in the cas-cade reaction between benzaldehyde and propanal.

Pd-Based Sulfated Zirconia Prepared by a SingleStep Sol–Gel Procedure for Lean NOx ReductionE. M. HOLMGREEN, M. M. YUNG and U. S. OZKAN, J. Mol. Catal.A: Chem., 2007, 270, (1–2), 101–111

The title catalysts (1) were obtained via a single-stepsol-gel procedure. Tetragonal zirconia was formed atlower temperatures; larger zirconia crystallites wereobtained when Pd was added to the gel. Raman spec-tra of (1) calcined at 700ºC showed both thetetragonal and the monoclinic phases, indicating asurface phase transition. (1) were active for the reduc-tion of NO2 with CH4 under lean conditions. (1)calcined at 700ºC were much more active than (1) cal-cined at 600ºC, despite the observed transition to themonoclinic phase.

157

Platinum Metals Rev., 2007, 51, (3)

XPS and 1H NMR Study of Thermally StabilizedRh/CeO2 Catalysts Submitted to Reduction/OxidationTreatmentsC. FORCE, E. ROMÁN, J. M. GUIL and J. SANZ, Langmuir, 2007,23, (8), 4569–4574

Rh/CeO2 (1) was prepared by incipient wetnessimpregnation of CeO2 with a solution of Rh(NO3)3.(1) was submitted to different H2 reduction, Ar+ sput-tering, and oxidation treatments. Below 473 K,reduction increased the amount of OH and Ce3+

species; above this temperature, reduction producedO vacancies at the surface of the support. From 1HNMR, the metal dispersion was deduced and themean size of the metal particles (38 Å) estimated.

HOMOGENEOUS CATALYSISScreening of a Modular Sugar-Based Phosphite-Oxazoline Ligand Library in AsymmetricPd-Catalyzed Heck ReactionsY. MATA, O. PÀMIES and M. DIÉGUEZ, Chem. Eur. J., 2007, 13,(12), 3296–3304

A library of phosphite-oxazoline ligands (1) derivedfrom D-glucosamine was synthesised. (1) were suc-cessfully screened in the Pd-catalysed Heck reactionof several substrates with high regio- (≤ 99%) andenantioselectivities (ee’s ≤ 99%) as well as withimproved activities. The catalytic activity was highlyaffected by the oxazoline and biarylphosphite sub-stituents and the axial chirality of the biaryl moiety of (1).

Development of a Concise Scaleable Synthesis of2-Chloro-5-(pyridin-2-yl) Pyrimidine via a NegishiCross-CouplingC. PÉREZ-BALADO, A. WILLEMSENS, D. ORMEROD, W.AELTERMAN and N. MERTENS, Org. Process Res. Dev., 2007,11, (2), 237–240

A Negishi cross-coupling between an in situ prepared2-pyridylzinc chloride and 5-iodo-2-chloropyrimidinecatalysed by Pd(PPh3)4 afforded 2-chloro-5-(pyridin-2-yl) pyrimidine (1) in one step. Chromatography can beomitted as a convenient purification was developed.The method has been used on a mini-plant scale toproduce 16 kg of (1). The Pd and Zn content of (1)was acceptable for the production of its derived API,a selective PDE-V inhibitor.

FUEL CELLSNanoscale Current Imaging of the ConductingChannels in Proton Exchange Membrane Fuel CellsD. A. BUSSIAN, J. R. O’DEA, H. METIU and S. K. BURATTO, NanoLett., 2007, 7, (2), 227–232

A Pt-coated AFM tip was used as a nanoscale cath-ode in a PEMFC. Inhomogeneous distributions ofconductive surface domains at several length scaleswere found. Phase current correlation microscopyshowed that a large number (~ 60%) of the aqueousdomains present at the surface of the operatingNafion membrane were inactive.

Surface-Modified Carbons as Platinum CatalystSupport for PEM Fuel CellsA. GUHA, W. LU, T. A. ZAWODZINSKI and D. A. SCHIRALDI,Carbon, 2007, 45, (7), 1506–1517

The ability of functionalised high surface areagraphitic (C nanofibres) and amorphous (activated C)C to homogeneously support Pt particles was investi-gated. Functionalisation by conc. acid treatmentcreated various O carrying functionalities on the Csurfaces. Chemical reduction of the Pt precursorcomplex, using milder reducing agents at 75–85ºC,and using ethylene glycol at 140ºC, gave the smallestPt particle sizes. XPS confirmed the existence of Pt in(primarily) its metallic state on the functionalised C.

Characterization and PEMFC Testing of Pt1–xMx

(M = Ru, Mo, Co, Ta, Au, Sn) Anode Electrocatalyst Composition SpreadsD. A. STEVENS, J. M. ROULEAU, R. E. MAR, A. BONAKDARPOUR,R. T. ATANASOSKI, A. K. SCHMOECKEL, M. K. DEBE and J. R.DAHN, J. Electrochem. Soc., 2007, 154, (6), B566–B576

Pt1–xMx random alloy samples were deposited viamagnetron sputtering through shadow masks onto ananostructured thin-film support for testing in a 64-electrode PEMFC. CO stripping voltammograms andH2 oxidation polarisation curves with pure H2 andreformate (≤ 50 ppm CO) were measured. Ru, Mo,and Sn were confirmed to improve the CO toleranceof Pt, although the intrinsic H2 oxidation activity of Ptdecreased significantly as the Sn content increased.

Autothermal Reforming of Gasoline on Rh-BasedMonolithic CatalystsA. QI, S. WANG, C. NI and D. WU, Int. J. Hydrogen Energy, 2007,32, (8), 981–991

0.3 wt.% Rh/3 wt.% MgO/20 wt.% CeO2-ZrO2

supported on cordierite monolith was used for ATR.At 650–800ºC, O2/C molar ratio of 0.38–0.45 andH2O/C ratio of 2.0, octane was fully converted intoreformate (+ small amount of CH4), whereas tolueneconverted into CH4-free reformate at a relativelyhigher temperature. A 1 kW gasoline fuel processorfor fuel cell application was successfully operated at a(H2 + CO) throughput of 0.9–1.0 m3 h–1 for 60 h.

MEDICAL USESCorrosion Behaviour of FePt-Based Bulk Magnetsin Artificial Saliva SolutionA. GEBERT, S. ROTH, R. GOPALAN, A. A. KÜNDIG and L.SCHULTZ, J. Alloys Compd., 2007, 436, (1–2), 309–312

The corrosion behaviour of Fe50Pt50, Fe35Pt35P30 andFe53Pt44C3 magnets was investigated in artificial salivasolution at 37ºC. Electrochemical polarisation mea-surements and SEM showed that the alloys exhibit aPt-like behaviour (highly stable). Enhanced corrosionactivity did not occur when these alloys were coupledwith commercial Fe- and CoCr-based dental alloys. Alow-corroding state was attained in combination withthe dental spring steel Fe-18Cr-18Mn-2Mo-1N.

158


METALS AND ALLOYSPalladium AlloyDERINGER NEY INC European Appl. 1,756,325

A family of Pd alloys are claimed which can be usedfor in-body medical devices or for electrical contactssuch as low-noise signal sliding or static contacts. Thealloys have high strength, radio opacity and biocom-patibility and are workable. Components may includePd and B plus at least one element selected from Ru,Ir, Pt, Re, W, Au, Zr, Co, Ni and Ta; preferably (in%): 45–99.95 Pd, 0.005–1.5 B, 0–25 Re and optional-ly 0–8 Ru. Alternative compositions include Pd andRu with at least one element selected from Ir, Pt, B,Re, W, Au, Zr, Co, Ni and Ta.

Osmium DiboridesREGENTS UNIV. CALIFORNIA European Appl. 1,761,463

Os compounds of formula OsxM1–xB2 (1) where M = Ru, Re or Fe and 0.01 ≤ x ≤ 1; except when x ≠ 1 and M = Re, then 0.01 ≤ x ≤ 0.3, are claimedwhich are ultra-hard and incompressible. (1) haveVickers hardness ≥ 2000 kg mm–2 and can be used inplace of other ultra-hard materials for abrasives, pro-tective coatings or the surfaces of cutting tools.

PHOTOCHEMISTRYElectroluminescent Iridium ComplexesOLED T LTD European Appl. 1,761,614

Electroluminescent devices incorporate an electro-luminescent layer containing complexes of Ir, Ru, Rh,Pd, Pt or Os, preferably Ir. Ligands may include 2-benzo[b]thiophenyl and benzimidazole, and givecomplexes such as bis[thiophen-2-yl-pyridine-C2,N´]-2-(2-pyridyl)-benzimidazole iridium.

Facial Tris-Cyclometallated ComplexesEASTMAN KODAK CO U.S. Appl. 2007/0,078,264

A process for forming a facial tris-cyclometallated Iror Rh complex isomer includes a step of heating asolution of a meridional Ir or Rh complex isomer inan unsubstituted or halogenated hydrocarbon solventor a mixture, to a temperature of ≥ 150ºC. Ligandsmay include at least one 1-phenylisoquinoline, 3-phenylisoquinoline or 2-phenylquinoline group,and optionally a phenylpyridine group.

ELECTRODEPOSITION AND COATINGSRuthenium Plating Seed LayerIBM CORP World Appl. 2007/044,305

A plating seed layer for an interconnect structurecontains an O/N transition region between bottomand top plating seed regions, which may contain Ru,Ir or alloys thereof, but are preferably both Ru. Aninterconnect conductive material may include Cu, Al,W or AlCu, preferably Cu.

Platinum Electrode Surface CoatingSECOND SIGHT MED. PROD. INC

World Appl. 2007/050,212A rough-configuration electrode surface coating

and method for its manufature is claimed, which canbe applied to a conductive substrate such as Ru, Rh,Ir, Pd, Pt, Ti, Zr, Nb, Ta, Cr, Mo, W, Mn, Re, Ni,Ag, Au or C. Pure Pt is coated onto the surface byelectroplating at a rate faster than would be requiredto produce shiny Pt, but slower than required toproduce Pt black, preferably between 0.05–1.0 μmmin–1. The coating gives an increase in surface areaof 5–500 times vs. the original substrate surface, andis biocompatible.

Nitrogenated Hard Bias Layers Containing PlatinumHITACHI GLOB. STORAGE TECHNOL.

U.S. Appl. 2007/0,091,515A read sensor for a magnetic head includes a sensor

stack, hard bias layers and lead layers. The hard biaslayers are formed from a nitrogenated Co-based alloysuch as nitrogenated CoPtCr or CoPt, by ion beamdeposition using a sputtering gas such as Xe with N2

as a reactive gas. Coercivity and squareness areimproved by the nitrogenation. The sensor can beused for a magnetic data storage device.

Platinum-Iron Alloy Plating SolutionCANON INC Japanese Appl. 2006-265,716

A solution for plating a magnetic PtFe alloy includesammonium hexachloroplatinate and an Fe compo-nent stabilised by a complexing agent such as tartaricor citric acid. The pH is between 6–9.5. An electrodeand an object to be plated are positioned in a vesselcontaining the solution, and voltage is applied to theelectrode to effect plating.

Osmium Coating for TEM SamplesSUMIKA CHEM. ANAL. SERV. LTD

Japanese Appl. 2006-329,743A sample of a material such as a semiconducting

material is prepared for transmission electron micro-scope analysis by coating the surface with a film ofOs. A preparatory protective coating of Pt, Pd, PtPd,AuPd or Au may first be applied by vacuum vapourdeposition. The prepared sample forms a thin filmintegrated circuit.

Push Switch with Rhodium CoatingFUJIKURA LTD Japanese Appl. 2006-351,255

A dome-type push switch includes a dome shapedmetal spring made from a thin plate of stainless steelor phosphor bronze with a surface coating of Rh. Thegood corrosion resistance and high hardness (Vickershardness Hv = 800–1000) of Rh means that improvedcorrosion resistance and wear resistance can beobtained even when using a thin layer of Rh.

NEW PATENTSDOI: 10.1595/147106707X210310


APPARATUS AND TECHNIQUEPirani Pressure Gauge with Platinum Alloy FilamentTHE BOC GROUP PLC European Appl. 1,771,710

A thermal conductivity pressure gauge includes aheated, coiled sensing filament made from an alloy ofPt and Ir, containing ≥ 70% Pt and preferably ~ 90% Pt and 10% Ir. The filament is arranged in anon-linear configuration, for example a V-shape, andcan be used in a corrosive environment to reliablymeasure pressures down to 10–4 mbar over a pro-longed period of time.

Palladium-Based Alloy for SensorSHANGHAI RUISHI INSTRUM. ELECTRON. CO LTD

Chinese Appl. 1,773,211An ultralow-temperature drift electric eddy-current

displacement and vibration sensor is claimed whichincludes a probe formed from a Pd-based inductivecoil, a coaxial connecting cable and a preamplifier.The coil is made from an alloy wire containing (inwt.%): 58.6–65.4 Pd, 13.4–20.2 Ag and 16.6–23.4 Au.A connecting cable includes a core wire with (inwt.%): 36–43 Cu, 10–17 Au and 45–52 Ag.

HETEROGENEOUS CATALYSISN2O Decomposition CatalystSTICHTING ENERGIE European Appl. 1,755,770

A catalyst can be used for decomposition of N2O ina gas which may also contain NOx, O2 and/or H2Oand has < 50 ppm hydrocarbon. A zeolite is firstloaded with 0.00001–4 wt.% of a metal selected fromRu, Rh, Os, Ir, Pt, Ag, Re and Au, preferably Ru, Rh,Os or Ir. Then a second metal (0.1–10 wt.%) isloaded, selected from Fe, Co, Ni, V, Cr, Mn or Cu,preferably Fe, Co or Ni.

Synthesis of C2 Oxygenates from Carbon MonoxideBP PLC European Appl. 1,755,780

A SiO2-supported catalyst for synthesis of C2-oxy-genates such as EtOH, CH3CHO and CH3CO2H, byhydrogenation of CO has the formula RhMnFeM1M2,where M1 = Li and/or Na, and M2 = Ru and/or Ir.Rh is 0.1–3 wt.% of total catalyst, and the weightratios of the other constituents vs. Rh are: 0.5–12 Mn,0.01–0.5 Fe, 0.01–1 M1 and 0.1–1.0 M2. The catalystis prepared from a solution containing compounds ofthe desired components, which is impregnated ontoSiO2 gel support and then dried at 283–473 K for 2 h–20 days. It may be reduced in situ in pure H2(g) ora gas containing H2, at 573–673 K for ≥ 1 h.

Platinum Group Metal Oxide SolsJOHNSON MATTHEY PLC European Appl. 1,761,335

A sol is formed from nanoparticles of Pt, Pd, Rh, Ir,Ru or Os oxides plus stabiliser ions dispersed in anaqueous liquid. Nanoparticles have average diameter< 10 nm and the molar ratio of metal to stabiliser ionsis ≥ 0.7. The sol can be contacted with a supportmaterial in the form of a powder, an aqueous slurry ora solid substrate to form a supported catalyst.

Iridium Catalyst System for Alkane MetathesisA. S. GOLDMAN et al. U.S. Appl. 2007/0,060,781

A method of making a liquid hydrocarbon fuel suchas gasoline or diesel fuel from synthesis gas producedby Fischer-Tropsch catalysis includes a step of con-verting low molecular weight alkanes, CnH2n+2

(n = 3–10), to higher molecular weight alkanes,CmH2m+2 (m = 4–40), using a dual catalyst system. Thisincludes a H transfer catalyst, preferably an Ir pincercomplex catalyst, and a metathesis catalyst. Both cat-alysts are heterogeneous and may be immobilised onthe same or separate solid supports.

Exhaust Gas Cleaning CatalystTOYOTA MOTOR CORP Japanese Appl. 2006-297,237

A catalyst for exhaust gas purification is claimedwhich controls sintering of Pt particles. Pt is support-ed on a metal oxide carrier containing Mg and Aloxides, where atom ratio of Mg to total Mg and Al is1/3–4/5, and particle size of MgO is ≤ 20 nm. MgOis supported in a spinel.

Nitric Acid Reduction CatalystKOBE STEEL LTD Japanese Appl. 2006-314,888

A HNO3 reduction catalyst composition includesPt, Cu and Sn and is effective for reduction of NO3

–,NO2

–, etc. in the presence of H2(g), including at highHNO3 concentrations and pH ≤ 2. The atomic ratioof each constituent is in the range (in at.% vs. Pt)10–50 Cu and 0.5–10 Sn, and the composition maybe supported on a solid support.

Hydrogenation CatalystDALIAN TECHNOL. UNIV. Chinese Appl. 1,775,361

Nanoparticulate hydrogenation catalysts, with diam-eter 5 nm–500 μm, are prepared using a chemicaldisplacement process. A salt of a metal N, selectedfrom Ru, Rh, Pd, Pt and Ir or a mixture, is combinedwith a metal M, selected from Zn, Al, Fe, Co and Ni,then a displacement reaction is carried out to givemetal N. When the molar ratio N:M is between1:10–1:10,000, a catalyst of the type N/M is formed,where N is supported on M. When the molar ratioN:M is between 1:1–1:10, a highly dispersed N colloidcatalyst is formed. Addition of a stabilising agent tothe colloid gives a supported catalyst of type N/S,where S = C, Al2O3, SiO2, MgO, ZrO2 or CeO2.

HOMOGENEOUS CATALYSISContinuous Metathesis with Ruthenium CatalystsBOEHRINGER INGELHEIM INT. European Appl. 1,765,497

A continuous process for carrying out a metathesisreaction, such as RCM, in the presence of a Ru cata-lyst, is claimed. The five-coordinated Ru complexcatalyst includes 2 anionic ligands, 2 neutral ligandsand 1 carbene ligand which may optionally be linkedto one or both of the neutral ligands. Reaction may becarried out in solution on a packed column with afixed retention time, in a sequence of one or morestirred vessels with a catalyst inactivation region in thefinal vessel, or in one or more microreactors.


Enantiomers of 3-Hydroxy-3-phenyl-propylaminesBOEHRINGER INGELHEIM PHARMA GmbH

European Appl. 1,765,766An improved process for industrial preparation of

enantiomerically pure 3-hydroxy-3-phenyl-propylamines uses asymmetrical hydrogenation as akey step, with an optional sequence of subsequentsteps. The catalyst system consists of Rh and chiral 4-(dicyclohexylphosphino)-2-(diphenylphosphinomethyl)-N-methylaminocarbonylpyrrolidine. Reaction iscarried out at 0–100ºC, 1–150 bar, for 2–48 h andoptionally in the presence of a protic diluent such asH2O and less than 1 equivalent of a weak base. Theprocess can be used in the synthesis of products suchas R-atomoxetine, S-fluoxetine or S-norfluoxetine.

ROMP with Fluorinated GroupsD. LAZZARI et al. U.S. Appl. 2007/0,037,940

New metathesis oligomers are claimed which aresubstituted with fluorinated groups, and can be pre-pared by ROMP from a polymerisable compositionincluding: a penta- or hexavalent Ru or Os carbenecatalyst which contains 2 anionic ligands; 2 or 3 mon-odentate, neutral electron donor ligands; and an aryl,arylthio or C3–C5 alkenyl carbene ligand. The prod-ucts can be used to increase the oil and waterrepellence of organic materials.

FUEL CELLSRuthenium-Rhodium Alloy Electrode CatalystsLG CHEM. LTD European Appl. 1,771,903

An electrode catalyst for a MEA is made from aRuRh alloy, where each element is present in 10–90mol%. Supports may include porous C, conductivepolymers or metal oxides. The catalyst has good O2

reduction activity and MeOH resistance and can beused for a cathode in PEMFC, DMFC, etc.

Platinum-Containing ElectrocatalystsHONDA MOTOR CO LTD World Appl. 2007/024,489

Compositions formed from alloys of Pt and W withone of Ni or Zr can be used as a thin film electrocat-alyst in a fuel cell assembly. A first compositioncontains (in at.%): 20–45 Pt, 30–70 W and 5–25 Ni;preferably 20–40 Pt, 35–55 W and 15–25 Ni. A sec-ond composition contains (in at.%): 20–45 Pt, 30–70W and 5–40 Zr; preferably 30–40 Pt, 30–45 W and15–35 Zr.

Hydrogen Purification MembranesD. J. EDLUND et al. U.S. Patent 7,195,663

A steam-reforming fuel processor for productionof substantially pure H2(g) from a C-containingfeedstock plus H2O includes a steam-reformingcatalyst and a H2 separation assembly. At least oneH2 selective membrane is included, made from Pdor a Pd alloy such as PdCu, which may furtherincorporate ~ 5–250 ppm C, preferably < 100 ppmC. The fuel processor as claimed may be used incombination with a fuel cell stack adapted toreceive the product H2(g) stream.

Gold-Platinum Alloy NanoparticlesC.-J. ZHONG et al. U.S. Patent 7,208,439

C-supported AuPt nanoparticle alloy catalysts forfuel cells, especially those using MeOH as fuel, areclaimed. The Au:Pt ratio can be controlled in therange 1:99–99:1, and is preferably between50:50–80:20. Supports may include C black orC/TiO2. Bifunctional electrocatalytic activity isdemonstrated towards CO and MeOH oxidation andO2 reduction. Both anodes and cathodes may be pre-pared from the same catalyst material.

Production of Fuel GasIDEMITSU KOSAN CO LTD Japanese Appl. 2006-286,552

Fuel gas for SOFCs to be operated in mid- and low-temperature ranges can be produced using areforming catalyst containing at least one of Ru, Pt,Rh, Pd and Ir, preferably Ru, on a support selectedfrom MnO and CeO2, which may further incorporateAl2O3. Reforming processes may include partial oxi-dation, self-thermal reforming, steam reforming orCO2(g) reforming.

Membrane Electrode AssemblySAMSUNG SDI CO LTD Korean Appl. 2006-0,108,108

A MEA for a PEMFC includes an anode and acathode facing one another, with a PEM between,and can be applied to a high-temperature type poly-benzimidazole-based polymer. At least one of theelectrodes includes an electrode base having a sheet-like catalyst layer of thickness 10–100 μm, whichcontains a catalyst selected from Pt, Ru, Os, PtRu,PtOs, PtPd or PtM (where M = Ga, Ti, V, Cr, Mn,Fe, Co, Ni, Cu or Zn).

MEDICAL USESDemethylcantharidin Platinum Complex IsomersCHINESE UNIV. HONG KONG European Appl. 1,749,831

Stereoisomers of the demethylcantharidin Pt com-plex are claimed to inhibit growth of cisplatin,carboplatin or oxaliplatin-sensitive or -resistanttumour cells. Isomers may include a cis-isomer, atrans-S,S(+)-isomer, a trans-R,R(–)-isomer or a trans-racemate and may be incorporated into apharmaceutical composition with a carrier.

Ruthenium Complexes for Treating CancersUNIV. LOUIS PASTEUR European Appl. 1,776,103

Novel Ru complexes are claimed which can be usedfor the treatment of various cancers, including cis-platin-resistant cancers, separately or in combinationwith other treatments. The claimed structures mayinclude 1 or 4 of either halogens or electron-donatingligands containing N, O, P or S, plus a metallocycleof 5–8 atoms coordinated via N.

Impeller for Rotary Ventricular Assist DeviceC. R. SHAMBAUGH et al. U.S. Appl. 2007/0,078,293

An impeller for an implantable blood pump is madefrom a magnetic alloy containing 70–80 wt.% Pt and20–30 wt.% Co. The alloy may be heat-treated toimprove its magnetic properties.

Why is platinum used over other catalytic mate-rials such as silver or gold?

The role of platinum in catalytic converters isto oxidise carbon monoxide (CO) and hydrocar-bons. Platinum is particularly effective at thisunder oxygen-excessive conditions, so is often themetal of choice for diesel applications. For petrol-powered vehicles (where there is a balancebetween reductants and oxidants in the exhaustgas), platinum and palladium can be equally effec-tive, and so the choice is often made on the basisof relative cost. The three-way catalyst used forpetrol vehicles must also be able to reduce NOx tonitrogen as well as oxidise CO and hydrocarbons– that is why rhodium is generally used in additionto platinum or palladium.

Of course, some of the other transition ele-ments are also capable of catalysing oxidationreactions. However, platinum has several advan-tages:– it has a high melting point;– its interactions with ‘poisons’ (such as sulfur

compounds) are limited to the metal surface;– it can be efficiently recycled.

Although its high melting point may seem irrel-evant, because the platinum will never come closeto that temperature during use, it does provide anindication of its overall thermal durability. In a cat-alytic converter, the metal is in the form ofnanoparticles, which are dispersed over the entiresurface of a highly porous support material. As thetemperature of the catalyst rises, the particles startto become mobile and can coalesce – this is calledsintering, and becomes particularly noticeable as themetal approaches its Tammann temperature, atwhich bulk mobility of the metal particlesbecomes measurable. This temperature is oftentaken to be half the material’s melting point on theabsolute temperature scale (1). Metals such as goldand silver have a Tammann temperature (seeTable I) that is well below the average exhaust-gas

temperature (600–700ºC) for a petrol car beingdriven on a motorway, and so this precludes theiruse in three-way catalysts.

In addition, metals such as silver and copperhave a high affinity for sulfur-containing mole-cules, with which they will react to formcompounds (such as metal sulfates or sulfides). Asthis happens, there will be progressively less metalavailable for the useful reactions to take place.Platinum is different because it tends not tobecome totally or irreversibly poisoned, i.e. sulfur-containing molecules inhibit rather than poisonplatinum-based catalysts (see also (2)).

S. E. GOLUNSKI

References1 C. N. Satterfield, “Heterogeneous Catalysis in

Industrial Practice”, 2nd Edn., 1996 reprint, KriegerPublishing, Melbourne, FL, U.S.A.

2 J. K. Dunleavy, Platinum Metals Rev., 2006, 50, (2),110


FINAL ANALYSIS

Why Use Platinum in CatalyticConverters?

DOI: 10.1595/147106707X205857

Table I

The Platinum Group Metals and Their NearestNeighbours, Showing Atomic Number, ChemicalSymbol and Tammann Temperature (ºC)

25Mn

485ºC

26Fe

630ºC

27Co

610ºC

28Ni

590ºC

29Cu

405ºC

43Tc

975ºC

44Ru

990ºC

45Rh

845ºC

46Pd

640ºC

47Ag

345ºC

75Re

1450ºC

76Os

1375ºC

77Ir

1085ºC

78Pt

750ºC

79Au

395ºC

The AuthorDr Stan Golunski is Technology Manager of GasPhase Catalysis at the Johnson Matthey TechnologyCentre, U.K. Since joining the company in 1989, hehas worked on fuel reforming, process catalysis,and catalytic aftertreatment for internal combustionengines.

Platinum Metals ReviewJohnson Matthey Plc, Precious Metals Marketing, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.

E-mail: [email protected]://www.platinummetalsreview.com/

platinum metals · pdf file · 2016-01-28platinum metals review e-issn...

Documents